Merge tag 'drm-next-2020-10-15' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "Not a major amount of change, the i915 trees got split into display
  and gt trees to better facilitate higher level review, and there's a
  major refactoring of i915 GEM locking to use more core kernel concepts
  (like ww-mutexes). msm gets per-process pagetables, older AMD SI cards
  get DC support, nouveau got a bump in displayport support with common
  code extraction from i915.

  Outside of drm this contains a couple of patches for hexint
  moduleparams which you've acked, and a virtio common code tree that
  you should also get via it's regular path.

  New driver:
   - Cadence MHDP8546 DisplayPort bridge driver

  core:
   - cross-driver scatterlist cleanups
   - devm_drm conversions
   - remove drm_dev_init
   - devm_drm_dev_alloc conversion

  ttm:
   - lots of refactoring and cleanups

  bridges:
   - chained bridge support in more drivers

  panel:
   - misc new panels

  scheduler:
   - cleanup priority levels

  displayport:
   - refactor i915 code into helpers for nouveau

  i915:
   - split into display and GT trees
   - WW locking refactoring in GEM
   - execbuf2 extension mechanism
   - syncobj timeline support
   - GEN 12 HOBL display powersaving
   - Rocket Lake display additions
   - Disable FBC on Tigerlake
   - Tigerlake Type-C + DP improvements
   - Hotplug interrupt refactoring

  amdgpu:
   - Sienna Cichlid updates
   - Navy Flounder updates
   - DCE6 (SI) support for DC
   - Plane rotation enabled
   - TMZ state info ioctl
   - PCIe DPC recovery support
   - DC interrupt handling refactor
   - OLED panel fixes

  amdkfd:
   - add SMI events for thermal throttling
   - SMI interface events ioctl update
   - process eviction counters

  radeon:
   - move to dma_ for allocations
   - expose sclk via sysfs

  msm:
   - DSI support for sm8150/sm8250
   - per-process GPU pagetable support
   - Displayport support

  mediatek:
   - move HDMI phy driver to PHY
   - convert mtk-dpi to bridge API
   - disable mt2701 tmds

  tegra:
   - bridge support

  exynos:
   - misc cleanups

  vc4:
   - dual display cleanups

  ast:
   - cleanups

  gma500:
   - conversion to GPIOd API

  hisilicon:
   - misc reworks

  ingenic:
   - clock handling and format improvements

  mcde:
   - DSI support

  mgag200:
   - desktop g200 support

  mxsfb:
   - i.MX7 + i.MX8M
   - alpha plane support

  panfrost:
   - devfreq support
   - amlogic SoC support

  ps8640:
   - EDID from eDP retrieval

  tidss:
   - AM65xx YUV workaround

  virtio:
   - virtio-gpu exported resources

  rcar-du:
   - R8A7742, R8A774E1 and R8A77961 support
   - YUV planar format fixes
   - non-visible plane handling
   - VSP device reference count fix
   - Kconfig fix to avoid displaying disabled options in .config"

* tag 'drm-next-2020-10-15' of git://anongit.freedesktop.org/drm/drm: (1494 commits)
  drm/ingenic: Fix bad revert
  drm/amdgpu: Fix invalid number of character '{' in amdgpu_acpi_init
  drm/amdgpu: Remove warning for virtual_display
  drm/amdgpu: kfd_initialized can be static
  drm/amd/pm: setup APU dpm clock table in SMU HW initialization
  drm/amdgpu: prevent spurious warning
  drm/amdgpu/swsmu: fix ARC build errors
  drm/amd/display: Fix OPTC_DATA_FORMAT programming
  drm/amd/display: Don't allow pstate if no support in blank
  drm/panfrost: increase readl_relaxed_poll_timeout values
  MAINTAINERS: Update entry for st7703 driver after the rename
  Revert "gpu/drm: ingenic: Add option to mmap GEM buffers cached"
  drm/amd/display: HDMI remote sink need mode validation for Linux
  drm/amd/display: Change to correct unit on audio rate
  drm/amd/display: Avoid set zero in the requested clk
  drm/amdgpu: align frag_end to covered address space
  drm/amdgpu: fix NULL pointer dereference for Renoir
  drm/vmwgfx: fix regression in thp code due to ttm init refactor.
  drm/amdgpu/swsmu: add interrupt work handler for smu11 parts
  drm/amdgpu/swsmu: add interrupt work function
  ...

1152 files changed, 59383 insertions, 17587 deletions

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 2f31579f91d4..81569009f884 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -100,7 +100,7 @@ obj-$(CONFIG_DRM_MSM) += msm/
 obj-$(CONFIG_DRM_TEGRA) += tegra/
 obj-$(CONFIG_DRM_STM) += stm/
 obj-$(CONFIG_DRM_STI) += sti/
-obj-$(CONFIG_DRM_IMX) += imx/
+obj-y 			+= imx/
 obj-$(CONFIG_DRM_INGENIC) += ingenic/
 obj-$(CONFIG_DRM_MEDIATEK) += mediatek/
 obj-$(CONFIG_DRM_MESON)	+= meson/
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 403ec3db29df..39976c7b100c 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -30,7 +30,7 @@ FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME)
 ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_PATH)/include \
 	-I$(FULL_AMD_PATH)/amdgpu \
-	-I$(FULL_AMD_PATH)/powerplay/inc \
+	-I$(FULL_AMD_PATH)/pm/inc \
 	-I$(FULL_AMD_PATH)/acp/include \
 	-I$(FULL_AMD_DISPLAY_PATH) \
 	-I$(FULL_AMD_DISPLAY_PATH)/include \
@@ -47,7 +47,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_encoders.o amdgpu_display.o amdgpu_i2c.o \
 	amdgpu_fb.o amdgpu_gem.o amdgpu_ring.o \
 	amdgpu_cs.o amdgpu_bios.o amdgpu_benchmark.o amdgpu_test.o \
-	amdgpu_pm.o atombios_dp.o amdgpu_afmt.o amdgpu_trace_points.o \
+	atombios_dp.o amdgpu_afmt.o amdgpu_trace_points.o \
 	atombios_encoders.o amdgpu_sa.o atombios_i2c.o \
 	amdgpu_dma_buf.o amdgpu_vm.o amdgpu_ib.o amdgpu_pll.o \
 	amdgpu_ucode.o amdgpu_bo_list.o amdgpu_ctx.o amdgpu_sync.o \
@@ -55,15 +55,15 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \
 	amdgpu_gmc.o amdgpu_mmhub.o amdgpu_xgmi.o amdgpu_csa.o amdgpu_ras.o amdgpu_vm_cpu.o \
 	amdgpu_vm_sdma.o amdgpu_discovery.o amdgpu_ras_eeprom.o amdgpu_nbio.o \
-	amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o
+	amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o
 
 amdgpu-$(CONFIG_PERF_EVENTS) += amdgpu_pmu.o
 
 # add asic specific block
-amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \
+amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o \
 	dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.o
 
-amdgpu-$(CONFIG_DRM_AMDGPU_SI)+= si.o gmc_v6_0.o gfx_v6_0.o si_ih.o si_dma.o dce_v6_0.o si_dpm.o si_smc.o \
+amdgpu-$(CONFIG_DRM_AMDGPU_SI)+= si.o gmc_v6_0.o gfx_v6_0.o si_ih.o si_dma.o dce_v6_0.o \
 	uvd_v3_1.o
 
 amdgpu-y += \
@@ -85,7 +85,7 @@ amdgpu-y += \
 
 # add UMC block
 amdgpu-y += \
-	umc_v6_1.o umc_v6_0.o
+	umc_v6_1.o umc_v6_0.o umc_v8_7.o
 
 # add IH block
 amdgpu-y += \
@@ -105,10 +105,6 @@ amdgpu-y += \
 	psp_v11_0.o \
 	psp_v12_0.o
 
-# add SMC block
-amdgpu-y += \
-	amdgpu_dpm.o
-
 # add DCE block
 amdgpu-y += \
 	dce_v10_0.o \
@@ -212,7 +208,7 @@ amdgpu-$(CONFIG_VGA_SWITCHEROO) += amdgpu_atpx_handler.o
 amdgpu-$(CONFIG_ACPI) += amdgpu_acpi.o
 amdgpu-$(CONFIG_HMM_MIRROR) += amdgpu_mn.o
 
-include $(FULL_AMD_PATH)/powerplay/Makefile
+include $(FULL_AMD_PATH)/pm/Makefile
 
 amdgpu-y += $(AMD_POWERPLAY_FILES)
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 327a0daf4a1d..87f095dc385c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -49,6 +49,8 @@
 #include <linux/rbtree.h>
 #include <linux/hashtable.h>
 #include <linux/dma-fence.h>
+#include <linux/pci.h>
+#include <linux/aer.h>
 
 #include <drm/ttm/ttm_bo_api.h>
 #include <drm/ttm/ttm_bo_driver.h>
@@ -102,6 +104,7 @@
 #include "amdgpu_mes.h"
 #include "amdgpu_umc.h"
 #include "amdgpu_mmhub.h"
+#include "amdgpu_gfxhub.h"
 #include "amdgpu_df.h"
 
 #define MAX_GPU_INSTANCE		16
@@ -178,6 +181,7 @@ extern uint amdgpu_dm_abm_level;
 extern struct amdgpu_mgpu_info mgpu_info;
 extern int amdgpu_ras_enable;
 extern uint amdgpu_ras_mask;
+extern int amdgpu_bad_page_threshold;
 extern int amdgpu_async_gfx_ring;
 extern int amdgpu_mcbp;
 extern int amdgpu_discovery;
@@ -187,9 +191,11 @@ extern int amdgpu_force_asic_type;
 #ifdef CONFIG_HSA_AMD
 extern int sched_policy;
 extern bool debug_evictions;
+extern bool no_system_mem_limit;
 #else
 static const int sched_policy = KFD_SCHED_POLICY_HWS;
 static const bool debug_evictions; /* = false */
+static const bool no_system_mem_limit;
 #endif
 
 extern int amdgpu_tmz;
@@ -201,6 +207,7 @@ extern int amdgpu_si_support;
 #ifdef CONFIG_DRM_AMDGPU_CIK
 extern int amdgpu_cik_support;
 #endif
+extern int amdgpu_num_kcq;
 
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
@@ -212,6 +219,8 @@ extern int amdgpu_cik_support;
 #define AMDGPUFB_CONN_LIMIT			4
 #define AMDGPU_BIOS_NUM_SCRATCH			16
 
+#define AMDGPU_VBIOS_VGA_ALLOCATION		(9 * 1024 * 1024) /* reserve 8MB for vga emulator and 1 MB for FB */
+
 /* hard reset data */
 #define AMDGPU_ASIC_RESET_DATA                  0x39d5e86b
 
@@ -245,6 +254,7 @@ struct amdgpu_fpriv;
 struct amdgpu_bo_va_mapping;
 struct amdgpu_atif;
 struct kfd_vm_fault_info;
+struct amdgpu_hive_info;
 
 enum amdgpu_cp_irq {
 	AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP = 0,
@@ -611,6 +621,8 @@ struct amdgpu_asic_funcs {
 	uint64_t (*get_pcie_replay_count)(struct amdgpu_device *adev);
 	/* device supports BACO */
 	bool (*supports_baco)(struct amdgpu_device *adev);
+	/* pre asic_init quirks */
+	void (*pre_asic_init)(struct amdgpu_device *adev);
 };
 
 /*
@@ -648,16 +660,6 @@ struct amdgpu_atcs {
 };
 
 /*
- * Firmware VRAM reservation
- */
-struct amdgpu_fw_vram_usage {
-	u64 start_offset;
-	u64 size;
-	struct amdgpu_bo *reserved_bo;
-	void *va;
-};
-
-/*
  * CGS
  */
 struct cgs_device *amdgpu_cgs_create_device(struct amdgpu_device *adev);
@@ -725,13 +727,13 @@ struct amd_powerplay {
 #define AMDGPU_MAX_DF_PERFMONS 4
 struct amdgpu_device {
 	struct device			*dev;
-	struct drm_device		*ddev;
 	struct pci_dev			*pdev;
+	struct drm_device		ddev;
 
 #ifdef CONFIG_DRM_AMD_ACP
 	struct amdgpu_acp		acp;
 #endif
-
+	struct amdgpu_hive_info *hive;
 	/* ASIC */
 	enum amd_asic_type		asic_type;
 	uint32_t			family;
@@ -765,7 +767,6 @@ struct amdgpu_device {
 	bool				is_atom_fw;
 	uint8_t				*bios;
 	uint32_t			bios_size;
-	struct amdgpu_bo		*stolen_vga_memory;
 	uint32_t			bios_scratch_reg_offset;
 	uint32_t			bios_scratch[AMDGPU_BIOS_NUM_SCRATCH];
 
@@ -881,6 +882,9 @@ struct amdgpu_device {
 	/* mmhub */
 	struct amdgpu_mmhub		mmhub;
 
+	/* gfxhub */
+	struct amdgpu_gfxhub		gfxhub;
+
 	/* gfx */
 	struct amdgpu_gfx		gfx;
 
@@ -917,11 +921,6 @@ struct amdgpu_device {
 	/* display related functionality */
 	struct amdgpu_display_manager dm;
 
-	/* discovery */
-	uint8_t				*discovery_bin;
-	uint32_t			discovery_tmr_size;
-	struct amdgpu_bo		*discovery_memory;
-
 	/* mes */
 	bool                            enable_mes;
 	struct amdgpu_mes               mes;
@@ -946,8 +945,6 @@ struct amdgpu_device {
 	struct delayed_work     delayed_init_work;
 
 	struct amdgpu_virt	virt;
-	/* firmware VRAM reservation */
-	struct amdgpu_fw_vram_usage fw_vram_usage;
 
 	/* link all shadow bo */
 	struct list_head                shadow_list;
@@ -961,9 +958,9 @@ struct amdgpu_device {
 	bool                            in_suspend;
 	bool				in_hibernate;
 
-	bool                            in_gpu_reset;
+	atomic_t 			in_gpu_reset;
 	enum pp_mp1_state               mp1_state;
-	struct mutex  lock_reset;
+	struct rw_semaphore reset_sem;
 	struct amdgpu_doorbell_index doorbell_index;
 
 	struct mutex			notifier_lock;
@@ -995,34 +992,60 @@ struct amdgpu_device {
 
 	atomic_t			throttling_logging_enabled;
 	struct ratelimit_state		throttling_logging_rs;
+	uint32_t			ras_features;
+
+	bool                            in_pci_err_recovery;
+	struct pci_saved_state          *pci_state;
 };
 
+static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
+{
+	return container_of(ddev, struct amdgpu_device, ddev);
+}
+
+static inline struct drm_device *adev_to_drm(struct amdgpu_device *adev)
+{
+	return &adev->ddev;
+}
+
 static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev)
 {
 	return container_of(bdev, struct amdgpu_device, mman.bdev);
 }
 
 int amdgpu_device_init(struct amdgpu_device *adev,
-		       struct drm_device *ddev,
-		       struct pci_dev *pdev,
 		       uint32_t flags);
 void amdgpu_device_fini(struct amdgpu_device *adev);
 int amdgpu_gpu_wait_for_idle(struct amdgpu_device *adev);
 
 void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
 			       uint32_t *buf, size_t size, bool write);
-uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg,
+uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
+			    uint32_t reg, uint32_t acc_flags);
+void amdgpu_device_wreg(struct amdgpu_device *adev,
+			uint32_t reg, uint32_t v,
 			uint32_t acc_flags);
-void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
-		    uint32_t acc_flags);
-void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
-		    uint32_t acc_flags);
+void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev,
+			     uint32_t reg, uint32_t v);
 void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value);
 uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset);
 
 u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg);
 void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v);
 
+u32 amdgpu_device_indirect_rreg(struct amdgpu_device *adev,
+				u32 pcie_index, u32 pcie_data,
+				u32 reg_addr);
+u64 amdgpu_device_indirect_rreg64(struct amdgpu_device *adev,
+				  u32 pcie_index, u32 pcie_data,
+				  u32 reg_addr);
+void amdgpu_device_indirect_wreg(struct amdgpu_device *adev,
+				 u32 pcie_index, u32 pcie_data,
+				 u32 reg_addr, u32 reg_data);
+void amdgpu_device_indirect_wreg64(struct amdgpu_device *adev,
+				   u32 pcie_index, u32 pcie_data,
+				   u32 reg_addr, u64 reg_data);
+
 bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type);
 bool amdgpu_device_has_dc_support(struct amdgpu_device *adev);
 
@@ -1033,8 +1056,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
  */
 #define AMDGPU_REGS_NO_KIQ    (1<<1)
 
-#define RREG32_NO_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
-#define WREG32_NO_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
+#define RREG32_NO_KIQ(reg) amdgpu_device_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
+#define WREG32_NO_KIQ(reg, v) amdgpu_device_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
 
 #define RREG32_KIQ(reg) amdgpu_kiq_rreg(adev, (reg))
 #define WREG32_KIQ(reg, v) amdgpu_kiq_wreg(adev, (reg), (v))
@@ -1042,9 +1065,9 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define RREG8(reg) amdgpu_mm_rreg8(adev, (reg))
 #define WREG8(reg, v) amdgpu_mm_wreg8(adev, (reg), (v))
 
-#define RREG32(reg) amdgpu_mm_rreg(adev, (reg), 0)
-#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_mm_rreg(adev, (reg), 0))
-#define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0)
+#define RREG32(reg) amdgpu_device_rreg(adev, (reg), 0)
+#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_device_rreg(adev, (reg), 0))
+#define WREG32(reg, v) amdgpu_device_wreg(adev, (reg), (v), 0)
 #define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
 #define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
 #define RREG32_PCIE(reg) adev->pcie_rreg(adev, (reg))
@@ -1090,7 +1113,7 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 		WREG32_SMC(_Reg, tmp);                          \
 	} while (0)
 
-#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_mm_rreg((adev), (reg), false))
+#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_device_rreg((adev), (reg), false))
 #define RREG32_IO(reg) amdgpu_io_rreg(adev, (reg))
 #define WREG32_IO(reg, v) amdgpu_io_wreg(adev, (reg), (v))
 
@@ -1141,10 +1164,12 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define amdgpu_asic_need_reset_on_init(adev) (adev)->asic_funcs->need_reset_on_init((adev))
 #define amdgpu_asic_get_pcie_replay_count(adev) ((adev)->asic_funcs->get_pcie_replay_count((adev)))
 #define amdgpu_asic_supports_baco(adev) (adev)->asic_funcs->supports_baco((adev))
+#define amdgpu_asic_pre_asic_init(adev) (adev)->asic_funcs->pre_asic_init((adev))
 
 #define amdgpu_inc_vram_lost(adev) atomic_inc(&((adev)->vram_lost_counter));
 
 /* Common functions */
+bool amdgpu_device_has_job_running(struct amdgpu_device *adev);
 bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev);
 int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 			      struct amdgpu_job* job);
@@ -1194,7 +1219,7 @@ static inline void *amdgpu_atpx_get_dhandle(void) { return NULL; }
 extern const struct drm_ioctl_desc amdgpu_ioctls_kms[];
 extern const int amdgpu_max_kms_ioctl;
 
-int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags);
+int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags);
 void amdgpu_driver_unload_kms(struct drm_device *dev);
 void amdgpu_driver_lastclose_kms(struct drm_device *dev);
 int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv);
@@ -1258,6 +1283,15 @@ static inline int amdgpu_dm_display_resume(struct amdgpu_device *adev) { return
 void amdgpu_register_gpu_instance(struct amdgpu_device *adev);
 void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev);
 
+pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev,
+					   pci_channel_state_t state);
+pci_ers_result_t amdgpu_pci_mmio_enabled(struct pci_dev *pdev);
+pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev);
+void amdgpu_pci_resume(struct pci_dev *pdev);
+
+bool amdgpu_device_cache_pci_state(struct pci_dev *pdev);
+bool amdgpu_device_load_pci_state(struct pci_dev *pdev);
+
 #include "amdgpu_object.h"
 
 /* used by df_v3_6.c and amdgpu_pmu.c */
@@ -1278,4 +1312,8 @@ static inline bool amdgpu_is_tmz(struct amdgpu_device *adev)
        return adev->gmc.tmz_enabled;
 }
 
+static inline int amdgpu_in_reset(struct amdgpu_device *adev)
+{
+	return atomic_read(&adev->in_gpu_reset);
+}
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
index 12247a32f9ef..d3e51d361179 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
@@ -136,9 +136,7 @@ static int acp_poweroff(struct generic_pm_domain *genpd)
 	 * 2. power off the acp tiles
 	 * 3. check and enter ulv state
 	 */
-		if (adev->powerplay.pp_funcs &&
-			adev->powerplay.pp_funcs->set_powergating_by_smu)
-			amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, true);
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, true);
 	}
 	return 0;
 }
@@ -157,8 +155,7 @@ static int acp_poweron(struct generic_pm_domain *genpd)
 	 * 2. turn on acp clock
 	 * 3. power on acp tiles
 	 */
-		if (adev->powerplay.pp_funcs->set_powergating_by_smu)
-			amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, false);
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, false);
 	}
 	return 0;
 }
@@ -529,9 +526,7 @@ static int acp_set_powergating_state(void *handle,
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool enable = (state == AMD_PG_STATE_GATE);
 
-	if (adev->powerplay.pp_funcs &&
-		adev->powerplay.pp_funcs->set_powergating_by_smu)
-		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, enable);
+	amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, enable);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 913c8f0513bd..165b02e267b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -463,11 +463,11 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev,
 
 		if (req.pending & ATIF_DGPU_DISPLAY_EVENT) {
 			if (adev->flags & AMD_IS_PX) {
-				pm_runtime_get_sync(adev->ddev->dev);
+				pm_runtime_get_sync(adev_to_drm(adev)->dev);
 				/* Just fire off a uevent and let userspace tell us what to do */
-				drm_helper_hpd_irq_event(adev->ddev);
-				pm_runtime_mark_last_busy(adev->ddev->dev);
-				pm_runtime_put_autosuspend(adev->ddev->dev);
+				drm_helper_hpd_irq_event(adev_to_drm(adev));
+				pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+				pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			}
 		}
 		/* TODO: check other events */
@@ -806,8 +806,8 @@ int amdgpu_acpi_init(struct amdgpu_device *adev)
 	}
 	adev->atif = atif;
 
-	if (atif->notifications.brightness_change) {
 #if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) || defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE)
+	if (atif->notifications.brightness_change) {
 		if (amdgpu_device_has_dc_support(adev)) {
 #if defined(CONFIG_DRM_AMD_DC)
 			struct amdgpu_display_manager *dm = &adev->dm;
@@ -817,7 +817,7 @@ int amdgpu_acpi_init(struct amdgpu_device *adev)
 			struct drm_encoder *tmp;
 
 			/* Find the encoder controlling the brightness */
-			list_for_each_entry(tmp, &adev->ddev->mode_config.encoder_list,
+			list_for_each_entry(tmp, &adev_to_drm(adev)->mode_config.encoder_list,
 					    head) {
 				struct amdgpu_encoder *enc = to_amdgpu_encoder(tmp);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 1b865fed74ca..0544460653b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -36,6 +36,8 @@
  */
 uint64_t amdgpu_amdkfd_total_mem_size;
 
+static bool kfd_initialized;
+
 int amdgpu_amdkfd_init(void)
 {
 	struct sysinfo si;
@@ -51,19 +53,26 @@ int amdgpu_amdkfd_init(void)
 #else
 	ret = -ENOENT;
 #endif
+	kfd_initialized = !ret;
 
 	return ret;
 }
 
 void amdgpu_amdkfd_fini(void)
 {
-	kgd2kfd_exit();
+	if (kfd_initialized) {
+		kgd2kfd_exit();
+		kfd_initialized = false;
+	}
 }
 
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 {
 	bool vf = amdgpu_sriov_vf(adev);
 
+	if (!kfd_initialized)
+		return;
+
 	adev->kfd.dev = kgd2kfd_probe((struct kgd_dev *)adev,
 				      adev->pdev, adev->asic_type, vf);
 
@@ -119,7 +128,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 			.gpuvm_size = min(adev->vm_manager.max_pfn
 					  << AMDGPU_GPU_PAGE_SHIFT,
 					  AMDGPU_GMC_HOLE_START),
-			.drm_render_minor = adev->ddev->render->index,
+			.drm_render_minor = adev_to_drm(adev)->render->index,
 			.sdma_doorbell_idx = adev->doorbell_index.sdma_engine,
 
 		};
@@ -160,7 +169,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 					adev->doorbell_index.last_non_cp;
 		}
 
-		kgd2kfd_device_init(adev->kfd.dev, adev->ddev, &gpu_resources);
+		kgd2kfd_device_init(adev->kfd.dev, adev_to_drm(adev), &gpu_resources);
 	}
 }
 
@@ -479,11 +488,11 @@ int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd,
 		goto out_put;
 
 	obj = dma_buf->priv;
-	if (obj->dev->driver != adev->ddev->driver)
+	if (obj->dev->driver != adev_to_drm(adev)->driver)
 		/* Can't handle buffers from different drivers */
 		goto out_put;
 
-	adev = obj->dev->dev_private;
+	adev = drm_to_adev(obj->dev);
 	bo = gem_to_amdgpu_bo(obj);
 	if (!(bo->preferred_domains & (AMDGPU_GEM_DOMAIN_VRAM |
 				    AMDGPU_GEM_DOMAIN_GTT)))
@@ -517,8 +526,9 @@ out_put:
 uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+	struct ttm_resource_manager *vram_man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 
-	return amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+	return amdgpu_vram_mgr_usage(vram_man);
 }
 
 uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd)
@@ -571,6 +581,13 @@ uint32_t amdgpu_amdkfd_get_asic_rev_id(struct kgd_dev *kgd)
 	return adev->rev_id;
 }
 
+int amdgpu_amdkfd_get_noretry(struct kgd_dev *kgd)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	return adev->gmc.noretry;
+}
+
 int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine,
 				uint32_t vmid, uint64_t gpu_addr,
 				uint32_t *ib_cmd, uint32_t ib_len)
@@ -612,6 +629,7 @@ int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine,
 	job->vmid = vmid;
 
 	ret = amdgpu_ib_schedule(ring, 1, ib, job, &f);
+
 	if (ret) {
 		DRM_ERROR("amdgpu: failed to schedule IB.\n");
 		goto err_ib_sched;
@@ -755,4 +773,8 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd)
 {
 }
+
+void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint32_t throttle_bitmask)
+{
+}
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index dfef5a7e0f5a..ea391ca7f2f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -181,6 +181,7 @@ uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd);
 uint64_t amdgpu_amdkfd_get_mmio_remap_phys_addr(struct kgd_dev *kgd);
 uint32_t amdgpu_amdkfd_get_num_gws(struct kgd_dev *kgd);
 uint32_t amdgpu_amdkfd_get_asic_rev_id(struct kgd_dev *kgd);
+int amdgpu_amdkfd_get_noretry(struct kgd_dev *kgd);
 uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src);
 
 /* Read user wptr from a specified user address space with page fault
@@ -270,5 +271,6 @@ int kgd2kfd_resume_mm(struct mm_struct *mm);
 int kgd2kfd_schedule_evict_and_restore_process(struct mm_struct *mm,
 					       struct dma_fence *fence);
 void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd);
+void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint32_t throttle_bitmask);
 
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 35d4a5ab0228..1afa8f122e7d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -283,22 +283,6 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
 	return 0;
 }
 
-static void kgd_set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
-		uint64_t page_table_base)
-{
-	struct amdgpu_device *adev = get_amdgpu_device(kgd);
-
-	if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid)) {
-		pr_err("trying to set page table base for wrong VMID %u\n",
-		       vmid);
-		return;
-	}
-
-	mmhub_v9_4_setup_vm_pt_regs(adev, vmid, page_table_base);
-
-	gfxhub_v1_0_setup_vm_pt_regs(adev, vmid, page_table_base);
-}
-
 const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -317,7 +301,7 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.wave_control_execute = kgd_gfx_v9_wave_control_execute,
 	.address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset,
 	.get_atc_vmid_pasid_mapping_info =
-			kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
-	.set_vm_context_page_table_base = kgd_set_vm_context_page_table_base,
-	.get_hive_id = amdgpu_amdkfd_get_hive_id,
+				kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
+	.set_vm_context_page_table_base =
+				kgd_gfx_v9_set_vm_context_page_table_base,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index ee531c3988d1..4763bab7a4d0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -32,7 +32,6 @@
 #include "v10_structs.h"
 #include "nv.h"
 #include "nvd.h"
-#include "gfxhub_v2_0.h"
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -542,7 +541,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
 	uint32_t temp;
 	struct v10_compute_mqd *m = get_mqd(mqd);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 #if 0
@@ -753,7 +752,7 @@ static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 	}
 
 	/* SDMA is on gfxhub as well for Navi1* series */
-	gfxhub_v2_0_setup_vm_pt_regs(adev, vmid, page_table_base);
+	adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
 }
 
 const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
@@ -776,6 +775,4 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info =
 			get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
-	.get_hive_id = amdgpu_amdkfd_get_hive_id,
-	.get_unique_id = amdgpu_amdkfd_get_unique_id,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index cdea1338c8dc..50016bf9c427 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -31,7 +31,6 @@
 #include "v10_structs.h"
 #include "nv.h"
 #include "nvd.h"
-#include "gfxhub_v2_1.h"
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -657,7 +656,7 @@ static void set_vm_context_page_table_base_v10_3(struct kgd_dev *kgd, uint32_t v
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
 
 	/* SDMA is on gfxhub as well for Navi1* series */
-	gfxhub_v2_1_setup_vm_pt_regs(adev, vmid, page_table_base);
+	adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
 }
 
 #if 0
@@ -822,7 +821,6 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.address_watch_get_offset = address_watch_get_offset_v10_3,
 	.get_atc_vmid_pasid_mapping_info = NULL,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3,
-	.get_hive_id = amdgpu_amdkfd_get_hive_id,
 #if 0
 	.enable_debug_trap = enable_debug_trap_v10_3,
 	.disable_debug_trap = disable_debug_trap_v10_3,
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 4d41317b9292..b91d27e39bad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -423,7 +423,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
 	unsigned long flags, end_jiffies;
 	int retry;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 	acquire_queue(kgd, pipe_id, queue_id);
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 35917d4b50f6..5ce0ce704a21 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -419,7 +419,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd,
 	int retry;
 	struct vi_mqd *m = get_mqd(mqd);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 	acquire_queue(kgd, pipe_id, queue_id);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 1abfe63c80fe..43b18863a8b8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -36,9 +36,7 @@
 #include "v9_structs.h"
 #include "soc15.h"
 #include "soc15d.h"
-#include "mmhub_v1_0.h"
-#include "gfxhub_v1_0.h"
-
+#include "gfx_v9_0.h"
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -552,7 +550,7 @@ int kgd_gfx_v9_hqd_destroy(struct kgd_dev *kgd, void *mqd,
 	uint32_t temp;
 	struct v9_mqd *m = get_mqd(mqd);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 	acquire_queue(kgd, pipe_id, queue_id);
@@ -690,7 +688,7 @@ uint32_t kgd_gfx_v9_address_watch_get_offset(struct kgd_dev *kgd,
 	return 0;
 }
 
-static void kgd_gfx_v9_set_vm_context_page_table_base(struct kgd_dev *kgd,
+void kgd_gfx_v9_set_vm_context_page_table_base(struct kgd_dev *kgd,
 			uint32_t vmid, uint64_t page_table_base)
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
@@ -701,9 +699,182 @@ static void kgd_gfx_v9_set_vm_context_page_table_base(struct kgd_dev *kgd,
 		return;
 	}
 
-	mmhub_v1_0_setup_vm_pt_regs(adev, vmid, page_table_base);
+	adev->mmhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
+
+	adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
+}
+
+static void lock_spi_csq_mutexes(struct amdgpu_device *adev)
+{
+	mutex_lock(&adev->srbm_mutex);
+	mutex_lock(&adev->grbm_idx_mutex);
+
+}
+
+static void unlock_spi_csq_mutexes(struct amdgpu_device *adev)
+{
+	mutex_unlock(&adev->grbm_idx_mutex);
+	mutex_unlock(&adev->srbm_mutex);
+}
+
+/**
+ * @get_wave_count: Read device registers to get number of waves in flight for
+ * a particular queue. The method also returns the VMID associated with the
+ * queue.
+ *
+ * @adev: Handle of device whose registers are to be read
+ * @queue_idx: Index of queue in the queue-map bit-field
+ * @wave_cnt: Output parameter updated with number of waves in flight
+ * @vmid: Output parameter updated with VMID of queue whose wave count
+ * is being collected
+ */
+static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
+		int *wave_cnt, int *vmid)
+{
+	int pipe_idx;
+	int queue_slot;
+	unsigned int reg_val;
+
+	/*
+	 * Program GRBM with appropriate MEID, PIPEID, QUEUEID and VMID
+	 * parameters to read out waves in flight. Get VMID if there are
+	 * non-zero waves in flight.
+	 */
+	*vmid = 0xFF;
+	*wave_cnt = 0;
+	pipe_idx = queue_idx / adev->gfx.mec.num_queue_per_pipe;
+	queue_slot = queue_idx % adev->gfx.mec.num_queue_per_pipe;
+	soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0);
+	reg_val = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) +
+			 queue_slot);
+	*wave_cnt = reg_val & SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK;
+	if (*wave_cnt != 0)
+		*vmid = (RREG32_SOC15(GC, 0, mmCP_HQD_VMID) &
+			 CP_HQD_VMID__VMID_MASK) >> CP_HQD_VMID__VMID__SHIFT;
+}
+
+/**
+ * @kgd_gfx_v9_get_cu_occupancy: Reads relevant registers associated with each
+ * shader engine and aggregates the number of waves that are in flight for the
+ * process whose pasid is provided as a parameter. The process could have ZERO
+ * or more queues running and submitting waves to compute units.
+ *
+ * @kgd: Handle of device from which to get number of waves in flight
+ * @pasid: Identifies the process for which this query call is invoked
+ * @wave_cnt: Output parameter updated with number of waves in flight that
+ * belong to process with given pasid
+ * @max_waves_per_cu: Output parameter updated with maximum number of waves
+ * possible per Compute Unit
+ *
+ * @note: It's possible that the device has too many queues (oversubscription)
+ * in which case a VMID could be remapped to a different PASID. This could lead
+ * to an iaccurate wave count. Following is a high-level sequence:
+ *    Time T1: vmid = getVmid(); vmid is associated with Pasid P1
+ *    Time T2: passId = getPasId(vmid); vmid is associated with Pasid P2
+ * In the sequence above wave count obtained from time T1 will be incorrectly
+ * lost or added to total wave count.
+ *
+ * The registers that provide the waves in flight are:
+ *
+ *  SPI_CSQ_WF_ACTIVE_STATUS - bit-map of queues per pipe. The bit is ON if a
+ *  queue is slotted, OFF if there is no queue. A process could have ZERO or
+ *  more queues slotted and submitting waves to be run on compute units. Even
+ *  when there is a queue it is possible there could be zero wave fronts, this
+ *  can happen when queue is waiting on top-of-pipe events - e.g. waitRegMem
+ *  command
+ *
+ *  For each bit that is ON from above:
+ *
+ *    Read (SPI_CSQ_WF_ACTIVE_COUNT_0 + queue_idx) register. It provides the
+ *    number of waves that are in flight for the queue at specified index. The
+ *    index ranges from 0 to 7.
+ *
+ *    If non-zero waves are in flight, read CP_HQD_VMID register to obtain VMID
+ *    of the wave(s).
+ *
+ *    Determine if VMID from above step maps to pasid provided as parameter. If
+ *    it matches agrregate the wave count. That the VMID will not match pasid is
+ *    a normal condition i.e. a device is expected to support multiple queues
+ *    from multiple proceses.
+ *
+ *  Reading registers referenced above involves programming GRBM appropriately
+ */
+static void kgd_gfx_v9_get_cu_occupancy(struct kgd_dev *kgd, int pasid,
+		int *pasid_wave_cnt, int *max_waves_per_cu)
+{
+	int qidx;
+	int vmid;
+	int se_idx;
+	int sh_idx;
+	int se_cnt;
+	int sh_cnt;
+	int wave_cnt;
+	int queue_map;
+	int pasid_tmp;
+	int max_queue_cnt;
+	int vmid_wave_cnt = 0;
+	struct amdgpu_device *adev;
+	DECLARE_BITMAP(cp_queue_bitmap, KGD_MAX_QUEUES);
+
+	adev = get_amdgpu_device(kgd);
+	lock_spi_csq_mutexes(adev);
+	soc15_grbm_select(adev, 1, 0, 0, 0);
+
+	/*
+	 * Iterate through the shader engines and arrays of the device
+	 * to get number of waves in flight
+	 */
+	bitmap_complement(cp_queue_bitmap, adev->gfx.mec.queue_bitmap,
+			  KGD_MAX_QUEUES);
+	max_queue_cnt = adev->gfx.mec.num_pipe_per_mec *
+			adev->gfx.mec.num_queue_per_pipe;
+	sh_cnt = adev->gfx.config.max_sh_per_se;
+	se_cnt = adev->gfx.config.max_shader_engines;
+	for (se_idx = 0; se_idx < se_cnt; se_idx++) {
+		for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) {
+
+			gfx_v9_0_select_se_sh(adev, se_idx, sh_idx, 0xffffffff);
+			queue_map = RREG32(SOC15_REG_OFFSET(GC, 0,
+					   mmSPI_CSQ_WF_ACTIVE_STATUS));
+
+			/*
+			 * Assumption: queue map encodes following schema: four
+			 * pipes per each micro-engine, with each pipe mapping
+			 * eight queues. This schema is true for GFX9 devices
+			 * and must be verified for newer device families
+			 */
+			for (qidx = 0; qidx < max_queue_cnt; qidx++) {
+
+				/* Skip qeueus that are not associated with
+				 * compute functions
+				 */
+				if (!test_bit(qidx, cp_queue_bitmap))
+					continue;
+
+				if (!(queue_map & (1 << qidx)))
+					continue;
+
+				/* Get number of waves in flight and aggregate them */
+				get_wave_count(adev, qidx, &wave_cnt, &vmid);
+				if (wave_cnt != 0) {
+					pasid_tmp =
+					  RREG32(SOC15_REG_OFFSET(OSSSYS, 0,
+						 mmIH_VMID_0_LUT) + vmid);
+					if (pasid_tmp == pasid)
+						vmid_wave_cnt += wave_cnt;
+				}
+			}
+		}
+	}
+
+	gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	soc15_grbm_select(adev, 0, 0, 0, 0);
+	unlock_spi_csq_mutexes(adev);
 
-	gfxhub_v1_0_setup_vm_pt_regs(adev, vmid, page_table_base);
+	/* Update the output parameters and return */
+	*pasid_wave_cnt = vmid_wave_cnt;
+	*max_waves_per_cu = adev->gfx.cu_info.simd_per_cu *
+				adev->gfx.cu_info.max_waves_per_simd;
 }
 
 const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
@@ -726,6 +897,5 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info =
 			kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
-	.get_hive_id = amdgpu_amdkfd_get_hive_id,
-	.get_unique_id = amdgpu_amdkfd_get_unique_id,
+	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index ff2bc72e6646..fc8934b86d93 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -60,3 +60,6 @@ uint32_t kgd_gfx_v9_address_watch_get_offset(struct kgd_dev *kgd,
 
 bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd,
 					uint8_t vmid, uint16_t *p_pasid);
+
+void kgd_gfx_v9_set_vm_context_page_table_base(struct kgd_dev *kgd,
+			uint32_t vmid, uint64_t page_table_base);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index d02c5c177a98..5da487b64a66 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -148,8 +148,12 @@ static int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
 
+	if (kfd_mem_limit.system_mem_used + system_mem_needed >
+	    kfd_mem_limit.max_system_mem_limit)
+		pr_debug("Set no_system_mem_limit=1 if using shared memory\n");
+
 	if ((kfd_mem_limit.system_mem_used + system_mem_needed >
-	     kfd_mem_limit.max_system_mem_limit) ||
+	     kfd_mem_limit.max_system_mem_limit && !no_system_mem_limit) ||
 	    (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
 	     kfd_mem_limit.max_ttm_mem_limit) ||
 	    (adev->kfd.vram_used + vram_needed >
@@ -562,7 +566,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr)
 
 	mutex_lock(&process_info->lock);
 
-	ret = amdgpu_ttm_tt_set_userptr(bo->tbo.ttm, user_addr, 0);
+	ret = amdgpu_ttm_tt_set_userptr(&bo->tbo, user_addr, 0);
 	if (ret) {
 		pr_err("%s: Failed to set userptr: %d\n", __func__, ret);
 		goto out;
@@ -1668,7 +1672,7 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd,
 		return -EINVAL;
 
 	obj = dma_buf->priv;
-	if (obj->dev->dev_private != adev)
+	if (drm_to_adev(obj->dev) != adev)
 		/* Can't handle buffers from other devices */
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
index 29f767e026e4..469352e2d6ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@@ -148,7 +148,7 @@ void amdgpu_atombios_i2c_init(struct amdgpu_device *adev)
 
 			if (i2c.valid) {
 				sprintf(stmp, "0x%x", i2c.i2c_id);
-				adev->i2c_bus[i] = amdgpu_i2c_create(adev->ddev, &i2c, stmp);
+				adev->i2c_bus[i] = amdgpu_i2c_create(adev_to_drm(adev), &i2c, stmp);
 			}
 			gpio = (ATOM_GPIO_I2C_ASSIGMENT *)
 				((u8 *)gpio + sizeof(ATOM_GPIO_I2C_ASSIGMENT));
@@ -541,7 +541,7 @@ bool amdgpu_atombios_get_connector_info_from_object_table(struct amdgpu_device *
 		}
 	}
 
-	amdgpu_link_encoder_connector(adev->ddev);
+	amdgpu_link_encoder_connector(adev_to_drm(adev));
 
 	return true;
 }
@@ -1786,9 +1786,9 @@ static int amdgpu_atombios_allocate_fb_scratch(struct amdgpu_device *adev)
 			(uint32_t)(ATOM_VRAM_BLOCK_SRIOV_MSG_SHARE_RESERVATION <<
 			ATOM_VRAM_OPERATION_FLAGS_SHIFT)) {
 			/* Firmware request VRAM reservation for SR-IOV */
-			adev->fw_vram_usage.start_offset = (start_addr &
+			adev->mman.fw_vram_usage_start_offset = (start_addr &
 				(~ATOM_VRAM_OPERATION_FLAGS_MASK)) << 10;
-			adev->fw_vram_usage.size = size << 10;
+			adev->mman.fw_vram_usage_size = size << 10;
 			/* Use the default scratch size */
 			usage_bytes = 0;
 		} else {
@@ -1882,7 +1882,7 @@ static void cail_mc_write(struct card_info *info, uint32_t reg, uint32_t val)
  */
 static void cail_reg_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
-	struct amdgpu_device *adev = info->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(info->dev);
 
 	WREG32(reg, val);
 }
@@ -1898,7 +1898,7 @@ static void cail_reg_write(struct card_info *info, uint32_t reg, uint32_t val)
  */
 static uint32_t cail_reg_read(struct card_info *info, uint32_t reg)
 {
-	struct amdgpu_device *adev = info->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(info->dev);
 	uint32_t r;
 
 	r = RREG32(reg);
@@ -1916,7 +1916,7 @@ static uint32_t cail_reg_read(struct card_info *info, uint32_t reg)
  */
 static void cail_ioreg_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
-	struct amdgpu_device *adev = info->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(info->dev);
 
 	WREG32_IO(reg, val);
 }
@@ -1932,7 +1932,7 @@ static void cail_ioreg_write(struct card_info *info, uint32_t reg, uint32_t val)
  */
 static uint32_t cail_ioreg_read(struct card_info *info, uint32_t reg)
 {
-	struct amdgpu_device *adev = info->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(info->dev);
 	uint32_t r;
 
 	r = RREG32_IO(reg);
@@ -1944,7 +1944,7 @@ static ssize_t amdgpu_atombios_get_vbios_version(struct device *dev,
 						 char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	struct atom_context *ctx = adev->mode_info.atom_context;
 
 	return snprintf(buf, PAGE_SIZE, "%s\n", ctx->vbios_version);
@@ -1995,7 +1995,7 @@ int amdgpu_atombios_init(struct amdgpu_device *adev)
 		return -ENOMEM;
 
 	adev->mode_info.atom_card_info = atom_card_info;
-	atom_card_info->dev = adev->ddev;
+	atom_card_info->dev = adev_to_drm(adev);
 	atom_card_info->reg_read = cail_reg_read;
 	atom_card_info->reg_write = cail_reg_write;
 	/* needed for iio ops */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index 1279053324f9..b4df6460e45a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -89,9 +89,9 @@ int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev)
 			(uint32_t)(ATOM_VRAM_BLOCK_SRIOV_MSG_SHARE_RESERVATION <<
 			ATOM_VRAM_OPERATION_FLAGS_SHIFT)) {
 			/* Firmware request VRAM reservation for SR-IOV */
-			adev->fw_vram_usage.start_offset = (start_addr &
+			adev->mman.fw_vram_usage_start_offset = (start_addr &
 				(~ATOM_VRAM_OPERATION_FLAGS_MASK)) << 10;
-			adev->fw_vram_usage.size = size << 10;
+			adev->mman.fw_vram_usage_size = size << 10;
 			/* Use the default scratch size */
 			usage_bytes = 0;
 		} else {
@@ -543,6 +543,7 @@ int amdgpu_mem_train_support(struct amdgpu_device *adev)
 		case HW_REV(11, 0, 0):
 		case HW_REV(11, 0, 5):
 		case HW_REV(11, 0, 7):
+		case HW_REV(11, 0, 11):
 			ret = 1;
 			break;
 		default:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index 3e35a8f2c5e5..7abe9500c0c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -616,7 +616,7 @@ static bool amdgpu_atpx_detect(void)
 	while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev)) != NULL) {
 		vga_count++;
 
-		has_atpx |= (amdgpu_atpx_pci_probe_handle(pdev) == true);
+		has_atpx |= amdgpu_atpx_pci_probe_handle(pdev);
 
 		parent_pdev = pci_upstream_bridge(pdev);
 		d3_supported |= parent_pdev && parent_pdev->bridge_d3;
@@ -626,7 +626,7 @@ static bool amdgpu_atpx_detect(void)
 	while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_OTHER << 8, pdev)) != NULL) {
 		vga_count++;
 
-		has_atpx |= (amdgpu_atpx_pci_probe_handle(pdev) == true);
+		has_atpx |= amdgpu_atpx_pci_probe_handle(pdev);
 
 		parent_pdev = pci_upstream_bridge(pdev);
 		d3_supported |= parent_pdev && parent_pdev->bridge_d3;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
index b1172d93c99c..6333cada1e09 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
@@ -417,26 +417,40 @@ static inline bool amdgpu_acpi_vfct_bios(struct amdgpu_device *adev)
 
 bool amdgpu_get_bios(struct amdgpu_device *adev)
 {
-	if (amdgpu_atrm_get_bios(adev))
+	if (amdgpu_atrm_get_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from ATRM\n");
 		goto success;
+	}
 
-	if (amdgpu_acpi_vfct_bios(adev))
+	if (amdgpu_acpi_vfct_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from VFCT\n");
 		goto success;
+	}
 
-	if (igp_read_bios_from_vram(adev))
+	if (igp_read_bios_from_vram(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from VRAM BAR\n");
 		goto success;
+	}
 
-	if (amdgpu_read_bios(adev))
+	if (amdgpu_read_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from ROM BAR\n");
 		goto success;
+	}
 
-	if (amdgpu_read_bios_from_rom(adev))
+	if (amdgpu_read_bios_from_rom(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from ROM\n");
 		goto success;
+	}
 
-	if (amdgpu_read_disabled_bios(adev))
+	if (amdgpu_read_disabled_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from disabled ROM BAR\n");
 		goto success;
+	}
 
-	if (amdgpu_read_platform_bios(adev))
+	if (amdgpu_read_platform_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from platform\n");
 		goto success;
+	}
 
 	DRM_ERROR("Unable to locate a BIOS ROM\n");
 	return false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 4053597b3af2..15c45b2a3983 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -265,7 +265,7 @@ error_free:
 int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	union drm_amdgpu_bo_list *args = data;
 	uint32_t handle = args->in.list_handle;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index a1aec205435d..65d1b23d7e74 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -26,6 +26,7 @@
 
 #include <drm/drm_edid.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_dp_helper.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
@@ -41,7 +42,7 @@
 void amdgpu_connector_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 
 	/* bail if the connector does not have hpd pin, e.g.,
@@ -279,7 +280,7 @@ amdgpu_connector_get_hardcoded_edid(struct amdgpu_device *adev)
 static void amdgpu_connector_get_edid(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 
 	if (amdgpu_connector->edid)
@@ -463,7 +464,7 @@ static int amdgpu_connector_set_property(struct drm_connector *connector,
 					  uint64_t val)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_encoder *encoder;
 	struct amdgpu_encoder *amdgpu_encoder;
 
@@ -834,7 +835,7 @@ static enum drm_mode_status amdgpu_connector_vga_mode_valid(struct drm_connector
 					    struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* XXX check mode bandwidth */
 
@@ -941,7 +942,7 @@ static bool
 amdgpu_connector_check_hpd_status_unchanged(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	enum drm_connector_status status;
 
@@ -972,7 +973,7 @@ static enum drm_connector_status
 amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	const struct drm_encoder_helper_funcs *encoder_funcs;
 	int r;
@@ -1159,7 +1160,7 @@ static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector
 					    struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 
 	/* XXX check mode bandwidth */
@@ -1311,7 +1312,7 @@ static bool amdgpu_connector_encoder_is_hbr2(struct drm_connector *connector)
 bool amdgpu_connector_is_dp12_capable(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if ((adev->clock.default_dispclk >= 53900) &&
 	    amdgpu_connector_encoder_is_hbr2(connector)) {
@@ -1325,7 +1326,7 @@ static enum drm_connector_status
 amdgpu_connector_dp_detect(struct drm_connector *connector, bool force)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	enum drm_connector_status ret = connector_status_disconnected;
 	struct amdgpu_connector_atom_dig *amdgpu_dig_connector = amdgpu_connector->con_priv;
@@ -1413,6 +1414,10 @@ out:
 		pm_runtime_put_autosuspend(connector->dev->dev);
 	}
 
+	drm_dp_set_subconnector_property(&amdgpu_connector->base,
+					 ret,
+					 amdgpu_dig_connector->dpcd,
+					 amdgpu_dig_connector->downstream_ports);
 	return ret;
 }
 
@@ -1521,7 +1526,7 @@ amdgpu_connector_add(struct amdgpu_device *adev,
 		      struct amdgpu_hpd *hpd,
 		      struct amdgpu_router *router)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	struct amdgpu_connector *amdgpu_connector;
@@ -1959,6 +1964,11 @@ amdgpu_connector_add(struct amdgpu_device *adev,
 	if (has_aux)
 		amdgpu_atombios_dp_aux_init(amdgpu_connector);
 
+	if (connector_type == DRM_MODE_CONNECTOR_DisplayPort ||
+	    connector_type == DRM_MODE_CONNECTOR_eDP) {
+		drm_connector_attach_dp_subconnector_property(&amdgpu_connector->base);
+	}
+
 	return;
 
 failed:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index a512ccbc4dea..12598a4b5c78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -299,7 +299,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 {
 	s64 time_us, increment_us;
 	u64 free_vram, total_vram, used_vram;
-
+	struct ttm_resource_manager *vram_man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 	/* Allow a maximum of 200 accumulated ms. This is basically per-IB
 	 * throttling.
 	 *
@@ -316,7 +316,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 	}
 
 	total_vram = adev->gmc.real_vram_size - atomic64_read(&adev->vram_pin_size);
-	used_vram = amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+	used_vram = amdgpu_vram_mgr_usage(vram_man);
 	free_vram = used_vram >= total_vram ? 0 : total_vram - used_vram;
 
 	spin_lock(&adev->mm_stats.lock);
@@ -363,7 +363,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 	if (!amdgpu_gmc_vram_full_visible(&adev->gmc)) {
 		u64 total_vis_vram = adev->gmc.visible_vram_size;
 		u64 used_vis_vram =
-			amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+		  amdgpu_vram_mgr_vis_usage(vram_man);
 
 		if (used_vis_vram < total_vis_vram) {
 			u64 free_vis_vram = total_vis_vram - used_vis_vram;
@@ -1275,13 +1275,24 @@ error_unlock:
 	return r;
 }
 
+static void trace_amdgpu_cs_ibs(struct amdgpu_cs_parser *parser)
+{
+	int i;
+
+	if (!trace_amdgpu_cs_enabled())
+		return;
+
+	for (i = 0; i < parser->job->num_ibs; i++)
+		trace_amdgpu_cs(parser, i);
+}
+
 int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	union drm_amdgpu_cs *cs = data;
 	struct amdgpu_cs_parser parser = {};
 	bool reserved_buffers = false;
-	int i, r;
+	int r;
 
 	if (amdgpu_ras_intr_triggered())
 		return -EHWPOISON;
@@ -1294,7 +1305,8 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 	r = amdgpu_cs_parser_init(&parser, data);
 	if (r) {
-		DRM_ERROR("Failed to initialize parser %d!\n", r);
+		if (printk_ratelimit())
+			DRM_ERROR("Failed to initialize parser %d!\n", r);
 		goto out;
 	}
 
@@ -1319,8 +1331,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 	reserved_buffers = true;
 
-	for (i = 0; i < parser.job->num_ibs; i++)
-		trace_amdgpu_cs(&parser, i);
+	trace_amdgpu_cs_ibs(&parser);
 
 	r = amdgpu_cs_vm_handling(&parser);
 	if (r)
@@ -1421,7 +1432,7 @@ static struct dma_fence *amdgpu_cs_get_fence(struct amdgpu_device *adev,
 int amdgpu_cs_fence_to_handle_ioctl(struct drm_device *dev, void *data,
 				    struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	union drm_amdgpu_fence_to_handle *info = data;
 	struct dma_fence *fence;
 	struct drm_syncobj *syncobj;
@@ -1597,7 +1608,7 @@ err_free_fence_array:
 int amdgpu_cs_wait_fences_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	union drm_amdgpu_wait_fences *wait = data;
 	uint32_t fence_count = wait->in.fence_count;
 	struct drm_amdgpu_fence *fences_user;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 8842c55d4490..c80d8339f58c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -46,7 +46,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = {
 static int amdgpu_ctx_priority_permit(struct drm_file *filp,
 				      enum drm_sched_priority priority)
 {
-	if (priority < 0 || priority >= DRM_SCHED_PRIORITY_MAX)
+	if (priority < 0 || priority >= DRM_SCHED_PRIORITY_COUNT)
 		return -EINVAL;
 
 	/* NORMAL and below are accessible by everyone */
@@ -65,7 +65,7 @@ static int amdgpu_ctx_priority_permit(struct drm_file *filp,
 static enum gfx_pipe_priority amdgpu_ctx_sched_prio_to_compute_prio(enum drm_sched_priority prio)
 {
 	switch (prio) {
-	case DRM_SCHED_PRIORITY_HIGH_HW:
+	case DRM_SCHED_PRIORITY_HIGH:
 	case DRM_SCHED_PRIORITY_KERNEL:
 		return AMDGPU_GFX_PIPE_PRIO_HIGH;
 	default:
@@ -114,7 +114,11 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 	scheds = adev->gpu_sched[hw_ip][hw_prio].sched;
 	num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds;
 
-	if (hw_ip == AMDGPU_HW_IP_VCN_ENC || hw_ip == AMDGPU_HW_IP_VCN_DEC) {
+	/* disable load balance if the hw engine retains context among dependent jobs */
+	if (hw_ip == AMDGPU_HW_IP_VCN_ENC ||
+	    hw_ip == AMDGPU_HW_IP_VCN_DEC ||
+	    hw_ip == AMDGPU_HW_IP_UVD_ENC ||
+	    hw_ip == AMDGPU_HW_IP_UVD) {
 		sched = drm_sched_pick_best(scheds, num_scheds);
 		scheds = &sched;
 		num_scheds = 1;
@@ -385,16 +389,15 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
 	enum drm_sched_priority priority;
 
 	union drm_amdgpu_ctx *args = data;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 
-	r = 0;
 	id = args->in.ctx_id;
-	priority = amdgpu_to_sched_priority(args->in.priority);
+	r = amdgpu_to_sched_priority(args->in.priority, &priority);
 
 	/* For backwards compatibility reasons, we need to accept
 	 * ioctls with garbage in the priority field */
-	if (priority == DRM_SCHED_PRIORITY_INVALID)
+	if (r == -EINVAL)
 		priority = DRM_SCHED_PRIORITY_NORMAL;
 
 	switch (args->in.op) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 193ffdb957b6..2d125b8b15ee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -34,6 +34,7 @@
 #include "amdgpu_pm.h"
 #include "amdgpu_dm_debugfs.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_rap.h"
 
 /**
  * amdgpu_debugfs_add_files - Add simple debugfs entries
@@ -68,8 +69,8 @@ int amdgpu_debugfs_add_files(struct amdgpu_device *adev,
 	adev->debugfs_count = i;
 #if defined(CONFIG_DEBUG_FS)
 	drm_debugfs_create_files(files, nfiles,
-				 adev->ddev->primary->debugfs_root,
-				 adev->ddev->primary);
+				 adev_to_drm(adev)->primary->debugfs_root,
+				 adev_to_drm(adev)->primary);
 #endif
 	return 0;
 }
@@ -100,14 +101,18 @@ static int amdgpu_debugfs_autodump_open(struct inode *inode, struct file *file)
 
 	file->private_data = adev;
 
-	mutex_lock(&adev->lock_reset);
+	ret = down_read_killable(&adev->reset_sem);
+	if (ret)
+		return ret;
+
 	if (adev->autodump.dumping.done) {
 		reinit_completion(&adev->autodump.dumping);
 		ret = 0;
 	} else {
 		ret = -EBUSY;
 	}
-	mutex_unlock(&adev->lock_reset);
+
+	up_read(&adev->reset_sem);
 
 	return ret;
 }
@@ -126,7 +131,7 @@ static unsigned int amdgpu_debugfs_autodump_poll(struct file *file, struct poll_
 
 	poll_wait(file, &adev->autodump.gpu_hang, poll_table);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return POLLIN | POLLRDNORM | POLLWRNORM;
 
 	return 0;
@@ -146,7 +151,7 @@ static void amdgpu_debugfs_autodump_init(struct amdgpu_device *adev)
 	init_waitqueue_head(&adev->autodump.gpu_hang);
 
 	debugfs_create_file("amdgpu_autodump", 0600,
-		adev->ddev->primary->debugfs_root,
+		adev_to_drm(adev)->primary->debugfs_root,
 		adev, &autodump_debug_fops);
 }
 
@@ -222,23 +227,23 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 
 	*pos &= (1UL << 22) - 1;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	if (use_bank) {
 		if ((sh_bank != 0xFFFFFFFF && sh_bank >= adev->gfx.config.max_sh_per_se) ||
 		    (se_bank != 0xFFFFFFFF && se_bank >= adev->gfx.config.max_shader_engines)) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return -EINVAL;
 		}
@@ -262,7 +267,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 		} else {
 			r = get_user(value, (uint32_t *)buf);
 			if (!r)
-				amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value, 0);
+				amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value);
 		}
 		if (r) {
 			result = r;
@@ -287,8 +292,8 @@ end:
 	if (pm_pg_lock)
 		mutex_unlock(&adev->pm.mutex);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -335,15 +340,15 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf,
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -353,8 +358,8 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf,
 		value = RREG32_PCIE(*pos >> 2);
 		r = put_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -365,8 +370,8 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf,
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -394,15 +399,15 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -411,8 +416,8 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user
 
 		r = get_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -425,8 +430,8 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -454,15 +459,15 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf,
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -472,8 +477,8 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf,
 		value = RREG32_DIDT(*pos >> 2);
 		r = put_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -484,8 +489,8 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf,
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -513,15 +518,15 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -530,8 +535,8 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user
 
 		r = get_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -544,8 +549,8 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -573,15 +578,15 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf,
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -591,8 +596,8 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf,
 		value = RREG32_SMC(*pos);
 		r = put_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -603,8 +608,8 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf,
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -632,15 +637,15 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -649,8 +654,8 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *
 
 		r = get_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			amdgpu_virt_disable_access_debugfs(adev);
 			return r;
 		}
@@ -663,8 +668,8 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	amdgpu_virt_disable_access_debugfs(adev);
 	return result;
@@ -791,22 +796,22 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf,
 
 	valuesize = sizeof(values);
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_dpm_read_sensor(adev, idx, &values[0], &valuesize);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r) {
 		amdgpu_virt_disable_access_debugfs(adev);
@@ -873,15 +878,15 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf,
 	wave = (*pos & GENMASK_ULL(36, 31)) >> 31;
 	simd = (*pos & GENMASK_ULL(44, 37)) >> 37;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_virt_enable_access_debugfs(adev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -896,8 +901,8 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf,
 	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (!x) {
 		amdgpu_virt_disable_access_debugfs(adev);
@@ -971,7 +976,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 	if (!data)
 		return -ENOMEM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0)
 		goto err;
 
@@ -994,8 +999,8 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	while (size) {
 		uint32_t value;
@@ -1017,7 +1022,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 	return result;
 
 err:
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 	kfree(data);
 	return r;
 }
@@ -1042,9 +1047,9 @@ static ssize_t amdgpu_debugfs_gfxoff_write(struct file *f, const char __user *bu
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -1053,8 +1058,8 @@ static ssize_t amdgpu_debugfs_gfxoff_write(struct file *f, const char __user *bu
 
 		r = get_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return r;
 		}
 
@@ -1066,8 +1071,8 @@ static ssize_t amdgpu_debugfs_gfxoff_write(struct file *f, const char __user *bu
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return result;
 }
@@ -1091,7 +1096,7 @@ static ssize_t amdgpu_debugfs_gfxoff_read(struct file *f, char __user *buf,
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0)
 		return r;
 
@@ -1100,15 +1105,15 @@ static ssize_t amdgpu_debugfs_gfxoff_read(struct file *f, char __user *buf,
 
 		r = amdgpu_get_gfx_off_status(adev, &value);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return r;
 		}
 
 		r = put_user(value, (uint32_t *)buf);
 		if (r) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return r;
 		}
 
@@ -1118,8 +1123,8 @@ static ssize_t amdgpu_debugfs_gfxoff_read(struct file *f, char __user *buf,
 		size -= 4;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return result;
 }
@@ -1211,7 +1216,7 @@ static const char *debugfs_regs_names[] = {
  */
 int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
 {
-	struct drm_minor *minor = adev->ddev->primary;
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
 	struct dentry *ent, *root = minor->debugfs_root;
 	unsigned int i;
 
@@ -1231,17 +1236,19 @@ static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r = 0, i;
 
 	r = pm_runtime_get_sync(dev->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	/* Avoid accidently unparking the sched thread during GPU reset */
-	mutex_lock(&adev->lock_reset);
+	r = down_read_killable(&adev->reset_sem);
+	if (r)
+		return r;
 
 	/* hold on the scheduler */
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
@@ -1268,7 +1275,7 @@ static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data)
 		kthread_unpark(ring->sched.thread);
 	}
 
-	mutex_unlock(&adev->lock_reset);
+	up_read(&adev->reset_sem);
 
 	pm_runtime_mark_last_busy(dev->dev);
 	pm_runtime_put_autosuspend(dev->dev);
@@ -1280,7 +1287,7 @@ static int amdgpu_debugfs_get_vbios_dump(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	seq_write(m, adev->bios, adev->bios_size);
 	return 0;
@@ -1290,12 +1297,12 @@ static int amdgpu_debugfs_evict_vram(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r;
 
 	r = pm_runtime_get_sync(dev->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -1311,12 +1318,12 @@ static int amdgpu_debugfs_evict_gtt(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r;
 
 	r = pm_runtime_get_sync(dev->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -1458,7 +1465,9 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
 		return -ENOMEM;
 
 	/* Avoid accidently unparking the sched thread during GPU reset */
-	mutex_lock(&adev->lock_reset);
+	r = down_read_killable(&adev->reset_sem);
+	if (r)
+		goto pro_end;
 
 	/* stop the scheduler */
 	kthread_park(ring->sched.thread);
@@ -1499,13 +1508,14 @@ failure:
 	/* restart the scheduler */
 	kthread_unpark(ring->sched.thread);
 
-	mutex_unlock(&adev->lock_reset);
+	up_read(&adev->reset_sem);
 
 	ttm_bo_unlock_delayed_workqueue(&adev->mman.bdev, resched);
 
+pro_end:
 	kfree(fences);
 
-	return 0;
+	return r;
 }
 
 static int amdgpu_debugfs_sclk_set(void *data, u64 val)
@@ -1517,9 +1527,9 @@ static int amdgpu_debugfs_sclk_set(void *data, u64 val)
 	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
 		return -EINVAL;
 
-	ret = pm_runtime_get_sync(adev->ddev->dev);
+	ret = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (ret < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return ret;
 	}
 
@@ -1532,8 +1542,8 @@ static int amdgpu_debugfs_sclk_set(void *data, u64 val)
 		return 0;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (ret)
 		return -EINVAL;
@@ -1553,7 +1563,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 
 	adev->debugfs_preempt =
 		debugfs_create_file("amdgpu_preempt_ib", 0600,
-				    adev->ddev->primary->debugfs_root, adev,
+				    adev_to_drm(adev)->primary->debugfs_root, adev,
 				    &fops_ib_preempt);
 	if (!(adev->debugfs_preempt)) {
 		DRM_ERROR("unable to create amdgpu_preempt_ib debugsfs file\n");
@@ -1562,7 +1572,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 
 	adev->smu.debugfs_sclk =
 		debugfs_create_file("amdgpu_force_sclk", 0200,
-				    adev->ddev->primary->debugfs_root, adev,
+				    adev_to_drm(adev)->primary->debugfs_root, adev,
 				    &fops_sclk_set);
 	if (!(adev->smu.debugfs_sclk)) {
 		DRM_ERROR("unable to create amdgpu_set_sclk debugsfs file\n");
@@ -1623,6 +1633,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 
 	amdgpu_debugfs_autodump_init(adev);
 
+	amdgpu_rap_debugfs_init(adev);
+
 	return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list,
 					ARRAY_SIZE(amdgpu_debugfs_list));
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index d0b8d0d341af..e8b41756c9f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -130,7 +130,7 @@ static ssize_t amdgpu_device_get_pcie_replay_count(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint64_t cnt = amdgpu_asic_get_pcie_replay_count(adev);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n", cnt);
@@ -155,7 +155,7 @@ static ssize_t amdgpu_device_get_product_name(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%s\n", adev->product_name);
 }
@@ -177,7 +177,7 @@ static ssize_t amdgpu_device_get_product_number(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%s\n", adev->product_number);
 }
@@ -199,7 +199,7 @@ static ssize_t amdgpu_device_get_serial_number(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%s\n", adev->serial);
 }
@@ -217,7 +217,7 @@ static DEVICE_ATTR(serial_number, S_IRUGO,
  */
 bool amdgpu_device_supports_boco(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (adev->flags & AMD_IS_PX)
 		return true;
@@ -234,7 +234,7 @@ bool amdgpu_device_supports_boco(struct drm_device *dev)
  */
 bool amdgpu_device_supports_baco(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	return amdgpu_asic_supports_baco(adev);
 }
@@ -301,10 +301,10 @@ void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
 }
 
 /*
- * MMIO register access helper functions.
+ * register access helper functions.
  */
 /**
- * amdgpu_mm_rreg - read a memory mapped IO register
+ * amdgpu_device_rreg - read a memory mapped IO or indirect register
  *
  * @adev: amdgpu_device pointer
  * @reg: dword aligned register offset
@@ -312,25 +312,29 @@ void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
  *
  * Returns the 32 bit value from the offset specified.
  */
-uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg,
-			uint32_t acc_flags)
+uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
+			    uint32_t reg, uint32_t acc_flags)
 {
 	uint32_t ret;
 
-	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))
-		return amdgpu_kiq_rreg(adev, reg);
-
-	if ((reg * 4) < adev->rmmio_size)
-		ret = readl(((void __iomem *)adev->rmmio) + (reg * 4));
-	else {
-		unsigned long flags;
+	if (adev->in_pci_err_recovery)
+		return 0;
 
-		spin_lock_irqsave(&adev->mmio_idx_lock, flags);
-		writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4));
-		ret = readl(((void __iomem *)adev->rmmio) + (mmMM_DATA * 4));
-		spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
+	if ((reg * 4) < adev->rmmio_size) {
+		if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
+		    amdgpu_sriov_runtime(adev) &&
+		    down_read_trylock(&adev->reset_sem)) {
+			ret = amdgpu_kiq_rreg(adev, reg);
+			up_read(&adev->reset_sem);
+		} else {
+			ret = readl(((void __iomem *)adev->rmmio) + (reg * 4));
+		}
+	} else {
+		ret = adev->pcie_rreg(adev, reg * 4);
 	}
-	trace_amdgpu_mm_rreg(adev->pdev->device, reg, ret);
+
+	trace_amdgpu_device_rreg(adev->pdev->device, reg, ret);
+
 	return ret;
 }
 
@@ -348,7 +352,11 @@ uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg,
  *
  * Returns the 8 bit value from the offset specified.
  */
-uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset) {
+uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset)
+{
+	if (adev->in_pci_err_recovery)
+		return 0;
+
 	if (offset < adev->rmmio_size)
 		return (readb(adev->rmmio + offset));
 	BUG();
@@ -369,31 +377,19 @@ uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset) {
  *
  * Writes the value specified to the offset specified.
  */
-void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) {
+void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value)
+{
+	if (adev->in_pci_err_recovery)
+		return;
+
 	if (offset < adev->rmmio_size)
 		writeb(value, adev->rmmio + offset);
 	else
 		BUG();
 }
 
-void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags)
-{
-	trace_amdgpu_mm_wreg(adev->pdev->device, reg, v);
-
-	if ((reg * 4) < adev->rmmio_size)
-		writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));
-	else {
-		unsigned long flags;
-
-		spin_lock_irqsave(&adev->mmio_idx_lock, flags);
-		writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4));
-		writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4));
-		spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
-	}
-}
-
 /**
- * amdgpu_mm_wreg - write to a memory mapped IO register
+ * amdgpu_device_wreg - write to a memory mapped IO or indirect register
  *
  * @adev: amdgpu_device pointer
  * @reg: dword aligned register offset
@@ -402,13 +398,27 @@ void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg,
  *
  * Writes the value specified to the offset specified.
  */
-void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
-		    uint32_t acc_flags)
+void amdgpu_device_wreg(struct amdgpu_device *adev,
+			uint32_t reg, uint32_t v,
+			uint32_t acc_flags)
 {
-	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))
-		return amdgpu_kiq_wreg(adev, reg, v);
+	if (adev->in_pci_err_recovery)
+		return;
+
+	if ((reg * 4) < adev->rmmio_size) {
+		if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
+		    amdgpu_sriov_runtime(adev) &&
+		    down_read_trylock(&adev->reset_sem)) {
+			amdgpu_kiq_wreg(adev, reg, v);
+			up_read(&adev->reset_sem);
+		} else {
+			writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));
+		}
+	} else {
+		adev->pcie_wreg(adev, reg * 4, v);
+	}
 
-	amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags);
+	trace_amdgpu_device_wreg(adev->pdev->device, reg, v);
 }
 
 /*
@@ -416,18 +426,20 @@ void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
  *
  * this function is invoked only the debugfs register access
  * */
-void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
-		    uint32_t acc_flags)
+void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev,
+			     uint32_t reg, uint32_t v)
 {
-	if (amdgpu_sriov_fullaccess(adev) &&
-		adev->gfx.rlc.funcs &&
-		adev->gfx.rlc.funcs->is_rlcg_access_range) {
+	if (adev->in_pci_err_recovery)
+		return;
 
+	if (amdgpu_sriov_fullaccess(adev) &&
+	    adev->gfx.rlc.funcs &&
+	    adev->gfx.rlc.funcs->is_rlcg_access_range) {
 		if (adev->gfx.rlc.funcs->is_rlcg_access_range(adev, reg))
 			return adev->gfx.rlc.funcs->rlcg_wreg(adev, reg, v);
+	} else {
+		writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));
 	}
-
-	amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags);
 }
 
 /**
@@ -440,6 +452,9 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t
  */
 u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg)
 {
+	if (adev->in_pci_err_recovery)
+		return 0;
+
 	if ((reg * 4) < adev->rio_mem_size)
 		return ioread32(adev->rio_mem + (reg * 4));
 	else {
@@ -459,6 +474,9 @@ u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg)
  */
 void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
 {
+	if (adev->in_pci_err_recovery)
+		return;
+
 	if ((reg * 4) < adev->rio_mem_size)
 		iowrite32(v, adev->rio_mem + (reg * 4));
 	else {
@@ -478,6 +496,9 @@ void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
  */
 u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index)
 {
+	if (adev->in_pci_err_recovery)
+		return 0;
+
 	if (index < adev->doorbell.num_doorbells) {
 		return readl(adev->doorbell.ptr + index);
 	} else {
@@ -498,6 +519,9 @@ u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index)
  */
 void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v)
 {
+	if (adev->in_pci_err_recovery)
+		return;
+
 	if (index < adev->doorbell.num_doorbells) {
 		writel(v, adev->doorbell.ptr + index);
 	} else {
@@ -516,6 +540,9 @@ void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v)
  */
 u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index)
 {
+	if (adev->in_pci_err_recovery)
+		return 0;
+
 	if (index < adev->doorbell.num_doorbells) {
 		return atomic64_read((atomic64_t *)(adev->doorbell.ptr + index));
 	} else {
@@ -536,6 +563,9 @@ u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index)
  */
 void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v)
 {
+	if (adev->in_pci_err_recovery)
+		return;
+
 	if (index < adev->doorbell.num_doorbells) {
 		atomic64_set((atomic64_t *)(adev->doorbell.ptr + index), v);
 	} else {
@@ -544,6 +574,135 @@ void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v)
 }
 
 /**
+ * amdgpu_device_indirect_rreg - read an indirect register
+ *
+ * @adev: amdgpu_device pointer
+ * @pcie_index: mmio register offset
+ * @pcie_data: mmio register offset
+ *
+ * Returns the value of indirect register @reg_addr
+ */
+u32 amdgpu_device_indirect_rreg(struct amdgpu_device *adev,
+				u32 pcie_index, u32 pcie_data,
+				u32 reg_addr)
+{
+	unsigned long flags;
+	u32 r;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_data_offset;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	r = readl(pcie_data_offset);
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+
+	return r;
+}
+
+/**
+ * amdgpu_device_indirect_rreg64 - read a 64bits indirect register
+ *
+ * @adev: amdgpu_device pointer
+ * @pcie_index: mmio register offset
+ * @pcie_data: mmio register offset
+ *
+ * Returns the value of indirect register @reg_addr
+ */
+u64 amdgpu_device_indirect_rreg64(struct amdgpu_device *adev,
+				  u32 pcie_index, u32 pcie_data,
+				  u32 reg_addr)
+{
+	unsigned long flags;
+	u64 r;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_data_offset;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+
+	/* read low 32 bits */
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	r = readl(pcie_data_offset);
+	/* read high 32 bits */
+	writel(reg_addr + 4, pcie_index_offset);
+	readl(pcie_index_offset);
+	r |= ((u64)readl(pcie_data_offset) << 32);
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+
+	return r;
+}
+
+/**
+ * amdgpu_device_indirect_wreg - write an indirect register address
+ *
+ * @adev: amdgpu_device pointer
+ * @pcie_index: mmio register offset
+ * @pcie_data: mmio register offset
+ * @reg_addr: indirect register offset
+ * @reg_data: indirect register data
+ *
+ */
+void amdgpu_device_indirect_wreg(struct amdgpu_device *adev,
+				 u32 pcie_index, u32 pcie_data,
+				 u32 reg_addr, u32 reg_data)
+{
+	unsigned long flags;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_data_offset;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	writel(reg_data, pcie_data_offset);
+	readl(pcie_data_offset);
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+}
+
+/**
+ * amdgpu_device_indirect_wreg64 - write a 64bits indirect register address
+ *
+ * @adev: amdgpu_device pointer
+ * @pcie_index: mmio register offset
+ * @pcie_data: mmio register offset
+ * @reg_addr: indirect register offset
+ * @reg_data: indirect register data
+ *
+ */
+void amdgpu_device_indirect_wreg64(struct amdgpu_device *adev,
+				   u32 pcie_index, u32 pcie_data,
+				   u32 reg_addr, u64 reg_data)
+{
+	unsigned long flags;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_data_offset;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+
+	/* write low 32 bits */
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	writel((u32)(reg_data & 0xffffffffULL), pcie_data_offset);
+	readl(pcie_data_offset);
+	/* write high 32 bits */
+	writel(reg_addr + 4, pcie_index_offset);
+	readl(pcie_index_offset);
+	writel((u32)(reg_data >> 32), pcie_data_offset);
+	readl(pcie_data_offset);
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+}
+
+/**
  * amdgpu_invalid_rreg - dummy reg read function
  *
  * @adev: amdgpu device pointer
@@ -652,6 +811,20 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev,
 }
 
 /**
+ * amdgpu_device_asic_init - Wrapper for atom asic_init
+ *
+ * @dev: drm_device pointer
+ *
+ * Does any asic specific work and then calls atom asic init.
+ */
+static int amdgpu_device_asic_init(struct amdgpu_device *adev)
+{
+	amdgpu_asic_pre_asic_init(adev);
+
+	return amdgpu_atom_asic_init(adev->mode_info.atom_context);
+}
+
+/**
  * amdgpu_device_vram_scratch_init - allocate the VRAM scratch page
  *
  * @adev: amdgpu device pointer
@@ -1197,6 +1370,15 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 
 	amdgpu_gmc_tmz_set(adev);
 
+	if (amdgpu_num_kcq == -1) {
+		amdgpu_num_kcq = 8;
+	} else if (amdgpu_num_kcq > 8 || amdgpu_num_kcq < 0) {
+		amdgpu_num_kcq = 8;
+		dev_warn(adev->dev, "set kernel compute queue number to 8 due to invalid parameter provided by user\n");
+	}
+
+	amdgpu_gmc_noretry_set(adev);
+
 	return 0;
 }
 
@@ -1209,7 +1391,8 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
  * Callback for the switcheroo driver.  Suspends or resumes the
  * the asics before or after it is powered up using ACPI methods.
  */
-static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_state state)
+static void amdgpu_switcheroo_set_state(struct pci_dev *pdev,
+					enum vga_switcheroo_state state)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	int r;
@@ -1223,7 +1406,7 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
 
 		pci_set_power_state(dev->pdev, PCI_D0);
-		pci_restore_state(dev->pdev);
+		amdgpu_device_load_pci_state(dev->pdev);
 		r = pci_enable_device(dev->pdev);
 		if (r)
 			DRM_WARN("pci_enable_device failed (%d)\n", r);
@@ -1236,7 +1419,7 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 		drm_kms_helper_poll_disable(dev);
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
 		amdgpu_device_suspend(dev, true);
-		pci_save_state(dev->pdev);
+		amdgpu_device_cache_pci_state(dev->pdev);
 		/* Shut down the device */
 		pci_disable_device(dev->pdev);
 		pci_set_power_state(dev->pdev, PCI_D3cold);
@@ -1502,7 +1685,7 @@ static void amdgpu_device_enable_virtual_display(struct amdgpu_device *adev)
 	adev->enable_virtual_display = false;
 
 	if (amdgpu_virtual_display) {
-		struct drm_device *ddev = adev->ddev;
+		struct drm_device *ddev = adev_to_drm(adev);
 		const char *pci_address_name = pci_name(ddev->pdev);
 		char *pciaddstr, *pciaddstr_tmp, *pciaddname_tmp, *pciaddname;
 
@@ -1561,7 +1744,7 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev)
 
 	adev->firmware.gpu_info_fw = NULL;
 
-	if (adev->discovery_bin) {
+	if (adev->mman.discovery_bin) {
 		amdgpu_discovery_get_gfx_info(adev);
 
 		/*
@@ -1929,7 +2112,7 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev)
 			if (adev->ip_blocks[i].status.hw == true)
 				break;
 
-			if (adev->in_gpu_reset || adev->in_suspend) {
+			if (amdgpu_in_reset(adev) || adev->in_suspend) {
 				r = adev->ip_blocks[i].version->funcs->resume(adev);
 				if (r) {
 					DRM_ERROR("resume of IP block <%s> failed %d\n",
@@ -2049,13 +2232,19 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 	 * it should be called after amdgpu_device_ip_hw_init_phase2  since
 	 * for some ASICs the RAS EEPROM code relies on SMU fully functioning
 	 * for I2C communication which only true at this point.
-	 * recovery_init may fail, but it can free all resources allocated by
-	 * itself and its failure should not stop amdgpu init process.
+	 *
+	 * amdgpu_ras_recovery_init may fail, but the upper only cares the
+	 * failure from bad gpu situation and stop amdgpu init process
+	 * accordingly. For other failed cases, it will still release all
+	 * the resource and print error message, rather than returning one
+	 * negative value to upper level.
 	 *
 	 * Note: theoretically, this should be called before all vram allocations
 	 * to protect retired page from abusing
 	 */
-	amdgpu_ras_recovery_init(adev);
+	r = amdgpu_ras_recovery_init(adev);
+	if (r)
+		goto init_failed;
 
 	if (adev->gmc.xgmi.num_physical_nodes > 1)
 		amdgpu_xgmi_add_device(adev);
@@ -2100,7 +2289,7 @@ static bool amdgpu_device_check_vram_lost(struct amdgpu_device *adev)
 			AMDGPU_RESET_MAGIC_NUM))
 		return true;
 
-	if (!adev->in_gpu_reset)
+	if (!amdgpu_in_reset(adev))
 		return false;
 
 	/*
@@ -2211,9 +2400,7 @@ static int amdgpu_device_enable_mgpu_fan_boost(void)
 		gpu_ins = &(mgpu_info.gpu_ins[i]);
 		adev = gpu_ins->adev;
 		if (!(adev->flags & AMD_IS_APU) &&
-		    !gpu_ins->mgpu_fan_enabled &&
-		    adev->powerplay.pp_funcs &&
-		    adev->powerplay.pp_funcs->enable_mgpu_fan_boost) {
+		    !gpu_ins->mgpu_fan_enabled) {
 			ret = amdgpu_dpm_enable_mgpu_fan_boost(adev);
 			if (ret)
 				break;
@@ -2568,17 +2755,16 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
 		AMD_IP_BLOCK_TYPE_IH,
 	};
 
-	for (i = 0; i < adev->num_ip_blocks; i++)
-		adev->ip_blocks[i].status.hw = false;
-
 	for (i = 0; i < ARRAY_SIZE(ip_order); i++) {
 		int j;
 		struct amdgpu_ip_block *block;
 
-		for (j = 0; j < adev->num_ip_blocks; j++) {
-			block = &adev->ip_blocks[j];
+		block = &adev->ip_blocks[i];
+		block->status.hw = false;
 
-			if (block->version->type != ip_order[i] ||
+		for (j = 0; j < ARRAY_SIZE(ip_order); j++) {
+
+			if (block->version->type != ip_order[j] ||
 				!block->status.valid)
 				continue;
 
@@ -2771,6 +2957,12 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type)
 {
 	switch (asic_type) {
 #if defined(CONFIG_DRM_AMD_DC)
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+#endif
 	case CHIP_BONAIRE:
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
@@ -2825,7 +3017,7 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type)
  */
 bool amdgpu_device_has_dc_support(struct amdgpu_device *adev)
 {
-	if (amdgpu_sriov_vf(adev))
+	if (amdgpu_sriov_vf(adev) || adev->enable_virtual_display)
 		return false;
 
 	return amdgpu_device_asic_has_dc_support(adev->asic_type);
@@ -2836,7 +3028,7 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 {
 	struct amdgpu_device *adev =
 		container_of(__work, struct amdgpu_device, xgmi_reset_work);
-	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, 0);
+	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
 
 	/* It's a bug to not have a hive within this function */
 	if (WARN_ON(!hive))
@@ -2851,13 +3043,13 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 	if (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
 
 		task_barrier_enter(&hive->tb);
-		adev->asic_reset_res = amdgpu_device_baco_enter(adev->ddev);
+		adev->asic_reset_res = amdgpu_device_baco_enter(adev_to_drm(adev));
 
 		if (adev->asic_reset_res)
 			goto fail;
 
 		task_barrier_exit(&hive->tb);
-		adev->asic_reset_res = amdgpu_device_baco_exit(adev->ddev);
+		adev->asic_reset_res = amdgpu_device_baco_exit(adev_to_drm(adev));
 
 		if (adev->asic_reset_res)
 			goto fail;
@@ -2873,7 +3065,8 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 fail:
 	if (adev->asic_reset_res)
 		DRM_WARN("ASIC reset failed with error, %d for drm dev, %s",
-			 adev->asic_reset_res, adev->ddev->unique);
+			 adev->asic_reset_res, adev_to_drm(adev)->unique);
+	amdgpu_put_xgmi_hive(hive);
 }
 
 static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
@@ -2952,12 +3145,11 @@ static const struct attribute *amdgpu_dev_attributes[] = {
 	NULL
 };
 
+
 /**
  * amdgpu_device_init - initialize the driver
  *
  * @adev: amdgpu_device pointer
- * @ddev: drm dev pointer
- * @pdev: pci dev pointer
  * @flags: driver flags
  *
  * Initializes the driver info and hw (all asics).
@@ -2965,18 +3157,15 @@ static const struct attribute *amdgpu_dev_attributes[] = {
  * Called at driver startup.
  */
 int amdgpu_device_init(struct amdgpu_device *adev,
-		       struct drm_device *ddev,
-		       struct pci_dev *pdev,
 		       uint32_t flags)
 {
+	struct drm_device *ddev = adev_to_drm(adev);
+	struct pci_dev *pdev = adev->pdev;
 	int r, i;
 	bool boco = false;
 	u32 max_MBps;
 
 	adev->shutdown = false;
-	adev->dev = &pdev->dev;
-	adev->ddev = ddev;
-	adev->pdev = pdev;
 	adev->flags = flags;
 
 	if (amdgpu_force_asic_type >= 0 && amdgpu_force_asic_type < CHIP_LAST)
@@ -3032,7 +3221,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->mn_lock);
 	mutex_init(&adev->virt.vf_errors.lock);
 	hash_init(adev->mn_hash);
-	mutex_init(&adev->lock_reset);
+	atomic_set(&adev->in_gpu_reset, 0);
+	init_rwsem(&adev->reset_sem);
 	mutex_init(&adev->psp.mutex);
 	mutex_init(&adev->notifier_lock);
 
@@ -3127,13 +3317,13 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	r = amdgpu_device_get_job_timeout_settings(adev);
 	if (r) {
 		dev_err(adev->dev, "invalid lockup_timeout parameter syntax\n");
-		return r;
+		goto failed_unmap;
 	}
 
 	/* early init functions */
 	r = amdgpu_device_ip_early_init(adev);
 	if (r)
-		return r;
+		goto failed_unmap;
 
 	/* doorbell bar mapping and doorbell index init*/
 	amdgpu_device_doorbell_init(adev);
@@ -3174,6 +3364,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		}
 	}
 
+	pci_enable_pcie_error_reporting(adev->ddev.pdev);
+
 	/* Post card if necessary */
 	if (amdgpu_device_need_post(adev)) {
 		if (!adev->bios) {
@@ -3182,7 +3374,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 			goto failed;
 		}
 		DRM_INFO("GPU posting now...\n");
-		r = amdgpu_atom_asic_init(adev->mode_info.atom_context);
+		r = amdgpu_device_asic_init(adev);
 		if (r) {
 			dev_err(adev->dev, "gpu post error!\n");
 			goto failed;
@@ -3220,7 +3412,7 @@ fence_driver_init:
 	}
 
 	/* init the mode config */
-	drm_mode_config_init(adev->ddev);
+	drm_mode_config_init(adev_to_drm(adev));
 
 	r = amdgpu_device_ip_init(adev);
 	if (r) {
@@ -3316,16 +3508,18 @@ fence_driver_init:
 		flush_delayed_work(&adev->delayed_init_work);
 
 	r = sysfs_create_files(&adev->dev->kobj, amdgpu_dev_attributes);
-	if (r) {
+	if (r)
 		dev_err(adev->dev, "Could not create amdgpu device attr\n");
-		return r;
-	}
 
 	if (IS_ENABLED(CONFIG_PERF_EVENTS))
 		r = amdgpu_pmu_init(adev);
 	if (r)
 		dev_err(adev->dev, "amdgpu_pmu_init failed\n");
 
+	/* Have stored pci confspace at hand for restore in sudden PCI error */
+	if (amdgpu_device_cache_pci_state(adev->pdev))
+		pci_restore_state(pdev);
+
 	return 0;
 
 failed:
@@ -3333,6 +3527,10 @@ failed:
 	if (boco)
 		vga_switcheroo_fini_domain_pm_ops(adev->dev);
 
+failed_unmap:
+	iounmap(adev->rmmio);
+	adev->rmmio = NULL;
+
 	return r;
 }
 
@@ -3346,31 +3544,33 @@ failed:
  */
 void amdgpu_device_fini(struct amdgpu_device *adev)
 {
-	int r;
-
-	DRM_INFO("amdgpu: finishing device.\n");
+	dev_info(adev->dev, "amdgpu: finishing device.\n");
 	flush_delayed_work(&adev->delayed_init_work);
 	adev->shutdown = true;
 
+	kfree(adev->pci_state);
+
 	/* make sure IB test finished before entering exclusive mode
 	 * to avoid preemption on IB test
 	 * */
-	if (amdgpu_sriov_vf(adev))
+	if (amdgpu_sriov_vf(adev)) {
 		amdgpu_virt_request_full_gpu(adev, false);
+		amdgpu_virt_fini_data_exchange(adev);
+	}
 
 	/* disable all interrupts */
 	amdgpu_irq_disable_all(adev);
 	if (adev->mode_info.mode_config_initialized){
 		if (!amdgpu_device_has_dc_support(adev))
-			drm_helper_force_disable_all(adev->ddev);
+			drm_helper_force_disable_all(adev_to_drm(adev));
 		else
-			drm_atomic_helper_shutdown(adev->ddev);
+			drm_atomic_helper_shutdown(adev_to_drm(adev));
 	}
 	amdgpu_fence_driver_fini(adev);
 	if (adev->pm_sysfs_en)
 		amdgpu_pm_sysfs_fini(adev);
 	amdgpu_fbdev_fini(adev);
-	r = amdgpu_device_ip_fini(adev);
+	amdgpu_device_ip_fini(adev);
 	release_firmware(adev->firmware.gpu_info_fw);
 	adev->firmware.gpu_info_fw = NULL;
 	adev->accel_working = false;
@@ -3388,7 +3588,7 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	     amdgpu_has_atpx_dgpu_power_cntl()) &&
 	    !pci_is_thunderbolt_attached(adev->pdev))
 		vga_switcheroo_unregister_client(adev->pdev);
-	if (amdgpu_device_supports_boco(adev->ddev))
+	if (amdgpu_device_supports_boco(adev_to_drm(adev)))
 		vga_switcheroo_fini_domain_pm_ops(adev->dev);
 	vga_client_register(adev->pdev, NULL, NULL, NULL);
 	if (adev->rio_mem)
@@ -3404,7 +3604,7 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	sysfs_remove_files(&adev->dev->kobj, amdgpu_dev_attributes);
 	if (IS_ENABLED(CONFIG_PERF_EVENTS))
 		amdgpu_pmu_fini(adev);
-	if (adev->discovery_bin)
+	if (adev->mman.discovery_bin)
 		amdgpu_discovery_fini(adev);
 }
 
@@ -3430,11 +3630,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 	struct drm_connector_list_iter iter;
 	int r;
 
-	if (dev == NULL || dev->dev_private == NULL) {
-		return -ENODEV;
-	}
-
-	adev = dev->dev_private;
+	adev = drm_to_adev(dev);
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
@@ -3522,7 +3718,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 {
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_crtc *crtc;
 	int r = 0;
 
@@ -3531,14 +3727,14 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 
 	/* post card */
 	if (amdgpu_device_need_post(adev)) {
-		r = amdgpu_atom_asic_init(adev->mode_info.atom_context);
+		r = amdgpu_device_asic_init(adev);
 		if (r)
-			DRM_ERROR("amdgpu asic init failed\n");
+			dev_err(adev->dev, "amdgpu asic init failed\n");
 	}
 
 	r = amdgpu_device_ip_resume(adev);
 	if (r) {
-		DRM_ERROR("amdgpu_device_ip_resume failed (%d).\n", r);
+		dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r);
 		return r;
 	}
 	amdgpu_fence_driver_resume(adev);
@@ -3562,7 +3758,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 				if (r == 0) {
 					r = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM);
 					if (r != 0)
-						DRM_ERROR("Failed to pin cursor BO (%d)\n", r);
+						dev_err(adev->dev, "Failed to pin cursor BO (%d)\n", r);
 					amdgpu_crtc->cursor_addr = amdgpu_bo_gpu_offset(aobj);
 					amdgpu_bo_unreserve(aobj);
 				}
@@ -3652,7 +3848,7 @@ static bool amdgpu_device_ip_check_soft_reset(struct amdgpu_device *adev)
 			adev->ip_blocks[i].status.hang =
 				adev->ip_blocks[i].version->funcs->check_soft_reset(adev);
 		if (adev->ip_blocks[i].status.hang) {
-			DRM_INFO("IP block:%s is hung!\n", adev->ip_blocks[i].version->funcs->name);
+			dev_info(adev->dev, "IP block:%s is hung!\n", adev->ip_blocks[i].version->funcs->name);
 			asic_hang = true;
 		}
 	}
@@ -3713,7 +3909,7 @@ static bool amdgpu_device_ip_need_full_reset(struct amdgpu_device *adev)
 		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE) ||
 		     adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP) {
 			if (adev->ip_blocks[i].status.hang) {
-				DRM_INFO("Some block need full reset!\n");
+				dev_info(adev->dev, "Some block need full reset!\n");
 				return true;
 			}
 		}
@@ -3801,7 +3997,7 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
 	else
 		tmo = msecs_to_jiffies(100);
 
-	DRM_INFO("recover vram bo from shadow start\n");
+	dev_info(adev->dev, "recover vram bo from shadow start\n");
 	mutex_lock(&adev->shadow_list_lock);
 	list_for_each_entry(shadow, &adev->shadow_list, shadow_list) {
 
@@ -3837,11 +4033,11 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
 	dma_fence_put(fence);
 
 	if (r < 0 || tmo <= 0) {
-		DRM_ERROR("recover vram bo from shadow failed, r is %ld, tmo is %ld\n", r, tmo);
+		dev_err(adev->dev, "recover vram bo from shadow failed, r is %ld, tmo is %ld\n", r, tmo);
 		return -EIO;
 	}
 
-	DRM_INFO("recover vram bo from shadow done\n");
+	dev_info(adev->dev, "recover vram bo from shadow done\n");
 	return 0;
 }
 
@@ -3876,7 +4072,7 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
 
 	amdgpu_virt_init_data_exchange(adev);
 	/* we need recover gart prior to run SMC/CP/SDMA resume */
-	amdgpu_gtt_mgr_recover(&adev->mman.bdev.man[TTM_PL_TT]);
+	amdgpu_gtt_mgr_recover(ttm_manager_type(&adev->mman.bdev, TTM_PL_TT));
 
 	r = amdgpu_device_fw_loading(adev);
 	if (r)
@@ -3902,6 +4098,34 @@ error:
 }
 
 /**
+ * amdgpu_device_has_job_running - check if there is any job in mirror list
+ *
+ * @adev: amdgpu device pointer
+ *
+ * check if there is any job in mirror list
+ */
+bool amdgpu_device_has_job_running(struct amdgpu_device *adev)
+{
+	int i;
+	struct drm_sched_job *job;
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (!ring || !ring->sched.thread)
+			continue;
+
+		spin_lock(&ring->sched.job_list_lock);
+		job = list_first_entry_or_null(&ring->sched.ring_mirror_list,
+				struct drm_sched_job, node);
+		spin_unlock(&ring->sched.job_list_lock);
+		if (job)
+			return true;
+	}
+	return false;
+}
+
+/**
  * amdgpu_device_should_recover_gpu - check if we should try GPU recovery
  *
  * @adev: amdgpu device pointer
@@ -3912,7 +4136,7 @@ error:
 bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev)
 {
 	if (!amdgpu_device_ip_check_soft_reset(adev)) {
-		DRM_INFO("Timeout, but no hardware hang detected.\n");
+		dev_info(adev->dev, "Timeout, but no hardware hang detected.\n");
 		return false;
 	}
 
@@ -3952,7 +4176,7 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev)
 	return true;
 
 disabled:
-		DRM_INFO("GPU recovery disabled.\n");
+		dev_info(adev->dev, "GPU recovery disabled.\n");
 		return false;
 }
 
@@ -3966,6 +4190,11 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 
 	amdgpu_debugfs_wait_dump(adev);
 
+	if (amdgpu_sriov_vf(adev)) {
+		/* stop the data exchange thread */
+		amdgpu_virt_fini_data_exchange(adev);
+	}
+
 	/* block all schedulers and reset given job's ring */
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
 		struct amdgpu_ring *ring = adev->rings[i];
@@ -3991,7 +4220,7 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 			r = amdgpu_device_ip_soft_reset(adev);
 			amdgpu_device_ip_post_soft_reset(adev);
 			if (r || amdgpu_device_ip_check_soft_reset(adev)) {
-				DRM_INFO("soft reset failed, will fallback to full reset!\n");
+				dev_info(adev->dev, "soft reset failed, will fallback to full reset!\n");
 				need_full_reset = true;
 			}
 		}
@@ -4007,7 +4236,8 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 
 static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 			       struct list_head *device_list_handle,
-			       bool *need_full_reset_arg)
+			       bool *need_full_reset_arg,
+			       bool skip_hw_reset)
 {
 	struct amdgpu_device *tmp_adev = NULL;
 	bool need_full_reset = *need_full_reset_arg, vram_lost = false;
@@ -4017,7 +4247,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 	 * ASIC reset has to be done on all HGMI hive nodes ASAP
 	 * to allow proper links negotiation in FW (within 1 sec)
 	 */
-	if (need_full_reset) {
+	if (!skip_hw_reset && need_full_reset) {
 		list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
 			/* For XGMI run all resets in parallel to speed up the process */
 			if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
@@ -4027,8 +4257,8 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 				r = amdgpu_asic_reset(tmp_adev);
 
 			if (r) {
-				DRM_ERROR("ASIC reset failed with error, %d for drm dev, %s",
-					 r, tmp_adev->ddev->unique);
+				dev_err(tmp_adev->dev, "ASIC reset failed with error, %d for drm dev, %s",
+					 r, adev_to_drm(tmp_adev)->unique);
 				break;
 			}
 		}
@@ -4060,8 +4290,8 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 	list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
 		if (need_full_reset) {
 			/* post card */
-			if (amdgpu_atom_asic_init(tmp_adev->mode_info.atom_context))
-				DRM_WARN("asic atom init failed!");
+			if (amdgpu_device_asic_init(tmp_adev))
+				dev_warn(tmp_adev->dev, "asic atom init failed!");
 
 			if (!r) {
 				dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n");
@@ -4075,8 +4305,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 					amdgpu_inc_vram_lost(tmp_adev);
 				}
 
-				r = amdgpu_gtt_mgr_recover(
-					&tmp_adev->mman.bdev.man[TTM_PL_TT]);
+				r = amdgpu_gtt_mgr_recover(ttm_manager_type(&tmp_adev->mman.bdev, TTM_PL_TT));
 				if (r)
 					goto out;
 
@@ -4103,8 +4332,23 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 
 				amdgpu_fbdev_set_suspend(tmp_adev, 0);
 
-				/* must succeed. */
-				amdgpu_ras_resume(tmp_adev);
+				/*
+				 * The GPU enters bad state once faulty pages
+				 * by ECC has reached the threshold, and ras
+				 * recovery is scheduled next. So add one check
+				 * here to break recovery if it indeed exceeds
+				 * bad page threshold, and remind user to
+				 * retire this GPU or setting one bigger
+				 * bad_page_threshold value to fix this once
+				 * probing driver again.
+				 */
+				if (!amdgpu_ras_check_err_threshold(tmp_adev)) {
+					/* must succeed. */
+					amdgpu_ras_resume(tmp_adev);
+				} else {
+					r = -EINVAL;
+					goto out;
+				}
 
 				/* Update PSP FW topology after reset */
 				if (hive && tmp_adev->gmc.xgmi.num_physical_nodes > 1)
@@ -4112,7 +4356,6 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 			}
 		}
 
-
 out:
 		if (!r) {
 			amdgpu_irq_gpu_reset_resume_helper(tmp_adev);
@@ -4137,16 +4380,19 @@ end:
 	return r;
 }
 
-static bool amdgpu_device_lock_adev(struct amdgpu_device *adev, bool trylock)
+static bool amdgpu_device_lock_adev(struct amdgpu_device *adev,
+				struct amdgpu_hive_info *hive)
 {
-	if (trylock) {
-		if (!mutex_trylock(&adev->lock_reset))
-			return false;
-	} else
-		mutex_lock(&adev->lock_reset);
+	if (atomic_cmpxchg(&adev->in_gpu_reset, 0, 1) != 0)
+		return false;
+
+	if (hive) {
+		down_write_nest_lock(&adev->reset_sem, &hive->hive_lock);
+	} else {
+		down_write(&adev->reset_sem);
+	}
 
 	atomic_inc(&adev->gpu_reset_counter);
-	adev->in_gpu_reset = true;
 	switch (amdgpu_asic_reset_method(adev)) {
 	case AMD_RESET_METHOD_MODE1:
 		adev->mp1_state = PP_MP1_STATE_SHUTDOWN;
@@ -4166,8 +4412,8 @@ static void amdgpu_device_unlock_adev(struct amdgpu_device *adev)
 {
 	amdgpu_vf_error_trans_all(adev);
 	adev->mp1_state = PP_MP1_STATE_NONE;
-	adev->in_gpu_reset = false;
-	mutex_unlock(&adev->lock_reset);
+	atomic_set(&adev->in_gpu_reset, 0);
+	up_write(&adev->reset_sem);
 }
 
 static void amdgpu_device_resume_display_audio(struct amdgpu_device *adev)
@@ -4277,12 +4523,15 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	 * We always reset all schedulers for device and all devices for XGMI
 	 * hive so that should take care of them too.
 	 */
-	hive = amdgpu_get_xgmi_hive(adev, true);
-	if (hive && !mutex_trylock(&hive->reset_lock)) {
-		DRM_INFO("Bailing on TDR for s_job:%llx, hive: %llx as another already in progress",
-			  job ? job->base.id : -1, hive->hive_id);
-		mutex_unlock(&hive->hive_lock);
-		return 0;
+	hive = amdgpu_get_xgmi_hive(adev);
+	if (hive) {
+		if (atomic_cmpxchg(&hive->in_reset, 0, 1) != 0) {
+			DRM_INFO("Bailing on TDR for s_job:%llx, hive: %llx as another already in progress",
+				job ? job->base.id : -1, hive->hive_id);
+			amdgpu_put_xgmi_hive(hive);
+			return 0;
+		}
+		mutex_lock(&hive->hive_lock);
 	}
 
 	/*
@@ -4304,11 +4553,11 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 
 	/* block all schedulers and reset given job's ring */
 	list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
-		if (!amdgpu_device_lock_adev(tmp_adev, !hive)) {
-			DRM_INFO("Bailing on TDR for s_job:%llx, as another already in progress",
+		if (!amdgpu_device_lock_adev(tmp_adev, hive)) {
+			dev_info(tmp_adev->dev, "Bailing on TDR for s_job:%llx, as another already in progress",
 				  job ? job->base.id : -1);
-			mutex_unlock(&hive->hive_lock);
-			return 0;
+			r = 0;
+			goto skip_recovery;
 		}
 
 		/*
@@ -4380,8 +4629,8 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 						 &need_full_reset);
 		/*TODO Should we stop ?*/
 		if (r) {
-			DRM_ERROR("GPU pre asic reset failed with err, %d for drm dev, %s ",
-				  r, tmp_adev->ddev->unique);
+			dev_err(tmp_adev->dev, "GPU pre asic reset failed with err, %d for drm dev, %s ",
+				  r, adev_to_drm(tmp_adev)->unique);
 			tmp_adev->asic_reset_res = r;
 		}
 	}
@@ -4393,7 +4642,7 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 		if (r)
 			adev->asic_reset_res = r;
 	} else {
-		r  = amdgpu_do_asic_reset(hive, device_list_handle, &need_full_reset);
+		r  = amdgpu_do_asic_reset(hive, device_list_handle, &need_full_reset, false);
 		if (r && r == -EAGAIN)
 			goto retry;
 	}
@@ -4417,7 +4666,7 @@ skip_hw_reset:
 		}
 
 		if (!amdgpu_device_has_dc_support(tmp_adev) && !job_signaled) {
-			drm_helper_resume_force_mode(tmp_adev->ddev);
+			drm_helper_resume_force_mode(adev_to_drm(tmp_adev));
 		}
 
 		tmp_adev->asic_reset_res = 0;
@@ -4441,9 +4690,11 @@ skip_sched_resume:
 		amdgpu_device_unlock_adev(tmp_adev);
 	}
 
+skip_recovery:
 	if (hive) {
-		mutex_unlock(&hive->reset_lock);
+		atomic_set(&hive->in_reset, 0);
 		mutex_unlock(&hive->hive_lock);
+		amdgpu_put_xgmi_hive(hive);
 	}
 
 	if (r)
@@ -4589,10 +4840,10 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 
 int amdgpu_device_baco_enter(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
-	if (!amdgpu_device_supports_baco(adev->ddev))
+	if (!amdgpu_device_supports_baco(adev_to_drm(adev)))
 		return -ENOTSUPP;
 
 	if (ras && ras->supported)
@@ -4603,11 +4854,11 @@ int amdgpu_device_baco_enter(struct drm_device *dev)
 
 int amdgpu_device_baco_exit(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 	int ret = 0;
 
-	if (!amdgpu_device_supports_baco(adev->ddev))
+	if (!amdgpu_device_supports_baco(adev_to_drm(adev)))
 		return -ENOTSUPP;
 
 	ret = amdgpu_dpm_baco_exit(adev);
@@ -4619,3 +4870,235 @@ int amdgpu_device_baco_exit(struct drm_device *dev)
 
 	return 0;
 }
+
+static void amdgpu_cancel_all_tdr(struct amdgpu_device *adev)
+{
+	int i;
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (!ring || !ring->sched.thread)
+			continue;
+
+		cancel_delayed_work_sync(&ring->sched.work_tdr);
+	}
+}
+
+/**
+ * amdgpu_pci_error_detected - Called when a PCI error is detected.
+ * @pdev: PCI device struct
+ * @state: PCI channel state
+ *
+ * Description: Called when a PCI error is detected.
+ *
+ * Return: PCI_ERS_RESULT_NEED_RESET or PCI_ERS_RESULT_DISCONNECT.
+ */
+pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_state_t state)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int i;
+
+	DRM_INFO("PCI error: detected callback, state(%d)!!\n", state);
+
+	if (adev->gmc.xgmi.num_physical_nodes > 1) {
+		DRM_WARN("No support for XGMI hive yet...");
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	switch (state) {
+	case pci_channel_io_normal:
+		return PCI_ERS_RESULT_CAN_RECOVER;
+	/* Fatal error, prepare for slot reset */
+	case pci_channel_io_frozen:		
+		/*		
+		 * Cancel and wait for all TDRs in progress if failing to
+		 * set  adev->in_gpu_reset in amdgpu_device_lock_adev
+		 *
+		 * Locking adev->reset_sem will prevent any external access
+		 * to GPU during PCI error recovery
+		 */
+		while (!amdgpu_device_lock_adev(adev, NULL))
+			amdgpu_cancel_all_tdr(adev);
+
+		/*
+		 * Block any work scheduling as we do for regular GPU reset
+		 * for the duration of the recovery
+		 */
+		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+			struct amdgpu_ring *ring = adev->rings[i];
+
+			if (!ring || !ring->sched.thread)
+				continue;
+
+			drm_sched_stop(&ring->sched, NULL);
+		}
+		return PCI_ERS_RESULT_NEED_RESET;
+	case pci_channel_io_perm_failure:
+		/* Permanent error, prepare for device removal */
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+/**
+ * amdgpu_pci_mmio_enabled - Enable MMIO and dump debug registers
+ * @pdev: pointer to PCI device
+ */
+pci_ers_result_t amdgpu_pci_mmio_enabled(struct pci_dev *pdev)
+{
+
+	DRM_INFO("PCI error: mmio enabled callback!!\n");
+
+	/* TODO - dump whatever for debugging purposes */
+
+	/* This called only if amdgpu_pci_error_detected returns
+	 * PCI_ERS_RESULT_CAN_RECOVER. Read/write to the device still
+	 * works, no need to reset slot.
+	 */
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * amdgpu_pci_slot_reset - Called when PCI slot has been reset.
+ * @pdev: PCI device struct
+ *
+ * Description: This routine is called by the pci error recovery
+ * code after the PCI slot has been reset, just before we
+ * should resume normal operations.
+ */
+pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int r, i;
+	bool need_full_reset = true;
+	u32 memsize;
+	struct list_head device_list;
+
+	DRM_INFO("PCI error: slot reset callback!!\n");
+
+	INIT_LIST_HEAD(&device_list);
+	list_add_tail(&adev->gmc.xgmi.head, &device_list);
+
+	/* wait for asic to come out of reset */
+	msleep(500);
+
+	/* Restore PCI confspace */
+	amdgpu_device_load_pci_state(pdev);
+
+	/* confirm  ASIC came out of reset */
+	for (i = 0; i < adev->usec_timeout; i++) {
+		memsize = amdgpu_asic_get_config_memsize(adev);
+
+		if (memsize != 0xffffffff)
+			break;
+		udelay(1);
+	}
+	if (memsize == 0xffffffff) {
+		r = -ETIME;
+		goto out;
+	}
+
+	adev->in_pci_err_recovery = true;	
+	r = amdgpu_device_pre_asic_reset(adev, NULL, &need_full_reset);
+	adev->in_pci_err_recovery = false;
+	if (r)
+		goto out;
+
+	r = amdgpu_do_asic_reset(NULL, &device_list, &need_full_reset, true);
+
+out:
+	if (!r) {
+		if (amdgpu_device_cache_pci_state(adev->pdev))
+			pci_restore_state(adev->pdev);
+
+		DRM_INFO("PCIe error recovery succeeded\n");
+	} else {
+		DRM_ERROR("PCIe error recovery failed, err:%d", r);
+		amdgpu_device_unlock_adev(adev);
+	}
+
+	return r ? PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * amdgpu_pci_resume() - resume normal ops after PCI reset
+ * @pdev: pointer to PCI device
+ *
+ * Called when the error recovery driver tells us that its
+ * OK to resume normal operation. Use completion to allow
+ * halted scsi ops to resume.
+ */
+void amdgpu_pci_resume(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int i;
+
+
+	DRM_INFO("PCI error: resume callback!!\n");
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (!ring || !ring->sched.thread)
+			continue;
+
+
+		drm_sched_resubmit_jobs(&ring->sched);
+		drm_sched_start(&ring->sched, true);
+	}
+
+	amdgpu_device_unlock_adev(adev);
+}
+
+bool amdgpu_device_cache_pci_state(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int r;
+
+	r = pci_save_state(pdev);
+	if (!r) {
+		kfree(adev->pci_state);
+
+		adev->pci_state = pci_store_saved_state(pdev);
+
+		if (!adev->pci_state) {
+			DRM_ERROR("Failed to store PCI saved state");
+			return false;
+		}
+	} else {
+		DRM_WARN("Failed to save PCI state, err:%d\n", r);
+		return false;
+	}
+
+	return true;
+}
+
+bool amdgpu_device_load_pci_state(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int r;
+
+	if (!adev->pci_state)
+		return false;
+
+	r = pci_load_saved_state(pdev, adev->pci_state);
+
+	if (!r) {
+		pci_restore_state(pdev);
+	} else {
+		DRM_WARN("Failed to load PCI state, err:%d\n", r);
+		return false;
+	}
+
+	return true;
+}
+
+
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h
index 61a26c15c8dd..373cdebe0e2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h
@@ -44,9 +44,9 @@ struct amdgpu_df_funcs {
 	void (*enable_ecc_force_par_wr_rmw)(struct amdgpu_device *adev,
 					    bool enable);
 	int (*pmc_start)(struct amdgpu_device *adev, uint64_t config,
-					 int is_enable);
+					 int is_add);
 	int (*pmc_stop)(struct amdgpu_device *adev, uint64_t config,
-					 int is_disable);
+					 int is_remove);
 	void (*pmc_get_count)(struct amdgpu_device *adev, uint64_t config,
 					 uint64_t *count);
 	uint64_t (*get_fica)(struct amdgpu_device *adev, uint32_t ficaa_val);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index a50ff2306504..bfb95143ba5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -136,7 +136,7 @@ static int amdgpu_discovery_read_binary(struct amdgpu_device *adev, uint8_t *bin
 	uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET;
 
 	amdgpu_device_vram_access(adev, pos, (uint32_t *)binary,
-				  adev->discovery_tmr_size, false);
+				  adev->mman.discovery_tmr_size, false);
 	return 0;
 }
 
@@ -168,18 +168,18 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	uint16_t checksum;
 	int r;
 
-	adev->discovery_tmr_size = DISCOVERY_TMR_SIZE;
-	adev->discovery_bin = kzalloc(adev->discovery_tmr_size, GFP_KERNEL);
-	if (!adev->discovery_bin)
+	adev->mman.discovery_tmr_size = DISCOVERY_TMR_SIZE;
+	adev->mman.discovery_bin = kzalloc(adev->mman.discovery_tmr_size, GFP_KERNEL);
+	if (!adev->mman.discovery_bin)
 		return -ENOMEM;
 
-	r = amdgpu_discovery_read_binary(adev, adev->discovery_bin);
+	r = amdgpu_discovery_read_binary(adev, adev->mman.discovery_bin);
 	if (r) {
 		DRM_ERROR("failed to read ip discovery binary\n");
 		goto out;
 	}
 
-	bhdr = (struct binary_header *)adev->discovery_bin;
+	bhdr = (struct binary_header *)adev->mman.discovery_bin;
 
 	if (le32_to_cpu(bhdr->binary_signature) != BINARY_SIGNATURE) {
 		DRM_ERROR("invalid ip discovery binary signature\n");
@@ -192,7 +192,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	size = bhdr->binary_size - offset;
 	checksum = bhdr->binary_checksum;
 
-	if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset,
+	if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset,
 					      size, checksum)) {
 		DRM_ERROR("invalid ip discovery binary checksum\n");
 		r = -EINVAL;
@@ -202,7 +202,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	info = &bhdr->table_list[IP_DISCOVERY];
 	offset = le16_to_cpu(info->offset);
 	checksum = le16_to_cpu(info->checksum);
-	ihdr = (struct ip_discovery_header *)(adev->discovery_bin + offset);
+	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin + offset);
 
 	if (le32_to_cpu(ihdr->signature) != DISCOVERY_TABLE_SIGNATURE) {
 		DRM_ERROR("invalid ip discovery data table signature\n");
@@ -210,7 +210,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 		goto out;
 	}
 
-	if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset,
+	if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset,
 					      ihdr->size, checksum)) {
 		DRM_ERROR("invalid ip discovery data table checksum\n");
 		r = -EINVAL;
@@ -220,9 +220,9 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	info = &bhdr->table_list[GC];
 	offset = le16_to_cpu(info->offset);
 	checksum = le16_to_cpu(info->checksum);
-	ghdr = (struct gpu_info_header *)(adev->discovery_bin + offset);
+	ghdr = (struct gpu_info_header *)(adev->mman.discovery_bin + offset);
 
-	if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset,
+	if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset,
 				              ghdr->size, checksum)) {
 		DRM_ERROR("invalid gc data table checksum\n");
 		r = -EINVAL;
@@ -232,16 +232,16 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	return 0;
 
 out:
-	kfree(adev->discovery_bin);
-	adev->discovery_bin = NULL;
+	kfree(adev->mman.discovery_bin);
+	adev->mman.discovery_bin = NULL;
 
 	return r;
 }
 
 void amdgpu_discovery_fini(struct amdgpu_device *adev)
 {
-	kfree(adev->discovery_bin);
-	adev->discovery_bin = NULL;
+	kfree(adev->mman.discovery_bin);
+	adev->mman.discovery_bin = NULL;
 }
 
 int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
@@ -265,8 +265,8 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 		return r;
 	}
 
-	bhdr = (struct binary_header *)adev->discovery_bin;
-	ihdr = (struct ip_discovery_header *)(adev->discovery_bin +
+	bhdr = (struct binary_header *)adev->mman.discovery_bin;
+	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
 	num_dies = le16_to_cpu(ihdr->num_dies);
 
@@ -274,7 +274,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 
 	for (i = 0; i < num_dies; i++) {
 		die_offset = le16_to_cpu(ihdr->die_info[i].die_offset);
-		dhdr = (struct die_header *)(adev->discovery_bin + die_offset);
+		dhdr = (struct die_header *)(adev->mman.discovery_bin + die_offset);
 		num_ips = le16_to_cpu(dhdr->num_ips);
 		ip_offset = die_offset + sizeof(*dhdr);
 
@@ -288,7 +288,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				le16_to_cpu(dhdr->die_id), num_ips);
 
 		for (j = 0; j < num_ips; j++) {
-			ip = (struct ip *)(adev->discovery_bin + ip_offset);
+			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
 			num_base_address = ip->num_base_address;
 
 			DRM_DEBUG("%s(%d) #%d v%d.%d.%d:\n",
@@ -337,24 +337,24 @@ int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id,
 	uint16_t num_ips;
 	int i, j;
 
-	if (!adev->discovery_bin) {
+	if (!adev->mman.discovery_bin) {
 		DRM_ERROR("ip discovery uninitialized\n");
 		return -EINVAL;
 	}
 
-	bhdr = (struct binary_header *)adev->discovery_bin;
-	ihdr = (struct ip_discovery_header *)(adev->discovery_bin +
+	bhdr = (struct binary_header *)adev->mman.discovery_bin;
+	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
 	num_dies = le16_to_cpu(ihdr->num_dies);
 
 	for (i = 0; i < num_dies; i++) {
 		die_offset = le16_to_cpu(ihdr->die_info[i].die_offset);
-		dhdr = (struct die_header *)(adev->discovery_bin + die_offset);
+		dhdr = (struct die_header *)(adev->mman.discovery_bin + die_offset);
 		num_ips = le16_to_cpu(dhdr->num_ips);
 		ip_offset = die_offset + sizeof(*dhdr);
 
 		for (j = 0; j < num_ips; j++) {
-			ip = (struct ip *)(adev->discovery_bin + ip_offset);
+			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
 
 			if (le16_to_cpu(ip->hw_id) == hw_id) {
 				if (major)
@@ -377,13 +377,13 @@ int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev)
 	struct binary_header *bhdr;
 	struct gc_info_v1_0 *gc_info;
 
-	if (!adev->discovery_bin) {
+	if (!adev->mman.discovery_bin) {
 		DRM_ERROR("ip discovery uninitialized\n");
 		return -EINVAL;
 	}
 
-	bhdr = (struct binary_header *)adev->discovery_bin;
-	gc_info = (struct gc_info_v1_0 *)(adev->discovery_bin +
+	bhdr = (struct binary_header *)adev->mman.discovery_bin;
+	gc_info = (struct gc_info_v1_0 *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[GC].offset));
 
 	adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->gc_num_se);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index 44c1f6e00635..7cc7af2a6822 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -93,7 +93,7 @@ static void amdgpu_display_flip_work_func(struct work_struct *__work)
 	 * targeted by the flip
 	 */
 	if (amdgpu_crtc->enabled &&
-	    (amdgpu_display_get_crtc_scanoutpos(adev->ddev, work->crtc_id, 0,
+	    (amdgpu_display_get_crtc_scanoutpos(adev_to_drm(adev), work->crtc_id, 0,
 						&vpos, &hpos, NULL, NULL,
 						&crtc->hwmode)
 	     & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) ==
@@ -152,7 +152,7 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc,
 				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
 	struct amdgpu_flip_work *work;
@@ -292,7 +292,7 @@ int amdgpu_display_crtc_set_config(struct drm_mode_set *set,
 
 	pm_runtime_mark_last_busy(dev->dev);
 
-	adev = dev->dev_private;
+	adev = drm_to_adev(dev);
 	/* if we have active crtcs and we don't have a power ref,
 	   take the current one */
 	if (active && !adev->have_disp_power_ref) {
@@ -619,51 +619,51 @@ int amdgpu_display_modeset_create_props(struct amdgpu_device *adev)
 	int sz;
 
 	adev->mode_info.coherent_mode_property =
-		drm_property_create_range(adev->ddev, 0 , "coherent", 0, 1);
+		drm_property_create_range(adev_to_drm(adev), 0, "coherent", 0, 1);
 	if (!adev->mode_info.coherent_mode_property)
 		return -ENOMEM;
 
 	adev->mode_info.load_detect_property =
-		drm_property_create_range(adev->ddev, 0, "load detection", 0, 1);
+		drm_property_create_range(adev_to_drm(adev), 0, "load detection", 0, 1);
 	if (!adev->mode_info.load_detect_property)
 		return -ENOMEM;
 
-	drm_mode_create_scaling_mode_property(adev->ddev);
+	drm_mode_create_scaling_mode_property(adev_to_drm(adev));
 
 	sz = ARRAY_SIZE(amdgpu_underscan_enum_list);
 	adev->mode_info.underscan_property =
-		drm_property_create_enum(adev->ddev, 0,
-				    "underscan",
-				    amdgpu_underscan_enum_list, sz);
+		drm_property_create_enum(adev_to_drm(adev), 0,
+					 "underscan",
+					 amdgpu_underscan_enum_list, sz);
 
 	adev->mode_info.underscan_hborder_property =
-		drm_property_create_range(adev->ddev, 0,
-					"underscan hborder", 0, 128);
+		drm_property_create_range(adev_to_drm(adev), 0,
+					  "underscan hborder", 0, 128);
 	if (!adev->mode_info.underscan_hborder_property)
 		return -ENOMEM;
 
 	adev->mode_info.underscan_vborder_property =
-		drm_property_create_range(adev->ddev, 0,
-					"underscan vborder", 0, 128);
+		drm_property_create_range(adev_to_drm(adev), 0,
+					  "underscan vborder", 0, 128);
 	if (!adev->mode_info.underscan_vborder_property)
 		return -ENOMEM;
 
 	sz = ARRAY_SIZE(amdgpu_audio_enum_list);
 	adev->mode_info.audio_property =
-		drm_property_create_enum(adev->ddev, 0,
+		drm_property_create_enum(adev_to_drm(adev), 0,
 					 "audio",
 					 amdgpu_audio_enum_list, sz);
 
 	sz = ARRAY_SIZE(amdgpu_dither_enum_list);
 	adev->mode_info.dither_property =
-		drm_property_create_enum(adev->ddev, 0,
+		drm_property_create_enum(adev_to_drm(adev), 0,
 					 "dither",
 					 amdgpu_dither_enum_list, sz);
 
 	if (amdgpu_device_has_dc_support(adev)) {
 		adev->mode_info.abm_level_property =
-			drm_property_create_range(adev->ddev, 0,
-						"abm level", 0, 4);
+			drm_property_create_range(adev_to_drm(adev), 0,
+						  "abm level", 0, 4);
 		if (!adev->mode_info.abm_level_property)
 			return -ENOMEM;
 	}
@@ -813,7 +813,7 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev,
 	int vbl_start, vbl_end, vtotal, ret = 0;
 	bool in_vbl = true;
 
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index 519ce4427fce..957934926b24 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -35,6 +35,7 @@
 #include "amdgpu_display.h"
 #include "amdgpu_gem.h"
 #include "amdgpu_dma_buf.h"
+#include "amdgpu_xgmi.h"
 #include <drm/amdgpu_drm.h>
 #include <linux/dma-buf.h>
 #include <linux/dma-fence-array.h>
@@ -302,7 +303,8 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach,
 
 	switch (bo->tbo.mem.mem_type) {
 	case TTM_PL_TT:
-		sgt = drm_prime_pages_to_sg(bo->tbo.ttm->pages,
+		sgt = drm_prime_pages_to_sg(obj->dev,
+					    bo->tbo.ttm->pages,
 					    bo->tbo.num_pages);
 		if (IS_ERR(sgt))
 			return sgt;
@@ -454,7 +456,7 @@ static struct drm_gem_object *
 amdgpu_dma_buf_create_obj(struct drm_device *dev, struct dma_buf *dma_buf)
 {
 	struct dma_resv *resv = dma_buf->resv;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_bo *bo;
 	struct amdgpu_bo_param bp;
 	int ret;
@@ -595,3 +597,36 @@ struct drm_gem_object *amdgpu_gem_prime_import(struct drm_device *dev,
 	obj->import_attach = attach;
 	return obj;
 }
+
+/**
+ * amdgpu_dmabuf_is_xgmi_accessible - Check if xgmi available for P2P transfer
+ *
+ * @adev: amdgpu_device pointer of the importer
+ * @bo: amdgpu buffer object
+ *
+ * Returns:
+ * True if dmabuf accessible over xgmi, false otherwise.
+ */
+bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev,
+				      struct amdgpu_bo *bo)
+{
+	struct drm_gem_object *obj = &bo->tbo.base;
+	struct drm_gem_object *gobj;
+
+	if (obj->import_attach) {
+		struct dma_buf *dma_buf = obj->import_attach->dmabuf;
+
+		if (dma_buf->ops != &amdgpu_dmabuf_ops)
+			/* No XGMI with non AMD GPUs */
+			return false;
+
+		gobj = dma_buf->priv;
+		bo = gem_to_amdgpu_bo(gobj);
+	}
+
+	if (amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) &&
+			(bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM))
+		return true;
+
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.h
index ec447a7b6b28..2c5c84a06bb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.h
@@ -29,6 +29,8 @@ struct dma_buf *amdgpu_gem_prime_export(struct drm_gem_object *gobj,
 					int flags);
 struct drm_gem_object *amdgpu_gem_prime_import(struct drm_device *dev,
 					    struct dma_buf *dma_buf);
+bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev,
+				      struct amdgpu_bo *bo);
 void *amdgpu_gem_prime_vmap(struct drm_gem_object *obj);
 void amdgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
 int amdgpu_gem_prime_mmap(struct drm_gem_object *obj,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 321032d3a51a..c241317edee7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -26,12 +26,12 @@
 #include <drm/drm_drv.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_vblank.h>
+#include <drm/drm_managed.h>
 #include "amdgpu_drv.h"
 
 #include <drm/drm_pciids.h>
 #include <linux/console.h>
 #include <linux/module.h>
-#include <linux/pci.h>
 #include <linux/pm_runtime.h>
 #include <linux/vga_switcheroo.h>
 #include <drm/drm_probe_helper.h>
@@ -88,9 +88,10 @@
  * - 3.37.0 - L2 is invalidated before SDMA IBs, needed for correctness
  * - 3.38.0 - Add AMDGPU_IB_FLAG_EMIT_MEM_SYNC
  * - 3.39.0 - DMABUF implicit sync does a full pipeline sync
+ * - 3.40.0 - Add AMDGPU_IDS_FLAGS_TMZ
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	39
+#define KMS_DRIVER_MINOR	40
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;
@@ -146,16 +147,18 @@ int amdgpu_async_gfx_ring = 1;
 int amdgpu_mcbp = 0;
 int amdgpu_discovery = -1;
 int amdgpu_mes = 0;
-int amdgpu_noretry;
+int amdgpu_noretry = -1;
 int amdgpu_force_asic_type = -1;
 int amdgpu_tmz = 0;
 int amdgpu_reset_method = -1; /* auto */
+int amdgpu_num_kcq = -1;
 
 struct amdgpu_mgpu_info mgpu_info = {
 	.mutex = __MUTEX_INITIALIZER(mgpu_info.mutex),
 };
 int amdgpu_ras_enable = -1;
 uint amdgpu_ras_mask = 0xffffffff;
+int amdgpu_bad_page_threshold = -1;
 
 /**
  * DOC: vramlimit (int)
@@ -393,12 +396,12 @@ MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default
 module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444);
 
 /**
- * DOC: ppfeaturemask (uint)
+ * DOC: ppfeaturemask (hexint)
  * Override power features enabled. See enum PP_FEATURE_MASK in drivers/gpu/drm/amd/include/amd_shared.h.
  * The default is the current set of stable power features.
  */
 MODULE_PARM_DESC(ppfeaturemask, "all power features enabled (default))");
-module_param_named(ppfeaturemask, amdgpu_pp_feature_mask, uint, 0444);
+module_param_named(ppfeaturemask, amdgpu_pp_feature_mask, hexint, 0444);
 
 /**
  * DOC: forcelongtraining (uint)
@@ -593,8 +596,13 @@ MODULE_PARM_DESC(mes,
 	"Enable Micro Engine Scheduler (0 = disabled (default), 1 = enabled)");
 module_param_named(mes, amdgpu_mes, int, 0444);
 
+/**
+ * DOC: noretry (int)
+ * Disable retry faults in the GPU memory controller.
+ * (0 = retry enabled, 1 = retry disabled, -1 auto (default))
+ */
 MODULE_PARM_DESC(noretry,
-	"Disable retry faults (0 = retry enabled (default), 1 = retry disabled)");
+	"Disable retry faults (0 = retry enabled, 1 = retry disabled, -1 auto (default))");
 module_param_named(noretry, amdgpu_noretry, int, 0644);
 
 /**
@@ -676,11 +684,14 @@ MODULE_PARM_DESC(debug_largebar,
  * Ignore CRAT table during KFD initialization. By default, KFD uses the ACPI CRAT
  * table to get information about AMD APUs. This option can serve as a workaround on
  * systems with a broken CRAT table.
+ *
+ * Default is auto (according to asic type, iommu_v2, and crat table, to decide
+ * whehter use CRAT)
  */
 int ignore_crat;
 module_param(ignore_crat, int, 0444);
 MODULE_PARM_DESC(ignore_crat,
-	"Ignore CRAT table during KFD initialization (0 = use CRAT (default), 1 = ignore CRAT)");
+	"Ignore CRAT table during KFD initialization (0 = auto (default), 1 = ignore CRAT)");
 
 /**
  * DOC: halt_if_hws_hang (int)
@@ -715,6 +726,15 @@ MODULE_PARM_DESC(queue_preemption_timeout_ms, "queue preemption timeout in ms (1
 bool debug_evictions;
 module_param(debug_evictions, bool, 0644);
 MODULE_PARM_DESC(debug_evictions, "enable eviction debug messages (false = default)");
+
+/**
+ * DOC: no_system_mem_limit(bool)
+ * Disable system memory limit, to support multiple process shared memory
+ */
+bool no_system_mem_limit;
+module_param(no_system_mem_limit, bool, 0644);
+MODULE_PARM_DESC(no_system_mem_limit, "disable system memory limit (false = default)");
+
 #endif
 
 /**
@@ -765,6 +785,19 @@ module_param_named(tmz, amdgpu_tmz, int, 0444);
 MODULE_PARM_DESC(reset_method, "GPU reset method (-1 = auto (default), 0 = legacy, 1 = mode0, 2 = mode1, 3 = mode2, 4 = baco)");
 module_param_named(reset_method, amdgpu_reset_method, int, 0444);
 
+/**
+ * DOC: bad_page_threshold (int)
+ * Bad page threshold is to specify the threshold value of faulty pages
+ * detected by RAS ECC, that may result in GPU entering bad status if total
+ * faulty pages by ECC exceed threshold value and leave it for user's further
+ * check.
+ */
+MODULE_PARM_DESC(bad_page_threshold, "Bad page threshold(-1 = auto(default value), 0 = disable bad page retirement)");
+module_param_named(bad_page_threshold, amdgpu_bad_page_threshold, int, 0444);
+
+MODULE_PARM_DESC(num_kcq, "number of kernel compute queue user want to setup (8 if set to greater than 8 or less than 0, only affect gfx 8+)");
+module_param_named(num_kcq, amdgpu_num_kcq, int, 0444);
+
 static const struct pci_device_id pciidlist[] = {
 #ifdef  CONFIG_DRM_AMDGPU_SI
 	{0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
@@ -1065,7 +1098,7 @@ static struct drm_driver kms_driver;
 static int amdgpu_pci_probe(struct pci_dev *pdev,
 			    const struct pci_device_id *ent)
 {
-	struct drm_device *dev;
+	struct drm_device *ddev;
 	struct amdgpu_device *adev;
 	unsigned long flags = ent->driver_data;
 	int ret, retry = 0;
@@ -1081,6 +1114,16 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 		return -ENODEV;
 	}
 
+	/* Due to hardware bugs, S/G Display on raven requires a 1:1 IOMMU mapping,
+	 * however, SME requires an indirect IOMMU mapping because the encryption
+	 * bit is beyond the DMA mask of the chip.
+	 */
+	if (mem_encrypt_active() && ((flags & AMD_ASIC_MASK) == CHIP_RAVEN)) {
+		dev_info(&pdev->dev,
+			 "SME is not compatible with RAVEN\n");
+		return -ENOTSUPP;
+	}
+
 #ifdef CONFIG_DRM_AMDGPU_SI
 	if (!amdgpu_si_support) {
 		switch (flags & AMD_ASIC_MASK) {
@@ -1121,36 +1164,39 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 	if (ret)
 		return ret;
 
-	dev = drm_dev_alloc(&kms_driver, &pdev->dev);
-	if (IS_ERR(dev))
-		return PTR_ERR(dev);
+	adev = devm_drm_dev_alloc(&pdev->dev, &kms_driver, typeof(*adev), ddev);
+	if (IS_ERR(adev))
+		return PTR_ERR(adev);
+
+	adev->dev  = &pdev->dev;
+	adev->pdev = pdev;
+	ddev = adev_to_drm(adev);
 
 	if (!supports_atomic)
-		dev->driver_features &= ~DRIVER_ATOMIC;
+		ddev->driver_features &= ~DRIVER_ATOMIC;
 
 	ret = pci_enable_device(pdev);
 	if (ret)
-		goto err_free;
-
-	dev->pdev = pdev;
+		return ret;
 
-	pci_set_drvdata(pdev, dev);
+	ddev->pdev = pdev;
+	pci_set_drvdata(pdev, ddev);
 
-	ret = amdgpu_driver_load_kms(dev, ent->driver_data);
+	ret = amdgpu_driver_load_kms(adev, ent->driver_data);
 	if (ret)
 		goto err_pci;
 
 retry_init:
-	ret = drm_dev_register(dev, ent->driver_data);
+	ret = drm_dev_register(ddev, ent->driver_data);
 	if (ret == -EAGAIN && ++retry <= 3) {
 		DRM_INFO("retry init %d\n", retry);
 		/* Don't request EX mode too frequently which is attacking */
 		msleep(5000);
 		goto retry_init;
-	} else if (ret)
+	} else if (ret) {
 		goto err_pci;
+	}
 
-	adev = dev->dev_private;
 	ret = amdgpu_debugfs_init(adev);
 	if (ret)
 		DRM_ERROR("Creating debugfs files failed (%d).\n", ret);
@@ -1159,8 +1205,6 @@ retry_init:
 
 err_pci:
 	pci_disable_device(pdev);
-err_free:
-	drm_dev_put(dev);
 	return ret;
 }
 
@@ -1177,14 +1221,13 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 	amdgpu_driver_unload_kms(dev);
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
-	drm_dev_put(dev);
 }
 
 static void
 amdgpu_pci_shutdown(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (amdgpu_ras_intr_triggered())
 		return;
@@ -1217,7 +1260,7 @@ static int amdgpu_pmops_resume(struct device *dev)
 static int amdgpu_pmops_freeze(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = drm_dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
 	int r;
 
 	adev->in_hibernate = true;
@@ -1253,7 +1296,7 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct drm_device *drm_dev = pci_get_drvdata(pdev);
-	struct amdgpu_device *adev = drm_dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
 	int ret, i;
 
 	if (!adev->runpm) {
@@ -1287,7 +1330,7 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev)
 		if (amdgpu_is_atpx_hybrid()) {
 			pci_ignore_hotplug(pdev);
 		} else {
-			pci_save_state(pdev);
+			amdgpu_device_cache_pci_state(pdev);
 			pci_disable_device(pdev);
 			pci_ignore_hotplug(pdev);
 			pci_set_power_state(pdev, PCI_D3cold);
@@ -1304,7 +1347,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct drm_device *drm_dev = pci_get_drvdata(pdev);
-	struct amdgpu_device *adev = drm_dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
 	int ret;
 
 	if (!adev->runpm)
@@ -1320,7 +1363,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev)
 			pci_set_master(pdev);
 		} else {
 			pci_set_power_state(pdev, PCI_D0);
-			pci_restore_state(pdev);
+			amdgpu_device_load_pci_state(pdev);
 			ret = pci_enable_device(pdev);
 			if (ret)
 				return ret;
@@ -1340,7 +1383,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev)
 static int amdgpu_pmops_runtime_idle(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = drm_dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
 	/* we don't want the main rpm_idle to call suspend - we want to autosuspend */
 	int ret = 1;
 
@@ -1499,6 +1542,13 @@ static struct drm_driver kms_driver = {
 	.patchlevel = KMS_DRIVER_PATCHLEVEL,
 };
 
+static struct pci_error_handlers amdgpu_pci_err_handler = {
+	.error_detected	= amdgpu_pci_error_detected,
+	.mmio_enabled	= amdgpu_pci_mmio_enabled,
+	.slot_reset	= amdgpu_pci_slot_reset,
+	.resume		= amdgpu_pci_resume,
+};
+
 static struct pci_driver amdgpu_kms_pci_driver = {
 	.name = DRIVER_NAME,
 	.id_table = pciidlist,
@@ -1506,10 +1556,9 @@ static struct pci_driver amdgpu_kms_pci_driver = {
 	.remove = amdgpu_pci_remove,
 	.shutdown = amdgpu_pci_shutdown,
 	.driver.pm = &amdgpu_pm_ops,
+	.err_handler = &amdgpu_pci_err_handler,
 };
 
-
-
 static int __init amdgpu_init(void)
 {
 	int r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
index 61fcf247a638..af4ef84e27a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
@@ -35,7 +35,7 @@
 void
 amdgpu_link_encoder_connector(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	struct amdgpu_connector *amdgpu_connector;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
index db731f573f98..e2c2eb45a793 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@@ -135,7 +135,7 @@ static int amdgpufb_create_pinned_object(struct amdgpu_fbdev *rfbdev,
 			       AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS     |
 			       AMDGPU_GEM_CREATE_VRAM_CLEARED;
 
-	info = drm_get_format_info(adev->ddev, mode_cmd);
+	info = drm_get_format_info(adev_to_drm(adev), mode_cmd);
 	cpp = info->cpp[0];
 
 	/* need to align pitch with crtc limits */
@@ -231,7 +231,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 		goto out;
 	}
 
-	ret = amdgpu_display_framebuffer_init(adev->ddev, &rfbdev->rfb,
+	ret = amdgpu_display_framebuffer_init(adev_to_drm(adev), &rfbdev->rfb,
 					      &mode_cmd, gobj);
 	if (ret) {
 		DRM_ERROR("failed to initialize framebuffer %d\n", ret);
@@ -254,7 +254,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	drm_fb_helper_fill_info(info, &rfbdev->helper, sizes);
 
 	/* setup aperture base/size for vesafb takeover */
-	info->apertures->ranges[0].base = adev->ddev->mode_config.fb_base;
+	info->apertures->ranges[0].base = adev_to_drm(adev)->mode_config.fb_base;
 	info->apertures->ranges[0].size = adev->gmc.aper_size;
 
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
@@ -270,7 +270,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	DRM_INFO("fb depth is %d\n", fb->format->depth);
 	DRM_INFO("   pitch is %d\n", fb->pitches[0]);
 
-	vga_switcheroo_client_fb_set(adev->ddev->pdev, info);
+	vga_switcheroo_client_fb_set(adev_to_drm(adev)->pdev, info);
 	return 0;
 
 out:
@@ -318,7 +318,7 @@ int amdgpu_fbdev_init(struct amdgpu_device *adev)
 		return 0;
 
 	/* don't init fbdev if there are no connectors */
-	if (list_empty(&adev->ddev->mode_config.connector_list))
+	if (list_empty(&adev_to_drm(adev)->mode_config.connector_list))
 		return 0;
 
 	/* select 8 bpp console on low vram cards */
@@ -332,10 +332,10 @@ int amdgpu_fbdev_init(struct amdgpu_device *adev)
 	rfbdev->adev = adev;
 	adev->mode_info.rfbdev = rfbdev;
 
-	drm_fb_helper_prepare(adev->ddev, &rfbdev->helper,
-			&amdgpu_fb_helper_funcs);
+	drm_fb_helper_prepare(adev_to_drm(adev), &rfbdev->helper,
+			      &amdgpu_fb_helper_funcs);
 
-	ret = drm_fb_helper_init(adev->ddev, &rfbdev->helper);
+	ret = drm_fb_helper_init(adev_to_drm(adev), &rfbdev->helper);
 	if (ret) {
 		kfree(rfbdev);
 		return ret;
@@ -343,7 +343,7 @@ int amdgpu_fbdev_init(struct amdgpu_device *adev)
 
 	/* disable all the possible outputs/crtcs before entering KMS mode */
 	if (!amdgpu_device_has_dc_support(adev))
-		drm_helper_disable_unused_functions(adev->ddev);
+		drm_helper_disable_unused_functions(adev_to_drm(adev));
 
 	drm_fb_helper_initial_config(&rfbdev->helper, bpp_sel);
 	return 0;
@@ -354,7 +354,7 @@ void amdgpu_fbdev_fini(struct amdgpu_device *adev)
 	if (!adev->mode_info.rfbdev)
 		return;
 
-	amdgpu_fbdev_destroy(adev->ddev, adev->mode_info.rfbdev);
+	amdgpu_fbdev_destroy(adev_to_drm(adev), adev->mode_info.rfbdev);
 	kfree(adev->mode_info.rfbdev);
 	adev->mode_info.rfbdev = NULL;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 58d4c219178a..fe2d495d08ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -155,7 +155,7 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 		       seq);
 	amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr,
 			       seq, flags | AMDGPU_FENCE_FLAG_INT);
-	pm_runtime_get_noresume(adev->ddev->dev);
+	pm_runtime_get_noresume(adev_to_drm(adev)->dev);
 	ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask];
 	if (unlikely(rcu_dereference_protected(*ptr, 1))) {
 		struct dma_fence *old;
@@ -284,8 +284,8 @@ bool amdgpu_fence_process(struct amdgpu_ring *ring)
 			BUG();
 
 		dma_fence_put(fence);
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 	} while (last_seq != seq);
 
 	return true;
@@ -700,7 +700,7 @@ static int amdgpu_debugfs_fence_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int i;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
@@ -749,7 +749,7 @@ static int amdgpu_debugfs_gpu_recover(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r;
 
 	r = pm_runtime_get_sync(dev->dev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
index e811fecc540f..8f4a8f8d8146 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
@@ -34,18 +34,31 @@
 
 static bool is_fru_eeprom_supported(struct amdgpu_device *adev)
 {
-	/* TODO: Gaming SKUs don't have the FRU EEPROM.
-	 * Use this hack to address hangs on modprobe on gaming SKUs
-	 * until a proper solution can be implemented by only supporting
-	 * the explicit chip IDs for VG20 Server cards
-	 *
-	 * TODO: Add list of supported Arcturus DIDs once confirmed
+	/* Only server cards have the FRU EEPROM
+	 * TODO: See if we can figure this out dynamically instead of
+	 * having to parse VBIOS versions.
 	 */
-	if ((adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a0) ||
-	    (adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a1) ||
-	    (adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a4))
-		return true;
-	return false;
+	struct atom_context *atom_ctx = adev->mode_info.atom_context;
+
+	/* VBIOS is of the format ###-DXXXYY-##. For SKU identification,
+	 * we can use just the "DXXX" portion. If there were more models, we
+	 * could convert the 3 characters to a hex integer and use a switch
+	 * for ease/speed/readability. For now, 2 string comparisons are
+	 * reasonable and not too expensive
+	 */
+	switch (adev->asic_type) {
+	case CHIP_VEGA20:
+		/* D161 and D163 are the VG20 server SKUs */
+		if (strnstr(atom_ctx->vbios_version, "D161",
+			    sizeof(atom_ctx->vbios_version)) ||
+		    strnstr(atom_ctx->vbios_version, "D163",
+			    sizeof(atom_ctx->vbios_version)))
+			return true;
+		else
+			return false;
+	default:
+		return false;
+	}
 }
 
 static int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h
index f29a8611d69b..1308d976d60e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h
@@ -26,4 +26,4 @@
 
 int amdgpu_fru_get_product_info(struct amdgpu_device *adev);
 
-#endif  // __AMDGPU_PRODINFO_H__
+#endif  // __AMDGPU_FRU_EEPROM_H__
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 7f9e50247413..aa7f230c71bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -93,7 +93,7 @@ retry:
 
 void amdgpu_gem_force_release(struct amdgpu_device *adev)
 {
-	struct drm_device *ddev = adev->ddev;
+	struct drm_device *ddev = adev_to_drm(adev);
 	struct drm_file *file;
 
 	mutex_lock(&ddev->filelist_mutex);
@@ -217,7 +217,7 @@ out_unlock:
 int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 			    struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	struct amdgpu_vm *vm = &fpriv->vm;
 	union drm_amdgpu_gem_create *args = data;
@@ -298,7 +298,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *filp)
 {
 	struct ttm_operation_ctx ctx = { true, false };
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_amdgpu_gem_userptr *args = data;
 	struct drm_gem_object *gobj;
 	struct amdgpu_bo *bo;
@@ -332,7 +332,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 	bo = gem_to_amdgpu_bo(gobj);
 	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);
+	r = amdgpu_ttm_tt_set_userptr(&bo->tbo, args->addr, args->flags);
 	if (r)
 		goto release_object;
 
@@ -587,7 +587,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 
 	struct drm_amdgpu_gem_va *args = data;
 	struct drm_gem_object *gobj;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	struct amdgpu_bo *abo;
 	struct amdgpu_bo_va *bo_va;
@@ -711,7 +711,7 @@ error_unref:
 int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_amdgpu_gem_op *args = data;
 	struct drm_gem_object *gobj;
 	struct amdgpu_vm_bo_base *base;
@@ -788,7 +788,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 			    struct drm_device *dev,
 			    struct drm_mode_create_dumb *args)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_gem_object *gobj;
 	uint32_t handle;
 	u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 78d37f92c7be..8c9bacfdbc30 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -202,40 +202,29 @@ bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev,
 
 void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev)
 {
-	int i, queue, pipe, mec;
+	int i, queue, pipe;
 	bool multipipe_policy = amdgpu_gfx_is_multipipe_capable(adev);
-
-	/* policy for amdgpu compute queue ownership */
-	for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) {
-		queue = i % adev->gfx.mec.num_queue_per_pipe;
-		pipe = (i / adev->gfx.mec.num_queue_per_pipe)
-			% adev->gfx.mec.num_pipe_per_mec;
-		mec = (i / adev->gfx.mec.num_queue_per_pipe)
-			/ adev->gfx.mec.num_pipe_per_mec;
-
-		/* we've run out of HW */
-		if (mec >= adev->gfx.mec.num_mec)
-			break;
-
-		if (multipipe_policy) {
-			/* policy: amdgpu owns the first two queues of the first MEC */
-			if (mec == 0 && queue < 2)
-				set_bit(i, adev->gfx.mec.queue_bitmap);
-		} else {
-			/* policy: amdgpu owns all queues in the first pipe */
-			if (mec == 0 && pipe == 0)
-				set_bit(i, adev->gfx.mec.queue_bitmap);
+	int max_queues_per_mec = min(adev->gfx.mec.num_pipe_per_mec *
+				     adev->gfx.mec.num_queue_per_pipe,
+				     adev->gfx.num_compute_rings);
+
+	if (multipipe_policy) {
+		/* policy: make queues evenly cross all pipes on MEC1 only */
+		for (i = 0; i < max_queues_per_mec; i++) {
+			pipe = i % adev->gfx.mec.num_pipe_per_mec;
+			queue = (i / adev->gfx.mec.num_pipe_per_mec) %
+				adev->gfx.mec.num_queue_per_pipe;
+
+			set_bit(pipe * adev->gfx.mec.num_queue_per_pipe + queue,
+					adev->gfx.mec.queue_bitmap);
 		}
+	} else {
+		/* policy: amdgpu owns all queues in the given pipe */
+		for (i = 0; i < max_queues_per_mec; ++i)
+			set_bit(i, adev->gfx.mec.queue_bitmap);
 	}
 
-	/* update the number of active compute rings */
-	adev->gfx.num_compute_rings =
-		bitmap_weight(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
-
-	/* If you hit this case and edited the policy, you probably just
-	 * need to increase AMDGPU_MAX_COMPUTE_RINGS */
-	if (WARN_ON(adev->gfx.num_compute_rings > AMDGPU_MAX_COMPUTE_RINGS))
-		adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS;
+	dev_dbg(adev->dev, "mec queue bitmap weight=%d\n", bitmap_weight(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES));
 }
 
 void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
@@ -571,8 +560,14 @@ void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable)
 	if (enable && !adev->gfx.gfx_off_state && !adev->gfx.gfx_off_req_count) {
 		schedule_delayed_work(&adev->gfx.gfx_off_delay_work, GFX_OFF_DELAY_ENABLE);
 	} else if (!enable && adev->gfx.gfx_off_state) {
-		if (!amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, false))
+		if (!amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, false)) {
 			adev->gfx.gfx_off_state = false;
+
+			if (adev->gfx.funcs->init_spm_golden) {
+				dev_dbg(adev->dev, "GFXOFF is disabled, re-init SPM golden settings\n");
+				amdgpu_gfx_init_spm_golden(adev);
+			}
+		}
 	}
 
 	mutex_unlock(&adev->gfx.gfx_off_mutex);
@@ -698,6 +693,9 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
 	struct amdgpu_ring *ring = &kiq->ring;
 
+	if (adev->in_pci_err_recovery)
+		return 0;
+
 	BUG_ON(!ring->funcs->emit_rreg);
 
 	spin_lock_irqsave(&kiq->ring_lock, flags);
@@ -724,7 +722,7 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
 	 *
 	 * also don't wait anymore for IRQ context
 	 * */
-	if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
+	if (r < 1 && (amdgpu_in_reset(adev) || in_interrupt()))
 		goto failed_kiq_read;
 
 	might_sleep();
@@ -748,7 +746,7 @@ failed_unlock:
 failed_kiq_read:
 	if (reg_val_offs)
 		amdgpu_device_wb_free(adev, reg_val_offs);
-	pr_err("failed to read reg:%x\n", reg);
+	dev_err(adev->dev, "failed to read reg:%x\n", reg);
 	return ~0;
 }
 
@@ -762,6 +760,9 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
 
 	BUG_ON(!ring->funcs->emit_wreg);
 
+	if (adev->in_pci_err_recovery)
+		return;
+
 	spin_lock_irqsave(&kiq->ring_lock, flags);
 	amdgpu_ring_alloc(ring, 32);
 	amdgpu_ring_emit_wreg(ring, reg, v);
@@ -782,7 +783,7 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
 	 *
 	 * also don't wait anymore for IRQ context
 	 * */
-	if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
+	if (r < 1 && (amdgpu_in_reset(adev) || in_interrupt()))
 		goto failed_kiq_write;
 
 	might_sleep();
@@ -801,5 +802,5 @@ failed_undo:
 	amdgpu_ring_undo(ring);
 	spin_unlock_irqrestore(&kiq->ring_lock, flags);
 failed_kiq_write:
-	pr_err("failed to write reg:%x\n", reg);
+	dev_err(adev->dev, "failed to write reg:%x\n", reg);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 1e7a2b0997c5..258498cbf1eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -216,6 +216,8 @@ struct amdgpu_gfx_funcs {
 	int (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if);
 	int (*query_ras_error_count) (struct amdgpu_device *adev, void *ras_error_status);
 	void (*reset_ras_error_count) (struct amdgpu_device *adev);
+	void (*init_spm_golden)(struct amdgpu_device *adev);
+	void (*query_ras_error_status) (struct amdgpu_device *adev);
 };
 
 struct sq_work {
@@ -324,6 +326,7 @@ struct amdgpu_gfx {
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance))
 #define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid))
+#define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev))
 
 /**
  * amdgpu_gfx_create_bitmask - create a bitmask
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfxhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfxhub.h
new file mode 100644
index 000000000000..66ebc2e3b2ad
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfxhub.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright 2020 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"),
+ * 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.
+ *
+ */
+#ifndef __AMDGPU_GFXHUB_H__
+#define __AMDGPU_GFXHUB_H__
+
+struct amdgpu_gfxhub_funcs {
+	u64 (*get_fb_location)(struct amdgpu_device *adev);
+	u64 (*get_mc_fb_offset)(struct amdgpu_device *adev);
+	void (*setup_vm_pt_regs)(struct amdgpu_device *adev, uint32_t vmid,
+			uint64_t page_table_base);
+	int (*gart_enable)(struct amdgpu_device *adev);
+
+	void (*gart_disable)(struct amdgpu_device *adev);
+	void (*set_fault_enable_default)(struct amdgpu_device *adev, bool value);
+	void (*init)(struct amdgpu_device *adev);
+	int (*get_xgmi_info)(struct amdgpu_device *adev);
+};
+
+struct amdgpu_gfxhub {
+	const struct amdgpu_gfxhub_funcs *funcs;
+};
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 34cbd6f6a56b..36604d751d62 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -27,6 +27,7 @@
 #include <linux/io-64-nonatomic-lo-hi.h>
 
 #include "amdgpu.h"
+#include "amdgpu_gmc.h"
 #include "amdgpu_ras.h"
 #include "amdgpu_xgmi.h"
 
@@ -411,3 +412,102 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev)
 		break;
 	}
 }
+
+/**
+ * amdgpu_noretry_set -- set per asic noretry defaults
+ * @adev: amdgpu_device pointer
+ *
+ * Set a per asic default for the no-retry parameter.
+ *
+ */
+void amdgpu_gmc_noretry_set(struct amdgpu_device *adev)
+{
+	struct amdgpu_gmc *gmc = &adev->gmc;
+
+	switch (adev->asic_type) {
+	case CHIP_RAVEN:
+		/* Raven currently has issues with noretry
+		 * regardless of what we decide for other
+		 * asics, we should leave raven with
+		 * noretry = 0 until we root cause the
+		 * issues.
+		 */
+		if (amdgpu_noretry == -1)
+			gmc->noretry = 0;
+		else
+			gmc->noretry = amdgpu_noretry;
+		break;
+	default:
+		/* default this to 0 for now, but we may want
+		 * to change this in the future for certain
+		 * GPUs as it can increase performance in
+		 * certain cases.
+		 */
+		if (amdgpu_noretry == -1)
+			gmc->noretry = 0;
+		else
+			gmc->noretry = amdgpu_noretry;
+		break;
+	}
+}
+
+void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type,
+				   bool enable)
+{
+	struct amdgpu_vmhub *hub;
+	u32 tmp, reg, i;
+
+	hub = &adev->vmhub[hub_type];
+	for (i = 0; i < 16; i++) {
+		reg = hub->vm_context0_cntl + hub->ctx_distance * i;
+
+		tmp = RREG32(reg);
+		if (enable)
+			tmp |= hub->vm_cntx_cntl_vm_fault;
+		else
+			tmp &= ~hub->vm_cntx_cntl_vm_fault;
+
+		WREG32(reg, tmp);
+	}
+}
+
+void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev)
+{
+	unsigned size;
+
+	/*
+	 * TODO:
+	 * Currently there is a bug where some memory client outside
+	 * of the driver writes to first 8M of VRAM on S3 resume,
+	 * this overrides GART which by default gets placed in first 8M and
+	 * causes VM_FAULTS once GTT is accessed.
+	 * Keep the stolen memory reservation until the while this is not solved.
+	 */
+	switch (adev->asic_type) {
+	case CHIP_VEGA10:
+	case CHIP_RAVEN:
+	case CHIP_RENOIR:
+		adev->mman.keep_stolen_vga_memory = true;
+		break;
+	default:
+		adev->mman.keep_stolen_vga_memory = false;
+		break;
+	}
+
+	if (!amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_DCE))
+		size = 0;
+	else
+		size = amdgpu_gmc_get_vbios_fb_size(adev);
+
+	/* set to 0 if the pre-OS buffer uses up most of vram */
+	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
+		size = 0;
+
+	if (size > AMDGPU_VBIOS_VGA_ALLOCATION) {
+		adev->mman.stolen_vga_size = AMDGPU_VBIOS_VGA_ALLOCATION;
+		adev->mman.stolen_extended_size = size - adev->mman.stolen_vga_size;
+	} else {
+		adev->mman.stolen_vga_size = size;
+		adev->mman.stolen_extended_size = 0;
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index acdb61cfa24c..aa0c83776ce0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -74,6 +74,12 @@ struct amdgpu_gmc_fault {
 /*
  * VMHUB structures, functions & helpers
  */
+struct amdgpu_vmhub_funcs {
+	void (*print_l2_protection_fault_status)(struct amdgpu_device *adev,
+						 uint32_t status);
+	uint32_t (*get_invalidate_req)(unsigned int vmid, uint32_t flush_type);
+};
+
 struct amdgpu_vmhub {
 	uint32_t	ctx0_ptb_addr_lo32;
 	uint32_t	ctx0_ptb_addr_hi32;
@@ -92,6 +98,10 @@ struct amdgpu_vmhub {
 	uint32_t	ctx_addr_distance; /* include LO32/HI32 */
 	uint32_t	eng_distance;
 	uint32_t	eng_addr_distance; /* include LO32/HI32 */
+
+	uint32_t	vm_cntx_cntl_vm_fault;
+
+	const struct amdgpu_vmhub_funcs *vmhub_funcs;
 };
 
 /*
@@ -121,6 +131,8 @@ struct amdgpu_gmc_funcs {
 	void (*get_vm_pte)(struct amdgpu_device *adev,
 			   struct amdgpu_bo_va_mapping *mapping,
 			   uint64_t *flags);
+	/* get the amount of memory used by the vbios for pre-OS console */
+	unsigned int (*get_vbios_fb_size)(struct amdgpu_device *adev);
 };
 
 struct amdgpu_xgmi {
@@ -203,7 +215,6 @@ struct amdgpu_gmc {
 	uint8_t			vram_vendor;
 	uint32_t                srbm_soft_reset;
 	bool			prt_warning;
-	uint64_t		stolen_size;
 	uint32_t		sdpif_register;
 	/* apertures */
 	u64			shared_aperture_start;
@@ -228,6 +239,7 @@ struct amdgpu_gmc {
 
 	struct amdgpu_xgmi xgmi;
 	struct amdgpu_irq_src	ecc_irq;
+	int noretry;
 };
 
 #define amdgpu_gmc_flush_gpu_tlb(adev, vmid, vmhub, type) ((adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (vmhub), (type)))
@@ -239,6 +251,7 @@ struct amdgpu_gmc {
 #define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags))
 #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags))
 #define amdgpu_gmc_get_vm_pte(adev, mapping, flags) (adev)->gmc.gmc_funcs->get_vm_pte((adev), (mapping), (flags))
+#define amdgpu_gmc_get_vbios_fb_size(adev) (adev)->gmc.gmc_funcs->get_vbios_fb_size((adev))
 
 /**
  * amdgpu_gmc_vram_full_visible - Check if full VRAM is visible through the BAR
@@ -288,5 +301,12 @@ void amdgpu_gmc_ras_fini(struct amdgpu_device *adev);
 int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev);
 
 extern void amdgpu_gmc_tmz_set(struct amdgpu_device *adev);
+extern void amdgpu_gmc_noretry_set(struct amdgpu_device *adev);
+
+extern void
+amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type,
+			      bool enable);
+
+void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
index 77fae40197ab..f203e4a6a3f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
@@ -24,11 +24,10 @@
 
 #include "amdgpu.h"
 
-struct amdgpu_gtt_mgr {
-	struct drm_mm mm;
-	spinlock_t lock;
-	atomic64_t available;
-};
+static inline struct amdgpu_gtt_mgr *to_gtt_mgr(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct amdgpu_gtt_mgr, manager);
+}
 
 struct amdgpu_gtt_node {
 	struct drm_mm_node node;
@@ -47,10 +46,11 @@ static ssize_t amdgpu_mem_info_gtt_total_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_TT);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n",
-			(adev->mman.bdev.man[TTM_PL_TT].size) * PAGE_SIZE);
+			man->size * PAGE_SIZE);
 }
 
 /**
@@ -65,10 +65,11 @@ static ssize_t amdgpu_mem_info_gtt_used_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_TT);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n",
-			amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]));
+			amdgpu_gtt_mgr_usage(man));
 }
 
 static DEVICE_ATTR(mem_info_gtt_total, S_IRUGO,
@@ -76,6 +77,7 @@ static DEVICE_ATTR(mem_info_gtt_total, S_IRUGO,
 static DEVICE_ATTR(mem_info_gtt_used, S_IRUGO,
 	           amdgpu_mem_info_gtt_used_show, NULL);
 
+static const struct ttm_resource_manager_func amdgpu_gtt_mgr_func;
 /**
  * amdgpu_gtt_mgr_init - init GTT manager and DRM MM
  *
@@ -84,24 +86,23 @@ static DEVICE_ATTR(mem_info_gtt_used, S_IRUGO,
  *
  * Allocate and initialize the GTT manager.
  */
-static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
-			       unsigned long p_size)
+int amdgpu_gtt_mgr_init(struct amdgpu_device *adev, uint64_t gtt_size)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_gtt_mgr *mgr;
+	struct amdgpu_gtt_mgr *mgr = &adev->mman.gtt_mgr;
+	struct ttm_resource_manager *man = &mgr->manager;
 	uint64_t start, size;
 	int ret;
 
-	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
-	if (!mgr)
-		return -ENOMEM;
+	man->use_tt = true;
+	man->func = &amdgpu_gtt_mgr_func;
+
+	ttm_resource_manager_init(man, gtt_size >> PAGE_SHIFT);
 
 	start = AMDGPU_GTT_MAX_TRANSFER_SIZE * AMDGPU_GTT_NUM_TRANSFER_WINDOWS;
 	size = (adev->gmc.gart_size >> PAGE_SHIFT) - start;
 	drm_mm_init(&mgr->mm, start, size);
 	spin_lock_init(&mgr->lock);
-	atomic64_set(&mgr->available, p_size);
-	man->priv = mgr;
+	atomic64_set(&mgr->available, gtt_size >> PAGE_SHIFT);
 
 	ret = device_create_file(adev->dev, &dev_attr_mem_info_gtt_total);
 	if (ret) {
@@ -114,6 +115,8 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
 		return ret;
 	}
 
+	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_TT, &mgr->manager);
+	ttm_resource_manager_set_used(man, true);
 	return 0;
 }
 
@@ -125,20 +128,27 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
  * Destroy and free the GTT manager, returns -EBUSY if ranges are still
  * allocated inside it.
  */
-static int amdgpu_gtt_mgr_fini(struct ttm_mem_type_manager *man)
+void amdgpu_gtt_mgr_fini(struct amdgpu_device *adev)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = &adev->mman.gtt_mgr;
+	struct ttm_resource_manager *man = &mgr->manager;
+	int ret;
+
+	ttm_resource_manager_set_used(man, false);
+
+	ret = ttm_resource_manager_force_list_clean(&adev->mman.bdev, man);
+	if (ret)
+		return;
+
 	spin_lock(&mgr->lock);
 	drm_mm_takedown(&mgr->mm);
 	spin_unlock(&mgr->lock);
-	kfree(mgr);
-	man->priv = NULL;
 
 	device_remove_file(adev->dev, &dev_attr_mem_info_gtt_total);
 	device_remove_file(adev->dev, &dev_attr_mem_info_gtt_used);
 
-	return 0;
+	ttm_resource_manager_cleanup(man);
+	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_TT, NULL);
 }
 
 /**
@@ -148,7 +158,7 @@ static int amdgpu_gtt_mgr_fini(struct ttm_mem_type_manager *man)
  *
  * Check if a mem object has already address space allocated.
  */
-bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_mem_reg *mem)
+bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_resource *mem)
 {
 	return mem->mm_node != NULL;
 }
@@ -163,12 +173,12 @@ bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_mem_reg *mem)
  *
  * Dummy, allocate the node but no space for it yet.
  */
-static int amdgpu_gtt_mgr_new(struct ttm_mem_type_manager *man,
+static int amdgpu_gtt_mgr_new(struct ttm_resource_manager *man,
 			      struct ttm_buffer_object *tbo,
 			      const struct ttm_place *place,
-			      struct ttm_mem_reg *mem)
+			      struct ttm_resource *mem)
 {
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = to_gtt_mgr(man);
 	struct amdgpu_gtt_node *node;
 	int r;
 
@@ -226,10 +236,10 @@ err_out:
  *
  * Free the allocated GTT again.
  */
-static void amdgpu_gtt_mgr_del(struct ttm_mem_type_manager *man,
-			       struct ttm_mem_reg *mem)
+static void amdgpu_gtt_mgr_del(struct ttm_resource_manager *man,
+			       struct ttm_resource *mem)
 {
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = to_gtt_mgr(man);
 	struct amdgpu_gtt_node *node = mem->mm_node;
 
 	if (node) {
@@ -249,17 +259,17 @@ static void amdgpu_gtt_mgr_del(struct ttm_mem_type_manager *man,
  *
  * Return how many bytes are used in the GTT domain
  */
-uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man)
+uint64_t amdgpu_gtt_mgr_usage(struct ttm_resource_manager *man)
 {
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = to_gtt_mgr(man);
 	s64 result = man->size - atomic64_read(&mgr->available);
 
 	return (result > 0 ? result : 0) * PAGE_SIZE;
 }
 
-int amdgpu_gtt_mgr_recover(struct ttm_mem_type_manager *man)
+int amdgpu_gtt_mgr_recover(struct ttm_resource_manager *man)
 {
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = to_gtt_mgr(man);
 	struct amdgpu_gtt_node *node;
 	struct drm_mm_node *mm_node;
 	int r = 0;
@@ -284,10 +294,10 @@ int amdgpu_gtt_mgr_recover(struct ttm_mem_type_manager *man)
  *
  * Dump the table content using printk.
  */
-static void amdgpu_gtt_mgr_debug(struct ttm_mem_type_manager *man,
+static void amdgpu_gtt_mgr_debug(struct ttm_resource_manager *man,
 				 struct drm_printer *printer)
 {
-	struct amdgpu_gtt_mgr *mgr = man->priv;
+	struct amdgpu_gtt_mgr *mgr = to_gtt_mgr(man);
 
 	spin_lock(&mgr->lock);
 	drm_mm_print(&mgr->mm, printer);
@@ -298,10 +308,8 @@ static void amdgpu_gtt_mgr_debug(struct ttm_mem_type_manager *man,
 		   amdgpu_gtt_mgr_usage(man) >> 20);
 }
 
-const struct ttm_mem_type_manager_func amdgpu_gtt_mgr_func = {
-	.init = amdgpu_gtt_mgr_init,
-	.takedown = amdgpu_gtt_mgr_fini,
-	.get_node = amdgpu_gtt_mgr_new,
-	.put_node = amdgpu_gtt_mgr_del,
+static const struct ttm_resource_manager_func amdgpu_gtt_mgr_func = {
+	.alloc = amdgpu_gtt_mgr_new,
+	.free = amdgpu_gtt_mgr_del,
 	.debug = amdgpu_gtt_mgr_debug
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
index 70dbe343f51d..47cad23a6b9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
@@ -40,7 +40,7 @@
 static int amdgpu_i2c_pre_xfer(struct i2c_adapter *i2c_adap)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_get_adapdata(i2c_adap);
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t temp;
 
@@ -82,7 +82,7 @@ static int amdgpu_i2c_pre_xfer(struct i2c_adapter *i2c_adap)
 static void amdgpu_i2c_post_xfer(struct i2c_adapter *i2c_adap)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_get_adapdata(i2c_adap);
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t temp;
 
@@ -101,7 +101,7 @@ static void amdgpu_i2c_post_xfer(struct i2c_adapter *i2c_adap)
 static int amdgpu_i2c_get_clock(void *i2c_priv)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_priv;
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t val;
 
@@ -116,7 +116,7 @@ static int amdgpu_i2c_get_clock(void *i2c_priv)
 static int amdgpu_i2c_get_data(void *i2c_priv)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_priv;
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t val;
 
@@ -130,7 +130,7 @@ static int amdgpu_i2c_get_data(void *i2c_priv)
 static void amdgpu_i2c_set_clock(void *i2c_priv, int clock)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_priv;
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t val;
 
@@ -143,7 +143,7 @@ static void amdgpu_i2c_set_clock(void *i2c_priv, int clock)
 static void amdgpu_i2c_set_data(void *i2c_priv, int data)
 {
 	struct amdgpu_i2c_chan *i2c = i2c_priv;
-	struct amdgpu_device *adev = i2c->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(i2c->dev);
 	struct amdgpu_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t val;
 
@@ -253,7 +253,7 @@ void amdgpu_i2c_add(struct amdgpu_device *adev,
 		    const struct amdgpu_i2c_bus_rec *rec,
 		    const char *name)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	int i;
 
 	for (i = 0; i < AMDGPU_MAX_I2C_BUS; i++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index dcd492170598..2f53fa0ae9a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -445,7 +445,7 @@ static int amdgpu_debugfs_sa_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	seq_printf(m, "--------------------- DELAYED --------------------- \n");
 	amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DELAYED],
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 0cc4c67f95f7..300ac73b4738 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -85,7 +85,7 @@ static void amdgpu_hotplug_work_func(struct work_struct *work)
 {
 	struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
 						  hotplug_work);
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
@@ -151,7 +151,7 @@ void amdgpu_irq_disable_all(struct amdgpu_device *adev)
 irqreturn_t amdgpu_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	irqreturn_t ret;
 
 	ret = amdgpu_ih_process(adev, &adev->irq.ih);
@@ -268,9 +268,9 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 		if (!adev->enable_virtual_display)
 			/* Disable vblank IRQs aggressively for power-saving */
 			/* XXX: can this be enabled for DC? */
-			adev->ddev->vblank_disable_immediate = true;
+			adev_to_drm(adev)->vblank_disable_immediate = true;
 
-		r = drm_vblank_init(adev->ddev, adev->mode_info.num_crtc);
+		r = drm_vblank_init(adev_to_drm(adev), adev->mode_info.num_crtc);
 		if (r)
 			return r;
 
@@ -284,14 +284,14 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 
 	adev->irq.installed = true;
 	/* Use vector 0 for MSI-X */
-	r = drm_irq_install(adev->ddev, pci_irq_vector(adev->pdev, 0));
+	r = drm_irq_install(adev_to_drm(adev), pci_irq_vector(adev->pdev, 0));
 	if (r) {
 		adev->irq.installed = false;
 		if (!amdgpu_device_has_dc_support(adev))
 			flush_work(&adev->hotplug_work);
 		return r;
 	}
-	adev->ddev->max_vblank_count = 0x00ffffff;
+	adev_to_drm(adev)->max_vblank_count = 0x00ffffff;
 
 	DRM_DEBUG("amdgpu: irq initialized.\n");
 	return 0;
@@ -311,7 +311,7 @@ void amdgpu_irq_fini(struct amdgpu_device *adev)
 	unsigned i, j;
 
 	if (adev->irq.installed) {
-		drm_irq_uninstall(adev->ddev);
+		drm_irq_uninstall(adev_to_drm(adev));
 		adev->irq.installed = false;
 		if (adev->irq.msi_enabled)
 			pci_free_irq_vectors(adev->pdev);
@@ -522,7 +522,7 @@ void amdgpu_irq_gpu_reset_resume_helper(struct amdgpu_device *adev)
 int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		   unsigned type)
 {
-	if (!adev->ddev->irq_enabled)
+	if (!adev_to_drm(adev)->irq_enabled)
 		return -ENOENT;
 
 	if (type >= src->num_types)
@@ -552,7 +552,7 @@ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		   unsigned type)
 {
-	if (!adev->ddev->irq_enabled)
+	if (!adev_to_drm(adev)->irq_enabled)
 		return -ENOENT;
 
 	if (type >= src->num_types)
@@ -583,7 +583,7 @@ int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 bool amdgpu_irq_enabled(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 			unsigned type)
 {
-	if (!adev->ddev->irq_enabled)
+	if (!adev_to_drm(adev)->irq_enabled)
 		return false;
 
 	if (type >= src->num_types)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 937029ad5271..dcfe8a3b03ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -251,7 +251,7 @@ void amdgpu_job_stop_all_jobs_on_sched(struct drm_gpu_scheduler *sched)
 	int i;
 
 	/* Signal all jobs not yet scheduled */
-	for (i = DRM_SCHED_PRIORITY_MAX - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
+	for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
 		struct drm_sched_rq *rq = &sched->sched_rq[i];
 
 		if (!rq)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index b403b2a88ee5..efda38349a03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -78,7 +78,7 @@ void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev)
  */
 void amdgpu_driver_unload_kms(struct drm_device *dev)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (adev == NULL)
 		return;
@@ -86,7 +86,7 @@ void amdgpu_driver_unload_kms(struct drm_device *dev)
 	amdgpu_unregister_gpu_instance(adev);
 
 	if (adev->rmmio == NULL)
-		goto done_free;
+		return;
 
 	if (adev->runpm) {
 		pm_runtime_get_sync(dev->dev);
@@ -94,12 +94,7 @@ void amdgpu_driver_unload_kms(struct drm_device *dev)
 	}
 
 	amdgpu_acpi_fini(adev);
-
 	amdgpu_device_fini(adev);
-
-done_free:
-	kfree(adev);
-	dev->dev_private = NULL;
 }
 
 void amdgpu_register_gpu_instance(struct amdgpu_device *adev)
@@ -130,22 +125,18 @@ void amdgpu_register_gpu_instance(struct amdgpu_device *adev)
 /**
  * amdgpu_driver_load_kms - Main load function for KMS.
  *
- * @dev: drm dev pointer
+ * @adev: pointer to struct amdgpu_device
  * @flags: device flags
  *
  * This is the main load function for KMS (all asics).
  * Returns 0 on success, error on failure.
  */
-int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
+int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags)
 {
-	struct amdgpu_device *adev;
+	struct drm_device *dev;
 	int r, acpi_status;
 
-	adev = kzalloc(sizeof(struct amdgpu_device), GFP_KERNEL);
-	if (adev == NULL) {
-		return -ENOMEM;
-	}
-	dev->dev_private = (void *)adev;
+	dev = adev_to_drm(adev);
 
 	if (amdgpu_has_atpx() &&
 	    (amdgpu_is_atpx_hybrid() ||
@@ -160,7 +151,7 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
 	 * properly initialize the GPU MC controller and permit
 	 * VRAM allocation
 	 */
-	r = amdgpu_device_init(adev, dev, dev->pdev, flags);
+	r = amdgpu_device_init(adev, flags);
 	if (r) {
 		dev_err(&dev->pdev->dev, "Fatal error during GPU init\n");
 		goto out;
@@ -186,7 +177,7 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
 			break;
 		case CHIP_VEGA10:
 			/* turn runpm on if noretry=0 */
-			if (!amdgpu_noretry)
+			if (!adev->gmc.noretry)
 				adev->runpm = true;
 			break;
 		default:
@@ -291,14 +282,25 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info,
 		fw_info->feature = 0;
 		break;
 	case AMDGPU_INFO_FW_TA:
-		if (query_fw->index > 1)
-			return -EINVAL;
-		if (query_fw->index == 0) {
+		switch (query_fw->index) {
+		case 0:
 			fw_info->ver = adev->psp.ta_fw_version;
 			fw_info->feature = adev->psp.ta_xgmi_ucode_version;
-		} else {
+			break;
+		case 1:
 			fw_info->ver = adev->psp.ta_fw_version;
 			fw_info->feature = adev->psp.ta_ras_ucode_version;
+			break;
+		case 2:
+			fw_info->ver = adev->psp.ta_fw_version;
+			fw_info->feature = adev->psp.ta_hdcp_ucode_version;
+			break;
+		case 3:
+			fw_info->ver = adev->psp.ta_fw_version;
+			fw_info->feature = adev->psp.ta_dtm_ucode_version;
+			break;
+		default:
+			return -EINVAL;
 		}
 		break;
 	case AMDGPU_INFO_FW_SDMA:
@@ -480,7 +482,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
  */
 static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_amdgpu_info *info = data;
 	struct amdgpu_mode_info *minfo = &adev->mode_info;
 	void __user *out = (void __user *)(uintptr_t)info->return_pointer;
@@ -595,13 +597,13 @@ 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 = amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+		ui64 = amdgpu_vram_mgr_usage(ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM));
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_VIS_VRAM_USAGE:
-		ui64 = amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+		ui64 = amdgpu_vram_mgr_vis_usage(ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM));
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_GTT_USAGE:
-		ui64 = amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]);
+		ui64 = amdgpu_gtt_mgr_usage(ttm_manager_type(&adev->mman.bdev, 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;
@@ -624,7 +626,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			min(adev->gmc.visible_vram_size -
 			    atomic64_read(&adev->visible_pin_size),
 			    vram_gtt.vram_size);
-		vram_gtt.gtt_size = adev->mman.bdev.man[TTM_PL_TT].size;
+		vram_gtt.gtt_size = ttm_manager_type(&adev->mman.bdev, TTM_PL_TT)->size;
 		vram_gtt.gtt_size *= PAGE_SIZE;
 		vram_gtt.gtt_size -= atomic64_read(&adev->gart_pin_size);
 		return copy_to_user(out, &vram_gtt,
@@ -632,14 +634,17 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 	}
 	case AMDGPU_INFO_MEMORY: {
 		struct drm_amdgpu_memory_info mem;
-
+		struct ttm_resource_manager *vram_man =
+			ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
+		struct ttm_resource_manager *gtt_man =
+			ttm_manager_type(&adev->mman.bdev, TTM_PL_TT);
 		memset(&mem, 0, sizeof(mem));
 		mem.vram.total_heap_size = adev->gmc.real_vram_size;
 		mem.vram.usable_heap_size = adev->gmc.real_vram_size -
 			atomic64_read(&adev->vram_pin_size) -
 			AMDGPU_VM_RESERVED_VRAM;
 		mem.vram.heap_usage =
-			amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+			amdgpu_vram_mgr_usage(vram_man);
 		mem.vram.max_allocation = mem.vram.usable_heap_size * 3 / 4;
 
 		mem.cpu_accessible_vram.total_heap_size =
@@ -649,16 +654,16 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			    atomic64_read(&adev->visible_pin_size),
 			    mem.vram.usable_heap_size);
 		mem.cpu_accessible_vram.heap_usage =
-			amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
+			amdgpu_vram_mgr_vis_usage(vram_man);
 		mem.cpu_accessible_vram.max_allocation =
 			mem.cpu_accessible_vram.usable_heap_size * 3 / 4;
 
-		mem.gtt.total_heap_size = adev->mman.bdev.man[TTM_PL_TT].size;
+		mem.gtt.total_heap_size = gtt_man->size;
 		mem.gtt.total_heap_size *= PAGE_SIZE;
 		mem.gtt.usable_heap_size = mem.gtt.total_heap_size -
 			atomic64_read(&adev->gart_pin_size);
 		mem.gtt.heap_usage =
-			amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]);
+			amdgpu_gtt_mgr_usage(gtt_man);
 		mem.gtt.max_allocation = mem.gtt.usable_heap_size * 3 / 4;
 
 		return copy_to_user(out, &mem,
@@ -742,6 +747,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			dev_info.ids_flags |= AMDGPU_IDS_FLAGS_FUSION;
 		if (amdgpu_mcbp || amdgpu_sriov_vf(adev))
 			dev_info.ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION;
+		if (amdgpu_is_tmz(adev))
+			dev_info.ids_flags |= AMDGPU_IDS_FLAGS_TMZ;
 
 		vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE;
 		vm_size -= AMDGPU_VA_RESERVED_SIZE;
@@ -995,7 +1002,7 @@ void amdgpu_driver_lastclose_kms(struct drm_device *dev)
  */
 int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv;
 	int r, pasid;
 
@@ -1080,7 +1087,7 @@ pm_put:
 void amdgpu_driver_postclose_kms(struct drm_device *dev,
 				 struct drm_file *file_priv)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
 	struct amdgpu_bo_list *list;
 	struct amdgpu_bo *pd;
@@ -1145,7 +1152,7 @@ u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	unsigned int pipe = crtc->index;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int vpos, hpos, stat;
 	u32 count;
 
@@ -1213,7 +1220,7 @@ int amdgpu_enable_vblank_kms(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	unsigned int pipe = crtc->index;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int idx = amdgpu_display_crtc_idx_to_irq_type(adev, pipe);
 
 	return amdgpu_irq_get(adev, &adev->crtc_irq, idx);
@@ -1230,7 +1237,7 @@ void amdgpu_disable_vblank_kms(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	unsigned int pipe = crtc->index;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int idx = amdgpu_display_crtc_idx_to_irq_type(adev, pipe);
 
 	amdgpu_irq_put(adev, &adev->crtc_irq, idx);
@@ -1266,7 +1273,7 @@ static int amdgpu_debugfs_firmware_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_amdgpu_info_firmware fw_info;
 	struct drm_amdgpu_query_fw query_fw;
 	struct atom_context *ctx = adev->mode_info.atom_context;
@@ -1389,13 +1396,31 @@ static int amdgpu_debugfs_firmware_info(struct seq_file *m, void *data)
 		   fw_info.feature, fw_info.ver);
 
 	query_fw.fw_type = AMDGPU_INFO_FW_TA;
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < 4; i++) {
 		query_fw.index = i;
 		ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);
 		if (ret)
 			continue;
-		seq_printf(m, "TA %s feature version: %u, firmware version: 0x%08x\n",
-				i ? "RAS" : "XGMI", fw_info.feature, fw_info.ver);
+		switch (query_fw.index) {
+		case 0:
+			seq_printf(m, "TA %s feature version: 0x%08x, firmware version: 0x%08x\n",
+					"RAS", fw_info.feature, fw_info.ver);
+			break;
+		case 1:
+			seq_printf(m, "TA %s feature version: 0x%08x, firmware version: 0x%08x\n",
+					"XGMI", fw_info.feature, fw_info.ver);
+			break;
+		case 2:
+			seq_printf(m, "TA %s feature version: 0x%08x, firmware version: 0x%08x\n",
+					"HDCP", fw_info.feature, fw_info.ver);
+			break;
+		case 3:
+			seq_printf(m, "TA %s feature version: 0x%08x, firmware version: 0x%08x\n",
+					"DTM", fw_info.feature, fw_info.ver);
+			break;
+		default:
+			return -EINVAL;
+		}
 	}
 
 	/* SMC */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
index e89fb35fec71..1ae9bdae7311 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
@@ -27,6 +27,20 @@ struct amdgpu_mmhub_funcs {
 	void (*query_ras_error_count)(struct amdgpu_device *adev,
 					void *ras_error_status);
 	void (*reset_ras_error_count)(struct amdgpu_device *adev);
+	u64 (*get_fb_location)(struct amdgpu_device *adev);
+	void (*init)(struct amdgpu_device *adev);
+	int (*gart_enable)(struct amdgpu_device *adev);
+	void (*set_fault_enable_default)(struct amdgpu_device *adev,
+			bool value);
+	void (*gart_disable)(struct amdgpu_device *adev);
+	int (*set_clockgating)(struct amdgpu_device *adev,
+			       enum amd_clockgating_state state);
+	void (*get_clockgating)(struct amdgpu_device *adev, u32 *flags);
+	void (*setup_vm_pt_regs)(struct amdgpu_device *adev, uint32_t vmid,
+				uint64_t page_table_base);
+	void (*update_power_gating)(struct amdgpu_device *adev,
+                                bool enable);
+	void (*query_ras_error_status)(struct amdgpu_device *adev);
 };
 
 struct amdgpu_mmhub {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 37ba07e2feb5..a04decb934b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -46,6 +46,7 @@
 
 #include <drm/drm_dp_mst_helper.h>
 #include "modules/inc/mod_freesync.h"
+#include "amdgpu_dm_irq_params.h"
 
 struct amdgpu_bo;
 struct amdgpu_device;
@@ -404,7 +405,8 @@ struct amdgpu_crtc {
 	struct amdgpu_flip_work *pflip_works;
 	enum amdgpu_flip_status pflip_status;
 	int deferred_flip_completion;
-	u32 last_flip_vblank;
+	/* parameters access from DM IRQ handler */
+	struct dm_irq_params dm_irq_params;
 	/* pll sharing */
 	struct amdgpu_atom_ss ss;
 	bool ss_enabled;
@@ -469,6 +471,7 @@ struct amdgpu_encoder {
 struct amdgpu_connector_atom_dig {
 	/* displayport */
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
+	u8 downstream_ports[DP_MAX_DOWNSTREAM_PORTS];
 	u8 dp_sink_type;
 	int dp_clock;
 	int dp_lane_count;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 5ac7b5561475..ac043baac05d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -136,8 +136,8 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
-			TTM_PL_FLAG_VRAM;
+		places[c].mem_type = TTM_PL_VRAM;
+		places[c].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED;
 
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
 			places[c].lpfn = visible_pfn;
@@ -152,7 +152,8 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 	if (domain & AMDGPU_GEM_DOMAIN_GTT) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_TT;
+		places[c].mem_type = TTM_PL_TT;
+		places[c].flags = 0;
 		if (flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 			places[c].flags |= TTM_PL_FLAG_WC |
 				TTM_PL_FLAG_UNCACHED;
@@ -164,7 +165,8 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 	if (domain & AMDGPU_GEM_DOMAIN_CPU) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_SYSTEM;
+		places[c].mem_type = TTM_PL_SYSTEM;
+		places[c].flags = 0;
 		if (flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 			places[c].flags |= TTM_PL_FLAG_WC |
 				TTM_PL_FLAG_UNCACHED;
@@ -176,28 +178,32 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 	if (domain & AMDGPU_GEM_DOMAIN_GDS) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_UNCACHED | AMDGPU_PL_FLAG_GDS;
+		places[c].mem_type = AMDGPU_PL_GDS;
+		places[c].flags = TTM_PL_FLAG_UNCACHED;
 		c++;
 	}
 
 	if (domain & AMDGPU_GEM_DOMAIN_GWS) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_UNCACHED | AMDGPU_PL_FLAG_GWS;
+		places[c].mem_type = AMDGPU_PL_GWS;
+		places[c].flags = TTM_PL_FLAG_UNCACHED;
 		c++;
 	}
 
 	if (domain & AMDGPU_GEM_DOMAIN_OA) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_FLAG_UNCACHED | AMDGPU_PL_FLAG_OA;
+		places[c].mem_type = AMDGPU_PL_OA;
+		places[c].flags = TTM_PL_FLAG_UNCACHED;
 		c++;
 	}
 
 	if (!c) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
-		places[c].flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM;
+		places[c].mem_type = TTM_PL_SYSTEM;
+		places[c].flags = TTM_PL_MASK_CACHING;
 		c++;
 	}
 
@@ -374,6 +380,9 @@ int amdgpu_bo_create_kernel_at(struct amdgpu_device *adev,
 	if (r)
 		return r;
 
+	if ((*bo_ptr) == NULL)
+		return 0;
+
 	/*
 	 * Remove the original mem node and create a new one at the request
 	 * position.
@@ -381,7 +390,7 @@ int amdgpu_bo_create_kernel_at(struct amdgpu_device *adev,
 	if (cpu_addr)
 		amdgpu_bo_kunmap(*bo_ptr);
 
-	ttm_bo_mem_put(&(*bo_ptr)->tbo, &(*bo_ptr)->tbo.mem);
+	ttm_resource_free(&(*bo_ptr)->tbo, &(*bo_ptr)->tbo.mem);
 
 	for (i = 0; i < (*bo_ptr)->placement.num_placement; ++i) {
 		(*bo_ptr)->placements[i].fpfn = offset >> PAGE_SHIFT;
@@ -442,14 +451,14 @@ void amdgpu_bo_free_kernel(struct amdgpu_bo **bo, u64 *gpu_addr,
 static bool amdgpu_bo_validate_size(struct amdgpu_device *adev,
 					  unsigned long size, u32 domain)
 {
-	struct ttm_mem_type_manager *man = NULL;
+	struct ttm_resource_manager *man = NULL;
 
 	/*
 	 * If GTT is part of requested domains the check must succeed to
 	 * allow fall back to GTT
 	 */
 	if (domain & AMDGPU_GEM_DOMAIN_GTT) {
-		man = &adev->mman.bdev.man[TTM_PL_TT];
+		man = ttm_manager_type(&adev->mman.bdev, TTM_PL_TT);
 
 		if (size < (man->size << PAGE_SHIFT))
 			return true;
@@ -458,7 +467,7 @@ static bool amdgpu_bo_validate_size(struct amdgpu_device *adev,
 	}
 
 	if (domain & AMDGPU_GEM_DOMAIN_VRAM) {
-		man = &adev->mman.bdev.man[TTM_PL_VRAM];
+		man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 
 		if (size < (man->size << PAGE_SHIFT))
 			return true;
@@ -552,7 +561,7 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 	bo = kzalloc(sizeof(struct amdgpu_bo), GFP_KERNEL);
 	if (bo == NULL)
 		return -ENOMEM;
-	drm_gem_private_object_init(adev->ddev, &bo->tbo.base, size);
+	drm_gem_private_object_init(adev_to_drm(adev), &bo->tbo.base, size);
 	INIT_LIST_HEAD(&bo->shadow_list);
 	bo->vm_bo = NULL;
 	bo->preferred_domains = bp->preferred_domain ? bp->preferred_domain :
@@ -591,7 +600,7 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 		amdgpu_cs_report_moved_bytes(adev, ctx.bytes_moved, 0);
 
 	if (bp->flags & AMDGPU_GEM_CREATE_VRAM_CLEARED &&
-	    bo->tbo.mem.placement & TTM_PL_FLAG_VRAM) {
+	    bo->tbo.mem.mem_type == TTM_PL_VRAM) {
 		struct dma_fence *fence;
 
 		r = amdgpu_fill_buffer(bo, 0, bo->tbo.base.resv, &fence);
@@ -1268,11 +1277,11 @@ int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
  */
 void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 			   bool evict,
-			   struct ttm_mem_reg *new_mem)
+			   struct ttm_resource *new_mem)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct amdgpu_bo *abo;
-	struct ttm_mem_reg *old_mem = &bo->mem;
+	struct ttm_resource *old_mem = &bo->mem;
 
 	if (!amdgpu_bo_is_amdgpu_bo(bo))
 		return;
@@ -1299,7 +1308,7 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 }
 
 /**
- * amdgpu_bo_move_notify - notification about a BO being released
+ * amdgpu_bo_release_notify - notification about a BO being released
  * @bo: pointer to a buffer object
  *
  * Wipes VRAM buffers whose contents should not be leaked before the
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index afa5189dba7d..5ddb6cf96030 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -160,7 +160,7 @@ static inline int amdgpu_bo_reserve(struct amdgpu_bo *bo, bool no_intr)
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	int r;
 
-	r = __ttm_bo_reserve(&bo->tbo, !no_intr, false, NULL);
+	r = ttm_bo_reserve(&bo->tbo, !no_intr, false, NULL);
 	if (unlikely(r != 0)) {
 		if (r != -ERESTARTSYS)
 			dev_err(adev->dev, "%p reserve failed\n", bo);
@@ -283,7 +283,7 @@ int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
 			   uint64_t *flags);
 void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 			   bool evict,
-			   struct ttm_mem_reg *new_mem);
+			   struct ttm_resource *new_mem);
 void amdgpu_bo_release_notify(struct ttm_buffer_object *bo);
 int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c
index 1311d6aec5d4..69af462db34d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c
@@ -226,7 +226,7 @@ static int init_pmu_by_type(struct amdgpu_device *adev,
 	pmu_entry->pmu.attr_groups = attr_groups;
 	pmu_entry->pmu_perf_type = pmu_perf_type;
 	snprintf(pmu_name, PMU_NAME_SIZE, "%s_%d",
-				pmu_file_prefix, adev->ddev->primary->index);
+				pmu_file_prefix, adev_to_drm(adev)->primary->index);
 
 	ret = perf_pmu_register(&pmu_entry->pmu, pmu_name, -1);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 06757681b2ce..18be544d8c1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -161,10 +161,12 @@ static int psp_sw_init(void *handle)
 	struct psp_context *psp = &adev->psp;
 	int ret;
 
-	ret = psp_init_microcode(psp);
-	if (ret) {
-		DRM_ERROR("Failed to load psp firmware!\n");
-		return ret;
+	if (!amdgpu_sriov_vf(adev)) {
+		ret = psp_init_microcode(psp);
+		if (ret) {
+			DRM_ERROR("Failed to load psp firmware!\n");
+			return ret;
+		}
 	}
 
 	ret = psp_memory_training_init(psp);
@@ -219,6 +221,9 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index,
 	int i;
 	struct amdgpu_device *adev = psp->adev;
 
+	if (psp->adev->in_pci_err_recovery)
+		return 0;
+
 	for (i = 0; i < adev->usec_timeout; i++) {
 		val = RREG32(reg_index);
 		if (check_changed) {
@@ -245,6 +250,9 @@ psp_cmd_submit_buf(struct psp_context *psp,
 	bool ras_intr = false;
 	bool skip_unsupport = false;
 
+	if (psp->adev->in_pci_err_recovery)
+		return 0;
+
 	mutex_lock(&psp->mutex);
 
 	memset(psp->cmd_buf_mem, 0, PSP_CMD_BUFFER_SIZE);
@@ -929,6 +937,7 @@ static int psp_ras_load(struct psp_context *psp)
 {
 	int ret;
 	struct psp_gfx_cmd_resp *cmd;
+	struct ta_ras_shared_memory *ras_cmd;
 
 	/*
 	 * TODO: bypass the loading in sriov for now
@@ -952,11 +961,20 @@ static int psp_ras_load(struct psp_context *psp)
 	ret = psp_cmd_submit_buf(psp, NULL, cmd,
 			psp->fence_buf_mc_addr);
 
+	ras_cmd = (struct ta_ras_shared_memory*)psp->ras.ras_shared_buf;
+
 	if (!ret) {
-		psp->ras.ras_initialized = true;
 		psp->ras.session_id = cmd->resp.session_id;
+
+		if (!ras_cmd->ras_status)
+			psp->ras.ras_initialized = true;
+		else
+			dev_warn(psp->adev->dev, "RAS Init Status: 0x%X\n", ras_cmd->ras_status);
 	}
 
+	if (ret || ras_cmd->ras_status)
+		amdgpu_ras_fini(psp->adev);
+
 	kfree(cmd);
 
 	return ret;
@@ -1429,6 +1447,168 @@ static int psp_dtm_terminate(struct psp_context *psp)
 }
 // DTM end
 
+// RAP start
+static int psp_rap_init_shared_buf(struct psp_context *psp)
+{
+	int ret;
+
+	/*
+	 * Allocate 16k memory aligned to 4k from Frame Buffer (local
+	 * physical) for rap ta <-> Driver
+	 */
+	ret = amdgpu_bo_create_kernel(psp->adev, PSP_RAP_SHARED_MEM_SIZE,
+				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				      &psp->rap_context.rap_shared_bo,
+				      &psp->rap_context.rap_shared_mc_addr,
+				      &psp->rap_context.rap_shared_buf);
+
+	return ret;
+}
+
+static int psp_rap_load(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	memset(psp->fw_pri_buf, 0, PSP_1_MEG);
+	memcpy(psp->fw_pri_buf, psp->ta_rap_start_addr, psp->ta_rap_ucode_size);
+
+	psp_prep_ta_load_cmd_buf(cmd,
+				 psp->fw_pri_mc_addr,
+				 psp->ta_rap_ucode_size,
+				 psp->rap_context.rap_shared_mc_addr,
+				 PSP_RAP_SHARED_MEM_SIZE);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
+
+	if (!ret) {
+		psp->rap_context.rap_initialized = true;
+		psp->rap_context.session_id = cmd->resp.session_id;
+		mutex_init(&psp->rap_context.mutex);
+	}
+
+	kfree(cmd);
+
+	return ret;
+}
+
+static int psp_rap_unload(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	psp_prep_ta_unload_cmd_buf(cmd, psp->rap_context.session_id);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
+
+	kfree(cmd);
+
+	return ret;
+}
+
+static int psp_rap_initialize(struct psp_context *psp)
+{
+	int ret;
+
+	/*
+	 * TODO: bypass the initialize in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	if (!psp->adev->psp.ta_rap_ucode_size ||
+	    !psp->adev->psp.ta_rap_start_addr) {
+		dev_info(psp->adev->dev, "RAP: optional rap ta ucode is not available\n");
+		return 0;
+	}
+
+	if (!psp->rap_context.rap_initialized) {
+		ret = psp_rap_init_shared_buf(psp);
+		if (ret)
+			return ret;
+	}
+
+	ret = psp_rap_load(psp);
+	if (ret)
+		return ret;
+
+	ret = psp_rap_invoke(psp, TA_CMD_RAP__INITIALIZE);
+	if (ret != TA_RAP_STATUS__SUCCESS) {
+		psp_rap_unload(psp);
+
+		amdgpu_bo_free_kernel(&psp->rap_context.rap_shared_bo,
+			      &psp->rap_context.rap_shared_mc_addr,
+			      &psp->rap_context.rap_shared_buf);
+
+		psp->rap_context.rap_initialized = false;
+
+		dev_warn(psp->adev->dev, "RAP TA initialize fail.\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int psp_rap_terminate(struct psp_context *psp)
+{
+	int ret;
+
+	if (!psp->rap_context.rap_initialized)
+		return 0;
+
+	ret = psp_rap_unload(psp);
+
+	psp->rap_context.rap_initialized = false;
+
+	/* free rap shared memory */
+	amdgpu_bo_free_kernel(&psp->rap_context.rap_shared_bo,
+			      &psp->rap_context.rap_shared_mc_addr,
+			      &psp->rap_context.rap_shared_buf);
+
+	return ret;
+}
+
+int psp_rap_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
+{
+	struct ta_rap_shared_memory *rap_cmd;
+	int ret;
+
+	if (!psp->rap_context.rap_initialized)
+		return -EINVAL;
+
+	if (ta_cmd_id != TA_CMD_RAP__INITIALIZE &&
+	    ta_cmd_id != TA_CMD_RAP__VALIDATE_L0)
+		return -EINVAL;
+
+	mutex_lock(&psp->rap_context.mutex);
+
+	rap_cmd = (struct ta_rap_shared_memory *)
+		  psp->rap_context.rap_shared_buf;
+	memset(rap_cmd, 0, sizeof(struct ta_rap_shared_memory));
+
+	rap_cmd->cmd_id = ta_cmd_id;
+	rap_cmd->validation_method_id = METHOD_A;
+
+	ret = psp_ta_invoke(psp, rap_cmd->cmd_id, psp->rap_context.session_id);
+	if (ret) {
+		mutex_unlock(&psp->rap_context.mutex);
+		return ret;
+	}
+
+	mutex_unlock(&psp->rap_context.mutex);
+
+	return rap_cmd->rap_status;
+}
+// RAP end
+
 static int psp_hw_start(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -1706,7 +1886,7 @@ static int psp_load_smu_fw(struct psp_context *psp)
 		return 0;
 
 
-	if (adev->in_gpu_reset && ras && ras->supported) {
+	if (amdgpu_in_reset(adev) && ras && ras->supported) {
 		ret = amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_UNLOAD);
 		if (ret) {
 			DRM_WARN("Failed to set MP1 state prepare for reload\n");
@@ -1821,7 +2001,7 @@ static int psp_load_fw(struct amdgpu_device *adev)
 	int ret;
 	struct psp_context *psp = &adev->psp;
 
-	if (amdgpu_sriov_vf(adev) && adev->in_gpu_reset) {
+	if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) {
 		psp_ring_stop(psp, PSP_RING_TYPE__KM); /* should not destroy ring, only stop */
 		goto skip_memalloc;
 	}
@@ -1891,6 +2071,11 @@ skip_memalloc:
 		if (ret)
 			dev_err(psp->adev->dev,
 				"DTM: Failed to initialize DTM\n");
+
+		ret = psp_rap_initialize(psp);
+		if (ret)
+			dev_err(psp->adev->dev,
+				"RAP: Failed to initialize RAP\n");
 	}
 
 	return 0;
@@ -1941,6 +2126,7 @@ static int psp_hw_fini(void *handle)
 
 	if (psp->adev->psp.ta_fw) {
 		psp_ras_terminate(psp);
+		psp_rap_terminate(psp);
 		psp_dtm_terminate(psp);
 		psp_hdcp_terminate(psp);
 	}
@@ -1999,6 +2185,11 @@ static int psp_suspend(void *handle)
 			DRM_ERROR("Failed to terminate dtm ta\n");
 			return ret;
 		}
+		ret = psp_rap_terminate(psp);
+		if (ret) {
+			DRM_ERROR("Failed to terminate rap ta\n");
+			return ret;
+		}
 	}
 
 	ret = psp_asd_unload(psp);
@@ -2077,6 +2268,11 @@ static int psp_resume(void *handle)
 		if (ret)
 			dev_err(psp->adev->dev,
 				"DTM: Failed to initialize DTM\n");
+
+		ret = psp_rap_initialize(psp);
+		if (ret)
+			dev_err(psp->adev->dev,
+				"RAP: Failed to initialize RAP\n");
 	}
 
 	mutex_unlock(&adev->firmware.mutex);
@@ -2342,6 +2538,11 @@ int parse_ta_bin_descriptor(struct psp_context *psp,
 		psp->ta_dtm_ucode_size     = le32_to_cpu(desc->size_bytes);
 		psp->ta_dtm_start_addr     = ucode_start_addr;
 		break;
+	case TA_FW_TYPE_PSP_RAP:
+		psp->ta_rap_ucode_version  = le32_to_cpu(desc->fw_version);
+		psp->ta_rap_ucode_size     = le32_to_cpu(desc->size_bytes);
+		psp->ta_rap_start_addr     = ucode_start_addr;
+		break;
 	default:
 		dev_warn(psp->adev->dev, "Unsupported TA type: %d\n", desc->fw_type);
 		break;
@@ -2420,7 +2621,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_read(struct device *dev,
 					 char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t fw_ver;
 	int ret;
 
@@ -2447,7 +2648,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev,
 						       size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	void *cpu_addr;
 	dma_addr_t dma_addr;
 	int ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 623888bf30cb..919d2fb7427b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -29,6 +29,7 @@
 #include "psp_gfx_if.h"
 #include "ta_xgmi_if.h"
 #include "ta_ras_if.h"
+#include "ta_rap_if.h"
 
 #define PSP_FENCE_BUFFER_SIZE	0x1000
 #define PSP_CMD_BUFFER_SIZE	0x1000
@@ -38,6 +39,7 @@
 #define PSP_TMR_SIZE	0x400000
 #define PSP_HDCP_SHARED_MEM_SIZE	0x4000
 #define PSP_DTM_SHARED_MEM_SIZE	0x4000
+#define PSP_RAP_SHARED_MEM_SIZE	0x4000
 #define PSP_SHARED_MEM_SIZE		0x4000
 
 struct psp_context;
@@ -159,6 +161,15 @@ struct psp_dtm_context {
 	struct mutex		mutex;
 };
 
+struct psp_rap_context {
+	bool			rap_initialized;
+	uint32_t		session_id;
+	struct amdgpu_bo	*rap_shared_bo;
+	uint64_t		rap_shared_mc_addr;
+	void			*rap_shared_buf;
+	struct mutex		mutex;
+};
+
 #define MEM_TRAIN_SYSTEM_SIGNATURE		0x54534942
 #define GDDR6_MEM_TRAINING_DATA_SIZE_IN_BYTES	0x1000
 #define GDDR6_MEM_TRAINING_OFFSET		0x8000
@@ -277,11 +288,16 @@ struct psp_context
 	uint32_t			ta_dtm_ucode_size;
 	uint8_t				*ta_dtm_start_addr;
 
+	uint32_t			ta_rap_ucode_version;
+	uint32_t			ta_rap_ucode_size;
+	uint8_t				*ta_rap_start_addr;
+
 	struct psp_asd_context		asd_context;
 	struct psp_xgmi_context		xgmi_context;
 	struct psp_ras_context		ras;
 	struct psp_hdcp_context 	hdcp_context;
 	struct psp_dtm_context		dtm_context;
+	struct psp_rap_context		rap_context;
 	struct mutex			mutex;
 	struct psp_memory_training_context mem_train_ctx;
 };
@@ -357,6 +373,7 @@ int psp_ras_trigger_error(struct psp_context *psp,
 
 int psp_hdcp_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
 int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
+int psp_rap_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
 
 int psp_rlc_autoload_start(struct psp_context *psp);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.c
new file mode 100644
index 000000000000..8da5356c36f1
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.c
@@ -0,0 +1,127 @@
+/*
+ * Copyright 2020 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"),
+ * 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 <linux/debugfs.h>
+#include <linux/pm_runtime.h>
+
+#include "amdgpu.h"
+#include "amdgpu_rap.h"
+
+/**
+ * DOC: AMDGPU RAP debugfs test interface
+ *
+ * how to use?
+ * echo opcode > <debugfs_dir>/dri/xxx/rap_test
+ *
+ * opcode:
+ * currently, only 2 is supported by Linux host driver,
+ * opcode 2 stands for TA_CMD_RAP__VALIDATE_L0, used to
+ * trigger L0 policy validation, you can refer more detail
+ * from header file ta_rap_if.h
+ *
+ */
+static ssize_t amdgpu_rap_debugfs_write(struct file *f, const char __user *buf,
+		size_t size, loff_t *pos)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
+	struct ta_rap_shared_memory *rap_shared_mem;
+	struct ta_rap_cmd_output_data *rap_cmd_output;
+	struct drm_device *dev = adev_to_drm(adev);
+	uint32_t op;
+	int ret;
+
+	if (*pos || size != 2)
+		return -EINVAL;
+
+	ret = kstrtouint_from_user(buf, size, *pos, &op);
+	if (ret)
+		return ret;
+
+	ret = pm_runtime_get_sync(dev->dev);
+	if (ret < 0) {
+		pm_runtime_put_autosuspend(dev->dev);
+		return ret;
+	}
+
+	/* make sure gfx core is on, RAP TA cann't handle
+	 * GFX OFF case currently.
+	 */
+	amdgpu_gfx_off_ctrl(adev, false);
+
+	switch (op) {
+	case 2:
+		ret = psp_rap_invoke(&adev->psp, op);
+
+		if (ret == TA_RAP_STATUS__SUCCESS) {
+			dev_info(adev->dev, "RAP L0 validate test success.\n");
+		} else {
+			rap_shared_mem = (struct ta_rap_shared_memory *)
+					 adev->psp.rap_context.rap_shared_buf;
+			rap_cmd_output = &(rap_shared_mem->rap_out_message.output);
+
+			dev_info(adev->dev, "RAP test failed, the output is:\n");
+			dev_info(adev->dev, "\tlast_subsection: 0x%08x.\n",
+				 rap_cmd_output->last_subsection);
+			dev_info(adev->dev, "\tnum_total_validate: 0x%08x.\n",
+				 rap_cmd_output->num_total_validate);
+			dev_info(adev->dev, "\tnum_valid: 0x%08x.\n",
+				 rap_cmd_output->num_valid);
+			dev_info(adev->dev, "\tlast_validate_addr: 0x%08x.\n",
+				 rap_cmd_output->last_validate_addr);
+			dev_info(adev->dev, "\tlast_validate_val: 0x%08x.\n",
+				 rap_cmd_output->last_validate_val);
+			dev_info(adev->dev, "\tlast_validate_val_exptd: 0x%08x.\n",
+				 rap_cmd_output->last_validate_val_exptd);
+		}
+		break;
+	default:
+		dev_info(adev->dev, "Unsupported op id: %d, ", op);
+		dev_info(adev->dev, "Only support op 2(L0 validate test).\n");
+	}
+
+	amdgpu_gfx_off_ctrl(adev, true);
+	pm_runtime_mark_last_busy(dev->dev);
+	pm_runtime_put_autosuspend(dev->dev);
+
+	return size;
+}
+
+static const struct file_operations amdgpu_rap_debugfs_ops = {
+	.owner = THIS_MODULE,
+	.read = NULL,
+	.write = amdgpu_rap_debugfs_write,
+	.llseek = default_llseek
+};
+
+void amdgpu_rap_debugfs_init(struct amdgpu_device *adev)
+{
+#if defined(CONFIG_DEBUG_FS)
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
+
+	if (!adev->psp.rap_context.rap_initialized)
+		return;
+
+	debugfs_create_file("rap_test", S_IWUSR, minor->debugfs_root,
+				adev, &amdgpu_rap_debugfs_ops);
+#endif
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.h
new file mode 100644
index 000000000000..ec6d7632d3a0
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rap.h
@@ -0,0 +1,30 @@
+/*
+ * Copyright 2020 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"),
+ * 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.
+ *
+ *
+ */
+#ifndef _AMDGPU_RAP_H
+#define _AMDGPU_RAP_H
+
+#include "amdgpu.h"
+
+void amdgpu_rap_debugfs_init(struct amdgpu_device *adev);
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 1bedb416eebd..8bf6a7c056bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -34,6 +34,8 @@
 #include "amdgpu_xgmi.h"
 #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
 
+static const char *RAS_FS_NAME = "ras";
+
 const char *ras_error_string[] = {
 	"none",
 	"parity",
@@ -62,13 +64,14 @@ const char *ras_block_string[] = {
 #define ras_err_str(i) (ras_error_string[ffs(i)])
 #define ras_block_str(i) (ras_block_string[i])
 
-#define AMDGPU_RAS_FLAG_INIT_BY_VBIOS		1
-#define AMDGPU_RAS_FLAG_INIT_NEED_RESET		2
 #define RAS_DEFAULT_FLAGS (AMDGPU_RAS_FLAG_INIT_BY_VBIOS)
 
 /* inject address is 52 bits */
 #define	RAS_UMC_INJECT_ADDR_LIMIT	(0x1ULL << 52)
 
+/* typical ECC bad page rate(1 bad page per 100MB VRAM) */
+#define RAS_BAD_PAGE_RATE		(100 * 1024 * 1024ULL)
+
 enum amdgpu_ras_retire_page_reservation {
 	AMDGPU_RAS_RETIRE_PAGE_RESERVED,
 	AMDGPU_RAS_RETIRE_PAGE_PENDING,
@@ -367,12 +370,19 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
 static ssize_t amdgpu_ras_debugfs_eeprom_write(struct file *f, const char __user *buf,
 		size_t size, loff_t *pos)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
+	struct amdgpu_device *adev =
+		(struct amdgpu_device *)file_inode(f)->i_private;
 	int ret;
 
-	ret = amdgpu_ras_eeprom_reset_table(&adev->psp.ras.ras->eeprom_control);
+	ret = amdgpu_ras_eeprom_reset_table(
+			&(amdgpu_ras_get_context(adev)->eeprom_control));
 
-	return ret == 1 ? size : -EIO;
+	if (ret == 1) {
+		amdgpu_ras_get_context(adev)->flags = RAS_DEFAULT_FLAGS;
+		return size;
+	} else {
+		return -EIO;
+	}
 }
 
 static const struct file_operations amdgpu_ras_debugfs_ctrl_ops = {
@@ -1017,45 +1027,13 @@ static ssize_t amdgpu_ras_sysfs_features_read(struct device *dev,
 	return scnprintf(buf, PAGE_SIZE, "feature mask: 0x%x\n", con->features);
 }
 
-static int amdgpu_ras_sysfs_create_feature_node(struct amdgpu_device *adev)
+static void amdgpu_ras_sysfs_remove_bad_page_node(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
-	struct attribute *attrs[] = {
-		&con->features_attr.attr,
-		NULL
-	};
-	struct bin_attribute *bin_attrs[] = {
-		&con->badpages_attr,
-		NULL
-	};
-	struct attribute_group group = {
-		.name = "ras",
-		.attrs = attrs,
-		.bin_attrs = bin_attrs,
-	};
 
-	con->features_attr = (struct device_attribute) {
-		.attr = {
-			.name = "features",
-			.mode = S_IRUGO,
-		},
-			.show = amdgpu_ras_sysfs_features_read,
-	};
-
-	con->badpages_attr = (struct bin_attribute) {
-		.attr = {
-			.name = "gpu_vram_bad_pages",
-			.mode = S_IRUGO,
-		},
-		.size = 0,
-		.private = NULL,
-		.read = amdgpu_ras_sysfs_badpages_read,
-	};
-
-	sysfs_attr_init(attrs[0]);
-	sysfs_bin_attr_init(bin_attrs[0]);
-
-	return sysfs_create_group(&adev->dev->kobj, &group);
+	sysfs_remove_file_from_group(&adev->dev->kobj,
+				&con->badpages_attr.attr,
+				RAS_FS_NAME);
 }
 
 static int amdgpu_ras_sysfs_remove_feature_node(struct amdgpu_device *adev)
@@ -1065,14 +1043,9 @@ static int amdgpu_ras_sysfs_remove_feature_node(struct amdgpu_device *adev)
 		&con->features_attr.attr,
 		NULL
 	};
-	struct bin_attribute *bin_attrs[] = {
-		&con->badpages_attr,
-		NULL
-	};
 	struct attribute_group group = {
-		.name = "ras",
+		.name = RAS_FS_NAME,
 		.attrs = attrs,
-		.bin_attrs = bin_attrs,
 	};
 
 	sysfs_remove_group(&adev->dev->kobj, &group);
@@ -1105,7 +1078,7 @@ int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
 
 	if (sysfs_add_file_to_group(&adev->dev->kobj,
 				&obj->sysfs_attr.attr,
-				"ras")) {
+				RAS_FS_NAME)) {
 		put_obj(obj);
 		return -EINVAL;
 	}
@@ -1125,7 +1098,7 @@ int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
 
 	sysfs_remove_file_from_group(&adev->dev->kobj,
 				&obj->sysfs_attr.attr,
-				"ras");
+				RAS_FS_NAME);
 	obj->attr_inuse = 0;
 	put_obj(obj);
 
@@ -1141,6 +1114,9 @@ static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
 		amdgpu_ras_sysfs_remove(adev, &obj->head);
 	}
 
+	if (amdgpu_bad_page_threshold != 0)
+		amdgpu_ras_sysfs_remove_bad_page_node(adev);
+
 	amdgpu_ras_sysfs_remove_feature_node(adev);
 
 	return 0;
@@ -1169,9 +1145,9 @@ static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
 static void amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
-	struct drm_minor *minor = adev->ddev->primary;
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
 
-	con->dir = debugfs_create_dir("ras", minor->debugfs_root);
+	con->dir = debugfs_create_dir(RAS_FS_NAME, minor->debugfs_root);
 	debugfs_create_file("ras_ctrl", S_IWUGO | S_IRUGO, con->dir,
 				adev, &amdgpu_ras_debugfs_ctrl_ops);
 	debugfs_create_file("ras_eeprom_reset", S_IWUGO | S_IRUGO, con->dir,
@@ -1187,6 +1163,13 @@ static void amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
 	 */
 	debugfs_create_bool("auto_reboot", S_IWUGO | S_IRUGO, con->dir,
 				&con->reboot);
+
+	/*
+	 * User could set this not to clean up hardware's error count register
+	 * of RAS IPs during ras recovery.
+	 */
+	debugfs_create_bool("disable_ras_err_cnt_harvest", 0644,
+			con->dir, &con->disable_ras_err_cnt_harvest);
 }
 
 void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
@@ -1211,6 +1194,7 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
 
 void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
 {
+#if defined(CONFIG_DEBUG_FS)
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_manager *obj;
 	struct ras_fs_if fs_info;
@@ -1233,6 +1217,7 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
 			amdgpu_ras_debugfs_create(adev, &fs_info);
 		}
 	}
+#endif
 }
 
 void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
@@ -1249,6 +1234,7 @@ void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
 
 static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
 {
+#if defined(CONFIG_DEBUG_FS)
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_manager *obj, *tmp;
 
@@ -1257,14 +1243,48 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
 	}
 
 	con->dir = NULL;
+#endif
 }
 /* debugfs end */
 
 /* ras fs */
-
+static BIN_ATTR(gpu_vram_bad_pages, S_IRUGO,
+		amdgpu_ras_sysfs_badpages_read, NULL, 0);
+static DEVICE_ATTR(features, S_IRUGO,
+		amdgpu_ras_sysfs_features_read, NULL);
 static int amdgpu_ras_fs_init(struct amdgpu_device *adev)
 {
-	amdgpu_ras_sysfs_create_feature_node(adev);
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct attribute_group group = {
+		.name = RAS_FS_NAME,
+	};
+	struct attribute *attrs[] = {
+		&con->features_attr.attr,
+		NULL
+	};
+	struct bin_attribute *bin_attrs[] = {
+		NULL,
+		NULL,
+	};
+	int r;
+
+	/* add features entry */
+	con->features_attr = dev_attr_features;
+	group.attrs = attrs;
+	sysfs_attr_init(attrs[0]);
+
+	if (amdgpu_bad_page_threshold != 0) {
+		/* add bad_page_features entry */
+		bin_attr_gpu_vram_bad_pages.private = NULL;
+		con->badpages_attr = bin_attr_gpu_vram_bad_pages;
+		bin_attrs[0] = &con->badpages_attr;
+		group.bin_attrs = bin_attrs;
+		sysfs_bin_attr_init(bin_attrs[0]);
+	}
+
+	r = sysfs_create_group(&adev->dev->kobj, &group);
+	if (r)
+		dev_err(adev->dev, "Failed to create RAS sysfs group!");
 
 	return 0;
 }
@@ -1456,6 +1476,45 @@ static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev)
 	}
 }
 
+/* Parse RdRspStatus and WrRspStatus */
+void amdgpu_ras_error_status_query(struct amdgpu_device *adev,
+		struct ras_query_if *info)
+{
+	/*
+	 * Only two block need to query read/write
+	 * RspStatus at current state
+	 */
+	switch (info->head.block) {
+	case AMDGPU_RAS_BLOCK__GFX:
+		if (adev->gfx.funcs->query_ras_error_status)
+			adev->gfx.funcs->query_ras_error_status(adev);
+		break;
+	case AMDGPU_RAS_BLOCK__MMHUB:
+		if (adev->mmhub.funcs->query_ras_error_status)
+			adev->mmhub.funcs->query_ras_error_status(adev);
+		break;
+	default:
+		break;
+	}
+}
+
+static void amdgpu_ras_query_err_status(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+
+	if (!con)
+		return;
+
+	list_for_each_entry(obj, &con->head, node) {
+		struct ras_query_if info = {
+			.head = obj->head,
+		};
+
+		amdgpu_ras_error_status_query(adev, &info);
+	}
+}
+
 /* recovery begin */
 
 /* return 0 on success.
@@ -1512,23 +1571,30 @@ static void amdgpu_ras_do_recovery(struct work_struct *work)
 	struct amdgpu_device *remote_adev = NULL;
 	struct amdgpu_device *adev = ras->adev;
 	struct list_head device_list, *device_list_handle =  NULL;
-	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, false);
-
-	/* Build list of devices to query RAS related errors */
-	if  (hive && adev->gmc.xgmi.num_physical_nodes > 1)
-		device_list_handle = &hive->device_list;
-	else {
-		INIT_LIST_HEAD(&device_list);
-		list_add_tail(&adev->gmc.xgmi.head, &device_list);
-		device_list_handle = &device_list;
-	}
 
-	list_for_each_entry(remote_adev, device_list_handle, gmc.xgmi.head) {
-		amdgpu_ras_log_on_err_counter(remote_adev);
+	if (!ras->disable_ras_err_cnt_harvest) {
+		struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
+
+		/* Build list of devices to query RAS related errors */
+		if  (hive && adev->gmc.xgmi.num_physical_nodes > 1) {
+			device_list_handle = &hive->device_list;
+		} else {
+			INIT_LIST_HEAD(&device_list);
+			list_add_tail(&adev->gmc.xgmi.head, &device_list);
+			device_list_handle = &device_list;
+		}
+
+		list_for_each_entry(remote_adev,
+				device_list_handle, gmc.xgmi.head) {
+			amdgpu_ras_query_err_status(remote_adev);
+			amdgpu_ras_log_on_err_counter(remote_adev);
+		}
+
+		amdgpu_put_xgmi_hive(hive);
 	}
 
 	if (amdgpu_device_should_recover_gpu(ras->adev))
-		amdgpu_device_gpu_recover(ras->adev, 0);
+		amdgpu_device_gpu_recover(ras->adev, NULL);
 	atomic_set(&ras->in_recovery, 0);
 }
 
@@ -1643,7 +1709,7 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev)
 	int ret = 0;
 
 	/* no bad page record, skip eeprom access */
-	if (!control->num_recs)
+	if (!control->num_recs || (amdgpu_bad_page_threshold == 0))
 		return ret;
 
 	bps = kcalloc(control->num_recs, sizeof(*bps), GFP_KERNEL);
@@ -1697,6 +1763,47 @@ out:
 	return ret;
 }
 
+static void amdgpu_ras_validate_threshold(struct amdgpu_device *adev,
+					uint32_t max_length)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	int tmp_threshold = amdgpu_bad_page_threshold;
+	u64 val;
+
+	/*
+	 * Justification of value bad_page_cnt_threshold in ras structure
+	 *
+	 * Generally, -1 <= amdgpu_bad_page_threshold <= max record length
+	 * in eeprom, and introduce two scenarios accordingly.
+	 *
+	 * Bad page retirement enablement:
+	 *    - If amdgpu_bad_page_threshold = -1,
+	 *      bad_page_cnt_threshold = typical value by formula.
+	 *
+	 *    - When the value from user is 0 < amdgpu_bad_page_threshold <
+	 *      max record length in eeprom, use it directly.
+	 *
+	 * Bad page retirement disablement:
+	 *    - If amdgpu_bad_page_threshold = 0, bad page retirement
+	 *      functionality is disabled, and bad_page_cnt_threshold will
+	 *      take no effect.
+	 */
+
+	if (tmp_threshold < -1)
+		tmp_threshold = -1;
+	else if (tmp_threshold > max_length)
+		tmp_threshold = max_length;
+
+	if (tmp_threshold == -1) {
+		val = adev->gmc.mc_vram_size;
+		do_div(val, RAS_BAD_PAGE_RATE);
+		con->bad_page_cnt_threshold = min(lower_32_bits(val),
+						max_length);
+	} else {
+		con->bad_page_cnt_threshold = tmp_threshold;
+	}
+}
+
 /* called in gpu recovery/init */
 int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev)
 {
@@ -1706,7 +1813,8 @@ int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev)
 	struct amdgpu_bo *bo = NULL;
 	int i, ret = 0;
 
-	if (!con || !con->eh_data)
+	/* Not reserve bad page when amdgpu_bad_page_threshold == 0. */
+	if (!con || !con->eh_data || (amdgpu_bad_page_threshold == 0))
 		return 0;
 
 	mutex_lock(&con->recovery_lock);
@@ -1774,6 +1882,8 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_err_handler_data **data;
+	uint32_t max_eeprom_records_len = 0;
+	bool exc_err_limit = false;
 	int ret;
 
 	if (con)
@@ -1792,8 +1902,15 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
 	atomic_set(&con->in_recovery, 0);
 	con->adev = adev;
 
-	ret = amdgpu_ras_eeprom_init(&con->eeprom_control);
-	if (ret)
+	max_eeprom_records_len = amdgpu_ras_eeprom_get_record_max_length();
+	amdgpu_ras_validate_threshold(adev, max_eeprom_records_len);
+
+	ret = amdgpu_ras_eeprom_init(&con->eeprom_control, &exc_err_limit);
+	/*
+	 * This calling fails when exc_err_limit is true or
+	 * ret != 0.
+	 */
+	if (exc_err_limit || ret)
 		goto free;
 
 	if (con->eeprom_control.num_recs) {
@@ -1817,6 +1934,15 @@ free:
 out:
 	dev_warn(adev->dev, "Failed to initialize ras recovery!\n");
 
+	/*
+	 * Except error threshold exceeding case, other failure cases in this
+	 * function would not fail amdgpu driver init.
+	 */
+	if (!exc_err_limit)
+		ret = 0;
+	else
+		ret = -EINVAL;
+
 	return ret;
 }
 
@@ -1856,6 +1982,16 @@ int amdgpu_ras_request_reset_on_boot(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int amdgpu_ras_check_asic_type(struct amdgpu_device *adev)
+{
+	if (adev->asic_type != CHIP_VEGA10 &&
+		adev->asic_type != CHIP_VEGA20 &&
+		adev->asic_type != CHIP_ARCTURUS)
+		return 1;
+	else
+		return 0;
+}
+
 /*
  * check hardware's ras ability which will be saved in hw_supported.
  * if hardware does not support ras, we can skip some ras initializtion and
@@ -1872,8 +2008,7 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
 	*supported = 0;
 
 	if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw ||
-	    (adev->asic_type != CHIP_VEGA20 &&
-	     adev->asic_type != CHIP_ARCTURUS))
+		amdgpu_ras_check_asic_type(adev))
 		return;
 
 	if (amdgpu_atomfirmware_mem_ecc_supported(adev)) {
@@ -1895,6 +2030,8 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
 
 	*supported = amdgpu_ras_enable == 0 ?
 			0 : *hw_supported & amdgpu_ras_mask;
+
+	adev->ras_features = *supported;
 }
 
 int amdgpu_ras_init(struct amdgpu_device *adev)
@@ -1917,9 +2054,9 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 
 	amdgpu_ras_check_supported(adev, &con->hw_supported,
 			&con->supported);
-	if (!con->hw_supported) {
+	if (!con->hw_supported || (adev->asic_type == CHIP_VEGA10)) {
 		r = 0;
-		goto err_out;
+		goto release_con;
 	}
 
 	con->features = 0;
@@ -1930,25 +2067,25 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 	if (adev->nbio.funcs->init_ras_controller_interrupt) {
 		r = adev->nbio.funcs->init_ras_controller_interrupt(adev);
 		if (r)
-			goto err_out;
+			goto release_con;
 	}
 
 	if (adev->nbio.funcs->init_ras_err_event_athub_interrupt) {
 		r = adev->nbio.funcs->init_ras_err_event_athub_interrupt(adev);
 		if (r)
-			goto err_out;
+			goto release_con;
 	}
 
 	if (amdgpu_ras_fs_init(adev)) {
 		r = -EINVAL;
-		goto err_out;
+		goto release_con;
 	}
 
 	dev_info(adev->dev, "RAS INFO: ras initialized successfully, "
 			"hardware ability[%x] ras_mask[%x]\n",
 			con->hw_supported, con->supported);
 	return 0;
-err_out:
+release_con:
 	amdgpu_ras_set_context(adev, NULL);
 	kfree(con);
 
@@ -1976,7 +2113,7 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev,
 			amdgpu_ras_request_reset_on_boot(adev,
 					ras_block->block);
 			return 0;
-		} else if (adev->in_suspend || adev->in_gpu_reset) {
+		} else if (adev->in_suspend || amdgpu_in_reset(adev)) {
 			/* in resume phase, if fail to enable ras,
 			 * clean up all ras fs nodes, and disable ras */
 			goto cleanup;
@@ -1985,7 +2122,7 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev,
 	}
 
 	/* in resume phase, no need to create ras fs node */
-	if (adev->in_suspend || adev->in_gpu_reset)
+	if (adev->in_suspend || amdgpu_in_reset(adev))
 		return 0;
 
 	if (ih_info->cb) {
@@ -2143,3 +2280,19 @@ bool amdgpu_ras_need_emergency_restart(struct amdgpu_device *adev)
 
 	return false;
 }
+
+bool amdgpu_ras_check_err_threshold(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	bool exc_err_limit = false;
+
+	if (con && (amdgpu_bad_page_threshold != 0))
+		amdgpu_ras_eeprom_check_err_threshold(&con->eeprom_control,
+						&exc_err_limit);
+
+	/*
+	 * We are only interested in variable exc_err_limit,
+	 * as it says if GPU is in bad state or not.
+	 */
+	return exc_err_limit;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index b2667342cf67..6b8d7bb83bb3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -31,6 +31,10 @@
 #include "ta_ras_if.h"
 #include "amdgpu_ras_eeprom.h"
 
+#define AMDGPU_RAS_FLAG_INIT_BY_VBIOS		(0x1 << 0)
+#define AMDGPU_RAS_FLAG_INIT_NEED_RESET		(0x1 << 1)
+#define AMDGPU_RAS_FLAG_SKIP_BAD_PAGE_RESV	(0x1 << 2)
+
 enum amdgpu_ras_block {
 	AMDGPU_RAS_BLOCK__UMC = 0,
 	AMDGPU_RAS_BLOCK__SDMA,
@@ -336,6 +340,12 @@ struct amdgpu_ras {
 	struct amdgpu_ras_eeprom_control eeprom_control;
 
 	bool error_query_ready;
+
+	/* bad page count threshold */
+	uint32_t bad_page_cnt_threshold;
+
+	/* disable ras error count harvest in recovery */
+	bool disable_ras_err_cnt_harvest;
 };
 
 struct ras_fs_data {
@@ -490,6 +500,8 @@ void amdgpu_ras_suspend(struct amdgpu_device *adev);
 unsigned long amdgpu_ras_query_error_count(struct amdgpu_device *adev,
 		bool is_ce);
 
+bool amdgpu_ras_check_err_threshold(struct amdgpu_device *adev);
+
 /* error handling functions */
 int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
 		struct eeprom_table_record *bps, int pages);
@@ -500,10 +512,14 @@ static inline int amdgpu_ras_reset_gpu(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
-	/* save bad page to eeprom before gpu reset,
-	 * i2c may be unstable in gpu reset
+	/*
+	 * Save bad page to eeprom before gpu reset, i2c may be unstable
+	 * in gpu reset.
+	 *
+	 * Also, exclude the case when ras recovery issuer is
+	 * eeprom page write itself.
 	 */
-	if (in_task())
+	if (!(ras->flags & AMDGPU_RAS_FLAG_SKIP_BAD_PAGE_RESV) && in_task())
 		amdgpu_ras_reserve_bad_pages(adev);
 
 	if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index c0096097bbcf..0e64c39a2372 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -46,6 +46,9 @@
 #define EEPROM_TABLE_HDR_VAL 0x414d4452
 #define EEPROM_TABLE_VER 0x00010000
 
+/* Bad GPU tag ‘BADG’ */
+#define EEPROM_TABLE_HDR_BAD 0x42414447
+
 /* Assume 2 Mbit size */
 #define EEPROM_SIZE_BYTES 256000
 #define EEPROM_PAGE__SIZE_BYTES 256
@@ -56,6 +59,15 @@
 
 #define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control))->adev
 
+static bool __is_ras_eeprom_supported(struct amdgpu_device *adev)
+{
+	if ((adev->asic_type == CHIP_VEGA20) ||
+	    (adev->asic_type == CHIP_ARCTURUS))
+		return true;
+
+	return false;
+}
+
 static bool __get_eeprom_i2c_addr_arct(struct amdgpu_device *adev,
 				       uint16_t *i2c_addr)
 {
@@ -213,6 +225,24 @@ static bool __validate_tbl_checksum(struct amdgpu_ras_eeprom_control *control,
 	return true;
 }
 
+static int amdgpu_ras_eeprom_correct_header_tag(
+				struct amdgpu_ras_eeprom_control *control,
+				uint32_t header)
+{
+	unsigned char buff[EEPROM_ADDRESS_SIZE + EEPROM_TABLE_HEADER_SIZE];
+	struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr;
+	int ret = 0;
+
+	memset(buff, 0, EEPROM_ADDRESS_SIZE + EEPROM_TABLE_HEADER_SIZE);
+
+	mutex_lock(&control->tbl_mutex);
+	hdr->header = header;
+	ret = __update_table_header(control, buff);
+	mutex_unlock(&control->tbl_mutex);
+
+	return ret;
+}
+
 int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 {
 	unsigned char buff[EEPROM_ADDRESS_SIZE + EEPROM_TABLE_HEADER_SIZE] = { 0 };
@@ -238,12 +268,14 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 
 }
 
-int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control)
+int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
+			bool *exceed_err_limit)
 {
 	int ret = 0;
 	struct amdgpu_device *adev = to_amdgpu_device(control);
 	unsigned char buff[EEPROM_ADDRESS_SIZE + EEPROM_TABLE_HEADER_SIZE] = { 0 };
 	struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr;
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 	struct i2c_msg msg = {
 			.addr	= 0,
 			.flags	= I2C_M_RD,
@@ -251,6 +283,11 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control)
 			.buf	= buff,
 	};
 
+	*exceed_err_limit = false;
+
+	if (!__is_ras_eeprom_supported(adev))
+		return 0;
+
 	/* Verify i2c adapter is initialized */
 	if (!adev->pm.smu_i2c.algo)
 		return -ENOENT;
@@ -279,6 +316,18 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control)
 		DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records",
 				 control->num_recs);
 
+	} else if ((hdr->header == EEPROM_TABLE_HDR_BAD) &&
+			(amdgpu_bad_page_threshold != 0)) {
+		if (ras->bad_page_cnt_threshold > control->num_recs) {
+			dev_info(adev->dev, "Using one valid bigger bad page "
+				"threshold and correcting eeprom header tag.\n");
+			ret = amdgpu_ras_eeprom_correct_header_tag(control,
+							EEPROM_TABLE_HDR_VAL);
+		} else {
+			*exceed_err_limit = true;
+			dev_err(adev->dev, "Exceeding the bad_page_threshold parameter, "
+				"disabling the GPU.\n");
+		}
 	} else {
 		DRM_INFO("Creating new EEPROM table");
 
@@ -375,6 +424,49 @@ static uint32_t __correct_eeprom_dest_address(uint32_t curr_address)
 	return curr_address;
 }
 
+int amdgpu_ras_eeprom_check_err_threshold(
+				struct amdgpu_ras_eeprom_control *control,
+				bool *exceed_err_limit)
+{
+	struct amdgpu_device *adev = to_amdgpu_device(control);
+	unsigned char buff[EEPROM_ADDRESS_SIZE +
+			EEPROM_TABLE_HEADER_SIZE] = { 0 };
+	struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr;
+	struct i2c_msg msg = {
+			.addr = control->i2c_address,
+			.flags = I2C_M_RD,
+			.len = EEPROM_ADDRESS_SIZE + EEPROM_TABLE_HEADER_SIZE,
+			.buf = buff,
+	};
+	int ret;
+
+	*exceed_err_limit = false;
+
+	if (!__is_ras_eeprom_supported(adev))
+		return 0;
+
+	/* read EEPROM table header */
+	mutex_lock(&control->tbl_mutex);
+	ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1);
+	if (ret < 1) {
+		dev_err(adev->dev, "Failed to read EEPROM table header.\n");
+		goto err;
+	}
+
+	__decode_table_header_from_buff(hdr, &buff[2]);
+
+	if (hdr->header == EEPROM_TABLE_HDR_BAD) {
+		dev_warn(adev->dev, "This GPU is in BAD status.");
+		dev_warn(adev->dev, "Please retire it or setting one bigger "
+				"threshold value when reloading driver.\n");
+		*exceed_err_limit = true;
+	}
+
+err:
+	mutex_unlock(&control->tbl_mutex);
+	return 0;
+}
+
 int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control,
 					    struct eeprom_table_record *records,
 					    bool write,
@@ -383,10 +475,12 @@ int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control,
 	int i, ret = 0;
 	struct i2c_msg *msgs, *msg;
 	unsigned char *buffs, *buff;
+	bool sched_ras_recovery = false;
 	struct eeprom_table_record *record;
 	struct amdgpu_device *adev = to_amdgpu_device(control);
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
-	if (adev->asic_type != CHIP_VEGA20 && adev->asic_type != CHIP_ARCTURUS)
+	if (!__is_ras_eeprom_supported(adev))
 		return 0;
 
 	buffs = kcalloc(num, EEPROM_ADDRESS_SIZE + EEPROM_TABLE_RECORD_SIZE,
@@ -402,11 +496,30 @@ int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control,
 		goto free_buff;
 	}
 
+	/*
+	 * If saved bad pages number exceeds the bad page threshold for
+	 * the whole VRAM, update table header to mark the BAD GPU tag
+	 * and schedule one ras recovery after eeprom write is done,
+	 * this can avoid the missing for latest records.
+	 *
+	 * This new header will be picked up and checked in the bootup
+	 * by ras recovery, which may break bootup process to notify
+	 * user this GPU is in bad state and to retire such GPU for
+	 * further check.
+	 */
+	if (write && (amdgpu_bad_page_threshold != 0) &&
+		((control->num_recs + num) >= ras->bad_page_cnt_threshold)) {
+		dev_warn(adev->dev,
+			"Saved bad pages(%d) reaches threshold value(%d).\n",
+			control->num_recs + num, ras->bad_page_cnt_threshold);
+		control->tbl_hdr.header = EEPROM_TABLE_HDR_BAD;
+		sched_ras_recovery = true;
+	}
+
 	/* In case of overflow just start from beginning to not lose newest records */
 	if (write && (control->next_addr + EEPROM_TABLE_RECORD_SIZE * num > EEPROM_SIZE_BYTES))
 		control->next_addr = EEPROM_RECORD_START;
 
-
 	/*
 	 * TODO Currently makes EEPROM writes for each record, this creates
 	 * internal fragmentation. Optimized the code to do full page write of
@@ -482,6 +595,20 @@ int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control,
 		__update_tbl_checksum(control, records, num, old_hdr_byte_sum);
 
 		__update_table_header(control, buffs);
+
+		if (sched_ras_recovery) {
+			/*
+			 * Before scheduling ras recovery, assert the related
+			 * flag first, which shall bypass common bad page
+			 * reservation execution in amdgpu_ras_reset_gpu.
+			 */
+			amdgpu_ras_get_context(adev)->flags |=
+				AMDGPU_RAS_FLAG_SKIP_BAD_PAGE_RESV;
+
+			dev_warn(adev->dev, "Conduct ras recovery due to bad "
+				"page threshold reached.\n");
+			amdgpu_ras_reset_gpu(adev);
+		}
 	} else if (!__validate_tbl_checksum(control, records, num)) {
 		DRM_WARN("EEPROM Table checksum mismatch!");
 		/* TODO Uncomment when EEPROM read/write is relliable */
@@ -499,6 +626,11 @@ free_buff:
 	return ret == num ? 0 : -EIO;
 }
 
+inline uint32_t amdgpu_ras_eeprom_get_record_max_length(void)
+{
+	return EEPROM_MAX_RECORD_NUM;
+}
+
 /* Used for testing if bugs encountered */
 #if 0
 void amdgpu_ras_eeprom_test(struct amdgpu_ras_eeprom_control *control)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
index 9e7d640920fb..c7a5e5c7c61e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
@@ -76,14 +76,21 @@ struct eeprom_table_record {
 	unsigned char mcumc_id;
 }__attribute__((__packed__));
 
-int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control);
+int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
+			bool *exceed_err_limit);
 int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control);
 
+int amdgpu_ras_eeprom_check_err_threshold(
+				struct amdgpu_ras_eeprom_control *control,
+				bool *exceed_err_limit);
+
 int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control,
 					    struct eeprom_table_record *records,
 					    bool write,
 					    int num);
 
+inline uint32_t amdgpu_ras_eeprom_get_record_max_length(void);
+
 void amdgpu_ras_eeprom_test(struct amdgpu_ras_eeprom_control *control);
 
 #endif // _AMDGPU_RAS_EEPROM_H
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 13ea8ebc421c..15ee13c3bd9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -267,7 +267,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 			&ring->sched;
 	}
 
-	for (i = 0; i < DRM_SCHED_PRIORITY_MAX; ++i)
+	for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_COUNT; ++i)
 		atomic_set(&ring->num_jobs[i], 0);
 
 	return 0;
@@ -420,7 +420,7 @@ int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
 			     struct amdgpu_ring *ring)
 {
 #if defined(CONFIG_DEBUG_FS)
-	struct drm_minor *minor = adev->ddev->primary;
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
 	struct dentry *ent, *root = minor->debugfs_root;
 	char name[32];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index da871d84b742..7112137689db 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -243,7 +243,7 @@ struct amdgpu_ring {
 	bool			has_compute_vm_bug;
 	bool			no_scheduler;
 
-	atomic_t		num_jobs[DRM_SCHED_PRIORITY_MAX];
+	atomic_t		num_jobs[DRM_SCHED_PRIORITY_COUNT];
 	struct mutex		priority_mutex;
 	/* protected by priority_mutex */
 	int			priority;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
index c799691dfa84..0da0a0d98672 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
@@ -32,24 +32,32 @@
 
 #include "amdgpu_vm.h"
 
-enum drm_sched_priority amdgpu_to_sched_priority(int amdgpu_priority)
+int amdgpu_to_sched_priority(int amdgpu_priority,
+			     enum drm_sched_priority *prio)
 {
 	switch (amdgpu_priority) {
 	case AMDGPU_CTX_PRIORITY_VERY_HIGH:
-		return DRM_SCHED_PRIORITY_HIGH_HW;
+		*prio = DRM_SCHED_PRIORITY_HIGH;
+		break;
 	case AMDGPU_CTX_PRIORITY_HIGH:
-		return DRM_SCHED_PRIORITY_HIGH_SW;
+		*prio = DRM_SCHED_PRIORITY_HIGH;
+		break;
 	case AMDGPU_CTX_PRIORITY_NORMAL:
-		return DRM_SCHED_PRIORITY_NORMAL;
+		*prio = DRM_SCHED_PRIORITY_NORMAL;
+		break;
 	case AMDGPU_CTX_PRIORITY_LOW:
 	case AMDGPU_CTX_PRIORITY_VERY_LOW:
-		return DRM_SCHED_PRIORITY_LOW;
+		*prio = DRM_SCHED_PRIORITY_MIN;
+		break;
 	case AMDGPU_CTX_PRIORITY_UNSET:
-		return DRM_SCHED_PRIORITY_UNSET;
+		*prio = DRM_SCHED_PRIORITY_UNSET;
+		break;
 	default:
 		WARN(1, "Invalid context priority %d\n", amdgpu_priority);
-		return DRM_SCHED_PRIORITY_INVALID;
+		return -EINVAL;
 	}
+
+	return 0;
 }
 
 static int amdgpu_sched_process_priority_override(struct amdgpu_device *adev,
@@ -115,13 +123,24 @@ int amdgpu_sched_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *filp)
 {
 	union drm_amdgpu_sched *args = data;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	enum drm_sched_priority priority;
 	int r;
 
-	priority = amdgpu_to_sched_priority(args->in.priority);
-	if (priority == DRM_SCHED_PRIORITY_INVALID)
+	/* First check the op, then the op's argument.
+	 */
+	switch (args->in.op) {
+	case AMDGPU_SCHED_OP_PROCESS_PRIORITY_OVERRIDE:
+	case AMDGPU_SCHED_OP_CONTEXT_PRIORITY_OVERRIDE:
+		break;
+	default:
+		DRM_ERROR("Invalid sched op specified: %d\n", args->in.op);
 		return -EINVAL;
+	}
+
+	r = amdgpu_to_sched_priority(args->in.priority, &priority);
+	if (r)
+		return r;
 
 	switch (args->in.op) {
 	case AMDGPU_SCHED_OP_PROCESS_PRIORITY_OVERRIDE:
@@ -136,7 +155,8 @@ int amdgpu_sched_ioctl(struct drm_device *dev, void *data,
 							   priority);
 		break;
 	default:
-		DRM_ERROR("Invalid sched op specified: %d\n", args->in.op);
+		/* Impossible.
+		 */
 		r = -EINVAL;
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.h
index 12299fd95691..67e5b2472f6a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.h
@@ -30,7 +30,8 @@ enum drm_sched_priority;
 struct drm_device;
 struct drm_file;
 
-enum drm_sched_priority amdgpu_to_sched_priority(int amdgpu_priority);
+int amdgpu_to_sched_priority(int amdgpu_priority,
+			     enum drm_sched_priority *prio);
 int amdgpu_sched_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *filp);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index 63e734a125fb..ee9480d14cbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -35,7 +35,7 @@
 #define AMDGPU_JOB_GET_TIMELINE_NAME(job) \
 	 job->base.s_fence->finished.ops->get_timeline_name(&job->base.s_fence->finished)
 
-TRACE_EVENT(amdgpu_mm_rreg,
+TRACE_EVENT(amdgpu_device_rreg,
 	    TP_PROTO(unsigned did, uint32_t reg, uint32_t value),
 	    TP_ARGS(did, reg, value),
 	    TP_STRUCT__entry(
@@ -54,7 +54,7 @@ TRACE_EVENT(amdgpu_mm_rreg,
 		      (unsigned long)__entry->value)
 );
 
-TRACE_EVENT(amdgpu_mm_wreg,
+TRACE_EVENT(amdgpu_device_wreg,
 	    TP_PROTO(unsigned did, uint32_t reg, uint32_t value),
 	    TP_ARGS(did, reg, value),
 	    TP_STRUCT__entry(
@@ -321,6 +321,49 @@ DEFINE_EVENT(amdgpu_vm_mapping, amdgpu_vm_bo_cs,
 	    TP_ARGS(mapping)
 );
 
+TRACE_EVENT(amdgpu_vm_update_ptes,
+	    TP_PROTO(struct amdgpu_vm_update_params *p,
+		     uint64_t start, uint64_t end,
+		     unsigned int nptes, uint64_t dst,
+		     uint64_t incr, uint64_t flags,
+		     pid_t pid, uint64_t vm_ctx),
+	TP_ARGS(p, start, end, nptes, dst, incr, flags, pid, vm_ctx),
+	TP_STRUCT__entry(
+			 __field(u64, start)
+			 __field(u64, end)
+			 __field(u64, flags)
+			 __field(unsigned int, nptes)
+			 __field(u64, incr)
+			 __field(pid_t, pid)
+			 __field(u64, vm_ctx)
+			 __dynamic_array(u64, dst, nptes)
+	),
+
+	TP_fast_assign(
+			unsigned int i;
+
+			__entry->start = start;
+			__entry->end = end;
+			__entry->flags = flags;
+			__entry->incr = incr;
+			__entry->nptes = nptes;
+			__entry->pid = pid;
+			__entry->vm_ctx = vm_ctx;
+			for (i = 0; i < nptes; ++i) {
+				u64 addr = p->pages_addr ? amdgpu_vm_map_gart(
+					p->pages_addr, dst) : dst;
+
+				((u64 *)__get_dynamic_array(dst))[i] = addr;
+				dst += incr;
+			}
+	),
+	TP_printk("pid:%u vm_ctx:0x%llx start:0x%010llx end:0x%010llx,"
+		  " flags:0x%llx, incr:%llu, dst:\n%s", __entry->pid,
+		  __entry->vm_ctx, __entry->start, __entry->end,
+		  __entry->flags, __entry->incr,  __print_array(
+		  __get_dynamic_array(dst), __entry->nptes, 8))
+);
+
 TRACE_EVENT(amdgpu_vm_set_ptes,
 	    TP_PROTO(uint64_t pe, uint64_t addr, unsigned count,
 		     uint32_t incr, uint64_t flags, bool direct),
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 978bae731398..8039d2399584 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -63,61 +63,16 @@
 
 #define AMDGPU_TTM_VRAM_MAX_DW_READ	(size_t)128
 
+static int amdgpu_ttm_backend_bind(struct ttm_bo_device *bdev,
+				   struct ttm_tt *ttm,
+				   struct ttm_resource *bo_mem);
 
-/**
- * amdgpu_init_mem_type - Initialize a memory manager for a specific type of
- * memory request.
- *
- * @bdev: The TTM BO device object (contains a reference to amdgpu_device)
- * @type: The type of memory requested
- * @man: The memory type manager for each domain
- *
- * This is called by ttm_bo_init_mm() when a buffer object is being
- * initialized.
- */
-static int amdgpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-				struct ttm_mem_type_manager *man)
+static int amdgpu_ttm_init_on_chip(struct amdgpu_device *adev,
+				    unsigned int type,
+				    uint64_t size)
 {
-	struct amdgpu_device *adev;
-
-	adev = amdgpu_ttm_adev(bdev);
-
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		/* System memory */
-		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_TT:
-		/* GTT memory  */
-		man->func = &amdgpu_gtt_mgr_func;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
-		break;
-	case TTM_PL_VRAM:
-		/* "On-card" video ram */
-		man->func = &amdgpu_vram_mgr_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED |
-			     TTM_MEMTYPE_FLAG_MAPPABLE;
-		man->available_caching = TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_WC;
-		man->default_caching = TTM_PL_FLAG_WC;
-		break;
-	case AMDGPU_PL_GDS:
-	case AMDGPU_PL_GWS:
-	case AMDGPU_PL_OA:
-		/* On-chip GDS memory*/
-		man->func = &ttm_bo_manager_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED;
-		man->available_caching = TTM_PL_FLAG_UNCACHED;
-		man->default_caching = TTM_PL_FLAG_UNCACHED;
-		break;
-	default:
-		DRM_ERROR("Unsupported memory type %u\n", (unsigned)type);
-		return -EINVAL;
-	}
-	return 0;
+	return ttm_range_man_init(&adev->mman.bdev, type,
+				  false, size >> PAGE_SHIFT);
 }
 
 /**
@@ -136,7 +91,8 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
 	static const struct ttm_place placements = {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_MASK_CACHING
 	};
 
 	/* Don't handle scatter gather BOs */
@@ -223,24 +179,6 @@ static int amdgpu_verify_access(struct ttm_buffer_object *bo, struct file *filp)
 }
 
 /**
- * amdgpu_move_null - Register memory for a buffer object
- *
- * @bo: The bo to assign the memory to
- * @new_mem: The memory to be assigned.
- *
- * Assign the memory from new_mem to the memory of the buffer object bo.
- */
-static void amdgpu_move_null(struct ttm_buffer_object *bo,
-			     struct ttm_mem_reg *new_mem)
-{
-	struct ttm_mem_reg *old_mem = &bo->mem;
-
-	BUG_ON(old_mem->mm_node != NULL);
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
-}
-
-/**
  * amdgpu_mm_node_addr - Compute the GPU relative offset of a GTT buffer.
  *
  * @bo: The bo to assign the memory to.
@@ -250,7 +188,7 @@ static void amdgpu_move_null(struct ttm_buffer_object *bo,
  */
 static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo,
 				    struct drm_mm_node *mm_node,
-				    struct ttm_mem_reg *mem)
+				    struct ttm_resource *mem)
 {
 	uint64_t addr = 0;
 
@@ -270,7 +208,7 @@ static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo,
  * @offset: The offset that drm_mm_node is used for finding.
  *
  */
-static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem,
+static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_resource *mem,
 					       uint64_t *offset)
 {
 	struct drm_mm_node *mm_node = mem->mm_node;
@@ -298,7 +236,7 @@ static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem,
  * the physical address for local memory.
  */
 static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo,
-				 struct ttm_mem_reg *mem,
+				 struct ttm_resource *mem,
 				 struct drm_mm_node *mm_node,
 				 unsigned num_pages, uint64_t offset,
 				 unsigned window, struct amdgpu_ring *ring,
@@ -521,9 +459,9 @@ error:
  * help move buffers to and from VRAM.
  */
 static int amdgpu_move_blit(struct ttm_buffer_object *bo,
-			    bool evict, bool no_wait_gpu,
-			    struct ttm_mem_reg *new_mem,
-			    struct ttm_mem_reg *old_mem)
+			    bool evict,
+			    struct ttm_resource *new_mem,
+			    struct ttm_resource *old_mem)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo);
@@ -562,9 +500,9 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
 
 	/* Always block for VM page tables before committing the new location */
 	if (bo->type == ttm_bo_type_kernel)
-		r = ttm_bo_move_accel_cleanup(bo, fence, true, new_mem);
+		r = ttm_bo_move_accel_cleanup(bo, fence, true, false, new_mem);
 	else
-		r = ttm_bo_pipeline_move(bo, fence, evict, new_mem);
+		r = ttm_bo_move_accel_cleanup(bo, fence, evict, true, new_mem);
 	dma_fence_put(fence);
 	return r;
 
@@ -582,10 +520,10 @@ error:
  */
 static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict,
 				struct ttm_operation_ctx *ctx,
-				struct ttm_mem_reg *new_mem)
+				struct ttm_resource *new_mem)
 {
-	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg tmp_mem;
+	struct ttm_resource *old_mem = &bo->mem;
+	struct ttm_resource tmp_mem;
 	struct ttm_place placements;
 	struct ttm_placement placement;
 	int r;
@@ -599,7 +537,8 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict,
 	placement.busy_placement = &placements;
 	placements.fpfn = 0;
 	placements.lpfn = 0;
-	placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
+	placements.mem_type = TTM_PL_TT;
+	placements.flags = TTM_PL_MASK_CACHING;
 	r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx);
 	if (unlikely(r)) {
 		pr_err("Failed to find GTT space for blit from VRAM\n");
@@ -612,14 +551,18 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict,
 		goto out_cleanup;
 	}
 
+	r = ttm_tt_populate(bo->bdev, bo->ttm, ctx);
+	if (unlikely(r))
+		goto out_cleanup;
+
 	/* Bind the memory to the GTT space */
-	r = ttm_tt_bind(bo->ttm, &tmp_mem, ctx);
+	r = amdgpu_ttm_backend_bind(bo->bdev, bo->ttm, &tmp_mem);
 	if (unlikely(r)) {
 		goto out_cleanup;
 	}
 
 	/* blit VRAM to GTT */
-	r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu, &tmp_mem, old_mem);
+	r = amdgpu_move_blit(bo, evict, &tmp_mem, old_mem);
 	if (unlikely(r)) {
 		goto out_cleanup;
 	}
@@ -627,7 +570,7 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict,
 	/* move BO (in tmp_mem) to new_mem */
 	r = ttm_bo_move_ttm(bo, ctx, new_mem);
 out_cleanup:
-	ttm_bo_mem_put(bo, &tmp_mem);
+	ttm_resource_free(bo, &tmp_mem);
 	return r;
 }
 
@@ -638,10 +581,10 @@ out_cleanup:
  */
 static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict,
 				struct ttm_operation_ctx *ctx,
-				struct ttm_mem_reg *new_mem)
+				struct ttm_resource *new_mem)
 {
-	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg tmp_mem;
+	struct ttm_resource *old_mem = &bo->mem;
+	struct ttm_resource tmp_mem;
 	struct ttm_placement placement;
 	struct ttm_place placements;
 	int r;
@@ -655,7 +598,8 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict,
 	placement.busy_placement = &placements;
 	placements.fpfn = 0;
 	placements.lpfn = 0;
-	placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
+	placements.mem_type = TTM_PL_TT;
+	placements.flags = TTM_PL_MASK_CACHING;
 	r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx);
 	if (unlikely(r)) {
 		pr_err("Failed to find GTT space for blit to VRAM\n");
@@ -669,12 +613,12 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict,
 	}
 
 	/* copy to VRAM */
-	r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu, new_mem, old_mem);
+	r = amdgpu_move_blit(bo, evict, new_mem, old_mem);
 	if (unlikely(r)) {
 		goto out_cleanup;
 	}
 out_cleanup:
-	ttm_bo_mem_put(bo, &tmp_mem);
+	ttm_resource_free(bo, &tmp_mem);
 	return r;
 }
 
@@ -684,7 +628,7 @@ out_cleanup:
  * Called by amdgpu_bo_move()
  */
 static bool amdgpu_mem_visible(struct amdgpu_device *adev,
-			       struct ttm_mem_reg *mem)
+			       struct ttm_resource *mem)
 {
 	struct drm_mm_node *nodes = mem->mm_node;
 
@@ -694,7 +638,7 @@ static bool amdgpu_mem_visible(struct amdgpu_device *adev,
 	if (mem->mem_type != TTM_PL_VRAM)
 		return false;
 
-	/* ttm_mem_reg_ioremap only supports contiguous memory */
+	/* ttm_resource_ioremap only supports contiguous memory */
 	if (nodes->size != mem->num_pages)
 		return false;
 
@@ -709,11 +653,11 @@ static bool amdgpu_mem_visible(struct amdgpu_device *adev,
  */
 static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 			  struct ttm_operation_ctx *ctx,
-			  struct ttm_mem_reg *new_mem)
+			  struct ttm_resource *new_mem)
 {
 	struct amdgpu_device *adev;
 	struct amdgpu_bo *abo;
-	struct ttm_mem_reg *old_mem = &bo->mem;
+	struct ttm_resource *old_mem = &bo->mem;
 	int r;
 
 	/* Can't move a pinned BO */
@@ -724,7 +668,7 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 	adev = amdgpu_ttm_adev(bo->bdev);
 
 	if (old_mem->mem_type == TTM_PL_SYSTEM && bo->ttm == NULL) {
-		amdgpu_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 	if ((old_mem->mem_type == TTM_PL_TT &&
@@ -732,7 +676,7 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 	    (old_mem->mem_type == TTM_PL_SYSTEM &&
 	     new_mem->mem_type == TTM_PL_TT)) {
 		/* bind is enough */
-		amdgpu_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 	if (old_mem->mem_type == AMDGPU_PL_GDS ||
@@ -742,7 +686,7 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 	    new_mem->mem_type == AMDGPU_PL_GWS ||
 	    new_mem->mem_type == AMDGPU_PL_OA) {
 		/* Nothing to save here */
-		amdgpu_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 
@@ -758,7 +702,7 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 		   new_mem->mem_type == TTM_PL_VRAM) {
 		r = amdgpu_move_ram_vram(bo, evict, ctx, new_mem);
 	} else {
-		r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu,
+		r = amdgpu_move_blit(bo, evict,
 				     new_mem, old_mem);
 	}
 
@@ -795,19 +739,12 @@ memcpy:
  *
  * Called by ttm_mem_io_reserve() ultimately via ttm_bo_vm_fault()
  */
-static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
+static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_resource *mem)
 {
-	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct drm_mm_node *mm_node = mem->mm_node;
+	size_t bus_size = (size_t)mem->num_pages << PAGE_SHIFT;
 
-	mem->bus.addr = NULL;
-	mem->bus.offset = 0;
-	mem->bus.size = mem->num_pages << PAGE_SHIFT;
-	mem->bus.base = 0;
-	mem->bus.is_iomem = false;
-	if (!(man->flags & TTM_MEMTYPE_FLAG_MAPPABLE))
-		return -EINVAL;
 	switch (mem->mem_type) {
 	case TTM_PL_SYSTEM:
 		/* system memory */
@@ -817,18 +754,18 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 	case TTM_PL_VRAM:
 		mem->bus.offset = mem->start << PAGE_SHIFT;
 		/* check if it's visible */
-		if ((mem->bus.offset + mem->bus.size) > adev->gmc.visible_vram_size)
+		if ((mem->bus.offset + bus_size) > adev->gmc.visible_vram_size)
 			return -EINVAL;
 		/* Only physically contiguous buffers apply. In a contiguous
 		 * buffer, size of the first mm_node would match the number of
-		 * pages in ttm_mem_reg.
+		 * pages in ttm_resource.
 		 */
 		if (adev->mman.aper_base_kaddr &&
 		    (mm_node->size == mem->num_pages))
 			mem->bus.addr = (u8 *)adev->mman.aper_base_kaddr +
 					mem->bus.offset;
 
-		mem->bus.base = adev->gmc.aper_base;
+		mem->bus.offset += adev->gmc.aper_base;
 		mem->bus.is_iomem = true;
 		break;
 	default:
@@ -840,12 +777,13 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
 					   unsigned long page_offset)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	uint64_t offset = (page_offset << PAGE_SHIFT);
 	struct drm_mm_node *mm;
 
 	mm = amdgpu_find_mm_node(&bo->mem, &offset);
-	return (bo->mem.bus.base >> PAGE_SHIFT) + mm->start +
-		(offset >> PAGE_SHIFT);
+	offset += adev->gmc.aper_base;
+	return mm->start + (offset >> PAGE_SHIFT);
 }
 
 /**
@@ -879,6 +817,7 @@ struct amdgpu_ttm_tt {
 	uint64_t		userptr;
 	struct task_struct	*usertask;
 	uint32_t		userflags;
+	bool			bound;
 #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR)
 	struct hmm_range	*range;
 #endif
@@ -1046,9 +985,10 @@ void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages)
  *
  * Called by amdgpu_ttm_backend_bind()
  **/
-static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
+static int amdgpu_ttm_tt_pin_userptr(struct ttm_bo_device *bdev,
+				     struct ttm_tt *ttm)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 	int r;
 
@@ -1083,9 +1023,10 @@ release_sg:
 /**
  * amdgpu_ttm_tt_unpin_userptr - Unpin and unmap userptr pages
  */
-static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm)
+static void amdgpu_ttm_tt_unpin_userptr(struct ttm_bo_device *bdev,
+					struct ttm_tt *ttm)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 
 	int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
@@ -1166,16 +1107,23 @@ gart_bind_fail:
  * Called by ttm_tt_bind() on behalf of ttm_bo_handle_move_mem().
  * This handles binding GTT memory to the device address space.
  */
-static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm,
-				   struct ttm_mem_reg *bo_mem)
+static int amdgpu_ttm_backend_bind(struct ttm_bo_device *bdev,
+				   struct ttm_tt *ttm,
+				   struct ttm_resource *bo_mem)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = (void*)ttm;
 	uint64_t flags;
 	int r = 0;
 
+	if (!bo_mem)
+		return -EINVAL;
+
+	if (gtt->bound)
+		return 0;
+
 	if (gtt->userptr) {
-		r = amdgpu_ttm_tt_pin_userptr(ttm);
+		r = amdgpu_ttm_tt_pin_userptr(bdev, ttm);
 		if (r) {
 			DRM_ERROR("failed to pin userptr\n");
 			return r;
@@ -1207,18 +1155,24 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm,
 	if (r)
 		DRM_ERROR("failed to bind %lu pages at 0x%08llX\n",
 			  ttm->num_pages, gtt->offset);
+	gtt->bound = true;
 	return r;
 }
 
 /**
- * amdgpu_ttm_alloc_gart - Allocate GART memory for buffer object
+ * amdgpu_ttm_alloc_gart - Make sure buffer object is accessible either
+ * through AGP or GART aperture.
+ *
+ * If bo is accessible through AGP aperture, then use AGP aperture
+ * to access bo; otherwise allocate logical space in GART aperture
+ * and map bo to GART aperture.
  */
 int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct ttm_operation_ctx ctx = { false, false };
 	struct amdgpu_ttm_tt *gtt = (void*)bo->ttm;
-	struct ttm_mem_reg tmp;
+	struct ttm_resource tmp;
 	struct ttm_placement placement;
 	struct ttm_place placements;
 	uint64_t addr, flags;
@@ -1241,8 +1195,8 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
 		placement.busy_placement = &placements;
 		placements.fpfn = 0;
 		placements.lpfn = adev->gmc.gart_size >> PAGE_SHIFT;
-		placements.flags = (bo->mem.placement & ~TTM_PL_MASK_MEM) |
-			TTM_PL_FLAG_TT;
+		placements.mem_type = TTM_PL_TT;
+		placements.flags = bo->mem.placement;
 
 		r = ttm_bo_mem_space(bo, &placement, &tmp, &ctx);
 		if (unlikely(r))
@@ -1255,11 +1209,11 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
 		gtt->offset = (u64)tmp.start << PAGE_SHIFT;
 		r = amdgpu_ttm_gart_bind(adev, bo, flags);
 		if (unlikely(r)) {
-			ttm_bo_mem_put(bo, &tmp);
+			ttm_resource_free(bo, &tmp);
 			return r;
 		}
 
-		ttm_bo_mem_put(bo, &bo->mem);
+		ttm_resource_free(bo, &bo->mem);
 		bo->mem = tmp;
 	}
 
@@ -1293,15 +1247,19 @@ int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo)
  * Called by ttm_tt_unbind() on behalf of ttm_bo_move_ttm() and
  * ttm_tt_destroy().
  */
-static void amdgpu_ttm_backend_unbind(struct ttm_tt *ttm)
+static void amdgpu_ttm_backend_unbind(struct ttm_bo_device *bdev,
+				      struct ttm_tt *ttm)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 	int r;
 
+	if (!gtt->bound)
+		return;
+
 	/* if the pages have userptr pinning then clear that first */
 	if (gtt->userptr)
-		amdgpu_ttm_tt_unpin_userptr(ttm);
+		amdgpu_ttm_tt_unpin_userptr(bdev, ttm);
 
 	if (gtt->offset == AMDGPU_BO_INVALID_OFFSET)
 		return;
@@ -1311,12 +1269,16 @@ static void amdgpu_ttm_backend_unbind(struct ttm_tt *ttm)
 	if (r)
 		DRM_ERROR("failed to unbind %lu pages at 0x%08llX\n",
 			  gtt->ttm.ttm.num_pages, gtt->offset);
+	gtt->bound = false;
 }
 
-static void amdgpu_ttm_backend_destroy(struct ttm_tt *ttm)
+static void amdgpu_ttm_backend_destroy(struct ttm_bo_device *bdev,
+				       struct ttm_tt *ttm)
 {
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 
+	amdgpu_ttm_backend_unbind(bdev, ttm);
+	ttm_tt_destroy_common(bdev, ttm);
 	if (gtt->usertask)
 		put_task_struct(gtt->usertask);
 
@@ -1324,12 +1286,6 @@ static void amdgpu_ttm_backend_destroy(struct ttm_tt *ttm)
 	kfree(gtt);
 }
 
-static struct ttm_backend_func amdgpu_backend_func = {
-	.bind = &amdgpu_ttm_backend_bind,
-	.unbind = &amdgpu_ttm_backend_unbind,
-	.destroy = &amdgpu_ttm_backend_destroy,
-};
-
 /**
  * amdgpu_ttm_tt_create - Create a ttm_tt object for a given BO
  *
@@ -1346,7 +1302,6 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 	if (gtt == NULL) {
 		return NULL;
 	}
-	gtt->ttm.ttm.func = &amdgpu_backend_func;
 	gtt->gobj = &bo->base;
 
 	/* allocate space for the uninitialized page entries */
@@ -1363,10 +1318,11 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
  * Map the pages of a ttm_tt object to an address space visible
  * to the underlying device.
  */
-static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm,
-			struct ttm_operation_ctx *ctx)
+static int amdgpu_ttm_tt_populate(struct ttm_bo_device *bdev,
+				  struct ttm_tt *ttm,
+				  struct ttm_operation_ctx *ctx)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 
 	/* user pages are bound by amdgpu_ttm_tt_pin_userptr() */
@@ -1376,7 +1332,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm,
 			return -ENOMEM;
 
 		ttm->page_flags |= TTM_PAGE_FLAG_SG;
-		ttm->state = tt_unbound;
+		ttm_tt_set_populated(ttm);
 		return 0;
 	}
 
@@ -1396,7 +1352,7 @@ 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;
+		ttm_tt_set_populated(ttm);
 		return 0;
 	}
 
@@ -1417,7 +1373,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm,
  * Unmaps pages of a ttm_tt object from the device address space and
  * unpopulates the page array backing it.
  */
-static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
+static void amdgpu_ttm_tt_unpopulate(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 	struct amdgpu_device *adev;
@@ -1441,7 +1397,7 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	if (ttm->page_flags & TTM_PAGE_FLAG_SG)
 		return;
 
-	adev = amdgpu_ttm_adev(ttm->bdev);
+	adev = amdgpu_ttm_adev(bdev);
 
 #ifdef CONFIG_SWIOTLB
 	if (adev->need_swiotlb && swiotlb_nr_tbl()) {
@@ -1458,21 +1414,26 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
  * amdgpu_ttm_tt_set_userptr - Initialize userptr GTT ttm_tt for the current
  * task
  *
- * @ttm: The ttm_tt object to bind this userptr object to
+ * @bo: The ttm_buffer_object to bind this userptr to
  * @addr:  The address in the current tasks VM space to use
  * @flags: Requirements of userptr object.
  *
  * Called by amdgpu_gem_userptr_ioctl() to bind userptr pages
  * to current task
  */
-int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
-			      uint32_t flags)
+int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,
+			      uint64_t addr, uint32_t flags)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt;
 
-	if (gtt == NULL)
-		return -EINVAL;
+	if (!bo->ttm) {
+		/* TODO: We want a separate TTM object type for userptrs */
+		bo->ttm = amdgpu_ttm_tt_create(bo, 0);
+		if (bo->ttm == NULL)
+			return -ENOMEM;
+	}
 
+	gtt = (void*)bo->ttm;
 	gtt->userptr = addr;
 	gtt->userflags = flags;
 
@@ -1558,7 +1519,7 @@ bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm)
  *
  * Figure out the flags to use for a VM PDE (Page Directory Entry).
  */
-uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem)
+uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem)
 {
 	uint64_t flags = 0;
 
@@ -1584,7 +1545,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem)
  * Figure out the flags to use for a VM PTE (Page Table Entry).
  */
 uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
-				 struct ttm_mem_reg *mem)
+				 struct ttm_resource *mem)
 {
 	uint64_t flags = amdgpu_ttm_tt_pde_flags(ttm, mem);
 
@@ -1742,7 +1703,9 @@ static struct ttm_bo_driver amdgpu_bo_driver = {
 	.ttm_tt_create = &amdgpu_ttm_tt_create,
 	.ttm_tt_populate = &amdgpu_ttm_tt_populate,
 	.ttm_tt_unpopulate = &amdgpu_ttm_tt_unpopulate,
-	.init_mem_type = &amdgpu_init_mem_type,
+	.ttm_tt_bind = &amdgpu_ttm_backend_bind,
+	.ttm_tt_unbind = &amdgpu_ttm_backend_unbind,
+	.ttm_tt_destroy = &amdgpu_ttm_backend_destroy,
 	.eviction_valuable = amdgpu_ttm_bo_eviction_valuable,
 	.evict_flags = &amdgpu_evict_flags,
 	.move = &amdgpu_bo_move,
@@ -1768,8 +1731,8 @@ static struct ttm_bo_driver amdgpu_bo_driver = {
  */
 static void amdgpu_ttm_fw_reserve_vram_fini(struct amdgpu_device *adev)
 {
-	amdgpu_bo_free_kernel(&adev->fw_vram_usage.reserved_bo,
-		NULL, &adev->fw_vram_usage.va);
+	amdgpu_bo_free_kernel(&adev->mman.fw_vram_usage_reserved_bo,
+		NULL, &adev->mman.fw_vram_usage_va);
 }
 
 /**
@@ -1783,19 +1746,19 @@ static int amdgpu_ttm_fw_reserve_vram_init(struct amdgpu_device *adev)
 {
 	uint64_t vram_size = adev->gmc.visible_vram_size;
 
-	adev->fw_vram_usage.va = NULL;
-	adev->fw_vram_usage.reserved_bo = NULL;
+	adev->mman.fw_vram_usage_va = NULL;
+	adev->mman.fw_vram_usage_reserved_bo = NULL;
 
-	if (adev->fw_vram_usage.size == 0 ||
-	    adev->fw_vram_usage.size > vram_size)
+	if (adev->mman.fw_vram_usage_size == 0 ||
+	    adev->mman.fw_vram_usage_size > vram_size)
 		return 0;
 
 	return amdgpu_bo_create_kernel_at(adev,
-					  adev->fw_vram_usage.start_offset,
-					  adev->fw_vram_usage.size,
+					  adev->mman.fw_vram_usage_start_offset,
+					  adev->mman.fw_vram_usage_size,
 					  AMDGPU_GEM_DOMAIN_VRAM,
-					  &adev->fw_vram_usage.reserved_bo,
-					  &adev->fw_vram_usage.va);
+					  &adev->mman.fw_vram_usage_reserved_bo,
+					  &adev->mman.fw_vram_usage_va);
 }
 
 /*
@@ -1827,7 +1790,7 @@ static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev)
 	memset(ctx, 0, sizeof(*ctx));
 
 	ctx->c2p_train_data_offset =
-		ALIGN((adev->gmc.mc_vram_size - adev->discovery_tmr_size - SZ_1M), SZ_1M);
+		ALIGN((adev->gmc.mc_vram_size - adev->mman.discovery_tmr_size - SZ_1M), SZ_1M);
 	ctx->p2c_train_data_offset =
 		(adev->gmc.mc_vram_size - GDDR6_MEM_TRAINING_OFFSET);
 	ctx->train_data_size =
@@ -1866,10 +1829,10 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	 * Otherwise, fallback to legacy approach to check and reserve tmr block for ip
 	 * discovery data and G6 memory training data respectively
 	 */
-	adev->discovery_tmr_size =
+	adev->mman.discovery_tmr_size =
 		amdgpu_atomfirmware_get_fw_reserved_fb_size(adev);
-	if (!adev->discovery_tmr_size)
-		adev->discovery_tmr_size = DISCOVERY_TMR_OFFSET;
+	if (!adev->mman.discovery_tmr_size)
+		adev->mman.discovery_tmr_size = DISCOVERY_TMR_OFFSET;
 
 	if (mem_train_support) {
 		/* reserve vram for mem train according to TMR location */
@@ -1889,14 +1852,14 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	}
 
 	ret = amdgpu_bo_create_kernel_at(adev,
-				adev->gmc.real_vram_size - adev->discovery_tmr_size,
-				adev->discovery_tmr_size,
+				adev->gmc.real_vram_size - adev->mman.discovery_tmr_size,
+				adev->mman.discovery_tmr_size,
 				AMDGPU_GEM_DOMAIN_VRAM,
-				&adev->discovery_memory,
+				&adev->mman.discovery_memory,
 				NULL);
 	if (ret) {
 		DRM_ERROR("alloc tmr failed(%d)!\n", ret);
-		amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
+		amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
 		return ret;
 	}
 
@@ -1917,15 +1880,14 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	uint64_t gtt_size;
 	int r;
 	u64 vis_vram_limit;
-	void *stolen_vga_buf;
 
 	mutex_init(&adev->mman.gtt_window_lock);
 
 	/* No others user of address space so set it to 0 */
 	r = ttm_bo_device_init(&adev->mman.bdev,
 			       &amdgpu_bo_driver,
-			       adev->ddev->anon_inode->i_mapping,
-			       adev->ddev->vma_offset_manager,
+			       adev_to_drm(adev)->anon_inode->i_mapping,
+			       adev_to_drm(adev)->vma_offset_manager,
 			       dma_addressing_limited(adev->dev));
 	if (r) {
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
@@ -1937,8 +1899,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	adev->mman.bdev.no_retry = true;
 
 	/* Initialize VRAM pool with all of VRAM divided into pages */
-	r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_VRAM,
-				adev->gmc.real_vram_size >> PAGE_SHIFT);
+	r = amdgpu_vram_mgr_init(adev);
 	if (r) {
 		DRM_ERROR("Failed initializing VRAM heap.\n");
 		return r;
@@ -1971,7 +1932,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	 * If IP discovery enabled, a block of memory should be
 	 * reserved for IP discovey.
 	 */
-	if (adev->discovery_bin) {
+	if (adev->mman.discovery_bin) {
 		r = amdgpu_ttm_reserve_tmr(adev);
 		if (r)
 			return r;
@@ -1981,10 +1942,17 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	 * This is used for VGA emulation and pre-OS scanout buffers to
 	 * avoid display artifacts while transitioning between pre-OS
 	 * and driver.  */
-	r = amdgpu_bo_create_kernel(adev, adev->gmc.stolen_size, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_VRAM,
-				    &adev->stolen_vga_memory,
-				    NULL, &stolen_vga_buf);
+	r = amdgpu_bo_create_kernel_at(adev, 0, adev->mman.stolen_vga_size,
+				       AMDGPU_GEM_DOMAIN_VRAM,
+				       &adev->mman.stolen_vga_memory,
+				       NULL);
+	if (r)
+		return r;
+	r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_vga_size,
+				       adev->mman.stolen_extended_size,
+				       AMDGPU_GEM_DOMAIN_VRAM,
+				       &adev->mman.stolen_extended_memory,
+				       NULL);
 	if (r)
 		return r;
 
@@ -2005,7 +1973,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		gtt_size = (uint64_t)amdgpu_gtt_size << 20;
 
 	/* Initialize GTT memory pool */
-	r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_TT, gtt_size >> PAGE_SHIFT);
+	r = amdgpu_gtt_mgr_init(adev, gtt_size);
 	if (r) {
 		DRM_ERROR("Failed initializing GTT heap.\n");
 		return r;
@@ -2014,22 +1982,19 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		 (unsigned)(gtt_size / (1024 * 1024)));
 
 	/* Initialize various on-chip memory pools */
-	r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_GDS,
-			   adev->gds.gds_size);
+	r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GDS, adev->gds.gds_size);
 	if (r) {
 		DRM_ERROR("Failed initializing GDS heap.\n");
 		return r;
 	}
 
-	r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_GWS,
-			   adev->gds.gws_size);
+	r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GWS, adev->gds.gws_size);
 	if (r) {
 		DRM_ERROR("Failed initializing gws heap.\n");
 		return r;
 	}
 
-	r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_OA,
-			   adev->gds.oa_size);
+	r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_OA, adev->gds.oa_size);
 	if (r) {
 		DRM_ERROR("Failed initializing oa heap.\n");
 		return r;
@@ -2043,9 +2008,10 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
  */
 void amdgpu_ttm_late_init(struct amdgpu_device *adev)
 {
-	void *stolen_vga_buf;
 	/* return the VGA stolen memory (if any) back to VRAM */
-	amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, &stolen_vga_buf);
+	if (!adev->mman.keep_stolen_vga_memory)
+		amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
 }
 
 /**
@@ -2057,19 +2023,22 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 		return;
 
 	amdgpu_ttm_training_reserve_vram_fini(adev);
+	/* return the stolen vga memory back to VRAM */
+	if (adev->mman.keep_stolen_vga_memory)
+		amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
 	/* return the IP Discovery TMR memory back to VRAM */
-	amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
 	amdgpu_ttm_fw_reserve_vram_fini(adev);
 
 	if (adev->mman.aper_base_kaddr)
 		iounmap(adev->mman.aper_base_kaddr);
 	adev->mman.aper_base_kaddr = NULL;
 
-	ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_VRAM);
-	ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_TT);
-	ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_GDS);
-	ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_GWS);
-	ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_OA);
+	amdgpu_vram_mgr_fini(adev);
+	amdgpu_gtt_mgr_fini(adev);
+	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_GDS);
+	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_GWS);
+	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_OA);
 	ttm_bo_device_release(&adev->mman.bdev);
 	adev->mman.initialized = false;
 	DRM_INFO("amdgpu: ttm finalized\n");
@@ -2086,11 +2055,11 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
  */
 void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 {
-	struct ttm_mem_type_manager *man = &adev->mman.bdev.man[TTM_PL_VRAM];
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 	uint64_t size;
 	int r;
 
-	if (!adev->mman.initialized || adev->in_gpu_reset ||
+	if (!adev->mman.initialized || amdgpu_in_reset(adev) ||
 	    adev->mman.buffer_funcs_enabled == enable)
 		return;
 
@@ -2101,7 +2070,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 		ring = adev->mman.buffer_funcs_ring;
 		sched = &ring->sched;
 		r = drm_sched_entity_init(&adev->mman.entity,
-				          DRM_SCHED_PRIORITY_KERNEL, &sched,
+					  DRM_SCHED_PRIORITY_KERNEL, &sched,
 					  1, NULL);
 		if (r) {
 			DRM_ERROR("Failed setting up TTM BO move entity (%d)\n",
@@ -2126,7 +2095,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 int amdgpu_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	struct drm_file *file_priv = filp->private_data;
-	struct amdgpu_device *adev = file_priv->minor->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(file_priv->minor->dev);
 
 	if (adev == NULL)
 		return -EINVAL;
@@ -2307,8 +2276,8 @@ 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 = (uintptr_t)node->info_ent->data;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
-	struct ttm_mem_type_manager *man = &adev->mman.bdev.man[ttm_pl];
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, ttm_pl);
 	struct drm_printer p = drm_seq_file_printer(m);
 
 	man->func->debug(man, &p);
@@ -2598,7 +2567,7 @@ int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev)
 #if defined(CONFIG_DEBUG_FS)
 	unsigned count;
 
-	struct drm_minor *minor = adev->ddev->primary;
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
 	struct dentry *ent, *root = minor->debugfs_root;
 
 	for (count = 0; count < ARRAY_SIZE(ttm_debugfs_entries); count++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 17c8d0d7bcc3..a87951b2f06d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -32,15 +32,26 @@
 #define AMDGPU_PL_GWS		(TTM_PL_PRIV + 1)
 #define AMDGPU_PL_OA		(TTM_PL_PRIV + 2)
 
-#define AMDGPU_PL_FLAG_GDS		(TTM_PL_FLAG_PRIV << 0)
-#define AMDGPU_PL_FLAG_GWS		(TTM_PL_FLAG_PRIV << 1)
-#define AMDGPU_PL_FLAG_OA		(TTM_PL_FLAG_PRIV << 2)
-
 #define AMDGPU_GTT_MAX_TRANSFER_SIZE	512
 #define AMDGPU_GTT_NUM_TRANSFER_WINDOWS	2
 
 #define AMDGPU_POISON	0xd0bed0be
 
+struct amdgpu_vram_mgr {
+	struct ttm_resource_manager manager;
+	struct drm_mm mm;
+	spinlock_t lock;
+	atomic64_t usage;
+	atomic64_t vis_usage;
+};
+
+struct amdgpu_gtt_mgr {
+	struct ttm_resource_manager manager;
+	struct drm_mm mm;
+	spinlock_t lock;
+	atomic64_t available;
+};
+
 struct amdgpu_mman {
 	struct ttm_bo_device		bdev;
 	bool				mem_global_referenced;
@@ -59,24 +70,46 @@ struct amdgpu_mman {
 	struct mutex				gtt_window_lock;
 	/* Scheduler entity for buffer moves */
 	struct drm_sched_entity			entity;
+
+	struct amdgpu_vram_mgr vram_mgr;
+	struct amdgpu_gtt_mgr gtt_mgr;
+
+	uint64_t		stolen_vga_size;
+	struct amdgpu_bo	*stolen_vga_memory;
+	uint64_t		stolen_extended_size;
+	struct amdgpu_bo	*stolen_extended_memory;
+	bool			keep_stolen_vga_memory;
+
+	/* discovery */
+	uint8_t				*discovery_bin;
+	uint32_t			discovery_tmr_size;
+	struct amdgpu_bo		*discovery_memory;
+
+	/* firmware VRAM reservation */
+	u64		fw_vram_usage_start_offset;
+	u64		fw_vram_usage_size;
+	struct amdgpu_bo	*fw_vram_usage_reserved_bo;
+	void		*fw_vram_usage_va;
 };
 
 struct amdgpu_copy_mem {
 	struct ttm_buffer_object	*bo;
-	struct ttm_mem_reg		*mem;
+	struct ttm_resource		*mem;
 	unsigned long			offset;
 };
 
-extern const struct ttm_mem_type_manager_func amdgpu_gtt_mgr_func;
-extern const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func;
+int amdgpu_gtt_mgr_init(struct amdgpu_device *adev, uint64_t gtt_size);
+void amdgpu_gtt_mgr_fini(struct amdgpu_device *adev);
+int amdgpu_vram_mgr_init(struct amdgpu_device *adev);
+void amdgpu_vram_mgr_fini(struct amdgpu_device *adev);
 
-bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_mem_reg *mem);
-uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man);
-int amdgpu_gtt_mgr_recover(struct ttm_mem_type_manager *man);
+bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_resource *mem);
+uint64_t amdgpu_gtt_mgr_usage(struct ttm_resource_manager *man);
+int amdgpu_gtt_mgr_recover(struct ttm_resource_manager *man);
 
 u64 amdgpu_vram_mgr_bo_visible_size(struct amdgpu_bo *bo);
 int amdgpu_vram_mgr_alloc_sgt(struct amdgpu_device *adev,
-			      struct ttm_mem_reg *mem,
+			      struct ttm_resource *mem,
 			      struct device *dev,
 			      enum dma_data_direction dir,
 			      struct sg_table **sgt);
@@ -84,8 +117,8 @@ void amdgpu_vram_mgr_free_sgt(struct amdgpu_device *adev,
 			      struct device *dev,
 			      enum dma_data_direction dir,
 			      struct sg_table *sgt);
-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);
+uint64_t amdgpu_vram_mgr_usage(struct ttm_resource_manager *man);
+uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_resource_manager *man);
 
 int amdgpu_ttm_init(struct amdgpu_device *adev);
 void amdgpu_ttm_late_init(struct amdgpu_device *adev);
@@ -130,8 +163,8 @@ static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm)
 #endif
 
 void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages);
-int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
-				     uint32_t flags);
+int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,
+			      uint64_t addr, uint32_t flags);
 bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm);
 struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm);
 bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
@@ -140,9 +173,9 @@ bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
 				       int *last_invalidated);
 bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm);
 bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm);
-uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem);
+uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem);
 uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
-				 struct ttm_mem_reg *mem);
+				 struct ttm_resource *mem);
 
 int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
index 183743c5fb7b..55fe19a2f332 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
@@ -408,7 +408,7 @@ static ssize_t show_##name(struct device *dev,				\
 			  char *buf)					\
 {									\
 	struct drm_device *ddev = dev_get_drvdata(dev);			\
-	struct amdgpu_device *adev = ddev->dev_private;			\
+	struct amdgpu_device *adev = drm_to_adev(ddev);			\
 									\
 	return snprintf(buf, PAGE_SIZE, "0x%08x\n", adev->field);	\
 }									\
@@ -628,7 +628,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev)
 	struct amdgpu_firmware_info *ucode = NULL;
 
  /* for baremetal, the ucode is allocated in gtt, so don't need to fill the bo when reset/suspend */
-	if (!amdgpu_sriov_vf(adev) && (adev->in_gpu_reset || adev->in_suspend))
+	if (!amdgpu_sriov_vf(adev) && (amdgpu_in_reset(adev) || adev->in_suspend))
 		return 0;
 	/*
 	 * if SMU loaded firmware, it needn't add SMC, UVD, and VCE
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
index 12a8bc8fca0b..3c23c6293ff9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
@@ -131,6 +131,7 @@ enum ta_fw_type {
 	TA_FW_TYPE_PSP_RAS,
 	TA_FW_TYPE_PSP_HDCP,
 	TA_FW_TYPE_PSP_DTM,
+	TA_FW_TYPE_PSP_RAP,
 };
 
 struct ta_fw_bin_desc {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
index af1b1ccf613c..262baf0f61ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@@ -125,8 +125,9 @@ int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev,
 				"detected in UMC block\n",
 				err_data->ue_count);
 
-		if (err_data->err_addr_cnt &&
-		    amdgpu_ras_add_bad_pages(adev, err_data->err_addr,
+		if ((amdgpu_bad_page_threshold != 0) &&
+			err_data->err_addr_cnt &&
+			amdgpu_ras_add_bad_pages(adev, err_data->err_addr,
 						err_data->err_addr_cnt))
 			dev_warn(adev->dev, "Failed to add ras bad page!\n");
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
index a615a1eb750b..183814493658 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
@@ -21,6 +21,20 @@
 #ifndef __AMDGPU_UMC_H__
 #define __AMDGPU_UMC_H__
 
+/*
+ * (addr / 256) * 8192, the higher 26 bits in ErrorAddr
+ * is the index of 8KB block
+ */
+#define ADDR_OF_8KB_BLOCK(addr)			(((addr) & ~0xffULL) << 5)
+/* channel index is the index of 256B block */
+#define ADDR_OF_256B_BLOCK(channel_index)	((channel_index) << 8)
+/* offset in 256B block */
+#define OFFSET_IN_256B_BLOCK(addr)		((addr) & 0xffULL)
+
+#define LOOP_UMC_INST(umc_inst) for ((umc_inst) = 0; (umc_inst) < adev->umc.umc_inst_num; (umc_inst)++)
+#define LOOP_UMC_CH_INST(ch_inst) for ((ch_inst) = 0; (ch_inst) < adev->umc.channel_inst_num; (ch_inst)++)
+#define LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) LOOP_UMC_INST((umc_inst)) LOOP_UMC_CH_INST((ch_inst))
+
 struct amdgpu_umc_funcs {
 	void (*err_cnt_init)(struct amdgpu_device *adev);
 	int (*ras_late_init)(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 1203c20491e6..d0aea5e39531 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -31,6 +31,12 @@
 #include "soc15.h"
 #include "nv.h"
 
+#define POPULATE_UCODE_INFO(vf2pf_info, ucode, ver) \
+	do { \
+		vf2pf_info->ucode_info[ucode].id = ucode; \
+		vf2pf_info->ucode_info[ucode].version = ver; \
+	} while (0)
+
 bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev)
 {
 	/* By now all MMIO pages except mailbox are blocked */
@@ -45,7 +51,7 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	if (adev->mode_info.num_crtc == 0)
 		adev->mode_info.num_crtc = 1;
 	adev->enable_virtual_display = true;
-	adev->ddev->driver->driver_features &= ~DRIVER_ATOMIC;
+	adev_to_drm(adev)->driver->driver_features &= ~DRIVER_ATOMIC;
 	adev->cg_flags = 0;
 	adev->pg_flags = 0;
 }
@@ -93,7 +99,7 @@ failed_undo:
 	amdgpu_ring_undo(ring);
 	spin_unlock_irqrestore(&kiq->ring_lock, flags);
 failed_kiq:
-	pr_err("failed to write reg %x wait reg %x\n", reg0, reg1);
+	dev_err(adev->dev, "failed to write reg %x wait reg %x\n", reg0, reg1);
 }
 
 /**
@@ -239,10 +245,10 @@ void amdgpu_virt_free_mm_table(struct amdgpu_device *adev)
 }
 
 
-int amdgpu_virt_fw_reserve_get_checksum(void *obj,
-					unsigned long obj_size,
-					unsigned int key,
-					unsigned int chksum)
+unsigned int amd_sriov_msg_checksum(void *obj,
+				unsigned long obj_size,
+				unsigned int key,
+				unsigned int checksum)
 {
 	unsigned int ret = key;
 	unsigned long i = 0;
@@ -252,9 +258,9 @@ int amdgpu_virt_fw_reserve_get_checksum(void *obj,
 	/* calculate checksum */
 	for (i = 0; i < obj_size; ++i)
 		ret += *(pos + i);
-	/* minus the chksum itself */
-	pos = (char *)&chksum;
-	for (i = 0; i < sizeof(chksum); ++i)
+	/* minus the checksum itself */
+	pos = (char *)&checksum;
+	for (i = 0; i < sizeof(checksum); ++i)
 		ret -= *(pos + i);
 	return ret;
 }
@@ -401,7 +407,7 @@ static void amdgpu_virt_add_bad_page(struct amdgpu_device *adev,
 	if (bp_block_size) {
 		bp_cnt = bp_block_size / sizeof(uint64_t);
 		for (bp_idx = 0; bp_idx < bp_cnt; bp_idx++) {
-			retired_page = *(uint64_t *)(adev->fw_vram_usage.va +
+			retired_page = *(uint64_t *)(adev->mman.fw_vram_usage_va +
 					bp_block_offset + bp_idx * sizeof(uint64_t));
 			bp.retired_page = retired_page;
 
@@ -415,33 +421,188 @@ static void amdgpu_virt_add_bad_page(struct amdgpu_device *adev,
 	}
 }
 
-void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
+static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev)
 {
-	uint32_t pf2vf_size = 0;
-	uint32_t checksum = 0;
+	struct amd_sriov_msg_pf2vf_info_header *pf2vf_info = adev->virt.fw_reserve.p_pf2vf;
+	uint32_t checksum;
 	uint32_t checkval;
-	char *str;
+
+	if (adev->virt.fw_reserve.p_pf2vf == NULL)
+		return -EINVAL;
+
+	if (pf2vf_info->size > 1024) {
+		DRM_ERROR("invalid pf2vf message size\n");
+		return -EINVAL;
+	}
+
+	switch (pf2vf_info->version) {
+	case 1:
+		checksum = ((struct amdgim_pf2vf_info_v1 *)pf2vf_info)->checksum;
+		checkval = amd_sriov_msg_checksum(
+			adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size,
+			adev->virt.fw_reserve.checksum_key, checksum);
+		if (checksum != checkval) {
+			DRM_ERROR("invalid pf2vf message\n");
+			return -EINVAL;
+		}
+
+		adev->virt.gim_feature =
+			((struct amdgim_pf2vf_info_v1 *)pf2vf_info)->feature_flags;
+		break;
+	case 2:
+		/* TODO: missing key, need to add it later */
+		checksum = ((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->checksum;
+		checkval = amd_sriov_msg_checksum(
+			adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size,
+			0, checksum);
+		if (checksum != checkval) {
+			DRM_ERROR("invalid pf2vf message\n");
+			return -EINVAL;
+		}
+
+		adev->virt.vf2pf_update_interval_ms =
+			((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->vf2pf_update_interval_ms;
+		adev->virt.gim_feature =
+			((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->feature_flags.all;
+
+		break;
+	default:
+		DRM_ERROR("invalid pf2vf version\n");
+		return -EINVAL;
+	}
+
+	/* correct too large or too little interval value */
+	if (adev->virt.vf2pf_update_interval_ms < 200 || adev->virt.vf2pf_update_interval_ms > 10000)
+		adev->virt.vf2pf_update_interval_ms = 2000;
+
+	return 0;
+}
+
+static void amdgpu_virt_populate_vf2pf_ucode_info(struct amdgpu_device *adev)
+{
+	struct amd_sriov_msg_vf2pf_info *vf2pf_info;
+	vf2pf_info = (struct amd_sriov_msg_vf2pf_info *) adev->virt.fw_reserve.p_vf2pf;
+
+	if (adev->virt.fw_reserve.p_vf2pf == NULL)
+		return;
+
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_VCE,      adev->vce.fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_UVD,      adev->uvd.fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MC,       adev->gmc.fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_ME,       adev->gfx.me_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_PFP,      adev->gfx.pfp_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_CE,       adev->gfx.ce_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC,      adev->gfx.rlc_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC_SRLC, adev->gfx.rlc_srlc_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC_SRLG, adev->gfx.rlc_srlg_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC_SRLS, adev->gfx.rlc_srls_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC,      adev->gfx.mec_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC2,     adev->gfx.mec2_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SOS,      adev->psp.sos_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_ASD,      adev->psp.asd_fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_TA_RAS,   adev->psp.ta_ras_ucode_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_TA_XGMI,  adev->psp.ta_xgmi_ucode_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SMC,      adev->pm.fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SDMA,     adev->sdma.instance[0].fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SDMA2,    adev->sdma.instance[1].fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_VCN,      adev->vcn.fw_version);
+	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_DMCU,     adev->dm.dmcu_fw_version);
+}
+
+static int amdgpu_virt_write_vf2pf_data(struct amdgpu_device *adev)
+{
+	struct amd_sriov_msg_vf2pf_info *vf2pf_info;
+	struct ttm_resource_manager *vram_man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
+
+	vf2pf_info = (struct amd_sriov_msg_vf2pf_info *) adev->virt.fw_reserve.p_vf2pf;
+
+	if (adev->virt.fw_reserve.p_vf2pf == NULL)
+		return -EINVAL;
+
+	memset(vf2pf_info, 0, sizeof(struct amd_sriov_msg_vf2pf_info));
+
+	vf2pf_info->header.size = sizeof(struct amd_sriov_msg_vf2pf_info);
+	vf2pf_info->header.version = AMD_SRIOV_MSG_FW_VRAM_VF2PF_VER;
+
+#ifdef MODULE
+	if (THIS_MODULE->version != NULL)
+		strcpy(vf2pf_info->driver_version, THIS_MODULE->version);
+	else
+#endif
+		strcpy(vf2pf_info->driver_version, "N/A");
+
+	vf2pf_info->pf2vf_version_required = 0; // no requirement, guest understands all
+	vf2pf_info->driver_cert = 0;
+	vf2pf_info->os_info.all = 0;
+
+	vf2pf_info->fb_usage = amdgpu_vram_mgr_usage(vram_man) >> 20;
+	vf2pf_info->fb_vis_usage = amdgpu_vram_mgr_vis_usage(vram_man) >> 20;
+	vf2pf_info->fb_size = adev->gmc.real_vram_size >> 20;
+	vf2pf_info->fb_vis_size = adev->gmc.visible_vram_size >> 20;
+
+	amdgpu_virt_populate_vf2pf_ucode_info(adev);
+
+	/* TODO: read dynamic info */
+	vf2pf_info->gfx_usage = 0;
+	vf2pf_info->compute_usage = 0;
+	vf2pf_info->encode_usage = 0;
+	vf2pf_info->decode_usage = 0;
+
+	vf2pf_info->checksum =
+		amd_sriov_msg_checksum(
+		vf2pf_info, vf2pf_info->header.size, 0, 0);
+
+	return 0;
+}
+
+void amdgpu_virt_update_vf2pf_work_item(struct work_struct *work)
+{
+	struct amdgpu_device *adev = container_of(work, struct amdgpu_device, virt.vf2pf_work.work);
+
+	amdgpu_virt_read_pf2vf_data(adev);
+	amdgpu_virt_write_vf2pf_data(adev);
+
+	schedule_delayed_work(&(adev->virt.vf2pf_work), adev->virt.vf2pf_update_interval_ms);
+}
+
+void amdgpu_virt_fini_data_exchange(struct amdgpu_device *adev)
+{
+	if (adev->virt.vf2pf_update_interval_ms != 0) {
+		DRM_INFO("clean up the vf2pf work item\n");
+		flush_delayed_work(&adev->virt.vf2pf_work);
+		cancel_delayed_work_sync(&adev->virt.vf2pf_work);
+	}
+}
+
+void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
+{
 	uint64_t bp_block_offset = 0;
 	uint32_t bp_block_size = 0;
-	struct amdgim_pf2vf_info_v2 *pf2vf_v2 = NULL;
+	struct amd_sriov_msg_pf2vf_info *pf2vf_v2 = NULL;
 
 	adev->virt.fw_reserve.p_pf2vf = NULL;
 	adev->virt.fw_reserve.p_vf2pf = NULL;
+	adev->virt.vf2pf_update_interval_ms = 0;
+
+	if (adev->mman.fw_vram_usage_va != NULL) {
+		adev->virt.vf2pf_update_interval_ms = 2000;
 
-	if (adev->fw_vram_usage.va != NULL) {
 		adev->virt.fw_reserve.p_pf2vf =
-			(struct amd_sriov_msg_pf2vf_info_header *)(
-			adev->fw_vram_usage.va + AMDGIM_DATAEXCHANGE_OFFSET);
-		AMDGPU_FW_VRAM_PF2VF_READ(adev, header.size, &pf2vf_size);
-		AMDGPU_FW_VRAM_PF2VF_READ(adev, checksum, &checksum);
-		AMDGPU_FW_VRAM_PF2VF_READ(adev, feature_flags, &adev->virt.gim_feature);
-
-		/* pf2vf message must be in 4K */
-		if (pf2vf_size > 0 && pf2vf_size < 4096) {
-			if (adev->virt.fw_reserve.p_pf2vf->version == 2) {
-				pf2vf_v2 = (struct amdgim_pf2vf_info_v2 *)adev->virt.fw_reserve.p_pf2vf;
-				bp_block_offset = ((uint64_t)pf2vf_v2->bp_block_offset_L & 0xFFFFFFFF) |
-						((((uint64_t)pf2vf_v2->bp_block_offset_H) << 32) & 0xFFFFFFFF00000000);
+			(struct amd_sriov_msg_pf2vf_info_header *)
+			(adev->mman.fw_vram_usage_va + (AMD_SRIOV_MSG_PF2VF_OFFSET_KB << 10));
+		adev->virt.fw_reserve.p_vf2pf =
+			(struct amd_sriov_msg_vf2pf_info_header *)
+			(adev->mman.fw_vram_usage_va + (AMD_SRIOV_MSG_VF2PF_OFFSET_KB << 10));
+
+		amdgpu_virt_read_pf2vf_data(adev);
+		amdgpu_virt_write_vf2pf_data(adev);
+
+		/* bad page handling for version 2 */
+		if (adev->virt.fw_reserve.p_pf2vf->version == 2) {
+				pf2vf_v2 = (struct amd_sriov_msg_pf2vf_info *)adev->virt.fw_reserve.p_pf2vf;
+
+				bp_block_offset = ((uint64_t)pf2vf_v2->bp_block_offset_low & 0xFFFFFFFF) |
+						((((uint64_t)pf2vf_v2->bp_block_offset_high) << 32) & 0xFFFFFFFF00000000);
 				bp_block_size = pf2vf_v2->bp_block_size;
 
 				if (bp_block_size && !adev->virt.ras_init_done)
@@ -450,37 +611,11 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
 				if (adev->virt.ras_init_done)
 					amdgpu_virt_add_bad_page(adev, bp_block_offset, bp_block_size);
 			}
+	}
 
-			checkval = amdgpu_virt_fw_reserve_get_checksum(
-				adev->virt.fw_reserve.p_pf2vf, pf2vf_size,
-				adev->virt.fw_reserve.checksum_key, checksum);
-			if (checkval == checksum) {
-				adev->virt.fw_reserve.p_vf2pf =
-					((void *)adev->virt.fw_reserve.p_pf2vf +
-					pf2vf_size);
-				memset((void *)adev->virt.fw_reserve.p_vf2pf, 0,
-					sizeof(amdgim_vf2pf_info));
-				AMDGPU_FW_VRAM_VF2PF_WRITE(adev, header.version,
-					AMDGPU_FW_VRAM_VF2PF_VER);
-				AMDGPU_FW_VRAM_VF2PF_WRITE(adev, header.size,
-					sizeof(amdgim_vf2pf_info));
-				AMDGPU_FW_VRAM_VF2PF_READ(adev, driver_version,
-					&str);
-#ifdef MODULE
-				if (THIS_MODULE->version != NULL)
-					strcpy(str, THIS_MODULE->version);
-				else
-#endif
-					strcpy(str, "N/A");
-				AMDGPU_FW_VRAM_VF2PF_WRITE(adev, driver_cert,
-					0);
-				AMDGPU_FW_VRAM_VF2PF_WRITE(adev, checksum,
-					amdgpu_virt_fw_reserve_get_checksum(
-					adev->virt.fw_reserve.p_vf2pf,
-					pf2vf_size,
-					adev->virt.fw_reserve.checksum_key, 0));
-			}
-		}
+	if (adev->virt.vf2pf_update_interval_ms != 0) {
+		INIT_DELAYED_WORK(&adev->virt.vf2pf_work, amdgpu_virt_update_vf2pf_work_item);
+		schedule_delayed_work(&(adev->virt.vf2pf_work), adev->virt.vf2pf_update_interval_ms);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index f826945989c7..8dd624c20f89 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -24,6 +24,8 @@
 #ifndef AMDGPU_VIRT_H
 #define AMDGPU_VIRT_H
 
+#include "amdgv_sriovmsg.h"
+
 #define AMDGPU_SRIOV_CAPS_SRIOV_VBIOS  (1 << 0) /* vBIOS is sr-iov ready */
 #define AMDGPU_SRIOV_CAPS_ENABLE_IOV   (1 << 1) /* sr-iov is enabled on this GPU */
 #define AMDGPU_SRIOV_CAPS_IS_VF        (1 << 2) /* this GPU is a virtual function */
@@ -79,7 +81,10 @@ struct amdgpu_virt_fw_reserve {
 	struct amd_sriov_msg_vf2pf_info_header *p_vf2pf;
 	unsigned int checksum_key;
 };
+
 /*
+ * Legacy GIM header
+ *
  * Defination between PF and VF
  * Structures forcibly aligned to 4 to keep the same style as PF.
  */
@@ -101,15 +106,7 @@ enum AMDGIM_FEATURE_FLAG {
 	AMDGIM_FEATURE_PP_ONE_VF = (1 << 4),
 };
 
-struct amd_sriov_msg_pf2vf_info_header {
-	/* the total structure size in byte. */
-	uint32_t size;
-	/* version of this structure, written by the GIM */
-	uint32_t version;
-	/* reserved */
-	uint32_t reserved[2];
-} __aligned(4);
-struct  amdgim_pf2vf_info_v1 {
+struct amdgim_pf2vf_info_v1 {
 	/* header contains size and version */
 	struct amd_sriov_msg_pf2vf_info_header header;
 	/* max_width * max_height */
@@ -128,54 +125,6 @@ struct  amdgim_pf2vf_info_v1 {
 	unsigned int checksum;
 } __aligned(4);
 
-struct  amdgim_pf2vf_info_v2 {
-	/* header contains size and version */
-	struct amd_sriov_msg_pf2vf_info_header header;
-	/* use private key from mailbox 2 to create chueksum */
-	uint32_t checksum;
-	/* The features flags of the GIM driver supports. */
-	uint32_t feature_flags;
-	/* max_width * max_height */
-	uint32_t uvd_enc_max_pixels_count;
-	/* 16x16 pixels/sec, codec independent */
-	uint32_t uvd_enc_max_bandwidth;
-	/* max_width * max_height */
-	uint32_t vce_enc_max_pixels_count;
-	/* 16x16 pixels/sec, codec independent */
-	uint32_t vce_enc_max_bandwidth;
-	/* Bad pages block position in BYTE */
-	uint32_t bp_block_offset_L;
-	uint32_t bp_block_offset_H;
-	/* Bad pages block size in BYTE */
-	uint32_t bp_block_size;
-	/* MEC FW position in kb from the start of VF visible frame buffer */
-	uint32_t mecfw_kboffset_L;
-	uint32_t mecfw_kboffset_H;
-	/* MEC FW size in KB */
-	uint32_t mecfw_ksize;
-	/* UVD FW position in kb from the start of VF visible frame buffer */
-	uint32_t uvdfw_kboffset_L;
-	uint32_t uvdfw_kboffset_H;
-	/* UVD FW size in KB */
-	uint32_t uvdfw_ksize;
-	/* VCE FW position in kb from the start of VF visible frame buffer */
-	uint32_t vcefw_kboffset_L;
-	uint32_t vcefw_kboffset_H;
-	/* VCE FW size in KB */
-	uint32_t vcefw_ksize;
-	uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 0, 0, (18 + sizeof(struct amd_sriov_msg_pf2vf_info_header)/sizeof(uint32_t)), 0)];
-} __aligned(4);
-
-
-struct amd_sriov_msg_vf2pf_info_header {
-	/* the total structure size in byte. */
-	uint32_t size;
-	/*version of this structure, written by the guest */
-	uint32_t version;
-	/* reserved */
-	uint32_t reserved[2];
-} __aligned(4);
-
 struct amdgim_vf2pf_info_v1 {
 	/* header contains size and version */
 	struct amd_sriov_msg_vf2pf_info_header header;
@@ -237,31 +186,6 @@ struct amdgim_vf2pf_info_v2 {
 	uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 64, 0, (12 + sizeof(struct amd_sriov_msg_vf2pf_info_header)/sizeof(uint32_t)), 0)];
 } __aligned(4);
 
-#define AMDGPU_FW_VRAM_VF2PF_VER 2
-typedef struct amdgim_vf2pf_info_v2 amdgim_vf2pf_info ;
-
-#define AMDGPU_FW_VRAM_VF2PF_WRITE(adev, field, val) \
-	do { \
-		((amdgim_vf2pf_info *)adev->virt.fw_reserve.p_vf2pf)->field = (val); \
-	} while (0)
-
-#define AMDGPU_FW_VRAM_VF2PF_READ(adev, field, val) \
-	do { \
-		(*val) = ((amdgim_vf2pf_info *)adev->virt.fw_reserve.p_vf2pf)->field; \
-	} while (0)
-
-#define AMDGPU_FW_VRAM_PF2VF_READ(adev, field, val) \
-	do { \
-		if (!adev->virt.fw_reserve.p_pf2vf) \
-			*(val) = 0; \
-		else { \
-			if (adev->virt.fw_reserve.p_pf2vf->version == 1) \
-				*(val) = ((struct amdgim_pf2vf_info_v1 *)adev->virt.fw_reserve.p_pf2vf)->field; \
-			if (adev->virt.fw_reserve.p_pf2vf->version == 2) \
-				*(val) = ((struct amdgim_pf2vf_info_v2 *)adev->virt.fw_reserve.p_pf2vf)->field; \
-		} \
-	} while (0)
-
 struct amdgpu_virt_ras_err_handler_data {
 	/* point to bad page records array */
 	struct eeprom_table_record *bps;
@@ -285,7 +209,7 @@ struct amdgpu_virt {
 	struct work_struct		flr_work;
 	struct amdgpu_mm_table		mm_table;
 	const struct amdgpu_virt_ops	*ops;
-	struct amdgpu_vf_error_buffer   vf_errors;
+	struct amdgpu_vf_error_buffer	vf_errors;
 	struct amdgpu_virt_fw_reserve	fw_reserve;
 	uint32_t gim_feature;
 	uint32_t reg_access_mode;
@@ -293,6 +217,10 @@ struct amdgpu_virt {
 	bool tdr_debug;
 	struct amdgpu_virt_ras_err_handler_data *virt_eh_data;
 	bool ras_init_done;
+
+	/* vf2pf message */
+	struct delayed_work vf2pf_work;
+	uint32_t vf2pf_update_interval_ms;
 };
 
 #define amdgpu_sriov_enabled(adev) \
@@ -325,9 +253,9 @@ static inline bool is_virtual_machine(void)
 #define amdgpu_sriov_is_pp_one_vf(adev) \
 	((adev)->virt.gim_feature & AMDGIM_FEATURE_PP_ONE_VF)
 #define amdgpu_sriov_is_debug(adev) \
-	((!adev->in_gpu_reset) && adev->virt.tdr_debug)
+	((!amdgpu_in_reset(adev)) && adev->virt.tdr_debug)
 #define amdgpu_sriov_is_normal(adev) \
-	((!adev->in_gpu_reset) && (!adev->virt.tdr_debug))
+	((!amdgpu_in_reset(adev)) && (!adev->virt.tdr_debug))
 
 bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev);
 void amdgpu_virt_init_setting(struct amdgpu_device *adev);
@@ -341,11 +269,9 @@ void amdgpu_virt_request_init_data(struct amdgpu_device *adev);
 int amdgpu_virt_wait_reset(struct amdgpu_device *adev);
 int amdgpu_virt_alloc_mm_table(struct amdgpu_device *adev);
 void amdgpu_virt_free_mm_table(struct amdgpu_device *adev);
-int amdgpu_virt_fw_reserve_get_checksum(void *obj, unsigned long obj_size,
-					unsigned int key,
-					unsigned int chksum);
 void amdgpu_virt_release_ras_err_handler_data(struct amdgpu_device *adev);
 void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev);
+void amdgpu_virt_fini_data_exchange(struct amdgpu_device *adev);
 void amdgpu_detect_virtualization(struct amdgpu_device *adev);
 
 bool amdgpu_virt_can_access_debugfs(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index cb1d7cddebc3..df110afa97bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -28,6 +28,7 @@
 #include <linux/dma-fence-array.h>
 #include <linux/interval_tree_generic.h>
 #include <linux/idr.h>
+#include <linux/dma-buf.h>
 
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
@@ -35,6 +36,7 @@
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_gmc.h"
 #include "amdgpu_xgmi.h"
+#include "amdgpu_dma_buf.h"
 
 /**
  * DOC: GPUVM
@@ -1500,6 +1502,8 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 
 			pt = cursor.entry->base.bo;
 			shift = parent_shift;
+			frag_end = max(frag_end, ALIGN(frag_start + 1,
+				   1ULL << shift));
 		}
 
 		/* Looks good so far, calculate parameters for the update */
@@ -1511,19 +1515,26 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 		entry_end = min(entry_end, end);
 
 		do {
+			struct amdgpu_vm *vm = params->vm;
 			uint64_t upd_end = min(entry_end, frag_end);
 			unsigned nptes = (upd_end - frag_start) >> shift;
+			uint64_t upd_flags = flags | AMDGPU_PTE_FRAG(frag);
 
 			/* This can happen when we set higher level PDs to
 			 * silent to stop fault floods.
 			 */
 			nptes = max(nptes, 1u);
+
+			trace_amdgpu_vm_update_ptes(params, frag_start, upd_end,
+						    nptes, dst, incr, upd_flags,
+						    vm->task_info.pid,
+						    vm->immediate.fence_context);
 			amdgpu_vm_update_flags(params, pt, cursor.level,
 					       pe_start, dst, nptes, incr,
-					       flags | AMDGPU_PTE_FRAG(frag));
+					       upd_flags);
 
 			pe_start += nptes * 8;
-			dst += (uint64_t)nptes * AMDGPU_GPU_PAGE_SIZE << shift;
+			dst += nptes * incr;
 
 			frag_start = upd_end;
 			if (frag_start >= frag_end) {
@@ -1691,13 +1702,13 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 		uint64_t max_entries;
 		uint64_t addr, last;
 
+		max_entries = mapping->last - start + 1;
 		if (nodes) {
 			addr = nodes->start << PAGE_SHIFT;
-			max_entries = (nodes->size - pfn) *
-				AMDGPU_GPU_PAGES_IN_CPU_PAGE;
+			max_entries = min((nodes->size - pfn) *
+				AMDGPU_GPU_PAGES_IN_CPU_PAGE, max_entries);
 		} else {
 			addr = 0;
-			max_entries = S64_MAX;
 		}
 
 		if (pages_addr) {
@@ -1727,7 +1738,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 			addr += pfn << PAGE_SHIFT;
 		}
 
-		last = min((uint64_t)mapping->last, start + max_entries - 1);
+		last = start + max_entries - 1;
 		r = amdgpu_vm_bo_update_mapping(adev, vm, false, false, resv,
 						start, last, flags, addr,
 						dma_addr, fence);
@@ -1765,7 +1776,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	struct amdgpu_vm *vm = bo_va->base.vm;
 	struct amdgpu_bo_va_mapping *mapping;
 	dma_addr_t *pages_addr = NULL;
-	struct ttm_mem_reg *mem;
+	struct ttm_resource *mem;
 	struct drm_mm_node *nodes;
 	struct dma_fence **last_update;
 	struct dma_resv *resv;
@@ -1778,15 +1789,24 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		nodes = NULL;
 		resv = vm->root.base.bo->tbo.base.resv;
 	} else {
+		struct drm_gem_object *obj = &bo->tbo.base;
 		struct ttm_dma_tt *ttm;
 
+		resv = bo->tbo.base.resv;
+		if (obj->import_attach && bo_va->is_xgmi) {
+			struct dma_buf *dma_buf = obj->import_attach->dmabuf;
+			struct drm_gem_object *gobj = dma_buf->priv;
+			struct amdgpu_bo *abo = gem_to_amdgpu_bo(gobj);
+
+			if (abo->tbo.mem.mem_type == TTM_PL_VRAM)
+				bo = gem_to_amdgpu_bo(gobj);
+		}
 		mem = &bo->tbo.mem;
 		nodes = mem->mm_node;
 		if (mem->mem_type == TTM_PL_TT) {
 			ttm = container_of(bo->tbo.ttm, struct ttm_dma_tt, ttm);
 			pages_addr = ttm->dma_address;
 		}
-		resv = bo->tbo.base.resv;
 	}
 
 	if (bo) {
@@ -2132,8 +2152,10 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
-	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) &&
-	    (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)) {
+	if (!bo)
+		return bo_va;
+
+	if (amdgpu_dmabuf_is_xgmi_accessible(adev, bo)) {
 		bo_va->is_xgmi = true;
 		/* Power up XGMI if it can be potentially used */
 		amdgpu_xgmi_set_pstate(adev, AMDGPU_XGMI_PSTATE_MAX_VEGA20);
@@ -3209,7 +3231,7 @@ void amdgpu_vm_manager_fini(struct amdgpu_device *adev)
 int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
 	union drm_amdgpu_vm *args = data;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	long timeout = msecs_to_jiffies(2000);
 	int r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index ffbc0cc87ccf..c6abb16e8018 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -98,7 +98,7 @@ struct amdgpu_bo_list_entry;
 #define AMDGPU_PTE_MTYPE_NV10(a)       ((uint64_t)(a) << 48)
 #define AMDGPU_PTE_MTYPE_NV10_MASK     AMDGPU_PTE_MTYPE_NV10(7ULL)
 
-/* How to programm VM fault handling */
+/* How to program VM fault handling */
 #define AMDGPU_VM_FAULT_STOP_NEVER	0
 #define AMDGPU_VM_FAULT_STOP_FIRST	1
 #define AMDGPU_VM_FAULT_STOP_ALWAYS	2
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
index 39c704a1fb0e..0786e7555554 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -59,7 +59,7 @@ static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p,
  *
  * @p: see amdgpu_vm_update_params definition
  * @bo: PD/PT to update
- * @pe: kmap addr of the page entry
+ * @pe: byte offset of the PDE/PTE, relative to start of PDB/PTB
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
index 189d46ea603b..db790574dc2e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -155,7 +155,7 @@ static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p,
  *
  * @p: see amdgpu_vm_update_params definition
  * @bo: PD/PT to update
- * @pe: addr of the page entry
+ * @pe: byte offset of the PDE/PTE, relative to start of PDB/PTB
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
@@ -187,7 +187,7 @@ static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p,
  *
  * @p: see amdgpu_vm_update_params definition
  * @bo: PD/PT to update
- * @pe: addr of the page entry
+ * @pe: byte offset of the PDE/PTE, relative to start of PDB/PTB
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 0739e259bf91..01c1171afbe0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -28,12 +28,15 @@
 #include "amdgpu_atomfirmware.h"
 #include "atom.h"
 
-struct amdgpu_vram_mgr {
-	struct drm_mm mm;
-	spinlock_t lock;
-	atomic64_t usage;
-	atomic64_t vis_usage;
-};
+static inline struct amdgpu_vram_mgr *to_vram_mgr(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct amdgpu_vram_mgr, manager);
+}
+
+static inline struct amdgpu_device *to_amdgpu_device(struct amdgpu_vram_mgr *mgr)
+{
+	return container_of(mgr, struct amdgpu_device, mman.vram_mgr);
+}
 
 /**
  * DOC: mem_info_vram_total
@@ -47,7 +50,7 @@ static ssize_t amdgpu_mem_info_vram_total_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.real_vram_size);
 }
@@ -64,7 +67,7 @@ static ssize_t amdgpu_mem_info_vis_vram_total_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.visible_vram_size);
 }
@@ -81,10 +84,11 @@ static ssize_t amdgpu_mem_info_vram_used_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]));
+			amdgpu_vram_mgr_usage(man));
 }
 
 /**
@@ -99,10 +103,11 @@ static ssize_t amdgpu_mem_info_vis_vram_used_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]));
+			amdgpu_vram_mgr_vis_usage(man));
 }
 
 static ssize_t amdgpu_mem_info_vram_vendor(struct device *dev,
@@ -110,7 +115,7 @@ static ssize_t amdgpu_mem_info_vram_vendor(struct device *dev,
 						 char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	switch (adev->gmc.vram_vendor) {
 	case SAMSUNG:
@@ -158,6 +163,8 @@ static const struct attribute *amdgpu_vram_mgr_attributes[] = {
 	NULL
 };
 
+static const struct ttm_resource_manager_func amdgpu_vram_mgr_func;
+
 /**
  * amdgpu_vram_mgr_init - init VRAM manager and DRM MM
  *
@@ -166,26 +173,26 @@ static const struct attribute *amdgpu_vram_mgr_attributes[] = {
  *
  * Allocate and initialize the VRAM manager.
  */
-static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
-				unsigned long p_size)
+int amdgpu_vram_mgr_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_vram_mgr *mgr;
+	struct amdgpu_vram_mgr *mgr = &adev->mman.vram_mgr;
+	struct ttm_resource_manager *man = &mgr->manager;
 	int ret;
 
-	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
-	if (!mgr)
-		return -ENOMEM;
+	ttm_resource_manager_init(man, adev->gmc.real_vram_size >> PAGE_SHIFT);
+
+	man->func = &amdgpu_vram_mgr_func;
 
-	drm_mm_init(&mgr->mm, 0, p_size);
+	drm_mm_init(&mgr->mm, 0, man->size);
 	spin_lock_init(&mgr->lock);
-	man->priv = mgr;
 
 	/* Add the two VRAM-related sysfs files */
 	ret = sysfs_create_files(&adev->dev->kobj, amdgpu_vram_mgr_attributes);
 	if (ret)
 		DRM_ERROR("Failed to register sysfs\n");
 
+	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_VRAM, &mgr->manager);
+	ttm_resource_manager_set_used(man, true);
 	return 0;
 }
 
@@ -197,18 +204,26 @@ static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
  * Destroy and free the VRAM manager, returns -EBUSY if ranges are still
  * allocated inside it.
  */
-static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
+void amdgpu_vram_mgr_fini(struct amdgpu_device *adev)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = &adev->mman.vram_mgr;
+	struct ttm_resource_manager *man = &mgr->manager;
+	int ret;
+
+	ttm_resource_manager_set_used(man, false);
+
+	ret = ttm_resource_manager_force_list_clean(&adev->mman.bdev, man);
+	if (ret)
+		return;
 
 	spin_lock(&mgr->lock);
 	drm_mm_takedown(&mgr->mm);
 	spin_unlock(&mgr->lock);
-	kfree(mgr);
-	man->priv = NULL;
+
 	sysfs_remove_files(&adev->dev->kobj, amdgpu_vram_mgr_attributes);
-	return 0;
+
+	ttm_resource_manager_cleanup(man);
+	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_VRAM, NULL);
 }
 
 /**
@@ -243,7 +258,7 @@ static u64 amdgpu_vram_mgr_vis_size(struct amdgpu_device *adev,
 u64 amdgpu_vram_mgr_bo_visible_size(struct amdgpu_bo *bo)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
-	struct ttm_mem_reg *mem = &bo->tbo.mem;
+	struct ttm_resource *mem = &bo->tbo.mem;
 	struct drm_mm_node *nodes = mem->mm_node;
 	unsigned pages = mem->num_pages;
 	u64 usage;
@@ -263,13 +278,13 @@ u64 amdgpu_vram_mgr_bo_visible_size(struct amdgpu_bo *bo)
 /**
  * amdgpu_vram_mgr_virt_start - update virtual start address
  *
- * @mem: ttm_mem_reg to update
+ * @mem: ttm_resource to update
  * @node: just allocated node
  *
  * Calculate a virtual BO start address to easily check if everything is CPU
  * accessible.
  */
-static void amdgpu_vram_mgr_virt_start(struct ttm_mem_reg *mem,
+static void amdgpu_vram_mgr_virt_start(struct ttm_resource *mem,
 				       struct drm_mm_node *node)
 {
 	unsigned long start;
@@ -292,13 +307,13 @@ static void amdgpu_vram_mgr_virt_start(struct ttm_mem_reg *mem,
  *
  * Allocate VRAM for the given BO.
  */
-static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
+static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man,
 			       struct ttm_buffer_object *tbo,
 			       const struct ttm_place *place,
-			       struct ttm_mem_reg *mem)
+			       struct ttm_resource *mem)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
+	struct amdgpu_device *adev = to_amdgpu_device(mgr);
 	struct drm_mm *mm = &mgr->mm;
 	struct drm_mm_node *nodes;
 	enum drm_mm_insert_mode mode;
@@ -410,11 +425,11 @@ error:
  *
  * Free the allocated VRAM again.
  */
-static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
-				struct ttm_mem_reg *mem)
+static void amdgpu_vram_mgr_del(struct ttm_resource_manager *man,
+				struct ttm_resource *mem)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
+	struct amdgpu_device *adev = to_amdgpu_device(mgr);
 	struct drm_mm_node *nodes = mem->mm_node;
 	uint64_t usage = 0, vis_usage = 0;
 	unsigned pages = mem->num_pages;
@@ -451,7 +466,7 @@ static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
  * Allocate and fill a sg table from a VRAM allocation.
  */
 int amdgpu_vram_mgr_alloc_sgt(struct amdgpu_device *adev,
-			      struct ttm_mem_reg *mem,
+			      struct ttm_resource *mem,
 			      struct device *dev,
 			      enum dma_data_direction dir,
 			      struct sg_table **sgt)
@@ -544,9 +559,9 @@ void amdgpu_vram_mgr_free_sgt(struct amdgpu_device *adev,
  *
  * Returns how many bytes are used in this domain.
  */
-uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man)
+uint64_t amdgpu_vram_mgr_usage(struct ttm_resource_manager *man)
 {
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
 
 	return atomic64_read(&mgr->usage);
 }
@@ -558,9 +573,9 @@ uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man)
  *
  * 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)
+uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_resource_manager *man)
 {
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
 
 	return atomic64_read(&mgr->vis_usage);
 }
@@ -573,10 +588,10 @@ uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_mem_type_manager *man)
  *
  * Dump the table content using printk.
  */
-static void amdgpu_vram_mgr_debug(struct ttm_mem_type_manager *man,
+static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man,
 				  struct drm_printer *printer)
 {
-	struct amdgpu_vram_mgr *mgr = man->priv;
+	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
 
 	spin_lock(&mgr->lock);
 	drm_mm_print(&mgr->mm, printer);
@@ -587,10 +602,8 @@ static void amdgpu_vram_mgr_debug(struct ttm_mem_type_manager *man,
 		   amdgpu_vram_mgr_vis_usage(man) >> 20);
 }
 
-const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func = {
-	.init		= amdgpu_vram_mgr_init,
-	.takedown	= amdgpu_vram_mgr_fini,
-	.get_node	= amdgpu_vram_mgr_new,
-	.put_node	= amdgpu_vram_mgr_del,
-	.debug		= amdgpu_vram_mgr_debug
+static const struct ttm_resource_manager_func amdgpu_vram_mgr_func = {
+	.alloc	= amdgpu_vram_mgr_new,
+	.free	= amdgpu_vram_mgr_del,
+	.debug	= amdgpu_vram_mgr_debug
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index e3a3755cb999..1162913c8bf4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -35,11 +35,9 @@
 
 static DEFINE_MUTEX(xgmi_mutex);
 
-#define AMDGPU_MAX_XGMI_HIVE			8
 #define AMDGPU_MAX_XGMI_DEVICE_PER_HIVE		4
 
-static struct amdgpu_hive_info xgmi_hives[AMDGPU_MAX_XGMI_HIVE];
-static unsigned hive_count = 0;
+static LIST_HEAD(xgmi_hive_list);
 
 static const int xgmi_pcs_err_status_reg_vg20[] = {
 	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS,
@@ -171,65 +169,53 @@ static const struct amdgpu_pcs_ras_field wafl_pcs_ras_fields[] = {
  *
  */
 
+static struct attribute amdgpu_xgmi_hive_id = {
+	.name = "xgmi_hive_id",
+	.mode = S_IRUGO
+};
 
-static ssize_t amdgpu_xgmi_show_hive_id(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct amdgpu_hive_info *hive =
-			container_of(attr, struct amdgpu_hive_info, dev_attr);
-
-	return snprintf(buf, PAGE_SIZE, "%llu\n", hive->hive_id);
-}
+static struct attribute *amdgpu_xgmi_hive_attrs[] = {
+	&amdgpu_xgmi_hive_id,
+	NULL
+};
 
-static int amdgpu_xgmi_sysfs_create(struct amdgpu_device *adev,
-				    struct amdgpu_hive_info *hive)
+static ssize_t amdgpu_xgmi_show_attrs(struct kobject *kobj,
+	struct attribute *attr, char *buf)
 {
-	int ret = 0;
+	struct amdgpu_hive_info *hive = container_of(
+		kobj, struct amdgpu_hive_info, kobj);
 
-	if (WARN_ON(hive->kobj))
-		return -EINVAL;
-
-	hive->kobj = kobject_create_and_add("xgmi_hive_info", &adev->dev->kobj);
-	if (!hive->kobj) {
-		dev_err(adev->dev, "XGMI: Failed to allocate sysfs entry!\n");
-		return -EINVAL;
-	}
-
-	hive->dev_attr = (struct device_attribute) {
-		.attr = {
-			.name = "xgmi_hive_id",
-			.mode = S_IRUGO,
-
-		},
-		.show = amdgpu_xgmi_show_hive_id,
-	};
+	if (attr == &amdgpu_xgmi_hive_id)
+		return snprintf(buf, PAGE_SIZE, "%llu\n", hive->hive_id);
 
-	ret = sysfs_create_file(hive->kobj, &hive->dev_attr.attr);
-	if (ret) {
-		dev_err(adev->dev, "XGMI: Failed to create device file xgmi_hive_id\n");
-		kobject_del(hive->kobj);
-		kobject_put(hive->kobj);
-		hive->kobj = NULL;
-	}
-
-	return ret;
+	return 0;
 }
 
-static void amdgpu_xgmi_sysfs_destroy(struct amdgpu_device *adev,
-				    struct amdgpu_hive_info *hive)
+static void amdgpu_xgmi_hive_release(struct kobject *kobj)
 {
-	sysfs_remove_file(hive->kobj, &hive->dev_attr.attr);
-	kobject_del(hive->kobj);
-	kobject_put(hive->kobj);
-	hive->kobj = NULL;
+	struct amdgpu_hive_info *hive = container_of(
+		kobj, struct amdgpu_hive_info, kobj);
+
+	mutex_destroy(&hive->hive_lock);
+	kfree(hive);
 }
 
+static const struct sysfs_ops amdgpu_xgmi_hive_ops = {
+	.show = amdgpu_xgmi_show_attrs,
+};
+
+struct kobj_type amdgpu_xgmi_hive_type = {
+	.release = amdgpu_xgmi_hive_release,
+	.sysfs_ops = &amdgpu_xgmi_hive_ops,
+	.default_attrs = amdgpu_xgmi_hive_attrs,
+};
+
 static ssize_t amdgpu_xgmi_show_device_id(struct device *dev,
 				     struct device_attribute *attr,
 				     char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.xgmi.node_id);
 
@@ -241,7 +227,7 @@ static ssize_t amdgpu_xgmi_show_error(struct device *dev,
 				      char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t ficaa_pie_ctl_in, ficaa_pie_status_in;
 	uint64_t fica_out;
 	unsigned int error_count = 0;
@@ -287,8 +273,8 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 
 
 	/* Create sysfs link to hive info folder on the first device */
-	if (adev != hive->adev) {
-		ret = sysfs_create_link(&adev->dev->kobj, hive->kobj,
+	if (hive->kobj.parent != (&adev->dev->kobj)) {
+		ret = sysfs_create_link(&adev->dev->kobj, &hive->kobj,
 					"xgmi_hive_info");
 		if (ret) {
 			dev_err(adev->dev, "XGMI: Failed to create link to hive info");
@@ -296,9 +282,9 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 		}
 	}
 
-	sprintf(node, "node%d", hive->number_devices);
+	sprintf(node, "node%d", atomic_read(&hive->number_devices));
 	/* Create sysfs link form the hive folder to yourself */
-	ret = sysfs_create_link(hive->kobj, &adev->dev->kobj, node);
+	ret = sysfs_create_link(&hive->kobj, &adev->dev->kobj, node);
 	if (ret) {
 		dev_err(adev->dev, "XGMI: Failed to create link from hive info");
 		goto remove_link;
@@ -308,7 +294,7 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 
 
 remove_link:
-	sysfs_remove_link(&adev->dev->kobj, adev->ddev->unique);
+	sysfs_remove_link(&adev->dev->kobj, adev_to_drm(adev)->unique);
 
 remove_file:
 	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
@@ -326,78 +312,96 @@ static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev,
 	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
 	device_remove_file(adev->dev, &dev_attr_xgmi_error);
 
-	if (adev != hive->adev)
+	if (hive->kobj.parent != (&adev->dev->kobj))
 		sysfs_remove_link(&adev->dev->kobj,"xgmi_hive_info");
 
-	sprintf(node, "node%d", hive->number_devices);
-	sysfs_remove_link(hive->kobj, node);
+	sprintf(node, "node%d", atomic_read(&hive->number_devices));
+	sysfs_remove_link(&hive->kobj, node);
 
 }
 
 
 
-struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock)
+struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev)
 {
-	int i;
-	struct amdgpu_hive_info *tmp;
+	struct amdgpu_hive_info *hive = NULL, *tmp = NULL;
+	int ret;
 
 	if (!adev->gmc.xgmi.hive_id)
 		return NULL;
 
+	if (adev->hive) {
+		kobject_get(&adev->hive->kobj);
+		return adev->hive;
+	}
+
 	mutex_lock(&xgmi_mutex);
 
-	for (i = 0 ; i < hive_count; ++i) {
-		tmp = &xgmi_hives[i];
-		if (tmp->hive_id == adev->gmc.xgmi.hive_id) {
-			if (lock)
-				mutex_lock(&tmp->hive_lock);
-			mutex_unlock(&xgmi_mutex);
-			return tmp;
+	if (!list_empty(&xgmi_hive_list)) {
+		list_for_each_entry_safe(hive, tmp, &xgmi_hive_list, node)  {
+			if (hive->hive_id == adev->gmc.xgmi.hive_id)
+				goto pro_end;
 		}
 	}
-	if (i >= AMDGPU_MAX_XGMI_HIVE) {
-		mutex_unlock(&xgmi_mutex);
-		return NULL;
+
+	hive = kzalloc(sizeof(*hive), GFP_KERNEL);
+	if (!hive) {
+		dev_err(adev->dev, "XGMI: allocation failed\n");
+		hive = NULL;
+		goto pro_end;
 	}
 
 	/* initialize new hive if not exist */
-	tmp = &xgmi_hives[hive_count++];
-
-	if (amdgpu_xgmi_sysfs_create(adev, tmp)) {
-		mutex_unlock(&xgmi_mutex);
-		return NULL;
+	ret = kobject_init_and_add(&hive->kobj,
+			&amdgpu_xgmi_hive_type,
+			&adev->dev->kobj,
+			"%s", "xgmi_hive_info");
+	if (ret) {
+		dev_err(adev->dev, "XGMI: failed initializing kobject for xgmi hive\n");
+		kfree(hive);
+		hive = NULL;
+		goto pro_end;
 	}
 
-	tmp->adev = adev;
-	tmp->hive_id = adev->gmc.xgmi.hive_id;
-	INIT_LIST_HEAD(&tmp->device_list);
-	mutex_init(&tmp->hive_lock);
-	mutex_init(&tmp->reset_lock);
-	task_barrier_init(&tmp->tb);
-
-	if (lock)
-		mutex_lock(&tmp->hive_lock);
-	tmp->pstate = AMDGPU_XGMI_PSTATE_UNKNOWN;
-	tmp->hi_req_gpu = NULL;
+	hive->hive_id = adev->gmc.xgmi.hive_id;
+	INIT_LIST_HEAD(&hive->device_list);
+	INIT_LIST_HEAD(&hive->node);
+	mutex_init(&hive->hive_lock);
+	atomic_set(&hive->in_reset, 0);
+	atomic_set(&hive->number_devices, 0);
+	task_barrier_init(&hive->tb);
+	hive->pstate = AMDGPU_XGMI_PSTATE_UNKNOWN;
+	hive->hi_req_gpu = NULL;
 	/*
 	 * hive pstate on boot is high in vega20 so we have to go to low
 	 * pstate on after boot.
 	 */
-	tmp->hi_req_count = AMDGPU_MAX_XGMI_DEVICE_PER_HIVE;
+	hive->hi_req_count = AMDGPU_MAX_XGMI_DEVICE_PER_HIVE;
+	list_add_tail(&hive->node, &xgmi_hive_list);
+
+pro_end:
+	if (hive)
+		kobject_get(&hive->kobj);
 	mutex_unlock(&xgmi_mutex);
+	return hive;
+}
 
-	return tmp;
+void amdgpu_put_xgmi_hive(struct amdgpu_hive_info *hive)
+{
+	if (hive)
+		kobject_put(&hive->kobj);
 }
 
 int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate)
 {
 	int ret = 0;
-	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, 0);
+	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
 	struct amdgpu_device *request_adev = hive->hi_req_gpu ?
 						hive->hi_req_gpu : adev;
 	bool is_hi_req = pstate == AMDGPU_XGMI_PSTATE_MAX_VEGA20;
 	bool init_low = hive->pstate == AMDGPU_XGMI_PSTATE_UNKNOWN;
 
+	amdgpu_put_xgmi_hive(hive);
 	/* fw bug so temporarily disable pstate switching */
 	return 0;
 
@@ -449,7 +453,7 @@ int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_dev
 
 	/* Each psp need to set the latest topology */
 	ret = psp_xgmi_set_topology_info(&adev->psp,
-					 hive->number_devices,
+					 atomic_read(&hive->number_devices),
 					 &adev->psp.xgmi_context.top_info);
 	if (ret)
 		dev_err(adev->dev,
@@ -511,7 +515,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 		adev->gmc.xgmi.node_id = adev->gmc.xgmi.physical_node_id + 16;
 	}
 
-	hive = amdgpu_get_xgmi_hive(adev, 1);
+	hive = amdgpu_get_xgmi_hive(adev);
 	if (!hive) {
 		ret = -EINVAL;
 		dev_err(adev->dev,
@@ -519,6 +523,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 			adev->gmc.xgmi.node_id, adev->gmc.xgmi.hive_id);
 		goto exit;
 	}
+	mutex_lock(&hive->hive_lock);
 
 	top_info = &adev->psp.xgmi_context.top_info;
 
@@ -526,7 +531,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 	list_for_each_entry(entry, &hive->device_list, head)
 		top_info->nodes[count++].node_id = entry->node_id;
 	top_info->num_nodes = count;
-	hive->number_devices = count;
+	atomic_set(&hive->number_devices, count);
 
 	task_barrier_add_task(&hive->tb);
 
@@ -541,7 +546,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 			}
 			ret = amdgpu_xgmi_update_topology(hive, tmp_adev);
 			if (ret)
-				goto exit;
+				goto exit_unlock;
 		}
 
 		/* get latest topology info for each device from psp */
@@ -554,7 +559,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 					tmp_adev->gmc.xgmi.node_id,
 					tmp_adev->gmc.xgmi.hive_id, ret);
 				/* To do : continue with some node failed or disable the whole hive */
-				goto exit;
+				goto exit_unlock;
 			}
 		}
 	}
@@ -562,39 +567,51 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 	if (!ret)
 		ret = amdgpu_xgmi_sysfs_add_dev_info(adev, hive);
 
-
+exit_unlock:
 	mutex_unlock(&hive->hive_lock);
 exit:
-	if (!ret)
+	if (!ret) {
+		adev->hive = hive;
 		dev_info(adev->dev, "XGMI: Add node %d, hive 0x%llx.\n",
 			 adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id);
-	else
+	} else {
+		amdgpu_put_xgmi_hive(hive);
 		dev_err(adev->dev, "XGMI: Failed to add node %d, hive 0x%llx ret: %d\n",
 			adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id,
 			ret);
+	}
 
 	return ret;
 }
 
 int amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 {
-	struct amdgpu_hive_info *hive;
+	struct amdgpu_hive_info *hive = adev->hive;
 
 	if (!adev->gmc.xgmi.supported)
 		return -EINVAL;
 
-	hive = amdgpu_get_xgmi_hive(adev, 1);
 	if (!hive)
 		return -EINVAL;
 
+	mutex_lock(&hive->hive_lock);
 	task_barrier_rem_task(&hive->tb);
 	amdgpu_xgmi_sysfs_rem_dev_info(adev, hive);
+	if (hive->hi_req_gpu == adev)
+		hive->hi_req_gpu = NULL;
+	list_del(&adev->gmc.xgmi.head);
 	mutex_unlock(&hive->hive_lock);
 
-	if(!(--hive->number_devices)){
-		amdgpu_xgmi_sysfs_destroy(adev, hive);
-		mutex_destroy(&hive->hive_lock);
-		mutex_destroy(&hive->reset_lock);
+	amdgpu_put_xgmi_hive(hive);
+	adev->hive = NULL;
+
+	if (atomic_dec_return(&hive->number_devices) == 0) {
+		/* Remove the hive from global hive list */
+		mutex_lock(&xgmi_mutex);
+		list_del(&hive->node);
+		mutex_unlock(&xgmi_mutex);
+
+		amdgpu_put_xgmi_hive(hive);
 	}
 
 	return psp_xgmi_terminate(&adev->psp);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index 6999eab16a72..148560d63554 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -27,13 +27,13 @@
 
 
 struct amdgpu_hive_info {
-	uint64_t		hive_id;
-	struct list_head	device_list;
-	int number_devices;
-	struct mutex hive_lock, reset_lock;
-	struct kobject *kobj;
-	struct device_attribute dev_attr;
-	struct amdgpu_device *adev;
+	struct kobject kobj;
+	uint64_t hive_id;
+	struct list_head device_list;
+	struct list_head node;
+	atomic_t number_devices;
+	struct mutex hive_lock;
+	atomic_t in_reset;
 	int hi_req_count;
 	struct amdgpu_device *hi_req_gpu;
 	struct task_barrier tb;
@@ -50,7 +50,8 @@ struct amdgpu_pcs_ras_field {
 	uint32_t pcs_err_shift;
 };
 
-struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock);
+struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev);
+void amdgpu_put_xgmi_hive(struct amdgpu_hive_info *hive);
 int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev);
 int amdgpu_xgmi_add_device(struct amdgpu_device *adev);
 int amdgpu_xgmi_remove_device(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
new file mode 100644
index 000000000000..5355827ed0ae
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
@@ -0,0 +1,276 @@
+/*
+ * Copyright 2018-2019 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"),
+ * 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.
+ *
+ */
+
+#ifndef AMDGV_SRIOV_MSG__H_
+#define AMDGV_SRIOV_MSG__H_
+
+/* unit in kilobytes */
+#define AMD_SRIOV_MSG_VBIOS_OFFSET              0
+#define AMD_SRIOV_MSG_VBIOS_SIZE_KB             64
+#define AMD_SRIOV_MSG_DATAEXCHANGE_OFFSET_KB    AMD_SRIOV_MSG_VBIOS_SIZE_KB
+#define AMD_SRIOV_MSG_DATAEXCHANGE_SIZE_KB      4
+
+/*
+ * layout
+ * 0           64KB        65KB        66KB
+ * |   VBIOS   |   PF2VF   |   VF2PF   |   Bad Page   | ...
+ * |   64KB    |   1KB     |   1KB     |
+ */
+#define AMD_SRIOV_MSG_SIZE_KB                   1
+#define AMD_SRIOV_MSG_PF2VF_OFFSET_KB           AMD_SRIOV_MSG_DATAEXCHANGE_OFFSET_KB
+#define AMD_SRIOV_MSG_VF2PF_OFFSET_KB           (AMD_SRIOV_MSG_PF2VF_OFFSET_KB + AMD_SRIOV_MSG_SIZE_KB)
+#define AMD_SRIOV_MSG_BAD_PAGE_OFFSET_KB        (AMD_SRIOV_MSG_VF2PF_OFFSET_KB + AMD_SRIOV_MSG_SIZE_KB)
+
+/*
+ * PF2VF history log:
+ * v1 defined in amdgim
+ * v2 current
+ *
+ * VF2PF history log:
+ * v1 defined in amdgim
+ * v2 defined in amdgim
+ * v3 current
+ */
+#define AMD_SRIOV_MSG_FW_VRAM_PF2VF_VER			2
+#define AMD_SRIOV_MSG_FW_VRAM_VF2PF_VER			3
+
+#define AMD_SRIOV_MSG_RESERVE_UCODE		24
+
+enum amd_sriov_ucode_engine_id {
+	AMD_SRIOV_UCODE_ID_VCE = 0,
+	AMD_SRIOV_UCODE_ID_UVD,
+	AMD_SRIOV_UCODE_ID_MC,
+	AMD_SRIOV_UCODE_ID_ME,
+	AMD_SRIOV_UCODE_ID_PFP,
+	AMD_SRIOV_UCODE_ID_CE,
+	AMD_SRIOV_UCODE_ID_RLC,
+	AMD_SRIOV_UCODE_ID_RLC_SRLC,
+	AMD_SRIOV_UCODE_ID_RLC_SRLG,
+	AMD_SRIOV_UCODE_ID_RLC_SRLS,
+	AMD_SRIOV_UCODE_ID_MEC,
+	AMD_SRIOV_UCODE_ID_MEC2,
+	AMD_SRIOV_UCODE_ID_SOS,
+	AMD_SRIOV_UCODE_ID_ASD,
+	AMD_SRIOV_UCODE_ID_TA_RAS,
+	AMD_SRIOV_UCODE_ID_TA_XGMI,
+	AMD_SRIOV_UCODE_ID_SMC,
+	AMD_SRIOV_UCODE_ID_SDMA,
+	AMD_SRIOV_UCODE_ID_SDMA2,
+	AMD_SRIOV_UCODE_ID_VCN,
+	AMD_SRIOV_UCODE_ID_DMCU,
+	AMD_SRIOV_UCODE_ID__MAX
+};
+
+#pragma pack(push, 1) 	// PF2VF / VF2PF data areas are byte packed
+
+union amd_sriov_msg_feature_flags {
+	struct {
+		uint32_t  error_log_collect  : 1;
+		uint32_t  host_load_ucodes   : 1;
+		uint32_t  host_flr_vramlost  : 1;
+		uint32_t  mm_bw_management   : 1;
+		uint32_t  pp_one_vf_mode     : 1;
+		uint32_t  reserved           : 27;
+	} flags;
+	uint32_t      all;
+};
+
+union amd_sriov_msg_os_info {
+	struct {
+		uint32_t  windows            : 1;
+		uint32_t  reserved           : 31;
+	} info;
+	uint32_t      all;
+};
+
+struct amd_sriov_msg_pf2vf_info_header {
+	/* the total structure size in byte */
+	uint32_t size;
+	/* version of this structure, written by the HOST */
+	uint32_t version;
+	/* reserved */
+	uint32_t reserved[2];
+};
+
+struct amd_sriov_msg_pf2vf_info {
+	/* header contains size and version */
+	struct amd_sriov_msg_pf2vf_info_header header;
+	/* use private key from mailbox 2 to create checksum */
+	uint32_t checksum;
+	/* The features flags of the HOST driver supports */
+	union amd_sriov_msg_feature_flags feature_flags;
+	/* (max_width * max_height * fps) / (16 * 16) */
+	uint32_t hevc_enc_max_mb_per_second;
+	/* (max_width * max_height) / (16 * 16) */
+	uint32_t hevc_enc_max_mb_per_frame;
+	/* (max_width * max_height * fps) / (16 * 16) */
+	uint32_t avc_enc_max_mb_per_second;
+	/* (max_width * max_height) / (16 * 16) */
+	uint32_t avc_enc_max_mb_per_frame;
+	/* MEC FW position in BYTE from the start of VF visible frame buffer */
+	uint64_t mecfw_offset;
+	/* MEC FW size in BYTE */
+	uint32_t mecfw_size;
+	/* UVD FW position in BYTE from the start of VF visible frame buffer */
+	uint64_t uvdfw_offset;
+	/* UVD FW size in BYTE */
+	uint32_t uvdfw_size;
+	/* VCE FW position in BYTE from the start of VF visible frame buffer */
+	uint64_t vcefw_offset;
+	/* VCE FW size in BYTE */
+	uint32_t vcefw_size;
+	/* Bad pages block position in BYTE */
+	uint32_t bp_block_offset_low;
+	uint32_t bp_block_offset_high;
+	/* Bad pages block size in BYTE */
+	uint32_t bp_block_size;
+	/* frequency for VF to update the VF2PF area in msec, 0 = manual */
+	uint32_t vf2pf_update_interval_ms;
+	/* identification in ROCm SMI */
+	uint64_t uuid;
+	uint32_t fcn_idx;
+	/* reserved */
+	uint32_t reserved[256-26];
+};
+
+struct amd_sriov_msg_vf2pf_info_header {
+	/* the total structure size in byte */
+	uint32_t size;
+	/* version of this structure, written by the guest */
+	uint32_t version;
+	/* reserved */
+	uint32_t reserved[2];
+};
+
+struct amd_sriov_msg_vf2pf_info {
+	/* header contains size and version */
+	struct amd_sriov_msg_vf2pf_info_header header;
+	uint32_t checksum;
+	/* driver version */
+	uint8_t  driver_version[64];
+	/* driver certification, 1=WHQL, 0=None */
+	uint32_t driver_cert;
+	/* guest OS type and version */
+	union amd_sriov_msg_os_info os_info;
+	/* guest fb information in the unit of MB */
+	uint32_t fb_usage;
+	/* guest gfx engine usage percentage */
+	uint32_t gfx_usage;
+	/* guest gfx engine health percentage */
+	uint32_t gfx_health;
+	/* guest compute engine usage percentage */
+	uint32_t compute_usage;
+	/* guest compute engine health percentage */
+	uint32_t compute_health;
+	/* guest avc engine usage percentage. 0xffff means N/A */
+	uint32_t avc_enc_usage;
+	/* guest avc engine health percentage. 0xffff means N/A */
+	uint32_t avc_enc_health;
+	/* guest hevc engine usage percentage. 0xffff means N/A */
+	uint32_t hevc_enc_usage;
+	/* guest hevc engine usage percentage. 0xffff means N/A */
+	uint32_t hevc_enc_health;
+	/* combined encode/decode usage */
+	uint32_t encode_usage;
+	uint32_t decode_usage;
+	/* Version of PF2VF that VF understands */
+	uint32_t pf2vf_version_required;
+	/* additional FB usage */
+	uint32_t fb_vis_usage;
+	uint32_t fb_vis_size;
+	uint32_t fb_size;
+	/* guest ucode data, each one is 1.25 Dword */
+	struct {
+		uint8_t  id;
+		uint32_t version;
+	} ucode_info[AMD_SRIOV_MSG_RESERVE_UCODE];
+
+	/* reserved */
+	uint32_t reserved[256-68];
+};
+
+/* mailbox message send from guest to host  */
+enum amd_sriov_mailbox_request_message {
+	MB_REQ_MSG_REQ_GPU_INIT_ACCESS = 1,
+	MB_REQ_MSG_REL_GPU_INIT_ACCESS,
+	MB_REQ_MSG_REQ_GPU_FINI_ACCESS,
+	MB_REQ_MSG_REL_GPU_FINI_ACCESS,
+	MB_REQ_MSG_REQ_GPU_RESET_ACCESS,
+	MB_REQ_MSG_REQ_GPU_INIT_DATA,
+
+	MB_REQ_MSG_LOG_VF_ERROR       = 200,
+};
+
+/* mailbox message send from host to guest  */
+enum amd_sriov_mailbox_response_message {
+	MB_RES_MSG_CLR_MSG_BUF = 0,
+	MB_RES_MSG_READY_TO_ACCESS_GPU = 1,
+	MB_RES_MSG_FLR_NOTIFICATION,
+	MB_RES_MSG_FLR_NOTIFICATION_COMPLETION,
+	MB_RES_MSG_SUCCESS,
+	MB_RES_MSG_FAIL,
+	MB_RES_MSG_QUERY_ALIVE,
+	MB_RES_MSG_GPU_INIT_DATA_READY,
+
+	MB_RES_MSG_TEXT_MESSAGE = 255
+};
+
+/* version data stored in MAILBOX_MSGBUF_RCV_DW1 for future expansion */
+enum amd_sriov_gpu_init_data_version {
+	GPU_INIT_DATA_READY_V1 = 1,
+};
+
+#pragma pack(pop)	// Restore previous packing option
+
+/* checksum function between host and guest */
+unsigned int amd_sriov_msg_checksum(void *obj,
+				unsigned long obj_size,
+				unsigned int key,
+				unsigned int checksum);
+
+/* assertion at compile time */
+#ifdef __linux__
+#define stringification(s) _stringification(s)
+#define _stringification(s) #s
+
+_Static_assert(
+	sizeof(struct amd_sriov_msg_vf2pf_info) == AMD_SRIOV_MSG_SIZE_KB << 10,
+	"amd_sriov_msg_vf2pf_info must be " stringification(AMD_SRIOV_MSG_SIZE_KB) " KB");
+
+_Static_assert(
+	sizeof(struct amd_sriov_msg_pf2vf_info) == AMD_SRIOV_MSG_SIZE_KB << 10,
+	"amd_sriov_msg_pf2vf_info must be " stringification(AMD_SRIOV_MSG_SIZE_KB) " KB");
+
+_Static_assert(
+	AMD_SRIOV_MSG_RESERVE_UCODE % 4 == 0,
+	"AMD_SRIOV_MSG_RESERVE_UCODE must be multiple of 4");
+
+_Static_assert(
+	AMD_SRIOV_MSG_RESERVE_UCODE > AMD_SRIOV_UCODE_ID__MAX,
+	"AMD_SRIOV_MSG_RESERVE_UCODE must be bigger than AMD_SRIOV_UCODE_ID__MAX");
+
+#undef _stringification
+#undef stringification
+#endif
+
+#endif /* AMDGV_SRIOV_MSG__H_ */
diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c
index 847ca9b3ce4e..3ea557864320 100644
--- a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c
@@ -73,6 +73,7 @@ int athub_v1_0_set_clockgating(struct amdgpu_device *adev,
 	case CHIP_VEGA12:
 	case CHIP_VEGA20:
 	case CHIP_RAVEN:
+	case CHIP_RENOIR:
 		athub_update_medium_grain_clock_gating(adev,
 				state == AMD_CG_STATE_GATE);
 		athub_update_medium_grain_light_sleep(adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c b/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c
index 213e62a28ba0..159a2a4385a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c
@@ -41,7 +41,7 @@ void amdgpu_atombios_crtc_overscan_setup(struct drm_crtc *crtc,
 				  struct drm_display_mode *adjusted_mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	SET_CRTC_OVERSCAN_PS_ALLOCATION args;
 	int index = GetIndexIntoMasterTable(COMMAND, SetCRTC_OverScan);
@@ -84,7 +84,7 @@ void amdgpu_atombios_crtc_overscan_setup(struct drm_crtc *crtc,
 void amdgpu_atombios_crtc_scaler_setup(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	ENABLE_SCALER_PS_ALLOCATION args;
 	int index = GetIndexIntoMasterTable(COMMAND, EnableScaler);
@@ -114,7 +114,7 @@ void amdgpu_atombios_crtc_lock(struct drm_crtc *crtc, int lock)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int index =
 	    GetIndexIntoMasterTable(COMMAND, UpdateCRTC_DoubleBufferRegisters);
 	ENABLE_CRTC_PS_ALLOCATION args;
@@ -131,7 +131,7 @@ void amdgpu_atombios_crtc_enable(struct drm_crtc *crtc, int state)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int index = GetIndexIntoMasterTable(COMMAND, EnableCRTC);
 	ENABLE_CRTC_PS_ALLOCATION args;
 
@@ -147,7 +147,7 @@ void amdgpu_atombios_crtc_blank(struct drm_crtc *crtc, int state)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int index = GetIndexIntoMasterTable(COMMAND, BlankCRTC);
 	BLANK_CRTC_PS_ALLOCATION args;
 
@@ -163,7 +163,7 @@ void amdgpu_atombios_crtc_powergate(struct drm_crtc *crtc, int state)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	int index = GetIndexIntoMasterTable(COMMAND, EnableDispPowerGating);
 	ENABLE_DISP_POWER_GATING_PS_ALLOCATION args;
 
@@ -192,7 +192,7 @@ void amdgpu_atombios_crtc_set_dtd_timing(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	SET_CRTC_USING_DTD_TIMING_PARAMETERS args;
 	int index = GetIndexIntoMasterTable(COMMAND, SetCRTC_UsingDTDTiming);
 	u16 misc = 0;
@@ -307,7 +307,7 @@ static u32 amdgpu_atombios_crtc_adjust_pll(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_encoder *encoder = amdgpu_crtc->encoder;
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -588,7 +588,7 @@ void amdgpu_atombios_crtc_program_pll(struct drm_crtc *crtc,
 				      struct amdgpu_atom_ss *ss)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u8 frev, crev;
 	int index = GetIndexIntoMasterTable(COMMAND, SetPixelClock);
 	union set_pixel_clock args;
@@ -749,7 +749,7 @@ int amdgpu_atombios_crtc_prepare_pll(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder =
 		to_amdgpu_encoder(amdgpu_crtc->encoder);
 	int encoder_mode = amdgpu_atombios_encoder_get_encoder_mode(amdgpu_crtc->encoder);
@@ -818,7 +818,7 @@ void amdgpu_atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder =
 		to_amdgpu_encoder(amdgpu_crtc->encoder);
 	u32 pll_clock = mode->clock;
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
index 9b74cfdba7b8..a3ba9ca11e98 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
@@ -60,7 +60,7 @@ static int amdgpu_atombios_dp_process_aux_ch(struct amdgpu_i2c_chan *chan,
 				      u8 delay, u8 *ack)
 {
 	struct drm_device *dev = chan->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	union aux_channel_transaction args;
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessAuxChannelTransaction);
 	unsigned char *base;
@@ -305,7 +305,7 @@ static u8 amdgpu_atombios_dp_encoder_service(struct amdgpu_device *adev,
 u8 amdgpu_atombios_dp_get_sinktype(struct amdgpu_connector *amdgpu_connector)
 {
 	struct drm_device *dev = amdgpu_connector->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	return amdgpu_atombios_dp_encoder_service(adev, ATOM_DP_ACTION_GET_SINK_TYPE, 0,
 					   amdgpu_connector->ddc_bus->rec.i2c_id, 0);
@@ -328,6 +328,22 @@ static void amdgpu_atombios_dp_probe_oui(struct amdgpu_connector *amdgpu_connect
 			      buf[0], buf[1], buf[2]);
 }
 
+static void amdgpu_atombios_dp_ds_ports(struct amdgpu_connector *amdgpu_connector)
+{
+	struct amdgpu_connector_atom_dig *dig_connector = amdgpu_connector->con_priv;
+	int ret;
+
+	if (dig_connector->dpcd[DP_DPCD_REV] > 0x10) {
+		ret = drm_dp_dpcd_read(&amdgpu_connector->ddc_bus->aux,
+				       DP_DOWNSTREAM_PORT_0,
+				       dig_connector->downstream_ports,
+				       DP_MAX_DOWNSTREAM_PORTS);
+		if (ret)
+			memset(dig_connector->downstream_ports, 0,
+			       DP_MAX_DOWNSTREAM_PORTS);
+	}
+}
+
 int amdgpu_atombios_dp_get_dpcd(struct amdgpu_connector *amdgpu_connector)
 {
 	struct amdgpu_connector_atom_dig *dig_connector = amdgpu_connector->con_priv;
@@ -343,7 +359,7 @@ int amdgpu_atombios_dp_get_dpcd(struct amdgpu_connector *amdgpu_connector)
 			      dig_connector->dpcd);
 
 		amdgpu_atombios_dp_probe_oui(amdgpu_connector);
-
+		amdgpu_atombios_dp_ds_ports(amdgpu_connector);
 		return 0;
 	}
 
@@ -702,7 +718,7 @@ void amdgpu_atombios_dp_link_train(struct drm_encoder *encoder,
 			    struct drm_connector *connector)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_connector *amdgpu_connector;
 	struct amdgpu_connector_atom_dig *dig_connector;
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c
index 1e94a9b652f7..8339c8c3a328 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c
@@ -70,7 +70,7 @@ u8
 amdgpu_atombios_encoder_get_backlight_level(struct amdgpu_encoder *amdgpu_encoder)
 {
 	struct drm_device *dev = amdgpu_encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (!(adev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU))
 		return 0;
@@ -84,7 +84,7 @@ amdgpu_atombios_encoder_set_backlight_level(struct amdgpu_encoder *amdgpu_encode
 {
 	struct drm_encoder *encoder = &amdgpu_encoder->base;
 	struct drm_device *dev = amdgpu_encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder_atom_dig *dig;
 
 	if (!(adev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU))
@@ -152,7 +152,7 @@ amdgpu_atombios_encoder_get_backlight_brightness(struct backlight_device *bd)
 	struct amdgpu_backlight_privdata *pdata = bl_get_data(bd);
 	struct amdgpu_encoder *amdgpu_encoder = pdata->encoder;
 	struct drm_device *dev = amdgpu_encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	return amdgpu_atombios_encoder_get_backlight_level_from_reg(adev);
 }
@@ -166,7 +166,7 @@ void amdgpu_atombios_encoder_init_backlight(struct amdgpu_encoder *amdgpu_encode
 				     struct drm_connector *drm_connector)
 {
 	struct drm_device *dev = amdgpu_encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct backlight_device *bd;
 	struct backlight_properties props;
 	struct amdgpu_backlight_privdata *pdata;
@@ -229,7 +229,7 @@ void
 amdgpu_atombios_encoder_fini_backlight(struct amdgpu_encoder *amdgpu_encoder)
 {
 	struct drm_device *dev = amdgpu_encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct backlight_device *bd = NULL;
 	struct amdgpu_encoder_atom_dig *dig;
 
@@ -319,7 +319,7 @@ static void
 amdgpu_atombios_encoder_setup_dac(struct drm_encoder *encoder, int action)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	DAC_ENCODER_CONTROL_PS_ALLOCATION args;
 	int index = 0;
@@ -382,7 +382,7 @@ static void
 amdgpu_atombios_encoder_setup_dvo(struct drm_encoder *encoder, int action)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	union dvo_encoder_control args;
 	int index = GetIndexIntoMasterTable(COMMAND, DVOEncoderControl);
@@ -573,7 +573,7 @@ amdgpu_atombios_encoder_setup_dig_encoder(struct drm_encoder *encoder,
 				   int action, int panel_mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -762,7 +762,7 @@ amdgpu_atombios_encoder_setup_dig_transmitter(struct drm_encoder *encoder, int a
 					      uint8_t lane_num, uint8_t lane_set)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1178,7 +1178,7 @@ amdgpu_atombios_encoder_set_edp_panel_power(struct drm_connector *connector,
 {
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	struct drm_device *dev = amdgpu_connector->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	union dig_transmitter_control args;
 	int index = GetIndexIntoMasterTable(COMMAND, UNIPHYTransmitterControl);
 	uint8_t frev, crev;
@@ -1225,7 +1225,7 @@ amdgpu_atombios_encoder_setup_external_encoder(struct drm_encoder *encoder,
 					int action)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder *ext_amdgpu_encoder = to_amdgpu_encoder(ext_encoder);
 	union external_encoder_control args;
@@ -1466,7 +1466,7 @@ void
 amdgpu_atombios_encoder_set_crtc_source(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
 	union crtc_source_param args;
@@ -1673,7 +1673,7 @@ amdgpu_atombios_encoder_set_crtc_source(struct drm_encoder *encoder)
 void
 amdgpu_atombios_encoder_init_dig(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_encoder *encoder;
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
@@ -1701,7 +1701,7 @@ amdgpu_atombios_encoder_dac_load_detect(struct drm_encoder *encoder,
 				 struct drm_connector *connector)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 
@@ -1751,7 +1751,7 @@ amdgpu_atombios_encoder_dac_detect(struct drm_encoder *encoder,
 			    struct drm_connector *connector)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	uint32_t bios_0_scratch;
@@ -1790,7 +1790,7 @@ amdgpu_atombios_encoder_dig_detect(struct drm_encoder *encoder,
 			    struct drm_connector *connector)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
 	struct drm_encoder *ext_encoder = amdgpu_get_external_encoder(encoder);
@@ -1848,7 +1848,7 @@ amdgpu_atombios_encoder_set_bios_scratch_regs(struct drm_connector *connector,
 				       bool connected)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector =
 	    to_amdgpu_connector(connector);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
@@ -1999,7 +1999,7 @@ struct amdgpu_encoder_atom_dig *
 amdgpu_atombios_encoder_get_lcd_info(struct amdgpu_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_mode_info *mode_info = &adev->mode_info;
 	int index = GetIndexIntoMasterTable(DATA, LVDS_Info);
 	uint16_t data_offset, misc;
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c b/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c
index b4cc7c55fa16..09a538465ffd 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c
@@ -40,7 +40,7 @@ static int amdgpu_atombios_i2c_process_i2c_ch(struct amdgpu_i2c_chan *chan,
 				       u8 *buf, u8 num)
 {
 	struct drm_device *dev = chan->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	PROCESS_I2C_CHANNEL_TRANSACTION_PS_ALLOCATION args;
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction);
 	unsigned char *base;
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index c2c67ab68a43..03ff8bd1fee8 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -1366,8 +1366,10 @@ static int cik_asic_reset(struct amdgpu_device *adev)
 	int r;
 
 	if (cik_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
+		dev_info(adev->dev, "BACO reset\n");
 		r = amdgpu_dpm_baco_reset(adev);
 	} else {
+		dev_info(adev->dev, "PCI CONFIG reset\n");
 		r = cik_asic_pci_config_reset(adev);
 	}
 
@@ -1919,6 +1921,10 @@ static uint64_t cik_get_pcie_replay_count(struct amdgpu_device *adev)
 	return (nak_r + nak_g);
 }
 
+static void cik_pre_asic_init(struct amdgpu_device *adev)
+{
+}
+
 static const struct amdgpu_asic_funcs cik_asic_funcs =
 {
 	.read_disabled_bios = &cik_read_disabled_bios,
@@ -1939,6 +1945,7 @@ static const struct amdgpu_asic_funcs cik_asic_funcs =
 	.need_reset_on_init = &cik_need_reset_on_init,
 	.get_pcie_replay_count = &cik_get_pcie_replay_count,
 	.supports_baco = &cik_asic_supports_baco,
+	.pre_asic_init = &cik_pre_asic_init,
 };
 
 static int cik_common_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c
index 401c99f0b2d0..db953e95f3d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c
@@ -316,14 +316,9 @@ static int cik_ih_sw_fini(void *handle)
 
 static int cik_ih_hw_init(void *handle)
 {
-	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	r = cik_ih_irq_init(adev);
-	if (r)
-		return r;
-
-	return 0;
+	return cik_ih_irq_init(adev);
 }
 
 static int cik_ih_hw_fini(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 84b45a019a36..5963cbe0d455 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -328,7 +328,7 @@ static void dce_v10_0_hpd_set_polarity(struct amdgpu_device *adev,
  */
 static void dce_v10_0_hpd_init(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -383,7 +383,7 @@ static void dce_v10_0_hpd_init(struct amdgpu_device *adev)
  */
 static void dce_v10_0_hpd_fini(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -504,7 +504,7 @@ void dce_v10_0_disable_dce(struct amdgpu_device *adev)
 static void dce_v10_0_program_fmt(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1209,7 +1209,7 @@ static struct amdgpu_audio_pin *dce_v10_0_audio_get_pin(struct amdgpu_device *ad
 
 static void dce_v10_0_afmt_audio_select_pin(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1226,7 +1226,7 @@ static void dce_v10_0_audio_write_latency_fields(struct drm_encoder *encoder,
 						struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1272,7 +1272,7 @@ static void dce_v10_0_audio_write_latency_fields(struct drm_encoder *encoder,
 static void dce_v10_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1328,7 +1328,7 @@ static void dce_v10_0_audio_write_speaker_allocation(struct drm_encoder *encoder
 static void dce_v10_0_audio_write_sad_regs(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1483,7 +1483,7 @@ static void dce_v10_0_audio_fini(struct amdgpu_device *adev)
 static void dce_v10_0_afmt_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_afmt_acr acr = amdgpu_afmt_acr(clock);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
@@ -1519,7 +1519,7 @@ static void dce_v10_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 					       void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	uint8_t *frame = buffer + 3;
@@ -1538,7 +1538,7 @@ static void dce_v10_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 static void dce_v10_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
@@ -1569,7 +1569,7 @@ static void dce_v10_0_afmt_setmode(struct drm_encoder *encoder,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1749,7 +1749,7 @@ static void dce_v10_0_afmt_setmode(struct drm_encoder *encoder,
 static void dce_v10_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 
@@ -1822,7 +1822,7 @@ static void dce_v10_0_vga_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 vga_control;
 
 	vga_control = RREG32(vga_control_regs[amdgpu_crtc->crtc_id]) & ~1;
@@ -1836,7 +1836,7 @@ static void dce_v10_0_grph_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (enable)
 		WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, 1);
@@ -1850,7 +1850,7 @@ static int dce_v10_0_crtc_do_set_base(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_framebuffer *target_fb;
 	struct drm_gem_object *obj;
 	struct amdgpu_bo *abo;
@@ -2095,7 +2095,7 @@ static void dce_v10_0_set_interleave(struct drm_crtc *crtc,
 				     struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	u32 tmp;
 
@@ -2111,7 +2111,7 @@ static void dce_v10_0_crtc_load_lut(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u16 *r, *g, *b;
 	int i;
 	u32 tmp;
@@ -2250,7 +2250,7 @@ static u32 dce_v10_0_pick_pll(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 pll_in_use;
 	int pll;
 
@@ -2285,7 +2285,7 @@ static u32 dce_v10_0_pick_pll(struct drm_crtc *crtc)
 
 static void dce_v10_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	uint32_t cur_lock;
 
@@ -2300,7 +2300,7 @@ static void dce_v10_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 static void dce_v10_0_hide_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	u32 tmp;
 
 	tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
@@ -2311,7 +2311,7 @@ static void dce_v10_0_hide_cursor(struct drm_crtc *crtc)
 static void dce_v10_0_show_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	u32 tmp;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
@@ -2329,7 +2329,7 @@ static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
 					int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	int xorigin = 0, yorigin = 0;
 
 	amdgpu_crtc->cursor_x = x;
@@ -2503,7 +2503,7 @@ static const struct drm_crtc_funcs dce_v10_0_crtc_funcs = {
 static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	unsigned type;
 
@@ -2557,7 +2557,7 @@ static void dce_v10_0_crtc_disable(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_atom_ss ss;
 	int i;
 
@@ -2701,7 +2701,7 @@ static int dce_v10_0_crtc_init(struct amdgpu_device *adev, int index)
 	if (amdgpu_crtc == NULL)
 		return -ENOMEM;
 
-	drm_crtc_init(adev->ddev, &amdgpu_crtc->base, &dce_v10_0_crtc_funcs);
+	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_v10_0_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
 	amdgpu_crtc->crtc_id = index;
@@ -2709,8 +2709,8 @@ static int dce_v10_0_crtc_init(struct amdgpu_device *adev, int index)
 
 	amdgpu_crtc->max_cursor_width = 128;
 	amdgpu_crtc->max_cursor_height = 128;
-	adev->ddev->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
-	adev->ddev->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
+	adev_to_drm(adev)->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
+	adev_to_drm(adev)->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
 
 	switch (amdgpu_crtc->crtc_id) {
 	case 0:
@@ -2792,24 +2792,24 @@ static int dce_v10_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
+	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
 
-	adev->ddev->mode_config.async_page_flip = true;
+	adev_to_drm(adev)->mode_config.async_page_flip = true;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	/* allocate crtcs */
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
@@ -2819,7 +2819,7 @@ static int dce_v10_0_sw_init(void *handle)
 	}
 
 	if (amdgpu_atombios_get_connector_info_from_object_table(adev))
-		amdgpu_display_print_display_setup(adev->ddev);
+		amdgpu_display_print_display_setup(adev_to_drm(adev));
 	else
 		return -EINVAL;
 
@@ -2832,7 +2832,7 @@ static int dce_v10_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	drm_kms_helper_poll_init(adev->ddev);
+	drm_kms_helper_poll_init(adev_to_drm(adev));
 
 	adev->mode_info.mode_config_initialized = true;
 	return 0;
@@ -2844,13 +2844,13 @@ static int dce_v10_0_sw_fini(void *handle)
 
 	kfree(adev->mode_info.bios_hardcoded_edid);
 
-	drm_kms_helper_poll_fini(adev->ddev);
+	drm_kms_helper_poll_fini(adev_to_drm(adev));
 
 	dce_v10_0_audio_fini(adev);
 
 	dce_v10_0_afmt_fini(adev);
 
-	drm_mode_config_cleanup(adev->ddev);
+	drm_mode_config_cleanup(adev_to_drm(adev));
 	adev->mode_info.mode_config_initialized = false;
 
 	return 0;
@@ -3157,14 +3157,14 @@ static int dce_v10_0_pageflip_irq(struct amdgpu_device *adev,
 	if (amdgpu_crtc == NULL)
 		return 0;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED) {
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 						 "AMDGPU_FLIP_SUBMITTED(%d)\n",
 						 amdgpu_crtc->pflip_status,
 						 AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return 0;
 	}
 
@@ -3176,7 +3176,7 @@ static int dce_v10_0_pageflip_irq(struct amdgpu_device *adev,
 	if (works->event)
 		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
 	schedule_work(&works->unpin_work);
@@ -3245,7 +3245,7 @@ static int dce_v10_0_crtc_irq(struct amdgpu_device *adev,
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
 		if (amdgpu_irq_enabled(adev, source, irq_type)) {
-			drm_handle_vblank(adev->ddev, crtc);
+			drm_handle_vblank(adev_to_drm(adev), crtc);
 		}
 		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
 
@@ -3345,7 +3345,7 @@ dce_v10_0_encoder_mode_set(struct drm_encoder *encoder,
 
 static void dce_v10_0_encoder_prepare(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
 
@@ -3385,7 +3385,7 @@ static void dce_v10_0_encoder_prepare(struct drm_encoder *encoder)
 static void dce_v10_0_encoder_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* need to call this here as we need the crtc set up */
 	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
@@ -3485,7 +3485,7 @@ static void dce_v10_0_encoder_add(struct amdgpu_device *adev,
 				 uint32_t supported_device,
 				 u16 caps)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_encoder *encoder;
 	struct amdgpu_encoder *amdgpu_encoder;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 01ce52266966..1954472c8e8f 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -346,7 +346,7 @@ static void dce_v11_0_hpd_set_polarity(struct amdgpu_device *adev,
  */
 static void dce_v11_0_hpd_init(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -400,7 +400,7 @@ static void dce_v11_0_hpd_init(struct amdgpu_device *adev)
  */
 static void dce_v11_0_hpd_fini(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -530,7 +530,7 @@ void dce_v11_0_disable_dce(struct amdgpu_device *adev)
 static void dce_v11_0_program_fmt(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1235,7 +1235,7 @@ static struct amdgpu_audio_pin *dce_v11_0_audio_get_pin(struct amdgpu_device *ad
 
 static void dce_v11_0_afmt_audio_select_pin(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1252,7 +1252,7 @@ static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
 						struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1298,7 +1298,7 @@ static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
 static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1354,7 +1354,7 @@ static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder
 static void dce_v11_0_audio_write_sad_regs(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1525,7 +1525,7 @@ static void dce_v11_0_audio_fini(struct amdgpu_device *adev)
 static void dce_v11_0_afmt_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_afmt_acr acr = amdgpu_afmt_acr(clock);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
@@ -1561,7 +1561,7 @@ static void dce_v11_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 					       void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	uint8_t *frame = buffer + 3;
@@ -1580,7 +1580,7 @@ static void dce_v11_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 static void dce_v11_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
@@ -1611,7 +1611,7 @@ static void dce_v11_0_afmt_setmode(struct drm_encoder *encoder,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1791,7 +1791,7 @@ static void dce_v11_0_afmt_setmode(struct drm_encoder *encoder,
 static void dce_v11_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 
@@ -1864,7 +1864,7 @@ static void dce_v11_0_vga_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 vga_control;
 
 	vga_control = RREG32(vga_control_regs[amdgpu_crtc->crtc_id]) & ~1;
@@ -1878,7 +1878,7 @@ static void dce_v11_0_grph_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (enable)
 		WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, 1);
@@ -1892,7 +1892,7 @@ static int dce_v11_0_crtc_do_set_base(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_framebuffer *target_fb;
 	struct drm_gem_object *obj;
 	struct amdgpu_bo *abo;
@@ -2137,7 +2137,7 @@ static void dce_v11_0_set_interleave(struct drm_crtc *crtc,
 				     struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	u32 tmp;
 
@@ -2153,7 +2153,7 @@ static void dce_v11_0_crtc_load_lut(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u16 *r, *g, *b;
 	int i;
 	u32 tmp;
@@ -2283,7 +2283,7 @@ static u32 dce_v11_0_pick_pll(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 pll_in_use;
 	int pll;
 
@@ -2364,7 +2364,7 @@ static u32 dce_v11_0_pick_pll(struct drm_crtc *crtc)
 
 static void dce_v11_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	uint32_t cur_lock;
 
@@ -2379,7 +2379,7 @@ static void dce_v11_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 static void dce_v11_0_hide_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	u32 tmp;
 
 	tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
@@ -2390,7 +2390,7 @@ static void dce_v11_0_hide_cursor(struct drm_crtc *crtc)
 static void dce_v11_0_show_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	u32 tmp;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
@@ -2408,7 +2408,7 @@ static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
 					int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	int xorigin = 0, yorigin = 0;
 
 	amdgpu_crtc->cursor_x = x;
@@ -2582,7 +2582,7 @@ static const struct drm_crtc_funcs dce_v11_0_crtc_funcs = {
 static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	unsigned type;
 
@@ -2636,7 +2636,7 @@ static void dce_v11_0_crtc_disable(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_atom_ss ss;
 	int i;
 
@@ -2706,7 +2706,7 @@ static int dce_v11_0_crtc_mode_set(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (!amdgpu_crtc->adjusted_clock)
 		return -EINVAL;
@@ -2809,7 +2809,7 @@ static int dce_v11_0_crtc_init(struct amdgpu_device *adev, int index)
 	if (amdgpu_crtc == NULL)
 		return -ENOMEM;
 
-	drm_crtc_init(adev->ddev, &amdgpu_crtc->base, &dce_v11_0_crtc_funcs);
+	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_v11_0_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
 	amdgpu_crtc->crtc_id = index;
@@ -2817,8 +2817,8 @@ static int dce_v11_0_crtc_init(struct amdgpu_device *adev, int index)
 
 	amdgpu_crtc->max_cursor_width = 128;
 	amdgpu_crtc->max_cursor_height = 128;
-	adev->ddev->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
-	adev->ddev->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
+	adev_to_drm(adev)->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
+	adev_to_drm(adev)->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
 
 	switch (amdgpu_crtc->crtc_id) {
 	case 0:
@@ -2913,24 +2913,24 @@ static int dce_v11_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
+	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
 
-	adev->ddev->mode_config.async_page_flip = true;
+	adev_to_drm(adev)->mode_config.async_page_flip = true;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 
 	/* allocate crtcs */
@@ -2941,7 +2941,7 @@ static int dce_v11_0_sw_init(void *handle)
 	}
 
 	if (amdgpu_atombios_get_connector_info_from_object_table(adev))
-		amdgpu_display_print_display_setup(adev->ddev);
+		amdgpu_display_print_display_setup(adev_to_drm(adev));
 	else
 		return -EINVAL;
 
@@ -2954,7 +2954,7 @@ static int dce_v11_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	drm_kms_helper_poll_init(adev->ddev);
+	drm_kms_helper_poll_init(adev_to_drm(adev));
 
 	adev->mode_info.mode_config_initialized = true;
 	return 0;
@@ -2966,13 +2966,13 @@ static int dce_v11_0_sw_fini(void *handle)
 
 	kfree(adev->mode_info.bios_hardcoded_edid);
 
-	drm_kms_helper_poll_fini(adev->ddev);
+	drm_kms_helper_poll_fini(adev_to_drm(adev));
 
 	dce_v11_0_audio_fini(adev);
 
 	dce_v11_0_afmt_fini(adev);
 
-	drm_mode_config_cleanup(adev->ddev);
+	drm_mode_config_cleanup(adev_to_drm(adev));
 	adev->mode_info.mode_config_initialized = false;
 
 	return 0;
@@ -3283,14 +3283,14 @@ static int dce_v11_0_pageflip_irq(struct amdgpu_device *adev,
 	if(amdgpu_crtc == NULL)
 		return 0;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 						 "AMDGPU_FLIP_SUBMITTED(%d)\n",
 						 amdgpu_crtc->pflip_status,
 						 AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return 0;
 	}
 
@@ -3302,7 +3302,7 @@ static int dce_v11_0_pageflip_irq(struct amdgpu_device *adev,
 	if(works->event)
 		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
 	schedule_work(&works->unpin_work);
@@ -3372,7 +3372,7 @@ static int dce_v11_0_crtc_irq(struct amdgpu_device *adev,
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
 		if (amdgpu_irq_enabled(adev, source, irq_type)) {
-			drm_handle_vblank(adev->ddev, crtc);
+			drm_handle_vblank(adev_to_drm(adev), crtc);
 		}
 		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
 
@@ -3471,7 +3471,7 @@ dce_v11_0_encoder_mode_set(struct drm_encoder *encoder,
 
 static void dce_v11_0_encoder_prepare(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
 
@@ -3511,7 +3511,7 @@ static void dce_v11_0_encoder_prepare(struct drm_encoder *encoder)
 static void dce_v11_0_encoder_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* need to call this here as we need the crtc set up */
 	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
@@ -3611,7 +3611,7 @@ static void dce_v11_0_encoder_add(struct amdgpu_device *adev,
 				 uint32_t supported_device,
 				 u16 caps)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_encoder *encoder;
 	struct amdgpu_encoder *amdgpu_encoder;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index cbddead3dafb..3a44753a80d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -279,7 +279,7 @@ static void dce_v6_0_hpd_set_polarity(struct amdgpu_device *adev,
  */
 static void dce_v6_0_hpd_init(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -324,7 +324,7 @@ static void dce_v6_0_hpd_init(struct amdgpu_device *adev)
  */
 static void dce_v6_0_hpd_fini(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -401,7 +401,7 @@ static void dce_v6_0_program_fmt(struct drm_encoder *encoder)
 {
 
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
@@ -1114,7 +1114,7 @@ static struct amdgpu_audio_pin *dce_v6_0_audio_get_pin(struct amdgpu_device *ade
 
 static void dce_v6_0_audio_select_pin(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 
@@ -1130,7 +1130,7 @@ static void dce_v6_0_audio_write_latency_fields(struct drm_encoder *encoder,
 						struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1174,7 +1174,7 @@ static void dce_v6_0_audio_write_latency_fields(struct drm_encoder *encoder,
 static void dce_v6_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1235,7 +1235,7 @@ static void dce_v6_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1392,7 +1392,7 @@ static void dce_v6_0_audio_fini(struct amdgpu_device *adev)
 static void dce_v6_0_audio_set_vbi_packet(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1408,7 +1408,7 @@ static void dce_v6_0_audio_set_acr(struct drm_encoder *encoder,
 				   uint32_t clock, int bpc)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_afmt_acr acr = amdgpu_afmt_acr(clock);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
@@ -1446,7 +1446,7 @@ static void dce_v6_0_audio_set_avi_infoframe(struct drm_encoder *encoder,
 					       struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1488,7 +1488,7 @@ static void dce_v6_0_audio_set_avi_infoframe(struct drm_encoder *encoder,
 static void dce_v6_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
 	int em = amdgpu_atombios_encoder_get_encoder_mode(encoder);
 	u32 tmp;
@@ -1522,7 +1522,7 @@ static void dce_v6_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 static void dce_v6_0_audio_set_packet(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1566,7 +1566,7 @@ static void dce_v6_0_audio_set_packet(struct drm_encoder *encoder)
 static void dce_v6_0_audio_set_mute(struct drm_encoder *encoder, bool mute)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1579,7 +1579,7 @@ static void dce_v6_0_audio_set_mute(struct drm_encoder *encoder, bool mute)
 static void dce_v6_0_audio_hdmi_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1616,7 +1616,7 @@ static void dce_v6_0_audio_hdmi_enable(struct drm_encoder *encoder, bool enable)
 static void dce_v6_0_audio_dp_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 tmp;
@@ -1645,7 +1645,7 @@ static void dce_v6_0_afmt_setmode(struct drm_encoder *encoder,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1714,7 +1714,7 @@ static void dce_v6_0_afmt_setmode(struct drm_encoder *encoder,
 static void dce_v6_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 
@@ -1788,7 +1788,7 @@ static void dce_v6_0_vga_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 vga_control;
 
 	vga_control = RREG32(vga_control_regs[amdgpu_crtc->crtc_id]) & ~1;
@@ -1799,7 +1799,7 @@ static void dce_v6_0_grph_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, enable ? 1 : 0);
 }
@@ -1810,7 +1810,7 @@ static int dce_v6_0_crtc_do_set_base(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_framebuffer *target_fb;
 	struct drm_gem_object *obj;
 	struct amdgpu_bo *abo;
@@ -2033,7 +2033,7 @@ static void dce_v6_0_set_interleave(struct drm_crtc *crtc,
 				    struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
@@ -2048,7 +2048,7 @@ static void dce_v6_0_crtc_load_lut(struct drm_crtc *crtc)
 
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u16 *r, *g, *b;
 	int i;
 
@@ -2148,7 +2148,7 @@ static u32 dce_v6_0_pick_pll(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 pll_in_use;
 	int pll;
 
@@ -2177,7 +2177,7 @@ static u32 dce_v6_0_pick_pll(struct drm_crtc *crtc)
 
 static void dce_v6_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	uint32_t cur_lock;
 
@@ -2192,7 +2192,7 @@ static void dce_v6_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 static void dce_v6_0_hide_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 
 	WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset,
 	       (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) |
@@ -2204,7 +2204,7 @@ static void dce_v6_0_hide_cursor(struct drm_crtc *crtc)
 static void dce_v6_0_show_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(amdgpu_crtc->cursor_addr));
@@ -2222,7 +2222,7 @@ static int dce_v6_0_cursor_move_locked(struct drm_crtc *crtc,
 				       int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	int xorigin = 0, yorigin = 0;
 
 	int w = amdgpu_crtc->cursor_width;
@@ -2397,7 +2397,7 @@ static const struct drm_crtc_funcs dce_v6_0_crtc_funcs = {
 static void dce_v6_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	unsigned type;
 
@@ -2447,7 +2447,7 @@ static void dce_v6_0_crtc_disable(struct drm_crtc *crtc)
 
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_atom_ss ss;
 	int i;
 
@@ -2591,7 +2591,7 @@ static int dce_v6_0_crtc_init(struct amdgpu_device *adev, int index)
 	if (amdgpu_crtc == NULL)
 		return -ENOMEM;
 
-	drm_crtc_init(adev->ddev, &amdgpu_crtc->base, &dce_v6_0_crtc_funcs);
+	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_v6_0_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
 	amdgpu_crtc->crtc_id = index;
@@ -2599,8 +2599,8 @@ static int dce_v6_0_crtc_init(struct amdgpu_device *adev, int index)
 
 	amdgpu_crtc->max_cursor_width = CURSOR_WIDTH;
 	amdgpu_crtc->max_cursor_height = CURSOR_HEIGHT;
-	adev->ddev->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
-	adev->ddev->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
+	adev_to_drm(adev)->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
+	adev_to_drm(adev)->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
 
 	amdgpu_crtc->crtc_offset = crtc_offsets[amdgpu_crtc->crtc_id];
 
@@ -2669,20 +2669,20 @@ static int dce_v6_0_sw_init(void *handle)
 
 	adev->mode_info.mode_config_initialized = true;
 
-	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
-	adev->ddev->mode_config.async_page_flip = true;
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
+	adev_to_drm(adev)->mode_config.async_page_flip = true;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	/* allocate crtcs */
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
@@ -2693,7 +2693,7 @@ static int dce_v6_0_sw_init(void *handle)
 
 	ret = amdgpu_atombios_get_connector_info_from_object_table(adev);
 	if (ret)
-		amdgpu_display_print_display_setup(adev->ddev);
+		amdgpu_display_print_display_setup(adev_to_drm(adev));
 	else
 		return -EINVAL;
 
@@ -2706,7 +2706,7 @@ static int dce_v6_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	drm_kms_helper_poll_init(adev->ddev);
+	drm_kms_helper_poll_init(adev_to_drm(adev));
 
 	return r;
 }
@@ -2717,12 +2717,12 @@ static int dce_v6_0_sw_fini(void *handle)
 
 	kfree(adev->mode_info.bios_hardcoded_edid);
 
-	drm_kms_helper_poll_fini(adev->ddev);
+	drm_kms_helper_poll_fini(adev_to_drm(adev));
 
 	dce_v6_0_audio_fini(adev);
 	dce_v6_0_afmt_fini(adev);
 
-	drm_mode_config_cleanup(adev->ddev);
+	drm_mode_config_cleanup(adev_to_drm(adev));
 	adev->mode_info.mode_config_initialized = false;
 
 	return 0;
@@ -2967,7 +2967,7 @@ static int dce_v6_0_crtc_irq(struct amdgpu_device *adev,
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
 		if (amdgpu_irq_enabled(adev, source, irq_type)) {
-			drm_handle_vblank(adev->ddev, crtc);
+			drm_handle_vblank(adev_to_drm(adev), crtc);
 		}
 		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
 		break;
@@ -3036,14 +3036,14 @@ static int dce_v6_0_pageflip_irq(struct amdgpu_device *adev,
 	if (amdgpu_crtc == NULL)
 		return 0;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 						"AMDGPU_FLIP_SUBMITTED(%d)\n",
 						amdgpu_crtc->pflip_status,
 						AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return 0;
 	}
 
@@ -3055,7 +3055,7 @@ static int dce_v6_0_pageflip_irq(struct amdgpu_device *adev,
 	if (works->event)
 		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
 	schedule_work(&works->unpin_work);
@@ -3146,7 +3146,7 @@ dce_v6_0_encoder_mode_set(struct drm_encoder *encoder,
 static void dce_v6_0_encoder_prepare(struct drm_encoder *encoder)
 {
 
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
 
@@ -3187,7 +3187,7 @@ static void dce_v6_0_encoder_commit(struct drm_encoder *encoder)
 {
 
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* need to call this here as we need the crtc set up */
 	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
@@ -3297,7 +3297,7 @@ static void dce_v6_0_encoder_add(struct amdgpu_device *adev,
 				 uint32_t supported_device,
 				 u16 caps)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_encoder *encoder;
 	struct amdgpu_encoder *amdgpu_encoder;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index fa0ad50b628c..3603e5f13077 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -273,7 +273,7 @@ static void dce_v8_0_hpd_set_polarity(struct amdgpu_device *adev,
  */
 static void dce_v8_0_hpd_init(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -318,7 +318,7 @@ static void dce_v8_0_hpd_init(struct amdgpu_device *adev)
  */
 static void dce_v8_0_hpd_fini(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	u32 tmp;
@@ -444,7 +444,7 @@ void dce_v8_0_disable_dce(struct amdgpu_device *adev)
 static void dce_v8_0_program_fmt(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1146,7 +1146,7 @@ static struct amdgpu_audio_pin *dce_v8_0_audio_get_pin(struct amdgpu_device *ade
 
 static void dce_v8_0_afmt_audio_select_pin(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 offset;
@@ -1164,7 +1164,7 @@ static void dce_v8_0_audio_write_latency_fields(struct drm_encoder *encoder,
 						struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1225,7 +1225,7 @@ static void dce_v8_0_audio_write_latency_fields(struct drm_encoder *encoder,
 static void dce_v8_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector;
@@ -1278,7 +1278,7 @@ static void dce_v8_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
 static void dce_v8_0_audio_write_sad_regs(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	u32 offset;
@@ -1446,7 +1446,7 @@ static void dce_v8_0_audio_fini(struct amdgpu_device *adev)
 static void dce_v8_0_afmt_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_afmt_acr acr = amdgpu_afmt_acr(clock);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
@@ -1469,7 +1469,7 @@ static void dce_v8_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 					       void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
@@ -1489,7 +1489,7 @@ static void dce_v8_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
 static void dce_v8_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
@@ -1516,7 +1516,7 @@ static void dce_v8_0_afmt_setmode(struct drm_encoder *encoder,
 				  struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -1678,7 +1678,7 @@ static void dce_v8_0_afmt_setmode(struct drm_encoder *encoder,
 static void dce_v8_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
 
@@ -1751,7 +1751,7 @@ static void dce_v8_0_vga_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 vga_control;
 
 	vga_control = RREG32(vga_control_regs[amdgpu_crtc->crtc_id]) & ~1;
@@ -1765,7 +1765,7 @@ static void dce_v8_0_grph_enable(struct drm_crtc *crtc, bool enable)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	if (enable)
 		WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, 1);
@@ -1779,7 +1779,7 @@ static int dce_v8_0_crtc_do_set_base(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_framebuffer *target_fb;
 	struct drm_gem_object *obj;
 	struct amdgpu_bo *abo;
@@ -2004,7 +2004,7 @@ static void dce_v8_0_set_interleave(struct drm_crtc *crtc,
 				    struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
@@ -2018,7 +2018,7 @@ static void dce_v8_0_crtc_load_lut(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u16 *r, *g, *b;
 	int i;
 
@@ -2140,7 +2140,7 @@ static u32 dce_v8_0_pick_pll(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 pll_in_use;
 	int pll;
 
@@ -2188,7 +2188,7 @@ static u32 dce_v8_0_pick_pll(struct drm_crtc *crtc)
 
 static void dce_v8_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	uint32_t cur_lock;
 
@@ -2203,7 +2203,7 @@ static void dce_v8_0_lock_cursor(struct drm_crtc *crtc, bool lock)
 static void dce_v8_0_hide_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 
 	WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset,
 	       (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) |
@@ -2213,7 +2213,7 @@ static void dce_v8_0_hide_cursor(struct drm_crtc *crtc)
 static void dce_v8_0_show_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(amdgpu_crtc->cursor_addr));
@@ -2230,7 +2230,7 @@ static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
 				       int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	int xorigin = 0, yorigin = 0;
 
 	amdgpu_crtc->cursor_x = x;
@@ -2404,7 +2404,7 @@ static const struct drm_crtc_funcs dce_v8_0_crtc_funcs = {
 static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	unsigned type;
 
@@ -2458,7 +2458,7 @@ static void dce_v8_0_crtc_disable(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_atom_ss ss;
 	int i;
 
@@ -2609,7 +2609,7 @@ static int dce_v8_0_crtc_init(struct amdgpu_device *adev, int index)
 	if (amdgpu_crtc == NULL)
 		return -ENOMEM;
 
-	drm_crtc_init(adev->ddev, &amdgpu_crtc->base, &dce_v8_0_crtc_funcs);
+	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_v8_0_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
 	amdgpu_crtc->crtc_id = index;
@@ -2617,8 +2617,8 @@ static int dce_v8_0_crtc_init(struct amdgpu_device *adev, int index)
 
 	amdgpu_crtc->max_cursor_width = CIK_CURSOR_WIDTH;
 	amdgpu_crtc->max_cursor_height = CIK_CURSOR_HEIGHT;
-	adev->ddev->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
-	adev->ddev->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
+	adev_to_drm(adev)->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
+	adev_to_drm(adev)->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
 
 	amdgpu_crtc->crtc_offset = crtc_offsets[amdgpu_crtc->crtc_id];
 
@@ -2689,24 +2689,24 @@ static int dce_v8_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
+	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
 
-	adev->ddev->mode_config.async_page_flip = true;
+	adev_to_drm(adev)->mode_config.async_page_flip = true;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	/* allocate crtcs */
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
@@ -2716,7 +2716,7 @@ static int dce_v8_0_sw_init(void *handle)
 	}
 
 	if (amdgpu_atombios_get_connector_info_from_object_table(adev))
-		amdgpu_display_print_display_setup(adev->ddev);
+		amdgpu_display_print_display_setup(adev_to_drm(adev));
 	else
 		return -EINVAL;
 
@@ -2729,7 +2729,7 @@ static int dce_v8_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	drm_kms_helper_poll_init(adev->ddev);
+	drm_kms_helper_poll_init(adev_to_drm(adev));
 
 	adev->mode_info.mode_config_initialized = true;
 	return 0;
@@ -2741,13 +2741,13 @@ static int dce_v8_0_sw_fini(void *handle)
 
 	kfree(adev->mode_info.bios_hardcoded_edid);
 
-	drm_kms_helper_poll_fini(adev->ddev);
+	drm_kms_helper_poll_fini(adev_to_drm(adev));
 
 	dce_v8_0_audio_fini(adev);
 
 	dce_v8_0_afmt_fini(adev);
 
-	drm_mode_config_cleanup(adev->ddev);
+	drm_mode_config_cleanup(adev_to_drm(adev));
 	adev->mode_info.mode_config_initialized = false;
 
 	return 0;
@@ -3057,7 +3057,7 @@ static int dce_v8_0_crtc_irq(struct amdgpu_device *adev,
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
 		if (amdgpu_irq_enabled(adev, source, irq_type)) {
-			drm_handle_vblank(adev->ddev, crtc);
+			drm_handle_vblank(adev_to_drm(adev), crtc);
 		}
 		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
 		break;
@@ -3126,14 +3126,14 @@ static int dce_v8_0_pageflip_irq(struct amdgpu_device *adev,
 	if (amdgpu_crtc == NULL)
 		return 0;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 						"AMDGPU_FLIP_SUBMITTED(%d)\n",
 						amdgpu_crtc->pflip_status,
 						AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return 0;
 	}
 
@@ -3145,7 +3145,7 @@ static int dce_v8_0_pageflip_irq(struct amdgpu_device *adev,
 	if (works->event)
 		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
 	schedule_work(&works->unpin_work);
@@ -3233,7 +3233,7 @@ dce_v8_0_encoder_mode_set(struct drm_encoder *encoder,
 
 static void dce_v8_0_encoder_prepare(struct drm_encoder *encoder)
 {
-	struct amdgpu_device *adev = encoder->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
 
@@ -3273,7 +3273,7 @@ static void dce_v8_0_encoder_prepare(struct drm_encoder *encoder)
 static void dce_v8_0_encoder_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	/* need to call this here as we need the crtc set up */
 	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
@@ -3373,7 +3373,7 @@ static void dce_v8_0_encoder_add(struct amdgpu_device *adev,
 				 uint32_t supported_device,
 				 u16 caps)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_encoder *encoder;
 	struct amdgpu_encoder *amdgpu_encoder;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
index d5ff7b6331ff..b4d4b76538d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
@@ -47,6 +47,9 @@ static void dce_virtual_set_display_funcs(struct amdgpu_device *adev);
 static void dce_virtual_set_irq_funcs(struct amdgpu_device *adev);
 static int dce_virtual_connector_encoder_init(struct amdgpu_device *adev,
 					      int index);
+static int dce_virtual_pageflip(struct amdgpu_device *adev,
+				unsigned crtc_id);
+static enum hrtimer_restart dce_virtual_vblank_timer_handle(struct hrtimer *vblank_timer);
 static void dce_virtual_set_crtc_vblank_interrupt_state(struct amdgpu_device *adev,
 							int crtc,
 							enum amdgpu_interrupt_state state);
@@ -132,7 +135,7 @@ static const struct drm_crtc_funcs dce_virtual_crtc_funcs = {
 static void dce_virtual_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	unsigned type;
 
@@ -171,8 +174,10 @@ static void dce_virtual_crtc_commit(struct drm_crtc *crtc)
 static void dce_virtual_crtc_disable(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
 
-	drm_crtc_vblank_off(crtc);
+	if (dev->num_crtcs)
+		drm_crtc_vblank_off(crtc);
 
 	amdgpu_crtc->enabled = false;
 	amdgpu_crtc->pll_id = ATOM_PPLL_INVALID;
@@ -235,7 +240,7 @@ static int dce_virtual_crtc_init(struct amdgpu_device *adev, int index)
 	if (amdgpu_crtc == NULL)
 		return -ENOMEM;
 
-	drm_crtc_init(adev->ddev, &amdgpu_crtc->base, &dce_virtual_crtc_funcs);
+	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_virtual_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
 	amdgpu_crtc->crtc_id = index;
@@ -247,6 +252,11 @@ static int dce_virtual_crtc_init(struct amdgpu_device *adev, int index)
 	amdgpu_crtc->vsync_timer_enabled = AMDGPU_IRQ_STATE_DISABLE;
 	drm_crtc_helper_add(&amdgpu_crtc->base, &dce_virtual_crtc_helper_funcs);
 
+	hrtimer_init(&amdgpu_crtc->vblank_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_set_expires(&amdgpu_crtc->vblank_timer, DCE_VIRTUAL_VBLANK_PERIOD);
+	amdgpu_crtc->vblank_timer.function = dce_virtual_vblank_timer_handle;
+	hrtimer_start(&amdgpu_crtc->vblank_timer,
+		      DCE_VIRTUAL_VBLANK_PERIOD, HRTIMER_MODE_REL);
 	return 0;
 }
 
@@ -374,24 +384,24 @@ static int dce_virtual_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->ddev->max_vblank_count = 0;
+	adev_to_drm(adev)->max_vblank_count = 0;
 
-	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
+	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	/* allocate crtcs, encoders, connectors */
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
@@ -403,7 +413,7 @@ static int dce_virtual_sw_init(void *handle)
 			return r;
 	}
 
-	drm_kms_helper_poll_init(adev->ddev);
+	drm_kms_helper_poll_init(adev_to_drm(adev));
 
 	adev->mode_info.mode_config_initialized = true;
 	return 0;
@@ -415,9 +425,9 @@ static int dce_virtual_sw_fini(void *handle)
 
 	kfree(adev->mode_info.bios_hardcoded_edid);
 
-	drm_kms_helper_poll_fini(adev->ddev);
+	drm_kms_helper_poll_fini(adev_to_drm(adev));
 
-	drm_mode_config_cleanup(adev->ddev);
+	drm_mode_config_cleanup(adev_to_drm(adev));
 	/* clear crtcs pointer to avoid dce irq finish routine access freed data */
 	memset(adev->mode_info.crtcs, 0, sizeof(adev->mode_info.crtcs[0]) * AMDGPU_MAX_CRTCS);
 	adev->mode_info.mode_config_initialized = false;
@@ -476,7 +486,7 @@ static int dce_virtual_hw_fini(void *handle)
 
 	for (i = 0; i<adev->mode_info.num_crtc; i++)
 		if (adev->mode_info.crtcs[i])
-			dce_virtual_set_crtc_vblank_interrupt_state(adev, i, AMDGPU_IRQ_STATE_DISABLE);
+			hrtimer_cancel(&adev->mode_info.crtcs[i]->vblank_timer);
 
 	return 0;
 }
@@ -602,7 +612,7 @@ static int dce_virtual_connector_encoder_init(struct amdgpu_device *adev,
 	if (!encoder)
 		return -ENOMEM;
 	encoder->possible_crtcs = 1 << index;
-	drm_encoder_init(adev->ddev, encoder, &dce_virtual_encoder_funcs,
+	drm_encoder_init(adev_to_drm(adev), encoder, &dce_virtual_encoder_funcs,
 			 DRM_MODE_ENCODER_VIRTUAL, NULL);
 	drm_encoder_helper_add(encoder, &dce_virtual_encoder_helper_funcs);
 
@@ -613,7 +623,7 @@ static int dce_virtual_connector_encoder_init(struct amdgpu_device *adev,
 	}
 
 	/* add a new connector */
-	drm_connector_init(adev->ddev, connector, &dce_virtual_connector_funcs,
+	drm_connector_init(adev_to_drm(adev), connector, &dce_virtual_connector_funcs,
 			   DRM_MODE_CONNECTOR_VIRTUAL);
 	drm_connector_helper_add(connector, &dce_virtual_connector_helper_funcs);
 	connector->display_info.subpixel_order = SubPixelHorizontalRGB;
@@ -663,14 +673,14 @@ static int dce_virtual_pageflip(struct amdgpu_device *adev,
 	if (amdgpu_crtc == NULL)
 		return 0;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED) {
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 			"AMDGPU_FLIP_SUBMITTED(%d)\n",
 			amdgpu_crtc->pflip_status,
 			AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return 0;
 	}
 
@@ -682,7 +692,7 @@ static int dce_virtual_pageflip(struct amdgpu_device *adev,
 	if (works->event)
 		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
 	amdgpu_bo_unref(&works->old_abo);
@@ -697,10 +707,16 @@ static enum hrtimer_restart dce_virtual_vblank_timer_handle(struct hrtimer *vbla
 	struct amdgpu_crtc *amdgpu_crtc = container_of(vblank_timer,
 						       struct amdgpu_crtc, vblank_timer);
 	struct drm_device *ddev = amdgpu_crtc->base.dev;
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct amdgpu_irq_src *source = adev->irq.client[AMDGPU_IRQ_CLIENTID_LEGACY].sources
+		[VISLANDS30_IV_SRCID_SMU_DISP_TIMER2_TRIGGER];
+	int irq_type = amdgpu_display_crtc_idx_to_irq_type(adev,
+						amdgpu_crtc->crtc_id);
 
-	drm_handle_vblank(ddev, amdgpu_crtc->crtc_id);
-	dce_virtual_pageflip(adev, amdgpu_crtc->crtc_id);
+	if (amdgpu_irq_enabled(adev, source, irq_type)) {
+		drm_handle_vblank(ddev, amdgpu_crtc->crtc_id);
+		dce_virtual_pageflip(adev, amdgpu_crtc->crtc_id);
+	}
 	hrtimer_start(vblank_timer, DCE_VIRTUAL_VBLANK_PERIOD,
 		      HRTIMER_MODE_REL);
 
@@ -716,21 +732,6 @@ static void dce_virtual_set_crtc_vblank_interrupt_state(struct amdgpu_device *ad
 		return;
 	}
 
-	if (state && !adev->mode_info.crtcs[crtc]->vsync_timer_enabled) {
-		DRM_DEBUG("Enable software vsync timer\n");
-		hrtimer_init(&adev->mode_info.crtcs[crtc]->vblank_timer,
-			     CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-		hrtimer_set_expires(&adev->mode_info.crtcs[crtc]->vblank_timer,
-				    DCE_VIRTUAL_VBLANK_PERIOD);
-		adev->mode_info.crtcs[crtc]->vblank_timer.function =
-			dce_virtual_vblank_timer_handle;
-		hrtimer_start(&adev->mode_info.crtcs[crtc]->vblank_timer,
-			      DCE_VIRTUAL_VBLANK_PERIOD, HRTIMER_MODE_REL);
-	} else if (!state && adev->mode_info.crtcs[crtc]->vsync_timer_enabled) {
-		DRM_DEBUG("Disable software vsync timer\n");
-		hrtimer_cancel(&adev->mode_info.crtcs[crtc]->vblank_timer);
-	}
-
 	adev->mode_info.crtcs[crtc]->vsync_timer_enabled = state;
 	DRM_DEBUG("[FM]set crtc %d vblank interrupt state %d\n", crtc, state);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c
index 1ab261836983..7b89fd2aa44a 100644
--- a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c
+++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c
@@ -251,7 +251,7 @@ static ssize_t df_v3_6_get_df_cntr_avail(struct device *dev,
 	int i, count;
 
 	ddev = dev_get_drvdata(dev);
-	adev = ddev->dev_private;
+	adev = drm_to_adev(ddev);
 	count = 0;
 
 	for (i = 0; i < DF_V3_6_MAX_COUNTERS; i++) {
@@ -455,7 +455,8 @@ static int df_v3_6_pmc_get_ctrl_settings(struct amdgpu_device *adev,
 					  uint32_t *lo_base_addr,
 					  uint32_t *hi_base_addr,
 					  uint32_t *lo_val,
-					  uint32_t *hi_val)
+					  uint32_t *hi_val,
+					  bool is_enable)
 {
 
 	uint32_t eventsel, instance, unitmask;
@@ -477,7 +478,8 @@ static int df_v3_6_pmc_get_ctrl_settings(struct amdgpu_device *adev,
 	instance_5432 = (instance >> 2) & 0xf;
 	instance_76 = (instance >> 6) & 0x3;
 
-	*lo_val = (unitmask << 8) | (instance_10 << 6) | eventsel | (1 << 22);
+	*lo_val = (unitmask << 8) | (instance_10 << 6) | eventsel;
+	*lo_val = is_enable ? *lo_val | (1 << 22) : *lo_val & ~(1 << 22);
 	*hi_val = (instance_76 << 29) | instance_5432;
 
 	DRM_DEBUG_DRIVER("config=%llx addr=%08x:%08x val=%08x:%08x",
@@ -572,14 +574,14 @@ static void df_v3_6_reset_perfmon_cntr(struct amdgpu_device *adev,
 }
 
 static int df_v3_6_pmc_start(struct amdgpu_device *adev, uint64_t config,
-			     int is_enable)
+			     int is_add)
 {
 	uint32_t lo_base_addr, hi_base_addr, lo_val, hi_val;
 	int err = 0, ret = 0;
 
 	switch (adev->asic_type) {
 	case CHIP_VEGA20:
-		if (is_enable)
+		if (is_add)
 			return df_v3_6_pmc_add_cntr(adev, config);
 
 		df_v3_6_reset_perfmon_cntr(adev, config);
@@ -589,7 +591,8 @@ static int df_v3_6_pmc_start(struct amdgpu_device *adev, uint64_t config,
 					&lo_base_addr,
 					&hi_base_addr,
 					&lo_val,
-					&hi_val);
+					&hi_val,
+					true);
 
 		if (ret)
 			return ret;
@@ -612,7 +615,7 @@ static int df_v3_6_pmc_start(struct amdgpu_device *adev, uint64_t config,
 }
 
 static int df_v3_6_pmc_stop(struct amdgpu_device *adev, uint64_t config,
-			    int is_disable)
+			    int is_remove)
 {
 	uint32_t lo_base_addr, hi_base_addr, lo_val, hi_val;
 	int ret = 0;
@@ -624,15 +627,17 @@ static int df_v3_6_pmc_stop(struct amdgpu_device *adev, uint64_t config,
 			&lo_base_addr,
 			&hi_base_addr,
 			&lo_val,
-			&hi_val);
+			&hi_val,
+			false);
 
 		if (ret)
 			return ret;
 
-		df_v3_6_reset_perfmon_cntr(adev, config);
 
-		if (is_disable)
+		if (is_remove) {
+			df_v3_6_reset_perfmon_cntr(adev, config);
 			df_v3_6_pmc_release_cntr(adev, config);
+		}
 
 		break;
 	default:
@@ -646,7 +651,7 @@ static void df_v3_6_pmc_get_count(struct amdgpu_device *adev,
 				  uint64_t config,
 				  uint64_t *count)
 {
-	uint32_t lo_base_addr, hi_base_addr, lo_val = 0, hi_val = 0;
+	uint32_t lo_base_addr = 0, hi_base_addr = 0, lo_val = 0, hi_val = 0;
 	*count = 0;
 
 	switch (adev->asic_type) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index f73ce9721233..9792ec737029 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -3307,6 +3307,29 @@ static void gfx_v10_0_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 	adev->gfx.kiq.pmf = &gfx_v10_0_kiq_pm4_funcs;
 }
 
+static void gfx_v10_0_init_spm_golden_registers(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
+	case CHIP_NAVI10:
+		soc15_program_register_sequence(adev,
+						golden_settings_gc_rlc_spm_10_0_nv10,
+						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_0_nv10));
+		break;
+	case CHIP_NAVI14:
+		soc15_program_register_sequence(adev,
+						golden_settings_gc_rlc_spm_10_1_nv14,
+						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_nv14));
+		break;
+	case CHIP_NAVI12:
+		soc15_program_register_sequence(adev,
+						golden_settings_gc_rlc_spm_10_1_2_nv12,
+						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_2_nv12));
+		break;
+	default:
+		break;
+	}
+}
+
 static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 {
 	switch (adev->asic_type) {
@@ -3317,9 +3340,6 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 		soc15_program_register_sequence(adev,
 						golden_settings_gc_10_0_nv10,
 						(const u32)ARRAY_SIZE(golden_settings_gc_10_0_nv10));
-		soc15_program_register_sequence(adev,
-						golden_settings_gc_rlc_spm_10_0_nv10,
-						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_0_nv10));
 		break;
 	case CHIP_NAVI14:
 		soc15_program_register_sequence(adev,
@@ -3328,9 +3348,6 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 		soc15_program_register_sequence(adev,
 						golden_settings_gc_10_1_nv14,
 						(const u32)ARRAY_SIZE(golden_settings_gc_10_1_nv14));
-		soc15_program_register_sequence(adev,
-						golden_settings_gc_rlc_spm_10_1_nv14,
-						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_nv14));
 		break;
 	case CHIP_NAVI12:
 		soc15_program_register_sequence(adev,
@@ -3339,9 +3356,6 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 		soc15_program_register_sequence(adev,
 						golden_settings_gc_10_1_2_nv12,
 						(const u32)ARRAY_SIZE(golden_settings_gc_10_1_2_nv12));
-		soc15_program_register_sequence(adev,
-						golden_settings_gc_rlc_spm_10_1_2_nv12,
-						(const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_2_nv12));
 		break;
 	case CHIP_SIENNA_CICHLID:
 		soc15_program_register_sequence(adev,
@@ -3360,6 +3374,7 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 	default:
 		break;
 	}
+	gfx_v10_0_init_spm_golden_registers(adev);
 }
 
 static void gfx_v10_0_scratch_init(struct amdgpu_device *adev)
@@ -3545,7 +3560,7 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev)
 		break;
 	}
 
-	if (adev->gfx.cp_fw_write_wait == false)
+	if (!adev->gfx.cp_fw_write_wait)
 		DRM_WARN_ONCE("CP firmware version too old, please update!");
 }
 
@@ -4025,21 +4040,23 @@ static int gfx_v10_0_mec_init(struct amdgpu_device *adev)
 	amdgpu_gfx_compute_queue_acquire(adev);
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX10_MEC_HPD_SIZE;
 
-	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) create HDP EOP bo failed\n", r);
-		gfx_v10_0_mec_fini(adev);
-		return r;
-	}
+	if (mec_hpd_size) {
+		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) create HDP EOP bo failed\n", r);
+			gfx_v10_0_mec_fini(adev);
+			return r;
+		}
 
-	memset(hpd, 0, mec_hpd_size);
+		memset(hpd, 0, mec_hpd_size);
 
-	amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
-	amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+	}
 
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) {
 		mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
@@ -4150,6 +4167,7 @@ static const struct amdgpu_gfx_funcs gfx_v10_0_gfx_funcs = {
 	.read_wave_sgprs = &gfx_v10_0_read_wave_sgprs,
 	.read_wave_vgprs = &gfx_v10_0_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v10_0_select_me_pipe_q,
+	.init_spm_golden = &gfx_v10_0_init_spm_golden_registers,
 };
 
 static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev)
@@ -6183,7 +6201,7 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 	struct v10_gfx_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.gfx_ring[0];
 
-	if (!adev->in_gpu_reset && !adev->in_suspend) {
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		memset((void *)mqd, 0, sizeof(*mqd));
 		mutex_lock(&adev->srbm_mutex);
 		nv_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
@@ -6195,7 +6213,7 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 		mutex_unlock(&adev->srbm_mutex);
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (adev->in_gpu_reset) {
+	} else if (amdgpu_in_reset(adev)) {
 		/* reset mqd with the backup copy */
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd));
@@ -6436,6 +6454,10 @@ static int gfx_v10_0_kiq_init_register(struct amdgpu_ring *ring)
 	struct v10_compute_mqd *mqd = ring->mqd_ptr;
 	int j;
 
+	/* inactivate the queue */
+	if (amdgpu_sriov_vf(adev))
+		WREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE, 0);
+
 	/* disable wptr polling */
 	WREG32_FIELD15(GC, 0, CP_PQ_WPTR_POLL_CNTL, EN, 0);
 
@@ -6544,7 +6566,7 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring)
 
 	gfx_v10_0_kiq_setting(ring);
 
-	if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
@@ -6580,7 +6602,7 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring)
 	struct v10_compute_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.compute_ring[0];
 
-	if (!adev->in_gpu_reset && !adev->in_suspend) {
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		memset((void *)mqd, 0, sizeof(*mqd));
 		mutex_lock(&adev->srbm_mutex);
 		nv_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
@@ -6590,7 +6612,7 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
@@ -6961,15 +6983,19 @@ static int gfx_v10_0_hw_fini(void *handle)
 
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
+
+	if (!adev->in_pci_err_recovery) {
 #ifndef BRING_UP_DEBUG
-	if (amdgpu_async_gfx_ring) {
-		r = gfx_v10_0_kiq_disable_kgq(adev);
-		if (r)
-			DRM_ERROR("KGQ disable failed\n");
-	}
+		if (amdgpu_async_gfx_ring) {
+			r = gfx_v10_0_kiq_disable_kgq(adev);
+			if (r)
+				DRM_ERROR("KGQ disable failed\n");
+		}
 #endif
-	if (amdgpu_gfx_disable_kcq(adev))
-		DRM_ERROR("KCQ disable failed\n");
+		if (amdgpu_gfx_disable_kcq(adev))
+			DRM_ERROR("KCQ disable failed\n");
+	}
+
 	if (amdgpu_sriov_vf(adev)) {
 		gfx_v10_0_cp_gfx_enable(adev, false);
 		/* Program KIQ position of RLC_CP_SCHEDULERS during destroy */
@@ -7036,8 +7062,7 @@ static int gfx_v10_0_soft_reset(void *handle)
 		   GRBM_STATUS__BCI_BUSY_MASK | GRBM_STATUS__SX_BUSY_MASK |
 		   GRBM_STATUS__TA_BUSY_MASK | GRBM_STATUS__DB_BUSY_MASK |
 		   GRBM_STATUS__CB_BUSY_MASK | GRBM_STATUS__GDS_BUSY_MASK |
-		   GRBM_STATUS__SPI_BUSY_MASK | GRBM_STATUS__GE_BUSY_NO_DMA_MASK
-		   | GRBM_STATUS__BCI_BUSY_MASK)) {
+		   GRBM_STATUS__SPI_BUSY_MASK | GRBM_STATUS__GE_BUSY_NO_DMA_MASK)) {
 		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
 						GRBM_SOFT_RESET, SOFT_RESET_CP,
 						1);
@@ -7162,7 +7187,7 @@ static int gfx_v10_0_early_init(void *handle)
 		break;
 	}
 
-	adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS;
+	adev->gfx.num_compute_rings = amdgpu_num_kcq;
 
 	gfx_v10_0_set_kiq_pm4_funcs(adev);
 	gfx_v10_0_set_ring_funcs(adev);
@@ -7430,7 +7455,6 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 	    (AMD_CG_SUPPORT_GFX_MGCG |
 	     AMD_CG_SUPPORT_GFX_CGLS |
 	     AMD_CG_SUPPORT_GFX_CGCG |
-	     AMD_CG_SUPPORT_GFX_CGLS |
 	     AMD_CG_SUPPORT_GFX_3D_CGCG |
 	     AMD_CG_SUPPORT_GFX_3D_CGLS))
 		gfx_v10_0_enable_gui_idle_interrupt(adev, enable);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 88f63d7ea371..94b7e0531d09 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1343,21 +1343,22 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
 	amdgpu_gfx_compute_queue_acquire(adev);
 
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX8_MEC_HPD_SIZE;
+	if (mec_hpd_size) {
+		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.mec.hpd_eop_obj,
+					      &adev->gfx.mec.hpd_eop_gpu_addr,
+					      (void **)&hpd);
+		if (r) {
+			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
+			return r;
+		}
 
-	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
-				      &adev->gfx.mec.hpd_eop_obj,
-				      &adev->gfx.mec.hpd_eop_gpu_addr,
-				      (void **)&hpd);
-	if (r) {
-		dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
-		return r;
-	}
-
-	memset(hpd, 0, mec_hpd_size);
+		memset(hpd, 0, mec_hpd_size);
 
-	amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
-	amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+	}
 
 	return 0;
 }
@@ -4632,7 +4633,7 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring)
 
 	gfx_v8_0_kiq_setting(ring);
 
-	if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation));
@@ -4669,7 +4670,7 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring)
 	struct vi_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.compute_ring[0];
 
-	if (!adev->in_gpu_reset && !adev->in_suspend) {
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		memset((void *)mqd, 0, sizeof(struct vi_mqd_allocation));
 		((struct vi_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct vi_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
@@ -4681,7 +4682,7 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct vi_mqd_allocation));
-	} else if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation));
@@ -5294,7 +5295,7 @@ static int gfx_v8_0_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	adev->gfx.num_gfx_rings = GFX8_NUM_GFX_RINGS;
-	adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS;
+	adev->gfx.num_compute_rings = amdgpu_num_kcq;
 	adev->gfx.funcs = &gfx_v8_0_gfx_funcs;
 	gfx_v8_0_set_ring_funcs(adev);
 	gfx_v8_0_set_irq_funcs(adev);
@@ -5342,10 +5343,9 @@ static int gfx_v8_0_late_init(void *handle)
 static void gfx_v8_0_enable_gfx_static_mg_power_gating(struct amdgpu_device *adev,
 						       bool enable)
 {
-	if (((adev->asic_type == CHIP_POLARIS11) ||
+	if ((adev->asic_type == CHIP_POLARIS11) ||
 	    (adev->asic_type == CHIP_POLARIS12) ||
-	    (adev->asic_type == CHIP_VEGAM)) &&
-	    adev->powerplay.pp_funcs->set_powergating_by_smu)
+	    (adev->asic_type == CHIP_VEGAM))
 		/* Send msg to SMU via Powerplay */
 		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, enable);
 
@@ -5879,8 +5879,7 @@ static int gfx_v8_0_tonga_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_CG,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS)) {
@@ -5901,8 +5900,7 @@ static int gfx_v8_0_tonga_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_MG,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	return 0;
@@ -5931,8 +5929,7 @@ static int gfx_v8_0_polaris_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_CG,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_3D_CGCG | AMD_CG_SUPPORT_GFX_3D_CGLS)) {
@@ -5951,8 +5948,7 @@ static int gfx_v8_0_polaris_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_3D,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS)) {
@@ -5973,8 +5969,7 @@ static int gfx_v8_0_polaris_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_MG,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_GFX_RLC_LS) {
@@ -5989,8 +5984,7 @@ static int gfx_v8_0_polaris_update_gfx_clock_gating(struct amdgpu_device *adev,
 				PP_BLOCK_GFX_RLC,
 				pp_support_state,
 				pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) {
@@ -6004,8 +5998,7 @@ static int gfx_v8_0_polaris_update_gfx_clock_gating(struct amdgpu_device *adev,
 			PP_BLOCK_GFX_CP,
 			pp_support_state,
 			pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index b95f22262a90..6959aebae6d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -49,6 +49,7 @@
 #include "amdgpu_ras.h"
 
 #include "gfx_v9_4.h"
+#include "gfx_v9_0.h"
 
 #include "asic_reg/pwr/pwr_10_0_offset.h"
 #include "asic_reg/pwr/pwr_10_0_sh_mask.h"
@@ -788,7 +789,6 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev);
 static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev,
                                  struct amdgpu_cu_info *cu_info);
 static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev);
-static void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, u32 instance);
 static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring);
 static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring);
 static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
@@ -1939,22 +1939,23 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
+	if (mec_hpd_size) {
+		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.mec.hpd_eop_obj,
+					      &adev->gfx.mec.hpd_eop_gpu_addr,
+					      (void **)&hpd);
+		if (r) {
+			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
+			gfx_v9_0_mec_fini(adev);
+			return r;
+		}
 
-	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
-				      &adev->gfx.mec.hpd_eop_obj,
-				      &adev->gfx.mec.hpd_eop_gpu_addr,
-				      (void **)&hpd);
-	if (r) {
-		dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
-		gfx_v9_0_mec_fini(adev);
-		return r;
-	}
-
-	memset(hpd, 0, mec_hpd_size);
+		memset(hpd, 0, mec_hpd_size);
 
-	amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
-	amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+	}
 
 	mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
 
@@ -2074,6 +2075,7 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_gfx_funcs = {
 	.ras_error_inject = &gfx_v9_4_ras_error_inject,
 	.query_ras_error_count = &gfx_v9_4_query_ras_error_count,
 	.reset_ras_error_count = &gfx_v9_4_reset_ras_error_count,
+	.query_ras_error_status = &gfx_v9_4_query_ras_error_status,
 };
 
 static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev)
@@ -2195,7 +2197,6 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev)
 static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 				      int mec, int pipe, int queue)
 {
-	int r;
 	unsigned irq_type;
 	struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id];
 	unsigned int hw_prio;
@@ -2220,13 +2221,8 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue) ?
 			AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL;
 	/* type-2 packets are deprecated on MEC, use type-3 instead */
-	r = amdgpu_ring_init(adev, ring, 1024,
-			     &adev->gfx.eop_irq, irq_type, hw_prio);
-	if (r)
-		return r;
-
-
-	return 0;
+	return amdgpu_ring_init(adev, ring, 1024,
+				&adev->gfx.eop_irq, irq_type, hw_prio);
 }
 
 static int gfx_v9_0_sw_init(void *handle)
@@ -2401,7 +2397,8 @@ static void gfx_v9_0_tiling_mode_table_init(struct amdgpu_device *adev)
 	/* TODO */
 }
 
-static void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, u32 instance)
+void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num,
+			   u32 instance)
 {
 	u32 data;
 
@@ -2559,14 +2556,14 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev)
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
 					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-					    !!amdgpu_noretry);
+					    !!adev->gmc.noretry);
 			WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, tmp);
 			WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, 0);
 		} else {
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
 					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-					    !!amdgpu_noretry);
+					    !!adev->gmc.noretry);
 			WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, tmp);
 			tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
 				(adev->gmc.private_aperture_start >> 48));
@@ -2799,7 +2796,7 @@ static void pwr_10_0_gfxip_control_over_cgpg(struct amdgpu_device *adev,
 	uint32_t default_data = 0;
 
 	default_data = data = RREG32(SOC15_REG_OFFSET(PWR, 0, mmPWR_MISC_CNTL_STATUS));
-	if (enable == true) {
+	if (enable) {
 		/* enable GFXIP control over CGPG */
 		data |= PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK;
 		if(default_data != data)
@@ -3685,7 +3682,7 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 
 	gfx_v9_0_kiq_setting(ring);
 
-	if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
@@ -3723,7 +3720,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 	struct v9_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.compute_ring[0];
 
-	if (!adev->in_gpu_reset && !adev->in_suspend) {
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
@@ -3735,7 +3732,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
-	} else if (adev->in_gpu_reset) { /* for GPU_RESET case */
+	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
@@ -3929,7 +3926,7 @@ static int gfx_v9_0_hw_fini(void *handle)
 	/* Use deinitialize sequence from CAIL when unbinding device from driver,
 	 * otherwise KIQ is hanging when binding back
 	 */
-	if (!adev->in_gpu_reset && !adev->in_suspend) {
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, adev->gfx.kiq.ring.me,
 				adev->gfx.kiq.ring.pipe,
@@ -4087,7 +4084,7 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev)
 	 *
 	 * also don't wait anymore for IRQ context
 	 * */
-	if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
+	if (r < 1 && (amdgpu_in_reset(adev) || in_interrupt()))
 		goto failed_kiq_read;
 
 	might_sleep();
@@ -4626,7 +4623,7 @@ static int gfx_v9_0_early_init(void *handle)
 		adev->gfx.num_gfx_rings = 0;
 	else
 		adev->gfx.num_gfx_rings = GFX9_NUM_GFX_RINGS;
-	adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS;
+	adev->gfx.num_compute_rings = amdgpu_num_kcq;
 	gfx_v9_0_set_kiq_pm4_funcs(adev);
 	gfx_v9_0_set_ring_funcs(adev);
 	gfx_v9_0_set_irq_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
index fa5a3fbaf6ab..dfe8d4841f58 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
@@ -26,9 +26,7 @@
 
 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);
-
-uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev);
-int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info);
+void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num,
+			   u32 instance);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
index 46351db36922..bc699d680ce8 100755..100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
@@ -57,10 +57,10 @@ static const struct soc15_reg_entry gfx_v9_4_edc_counter_regs[] = {
 	/* SPI */
 	{ SOC15_REG_ENTRY(GC, 0, mmSPI_EDC_CNT), 0, 4, 1 },
 	/* SQ */
-	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 0, 4, 16 },
-	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_DED_CNT), 0, 4, 16 },
-	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_INFO), 0, 4, 16 },
-	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_SEC_CNT), 0, 4, 16 },
+	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_CNT), 0, 8, 16 },
+	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_DED_CNT), 0, 8, 16 },
+	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_INFO), 0, 8, 16 },
+	{ SOC15_REG_ENTRY(GC, 0, mmSQ_EDC_SEC_CNT), 0, 8, 16 },
 	/* SQC */
 	{ SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT), 0, 4, 6 },
 	{ SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT2), 0, 4, 6 },
@@ -992,3 +992,32 @@ int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev, void *inject_if)
 
 	return ret;
 }
+
+static const struct soc15_reg_entry gfx_v9_4_rdrsp_status_regs =
+	{ SOC15_REG_ENTRY(GC, 0, mmGCEA_ERR_STATUS), 0, 1, 32 };
+
+void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev)
+{
+	uint32_t i, j;
+	uint32_t reg_value;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
+		return;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	for (i = 0; i < gfx_v9_4_rdrsp_status_regs.se_num; i++) {
+		for (j = 0; j < gfx_v9_4_rdrsp_status_regs.instance;
+		     j++) {
+			gfx_v9_4_select_se_sh(adev, i, 0, j);
+			reg_value = RREG32(SOC15_REG_ENTRY_OFFSET(
+				gfx_v9_4_rdrsp_status_regs));
+			if (reg_value)
+				dev_warn(adev->dev, "GCEA err detected at instance: %d, status: 0x%x!\n",
+						j, reg_value);
+		}
+	}
+
+	gfx_v9_4_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h
index 1ffecc5c0f0a..875f18473a98 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h
@@ -34,4 +34,6 @@ int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev,
 
 void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev);
 
+void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev);
+
 #endif /* __GFX_V9_4_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
index 529e46386a50..fad887a66886 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
@@ -245,7 +245,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_CNTL,
 				    i * hub->ctx_distance, tmp);
 		WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
@@ -403,3 +403,13 @@ void gfxhub_v1_0_init(struct amdgpu_device *adev)
 	hub->eng_addr_distance = mmVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
 		mmVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
 }
+
+
+const struct amdgpu_gfxhub_funcs gfxhub_v1_0_funcs = {
+	.get_mc_fb_offset = gfxhub_v1_0_get_mc_fb_offset,
+	.setup_vm_pt_regs = gfxhub_v1_0_setup_vm_pt_regs,
+	.gart_enable = gfxhub_v1_0_gart_enable,
+	.gart_disable = gfxhub_v1_0_gart_disable,
+	.set_fault_enable_default = gfxhub_v1_0_set_fault_enable_default,
+	.init = gfxhub_v1_0_init,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
index 92d3a70cd9b1..0c46672bbf49 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
@@ -33,4 +33,5 @@ u64 gfxhub_v1_0_get_mc_fb_offset(struct amdgpu_device *adev);
 void gfxhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base);
 
+extern const struct amdgpu_gfxhub_funcs gfxhub_v1_0_funcs;
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.c
index c0ab71df0d90..1e24b6d51e41 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.c
@@ -21,6 +21,7 @@
  *
  */
 #include "amdgpu.h"
+#include "gfxhub_v1_0.h"
 #include "gfxhub_v1_1.h"
 
 #include "gc/gc_9_2_1_offset.h"
@@ -28,7 +29,7 @@
 
 #include "soc15_common.h"
 
-int gfxhub_v1_1_get_xgmi_info(struct amdgpu_device *adev)
+static int gfxhub_v1_1_get_xgmi_info(struct amdgpu_device *adev)
 {
 	u32 xgmi_lfb_cntl = RREG32_SOC15(GC, 0, mmMC_VM_XGMI_LFB_CNTL);
 	u32 max_region =
@@ -66,3 +67,13 @@ int gfxhub_v1_1_get_xgmi_info(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+const struct amdgpu_gfxhub_funcs gfxhub_v1_1_funcs = {
+	.get_mc_fb_offset = gfxhub_v1_0_get_mc_fb_offset,
+	.setup_vm_pt_regs = gfxhub_v1_0_setup_vm_pt_regs,
+	.gart_enable = gfxhub_v1_0_gart_enable,
+	.gart_disable = gfxhub_v1_0_gart_disable,
+	.set_fault_enable_default = gfxhub_v1_0_set_fault_enable_default,
+	.init = gfxhub_v1_0_init,
+	.get_xgmi_info = gfxhub_v1_1_get_xgmi_info,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.h
index d753cf28a0a6..ae5759ffbee3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_1.h
@@ -24,6 +24,6 @@
 #ifndef __GFXHUB_V1_1_H__
 #define __GFXHUB_V1_1_H__
 
-int gfxhub_v1_1_get_xgmi_info(struct amdgpu_device *adev);
+extern const struct amdgpu_gfxhub_funcs gfxhub_v1_1_funcs;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
index 394e6f56948a..456360bf58fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
@@ -31,7 +31,78 @@
 
 #include "soc15_common.h"
 
-u64 gfxhub_v2_0_get_fb_location(struct amdgpu_device *adev)
+static const char *gfxhub_client_ids[] = {
+	"CB/DB",
+	"Reserved",
+	"GE1",
+	"GE2",
+	"CPF",
+	"CPC",
+	"CPG",
+	"RLC",
+	"TCP",
+	"SQC (inst)",
+	"SQC (data)",
+	"SQG",
+	"Reserved",
+	"SDMA0",
+	"SDMA1",
+	"GCR",
+	"SDMA2",
+	"SDMA3",
+};
+
+static uint32_t gfxhub_v2_0_get_invalidate_req(unsigned int vmid,
+					       uint32_t flush_type)
+{
+	u32 req = 0;
+
+	/* invalidate using legacy mode on vmid*/
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
+			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
+			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);
+
+	return req;
+}
+
+static void
+gfxhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
+					     uint32_t status)
+{
+	u32 cid = REG_GET_FIELD(status,
+				GCVM_L2_PROTECTION_FAULT_STATUS, CID);
+
+	dev_err(adev->dev,
+		"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
+		status);
+	dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+		cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid],
+		cid);
+	dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
+	dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR));
+	dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS));
+	dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
+	dev_err(adev->dev, "\t RW: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, RW));
+}
+
+static u64 gfxhub_v2_0_get_fb_location(struct amdgpu_device *adev)
 {
 	u64 base = RREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_BASE);
 
@@ -41,12 +112,12 @@ u64 gfxhub_v2_0_get_fb_location(struct amdgpu_device *adev)
 	return base;
 }
 
-u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev)
+static u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev)
 {
 	return (u64)RREG32_SOC15(GC, 0, mmGCMC_VM_FB_OFFSET) << 24;
 }
 
-void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+static void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
@@ -82,11 +153,6 @@ static void gfxhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev)
 	uint64_t value;
 
 	if (!amdgpu_sriov_vf(adev)) {
-		/*
-		 * the new L1 policy will block SRIOV guest from writing
-		 * these regs, and they will be programed at host.
-		 * so skip programing these regs.
-		 */
 		/* Disable AGP. */
 		WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BASE, 0);
 		WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_TOP, 0);
@@ -247,7 +313,7 @@ static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, GCVM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_CNTL,
 				    i * hub->ctx_distance, tmp);
 		WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
@@ -276,7 +342,7 @@ static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
-int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev)
+static int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev)
 {
 	/* GART Enable. */
 	gfxhub_v2_0_init_gart_aperture_regs(adev);
@@ -292,7 +358,7 @@ int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev)
+static void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 	u32 tmp;
@@ -323,7 +389,7 @@ void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev)
  * @adev: amdgpu_device pointer
  * @value: true redirects VM faults to the default page
  */
-void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
+static void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
 					  bool value)
 {
 	u32 tmp;
@@ -360,7 +426,12 @@ void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
 	WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
-void gfxhub_v2_0_init(struct amdgpu_device *adev)
+static const struct amdgpu_vmhub_funcs gfxhub_v2_0_vmhub_funcs = {
+	.print_l2_protection_fault_status = gfxhub_v2_0_print_l2_protection_fault_status,
+	.get_invalidate_req = gfxhub_v2_0_get_invalidate_req,
+};
+
+static void gfxhub_v2_0_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 
@@ -390,4 +461,24 @@ void gfxhub_v2_0_init(struct amdgpu_device *adev)
 		mmGCVM_INVALIDATE_ENG0_REQ;
 	hub->eng_addr_distance = mmGCVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
 		mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+	hub->vm_cntx_cntl_vm_fault = GCVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
+
+	hub->vmhub_funcs = &gfxhub_v2_0_vmhub_funcs;
 }
+
+const struct amdgpu_gfxhub_funcs gfxhub_v2_0_funcs = {
+	.get_fb_location = gfxhub_v2_0_get_fb_location,
+	.get_mc_fb_offset = gfxhub_v2_0_get_mc_fb_offset,
+	.setup_vm_pt_regs = gfxhub_v2_0_setup_vm_pt_regs,
+	.gart_enable = gfxhub_v2_0_gart_enable,
+	.gart_disable = gfxhub_v2_0_gart_disable,
+	.set_fault_enable_default = gfxhub_v2_0_set_fault_enable_default,
+	.init = gfxhub_v2_0_init,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
index 392b8cd94fc0..9ddc35cd53d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.h
@@ -24,14 +24,6 @@
 #ifndef __GFXHUB_V2_0_H__
 #define __GFXHUB_V2_0_H__
 
-u64 gfxhub_v2_0_get_fb_location(struct amdgpu_device *adev);
-int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev);
-void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev);
-void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
-					  bool value);
-void gfxhub_v2_0_init(struct amdgpu_device *adev);
-u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev);
-void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
-				uint64_t page_table_base);
+extern const struct amdgpu_gfxhub_funcs gfxhub_v2_0_funcs;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
index 5d2505956f84..724bb29e9bb4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
@@ -31,7 +31,78 @@
 
 #include "soc15_common.h"
 
-u64 gfxhub_v2_1_get_fb_location(struct amdgpu_device *adev)
+static const char *gfxhub_client_ids[] = {
+	"CB/DB",
+	"Reserved",
+	"GE1",
+	"GE2",
+	"CPF",
+	"CPC",
+	"CPG",
+	"RLC",
+	"TCP",
+	"SQC (inst)",
+	"SQC (data)",
+	"SQG",
+	"Reserved",
+	"SDMA0",
+	"SDMA1",
+	"GCR",
+	"SDMA2",
+	"SDMA3",
+};
+
+static uint32_t gfxhub_v2_1_get_invalidate_req(unsigned int vmid,
+					       uint32_t flush_type)
+{
+	u32 req = 0;
+
+	/* invalidate using legacy mode on vmid*/
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
+			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
+	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
+			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);
+
+	return req;
+}
+
+static void
+gfxhub_v2_1_print_l2_protection_fault_status(struct amdgpu_device *adev,
+					     uint32_t status)
+{
+	u32 cid = REG_GET_FIELD(status,
+				GCVM_L2_PROTECTION_FAULT_STATUS, CID);
+
+	dev_err(adev->dev,
+		"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
+		status);
+	dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+		cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid],
+		cid);
+	dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
+	dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR));
+	dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS));
+	dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
+	dev_err(adev->dev, "\t RW: 0x%lx\n",
+		REG_GET_FIELD(status,
+		GCVM_L2_PROTECTION_FAULT_STATUS, RW));
+}
+
+static u64 gfxhub_v2_1_get_fb_location(struct amdgpu_device *adev)
 {
 	u64 base = RREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_BASE);
 
@@ -41,12 +112,12 @@ u64 gfxhub_v2_1_get_fb_location(struct amdgpu_device *adev)
 	return base;
 }
 
-u64 gfxhub_v2_1_get_mc_fb_offset(struct amdgpu_device *adev)
+static u64 gfxhub_v2_1_get_mc_fb_offset(struct amdgpu_device *adev)
 {
 	return (u64)RREG32_SOC15(GC, 0, mmGCMC_VM_FB_OFFSET) << 24;
 }
 
-void gfxhub_v2_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+static void gfxhub_v2_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
@@ -248,7 +319,7 @@ static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, GCVM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_CNTL,
 				    i * hub->ctx_distance, tmp);
 		WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
@@ -277,7 +348,7 @@ static void gfxhub_v2_1_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
-int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev)
+static int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev)
 {
 	if (amdgpu_sriov_vf(adev)) {
 		/*
@@ -305,7 +376,7 @@ int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev)
+static void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 	u32 tmp;
@@ -334,7 +405,7 @@ void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev)
  * @adev: amdgpu_device pointer
  * @value: true redirects VM faults to the default page
  */
-void gfxhub_v2_1_set_fault_enable_default(struct amdgpu_device *adev,
+static void gfxhub_v2_1_set_fault_enable_default(struct amdgpu_device *adev,
 					  bool value)
 {
 	u32 tmp;
@@ -378,7 +449,12 @@ void gfxhub_v2_1_set_fault_enable_default(struct amdgpu_device *adev,
 	WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
-void gfxhub_v2_1_init(struct amdgpu_device *adev)
+static const struct amdgpu_vmhub_funcs gfxhub_v2_1_vmhub_funcs = {
+	.print_l2_protection_fault_status = gfxhub_v2_1_print_l2_protection_fault_status,
+	.get_invalidate_req = gfxhub_v2_1_get_invalidate_req,
+};
+
+static void gfxhub_v2_1_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 
@@ -408,9 +484,19 @@ void gfxhub_v2_1_init(struct amdgpu_device *adev)
 		mmGCVM_INVALIDATE_ENG0_REQ;
 	hub->eng_addr_distance = mmGCVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
 		mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+	hub->vm_cntx_cntl_vm_fault = GCVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		GCVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
+
+	hub->vmhub_funcs = &gfxhub_v2_1_vmhub_funcs;
 }
 
-int gfxhub_v2_1_get_xgmi_info(struct amdgpu_device *adev)
+static int gfxhub_v2_1_get_xgmi_info(struct amdgpu_device *adev)
 {
 	u32 xgmi_lfb_cntl = RREG32_SOC15(GC, 0, mmGCMC_VM_XGMI_LFB_CNTL);
 	u32 max_region =
@@ -445,3 +531,14 @@ int gfxhub_v2_1_get_xgmi_info(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+const struct amdgpu_gfxhub_funcs gfxhub_v2_1_funcs = {
+	.get_fb_location = gfxhub_v2_1_get_fb_location,
+	.get_mc_fb_offset = gfxhub_v2_1_get_mc_fb_offset,
+	.setup_vm_pt_regs = gfxhub_v2_1_setup_vm_pt_regs,
+	.gart_enable = gfxhub_v2_1_gart_enable,
+	.gart_disable = gfxhub_v2_1_gart_disable,
+	.set_fault_enable_default = gfxhub_v2_1_set_fault_enable_default,
+	.init = gfxhub_v2_1_init,
+	.get_xgmi_info = gfxhub_v2_1_get_xgmi_info,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.h
index 3452a4e9a3da..f75c2eccfad9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.h
@@ -24,16 +24,6 @@
 #ifndef __GFXHUB_V2_1_H__
 #define __GFXHUB_V2_1_H__
 
-u64 gfxhub_v2_1_get_fb_location(struct amdgpu_device *adev);
-int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev);
-void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev);
-void gfxhub_v2_1_set_fault_enable_default(struct amdgpu_device *adev,
-					  bool value);
-void gfxhub_v2_1_init(struct amdgpu_device *adev);
-u64 gfxhub_v2_1_get_mc_fb_offset(struct amdgpu_device *adev);
-void gfxhub_v2_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
-				uint64_t page_table_base);
-
-int gfxhub_v2_1_get_xgmi_info(struct amdgpu_device *adev);
+extern const struct amdgpu_gfxhub_funcs gfxhub_v2_1_funcs;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index ec90c62078d9..dbc8b76b9b78 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -25,11 +25,10 @@
 #include "amdgpu.h"
 #include "amdgpu_atomfirmware.h"
 #include "gmc_v10_0.h"
+#include "umc_v8_7.h"
 
 #include "hdp/hdp_5_0_0_offset.h"
 #include "hdp/hdp_5_0_0_sh_mask.h"
-#include "gc/gc_10_1_0_sh_mask.h"
-#include "mmhub/mmhub_2_0_0_sh_mask.h"
 #include "athub/athub_2_0_0_sh_mask.h"
 #include "athub/athub_2_0_0_offset.h"
 #include "dcn/dcn_2_0_0_offset.h"
@@ -57,68 +56,31 @@ static const struct soc15_reg_golden golden_settings_navi10_hdp[] =
 };
 #endif
 
+static int gmc_v10_0_ecc_interrupt_state(struct amdgpu_device *adev,
+					 struct amdgpu_irq_src *src,
+					 unsigned type,
+					 enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
 static int
 gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 				   struct amdgpu_irq_src *src, unsigned type,
 				   enum amdgpu_interrupt_state state)
 {
-	struct amdgpu_vmhub *hub;
-	u32 tmp, reg, bits[AMDGPU_MAX_VMHUBS], i;
-
-	bits[AMDGPU_GFXHUB_0] = GCVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		GCVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
-
-	bits[AMDGPU_MMHUB_0] = MMVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
-		MMVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
-
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 		/* MM HUB */
-		hub = &adev->vmhub[AMDGPU_MMHUB_0];
-		for (i = 0; i < 16; i++) {
-			reg = hub->vm_context0_cntl + hub->ctx_distance * i;
-			tmp = RREG32(reg);
-			tmp &= ~bits[AMDGPU_MMHUB_0];
-			WREG32(reg, tmp);
-		}
-
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, false);
 		/* GFX HUB */
-		hub = &adev->vmhub[AMDGPU_GFXHUB_0];
-		for (i = 0; i < 16; i++) {
-			reg = hub->vm_context0_cntl + hub->ctx_distance * i;
-			tmp = RREG32(reg);
-			tmp &= ~bits[AMDGPU_GFXHUB_0];
-			WREG32(reg, tmp);
-		}
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, false);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
 		/* MM HUB */
-		hub = &adev->vmhub[AMDGPU_MMHUB_0];
-		for (i = 0; i < 16; i++) {
-			reg = hub->vm_context0_cntl + hub->ctx_distance * i;
-			tmp = RREG32(reg);
-			tmp |= bits[AMDGPU_MMHUB_0];
-			WREG32(reg, tmp);
-		}
-
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, true);
 		/* GFX HUB */
-		hub = &adev->vmhub[AMDGPU_GFXHUB_0];
-		for (i = 0; i < 16; i++) {
-			reg = hub->vm_context0_cntl + hub->ctx_distance * i;
-			tmp = RREG32(reg);
-			tmp |= bits[AMDGPU_GFXHUB_0];
-			WREG32(reg, tmp);
-		}
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, true);
 		break;
 	default:
 		break;
@@ -166,29 +128,8 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,
 			task_info.task_name, task_info.pid);
 		dev_err(adev->dev, "  in page starting at address 0x%016llx from client %d\n",
 			addr, entry->client_id);
-		if (!amdgpu_sriov_vf(adev)) {
-			dev_err(adev->dev,
-				"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
-				status);
-			dev_err(adev->dev, "\t Faulty UTCL2 client ID: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, CID));
-			dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
-			dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR));
-			dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS));
-			dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
-			dev_err(adev->dev, "\t RW: 0x%lx\n",
-				REG_GET_FIELD(status,
-				GCVM_L2_PROTECTION_FAULT_STATUS, RW));
-		}
+		if (!amdgpu_sriov_vf(adev))
+			hub->vmhub_funcs->print_l2_protection_fault_status(adev, status);
 	}
 
 	return 0;
@@ -199,30 +140,20 @@ static const struct amdgpu_irq_src_funcs gmc_v10_0_irq_funcs = {
 	.process = gmc_v10_0_process_interrupt,
 };
 
-static void gmc_v10_0_set_irq_funcs(struct amdgpu_device *adev)
+static const struct amdgpu_irq_src_funcs gmc_v10_0_ecc_funcs = {
+	.set = gmc_v10_0_ecc_interrupt_state,
+	.process = amdgpu_umc_process_ecc_irq,
+};
+
+ static void gmc_v10_0_set_irq_funcs(struct amdgpu_device *adev)
 {
 	adev->gmc.vm_fault.num_types = 1;
 	adev->gmc.vm_fault.funcs = &gmc_v10_0_irq_funcs;
-}
 
-static uint32_t gmc_v10_0_get_invalidate_req(unsigned int vmid,
-					     uint32_t flush_type)
-{
-	u32 req = 0;
-
-	/* invalidate using legacy mode on vmid*/
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
-			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
-	req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
-			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);
-
-	return req;
+	if (!amdgpu_sriov_vf(adev)) {
+		adev->gmc.ecc_irq.num_types = 1;
+		adev->gmc.ecc_irq.funcs = &gmc_v10_0_ecc_funcs;
+	}
 }
 
 /**
@@ -265,7 +196,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 {
 	bool use_semaphore = gmc_v10_0_use_invalidate_semaphore(adev, vmhub);
 	struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];
-	u32 inv_req = gmc_v10_0_get_invalidate_req(vmid, flush_type);
+	u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type);
 	u32 tmp;
 	/* Use register 17 for GART */
 	const unsigned eng = 17;
@@ -356,16 +287,17 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	 */
 	if (adev->gfx.kiq.ring.sched.ready &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
-	    !adev->in_gpu_reset) {
-
+	    down_read_trylock(&adev->reset_sem)) {
 		struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];
 		const unsigned eng = 17;
-		u32 inv_req = gmc_v10_0_get_invalidate_req(vmid, flush_type);
+		u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type);
 		u32 req = hub->vm_inv_eng0_req + hub->eng_distance * eng;
 		u32 ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng;
 
 		amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req,
 				1 << vmid);
+
+		up_read(&adev->reset_sem);
 		return;
 	}
 
@@ -381,7 +313,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 
 	if (!adev->mman.buffer_funcs_enabled ||
 	    !adev->ib_pool_ready ||
-	    adev->in_gpu_reset ||
+	    amdgpu_in_reset(adev) ||
 	    ring->sched.ready == false) {
 		gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB_0, 0);
 		mutex_unlock(&adev->mman.gtt_window_lock);
@@ -459,7 +391,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		spin_unlock(&adev->gfx.kiq.ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout);
 		if (r < 1) {
-			DRM_ERROR("wait for kiq fence error: %ld.\n", r);
+			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
 			return -ETIME;
 		}
 
@@ -491,7 +423,7 @@ static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
 {
 	bool use_semaphore = gmc_v10_0_use_invalidate_semaphore(ring->adev, ring->funcs->vmhub);
 	struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->funcs->vmhub];
-	uint32_t req = gmc_v10_0_get_invalidate_req(vmid, 0);
+	uint32_t req = hub->vmhub_funcs->get_invalidate_req(vmid, 0);
 	unsigned eng = ring->vm_inv_eng;
 
 	/*
@@ -641,6 +573,28 @@ static void gmc_v10_0_get_vm_pte(struct amdgpu_device *adev,
 	}
 }
 
+static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev)
+{
+	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
+	unsigned size;
+
+	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
+	} else {
+		u32 viewport;
+		u32 pitch;
+
+		viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
+		pitch = RREG32_SOC15(DCE, 0, mmHUBPREQ0_DCSURF_SURFACE_PITCH);
+		size = (REG_GET_FIELD(viewport,
+					HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
+				REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) *
+				4);
+	}
+
+	return size;
+}
+
 static const struct amdgpu_gmc_funcs gmc_v10_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v10_0_flush_gpu_tlb,
 	.flush_gpu_tlb_pasid = gmc_v10_0_flush_gpu_tlb_pasid,
@@ -648,7 +602,8 @@ static const struct amdgpu_gmc_funcs gmc_v10_0_gmc_funcs = {
 	.emit_pasid_mapping = gmc_v10_0_emit_pasid_mapping,
 	.map_mtype = gmc_v10_0_map_mtype,
 	.get_vm_pde = gmc_v10_0_get_vm_pde,
-	.get_vm_pte = gmc_v10_0_get_vm_pte
+	.get_vm_pte = gmc_v10_0_get_vm_pte,
+	.get_vbios_fb_size = gmc_v10_0_get_vbios_fb_size,
 };
 
 static void gmc_v10_0_set_gmc_funcs(struct amdgpu_device *adev)
@@ -657,12 +612,51 @@ static void gmc_v10_0_set_gmc_funcs(struct amdgpu_device *adev)
 		adev->gmc.gmc_funcs = &gmc_v10_0_gmc_funcs;
 }
 
+static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
+	case CHIP_SIENNA_CICHLID:
+		adev->umc.max_ras_err_cnt_per_query = UMC_V8_7_TOTAL_CHANNEL_NUM;
+		adev->umc.channel_inst_num = UMC_V8_7_CHANNEL_INSTANCE_NUM;
+		adev->umc.umc_inst_num = UMC_V8_7_UMC_INSTANCE_NUM;
+		adev->umc.channel_offs = UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA;
+		adev->umc.channel_idx_tbl = &umc_v8_7_channel_idx_tbl[0][0];
+		adev->umc.funcs = &umc_v8_7_funcs;
+		break;
+	default:
+		break;
+	}
+}
+
+
+static void gmc_v10_0_set_mmhub_funcs(struct amdgpu_device *adev)
+{
+	adev->mmhub.funcs = &mmhub_v2_0_funcs;
+}
+
+static void gmc_v10_0_set_gfxhub_funcs(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
+	case CHIP_SIENNA_CICHLID:
+	case CHIP_NAVY_FLOUNDER:
+		adev->gfxhub.funcs = &gfxhub_v2_1_funcs;
+		break;
+	default:
+		adev->gfxhub.funcs = &gfxhub_v2_0_funcs;
+		break;
+	}
+}
+
+
 static int gmc_v10_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	gmc_v10_0_set_mmhub_funcs(adev);
+	gmc_v10_0_set_gfxhub_funcs(adev);
 	gmc_v10_0_set_gmc_funcs(adev);
 	gmc_v10_0_set_irq_funcs(adev);
+	gmc_v10_0_set_umc_funcs(adev);
 
 	adev->gmc.shared_aperture_start = 0x2000000000000000ULL;
 	adev->gmc.shared_aperture_end =
@@ -685,6 +679,10 @@ static int gmc_v10_0_late_init(void *handle)
 	if (r)
 		return r;
 
+	r = amdgpu_gmc_ras_late_init(adev);
+	if (r)
+		return r;
+
 	return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0);
 }
 
@@ -693,11 +691,7 @@ static void gmc_v10_0_vram_gtt_location(struct amdgpu_device *adev,
 {
 	u64 base = 0;
 
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		base = gfxhub_v2_1_get_fb_location(adev);
-	else
-		base = gfxhub_v2_0_get_fb_location(adev);
+	base = adev->gfxhub.funcs->get_fb_location(adev);
 
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
@@ -706,11 +700,7 @@ static void gmc_v10_0_vram_gtt_location(struct amdgpu_device *adev,
 	amdgpu_gmc_gart_location(adev, mc);
 
 	/* base offset of vram pages */
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		adev->vm_manager.vram_base_offset = gfxhub_v2_1_get_mc_fb_offset(adev);
-	else
-		adev->vm_manager.vram_base_offset = gfxhub_v2_0_get_mc_fb_offset(adev);
+	adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev);
 
 	/* add the xgmi offset of the physical node */
 	adev->vm_manager.vram_base_offset +=
@@ -789,48 +779,14 @@ static int gmc_v10_0_gart_init(struct amdgpu_device *adev)
 	return amdgpu_gart_table_vram_alloc(adev);
 }
 
-static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev)
-{
-	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
-	unsigned size;
-
-	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
-		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
-	} else {
-		u32 viewport;
-		u32 pitch;
-
-		viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
-		pitch = RREG32_SOC15(DCE, 0, mmHUBPREQ0_DCSURF_SURFACE_PITCH);
-		size = (REG_GET_FIELD(viewport,
-					HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
-				REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) *
-				4);
-	}
-	/* return 0 if the pre-OS buffer uses up most of vram */
-	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024)) {
-		DRM_ERROR("Warning: pre-OS buffer uses most of vram, \
-				be aware of gart table overwrite\n");
-		return 0;
-	}
-
-	return size;
-}
-
-
-
 static int gmc_v10_0_sw_init(void *handle)
 {
 	int r, vram_width = 0, vram_type = 0, vram_vendor = 0;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		gfxhub_v2_1_init(adev);
-	else
-		gfxhub_v2_0_init(adev);
+	adev->gfxhub.funcs->init(adev);
 
-	mmhub_v2_0_init(adev);
+	adev->mmhub.funcs->init(adev);
 
 	spin_lock_init(&adev->gmc.invalidate_lock);
 
@@ -878,6 +834,14 @@ static int gmc_v10_0_sw_init(void *handle)
 	if (r)
 		return r;
 
+	if (!amdgpu_sriov_vf(adev)) {
+		/* interrupt sent to DF. */
+		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DF, 0,
+				      &adev->gmc.ecc_irq);
+		if (r)
+			return r;
+	}
+
 	/*
 	 * Set the internal MC address mask This is the max address of the GPU's
 	 * internal address space.
@@ -891,7 +855,7 @@ static int gmc_v10_0_sw_init(void *handle)
 	}
 
 	if (adev->gmc.xgmi.supported) {
-		r = gfxhub_v2_1_get_xgmi_info(adev);
+		r = adev->gfxhub.funcs->get_xgmi_info(adev);
 		if (r)
 			return r;
 	}
@@ -900,7 +864,7 @@ static int gmc_v10_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->gmc.stolen_size = gmc_v10_0_get_vbios_fb_size(adev);
+	amdgpu_gmc_get_vbios_allocations(adev);
 
 	/* Memory manager */
 	r = amdgpu_bo_init(adev);
@@ -983,15 +947,11 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		r = gfxhub_v2_1_gart_enable(adev);
-	else
-		r = gfxhub_v2_0_gart_enable(adev);
+	r = adev->gfxhub.funcs->gart_enable(adev);
 	if (r)
 		return r;
 
-	r = mmhub_v2_0_gart_enable(adev);
+	r = adev->mmhub.funcs->gart_enable(adev);
 	if (r)
 		return r;
 
@@ -1008,12 +968,8 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
 
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		gfxhub_v2_1_set_fault_enable_default(adev, value);
-	else
-		gfxhub_v2_0_set_fault_enable_default(adev, value);
-	mmhub_v2_0_set_fault_enable_default(adev, value);
+	adev->gfxhub.funcs->set_fault_enable_default(adev, value);
+	adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0);
 	gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0);
 
@@ -1038,6 +994,9 @@ static int gmc_v10_0_hw_init(void *handle)
 	if (r)
 		return r;
 
+	if (adev->umc.funcs && adev->umc.funcs->init_registers)
+		adev->umc.funcs->init_registers(adev);
+
 	return 0;
 }
 
@@ -1050,12 +1009,8 @@ static int gmc_v10_0_hw_init(void *handle)
  */
 static void gmc_v10_0_gart_disable(struct amdgpu_device *adev)
 {
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
-		gfxhub_v2_1_gart_disable(adev);
-	else
-		gfxhub_v2_0_gart_disable(adev);
-	mmhub_v2_0_gart_disable(adev);
+	adev->gfxhub.funcs->gart_disable(adev);
+	adev->mmhub.funcs->gart_disable(adev);
 	amdgpu_gart_table_vram_unpin(adev);
 }
 
@@ -1069,6 +1024,7 @@ static int gmc_v10_0_hw_fini(void *handle)
 		return 0;
 	}
 
+	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 	gmc_v10_0_gart_disable(adev);
 
@@ -1121,7 +1077,7 @@ static int gmc_v10_0_set_clockgating_state(void *handle,
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	r = mmhub_v2_0_set_clockgating(adev, state);
+	r = adev->mmhub.funcs->set_clockgating(adev, state);
 	if (r)
 		return r;
 
@@ -1136,7 +1092,7 @@ static void gmc_v10_0_get_clockgating_state(void *handle, u32 *flags)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	mmhub_v2_0_get_clockgating(adev, flags);
+	adev->mmhub.funcs->get_clockgating(adev, flags);
 
 	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
 	    adev->asic_type == CHIP_NAVY_FLOUNDER)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index 538e7ee35cdf..95a9117e9564 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -805,16 +805,13 @@ static unsigned gmc_v6_0_get_vbios_fb_size(struct amdgpu_device *adev)
 	unsigned size;
 
 	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
-		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
 	} else {
 		u32 viewport = RREG32(mmVIEWPORT_SIZE);
 		size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
 			REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) *
 			4);
 	}
-	/* return 0 if the pre-OS buffer uses up most of vram */
-	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
-		return 0;
 	return size;
 }
 
@@ -862,7 +859,7 @@ static int gmc_v6_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->gmc.stolen_size = gmc_v6_0_get_vbios_fb_size(adev);
+	amdgpu_gmc_get_vbios_allocations(adev);
 
 	r = amdgpu_bo_init(adev);
 	if (r)
@@ -1136,6 +1133,7 @@ static const struct amdgpu_gmc_funcs gmc_v6_0_gmc_funcs = {
 	.set_prt = gmc_v6_0_set_prt,
 	.get_vm_pde = gmc_v6_0_get_vm_pde,
 	.get_vm_pte = gmc_v6_0_get_vm_pte,
+	.get_vbios_fb_size = gmc_v6_0_get_vbios_fb_size,
 };
 
 static const struct amdgpu_irq_src_funcs gmc_v6_0_irq_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index e18296dc1386..80c146df338a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -434,7 +434,7 @@ static int gmc_v7_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	int vmid;
 	unsigned int tmp;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 	for (vmid = 1; vmid < 16; vmid++) {
@@ -970,16 +970,14 @@ static unsigned gmc_v7_0_get_vbios_fb_size(struct amdgpu_device *adev)
 	unsigned size;
 
 	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
-		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
 	} else {
 		u32 viewport = RREG32(mmVIEWPORT_SIZE);
 		size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
 			REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) *
 			4);
 	}
-	/* return 0 if the pre-OS buffer uses up most of vram */
-	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
-		return 0;
+
 	return size;
 }
 
@@ -1035,7 +1033,7 @@ static int gmc_v7_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->gmc.stolen_size = gmc_v7_0_get_vbios_fb_size(adev);
+	amdgpu_gmc_get_vbios_allocations(adev);
 
 	/* Memory manager */
 	r = amdgpu_bo_init(adev);
@@ -1372,7 +1370,8 @@ static const struct amdgpu_gmc_funcs gmc_v7_0_gmc_funcs = {
 	.emit_pasid_mapping = gmc_v7_0_emit_pasid_mapping,
 	.set_prt = gmc_v7_0_set_prt,
 	.get_vm_pde = gmc_v7_0_get_vm_pde,
-	.get_vm_pte = gmc_v7_0_get_vm_pte
+	.get_vm_pte = gmc_v7_0_get_vm_pte,
+	.get_vbios_fb_size = gmc_v7_0_get_vbios_fb_size,
 };
 
 static const struct amdgpu_irq_src_funcs gmc_v7_0_irq_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index a9e722b8a458..9ab65ca7df77 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -635,7 +635,7 @@ static int gmc_v8_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	int vmid;
 	unsigned int tmp;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
 	for (vmid = 1; vmid < 16; vmid++) {
@@ -1087,16 +1087,14 @@ static unsigned gmc_v8_0_get_vbios_fb_size(struct amdgpu_device *adev)
 	unsigned size;
 
 	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
-		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
 	} else {
 		u32 viewport = RREG32(mmVIEWPORT_SIZE);
 		size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
 			REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) *
 			4);
 	}
-	/* return 0 if the pre-OS buffer uses up most of vram */
-	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
-		return 0;
+
 	return size;
 }
 
@@ -1160,7 +1158,7 @@ static int gmc_v8_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->gmc.stolen_size = gmc_v8_0_get_vbios_fb_size(adev);
+	amdgpu_gmc_get_vbios_allocations(adev);
 
 	/* Memory manager */
 	r = amdgpu_bo_init(adev);
@@ -1739,7 +1737,8 @@ static const struct amdgpu_gmc_funcs gmc_v8_0_gmc_funcs = {
 	.emit_pasid_mapping = gmc_v8_0_emit_pasid_mapping,
 	.set_prt = gmc_v8_0_set_prt,
 	.get_vm_pde = gmc_v8_0_get_vm_pde,
-	.get_vm_pte = gmc_v8_0_get_vm_pte
+	.get_vm_pte = gmc_v8_0_get_vm_pte,
+	.get_vbios_fb_size = gmc_v8_0_get_vbios_fb_size,
 };
 
 static const struct amdgpu_irq_src_funcs gmc_v8_0_irq_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index b67ba38a195f..3ebbddb63705 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -67,6 +67,221 @@
 #define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT__SHIFT                                       0x10
 #define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_WIDTH_MASK                                          0x00003FFFL
 #define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT_MASK                                         0x3FFF0000L
+#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0                                                                  0x049d
+#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0_BASE_IDX                                                         2
+
+
+static const char *gfxhub_client_ids[] = {
+	"CB",
+	"DB",
+	"IA",
+	"WD",
+	"CPF",
+	"CPC",
+	"CPG",
+	"RLC",
+	"TCP",
+	"SQC (inst)",
+	"SQC (data)",
+	"SQG",
+	"PA",
+};
+
+static const char *mmhub_client_ids_raven[][2] = {
+	[0][0] = "MP1",
+	[1][0] = "MP0",
+	[2][0] = "VCN",
+	[3][0] = "VCNU",
+	[4][0] = "HDP",
+	[5][0] = "DCE",
+	[13][0] = "UTCL2",
+	[19][0] = "TLS",
+	[26][0] = "OSS",
+	[27][0] = "SDMA0",
+	[0][1] = "MP1",
+	[1][1] = "MP0",
+	[2][1] = "VCN",
+	[3][1] = "VCNU",
+	[4][1] = "HDP",
+	[5][1] = "XDP",
+	[6][1] = "DBGU0",
+	[7][1] = "DCE",
+	[8][1] = "DCEDWB0",
+	[9][1] = "DCEDWB1",
+	[26][1] = "OSS",
+	[27][1] = "SDMA0",
+};
+
+static const char *mmhub_client_ids_renoir[][2] = {
+	[0][0] = "MP1",
+	[1][0] = "MP0",
+	[2][0] = "HDP",
+	[4][0] = "DCEDMC",
+	[5][0] = "DCEVGA",
+	[13][0] = "UTCL2",
+	[19][0] = "TLS",
+	[26][0] = "OSS",
+	[27][0] = "SDMA0",
+	[28][0] = "VCN",
+	[29][0] = "VCNU",
+	[30][0] = "JPEG",
+	[0][1] = "MP1",
+	[1][1] = "MP0",
+	[2][1] = "HDP",
+	[3][1] = "XDP",
+	[6][1] = "DBGU0",
+	[7][1] = "DCEDMC",
+	[8][1] = "DCEVGA",
+	[9][1] = "DCEDWB",
+	[26][1] = "OSS",
+	[27][1] = "SDMA0",
+	[28][1] = "VCN",
+	[29][1] = "VCNU",
+	[30][1] = "JPEG",
+};
+
+static const char *mmhub_client_ids_vega10[][2] = {
+	[0][0] = "MP0",
+	[1][0] = "UVD",
+	[2][0] = "UVDU",
+	[3][0] = "HDP",
+	[13][0] = "UTCL2",
+	[14][0] = "OSS",
+	[15][0] = "SDMA1",
+	[32+0][0] = "VCE0",
+	[32+1][0] = "VCE0U",
+	[32+2][0] = "XDMA",
+	[32+3][0] = "DCE",
+	[32+4][0] = "MP1",
+	[32+14][0] = "SDMA0",
+	[0][1] = "MP0",
+	[1][1] = "UVD",
+	[2][1] = "UVDU",
+	[3][1] = "DBGU0",
+	[4][1] = "HDP",
+	[5][1] = "XDP",
+	[14][1] = "OSS",
+	[15][1] = "SDMA0",
+	[32+0][1] = "VCE0",
+	[32+1][1] = "VCE0U",
+	[32+2][1] = "XDMA",
+	[32+3][1] = "DCE",
+	[32+4][1] = "DCEDWB",
+	[32+5][1] = "MP1",
+	[32+6][1] = "DBGU1",
+	[32+14][1] = "SDMA1",
+};
+
+static const char *mmhub_client_ids_vega12[][2] = {
+	[0][0] = "MP0",
+	[1][0] = "VCE0",
+	[2][0] = "VCE0U",
+	[3][0] = "HDP",
+	[13][0] = "UTCL2",
+	[14][0] = "OSS",
+	[15][0] = "SDMA1",
+	[32+0][0] = "DCE",
+	[32+1][0] = "XDMA",
+	[32+2][0] = "UVD",
+	[32+3][0] = "UVDU",
+	[32+4][0] = "MP1",
+	[32+15][0] = "SDMA0",
+	[0][1] = "MP0",
+	[1][1] = "VCE0",
+	[2][1] = "VCE0U",
+	[3][1] = "DBGU0",
+	[4][1] = "HDP",
+	[5][1] = "XDP",
+	[14][1] = "OSS",
+	[15][1] = "SDMA0",
+	[32+0][1] = "DCE",
+	[32+1][1] = "DCEDWB",
+	[32+2][1] = "XDMA",
+	[32+3][1] = "UVD",
+	[32+4][1] = "UVDU",
+	[32+5][1] = "MP1",
+	[32+6][1] = "DBGU1",
+	[32+15][1] = "SDMA1",
+};
+
+static const char *mmhub_client_ids_vega20[][2] = {
+	[0][0] = "XDMA",
+	[1][0] = "DCE",
+	[2][0] = "VCE0",
+	[3][0] = "VCE0U",
+	[4][0] = "UVD",
+	[5][0] = "UVD1U",
+	[13][0] = "OSS",
+	[14][0] = "HDP",
+	[15][0] = "SDMA0",
+	[32+0][0] = "UVD",
+	[32+1][0] = "UVDU",
+	[32+2][0] = "MP1",
+	[32+3][0] = "MP0",
+	[32+12][0] = "UTCL2",
+	[32+14][0] = "SDMA1",
+	[0][1] = "XDMA",
+	[1][1] = "DCE",
+	[2][1] = "DCEDWB",
+	[3][1] = "VCE0",
+	[4][1] = "VCE0U",
+	[5][1] = "UVD1",
+	[6][1] = "UVD1U",
+	[7][1] = "DBGU0",
+	[8][1] = "XDP",
+	[13][1] = "OSS",
+	[14][1] = "HDP",
+	[15][1] = "SDMA0",
+	[32+0][1] = "UVD",
+	[32+1][1] = "UVDU",
+	[32+2][1] = "DBGU1",
+	[32+3][1] = "MP1",
+	[32+4][1] = "MP0",
+	[32+14][1] = "SDMA1",
+};
+
+static const char *mmhub_client_ids_arcturus[][2] = {
+	[2][0] = "MP1",
+	[3][0] = "MP0",
+	[10][0] = "UTCL2",
+	[13][0] = "OSS",
+	[14][0] = "HDP",
+	[15][0] = "SDMA0",
+	[32+15][0] = "SDMA1",
+	[64+15][0] = "SDMA2",
+	[96+15][0] = "SDMA3",
+	[128+15][0] = "SDMA4",
+	[160+11][0] = "JPEG",
+	[160+12][0] = "VCN",
+	[160+13][0] = "VCNU",
+	[160+15][0] = "SDMA5",
+	[192+10][0] = "UTCL2",
+	[192+11][0] = "JPEG1",
+	[192+12][0] = "VCN1",
+	[192+13][0] = "VCN1U",
+	[192+15][0] = "SDMA6",
+	[224+15][0] = "SDMA7",
+	[0][1] = "DBGU1",
+	[1][1] = "XDP",
+	[2][1] = "MP1",
+	[3][1] = "MP0",
+	[13][1] = "OSS",
+	[14][1] = "HDP",
+	[15][1] = "SDMA0",
+	[32+15][1] = "SDMA1",
+	[64+15][1] = "SDMA2",
+	[96+15][1] = "SDMA3",
+	[128+15][1] = "SDMA4",
+	[160+11][1] = "JPEG",
+	[160+12][1] = "VCN",
+	[160+13][1] = "VCNU",
+	[160+15][1] = "SDMA5",
+	[192+11][1] = "JPEG1",
+	[192+12][1] = "VCN1",
+	[192+13][1] = "VCN1U",
+	[192+15][1] = "SDMA6",
+	[224+15][1] = "SDMA7",
+};
 
 static const u32 golden_settings_vega10_hdp[] =
 {
@@ -300,9 +515,10 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 {
 	struct amdgpu_vmhub *hub;
 	bool retry_fault = !!(entry->src_data[1] & 0x80);
-	uint32_t status = 0;
+	uint32_t status = 0, cid = 0, rw = 0;
 	u64 addr;
 	char hub_name[10];
+	const char *mmhub_cid;
 
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
@@ -337,6 +553,10 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 			RREG32(hub->vm_l2_pro_fault_status);
 
 		status = RREG32(hub->vm_l2_pro_fault_status);
+		cid = REG_GET_FIELD(status,
+				    VM_L2_PROTECTION_FAULT_STATUS, CID);
+		rw = REG_GET_FIELD(status,
+				   VM_L2_PROTECTION_FAULT_STATUS, RW);
 		WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
 	}
 
@@ -359,9 +579,37 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 			dev_err(adev->dev,
 				"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
 				status);
-			dev_err(adev->dev, "\t Faulty UTCL2 client ID: 0x%lx\n",
-				REG_GET_FIELD(status,
-				VM_L2_PROTECTION_FAULT_STATUS, CID));
+			if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) {
+				dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+					cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid],
+					cid);
+			} else {
+				switch (adev->asic_type) {
+				case CHIP_VEGA10:
+					mmhub_cid = mmhub_client_ids_vega10[cid][rw];
+					break;
+				case CHIP_VEGA12:
+					mmhub_cid = mmhub_client_ids_vega12[cid][rw];
+					break;
+				case CHIP_VEGA20:
+					mmhub_cid = mmhub_client_ids_vega20[cid][rw];
+					break;
+				case CHIP_ARCTURUS:
+					mmhub_cid = mmhub_client_ids_arcturus[cid][rw];
+					break;
+				case CHIP_RAVEN:
+					mmhub_cid = mmhub_client_ids_raven[cid][rw];
+					break;
+				case CHIP_RENOIR:
+					mmhub_cid = mmhub_client_ids_renoir[cid][rw];
+					break;
+				default:
+					mmhub_cid = NULL;
+					break;
+				}
+				dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+					mmhub_cid ? mmhub_cid : "unknown", cid);
+			}
 			dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
 				REG_GET_FIELD(status,
 				VM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
@@ -374,10 +622,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 			dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
 				REG_GET_FIELD(status,
 				VM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
-			dev_err(adev->dev, "\t RW: 0x%lx\n",
-				REG_GET_FIELD(status,
-				VM_L2_PROTECTION_FAULT_STATUS, RW));
-
+			dev_err(adev->dev, "\t RW: 0x%x\n", rw);
 		}
 	}
 
@@ -500,13 +745,14 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	 * as GFXOFF under bare metal
 	 */
 	if (adev->gfx.kiq.ring.sched.ready &&
-			(amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
-			!adev->in_gpu_reset) {
+	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
+	    down_read_trylock(&adev->reset_sem)) {
 		uint32_t req = hub->vm_inv_eng0_req + hub->eng_distance * eng;
 		uint32_t ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng;
 
 		amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req,
 						   1 << vmid);
+		up_read(&adev->reset_sem);
 		return;
 	}
 
@@ -596,10 +842,10 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EIO;
 
-	if (ring->sched.ready) {
+	if (ring->sched.ready && down_read_trylock(&adev->reset_sem)) {
 		/* Vega20+XGMI caches PTEs in TC and TLB. Add a
 		 * heavy-weight TLB flush (type 2), which flushes
 		 * both. Due to a race condition with concurrent
@@ -626,6 +872,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		if (r) {
 			amdgpu_ring_undo(ring);
 			spin_unlock(&adev->gfx.kiq.ring_lock);
+			up_read(&adev->reset_sem);
 			return -ETIME;
 		}
 
@@ -633,10 +880,11 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		spin_unlock(&adev->gfx.kiq.ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout);
 		if (r < 1) {
-			DRM_ERROR("wait for kiq fence error: %ld.\n", r);
+			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
+			up_read(&adev->reset_sem);
 			return -ETIME;
 		}
-
+		up_read(&adev->reset_sem);
 		return 0;
 	}
 
@@ -826,6 +1074,41 @@ static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev,
 		*flags |= AMDGPU_PTE_SNOOPED;
 }
 
+static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
+{
+	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
+	unsigned size;
+
+	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
+	} else {
+		u32 viewport;
+
+		switch (adev->asic_type) {
+		case CHIP_RAVEN:
+		case CHIP_RENOIR:
+			viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
+			size = (REG_GET_FIELD(viewport,
+					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
+				REG_GET_FIELD(viewport,
+					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_WIDTH) *
+				4);
+			break;
+		case CHIP_VEGA10:
+		case CHIP_VEGA12:
+		case CHIP_VEGA20:
+		default:
+			viewport = RREG32_SOC15(DCE, 0, mmSCL0_VIEWPORT_SIZE);
+			size = (REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
+				REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_WIDTH) *
+				4);
+			break;
+		}
+	}
+
+	return size;
+}
+
 static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
 	.flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid,
@@ -833,7 +1116,8 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping,
 	.map_mtype = gmc_v9_0_map_mtype,
 	.get_vm_pde = gmc_v9_0_get_vm_pde,
-	.get_vm_pte = gmc_v9_0_get_vm_pte
+	.get_vm_pte = gmc_v9_0_get_vm_pte,
+	.get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size,
 };
 
 static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev)
@@ -871,13 +1155,24 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 static void gmc_v9_0_set_mmhub_funcs(struct amdgpu_device *adev)
 {
 	switch (adev->asic_type) {
-	case CHIP_VEGA20:
+	case CHIP_ARCTURUS:
+		adev->mmhub.funcs = &mmhub_v9_4_funcs;
+		break;
+	default:
 		adev->mmhub.funcs = &mmhub_v1_0_funcs;
 		break;
+	}
+}
+
+static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
 	case CHIP_ARCTURUS:
-		adev->mmhub.funcs = &mmhub_v9_4_funcs;
+	case CHIP_VEGA20:
+		adev->gfxhub.funcs = &gfxhub_v1_1_funcs;
 		break;
 	default:
+		adev->gfxhub.funcs = &gfxhub_v1_0_funcs;
 		break;
 	}
 }
@@ -890,6 +1185,7 @@ static int gmc_v9_0_early_init(void *handle)
 	gmc_v9_0_set_irq_funcs(adev);
 	gmc_v9_0_set_umc_funcs(adev);
 	gmc_v9_0_set_mmhub_funcs(adev);
+	gmc_v9_0_set_gfxhub_funcs(adev);
 
 	adev->gmc.shared_aperture_start = 0x2000000000000000ULL;
 	adev->gmc.shared_aperture_end =
@@ -901,57 +1197,26 @@ static int gmc_v9_0_early_init(void *handle)
 	return 0;
 }
 
-static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev)
-{
-
-	/*
-	 * TODO:
-	 * Currently there is a bug where some memory client outside
-	 * of the driver writes to first 8M of VRAM on S3 resume,
-	 * this overrides GART which by default gets placed in first 8M and
-	 * causes VM_FAULTS once GTT is accessed.
-	 * Keep the stolen memory reservation until the while this is not solved.
-	 * Also check code in gmc_v9_0_get_vbios_fb_size and gmc_v9_0_late_init
-	 */
-	switch (adev->asic_type) {
-	case CHIP_VEGA10:
-	case CHIP_RAVEN:
-	case CHIP_ARCTURUS:
-	case CHIP_RENOIR:
-		return true;
-	case CHIP_VEGA12:
-	case CHIP_VEGA20:
-	default:
-		return false;
-	}
-}
-
 static int gmc_v9_0_late_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int r;
 
-	if (!gmc_v9_0_keep_stolen_memory(adev))
-		amdgpu_bo_late_init(adev);
+	amdgpu_bo_late_init(adev);
 
 	r = amdgpu_gmc_allocate_vm_inv_eng(adev);
 	if (r)
 		return r;
-	/* Check if ecc is available */
+
+	/*
+	 * Workaround performance drop issue with VBIOS enables partial
+	 * writes, while disables HBM ECC for vega10.
+	 */
 	if (!amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_VEGA10)) {
-		r = amdgpu_atomfirmware_mem_ecc_supported(adev);
-		if (!r) {
-			DRM_INFO("ECC is not present.\n");
+		if (!(adev->ras_features & (1 << AMDGPU_RAS_BLOCK__UMC))) {
 			if (adev->df.funcs->enable_ecc_force_par_wr_rmw)
 				adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false);
-		} else
-			DRM_INFO("ECC is active.\n");
-
-		r = amdgpu_atomfirmware_sram_ecc_supported(adev);
-		if (!r)
-			DRM_INFO("SRAM ECC is not present.\n");
-		else
-			DRM_INFO("SRAM ECC is active.\n");
+		}
 	}
 
 	if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count)
@@ -969,10 +1234,8 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
 {
 	u64 base = 0;
 
-	if (adev->asic_type == CHIP_ARCTURUS)
-		base = mmhub_v9_4_get_fb_location(adev);
-	else if (!amdgpu_sriov_vf(adev))
-		base = mmhub_v1_0_get_fb_location(adev);
+	if (!amdgpu_sriov_vf(adev))
+		base = adev->mmhub.funcs->get_fb_location(adev);
 
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
@@ -980,7 +1243,7 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
 	amdgpu_gmc_gart_location(adev, mc);
 	amdgpu_gmc_agp_location(adev, mc);
 	/* base offset of vram pages */
-	adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev);
+	adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev);
 
 	/* XXX: add the xgmi offset of the physical node? */
 	adev->vm_manager.vram_base_offset +=
@@ -1015,7 +1278,7 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev)
 
 #ifdef CONFIG_X86_64
 	if (adev->flags & AMD_IS_APU) {
-		adev->gmc.aper_base = gfxhub_v1_0_get_mc_fb_offset(adev);
+		adev->gmc.aper_base = adev->gfxhub.funcs->get_mc_fb_offset(adev);
 		adev->gmc.aper_size = adev->gmc.real_vram_size;
 	}
 #endif
@@ -1066,50 +1329,18 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev)
 	return amdgpu_gart_table_vram_alloc(adev);
 }
 
-static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
+/**
+ * gmc_v9_0_save_registers - saves regs
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * This saves potential register values that should be
+ * restored upon resume
+ */
+static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
 {
-	u32 d1vga_control;
-	unsigned size;
-
-	/*
-	 * TODO Remove once GART corruption is resolved
-	 * Check related code in gmc_v9_0_sw_fini
-	 * */
-	if (gmc_v9_0_keep_stolen_memory(adev))
-		return 9 * 1024 * 1024;
-
-	d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
-	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
-		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
-	} else {
-		u32 viewport;
-
-		switch (adev->asic_type) {
-		case CHIP_RAVEN:
-		case CHIP_RENOIR:
-			viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
-			size = (REG_GET_FIELD(viewport,
-					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
-				REG_GET_FIELD(viewport,
-					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_WIDTH) *
-				4);
-			break;
-		case CHIP_VEGA10:
-		case CHIP_VEGA12:
-		case CHIP_VEGA20:
-		default:
-			viewport = RREG32_SOC15(DCE, 0, mmSCL0_VIEWPORT_SIZE);
-			size = (REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
-				REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_WIDTH) *
-				4);
-			break;
-		}
-	}
-	/* return 0 if the pre-OS buffer uses up most of vram */
-	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
-		return 0;
-
-	return size;
+	if (adev->asic_type == CHIP_RAVEN)
+		adev->gmc.sdpif_register = RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0);
 }
 
 static int gmc_v9_0_sw_init(void *handle)
@@ -1117,11 +1348,9 @@ static int gmc_v9_0_sw_init(void *handle)
 	int r, vram_width = 0, vram_type = 0, vram_vendor = 0;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	gfxhub_v1_0_init(adev);
-	if (adev->asic_type == CHIP_ARCTURUS)
-		mmhub_v9_4_init(adev);
-	else
-		mmhub_v1_0_init(adev);
+	adev->gfxhub.funcs->init(adev);
+
+	adev->mmhub.funcs->init(adev);
 
 	spin_lock_init(&adev->gmc.invalidate_lock);
 
@@ -1233,7 +1462,7 @@ static int gmc_v9_0_sw_init(void *handle)
 	adev->need_swiotlb = drm_need_swiotlb(44);
 
 	if (adev->gmc.xgmi.supported) {
-		r = gfxhub_v1_1_get_xgmi_info(adev);
+		r = adev->gfxhub.funcs->get_xgmi_info(adev);
 		if (r)
 			return r;
 	}
@@ -1242,7 +1471,7 @@ static int gmc_v9_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->gmc.stolen_size = gmc_v9_0_get_vbios_fb_size(adev);
+	amdgpu_gmc_get_vbios_allocations(adev);
 
 	/* Memory manager */
 	r = amdgpu_bo_init(adev);
@@ -1268,21 +1497,18 @@ static int gmc_v9_0_sw_init(void *handle)
 
 	amdgpu_vm_manager_init(adev);
 
+	gmc_v9_0_save_registers(adev);
+
 	return 0;
 }
 
 static int gmc_v9_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	void *stolen_vga_buf;
 
 	amdgpu_gmc_ras_fini(adev);
 	amdgpu_gem_force_release(adev);
 	amdgpu_vm_manager_fini(adev);
-
-	if (gmc_v9_0_keep_stolen_memory(adev))
-		amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, &stolen_vga_buf);
-
 	amdgpu_gart_table_vram_free(adev);
 	amdgpu_bo_fini(adev);
 	amdgpu_gart_fini(adev);
@@ -1326,10 +1552,13 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev)
  *
  * This restores register values, saved at suspend.
  */
-static void gmc_v9_0_restore_registers(struct amdgpu_device *adev)
+void gmc_v9_0_restore_registers(struct amdgpu_device *adev)
 {
-	if (adev->asic_type == CHIP_RAVEN)
-		WREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0, adev->gmc.sdpif_register);
+	if (adev->asic_type == CHIP_RAVEN) {
+		WREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0, adev->gmc.sdpif_register);
+		WARN_ON(adev->gmc.sdpif_register !=
+			RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0));
+	}
 }
 
 /**
@@ -1349,14 +1578,11 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	r = gfxhub_v1_0_gart_enable(adev);
+	r = adev->gfxhub.funcs->gart_enable(adev);
 	if (r)
 		return r;
 
-	if (adev->asic_type == CHIP_ARCTURUS)
-		r = mmhub_v9_4_gart_enable(adev);
-	else
-		r = mmhub_v1_0_gart_enable(adev);
+	r = adev->mmhub.funcs->gart_enable(adev);
 	if (r)
 		return r;
 
@@ -1391,11 +1617,10 @@ static int gmc_v9_0_hw_init(void *handle)
 						golden_settings_vega10_hdp,
 						ARRAY_SIZE(golden_settings_vega10_hdp));
 
+	if (adev->mmhub.funcs->update_power_gating)
+		adev->mmhub.funcs->update_power_gating(adev, true);
+
 	switch (adev->asic_type) {
-	case CHIP_RAVEN:
-		/* TODO for renoir */
-		mmhub_v1_0_update_power_gating(adev, true);
-		break;
 	case CHIP_ARCTURUS:
 		WREG32_FIELD15(HDP, 0, HDP_MMHUB_CNTL, HDP_MMHUB_GCC, 1);
 		break;
@@ -1420,11 +1645,8 @@ static int gmc_v9_0_hw_init(void *handle)
 		value = true;
 
 	if (!amdgpu_sriov_vf(adev)) {
-		gfxhub_v1_0_set_fault_enable_default(adev, value);
-		if (adev->asic_type == CHIP_ARCTURUS)
-			mmhub_v9_4_set_fault_enable_default(adev, value);
-		else
-			mmhub_v1_0_set_fault_enable_default(adev, value);
+		adev->gfxhub.funcs->set_fault_enable_default(adev, value);
+		adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	}
 	for (i = 0; i < adev->num_vmhubs; ++i)
 		gmc_v9_0_flush_gpu_tlb(adev, 0, i, 0);
@@ -1438,20 +1660,6 @@ static int gmc_v9_0_hw_init(void *handle)
 }
 
 /**
- * gmc_v9_0_save_registers - saves regs
- *
- * @adev: amdgpu_device pointer
- *
- * This saves potential register values that should be
- * restored upon resume
- */
-static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
-{
-	if (adev->asic_type == CHIP_RAVEN)
-		adev->gmc.sdpif_register = RREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0);
-}
-
-/**
  * gmc_v9_0_gart_disable - gart disable
  *
  * @adev: amdgpu_device pointer
@@ -1460,11 +1668,8 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
  */
 static void gmc_v9_0_gart_disable(struct amdgpu_device *adev)
 {
-	gfxhub_v1_0_gart_disable(adev);
-	if (adev->asic_type == CHIP_ARCTURUS)
-		mmhub_v9_4_gart_disable(adev);
-	else
-		mmhub_v1_0_gart_disable(adev);
+	adev->gfxhub.funcs->gart_disable(adev);
+	adev->mmhub.funcs->gart_disable(adev);
 	amdgpu_gart_table_vram_unpin(adev);
 }
 
@@ -1487,16 +1692,9 @@ static int gmc_v9_0_hw_fini(void *handle)
 
 static int gmc_v9_0_suspend(void *handle)
 {
-	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	r = gmc_v9_0_hw_fini(adev);
-	if (r)
-		return r;
-
-	gmc_v9_0_save_registers(adev);
-
-	return 0;
+	return gmc_v9_0_hw_fini(adev);
 }
 
 static int gmc_v9_0_resume(void *handle)
@@ -1504,7 +1702,6 @@ static int gmc_v9_0_resume(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	gmc_v9_0_restore_registers(adev);
 	r = gmc_v9_0_hw_init(adev);
 	if (r)
 		return r;
@@ -1537,10 +1734,7 @@ static int gmc_v9_0_set_clockgating_state(void *handle,
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (adev->asic_type == CHIP_ARCTURUS)
-		mmhub_v9_4_set_clockgating(adev, state);
-	else
-		mmhub_v1_0_set_clockgating(adev, state);
+	adev->mmhub.funcs->set_clockgating(adev, state);
 
 	athub_v1_0_set_clockgating(adev, state);
 
@@ -1551,10 +1745,7 @@ static void gmc_v9_0_get_clockgating_state(void *handle, u32 *flags)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (adev->asic_type == CHIP_ARCTURUS)
-		mmhub_v9_4_get_clockgating(adev, flags);
-	else
-		mmhub_v1_0_get_clockgating(adev, flags);
+	adev->mmhub.funcs->get_clockgating(adev, flags);
 
 	athub_v1_0_get_clockgating(adev, flags);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h
index e0585e8c6c1b..c415c439f690 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h
@@ -26,4 +26,6 @@
 
 extern const struct amd_ip_funcs gmc_v9_0_ip_funcs;
 extern const struct amdgpu_ip_block_version gmc_v9_0_ip_block;
+
+void gmc_v9_0_restore_registers(struct amdgpu_device *adev);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index 7a51c615d22d..845306f63cdb 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -55,22 +55,18 @@ static int amdgpu_ih_clientid_jpeg[] = {
 static int jpeg_v2_5_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	if (adev->asic_type == CHIP_ARCTURUS) {
-		u32 harvest;
-		int i;
-
-		adev->jpeg.num_jpeg_inst = JPEG25_MAX_HW_INSTANCES_ARCTURUS;
-		for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) {
-			harvest = RREG32_SOC15(JPEG, i, mmCC_UVD_HARVESTING);
-			if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK)
-				adev->jpeg.harvest_config |= 1 << i;
-		}
+	u32 harvest;
+	int i;
 
-		if (adev->jpeg.harvest_config == (AMDGPU_JPEG_HARVEST_JPEG0 |
-						 AMDGPU_JPEG_HARVEST_JPEG1))
-			return -ENOENT;
-	} else
-		adev->jpeg.num_jpeg_inst = 1;
+	adev->jpeg.num_jpeg_inst = JPEG25_MAX_HW_INSTANCES_ARCTURUS;
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) {
+		harvest = RREG32_SOC15(JPEG, i, mmCC_UVD_HARVESTING);
+		if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK)
+			adev->jpeg.harvest_config |= 1 << i;
+	}
+	if (adev->jpeg.harvest_config == (AMDGPU_JPEG_HARVEST_JPEG0 |
+					 AMDGPU_JPEG_HARVEST_JPEG1))
+		return -ENOENT;
 
 	jpeg_v2_5_set_dec_ring_funcs(adev);
 	jpeg_v2_5_set_irq_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index c41e5590a701..3a0dff53654d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -460,15 +460,10 @@ static bool jpeg_v3_0_is_idle(void *handle)
 static int jpeg_v3_0_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int ret;
 
-	ret = SOC15_WAIT_ON_RREG(JPEG, 0, mmUVD_JRBC_STATUS,
+	return SOC15_WAIT_ON_RREG(JPEG, 0, mmUVD_JRBC_STATUS,
 		UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
 		UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
-	if (ret)
-		return ret;
-
-	return ret;
 }
 
 static int jpeg_v3_0_set_clockgating_state(void *handle,
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index 4b746584a797..1c22d8393b21 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -832,7 +832,6 @@ static int mes_v10_1_queue_init(struct amdgpu_device *adev)
 static int mes_v10_1_ring_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
-	int r;
 
 	ring = &adev->mes.ring;
 
@@ -849,11 +848,7 @@ static int mes_v10_1_ring_init(struct amdgpu_device *adev)
 	ring->no_scheduler = true;
 	sprintf(ring->name, "mes_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
-	r = amdgpu_ring_init(adev, ring, 1024, NULL, 0, AMDGPU_RING_PRIO_DEFAULT);
-	if (r)
-		return r;
-
-	return 0;
+	return amdgpu_ring_init(adev, ring, 1024, NULL, 0, AMDGPU_RING_PRIO_DEFAULT);
 }
 
 static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index dffcb93ecee5..f84701c562bf 100755..100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -34,7 +34,7 @@
 #define mmDAGB0_CNTL_MISC2_RV 0x008f
 #define mmDAGB0_CNTL_MISC2_RV_BASE_IDX 0
 
-u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev)
+static u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev)
 {
 	u64 base = RREG32_SOC15(MMHUB, 0, mmMC_VM_FB_LOCATION_BASE);
 	u64 top = RREG32_SOC15(MMHUB, 0, mmMC_VM_FB_LOCATION_TOP);
@@ -51,7 +51,7 @@ u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev)
 	return base;
 }
 
-void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+static void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
@@ -268,7 +268,7 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_CNTL,
 				    i * hub->ctx_distance, tmp);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
@@ -297,20 +297,19 @@ static void mmhub_v1_0_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
-void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
+static void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
 				bool enable)
 {
 	if (amdgpu_sriov_vf(adev))
 		return;
 
 	if (enable && adev->pg_flags & AMD_PG_SUPPORT_MMHUB) {
-		if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->set_powergating_by_smu)
-			amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GMC, true);
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GMC, true);
 
 	}
 }
 
-int mmhub_v1_0_gart_enable(struct amdgpu_device *adev)
+static int mmhub_v1_0_gart_enable(struct amdgpu_device *adev)
 {
 	if (amdgpu_sriov_vf(adev)) {
 		/*
@@ -338,7 +337,7 @@ int mmhub_v1_0_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-void mmhub_v1_0_gart_disable(struct amdgpu_device *adev)
+static void mmhub_v1_0_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
 	u32 tmp;
@@ -373,7 +372,7 @@ void mmhub_v1_0_gart_disable(struct amdgpu_device *adev)
  * @adev: amdgpu_device pointer
  * @value: true redirects VM faults to the default page
  */
-void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
+static void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 {
 	u32 tmp;
 
@@ -415,7 +414,7 @@ void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
-void mmhub_v1_0_init(struct amdgpu_device *adev)
+static void mmhub_v1_0_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
 
@@ -525,7 +524,7 @@ static void mmhub_v1_0_update_medium_grain_light_sleep(struct amdgpu_device *ade
 		WREG32_SOC15(MMHUB, 0, mmATC_L2_MISC_CG, data);
 }
 
-int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev,
+static int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev,
 			       enum amd_clockgating_state state)
 {
 	if (amdgpu_sriov_vf(adev))
@@ -549,7 +548,7 @@ int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev,
 	return 0;
 }
 
-void mmhub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags)
+static void mmhub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags)
 {
 	int data, data1;
 
@@ -781,4 +780,13 @@ const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = {
 	.ras_late_init = amdgpu_mmhub_ras_late_init,
 	.query_ras_error_count = mmhub_v1_0_query_ras_error_count,
 	.reset_ras_error_count = mmhub_v1_0_reset_ras_error_count,
+	.get_fb_location = mmhub_v1_0_get_fb_location,
+	.init = mmhub_v1_0_init,
+	.gart_enable = mmhub_v1_0_gart_enable,
+	.set_fault_enable_default = mmhub_v1_0_set_fault_enable_default,
+	.gart_disable = mmhub_v1_0_gart_disable,
+	.set_clockgating = mmhub_v1_0_set_clockgating,
+	.get_clockgating = mmhub_v1_0_get_clockgating,
+	.setup_vm_pt_regs = mmhub_v1_0_setup_vm_pt_regs,
+	.update_power_gating = mmhub_v1_0_update_power_gating,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
index c43319e8f945..d77f5b65a618 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
@@ -25,18 +25,4 @@
 
 extern const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs;
 
-u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev);
-int mmhub_v1_0_gart_enable(struct amdgpu_device *adev);
-void mmhub_v1_0_gart_disable(struct amdgpu_device *adev);
-void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev,
-					 bool value);
-void mmhub_v1_0_init(struct amdgpu_device *adev);
-int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev,
-			       enum amd_clockgating_state state);
-void mmhub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags);
-void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
-                                bool enable);
-void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
-				uint64_t page_table_base);
-
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
index c79fc54bc3c4..2063700f0bc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
@@ -36,7 +36,130 @@
 #define mmDAGB0_CNTL_MISC2_Sienna_Cichlid                       0x0070
 #define mmDAGB0_CNTL_MISC2_Sienna_Cichlid_BASE_IDX              0
 
-void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+static const char *mmhub_client_ids_navi1x[][2] = {
+	[3][0] = "DCEDMC",
+	[4][0] = "DCEVGA",
+	[5][0] = "MP0",
+	[6][0] = "MP1",
+	[13][0] = "VMC",
+	[14][0] = "HDP",
+	[15][0] = "OSS",
+	[16][0] = "VCNU",
+	[17][0] = "JPEG",
+	[18][0] = "VCN",
+	[3][1] = "DCEDMC",
+	[4][1] = "DCEXFC",
+	[5][1] = "DCEVGA",
+	[6][1] = "DCEDWB",
+	[7][1] = "MP0",
+	[8][1] = "MP1",
+	[9][1] = "DBGU1",
+	[10][1] = "DBGU0",
+	[11][1] = "XDP",
+	[14][1] = "HDP",
+	[15][1] = "OSS",
+	[16][1] = "VCNU",
+	[17][1] = "JPEG",
+	[18][1] = "VCN",
+};
+
+static const char *mmhub_client_ids_sienna_cichlid[][2] = {
+	[3][0] = "DCEDMC",
+	[4][0] = "DCEVGA",
+	[5][0] = "MP0",
+	[6][0] = "MP1",
+	[8][0] = "VMC",
+	[9][0] = "VCNU0",
+	[10][0] = "JPEG",
+	[12][0] = "VCNU1",
+	[13][0] = "VCN1",
+	[14][0] = "HDP",
+	[15][0] = "OSS",
+	[32+11][0] = "VCN0",
+	[0][1] = "DBGU0",
+	[1][1] = "DBGU1",
+	[2][1] = "DCEDWB",
+	[3][1] = "DCEDMC",
+	[4][1] = "DCEVGA",
+	[5][1] = "MP0",
+	[6][1] = "MP1",
+	[7][1] = "XDP",
+	[9][1] = "VCNU0",
+	[10][1] = "JPEG",
+	[11][1] = "VCN0",
+	[12][1] = "VCNU1",
+	[13][1] = "VCN1",
+	[14][1] = "HDP",
+	[15][1] = "OSS",
+};
+
+static uint32_t mmhub_v2_0_get_invalidate_req(unsigned int vmid,
+					      uint32_t flush_type)
+{
+	u32 req = 0;
+
+	/* invalidate using legacy mode on vmid*/
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
+	req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);
+
+	return req;
+}
+
+static void
+mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
+					     uint32_t status)
+{
+	uint32_t cid, rw;
+	const char *mmhub_cid = NULL;
+
+	cid = REG_GET_FIELD(status,
+			    MMVM_L2_PROTECTION_FAULT_STATUS, CID);
+	rw = REG_GET_FIELD(status,
+			   MMVM_L2_PROTECTION_FAULT_STATUS, RW);
+
+	dev_err(adev->dev,
+		"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
+		status);
+	switch (adev->asic_type) {
+	case CHIP_NAVI10:
+	case CHIP_NAVI12:
+	case CHIP_NAVI14:
+		mmhub_cid = mmhub_client_ids_navi1x[cid][rw];
+		break;
+	case CHIP_SIENNA_CICHLID:
+	case CHIP_NAVY_FLOUNDER:
+		mmhub_cid = mmhub_client_ids_sienna_cichlid[cid][rw];
+		break;
+	default:
+		mmhub_cid = NULL;
+		break;
+	}
+	dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+		mmhub_cid ? mmhub_cid : "unknown", cid);
+	dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		MMVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
+	dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		MMVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR));
+	dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
+		REG_GET_FIELD(status,
+		MMVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS));
+	dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
+		REG_GET_FIELD(status,
+		MMVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
+	dev_err(adev->dev, "\t RW: 0x%x\n", rw);
+}
+
+static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
@@ -78,11 +201,6 @@ static void mmhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev)
 	WREG32_SOC15(MMHUB, 0, mmMMMC_VM_AGP_BOT, 0x00FFFFFF);
 
 	if (!amdgpu_sriov_vf(adev)) {
-		/*
-		 * the new L1 policy will block SRIOV guest from writing
-		 * these regs, and they will be programed at host.
-		 * so skip programing these regs.
-		 */
 		/* Program the system aperture low logical page number. */
 		WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_LOW_ADDR,
 			     adev->gmc.vram_start >> 18);
@@ -251,7 +369,7 @@ static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT1_CNTL,
 				    i * hub->ctx_distance, tmp);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
@@ -280,7 +398,7 @@ static void mmhub_v2_0_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
-int mmhub_v2_0_gart_enable(struct amdgpu_device *adev)
+static int mmhub_v2_0_gart_enable(struct amdgpu_device *adev)
 {
 	/* GART Enable. */
 	mmhub_v2_0_init_gart_aperture_regs(adev);
@@ -296,7 +414,7 @@ int mmhub_v2_0_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-void mmhub_v2_0_gart_disable(struct amdgpu_device *adev)
+static void mmhub_v2_0_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
 	u32 tmp;
@@ -327,7 +445,7 @@ void mmhub_v2_0_gart_disable(struct amdgpu_device *adev)
  * @adev: amdgpu_device pointer
  * @value: true redirects VM faults to the default page
  */
-void mmhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
+static void mmhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 {
 	u32 tmp;
 
@@ -370,7 +488,12 @@ void mmhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 	WREG32_SOC15(MMHUB, 0, mmMMVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
-void mmhub_v2_0_init(struct amdgpu_device *adev)
+static const struct amdgpu_vmhub_funcs mmhub_v2_0_vmhub_funcs = {
+	.print_l2_protection_fault_status = mmhub_v2_0_print_l2_protection_fault_status,
+	.get_invalidate_req = mmhub_v2_0_get_invalidate_req,
+};
+
+static void mmhub_v2_0_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
 
@@ -400,6 +523,16 @@ void mmhub_v2_0_init(struct amdgpu_device *adev)
 		mmMMVM_INVALIDATE_ENG0_REQ;
 	hub->eng_addr_distance = mmMMVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
 		mmMMVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+	hub->vm_cntx_cntl_vm_fault = MMVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+		MMVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
+
+	hub->vmhub_funcs = &mmhub_v2_0_vmhub_funcs;
 }
 
 static void mmhub_v2_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
@@ -490,7 +623,7 @@ static void mmhub_v2_0_update_medium_grain_light_sleep(struct amdgpu_device *ade
 	}
 }
 
-int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev,
+static int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev,
 			       enum amd_clockgating_state state)
 {
 	if (amdgpu_sriov_vf(adev))
@@ -514,7 +647,7 @@ int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev,
 	return 0;
 }
 
-void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u32 *flags)
+static void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u32 *flags)
 {
 	int data, data1;
 
@@ -547,3 +680,14 @@ void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u32 *flags)
 	if (data & MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK)
 		*flags |= AMD_CG_SUPPORT_MC_LS;
 }
+
+const struct amdgpu_mmhub_funcs mmhub_v2_0_funcs = {
+	.ras_late_init = amdgpu_mmhub_ras_late_init,
+	.init = mmhub_v2_0_init,
+	.gart_enable = mmhub_v2_0_gart_enable,
+	.set_fault_enable_default = mmhub_v2_0_set_fault_enable_default,
+	.gart_disable = mmhub_v2_0_gart_disable,
+	.set_clockgating = mmhub_v2_0_set_clockgating,
+	.get_clockgating = mmhub_v2_0_get_clockgating,
+	.setup_vm_pt_regs = mmhub_v2_0_setup_vm_pt_regs,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
index 3ea4344f0315..f80f461d67da 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.h
@@ -23,15 +23,6 @@
 #ifndef __MMHUB_V2_0_H__
 #define __MMHUB_V2_0_H__
 
-int mmhub_v2_0_gart_enable(struct amdgpu_device *adev);
-void mmhub_v2_0_gart_disable(struct amdgpu_device *adev);
-void mmhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev,
-					 bool value);
-void mmhub_v2_0_init(struct amdgpu_device *adev);
-int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev,
-			       enum amd_clockgating_state state);
-void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u32 *flags);
-void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
-				uint64_t page_table_base);
+extern const struct amdgpu_mmhub_funcs mmhub_v2_0_funcs;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
index 9979f54fef57..66748bb01b52 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
@@ -36,7 +36,7 @@
 #define MMHUB_NUM_INSTANCES			2
 #define MMHUB_INSTANCE_REGISTER_OFFSET		0x3000
 
-u64 mmhub_v9_4_get_fb_location(struct amdgpu_device *adev)
+static u64 mmhub_v9_4_get_fb_location(struct amdgpu_device *adev)
 {
 	/* The base should be same b/t 2 mmhubs on Acrturus. Read one here. */
 	u64 base = RREG32_SOC15(MMHUB, 0, mmVMSHAREDVC0_MC_VM_FB_LOCATION_BASE);
@@ -97,7 +97,7 @@ static void mmhub_v9_4_init_gart_aperture_regs(struct amdgpu_device *adev,
 			    (u32)(adev->gmc.gart_end >> 44));
 }
 
-void mmhub_v9_4_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+static void mmhub_v9_4_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
 	int i;
@@ -330,7 +330,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !amdgpu_noretry);
+				    !adev->gmc.noretry);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVML2VC0_VM_CONTEXT1_CNTL,
 				    hubid * MMHUB_INSTANCE_REGISTER_OFFSET +
 				    i * hub->ctx_distance, tmp);
@@ -375,7 +375,7 @@ static void mmhub_v9_4_program_invalidation(struct amdgpu_device *adev,
 	}
 }
 
-int mmhub_v9_4_gart_enable(struct amdgpu_device *adev)
+static int mmhub_v9_4_gart_enable(struct amdgpu_device *adev)
 {
 	int i;
 
@@ -397,7 +397,7 @@ int mmhub_v9_4_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-void mmhub_v9_4_gart_disable(struct amdgpu_device *adev)
+static void mmhub_v9_4_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
 	u32 tmp;
@@ -442,7 +442,7 @@ void mmhub_v9_4_gart_disable(struct amdgpu_device *adev)
  * @adev: amdgpu_device pointer
  * @value: true redirects VM faults to the default page
  */
-void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool value)
+static void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 {
 	u32 tmp;
 	int i;
@@ -500,7 +500,7 @@ void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 	}
 }
 
-void mmhub_v9_4_init(struct amdgpu_device *adev)
+static void mmhub_v9_4_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub[MMHUB_NUM_INSTANCES] =
 		{&adev->vmhub[AMDGPU_MMHUB_0], &adev->vmhub[AMDGPU_MMHUB_1]};
@@ -630,7 +630,7 @@ static void mmhub_v9_4_update_medium_grain_light_sleep(struct amdgpu_device *ade
 	}
 }
 
-int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev,
+static int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev,
 			       enum amd_clockgating_state state)
 {
 	if (amdgpu_sriov_vf(adev))
@@ -650,7 +650,7 @@ int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev,
 	return 0;
 }
 
-void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u32 *flags)
+static void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u32 *flags)
 {
 	int data, data1;
 
@@ -1624,8 +1624,45 @@ static void mmhub_v9_4_reset_ras_error_count(struct amdgpu_device *adev)
 	}
 }
 
+static const struct soc15_reg_entry mmhub_v9_4_err_status_regs[] = {
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA0_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA1_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA2_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA3_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA4_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA5_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA6_ERR_STATUS), 0, 0, 0 },
+	{ SOC15_REG_ENTRY(MMHUB, 0, mmMMEA7_ERR_STATUS), 0, 0, 0 },
+};
+
+static void mmhub_v9_4_query_ras_error_status(struct amdgpu_device *adev)
+{
+	int i;
+	uint32_t reg_value;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB))
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(mmhub_v9_4_err_status_regs); i++) {
+		reg_value =
+			RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v9_4_err_status_regs[i]));
+		if (reg_value)
+			dev_warn(adev->dev, "MMHUB EA err detected at instance: %d, status: 0x%x!\n",
+					i, reg_value);
+	}
+}
+
 const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs = {
 	.ras_late_init = amdgpu_mmhub_ras_late_init,
 	.query_ras_error_count = mmhub_v9_4_query_ras_error_count,
 	.reset_ras_error_count = mmhub_v9_4_reset_ras_error_count,
+	.get_fb_location = mmhub_v9_4_get_fb_location,
+	.init = mmhub_v9_4_init,
+	.gart_enable = mmhub_v9_4_gart_enable,
+	.set_fault_enable_default = mmhub_v9_4_set_fault_enable_default,
+	.gart_disable = mmhub_v9_4_gart_disable,
+	.set_clockgating = mmhub_v9_4_set_clockgating,
+	.get_clockgating = mmhub_v9_4_get_clockgating,
+	.setup_vm_pt_regs = mmhub_v9_4_setup_vm_pt_regs,
+	.query_ras_error_status = mmhub_v9_4_query_ras_error_status,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h
index 1b979773776c..92404a8f66f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h
@@ -25,16 +25,4 @@
 
 extern const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs;
 
-u64 mmhub_v9_4_get_fb_location(struct amdgpu_device *adev);
-int mmhub_v9_4_gart_enable(struct amdgpu_device *adev);
-void mmhub_v9_4_gart_disable(struct amdgpu_device *adev);
-void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev,
-					 bool value);
-void mmhub_v9_4_init(struct amdgpu_device *adev);
-int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev,
-			       enum amd_clockgating_state state);
-void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u32 *flags);
-void mmhub_v9_4_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
-				uint64_t page_table_base);
-
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index 5fd67e1cc2a0..f5ce9a9f4cf5 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -238,19 +238,15 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
 	int timeout = AI_MAILBOX_POLL_FLR_TIMEDOUT;
-	int locked;
 
 	/* block amdgpu_gpu_recover till msg FLR COMPLETE received,
 	 * otherwise the mailbox msg will be ruined/reseted by
 	 * the VF FLR.
-	 *
-	 * we can unlock the lock_reset to allow "amdgpu_job_timedout"
-	 * to run gpu_recover() after FLR_NOTIFICATION_CMPL received
-	 * which means host side had finished this VF's FLR.
 	 */
-	locked = mutex_trylock(&adev->lock_reset);
-	if (locked)
-		adev->in_gpu_reset = true;
+	if (!down_read_trylock(&adev->reset_sem))
+		return;
+
+	atomic_set(&adev->in_gpu_reset, 1);
 
 	do {
 		if (xgpu_ai_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL)
@@ -261,14 +257,13 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 	} while (timeout > 1);
 
 flr_done:
-	if (locked) {
-		adev->in_gpu_reset = false;
-		mutex_unlock(&adev->lock_reset);
-	}
+	atomic_set(&adev->in_gpu_reset, 0);
+	up_read(&adev->reset_sem);
 
 	/* Trigger recovery for world switch failure if no TDR */
 	if (amdgpu_device_should_recover_gpu(adev)
-		&& adev->sdma_timeout == MAX_SCHEDULE_TIMEOUT)
+		&& (!amdgpu_device_has_job_running(adev) ||
+		adev->sdma_timeout == MAX_SCHEDULE_TIMEOUT))
 		amdgpu_device_gpu_recover(adev, NULL);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
index ce2bf1fb79ed..666ed99cc14b 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
@@ -259,19 +259,15 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
 	int timeout = NV_MAILBOX_POLL_FLR_TIMEDOUT;
-	int locked;
 
 	/* block amdgpu_gpu_recover till msg FLR COMPLETE received,
 	 * otherwise the mailbox msg will be ruined/reseted by
 	 * the VF FLR.
-	 *
-	 * we can unlock the lock_reset to allow "amdgpu_job_timedout"
-	 * to run gpu_recover() after FLR_NOTIFICATION_CMPL received
-	 * which means host side had finished this VF's FLR.
 	 */
-	locked = mutex_trylock(&adev->lock_reset);
-	if (locked)
-		adev->in_gpu_reset = true;
+	if (!down_read_trylock(&adev->reset_sem))
+		return;
+
+	atomic_set(&adev->in_gpu_reset, 1);
 
 	do {
 		if (xgpu_nv_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL)
@@ -282,14 +278,13 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 	} while (timeout > 1);
 
 flr_done:
-	if (locked) {
-		adev->in_gpu_reset = false;
-		mutex_unlock(&adev->lock_reset);
-	}
+	atomic_set(&adev->in_gpu_reset, 0);
+	up_read(&adev->reset_sem);
 
 	/* Trigger recovery for world switch failure if no TDR */
 	if (amdgpu_device_should_recover_gpu(adev)
-		&& (adev->sdma_timeout == MAX_SCHEDULE_TIMEOUT ||
+		&& (!amdgpu_device_has_job_running(adev) ||
+		adev->sdma_timeout == MAX_SCHEDULE_TIMEOUT ||
 		adev->gfx_timeout == MAX_SCHEDULE_TIMEOUT ||
 		adev->compute_timeout == MAX_SCHEDULE_TIMEOUT ||
 		adev->video_timeout == MAX_SCHEDULE_TIMEOUT))
diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
index 350f1bf063c6..74b1e7dc49a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
@@ -306,7 +306,8 @@ static int navi10_ih_irq_init(struct amdgpu_device *adev)
 	} else {
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl);
 	}
-	navi10_ih_reroute_ih(adev);
+	if (adev->irq.ih1.ring_size)
+		navi10_ih_reroute_ih(adev);
 
 	if (unlikely(adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT)) {
 		if (ih->use_bus_addr) {
@@ -668,19 +669,26 @@ static int navi10_ih_sw_init(void *handle)
 	adev->irq.ih.use_doorbell = true;
 	adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true);
-	if (r)
-		return r;
+	adev->irq.ih1.ring_size = 0;
+	adev->irq.ih2.ring_size = 0;
 
-	adev->irq.ih1.use_doorbell = true;
-	adev->irq.ih1.doorbell_index = (adev->doorbell_index.ih + 1) << 1;
+	if (adev->asic_type < CHIP_NAVI10) {
+		r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true);
+		if (r)
+			return r;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih2, PAGE_SIZE, true);
-	if (r)
-		return r;
+		adev->irq.ih1.use_doorbell = true;
+		adev->irq.ih1.doorbell_index =
+					(adev->doorbell_index.ih + 1) << 1;
+
+		r = amdgpu_ih_ring_init(adev, &adev->irq.ih2, PAGE_SIZE, true);
+		if (r)
+			return r;
 
-	adev->irq.ih2.use_doorbell = true;
-	adev->irq.ih2.doorbell_index = (adev->doorbell_index.ih + 2) << 1;
+		adev->irq.ih2.use_doorbell = true;
+		adev->irq.ih2.doorbell_index =
+					(adev->doorbell_index.ih + 2) << 1;
+	}
 
 	r = amdgpu_irq_init(adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index e629156173d3..eadc9526d33f 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -302,6 +302,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device
 	uint32_t bif_doorbell_intr_cntl;
 	struct ras_manager *obj = amdgpu_ras_find_obj(adev, adev->nbio.ras_if);
 	struct ras_err_data err_data = {0, 0, 0, NULL};
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL);
 	if (REG_GET_FIELD(bif_doorbell_intr_cntl,
@@ -312,28 +313,31 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device
 						RAS_CNTLR_INTERRUPT_CLEAR, 1);
 		WREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL, bif_doorbell_intr_cntl);
 
-		/*
-		 * clear error status after ras_controller_intr according to
-		 * hw team and count ue number for query
-		 */
-		nbio_v7_4_query_ras_error_count(adev, &err_data);
-
-		/* logging on error counter and printing for awareness */
-		obj->err_data.ue_count += err_data.ue_count;
-		obj->err_data.ce_count += err_data.ce_count;
-
-		if (err_data.ce_count)
-			dev_info(adev->dev, "%ld correctable hardware "
-					"errors detected in %s block, "
-					"no user action is needed.\n",
-					obj->err_data.ce_count,
-					adev->nbio.ras_if->name);
-
-		if (err_data.ue_count)
-			dev_info(adev->dev, "%ld uncorrectable hardware "
-					"errors detected in %s block\n",
-					obj->err_data.ue_count,
-					adev->nbio.ras_if->name);
+		if (!ras->disable_ras_err_cnt_harvest) {
+			/*
+			 * clear error status after ras_controller_intr
+			 * according to hw team and count ue number
+			 * for query
+			 */
+			nbio_v7_4_query_ras_error_count(adev, &err_data);
+
+			/* logging on error cnt and printing for awareness */
+			obj->err_data.ue_count += err_data.ue_count;
+			obj->err_data.ce_count += err_data.ce_count;
+
+			if (err_data.ce_count)
+				dev_info(adev->dev, "%ld correctable hardware "
+						"errors detected in %s block, "
+						"no user action is needed.\n",
+						obj->err_data.ce_count,
+						adev->nbio.ras_if->name);
+
+			if (err_data.ue_count)
+				dev_info(adev->dev, "%ld uncorrectable hardware "
+						"errors detected in %s block\n",
+						obj->err_data.ue_count,
+						adev->nbio.ras_if->name);
+		}
 
 		dev_info(adev->dev, "RAS controller interrupt triggered "
 					"by NBIF error\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index ca11253e787c..1ce741a0c6a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -69,75 +69,40 @@ static const struct amd_ip_funcs nv_common_ip_funcs;
  */
 static u32 nv_pcie_rreg(struct amdgpu_device *adev, u32 reg)
 {
-	unsigned long flags, address, data;
-	u32 r;
+	unsigned long address, data;
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	WREG32(address, reg);
-	(void)RREG32(address);
-	r = RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
-	return r;
+	return amdgpu_device_indirect_rreg(adev, address, data, reg);
 }
 
 static void nv_pcie_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
 {
-	unsigned long flags, address, data;
+	unsigned long address, data;
 
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	WREG32(address, reg);
-	(void)RREG32(address);
-	WREG32(data, v);
-	(void)RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+	amdgpu_device_indirect_wreg(adev, address, data, reg, v);
 }
 
 static u64 nv_pcie_rreg64(struct amdgpu_device *adev, u32 reg)
 {
-	unsigned long flags, address, data;
-	u64 r;
+	unsigned long address, data;
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	/* read low 32 bit */
-	WREG32(address, reg);
-	(void)RREG32(address);
-	r = RREG32(data);
-
-	/* read high 32 bit*/
-	WREG32(address, reg + 4);
-	(void)RREG32(address);
-	r |= ((u64)RREG32(data) << 32);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
-	return r;
+	return amdgpu_device_indirect_rreg64(adev, address, data, reg);
 }
 
 static void nv_pcie_wreg64(struct amdgpu_device *adev, u32 reg, u64 v)
 {
-	unsigned long flags, address, data;
+	unsigned long address, data;
 
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	/* write low 32 bit */
-	WREG32(address, reg);
-	(void)RREG32(address);
-	WREG32(data, (u32)(v & 0xffffffffULL));
-	(void)RREG32(data);
-
-	/* write high 32 bit */
-	WREG32(address, reg + 4);
-	(void)RREG32(address);
-	WREG32(data, (u32)(v >> 32));
-	(void)RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+	amdgpu_device_indirect_wreg64(adev, address, data, reg, v);
 }
 
 static u32 nv_didt_rreg(struct amdgpu_device *adev, u32 reg)
@@ -311,7 +276,7 @@ static int nv_asic_mode1_reset(struct amdgpu_device *adev)
 	/* disable BM */
 	pci_clear_master(adev->pdev);
 
-	pci_save_state(adev->pdev);
+	amdgpu_device_cache_pci_state(adev->pdev);
 
 	if (amdgpu_dpm_is_mode1_reset_supported(adev)) {
 		dev_info(adev->dev, "GPU smu mode1 reset\n");
@@ -323,7 +288,7 @@ static int nv_asic_mode1_reset(struct amdgpu_device *adev)
 
 	if (ret)
 		dev_err(adev->dev, "GPU mode1 reset failed\n");
-	pci_restore_state(adev->pdev);
+	amdgpu_device_load_pci_state(adev->pdev);
 
 	/* wait for asic to come out of reset */
 	for (i = 0; i < adev->usec_timeout; i++) {
@@ -380,7 +345,7 @@ static int nv_asic_reset(struct amdgpu_device *adev)
 	struct smu_context *smu = &adev->smu;
 
 	if (nv_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
-		dev_info(adev->dev, "GPU BACO reset\n");
+		dev_info(adev->dev, "BACO reset\n");
 
 		ret = smu_baco_enter(smu);
 		if (ret)
@@ -388,8 +353,10 @@ static int nv_asic_reset(struct amdgpu_device *adev)
 		ret = smu_baco_exit(smu);
 		if (ret)
 			return ret;
-	} else
+	} else {
+		dev_info(adev->dev, "MODE1 reset\n");
 		ret = nv_asic_mode1_reset(adev);
+	}
 
 	return ret;
 }
@@ -619,7 +586,7 @@ static void nv_invalidate_hdp(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
 	if (!ring || !ring->funcs->emit_wreg) {
-		WREG32_SOC15_NO_KIQ(NBIO, 0, mmHDP_READ_CACHE_INVALIDATE, 1);
+		WREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE, 1);
 	} else {
 		amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
 					HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1);
@@ -690,6 +657,10 @@ static void nv_init_doorbell_index(struct amdgpu_device *adev)
 	adev->doorbell_index.sdma_doorbell_range = 20;
 }
 
+static void nv_pre_asic_init(struct amdgpu_device *adev)
+{
+}
+
 static const struct amdgpu_asic_funcs nv_asic_funcs =
 {
 	.read_disabled_bios = &nv_read_disabled_bios,
@@ -709,6 +680,7 @@ static const struct amdgpu_asic_funcs nv_asic_funcs =
 	.need_reset_on_init = &nv_need_reset_on_init,
 	.get_pcie_replay_count = &nv_get_pcie_replay_count,
 	.supports_baco = &nv_asic_supports_baco,
+	.pre_asic_init = &nv_pre_asic_init,
 };
 
 static int nv_common_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
index cbc04a5c0fe1..1ef2f5b1d828 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
@@ -83,19 +83,6 @@ struct psp_gfx_ctrl
 */
 #define GFX_FLAG_RESPONSE               0x80000000
 
-/* Gbr IH registers ID */
-enum ih_reg_id {
-	IH_RB		= 0,		// IH_RB_CNTL
-	IH_RB_RNG1	= 1,		// IH_RB_CNTL_RING1
-	IH_RB_RNG2	= 2,		// IH_RB_CNTL_RING2
-};
-
-/* Command to setup Gibraltar IH register */
-struct psp_gfx_cmd_gbr_ih_reg {
-	uint32_t		reg_value;	/* Value to be set to the IH_RB_CNTL... register*/
-	enum ih_reg_id		reg_id;		/* ID of the register */
-};
-
 /* TEE Gfx Command IDs for the ring buffer interface. */
 enum psp_gfx_cmd_id
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
index 6c9614f77d33..75489313dbad 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
@@ -38,6 +38,8 @@
 #include "oss/osssys_4_0_sh_mask.h"
 
 MODULE_FIRMWARE("amdgpu/renoir_asd.bin");
+MODULE_FIRMWARE("amdgpu/renoir_ta.bin");
+
 /* address block */
 #define smnMP1_FIRMWARE_FLAGS		0x3010024
 
@@ -45,7 +47,10 @@ static int psp_v12_0_init_microcode(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
 	const char *chip_name;
+	char fw_name[30];
 	int err = 0;
+	const struct ta_firmware_header_v1_0 *ta_hdr;
+	DRM_DEBUG("\n");
 
 	switch (adev->asic_type) {
 	case CHIP_RENOIR:
@@ -56,6 +61,55 @@ static int psp_v12_0_init_microcode(struct psp_context *psp)
 	}
 
 	err = psp_init_asd_microcode(psp, chip_name);
+	if (err)
+		goto out;
+
+	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name);
+	err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev);
+	if (err) {
+		release_firmware(adev->psp.ta_fw);
+		adev->psp.ta_fw = NULL;
+		dev_info(adev->dev,
+			 "psp v12.0: Failed to load firmware \"%s\"\n",
+			 fw_name);
+	} else {
+		err = amdgpu_ucode_validate(adev->psp.ta_fw);
+		if (err)
+			goto out2;
+
+		ta_hdr = (const struct ta_firmware_header_v1_0 *)
+				 adev->psp.ta_fw->data;
+		adev->psp.ta_hdcp_ucode_version =
+			le32_to_cpu(ta_hdr->ta_hdcp_ucode_version);
+		adev->psp.ta_hdcp_ucode_size =
+			le32_to_cpu(ta_hdr->ta_hdcp_size_bytes);
+		adev->psp.ta_hdcp_start_addr =
+			(uint8_t *)ta_hdr +
+			le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes);
+
+		adev->psp.ta_fw_version = le32_to_cpu(ta_hdr->header.ucode_version);
+
+		adev->psp.ta_dtm_ucode_version =
+			le32_to_cpu(ta_hdr->ta_dtm_ucode_version);
+		adev->psp.ta_dtm_ucode_size =
+			le32_to_cpu(ta_hdr->ta_dtm_size_bytes);
+		adev->psp.ta_dtm_start_addr =
+			(uint8_t *)adev->psp.ta_hdcp_start_addr +
+			le32_to_cpu(ta_hdr->ta_dtm_offset_bytes);
+	}
+
+	return 0;
+
+out2:
+	release_firmware(adev->psp.ta_fw);
+	adev->psp.ta_fw = NULL;
+out:
+	if (err) {
+		dev_err(adev->dev,
+			"psp v12.0: Failed to load firmware \"%s\"\n",
+			fw_name);
+	}
+
 	return err;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 856c50386c86..86fb1eddf5a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -592,6 +592,9 @@ static int sdma_v4_0_init_microcode(struct amdgpu_device *adev)
 	struct amdgpu_firmware_info *info = NULL;
 	const struct common_firmware_header *header = NULL;
 
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	DRM_DEBUG("\n");
 
 	switch (adev->asic_type) {
@@ -1000,7 +1003,7 @@ static void sdma_v4_0_page_stop(struct amdgpu_device *adev)
 		sdma[i] = &adev->sdma.instance[i].page;
 
 		if ((adev->mman.buffer_funcs_ring == sdma[i]) &&
-			(unset == false)) {
+			(!unset)) {
 			amdgpu_ttm_set_buffer_funcs_status(adev, false);
 			unset = true;
 		}
@@ -1063,6 +1066,15 @@ static void sdma_v4_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
 			WREG32_SDMA(i, mmSDMA0_PHASE2_QUANTUM, phase_quantum);
 		}
 		WREG32_SDMA(i, mmSDMA0_CNTL, f32_cntl);
+
+		/*
+		 * Enable SDMA utilization. Its only supported on
+		 * Arcturus for the moment and firmware version 14
+		 * and above.
+		 */
+		if (adev->asic_type == CHIP_ARCTURUS &&
+		    adev->sdma.instance[i].fw_version >= 14)
+			WREG32_SDMA(i, mmSDMA0_PUB_DUMMY_REG2, enable);
 	}
 
 }
@@ -1080,7 +1092,7 @@ static void sdma_v4_0_enable(struct amdgpu_device *adev, bool enable)
 	u32 f32_cntl;
 	int i;
 
-	if (enable == false) {
+	if (!enable) {
 		sdma_v4_0_gfx_stop(adev);
 		sdma_v4_0_rlc_stop(adev);
 		if (adev->sdma.has_page_queue)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index e2232dd12d8e..9c72b95b7463 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -203,6 +203,9 @@ static int sdma_v5_0_init_microcode(struct amdgpu_device *adev)
 	const struct common_firmware_header *header = NULL;
 	const struct sdma_firmware_header_v1_0 *hdr;
 
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	DRM_DEBUG("\n");
 
 	switch (adev->asic_type) {
@@ -616,7 +619,7 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable)
 	u32 f32_cntl;
 	int i;
 
-	if (enable == false) {
+	if (!enable) {
 		sdma_v5_0_gfx_stop(adev);
 		sdma_v5_0_rlc_stop(adev);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index 46a9617fee5f..9f3952723c63 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -148,6 +148,9 @@ static int sdma_v5_2_init_microcode(struct amdgpu_device *adev)
 	struct amdgpu_firmware_info *info = NULL;
 	const struct common_firmware_header *header = NULL;
 
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	DRM_DEBUG("\n");
 
 	switch (adev->asic_type) {
@@ -559,7 +562,7 @@ static void sdma_v5_2_enable(struct amdgpu_device *adev, bool enable)
 	u32 f32_cntl;
 	int i;
 
-	if (enable == false) {
+	if (!enable) {
 		sdma_v5_2_gfx_stop(adev);
 		sdma_v5_2_rlc_stop(adev);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c
index 1b449291f068..e5e336fd9e94 100644
--- a/drivers/gpu/drm/amd/amdgpu/si.c
+++ b/drivers/gpu/drm/amd/amdgpu/si.c
@@ -52,6 +52,8 @@
 #include "bif/bif_3_0_d.h"
 #include "bif/bif_3_0_sh_mask.h"
 
+#include "amdgpu_dm.h"
+
 static const u32 tahiti_golden_registers[] =
 {
 	mmAZALIA_SCLK_CONTROL, 0x00000030, 0x00000011,
@@ -1215,10 +1217,100 @@ static bool si_read_bios_from_rom(struct amdgpu_device *adev,
 	return true;
 }
 
-//xxx: not implemented
+static void si_set_clk_bypass_mode(struct amdgpu_device *adev)
+{
+	u32 tmp, i;
+
+	tmp = RREG32(CG_SPLL_FUNC_CNTL);
+	tmp |= SPLL_BYPASS_EN;
+	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+
+	tmp = RREG32(CG_SPLL_FUNC_CNTL_2);
+	tmp |= SPLL_CTLREQ_CHG;
+	WREG32(CG_SPLL_FUNC_CNTL_2, tmp);
+
+	for (i = 0; i < adev->usec_timeout; i++) {
+		if (RREG32(SPLL_STATUS) & SPLL_CHG_STATUS)
+			break;
+		udelay(1);
+	}
+
+	tmp = RREG32(CG_SPLL_FUNC_CNTL_2);
+	tmp &= ~(SPLL_CTLREQ_CHG | SCLK_MUX_UPDATE);
+	WREG32(CG_SPLL_FUNC_CNTL_2, tmp);
+
+	tmp = RREG32(MPLL_CNTL_MODE);
+	tmp &= ~MPLL_MCLK_SEL;
+	WREG32(MPLL_CNTL_MODE, tmp);
+}
+
+static void si_spll_powerdown(struct amdgpu_device *adev)
+{
+	u32 tmp;
+
+	tmp = RREG32(SPLL_CNTL_MODE);
+	tmp |= SPLL_SW_DIR_CONTROL;
+	WREG32(SPLL_CNTL_MODE, tmp);
+
+	tmp = RREG32(CG_SPLL_FUNC_CNTL);
+	tmp |= SPLL_RESET;
+	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+
+	tmp = RREG32(CG_SPLL_FUNC_CNTL);
+	tmp |= SPLL_SLEEP;
+	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+
+	tmp = RREG32(SPLL_CNTL_MODE);
+	tmp &= ~SPLL_SW_DIR_CONTROL;
+	WREG32(SPLL_CNTL_MODE, tmp);
+}
+
+static int si_gpu_pci_config_reset(struct amdgpu_device *adev)
+{
+	u32 i;
+	int r = -EINVAL;
+
+	dev_info(adev->dev, "GPU pci config reset\n");
+
+	/* set mclk/sclk to bypass */
+	si_set_clk_bypass_mode(adev);
+	/* powerdown spll */
+	si_spll_powerdown(adev);
+	/* disable BM */
+	pci_clear_master(adev->pdev);
+	/* reset */
+	amdgpu_device_pci_config_reset(adev);
+
+	udelay(100);
+
+	/* wait for asic to come out of reset */
+	for (i = 0; i < adev->usec_timeout; i++) {
+		if (RREG32(mmCONFIG_MEMSIZE) != 0xffffffff) {
+			/* enable BM */
+			pci_set_master(adev->pdev);
+			adev->has_hw_reset = true;
+			r = 0;
+			break;
+		}
+		udelay(1);
+	}
+
+	return r;
+}
+
 static int si_asic_reset(struct amdgpu_device *adev)
 {
-	return 0;
+	int r;
+
+	dev_info(adev->dev, "PCI CONFIG reset\n");
+
+	amdgpu_atombios_scratch_regs_engine_hung(adev, true);
+
+	r = si_gpu_pci_config_reset(adev);
+
+	amdgpu_atombios_scratch_regs_engine_hung(adev, false);
+
+	return r;
 }
 
 static bool si_asic_supports_baco(struct amdgpu_device *adev)
@@ -1247,7 +1339,7 @@ static void si_vga_set_state(struct amdgpu_device *adev, bool state)
 	uint32_t temp;
 
 	temp = RREG32(CONFIG_CNTL);
-	if (state == false) {
+	if (!state) {
 		temp &= ~(1<<0);
 		temp |= (1<<1);
 	} else {
@@ -1779,6 +1871,10 @@ static int si_set_vce_clocks(struct amdgpu_device *adev, u32 evclk, u32 ecclk)
 	return 0;
 }
 
+static void si_pre_asic_init(struct amdgpu_device *adev)
+{
+}
+
 static const struct amdgpu_asic_funcs si_asic_funcs =
 {
 	.read_disabled_bios = &si_read_disabled_bios,
@@ -1800,6 +1896,7 @@ static const struct amdgpu_asic_funcs si_asic_funcs =
 	.need_reset_on_init = &si_need_reset_on_init,
 	.get_pcie_replay_count = &si_get_pcie_replay_count,
 	.supports_baco = &si_asic_supports_baco,
+	.pre_asic_init = &si_pre_asic_init,
 };
 
 static uint32_t si_get_rev_id(struct amdgpu_device *adev)
@@ -2546,6 +2643,10 @@ int si_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &si_smu_ip_block);
 		if (adev->enable_virtual_display)
 			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
+#if defined(CONFIG_DRM_AMD_DC) && defined(CONFIG_DRM_AMD_DC_SI)
+		else if (amdgpu_device_has_dc_support(adev))
+			amdgpu_device_ip_block_add(adev, &dm_ip_block);
+#endif
 		else
 			amdgpu_device_ip_block_add(adev, &dce_v6_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &uvd_v3_1_ip_block);
@@ -2560,6 +2661,10 @@ int si_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &si_smu_ip_block);
 		if (adev->enable_virtual_display)
 			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
+#if defined(CONFIG_DRM_AMD_DC) && defined(CONFIG_DRM_AMD_DC_SI)
+		else if (amdgpu_device_has_dc_support(adev))
+			amdgpu_device_ip_block_add(adev, &dm_ip_block);
+#endif
 		else
 			amdgpu_device_ip_block_add(adev, &dce_v6_4_ip_block);
 		amdgpu_device_ip_block_add(adev, &uvd_v3_1_ip_block);
diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
index d55bf64770c4..7fb240c4990c 100644
--- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
+++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
@@ -508,14 +508,9 @@ static bool smu_v11_0_i2c_bus_lock(struct i2c_adapter *control)
 	struct amdgpu_device *adev = to_amdgpu_device(control);
 
 	/* Send  PPSMC_MSG_RequestI2CBus */
-	if (!adev->powerplay.pp_funcs->smu_i2c_bus_access)
-		goto Fail;
-
-
-	if (!adev->powerplay.pp_funcs->smu_i2c_bus_access(adev->powerplay.pp_handle, true))
+	if (!amdgpu_dpm_smu_i2c_bus_access(adev, true))
 		return true;
 
-Fail:
 	return false;
 }
 
@@ -523,16 +518,10 @@ static bool smu_v11_0_i2c_bus_unlock(struct i2c_adapter *control)
 {
 	struct amdgpu_device *adev = to_amdgpu_device(control);
 
-	/* Send  PPSMC_MSG_RequestI2CBus */
-	if (!adev->powerplay.pp_funcs->smu_i2c_bus_access)
-		goto Fail;
-
 	/* Send  PPSMC_MSG_ReleaseI2CBus */
-	if (!adev->powerplay.pp_funcs->smu_i2c_bus_access(adev->powerplay.pp_handle,
-							     false))
+	if (!amdgpu_dpm_smu_i2c_bus_access(adev, false))
 		return true;
 
-Fail:
 	return false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index c28ebf41530a..afcccc6c0fc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -101,75 +101,40 @@
  */
 static u32 soc15_pcie_rreg(struct amdgpu_device *adev, u32 reg)
 {
-	unsigned long flags, address, data;
-	u32 r;
+	unsigned long address, data;
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	WREG32(address, reg);
-	(void)RREG32(address);
-	r = RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
-	return r;
+	return amdgpu_device_indirect_rreg(adev, address, data, reg);
 }
 
 static void soc15_pcie_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
 {
-	unsigned long flags, address, data;
+	unsigned long address, data;
 
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	WREG32(address, reg);
-	(void)RREG32(address);
-	WREG32(data, v);
-	(void)RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+	amdgpu_device_indirect_wreg(adev, address, data, reg, v);
 }
 
 static u64 soc15_pcie_rreg64(struct amdgpu_device *adev, u32 reg)
 {
-	unsigned long flags, address, data;
-	u64 r;
+	unsigned long address, data;
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	/* read low 32 bit */
-	WREG32(address, reg);
-	(void)RREG32(address);
-	r = RREG32(data);
-
-	/* read high 32 bit*/
-	WREG32(address, reg + 4);
-	(void)RREG32(address);
-	r |= ((u64)RREG32(data) << 32);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
-	return r;
+	return amdgpu_device_indirect_rreg64(adev, address, data, reg);
 }
 
 static void soc15_pcie_wreg64(struct amdgpu_device *adev, u32 reg, u64 v)
 {
-	unsigned long flags, address, data;
+	unsigned long address, data;
 
 	address = adev->nbio.funcs->get_pcie_index_offset(adev);
 	data = adev->nbio.funcs->get_pcie_data_offset(adev);
 
-	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
-	/* write low 32 bit */
-	WREG32(address, reg);
-	(void)RREG32(address);
-	WREG32(data, (u32)(v & 0xffffffffULL));
-	(void)RREG32(data);
-
-	/* write high 32 bit */
-	WREG32(address, reg + 4);
-	(void)RREG32(address);
-	WREG32(data, (u32)(v >> 32));
-	(void)RREG32(data);
-	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+	amdgpu_device_indirect_wreg64(adev, address, data, reg, v);
 }
 
 static u32 soc15_uvd_ctx_rreg(struct amdgpu_device *adev, u32 reg)
@@ -484,13 +449,13 @@ static int soc15_asic_mode1_reset(struct amdgpu_device *adev)
 	/* disable BM */
 	pci_clear_master(adev->pdev);
 
-	pci_save_state(adev->pdev);
+	amdgpu_device_cache_pci_state(adev->pdev);
 
 	ret = psp_gpu_reset(adev);
 	if (ret)
 		dev_err(adev->dev, "GPU mode1 reset failed\n");
 
-	pci_restore_state(adev->pdev);
+	amdgpu_device_load_pci_state(adev->pdev);
 
 	/* wait for asic to come out of reset */
 	for (i = 0; i < adev->usec_timeout; i++) {
@@ -580,10 +545,13 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
 
 	switch (soc15_asic_reset_method(adev)) {
 		case AMD_RESET_METHOD_BACO:
+			dev_info(adev->dev, "BACO reset\n");
 			return soc15_asic_baco_reset(adev);
 		case AMD_RESET_METHOD_MODE2:
+			dev_info(adev->dev, "MODE2 reset\n");
 			return amdgpu_dpm_mode2_reset(adev);
 		default:
+			dev_info(adev->dev, "MODE1 reset\n");
 			return soc15_asic_mode1_reset(adev);
 	}
 }
@@ -1026,6 +994,11 @@ static uint64_t soc15_get_pcie_replay_count(struct amdgpu_device *adev)
 	return (nak_r + nak_g);
 }
 
+static void soc15_pre_asic_init(struct amdgpu_device *adev)
+{
+	gmc_v9_0_restore_registers(adev);
+}
+
 static const struct amdgpu_asic_funcs soc15_asic_funcs =
 {
 	.read_disabled_bios = &soc15_read_disabled_bios,
@@ -1046,6 +1019,7 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs =
 	.need_reset_on_init = &soc15_need_reset_on_init,
 	.get_pcie_replay_count = &soc15_get_pcie_replay_count,
 	.supports_baco = &soc15_supports_baco,
+	.pre_asic_init = &soc15_pre_asic_init,
 };
 
 static const struct amdgpu_asic_funcs vega20_asic_funcs =
@@ -1069,6 +1043,7 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs =
 	.need_reset_on_init = &soc15_need_reset_on_init,
 	.get_pcie_replay_count = &soc15_get_pcie_replay_count,
 	.supports_baco = &soc15_supports_baco,
+	.pre_asic_init = &soc15_pre_asic_init,
 };
 
 static int soc15_common_early_init(void *handle)
@@ -1449,7 +1424,8 @@ static void soc15_update_hdp_light_sleep(struct amdgpu_device *adev, bool enable
 	uint32_t def, data;
 
 	if (adev->asic_type == CHIP_VEGA20 ||
-		adev->asic_type == CHIP_ARCTURUS) {
+		adev->asic_type == CHIP_ARCTURUS ||
+		adev->asic_type == CHIP_RENOIR) {
 		def = data = RREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_CTRL));
 
 		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS))
diff --git a/drivers/gpu/drm/amd/amdgpu/ta_rap_if.h b/drivers/gpu/drm/amd/amdgpu/ta_rap_if.h
new file mode 100644
index 000000000000..f14833fae07c
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/ta_rap_if.h
@@ -0,0 +1,84 @@
+/*
+ * Copyright 2020 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"),
+ * 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.
+ *
+ */
+
+#ifndef _TA_RAP_IF_H
+#define _TA_RAP_IF_H
+
+/* Responses have bit 31 set */
+#define RSP_ID_MASK (1U << 31)
+#define RSP_ID(cmdId) (((uint32_t)(cmdId)) | RSP_ID_MASK)
+
+enum ta_rap_status {
+	TA_RAP_STATUS__SUCCESS                              = 1,
+	TA_RAP_STATUS__ERROR_GENERIC_FAILURE                = 2,
+	TA_RAP_STATUS__ERROR_CMD_NOT_SUPPORTED              = 3,
+	TA_RAP_STATUS__ERROR_INVALID_VALIDATION_METHOD      = 4,
+	TA_RAP_STATUS__ERROR_NULL_POINTER                   = 5,
+	TA_RAP_STATUS__ERROR_NOT_INITIALIZED                = 6,
+	TA_RAP_STATUS__ERROR_VALIDATION_FAILED              = 7,
+	TA_RAP_STATUS__ERROR_ASIC_NOT_SUPPORTED             = 8,
+	TA_RAP_STATUS__ERROR_OPERATION_NOT_PERMISSABLE      = 9,
+	TA_RAP_STATUS__ERROR_ALREADY_INIT                   = 10,
+};
+
+enum ta_rap_cmd {
+	TA_CMD_RAP__INITIALIZE              = 1,
+	TA_CMD_RAP__VALIDATE_L0             = 2,
+};
+
+enum ta_rap_validation_method {
+	METHOD_A           = 1,
+};
+
+struct ta_rap_cmd_input_data {
+	uint8_t reserved[8];
+};
+
+struct ta_rap_cmd_output_data {
+	uint32_t    last_subsection;
+	uint32_t    num_total_validate;
+	uint32_t    num_valid;
+	uint32_t    last_validate_addr;
+	uint32_t    last_validate_val;
+	uint32_t    last_validate_val_exptd;
+};
+
+union ta_rap_cmd_input {
+	struct ta_rap_cmd_input_data input;
+};
+
+union ta_rap_cmd_output {
+	struct ta_rap_cmd_output_data output;
+};
+
+struct ta_rap_shared_memory {
+	uint32_t                    cmd_id;
+	uint32_t                    validation_method_id;
+	uint32_t                    resp_id;
+	enum ta_rap_status          rap_status;
+	union ta_rap_cmd_input      rap_in_message;
+	union ta_rap_cmd_output     rap_out_message;
+	uint8_t                     reserved[64];
+};
+
+#endif  // #define _TA_RAP_IF_H
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
index 418cf097c918..5288617ca552 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
@@ -32,20 +32,6 @@
 
 #define UMC_6_INST_DIST			0x40000
 
-/*
- * (addr / 256) * 8192, the higher 26 bits in ErrorAddr
- * is the index of 8KB block
- */
-#define ADDR_OF_8KB_BLOCK(addr)			(((addr) & ~0xffULL) << 5)
-/* channel index is the index of 256B block */
-#define ADDR_OF_256B_BLOCK(channel_index)	((channel_index) << 8)
-/* offset in 256B block */
-#define OFFSET_IN_256B_BLOCK(addr)		((addr) & 0xffULL)
-
-#define LOOP_UMC_INST(umc_inst) for ((umc_inst) = 0; (umc_inst) < adev->umc.umc_inst_num; (umc_inst)++)
-#define LOOP_UMC_CH_INST(ch_inst) for ((ch_inst) = 0; (ch_inst) < adev->umc.channel_inst_num; (ch_inst)++)
-#define LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) LOOP_UMC_INST((umc_inst)) LOOP_UMC_CH_INST((ch_inst))
-
 const uint32_t
 	umc_v6_1_channel_idx_tbl[UMC_V6_1_UMC_INSTANCE_NUM][UMC_V6_1_CHANNEL_INSTANCE_NUM] = {
 		{2, 18, 11, 27},	{4, 20, 13, 29},
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c
new file mode 100644
index 000000000000..5665c77a9d58
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c
@@ -0,0 +1,331 @@
+/*
+ * Copyright 2020 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"),
+ * 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 "umc_v8_7.h"
+#include "amdgpu_ras.h"
+#include "amdgpu.h"
+
+#include "rsmu/rsmu_0_0_2_offset.h"
+#include "rsmu/rsmu_0_0_2_sh_mask.h"
+#include "umc/umc_8_7_0_offset.h"
+#include "umc/umc_8_7_0_sh_mask.h"
+
+#define UMC_8_INST_DIST			0x40000
+
+const uint32_t
+	umc_v8_7_channel_idx_tbl[UMC_V8_7_UMC_INSTANCE_NUM][UMC_V8_7_CHANNEL_INSTANCE_NUM] = {
+		{2, 11},  {4, 13},
+		{1, 8},   {7, 14},
+		{10, 3},  {12, 5},
+		{9, 0},   {15, 6}
+};
+
+static inline uint32_t get_umc_8_reg_offset(struct amdgpu_device *adev,
+					    uint32_t umc_inst,
+					    uint32_t ch_inst)
+{
+	return adev->umc.channel_offs*ch_inst + UMC_8_INST_DIST*umc_inst;
+}
+
+static void umc_v8_7_clear_error_count_per_channel(struct amdgpu_device *adev,
+					uint32_t umc_reg_offset)
+{
+	uint32_t ecc_err_cnt_addr;
+	uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr;
+
+	ecc_err_cnt_sel_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCntSel);
+	ecc_err_cnt_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCnt);
+
+	/* select the lower chip */
+	ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr +
+					umc_reg_offset) * 4);
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel,
+					UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 0);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4,
+			ecc_err_cnt_sel);
+
+	/* clear lower chip error count */
+	WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4,
+			UMC_V8_7_CE_CNT_INIT);
+
+	/* select the higher chip */
+	ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr +
+					umc_reg_offset) * 4);
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel,
+					UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 1);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4,
+			ecc_err_cnt_sel);
+
+	/* clear higher chip error count */
+	WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4,
+			UMC_V8_7_CE_CNT_INIT);
+}
+
+static void umc_v8_7_clear_error_count(struct amdgpu_device *adev)
+{
+	uint32_t umc_inst        = 0;
+	uint32_t ch_inst         = 0;
+	uint32_t umc_reg_offset  = 0;
+
+	LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) {
+		umc_reg_offset = get_umc_8_reg_offset(adev,
+						umc_inst,
+						ch_inst);
+
+		umc_v8_7_clear_error_count_per_channel(adev,
+						umc_reg_offset);
+	}
+}
+
+static void umc_v8_7_query_correctable_error_count(struct amdgpu_device *adev,
+						   uint32_t umc_reg_offset,
+						   unsigned long *error_count)
+{
+	uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr;
+	uint32_t ecc_err_cnt, ecc_err_cnt_addr;
+	uint64_t mc_umc_status;
+	uint32_t mc_umc_status_addr;
+
+	/* UMC 8_7_2 registers */
+	ecc_err_cnt_sel_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCntSel);
+	ecc_err_cnt_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCnt);
+	mc_umc_status_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_STATUST0);
+
+	/* select the lower chip and check the error count */
+	ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4);
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 0);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel);
+
+	ecc_err_cnt = RREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4);
+	*error_count +=
+		(REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_GeccErrCnt, GeccErrCnt) -
+		 UMC_V8_7_CE_CNT_INIT);
+
+	/* select the higher chip and check the err counter */
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 1);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel);
+
+	ecc_err_cnt = RREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4);
+	*error_count +=
+		(REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_GeccErrCnt, GeccErrCnt) -
+		 UMC_V8_7_CE_CNT_INIT);
+
+	/* check for SRAM correctable error
+	  MCUMC_STATUS is a 64 bit register */
+	mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4);
+	if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, ErrorCodeExt) == 6 &&
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)
+		*error_count += 1;
+}
+
+static void umc_v8_7_querry_uncorrectable_error_count(struct amdgpu_device *adev,
+						      uint32_t umc_reg_offset,
+						      unsigned long *error_count)
+{
+	uint64_t mc_umc_status;
+	uint32_t mc_umc_status_addr;
+
+	mc_umc_status_addr = SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_STATUST0);
+
+	/* check the MCUMC_STATUS */
+	mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4);
+	if ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) &&
+	    (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1 ||
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 ||
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 ||
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 ||
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1))
+		*error_count += 1;
+}
+
+static void umc_v8_7_query_ras_error_count(struct amdgpu_device *adev,
+					   void *ras_error_status)
+{
+	struct ras_err_data* err_data = (struct ras_err_data*)ras_error_status;
+
+	uint32_t umc_inst        = 0;
+	uint32_t ch_inst         = 0;
+	uint32_t umc_reg_offset  = 0;
+
+	LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) {
+		umc_reg_offset = get_umc_8_reg_offset(adev,
+						      umc_inst,
+						      ch_inst);
+
+		umc_v8_7_query_correctable_error_count(adev,
+						       umc_reg_offset,
+						       &(err_data->ce_count));
+		umc_v8_7_querry_uncorrectable_error_count(adev,
+							  umc_reg_offset,
+							  &(err_data->ue_count));
+	}
+
+	umc_v8_7_clear_error_count(adev);
+}
+
+static void umc_v8_7_query_error_address(struct amdgpu_device *adev,
+					 struct ras_err_data *err_data,
+					 uint32_t umc_reg_offset,
+					 uint32_t ch_inst,
+					 uint32_t umc_inst)
+{
+	uint32_t lsb, mc_umc_status_addr;
+	uint64_t mc_umc_status, err_addr, retired_page, mc_umc_addrt0;
+	struct eeprom_table_record *err_rec;
+	uint32_t channel_index = adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst];
+
+	mc_umc_status_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_STATUST0);
+	mc_umc_addrt0 =
+		SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_ADDRT0);
+
+	mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4);
+
+	if (mc_umc_status == 0)
+		return;
+
+	if (!err_data->err_addr) {
+		/* clear umc status */
+		WREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4, 0x0ULL);
+		return;
+	}
+
+	err_rec = &err_data->err_addr[err_data->err_addr_cnt];
+
+	/* calculate error address if ue/ce error is detected */
+	if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
+	    (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 ||
+	    REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)) {
+
+		err_addr = RREG64_PCIE((mc_umc_addrt0 + umc_reg_offset) * 4);
+		/* the lowest lsb bits should be ignored */
+		lsb = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, LSB);
+		err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr);
+		err_addr &= ~((0x1ULL << lsb) - 1);
+
+		/* translate umc channel address to soc pa, 3 parts are included */
+		retired_page = ADDR_OF_8KB_BLOCK(err_addr) |
+				ADDR_OF_256B_BLOCK(channel_index) |
+				OFFSET_IN_256B_BLOCK(err_addr);
+
+		/* we only save ue error information currently, ce is skipped */
+		if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC)
+				== 1) {
+			err_rec->address = err_addr;
+			/* page frame address is saved */
+			err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT;
+			err_rec->ts = (uint64_t)ktime_get_real_seconds();
+			err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE;
+			err_rec->cu = 0;
+			err_rec->mem_channel = channel_index;
+			err_rec->mcumc_id = umc_inst;
+
+			err_data->err_addr_cnt++;
+		}
+	}
+
+	/* clear umc status */
+	WREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4, 0x0ULL);
+}
+
+static void umc_v8_7_query_ras_error_address(struct amdgpu_device *adev,
+					     void *ras_error_status)
+{
+	struct ras_err_data* err_data = (struct ras_err_data*)ras_error_status;
+
+	uint32_t umc_inst        = 0;
+	uint32_t ch_inst         = 0;
+	uint32_t umc_reg_offset  = 0;
+
+	LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) {
+		umc_reg_offset = get_umc_8_reg_offset(adev,
+						      umc_inst,
+						      ch_inst);
+
+		umc_v8_7_query_error_address(adev,
+					     err_data,
+					     umc_reg_offset,
+					     ch_inst,
+					     umc_inst);
+	}
+}
+
+static void umc_v8_7_err_cnt_init_per_channel(struct amdgpu_device *adev,
+					      uint32_t umc_reg_offset)
+{
+	uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr;
+	uint32_t ecc_err_cnt_addr;
+
+	ecc_err_cnt_sel_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCntSel);
+	ecc_err_cnt_addr =
+		SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_GeccErrCnt);
+
+	/* select the lower chip and check the error count */
+	ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4);
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 0);
+	/* set ce error interrupt type to APIC based interrupt */
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_GeccErrCntSel,
+					GeccErrInt, 0x1);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel);
+	/* set error count to initial value */
+	WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, UMC_V8_7_CE_CNT_INIT);
+
+	/* select the higher chip and check the err counter */
+	ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_GeccErrCntSel,
+					GeccErrCntCsSel, 1);
+	WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel);
+	WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, UMC_V8_7_CE_CNT_INIT);
+}
+
+static void umc_v8_7_err_cnt_init(struct amdgpu_device *adev)
+{
+	uint32_t umc_inst        = 0;
+	uint32_t ch_inst         = 0;
+	uint32_t umc_reg_offset  = 0;
+
+	LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) {
+		umc_reg_offset = get_umc_8_reg_offset(adev,
+						      umc_inst,
+						      ch_inst);
+
+		umc_v8_7_err_cnt_init_per_channel(adev, umc_reg_offset);
+	}
+}
+
+const struct amdgpu_umc_funcs umc_v8_7_funcs = {
+	.err_cnt_init = umc_v8_7_err_cnt_init,
+	.ras_late_init = amdgpu_umc_ras_late_init,
+	.query_ras_error_count = umc_v8_7_query_ras_error_count,
+	.query_ras_error_address = umc_v8_7_query_ras_error_address,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h
new file mode 100644
index 000000000000..d4d0468e3df5
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright 2020 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"),
+ * 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.
+ *
+ */
+#ifndef __UMC_V8_7_H__
+#define __UMC_V8_7_H__
+
+#include "soc15_common.h"
+#include "amdgpu.h"
+
+/* HBM  Memory Channel Width */
+#define UMC_V8_7_HBM_MEMORY_CHANNEL_WIDTH	128
+/* number of umc channel instance with memory map register access */
+#define UMC_V8_7_CHANNEL_INSTANCE_NUM		2
+/* number of umc instance with memory map register access */
+#define UMC_V8_7_UMC_INSTANCE_NUM		8
+/* total channel instances in one umc block */
+#define UMC_V8_7_TOTAL_CHANNEL_NUM	(UMC_V8_7_CHANNEL_INSTANCE_NUM * UMC_V8_7_UMC_INSTANCE_NUM)
+/* UMC regiser per channel offset */
+#define UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA	0x400
+
+/* EccErrCnt max value */
+#define UMC_V8_7_CE_CNT_MAX		0xffff
+/* umc ce interrupt threshold */
+#define UMC_V8_7_CE_INT_THRESHOLD	0xffff
+/* umc ce count initial value */
+#define UMC_V8_7_CE_CNT_INIT	(UMC_V8_7_CE_CNT_MAX - UMC_V8_7_CE_INT_THRESHOLD)
+
+extern const struct amdgpu_umc_funcs umc_v8_7_funcs;
+extern const uint32_t
+	umc_v8_7_channel_idx_tbl[UMC_V8_7_UMC_INSTANCE_NUM][UMC_V8_7_CHANNEL_INSTANCE_NUM];
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index 3cafba726587..b0c0c438fc93 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -348,7 +348,7 @@ static int uvd_v4_2_start(struct amdgpu_device *adev)
 	/* Set the write pointer delay */
 	WREG32(mmUVD_RBC_RB_WPTR_CNTL, 0);
 
-	/* programm the 4GB memory segment for rptr and ring buffer */
+	/* program the 4GB memory segment for rptr and ring buffer */
 	WREG32(mmUVD_LMI_EXT40_ADDR, upper_32_bits(ring->gpu_addr) |
 				   (0x7 << 16) | (0x1 << 31));
 
@@ -541,7 +541,7 @@ static void uvd_v4_2_mc_resume(struct amdgpu_device *adev)
 	uint64_t addr;
 	uint32_t size;
 
-	/* programm the VCPU memory controller bits 0-27 */
+	/* program the VCPU memory controller bits 0-27 */
 	addr = (adev->uvd.inst->gpu_addr + AMDGPU_UVD_FIRMWARE_OFFSET) >> 3;
 	size = AMDGPU_UVD_FIRMWARE_SIZE(adev) >> 3;
 	WREG32(mmUVD_VCPU_CACHE_OFFSET0, addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index a566ff926e90..6e57001f6d0a 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -253,7 +253,7 @@ static void uvd_v5_0_mc_resume(struct amdgpu_device *adev)
 	uint64_t offset;
 	uint32_t size;
 
-	/* programm memory controller bits 0-27 */
+	/* program memory controller bits 0-27 */
 	WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
 			lower_32_bits(adev->uvd.inst->gpu_addr));
 	WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
@@ -404,7 +404,7 @@ static int uvd_v5_0_start(struct amdgpu_device *adev)
 	/* set the wb address */
 	WREG32(mmUVD_RBC_RB_RPTR_ADDR, (upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32(mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 			lower_32_bits(ring->gpu_addr));
 	WREG32(mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 0a880bc101b8..666bfa4a0b8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -583,7 +583,7 @@ static void uvd_v6_0_mc_resume(struct amdgpu_device *adev)
 	uint64_t offset;
 	uint32_t size;
 
-	/* programm memory controller bits 0-27 */
+	/* program memory controller bits 0-27 */
 	WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
 			lower_32_bits(adev->uvd.inst->gpu_addr));
 	WREG32(mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
@@ -825,7 +825,7 @@ static int uvd_v6_0_start(struct amdgpu_device *adev)
 	/* set the wb address */
 	WREG32(mmUVD_RBC_RB_RPTR_ADDR, (upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32(mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 			lower_32_bits(ring->gpu_addr));
 	WREG32(mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
@@ -1240,8 +1240,8 @@ static int uvd_v6_0_process_interrupt(struct amdgpu_device *adev,
 		break;
 	}
 
-	if (false == int_handled)
-			DRM_ERROR("Unhandled interrupt: %d %d\n",
+	if (!int_handled)
+		DRM_ERROR("Unhandled interrupt: %d %d\n",
 			  entry->src_id, entry->src_data[0]);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index e07e3fae99b5..b44c8677ce8d 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -1073,7 +1073,7 @@ static int uvd_v7_0_start(struct amdgpu_device *adev)
 		WREG32_SOC15(UVD, k, mmUVD_RBC_RB_RPTR_ADDR,
 				(upper_32_bits(ring->gpu_addr) >> 2));
 
-		/* programm the RB_BASE for ring buffer */
+		/* program the RB_BASE for ring buffer */
 		WREG32_SOC15(UVD, k, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 				lower_32_bits(ring->gpu_addr));
 		WREG32_SOC15(UVD, k, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 927c330fad21..73699eafb51e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -910,7 +910,7 @@ static int vcn_v1_0_start_spg_mode(struct amdgpu_device *adev)
 	WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR_ADDR,
 			(upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 			lower_32_bits(ring->gpu_addr));
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
@@ -1068,7 +1068,7 @@ static int vcn_v1_0_start_dpg_mode(struct amdgpu_device *adev)
 	WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR_ADDR,
 								(upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 								lower_32_bits(ring->gpu_addr));
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 23a9eb5b2c8a..e5d29dee0c88 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -900,7 +900,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
 	WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR_ADDR,
 		(upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 		lower_32_bits(ring->gpu_addr));
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
@@ -1060,7 +1060,7 @@ static int vcn_v2_0_start(struct amdgpu_device *adev)
 	WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_CNTL, tmp);
 
 	fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET;
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 		lower_32_bits(ring->gpu_addr));
 	WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index e99bef6e2354..0f1d3ef8baa7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -80,23 +80,18 @@ static int vcn_v2_5_early_init(void *handle)
 		adev->vcn.harvest_config = 0;
 		adev->vcn.num_enc_rings = 1;
 	} else {
-		if (adev->asic_type == CHIP_ARCTURUS) {
-			u32 harvest;
-			int i;
-
-			adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS;
-			for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-				harvest = RREG32_SOC15(VCN, i, mmCC_UVD_HARVESTING);
-				if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK)
-					adev->vcn.harvest_config |= 1 << i;
-			}
-
-			if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 |
-						AMDGPU_VCN_HARVEST_VCN1))
-				/* both instances are harvested, disable the block */
-				return -ENOENT;
-		} else
-			adev->vcn.num_vcn_inst = 1;
+		u32 harvest;
+		int i;
+		adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS;
+		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+			harvest = RREG32_SOC15(VCN, i, mmCC_UVD_HARVESTING);
+			if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK)
+				adev->vcn.harvest_config |= 1 << i;
+		}
+		if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 |
+					AMDGPU_VCN_HARVEST_VCN1))
+			/* both instances are harvested, disable the block */
+			return -ENOENT;
 
 		adev->vcn.num_enc_rings = 2;
 	}
@@ -887,7 +882,7 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo
 	WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR_ADDR,
 		(upper_32_bits(ring->gpu_addr) >> 2));
 
-	/* programm the RB_BASE for ring buffer */
+	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(VCN, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 		lower_32_bits(ring->gpu_addr));
 	WREG32_SOC15(VCN, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
@@ -1067,7 +1062,7 @@ static int vcn_v2_5_start(struct amdgpu_device *adev)
 		WREG32_SOC15(VCN, i, mmUVD_RBC_RB_CNTL, tmp);
 
 		fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET;
-		/* programm the RB_BASE for ring buffer */
+		/* program the RB_BASE for ring buffer */
 		WREG32_SOC15(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW,
 			lower_32_bits(ring->gpu_addr));
 		WREG32_SOC15(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH,
@@ -1108,7 +1103,7 @@ static int vcn_v2_5_mmsch_start(struct amdgpu_device *adev,
 {
 	uint32_t data = 0, loop = 0, size = 0;
 	uint64_t addr = table->gpu_addr;
-	struct mmsch_v1_1_init_header *header = NULL;;
+	struct mmsch_v1_1_init_header *header = NULL;
 
 	header = (struct mmsch_v1_1_init_header *)table->cpu_addr;
 	size = header->total_size;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 3a805eaf6f11..e074f7ed388c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -198,7 +198,7 @@ static int vcn_v3_0_sw_init(void *handle)
 		} else {
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 8 * i;
 		}
-		if (i != 0)
+		if (adev->asic_type == CHIP_SIENNA_CICHLID && i != 0)
 			ring->no_scheduler = true;
 		sprintf(ring->name, "vcn_dec_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0,
@@ -222,7 +222,7 @@ static int vcn_v3_0_sw_init(void *handle)
 			} else {
 				ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + j + 8 * i;
 			}
-			if (i != 1)
+			if (adev->asic_type == CHIP_SIENNA_CICHLID && i != 1)
 				ring->no_scheduler = true;
 			sprintf(ring->name, "vcn_enc_%d.%d", i, j);
 			r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index f6f2ed0830b1..9bcd0eebc6d7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -752,8 +752,10 @@ static int vi_asic_reset(struct amdgpu_device *adev)
 	int r;
 
 	if (vi_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
+		dev_info(adev->dev, "BACO reset\n");
 		r = amdgpu_dpm_baco_reset(adev);
 	} else {
+		dev_info(adev->dev, "PCI CONFIG reset\n");
 		r = vi_asic_pci_config_reset(adev);
 	}
 
@@ -1066,6 +1068,10 @@ static bool vi_need_reset_on_init(struct amdgpu_device *adev)
 	return false;
 }
 
+static void vi_pre_asic_init(struct amdgpu_device *adev)
+{
+}
+
 static const struct amdgpu_asic_funcs vi_asic_funcs =
 {
 	.read_disabled_bios = &vi_read_disabled_bios,
@@ -1086,6 +1092,7 @@ static const struct amdgpu_asic_funcs vi_asic_funcs =
 	.need_reset_on_init = &vi_need_reset_on_init,
 	.get_pcie_replay_count = &vi_get_pcie_replay_count,
 	.supports_baco = &vi_asic_supports_baco,
+	.pre_asic_init = &vi_pre_asic_init,
 };
 
 #define CZ_REV_BRISTOL(rev)	 \
@@ -1507,8 +1514,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_MC,
 			       pp_support_state,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & (AMD_CG_SUPPORT_SDMA_LS | AMD_CG_SUPPORT_SDMA_MGCG)) {
@@ -1526,8 +1532,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_SDMA,
 			       pp_support_state,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS | AMD_CG_SUPPORT_HDP_MGCG)) {
@@ -1545,8 +1550,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_HDP,
 			       pp_support_state,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 
@@ -1560,8 +1564,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_BIF,
 			       PP_STATE_SUPPORT_LS,
 			        pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 	if (adev->cg_flags & AMD_CG_SUPPORT_BIF_MGCG) {
 		if (state == AMD_CG_STATE_UNGATE)
@@ -1573,8 +1576,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_BIF,
 			       PP_STATE_SUPPORT_CG,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_DRM_LS) {
@@ -1588,8 +1590,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_DRM,
 			       PP_STATE_SUPPORT_LS,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_ROM_MGCG) {
@@ -1603,8 +1604,7 @@ static int vi_common_set_clockgating_state_by_smu(void *handle,
 			       PP_BLOCK_SYS_ROM,
 			       PP_STATE_SUPPORT_CG,
 			       pp_state);
-		if (adev->powerplay.pp_funcs->set_clockgating_by_smu)
-			amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
+		amdgpu_dpm_set_clockgating_by_smu(adev, msg_id);
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
index 577d901fdb63..affbca7c0050 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
@@ -911,7 +911,7 @@ static const uint32_t cwsr_trap_nv1x_hex[] = {
 	0x705d0000, 0x807c817c,
 	0x8070ff70, 0x00000080,
 	0xbf0a7b7c, 0xbf85fff8,
-	0xbf82014f, 0xbef4037e,
+	0xbf820151, 0xbef4037e,
 	0x8775ff7f, 0x0000ffff,
 	0x8875ff75, 0x00040000,
 	0xbef60380, 0xbef703ff,
@@ -1024,61 +1024,62 @@ static const uint32_t cwsr_trap_nv1x_hex[] = {
 	0xbe883108, 0xbe8a310a,
 	0xbe8c310c, 0xbe8e310e,
 	0xbf06807c, 0xbf84fff0,
-	0xb9782a05, 0x80788178,
-	0xbf0d9972, 0xbf850002,
-	0x8f788978, 0xbf820001,
-	0x8f788a78, 0xb96e1e06,
-	0x8f6e8a6e, 0x80786e78,
-	0x8078ff78, 0x00000200,
-	0xbef603ff, 0x01000000,
-	0xf4211bfa, 0xf0000000,
-	0x80788478, 0xf4211b3a,
+	0xba80f801, 0x00000000,
+	0xbf8a0000, 0xb9782a05,
+	0x80788178, 0xbf0d9972,
+	0xbf850002, 0x8f788978,
+	0xbf820001, 0x8f788a78,
+	0xb96e1e06, 0x8f6e8a6e,
+	0x80786e78, 0x8078ff78,
+	0x00000200, 0xbef603ff,
+	0x01000000, 0xf4211bfa,
 	0xf0000000, 0x80788478,
-	0xf4211b7a, 0xf0000000,
-	0x80788478, 0xf4211c3a,
+	0xf4211b3a, 0xf0000000,
+	0x80788478, 0xf4211b7a,
 	0xf0000000, 0x80788478,
-	0xf4211c7a, 0xf0000000,
-	0x80788478, 0xf4211eba,
+	0xf4211c3a, 0xf0000000,
+	0x80788478, 0xf4211c7a,
 	0xf0000000, 0x80788478,
-	0xf4211efa, 0xf0000000,
-	0x80788478, 0xf4211e7a,
+	0xf4211eba, 0xf0000000,
+	0x80788478, 0xf4211efa,
 	0xf0000000, 0x80788478,
-	0xf4211cfa, 0xf0000000,
-	0x80788478, 0xf4211bba,
+	0xf4211e7a, 0xf0000000,
+	0x80788478, 0xf4211cfa,
 	0xf0000000, 0x80788478,
-	0xbf8cc07f, 0xb9eef814,
 	0xf4211bba, 0xf0000000,
 	0x80788478, 0xbf8cc07f,
-	0xb9eef815, 0xbefc036f,
-	0xbefe0370, 0xbeff0371,
-	0x876f7bff, 0x000003ff,
-	0xb9ef4803, 0xb9f9f816,
-	0x876f7bff, 0xfffff800,
-	0x906f8b6f, 0xb9efa2c3,
-	0xb9f3f801, 0xb96e2a05,
-	0x806e816e, 0xbf0d9972,
-	0xbf850002, 0x8f6e896e,
-	0xbf820001, 0x8f6e8a6e,
-	0x806eff6e, 0x00000200,
-	0x806e746e, 0x826f8075,
-	0x876fff6f, 0x0000ffff,
-	0xf4091c37, 0xfa000050,
-	0xf4091d37, 0xfa000060,
-	0xf4011e77, 0xfa000074,
-	0xbf8cc07f, 0x876fff6d,
-	0xfc000000, 0x906f9a6f,
-	0x8f6f906f, 0xbeee0380,
+	0xb9eef814, 0xf4211bba,
+	0xf0000000, 0x80788478,
+	0xbf8cc07f, 0xb9eef815,
+	0xbefc036f, 0xbefe0370,
+	0xbeff0371, 0x876f7bff,
+	0x000003ff, 0xb9ef4803,
+	0xb9f9f816, 0x876f7bff,
+	0xfffff800, 0x906f8b6f,
+	0xb9efa2c3, 0xb9f3f801,
+	0xb96e2a05, 0x806e816e,
+	0xbf0d9972, 0xbf850002,
+	0x8f6e896e, 0xbf820001,
+	0x8f6e8a6e, 0x806eff6e,
+	0x00000200, 0x806e746e,
+	0x826f8075, 0x876fff6f,
+	0x0000ffff, 0xf4091c37,
+	0xfa000050, 0xf4091d37,
+	0xfa000060, 0xf4011e77,
+	0xfa000074, 0xbf8cc07f,
+	0x876fff6d, 0xfc000000,
+	0x906f9a6f, 0x8f6f906f,
+	0xbeee0380, 0x886e6f6e,
+	0x876fff6d, 0x02000000,
+	0x906f996f, 0x8f6f8f6f,
 	0x886e6f6e, 0x876fff6d,
-	0x02000000, 0x906f996f,
-	0x8f6f8f6f, 0x886e6f6e,
-	0x876fff6d, 0x01000000,
-	0x906f986f, 0x8f6f996f,
-	0x886e6f6e, 0x876fff7a,
-	0x00800000, 0x906f976f,
-	0xb9eef807, 0x876dff6d,
-	0x0000ffff, 0x87fe7e7e,
-	0x87ea6a6a, 0xb9faf802,
-	0xbf8a0000, 0xbe80226c,
+	0x01000000, 0x906f986f,
+	0x8f6f996f, 0x886e6f6e,
+	0x876fff7a, 0x00800000,
+	0x906f976f, 0xb9eef807,
+	0x876dff6d, 0x0000ffff,
+	0x87fe7e7e, 0x87ea6a6a,
+	0xb9faf802, 0xbe80226c,
 	0xbf810000, 0xbf9f0000,
 	0xbf9f0000, 0xbf9f0000,
 	0xbf9f0000, 0xbf9f0000,
@@ -1807,7 +1808,7 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
 	0x705d0000, 0x807c817c,
 	0x8070ff70, 0x00000080,
 	0xbf0a7b7c, 0xbf85fff8,
-	0xbf82013a, 0xbef4037e,
+	0xbf82013c, 0xbef4037e,
 	0x8775ff7f, 0x0000ffff,
 	0x8875ff75, 0x00040000,
 	0xbef60380, 0xbef703ff,
@@ -1920,50 +1921,51 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
 	0xbe883108, 0xbe8a310a,
 	0xbe8c310c, 0xbe8e310e,
 	0xbf06807c, 0xbf84fff0,
-	0xb9782a05, 0x80788178,
-	0xbf0d9972, 0xbf850002,
-	0x8f788978, 0xbf820001,
-	0x8f788a78, 0xb96e1e06,
-	0x8f6e8a6e, 0x80786e78,
-	0x8078ff78, 0x00000200,
-	0xbef603ff, 0x01000000,
-	0xf4211bfa, 0xf0000000,
-	0x80788478, 0xf4211b3a,
+	0xba80f801, 0x00000000,
+	0xbf8a0000, 0xb9782a05,
+	0x80788178, 0xbf0d9972,
+	0xbf850002, 0x8f788978,
+	0xbf820001, 0x8f788a78,
+	0xb96e1e06, 0x8f6e8a6e,
+	0x80786e78, 0x8078ff78,
+	0x00000200, 0xbef603ff,
+	0x01000000, 0xf4211bfa,
 	0xf0000000, 0x80788478,
-	0xf4211b7a, 0xf0000000,
-	0x80788478, 0xf4211c3a,
+	0xf4211b3a, 0xf0000000,
+	0x80788478, 0xf4211b7a,
 	0xf0000000, 0x80788478,
-	0xf4211c7a, 0xf0000000,
-	0x80788478, 0xf4211eba,
+	0xf4211c3a, 0xf0000000,
+	0x80788478, 0xf4211c7a,
 	0xf0000000, 0x80788478,
-	0xf4211efa, 0xf0000000,
-	0x80788478, 0xf4211e7a,
+	0xf4211eba, 0xf0000000,
+	0x80788478, 0xf4211efa,
 	0xf0000000, 0x80788478,
-	0xf4211cfa, 0xf0000000,
-	0x80788478, 0xf4211bba,
+	0xf4211e7a, 0xf0000000,
+	0x80788478, 0xf4211cfa,
 	0xf0000000, 0x80788478,
-	0xbf8cc07f, 0xb9eef814,
 	0xf4211bba, 0xf0000000,
 	0x80788478, 0xbf8cc07f,
-	0xb9eef815, 0xbefc036f,
-	0xbefe0370, 0xbeff0371,
-	0x876f7bff, 0x000003ff,
-	0xb9ef4803, 0x876f7bff,
-	0xfffff800, 0x906f8b6f,
-	0xb9efa2c3, 0xb9f3f801,
-	0xb96e2a05, 0x806e816e,
-	0xbf0d9972, 0xbf850002,
-	0x8f6e896e, 0xbf820001,
-	0x8f6e8a6e, 0x806eff6e,
-	0x00000200, 0x806e746e,
-	0x826f8075, 0x876fff6f,
-	0x0000ffff, 0xf4091c37,
-	0xfa000050, 0xf4091d37,
-	0xfa000060, 0xf4011e77,
-	0xfa000074, 0xbf8cc07f,
-	0x876dff6d, 0x0000ffff,
-	0x87fe7e7e, 0x87ea6a6a,
-	0xb9faf802, 0xbf8a0000,
+	0xb9eef814, 0xf4211bba,
+	0xf0000000, 0x80788478,
+	0xbf8cc07f, 0xb9eef815,
+	0xbefc036f, 0xbefe0370,
+	0xbeff0371, 0x876f7bff,
+	0x000003ff, 0xb9ef4803,
+	0x876f7bff, 0xfffff800,
+	0x906f8b6f, 0xb9efa2c3,
+	0xb9f3f801, 0xb96e2a05,
+	0x806e816e, 0xbf0d9972,
+	0xbf850002, 0x8f6e896e,
+	0xbf820001, 0x8f6e8a6e,
+	0x806eff6e, 0x00000200,
+	0x806e746e, 0x826f8075,
+	0x876fff6f, 0x0000ffff,
+	0xf4091c37, 0xfa000050,
+	0xf4091d37, 0xfa000060,
+	0xf4011e77, 0xfa000074,
+	0xbf8cc07f, 0x876dff6d,
+	0x0000ffff, 0x87fe7e7e,
+	0x87ea6a6a, 0xb9faf802,
 	0xbe80226c, 0xbf810000,
 	0xbf9f0000, 0xbf9f0000,
 	0xbf9f0000, 0xbf9f0000,
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
index 5b220f2a7501..5081f91190b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
@@ -894,6 +894,11 @@ L_RESTORE_SGPR:
 	s_cmp_eq_u32	m0, 0							//scc = (m0 < s_sgpr_save_num) ? 1 : 0
 	s_cbranch_scc0	L_RESTORE_SGPR_LOOP
 
+	// s_barrier with MODE.DEBUG_EN=1, STATUS.PRIV=1 incorrectly asserts debug exception.
+	// Clear DEBUG_EN before and restore MODE after the barrier.
+	s_setreg_imm32_b32	hwreg(HW_REG_MODE), 0
+	s_barrier								//barrier to ensure the readiness of LDS before access attemps from any other wave in the same TG
+
 	/* restore HW registers */
 L_RESTORE_HWREG:
 	// HWREG SR memory offset : size(VGPR)+size(SVGPR)+size(SGPR)
@@ -976,8 +981,6 @@ L_RESTORE_HWREG:
 	s_and_b64	vcc, vcc, vcc						// Restore STATUS.VCCZ, not writable by s_setreg_b32
 	s_setreg_b32	hwreg(HW_REG_STATUS), s_restore_status			// SCC is included, which is changed by previous salu
 
-	s_barrier								//barrier to ensure the readiness of LDS before access attemps from any other wave in the same TG
-
 	s_rfe_b64	s_restore_pc_lo						//Return to the main shader program and resume execution
 
 L_END_PGM:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index e9b96ad3d9a5..222f1df1a6b6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -97,6 +97,7 @@ void kfd_chardev_exit(void)
 	device_destroy(kfd_class, MKDEV(kfd_char_dev_major, 0));
 	class_destroy(kfd_class);
 	unregister_chrdev(kfd_char_dev_major, kfd_dev_name);
+	kfd_device = NULL;
 }
 
 struct device *kfd_chardev(void)
@@ -1254,7 +1255,7 @@ bool kfd_dev_is_large_bar(struct kfd_dev *dev)
 		return true;
 	}
 
-	if (dev->device_info->needs_iommu_device)
+	if (dev->use_iommu_v2)
 		return false;
 
 	amdgpu_amdkfd_get_local_mem_info(dev->kgd, &mem_info);
@@ -1290,18 +1291,6 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 		return -EINVAL;
 	}
 
-	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) {
-		if (args->size != kfd_doorbell_process_slice(dev))
-			return -EINVAL;
-		offset = kfd_get_process_doorbells(dev, p);
-	} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) {
-		if (args->size != PAGE_SIZE)
-			return -EINVAL;
-		offset = amdgpu_amdkfd_get_mmio_remap_phys_addr(dev->kgd);
-		if (!offset)
-			return -ENOMEM;
-	}
-
 	mutex_lock(&p->mutex);
 
 	pdd = kfd_bind_process_to_device(dev, p);
@@ -1310,6 +1299,24 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 		goto err_unlock;
 	}
 
+	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) {
+		if (args->size != kfd_doorbell_process_slice(dev)) {
+			err = -EINVAL;
+			goto err_unlock;
+		}
+		offset = kfd_get_process_doorbells(pdd);
+	} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) {
+		if (args->size != PAGE_SIZE) {
+			err = -EINVAL;
+			goto err_unlock;
+		}
+		offset = amdgpu_amdkfd_get_mmio_remap_phys_addr(dev->kgd);
+		if (!offset) {
+			err = -ENOMEM;
+			goto err_unlock;
+		}
+	}
+
 	err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		dev->kgd, args->va_addr, args->size,
 		pdd->vm, (struct kgd_mem **) &mem, &offset,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 6a250f8fcfb8..d2981524dba0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -742,6 +742,22 @@ static int kfd_fill_gpu_cache_info(struct kfd_dev *kdev,
 	return 0;
 }
 
+static bool kfd_ignore_crat(void)
+{
+	bool ret;
+
+	if (ignore_crat)
+		return true;
+
+#ifndef KFD_SUPPORT_IOMMU_V2
+	ret = true;
+#else
+	ret = false;
+#endif
+
+	return ret;
+}
+
 /*
  * kfd_create_crat_image_acpi - Allocates memory for CRAT image and
  * copies CRAT from ACPI (if available).
@@ -776,12 +792,13 @@ int kfd_create_crat_image_acpi(void **crat_image, size_t *size)
 		return -EINVAL;
 	}
 
-	if (ignore_crat) {
+	if (kfd_ignore_crat()) {
 		pr_info("CRAT table disabled by module option\n");
 		return -ENODATA;
 	}
 
-	pcrat_image = kmemdup(crat_table, crat_table->length, GFP_KERNEL);
+	pcrat_image = kvmalloc(crat_table->length, GFP_KERNEL);
+	memcpy(pcrat_image, crat_table, crat_table->length);
 	if (!pcrat_image)
 		return -ENOMEM;
 
@@ -793,11 +810,10 @@ int kfd_create_crat_image_acpi(void **crat_image, size_t *size)
 
 /* Memory required to create Virtual CRAT.
  * Since there is no easy way to predict the amount of memory required, the
- * following amount are allocated for CPU and GPU Virtual CRAT. This is
+ * following amount is allocated for GPU Virtual CRAT. This is
  * expected to cover all known conditions. But to be safe additional check
  * is put in the code to ensure we don't overwrite.
  */
-#define VCRAT_SIZE_FOR_CPU	(2 * PAGE_SIZE)
 #define VCRAT_SIZE_FOR_GPU	(4 * PAGE_SIZE)
 
 /* kfd_fill_cu_for_cpu - Fill in Compute info for the given CPU NUMA node
@@ -948,7 +964,7 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 #endif
 	int ret = 0;
 
-	if (!pcrat_image || avail_size < VCRAT_SIZE_FOR_CPU)
+	if (!pcrat_image)
 		return -EINVAL;
 
 	/* Fill in CRAT Header.
@@ -1348,30 +1364,37 @@ int kfd_create_crat_image_virtual(void **crat_image, size_t *size,
 				  uint32_t proximity_domain)
 {
 	void *pcrat_image = NULL;
-	int ret = 0;
+	int ret = 0, num_nodes;
+	size_t dyn_size;
 
 	if (!crat_image)
 		return -EINVAL;
 
 	*crat_image = NULL;
 
-	/* Allocate one VCRAT_SIZE_FOR_CPU for CPU virtual CRAT image and
-	 * VCRAT_SIZE_FOR_GPU for GPU virtual CRAT image. This should cover
-	 * all the current conditions. A check is put not to overwrite beyond
-	 * allocated size
+	/* Allocate the CPU Virtual CRAT size based on the number of online
+	 * nodes. Allocate VCRAT_SIZE_FOR_GPU for GPU virtual CRAT image.
+	 * This should cover all the current conditions. A check is put not
+	 * to overwrite beyond allocated size for GPUs
 	 */
 	switch (flags) {
 	case COMPUTE_UNIT_CPU:
-		pcrat_image = kmalloc(VCRAT_SIZE_FOR_CPU, GFP_KERNEL);
+		num_nodes = num_online_nodes();
+		dyn_size = sizeof(struct crat_header) +
+			num_nodes * (sizeof(struct crat_subtype_computeunit) +
+			sizeof(struct crat_subtype_memory) +
+			(num_nodes - 1) * sizeof(struct crat_subtype_iolink));
+		pcrat_image = kvmalloc(dyn_size, GFP_KERNEL);
 		if (!pcrat_image)
 			return -ENOMEM;
-		*size = VCRAT_SIZE_FOR_CPU;
+		*size = dyn_size;
+		pr_debug("CRAT size is %ld", dyn_size);
 		ret = kfd_create_vcrat_image_cpu(pcrat_image, size);
 		break;
 	case COMPUTE_UNIT_GPU:
 		if (!kdev)
 			return -EINVAL;
-		pcrat_image = kmalloc(VCRAT_SIZE_FOR_GPU, GFP_KERNEL);
+		pcrat_image = kvmalloc(VCRAT_SIZE_FOR_GPU, GFP_KERNEL);
 		if (!pcrat_image)
 			return -ENOMEM;
 		*size = VCRAT_SIZE_FOR_GPU;
@@ -1390,7 +1413,7 @@ int kfd_create_crat_image_virtual(void **crat_image, size_t *size,
 	if (!ret)
 		*crat_image = pcrat_image;
 	else
-		kfree(pcrat_image);
+		kvfree(pcrat_image);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 4bfedaab183f..903170e59342 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -29,6 +29,7 @@
 #include "cwsr_trap_handler.h"
 #include "kfd_iommu.h"
 #include "amdgpu_amdkfd.h"
+#include "kfd_smi_events.h"
 
 #define MQD_SIZE_ALIGNED 768
 
@@ -115,6 +116,7 @@ static const struct kfd_device_info carrizo_device_info = {
 	.num_xgmi_sdma_engines = 0,
 	.num_sdma_queues_per_engine = 2,
 };
+#endif
 
 static const struct kfd_device_info raven_device_info = {
 	.asic_family = CHIP_RAVEN,
@@ -133,7 +135,6 @@ static const struct kfd_device_info raven_device_info = {
 	.num_xgmi_sdma_engines = 0,
 	.num_sdma_queues_per_engine = 2,
 };
-#endif
 
 static const struct kfd_device_info hawaii_device_info = {
 	.asic_family = CHIP_HAWAII,
@@ -502,8 +503,8 @@ static const struct kfd_device_info *kfd_supported_devices[][2] = {
 #ifdef KFD_SUPPORT_IOMMU_V2
 	[CHIP_KAVERI] = {&kaveri_device_info, NULL},
 	[CHIP_CARRIZO] = {&carrizo_device_info, NULL},
-	[CHIP_RAVEN] = {&raven_device_info, NULL},
 #endif
+	[CHIP_RAVEN] = {&raven_device_info, NULL},
 	[CHIP_HAWAII] = {&hawaii_device_info, NULL},
 	[CHIP_TONGA] = {&tonga_device_info, NULL},
 	[CHIP_FIJI] = {&fiji_device_info, &fiji_vf_device_info},
@@ -582,6 +583,8 @@ struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd,
 
 	atomic_set(&kfd->sram_ecc_flag, 0);
 
+	ida_init(&kfd->doorbell_ida);
+
 	return kfd;
 }
 
@@ -711,11 +714,11 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		goto kfd_doorbell_error;
 	}
 
-	if (kfd->kfd2kgd->get_hive_id)
-		kfd->hive_id = kfd->kfd2kgd->get_hive_id(kfd->kgd);
+	kfd->hive_id = amdgpu_amdkfd_get_hive_id(kfd->kgd);
+
+	kfd->unique_id = amdgpu_amdkfd_get_unique_id(kfd->kgd);
 
-	if (kfd->kfd2kgd->get_unique_id)
-		kfd->unique_id = kfd->kfd2kgd->get_unique_id(kfd->kgd);
+	kfd->noretry = amdgpu_amdkfd_get_noretry(kfd->kgd);
 
 	if (kfd_interrupt_init(kfd)) {
 		dev_err(kfd_device, "Error initializing interrupts\n");
@@ -737,6 +740,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		goto gws_error;
 	}
 
+	/* If CRAT is broken, won't set iommu enabled */
+	kfd_double_confirm_iommu_support(kfd);
+
 	if (kfd_iommu_device_init(kfd)) {
 		dev_err(kfd_device, "Error initializing iommuv2\n");
 		goto device_iommu_error;
@@ -796,6 +802,7 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
 		kfd_interrupt_exit(kfd);
 		kfd_topology_remove_device(kfd);
 		kfd_doorbell_fini(kfd);
+		ida_destroy(&kfd->doorbell_ida);
 		kfd_gtt_sa_fini(kfd);
 		amdgpu_amdkfd_free_gtt_mem(kfd->kgd, kfd->gtt_mem);
 		if (kfd->gws)
@@ -810,6 +817,8 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 	if (!kfd->init_complete)
 		return 0;
 
+	kfd_smi_event_update_gpu_reset(kfd, false);
+
 	kfd->dqm->ops.pre_reset(kfd->dqm);
 
 	kgd2kfd_suspend(kfd, false);
@@ -838,6 +847,8 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd)
 
 	atomic_set(&kfd->sram_ecc_flag, 0);
 
+	kfd_smi_event_update_gpu_reset(kfd, true);
+
 	return 0;
 }
 
@@ -1245,6 +1256,12 @@ void kfd_dec_compute_active(struct kfd_dev *kfd)
 	WARN_ONCE(count < 0, "Compute profile ref. count error");
 }
 
+void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint32_t throttle_bitmask)
+{
+	if (kfd)
+		kfd_smi_event_update_thermal_throttling(kfd, throttle_bitmask);
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 /* This function will send a package to HIQ to hang the HWS
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 a8d316711625..c0ae04a08625 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -153,30 +153,6 @@ static void decrement_queue_count(struct device_queue_manager *dqm,
 		dqm->active_cp_queue_count--;
 }
 
-int read_sdma_queue_counter(uint64_t q_rptr, uint64_t *val)
-{
-	int ret;
-	uint64_t tmp = 0;
-
-	if (!val)
-		return -EINVAL;
-	/*
-	 * SDMA activity counter is stored at queue's RPTR + 0x8 location.
-	 */
-	if (!access_ok((const void __user *)(q_rptr +
-					sizeof(uint64_t)), sizeof(uint64_t))) {
-		pr_err("Can't access sdma queue activity counter\n");
-		return -EFAULT;
-	}
-
-	ret = get_user(tmp, (uint64_t *)(q_rptr + sizeof(uint64_t)));
-	if (!ret) {
-		*val = tmp;
-	}
-
-	return ret;
-}
-
 static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q)
 {
 	struct kfd_dev *dev = qpd->dqm->dev;
@@ -215,9 +191,8 @@ static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q)
 	}
 
 	q->properties.doorbell_off =
-		kfd_get_doorbell_dw_offset_in_bar(dev, q->process,
+		kfd_get_doorbell_dw_offset_in_bar(dev, qpd_to_pdd(qpd),
 					  q->doorbell_id);
-
 	return 0;
 }
 
@@ -552,7 +527,7 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
 	    (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI)) {
-		retval = read_sdma_queue_counter((uint64_t)q->properties.read_ptr,
+		retval = read_sdma_queue_counter((uint64_t __user *)q->properties.read_ptr,
 							&sdma_val);
 		if (retval)
 			pr_err("Failed to read SDMA queue counter for queue: %d\n",
@@ -674,9 +649,10 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm,
 		goto out;
 
 	pdd = qpd_to_pdd(qpd);
-	pr_info_ratelimited("Evicting PASID 0x%x queues\n",
+	pr_debug_ratelimited("Evicting PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
+	pdd->last_evict_timestamp = get_jiffies_64();
 	/* Mark all queues as evicted. Deactivate all active queues on
 	 * the qpd.
 	 */
@@ -724,7 +700,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		goto out;
 
 	pdd = qpd_to_pdd(qpd);
-	pr_info_ratelimited("Evicting PASID 0x%x queues\n",
+	pr_debug_ratelimited("Evicting PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
 	/* Mark all queues as evicted. Deactivate all active queues on
@@ -738,6 +714,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		q->properties.is_active = false;
 		decrement_queue_count(dqm, q->properties.type);
 	}
+	pdd->last_evict_timestamp = get_jiffies_64();
 	retval = execute_queues_cpsch(dqm,
 				qpd->is_debug ?
 				KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES :
@@ -756,6 +733,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
 	struct mqd_manager *mqd_mgr;
 	struct kfd_process_device *pdd;
 	uint64_t pd_base;
+	uint64_t eviction_duration;
 	int retval, ret = 0;
 
 	pdd = qpd_to_pdd(qpd);
@@ -770,7 +748,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
 		goto out;
 	}
 
-	pr_info_ratelimited("Restoring PASID 0x%x queues\n",
+	pr_debug_ratelimited("Restoring PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
 	/* Update PD Base in QPD */
@@ -823,6 +801,8 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
 			ret = retval;
 	}
 	qpd->evicted = 0;
+	eviction_duration = get_jiffies_64() - pdd->last_evict_timestamp;
+	atomic64_add(eviction_duration, &pdd->evict_duration_counter);
 out:
 	if (mm)
 		mmput(mm);
@@ -836,6 +816,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 	struct queue *q;
 	struct kfd_process_device *pdd;
 	uint64_t pd_base;
+	uint64_t eviction_duration;
 	int retval = 0;
 
 	pdd = qpd_to_pdd(qpd);
@@ -850,7 +831,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 		goto out;
 	}
 
-	pr_info_ratelimited("Restoring PASID 0x%x queues\n",
+	pr_debug_ratelimited("Restoring PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
 	/* Update PD Base in QPD */
@@ -869,6 +850,8 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 	retval = execute_queues_cpsch(dqm,
 				KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 	qpd->evicted = 0;
+	eviction_duration = get_jiffies_64() - pdd->last_evict_timestamp;
+	atomic64_add(eviction_duration, &pdd->evict_duration_counter);
 out:
 	dqm_unlock(dqm);
 	return retval;
@@ -1475,7 +1458,7 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
 	    (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI)) {
-		retval = read_sdma_queue_counter((uint64_t)q->properties.read_ptr,
+		retval = read_sdma_queue_counter((uint64_t __user *)q->properties.read_ptr,
 							&sdma_val);
 		if (retval)
 			pr_err("Failed to read SDMA queue counter for queue: %d\n",
@@ -1989,6 +1972,7 @@ int kfd_process_vm_fault(struct device_queue_manager *dqm, u32 pasid)
 
 	if (!p)
 		return -EINVAL;
+	WARN(debug_evictions, "Evicting pid %d", p->lead_thread->pid);
 	pdd = kfd_get_process_device_data(dqm->dev, p);
 	if (pdd)
 		ret = dqm->ops.evict_process_queues(dqm, &pdd->qpd);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index 49d8e324c636..16262e5d93f5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -251,5 +251,11 @@ static inline void dqm_unlock(struct device_queue_manager *dqm)
 	mutex_unlock(&dqm->lock_hidden);
 }
 
-int read_sdma_queue_counter(uint64_t q_rptr, uint64_t *val);
+static inline int read_sdma_queue_counter(uint64_t __user *q_rptr, uint64_t *val)
+{
+        /*
+         * SDMA activity counter is stored at queue's RPTR + 0x8 location.
+         */
+	return get_user(*val, q_rptr + 1);
+}
 #endif /* KFD_DEVICE_QUEUE_MANAGER_H_ */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
index 95a82ac455f2..eca6331efa94 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
@@ -61,8 +61,8 @@ static int update_qpd_v9(struct device_queue_manager *dqm,
 		qpd->sh_mem_config =
 				SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
 					SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
-		if (amdgpu_noretry &&
-		    !dqm->dev->device_info->needs_iommu_device)
+		if (dqm->dev->noretry &&
+		    !dqm->dev->use_iommu_v2)
 			qpd->sh_mem_config |=
 				1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
index 8e0c00b9555e..768d153acff4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
@@ -31,9 +31,6 @@
  * kernel queues using the first doorbell page reserved for the kernel.
  */
 
-static DEFINE_IDA(doorbell_ida);
-static unsigned int max_doorbell_slices;
-
 /*
  * Each device exposes a doorbell aperture, a PCI MMIO aperture that
  * receives 32-bit writes that are passed to queues as wptr values.
@@ -84,9 +81,9 @@ int kfd_doorbell_init(struct kfd_dev *kfd)
 	else
 		return -ENOSPC;
 
-	if (!max_doorbell_slices ||
-	    doorbell_process_limit < max_doorbell_slices)
-		max_doorbell_slices = doorbell_process_limit;
+	if (!kfd->max_doorbell_slices ||
+	    doorbell_process_limit < kfd->max_doorbell_slices)
+		kfd->max_doorbell_slices = doorbell_process_limit;
 
 	kfd->doorbell_base = kfd->shared_resources.doorbell_physical_address +
 				doorbell_start_offset;
@@ -130,6 +127,7 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
 		      struct vm_area_struct *vma)
 {
 	phys_addr_t address;
+	struct kfd_process_device *pdd;
 
 	/*
 	 * For simplicitly we only allow mapping of the entire doorbell
@@ -138,9 +136,12 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
 	if (vma->vm_end - vma->vm_start != kfd_doorbell_process_slice(dev))
 		return -EINVAL;
 
-	/* Calculate physical address of doorbell */
-	address = kfd_get_process_doorbells(dev, process);
+	pdd = kfd_get_process_device_data(dev, process);
+	if (!pdd)
+		return -EINVAL;
 
+	/* Calculate physical address of doorbell */
+	address = kfd_get_process_doorbells(pdd);
 	vma->vm_flags |= VM_IO | VM_DONTCOPY | VM_DONTEXPAND | VM_NORESERVE |
 				VM_DONTDUMP | VM_PFNMAP;
 
@@ -226,7 +227,7 @@ void write_kernel_doorbell64(void __iomem *db, u64 value)
 }
 
 unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd,
-					struct kfd_process *process,
+					struct kfd_process_device *pdd,
 					unsigned int doorbell_id)
 {
 	/*
@@ -236,7 +237,7 @@ unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd,
 	 * units regardless of the ASIC-dependent doorbell size.
 	 */
 	return kfd->doorbell_base_dw_offset +
-		process->doorbell_index
+		pdd->doorbell_index
 		* kfd_doorbell_process_slice(kfd) / sizeof(u32) +
 		doorbell_id * kfd->device_info->doorbell_size / sizeof(u32);
 }
@@ -251,25 +252,24 @@ uint64_t kfd_get_number_elems(struct kfd_dev *kfd)
 
 }
 
-phys_addr_t kfd_get_process_doorbells(struct kfd_dev *dev,
-					struct kfd_process *process)
+phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd)
 {
-	return dev->doorbell_base +
-		process->doorbell_index * kfd_doorbell_process_slice(dev);
+	return pdd->dev->doorbell_base +
+		pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev);
 }
 
-int kfd_alloc_process_doorbells(struct kfd_process *process)
+int kfd_alloc_process_doorbells(struct kfd_dev *kfd, unsigned int *doorbell_index)
 {
-	int r = ida_simple_get(&doorbell_ida, 1, max_doorbell_slices,
+	int r = ida_simple_get(&kfd->doorbell_ida, 1, kfd->max_doorbell_slices,
 				GFP_KERNEL);
 	if (r > 0)
-		process->doorbell_index = r;
+		*doorbell_index = r;
 
 	return r;
 }
 
-void kfd_free_process_doorbells(struct kfd_process *process)
+void kfd_free_process_doorbells(struct kfd_dev *kfd, unsigned int doorbell_index)
 {
-	if (process->doorbell_index)
-		ida_simple_remove(&doorbell_ida, process->doorbell_index);
+	if (doorbell_index)
+		ida_simple_remove(&kfd->doorbell_ida, doorbell_index);
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
index c1166c40ac15..3c22909470f2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
@@ -321,7 +321,7 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id)
 	pdd->lds_base = MAKE_LDS_APP_BASE_VI();
 	pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base);
 
-	if (!pdd->dev->device_info->needs_iommu_device) {
+	if (!pdd->dev->use_iommu_v2) {
 		/* dGPUs: SVM aperture starting at 0
 		 * with small reserved space for kernel.
 		 * Set them to CANONICAL addresses.
@@ -425,7 +425,7 @@ int kfd_init_apertures(struct kfd_process *process)
 				return -EINVAL;
 			}
 
-			if (!dev->device_info->needs_iommu_device) {
+			if (!dev->use_iommu_v2) {
 				/* dGPUs: the reserved space for kernel
 				 * before SVM
 				 */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
index e8ef3886688b..66bbca61e3ef 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
@@ -41,7 +41,7 @@ int kfd_iommu_check_device(struct kfd_dev *kfd)
 	struct amd_iommu_device_info iommu_info;
 	int err;
 
-	if (!kfd->device_info->needs_iommu_device)
+	if (!kfd->use_iommu_v2)
 		return -ENODEV;
 
 	iommu_info.flags = 0;
@@ -63,7 +63,7 @@ int kfd_iommu_device_init(struct kfd_dev *kfd)
 	unsigned int pasid_limit;
 	int err;
 
-	if (!kfd->device_info->needs_iommu_device)
+	if (!kfd->use_iommu_v2)
 		return 0;
 
 	iommu_info.flags = 0;
@@ -109,7 +109,7 @@ int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd)
 	struct kfd_process *p = pdd->process;
 	int err;
 
-	if (!dev->device_info->needs_iommu_device || pdd->bound == PDD_BOUND)
+	if (!dev->use_iommu_v2 || pdd->bound == PDD_BOUND)
 		return 0;
 
 	if (unlikely(pdd->bound == PDD_BOUND_SUSPENDED)) {
@@ -284,7 +284,7 @@ static void kfd_unbind_processes_from_device(struct kfd_dev *kfd)
  */
 void kfd_iommu_suspend(struct kfd_dev *kfd)
 {
-	if (!kfd->device_info->needs_iommu_device)
+	if (!kfd->use_iommu_v2)
 		return;
 
 	kfd_unbind_processes_from_device(kfd);
@@ -304,7 +304,7 @@ int kfd_iommu_resume(struct kfd_dev *kfd)
 	unsigned int pasid_limit;
 	int err;
 
-	if (!kfd->device_info->needs_iommu_device)
+	if (!kfd->use_iommu_v2)
 		return 0;
 
 	pasid_limit = kfd_get_pasid_limit();
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index f4b7f7e6c40e..5e90fe642192 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -70,6 +70,7 @@ err_create_wq:
 err_topology:
 	kfd_chardev_exit();
 err_ioctl:
+	pr_err("KFD is disabled due to module initialization failure\n");
 	return err;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 922ae138ab85..c77cf23032ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -297,6 +297,9 @@ struct kfd_dev {
 
 	bool pci_atomic_requested;
 
+	/* Use IOMMU v2 flag */
+	bool use_iommu_v2;
+
 	/* SRAM ECC flag */
 	atomic_t sram_ecc_flag;
 
@@ -309,6 +312,13 @@ struct kfd_dev {
 	/* Clients watching SMI events */
 	struct list_head smi_clients;
 	spinlock_t smi_lock;
+
+	uint32_t reset_seq_num;
+
+	struct ida doorbell_ida;
+	unsigned int max_doorbell_slices;
+
+	int noretry;
 };
 
 enum kfd_mempool {
@@ -626,7 +636,7 @@ enum kfd_pdd_bound {
 	PDD_BOUND_SUSPENDED,
 };
 
-#define MAX_SYSFS_FILENAME_LEN 11
+#define MAX_SYSFS_FILENAME_LEN 15
 
 /*
  * SDMA counter runs at 100MHz frequency.
@@ -687,6 +697,39 @@ struct kfd_process_device {
 	uint64_t sdma_past_activity_counter;
 	struct attribute attr_sdma;
 	char sdma_filename[MAX_SYSFS_FILENAME_LEN];
+
+	/* Eviction activity tracking */
+	uint64_t last_evict_timestamp;
+	atomic64_t evict_duration_counter;
+	struct attribute attr_evict;
+
+	struct kobject *kobj_stats;
+	unsigned int doorbell_index;
+
+	/*
+	 * @cu_occupancy: Reports occupancy of Compute Units (CU) of a process
+	 * that is associated with device encoded by "this" struct instance. The
+	 * value reflects CU usage by all of the waves launched by this process
+	 * on this device. A very important property of occupancy parameter is
+	 * that its value is a snapshot of current use.
+	 *
+	 * Following is to be noted regarding how this parameter is reported:
+	 *
+	 *  The number of waves that a CU can launch is limited by couple of
+	 *  parameters. These are encoded by struct amdgpu_cu_info instance
+	 *  that is part of every device definition. For GFX9 devices this
+	 *  translates to 40 waves (simd_per_cu * max_waves_per_simd) when waves
+	 *  do not use scratch memory and 32 waves (max_scratch_slots_per_cu)
+	 *  when they do use scratch memory. This could change for future
+	 *  devices and therefore this example should be considered as a guide.
+	 *
+	 *  All CU's of a device are available for the process. This may not be true
+	 *  under certain conditions - e.g. CU masking.
+	 *
+	 *  Finally number of CU's that are occupied by a process is affected by both
+	 *  number of CU's a device has along with number of other competing processes
+	 */
+	struct attribute attr_cu_occupancy;
 };
 
 #define qpd_to_pdd(x) container_of(x, struct kfd_process_device, qpd)
@@ -724,7 +767,6 @@ struct kfd_process {
 	struct mmu_notifier mmu_notifier;
 
 	u32 pasid;
-	unsigned int doorbell_index;
 
 	/*
 	 * List of kfd_process_device structures,
@@ -857,13 +899,13 @@ u32 read_kernel_doorbell(u32 __iomem *db);
 void write_kernel_doorbell(void __iomem *db, u32 value);
 void write_kernel_doorbell64(void __iomem *db, u64 value);
 unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd,
-					struct kfd_process *process,
+					struct kfd_process_device *pdd,
 					unsigned int doorbell_id);
-phys_addr_t kfd_get_process_doorbells(struct kfd_dev *dev,
-					struct kfd_process *process);
-int kfd_alloc_process_doorbells(struct kfd_process *process);
-void kfd_free_process_doorbells(struct kfd_process *process);
-
+phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd);
+int kfd_alloc_process_doorbells(struct kfd_dev *kfd,
+				unsigned int *doorbell_index);
+void kfd_free_process_doorbells(struct kfd_dev *kfd,
+				unsigned int doorbell_index);
 /* GTT Sub-Allocator */
 
 int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
@@ -892,6 +934,7 @@ struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev);
 struct kfd_dev *kfd_device_by_kgd(const struct kgd_dev *kgd);
 int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev);
 int kfd_numa_node_to_apic_id(int numa_node_id);
+void kfd_double_confirm_iommu_support(struct kfd_dev *gpu);
 
 /* Interrupts */
 int kfd_interrupt_init(struct kfd_dev *dev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 627793029033..65803e153a22 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -87,7 +87,7 @@ struct kfd_sdma_activity_handler_workarea {
 };
 
 struct temp_sdma_queue_list {
-	uint64_t rptr;
+	uint64_t __user *rptr;
 	uint64_t sdma_val;
 	unsigned int queue_id;
 	struct list_head list;
@@ -159,7 +159,7 @@ static void kfd_sdma_activity_worker(struct work_struct *work)
 		}
 
 		INIT_LIST_HEAD(&sdma_q->list);
-		sdma_q->rptr = (uint64_t)q->properties.read_ptr;
+		sdma_q->rptr = (uint64_t __user *)q->properties.read_ptr;
 		sdma_q->queue_id = q->properties.queue_id;
 		list_add_tail(&sdma_q->list, &sdma_q_list.list);
 	}
@@ -218,7 +218,7 @@ static void kfd_sdma_activity_worker(struct work_struct *work)
 			continue;
 
 		list_for_each_entry_safe(sdma_q, next, &sdma_q_list.list, list) {
-			if (((uint64_t)q->properties.read_ptr == sdma_q->rptr) &&
+			if (((uint64_t __user *)q->properties.read_ptr == sdma_q->rptr) &&
 			     (sdma_q->queue_id == q->properties.queue_id)) {
 				list_del(&sdma_q->list);
 				kfree(sdma_q);
@@ -249,6 +249,52 @@ cleanup:
 	}
 }
 
+/**
+ * @kfd_get_cu_occupancy() - Collect number of waves in-flight on this device
+ * by current process. Translates acquired wave count into number of compute units
+ * that are occupied.
+ *
+ * @atr: Handle of attribute that allows reporting of wave count. The attribute
+ * handle encapsulates GPU device it is associated with, thereby allowing collection
+ * of waves in flight, etc
+ *
+ * @buffer: Handle of user provided buffer updated with wave count
+ *
+ * Return: Number of bytes written to user buffer or an error value
+ */
+static int kfd_get_cu_occupancy(struct attribute *attr, char *buffer)
+{
+	int cu_cnt;
+	int wave_cnt;
+	int max_waves_per_cu;
+	struct kfd_dev *dev = NULL;
+	struct kfd_process *proc = NULL;
+	struct kfd_process_device *pdd = NULL;
+
+	pdd = container_of(attr, struct kfd_process_device, attr_cu_occupancy);
+	dev = pdd->dev;
+	if (dev->kfd2kgd->get_cu_occupancy == NULL)
+		return -EINVAL;
+
+	cu_cnt = 0;
+	proc = pdd->process;
+	if (pdd->qpd.queue_count == 0) {
+		pr_debug("Gpu-Id: %d has no active queues for process %d\n",
+			 dev->id, proc->pasid);
+		return snprintf(buffer, PAGE_SIZE, "%d\n", cu_cnt);
+	}
+
+	/* Collect wave count from device if it supports */
+	wave_cnt = 0;
+	max_waves_per_cu = 0;
+	dev->kfd2kgd->get_cu_occupancy(dev->kgd, proc->pasid, &wave_cnt,
+			&max_waves_per_cu);
+
+	/* Translate wave count to number of compute units */
+	cu_cnt = (wave_cnt + (max_waves_per_cu - 1)) / max_waves_per_cu;
+	return snprintf(buffer, PAGE_SIZE, "%d\n", cu_cnt);
+}
+
 static ssize_t kfd_procfs_show(struct kobject *kobj, struct attribute *attr,
 			       char *buffer)
 {
@@ -270,6 +316,7 @@ static ssize_t kfd_procfs_show(struct kobject *kobj, struct attribute *attr,
 					kfd_sdma_activity_worker);
 
 		sdma_activity_work_handler.pdd = pdd;
+		sdma_activity_work_handler.sdma_activity_counter = 0;
 
 		schedule_work(&sdma_activity_work_handler.sdma_activity_work);
 
@@ -344,6 +391,32 @@ static ssize_t kfd_procfs_queue_show(struct kobject *kobj,
 	return 0;
 }
 
+static ssize_t kfd_procfs_stats_show(struct kobject *kobj,
+				     struct attribute *attr, char *buffer)
+{
+	if (strcmp(attr->name, "evicted_ms") == 0) {
+		struct kfd_process_device *pdd = container_of(attr,
+				struct kfd_process_device,
+				attr_evict);
+		uint64_t evict_jiffies;
+
+		evict_jiffies = atomic64_read(&pdd->evict_duration_counter);
+
+		return snprintf(buffer,
+				PAGE_SIZE,
+				"%llu\n",
+				jiffies64_to_msecs(evict_jiffies));
+
+	/* Sysfs handle that gets CU occupancy is per device */
+	} else if (strcmp(attr->name, "cu_occupancy") == 0) {
+		return kfd_get_cu_occupancy(attr, buffer);
+	} else {
+		pr_err("Invalid attribute");
+	}
+
+	return 0;
+}
+
 static struct attribute attr_queue_size = {
 	.name = "size",
 	.mode = KFD_SYSFS_FILE_MODE
@@ -375,6 +448,19 @@ static struct kobj_type procfs_queue_type = {
 	.default_attrs = procfs_queue_attrs,
 };
 
+static const struct sysfs_ops procfs_stats_ops = {
+	.show = kfd_procfs_stats_show,
+};
+
+static struct attribute *procfs_stats_attrs[] = {
+	NULL
+};
+
+static struct kobj_type procfs_stats_type = {
+	.sysfs_ops = &procfs_stats_ops,
+	.default_attrs = procfs_stats_attrs,
+};
+
 int kfd_procfs_add_queue(struct queue *q)
 {
 	struct kfd_process *proc;
@@ -416,6 +502,72 @@ static int kfd_sysfs_create_file(struct kfd_process *p, struct attribute *attr,
 	return ret;
 }
 
+static int kfd_procfs_add_sysfs_stats(struct kfd_process *p)
+{
+	int ret = 0;
+	struct kfd_process_device *pdd;
+	char stats_dir_filename[MAX_SYSFS_FILENAME_LEN];
+
+	if (!p)
+		return -EINVAL;
+
+	if (!p->kobj)
+		return -EFAULT;
+
+	/*
+	 * Create sysfs files for each GPU:
+	 * - proc/<pid>/stats_<gpuid>/
+	 * - proc/<pid>/stats_<gpuid>/evicted_ms
+	 * - proc/<pid>/stats_<gpuid>/cu_occupancy
+	 */
+	list_for_each_entry(pdd, &p->per_device_data, per_device_list) {
+		struct kobject *kobj_stats;
+
+		snprintf(stats_dir_filename, MAX_SYSFS_FILENAME_LEN,
+				"stats_%u", pdd->dev->id);
+		kobj_stats = kfd_alloc_struct(kobj_stats);
+		if (!kobj_stats)
+			return -ENOMEM;
+
+		ret = kobject_init_and_add(kobj_stats,
+						&procfs_stats_type,
+						p->kobj,
+						stats_dir_filename);
+
+		if (ret) {
+			pr_warn("Creating KFD proc/stats_%s folder failed",
+					stats_dir_filename);
+			kobject_put(kobj_stats);
+			goto err;
+		}
+
+		pdd->kobj_stats = kobj_stats;
+		pdd->attr_evict.name = "evicted_ms";
+		pdd->attr_evict.mode = KFD_SYSFS_FILE_MODE;
+		sysfs_attr_init(&pdd->attr_evict);
+		ret = sysfs_create_file(kobj_stats, &pdd->attr_evict);
+		if (ret)
+			pr_warn("Creating eviction stats for gpuid %d failed",
+					(int)pdd->dev->id);
+
+		/* Add sysfs file to report compute unit occupancy */
+		if (pdd->dev->kfd2kgd->get_cu_occupancy != NULL) {
+			pdd->attr_cu_occupancy.name = "cu_occupancy";
+			pdd->attr_cu_occupancy.mode = KFD_SYSFS_FILE_MODE;
+			sysfs_attr_init(&pdd->attr_cu_occupancy);
+			ret = sysfs_create_file(kobj_stats,
+						&pdd->attr_cu_occupancy);
+			if (ret)
+				pr_warn("Creating %s failed for gpuid: %d",
+					pdd->attr_cu_occupancy.name,
+					(int)pdd->dev->id);
+		}
+	}
+err:
+	return ret;
+}
+
+
 static int kfd_procfs_add_sysfs_files(struct kfd_process *p)
 {
 	int ret = 0;
@@ -451,7 +603,6 @@ static int kfd_procfs_add_sysfs_files(struct kfd_process *p)
 	return ret;
 }
 
-
 void kfd_procfs_del_queue(struct queue *q)
 {
 	if (!q)
@@ -659,6 +810,11 @@ struct kfd_process *kfd_create_process(struct file *filep)
 		if (!process->kobj_queues)
 			pr_warn("Creating KFD proc/queues folder failed");
 
+		ret = kfd_procfs_add_sysfs_stats(process);
+		if (ret)
+			pr_warn("Creating sysfs stats dir for pid %d failed",
+				(int)process->lead_thread->pid);
+
 		ret = kfd_procfs_add_sysfs_files(process);
 		if (ret)
 			pr_warn("Creating sysfs usage file for pid %d failed",
@@ -780,6 +936,8 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 		kfree(pdd->qpd.doorbell_bitmap);
 		idr_destroy(&pdd->alloc_idr);
 
+		kfd_free_process_doorbells(pdd->dev, pdd->doorbell_index);
+
 		/*
 		 * before destroying pdd, make sure to report availability
 		 * for auto suspend
@@ -815,6 +973,12 @@ static void kfd_process_wq_release(struct work_struct *work)
 		list_for_each_entry(pdd, &p->per_device_data, per_device_list) {
 			sysfs_remove_file(p->kobj, &pdd->attr_vram);
 			sysfs_remove_file(p->kobj, &pdd->attr_sdma);
+			sysfs_remove_file(p->kobj, &pdd->attr_evict);
+			if (pdd->dev->kfd2kgd->get_cu_occupancy != NULL)
+				sysfs_remove_file(p->kobj, &pdd->attr_cu_occupancy);
+			kobject_del(pdd->kobj_stats);
+			kobject_put(pdd->kobj_stats);
+			pdd->kobj_stats = NULL;
 		}
 
 		kobject_del(p->kobj);
@@ -832,8 +996,6 @@ static void kfd_process_wq_release(struct work_struct *work)
 	kfd_event_free_process(p);
 
 	kfd_pasid_free(p->pasid);
-	kfd_free_process_doorbells(p);
-
 	mutex_destroy(&p->mutex);
 
 	put_task_struct(p->lead_thread);
@@ -1011,9 +1173,6 @@ static struct kfd_process *create_process(const struct task_struct *thread)
 	if (process->pasid == 0)
 		goto err_alloc_pasid;
 
-	if (kfd_alloc_process_doorbells(process) < 0)
-		goto err_alloc_doorbells;
-
 	err = pqm_init(&process->pqm, process);
 	if (err != 0)
 		goto err_process_pqm_init;
@@ -1041,8 +1200,6 @@ err_register_notifier:
 err_init_apertures:
 	pqm_uninit(&process->pqm);
 err_process_pqm_init:
-	kfd_free_process_doorbells(process);
-err_alloc_doorbells:
 	kfd_pasid_free(process->pasid);
 err_alloc_pasid:
 	mutex_destroy(&process->mutex);
@@ -1105,10 +1262,14 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 	if (!pdd)
 		return NULL;
 
+	if (kfd_alloc_process_doorbells(dev, &pdd->doorbell_index) < 0) {
+		pr_err("Failed to alloc doorbell for pdd\n");
+		goto err_free_pdd;
+	}
+
 	if (init_doorbell_bitmap(&pdd->qpd, dev)) {
 		pr_err("Failed to init doorbell for process\n");
-		kfree(pdd);
-		return NULL;
+		goto err_free_pdd;
 	}
 
 	pdd->dev = dev;
@@ -1124,12 +1285,17 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 	pdd->runtime_inuse = false;
 	pdd->vram_usage = 0;
 	pdd->sdma_past_activity_counter = 0;
+	atomic64_set(&pdd->evict_duration_counter, 0);
 	list_add(&pdd->per_device_list, &p->per_device_data);
 
 	/* Init idr used for memory handle translation */
 	idr_init(&pdd->alloc_idr);
 
 	return pdd;
+
+err_free_pdd:
+	kfree(pdd);
+	return NULL;
 }
 
 /**
@@ -1487,6 +1653,7 @@ void kfd_suspend_all_processes(void)
 	unsigned int temp;
 	int idx = srcu_read_lock(&kfd_processes_srcu);
 
+	WARN(debug_evictions, "Evicting all processes");
 	hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
 		cancel_delayed_work_sync(&p->eviction_work);
 		cancel_delayed_work_sync(&p->restore_work);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index 7b348bf9df21..17d1736367ea 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -24,6 +24,7 @@
 #include <linux/wait.h>
 #include <linux/anon_inodes.h>
 #include <uapi/linux/kfd_ioctl.h>
+#include "amdgpu.h"
 #include "amdgpu_vm.h"
 #include "kfd_priv.h"
 #include "kfd_smi_events.h"
@@ -148,15 +149,94 @@ static int kfd_smi_ev_release(struct inode *inode, struct file *filep)
 	return 0;
 }
 
+static void add_event_to_kfifo(struct kfd_dev *dev, unsigned int smi_event,
+			      char *event_msg, int len)
+{
+	struct kfd_smi_client *client;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(client, &dev->smi_clients, list) {
+		if (!(READ_ONCE(client->events) &
+				KFD_SMI_EVENT_MASK_FROM_INDEX(smi_event)))
+			continue;
+		spin_lock(&client->lock);
+		if (kfifo_avail(&client->fifo) >= len) {
+			kfifo_in(&client->fifo, event_msg, len);
+			wake_up_all(&client->wait_queue);
+		} else {
+			pr_debug("smi_event(EventID: %u): no space left\n",
+					smi_event);
+		}
+		spin_unlock(&client->lock);
+	}
+
+	rcu_read_unlock();
+}
+
+void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset)
+{
+	/*
+	 * GpuReset msg = Reset seq number (incremented for
+	 * every reset message sent before GPU reset).
+	 * 1 byte event + 1 byte space + 8 bytes seq num +
+	 * 1 byte \n + 1 byte \0 = 12
+	 */
+	char fifo_in[12];
+	int len;
+	unsigned int event;
+
+	if (list_empty(&dev->smi_clients))
+		return;
+
+	memset(fifo_in, 0x0, sizeof(fifo_in));
+
+	if (post_reset) {
+		event = KFD_SMI_EVENT_GPU_POST_RESET;
+	} else {
+		event = KFD_SMI_EVENT_GPU_PRE_RESET;
+		++(dev->reset_seq_num);
+	}
+
+	len = snprintf(fifo_in, sizeof(fifo_in), "%x %x\n", event,
+						dev->reset_seq_num);
+
+	add_event_to_kfifo(dev, event, fifo_in, len);
+}
+
+void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev,
+					     uint32_t throttle_bitmask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)dev->kgd;
+	/*
+	 * ThermalThrottle msg = throttle_bitmask(8):
+	 * 			 thermal_interrupt_count(16):
+	 * 1 byte event + 1 byte space + 8 byte throttle_bitmask +
+	 * 1 byte : + 16 byte thermal_interupt_counter + 1 byte \n +
+	 * 1 byte \0 = 29
+	 */
+	char fifo_in[29];
+	int len;
+
+	if (list_empty(&dev->smi_clients))
+		return;
+
+	len = snprintf(fifo_in, sizeof(fifo_in), "%x %x:%llx\n",
+		       KFD_SMI_EVENT_THERMAL_THROTTLE, throttle_bitmask,
+		       atomic64_read(&adev->smu.throttle_int_counter));
+
+	add_event_to_kfifo(dev, KFD_SMI_EVENT_THERMAL_THROTTLE,	fifo_in, len);
+}
+
 void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)dev->kgd;
 	struct amdgpu_task_info task_info;
 	/* VmFault msg = (hex)uint32_pid(8) + :(1) + task name(16) = 25 */
-	/* 16 bytes event + 1 byte space + 25 bytes msg + 1 byte \n = 43
+	/* 1 byte event + 1 byte space + 25 bytes msg + 1 byte \n +
+	 * 1 byte \0 = 29
 	 */
-	char fifo_in[43];
-	struct kfd_smi_client *client;
+	char fifo_in[29];
 	int len;
 
 	if (list_empty(&dev->smi_clients))
@@ -168,25 +248,10 @@ void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid)
 	if (!task_info.pid)
 		return;
 
-	len = snprintf(fifo_in, 43, "%x %x:%s\n", KFD_SMI_EVENT_VMFAULT,
+	len = snprintf(fifo_in, sizeof(fifo_in), "%x %x:%s\n", KFD_SMI_EVENT_VMFAULT,
 		task_info.pid, task_info.task_name);
 
-	rcu_read_lock();
-
-	list_for_each_entry_rcu(client, &dev->smi_clients, list) {
-		if (!(READ_ONCE(client->events) & KFD_SMI_EVENT_VMFAULT))
-			continue;
-		spin_lock(&client->lock);
-		if (kfifo_avail(&client->fifo) >= len) {
-			kfifo_in(&client->fifo, fifo_in, len);
-			wake_up_all(&client->wait_queue);
-		}
-		else
-			pr_debug("smi_event(vmfault): no space left\n");
-		spin_unlock(&client->lock);
-	}
-
-	rcu_read_unlock();
+	add_event_to_kfifo(dev, KFD_SMI_EVENT_VMFAULT, fifo_in, len);
 }
 
 int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
index a9cb218fef96..b9b0438202e2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
@@ -25,5 +25,8 @@
 
 int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd);
 void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid);
+void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev,
+					     uint32_t throttle_bitmask);
+void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index f185f6cbc05c..2b31c3066aaa 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -446,7 +446,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 	sysfs_show_32bit_prop(buffer, offs, "cpu_cores_count",
 			      dev->node_props.cpu_cores_count);
 	sysfs_show_32bit_prop(buffer, offs, "simd_count",
-			      dev->node_props.simd_count);
+			      dev->gpu ? dev->node_props.simd_count : 0);
 	sysfs_show_32bit_prop(buffer, offs, "mem_banks_count",
 			      dev->node_props.mem_banks_count);
 	sysfs_show_32bit_prop(buffer, offs, "caches_count",
@@ -1139,7 +1139,7 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 		/* Discrete GPUs need their own topology device list
 		 * entries. Don't assign them to CPU/APU nodes.
 		 */
-		if (!gpu->device_info->needs_iommu_device &&
+		if (!gpu->use_iommu_v2 &&
 		    dev->node_props.cpu_cores_count)
 			continue;
 
@@ -1239,7 +1239,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	void *crat_image = NULL;
 	size_t image_size = 0;
 	int proximity_domain;
-	struct amdgpu_ras *ctx;
+	struct amdgpu_device *adev;
 
 	INIT_LIST_HEAD(&temp_topology_device_list);
 
@@ -1388,7 +1388,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	* Overwrite ATS capability according to needs_iommu_device to fix
 	* potential missing corresponding bit in CRAT of BIOS.
 	*/
-	if (dev->gpu->device_info->needs_iommu_device)
+	if (dev->gpu->use_iommu_v2)
 		dev->node_props.capability |= HSA_CAP_ATS_PRESENT;
 	else
 		dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT;
@@ -1404,19 +1404,17 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 		dev->node_props.max_waves_per_simd = 10;
 	}
 
-	ctx = amdgpu_ras_get_context((struct amdgpu_device *)(dev->gpu->kgd));
-	if (ctx) {
-		/* kfd only concerns sram ecc on GFX/SDMA and HBM ecc on UMC */
-		dev->node_props.capability |=
-			(((ctx->features & BIT(AMDGPU_RAS_BLOCK__SDMA)) != 0) ||
-			 ((ctx->features & BIT(AMDGPU_RAS_BLOCK__GFX)) != 0)) ?
-			HSA_CAP_SRAM_EDCSUPPORTED : 0;
-		dev->node_props.capability |= ((ctx->features & BIT(AMDGPU_RAS_BLOCK__UMC)) != 0) ?
-			HSA_CAP_MEM_EDCSUPPORTED : 0;
-
-		dev->node_props.capability |= (ctx->features != 0) ?
+	adev = (struct amdgpu_device *)(dev->gpu->kgd);
+	/* kfd only concerns sram ecc on GFX and HBM ecc on UMC */
+	dev->node_props.capability |=
+		((adev->ras_features & BIT(AMDGPU_RAS_BLOCK__GFX)) != 0) ?
+		HSA_CAP_SRAM_EDCSUPPORTED : 0;
+	dev->node_props.capability |= ((adev->ras_features & BIT(AMDGPU_RAS_BLOCK__UMC)) != 0) ?
+		HSA_CAP_MEM_EDCSUPPORTED : 0;
+
+	if (adev->asic_type != CHIP_VEGA10)
+		dev->node_props.capability |= (adev->ras_features != 0) ?
 			HSA_CAP_RASEVENTNOTIFY : 0;
-	}
 
 	kfd_debug_print_topology();
 
@@ -1515,6 +1513,29 @@ int kfd_numa_node_to_apic_id(int numa_node_id)
 	return kfd_cpumask_to_apic_id(cpumask_of_node(numa_node_id));
 }
 
+void kfd_double_confirm_iommu_support(struct kfd_dev *gpu)
+{
+	struct kfd_topology_device *dev;
+
+	gpu->use_iommu_v2 = false;
+
+	if (!gpu->device_info->needs_iommu_device)
+		return;
+
+	down_read(&topology_lock);
+
+	/* Only use IOMMUv2 if there is an APU topology node with no GPU
+	 * assigned yet. This GPU will be assigned to it.
+	 */
+	list_for_each_entry(dev, &topology_device_list, list)
+		if (dev->node_props.cpu_cores_count &&
+		    dev->node_props.simd_count &&
+		    !dev->gpu)
+			gpu->use_iommu_v2 = true;
+
+	up_read(&topology_lock);
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 int kfd_debugfs_hqds_by_device(struct seq_file *m, void *data)
diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig
index 34ae4f3a32f4..f24abf428534 100644
--- a/drivers/gpu/drm/amd/display/Kconfig
+++ b/drivers/gpu/drm/amd/display/Kconfig
@@ -6,7 +6,7 @@ config DRM_AMD_DC
 	bool "AMD DC - Enable new display engine"
 	default y
 	select SND_HDA_COMPONENT if SND_HDA_CORE
-	select DRM_AMD_DC_DCN if (X86 || PPC64) && !(KCOV_INSTRUMENT_ALL && KCOV_ENABLE_COMPARISONS)
+	select DRM_AMD_DC_DCN if (X86 || PPC64 || (ARM64 && KERNEL_MODE_NEON)) && !(KCOV_INSTRUMENT_ALL && KCOV_ENABLE_COMPARISONS)
 	help
 	  Choose this option if you want to use the new display engine
 	  support for AMDGPU. This adds required support for Vega and
@@ -31,6 +31,14 @@ config DRM_AMD_DC_HDCP
 	help
 	  Choose this option if you want to support HDCP authentication.
 
+config DRM_AMD_DC_SI
+	bool "AMD DC support for Southern Islands ASICs"
+	default n
+	help
+	  Choose this option to enable new AMD DC support for SI asics
+	  by default. This includes Tahiti, Pitcairn, Cape Verde, Oland.
+	  Hainan is not supported by AMD DC and it has no physical DCE6.
+
 config DEBUG_KERNEL_DC
 	bool "Enable kgdb break in DC"
 	depends on DRM_AMD_DC
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index a717a4904268..bb1bc7f5d149 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -127,6 +127,42 @@ MODULE_FIRMWARE(FIRMWARE_NAVI12_DMCU);
 static int amdgpu_dm_init(struct amdgpu_device *adev);
 static void amdgpu_dm_fini(struct amdgpu_device *adev);
 
+static enum drm_mode_subconnector get_subconnector_type(struct dc_link *link)
+{
+	switch (link->dpcd_caps.dongle_type) {
+	case DISPLAY_DONGLE_NONE:
+		return DRM_MODE_SUBCONNECTOR_Native;
+	case DISPLAY_DONGLE_DP_VGA_CONVERTER:
+		return DRM_MODE_SUBCONNECTOR_VGA;
+	case DISPLAY_DONGLE_DP_DVI_CONVERTER:
+	case DISPLAY_DONGLE_DP_DVI_DONGLE:
+		return DRM_MODE_SUBCONNECTOR_DVID;
+	case DISPLAY_DONGLE_DP_HDMI_CONVERTER:
+	case DISPLAY_DONGLE_DP_HDMI_DONGLE:
+		return DRM_MODE_SUBCONNECTOR_HDMIA;
+	case DISPLAY_DONGLE_DP_HDMI_MISMATCHED_DONGLE:
+	default:
+		return DRM_MODE_SUBCONNECTOR_Unknown;
+	}
+}
+
+static void update_subconnector_property(struct amdgpu_dm_connector *aconnector)
+{
+	struct dc_link *link = aconnector->dc_link;
+	struct drm_connector *connector = &aconnector->base;
+	enum drm_mode_subconnector subconnector = DRM_MODE_SUBCONNECTOR_Unknown;
+
+	if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort)
+		return;
+
+	if (aconnector->dc_sink)
+		subconnector = get_subconnector_type(link);
+
+	drm_object_property_set_value(&connector->base,
+			connector->dev->mode_config.dp_subconnector_property,
+			subconnector);
+}
+
 /*
  * initializes drm_device display related structures, based on the information
  * provided by DAL. The drm strcutures are: drm_crtc, drm_connector,
@@ -171,7 +207,7 @@ static void amdgpu_dm_set_psr_caps(struct dc_link *link);
 static bool amdgpu_dm_psr_enable(struct dc_stream_state *stream);
 static bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream);
 static bool amdgpu_dm_psr_disable(struct dc_stream_state *stream);
-
+static bool amdgpu_dm_psr_disable_all(struct amdgpu_display_manager *dm);
 
 /*
  * dm_vblank_get_counter
@@ -192,17 +228,14 @@ static u32 dm_vblank_get_counter(struct amdgpu_device *adev, int crtc)
 		return 0;
 	else {
 		struct amdgpu_crtc *acrtc = adev->mode_info.crtcs[crtc];
-		struct dm_crtc_state *acrtc_state = to_dm_crtc_state(
-				acrtc->base.state);
-
 
-		if (acrtc_state->stream == NULL) {
+		if (acrtc->dm_irq_params.stream == NULL) {
 			DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n",
 				  crtc);
 			return 0;
 		}
 
-		return dc_stream_get_vblank_counter(acrtc_state->stream);
+		return dc_stream_get_vblank_counter(acrtc->dm_irq_params.stream);
 	}
 }
 
@@ -215,10 +248,8 @@ static int dm_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
 		return -EINVAL;
 	else {
 		struct amdgpu_crtc *acrtc = adev->mode_info.crtcs[crtc];
-		struct dm_crtc_state *acrtc_state = to_dm_crtc_state(
-						acrtc->base.state);
 
-		if (acrtc_state->stream ==  NULL) {
+		if (acrtc->dm_irq_params.stream ==  NULL) {
 			DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n",
 				  crtc);
 			return 0;
@@ -228,7 +259,7 @@ static int dm_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
 		 * TODO rework base driver to use values directly.
 		 * for now parse it back into reg-format
 		 */
-		dc_stream_get_scanoutpos(acrtc_state->stream,
+		dc_stream_get_scanoutpos(acrtc->dm_irq_params.stream,
 					 &v_blank_start,
 					 &v_blank_end,
 					 &h_position,
@@ -268,7 +299,7 @@ static struct amdgpu_crtc *
 get_crtc_by_otg_inst(struct amdgpu_device *adev,
 		     int otg_inst)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_crtc *crtc;
 	struct amdgpu_crtc *amdgpu_crtc;
 
@@ -287,6 +318,14 @@ get_crtc_by_otg_inst(struct amdgpu_device *adev,
 	return NULL;
 }
 
+static inline bool amdgpu_dm_vrr_active_irq(struct amdgpu_crtc *acrtc)
+{
+	return acrtc->dm_irq_params.freesync_config.state ==
+		       VRR_STATE_ACTIVE_VARIABLE ||
+	       acrtc->dm_irq_params.freesync_config.state ==
+		       VRR_STATE_ACTIVE_FIXED;
+}
+
 static inline bool amdgpu_dm_vrr_active(struct dm_crtc_state *dm_state)
 {
 	return dm_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE ||
@@ -307,7 +346,6 @@ static void dm_pflip_high_irq(void *interrupt_params)
 	struct amdgpu_device *adev = irq_params->adev;
 	unsigned long flags;
 	struct drm_pending_vblank_event *e;
-	struct dm_crtc_state *acrtc_state;
 	uint32_t vpos, hpos, v_blank_start, v_blank_end;
 	bool vrr_active;
 
@@ -320,7 +358,7 @@ static void dm_pflip_high_irq(void *interrupt_params)
 		return;
 	}
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 
 	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d !=AMDGPU_FLIP_SUBMITTED(%d) on crtc:%d[%p] \n",
@@ -328,7 +366,7 @@ static void dm_pflip_high_irq(void *interrupt_params)
 						 AMDGPU_FLIP_SUBMITTED,
 						 amdgpu_crtc->crtc_id,
 						 amdgpu_crtc);
-		spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 		return;
 	}
 
@@ -339,12 +377,11 @@ static void dm_pflip_high_irq(void *interrupt_params)
 	if (!e)
 		WARN_ON(1);
 
-	acrtc_state = to_dm_crtc_state(amdgpu_crtc->base.state);
-	vrr_active = amdgpu_dm_vrr_active(acrtc_state);
+	vrr_active = amdgpu_dm_vrr_active_irq(amdgpu_crtc);
 
 	/* Fixed refresh rate, or VRR scanout position outside front-porch? */
 	if (!vrr_active ||
-	    !dc_stream_get_scanoutpos(acrtc_state->stream, &v_blank_start,
+	    !dc_stream_get_scanoutpos(amdgpu_crtc->dm_irq_params.stream, &v_blank_start,
 				      &v_blank_end, &hpos, &vpos) ||
 	    (vpos < v_blank_start)) {
 		/* Update to correct count and vblank timestamp if racing with
@@ -380,7 +417,7 @@ static void dm_pflip_high_irq(void *interrupt_params)
 		e->sequence = drm_crtc_vblank_count(&amdgpu_crtc->base);
 		e->pipe = amdgpu_crtc->crtc_id;
 
-		list_add_tail(&e->base.link, &adev->ddev->vblank_event_list);
+		list_add_tail(&e->base.link, &adev_to_drm(adev)->vblank_event_list);
 		e = NULL;
 	}
 
@@ -389,11 +426,11 @@ static void dm_pflip_high_irq(void *interrupt_params)
 	 * of pageflip completion, so last_flip_vblank is the forbidden count
 	 * for queueing new pageflips if vsync + VRR is enabled.
 	 */
-	amdgpu_crtc->last_flip_vblank =
+	amdgpu_crtc->dm_irq_params.last_flip_vblank =
 		amdgpu_get_vblank_counter_kms(&amdgpu_crtc->base);
 
 	amdgpu_crtc->pflip_status = AMDGPU_FLIP_NONE;
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	DRM_DEBUG_DRIVER("crtc:%d[%p], pflip_stat:AMDGPU_FLIP_NONE, vrr[%d]-fp %d\n",
 			 amdgpu_crtc->crtc_id, amdgpu_crtc,
@@ -405,17 +442,17 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 	struct common_irq_params *irq_params = interrupt_params;
 	struct amdgpu_device *adev = irq_params->adev;
 	struct amdgpu_crtc *acrtc;
-	struct dm_crtc_state *acrtc_state;
 	unsigned long flags;
+	int vrr_active;
 
 	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VUPDATE);
 
 	if (acrtc) {
-		acrtc_state = to_dm_crtc_state(acrtc->base.state);
+		vrr_active = amdgpu_dm_vrr_active_irq(acrtc);
 
 		DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n",
 			      acrtc->crtc_id,
-			      amdgpu_dm_vrr_active(acrtc_state));
+			      vrr_active);
 
 		/* Core vblank handling is done here after end of front-porch in
 		 * vrr mode, as vblank timestamping will give valid results
@@ -423,23 +460,23 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 		 * page-flip completion events that have been queued to us
 		 * if a pageflip happened inside front-porch.
 		 */
-		if (amdgpu_dm_vrr_active(acrtc_state)) {
+		if (vrr_active) {
 			drm_crtc_handle_vblank(&acrtc->base);
 
 			/* BTR processing for pre-DCE12 ASICs */
-			if (acrtc_state->stream &&
+			if (acrtc->dm_irq_params.stream &&
 			    adev->family < AMDGPU_FAMILY_AI) {
-				spin_lock_irqsave(&adev->ddev->event_lock, flags);
+				spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 				mod_freesync_handle_v_update(
 				    adev->dm.freesync_module,
-				    acrtc_state->stream,
-				    &acrtc_state->vrr_params);
+				    acrtc->dm_irq_params.stream,
+				    &acrtc->dm_irq_params.vrr_params);
 
 				dc_stream_adjust_vmin_vmax(
 				    adev->dm.dc,
-				    acrtc_state->stream,
-				    &acrtc_state->vrr_params.adjust);
-				spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+				    acrtc->dm_irq_params.stream,
+				    &acrtc->dm_irq_params.vrr_params.adjust);
+				spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 			}
 		}
 	}
@@ -457,18 +494,17 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	struct common_irq_params *irq_params = interrupt_params;
 	struct amdgpu_device *adev = irq_params->adev;
 	struct amdgpu_crtc *acrtc;
-	struct dm_crtc_state *acrtc_state;
 	unsigned long flags;
+	int vrr_active;
 
 	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK);
 	if (!acrtc)
 		return;
 
-	acrtc_state = to_dm_crtc_state(acrtc->base.state);
+	vrr_active = amdgpu_dm_vrr_active_irq(acrtc);
 
 	DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d, planes:%d\n", acrtc->crtc_id,
-			 amdgpu_dm_vrr_active(acrtc_state),
-			 acrtc_state->active_planes);
+		      vrr_active, acrtc->dm_irq_params.active_planes);
 
 	/**
 	 * Core vblank handling at start of front-porch is only possible
@@ -476,7 +512,7 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	 * valid results while done in front-porch. Otherwise defer it
 	 * to dm_vupdate_high_irq after end of front-porch.
 	 */
-	if (!amdgpu_dm_vrr_active(acrtc_state))
+	if (!vrr_active)
 		drm_crtc_handle_vblank(&acrtc->base);
 
 	/**
@@ -489,16 +525,18 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	if (adev->family < AMDGPU_FAMILY_AI)
 		return;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 
-	if (acrtc_state->stream && acrtc_state->vrr_params.supported &&
-	    acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) {
+	if (acrtc->dm_irq_params.stream &&
+	    acrtc->dm_irq_params.vrr_params.supported &&
+	    acrtc->dm_irq_params.freesync_config.state ==
+		    VRR_STATE_ACTIVE_VARIABLE) {
 		mod_freesync_handle_v_update(adev->dm.freesync_module,
-					     acrtc_state->stream,
-					     &acrtc_state->vrr_params);
+					     acrtc->dm_irq_params.stream,
+					     &acrtc->dm_irq_params.vrr_params);
 
-		dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc_state->stream,
-					   &acrtc_state->vrr_params.adjust);
+		dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc->dm_irq_params.stream,
+					   &acrtc->dm_irq_params.vrr_params.adjust);
 	}
 
 	/*
@@ -513,7 +551,7 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	 */
 	if (adev->family >= AMDGPU_FAMILY_RV &&
 	    acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED &&
-	    acrtc_state->active_planes == 0) {
+	    acrtc->dm_irq_params.active_planes == 0) {
 		if (acrtc->event) {
 			drm_crtc_send_vblank_event(&acrtc->base, acrtc->event);
 			acrtc->event = NULL;
@@ -522,7 +560,7 @@ static void dm_crtc_high_irq(void *interrupt_params)
 		acrtc->pflip_status = AMDGPU_FLIP_NONE;
 	}
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 }
 
 static int dm_set_clockgating_state(void *handle,
@@ -544,7 +582,7 @@ static int dm_early_init(void* handle);
 static void amdgpu_dm_fbc_init(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct dm_comressor_info *compressor = &adev->dm.compressor;
 	struct amdgpu_dm_connector *aconn = to_amdgpu_dm_connector(connector);
 	struct drm_display_mode *mode;
@@ -586,7 +624,7 @@ static int amdgpu_dm_audio_component_get_eld(struct device *kdev, int port,
 					  unsigned char *buf, int max_bytes)
 {
 	struct drm_device *dev = dev_get_drvdata(kdev);
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 	struct amdgpu_dm_connector *aconnector;
@@ -625,7 +663,7 @@ static int amdgpu_dm_audio_component_bind(struct device *kdev,
 				       struct device *hda_kdev, void *data)
 {
 	struct drm_device *dev = dev_get_drvdata(kdev);
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_audio_component *acomp = data;
 
 	acomp->ops = &amdgpu_dm_audio_component_ops;
@@ -639,7 +677,7 @@ static void amdgpu_dm_audio_component_unbind(struct device *kdev,
 					  struct device *hda_kdev, void *data)
 {
 	struct drm_device *dev = dev_get_drvdata(kdev);
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_audio_component *acomp = data;
 
 	acomp->ops = NULL;
@@ -842,6 +880,45 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 	return 0;
 }
 
+static void amdgpu_check_debugfs_connector_property_change(struct amdgpu_device *adev,
+							   struct drm_atomic_state *state)
+{
+	struct drm_connector *connector;
+	struct drm_crtc *crtc;
+	struct amdgpu_dm_connector *amdgpu_dm_connector;
+	struct drm_connector_state *conn_state;
+	struct dm_crtc_state *acrtc_state;
+	struct drm_crtc_state *crtc_state;
+	struct dc_stream_state *stream;
+	struct drm_device *dev = adev_to_drm(adev);
+
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+
+		amdgpu_dm_connector = to_amdgpu_dm_connector(connector);
+		conn_state = connector->state;
+
+		if (!(conn_state && conn_state->crtc))
+			continue;
+
+		crtc = conn_state->crtc;
+		acrtc_state = to_dm_crtc_state(crtc->state);
+
+		if (!(acrtc_state && acrtc_state->stream))
+			continue;
+
+		stream = acrtc_state->stream;
+
+		if (amdgpu_dm_connector->dsc_settings.dsc_force_enable ||
+		    amdgpu_dm_connector->dsc_settings.dsc_num_slices_v ||
+		    amdgpu_dm_connector->dsc_settings.dsc_num_slices_h ||
+		    amdgpu_dm_connector->dsc_settings.dsc_bits_per_pixel) {
+			conn_state = drm_atomic_get_connector_state(state, connector);
+			crtc_state = drm_atomic_get_crtc_state(state, crtc);
+			crtc_state->mode_changed = true;
+		}
+	}
+}
+
 static int amdgpu_dm_init(struct amdgpu_device *adev)
 {
 	struct dc_init_data init_data;
@@ -850,7 +927,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 #endif
 	int r;
 
-	adev->dm.ddev = adev->ddev;
+	adev->dm.ddev = adev_to_drm(adev);
 	adev->dm.adev = adev;
 
 	/* Zero all the fields */
@@ -986,10 +1063,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	/* TODO: Add_display_info? */
 
 	/* TODO use dynamic cursor width */
-	adev->ddev->mode_config.cursor_width = adev->dm.dc->caps.max_cursor_size;
-	adev->ddev->mode_config.cursor_height = adev->dm.dc->caps.max_cursor_size;
+	adev_to_drm(adev)->mode_config.cursor_width = adev->dm.dc->caps.max_cursor_size;
+	adev_to_drm(adev)->mode_config.cursor_height = adev->dm.dc->caps.max_cursor_size;
 
-	if (drm_vblank_init(adev->ddev, adev->dm.display_indexes_num)) {
+	if (drm_vblank_init(adev_to_drm(adev), adev->dm.display_indexes_num)) {
 		DRM_ERROR(
 		"amdgpu: failed to initialize sw for display support.\n");
 		goto error;
@@ -1066,6 +1143,12 @@ static int load_dmcu_fw(struct amdgpu_device *adev)
 	const struct dmcu_firmware_header_v1_0 *hdr;
 
 	switch(adev->asic_type) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+#endif
 	case CHIP_BONAIRE:
 	case CHIP_HAWAII:
 	case CHIP_KAVERI:
@@ -1383,9 +1466,6 @@ static int dm_late_init(void *handle)
 	struct dmcu *dmcu = NULL;
 	bool ret = true;
 
-	if (!adev->dm.fw_dmcu && !adev->dm.dmub_fw)
-		return detect_mst_link_for_all_connectors(adev->ddev);
-
 	dmcu = adev->dm.dc->res_pool->dmcu;
 
 	for (i = 0; i < 16; i++)
@@ -1414,7 +1494,7 @@ static int dm_late_init(void *handle)
 	if (!ret)
 		return -EINVAL;
 
-	return detect_mst_link_for_all_connectors(adev->ddev);
+	return detect_mst_link_for_all_connectors(adev_to_drm(adev));
 }
 
 static void s3_handle_mst(struct drm_device *dev, bool suspend)
@@ -1652,7 +1732,7 @@ static int dm_suspend(void *handle)
 	struct amdgpu_display_manager *dm = &adev->dm;
 	int ret = 0;
 
-	if (adev->in_gpu_reset) {
+	if (amdgpu_in_reset(adev)) {
 		mutex_lock(&dm->dc_lock);
 		dm->cached_dc_state = dc_copy_state(dm->dc->current_state);
 
@@ -1666,9 +1746,9 @@ static int dm_suspend(void *handle)
 	}
 
 	WARN_ON(adev->dm.cached_state);
-	adev->dm.cached_state = drm_atomic_helper_suspend(adev->ddev);
+	adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev));
 
-	s3_handle_mst(adev->ddev, true);
+	s3_handle_mst(adev_to_drm(adev), true);
 
 	amdgpu_dm_irq_suspend(adev);
 
@@ -1822,7 +1902,7 @@ cleanup:
 static int dm_resume(void *handle)
 {
 	struct amdgpu_device *adev = handle;
-	struct drm_device *ddev = adev->ddev;
+	struct drm_device *ddev = adev_to_drm(adev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_connector *connector;
@@ -1838,7 +1918,7 @@ static int dm_resume(void *handle)
 	struct dc_state *dc_state;
 	int i, r, j;
 
-	if (adev->in_gpu_reset) {
+	if (amdgpu_in_reset(adev)) {
 		dc_state = dm->cached_dc_state;
 
 		r = dm_dmub_hw_init(adev);
@@ -2044,7 +2124,7 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 		return;
 
 	conn_base = &aconnector->base;
-	adev = conn_base->dev->dev_private;
+	adev = drm_to_adev(conn_base->dev);
 	dm = &adev->dm;
 	caps = &dm->backlight_caps;
 	caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps;
@@ -2095,7 +2175,6 @@ void amdgpu_dm_update_connector_after_detect(
 	if (aconnector->mst_mgr.mst_state == true)
 		return;
 
-
 	sink = aconnector->dc_link->local_sink;
 	if (sink)
 		dc_sink_retain(sink);
@@ -2222,6 +2301,8 @@ void amdgpu_dm_update_connector_after_detect(
 
 	mutex_unlock(&dev->mode_config.mutex);
 
+	update_subconnector_property(aconnector);
+
 	if (sink)
 		dc_sink_release(sink);
 }
@@ -2233,7 +2314,7 @@ static void handle_hpd_irq(void *param)
 	struct drm_device *dev = connector->dev;
 	enum dc_connection_type new_connection_type = dc_connection_none;
 #ifdef CONFIG_DRM_AMD_DC_HDCP
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 #endif
 
 	/*
@@ -2366,7 +2447,7 @@ static void handle_hpd_rx_irq(void *param)
 	enum dc_connection_type new_connection_type = dc_connection_none;
 #ifdef CONFIG_DRM_AMD_DC_HDCP
 	union hpd_irq_data hpd_irq_data;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	memset(&hpd_irq_data, 0, sizeof(hpd_irq_data));
 #endif
@@ -2437,7 +2518,7 @@ static void handle_hpd_rx_irq(void *param)
 
 static void register_hpd_handlers(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct amdgpu_dm_connector *aconnector;
 	const struct dc_link *dc_link;
@@ -2474,6 +2555,89 @@ static void register_hpd_handlers(struct amdgpu_device *adev)
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/* Register IRQ sources and initialize IRQ callbacks */
+static int dce60_register_irq_handlers(struct amdgpu_device *adev)
+{
+	struct dc *dc = adev->dm.dc;
+	struct common_irq_params *c_irq_params;
+	struct dc_interrupt_params int_params = {0};
+	int r;
+	int i;
+	unsigned client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
+
+	int_params.requested_polarity = INTERRUPT_POLARITY_DEFAULT;
+	int_params.current_polarity = INTERRUPT_POLARITY_DEFAULT;
+
+	/*
+	 * Actions of amdgpu_irq_add_id():
+	 * 1. Register a set() function with base driver.
+	 *    Base driver will call set() function to enable/disable an
+	 *    interrupt in DC hardware.
+	 * 2. Register amdgpu_dm_irq_handler().
+	 *    Base driver will call amdgpu_dm_irq_handler() for ALL interrupts
+	 *    coming from DC hardware.
+	 *    amdgpu_dm_irq_handler() will re-direct the interrupt to DC
+	 *    for acknowledging and handling. */
+
+	/* Use VBLANK interrupt */
+	for (i = 0; i < adev->mode_info.num_crtc; i++) {
+		r = amdgpu_irq_add_id(adev, client_id, i+1 , &adev->crtc_irq);
+		if (r) {
+			DRM_ERROR("Failed to add crtc irq id!\n");
+			return r;
+		}
+
+		int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT;
+		int_params.irq_source =
+			dc_interrupt_to_irq_source(dc, i+1 , 0);
+
+		c_irq_params = &adev->dm.vblank_params[int_params.irq_source - DC_IRQ_SOURCE_VBLANK1];
+
+		c_irq_params->adev = adev;
+		c_irq_params->irq_src = int_params.irq_source;
+
+		amdgpu_dm_irq_register_interrupt(adev, &int_params,
+				dm_crtc_high_irq, c_irq_params);
+	}
+
+	/* Use GRPH_PFLIP interrupt */
+	for (i = VISLANDS30_IV_SRCID_D1_GRPH_PFLIP;
+			i <= VISLANDS30_IV_SRCID_D6_GRPH_PFLIP; i += 2) {
+		r = amdgpu_irq_add_id(adev, client_id, i, &adev->pageflip_irq);
+		if (r) {
+			DRM_ERROR("Failed to add page flip irq id!\n");
+			return r;
+		}
+
+		int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT;
+		int_params.irq_source =
+			dc_interrupt_to_irq_source(dc, i, 0);
+
+		c_irq_params = &adev->dm.pflip_params[int_params.irq_source - DC_IRQ_SOURCE_PFLIP_FIRST];
+
+		c_irq_params->adev = adev;
+		c_irq_params->irq_src = int_params.irq_source;
+
+		amdgpu_dm_irq_register_interrupt(adev, &int_params,
+				dm_pflip_high_irq, c_irq_params);
+
+	}
+
+	/* HPD */
+	r = amdgpu_irq_add_id(adev, client_id,
+			VISLANDS30_IV_SRCID_HOTPLUG_DETECT_A, &adev->hpd_irq);
+	if (r) {
+		DRM_ERROR("Failed to add hpd irq id!\n");
+		return r;
+	}
+
+	register_hpd_handlers(adev);
+
+	return 0;
+}
+#endif
+
 /* Register IRQ sources and initialize IRQ callbacks */
 static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 {
@@ -2704,7 +2868,7 @@ static int dm_atomic_get_state(struct drm_atomic_state *state,
 			       struct dm_atomic_state **dm_state)
 {
 	struct drm_device *dev = state->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 	struct drm_private_state *priv_state;
 
@@ -2724,7 +2888,7 @@ static struct dm_atomic_state *
 dm_atomic_get_new_state(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 	struct drm_private_obj *obj;
 	struct drm_private_state *new_obj_state;
@@ -2738,24 +2902,6 @@ dm_atomic_get_new_state(struct drm_atomic_state *state)
 	return NULL;
 }
 
-static struct dm_atomic_state *
-dm_atomic_get_old_state(struct drm_atomic_state *state)
-{
-	struct drm_device *dev = state->dev;
-	struct amdgpu_device *adev = dev->dev_private;
-	struct amdgpu_display_manager *dm = &adev->dm;
-	struct drm_private_obj *obj;
-	struct drm_private_state *old_obj_state;
-	int i;
-
-	for_each_old_private_obj_in_state(state, obj, old_obj_state, i) {
-		if (obj->funcs == dm->atomic_obj.funcs)
-			return to_dm_atomic_state(old_obj_state);
-	}
-
-	return NULL;
-}
-
 static struct drm_private_state *
 dm_atomic_duplicate_state(struct drm_private_obj *obj)
 {
@@ -2803,18 +2949,18 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev)
 
 	adev->mode_info.mode_config_initialized = true;
 
-	adev->ddev->mode_config.funcs = (void *)&amdgpu_dm_mode_funcs;
-	adev->ddev->mode_config.helper_private = &amdgpu_dm_mode_config_helperfuncs;
+	adev_to_drm(adev)->mode_config.funcs = (void *)&amdgpu_dm_mode_funcs;
+	adev_to_drm(adev)->mode_config.helper_private = &amdgpu_dm_mode_config_helperfuncs;
 
-	adev->ddev->mode_config.max_width = 16384;
-	adev->ddev->mode_config.max_height = 16384;
+	adev_to_drm(adev)->mode_config.max_width = 16384;
+	adev_to_drm(adev)->mode_config.max_height = 16384;
 
-	adev->ddev->mode_config.preferred_depth = 24;
-	adev->ddev->mode_config.prefer_shadow = 1;
+	adev_to_drm(adev)->mode_config.preferred_depth = 24;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 	/* indicates support for immediate flip */
-	adev->ddev->mode_config.async_page_flip = true;
+	adev_to_drm(adev)->mode_config.async_page_flip = true;
 
-	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
+	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
 	state = kzalloc(sizeof(*state), GFP_KERNEL);
 	if (!state)
@@ -2828,7 +2974,7 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev)
 
 	dc_resource_state_copy_construct_current(adev->dm.dc, state->context);
 
-	drm_atomic_private_obj_init(adev->ddev,
+	drm_atomic_private_obj_init(adev_to_drm(adev),
 				    &adev->dm.atomic_obj,
 				    &state->base,
 				    &dm_atomic_state_funcs);
@@ -3000,13 +3146,13 @@ amdgpu_dm_register_backlight_device(struct amdgpu_display_manager *dm)
 	props.type = BACKLIGHT_RAW;
 
 	snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d",
-			dm->adev->ddev->primary->index);
+		 adev_to_drm(dm->adev)->primary->index);
 
 	dm->backlight_dev = backlight_device_register(bl_name,
-			dm->adev->ddev->dev,
-			dm,
-			&amdgpu_dm_backlight_ops,
-			&props);
+						      adev_to_drm(dm->adev)->dev,
+						      dm,
+						      &amdgpu_dm_backlight_ops,
+						      &props);
 
 	if (IS_ERR(dm->backlight_dev))
 		DRM_ERROR("DM: Backlight registration failed!\n");
@@ -3212,6 +3358,17 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 
 	/* Software is initialized. Now we can register interrupt handlers. */
 	switch (adev->asic_type) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+		if (dce60_register_irq_handlers(dm->adev)) {
+			DRM_ERROR("DM: Failed to initialize IRQ\n");
+			goto fail;
+		}
+		break;
+#endif
 	case CHIP_BONAIRE:
 	case CHIP_HAWAII:
 	case CHIP_KAVERI:
@@ -3254,9 +3411,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 		goto fail;
 	}
 
-	/* No userspace support. */
-	dm->dc->debug.disable_tri_buf = true;
-
 	return 0;
 fail:
 	kfree(aencoder);
@@ -3312,14 +3466,14 @@ static ssize_t s3_debug_store(struct device *device,
 	int ret;
 	int s3_state;
 	struct drm_device *drm_dev = dev_get_drvdata(device);
-	struct amdgpu_device *adev = drm_dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
 
 	ret = kstrtoint(buf, 0, &s3_state);
 
 	if (ret == 0) {
 		if (s3_state) {
 			dm_resume(adev);
-			drm_kms_helper_hotplug_event(adev->ddev);
+			drm_kms_helper_hotplug_event(adev_to_drm(adev));
 		} else
 			dm_suspend(adev);
 	}
@@ -3336,6 +3490,20 @@ static int dm_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	switch (adev->asic_type) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+		adev->mode_info.num_crtc = 6;
+		adev->mode_info.num_hpd = 6;
+		adev->mode_info.num_dig = 6;
+		break;
+	case CHIP_OLAND:
+		adev->mode_info.num_crtc = 2;
+		adev->mode_info.num_hpd = 2;
+		adev->mode_info.num_dig = 2;
+		break;
+#endif
 	case CHIP_BONAIRE:
 	case CHIP_HAWAII:
 		adev->mode_info.num_crtc = 6;
@@ -3432,7 +3600,7 @@ static int dm_early_init(void *handle)
 	 */
 #if defined(CONFIG_DEBUG_KERNEL_DC)
 	device_create_file(
-		adev->ddev->dev,
+		adev_to_drm(adev)->dev,
 		&dev_attr_s3_debug);
 #endif
 
@@ -3443,21 +3611,12 @@ static bool modeset_required(struct drm_crtc_state *crtc_state,
 			     struct dc_stream_state *new_stream,
 			     struct dc_stream_state *old_stream)
 {
-	if (!drm_atomic_crtc_needs_modeset(crtc_state))
-		return false;
-
-	if (!crtc_state->enable)
-		return false;
-
-	return crtc_state->active;
+	return crtc_state->active && drm_atomic_crtc_needs_modeset(crtc_state);
 }
 
 static bool modereset_required(struct drm_crtc_state *crtc_state)
 {
-	if (!drm_atomic_crtc_needs_modeset(crtc_state))
-		return false;
-
-	return !crtc_state->enable || !crtc_state->active;
+	return !crtc_state->active && drm_atomic_crtc_needs_modeset(crtc_state);
 }
 
 static void amdgpu_dm_encoder_destroy(struct drm_encoder *encoder)
@@ -3530,8 +3689,17 @@ static int fill_dc_scaling_info(const struct drm_plane_state *state,
 static int get_fb_info(const struct amdgpu_framebuffer *amdgpu_fb,
 		       uint64_t *tiling_flags, bool *tmz_surface)
 {
-	struct amdgpu_bo *rbo = gem_to_amdgpu_bo(amdgpu_fb->base.obj[0]);
-	int r = amdgpu_bo_reserve(rbo, false);
+	struct amdgpu_bo *rbo;
+	int r;
+
+	if (!amdgpu_fb) {
+		*tiling_flags = 0;
+		*tmz_surface = false;
+		return 0;
+	}
+
+	rbo = gem_to_amdgpu_bo(amdgpu_fb->base.obj[0]);
+	r = amdgpu_bo_reserve(rbo, false);
 
 	if (unlikely(r)) {
 		/* Don't show error message when returning -ERESTARTSYS */
@@ -3954,13 +4122,10 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev,
 				    struct drm_crtc_state *crtc_state)
 {
 	struct dm_crtc_state *dm_crtc_state = to_dm_crtc_state(crtc_state);
-	const struct amdgpu_framebuffer *amdgpu_fb =
-		to_amdgpu_framebuffer(plane_state->fb);
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
 	struct dc_scaling_info scaling_info;
 	struct dc_plane_info plane_info;
-	uint64_t tiling_flags;
 	int ret;
-	bool tmz_surface = false;
 	bool force_disable_dcc = false;
 
 	ret = fill_dc_scaling_info(plane_state, &scaling_info);
@@ -3972,15 +4137,12 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev,
 	dc_plane_state->clip_rect = scaling_info.clip_rect;
 	dc_plane_state->scaling_quality = scaling_info.scaling_quality;
 
-	ret = get_fb_info(amdgpu_fb, &tiling_flags, &tmz_surface);
-	if (ret)
-		return ret;
-
 	force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend;
-	ret = fill_dc_plane_info_and_addr(adev, plane_state, tiling_flags,
+	ret = fill_dc_plane_info_and_addr(adev, plane_state,
+					  dm_plane_state->tiling_flags,
 					  &plane_info,
 					  &dc_plane_state->address,
-					  tmz_surface,
+					  dm_plane_state->tmz_surface,
 					  force_disable_dcc);
 	if (ret)
 		return ret;
@@ -4562,7 +4724,11 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 							     dc_link_get_link_cap(aconnector->dc_link));
 
 #if defined(CONFIG_DRM_AMD_DC_DCN)
-		if (dsc_caps.is_dsc_supported)
+		if (aconnector->dsc_settings.dsc_force_enable != DSC_CLK_FORCE_DISABLE && dsc_caps.is_dsc_supported) {
+			/* Set DSC policy according to dsc_clock_en */
+			dc_dsc_policy_set_enable_dsc_when_not_needed(
+				aconnector->dsc_settings.dsc_force_enable == DSC_CLK_FORCE_ENABLE);
+
 			if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc->res_pool->dscs[0],
 						  &dsc_caps,
 						  aconnector->dc_link->ctx->dc->debug.dsc_min_slice_height_override,
@@ -4570,6 +4736,19 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 						  &stream->timing,
 						  &stream->timing.dsc_cfg))
 				stream->timing.flags.DSC = 1;
+			/* Overwrite the stream flag if DSC is enabled through debugfs */
+			if (aconnector->dsc_settings.dsc_force_enable == DSC_CLK_FORCE_ENABLE)
+				stream->timing.flags.DSC = 1;
+
+			if (stream->timing.flags.DSC && aconnector->dsc_settings.dsc_num_slices_h)
+				stream->timing.dsc_cfg.num_slices_h = aconnector->dsc_settings.dsc_num_slices_h;
+
+			if (stream->timing.flags.DSC && aconnector->dsc_settings.dsc_num_slices_v)
+				stream->timing.dsc_cfg.num_slices_v = aconnector->dsc_settings.dsc_num_slices_v;
+
+			if (stream->timing.flags.DSC && aconnector->dsc_settings.dsc_bits_per_pixel)
+				stream->timing.dsc_cfg.bits_per_pixel = aconnector->dsc_settings.dsc_bits_per_pixel;
+		}
 #endif
 	}
 
@@ -4583,7 +4762,8 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 	update_stream_signal(stream, sink);
 
 	if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
-		mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket, false, false);
+		mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket);
+
 	if (stream->link->psr_settings.psr_feature_enabled) {
 		//
 		// should decide stream support vsc sdp colorimetry capability
@@ -4663,7 +4843,6 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc)
 	}
 
 	state->active_planes = cur->active_planes;
-	state->vrr_params = cur->vrr_params;
 	state->vrr_infopacket = cur->vrr_infopacket;
 	state->abm_level = cur->abm_level;
 	state->vrr_supported = cur->vrr_supported;
@@ -4681,7 +4860,7 @@ static inline int dm_set_vupdate_irq(struct drm_crtc *crtc, bool enable)
 {
 	enum dc_irq_source irq_source;
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	int rc;
 
 	irq_source = IRQ_TYPE_VUPDATE + acrtc->otg_inst;
@@ -4697,7 +4876,7 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
 {
 	enum dc_irq_source irq_source;
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
 	int rc = 0;
 
@@ -4764,6 +4943,8 @@ amdgpu_dm_connector_detect(struct drm_connector *connector, bool force)
 	else
 		connected = (aconnector->base.force == DRM_FORCE_ON);
 
+	update_subconnector_property(aconnector);
+
 	return (connected ? connector_status_connected :
 			connector_status_disconnected);
 }
@@ -4774,7 +4955,7 @@ int amdgpu_dm_connector_atomic_set_property(struct drm_connector *connector,
 					    uint64_t val)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct dm_connector_state *dm_old_state =
 		to_dm_connector_state(connector->state);
 	struct dm_connector_state *dm_new_state =
@@ -4829,7 +5010,7 @@ int amdgpu_dm_connector_atomic_get_property(struct drm_connector *connector,
 					    uint64_t *val)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct dm_connector_state *dm_state =
 		to_dm_connector_state(state);
 	int ret = -EINVAL;
@@ -4879,9 +5060,10 @@ static void amdgpu_dm_connector_destroy(struct drm_connector *connector)
 {
 	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
 	const struct dc_link *link = aconnector->dc_link;
-	struct amdgpu_device *adev = connector->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(connector->dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 
+	drm_atomic_private_obj_fini(&aconnector->mst_mgr.base);
 #if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) ||\
 	defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE)
 
@@ -5064,7 +5246,7 @@ create_validate_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 				const struct dc_stream_state *old_stream)
 {
 	struct drm_connector *connector = &aconnector->base;
-	struct amdgpu_device *adev = connector->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(connector->dev);
 	struct dc_stream_state *stream;
 	const struct drm_connector_state *drm_state = dm_state ? &dm_state->base : NULL;
 	int requested_bpc = drm_state ? drm_state->max_requested_bpc : 8;
@@ -5328,7 +5510,7 @@ static void dm_update_crtc_active_planes(struct drm_crtc *crtc,
 static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc,
 				       struct drm_crtc_state *state)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct dc *dc = adev->dm.dc;
 	struct dm_crtc_state *dm_crtc_state = to_dm_crtc_state(state);
 	int ret = -EINVAL;
@@ -5549,6 +5731,10 @@ dm_drm_plane_duplicate_state(struct drm_plane *plane)
 		dc_plane_state_retain(dm_plane_state->dc_state);
 	}
 
+	/* Framebuffer hasn't been updated yet, so retain old flags. */
+	dm_plane_state->tiling_flags = old_dm_plane_state->tiling_flags;
+	dm_plane_state->tmz_surface = old_dm_plane_state->tmz_surface;
+
 	return &dm_plane_state->base;
 }
 
@@ -5583,14 +5769,8 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 	struct list_head list;
 	struct ttm_validate_buffer tv;
 	struct ww_acquire_ctx ticket;
-	uint64_t tiling_flags;
 	uint32_t domain;
 	int r;
-	bool tmz_surface = false;
-	bool force_disable_dcc = false;
-
-	dm_plane_state_old = to_dm_plane_state(plane->state);
-	dm_plane_state_new = to_dm_plane_state(new_state);
 
 	if (!new_state->fb) {
 		DRM_DEBUG_DRIVER("No FB bound\n");
@@ -5634,27 +5814,35 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 		return r;
 	}
 
-	amdgpu_bo_get_tiling_flags(rbo, &tiling_flags);
-
-	tmz_surface = amdgpu_bo_encrypted(rbo);
-
 	ttm_eu_backoff_reservation(&ticket, &list);
 
 	afb->address = amdgpu_bo_gpu_offset(rbo);
 
 	amdgpu_bo_ref(rbo);
 
+	/**
+	 * We don't do surface updates on planes that have been newly created,
+	 * but we also don't have the afb->address during atomic check.
+	 *
+	 * Fill in buffer attributes depending on the address here, but only on
+	 * newly created planes since they're not being used by DC yet and this
+	 * won't modify global state.
+	 */
+	dm_plane_state_old = to_dm_plane_state(plane->state);
+	dm_plane_state_new = to_dm_plane_state(new_state);
+
 	if (dm_plane_state_new->dc_state &&
-			dm_plane_state_old->dc_state != dm_plane_state_new->dc_state) {
-		struct dc_plane_state *plane_state = dm_plane_state_new->dc_state;
+	    dm_plane_state_old->dc_state != dm_plane_state_new->dc_state) {
+		struct dc_plane_state *plane_state =
+			dm_plane_state_new->dc_state;
+		bool force_disable_dcc = !plane_state->dcc.enable;
 
-		force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend;
 		fill_plane_buffer_attributes(
 			adev, afb, plane_state->format, plane_state->rotation,
-			tiling_flags, &plane_state->tiling_info,
-			&plane_state->plane_size, &plane_state->dcc,
-			&plane_state->address, tmz_surface,
-			force_disable_dcc);
+			dm_plane_state_new->tiling_flags,
+			&plane_state->tiling_info, &plane_state->plane_size,
+			&plane_state->dcc, &plane_state->address,
+			dm_plane_state_new->tmz_surface, force_disable_dcc);
 	}
 
 	return 0;
@@ -5695,7 +5883,7 @@ static int dm_plane_helper_check_state(struct drm_plane_state *state,
 static int dm_plane_atomic_check(struct drm_plane *plane,
 				 struct drm_plane_state *state)
 {
-	struct amdgpu_device *adev = plane->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(plane->dev);
 	struct dc *dc = adev->dm.dc;
 	struct dm_plane_state *dm_plane_state;
 	struct dc_scaling_info scaling_info;
@@ -5864,7 +6052,7 @@ static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
 	num_formats = get_plane_formats(plane, plane_cap, formats,
 					ARRAY_SIZE(formats));
 
-	res = drm_universal_plane_init(dm->adev->ddev, plane, possible_crtcs,
+	res = drm_universal_plane_init(adev_to_drm(dm->adev), plane, possible_crtcs,
 				       &dm_plane_funcs, formats, num_formats,
 				       NULL, plane->type, NULL);
 	if (res)
@@ -5898,8 +6086,9 @@ static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
 		DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_90 |
 		DRM_MODE_ROTATE_180 | DRM_MODE_ROTATE_270;
 
-	drm_plane_create_rotation_property(plane, DRM_MODE_ROTATE_0,
-					   supported_rotations);
+	if (dm->adev->asic_type >= CHIP_BONAIRE)
+		drm_plane_create_rotation_property(plane, DRM_MODE_ROTATE_0,
+						   supported_rotations);
 
 	drm_plane_helper_add(plane, &dm_plane_helper_funcs);
 
@@ -6169,7 +6358,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 				     struct dc_link *link,
 				     int link_index)
 {
-	struct amdgpu_device *adev = dm->ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dm->ddev);
 
 	/*
 	 * Some of the properties below require access to state, like bpc.
@@ -6420,7 +6609,7 @@ static int amdgpu_dm_encoder_init(struct drm_device *dev,
 				  struct amdgpu_encoder *aencoder,
 				  uint32_t link_index)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	int res = drm_encoder_init(dev,
 				   &aencoder->base,
@@ -6605,7 +6794,7 @@ static int get_cursor_position(struct drm_plane *plane, struct drm_crtc *crtc,
 static void handle_cursor_update(struct drm_plane *plane,
 				 struct drm_plane_state *old_plane_state)
 {
-	struct amdgpu_device *adev = plane->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(plane->dev);
 	struct amdgpu_framebuffer *afb = to_amdgpu_framebuffer(plane->state->fb);
 	struct drm_crtc *crtc = afb ? plane->state->crtc : old_plane_state->crtc;
 	struct dm_crtc_state *crtc_state = crtc ? to_dm_crtc_state(crtc->state) : NULL;
@@ -6694,6 +6883,7 @@ static void update_freesync_state_on_stream(
 	struct mod_vrr_params vrr_params;
 	struct dc_info_packet vrr_infopacket = {0};
 	struct amdgpu_device *adev = dm->adev;
+	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(new_crtc_state->base.crtc);
 	unsigned long flags;
 
 	if (!new_stream)
@@ -6707,8 +6897,8 @@ static void update_freesync_state_on_stream(
 	if (!new_stream->timing.h_total || !new_stream->timing.v_total)
 		return;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
-	vrr_params = new_crtc_state->vrr_params;
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
+        vrr_params = acrtc->dm_irq_params.vrr_params;
 
 	if (surface) {
 		mod_freesync_handle_preflip(
@@ -6739,7 +6929,7 @@ static void update_freesync_state_on_stream(
 		&vrr_infopacket);
 
 	new_crtc_state->freesync_timing_changed |=
-		(memcmp(&new_crtc_state->vrr_params.adjust,
+		(memcmp(&acrtc->dm_irq_params.vrr_params.adjust,
 			&vrr_params.adjust,
 			sizeof(vrr_params.adjust)) != 0);
 
@@ -6748,10 +6938,10 @@ static void update_freesync_state_on_stream(
 			&vrr_infopacket,
 			sizeof(vrr_infopacket)) != 0);
 
-	new_crtc_state->vrr_params = vrr_params;
+	acrtc->dm_irq_params.vrr_params = vrr_params;
 	new_crtc_state->vrr_infopacket = vrr_infopacket;
 
-	new_stream->adjust = new_crtc_state->vrr_params.adjust;
+	new_stream->adjust = acrtc->dm_irq_params.vrr_params.adjust;
 	new_stream->vrr_infopacket = vrr_infopacket;
 
 	if (new_crtc_state->freesync_vrr_info_changed)
@@ -6760,10 +6950,10 @@ static void update_freesync_state_on_stream(
 			      (int)new_crtc_state->base.vrr_enabled,
 			      (int)vrr_params.state);
 
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 }
 
-static void pre_update_freesync_state_on_stream(
+static void update_stream_irq_parameters(
 	struct amdgpu_display_manager *dm,
 	struct dm_crtc_state *new_crtc_state)
 {
@@ -6771,6 +6961,7 @@ static void pre_update_freesync_state_on_stream(
 	struct mod_vrr_params vrr_params;
 	struct mod_freesync_config config = new_crtc_state->freesync_config;
 	struct amdgpu_device *adev = dm->adev;
+	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(new_crtc_state->base.crtc);
 	unsigned long flags;
 
 	if (!new_stream)
@@ -6783,8 +6974,8 @@ static void pre_update_freesync_state_on_stream(
 	if (!new_stream->timing.h_total || !new_stream->timing.v_total)
 		return;
 
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
-	vrr_params = new_crtc_state->vrr_params;
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
+	vrr_params = acrtc->dm_irq_params.vrr_params;
 
 	if (new_crtc_state->vrr_supported &&
 	    config.min_refresh_in_uhz &&
@@ -6801,12 +6992,15 @@ static void pre_update_freesync_state_on_stream(
 				      &config, &vrr_params);
 
 	new_crtc_state->freesync_timing_changed |=
-		(memcmp(&new_crtc_state->vrr_params.adjust,
-			&vrr_params.adjust,
-			sizeof(vrr_params.adjust)) != 0);
+		(memcmp(&acrtc->dm_irq_params.vrr_params.adjust,
+			&vrr_params.adjust, sizeof(vrr_params.adjust)) != 0);
 
-	new_crtc_state->vrr_params = vrr_params;
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	new_crtc_state->freesync_config = config;
+	/* Copy state for access from DM IRQ handler */
+	acrtc->dm_irq_params.freesync_config = config;
+	acrtc->dm_irq_params.active_planes = new_crtc_state->active_planes;
+	acrtc->dm_irq_params.vrr_params = vrr_params;
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 }
 
 static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state,
@@ -6876,8 +7070,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 	long r;
 	unsigned long flags;
 	struct amdgpu_bo *abo;
-	uint64_t tiling_flags;
-	bool tmz_surface = false;
 	uint32_t target_vblank, last_flip_vblank;
 	bool vrr_active = amdgpu_dm_vrr_active(acrtc_state);
 	bool pflip_present = false;
@@ -6961,28 +7153,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		if (unlikely(r <= 0))
 			DRM_ERROR("Waiting for fences timed out!");
 
-		/*
-		 * TODO This might fail and hence better not used, wait
-		 * explicitly on fences instead
-		 * and in general should be called for
-		 * blocking commit to as per framework helpers
-		 */
-		r = amdgpu_bo_reserve(abo, true);
-		if (unlikely(r != 0))
-			DRM_ERROR("failed to reserve buffer before flip\n");
-
-		amdgpu_bo_get_tiling_flags(abo, &tiling_flags);
-
-		tmz_surface = amdgpu_bo_encrypted(abo);
-
-		amdgpu_bo_unreserve(abo);
-
 		fill_dc_plane_info_and_addr(
-			dm->adev, new_plane_state, tiling_flags,
+			dm->adev, new_plane_state,
+			dm_new_plane_state->tiling_flags,
 			&bundle->plane_infos[planes_count],
 			&bundle->flip_addrs[planes_count].address,
-			tmz_surface,
-			false);
+			dm_new_plane_state->tmz_surface, false);
 
 		DRM_DEBUG_DRIVER("plane: id=%d dcc_en=%d\n",
 				 new_plane_state->plane->index,
@@ -7047,7 +7223,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			 * on late submission of flips.
 			 */
 			spin_lock_irqsave(&pcrtc->dev->event_lock, flags);
-			last_flip_vblank = acrtc_attach->last_flip_vblank;
+			last_flip_vblank = acrtc_attach->dm_irq_params.last_flip_vblank;
 			spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags);
 		}
 
@@ -7131,7 +7307,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			spin_lock_irqsave(&pcrtc->dev->event_lock, flags);
 			dc_stream_adjust_vmin_vmax(
 				dm->dc, acrtc_state->stream,
-				&acrtc_state->vrr_params.adjust);
+				&acrtc_attach->dm_irq_params.vrr_params.adjust);
 			spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags);
 		}
 		mutex_lock(&dm->dc_lock);
@@ -7160,9 +7336,8 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		 * on some ASICs).
 		 */
 		if (dm_old_crtc_state->active_planes != acrtc_state->active_planes)
-			dm_update_pflip_irq_state(
-				(struct amdgpu_device *)dev->dev_private,
-				acrtc_attach);
+			dm_update_pflip_irq_state(drm_to_adev(dev),
+						  acrtc_attach);
 
 		if ((acrtc_state->update_type > UPDATE_TYPE_FAST) &&
 				acrtc_state->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED &&
@@ -7192,7 +7367,7 @@ cleanup:
 static void amdgpu_dm_commit_audio(struct drm_device *dev,
 				   struct drm_atomic_state *state)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_connector *connector;
 	struct drm_connector_state *old_con_state, *new_con_state;
@@ -7282,34 +7457,6 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev,
 				   struct drm_atomic_state *state,
 				   bool nonblock)
 {
-	struct drm_crtc *crtc;
-	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
-	struct amdgpu_device *adev = dev->dev_private;
-	int i;
-
-	/*
-	 * We evade vblank and pflip interrupts on CRTCs that are undergoing
-	 * a modeset, being disabled, or have no active planes.
-	 *
-	 * It's done in atomic commit rather than commit tail for now since
-	 * some of these interrupt handlers access the current CRTC state and
-	 * potentially the stream pointer itself.
-	 *
-	 * Since the atomic state is swapped within atomic commit and not within
-	 * commit tail this would leave to new state (that hasn't been committed yet)
-	 * being accesssed from within the handlers.
-	 *
-	 * TODO: Fix this so we can do this in commit tail and not have to block
-	 * in atomic check.
-	 */
-	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
-		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
-
-		if (old_crtc_state->active &&
-		    (!new_crtc_state->active ||
-		     drm_atomic_crtc_needs_modeset(new_crtc_state)))
-			manage_dm_interrupts(adev, acrtc, false);
-	}
 	/*
 	 * Add check here for SoC's that support hardware cursor plane, to
 	 * unset legacy_cursor_update
@@ -7331,7 +7478,7 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev,
 static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 	struct dm_atomic_state *dm_state;
 	struct dc_state *dc_state = NULL, *dc_state_temp = NULL;
@@ -7344,8 +7491,10 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 	struct drm_connector_state *old_con_state, *new_con_state;
 	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
 	int crtc_disable_count = 0;
+	bool mode_set_reset_required = false;
 
 	drm_atomic_helper_update_legacy_modeset_state(dev, state);
+	drm_atomic_helper_calc_timestamping_constants(state);
 
 	dm_state = dm_atomic_get_new_state(state);
 	if (dm_state && dm_state->context) {
@@ -7358,6 +7507,20 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		dc_resource_state_copy_construct_current(dm->dc, dc_state);
 	}
 
+	for_each_oldnew_crtc_in_state (state, crtc, old_crtc_state,
+				       new_crtc_state, i) {
+		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
+
+		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+
+		if (old_crtc_state->active &&
+		    (!new_crtc_state->active ||
+		     drm_atomic_crtc_needs_modeset(new_crtc_state))) {
+			manage_dm_interrupts(adev, acrtc, false);
+			dc_stream_release(dm_old_crtc_state->stream);
+		}
+	}
+
 	/* update changed items */
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
@@ -7420,19 +7583,21 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 			acrtc->enabled = true;
 			acrtc->hw_mode = new_crtc_state->mode;
 			crtc->hwmode = new_crtc_state->mode;
+			mode_set_reset_required = true;
 		} else if (modereset_required(new_crtc_state)) {
 			DRM_DEBUG_DRIVER("Atomic commit: RESET. crtc id %d:[%p]\n", acrtc->crtc_id, acrtc);
 			/* i.e. reset mode */
-			if (dm_old_crtc_state->stream) {
-				if (dm_old_crtc_state->stream->link->psr_settings.psr_allow_active)
-					amdgpu_dm_psr_disable(dm_old_crtc_state->stream);
-
+			if (dm_old_crtc_state->stream)
 				remove_stream(adev, acrtc, dm_old_crtc_state->stream);
-			}
+			mode_set_reset_required = true;
 		}
 	} /* for_each_crtc_in_state() */
 
 	if (dc_state) {
+		/* if there mode set or reset, disable eDP PSR */
+		if (mode_set_reset_required)
+			amdgpu_dm_psr_disable_all(dm);
+
 		dm_enable_per_frame_crtc_master_sync(dc_state);
 		mutex_lock(&dm->dc_lock);
 		WARN_ON(!dc_commit_state(dm->dc, dc_state));
@@ -7451,7 +7616,6 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 			if (!status)
 				status = dc_stream_get_status_from_state(dc_state,
 									 dm_new_crtc_state->stream);
-
 			if (!status)
 				DC_ERR("got no status for stream %p on acrtc%p\n", dm_new_crtc_state->stream, acrtc);
 			else
@@ -7577,8 +7741,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
 
-		/* Update freesync active state. */
-		pre_update_freesync_state_on_stream(dm, dm_new_crtc_state);
+		/* For freesync config update on crtc state and params for irq */
+		update_stream_irq_parameters(dm, dm_new_crtc_state);
 
 		/* Handle vrr on->off / off->on transitions */
 		amdgpu_dm_handle_vrr_transition(dm_old_crtc_state,
@@ -7594,10 +7758,15 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
 
+		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
+
 		if (new_crtc_state->active &&
 		    (!old_crtc_state->active ||
 		     drm_atomic_crtc_needs_modeset(new_crtc_state))) {
+			dc_stream_retain(dm_new_crtc_state->stream);
+			acrtc->dm_irq_params.stream = dm_new_crtc_state->stream;
 			manage_dm_interrupts(adev, acrtc, true);
+
 #ifdef CONFIG_DEBUG_FS
 			/**
 			 * Frontend may have changed so reapply the CRC capture
@@ -7634,7 +7803,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 	 * send vblank event on all events not handled in flip and
 	 * mark consumed event for drm_atomic_helper_commit_hw_done
 	 */
-	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
 
 		if (new_crtc_state->event)
@@ -7642,7 +7811,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 
 		new_crtc_state->event = NULL;
 	}
-	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 
 	/* Signal HW programming completion */
 	drm_atomic_helper_commit_hw_done(state);
@@ -7841,8 +8010,6 @@ static void reset_freesync_config_for_crtc(
 {
 	new_crtc_state->vrr_supported = false;
 
-	memset(&new_crtc_state->vrr_params, 0,
-	       sizeof(new_crtc_state->vrr_params));
 	memset(&new_crtc_state->vrr_infopacket, 0,
 	       sizeof(new_crtc_state->vrr_infopacket));
 }
@@ -7914,6 +8081,13 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 			goto fail;
 		}
 
+		/*
+		 * TODO: Check VSDB bits to decide whether this should
+		 * be enabled or not.
+		 */
+		new_stream->triggered_crtc_reset.enabled =
+			dm->force_timing_sync;
+
 		dm_new_crtc_state->abm_level = dm_new_conn_state->abm_level;
 
 		ret = fill_hdr_info_packet(drm_new_conn_state,
@@ -8033,8 +8207,7 @@ skip_modeset:
 	 * We want to do dc stream updates that do not require a
 	 * full modeset below.
 	 */
-	if (!(enable && aconnector && new_crtc_state->enable &&
-	      new_crtc_state->active))
+	if (!(enable && aconnector && new_crtc_state->active))
 		return 0;
 	/*
 	 * Given above conditions, the dc state cannot be NULL because:
@@ -8125,6 +8298,8 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 	 * TODO: Come up with a more elegant solution for this.
 	 */
 	for_each_oldnew_plane_in_state(state, other, old_other_state, new_other_state, i) {
+		struct dm_plane_state *old_dm_plane_state, *new_dm_plane_state;
+
 		if (other->type == DRM_PLANE_TYPE_CURSOR)
 			continue;
 
@@ -8135,9 +8310,45 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 		if (old_other_state->crtc != new_other_state->crtc)
 			return true;
 
-		/* TODO: Remove this once we can handle fast format changes. */
-		if (old_other_state->fb && new_other_state->fb &&
-		    old_other_state->fb->format != new_other_state->fb->format)
+		/* Src/dst size and scaling updates. */
+		if (old_other_state->src_w != new_other_state->src_w ||
+		    old_other_state->src_h != new_other_state->src_h ||
+		    old_other_state->crtc_w != new_other_state->crtc_w ||
+		    old_other_state->crtc_h != new_other_state->crtc_h)
+			return true;
+
+		/* Rotation / mirroring updates. */
+		if (old_other_state->rotation != new_other_state->rotation)
+			return true;
+
+		/* Blending updates. */
+		if (old_other_state->pixel_blend_mode !=
+		    new_other_state->pixel_blend_mode)
+			return true;
+
+		/* Alpha updates. */
+		if (old_other_state->alpha != new_other_state->alpha)
+			return true;
+
+		/* Colorspace changes. */
+		if (old_other_state->color_range != new_other_state->color_range ||
+		    old_other_state->color_encoding != new_other_state->color_encoding)
+			return true;
+
+		/* Framebuffer checks fall at the end. */
+		if (!old_other_state->fb || !new_other_state->fb)
+			continue;
+
+		/* Pixel format changes can require bandwidth updates. */
+		if (old_other_state->fb->format != new_other_state->fb->format)
+			return true;
+
+		old_dm_plane_state = to_dm_plane_state(old_other_state);
+		new_dm_plane_state = to_dm_plane_state(new_other_state);
+
+		/* Tiling and DCC changes also require bandwidth updates. */
+		if (old_dm_plane_state->tiling_flags !=
+		    new_dm_plane_state->tiling_flags)
 			return true;
 	}
 
@@ -8217,8 +8428,7 @@ static int dm_update_plane_state(struct dc *dc,
 				dm_old_plane_state->dc_state,
 				dm_state->context)) {
 
-			ret = EINVAL;
-			return ret;
+			return -EINVAL;
 		}
 
 
@@ -8259,7 +8469,7 @@ static int dm_update_plane_state(struct dc *dc,
 				plane->base.id, new_plane_crtc->base.id);
 
 		ret = fill_dc_plane_attributes(
-			new_plane_crtc->dev->dev_private,
+			drm_to_adev(new_plane_crtc->dev),
 			dc_new_plane_state,
 			new_plane_state,
 			new_crtc_state);
@@ -8305,169 +8515,6 @@ static int dm_update_plane_state(struct dc *dc,
 	return ret;
 }
 
-static int
-dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm,
-				    struct drm_atomic_state *state,
-				    enum surface_update_type *out_type)
-{
-	struct dc *dc = dm->dc;
-	struct dm_atomic_state *dm_state = NULL, *old_dm_state = NULL;
-	int i, j, num_plane, ret = 0;
-	struct drm_plane_state *old_plane_state, *new_plane_state;
-	struct dm_plane_state *new_dm_plane_state, *old_dm_plane_state;
-	struct drm_crtc *new_plane_crtc;
-	struct drm_plane *plane;
-
-	struct drm_crtc *crtc;
-	struct drm_crtc_state *new_crtc_state, *old_crtc_state;
-	struct dm_crtc_state *new_dm_crtc_state, *old_dm_crtc_state;
-	struct dc_stream_status *status = NULL;
-	enum surface_update_type update_type = UPDATE_TYPE_FAST;
-	struct surface_info_bundle {
-		struct dc_surface_update surface_updates[MAX_SURFACES];
-		struct dc_plane_info plane_infos[MAX_SURFACES];
-		struct dc_scaling_info scaling_infos[MAX_SURFACES];
-		struct dc_flip_addrs flip_addrs[MAX_SURFACES];
-		struct dc_stream_update stream_update;
-	} *bundle;
-
-	bundle = kzalloc(sizeof(*bundle), GFP_KERNEL);
-
-	if (!bundle) {
-		DRM_ERROR("Failed to allocate update bundle\n");
-		/* Set type to FULL to avoid crashing in DC*/
-		update_type = UPDATE_TYPE_FULL;
-		goto cleanup;
-	}
-
-	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
-
-		memset(bundle, 0, sizeof(struct surface_info_bundle));
-
-		new_dm_crtc_state = to_dm_crtc_state(new_crtc_state);
-		old_dm_crtc_state = to_dm_crtc_state(old_crtc_state);
-		num_plane = 0;
-
-		if (new_dm_crtc_state->stream != old_dm_crtc_state->stream) {
-			update_type = UPDATE_TYPE_FULL;
-			goto cleanup;
-		}
-
-		if (!new_dm_crtc_state->stream)
-			continue;
-
-		for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, j) {
-			const struct amdgpu_framebuffer *amdgpu_fb =
-				to_amdgpu_framebuffer(new_plane_state->fb);
-			struct dc_plane_info *plane_info = &bundle->plane_infos[num_plane];
-			struct dc_flip_addrs *flip_addr = &bundle->flip_addrs[num_plane];
-			struct dc_scaling_info *scaling_info = &bundle->scaling_infos[num_plane];
-			uint64_t tiling_flags;
-			bool tmz_surface = false;
-
-			new_plane_crtc = new_plane_state->crtc;
-			new_dm_plane_state = to_dm_plane_state(new_plane_state);
-			old_dm_plane_state = to_dm_plane_state(old_plane_state);
-
-			if (plane->type == DRM_PLANE_TYPE_CURSOR)
-				continue;
-
-			if (new_dm_plane_state->dc_state != old_dm_plane_state->dc_state) {
-				update_type = UPDATE_TYPE_FULL;
-				goto cleanup;
-			}
-
-			if (crtc != new_plane_crtc)
-				continue;
-
-			bundle->surface_updates[num_plane].surface =
-					new_dm_plane_state->dc_state;
-
-			if (new_crtc_state->mode_changed) {
-				bundle->stream_update.dst = new_dm_crtc_state->stream->dst;
-				bundle->stream_update.src = new_dm_crtc_state->stream->src;
-			}
-
-			if (new_crtc_state->color_mgmt_changed) {
-				bundle->surface_updates[num_plane].gamma =
-						new_dm_plane_state->dc_state->gamma_correction;
-				bundle->surface_updates[num_plane].in_transfer_func =
-						new_dm_plane_state->dc_state->in_transfer_func;
-				bundle->surface_updates[num_plane].gamut_remap_matrix =
-						&new_dm_plane_state->dc_state->gamut_remap_matrix;
-				bundle->stream_update.gamut_remap =
-						&new_dm_crtc_state->stream->gamut_remap_matrix;
-				bundle->stream_update.output_csc_transform =
-						&new_dm_crtc_state->stream->csc_color_matrix;
-				bundle->stream_update.out_transfer_func =
-						new_dm_crtc_state->stream->out_transfer_func;
-			}
-
-			ret = fill_dc_scaling_info(new_plane_state,
-						   scaling_info);
-			if (ret)
-				goto cleanup;
-
-			bundle->surface_updates[num_plane].scaling_info = scaling_info;
-
-			if (amdgpu_fb) {
-				ret = get_fb_info(amdgpu_fb, &tiling_flags, &tmz_surface);
-				if (ret)
-					goto cleanup;
-
-				ret = fill_dc_plane_info_and_addr(
-					dm->adev, new_plane_state, tiling_flags,
-					plane_info,
-					&flip_addr->address, tmz_surface,
-					false);
-				if (ret)
-					goto cleanup;
-
-				bundle->surface_updates[num_plane].plane_info = plane_info;
-				bundle->surface_updates[num_plane].flip_addr = flip_addr;
-			}
-
-			num_plane++;
-		}
-
-		if (num_plane == 0)
-			continue;
-
-		ret = dm_atomic_get_state(state, &dm_state);
-		if (ret)
-			goto cleanup;
-
-		old_dm_state = dm_atomic_get_old_state(state);
-		if (!old_dm_state) {
-			ret = -EINVAL;
-			goto cleanup;
-		}
-
-		status = dc_stream_get_status_from_state(old_dm_state->context,
-							 new_dm_crtc_state->stream);
-		bundle->stream_update.stream = new_dm_crtc_state->stream;
-		/*
-		 * TODO: DC modifies the surface during this call so we need
-		 * to lock here - find a way to do this without locking.
-		 */
-		mutex_lock(&dm->dc_lock);
-		update_type = dc_check_update_surfaces_for_stream(
-				dc,	bundle->surface_updates, num_plane,
-				&bundle->stream_update, status);
-		mutex_unlock(&dm->dc_lock);
-
-		if (update_type > UPDATE_TYPE_MED) {
-			update_type = UPDATE_TYPE_FULL;
-			goto cleanup;
-		}
-	}
-
-cleanup:
-	kfree(bundle);
-
-	*out_type = update_type;
-	return ret;
-}
 #if defined(CONFIG_DRM_AMD_DC_DCN)
 static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm_crtc *crtc)
 {
@@ -8508,8 +8555,7 @@ static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm
  * acquired. For full updates case which removes/adds/updates streams on one
  * CRTC while flipping on another CRTC, acquiring global lock will guarantee
  * that any such full update commit will wait for completion of any outstanding
- * flip using DRMs synchronization events. See
- * dm_determine_update_type_for_commit()
+ * flip using DRMs synchronization events.
  *
  * Note that DM adds the affected connectors for all CRTCs in state, when that
  * might not seem necessary. This is because DC stream creation requires the
@@ -8521,7 +8567,7 @@ static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm
 static int amdgpu_dm_atomic_check(struct drm_device *dev,
 				  struct drm_atomic_state *state)
 {
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct dm_atomic_state *dm_state = NULL;
 	struct dc *dc = adev->dm.dc;
 	struct drm_connector *connector;
@@ -8530,17 +8576,12 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
 	struct drm_plane *plane;
 	struct drm_plane_state *old_plane_state, *new_plane_state;
-	enum surface_update_type update_type = UPDATE_TYPE_FAST;
-	enum surface_update_type overall_update_type = UPDATE_TYPE_FAST;
 	enum dc_status status;
 	int ret, i;
-
-	/*
-	 * This bool will be set for true for any modeset/reset
-	 * or plane update which implies non fast surface update.
-	 */
 	bool lock_and_validation_needed = false;
 
+	amdgpu_check_debugfs_connector_property_change(adev, state);
+
 	ret = drm_atomic_helper_check_modeset(dev, state);
 	if (ret)
 		goto fail;
@@ -8633,6 +8674,17 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		}
 	}
 
+	/* Prepass for updating tiling flags on new planes. */
+	for_each_new_plane_in_state(state, plane, new_plane_state, i) {
+		struct dm_plane_state *new_dm_plane_state = to_dm_plane_state(new_plane_state);
+		struct amdgpu_framebuffer *new_afb = to_amdgpu_framebuffer(new_plane_state->fb);
+
+		ret = get_fb_info(new_afb, &new_dm_plane_state->tiling_flags,
+				  &new_dm_plane_state->tmz_surface);
+		if (ret)
+			goto fail;
+	}
+
 	/* Remove exiting planes if they are modified */
 	for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) {
 		ret = dm_update_plane_state(dc, state, plane,
@@ -8721,27 +8773,23 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		if (!is_scaling_state_different(dm_new_con_state, dm_old_con_state))
 			continue;
 
-		overall_update_type = UPDATE_TYPE_FULL;
 		lock_and_validation_needed = true;
 	}
 
-	ret = dm_determine_update_type_for_commit(&adev->dm, state, &update_type);
-	if (ret)
-		goto fail;
-
-	if (overall_update_type < update_type)
-		overall_update_type = update_type;
-
-	/*
-	 * lock_and_validation_needed was an old way to determine if we need to set
-	 * the global lock. Leaving it in to check if we broke any corner cases
-	 * lock_and_validation_needed true = UPDATE_TYPE_FULL or UPDATE_TYPE_MED
-	 * lock_and_validation_needed false = UPDATE_TYPE_FAST
+	/**
+	 * Streams and planes are reset when there are changes that affect
+	 * bandwidth. Anything that affects bandwidth needs to go through
+	 * DC global validation to ensure that the configuration can be applied
+	 * to hardware.
+	 *
+	 * We have to currently stall out here in atomic_check for outstanding
+	 * commits to finish in this case because our IRQ handlers reference
+	 * DRM state directly - we can end up disabling interrupts too early
+	 * if we don't.
+	 *
+	 * TODO: Remove this stall and drop DM state private objects.
 	 */
-	if (lock_and_validation_needed && overall_update_type <= UPDATE_TYPE_FAST)
-		WARN(1, "Global lock should be Set, overall_update_type should be UPDATE_TYPE_MED or UPDATE_TYPE_FULL");
-
-	if (overall_update_type > UPDATE_TYPE_FAST) {
+	if (lock_and_validation_needed) {
 		ret = dm_atomic_get_state(state, &dm_state);
 		if (ret)
 			goto fail;
@@ -8823,7 +8871,9 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		struct dm_crtc_state *dm_new_crtc_state =
 			to_dm_crtc_state(new_crtc_state);
 
-		dm_new_crtc_state->update_type = (int)overall_update_type;
+		dm_new_crtc_state->update_type = lock_and_validation_needed ?
+							 UPDATE_TYPE_FULL :
+							 UPDATE_TYPE_FAST;
 	}
 
 	/* Must be success */
@@ -8872,7 +8922,7 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 	struct dm_connector_state *dm_con_state = NULL;
 
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	bool freesync_capable = false;
 
 	if (!connector->state) {
@@ -9071,3 +9121,34 @@ static bool amdgpu_dm_psr_disable(struct dc_stream_state *stream)
 
 	return dc_link_set_psr_allow_active(stream->link, false, true);
 }
+
+/*
+ * amdgpu_dm_psr_disable() - disable psr f/w
+ * if psr is enabled on any stream
+ *
+ * Return: true if success
+ */
+static bool amdgpu_dm_psr_disable_all(struct amdgpu_display_manager *dm)
+{
+	DRM_DEBUG_DRIVER("Disabling psr if psr is enabled on any stream\n");
+	return dc_set_psr_allow_active(dm->dc, false);
+}
+
+void amdgpu_dm_trigger_timing_sync(struct drm_device *dev)
+{
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	struct dc *dc = adev->dm.dc;
+	int i;
+
+	mutex_lock(&adev->dm.dc_lock);
+	if (dc->current_state) {
+		for (i = 0; i < dc->current_state->stream_count; ++i)
+			dc->current_state->streams[i]
+				->triggered_crtc_reset.enabled =
+				adev->dm.force_timing_sync;
+
+		dm_enable_per_frame_crtc_master_sync(dc->current_state);
+		dc_trigger_sync(dc, dc->current_state);
+	}
+	mutex_unlock(&adev->dm.dc_lock);
+}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index dd1559c743c2..9c1e003d9c29 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -340,6 +340,20 @@ struct amdgpu_display_manager {
 	 * fake encoders used for DP MST.
 	 */
 	struct amdgpu_encoder mst_encoders[AMDGPU_DM_MAX_CRTC];
+	bool force_timing_sync;
+};
+
+enum dsc_clock_force_state {
+	DSC_CLK_FORCE_DEFAULT = 0,
+	DSC_CLK_FORCE_ENABLE,
+	DSC_CLK_FORCE_DISABLE,
+};
+
+struct dsc_preferred_settings {
+	enum dsc_clock_force_state dsc_force_enable;
+	uint32_t dsc_num_slices_v;
+	uint32_t dsc_num_slices_h;
+	uint32_t dsc_bits_per_pixel;
 };
 
 struct amdgpu_dm_connector {
@@ -389,6 +403,7 @@ struct amdgpu_dm_connector {
 	uint32_t debugfs_dpcd_size;
 #endif
 	bool force_yuv420_output;
+	struct dsc_preferred_settings dsc_settings;
 };
 
 #define to_amdgpu_dm_connector(x) container_of(x, struct amdgpu_dm_connector, base)
@@ -403,6 +418,8 @@ struct dc_plane_state;
 struct dm_plane_state {
 	struct drm_plane_state base;
 	struct dc_plane_state *dc_state;
+	uint64_t tiling_flags;
+	bool tmz_surface;
 };
 
 struct dm_crtc_state {
@@ -423,7 +440,6 @@ struct dm_crtc_state {
 
 	bool vrr_supported;
 	struct mod_freesync_config freesync_config;
-	struct mod_vrr_params vrr_params;
 	struct dc_info_packet vrr_infopacket;
 
 	int abm_level;
@@ -485,6 +501,8 @@ void dm_restore_drm_connector_state(struct drm_device *dev,
 void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 					struct edid *edid);
 
+void amdgpu_dm_trigger_timing_sync(struct drm_device *dev);
+
 #define MAX_COLOR_LUT_ENTRIES 4096
 /* Legacy gamm LUT users such as X doesn't like large LUT sizes */
 #define MAX_COLOR_LEGACY_LUT_ENTRIES 256
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index b321ff654df4..5df05f0d18bc 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -308,8 +308,7 @@ static int __set_input_tf(struct dc_transfer_func *func,
 int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 {
 	struct dc_stream_state *stream = crtc->stream;
-	struct amdgpu_device *adev =
-		(struct amdgpu_device *)crtc->base.state->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->base.state->dev);
 	bool has_rom = adev->asic_type <= CHIP_RAVEN;
 	struct drm_color_ctm *ctm = NULL;
 	const struct drm_color_lut *degamma_lut, *regamma_lut;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
index eaad9099bc0b..d0699e98db92 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
@@ -101,7 +101,7 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc,
 					struct dm_crtc_state *dm_crtc_state,
 					enum amdgpu_dm_pipe_crc_source source)
 {
-	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct dc_stream_state *stream_state = dm_crtc_state->stream;
 	bool enable = amdgpu_dm_is_valid_crc_source(source);
 	int ret = 0;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index e5a6d9115949..8cd646eef096 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -49,6 +49,10 @@ struct dmub_debugfs_trace_entry {
 	uint32_t param1;
 };
 
+static inline const char *yesno(bool v)
+{
+	return v ? "yes" : "no";
+}
 
 /* parse_write_buffer_into_params - Helper function to parse debugfs write buffer into an array
  *
@@ -107,7 +111,6 @@ static int parse_write_buffer_into_params(char *wr_buf, uint32_t wr_buf_size,
 
 	if (*param_nums > max_param_num)
 		*param_nums = max_param_num;
-;
 
 	wr_buf_ptr = wr_buf; /* reset buf pointer */
 	wr_buf_count = 0; /* number of char already checked */
@@ -261,7 +264,7 @@ static ssize_t dp_link_settings_write(struct file *f, const char __user *buf,
 	if (!wr_buf)
 		return -ENOSPC;
 
-	if (parse_write_buffer_into_params(wr_buf, wr_buf_size,
+	if (parse_write_buffer_into_params(wr_buf, size,
 					   (long *)param, buf,
 					   max_param_num,
 					   &param_nums)) {
@@ -420,7 +423,7 @@ static ssize_t dp_phy_settings_write(struct file *f, const char __user *buf,
 	if (!wr_buf)
 		return -ENOSPC;
 
-	if (parse_write_buffer_into_params(wr_buf, wr_buf_size,
+	if (parse_write_buffer_into_params(wr_buf, size,
 					   (long *)param, buf,
 					   max_param_num,
 					   &param_nums)) {
@@ -572,7 +575,7 @@ static ssize_t dp_phy_test_pattern_debugfs_write(struct file *f, const char __us
 	if (!wr_buf)
 		return -ENOSPC;
 
-	if (parse_write_buffer_into_params(wr_buf, wr_buf_size,
+	if (parse_write_buffer_into_params(wr_buf, size,
 					   (long *)param, buf,
 					   max_param_num,
 					   &param_nums)) {
@@ -905,7 +908,7 @@ static ssize_t dp_dpcd_address_write(struct file *f, const char __user *buf,
 	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
 
 	if (size < sizeof(connector->debugfs_dpcd_address))
-		return 0;
+		return -EINVAL;
 
 	r = copy_from_user(&connector->debugfs_dpcd_address,
 			buf, sizeof(connector->debugfs_dpcd_address));
@@ -920,7 +923,7 @@ static ssize_t dp_dpcd_size_write(struct file *f, const char __user *buf,
 	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
 
 	if (size < sizeof(connector->debugfs_dpcd_size))
-		return 0;
+		return -EINVAL;
 
 	r = copy_from_user(&connector->debugfs_dpcd_size,
 			buf, sizeof(connector->debugfs_dpcd_size));
@@ -940,8 +943,8 @@ static ssize_t dp_dpcd_data_write(struct file *f, const char __user *buf,
 	struct dc_link *link = connector->dc_link;
 	uint32_t write_size = connector->debugfs_dpcd_size;
 
-	if (size < write_size)
-		return 0;
+	if (!write_size || size < write_size)
+		return -EINVAL;
 
 	data = kzalloc(write_size, GFP_KERNEL);
 	if (!data)
@@ -964,7 +967,7 @@ static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf,
 	struct dc_link *link = connector->dc_link;
 	uint32_t read_size = connector->debugfs_dpcd_size;
 
-	if (size < read_size)
+	if (!read_size || size < read_size)
 		return 0;
 
 	data = kzalloc(read_size, GFP_KERNEL);
@@ -980,6 +983,190 @@ static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf,
 	return read_size - r;
 }
 
+/* function: Read link's DSC & FEC capabilities
+ *
+ *
+ * Access it with the following command (you need to specify
+ * connector like DP-1):
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dp_dsc_fec_support
+ *
+ */
+static int dp_dsc_fec_support_show(struct seq_file *m, void *data)
+{
+	struct drm_connector *connector = m->private;
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_device *dev = connector->dev;
+	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
+	int ret = 0;
+	bool try_again = false;
+	bool is_fec_supported = false;
+	bool is_dsc_supported = false;
+	struct dpcd_caps dpcd_caps;
+
+	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
+	do {
+		try_again = false;
+		ret = drm_modeset_lock(&dev->mode_config.connection_mutex, &ctx);
+		if (ret) {
+			if (ret == -EDEADLK) {
+				ret = drm_modeset_backoff(&ctx);
+				if (!ret) {
+					try_again = true;
+					continue;
+				}
+			}
+			break;
+		}
+		if (connector->status != connector_status_connected) {
+			ret = -ENODEV;
+			break;
+		}
+		dpcd_caps = aconnector->dc_link->dpcd_caps;
+		if (aconnector->port) {
+			/* aconnector sets dsc_aux during get_modes call
+			 * if MST connector has it means it can either
+			 * enable DSC on the sink device or on MST branch
+			 * its connected to.
+			 */
+			if (aconnector->dsc_aux) {
+				is_fec_supported = true;
+				is_dsc_supported = true;
+			}
+		} else {
+			is_fec_supported = dpcd_caps.fec_cap.raw & 0x1;
+			is_dsc_supported = dpcd_caps.dsc_caps.dsc_basic_caps.raw[0] & 0x1;
+		}
+	} while (try_again);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	seq_printf(m, "FEC_Sink_Support: %s\n", yesno(is_fec_supported));
+	seq_printf(m, "DSC_Sink_Support: %s\n", yesno(is_dsc_supported));
+
+	return ret;
+}
+
+/* function: Trigger virtual HPD redetection on connector
+ *
+ * This function will perform link rediscovery, link disable
+ * and enable, and dm connector state update.
+ *
+ * Retrigger HPD on an existing connector by echoing 1 into
+ * its respectful "trigger_hotplug" debugfs entry:
+ *
+ *	echo 1 > /sys/kernel/debug/dri/0/DP-X/trigger_hotplug
+ *
+ * This function can perform HPD unplug:
+ *
+ *	echo 0 > /sys/kernel/debug/dri/0/DP-X/trigger_hotplug
+ *
+ */
+static ssize_t dp_trigger_hotplug(struct file *f, const char __user *buf,
+							size_t size, loff_t *pos)
+{
+	struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private;
+	struct drm_connector *connector = &aconnector->base;
+	struct dc_link *link = NULL;
+	struct drm_device *dev = connector->dev;
+	enum dc_connection_type new_connection_type = dc_connection_none;
+	char *wr_buf = NULL;
+	uint32_t wr_buf_size = 42;
+	int max_param_num = 1;
+	long param[1] = {0};
+	uint8_t param_nums = 0;
+
+	if (!aconnector || !aconnector->dc_link)
+		return -EINVAL;
+
+	if (size == 0)
+		return -EINVAL;
+
+	wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL);
+
+	if (!wr_buf) {
+		DRM_DEBUG_DRIVER("no memory to allocate write buffer\n");
+		return -ENOSPC;
+	}
+
+	if (parse_write_buffer_into_params(wr_buf, size,
+						(long *)param, buf,
+						max_param_num,
+						&param_nums)) {
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param_nums <= 0) {
+		DRM_DEBUG_DRIVER("user data not be read\n");
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param[0] == 1) {
+		mutex_lock(&aconnector->hpd_lock);
+
+		if (!dc_link_detect_sink(aconnector->dc_link, &new_connection_type) &&
+			new_connection_type != dc_connection_none)
+			goto unlock;
+
+		if (!dc_link_detect(aconnector->dc_link, DETECT_REASON_HPD))
+			goto unlock;
+
+		amdgpu_dm_update_connector_after_detect(aconnector);
+
+		drm_modeset_lock_all(dev);
+		dm_restore_drm_connector_state(dev, connector);
+		drm_modeset_unlock_all(dev);
+
+		drm_kms_helper_hotplug_event(dev);
+	} else if (param[0] == 0) {
+		if (!aconnector->dc_link)
+			goto unlock;
+
+		link = aconnector->dc_link;
+
+		if (link->local_sink) {
+			dc_sink_release(link->local_sink);
+			link->local_sink = NULL;
+		}
+
+		link->dpcd_sink_count = 0;
+		link->type = dc_connection_none;
+		link->dongle_max_pix_clk = 0;
+
+		amdgpu_dm_update_connector_after_detect(aconnector);
+
+		drm_modeset_lock_all(dev);
+		dm_restore_drm_connector_state(dev, connector);
+		drm_modeset_unlock_all(dev);
+
+		drm_kms_helper_hotplug_event(dev);
+	}
+
+unlock:
+	mutex_unlock(&aconnector->hpd_lock);
+
+	kfree(wr_buf);
+	return size;
+}
+
+/* function: read DSC status on the connector
+ *
+ * The read function: dp_dsc_clock_en_read
+ * returns current status of DSC clock on the connector.
+ * The return is a boolean flag: 1 or 0.
+ *
+ * Access it with the following command (you need to specify
+ * connector like DP-1):
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_clock_en
+ *
+ * Expected output:
+ * 1 - means that DSC is currently enabled
+ * 0 - means that DSC is disabled
+ */
 static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1037,6 +1224,105 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: write force DSC on the connector
+ *
+ * The write function: dp_dsc_clock_en_write
+ * enables to force DSC on the connector.
+ * User can write to either force enable or force disable DSC
+ * on the next modeset or set it to driver default
+ *
+ * Accepted inputs:
+ * 0 - default DSC enablement policy
+ * 1 - force enable DSC on the connector
+ * 2 - force disable DSC on the connector (might cause fail in atomic_check)
+ *
+ * Writing DSC settings is done with the following command:
+ * - To force enable DSC (you need to specify
+ * connector like DP-1):
+ *
+ *	echo 0x1 > /sys/kernel/debug/dri/0/DP-X/dsc_clock_en
+ *
+ * - To return to default state set the flag to zero and
+ * let driver deal with DSC automatically
+ * (you need to specify connector like DP-1):
+ *
+ *	echo 0x0 > /sys/kernel/debug/dri/0/DP-X/dsc_clock_en
+ *
+ */
+static ssize_t dp_dsc_clock_en_write(struct file *f, const char __user *buf,
+				     size_t size, loff_t *pos)
+{
+	struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private;
+	struct pipe_ctx *pipe_ctx;
+	int i;
+	char *wr_buf = NULL;
+	uint32_t wr_buf_size = 42;
+	int max_param_num = 1;
+	long param[1] = {0};
+	uint8_t param_nums = 0;
+
+	if (size == 0)
+		return -EINVAL;
+
+	wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL);
+
+	if (!wr_buf) {
+		DRM_DEBUG_DRIVER("no memory to allocate write buffer\n");
+		return -ENOSPC;
+	}
+
+	if (parse_write_buffer_into_params(wr_buf, size,
+					    (long *)param, buf,
+					    max_param_num,
+					    &param_nums)) {
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param_nums <= 0) {
+		DRM_DEBUG_DRIVER("user data not be read\n");
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < MAX_PIPES; i++) {
+		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe_ctx && pipe_ctx->stream &&
+			    pipe_ctx->stream->link == aconnector->dc_link)
+				break;
+	}
+
+	if (!pipe_ctx || !pipe_ctx->stream)
+		goto done;
+
+	if (param[0] == 1)
+		aconnector->dsc_settings.dsc_force_enable = DSC_CLK_FORCE_ENABLE;
+	else if (param[0] == 2)
+		aconnector->dsc_settings.dsc_force_enable = DSC_CLK_FORCE_DISABLE;
+	else
+		aconnector->dsc_settings.dsc_force_enable = DSC_CLK_FORCE_DEFAULT;
+
+done:
+	kfree(wr_buf);
+	return size;
+}
+
+/* function: read DSC slice width parameter on the connector
+ *
+ * The read function: dp_dsc_slice_width_read
+ * returns dsc slice width used in the current configuration
+ * The return is an integer: 0 or other positive number
+ *
+ * Access the status with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_slice_width
+ *
+ * 0 - means that DSC is disabled
+ *
+ * Any other number more than zero represents the
+ * slice width currently used by DSC in pixels
+ *
+ */
 static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1094,6 +1380,103 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: write DSC slice width parameter
+ *
+ * The write function: dp_dsc_slice_width_write
+ * overwrites automatically generated DSC configuration
+ * of slice width.
+ *
+ * The user has to write the slice width divisible by the
+ * picture width.
+ *
+ * Also the user has to write width in hexidecimal
+ * rather than in decimal.
+ *
+ * Writing DSC settings is done with the following command:
+ * - To force overwrite slice width: (example sets to 1920 pixels)
+ *
+ *	echo 0x780 > /sys/kernel/debug/dri/0/DP-X/dsc_slice_width
+ *
+ *  - To stop overwriting and let driver find the optimal size,
+ * set the width to zero:
+ *
+ *	echo 0x0 > /sys/kernel/debug/dri/0/DP-X/dsc_slice_width
+ *
+ */
+static ssize_t dp_dsc_slice_width_write(struct file *f, const char __user *buf,
+				     size_t size, loff_t *pos)
+{
+	struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private;
+	struct pipe_ctx *pipe_ctx;
+	int i;
+	char *wr_buf = NULL;
+	uint32_t wr_buf_size = 42;
+	int max_param_num = 1;
+	long param[1] = {0};
+	uint8_t param_nums = 0;
+
+	if (size == 0)
+		return -EINVAL;
+
+	wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL);
+
+	if (!wr_buf) {
+		DRM_DEBUG_DRIVER("no memory to allocate write buffer\n");
+		return -ENOSPC;
+	}
+
+	if (parse_write_buffer_into_params(wr_buf, size,
+					    (long *)param, buf,
+					    max_param_num,
+					    &param_nums)) {
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param_nums <= 0) {
+		DRM_DEBUG_DRIVER("user data not be read\n");
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < MAX_PIPES; i++) {
+		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe_ctx && pipe_ctx->stream &&
+			    pipe_ctx->stream->link == aconnector->dc_link)
+				break;
+	}
+
+	if (!pipe_ctx || !pipe_ctx->stream)
+		goto done;
+
+	if (param[0] > 0)
+		aconnector->dsc_settings.dsc_num_slices_h = DIV_ROUND_UP(
+					pipe_ctx->stream->timing.h_addressable,
+					param[0]);
+	else
+		aconnector->dsc_settings.dsc_num_slices_h = 0;
+
+done:
+	kfree(wr_buf);
+	return size;
+}
+
+/* function: read DSC slice height parameter on the connector
+ *
+ * The read function: dp_dsc_slice_height_read
+ * returns dsc slice height used in the current configuration
+ * The return is an integer: 0 or other positive number
+ *
+ * Access the status with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_slice_height
+ *
+ * 0 - means that DSC is disabled
+ *
+ * Any other number more than zero represents the
+ * slice height currently used by DSC in pixels
+ *
+ */
 static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1151,6 +1534,99 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: write DSC slice height parameter
+ *
+ * The write function: dp_dsc_slice_height_write
+ * overwrites automatically generated DSC configuration
+ * of slice height.
+ *
+ * The user has to write the slice height divisible by the
+ * picture height.
+ *
+ * Also the user has to write height in hexidecimal
+ * rather than in decimal.
+ *
+ * Writing DSC settings is done with the following command:
+ * - To force overwrite slice height (example sets to 128 pixels):
+ *
+ *	echo 0x80 > /sys/kernel/debug/dri/0/DP-X/dsc_slice_height
+ *
+ *  - To stop overwriting and let driver find the optimal size,
+ * set the height to zero:
+ *
+ *	echo 0x0 > /sys/kernel/debug/dri/0/DP-X/dsc_slice_height
+ *
+ */
+static ssize_t dp_dsc_slice_height_write(struct file *f, const char __user *buf,
+				     size_t size, loff_t *pos)
+{
+	struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private;
+	struct pipe_ctx *pipe_ctx;
+	int i;
+	char *wr_buf = NULL;
+	uint32_t wr_buf_size = 42;
+	int max_param_num = 1;
+	uint8_t param_nums = 0;
+	long param[1] = {0};
+
+	if (size == 0)
+		return -EINVAL;
+
+	wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL);
+
+	if (!wr_buf) {
+		DRM_DEBUG_DRIVER("no memory to allocate write buffer\n");
+		return -ENOSPC;
+	}
+
+	if (parse_write_buffer_into_params(wr_buf, size,
+					    (long *)param, buf,
+					    max_param_num,
+					    &param_nums)) {
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param_nums <= 0) {
+		DRM_DEBUG_DRIVER("user data not be read\n");
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < MAX_PIPES; i++) {
+		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe_ctx && pipe_ctx->stream &&
+			    pipe_ctx->stream->link == aconnector->dc_link)
+				break;
+	}
+
+	if (!pipe_ctx || !pipe_ctx->stream)
+		goto done;
+
+	if (param[0] > 0)
+		aconnector->dsc_settings.dsc_num_slices_v = DIV_ROUND_UP(
+					pipe_ctx->stream->timing.v_addressable,
+					param[0]);
+	else
+		aconnector->dsc_settings.dsc_num_slices_v = 0;
+
+done:
+	kfree(wr_buf);
+	return size;
+}
+
+/* function: read DSC target rate on the connector in bits per pixel
+ *
+ * The read function: dp_dsc_bits_per_pixel_read
+ * returns target rate of compression in bits per pixel
+ * The return is an integer: 0 or other positive integer
+ *
+ * Access it with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_bits_per_pixel
+ *
+ *  0 - means that DSC is disabled
+ */
 static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1208,6 +1684,94 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: write DSC target rate in bits per pixel
+ *
+ * The write function: dp_dsc_bits_per_pixel_write
+ * overwrites automatically generated DSC configuration
+ * of DSC target bit rate.
+ *
+ * Also the user has to write bpp in hexidecimal
+ * rather than in decimal.
+ *
+ * Writing DSC settings is done with the following command:
+ * - To force overwrite rate (example sets to 256 bpp x 1/16):
+ *
+ *	echo 0x100 > /sys/kernel/debug/dri/0/DP-X/dsc_bits_per_pixel
+ *
+ *  - To stop overwriting and let driver find the optimal rate,
+ * set the rate to zero:
+ *
+ *	echo 0x0 > /sys/kernel/debug/dri/0/DP-X/dsc_bits_per_pixel
+ *
+ */
+static ssize_t dp_dsc_bits_per_pixel_write(struct file *f, const char __user *buf,
+				     size_t size, loff_t *pos)
+{
+	struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private;
+	struct pipe_ctx *pipe_ctx;
+	int i;
+	char *wr_buf = NULL;
+	uint32_t wr_buf_size = 42;
+	int max_param_num = 1;
+	uint8_t param_nums = 0;
+	long param[1] = {0};
+
+	if (size == 0)
+		return -EINVAL;
+
+	wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL);
+
+	if (!wr_buf) {
+		DRM_DEBUG_DRIVER("no memory to allocate write buffer\n");
+		return -ENOSPC;
+	}
+
+	if (parse_write_buffer_into_params(wr_buf, size,
+					    (long *)param, buf,
+					    max_param_num,
+					    &param_nums)) {
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	if (param_nums <= 0) {
+		DRM_DEBUG_DRIVER("user data not be read\n");
+		kfree(wr_buf);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < MAX_PIPES; i++) {
+		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe_ctx && pipe_ctx->stream &&
+			    pipe_ctx->stream->link == aconnector->dc_link)
+				break;
+	}
+
+	if (!pipe_ctx || !pipe_ctx->stream)
+		goto done;
+
+	aconnector->dsc_settings.dsc_bits_per_pixel = param[0];
+
+done:
+	kfree(wr_buf);
+	return size;
+}
+
+/* function: read DSC picture width parameter on the connector
+ *
+ * The read function: dp_dsc_pic_width_read
+ * returns dsc picture width used in the current configuration
+ * It is the same as h_addressable of the current
+ * display's timing
+ * The return is an integer: 0 or other positive integer
+ * If 0 then DSC is disabled.
+ *
+ * Access it with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_pic_width
+ *
+ * 0 - means that DSC is disabled
+ */
 static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1322,6 +1886,21 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: read DSC chunk size parameter on the connector
+ *
+ * The read function: dp_dsc_chunk_size_read
+ * returns dsc chunk size set in the current configuration
+ * The value is calculated automatically by DSC code
+ * and depends on slice parameters and bpp target rate
+ * The return is an integer: 0 or other positive integer
+ * If 0 then DSC is disabled.
+ *
+ * Access it with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_chunk_size
+ *
+ * 0 - means that DSC is disabled
+ */
 static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1379,6 +1958,21 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf,
 	return result;
 }
 
+/* function: read DSC slice bpg offset on the connector
+ *
+ * The read function: dp_dsc_slice_bpg_offset_read
+ * returns dsc bpg slice offset set in the current configuration
+ * The value is calculated automatically by DSC code
+ * and depends on slice parameters and bpp target rate
+ * The return is an integer: 0 or other positive integer
+ * If 0 then DSC is disabled.
+ *
+ * Access it with the following command:
+ *
+ *	cat /sys/kernel/debug/dri/0/DP-X/dsc_slice_bpg_offset
+ *
+ * 0 - means that DSC is disabled
+ */
 static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf,
 				    size_t size, loff_t *pos)
 {
@@ -1436,6 +2030,7 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf,
 	return result;
 }
 
+DEFINE_SHOW_ATTRIBUTE(dp_dsc_fec_support);
 DEFINE_SHOW_ATTRIBUTE(dmub_fw_state);
 DEFINE_SHOW_ATTRIBUTE(dmub_tracebuffer);
 DEFINE_SHOW_ATTRIBUTE(output_bpc);
@@ -1446,24 +2041,28 @@ DEFINE_SHOW_ATTRIBUTE(hdcp_sink_capability);
 static const struct file_operations dp_dsc_clock_en_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_dsc_clock_en_read,
+	.write = dp_dsc_clock_en_write,
 	.llseek = default_llseek
 };
 
 static const struct file_operations dp_dsc_slice_width_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_dsc_slice_width_read,
+	.write = dp_dsc_slice_width_write,
 	.llseek = default_llseek
 };
 
 static const struct file_operations dp_dsc_slice_height_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_dsc_slice_height_read,
+	.write = dp_dsc_slice_height_write,
 	.llseek = default_llseek
 };
 
 static const struct file_operations dp_dsc_bits_per_pixel_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_dsc_bits_per_pixel_read,
+	.write = dp_dsc_bits_per_pixel_write,
 	.llseek = default_llseek
 };
 
@@ -1491,6 +2090,12 @@ static const struct file_operations dp_dsc_slice_bpg_offset_debugfs_fops = {
 	.llseek = default_llseek
 };
 
+static const struct file_operations dp_trigger_hotplug_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.write = dp_trigger_hotplug,
+	.llseek = default_llseek
+};
+
 static const struct file_operations dp_link_settings_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_link_settings_read,
@@ -1541,6 +2146,7 @@ static const struct {
 	const struct file_operations *fops;
 } dp_debugfs_entries[] = {
 		{"link_settings", &dp_link_settings_debugfs_fops},
+		{"trigger_hotplug", &dp_trigger_hotplug_debugfs_fops},
 		{"phy_settings", &dp_phy_settings_debugfs_fop},
 		{"test_pattern", &dp_phy_test_pattern_fops},
 #ifdef CONFIG_DRM_AMD_DC_HDCP
@@ -1557,7 +2163,8 @@ static const struct {
 		{"dsc_pic_width", &dp_dsc_pic_width_debugfs_fops},
 		{"dsc_pic_height", &dp_dsc_pic_height_debugfs_fops},
 		{"dsc_chunk_size", &dp_dsc_chunk_size_debugfs_fops},
-		{"dsc_slice_bpg", &dp_dsc_slice_bpg_offset_debugfs_fops}
+		{"dsc_slice_bpg", &dp_dsc_slice_bpg_offset_debugfs_fops},
+		{"dp_dsc_fec_support", &dp_dsc_fec_support_fops}
 };
 
 #ifdef CONFIG_DRM_AMD_DC_HDCP
@@ -1721,7 +2328,7 @@ static int current_backlight_read(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 
 	unsigned int backlight = dc_link_get_backlight_level(dm->backlight_link);
@@ -1739,7 +2346,7 @@ static int target_backlight_read(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_display_manager *dm = &adev->dm;
 
 	unsigned int backlight = dc_link_get_target_backlight_pwm(dm->backlight_link);
@@ -1778,6 +2385,38 @@ static const struct drm_info_list amdgpu_dm_debugfs_list[] = {
 };
 
 /*
+ * Sets the force_timing_sync debug optino from the given string.
+ * All connected displays will be force synchronized immediately.
+ * Usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_force_timing_sync
+ */
+static int force_timing_sync_set(void *data, u64 val)
+{
+	struct amdgpu_device *adev = data;
+
+	adev->dm.force_timing_sync = (bool)val;
+
+	amdgpu_dm_trigger_timing_sync(adev_to_drm(adev));
+
+	return 0;
+}
+
+/*
+ * Gets the force_timing_sync debug option value into the given buffer.
+ * Usage: cat /sys/kernel/debug/dri/0/amdgpu_dm_force_timing_sync
+ */
+static int force_timing_sync_get(void *data, u64 *val)
+{
+	struct amdgpu_device *adev = data;
+
+	*val = adev->dm.force_timing_sync;
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(force_timing_sync_ops, force_timing_sync_get,
+			 force_timing_sync_set, "%llu\n");
+
+/*
  * Sets the DC visual confirm debug option from the given string.
  * Example usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_visual_confirm
  */
@@ -1815,7 +2454,7 @@ int dtn_debugfs_init(struct amdgpu_device *adev)
 		.llseek = default_llseek
 	};
 
-	struct drm_minor *minor = adev->ddev->primary;
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
 	struct dentry *root = minor->debugfs_root;
 	int ret;
 
@@ -1836,5 +2475,8 @@ int dtn_debugfs_init(struct amdgpu_device *adev)
 	debugfs_create_file_unsafe("amdgpu_dm_dmub_fw_state", 0644, root,
 				   adev, &dmub_fw_state_fops);
 
+	debugfs_create_file_unsafe("amdgpu_dm_force_timing_sync", 0644, root,
+				   adev, &force_timing_sync_ops);
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
index cbcf504f73a5..357778556b06 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
@@ -719,7 +719,7 @@ void amdgpu_dm_set_irq_funcs(struct amdgpu_device *adev)
  */
 void amdgpu_dm_hpd_init(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 
@@ -755,7 +755,7 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev)
  */
 void amdgpu_dm_hpd_fini(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
new file mode 100644
index 000000000000..45825a34f8eb
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright 2020 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"),
+ * 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: AMD
+ *
+ */
+
+#ifndef __AMDGPU_DM_IRQ_PARAMS_H__
+#define __AMDGPU_DM_IRQ_PARAMS_H__
+
+struct dm_irq_params {
+	u32 last_flip_vblank;
+	struct mod_vrr_params vrr_params;
+	struct dc_stream_state *stream;
+	int active_planes;
+	struct mod_freesync_config freesync_config;
+};
+
+#endif /* __AMDGPU_DM_IRQ_PARAMS_H__ */
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 336aaa09be46..db741e47d194 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -26,6 +26,7 @@
 #include <linux/version.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_dp_mst_helper.h>
+#include <drm/drm_dp_helper.h>
 #include "dm_services.h"
 #include "amdgpu.h"
 #include "amdgpu_dm.h"
@@ -158,7 +159,20 @@ static bool validate_dsc_caps_on_connector(struct amdgpu_dm_connector *aconnecto
 	u8 dsc_caps[16] = { 0 };
 
 	aconnector->dsc_aux = drm_dp_mst_dsc_aux_for_port(port);
+#if defined(CONFIG_HP_HOOK_WORKAROUND)
+	/*
+	 * drm_dp_mst_dsc_aux_for_port() will return NULL for certain configs
+	 * because it only check the dsc/fec caps of the "port variable" and not the dock
+	 *
+	 * This case will return NULL: DSC capabe MST dock connected to a non fec/dsc capable display
+	 *
+	 * Workaround: explicitly check the use case above and use the mst dock's aux as dsc_aux
+	 *
+	 */
 
+	if (!aconnector->dsc_aux && !port->parent->port_parent)
+		aconnector->dsc_aux = &aconnector->mst_port->dm_dp_aux.aux;
+#endif
 	if (!aconnector->dsc_aux)
 		return false;
 
@@ -241,7 +255,7 @@ dm_mst_atomic_best_encoder(struct drm_connector *connector,
 			   struct drm_connector_state *connector_state)
 {
 	struct drm_device *dev = connector->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(connector_state->crtc);
 
 	return &adev->dm.mst_encoders[acrtc->crtc_id].base;
@@ -310,7 +324,7 @@ static const struct drm_encoder_funcs amdgpu_dm_encoder_funcs = {
 void
 dm_dp_create_fake_mst_encoders(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	int i;
 
 	for (i = 0; i < adev->dm.display_indexes_num; i++) {
@@ -337,7 +351,7 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
 {
 	struct amdgpu_dm_connector *master = container_of(mgr, struct amdgpu_dm_connector, mst_mgr);
 	struct drm_device *dev = master->base.dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_connector *connector;
 	int i;
@@ -426,11 +440,13 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
 	aconnector->mst_mgr.cbs = &dm_mst_cbs;
 	drm_dp_mst_topology_mgr_init(
 		&aconnector->mst_mgr,
-		dm->adev->ddev,
+		adev_to_drm(dm->adev),
 		&aconnector->dm_dp_aux.aux,
 		16,
 		4,
 		aconnector->connector_id);
+
+	drm_connector_attach_dp_subconnector_property(&aconnector->base);
 }
 
 int dm_mst_get_pbn_divider(struct dc_link *link)
@@ -450,6 +466,10 @@ struct dsc_mst_fairness_params {
 	struct dc_dsc_bw_range bw_range;
 	bool compression_possible;
 	struct drm_dp_mst_port *port;
+	enum dsc_clock_force_state clock_force_enable;
+	uint32_t num_slices_h;
+	uint32_t num_slices_v;
+	uint32_t bpp_overwrite;
 };
 
 struct dsc_mst_fairness_vars {
@@ -483,7 +503,17 @@ static void set_dsc_configs_from_fairness_vars(struct dsc_mst_fairness_params *p
 					params[i].timing,
 					&params[i].timing->dsc_cfg)) {
 			params[i].timing->flags.DSC = 1;
-			params[i].timing->dsc_cfg.bits_per_pixel = vars[i].bpp_x16;
+
+			if (params[i].bpp_overwrite)
+				params[i].timing->dsc_cfg.bits_per_pixel = params[i].bpp_overwrite;
+			else
+				params[i].timing->dsc_cfg.bits_per_pixel = vars[i].bpp_x16;
+
+			if (params[i].num_slices_h)
+				params[i].timing->dsc_cfg.num_slices_h = params[i].num_slices_h;
+
+			if (params[i].num_slices_v)
+				params[i].timing->dsc_cfg.num_slices_v = params[i].num_slices_v;
 		} else {
 			params[i].timing->flags.DSC = 0;
 		}
@@ -615,7 +645,9 @@ static void try_disable_dsc(struct drm_atomic_state *state,
 	int remaining_to_try = 0;
 
 	for (i = 0; i < count; i++) {
-		if (vars[i].dsc_enabled && vars[i].bpp_x16 == params[i].bw_range.max_target_bpp_x16) {
+		if (vars[i].dsc_enabled
+				&& vars[i].bpp_x16 == params[i].bw_range.max_target_bpp_x16
+				&& !params[i].clock_force_enable == DSC_CLK_FORCE_DEFAULT) {
 			kbps_increase[i] = params[i].bw_range.stream_kbps - params[i].bw_range.max_kbps;
 			tried[i] = false;
 			remaining_to_try += 1;
@@ -676,6 +708,7 @@ static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state,
 	struct dsc_mst_fairness_vars vars[MAX_PIPES];
 	struct amdgpu_dm_connector *aconnector;
 	int count = 0;
+	bool debugfs_overwrite = false;
 
 	memset(params, 0, sizeof(params));
 
@@ -694,6 +727,12 @@ static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state,
 		params[count].sink = stream->sink;
 		aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
 		params[count].port = aconnector->port;
+		params[count].clock_force_enable = aconnector->dsc_settings.dsc_force_enable;
+		if (params[count].clock_force_enable == DSC_CLK_FORCE_ENABLE)
+			debugfs_overwrite = true;
+		params[count].num_slices_h = aconnector->dsc_settings.dsc_num_slices_h;
+		params[count].num_slices_v = aconnector->dsc_settings.dsc_num_slices_v;
+		params[count].bpp_overwrite = aconnector->dsc_settings.dsc_bits_per_pixel;
 		params[count].compression_possible = stream->sink->dsc_caps.dsc_dec_caps.is_dsc_supported;
 		dc_dsc_get_policy_for_timing(params[count].timing, &dsc_policy);
 		if (!dc_dsc_compute_bandwidth_range(
@@ -719,14 +758,14 @@ static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state,
 						 dm_mst_get_pbn_divider(dc_link)) < 0)
 			return false;
 	}
-	if (!drm_dp_mst_atomic_check(state)) {
+	if (!drm_dp_mst_atomic_check(state) && !debugfs_overwrite) {
 		set_dsc_configs_from_fairness_vars(params, vars, count);
 		return true;
 	}
 
 	/* Try max compression */
 	for (i = 0; i < count; i++) {
-		if (params[i].compression_possible) {
+		if (params[i].compression_possible && params[i].clock_force_enable != DSC_CLK_FORCE_DISABLE) {
 			vars[i].pbn = kbps_to_peak_pbn(params[i].bw_range.min_kbps);
 			vars[i].dsc_enabled = true;
 			vars[i].bpp_x16 = params[i].bw_range.min_target_bpp_x16;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index c5f2216e59c4..6e575ffe34d0 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -592,9 +592,6 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
 	if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges)
 		pp_funcs->set_watermarks_for_clocks_ranges(pp_handle,
 							   &wm_with_clock_ranges);
-	else if (adev->smu.ppt_funcs)
-		smu_set_watermarks_for_clock_ranges(&adev->smu,
-				&wm_with_clock_ranges);
 }
 
 void pp_rv_set_pme_wa_enable(struct pp_smu *pp)
@@ -667,49 +664,8 @@ static enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp,
 {
 	const struct dc_context *ctx = pp->dm;
 	struct amdgpu_device *adev = ctx->driver_context;
-	struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges;
-	struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks =
-			wm_with_clock_ranges.wm_dmif_clocks_ranges;
-	struct dm_pp_clock_range_for_mcif_wm_set_soc15 *wm_soc_clocks =
-			wm_with_clock_ranges.wm_mcif_clocks_ranges;
-	int32_t i;
 
-	wm_with_clock_ranges.num_wm_dmif_sets = ranges->num_reader_wm_sets;
-	wm_with_clock_ranges.num_wm_mcif_sets = ranges->num_writer_wm_sets;
-
-	for (i = 0; i < wm_with_clock_ranges.num_wm_dmif_sets; i++) {
-		if (ranges->reader_wm_sets[i].wm_inst > 3)
-			wm_dce_clocks[i].wm_set_id = WM_SET_A;
-		else
-			wm_dce_clocks[i].wm_set_id =
-					ranges->reader_wm_sets[i].wm_inst;
-		wm_dce_clocks[i].wm_max_dcfclk_clk_in_khz =
-			ranges->reader_wm_sets[i].max_drain_clk_mhz * 1000;
-		wm_dce_clocks[i].wm_min_dcfclk_clk_in_khz =
-			ranges->reader_wm_sets[i].min_drain_clk_mhz * 1000;
-		wm_dce_clocks[i].wm_max_mem_clk_in_khz =
-			ranges->reader_wm_sets[i].max_fill_clk_mhz * 1000;
-		wm_dce_clocks[i].wm_min_mem_clk_in_khz =
-			ranges->reader_wm_sets[i].min_fill_clk_mhz * 1000;
-	}
-
-	for (i = 0; i < wm_with_clock_ranges.num_wm_mcif_sets; i++) {
-		if (ranges->writer_wm_sets[i].wm_inst > 3)
-			wm_soc_clocks[i].wm_set_id = WM_SET_A;
-		else
-			wm_soc_clocks[i].wm_set_id =
-					ranges->writer_wm_sets[i].wm_inst;
-		wm_soc_clocks[i].wm_max_socclk_clk_in_khz =
-			ranges->writer_wm_sets[i].max_fill_clk_mhz * 1000;
-		wm_soc_clocks[i].wm_min_socclk_clk_in_khz =
-			ranges->writer_wm_sets[i].min_fill_clk_mhz * 1000;
-		wm_soc_clocks[i].wm_max_mem_clk_in_khz =
-			ranges->writer_wm_sets[i].max_drain_clk_mhz * 1000;
-		wm_soc_clocks[i].wm_min_mem_clk_in_khz =
-			ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000;
-	}
-
-	smu_set_watermarks_for_clock_ranges(&adev->smu,	&wm_with_clock_ranges);
+	smu_set_watermarks_for_clock_ranges(&adev->smu, ranges);
 
 	return PP_SMU_RESULT_OK;
 }
@@ -810,7 +766,7 @@ pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz)
 }
 
 static enum pp_smu_status pp_nv_set_pstate_handshake_support(
-	struct pp_smu *pp, BOOLEAN pstate_handshake_supported)
+	struct pp_smu *pp, bool pstate_handshake_supported)
 {
 	const struct dc_context *ctx = pp->dm;
 	struct amdgpu_device *adev = ctx->driver_context;
@@ -920,60 +876,8 @@ static enum pp_smu_status pp_rn_set_wm_ranges(struct pp_smu *pp,
 {
 	const struct dc_context *ctx = pp->dm;
 	struct amdgpu_device *adev = ctx->driver_context;
-	struct smu_context *smu = &adev->smu;
-	struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges;
-	struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks =
-			wm_with_clock_ranges.wm_dmif_clocks_ranges;
-	struct dm_pp_clock_range_for_mcif_wm_set_soc15 *wm_soc_clocks =
-			wm_with_clock_ranges.wm_mcif_clocks_ranges;
-	int32_t i;
-
-	if (!smu->ppt_funcs)
-		return PP_SMU_RESULT_UNSUPPORTED;
-
-	wm_with_clock_ranges.num_wm_dmif_sets = ranges->num_reader_wm_sets;
-	wm_with_clock_ranges.num_wm_mcif_sets = ranges->num_writer_wm_sets;
-
-	for (i = 0; i < wm_with_clock_ranges.num_wm_dmif_sets; i++) {
-		if (ranges->reader_wm_sets[i].wm_inst > 3)
-			wm_dce_clocks[i].wm_set_id = WM_SET_A;
-		else
-			wm_dce_clocks[i].wm_set_id =
-					ranges->reader_wm_sets[i].wm_inst;
-
-		wm_dce_clocks[i].wm_min_dcfclk_clk_in_khz =
-			ranges->reader_wm_sets[i].min_drain_clk_mhz;
-
-		wm_dce_clocks[i].wm_max_dcfclk_clk_in_khz =
-			ranges->reader_wm_sets[i].max_drain_clk_mhz;
-
-		wm_dce_clocks[i].wm_min_mem_clk_in_khz =
-			ranges->reader_wm_sets[i].min_fill_clk_mhz;
-
-		wm_dce_clocks[i].wm_max_mem_clk_in_khz =
-			ranges->reader_wm_sets[i].max_fill_clk_mhz;
-	}
-
-	for (i = 0; i < wm_with_clock_ranges.num_wm_mcif_sets; i++) {
-		if (ranges->writer_wm_sets[i].wm_inst > 3)
-			wm_soc_clocks[i].wm_set_id = WM_SET_A;
-		else
-			wm_soc_clocks[i].wm_set_id =
-					ranges->writer_wm_sets[i].wm_inst;
-		wm_soc_clocks[i].wm_min_socclk_clk_in_khz =
-				ranges->writer_wm_sets[i].min_fill_clk_mhz;
-
-		wm_soc_clocks[i].wm_max_socclk_clk_in_khz =
-			ranges->writer_wm_sets[i].max_fill_clk_mhz;
-
-		wm_soc_clocks[i].wm_min_mem_clk_in_khz =
-			ranges->writer_wm_sets[i].min_drain_clk_mhz;
-
-		wm_soc_clocks[i].wm_max_mem_clk_in_khz =
-			ranges->writer_wm_sets[i].max_drain_clk_mhz;
-	}
 
-	smu_set_watermarks_for_clock_ranges(&adev->smu, &wm_with_clock_ranges);
+	smu_set_watermarks_for_clock_ranges(&adev->smu, ranges);
 
 	return PP_SMU_RESULT_OK;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile
index e0f4f1be1618..047b1e2dd8f1 100644
--- a/drivers/gpu/drm/amd/display/dc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/Makefile
@@ -43,6 +43,10 @@ DC_LIBS += dce110
 DC_LIBS += dce100
 DC_LIBS += dce80
 
+ifdef CONFIG_DRM_AMD_DC_SI
+DC_LIBS += dce60
+endif
+
 ifdef CONFIG_DRM_AMD_DC_HDCP
 DC_LIBS += hdcp
 endif
diff --git a/drivers/gpu/drm/amd/display/dc/bios/Makefile b/drivers/gpu/drm/amd/display/dc/bios/Makefile
index 239e86bbec5a..ed6b5e9763f6 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/bios/Makefile
@@ -32,6 +32,15 @@ AMD_DAL_BIOS = $(addprefix $(AMDDALPATH)/dc/bios/,$(BIOS))
 AMD_DISPLAY_FILES += $(AMD_DAL_BIOS)
 
 ###############################################################################
+# DCE 6x
+###############################################################################
+# All DCE6.x are derived from DCE6.0, so 6.0 MUST be defined if ANY of
+# DCE6.x is compiled.
+ifdef CONFIG_DRM_AMD_DC_SI
+AMD_DISPLAY_FILES += $(AMDDALPATH)/dc/bios/dce60/command_table_helper_dce60.o
+endif
+
+###############################################################################
 # DCE 8x
 ###############################################################################
 # All DCE8.x are derived from DCE8.0, so 8.0 MUST be defined if ANY of
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
index 2d5c7daaee23..29d64e7e304f 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -847,6 +847,73 @@ static enum bp_result bios_parser_get_spread_spectrum_info(
 	return result;
 }
 
+static enum bp_result get_soc_bb_info_v4_4(
+	struct bios_parser *bp,
+	struct bp_soc_bb_info *soc_bb_info)
+{
+	enum bp_result result = BP_RESULT_OK;
+	struct atom_display_controller_info_v4_4 *disp_cntl_tbl = NULL;
+
+	if (!soc_bb_info)
+		return BP_RESULT_BADINPUT;
+
+	if (!DATA_TABLES(dce_info))
+		return BP_RESULT_BADBIOSTABLE;
+
+	if (!DATA_TABLES(smu_info))
+		return BP_RESULT_BADBIOSTABLE;
+
+	disp_cntl_tbl =  GET_IMAGE(struct atom_display_controller_info_v4_4,
+							DATA_TABLES(dce_info));
+	if (!disp_cntl_tbl)
+		return BP_RESULT_BADBIOSTABLE;
+
+	soc_bb_info->dram_clock_change_latency_100ns = disp_cntl_tbl->max_mclk_chg_lat;
+	soc_bb_info->dram_sr_enter_exit_latency_100ns = disp_cntl_tbl->max_sr_enter_exit_lat;
+	soc_bb_info->dram_sr_exit_latency_100ns = disp_cntl_tbl->max_sr_exit_lat;
+
+	return result;
+}
+
+static enum bp_result bios_parser_get_soc_bb_info(
+	struct dc_bios *dcb,
+	struct bp_soc_bb_info *soc_bb_info)
+{
+	struct bios_parser *bp = BP_FROM_DCB(dcb);
+	enum bp_result result = BP_RESULT_UNSUPPORTED;
+	struct atom_common_table_header *header;
+	struct atom_data_revision tbl_revision;
+
+	if (!soc_bb_info) /* check for bad input */
+		return BP_RESULT_BADINPUT;
+
+	if (!DATA_TABLES(dce_info))
+		return BP_RESULT_UNSUPPORTED;
+
+	header = GET_IMAGE(struct atom_common_table_header,
+						DATA_TABLES(dce_info));
+	get_atom_data_table_revision(header, &tbl_revision);
+
+	switch (tbl_revision.major) {
+	case 4:
+		switch (tbl_revision.minor) {
+		case 1:
+		case 2:
+		case 3:
+			break;
+		case 4:
+			result = get_soc_bb_info_v4_4(bp, soc_bb_info);
+		default:
+			break;
+		}
+		break;
+	default:
+		break;
+	}
+
+	return result;
+}
+
 static enum bp_result get_embedded_panel_info_v2_1(
 		struct bios_parser *bp,
 		struct embedded_panel_info *info)
@@ -2222,7 +2289,9 @@ static const struct dc_vbios_funcs vbios_funcs = {
 
 	.get_atom_dc_golden_table = bios_get_atom_dc_golden_table,
 
-	.enable_lvtma_control = bios_parser_enable_lvtma_control
+	.enable_lvtma_control = bios_parser_enable_lvtma_control,
+
+	.get_soc_bb_info = bios_parser_get_soc_bb_info,
 };
 
 static bool bios_parser2_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.c b/drivers/gpu/drm/amd/display/dc/bios/command_table.c
index 5815983caaf8..070459e3e407 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c
@@ -1877,9 +1877,7 @@ static enum bp_result set_crtc_using_dtd_timing_v3(
 			 * but it is 4 either from Edid data (spec CEA 861)
 			 * or CEA timing table.
 			 */
-			params.usV_SyncOffset =
-					cpu_to_le16(le16_to_cpu(params.usV_SyncOffset) + 1);
-
+			le16_add_cpu(&params.usV_SyncOffset, 1);
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
index eb3ae5c3677c..25bdf1c38e0a 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
@@ -569,10 +569,7 @@ static enum bp_result set_crtc_using_dtd_timing_v3(
 			 * but it is 4 either from Edid data (spec CEA 861)
 			 * or CEA timing table.
 			 */
-			params.v_syncoffset =
-				cpu_to_le16(le16_to_cpu(params.v_syncoffset) +
-						1);
-
+			le16_add_cpu(&params.v_syncoffset, 1);
 		}
 	}
 
@@ -923,11 +920,39 @@ static void init_enable_lvtma_control(struct bios_parser *bp)
 
 }
 
+static void enable_lvtma_control_dmcub(
+	struct dc_dmub_srv *dmcub,
+	uint8_t uc_pwr_on)
+{
+
+	union dmub_rb_cmd cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+
+	cmd.cmd_common.header.type = DMUB_CMD__VBIOS;
+	cmd.cmd_common.header.sub_type =
+			DMUB_CMD__VBIOS_LVTMA_CONTROL;
+	cmd.cmd_common.cmd_buffer[0] =
+			uc_pwr_on;
+
+	dc_dmub_srv_cmd_queue(dmcub, &cmd);
+	dc_dmub_srv_cmd_execute(dmcub);
+	dc_dmub_srv_wait_idle(dmcub);
+
+}
+
 static enum bp_result enable_lvtma_control(
 	struct bios_parser *bp,
 	uint8_t uc_pwr_on)
 {
 	enum bp_result result = BP_RESULT_FAILURE;
+
+	if (bp->base.ctx->dc->ctx->dmub_srv &&
+	    bp->base.ctx->dc->debug.dmub_command_table) {
+		enable_lvtma_control_dmcub(bp->base.ctx->dmub_srv,
+				uc_pwr_on);
+		return BP_RESULT_OK;
+	}
 	return result;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c
index 253bbb1eea60..48b4ef03fc8f 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.c
@@ -36,6 +36,14 @@ bool dal_bios_parser_init_cmd_tbl_helper(
 	enum dce_version dce)
 {
 	switch (dce) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case DCE_VERSION_6_0:
+	case DCE_VERSION_6_1:
+	case DCE_VERSION_6_4:
+		*h = dal_cmd_tbl_helper_dce60_get_table();
+		return true;
+#endif
+
 	case DCE_VERSION_8_0:
 	case DCE_VERSION_8_1:
 	case DCE_VERSION_8_3:
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.h b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.h
index 4c3789df253d..dfd30aaf4032 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper.h
@@ -26,6 +26,9 @@
 #ifndef __DAL_COMMAND_TABLE_HELPER_H__
 #define __DAL_COMMAND_TABLE_HELPER_H__
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/command_table_helper_dce60.h"
+#endif
 #include "dce80/command_table_helper_dce80.h"
 #include "dce110/command_table_helper_dce110.h"
 #include "dce112/command_table_helper_dce112.h"
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c
index 21ff6b686f5f..74c498b6774d 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c
@@ -37,6 +37,14 @@ bool dal_bios_parser_init_cmd_tbl_helper2(
 	enum dce_version dce)
 {
 	switch (dce) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case DCE_VERSION_6_0:
+	case DCE_VERSION_6_1:
+	case DCE_VERSION_6_4:
+		*h = dal_cmd_tbl_helper_dce60_get_table();
+		return true;
+#endif
+
 	case DCE_VERSION_8_0:
 	case DCE_VERSION_8_1:
 	case DCE_VERSION_8_3:
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.h b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.h
index 785fcb20a1b9..66e0a3e73768 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.h
@@ -26,6 +26,9 @@
 #ifndef __DAL_COMMAND_TABLE_HELPER2_H__
 #define __DAL_COMMAND_TABLE_HELPER2_H__
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/command_table_helper_dce60.h"
+#endif
 #include "dce80/command_table_helper_dce80.h"
 #include "dce110/command_table_helper_dce110.h"
 #include "dce112/command_table_helper2_dce112.h"
diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.c b/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.c
new file mode 100644
index 000000000000..710221b4f5c5
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.c
@@ -0,0 +1,354 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include "dm_services.h"
+
+#include "atom.h"
+
+#include "include/grph_object_id.h"
+#include "include/grph_object_defs.h"
+#include "include/bios_parser_types.h"
+
+#include "../command_table_helper.h"
+
+static uint8_t encoder_action_to_atom(enum bp_encoder_control_action action)
+{
+	uint8_t atom_action = 0;
+
+	switch (action) {
+	case ENCODER_CONTROL_ENABLE:
+		atom_action = ATOM_ENABLE;
+		break;
+	case ENCODER_CONTROL_DISABLE:
+		atom_action = ATOM_DISABLE;
+		break;
+	case ENCODER_CONTROL_SETUP:
+		atom_action = ATOM_ENCODER_CMD_SETUP;
+		break;
+	case ENCODER_CONTROL_INIT:
+		atom_action = ATOM_ENCODER_INIT;
+		break;
+	default:
+		BREAK_TO_DEBUGGER(); /* Unhandle action in driver.!! */
+		break;
+	}
+
+	return atom_action;
+}
+
+static bool engine_bp_to_atom(enum engine_id id, uint32_t *atom_engine_id)
+{
+	bool result = false;
+
+	if (atom_engine_id != NULL)
+		switch (id) {
+		case ENGINE_ID_DIGA:
+			*atom_engine_id = ASIC_INT_DIG1_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGB:
+			*atom_engine_id = ASIC_INT_DIG2_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGC:
+			*atom_engine_id = ASIC_INT_DIG3_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGD:
+			*atom_engine_id = ASIC_INT_DIG4_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGE:
+			*atom_engine_id = ASIC_INT_DIG5_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGF:
+			*atom_engine_id = ASIC_INT_DIG6_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DIGG:
+			*atom_engine_id = ASIC_INT_DIG7_ENCODER_ID;
+			result = true;
+			break;
+		case ENGINE_ID_DACA:
+			*atom_engine_id = ASIC_INT_DAC1_ENCODER_ID;
+			result = true;
+			break;
+		default:
+			break;
+		}
+
+	return result;
+}
+
+static bool clock_source_id_to_atom(
+	enum clock_source_id id,
+	uint32_t *atom_pll_id)
+{
+	bool result = true;
+
+	if (atom_pll_id != NULL)
+		switch (id) {
+		case CLOCK_SOURCE_ID_PLL0:
+			*atom_pll_id = ATOM_PPLL0;
+			break;
+		case CLOCK_SOURCE_ID_PLL1:
+			*atom_pll_id = ATOM_PPLL1;
+			break;
+		case CLOCK_SOURCE_ID_PLL2:
+			*atom_pll_id = ATOM_PPLL2;
+			break;
+		case CLOCK_SOURCE_ID_EXTERNAL:
+			*atom_pll_id = ATOM_PPLL_INVALID;
+			break;
+		case CLOCK_SOURCE_ID_DFS:
+			*atom_pll_id = ATOM_EXT_PLL1;
+			break;
+		case CLOCK_SOURCE_ID_VCE:
+			/* for VCE encoding,
+			 * we need to pass in ATOM_PPLL_INVALID
+			 */
+			*atom_pll_id = ATOM_PPLL_INVALID;
+			break;
+		case CLOCK_SOURCE_ID_DP_DTO:
+			/* When programming DP DTO PLL ID should be invalid */
+			*atom_pll_id = ATOM_PPLL_INVALID;
+			break;
+		case CLOCK_SOURCE_ID_UNDEFINED:
+			BREAK_TO_DEBUGGER(); /* check when this will happen! */
+			*atom_pll_id = ATOM_PPLL_INVALID;
+			result = false;
+			break;
+		default:
+			result = false;
+			break;
+		}
+
+	return result;
+}
+
+static uint8_t clock_source_id_to_atom_phy_clk_src_id(
+		enum clock_source_id id)
+{
+	uint8_t atom_phy_clk_src_id = 0;
+
+	switch (id) {
+	case CLOCK_SOURCE_ID_PLL0:
+		atom_phy_clk_src_id = ATOM_TRANSMITTER_CONFIG_V5_P0PLL;
+		break;
+	case CLOCK_SOURCE_ID_PLL1:
+		atom_phy_clk_src_id = ATOM_TRANSMITTER_CONFIG_V5_P1PLL;
+		break;
+	case CLOCK_SOURCE_ID_PLL2:
+		atom_phy_clk_src_id = ATOM_TRANSMITTER_CONFIG_V5_P2PLL;
+		break;
+	case CLOCK_SOURCE_ID_EXTERNAL:
+		atom_phy_clk_src_id = ATOM_TRANSMITTER_CONFIG_V5_REFCLK_SRC_EXT;
+		break;
+	default:
+		atom_phy_clk_src_id = ATOM_TRANSMITTER_CONFIG_V5_P1PLL;
+		break;
+	}
+
+	return atom_phy_clk_src_id >> 2;
+}
+
+static uint8_t signal_type_to_atom_dig_mode(enum signal_type s)
+{
+	uint8_t atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_DP;
+
+	switch (s) {
+	case SIGNAL_TYPE_DISPLAY_PORT:
+	case SIGNAL_TYPE_EDP:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_DP;
+		break;
+	case SIGNAL_TYPE_LVDS:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_LVDS;
+		break;
+	case SIGNAL_TYPE_DVI_SINGLE_LINK:
+	case SIGNAL_TYPE_DVI_DUAL_LINK:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_DVI;
+		break;
+	case SIGNAL_TYPE_HDMI_TYPE_A:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_HDMI;
+		break;
+	case SIGNAL_TYPE_DISPLAY_PORT_MST:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_DP_MST;
+		break;
+	default:
+		atom_dig_mode = ATOM_TRANSMITTER_DIGMODE_V5_DVI;
+		break;
+	}
+
+	return atom_dig_mode;
+}
+
+static uint8_t hpd_sel_to_atom(enum hpd_source_id id)
+{
+	uint8_t atom_hpd_sel = 0;
+
+	switch (id) {
+	case HPD_SOURCEID1:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD1_SEL;
+		break;
+	case HPD_SOURCEID2:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD2_SEL;
+		break;
+	case HPD_SOURCEID3:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD3_SEL;
+		break;
+	case HPD_SOURCEID4:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD4_SEL;
+		break;
+	case HPD_SOURCEID5:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD5_SEL;
+		break;
+	case HPD_SOURCEID6:
+		atom_hpd_sel = ATOM_TRANSMITTER_CONFIG_V5_HPD6_SEL;
+		break;
+	case HPD_SOURCEID_UNKNOWN:
+	default:
+		atom_hpd_sel = 0;
+		break;
+	}
+	return atom_hpd_sel >> 4;
+}
+
+static uint8_t dig_encoder_sel_to_atom(enum engine_id id)
+{
+	uint8_t atom_dig_encoder_sel = 0;
+
+	switch (id) {
+	case ENGINE_ID_DIGA:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL;
+		break;
+	case ENGINE_ID_DIGB:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGB_SEL;
+		break;
+	case ENGINE_ID_DIGC:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGC_SEL;
+		break;
+	case ENGINE_ID_DIGD:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGD_SEL;
+		break;
+	case ENGINE_ID_DIGE:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGE_SEL;
+		break;
+	case ENGINE_ID_DIGF:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGF_SEL;
+		break;
+	case ENGINE_ID_DIGG:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGG_SEL;
+		break;
+	default:
+		atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL;
+		break;
+	}
+
+	return atom_dig_encoder_sel;
+}
+
+static uint8_t phy_id_to_atom(enum transmitter t)
+{
+	uint8_t atom_phy_id;
+
+	switch (t) {
+	case TRANSMITTER_UNIPHY_A:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYA;
+		break;
+	case TRANSMITTER_UNIPHY_B:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYB;
+		break;
+	case TRANSMITTER_UNIPHY_C:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYC;
+		break;
+	case TRANSMITTER_UNIPHY_D:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYD;
+		break;
+	case TRANSMITTER_UNIPHY_E:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYE;
+		break;
+	case TRANSMITTER_UNIPHY_F:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYF;
+		break;
+	case TRANSMITTER_UNIPHY_G:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYG;
+		break;
+	default:
+		atom_phy_id = ATOM_PHY_ID_UNIPHYA;
+		break;
+	}
+	return atom_phy_id;
+}
+
+static uint8_t disp_power_gating_action_to_atom(
+	enum bp_pipe_control_action action)
+{
+	uint8_t atom_pipe_action = 0;
+
+	switch (action) {
+	case ASIC_PIPE_DISABLE:
+		atom_pipe_action = ATOM_DISABLE;
+		break;
+	case ASIC_PIPE_ENABLE:
+		atom_pipe_action = ATOM_ENABLE;
+		break;
+	case ASIC_PIPE_INIT:
+		atom_pipe_action = ATOM_INIT;
+		break;
+	default:
+		BREAK_TO_DEBUGGER(); /* Unhandle action in driver! */
+		break;
+	}
+
+	return atom_pipe_action;
+}
+
+static const struct command_table_helper command_table_helper_funcs = {
+	.controller_id_to_atom = dal_cmd_table_helper_controller_id_to_atom,
+	.encoder_action_to_atom = encoder_action_to_atom,
+	.engine_bp_to_atom = engine_bp_to_atom,
+	.clock_source_id_to_atom = clock_source_id_to_atom,
+	.clock_source_id_to_atom_phy_clk_src_id =
+		clock_source_id_to_atom_phy_clk_src_id,
+	.signal_type_to_atom_dig_mode = signal_type_to_atom_dig_mode,
+	.hpd_sel_to_atom = hpd_sel_to_atom,
+	.dig_encoder_sel_to_atom = dig_encoder_sel_to_atom,
+	.phy_id_to_atom = phy_id_to_atom,
+	.disp_power_gating_action_to_atom = disp_power_gating_action_to_atom,
+	.assign_control_parameter =
+		dal_cmd_table_helper_assign_control_parameter,
+	.clock_source_id_to_ref_clk_src =
+		dal_cmd_table_helper_clock_source_id_to_ref_clk_src,
+	.transmitter_bp_to_atom = dal_cmd_table_helper_transmitter_bp_to_atom,
+	.encoder_id_to_atom = dal_cmd_table_helper_encoder_id_to_atom,
+	.encoder_mode_bp_to_atom =
+		dal_cmd_table_helper_encoder_mode_bp_to_atom,
+};
+
+const struct command_table_helper *dal_cmd_tbl_helper_dce60_get_table(void)
+{
+	return &command_table_helper_funcs;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.h b/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.h
new file mode 100644
index 000000000000..f733be553d5a
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/bios/dce60/command_table_helper_dce60.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DAL_COMMAND_TABLE_HELPER_DCE60_H__
+#define __DAL_COMMAND_TABLE_HELPER_DCE60_H__
+
+struct command_table_helper;
+
+const struct command_table_helper *dal_cmd_tbl_helper_dce60_get_table(void);
+
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/Makefile b/drivers/gpu/drm/amd/display/dc/calcs/Makefile
index 4674aca8f206..64f515d74410 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile
@@ -33,6 +33,10 @@ ifdef CONFIG_PPC64
 calcs_ccflags := -mhard-float -maltivec
 endif
 
+ifdef CONFIG_ARM64
+calcs_rcflags := -mgeneral-regs-only
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 IS_OLD_GCC = 1
@@ -53,6 +57,9 @@ endif
 CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_rcflags)
 
 BW_CALCS = dce_calcs.o bw_fixed.o custom_float.o
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile
index 6874276bb2a1..1a495759a034 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile
@@ -30,6 +30,17 @@ AMD_DAL_CLK_MGR = $(addprefix $(AMDDALPATH)/dc/clk_mgr/,$(CLK_MGR))
 AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR)
 
 
+ifdef CONFIG_DRM_AMD_DC_SI
+###############################################################################
+# DCE 60
+###############################################################################
+CLK_MGR_DCE60 = dce60_clk_mgr.o
+
+AMD_DAL_CLK_MGR_DCE60 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dce60/,$(CLK_MGR_DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCE60)
+endif
+
 ###############################################################################
 # DCE 100 and DCE8x
 ###############################################################################
@@ -93,6 +104,13 @@ ifdef CONFIG_PPC64
 CFLAGS_$(AMDDALPATH)/dc/clk_mgr/dcn21/rn_clk_mgr.o := $(call cc-option,-mno-gnu-attribute)
 endif
 
+# prevent build errors:
+# ...: '-mgeneral-regs-only' is incompatible with the use of floating-point types
+# this file is unused on arm64, just like on ppc64
+ifdef CONFIG_ARM64
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/clk_mgr/dcn21/rn_clk_mgr.o := -mgeneral-regs-only
+endif
+
 AMD_DAL_CLK_MGR_DCN21 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn21/,$(CLK_MGR_DCN21))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN21)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
index 6a345d43028c..efb909ef7a0f 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
@@ -34,6 +34,7 @@
 #include "dce110/dce110_clk_mgr.h"
 #include "dce112/dce112_clk_mgr.h"
 #include "dce120/dce120_clk_mgr.h"
+#include "dce60/dce60_clk_mgr.h"
 #include "dcn10/rv1_clk_mgr.h"
 #include "dcn10/rv2_clk_mgr.h"
 #include "dcn20/dcn20_clk_mgr.h"
@@ -123,6 +124,11 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p
 	}
 
 	switch (asic_id.chip_family) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case FAMILY_SI:
+		dce60_clk_mgr_construct(ctx, clk_mgr);
+		break;
+#endif
 	case FAMILY_CI:
 	case FAMILY_KV:
 		dce_clk_mgr_construct(ctx, clk_mgr);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
index d031bd3d3072..807dca8f7d7a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
@@ -79,8 +79,7 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz)
 	memset(&dce_clk_params, 0, sizeof(dce_clk_params));
 
 	/* Make sure requested clock isn't lower than minimum threshold*/
-	if (requested_clk_khz > 0)
-		requested_clk_khz = max(requested_clk_khz,
+	requested_clk_khz = max(requested_clk_khz,
 				clk_mgr_dce->base.dentist_vco_freq_khz / 62);
 
 	dce_clk_params.target_clock_frequency = requested_clk_khz;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
new file mode 100644
index 000000000000..0267644717b2
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
@@ -0,0 +1,174 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+
+#include "dccg.h"
+#include "clk_mgr_internal.h"
+#include "dce100/dce_clk_mgr.h"
+#include "dce110/dce110_clk_mgr.h"
+#include "dce60_clk_mgr.h"
+#include "reg_helper.h"
+#include "dmcu.h"
+#include "core_types.h"
+#include "dal_asic_id.h"
+
+/*
+ * Currently the register shifts and masks in this file are used for dce60
+ * which has no DPREFCLK_CNTL register
+ * TODO: remove this when DENTIST_DISPCLK_CNTL
+ * is moved to dccg, where it belongs
+ */
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#define REG(reg) \
+	(clk_mgr->regs->reg)
+
+#undef FN
+#define FN(reg_name, field_name) \
+	clk_mgr->clk_mgr_shift->field_name, clk_mgr->clk_mgr_mask->field_name
+
+/* set register offset */
+#define SR(reg_name)\
+	.reg_name = mm ## reg_name
+
+static const struct clk_mgr_registers disp_clk_regs = {
+		CLK_COMMON_REG_LIST_DCE60_BASE()
+};
+
+static const struct clk_mgr_shift disp_clk_shift = {
+		CLK_COMMON_MASK_SH_LIST_DCE60_COMMON_BASE(__SHIFT)
+};
+
+static const struct clk_mgr_mask disp_clk_mask = {
+		CLK_COMMON_MASK_SH_LIST_DCE60_COMMON_BASE(_MASK)
+};
+
+
+/* Max clock values for each state indexed by "enum clocks_state": */
+static const struct state_dependent_clocks dce60_max_clks_by_state[] = {
+/* ClocksStateInvalid - should not be used */
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/* ClocksStateUltraLow - not expected to be used for DCE 6.0 */
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/* ClocksStateLow */
+{ .display_clk_khz = 352000, .pixel_clk_khz = 330000},
+/* ClocksStateNominal */
+{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 },
+/* ClocksStatePerformance */
+{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 } };
+
+static int dce60_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+	int dprefclk_wdivider;
+	int dp_ref_clk_khz;
+	int target_div;
+
+	/* DCE6 has no DPREFCLK_CNTL to read DP Reference Clock source */
+
+	/* Read the mmDENTIST_DISPCLK_CNTL to get the currently
+	 * programmed DID DENTIST_DPREFCLK_WDIVIDER*/
+	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, &dprefclk_wdivider);
+
+	/* Convert DENTIST_DPREFCLK_WDIVIDERto actual divider*/
+	target_div = dentist_get_divider_from_did(dprefclk_wdivider);
+
+	/* Calculate the current DFS clock, in kHz.*/
+	dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
+		* clk_mgr->base.dentist_vco_freq_khz) / target_div;
+
+	return dce_adjust_dp_ref_freq_for_ss(clk_mgr, dp_ref_clk_khz);
+}
+
+static void dce60_pplib_apply_display_requirements(
+	struct dc *dc,
+	struct dc_state *context)
+{
+	struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
+
+	pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context);
+
+	dce110_fill_display_configs(context, pp_display_cfg);
+
+	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)
+		dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
+}
+
+static void dce60_update_clocks(struct clk_mgr *clk_mgr_base,
+			struct dc_state *context,
+			bool safe_to_lower)
+{
+	struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+	struct dm_pp_power_level_change_request level_change_req;
+	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
+
+	/*TODO: W/A for dal3 linux, investigate why this works */
+	if (!clk_mgr_dce->dfs_bypass_active)
+		patched_disp_clk = patched_disp_clk * 115 / 100;
+
+	level_change_req.power_level = dce_get_required_clocks_state(clk_mgr_base, context);
+	/* get max clock state from PPLIB */
+	if ((level_change_req.power_level < clk_mgr_dce->cur_min_clks_state && safe_to_lower)
+			|| level_change_req.power_level > clk_mgr_dce->cur_min_clks_state) {
+		if (dm_pp_apply_power_level_change_request(clk_mgr_base->ctx, &level_change_req))
+			clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
+	}
+
+	if (should_set_clock(safe_to_lower, patched_disp_clk, clk_mgr_base->clks.dispclk_khz)) {
+		patched_disp_clk = dce_set_clock(clk_mgr_base, patched_disp_clk);
+		clk_mgr_base->clks.dispclk_khz = patched_disp_clk;
+	}
+	dce60_pplib_apply_display_requirements(clk_mgr_base->ctx->dc, context);
+}
+
+
+
+
+
+
+
+
+static struct clk_mgr_funcs dce60_funcs = {
+	.get_dp_ref_clk_frequency = dce60_get_dp_ref_freq_khz,
+	.update_clocks = dce60_update_clocks
+};
+
+void dce60_clk_mgr_construct(
+		struct dc_context *ctx,
+		struct clk_mgr_internal *clk_mgr)
+{
+	dce_clk_mgr_construct(ctx, clk_mgr);
+
+	memcpy(clk_mgr->max_clks_by_state,
+		dce60_max_clks_by_state,
+		sizeof(dce60_max_clks_by_state));
+
+	clk_mgr->regs = &disp_clk_regs;
+	clk_mgr->clk_mgr_shift = &disp_clk_shift;
+	clk_mgr->clk_mgr_mask = &disp_clk_mask;
+	clk_mgr->base.funcs = &dce60_funcs;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.h
new file mode 100644
index 000000000000..eca3e5168089
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.h
@@ -0,0 +1,36 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+
+#ifndef DAL_DC_DCE_DCE60_CLK_MGR_H_
+#define DAL_DC_DCE_DCE60_CLK_MGR_H_
+
+#include "dc.h"
+
+void dce60_clk_mgr_construct(
+		struct dc_context *ctx,
+		struct clk_mgr_internal *clk_mgr_dce);
+
+#endif /* DAL_DC_DCE_DCE60_CLK_MGR_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
index 21a3073c8929..2f8fee05547a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
@@ -761,6 +761,7 @@ void rn_clk_mgr_construct(
 {
 	struct dc_debug_options *debug = &ctx->dc->debug;
 	struct dpm_clocks clock_table = { 0 };
+	enum pp_smu_status status = 0;
 
 	clk_mgr->base.ctx = ctx;
 	clk_mgr->base.funcs = &dcn21_funcs;
@@ -817,8 +818,10 @@ void rn_clk_mgr_construct(
 	clk_mgr->base.bw_params = &rn_bw_params;
 
 	if (pp_smu && pp_smu->rn_funcs.get_dpm_clock_table) {
-		pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table);
-		if (ctx->dc_bios && ctx->dc_bios->integrated_info) {
+		status = pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table);
+
+		if (status == PP_SMU_RESULT_OK &&
+		    ctx->dc_bios && ctx->dc_bios->integrated_info) {
 			rn_clk_mgr_helper_populate_bw_params (clk_mgr->base.bw_params, &clock_table, ctx->dc_bios->integrated_info);
 		}
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
index 9133646f6d5f..b0e9b0509568 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
@@ -554,8 +554,7 @@ void dcn3_clk_mgr_construct(
 
 void dcn3_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr)
 {
-	if (clk_mgr->base.bw_params)
-		kfree(clk_mgr->base.bw_params);
+	kfree(clk_mgr->base.bw_params);
 
 	if (clk_mgr->wm_range_table)
 		dm_helpers_free_gpu_mem(clk_mgr->base.ctx, DC_MEM_ALLOC_TYPE_GART,
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 92eb1ca1634f..2a725a5fba40 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -735,6 +735,8 @@ static bool dc_construct(struct dc *dc,
 	dc->clk_mgr->force_smu_not_present = init_params->force_smu_not_present;
 #endif
 
+	dc->debug.force_ignore_link_settings = init_params->force_ignore_link_settings;
+
 	if (dc->res_pool->funcs->update_bw_bounding_box)
 		dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params);
 
@@ -842,6 +844,60 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 	dc_release_state(current_ctx);
 }
 
+static void disable_vbios_mode_if_required(
+		struct dc *dc,
+		struct dc_state *context)
+{
+	unsigned int i;
+
+	/* check if timing_changed, disable stream*/
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_stream_state *stream = NULL;
+		struct dc_link *link = NULL;
+		struct pipe_ctx *pipe = NULL;
+
+		pipe = &context->res_ctx.pipe_ctx[i];
+		stream = pipe->stream;
+		if (stream == NULL)
+			continue;
+
+		if (stream->link->local_sink &&
+			stream->link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
+			link = stream->link;
+		}
+
+		if (link != NULL) {
+			unsigned int enc_inst, tg_inst = 0;
+			unsigned int pix_clk_100hz;
+
+			enc_inst = link->link_enc->funcs->get_dig_frontend(link->link_enc);
+			if (enc_inst != ENGINE_ID_UNKNOWN) {
+				for (i = 0; i < dc->res_pool->stream_enc_count; i++) {
+					if (dc->res_pool->stream_enc[i]->id == enc_inst) {
+						tg_inst = dc->res_pool->stream_enc[i]->funcs->dig_source_otg(
+							dc->res_pool->stream_enc[i]);
+						break;
+					}
+				}
+
+				dc->res_pool->dp_clock_source->funcs->get_pixel_clk_frequency_100hz(
+					dc->res_pool->dp_clock_source,
+					tg_inst, &pix_clk_100hz);
+
+				if (link->link_status.link_active) {
+					uint32_t requested_pix_clk_100hz =
+						pipe->stream_res.pix_clk_params.requested_pix_clk_100hz;
+
+					if (pix_clk_100hz != requested_pix_clk_100hz) {
+						core_link_disable_stream(pipe);
+						pipe->stream->dpms_off = false;
+					}
+				}
+			}
+		}
+	}
+}
+
 static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context)
 {
 	int i;
@@ -1238,6 +1294,27 @@ bool dc_enable_stereo(
 	return ret;
 }
 
+void dc_trigger_sync(struct dc *dc, struct dc_state *context)
+{
+	if (context->stream_count > 1 && !dc->debug.disable_timing_sync) {
+		enable_timing_multisync(dc, context);
+		program_timing_sync(dc, context);
+	}
+}
+
+static uint8_t get_stream_mask(struct dc *dc, struct dc_state *context)
+{
+	int i;
+	unsigned int stream_mask = 0;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (context->res_ctx.pipe_ctx[i].stream)
+			stream_mask |= 1 << i;
+	}
+
+	return stream_mask;
+}
+
 /*
  * Applies given context to HW and copy it into current context.
  * It's up to the user to release the src context afterwards.
@@ -1257,15 +1334,17 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	for (i = 0; i < context->stream_count; i++)
 		dc_streams[i] =  context->streams[i];
 
-	if (!dcb->funcs->is_accelerated_mode(dcb))
+	if (!dcb->funcs->is_accelerated_mode(dcb)) {
+		disable_vbios_mode_if_required(dc, context);
 		dc->hwss.enable_accelerated_mode(dc, context);
+	}
 
-	for (i = 0; i < context->stream_count; i++) {
+	for (i = 0; i < context->stream_count; i++)
 		if (context->streams[i]->apply_seamless_boot_optimization)
 			dc->optimize_seamless_boot_streams++;
-	}
 
-	if (dc->optimize_seamless_boot_streams == 0)
+	if (context->stream_count > dc->optimize_seamless_boot_streams ||
+		context->stream_count == 0)
 		dc->hwss.prepare_bandwidth(dc, context);
 
 	disable_dangling_plane(dc, context);
@@ -1297,10 +1376,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	if (result != DC_OK)
 		return result;
 
-	if (context->stream_count > 1 && !dc->debug.disable_timing_sync) {
-		enable_timing_multisync(dc, context);
-		program_timing_sync(dc, context);
-	}
+	dc_trigger_sync(dc, context);
 
 	/* Program all planes within new context*/
 	if (dc->hwss.program_front_end_for_ctx) {
@@ -1350,13 +1426,19 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 
 	dc_enable_stereo(dc, context, dc_streams, context->stream_count);
 
-	if (dc->optimize_seamless_boot_streams == 0) {
+	if (context->stream_count > dc->optimize_seamless_boot_streams ||
+		context->stream_count == 0) {
 		/* Must wait for no flips to be pending before doing optimize bw */
 		wait_for_no_pipes_pending(dc, context);
 		/* pplib is notified if disp_num changed */
 		dc->hwss.optimize_bandwidth(dc, context);
 	}
 
+	context->stream_mask = get_stream_mask(dc, context);
+
+	if (context->stream_mask != dc->current_state->stream_mask)
+		dc_dmub_srv_notify_stream_mask(dc->ctx->dmub_srv, context->stream_mask);
+
 	for (i = 0; i < context->stream_count; i++)
 		context->streams[i]->mode_changed = false;
 
@@ -1476,13 +1558,8 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc)
 	return true;
 }
 
-struct dc_state *dc_create_state(struct dc *dc)
+static void init_state(struct dc *dc, struct dc_state *context)
 {
-	struct dc_state *context = kvzalloc(sizeof(struct dc_state),
-					    GFP_KERNEL);
-
-	if (!context)
-		return NULL;
 	/* Each context must have their own instance of VBA and in order to
 	 * initialize and obtain IP and SOC the base DML instance from DC is
 	 * initially copied into every context
@@ -1490,6 +1567,17 @@ struct dc_state *dc_create_state(struct dc *dc)
 #ifdef CONFIG_DRM_AMD_DC_DCN
 	memcpy(&context->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib));
 #endif
+}
+
+struct dc_state *dc_create_state(struct dc *dc)
+{
+	struct dc_state *context = kzalloc(sizeof(struct dc_state),
+					   GFP_KERNEL);
+
+	if (!context)
+		return NULL;
+
+	init_state(dc, context);
 
 	kref_init(&context->refcount);
 
@@ -2295,6 +2383,7 @@ static void commit_planes_for_stream(struct dc *dc,
 		enum surface_update_type update_type,
 		struct dc_state *context)
 {
+	bool mpcc_disconnected = false;
 	int i, j;
 	struct pipe_ctx *top_pipe_to_program = NULL;
 
@@ -2325,6 +2414,15 @@ static void commit_planes_for_stream(struct dc *dc,
 		context_clock_trace(dc, context);
 	}
 
+	if (update_type != UPDATE_TYPE_FAST && dc->hwss.interdependent_update_lock &&
+		dc->hwss.disconnect_pipes && dc->hwss.wait_for_pending_cleared){
+		dc->hwss.interdependent_update_lock(dc, context, true);
+		mpcc_disconnected = dc->hwss.disconnect_pipes(dc, context);
+		dc->hwss.interdependent_update_lock(dc, context, false);
+		if (mpcc_disconnected)
+			dc->hwss.wait_for_pending_cleared(dc, context);
+	}
+
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
 		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
 
@@ -2400,8 +2498,7 @@ static void commit_planes_for_stream(struct dc *dc,
 				plane_state->triplebuffer_flips = false;
 				if (update_type == UPDATE_TYPE_FAST &&
 					dc->hwss.program_triplebuffer != NULL &&
-					!plane_state->flip_immediate &&
-					!dc->debug.disable_tri_buf) {
+					!plane_state->flip_immediate && dc->debug.enable_tri_buf) {
 						/*triple buffer for VUpdate  only*/
 						plane_state->triplebuffer_flips = true;
 				}
@@ -2428,8 +2525,7 @@ static void commit_planes_for_stream(struct dc *dc,
 
 			ASSERT(!pipe_ctx->plane_state->triplebuffer_flips);
 
-			if (dc->hwss.program_triplebuffer != NULL &&
-				!dc->debug.disable_tri_buf) {
+			if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) {
 				/*turn off triple buffer for full update*/
 				dc->hwss.program_triplebuffer(
 					dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips);
@@ -2494,8 +2590,7 @@ static void commit_planes_for_stream(struct dc *dc,
 				if (pipe_ctx->plane_state != plane_state)
 					continue;
 				/*program triple buffer after lock based on flip type*/
-				if (dc->hwss.program_triplebuffer != NULL &&
-					!dc->debug.disable_tri_buf) {
+				if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) {
 					/*only enable triplebuffer for  fast_update*/
 					dc->hwss.program_triplebuffer(
 						dc, pipe_ctx, plane_state->triplebuffer_flips);
@@ -2621,7 +2716,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 
 	copy_stream_update_to_stream(dc, context, stream, stream_update);
 
-	if (update_type > UPDATE_TYPE_FAST) {
+	if (update_type >= UPDATE_TYPE_FULL) {
 		if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) {
 			DC_ERROR("Mode validation failed for stream update!\n");
 			dc_release_state(context);
@@ -2933,6 +3028,30 @@ void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_
 		dc->hwss.get_clock(dc, clock_type, clock_cfg);
 }
 
+/* enable/disable eDP PSR without specify stream for eDP */
+bool dc_set_psr_allow_active(struct dc *dc, bool enable)
+{
+	int i;
+
+	for (i = 0; i < dc->current_state->stream_count ; i++) {
+		struct dc_link *link;
+		struct dc_stream_state *stream = dc->current_state->streams[i];
+
+		link = stream->link;
+		if (!link)
+			continue;
+
+		if (link->psr_settings.psr_feature_enabled) {
+			if (enable && !link->psr_settings.psr_allow_active)
+				return dc_link_set_psr_allow_active(link, true, false);
+			else if (!enable && link->psr_settings.psr_allow_active)
+				return dc_link_set_psr_allow_active(link, false, true);
+		}
+	}
+
+	return true;
+}
+
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 
 void dc_allow_idle_optimizations(struct dc *dc, bool allow)
@@ -2979,4 +3098,10 @@ void dc_lock_memory_clock_frequency(struct dc *dc)
 		if (dc->current_state->res_ctx.pipe_ctx[i].plane_state)
 			core_link_enable_stream(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]);
 }
+
+bool dc_is_plane_eligible_for_idle_optimizaitons(struct dc *dc,
+						 struct dc_plane_state *plane)
+{
+	return false;
+}
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index c026b393f3c5..2a9080400bdd 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -177,7 +177,7 @@ static bool is_ycbcr709_limited_type(
 		ret = true;
 	return ret;
 }
-enum dc_color_space_type get_color_space_type(enum dc_color_space color_space)
+static enum dc_color_space_type get_color_space_type(enum dc_color_space color_space)
 {
 	enum dc_color_space_type type = COLOR_SPACE_RGB_TYPE;
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
index 437d1a7a16fe..fec87a2e210c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -2441,7 +2441,7 @@ enum dc_status dc_link_validate_mode_timing(
 	/* A hack to avoid failing any modes for EDID override feature on
 	 * topology change such as lower quality cable for DP or different dongle
 	 */
-	if (link->remote_sinks[0])
+	if (link->remote_sinks[0] && link->remote_sinks[0]->sink_signal == SIGNAL_TYPE_VIRTUAL)
 		return DC_OK;
 
 	/* Passive Dongle */
@@ -2566,7 +2566,7 @@ bool dc_link_set_psr_allow_active(struct dc_link *link, bool allow_active, bool
 	link->psr_settings.psr_allow_active = allow_active;
 
 	if (psr != NULL && link->psr_settings.psr_feature_enabled)
-		psr->funcs->psr_enable(psr, allow_active);
+		psr->funcs->psr_enable(psr, allow_active, wait);
 	else if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_settings.psr_feature_enabled)
 		dmcu->funcs->set_psr_enable(dmcu, allow_active, wait);
 	else
@@ -2946,7 +2946,7 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx)
 	pbn = get_pbn_from_timing(pipe_ctx);
 	avg_time_slots_per_mtp = dc_fixpt_div(pbn, pbn_per_slot);
 
-	stream_encoder->funcs->set_mst_bandwidth(
+	stream_encoder->funcs->set_throttled_vcp_size(
 		stream_encoder,
 		avg_time_slots_per_mtp);
 
@@ -2974,7 +2974,7 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx)
 	 */
 
 	/* slot X.Y */
-	stream_encoder->funcs->set_mst_bandwidth(
+	stream_encoder->funcs->set_throttled_vcp_size(
 		stream_encoder,
 		avg_time_slots_per_mtp);
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
index b984eecca58b..dec12de37642 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
@@ -148,14 +148,6 @@ static uint32_t dal_ddc_i2c_payloads_get_count(struct i2c_payloads *p)
 	return p->payloads.count;
 }
 
-static void dal_ddc_i2c_payloads_destroy(struct i2c_payloads *p)
-{
-	if (!p)
-		return;
-
-	dal_vector_destruct(&p->payloads);
-}
-
 #define DDC_MIN(a, b) (((a) < (b)) ? (a) : (b))
 
 void dal_ddc_i2c_payloads_add(
@@ -582,7 +574,7 @@ bool dal_ddc_service_query_ddc_data(
 				ddc->link,
 				&command);
 
-		dal_ddc_i2c_payloads_destroy(&payloads);
+		dal_vector_destruct(&payloads.payloads);
 	}
 
 	return success;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
index b2be6ad5101d..ff1e9963ec7a 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
@@ -49,14 +49,31 @@ static struct dc_link_settings get_common_supported_link_settings(
 		struct dc_link_settings link_setting_a,
 		struct dc_link_settings link_setting_b);
 
-static uint32_t get_training_aux_rd_interval(
-	struct dc_link *link,
-	uint32_t default_wait_in_micro_secs)
+static uint32_t get_cr_training_aux_rd_interval(struct dc_link *link,
+		const struct dc_link_settings *link_settings)
 {
 	union training_aux_rd_interval training_rd_interval;
+	uint32_t wait_in_micro_secs = 100;
 
 	memset(&training_rd_interval, 0, sizeof(training_rd_interval));
+	core_link_read_dpcd(
+			link,
+			DP_TRAINING_AUX_RD_INTERVAL,
+			(uint8_t *)&training_rd_interval,
+			sizeof(training_rd_interval));
+	if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL)
+		wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000;
+	return wait_in_micro_secs;
+}
+
+static uint32_t get_eq_training_aux_rd_interval(
+	struct dc_link *link,
+	const struct dc_link_settings *link_settings)
+{
+	union training_aux_rd_interval training_rd_interval;
+	uint32_t wait_in_micro_secs = 400;
 
+	memset(&training_rd_interval, 0, sizeof(training_rd_interval));
 	/* overwrite the delay if rev > 1.1*/
 	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12) {
 		/* DP 1.2 or later - retrieve delay through
@@ -68,10 +85,10 @@ static uint32_t get_training_aux_rd_interval(
 			sizeof(training_rd_interval));
 
 		if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL)
-			default_wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000;
+			wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000;
 	}
 
-	return default_wait_in_micro_secs;
+	return wait_in_micro_secs;
 }
 
 static void wait_for_training_aux_rd_interval(
@@ -101,7 +118,16 @@ static void dpcd_set_training_pattern(
 		dpcd_pattern.v1_4.TRAINING_PATTERN_SET);
 }
 
-static enum dc_dp_training_pattern get_supported_tp(struct dc_link *link)
+static enum dc_dp_training_pattern decide_cr_training_pattern(
+		const struct dc_link_settings *link_settings)
+{
+	enum dc_dp_training_pattern pattern = DP_TRAINING_PATTERN_SEQUENCE_1;
+
+	return pattern;
+}
+
+static enum dc_dp_training_pattern decide_eq_training_pattern(struct dc_link *link,
+		const struct dc_link_settings *link_settings)
 {
 	enum dc_dp_training_pattern highest_tp = DP_TRAINING_PATTERN_SEQUENCE_2;
 	struct encoder_feature_support *features = &link->link_enc->features;
@@ -132,7 +158,6 @@ static void dpcd_set_link_settings(
 
 	union down_spread_ctrl downspread = { {0} };
 	union lane_count_set lane_count_set = { {0} };
-	enum dc_dp_training_pattern dp_tr_pattern;
 
 	downspread.raw = (uint8_t)
 	(lt_settings->link_settings.link_spread);
@@ -143,9 +168,8 @@ static void dpcd_set_link_settings(
 	lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing;
 	lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0;
 
-	dp_tr_pattern = get_supported_tp(link);
 
-	if (dp_tr_pattern != DP_TRAINING_PATTERN_SEQUENCE_4) {
+	if (lt_settings->pattern_for_eq < DP_TRAINING_PATTERN_SEQUENCE_4) {
 		lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED =
 				link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED;
 	}
@@ -373,34 +397,30 @@ static void dpcd_set_lt_pattern_and_lane_settings(
 static bool is_cr_done(enum dc_lane_count ln_count,
 	union lane_status *dpcd_lane_status)
 {
-	bool done = true;
 	uint32_t lane;
 	/*LANEx_CR_DONE bits All 1's?*/
 	for (lane = 0; lane < (uint32_t)(ln_count); lane++) {
 		if (!dpcd_lane_status[lane].bits.CR_DONE_0)
-			done = false;
+			return false;
 	}
-	return done;
-
+	return true;
 }
 
 static bool is_ch_eq_done(enum dc_lane_count ln_count,
 	union lane_status *dpcd_lane_status,
 	union lane_align_status_updated *lane_status_updated)
 {
-	bool done = true;
 	uint32_t lane;
 	if (!lane_status_updated->bits.INTERLANE_ALIGN_DONE)
-		done = false;
+		return false;
 	else {
 		for (lane = 0; lane < (uint32_t)(ln_count); lane++) {
 			if (!dpcd_lane_status[lane].bits.SYMBOL_LOCKED_0 ||
 				!dpcd_lane_status[lane].bits.CHANNEL_EQ_DONE_0)
-				done = false;
+				return false;
 		}
 	}
-	return done;
-
+	return true;
 }
 
 static void update_drive_settings(
@@ -979,7 +999,7 @@ static void start_clock_recovery_pattern_early(struct dc_link *link,
 {
 	DC_LOG_HW_LINK_TRAINING("%s\n GPU sends TPS1. Wait 400us.\n",
 			__func__);
-	dp_set_hw_training_pattern(link, DP_TRAINING_PATTERN_SEQUENCE_1, offset);
+	dp_set_hw_training_pattern(link, lt_settings->pattern_for_cr, offset);
 	dp_set_hw_lane_settings(link, lt_settings, offset);
 	udelay(400);
 }
@@ -994,7 +1014,6 @@ static enum link_training_result perform_clock_recovery_sequence(
 	uint32_t wait_time_microsec;
 	struct link_training_settings req_settings;
 	enum dc_lane_count lane_count = lt_settings->link_settings.lane_count;
-	enum dc_dp_training_pattern tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_1;
 	union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX];
 	union lane_align_status_updated dpcd_lane_status_updated;
 
@@ -1002,7 +1021,7 @@ static enum link_training_result perform_clock_recovery_sequence(
 	retry_count = 0;
 
 	if (!link->ctx->dc->work_arounds.lt_early_cr_pattern)
-		dp_set_hw_training_pattern(link, tr_pattern, offset);
+		dp_set_hw_training_pattern(link, lt_settings->pattern_for_cr, offset);
 
 	/* najeeb - The synaptics MST hub can put the LT in
 	* infinite loop by switching the VS
@@ -1029,7 +1048,7 @@ static enum link_training_result perform_clock_recovery_sequence(
 			dpcd_set_lt_pattern_and_lane_settings(
 					link,
 					lt_settings,
-					tr_pattern,
+					lt_settings->pattern_for_cr,
 					offset);
 		else
 			dpcd_set_lane_settings(
@@ -1113,7 +1132,7 @@ static inline enum link_training_result perform_link_training_int(
 	 * TPS4 must be used instead of POST_LT_ADJ_REQ.
 	 */
 	if (link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED != 1 ||
-			get_supported_tp(link) == DP_TRAINING_PATTERN_SEQUENCE_4)
+			lt_settings->pattern_for_eq == DP_TRAINING_PATTERN_SEQUENCE_4)
 		return status;
 
 	if (status == LINK_TRAINING_SUCCESS &&
@@ -1245,17 +1264,21 @@ static void initialize_training_settings(
 	if (overrides->cr_pattern_time != NULL)
 		lt_settings->cr_pattern_time = *overrides->cr_pattern_time;
 	else
-		lt_settings->cr_pattern_time = get_training_aux_rd_interval(link, 100);
+		lt_settings->cr_pattern_time = get_cr_training_aux_rd_interval(link, link_setting);
 
 	if (overrides->eq_pattern_time != NULL)
 		lt_settings->eq_pattern_time = *overrides->eq_pattern_time;
 	else
-		lt_settings->eq_pattern_time = get_training_aux_rd_interval(link, 400);
+		lt_settings->eq_pattern_time = get_eq_training_aux_rd_interval(link, link_setting);
 
+	if (overrides->pattern_for_cr != NULL)
+		lt_settings->pattern_for_cr = *overrides->pattern_for_cr;
+	else
+		lt_settings->pattern_for_cr = decide_cr_training_pattern(link_setting);
 	if (overrides->pattern_for_eq != NULL)
 		lt_settings->pattern_for_eq = *overrides->pattern_for_eq;
 	else
-		lt_settings->pattern_for_eq = get_supported_tp(link);
+		lt_settings->pattern_for_eq = decide_eq_training_pattern(link, link_setting);
 
 	if (overrides->enhanced_framing != NULL)
 		lt_settings->enhanced_framing = *overrides->enhanced_framing;
@@ -1457,7 +1480,6 @@ bool dc_link_dp_perform_link_training_skip_aux(
 	const struct dc_link_settings *link_setting)
 {
 	struct link_training_settings lt_settings;
-	enum dc_dp_training_pattern pattern_for_cr = DP_TRAINING_PATTERN_SEQUENCE_1;
 
 	initialize_training_settings(
 			link,
@@ -1468,7 +1490,7 @@ bool dc_link_dp_perform_link_training_skip_aux(
 	/* 1. Perform_clock_recovery_sequence. */
 
 	/* transmit training pattern for clock recovery */
-	dp_set_hw_training_pattern(link, pattern_for_cr, DPRX);
+	dp_set_hw_training_pattern(link, lt_settings.pattern_for_cr, DPRX);
 
 	/* call HWSS to set lane settings*/
 	dp_set_hw_lane_settings(link, &lt_settings, DPRX);
@@ -1610,6 +1632,9 @@ bool perform_link_training_with_retries(
 
 	for (j = 0; j < attempts; ++j) {
 
+		DC_LOG_HW_LINK_TRAINING("%s: Beginning link training attempt %u of %d\n",
+			__func__, (unsigned int)j + 1, attempts);
+
 		dp_enable_link_phy(
 			link,
 			signal,
@@ -1638,6 +1663,9 @@ bool perform_link_training_with_retries(
 		if (j == (attempts - 1))
 			break;
 
+		DC_LOG_WARNING("%s: Link training attempt %u of %d failed\n",
+			__func__, (unsigned int)j + 1, attempts);
+
 		dp_disable_link_phy(link, signal);
 
 		msleep(delay_between_attempts);
@@ -2431,6 +2459,12 @@ static bool decide_edp_link_settings(struct dc_link *link, struct dc_link_settin
 	return false;
 }
 
+static bool decide_mst_link_settings(const struct dc_link *link, struct dc_link_settings *link_setting)
+{
+	*link_setting = link->verified_link_cap;
+	return true;
+}
+
 void decide_link_settings(struct dc_stream_state *stream,
 	struct dc_link_settings *link_setting)
 {
@@ -2456,11 +2490,9 @@ void decide_link_settings(struct dc_stream_state *stream,
 	 * TODO: add MST specific link training routine
 	 */
 	if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
-		*link_setting = link->verified_link_cap;
-		return;
-	}
-
-	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+		if (decide_mst_link_settings(link, link_setting))
+			return;
+	} else if (link->connector_signal == SIGNAL_TYPE_EDP) {
 		if (decide_edp_link_settings(link, link_setting, req_bw))
 			return;
 	} else if (decide_dp_link_settings(link, link_setting, req_bw))
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
index dd88eb348dfa..11a619befb42 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
@@ -104,6 +104,12 @@ void dp_enable_link_phy(
 	struct clock_source *dp_cs =
 			link->dc->res_pool->dp_clock_source;
 	unsigned int i;
+
+	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+		link->dc->hwss.edp_power_control(link, true);
+		link->dc->hwss.edp_wait_for_hpd_ready(link, true);
+	}
+
 	/* If the current pixel clock source is not DTO(happens after
 	 * switching from HDMI passive dongle to DP on the same connector),
 	 * switch the pixel clock source to DTO.
@@ -223,6 +229,8 @@ void dp_disable_link_phy(struct dc_link *link, enum signal_type signal)
 		dp_receiver_power_ctrl(link, false);
 
 	if (signal == SIGNAL_TYPE_EDP) {
+		if (link->dc->hwss.edp_backlight_control)
+			link->dc->hwss.edp_backlight_control(link, false);
 		link->link_enc->funcs->disable_output(link->link_enc, signal);
 		link->dc->hwss.edp_power_control(link, false);
 	} else {
@@ -485,13 +493,15 @@ void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 				OPTC_DSC_DISABLED, 0, 0);
 
 		/* disable DSC in stream encoder */
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(
-					pipe_ctx->stream_res.stream_enc,
-					OPTC_DSC_DISABLED, 0, 0);
-
-			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
-					pipe_ctx->stream_res.stream_enc, false, NULL);
+		if (dc_is_dp_signal(stream->signal)) {
+
+			if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+				pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(
+						pipe_ctx->stream_res.stream_enc,
+						OPTC_DSC_DISABLED, 0, 0);
+				pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
+							pipe_ctx->stream_res.stream_enc, false, NULL);
+			}
 		}
 
 		/* disable DSC block */
@@ -528,7 +538,6 @@ out:
 bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 {
 	struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc;
-	struct dc *dc = pipe_ctx->stream->ctx->dc;
 	struct dc_stream_state *stream = pipe_ctx->stream;
 
 	if (!pipe_ctx->stream->timing.flags.DSC || !dsc)
@@ -551,7 +560,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 
 		DC_LOG_DSC(" ");
 		dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]);
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+		if (dc_is_dp_signal(stream->signal)) {
 			DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id);
 			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
 									pipe_ctx->stream_res.stream_enc,
@@ -560,7 +569,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 		}
 	} else {
 		/* disable DSC PPS in stream encoder */
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+		if (dc_is_dp_signal(stream->signal)) {
 			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
 						pipe_ctx->stream_res.stream_enc, false, NULL);
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 7b5f90ebb133..e430148e47cf 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -42,6 +42,9 @@
 #include "virtual/virtual_stream_encoder.h"
 #include "dpcd_defs.h"
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/dce60_resource.h"
+#endif
 #include "dce80/dce80_resource.h"
 #include "dce100/dce100_resource.h"
 #include "dce110/dce110_resource.h"
@@ -63,6 +66,18 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id)
 	enum dce_version dc_version = DCE_VERSION_UNKNOWN;
 	switch (asic_id.chip_family) {
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case FAMILY_SI:
+		if (ASIC_REV_IS_TAHITI_P(asic_id.hw_internal_rev) ||
+		    ASIC_REV_IS_PITCAIRN_PM(asic_id.hw_internal_rev) ||
+		    ASIC_REV_IS_CAPEVERDE_M(asic_id.hw_internal_rev))
+			dc_version = DCE_VERSION_6_0;
+		else if (ASIC_REV_IS_OLAND_M(asic_id.hw_internal_rev))
+			dc_version = DCE_VERSION_6_4;
+		else
+			dc_version = DCE_VERSION_6_1;
+		break;
+#endif
 	case FAMILY_CI:
 		dc_version = DCE_VERSION_8_0;
 		break;
@@ -129,6 +144,20 @@ struct resource_pool *dc_create_resource_pool(struct dc  *dc,
 	struct resource_pool *res_pool = NULL;
 
 	switch (dc_version) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case DCE_VERSION_6_0:
+		res_pool = dce60_create_resource_pool(
+			init_data->num_virtual_links, dc);
+		break;
+	case DCE_VERSION_6_1:
+		res_pool = dce61_create_resource_pool(
+			init_data->num_virtual_links, dc);
+		break;
+	case DCE_VERSION_6_4:
+		res_pool = dce64_create_resource_pool(
+			init_data->num_virtual_links, dc);
+		break;
+#endif
 	case DCE_VERSION_8_0:
 		res_pool = dce80_create_resource_pool(
 				init_data->num_virtual_links, dc);
@@ -753,11 +782,18 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx)
 
 	calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx);
 
-	data->recout.x = stream->dst.x;
-	if (stream->src.x < surf_clip.x)
-		data->recout.x += (surf_clip.x - stream->src.x) * stream->dst.width
+	/*
+	 * Only the leftmost ODM pipe should be offset by a nonzero distance
+	 */
+	if (!pipe_ctx->prev_odm_pipe) {
+		data->recout.x = stream->dst.x;
+		if (stream->src.x < surf_clip.x)
+			data->recout.x += (surf_clip.x - stream->src.x) * stream->dst.width
 						/ stream->src.width;
 
+	} else
+		data->recout.x = 0;
+
 	data->recout.width = surf_clip.width * stream->dst.width / stream->src.width;
 	if (data->recout.width + data->recout.x > stream->dst.x + stream->dst.width)
 		data->recout.width = stream->dst.x + stream->dst.width - data->recout.x;
@@ -928,7 +964,7 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx)
 {
 	const struct dc_plane_state *plane_state = pipe_ctx->plane_state;
 	const struct dc_stream_state *stream = pipe_ctx->stream;
-	struct pipe_ctx *odm_pipe = pipe_ctx->prev_odm_pipe;
+	struct pipe_ctx *odm_pipe = pipe_ctx;
 	struct scaler_data *data = &pipe_ctx->plane_res.scl_data;
 	struct rect src = pipe_ctx->plane_state->src_rect;
 	int recout_skip_h, recout_skip_v, surf_size_h, surf_size_v;
@@ -959,21 +995,24 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx)
 		swap(src.width, src.height);
 	}
 
+	/*modified recout_skip_h calculation due to odm having no recout offset*/
+	while (odm_pipe->prev_odm_pipe) {
+		odm_idx++;
+		odm_pipe = odm_pipe->prev_odm_pipe;
+	}
+	/*odm_pipe is the leftmost pipe in the ODM group*/
+	recout_skip_h = odm_idx * data->recout.width;
+
 	/* Recout matching initial vp offset = recout_offset - (stream dst offset +
 	 *			((surf dst offset - stream src offset) * 1/ stream scaling ratio)
 	 *			- (surf surf_src offset * 1/ full scl ratio))
 	 */
-	recout_skip_h = data->recout.x - (stream->dst.x + (plane_state->dst_rect.x - stream->src.x)
+	recout_skip_h += odm_pipe->plane_res.scl_data.recout.x
+				- (stream->dst.x + (plane_state->dst_rect.x - stream->src.x)
 					* stream->dst.width / stream->src.width -
 					src.x * plane_state->dst_rect.width / src.width
 					* stream->dst.width / stream->src.width);
-	/*modified recout_skip_h calculation due to odm having no recout offset*/
-	while (odm_pipe) {
-		odm_idx++;
-		odm_pipe = odm_pipe->prev_odm_pipe;
-	}
-	if (odm_idx)
-		recout_skip_h += odm_idx * data->recout.width;
+
 
 	recout_skip_v = data->recout.y - (stream->dst.y + (plane_state->dst_rect.y - stream->src.y)
 					* stream->dst.height / stream->src.height -
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 0257a900fe2b..d48fd87d3b95 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -123,7 +123,6 @@ static bool dc_stream_construct(struct dc_stream_state *stream,
 		return false;
 	}
 	stream->out_transfer_func->type = TF_TYPE_BYPASS;
-	stream->out_transfer_func->ctx = stream->ctx;
 
 	stream->stream_id = stream->ctx->dc_stream_id_count;
 	stream->ctx->dc_stream_id_count++;
@@ -298,7 +297,7 @@ bool dc_stream_set_cursor_attributes(
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	/* disable idle optimizations while updating cursor */
 	if (dc->idle_optimizations_allowed) {
-		dc->hwss.apply_idle_power_optimizations(dc, false);
+		dc_allow_idle_optimizations(dc, false);
 		reset_idle_optimizations = true;
 	}
 
@@ -326,7 +325,7 @@ bool dc_stream_set_cursor_attributes(
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	/* re-enable idle optimizations if necessary */
 	if (reset_idle_optimizations)
-		dc->hwss.apply_idle_power_optimizations(dc, true);
+		dc_allow_idle_optimizations(dc, true);
 
 #endif
 	return true;
@@ -359,9 +358,8 @@ bool dc_stream_set_cursor_position(
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 
 	/* disable idle optimizations if enabling cursor */
-	if (dc->idle_optimizations_allowed &&
-			!stream->cursor_position.enable && position->enable) {
-		dc->hwss.apply_idle_power_optimizations(dc, false);
+	if (dc->idle_optimizations_allowed && !stream->cursor_position.enable && position->enable) {
+		dc_allow_idle_optimizations(dc, false);
 		reset_idle_optimizations = true;
 	}
 
@@ -392,7 +390,7 @@ bool dc_stream_set_cursor_position(
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	/* re-enable idle optimizations if necessary */
 	if (reset_idle_optimizations)
-		dc->hwss.apply_idle_power_optimizations(dc, true);
+		dc_allow_idle_optimizations(dc, true);
 
 #endif
 	return true;
@@ -708,3 +706,4 @@ void dc_stream_log(const struct dc *dc, const struct dc_stream_state *stream)
 			"\tlink: %d\n",
 			stream->link->link_index);
 }
+
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
index ea1229a3e2b2..3d7d27435f15 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
@@ -48,22 +48,17 @@ static void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *pl
 	plane_state->in_transfer_func = dc_create_transfer_func();
 	if (plane_state->in_transfer_func != NULL) {
 		plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
-		plane_state->in_transfer_func->ctx = ctx;
 	}
 	plane_state->in_shaper_func = dc_create_transfer_func();
 	if (plane_state->in_shaper_func != NULL) {
 		plane_state->in_shaper_func->type = TF_TYPE_BYPASS;
-		plane_state->in_shaper_func->ctx = ctx;
 	}
 
 	plane_state->lut3d_func = dc_create_3dlut_func();
-	if (plane_state->lut3d_func != NULL) {
-		plane_state->lut3d_func->ctx = ctx;
-	}
+
 	plane_state->blend_tf = dc_create_transfer_func();
 	if (plane_state->blend_tf != NULL) {
 		plane_state->blend_tf->type = TF_TYPE_BYPASS;
-		plane_state->blend_tf->ctx = ctx;
 	}
 
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index f50ef4255020..82fe0ab56e3a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -42,7 +42,7 @@
 #include "inc/hw/dmcu.h"
 #include "dml/display_mode_lib.h"
 
-#define DC_VER "3.2.95"
+#define DC_VER "3.2.104"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
@@ -476,7 +476,7 @@ struct dc_debug_options {
 	unsigned int force_odm_combine_4to1; //bit vector based on otg inst
 #endif
 	unsigned int force_fclk_khz;
-	bool disable_tri_buf;
+	bool enable_tri_buf;
 	bool dmub_offload_enabled;
 	bool dmcub_emulation;
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
@@ -503,6 +503,7 @@ struct dc_debug_options {
 	bool usbc_combo_phy_reset_wa;
 	bool disable_dsc;
 	bool enable_dram_clock_change_one_display_vactive;
+	bool force_ignore_link_settings;
 };
 
 struct dc_debug_data {
@@ -660,6 +661,7 @@ struct dc_init_data {
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	bool force_smu_not_present;
 #endif
+	bool force_ignore_link_settings;
 };
 
 struct dc_callback_init {
@@ -745,7 +747,6 @@ struct dc_transfer_func {
 	enum dc_transfer_func_predefined tf;
 	/* FP16 1.0 reference level in nits, default is 80 nits, only for PQ*/
 	uint32_t sdr_ref_white_level;
-	struct dc_context *ctx;
 	union {
 		struct pwl_params pwl;
 		struct dc_transfer_func_distributed_points tf_pts;
@@ -772,7 +773,6 @@ struct dc_3dlut {
 	struct tetrahedral_params lut_3d;
 	struct fixed31_32 hdr_multiplier;
 	union dc_3dlut_state state;
-	struct dc_context *ctx;
 };
 /*
  * This structure is filled in by dc_surface_get_status and contains
@@ -1250,6 +1250,9 @@ enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32
 void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg);
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 
+bool dc_is_plane_eligible_for_idle_optimizations(struct dc *dc,
+						 struct dc_plane_state *plane);
+
 void dc_allow_idle_optimizations(struct dc *dc, bool allow);
 
 /*
@@ -1265,6 +1268,9 @@ void dc_unlock_memory_clock_frequency(struct dc *dc);
 void dc_lock_memory_clock_frequency(struct dc *dc);
 
 #endif
+
+bool dc_set_psr_allow_active(struct dc *dc, bool enable);
+
 /*******************************************************************************
  * DSC Interfaces
  ******************************************************************************/
diff --git a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
index 0811f941f430..e146e3cba8eb 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
@@ -140,6 +140,10 @@ struct dc_vbios_funcs {
 	enum bp_result (*enable_lvtma_control)(
 		struct dc_bios *bios,
 		uint8_t uc_pwr_on);
+
+	enum bp_result (*get_soc_bb_info)(
+		struct dc_bios *dcb,
+		struct bp_soc_bb_info *soc_bb_info);
 };
 
 struct bios_registers {
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index eea2429ac67d..b98754811977 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -132,3 +132,19 @@ void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv)
 		/* Continue spinning so we don't hang the ASIC. */
 	}
 }
+
+bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
+				    unsigned int stream_mask)
+{
+	struct dmub_srv *dmub;
+	const uint32_t timeout = 30;
+
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
+
+	dmub = dc_dmub_srv->dmub;
+
+	return dmub_srv_send_gpint_command(
+		       dmub, DMUB_GPINT__IDLE_OPT_NOTIFY_STREAM_MASK,
+		       stream_mask, timeout) == DMUB_STATUS_OK;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index a3a09ccb6d26..bb4ab61887e4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -56,4 +56,6 @@ void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv);
 
 void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv);
 
+bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
+				    unsigned int stream_mask);
 #endif /* _DMUB_DC_SRV_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index a8a3b0643505..80a2191a3115 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -123,6 +123,7 @@ struct dc_link_training_overrides {
 
 	uint16_t *cr_pattern_time;
 	uint16_t *eq_pattern_time;
+	enum dc_dp_training_pattern *pattern_for_cr;
 	enum dc_dp_training_pattern *pattern_for_eq;
 
 	enum dc_link_spread *downspread;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dsc.h b/drivers/gpu/drm/amd/display/dc/dc_dsc.h
index 3800340a5b4f..768ab38d41cf 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dsc.h
@@ -51,6 +51,7 @@ struct dc_dsc_policy {
 	int min_slice_height; // Must not be less than 8
 	uint32_t max_target_bpp;
 	uint32_t min_target_bpp;
+	bool enable_dsc_when_not_needed;
 };
 
 bool dc_dsc_parse_dsc_dpcd(const struct dc *dc,
@@ -80,4 +81,6 @@ void dc_dsc_get_policy_for_timing(const struct dc_crtc_timing *timing,
 
 void dc_dsc_policy_set_max_target_bpp_limit(uint32_t limit);
 
+void dc_dsc_policy_set_enable_dsc_when_not_needed(bool enable);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index e002ef706e1d..266b93a705d5 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -237,6 +237,8 @@ enum dc_detect_reason {
 	DETECT_REASON_BOOT,
 	DETECT_REASON_HPD,
 	DETECT_REASON_HPDRX,
+	DETECT_REASON_FALLBACK,
+	DETECT_REASON_RETRAIN
 };
 
 bool dc_link_detect(struct dc_link *dc_link, enum dc_detect_reason reason);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index d9888f316da6..c246af7c584b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -397,6 +397,8 @@ bool dc_enable_stereo(
 	struct dc_stream_state *streams[],
 	uint8_t stream_count);
 
+/* Triggers multi-stream synchronization. */
+void dc_trigger_sync(struct dc *dc, struct dc_state *context);
 
 enum surface_update_type dc_check_update_surfaces_for_stream(
 		struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 946ba929c6f6..c47a19719de2 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -122,7 +122,7 @@ struct dc_context {
 };
 
 
-#define DC_MAX_EDID_BUFFER_SIZE 1024
+#define DC_MAX_EDID_BUFFER_SIZE 1280
 #define DC_EDID_BLOCK_SIZE 128
 #define MAX_SURFACE_NUM 4
 #define NUM_PIXEL_FORMATS 10
@@ -233,6 +233,7 @@ struct dc_panel_patch {
 	unsigned int skip_scdc_overwrite;
 	unsigned int delay_ignore_msa;
 	unsigned int disable_fec;
+	unsigned int extra_t3_ms;
 };
 
 struct dc_edid_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
index a44effcda49f..e84d21605854 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
@@ -46,6 +46,8 @@
 	SR(BL1_PWM_USER_LEVEL), \
 	SR(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES), \
 	SR(DC_ABM1_HGLS_REG_READ_PROGRESS), \
+	SR(DC_ABM1_ACE_OFFSET_SLOPE_0), \
+	SR(DC_ABM1_ACE_THRES_12), \
 	SR(BIOS_SCRATCH_2)
 
 #define ABM_DCN10_REG_LIST(id)\
@@ -60,6 +62,8 @@
 	SRI(BL1_PWM_USER_LEVEL, ABM, id), \
 	SRI(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, ABM, id), \
 	SRI(DC_ABM1_HGLS_REG_READ_PROGRESS, ABM, id), \
+	SRI(DC_ABM1_ACE_OFFSET_SLOPE_0, ABM, id), \
+	SRI(DC_ABM1_ACE_THRES_12, ABM, id), \
 	NBIO_SR(BIOS_SCRATCH_2)
 
 #define ABM_DCN20_REG_LIST() \
@@ -74,10 +78,12 @@
 	SR(BL1_PWM_USER_LEVEL), \
 	SR(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES), \
 	SR(DC_ABM1_HGLS_REG_READ_PROGRESS), \
+	SR(DC_ABM1_ACE_OFFSET_SLOPE_0), \
+	SR(DC_ABM1_ACE_THRES_12), \
 	NBIO_SR(BIOS_SCRATCH_2)
 
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
-#define ABM_DCN301_REG_LIST(id)\
+#define ABM_DCN30_REG_LIST(id)\
 	ABM_COMMON_REG_LIST_DCE_BASE(), \
 	SRI(DC_ABM1_HG_SAMPLE_RATE, ABM, id), \
 	SRI(DC_ABM1_LS_SAMPLE_RATE, ABM, id), \
@@ -89,6 +95,8 @@
 	SRI(BL1_PWM_USER_LEVEL, ABM, id), \
 	SRI(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, ABM, id), \
 	SRI(DC_ABM1_HGLS_REG_READ_PROGRESS, ABM, id), \
+	SRI(DC_ABM1_ACE_OFFSET_SLOPE_0, ABM, id), \
+	SRI(DC_ABM1_ACE_THRES_12, ABM, id), \
 	NBIO_SR(BIOS_SCRATCH_2)
 #endif
 
@@ -208,6 +216,8 @@ struct dce_abm_registers {
 	uint32_t BL1_PWM_USER_LEVEL;
 	uint32_t DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES;
 	uint32_t DC_ABM1_HGLS_REG_READ_PROGRESS;
+	uint32_t DC_ABM1_ACE_OFFSET_SLOPE_0;
+	uint32_t DC_ABM1_ACE_THRES_12;
 	uint32_t MASTER_COMM_CNTL_REG;
 	uint32_t MASTER_COMM_CMD_REG;
 	uint32_t MASTER_COMM_DATA_REG1;
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c
index 408046579712..2a2a0fdb9253 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c
@@ -867,6 +867,98 @@ void dce_aud_wall_dto_setup(
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_aud_wall_dto_setup(
+	struct audio *audio,
+	enum signal_type signal,
+	const struct audio_crtc_info *crtc_info,
+	const struct audio_pll_info *pll_info)
+{
+	struct dce_audio *aud = DCE_AUD(audio);
+
+	struct azalia_clock_info clock_info = { 0 };
+
+	if (dc_is_hdmi_signal(signal)) {
+		uint32_t src_sel;
+
+		/*DTO0 Programming goal:
+		-generate 24MHz, 128*Fs from 24MHz
+		-use DTO0 when an active HDMI port is connected
+		(optionally a DP is connected) */
+
+		/* calculate DTO settings */
+		get_azalia_clock_info_hdmi(
+			crtc_info->requested_pixel_clock_100Hz,
+			crtc_info->calculated_pixel_clock_100Hz,
+			&clock_info);
+
+		DC_LOG_HW_AUDIO("\n%s:Input::requested_pixel_clock_100Hz = %d"\
+				"calculated_pixel_clock_100Hz =%d\n"\
+				"audio_dto_module = %d audio_dto_phase =%d \n\n", __func__,\
+				crtc_info->requested_pixel_clock_100Hz,\
+				crtc_info->calculated_pixel_clock_100Hz,\
+				clock_info.audio_dto_module,\
+				clock_info.audio_dto_phase);
+
+		/* On TN/SI, Program DTO source select and DTO select before
+		programming DTO modulo and DTO phase. These bits must be
+		programmed first, otherwise there will be no HDMI audio at boot
+		up. This is a HW sequence change (different from old ASICs).
+		Caution when changing this programming sequence.
+
+		HDMI enabled, using DTO0
+		program master CRTC for DTO0 */
+		src_sel = pll_info->dto_source - DTO_SOURCE_ID0;
+		REG_UPDATE_2(DCCG_AUDIO_DTO_SOURCE,
+			DCCG_AUDIO_DTO0_SOURCE_SEL, src_sel,
+			DCCG_AUDIO_DTO_SEL, 0);
+
+		/* module */
+		REG_UPDATE(DCCG_AUDIO_DTO0_MODULE,
+			DCCG_AUDIO_DTO0_MODULE, clock_info.audio_dto_module);
+
+		/* phase */
+		REG_UPDATE(DCCG_AUDIO_DTO0_PHASE,
+			DCCG_AUDIO_DTO0_PHASE, clock_info.audio_dto_phase);
+	} else {
+		/*DTO1 Programming goal:
+		-generate 24MHz, 128*Fs from 24MHz (DCE6 does not support 512*Fs)
+		-default is to used DTO1, and switch to DTO0 when an audio
+		master HDMI port is connected
+		-use as default for DP
+
+		calculate DTO settings */
+		get_azalia_clock_info_dp(
+			crtc_info->requested_pixel_clock_100Hz,
+			pll_info,
+			&clock_info);
+
+		/* Program DTO select before programming DTO modulo and DTO
+		phase. default to use DTO1 */
+
+		REG_UPDATE(DCCG_AUDIO_DTO_SOURCE,
+				DCCG_AUDIO_DTO_SEL, 1);
+
+			/* DCCG_AUDIO_DTO2_USE_512FBR_DTO, 1)
+			 * Cannot select 512fs for DP
+			 *
+			 * DCE6 has no DCCG_AUDIO_DTO2_USE_512FBR_DTO mask
+			*/
+
+		/* module */
+		REG_UPDATE(DCCG_AUDIO_DTO1_MODULE,
+				DCCG_AUDIO_DTO1_MODULE, clock_info.audio_dto_module);
+
+		/* phase */
+		REG_UPDATE(DCCG_AUDIO_DTO1_PHASE,
+				DCCG_AUDIO_DTO1_PHASE, clock_info.audio_dto_phase);
+
+		/* DCE6 has no DCCG_AUDIO_DTO2_USE_512FBR_DTO mask in DCCG_AUDIO_DTO_SOURCE reg */
+
+	}
+}
+#endif
+
 static bool dce_aud_endpoint_valid(struct audio *audio)
 {
 	uint32_t value;
@@ -926,6 +1018,19 @@ static const struct audio_funcs funcs = {
 	.az_configure = dce_aud_az_configure,
 	.destroy = dce_aud_destroy,
 };
+
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct audio_funcs dce60_funcs = {
+	.endpoint_valid = dce_aud_endpoint_valid,
+	.hw_init = dce_aud_hw_init,
+	.wall_dto_setup = dce60_aud_wall_dto_setup,
+	.az_enable = dce_aud_az_enable,
+	.az_disable = dce_aud_az_disable,
+	.az_configure = dce_aud_az_configure,
+	.destroy = dce_aud_destroy,
+};
+#endif
+
 void dce_aud_destroy(struct audio **audio)
 {
 	struct dce_audio *aud = DCE_AUD(*audio);
@@ -959,3 +1064,29 @@ struct audio *dce_audio_create(
 	return &audio->base;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+struct audio *dce60_audio_create(
+		struct dc_context *ctx,
+		unsigned int inst,
+		const struct dce_audio_registers *reg,
+		const struct dce_audio_shift *shifts,
+		const struct dce_audio_mask *masks
+		)
+{
+	struct dce_audio *audio = kzalloc(sizeof(*audio), GFP_KERNEL);
+
+	if (audio == NULL) {
+		ASSERT_CRITICAL(audio);
+		return NULL;
+	}
+
+	audio->base.ctx = ctx;
+	audio->base.inst = inst;
+	audio->base.funcs = &dce60_funcs;
+
+	audio->regs = reg;
+	audio->shifts = shifts;
+	audio->masks = masks;
+	return &audio->base;
+}
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h
index 1392fab0860b..5622d5e32d81 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h
@@ -64,6 +64,20 @@
 		SF(AZALIA_F0_CODEC_ENDPOINT_INDEX, AZALIA_ENDPOINT_REG_INDEX, mask_sh),\
 		SF(AZALIA_F0_CODEC_ENDPOINT_DATA, AZALIA_ENDPOINT_REG_DATA, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define AUD_DCE60_MASK_SH_LIST(mask_sh)\
+		SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL, mask_sh),\
+		SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, mask_sh),\
+		SF(DCCG_AUDIO_DTO0_MODULE, DCCG_AUDIO_DTO0_MODULE, mask_sh),\
+		SF(DCCG_AUDIO_DTO0_PHASE, DCCG_AUDIO_DTO0_PHASE, mask_sh),\
+		SF(DCCG_AUDIO_DTO1_MODULE, DCCG_AUDIO_DTO1_MODULE, mask_sh),\
+		SF(DCCG_AUDIO_DTO1_PHASE, DCCG_AUDIO_DTO1_PHASE, mask_sh),\
+		SF(AZALIA_F0_CODEC_FUNCTION_PARAMETER_SUPPORTED_SIZE_RATES, AUDIO_RATE_CAPABILITIES, mask_sh),\
+		SF(AZALIA_F0_CODEC_FUNCTION_PARAMETER_POWER_STATES, CLKSTOP, mask_sh),\
+		SF(AZALIA_F0_CODEC_FUNCTION_PARAMETER_POWER_STATES, EPSS, mask_sh), \
+		SF(AZALIA_F0_CODEC_ENDPOINT_INDEX, AZALIA_ENDPOINT_REG_INDEX, mask_sh),\
+		SF(AZALIA_F0_CODEC_ENDPOINT_DATA, AZALIA_ENDPOINT_REG_DATA, mask_sh)
+#endif
 
 struct dce_audio_registers {
 	uint32_t AZALIA_F0_CODEC_ENDPOINT_INDEX;
@@ -135,6 +149,15 @@ struct audio *dce_audio_create(
 		const struct dce_audio_shift *shifts,
 		const struct dce_audio_mask *masks);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+struct audio *dce60_audio_create(
+		struct dc_context *ctx,
+		unsigned int inst,
+		const struct dce_audio_registers *reg,
+		const struct dce_audio_shift *shifts,
+		const struct dce_audio_mask *masks);
+#endif
+
 void dce_aud_destroy(struct audio **audio);
 
 void dce_aud_hw_init(struct audio *audio);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
index 5e044c2d3d6d..93e7f34d4775 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
@@ -46,6 +46,24 @@
 	SR(SMU_INTERRUPT_CONTROL), \
 	SR(DC_DMCU_SCRATCH)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define DMCU_DCE60_REG_LIST() \
+	SR(DMCU_CTRL), \
+	SR(DMCU_STATUS), \
+	SR(DMCU_RAM_ACCESS_CTRL), \
+	SR(DMCU_IRAM_WR_CTRL), \
+	SR(DMCU_IRAM_WR_DATA), \
+	SR(MASTER_COMM_DATA_REG1), \
+	SR(MASTER_COMM_DATA_REG2), \
+	SR(MASTER_COMM_DATA_REG3), \
+	SR(MASTER_COMM_CMD_REG), \
+	SR(MASTER_COMM_CNTL_REG), \
+	SR(DMCU_IRAM_RD_CTRL), \
+	SR(DMCU_IRAM_RD_DATA), \
+	SR(DMCU_INTERRUPT_TO_UC_EN_MASK), \
+	SR(DC_DMCU_SCRATCH)
+#endif
+
 #define DMCU_DCE80_REG_LIST() \
 	SR(DMCU_CTRL), \
 	SR(DMCU_STATUS), \
@@ -104,6 +122,25 @@
 			STATIC_SCREEN4_INT_TO_UC_EN, mask_sh), \
 	DMCU_SF(SMU_INTERRUPT_CONTROL, DC_SMU_INT_ENABLE, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define DMCU_MASK_SH_LIST_DCE60(mask_sh) \
+	DMCU_SF(DMCU_CTRL, \
+			DMCU_ENABLE, mask_sh), \
+	DMCU_SF(DMCU_STATUS, \
+			UC_IN_STOP_MODE, mask_sh), \
+	DMCU_SF(DMCU_STATUS, \
+			UC_IN_RESET, mask_sh), \
+	DMCU_SF(DMCU_RAM_ACCESS_CTRL, \
+			IRAM_HOST_ACCESS_EN, mask_sh), \
+	DMCU_SF(DMCU_RAM_ACCESS_CTRL, \
+			IRAM_WR_ADDR_AUTO_INC, mask_sh), \
+	DMCU_SF(DMCU_RAM_ACCESS_CTRL, \
+			IRAM_RD_ADDR_AUTO_INC, mask_sh), \
+	DMCU_SF(MASTER_COMM_CMD_REG, \
+			MASTER_COMM_CMD_REG_BYTE0, mask_sh), \
+	DMCU_SF(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, mask_sh)
+#endif
+
 #define DMCU_MASK_SH_LIST_DCE80(mask_sh) \
 	DMCU_SF(DMCU_CTRL, \
 			DMCU_ENABLE, mask_sh), \
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c
index e1c5839a80dc..4202fadb2c0e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c
@@ -85,6 +85,15 @@ void dce_pipe_control_lock(struct dc *dc,
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_pipe_control_lock(struct dc *dc,
+		struct pipe_ctx *pipe,
+		bool lock)
+{
+	/* DCE6 has no BLND_V_UPDATE_LOCK register */
+}
+#endif
+
 void dce_set_blender_mode(struct dce_hwseq *hws,
 	unsigned int blnd_inst,
 	enum blnd_mode mode)
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
index 66b88d6ba398..70bbc1311327 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
@@ -110,6 +110,12 @@
 	SR(BLNDV_CONTROL),\
 	HWSEQ_PIXEL_RATE_REG_LIST(CRTC)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define HWSEQ_DCE6_REG_LIST() \
+	HWSEQ_DCEF_REG_LIST_DCE8(), \
+	HWSEQ_PIXEL_RATE_REG_LIST(CRTC)
+#endif
+
 #define HWSEQ_DCE8_REG_LIST() \
 	HWSEQ_DCEF_REG_LIST_DCE8(), \
 	HWSEQ_BLND_REG_LIST(), \
@@ -488,6 +494,12 @@ struct dce_hwseq_registers {
 	HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh),\
 	HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PIXEL_RATE_PLL_SOURCE, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define HWSEQ_DCE6_MASK_SH_LIST(mask_sh)\
+	.DCFE_CLOCK_ENABLE = CRTC_DCFE_CLOCK_CONTROL__CRTC_DCFE_CLOCK_ENABLE ## mask_sh, \
+	HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_)
+#endif
+
 #define HWSEQ_DCE8_MASK_SH_LIST(mask_sh)\
 	.DCFE_CLOCK_ENABLE = CRTC_DCFE_CLOCK_CONTROL__CRTC_DCFE_CLOCK_ENABLE ## mask_sh, \
 	HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_V_UPDATE_LOCK, mask_sh),\
@@ -836,6 +848,12 @@ void dce_pipe_control_lock(struct dc *dc,
 void dce_set_blender_mode(struct dce_hwseq *hws,
 	unsigned int blnd_inst, enum blnd_mode mode);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_pipe_control_lock(struct dc *dc,
+		struct pipe_ctx *pipe,
+		bool lock);
+#endif
+
 void dce_clock_gating_power_up(struct dce_hwseq *hws,
 		bool enable);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c
index ce30dbf579d4..80569a2734eb 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.c
@@ -231,6 +231,22 @@ static void dce_ipp_set_degamma(
 		  CURSOR2_DEGAMMA_MODE, degamma_type);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_ipp_set_degamma(
+	struct input_pixel_processor *ipp,
+	enum ipp_degamma_mode mode)
+{
+	struct dce_ipp *ipp_dce = TO_DCE_IPP(ipp);
+	uint32_t degamma_type = (mode == IPP_DEGAMMA_MODE_HW_sRGB) ? 1 : 0;
+
+	ASSERT(mode == IPP_DEGAMMA_MODE_BYPASS || mode == IPP_DEGAMMA_MODE_HW_sRGB);
+	/* DCE6 does not have CURSOR2_DEGAMMA_MODE bit in DEGAMMA_CONTROL reg */
+	REG_SET_2(DEGAMMA_CONTROL, 0,
+		  GRPH_DEGAMMA_MODE, degamma_type,
+		  CURSOR_DEGAMMA_MODE, degamma_type);
+}
+#endif
+
 static const struct ipp_funcs dce_ipp_funcs = {
 	.ipp_cursor_set_attributes = dce_ipp_cursor_set_attributes,
 	.ipp_cursor_set_position = dce_ipp_cursor_set_position,
@@ -239,6 +255,17 @@ static const struct ipp_funcs dce_ipp_funcs = {
 	.ipp_set_degamma = dce_ipp_set_degamma
 };
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct ipp_funcs dce60_ipp_funcs = {
+	.ipp_cursor_set_attributes = dce_ipp_cursor_set_attributes,
+	.ipp_cursor_set_position = dce_ipp_cursor_set_position,
+	.ipp_program_prescale = dce_ipp_program_prescale,
+	.ipp_program_input_lut = dce_ipp_program_input_lut,
+	.ipp_set_degamma = dce60_ipp_set_degamma
+};
+#endif
+
+
 /*****************************************/
 /* Constructor, Destructor               */
 /*****************************************/
@@ -260,6 +287,25 @@ void dce_ipp_construct(
 	ipp_dce->ipp_mask = ipp_mask;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_ipp_construct(
+	struct dce_ipp *ipp_dce,
+	struct dc_context *ctx,
+	int inst,
+	const struct dce_ipp_registers *regs,
+	const struct dce_ipp_shift *ipp_shift,
+	const struct dce_ipp_mask *ipp_mask)
+{
+	ipp_dce->base.ctx = ctx;
+	ipp_dce->base.inst = inst;
+	ipp_dce->base.funcs = &dce60_ipp_funcs;
+
+	ipp_dce->regs = regs;
+	ipp_dce->ipp_shift = ipp_shift;
+	ipp_dce->ipp_mask = ipp_mask;
+}
+#endif
+
 void dce_ipp_destroy(struct input_pixel_processor **ipp)
 {
 	kfree(TO_DCE_IPP(*ipp));
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.h b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.h
index ca04e97d44c3..0028d4bdd81b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_ipp.h
@@ -147,6 +147,46 @@
 	IPP_SF(DCP0_DEGAMMA_CONTROL, CURSOR_DEGAMMA_MODE, mask_sh), \
 	IPP_SF(DCP0_DEGAMMA_CONTROL, CURSOR2_DEGAMMA_MODE, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define IPP_DCE60_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh) \
+	IPP_SF(CUR_UPDATE, CURSOR_UPDATE_LOCK, mask_sh), \
+	IPP_SF(CUR_CONTROL, CURSOR_EN, mask_sh), \
+	IPP_SF(CUR_CONTROL, CURSOR_MODE, mask_sh), \
+	IPP_SF(CUR_CONTROL, CURSOR_2X_MAGNIFY, mask_sh), \
+	IPP_SF(CUR_CONTROL, CUR_INV_TRANS_CLAMP, mask_sh), \
+	IPP_SF(CUR_POSITION, CURSOR_X_POSITION, mask_sh), \
+	IPP_SF(CUR_POSITION, CURSOR_Y_POSITION, mask_sh), \
+	IPP_SF(CUR_HOT_SPOT, CURSOR_HOT_SPOT_X, mask_sh), \
+	IPP_SF(CUR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \
+	IPP_SF(CUR_COLOR1, CUR_COLOR1_BLUE, mask_sh), \
+	IPP_SF(CUR_COLOR1, CUR_COLOR1_GREEN, mask_sh), \
+	IPP_SF(CUR_COLOR1, CUR_COLOR1_RED, mask_sh), \
+	IPP_SF(CUR_COLOR2, CUR_COLOR2_BLUE, mask_sh), \
+	IPP_SF(CUR_COLOR2, CUR_COLOR2_GREEN, mask_sh), \
+	IPP_SF(CUR_COLOR2, CUR_COLOR2_RED, mask_sh), \
+	IPP_SF(CUR_SIZE, CURSOR_WIDTH, mask_sh), \
+	IPP_SF(CUR_SIZE, CURSOR_HEIGHT, mask_sh), \
+	IPP_SF(CUR_SURFACE_ADDRESS_HIGH, CURSOR_SURFACE_ADDRESS_HIGH, mask_sh), \
+	IPP_SF(CUR_SURFACE_ADDRESS, CURSOR_SURFACE_ADDRESS, mask_sh), \
+	IPP_SF(PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_BYPASS, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_R, GRPH_PRESCALE_SCALE_R, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_R, GRPH_PRESCALE_BIAS_R, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_G, GRPH_PRESCALE_SCALE_G, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_G, GRPH_PRESCALE_BIAS_G, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_B, GRPH_PRESCALE_SCALE_B, mask_sh), \
+	IPP_SF(PRESCALE_VALUES_GRPH_B, GRPH_PRESCALE_BIAS_B, mask_sh), \
+	IPP_SF(INPUT_GAMMA_CONTROL, GRPH_INPUT_GAMMA_MODE, mask_sh), \
+	IPP_SF(DC_LUT_WRITE_EN_MASK, DC_LUT_WRITE_EN_MASK, mask_sh), \
+	IPP_SF(DC_LUT_RW_MODE, DC_LUT_RW_MODE, mask_sh), \
+	IPP_SF(DC_LUT_CONTROL, DC_LUT_DATA_R_FORMAT, mask_sh), \
+	IPP_SF(DC_LUT_CONTROL, DC_LUT_DATA_G_FORMAT, mask_sh), \
+	IPP_SF(DC_LUT_CONTROL, DC_LUT_DATA_B_FORMAT, mask_sh), \
+	IPP_SF(DC_LUT_RW_INDEX, DC_LUT_RW_INDEX, mask_sh), \
+	IPP_SF(DC_LUT_SEQ_COLOR, DC_LUT_SEQ_COLOR, mask_sh), \
+	IPP_SF(DEGAMMA_CONTROL, GRPH_DEGAMMA_MODE, mask_sh), \
+	IPP_SF(DEGAMMA_CONTROL, CURSOR_DEGAMMA_MODE, mask_sh)
+#endif
+
 #define IPP_REG_FIELD_LIST(type) \
 	type CURSOR_UPDATE_LOCK; \
 	type CURSOR_EN; \
@@ -233,6 +273,15 @@ void dce_ipp_construct(struct dce_ipp *ipp_dce,
 	const struct dce_ipp_shift *ipp_shift,
 	const struct dce_ipp_mask *ipp_mask);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_ipp_construct(struct dce_ipp *ipp_dce,
+	struct dc_context *ctx,
+	int inst,
+	const struct dce_ipp_registers *regs,
+	const struct dce_ipp_shift *ipp_shift,
+	const struct dce_ipp_mask *ipp_mask);
+#endif
+
 void dce_ipp_destroy(struct input_pixel_processor **ipp);
 
 #endif /* _DCE_IPP_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
index 8d8c84c81b34..b409f6b2bfd8 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
@@ -425,6 +425,59 @@ static void set_dp_phy_pattern_hbr2_compliance_cp2520_2(
 	enable_phy_bypass_mode(enc110, false);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_set_dp_phy_pattern_hbr2_compliance_cp2520_2(
+	struct dce110_link_encoder *enc110,
+	unsigned int cp2520_pattern)
+{
+
+	/* previously there is a register DP_HBR2_EYE_PATTERN
+	 * that is enabled to get the pattern.
+	 * But it does not work with the latest spec change,
+	 * so we are programming the following registers manually.
+	 *
+	 * The following settings have been confirmed
+	 * by Nick Chorney and Sandra Liu */
+
+	/* Disable PHY Bypass mode to setup the test pattern */
+
+	enable_phy_bypass_mode(enc110, false);
+
+	/* Setup DIG encoder in DP SST mode */
+	enc110->base.funcs->setup(&enc110->base, SIGNAL_TYPE_DISPLAY_PORT);
+
+	/* ensure normal panel mode. */
+	setup_panel_mode(enc110, DP_PANEL_MODE_DEFAULT);
+
+	/* no vbid after BS (SR)
+	 * DP_LINK_FRAMING_CNTL changed history Sandra Liu
+	 * 11000260 / 11000104 / 110000FC */
+	REG_UPDATE_3(DP_LINK_FRAMING_CNTL,
+			DP_IDLE_BS_INTERVAL, 0xFC,
+			DP_VBID_DISABLE, 1,
+			DP_VID_ENHANCED_FRAME_MODE, 1);
+
+	/* DCE6 has no DP_DPHY_SCRAM_CNTL register, skip swap BS with SR */
+
+	/* select cp2520 patterns */
+	if (REG(DP_DPHY_HBR2_PATTERN_CONTROL))
+		REG_UPDATE(DP_DPHY_HBR2_PATTERN_CONTROL,
+				DP_DPHY_HBR2_PATTERN_CONTROL, cp2520_pattern);
+	else
+		/* pre-DCE11 can only generate CP2520 pattern 2 */
+		ASSERT(cp2520_pattern == 2);
+
+	/* set link training complete */
+	set_link_training_complete(enc110, true);
+
+	/* disable video stream */
+	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, 0);
+
+	/* Disable PHY Bypass mode to setup the test pattern */
+	enable_phy_bypass_mode(enc110, false);
+}
+#endif
+
 static void set_dp_phy_pattern_passthrough_mode(
 	struct dce110_link_encoder *enc110,
 	enum dp_panel_mode panel_mode)
@@ -452,6 +505,35 @@ static void set_dp_phy_pattern_passthrough_mode(
 	disable_prbs_mode(enc110);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_set_dp_phy_pattern_passthrough_mode(
+	struct dce110_link_encoder *enc110,
+	enum dp_panel_mode panel_mode)
+{
+	/* program correct panel mode */
+	setup_panel_mode(enc110, panel_mode);
+
+	/* restore LINK_FRAMING_CNTL
+	 * in case we were doing HBR2 compliance pattern before
+	 */
+	REG_UPDATE_3(DP_LINK_FRAMING_CNTL,
+			DP_IDLE_BS_INTERVAL, 0x2000,
+			DP_VBID_DISABLE, 0,
+			DP_VID_ENHANCED_FRAME_MODE, 1);
+
+	/* DCE6 has no DP_DPHY_SCRAM_CNTL register, skip DPHY_SCRAMBLER_BS_COUNT restore */
+
+	/* set link training complete */
+	set_link_training_complete(enc110, true);
+
+	/* Disable PHY Bypass mode to setup the test pattern */
+	enable_phy_bypass_mode(enc110, false);
+
+	/* Disable PRBS mode */
+	disable_prbs_mode(enc110);
+}
+#endif
+
 /* return value is bit-vector */
 static uint8_t get_frontend_source(
 	enum engine_id engine)
@@ -490,6 +572,20 @@ static void configure_encoder(
 	REG_UPDATE(DP_DPHY_SCRAM_CNTL, DPHY_SCRAMBLER_ADVANCE, 1);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_configure_encoder(
+	struct dce110_link_encoder *enc110,
+	const struct dc_link_settings *link_settings)
+{
+	/* set number of lanes */
+
+	REG_SET(DP_CONFIG, 0,
+			DP_UDI_LANES, link_settings->lane_count - LANE_COUNT_ONE);
+
+	/* DCE6 has no DP_DPHY_SCRAM_CNTL register, skip setup scrambler */
+}
+#endif
+
 static void aux_initialize(
 	struct dce110_link_encoder *enc110)
 {
@@ -1059,6 +1155,87 @@ void dce110_link_encoder_enable_dp_mst_output(
 		BREAK_TO_DEBUGGER();
 	}
 }
+
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/* enables DP PHY output */
+void dce60_link_encoder_enable_dp_output(
+	struct link_encoder *enc,
+	const struct dc_link_settings *link_settings,
+	enum clock_source_id clock_source)
+{
+	struct dce110_link_encoder *enc110 = TO_DCE110_LINK_ENC(enc);
+	struct bp_transmitter_control cntl = { 0 };
+	enum bp_result result;
+
+	/* Enable the PHY */
+
+	/* number_of_lanes is used for pixel clock adjust,
+	 * but it's not passed to asic_control.
+	 * We need to set number of lanes manually.
+	 */
+	dce60_configure_encoder(enc110, link_settings);
+	cntl.connector_obj_id = enc110->base.connector;
+	cntl.action = TRANSMITTER_CONTROL_ENABLE;
+	cntl.engine_id = enc->preferred_engine;
+	cntl.transmitter = enc110->base.transmitter;
+	cntl.pll_id = clock_source;
+	cntl.signal = SIGNAL_TYPE_DISPLAY_PORT;
+	cntl.lanes_number = link_settings->lane_count;
+	cntl.hpd_sel = enc110->base.hpd_source;
+	cntl.pixel_clock = link_settings->link_rate
+						* LINK_RATE_REF_FREQ_IN_KHZ;
+	/* TODO: check if undefined works */
+	cntl.color_depth = COLOR_DEPTH_UNDEFINED;
+
+	result = link_transmitter_control(enc110, &cntl);
+
+	if (result != BP_RESULT_OK) {
+		DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n",
+			__func__);
+		BREAK_TO_DEBUGGER();
+	}
+}
+
+/* enables DP PHY output in MST mode */
+void dce60_link_encoder_enable_dp_mst_output(
+	struct link_encoder *enc,
+	const struct dc_link_settings *link_settings,
+	enum clock_source_id clock_source)
+{
+	struct dce110_link_encoder *enc110 = TO_DCE110_LINK_ENC(enc);
+	struct bp_transmitter_control cntl = { 0 };
+	enum bp_result result;
+
+	/* Enable the PHY */
+
+	/* number_of_lanes is used for pixel clock adjust,
+	 * but it's not passed to asic_control.
+	 * We need to set number of lanes manually.
+	 */
+	dce60_configure_encoder(enc110, link_settings);
+
+	cntl.action = TRANSMITTER_CONTROL_ENABLE;
+	cntl.engine_id = ENGINE_ID_UNKNOWN;
+	cntl.transmitter = enc110->base.transmitter;
+	cntl.pll_id = clock_source;
+	cntl.signal = SIGNAL_TYPE_DISPLAY_PORT_MST;
+	cntl.lanes_number = link_settings->lane_count;
+	cntl.hpd_sel = enc110->base.hpd_source;
+	cntl.pixel_clock = link_settings->link_rate
+						* LINK_RATE_REF_FREQ_IN_KHZ;
+	/* TODO: check if undefined works */
+	cntl.color_depth = COLOR_DEPTH_UNDEFINED;
+
+	result = link_transmitter_control(enc110, &cntl);
+
+	if (result != BP_RESULT_OK) {
+		DC_LOG_ERROR("%s: Failed to execute VBIOS command table!\n",
+			__func__);
+		BREAK_TO_DEBUGGER();
+	}
+}
+#endif
+
 /*
  * @brief
  * Disable transmitter and its encoder
@@ -1208,6 +1385,63 @@ void dce110_link_encoder_dp_set_phy_pattern(
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/* set DP PHY test and training patterns */
+void dce60_link_encoder_dp_set_phy_pattern(
+	struct link_encoder *enc,
+	const struct encoder_set_dp_phy_pattern_param *param)
+{
+	struct dce110_link_encoder *enc110 = TO_DCE110_LINK_ENC(enc);
+
+	switch (param->dp_phy_pattern) {
+	case DP_TEST_PATTERN_TRAINING_PATTERN1:
+		dce110_link_encoder_set_dp_phy_pattern_training_pattern(enc, 0);
+		break;
+	case DP_TEST_PATTERN_TRAINING_PATTERN2:
+		dce110_link_encoder_set_dp_phy_pattern_training_pattern(enc, 1);
+		break;
+	case DP_TEST_PATTERN_TRAINING_PATTERN3:
+		dce110_link_encoder_set_dp_phy_pattern_training_pattern(enc, 2);
+		break;
+	case DP_TEST_PATTERN_TRAINING_PATTERN4:
+		dce110_link_encoder_set_dp_phy_pattern_training_pattern(enc, 3);
+		break;
+	case DP_TEST_PATTERN_D102:
+		set_dp_phy_pattern_d102(enc110);
+		break;
+	case DP_TEST_PATTERN_SYMBOL_ERROR:
+		set_dp_phy_pattern_symbol_error(enc110);
+		break;
+	case DP_TEST_PATTERN_PRBS7:
+		set_dp_phy_pattern_prbs7(enc110);
+		break;
+	case DP_TEST_PATTERN_80BIT_CUSTOM:
+		set_dp_phy_pattern_80bit_custom(
+			enc110, param->custom_pattern);
+		break;
+	case DP_TEST_PATTERN_CP2520_1:
+		dce60_set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc110, 1);
+		break;
+	case DP_TEST_PATTERN_CP2520_2:
+		dce60_set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc110, 2);
+		break;
+	case DP_TEST_PATTERN_CP2520_3:
+		dce60_set_dp_phy_pattern_hbr2_compliance_cp2520_2(enc110, 3);
+		break;
+	case DP_TEST_PATTERN_VIDEO_MODE: {
+		dce60_set_dp_phy_pattern_passthrough_mode(
+			enc110, param->dp_panel_mode);
+		break;
+	}
+
+	default:
+		/* invalid phy pattern */
+		ASSERT_CRITICAL(false);
+		break;
+	}
+}
+#endif
+
 static void fill_stream_allocation_row_info(
 	const struct link_mst_stream_allocation *stream_allocation,
 	uint32_t *src,
@@ -1407,3 +1641,138 @@ void dce110_link_encoder_get_max_link_cap(struct link_encoder *enc,
 
 	*link_settings = max_link_cap;
 }
+
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct link_encoder_funcs dce60_lnk_enc_funcs = {
+	.validate_output_with_stream =
+		dce110_link_encoder_validate_output_with_stream,
+	.hw_init = dce110_link_encoder_hw_init,
+	.setup = dce110_link_encoder_setup,
+	.enable_tmds_output = dce110_link_encoder_enable_tmds_output,
+	.enable_dp_output = dce60_link_encoder_enable_dp_output,
+	.enable_dp_mst_output = dce60_link_encoder_enable_dp_mst_output,
+	.enable_lvds_output = dce110_link_encoder_enable_lvds_output,
+	.disable_output = dce110_link_encoder_disable_output,
+	.dp_set_lane_settings = dce110_link_encoder_dp_set_lane_settings,
+	.dp_set_phy_pattern = dce60_link_encoder_dp_set_phy_pattern,
+	.update_mst_stream_allocation_table =
+		dce110_link_encoder_update_mst_stream_allocation_table,
+	.psr_program_dp_dphy_fast_training =
+			dce110_psr_program_dp_dphy_fast_training,
+	.psr_program_secondary_packet = dce110_psr_program_secondary_packet,
+	.connect_dig_be_to_fe = dce110_link_encoder_connect_dig_be_to_fe,
+	.enable_hpd = dce110_link_encoder_enable_hpd,
+	.disable_hpd = dce110_link_encoder_disable_hpd,
+	.is_dig_enabled = dce110_is_dig_enabled,
+	.destroy = dce110_link_encoder_destroy,
+	.get_max_link_cap = dce110_link_encoder_get_max_link_cap
+};
+
+void dce60_link_encoder_construct(
+	struct dce110_link_encoder *enc110,
+	const struct encoder_init_data *init_data,
+	const struct encoder_feature_support *enc_features,
+	const struct dce110_link_enc_registers *link_regs,
+	const struct dce110_link_enc_aux_registers *aux_regs,
+	const struct dce110_link_enc_hpd_registers *hpd_regs)
+{
+	struct bp_encoder_cap_info bp_cap_info = {0};
+	const struct dc_vbios_funcs *bp_funcs = init_data->ctx->dc_bios->funcs;
+	enum bp_result result = BP_RESULT_OK;
+
+	enc110->base.funcs = &dce60_lnk_enc_funcs;
+	enc110->base.ctx = init_data->ctx;
+	enc110->base.id = init_data->encoder;
+
+	enc110->base.hpd_source = init_data->hpd_source;
+	enc110->base.connector = init_data->connector;
+
+	enc110->base.preferred_engine = ENGINE_ID_UNKNOWN;
+
+	enc110->base.features = *enc_features;
+
+	enc110->base.transmitter = init_data->transmitter;
+
+	/* set the flag to indicate whether driver poll the I2C data pin
+	 * while doing the DP sink detect
+	 */
+
+/*	if (dal_adapter_service_is_feature_supported(as,
+		FEATURE_DP_SINK_DETECT_POLL_DATA_PIN))
+		enc110->base.features.flags.bits.
+			DP_SINK_DETECT_POLL_DATA_PIN = true;*/
+
+	enc110->base.output_signals =
+		SIGNAL_TYPE_DVI_SINGLE_LINK |
+		SIGNAL_TYPE_DVI_DUAL_LINK |
+		SIGNAL_TYPE_LVDS |
+		SIGNAL_TYPE_DISPLAY_PORT |
+		SIGNAL_TYPE_DISPLAY_PORT_MST |
+		SIGNAL_TYPE_EDP |
+		SIGNAL_TYPE_HDMI_TYPE_A;
+
+	/* For DCE 8.0 and 8.1, by design, UNIPHY is hardwired to DIG_BE.
+	 * SW always assign DIG_FE 1:1 mapped to DIG_FE for non-MST UNIPHY.
+	 * SW assign DIG_FE to non-MST UNIPHY first and MST last. So prefer
+	 * DIG is per UNIPHY and used by SST DP, eDP, HDMI, DVI and LVDS.
+	 * Prefer DIG assignment is decided by board design.
+	 * For DCE 8.0, there are only max 6 UNIPHYs, we assume board design
+	 * and VBIOS will filter out 7 UNIPHY for DCE 8.0.
+	 * By this, adding DIGG should not hurt DCE 8.0.
+	 * This will let DCE 8.1 share DCE 8.0 as much as possible
+	 */
+
+	enc110->link_regs = link_regs;
+	enc110->aux_regs = aux_regs;
+	enc110->hpd_regs = hpd_regs;
+
+	switch (enc110->base.transmitter) {
+	case TRANSMITTER_UNIPHY_A:
+		enc110->base.preferred_engine = ENGINE_ID_DIGA;
+	break;
+	case TRANSMITTER_UNIPHY_B:
+		enc110->base.preferred_engine = ENGINE_ID_DIGB;
+	break;
+	case TRANSMITTER_UNIPHY_C:
+		enc110->base.preferred_engine = ENGINE_ID_DIGC;
+	break;
+	case TRANSMITTER_UNIPHY_D:
+		enc110->base.preferred_engine = ENGINE_ID_DIGD;
+	break;
+	case TRANSMITTER_UNIPHY_E:
+		enc110->base.preferred_engine = ENGINE_ID_DIGE;
+	break;
+	case TRANSMITTER_UNIPHY_F:
+		enc110->base.preferred_engine = ENGINE_ID_DIGF;
+	break;
+	case TRANSMITTER_UNIPHY_G:
+		enc110->base.preferred_engine = ENGINE_ID_DIGG;
+	break;
+	default:
+		ASSERT_CRITICAL(false);
+		enc110->base.preferred_engine = ENGINE_ID_UNKNOWN;
+	}
+
+	/* default to one to mirror Windows behavior */
+	enc110->base.features.flags.bits.HDMI_6GB_EN = 1;
+
+	result = bp_funcs->get_encoder_cap_info(enc110->base.ctx->dc_bios,
+						enc110->base.id, &bp_cap_info);
+
+	/* Override features with DCE-specific values */
+	if (BP_RESULT_OK == result) {
+		enc110->base.features.flags.bits.IS_HBR2_CAPABLE =
+				bp_cap_info.DP_HBR2_EN;
+		enc110->base.features.flags.bits.IS_HBR3_CAPABLE =
+				bp_cap_info.DP_HBR3_EN;
+		enc110->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN;
+	} else {
+		DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n",
+				__func__,
+				result);
+	}
+	if (enc110->base.ctx->dc->debug.hdmi20_disable) {
+		enc110->base.features.flags.bits.HDMI_6GB_EN = 0;
+	}
+}
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h
index 66027d496778..cb714a48b171 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h
@@ -76,6 +76,34 @@
 	SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \
 	SR(DCI_MEM_PWR_STATUS)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define LE_DCE60_REG_LIST(id)\
+	SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \
+	SR(DMCU_RAM_ACCESS_CTRL), \
+	SR(DMCU_IRAM_RD_CTRL), \
+	SR(DMCU_IRAM_RD_DATA), \
+	SR(DMCU_INTERRUPT_TO_UC_EN_MASK), \
+	SRI(DIG_BE_CNTL, DIG, id), \
+	SRI(DIG_BE_EN_CNTL, DIG, id), \
+	SRI(DP_CONFIG, DP, id), \
+	SRI(DP_DPHY_CNTL, DP, id), \
+	SRI(DP_DPHY_PRBS_CNTL, DP, id), \
+	SRI(DP_DPHY_SYM0, DP, id), \
+	SRI(DP_DPHY_SYM1, DP, id), \
+	SRI(DP_DPHY_SYM2, DP, id), \
+	SRI(DP_DPHY_TRAINING_PATTERN_SEL, DP, id), \
+	SRI(DP_LINK_CNTL, DP, id), \
+	SRI(DP_LINK_FRAMING_CNTL, DP, id), \
+	SRI(DP_MSE_SAT0, DP, id), \
+	SRI(DP_MSE_SAT1, DP, id), \
+	SRI(DP_MSE_SAT2, DP, id), \
+	SRI(DP_MSE_SAT_UPDATE, DP, id), \
+	SRI(DP_SEC_CNTL, DP, id), \
+	SRI(DP_VID_STREAM_CNTL, DP, id), \
+	SRI(DP_DPHY_FAST_TRAINING, DP, id), \
+	SRI(DP_SEC_CNTL1, DP, id)
+#endif
+
 #define LE_DCE80_REG_LIST(id)\
 	SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \
 	LE_COMMON_REG_LIST_BASE(id)
@@ -171,6 +199,16 @@ void dce110_link_encoder_construct(
 	const struct dce110_link_enc_aux_registers *aux_regs,
 	const struct dce110_link_enc_hpd_registers *hpd_regs);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_link_encoder_construct(
+	struct dce110_link_encoder *enc110,
+	const struct encoder_init_data *init_data,
+	const struct encoder_feature_support *enc_features,
+	const struct dce110_link_enc_registers *link_regs,
+	const struct dce110_link_enc_aux_registers *aux_regs,
+	const struct dce110_link_enc_hpd_registers *hpd_regs);
+#endif
+
 bool dce110_link_encoder_validate_dvi_output(
 	const struct dce110_link_encoder *enc110,
 	enum signal_type connector_signal,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
index 51481e922eb9..79a6f261a0da 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c
@@ -174,6 +174,22 @@ static void program_urgency_watermark(
 		URGENCY_HIGH_WATERMARK, urgency_high_wm);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_program_urgency_watermark(
+	struct dce_mem_input *dce_mi,
+	uint32_t wm_select,
+	uint32_t urgency_low_wm,
+	uint32_t urgency_high_wm)
+{
+	REG_UPDATE(DPG_PIPE_ARBITRATION_CONTROL3,
+		URGENCY_WATERMARK_MASK, wm_select);
+
+	REG_SET_2(DPG_PIPE_URGENCY_CONTROL, 0,
+		URGENCY_LOW_WATERMARK, urgency_low_wm,
+		URGENCY_HIGH_WATERMARK, urgency_high_wm);
+}
+#endif
+
 static void dce120_program_urgency_watermark(
 	struct dce_mem_input *dce_mi,
 	uint32_t wm_select,
@@ -193,6 +209,25 @@ static void dce120_program_urgency_watermark(
 
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_program_nbp_watermark(
+	struct dce_mem_input *dce_mi,
+	uint32_t wm_select,
+	uint32_t nbp_wm)
+{
+	REG_UPDATE(DPG_PIPE_NB_PSTATE_CHANGE_CONTROL,
+		NB_PSTATE_CHANGE_WATERMARK_MASK, wm_select);
+
+	REG_UPDATE_3(DPG_PIPE_NB_PSTATE_CHANGE_CONTROL,
+		NB_PSTATE_CHANGE_ENABLE, 1,
+		NB_PSTATE_CHANGE_URGENT_DURING_REQUEST, 1,
+		NB_PSTATE_CHANGE_NOT_SELF_REFRESH_DURING_REQUEST, 1);
+
+	REG_UPDATE(DPG_PIPE_NB_PSTATE_CHANGE_CONTROL,
+		NB_PSTATE_CHANGE_WATERMARK, nbp_wm);
+}
+#endif
+
 static void program_nbp_watermark(
 	struct dce_mem_input *dce_mi,
 	uint32_t wm_select,
@@ -225,6 +260,20 @@ static void program_nbp_watermark(
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_program_stutter_watermark(
+	struct dce_mem_input *dce_mi,
+	uint32_t wm_select,
+	uint32_t stutter_mark)
+{
+	REG_UPDATE(DPG_PIPE_STUTTER_CONTROL,
+		STUTTER_EXIT_SELF_REFRESH_WATERMARK_MASK, wm_select);
+
+	REG_UPDATE(DPG_PIPE_STUTTER_CONTROL,
+		STUTTER_EXIT_SELF_REFRESH_WATERMARK, stutter_mark);
+}
+#endif
+
 static void dce120_program_stutter_watermark(
 	struct dce_mem_input *dce_mi,
 	uint32_t wm_select,
@@ -286,6 +335,34 @@ static void dce_mi_program_display_marks(
 	program_stutter_watermark(dce_mi, 1, stutter_exit.d_mark); /* set d */
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_mi_program_display_marks(
+	struct mem_input *mi,
+	struct dce_watermarks nbp,
+	struct dce_watermarks stutter_exit,
+	struct dce_watermarks stutter_enter,
+	struct dce_watermarks urgent,
+	uint32_t total_dest_line_time_ns)
+{
+	struct dce_mem_input *dce_mi = TO_DCE_MEM_INPUT(mi);
+	uint32_t stutter_en = mi->ctx->dc->debug.disable_stutter ? 0 : 1;
+
+	dce60_program_urgency_watermark(dce_mi, 2, /* set a */
+			urgent.a_mark, total_dest_line_time_ns);
+	dce60_program_urgency_watermark(dce_mi, 1, /* set d */
+			urgent.d_mark, total_dest_line_time_ns);
+
+	REG_UPDATE_2(DPG_PIPE_STUTTER_CONTROL,
+		STUTTER_ENABLE, stutter_en,
+		STUTTER_IGNORE_FBC, 1);
+	dce60_program_nbp_watermark(dce_mi, 2, nbp.a_mark); /* set a */
+	dce60_program_nbp_watermark(dce_mi, 1, nbp.d_mark); /* set d */
+
+	dce60_program_stutter_watermark(dce_mi, 2, stutter_exit.a_mark); /* set a */
+	dce60_program_stutter_watermark(dce_mi, 1, stutter_exit.d_mark); /* set d */
+}
+#endif
+
 static void dce112_mi_program_display_marks(struct mem_input *mi,
 	struct dce_watermarks nbp,
 	struct dce_watermarks stutter_exit,
@@ -369,7 +446,7 @@ static void program_tiling(
 		 */
 	}
 
-	if (dce_mi->masks->GRPH_ARRAY_MODE) { /* GFX8 */
+	if (dce_mi->masks->GRPH_MICRO_TILE_MODE) { /* GFX8 */
 		REG_UPDATE_9(GRPH_CONTROL,
 				GRPH_NUM_BANKS, info->gfx8.num_banks,
 				GRPH_BANK_WIDTH, info->gfx8.bank_width,
@@ -385,6 +462,23 @@ static void program_tiling(
 				GRPH_Z, 0);
 				*/
 	}
+
+	if (dce_mi->masks->GRPH_ARRAY_MODE) { /* GFX6 but reuses gfx8 struct */
+		REG_UPDATE_8(GRPH_CONTROL,
+				GRPH_NUM_BANKS, info->gfx8.num_banks,
+				GRPH_BANK_WIDTH, info->gfx8.bank_width,
+				GRPH_BANK_HEIGHT, info->gfx8.bank_height,
+				GRPH_MACRO_TILE_ASPECT, info->gfx8.tile_aspect,
+				GRPH_TILE_SPLIT, info->gfx8.tile_split,
+				/* DCE6 has no GRPH_MICRO_TILE_MODE mask */
+				GRPH_PIPE_CONFIG, info->gfx8.pipe_config,
+				GRPH_ARRAY_MODE, info->gfx8.array_mode,
+				GRPH_COLOR_EXPANSION_MODE, 1);
+		/* 01 - DCP_GRPH_COLOR_EXPANSION_MODE_ZEXP: zero expansion for YCbCr */
+		/*
+				GRPH_Z, 0);
+				*/
+	}
 }
 
 
@@ -429,6 +523,36 @@ static void program_size_and_rotation(
 			GRPH_ROTATION_ANGLE, rotation_angles[rotation]);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_program_size(
+	struct dce_mem_input *dce_mi,
+	enum dc_rotation_angle rotation, /* not used in DCE6 */
+	const struct plane_size *plane_size)
+{
+	struct rect hw_rect = plane_size->surface_size;
+	/* DCE6 has no HW rotation, skip rotation_angles declaration */
+
+	/* DCE6 has no HW rotation, skip ROTATION_ANGLE_* processing */
+
+	REG_SET(GRPH_X_START, 0,
+			GRPH_X_START, hw_rect.x);
+
+	REG_SET(GRPH_Y_START, 0,
+			GRPH_Y_START, hw_rect.y);
+
+	REG_SET(GRPH_X_END, 0,
+			GRPH_X_END, hw_rect.width);
+
+	REG_SET(GRPH_Y_END, 0,
+			GRPH_Y_END, hw_rect.height);
+
+	REG_SET(GRPH_PITCH, 0,
+			GRPH_PITCH, plane_size->surface_pitch);
+
+	/* DCE6 has no HW_ROTATION register, skip setting rotation_angles */
+}
+#endif
+
 static void program_grph_pixel_format(
 	struct dce_mem_input *dce_mi,
 	enum surface_pixel_format format)
@@ -521,6 +645,28 @@ static void dce_mi_program_surface_config(
 		program_grph_pixel_format(dce_mi, format);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_mi_program_surface_config(
+	struct mem_input *mi,
+	enum surface_pixel_format format,
+	union dc_tiling_info *tiling_info,
+	struct plane_size *plane_size,
+	enum dc_rotation_angle rotation, /* not used in DCE6 */
+	struct dc_plane_dcc_param *dcc,
+	bool horizontal_mirror)
+{
+	struct dce_mem_input *dce_mi = TO_DCE_MEM_INPUT(mi);
+	REG_UPDATE(GRPH_ENABLE, GRPH_ENABLE, 1);
+
+	program_tiling(dce_mi, tiling_info);
+	dce60_program_size(dce_mi, rotation, plane_size);
+
+	if (format >= SURFACE_PIXEL_FORMAT_GRPH_BEGIN &&
+		format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
+		program_grph_pixel_format(dce_mi, format);
+}
+#endif
+
 static uint32_t get_dmif_switch_time_us(
 	uint32_t h_total,
 	uint32_t v_total,
@@ -741,6 +887,20 @@ static const struct mem_input_funcs dce_mi_funcs = {
 	.mem_input_is_flip_pending = dce_mi_is_flip_pending
 };
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct mem_input_funcs dce60_mi_funcs = {
+	.mem_input_program_display_marks = dce60_mi_program_display_marks,
+	.allocate_mem_input = dce_mi_allocate_dmif,
+	.free_mem_input = dce_mi_free_dmif,
+	.mem_input_program_surface_flip_and_addr =
+			dce_mi_program_surface_flip_and_addr,
+	.mem_input_program_pte_vm = dce_mi_program_pte_vm,
+	.mem_input_program_surface_config =
+			dce60_mi_program_surface_config,
+	.mem_input_is_flip_pending = dce_mi_is_flip_pending
+};
+#endif
+
 static const struct mem_input_funcs dce112_mi_funcs = {
 	.mem_input_program_display_marks = dce112_mi_program_display_marks,
 	.allocate_mem_input = dce_mi_allocate_dmif,
@@ -783,6 +943,20 @@ void dce_mem_input_construct(
 	dce_mi->masks = mi_mask;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_mem_input_construct(
+	struct dce_mem_input *dce_mi,
+	struct dc_context *ctx,
+	int inst,
+	const struct dce_mem_input_registers *regs,
+	const struct dce_mem_input_shift *mi_shift,
+	const struct dce_mem_input_mask *mi_mask)
+{
+	dce_mem_input_construct(dce_mi, ctx, inst, regs, mi_shift, mi_mask);
+	dce_mi->base.funcs = &dce60_mi_funcs;
+}
+#endif
+
 void dce112_mem_input_construct(
 	struct dce_mem_input *dce_mi,
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.h b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.h
index d15b0d7f47fc..23db5c72f07e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.h
@@ -58,6 +58,31 @@
 	SRI(DVMM_PTE_CONTROL, DCP, id),\
 	SRI(DVMM_PTE_ARB_CONTROL, DCP, id)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define MI_DCE6_REG_LIST(id)\
+	SRI(GRPH_ENABLE, DCP, id),\
+	SRI(GRPH_CONTROL, DCP, id),\
+	SRI(GRPH_X_START, DCP, id),\
+	SRI(GRPH_Y_START, DCP, id),\
+	SRI(GRPH_X_END, DCP, id),\
+	SRI(GRPH_Y_END, DCP, id),\
+	SRI(GRPH_PITCH, DCP, id),\
+	SRI(GRPH_SWAP_CNTL, DCP, id),\
+	SRI(PRESCALE_GRPH_CONTROL, DCP, id),\
+	SRI(GRPH_UPDATE, DCP, id),\
+	SRI(GRPH_FLIP_CONTROL, DCP, id),\
+	SRI(GRPH_PRIMARY_SURFACE_ADDRESS, DCP, id),\
+	SRI(GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, DCP, id),\
+	SRI(GRPH_SECONDARY_SURFACE_ADDRESS, DCP, id),\
+	SRI(GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, DCP, id),\
+	SRI(DPG_PIPE_ARBITRATION_CONTROL1, DMIF_PG, id),\
+	SRI(DPG_PIPE_ARBITRATION_CONTROL3, DMIF_PG, id),\
+	SRI(DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, DMIF_PG, id),\
+	SRI(DPG_PIPE_URGENCY_CONTROL, DMIF_PG, id),\
+	SRI(DPG_PIPE_STUTTER_CONTROL, DMIF_PG, id),\
+	SRI(DMIF_BUFFER_CONTROL, PIPE, id)
+#endif
+
 #define MI_DCE8_REG_LIST(id)\
 	MI_DCE_BASE_REG_LIST(id),\
 	SRI(DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, DMIF_PG, id)
@@ -104,6 +129,9 @@ struct dce_mem_input_registers {
 	uint32_t GRPH_SECONDARY_SURFACE_ADDRESS_HIGH;
 	/* DMIF_PG */
 	uint32_t DPG_PIPE_ARBITRATION_CONTROL1;
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	uint32_t DPG_PIPE_ARBITRATION_CONTROL3;
+#endif
 	uint32_t DPG_WATERMARK_MASK_CONTROL;
 	uint32_t DPG_PIPE_URGENCY_CONTROL;
 	uint32_t DPG_PIPE_URGENT_LEVEL_CONTROL;
@@ -126,6 +154,18 @@ struct dce_mem_input_registers {
 #define SFB(blk_name, reg_name, field_name, post_fix)\
 	.field_name = blk_name ## reg_name ## __ ## field_name ## post_fix
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define MI_GFX6_TILE_MASK_SH_LIST(mask_sh, blk)\
+	SFB(blk, GRPH_CONTROL, GRPH_NUM_BANKS, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_BANK_WIDTH, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_BANK_HEIGHT, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_MACRO_TILE_ASPECT, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_TILE_SPLIT, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_PIPE_CONFIG, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_ARRAY_MODE, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_COLOR_EXPANSION_MODE, mask_sh)
+#endif
+
 #define MI_GFX8_TILE_MASK_SH_LIST(mask_sh, blk)\
 	SFB(blk, GRPH_CONTROL, GRPH_NUM_BANKS, mask_sh),\
 	SFB(blk, GRPH_CONTROL, GRPH_BANK_WIDTH, mask_sh),\
@@ -162,6 +202,32 @@ struct dce_mem_input_registers {
 	SFB(blk, GRPH_UPDATE, GRPH_UPDATE_LOCK, mask_sh),\
 	SFB(blk, GRPH_FLIP_CONTROL, GRPH_SURFACE_UPDATE_H_RETRACE_EN, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define MI_DCP_MASK_SH_LIST_DCE6(mask_sh, blk)\
+	SFB(blk, GRPH_ENABLE, GRPH_ENABLE, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_DEPTH, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_FORMAT, mask_sh),\
+	SFB(blk, GRPH_CONTROL, GRPH_NUM_BANKS, mask_sh),\
+	SFB(blk, GRPH_X_START, GRPH_X_START, mask_sh),\
+	SFB(blk, GRPH_Y_START, GRPH_Y_START, mask_sh),\
+	SFB(blk, GRPH_X_END, GRPH_X_END, mask_sh),\
+	SFB(blk, GRPH_Y_END, GRPH_Y_END, mask_sh),\
+	SFB(blk, GRPH_PITCH, GRPH_PITCH, mask_sh),\
+	SFB(blk, GRPH_SWAP_CNTL, GRPH_RED_CROSSBAR, mask_sh),\
+	SFB(blk, GRPH_SWAP_CNTL, GRPH_BLUE_CROSSBAR, mask_sh),\
+	SFB(blk, PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_SELECT, mask_sh),\
+	SFB(blk, PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_R_SIGN, mask_sh),\
+	SFB(blk, PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_G_SIGN, mask_sh),\
+	SFB(blk, PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_B_SIGN, mask_sh),\
+	SFB(blk, GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, mask_sh),\
+	SFB(blk, GRPH_SECONDARY_SURFACE_ADDRESS, GRPH_SECONDARY_SURFACE_ADDRESS, mask_sh),\
+	SFB(blk, GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, mask_sh),\
+	SFB(blk, GRPH_PRIMARY_SURFACE_ADDRESS, GRPH_PRIMARY_SURFACE_ADDRESS, mask_sh),\
+	SFB(blk, GRPH_UPDATE, GRPH_SURFACE_UPDATE_PENDING, mask_sh),\
+	SFB(blk, GRPH_UPDATE, GRPH_UPDATE_LOCK, mask_sh),\
+	SFB(blk, GRPH_FLIP_CONTROL, GRPH_SURFACE_UPDATE_H_RETRACE_EN, mask_sh)
+#endif
+
 #define MI_DCP_DCE11_MASK_SH_LIST(mask_sh, blk)\
 	SFB(blk, GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, mask_sh)
 
@@ -172,6 +238,33 @@ struct dce_mem_input_registers {
 	SFB(blk, DVMM_PTE_ARB_CONTROL, DVMM_PTE_REQ_PER_CHUNK, mask_sh),\
 	SFB(blk, DVMM_PTE_ARB_CONTROL, DVMM_MAX_PTE_REQ_OUTSTANDING, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define MI_DMIF_PG_MASK_SH_LIST_DCE6(mask_sh, blk)\
+	SFB(blk, DPG_PIPE_ARBITRATION_CONTROL1, PIXEL_DURATION, mask_sh),\
+	SFB(blk, DPG_PIPE_URGENCY_CONTROL, URGENCY_LOW_WATERMARK, mask_sh),\
+	SFB(blk, DPG_PIPE_URGENCY_CONTROL, URGENCY_HIGH_WATERMARK, mask_sh),\
+	SFB(blk, DPG_PIPE_STUTTER_CONTROL, STUTTER_ENABLE, mask_sh),\
+	SFB(blk, DPG_PIPE_STUTTER_CONTROL, STUTTER_IGNORE_FBC, mask_sh),\
+	SF(PIPE0_DMIF_BUFFER_CONTROL, DMIF_BUFFERS_ALLOCATED, mask_sh),\
+	SF(PIPE0_DMIF_BUFFER_CONTROL, DMIF_BUFFERS_ALLOCATION_COMPLETED, mask_sh)
+
+#define MI_DMIF_PG_MASK_SH_DCE6(mask_sh, blk)\
+	SFB(blk, DPG_PIPE_ARBITRATION_CONTROL3, URGENCY_WATERMARK_MASK, mask_sh),\
+	SFB(blk, DPG_PIPE_STUTTER_CONTROL, STUTTER_EXIT_SELF_REFRESH_WATERMARK_MASK, mask_sh),\
+	SFB(blk, DPG_PIPE_STUTTER_CONTROL, STUTTER_EXIT_SELF_REFRESH_WATERMARK, mask_sh),\
+	SFB(blk, DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, NB_PSTATE_CHANGE_WATERMARK_MASK, mask_sh),\
+	SFB(blk, DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, NB_PSTATE_CHANGE_ENABLE, mask_sh),\
+	SFB(blk, DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, NB_PSTATE_CHANGE_URGENT_DURING_REQUEST, mask_sh),\
+	SFB(blk, DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, NB_PSTATE_CHANGE_NOT_SELF_REFRESH_DURING_REQUEST, mask_sh),\
+	SFB(blk, DPG_PIPE_NB_PSTATE_CHANGE_CONTROL, NB_PSTATE_CHANGE_WATERMARK, mask_sh)
+
+#define MI_DCE6_MASK_SH_LIST(mask_sh)\
+	MI_DCP_MASK_SH_LIST_DCE6(mask_sh, ),\
+	MI_DMIF_PG_MASK_SH_LIST_DCE6(mask_sh, ),\
+	MI_DMIF_PG_MASK_SH_DCE6(mask_sh, ),\
+	MI_GFX6_TILE_MASK_SH_LIST(mask_sh, )
+#endif
+
 #define MI_DMIF_PG_MASK_SH_LIST(mask_sh, blk)\
 	SFB(blk, DPG_PIPE_ARBITRATION_CONTROL1, PIXEL_DURATION, mask_sh),\
 	SFB(blk, DPG_WATERMARK_MASK_CONTROL, URGENCY_WATERMARK_MASK, mask_sh),\
@@ -345,6 +438,16 @@ void dce_mem_input_construct(
 	const struct dce_mem_input_shift *mi_shift,
 	const struct dce_mem_input_mask *mi_mask);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_mem_input_construct(
+	struct dce_mem_input *dce_mi,
+	struct dc_context *ctx,
+	int inst,
+	const struct dce_mem_input_registers *regs,
+	const struct dce_mem_input_shift *mi_shift,
+	const struct dce_mem_input_mask *mi_mask);
+#endif
+
 void dce112_mem_input_construct(
 	struct dce_mem_input *dce_mi,
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_opp.c b/drivers/gpu/drm/amd/display/dc/dce/dce_opp.c
index 51081d9ae3fb..e459ae65aaf7 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_opp.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_opp.c
@@ -141,6 +141,47 @@ static void set_truncation(
 				params->flags.TRUNCATE_MODE);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/**
+ *	dce60_set_truncation
+ *	1) set truncation depth: 0 for 18 bpp or 1 for 24 bpp
+ *	2) enable truncation
+ *	3) HW remove 12bit FMT support for DCE11 power saving reason.
+ */
+static void dce60_set_truncation(
+		struct dce110_opp *opp110,
+		const struct bit_depth_reduction_params *params)
+{
+	/* DCE6 has no FMT_TRUNCATE_MODE bit in FMT_BIT_DEPTH_CONTROL reg */
+
+	/*Disable truncation*/
+	REG_UPDATE_2(FMT_BIT_DEPTH_CONTROL,
+			FMT_TRUNCATE_EN, 0,
+			FMT_TRUNCATE_DEPTH, 0);
+
+	if (params->pixel_encoding == PIXEL_ENCODING_YCBCR422) {
+		/*  8bpc trunc on YCbCr422*/
+		if (params->flags.TRUNCATE_DEPTH == 1)
+			REG_UPDATE_2(FMT_BIT_DEPTH_CONTROL,
+					FMT_TRUNCATE_EN, 1,
+					FMT_TRUNCATE_DEPTH, 1);
+		else if (params->flags.TRUNCATE_DEPTH == 2)
+			/*  10bpc trunc on YCbCr422*/
+			REG_UPDATE_2(FMT_BIT_DEPTH_CONTROL,
+					FMT_TRUNCATE_EN, 1,
+					FMT_TRUNCATE_DEPTH, 2);
+		return;
+	}
+	/* on other format-to do */
+	if (params->flags.TRUNCATE_ENABLED == 0)
+		return;
+	/*Set truncation depth and Enable truncation*/
+	REG_UPDATE_2(FMT_BIT_DEPTH_CONTROL,
+				FMT_TRUNCATE_EN, 1,
+				FMT_TRUNCATE_DEPTH,
+				params->flags.TRUNCATE_DEPTH);
+}
+#endif
 
 /**
  *	set_spatial_dither
@@ -373,6 +414,57 @@ void dce110_opp_set_clamping(
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/**
+ *	Set Clamping for DCE6 parts
+ *	1) Set clamping format based on bpc - 0 for 6bpc (No clamping)
+ *		1 for 8 bpc
+ *		2 for 10 bpc
+ *		3 for 12 bpc
+ *		7 for programable
+ *	2) Enable clamp if Limited range requested
+ */
+void dce60_opp_set_clamping(
+	struct dce110_opp *opp110,
+	const struct clamping_and_pixel_encoding_params *params)
+{
+	REG_SET_2(FMT_CLAMP_CNTL, 0,
+		FMT_CLAMP_DATA_EN, 0,
+		FMT_CLAMP_COLOR_FORMAT, 0);
+
+	switch (params->clamping_level) {
+	case CLAMPING_FULL_RANGE:
+		break;
+	case CLAMPING_LIMITED_RANGE_8BPC:
+		REG_SET_2(FMT_CLAMP_CNTL, 0,
+			FMT_CLAMP_DATA_EN, 1,
+			FMT_CLAMP_COLOR_FORMAT, 1);
+		break;
+	case CLAMPING_LIMITED_RANGE_10BPC:
+		REG_SET_2(FMT_CLAMP_CNTL, 0,
+			FMT_CLAMP_DATA_EN, 1,
+			FMT_CLAMP_COLOR_FORMAT, 2);
+		break;
+	case CLAMPING_LIMITED_RANGE_12BPC:
+		REG_SET_2(FMT_CLAMP_CNTL, 0,
+			FMT_CLAMP_DATA_EN, 1,
+			FMT_CLAMP_COLOR_FORMAT, 3);
+		break;
+	case CLAMPING_LIMITED_RANGE_PROGRAMMABLE:
+		/*Set clamp control*/
+		REG_SET_2(FMT_CLAMP_CNTL, 0,
+			FMT_CLAMP_DATA_EN, 1,
+			FMT_CLAMP_COLOR_FORMAT, 7);
+
+		/* DCE6 does have FMT_CLAMP_COMPONENT_{R,G,B} registers */
+
+		break;
+	default:
+		break;
+	}
+}
+#endif
+
 /**
  *	set_pixel_encoding
  *
@@ -408,6 +500,39 @@ static void set_pixel_encoding(
 
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/**
+ *	dce60_set_pixel_encoding
+ *	DCE6 has no FMT_SUBSAMPLING_{MODE,ORDER} bits in FMT_CONTROL reg
+ *	Set Pixel Encoding
+ *		0: RGB 4:4:4 or YCbCr 4:4:4 or YOnly
+ *		1: YCbCr 4:2:2
+ */
+static void dce60_set_pixel_encoding(
+	struct dce110_opp *opp110,
+	const struct clamping_and_pixel_encoding_params *params)
+{
+	if (opp110->opp_mask->FMT_CBCR_BIT_REDUCTION_BYPASS)
+		REG_UPDATE_2(FMT_CONTROL,
+				FMT_PIXEL_ENCODING, 0,
+				FMT_CBCR_BIT_REDUCTION_BYPASS, 0);
+	else
+		REG_UPDATE(FMT_CONTROL,
+				FMT_PIXEL_ENCODING, 0);
+
+	if (params->pixel_encoding == PIXEL_ENCODING_YCBCR422) {
+		REG_UPDATE(FMT_CONTROL,
+				FMT_PIXEL_ENCODING, 1);
+	}
+	if (params->pixel_encoding == PIXEL_ENCODING_YCBCR420) {
+		REG_UPDATE_2(FMT_CONTROL,
+				FMT_PIXEL_ENCODING, 2,
+				FMT_CBCR_BIT_REDUCTION_BYPASS, 1);
+	}
+
+}
+#endif
+
 void dce110_opp_program_bit_depth_reduction(
 	struct output_pixel_processor *opp,
 	const struct bit_depth_reduction_params *params)
@@ -419,6 +544,19 @@ void dce110_opp_program_bit_depth_reduction(
 	set_temporal_dither(opp110, params);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_opp_program_bit_depth_reduction(
+	struct output_pixel_processor *opp,
+	const struct bit_depth_reduction_params *params)
+{
+	struct dce110_opp *opp110 = TO_DCE110_OPP(opp);
+
+	dce60_set_truncation(opp110, params);
+	set_spatial_dither(opp110, params);
+	set_temporal_dither(opp110, params);
+}
+#endif
+
 void dce110_opp_program_clamping_and_pixel_encoding(
 	struct output_pixel_processor *opp,
 	const struct clamping_and_pixel_encoding_params *params)
@@ -429,6 +567,19 @@ void dce110_opp_program_clamping_and_pixel_encoding(
 	set_pixel_encoding(opp110, params);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_opp_program_clamping_and_pixel_encoding(
+	struct output_pixel_processor *opp,
+	const struct clamping_and_pixel_encoding_params *params)
+{
+	struct dce110_opp *opp110 = TO_DCE110_OPP(opp);
+
+	dce60_opp_set_clamping(opp110, params);
+	dce60_set_pixel_encoding(opp110, params);
+}
+#endif
+
+
 static void program_formatter_420_memory(struct output_pixel_processor *opp)
 {
 	struct dce110_opp *opp110 = TO_DCE110_OPP(opp);
@@ -526,7 +677,32 @@ void dce110_opp_program_fmt(
 	return;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_opp_program_fmt(
+	struct output_pixel_processor *opp,
+	struct bit_depth_reduction_params *fmt_bit_depth,
+	struct clamping_and_pixel_encoding_params *clamping)
+{
+	/* dithering is affected by <CrtcSourceSelect>, hence should be
+	 * programmed afterwards */
+
+	if (clamping->pixel_encoding == PIXEL_ENCODING_YCBCR420)
+		program_formatter_420_memory(opp);
+
+	dce60_opp_program_bit_depth_reduction(
+		opp,
+		fmt_bit_depth);
+
+	dce60_opp_program_clamping_and_pixel_encoding(
+		opp,
+		clamping);
+
+	if (clamping->pixel_encoding == PIXEL_ENCODING_YCBCR420)
+		program_formatter_reset_dig_resync_fifo(opp);
 
+	return;
+}
+#endif
 
 
 
@@ -541,6 +717,15 @@ static const struct opp_funcs funcs = {
 	.opp_program_bit_depth_reduction = dce110_opp_program_bit_depth_reduction
 };
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct opp_funcs dce60_opp_funcs = {
+	.opp_set_dyn_expansion = dce110_opp_set_dyn_expansion,
+	.opp_destroy = dce110_opp_destroy,
+	.opp_program_fmt = dce60_opp_program_fmt,
+	.opp_program_bit_depth_reduction = dce60_opp_program_bit_depth_reduction
+};
+#endif
+
 void dce110_opp_construct(struct dce110_opp *opp110,
 	struct dc_context *ctx,
 	uint32_t inst,
@@ -559,6 +744,26 @@ void dce110_opp_construct(struct dce110_opp *opp110,
 	opp110->opp_mask = opp_mask;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_opp_construct(struct dce110_opp *opp110,
+	struct dc_context *ctx,
+	uint32_t inst,
+	const struct dce_opp_registers *regs,
+	const struct dce_opp_shift *opp_shift,
+	const struct dce_opp_mask *opp_mask)
+{
+	opp110->base.funcs = &dce60_opp_funcs;
+
+	opp110->base.ctx = ctx;
+
+	opp110->base.inst = inst;
+
+	opp110->regs = regs;
+	opp110->opp_shift = opp_shift;
+	opp110->opp_mask = opp_mask;
+}
+#endif
+
 void dce110_opp_destroy(struct output_pixel_processor **opp)
 {
 	if (*opp)
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_opp.h b/drivers/gpu/drm/amd/display/dc/dce/dce_opp.h
index 2ab0147cbd9d..4d484ef60f35 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_opp.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_opp.h
@@ -81,6 +81,17 @@ enum dce110_opp_reg_type {
 	OPP_COMMON_REG_LIST_BASE(id), \
 	SRI(CONTROL, FMT_MEMORY, id)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define OPP_DCE_60_REG_LIST(id) \
+	SRI(FMT_DYNAMIC_EXP_CNTL, FMT, id), \
+	SRI(FMT_BIT_DEPTH_CONTROL, FMT, id), \
+	SRI(FMT_CONTROL, FMT, id), \
+	SRI(FMT_DITHER_RAND_R_SEED, FMT, id), \
+	SRI(FMT_DITHER_RAND_G_SEED, FMT, id), \
+	SRI(FMT_DITHER_RAND_B_SEED, FMT, id), \
+	SRI(FMT_CLAMP_CNTL, FMT, id)
+#endif
+
 #define OPP_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
@@ -192,6 +203,35 @@ enum dce110_opp_reg_type {
 	OPP_SF(FMT0_FMT_CONTROL, FMT_SUBSAMPLING_ORDER, mask_sh),\
 	OPP_SF(FMT0_FMT_CONTROL, FMT_CBCR_BIT_REDUCTION_BYPASS, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define OPP_COMMON_MASK_SH_LIST_DCE_60(mask_sh)\
+	OPP_SF(FMT_DYNAMIC_EXP_CNTL, FMT_DYNAMIC_EXP_EN, mask_sh),\
+	OPP_SF(FMT_DYNAMIC_EXP_CNTL, FMT_DYNAMIC_EXP_MODE, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_EN, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_DEPTH, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_EN, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_DEPTH, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_MODE, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_HIGHPASS_RANDOM_ENABLE, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_FRAME_RANDOM_ENABLE, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_RGB_RANDOM_ENABLE, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_DITHER_EN, mask_sh),\
+	OPP_SF(FMT_DITHER_RAND_R_SEED, FMT_RAND_R_SEED, mask_sh),\
+	OPP_SF(FMT_DITHER_RAND_G_SEED, FMT_RAND_G_SEED, mask_sh),\
+	OPP_SF(FMT_DITHER_RAND_B_SEED, FMT_RAND_B_SEED, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_DITHER_EN, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_DITHER_RESET, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_DITHER_OFFSET, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_DITHER_DEPTH, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_TEMPORAL_LEVEL, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_25FRC_SEL, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_50FRC_SEL, mask_sh),\
+	OPP_SF(FMT_BIT_DEPTH_CONTROL, FMT_75FRC_SEL, mask_sh),\
+	OPP_SF(FMT_CLAMP_CNTL, FMT_CLAMP_DATA_EN, mask_sh),\
+	OPP_SF(FMT_CLAMP_CNTL, FMT_CLAMP_COLOR_FORMAT, mask_sh),\
+	OPP_SF(FMT_CONTROL, FMT_PIXEL_ENCODING, mask_sh)
+#endif
+
 #define OPP_REG_FIELD_LIST(type) \
 	type FMT_DYNAMIC_EXP_EN; \
 	type FMT_DYNAMIC_EXP_MODE; \
@@ -279,6 +319,15 @@ void dce110_opp_construct(struct dce110_opp *opp110,
 	const struct dce_opp_shift *opp_shift,
 	const struct dce_opp_mask *opp_mask);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_opp_construct(struct dce110_opp *opp110,
+	struct dc_context *ctx,
+	uint32_t inst,
+	const struct dce_opp_registers *regs,
+	const struct dce_opp_shift *opp_shift,
+	const struct dce_opp_mask *opp_mask);
+#endif
+
 void dce110_opp_destroy(struct output_pixel_processor **opp);
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c
index 43781e77be43..74f7619d4154 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c
@@ -46,13 +46,14 @@
 #define FN(reg_name, field_name) \
 	dce_panel_cntl->shift->field_name, dce_panel_cntl->mask->field_name
 
-static unsigned int calculate_16_bit_backlight_from_pwm(struct dce_panel_cntl *dce_panel_cntl)
+static unsigned int dce_get_16_bit_backlight_from_pwm(struct panel_cntl *panel_cntl)
 {
 	uint64_t current_backlight;
 	uint32_t round_result;
 	uint32_t pwm_period_cntl, bl_period, bl_int_count;
 	uint32_t bl_pwm_cntl, bl_pwm, fractional_duty_cycle_en;
 	uint32_t bl_period_mask, bl_pwm_mask;
+	struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl);
 
 	pwm_period_cntl = REG_READ(BL_PWM_PERIOD_CNTL);
 	REG_GET(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, &bl_period);
@@ -75,7 +76,7 @@ static unsigned int calculate_16_bit_backlight_from_pwm(struct dce_panel_cntl *d
 	else
 		bl_pwm &= 0xFFFF;
 
-	current_backlight = bl_pwm << (1 + bl_int_count);
+	current_backlight = (uint64_t)bl_pwm << (1 + bl_int_count);
 
 	if (bl_period == 0)
 		bl_period = 0xFFFF;
@@ -150,7 +151,7 @@ static uint32_t dce_panel_cntl_hw_init(struct panel_cntl *panel_cntl)
 	REG_UPDATE(BL_PWM_GRP1_REG_LOCK,
 			BL_PWM_GRP1_REG_LOCK, 0);
 
-	current_backlight = calculate_16_bit_backlight_from_pwm(dce_panel_cntl);
+	current_backlight = dce_get_16_bit_backlight_from_pwm(panel_cntl);
 
 	return current_backlight;
 }
@@ -158,11 +159,15 @@ static uint32_t dce_panel_cntl_hw_init(struct panel_cntl *panel_cntl)
 static bool dce_is_panel_backlight_on(struct panel_cntl *panel_cntl)
 {
 	struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl);
-	uint32_t value;
+	uint32_t blon, blon_ovrd, pwrseq_target_state;
 
-	REG_GET(PWRSEQ_CNTL, LVTMA_BLON, &value);
+	REG_GET_2(PWRSEQ_CNTL, LVTMA_BLON, &blon, LVTMA_BLON_OVRD, &blon_ovrd);
+	REG_GET(PWRSEQ_CNTL, LVTMA_PWRSEQ_TARGET_STATE, &pwrseq_target_state);
 
-	return value;
+	if (blon_ovrd)
+		return blon;
+	else
+		return pwrseq_target_state;
 }
 
 static bool dce_is_panel_powered_on(struct panel_cntl *panel_cntl)
@@ -273,6 +278,7 @@ static const struct panel_cntl_funcs dce_link_panel_cntl_funcs = {
 	.is_panel_powered_on = dce_is_panel_powered_on,
 	.store_backlight_level = dce_store_backlight_level,
 	.driver_set_backlight = dce_driver_set_backlight,
+	.get_current_backlight = dce_get_16_bit_backlight_from_pwm,
 };
 
 void dce_panel_cntl_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h
index 99c68ca9c7e0..6bd1196083a3 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h
@@ -54,15 +54,17 @@
 	SR(BL_PWM_CNTL2), \
 	SR(BL_PWM_PERIOD_CNTL), \
 	SR(BL_PWM_GRP1_REG_LOCK), \
-	SR(BIOS_SCRATCH_2)
+	NBIO_SR(BIOS_SCRATCH_2)
 
 #define DCE_PANEL_CNTL_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
 #define DCE_PANEL_CNTL_MASK_SH_LIST(mask_sh) \
 	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\
+	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_BLON_OVRD, mask_sh),\
 	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\
 	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\
+	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_PWRSEQ_TARGET_STATE, mask_sh), \
 	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh), \
 	DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_REF_DIV, BL_PWM_REF_DIV, mask_sh), \
 	DCE_PANEL_CNTL_SF(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, mask_sh), \
@@ -76,8 +78,10 @@
 
 #define DCE_PANEL_CNTL_REG_FIELD_LIST(type) \
 	type LVTMA_BLON;\
+	type LVTMA_BLON_OVRD;\
 	type LVTMA_DIGON;\
 	type LVTMA_DIGON_OVRD;\
+	type LVTMA_PWRSEQ_TARGET_STATE; \
 	type LVTMA_PWRSEQ_TARGET_STATE_R; \
 	type BL_PWM_REF_DIV; \
 	type BL_PWM_EN; \
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
index 4cdaaf4d881c..5054bb567b74 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
@@ -710,7 +710,7 @@ static void dce110_stream_encoder_lvds_set_stream_attribute(
 	ASSERT(crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB);
 }
 
-static void dce110_stream_encoder_set_mst_bandwidth(
+static void dce110_stream_encoder_set_throttled_vcp_size(
 	struct stream_encoder *enc,
 	struct fixed31_32 avg_time_slots_per_mtp)
 {
@@ -1621,8 +1621,8 @@ static const struct stream_encoder_funcs dce110_str_enc_funcs = {
 		dce110_stream_encoder_dvi_set_stream_attribute,
 	.lvds_set_stream_attribute =
 		dce110_stream_encoder_lvds_set_stream_attribute,
-	.set_mst_bandwidth =
-		dce110_stream_encoder_set_mst_bandwidth,
+	.set_throttled_vcp_size =
+		dce110_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =
 		dce110_stream_encoder_update_hdmi_info_packets,
 	.stop_hdmi_info_packets =
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c
index ab63d0d0304c..2a32b66959ba 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c
@@ -146,6 +146,33 @@ static bool setup_scaling_configuration(
 	return true;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static bool dce60_setup_scaling_configuration(
+	struct dce_transform *xfm_dce,
+	const struct scaler_data *data)
+{
+	REG_SET(SCL_BYPASS_CONTROL, 0, SCL_BYPASS_MODE, 0);
+
+	if (data->taps.h_taps + data->taps.v_taps <= 2) {
+		/* Set bypass */
+
+		/* DCE6 has no SCL_MODE register, skip scale mode programming */
+
+		return false;
+	}
+
+	REG_SET_2(SCL_TAP_CONTROL, 0,
+			SCL_H_NUM_OF_TAPS, data->taps.h_taps - 1,
+			SCL_V_NUM_OF_TAPS, data->taps.v_taps - 1);
+
+	/* DCE6 has no SCL_MODE register, skip scale mode programming */
+
+	/* DCE6 has no SCL_BOUNDARY_MODE bit, skip replace out of bound pixels */
+
+	return true;
+}
+#endif
+
 static void program_overscan(
 		struct dce_transform *xfm_dce,
 		const struct scaler_data *data)
@@ -279,6 +306,36 @@ static void calculate_inits(
 	inits->v_init.fraction = dc_fixpt_u0d19(v_init) << 5;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_calculate_inits(
+	struct dce_transform *xfm_dce,
+	const struct scaler_data *data,
+	struct sclh_ratios_inits *inits)
+{
+	struct fixed31_32 v_init;
+
+	inits->h_int_scale_ratio =
+		dc_fixpt_u2d19(data->ratios.horz) << 5;
+	inits->v_int_scale_ratio =
+		dc_fixpt_u2d19(data->ratios.vert) << 5;
+
+	/* DCE6 h_init_luma setting inspired by DCE110 */
+	inits->h_init_luma.integer = 1;
+
+	/* DCE6 h_init_chroma setting inspired by DCE110 */
+	inits->h_init_chroma.integer = 1;
+
+	v_init =
+		dc_fixpt_div_int(
+			dc_fixpt_add(
+				data->ratios.vert,
+				dc_fixpt_from_int(data->taps.v_taps + 1)),
+				2);
+	inits->v_init.integer = dc_fixpt_floor(v_init);
+	inits->v_init.fraction = dc_fixpt_u0d19(v_init) << 5;
+}
+#endif
+
 static void program_scl_ratios_inits(
 	struct dce_transform *xfm_dce,
 	struct scl_ratios_inits *inits)
@@ -301,6 +358,36 @@ static void program_scl_ratios_inits(
 	REG_WRITE(SCL_AUTOMATIC_MODE_CONTROL, 0);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_program_scl_ratios_inits(
+	struct dce_transform *xfm_dce,
+	struct sclh_ratios_inits *inits)
+{
+
+	REG_SET(SCL_HORZ_FILTER_SCALE_RATIO, 0,
+			SCL_H_SCALE_RATIO, inits->h_int_scale_ratio);
+
+	REG_SET(SCL_VERT_FILTER_SCALE_RATIO, 0,
+			SCL_V_SCALE_RATIO, inits->v_int_scale_ratio);
+
+	/* DCE6 has SCL_HORZ_FILTER_INIT_RGB_LUMA register */
+	REG_SET_2(SCL_HORZ_FILTER_INIT_RGB_LUMA, 0,
+			SCL_H_INIT_INT_RGB_Y, inits->h_init_luma.integer,
+			SCL_H_INIT_FRAC_RGB_Y, inits->h_init_luma.fraction);
+
+	/* DCE6 has SCL_HORZ_FILTER_INIT_CHROMA register */
+	REG_SET_2(SCL_HORZ_FILTER_INIT_CHROMA, 0,
+			SCL_H_INIT_INT_CBCR, inits->h_init_chroma.integer,
+			SCL_H_INIT_FRAC_CBCR, inits->h_init_chroma.fraction);
+
+	REG_SET_2(SCL_VERT_FILTER_INIT, 0,
+			SCL_V_INIT_INT, inits->v_init.integer,
+			SCL_V_INIT_FRAC, inits->v_init.fraction);
+
+	REG_WRITE(SCL_AUTOMATIC_MODE_CONTROL, 0);
+}
+#endif
+
 static const uint16_t *get_filter_coeffs_16p(int taps, struct fixed31_32 ratio)
 {
 	if (taps == 4)
@@ -399,6 +486,91 @@ static void dce_transform_set_scaler(
 	REG_UPDATE(LB_DATA_FORMAT, ALPHA_EN, data->lb_params.alpha_en);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_transform_set_scaler(
+	struct transform *xfm,
+	const struct scaler_data *data)
+{
+	struct dce_transform *xfm_dce = TO_DCE_TRANSFORM(xfm);
+	bool is_scaling_required;
+	bool filter_updated = false;
+	const uint16_t *coeffs_v, *coeffs_h;
+
+	/*Use whole line buffer memory always*/
+	REG_SET(DC_LB_MEMORY_SPLIT, 0,
+		DC_LB_MEMORY_CONFIG, 0);
+
+	REG_SET(DC_LB_MEM_SIZE, 0,
+		DC_LB_MEM_SIZE, xfm_dce->lb_memory_size);
+
+	/* Clear SCL_F_SHARP_CONTROL value to 0 */
+	REG_WRITE(SCL_F_SHARP_CONTROL, 0);
+
+	/* 1. Program overscan */
+	program_overscan(xfm_dce, data);
+
+	/* 2. Program taps and configuration */
+	is_scaling_required = dce60_setup_scaling_configuration(xfm_dce, data);
+
+	if (is_scaling_required) {
+		/* 3. Calculate and program ratio, DCE6 filter initialization */
+		struct sclh_ratios_inits inits = { 0 };
+
+		/* DCE6 has specific calculate_inits() function */
+		dce60_calculate_inits(xfm_dce, data, &inits);
+
+		/* DCE6 has specific program_scl_ratios_inits() function */
+		dce60_program_scl_ratios_inits(xfm_dce, &inits);
+
+		coeffs_v = get_filter_coeffs_16p(data->taps.v_taps, data->ratios.vert);
+		coeffs_h = get_filter_coeffs_16p(data->taps.h_taps, data->ratios.horz);
+
+		if (coeffs_v != xfm_dce->filter_v || coeffs_h != xfm_dce->filter_h) {
+			/* 4. Program vertical filters */
+			if (xfm_dce->filter_v == NULL)
+				REG_SET(SCL_VERT_FILTER_CONTROL, 0,
+						SCL_V_2TAP_HARDCODE_COEF_EN, 0);
+			program_multi_taps_filter(
+					xfm_dce,
+					data->taps.v_taps,
+					coeffs_v,
+					FILTER_TYPE_RGB_Y_VERTICAL);
+			program_multi_taps_filter(
+					xfm_dce,
+					data->taps.v_taps,
+					coeffs_v,
+					FILTER_TYPE_ALPHA_VERTICAL);
+
+			/* 5. Program horizontal filters */
+			if (xfm_dce->filter_h == NULL)
+				REG_SET(SCL_HORZ_FILTER_CONTROL, 0,
+						SCL_H_2TAP_HARDCODE_COEF_EN, 0);
+			program_multi_taps_filter(
+					xfm_dce,
+					data->taps.h_taps,
+					coeffs_h,
+					FILTER_TYPE_RGB_Y_HORIZONTAL);
+			program_multi_taps_filter(
+					xfm_dce,
+					data->taps.h_taps,
+					coeffs_h,
+					FILTER_TYPE_ALPHA_HORIZONTAL);
+
+			xfm_dce->filter_v = coeffs_v;
+			xfm_dce->filter_h = coeffs_h;
+			filter_updated = true;
+		}
+	}
+
+	/* 6. Program the viewport */
+	program_viewport(xfm_dce, &data->viewport);
+
+	/* DCE6 has no SCL_COEF_UPDATE_COMPLETE bit to flip to new coefficient memory */
+
+	/* DCE6 DATA_FORMAT register does not support ALPHA_EN */
+}
+#endif
+
 /*****************************************************************************
  * set_clamp
  *
@@ -664,6 +836,67 @@ static void program_bit_depth_reduction(
 		   bit_depth_params->flags.HIGHPASS_RANDOM);
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+/*****************************************************************************
+ * dce60_transform_bit_depth_reduction program
+ *
+ * @brief
+ *     Programs the DCP bit depth reduction registers (Clamp, Round/Truncate,
+ *      Dither) for dce
+ *
+ * @param depth : bit depth to set the clamp to (should match denorm)
+ *
+ ******************************************************************************/
+static void dce60_program_bit_depth_reduction(
+	struct dce_transform *xfm_dce,
+	enum dc_color_depth depth,
+	const struct bit_depth_reduction_params *bit_depth_params)
+{
+	enum dcp_out_trunc_round_depth trunc_round_depth;
+	enum dcp_out_trunc_round_mode trunc_mode;
+	bool spatial_dither_enable;
+
+	ASSERT(depth < COLOR_DEPTH_121212); /* Invalid clamp bit depth */
+
+	spatial_dither_enable = bit_depth_params->flags.SPATIAL_DITHER_ENABLED;
+	/* Default to 12 bit truncation without rounding */
+	trunc_round_depth = DCP_OUT_TRUNC_ROUND_DEPTH_12BIT;
+	trunc_mode = DCP_OUT_TRUNC_ROUND_MODE_TRUNCATE;
+
+	if (bit_depth_params->flags.TRUNCATE_ENABLED) {
+		/* Don't enable dithering if truncation is enabled */
+		spatial_dither_enable = false;
+		trunc_mode = bit_depth_params->flags.TRUNCATE_MODE ?
+			     DCP_OUT_TRUNC_ROUND_MODE_ROUND :
+			     DCP_OUT_TRUNC_ROUND_MODE_TRUNCATE;
+
+		if (bit_depth_params->flags.TRUNCATE_DEPTH == 0 ||
+		    bit_depth_params->flags.TRUNCATE_DEPTH == 1)
+			trunc_round_depth = DCP_OUT_TRUNC_ROUND_DEPTH_8BIT;
+		else if (bit_depth_params->flags.TRUNCATE_DEPTH == 2)
+			trunc_round_depth = DCP_OUT_TRUNC_ROUND_DEPTH_10BIT;
+		else {
+			/*
+			 * Invalid truncate/round depth. Setting here to 12bit
+			 * to prevent use-before-initialize errors.
+			 */
+			trunc_round_depth = DCP_OUT_TRUNC_ROUND_DEPTH_12BIT;
+			BREAK_TO_DEBUGGER();
+		}
+	}
+
+	/* DCE6 has no OUT_CLAMP_CONTROL_* registers - set_clamp() is skipped */
+	set_round(xfm_dce, trunc_mode, trunc_round_depth);
+	set_dither(xfm_dce,
+		   spatial_dither_enable,
+		   DCP_SPATIAL_DITHER_MODE_A_AA_A,
+		   DCP_SPATIAL_DITHER_DEPTH_30BPP,
+		   bit_depth_params->flags.FRAME_RANDOM,
+		   bit_depth_params->flags.RGB_RANDOM,
+		   bit_depth_params->flags.HIGHPASS_RANDOM);
+}
+#endif
+
 static int dce_transform_get_max_num_of_supported_lines(
 	struct dce_transform *xfm_dce,
 	enum lb_pixel_depth depth,
@@ -797,6 +1030,59 @@ static void dce_transform_set_pixel_storage_depth(
 	}
 }
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static void dce60_transform_set_pixel_storage_depth(
+	struct transform *xfm,
+	enum lb_pixel_depth depth,
+	const struct bit_depth_reduction_params *bit_depth_params)
+{
+	struct dce_transform *xfm_dce = TO_DCE_TRANSFORM(xfm);
+	int pixel_depth, expan_mode;
+	enum dc_color_depth color_depth;
+
+	switch (depth) {
+	case LB_PIXEL_DEPTH_18BPP:
+		color_depth = COLOR_DEPTH_666;
+		pixel_depth = 2;
+		expan_mode  = 1;
+		break;
+	case LB_PIXEL_DEPTH_24BPP:
+		color_depth = COLOR_DEPTH_888;
+		pixel_depth = 1;
+		expan_mode  = 1;
+		break;
+	case LB_PIXEL_DEPTH_30BPP:
+		color_depth = COLOR_DEPTH_101010;
+		pixel_depth = 0;
+		expan_mode  = 1;
+		break;
+	case LB_PIXEL_DEPTH_36BPP:
+		color_depth = COLOR_DEPTH_121212;
+		pixel_depth = 3;
+		expan_mode  = 0;
+		break;
+	default:
+		color_depth = COLOR_DEPTH_101010;
+		pixel_depth = 0;
+		expan_mode  = 1;
+		BREAK_TO_DEBUGGER();
+		break;
+	}
+
+	set_denormalization(xfm_dce, color_depth);
+	dce60_program_bit_depth_reduction(xfm_dce, color_depth, bit_depth_params);
+
+	/* DATA_FORMAT in DCE6 does not have PIXEL_DEPTH and PIXEL_EXPAN_MODE masks */
+
+	if (!(xfm_dce->lb_pixel_depth_supported & depth)) {
+		/*we should use unsupported capabilities
+		 *  unless it is required by w/a*/
+		DC_LOG_WARNING("%s: Capability not supported",
+			__func__);
+	}
+}
+#endif
+
 static void program_gamut_remap(
 	struct dce_transform *xfm_dce,
 	const uint16_t *reg_val)
@@ -1335,6 +1621,21 @@ static const struct transform_funcs dce_transform_funcs = {
 	.transform_get_optimal_number_of_taps = dce_transform_get_optimal_number_of_taps
 };
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+static const struct transform_funcs dce60_transform_funcs = {
+	.transform_reset = dce_transform_reset,
+	.transform_set_scaler = dce60_transform_set_scaler,
+	.transform_set_gamut_remap = dce_transform_set_gamut_remap,
+	.opp_set_csc_adjustment = dce110_opp_set_csc_adjustment,
+	.opp_set_csc_default = dce110_opp_set_csc_default,
+	.opp_power_on_regamma_lut = dce110_opp_power_on_regamma_lut,
+	.opp_program_regamma_pwl = dce110_opp_program_regamma_pwl,
+	.opp_set_regamma_mode = dce110_opp_set_regamma_mode,
+	.transform_set_pixel_storage_depth = dce60_transform_set_pixel_storage_depth,
+	.transform_get_optimal_number_of_taps = dce_transform_get_optimal_number_of_taps
+};
+#endif
+
 /*****************************************/
 /* Constructor, Destructor               */
 /*****************************************/
@@ -1365,3 +1666,32 @@ void dce_transform_construct(
 	xfm_dce->lb_bits_per_entry = LB_BITS_PER_ENTRY;
 	xfm_dce->lb_memory_size = LB_TOTAL_NUMBER_OF_ENTRIES; /*0x6B0*/
 }
+
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_transform_construct(
+	struct dce_transform *xfm_dce,
+	struct dc_context *ctx,
+	uint32_t inst,
+	const struct dce_transform_registers *regs,
+	const struct dce_transform_shift *xfm_shift,
+	const struct dce_transform_mask *xfm_mask)
+{
+	xfm_dce->base.ctx = ctx;
+
+	xfm_dce->base.inst = inst;
+	xfm_dce->base.funcs = &dce60_transform_funcs;
+
+	xfm_dce->regs = regs;
+	xfm_dce->xfm_shift = xfm_shift;
+	xfm_dce->xfm_mask = xfm_mask;
+
+	xfm_dce->prescaler_on = true;
+	xfm_dce->lb_pixel_depth_supported =
+			LB_PIXEL_DEPTH_18BPP |
+			LB_PIXEL_DEPTH_24BPP |
+			LB_PIXEL_DEPTH_30BPP;
+
+	xfm_dce->lb_bits_per_entry = LB_BITS_PER_ENTRY;
+	xfm_dce->lb_memory_size = LB_TOTAL_NUMBER_OF_ENTRIES; /*0x6B0*/
+}
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.h b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.h
index 948281d8b6af..cbce194ec7b8 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.h
@@ -108,6 +108,68 @@
 	SRI(DCFE_MEM_PWR_CTRL, DCFE, id), \
 	SRI(DCFE_MEM_PWR_STATUS, DCFE, id)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define XFM_COMMON_REG_LIST_DCE60_BASE(id) \
+	SRI(DATA_FORMAT, LB, id), \
+	SRI(GAMUT_REMAP_CONTROL, DCP, id), \
+	SRI(GAMUT_REMAP_C11_C12, DCP, id), \
+	SRI(GAMUT_REMAP_C13_C14, DCP, id), \
+	SRI(GAMUT_REMAP_C21_C22, DCP, id), \
+	SRI(GAMUT_REMAP_C23_C24, DCP, id), \
+	SRI(GAMUT_REMAP_C31_C32, DCP, id), \
+	SRI(GAMUT_REMAP_C33_C34, DCP, id), \
+	SRI(OUTPUT_CSC_C11_C12, DCP, id), \
+	SRI(OUTPUT_CSC_C13_C14, DCP, id), \
+	SRI(OUTPUT_CSC_C21_C22, DCP, id), \
+	SRI(OUTPUT_CSC_C23_C24, DCP, id), \
+	SRI(OUTPUT_CSC_C31_C32, DCP, id), \
+	SRI(OUTPUT_CSC_C33_C34, DCP, id), \
+	SRI(OUTPUT_CSC_CONTROL, DCP, id), \
+	SRI(REGAMMA_CNTLA_START_CNTL, DCP, id), \
+	SRI(REGAMMA_CNTLA_SLOPE_CNTL, DCP, id), \
+	SRI(REGAMMA_CNTLA_END_CNTL1, DCP, id), \
+	SRI(REGAMMA_CNTLA_END_CNTL2, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_0_1, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_2_3, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_4_5, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_6_7, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_8_9, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_10_11, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_12_13, DCP, id), \
+	SRI(REGAMMA_CNTLA_REGION_14_15, DCP, id), \
+	SRI(REGAMMA_LUT_WRITE_EN_MASK, DCP, id), \
+	SRI(REGAMMA_LUT_INDEX, DCP, id), \
+	SRI(REGAMMA_LUT_DATA, DCP, id), \
+	SRI(REGAMMA_CONTROL, DCP, id), \
+	SRI(DENORM_CONTROL, DCP, id), \
+	SRI(DCP_SPATIAL_DITHER_CNTL, DCP, id), \
+	SRI(OUT_ROUND_CONTROL, DCP, id), \
+	SRI(SCL_TAP_CONTROL, SCL, id), \
+	SRI(SCL_CONTROL, SCL, id), \
+	SRI(SCL_BYPASS_CONTROL, SCL, id), \
+	SRI(EXT_OVERSCAN_LEFT_RIGHT, SCL, id), \
+	SRI(EXT_OVERSCAN_TOP_BOTTOM, SCL, id), \
+	SRI(SCL_VERT_FILTER_CONTROL, SCL, id), \
+	SRI(SCL_HORZ_FILTER_CONTROL, SCL, id), \
+	SRI(SCL_COEF_RAM_SELECT, SCL, id), \
+	SRI(SCL_COEF_RAM_TAP_DATA, SCL, id), \
+	SRI(VIEWPORT_START, SCL, id), \
+	SRI(VIEWPORT_SIZE, SCL, id), \
+	SRI(SCL_HORZ_FILTER_SCALE_RATIO, SCL, id), \
+	SRI(SCL_VERT_FILTER_SCALE_RATIO, SCL, id), \
+	SRI(SCL_VERT_FILTER_INIT, SCL, id), \
+	SRI(SCL_AUTOMATIC_MODE_CONTROL, SCL, id), \
+	SRI(DC_LB_MEMORY_SPLIT, LB, id), \
+	SRI(DC_LB_MEM_SIZE, LB, id), \
+	SRI(DCFE_MEM_LIGHT_SLEEP_CNTL, CRTC, id), \
+	SRI(SCL_UPDATE, SCL, id), \
+	SRI(SCL_F_SHARP_CONTROL, SCL, id)
+
+#define XFM_COMMON_REG_LIST_DCE60(id) \
+	XFM_COMMON_REG_LIST_DCE60_BASE(id), \
+	SRI(DCFE_MEM_LIGHT_SLEEP_CNTL, CRTC, id)
+#endif
+
 #define XFM_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
@@ -204,6 +266,83 @@
 	XFM_SF(DCFE_MEM_PWR_STATUS, DCP_REGAMMA_MEM_PWR_STATE, mask_sh),\
 	XFM_SF(SCL_MODE, SCL_PSCL_EN, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define XFM_COMMON_MASK_SH_LIST_DCE60(mask_sh) \
+	XFM_COMMON_MASK_SH_LIST_DCE60_COMMON_BASE(mask_sh), \
+	OPP_SF(DCFE_MEM_LIGHT_SLEEP_CNTL, REGAMMA_LUT_LIGHT_SLEEP_DIS, mask_sh),\
+	OPP_SF(DCFE_MEM_LIGHT_SLEEP_CNTL, DCP_LUT_LIGHT_SLEEP_DIS, mask_sh),\
+	OPP_SF(DCFE_MEM_LIGHT_SLEEP_CNTL, REGAMMA_LUT_MEM_PWR_STATE, mask_sh)
+
+#define XFM_COMMON_MASK_SH_LIST_DCE60_COMMON_BASE(mask_sh) \
+	XFM_SF(OUT_ROUND_CONTROL, OUT_ROUND_TRUNC_MODE, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_SPATIAL_DITHER_EN, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_SPATIAL_DITHER_MODE, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_SPATIAL_DITHER_DEPTH, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_FRAME_RANDOM_ENABLE, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_RGB_RANDOM_ENABLE, mask_sh), \
+	XFM_SF(DCP_SPATIAL_DITHER_CNTL, DCP_HIGHPASS_RANDOM_ENABLE, mask_sh), \
+	XFM_SF(DENORM_CONTROL, DENORM_MODE, mask_sh), \
+	XFM_SF(DATA_FORMAT, INTERLEAVE_EN, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C11_C12, GAMUT_REMAP_C11, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C11_C12, GAMUT_REMAP_C12, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C13_C14, GAMUT_REMAP_C13, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C13_C14, GAMUT_REMAP_C14, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C21_C22, GAMUT_REMAP_C21, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C21_C22, GAMUT_REMAP_C22, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C23_C24, GAMUT_REMAP_C23, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C23_C24, GAMUT_REMAP_C24, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C31_C32, GAMUT_REMAP_C31, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C31_C32, GAMUT_REMAP_C32, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C33_C34, GAMUT_REMAP_C33, mask_sh), \
+	XFM_SF(GAMUT_REMAP_C33_C34, GAMUT_REMAP_C34, mask_sh), \
+	XFM_SF(GAMUT_REMAP_CONTROL, GRPH_GAMUT_REMAP_MODE, mask_sh), \
+	XFM_SF(OUTPUT_CSC_C11_C12, OUTPUT_CSC_C11, mask_sh),\
+	XFM_SF(OUTPUT_CSC_C11_C12, OUTPUT_CSC_C12, mask_sh),\
+	XFM_SF(OUTPUT_CSC_CONTROL, OUTPUT_CSC_GRPH_MODE, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_START_CNTL, REGAMMA_CNTLA_EXP_REGION_START, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_START_CNTL, REGAMMA_CNTLA_EXP_REGION_START_SEGMENT, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_SLOPE_CNTL, REGAMMA_CNTLA_EXP_REGION_LINEAR_SLOPE, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_END_CNTL1, REGAMMA_CNTLA_EXP_REGION_END, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_END_CNTL2, REGAMMA_CNTLA_EXP_REGION_END_BASE, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_END_CNTL2, REGAMMA_CNTLA_EXP_REGION_END_SLOPE, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_REGION_0_1, REGAMMA_CNTLA_EXP_REGION0_LUT_OFFSET, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_REGION_0_1, REGAMMA_CNTLA_EXP_REGION0_NUM_SEGMENTS, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_REGION_0_1, REGAMMA_CNTLA_EXP_REGION1_LUT_OFFSET, mask_sh),\
+	XFM_SF(REGAMMA_CNTLA_REGION_0_1, REGAMMA_CNTLA_EXP_REGION1_NUM_SEGMENTS, mask_sh),\
+	XFM_SF(REGAMMA_LUT_WRITE_EN_MASK, REGAMMA_LUT_WRITE_EN_MASK, mask_sh),\
+	XFM_SF(REGAMMA_CONTROL, GRPH_REGAMMA_MODE, mask_sh),\
+	XFM_SF(SCL_TAP_CONTROL, SCL_H_NUM_OF_TAPS, mask_sh), \
+	XFM_SF(SCL_TAP_CONTROL, SCL_V_NUM_OF_TAPS, mask_sh), \
+	XFM_SF(SCL_BYPASS_CONTROL, SCL_BYPASS_MODE, mask_sh), \
+	XFM_SF(EXT_OVERSCAN_LEFT_RIGHT, EXT_OVERSCAN_LEFT, mask_sh), \
+	XFM_SF(EXT_OVERSCAN_LEFT_RIGHT, EXT_OVERSCAN_RIGHT, mask_sh), \
+	XFM_SF(EXT_OVERSCAN_TOP_BOTTOM, EXT_OVERSCAN_TOP, mask_sh), \
+	XFM_SF(EXT_OVERSCAN_TOP_BOTTOM, EXT_OVERSCAN_BOTTOM, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_SELECT, SCL_C_RAM_FILTER_TYPE, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_SELECT, SCL_C_RAM_PHASE, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_SELECT, SCL_C_RAM_TAP_PAIR_IDX, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_TAP_DATA, SCL_C_RAM_EVEN_TAP_COEF_EN, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_TAP_DATA, SCL_C_RAM_EVEN_TAP_COEF, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_TAP_DATA, SCL_C_RAM_ODD_TAP_COEF_EN, mask_sh), \
+	XFM_SF(SCL_COEF_RAM_TAP_DATA, SCL_C_RAM_ODD_TAP_COEF, mask_sh), \
+	XFM_SF(VIEWPORT_START, VIEWPORT_X_START, mask_sh), \
+	XFM_SF(VIEWPORT_START, VIEWPORT_Y_START, mask_sh), \
+	XFM_SF(VIEWPORT_SIZE, VIEWPORT_HEIGHT, mask_sh), \
+	XFM_SF(VIEWPORT_SIZE, VIEWPORT_WIDTH, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_SCALE_RATIO, SCL_H_SCALE_RATIO, mask_sh), \
+	XFM_SF(SCL_VERT_FILTER_SCALE_RATIO, SCL_V_SCALE_RATIO, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_INIT_RGB_LUMA, SCL_H_INIT_INT_RGB_Y, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_INIT_RGB_LUMA, SCL_H_INIT_FRAC_RGB_Y, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_INIT_CHROMA, SCL_H_INIT_INT_CBCR, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_INIT_CHROMA, SCL_H_INIT_FRAC_CBCR, mask_sh), \
+	XFM_SF(SCL_VERT_FILTER_INIT, SCL_V_INIT_INT, mask_sh), \
+	XFM_SF(SCL_VERT_FILTER_INIT, SCL_V_INIT_FRAC, mask_sh), \
+	XFM_SF(SCL_HORZ_FILTER_CONTROL, SCL_H_FILTER_PICK_NEAREST, mask_sh), \
+	XFM_SF(SCL_VERT_FILTER_CONTROL, SCL_V_FILTER_PICK_NEAREST, mask_sh), \
+	XFM_SF(DC_LB_MEMORY_SPLIT, DC_LB_MEMORY_CONFIG, mask_sh), \
+	XFM_SF(DC_LB_MEM_SIZE, DC_LB_MEM_SIZE, mask_sh)
+#endif
+
 #define XFM_COMMON_MASK_SH_LIST_SOC_BASE(mask_sh) \
 	XFM_SF(DCP0_OUT_CLAMP_CONTROL_B_CB, OUT_CLAMP_MIN_B_CB, mask_sh), \
 	XFM_SF(DCP0_OUT_CLAMP_CONTROL_B_CB, OUT_CLAMP_MAX_B_CB, mask_sh), \
@@ -302,6 +441,7 @@
 	type DCP_RGB_RANDOM_ENABLE; \
 	type DCP_HIGHPASS_RANDOM_ENABLE; \
 	type DENORM_MODE; \
+	type INTERLEAVE_EN; \
 	type PIXEL_DEPTH; \
 	type PIXEL_EXPAN_MODE; \
 	type GAMUT_REMAP_C11; \
@@ -365,12 +505,20 @@
 	type SCL_V_SCALE_RATIO; \
 	type SCL_H_INIT_INT; \
 	type SCL_H_INIT_FRAC; \
+	type SCL_H_INIT_INT_RGB_Y; \
+	type SCL_H_INIT_FRAC_RGB_Y; \
+	type SCL_H_INIT_INT_CBCR; \
+	type SCL_H_INIT_FRAC_CBCR; \
 	type SCL_V_INIT_INT; \
 	type SCL_V_INIT_FRAC; \
+	type DC_LB_MEMORY_CONFIG; \
+	type DC_LB_MEM_SIZE; \
 	type LB_MEMORY_CONFIG; \
 	type LB_MEMORY_SIZE; \
 	type SCL_V_2TAP_HARDCODE_COEF_EN; \
 	type SCL_H_2TAP_HARDCODE_COEF_EN; \
+	type SCL_V_FILTER_PICK_NEAREST; \
+	type SCL_H_FILTER_PICK_NEAREST; \
 	type SCL_COEF_UPDATE_COMPLETE; \
 	type ALPHA_EN
 
@@ -383,6 +531,9 @@ struct dce_transform_mask {
 };
 
 struct dce_transform_registers {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	uint32_t DATA_FORMAT;
+#endif
 	uint32_t LB_DATA_FORMAT;
 	uint32_t GAMUT_REMAP_CONTROL;
 	uint32_t GAMUT_REMAP_C11_C12;
@@ -438,8 +589,16 @@ struct dce_transform_registers {
 	uint32_t SCL_HORZ_FILTER_SCALE_RATIO;
 	uint32_t SCL_VERT_FILTER_SCALE_RATIO;
 	uint32_t SCL_HORZ_FILTER_INIT;
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	uint32_t SCL_HORZ_FILTER_INIT_RGB_LUMA;
+	uint32_t SCL_HORZ_FILTER_INIT_CHROMA;
+#endif
 	uint32_t SCL_VERT_FILTER_INIT;
 	uint32_t SCL_AUTOMATIC_MODE_CONTROL;
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	uint32_t DC_LB_MEMORY_SPLIT;
+	uint32_t DC_LB_MEM_SIZE;
+#endif
 	uint32_t LB_MEMORY_CTRL;
 	uint32_t SCL_UPDATE;
 	uint32_t SCL_F_SHARP_CONTROL;
@@ -457,6 +616,16 @@ struct scl_ratios_inits {
 	struct init_int_and_frac v_init;
 };
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+struct sclh_ratios_inits {
+	uint32_t h_int_scale_ratio;
+	uint32_t v_int_scale_ratio;
+	struct init_int_and_frac h_init_luma;
+	struct init_int_and_frac h_init_chroma;
+	struct init_int_and_frac v_init;
+};
+#endif
+
 enum ram_filter_type {
 	FILTER_TYPE_RGB_Y_VERTICAL	= 0, /* 0 - RGB/Y Vertical filter */
 	FILTER_TYPE_CBCR_VERTICAL	= 1, /* 1 - CbCr  Vertical filter */
@@ -489,6 +658,15 @@ void dce_transform_construct(struct dce_transform *xfm_dce,
 	const struct dce_transform_shift *xfm_shift,
 	const struct dce_transform_mask *xfm_mask);
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+void dce60_transform_construct(struct dce_transform *xfm_dce,
+	struct dc_context *ctx,
+	uint32_t inst,
+	const struct dce_transform_registers *regs,
+	const struct dce_transform_shift *xfm_shift,
+	const struct dce_transform_mask *xfm_mask);
+#endif
+
 bool dce_transform_get_optimal_number_of_taps(
 	struct transform *xfm,
 	struct scaler_data *scl_data,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
index 5167d6b8a48d..67af67ef2865 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@@ -119,10 +119,11 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state *
 /**
  * Enable/Disable PSR.
  */
-static void dmub_psr_enable(struct dmub_psr *dmub, bool enable)
+static void dmub_psr_enable(struct dmub_psr *dmub, bool enable, bool wait)
 {
 	union dmub_rb_cmd cmd;
 	struct dc_context *dc = dmub->ctx;
+	uint32_t retry_count, psr_state = 0;
 
 	cmd.psr_enable.header.type = DMUB_CMD__PSR;
 
@@ -136,6 +137,30 @@ static void dmub_psr_enable(struct dmub_psr *dmub, bool enable)
 	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
 	dc_dmub_srv_cmd_execute(dc->dmub_srv);
 	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+	/* Below loops 1000 x 500us = 500 ms.
+	 *  Exit PSR may need to wait 1-2 frames to power up. Timeout after at
+	 *  least a few frames. Should never hit the max retry assert below.
+	 */
+	if (wait) {
+		for (retry_count = 0; retry_count <= 1000; retry_count++) {
+			dmub_psr_get_state(dmub, &psr_state);
+
+			if (enable) {
+				if (psr_state != 0)
+					break;
+			} else {
+				if (psr_state == 0)
+					break;
+			}
+
+			udelay(500);
+		}
+
+		/* assert if max retry hit */
+		if (retry_count >= 1000)
+			ASSERT(0);
+	}
 }
 
 /**
@@ -231,10 +256,11 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub,
 	copy_settings_data->smu_optimizations_en		= psr_context->allow_smu_optimizations;
 	copy_settings_data->frame_delay				= psr_context->frame_delay;
 	copy_settings_data->frame_cap_ind			= psr_context->psrFrameCaptureIndicationReq;
+	copy_settings_data->init_sdp_deadline			= psr_context->sdpTransmitLineNumDeadline;
+	copy_settings_data->debug.u32All = 0;
 	copy_settings_data->debug.bitfields.visual_confirm	= dc->dc->debug.visual_confirm == VISUAL_CONFIRM_PSR ?
 									true : false;
-	copy_settings_data->debug.bitfields.use_hw_lock_mgr	= 1;
-	copy_settings_data->init_sdp_deadline			= psr_context->sdpTransmitLineNumDeadline;
+	copy_settings_data->debug.bitfields.use_hw_lock_mgr		= 1;
 
 	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
 	dc_dmub_srv_cmd_execute(dc->dmub_srv);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h
index f404fecd6410..dc121ed92d2e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h
@@ -36,7 +36,7 @@ struct dmub_psr {
 
 struct dmub_psr_funcs {
 	bool (*psr_copy_settings)(struct dmub_psr *dmub, struct dc_link *link, struct psr_context *psr_context);
-	void (*psr_enable)(struct dmub_psr *dmub, bool enable);
+	void (*psr_enable)(struct dmub_psr *dmub, bool enable, bool wait);
 	void (*psr_get_state)(struct dmub_psr *dmub, uint32_t *psr_state);
 	void (*psr_set_level)(struct dmub_psr *dmub, uint16_t psr_level);
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 45c9e9027886..3ac6c7b65a45 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -720,6 +720,7 @@ void dce110_edp_wait_for_hpd_ready(
 	struct dc_context *ctx = link->ctx;
 	struct graphics_object_id connector = link->link_enc->connector;
 	struct gpio *hpd;
+	struct dc_sink *sink = link->local_sink;
 	bool edp_hpd_high = false;
 	uint32_t time_elapsed = 0;
 	uint32_t timeout = power_up ?
@@ -752,6 +753,14 @@ void dce110_edp_wait_for_hpd_ready(
 		return;
 	}
 
+	if (sink != NULL) {
+		if (sink->edid_caps.panel_patch.extra_t3_ms > 0) {
+			int extra_t3_in_ms = sink->edid_caps.panel_patch.extra_t3_ms;
+
+			msleep(extra_t3_in_ms);
+		}
+	}
+
 	dal_gpio_open(hpd, GPIO_MODE_INTERRUPT);
 
 	/* wait until timeout or panel detected */
@@ -801,37 +810,66 @@ void dce110_edp_power_control(
 
 	if (power_up !=
 		link->panel_cntl->funcs->is_panel_powered_on(link->panel_cntl)) {
+
+		unsigned long long current_ts = dm_get_timestamp(ctx);
+		unsigned long long time_since_edp_poweroff_ms =
+				div64_u64(dm_get_elapse_time_in_ns(
+						ctx,
+						current_ts,
+						link->link_trace.time_stamp.edp_poweroff), 1000000);
+		unsigned long long time_since_edp_poweron_ms =
+				div64_u64(dm_get_elapse_time_in_ns(
+						ctx,
+						current_ts,
+						link->link_trace.time_stamp.edp_poweron), 1000000);
+		DC_LOG_HW_RESUME_S3(
+				"%s: transition: power_up=%d current_ts=%llu edp_poweroff=%llu edp_poweron=%llu time_since_edp_poweroff_ms=%llu time_since_edp_poweron_ms=%llu",
+				__func__,
+				power_up,
+				current_ts,
+				link->link_trace.time_stamp.edp_poweroff,
+				link->link_trace.time_stamp.edp_poweron,
+				time_since_edp_poweroff_ms,
+				time_since_edp_poweron_ms);
+
 		/* Send VBIOS command to prompt eDP panel power */
 		if (power_up) {
-			unsigned long long current_ts = dm_get_timestamp(ctx);
-			unsigned long long duration_in_ms =
-					div64_u64(dm_get_elapse_time_in_ns(
-							ctx,
-							current_ts,
-							link->link_trace.time_stamp.edp_poweroff), 1000000);
-			unsigned long long wait_time_ms = 0;
-
-			/* max 500ms from LCDVDD off to on */
-			unsigned long long edp_poweroff_time_ms = 500;
+			/* edp requires a min of 500ms from LCDVDD off to on */
+			unsigned long long remaining_min_edp_poweroff_time_ms = 500;
 
+			/* add time defined by a patch, if any (usually patch extra_t12_ms is 0) */
 			if (link->local_sink != NULL)
-				edp_poweroff_time_ms =
-						500 + link->local_sink->edid_caps.panel_patch.extra_t12_ms;
-			if (link->link_trace.time_stamp.edp_poweroff == 0)
-				wait_time_ms = edp_poweroff_time_ms;
-			else if (duration_in_ms < edp_poweroff_time_ms)
-				wait_time_ms = edp_poweroff_time_ms - duration_in_ms;
-
-			if (wait_time_ms) {
-				msleep(wait_time_ms);
-				dm_output_to_console("%s: wait %lld ms to power on eDP.\n",
-						__func__, wait_time_ms);
+				remaining_min_edp_poweroff_time_ms +=
+					link->local_sink->edid_caps.panel_patch.extra_t12_ms;
+
+			/* Adjust remaining_min_edp_poweroff_time_ms if this is not the first time. */
+			if (link->link_trace.time_stamp.edp_poweroff != 0) {
+				if (time_since_edp_poweroff_ms < remaining_min_edp_poweroff_time_ms)
+					remaining_min_edp_poweroff_time_ms =
+						remaining_min_edp_poweroff_time_ms - time_since_edp_poweroff_ms;
+				else
+					remaining_min_edp_poweroff_time_ms = 0;
 			}
 
+			if (remaining_min_edp_poweroff_time_ms) {
+				DC_LOG_HW_RESUME_S3(
+						"%s: remaining_min_edp_poweroff_time_ms=%llu: begin wait.\n",
+						__func__, remaining_min_edp_poweroff_time_ms);
+				msleep(remaining_min_edp_poweroff_time_ms);
+				DC_LOG_HW_RESUME_S3(
+						"%s: remaining_min_edp_poweroff_time_ms=%llu: end wait.\n",
+						__func__, remaining_min_edp_poweroff_time_ms);
+				dm_output_to_console("%s: wait %lld ms to power on eDP.\n",
+						__func__, remaining_min_edp_poweroff_time_ms);
+			} else {
+				DC_LOG_HW_RESUME_S3(
+						"%s: remaining_min_edp_poweroff_time_ms=%llu: no wait required.\n",
+						__func__, remaining_min_edp_poweroff_time_ms);
+			}
 		}
 
 		DC_LOG_HW_RESUME_S3(
-				"%s: Panel Power action: %s\n",
+				"%s: BEGIN: Panel Power action: %s\n",
 				__func__, (power_up ? "On":"Off"));
 
 		cntl.action = power_up ?
@@ -855,12 +893,23 @@ void dce110_edp_power_control(
 
 		bp_result = link_transmitter_control(ctx->dc_bios, &cntl);
 
+		DC_LOG_HW_RESUME_S3(
+				"%s: END: Panel Power action: %s bp_result=%u\n",
+				__func__, (power_up ? "On":"Off"),
+				bp_result);
+
 		if (!power_up)
 			/*save driver power off time stamp*/
 			link->link_trace.time_stamp.edp_poweroff = dm_get_timestamp(ctx);
 		else
 			link->link_trace.time_stamp.edp_poweron = dm_get_timestamp(ctx);
 
+		DC_LOG_HW_RESUME_S3(
+				"%s: updated values: edp_poweroff=%llu edp_poweron=%llu\n",
+				__func__,
+				link->link_trace.time_stamp.edp_poweroff,
+				link->link_trace.time_stamp.edp_poweron);
+
 		if (bp_result != BP_RESULT_OK)
 			DC_LOG_ERROR(
 					"%s: Panel Power bp_result: %d\n",
@@ -1605,7 +1654,7 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 		// enable fastboot if backend is enabled on eDP
 		if (edp_link->link_enc->funcs->is_dig_enabled(edp_link->link_enc)) {
 			/* Set optimization flag on eDP stream*/
-			if (edp_stream) {
+			if (edp_stream && edp_link->link_status.link_active) {
 				edp_stream->apply_edp_fast_boot_optimization = true;
 				can_apply_edp_fast_boot = true;
 			}
@@ -2688,7 +2737,7 @@ static void program_output_csc(struct dc *dc,
 	}
 }
 
-void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx)
+static void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_cursor_position pos_cpy = pipe_ctx->stream->cursor_position;
 	struct input_pixel_processor *ipp = pipe_ctx->plane_res.ipp;
@@ -2733,7 +2782,7 @@ void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx)
 		mi->funcs->set_cursor_position(mi, &pos_cpy, &param);
 }
 
-void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx)
+static void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_cursor_attributes *attributes = &pipe_ctx->stream->cursor_attributes;
 
@@ -2841,6 +2890,7 @@ static const struct hw_sequencer_funcs dce110_funcs = {
 	.setup_stereo = NULL,
 	.set_avmute = dce110_set_avmute,
 	.wait_for_mpcc_disconnect = dce110_wait_for_mpcc_disconnect,
+	.edp_backlight_control = dce110_edp_backlight_control,
 	.edp_power_control = dce110_edp_power_control,
 	.edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready,
 	.set_cursor_position = dce110_set_cursor_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
index 5d83e8174005..0853bc9917c7 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -1017,7 +1017,7 @@ enum dc_status dce112_add_stream_to_ctx(
 		struct dc_state *new_ctx,
 		struct dc_stream_state *dc_stream)
 {
-	enum dc_status result = DC_ERROR_UNEXPECTED;
+	enum dc_status result;
 
 	result = resource_map_pool_resources(dc, new_ctx, dc_stream);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/Makefile b/drivers/gpu/drm/amd/display/dc/dce60/Makefile
new file mode 100644
index 000000000000..7036c3bd0f87
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/Makefile
@@ -0,0 +1,34 @@
+#
+# Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+#
+# 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.
+#
+#
+# Makefile for the 'controller' sub-component of DAL.
+# It provides the control and status of HW CRTC block.
+
+DCE60 = dce60_timing_generator.o dce60_hw_sequencer.o \
+	dce60_resource.o
+
+AMD_DAL_DCE60 = $(addprefix $(AMDDALPATH)/dc/dce60/,$(DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_DCE60)
+
+
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.c
new file mode 100644
index 000000000000..920c7ae29d53
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.c
@@ -0,0 +1,432 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include "dm_services.h"
+#include "dc.h"
+#include "core_types.h"
+#include "dce60_hw_sequencer.h"
+
+#include "dce/dce_hwseq.h"
+#include "dce110/dce110_hw_sequencer.h"
+#include "dce100/dce100_hw_sequencer.h"
+
+/* include DCE6 register header files */
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#define DC_LOGGER_INIT()
+
+/*******************************************************************************
+ * Private definitions
+ ******************************************************************************/
+
+/***************************PIPE_CONTROL***********************************/
+
+/*
+ *  Check if FBC can be enabled
+ */
+static bool dce60_should_enable_fbc(struct dc *dc,
+		struct dc_state *context,
+		uint32_t *pipe_idx)
+{
+	uint32_t i;
+	struct pipe_ctx *pipe_ctx = NULL;
+	struct resource_context *res_ctx = &context->res_ctx;
+	unsigned int underlay_idx = dc->res_pool->underlay_pipe_index;
+
+
+	ASSERT(dc->fbc_compressor);
+
+	/* FBC memory should be allocated */
+	if (!dc->ctx->fbc_gpu_addr)
+		return false;
+
+	/* Only supports single display */
+	if (context->stream_count != 1)
+		return false;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (res_ctx->pipe_ctx[i].stream) {
+
+			pipe_ctx = &res_ctx->pipe_ctx[i];
+
+			if (!pipe_ctx)
+				continue;
+
+			/* fbc not applicable on underlay pipe */
+			if (pipe_ctx->pipe_idx != underlay_idx) {
+				*pipe_idx = i;
+				break;
+			}
+		}
+	}
+
+	if (i == dc->res_pool->pipe_count)
+		return false;
+
+	if (!pipe_ctx->stream->link)
+		return false;
+
+	/* Only supports eDP */
+	if (pipe_ctx->stream->link->connector_signal != SIGNAL_TYPE_EDP)
+		return false;
+
+	/* PSR should not be enabled */
+	if (pipe_ctx->stream->link->psr_settings.psr_feature_enabled)
+		return false;
+
+	/* Nothing to compress */
+	if (!pipe_ctx->plane_state)
+		return false;
+
+	/* Only for non-linear tiling */
+	if (pipe_ctx->plane_state->tiling_info.gfx8.array_mode == DC_ARRAY_LINEAR_GENERAL)
+		return false;
+
+	return true;
+}
+
+/*
+ *  Enable FBC
+ */
+static void dce60_enable_fbc(
+		struct dc *dc,
+		struct dc_state *context)
+{
+	uint32_t pipe_idx = 0;
+
+	if (dce60_should_enable_fbc(dc, context, &pipe_idx)) {
+		/* Program GRPH COMPRESSED ADDRESS and PITCH */
+		struct compr_addr_and_pitch_params params = {0, 0, 0};
+		struct compressor *compr = dc->fbc_compressor;
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[pipe_idx];
+
+		params.source_view_width = pipe_ctx->stream->timing.h_addressable;
+		params.source_view_height = pipe_ctx->stream->timing.v_addressable;
+		params.inst = pipe_ctx->stream_res.tg->inst;
+		compr->compr_surface_address.quad_part = dc->ctx->fbc_gpu_addr;
+
+		compr->funcs->surface_address_and_pitch(compr, &params);
+		compr->funcs->set_fbc_invalidation_triggers(compr, 1);
+
+		compr->funcs->enable_fbc(compr, &params);
+	}
+}
+
+
+/*******************************************************************************
+ * Front End programming
+ ******************************************************************************/
+
+static void dce60_set_default_colors(struct pipe_ctx *pipe_ctx)
+{
+	struct default_adjustment default_adjust = { 0 };
+
+	default_adjust.force_hw_default = false;
+	default_adjust.in_color_space = pipe_ctx->plane_state->color_space;
+	default_adjust.out_color_space = pipe_ctx->stream->output_color_space;
+	default_adjust.csc_adjust_type = GRAPHICS_CSC_ADJUST_TYPE_SW;
+	default_adjust.surface_pixel_format = pipe_ctx->plane_res.scl_data.format;
+
+	/* display color depth */
+	default_adjust.color_depth =
+		pipe_ctx->stream->timing.display_color_depth;
+
+	/* Lb color depth */
+	default_adjust.lb_color_depth = pipe_ctx->plane_res.scl_data.lb_params.depth;
+
+	pipe_ctx->plane_res.xfm->funcs->opp_set_csc_default(
+					pipe_ctx->plane_res.xfm, &default_adjust);
+}
+
+/*******************************************************************************
+ * In order to turn on surface we will program
+ * CRTC
+ *
+ * DCE6 has no bottom_pipe and no Blender HW
+ * We need to set 'blank_target' to false in order to turn on the display
+ *
+ * |-----------|------------|---------|
+ * |curr pipe  | set_blank  |         |
+ * |Surface    |blank_target|  CRCT   |
+ * |visibility |  argument  |         |
+ * |-----------|------------|---------|
+ * |    off    |   true     | blank   |
+ * |    on     |   false    | unblank |
+ * |-----------|------------|---------|
+ *
+ ******************************************************************************/
+static void dce60_program_surface_visibility(const struct dc *dc,
+		struct pipe_ctx *pipe_ctx)
+{
+	bool blank_target = false;
+
+	/* DCE6 has no bottom_pipe and no Blender HW */
+
+	if (!pipe_ctx->plane_state->visible)
+		blank_target = true;
+
+	/* DCE6 skip dce_set_blender_mode() but then proceed to 'unblank' CRTC */
+	pipe_ctx->stream_res.tg->funcs->set_blank(pipe_ctx->stream_res.tg, blank_target);
+
+}
+
+
+static void dce60_get_surface_visual_confirm_color(const struct pipe_ctx *pipe_ctx,
+		struct tg_color *color)
+{
+	uint32_t color_value = MAX_TG_COLOR_VALUE * (4 - pipe_ctx->stream_res.tg->inst) / 4;
+
+	switch (pipe_ctx->plane_res.scl_data.format) {
+	case PIXEL_FORMAT_ARGB8888:
+		/* set boarder color to red */
+		color->color_r_cr = color_value;
+		break;
+
+	case PIXEL_FORMAT_ARGB2101010:
+		/* set boarder color to blue */
+		color->color_b_cb = color_value;
+		break;
+	case PIXEL_FORMAT_420BPP8:
+		/* set boarder color to green */
+		color->color_g_y = color_value;
+		break;
+	case PIXEL_FORMAT_420BPP10:
+		/* set boarder color to yellow */
+		color->color_g_y = color_value;
+		color->color_r_cr = color_value;
+		break;
+	case PIXEL_FORMAT_FP16:
+		/* set boarder color to white */
+		color->color_r_cr = color_value;
+		color->color_b_cb = color_value;
+		color->color_g_y = color_value;
+		break;
+	default:
+		break;
+	}
+}
+
+static void dce60_program_scaler(const struct dc *dc,
+		const struct pipe_ctx *pipe_ctx)
+{
+	struct tg_color color = {0};
+
+	/* DCE6 skips DCN TOFPGA check for transform_set_pixel_storage_depth == NULL */
+
+	if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE)
+		dce60_get_surface_visual_confirm_color(pipe_ctx, &color);
+	else
+		color_space_to_black_color(dc,
+				pipe_ctx->stream->output_color_space,
+				&color);
+
+	pipe_ctx->plane_res.xfm->funcs->transform_set_pixel_storage_depth(
+		pipe_ctx->plane_res.xfm,
+		pipe_ctx->plane_res.scl_data.lb_params.depth,
+		&pipe_ctx->stream->bit_depth_params);
+
+	if (pipe_ctx->stream_res.tg->funcs->set_overscan_blank_color) {
+		/*
+		 * The way 420 is packed, 2 channels carry Y component, 1 channel
+		 * alternate between Cb and Cr, so both channels need the pixel
+		 * value for Y
+		 */
+		if (pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)
+			color.color_r_cr = color.color_g_y;
+
+		pipe_ctx->stream_res.tg->funcs->set_overscan_blank_color(
+				pipe_ctx->stream_res.tg,
+				&color);
+	}
+
+	pipe_ctx->plane_res.xfm->funcs->transform_set_scaler(pipe_ctx->plane_res.xfm,
+		&pipe_ctx->plane_res.scl_data);
+}
+
+static void
+dce60_program_front_end_for_pipe(
+		struct dc *dc, struct pipe_ctx *pipe_ctx)
+{
+	struct mem_input *mi = pipe_ctx->plane_res.mi;
+	struct dc_plane_state *plane_state = pipe_ctx->plane_state;
+	struct xfm_grph_csc_adjustment adjust;
+	struct out_csc_color_matrix tbl_entry;
+	unsigned int i;
+	struct dce_hwseq *hws = dc->hwseq;
+
+	DC_LOGGER_INIT();
+	memset(&tbl_entry, 0, sizeof(tbl_entry));
+
+	memset(&adjust, 0, sizeof(adjust));
+	adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS;
+
+	dce_enable_fe_clock(dc->hwseq, mi->inst, true);
+
+	dce60_set_default_colors(pipe_ctx);
+	if (pipe_ctx->stream->csc_color_matrix.enable_adjustment
+			== true) {
+		tbl_entry.color_space =
+			pipe_ctx->stream->output_color_space;
+
+		for (i = 0; i < 12; i++)
+			tbl_entry.regval[i] =
+			pipe_ctx->stream->csc_color_matrix.matrix[i];
+
+		pipe_ctx->plane_res.xfm->funcs->opp_set_csc_adjustment
+				(pipe_ctx->plane_res.xfm, &tbl_entry);
+	}
+
+	if (pipe_ctx->stream->gamut_remap_matrix.enable_remap == true) {
+		adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW;
+
+		for (i = 0; i < CSC_TEMPERATURE_MATRIX_SIZE; i++)
+			adjust.temperature_matrix[i] =
+				pipe_ctx->stream->gamut_remap_matrix.matrix[i];
+	}
+
+	pipe_ctx->plane_res.xfm->funcs->transform_set_gamut_remap(pipe_ctx->plane_res.xfm, &adjust);
+
+	pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0;
+
+	dce60_program_scaler(dc, pipe_ctx);
+
+	mi->funcs->mem_input_program_surface_config(
+			mi,
+			plane_state->format,
+			&plane_state->tiling_info,
+			&plane_state->plane_size,
+			plane_state->rotation,
+			NULL,
+			false);
+	if (mi->funcs->set_blank)
+		mi->funcs->set_blank(mi, pipe_ctx->plane_state->visible);
+
+	if (dc->config.gpu_vm_support)
+		mi->funcs->mem_input_program_pte_vm(
+				pipe_ctx->plane_res.mi,
+				plane_state->format,
+				&plane_state->tiling_info,
+				plane_state->rotation);
+
+	/* Moved programming gamma from dc to hwss */
+	if (pipe_ctx->plane_state->update_flags.bits.full_update ||
+			pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change ||
+			pipe_ctx->plane_state->update_flags.bits.gamma_change)
+		hws->funcs.set_input_transfer_func(dc, pipe_ctx, pipe_ctx->plane_state);
+
+	if (pipe_ctx->plane_state->update_flags.bits.full_update)
+		hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream);
+
+	DC_LOG_SURFACE(
+			"Pipe:%d %p: addr hi:0x%x, "
+			"addr low:0x%x, "
+			"src: %d, %d, %d,"
+			" %d; dst: %d, %d, %d, %d;"
+			"clip: %d, %d, %d, %d\n",
+			pipe_ctx->pipe_idx,
+			(void *) pipe_ctx->plane_state,
+			pipe_ctx->plane_state->address.grph.addr.high_part,
+			pipe_ctx->plane_state->address.grph.addr.low_part,
+			pipe_ctx->plane_state->src_rect.x,
+			pipe_ctx->plane_state->src_rect.y,
+			pipe_ctx->plane_state->src_rect.width,
+			pipe_ctx->plane_state->src_rect.height,
+			pipe_ctx->plane_state->dst_rect.x,
+			pipe_ctx->plane_state->dst_rect.y,
+			pipe_ctx->plane_state->dst_rect.width,
+			pipe_ctx->plane_state->dst_rect.height,
+			pipe_ctx->plane_state->clip_rect.x,
+			pipe_ctx->plane_state->clip_rect.y,
+			pipe_ctx->plane_state->clip_rect.width,
+			pipe_ctx->plane_state->clip_rect.height);
+
+	DC_LOG_SURFACE(
+			"Pipe %d: width, height, x, y\n"
+			"viewport:%d, %d, %d, %d\n"
+			"recout:  %d, %d, %d, %d\n",
+			pipe_ctx->pipe_idx,
+			pipe_ctx->plane_res.scl_data.viewport.width,
+			pipe_ctx->plane_res.scl_data.viewport.height,
+			pipe_ctx->plane_res.scl_data.viewport.x,
+			pipe_ctx->plane_res.scl_data.viewport.y,
+			pipe_ctx->plane_res.scl_data.recout.width,
+			pipe_ctx->plane_res.scl_data.recout.height,
+			pipe_ctx->plane_res.scl_data.recout.x,
+			pipe_ctx->plane_res.scl_data.recout.y);
+}
+
+static void dce60_apply_ctx_for_surface(
+		struct dc *dc,
+		const struct dc_stream_state *stream,
+		int num_planes,
+		struct dc_state *context)
+{
+	int i;
+
+	if (num_planes == 0)
+		return;
+
+	if (dc->fbc_compressor)
+		dc->fbc_compressor->funcs->disable_fbc(dc->fbc_compressor);
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+
+		if (pipe_ctx->stream != stream)
+			continue;
+
+		/* Need to allocate mem before program front end for Fiji */
+		pipe_ctx->plane_res.mi->funcs->allocate_mem_input(
+				pipe_ctx->plane_res.mi,
+				pipe_ctx->stream->timing.h_total,
+				pipe_ctx->stream->timing.v_total,
+				pipe_ctx->stream->timing.pix_clk_100hz / 10,
+				context->stream_count);
+
+		dce60_program_front_end_for_pipe(dc, pipe_ctx);
+
+		dc->hwss.update_plane_addr(dc, pipe_ctx);
+
+		dce60_program_surface_visibility(dc, pipe_ctx);
+
+	}
+
+	if (dc->fbc_compressor)
+		dce60_enable_fbc(dc, context);
+}
+
+void dce60_hw_sequencer_construct(struct dc *dc)
+{
+	dce110_hw_sequencer_construct(dc);
+
+	dc->hwseq->funcs.enable_display_power_gating = dce100_enable_display_power_gating;
+	dc->hwss.apply_ctx_for_surface = dce60_apply_ctx_for_surface;
+	dc->hwss.cursor_lock = dce60_pipe_control_lock;
+	dc->hwss.pipe_control_lock = dce60_pipe_control_lock;
+	dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth;
+	dc->hwss.optimize_bandwidth = dce100_optimize_bandwidth;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.h
new file mode 100644
index 000000000000..f3b2d8b60d5b
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DC_HWSS_DCE60_H__
+#define __DC_HWSS_DCE60_H__
+
+#include "core_types.h"
+#include "hw_sequencer_private.h"
+
+struct dc;
+
+void dce60_hw_sequencer_construct(struct dc *dc);
+
+#endif /* __DC_HWSS_DCE60_H__ */
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c
new file mode 100644
index 000000000000..5a5a9cb77acb
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c
@@ -0,0 +1,1527 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include <linux/slab.h>
+
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#include "dm_services.h"
+
+#include "link_encoder.h"
+#include "stream_encoder.h"
+
+#include "resource.h"
+#include "include/irq_service_interface.h"
+#include "irq/dce60/irq_service_dce60.h"
+#include "dce110/dce110_timing_generator.h"
+#include "dce110/dce110_resource.h"
+#include "dce60/dce60_timing_generator.h"
+#include "dce/dce_mem_input.h"
+#include "dce/dce_link_encoder.h"
+#include "dce/dce_stream_encoder.h"
+#include "dce/dce_ipp.h"
+#include "dce/dce_transform.h"
+#include "dce/dce_opp.h"
+#include "dce/dce_clock_source.h"
+#include "dce/dce_audio.h"
+#include "dce/dce_hwseq.h"
+#include "dce60/dce60_hw_sequencer.h"
+#include "dce100/dce100_resource.h"
+#include "dce/dce_panel_cntl.h"
+
+#include "reg_helper.h"
+
+#include "dce/dce_dmcu.h"
+#include "dce/dce_aux.h"
+#include "dce/dce_abm.h"
+#include "dce/dce_i2c.h"
+/* TODO remove this include */
+
+#ifndef mmMC_HUB_RDREQ_DMIF_LIMIT
+#include "gmc/gmc_6_0_d.h"
+#include "gmc/gmc_6_0_sh_mask.h"
+#endif
+
+#ifndef mmDP_DPHY_INTERNAL_CTRL
+#define mmDP_DPHY_INTERNAL_CTRL                         0x1CDE
+#define mmDP0_DP_DPHY_INTERNAL_CTRL                     0x1CDE
+#define mmDP1_DP_DPHY_INTERNAL_CTRL                     0x1FDE
+#define mmDP2_DP_DPHY_INTERNAL_CTRL                     0x42DE
+#define mmDP3_DP_DPHY_INTERNAL_CTRL                     0x45DE
+#define mmDP4_DP_DPHY_INTERNAL_CTRL                     0x48DE
+#define mmDP5_DP_DPHY_INTERNAL_CTRL                     0x4BDE
+#endif
+
+
+#ifndef mmBIOS_SCRATCH_2
+	#define mmBIOS_SCRATCH_2 0x05CB
+	#define mmBIOS_SCRATCH_3 0x05CC
+	#define mmBIOS_SCRATCH_6 0x05CF
+#endif
+
+#ifndef mmDP_DPHY_FAST_TRAINING
+	#define mmDP_DPHY_FAST_TRAINING                         0x1CCE
+	#define mmDP0_DP_DPHY_FAST_TRAINING                     0x1CCE
+	#define mmDP1_DP_DPHY_FAST_TRAINING                     0x1FCE
+	#define mmDP2_DP_DPHY_FAST_TRAINING                     0x42CE
+	#define mmDP3_DP_DPHY_FAST_TRAINING                     0x45CE
+	#define mmDP4_DP_DPHY_FAST_TRAINING                     0x48CE
+	#define mmDP5_DP_DPHY_FAST_TRAINING                     0x4BCE
+#endif
+
+
+#ifndef mmHPD_DC_HPD_CONTROL
+	#define mmHPD_DC_HPD_CONTROL                            0x189A
+	#define mmHPD0_DC_HPD_CONTROL                           0x189A
+	#define mmHPD1_DC_HPD_CONTROL                           0x18A2
+	#define mmHPD2_DC_HPD_CONTROL                           0x18AA
+	#define mmHPD3_DC_HPD_CONTROL                           0x18B2
+	#define mmHPD4_DC_HPD_CONTROL                           0x18BA
+	#define mmHPD5_DC_HPD_CONTROL                           0x18C2
+#endif
+
+#define DCE11_DIG_FE_CNTL 0x4a00
+#define DCE11_DIG_BE_CNTL 0x4a47
+#define DCE11_DP_SEC 0x4ac3
+
+static const struct dce110_timing_generator_offsets dce60_tg_offsets[] = {
+		{
+			.crtc = (mmCRTC0_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp =  (mmGRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG0_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		},
+		{
+			.crtc = (mmCRTC1_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp = (mmDCP1_GRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG1_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		},
+		{
+			.crtc = (mmCRTC2_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp = (mmDCP2_GRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG2_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		},
+		{
+			.crtc = (mmCRTC3_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp = (mmDCP3_GRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG3_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		},
+		{
+			.crtc = (mmCRTC4_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp = (mmDCP4_GRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG4_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		},
+		{
+			.crtc = (mmCRTC5_CRTC_CONTROL - mmCRTC_CONTROL),
+			.dcp = (mmDCP5_GRPH_CONTROL - mmGRPH_CONTROL),
+			.dmif = (mmDMIF_PG5_DPG_PIPE_ARBITRATION_CONTROL3
+					- mmDPG_PIPE_ARBITRATION_CONTROL3),
+		}
+};
+
+/* set register offset */
+#define SR(reg_name)\
+	.reg_name = mm ## reg_name
+
+/* set register offset with instance */
+#define SRI(reg_name, block, id)\
+	.reg_name = mm ## block ## id ## _ ## reg_name
+
+#define ipp_regs(id)\
+[id] = {\
+		IPP_COMMON_REG_LIST_DCE_BASE(id)\
+}
+
+static const struct dce_ipp_registers ipp_regs[] = {
+		ipp_regs(0),
+		ipp_regs(1),
+		ipp_regs(2),
+		ipp_regs(3),
+		ipp_regs(4),
+		ipp_regs(5)
+};
+
+static const struct dce_ipp_shift ipp_shift = {
+		IPP_DCE60_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
+};
+
+static const struct dce_ipp_mask ipp_mask = {
+		IPP_DCE60_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
+};
+
+#define transform_regs(id)\
+[id] = {\
+		XFM_COMMON_REG_LIST_DCE60(id)\
+}
+
+static const struct dce_transform_registers xfm_regs[] = {
+		transform_regs(0),
+		transform_regs(1),
+		transform_regs(2),
+		transform_regs(3),
+		transform_regs(4),
+		transform_regs(5)
+};
+
+static const struct dce_transform_shift xfm_shift = {
+		XFM_COMMON_MASK_SH_LIST_DCE60(__SHIFT)
+};
+
+static const struct dce_transform_mask xfm_mask = {
+		XFM_COMMON_MASK_SH_LIST_DCE60(_MASK)
+};
+
+#define aux_regs(id)\
+[id] = {\
+	AUX_REG_LIST(id)\
+}
+
+static const struct dce110_link_enc_aux_registers link_enc_aux_regs[] = {
+	aux_regs(0),
+	aux_regs(1),
+	aux_regs(2),
+	aux_regs(3),
+	aux_regs(4),
+	aux_regs(5)
+};
+
+#define hpd_regs(id)\
+[id] = {\
+	HPD_REG_LIST(id)\
+}
+
+static const struct dce110_link_enc_hpd_registers link_enc_hpd_regs[] = {
+		hpd_regs(0),
+		hpd_regs(1),
+		hpd_regs(2),
+		hpd_regs(3),
+		hpd_regs(4),
+		hpd_regs(5)
+};
+
+#define link_regs(id)\
+[id] = {\
+	LE_DCE60_REG_LIST(id)\
+}
+
+static const struct dce110_link_enc_registers link_enc_regs[] = {
+	link_regs(0),
+	link_regs(1),
+	link_regs(2),
+	link_regs(3),
+	link_regs(4),
+	link_regs(5)
+};
+
+#define stream_enc_regs(id)\
+[id] = {\
+	SE_COMMON_REG_LIST_DCE_BASE(id),\
+	.AFMT_CNTL = 0,\
+}
+
+static const struct dce110_stream_enc_registers stream_enc_regs[] = {
+	stream_enc_regs(0),
+	stream_enc_regs(1),
+	stream_enc_regs(2),
+	stream_enc_regs(3),
+	stream_enc_regs(4),
+	stream_enc_regs(5)
+};
+
+static const struct dce_stream_encoder_shift se_shift = {
+		SE_COMMON_MASK_SH_LIST_DCE80_100(__SHIFT)
+};
+
+static const struct dce_stream_encoder_mask se_mask = {
+		SE_COMMON_MASK_SH_LIST_DCE80_100(_MASK)
+};
+
+static const struct dce_panel_cntl_registers panel_cntl_regs[] = {
+	{ DCE_PANEL_CNTL_REG_LIST() }
+};
+
+static const struct dce_panel_cntl_shift panel_cntl_shift = {
+	DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct dce_panel_cntl_mask panel_cntl_mask = {
+	DCE_PANEL_CNTL_MASK_SH_LIST(_MASK)
+};
+
+#define opp_regs(id)\
+[id] = {\
+	OPP_DCE_60_REG_LIST(id),\
+}
+
+static const struct dce_opp_registers opp_regs[] = {
+	opp_regs(0),
+	opp_regs(1),
+	opp_regs(2),
+	opp_regs(3),
+	opp_regs(4),
+	opp_regs(5)
+};
+
+static const struct dce_opp_shift opp_shift = {
+	OPP_COMMON_MASK_SH_LIST_DCE_60(__SHIFT)
+};
+
+static const struct dce_opp_mask opp_mask = {
+	OPP_COMMON_MASK_SH_LIST_DCE_60(_MASK)
+};
+
+static const struct dce110_aux_registers_shift aux_shift = {
+	DCE10_AUX_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct dce110_aux_registers_mask aux_mask = {
+	DCE10_AUX_MASK_SH_LIST(_MASK)
+};
+
+#define aux_engine_regs(id)\
+[id] = {\
+	AUX_COMMON_REG_LIST(id), \
+	.AUX_RESET_MASK = 0 \
+}
+
+static const struct dce110_aux_registers aux_engine_regs[] = {
+		aux_engine_regs(0),
+		aux_engine_regs(1),
+		aux_engine_regs(2),
+		aux_engine_regs(3),
+		aux_engine_regs(4),
+		aux_engine_regs(5)
+};
+
+#define audio_regs(id)\
+[id] = {\
+	AUD_COMMON_REG_LIST(id)\
+}
+
+static const struct dce_audio_registers audio_regs[] = {
+	audio_regs(0),
+	audio_regs(1),
+	audio_regs(2),
+	audio_regs(3),
+	audio_regs(4),
+	audio_regs(5),
+};
+
+static const struct dce_audio_shift audio_shift = {
+		AUD_DCE60_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct dce_audio_mask audio_mask = {
+		AUD_DCE60_MASK_SH_LIST(_MASK)
+};
+
+#define clk_src_regs(id)\
+[id] = {\
+	CS_COMMON_REG_LIST_DCE_80(id),\
+}
+
+
+static const struct dce110_clk_src_regs clk_src_regs[] = {
+	clk_src_regs(0),
+	clk_src_regs(1),
+	clk_src_regs(2)
+};
+
+static const struct dce110_clk_src_shift cs_shift = {
+		CS_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
+};
+
+static const struct dce110_clk_src_mask cs_mask = {
+		CS_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
+};
+
+static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3,
+	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6
+};
+
+static const struct resource_caps res_cap = {
+		.num_timing_generator = 6,
+		.num_audio = 6,
+		.num_stream_encoder = 6,
+		.num_pll = 2,
+		.num_ddc = 6,
+};
+
+static const struct resource_caps res_cap_61 = {
+		.num_timing_generator = 4,
+		.num_audio = 6,
+		.num_stream_encoder = 6,
+		.num_pll = 3,
+		.num_ddc = 6,
+};
+
+static const struct resource_caps res_cap_64 = {
+		.num_timing_generator = 2,
+		.num_audio = 2,
+		.num_stream_encoder = 2,
+		.num_pll = 2,
+		.num_ddc = 2,
+};
+
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCE_RGB,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = false,
+			.fp16 = false
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 1,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 1,
+			.fp16 = 1
+	}
+};
+
+static const struct dce_dmcu_registers dmcu_regs = {
+		DMCU_DCE60_REG_LIST()
+};
+
+static const struct dce_dmcu_shift dmcu_shift = {
+		DMCU_MASK_SH_LIST_DCE60(__SHIFT)
+};
+
+static const struct dce_dmcu_mask dmcu_mask = {
+		DMCU_MASK_SH_LIST_DCE60(_MASK)
+};
+static const struct dce_abm_registers abm_regs = {
+		ABM_DCE110_COMMON_REG_LIST()
+};
+
+static const struct dce_abm_shift abm_shift = {
+		ABM_MASK_SH_LIST_DCE110(__SHIFT)
+};
+
+static const struct dce_abm_mask abm_mask = {
+		ABM_MASK_SH_LIST_DCE110(_MASK)
+};
+
+#define CTX  ctx
+#define REG(reg) mm ## reg
+
+#ifndef mmCC_DC_HDMI_STRAPS
+#define mmCC_DC_HDMI_STRAPS 0x1918
+#define CC_DC_HDMI_STRAPS__HDMI_DISABLE_MASK 0x40
+#define CC_DC_HDMI_STRAPS__HDMI_DISABLE__SHIFT 0x6
+#define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER_MASK 0x700
+#define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER__SHIFT 0x8
+#endif
+
+static int map_transmitter_id_to_phy_instance(
+	enum transmitter transmitter)
+{
+	switch (transmitter) {
+	case TRANSMITTER_UNIPHY_A:
+		return 0;
+	break;
+	case TRANSMITTER_UNIPHY_B:
+		return 1;
+	break;
+	case TRANSMITTER_UNIPHY_C:
+		return 2;
+	break;
+	case TRANSMITTER_UNIPHY_D:
+		return 3;
+	break;
+	case TRANSMITTER_UNIPHY_E:
+		return 4;
+	break;
+	case TRANSMITTER_UNIPHY_F:
+		return 5;
+	break;
+	case TRANSMITTER_UNIPHY_G:
+		return 6;
+	break;
+	default:
+		ASSERT(0);
+		return 0;
+	}
+}
+
+static void read_dce_straps(
+	struct dc_context *ctx,
+	struct resource_straps *straps)
+{
+	REG_GET_2(CC_DC_HDMI_STRAPS,
+			HDMI_DISABLE, &straps->hdmi_disable,
+			AUDIO_STREAM_NUMBER, &straps->audio_stream_number);
+
+	REG_GET(DC_PINSTRAPS, DC_PINSTRAPS_AUDIO, &straps->dc_pinstraps_audio);
+}
+
+static struct audio *create_audio(
+		struct dc_context *ctx, unsigned int inst)
+{
+	return dce60_audio_create(ctx, inst,
+			&audio_regs[inst], &audio_shift, &audio_mask);
+}
+
+static struct timing_generator *dce60_timing_generator_create(
+		struct dc_context *ctx,
+		uint32_t instance,
+		const struct dce110_timing_generator_offsets *offsets)
+{
+	struct dce110_timing_generator *tg110 =
+		kzalloc(sizeof(struct dce110_timing_generator), GFP_KERNEL);
+
+	if (!tg110)
+		return NULL;
+
+	dce60_timing_generator_construct(tg110, ctx, instance, offsets);
+	return &tg110->base;
+}
+
+static struct output_pixel_processor *dce60_opp_create(
+	struct dc_context *ctx,
+	uint32_t inst)
+{
+	struct dce110_opp *opp =
+		kzalloc(sizeof(struct dce110_opp), GFP_KERNEL);
+
+	if (!opp)
+		return NULL;
+
+	dce60_opp_construct(opp,
+			     ctx, inst, &opp_regs[inst], &opp_shift, &opp_mask);
+	return &opp->base;
+}
+
+struct dce_aux *dce60_aux_engine_create(
+	struct dc_context *ctx,
+	uint32_t inst)
+{
+	struct aux_engine_dce110 *aux_engine =
+		kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL);
+
+	if (!aux_engine)
+		return NULL;
+
+	dce110_aux_engine_construct(aux_engine, ctx, inst,
+				    SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD,
+				    &aux_engine_regs[inst],
+					&aux_mask,
+					&aux_shift,
+					ctx->dc->caps.extended_aux_timeout_support);
+
+	return &aux_engine->base;
+}
+#define i2c_inst_regs(id) { I2C_HW_ENGINE_COMMON_REG_LIST(id) }
+
+static const struct dce_i2c_registers i2c_hw_regs[] = {
+		i2c_inst_regs(1),
+		i2c_inst_regs(2),
+		i2c_inst_regs(3),
+		i2c_inst_regs(4),
+		i2c_inst_regs(5),
+		i2c_inst_regs(6),
+};
+
+static const struct dce_i2c_shift i2c_shifts = {
+		I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
+};
+
+static const struct dce_i2c_mask i2c_masks = {
+		I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
+};
+
+struct dce_i2c_hw *dce60_i2c_hw_create(
+	struct dc_context *ctx,
+	uint32_t inst)
+{
+	struct dce_i2c_hw *dce_i2c_hw =
+		kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL);
+
+	if (!dce_i2c_hw)
+		return NULL;
+
+	dce_i2c_hw_construct(dce_i2c_hw, ctx, inst,
+				    &i2c_hw_regs[inst], &i2c_shifts, &i2c_masks);
+
+	return dce_i2c_hw;
+}
+
+struct dce_i2c_sw *dce60_i2c_sw_create(
+	struct dc_context *ctx)
+{
+	struct dce_i2c_sw *dce_i2c_sw =
+		kzalloc(sizeof(struct dce_i2c_sw), GFP_KERNEL);
+
+	if (!dce_i2c_sw)
+		return NULL;
+
+	dce_i2c_sw_construct(dce_i2c_sw, ctx);
+
+	return dce_i2c_sw;
+}
+static struct stream_encoder *dce60_stream_encoder_create(
+	enum engine_id eng_id,
+	struct dc_context *ctx)
+{
+	struct dce110_stream_encoder *enc110 =
+		kzalloc(sizeof(struct dce110_stream_encoder), GFP_KERNEL);
+
+	if (!enc110)
+		return NULL;
+
+	dce110_stream_encoder_construct(enc110, ctx, ctx->dc_bios, eng_id,
+					&stream_enc_regs[eng_id],
+					&se_shift, &se_mask);
+	return &enc110->base;
+}
+
+#define SRII(reg_name, block, id)\
+	.reg_name[id] = mm ## block ## id ## _ ## reg_name
+
+static const struct dce_hwseq_registers hwseq_reg = {
+		HWSEQ_DCE6_REG_LIST()
+};
+
+static const struct dce_hwseq_shift hwseq_shift = {
+		HWSEQ_DCE6_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct dce_hwseq_mask hwseq_mask = {
+		HWSEQ_DCE6_MASK_SH_LIST(_MASK)
+};
+
+static struct dce_hwseq *dce60_hwseq_create(
+	struct dc_context *ctx)
+{
+	struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL);
+
+	if (hws) {
+		hws->ctx = ctx;
+		hws->regs = &hwseq_reg;
+		hws->shifts = &hwseq_shift;
+		hws->masks = &hwseq_mask;
+	}
+	return hws;
+}
+
+static const struct resource_create_funcs res_create_funcs = {
+	.read_dce_straps = read_dce_straps,
+	.create_audio = create_audio,
+	.create_stream_encoder = dce60_stream_encoder_create,
+	.create_hwseq = dce60_hwseq_create,
+};
+
+#define mi_inst_regs(id) { \
+	MI_DCE6_REG_LIST(id), \
+	.MC_HUB_RDREQ_DMIF_LIMIT = mmMC_HUB_RDREQ_DMIF_LIMIT \
+}
+static const struct dce_mem_input_registers mi_regs[] = {
+		mi_inst_regs(0),
+		mi_inst_regs(1),
+		mi_inst_regs(2),
+		mi_inst_regs(3),
+		mi_inst_regs(4),
+		mi_inst_regs(5),
+};
+
+static const struct dce_mem_input_shift mi_shifts = {
+		MI_DCE6_MASK_SH_LIST(__SHIFT),
+		.ENABLE = MC_HUB_RDREQ_DMIF_LIMIT__ENABLE__SHIFT
+};
+
+static const struct dce_mem_input_mask mi_masks = {
+		MI_DCE6_MASK_SH_LIST(_MASK),
+		.ENABLE = MC_HUB_RDREQ_DMIF_LIMIT__ENABLE_MASK
+};
+
+static struct mem_input *dce60_mem_input_create(
+	struct dc_context *ctx,
+	uint32_t inst)
+{
+	struct dce_mem_input *dce_mi = kzalloc(sizeof(struct dce_mem_input),
+					       GFP_KERNEL);
+
+	if (!dce_mi) {
+		BREAK_TO_DEBUGGER();
+		return NULL;
+	}
+
+	dce60_mem_input_construct(dce_mi, ctx, inst, &mi_regs[inst], &mi_shifts, &mi_masks);
+	dce_mi->wa.single_head_rdreq_dmif_limit = 2;
+	return &dce_mi->base;
+}
+
+static void dce60_transform_destroy(struct transform **xfm)
+{
+	kfree(TO_DCE_TRANSFORM(*xfm));
+	*xfm = NULL;
+}
+
+static struct transform *dce60_transform_create(
+	struct dc_context *ctx,
+	uint32_t inst)
+{
+	struct dce_transform *transform =
+		kzalloc(sizeof(struct dce_transform), GFP_KERNEL);
+
+	if (!transform)
+		return NULL;
+
+	dce60_transform_construct(transform, ctx, inst,
+				&xfm_regs[inst], &xfm_shift, &xfm_mask);
+	transform->prescaler_on = false;
+	return &transform->base;
+}
+
+static const struct encoder_feature_support link_enc_feature = {
+		.max_hdmi_deep_color = COLOR_DEPTH_121212,
+		.max_hdmi_pixel_clock = 297000,
+		.flags.bits.IS_HBR2_CAPABLE = true,
+		.flags.bits.IS_TPS3_CAPABLE = true
+};
+
+struct link_encoder *dce60_link_encoder_create(
+	const struct encoder_init_data *enc_init_data)
+{
+	struct dce110_link_encoder *enc110 =
+		kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL);
+	int link_regs_id;
+
+	if (!enc110)
+		return NULL;
+
+	link_regs_id =
+		map_transmitter_id_to_phy_instance(enc_init_data->transmitter);
+
+	dce60_link_encoder_construct(enc110,
+				      enc_init_data,
+				      &link_enc_feature,
+				      &link_enc_regs[link_regs_id],
+				      &link_enc_aux_regs[enc_init_data->channel - 1],
+				      &link_enc_hpd_regs[enc_init_data->hpd_source]);
+	return &enc110->base;
+}
+
+static struct panel_cntl *dce60_panel_cntl_create(const struct panel_cntl_init_data *init_data)
+{
+	struct dce_panel_cntl *panel_cntl =
+		kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL);
+
+	if (!panel_cntl)
+		return NULL;
+
+	dce_panel_cntl_construct(panel_cntl,
+			init_data,
+			&panel_cntl_regs[init_data->inst],
+			&panel_cntl_shift,
+			&panel_cntl_mask);
+
+	return &panel_cntl->base;
+}
+
+struct clock_source *dce60_clock_source_create(
+	struct dc_context *ctx,
+	struct dc_bios *bios,
+	enum clock_source_id id,
+	const struct dce110_clk_src_regs *regs,
+	bool dp_clk_src)
+{
+	struct dce110_clk_src *clk_src =
+		kzalloc(sizeof(struct dce110_clk_src), GFP_KERNEL);
+
+	if (!clk_src)
+		return NULL;
+
+	if (dce110_clk_src_construct(clk_src, ctx, bios, id,
+			regs, &cs_shift, &cs_mask)) {
+		clk_src->base.dp_clk_src = dp_clk_src;
+		return &clk_src->base;
+	}
+
+	kfree(clk_src);
+	BREAK_TO_DEBUGGER();
+	return NULL;
+}
+
+void dce60_clock_source_destroy(struct clock_source **clk_src)
+{
+	kfree(TO_DCE110_CLK_SRC(*clk_src));
+	*clk_src = NULL;
+}
+
+static struct input_pixel_processor *dce60_ipp_create(
+	struct dc_context *ctx, uint32_t inst)
+{
+	struct dce_ipp *ipp = kzalloc(sizeof(struct dce_ipp), GFP_KERNEL);
+
+	if (!ipp) {
+		BREAK_TO_DEBUGGER();
+		return NULL;
+	}
+
+	dce60_ipp_construct(ipp, ctx, inst,
+			&ipp_regs[inst], &ipp_shift, &ipp_mask);
+	return &ipp->base;
+}
+
+static void dce60_resource_destruct(struct dce110_resource_pool *pool)
+{
+	unsigned int i;
+
+	for (i = 0; i < pool->base.pipe_count; i++) {
+		if (pool->base.opps[i] != NULL)
+			dce110_opp_destroy(&pool->base.opps[i]);
+
+		if (pool->base.transforms[i] != NULL)
+			dce60_transform_destroy(&pool->base.transforms[i]);
+
+		if (pool->base.ipps[i] != NULL)
+			dce_ipp_destroy(&pool->base.ipps[i]);
+
+		if (pool->base.mis[i] != NULL) {
+			kfree(TO_DCE_MEM_INPUT(pool->base.mis[i]));
+			pool->base.mis[i] = NULL;
+		}
+
+		if (pool->base.timing_generators[i] != NULL)	{
+			kfree(DCE110TG_FROM_TG(pool->base.timing_generators[i]));
+			pool->base.timing_generators[i] = NULL;
+		}
+	}
+
+	for (i = 0; i < pool->base.res_cap->num_ddc; i++) {
+		if (pool->base.engines[i] != NULL)
+			dce110_engine_destroy(&pool->base.engines[i]);
+		if (pool->base.hw_i2cs[i] != NULL) {
+			kfree(pool->base.hw_i2cs[i]);
+			pool->base.hw_i2cs[i] = NULL;
+		}
+		if (pool->base.sw_i2cs[i] != NULL) {
+			kfree(pool->base.sw_i2cs[i]);
+			pool->base.sw_i2cs[i] = NULL;
+		}
+	}
+
+	for (i = 0; i < pool->base.stream_enc_count; i++) {
+		if (pool->base.stream_enc[i] != NULL)
+			kfree(DCE110STRENC_FROM_STRENC(pool->base.stream_enc[i]));
+	}
+
+	for (i = 0; i < pool->base.clk_src_count; i++) {
+		if (pool->base.clock_sources[i] != NULL) {
+			dce60_clock_source_destroy(&pool->base.clock_sources[i]);
+		}
+	}
+
+	if (pool->base.abm != NULL)
+			dce_abm_destroy(&pool->base.abm);
+
+	if (pool->base.dmcu != NULL)
+			dce_dmcu_destroy(&pool->base.dmcu);
+
+	if (pool->base.dp_clock_source != NULL)
+		dce60_clock_source_destroy(&pool->base.dp_clock_source);
+
+	for (i = 0; i < pool->base.audio_count; i++)	{
+		if (pool->base.audios[i] != NULL) {
+			dce_aud_destroy(&pool->base.audios[i]);
+		}
+	}
+
+	if (pool->base.irqs != NULL) {
+		dal_irq_service_destroy(&pool->base.irqs);
+	}
+}
+
+bool dce60_validate_bandwidth(
+	struct dc *dc,
+	struct dc_state *context,
+	bool fast_validate)
+{
+	int i;
+	bool at_least_one_pipe = false;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (context->res_ctx.pipe_ctx[i].stream)
+			at_least_one_pipe = true;
+	}
+
+	if (at_least_one_pipe) {
+		/* TODO implement when needed but for now hardcode max value*/
+		context->bw_ctx.bw.dce.dispclk_khz = 681000;
+		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
+	} else {
+		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		context->bw_ctx.bw.dce.yclk_khz = 0;
+	}
+
+	return true;
+}
+
+static bool dce60_validate_surface_sets(
+		struct dc_state *context)
+{
+	int i;
+
+	for (i = 0; i < context->stream_count; i++) {
+		if (context->stream_status[i].plane_count == 0)
+			continue;
+
+		if (context->stream_status[i].plane_count > 1)
+			return false;
+
+		if (context->stream_status[i].plane_states[0]->format
+				>= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
+			return false;
+	}
+
+	return true;
+}
+
+enum dc_status dce60_validate_global(
+		struct dc *dc,
+		struct dc_state *context)
+{
+	if (!dce60_validate_surface_sets(context))
+		return DC_FAIL_SURFACE_VALIDATE;
+
+	return DC_OK;
+}
+
+static void dce60_destroy_resource_pool(struct resource_pool **pool)
+{
+	struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool);
+
+	dce60_resource_destruct(dce110_pool);
+	kfree(dce110_pool);
+	*pool = NULL;
+}
+
+static const struct resource_funcs dce60_res_pool_funcs = {
+	.destroy = dce60_destroy_resource_pool,
+	.link_enc_create = dce60_link_encoder_create,
+	.panel_cntl_create = dce60_panel_cntl_create,
+	.validate_bandwidth = dce60_validate_bandwidth,
+	.validate_plane = dce100_validate_plane,
+	.add_stream_to_ctx = dce100_add_stream_to_ctx,
+	.validate_global = dce60_validate_global,
+	.find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link
+};
+
+static bool dce60_construct(
+	uint8_t num_virtual_links,
+	struct dc *dc,
+	struct dce110_resource_pool *pool)
+{
+	unsigned int i;
+	struct dc_context *ctx = dc->ctx;
+	struct dc_bios *bp;
+
+	ctx->dc_bios->regs = &bios_regs;
+
+	pool->base.res_cap = &res_cap;
+	pool->base.funcs = &dce60_res_pool_funcs;
+
+
+	/*************************************************
+	 *  Resource + asic cap harcoding                *
+	 *************************************************/
+	pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE;
+	pool->base.pipe_count = res_cap.num_timing_generator;
+	pool->base.timing_generator_count = res_cap.num_timing_generator;
+	dc->caps.max_downscale_ratio = 200;
+	dc->caps.i2c_speed_in_khz = 40;
+	dc->caps.max_cursor_size = 64;
+	dc->caps.dual_link_dvi = true;
+	dc->caps.extended_aux_timeout_support = false;
+
+	/*************************************************
+	 *  Create resources                             *
+	 *************************************************/
+
+	bp = ctx->dc_bios;
+
+	if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], false);
+		pool->base.clock_sources[1] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clk_src_count = 2;
+
+	} else {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clk_src_count = 1;
+	}
+
+	if (pool->base.dp_clock_source == NULL) {
+		dm_error("DC: failed to create dp clock source!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.clk_src_count; i++) {
+		if (pool->base.clock_sources[i] == NULL) {
+			dm_error("DC: failed to create clock sources!\n");
+			BREAK_TO_DEBUGGER();
+			goto res_create_fail;
+		}
+	}
+
+	pool->base.dmcu = dce_dmcu_create(ctx,
+			&dmcu_regs,
+			&dmcu_shift,
+			&dmcu_mask);
+	if (pool->base.dmcu == NULL) {
+		dm_error("DC: failed to create dmcu!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	pool->base.abm = dce_abm_create(ctx,
+			&abm_regs,
+			&abm_shift,
+			&abm_mask);
+	if (pool->base.abm == NULL) {
+		dm_error("DC: failed to create abm!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	{
+		struct irq_service_init_data init_data;
+		init_data.ctx = dc->ctx;
+		pool->base.irqs = dal_irq_service_dce60_create(&init_data);
+		if (!pool->base.irqs)
+			goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.pipe_count; i++) {
+		pool->base.timing_generators[i] = dce60_timing_generator_create(
+				ctx, i, &dce60_tg_offsets[i]);
+		if (pool->base.timing_generators[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create tg!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.mis[i] = dce60_mem_input_create(ctx, i);
+		if (pool->base.mis[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create memory input!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.ipps[i] = dce60_ipp_create(ctx, i);
+		if (pool->base.ipps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create input pixel processor!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.transforms[i] = dce60_transform_create(ctx, i);
+		if (pool->base.transforms[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create transform!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.opps[i] = dce60_opp_create(ctx, i);
+		if (pool->base.opps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create output pixel processor!\n");
+			goto res_create_fail;
+		}
+	}
+
+	for (i = 0; i < pool->base.res_cap->num_ddc; i++) {
+		pool->base.engines[i] = dce60_aux_engine_create(ctx, i);
+		if (pool->base.engines[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create aux engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.hw_i2cs[i] = dce60_i2c_hw_create(ctx, i);
+		if (pool->base.hw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create i2c engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.sw_i2cs[i] = dce60_i2c_sw_create(ctx);
+		if (pool->base.sw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create sw i2c!!\n");
+			goto res_create_fail;
+		}
+	}
+
+	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
+	dc->caps.disable_dp_clk_share = true;
+
+	if (!resource_construct(num_virtual_links, dc, &pool->base,
+			&res_create_funcs))
+		goto res_create_fail;
+
+	/* Create hardware sequencer */
+	dce60_hw_sequencer_construct(dc);
+
+	return true;
+
+res_create_fail:
+	dce60_resource_destruct(pool);
+	return false;
+}
+
+struct resource_pool *dce60_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc)
+{
+	struct dce110_resource_pool *pool =
+		kzalloc(sizeof(struct dce110_resource_pool), GFP_KERNEL);
+
+	if (!pool)
+		return NULL;
+
+	if (dce60_construct(num_virtual_links, dc, pool))
+		return &pool->base;
+
+	BREAK_TO_DEBUGGER();
+	return NULL;
+}
+
+static bool dce61_construct(
+	uint8_t num_virtual_links,
+	struct dc *dc,
+	struct dce110_resource_pool *pool)
+{
+	unsigned int i;
+	struct dc_context *ctx = dc->ctx;
+	struct dc_bios *bp;
+
+	ctx->dc_bios->regs = &bios_regs;
+
+	pool->base.res_cap = &res_cap_61;
+	pool->base.funcs = &dce60_res_pool_funcs;
+
+
+	/*************************************************
+	 *  Resource + asic cap harcoding                *
+	 *************************************************/
+	pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE;
+	pool->base.pipe_count = res_cap_61.num_timing_generator;
+	pool->base.timing_generator_count = res_cap_61.num_timing_generator;
+	dc->caps.max_downscale_ratio = 200;
+	dc->caps.i2c_speed_in_khz = 40;
+	dc->caps.max_cursor_size = 64;
+	dc->caps.is_apu = true;
+
+	/*************************************************
+	 *  Create resources                             *
+	 *************************************************/
+
+	bp = ctx->dc_bios;
+
+	if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], false);
+		pool->base.clock_sources[1] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clock_sources[2] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
+		pool->base.clk_src_count = 3;
+
+	} else {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clock_sources[1] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
+		pool->base.clk_src_count = 2;
+	}
+
+	if (pool->base.dp_clock_source == NULL) {
+		dm_error("DC: failed to create dp clock source!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.clk_src_count; i++) {
+		if (pool->base.clock_sources[i] == NULL) {
+			dm_error("DC: failed to create clock sources!\n");
+			BREAK_TO_DEBUGGER();
+			goto res_create_fail;
+		}
+	}
+
+	pool->base.dmcu = dce_dmcu_create(ctx,
+			&dmcu_regs,
+			&dmcu_shift,
+			&dmcu_mask);
+	if (pool->base.dmcu == NULL) {
+		dm_error("DC: failed to create dmcu!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	pool->base.abm = dce_abm_create(ctx,
+			&abm_regs,
+			&abm_shift,
+			&abm_mask);
+	if (pool->base.abm == NULL) {
+		dm_error("DC: failed to create abm!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	{
+		struct irq_service_init_data init_data;
+		init_data.ctx = dc->ctx;
+		pool->base.irqs = dal_irq_service_dce60_create(&init_data);
+		if (!pool->base.irqs)
+			goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.pipe_count; i++) {
+		pool->base.timing_generators[i] = dce60_timing_generator_create(
+				ctx, i, &dce60_tg_offsets[i]);
+		if (pool->base.timing_generators[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create tg!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.mis[i] = dce60_mem_input_create(ctx, i);
+		if (pool->base.mis[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create memory input!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.ipps[i] = dce60_ipp_create(ctx, i);
+		if (pool->base.ipps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create input pixel processor!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.transforms[i] = dce60_transform_create(ctx, i);
+		if (pool->base.transforms[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create transform!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.opps[i] = dce60_opp_create(ctx, i);
+		if (pool->base.opps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create output pixel processor!\n");
+			goto res_create_fail;
+		}
+	}
+
+	for (i = 0; i < pool->base.res_cap->num_ddc; i++) {
+		pool->base.engines[i] = dce60_aux_engine_create(ctx, i);
+		if (pool->base.engines[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create aux engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.hw_i2cs[i] = dce60_i2c_hw_create(ctx, i);
+		if (pool->base.hw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create i2c engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.sw_i2cs[i] = dce60_i2c_sw_create(ctx);
+		if (pool->base.sw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create sw i2c!!\n");
+			goto res_create_fail;
+		}
+	}
+
+	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
+	dc->caps.disable_dp_clk_share = true;
+
+	if (!resource_construct(num_virtual_links, dc, &pool->base,
+			&res_create_funcs))
+		goto res_create_fail;
+
+	/* Create hardware sequencer */
+	dce60_hw_sequencer_construct(dc);
+
+	return true;
+
+res_create_fail:
+	dce60_resource_destruct(pool);
+	return false;
+}
+
+struct resource_pool *dce61_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc)
+{
+	struct dce110_resource_pool *pool =
+		kzalloc(sizeof(struct dce110_resource_pool), GFP_KERNEL);
+
+	if (!pool)
+		return NULL;
+
+	if (dce61_construct(num_virtual_links, dc, pool))
+		return &pool->base;
+
+	BREAK_TO_DEBUGGER();
+	return NULL;
+}
+
+static bool dce64_construct(
+	uint8_t num_virtual_links,
+	struct dc *dc,
+	struct dce110_resource_pool *pool)
+{
+	unsigned int i;
+	struct dc_context *ctx = dc->ctx;
+	struct dc_bios *bp;
+
+	ctx->dc_bios->regs = &bios_regs;
+
+	pool->base.res_cap = &res_cap_64;
+	pool->base.funcs = &dce60_res_pool_funcs;
+
+
+	/*************************************************
+	 *  Resource + asic cap harcoding                *
+	 *************************************************/
+	pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE;
+	pool->base.pipe_count = res_cap_64.num_timing_generator;
+	pool->base.timing_generator_count = res_cap_64.num_timing_generator;
+	dc->caps.max_downscale_ratio = 200;
+	dc->caps.i2c_speed_in_khz = 40;
+	dc->caps.max_cursor_size = 64;
+	dc->caps.is_apu = true;
+
+	/*************************************************
+	 *  Create resources                             *
+	 *************************************************/
+
+	bp = ctx->dc_bios;
+
+	if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[0], false);
+		pool->base.clock_sources[1] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[1], false);
+		pool->base.clk_src_count = 2;
+
+	} else {
+		pool->base.dp_clock_source =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[0], true);
+
+		pool->base.clock_sources[0] =
+				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[1], false);
+		pool->base.clk_src_count = 1;
+	}
+
+	if (pool->base.dp_clock_source == NULL) {
+		dm_error("DC: failed to create dp clock source!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.clk_src_count; i++) {
+		if (pool->base.clock_sources[i] == NULL) {
+			dm_error("DC: failed to create clock sources!\n");
+			BREAK_TO_DEBUGGER();
+			goto res_create_fail;
+		}
+	}
+
+	pool->base.dmcu = dce_dmcu_create(ctx,
+			&dmcu_regs,
+			&dmcu_shift,
+			&dmcu_mask);
+	if (pool->base.dmcu == NULL) {
+		dm_error("DC: failed to create dmcu!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	pool->base.abm = dce_abm_create(ctx,
+			&abm_regs,
+			&abm_shift,
+			&abm_mask);
+	if (pool->base.abm == NULL) {
+		dm_error("DC: failed to create abm!\n");
+		BREAK_TO_DEBUGGER();
+		goto res_create_fail;
+	}
+
+	{
+		struct irq_service_init_data init_data;
+		init_data.ctx = dc->ctx;
+		pool->base.irqs = dal_irq_service_dce60_create(&init_data);
+		if (!pool->base.irqs)
+			goto res_create_fail;
+	}
+
+	for (i = 0; i < pool->base.pipe_count; i++) {
+		pool->base.timing_generators[i] = dce60_timing_generator_create(
+				ctx, i, &dce60_tg_offsets[i]);
+		if (pool->base.timing_generators[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create tg!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.mis[i] = dce60_mem_input_create(ctx, i);
+		if (pool->base.mis[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create memory input!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.ipps[i] = dce60_ipp_create(ctx, i);
+		if (pool->base.ipps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create input pixel processor!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.transforms[i] = dce60_transform_create(ctx, i);
+		if (pool->base.transforms[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create transform!\n");
+			goto res_create_fail;
+		}
+
+		pool->base.opps[i] = dce60_opp_create(ctx, i);
+		if (pool->base.opps[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error("DC: failed to create output pixel processor!\n");
+			goto res_create_fail;
+		}
+	}
+
+	for (i = 0; i < pool->base.res_cap->num_ddc; i++) {
+		pool->base.engines[i] = dce60_aux_engine_create(ctx, i);
+		if (pool->base.engines[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create aux engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.hw_i2cs[i] = dce60_i2c_hw_create(ctx, i);
+		if (pool->base.hw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create i2c engine!!\n");
+			goto res_create_fail;
+		}
+		pool->base.sw_i2cs[i] = dce60_i2c_sw_create(ctx);
+		if (pool->base.sw_i2cs[i] == NULL) {
+			BREAK_TO_DEBUGGER();
+			dm_error(
+				"DC:failed to create sw i2c!!\n");
+			goto res_create_fail;
+		}
+	}
+
+	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
+	dc->caps.disable_dp_clk_share = true;
+
+	if (!resource_construct(num_virtual_links, dc, &pool->base,
+			&res_create_funcs))
+		goto res_create_fail;
+
+	/* Create hardware sequencer */
+	dce60_hw_sequencer_construct(dc);
+
+	return true;
+
+res_create_fail:
+	dce60_resource_destruct(pool);
+	return false;
+}
+
+struct resource_pool *dce64_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc)
+{
+	struct dce110_resource_pool *pool =
+		kzalloc(sizeof(struct dce110_resource_pool), GFP_KERNEL);
+
+	if (!pool)
+		return NULL;
+
+	if (dce64_construct(num_virtual_links, dc, pool))
+		return &pool->base;
+
+	BREAK_TO_DEBUGGER();
+	return NULL;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.h b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.h
new file mode 100644
index 000000000000..5d653a76b0b0
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.h
@@ -0,0 +1,47 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DC_RESOURCE_DCE60_H__
+#define __DC_RESOURCE_DCE60_H__
+
+#include "core_types.h"
+
+struct dc;
+struct resource_pool;
+
+struct resource_pool *dce60_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc);
+
+struct resource_pool *dce61_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc);
+
+struct resource_pool *dce64_create_resource_pool(
+	uint8_t num_virtual_links,
+	struct dc *dc);
+
+#endif /* __DC_RESOURCE_DCE60_H__ */
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.c
new file mode 100644
index 000000000000..fc1af0ff0ca4
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.c
@@ -0,0 +1,266 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include "dm_services.h"
+
+/* include DCE6 register header files */
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#include "dc_types.h"
+
+#include "include/grph_object_id.h"
+#include "include/logger_interface.h"
+#include "../dce110/dce110_timing_generator.h"
+#include "dce60_timing_generator.h"
+
+#include "timing_generator.h"
+
+enum black_color_format {
+	BLACK_COLOR_FORMAT_RGB_FULLRANGE = 0,	/* used as index in array */
+	BLACK_COLOR_FORMAT_RGB_LIMITED,
+	BLACK_COLOR_FORMAT_YUV_TV,
+	BLACK_COLOR_FORMAT_YUV_CV,
+	BLACK_COLOR_FORMAT_YUV_SUPER_AA,
+
+	BLACK_COLOR_FORMAT_COUNT
+};
+
+static const struct dce110_timing_generator_offsets reg_offsets[] = {
+{
+	.crtc = (mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP0_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+},
+{
+	.crtc = (mmCRTC1_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP1_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+},
+{
+	.crtc = (mmCRTC2_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP2_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+},
+{
+	.crtc = (mmCRTC3_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP3_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+},
+{
+	.crtc = (mmCRTC4_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP4_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+},
+{
+	.crtc = (mmCRTC5_DCFE_MEM_LIGHT_SLEEP_CNTL - mmCRTC0_DCFE_MEM_LIGHT_SLEEP_CNTL),
+	.dcp = (mmDCP5_GRPH_CONTROL - mmDCP0_GRPH_CONTROL),
+}
+};
+
+#define NUMBER_OF_FRAME_TO_WAIT_ON_TRIGGERED_RESET 10
+
+#define MAX_H_TOTAL (CRTC_H_TOTAL__CRTC_H_TOTAL_MASK + 1)
+#define MAX_V_TOTAL (CRTC_V_TOTAL__CRTC_V_TOTAL_MASKhw + 1)
+
+#define CRTC_REG(reg) (reg + tg110->offsets.crtc)
+#define DCP_REG(reg) (reg + tg110->offsets.dcp)
+#define DMIF_REG(reg) (reg + tg110->offsets.dmif)
+
+static void program_pix_dur(struct timing_generator *tg, uint32_t pix_clk_100hz)
+{
+	uint64_t pix_dur;
+	uint32_t addr = mmDMIF_PG0_DPG_PIPE_ARBITRATION_CONTROL1
+					+ DCE110TG_FROM_TG(tg)->offsets.dmif;
+	uint32_t value = dm_read_reg(tg->ctx, addr);
+
+	if (pix_clk_100hz == 0)
+		return;
+
+	pix_dur = div_u64(10000000000ull, pix_clk_100hz);
+
+	set_reg_field_value(
+		value,
+		pix_dur,
+		DPG_PIPE_ARBITRATION_CONTROL1,
+		PIXEL_DURATION);
+
+	dm_write_reg(tg->ctx, addr, value);
+}
+
+static void program_timing(struct timing_generator *tg,
+	const struct dc_crtc_timing *timing,
+	int vready_offset,
+	int vstartup_start,
+	int vupdate_offset,
+	int vupdate_width,
+	const enum signal_type signal,
+	bool use_vbios)
+{
+	if (!use_vbios)
+		program_pix_dur(tg, timing->pix_clk_100hz);
+
+	dce110_tg_program_timing(tg, timing, 0, 0, 0, 0, 0, use_vbios);
+}
+
+static void dce60_timing_generator_enable_advanced_request(
+	struct timing_generator *tg,
+	bool enable,
+	const struct dc_crtc_timing *timing)
+{
+	struct dce110_timing_generator *tg110 = DCE110TG_FROM_TG(tg);
+	uint32_t addr = CRTC_REG(mmCRTC_START_LINE_CONTROL);
+	uint32_t value = dm_read_reg(tg->ctx, addr);
+	/* DCE6 has CRTC_PREFETCH_EN bit in CRTC_CONTROL register */
+	uint32_t addr2 = CRTC_REG(mmCRTC_CONTROL);
+	uint32_t value2 = dm_read_reg(tg->ctx, addr2);
+
+	/* DCE6 does not support CRTC_LEGACY_REQUESTOR_EN bit
+	   so here is not possible to set bit based on enable argument */
+
+	if ((timing->v_sync_width + timing->v_front_porch) <= 3) {
+		set_reg_field_value(
+			value,
+			3,
+			CRTC_START_LINE_CONTROL,
+			CRTC_ADVANCED_START_LINE_POSITION);
+		set_reg_field_value(
+			value2,
+			0,
+			CRTC_CONTROL,
+			CRTC_PREFETCH_EN);
+	} else {
+		set_reg_field_value(
+			value,
+			4,
+			CRTC_START_LINE_CONTROL,
+			CRTC_ADVANCED_START_LINE_POSITION);
+		set_reg_field_value(
+			value2,
+			1,
+			CRTC_CONTROL,
+			CRTC_PREFETCH_EN);
+	}
+
+	set_reg_field_value(
+		value,
+		1,
+		CRTC_START_LINE_CONTROL,
+		CRTC_PROGRESSIVE_START_LINE_EARLY);
+
+	set_reg_field_value(
+		value,
+		1,
+		CRTC_START_LINE_CONTROL,
+		CRTC_INTERLACE_START_LINE_EARLY);
+
+	dm_write_reg(tg->ctx, addr, value);
+	dm_write_reg(tg->ctx, addr2, value2);
+}
+
+static bool dce60_is_tg_enabled(struct timing_generator *tg)
+{
+	uint32_t addr = 0;
+	uint32_t value = 0;
+	uint32_t field = 0;
+	struct dce110_timing_generator *tg110 = DCE110TG_FROM_TG(tg);
+
+	addr = CRTC_REG(mmCRTC_CONTROL);
+	value = dm_read_reg(tg->ctx, addr);
+	field = get_reg_field_value(value, CRTC_CONTROL,
+				    CRTC_CURRENT_MASTER_EN_STATE);
+	return field == 1;
+}
+
+bool dce60_configure_crc(struct timing_generator *tg,
+			  const struct crc_params *params)
+{
+	/* Cannot configure crc on a CRTC that is disabled */
+	if (!dce60_is_tg_enabled(tg))
+		return false;
+
+	/* DCE6 has no CRTC_CRC_CNTL register, nothing to do */
+
+	return true;
+}
+
+static const struct timing_generator_funcs dce60_tg_funcs = {
+		.validate_timing = dce110_tg_validate_timing,
+		.program_timing = program_timing,
+		.enable_crtc = dce110_timing_generator_enable_crtc,
+		.disable_crtc = dce110_timing_generator_disable_crtc,
+		.is_counter_moving = dce110_timing_generator_is_counter_moving,
+		.get_position = dce110_timing_generator_get_position,
+		.get_frame_count = dce110_timing_generator_get_vblank_counter,
+		.get_scanoutpos = dce110_timing_generator_get_crtc_scanoutpos,
+		.set_early_control = dce110_timing_generator_set_early_control,
+		.wait_for_state = dce110_tg_wait_for_state,
+		.set_blank = dce110_tg_set_blank,
+		.is_blanked = dce110_tg_is_blanked,
+		.set_colors = dce110_tg_set_colors,
+		.set_overscan_blank_color =
+				dce110_timing_generator_set_overscan_color_black,
+		.set_blank_color = dce110_timing_generator_program_blank_color,
+		.disable_vga = dce110_timing_generator_disable_vga,
+		.did_triggered_reset_occur =
+				dce110_timing_generator_did_triggered_reset_occur,
+		.setup_global_swap_lock =
+				dce110_timing_generator_setup_global_swap_lock,
+		.enable_reset_trigger = dce110_timing_generator_enable_reset_trigger,
+		.disable_reset_trigger = dce110_timing_generator_disable_reset_trigger,
+		.tear_down_global_swap_lock =
+				dce110_timing_generator_tear_down_global_swap_lock,
+		.set_drr = dce110_timing_generator_set_drr,
+		.set_static_screen_control =
+			dce110_timing_generator_set_static_screen_control,
+		.set_test_pattern = dce110_timing_generator_set_test_pattern,
+		.arm_vert_intr = dce110_arm_vert_intr,
+
+		/* DCE6.0 overrides */
+		.enable_advanced_request =
+				dce60_timing_generator_enable_advanced_request,
+		.configure_crc = dce60_configure_crc,
+		.get_crc = dce110_get_crc,
+};
+
+void dce60_timing_generator_construct(
+	struct dce110_timing_generator *tg110,
+	struct dc_context *ctx,
+	uint32_t instance,
+	const struct dce110_timing_generator_offsets *offsets)
+{
+	tg110->controller_id = CONTROLLER_ID_D0 + instance;
+	tg110->base.inst = instance;
+	tg110->offsets = *offsets;
+	tg110->derived_offsets = reg_offsets[instance];
+
+	tg110->base.funcs = &dce60_tg_funcs;
+
+	tg110->base.ctx = ctx;
+	tg110->base.bp = ctx->dc_bios;
+
+	tg110->max_h_total = CRTC_H_TOTAL__CRTC_H_TOTAL_MASK + 1;
+	tg110->max_v_total = CRTC_V_TOTAL__CRTC_V_TOTAL_MASK + 1;
+
+	tg110->min_h_blank = 56;
+	tg110->min_h_front_porch = 4;
+	tg110->min_h_back_porch = 4;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.h b/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.h
new file mode 100644
index 000000000000..81d831233cc5
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_timing_generator.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DC_TIMING_GENERATOR_DCE60_H__
+#define __DC_TIMING_GENERATOR_DCE60_H__
+
+#include "timing_generator.h"
+#include "../include/grph_object_id.h"
+
+/* DCE6.0 implementation inherits from DCE11.0 */
+void dce60_timing_generator_construct(
+	struct dce110_timing_generator *tg,
+	struct dc_context *ctx,
+	uint32_t instance,
+	const struct dce110_timing_generator_offsets *offsets);
+
+#endif /* __DC_TIMING_GENERATOR_DCE60_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
index 62ad1a11bff9..733e6e6e43bd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
@@ -31,4 +31,11 @@ DCN10 = dcn10_init.o dcn10_resource.o dcn10_ipp.o dcn10_hw_sequencer.o \
 
 AMD_DAL_DCN10 = $(addprefix $(AMDDALPATH)/dc/dcn10/,$(DCN10))
 
+# fix:
+# ...: '-mgeneral-regs-only' is incompatible with the use of floating-point types
+# aarch64 does not support soft-float, so use hard-float and handle this in code
+ifdef CONFIG_ARM64
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dcn10/dcn10_resource.o := -mgeneral-regs-only
+endif
+
 AMD_DISPLAY_FILES += $(AMD_DAL_DCN10)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
index 47a39eb9400b..7a00fe525dfb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
@@ -325,8 +325,6 @@ bool cm_helper_translate_curve_to_hw_format(
 	if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS)
 		return false;
 
-	PERF_TRACE_CTX(output_tf->ctx);
-
 	corner_points = lut_params->corner_points;
 	rgb_resulted = lut_params->rgb_resulted;
 	hw_points = 0;
@@ -524,8 +522,6 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
 	if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS)
 		return false;
 
-	PERF_TRACE_CTX(output_tf->ctx);
-
 	corner_points = lut_params->corner_points;
 	rgb_resulted = lut_params->rgb_resulted;
 	hw_points = 0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
index cedf359a00f5..db5615a51fea 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
@@ -734,6 +734,9 @@ bool hubp1_is_flip_pending(struct hubp *hubp)
 	struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp);
 	struct dc_plane_address earliest_inuse_address;
 
+	if (hubp && hubp->power_gated)
+		return false;
+
 	REG_GET(DCSURF_FLIP_CONTROL,
 			SURFACE_FLIP_PENDING, &flip_pending);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index fa643ec5a876..d0f3bf953d02 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -2377,14 +2377,6 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 				&blnd_cfg.black_color);
 	}
 
-	/*
-	 * The way 420 is packed, 2 channels carry Y component, 1 channel
-	 * alternate between Cb and Cr, so both channels need the pixel
-	 * value for Y
-	 */
-	if (pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		blnd_cfg.black_color.color_r_cr = blnd_cfg.black_color.color_g_y;
-
 	if (per_pixel_alpha)
 		blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA;
 	else
@@ -2769,6 +2761,154 @@ static struct pipe_ctx *dcn10_find_top_pipe_for_stream(
 	return NULL;
 }
 
+bool dcn10_disconnect_pipes(
+		struct dc *dc,
+		struct dc_state *context)
+{
+		bool found_pipe = false;
+		int i, j;
+		struct dce_hwseq *hws = dc->hwseq;
+		struct dc_state *old_ctx = dc->current_state;
+		bool mpcc_disconnected = false;
+		struct pipe_ctx *old_pipe;
+		struct pipe_ctx *new_pipe;
+		DC_LOGGER_INIT(dc->ctx->logger);
+
+		/* Set pipe update flags and lock pipes */
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+			new_pipe = &context->res_ctx.pipe_ctx[i];
+			new_pipe->update_flags.raw = 0;
+
+			if (!old_pipe->plane_state && !new_pipe->plane_state)
+				continue;
+
+			if (old_pipe->plane_state && !new_pipe->plane_state)
+				new_pipe->update_flags.bits.disable = 1;
+
+			/* Check for scl update */
+			if (memcmp(&old_pipe->plane_res.scl_data, &new_pipe->plane_res.scl_data, sizeof(struct scaler_data)))
+					new_pipe->update_flags.bits.scaler = 1;
+
+			/* Check for vp update */
+			if (memcmp(&old_pipe->plane_res.scl_data.viewport, &new_pipe->plane_res.scl_data.viewport, sizeof(struct rect))
+					|| memcmp(&old_pipe->plane_res.scl_data.viewport_c,
+						&new_pipe->plane_res.scl_data.viewport_c, sizeof(struct rect)))
+				new_pipe->update_flags.bits.viewport = 1;
+
+		}
+
+		if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
+			/* Disconnect mpcc here only if losing pipe split*/
+			for (i = 0; i < dc->res_pool->pipe_count; i++) {
+				if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable &&
+					old_ctx->res_ctx.pipe_ctx[i].top_pipe) {
+
+					/* Find the top pipe in the new ctx for the bottom pipe that we
+					 * want to remove by comparing the streams and planes. If both
+					 * pipes are being disabled then do it in the regular pipe
+					 * programming sequence
+					 */
+					for (j = 0; j < dc->res_pool->pipe_count; j++) {
+						if (old_ctx->res_ctx.pipe_ctx[i].top_pipe->stream == context->res_ctx.pipe_ctx[j].stream &&
+							old_ctx->res_ctx.pipe_ctx[i].top_pipe->plane_state == context->res_ctx.pipe_ctx[j].plane_state &&
+							!context->res_ctx.pipe_ctx[j].top_pipe &&
+							!context->res_ctx.pipe_ctx[j].update_flags.bits.disable) {
+							found_pipe = true;
+							break;
+						}
+					}
+
+					// Disconnect if the top pipe lost it's pipe split
+					if (found_pipe && !context->res_ctx.pipe_ctx[j].bottom_pipe) {
+						hws->funcs.plane_atomic_disconnect(dc, &dc->current_state->res_ctx.pipe_ctx[i]);
+						DC_LOG_DC("Reset mpcc for pipe %d\n", dc->current_state->res_ctx.pipe_ctx[i].pipe_idx);
+						mpcc_disconnected = true;
+					}
+				}
+				found_pipe = false;
+			}
+		}
+
+		if (mpcc_disconnected) {
+			for (i = 0; i < dc->res_pool->pipe_count; i++) {
+				struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+				struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+				struct dc_plane_state *plane_state = pipe_ctx->plane_state;
+				struct hubp *hubp = pipe_ctx->plane_res.hubp;
+
+				if (!pipe_ctx || !plane_state || !pipe_ctx->stream)
+					continue;
+
+				// Only update scaler and viewport here if we lose a pipe split.
+				// This is to prevent half the screen from being black when we
+				// unlock after disconnecting MPCC.
+				if (!(old_pipe && !pipe_ctx->top_pipe &&
+					!pipe_ctx->bottom_pipe && old_pipe->bottom_pipe))
+					continue;
+
+				if (pipe_ctx->update_flags.raw || pipe_ctx->plane_state->update_flags.raw || pipe_ctx->stream->update_flags.raw) {
+					if (pipe_ctx->update_flags.bits.scaler ||
+						plane_state->update_flags.bits.scaling_change ||
+						plane_state->update_flags.bits.position_change ||
+						plane_state->update_flags.bits.per_pixel_alpha_change ||
+						pipe_ctx->stream->update_flags.bits.scaling) {
+
+						pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->plane_state->per_pixel_alpha;
+						ASSERT(pipe_ctx->plane_res.scl_data.lb_params.depth == LB_PIXEL_DEPTH_30BPP);
+						/* scaler configuration */
+						pipe_ctx->plane_res.dpp->funcs->dpp_set_scaler(
+						pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data);
+					}
+
+					if (pipe_ctx->update_flags.bits.viewport ||
+						(context == dc->current_state && plane_state->update_flags.bits.position_change) ||
+						(context == dc->current_state && plane_state->update_flags.bits.scaling_change) ||
+						(context == dc->current_state && pipe_ctx->stream->update_flags.bits.scaling)) {
+
+						hubp->funcs->mem_program_viewport(
+							hubp,
+							&pipe_ctx->plane_res.scl_data.viewport,
+							&pipe_ctx->plane_res.scl_data.viewport_c);
+					}
+				}
+			}
+		}
+	return mpcc_disconnected;
+}
+
+void dcn10_wait_for_pending_cleared(struct dc *dc,
+		struct dc_state *context)
+{
+		struct pipe_ctx *pipe_ctx;
+		struct timing_generator *tg;
+		int i;
+
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			pipe_ctx = &context->res_ctx.pipe_ctx[i];
+			tg = pipe_ctx->stream_res.tg;
+
+			/*
+			 * Only wait for top pipe's tg penindg bit
+			 * Also skip if pipe is disabled.
+			 */
+			if (pipe_ctx->top_pipe ||
+			    !pipe_ctx->stream || !pipe_ctx->plane_state ||
+			    !tg->funcs->is_tg_enabled(tg))
+				continue;
+
+			/*
+			 * Wait for VBLANK then VACTIVE to ensure we get VUPDATE.
+			 * For some reason waiting for OTG_UPDATE_PENDING cleared
+			 * seems to not trigger the update right away, and if we
+			 * lock again before VUPDATE then we don't get a separated
+			 * operation.
+			 */
+			pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, CRTC_STATE_VBLANK);
+			pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, CRTC_STATE_VACTIVE);
+		}
+}
+
 void dcn10_apply_ctx_for_surface(
 		struct dc *dc,
 		const struct dc_stream_state *stream,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
index 6d891166da8a..e5691e499023 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
@@ -194,6 +194,12 @@ void dcn10_get_surface_visual_confirm_color(
 void dcn10_get_hdr_visual_confirm_color(
 		struct pipe_ctx *pipe_ctx,
 		struct tg_color *color);
+bool dcn10_disconnect_pipes(
+		struct dc *dc,
+		struct dc_state *context);
+
+void dcn10_wait_for_pending_cleared(struct dc *dc,
+		struct dc_state *context);
 void dcn10_set_hdr_multiplier(struct pipe_ctx *pipe_ctx);
 void dcn10_verify_allow_pstate_change_high(struct dc *dc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c
index 5c98b71c1d47..b24c8ae8b1ec 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c
@@ -34,6 +34,8 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = dcn10_apply_ctx_for_surface,
 	.post_unlock_program_front_end = dcn10_post_unlock_program_front_end,
+	.disconnect_pipes = dcn10_disconnect_pipes,
+	.wait_for_pending_cleared = dcn10_wait_for_pending_cleared,
 	.update_plane_addr = dcn10_update_plane_addr,
 	.update_dchub = dcn10_update_dchub,
 	.update_pending_status = dcn10_update_pending_status,
@@ -64,6 +66,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
 	.get_hw_state = dcn10_get_hw_state,
 	.clear_status_bits = dcn10_clear_status_bits,
 	.wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect,
+	.edp_backlight_control = dce110_edp_backlight_control,
 	.edp_power_control = dce110_edp_power_control,
 	.edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready,
 	.set_cursor_position = dcn10_set_cursor_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index 2972392f9788..800be2693fac 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -288,6 +288,17 @@ void optc1_program_timing(
 	if (optc1_is_two_pixels_per_containter(&patched_crtc_timing) || optc1->opp_count == 2)
 		h_div = H_TIMING_DIV_BY2;
 
+	if (REG(OPTC_DATA_FORMAT_CONTROL)) {
+		uint32_t data_fmt = 0;
+
+		if (patched_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR422)
+			data_fmt = 1;
+		else if (patched_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)
+			data_fmt = 2;
+
+		REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DATA_FORMAT, data_fmt);
+	}
+
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	if (optc1->tg_mask->OTG_H_TIMING_DIV_MODE != 0) {
 		if (optc1->opp_count == 4)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index 8939541ad7af..a78712caf124 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -798,7 +798,7 @@ static const struct encoder_feature_support link_enc_feature = {
 		.max_hdmi_deep_color = COLOR_DEPTH_121212,
 		.max_hdmi_pixel_clock = 600000,
 		.hdmi_ycbcr420_supported = true,
-		.dp_ycbcr420_supported = false,
+		.dp_ycbcr420_supported = true,
 		.flags.bits.IS_HBR2_CAPABLE = true,
 		.flags.bits.IS_HBR3_CAPABLE = true,
 		.flags.bits.IS_TPS3_CAPABLE = true,
@@ -1339,6 +1339,47 @@ static uint32_t read_pipe_fuses(struct dc_context *ctx)
 	return value;
 }
 
+/*
+ * Some architectures don't support soft-float (e.g. aarch64), on those
+ * this function has to be called with hardfloat enabled, make sure not
+ * to inline it so whatever fp stuff is done stays inside
+ */
+static noinline void dcn10_resource_construct_fp(
+	struct dc *dc)
+{
+	if (dc->ctx->dce_version == DCN_VERSION_1_01) {
+		struct dcn_soc_bounding_box *dcn_soc = dc->dcn_soc;
+		struct dcn_ip_params *dcn_ip = dc->dcn_ip;
+		struct display_mode_lib *dml = &dc->dml;
+
+		dml->ip.max_num_dpp = 3;
+		/* TODO how to handle 23.84? */
+		dcn_soc->dram_clock_change_latency = 23;
+		dcn_ip->max_num_dpp = 3;
+	}
+	if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) {
+		dc->dcn_soc->urgent_latency = 3;
+		dc->debug.disable_dmcu = true;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 41.60f;
+	}
+
+
+	dc->dcn_soc->number_of_channels = dc->ctx->asic_id.vram_width / ddr4_dram_width;
+	ASSERT(dc->dcn_soc->number_of_channels < 3);
+	if (dc->dcn_soc->number_of_channels == 0)/*old sbios bug*/
+		dc->dcn_soc->number_of_channels = 2;
+
+	if (dc->dcn_soc->number_of_channels == 1) {
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 19.2f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = 17.066f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = 14.933f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = 12.8f;
+		if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) {
+			dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 20.80f;
+		}
+	}
+}
+
 static bool dcn10_resource_construct(
 	uint8_t num_virtual_links,
 	struct dc *dc,
@@ -1490,37 +1531,15 @@ static bool dcn10_resource_construct(
 	memcpy(dc->dcn_ip, &dcn10_ip_defaults, sizeof(dcn10_ip_defaults));
 	memcpy(dc->dcn_soc, &dcn10_soc_defaults, sizeof(dcn10_soc_defaults));
 
-	if (dc->ctx->dce_version == DCN_VERSION_1_01) {
-		struct dcn_soc_bounding_box *dcn_soc = dc->dcn_soc;
-		struct dcn_ip_params *dcn_ip = dc->dcn_ip;
-		struct display_mode_lib *dml = &dc->dml;
-
-		dml->ip.max_num_dpp = 3;
-		/* TODO how to handle 23.84? */
-		dcn_soc->dram_clock_change_latency = 23;
-		dcn_ip->max_num_dpp = 3;
-	}
-	if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) {
-		dc->dcn_soc->urgent_latency = 3;
-		dc->debug.disable_dmcu = true;
-		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 41.60f;
-	}
-
-
-	dc->dcn_soc->number_of_channels = dc->ctx->asic_id.vram_width / ddr4_dram_width;
-	ASSERT(dc->dcn_soc->number_of_channels < 3);
-	if (dc->dcn_soc->number_of_channels == 0)/*old sbios bug*/
-		dc->dcn_soc->number_of_channels = 2;
-
-	if (dc->dcn_soc->number_of_channels == 1) {
-		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 19.2f;
-		dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = 17.066f;
-		dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = 14.933f;
-		dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = 12.8f;
-		if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) {
-			dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 20.80f;
-		}
-	}
+#if defined(CONFIG_ARM64)
+	/* Aarch64 does not support -msoft-float/-mfloat-abi=soft */
+	DC_FP_START();
+	dcn10_resource_construct_fp(dc);
+	DC_FP_END();
+#else
+	/* Other architectures we build for build this with soft-float */
+	dcn10_resource_construct_fp(dc);
+#endif
 
 	pool->base.pp_smu = dcn10_pp_smu_create(ctx);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
index 842abb4c475b..f70fcadf1ee5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
@@ -619,7 +619,7 @@ void enc1_stream_encoder_dvi_set_stream_attribute(
 	enc1_stream_encoder_set_stream_attribute_helper(enc1, crtc_timing);
 }
 
-void enc1_stream_encoder_set_mst_bandwidth(
+void enc1_stream_encoder_set_throttled_vcp_size(
 	struct stream_encoder *enc,
 	struct fixed31_32 avg_time_slots_per_mtp)
 {
@@ -896,10 +896,10 @@ void enc1_stream_encoder_dp_blank(
 	 */
 	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_DIS_DEFER, 2);
 	/* Larger delay to wait until VBLANK - use max retry of
-	 * 10us*5000=50ms. This covers 41.7ms of minimum 24 Hz mode +
+	 * 10us*10200=102ms. This covers 100.0ms of minimum 10 Hz mode +
 	 * a little more because we may not trust delay accuracy.
 	 */
-	max_retries = DP_BLANK_MAX_RETRY * 250;
+	max_retries = DP_BLANK_MAX_RETRY * 501;
 
 	/* disable DP stream */
 	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, 0);
@@ -1616,8 +1616,8 @@ static const struct stream_encoder_funcs dcn10_str_enc_funcs = {
 		enc1_stream_encoder_hdmi_set_stream_attribute,
 	.dvi_set_stream_attribute =
 		enc1_stream_encoder_dvi_set_stream_attribute,
-	.set_mst_bandwidth =
-		enc1_stream_encoder_set_mst_bandwidth,
+	.set_throttled_vcp_size =
+		enc1_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =
 		enc1_stream_encoder_update_hdmi_info_packets,
 	.stop_hdmi_info_packets =
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
index 30eae7459d50..b99d2527cf03 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
@@ -588,7 +588,7 @@ void enc1_stream_encoder_dvi_set_stream_attribute(
 	struct dc_crtc_timing *crtc_timing,
 	bool is_dual_link);
 
-void enc1_stream_encoder_set_mst_bandwidth(
+void enc1_stream_encoder_set_throttled_vcp_size(
 	struct stream_encoder *enc,
 	struct fixed31_32 avg_time_slots_per_mtp);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
index 5fcaf78334ff..624cb1341ef1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
@@ -17,6 +17,10 @@ ifdef CONFIG_PPC64
 CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec
 endif
 
+ifdef CONFIG_ARM64
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mgeneral-regs-only
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 IS_OLD_GCC = 1
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
index 667640c4b288..1118e33aaa2c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
@@ -94,6 +94,7 @@
 	DSC_SF(DSC_TOP0_DSC_TOP_CONTROL, DSC_CLOCK_EN, mask_sh), \
 	DSC_SF(DSC_TOP0_DSC_TOP_CONTROL, DSC_DISPCLK_R_GATE_DIS, mask_sh), \
 	DSC_SF(DSC_TOP0_DSC_TOP_CONTROL, DSC_DSCCLK_R_GATE_DIS, mask_sh), \
+	DSC_SF(DSC_TOP0_DSC_DEBUG_CONTROL, DSC_DBG_EN, mask_sh), \
 	DSC_SF(DSCC0_DSCC_CONFIG0, ICH_RESET_AT_END_OF_LINE, mask_sh), \
 	DSC_SF(DSCC0_DSCC_CONFIG0, NUMBER_OF_SLICES_PER_LINE, mask_sh), \
 	DSC_SF(DSCC0_DSCC_CONFIG0, ALTERNATE_ICH_ENCODING_EN, mask_sh), \
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c
index bb920d0e0b89..368818d2dfc6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c
@@ -908,6 +908,9 @@ bool hubp2_is_flip_pending(struct hubp *hubp)
 	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
 	struct dc_plane_address earliest_inuse_address;
 
+	if (hubp && hubp->power_gated)
+		return false;
+
 	REG_GET(DCSURF_FLIP_CONTROL,
 			SURFACE_FLIP_PENDING, &flip_pending);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index c8cfd3ba1c15..01530e686f43 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -1251,6 +1251,11 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx
 		return;
 	}
 
+	/* Detect plane change */
+	if (old_pipe->plane_state != new_pipe->plane_state) {
+		new_pipe->update_flags.bits.plane_changed = true;
+	}
+
 	/* Detect top pipe only changes */
 	if (!new_pipe->top_pipe && !new_pipe->prev_odm_pipe) {
 		/* Detect odm changes */
@@ -1392,6 +1397,7 @@ static void dcn20_update_dchubp_dpp(
 			&pipe_ctx->ttu_regs);
 
 	if (pipe_ctx->update_flags.bits.enable ||
+			pipe_ctx->update_flags.bits.plane_changed ||
 			plane_state->update_flags.bits.bpp_change ||
 			plane_state->update_flags.bits.input_csc_change ||
 			plane_state->update_flags.bits.color_space_change ||
@@ -1414,6 +1420,7 @@ static void dcn20_update_dchubp_dpp(
 	}
 
 	if (pipe_ctx->update_flags.bits.mpcc
+			|| pipe_ctx->update_flags.bits.plane_changed
 			|| plane_state->update_flags.bits.global_alpha_change
 			|| plane_state->update_flags.bits.per_pixel_alpha_change) {
 		// MPCC inst is equal to pipe index in practice
@@ -1515,6 +1522,7 @@ static void dcn20_update_dchubp_dpp(
 	}
 
 	if (pipe_ctx->update_flags.bits.enable ||
+			pipe_ctx->update_flags.bits.plane_changed ||
 			pipe_ctx->update_flags.bits.opp_changed ||
 			plane_state->update_flags.bits.pixel_format_change ||
 			plane_state->update_flags.bits.horizontal_mirror_change ||
@@ -1539,7 +1547,9 @@ static void dcn20_update_dchubp_dpp(
 		hubp->power_gated = false;
 	}
 
-	if (pipe_ctx->update_flags.bits.enable || plane_state->update_flags.bits.addr_update)
+	if (pipe_ctx->update_flags.bits.enable ||
+		pipe_ctx->update_flags.bits.plane_changed ||
+		plane_state->update_flags.bits.addr_update)
 		hws->funcs.update_plane_addr(dc, pipe_ctx);
 
 
@@ -1632,16 +1642,26 @@ void dcn20_program_front_end_for_ctx(
 	struct dce_hwseq *hws = dc->hwseq;
 	DC_LOGGER_INIT(dc->ctx->logger);
 
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+	/* Carry over GSL groups in case the context is changing. */
+       for (i = 0; i < dc->res_pool->pipe_count; i++) {
+               struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+               struct pipe_ctx *old_pipe_ctx =
+                       &dc->current_state->res_ctx.pipe_ctx[i];
+
+               if (pipe_ctx->stream == old_pipe_ctx->stream)
+                       pipe_ctx->stream_res.gsl_group =
+                               old_pipe_ctx->stream_res.gsl_group;
+       }
+
+	if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) {
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
-		if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->plane_state) {
-			ASSERT(!pipe_ctx->plane_state->triplebuffer_flips);
-			if (dc->hwss.program_triplebuffer != NULL &&
-				!dc->debug.disable_tri_buf) {
+			if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->plane_state) {
+				ASSERT(!pipe_ctx->plane_state->triplebuffer_flips);
 				/*turn off triple buffer for full update*/
 				dc->hwss.program_triplebuffer(
-					dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips);
+						dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips);
 			}
 		}
 	}
@@ -1909,9 +1929,9 @@ void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	if (pipe_ctx->stream_res.dsc) {
 		struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe;
 
-		dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, true);
+		hws->funcs.dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, true);
 		while (odm_pipe) {
-			dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, true);
+			hws->funcs.dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, true);
 			odm_pipe = odm_pipe->next_odm_pipe;
 		}
 	}
@@ -1924,9 +1944,9 @@ void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	if (pipe_ctx->stream_res.dsc) {
 		struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe;
 
-		dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, false);
+		hws->funcs.dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, false);
 		while (odm_pipe) {
-			dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, false);
+			hws->funcs.dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, false);
 			odm_pipe = odm_pipe->next_odm_pipe;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
index 3dde6f26de47..072193c5ffe6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
@@ -34,6 +34,8 @@ static const struct hw_sequencer_funcs dcn20_funcs = {
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = NULL,
 	.program_front_end_for_ctx = dcn20_program_front_end_for_ctx,
+	.disconnect_pipes = dcn10_disconnect_pipes,
+	.wait_for_pending_cleared = dcn10_wait_for_pending_cleared,
 	.post_unlock_program_front_end = dcn20_post_unlock_program_front_end,
 	.update_plane_addr = dcn20_update_plane_addr,
 	.update_dchub = dcn10_update_dchub,
@@ -66,6 +68,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = {
 	.get_hw_state = dcn10_get_hw_state,
 	.clear_status_bits = dcn10_clear_status_bits,
 	.wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect,
+	.edp_backlight_control = dce110_edp_backlight_control,
 	.edp_power_control = dce110_edp_power_control,
 	.edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready,
 	.set_cursor_position = dcn10_set_cursor_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h
index dcbf28dd72d4..864acd695cbb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h
@@ -231,8 +231,6 @@
 	SRI(RDPCSTX_PHY_FUSE3, RDPCSTX, id), \
 	SRI(DPCSTX_TX_CLOCK_CNTL, DPCSTX, id), \
 	SRI(DPCSTX_TX_CNTL, DPCSTX, id), \
-	SRI(DPCSTX_DEBUG_CONFIG, DPCSTX, id), \
-	SRI(RDPCSTX_DEBUG_CONFIG, RDPCSTX, id), \
 	SR(RDPCSTX0_RDPCSTX_SCRATCH)
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
index 8c16967fe018..d8b18c515d06 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
@@ -239,7 +239,6 @@ void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_c
 	int mpcc_hactive = (timing->h_addressable + timing->h_border_left + timing->h_border_right)
 			/ opp_cnt;
 	uint32_t memory_mask;
-	uint32_t data_fmt = 0;
 
 	ASSERT(opp_cnt == 2);
 
@@ -262,13 +261,6 @@ void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_c
 		REG_SET(OPTC_MEMORY_CONFIG, 0,
 			OPTC_MEM_SEL, memory_mask);
 
-	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-		data_fmt = 1;
-	else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		data_fmt = 2;
-
-	REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DATA_FORMAT, data_fmt);
-
 	REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0,
 			OPTC_NUM_OF_INPUT_SEGMENT, 1,
 			OPTC_SEG0_SRC_SEL, opp_id[0],
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index f31f48dd0da2..d50a9c370637 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -150,7 +150,6 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = {
 	.dispclk_delay_subtotal = 87, //
 	.dcfclk_cstate_latency = 10, // SRExitTime
 	.max_inter_dcn_tile_repeaters = 8,
-
 	.xfc_supported = true,
 	.xfc_fill_bw_overhead_percent = 10.0,
 	.xfc_fill_constant_bytes = 0,
@@ -298,8 +297,8 @@ static struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = {
 			},
 		},
 	.num_states = 5,
-	.sr_exit_time_us = 8.6,
-	.sr_enter_plus_exit_time_us = 10.9,
+	.sr_exit_time_us = 11.6,
+	.sr_enter_plus_exit_time_us = 13.9,
 	.urgent_latency_us = 4.0,
 	.urgent_latency_pixel_data_only_us = 4.0,
 	.urgent_latency_pixel_mixed_with_vm_data_us = 4.0,
@@ -1075,7 +1074,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.disable_pplib_wm_range = false,
 		.scl_reset_length10 = true,
 		.sanity_checks = false,
-		.disable_tri_buf = true,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
 };
 
@@ -1092,6 +1090,7 @@ static const struct dc_debug_options debug_defaults_diags = {
 		.disable_stutter = true,
 		.scl_reset_length10 = true,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
+		.enable_tri_buf = true,
 };
 
 void dcn20_dpp_destroy(struct dpp **dpp)
@@ -2203,9 +2202,9 @@ int dcn20_populate_dml_pipes_from_context(
 		/* todo: default max for now, until there is logic reflecting this in dc*/
 		pipes[pipe_cnt].dout.output_bpc = 12;
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
-		/*fill up the audio sample rate*/
+		/*fill up the audio sample rate (unit in kHz)*/
 		get_audio_check(&res_ctx->pipe_ctx[i].stream->audio_info, &aud_check);
-		pipes[pipe_cnt].dout.max_audio_sample_rate = aud_check.max_audiosample_rate;
+		pipes[pipe_cnt].dout.max_audio_sample_rate = aud_check.max_audiosample_rate / 1000;
 #endif
 		/*
 		 * For graphic plane, cursor number is 1, nv12 is 0
@@ -3209,6 +3208,9 @@ static noinline bool dcn20_validate_bandwidth_fp(struct dc *dc,
 	context->bw_ctx.dml.soc.allow_dram_clock_one_display_vactive =
 		dc->debug.enable_dram_clock_change_one_display_vactive;
 
+	/*Unsafe due to current pipe merge and split logic*/
+	ASSERT(context != dc->current_state);
+
 	if (fast_validate) {
 		return dcn20_validate_bandwidth_internal(dc, context, true);
 	}
@@ -3320,7 +3322,7 @@ enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_stat
 	return DC_OK;
 }
 
-static struct resource_funcs dcn20_res_pool_funcs = {
+static const struct resource_funcs dcn20_res_pool_funcs = {
 	.destroy = dcn20_destroy_resource_pool,
 	.link_enc_create = dcn20_link_encoder_create,
 	.panel_cntl_create = dcn20_panel_cntl_create,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
index 2c1959845c29..cdd39ee9761d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h
@@ -95,7 +95,6 @@ struct display_stream_compressor *dcn20_dsc_create(
 	struct dc_context *ctx, uint32_t inst);
 void dcn20_dsc_destroy(struct display_stream_compressor **dsc);
 
-void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb);
 void dcn20_cap_soc_clocks(
 		struct _vcs_dpi_soc_bounding_box_st *bb,
 		struct pp_smu_nv_clock_table max_clocks);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
index e3984f02b7b3..4075ae111530 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c
@@ -561,8 +561,8 @@ static const struct stream_encoder_funcs dcn20_str_enc_funcs = {
 		enc1_stream_encoder_hdmi_set_stream_attribute,
 	.dvi_set_stream_attribute =
 		enc1_stream_encoder_dvi_set_stream_attribute,
-	.set_mst_bandwidth =
-		enc1_stream_encoder_set_mst_bandwidth,
+	.set_throttled_vcp_size =
+		enc1_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =
 		enc2_stream_encoder_update_hdmi_info_packets,
 	.stop_hdmi_info_packets =
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
index 07684d3e375a..51a2f3d4c194 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
@@ -13,6 +13,10 @@ ifdef CONFIG_PPC64
 CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec
 endif
 
+ifdef CONFIG_ARM64
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mgeneral-regs-only
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 IS_OLD_GCC = 1
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
index b187f71afa65..2b7396c9fcb4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
@@ -35,6 +35,8 @@ static const struct hw_sequencer_funcs dcn21_funcs = {
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = NULL,
 	.program_front_end_for_ctx = dcn20_program_front_end_for_ctx,
+	.disconnect_pipes = dcn10_disconnect_pipes,
+	.wait_for_pending_cleared = dcn10_wait_for_pending_cleared,
 	.post_unlock_program_front_end = dcn20_post_unlock_program_front_end,
 	.update_plane_addr = dcn20_update_plane_addr,
 	.update_dchub = dcn10_update_dchub,
@@ -67,6 +69,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = {
 	.get_hw_state = dcn10_get_hw_state,
 	.clear_status_bits = dcn10_clear_status_bits,
 	.wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect,
+	.edp_backlight_control = dce110_edp_backlight_control,
 	.edp_power_control = dce110_edp_power_control,
 	.edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready,
 	.set_cursor_position = dcn10_set_cursor_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index 88d41a385add..e73785e74cba 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -894,6 +894,8 @@ static const struct dc_debug_options debug_defaults_diags = {
 		.disable_pplib_wm_range = true,
 		.disable_stutter = true,
 		.disable_48mhz_pwrdwn = true,
+		.disable_psr = true,
+		.enable_tri_buf = true
 };
 
 enum dcn20_clk_src_array_id {
@@ -1184,6 +1186,9 @@ bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context,
 
 	BW_VAL_TRACE_COUNT();
 
+	/*Unsafe due to current pipe merge and split logic*/
+	ASSERT(context != dc->current_state);
+
 	out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel);
 
 	if (pipe_cnt == 0)
@@ -1754,7 +1759,7 @@ enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_stat
 	return result;
 }
 
-static struct resource_funcs dcn21_res_pool_funcs = {
+static const struct resource_funcs dcn21_res_pool_funcs = {
 	.destroy = dcn21_destroy_resource_pool,
 	.link_enc_create = dcn21_link_encoder_create,
 	.panel_cntl_create = dcn21_panel_cntl_create,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c
index a139a87a1a81..41a1d0e9b7e2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c
@@ -122,8 +122,6 @@ bool cm3_helper_translate_curve_to_hw_format(
 	if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS)
 		return false;
 
-	PERF_TRACE_CTX(output_tf->ctx);
-
 	corner_points = lut_params->corner_points;
 	rgb_resulted = lut_params->rgb_resulted;
 	hw_points = 0;
@@ -314,8 +312,6 @@ bool cm3_helper_translate_curve_to_degamma_hw_format(
 	if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS)
 		return false;
 
-	PERF_TRACE_CTX(output_tf->ctx);
-
 	corner_points = lut_params->corner_points;
 	rgb_resulted = lut_params->rgb_resulted;
 	hw_points = 0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
index f5e80a0db72b..6c0f7ef0a3df 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
@@ -790,8 +790,8 @@ static const struct stream_encoder_funcs dcn30_str_enc_funcs = {
 		enc3_stream_encoder_hdmi_set_stream_attribute,
 	.dvi_set_stream_attribute =
 		enc3_stream_encoder_dvi_set_stream_attribute,
-	.set_mst_bandwidth =
-		enc1_stream_encoder_set_mst_bandwidth,
+	.set_throttled_vcp_size =
+		enc1_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =
 		enc3_stream_encoder_update_hdmi_info_packets,
 	.stop_hdmi_info_packets =
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index a5d750ed569e..204773ffc376 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -35,7 +35,6 @@
 #include "dcn30_dpp.h"
 #include "dcn10/dcn10_cm_common.h"
 #include "dcn30_cm_common.h"
-#include "clk_mgr.h"
 #include "reg_helper.h"
 #include "abm.h"
 #include "clk_mgr.h"
@@ -220,15 +219,13 @@ static void dcn30_set_writeback(
 		struct dc_writeback_info *wb_info,
 		struct dc_state *context)
 {
-	struct dwbc *dwb;
 	struct mcif_wb *mcif_wb;
 	struct mcif_buf_params *mcif_buf_params;
 
 	ASSERT(wb_info->dwb_pipe_inst < MAX_DWB_PIPES);
 	ASSERT(wb_info->wb_enabled);
 	ASSERT(wb_info->mpcc_inst >= 0);
-	ASSERT(wb_info->mpcc_inst < 4);
-	dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst];
+	ASSERT(wb_info->mpcc_inst < dc->res_pool->mpcc_count);
 	mcif_wb = dc->res_pool->mcif_wb[wb_info->dwb_pipe_inst];
 	mcif_buf_params = &wb_info->mcif_buf_params;
 
@@ -692,26 +689,23 @@ void dcn30_program_dmdata_engine(struct pipe_ctx *pipe_ctx)
 
 bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable)
 {
-	unsigned int surface_size;
-
 	if (!dc->ctx->dmub_srv)
 		return false;
 
 	if (enable) {
-		if (dc->current_state
-				&& dc->current_state->stream_count == 1 // single display only
-				&& dc->current_state->stream_status[0].plane_count == 1 // single surface only
-				&& dc->current_state->stream_status[0].plane_states[0]->address.page_table_base.quad_part == 0 // no VM
-				// Only 8 and 16 bit formats
-				&& dc->current_state->stream_status[0].plane_states[0]->format <= SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F
-				&& dc->current_state->stream_status[0].plane_states[0]->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB8888) {
-
-			surface_size = dc->current_state->stream_status[0].plane_states[0]->plane_size.surface_pitch *
-					dc->current_state->stream_status[0].plane_states[0]->plane_size.surface_size.height *
-					(dc->current_state->stream_status[0].plane_states[0]->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 ? 8 : 4);
-
+		if (dc->current_state) {
+			int i;
+
+			/* First, check no-memory-requests case */
+			for (i = 0; i < dc->current_state->stream_count; i++) {
+				if (dc->current_state->stream_status[i]
+					    .plane_count)
+					/* Fail eligibility on a visible stream */
+					break;
+			}
 		}
 
+		/* No applicable optimizations */
 		return false;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
index 9afee7160490..7c90c2222506 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
@@ -35,6 +35,8 @@ static const struct hw_sequencer_funcs dcn30_funcs = {
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = NULL,
 	.program_front_end_for_ctx = dcn20_program_front_end_for_ctx,
+	.disconnect_pipes = dcn10_disconnect_pipes,
+	.wait_for_pending_cleared = dcn10_wait_for_pending_cleared,
 	.post_unlock_program_front_end = dcn20_post_unlock_program_front_end,
 	.update_plane_addr = dcn20_update_plane_addr,
 	.update_dchub = dcn10_update_dchub,
@@ -67,6 +69,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = {
 	.get_hw_state = dcn10_get_hw_state,
 	.clear_status_bits = dcn10_clear_status_bits,
 	.wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect,
+	.edp_backlight_control = dce110_edp_backlight_control,
 	.edp_power_control = dce110_edp_power_control,
 	.edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready,
 	.set_cursor_position = dcn10_set_cursor_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
index 224c8d145eba..b1f228fc119a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
@@ -179,8 +179,7 @@ void optc3_set_dsc_config(struct timing_generator *optc,
 
 }
 
-
-static void optc3_set_odm_bypass(struct timing_generator *optc,
+void optc3_set_odm_bypass(struct timing_generator *optc,
 		const struct dc_crtc_timing *dc_crtc_timing)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
@@ -210,7 +209,6 @@ static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, in
 	int mpcc_hactive = (timing->h_addressable + timing->h_border_left + timing->h_border_right)
 			/ opp_cnt;
 	uint32_t memory_mask = 0;
-	uint32_t data_fmt = 0;
 
 	/* TODO: In pseudocode but does not affect maximus, delete comment if we dont need on asic
 	 * REG_SET(OTG_GLOBAL_CONTROL2, 0, GLOBAL_UPDATE_LOCK_EN, 1);
@@ -241,13 +239,6 @@ static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, in
 		REG_SET(OPTC_MEMORY_CONFIG, 0,
 			OPTC_MEM_SEL, memory_mask);
 
-	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-		data_fmt = 1;
-	else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		data_fmt = 2;
-
-	REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DATA_FORMAT, data_fmt);
-
 	if (opp_cnt == 2) {
 		REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0,
 				OPTC_NUM_OF_INPUT_SEGMENT, 1,
@@ -277,7 +268,7 @@ static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, in
  *
  * Options: any time,  start of frame, dp start of frame (range timing)
  */
-void optc3_set_timing_double_buffer(struct timing_generator *optc, bool enable)
+static void optc3_set_timing_double_buffer(struct timing_generator *optc, bool enable)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
 	uint32_t mode = enable ? 2 : 0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h
index 33f13c1e7520..379616831636 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h
@@ -339,4 +339,8 @@ void optc3_set_dsc_config(struct timing_generator *optc,
 
 void optc3_set_timing_db_mode(struct timing_generator *optc, bool enable);
 
+void optc3_set_odm_bypass(struct timing_generator *optc,
+		const struct dc_crtc_timing *dc_crtc_timing);
+void optc3_tg_init(struct timing_generator *optc);
+
 #endif /* __DC_OPTC_DCN30_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index ebe0cc5b833b..24fb39a11e5d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -79,6 +79,7 @@
 
 #include "reg_helper.h"
 #include "dce/dmub_abm.h"
+#include "dce/dmub_psr.h"
 #include "dce/dce_aux.h"
 #include "dce/dce_i2c.h"
 
@@ -340,7 +341,7 @@ static const struct dce110_clk_src_mask cs_mask = {
 
 #define abm_regs(id)\
 [id] = {\
-		ABM_DCN301_REG_LIST(id)\
+		ABM_DCN30_REG_LIST(id)\
 }
 
 static const struct dce_abm_registers abm_regs[] = {
@@ -491,6 +492,7 @@ static const struct dcn10_link_enc_hpd_registers link_enc_hpd_regs[] = {
 [id] = {\
 	LE_DCN3_REG_LIST(id), \
 	UNIPHY_DCN2_REG_LIST(phyid), \
+	DPCS_DCN2_REG_LIST(id), \
 	SRI(DP_DPHY_INTERNAL_CTRL, DP, id) \
 }
 
@@ -831,7 +833,7 @@ static const struct dc_plane_cap plane_cap = {
 };
 
 static const struct dc_debug_options debug_defaults_drv = {
-	.disable_dmcu = true,
+	.disable_dmcu = true, //No DMCU on DCN30
 	.force_abm_enable = false,
 	.timing_trace = false,
 	.clock_trace = true,
@@ -848,10 +850,11 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.underflow_assert_delay_us = 0xFFFFFFFF,
 	.dwb_fi_phase = -1, // -1 = disable,
 	.dmub_command_table = true,
+	.disable_psr = false,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
+	.disable_dmcu = true, //No dmcu on DCN30
 	.force_abm_enable = false,
 	.timing_trace = true,
 	.clock_trace = true,
@@ -864,6 +867,8 @@ static const struct dc_debug_options debug_defaults_diags = {
 	.scl_reset_length10 = true,
 	.dwb_fi_phase = -1, // -1 = disable
 	.dmub_command_table = true,
+	.disable_psr = true,
+	.enable_tri_buf = true,
 };
 
 void dcn30_dpp_destroy(struct dpp **dpp)
@@ -872,7 +877,7 @@ void dcn30_dpp_destroy(struct dpp **dpp)
 	*dpp = NULL;
 }
 
-struct dpp *dcn30_dpp_create(
+static struct dpp *dcn30_dpp_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -890,7 +895,8 @@ struct dpp *dcn30_dpp_create(
 	kfree(dpp);
 	return NULL;
 }
-struct output_pixel_processor *dcn30_opp_create(
+
+static struct output_pixel_processor *dcn30_opp_create(
 	struct dc_context *ctx, uint32_t inst)
 {
 	struct dcn20_opp *opp =
@@ -906,7 +912,7 @@ struct output_pixel_processor *dcn30_opp_create(
 	return &opp->base;
 }
 
-struct dce_aux *dcn30_aux_engine_create(
+static struct dce_aux *dcn30_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -925,6 +931,7 @@ struct dce_aux *dcn30_aux_engine_create(
 
 	return &aux_engine->base;
 }
+
 #define i2c_inst_regs(id) { I2C_HW_ENGINE_COMMON_REG_LIST(id) }
 
 static const struct dce_i2c_registers i2c_hw_regs[] = {
@@ -944,7 +951,7 @@ static const struct dce_i2c_mask i2c_masks = {
 		I2C_COMMON_MASK_SH_LIST_DCN2(_MASK)
 };
 
-struct dce_i2c_hw *dcn30_i2c_hw_create(
+static struct dce_i2c_hw *dcn30_i2c_hw_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -959,6 +966,7 @@ struct dce_i2c_hw *dcn30_i2c_hw_create(
 
 	return dce_i2c_hw;
 }
+
 static struct mpc *dcn30_mpc_create(
 		struct dc_context *ctx,
 		int num_mpcc,
@@ -1009,7 +1017,7 @@ struct hubbub *dcn30_hubbub_create(struct dc_context *ctx)
 	return &hubbub3->base;
 }
 
-struct timing_generator *dcn30_timing_generator_create(
+static struct timing_generator *dcn30_timing_generator_create(
 		struct dc_context *ctx,
 		uint32_t instance)
 {
@@ -1043,7 +1051,7 @@ static const struct encoder_feature_support link_enc_feature = {
 		.flags.bits.IS_TPS4_CAPABLE = true
 };
 
-struct link_encoder *dcn30_link_encoder_create(
+static struct link_encoder *dcn30_link_encoder_create(
 	const struct encoder_init_data *enc_init_data)
 {
 	struct dcn20_link_encoder *enc20 =
@@ -1064,7 +1072,7 @@ struct link_encoder *dcn30_link_encoder_create(
 	return &enc20->enc10.base;
 }
 
-struct panel_cntl *dcn30_panel_cntl_create(const struct panel_cntl_init_data *init_data)
+static struct panel_cntl *dcn30_panel_cntl_create(const struct panel_cntl_init_data *init_data)
 {
 	struct dce_panel_cntl *panel_cntl =
 		kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL);
@@ -1308,11 +1316,14 @@ static void dcn30_resource_destruct(struct dcn30_resource_pool *pool)
 			dce_abm_destroy(&pool->base.multiple_abms[i]);
 	}
 
+	if (pool->base.psr != NULL)
+		dmub_psr_destroy(&pool->base.psr);
+
 	if (pool->base.dccg != NULL)
 		dcn_dccg_destroy(&pool->base.dccg);
 }
 
-struct hubp *dcn30_hubp_create(
+static struct hubp *dcn30_hubp_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -1331,7 +1342,7 @@ struct hubp *dcn30_hubp_create(
 	return NULL;
 }
 
-bool dcn30_dwbc_create(struct dc_context *ctx, struct resource_pool *pool)
+static bool dcn30_dwbc_create(struct dc_context *ctx, struct resource_pool *pool)
 {
 	int i;
 	uint32_t pipe_count = pool->res_cap->num_dwb;
@@ -1356,7 +1367,7 @@ bool dcn30_dwbc_create(struct dc_context *ctx, struct resource_pool *pool)
 	return true;
 }
 
-bool dcn30_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool)
+static bool dcn30_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool)
 {
 	int i;
 	uint32_t pipe_count = pool->res_cap->num_dwb;
@@ -1817,6 +1828,22 @@ static bool init_soc_bounding_box(struct dc *dc,
 	loaded_ip->max_num_dpp = pool->base.pipe_count;
 	loaded_ip->clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
 	dcn20_patch_bounding_box(dc, loaded_bb);
+
+	if (!bb && dc->ctx->dc_bios->funcs->get_soc_bb_info) {
+		struct bp_soc_bb_info bb_info = {0};
+
+		if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dcn3_0_soc.dram_clock_change_latency_us = bb_info.dram_clock_change_latency_100ns * 10;
+
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dcn3_0_soc.sr_enter_plus_exit_time_us = bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dcn3_0_soc.sr_exit_time_us = bb_info.dram_sr_exit_latency_100ns * 10;
+		}
+	}
+
 	return true;
 }
 
@@ -1872,6 +1899,48 @@ static bool dcn30_split_stream_for_mpc_or_odm(
 	return true;
 }
 
+static struct pipe_ctx *dcn30_find_split_pipe(
+		struct dc *dc,
+		struct dc_state *context,
+		int old_index)
+{
+	struct pipe_ctx *pipe = NULL;
+	int i;
+
+	if (old_index >= 0 && context->res_ctx.pipe_ctx[old_index].stream == NULL) {
+		pipe = &context->res_ctx.pipe_ctx[old_index];
+		pipe->pipe_idx = old_index;
+	}
+
+	if (!pipe)
+		for (i = dc->res_pool->pipe_count - 1; i >= 0; i--) {
+			if (dc->current_state->res_ctx.pipe_ctx[i].top_pipe == NULL
+					&& dc->current_state->res_ctx.pipe_ctx[i].prev_odm_pipe == NULL) {
+				if (context->res_ctx.pipe_ctx[i].stream == NULL) {
+					pipe = &context->res_ctx.pipe_ctx[i];
+					pipe->pipe_idx = i;
+					break;
+				}
+			}
+		}
+
+	/*
+	 * May need to fix pipes getting tossed from 1 opp to another on flip
+	 * Add for debugging transient underflow during topology updates:
+	 * ASSERT(pipe);
+	 */
+	if (!pipe)
+		for (i = dc->res_pool->pipe_count - 1; i >= 0; i--) {
+			if (context->res_ctx.pipe_ctx[i].stream == NULL) {
+				pipe = &context->res_ctx.pipe_ctx[i];
+				pipe->pipe_idx = i;
+				break;
+			}
+		}
+
+	return pipe;
+}
+
 static bool dcn30_internal_validate_bw(
 		struct dc *dc,
 		struct dc_state *context,
@@ -1997,6 +2066,7 @@ static bool dcn30_internal_validate_bw(
 				dcn20_release_dsc(&context->res_ctx, dc->res_pool, &pipe->stream_res.dsc);
 			memset(&pipe->plane_res, 0, sizeof(pipe->plane_res));
 			memset(&pipe->stream_res, 0, sizeof(pipe->stream_res));
+			repopulate_pipes = true;
 		} else if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) {
 			struct pipe_ctx *top_pipe = pipe->top_pipe;
 			struct pipe_ctx *bottom_pipe = pipe->bottom_pipe;
@@ -2011,6 +2081,7 @@ static bool dcn30_internal_validate_bw(
 			pipe->stream = NULL;
 			memset(&pipe->plane_res, 0, sizeof(pipe->plane_res));
 			memset(&pipe->stream_res, 0, sizeof(pipe->stream_res));
+			repopulate_pipes = true;
 		} else
 			ASSERT(0); /* Should never try to merge master pipe */
 
@@ -2018,8 +2089,10 @@ static bool dcn30_internal_validate_bw(
 
 	for (i = 0, pipe_idx = -1; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+		struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
 		struct pipe_ctx *hsplit_pipe = NULL;
 		bool odm;
+		int old_index = -1;
 
 		if (!pipe->stream || newly_split[i])
 			continue;
@@ -2031,7 +2104,20 @@ static bool dcn30_internal_validate_bw(
 			continue;
 
 		if (split[i]) {
-			hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe);
+			if (odm) {
+				if (split[i] == 4 && old_pipe->next_odm_pipe->next_odm_pipe)
+					old_index = old_pipe->next_odm_pipe->next_odm_pipe->pipe_idx;
+				else if (old_pipe->next_odm_pipe)
+					old_index = old_pipe->next_odm_pipe->pipe_idx;
+			} else {
+				if (split[i] == 4 && old_pipe->bottom_pipe->bottom_pipe &&
+						old_pipe->bottom_pipe->bottom_pipe->plane_state == old_pipe->plane_state)
+					old_index = old_pipe->bottom_pipe->bottom_pipe->pipe_idx;
+				else if (old_pipe->bottom_pipe &&
+						old_pipe->bottom_pipe->plane_state == old_pipe->plane_state)
+					old_index = old_pipe->bottom_pipe->pipe_idx;
+			}
+			hsplit_pipe = dcn30_find_split_pipe(dc, context, old_index);
 			ASSERT(hsplit_pipe);
 			if (!hsplit_pipe)
 				goto validate_fail;
@@ -2045,8 +2131,16 @@ static bool dcn30_internal_validate_bw(
 			repopulate_pipes = true;
 		}
 		if (split[i] == 4) {
-			struct pipe_ctx *pipe_4to1 = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe);
+			struct pipe_ctx *pipe_4to1;
 
+			if (odm && old_pipe->next_odm_pipe)
+				old_index = old_pipe->next_odm_pipe->pipe_idx;
+			else if (!odm && old_pipe->bottom_pipe &&
+						old_pipe->bottom_pipe->plane_state == old_pipe->plane_state)
+				old_index = old_pipe->bottom_pipe->pipe_idx;
+			else
+				old_index = -1;
+			pipe_4to1 = dcn30_find_split_pipe(dc, context, old_index);
 			ASSERT(pipe_4to1);
 			if (!pipe_4to1)
 				goto validate_fail;
@@ -2056,7 +2150,14 @@ static bool dcn30_internal_validate_bw(
 				goto validate_fail;
 			newly_split[pipe_4to1->pipe_idx] = true;
 
-			pipe_4to1 = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe);
+			if (odm && old_pipe->next_odm_pipe->next_odm_pipe->next_odm_pipe)
+				old_index = old_pipe->next_odm_pipe->next_odm_pipe->next_odm_pipe->pipe_idx;
+			else if (!odm && old_pipe->bottom_pipe->bottom_pipe->bottom_pipe &&
+						old_pipe->bottom_pipe->bottom_pipe->bottom_pipe->plane_state == old_pipe->plane_state)
+				old_index = old_pipe->bottom_pipe->bottom_pipe->bottom_pipe->pipe_idx;
+			else
+				old_index = -1;
+			pipe_4to1 = dcn30_find_split_pipe(dc, context, old_index);
 			ASSERT(pipe_4to1);
 			if (!pipe_4to1)
 				goto validate_fail;
@@ -2100,7 +2201,7 @@ validate_out:
 	return out;
 }
 
-static void dcn30_calculate_wm(
+void dcn30_calculate_wm_and_dlg(
 		struct dc *dc, struct dc_state *context,
 		display_e2e_pipe_params_st *pipes,
 		int pipe_cnt,
@@ -2108,6 +2209,8 @@ static void dcn30_calculate_wm(
 {
 	int i, pipe_idx;
 	double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
+	bool pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] !=
+			dm_dram_clock_change_unsupported;
 
 	if (context->bw_ctx.dml.soc.min_dcfclk > dcfclk)
 		dcfclk = context->bw_ctx.dml.soc.min_dcfclk;
@@ -2141,30 +2244,12 @@ static void dcn30_calculate_wm(
 	pipes[0].clks_cfg.voltage = vlevel;
 	pipes[0].clks_cfg.dcfclk_mhz = dcfclk;
 
-	/* Set C:
-	 * DCFCLK: Min Required
-	 * FCLK(proportional to UCLK): 1GHz or Max
-	 * pstate latency overriden to 5us
-	 */
-	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) {
-		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us;
-		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us;
-		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us;
-	}
-	context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-
 	/* Set D:
 	 * DCFCLK: Min Required
 	 * FCLK(proportional to UCLK): 1GHz or Max
 	 * sr_enter_exit = 4, sr_exit = 2us
 	 */
+	/*
 	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid) {
 		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us;
 		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us;
@@ -2178,26 +2263,72 @@ static void dcn30_calculate_wm(
 	context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	*/
 
-	/* Set A:
+	/* Set C:
 	 * DCFCLK: Min Required
 	 * FCLK(proportional to UCLK): 1GHz or Max
-	 *
-	 * Set A calculated last so that following calculations are based on Set A
+	 * pstate latency overridden to 5us
 	 */
-	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) {
-		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
-		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
-		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
+	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) {
+		unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed;
+		unsigned int min_dram_speed_mts_margin = 160;
+
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us;
+
+		if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_unsupported)
+			min_dram_speed_mts = dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz * 16;
+
+		for (i = 3; i > 0; i--) {
+			if ((min_dram_speed_mts + min_dram_speed_mts_margin > dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts) &&
+					(min_dram_speed_mts - min_dram_speed_mts_margin < dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts))
+				context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[i].dummy_pstate_latency_us;
+		}
+
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us;
+		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us;
+	}
+	context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+
+	if (!pstate_en) {
+		/* The only difference between A and C is p-state latency, if p-state is not supported we want to
+		 * calculate DLG based on dummy p-state latency, and max out the set A p-state watermark
+		 */
+		context->bw_ctx.bw.dcn.watermarks.a = context->bw_ctx.bw.dcn.watermarks.c;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 0x13FFFF;
+	} else {
+		/* Set A:
+		 * DCFCLK: Min Required
+		 * FCLK(proportional to UCLK): 1GHz or Max
+		 *
+		 * Set A calculated last so that following calculations are based on Set A
+		 */
+		if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) {
+			context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
+			context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
+			context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
+		}
+		context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	}
-	context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+
+	context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod;
+
+	/* Make set D = set A until set D is enabled */
+	context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		if (!context->res_ctx.pipe_ctx[i].stream)
@@ -2217,6 +2348,13 @@ static void dcn30_calculate_wm(
 
 		pipe_idx++;
 	}
+
+	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+
+	if (!pstate_en)
+		/* Restore full p-state latency */
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+				dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
 }
 
 bool dcn30_validate_bandwidth(struct dc *dc,
@@ -2249,8 +2387,7 @@ bool dcn30_validate_bandwidth(struct dc *dc,
 		goto validate_out;
 	}
 
-	dcn30_calculate_wm(dc, context, pipes, pipe_cnt, vlevel);
-	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+	dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel);
 
 	BW_VAL_TRACE_END_WATERMARKS();
 
@@ -2293,7 +2430,7 @@ static void get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts,
                (dcn3_0_soc.return_bus_width_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100));
 }
 
-static void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
+void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
 {
 	unsigned int i, j;
 	unsigned int num_states = 0;
@@ -2413,14 +2550,16 @@ static void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw
 		dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_0_soc, &dcn3_0_ip, DML_PROJECT_DCN30);
 }
 
-static struct resource_funcs dcn30_res_pool_funcs = {
+static const struct resource_funcs dcn30_res_pool_funcs = {
 	.destroy = dcn30_destroy_resource_pool,
 	.link_enc_create = dcn30_link_encoder_create,
 	.panel_cntl_create = dcn30_panel_cntl_create,
 	.validate_bandwidth = dcn30_validate_bandwidth,
+	.calculate_wm_and_dlg = dcn30_calculate_wm_and_dlg,
 	.populate_dml_pipes = dcn30_populate_dml_pipes_from_context,
 	.acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer,
 	.add_stream_to_ctx = dcn30_add_stream_to_ctx,
+	.add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource,
 	.remove_stream_from_ctx = dcn20_remove_stream_from_ctx,
 	.populate_dml_writeback_from_context = dcn30_populate_dml_writeback_from_context,
 	.set_mcif_arb_params = dcn30_set_mcif_arb_params,
@@ -2618,6 +2757,14 @@ static bool dcn30_resource_construct(
 		}
 	}
 	pool->base.timing_generator_count = i;
+	/* PSR */
+	pool->base.psr = dmub_psr_create(ctx);
+
+	if (pool->base.psr == NULL) {
+		dm_error("DC: failed to create PSR obj!\n");
+		BREAK_TO_DEBUGGER();
+		goto create_fail;
+	}
 
 	/* ABM */
 	for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) {
@@ -2684,7 +2831,7 @@ static bool dcn30_resource_construct(
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
 			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
 			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
 	dcn30_hw_sequencer_construct(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
index 4b4a4d81c1e3..d163812af858 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
@@ -55,6 +55,11 @@ unsigned int dcn30_calc_max_scaled_time(
 
 bool dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context,
 		bool fast_validate);
+void dcn30_calculate_wm_and_dlg(
+		struct dc *dc, struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel);
 void dcn30_populate_dml_writeback_from_context(
 		struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes);
 
@@ -79,4 +84,7 @@ enum dc_status dcn30_add_stream_to_ctx(
 		struct dc *dc,
 		struct dc_state *new_ctx,
 		struct dc_stream_state *dc_stream);
+
+void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
+
 #endif /* _DCN30_RESOURCE_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
index ae608c329366..3586934df25f 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
@@ -30,8 +30,6 @@
  * interface to PPLIB/SMU to setup clocks and pstate requirements on SoC
  */
 
-typedef bool BOOLEAN;
-
 enum pp_smu_ver {
 	/*
 	 * PP_SMU_INTERFACE_X should be interpreted as the interface defined
@@ -240,7 +238,7 @@ struct pp_smu_funcs_nv {
 	 * DC hardware
 	 */
 	enum pp_smu_status (*set_pstate_handshake_support)(struct pp_smu *pp,
-			BOOLEAN pstate_handshake_supported);
+			bool pstate_handshake_supported);
 };
 
 #define PP_SMU_NUM_SOCCLK_DPM_LEVELS  8
diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile
index 417331438c30..dbc7e2abe379 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
@@ -33,6 +33,10 @@ ifdef CONFIG_PPC64
 dml_ccflags := -mhard-float -maltivec
 endif
 
+ifdef CONFIG_ARM64
+dml_rcflags := -mgeneral-regs-only
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 IS_OLD_GCC = 1
@@ -60,6 +64,13 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_rcflags)
 endif
 ifdef CONFIG_DRM_AMD_DC_DCN3_0
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) -Wframe-larger-than=2048
@@ -67,6 +78,8 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags)
 endif
 CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_rcflags)
 
 DML = display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o \
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c
index 80170f9721ce..860e72a51534 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c
@@ -2635,15 +2635,14 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP
 	}
 
 	if (mode_lib->vba.DRAMClockChangeSupportsVActive &&
-			mode_lib->vba.MinActiveDRAMClockChangeMargin > 60) {
+			mode_lib->vba.MinActiveDRAMClockChangeMargin > 60 &&
+			mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) {
 		mode_lib->vba.DRAMClockChangeWatermark += 25;
 
 		for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) {
-			if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) {
-				if (mode_lib->vba.DRAMClockChangeWatermark >
-				dml_max(mode_lib->vba.StutterEnterPlusExitWatermark, mode_lib->vba.UrgentWatermark))
-					mode_lib->vba.MinTTUVBlank[k] += 25;
-			}
+			if (mode_lib->vba.DRAMClockChangeWatermark >
+			dml_max(mode_lib->vba.StutterEnterPlusExitWatermark, mode_lib->vba.UrgentWatermark))
+				mode_lib->vba.MinTTUVBlank[k] += 25;
 		}
 
 		mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h
index 1e557ddcb638..d0b90947f540 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h
@@ -33,7 +33,7 @@ struct display_mode_lib;
 
 // Function: dml_rq_dlg_get_rq_reg
 //  Main entry point for test to get the register values out of this DML class.
-//  This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
 //  and then populate the rq_regs struct
 // Input:
 //  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h
index 0d53e871a9d1..27cf8bed9376 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h
@@ -33,7 +33,7 @@ struct display_mode_lib;
 
 // Function: dml_rq_dlg_get_rq_reg
 //  Main entry point for test to get the register values out of this DML class.
-//  This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
 //  and then populate the rq_regs struct
 // Input:
 //  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
index a576eed94d9b..367c82b5ab4c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
@@ -1294,7 +1294,7 @@ static unsigned int CalculateVMAndRowBytes(
 	unsigned int MacroTileHeight;
 	unsigned int ExtraDPDEBytesFrame;
 	unsigned int PDEAndMetaPTEBytesFrame;
-	unsigned int PixelPTEReqHeightPTEs;
+	unsigned int PixelPTEReqHeightPTEs = 0;
 
 	if (DCCEnable == true) {
 		*MetaRequestHeight = 8 * BlockHeight256Bytes;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index 2beb284f89b0..9e0ae18e71fa 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -597,7 +597,8 @@ static void CalculateStutterEfficiency(
 		double meta_row_bw[],
 		double dpte_row_bw[],
 		double *StutterEfficiencyNotIncludingVBlank,
-		double *StutterEfficiency);
+		double *StutterEfficiency,
+		double *StutterPeriodOut);
 
 static void CalculateSwathAndDETConfiguration(
 		bool ForceSingleDPP,
@@ -3134,7 +3135,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 			v->meta_row_bw,
 			v->dpte_row_bw,
 			&v->StutterEfficiencyNotIncludingVBlank,
-			&v->StutterEfficiency);
+			&v->StutterEfficiency,
+			&v->StutterPeriod);
 }
 
 static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib)
@@ -3235,7 +3237,7 @@ static bool CalculateBytePerPixelAnd256BBlockSizes(
 		*BytePerPixelDETC = 0;
 		*BytePerPixelY = 4;
 		*BytePerPixelC = 0;
-	} else if (SourcePixelFormat == dm_444_16 || SourcePixelFormat == dm_444_16) {
+	} else if (SourcePixelFormat == dm_444_16) {
 		*BytePerPixelDETY = 2;
 		*BytePerPixelDETC = 0;
 		*BytePerPixelY = 2;
@@ -5305,7 +5307,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 			ViewportExceedsSurface = true;
 
 		if (v->SourcePixelFormat[k] != dm_444_64 && v->SourcePixelFormat[k] != dm_444_32 && v->SourcePixelFormat[k] != dm_444_16
-				&& v->SourcePixelFormat[k] != dm_444_16 && v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
+				&& v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
 			if (v->ViewportWidthChroma[k] > v->SurfaceWidthC[k] || v->ViewportHeightChroma[k] > v->SurfaceHeightC[k]) {
 				ViewportExceedsSurface = true;
 			}
@@ -5515,7 +5517,7 @@ static void CalculateWatermarksAndDRAMSpeedChangeSupport(
 			if (WritebackPixelFormat[k] == dm_444_64) {
 				WritebackDRAMClockChangeLatencyHiding = WritebackDRAMClockChangeLatencyHiding / 2;
 			}
-			if (mode_lib->vba.WritebackConfiguration == dm_whole_buffer_for_single_stream_interleave || mode_lib->vba.WritebackConfiguration == dm_whole_buffer_for_single_stream_interleave) {
+			if (mode_lib->vba.WritebackConfiguration == dm_whole_buffer_for_single_stream_interleave) {
 				WritebackDRAMClockChangeLatencyHiding = WritebackDRAMClockChangeLatencyHiding * 2;
 			}
 			WritebackDRAMClockChangeLatencyMargin = WritebackDRAMClockChangeLatencyHiding - mode_lib->vba.WritebackDRAMClockChangeWatermark;
@@ -5556,7 +5558,7 @@ static void CalculateWatermarksAndDRAMSpeedChangeSupport(
 		}
 	}
 
-	if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) {
+	if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0 && PrefetchMode == 0) {
 		*DRAMClockChangeSupport = dm_dram_clock_change_vactive;
 	} else if (((mode_lib->vba.SynchronizedVBlank == true || mode_lib->vba.TotalNumberOfActiveOTG == 1 || SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank > 0) && PrefetchMode == 0)) {
 		*DRAMClockChangeSupport = dm_dram_clock_change_vblank;
@@ -6151,7 +6153,8 @@ static void CalculateStutterEfficiency(
 		double meta_row_bw[],
 		double dpte_row_bw[],
 		double *StutterEfficiencyNotIncludingVBlank,
-		double *StutterEfficiency)
+		double *StutterEfficiency,
+		double *StutterPeriodOut)
 {
 	double FullDETBufferingTimeY[DC__NUM_DPP__MAX] = { 0 };
 	double FrameTimeForMinFullDETBufferingTime = 0;
@@ -6262,6 +6265,9 @@ static void CalculateStutterEfficiency(
 	}
 
 	*StutterEfficiency =  (*StutterEfficiencyNotIncludingVBlank / 100.0 * (FrameTimeForMinFullDETBufferingTime - SmallestVBlank) + SmallestVBlank) / FrameTimeForMinFullDETBufferingTime * 100;
+
+	if (StutterPeriodOut)
+		*StutterPeriodOut = StutterPeriod;
 }
 
 static void CalculateSwathAndDETConfiguration(
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
index 5bb10f6e300d..416bf6fb67bd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
@@ -279,7 +279,7 @@ static bool CalculateBytePerPixelAnd256BBlockSizes(
 		*BytePerPixelDETC = 0;
 		*BytePerPixelY = 4;
 		*BytePerPixelC = 0;
-	} else if (SourcePixelFormat == dm_444_16 || SourcePixelFormat == dm_444_16) {
+	} else if (SourcePixelFormat == dm_444_16) {
 		*BytePerPixelDETY = 2;
 		*BytePerPixelDETC = 0;
 		*BytePerPixelY = 2;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h
index e5b17e1104c6..c04965cceff3 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h
@@ -32,7 +32,7 @@ struct display_mode_lib;
 
 // Function: dml_rq_dlg_get_rq_reg
 //  Main entry point for test to get the register values out of this DML class.
-//  This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
 //  and then populate the rq_regs struct
 // Input:
 //  pipe_param - pipe source configuration (e.g. vp, pitch, scaling, dest, etc.)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
index afdd4f0d9d71..b32093136089 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
@@ -467,7 +467,7 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib)
 		mode_lib->vba.AudioSampleLayout[mode_lib->vba.NumberOfActivePlanes] =
 			1;
 		mode_lib->vba.DRAMClockChangeLatencyOverride = 0.0;
-		mode_lib->vba.DSCEnabled[mode_lib->vba.NumberOfActivePlanes] = dout->dsc_enable;;
+		mode_lib->vba.DSCEnabled[mode_lib->vba.NumberOfActivePlanes] = dout->dsc_enable;
 		mode_lib->vba.DSCEnable[mode_lib->vba.NumberOfActivePlanes] = dout->dsc_enable;
 		mode_lib->vba.NumberOfDSCSlices[mode_lib->vba.NumberOfActivePlanes] =
 				dout->dsc_slices;
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/Makefile b/drivers/gpu/drm/amd/display/dc/dsc/Makefile
index ea29cf95d470..f2624a1156e5 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile
@@ -10,6 +10,10 @@ ifdef CONFIG_PPC64
 dsc_ccflags := -mhard-float -maltivec
 endif
 
+ifdef CONFIG_ARM64
+dsc_rcflags := -mgeneral-regs-only
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 IS_OLD_GCC = 1
@@ -28,6 +32,7 @@ endif
 endif
 
 CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_rcflags)
 
 DSC = dc_dsc.o rc_calc.o rc_calc_dpi.o
 
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index 8cdaa6eef5d3..4c844cfaa956 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -34,6 +34,9 @@
 /* default DSC policy target bitrate limit is 16bpp */
 static uint32_t dsc_policy_max_target_bpp_limit = 16;
 
+/* default DSC policy enables DSC only when needed */
+static bool dsc_policy_enable_dsc_when_not_needed;
+
 static uint32_t dc_dsc_bandwidth_in_kbps_from_timing(
 	const struct dc_crtc_timing *timing)
 {
@@ -189,8 +192,10 @@ static bool dsc_throughput_from_dpcd(int dpcd_throughput, int *throughput)
 }
 
 
-static bool dsc_bpp_increment_div_from_dpcd(int bpp_increment_dpcd, uint32_t *bpp_increment_div)
+static bool dsc_bpp_increment_div_from_dpcd(uint8_t bpp_increment_dpcd, uint32_t *bpp_increment_div)
 {
+	// Mask bpp increment dpcd field to avoid reading other fields
+	bpp_increment_dpcd &= 0x7;
 
 	switch (bpp_increment_dpcd) {
 	case 0:
@@ -360,7 +365,7 @@ static bool decide_dsc_target_bpp_x16(
 
 	get_dsc_bandwidth_range(policy->min_target_bpp, policy->max_target_bpp,
 			dsc_common_caps, timing, &range);
-	if (target_bandwidth_kbps >= range.stream_kbps) {
+	if (!policy->enable_dsc_when_not_needed && target_bandwidth_kbps >= range.stream_kbps) {
 		/* enough bandwidth without dsc */
 		*target_bpp_x16 = 0;
 		should_use_dsc = false;
@@ -961,9 +966,20 @@ void dc_dsc_get_policy_for_timing(const struct dc_crtc_timing *timing, struct dc
 	/* internal upper limit, default 16 bpp */
 	if (policy->max_target_bpp > dsc_policy_max_target_bpp_limit)
 		policy->max_target_bpp = dsc_policy_max_target_bpp_limit;
+
+	/* enable DSC when not needed, default false */
+	if (dsc_policy_enable_dsc_when_not_needed)
+		policy->enable_dsc_when_not_needed = dsc_policy_enable_dsc_when_not_needed;
+	else
+		policy->enable_dsc_when_not_needed = false;
 }
 
 void dc_dsc_policy_set_max_target_bpp_limit(uint32_t limit)
 {
 	dsc_policy_max_target_bpp_limit = limit;
 }
+
+void dc_dsc_policy_set_enable_dsc_when_not_needed(bool enable)
+{
+	dsc_policy_enable_dsc_when_not_needed = enable;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/Makefile b/drivers/gpu/drm/amd/display/dc/gpio/Makefile
index 0f2f4508e564..74c0943ed644 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/gpio/Makefile
@@ -31,6 +31,18 @@ AMD_DAL_GPIO = $(addprefix $(AMDDALPATH)/dc/gpio/,$(GPIO))
 AMD_DISPLAY_FILES += $(AMD_DAL_GPIO)
 
 ###############################################################################
+# DCE 6x
+###############################################################################
+# all DCE6.x are derived from DCE6.0
+ifdef CONFIG_DRM_AMD_DC_SI
+GPIO_DCE60 = hw_translate_dce60.o hw_factory_dce60.o
+
+AMD_DAL_GPIO_DCE60 = $(addprefix $(AMDDALPATH)/dc/gpio/dce60/,$(GPIO_DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCE60)
+endif
+
+###############################################################################
 # DCE 8x
 ###############################################################################
 # all DCE8.x are derived from DCE8.0
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c
index cf98aa827a9a..e883864cff3c 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c
@@ -162,7 +162,7 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en)
 }
 
 
-/* fucntion table */
+/* function table */
 static const struct hw_factory_funcs funcs = {
 	.init_ddc_data = dal_hw_ddc_init,
 	.init_generic = NULL,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.c b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.c
new file mode 100644
index 000000000000..cc69acd8ada7
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.c
@@ -0,0 +1,175 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include "dm_services.h"
+#include "include/gpio_types.h"
+#include "../hw_factory.h"
+
+#include "hw_factory_dce60.h"
+
+#include "../hw_gpio.h"
+#include "../hw_ddc.h"
+#include "../hw_hpd.h"
+#include "../hw_generic.h"
+
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+
+#define REG(reg_name)\
+		mm ## reg_name
+
+#include "reg_helper.h"
+#include "../hpd_regs.h"
+
+#define HPD_REG_LIST_DCE6(id) \
+	HPD_GPIO_REG_LIST(id), \
+	.int_status = mmDC_HPD ## id ## _INT_STATUS,\
+	.toggle_filt_cntl = mmDC_HPD ## id ## _TOGGLE_FILT_CNTL
+
+#define HPD_MASK_SH_LIST_DCE6(mask_sh) \
+		.DC_HPD_SENSE_DELAYED = DC_HPD1_INT_STATUS__DC_HPD1_SENSE_DELAYED ## mask_sh,\
+		.DC_HPD_SENSE = DC_HPD1_INT_STATUS__DC_HPD1_SENSE ## mask_sh,\
+		.DC_HPD_CONNECT_INT_DELAY = DC_HPD1_TOGGLE_FILT_CNTL__DC_HPD1_CONNECT_INT_DELAY ## mask_sh,\
+		.DC_HPD_DISCONNECT_INT_DELAY = DC_HPD1_TOGGLE_FILT_CNTL__DC_HPD1_DISCONNECT_INT_DELAY ## mask_sh
+
+#define hpd_regs(id) \
+{\
+	HPD_REG_LIST_DCE6(id)\
+}
+
+static const struct hpd_registers hpd_regs[] = {
+	hpd_regs(1),
+	hpd_regs(2),
+	hpd_regs(3),
+	hpd_regs(4),
+	hpd_regs(5),
+	hpd_regs(6)
+};
+
+static const struct hpd_sh_mask hpd_shift = {
+		HPD_MASK_SH_LIST_DCE6(__SHIFT)
+};
+
+static const struct hpd_sh_mask hpd_mask = {
+		HPD_MASK_SH_LIST_DCE6(_MASK)
+};
+
+#include "../ddc_regs.h"
+
+ /* set field name */
+#define SF_DDC(reg_name, field_name, post_fix)\
+	.field_name = reg_name ## __ ## field_name ## post_fix
+
+static const struct ddc_registers ddc_data_regs[] = {
+	ddc_data_regs(1),
+	ddc_data_regs(2),
+	ddc_data_regs(3),
+	ddc_data_regs(4),
+	ddc_data_regs(5),
+	ddc_data_regs(6),
+	ddc_vga_data_regs,
+	ddc_i2c_data_regs
+};
+
+static const struct ddc_registers ddc_clk_regs[] = {
+	ddc_clk_regs(1),
+	ddc_clk_regs(2),
+	ddc_clk_regs(3),
+	ddc_clk_regs(4),
+	ddc_clk_regs(5),
+	ddc_clk_regs(6),
+	ddc_vga_clk_regs,
+	ddc_i2c_clk_regs
+};
+
+static const struct ddc_sh_mask ddc_shift = {
+		DDC_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct ddc_sh_mask ddc_mask = {
+		DDC_MASK_SH_LIST(_MASK)
+};
+
+static void define_ddc_registers(
+		struct hw_gpio_pin *pin,
+		uint32_t en)
+{
+	struct hw_ddc *ddc = HW_DDC_FROM_BASE(pin);
+
+	switch (pin->id) {
+	case GPIO_ID_DDC_DATA:
+		ddc->regs = &ddc_data_regs[en];
+		ddc->base.regs = &ddc_data_regs[en].gpio;
+		break;
+	case GPIO_ID_DDC_CLOCK:
+		ddc->regs = &ddc_clk_regs[en];
+		ddc->base.regs = &ddc_clk_regs[en].gpio;
+		break;
+	default:
+		ASSERT_CRITICAL(false);
+		return;
+	}
+
+	ddc->shifts = &ddc_shift;
+	ddc->masks = &ddc_mask;
+
+}
+
+static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en)
+{
+	struct hw_hpd *hpd = HW_HPD_FROM_BASE(pin);
+
+	hpd->regs = &hpd_regs[en];
+	hpd->shifts = &hpd_shift;
+	hpd->masks = &hpd_mask;
+	hpd->base.regs = &hpd_regs[en].gpio;
+}
+
+static const struct hw_factory_funcs funcs = {
+	.init_ddc_data = dal_hw_ddc_init,
+	.init_generic = NULL,
+	.init_hpd = dal_hw_hpd_init,
+	.get_ddc_pin = dal_hw_ddc_get_pin,
+	.get_hpd_pin = dal_hw_hpd_get_pin,
+	.get_generic_pin = NULL,
+	.define_hpd_registers = define_hpd_registers,
+	.define_ddc_registers = define_ddc_registers
+};
+
+void dal_hw_factory_dce60_init(
+	struct hw_factory *factory)
+{
+	factory->number_of_pins[GPIO_ID_DDC_DATA] = 8;
+	factory->number_of_pins[GPIO_ID_DDC_CLOCK] = 8;
+	factory->number_of_pins[GPIO_ID_GENERIC] = 7;
+	factory->number_of_pins[GPIO_ID_HPD] = 6;
+	factory->number_of_pins[GPIO_ID_GPIO_PAD] = 31;
+	factory->number_of_pins[GPIO_ID_VIP_PAD] = 0;
+	factory->number_of_pins[GPIO_ID_SYNC] = 2;
+	factory->number_of_pins[GPIO_ID_GSL] = 4;
+
+	factory->funcs = &funcs;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.h b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.h
new file mode 100644
index 000000000000..1fd54ff8979c
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_factory_dce60.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DAL_HW_FACTORY_DCE60_H__
+#define __DAL_HW_FACTORY_DCE60_H__
+
+void dal_hw_factory_dce60_init(
+	struct hw_factory *factory);
+
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.c b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.c
new file mode 100644
index 000000000000..255df31ec577
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.c
@@ -0,0 +1,411 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include "dm_services.h"
+
+/*
+ * Pre-requisites: headers required by header of this unit
+ */
+#include "include/gpio_types.h"
+#include "../hw_translate.h"
+
+#include "hw_translate_dce60.h"
+
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+#include "smu/smu_6_0_d.h"
+
+/*
+ * @brief
+ * Returns index of first bit (starting with LSB) which is set
+ */
+static uint32_t index_from_vector(
+	uint32_t vector)
+{
+	uint32_t result = 0;
+	uint32_t mask = 1;
+
+	do {
+		if (vector == mask)
+			return result;
+
+		++result;
+		mask <<= 1;
+	} while (mask);
+
+	BREAK_TO_DEBUGGER();
+
+	return GPIO_ENUM_UNKNOWN;
+}
+
+static bool offset_to_id(
+	uint32_t offset,
+	uint32_t mask,
+	enum gpio_id *id,
+	uint32_t *en)
+{
+	switch (offset) {
+	/* GENERIC */
+	case mmDC_GPIO_GENERIC_A:
+		*id = GPIO_ID_GENERIC;
+		switch (mask) {
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK:
+			*en = GPIO_GENERIC_A;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK:
+			*en = GPIO_GENERIC_B;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK:
+			*en = GPIO_GENERIC_C;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK:
+			*en = GPIO_GENERIC_D;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK:
+			*en = GPIO_GENERIC_E;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK:
+			*en = GPIO_GENERIC_F;
+			return true;
+		case DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK:
+			*en = GPIO_GENERIC_G;
+			return true;
+		default:
+			BREAK_TO_DEBUGGER();
+			return false;
+		}
+	break;
+	/* HPD */
+	case mmDC_GPIO_HPD_A:
+		*id = GPIO_ID_HPD;
+		switch (mask) {
+		case DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK:
+			*en = GPIO_HPD_1;
+			return true;
+		case DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK:
+			*en = GPIO_HPD_2;
+			return true;
+		case DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK:
+			*en = GPIO_HPD_3;
+			return true;
+		case DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK:
+			*en = GPIO_HPD_4;
+			return true;
+		case DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK:
+			*en = GPIO_HPD_5;
+			return true;
+		case DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK:
+			*en = GPIO_HPD_6;
+			return true;
+		default:
+			BREAK_TO_DEBUGGER();
+			return false;
+		}
+	break;
+	/* SYNCA */
+	case mmDC_GPIO_SYNCA_A:
+		*id = GPIO_ID_SYNC;
+		switch (mask) {
+		case DC_GPIO_SYNCA_A__DC_GPIO_HSYNCA_A_MASK:
+			*en = GPIO_SYNC_HSYNC_A;
+			return true;
+		case DC_GPIO_SYNCA_A__DC_GPIO_VSYNCA_A_MASK:
+			*en = GPIO_SYNC_VSYNC_A;
+			return true;
+		default:
+			BREAK_TO_DEBUGGER();
+			return false;
+		}
+	break;
+	/* mmDC_GPIO_GENLK_MASK */
+	case mmDC_GPIO_GENLK_A:
+		*id = GPIO_ID_GSL;
+		switch (mask) {
+		case DC_GPIO_GENLK_A__DC_GPIO_GENLK_CLK_A_MASK:
+			*en = GPIO_GSL_GENLOCK_CLOCK;
+			return true;
+		case DC_GPIO_GENLK_A__DC_GPIO_GENLK_VSYNC_A_MASK:
+			*en = GPIO_GSL_GENLOCK_VSYNC;
+			return true;
+		case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_A_A_MASK:
+			*en = GPIO_GSL_SWAPLOCK_A;
+			return true;
+		case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_B_A_MASK:
+			*en = GPIO_GSL_SWAPLOCK_B;
+			return true;
+		default:
+			BREAK_TO_DEBUGGER();
+			return false;
+		}
+	break;
+	/* GPIOPAD */
+	case mmGPIOPAD_A:
+		*id = GPIO_ID_GPIO_PAD;
+		*en = index_from_vector(mask);
+		return (*en <= GPIO_GPIO_PAD_MAX);
+	/* DDC */
+	/* we don't care about the GPIO_ID for DDC
+	 * in DdcHandle it will use GPIO_ID_DDC_DATA/GPIO_ID_DDC_CLOCK
+	 * directly in the create method */
+	case mmDC_GPIO_DDC1_A:
+		*en = GPIO_DDC_LINE_DDC1;
+		return true;
+	case mmDC_GPIO_DDC2_A:
+		*en = GPIO_DDC_LINE_DDC2;
+		return true;
+	case mmDC_GPIO_DDC3_A:
+		*en = GPIO_DDC_LINE_DDC3;
+		return true;
+	case mmDC_GPIO_DDC4_A:
+		*en = GPIO_DDC_LINE_DDC4;
+		return true;
+	case mmDC_GPIO_DDC5_A:
+		*en = GPIO_DDC_LINE_DDC5;
+		return true;
+	case mmDC_GPIO_DDC6_A:
+		*en = GPIO_DDC_LINE_DDC6;
+		return true;
+	case mmDC_GPIO_DDCVGA_A:
+		*en = GPIO_DDC_LINE_DDC_VGA;
+		return true;
+	/* GPIO_I2CPAD */
+	case mmDC_GPIO_I2CPAD_A:
+		*en = GPIO_DDC_LINE_I2C_PAD;
+		return true;
+	/* Not implemented */
+	case mmDC_GPIO_PWRSEQ_A:
+	case mmDC_GPIO_PAD_STRENGTH_1:
+	case mmDC_GPIO_PAD_STRENGTH_2:
+	case mmDC_GPIO_DEBUG:
+		return false;
+	/* UNEXPECTED */
+	default:
+		BREAK_TO_DEBUGGER();
+		return false;
+	}
+}
+
+static bool id_to_offset(
+	enum gpio_id id,
+	uint32_t en,
+	struct gpio_pin_info *info)
+{
+	bool result = true;
+
+	switch (id) {
+	case GPIO_ID_DDC_DATA:
+		info->mask = DC_GPIO_DDC6_A__DC_GPIO_DDC6DATA_A_MASK;
+		switch (en) {
+		case GPIO_DDC_LINE_DDC1:
+			info->offset = mmDC_GPIO_DDC1_A;
+		break;
+		case GPIO_DDC_LINE_DDC2:
+			info->offset = mmDC_GPIO_DDC2_A;
+		break;
+		case GPIO_DDC_LINE_DDC3:
+			info->offset = mmDC_GPIO_DDC3_A;
+		break;
+		case GPIO_DDC_LINE_DDC4:
+			info->offset = mmDC_GPIO_DDC4_A;
+		break;
+		case GPIO_DDC_LINE_DDC5:
+			info->offset = mmDC_GPIO_DDC5_A;
+		break;
+		case GPIO_DDC_LINE_DDC6:
+			info->offset = mmDC_GPIO_DDC6_A;
+		break;
+		case GPIO_DDC_LINE_DDC_VGA:
+			info->offset = mmDC_GPIO_DDCVGA_A;
+		break;
+		case GPIO_DDC_LINE_I2C_PAD:
+			info->offset = mmDC_GPIO_I2CPAD_A;
+		break;
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_DDC_CLOCK:
+		info->mask = DC_GPIO_DDC6_A__DC_GPIO_DDC6CLK_A_MASK;
+		switch (en) {
+		case GPIO_DDC_LINE_DDC1:
+			info->offset = mmDC_GPIO_DDC1_A;
+		break;
+		case GPIO_DDC_LINE_DDC2:
+			info->offset = mmDC_GPIO_DDC2_A;
+		break;
+		case GPIO_DDC_LINE_DDC3:
+			info->offset = mmDC_GPIO_DDC3_A;
+		break;
+		case GPIO_DDC_LINE_DDC4:
+			info->offset = mmDC_GPIO_DDC4_A;
+		break;
+		case GPIO_DDC_LINE_DDC5:
+			info->offset = mmDC_GPIO_DDC5_A;
+		break;
+		case GPIO_DDC_LINE_DDC6:
+			info->offset = mmDC_GPIO_DDC6_A;
+		break;
+		case GPIO_DDC_LINE_DDC_VGA:
+			info->offset = mmDC_GPIO_DDCVGA_A;
+		break;
+		case GPIO_DDC_LINE_I2C_PAD:
+			info->offset = mmDC_GPIO_I2CPAD_A;
+		break;
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_GENERIC:
+		info->offset = mmDC_GPIO_GENERIC_A;
+		switch (en) {
+		case GPIO_GENERIC_A:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK;
+		break;
+		case GPIO_GENERIC_B:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK;
+		break;
+		case GPIO_GENERIC_C:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK;
+		break;
+		case GPIO_GENERIC_D:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK;
+		break;
+		case GPIO_GENERIC_E:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK;
+		break;
+		case GPIO_GENERIC_F:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK;
+		break;
+		case GPIO_GENERIC_G:
+			info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK;
+		break;
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_HPD:
+		info->offset = mmDC_GPIO_HPD_A;
+		switch (en) {
+		case GPIO_HPD_1:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK;
+		break;
+		case GPIO_HPD_2:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK;
+		break;
+		case GPIO_HPD_3:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK;
+		break;
+		case GPIO_HPD_4:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK;
+		break;
+		case GPIO_HPD_5:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK;
+		break;
+		case GPIO_HPD_6:
+			info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK;
+		break;
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_SYNC:
+		switch (en) {
+		case GPIO_SYNC_HSYNC_A:
+			info->offset = mmDC_GPIO_SYNCA_A;
+			info->mask = DC_GPIO_SYNCA_A__DC_GPIO_HSYNCA_A_MASK;
+		break;
+		case GPIO_SYNC_VSYNC_A:
+			info->offset = mmDC_GPIO_SYNCA_A;
+			info->mask = DC_GPIO_SYNCA_A__DC_GPIO_VSYNCA_A_MASK;
+		break;
+		case GPIO_SYNC_HSYNC_B:
+		case GPIO_SYNC_VSYNC_B:
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_GSL:
+		switch (en) {
+		case GPIO_GSL_GENLOCK_CLOCK:
+			info->offset = mmDC_GPIO_GENLK_A;
+			info->mask = DC_GPIO_GENLK_A__DC_GPIO_GENLK_CLK_A_MASK;
+		break;
+		case GPIO_GSL_GENLOCK_VSYNC:
+			info->offset = mmDC_GPIO_GENLK_A;
+			info->mask =
+				DC_GPIO_GENLK_A__DC_GPIO_GENLK_VSYNC_A_MASK;
+		break;
+		case GPIO_GSL_SWAPLOCK_A:
+			info->offset = mmDC_GPIO_GENLK_A;
+			info->mask = DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_A_A_MASK;
+		break;
+		case GPIO_GSL_SWAPLOCK_B:
+			info->offset = mmDC_GPIO_GENLK_A;
+			info->mask = DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_B_A_MASK;
+		break;
+		default:
+			BREAK_TO_DEBUGGER();
+			result = false;
+		}
+	break;
+	case GPIO_ID_GPIO_PAD:
+		info->offset = mmGPIOPAD_A;
+		info->mask = (1 << en);
+		result = (info->mask <= GPIO_GPIO_PAD_MAX);
+	break;
+	case GPIO_ID_VIP_PAD:
+	default:
+		BREAK_TO_DEBUGGER();
+		result = false;
+	}
+
+	if (result) {
+		info->offset_y = info->offset + 2;
+		info->offset_en = info->offset + 1;
+		info->offset_mask = info->offset - 1;
+
+		info->mask_y = info->mask;
+		info->mask_en = info->mask;
+		info->mask_mask = info->mask;
+	}
+
+	return result;
+}
+
+static const struct hw_translate_funcs funcs = {
+		.offset_to_id = offset_to_id,
+		.id_to_offset = id_to_offset,
+};
+
+void dal_hw_translate_dce60_init(
+	struct hw_translate *translate)
+{
+	translate->funcs = &funcs;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.h b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.h
new file mode 100644
index 000000000000..1e811f35cec7
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dce60/hw_translate_dce60.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DAL_HW_TRANSLATE_DCE60_H__
+#define __DAL_HW_TRANSLATE_DCE60_H__
+
+void dal_hw_translate_dce60_init(
+	struct hw_translate *tr);
+
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c
index b38c96c9fed3..7d36b56346a6 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c
@@ -194,7 +194,7 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en)
 }
 
 
-/* fucntion table */
+/* function table */
 static const struct hw_factory_funcs funcs = {
 	.init_ddc_data = dal_hw_ddc_init,
 	.init_generic = dal_hw_generic_init,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c
index 83f798cb8b21..9b63c6c0cc84 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c
@@ -221,7 +221,7 @@ static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en)
 	generic->base.regs = &generic_regs[en].gpio;
 }
 
-/* fucntion table */
+/* function table */
 static const struct hw_factory_funcs funcs = {
 	.init_ddc_data = dal_hw_ddc_init,
 	.init_generic = dal_hw_generic_init,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c
index 907c5911eb9e..2f57ee6deabc 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c
@@ -202,7 +202,7 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en)
 }
 
 
-/* fucntion table */
+/* function table */
 static const struct hw_factory_funcs funcs = {
 	.init_ddc_data = dal_hw_ddc_init,
 	.init_generic = dal_hw_generic_init,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c
index 7e7fb6572107..3be2c90b0c61 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c
@@ -218,7 +218,7 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en)
 }
 
 
-/* fucntion table */
+/* function table */
 static const struct hw_factory_funcs funcs = {
 	.init_ddc_data = dal_hw_ddc_init,
 	.init_generic = dal_hw_generic_init,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c
index e5cfe28bc7bf..6fc8a6e9dc15 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c
@@ -42,6 +42,9 @@
  * Post-requisites: headers required by this unit
  */
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/hw_factory_dce60.h"
+#endif
 #include "dce80/hw_factory_dce80.h"
 #include "dce110/hw_factory_dce110.h"
 #include "dce120/hw_factory_dce120.h"
@@ -71,6 +74,13 @@ bool dal_hw_factory_init(
 	}
 
 	switch (dce_version) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case DCE_VERSION_6_0:
+	case DCE_VERSION_6_1:
+	case DCE_VERSION_6_4:
+		dal_hw_factory_dce60_init(factory);
+		return true;
+#endif
 	case DCE_VERSION_8_0:
 	case DCE_VERSION_8_1:
 	case DCE_VERSION_8_3:
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c
index efea7cb0f17c..3a93c945e57d 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c
@@ -40,6 +40,9 @@
  * Post-requisites: headers required by this unit
  */
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/hw_translate_dce60.h"
+#endif
 #include "dce80/hw_translate_dce80.h"
 #include "dce110/hw_translate_dce110.h"
 #include "dce120/hw_translate_dce120.h"
@@ -69,6 +72,13 @@ bool dal_hw_translate_init(
 	}
 
 	switch (dce_version) {
+#if defined(CONFIG_DRM_AMD_DC_SI)
+	case DCE_VERSION_6_0:
+	case DCE_VERSION_6_1:
+	case DCE_VERSION_6_4:
+		dal_hw_translate_dce60_init(translate);
+		return true;
+#endif
 	case DCE_VERSION_8_0:
 	case DCE_VERSION_8_1:
 	case DCE_VERSION_8_3:
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index 329395ee7461..6e6bc66e49f0 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -101,7 +101,7 @@ struct resource_funcs {
 					struct dc *dc,
 					struct dc_state *context,
 					bool fast_validate);
-	void (*calculate_wm)(
+	void (*calculate_wm_and_dlg)(
 				struct dc *dc, struct dc_state *context,
 				display_e2e_pipe_params_st *pipes,
 				int pipe_cnt,
@@ -300,6 +300,7 @@ union pipe_update_flags {
 		uint32_t gamut_remap : 1;
 		uint32_t scaler : 1;
 		uint32_t viewport : 1;
+		uint32_t plane_changed : 1;
 	} bits;
 	uint32_t raw;
 };
@@ -396,6 +397,7 @@ struct dc_state {
 	struct dc_stream_state *streams[MAX_PIPES];
 	struct dc_stream_status stream_status[MAX_PIPES];
 	uint8_t stream_count;
+	uint8_t stream_mask;
 
 	struct resource_context res_ctx;
 
@@ -410,6 +412,10 @@ struct dc_state {
 	struct clk_mgr *clk_mgr;
 
 	struct kref refcount;
+
+	struct {
+		unsigned int stutter_period_us;
+	} perf_params;
 };
 
 #endif /* _CORE_TYPES_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
index 72743058836d..949b61351ede 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
@@ -89,6 +89,11 @@ enum dentist_divider_range {
 	.DPREFCLK_CNTL = mmDPREFCLK_CNTL, \
 	.DENTIST_DISPCLK_CNTL = mmDENTIST_DISPCLK_CNTL
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define CLK_COMMON_REG_LIST_DCE60_BASE() \
+	SR(DENTIST_DISPCLK_CNTL)
+#endif
+
 #define CLK_COMMON_REG_LIST_DCN_BASE() \
 	SR(DENTIST_DISPCLK_CNTL)
 
@@ -115,6 +120,12 @@ enum dentist_divider_range {
 	CLK_SF(DPREFCLK_CNTL, DPREFCLK_SRC_SEL, mask_sh), \
 	CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, mask_sh)
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#define CLK_COMMON_MASK_SH_LIST_DCE60_COMMON_BASE(mask_sh) \
+	CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh),\
+	CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)
+#endif
+
 #define CLK_COMMON_MASK_SH_LIST_DCN_COMMON_BASE(mask_sh) \
 	CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh),\
 	CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h
index f9ab5abb6462..48eac622c6a0 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h
@@ -49,6 +49,7 @@ struct panel_cntl_funcs {
 	void (*store_backlight_level)(struct panel_cntl *panel_cntl);
 	void (*driver_set_backlight)(struct panel_cntl *panel_cntl,
 			uint32_t backlight_pwm_u16_16);
+	uint32_t (*get_current_backlight)(struct panel_cntl *panel_cntl);
 };
 
 struct panel_cntl_init_data {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
index 11ce06e69d3f..0184cefb083b 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
@@ -143,7 +143,7 @@ struct stream_encoder_funcs {
 		struct stream_encoder *enc,
 		struct dc_crtc_timing *crtc_timing);
 
-	void (*set_mst_bandwidth)(
+	void (*set_throttled_vcp_size)(
 		struct stream_encoder *enc,
 		struct fixed31_32 avg_time_slots_per_mtp);
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index 3c986717dcd5..64c1be818b0e 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -67,6 +67,10 @@ struct hw_sequencer_funcs {
 			int num_planes, struct dc_state *context);
 	void (*program_front_end_for_ctx)(struct dc *dc,
 			struct dc_state *context);
+	bool (*disconnect_pipes)(struct dc *dc,
+			struct dc_state *context);
+	void (*wait_for_pending_cleared)(struct dc *dc,
+			struct dc_state *context);
 	void (*post_unlock_program_front_end)(struct dc *dc,
 			struct dc_state *context);
 	void (*update_plane_addr)(const struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/irq/Makefile b/drivers/gpu/drm/amd/display/dc/irq/Makefile
index 3352b79fb1cb..405c25322607 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/irq/Makefile
@@ -31,6 +31,17 @@ AMD_DAL_IRQ = $(addprefix $(AMDDALPATH)/dc/irq/,$(IRQ))
 AMD_DISPLAY_FILES += $(AMD_DAL_IRQ)
 
 ###############################################################################
+# DCE 6x
+###############################################################################
+ifdef CONFIG_DRM_AMD_DC_SI
+IRQ_DCE60 = irq_service_dce60.o
+
+AMD_DAL_IRQ_DCE60 = $(addprefix $(AMDDALPATH)/dc/irq/dce60/,$(IRQ_DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCE60)
+endif
+
+###############################################################################
 # DCE 8x
 ###############################################################################
 IRQ_DCE80 = irq_service_dce80.o
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c
new file mode 100644
index 000000000000..524481885fd0
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c
@@ -0,0 +1,395 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#include <linux/slab.h>
+
+#include "dm_services.h"
+
+#include "include/logger_interface.h"
+
+#include "irq_service_dce60.h"
+#include "../dce110/irq_service_dce110.h"
+
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#include "ivsrcid/ivsrcid_vislands30.h"
+
+#define VISLANDS30_IV_SRCID_D1_VBLANK                        1
+#define VISLANDS30_IV_SRCID_D2_VBLANK                        2
+#define VISLANDS30_IV_SRCID_D3_VBLANK                        3
+#define VISLANDS30_IV_SRCID_D4_VBLANK                        4
+#define VISLANDS30_IV_SRCID_D5_VBLANK                        5
+#define VISLANDS30_IV_SRCID_D6_VBLANK                        6
+
+#include "dc_types.h"
+
+static bool hpd_ack(
+	struct irq_service *irq_service,
+	const struct irq_source_info *info)
+{
+	uint32_t addr = info->status_reg;
+	uint32_t value = dm_read_reg(irq_service->ctx, addr);
+	uint32_t current_status =
+		get_reg_field_value(
+			value,
+			DC_HPD1_INT_STATUS,
+			DC_HPD1_SENSE_DELAYED);
+
+	dal_irq_service_ack_generic(irq_service, info);
+
+	value = dm_read_reg(irq_service->ctx, info->enable_reg);
+
+	set_reg_field_value(
+		value,
+		current_status ? 0 : 1,
+		DC_HPD1_INT_CONTROL,
+		DC_HPD1_INT_POLARITY);
+
+	dm_write_reg(irq_service->ctx, info->enable_reg, value);
+
+	return true;
+}
+
+static const struct irq_source_info_funcs hpd_irq_info_funcs = {
+	.set = NULL,
+	.ack = hpd_ack
+};
+
+static const struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
+static const struct irq_source_info_funcs pflip_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
+static const struct irq_source_info_funcs vblank_irq_info_funcs = {
+	.set = dce110_vblank_set,
+	.ack = NULL
+};
+
+static const struct irq_source_info_funcs vblank_irq_info_funcs_dce60 = {
+	.set = NULL,
+	.ack = NULL
+};
+
+#define hpd_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_INVALID + reg_num] = {\
+		.enable_reg = mmDC_HPD ## reg_num ## _INT_CONTROL,\
+		.enable_mask = DC_HPD1_INT_CONTROL__DC_HPD1_INT_EN_MASK,\
+		.enable_value = {\
+			DC_HPD1_INT_CONTROL__DC_HPD1_INT_EN_MASK,\
+			~DC_HPD1_INT_CONTROL__DC_HPD1_INT_EN_MASK\
+		},\
+		.ack_reg = mmDC_HPD ## reg_num ## _INT_CONTROL,\
+		.ack_mask = DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK,\
+		.ack_value = DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK,\
+		.status_reg = mmDC_HPD ## reg_num ## _INT_STATUS,\
+		.funcs = &hpd_irq_info_funcs\
+	}
+
+#define hpd_rx_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_HPD6 + reg_num] = {\
+		.enable_reg = mmDC_HPD ## reg_num ## _INT_CONTROL,\
+		.enable_mask = DC_HPD1_INT_CONTROL__DC_HPD1_RX_INT_EN_MASK,\
+		.enable_value = {\
+				DC_HPD1_INT_CONTROL__DC_HPD1_RX_INT_EN_MASK,\
+			~DC_HPD1_INT_CONTROL__DC_HPD1_RX_INT_EN_MASK },\
+		.ack_reg = mmDC_HPD ## reg_num ## _INT_CONTROL,\
+		.ack_mask = DC_HPD1_INT_CONTROL__DC_HPD1_RX_INT_ACK_MASK,\
+		.ack_value = DC_HPD1_INT_CONTROL__DC_HPD1_RX_INT_ACK_MASK,\
+		.status_reg = mmDC_HPD ## reg_num ## _INT_STATUS,\
+		.funcs = &hpd_rx_irq_info_funcs\
+	}
+
+#define pflip_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_PFLIP1 + reg_num] = {\
+		.enable_reg = mmDCP ## reg_num ## _GRPH_INTERRUPT_CONTROL,\
+		.enable_mask =\
+		GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK,\
+		.enable_value = {\
+			GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK,\
+			~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK},\
+		.ack_reg = mmDCP ## reg_num ## _GRPH_INTERRUPT_STATUS,\
+		.ack_mask = GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK,\
+		.ack_value = GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK,\
+		.status_reg = mmDCP ## reg_num ##_GRPH_INTERRUPT_STATUS,\
+		.funcs = &pflip_irq_info_funcs\
+ 	}
+
+#define vupdate_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
+		.enable_reg = mmCRTC ## reg_num ## _CRTC_INTERRUPT_CONTROL,\
+		.enable_mask =\
+		CRTC_INTERRUPT_CONTROL__CRTC_V_UPDATE_INT_MSK_MASK,\
+		.enable_value = {\
+			CRTC_INTERRUPT_CONTROL__CRTC_V_UPDATE_INT_MSK_MASK,\
+			~CRTC_INTERRUPT_CONTROL__CRTC_V_UPDATE_INT_MSK_MASK},\
+		.ack_reg = mmCRTC ## reg_num ## _CRTC_V_UPDATE_INT_STATUS,\
+		.ack_mask =\
+		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
+		.ack_value =\
+		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
+		.funcs = &vblank_irq_info_funcs\
+	}
+
+#define vblank_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\
+		.enable_reg = mmLB ## reg_num ## _INT_MASK,\
+		.enable_mask =\
+			INT_MASK__VBLANK_INT_MASK,\
+		.enable_value = {\
+			INT_MASK__VBLANK_INT_MASK,\
+			~INT_MASK__VBLANK_INT_MASK},\
+		.ack_reg = mmLB ## reg_num ## _VBLANK_STATUS,\
+		.ack_mask =\
+		VBLANK_STATUS__VBLANK_ACK_MASK,\
+		.ack_value =\
+		VBLANK_STATUS__VBLANK_ACK_MASK,\
+		.funcs = &vblank_irq_info_funcs_dce60\
+	}
+
+#define dummy_irq_entry() \
+	{\
+		.funcs = &dummy_irq_info_funcs\
+	}
+
+#define i2c_int_entry(reg_num) \
+	[DC_IRQ_SOURCE_I2C_DDC ## reg_num] = dummy_irq_entry()
+
+#define dp_sink_int_entry(reg_num) \
+	[DC_IRQ_SOURCE_DPSINK ## reg_num] = dummy_irq_entry()
+
+#define gpio_pad_int_entry(reg_num) \
+	[DC_IRQ_SOURCE_GPIOPAD ## reg_num] = dummy_irq_entry()
+
+#define dc_underflow_int_entry(reg_num) \
+	[DC_IRQ_SOURCE_DC ## reg_num ## UNDERFLOW] = dummy_irq_entry()
+
+
+static const struct irq_source_info_funcs dummy_irq_info_funcs = {
+	.set = dal_irq_service_dummy_set,
+	.ack = dal_irq_service_dummy_ack
+};
+
+static const struct irq_source_info
+irq_source_info_dce60[DAL_IRQ_SOURCES_NUMBER] = {
+	[DC_IRQ_SOURCE_INVALID] = dummy_irq_entry(),
+	hpd_int_entry(1),
+	hpd_int_entry(2),
+	hpd_int_entry(3),
+	hpd_int_entry(4),
+	hpd_int_entry(5),
+	hpd_int_entry(6),
+	hpd_rx_int_entry(1),
+	hpd_rx_int_entry(2),
+	hpd_rx_int_entry(3),
+	hpd_rx_int_entry(4),
+	hpd_rx_int_entry(5),
+	hpd_rx_int_entry(6),
+	i2c_int_entry(1),
+	i2c_int_entry(2),
+	i2c_int_entry(3),
+	i2c_int_entry(4),
+	i2c_int_entry(5),
+	i2c_int_entry(6),
+	dp_sink_int_entry(1),
+	dp_sink_int_entry(2),
+	dp_sink_int_entry(3),
+	dp_sink_int_entry(4),
+	dp_sink_int_entry(5),
+	dp_sink_int_entry(6),
+	[DC_IRQ_SOURCE_TIMER] = dummy_irq_entry(),
+	pflip_int_entry(0),
+	pflip_int_entry(1),
+	pflip_int_entry(2),
+	pflip_int_entry(3),
+	pflip_int_entry(4),
+	pflip_int_entry(5),
+	[DC_IRQ_SOURCE_PFLIP_UNDERLAY0] = dummy_irq_entry(),
+	gpio_pad_int_entry(0),
+	gpio_pad_int_entry(1),
+	gpio_pad_int_entry(2),
+	gpio_pad_int_entry(3),
+	gpio_pad_int_entry(4),
+	gpio_pad_int_entry(5),
+	gpio_pad_int_entry(6),
+	gpio_pad_int_entry(7),
+	gpio_pad_int_entry(8),
+	gpio_pad_int_entry(9),
+	gpio_pad_int_entry(10),
+	gpio_pad_int_entry(11),
+	gpio_pad_int_entry(12),
+	gpio_pad_int_entry(13),
+	gpio_pad_int_entry(14),
+	gpio_pad_int_entry(15),
+	gpio_pad_int_entry(16),
+	gpio_pad_int_entry(17),
+	gpio_pad_int_entry(18),
+	gpio_pad_int_entry(19),
+	gpio_pad_int_entry(20),
+	gpio_pad_int_entry(21),
+	gpio_pad_int_entry(22),
+	gpio_pad_int_entry(23),
+	gpio_pad_int_entry(24),
+	gpio_pad_int_entry(25),
+	gpio_pad_int_entry(26),
+	gpio_pad_int_entry(27),
+	gpio_pad_int_entry(28),
+	gpio_pad_int_entry(29),
+	gpio_pad_int_entry(30),
+	dc_underflow_int_entry(1),
+	dc_underflow_int_entry(2),
+	dc_underflow_int_entry(3),
+	dc_underflow_int_entry(4),
+	dc_underflow_int_entry(5),
+	dc_underflow_int_entry(6),
+	[DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(),
+	[DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(),
+	vupdate_int_entry(0),
+	vupdate_int_entry(1),
+	vupdate_int_entry(2),
+	vupdate_int_entry(3),
+	vupdate_int_entry(4),
+	vupdate_int_entry(5),
+	vblank_int_entry(0),
+	vblank_int_entry(1),
+	vblank_int_entry(2),
+	vblank_int_entry(3),
+	vblank_int_entry(4),
+	vblank_int_entry(5),
+};
+
+enum dc_irq_source to_dal_irq_source_dce60(
+		struct irq_service *irq_service,
+		uint32_t src_id,
+		uint32_t ext_id)
+{
+	switch (src_id) {
+	case VISLANDS30_IV_SRCID_D1_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK1;
+	case VISLANDS30_IV_SRCID_D2_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK2;
+	case VISLANDS30_IV_SRCID_D3_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK3;
+	case VISLANDS30_IV_SRCID_D4_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK4;
+	case VISLANDS30_IV_SRCID_D5_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK5;
+	case VISLANDS30_IV_SRCID_D6_VBLANK:
+		return DC_IRQ_SOURCE_VBLANK6;
+	case VISLANDS30_IV_SRCID_D1_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE1;
+	case VISLANDS30_IV_SRCID_D2_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE2;
+	case VISLANDS30_IV_SRCID_D3_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE3;
+	case VISLANDS30_IV_SRCID_D4_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE4;
+	case VISLANDS30_IV_SRCID_D5_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE5;
+	case VISLANDS30_IV_SRCID_D6_V_UPDATE_INT:
+		return DC_IRQ_SOURCE_VUPDATE6;
+	case VISLANDS30_IV_SRCID_D1_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP1;
+	case VISLANDS30_IV_SRCID_D2_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP2;
+	case VISLANDS30_IV_SRCID_D3_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP3;
+	case VISLANDS30_IV_SRCID_D4_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP4;
+	case VISLANDS30_IV_SRCID_D5_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP5;
+	case VISLANDS30_IV_SRCID_D6_GRPH_PFLIP:
+		return DC_IRQ_SOURCE_PFLIP6;
+
+	case VISLANDS30_IV_SRCID_HOTPLUG_DETECT_A:
+		/* generic src_id for all HPD and HPDRX interrupts */
+		switch (ext_id) {
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_A:
+			return DC_IRQ_SOURCE_HPD1;
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_B:
+			return DC_IRQ_SOURCE_HPD2;
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_C:
+			return DC_IRQ_SOURCE_HPD3;
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_D:
+			return DC_IRQ_SOURCE_HPD4;
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_E:
+			return DC_IRQ_SOURCE_HPD5;
+		case VISLANDS30_IV_EXTID_HOTPLUG_DETECT_F:
+			return DC_IRQ_SOURCE_HPD6;
+		case VISLANDS30_IV_EXTID_HPD_RX_A:
+			return DC_IRQ_SOURCE_HPD1RX;
+		case VISLANDS30_IV_EXTID_HPD_RX_B:
+			return DC_IRQ_SOURCE_HPD2RX;
+		case VISLANDS30_IV_EXTID_HPD_RX_C:
+			return DC_IRQ_SOURCE_HPD3RX;
+		case VISLANDS30_IV_EXTID_HPD_RX_D:
+			return DC_IRQ_SOURCE_HPD4RX;
+		case VISLANDS30_IV_EXTID_HPD_RX_E:
+			return DC_IRQ_SOURCE_HPD5RX;
+		case VISLANDS30_IV_EXTID_HPD_RX_F:
+			return DC_IRQ_SOURCE_HPD6RX;
+		default:
+			return DC_IRQ_SOURCE_INVALID;
+		}
+		break;
+
+	default:
+		return DC_IRQ_SOURCE_INVALID;
+	}
+}
+
+static const struct irq_service_funcs irq_service_funcs_dce60 = {
+		.to_dal_irq_source = to_dal_irq_source_dce60
+};
+
+static void dce60_irq_construct(
+	struct irq_service *irq_service,
+	struct irq_service_init_data *init_data)
+{
+	dal_irq_service_construct(irq_service, init_data);
+
+	irq_service->info = irq_source_info_dce60;
+	irq_service->funcs = &irq_service_funcs_dce60;
+}
+
+struct irq_service *dal_irq_service_dce60_create(
+	struct irq_service_init_data *init_data)
+{
+	struct irq_service *irq_service = kzalloc(sizeof(*irq_service),
+						  GFP_KERNEL);
+
+	if (!irq_service)
+		return NULL;
+
+	dce60_irq_construct(irq_service, init_data);
+	return irq_service;
+}
+
+
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.h b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.h
new file mode 100644
index 000000000000..294db29e8115
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.h
@@ -0,0 +1,40 @@
+/*
+ * Copyright 2020 Mauro Rossi <issor.oruam@gmail.com>
+ *
+ * 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: AMD
+ *
+ */
+
+#ifndef __DAL_IRQ_SERVICE_DCE60_H__
+#define __DAL_IRQ_SERVICE_DCE60_H__
+
+#include "../irq_service.h"
+
+enum dc_irq_source to_dal_irq_source_dce60(
+		struct irq_service *irq_service,
+		uint32_t src_id,
+		uint32_t ext_id);
+
+struct irq_service *dal_irq_service_dce60_create(
+	struct irq_service_init_data *init_data);
+
+#endif
+
diff --git a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
index 33053b9fe6bd..6bf27bde8724 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
@@ -32,6 +32,9 @@
 
 #include "dce110/irq_service_dce110.h"
 
+#if defined(CONFIG_DRM_AMD_DC_SI)
+#include "dce60/irq_service_dce60.h"
+#endif
 
 #include "dce80/irq_service_dce80.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h
index c3bbfe397e8d..330acaaed79a 100644
--- a/drivers/gpu/drm/amd/display/dc/os_types.h
+++ b/drivers/gpu/drm/amd/display/dc/os_types.h
@@ -55,6 +55,10 @@
 #include <asm/fpu/api.h>
 #define DC_FP_START() kernel_fpu_begin()
 #define DC_FP_END() kernel_fpu_end()
+#elif defined(CONFIG_ARM64)
+#include <asm/neon.h>
+#define DC_FP_START() kernel_neon_begin()
+#define DC_FP_END() kernel_neon_end()
 #elif defined(CONFIG_PPC64)
 #include <asm/switch_to.h>
 #include <asm/cputable.h>
diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
index b8040da94b9d..1053b165c139 100644
--- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
@@ -46,9 +46,10 @@ static void virtual_stream_encoder_dvi_set_stream_attribute(
 	struct dc_crtc_timing *crtc_timing,
 	bool is_dual_link) {}
 
-static void virtual_stream_encoder_set_mst_bandwidth(
+static void virtual_stream_encoder_set_throttled_vcp_size(
 	struct stream_encoder *enc,
-	struct fixed31_32 avg_time_slots_per_mtp) {}
+	struct fixed31_32 avg_time_slots_per_mtp)
+{}
 
 static void virtual_stream_encoder_update_hdmi_info_packets(
 	struct stream_encoder *enc,
@@ -87,6 +88,23 @@ static void virtual_enc_dp_set_odm_combine(
 	bool odm_combine)
 {}
 
+static void virtual_dig_connect_to_otg(
+		struct stream_encoder *enc,
+		int tg_inst)
+{}
+
+static void virtual_setup_stereo_sync(
+			struct stream_encoder *enc,
+			int tg_inst,
+			bool enable)
+{}
+
+static void virtual_stream_encoder_set_dsc_pps_info_packet(
+		struct stream_encoder *enc,
+		bool enable,
+		uint8_t *dsc_packed_pps)
+{}
+
 static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 	.dp_set_odm_combine =
 		virtual_enc_dp_set_odm_combine,
@@ -96,8 +114,8 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 		virtual_stream_encoder_hdmi_set_stream_attribute,
 	.dvi_set_stream_attribute =
 		virtual_stream_encoder_dvi_set_stream_attribute,
-	.set_mst_bandwidth =
-		virtual_stream_encoder_set_mst_bandwidth,
+	.set_throttled_vcp_size =
+		virtual_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =
 		virtual_stream_encoder_update_hdmi_info_packets,
 	.stop_hdmi_info_packets =
@@ -114,6 +132,9 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 	.audio_mute_control = virtual_audio_mute_control,
 	.set_avmute = virtual_stream_encoder_set_avmute,
 	.hdmi_reset_stream_attribute = virtual_stream_encoder_reset_hdmi_stream_attribute,
+	.dig_connect_to_otg = virtual_dig_connect_to_otg,
+	.setup_stereo_sync = virtual_setup_stereo_sync,
+	.dp_set_dsc_pps_info_packet = virtual_stream_encoder_set_dsc_pps_info_packet,
 };
 
 bool virtual_stream_encoder_construct(
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index e013875b89ed..d103ec1eaa73 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -36,11 +36,20 @@
 
 /* Firmware versioning. */
 #ifdef DMUB_EXPOSE_VERSION
-#define DMUB_FW_VERSION_GIT_HASH 0xe6d590b09
+#define DMUB_FW_VERSION_GIT_HASH 0x9cf8f05fe
 #define DMUB_FW_VERSION_MAJOR 0
 #define DMUB_FW_VERSION_MINOR 0
-#define DMUB_FW_VERSION_REVISION 25
-#define DMUB_FW_VERSION_UCODE ((DMUB_FW_VERSION_MAJOR << 24) | (DMUB_FW_VERSION_MINOR << 16) | DMUB_FW_VERSION_REVISION)
+#define DMUB_FW_VERSION_REVISION 35
+#define DMUB_FW_VERSION_TEST 0
+#define DMUB_FW_VERSION_VBIOS 0
+#define DMUB_FW_VERSION_HOTFIX 0
+#define DMUB_FW_VERSION_UCODE (((DMUB_FW_VERSION_MAJOR & 0xFF) << 24) | \
+		((DMUB_FW_VERSION_MINOR & 0xFF) << 16) | \
+		((DMUB_FW_VERSION_REVISION & 0xFF) << 8) | \
+		((DMUB_FW_VERSION_TEST & 0x1) << 7) | \
+		((DMUB_FW_VERSION_VBIOS & 0x1) << 6) | \
+		(DMUB_FW_VERSION_HOTFIX & 0x3F))
+
 #endif
 
 //<DMUB_TYPES>==================================================================
@@ -48,6 +57,7 @@
 
 #define SET_ABM_PIPE_GRADUALLY_DISABLE           0
 #define SET_ABM_PIPE_IMMEDIATELY_DISABLE         255
+#define SET_ABM_PIPE_IMMEDIATE_KEEP_GAIN_DISABLE 254
 #define SET_ABM_PIPE_NORMAL                      1
 
 /* Maximum number of streams on any ASIC. */
@@ -60,10 +70,6 @@
 #define PHYSICAL_ADDRESS_LOC union large_integer
 #endif
 
-#if defined(__cplusplus)
-extern "C" {
-#endif
-
 #ifndef dmub_memcpy
 #define dmub_memcpy(dest, source, bytes) memcpy((dest), (source), (bytes))
 #endif
@@ -72,6 +78,10 @@ extern "C" {
 #define dmub_memset(dest, val, bytes) memset((dest), (val), (bytes))
 #endif
 
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
 #ifndef dmub_udelay
 #define dmub_udelay(microseconds) udelay(microseconds)
 #endif
@@ -88,6 +98,7 @@ union dmub_psr_debug_flags {
 	struct {
 		uint32_t visual_confirm : 1;
 		uint32_t use_hw_lock_mgr : 1;
+		uint32_t log_line_nums : 1;
 	} bitfields;
 
 	uint32_t u32All;
@@ -160,7 +171,7 @@ union dmub_fw_boot_status {
 		uint32_t dal_fw : 1;
 		uint32_t mailbox_rdy : 1;
 		uint32_t optimized_init_done : 1;
-		uint32_t reserved : 29;
+		uint32_t restore_required : 1;
 	} bits;
 	uint32_t all;
 };
@@ -169,6 +180,7 @@ enum dmub_fw_boot_status_bit {
 	DMUB_FW_BOOT_STATUS_BIT_DAL_FIRMWARE = (1 << 0),
 	DMUB_FW_BOOT_STATUS_BIT_MAILBOX_READY = (1 << 1),
 	DMUB_FW_BOOT_STATUS_BIT_OPTIMIZED_INIT_DONE = (1 << 2),
+	DMUB_FW_BOOT_STATUS_BIT_RESTORE_REQUIRED = (1 << 3),
 };
 
 /* Register bit definition for SCRATCH15 */
@@ -204,6 +216,7 @@ enum dmub_cmd_vbios_type {
 	DMUB_CMD__VBIOS_DIG1_TRANSMITTER_CONTROL = 1,
 	DMUB_CMD__VBIOS_SET_PIXEL_CLOCK = 2,
 	DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING = 3,
+	DMUB_CMD__VBIOS_LVTMA_CONTROL = 15,
 };
 
 //==============================================================================
@@ -287,9 +300,17 @@ enum dmub_cmd_type {
 	DMUB_CMD__PSR = 64,
 	DMUB_CMD__ABM = 66,
 	DMUB_CMD__HW_LOCK = 69,
+	DMUB_CMD__DP_AUX_ACCESS = 70,
+	DMUB_CMD__OUTBOX1_ENABLE = 71,
 	DMUB_CMD__VBIOS = 128,
 };
 
+enum dmub_out_cmd_type {
+	DMUB_OUT_CMD__NULL = 0,
+	DMUB_OUT_CMD__DP_AUX_REPLY = 1,
+	DMUB_OUT_CMD__DP_HPD_NOTIFY = 2,
+};
+
 #pragma pack(push, 1)
 
 struct dmub_cmd_header {
@@ -445,6 +466,78 @@ struct dmub_rb_cmd_dpphy_init {
 	uint8_t reserved[60];
 };
 
+enum dp_aux_request_action {
+	DP_AUX_REQ_ACTION_I2C_WRITE		= 0x00,
+	DP_AUX_REQ_ACTION_I2C_READ		= 0x10,
+	DP_AUX_REQ_ACTION_I2C_STATUS_REQ	= 0x20,
+	DP_AUX_REQ_ACTION_I2C_WRITE_MOT		= 0x40,
+	DP_AUX_REQ_ACTION_I2C_READ_MOT		= 0x50,
+	DP_AUX_REQ_ACTION_I2C_STATUS_REQ_MOT	= 0x60,
+	DP_AUX_REQ_ACTION_DPCD_WRITE		= 0x80,
+	DP_AUX_REQ_ACTION_DPCD_READ		= 0x90
+};
+
+/* DP AUX command */
+struct aux_transaction_parameters {
+	uint8_t is_i2c_over_aux;
+	uint8_t action;
+	uint8_t length;
+	uint8_t pad;
+	uint32_t address;
+	uint8_t data[16];
+};
+
+struct dmub_cmd_dp_aux_control_data {
+	uint32_t handle;
+	uint8_t port_index;
+	uint8_t sw_crc_enabled;
+	uint16_t timeout;
+	struct aux_transaction_parameters dpaux;
+};
+
+struct dmub_rb_cmd_dp_aux_access {
+	struct dmub_cmd_header header;
+	struct dmub_cmd_dp_aux_control_data aux_control;
+};
+
+struct dmub_rb_cmd_outbox1_enable {
+	struct dmub_cmd_header header;
+	uint32_t enable;
+};
+
+/* DP AUX Reply command - OutBox Cmd */
+struct aux_reply_data {
+	uint8_t command;
+	uint8_t length;
+	uint8_t pad[2];
+	uint8_t data[16];
+};
+
+struct aux_reply_control_data {
+	uint32_t handle;
+	uint8_t phy_port_index;
+	uint8_t result;
+	uint16_t pad;
+};
+
+struct dmub_rb_cmd_dp_aux_reply {
+	struct dmub_cmd_header header;
+	struct aux_reply_control_data control;
+	struct aux_reply_data reply_data;
+};
+
+struct dp_hpd_data {
+	uint8_t phy_port_index;
+	uint8_t hpd_type;
+	uint8_t hpd_status;
+	uint8_t pad;
+};
+
+struct dmub_rb_cmd_dp_hpd_notify {
+	struct dmub_cmd_header header;
+	struct dp_hpd_data hpd_data;
+};
+
 /*
  * Command IDs should be treated as stable ABI.
  * Do not reuse or modify IDs.
@@ -674,8 +767,15 @@ union dmub_rb_cmd {
 	struct dmub_rb_cmd_abm_set_ambient_level abm_set_ambient_level;
 	struct dmub_rb_cmd_abm_set_pwm_frac abm_set_pwm_frac;
 	struct dmub_rb_cmd_abm_init_config abm_init_config;
+	struct dmub_rb_cmd_dp_aux_access dp_aux_access;
+	struct dmub_rb_cmd_outbox1_enable outbox1_enable;
 };
 
+union dmub_rb_out_cmd {
+	struct dmub_rb_cmd_common cmd_common;
+	struct dmub_rb_cmd_dp_aux_reply dp_aux_reply;
+	struct dmub_rb_cmd_dp_hpd_notify dp_hpd_notify;
+};
 #pragma pack(pop)
 
 
@@ -748,6 +848,25 @@ static inline bool dmub_rb_push_front(struct dmub_rb *rb,
 	return true;
 }
 
+static inline bool dmub_rb_out_push_front(struct dmub_rb *rb,
+				      const union dmub_rb_out_cmd *cmd)
+{
+	uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt;
+	const uint8_t *src = (uint8_t *)cmd;
+
+	if (dmub_rb_full(rb))
+		return false;
+
+	dmub_memcpy(dst, src, DMUB_RB_CMD_SIZE);
+
+	rb->wrpt += DMUB_RB_CMD_SIZE;
+
+	if (rb->wrpt >= rb->capacity)
+		rb->wrpt %= rb->capacity;
+
+	return true;
+}
+
 static inline bool dmub_rb_front(struct dmub_rb *rb,
 				 union dmub_rb_cmd  *cmd)
 {
@@ -761,6 +880,23 @@ static inline bool dmub_rb_front(struct dmub_rb *rb,
 	return true;
 }
 
+static inline bool dmub_rb_out_front(struct dmub_rb *rb,
+				 union dmub_rb_out_cmd  *cmd)
+{
+	const uint64_t volatile *src = (const uint64_t volatile *)(rb->base_address) + rb->rptr / sizeof(uint64_t);
+	uint64_t *dst = (uint64_t *)cmd;
+	int i;
+
+	if (dmub_rb_empty(rb))
+		return false;
+
+	// copying data
+	for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++)
+		*dst++ = *src++;
+
+	return true;
+}
+
 static inline bool dmub_rb_pop_front(struct dmub_rb *rb)
 {
 	if (dmub_rb_empty(rb))
@@ -781,12 +917,10 @@ static inline void dmub_rb_flush_pending(const struct dmub_rb *rb)
 
 	while (rptr != wptr) {
 		uint64_t volatile *data = (uint64_t volatile *)rb->base_address + rptr / sizeof(uint64_t);
-		//uint64_t volatile *p = (uint64_t volatile *)data;
-		uint64_t temp;
 		int i;
 
 		for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++)
-			temp = *data++;
+			*data++;
 
 		rptr += DMUB_RB_CMD_SIZE;
 		if (rptr >= rb->capacity)
diff --git a/drivers/gpu/drm/amd/display/include/bios_parser_types.h b/drivers/gpu/drm/amd/display/include/bios_parser_types.h
index 21011edea337..7c782924c941 100644
--- a/drivers/gpu/drm/amd/display/include/bios_parser_types.h
+++ b/drivers/gpu/drm/amd/display/include/bios_parser_types.h
@@ -318,4 +318,10 @@ struct bp_encoder_cap_info {
 	uint32_t RESERVED:27;
 };
 
+struct bp_soc_bb_info {
+	uint32_t dram_clock_change_latency_100ns;
+	uint32_t dram_sr_exit_latency_100ns;
+	uint32_t dram_sr_enter_exit_latency_100ns;
+};
+
 #endif /*__DAL_BIOS_PARSER_TYPES_H__ */
diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h
index abeb58d544b1..b267987aed06 100644
--- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h
+++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h
@@ -30,6 +30,34 @@
  * ASIC internal revision ID
  */
 
+/* DCE60 (based on si_id.h in GPUOpen-Tools CodeXL) */
+#define SI_TAHITI_P_A0    0x01
+#define SI_TAHITI_P_B0    0x05
+#define SI_TAHITI_P_B1    0x06
+#define SI_PITCAIRN_PM_A0 0x14
+#define SI_PITCAIRN_PM_A1 0x15
+#define SI_CAPEVERDE_M_A0 0x28
+#define SI_CAPEVERDE_M_A1 0x29
+#define SI_OLAND_M_A0     0x3C
+#define SI_HAINAN_V_A0    0x46
+
+#define SI_UNKNOWN        0xFF
+
+#define ASIC_REV_IS_TAHITI_P(rev) \
+	((rev >= SI_TAHITI_P_A0) && (rev < SI_PITCAIRN_PM_A0))
+
+#define ASIC_REV_IS_PITCAIRN_PM(rev) \
+	((rev >= SI_PITCAIRN_PM_A0) && (rev < SI_CAPEVERDE_M_A0))
+
+#define ASIC_REV_IS_CAPEVERDE_M(rev) \
+	((rev >= SI_CAPEVERDE_M_A0) && (rev < SI_OLAND_M_A0))
+
+#define ASIC_REV_IS_OLAND_M(rev) \
+	((rev >= SI_OLAND_M_A0) && (rev < SI_HAINAN_V_A0))
+
+#define ASIC_REV_IS_HAINAN_V(rev) \
+	((rev >= SI_HAINAN_V_A0) && (rev < SI_UNKNOWN))
+
 /* DCE80 (based on ci_id.h in Perforce) */
 #define	CI_BONAIRE_M_A0 0x14
 #define	CI_BONAIRE_M_A1	0x15
@@ -181,6 +209,17 @@ enum {
 /*
  * ASIC chip ID
  */
+
+/* DCE60 */
+#define DEVICE_ID_SI_TAHITI_P_6780 0x6780
+#define DEVICE_ID_SI_PITCAIRN_PM_6800 0x6800
+#define DEVICE_ID_SI_PITCAIRN_PM_6808 0x6808
+#define DEVICE_ID_SI_CAPEVERDE_M_6820 0x6820
+#define DEVICE_ID_SI_CAPEVERDE_M_6828 0x6828
+#define DEVICE_ID_SI_OLAND_M_6600 0x6600
+#define DEVICE_ID_SI_OLAND_M_6608 0x6608
+#define DEVICE_ID_SI_HAINAN_V_6660 0x6660
+
 /* DCE80 */
 #define DEVICE_ID_KALINDI_9834 0x9834
 #define DEVICE_ID_TEMASH_9839 0x9839
@@ -190,6 +229,7 @@ enum {
 #define DEVICE_ID_RENOIR_1636 0x1636
 
 /* Asic Family IDs for different asic family. */
+#define FAMILY_SI 110 /* Southern Islands: Tahiti (P), Pitcairn (PM), Cape Verde (M), Oland (M), Hainan (V) */
 #define FAMILY_CI 120 /* Sea Islands: Hawaii (P), Bonaire (M) */
 #define FAMILY_KV 125 /* Fusion => Kaveri: Spectre, Spooky; Kabini: Kalindi */
 #define FAMILY_VI 130 /* Volcanic Islands: Iceland (V), Tonga (M) */
diff --git a/drivers/gpu/drm/amd/display/include/dal_types.h b/drivers/gpu/drm/amd/display/include/dal_types.h
index b67c9fa6b9cd..8aaa3af69202 100644
--- a/drivers/gpu/drm/amd/display/include/dal_types.h
+++ b/drivers/gpu/drm/amd/display/include/dal_types.h
@@ -34,6 +34,9 @@ struct dc_bios;
 
 enum dce_version {
 	DCE_VERSION_UNKNOWN = (-1),
+	DCE_VERSION_6_0,
+	DCE_VERSION_6_1,
+	DCE_VERSION_6_4,
 	DCE_VERSION_8_0,
 	DCE_VERSION_8_1,
 	DCE_VERSION_8_3,
diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h
index 550f46e9b95f..7392a89e771f 100644
--- a/drivers/gpu/drm/amd/display/include/link_service_types.h
+++ b/drivers/gpu/drm/amd/display/include/link_service_types.h
@@ -80,6 +80,7 @@ struct link_training_settings {
 
 	uint16_t cr_pattern_time;
 	uint16_t eq_pattern_time;
+	enum dc_dp_training_pattern pattern_for_cr;
 	enum dc_dp_training_pattern pattern_for_eq;
 
 	bool enhanced_framing;
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
index e9fbd94f8635..20e554e771d1 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
@@ -470,6 +470,14 @@ enum mod_hdcp_status mod_hdcp_process_event(struct mod_hdcp *hdcp,
 		if (reset_status != MOD_HDCP_STATUS_SUCCESS)
 			push_error_status(hdcp, reset_status);
 	}
+
+	/* Clear CP_IRQ status if needed */
+	if (event_ctx.event == MOD_HDCP_EVENT_CPIRQ) {
+		status = mod_hdcp_clear_cp_irq_status(hdcp);
+		if (status != MOD_HDCP_STATUS_SUCCESS)
+			push_error_status(hdcp, status);
+	}
+
 	return status;
 }
 
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
index b0cefed2eb02..6c678cfb82e3 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
@@ -386,6 +386,7 @@ enum mod_hdcp_status mod_hdcp_write_eks(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_write_repeater_auth_ack(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_write_stream_manage(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp);
+enum mod_hdcp_status mod_hdcp_clear_cp_irq_status(struct mod_hdcp *hdcp);
 
 /* hdcp version helpers */
 static inline uint8_t is_dp_hdcp(struct mod_hdcp *hdcp)
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
index bb5130f4228d..f7b5583ee609 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
@@ -30,6 +30,8 @@
 #define KSV_READ_SIZE 0xf	/* 0x6803b - 0x6802c */
 #define HDCP_MAX_AUX_TRANSACTION_SIZE 16
 
+#define DP_CP_IRQ (1 << 2)
+
 enum mod_hdcp_ddc_message_id {
 	MOD_HDCP_MESSAGE_ID_INVALID = -1,
 
@@ -645,3 +647,18 @@ enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp)
 		status = MOD_HDCP_STATUS_INVALID_OPERATION;
 	return status;
 }
+
+enum mod_hdcp_status mod_hdcp_clear_cp_irq_status(struct mod_hdcp *hdcp)
+{
+	uint8_t clear_cp_irq_bit = DP_CP_IRQ;
+	uint32_t size = 1;
+
+	if (is_dp_hdcp(hdcp)) {
+		uint32_t cp_irq_addrs = (hdcp->connection.link.dp.rev >= 0x14)
+				? DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0:DP_DEVICE_SERVICE_IRQ_VECTOR;
+		return hdcp->config.ddc.funcs.write_dpcd(hdcp->config.ddc.handle, cp_irq_addrs,
+				&clear_cp_irq_bit, size) ? MOD_HDCP_STATUS_SUCCESS : MOD_HDCP_STATUS_DDC_FAILURE;
+	}
+
+	return MOD_HDCP_STATUS_INVALID_OPERATION;
+}
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
index 13c57ff2abdc..1ab813b4fd14 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h
@@ -37,6 +37,6 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream,
 		struct dc_info_packet *info_packet);
 
 void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
-		struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue);
+		struct dc_info_packet *info_packet);
 
 #endif
diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
index 7cd8a43d1889..0fdf7a3e96de 100644
--- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
+++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c
@@ -421,15 +421,13 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream,
  *****************************************************************************
  */
 void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
-		struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue)
+		struct dc_info_packet *info_packet)
 {
 		unsigned int length = 5;
 		bool hdmi_vic_mode = false;
 		uint8_t checksum = 0;
 		uint32_t i = 0;
 		enum dc_timing_3d_format format;
-		bool bALLM = (bool)ALLMEnabled;
-		bool bALLMVal = (bool)ALLMValue;
 
 		info_packet->valid = false;
 		format = stream->timing.timing_3d_format;
@@ -442,20 +440,13 @@ void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
 				&& format == TIMING_3D_FORMAT_NONE)
 			hdmi_vic_mode = true;
 
-		if ((format == TIMING_3D_FORMAT_NONE) && !hdmi_vic_mode && !bALLM)
+		if ((format == TIMING_3D_FORMAT_NONE) && !hdmi_vic_mode)
 			return;
 
 		info_packet->sb[1] = 0x03;
 		info_packet->sb[2] = 0x0C;
 		info_packet->sb[3] = 0x00;
 
-		if (bALLM) {
-			info_packet->sb[1] = 0xD8;
-			info_packet->sb[2] = 0x5D;
-			info_packet->sb[3] = 0xC4;
-			info_packet->sb[4] = HF_VSIF_VERSION;
-		}
-
 		if (format != TIMING_3D_FORMAT_NONE)
 			info_packet->sb[4] = (2 << 5);
 
@@ -490,9 +481,6 @@ void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream,
 		info_packet->hb1 = 0x01;
 		info_packet->hb2 = (uint8_t) (length);
 
-		if (bALLM)
-			info_packet->sb[5] = (info_packet->sb[5] & ~0x02) | (bALLMVal << 1);
-
 		checksum += info_packet->hb0;
 		checksum += info_packet->hb1;
 		checksum += info_packet->hb2;
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index e98c84ef206f..10dc481ecbc4 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -47,6 +47,40 @@ enum amd_apu_flags {
 	AMD_APU_IS_RENOIR = 0x00000008UL,
 };
 
+/**
+* DOC: IP Blocks
+*
+* GPUs are composed of IP (intellectual property) blocks. These
+* IP blocks provide various functionalities: display, graphics,
+* video decode, etc. The IP blocks that comprise a particular GPU
+* are listed in the GPU's respective SoC file. amdgpu_device.c
+* acquires the list of IP blocks for the GPU in use on initialization.
+* It can then operate on this list to perform standard driver operations
+* such as: init, fini, suspend, resume, etc.
+* 
+*
+* IP block implementations are named using the following convention:
+* <functionality>_v<version> (E.g.: gfx_v6_0).
+*/
+
+/**
+* enum amd_ip_block_type - Used to classify IP blocks by functionality.
+*
+* @AMD_IP_BLOCK_TYPE_COMMON: GPU Family
+* @AMD_IP_BLOCK_TYPE_GMC: Graphics Memory Controller
+* @AMD_IP_BLOCK_TYPE_IH: Interrupt Handler
+* @AMD_IP_BLOCK_TYPE_SMC: System Management Controller
+* @AMD_IP_BLOCK_TYPE_PSP: Platform Security Processor
+* @AMD_IP_BLOCK_TYPE_DCE: Display and Compositing Engine
+* @AMD_IP_BLOCK_TYPE_GFX: Graphics and Compute Engine
+* @AMD_IP_BLOCK_TYPE_SDMA: System DMA Engine
+* @AMD_IP_BLOCK_TYPE_UVD: Unified Video Decoder
+* @AMD_IP_BLOCK_TYPE_VCE: Video Compression Engine
+* @AMD_IP_BLOCK_TYPE_ACP: Audio Co-Processor
+* @AMD_IP_BLOCK_TYPE_VCN: Video Core/Codec Next
+* @AMD_IP_BLOCK_TYPE_MES: Micro-Engine Scheduler
+* @AMD_IP_BLOCK_TYPE_JPEG: JPEG Engine
+*/
 enum amd_ip_block_type {
 	AMD_IP_BLOCK_TYPE_COMMON,
 	AMD_IP_BLOCK_TYPE_GMC,
@@ -128,6 +162,34 @@ enum amd_powergating_state {
 #define AMD_PG_SUPPORT_ATHUB			(1 << 16)
 #define AMD_PG_SUPPORT_JPEG			(1 << 17)
 
+/**
+ * enum PP_FEATURE_MASK - Used to mask power play features.
+ *
+ * @PP_SCLK_DPM_MASK: Dynamic adjustment of the system (graphics) clock.
+ * @PP_MCLK_DPM_MASK: Dynamic adjustment of the memory clock.
+ * @PP_PCIE_DPM_MASK: Dynamic adjustment of PCIE clocks and lanes.
+ * @PP_SCLK_DEEP_SLEEP_MASK: System (graphics) clock deep sleep.
+ * @PP_POWER_CONTAINMENT_MASK: Power containment.
+ * @PP_UVD_HANDSHAKE_MASK: Unified video decoder handshake.
+ * @PP_SMC_VOLTAGE_CONTROL_MASK: Dynamic voltage control.
+ * @PP_VBI_TIME_SUPPORT_MASK: Vertical blank interval support.
+ * @PP_ULV_MASK: Ultra low voltage.
+ * @PP_ENABLE_GFX_CG_THRU_SMU: SMU control of GFX engine clockgating.
+ * @PP_CLOCK_STRETCH_MASK: Clock stretching.
+ * @PP_OD_FUZZY_FAN_CONTROL_MASK: Overdrive fuzzy fan control.
+ * @PP_SOCCLK_DPM_MASK: Dynamic adjustment of the SoC clock.
+ * @PP_DCEFCLK_DPM_MASK: Dynamic adjustment of the Display Controller Engine Fabric clock.
+ * @PP_OVERDRIVE_MASK: Over- and under-clocking support.
+ * @PP_GFXOFF_MASK: Dynamic graphics engine power control.
+ * @PP_ACG_MASK: Adaptive clock generator.
+ * @PP_STUTTER_MODE: Stutter mode.
+ * @PP_AVFS_MASK: Adaptive voltage and frequency scaling.
+ *
+ * To override these settings on boot, append amdgpu.ppfeaturemask=<mask> to
+ * the kernel's command line parameters. This is usually done through a system's
+ * boot loader (E.g. GRUB). If manually loading the driver, pass
+ * ppfeaturemask=<mask> as a modprobe parameter.
+ */
 enum PP_FEATURE_MASK {
 	PP_SCLK_DPM_MASK = 0x1,
 	PP_MCLK_DPM_MASK = 0x2,
@@ -165,56 +227,59 @@ enum DC_DEBUG_MASK {
 };
 
 enum amd_dpm_forced_level;
+
 /**
  * struct amd_ip_funcs - general hooks for managing amdgpu IP Blocks
+ * @name: Name of IP block
+ * @early_init: sets up early driver state (pre sw_init),
+ *              does not configure hw - Optional
+ * @late_init: sets up late driver/hw state (post hw_init) - Optional
+ * @sw_init: sets up driver state, does not configure hw
+ * @sw_fini: tears down driver state, does not configure hw
+ * @hw_init: sets up the hw state
+ * @hw_fini: tears down the hw state
+ * @late_fini: final cleanup
+ * @suspend: handles IP specific hw/sw changes for suspend
+ * @resume: handles IP specific hw/sw changes for resume
+ * @is_idle: returns current IP block idle status
+ * @wait_for_idle: poll for idle
+ * @check_soft_reset: check soft reset the IP block
+ * @pre_soft_reset: pre soft reset the IP block
+ * @soft_reset: soft reset the IP block
+ * @post_soft_reset: post soft reset the IP block
+ * @set_clockgating_state: enable/disable cg for the IP block
+ * @set_powergating_state: enable/disable pg for the IP block
+ * @get_clockgating_state: get current clockgating status
+ * @enable_umd_pstate: enable UMD powerstate
+ *
+ * These hooks provide an interface for controlling the operational state
+ * of IP blocks. After acquiring a list of IP blocks for the GPU in use,
+ * the driver can make chip-wide state changes by walking this list and
+ * making calls to hooks from each IP block. This list is ordered to ensure
+ * that the driver initializes the IP blocks in a safe sequence.
  */
 struct amd_ip_funcs {
-	/** @name: Name of IP block */
 	char *name;
-	/**
-	 * @early_init:
-	 *
-	 * sets up early driver state (pre sw_init),
-	 * does not configure hw - Optional
-	 */
 	int (*early_init)(void *handle);
-	/** @late_init: sets up late driver/hw state (post hw_init) - Optional */
 	int (*late_init)(void *handle);
-	/** @sw_init: sets up driver state, does not configure hw */
 	int (*sw_init)(void *handle);
-	/** @sw_fini: tears down driver state, does not configure hw */
 	int (*sw_fini)(void *handle);
-	/** @hw_init: sets up the hw state */
 	int (*hw_init)(void *handle);
-	/** @hw_fini: tears down the hw state */
 	int (*hw_fini)(void *handle);
-	/** @late_fini: final cleanup */
 	void (*late_fini)(void *handle);
-	/** @suspend: handles IP specific hw/sw changes for suspend */
 	int (*suspend)(void *handle);
-	/** @resume: handles IP specific hw/sw changes for resume */
 	int (*resume)(void *handle);
-	/** @is_idle: returns current IP block idle status */
 	bool (*is_idle)(void *handle);
-	/** @wait_for_idle: poll for idle */
 	int (*wait_for_idle)(void *handle);
-	/** @check_soft_reset: check soft reset the IP block */
 	bool (*check_soft_reset)(void *handle);
-	/** @pre_soft_reset: pre soft reset the IP block */
 	int (*pre_soft_reset)(void *handle);
-	/** @soft_reset: soft reset the IP block */
 	int (*soft_reset)(void *handle);
-	/** @post_soft_reset: post soft reset the IP block */
 	int (*post_soft_reset)(void *handle);
-	/** @set_clockgating_state: enable/disable cg for the IP block */
 	int (*set_clockgating_state)(void *handle,
 				     enum amd_clockgating_state state);
-	/** @set_powergating_state: enable/disable pg for the IP block */
 	int (*set_powergating_state)(void *handle,
 				     enum amd_powergating_state state);
-	/** @get_clockgating_state: get current clockgating status */
 	void (*get_clockgating_state)(void *handle, u32 *flags);
-	/** @enable_umd_pstate: enable UMD powerstate */
 	int (*enable_umd_pstate)(void *handle, enum amd_dpm_forced_level *level);
 };
 
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h
index 27bb8c1ab858..b6f74bf4af02 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_12_0_offset.h
@@ -7376,8 +7376,6 @@
 #define mmCRTC4_CRTC_DRR_CONTROL                                                                       0x0f3e
 #define mmCRTC4_CRTC_DRR_CONTROL_BASE_IDX                                                              2
 
-#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0                                                                  0x395d
-#define mmDCHUBBUB_SDPIF_MMIO_CNTRL_0_BASE_IDX                                                         2
 
 // addressBlock: dce_dc_fmt4_dispdec
 // base address: 0x2000
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_d.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_d.h
index ae798f768853..9de01ae574c0 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_d.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_d.h
@@ -4444,14 +4444,90 @@
 
 /* Registers that spilled out of sid.h */
 #define mmDATA_FORMAT                              0x1AC0
+#define mmLB0_DATA_FORMAT                               0x1AC0
+#define mmLB1_DATA_FORMAT                               0x1DC0
+#define mmLB2_DATA_FORMAT                               0x40C0
+#define mmLB3_DATA_FORMAT                               0x43C0
+#define mmLB4_DATA_FORMAT                               0x46C0
+#define mmLB5_DATA_FORMAT                               0x49C0
 #define mmDESKTOP_HEIGHT                           0x1AC1
+#define mmLB0_DESKTOP_HEIGHT                            0x1AC1
+#define mmLB1_DESKTOP_HEIGHT                            0x1DC1
+#define mmLB2_DESKTOP_HEIGHT                            0x40C1
+#define mmLB3_DESKTOP_HEIGHT                            0x43C1
+#define mmLB4_DESKTOP_HEIGHT                            0x46C1
+#define mmLB5_DESKTOP_HEIGHT                            0x49C1
 #define mmDC_LB_MEMORY_SPLIT                       0x1AC3
+#define mmLB0_DC_LB_MEMORY_SPLIT                        0x1AC3
+#define mmLB1_DC_LB_MEMORY_SPLIT                        0x1DC3
+#define mmLB2_DC_LB_MEMORY_SPLIT                        0x40C3
+#define mmLB3_DC_LB_MEMORY_SPLIT                        0x43C3
+#define mmLB4_DC_LB_MEMORY_SPLIT                        0x46C3
+#define mmLB5_DC_LB_MEMORY_SPLIT                        0x49C3
+#define mmDC_LB_MEM_SIZE                                0x1AC4
+#define mmLB0_DC_LB_MEM_SIZE                            0x1AC4
+#define mmLB1_DC_LB_MEM_SIZE                            0x1DC4
+#define mmLB2_DC_LB_MEM_SIZE                            0x40C4
+#define mmLB3_DC_LB_MEM_SIZE                            0x43C4
+#define mmLB4_DC_LB_MEM_SIZE                            0x46C4
+#define mmLB5_DC_LB_MEM_SIZE                            0x49C4
 #define mmPRIORITY_A_CNT                           0x1AC6
+#define mmLB0_PRIORITY_A_CNT                            0x1AC6
+#define mmLB1_PRIORITY_A_CNT                            0x1DC6
+#define mmLB2_PRIORITY_A_CNT                            0x40C6
+#define mmLB3_PRIORITY_A_CNT                            0x43C6
+#define mmLB4_PRIORITY_A_CNT                            0x46C6
+#define mmLB5_PRIORITY_A_CNT                            0x49C6
 #define mmPRIORITY_B_CNT                           0x1AC7
+#define mmLB0_PRIORITY_B_CNT                            0x1AC7
+#define mmLB1_PRIORITY_B_CNT                            0x1DC7
+#define mmLB2_PRIORITY_B_CNT                            0x40C7
+#define mmLB3_PRIORITY_B_CNT                            0x43C7
+#define mmLB4_PRIORITY_B_CNT                            0x46C7
+#define mmLB5_PRIORITY_B_CNT                            0x49C7
 #define mmDPG_PIPE_ARBITRATION_CONTROL3            0x1B32
+#define mmDMIF_PG0_DPG_PIPE_ARBITRATION_CONTROL3        0x1B32
+#define mmDMIF_PG1_DPG_PIPE_ARBITRATION_CONTROL3        0x1E32
+#define mmDMIF_PG2_DPG_PIPE_ARBITRATION_CONTROL3        0x4132
+#define mmDMIF_PG3_DPG_PIPE_ARBITRATION_CONTROL3        0x4432
+#define mmDMIF_PG4_DPG_PIPE_ARBITRATION_CONTROL3        0x4732
+#define mmDMIF_PG5_DPG_PIPE_ARBITRATION_CONTROL3        0x4A32
 #define mmINT_MASK                                 0x1AD0
+#define mmLB0_INT_MASK                                  0x1AD0
+#define mmLB1_INT_MASK                                  0x1DD0
+#define mmLB2_INT_MASK                                  0x40D0
+#define mmLB3_INT_MASK                                  0x43D0
+#define mmLB4_INT_MASK                                  0x46D0
+#define mmLB5_INT_MASK                                  0x49D0
 #define mmVLINE_STATUS                             0x1AEE
+#define mmLB0_VLINE_STATUS                              0x1AEE
+#define mmLB1_VLINE_STATUS                              0x1DEE
+#define mmLB2_VLINE_STATUS                              0x40EE
+#define mmLB3_VLINE_STATUS                              0x43EE
+#define mmLB4_VLINE_STATUS                              0x46EE
+#define mmLB5_VLINE_STATUS                              0x49EE
 #define mmVBLANK_STATUS                            0x1AEF
+#define mmLB0_VBLANK_STATUS                             0x1AEF
+#define mmLB1_VBLANK_STATUS                             0x1DEF
+#define mmLB2_VBLANK_STATUS                             0x40EF
+#define mmLB3_VBLANK_STATUS                             0x43EF
+#define mmLB4_VBLANK_STATUS                             0x46EF
+#define mmLB5_VBLANK_STATUS                             0x49EF
 
+#define mmSCL_HORZ_FILTER_INIT_RGB_LUMA            0x1B4C
+#define mmSCL0_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x1B4C
+#define mmSCL1_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x1E4C
+#define mmSCL2_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x414C
+#define mmSCL3_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x444C
+#define mmSCL4_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x474C
+#define mmSCL5_SCL_HORZ_FILTER_INIT_RGB_LUMA            0x4A4C
+
+#define mmSCL_HORZ_FILTER_INIT_CHROMA              0x1B4D
+#define mmSCL0_SCL_HORZ_FILTER_INIT_CHROMA              0x1B4D
+#define mmSCL1_SCL_HORZ_FILTER_INIT_CHROMA              0x1E4D
+#define mmSCL2_SCL_HORZ_FILTER_INIT_CHROMA              0x414D
+#define mmSCL3_SCL_HORZ_FILTER_INIT_CHROMA              0x444D
+#define mmSCL4_SCL_HORZ_FILTER_INIT_CHROMA              0x474D
+#define mmSCL5_SCL_HORZ_FILTER_INIT_CHROMA              0x4A4D
 
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
index abe05bc80752..41c4a46ce357 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
@@ -2076,6 +2076,8 @@
 #define CRTC_CONTROL__CRTC_START_POINT_CNTL__SHIFT 0x0000000c
 #define CRTC_CONTROL__CRTC_SYNC_RESET_SEL_MASK 0x00000010L
 #define CRTC_CONTROL__CRTC_SYNC_RESET_SEL__SHIFT 0x00000004
+#define CRTC_CONTROL__CRTC_PREFETCH_EN_MASK                0x10000000L
+#define	CRTC_CONTROL__CRTC_PREFETCH_EN__SHIFT              0x0000001c
 #define CRTC_COUNT_CONTROL__CRTC_HORZ_COUNT_BY2_EN_MASK 0x00000001L
 #define CRTC_COUNT_CONTROL__CRTC_HORZ_COUNT_BY2_EN__SHIFT 0x00000000
 #define CRTC_COUNT_CONTROL__CRTC_HORZ_REPETITION_COUNT_MASK 0x0000001eL
@@ -6364,6 +6366,8 @@
 #define DPG_PIPE_ARBITRATION_CONTROL2__TIME_WEIGHT__SHIFT 0x00000000
 #define DPG_PIPE_ARBITRATION_CONTROL2__URGENCY_WEIGHT_MASK 0xffff0000L
 #define DPG_PIPE_ARBITRATION_CONTROL2__URGENCY_WEIGHT__SHIFT 0x00000010
+#define DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK_MASK 0x00030000L
+#define	DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK__SHIFT 0x00000010
 #define DPG_PIPE_DPM_CONTROL__DPM_ENABLE_MASK 0x00000001L
 #define DPG_PIPE_DPM_CONTROL__DPM_ENABLE__SHIFT 0x00000000
 #define DPG_PIPE_DPM_CONTROL__MCLK_CHANGE_ENABLE_MASK 0x00000010L
@@ -6384,6 +6388,8 @@
 #define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_NOT_SELF_REFRESH_DURING_REQUEST__SHIFT 0x00000008
 #define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_URGENT_DURING_REQUEST_MASK 0x00000010L
 #define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_URGENT_DURING_REQUEST__SHIFT 0x00000004
+#define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_WATERMARK_MASK_MASK 0x00003000L
+#define	DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_WATERMARK_MASK__SHIFT 0x0000000c
 #define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_WATERMARK_MASK 0xffff0000L
 #define DPG_PIPE_NB_PSTATE_CHANGE_CONTROL__NB_PSTATE_CHANGE_WATERMARK__SHIFT 0x00000010
 #define DPG_PIPE_STUTTER_CONTROL_NONLPTCH__STUTTER_ENABLE_NONLPTCH_MASK 0x00000001L
@@ -6406,6 +6412,8 @@
 #define DPG_PIPE_STUTTER_CONTROL_NONLPTCH__STUTTER_WM_HIGH_FORCE_ON_NONLPTCH__SHIFT 0x00000008
 #define DPG_PIPE_STUTTER_CONTROL__STUTTER_ENABLE_MASK 0x00000001L
 #define DPG_PIPE_STUTTER_CONTROL__STUTTER_ENABLE__SHIFT 0x00000000
+#define DPG_PIPE_STUTTER_CONTROL__STUTTER_EXIT_SELF_REFRESH_WATERMARK_MASK_MASK 0x00003000L
+#define	DPG_PIPE_STUTTER_CONTROL__STUTTER_EXIT_SELF_REFRESH_WATERMARK_MASK__SHIFT 0x0000000c
 #define DPG_PIPE_STUTTER_CONTROL__STUTTER_EXIT_SELF_REFRESH_WATERMARK_MASK 0xffff0000L
 #define DPG_PIPE_STUTTER_CONTROL__STUTTER_EXIT_SELF_REFRESH_WATERMARK__SHIFT 0x00000010
 #define DPG_PIPE_STUTTER_CONTROL__STUTTER_IGNORE_CURSOR_MASK 0x00000010L
@@ -7256,6 +7264,8 @@
 #define GRPH_CONTROL__GRPH_FORMAT__SHIFT 0x00000008
 #define GRPH_CONTROL__GRPH_MACRO_TILE_ASPECT_MASK 0x000c0000L
 #define GRPH_CONTROL__GRPH_MACRO_TILE_ASPECT__SHIFT 0x00000012
+#define GRPH_CONTROL__GRPH_ARRAY_MODE_MASK                 0x00f00000L
+#define	GRPH_CONTROL__GRPH_ARRAY_MODE__SHIFT               0x00000014
 #define GRPH_CONTROL__GRPH_NUM_BANKS_MASK 0x0000000cL
 #define GRPH_CONTROL__GRPH_NUM_BANKS__SHIFT 0x00000002
 #define GRPH_CONTROL__GRPH_PIPE_CONFIG_MASK 0x1f000000L
@@ -9835,4 +9845,98 @@
 #define XDMA_TEST_DEBUG_INDEX__XDMA_TEST_DEBUG_WRITE_EN_MASK 0x00000100L
 #define XDMA_TEST_DEBUG_INDEX__XDMA_TEST_DEBUG_WRITE_EN__SHIFT 0x00000008
 
+// DATA_FORMAT
+#define DATA_FORMAT__INTERLEAVE_EN_MASK                    0x00000001L
+#define	DATA_FORMAT__INTERLEAVE_EN__SHIFT                  0x00000000
+#define DATA_FORMAT__RESET_REQ_AT_EOL_MASK                 0x00000010L
+#define	DATA_FORMAT__RESET_REQ_AT_EOL__SHIFT               0x00000004
+#define DATA_FORMAT__PREFETCH_MASK                         0x00001000L
+#define	DATA_FORMAT__PREFETCH__SHIFT                       0x0000000c
+#define DATA_FORMAT__SOF_READ_PT_MASK                      0x001f0000L
+#define	DATA_FORMAT__SOF_READ_PT__SHIFT                    0x00000010
+#define DATA_FORMAT__REQUEST_MODE_MASK                     0x03000000L
+#define	DATA_FORMAT__REQUEST_MODE__SHIFT                   0x00000018
+#define DATA_FORMAT__ALLOW_REQ_MODE_1_2_MASK               0x10000000L
+#define	DATA_FORMAT__ALLOW_REQ_MODE_1_2__SHIFT             0x0000001c
+
+
+// DC_LB_MEMORY_SPLIT
+#define DC_LB_MEMORY_SPLIT__LB_NUM_PARTITIONS_MASK         0x000f0000L
+#define	DC_LB_MEMORY_SPLIT__LB_NUM_PARTITIONS__SHIFT       0x00000010
+#define DC_LB_MEMORY_SPLIT__DC_LB_MEMORY_CONFIG_MASK       0x00300000L
+#define	DC_LB_MEMORY_SPLIT__DC_LB_MEMORY_CONFIG__SHIFT     0x00000014
+
+// DC_LB_MEM_SIZE
+#define DC_LB_MEM_SIZE__DC_LB_MEM_SIZE_MASK                0x000007ffL
+#define	DC_LB_MEM_SIZE__DC_LB_MEM_SIZE__SHIFT              0x00000000
+
+// SCL_TAP_CONTROL
+#define SCL_TAP_CONTROL__SCL_V_NUM_OF_TAPS_MASK            0x00000007L
+#define	SCL_TAP_CONTROL__SCL_V_NUM_OF_TAPS__SHIFT          0x00000000
+#define SCL_TAP_CONTROL__SCL_H_NUM_OF_TAPS_MASK            0x00000f00L
+#define	SCL_TAP_CONTROL__SCL_H_NUM_OF_TAPS__SHIFT          0x00000008
+
+// INT_MASK
+#define INT_MASK__VBLANK_INT_MASK                          0x00000001L
+#define	INT_MASK__VBLANK_INT__SHIFT                        0x00000000
+#define INT_MASK__VLINE_INT_MASK                           0x00000010L
+#define	INT_MASK__VLINE_INT__SHIFT                         0x00000004
+
+// PRIORITY_A_CNT
+#define PRIORITY_A_CNT__PRIORITY_MARK_A_MASK               0x00007fffL
+#define	PRIORITY_A_CNT__PRIORITY_MARK_A__SHIFT             0x00000000
+#define PRIORITY_A_CNT__PRIORITY_A_OFF_MASK                0x00010000L
+#define	PRIORITY_A_CNT__PRIORITY_A_OFF__SHIFT              0x00000010
+#define PRIORITY_A_CNT__PRIORITY_A_ALWAYS_ON_MASK          0x00100000L
+#define	PRIORITY_A_CNT__PRIORITY_A_ALWAYS_ON__SHIFT        0x00000014
+#define PRIORITY_A_CNT__PRIORITY_A_FORCE_MASK_MASK         0x01000000L
+#define	PRIORITY_A_CNT__PRIORITY_A_FORCE_MASK__SHIFT       0x00000018
+
+// PRIORITY_B_CNT
+#define PRIORITY_B_CNT__PRIORITY_MARK_B_MASK               0x00007fffL
+#define	PRIORITY_B_CNT__PRIORITY_MARK_B__SHIFT             0x00000000
+#define PRIORITY_B_CNT__PRIORITY_B_OFF_MASK                0x00010000L
+#define	PRIORITY_B_CNT__PRIORITY_B_OFF__SHIFT              0x00000010
+#define PRIORITY_B_CNT__PRIORITY_B_ALWAYS_ON_MASK          0x00100000L
+#define	PRIORITY_B_CNT__PRIORITY_B_ALWAYS_ON__SHIFT        0x00000014
+#define PRIORITY_B_CNT__PRIORITY_B_FORCE_MASK_MASK         0x01000000L
+#define	PRIORITY_B_CNT__PRIORITY_B_FORCE_MASK__SHIFT       0x00000018
+
+// VLINE_STATUS
+#define VLINE_STATUS__VLINE_OCCURRED_MASK                  0x00000001L
+#define	VLINE_STATUS__VLINE_OCCURRED__SHIFT                0x00000000
+#define VLINE_STATUS__VLINE_ACK_MASK                       0x00000010L
+#define	VLINE_STATUS__VLINE_ACK__SHIFT                     0x00000004
+#define VLINE_STATUS__VLINE_STAT_MASK                      0x00001000L
+#define	VLINE_STATUS__VLINE_STAT__SHIFT                    0x0000000c
+#define VLINE_STATUS__VLINE_INTERRUPT_MASK                 0x00010000L
+#define	VLINE_STATUS__VLINE_INTERRUPT__SHIFT               0x00000010
+#define VLINE_STATUS__VLINE_INTERRUPT_TYPE_MASK            0x00020000L
+#define	VLINE_STATUS__VLINE_INTERRUPT_TYPE__SHIFT          0x00000011
+
+// VBLANK_STATUS
+#define VBLANK_STATUS__VBLANK_OCCURRED_MASK                0x00000001L
+#define	VBLANK_STATUS__VBLANK_OCCURRED__SHIFT              0x00000000
+#define VBLANK_STATUS__VBLANK_ACK_MASK                     0x00000010L
+#define	VBLANK_STATUS__VBLANK_ACK__SHIFT                   0x00000004
+#define VBLANK_STATUS__VBLANK_STAT_MASK                    0x00001000L
+#define	VBLANK_STATUS__VBLANK_STAT__SHIFT                  0x0000000c
+#define VBLANK_STATUS__VBLANK_INTERRUPT_MASK               0x00010000L
+#define	VBLANK_STATUS__VBLANK_INTERRUPT__SHIFT             0x00000010
+#define VBLANK_STATUS__VBLANK_INTERRUPT_TYPE_MASK          0x00020000L
+#define	VBLANK_STATUS__VBLANK_INTERRUPT_TYPE__SHIFT        0x00000011
+
+// SCL_HORZ_FILTER_INIT_RGB_LUMA
+#define SCL_HORZ_FILTER_INIT_RGB_LUMA__SCL_H_INIT_FRAC_RGB_Y_MASK 0x0000ffffL
+#define	SCL_HORZ_FILTER_INIT_RGB_LUMA__SCL_H_INIT_FRAC_RGB_Y__SHIFT 0x00000000
+#define SCL_HORZ_FILTER_INIT_RGB_LUMA__SCL_H_INIT_INT_RGB_Y_MASK 0x000f0000L
+#define	SCL_HORZ_FILTER_INIT_RGB_LUMA__SCL_H_INIT_INT_RGB_Y__SHIFT 0x00000010
+
+// SCL_HORZ_FILTER_INIT_CHROMA
+#define SCL_HORZ_FILTER_INIT_CHROMA__SCL_H_INIT_FRAC_CBCR_MASK 0x0000ffffL
+#define	SCL_HORZ_FILTER_INIT_CHROMA__SCL_H_INIT_FRAC_CBCR__SHIFT 0x00000000
+#define SCL_HORZ_FILTER_INIT_CHROMA__SCL_H_INIT_INT_CBCR_MASK 0x00070000L
+#define	SCL_HORZ_FILTER_INIT_CHROMA__SCL_H_INIT_INT_CBCR__SHIFT 0x00000010
+
+
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h
index cf166b591bc5..cf166b591bc5 100755..100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h
index 0e0319e98c07..ea683f452bb3 100755..100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h
@@ -50271,6 +50271,10 @@
 #define DSC_TOP0_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP0_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP0_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP0_DSC_DEBUG_CONTROL
+#define DSC_TOP0_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP0_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
+
 
 // addressBlock: dce_dc_dsc0_dispdec_dsccif_dispdec
 //DSCCIF0_DSCCIF_CONFIG0
@@ -50789,6 +50793,9 @@
 #define DSC_TOP1_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP1_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP1_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP1_DSC_DEBUG_CONTROL
+#define DSC_TOP1_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP1_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
 
 
 // addressBlock: dce_dc_dsc1_dispdec_dsccif_dispdec
@@ -51308,6 +51315,10 @@
 #define DSC_TOP2_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP2_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP2_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP2_DSC_DEBUG_CONTROL
+#define DSC_TOP2_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP2_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
+
 
 // addressBlock: dce_dc_dsc2_dispdec_dsccif_dispdec
 //DSCCIF2_DSCCIF_CONFIG0
@@ -51826,6 +51837,9 @@
 #define DSC_TOP3_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP3_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP3_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP3_DSC_DEBUG_CONTROL
+#define DSC_TOP3_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP3_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
 
 
 // addressBlock: dce_dc_dsc3_dispdec_dsccif_dispdec
@@ -52346,6 +52360,10 @@
 #define DSC_TOP4_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP4_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP4_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP4_DSC_DEBUG_CONTROL
+#define DSC_TOP4_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP4_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
+
 
 // addressBlock: dce_dc_dsc4_dispdec_dsccif_dispdec
 //DSCCIF4_DSCCIF_CONFIG0
@@ -52864,6 +52882,10 @@
 #define DSC_TOP5_DSC_TOP_CONTROL__DSC_CLOCK_EN_MASK                                                           0x00000001L
 #define DSC_TOP5_DSC_TOP_CONTROL__DSC_DISPCLK_R_GATE_DIS_MASK                                                 0x00000010L
 #define DSC_TOP5_DSC_TOP_CONTROL__DSC_DSCCLK_R_GATE_DIS_MASK                                                  0x00000100L
+//DSC_TOP5_DSC_DEBUG_CONTROL
+#define DSC_TOP5_DSC_DEBUG_CONTROL__DSC_DBG_EN__SHIFT                                                         0x0
+#define DSC_TOP5_DSC_DEBUG_CONTROL__DSC_DBG_EN_MASK                                                           0x00000001L
+
 
 // addressBlock: dce_dc_dsc5_dispdec_dsccif_dispdec
 //DSCCIF5_DSCCIF_CONFIG0
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h
index 67faaf68e9d7..67faaf68e9d7 100755..100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h
index b4ef50a72868..b4ef50a72868 100755..100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
index 644a9fa71bb2..66a4151fa676 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
@@ -9184,6 +9184,8 @@
 #define mmRLC_GPM_THREAD_ENABLE_BASE_IDX                                                               1
 #define mmRLC_RLCG_DOORBELL_RANGE                                                                      0x4c47
 #define mmRLC_RLCG_DOORBELL_RANGE_BASE_IDX                                                             1
+#define mmRLC_CGTT_MGCG_OVERRIDE                                                                       0x4c48
+#define mmRLC_CGTT_MGCG_OVERRIDE_BASE_IDX                                                              1
 #define mmRLC_CGCG_CGLS_CTRL                                                                           0x4c49
 #define mmRLC_CGCG_CGLS_CTRL_BASE_IDX                                                                  1
 #define mmRLC_CGCG_RAMP_CTRL                                                                           0x4c4a
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
index 2e449fcff893..aed799d9a0e8 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
@@ -32365,6 +32365,31 @@
 #define RLC_RLCG_DOORBELL_CNTL__DOORBELL_3_MODE_MASK                                                          0x000000C0L
 #define RLC_RLCG_DOORBELL_CNTL__DOORBELL_ID_MASK                                                              0x001F0000L
 #define RLC_RLCG_DOORBELL_CNTL__DOORBELL_ID_EN_MASK                                                           0x00200000L
+//RLC_CGTT_MGCG_OVERRIDE
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_0__SHIFT                                                             0x0
+#define RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE__SHIFT                                                 0x1
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE__SHIFT                                                    0x2
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE__SHIFT                                                    0x3
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE__SHIFT                                                    0x4
+#define RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE__SHIFT                                                0x5
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE__SHIFT                                                    0x6
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_GFX3D_CG_OVERRIDE__SHIFT                                                0x7
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE__SHIFT                                                    0x8
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_15_9__SHIFT                                                          0x9
+#define RLC_CGTT_MGCG_OVERRIDE__ENABLE_CGTS_LEGACY__SHIFT                                                     0x10
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_31_17__SHIFT                                                         0x11
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_0_MASK                                                               0x00000001L
+#define RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK                                                   0x00000002L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK                                                      0x00000004L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK                                                      0x00000008L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK                                                      0x00000010L
+#define RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK                                                  0x00000020L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK                                                      0x00000040L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_GFX3D_CG_OVERRIDE_MASK                                                  0x00000080L
+#define RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK                                                      0x00000100L
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_15_9_MASK                                                            0x0000FE00L
+#define RLC_CGTT_MGCG_OVERRIDE__ENABLE_CGTS_LEGACY_MASK                                                       0x00010000L
+#define RLC_CGTT_MGCG_OVERRIDE__RESERVED_31_17_MASK                                                           0xFFFE0000L
 //RLC_RLCG_DOORBELL_STAT
 #define RLC_RLCG_DOORBELL_STAT__DOORBELL_0_VALID__SHIFT                                                       0x0
 #define RLC_RLCG_DOORBELL_STAT__DOORBELL_1_VALID__SHIFT                                                       0x1
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_1_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_1_offset.h
index f41556abfbbc..629a8a3b55e9 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_1_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_1_offset.h
@@ -205,6 +205,8 @@
 #define mmGCEA_EDC_CNT2_BASE_IDX                                                                       0
 #define mmGCEA_EDC_CNT3                                                                                0x071b
 #define mmGCEA_EDC_CNT3_BASE_IDX                                                                       0
+#define mmGCEA_ERR_STATUS                                                                              0x0712
+#define mmGCEA_ERR_STATUS_BASE_IDX                                                                     0
 
 // addressBlock: gc_gfxudec
 // base address: 0x30000
@@ -261,4 +263,4 @@
 #define mmRLC_EDC_CNT2                                                                                 0x4d41
 #define mmRLC_EDC_CNT2_BASE_IDX                                                                        1
 
-#endif
\ No newline at end of file
+#endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_offset.h
new file mode 100644
index 000000000000..3685766c4d56
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_offset.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright (C) 2020  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"),
+ * 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) 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.
+ */
+#ifndef _umc_8_7_0_OFFSET_HEADER
+#define _umc_8_7_0_OFFSET_HEADER
+
+#define mmUMCCH0_0_GeccErrCntSel                                  0x0328
+#define mmUMCCH0_0_GeccErrCntSel_BASE_IDX                         0
+#define mmUMCCH0_0_GeccErrCnt                                     0x0329
+#define mmUMCCH0_0_GeccErrCnt_BASE_IDX                            0
+#define mmMCA_UMC_UMC0_MCUMC_STATUST0                             0x03c2
+#define mmMCA_UMC_UMC0_MCUMC_STATUST0_BASE_IDX                    0
+#define mmMCA_UMC_UMC0_MCUMC_ADDRT0                               0x03c4
+#define mmMCA_UMC_UMC0_MCUMC_ADDRT0_BASE_IDX      		  0
+
+#endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_sh_mask.h
new file mode 100644
index 000000000000..4c5097fa0c09
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/umc/umc_8_7_0_sh_mask.h
@@ -0,0 +1,79 @@
+#ifndef _umc_8_7_0_SH_MASK_HEADER

+#define _umc_8_7_0_SH_MASK_HEADER

+

+//UMCCH0_0_GeccErrCntSel

+#define UMCCH0_0_GeccErrCntSel__GeccErrCntCsSel__SHIFT                                                        0x0

+#define UMCCH0_0_GeccErrCntSel__GeccErrInt__SHIFT                                                             0xc

+#define UMCCH0_0_GeccErrCntSel__GeccErrCntEn__SHIFT                                                           0xf

+#define UMCCH0_0_GeccErrCntSel__PoisonCntEn__SHIFT                                                            0x10

+#define UMCCH0_0_GeccErrCntSel__GeccErrCntCsSel_MASK                                                          0x0000000FL

+#define UMCCH0_0_GeccErrCntSel__GeccErrInt_MASK                                                               0x00003000L

+#define UMCCH0_0_GeccErrCntSel__GeccErrCntEn_MASK                                                             0x00008000L

+#define UMCCH0_0_GeccErrCntSel__PoisonCntEn_MASK                                                              0x00030000L

+//UMCCH0_0_GeccErrCnt

+#define UMCCH0_0_GeccErrCnt__GeccErrCnt__SHIFT                                                                0x0

+#define UMCCH0_0_GeccErrCnt__GeccUnCorrErrCnt__SHIFT                                                          0x10

+#define UMCCH0_0_GeccErrCnt__GeccErrCnt_MASK                                                                  0x0000FFFFL

+#define UMCCH0_0_GeccErrCnt__GeccUnCorrErrCnt_MASK                                                            0xFFFF0000L

+//MCA_UMC_UMC0_MCUMC_STATUST0

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrorCode__SHIFT                                                         0x0

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrorCodeExt__SHIFT                                                      0x10

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV22__SHIFT                                                          0x16

+#define MCA_UMC_UMC0_MCUMC_STATUST0__AddrLsb__SHIFT                                                           0x18

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV30__SHIFT                                                          0x1e

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrCoreId__SHIFT                                                         0x20

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV38__SHIFT                                                          0x26

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Scrub__SHIFT                                                             0x28

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV41__SHIFT                                                          0x29

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Poison__SHIFT                                                            0x2b

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Deferred__SHIFT                                                          0x2c

+#define MCA_UMC_UMC0_MCUMC_STATUST0__UECC__SHIFT                                                              0x2d

+#define MCA_UMC_UMC0_MCUMC_STATUST0__CECC__SHIFT                                                              0x2e

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV47__SHIFT                                                          0x2f

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Transparent__SHIFT                                                       0x34

+#define MCA_UMC_UMC0_MCUMC_STATUST0__SyndV__SHIFT                                                             0x35

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV54__SHIFT                                                          0x36

+#define MCA_UMC_UMC0_MCUMC_STATUST0__TCC__SHIFT                                                               0x37

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrCoreIdVal__SHIFT                                                      0x38

+#define MCA_UMC_UMC0_MCUMC_STATUST0__PCC__SHIFT                                                               0x39

+#define MCA_UMC_UMC0_MCUMC_STATUST0__AddrV__SHIFT                                                             0x3a

+#define MCA_UMC_UMC0_MCUMC_STATUST0__MiscV__SHIFT                                                             0x3b

+#define MCA_UMC_UMC0_MCUMC_STATUST0__En__SHIFT                                                                0x3c

+#define MCA_UMC_UMC0_MCUMC_STATUST0__UC__SHIFT                                                                0x3d

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Overflow__SHIFT                                                          0x3e

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Val__SHIFT                                                               0x3f

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrorCode_MASK                                                           0x000000000000FFFFL

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrorCodeExt_MASK                                                        0x00000000003F0000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV22_MASK                                                            0x0000000000C00000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__AddrLsb_MASK                                                             0x000000003F000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV30_MASK                                                            0x00000000C0000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrCoreId_MASK                                                           0x0000003F00000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV38_MASK                                                            0x000000C000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Scrub_MASK                                                               0x0000010000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV41_MASK                                                            0x0000060000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Poison_MASK                                                              0x0000080000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Deferred_MASK                                                            0x0000100000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__UECC_MASK                                                                0x0000200000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__CECC_MASK                                                                0x0000400000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV47_MASK                                                            0x000F800000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Transparent_MASK                                                         0x0010000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__SyndV_MASK                                                               0x0020000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__RESERV54_MASK                                                            0x0040000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__TCC_MASK                                                                 0x0080000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__ErrCoreIdVal_MASK                                                        0x0100000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__PCC_MASK                                                                 0x0200000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__AddrV_MASK                                                               0x0400000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__MiscV_MASK                                                               0x0800000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__En_MASK                                                                  0x1000000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__UC_MASK                                                                  0x2000000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Overflow_MASK                                                            0x4000000000000000L

+#define MCA_UMC_UMC0_MCUMC_STATUST0__Val_MASK                                                                 0x8000000000000000L

+//MCA_UMC_UMC0_MCUMC_ADDRT0

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__ErrorAddr__SHIFT                                                           0x0

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__LSB__SHIFT                                                                 0x38

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__Reserved__SHIFT                                                            0x3e

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__ErrorAddr_MASK                                                             0x00FFFFFFFFFFFFFFL

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__LSB_MASK                                                                   0x3F00000000000000L

+#define MCA_UMC_UMC0_MCUMC_ADDRT0__Reserved_MASK                                                              0xC000000000000000L

+

+#endif

diff --git a/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_offset.h
index 07aceffb108a..524ba4421c17 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_offset.h
@@ -151,6 +151,8 @@
 #define mmUVD_LMI_CTRL2_BASE_IDX                                                                       1
 #define mmUVD_MASTINT_EN                                                                               0x0540
 #define mmUVD_MASTINT_EN_BASE_IDX                                                                      1
+#define mmUVD_FW_STATUS                                                                                0x0557
+#define mmUVD_FW_STATUS_BASE_IDX                                                                       1
 #define mmJPEG_CGC_CTRL                                                                                0x0565
 #define mmJPEG_CGC_CTRL_BASE_IDX                                                                       1
 #define mmUVD_LMI_CTRL                                                                                 0x0566
@@ -219,4 +221,5 @@
 #define mmUVD_CONTEXT_ID2_BASE_IDX                                                                     1
 
 
+
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_sh_mask.h
index b427f73bd536..919be1842bd5 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_7_0_sh_mask.h
@@ -807,5 +807,25 @@
 #define UVD_CONTEXT_ID2__CONTEXT_ID2__SHIFT                                                                   0x0
 #define UVD_CONTEXT_ID2__CONTEXT_ID2_MASK                                                                     0xFFFFFFFFL
 
+//UVD_FW_STATUS
+#define UVD_FW_STATUS__BUSY__SHIFT                                                                            0x0
+#define UVD_FW_STATUS__ACTIVE__SHIFT                                                                          0x1
+#define UVD_FW_STATUS__SEND_EFUSE_REQ__SHIFT                                                                  0x2
+#define UVD_FW_STATUS__DONE__SHIFT                                                                            0x8
+#define UVD_FW_STATUS__PASS__SHIFT                                                                            0x10
+#define UVD_FW_STATUS__FAIL__SHIFT                                                                            0x11
+#define UVD_FW_STATUS__INVALID_LEN__SHIFT                                                                     0x12
+#define UVD_FW_STATUS__INVALID_0_PADDING__SHIFT                                                               0x13
+#define UVD_FW_STATUS__INVALID_NONCE__SHIFT                                                                   0x14
+#define UVD_FW_STATUS__BUSY_MASK                                                                              0x00000001L
+#define UVD_FW_STATUS__ACTIVE_MASK                                                                            0x00000002L
+#define UVD_FW_STATUS__SEND_EFUSE_REQ_MASK                                                                    0x00000004L
+#define UVD_FW_STATUS__DONE_MASK                                                                              0x00000100L
+#define UVD_FW_STATUS__PASS_MASK                                                                              0x00010000L
+#define UVD_FW_STATUS__FAIL_MASK                                                                              0x00020000L
+#define UVD_FW_STATUS__INVALID_LEN_MASK                                                                       0x00040000L
+#define UVD_FW_STATUS__INVALID_0_PADDING_MASK                                                                 0x00080000L
+#define UVD_FW_STATUS__INVALID_NONCE_MASK                                                                     0x00100000L
+
 
 #endif
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 301de493377a..95c656d205ed 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -212,10 +212,15 @@ struct tile_config {
  * IH ring entry. This function allows the KFD ISR to get the VMID
  * from the fault status register as early as possible.
  *
- * @get_hive_id: Returns hive id of current  device,  0 if xgmi is not enabled
+ * @get_cu_occupancy: Function pointer that returns to caller the number
+ * of wave fronts that are in flight for all of the queues of a process
+ * as identified by its pasid. It is important to note that the value
+ * returned by this function is a snapshot of current moment and cannot
+ * guarantee any minimum for the number of waves in-flight. This function
+ * is defined for devices that belong to GFX9 and later GFX families. Care
+ * must be taken in calling this function as it is not defined for devices
+ * that belong to GFX8 and below GFX families.
  *
- * @get_unique_id: Returns uuid id of current  device
- * 
  * This structure contains function pointers to services that the kgd driver
  * provides to amdkfd driver.
  *
@@ -290,9 +295,9 @@ struct kfd2kgd_calls {
 	void (*set_vm_context_page_table_base)(struct kgd_dev *kgd,
 			uint32_t vmid, uint64_t page_table_base);
 	uint32_t (*read_vmid_from_vmfault_reg)(struct kgd_dev *kgd);
-	uint64_t (*get_hive_id)(struct kgd_dev *kgd);
-	uint64_t (*get_unique_id)(struct kgd_dev *kgd);
 
+	void (*get_cu_occupancy)(struct kgd_dev *kgd, int pasid, int *wave_cnt,
+			int *max_waves_per_cu);
 };
 
 #endif	/* KGD_KFD_INTERFACE_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
index a7f92d0b3a90..94132c70d7af 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -281,6 +281,7 @@ struct amd_pm_funcs {
 	int (*get_power_limit)(void *handle, uint32_t *limit, bool default_limit);
 	int (*get_power_profile_mode)(void *handle, char *buf);
 	int (*set_power_profile_mode)(void *handle, long *input, uint32_t size);
+	int (*set_fine_grain_clk_vol)(void *handle, uint32_t type, long *input, uint32_t size);
 	int (*odn_edit_dpm_table)(void *handle, uint32_t type, long *input, uint32_t size);
 	int (*set_mp1_state)(void *handle, enum pp_mp1_state mp1_state);
 	int (*smu_i2c_bus_access)(void *handle, bool acquire);
@@ -322,6 +323,115 @@ struct amd_pm_funcs {
 	int (*asic_reset_mode_2)(void *handle);
 	int (*set_df_cstate)(void *handle, enum pp_df_cstate state);
 	int (*set_xgmi_pstate)(void *handle, uint32_t pstate);
+	ssize_t (*get_gpu_metrics)(void *handle, void **table);
+};
+
+struct metrics_table_header {
+	uint16_t			structure_size;
+	uint8_t				format_revision;
+	uint8_t				content_revision;
+};
+
+struct gpu_metrics_v1_0 {
+	struct metrics_table_header	common_header;
+
+	/* Driver attached timestamp (in ns) */
+	uint64_t			system_clock_counter;
+
+	/* Temperature */
+	uint16_t			temperature_edge;
+	uint16_t			temperature_hotspot;
+	uint16_t			temperature_mem;
+	uint16_t			temperature_vrgfx;
+	uint16_t			temperature_vrsoc;
+	uint16_t			temperature_vrmem;
+
+	/* Utilization */
+	uint16_t			average_gfx_activity;
+	uint16_t			average_umc_activity; // memory controller
+	uint16_t			average_mm_activity; // UVD or VCN
+
+	/* Power/Energy */
+	uint16_t			average_socket_power;
+	uint32_t			energy_accumulator;
+
+	/* Average clocks */
+	uint16_t			average_gfxclk_frequency;
+	uint16_t			average_socclk_frequency;
+	uint16_t			average_uclk_frequency;
+	uint16_t			average_vclk0_frequency;
+	uint16_t			average_dclk0_frequency;
+	uint16_t			average_vclk1_frequency;
+	uint16_t			average_dclk1_frequency;
+
+	/* Current clocks */
+	uint16_t			current_gfxclk;
+	uint16_t			current_socclk;
+	uint16_t			current_uclk;
+	uint16_t			current_vclk0;
+	uint16_t			current_dclk0;
+	uint16_t			current_vclk1;
+	uint16_t			current_dclk1;
+
+	/* Throttle status */
+	uint32_t			throttle_status;
+
+	/* Fans */
+	uint16_t			current_fan_speed;
+
+	/* Link width/speed */
+	uint8_t				pcie_link_width;
+	uint8_t				pcie_link_speed; // in 0.1 GT/s
+};
+
+struct gpu_metrics_v2_0 {
+	struct metrics_table_header	common_header;
+
+	/* Driver attached timestamp (in ns) */
+	uint64_t			system_clock_counter;
+
+	/* Temperature */
+	uint16_t			temperature_gfx; // gfx temperature on APUs
+	uint16_t			temperature_soc; // soc temperature on APUs
+	uint16_t			temperature_core[8]; // CPU core temperature on APUs
+	uint16_t			temperature_l3[2];
+
+	/* Utilization */
+	uint16_t			average_gfx_activity;
+	uint16_t			average_mm_activity; // UVD or VCN
+
+	/* Power/Energy */
+	uint16_t			average_socket_power; // dGPU + APU power on A + A platform
+	uint16_t			average_cpu_power;
+	uint16_t			average_soc_power;
+	uint16_t			average_gfx_power;
+	uint16_t			average_core_power[8]; // CPU core power on APUs
+
+	/* Average clocks */
+	uint16_t			average_gfxclk_frequency;
+	uint16_t			average_socclk_frequency;
+	uint16_t			average_uclk_frequency;
+	uint16_t			average_fclk_frequency;
+	uint16_t			average_vclk_frequency;
+	uint16_t			average_dclk_frequency;
+
+	/* Current clocks */
+	uint16_t			current_gfxclk;
+	uint16_t			current_socclk;
+	uint16_t			current_uclk;
+	uint16_t			current_fclk;
+	uint16_t			current_vclk;
+	uint16_t			current_dclk;
+	uint16_t			current_coreclk[8]; // CPU core clocks
+	uint16_t			current_l3clk[2];
+
+	/* Throttle status */
+	uint32_t			throttle_status;
+
+	/* Fans */
+	uint16_t			fan_pwm;
+
+	uint16_t			padding;
 };
 
 #endif
diff --git a/drivers/gpu/drm/amd/pm/Makefile b/drivers/gpu/drm/amd/pm/Makefile
new file mode 100644
index 000000000000..f01e86030cd1
--- /dev/null
+++ b/drivers/gpu/drm/amd/pm/Makefile
@@ -0,0 +1,46 @@
+#
+# Copyright 2017 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"),
+# 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.
+#
+
+subdir-ccflags-y += \
+		-I$(FULL_AMD_PATH)/pm/inc/  \
+		-I$(FULL_AMD_PATH)/include/asic_reg  \
+		-I$(FULL_AMD_PATH)/include  \
+		-I$(FULL_AMD_PATH)/pm/swsmu \
+		-I$(FULL_AMD_PATH)/pm/swsmu/smu11 \
+		-I$(FULL_AMD_PATH)/pm/swsmu/smu12 \
+		-I$(FULL_AMD_PATH)/pm/powerplay \
+		-I$(FULL_AMD_PATH)/pm/powerplay/smumgr\
+		-I$(FULL_AMD_PATH)/pm/powerplay/hwmgr
+
+AMD_PM_PATH = ../pm
+
+PM_LIBS = swsmu powerplay
+
+AMD_PM = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/pm/,$(PM_LIBS)))
+
+include $(AMD_PM)
+
+PM_MGR = amdgpu_dpm.o amdgpu_pm.o
+
+AMD_PM_POWER = $(addprefix $(AMD_PM_PATH)/,$(PM_MGR))
+
+AMD_POWERPLAY_FILES += $(AMD_PM_POWER)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 2082c0acd216..17a45baff638 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -28,6 +28,11 @@
 #include "amdgpu_dpm.h"
 #include "atom.h"
 #include "amd_pcie.h"
+#include "amdgpu_display.h"
+#include "hwmgr.h"
+#include <linux/power_supply.h>
+
+#define WIDTH_4K 3840
 
 void amdgpu_dpm_print_class_info(u32 class, u32 class2)
 {
@@ -117,7 +122,7 @@ void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev,
 
 void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev)
 {
-	struct drm_device *ddev = adev->ddev;
+	struct drm_device *ddev = adev_to_drm(adev);
 	struct drm_crtc *crtc;
 	struct amdgpu_crtc *amdgpu_crtc;
 
@@ -138,7 +143,7 @@ void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev)
 
 u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_crtc *crtc;
 	struct amdgpu_crtc *amdgpu_crtc;
 	u32 vblank_in_pixels;
@@ -165,7 +170,7 @@ u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev)
 
 u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev)
 {
-	struct drm_device *dev = adev->ddev;
+	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_crtc *crtc;
 	struct amdgpu_crtc *amdgpu_crtc;
 	u32 vrefresh = 0;
@@ -1110,8 +1115,6 @@ int amdgpu_dpm_baco_reset(struct amdgpu_device *adev)
 	struct smu_context *smu = &adev->smu;
 	int ret = 0;
 
-	dev_info(adev->dev, "GPU BACO reset\n");
-
 	if (is_support_sw_smu(adev)) {
 		ret = smu_baco_enter(smu);
 		if (ret)
@@ -1216,3 +1219,469 @@ int amdgpu_dpm_allow_xgmi_power_down(struct amdgpu_device *adev, bool en)
 
 	return 0;
 }
+
+int amdgpu_dpm_enable_mgpu_fan_boost(struct amdgpu_device *adev)
+{
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs =
+			adev->powerplay.pp_funcs;
+	struct smu_context *smu = &adev->smu;
+	int ret = 0;
+
+	if (is_support_sw_smu(adev))
+		ret = smu_enable_mgpu_fan_boost(smu);
+	else if (pp_funcs && pp_funcs->enable_mgpu_fan_boost)
+		ret = pp_funcs->enable_mgpu_fan_boost(pp_handle);
+
+	return ret;
+}
+
+int amdgpu_dpm_set_clockgating_by_smu(struct amdgpu_device *adev,
+				      uint32_t msg_id)
+{
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs =
+			adev->powerplay.pp_funcs;
+	int ret = 0;
+
+	if (pp_funcs && pp_funcs->set_clockgating_by_smu)
+		ret = pp_funcs->set_clockgating_by_smu(pp_handle,
+						       msg_id);
+
+	return ret;
+}
+
+int amdgpu_dpm_smu_i2c_bus_access(struct amdgpu_device *adev,
+				  bool acquire)
+{
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs =
+			adev->powerplay.pp_funcs;
+	int ret = -EOPNOTSUPP;
+
+	if (pp_funcs && pp_funcs->smu_i2c_bus_access)
+		ret = pp_funcs->smu_i2c_bus_access(pp_handle,
+						   acquire);
+
+	return ret;
+}
+
+void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev)
+{
+	if (adev->pm.dpm_enabled) {
+		mutex_lock(&adev->pm.mutex);
+		if (power_supply_is_system_supplied() > 0)
+			adev->pm.ac_power = true;
+		else
+			adev->pm.ac_power = false;
+		if (adev->powerplay.pp_funcs &&
+		    adev->powerplay.pp_funcs->enable_bapm)
+			amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power);
+		mutex_unlock(&adev->pm.mutex);
+
+		if (is_support_sw_smu(adev))
+			smu_set_ac_dc(&adev->smu);
+	}
+}
+
+int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size)
+{
+	int ret = 0;
+
+	if (!data || !size)
+		return -EINVAL;
+
+	if (is_support_sw_smu(adev))
+		ret = smu_read_sensor(&adev->smu, sensor, data, size);
+	else {
+		if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->read_sensor)
+			ret = adev->powerplay.pp_funcs->read_sensor((adev)->powerplay.pp_handle,
+								    sensor, data, size);
+		else
+			ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+void amdgpu_dpm_thermal_work_handler(struct work_struct *work)
+{
+	struct amdgpu_device *adev =
+		container_of(work, struct amdgpu_device,
+			     pm.dpm.thermal.work);
+	/* switch to the thermal state */
+	enum amd_pm_state_type dpm_state = POWER_STATE_TYPE_INTERNAL_THERMAL;
+	int temp, size = sizeof(temp);
+
+	if (!adev->pm.dpm_enabled)
+		return;
+
+	if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP,
+				    (void *)&temp, &size)) {
+		if (temp < adev->pm.dpm.thermal.min_temp)
+			/* switch back the user state */
+			dpm_state = adev->pm.dpm.user_state;
+	} else {
+		if (adev->pm.dpm.thermal.high_to_low)
+			/* switch back the user state */
+			dpm_state = adev->pm.dpm.user_state;
+	}
+	mutex_lock(&adev->pm.mutex);
+	if (dpm_state == POWER_STATE_TYPE_INTERNAL_THERMAL)
+		adev->pm.dpm.thermal_active = true;
+	else
+		adev->pm.dpm.thermal_active = false;
+	adev->pm.dpm.state = dpm_state;
+	mutex_unlock(&adev->pm.mutex);
+
+	amdgpu_pm_compute_clocks(adev);
+}
+
+static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev,
+						     enum amd_pm_state_type dpm_state)
+{
+	int i;
+	struct amdgpu_ps *ps;
+	u32 ui_class;
+	bool single_display = (adev->pm.dpm.new_active_crtc_count < 2) ?
+		true : false;
+
+	/* check if the vblank period is too short to adjust the mclk */
+	if (single_display && adev->powerplay.pp_funcs->vblank_too_short) {
+		if (amdgpu_dpm_vblank_too_short(adev))
+			single_display = false;
+	}
+
+	/* certain older asics have a separare 3D performance state,
+	 * so try that first if the user selected performance
+	 */
+	if (dpm_state == POWER_STATE_TYPE_PERFORMANCE)
+		dpm_state = POWER_STATE_TYPE_INTERNAL_3DPERF;
+	/* balanced states don't exist at the moment */
+	if (dpm_state == POWER_STATE_TYPE_BALANCED)
+		dpm_state = POWER_STATE_TYPE_PERFORMANCE;
+
+restart_search:
+	/* Pick the best power state based on current conditions */
+	for (i = 0; i < adev->pm.dpm.num_ps; i++) {
+		ps = &adev->pm.dpm.ps[i];
+		ui_class = ps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK;
+		switch (dpm_state) {
+		/* user states */
+		case POWER_STATE_TYPE_BATTERY:
+			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY) {
+				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
+					if (single_display)
+						return ps;
+				} else
+					return ps;
+			}
+			break;
+		case POWER_STATE_TYPE_BALANCED:
+			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BALANCED) {
+				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
+					if (single_display)
+						return ps;
+				} else
+					return ps;
+			}
+			break;
+		case POWER_STATE_TYPE_PERFORMANCE:
+			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE) {
+				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
+					if (single_display)
+						return ps;
+				} else
+					return ps;
+			}
+			break;
+		/* internal states */
+		case POWER_STATE_TYPE_INTERNAL_UVD:
+			if (adev->pm.dpm.uvd_ps)
+				return adev->pm.dpm.uvd_ps;
+			else
+				break;
+		case POWER_STATE_TYPE_INTERNAL_UVD_SD:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_SDSTATE)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_UVD_HD:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_HDSTATE)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_HD2STATE)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
+			if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_MVC)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_BOOT:
+			return adev->pm.dpm.boot_ps;
+		case POWER_STATE_TYPE_INTERNAL_THERMAL:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_THERMAL)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_ACPI:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_ACPI)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_ULV:
+			if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV)
+				return ps;
+			break;
+		case POWER_STATE_TYPE_INTERNAL_3DPERF:
+			if (ps->class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE)
+				return ps;
+			break;
+		default:
+			break;
+		}
+	}
+	/* use a fallback state if we didn't match */
+	switch (dpm_state) {
+	case POWER_STATE_TYPE_INTERNAL_UVD_SD:
+		dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+		goto restart_search;
+	case POWER_STATE_TYPE_INTERNAL_UVD_HD:
+	case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
+	case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
+		if (adev->pm.dpm.uvd_ps) {
+			return adev->pm.dpm.uvd_ps;
+		} else {
+			dpm_state = POWER_STATE_TYPE_PERFORMANCE;
+			goto restart_search;
+		}
+	case POWER_STATE_TYPE_INTERNAL_THERMAL:
+		dpm_state = POWER_STATE_TYPE_INTERNAL_ACPI;
+		goto restart_search;
+	case POWER_STATE_TYPE_INTERNAL_ACPI:
+		dpm_state = POWER_STATE_TYPE_BATTERY;
+		goto restart_search;
+	case POWER_STATE_TYPE_BATTERY:
+	case POWER_STATE_TYPE_BALANCED:
+	case POWER_STATE_TYPE_INTERNAL_3DPERF:
+		dpm_state = POWER_STATE_TYPE_PERFORMANCE;
+		goto restart_search;
+	default:
+		break;
+	}
+
+	return NULL;
+}
+
+static void amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev)
+{
+	struct amdgpu_ps *ps;
+	enum amd_pm_state_type dpm_state;
+	int ret;
+	bool equal = false;
+
+	/* if dpm init failed */
+	if (!adev->pm.dpm_enabled)
+		return;
+
+	if (adev->pm.dpm.user_state != adev->pm.dpm.state) {
+		/* add other state override checks here */
+		if ((!adev->pm.dpm.thermal_active) &&
+		    (!adev->pm.dpm.uvd_active))
+			adev->pm.dpm.state = adev->pm.dpm.user_state;
+	}
+	dpm_state = adev->pm.dpm.state;
+
+	ps = amdgpu_dpm_pick_power_state(adev, dpm_state);
+	if (ps)
+		adev->pm.dpm.requested_ps = ps;
+	else
+		return;
+
+	if (amdgpu_dpm == 1 && adev->powerplay.pp_funcs->print_power_state) {
+		printk("switching from power state:\n");
+		amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps);
+		printk("switching to power state:\n");
+		amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps);
+	}
+
+	/* update whether vce is active */
+	ps->vce_active = adev->pm.dpm.vce_active;
+	if (adev->powerplay.pp_funcs->display_configuration_changed)
+		amdgpu_dpm_display_configuration_changed(adev);
+
+	ret = amdgpu_dpm_pre_set_power_state(adev);
+	if (ret)
+		return;
+
+	if (adev->powerplay.pp_funcs->check_state_equal) {
+		if (0 != amdgpu_dpm_check_state_equal(adev, adev->pm.dpm.current_ps, adev->pm.dpm.requested_ps, &equal))
+			equal = false;
+	}
+
+	if (equal)
+		return;
+
+	amdgpu_dpm_set_power_state(adev);
+	amdgpu_dpm_post_set_power_state(adev);
+
+	adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs;
+	adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count;
+
+	if (adev->powerplay.pp_funcs->force_performance_level) {
+		if (adev->pm.dpm.thermal_active) {
+			enum amd_dpm_forced_level level = adev->pm.dpm.forced_level;
+			/* force low perf level for thermal */
+			amdgpu_dpm_force_performance_level(adev, AMD_DPM_FORCED_LEVEL_LOW);
+			/* save the user's level */
+			adev->pm.dpm.forced_level = level;
+		} else {
+			/* otherwise, user selected level */
+			amdgpu_dpm_force_performance_level(adev, adev->pm.dpm.forced_level);
+		}
+	}
+}
+
+void amdgpu_pm_compute_clocks(struct amdgpu_device *adev)
+{
+	int i = 0;
+
+	if (!adev->pm.dpm_enabled)
+		return;
+
+	if (adev->mode_info.num_crtc)
+		amdgpu_display_bandwidth_update(adev);
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
+		struct amdgpu_ring *ring = adev->rings[i];
+		if (ring && ring->sched.ready)
+			amdgpu_fence_wait_empty(ring);
+	}
+
+	if (is_support_sw_smu(adev)) {
+		struct smu_dpm_context *smu_dpm = &adev->smu.smu_dpm;
+		smu_handle_task(&adev->smu,
+				smu_dpm->dpm_level,
+				AMD_PP_TASK_DISPLAY_CONFIG_CHANGE,
+				true);
+	} else {
+		if (adev->powerplay.pp_funcs->dispatch_tasks) {
+			if (!amdgpu_device_has_dc_support(adev)) {
+				mutex_lock(&adev->pm.mutex);
+				amdgpu_dpm_get_active_displays(adev);
+				adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count;
+				adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev);
+				adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev);
+				/* we have issues with mclk switching with refresh rates over 120 hz on the non-DC code. */
+				if (adev->pm.pm_display_cfg.vrefresh > 120)
+					adev->pm.pm_display_cfg.min_vblank_time = 0;
+				if (adev->powerplay.pp_funcs->display_configuration_change)
+					adev->powerplay.pp_funcs->display_configuration_change(
+									adev->powerplay.pp_handle,
+									&adev->pm.pm_display_cfg);
+				mutex_unlock(&adev->pm.mutex);
+			}
+			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, NULL);
+		} else {
+			mutex_lock(&adev->pm.mutex);
+			amdgpu_dpm_get_active_displays(adev);
+			amdgpu_dpm_change_power_state_locked(adev);
+			mutex_unlock(&adev->pm.mutex);
+		}
+	}
+}
+
+void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable)
+{
+	int ret = 0;
+
+	if (adev->family == AMDGPU_FAMILY_SI) {
+		mutex_lock(&adev->pm.mutex);
+		if (enable) {
+			adev->pm.dpm.uvd_active = true;
+			adev->pm.dpm.state = POWER_STATE_TYPE_INTERNAL_UVD;
+		} else {
+			adev->pm.dpm.uvd_active = false;
+		}
+		mutex_unlock(&adev->pm.mutex);
+
+		amdgpu_pm_compute_clocks(adev);
+	} else {
+		ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable);
+		if (ret)
+			DRM_ERROR("Dpm %s uvd failed, ret = %d. \n",
+				  enable ? "enable" : "disable", ret);
+
+		/* enable/disable Low Memory PState for UVD (4k videos) */
+		if (adev->asic_type == CHIP_STONEY &&
+			adev->uvd.decode_image_width >= WIDTH_4K) {
+			struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
+
+			if (hwmgr && hwmgr->hwmgr_func &&
+			    hwmgr->hwmgr_func->update_nbdpm_pstate)
+				hwmgr->hwmgr_func->update_nbdpm_pstate(hwmgr,
+								       !enable,
+								       true);
+		}
+	}
+}
+
+void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable)
+{
+	int ret = 0;
+
+	if (adev->family == AMDGPU_FAMILY_SI) {
+		mutex_lock(&adev->pm.mutex);
+		if (enable) {
+			adev->pm.dpm.vce_active = true;
+			/* XXX select vce level based on ring/task */
+			adev->pm.dpm.vce_level = AMD_VCE_LEVEL_AC_ALL;
+		} else {
+			adev->pm.dpm.vce_active = false;
+		}
+		mutex_unlock(&adev->pm.mutex);
+
+		amdgpu_pm_compute_clocks(adev);
+	} else {
+		ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VCE, !enable);
+		if (ret)
+			DRM_ERROR("Dpm %s vce failed, ret = %d. \n",
+				  enable ? "enable" : "disable", ret);
+	}
+}
+
+void amdgpu_pm_print_power_states(struct amdgpu_device *adev)
+{
+	int i;
+
+	if (adev->powerplay.pp_funcs->print_power_state == NULL)
+		return;
+
+	for (i = 0; i < adev->pm.dpm.num_ps; i++)
+		amdgpu_dpm_print_power_state(adev, &adev->pm.dpm.ps[i]);
+
+}
+
+void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable)
+{
+	int ret = 0;
+
+	ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_JPEG, !enable);
+	if (ret)
+		DRM_ERROR("Dpm %s jpeg failed, ret = %d. \n",
+			  enable ? "enable" : "disable", ret);
+}
+
+int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version)
+{
+	int r;
+
+	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->load_firmware) {
+		r = adev->powerplay.pp_funcs->load_firmware(adev->powerplay.pp_handle);
+		if (r) {
+			pr_err("smu firmware loading failed\n");
+			return r;
+		}
+		*smu_version = adev->pm.fw_version;
+	}
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index e4dbf14320b6..529816637c73 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -29,17 +29,14 @@
 #include "amdgpu_drv.h"
 #include "amdgpu_pm.h"
 #include "amdgpu_dpm.h"
-#include "amdgpu_display.h"
 #include "amdgpu_smu.h"
 #include "atom.h"
-#include <linux/power_supply.h>
 #include <linux/pci.h>
 #include <linux/hwmon.h>
 #include <linux/hwmon-sysfs.h>
 #include <linux/nospec.h>
 #include <linux/pm_runtime.h>
 #include "hwmgr.h"
-#define WIDTH_4K 3840
 
 static const struct cg_flag_name clocks[] = {
 	{AMD_CG_SUPPORT_GFX_MGCG, "Graphics Medium Grain Clock Gating"},
@@ -81,45 +78,6 @@ static const struct hwmon_temp_label {
 	{PP_TEMP_MEM, "mem"},
 };
 
-void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev)
-{
-	if (adev->pm.dpm_enabled) {
-		mutex_lock(&adev->pm.mutex);
-		if (power_supply_is_system_supplied() > 0)
-			adev->pm.ac_power = true;
-		else
-			adev->pm.ac_power = false;
-		if (adev->powerplay.pp_funcs &&
-		    adev->powerplay.pp_funcs->enable_bapm)
-			amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power);
-		mutex_unlock(&adev->pm.mutex);
-
-		if (is_support_sw_smu(adev))
-			smu_set_ac_dc(&adev->smu);
-	}
-}
-
-int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor,
-			   void *data, uint32_t *size)
-{
-	int ret = 0;
-
-	if (!data || !size)
-		return -EINVAL;
-
-	if (is_support_sw_smu(adev))
-		ret = smu_read_sensor(&adev->smu, sensor, data, size);
-	else {
-		if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->read_sensor)
-			ret = adev->powerplay.pp_funcs->read_sensor((adev)->powerplay.pp_handle,
-								    sensor, data, size);
-		else
-			ret = -EINVAL;
-	}
-
-	return ret;
-}
-
 /**
  * DOC: power_dpm_state
  *
@@ -159,11 +117,11 @@ static ssize_t amdgpu_get_power_dpm_state(struct device *dev,
 					  char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amd_pm_state_type pm;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -197,11 +155,11 @@ static ssize_t amdgpu_set_power_dpm_state(struct device *dev,
 					  size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amd_pm_state_type  state;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (strncmp("battery", buf, strlen("battery")) == 0)
@@ -303,11 +261,11 @@ static ssize_t amdgpu_get_power_dpm_force_performance_level(struct device *dev,
 							    char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amd_dpm_forced_level level = 0xff;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -344,12 +302,12 @@ static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev,
 							    size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amd_dpm_forced_level level;
 	enum amd_dpm_forced_level current_level = 0xff;
 	int ret = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (strncmp("low", buf, strlen("low")) == 0) {
@@ -449,11 +407,11 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	struct pp_states_info data;
 	int i, buf_len, ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -491,13 +449,13 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	struct pp_states_info data;
 	struct smu_context *smu = &adev->smu;
 	enum amd_pm_state_type pm = 0;
 	int i = 0, ret = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -536,9 +494,9 @@ static ssize_t amdgpu_get_pp_force_state(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (adev->pp_force_state_enabled)
@@ -553,12 +511,12 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amd_pm_state_type state = 0;
 	unsigned long idx;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (strlen(buf) == 1)
@@ -614,11 +572,11 @@ static ssize_t amdgpu_get_pp_table(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	char *table = NULL;
 	int size, ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -659,10 +617,10 @@ static ssize_t amdgpu_set_pp_table(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -694,6 +652,52 @@ static ssize_t amdgpu_set_pp_table(struct device *dev,
  * in each power level within a power state.  The pp_od_clk_voltage is used for
  * this.
  *
+ * Note that the actual memory controller clock rate are exposed, not
+ * the effective memory clock of the DRAMs. To translate it, use the
+ * following formula:
+ *
+ * Clock conversion (Mhz):
+ *
+ * HBM: effective_memory_clock = memory_controller_clock * 1
+ *
+ * G5: effective_memory_clock = memory_controller_clock * 1
+ *
+ * G6: effective_memory_clock = memory_controller_clock * 2
+ *
+ * DRAM data rate (MT/s):
+ *
+ * HBM: effective_memory_clock * 2 = data_rate
+ *
+ * G5: effective_memory_clock * 4 = data_rate
+ *
+ * G6: effective_memory_clock * 8 = data_rate
+ *
+ * Bandwidth (MB/s):
+ *
+ * data_rate * vram_bit_width / 8 = memory_bandwidth
+ *
+ * Some examples:
+ *
+ * G5 on RX460:
+ *
+ * memory_controller_clock = 1750 Mhz
+ *
+ * effective_memory_clock = 1750 Mhz * 1 = 1750 Mhz
+ *
+ * data rate = 1750 * 4 = 7000 MT/s
+ *
+ * memory_bandwidth = 7000 * 128 bits / 8 = 112000 MB/s
+ *
+ * G6 on RX5700:
+ *
+ * memory_controller_clock = 875 Mhz
+ *
+ * effective_memory_clock = 875 Mhz * 2 = 1750 Mhz
+ *
+ * data rate = 1750 * 8 = 14000 MT/s
+ *
+ * memory_bandwidth = 14000 * 256 bits / 8 = 448000 MB/s
+ *
  * < For Vega10 and previous ASICs >
  *
  * Reading the file will display:
@@ -759,7 +763,7 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t parameter_size = 0;
 	long parameter[64];
@@ -769,7 +773,7 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev,
 	const char delimiter[3] = {' ', '\n', '\0'};
 	uint32_t type;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (count > 127)
@@ -796,7 +800,8 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev,
 		tmp_str++;
 	while (isspace(*++tmp_str));
 
-	while ((sub_str = strsep(&tmp_str, delimiter)) != NULL) {
+	while (tmp_str[0]) {
+		sub_str = strsep(&tmp_str, delimiter);
 		ret = kstrtol(sub_str, 0, &parameter[parameter_size]);
 		if (ret)
 			return -EINVAL;
@@ -822,6 +827,18 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev,
 			return -EINVAL;
 		}
 	} else {
+
+		if (adev->powerplay.pp_funcs->set_fine_grain_clk_vol) {
+			ret = amdgpu_dpm_set_fine_grain_clk_vol(adev, type,
+								parameter,
+								parameter_size);
+			if (ret) {
+				pm_runtime_mark_last_busy(ddev->dev);
+				pm_runtime_put_autosuspend(ddev->dev);
+				return -EINVAL;
+			}
+		}
+
 		if (adev->powerplay.pp_funcs->odn_edit_dpm_table) {
 			ret = amdgpu_dpm_odn_edit_dpm_table(adev, type,
 						parameter, parameter_size);
@@ -858,11 +875,11 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -912,11 +929,11 @@ static ssize_t amdgpu_set_pp_features(struct device *dev,
 				      size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint64_t featuremask;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = kstrtou64(buf, 0, &featuremask);
@@ -957,11 +974,11 @@ static ssize_t amdgpu_get_pp_features(struct device *dev,
 				      char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1018,11 +1035,11 @@ static ssize_t amdgpu_get_pp_dpm_sclk(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1066,7 +1083,8 @@ static ssize_t amdgpu_read_mask(const char *buf, size_t count, uint32_t *mask)
 	memcpy(buf_cpy, buf, bytes);
 	buf_cpy[bytes] = '\0';
 	tmp = buf_cpy;
-	while ((sub_str = strsep(&tmp, delimiter)) != NULL) {
+	while (tmp[0]) {
+		sub_str = strsep(&tmp, delimiter);
 		if (strlen(sub_str)) {
 			ret = kstrtol(sub_str, 0, &level);
 			if (ret)
@@ -1085,11 +1103,11 @@ static ssize_t amdgpu_set_pp_dpm_sclk(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t mask = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1121,11 +1139,11 @@ static ssize_t amdgpu_get_pp_dpm_mclk(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1153,11 +1171,11 @@ static ssize_t amdgpu_set_pp_dpm_mclk(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t mask = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1189,11 +1207,11 @@ static ssize_t amdgpu_get_pp_dpm_socclk(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1221,11 +1239,11 @@ static ssize_t amdgpu_set_pp_dpm_socclk(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t mask = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1259,11 +1277,11 @@ static ssize_t amdgpu_get_pp_dpm_fclk(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1291,11 +1309,11 @@ static ssize_t amdgpu_set_pp_dpm_fclk(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t mask = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1329,11 +1347,11 @@ static ssize_t amdgpu_get_pp_dpm_dcefclk(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1361,11 +1379,11 @@ static ssize_t amdgpu_set_pp_dpm_dcefclk(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t mask = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1399,11 +1417,11 @@ static ssize_t amdgpu_get_pp_dpm_pcie(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1431,11 +1449,11 @@ static ssize_t amdgpu_set_pp_dpm_pcie(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	uint32_t mask = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = amdgpu_read_mask(buf, count, &mask);
@@ -1469,11 +1487,11 @@ static ssize_t amdgpu_get_pp_sclk_od(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t value = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1499,11 +1517,11 @@ static ssize_t amdgpu_set_pp_sclk_od(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	long int value;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = kstrtol(buf, 0, &value);
@@ -1542,11 +1560,11 @@ static ssize_t amdgpu_get_pp_mclk_od(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t value = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1572,11 +1590,11 @@ static ssize_t amdgpu_set_pp_mclk_od(struct device *dev,
 		size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int ret;
 	long int value;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = kstrtol(buf, 0, &value);
@@ -1635,11 +1653,11 @@ static ssize_t amdgpu_get_pp_power_profile_mode(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	ssize_t size;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	ret = pm_runtime_get_sync(ddev->dev);
@@ -1669,7 +1687,7 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 {
 	int ret;
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint32_t parameter_size = 0;
 	long parameter[64];
 	char *sub_str, buf_cpy[128];
@@ -1679,7 +1697,7 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 	long int profile_mode = 0;
 	const char delimiter[3] = {' ', '\n', '\0'};
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	tmp[0] = *(buf);
@@ -1695,7 +1713,8 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 			i++;
 		memcpy(buf_cpy, buf, count-i);
 		tmp_str = buf_cpy;
-		while ((sub_str = strsep(&tmp_str, delimiter)) != NULL) {
+		while (tmp_str[0]) {
+			sub_str = strsep(&tmp_str, delimiter);
 			ret = kstrtol(sub_str, 0, &parameter[parameter_size]);
 			if (ret)
 				return -EINVAL;
@@ -1739,10 +1758,10 @@ static ssize_t amdgpu_get_gpu_busy_percent(struct device *dev,
 					   char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int r, value, size = sizeof(value);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	r = pm_runtime_get_sync(ddev->dev);
@@ -1777,10 +1796,10 @@ static ssize_t amdgpu_get_mem_busy_percent(struct device *dev,
 					   char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int r, value, size = sizeof(value);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	r = pm_runtime_get_sync(ddev->dev);
@@ -1819,11 +1838,11 @@ static ssize_t amdgpu_get_pcie_bw(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	uint64_t count0 = 0, count1 = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (adev->flags & AMD_IS_APU)
@@ -1862,9 +1881,9 @@ static ssize_t amdgpu_get_unique_id(struct device *dev,
 		char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (adev->unique_id)
@@ -1893,10 +1912,10 @@ static ssize_t amdgpu_get_thermal_throttling_logging(struct device *dev,
 						     char *buf)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 
 	return snprintf(buf, PAGE_SIZE, "%s: thermal throttling logging %s, with interval %d seconds\n",
-			adev->ddev->unique,
+			adev_to_drm(adev)->unique,
 			atomic_read(&adev->throttling_logging_enabled) ? "enabled" : "disabled",
 			adev->throttling_logging_rs.interval / HZ + 1);
 }
@@ -1907,7 +1926,7 @@ static ssize_t amdgpu_set_thermal_throttling_logging(struct device *dev,
 						     size_t count)
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = ddev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(ddev);
 	long throttling_logging_interval;
 	unsigned long flags;
 	int ret = 0;
@@ -1940,6 +1959,57 @@ static ssize_t amdgpu_set_thermal_throttling_logging(struct device *dev,
 	return count;
 }
 
+/**
+ * DOC: gpu_metrics
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current gpu
+ * metrics data. The file gpu_metrics is used for this. Reading the
+ * file will dump all the current gpu metrics data.
+ *
+ * These data include temperature, frequency, engines utilization,
+ * power consume, throttler status, fan speed and cpu core statistics(
+ * available for APU only). That's it will give a snapshot of all sensors
+ * at the same time.
+ */
+static ssize_t amdgpu_get_gpu_metrics(struct device *dev,
+				      struct device_attribute *attr,
+				      char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	void *gpu_metrics;
+	ssize_t size = 0;
+	int ret;
+
+	if (amdgpu_in_reset(adev))
+		return -EPERM;
+
+	ret = pm_runtime_get_sync(ddev->dev);
+	if (ret < 0) {
+		pm_runtime_put_autosuspend(ddev->dev);
+		return ret;
+	}
+
+	if (is_support_sw_smu(adev))
+		size = smu_sys_get_gpu_metrics(&adev->smu, &gpu_metrics);
+	else if (adev->powerplay.pp_funcs->get_gpu_metrics)
+		size = amdgpu_dpm_get_gpu_metrics(adev, &gpu_metrics);
+
+	if (size <= 0)
+		goto out;
+
+	if (size >= PAGE_SIZE)
+		size = PAGE_SIZE - 1;
+
+	memcpy(buf, gpu_metrics, size);
+
+out:
+	pm_runtime_mark_last_busy(ddev->dev);
+	pm_runtime_put_autosuspend(ddev->dev);
+
+	return size;
+}
+
 static struct amdgpu_device_attr amdgpu_device_attrs[] = {
 	AMDGPU_DEVICE_ATTR_RW(power_dpm_state,				ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF),
 	AMDGPU_DEVICE_ATTR_RW(power_dpm_force_performance_level,	ATTR_FLAG_BASIC),
@@ -1963,6 +2033,7 @@ static struct amdgpu_device_attr amdgpu_device_attrs[] = {
 	AMDGPU_DEVICE_ATTR_RW(pp_features,				ATTR_FLAG_BASIC),
 	AMDGPU_DEVICE_ATTR_RO(unique_id,				ATTR_FLAG_BASIC),
 	AMDGPU_DEVICE_ATTR_RW(thermal_throttling_logging,		ATTR_FLAG_BASIC),
+	AMDGPU_DEVICE_ATTR_RO(gpu_metrics,				ATTR_FLAG_BASIC),
 };
 
 static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr,
@@ -2012,6 +2083,9 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 	} else if (DEVICE_ATTR_IS(pp_features)) {
 		if (adev->flags & AMD_IS_APU || asic_type < CHIP_VEGA10)
 			*states = ATTR_STATE_UNSUPPORTED;
+	} else if (DEVICE_ATTR_IS(gpu_metrics)) {
+		if (asic_type < CHIP_VEGA12)
+			*states = ATTR_STATE_UNSUPPORTED;
 	}
 
 	if (asic_type == CHIP_ARCTURUS) {
@@ -2131,15 +2205,15 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 	int channel = to_sensor_dev_attr(attr)->index;
 	int r, temp = 0, size = sizeof(temp);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (channel >= PP_TEMP_MAX)
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2164,8 +2238,8 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 		break;
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2267,12 +2341,12 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev,
 	u32 pwm_mode = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	ret = pm_runtime_get_sync(adev->ddev->dev);
+	ret = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (ret < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return ret;
 	}
 
@@ -2280,16 +2354,16 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev,
 		pwm_mode = smu_get_fan_control_mode(&adev->smu);
 	} else {
 		if (!adev->powerplay.pp_funcs->get_fan_control_mode) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return -EINVAL;
 		}
 
 		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return sprintf(buf, "%i\n", pwm_mode);
 }
@@ -2303,16 +2377,16 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev,
 	int err, ret;
 	int value;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	err = kstrtoint(buf, 10, &value);
 	if (err)
 		return err;
 
-	ret = pm_runtime_get_sync(adev->ddev->dev);
+	ret = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (ret < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return ret;
 	}
 
@@ -2320,16 +2394,16 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev,
 		smu_set_fan_control_mode(&adev->smu, value);
 	} else {
 		if (!adev->powerplay.pp_funcs->set_fan_control_mode) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return -EINVAL;
 		}
 
 		amdgpu_dpm_set_fan_control_mode(adev, value);
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return count;
 }
@@ -2357,12 +2431,12 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev,
 	u32 value;
 	u32 pwm_mode;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2373,15 +2447,15 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev,
 
 	if (pwm_mode != AMD_FAN_CTRL_MANUAL) {
 		pr_info("manual fan speed control should be enabled first\n");
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return -EINVAL;
 	}
 
 	err = kstrtou32(buf, 10, &value);
 	if (err) {
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2394,8 +2468,8 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2411,12 +2485,12 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev,
 	int err;
 	u32 speed = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2427,8 +2501,8 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2446,12 +2520,12 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev,
 	int err;
 	u32 speed = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2462,8 +2536,8 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2480,20 +2554,20 @@ static ssize_t amdgpu_hwmon_get_fan1_min(struct device *dev,
 	u32 size = sizeof(min_rpm);
 	int r;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM,
 				   (void *)&min_rpm, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2510,20 +2584,20 @@ static ssize_t amdgpu_hwmon_get_fan1_max(struct device *dev,
 	u32 size = sizeof(max_rpm);
 	int r;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM,
 				   (void *)&max_rpm, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2539,12 +2613,12 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev,
 	int err;
 	u32 rpm = 0;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2555,8 +2629,8 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2573,12 +2647,12 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev,
 	u32 value;
 	u32 pwm_mode;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2588,15 +2662,15 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev,
 		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
 
 	if (pwm_mode != AMD_FAN_CTRL_MANUAL) {
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return -ENODATA;
 	}
 
 	err = kstrtou32(buf, 10, &value);
 	if (err) {
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2607,8 +2681,8 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2624,12 +2698,12 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev,
 	u32 pwm_mode = 0;
 	int ret;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	ret = pm_runtime_get_sync(adev->ddev->dev);
+	ret = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (ret < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return ret;
 	}
 
@@ -2637,16 +2711,16 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev,
 		pwm_mode = smu_get_fan_control_mode(&adev->smu);
 	} else {
 		if (!adev->powerplay.pp_funcs->get_fan_control_mode) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return -EINVAL;
 		}
 
 		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return sprintf(buf, "%i\n", pwm_mode == AMD_FAN_CTRL_AUTO ? 0 : 1);
 }
@@ -2661,7 +2735,7 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev,
 	int value;
 	u32 pwm_mode;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	err = kstrtoint(buf, 10, &value);
@@ -2675,9 +2749,9 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev,
 	else
 		return -EINVAL;
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2685,15 +2759,15 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev,
 		smu_set_fan_control_mode(&adev->smu, pwm_mode);
 	} else {
 		if (!adev->powerplay.pp_funcs->set_fan_control_mode) {
-			pm_runtime_mark_last_busy(adev->ddev->dev);
-			pm_runtime_put_autosuspend(adev->ddev->dev);
+			pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+			pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 			return -EINVAL;
 		}
 		amdgpu_dpm_set_fan_control_mode(adev, pwm_mode);
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return count;
 }
@@ -2706,12 +2780,12 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev,
 	u32 vddgfx;
 	int r, size = sizeof(vddgfx);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2719,8 +2793,8 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev,
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDGFX,
 				   (void *)&vddgfx, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2743,16 +2817,16 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev,
 	u32 vddnb;
 	int r, size = sizeof(vddnb);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	/* only APUs have vddnb */
 	if  (!(adev->flags & AMD_IS_APU))
 		return -EINVAL;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2760,8 +2834,8 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev,
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDNB,
 				   (void *)&vddnb, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2785,12 +2859,12 @@ static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev,
 	int r, size = sizeof(u32);
 	unsigned uw;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2798,8 +2872,8 @@ static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev,
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_POWER,
 				   (void *)&query, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2826,12 +2900,12 @@ static ssize_t amdgpu_hwmon_show_power_cap_max(struct device *dev,
 	ssize_t size;
 	int r;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2845,8 +2919,8 @@ static ssize_t amdgpu_hwmon_show_power_cap_max(struct device *dev,
 		size = snprintf(buf, PAGE_SIZE, "\n");
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return size;
 }
@@ -2860,12 +2934,12 @@ static ssize_t amdgpu_hwmon_show_power_cap(struct device *dev,
 	ssize_t size;
 	int r;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2879,8 +2953,8 @@ static ssize_t amdgpu_hwmon_show_power_cap(struct device *dev,
 		size = snprintf(buf, PAGE_SIZE, "\n");
 	}
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	return size;
 }
@@ -2895,7 +2969,7 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 	int err;
 	u32 value;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	if (amdgpu_sriov_vf(adev))
@@ -2908,9 +2982,9 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 	value = value / 1000000; /* convert to Watt */
 
 
-	err = pm_runtime_get_sync(adev->ddev->dev);
+	err = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (err < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return err;
 	}
 
@@ -2921,8 +2995,8 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 	else
 		err = -EINVAL;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (err)
 		return err;
@@ -2938,12 +3012,12 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev,
 	uint32_t sclk;
 	int r, size = sizeof(sclk);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2951,8 +3025,8 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev,
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_SCLK,
 				   (void *)&sclk, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -2975,12 +3049,12 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev,
 	uint32_t mclk;
 	int r, size = sizeof(mclk);
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
-	r = pm_runtime_get_sync(adev->ddev->dev);
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 		return r;
 	}
 
@@ -2988,8 +3062,8 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev,
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_MCLK,
 				   (void *)&mclk, &size);
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
 
 	if (r)
 		return r;
@@ -3249,14 +3323,18 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	}
 
 	if (((adev->flags & AMD_IS_APU) ||
-	     adev->family == AMDGPU_FAMILY_SI ||	/* not implemented yet */
-	     adev->family == AMDGPU_FAMILY_KV) &&	/* not implemented yet */
-	    (attr == &sensor_dev_attr_power1_average.dev_attr.attr ||
-	     attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
+	     adev->family == AMDGPU_FAMILY_SI) &&	/* not implemented yet */
+	    (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
 	     attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr||
 	     attr == &sensor_dev_attr_power1_cap.dev_attr.attr))
 		return 0;
 
+	if (((adev->family == AMDGPU_FAMILY_SI) ||
+	     ((adev->flags & AMD_IS_APU) &&
+	      (adev->asic_type < CHIP_RENOIR))) &&	/* not implemented yet */
+	    (attr == &sensor_dev_attr_power1_average.dev_attr.attr))
+		return 0;
+
 	if (!is_support_sw_smu(adev)) {
 		/* hide max/min values if we can't both query and manage the fan */
 		if ((!adev->powerplay.pp_funcs->set_fan_speed_percent &&
@@ -3321,338 +3399,6 @@ static const struct attribute_group *hwmon_groups[] = {
 	NULL
 };
 
-void amdgpu_dpm_thermal_work_handler(struct work_struct *work)
-{
-	struct amdgpu_device *adev =
-		container_of(work, struct amdgpu_device,
-			     pm.dpm.thermal.work);
-	/* switch to the thermal state */
-	enum amd_pm_state_type dpm_state = POWER_STATE_TYPE_INTERNAL_THERMAL;
-	int temp, size = sizeof(temp);
-
-	if (!adev->pm.dpm_enabled)
-		return;
-
-	if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP,
-				    (void *)&temp, &size)) {
-		if (temp < adev->pm.dpm.thermal.min_temp)
-			/* switch back the user state */
-			dpm_state = adev->pm.dpm.user_state;
-	} else {
-		if (adev->pm.dpm.thermal.high_to_low)
-			/* switch back the user state */
-			dpm_state = adev->pm.dpm.user_state;
-	}
-	mutex_lock(&adev->pm.mutex);
-	if (dpm_state == POWER_STATE_TYPE_INTERNAL_THERMAL)
-		adev->pm.dpm.thermal_active = true;
-	else
-		adev->pm.dpm.thermal_active = false;
-	adev->pm.dpm.state = dpm_state;
-	mutex_unlock(&adev->pm.mutex);
-
-	amdgpu_pm_compute_clocks(adev);
-}
-
-static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev,
-						     enum amd_pm_state_type dpm_state)
-{
-	int i;
-	struct amdgpu_ps *ps;
-	u32 ui_class;
-	bool single_display = (adev->pm.dpm.new_active_crtc_count < 2) ?
-		true : false;
-
-	/* check if the vblank period is too short to adjust the mclk */
-	if (single_display && adev->powerplay.pp_funcs->vblank_too_short) {
-		if (amdgpu_dpm_vblank_too_short(adev))
-			single_display = false;
-	}
-
-	/* certain older asics have a separare 3D performance state,
-	 * so try that first if the user selected performance
-	 */
-	if (dpm_state == POWER_STATE_TYPE_PERFORMANCE)
-		dpm_state = POWER_STATE_TYPE_INTERNAL_3DPERF;
-	/* balanced states don't exist at the moment */
-	if (dpm_state == POWER_STATE_TYPE_BALANCED)
-		dpm_state = POWER_STATE_TYPE_PERFORMANCE;
-
-restart_search:
-	/* Pick the best power state based on current conditions */
-	for (i = 0; i < adev->pm.dpm.num_ps; i++) {
-		ps = &adev->pm.dpm.ps[i];
-		ui_class = ps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK;
-		switch (dpm_state) {
-		/* user states */
-		case POWER_STATE_TYPE_BATTERY:
-			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY) {
-				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
-					if (single_display)
-						return ps;
-				} else
-					return ps;
-			}
-			break;
-		case POWER_STATE_TYPE_BALANCED:
-			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BALANCED) {
-				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
-					if (single_display)
-						return ps;
-				} else
-					return ps;
-			}
-			break;
-		case POWER_STATE_TYPE_PERFORMANCE:
-			if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE) {
-				if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) {
-					if (single_display)
-						return ps;
-				} else
-					return ps;
-			}
-			break;
-		/* internal states */
-		case POWER_STATE_TYPE_INTERNAL_UVD:
-			if (adev->pm.dpm.uvd_ps)
-				return adev->pm.dpm.uvd_ps;
-			else
-				break;
-		case POWER_STATE_TYPE_INTERNAL_UVD_SD:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_SDSTATE)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_UVD_HD:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_HDSTATE)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_HD2STATE)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
-			if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_MVC)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_BOOT:
-			return adev->pm.dpm.boot_ps;
-		case POWER_STATE_TYPE_INTERNAL_THERMAL:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_THERMAL)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_ACPI:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_ACPI)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_ULV:
-			if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV)
-				return ps;
-			break;
-		case POWER_STATE_TYPE_INTERNAL_3DPERF:
-			if (ps->class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE)
-				return ps;
-			break;
-		default:
-			break;
-		}
-	}
-	/* use a fallback state if we didn't match */
-	switch (dpm_state) {
-	case POWER_STATE_TYPE_INTERNAL_UVD_SD:
-		dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
-		goto restart_search;
-	case POWER_STATE_TYPE_INTERNAL_UVD_HD:
-	case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
-	case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
-		if (adev->pm.dpm.uvd_ps) {
-			return adev->pm.dpm.uvd_ps;
-		} else {
-			dpm_state = POWER_STATE_TYPE_PERFORMANCE;
-			goto restart_search;
-		}
-	case POWER_STATE_TYPE_INTERNAL_THERMAL:
-		dpm_state = POWER_STATE_TYPE_INTERNAL_ACPI;
-		goto restart_search;
-	case POWER_STATE_TYPE_INTERNAL_ACPI:
-		dpm_state = POWER_STATE_TYPE_BATTERY;
-		goto restart_search;
-	case POWER_STATE_TYPE_BATTERY:
-	case POWER_STATE_TYPE_BALANCED:
-	case POWER_STATE_TYPE_INTERNAL_3DPERF:
-		dpm_state = POWER_STATE_TYPE_PERFORMANCE;
-		goto restart_search;
-	default:
-		break;
-	}
-
-	return NULL;
-}
-
-static void amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev)
-{
-	struct amdgpu_ps *ps;
-	enum amd_pm_state_type dpm_state;
-	int ret;
-	bool equal = false;
-
-	/* if dpm init failed */
-	if (!adev->pm.dpm_enabled)
-		return;
-
-	if (adev->pm.dpm.user_state != adev->pm.dpm.state) {
-		/* add other state override checks here */
-		if ((!adev->pm.dpm.thermal_active) &&
-		    (!adev->pm.dpm.uvd_active))
-			adev->pm.dpm.state = adev->pm.dpm.user_state;
-	}
-	dpm_state = adev->pm.dpm.state;
-
-	ps = amdgpu_dpm_pick_power_state(adev, dpm_state);
-	if (ps)
-		adev->pm.dpm.requested_ps = ps;
-	else
-		return;
-
-	if (amdgpu_dpm == 1 && adev->powerplay.pp_funcs->print_power_state) {
-		printk("switching from power state:\n");
-		amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps);
-		printk("switching to power state:\n");
-		amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps);
-	}
-
-	/* update whether vce is active */
-	ps->vce_active = adev->pm.dpm.vce_active;
-	if (adev->powerplay.pp_funcs->display_configuration_changed)
-		amdgpu_dpm_display_configuration_changed(adev);
-
-	ret = amdgpu_dpm_pre_set_power_state(adev);
-	if (ret)
-		return;
-
-	if (adev->powerplay.pp_funcs->check_state_equal) {
-		if (0 != amdgpu_dpm_check_state_equal(adev, adev->pm.dpm.current_ps, adev->pm.dpm.requested_ps, &equal))
-			equal = false;
-	}
-
-	if (equal)
-		return;
-
-	amdgpu_dpm_set_power_state(adev);
-	amdgpu_dpm_post_set_power_state(adev);
-
-	adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs;
-	adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count;
-
-	if (adev->powerplay.pp_funcs->force_performance_level) {
-		if (adev->pm.dpm.thermal_active) {
-			enum amd_dpm_forced_level level = adev->pm.dpm.forced_level;
-			/* force low perf level for thermal */
-			amdgpu_dpm_force_performance_level(adev, AMD_DPM_FORCED_LEVEL_LOW);
-			/* save the user's level */
-			adev->pm.dpm.forced_level = level;
-		} else {
-			/* otherwise, user selected level */
-			amdgpu_dpm_force_performance_level(adev, adev->pm.dpm.forced_level);
-		}
-	}
-}
-
-void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable)
-{
-	int ret = 0;
-
-	if (adev->family == AMDGPU_FAMILY_SI) {
-		mutex_lock(&adev->pm.mutex);
-		if (enable) {
-			adev->pm.dpm.uvd_active = true;
-			adev->pm.dpm.state = POWER_STATE_TYPE_INTERNAL_UVD;
-		} else {
-			adev->pm.dpm.uvd_active = false;
-		}
-		mutex_unlock(&adev->pm.mutex);
-
-		amdgpu_pm_compute_clocks(adev);
-	} else {
-		ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable);
-		if (ret)
-			DRM_ERROR("Dpm %s uvd failed, ret = %d. \n",
-				  enable ? "enable" : "disable", ret);
-
-		/* enable/disable Low Memory PState for UVD (4k videos) */
-		if (adev->asic_type == CHIP_STONEY &&
-			adev->uvd.decode_image_width >= WIDTH_4K) {
-			struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
-
-			if (hwmgr && hwmgr->hwmgr_func &&
-			    hwmgr->hwmgr_func->update_nbdpm_pstate)
-				hwmgr->hwmgr_func->update_nbdpm_pstate(hwmgr,
-								       !enable,
-								       true);
-		}
-	}
-}
-
-void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable)
-{
-	int ret = 0;
-
-	if (adev->family == AMDGPU_FAMILY_SI) {
-		mutex_lock(&adev->pm.mutex);
-		if (enable) {
-			adev->pm.dpm.vce_active = true;
-			/* XXX select vce level based on ring/task */
-			adev->pm.dpm.vce_level = AMD_VCE_LEVEL_AC_ALL;
-		} else {
-			adev->pm.dpm.vce_active = false;
-		}
-		mutex_unlock(&adev->pm.mutex);
-
-		amdgpu_pm_compute_clocks(adev);
-	} else {
-		ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VCE, !enable);
-		if (ret)
-			DRM_ERROR("Dpm %s vce failed, ret = %d. \n",
-				  enable ? "enable" : "disable", ret);
-	}
-}
-
-void amdgpu_pm_print_power_states(struct amdgpu_device *adev)
-{
-	int i;
-
-	if (adev->powerplay.pp_funcs->print_power_state == NULL)
-		return;
-
-	for (i = 0; i < adev->pm.dpm.num_ps; i++)
-		amdgpu_dpm_print_power_state(adev, &adev->pm.dpm.ps[i]);
-
-}
-
-void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable)
-{
-	int ret = 0;
-
-	ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_JPEG, !enable);
-	if (ret)
-		DRM_ERROR("Dpm %s jpeg failed, ret = %d. \n",
-			  enable ? "enable" : "disable", ret);
-}
-
-int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version)
-{
-	int r;
-
-	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->load_firmware) {
-		r = adev->powerplay.pp_funcs->load_firmware(adev->powerplay.pp_handle);
-		if (r) {
-			pr_err("smu firmware loading failed\n");
-			return r;
-		}
-		*smu_version = adev->pm.fw_version;
-	}
-	return 0;
-}
-
 int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 {
 	int ret;
@@ -3713,55 +3459,6 @@ void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev)
 	amdgpu_device_attr_remove_groups(adev, &adev->pm.pm_attr_list);
 }
 
-void amdgpu_pm_compute_clocks(struct amdgpu_device *adev)
-{
-	int i = 0;
-
-	if (!adev->pm.dpm_enabled)
-		return;
-
-	if (adev->mode_info.num_crtc)
-		amdgpu_display_bandwidth_update(adev);
-
-	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
-		struct amdgpu_ring *ring = adev->rings[i];
-		if (ring && ring->sched.ready)
-			amdgpu_fence_wait_empty(ring);
-	}
-
-	if (is_support_sw_smu(adev)) {
-		struct smu_dpm_context *smu_dpm = &adev->smu.smu_dpm;
-		smu_handle_task(&adev->smu,
-				smu_dpm->dpm_level,
-				AMD_PP_TASK_DISPLAY_CONFIG_CHANGE,
-				true);
-	} else {
-		if (adev->powerplay.pp_funcs->dispatch_tasks) {
-			if (!amdgpu_device_has_dc_support(adev)) {
-				mutex_lock(&adev->pm.mutex);
-				amdgpu_dpm_get_active_displays(adev);
-				adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count;
-				adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev);
-				adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev);
-				/* we have issues with mclk switching with refresh rates over 120 hz on the non-DC code. */
-				if (adev->pm.pm_display_cfg.vrefresh > 120)
-					adev->pm.pm_display_cfg.min_vblank_time = 0;
-				if (adev->powerplay.pp_funcs->display_configuration_change)
-					adev->powerplay.pp_funcs->display_configuration_change(
-									adev->powerplay.pp_handle,
-									&adev->pm.pm_display_cfg);
-				mutex_unlock(&adev->pm.mutex);
-			}
-			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, NULL);
-		} else {
-			mutex_lock(&adev->pm.mutex);
-			amdgpu_dpm_get_active_displays(adev);
-			amdgpu_dpm_change_power_state_locked(adev);
-			mutex_unlock(&adev->pm.mutex);
-		}
-	}
-}
-
 /*
  * Debugfs info
  */
@@ -3869,11 +3566,11 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct amdgpu_device *adev = dev->dev_private;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 	u32 flags = 0;
 	int r;
 
-	if (adev->in_gpu_reset)
+	if (amdgpu_in_reset(adev))
 		return -EPERM;
 
 	r = pm_runtime_get_sync(dev->dev);
@@ -3882,11 +3579,6 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data)
 		return r;
 	}
 
-	amdgpu_device_ip_get_clockgating_state(adev, &flags);
-	seq_printf(m, "Clock Gating Flags Mask: 0x%x\n", flags);
-	amdgpu_parse_cg_state(m, flags);
-	seq_printf(m, "\n");
-
 	if (!adev->pm.dpm_enabled) {
 		seq_printf(m, "dpm not enabled\n");
 		pm_runtime_mark_last_busy(dev->dev);
@@ -3906,7 +3598,16 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data)
 	} else {
 		r = amdgpu_debugfs_pm_info_pp(m, adev);
 	}
+	if (r)
+		goto out;
+
+	amdgpu_device_ip_get_clockgating_state(adev, &flags);
+
+	seq_printf(m, "Clock Gating Flags Mask: 0x%x\n", flags);
+	amdgpu_parse_cg_state(m, flags);
+	seq_printf(m, "\n");
 
+out:
 	pm_runtime_mark_last_busy(dev->dev);
 	pm_runtime_put_autosuspend(dev->dev);
 
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amd_powerplay.h b/drivers/gpu/drm/amd/pm/inc/amd_powerplay.h
index fe3665965416..fe3665965416 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/amd_powerplay.h
+++ b/drivers/gpu/drm/amd/pm/inc/amd_powerplay.h
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
index aa27fe65cdfa..f6e0e7d8a007 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
@@ -341,10 +341,6 @@ enum amdgpu_pcie_gen {
 		((adev)->powerplay.pp_funcs->reset_power_profile_state(\
 			(adev)->powerplay.pp_handle, request))
 
-#define amdgpu_dpm_set_clockgating_by_smu(adev, msg_id) \
-		((adev)->powerplay.pp_funcs->set_clockgating_by_smu(\
-			(adev)->powerplay.pp_handle, msg_id))
-
 #define amdgpu_dpm_get_power_profile_mode(adev, buf) \
 		((adev)->powerplay.pp_funcs->get_power_profile_mode(\
 			(adev)->powerplay.pp_handle, buf))
@@ -353,14 +349,14 @@ enum amdgpu_pcie_gen {
 		((adev)->powerplay.pp_funcs->set_power_profile_mode(\
 			(adev)->powerplay.pp_handle, parameter, size))
 
+#define amdgpu_dpm_set_fine_grain_clk_vol(adev, type, parameter, size) \
+		((adev)->powerplay.pp_funcs->set_fine_grain_clk_vol(\
+			(adev)->powerplay.pp_handle, type, parameter, size))
+
 #define amdgpu_dpm_odn_edit_dpm_table(adev, type, parameter, size) \
 		((adev)->powerplay.pp_funcs->odn_edit_dpm_table(\
 			(adev)->powerplay.pp_handle, type, parameter, size))
 
-#define amdgpu_dpm_enable_mgpu_fan_boost(adev) \
-		((adev)->powerplay.pp_funcs->enable_mgpu_fan_boost(\
-			(adev)->powerplay.pp_handle))
-
 #define amdgpu_dpm_get_ppfeature_status(adev, buf) \
 		((adev)->powerplay.pp_funcs->get_ppfeature_status(\
 			(adev)->powerplay.pp_handle, (buf)))
@@ -369,6 +365,9 @@ enum amdgpu_pcie_gen {
 		((adev)->powerplay.pp_funcs->set_ppfeature_status(\
 			(adev)->powerplay.pp_handle, (ppfeatures)))
 
+#define amdgpu_dpm_get_gpu_metrics(adev, table) \
+		((adev)->powerplay.pp_funcs->get_gpu_metrics((adev)->powerplay.pp_handle, table))
+
 struct amdgpu_dpm {
 	struct amdgpu_ps        *ps;
 	/* number of valid power states */
@@ -545,4 +544,26 @@ int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev,
 
 int amdgpu_dpm_allow_xgmi_power_down(struct amdgpu_device *adev, bool en);
 
+int amdgpu_dpm_enable_mgpu_fan_boost(struct amdgpu_device *adev);
+
+int amdgpu_dpm_set_clockgating_by_smu(struct amdgpu_device *adev,
+				      uint32_t msg_id);
+
+int amdgpu_dpm_smu_i2c_bus_access(struct amdgpu_device *adev,
+				  bool acquire);
+
+void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev);
+
+int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size);
+
+void amdgpu_dpm_thermal_work_handler(struct work_struct *work);
+
+void amdgpu_pm_compute_clocks(struct amdgpu_device *adev);
+void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable);
+void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable);
+void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable);
+void amdgpu_pm_print_power_states(struct amdgpu_device *adev);
+int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h
index d9ae2b49a402..45a22e101d15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h
@@ -79,18 +79,10 @@ struct amdgpu_device_attr_entry {
 			     amdgpu_get_##_name, NULL,			\
 			     _flags, ##__VA_ARGS__)
 
-void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev);
 int amdgpu_pm_sysfs_init(struct amdgpu_device *adev);
 int amdgpu_pm_virt_sysfs_init(struct amdgpu_device *adev);
 void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev);
 void amdgpu_pm_virt_sysfs_fini(struct amdgpu_device *adev);
-void amdgpu_pm_print_power_states(struct amdgpu_device *adev);
-int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version);
-void amdgpu_pm_compute_clocks(struct amdgpu_device *adev);
-void amdgpu_dpm_thermal_work_handler(struct work_struct *work);
-void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable);
-void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable);
-void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable);
 
 int amdgpu_debugfs_pm_init(struct amdgpu_device *adev);
 
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h
index 074458eb5407..44fd0cd069de 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h
@@ -270,10 +270,14 @@ struct smu_table_context
 	 */
 	struct smu_table		driver_table;
 	struct smu_table		memory_pool;
+	struct smu_table		dummy_read_1_table;
 	uint8_t                         thermal_controller_type;
 
 	void				*overdrive_table;
 	void                            *boot_overdrive_table;
+
+	uint32_t			gpu_metrics_table_size;
+	void				*gpu_metrics_table;
 };
 
 struct smu_dpm_context {
@@ -448,6 +452,11 @@ struct smu_context
 	bool dc_controlled_by_gpio;
 
 	struct work_struct throttling_logging_work;
+	atomic64_t throttle_int_counter;
+	struct work_struct interrupt_work;
+
+	unsigned fan_max_rpm;
+	unsigned manual_fan_speed_rpm;
 };
 
 struct i2c_adapter;
@@ -491,10 +500,9 @@ struct pptable_funcs {
 	int (*notify_smc_display_config)(struct smu_context *smu);
 	int (*set_cpu_power_state)(struct smu_context *smu);
 	bool (*is_dpm_running)(struct smu_context *smu);
-	int (*get_fan_speed_percent)(struct smu_context *smu, uint32_t *speed);
 	int (*get_fan_speed_rpm)(struct smu_context *smu, uint32_t *speed);
 	int (*set_watermarks_table)(struct smu_context *smu,
-				    struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges);
+				    struct pp_smu_wm_range_sets *clock_ranges);
 	int (*get_thermal_temperature_range)(struct smu_context *smu, struct smu_temperature_range *range);
 	int (*get_uclk_dpm_states)(struct smu_context *smu, uint32_t *clocks_in_khz, uint32_t *num_states);
 	int (*set_default_od_settings)(struct smu_context *smu);
@@ -567,7 +575,6 @@ struct pptable_funcs {
 	int (*conv_power_profile_to_pplib_workload)(int power_profile);
 	uint32_t (*get_fan_control_mode)(struct smu_context *smu);
 	int (*set_fan_control_mode)(struct smu_context *smu, uint32_t mode);
-	int (*set_fan_speed_percent)(struct smu_context *smu, uint32_t speed);
 	int (*set_fan_speed_rpm)(struct smu_context *smu, uint32_t speed);
 	int (*set_xgmi_pstate)(struct smu_context *smu, uint32_t pstate);
 	int (*gfx_off_control)(struct smu_context *smu, bool enable);
@@ -585,11 +592,17 @@ struct pptable_funcs {
 	int (*mode2_reset)(struct smu_context *smu);
 	int (*get_dpm_ultimate_freq)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max);
 	int (*set_soft_freq_limited_range)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max);
-	int (*disable_umc_cdr_12gbps_workaround)(struct smu_context *smu);
 	int (*set_power_source)(struct smu_context *smu, enum smu_power_src_type power_src);
 	void (*log_thermal_throttling_event)(struct smu_context *smu);
 	size_t (*get_pp_feature_mask)(struct smu_context *smu, char *buf);
 	int (*set_pp_feature_mask)(struct smu_context *smu, uint64_t new_mask);
+	ssize_t (*get_gpu_metrics)(struct smu_context *smu, void **table);
+	int (*enable_mgpu_fan_boost)(struct smu_context *smu);
+	int (*gfx_ulv_control)(struct smu_context *smu, bool enablement);
+	int (*deep_sleep_control)(struct smu_context *smu, bool enablement);
+	int (*get_fan_parameters)(struct smu_context *smu);
+	int (*post_init)(struct smu_context *smu);
+	void (*interrupt_work)(struct smu_context *smu);
 };
 
 typedef enum {
@@ -693,7 +706,6 @@ int smu_set_fan_speed_percent(struct smu_context *smu, uint32_t speed);
 int smu_get_fan_speed_rpm(struct smu_context *smu, uint32_t *speed);
 
 int smu_set_deep_sleep_dcefclk(struct smu_context *smu, int clk);
-int smu_set_active_display_count(struct smu_context *smu, uint32_t count);
 
 int smu_get_clock_by_type(struct smu_context *smu,
 			  enum amd_pp_clock_type type,
@@ -745,7 +757,7 @@ enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu);
 int smu_write_watermarks_table(struct smu_context *smu);
 int smu_set_watermarks_for_clock_ranges(
 		struct smu_context *smu,
-		struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges);
+		struct pp_smu_wm_range_sets *clock_ranges);
 
 /* smu to display interface */
 extern int smu_display_configuration_change(struct smu_context *smu, const
@@ -792,5 +804,9 @@ int smu_get_dpm_clock_table(struct smu_context *smu,
 
 int smu_get_status_gfxoff(struct amdgpu_device *adev, uint32_t *value);
 
+ssize_t smu_sys_get_gpu_metrics(struct smu_context *smu, void **table);
+
+int smu_enable_mgpu_fan_boost(struct smu_context *smu);
+
 #endif
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/arcturus_ppsmc.h
index 79afb132164e..79afb132164e 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/arcturus_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/cz_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/cz_ppsmc.h
index 9b698780aed8..9b698780aed8 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/cz_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/cz_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/fiji_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/fiji_ppsmc.h
index 7ae494569a60..7ae494569a60 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/fiji_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/fiji_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h b/drivers/gpu/drm/amd/pm/inc/hardwaremanager.h
index 6e0be6027705..6e0be6027705 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h
+++ b/drivers/gpu/drm/amd/pm/inc/hardwaremanager.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/inc/hwmgr.h
index 15ed6cbdf366..3898a95ec28b 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/inc/hwmgr.h
@@ -340,6 +340,9 @@ struct pp_hwmgr_func {
 	int (*odn_edit_dpm_table)(struct pp_hwmgr *hwmgr,
 					enum PP_OD_DPM_TABLE_COMMAND type,
 					long *input, uint32_t size);
+	int (*set_fine_grain_clk_vol)(struct pp_hwmgr *hwmgr,
+				      enum PP_OD_DPM_TABLE_COMMAND type,
+				      long *input, uint32_t size);
 	int (*set_power_limit)(struct pp_hwmgr *hwmgr, uint32_t n);
 	int (*powergate_mmhub)(struct pp_hwmgr *hwmgr);
 	int (*smus_notify_pwe)(struct pp_hwmgr *hwmgr);
@@ -347,6 +350,8 @@ struct pp_hwmgr_func {
 	int (*enable_mgpu_fan_boost)(struct pp_hwmgr *hwmgr);
 	int (*set_hard_min_dcefclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock);
 	int (*set_hard_min_fclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock);
+	int (*set_hard_min_gfxclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock);
+	int (*set_soft_max_gfxclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock);
 	int (*get_asic_baco_capability)(struct pp_hwmgr *hwmgr, bool *cap);
 	int (*get_asic_baco_state)(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
 	int (*set_asic_baco_state)(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
@@ -359,6 +364,7 @@ struct pp_hwmgr_func {
 	int (*set_xgmi_pstate)(struct pp_hwmgr *hwmgr, uint32_t pstate);
 	int (*disable_power_features_for_compute_performance)(struct pp_hwmgr *hwmgr,
 					bool disable);
+	ssize_t (*get_gpu_metrics)(struct pp_hwmgr *hwmgr, void **table);
 };
 
 struct pp_table_func {
diff --git a/drivers/gpu/drm/amd/powerplay/inc/polaris10_pwrvirus.h b/drivers/gpu/drm/amd/pm/inc/polaris10_pwrvirus.h
index 6a53b7e74ccd..6a53b7e74ccd 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/polaris10_pwrvirus.h
+++ b/drivers/gpu/drm/amd/pm/inc/polaris10_pwrvirus.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/power_state.h b/drivers/gpu/drm/amd/pm/inc/power_state.h
index a5f2227a3971..a5f2227a3971 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/power_state.h
+++ b/drivers/gpu/drm/amd/pm/inc/power_state.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h b/drivers/gpu/drm/amd/pm/inc/pp_debug.h
index cea65093b6ad..cea65093b6ad 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/pp_debug.h
+++ b/drivers/gpu/drm/amd/pm/inc/pp_debug.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/pp_endian.h b/drivers/gpu/drm/amd/pm/inc/pp_endian.h
index f49d1963fe85..f49d1963fe85 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/pp_endian.h
+++ b/drivers/gpu/drm/amd/pm/inc/pp_endian.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/pp_thermal.h b/drivers/gpu/drm/amd/pm/inc/pp_thermal.h
index 3e30768f9e1c..3e30768f9e1c 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/pp_thermal.h
+++ b/drivers/gpu/drm/amd/pm/inc/pp_thermal.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/ppinterrupt.h b/drivers/gpu/drm/amd/pm/inc/ppinterrupt.h
index c067e0925b6b..c067e0925b6b 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/ppinterrupt.h
+++ b/drivers/gpu/drm/amd/pm/inc/ppinterrupt.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/rv_ppsmc.h
index df4677da736c..df4677da736c 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/rv_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu10.h b/drivers/gpu/drm/amd/pm/inc/smu10.h
index b96520528240..b96520528240 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu10.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu10.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu10_driver_if.h b/drivers/gpu/drm/amd/pm/inc/smu10_driver_if.h
index dea8fe93da63..c498158771cc 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu10_driver_if.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu10_driver_if.h
@@ -54,7 +54,8 @@ typedef struct {
 	uint16_t MaxMclk;
 
 	uint8_t  WmSetting;
-	uint8_t  Padding[3];
+	uint8_t  WmType;
+	uint8_t  Padding[2];
 } WatermarkRowGeneric_t;
 
 #define NUM_WM_RANGES 4
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if.h
index fdc6b7a57bc9..fdc6b7a57bc9 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h
index 43d43d6addc0..43d43d6addc0 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h
index 4b2da98afcd2..246d3951a78a 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h
@@ -885,6 +885,45 @@ typedef struct {
 } SmuMetrics_t;
 
 typedef struct {
+  uint16_t CurrClock[PPCLK_COUNT];
+  uint16_t AverageGfxclkFrequency;
+  uint16_t AverageSocclkFrequency;
+  uint16_t AverageUclkFrequency  ;
+  uint16_t AverageGfxActivity    ;
+  uint16_t AverageUclkActivity   ;
+  uint8_t  CurrSocVoltageOffset  ;
+  uint8_t  CurrGfxVoltageOffset  ;
+  uint8_t  CurrMemVidOffset      ;
+  uint8_t  Padding8              ;
+  uint16_t AverageSocketPower    ;
+  uint16_t TemperatureEdge       ;
+  uint16_t TemperatureHotspot    ;
+  uint16_t TemperatureMem        ;
+  uint16_t TemperatureVrGfx      ;
+  uint16_t TemperatureVrMem0     ;
+  uint16_t TemperatureVrMem1     ;
+  uint16_t TemperatureVrSoc      ;
+  uint16_t TemperatureLiquid0    ;
+  uint16_t TemperatureLiquid1    ;
+  uint16_t TemperaturePlx        ;
+  uint16_t Padding16             ;
+  uint32_t ThrottlerStatus       ;
+
+  uint8_t  LinkDpmLevel;
+  uint8_t  Padding8_2;
+  uint16_t CurrFanSpeed;
+
+  uint32_t EnergyAccumulator;
+  uint16_t AverageVclkFrequency  ;
+  uint16_t AverageDclkFrequency  ;
+  uint16_t VcnActivityPercentage ;
+  uint16_t padding16_2;
+
+  // Padding - ignore
+  uint32_t     MmHubPadding[8]; // SMU internal use
+} SmuMetrics_NV12_t;
+
+typedef struct {
   uint16_t MinClock; // This is either DCEFCLK or SOCCLK (in MHz)
   uint16_t MaxClock; // This is either DCEFCLK or SOCCLK (in MHz)
   uint16_t MinUclk;
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_sienna_cichlid.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_sienna_cichlid.h
index aa2708fccb6d..1275246769d9 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_sienna_cichlid.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_sienna_cichlid.h
@@ -27,9 +27,9 @@
 // *** IMPORTANT ***
 // SMU TEAM: Always increment the interface version if 
 // any structure is changed in this file
-#define SMU11_DRIVER_IF_VERSION 0x34
+#define SMU11_DRIVER_IF_VERSION 0x39
 
-#define PPTABLE_Sienna_Cichlid_SMU_VERSION 5
+#define PPTABLE_Sienna_Cichlid_SMU_VERSION 6
 
 #define NUM_GFXCLK_DPM_LEVELS  16
 #define NUM_SMNCLK_DPM_LEVELS  2
@@ -127,7 +127,7 @@
 #define FEATURE_DF_CSTATE_BIT           45
 #define FEATURE_2_STEP_PSTATE_BIT       46
 #define FEATURE_SMNCLK_DPM_BIT          47
-#define FEATURE_SPARE_48_BIT            48
+#define FEATURE_PERLINK_GMIDOWN_BIT     48
 #define FEATURE_GFX_EDC_BIT             49
 #define FEATURE_SPARE_50_BIT            50
 #define FEATURE_SPARE_51_BIT            51
@@ -793,8 +793,18 @@ typedef struct {
 
   // SECTION: Sku Reserved
   uint8_t          CustomerVariant;
-  uint8_t          Spare[3];
-  uint32_t         SkuReserved[14];
+
+  //VC BTC parameters are only applicable to VDD_GFX domain
+  uint8_t          VcBtcEnabled;
+  uint16_t         VcBtcVminT0;                 // T0_VMIN
+  uint16_t         VcBtcFixedVminAgingOffset;   // FIXED_VMIN_AGING_OFFSET 
+  uint16_t         VcBtcVmin2PsmDegrationGb;    // VMIN_TO_PSM_DEGRADATION_GB 
+  uint32_t         VcBtcPsmA;                   // A_PSM
+  uint32_t         VcBtcPsmB;                   // B_PSM
+  uint32_t         VcBtcVminA;                  // A_VMIN
+  uint32_t         VcBtcVminB;                  // B_VMIN  
+  
+  uint32_t         SkuReserved[9];
 
 
   // MAJOR SECTION: BOARD PARAMETERS
@@ -952,7 +962,7 @@ typedef struct {
   uint8_t                FanLinearPwmPoints[NUM_OD_FAN_MAX_POINTS];
   uint8_t                FanLinearTempPoints[NUM_OD_FAN_MAX_POINTS];
   uint16_t               MaxOpTemp;            // Degree Celcius
-  uint16_t               Padding_16[1];
+  int16_t                VddGfxOffset;         // in mV
   uint8_t                FanZeroRpmEnable;
   uint8_t                FanZeroRpmStopTemp;
   uint8_t                FanMode;
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h b/drivers/gpu/drm/amd/pm/inc/smu12_driver_if.h
index e9315eb5b48e..e9315eb5b48e 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu12_driver_if.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7.h b/drivers/gpu/drm/amd/pm/inc/smu7.h
index e14072d45918..e14072d45918 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu7.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu71.h b/drivers/gpu/drm/amd/pm/inc/smu71.h
index 71c9b2d28640..71c9b2d28640 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu71.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu71.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu71_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu71_discrete.h
index c0e3936d5c2e..c0e3936d5c2e 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu71_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu71_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu72.h b/drivers/gpu/drm/amd/pm/inc/smu72.h
index 9ad1cefff79f..9ad1cefff79f 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu72.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu72.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu72_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu72_discrete.h
index 2aefbb85f620..2aefbb85f620 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu72_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu72_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu73.h b/drivers/gpu/drm/amd/pm/inc/smu73.h
index c6b12a4c00db..c6b12a4c00db 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu73.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu73.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu73_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu73_discrete.h
index 5916be08a7fe..5916be08a7fe 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu73_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu73_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu74.h b/drivers/gpu/drm/amd/pm/inc/smu74.h
index fd10a9fa843d..fd10a9fa843d 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu74.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu74.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu74_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu74_discrete.h
index 899d6d8108c2..899d6d8108c2 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu74_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu74_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu75.h b/drivers/gpu/drm/amd/pm/inc/smu75.h
index 771523001533..771523001533 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu75.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu75.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu75_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu75_discrete.h
index b64e58a22ddf..b64e58a22ddf 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu75_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu75_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h b/drivers/gpu/drm/amd/pm/inc/smu7_common.h
index 94bf7b649c20..94bf7b649c20 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu7_common.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_discrete.h b/drivers/gpu/drm/amd/pm/inc/smu7_discrete.h
index ee876745dd12..ee876745dd12 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7_discrete.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu7_discrete.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_fusion.h b/drivers/gpu/drm/amd/pm/inc/smu7_fusion.h
index 78ada9ffd508..78ada9ffd508 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7_fusion.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu7_fusion.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/smu7_ppsmc.h
index 6e19f4c7cf8f..6e19f4c7cf8f 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu7_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu8.h b/drivers/gpu/drm/amd/pm/inc/smu8.h
index d758d07b6a31..d758d07b6a31 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu8.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu8.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu8_fusion.h b/drivers/gpu/drm/amd/pm/inc/smu8_fusion.h
index 0c37c94e9414..0c37c94e9414 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu8_fusion.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu8_fusion.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu9.h b/drivers/gpu/drm/amd/pm/inc/smu9.h
index 70ac4d477be2..70ac4d477be2 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu9.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu9.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h b/drivers/gpu/drm/amd/pm/inc/smu9_driver_if.h
index 2818c98ff5ca..2818c98ff5ca 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu9_driver_if.h
diff --git a/drivers/gpu/drm/amd/pm/inc/smu_11_0_cdr_table.h b/drivers/gpu/drm/amd/pm/inc/smu_11_0_cdr_table.h
new file mode 100644
index 000000000000..beab6d7b28b7
--- /dev/null
+++ b/drivers/gpu/drm/amd/pm/inc/smu_11_0_cdr_table.h
@@ -0,0 +1,194 @@
+/*
+ * Copyright 2020 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"),
+ * 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.
+ *
+ */
+
+
+#ifndef SMU_11_0_CDR_TABLE
+#define SMU_11_0_CDR_TABLE
+
+
+#pragma pack(push, 1)
+
+/// CDR table : PRBS sequence for DQ toggles
+
+/*static unsigned int NoDbiPrbs7[] =
+{
+//256 bytes, 256 byte aligned
+0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+};
+
+
+static unsigned int DbiPrbs7[] =
+{
+// 256 bytes, 256 byte aligned
+0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+};
+*/
+
+
+//4096 bytes, 256 byte aligned
+static unsigned int NoDbiPrbs7[] =
+{
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+    0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f,
+    0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f,
+    0xf0f00f0f, 0x0f0ff0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0xf0f0ffff,
+};
+
+// 4096 bytes, 256 byte aligned
+static unsigned int DbiPrbs7[] =
+{
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+    0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff,
+    0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff,
+    0x0000ffff, 0xffff0000, 0xffff0000, 0x00000000, 0xffff0000, 0x00000000, 0xffffffff, 0x00000000, 0xffffffff, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0x00000000, 0x00000000, 0x00000000, 0x0000ffff,
+};
+
+#pragma pack(pop)
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h b/drivers/gpu/drm/amd/pm/inc/smu_types.h
index 7b585e205a5a..35fc46d3c9c0 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_types.h
@@ -173,6 +173,9 @@
 	__SMU_DUMMY_MAP(GmiPwrDnControl), \
 	__SMU_DUMMY_MAP(DAL_DISABLE_DUMMY_PSTATE_CHANGE), \
 	__SMU_DUMMY_MAP(DAL_ENABLE_DUMMY_PSTATE_CHANGE), \
+	__SMU_DUMMY_MAP(SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_HIGH), \
+	__SMU_DUMMY_MAP(SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_LOW), \
+	__SMU_DUMMY_MAP(GET_UMC_FW_WA), \
 	__SMU_DUMMY_MAP(Mode1Reset), \
 
 #undef __SMU_DUMMY_MAP
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_ucode_xfer_cz.h b/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_cz.h
index eb0f79f9c876..eb0f79f9c876 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_ucode_xfer_cz.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_cz.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_ucode_xfer_vi.h b/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_vi.h
index 880152c0f775..880152c0f775 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_ucode_xfer_vi.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_vi.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/pm/inc/smu_v11_0.h
index 6a42331aba8a..2d1c3babaa3a 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v11_0.h
@@ -28,10 +28,10 @@
 #define SMU11_DRIVER_IF_VERSION_INV 0xFFFFFFFF
 #define SMU11_DRIVER_IF_VERSION_ARCT 0x17
 #define SMU11_DRIVER_IF_VERSION_NV10 0x36
-#define SMU11_DRIVER_IF_VERSION_NV12 0x33
+#define SMU11_DRIVER_IF_VERSION_NV12 0x36
 #define SMU11_DRIVER_IF_VERSION_NV14 0x36
-#define SMU11_DRIVER_IF_VERSION_Sienna_Cichlid 0x34
-#define SMU11_DRIVER_IF_VERSION_Navy_Flounder 0x3
+#define SMU11_DRIVER_IF_VERSION_Sienna_Cichlid 0x39
+#define SMU11_DRIVER_IF_VERSION_Navy_Flounder 0x5
 
 /* MP Apertures */
 #define MP0_Public			0x03800000
@@ -200,12 +200,12 @@ int
 smu_v11_0_set_fan_control_mode(struct smu_context *smu,
 			       uint32_t mode);
 
-int
-smu_v11_0_set_fan_speed_percent(struct smu_context *smu, uint32_t speed);
-
 int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu,
 				       uint32_t speed);
 
+int smu_v11_0_get_fan_speed_rpm(struct smu_context *smu,
+				uint32_t *speed);
+
 int smu_v11_0_set_xgmi_pstate(struct smu_context *smu,
 				     uint32_t pstate);
 
@@ -264,5 +264,23 @@ int smu_v11_0_get_dpm_level_range(struct smu_context *smu,
 				  uint32_t *min_value,
 				  uint32_t *max_value);
 
+int smu_v11_0_get_current_pcie_link_width_level(struct smu_context *smu);
+
+int smu_v11_0_get_current_pcie_link_width(struct smu_context *smu);
+
+int smu_v11_0_get_current_pcie_link_speed_level(struct smu_context *smu);
+
+int smu_v11_0_get_current_pcie_link_speed(struct smu_context *smu);
+
+void smu_v11_0_init_gpu_metrics_v1_0(struct gpu_metrics_v1_0 *gpu_metrics);
+
+int smu_v11_0_gfx_ulv_control(struct smu_context *smu,
+			      bool enablement);
+
+int smu_v11_0_deep_sleep_control(struct smu_context *smu,
+				 bool enablement);
+
+void smu_v11_0_interrupt_work(struct smu_context *smu);
+
 #endif
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_7_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_ppsmc.h
index 35dd6072cc45..35dd6072cc45 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_7_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_7_pptable.h b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_pptable.h
index 247c6e9632ba..247c6e9632ba 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_7_pptable.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_pptable.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_ppsmc.h
index 406bfd187ce8..26181b679098 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_ppsmc.h
@@ -123,7 +123,14 @@
 #define PPSMC_MSG_DALDisableDummyPstateChange    0x49
 #define PPSMC_MSG_DALEnableDummyPstateChange     0x4A
 
-#define PPSMC_Message_Count                      0x4B
+#define PPSMC_MSG_SetMGpuFanBoostLimitRpm        0x4C
+
+#define PPSMC_MSG_SetDriverDummyTableDramAddrHigh 0x4E
+#define PPSMC_MSG_SetDriverDummyTableDramAddrLow  0x4F
+
+#define PPSMC_MSG_GetUMCFWWA                     0x50
+
+#define PPSMC_Message_Count                      0x51
 
 typedef uint32_t PPSMC_Result;
 typedef uint32_t PPSMC_Msg;
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_pptable.h b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_pptable.h
index 7a63cf8e85ed..7a63cf8e85ed 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_pptable.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v11_0_pptable.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h b/drivers/gpu/drm/amd/pm/inc/smu_v12_0.h
index 02de3b6199e5..fa2e8cb07967 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v12_0.h
@@ -60,5 +60,7 @@ int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_
 
 int smu_v12_0_set_driver_table_location(struct smu_context *smu);
 
+void smu_v12_0_init_gpu_metrics_v2_0(struct gpu_metrics_v2_0 *gpu_metrics);
+
 #endif
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/smu_v12_0_ppsmc.h
index 9ac9f3bd3664..9ac9f3bd3664 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v12_0_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h b/drivers/gpu/drm/amd/pm/inc/smumgr.h
index ad100b533d04..ad100b533d04 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h
+++ b/drivers/gpu/drm/amd/pm/inc/smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/tonga_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/tonga_ppsmc.h
index 63631296d751..63631296d751 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/tonga_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/tonga_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega10_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/vega10_ppsmc.h
index 715b5a168831..715b5a168831 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega10_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/vega10_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega12/smu9_driver_if.h b/drivers/gpu/drm/amd/pm/inc/vega12/smu9_driver_if.h
index b6ffd08784e7..b6ffd08784e7 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega12/smu9_driver_if.h
+++ b/drivers/gpu/drm/amd/pm/inc/vega12/smu9_driver_if.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega12_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/vega12_ppsmc.h
index f985c78d746a..f985c78d746a 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega12_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/vega12_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/pm/inc/vega20_ppsmc.h
index 0c66f0fe1aaf..0c66f0fe1aaf 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/inc/vega20_ppsmc.h
diff --git a/drivers/gpu/drm/amd/powerplay/Makefile b/drivers/gpu/drm/amd/pm/powerplay/Makefile
index e9c48f99f71b..0fb114adc79f 100644
--- a/drivers/gpu/drm/amd/powerplay/Makefile
+++ b/drivers/gpu/drm/amd/pm/powerplay/Makefile
@@ -1,5 +1,5 @@
 #
-# Copyright 2017 Advanced Micro Devices, Inc.
+# Copyright 2020 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"),
@@ -20,25 +20,20 @@
 # OTHER DEALINGS IN THE SOFTWARE.
 #
 
-subdir-ccflags-y += \
-		-I$(FULL_AMD_PATH)/powerplay/inc/  \
-		-I$(FULL_AMD_PATH)/include/asic_reg  \
-		-I$(FULL_AMD_PATH)/include  \
-		-I$(FULL_AMD_PATH)/powerplay/smumgr\
-		-I$(FULL_AMD_PATH)/powerplay/hwmgr
-
-AMD_PP_PATH = ../powerplay
+AMD_PP_PATH = ../pm/powerplay
 
 PP_LIBS = smumgr hwmgr
 
-AMD_POWERPLAY = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/powerplay/,$(PP_LIBS)))
+AMD_POWERPLAY = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/pm/powerplay/,$(PP_LIBS)))
 
 include $(AMD_POWERPLAY)
 
-POWER_MGR = amd_powerplay.o amdgpu_smu.o smu_v11_0.o \
-	    smu_v12_0.o arcturus_ppt.o navi10_ppt.o \
-	    renoir_ppt.o sienna_cichlid_ppt.o smu_cmn.o
+POWER_MGR-y = amd_powerplay.o
+
+POWER_MGR-$(CONFIG_DRM_AMDGPU_CIK)+= kv_dpm.o kv_smc.o
+
+POWER_MGR-$(CONFIG_DRM_AMDGPU_SI)+= si_dpm.o si_smc.o
 
-AMD_PP_POWER = $(addprefix $(AMD_PP_PATH)/,$(POWER_MGR))
+AMD_PP_POWER = $(addprefix $(AMD_PP_PATH)/,$(POWER_MGR-y))
 
 AMD_POWERPLAY_FILES += $(AMD_PP_POWER)
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
index 7e6dcdf7df73..eab9768029c1 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
@@ -911,6 +911,19 @@ static int pp_set_power_profile_mode(void *handle, long *input, uint32_t size)
 	return ret;
 }
 
+static int pp_set_fine_grain_clk_vol(void *handle, uint32_t type, long *input, uint32_t size)
+{
+	struct pp_hwmgr *hwmgr = handle;
+
+	if (!hwmgr || !hwmgr->pm_en)
+		return -EINVAL;
+
+	if (hwmgr->hwmgr_func->set_fine_grain_clk_vol == NULL)
+		return 0;
+
+	return hwmgr->hwmgr_func->set_fine_grain_clk_vol(hwmgr, type, input, size);
+}
+
 static int pp_odn_edit_dpm_table(void *handle, uint32_t type, long *input, uint32_t size)
 {
 	struct pp_hwmgr *hwmgr = handle;
@@ -920,7 +933,7 @@ static int pp_odn_edit_dpm_table(void *handle, uint32_t type, long *input, uint3
 
 	if (hwmgr->hwmgr_func->odn_edit_dpm_table == NULL) {
 		pr_info_ratelimited("%s was not implemented.\n", __func__);
-		return -EINVAL;
+		return 0;
 	}
 
 	return hwmgr->hwmgr_func->odn_edit_dpm_table(hwmgr, type, input, size);
@@ -1598,6 +1611,24 @@ static int pp_set_xgmi_pstate(void *handle, uint32_t pstate)
 	return 0;
 }
 
+static ssize_t pp_get_gpu_metrics(void *handle, void **table)
+{
+	struct pp_hwmgr *hwmgr = handle;
+	ssize_t size;
+
+	if (!hwmgr)
+		return -EINVAL;
+
+	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_gpu_metrics)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&hwmgr->smu_lock);
+	size = hwmgr->hwmgr_func->get_gpu_metrics(hwmgr, table);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return size;
+}
+
 static const struct amd_pm_funcs pp_dpm_funcs = {
 	.load_firmware = pp_dpm_load_fw,
 	.wait_for_fw_loading_complete = pp_dpm_fw_loading_complete,
@@ -1627,6 +1658,7 @@ static const struct amd_pm_funcs pp_dpm_funcs = {
 	.set_powergating_by_smu = pp_set_powergating_by_smu,
 	.get_power_profile_mode = pp_get_power_profile_mode,
 	.set_power_profile_mode = pp_set_power_profile_mode,
+	.set_fine_grain_clk_vol = pp_set_fine_grain_clk_vol,
 	.odn_edit_dpm_table = pp_odn_edit_dpm_table,
 	.set_mp1_state = pp_dpm_set_mp1_state,
 	.set_power_limit = pp_set_power_limit,
@@ -1658,4 +1690,5 @@ static const struct amd_pm_funcs pp_dpm_funcs = {
 	.smu_i2c_bus_access = pp_smu_i2c_bus_access,
 	.set_df_cstate = pp_set_df_cstate,
 	.set_xgmi_pstate = pp_set_xgmi_pstate,
+	.get_gpu_metrics = pp_get_gpu_metrics,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_dpm.h b/drivers/gpu/drm/amd/pm/powerplay/cik_dpm.h
index 2fcc4b60153c..2fcc4b60153c 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_dpm.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/cik_dpm.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/Makefile
index 2773966ae434..2773966ae434 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/Makefile
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ci_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c
index 3be40114e63d..3be40114e63d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ci_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ci_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.h
index 17041f187020..17041f187020 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ci_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c
index 1c73776bd606..1c73776bd606 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.h
index 8393eb62706d..8393eb62706d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c
index c0368f2dfb21..c0368f2dfb21 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.h
index 47f402900bdb..47f402900bdb 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hardwaremanager.c
index 9454ab50f9a1..1f9b9facdf1f 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hardwaremanager.c
@@ -271,7 +271,10 @@ int phm_start_thermal_controller(struct pp_hwmgr *hwmgr)
 
 bool phm_check_smc_update_required_for_display_configuration(struct pp_hwmgr *hwmgr)
 {
-	PHM_FUNC_CHECK(hwmgr);
+	if (hwmgr == NULL ||
+	    hwmgr->hwmgr_func == NULL)
+		return false;
+
 	if (hwmgr->pp_one_vf)
 		return false;
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c
index f48fdc7f0382..f48fdc7f0382 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr_ppt.h
index c0193e09d58a..c0193e09d58a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr_ppt.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c
index 8f8e296f2fe9..8f8e296f2fe9 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.h
index 87a5fa0a157a..87a5fa0a157a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_overdriver.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_overdriver.c
index 8de384bf9a8f..8de384bf9a8f 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_overdriver.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_overdriver.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_overdriver.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_overdriver.h
index 4112a9398163..4112a9398163 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_overdriver.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_overdriver.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_psm.c
index 31a32a79cfc2..31a32a79cfc2 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_psm.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_psm.h
index b62d55f1f289..b62d55f1f289 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pp_psm.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
index 01dc46dc9c8a..01dc46dc9c8a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h
index 3ee54f182943..76ed2e413594 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomctrl.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h
@@ -26,15 +26,6 @@
 
 #include "hwmgr.h"
 
-#define MEM_TYPE_GDDR5  0x50
-#define MEM_TYPE_GDDR4  0x40
-#define MEM_TYPE_GDDR3  0x30
-#define MEM_TYPE_DDR2   0x20
-#define MEM_TYPE_GDDR1  0x10
-#define MEM_TYPE_DDR3   0xb0
-#define MEM_TYPE_MASK   0xF0
-
-
 /* As returned from PowerConnectorDetectionTable. */
 #define PP_ATOM_POWER_BUDGET_DISABLE_OVERDRIVE  0x80
 #define PP_ATOM_POWER_BUDGET_SHOW_WARNING       0x40
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c
index 615cf2c09e54..615cf2c09e54 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h
index b7e2651b570b..b7e2651b570b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/ppevvmath.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h
index 8f50a038396c..8f50a038396c 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/ppevvmath.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pppcielanes.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pppcielanes.c
index 186496a34cbe..186496a34cbe 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pppcielanes.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pppcielanes.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pppcielanes.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pppcielanes.h
index 70b163b35570..70b163b35570 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pppcielanes.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pppcielanes.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pptable_v1_0.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h
index 1e870f58dd12..1e870f58dd12 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pptable_v1_0.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/process_pptables_v1_0.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c
index b760f95e7fa7..b760f95e7fa7 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/process_pptables_v1_0.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/process_pptables_v1_0.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.h
index b9710abdff01..b9710abdff01 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/process_pptables_v1_0.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c
index 719597c5d27d..719597c5d27d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.h
index baddaa75693b..baddaa75693b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
index 43f7adff6cb7..cf60f3992303 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
@@ -242,6 +242,34 @@ static int smu10_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t cloc
 	return 0;
 }
 
+static int smu10_set_hard_min_gfxclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock)
+{
+	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
+
+	if (clock && smu10_data->gfx_actual_soft_min_freq != clock) {
+		smu10_data->gfx_actual_soft_min_freq = clock;
+		smum_send_msg_to_smc_with_parameter(hwmgr,
+					PPSMC_MSG_SetHardMinGfxClk,
+					smu10_data->gfx_actual_soft_min_freq,
+					NULL);
+	}
+	return 0;
+}
+
+static int smu10_set_soft_max_gfxclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock)
+{
+	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
+
+	if (clock && smu10_data->gfx_max_freq_limit != (clock * 100))  {
+		smu10_data->gfx_max_freq_limit = clock * 100;
+		smum_send_msg_to_smc_with_parameter(hwmgr,
+					PPSMC_MSG_SetSoftMaxGfxClk,
+					clock,
+					NULL);
+	}
+	return 0;
+}
+
 static int smu10_set_active_display_count(struct pp_hwmgr *hwmgr, uint32_t count)
 {
 	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
@@ -527,6 +555,9 @@ static int smu10_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 	hwmgr->pstate_sclk = SMU10_UMD_PSTATE_GFXCLK * 100;
 	hwmgr->pstate_mclk = SMU10_UMD_PSTATE_FCLK * 100;
 
+	/* enable the pp_od_clk_voltage sysfs file */
+	hwmgr->od_enabled = 1;
+
 	return result;
 }
 
@@ -650,7 +681,7 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 						NULL);
 		smum_send_msg_to_smc_with_parameter(hwmgr,
 						PPSMC_MSG_SetHardMinVcn,
-						SMU10_UMD_PSTATE_VCE,
+						SMU10_UMD_PSTATE_PROFILE_VCE,
 						NULL);
 
 		smum_send_msg_to_smc_with_parameter(hwmgr,
@@ -667,7 +698,7 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 						NULL);
 		smum_send_msg_to_smc_with_parameter(hwmgr,
 						PPSMC_MSG_SetSoftMaxVcn,
-						SMU10_UMD_PSTATE_VCE,
+						SMU10_UMD_PSTATE_PROFILE_VCE,
 						NULL);
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
@@ -949,6 +980,26 @@ static int smu10_print_clock_levels(struct pp_hwmgr *hwmgr,
 					((mclk_table->entries[i].clk / 100)
 					 == now) ? "*" : "");
 		break;
+	case OD_SCLK:
+		if (hwmgr->od_enabled) {
+			size = sprintf(buf, "%s:\n", "OD_SCLK");
+
+			size += sprintf(buf + size, "0: %10uMhz\n",
+			(data->gfx_actual_soft_min_freq > 0) ? data->gfx_actual_soft_min_freq : data->gfx_min_freq_limit/100);
+			size += sprintf(buf + size, "1: %10uMhz\n", data->gfx_max_freq_limit/100);
+		}
+		break;
+	case OD_RANGE:
+		if (hwmgr->od_enabled) {
+			uint32_t min_freq, max_freq = 0;
+			smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMinGfxclkFrequency, &min_freq);
+			smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxGfxclkFrequency, &max_freq);
+
+			size = sprintf(buf, "%s:\n", "OD_RANGE");
+			size += sprintf(buf + size, "SCLK: %7uMHz %10uMHz\n",
+				min_freq, max_freq);
+		}
+		break;
 	default:
 		break;
 	}
@@ -1183,8 +1234,19 @@ static int smu10_set_watermarks_for_clocks_ranges(struct pp_hwmgr *hwmgr,
 	struct smu10_hwmgr *data = hwmgr->backend;
 	struct dm_pp_wm_sets_with_clock_ranges_soc15 *wm_with_clock_ranges = clock_ranges;
 	Watermarks_t *table = &(data->water_marks_table);
+	struct amdgpu_device *adev = hwmgr->adev;
+	int i;
 
 	smu_set_watermarks_for_clocks_ranges(table,wm_with_clock_ranges);
+
+	if (adev->apu_flags & AMD_APU_IS_RAVEN2) {
+		for (i = 0; i < NUM_WM_RANGES; i++)
+			table->WatermarkRow[WM_DCFCLK][i].WmType = (uint8_t)0;
+
+		for (i = 0; i < NUM_WM_RANGES; i++)
+			table->WatermarkRow[WM_SOCCLK][i].WmType = (uint8_t)0;
+	}
+
 	smum_smc_table_manager(hwmgr, (uint8_t *)table, (uint16_t)SMU10_WMTABLE, false);
 	data->water_marks_exist = true;
 	return 0;
@@ -1350,6 +1412,32 @@ static int smu10_asic_reset(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mod
 						   NULL);
 }
 
+static int smu10_set_fine_grain_clk_vol(struct pp_hwmgr *hwmgr,
+					enum PP_OD_DPM_TABLE_COMMAND type,
+					long *input, uint32_t size)
+{
+	if (!hwmgr->od_enabled) {
+		pr_err("Fine grain not support\n");
+		return -EINVAL;
+	}
+
+	if (size != 2) {
+		pr_err("Input parameter number not correct\n");
+		return -EINVAL;
+	}
+
+	if (type == PP_OD_EDIT_SCLK_VDDC_TABLE) {
+		if (input[0] == 0)
+			smu10_set_hard_min_gfxclk_by_freq(hwmgr, input[1]);
+		else if (input[0] == 1)
+			smu10_set_soft_max_gfxclk_by_freq(hwmgr, input[1]);
+		else
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
 static const struct pp_hwmgr_func smu10_hwmgr_funcs = {
 	.backend_init = smu10_hwmgr_backend_init,
 	.backend_fini = smu10_hwmgr_backend_fini,
@@ -1390,9 +1478,12 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = {
 	.powergate_sdma = smu10_powergate_sdma,
 	.set_hard_min_dcefclk_by_freq = smu10_set_hard_min_dcefclk_by_freq,
 	.set_hard_min_fclk_by_freq = smu10_set_hard_min_fclk_by_freq,
+	.set_hard_min_gfxclk_by_freq = smu10_set_hard_min_gfxclk_by_freq,
+	.set_soft_max_gfxclk_by_freq = smu10_set_soft_max_gfxclk_by_freq,
 	.get_power_profile_mode = smu10_get_power_profile_mode,
 	.set_power_profile_mode = smu10_set_power_profile_mode,
 	.asic_reset = smu10_asic_reset,
+	.set_fine_grain_clk_vol = smu10_set_fine_grain_clk_vol,
 };
 
 int smu10_init_function_pointers(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.h
index 0f969de10fab..6c9b5f060902 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.h
@@ -284,7 +284,7 @@ struct smu10_hwmgr {
 	uint32_t                        dclk_soft_min;
 	uint32_t                        gfx_actual_soft_min_freq;
 	uint32_t                        gfx_min_freq_limit;
-	uint32_t                        gfx_max_freq_limit;
+	uint32_t                        gfx_max_freq_limit; /* in 10Khz*/
 
 	bool                           vcn_power_gated;
 	bool                           vcn_dpg_mode;
@@ -310,6 +310,7 @@ int smu10_init_function_pointers(struct pp_hwmgr *hwmgr);
 #define SMU10_UMD_PSTATE_SOCCLK                 626
 #define SMU10_UMD_PSTATE_FCLK                   933
 #define SMU10_UMD_PSTATE_VCE                    0x03C00320
+#define SMU10_UMD_PSTATE_PROFILE_VCE            0x02AD0229
 
 #define SMU10_UMD_PSTATE_PEAK_SOCCLK            757
 #define SMU10_UMD_PSTATE_PEAK_FCLK              1200
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_inc.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_inc.h
index edb68e302f6f..edb68e302f6f 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_inc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_inc.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_baco.c
index 044cda005aed..044cda005aed 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_baco.h
index be0d98abb536..be0d98abb536 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_clockpowergating.c
index f2bda3bcbbde..f2bda3bcbbde 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_clockpowergating.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_clockpowergating.h
index fc8f8a6acc72..fc8f8a6acc72 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_clockpowergating.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_dyn_defaults.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_dyn_defaults.h
index 3477d4dfff70..3477d4dfff70 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_dyn_defaults.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_dyn_defaults.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c
index 4a3b64aa21ce..3bf8be4d107b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c
@@ -1585,9 +1585,19 @@ static void smu7_init_dpm_defaults(struct pp_hwmgr *hwmgr)
 	data->current_profile_setting.sclk_down_hyst = 100;
 	data->current_profile_setting.sclk_activity = SMU7_SCLK_TARGETACTIVITY_DFLT;
 	data->current_profile_setting.bupdate_mclk = 1;
-	data->current_profile_setting.mclk_up_hyst = 0;
-	data->current_profile_setting.mclk_down_hyst = 100;
-	data->current_profile_setting.mclk_activity = SMU7_MCLK_TARGETACTIVITY_DFLT;
+	if (adev->gmc.vram_width == 256) {
+		data->current_profile_setting.mclk_up_hyst = 10;
+		data->current_profile_setting.mclk_down_hyst = 60;
+		data->current_profile_setting.mclk_activity = 25;
+	} else if (adev->gmc.vram_width == 128) {
+		data->current_profile_setting.mclk_up_hyst = 5;
+		data->current_profile_setting.mclk_down_hyst = 16;
+		data->current_profile_setting.mclk_activity = 20;
+	} else if (adev->gmc.vram_width == 64) {
+		data->current_profile_setting.mclk_up_hyst = 3;
+		data->current_profile_setting.mclk_down_hyst = 16;
+		data->current_profile_setting.mclk_activity = 20;
+	}
 	hwmgr->workload_mask = 1 << hwmgr->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D];
 	hwmgr->power_profile_mode = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
 	hwmgr->default_power_profile_mode = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.h
index 69d361f8dfca..69d361f8dfca 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c
index 5d4971576111..5d4971576111 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.h
index 22f86b6bf1be..22f86b6bf1be 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_thermal.c
index 0b30f73649a8..0b30f73649a8 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_thermal.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_thermal.h
index 42c1ba0fad78..42c1ba0fad78 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_thermal.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c
index 35ed47ebaf09..35ed47ebaf09 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.h
index 05a06083e1b8..05a06083e1b8 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu9_baco.c
index de0a37f7c632..de0a37f7c632 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu9_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu9_baco.h
index 84e90f801ac3..84e90f801ac3 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu9_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c
index 60b5ca974356..60b5ca974356 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h
index ad33983a8064..ad33983a8064 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c
index ea743bea8e29..ea743bea8e29 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.h
index 5dc16cc8a295..5dc16cc8a295 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c
index 46bb16c29cf6..46bb16c29cf6 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.h
index 96d793f026a5..96d793f026a5 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c
index c378a000c934..7eada3098ffc 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c
@@ -4659,7 +4659,7 @@ static int vega10_display_configuration_changed_task(struct pp_hwmgr *hwmgr)
 	if ((data->water_marks_bitmap & WaterMarksExist) &&
 			!(data->water_marks_bitmap & WaterMarksLoaded)) {
 		result = smum_smc_table_manager(hwmgr, (uint8_t *)wm_table, WMTABLE, false);
-		PP_ASSERT_WITH_CODE(result, "Failed to update WMTABLE!", return EINVAL);
+		PP_ASSERT_WITH_CODE(result, "Failed to update WMTABLE!", return -EINVAL);
 		data->water_marks_bitmap |= WaterMarksLoaded;
 	}
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.h
index f752b4ad0c8a..f752b4ad0c8a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_inc.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_inc.h
index faf7ac044348..faf7ac044348 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_inc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_inc.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c
index 9757d47dd6b8..9757d47dd6b8 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.h
index b95771ab89cd..b95771ab89cd 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h
index c934e9612c1b..c934e9612c1b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c
index f29af5ca0aa0..f29af5ca0aa0 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.h
index da5fbec9b0cd..da5fbec9b0cd 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_thermal.c
index 952cd3d7240e..952cd3d7240e 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_thermal.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_thermal.h
index 4a0ede7c1f07..4a0ede7c1f07 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_thermal.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c
index bc53cce4f32d..bc53cce4f32d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.h
index 57b72e5a95ae..57b72e5a95ae 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c
index a678a67f1c0d..dc206fa88c5e 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c
@@ -47,6 +47,13 @@
 #include "pp_thermal.h"
 #include "vega12_baco.h"
 
+#define smnPCIE_LC_SPEED_CNTL			0x11140290
+#define smnPCIE_LC_LINK_WIDTH_CNTL		0x11140288
+
+#define LINK_WIDTH_MAX				6
+#define LINK_SPEED_MAX				3
+static int link_width[] = {0, 1, 2, 4, 8, 12, 16};
+static int link_speed[] = {25, 50, 80, 160};
 
 static int vega12_force_clock_level(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, uint32_t mask);
@@ -1255,22 +1262,29 @@ static uint32_t vega12_dpm_get_mclk(struct pp_hwmgr *hwmgr, bool low)
 	return (mem_clk * 100);
 }
 
-static int vega12_get_metrics_table(struct pp_hwmgr *hwmgr, SmuMetrics_t *metrics_table)
+static int vega12_get_metrics_table(struct pp_hwmgr *hwmgr,
+				    SmuMetrics_t *metrics_table,
+				    bool bypass_cache)
 {
 	struct vega12_hwmgr *data =
 			(struct vega12_hwmgr *)(hwmgr->backend);
 	int ret = 0;
 
-	if (!data->metrics_time || time_after(jiffies, data->metrics_time + HZ / 2)) {
-		ret = smum_smc_table_manager(hwmgr, (uint8_t *)metrics_table,
-				TABLE_SMU_METRICS, true);
+	if (bypass_cache ||
+	    !data->metrics_time ||
+	    time_after(jiffies, data->metrics_time + msecs_to_jiffies(1))) {
+		ret = smum_smc_table_manager(hwmgr,
+					     (uint8_t *)(&data->metrics_table),
+					     TABLE_SMU_METRICS,
+					     true);
 		if (ret) {
 			pr_info("Failed to export SMU metrics table!\n");
 			return ret;
 		}
-		memcpy(&data->metrics_table, metrics_table, sizeof(SmuMetrics_t));
 		data->metrics_time = jiffies;
-	} else
+	}
+
+	if (metrics_table)
 		memcpy(metrics_table, &data->metrics_table, sizeof(SmuMetrics_t));
 
 	return ret;
@@ -1281,7 +1295,7 @@ static int vega12_get_gpu_power(struct pp_hwmgr *hwmgr, uint32_t *query)
 	SmuMetrics_t metrics_table;
 	int ret = 0;
 
-	ret = vega12_get_metrics_table(hwmgr, &metrics_table);
+	ret = vega12_get_metrics_table(hwmgr, &metrics_table, false);
 	if (ret)
 		return ret;
 
@@ -1332,7 +1346,7 @@ static int vega12_get_current_activity_percent(
 	SmuMetrics_t metrics_table;
 	int ret = 0;
 
-	ret = vega12_get_metrics_table(hwmgr, &metrics_table);
+	ret = vega12_get_metrics_table(hwmgr, &metrics_table, false);
 	if (ret)
 		return ret;
 
@@ -1380,7 +1394,7 @@ static int vega12_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 		*size = 4;
 		break;
 	case AMDGPU_PP_SENSOR_HOTSPOT_TEMP:
-		ret = vega12_get_metrics_table(hwmgr, &metrics_table);
+		ret = vega12_get_metrics_table(hwmgr, &metrics_table, false);
 		if (ret)
 			return ret;
 
@@ -1389,7 +1403,7 @@ static int vega12_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 		*size = 4;
 		break;
 	case AMDGPU_PP_SENSOR_MEM_TEMP:
-		ret = vega12_get_metrics_table(hwmgr, &metrics_table);
+		ret = vega12_get_metrics_table(hwmgr, &metrics_table, false);
 		if (ret)
 			return ret;
 
@@ -2095,6 +2109,46 @@ static int vega12_set_ppfeature_status(struct pp_hwmgr *hwmgr, uint64_t new_ppfe
 	return 0;
 }
 
+static int vega12_get_current_pcie_link_width_level(struct pp_hwmgr *hwmgr)
+{
+	struct amdgpu_device *adev = hwmgr->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
+		PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
+		>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+}
+
+static int vega12_get_current_pcie_link_width(struct pp_hwmgr *hwmgr)
+{
+	uint32_t width_level;
+
+	width_level = vega12_get_current_pcie_link_width_level(hwmgr);
+	if (width_level > LINK_WIDTH_MAX)
+		width_level = 0;
+
+	return link_width[width_level];
+}
+
+static int vega12_get_current_pcie_link_speed_level(struct pp_hwmgr *hwmgr)
+{
+	struct amdgpu_device *adev = hwmgr->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
+		PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
+		>> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
+}
+
+static int vega12_get_current_pcie_link_speed(struct pp_hwmgr *hwmgr)
+{
+	uint32_t speed_level;
+
+	speed_level = vega12_get_current_pcie_link_speed_level(hwmgr);
+	if (speed_level > LINK_SPEED_MAX)
+		speed_level = 0;
+
+	return link_speed[speed_level];
+}
+
 static int vega12_print_clock_levels(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, char *buf)
 {
@@ -2390,7 +2444,7 @@ static int vega12_display_configuration_changed_task(struct pp_hwmgr *hwmgr)
 			!(data->water_marks_bitmap & WaterMarksLoaded)) {
 		result = smum_smc_table_manager(hwmgr,
 						(uint8_t *)wm_table, TABLE_WATERMARKS, false);
-		PP_ASSERT_WITH_CODE(result, "Failed to update WMTABLE!", return EINVAL);
+		PP_ASSERT_WITH_CODE(result, "Failed to update WMTABLE!", return -EINVAL);
 		data->water_marks_bitmap |= WaterMarksLoaded;
 	}
 
@@ -2682,6 +2736,69 @@ static int vega12_set_mp1_state(struct pp_hwmgr *hwmgr,
 	return 0;
 }
 
+static void vega12_init_gpu_metrics_v1_0(struct gpu_metrics_v1_0 *gpu_metrics)
+{
+	memset(gpu_metrics, 0xFF, sizeof(struct gpu_metrics_v1_0));
+
+	gpu_metrics->common_header.structure_size =
+				sizeof(struct gpu_metrics_v1_0);
+	gpu_metrics->common_header.format_revision = 1;
+	gpu_metrics->common_header.content_revision = 0;
+
+	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
+}
+
+static ssize_t vega12_get_gpu_metrics(struct pp_hwmgr *hwmgr,
+				      void **table)
+{
+	struct vega12_hwmgr *data =
+			(struct vega12_hwmgr *)(hwmgr->backend);
+	struct gpu_metrics_v1_0 *gpu_metrics =
+			&data->gpu_metrics_table;
+	SmuMetrics_t metrics;
+	uint32_t fan_speed_rpm;
+	int ret;
+
+	ret = vega12_get_metrics_table(hwmgr, &metrics, true);
+	if (ret)
+		return ret;
+
+	vega12_init_gpu_metrics_v1_0(gpu_metrics);
+
+	gpu_metrics->temperature_edge = metrics.TemperatureEdge;
+	gpu_metrics->temperature_hotspot = metrics.TemperatureHotspot;
+	gpu_metrics->temperature_mem = metrics.TemperatureHBM;
+	gpu_metrics->temperature_vrgfx = metrics.TemperatureVrGfx;
+	gpu_metrics->temperature_vrmem = metrics.TemperatureVrMem;
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_umc_activity = metrics.AverageUclkActivity;
+
+	gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequency;
+	gpu_metrics->average_socclk_frequency = metrics.AverageSocclkFrequency;
+	gpu_metrics->average_uclk_frequency = metrics.AverageUclkFrequency;
+
+	gpu_metrics->current_gfxclk = metrics.CurrClock[PPCLK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.CurrClock[PPCLK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.CurrClock[PPCLK_UCLK];
+	gpu_metrics->current_vclk0 = metrics.CurrClock[PPCLK_VCLK];
+	gpu_metrics->current_dclk0 = metrics.CurrClock[PPCLK_DCLK];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	vega12_fan_ctrl_get_fan_speed_rpm(hwmgr, &fan_speed_rpm);
+	gpu_metrics->current_fan_speed = (uint16_t)fan_speed_rpm;
+
+	gpu_metrics->pcie_link_width =
+			vega12_get_current_pcie_link_width(hwmgr);
+	gpu_metrics->pcie_link_speed =
+			vega12_get_current_pcie_link_speed(hwmgr);
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v1_0);
+}
+
 static const struct pp_hwmgr_func vega12_hwmgr_funcs = {
 	.backend_init = vega12_hwmgr_backend_init,
 	.backend_fini = vega12_hwmgr_backend_fini,
@@ -2739,6 +2856,7 @@ static const struct pp_hwmgr_func vega12_hwmgr_funcs = {
 	.get_ppfeature_status = vega12_get_ppfeature_status,
 	.set_ppfeature_status = vega12_set_ppfeature_status,
 	.set_mp1_state = vega12_set_mp1_state,
+	.get_gpu_metrics = vega12_get_gpu_metrics,
 };
 
 int vega12_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h
index 73875399666a..aa63ae41942d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h
@@ -399,6 +399,7 @@ struct vega12_hwmgr {
 
 	unsigned long                  metrics_time;
 	SmuMetrics_t                   metrics_table;
+	struct gpu_metrics_v1_0        gpu_metrics_table;
 };
 
 #define VEGA12_DPM2_NEAR_TDP_DEC                      10
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_inc.h
index e6d9e84059e1..0d08c57d3bca 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_inc.h
@@ -35,7 +35,6 @@
 #include "asic_reg/gc/gc_9_2_1_sh_mask.h"
 
 #include "asic_reg/nbio/nbio_6_1_offset.h"
-#include "asic_reg/nbio/nbio_6_1_offset.h"
 #include "asic_reg/nbio/nbio_6_1_sh_mask.h"
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h
index bf4f5095b80d..bf4f5095b80d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_pptable.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.c
index 195d8539fbb4..740e2fc7a034 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.c
@@ -252,7 +252,7 @@ static int init_powerplay_table_information(
 	phm_copy_clock_limits_array(hwmgr, &pptable_information->power_saving_clock_max, powerplay_table->PowerSavingClockMax, ATOM_VEGA12_PPCLOCK_COUNT);
 	phm_copy_clock_limits_array(hwmgr, &pptable_information->power_saving_clock_min, powerplay_table->PowerSavingClockMin, ATOM_VEGA12_PPCLOCK_COUNT);
 
-	pptable_information->smc_pptable = (PPTable_t *)kmalloc(sizeof(PPTable_t), GFP_KERNEL);
+	pptable_information->smc_pptable = kmalloc(sizeof(PPTable_t), GFP_KERNEL);
 	if (pptable_information->smc_pptable == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.h
index 65652ae65929..65652ae65929 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_thermal.c
index 7ace439dcde7..7ace439dcde7 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_thermal.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_thermal.h
index 0d8ed039ab12..0d8ed039ab12 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_thermal.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c
index 2a28c9df15a0..2a28c9df15a0 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.h
index f06471e712dc..f06471e712dc 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c
index ea70d736f6a8..da84012b7fd5 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c
@@ -55,6 +55,11 @@
 #define smnPCIE_LC_SPEED_CNTL			0x11140290
 #define smnPCIE_LC_LINK_WIDTH_CNTL		0x11140288
 
+#define LINK_WIDTH_MAX				6
+#define LINK_SPEED_MAX				3
+static int link_width[] = {0, 1, 2, 4, 8, 12, 16};
+static int link_speed[] = {25, 50, 80, 160};
+
 static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr)
 {
 	struct vega20_hwmgr *data =
@@ -484,7 +489,7 @@ static int vega20_setup_asic_task(struct pp_hwmgr *hwmgr)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
 	int ret = 0;
-	bool use_baco = (adev->in_gpu_reset &&
+	bool use_baco = (amdgpu_in_reset(adev) &&
 			 (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) ||
 		(adev->in_runpm && amdgpu_asic_supports_baco(adev));
 
@@ -2067,22 +2072,29 @@ static uint32_t vega20_dpm_get_mclk(struct pp_hwmgr *hwmgr, bool low)
 	return (mem_clk * 100);
 }
 
-static int vega20_get_metrics_table(struct pp_hwmgr *hwmgr, SmuMetrics_t *metrics_table)
+static int vega20_get_metrics_table(struct pp_hwmgr *hwmgr,
+				    SmuMetrics_t *metrics_table,
+				    bool bypass_cache)
 {
 	struct vega20_hwmgr *data =
 			(struct vega20_hwmgr *)(hwmgr->backend);
 	int ret = 0;
 
-	if (!data->metrics_time || time_after(jiffies, data->metrics_time + HZ / 2)) {
-		ret = smum_smc_table_manager(hwmgr, (uint8_t *)metrics_table,
-				TABLE_SMU_METRICS, true);
+	if (bypass_cache ||
+	    !data->metrics_time ||
+	    time_after(jiffies, data->metrics_time + msecs_to_jiffies(1))) {
+		ret = smum_smc_table_manager(hwmgr,
+					     (uint8_t *)(&data->metrics_table),
+					     TABLE_SMU_METRICS,
+					     true);
 		if (ret) {
 			pr_info("Failed to export SMU metrics table!\n");
 			return ret;
 		}
-		memcpy(&data->metrics_table, metrics_table, sizeof(SmuMetrics_t));
 		data->metrics_time = jiffies;
-	} else
+	}
+
+	if (metrics_table)
 		memcpy(metrics_table, &data->metrics_table, sizeof(SmuMetrics_t));
 
 	return ret;
@@ -2094,7 +2106,7 @@ static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr,
 	int ret = 0;
 	SmuMetrics_t metrics_table;
 
-	ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+	ret = vega20_get_metrics_table(hwmgr, &metrics_table, false);
 	if (ret)
 		return ret;
 
@@ -2132,7 +2144,7 @@ static int vega20_get_current_activity_percent(struct pp_hwmgr *hwmgr,
 	int ret = 0;
 	SmuMetrics_t metrics_table;
 
-	ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+	ret = vega20_get_metrics_table(hwmgr, &metrics_table, false);
 	if (ret)
 		return ret;
 
@@ -2162,7 +2174,7 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 
 	switch (idx) {
 	case AMDGPU_PP_SENSOR_GFX_SCLK:
-		ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+		ret = vega20_get_metrics_table(hwmgr, &metrics_table, false);
 		if (ret)
 			return ret;
 
@@ -2187,7 +2199,7 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 		*size = 4;
 		break;
 	case AMDGPU_PP_SENSOR_EDGE_TEMP:
-		ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+		ret = vega20_get_metrics_table(hwmgr, &metrics_table, false);
 		if (ret)
 			return ret;
 
@@ -2196,7 +2208,7 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 		*size = 4;
 		break;
 	case AMDGPU_PP_SENSOR_MEM_TEMP:
-		ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+		ret = vega20_get_metrics_table(hwmgr, &metrics_table, false);
 		if (ret)
 			return ret;
 
@@ -3259,6 +3271,46 @@ static int vega20_set_ppfeature_status(struct pp_hwmgr *hwmgr, uint64_t new_ppfe
 	return 0;
 }
 
+static int vega20_get_current_pcie_link_width_level(struct pp_hwmgr *hwmgr)
+{
+	struct amdgpu_device *adev = hwmgr->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
+		PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
+		>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+}
+
+static int vega20_get_current_pcie_link_width(struct pp_hwmgr *hwmgr)
+{
+	uint32_t width_level;
+
+	width_level = vega20_get_current_pcie_link_width_level(hwmgr);
+	if (width_level > LINK_WIDTH_MAX)
+		width_level = 0;
+
+	return link_width[width_level];
+}
+
+static int vega20_get_current_pcie_link_speed_level(struct pp_hwmgr *hwmgr)
+{
+	struct amdgpu_device *adev = hwmgr->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
+		PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
+		>> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
+}
+
+static int vega20_get_current_pcie_link_speed(struct pp_hwmgr *hwmgr)
+{
+	uint32_t speed_level;
+
+	speed_level = vega20_get_current_pcie_link_speed_level(hwmgr);
+	if (speed_level > LINK_SPEED_MAX)
+		speed_level = 0;
+
+	return link_speed[speed_level];
+}
+
 static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, char *buf)
 {
@@ -3271,7 +3323,6 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 	struct phm_ppt_v3_information *pptable_information =
 		(struct phm_ppt_v3_information *)hwmgr->pptable;
 	PPTable_t *pptable = (PPTable_t *)pptable_information->smc_pptable;
-	struct amdgpu_device *adev = hwmgr->adev;
 	struct pp_clock_levels_with_latency clocks;
 	struct vega20_single_dpm_table *fclk_dpm_table =
 			&(data->dpm_table.fclk_table);
@@ -3365,12 +3416,10 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 		break;
 
 	case PP_PCIE:
-		current_gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
-			     PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
-			    >> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
-		current_lane_width = (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
-			      PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
-			    >> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+		current_gen_speed =
+			vega20_get_current_pcie_link_speed_level(hwmgr);
+		current_lane_width =
+			vega20_get_current_pcie_link_width_level(hwmgr);
 		for (i = 0; i < NUM_LINK_LEVELS; i++) {
 			if (i == 1 && data->pcie_parameters_override) {
 				gen_speed = data->pcie_gen_level1;
@@ -4212,6 +4261,72 @@ static int vega20_set_xgmi_pstate(struct pp_hwmgr *hwmgr,
 	return ret;
 }
 
+static void vega20_init_gpu_metrics_v1_0(struct gpu_metrics_v1_0 *gpu_metrics)
+{
+	memset(gpu_metrics, 0xFF, sizeof(struct gpu_metrics_v1_0));
+
+	gpu_metrics->common_header.structure_size =
+				sizeof(struct gpu_metrics_v1_0);
+	gpu_metrics->common_header.format_revision = 1;
+	gpu_metrics->common_header.content_revision = 0;
+
+	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
+}
+
+static ssize_t vega20_get_gpu_metrics(struct pp_hwmgr *hwmgr,
+				      void **table)
+{
+	struct vega20_hwmgr *data =
+			(struct vega20_hwmgr *)(hwmgr->backend);
+	struct gpu_metrics_v1_0 *gpu_metrics =
+			&data->gpu_metrics_table;
+	SmuMetrics_t metrics;
+	uint32_t fan_speed_rpm;
+	int ret;
+
+	ret = vega20_get_metrics_table(hwmgr, &metrics, true);
+	if (ret)
+		return ret;
+
+	vega20_init_gpu_metrics_v1_0(gpu_metrics);
+
+	gpu_metrics->temperature_edge = metrics.TemperatureEdge;
+	gpu_metrics->temperature_hotspot = metrics.TemperatureHotspot;
+	gpu_metrics->temperature_mem = metrics.TemperatureHBM;
+	gpu_metrics->temperature_vrgfx = metrics.TemperatureVrGfx;
+	gpu_metrics->temperature_vrsoc = metrics.TemperatureVrSoc;
+	gpu_metrics->temperature_vrmem = metrics.TemperatureVrMem0;
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_umc_activity = metrics.AverageUclkActivity;
+
+	gpu_metrics->average_socket_power = metrics.AverageSocketPower;
+
+	gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequency;
+	gpu_metrics->average_socclk_frequency = metrics.AverageSocclkFrequency;
+	gpu_metrics->average_uclk_frequency = metrics.AverageUclkFrequency;
+
+	gpu_metrics->current_gfxclk = metrics.CurrClock[PPCLK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.CurrClock[PPCLK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.CurrClock[PPCLK_UCLK];
+	gpu_metrics->current_vclk0 = metrics.CurrClock[PPCLK_VCLK];
+	gpu_metrics->current_dclk0 = metrics.CurrClock[PPCLK_DCLK];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	vega20_fan_ctrl_get_fan_speed_rpm(hwmgr, &fan_speed_rpm);
+	gpu_metrics->current_fan_speed = (uint16_t)fan_speed_rpm;
+
+	gpu_metrics->pcie_link_width =
+			vega20_get_current_pcie_link_width(hwmgr);
+	gpu_metrics->pcie_link_speed =
+			vega20_get_current_pcie_link_speed(hwmgr);
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v1_0);
+}
+
 static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
 	/* init/fini related */
 	.backend_init = vega20_hwmgr_backend_init,
@@ -4282,6 +4397,7 @@ static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
 	.smu_i2c_bus_access = vega20_smu_i2c_bus_access,
 	.set_df_cstate = vega20_set_df_cstate,
 	.set_xgmi_pstate = vega20_set_xgmi_pstate,
+	.get_gpu_metrics = vega20_get_gpu_metrics,
 };
 
 int vega20_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h
index 2c3125f82b24..075c0094da9c 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h
@@ -527,6 +527,7 @@ struct vega20_hwmgr {
 
 	unsigned long                  metrics_time;
 	SmuMetrics_t                   metrics_table;
+	struct gpu_metrics_v1_0        gpu_metrics_table;
 
 	bool                           pcie_parameters_override;
 	uint32_t                       pcie_gen_level1;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_inc.h
index 613cb1989b3d..613cb1989b3d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_inc.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_powertune.c
index d7cc3d2d9e17..d7cc3d2d9e17 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_powertune.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_powertune.h
index d68c734c0f4e..d68c734c0f4e 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_powertune.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h
index 2222e29405c6..2222e29405c6 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_pptable.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c
index 7a7f15d0c53a..1f9082539457 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_processpptables.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c
@@ -890,14 +890,12 @@ static int init_powerplay_table_information(
 				power_saving_clock_count);
 	}
 
-	pptable_information->smc_pptable = (PPTable_t *)kmalloc(sizeof(PPTable_t), GFP_KERNEL);
+	pptable_information->smc_pptable = kmemdup(&(powerplay_table->smcPPTable),
+						   sizeof(PPTable_t),
+						   GFP_KERNEL);
 	if (pptable_information->smc_pptable == NULL)
 		return -ENOMEM;
 
-	memcpy(pptable_information->smc_pptable,
-			&(powerplay_table->smcPPTable),
-			sizeof(PPTable_t));
-
 
 	result = append_vbios_pptable(hwmgr, (pptable_information->smc_pptable));
 	if (result)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_processpptables.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.h
index 846c2cb40b35..846c2cb40b35 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_processpptables.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.h
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_thermal.c
index 364162ddaa9c..364162ddaa9c 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_thermal.c
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_thermal.h
index 2d1769bbd24e..2d1769bbd24e 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_thermal.h
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.c
index 4b3faaccecb9..4b3faaccecb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.c
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.h b/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.h
index 6df0ed41317c..6df0ed41317c 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.h
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_smc.c b/drivers/gpu/drm/amd/pm/powerplay/kv_smc.c
index 2d9ab6b8be66..2d9ab6b8be66 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_smc.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/kv_smc.c
diff --git a/drivers/gpu/drm/amd/amdgpu/ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/ppsmc.h
index 8463245f424f..8463245f424f 100644
--- a/drivers/gpu/drm/amd/amdgpu/ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/ppsmc.h
diff --git a/drivers/gpu/drm/amd/amdgpu/r600_dpm.h b/drivers/gpu/drm/amd/pm/powerplay/r600_dpm.h
index 055321f61ca7..055321f61ca7 100644
--- a/drivers/gpu/drm/amd/amdgpu/r600_dpm.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/r600_dpm.h
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/pm/powerplay/si_dpm.c
index b5986d19dc08..b5986d19dc08 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/si_dpm.c
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.h b/drivers/gpu/drm/amd/pm/powerplay/si_dpm.h
index bc0be6818e21..bc0be6818e21 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dpm.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/si_dpm.h
diff --git a/drivers/gpu/drm/amd/amdgpu/si_smc.c b/drivers/gpu/drm/amd/pm/powerplay/si_smc.c
index 8f994ffa9cd1..8f994ffa9cd1 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_smc.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/si_smc.c
diff --git a/drivers/gpu/drm/amd/amdgpu/sislands_smc.h b/drivers/gpu/drm/amd/pm/powerplay/sislands_smc.h
index d2930eceaf3c..d2930eceaf3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sislands_smc.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/sislands_smc.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/Makefile b/drivers/gpu/drm/amd/pm/powerplay/smumgr/Makefile
index 6c59c61a0d81..6c59c61a0d81 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/Makefile
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/Makefile
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c
index e4d1f3d66ef4..e4d1f3d66ef4 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.h
index a8282705c569..a8282705c569 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c
index ecb9ee46d6b3..ecb9ee46d6b3 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.h
index 6d3746268ccf..6d3746268ccf 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c
index 431ad2fd38df..431ad2fd38df 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.h
index f32c506779c9..f32c506779c9 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c
index c3d2e6dcf62a..c3d2e6dcf62a 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.h
index 1ec425df9eda..1ec425df9eda 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c
index ea2279bb8cbf..ea2279bb8cbf 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.h
index 9c2be74a2b2f..9c2be74a2b2f 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
index aae25243eb10..aae25243eb10 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h
index e7303dc8c260..e7303dc8c260 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu8_smumgr.c
index 76d4f12ceedf..76d4f12ceedf 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu8_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu8_smumgr.h
index c7b61222d258..c7b61222d258 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu8_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu9_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c
index adfbcbe5d113..8a9aee85043e 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu9_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c
@@ -61,9 +61,6 @@ static uint32_t smu9_wait_for_response(struct pp_hwmgr *hwmgr)
 	uint32_t reg;
 	uint32_t ret;
 
-	/* Due to the L1 policy problem under SRIOV, we have to use
-	 * mmMP1_SMN_C2PMSG_103 as the driver response register
-	 */
 	if (hwmgr->pp_one_vf) {
 		reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_SMN_C2PMSG_103);
 
@@ -148,10 +145,6 @@ int smu9_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr,
 
 	smu9_wait_for_response(hwmgr);
 
-	/* Due to the L1 policy problem under SRIOV, we have to use
-	 * mmMP1_SMN_C2PMSG_101 as the driver message register and
-	 * mmMP1_SMN_C2PMSG_102 as the driver parameter register.
-	 */
 	if (hwmgr->pp_one_vf) {
 		WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_103, 0);
 		WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_102, parameter);
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu9_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.h
index 1462279ca128..1462279ca128 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu9_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smumgr.c
index b6fb48066841..b6fb48066841 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c
index 4bfadb49521b..4bfadb49521b 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.h
index d664fedd3d85..d664fedd3d85 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega10_smumgr.c
index 1e222c5d91a4..daf122f24f23 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega10_smumgr.c
@@ -209,11 +209,13 @@ static int vega10_smu_init(struct pp_hwmgr *hwmgr)
 	int ret;
 	struct cgs_firmware_info info = {0};
 
-	ret = cgs_get_firmware_info(hwmgr->device,
-				    CGS_UCODE_ID_SMU,
-				    &info);
-	if (ret || !info.kptr)
-		return -EINVAL;
+	if (!amdgpu_sriov_vf((struct amdgpu_device *)hwmgr->adev)) {
+		ret = cgs_get_firmware_info(hwmgr->device,
+						CGS_UCODE_ID_SMU,
+						&info);
+		if (ret || !info.kptr)
+			return -EINVAL;
+	}
 
 	priv = kzalloc(sizeof(struct vega10_smumgr), GFP_KERNEL);
 
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega10_smumgr.h
index bad760f22624..bad760f22624 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega10_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega12_smumgr.c
index f54df76537e4..f54df76537e4 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega12_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega12_smumgr.h
index aeec965ce81f..aeec965ce81f 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega12_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c
index cf43629d29d2..cf43629d29d2 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.h
index 62ebbfd6068f..62ebbfd6068f 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.h
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c
index 0ecc18b55ffb..0ecc18b55ffb 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.h
index 2b6558238500..2b6558238500 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.h
diff --git a/drivers/gpu/drm/amd/pm/swsmu/Makefile b/drivers/gpu/drm/amd/pm/swsmu/Makefile
new file mode 100644
index 000000000000..6f281990b7b4
--- /dev/null
+++ b/drivers/gpu/drm/amd/pm/swsmu/Makefile
@@ -0,0 +1,36 @@
+#
+# Copyright 2020 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"),
+# 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.
+#
+
+AMD_SWSMU_PATH = ../pm/swsmu
+
+SWSMU_LIBS = smu11 smu12
+
+AMD_SWSMU = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/pm/swsmu/,$(SWSMU_LIBS)))
+
+include $(AMD_SWSMU)
+
+SWSMU_MGR = amdgpu_smu.o \
+	    smu_cmn.o \
+
+AMD_SWSMU_POWER = $(addprefix $(AMD_SWSMU_PATH)/,$(SWSMU_MGR))
+
+AMD_POWERPLAY_FILES += $(AMD_SWSMU_POWER)
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 8dc5abb6931e..942793947b4b 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -361,20 +361,16 @@ static int smu_get_driver_allowed_feature_mask(struct smu_context *smu)
 	int ret = 0;
 	uint32_t allowed_feature_mask[SMU_FEATURE_MAX/32];
 
-	mutex_lock(&feature->mutex);
 	bitmap_zero(feature->allowed, SMU_FEATURE_MAX);
-	mutex_unlock(&feature->mutex);
 
 	ret = smu_get_allowed_feature_mask(smu, allowed_feature_mask,
 					     SMU_FEATURE_MAX/32);
 	if (ret)
 		return ret;
 
-	mutex_lock(&feature->mutex);
 	bitmap_or(feature->allowed, feature->allowed,
 		      (unsigned long *)allowed_feature_mask,
 		      feature->feature_num);
-	mutex_unlock(&feature->mutex);
 
 	return ret;
 }
@@ -473,6 +469,12 @@ static int smu_late_init(void *handle)
 	if (!smu->pm_enabled)
 		return 0;
 
+	ret = smu_post_init(smu);
+	if (ret) {
+		dev_err(adev->dev, "Failed to post smu init!\n");
+		return ret;
+	}
+
 	ret = smu_set_default_od_settings(smu);
 	if (ret) {
 		dev_err(adev->dev, "Failed to setup default OD settings!\n");
@@ -493,6 +495,8 @@ static int smu_late_init(void *handle)
 
 	smu_get_unique_id(smu);
 
+	smu_get_fan_parameters(smu);
+
 	smu_handle_task(&adev->smu,
 			smu->smu_dpm.dpm_level,
 			AMD_PP_TASK_COMPLETE_INIT,
@@ -565,9 +569,6 @@ static int smu_fini_fb_allocations(struct smu_context *smu)
 	struct smu_table *tables = smu_table->tables;
 	struct smu_table *driver_table = &(smu_table->driver_table);
 
-	if (!tables)
-		return 0;
-
 	if (tables[SMU_TABLE_PMSTATUSLOG].mc_address)
 		amdgpu_bo_free_kernel(&tables[SMU_TABLE_PMSTATUSLOG].bo,
 				      &tables[SMU_TABLE_PMSTATUSLOG].mc_address,
@@ -644,6 +645,45 @@ static int smu_free_memory_pool(struct smu_context *smu)
 	return 0;
 }
 
+static int smu_alloc_dummy_read_table(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *dummy_read_1_table =
+			&smu_table->dummy_read_1_table;
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0;
+
+	dummy_read_1_table->size = 0x40000;
+	dummy_read_1_table->align = PAGE_SIZE;
+	dummy_read_1_table->domain = AMDGPU_GEM_DOMAIN_VRAM;
+
+	ret = amdgpu_bo_create_kernel(adev,
+				      dummy_read_1_table->size,
+				      dummy_read_1_table->align,
+				      dummy_read_1_table->domain,
+				      &dummy_read_1_table->bo,
+				      &dummy_read_1_table->mc_address,
+				      &dummy_read_1_table->cpu_addr);
+	if (ret)
+		dev_err(adev->dev, "VRAM allocation for dummy read table failed!\n");
+
+	return ret;
+}
+
+static void smu_free_dummy_read_table(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *dummy_read_1_table =
+			&smu_table->dummy_read_1_table;
+
+
+	amdgpu_bo_free_kernel(&dummy_read_1_table->bo,
+			      &dummy_read_1_table->mc_address,
+			      &dummy_read_1_table->cpu_addr);
+
+	memset(dummy_read_1_table, 0, sizeof(struct smu_table));
+}
+
 static int smu_smc_table_sw_init(struct smu_context *smu)
 {
 	int ret;
@@ -679,6 +719,10 @@ static int smu_smc_table_sw_init(struct smu_context *smu)
 	if (ret)
 		return ret;
 
+	ret = smu_alloc_dummy_read_table(smu);
+	if (ret)
+		return ret;
+
 	ret = smu_i2c_init(smu, &smu->adev->pm.smu_i2c);
 	if (ret)
 		return ret;
@@ -692,6 +736,8 @@ static int smu_smc_table_sw_fini(struct smu_context *smu)
 
 	smu_i2c_fini(smu, &smu->adev->pm.smu_i2c);
 
+	smu_free_dummy_read_table(smu);
+
 	ret = smu_free_memory_pool(smu);
 	if (ret)
 		return ret;
@@ -723,6 +769,19 @@ static void smu_throttling_logging_work_fn(struct work_struct *work)
 	smu_log_thermal_throttling(smu);
 }
 
+static void smu_interrupt_work_fn(struct work_struct *work)
+{
+	struct smu_context *smu = container_of(work, struct smu_context,
+					       interrupt_work);
+
+	mutex_lock(&smu->mutex);
+
+	if (smu->ppt_funcs && smu->ppt_funcs->interrupt_work)
+		smu->ppt_funcs->interrupt_work(smu);
+
+	mutex_unlock(&smu->mutex);
+}
+
 static int smu_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -745,6 +804,8 @@ static int smu_sw_init(void *handle)
 	mutex_init(&smu->message_lock);
 
 	INIT_WORK(&smu->throttling_logging_work, smu_throttling_logging_work_fn);
+	INIT_WORK(&smu->interrupt_work, smu_interrupt_work_fn);
+	atomic64_set(&smu->throttle_int_counter, 0);
 	smu->watermarks_bitmap = 0;
 	smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 	smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
@@ -774,10 +835,13 @@ static int smu_sw_init(void *handle)
 
 	smu->smu_dpm.dpm_level = AMD_DPM_FORCED_LEVEL_AUTO;
 	smu->smu_dpm.requested_dpm_level = AMD_DPM_FORCED_LEVEL_AUTO;
-	ret = smu_init_microcode(smu);
-	if (ret) {
-		dev_err(adev->dev, "Failed to load smu firmware!\n");
-		return ret;
+
+	if (!amdgpu_sriov_vf(adev)) {
+		ret = smu_init_microcode(smu);
+		if (ret) {
+			dev_err(adev->dev, "Failed to load smu firmware!\n");
+			return ret;
+		}
 	}
 
 	ret = smu_smc_table_sw_init(smu);
@@ -955,21 +1019,14 @@ static int smu_smc_hw_setup(struct smu_context *smu)
 		return ret;
 	}
 
-	ret = smu_disable_umc_cdr_12gbps_workaround(smu);
-	if (ret) {
-		dev_err(adev->dev, "Workaround failed to disable UMC CDR feature on 12Gbps SKU!\n");
-		return ret;
-	}
-
 	/*
-	 * For Navi1X, manually switch it to AC mode as PMFW
-	 * may boot it with DC mode.
+	 * Set initialized values (get from vbios) to dpm tables context such as
+	 * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each
+	 * type of clks.
 	 */
-	ret = smu_set_power_source(smu,
-				   adev->pm.ac_power ? SMU_POWER_SOURCE_AC :
-				   SMU_POWER_SOURCE_DC);
+	ret = smu_set_default_dpm_table(smu);
 	if (ret) {
-		dev_err(adev->dev, "Failed to switch to %s mode!\n", adev->pm.ac_power ? "AC" : "DC");
+		dev_err(adev->dev, "Failed to setup default dpm clock tables!\n");
 		return ret;
 	}
 
@@ -1109,7 +1166,7 @@ static int smu_disable_dpms(struct smu_context *smu)
 	struct amdgpu_device *adev = smu->adev;
 	int ret = 0;
 	bool use_baco = !smu->is_apu &&
-		((adev->in_gpu_reset &&
+		((amdgpu_in_reset(adev) &&
 		  (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) ||
 		 ((adev->in_runpm || adev->in_hibernate) && amdgpu_asic_supports_baco(adev)));
 
@@ -1165,6 +1222,7 @@ static int smu_smc_hw_cleanup(struct smu_context *smu)
 	int ret = 0;
 
 	cancel_work_sync(&smu->throttling_logging_work);
+	cancel_work_sync(&smu->interrupt_work);
 
 	ret = smu_disable_thermal_alert(smu);
 	if (ret) {
@@ -1185,7 +1243,6 @@ static int smu_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct smu_context *smu = &adev->smu;
-	int ret = 0;
 
 	if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev))
 		return 0;
@@ -1201,11 +1258,7 @@ static int smu_hw_fini(void *handle)
 
 	adev->pm.dpm_enabled = false;
 
-	ret = smu_smc_hw_cleanup(smu);
-	if (ret)
-		return ret;
-
-	return 0;
+	return smu_smc_hw_cleanup(smu);
 }
 
 int smu_reset(struct smu_context *smu)
@@ -1445,6 +1498,8 @@ static int smu_enable_umd_pstate(void *handle,
 			amdgpu_device_ip_set_clockgating_state(smu->adev,
 							       AMD_IP_BLOCK_TYPE_GFX,
 							       AMD_CG_STATE_UNGATE);
+			smu_gfx_ulv_control(smu, false);
+			smu_deep_sleep_control(smu, false);
 		}
 	} else {
 		/* exit umd pstate, restore level, enable gfx cg*/
@@ -1452,6 +1507,8 @@ static int smu_enable_umd_pstate(void *handle,
 			if (*level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)
 				*level = smu_dpm_ctx->saved_dpm_level;
 			smu_dpm_ctx->enable_umd_pstate = false;
+			smu_deep_sleep_control(smu, true);
+			smu_gfx_ulv_control(smu, true);
 			amdgpu_device_ip_set_clockgating_state(smu->adev,
 							       AMD_IP_BLOCK_TYPE_GFX,
 							       AMD_CG_STATE_GATE);
@@ -1783,25 +1840,19 @@ int smu_write_watermarks_table(struct smu_context *smu)
 }
 
 int smu_set_watermarks_for_clock_ranges(struct smu_context *smu,
-		struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges)
+		struct pp_smu_wm_range_sets *clock_ranges)
 {
 	int ret = 0;
 
 	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
 		return -EOPNOTSUPP;
 
+	if (smu->disable_watermark)
+		return 0;
+
 	mutex_lock(&smu->mutex);
 
-	if (!smu->disable_watermark &&
-			smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) &&
-			smu_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) {
-		ret = smu_set_watermarks_table(smu, clock_ranges);
-
-		if (!(smu->watermarks_bitmap & WATERMARKS_EXIST)) {
-			smu->watermarks_bitmap |= WATERMARKS_EXIST;
-			smu->watermarks_bitmap &= ~WATERMARKS_LOADED;
-		}
-	}
+	ret = smu_set_watermarks_table(smu, clock_ranges);
 
 	mutex_unlock(&smu->mutex);
 
@@ -2191,31 +2242,44 @@ int smu_set_fan_control_mode(struct smu_context *smu, int value)
 int smu_get_fan_speed_percent(struct smu_context *smu, uint32_t *speed)
 {
 	int ret = 0;
+	uint32_t percent;
+	uint32_t current_rpm;
 
 	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
 		return -EOPNOTSUPP;
 
 	mutex_lock(&smu->mutex);
 
-	if (smu->ppt_funcs->get_fan_speed_percent)
-		ret = smu->ppt_funcs->get_fan_speed_percent(smu, speed);
+	if (smu->ppt_funcs->get_fan_speed_rpm) {
+		ret = smu->ppt_funcs->get_fan_speed_rpm(smu, &current_rpm);
+		if (!ret) {
+			percent = current_rpm * 100 / smu->fan_max_rpm;
+			*speed = percent > 100 ? 100 : percent;
+		}
+	}
 
 	mutex_unlock(&smu->mutex);
 
+
 	return ret;
 }
 
 int smu_set_fan_speed_percent(struct smu_context *smu, uint32_t speed)
 {
 	int ret = 0;
+	uint32_t rpm;
 
 	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
 		return -EOPNOTSUPP;
 
 	mutex_lock(&smu->mutex);
 
-	if (smu->ppt_funcs->set_fan_speed_percent)
-		ret = smu->ppt_funcs->set_fan_speed_percent(smu, speed);
+	if (smu->ppt_funcs->set_fan_speed_rpm) {
+		if (speed > 100)
+			speed = 100;
+		rpm = speed * smu->fan_max_rpm / 100;
+		ret = smu->ppt_funcs->set_fan_speed_rpm(smu, rpm);
+	}
 
 	mutex_unlock(&smu->mutex);
 
@@ -2255,19 +2319,6 @@ int smu_set_deep_sleep_dcefclk(struct smu_context *smu, int clk)
 	return ret;
 }
 
-int smu_set_active_display_count(struct smu_context *smu, uint32_t count)
-{
-	int ret = 0;
-
-	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
-		return -EOPNOTSUPP;
-
-	if (smu->ppt_funcs->set_active_display_count)
-		ret = smu->ppt_funcs->set_active_display_count(smu, count);
-
-	return ret;
-}
-
 int smu_get_clock_by_type(struct smu_context *smu,
 			  enum amd_pp_clock_type type,
 			  struct amd_pp_clocks *clocks)
@@ -2637,3 +2688,40 @@ int smu_get_dpm_clock_table(struct smu_context *smu,
 
 	return ret;
 }
+
+ssize_t smu_sys_get_gpu_metrics(struct smu_context *smu,
+				void **table)
+{
+	ssize_t size;
+
+	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
+		return -EOPNOTSUPP;
+
+	if (!smu->ppt_funcs->get_gpu_metrics)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&smu->mutex);
+
+	size = smu->ppt_funcs->get_gpu_metrics(smu, table);
+
+	mutex_unlock(&smu->mutex);
+
+	return size;
+}
+
+int smu_enable_mgpu_fan_boost(struct smu_context *smu)
+{
+	int ret = 0;
+
+	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&smu->mutex);
+
+	if (smu->ppt_funcs->enable_mgpu_fan_boost)
+		ret = smu->ppt_funcs->enable_mgpu_fan_boost(smu);
+
+	mutex_unlock(&smu->mutex);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/Makefile b/drivers/gpu/drm/amd/pm/swsmu/smu11/Makefile
new file mode 100644
index 000000000000..f98d97192635
--- /dev/null
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/Makefile
@@ -0,0 +1,33 @@
+#
+# Copyright 2020 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"),
+# 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.
+#
+#
+# Makefile for the 'smu manager' sub-component of powerplay.
+# It provides the smu management services for the driver.
+
+SMU11_MGR = arcturus_ppt.o \
+	    navi10_ppt.o \
+	    sienna_cichlid_ppt.o \
+	    smu_v11_0.o
+
+AMD_SWSMU_SMU11MGR = $(addprefix $(AMD_SWSMU_PATH)/smu11/,$(SMU11_MGR))
+
+AMD_POWERPLAY_FILES += $(AMD_SWSMU_SMU11MGR)
diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index 9582b38162f0..fc376281e629 100644
--- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -79,6 +79,8 @@
 /* possible frequency drift (1Mhz) */
 #define EPSILON				1
 
+#define smnPCIE_ESM_CTRL			0x111003D0
+
 static const struct cmn2asic_msg_mapping arcturus_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
 	MSG_MAP(GetSmuVersion,			     PPSMC_MSG_GetSmuVersion,			1),
@@ -234,6 +236,13 @@ static int arcturus_tables_init(struct smu_context *smu)
 		return -ENOMEM;
 	smu_table->metrics_time = 0;
 
+	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_0);
+	smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
+	if (!smu_table->gpu_metrics_table) {
+		kfree(smu_table->metrics_table);
+		return -ENOMEM;
+	}
+
 	return 0;
 }
 
@@ -377,11 +386,9 @@ static int arcturus_check_powerplay_table(struct smu_context *smu)
 		table_context->power_play_table;
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
 
-	mutex_lock(&smu_baco->mutex);
 	if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_BACO ||
 	    powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_MACO)
 		smu_baco->platform_support = true;
-	mutex_unlock(&smu_baco->mutex);
 
 	table_context->thermal_controller_type =
 		powerplay_table->thermal_controller_type;
@@ -542,19 +549,12 @@ static int arcturus_get_smu_metrics_data(struct smu_context *smu,
 
 	mutex_lock(&smu->metrics_lock);
 
-	if (!smu_table->metrics_time ||
-	     time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(1))) {
-		ret = smu_cmn_update_table(smu,
-				       SMU_TABLE_SMU_METRICS,
-				       0,
-				       smu_table->metrics_table,
-				       false);
-		if (ret) {
-			dev_info(smu->adev->dev, "Failed to export SMU metrics table!\n");
-			mutex_unlock(&smu->metrics_lock);
-			return ret;
-		}
-		smu_table->metrics_time = jiffies;
+	ret = smu_cmn_get_metrics_table_locked(smu,
+					       NULL,
+					       false);
+	if (ret) {
+		mutex_unlock(&smu->metrics_lock);
+		return ret;
 	}
 
 	switch (member) {
@@ -897,9 +897,10 @@ static int arcturus_force_clk_levels(struct smu_context *smu,
 		return ret;
 	}
 
-	if (smu_version >= 0x361200) {
+	if ((smu_version >= 0x361200) &&
+	    (smu_version <= 0x361a00)) {
 		dev_err(smu->adev->dev, "Forcing clock level is not supported with "
-		       "54.18 and onwards SMU firmwares\n");
+		       "54.18 - 54.26(included) SMU firmwares\n");
 		return -EOPNOTSUPP;
 	}
 
@@ -1120,29 +1121,23 @@ static int arcturus_get_fan_speed_rpm(struct smu_context *smu,
 	if (!speed)
 		return -EINVAL;
 
-	return arcturus_get_smu_metrics_data(smu,
-					     METRICS_CURR_FANSPEED,
-					     speed);
+	switch (smu_v11_0_get_fan_control_mode(smu)) {
+	case AMD_FAN_CTRL_AUTO:
+		return arcturus_get_smu_metrics_data(smu,
+						     METRICS_CURR_FANSPEED,
+						     speed);
+	default:
+		return smu_v11_0_get_fan_speed_rpm(smu, speed);
+	}
 }
 
-static int arcturus_get_fan_speed_percent(struct smu_context *smu,
-					  uint32_t *speed)
+static int arcturus_get_fan_parameters(struct smu_context *smu)
 {
 	PPTable_t *pptable = smu->smu_table.driver_pptable;
-	uint32_t percent, current_rpm;
-	int ret = 0;
-
-	if (!speed)
-		return -EINVAL;
-
-	ret = arcturus_get_fan_speed_rpm(smu, &current_rpm);
-	if (ret)
-		return ret;
 
-	percent = current_rpm * 100 / pptable->FanMaximumRpm;
-	*speed = percent > 100 ? 100 : percent;
+	smu->fan_max_rpm = pptable->FanMaximumRpm;
 
-	return ret;
+	return 0;
 }
 
 static int arcturus_get_power_limit(struct smu_context *smu)
@@ -1392,9 +1387,10 @@ static int arcturus_set_performance_level(struct smu_context *smu,
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
-		if (smu_version >= 0x361200) {
+		if ((smu_version >= 0x361200) &&
+		    (smu_version <= 0x361a00)) {
 			dev_err(smu->adev->dev, "Forcing clock level is not supported with "
-			       "54.18 and onwards SMU firmwares\n");
+			       "54.18 - 54.26(included) SMU firmwares\n");
 			return -EOPNOTSUPP;
 		}
 		break;
@@ -2240,6 +2236,77 @@ static void arcturus_log_thermal_throttling_event(struct smu_context *smu)
 
 	dev_warn(adev->dev, "WARN: GPU thermal throttling temperature reached, expect performance decrease. %s.\n",
 			log_buf);
+	kgd2kfd_smi_event_throttle(smu->adev->kfd.dev, throttler_status);
+}
+
+static int arcturus_get_current_pcie_link_speed(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t esm_ctrl;
+
+	/* TODO: confirm this on real target */
+	esm_ctrl = RREG32_PCIE(smnPCIE_ESM_CTRL);
+	if ((esm_ctrl >> 15) & 0x1FFFF)
+		return (((esm_ctrl >> 8) & 0x3F) + 128);
+
+	return smu_v11_0_get_current_pcie_link_speed(smu);
+}
+
+static ssize_t arcturus_get_gpu_metrics(struct smu_context *smu,
+					void **table)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct gpu_metrics_v1_0 *gpu_metrics =
+		(struct gpu_metrics_v1_0 *)smu_table->gpu_metrics_table;
+	SmuMetrics_t metrics;
+	int ret = 0;
+
+	ret = smu_cmn_get_metrics_table(smu,
+					&metrics,
+					true);
+	if (ret)
+		return ret;
+
+	smu_v11_0_init_gpu_metrics_v1_0(gpu_metrics);
+
+	gpu_metrics->temperature_edge = metrics.TemperatureEdge;
+	gpu_metrics->temperature_hotspot = metrics.TemperatureHotspot;
+	gpu_metrics->temperature_mem = metrics.TemperatureHBM;
+	gpu_metrics->temperature_vrgfx = metrics.TemperatureVrGfx;
+	gpu_metrics->temperature_vrsoc = metrics.TemperatureVrSoc;
+	gpu_metrics->temperature_vrmem = metrics.TemperatureVrMem;
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_umc_activity = metrics.AverageUclkActivity;
+	gpu_metrics->average_mm_activity = metrics.VcnActivityPercentage;
+
+	gpu_metrics->average_socket_power = metrics.AverageSocketPower;
+	gpu_metrics->energy_accumulator = metrics.EnergyAccumulator;
+
+	gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequency;
+	gpu_metrics->average_socclk_frequency = metrics.AverageSocclkFrequency;
+	gpu_metrics->average_uclk_frequency = metrics.AverageUclkFrequency;
+	gpu_metrics->average_vclk0_frequency = metrics.AverageVclkFrequency;
+	gpu_metrics->average_dclk0_frequency = metrics.AverageDclkFrequency;
+
+	gpu_metrics->current_gfxclk = metrics.CurrClock[PPCLK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.CurrClock[PPCLK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.CurrClock[PPCLK_UCLK];
+	gpu_metrics->current_vclk0 = metrics.CurrClock[PPCLK_VCLK];
+	gpu_metrics->current_dclk0 = metrics.CurrClock[PPCLK_DCLK];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	gpu_metrics->current_fan_speed = metrics.CurrFanSpeed;
+
+	gpu_metrics->pcie_link_width =
+			smu_v11_0_get_current_pcie_link_width(smu);
+	gpu_metrics->pcie_link_speed =
+			arcturus_get_current_pcie_link_speed(smu);
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v1_0);
 }
 
 static const struct pptable_funcs arcturus_ppt_funcs = {
@@ -2254,7 +2321,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = {
 	.print_clk_levels = arcturus_print_clk_levels,
 	.force_clk_levels = arcturus_force_clk_levels,
 	.read_sensor = arcturus_read_sensor,
-	.get_fan_speed_percent = arcturus_get_fan_speed_percent,
 	.get_fan_speed_rpm = arcturus_get_fan_speed_rpm,
 	.get_power_profile_mode = arcturus_get_power_profile_mode,
 	.set_power_profile_mode = arcturus_set_power_profile_mode,
@@ -2300,7 +2366,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = {
 	.display_clock_voltage_request = smu_v11_0_display_clock_voltage_request,
 	.get_fan_control_mode = smu_v11_0_get_fan_control_mode,
 	.set_fan_control_mode = smu_v11_0_set_fan_control_mode,
-	.set_fan_speed_percent = smu_v11_0_set_fan_speed_percent,
 	.set_fan_speed_rpm = smu_v11_0_set_fan_speed_rpm,
 	.set_xgmi_pstate = smu_v11_0_set_xgmi_pstate,
 	.gfx_off_control = smu_v11_0_gfx_off_control,
@@ -2319,6 +2384,11 @@ static const struct pptable_funcs arcturus_ppt_funcs = {
 	.log_thermal_throttling_event = arcturus_log_thermal_throttling_event,
 	.get_pp_feature_mask = smu_cmn_get_pp_feature_mask,
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
+	.get_gpu_metrics = arcturus_get_gpu_metrics,
+	.gfx_ulv_control = smu_v11_0_gfx_ulv_control,
+	.deep_sleep_control = smu_v11_0_deep_sleep_control,
+	.get_fan_parameters = arcturus_get_fan_parameters,
+	.interrupt_work = smu_v11_0_interrupt_work,
 };
 
 void arcturus_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.h
index d756b16924b8..d756b16924b8 100644
--- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.h
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
index b1547a83e721..8d8081c6bd38 100644
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
@@ -45,6 +45,7 @@
 
 #include "asic_reg/mp/mp_11_0_sh_mask.h"
 #include "smu_cmn.h"
+#include "smu_11_0_cdr_table.h"
 
 /*
  * DO NOT use these for err/warn/info/debug messages.
@@ -138,6 +139,10 @@ static struct cmn2asic_msg_mapping navi10_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(DAL_ENABLE_DUMMY_PSTATE_CHANGE,	PPSMC_MSG_DALEnableDummyPstateChange,	0),
 	MSG_MAP(GetVoltageByDpm,		PPSMC_MSG_GetVoltageByDpm,		0),
 	MSG_MAP(GetVoltageByDpmOverdrive,	PPSMC_MSG_GetVoltageByDpmOverdrive,	0),
+	MSG_MAP(SetMGpuFanBoostLimitRpm,	PPSMC_MSG_SetMGpuFanBoostLimitRpm,	0),
+	MSG_MAP(SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_HIGH, PPSMC_MSG_SetDriverDummyTableDramAddrHigh, 0),
+	MSG_MAP(SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_LOW, PPSMC_MSG_SetDriverDummyTableDramAddrLow, 0),
+	MSG_MAP(GET_UMC_FW_WA,			PPSMC_MSG_GetUMCFWWA,			0),
 };
 
 static struct cmn2asic_mapping navi10_clk_map[SMU_CLK_COUNT] = {
@@ -278,9 +283,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
 				| FEATURE_MASK(FEATURE_FW_CTF_BIT)
 				| FEATURE_MASK(FEATURE_OUT_OF_BAND_MONITOR_BIT);
 
-	if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT);
-
 	if (adev->pm.pp_feature & PP_SCLK_DPM_MASK)
 		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT);
 
@@ -290,11 +292,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
 	if (adev->pm.pp_feature & PP_DCEFCLK_DPM_MASK)
 		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCEFCLK_BIT);
 
-	if (adev->pm.pp_feature & PP_MCLK_DPM_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT)
-				| FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT)
-				| FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT);
-
 	if (adev->pm.pp_feature & PP_ULV_MASK)
 		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT);
 
@@ -319,19 +316,24 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
 	if (smu->dc_controlled_by_gpio)
 		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT);
 
-	/* disable DPM UCLK and DS SOCCLK on navi10 A0 secure board */
-	if (is_asic_secure(smu)) {
-		/* only for navi10 A0 */
-		if ((adev->asic_type == CHIP_NAVI10) &&
-			(adev->rev_id == 0)) {
-			*(uint64_t *)feature_mask &=
-					~(FEATURE_MASK(FEATURE_DPM_UCLK_BIT)
-					  | FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT)
-					  | FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT));
-			*(uint64_t *)feature_mask &=
-					~FEATURE_MASK(FEATURE_DS_SOCCLK_BIT);
-		}
-	}
+	if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)
+		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT);
+
+	/* DPM UCLK enablement should be skipped for navi10 A0 secure board */
+	if (!(is_asic_secure(smu) &&
+	     (adev->asic_type == CHIP_NAVI10) &&
+	     (adev->rev_id == 0)) &&
+	    (adev->pm.pp_feature & PP_MCLK_DPM_MASK))
+		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT)
+				| FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT)
+				| FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT);
+
+	/* DS SOCCLK enablement should be skipped for navi10 A0 secure board */
+	if (is_asic_secure(smu) &&
+	    (adev->asic_type == CHIP_NAVI10) &&
+	    (adev->rev_id == 0))
+		*(uint64_t *)feature_mask &=
+				~FEATURE_MASK(FEATURE_DS_SOCCLK_BIT);
 
 	return 0;
 }
@@ -346,11 +348,9 @@ static int navi10_check_powerplay_table(struct smu_context *smu)
 	if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_HARDWAREDC)
 		smu->dc_controlled_by_gpio = true;
 
-	mutex_lock(&smu_baco->mutex);
 	if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_BACO ||
 	    powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_MACO)
 		smu_baco->platform_support = true;
-	mutex_unlock(&smu_baco->mutex);
 
 	table_context->thermal_controller_type =
 		powerplay_table->thermal_controller_type;
@@ -456,13 +456,18 @@ static int navi10_tables_init(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
 	struct smu_table *tables = smu_table->tables;
+	struct amdgpu_device *adev = smu->adev;
 
 	SMU_TABLE_INIT(tables, SMU_TABLE_PPTABLE, sizeof(PPTable_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_WATERMARKS, sizeof(Watermarks_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
-	SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(SmuMetrics_t),
-		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	if (adev->asic_type == CHIP_NAVI12)
+		SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(SmuMetrics_NV12_t),
+			       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	else
+		SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(SmuMetrics_t),
+			       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTable_t),
@@ -473,16 +478,30 @@ static int navi10_tables_init(struct smu_context *smu)
 		       sizeof(DpmActivityMonitorCoeffInt_t), PAGE_SIZE,
 		       AMDGPU_GEM_DOMAIN_VRAM);
 
-	smu_table->metrics_table = kzalloc(sizeof(SmuMetrics_t), GFP_KERNEL);
+	smu_table->metrics_table = kzalloc(adev->asic_type == CHIP_NAVI12 ?
+					   sizeof(SmuMetrics_NV12_t) :
+					   sizeof(SmuMetrics_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
-		return -ENOMEM;
+		goto err0_out;
 	smu_table->metrics_time = 0;
 
+	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_0);
+	smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
+	if (!smu_table->gpu_metrics_table)
+		goto err1_out;
+
 	smu_table->watermarks_table = kzalloc(sizeof(Watermarks_t), GFP_KERNEL);
 	if (!smu_table->watermarks_table)
-		return -ENOMEM;
+		goto err2_out;
 
 	return 0;
+
+err2_out:
+	kfree(smu_table->gpu_metrics_table);
+err1_out:
+	kfree(smu_table->metrics_table);
+err0_out:
+	return -ENOMEM;
 }
 
 static int navi10_get_smu_metrics_data(struct smu_context *smu,
@@ -490,23 +509,22 @@ static int navi10_get_smu_metrics_data(struct smu_context *smu,
 				       uint32_t *value)
 {
 	struct smu_table_context *smu_table= &smu->smu_table;
+	/*
+	 * This works for NV12 also. As although NV12 uses a different
+	 * SmuMetrics structure from other NV1X ASICs, they share the
+	 * same offsets for the heading parts(those members used here).
+	 */
 	SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table;
 	int ret = 0;
 
 	mutex_lock(&smu->metrics_lock);
-	if (!smu_table->metrics_time ||
-	     time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(1))) {
-		ret = smu_cmn_update_table(smu,
-				       SMU_TABLE_SMU_METRICS,
-				       0,
-				       smu_table->metrics_table,
-				       false);
-		if (ret) {
-			dev_info(smu->adev->dev, "Failed to export SMU metrics table!\n");
-			mutex_unlock(&smu->metrics_lock);
-			return ret;
-		}
-		smu_table->metrics_time = jiffies;
+
+	ret = smu_cmn_get_metrics_table_locked(smu,
+					       NULL,
+					       false);
+	if (ret) {
+		mutex_unlock(&smu->metrics_lock);
+		return ret;
 	}
 
 	switch (member) {
@@ -909,7 +927,6 @@ static int navi10_print_clk_levels(struct smu_context *smu,
 	uint32_t gen_speed, lane_width;
 	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
 	struct smu_11_0_dpm_context *dpm_context = smu_dpm->dpm_context;
-	struct amdgpu_device *adev = smu->adev;
 	PPTable_t *pptable = (PPTable_t *)table_context->driver_pptable;
 	OverDriveTable_t *od_table =
 		(OverDriveTable_t *)table_context->overdrive_table;
@@ -963,12 +980,8 @@ static int navi10_print_clk_levels(struct smu_context *smu,
 		}
 		break;
 	case SMU_PCIE:
-		gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
-			     PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
-			>> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
-		lane_width = (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
-			      PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
-			>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+		gen_speed = smu_v11_0_get_current_pcie_link_speed_level(smu);
+		lane_width = smu_v11_0_get_current_pcie_link_width_level(smu);
 		for (i = 0; i < NUM_LINK_LEVELS; i++)
 			size += sprintf(buf + size, "%d: %s %s %dMhz %s\n", i,
 					(dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 0) ? "2.5GT/s," :
@@ -1348,27 +1361,23 @@ static int navi10_get_fan_speed_rpm(struct smu_context *smu,
 	if (!speed)
 		return -EINVAL;
 
-	return navi10_get_smu_metrics_data(smu,
-					   METRICS_CURR_FANSPEED,
-					   speed);
+	switch (smu_v11_0_get_fan_control_mode(smu)) {
+	case AMD_FAN_CTRL_AUTO:
+		return navi10_get_smu_metrics_data(smu,
+						   METRICS_CURR_FANSPEED,
+						   speed);
+	default:
+		return smu_v11_0_get_fan_speed_rpm(smu, speed);
+	}
 }
 
-static int navi10_get_fan_speed_percent(struct smu_context *smu,
-					uint32_t *speed)
+static int navi10_get_fan_parameters(struct smu_context *smu)
 {
-	int ret = 0;
-	uint32_t percent = 0;
-	uint32_t current_rpm;
 	PPTable_t *pptable = smu->smu_table.driver_pptable;
 
-	ret = navi10_get_fan_speed_rpm(smu, &current_rpm);
-	if (ret)
-		return ret;
+	smu->fan_max_rpm = pptable->FanMaximumRpm;
 
-	percent = current_rpm * 100 / pptable->FanMaximumRpm;
-	*speed = percent > 100 ? 100 : percent;
-
-	return ret;
+	return 0;
 }
 
 static int navi10_get_power_profile_mode(struct smu_context *smu, char *buf)
@@ -1592,57 +1601,43 @@ static int navi10_notify_smc_display_config(struct smu_context *smu)
 }
 
 static int navi10_set_watermarks_table(struct smu_context *smu,
-				       struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges)
+				       struct pp_smu_wm_range_sets *clock_ranges)
 {
 	Watermarks_t *table = smu->smu_table.watermarks_table;
 	int ret = 0;
 	int i;
 
 	if (clock_ranges) {
-		if (clock_ranges->num_wm_dmif_sets > 4 ||
-		    clock_ranges->num_wm_mcif_sets > 4)
+		if (clock_ranges->num_reader_wm_sets > NUM_WM_RANGES ||
+		    clock_ranges->num_writer_wm_sets > NUM_WM_RANGES)
 			return -EINVAL;
 
-		for (i = 0; i < clock_ranges->num_wm_dmif_sets; i++) {
-			table->WatermarkRow[1][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MinUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MaxUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].WmSetting = (uint8_t)
-					clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
+		for (i = 0; i < clock_ranges->num_reader_wm_sets; i++) {
+			table->WatermarkRow[WM_DCEFCLK][i].MinClock =
+				clock_ranges->reader_wm_sets[i].min_drain_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MaxClock =
+				clock_ranges->reader_wm_sets[i].max_drain_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MinUclk =
+				clock_ranges->reader_wm_sets[i].min_fill_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MaxUclk =
+				clock_ranges->reader_wm_sets[i].max_fill_clk_mhz;
+
+			table->WatermarkRow[WM_DCEFCLK][i].WmSetting =
+				clock_ranges->reader_wm_sets[i].wm_inst;
 		}
 
-		for (i = 0; i < clock_ranges->num_wm_mcif_sets; i++) {
-			table->WatermarkRow[0][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MinUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MaxUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].WmSetting = (uint8_t)
-					clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
+		for (i = 0; i < clock_ranges->num_writer_wm_sets; i++) {
+			table->WatermarkRow[WM_SOCCLK][i].MinClock =
+				clock_ranges->writer_wm_sets[i].min_fill_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MaxClock =
+				clock_ranges->writer_wm_sets[i].max_fill_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MinUclk =
+				clock_ranges->writer_wm_sets[i].min_drain_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MaxUclk =
+				clock_ranges->writer_wm_sets[i].max_drain_clk_mhz;
+
+			table->WatermarkRow[WM_SOCCLK][i].WmSetting =
+				clock_ranges->writer_wm_sets[i].wm_inst;
 		}
 
 		smu->watermarks_bitmap |= WATERMARKS_EXIST;
@@ -2186,59 +2181,46 @@ static int navi10_run_btc(struct smu_context *smu)
 	return ret;
 }
 
-static int navi10_dummy_pstate_control(struct smu_context *smu, bool enable)
+static bool navi10_need_umc_cdr_workaround(struct smu_context *smu)
 {
-	int result = 0;
-
-	if (!enable)
-		result = smu_cmn_send_smc_msg(smu, SMU_MSG_DAL_DISABLE_DUMMY_PSTATE_CHANGE, NULL);
-	else
-		result = smu_cmn_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE, NULL);
-
-	return result;
-}
+	struct amdgpu_device *adev = smu->adev;
 
-static inline bool navi10_need_umc_cdr_12gbps_workaround(struct amdgpu_device *adev)
-{
-	if (adev->asic_type != CHIP_NAVI10)
+	if (!smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT))
 		return false;
 
-	if (adev->pdev->device == 0x731f &&
-	    (adev->pdev->revision == 0xc2 ||
-	     adev->pdev->revision == 0xc3 ||
-	     adev->pdev->revision == 0xca ||
-	     adev->pdev->revision == 0xcb))
+	if (adev->asic_type == CHIP_NAVI10 ||
+	    adev->asic_type == CHIP_NAVI14)
 		return true;
-	else
-		return false;
+
+	return false;
 }
 
-static int navi10_disable_umc_cdr_12gbps_workaround(struct smu_context *smu)
+static int navi10_umc_hybrid_cdr_workaround(struct smu_context *smu)
 {
 	uint32_t uclk_count, uclk_min, uclk_max;
-	uint32_t smu_version;
 	int ret = 0;
 
-	if (!navi10_need_umc_cdr_12gbps_workaround(smu->adev))
-		return 0;
-
-	ret = smu_cmn_get_smc_version(smu, NULL, &smu_version);
-	if (ret)
-		return ret;
-
-	/* This workaround is available only for 42.50 or later SMC firmwares */
-	if (smu_version < 0x2A3200)
+	/* This workaround can be applied only with uclk dpm enabled */
+	if (!smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT))
 		return 0;
 
 	ret = smu_v11_0_get_dpm_level_count(smu, SMU_UCLK, &uclk_count);
 	if (ret)
 		return ret;
 
-	ret = smu_v11_0_get_dpm_freq_by_index(smu, SMU_UCLK, (uint16_t)0, &uclk_min);
+	ret = smu_v11_0_get_dpm_freq_by_index(smu, SMU_UCLK, (uint16_t)(uclk_count - 1), &uclk_max);
 	if (ret)
 		return ret;
 
-	ret = smu_v11_0_get_dpm_freq_by_index(smu, SMU_UCLK, (uint16_t)(uclk_count - 1), &uclk_max);
+	/*
+	 * The NAVI10_UMC_HYBRID_CDR_WORKAROUND_UCLK_THRESHOLD is 750Mhz.
+	 * This workaround is needed only when the max uclk frequency
+	 * not greater than that.
+	 */
+	if (uclk_max > 0x2EE)
+		return 0;
+
+	ret = smu_v11_0_get_dpm_freq_by_index(smu, SMU_UCLK, (uint16_t)0, &uclk_min);
 	if (ret)
 		return ret;
 
@@ -2255,8 +2237,97 @@ static int navi10_disable_umc_cdr_12gbps_workaround(struct smu_context *smu)
 	/*
 	 * In this case, SMU already disabled dummy pstate during enablement
 	 * of UCLK DPM, we have to re-enabled it.
-	 * */
-	return navi10_dummy_pstate_control(smu, true);
+	 */
+	return smu_cmn_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE, NULL);
+}
+
+static int navi10_set_dummy_pstates_table_location(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *dummy_read_table =
+				&smu_table->dummy_read_1_table;
+	char *dummy_table = dummy_read_table->cpu_addr;
+	int ret = 0;
+	uint32_t i;
+
+	for (i = 0; i < 0x40000; i += 0x1000 * 2) {
+		memcpy(dummy_table, &NoDbiPrbs7[0], 0x1000);
+		dummy_table += 0x1000;
+		memcpy(dummy_table, &DbiPrbs7[0], 0x1000);
+		dummy_table += 0x1000;
+	}
+
+	amdgpu_asic_flush_hdp(smu->adev, NULL);
+
+	ret = smu_cmn_send_smc_msg_with_param(smu,
+					      SMU_MSG_SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_HIGH,
+					      upper_32_bits(dummy_read_table->mc_address),
+					      NULL);
+	if (ret)
+		return ret;
+
+	return smu_cmn_send_smc_msg_with_param(smu,
+					       SMU_MSG_SET_DRIVER_DUMMY_TABLE_DRAM_ADDR_LOW,
+					       lower_32_bits(dummy_read_table->mc_address),
+					       NULL);
+}
+
+static int navi10_run_umc_cdr_workaround(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint8_t umc_fw_greater_than_v136 = false;
+	uint8_t umc_fw_disable_cdr = false;
+	uint32_t pmfw_version;
+	uint32_t param;
+	int ret = 0;
+
+	if (!navi10_need_umc_cdr_workaround(smu))
+		return 0;
+
+	ret = smu_cmn_get_smc_version(smu, NULL, &pmfw_version);
+	if (ret) {
+		dev_err(adev->dev, "Failed to get smu version!\n");
+		return ret;
+	}
+
+	/*
+	 * The messages below are only supported by Navi10 42.53.0 and later
+	 * PMFWs and Navi14 53.29.0 and later PMFWs.
+	 * - PPSMC_MSG_SetDriverDummyTableDramAddrHigh
+	 * - PPSMC_MSG_SetDriverDummyTableDramAddrLow
+	 * - PPSMC_MSG_GetUMCFWWA
+	 */
+	if (((adev->asic_type == CHIP_NAVI10) && (pmfw_version >= 0x2a3500)) ||
+	    ((adev->asic_type == CHIP_NAVI14) && (pmfw_version >= 0x351D00))) {
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_GET_UMC_FW_WA,
+						      0,
+						      &param);
+		if (ret)
+			return ret;
+
+		/* First bit indicates if the UMC f/w is above v137 */
+		umc_fw_greater_than_v136 = param & 0x1;
+
+		/* Second bit indicates if hybrid-cdr is disabled */
+		umc_fw_disable_cdr = param & 0x2;
+
+		/* w/a only allowed if UMC f/w is <= 136 */
+		if (umc_fw_greater_than_v136)
+			return 0;
+
+		if (umc_fw_disable_cdr) {
+			if (adev->asic_type == CHIP_NAVI10)
+				return navi10_umc_hybrid_cdr_workaround(smu);
+		} else {
+			return navi10_set_dummy_pstates_table_location(smu);
+		}
+	} else {
+		if (adev->asic_type == CHIP_NAVI10)
+			return navi10_umc_hybrid_cdr_workaround(smu);
+	}
+
+	return 0;
 }
 
 static void navi10_fill_i2c_req(SwI2cRequest_t  *req, bool write,
@@ -2486,6 +2557,130 @@ static void navi10_i2c_control_fini(struct smu_context *smu, struct i2c_adapter
 	i2c_del_adapter(control);
 }
 
+static ssize_t navi10_get_gpu_metrics(struct smu_context *smu,
+				      void **table)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct gpu_metrics_v1_0 *gpu_metrics =
+		(struct gpu_metrics_v1_0 *)smu_table->gpu_metrics_table;
+	struct amdgpu_device *adev = smu->adev;
+	SmuMetrics_NV12_t nv12_metrics = { 0 };
+	SmuMetrics_t metrics;
+	int ret = 0;
+
+	mutex_lock(&smu->metrics_lock);
+
+	ret = smu_cmn_get_metrics_table_locked(smu,
+					       NULL,
+					       true);
+	if (ret) {
+		mutex_unlock(&smu->metrics_lock);
+		return ret;
+	}
+
+	memcpy(&metrics, smu_table->metrics_table, sizeof(SmuMetrics_t));
+	if (adev->asic_type == CHIP_NAVI12)
+		memcpy(&nv12_metrics, smu_table->metrics_table, sizeof(SmuMetrics_NV12_t));
+
+	mutex_unlock(&smu->metrics_lock);
+
+	smu_v11_0_init_gpu_metrics_v1_0(gpu_metrics);
+
+	gpu_metrics->temperature_edge = metrics.TemperatureEdge;
+	gpu_metrics->temperature_hotspot = metrics.TemperatureHotspot;
+	gpu_metrics->temperature_mem = metrics.TemperatureMem;
+	gpu_metrics->temperature_vrgfx = metrics.TemperatureVrGfx;
+	gpu_metrics->temperature_vrsoc = metrics.TemperatureVrSoc;
+	gpu_metrics->temperature_vrmem = metrics.TemperatureVrMem0;
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_umc_activity = metrics.AverageUclkActivity;
+
+	gpu_metrics->average_socket_power = metrics.AverageSocketPower;
+
+	gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequency;
+	gpu_metrics->average_socclk_frequency = metrics.AverageSocclkFrequency;
+	gpu_metrics->average_uclk_frequency = metrics.AverageUclkFrequency;
+
+	if (adev->asic_type == CHIP_NAVI12) {
+		gpu_metrics->energy_accumulator = nv12_metrics.EnergyAccumulator;
+		gpu_metrics->average_vclk0_frequency = nv12_metrics.AverageVclkFrequency;
+		gpu_metrics->average_dclk0_frequency = nv12_metrics.AverageDclkFrequency;
+		gpu_metrics->average_mm_activity = nv12_metrics.VcnActivityPercentage;
+	}
+
+	gpu_metrics->current_gfxclk = metrics.CurrClock[PPCLK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.CurrClock[PPCLK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.CurrClock[PPCLK_UCLK];
+	gpu_metrics->current_vclk0 = metrics.CurrClock[PPCLK_VCLK];
+	gpu_metrics->current_dclk0 = metrics.CurrClock[PPCLK_DCLK];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	gpu_metrics->current_fan_speed = metrics.CurrFanSpeed;
+
+	gpu_metrics->pcie_link_width =
+			smu_v11_0_get_current_pcie_link_width(smu);
+	gpu_metrics->pcie_link_speed =
+			smu_v11_0_get_current_pcie_link_speed(smu);
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v1_0);
+}
+
+static int navi10_enable_mgpu_fan_boost(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t param = 0;
+
+	/* Navi12 does not support this */
+	if (adev->asic_type == CHIP_NAVI12)
+		return 0;
+
+	/* Workaround for WS SKU */
+	if (adev->pdev->device == 0x7312 &&
+	    adev->pdev->revision == 0)
+		param = 0xD188;
+
+	return smu_cmn_send_smc_msg_with_param(smu,
+					       SMU_MSG_SetMGpuFanBoostLimitRpm,
+					       param,
+					       NULL);
+}
+
+static int navi10_post_smu_init(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0;
+
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
+	ret = navi10_run_umc_cdr_workaround(smu);
+	if (ret) {
+		dev_err(adev->dev, "Failed to apply umc cdr workaround!\n");
+		return ret;
+	}
+
+	if (!smu->dc_controlled_by_gpio) {
+		/*
+		 * For Navi1X, manually switch it to AC mode as PMFW
+		 * may boot it with DC mode.
+		 */
+		ret = smu_v11_0_set_power_source(smu,
+						 adev->pm.ac_power ?
+						 SMU_POWER_SOURCE_AC :
+						 SMU_POWER_SOURCE_DC);
+		if (ret) {
+			dev_err(adev->dev, "Failed to switch to %s mode!\n",
+					adev->pm.ac_power ? "AC" : "DC");
+			return ret;
+		}
+	}
+
+	return ret;
+}
 
 static const struct pptable_funcs navi10_ppt_funcs = {
 	.get_allowed_feature_mask = navi10_get_allowed_feature_mask,
@@ -2502,7 +2697,6 @@ static const struct pptable_funcs navi10_ppt_funcs = {
 	.display_config_changed = navi10_display_config_changed,
 	.notify_smc_display_config = navi10_notify_smc_display_config,
 	.is_dpm_running = navi10_is_dpm_running,
-	.get_fan_speed_percent = navi10_get_fan_speed_percent,
 	.get_fan_speed_rpm = navi10_get_fan_speed_rpm,
 	.get_power_profile_mode = navi10_get_power_profile_mode,
 	.set_power_profile_mode = navi10_set_power_profile_mode,
@@ -2546,7 +2740,6 @@ static const struct pptable_funcs navi10_ppt_funcs = {
 	.display_clock_voltage_request = smu_v11_0_display_clock_voltage_request,
 	.get_fan_control_mode = smu_v11_0_get_fan_control_mode,
 	.set_fan_control_mode = smu_v11_0_set_fan_control_mode,
-	.set_fan_speed_percent = smu_v11_0_set_fan_speed_percent,
 	.set_fan_speed_rpm = smu_v11_0_set_fan_speed_rpm,
 	.set_xgmi_pstate = smu_v11_0_set_xgmi_pstate,
 	.gfx_off_control = smu_v11_0_gfx_off_control,
@@ -2563,10 +2756,16 @@ static const struct pptable_funcs navi10_ppt_funcs = {
 	.set_default_od_settings = navi10_set_default_od_settings,
 	.od_edit_dpm_table = navi10_od_edit_dpm_table,
 	.run_btc = navi10_run_btc,
-	.disable_umc_cdr_12gbps_workaround = navi10_disable_umc_cdr_12gbps_workaround,
 	.set_power_source = smu_v11_0_set_power_source,
 	.get_pp_feature_mask = smu_cmn_get_pp_feature_mask,
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
+	.get_gpu_metrics = navi10_get_gpu_metrics,
+	.enable_mgpu_fan_boost = navi10_enable_mgpu_fan_boost,
+	.gfx_ulv_control = smu_v11_0_gfx_ulv_control,
+	.deep_sleep_control = smu_v11_0_deep_sleep_control,
+	.get_fan_parameters = navi10_get_fan_parameters,
+	.post_init = navi10_post_smu_init,
+	.interrupt_work = smu_v11_0_interrupt_work,
 };
 
 void navi10_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.h
index 2abb4ba01db1..84dc5a1b6830 100644
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.h
@@ -49,9 +49,6 @@
 
 #define NAVI10_VOLTAGE_SCALE (4)
 
-#define smnPCIE_LC_SPEED_CNTL			0x11140290
-#define smnPCIE_LC_LINK_WIDTH_CNTL		0x11140288
-
 extern void navi10_set_ppt_funcs(struct smu_context *smu);
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index ace682fde22f..c27806fd07e0 100644
--- a/drivers/gpu/drm/amd/powerplay/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -126,6 +126,7 @@ static struct cmn2asic_msg_mapping sienna_cichlid_message_map[SMU_MSG_MAX_COUNT]
 	MSG_MAP(BacoAudioD3PME,			PPSMC_MSG_BacoAudioD3PME,              0),
 	MSG_MAP(ArmD3,				PPSMC_MSG_ArmD3,                       0),
 	MSG_MAP(Mode1Reset,                     PPSMC_MSG_Mode1Reset,		       0),
+	MSG_MAP(SetMGpuFanBoostLimitRpm,	PPSMC_MSG_SetMGpuFanBoostLimitRpm,     0),
 };
 
 static struct cmn2asic_mapping sienna_cichlid_clk_map[SMU_CLK_COUNT] = {
@@ -297,11 +298,9 @@ static int sienna_cichlid_check_powerplay_table(struct smu_context *smu)
 		table_context->power_play_table;
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
 
-	mutex_lock(&smu_baco->mutex);
 	if (powerplay_table->platform_caps & SMU_11_0_7_PP_PLATFORM_CAP_BACO ||
 	    powerplay_table->platform_caps & SMU_11_0_7_PP_PLATFORM_CAP_MACO)
 		smu_baco->platform_support = true;
-	mutex_unlock(&smu_baco->mutex);
 
 	table_context->thermal_controller_type =
 		powerplay_table->thermal_controller_type;
@@ -388,14 +387,26 @@ static int sienna_cichlid_tables_init(struct smu_context *smu)
 
 	smu_table->metrics_table = kzalloc(sizeof(SmuMetrics_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
-		return -ENOMEM;
+		goto err0_out;
 	smu_table->metrics_time = 0;
 
+	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_0);
+	smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
+	if (!smu_table->gpu_metrics_table)
+		goto err1_out;
+
 	smu_table->watermarks_table = kzalloc(sizeof(Watermarks_t), GFP_KERNEL);
 	if (!smu_table->watermarks_table)
-		return -ENOMEM;
+		goto err2_out;
 
 	return 0;
+
+err2_out:
+	kfree(smu_table->gpu_metrics_table);
+err1_out:
+	kfree(smu_table->metrics_table);
+err0_out:
+	return -ENOMEM;
 }
 
 static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu,
@@ -407,19 +418,13 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu,
 	int ret = 0;
 
 	mutex_lock(&smu->metrics_lock);
-	if (!smu_table->metrics_time ||
-	     time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(1))) {
-		ret = smu_cmn_update_table(smu,
-				       SMU_TABLE_SMU_METRICS,
-				       0,
-				       smu_table->metrics_table,
-				       false);
-		if (ret) {
-			dev_info(smu->adev->dev, "Failed to export SMU metrics table!\n");
-			mutex_unlock(&smu->metrics_lock);
-			return ret;
-		}
-		smu_table->metrics_time = jiffies;
+
+	ret = smu_cmn_get_metrics_table_locked(smu,
+					       NULL,
+					       false);
+	if (ret) {
+		mutex_unlock(&smu->metrics_lock);
+		return ret;
 	}
 
 	switch (member) {
@@ -954,12 +959,8 @@ static int sienna_cichlid_print_clk_levels(struct smu_context *smu,
 		}
 		break;
 	case SMU_PCIE:
-		gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
-			     PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
-			>> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
-		lane_width = (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
-			      PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
-			>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+		gen_speed = smu_v11_0_get_current_pcie_link_speed(smu);
+		lane_width = smu_v11_0_get_current_pcie_link_width(smu);
 		for (i = 0; i < NUM_LINK_LEVELS; i++)
 			size += sprintf(buf + size, "%d: %s %s %dMhz %s\n", i,
 					(dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 0) ? "2.5GT/s," :
@@ -1166,27 +1167,23 @@ static int sienna_cichlid_get_fan_speed_rpm(struct smu_context *smu,
 	if (!speed)
 		return -EINVAL;
 
-	return sienna_cichlid_get_smu_metrics_data(smu,
-						   METRICS_CURR_FANSPEED,
-						   speed);
+	switch (smu_v11_0_get_fan_control_mode(smu)) {
+	case AMD_FAN_CTRL_AUTO:
+		return sienna_cichlid_get_smu_metrics_data(smu,
+							   METRICS_CURR_FANSPEED,
+							   speed);
+	default:
+		return smu_v11_0_get_fan_speed_rpm(smu, speed);
+	}
 }
 
-static int sienna_cichlid_get_fan_speed_percent(struct smu_context *smu,
-					uint32_t *speed)
+static int sienna_cichlid_get_fan_parameters(struct smu_context *smu)
 {
-	int ret = 0;
-	uint32_t percent = 0;
-	uint32_t current_rpm;
 	PPTable_t *pptable = smu->smu_table.driver_pptable;
 
-	ret = sienna_cichlid_get_fan_speed_rpm(smu, &current_rpm);
-	if (ret)
-		return ret;
-
-	percent = current_rpm * 100 / pptable->FanMaximumRpm;
-	*speed = percent > 100 ? 100 : percent;
+	smu->fan_max_rpm = pptable->FanMaximumRpm;
 
-	return ret;
+	return 0;
 }
 
 static int sienna_cichlid_get_power_profile_mode(struct smu_context *smu, char *buf)
@@ -1410,58 +1407,43 @@ static int sienna_cichlid_notify_smc_display_config(struct smu_context *smu)
 }
 
 static int sienna_cichlid_set_watermarks_table(struct smu_context *smu,
-					       struct dm_pp_wm_sets_with_clock_ranges_soc15
-					       *clock_ranges)
+					       struct pp_smu_wm_range_sets *clock_ranges)
 {
 	Watermarks_t *table = smu->smu_table.watermarks_table;
 	int ret = 0;
 	int i;
 
 	if (clock_ranges) {
-		if (clock_ranges->num_wm_dmif_sets > 4 ||
-		    clock_ranges->num_wm_mcif_sets > 4)
+		if (clock_ranges->num_reader_wm_sets > NUM_WM_RANGES ||
+		    clock_ranges->num_writer_wm_sets > NUM_WM_RANGES)
 			return -EINVAL;
 
-		for (i = 0; i < clock_ranges->num_wm_dmif_sets; i++) {
-			table->WatermarkRow[1][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MinUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].MaxUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[1][i].WmSetting = (uint8_t)
-					clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
+		for (i = 0; i < clock_ranges->num_reader_wm_sets; i++) {
+			table->WatermarkRow[WM_DCEFCLK][i].MinClock =
+				clock_ranges->reader_wm_sets[i].min_drain_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MaxClock =
+				clock_ranges->reader_wm_sets[i].max_drain_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MinUclk =
+				clock_ranges->reader_wm_sets[i].min_fill_clk_mhz;
+			table->WatermarkRow[WM_DCEFCLK][i].MaxUclk =
+				clock_ranges->reader_wm_sets[i].max_fill_clk_mhz;
+
+			table->WatermarkRow[WM_DCEFCLK][i].WmSetting =
+				clock_ranges->reader_wm_sets[i].wm_inst;
 		}
 
-		for (i = 0; i < clock_ranges->num_wm_mcif_sets; i++) {
-			table->WatermarkRow[0][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MinUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].MaxUclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz /
-				1000));
-			table->WatermarkRow[0][i].WmSetting = (uint8_t)
-					clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
+		for (i = 0; i < clock_ranges->num_writer_wm_sets; i++) {
+			table->WatermarkRow[WM_SOCCLK][i].MinClock =
+				clock_ranges->writer_wm_sets[i].min_fill_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MaxClock =
+				clock_ranges->writer_wm_sets[i].max_fill_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MinUclk =
+				clock_ranges->writer_wm_sets[i].min_drain_clk_mhz;
+			table->WatermarkRow[WM_SOCCLK][i].MaxUclk =
+				clock_ranges->writer_wm_sets[i].max_drain_clk_mhz;
+
+			table->WatermarkRow[WM_SOCCLK][i].WmSetting =
+				clock_ranges->writer_wm_sets[i].wm_inst;
 		}
 
 		smu->watermarks_bitmap |= WATERMARKS_EXIST;
@@ -2292,11 +2274,6 @@ static void sienna_cichlid_dump_pptable(struct smu_context *smu)
 	dev_info(smu->adev->dev, "SkuReserved[6] = 0x%x\n", pptable->SkuReserved[6]);
 	dev_info(smu->adev->dev, "SkuReserved[7] = 0x%x\n", pptable->SkuReserved[7]);
 	dev_info(smu->adev->dev, "SkuReserved[8] = 0x%x\n", pptable->SkuReserved[8]);
-	dev_info(smu->adev->dev, "SkuReserved[9] = 0x%x\n", pptable->SkuReserved[9]);
-	dev_info(smu->adev->dev, "SkuReserved[10] = 0x%x\n", pptable->SkuReserved[10]);
-	dev_info(smu->adev->dev, "SkuReserved[11] = 0x%x\n", pptable->SkuReserved[11]);
-	dev_info(smu->adev->dev, "SkuReserved[12] = 0x%x\n", pptable->SkuReserved[12]);
-	dev_info(smu->adev->dev, "SkuReserved[13] = 0x%x\n", pptable->SkuReserved[13]);
 
 	dev_info(smu->adev->dev, "GamingClk[0] = 0x%x\n", pptable->GamingClk[0]);
 	dev_info(smu->adev->dev, "GamingClk[1] = 0x%x\n", pptable->GamingClk[1]);
@@ -2666,6 +2643,76 @@ static void sienna_cichlid_i2c_control_fini(struct smu_context *smu, struct i2c_
 	i2c_del_adapter(control);
 }
 
+static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu,
+					      void **table)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct gpu_metrics_v1_0 *gpu_metrics =
+		(struct gpu_metrics_v1_0 *)smu_table->gpu_metrics_table;
+	SmuMetrics_t metrics;
+	int ret = 0;
+
+	ret = smu_cmn_get_metrics_table(smu,
+					&metrics,
+					true);
+	if (ret)
+		return ret;
+
+	smu_v11_0_init_gpu_metrics_v1_0(gpu_metrics);
+
+	gpu_metrics->temperature_edge = metrics.TemperatureEdge;
+	gpu_metrics->temperature_hotspot = metrics.TemperatureHotspot;
+	gpu_metrics->temperature_mem = metrics.TemperatureMem;
+	gpu_metrics->temperature_vrgfx = metrics.TemperatureVrGfx;
+	gpu_metrics->temperature_vrsoc = metrics.TemperatureVrSoc;
+	gpu_metrics->temperature_vrmem = metrics.TemperatureVrMem0;
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_umc_activity = metrics.AverageUclkActivity;
+	gpu_metrics->average_mm_activity = metrics.VcnActivityPercentage;
+
+	gpu_metrics->average_socket_power = metrics.AverageSocketPower;
+	gpu_metrics->energy_accumulator = metrics.EnergyAccumulator;
+
+	if (metrics.AverageGfxActivity <= SMU_11_0_7_GFX_BUSY_THRESHOLD)
+		gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequencyPostDs;
+	else
+		gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequencyPreDs;
+	gpu_metrics->average_uclk_frequency = metrics.AverageUclkFrequencyPostDs;
+	gpu_metrics->average_vclk0_frequency = metrics.AverageVclk0Frequency;
+	gpu_metrics->average_dclk0_frequency = metrics.AverageDclk0Frequency;
+	gpu_metrics->average_vclk1_frequency = metrics.AverageVclk1Frequency;
+	gpu_metrics->average_dclk1_frequency = metrics.AverageDclk1Frequency;
+
+	gpu_metrics->current_gfxclk = metrics.CurrClock[PPCLK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.CurrClock[PPCLK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.CurrClock[PPCLK_UCLK];
+	gpu_metrics->current_vclk0 = metrics.CurrClock[PPCLK_VCLK_0];
+	gpu_metrics->current_dclk0 = metrics.CurrClock[PPCLK_DCLK_0];
+	gpu_metrics->current_vclk1 = metrics.CurrClock[PPCLK_VCLK_1];
+	gpu_metrics->current_dclk1 = metrics.CurrClock[PPCLK_DCLK_1];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	gpu_metrics->current_fan_speed = metrics.CurrFanSpeed;
+
+	gpu_metrics->pcie_link_width =
+			smu_v11_0_get_current_pcie_link_width(smu);
+	gpu_metrics->pcie_link_speed =
+			smu_v11_0_get_current_pcie_link_speed(smu);
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v1_0);
+}
+
+static int sienna_cichlid_enable_mgpu_fan_boost(struct smu_context *smu)
+{
+	return smu_cmn_send_smc_msg_with_param(smu,
+					       SMU_MSG_SetMGpuFanBoostLimitRpm,
+					       0,
+					       NULL);
+}
 
 static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.get_allowed_feature_mask = sienna_cichlid_get_allowed_feature_mask,
@@ -2681,7 +2728,6 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.display_config_changed = sienna_cichlid_display_config_changed,
 	.notify_smc_display_config = sienna_cichlid_notify_smc_display_config,
 	.is_dpm_running = sienna_cichlid_is_dpm_running,
-	.get_fan_speed_percent = sienna_cichlid_get_fan_speed_percent,
 	.get_fan_speed_rpm = sienna_cichlid_get_fan_speed_rpm,
 	.get_power_profile_mode = sienna_cichlid_get_power_profile_mode,
 	.set_power_profile_mode = sienna_cichlid_set_power_profile_mode,
@@ -2725,7 +2771,6 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.display_clock_voltage_request = smu_v11_0_display_clock_voltage_request,
 	.get_fan_control_mode = smu_v11_0_get_fan_control_mode,
 	.set_fan_control_mode = smu_v11_0_set_fan_control_mode,
-	.set_fan_speed_percent = smu_v11_0_set_fan_speed_percent,
 	.set_fan_speed_rpm = smu_v11_0_set_fan_speed_rpm,
 	.set_xgmi_pstate = smu_v11_0_set_xgmi_pstate,
 	.gfx_off_control = smu_v11_0_gfx_off_control,
@@ -2744,6 +2789,12 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.run_btc = sienna_cichlid_run_btc,
 	.get_pp_feature_mask = smu_cmn_get_pp_feature_mask,
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
+	.get_gpu_metrics = sienna_cichlid_get_gpu_metrics,
+	.enable_mgpu_fan_boost = sienna_cichlid_enable_mgpu_fan_boost,
+	.gfx_ulv_control = smu_v11_0_gfx_ulv_control,
+	.deep_sleep_control = smu_v11_0_deep_sleep_control,
+	.get_fan_parameters = sienna_cichlid_get_fan_parameters,
+	.interrupt_work = smu_v11_0_interrupt_work,
 };
 
 void sienna_cichlid_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/sienna_cichlid_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.h
index 8078886e4cbc..57e120c440ea 100644
--- a/drivers/gpu/drm/amd/powerplay/sienna_cichlid_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.h
@@ -31,7 +31,4 @@ typedef enum {
 
 extern void sienna_cichlid_set_ppt_funcs(struct smu_context *smu);
 
-#define smnPCIE_LC_SPEED_CNTL                   0x11140290
-#define smnPCIE_LC_LINK_WIDTH_CNTL              0x11140288
-
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
index 7b950a582a28..2380759ddf48 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
@@ -67,6 +67,19 @@ MODULE_FIRMWARE("amdgpu/navy_flounder_smc.bin");
 
 #define SMU11_MODE1_RESET_WAIT_TIME_IN_MS 500  //500ms
 
+#define LINK_WIDTH_MAX				6
+#define LINK_SPEED_MAX				3
+
+#define smnPCIE_LC_LINK_WIDTH_CNTL		0x11140288
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK 0x00000070L
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT 0x4
+#define smnPCIE_LC_SPEED_CNTL			0x11140290
+#define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK 0xC000
+#define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT 0xE
+
+static int link_width[] = {0, 1, 2, 4, 8, 12, 16};
+static int link_speed[] = {25, 50, 80, 160};
+
 int smu_v11_0_init_microcode(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
@@ -309,39 +322,42 @@ int smu_v11_0_setup_pptable(struct smu_context *smu)
 	void *table;
 	uint16_t version_major, version_minor;
 
-	hdr = (const struct smc_firmware_header_v1_0 *) adev->pm.fw->data;
-	version_major = le16_to_cpu(hdr->header.header_version_major);
-	version_minor = le16_to_cpu(hdr->header.header_version_minor);
-	if ((version_major == 2 && smu->smu_table.boot_values.pp_table_id > 0) ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER) {
-		dev_info(adev->dev, "use driver provided pptable %d\n", smu->smu_table.boot_values.pp_table_id);
-		switch (version_minor) {
-		case 0:
-			ret = smu_v11_0_set_pptable_v2_0(smu, &table, &size);
-			break;
-		case 1:
-			ret = smu_v11_0_set_pptable_v2_1(smu, &table, &size,
-							 smu->smu_table.boot_values.pp_table_id);
-			break;
-		default:
-			ret = -EINVAL;
-			break;
+	if (!amdgpu_sriov_vf(adev)) {
+		hdr = (const struct smc_firmware_header_v1_0 *) adev->pm.fw->data;
+		version_major = le16_to_cpu(hdr->header.header_version_major);
+		version_minor = le16_to_cpu(hdr->header.header_version_minor);
+		if ((version_major == 2 && smu->smu_table.boot_values.pp_table_id > 0) ||
+			adev->asic_type == CHIP_NAVY_FLOUNDER) {
+			dev_info(adev->dev, "use driver provided pptable %d\n", smu->smu_table.boot_values.pp_table_id);
+			switch (version_minor) {
+			case 0:
+				ret = smu_v11_0_set_pptable_v2_0(smu, &table, &size);
+				break;
+			case 1:
+				ret = smu_v11_0_set_pptable_v2_1(smu, &table, &size,
+								smu->smu_table.boot_values.pp_table_id);
+				break;
+			default:
+				ret = -EINVAL;
+				break;
+			}
+			if (ret)
+				return ret;
+			goto out;
 		}
-		if (ret)
-			return ret;
+	}
 
-	} else {
-		dev_info(adev->dev, "use vbios provided pptable\n");
-		index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
-						    powerplayinfo);
+	dev_info(adev->dev, "use vbios provided pptable\n");
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+						powerplayinfo);
 
-		ret = amdgpu_atombios_get_data_table(adev, index, &atom_table_size, &frev, &crev,
-					      (uint8_t **)&table);
-		if (ret)
-			return ret;
-		size = atom_table_size;
-	}
+	ret = amdgpu_atombios_get_data_table(adev, index, &atom_table_size, &frev, &crev,
+						(uint8_t **)&table);
+	if (ret)
+		return ret;
+	size = atom_table_size;
 
+out:
 	if (!smu->smu_table.power_play_table)
 		smu->smu_table.power_play_table = table;
 	if (!smu->smu_table.power_play_table_size)
@@ -404,10 +420,12 @@ int smu_v11_0_fini_smc_tables(struct smu_context *smu)
 	struct smu_table_context *smu_table = &smu->smu_table;
 	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
 
+	kfree(smu_table->gpu_metrics_table);
 	kfree(smu_table->boot_overdrive_table);
 	kfree(smu_table->overdrive_table);
 	kfree(smu_table->max_sustainable_clocks);
 	kfree(smu_table->driver_pptable);
+	smu_table->gpu_metrics_table = NULL;
 	smu_table->boot_overdrive_table = NULL;
 	smu_table->overdrive_table = NULL;
 	smu_table->max_sustainable_clocks = NULL;
@@ -438,9 +456,6 @@ int smu_v11_0_init_power(struct smu_context *smu)
 {
 	struct smu_power_context *smu_power = &smu->smu_power;
 
-	if (smu_power->power_context || smu_power->power_context_size != 0)
-		return -EINVAL;
-
 	smu_power->power_context = kzalloc(sizeof(struct smu_11_0_dpm_context),
 					   GFP_KERNEL);
 	if (!smu_power->power_context)
@@ -454,9 +469,6 @@ int smu_v11_0_fini_power(struct smu_context *smu)
 {
 	struct smu_power_context *smu_power = &smu->smu_power;
 
-	if (!smu_power->power_context || smu_power->power_context_size == 0)
-		return -EINVAL;
-
 	kfree(smu_power->power_context);
 	smu_power->power_context = NULL;
 	smu_power->power_context_size = 0;
@@ -685,18 +697,16 @@ int smu_v11_0_set_tool_table_location(struct smu_context *smu)
 
 int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count)
 {
-	int ret = 0;
 	struct amdgpu_device *adev = smu->adev;
 
 	/* Navy_Flounder do not support to change display num currently */
 	if (adev->asic_type == CHIP_NAVY_FLOUNDER)
 		return 0;
 
-	if (!smu->pm_enabled)
-		return ret;
-
-	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count, NULL);
-	return ret;
+	return smu_cmn_send_smc_msg_with_param(smu,
+					       SMU_MSG_NumOfDisplays,
+					       count,
+					       NULL);
 }
 
 
@@ -706,7 +716,6 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu)
 	int ret = 0;
 	uint32_t feature_mask[2];
 
-	mutex_lock(&feature->mutex);
 	if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64)
 		goto failed;
 
@@ -723,7 +732,6 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu)
 		goto failed;
 
 failed:
-	mutex_unlock(&feature->mutex);
 	return ret;
 }
 
@@ -760,9 +768,6 @@ int smu_v11_0_notify_display_change(struct smu_context *smu)
 {
 	int ret = 0;
 
-	if (!smu->pm_enabled)
-		return ret;
-
 	if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) &&
 	    smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM)
 		ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL);
@@ -932,12 +937,45 @@ int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n)
 	return 0;
 }
 
+static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu)
+{
+	return smu_cmn_send_smc_msg(smu,
+				SMU_MSG_ReenableAcDcInterrupt,
+				NULL);
+}
+
+static int smu_v11_0_process_pending_interrupt(struct smu_context *smu)
+{
+	int ret = 0;
+
+	if (smu->dc_controlled_by_gpio &&
+	    smu_cmn_feature_is_enabled(smu, SMU_FEATURE_ACDC_BIT))
+		ret = smu_v11_0_ack_ac_dc_interrupt(smu);
+
+	return ret;
+}
+
+void smu_v11_0_interrupt_work(struct smu_context *smu)
+{
+	if (smu_v11_0_ack_ac_dc_interrupt(smu))
+		dev_err(smu->adev->dev, "Ack AC/DC interrupt Failed!\n");
+}
+
 int smu_v11_0_enable_thermal_alert(struct smu_context *smu)
 {
-	if (smu->smu_table.thermal_controller_type)
-		return amdgpu_irq_get(smu->adev, &smu->irq_source, 0);
+	int ret = 0;
 
-	return 0;
+	if (smu->smu_table.thermal_controller_type) {
+		ret = amdgpu_irq_get(smu->adev, &smu->irq_source, 0);
+		if (ret)
+			return ret;
+	}
+
+	/*
+	 * After init there might have been missed interrupts triggered
+	 * before driver registers for interrupt (Ex. AC/DC).
+	 */
+	return smu_v11_0_process_pending_interrupt(smu);
 }
 
 int smu_v11_0_disable_thermal_alert(struct smu_context *smu)
@@ -1085,35 +1123,6 @@ smu_v11_0_set_fan_static_mode(struct smu_context *smu, uint32_t mode)
 }
 
 int
-smu_v11_0_set_fan_speed_percent(struct smu_context *smu, uint32_t speed)
-{
-	struct amdgpu_device *adev = smu->adev;
-	uint32_t duty100, duty;
-	uint64_t tmp64;
-
-	if (speed > 100)
-		speed = 100;
-
-	if (smu_v11_0_auto_fan_control(smu, 0))
-		return -EINVAL;
-
-	duty100 = REG_GET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL1),
-				CG_FDO_CTRL1, FMAX_DUTY100);
-	if (!duty100)
-		return -EINVAL;
-
-	tmp64 = (uint64_t)speed * duty100;
-	do_div(tmp64, 100);
-	duty = (uint32_t)tmp64;
-
-	WREG32_SOC15(THM, 0, mmCG_FDO_CTRL0,
-		     REG_SET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL0),
-				   CG_FDO_CTRL0, FDO_STATIC_DUTY, duty));
-
-	return smu_v11_0_set_fan_static_mode(smu, FDO_PWM_MODE_STATIC);
-}
-
-int
 smu_v11_0_set_fan_control_mode(struct smu_context *smu,
 			       uint32_t mode)
 {
@@ -1121,7 +1130,7 @@ smu_v11_0_set_fan_control_mode(struct smu_context *smu,
 
 	switch (mode) {
 	case AMD_FAN_CTRL_NONE:
-		ret = smu_v11_0_set_fan_speed_percent(smu, 100);
+		ret = smu_v11_0_set_fan_speed_rpm(smu, smu->fan_max_rpm);
 		break;
 	case AMD_FAN_CTRL_MANUAL:
 		ret = smu_v11_0_auto_fan_control(smu, 0);
@@ -1167,15 +1176,34 @@ int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu,
 	return ret;
 }
 
+int smu_v11_0_get_fan_speed_rpm(struct smu_context *smu,
+				uint32_t *speed)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t tach_period, crystal_clock_freq;
+	uint64_t tmp64;
+
+	tach_period = REG_GET_FIELD(RREG32_SOC15(THM, 0, mmCG_TACH_CTRL),
+				    CG_TACH_CTRL, TARGET_PERIOD);
+	if (!tach_period)
+		return -EINVAL;
+
+	crystal_clock_freq = amdgpu_asic_get_xclk(adev);
+
+	tmp64 = (uint64_t)crystal_clock_freq * 60 * 10000;
+	do_div(tmp64, (tach_period * 8));
+	*speed = (uint32_t)tmp64;
+
+	return 0;
+}
+
 int smu_v11_0_set_xgmi_pstate(struct smu_context *smu,
 				     uint32_t pstate)
 {
-	int ret = 0;
-	ret = smu_cmn_send_smc_msg_with_param(smu,
-					  SMU_MSG_SetXgmiMode,
-					  pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3,
+	return smu_cmn_send_smc_msg_with_param(smu,
+					       SMU_MSG_SetXgmiMode,
+					       pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3,
 					  NULL);
-	return ret;
 }
 
 static int smu_v11_0_set_irq_state(struct amdgpu_device *adev,
@@ -1243,13 +1271,6 @@ static int smu_v11_0_set_irq_state(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu)
-{
-	return smu_cmn_send_smc_msg(smu,
-				SMU_MSG_ReenableAcDcInterrupt,
-				NULL);
-}
-
 #define THM_11_0__SRCID__THM_DIG_THERM_L2H		0		/* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH  */
 #define THM_11_0__SRCID__THM_DIG_THERM_H2L		1		/* ASIC_TEMP < CG_THERMAL_INT.DIG_THERM_INTL  */
 
@@ -1305,13 +1326,18 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev,
 			switch (ctxid) {
 			case 0x3:
 				dev_dbg(adev->dev, "Switched to AC mode!\n");
-				smu_v11_0_ack_ac_dc_interrupt(&adev->smu);
+				schedule_work(&smu->interrupt_work);
 				break;
 			case 0x4:
 				dev_dbg(adev->dev, "Switched to DC mode!\n");
-				smu_v11_0_ack_ac_dc_interrupt(&adev->smu);
+				schedule_work(&smu->interrupt_work);
 				break;
 			case 0x7:
+				/*
+				 * Increment the throttle interrupt counter
+				 */
+				atomic64_inc(&smu->throttle_int_counter);
+
 				if (!atomic_read(&adev->throttling_logging_enabled))
 					return 0;
 
@@ -1401,11 +1427,7 @@ int smu_v11_0_get_max_sustainable_clocks_by_dc(struct smu_context *smu,
 
 int smu_v11_0_set_azalia_d3_pme(struct smu_context *smu)
 {
-	int ret = 0;
-
-	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME, NULL);
-
-	return ret;
+	return smu_cmn_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME, NULL);
 }
 
 static int smu_v11_0_baco_set_armd3_sequence(struct smu_context *smu, enum smu_v11_0_baco_seq baco_seq)
@@ -1416,13 +1438,8 @@ static int smu_v11_0_baco_set_armd3_sequence(struct smu_context *smu, enum smu_v
 bool smu_v11_0_baco_is_support(struct smu_context *smu)
 {
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
-	bool baco_support;
-
-	mutex_lock(&smu_baco->mutex);
-	baco_support = smu_baco->platform_support;
-	mutex_unlock(&smu_baco->mutex);
 
-	if (!baco_support)
+	if (!smu_baco->platform_support)
 		return false;
 
 	/* Arcturus does not support this bit mask */
@@ -1509,13 +1526,7 @@ int smu_v11_0_baco_enter(struct smu_context *smu)
 
 int smu_v11_0_baco_exit(struct smu_context *smu)
 {
-	int ret = 0;
-
-	ret = smu_v11_0_baco_set_state(smu, SMU_BACO_STATE_EXIT);
-	if (ret)
-		return ret;
-
-	return ret;
+	return smu_v11_0_baco_set_state(smu, SMU_BACO_STATE_EXIT);
 }
 
 int smu_v11_0_mode1_reset(struct smu_context *smu)
@@ -1913,3 +1924,99 @@ int smu_v11_0_get_dpm_level_range(struct smu_context *smu,
 
 	return ret;
 }
+
+int smu_v11_0_get_current_pcie_link_width_level(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
+		PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
+		>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+}
+
+int smu_v11_0_get_current_pcie_link_width(struct smu_context *smu)
+{
+	uint32_t width_level;
+
+	width_level = smu_v11_0_get_current_pcie_link_width_level(smu);
+	if (width_level > LINK_WIDTH_MAX)
+		width_level = 0;
+
+	return link_width[width_level];
+}
+
+int smu_v11_0_get_current_pcie_link_speed_level(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	return (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
+		PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
+		>> PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
+}
+
+int smu_v11_0_get_current_pcie_link_speed(struct smu_context *smu)
+{
+	uint32_t speed_level;
+
+	speed_level = smu_v11_0_get_current_pcie_link_speed_level(smu);
+	if (speed_level > LINK_SPEED_MAX)
+		speed_level = 0;
+
+	return link_speed[speed_level];
+}
+
+void smu_v11_0_init_gpu_metrics_v1_0(struct gpu_metrics_v1_0 *gpu_metrics)
+{
+	memset(gpu_metrics, 0xFF, sizeof(struct gpu_metrics_v1_0));
+
+	gpu_metrics->common_header.structure_size =
+				sizeof(struct gpu_metrics_v1_0);
+	gpu_metrics->common_header.format_revision = 1;
+	gpu_metrics->common_header.content_revision = 0;
+
+	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
+}
+
+int smu_v11_0_gfx_ulv_control(struct smu_context *smu,
+			      bool enablement)
+{
+	int ret = 0;
+
+	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_GFX_ULV_BIT))
+		ret = smu_cmn_feature_set_enabled(smu, SMU_FEATURE_GFX_ULV_BIT, enablement);
+
+	return ret;
+}
+
+int smu_v11_0_deep_sleep_control(struct smu_context *smu,
+				 bool enablement)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0;
+
+	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DS_GFXCLK_BIT)) {
+		ret = smu_cmn_feature_set_enabled(smu, SMU_FEATURE_DS_GFXCLK_BIT, enablement);
+		if (ret) {
+			dev_err(adev->dev, "Failed to %s GFXCLK DS!\n", enablement ? "enable" : "disable");
+			return ret;
+		}
+	}
+
+	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DS_SOCCLK_BIT)) {
+		ret = smu_cmn_feature_set_enabled(smu, SMU_FEATURE_DS_SOCCLK_BIT, enablement);
+		if (ret) {
+			dev_err(adev->dev, "Failed to %s SOCCLK DS!\n", enablement ? "enable" : "disable");
+			return ret;
+		}
+	}
+
+	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DS_LCLK_BIT)) {
+		ret = smu_cmn_feature_set_enabled(smu, SMU_FEATURE_DS_LCLK_BIT, enablement);
+		if (ret) {
+			dev_err(adev->dev, "Failed to %s LCLK DS!\n", enablement ? "enable" : "disable");
+			return ret;
+		}
+	}
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/Makefile b/drivers/gpu/drm/amd/pm/swsmu/smu12/Makefile
new file mode 100644
index 000000000000..67e53f7da3ce
--- /dev/null
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/Makefile
@@ -0,0 +1,31 @@
+#
+# Copyright 2020 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"),
+# 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.
+#
+#
+# Makefile for the 'smu manager' sub-component of powerplay.
+# It provides the smu management services for the driver.
+
+SMU12_MGR = renoir_ppt.o \
+	    smu_v12_0.o
+
+AMD_SWSMU_SMU12MGR = $(addprefix $(AMD_SWSMU_PATH)/smu12/,$(SMU12_MGR))
+
+AMD_POWERPLAY_FILES += $(AMD_SWSMU_SMU12MGR)
diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
index 15263cf210d5..66c1026489be 100644
--- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
@@ -128,30 +128,6 @@ static struct cmn2asic_mapping renoir_workload_map[PP_SMC_POWER_PROFILE_COUNT] =
 	WORKLOAD_MAP(PP_SMC_POWER_PROFILE_CUSTOM,		WORKLOAD_PPLIB_CUSTOM_BIT),
 };
 
-static int renoir_get_metrics_table(struct smu_context *smu,
-				    SmuMetrics_t *metrics_table)
-{
-	struct smu_table_context *smu_table= &smu->smu_table;
-	int ret = 0;
-
-	mutex_lock(&smu->metrics_lock);
-	if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(100))) {
-		ret = smu_cmn_update_table(smu, SMU_TABLE_SMU_METRICS, 0,
-				(void *)smu_table->metrics_table, false);
-		if (ret) {
-			dev_info(smu->adev->dev, "Failed to export SMU metrics table!\n");
-			mutex_unlock(&smu->metrics_lock);
-			return ret;
-		}
-		smu_table->metrics_time = jiffies;
-	}
-
-	memcpy(metrics_table, smu_table->metrics_table, sizeof(SmuMetrics_t));
-	mutex_unlock(&smu->metrics_lock);
-
-	return ret;
-}
-
 static int renoir_init_smc_tables(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
@@ -166,18 +142,32 @@ static int renoir_init_smc_tables(struct smu_context *smu)
 
 	smu_table->clocks_table = kzalloc(sizeof(DpmClocks_t), GFP_KERNEL);
 	if (!smu_table->clocks_table)
-		return -ENOMEM;
+		goto err0_out;
 
 	smu_table->metrics_table = kzalloc(sizeof(SmuMetrics_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
-		return -ENOMEM;
+		goto err1_out;
 	smu_table->metrics_time = 0;
 
 	smu_table->watermarks_table = kzalloc(sizeof(Watermarks_t), GFP_KERNEL);
 	if (!smu_table->watermarks_table)
-		return -ENOMEM;
+		goto err2_out;
+
+	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v2_0);
+	smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
+	if (!smu_table->gpu_metrics_table)
+		goto err3_out;
 
 	return 0;
+
+err3_out:
+	kfree(smu_table->watermarks_table);
+err2_out:
+	kfree(smu_table->metrics_table);
+err1_out:
+	kfree(smu_table->clocks_table);
+err0_out:
+	return -ENOMEM;
 }
 
 /**
@@ -363,7 +353,7 @@ static int renoir_print_clk_levels(struct smu_context *smu,
 
 	memset(&metrics, 0, sizeof(metrics));
 
-	ret = renoir_get_metrics_table(smu, &metrics);
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
 	if (ret)
 		return ret;
 
@@ -509,7 +499,7 @@ static int renoir_get_current_clk_freq_by_table(struct smu_context *smu,
 	int ret = 0, clk_id = 0;
 	SmuMetrics_t metrics;
 
-	ret = renoir_get_metrics_table(smu, &metrics);
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
 	if (ret)
 		return ret;
 
@@ -592,7 +582,7 @@ static int renoir_get_gpu_temperature(struct smu_context *smu, uint32_t *value)
 	if (!value)
 		return -EINVAL;
 
-	ret = renoir_get_metrics_table(smu, &metrics);
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
 	if (ret)
 		return ret;
 
@@ -612,7 +602,7 @@ static int renoir_get_current_activity_percent(struct smu_context *smu,
 	if (!value)
 		return -EINVAL;
 
-	ret = renoir_get_metrics_table(smu, &metrics);
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
 	if (ret)
 		return ret;
 
@@ -628,6 +618,44 @@ static int renoir_get_current_activity_percent(struct smu_context *smu,
 	return 0;
 }
 
+static int renoir_get_vddc(struct smu_context *smu, uint32_t *value,
+			   unsigned int index)
+{
+	int ret = 0;
+	SmuMetrics_t metrics;
+
+	if (index >= 2)
+		return -EINVAL;
+
+	if (!value)
+		return -EINVAL;
+
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
+	if (ret)
+		return ret;
+
+	*value = metrics.Voltage[index];
+
+	return 0;
+}
+
+static int renoir_get_power(struct smu_context *smu, uint32_t *value)
+{
+	int ret = 0;
+	SmuMetrics_t metrics;
+
+	if (!value)
+		return -EINVAL;
+
+	ret = smu_cmn_get_metrics_table(smu, &metrics, false);
+	if (ret)
+		return ret;
+
+	*value = metrics.CurrentSocketPower << 8;
+
+	return 0;
+}
+
 /**
  * This interface get dpm clock table for dc
  */
@@ -806,9 +834,59 @@ static int renoir_set_performance_level(struct smu_context *smu,
 		ret = renoir_force_dpm_limit_value(smu, false);
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
-	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
 		ret = renoir_unforce_dpm_levels(smu);
 		break;
+	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetHardMinGfxClk,
+						      RENOIR_UMD_PSTATE_GFXCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetHardMinFclkByFreq,
+						      RENOIR_UMD_PSTATE_FCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetHardMinSocclkByFreq,
+						      RENOIR_UMD_PSTATE_SOCCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetHardMinVcn,
+						      RENOIR_UMD_PSTATE_VCNCLK,
+						      NULL);
+		if (ret)
+			return ret;
+
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetSoftMaxGfxClk,
+						      RENOIR_UMD_PSTATE_GFXCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetSoftMaxFclkByFreq,
+						      RENOIR_UMD_PSTATE_FCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetSoftMaxSocclkByFreq,
+						      RENOIR_UMD_PSTATE_SOCCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		ret = smu_cmn_send_smc_msg_with_param(smu,
+						      SMU_MSG_SetSoftMaxVcn,
+						      RENOIR_UMD_PSTATE_VCNCLK,
+						      NULL);
+		if (ret)
+			return ret;
+		break;
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
 		ret = renoir_get_profiling_clk_mask(smu, level,
@@ -837,50 +915,48 @@ static int renoir_set_performance_level(struct smu_context *smu,
  */
 static int renoir_set_watermarks_table(
 		struct smu_context *smu,
-		struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges)
+		struct pp_smu_wm_range_sets *clock_ranges)
 {
 	Watermarks_t *table = smu->smu_table.watermarks_table;
 	int ret = 0;
 	int i;
 
 	if (clock_ranges) {
-		if (clock_ranges->num_wm_dmif_sets > 4 ||
-				clock_ranges->num_wm_mcif_sets > 4)
+		if (clock_ranges->num_reader_wm_sets > NUM_WM_RANGES ||
+		    clock_ranges->num_writer_wm_sets > NUM_WM_RANGES)
 			return -EINVAL;
 
 		/* save into smu->smu_table.tables[SMU_TABLE_WATERMARKS]->cpu_addr*/
-		for (i = 0; i < clock_ranges->num_wm_dmif_sets; i++) {
+		for (i = 0; i < clock_ranges->num_reader_wm_sets; i++) {
 			table->WatermarkRow[WM_DCFCLK][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz));
+				clock_ranges->reader_wm_sets[i].min_drain_clk_mhz;
 			table->WatermarkRow[WM_DCFCLK][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz));
+				clock_ranges->reader_wm_sets[i].max_drain_clk_mhz;
 			table->WatermarkRow[WM_DCFCLK][i].MinMclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz));
+				clock_ranges->reader_wm_sets[i].min_fill_clk_mhz;
 			table->WatermarkRow[WM_DCFCLK][i].MaxMclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz));
-			table->WatermarkRow[WM_DCFCLK][i].WmSetting = (uint8_t)
-					clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
+				clock_ranges->reader_wm_sets[i].max_fill_clk_mhz;
+
+			table->WatermarkRow[WM_DCFCLK][i].WmSetting =
+				clock_ranges->reader_wm_sets[i].wm_inst;
+			table->WatermarkRow[WM_DCFCLK][i].WmType =
+				clock_ranges->reader_wm_sets[i].wm_type;
 		}
 
-		for (i = 0; i < clock_ranges->num_wm_mcif_sets; i++) {
+		for (i = 0; i < clock_ranges->num_writer_wm_sets; i++) {
 			table->WatermarkRow[WM_SOCCLK][i].MinClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz));
+				clock_ranges->writer_wm_sets[i].min_fill_clk_mhz;
 			table->WatermarkRow[WM_SOCCLK][i].MaxClock =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz));
+				clock_ranges->writer_wm_sets[i].max_fill_clk_mhz;
 			table->WatermarkRow[WM_SOCCLK][i].MinMclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz));
+				clock_ranges->writer_wm_sets[i].min_drain_clk_mhz;
 			table->WatermarkRow[WM_SOCCLK][i].MaxMclk =
-				cpu_to_le16((uint16_t)
-				(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz));
-			table->WatermarkRow[WM_SOCCLK][i].WmSetting = (uint8_t)
-					clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
+				clock_ranges->writer_wm_sets[i].max_drain_clk_mhz;
+
+			table->WatermarkRow[WM_SOCCLK][i].WmSetting =
+				clock_ranges->writer_wm_sets[i].wm_inst;
+			table->WatermarkRow[WM_SOCCLK][i].WmType =
+				clock_ranges->writer_wm_sets[i].wm_type;
 		}
 
 		smu->watermarks_bitmap |= WATERMARKS_EXIST;
@@ -964,6 +1040,18 @@ static int renoir_read_sensor(struct smu_context *smu,
 		*(uint32_t *)data *= 100;
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VDDGFX:
+		ret = renoir_get_vddc(smu, (uint32_t *)data, 0);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_VDDNB:
+		ret = renoir_get_vddc(smu, (uint32_t *)data, 1);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_GPU_POWER:
+		ret = renoir_get_power(smu, (uint32_t *)data);
+		*size = 4;
+		break;
 	default:
 		ret = -EOPNOTSUPP;
 		break;
@@ -989,6 +1077,65 @@ static bool renoir_is_dpm_running(struct smu_context *smu)
 
 }
 
+static ssize_t renoir_get_gpu_metrics(struct smu_context *smu,
+				      void **table)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct gpu_metrics_v2_0 *gpu_metrics =
+		(struct gpu_metrics_v2_0 *)smu_table->gpu_metrics_table;
+	SmuMetrics_t metrics;
+	int ret = 0;
+
+	ret = smu_cmn_get_metrics_table(smu, &metrics, true);
+	if (ret)
+		return ret;
+
+	smu_v12_0_init_gpu_metrics_v2_0(gpu_metrics);
+
+	gpu_metrics->temperature_gfx = metrics.GfxTemperature;
+	gpu_metrics->temperature_soc = metrics.SocTemperature;
+	memcpy(&gpu_metrics->temperature_core[0],
+		&metrics.CoreTemperature[0],
+		sizeof(uint16_t) * 8);
+	gpu_metrics->temperature_l3[0] = metrics.L3Temperature[0];
+	gpu_metrics->temperature_l3[1] = metrics.L3Temperature[1];
+
+	gpu_metrics->average_gfx_activity = metrics.AverageGfxActivity;
+	gpu_metrics->average_mm_activity = metrics.AverageUvdActivity;
+
+	gpu_metrics->average_socket_power = metrics.CurrentSocketPower;
+	gpu_metrics->average_cpu_power = metrics.Power[0];
+	gpu_metrics->average_soc_power = metrics.Power[1];
+	memcpy(&gpu_metrics->average_core_power[0],
+		&metrics.CorePower[0],
+		sizeof(uint16_t) * 8);
+
+	gpu_metrics->average_gfxclk_frequency = metrics.AverageGfxclkFrequency;
+	gpu_metrics->average_socclk_frequency = metrics.AverageSocclkFrequency;
+	gpu_metrics->average_fclk_frequency = metrics.AverageFclkFrequency;
+	gpu_metrics->average_vclk_frequency = metrics.AverageVclkFrequency;
+
+	gpu_metrics->current_gfxclk = metrics.ClockFrequency[CLOCK_GFXCLK];
+	gpu_metrics->current_socclk = metrics.ClockFrequency[CLOCK_SOCCLK];
+	gpu_metrics->current_uclk = metrics.ClockFrequency[CLOCK_UMCCLK];
+	gpu_metrics->current_fclk = metrics.ClockFrequency[CLOCK_FCLK];
+	gpu_metrics->current_vclk = metrics.ClockFrequency[CLOCK_VCLK];
+	gpu_metrics->current_dclk = metrics.ClockFrequency[CLOCK_DCLK];
+	memcpy(&gpu_metrics->current_coreclk[0],
+		&metrics.CoreFrequency[0],
+		sizeof(uint16_t) * 8);
+	gpu_metrics->current_l3clk[0] = metrics.L3Frequency[0];
+	gpu_metrics->current_l3clk[1] = metrics.L3Frequency[1];
+
+	gpu_metrics->throttle_status = metrics.ThrottlerStatus;
+
+	gpu_metrics->fan_pwm = metrics.FanPwm;
+
+	*table = (void *)gpu_metrics;
+
+	return sizeof(struct gpu_metrics_v2_0);
+}
+
 static const struct pptable_funcs renoir_ppt_funcs = {
 	.set_power_state = NULL,
 	.print_clk_levels = renoir_print_clk_levels,
@@ -1023,6 +1170,7 @@ static const struct pptable_funcs renoir_ppt_funcs = {
 	.is_dpm_running = renoir_is_dpm_running,
 	.get_pp_feature_mask = smu_cmn_get_pp_feature_mask,
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
+	.get_gpu_metrics = renoir_get_gpu_metrics,
 };
 
 void renoir_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.h
index 8c3f004cdf8d..11c3c22fecbe 100644
--- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.h
@@ -29,5 +29,6 @@ extern void renoir_set_ppt_funcs(struct smu_context *smu);
 #define RENOIR_UMD_PSTATE_GFXCLK       700
 #define RENOIR_UMD_PSTATE_SOCCLK       678
 #define RENOIR_UMD_PSTATE_FCLK         800
+#define RENOIR_UMD_PSTATE_VCNCLK       0x022D01D8
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c
index 31456437bb18..660f403d5770 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c
@@ -274,3 +274,15 @@ int smu_v12_0_set_driver_table_location(struct smu_context *smu)
 
 	return ret;
 }
+
+void smu_v12_0_init_gpu_metrics_v2_0(struct gpu_metrics_v2_0 *gpu_metrics)
+{
+	memset(gpu_metrics, 0xFF, sizeof(struct gpu_metrics_v2_0));
+
+	gpu_metrics->common_header.structure_size =
+				sizeof(struct gpu_metrics_v2_0);
+	gpu_metrics->common_header.format_revision = 2;
+	gpu_metrics->common_header.content_revision = 0;
+
+	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
+}
diff --git a/drivers/gpu/drm/amd/powerplay/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
index 5c23c44c33bd..c30d3338825f 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_cmn.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
@@ -112,6 +112,9 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu,
 	struct amdgpu_device *adev = smu->adev;
 	int ret = 0, index = 0;
 
+	if (smu->adev->in_pci_err_recovery)
+		return 0;
+
 	index = smu_cmn_to_asic_specific_index(smu,
 					       CMN2ASIC_MAPPING_MSG,
 					       msg);
@@ -343,9 +346,9 @@ int smu_cmn_get_enabled_mask(struct smu_context *smu,
 	return ret;
 }
 
-static int smu_cmn_feature_update_enable_state(struct smu_context *smu,
-					       uint64_t feature_mask,
-					       bool enabled)
+int smu_cmn_feature_update_enable_state(struct smu_context *smu,
+					uint64_t feature_mask,
+					bool enabled)
 {
 	struct smu_feature *feature = &smu->smu_feature;
 	int ret = 0;
@@ -604,7 +607,7 @@ int smu_cmn_update_table(struct smu_context *smu,
 		memcpy(table_data, table->cpu_addr, table_size);
 	}
 
-	return ret;
+	return 0;
 }
 
 int smu_cmn_write_watermarks_table(struct smu_context *smu)
@@ -631,3 +634,48 @@ int smu_cmn_write_pptable(struct smu_context *smu)
 				    pptable,
 				    true);
 }
+
+int smu_cmn_get_metrics_table_locked(struct smu_context *smu,
+				     void *metrics_table,
+				     bool bypass_cache)
+{
+	struct smu_table_context *smu_table= &smu->smu_table;
+	uint32_t table_size =
+		smu_table->tables[SMU_TABLE_SMU_METRICS].size;
+	int ret = 0;
+
+	if (bypass_cache ||
+	    !smu_table->metrics_time ||
+	    time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(1))) {
+		ret = smu_cmn_update_table(smu,
+				       SMU_TABLE_SMU_METRICS,
+				       0,
+				       smu_table->metrics_table,
+				       false);
+		if (ret) {
+			dev_info(smu->adev->dev, "Failed to export SMU metrics table!\n");
+			return ret;
+		}
+		smu_table->metrics_time = jiffies;
+	}
+
+	if (metrics_table)
+		memcpy(metrics_table, smu_table->metrics_table, table_size);
+
+	return 0;
+}
+
+int smu_cmn_get_metrics_table(struct smu_context *smu,
+			      void *metrics_table,
+			      bool bypass_cache)
+{
+	int ret = 0;
+
+	mutex_lock(&smu->metrics_lock);
+	ret = smu_cmn_get_metrics_table_locked(smu,
+					       metrics_table,
+					       bypass_cache);
+	mutex_unlock(&smu->metrics_lock);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/powerplay/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
index 98face8c5fd6..ab577be23c15 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_cmn.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
@@ -52,6 +52,10 @@ int smu_cmn_get_enabled_mask(struct smu_context *smu,
 			     uint32_t *feature_mask,
 			     uint32_t num);
 
+int smu_cmn_feature_update_enable_state(struct smu_context *smu,
+					uint64_t feature_mask,
+					bool enabled);
+
 int smu_cmn_feature_set_enabled(struct smu_context *smu,
 				enum smu_feature_mask mask,
 				bool enable);
@@ -79,5 +83,13 @@ int smu_cmn_write_watermarks_table(struct smu_context *smu);
 
 int smu_cmn_write_pptable(struct smu_context *smu);
 
+int smu_cmn_get_metrics_table_locked(struct smu_context *smu,
+				     void *metrics_table,
+				     bool bypass_cache);
+
+int smu_cmn_get_metrics_table(struct smu_context *smu,
+			      void *metrics_table,
+			      bool bypass_cache);
+
 #endif
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/smu_internal.h b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h
index 264073d4e263..c5adbe46ba0d 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_internal.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h
@@ -42,6 +42,7 @@
 #define smu_check_fw_version(smu)					smu_ppt_funcs(check_fw_version, 0, smu)
 #define smu_write_pptable(smu)						smu_ppt_funcs(write_pptable, 0, smu)
 #define smu_set_min_dcef_deep_sleep(smu, clk)				smu_ppt_funcs(set_min_dcef_deep_sleep, 0, smu, clk)
+#define smu_set_active_display_count(smu, count)			smu_ppt_funcs(set_active_display_count, 0, smu, count)
 #define smu_set_driver_table_location(smu)				smu_ppt_funcs(set_driver_table_location, 0, smu)
 #define smu_set_tool_table_location(smu)				smu_ppt_funcs(set_tool_table_location, 0, smu)
 #define smu_notify_memory_pool_location(smu)				smu_ppt_funcs(notify_memory_pool_location, 0, smu)
@@ -83,7 +84,6 @@
 #define smu_asic_set_performance_level(smu, level)			smu_ppt_funcs(set_performance_level, -EINVAL, smu, level)
 #define smu_dump_pptable(smu)						smu_ppt_funcs(dump_pptable, 0, smu)
 #define smu_update_pcie_parameters(smu, pcie_gen_cap, pcie_width_cap)	smu_ppt_funcs(update_pcie_parameters, 0, smu, pcie_gen_cap, pcie_width_cap)
-#define smu_disable_umc_cdr_12gbps_workaround(smu)			smu_ppt_funcs(disable_umc_cdr_12gbps_workaround, 0, smu)
 #define smu_set_power_source(smu, power_src)				smu_ppt_funcs(set_power_source, 0, smu, power_src)
 #define smu_i2c_init(smu, control)					smu_ppt_funcs(i2c_init, 0, smu, control)
 #define smu_i2c_fini(smu, control)					smu_ppt_funcs(i2c_fini, 0, smu, control)
@@ -92,6 +92,10 @@
 #define smu_get_asic_power_limits(smu)					smu_ppt_funcs(get_power_limit, 0, smu)
 #define smu_get_pp_feature_mask(smu, buf)				smu_ppt_funcs(get_pp_feature_mask, 0, smu, buf)
 #define smu_set_pp_feature_mask(smu, new_mask)				smu_ppt_funcs(set_pp_feature_mask, 0, smu, new_mask)
+#define smu_gfx_ulv_control(smu, enablement)				smu_ppt_funcs(gfx_ulv_control, 0, smu, enablement)
+#define smu_deep_sleep_control(smu, enablement)				smu_ppt_funcs(deep_sleep_control, 0, smu, enablement)
+#define smu_get_fan_parameters(smu)					smu_ppt_funcs(get_fan_parameters, 0, smu)
+#define smu_post_init(smu)						smu_ppt_funcs(post_init, 0, smu)
 
 #endif
 #endif
diff --git a/drivers/gpu/drm/arm/malidp_planes.c b/drivers/gpu/drm/arm/malidp_planes.c
index ab45ac445045..351a85088d0e 100644
--- a/drivers/gpu/drm/arm/malidp_planes.c
+++ b/drivers/gpu/drm/arm/malidp_planes.c
@@ -346,7 +346,7 @@ static bool malidp_check_pages_threshold(struct malidp_plane_state *ms,
 		if (cma_obj->sgt)
 			sgt = cma_obj->sgt;
 		else
-			sgt = obj->dev->driver->gem_prime_get_sg_table(obj);
+			sgt = obj->funcs->get_sg_table(obj);
 
 		if (!sgt)
 			return false;
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 38dfaa46d306..a887b6a5f8bd 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -757,7 +757,7 @@ static int armada_drm_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 static void armada_drm_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
-	struct armada_private *priv = crtc->dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(crtc->dev);
 
 	if (dcrtc->cursor_obj)
 		drm_gem_object_put(&dcrtc->cursor_obj->obj);
@@ -901,7 +901,7 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	struct resource *res, int irq, const struct armada_variant *variant,
 	struct device_node *port)
 {
-	struct armada_private *priv = drm->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(drm);
 	struct armada_crtc *dcrtc;
 	struct drm_plane *primary;
 	void __iomem *base;
diff --git a/drivers/gpu/drm/armada/armada_debugfs.c b/drivers/gpu/drm/armada/armada_debugfs.c
index c6fc2f1d58e9..29f4b52e3c8d 100644
--- a/drivers/gpu/drm/armada/armada_debugfs.c
+++ b/drivers/gpu/drm/armada/armada_debugfs.c
@@ -19,7 +19,7 @@ static int armada_debugfs_gem_linear_show(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 	struct drm_printer p = drm_seq_file_printer(m);
 
 	mutex_lock(&priv->linear_lock);
diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
index a11bdaccbb33..6a5a87932576 100644
--- a/drivers/gpu/drm/armada/armada_drm.h
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -73,6 +73,8 @@ struct armada_private {
 #endif
 };
 
+#define drm_to_armada_dev(dev) container_of(dev, struct armada_private, drm)
+
 int armada_fbdev_init(struct drm_device *);
 void armada_fbdev_fini(struct drm_device *);
 
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 5fc25c3f445c..980d3f1f8f16 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -87,24 +87,13 @@ static int armada_drm_bind(struct device *dev)
 				     "armada-drm"))
 		return -EBUSY;
 
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
-
-	/*
-	 * The drm_device structure must be at the start of
-	 * armada_private for drm_dev_put() to work correctly.
-	 */
-	BUILD_BUG_ON(offsetof(struct armada_private, drm) != 0);
-
-	ret = drm_dev_init(&priv->drm, &armada_drm_driver, dev);
-	if (ret) {
-		dev_err(dev, "[" DRM_NAME ":%s] drm_dev_init failed: %d\n",
-			__func__, ret);
-		kfree(priv);
-		return ret;
+	priv = devm_drm_dev_alloc(dev, &armada_drm_driver,
+				  struct armada_private, drm);
+	if (IS_ERR(priv)) {
+		dev_err(dev, "[" DRM_NAME ":%s] devm_drm_dev_alloc failed: %li\n",
+			__func__, PTR_ERR(priv));
+		return PTR_ERR(priv);
 	}
-	drmm_add_final_kfree(&priv->drm, priv);
 
 	/* Remove early framebuffers */
 	ret = drm_fb_helper_remove_conflicting_framebuffers(NULL,
@@ -117,8 +106,6 @@ static int armada_drm_bind(struct device *dev)
 		return ret;
 	}
 
-	priv->drm.dev_private = priv;
-
 	dev_set_drvdata(dev, &priv->drm);
 
 	/* Mode setting support */
@@ -174,14 +161,13 @@ static int armada_drm_bind(struct device *dev)
  err_kms:
 	drm_mode_config_cleanup(&priv->drm);
 	drm_mm_takedown(&priv->linear);
-	drm_dev_put(&priv->drm);
 	return ret;
 }
 
 static void armada_drm_unbind(struct device *dev)
 {
 	struct drm_device *drm = dev_get_drvdata(dev);
-	struct armada_private *priv = drm->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(drm);
 
 	drm_kms_helper_poll_fini(&priv->drm);
 	armada_fbdev_fini(&priv->drm);
@@ -194,8 +180,6 @@ static void armada_drm_unbind(struct device *dev)
 
 	drm_mode_config_cleanup(&priv->drm);
 	drm_mm_takedown(&priv->linear);
-
-	drm_dev_put(&priv->drm);
 }
 
 static int compare_of(struct device *dev, void *data)
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 0c4601275507..38f5170c0fea 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -117,7 +117,7 @@ static const struct drm_fb_helper_funcs armada_fb_helper_funcs = {
 
 int armada_fbdev_init(struct drm_device *dev)
 {
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 	struct drm_fb_helper *fbh;
 	int ret;
 
@@ -151,7 +151,7 @@ int armada_fbdev_init(struct drm_device *dev)
 
 void armada_fbdev_fini(struct drm_device *dev)
 {
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 	struct drm_fb_helper *fbh = priv->fbdev;
 
 	if (fbh) {
diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c
index 8005614d2e6b..6654bccd9466 100644
--- a/drivers/gpu/drm/armada/armada_gem.c
+++ b/drivers/gpu/drm/armada/armada_gem.c
@@ -39,7 +39,7 @@ static size_t roundup_gem_size(size_t size)
 void armada_gem_free_object(struct drm_gem_object *obj)
 {
 	struct armada_gem_object *dobj = drm_to_armada_gem(obj);
-	struct armada_private *priv = obj->dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(obj->dev);
 
 	DRM_DEBUG_DRIVER("release obj %p\n", dobj);
 
@@ -77,7 +77,7 @@ void armada_gem_free_object(struct drm_gem_object *obj)
 int
 armada_gem_linear_back(struct drm_device *dev, struct armada_gem_object *obj)
 {
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 	size_t size = obj->obj.size;
 
 	if (obj->page || obj->linear)
@@ -379,7 +379,7 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 	struct armada_gem_object *dobj = drm_to_armada_gem(obj);
 	struct scatterlist *sg;
 	struct sg_table *sgt;
-	int i, num;
+	int i;
 
 	sgt = kmalloc(sizeof(*sgt), GFP_KERNEL);
 	if (!sgt)
@@ -395,22 +395,18 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 
 		mapping = dobj->obj.filp->f_mapping;
 
-		for_each_sg(sgt->sgl, sg, count, i) {
+		for_each_sgtable_sg(sgt, sg, i) {
 			struct page *page;
 
 			page = shmem_read_mapping_page(mapping, i);
-			if (IS_ERR(page)) {
-				num = i;
+			if (IS_ERR(page))
 				goto release;
-			}
 
 			sg_set_page(sg, page, PAGE_SIZE, 0);
 		}
 
-		if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0) {
-			num = sgt->nents;
+		if (dma_map_sgtable(attach->dev, sgt, dir, 0))
 			goto release;
-		}
 	} else if (dobj->page) {
 		/* Single contiguous page */
 		if (sg_alloc_table(sgt, 1, GFP_KERNEL))
@@ -418,7 +414,7 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 
 		sg_set_page(sgt->sgl, dobj->page, dobj->obj.size, 0);
 
-		if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0)
+		if (dma_map_sgtable(attach->dev, sgt, dir, 0))
 			goto free_table;
 	} else if (dobj->linear) {
 		/* Single contiguous physical region - no struct page */
@@ -432,8 +428,9 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 	return sgt;
 
  release:
-	for_each_sg(sgt->sgl, sg, num, i)
-		put_page(sg_page(sg));
+	for_each_sgtable_sg(sgt, sg, i)
+		if (sg_page(sg))
+			put_page(sg_page(sg));
  free_table:
 	sg_free_table(sgt);
  free_sgt:
@@ -449,11 +446,12 @@ static void armada_gem_prime_unmap_dma_buf(struct dma_buf_attachment *attach,
 	int i;
 
 	if (!dobj->linear)
-		dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir);
+		dma_unmap_sgtable(attach->dev, sgt, dir, 0);
 
 	if (dobj->obj.filp) {
 		struct scatterlist *sg;
-		for_each_sg(sgt->sgl, sg, sgt->nents, i)
+
+		for_each_sgtable_sg(sgt, sg, i)
 			put_page(sg_page(sg));
 	}
 
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 07f0da4d9ba1..30e01101f59e 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -344,7 +344,7 @@ static int armada_overlay_set_property(struct drm_plane *plane,
 	struct drm_plane_state *state, struct drm_property *property,
 	uint64_t val)
 {
-	struct armada_private *priv = plane->dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(plane->dev);
 
 #define K2R(val) (((val) >> 0) & 0xff)
 #define K2G(val) (((val) >> 8) & 0xff)
@@ -412,7 +412,7 @@ static int armada_overlay_get_property(struct drm_plane *plane,
 	const struct drm_plane_state *state, struct drm_property *property,
 	uint64_t *val)
 {
-	struct armada_private *priv = plane->dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(plane->dev);
 
 #define C2K(c,s)	(((c) >> (s)) & 0xff)
 #define R2BGR(r,g,b,s)	(C2K(r,s) << 0 | C2K(g,s) << 8 | C2K(b,s) << 16)
@@ -505,7 +505,7 @@ static const struct drm_prop_enum_list armada_drm_colorkey_enum_list[] = {
 
 static int armada_overlay_create_properties(struct drm_device *dev)
 {
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 
 	if (priv->colorkey_prop)
 		return 0;
@@ -539,7 +539,7 @@ static int armada_overlay_create_properties(struct drm_device *dev)
 
 int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 {
-	struct armada_private *priv = dev->dev_private;
+	struct armada_private *priv = drm_to_armada_dev(dev);
 	struct drm_mode_object *mobj;
 	struct drm_plane *overlay;
 	int ret;
diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c
index 903f4f304647..2b424b2b85cc 100644
--- a/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c
+++ b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c
@@ -63,15 +63,21 @@ static const struct drm_mode_config_funcs aspeed_gfx_mode_config_funcs = {
 	.atomic_commit		= drm_atomic_helper_commit,
 };
 
-static void aspeed_gfx_setup_mode_config(struct drm_device *drm)
+static int aspeed_gfx_setup_mode_config(struct drm_device *drm)
 {
-	drm_mode_config_init(drm);
+	int ret;
+
+	ret = drmm_mode_config_init(drm);
+	if (ret)
+		return ret;
 
 	drm->mode_config.min_width = 0;
 	drm->mode_config.min_height = 0;
 	drm->mode_config.max_width = 800;
 	drm->mode_config.max_height = 600;
 	drm->mode_config.funcs = &aspeed_gfx_mode_config_funcs;
+
+	return ret;
 }
 
 static irqreturn_t aspeed_gfx_irq_handler(int irq, void *data)
@@ -144,7 +150,9 @@ static int aspeed_gfx_load(struct drm_device *drm)
 	writel(0, priv->base + CRT_CTRL1);
 	writel(0, priv->base + CRT_CTRL2);
 
-	aspeed_gfx_setup_mode_config(drm);
+	ret = aspeed_gfx_setup_mode_config(drm);
+	if (ret < 0)
+		return ret;
 
 	ret = drm_vblank_init(drm, 1);
 	if (ret < 0) {
@@ -179,7 +187,6 @@ static int aspeed_gfx_load(struct drm_device *drm)
 static void aspeed_gfx_unload(struct drm_device *drm)
 {
 	drm_kms_helper_poll_fini(drm);
-	drm_mode_config_cleanup(drm);
 }
 
 DEFINE_DRM_GEM_CMA_FOPS(fops);
diff --git a/drivers/gpu/drm/ast/ast_cursor.c b/drivers/gpu/drm/ast/ast_cursor.c
index acf0d23514e8..e0f4613918ad 100644
--- a/drivers/gpu/drm/ast/ast_cursor.c
+++ b/drivers/gpu/drm/ast/ast_cursor.c
@@ -47,7 +47,7 @@ static void ast_cursor_fini(struct ast_private *ast)
 
 static void ast_cursor_release(struct drm_device *dev, void *ptr)
 {
-	struct ast_private *ast = dev->dev_private;
+	struct ast_private *ast = to_ast_private(dev);
 
 	ast_cursor_fini(ast);
 }
@@ -57,7 +57,7 @@ static void ast_cursor_release(struct drm_device *dev, void *ptr)
  */
 int ast_cursor_init(struct ast_private *ast)
 {
-	struct drm_device *dev = ast->dev;
+	struct drm_device *dev = &ast->base;
 	size_t size, i;
 	struct drm_gem_vram_object *gbo;
 	void __iomem *vaddr;
@@ -168,7 +168,7 @@ static void update_cursor_image(u8 __iomem *dst, const u8 *src, int width, int h
 
 int ast_cursor_blit(struct ast_private *ast, struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = ast->dev;
+	struct drm_device *dev = &ast->base;
 	struct drm_gem_vram_object *gbo;
 	int ret;
 	void *src;
@@ -217,7 +217,7 @@ static void ast_cursor_set_base(struct ast_private *ast, u64 address)
 
 void ast_cursor_page_flip(struct ast_private *ast)
 {
-	struct drm_device *dev = ast->dev;
+	struct drm_device *dev = &ast->base;
 	struct drm_gem_vram_object *gbo;
 	s64 off;
 
@@ -253,7 +253,8 @@ void ast_cursor_show(struct ast_private *ast, int x, int y,
 		     unsigned int offset_x, unsigned int offset_y)
 {
 	u8 x_offset, y_offset;
-	u8 __iomem *dst, __iomem *sig;
+	u8 __iomem *dst;
+	u8 __iomem *sig;
 	u8 jreg;
 
 	dst = ast->cursor.vaddr[ast->cursor.next_index];
diff --git a/drivers/gpu/drm/ast/ast_dp501.c b/drivers/gpu/drm/ast/ast_dp501.c
index 4b85a504825a..88121c0e0d05 100644
--- a/drivers/gpu/drm/ast/ast_dp501.c
+++ b/drivers/gpu/drm/ast/ast_dp501.c
@@ -8,11 +8,24 @@
 
 MODULE_FIRMWARE("ast_dp501_fw.bin");
 
+static void ast_release_firmware(void *data)
+{
+	struct ast_private *ast = data;
+
+	release_firmware(ast->dp501_fw);
+	ast->dp501_fw = NULL;
+}
+
 static int ast_load_dp501_microcode(struct drm_device *dev)
 {
 	struct ast_private *ast = to_ast_private(dev);
+	int ret;
+
+	ret = request_firmware(&ast->dp501_fw, "ast_dp501_fw.bin", dev->dev);
+	if (ret)
+		return ret;
 
-	return request_firmware(&ast->dp501_fw, "ast_dp501_fw.bin", dev->dev);
+	return devm_add_action_or_reset(dev->dev, ast_release_firmware, ast);
 }
 
 static void send_ack(struct ast_private *ast)
@@ -435,11 +448,3 @@ void ast_init_3rdtx(struct drm_device *dev)
 		}
 	}
 }
-
-void ast_release_firmware(struct drm_device *dev)
-{
-	struct ast_private *ast = to_ast_private(dev);
-
-	release_firmware(ast->dp501_fw);
-	ast->dp501_fw = NULL;
-}
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 0b58f7aee6b0..f0b4af1c390a 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -43,9 +43,33 @@ int ast_modeset = -1;
 MODULE_PARM_DESC(modeset, "Disable/Enable modesetting");
 module_param_named(modeset, ast_modeset, int, 0400);
 
-#define PCI_VENDOR_ASPEED 0x1a03
+/*
+ * DRM driver
+ */
+
+DEFINE_DRM_GEM_FOPS(ast_fops);
+
+static struct drm_driver ast_driver = {
+	.driver_features = DRIVER_ATOMIC |
+			   DRIVER_GEM |
+			   DRIVER_MODESET,
+
+	.fops = &ast_fops,
+	.name = DRIVER_NAME,
+	.desc = DRIVER_DESC,
+	.date = DRIVER_DATE,
+	.major = DRIVER_MAJOR,
+	.minor = DRIVER_MINOR,
+	.patchlevel = DRIVER_PATCHLEVEL,
+
+	DRM_GEM_VRAM_DRIVER
+};
+
+/*
+ * PCI driver
+ */
 
-static struct drm_driver driver;
+#define PCI_VENDOR_ASPEED 0x1a03
 
 #define AST_VGA_DEVICE(id, info) {		\
 	.class = PCI_BASE_CLASS_DISPLAY << 16,	\
@@ -56,13 +80,13 @@ static struct drm_driver driver;
 	.subdevice = PCI_ANY_ID,		\
 	.driver_data = (unsigned long) info }
 
-static const struct pci_device_id pciidlist[] = {
+static const struct pci_device_id ast_pciidlist[] = {
 	AST_VGA_DEVICE(PCI_CHIP_AST2000, NULL),
 	AST_VGA_DEVICE(PCI_CHIP_AST2100, NULL),
 	{0, 0, 0},
 };
 
-MODULE_DEVICE_TABLE(pci, pciidlist);
+MODULE_DEVICE_TABLE(pci, ast_pciidlist);
 
 static void ast_kick_out_firmware_fb(struct pci_dev *pdev)
 {
@@ -85,6 +109,7 @@ static void ast_kick_out_firmware_fb(struct pci_dev *pdev)
 
 static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
+	struct ast_private *ast;
 	struct drm_device *dev;
 	int ret;
 
@@ -94,41 +119,25 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		return ret;
 
-	dev = drm_dev_alloc(&driver, &pdev->dev);
-	if (IS_ERR(dev))
-		return  PTR_ERR(dev);
-
-	dev->pdev = pdev;
-	pci_set_drvdata(pdev, dev);
-
-	ret = ast_driver_load(dev, ent->driver_data);
-	if (ret)
-		goto err_drm_dev_put;
+	ast = ast_device_create(&ast_driver, pdev, ent->driver_data);
+	if (IS_ERR(ast))
+		return PTR_ERR(ast);
+	dev = &ast->base;
 
 	ret = drm_dev_register(dev, ent->driver_data);
 	if (ret)
-		goto err_ast_driver_unload;
+		return ret;
 
 	drm_fbdev_generic_setup(dev, 32);
 
 	return 0;
-
-err_ast_driver_unload:
-	ast_driver_unload(dev);
-err_drm_dev_put:
-	drm_dev_put(dev);
-	return ret;
-
 }
 
-static void
-ast_pci_remove(struct pci_dev *pdev)
+static void ast_pci_remove(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 
 	drm_dev_unregister(dev);
-	ast_driver_unload(dev);
-	drm_dev_put(dev);
 }
 
 static int ast_drm_freeze(struct drm_device *dev)
@@ -217,30 +226,12 @@ static const struct dev_pm_ops ast_pm_ops = {
 
 static struct pci_driver ast_pci_driver = {
 	.name = DRIVER_NAME,
-	.id_table = pciidlist,
+	.id_table = ast_pciidlist,
 	.probe = ast_pci_probe,
 	.remove = ast_pci_remove,
 	.driver.pm = &ast_pm_ops,
 };
 
-DEFINE_DRM_GEM_FOPS(ast_fops);
-
-static struct drm_driver driver = {
-	.driver_features = DRIVER_ATOMIC |
-			   DRIVER_GEM |
-			   DRIVER_MODESET,
-
-	.fops = &ast_fops,
-	.name = DRIVER_NAME,
-	.desc = DRIVER_DESC,
-	.date = DRIVER_DATE,
-	.major = DRIVER_MAJOR,
-	.minor = DRIVER_MINOR,
-	.patchlevel = DRIVER_PATCHLEVEL,
-
-	DRM_GEM_VRAM_DRIVER
-};
-
 static int __init ast_init(void)
 {
 	if (vgacon_text_force() && ast_modeset == -1)
@@ -261,4 +252,3 @@ module_exit(ast_exit);
 MODULE_AUTHOR(DRIVER_AUTHOR);
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE("GPL and additional rights");
-
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index e3a264ac7ee2..467049ca8430 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -98,9 +98,25 @@ enum ast_tx_chip {
 #define AST_HWC_SIGNATURE_HOTSPOTX	0x14
 #define AST_HWC_SIGNATURE_HOTSPOTY	0x18
 
+struct ast_i2c_chan {
+	struct i2c_adapter adapter;
+	struct drm_device *dev;
+	struct i2c_algo_bit_data bit;
+};
+
+struct ast_connector {
+	struct drm_connector base;
+	struct ast_i2c_chan *i2c;
+};
+
+static inline struct ast_connector *
+to_ast_connector(struct drm_connector *connector)
+{
+	return container_of(connector, struct ast_connector, base);
+}
 
 struct ast_private {
-	struct drm_device *dev;
+	struct drm_device base;
 
 	void __iomem *regs;
 	void __iomem *ioregs;
@@ -119,9 +135,11 @@ struct ast_private {
 		unsigned int next_index;
 	} cursor;
 
-	struct drm_encoder encoder;
 	struct drm_plane primary_plane;
 	struct drm_plane cursor_plane;
+	struct drm_crtc crtc;
+	struct drm_encoder encoder;
+	struct ast_connector connector;
 
 	bool support_wide_screen;
 	enum {
@@ -138,11 +156,12 @@ struct ast_private {
 
 static inline struct ast_private *to_ast_private(struct drm_device *dev)
 {
-	return dev->dev_private;
+	return container_of(dev, struct ast_private, base);
 }
 
-int ast_driver_load(struct drm_device *dev, unsigned long flags);
-void ast_driver_unload(struct drm_device *dev);
+struct ast_private *ast_device_create(struct drm_driver *drv,
+				      struct pci_dev *pdev,
+				      unsigned long flags);
 
 #define AST_IO_AR_PORT_WRITE		(0x40)
 #define AST_IO_MISC_PORT_WRITE		(0x42)
@@ -158,6 +177,8 @@ void ast_driver_unload(struct drm_device *dev);
 
 #define AST_IO_MM_OFFSET		(0x380)
 
+#define AST_IO_VGAIR1_VREFRESH		BIT(3)
+
 #define __ast_read(x) \
 static inline u##x ast_read##x(struct ast_private *ast, u32 reg) { \
 u##x val = 0;\
@@ -226,19 +247,6 @@ static inline void ast_open_key(struct ast_private *ast)
 
 #define AST_VIDMEM_DEFAULT_SIZE AST_VIDMEM_SIZE_8M
 
-struct ast_i2c_chan {
-	struct i2c_adapter adapter;
-	struct drm_device *dev;
-	struct i2c_algo_bit_data bit;
-};
-
-struct ast_connector {
-	struct drm_connector base;
-	struct ast_i2c_chan *i2c;
-};
-
-#define to_ast_connector(x) container_of(x, struct ast_connector, base)
-
 struct ast_vbios_stdtable {
 	u8 misc;
 	u8 seq[4];
@@ -305,7 +313,6 @@ bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size);
 bool ast_dp501_read_edid(struct drm_device *dev, u8 *ediddata);
 u8 ast_get_dp501_max_clk(struct drm_device *dev);
 void ast_init_3rdtx(struct drm_device *dev);
-void ast_release_firmware(struct drm_device *dev);
 
 /* ast_cursor.c */
 int ast_cursor_init(struct ast_private *ast);
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 6a9fba051d13..77066bca8793 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -30,8 +30,10 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_gem_vram_helper.h>
+#include <drm/drm_managed.h>
 
 #include "ast_drv.h"
 
@@ -230,11 +232,11 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post)
 			ast->tx_chip_type = AST_TX_SIL164;
 			break;
 		case 0x08:
-			ast->dp501_fw_addr = kzalloc(32*1024, GFP_KERNEL);
+			ast->dp501_fw_addr = drmm_kzalloc(dev, 32*1024, GFP_KERNEL);
 			if (ast->dp501_fw_addr) {
 				/* backup firmware */
 				if (ast_backup_fw(dev, ast->dp501_fw_addr, 32*1024)) {
-					kfree(ast->dp501_fw_addr);
+					drmm_kfree(dev, ast->dp501_fw_addr);
 					ast->dp501_fw_addr = NULL;
 				}
 			}
@@ -378,24 +380,38 @@ static int ast_get_dram_info(struct drm_device *dev)
 	return 0;
 }
 
-int ast_driver_load(struct drm_device *dev, unsigned long flags)
+/*
+ * Run this function as part of the HW device cleanup; not
+ * when the DRM device gets released.
+ */
+static void ast_device_release(void *data)
 {
+	struct ast_private *ast = data;
+
+	/* enable standard VGA decode */
+	ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa1, 0x04);
+}
+
+struct ast_private *ast_device_create(struct drm_driver *drv,
+				      struct pci_dev *pdev,
+				      unsigned long flags)
+{
+	struct drm_device *dev;
 	struct ast_private *ast;
 	bool need_post;
 	int ret = 0;
 
-	ast = kzalloc(sizeof(struct ast_private), GFP_KERNEL);
-	if (!ast)
-		return -ENOMEM;
+	ast = devm_drm_dev_alloc(&pdev->dev, drv, struct ast_private, base);
+	if (IS_ERR(ast))
+		return ast;
+	dev = &ast->base;
 
-	dev->dev_private = ast;
-	ast->dev = dev;
+	dev->pdev = pdev;
+	pci_set_drvdata(pdev, dev);
 
 	ast->regs = pci_iomap(dev->pdev, 1, 0);
-	if (!ast->regs) {
-		ret = -EIO;
-		goto out_free;
-	}
+	if (!ast->regs)
+		return ERR_PTR(-EIO);
 
 	/*
 	 * If we don't have IO space at all, use MMIO now and
@@ -410,17 +426,16 @@ int ast_driver_load(struct drm_device *dev, unsigned long flags)
 	/* "map" IO regs if the above hasn't done so already */
 	if (!ast->ioregs) {
 		ast->ioregs = pci_iomap(dev->pdev, 2, 0);
-		if (!ast->ioregs) {
-			ret = -EIO;
-			goto out_free;
-		}
+		if (!ast->ioregs)
+			return ERR_PTR(-EIO);
 	}
 
 	ast_detect_chip(dev, &need_post);
 
 	ret = ast_get_dram_info(dev);
 	if (ret)
-		goto out_free;
+		return ERR_PTR(ret);
+
 	drm_info(dev, "dram MCLK=%u Mhz type=%d bus_width=%d\n",
 		 ast->mclk, ast->dram_type, ast->dram_bus_width);
 
@@ -429,28 +444,15 @@ int ast_driver_load(struct drm_device *dev, unsigned long flags)
 
 	ret = ast_mm_init(ast);
 	if (ret)
-		goto out_free;
+		return ERR_PTR(ret);
 
 	ret = ast_mode_config_init(ast);
 	if (ret)
-		goto out_free;
+		return ERR_PTR(ret);
 
-	return 0;
-out_free:
-	kfree(ast);
-	dev->dev_private = NULL;
-	return ret;
-}
-
-void ast_driver_unload(struct drm_device *dev)
-{
-	struct ast_private *ast = to_ast_private(dev);
-
-	/* enable standard VGA decode */
-	ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa1, 0x04);
-
-	ast_release_firmware(dev);
-	kfree(ast->dp501_fw_addr);
+	ret = devm_add_action_or_reset(dev->dev, ast_device_release, ast);
+	if (ret)
+		return ERR_PTR(ret);
 
-	kfree(ast);
+	return ast;
 }
diff --git a/drivers/gpu/drm/ast/ast_mm.c b/drivers/gpu/drm/ast/ast_mm.c
index 9186ec3ebbe0..8392ebde504b 100644
--- a/drivers/gpu/drm/ast/ast_mm.c
+++ b/drivers/gpu/drm/ast/ast_mm.c
@@ -85,9 +85,9 @@ static void ast_mm_release(struct drm_device *dev, void *ptr)
 
 int ast_mm_init(struct ast_private *ast)
 {
+	struct drm_device *dev = &ast->base;
 	u32 vram_size;
 	int ret;
-	struct drm_device *dev = ast->dev;
 
 	vram_size = ast_get_vram_size(ast);
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index 154cd877d9d1..834a156e3a75 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -514,6 +514,16 @@ static void ast_set_start_address_crt1(struct ast_private *ast,
 
 }
 
+static void ast_wait_for_vretrace(struct ast_private *ast)
+{
+	unsigned long timeout = jiffies + HZ;
+	u8 vgair1;
+
+	do {
+		vgair1 = ast_io_read8(ast, AST_IO_INPUT_STATUS1_READ);
+	} while (!(vgair1 & AST_IO_VGAIR1_VREFRESH) && time_before(jiffies, timeout));
+}
+
 /*
  * Primary plane
  */
@@ -562,13 +572,24 @@ ast_primary_plane_helper_atomic_update(struct drm_plane *plane,
 	struct drm_plane_state *state = plane->state;
 	struct drm_gem_vram_object *gbo;
 	s64 gpu_addr;
+	struct drm_framebuffer *fb = state->fb;
+	struct drm_framebuffer *old_fb = old_state->fb;
+
+	if (!old_fb || (fb->format != old_fb->format)) {
+		struct drm_crtc_state *crtc_state = state->crtc->state;
+		struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state);
+		struct ast_vbios_mode_info *vbios_mode_info = &ast_crtc_state->vbios_mode_info;
 
-	gbo = drm_gem_vram_of_gem(state->fb->obj[0]);
+		ast_set_color_reg(ast, fb->format);
+		ast_set_vbios_color_reg(ast, fb->format, vbios_mode_info);
+	}
+
+	gbo = drm_gem_vram_of_gem(fb->obj[0]);
 	gpu_addr = drm_gem_vram_offset(gbo);
 	if (drm_WARN_ON_ONCE(dev, gpu_addr < 0))
 		return; /* Bug: we didn't pin the BO to VRAM in prepare_fb. */
 
-	ast_set_offset_reg(ast, state->fb);
+	ast_set_offset_reg(ast, fb);
 	ast_set_start_address_crt1(ast, (u32)gpu_addr);
 
 	ast_set_index_reg_mask(ast, AST_IO_SEQ_PORT, 0x1, 0xdf, 0x00);
@@ -663,7 +684,7 @@ ast_cursor_plane_helper_atomic_update(struct drm_plane *plane,
 {
 	struct drm_plane_state *state = plane->state;
 	struct drm_framebuffer *fb = state->fb;
-	struct ast_private *ast = plane->dev->dev_private;
+	struct ast_private *ast = to_ast_private(plane->dev);
 	unsigned int offset_x, offset_y;
 
 	offset_x = AST_MAX_HWC_WIDTH - fb->width;
@@ -733,6 +754,7 @@ static void ast_crtc_dpms(struct drm_crtc *crtc, int mode)
 static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc,
 					struct drm_crtc_state *state)
 {
+	struct drm_device *dev = crtc->dev;
 	struct ast_crtc_state *ast_state;
 	const struct drm_format_info *format;
 	bool succ;
@@ -743,8 +765,8 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc,
 	ast_state = to_ast_crtc_state(state);
 
 	format = ast_state->format;
-	if (!format)
-		return 0;
+	if (drm_WARN_ON_ONCE(dev, !format))
+		return -EINVAL; /* BUG: We didn't set format in primary check(). */
 
 	succ = ast_get_vbios_mode_info(format, &state->mode,
 				       &state->adjusted_mode,
@@ -755,39 +777,17 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void ast_crtc_helper_atomic_begin(struct drm_crtc *crtc,
-					 struct drm_crtc_state *old_crtc_state)
-{
-	struct ast_private *ast = to_ast_private(crtc->dev);
-
-	ast_open_key(ast);
-}
-
-static void ast_crtc_helper_atomic_flush(struct drm_crtc *crtc,
-					 struct drm_crtc_state *old_crtc_state)
+static void
+ast_crtc_helper_atomic_enable(struct drm_crtc *crtc,
+			      struct drm_crtc_state *old_crtc_state)
 {
 	struct drm_device *dev = crtc->dev;
 	struct ast_private *ast = to_ast_private(dev);
-	struct ast_crtc_state *ast_state;
-	const struct drm_format_info *format;
-	struct ast_vbios_mode_info *vbios_mode_info;
-	struct drm_display_mode *adjusted_mode;
-
-	ast_state = to_ast_crtc_state(crtc->state);
-
-	format = ast_state->format;
-	if (!format)
-		return;
-
-	vbios_mode_info = &ast_state->vbios_mode_info;
-
-	ast_set_color_reg(ast, format);
-	ast_set_vbios_color_reg(ast, format, vbios_mode_info);
-
-	if (!crtc->state->mode_changed)
-		return;
-
-	adjusted_mode = &crtc->state->adjusted_mode;
+	struct drm_crtc_state *crtc_state = crtc->state;
+	struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state);
+	struct ast_vbios_mode_info *vbios_mode_info =
+		&ast_crtc_state->vbios_mode_info;
+	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 
 	ast_set_vbios_mode_reg(ast, adjusted_mode, vbios_mode_info);
 	ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa1, 0x06);
@@ -796,12 +796,7 @@ static void ast_crtc_helper_atomic_flush(struct drm_crtc *crtc,
 	ast_set_dclk_reg(ast, adjusted_mode, vbios_mode_info);
 	ast_set_crtthd_reg(ast);
 	ast_set_sync_reg(ast, adjusted_mode, vbios_mode_info);
-}
 
-static void
-ast_crtc_helper_atomic_enable(struct drm_crtc *crtc,
-			      struct drm_crtc_state *old_crtc_state)
-{
 	ast_crtc_dpms(crtc, DRM_MODE_DPMS_ON);
 }
 
@@ -809,13 +804,32 @@ static void
 ast_crtc_helper_atomic_disable(struct drm_crtc *crtc,
 			       struct drm_crtc_state *old_crtc_state)
 {
+	struct drm_device *dev = crtc->dev;
+	struct ast_private *ast = to_ast_private(dev);
+
 	ast_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+
+	/*
+	 * HW cursors require the underlying primary plane and CRTC to
+	 * display a valid mode and image. This is not the case during
+	 * full modeset operations. So we temporarily disable any active
+	 * plane, including the HW cursor. Each plane's atomic_update()
+	 * helper will re-enable it if necessary.
+	 *
+	 * We only do this during *full* modesets. It does not affect
+	 * simple pageflips on the planes.
+	 */
+	drm_atomic_helper_disable_planes_on_crtc(old_crtc_state, false);
+
+	/*
+	 * Ensure that no scanout takes place before reprogramming mode
+	 * and format registers.
+	 */
+	ast_wait_for_vretrace(ast);
 }
 
 static const struct drm_crtc_helper_funcs ast_crtc_helper_funcs = {
 	.atomic_check = ast_crtc_helper_atomic_check,
-	.atomic_begin = ast_crtc_helper_atomic_begin,
-	.atomic_flush = ast_crtc_helper_atomic_flush,
 	.atomic_enable = ast_crtc_helper_atomic_enable,
 	.atomic_disable = ast_crtc_helper_atomic_disable,
 };
@@ -831,12 +845,6 @@ static void ast_crtc_reset(struct drm_crtc *crtc)
 	__drm_atomic_helper_crtc_reset(crtc, &ast_state->base);
 }
 
-static void ast_crtc_destroy(struct drm_crtc *crtc)
-{
-	drm_crtc_cleanup(crtc);
-	kfree(crtc);
-}
-
 static struct drm_crtc_state *
 ast_crtc_atomic_duplicate_state(struct drm_crtc *crtc)
 {
@@ -872,7 +880,7 @@ static void ast_crtc_atomic_destroy_state(struct drm_crtc *crtc,
 static const struct drm_crtc_funcs ast_crtc_funcs = {
 	.reset = ast_crtc_reset,
 	.gamma_set = drm_atomic_helper_legacy_gamma_set,
-	.destroy = ast_crtc_destroy,
+	.destroy = drm_crtc_cleanup,
 	.set_config = drm_atomic_helper_set_config,
 	.page_flip = drm_atomic_helper_page_flip,
 	.atomic_duplicate_state = ast_crtc_atomic_duplicate_state,
@@ -882,27 +890,19 @@ static const struct drm_crtc_funcs ast_crtc_funcs = {
 static int ast_crtc_init(struct drm_device *dev)
 {
 	struct ast_private *ast = to_ast_private(dev);
-	struct drm_crtc *crtc;
+	struct drm_crtc *crtc = &ast->crtc;
 	int ret;
 
-	crtc = kzalloc(sizeof(*crtc), GFP_KERNEL);
-	if (!crtc)
-		return -ENOMEM;
-
 	ret = drm_crtc_init_with_planes(dev, crtc, &ast->primary_plane,
 					&ast->cursor_plane, &ast_crtc_funcs,
 					NULL);
 	if (ret)
-		goto err_kfree;
+		return ret;
 
 	drm_mode_crtc_set_gamma_size(crtc, 256);
 	drm_crtc_helper_add(crtc, &ast_crtc_helper_funcs);
 
 	return 0;
-
-err_kfree:
-	kfree(crtc);
-	return ret;
 }
 
 /*
@@ -1021,7 +1021,6 @@ static void ast_connector_destroy(struct drm_connector *connector)
 	struct ast_connector *ast_connector = to_ast_connector(connector);
 	ast_i2c_destroy(ast_connector->i2c);
 	drm_connector_cleanup(connector);
-	kfree(connector);
 }
 
 static const struct drm_connector_helper_funcs ast_connector_helper_funcs = {
@@ -1039,15 +1038,11 @@ static const struct drm_connector_funcs ast_connector_funcs = {
 
 static int ast_connector_init(struct drm_device *dev)
 {
-	struct ast_connector *ast_connector;
-	struct drm_connector *connector;
-	struct drm_encoder *encoder;
-
-	ast_connector = kzalloc(sizeof(struct ast_connector), GFP_KERNEL);
-	if (!ast_connector)
-		return -ENOMEM;
+	struct ast_private *ast = to_ast_private(dev);
+	struct ast_connector *ast_connector = &ast->connector;
+	struct drm_connector *connector = &ast_connector->base;
+	struct drm_encoder *encoder = &ast->encoder;
 
-	connector = &ast_connector->base;
 	ast_connector->i2c = ast_i2c_create(dev);
 	if (!ast_connector->i2c)
 		drm_err(dev, "failed to add ddc bus for connector\n");
@@ -1064,7 +1059,6 @@ static int ast_connector_init(struct drm_device *dev)
 
 	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
 
-	encoder = list_first_entry(&dev->mode_config.encoder_list, struct drm_encoder, head);
 	drm_connector_attach_encoder(connector, encoder);
 
 	return 0;
@@ -1074,6 +1068,11 @@ static int ast_connector_init(struct drm_device *dev)
  * Mode config
  */
 
+static const struct drm_mode_config_helper_funcs
+ast_mode_config_helper_funcs = {
+	.atomic_commit_tail = drm_atomic_helper_commit_tail_rpm,
+};
+
 static const struct drm_mode_config_funcs ast_mode_config_funcs = {
 	.fb_create = drm_gem_fb_create,
 	.mode_valid = drm_vram_helper_mode_valid,
@@ -1083,7 +1082,7 @@ static const struct drm_mode_config_funcs ast_mode_config_funcs = {
 
 int ast_mode_config_init(struct ast_private *ast)
 {
-	struct drm_device *dev = ast->dev;
+	struct drm_device *dev = &ast->base;
 	int ret;
 
 	ret = ast_cursor_init(ast);
@@ -1099,7 +1098,7 @@ int ast_mode_config_init(struct ast_private *ast)
 	dev->mode_config.min_height = 0;
 	dev->mode_config.preferred_depth = 24;
 	dev->mode_config.prefer_shadow = 1;
-	dev->mode_config.fb_base = pci_resource_start(ast->dev->pdev, 0);
+	dev->mode_config.fb_base = pci_resource_start(dev->pdev, 0);
 
 	if (ast->chip == AST2100 ||
 	    ast->chip == AST2200 ||
@@ -1113,6 +1112,8 @@ int ast_mode_config_init(struct ast_private *ast)
 		dev->mode_config.max_height = 1200;
 	}
 
+	dev->mode_config.helper_private = &ast_mode_config_helper_funcs;
+
 	memset(&ast->primary_plane, 0, sizeof(ast->primary_plane));
 	ret = drm_universal_plane_init(dev, &ast->primary_plane, 0x01,
 				       &ast_primary_plane_funcs,
diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c
index c043fe717553..8902c2f84bf9 100644
--- a/drivers/gpu/drm/ast/ast_post.c
+++ b/drivers/gpu/drm/ast/ast_post.c
@@ -365,12 +365,12 @@ static void ast_init_dram_reg(struct drm_device *dev)
 
 void ast_post_gpu(struct drm_device *dev)
 {
-	u32 reg;
 	struct ast_private *ast = to_ast_private(dev);
+	u32 reg;
 
-	pci_read_config_dword(ast->dev->pdev, 0x04, &reg);
+	pci_read_config_dword(dev->pdev, 0x04, &reg);
 	reg |= 0x3;
-	pci_write_config_dword(ast->dev->pdev, 0x04, reg);
+	pci_write_config_dword(dev->pdev, 0x04, reg);
 
 	ast_enable_vga(dev);
 	ast_open_key(ast);
diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 43271c21d3fc..ef91646441b1 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -48,6 +48,19 @@ config DRM_DISPLAY_CONNECTOR
 	  on ARM-based platforms. Saying Y here when this driver is not needed
 	  will not cause any issue.
 
+config DRM_LONTIUM_LT9611
+	tristate "Lontium LT9611 DSI/HDMI bridge"
+	select SND_SOC_HDMI_CODEC if SND_SOC
+	depends on OF
+	select DRM_PANEL_BRIDGE
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	help
+	  Driver for Lontium LT9611 DSI to HDMI bridge
+	  chip driver that converts dual DSI and I2S to
+	  HDMI signals
+	  Please say Y if you have such hardware.
+
 config DRM_LVDS_CODEC
 	tristate "Transparent LVDS encoders and decoders support"
 	depends on OF
@@ -153,6 +166,14 @@ config DRM_THINE_THC63LVD1024
 	help
 	  Thine THC63LVD1024 LVDS/parallel converter driver.
 
+config DRM_TOSHIBA_TC358762
+	tristate "TC358762 DSI/DPI bridge"
+	depends on OF
+	select DRM_MIPI_DSI
+	select DRM_PANEL_BRIDGE
+	help
+	  Toshiba TC358762 DSI/DPI bridge driver.
+
 config DRM_TOSHIBA_TC358764
 	tristate "TC358764 DSI/LVDS bridge"
 	depends on OF
@@ -181,6 +202,16 @@ config DRM_TOSHIBA_TC358768
 	help
 	  Toshiba TC358768AXBG/TC358778XBG DSI bridge chip driver.
 
+config DRM_TOSHIBA_TC358775
+	tristate "Toshiba TC358775 DSI/LVDS bridge"
+	depends on OF
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	select DRM_PANEL
+	select DRM_MIPI_DSI
+	help
+	  Toshiba TC358775 DSI/LVDS bridge chip driver.
+
 config DRM_TI_TFP410
 	tristate "TI TFP410 DVI/HDMI bridge"
 	depends on OF
@@ -210,6 +241,8 @@ source "drivers/gpu/drm/bridge/analogix/Kconfig"
 
 source "drivers/gpu/drm/bridge/adv7511/Kconfig"
 
+source "drivers/gpu/drm/bridge/cadence/Kconfig"
+
 source "drivers/gpu/drm/bridge/synopsys/Kconfig"
 
 endmenu
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index d63d4b7e4347..2b3aff104e46 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -2,6 +2,7 @@
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
 obj-$(CONFIG_DRM_DISPLAY_CONNECTOR) += display-connector.o
+obj-$(CONFIG_DRM_LONTIUM_LT9611) += lontium-lt9611.o
 obj-$(CONFIG_DRM_LVDS_CODEC) += lvds-codec.o
 obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
@@ -12,9 +13,11 @@ obj-$(CONFIG_DRM_SII902X) += sii902x.o
 obj-$(CONFIG_DRM_SII9234) += sii9234.o
 obj-$(CONFIG_DRM_SIMPLE_BRIDGE) += simple-bridge.o
 obj-$(CONFIG_DRM_THINE_THC63LVD1024) += thc63lvd1024.o
+obj-$(CONFIG_DRM_TOSHIBA_TC358762) += tc358762.o
 obj-$(CONFIG_DRM_TOSHIBA_TC358764) += tc358764.o
 obj-$(CONFIG_DRM_TOSHIBA_TC358767) += tc358767.o
 obj-$(CONFIG_DRM_TOSHIBA_TC358768) += tc358768.o
+obj-$(CONFIG_DRM_TOSHIBA_TC358775) += tc358775.o
 obj-$(CONFIG_DRM_I2C_ADV7511) += adv7511/
 obj-$(CONFIG_DRM_TI_SN65DSI86) += ti-sn65dsi86.o
 obj-$(CONFIG_DRM_TI_TFP410) += ti-tfp410.o
@@ -22,4 +25,5 @@ obj-$(CONFIG_DRM_TI_TPD12S015) += ti-tpd12s015.o
 obj-$(CONFIG_DRM_NWL_MIPI_DSI) += nwl-dsi.o
 
 obj-y += analogix/
+obj-y += cadence/
 obj-y += synopsys/
diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
index f082b4ed4878..d9164fab044d 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
@@ -507,10 +507,6 @@ static const struct drm_connector_helper_funcs anx6345_connector_helper_funcs =
 static void
 anx6345_connector_destroy(struct drm_connector *connector)
 {
-	struct anx6345 *anx6345 = connector_to_anx6345(connector);
-
-	if (anx6345->panel)
-		drm_panel_detach(anx6345->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -575,14 +571,6 @@ static int anx6345_bridge_attach(struct drm_bridge *bridge,
 		return err;
 	}
 
-	if (anx6345->panel) {
-		err = drm_panel_attach(anx6345->panel, &anx6345->connector);
-		if (err) {
-			DRM_ERROR("Failed to attach panel: %d\n", err);
-			return err;
-		}
-	}
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
index 76736fb8ed94..aa1bb86293fd 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
@@ -1265,14 +1265,6 @@ static int analogix_dp_bridge_attach(struct drm_bridge *bridge,
 		}
 	}
 
-	if (dp->plat_data->panel) {
-		ret = drm_panel_attach(dp->plat_data->panel, &dp->connector);
-		if (ret) {
-			DRM_ERROR("Failed to attach panel\n");
-			return ret;
-		}
-	}
-
 	return 0;
 }
 
@@ -1803,7 +1795,6 @@ void analogix_dp_unbind(struct analogix_dp_device *dp)
 	if (dp->plat_data->panel) {
 		if (drm_panel_unprepare(dp->plat_data->panel))
 			DRM_ERROR("failed to turnoff the panel\n");
-		drm_panel_detach(dp->plat_data->panel);
 	}
 
 	drm_dp_aux_unregister(&dp->aux);
diff --git a/drivers/gpu/drm/bridge/cadence/Kconfig b/drivers/gpu/drm/bridge/cadence/Kconfig
new file mode 100644
index 000000000000..511d67b16d14
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/Kconfig
@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config DRM_CDNS_MHDP8546
+	tristate "Cadence DPI/DP bridge"
+	select DRM_KMS_HELPER
+	select DRM_PANEL_BRIDGE
+	depends on OF
+	help
+	  Support Cadence DPI to DP bridge. This is an internal
+	  bridge and is meant to be directly embedded in a SoC.
+	  It takes a DPI stream as input and outputs it encoded
+	  in DP format.
+
+if DRM_CDNS_MHDP8546
+
+config DRM_CDNS_MHDP8546_J721E
+	depends on ARCH_K3_J721E_SOC || COMPILE_TEST
+	bool "J721E Cadence DPI/DP wrapper support"
+	default y
+	help
+	  Support J721E Cadence DPI/DP wrapper. This is a wrapper
+	  which adds support for J721E related platform ops. It
+	  initializes the J721E Display Port and sets up the
+	  clock and data muxes.
+endif
diff --git a/drivers/gpu/drm/bridge/cadence/Makefile b/drivers/gpu/drm/bridge/cadence/Makefile
new file mode 100644
index 000000000000..8f647991b374
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_DRM_CDNS_MHDP8546) += cdns-mhdp8546.o
+cdns-mhdp8546-y := cdns-mhdp8546-core.o
+cdns-mhdp8546-$(CONFIG_DRM_CDNS_MHDP8546_J721E) += cdns-mhdp8546-j721e.o
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
new file mode 100644
index 000000000000..d0c65610ebb5
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
@@ -0,0 +1,2532 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Cadence MHDP8546 DP bridge driver.
+ *
+ * Copyright (C) 2020 Cadence Design Systems, Inc.
+ *
+ * Authors: Quentin Schulz <quentin.schulz@free-electrons.com>
+ *          Swapnil Jakhade <sjakhade@cadence.com>
+ *          Yuti Amonkar <yamonkar@cadence.com>
+ *          Tomi Valkeinen <tomi.valkeinen@ti.com>
+ *          Jyri Sarha <jsarha@ti.com>
+ *
+ * TODO:
+ *     - Implement optimized mailbox communication using mailbox interrupts
+ *     - Add support for power management
+ *     - Add support for features like audio, MST and fast link training
+ *     - Implement request_fw_cancel to handle HW_STATE
+ *     - Fix asynchronous loading of firmware implementation
+ *     - Add DRM helper function for cdns_mhdp_lower_link_rate
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/firmware.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/irq.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/phy/phy.h>
+#include <linux/phy/phy-dp.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/wait.h>
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+
+#include <asm/unaligned.h>
+
+#include "cdns-mhdp8546-core.h"
+
+#include "cdns-mhdp8546-j721e.h"
+
+static int cdns_mhdp_mailbox_read(struct cdns_mhdp_device *mhdp)
+{
+	int ret, empty;
+
+	WARN_ON(!mutex_is_locked(&mhdp->mbox_mutex));
+
+	ret = readx_poll_timeout(readl, mhdp->regs + CDNS_MAILBOX_EMPTY,
+				 empty, !empty, MAILBOX_RETRY_US,
+				 MAILBOX_TIMEOUT_US);
+	if (ret < 0)
+		return ret;
+
+	return readl(mhdp->regs + CDNS_MAILBOX_RX_DATA) & 0xff;
+}
+
+static int cdns_mhdp_mailbox_write(struct cdns_mhdp_device *mhdp, u8 val)
+{
+	int ret, full;
+
+	WARN_ON(!mutex_is_locked(&mhdp->mbox_mutex));
+
+	ret = readx_poll_timeout(readl, mhdp->regs + CDNS_MAILBOX_FULL,
+				 full, !full, MAILBOX_RETRY_US,
+				 MAILBOX_TIMEOUT_US);
+	if (ret < 0)
+		return ret;
+
+	writel(val, mhdp->regs + CDNS_MAILBOX_TX_DATA);
+
+	return 0;
+}
+
+static int cdns_mhdp_mailbox_recv_header(struct cdns_mhdp_device *mhdp,
+					 u8 module_id, u8 opcode,
+					 u16 req_size)
+{
+	u32 mbox_size, i;
+	u8 header[4];
+	int ret;
+
+	/* read the header of the message */
+	for (i = 0; i < sizeof(header); i++) {
+		ret = cdns_mhdp_mailbox_read(mhdp);
+		if (ret < 0)
+			return ret;
+
+		header[i] = ret;
+	}
+
+	mbox_size = get_unaligned_be16(header + 2);
+
+	if (opcode != header[0] || module_id != header[1] ||
+	    req_size != mbox_size) {
+		/*
+		 * If the message in mailbox is not what we want, we need to
+		 * clear the mailbox by reading its contents.
+		 */
+		for (i = 0; i < mbox_size; i++)
+			if (cdns_mhdp_mailbox_read(mhdp) < 0)
+				break;
+
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int cdns_mhdp_mailbox_recv_data(struct cdns_mhdp_device *mhdp,
+				       u8 *buff, u16 buff_size)
+{
+	u32 i;
+	int ret;
+
+	for (i = 0; i < buff_size; i++) {
+		ret = cdns_mhdp_mailbox_read(mhdp);
+		if (ret < 0)
+			return ret;
+
+		buff[i] = ret;
+	}
+
+	return 0;
+}
+
+static int cdns_mhdp_mailbox_send(struct cdns_mhdp_device *mhdp, u8 module_id,
+				  u8 opcode, u16 size, u8 *message)
+{
+	u8 header[4];
+	int ret, i;
+
+	header[0] = opcode;
+	header[1] = module_id;
+	put_unaligned_be16(size, header + 2);
+
+	for (i = 0; i < sizeof(header); i++) {
+		ret = cdns_mhdp_mailbox_write(mhdp, header[i]);
+		if (ret)
+			return ret;
+	}
+
+	for (i = 0; i < size; i++) {
+		ret = cdns_mhdp_mailbox_write(mhdp, message[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static
+int cdns_mhdp_reg_read(struct cdns_mhdp_device *mhdp, u32 addr, u32 *value)
+{
+	u8 msg[4], resp[8];
+	int ret;
+
+	put_unaligned_be32(addr, msg);
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_GENERAL,
+				     GENERAL_REGISTER_READ,
+				     sizeof(msg), msg);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_GENERAL,
+					    GENERAL_REGISTER_READ,
+					    sizeof(resp));
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, resp, sizeof(resp));
+	if (ret)
+		goto out;
+
+	/* Returned address value should be the same as requested */
+	if (memcmp(msg, resp, sizeof(msg))) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	*value = get_unaligned_be32(resp + 4);
+
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+	if (ret) {
+		dev_err(mhdp->dev, "Failed to read register\n");
+		*value = 0;
+	}
+
+	return ret;
+}
+
+static
+int cdns_mhdp_reg_write(struct cdns_mhdp_device *mhdp, u16 addr, u32 val)
+{
+	u8 msg[6];
+	int ret;
+
+	put_unaligned_be16(addr, msg);
+	put_unaligned_be32(val, msg + 2);
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_WRITE_REGISTER, sizeof(msg), msg);
+
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	return ret;
+}
+
+static
+int cdns_mhdp_reg_write_bit(struct cdns_mhdp_device *mhdp, u16 addr,
+			    u8 start_bit, u8 bits_no, u32 val)
+{
+	u8 field[8];
+	int ret;
+
+	put_unaligned_be16(addr, field);
+	field[2] = start_bit;
+	field[3] = bits_no;
+	put_unaligned_be32(val, field + 4);
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_WRITE_FIELD, sizeof(field), field);
+
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	return ret;
+}
+
+static
+int cdns_mhdp_dpcd_read(struct cdns_mhdp_device *mhdp,
+			u32 addr, u8 *data, u16 len)
+{
+	u8 msg[5], reg[5];
+	int ret;
+
+	put_unaligned_be16(len, msg);
+	put_unaligned_be24(addr, msg + 2);
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_READ_DPCD, sizeof(msg), msg);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+					    DPTX_READ_DPCD,
+					    sizeof(reg) + len);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, reg, sizeof(reg));
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, data, len);
+
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	return ret;
+}
+
+static
+int cdns_mhdp_dpcd_write(struct cdns_mhdp_device *mhdp, u32 addr, u8 value)
+{
+	u8 msg[6], reg[5];
+	int ret;
+
+	put_unaligned_be16(1, msg);
+	put_unaligned_be24(addr, msg + 2);
+	msg[5] = value;
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_WRITE_DPCD, sizeof(msg), msg);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+					    DPTX_WRITE_DPCD, sizeof(reg));
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, reg, sizeof(reg));
+	if (ret)
+		goto out;
+
+	if (addr != get_unaligned_be24(reg + 2))
+		ret = -EINVAL;
+
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	if (ret)
+		dev_err(mhdp->dev, "dpcd write failed: %d\n", ret);
+	return ret;
+}
+
+static
+int cdns_mhdp_set_firmware_active(struct cdns_mhdp_device *mhdp, bool enable)
+{
+	u8 msg[5];
+	int ret, i;
+
+	msg[0] = GENERAL_MAIN_CONTROL;
+	msg[1] = MB_MODULE_ID_GENERAL;
+	msg[2] = 0;
+	msg[3] = 1;
+	msg[4] = enable ? FW_ACTIVE : FW_STANDBY;
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	for (i = 0; i < sizeof(msg); i++) {
+		ret = cdns_mhdp_mailbox_write(mhdp, msg[i]);
+		if (ret)
+			goto out;
+	}
+
+	/* read the firmware state */
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, msg, sizeof(msg));
+	if (ret)
+		goto out;
+
+	ret = 0;
+
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	if (ret < 0)
+		dev_err(mhdp->dev, "set firmware active failed\n");
+	return ret;
+}
+
+static
+int cdns_mhdp_get_hpd_status(struct cdns_mhdp_device *mhdp)
+{
+	u8 status;
+	int ret;
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_HPD_STATE, 0, NULL);
+	if (ret)
+		goto err_get_hpd;
+
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+					    DPTX_HPD_STATE,
+					    sizeof(status));
+	if (ret)
+		goto err_get_hpd;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, &status, sizeof(status));
+	if (ret)
+		goto err_get_hpd;
+
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	dev_dbg(mhdp->dev, "%s: HPD %splugged\n", __func__,
+		status ? "" : "un");
+
+	return status;
+
+err_get_hpd:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	return ret;
+}
+
+static
+int cdns_mhdp_get_edid_block(void *data, u8 *edid,
+			     unsigned int block, size_t length)
+{
+	struct cdns_mhdp_device *mhdp = data;
+	u8 msg[2], reg[2], i;
+	int ret;
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	for (i = 0; i < 4; i++) {
+		msg[0] = block / 2;
+		msg[1] = block % 2;
+
+		ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+					     DPTX_GET_EDID, sizeof(msg), msg);
+		if (ret)
+			continue;
+
+		ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+						    DPTX_GET_EDID,
+						    sizeof(reg) + length);
+		if (ret)
+			continue;
+
+		ret = cdns_mhdp_mailbox_recv_data(mhdp, reg, sizeof(reg));
+		if (ret)
+			continue;
+
+		ret = cdns_mhdp_mailbox_recv_data(mhdp, edid, length);
+		if (ret)
+			continue;
+
+		if (reg[0] == length && reg[1] == block / 2)
+			break;
+	}
+
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	if (ret)
+		dev_err(mhdp->dev, "get block[%d] edid failed: %d\n",
+			block, ret);
+
+	return ret;
+}
+
+static
+int cdns_mhdp_read_hpd_event(struct cdns_mhdp_device *mhdp)
+{
+	u8 event = 0;
+	int ret;
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_READ_EVENT, 0, NULL);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+					    DPTX_READ_EVENT, sizeof(event));
+	if (ret < 0)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, &event, sizeof(event));
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	if (ret < 0)
+		return ret;
+
+	dev_dbg(mhdp->dev, "%s: %s%s%s%s\n", __func__,
+		(event & DPTX_READ_EVENT_HPD_TO_HIGH) ? "TO_HIGH " : "",
+		(event & DPTX_READ_EVENT_HPD_TO_LOW) ? "TO_LOW " : "",
+		(event & DPTX_READ_EVENT_HPD_PULSE) ? "PULSE " : "",
+		(event & DPTX_READ_EVENT_HPD_STATE) ? "HPD_STATE " : "");
+
+	return event;
+}
+
+static
+int cdns_mhdp_adjust_lt(struct cdns_mhdp_device *mhdp, unsigned int nlanes,
+			unsigned int udelay, const u8 *lanes_data,
+			u8 link_status[DP_LINK_STATUS_SIZE])
+{
+	u8 payload[7];
+	u8 hdr[5]; /* For DPCD read response header */
+	u32 addr;
+	int ret;
+
+	if (nlanes != 4 && nlanes != 2 && nlanes != 1) {
+		dev_err(mhdp->dev, "invalid number of lanes: %u\n", nlanes);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	payload[0] = nlanes;
+	put_unaligned_be16(udelay, payload + 1);
+	memcpy(payload + 3, lanes_data, nlanes);
+
+	mutex_lock(&mhdp->mbox_mutex);
+
+	ret = cdns_mhdp_mailbox_send(mhdp, MB_MODULE_ID_DP_TX,
+				     DPTX_ADJUST_LT,
+				     sizeof(payload), payload);
+	if (ret)
+		goto out;
+
+	/* Yes, read the DPCD read command response */
+	ret = cdns_mhdp_mailbox_recv_header(mhdp, MB_MODULE_ID_DP_TX,
+					    DPTX_READ_DPCD,
+					    sizeof(hdr) + DP_LINK_STATUS_SIZE);
+	if (ret)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, hdr, sizeof(hdr));
+	if (ret)
+		goto out;
+
+	addr = get_unaligned_be24(hdr + 2);
+	if (addr != DP_LANE0_1_STATUS)
+		goto out;
+
+	ret = cdns_mhdp_mailbox_recv_data(mhdp, link_status,
+					  DP_LINK_STATUS_SIZE);
+
+out:
+	mutex_unlock(&mhdp->mbox_mutex);
+
+	if (ret)
+		dev_err(mhdp->dev, "Failed to adjust Link Training.\n");
+
+	return ret;
+}
+
+/**
+ * cdns_mhdp_link_power_up() - power up a DisplayPort link
+ * @aux: DisplayPort AUX channel
+ * @link: pointer to a structure containing the link configuration
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+static
+int cdns_mhdp_link_power_up(struct drm_dp_aux *aux, struct cdns_mhdp_link *link)
+{
+	u8 value;
+	int err;
+
+	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
+	if (link->revision < 0x11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D0;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	/*
+	 * According to the DP 1.1 specification, a "Sink Device must exit the
+	 * power saving state within 1 ms" (Section 2.5.3.1, Table 5-52, "Sink
+	 * Control Field" (register 0x600).
+	 */
+	usleep_range(1000, 2000);
+
+	return 0;
+}
+
+/**
+ * cdns_mhdp_link_power_down() - power down a DisplayPort link
+ * @aux: DisplayPort AUX channel
+ * @link: pointer to a structure containing the link configuration
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+static
+int cdns_mhdp_link_power_down(struct drm_dp_aux *aux,
+			      struct cdns_mhdp_link *link)
+{
+	u8 value;
+	int err;
+
+	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
+	if (link->revision < 0x11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D3;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+/**
+ * cdns_mhdp_link_configure() - configure a DisplayPort link
+ * @aux: DisplayPort AUX channel
+ * @link: pointer to a structure containing the link configuration
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+static
+int cdns_mhdp_link_configure(struct drm_dp_aux *aux,
+			     struct cdns_mhdp_link *link)
+{
+	u8 values[2];
+	int err;
+
+	values[0] = drm_dp_link_rate_to_bw_code(link->rate);
+	values[1] = link->num_lanes;
+
+	if (link->capabilities & DP_LINK_CAP_ENHANCED_FRAMING)
+		values[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+
+	err = drm_dp_dpcd_write(aux, DP_LINK_BW_SET, values, sizeof(values));
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static unsigned int cdns_mhdp_max_link_rate(struct cdns_mhdp_device *mhdp)
+{
+	return min(mhdp->host.link_rate, mhdp->sink.link_rate);
+}
+
+static u8 cdns_mhdp_max_num_lanes(struct cdns_mhdp_device *mhdp)
+{
+	return min(mhdp->sink.lanes_cnt, mhdp->host.lanes_cnt);
+}
+
+static u8 cdns_mhdp_eq_training_pattern_supported(struct cdns_mhdp_device *mhdp)
+{
+	return fls(mhdp->host.pattern_supp & mhdp->sink.pattern_supp);
+}
+
+static bool cdns_mhdp_get_ssc_supported(struct cdns_mhdp_device *mhdp)
+{
+	/* Check if SSC is supported by both sides */
+	return mhdp->host.ssc && mhdp->sink.ssc;
+}
+
+static enum drm_connector_status cdns_mhdp_detect(struct cdns_mhdp_device *mhdp)
+{
+	dev_dbg(mhdp->dev, "%s: %d\n", __func__, mhdp->plugged);
+
+	if (mhdp->plugged)
+		return connector_status_connected;
+	else
+		return connector_status_disconnected;
+}
+
+static int cdns_mhdp_check_fw_version(struct cdns_mhdp_device *mhdp)
+{
+	u32 major_num, minor_num, revision;
+	u32 fw_ver, lib_ver;
+
+	fw_ver = (readl(mhdp->regs + CDNS_VER_H) << 8)
+	       | readl(mhdp->regs + CDNS_VER_L);
+
+	lib_ver = (readl(mhdp->regs + CDNS_LIB_H_ADDR) << 8)
+		| readl(mhdp->regs + CDNS_LIB_L_ADDR);
+
+	if (lib_ver < 33984) {
+		/*
+		 * Older FW versions with major number 1, used to store FW
+		 * version information by storing repository revision number
+		 * in registers. This is for identifying these FW versions.
+		 */
+		major_num = 1;
+		minor_num = 2;
+		if (fw_ver == 26098) {
+			revision = 15;
+		} else if (lib_ver == 0 && fw_ver == 0) {
+			revision = 17;
+		} else {
+			dev_err(mhdp->dev, "Unsupported FW version: fw_ver = %u, lib_ver = %u\n",
+				fw_ver, lib_ver);
+			return -ENODEV;
+		}
+	} else {
+		/* To identify newer FW versions with major number 2 onwards. */
+		major_num = fw_ver / 10000;
+		minor_num = (fw_ver / 100) % 100;
+		revision = (fw_ver % 10000) % 100;
+	}
+
+	dev_dbg(mhdp->dev, "FW version: v%u.%u.%u\n", major_num, minor_num,
+		revision);
+	return 0;
+}
+
+static int cdns_mhdp_fw_activate(const struct firmware *fw,
+				 struct cdns_mhdp_device *mhdp)
+{
+	unsigned int reg;
+	int ret;
+
+	/* Release uCPU reset and stall it. */
+	writel(CDNS_CPU_STALL, mhdp->regs + CDNS_APB_CTRL);
+
+	memcpy_toio(mhdp->regs + CDNS_MHDP_IMEM, fw->data, fw->size);
+
+	/* Leave debug mode, release stall */
+	writel(0, mhdp->regs + CDNS_APB_CTRL);
+
+	/*
+	 * Wait for the KEEP_ALIVE "message" on the first 8 bits.
+	 * Updated each sched "tick" (~2ms)
+	 */
+	ret = readl_poll_timeout(mhdp->regs + CDNS_KEEP_ALIVE, reg,
+				 reg & CDNS_KEEP_ALIVE_MASK, 500,
+				 CDNS_KEEP_ALIVE_TIMEOUT);
+	if (ret) {
+		dev_err(mhdp->dev,
+			"device didn't give any life sign: reg %d\n", reg);
+		return ret;
+	}
+
+	ret = cdns_mhdp_check_fw_version(mhdp);
+	if (ret)
+		return ret;
+
+	/* Init events to 0 as it's not cleared by FW at boot but on read */
+	readl(mhdp->regs + CDNS_SW_EVENT0);
+	readl(mhdp->regs + CDNS_SW_EVENT1);
+	readl(mhdp->regs + CDNS_SW_EVENT2);
+	readl(mhdp->regs + CDNS_SW_EVENT3);
+
+	/* Activate uCPU */
+	ret = cdns_mhdp_set_firmware_active(mhdp, true);
+	if (ret)
+		return ret;
+
+	spin_lock(&mhdp->start_lock);
+
+	mhdp->hw_state = MHDP_HW_READY;
+
+	/*
+	 * Here we must keep the lock while enabling the interrupts
+	 * since it would otherwise be possible that interrupt enable
+	 * code is executed after the bridge is detached. The similar
+	 * situation is not possible in attach()/detach() callbacks
+	 * since the hw_state changes from MHDP_HW_READY to
+	 * MHDP_HW_STOPPED happens only due to driver removal when
+	 * bridge should already be detached.
+	 */
+	if (mhdp->bridge_attached)
+		writel(~(u32)CDNS_APB_INT_MASK_SW_EVENT_INT,
+		       mhdp->regs + CDNS_APB_INT_MASK);
+
+	spin_unlock(&mhdp->start_lock);
+
+	wake_up(&mhdp->fw_load_wq);
+	dev_dbg(mhdp->dev, "DP FW activated\n");
+
+	return 0;
+}
+
+static void cdns_mhdp_fw_cb(const struct firmware *fw, void *context)
+{
+	struct cdns_mhdp_device *mhdp = context;
+	bool bridge_attached;
+	int ret;
+
+	dev_dbg(mhdp->dev, "firmware callback\n");
+
+	if (!fw || !fw->data) {
+		dev_err(mhdp->dev, "%s: No firmware.\n", __func__);
+		return;
+	}
+
+	ret = cdns_mhdp_fw_activate(fw, mhdp);
+
+	release_firmware(fw);
+
+	if (ret)
+		return;
+
+	/*
+	 *  XXX how to make sure the bridge is still attached when
+	 *      calling drm_kms_helper_hotplug_event() after releasing
+	 *      the lock? We should not hold the spin lock when
+	 *      calling drm_kms_helper_hotplug_event() since it may
+	 *      cause a dead lock. FB-dev console calls detect from the
+	 *      same thread just down the call stack started here.
+	 */
+	spin_lock(&mhdp->start_lock);
+	bridge_attached = mhdp->bridge_attached;
+	spin_unlock(&mhdp->start_lock);
+	if (bridge_attached) {
+		if (mhdp->connector.dev)
+			drm_kms_helper_hotplug_event(mhdp->bridge.dev);
+		else
+			drm_bridge_hpd_notify(&mhdp->bridge, cdns_mhdp_detect(mhdp));
+	}
+}
+
+static int cdns_mhdp_load_firmware(struct cdns_mhdp_device *mhdp)
+{
+	int ret;
+
+	ret = request_firmware_nowait(THIS_MODULE, true, FW_NAME, mhdp->dev,
+				      GFP_KERNEL, mhdp, cdns_mhdp_fw_cb);
+	if (ret) {
+		dev_err(mhdp->dev, "failed to load firmware (%s), ret: %d\n",
+			FW_NAME, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static ssize_t cdns_mhdp_transfer(struct drm_dp_aux *aux,
+				  struct drm_dp_aux_msg *msg)
+{
+	struct cdns_mhdp_device *mhdp = dev_get_drvdata(aux->dev);
+	int ret;
+
+	if (msg->request != DP_AUX_NATIVE_WRITE &&
+	    msg->request != DP_AUX_NATIVE_READ)
+		return -EOPNOTSUPP;
+
+	if (msg->request == DP_AUX_NATIVE_WRITE) {
+		const u8 *buf = msg->buffer;
+		unsigned int i;
+
+		for (i = 0; i < msg->size; ++i) {
+			ret = cdns_mhdp_dpcd_write(mhdp,
+						   msg->address + i, buf[i]);
+			if (!ret)
+				continue;
+
+			dev_err(mhdp->dev,
+				"Failed to write DPCD addr %u\n",
+				msg->address + i);
+
+			return ret;
+		}
+	} else {
+		ret = cdns_mhdp_dpcd_read(mhdp, msg->address,
+					  msg->buffer, msg->size);
+		if (ret) {
+			dev_err(mhdp->dev,
+				"Failed to read DPCD addr %u\n",
+				msg->address);
+
+			return ret;
+		}
+	}
+
+	return msg->size;
+}
+
+static int cdns_mhdp_link_training_init(struct cdns_mhdp_device *mhdp)
+{
+	union phy_configure_opts phy_cfg;
+	u32 reg32;
+	int ret;
+
+	drm_dp_dpcd_writeb(&mhdp->aux, DP_TRAINING_PATTERN_SET,
+			   DP_TRAINING_PATTERN_DISABLE);
+
+	/* Reset PHY configuration */
+	reg32 = CDNS_PHY_COMMON_CONFIG | CDNS_PHY_TRAINING_TYPE(1);
+	if (!mhdp->host.scrambler)
+		reg32 |= CDNS_PHY_SCRAMBLER_BYPASS;
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_PHY_CONFIG, reg32);
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_ENHNCD,
+			    mhdp->sink.enhanced & mhdp->host.enhanced);
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_LANE_EN,
+			    CDNS_DP_LANE_EN_LANES(mhdp->link.num_lanes));
+
+	cdns_mhdp_link_configure(&mhdp->aux, &mhdp->link);
+	phy_cfg.dp.link_rate = mhdp->link.rate / 100;
+	phy_cfg.dp.lanes = mhdp->link.num_lanes;
+
+	memset(phy_cfg.dp.voltage, 0, sizeof(phy_cfg.dp.voltage));
+	memset(phy_cfg.dp.pre, 0, sizeof(phy_cfg.dp.pre));
+
+	phy_cfg.dp.ssc = cdns_mhdp_get_ssc_supported(mhdp);
+	phy_cfg.dp.set_lanes = true;
+	phy_cfg.dp.set_rate = true;
+	phy_cfg.dp.set_voltages = true;
+	ret = phy_configure(mhdp->phy,  &phy_cfg);
+	if (ret) {
+		dev_err(mhdp->dev, "%s: phy_configure() failed: %d\n",
+			__func__, ret);
+		return ret;
+	}
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_PHY_CONFIG,
+			    CDNS_PHY_COMMON_CONFIG |
+			    CDNS_PHY_TRAINING_EN |
+			    CDNS_PHY_TRAINING_TYPE(1) |
+			    CDNS_PHY_SCRAMBLER_BYPASS);
+
+	drm_dp_dpcd_writeb(&mhdp->aux, DP_TRAINING_PATTERN_SET,
+			   DP_TRAINING_PATTERN_1 | DP_LINK_SCRAMBLING_DISABLE);
+
+	return 0;
+}
+
+static void cdns_mhdp_get_adjust_train(struct cdns_mhdp_device *mhdp,
+				       u8 link_status[DP_LINK_STATUS_SIZE],
+				       u8 lanes_data[CDNS_DP_MAX_NUM_LANES],
+				       union phy_configure_opts *phy_cfg)
+{
+	u8 adjust, max_pre_emph, max_volt_swing;
+	u8 set_volt, set_pre;
+	unsigned int i;
+
+	max_pre_emph = CDNS_PRE_EMPHASIS(mhdp->host.pre_emphasis)
+			   << DP_TRAIN_PRE_EMPHASIS_SHIFT;
+	max_volt_swing = CDNS_VOLT_SWING(mhdp->host.volt_swing);
+
+	for (i = 0; i < mhdp->link.num_lanes; i++) {
+		/* Check if Voltage swing and pre-emphasis are within limits */
+		adjust = drm_dp_get_adjust_request_voltage(link_status, i);
+		set_volt = min(adjust, max_volt_swing);
+
+		adjust = drm_dp_get_adjust_request_pre_emphasis(link_status, i);
+		set_pre = min(adjust, max_pre_emph)
+			  >> DP_TRAIN_PRE_EMPHASIS_SHIFT;
+
+		/*
+		 * Voltage swing level and pre-emphasis level combination is
+		 * not allowed: leaving pre-emphasis as-is, and adjusting
+		 * voltage swing.
+		 */
+		if (set_volt + set_pre > 3)
+			set_volt = 3 - set_pre;
+
+		phy_cfg->dp.voltage[i] = set_volt;
+		lanes_data[i] = set_volt;
+
+		if (set_volt == max_volt_swing)
+			lanes_data[i] |= DP_TRAIN_MAX_SWING_REACHED;
+
+		phy_cfg->dp.pre[i] = set_pre;
+		lanes_data[i] |= (set_pre << DP_TRAIN_PRE_EMPHASIS_SHIFT);
+
+		if (set_pre == (max_pre_emph >> DP_TRAIN_PRE_EMPHASIS_SHIFT))
+			lanes_data[i] |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+	}
+}
+
+static
+void cdns_mhdp_set_adjust_request_voltage(u8 link_status[DP_LINK_STATUS_SIZE],
+					  unsigned int lane, u8 volt)
+{
+	unsigned int s = ((lane & 1) ?
+			  DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT :
+			  DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT);
+	unsigned int idx = DP_ADJUST_REQUEST_LANE0_1 - DP_LANE0_1_STATUS + (lane >> 1);
+
+	link_status[idx] &= ~(DP_ADJUST_VOLTAGE_SWING_LANE0_MASK << s);
+	link_status[idx] |= volt << s;
+}
+
+static
+void cdns_mhdp_set_adjust_request_pre_emphasis(u8 link_status[DP_LINK_STATUS_SIZE],
+					       unsigned int lane, u8 pre_emphasis)
+{
+	unsigned int s = ((lane & 1) ?
+			  DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT :
+			  DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT);
+	unsigned int idx = DP_ADJUST_REQUEST_LANE0_1 - DP_LANE0_1_STATUS + (lane >> 1);
+
+	link_status[idx] &= ~(DP_ADJUST_PRE_EMPHASIS_LANE0_MASK << s);
+	link_status[idx] |= pre_emphasis << s;
+}
+
+static void cdns_mhdp_adjust_requested_eq(struct cdns_mhdp_device *mhdp,
+					  u8 link_status[DP_LINK_STATUS_SIZE])
+{
+	u8 max_pre = CDNS_PRE_EMPHASIS(mhdp->host.pre_emphasis);
+	u8 max_volt = CDNS_VOLT_SWING(mhdp->host.volt_swing);
+	unsigned int i;
+	u8 volt, pre;
+
+	for (i = 0; i < mhdp->link.num_lanes; i++) {
+		volt = drm_dp_get_adjust_request_voltage(link_status, i);
+		pre = drm_dp_get_adjust_request_pre_emphasis(link_status, i);
+		if (volt + pre > 3)
+			cdns_mhdp_set_adjust_request_voltage(link_status, i,
+							     3 - pre);
+		if (mhdp->host.volt_swing & CDNS_FORCE_VOLT_SWING)
+			cdns_mhdp_set_adjust_request_voltage(link_status, i,
+							     max_volt);
+		if (mhdp->host.pre_emphasis & CDNS_FORCE_PRE_EMPHASIS)
+			cdns_mhdp_set_adjust_request_pre_emphasis(link_status,
+								  i, max_pre);
+	}
+}
+
+static void cdns_mhdp_print_lt_status(const char *prefix,
+				      struct cdns_mhdp_device *mhdp,
+				      union phy_configure_opts *phy_cfg)
+{
+	char vs[8] = "0/0/0/0";
+	char pe[8] = "0/0/0/0";
+	unsigned int i;
+
+	for (i = 0; i < mhdp->link.num_lanes; i++) {
+		vs[i * 2] = '0' + phy_cfg->dp.voltage[i];
+		pe[i * 2] = '0' + phy_cfg->dp.pre[i];
+	}
+
+	vs[i * 2 - 1] = '\0';
+	pe[i * 2 - 1] = '\0';
+
+	dev_dbg(mhdp->dev, "%s, %u lanes, %u Mbps, vs %s, pe %s\n",
+		prefix,
+		mhdp->link.num_lanes, mhdp->link.rate / 100,
+		vs, pe);
+}
+
+static bool cdns_mhdp_link_training_channel_eq(struct cdns_mhdp_device *mhdp,
+					       u8 eq_tps,
+					       unsigned int training_interval)
+{
+	u8 lanes_data[CDNS_DP_MAX_NUM_LANES], fail_counter_short = 0;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+	union phy_configure_opts phy_cfg;
+	u32 reg32;
+	int ret;
+	bool r;
+
+	dev_dbg(mhdp->dev, "Starting EQ phase\n");
+
+	/* Enable link training TPS[eq_tps] in PHY */
+	reg32 = CDNS_PHY_COMMON_CONFIG | CDNS_PHY_TRAINING_EN |
+		CDNS_PHY_TRAINING_TYPE(eq_tps);
+	if (eq_tps != 4)
+		reg32 |= CDNS_PHY_SCRAMBLER_BYPASS;
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_PHY_CONFIG, reg32);
+
+	drm_dp_dpcd_writeb(&mhdp->aux, DP_TRAINING_PATTERN_SET,
+			   (eq_tps != 4) ? eq_tps | DP_LINK_SCRAMBLING_DISABLE :
+			   CDNS_DP_TRAINING_PATTERN_4);
+
+	drm_dp_dpcd_read_link_status(&mhdp->aux, link_status);
+
+	do {
+		cdns_mhdp_get_adjust_train(mhdp, link_status, lanes_data,
+					   &phy_cfg);
+		phy_cfg.dp.lanes = mhdp->link.num_lanes;
+		phy_cfg.dp.ssc = cdns_mhdp_get_ssc_supported(mhdp);
+		phy_cfg.dp.set_lanes = false;
+		phy_cfg.dp.set_rate = false;
+		phy_cfg.dp.set_voltages = true;
+		ret = phy_configure(mhdp->phy,  &phy_cfg);
+		if (ret) {
+			dev_err(mhdp->dev, "%s: phy_configure() failed: %d\n",
+				__func__, ret);
+			goto err;
+		}
+
+		cdns_mhdp_adjust_lt(mhdp, mhdp->link.num_lanes,
+				    training_interval, lanes_data, link_status);
+
+		r = drm_dp_clock_recovery_ok(link_status, mhdp->link.num_lanes);
+		if (!r)
+			goto err;
+
+		if (drm_dp_channel_eq_ok(link_status, mhdp->link.num_lanes)) {
+			cdns_mhdp_print_lt_status("EQ phase ok", mhdp,
+						  &phy_cfg);
+			return true;
+		}
+
+		fail_counter_short++;
+
+		cdns_mhdp_adjust_requested_eq(mhdp, link_status);
+	} while (fail_counter_short < 5);
+
+err:
+	cdns_mhdp_print_lt_status("EQ phase failed", mhdp, &phy_cfg);
+
+	return false;
+}
+
+static void cdns_mhdp_adjust_requested_cr(struct cdns_mhdp_device *mhdp,
+					  u8 link_status[DP_LINK_STATUS_SIZE],
+					  u8 *req_volt, u8 *req_pre)
+{
+	const u8 max_volt = CDNS_VOLT_SWING(mhdp->host.volt_swing);
+	const u8 max_pre = CDNS_PRE_EMPHASIS(mhdp->host.pre_emphasis);
+	unsigned int i;
+
+	for (i = 0; i < mhdp->link.num_lanes; i++) {
+		u8 val;
+
+		val = mhdp->host.volt_swing & CDNS_FORCE_VOLT_SWING ?
+		      max_volt : req_volt[i];
+		cdns_mhdp_set_adjust_request_voltage(link_status, i, val);
+
+		val = mhdp->host.pre_emphasis & CDNS_FORCE_PRE_EMPHASIS ?
+		      max_pre : req_pre[i];
+		cdns_mhdp_set_adjust_request_pre_emphasis(link_status, i, val);
+	}
+}
+
+static
+void cdns_mhdp_validate_cr(struct cdns_mhdp_device *mhdp, bool *cr_done,
+			   bool *same_before_adjust, bool *max_swing_reached,
+			   u8 before_cr[CDNS_DP_MAX_NUM_LANES],
+			   u8 after_cr[DP_LINK_STATUS_SIZE], u8 *req_volt,
+			   u8 *req_pre)
+{
+	const u8 max_volt = CDNS_VOLT_SWING(mhdp->host.volt_swing);
+	const u8 max_pre = CDNS_PRE_EMPHASIS(mhdp->host.pre_emphasis);
+	bool same_pre, same_volt;
+	unsigned int i;
+	u8 adjust;
+
+	*same_before_adjust = false;
+	*max_swing_reached = false;
+	*cr_done = drm_dp_clock_recovery_ok(after_cr, mhdp->link.num_lanes);
+
+	for (i = 0; i < mhdp->link.num_lanes; i++) {
+		adjust = drm_dp_get_adjust_request_voltage(after_cr, i);
+		req_volt[i] = min(adjust, max_volt);
+
+		adjust = drm_dp_get_adjust_request_pre_emphasis(after_cr, i) >>
+		      DP_TRAIN_PRE_EMPHASIS_SHIFT;
+		req_pre[i] = min(adjust, max_pre);
+
+		same_pre = (before_cr[i] & DP_TRAIN_PRE_EMPHASIS_MASK) ==
+			   req_pre[i] << DP_TRAIN_PRE_EMPHASIS_SHIFT;
+		same_volt = (before_cr[i] & DP_TRAIN_VOLTAGE_SWING_MASK) ==
+			    req_volt[i];
+		if (same_pre && same_volt)
+			*same_before_adjust = true;
+
+		/* 3.1.5.2 in DP Standard v1.4. Table 3-1 */
+		if (!*cr_done && req_volt[i] + req_pre[i] >= 3) {
+			*max_swing_reached = true;
+			return;
+		}
+	}
+}
+
+static bool cdns_mhdp_link_training_cr(struct cdns_mhdp_device *mhdp)
+{
+	u8 lanes_data[CDNS_DP_MAX_NUM_LANES],
+	fail_counter_short = 0, fail_counter_cr_long = 0;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+	bool cr_done;
+	union phy_configure_opts phy_cfg;
+	int ret;
+
+	dev_dbg(mhdp->dev, "Starting CR phase\n");
+
+	ret = cdns_mhdp_link_training_init(mhdp);
+	if (ret)
+		goto err;
+
+	drm_dp_dpcd_read_link_status(&mhdp->aux, link_status);
+
+	do {
+		u8 requested_adjust_volt_swing[CDNS_DP_MAX_NUM_LANES] = {};
+		u8 requested_adjust_pre_emphasis[CDNS_DP_MAX_NUM_LANES] = {};
+		bool same_before_adjust, max_swing_reached;
+
+		cdns_mhdp_get_adjust_train(mhdp, link_status, lanes_data,
+					   &phy_cfg);
+		phy_cfg.dp.lanes = mhdp->link.num_lanes;
+		phy_cfg.dp.ssc = cdns_mhdp_get_ssc_supported(mhdp);
+		phy_cfg.dp.set_lanes = false;
+		phy_cfg.dp.set_rate = false;
+		phy_cfg.dp.set_voltages = true;
+		ret = phy_configure(mhdp->phy,  &phy_cfg);
+		if (ret) {
+			dev_err(mhdp->dev, "%s: phy_configure() failed: %d\n",
+				__func__, ret);
+			goto err;
+		}
+
+		cdns_mhdp_adjust_lt(mhdp, mhdp->link.num_lanes, 100,
+				    lanes_data, link_status);
+
+		cdns_mhdp_validate_cr(mhdp, &cr_done, &same_before_adjust,
+				      &max_swing_reached, lanes_data,
+				      link_status,
+				      requested_adjust_volt_swing,
+				      requested_adjust_pre_emphasis);
+
+		if (max_swing_reached) {
+			dev_err(mhdp->dev, "CR: max swing reached\n");
+			goto err;
+		}
+
+		if (cr_done) {
+			cdns_mhdp_print_lt_status("CR phase ok", mhdp,
+						  &phy_cfg);
+			return true;
+		}
+
+		/* Not all CR_DONE bits set */
+		fail_counter_cr_long++;
+
+		if (same_before_adjust) {
+			fail_counter_short++;
+			continue;
+		}
+
+		fail_counter_short = 0;
+		/*
+		 * Voltage swing/pre-emphasis adjust requested
+		 * during CR phase
+		 */
+		cdns_mhdp_adjust_requested_cr(mhdp, link_status,
+					      requested_adjust_volt_swing,
+					      requested_adjust_pre_emphasis);
+	} while (fail_counter_short < 5 && fail_counter_cr_long < 10);
+
+err:
+	cdns_mhdp_print_lt_status("CR phase failed", mhdp, &phy_cfg);
+
+	return false;
+}
+
+static void cdns_mhdp_lower_link_rate(struct cdns_mhdp_link *link)
+{
+	switch (drm_dp_link_rate_to_bw_code(link->rate)) {
+	case DP_LINK_BW_2_7:
+		link->rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_1_62);
+		break;
+	case DP_LINK_BW_5_4:
+		link->rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_2_7);
+		break;
+	case DP_LINK_BW_8_1:
+		link->rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_5_4);
+		break;
+	}
+}
+
+static int cdns_mhdp_link_training(struct cdns_mhdp_device *mhdp,
+				   unsigned int training_interval)
+{
+	u32 reg32;
+	const u8 eq_tps = cdns_mhdp_eq_training_pattern_supported(mhdp);
+	int ret;
+
+	while (1) {
+		if (!cdns_mhdp_link_training_cr(mhdp)) {
+			if (drm_dp_link_rate_to_bw_code(mhdp->link.rate) !=
+			    DP_LINK_BW_1_62) {
+				dev_dbg(mhdp->dev,
+					"Reducing link rate during CR phase\n");
+				cdns_mhdp_lower_link_rate(&mhdp->link);
+
+				continue;
+			} else if (mhdp->link.num_lanes > 1) {
+				dev_dbg(mhdp->dev,
+					"Reducing lanes number during CR phase\n");
+				mhdp->link.num_lanes >>= 1;
+				mhdp->link.rate = cdns_mhdp_max_link_rate(mhdp);
+
+				continue;
+			}
+
+			dev_err(mhdp->dev,
+				"Link training failed during CR phase\n");
+			goto err;
+		}
+
+		if (cdns_mhdp_link_training_channel_eq(mhdp, eq_tps,
+						       training_interval))
+			break;
+
+		if (mhdp->link.num_lanes > 1) {
+			dev_dbg(mhdp->dev,
+				"Reducing lanes number during EQ phase\n");
+			mhdp->link.num_lanes >>= 1;
+
+			continue;
+		} else if (drm_dp_link_rate_to_bw_code(mhdp->link.rate) !=
+			   DP_LINK_BW_1_62) {
+			dev_dbg(mhdp->dev,
+				"Reducing link rate during EQ phase\n");
+			cdns_mhdp_lower_link_rate(&mhdp->link);
+			mhdp->link.num_lanes = cdns_mhdp_max_num_lanes(mhdp);
+
+			continue;
+		}
+
+		dev_err(mhdp->dev, "Link training failed during EQ phase\n");
+		goto err;
+	}
+
+	dev_dbg(mhdp->dev, "Link training ok. Lanes: %u, Rate %u Mbps\n",
+		mhdp->link.num_lanes, mhdp->link.rate / 100);
+
+	drm_dp_dpcd_writeb(&mhdp->aux, DP_TRAINING_PATTERN_SET,
+			   mhdp->host.scrambler ? 0 :
+			   DP_LINK_SCRAMBLING_DISABLE);
+
+	ret = cdns_mhdp_reg_read(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, &reg32);
+	if (ret < 0) {
+		dev_err(mhdp->dev,
+			"Failed to read CDNS_DP_FRAMER_GLOBAL_CONFIG %d\n",
+			ret);
+		return ret;
+	}
+	reg32 &= ~GENMASK(1, 0);
+	reg32 |= CDNS_DP_NUM_LANES(mhdp->link.num_lanes);
+	reg32 |= CDNS_DP_WR_FAILING_EDGE_VSYNC;
+	reg32 |= CDNS_DP_FRAMER_EN;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, reg32);
+
+	/* Reset PHY config */
+	reg32 = CDNS_PHY_COMMON_CONFIG | CDNS_PHY_TRAINING_TYPE(1);
+	if (!mhdp->host.scrambler)
+		reg32 |= CDNS_PHY_SCRAMBLER_BYPASS;
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_PHY_CONFIG, reg32);
+
+	return 0;
+err:
+	/* Reset PHY config */
+	reg32 = CDNS_PHY_COMMON_CONFIG | CDNS_PHY_TRAINING_TYPE(1);
+	if (!mhdp->host.scrambler)
+		reg32 |= CDNS_PHY_SCRAMBLER_BYPASS;
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_PHY_CONFIG, reg32);
+
+	drm_dp_dpcd_writeb(&mhdp->aux, DP_TRAINING_PATTERN_SET,
+			   DP_TRAINING_PATTERN_DISABLE);
+
+	return -EIO;
+}
+
+static u32 cdns_mhdp_get_training_interval_us(struct cdns_mhdp_device *mhdp,
+					      u32 interval)
+{
+	if (interval == 0)
+		return 400;
+	if (interval < 5)
+		return 4000 << (interval - 1);
+	dev_err(mhdp->dev,
+		"wrong training interval returned by DPCD: %d\n", interval);
+	return 0;
+}
+
+static void cdns_mhdp_fill_host_caps(struct cdns_mhdp_device *mhdp)
+{
+	unsigned int link_rate;
+
+	/* Get source capabilities based on PHY attributes */
+
+	mhdp->host.lanes_cnt = mhdp->phy->attrs.bus_width;
+	if (!mhdp->host.lanes_cnt)
+		mhdp->host.lanes_cnt = 4;
+
+	link_rate = mhdp->phy->attrs.max_link_rate;
+	if (!link_rate)
+		link_rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_8_1);
+	else
+		/* PHY uses Mb/s, DRM uses tens of kb/s. */
+		link_rate *= 100;
+
+	mhdp->host.link_rate = link_rate;
+	mhdp->host.volt_swing = CDNS_VOLT_SWING(3);
+	mhdp->host.pre_emphasis = CDNS_PRE_EMPHASIS(3);
+	mhdp->host.pattern_supp = CDNS_SUPPORT_TPS(1) |
+				  CDNS_SUPPORT_TPS(2) | CDNS_SUPPORT_TPS(3) |
+				  CDNS_SUPPORT_TPS(4);
+	mhdp->host.lane_mapping = CDNS_LANE_MAPPING_NORMAL;
+	mhdp->host.fast_link = false;
+	mhdp->host.enhanced = true;
+	mhdp->host.scrambler = true;
+	mhdp->host.ssc = false;
+}
+
+static void cdns_mhdp_fill_sink_caps(struct cdns_mhdp_device *mhdp,
+				     u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	mhdp->sink.link_rate = mhdp->link.rate;
+	mhdp->sink.lanes_cnt = mhdp->link.num_lanes;
+	mhdp->sink.enhanced = !!(mhdp->link.capabilities &
+				 DP_LINK_CAP_ENHANCED_FRAMING);
+
+	/* Set SSC support */
+	mhdp->sink.ssc = !!(dpcd[DP_MAX_DOWNSPREAD] &
+				  DP_MAX_DOWNSPREAD_0_5);
+
+	/* Set TPS support */
+	mhdp->sink.pattern_supp = CDNS_SUPPORT_TPS(1) | CDNS_SUPPORT_TPS(2);
+	if (drm_dp_tps3_supported(dpcd))
+		mhdp->sink.pattern_supp |= CDNS_SUPPORT_TPS(3);
+	if (drm_dp_tps4_supported(dpcd))
+		mhdp->sink.pattern_supp |= CDNS_SUPPORT_TPS(4);
+
+	/* Set fast link support */
+	mhdp->sink.fast_link = !!(dpcd[DP_MAX_DOWNSPREAD] &
+				  DP_NO_AUX_HANDSHAKE_LINK_TRAINING);
+}
+
+static int cdns_mhdp_link_up(struct cdns_mhdp_device *mhdp)
+{
+	u8 dpcd[DP_RECEIVER_CAP_SIZE], amp[2];
+	u32 resp, interval, interval_us;
+	u8 ext_cap_chk = 0;
+	unsigned int addr;
+	int err;
+
+	WARN_ON(!mutex_is_locked(&mhdp->link_mutex));
+
+	drm_dp_dpcd_readb(&mhdp->aux, DP_TRAINING_AUX_RD_INTERVAL,
+			  &ext_cap_chk);
+
+	if (ext_cap_chk & DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT)
+		addr = DP_DP13_DPCD_REV;
+	else
+		addr = DP_DPCD_REV;
+
+	err = drm_dp_dpcd_read(&mhdp->aux, addr, dpcd, DP_RECEIVER_CAP_SIZE);
+	if (err < 0) {
+		dev_err(mhdp->dev, "Failed to read receiver capabilities\n");
+		return err;
+	}
+
+	mhdp->link.revision = dpcd[0];
+	mhdp->link.rate = drm_dp_bw_code_to_link_rate(dpcd[1]);
+	mhdp->link.num_lanes = dpcd[2] & DP_MAX_LANE_COUNT_MASK;
+
+	if (dpcd[2] & DP_ENHANCED_FRAME_CAP)
+		mhdp->link.capabilities |= DP_LINK_CAP_ENHANCED_FRAMING;
+
+	dev_dbg(mhdp->dev, "Set sink device power state via DPCD\n");
+	cdns_mhdp_link_power_up(&mhdp->aux, &mhdp->link);
+
+	cdns_mhdp_fill_sink_caps(mhdp, dpcd);
+
+	mhdp->link.rate = cdns_mhdp_max_link_rate(mhdp);
+	mhdp->link.num_lanes = cdns_mhdp_max_num_lanes(mhdp);
+
+	/* Disable framer for link training */
+	err = cdns_mhdp_reg_read(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, &resp);
+	if (err < 0) {
+		dev_err(mhdp->dev,
+			"Failed to read CDNS_DP_FRAMER_GLOBAL_CONFIG %d\n",
+			err);
+		return err;
+	}
+
+	resp &= ~CDNS_DP_FRAMER_EN;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, resp);
+
+	/* Spread AMP if required, enable 8b/10b coding */
+	amp[0] = cdns_mhdp_get_ssc_supported(mhdp) ? DP_SPREAD_AMP_0_5 : 0;
+	amp[1] = DP_SET_ANSI_8B10B;
+	drm_dp_dpcd_write(&mhdp->aux, DP_DOWNSPREAD_CTRL, amp, 2);
+
+	if (mhdp->host.fast_link & mhdp->sink.fast_link) {
+		dev_err(mhdp->dev, "fastlink not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	interval = dpcd[DP_TRAINING_AUX_RD_INTERVAL] & DP_TRAINING_AUX_RD_MASK;
+	interval_us = cdns_mhdp_get_training_interval_us(mhdp, interval);
+	if (!interval_us ||
+	    cdns_mhdp_link_training(mhdp, interval_us)) {
+		dev_err(mhdp->dev, "Link training failed. Exiting.\n");
+		return -EIO;
+	}
+
+	mhdp->link_up = true;
+
+	return 0;
+}
+
+static void cdns_mhdp_link_down(struct cdns_mhdp_device *mhdp)
+{
+	WARN_ON(!mutex_is_locked(&mhdp->link_mutex));
+
+	if (mhdp->plugged)
+		cdns_mhdp_link_power_down(&mhdp->aux, &mhdp->link);
+
+	mhdp->link_up = false;
+}
+
+static struct edid *cdns_mhdp_get_edid(struct cdns_mhdp_device *mhdp,
+				       struct drm_connector *connector)
+{
+	if (!mhdp->plugged)
+		return NULL;
+
+	return drm_do_get_edid(connector, cdns_mhdp_get_edid_block, mhdp);
+}
+
+static int cdns_mhdp_get_modes(struct drm_connector *connector)
+{
+	struct cdns_mhdp_device *mhdp = connector_to_mhdp(connector);
+	struct edid *edid;
+	int num_modes;
+
+	if (!mhdp->plugged)
+		return 0;
+
+	edid = cdns_mhdp_get_edid(mhdp, connector);
+	if (!edid) {
+		dev_err(mhdp->dev, "Failed to read EDID\n");
+		return 0;
+	}
+
+	drm_connector_update_edid_property(connector, edid);
+	num_modes = drm_add_edid_modes(connector, edid);
+	kfree(edid);
+
+	/*
+	 * HACK: Warn about unsupported display formats until we deal
+	 *       with them correctly.
+	 */
+	if (connector->display_info.color_formats &&
+	    !(connector->display_info.color_formats &
+	      mhdp->display_fmt.color_format))
+		dev_warn(mhdp->dev,
+			 "%s: No supported color_format found (0x%08x)\n",
+			__func__, connector->display_info.color_formats);
+
+	if (connector->display_info.bpc &&
+	    connector->display_info.bpc < mhdp->display_fmt.bpc)
+		dev_warn(mhdp->dev, "%s: Display bpc only %d < %d\n",
+			 __func__, connector->display_info.bpc,
+			 mhdp->display_fmt.bpc);
+
+	return num_modes;
+}
+
+static int cdns_mhdp_connector_detect(struct drm_connector *conn,
+				      struct drm_modeset_acquire_ctx *ctx,
+				      bool force)
+{
+	struct cdns_mhdp_device *mhdp = connector_to_mhdp(conn);
+
+	return cdns_mhdp_detect(mhdp);
+}
+
+static u32 cdns_mhdp_get_bpp(struct cdns_mhdp_display_fmt *fmt)
+{
+	u32 bpp;
+
+	if (fmt->y_only)
+		return fmt->bpc;
+
+	switch (fmt->color_format) {
+	case DRM_COLOR_FORMAT_RGB444:
+	case DRM_COLOR_FORMAT_YCRCB444:
+		bpp = fmt->bpc * 3;
+		break;
+	case DRM_COLOR_FORMAT_YCRCB422:
+		bpp = fmt->bpc * 2;
+		break;
+	case DRM_COLOR_FORMAT_YCRCB420:
+		bpp = fmt->bpc * 3 / 2;
+		break;
+	default:
+		bpp = fmt->bpc * 3;
+		WARN_ON(1);
+	}
+	return bpp;
+}
+
+static
+bool cdns_mhdp_bandwidth_ok(struct cdns_mhdp_device *mhdp,
+			    const struct drm_display_mode *mode,
+			    unsigned int lanes, unsigned int rate)
+{
+	u32 max_bw, req_bw, bpp;
+
+	/*
+	 * mode->clock is expressed in kHz. Multiplying by bpp and dividing by 8
+	 * we get the number of kB/s. DisplayPort applies a 8b-10b encoding, the
+	 * value thus equals the bandwidth in 10kb/s units, which matches the
+	 * units of the rate parameter.
+	 */
+
+	bpp = cdns_mhdp_get_bpp(&mhdp->display_fmt);
+	req_bw = mode->clock * bpp / 8;
+	max_bw = lanes * rate;
+	if (req_bw > max_bw) {
+		dev_dbg(mhdp->dev,
+			"Unsupported Mode: %s, Req BW: %u, Available Max BW:%u\n",
+			mode->name, req_bw, max_bw);
+
+		return false;
+	}
+
+	return true;
+}
+
+static
+enum drm_mode_status cdns_mhdp_mode_valid(struct drm_connector *conn,
+					  struct drm_display_mode *mode)
+{
+	struct cdns_mhdp_device *mhdp = connector_to_mhdp(conn);
+
+	mutex_lock(&mhdp->link_mutex);
+
+	if (!cdns_mhdp_bandwidth_ok(mhdp, mode, mhdp->link.num_lanes,
+				    mhdp->link.rate)) {
+		mutex_unlock(&mhdp->link_mutex);
+		return MODE_CLOCK_HIGH;
+	}
+
+	mutex_unlock(&mhdp->link_mutex);
+	return MODE_OK;
+}
+
+static const struct drm_connector_helper_funcs cdns_mhdp_conn_helper_funcs = {
+	.detect_ctx = cdns_mhdp_connector_detect,
+	.get_modes = cdns_mhdp_get_modes,
+	.mode_valid = cdns_mhdp_mode_valid,
+};
+
+static const struct drm_connector_funcs cdns_mhdp_conn_funcs = {
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+	.reset = drm_atomic_helper_connector_reset,
+	.destroy = drm_connector_cleanup,
+};
+
+static int cdns_mhdp_connector_init(struct cdns_mhdp_device *mhdp)
+{
+	u32 bus_format = MEDIA_BUS_FMT_RGB121212_1X36;
+	struct drm_connector *conn = &mhdp->connector;
+	struct drm_bridge *bridge = &mhdp->bridge;
+	int ret;
+
+	if (!bridge->encoder) {
+		dev_err(mhdp->dev, "Parent encoder object not found");
+		return -ENODEV;
+	}
+
+	conn->polled = DRM_CONNECTOR_POLL_HPD;
+
+	ret = drm_connector_init(bridge->dev, conn, &cdns_mhdp_conn_funcs,
+				 DRM_MODE_CONNECTOR_DisplayPort);
+	if (ret) {
+		dev_err(mhdp->dev, "Failed to initialize connector with drm\n");
+		return ret;
+	}
+
+	drm_connector_helper_add(conn, &cdns_mhdp_conn_helper_funcs);
+
+	ret = drm_display_info_set_bus_formats(&conn->display_info,
+					       &bus_format, 1);
+	if (ret)
+		return ret;
+
+	ret = drm_connector_attach_encoder(conn, bridge->encoder);
+	if (ret) {
+		dev_err(mhdp->dev, "Failed to attach connector to encoder\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int cdns_mhdp_attach(struct drm_bridge *bridge,
+			    enum drm_bridge_attach_flags flags)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+	bool hw_ready;
+	int ret;
+
+	dev_dbg(mhdp->dev, "%s\n", __func__);
+
+	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) {
+		ret = cdns_mhdp_connector_init(mhdp);
+		if (ret)
+			return ret;
+	}
+
+	spin_lock(&mhdp->start_lock);
+
+	mhdp->bridge_attached = true;
+	hw_ready = mhdp->hw_state == MHDP_HW_READY;
+
+	spin_unlock(&mhdp->start_lock);
+
+	/* Enable SW event interrupts */
+	if (hw_ready)
+		writel(~(u32)CDNS_APB_INT_MASK_SW_EVENT_INT,
+		       mhdp->regs + CDNS_APB_INT_MASK);
+
+	return 0;
+}
+
+static void cdns_mhdp_configure_video(struct cdns_mhdp_device *mhdp,
+				      const struct drm_display_mode *mode)
+{
+	unsigned int dp_framer_sp = 0, msa_horizontal_1,
+		msa_vertical_1, bnd_hsync2vsync, hsync2vsync_pol_ctrl,
+		misc0 = 0, misc1 = 0, pxl_repr,
+		front_porch, back_porch, msa_h0, msa_v0, hsync, vsync,
+		dp_vertical_1;
+	u8 stream_id = mhdp->stream_id;
+	u32 bpp, bpc, pxlfmt, framer;
+	int ret;
+
+	pxlfmt = mhdp->display_fmt.color_format;
+	bpc = mhdp->display_fmt.bpc;
+
+	/*
+	 * If YCBCR supported and stream not SD, use ITU709
+	 * Need to handle ITU version with YCBCR420 when supported
+	 */
+	if ((pxlfmt == DRM_COLOR_FORMAT_YCRCB444 ||
+	     pxlfmt == DRM_COLOR_FORMAT_YCRCB422) && mode->crtc_vdisplay >= 720)
+		misc0 = DP_YCBCR_COEFFICIENTS_ITU709;
+
+	bpp = cdns_mhdp_get_bpp(&mhdp->display_fmt);
+
+	switch (pxlfmt) {
+	case DRM_COLOR_FORMAT_RGB444:
+		pxl_repr = CDNS_DP_FRAMER_RGB << CDNS_DP_FRAMER_PXL_FORMAT;
+		misc0 |= DP_COLOR_FORMAT_RGB;
+		break;
+	case DRM_COLOR_FORMAT_YCRCB444:
+		pxl_repr = CDNS_DP_FRAMER_YCBCR444 << CDNS_DP_FRAMER_PXL_FORMAT;
+		misc0 |= DP_COLOR_FORMAT_YCbCr444 | DP_TEST_DYNAMIC_RANGE_CEA;
+		break;
+	case DRM_COLOR_FORMAT_YCRCB422:
+		pxl_repr = CDNS_DP_FRAMER_YCBCR422 << CDNS_DP_FRAMER_PXL_FORMAT;
+		misc0 |= DP_COLOR_FORMAT_YCbCr422 | DP_TEST_DYNAMIC_RANGE_CEA;
+		break;
+	case DRM_COLOR_FORMAT_YCRCB420:
+		pxl_repr = CDNS_DP_FRAMER_YCBCR420 << CDNS_DP_FRAMER_PXL_FORMAT;
+		break;
+	default:
+		pxl_repr = CDNS_DP_FRAMER_Y_ONLY << CDNS_DP_FRAMER_PXL_FORMAT;
+	}
+
+	switch (bpc) {
+	case 6:
+		misc0 |= DP_TEST_BIT_DEPTH_6;
+		pxl_repr |= CDNS_DP_FRAMER_6_BPC;
+		break;
+	case 8:
+		misc0 |= DP_TEST_BIT_DEPTH_8;
+		pxl_repr |= CDNS_DP_FRAMER_8_BPC;
+		break;
+	case 10:
+		misc0 |= DP_TEST_BIT_DEPTH_10;
+		pxl_repr |= CDNS_DP_FRAMER_10_BPC;
+		break;
+	case 12:
+		misc0 |= DP_TEST_BIT_DEPTH_12;
+		pxl_repr |= CDNS_DP_FRAMER_12_BPC;
+		break;
+	case 16:
+		misc0 |= DP_TEST_BIT_DEPTH_16;
+		pxl_repr |= CDNS_DP_FRAMER_16_BPC;
+		break;
+	}
+
+	bnd_hsync2vsync = CDNS_IP_BYPASS_V_INTERFACE;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		bnd_hsync2vsync |= CDNS_IP_DET_INTERLACE_FORMAT;
+
+	cdns_mhdp_reg_write(mhdp, CDNS_BND_HSYNC2VSYNC(stream_id),
+			    bnd_hsync2vsync);
+
+	hsync2vsync_pol_ctrl = 0;
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		hsync2vsync_pol_ctrl |= CDNS_H2V_HSYNC_POL_ACTIVE_LOW;
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		hsync2vsync_pol_ctrl |= CDNS_H2V_VSYNC_POL_ACTIVE_LOW;
+	cdns_mhdp_reg_write(mhdp, CDNS_HSYNC2VSYNC_POL_CTRL(stream_id),
+			    hsync2vsync_pol_ctrl);
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_PXL_REPR(stream_id), pxl_repr);
+
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		dp_framer_sp |= CDNS_DP_FRAMER_INTERLACE;
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		dp_framer_sp |= CDNS_DP_FRAMER_HSYNC_POL_LOW;
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		dp_framer_sp |= CDNS_DP_FRAMER_VSYNC_POL_LOW;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_SP(stream_id), dp_framer_sp);
+
+	front_porch = mode->crtc_hsync_start - mode->crtc_hdisplay;
+	back_porch = mode->crtc_htotal - mode->crtc_hsync_end;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRONT_BACK_PORCH(stream_id),
+			    CDNS_DP_FRONT_PORCH(front_porch) |
+			    CDNS_DP_BACK_PORCH(back_porch));
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_BYTE_COUNT(stream_id),
+			    mode->crtc_hdisplay * bpp / 8);
+
+	msa_h0 = mode->crtc_htotal - mode->crtc_hsync_start;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_HORIZONTAL_0(stream_id),
+			    CDNS_DP_MSAH0_H_TOTAL(mode->crtc_htotal) |
+			    CDNS_DP_MSAH0_HSYNC_START(msa_h0));
+
+	hsync = mode->crtc_hsync_end - mode->crtc_hsync_start;
+	msa_horizontal_1 = CDNS_DP_MSAH1_HSYNC_WIDTH(hsync) |
+			   CDNS_DP_MSAH1_HDISP_WIDTH(mode->crtc_hdisplay);
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		msa_horizontal_1 |= CDNS_DP_MSAH1_HSYNC_POL_LOW;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_HORIZONTAL_1(stream_id),
+			    msa_horizontal_1);
+
+	msa_v0 = mode->crtc_vtotal - mode->crtc_vsync_start;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_VERTICAL_0(stream_id),
+			    CDNS_DP_MSAV0_V_TOTAL(mode->crtc_vtotal) |
+			    CDNS_DP_MSAV0_VSYNC_START(msa_v0));
+
+	vsync = mode->crtc_vsync_end - mode->crtc_vsync_start;
+	msa_vertical_1 = CDNS_DP_MSAV1_VSYNC_WIDTH(vsync) |
+			 CDNS_DP_MSAV1_VDISP_WIDTH(mode->crtc_vdisplay);
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		msa_vertical_1 |= CDNS_DP_MSAV1_VSYNC_POL_LOW;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_VERTICAL_1(stream_id),
+			    msa_vertical_1);
+
+	if ((mode->flags & DRM_MODE_FLAG_INTERLACE) &&
+	    mode->crtc_vtotal % 2 == 0)
+		misc1 = DP_TEST_INTERLACED;
+	if (mhdp->display_fmt.y_only)
+		misc1 |= CDNS_DP_TEST_COLOR_FORMAT_RAW_Y_ONLY;
+	/* Use VSC SDP for Y420 */
+	if (pxlfmt == DRM_COLOR_FORMAT_YCRCB420)
+		misc1 = CDNS_DP_TEST_VSC_SDP;
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_MISC(stream_id),
+			    misc0 | (misc1 << 8));
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_HORIZONTAL(stream_id),
+			    CDNS_DP_H_HSYNC_WIDTH(hsync) |
+			    CDNS_DP_H_H_TOTAL(mode->crtc_hdisplay));
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_VERTICAL_0(stream_id),
+			    CDNS_DP_V0_VHEIGHT(mode->crtc_vdisplay) |
+			    CDNS_DP_V0_VSTART(msa_v0));
+
+	dp_vertical_1 = CDNS_DP_V1_VTOTAL(mode->crtc_vtotal);
+	if ((mode->flags & DRM_MODE_FLAG_INTERLACE) &&
+	    mode->crtc_vtotal % 2 == 0)
+		dp_vertical_1 |= CDNS_DP_V1_VTOTAL_EVEN;
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_VERTICAL_1(stream_id), dp_vertical_1);
+
+	cdns_mhdp_reg_write_bit(mhdp, CDNS_DP_VB_ID(stream_id), 2, 1,
+				(mode->flags & DRM_MODE_FLAG_INTERLACE) ?
+				CDNS_DP_VB_ID_INTERLACED : 0);
+
+	ret = cdns_mhdp_reg_read(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, &framer);
+	if (ret < 0) {
+		dev_err(mhdp->dev,
+			"Failed to read CDNS_DP_FRAMER_GLOBAL_CONFIG %d\n",
+			ret);
+		return;
+	}
+	framer |= CDNS_DP_FRAMER_EN;
+	framer &= ~CDNS_DP_NO_VIDEO_MODE;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, framer);
+}
+
+static void cdns_mhdp_sst_enable(struct cdns_mhdp_device *mhdp,
+				 const struct drm_display_mode *mode)
+{
+	u32 rate, vs, required_bandwidth, available_bandwidth;
+	s32 line_thresh1, line_thresh2, line_thresh = 0;
+	int pxlclock = mode->crtc_clock;
+	u32 tu_size = 64;
+	u32 bpp;
+
+	/* Get rate in MSymbols per second per lane */
+	rate = mhdp->link.rate / 1000;
+
+	bpp = cdns_mhdp_get_bpp(&mhdp->display_fmt);
+
+	required_bandwidth = pxlclock * bpp / 8;
+	available_bandwidth = mhdp->link.num_lanes * rate;
+
+	vs = tu_size * required_bandwidth / available_bandwidth;
+	vs /= 1000;
+
+	if (vs == tu_size)
+		vs = tu_size - 1;
+
+	line_thresh1 = ((vs + 1) << 5) * 8 / bpp;
+	line_thresh2 = (pxlclock << 5) / 1000 / rate * (vs + 1) - (1 << 5);
+	line_thresh = line_thresh1 - line_thresh2 / (s32)mhdp->link.num_lanes;
+	line_thresh = (line_thresh >> 5) + 2;
+
+	mhdp->stream_id = 0;
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_TU,
+			    CDNS_DP_FRAMER_TU_VS(vs) |
+			    CDNS_DP_FRAMER_TU_SIZE(tu_size) |
+			    CDNS_DP_FRAMER_TU_CNT_RST_EN);
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_LINE_THRESH(0),
+			    line_thresh & GENMASK(5, 0));
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_STREAM_CONFIG_2(0),
+			    CDNS_DP_SC2_TU_VS_DIFF((tu_size - vs > 3) ?
+						   0 : tu_size - vs));
+
+	cdns_mhdp_configure_video(mhdp, mode);
+}
+
+static void cdns_mhdp_atomic_enable(struct drm_bridge *bridge,
+				    struct drm_bridge_state *bridge_state)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+	struct drm_atomic_state *state = bridge_state->base.state;
+	struct cdns_mhdp_bridge_state *mhdp_state;
+	struct drm_crtc_state *crtc_state;
+	struct drm_connector *connector;
+	struct drm_connector_state *conn_state;
+	struct drm_bridge_state *new_state;
+	const struct drm_display_mode *mode;
+	u32 resp;
+	int ret;
+
+	dev_dbg(mhdp->dev, "bridge enable\n");
+
+	mutex_lock(&mhdp->link_mutex);
+
+	if (mhdp->plugged && !mhdp->link_up) {
+		ret = cdns_mhdp_link_up(mhdp);
+		if (ret < 0)
+			goto out;
+	}
+
+	if (mhdp->info && mhdp->info->ops && mhdp->info->ops->enable)
+		mhdp->info->ops->enable(mhdp);
+
+	/* Enable VIF clock for stream 0 */
+	ret = cdns_mhdp_reg_read(mhdp, CDNS_DPTX_CAR, &resp);
+	if (ret < 0) {
+		dev_err(mhdp->dev, "Failed to read CDNS_DPTX_CAR %d\n", ret);
+		goto out;
+	}
+
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_CAR,
+			    resp | CDNS_VIF_CLK_EN | CDNS_VIF_CLK_RSTN);
+
+	connector = drm_atomic_get_new_connector_for_encoder(state,
+							     bridge->encoder);
+	if (WARN_ON(!connector))
+		goto out;
+
+	conn_state = drm_atomic_get_new_connector_state(state, connector);
+	if (WARN_ON(!conn_state))
+		goto out;
+
+	crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+	if (WARN_ON(!crtc_state))
+		goto out;
+
+	mode = &crtc_state->adjusted_mode;
+
+	new_state = drm_atomic_get_new_bridge_state(state, bridge);
+	if (WARN_ON(!new_state))
+		goto out;
+
+	if (!cdns_mhdp_bandwidth_ok(mhdp, mode, mhdp->link.num_lanes,
+				    mhdp->link.rate)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	cdns_mhdp_sst_enable(mhdp, mode);
+
+	mhdp_state = to_cdns_mhdp_bridge_state(new_state);
+
+	mhdp_state->current_mode = drm_mode_duplicate(bridge->dev, mode);
+	drm_mode_set_name(mhdp_state->current_mode);
+
+	dev_dbg(mhdp->dev, "%s: Enabling mode %s\n", __func__, mode->name);
+
+	mhdp->bridge_enabled = true;
+
+out:
+	mutex_unlock(&mhdp->link_mutex);
+	if (ret < 0)
+		schedule_work(&mhdp->modeset_retry_work);
+}
+
+static void cdns_mhdp_atomic_disable(struct drm_bridge *bridge,
+				     struct drm_bridge_state *bridge_state)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+	u32 resp;
+
+	dev_dbg(mhdp->dev, "%s\n", __func__);
+
+	mutex_lock(&mhdp->link_mutex);
+
+	mhdp->bridge_enabled = false;
+	cdns_mhdp_reg_read(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, &resp);
+	resp &= ~CDNS_DP_FRAMER_EN;
+	resp |= CDNS_DP_NO_VIDEO_MODE;
+	cdns_mhdp_reg_write(mhdp, CDNS_DP_FRAMER_GLOBAL_CONFIG, resp);
+
+	cdns_mhdp_link_down(mhdp);
+
+	/* Disable VIF clock for stream 0 */
+	cdns_mhdp_reg_read(mhdp, CDNS_DPTX_CAR, &resp);
+	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_CAR,
+			    resp & ~(CDNS_VIF_CLK_EN | CDNS_VIF_CLK_RSTN));
+
+	if (mhdp->info && mhdp->info->ops && mhdp->info->ops->disable)
+		mhdp->info->ops->disable(mhdp);
+
+	mutex_unlock(&mhdp->link_mutex);
+}
+
+static void cdns_mhdp_detach(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+
+	dev_dbg(mhdp->dev, "%s\n", __func__);
+
+	spin_lock(&mhdp->start_lock);
+
+	mhdp->bridge_attached = false;
+
+	spin_unlock(&mhdp->start_lock);
+
+	writel(~0, mhdp->regs + CDNS_APB_INT_MASK);
+}
+
+static struct drm_bridge_state *
+cdns_mhdp_bridge_atomic_duplicate_state(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_bridge_state *state;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return NULL;
+
+	__drm_atomic_helper_bridge_duplicate_state(bridge, &state->base);
+
+	return &state->base;
+}
+
+static void
+cdns_mhdp_bridge_atomic_destroy_state(struct drm_bridge *bridge,
+				      struct drm_bridge_state *state)
+{
+	struct cdns_mhdp_bridge_state *cdns_mhdp_state;
+
+	cdns_mhdp_state = to_cdns_mhdp_bridge_state(state);
+
+	if (cdns_mhdp_state->current_mode) {
+		drm_mode_destroy(bridge->dev, cdns_mhdp_state->current_mode);
+		cdns_mhdp_state->current_mode = NULL;
+	}
+
+	kfree(cdns_mhdp_state);
+}
+
+static struct drm_bridge_state *
+cdns_mhdp_bridge_atomic_reset(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_bridge_state *cdns_mhdp_state;
+
+	cdns_mhdp_state = kzalloc(sizeof(*cdns_mhdp_state), GFP_KERNEL);
+	if (!cdns_mhdp_state)
+		return NULL;
+
+	 __drm_atomic_helper_bridge_reset(bridge, &cdns_mhdp_state->base);
+
+	return &cdns_mhdp_state->base;
+}
+
+static int cdns_mhdp_atomic_check(struct drm_bridge *bridge,
+				  struct drm_bridge_state *bridge_state,
+				  struct drm_crtc_state *crtc_state,
+				  struct drm_connector_state *conn_state)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+	const struct drm_display_mode *mode = &crtc_state->adjusted_mode;
+
+	mutex_lock(&mhdp->link_mutex);
+
+	if (!cdns_mhdp_bandwidth_ok(mhdp, mode, mhdp->link.num_lanes,
+				    mhdp->link.rate)) {
+		dev_err(mhdp->dev, "%s: Not enough BW for %s (%u lanes at %u Mbps)\n",
+			__func__, mode->name, mhdp->link.num_lanes,
+			mhdp->link.rate / 100);
+		mutex_unlock(&mhdp->link_mutex);
+		return -EINVAL;
+	}
+
+	mutex_unlock(&mhdp->link_mutex);
+	return 0;
+}
+
+static enum drm_connector_status cdns_mhdp_bridge_detect(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+
+	return cdns_mhdp_detect(mhdp);
+}
+
+static struct edid *cdns_mhdp_bridge_get_edid(struct drm_bridge *bridge,
+					      struct drm_connector *connector)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+
+	return cdns_mhdp_get_edid(mhdp, connector);
+}
+
+static void cdns_mhdp_bridge_hpd_enable(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+
+	/* Enable SW event interrupts */
+	if (mhdp->bridge_attached)
+		writel(~(u32)CDNS_APB_INT_MASK_SW_EVENT_INT,
+		       mhdp->regs + CDNS_APB_INT_MASK);
+}
+
+static void cdns_mhdp_bridge_hpd_disable(struct drm_bridge *bridge)
+{
+	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
+
+	writel(CDNS_APB_INT_MASK_SW_EVENT_INT, mhdp->regs + CDNS_APB_INT_MASK);
+}
+
+static const struct drm_bridge_funcs cdns_mhdp_bridge_funcs = {
+	.atomic_enable = cdns_mhdp_atomic_enable,
+	.atomic_disable = cdns_mhdp_atomic_disable,
+	.atomic_check = cdns_mhdp_atomic_check,
+	.attach = cdns_mhdp_attach,
+	.detach = cdns_mhdp_detach,
+	.atomic_duplicate_state = cdns_mhdp_bridge_atomic_duplicate_state,
+	.atomic_destroy_state = cdns_mhdp_bridge_atomic_destroy_state,
+	.atomic_reset = cdns_mhdp_bridge_atomic_reset,
+	.detect = cdns_mhdp_bridge_detect,
+	.get_edid = cdns_mhdp_bridge_get_edid,
+	.hpd_enable = cdns_mhdp_bridge_hpd_enable,
+	.hpd_disable = cdns_mhdp_bridge_hpd_disable,
+};
+
+static bool cdns_mhdp_detect_hpd(struct cdns_mhdp_device *mhdp, bool *hpd_pulse)
+{
+	int hpd_event, hpd_status;
+
+	*hpd_pulse = false;
+
+	hpd_event = cdns_mhdp_read_hpd_event(mhdp);
+
+	/* Getting event bits failed, bail out */
+	if (hpd_event < 0) {
+		dev_warn(mhdp->dev, "%s: read event failed: %d\n",
+			 __func__, hpd_event);
+		return false;
+	}
+
+	hpd_status = cdns_mhdp_get_hpd_status(mhdp);
+	if (hpd_status < 0) {
+		dev_warn(mhdp->dev, "%s: get hpd status failed: %d\n",
+			 __func__, hpd_status);
+		return false;
+	}
+
+	if (hpd_event & DPTX_READ_EVENT_HPD_PULSE)
+		*hpd_pulse = true;
+
+	return !!hpd_status;
+}
+
+static int cdns_mhdp_update_link_status(struct cdns_mhdp_device *mhdp)
+{
+	struct cdns_mhdp_bridge_state *cdns_bridge_state;
+	struct drm_display_mode *current_mode;
+	bool old_plugged = mhdp->plugged;
+	struct drm_bridge_state *state;
+	u8 status[DP_LINK_STATUS_SIZE];
+	bool hpd_pulse;
+	int ret = 0;
+
+	mutex_lock(&mhdp->link_mutex);
+
+	mhdp->plugged = cdns_mhdp_detect_hpd(mhdp, &hpd_pulse);
+
+	if (!mhdp->plugged) {
+		cdns_mhdp_link_down(mhdp);
+		mhdp->link.rate = mhdp->host.link_rate;
+		mhdp->link.num_lanes = mhdp->host.lanes_cnt;
+		goto out;
+	}
+
+	/*
+	 * If we get a HPD pulse event and we were and still are connected,
+	 * check the link status. If link status is ok, there's nothing to do
+	 * as we don't handle DP interrupts. If link status is bad, continue
+	 * with full link setup.
+	 */
+	if (hpd_pulse && old_plugged == mhdp->plugged) {
+		ret = drm_dp_dpcd_read_link_status(&mhdp->aux, status);
+
+		/*
+		 * If everything looks fine, just return, as we don't handle
+		 * DP IRQs.
+		 */
+		if (ret > 0 &&
+		    drm_dp_channel_eq_ok(status, mhdp->link.num_lanes) &&
+		    drm_dp_clock_recovery_ok(status, mhdp->link.num_lanes))
+			goto out;
+
+		/* If link is bad, mark link as down so that we do a new LT */
+		mhdp->link_up = false;
+	}
+
+	if (!mhdp->link_up) {
+		ret = cdns_mhdp_link_up(mhdp);
+		if (ret < 0)
+			goto out;
+	}
+
+	if (mhdp->bridge_enabled) {
+		state = drm_priv_to_bridge_state(mhdp->bridge.base.state);
+		if (!state) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		cdns_bridge_state = to_cdns_mhdp_bridge_state(state);
+		if (!cdns_bridge_state) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		current_mode = cdns_bridge_state->current_mode;
+		if (!current_mode) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		if (!cdns_mhdp_bandwidth_ok(mhdp, current_mode, mhdp->link.num_lanes,
+					    mhdp->link.rate)) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		dev_dbg(mhdp->dev, "%s: Enabling mode %s\n", __func__,
+			current_mode->name);
+
+		cdns_mhdp_sst_enable(mhdp, current_mode);
+	}
+out:
+	mutex_unlock(&mhdp->link_mutex);
+	return ret;
+}
+
+static void cdns_mhdp_modeset_retry_fn(struct work_struct *work)
+{
+	struct cdns_mhdp_device *mhdp;
+	struct drm_connector *conn;
+
+	mhdp = container_of(work, typeof(*mhdp), modeset_retry_work);
+
+	conn = &mhdp->connector;
+
+	/* Grab the locks before changing connector property */
+	mutex_lock(&conn->dev->mode_config.mutex);
+
+	/*
+	 * Set connector link status to BAD and send a Uevent to notify
+	 * userspace to do a modeset.
+	 */
+	drm_connector_set_link_status_property(conn, DRM_MODE_LINK_STATUS_BAD);
+	mutex_unlock(&conn->dev->mode_config.mutex);
+
+	/* Send Hotplug uevent so userspace can reprobe */
+	drm_kms_helper_hotplug_event(mhdp->bridge.dev);
+}
+
+static irqreturn_t cdns_mhdp_irq_handler(int irq, void *data)
+{
+	struct cdns_mhdp_device *mhdp = data;
+	u32 apb_stat, sw_ev0;
+	bool bridge_attached;
+	int ret;
+
+	apb_stat = readl(mhdp->regs + CDNS_APB_INT_STATUS);
+	if (!(apb_stat & CDNS_APB_INT_MASK_SW_EVENT_INT))
+		return IRQ_NONE;
+
+	sw_ev0 = readl(mhdp->regs + CDNS_SW_EVENT0);
+
+	/*
+	 *  Calling drm_kms_helper_hotplug_event() when not attached
+	 *  to drm device causes an oops because the drm_bridge->dev
+	 *  is NULL. See cdns_mhdp_fw_cb() comments for details about the
+	 *  problems related drm_kms_helper_hotplug_event() call.
+	 */
+	spin_lock(&mhdp->start_lock);
+	bridge_attached = mhdp->bridge_attached;
+	spin_unlock(&mhdp->start_lock);
+
+	if (bridge_attached && (sw_ev0 & CDNS_DPTX_HPD)) {
+		ret = cdns_mhdp_update_link_status(mhdp);
+		if (mhdp->connector.dev) {
+			if (ret < 0)
+				schedule_work(&mhdp->modeset_retry_work);
+			else
+				drm_kms_helper_hotplug_event(mhdp->bridge.dev);
+		} else {
+			drm_bridge_hpd_notify(&mhdp->bridge, cdns_mhdp_detect(mhdp));
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int cdns_mhdp_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct cdns_mhdp_device *mhdp;
+	unsigned long rate;
+	struct clk *clk;
+	int ret;
+	int irq;
+
+	mhdp = devm_kzalloc(dev, sizeof(*mhdp), GFP_KERNEL);
+	if (!mhdp)
+		return -ENOMEM;
+
+	clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(clk)) {
+		dev_err(dev, "couldn't get clk: %ld\n", PTR_ERR(clk));
+		return PTR_ERR(clk);
+	}
+
+	mhdp->clk = clk;
+	mhdp->dev = dev;
+	mutex_init(&mhdp->mbox_mutex);
+	mutex_init(&mhdp->link_mutex);
+	spin_lock_init(&mhdp->start_lock);
+
+	drm_dp_aux_init(&mhdp->aux);
+	mhdp->aux.dev = dev;
+	mhdp->aux.transfer = cdns_mhdp_transfer;
+
+	mhdp->regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(mhdp->regs)) {
+		dev_err(dev, "Failed to get memory resource\n");
+		return PTR_ERR(mhdp->regs);
+	}
+
+	mhdp->phy = devm_of_phy_get_by_index(dev, pdev->dev.of_node, 0);
+	if (IS_ERR(mhdp->phy)) {
+		dev_err(dev, "no PHY configured\n");
+		return PTR_ERR(mhdp->phy);
+	}
+
+	platform_set_drvdata(pdev, mhdp);
+
+	mhdp->info = of_device_get_match_data(dev);
+
+	clk_prepare_enable(clk);
+
+	pm_runtime_enable(dev);
+	ret = pm_runtime_get_sync(dev);
+	if (ret < 0) {
+		dev_err(dev, "pm_runtime_get_sync failed\n");
+		pm_runtime_disable(dev);
+		goto clk_disable;
+	}
+
+	if (mhdp->info && mhdp->info->ops && mhdp->info->ops->init) {
+		ret = mhdp->info->ops->init(mhdp);
+		if (ret != 0) {
+			dev_err(dev, "MHDP platform initialization failed: %d\n",
+				ret);
+			goto runtime_put;
+		}
+	}
+
+	rate = clk_get_rate(clk);
+	writel(rate % 1000000, mhdp->regs + CDNS_SW_CLK_L);
+	writel(rate / 1000000, mhdp->regs + CDNS_SW_CLK_H);
+
+	dev_dbg(dev, "func clk rate %lu Hz\n", rate);
+
+	writel(~0, mhdp->regs + CDNS_APB_INT_MASK);
+
+	irq = platform_get_irq(pdev, 0);
+	ret = devm_request_threaded_irq(mhdp->dev, irq, NULL,
+					cdns_mhdp_irq_handler, IRQF_ONESHOT,
+					"mhdp8546", mhdp);
+	if (ret) {
+		dev_err(dev, "cannot install IRQ %d\n", irq);
+		ret = -EIO;
+		goto plat_fini;
+	}
+
+	cdns_mhdp_fill_host_caps(mhdp);
+
+	/* Initialize link rate and num of lanes to host values */
+	mhdp->link.rate = mhdp->host.link_rate;
+	mhdp->link.num_lanes = mhdp->host.lanes_cnt;
+
+	/* The only currently supported format */
+	mhdp->display_fmt.y_only = false;
+	mhdp->display_fmt.color_format = DRM_COLOR_FORMAT_RGB444;
+	mhdp->display_fmt.bpc = 8;
+
+	mhdp->bridge.of_node = pdev->dev.of_node;
+	mhdp->bridge.funcs = &cdns_mhdp_bridge_funcs;
+	mhdp->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID |
+			   DRM_BRIDGE_OP_HPD;
+	mhdp->bridge.type = DRM_MODE_CONNECTOR_DisplayPort;
+	if (mhdp->info)
+		mhdp->bridge.timings = mhdp->info->timings;
+
+	ret = phy_init(mhdp->phy);
+	if (ret) {
+		dev_err(mhdp->dev, "Failed to initialize PHY: %d\n", ret);
+		goto plat_fini;
+	}
+
+	/* Initialize the work for modeset in case of link train failure */
+	INIT_WORK(&mhdp->modeset_retry_work, cdns_mhdp_modeset_retry_fn);
+
+	init_waitqueue_head(&mhdp->fw_load_wq);
+
+	ret = cdns_mhdp_load_firmware(mhdp);
+	if (ret)
+		goto phy_exit;
+
+	drm_bridge_add(&mhdp->bridge);
+
+	return 0;
+
+phy_exit:
+	phy_exit(mhdp->phy);
+plat_fini:
+	if (mhdp->info && mhdp->info->ops && mhdp->info->ops->exit)
+		mhdp->info->ops->exit(mhdp);
+runtime_put:
+	pm_runtime_put_sync(dev);
+	pm_runtime_disable(dev);
+clk_disable:
+	clk_disable_unprepare(mhdp->clk);
+
+	return ret;
+}
+
+static int cdns_mhdp_remove(struct platform_device *pdev)
+{
+	struct cdns_mhdp_device *mhdp = dev_get_drvdata(&pdev->dev);
+	unsigned long timeout = msecs_to_jiffies(100);
+	bool stop_fw = false;
+	int ret;
+
+	drm_bridge_remove(&mhdp->bridge);
+
+	ret = wait_event_timeout(mhdp->fw_load_wq,
+				 mhdp->hw_state == MHDP_HW_READY,
+				 timeout);
+	if (ret == 0)
+		dev_err(mhdp->dev, "%s: Timeout waiting for fw loading\n",
+			__func__);
+	else
+		stop_fw = true;
+
+	spin_lock(&mhdp->start_lock);
+	mhdp->hw_state = MHDP_HW_STOPPED;
+	spin_unlock(&mhdp->start_lock);
+
+	if (stop_fw)
+		ret = cdns_mhdp_set_firmware_active(mhdp, false);
+
+	phy_exit(mhdp->phy);
+
+	if (mhdp->info && mhdp->info->ops && mhdp->info->ops->exit)
+		mhdp->info->ops->exit(mhdp);
+
+	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+
+	cancel_work_sync(&mhdp->modeset_retry_work);
+	flush_scheduled_work();
+
+	clk_disable_unprepare(mhdp->clk);
+
+	return ret;
+}
+
+static const struct of_device_id mhdp_ids[] = {
+	{ .compatible = "cdns,mhdp8546", },
+#ifdef CONFIG_DRM_CDNS_MHDP8546_J721E
+	{ .compatible = "ti,j721e-mhdp8546",
+	  .data = &(const struct cdns_mhdp_platform_info) {
+		  .timings = &mhdp_ti_j721e_bridge_timings,
+		  .ops = &mhdp_ti_j721e_ops,
+	  },
+	},
+#endif
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, mhdp_ids);
+
+static struct platform_driver mhdp_driver = {
+	.driver	= {
+		.name		= "cdns-mhdp8546",
+		.of_match_table	= of_match_ptr(mhdp_ids),
+	},
+	.probe	= cdns_mhdp_probe,
+	.remove	= cdns_mhdp_remove,
+};
+module_platform_driver(mhdp_driver);
+
+MODULE_FIRMWARE(FW_NAME);
+
+MODULE_AUTHOR("Quentin Schulz <quentin.schulz@free-electrons.com>");
+MODULE_AUTHOR("Swapnil Jakhade <sjakhade@cadence.com>");
+MODULE_AUTHOR("Yuti Amonkar <yamonkar@cadence.com>");
+MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>");
+MODULE_AUTHOR("Jyri Sarha <jsarha@ti.com>");
+MODULE_DESCRIPTION("Cadence MHDP8546 DP bridge driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:cdns-mhdp8546");
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h
new file mode 100644
index 000000000000..5897a85e3159
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h
@@ -0,0 +1,400 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Cadence MHDP8546 DP bridge driver.
+ *
+ * Copyright (C) 2020 Cadence Design Systems, Inc.
+ *
+ * Author: Quentin Schulz <quentin.schulz@free-electrons.com>
+ *         Swapnil Jakhade <sjakhade@cadence.com>
+ */
+
+#ifndef CDNS_MHDP8546_CORE_H
+#define CDNS_MHDP8546_CORE_H
+
+#include <linux/bits.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+
+#include <drm/drm_bridge.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_dp_helper.h>
+
+struct clk;
+struct device;
+struct phy;
+
+/* Register offsets */
+#define CDNS_APB_CTRL				0x00000
+#define CDNS_CPU_STALL				BIT(3)
+
+#define CDNS_MAILBOX_FULL			0x00008
+#define CDNS_MAILBOX_EMPTY			0x0000c
+#define CDNS_MAILBOX_TX_DATA			0x00010
+#define CDNS_MAILBOX_RX_DATA			0x00014
+#define CDNS_KEEP_ALIVE				0x00018
+#define CDNS_KEEP_ALIVE_MASK			GENMASK(7, 0)
+
+#define CDNS_VER_L				0x0001C
+#define CDNS_VER_H				0x00020
+#define CDNS_LIB_L_ADDR				0x00024
+#define CDNS_LIB_H_ADDR				0x00028
+
+#define CDNS_MB_INT_MASK			0x00034
+#define CDNS_MB_INT_STATUS			0x00038
+
+#define CDNS_SW_CLK_L				0x0003c
+#define CDNS_SW_CLK_H				0x00040
+
+#define CDNS_SW_EVENT0				0x00044
+#define CDNS_DPTX_HPD				BIT(0)
+
+#define CDNS_SW_EVENT1				0x00048
+#define CDNS_SW_EVENT2				0x0004c
+#define CDNS_SW_EVENT3				0x00050
+
+#define CDNS_APB_INT_MASK			0x0006C
+#define CDNS_APB_INT_MASK_MAILBOX_INT		BIT(0)
+#define CDNS_APB_INT_MASK_SW_EVENT_INT		BIT(1)
+
+#define CDNS_APB_INT_STATUS			0x00070
+
+#define CDNS_DPTX_CAR				0x00904
+#define CDNS_VIF_CLK_EN				BIT(0)
+#define CDNS_VIF_CLK_RSTN			BIT(1)
+
+#define CDNS_SOURCE_VIDEO_IF(s)			(0x00b00 + ((s) * 0x20))
+#define CDNS_BND_HSYNC2VSYNC(s)			(CDNS_SOURCE_VIDEO_IF(s) + \
+						 0x00)
+#define CDNS_IP_DTCT_WIN			GENMASK(11, 0)
+#define CDNS_IP_DET_INTERLACE_FORMAT		BIT(12)
+#define CDNS_IP_BYPASS_V_INTERFACE		BIT(13)
+
+#define CDNS_HSYNC2VSYNC_POL_CTRL(s)		(CDNS_SOURCE_VIDEO_IF(s) + \
+						 0x10)
+#define CDNS_H2V_HSYNC_POL_ACTIVE_LOW		BIT(1)
+#define CDNS_H2V_VSYNC_POL_ACTIVE_LOW		BIT(2)
+
+#define CDNS_DPTX_PHY_CONFIG			0x02000
+#define CDNS_PHY_TRAINING_EN			BIT(0)
+#define CDNS_PHY_TRAINING_TYPE(x)		(((x) & GENMASK(3, 0)) << 1)
+#define CDNS_PHY_SCRAMBLER_BYPASS		BIT(5)
+#define CDNS_PHY_ENCODER_BYPASS			BIT(6)
+#define CDNS_PHY_SKEW_BYPASS			BIT(7)
+#define CDNS_PHY_TRAINING_AUTO			BIT(8)
+#define CDNS_PHY_LANE0_SKEW(x)			(((x) & GENMASK(2, 0)) << 9)
+#define CDNS_PHY_LANE1_SKEW(x)			(((x) & GENMASK(2, 0)) << 12)
+#define CDNS_PHY_LANE2_SKEW(x)			(((x) & GENMASK(2, 0)) << 15)
+#define CDNS_PHY_LANE3_SKEW(x)			(((x) & GENMASK(2, 0)) << 18)
+#define CDNS_PHY_COMMON_CONFIG			(CDNS_PHY_LANE1_SKEW(1) | \
+						CDNS_PHY_LANE2_SKEW(2) |  \
+						CDNS_PHY_LANE3_SKEW(3))
+#define CDNS_PHY_10BIT_EN			BIT(21)
+
+#define CDNS_DP_FRAMER_GLOBAL_CONFIG		0x02200
+#define CDNS_DP_NUM_LANES(x)			((x) - 1)
+#define CDNS_DP_MST_EN				BIT(2)
+#define CDNS_DP_FRAMER_EN			BIT(3)
+#define CDNS_DP_RATE_GOVERNOR_EN		BIT(4)
+#define CDNS_DP_NO_VIDEO_MODE			BIT(5)
+#define CDNS_DP_DISABLE_PHY_RST			BIT(6)
+#define CDNS_DP_WR_FAILING_EDGE_VSYNC		BIT(7)
+
+#define CDNS_DP_FRAMER_TU			0x02208
+#define CDNS_DP_FRAMER_TU_SIZE(x)		(((x) & GENMASK(6, 0)) << 8)
+#define CDNS_DP_FRAMER_TU_VS(x)			((x) & GENMASK(5, 0))
+#define CDNS_DP_FRAMER_TU_CNT_RST_EN		BIT(15)
+
+#define CDNS_DP_MTPH_CONTROL			0x02264
+#define CDNS_DP_MTPH_ECF_EN			BIT(0)
+#define CDNS_DP_MTPH_ACT_EN			BIT(1)
+#define CDNS_DP_MTPH_LVP_EN			BIT(2)
+
+#define CDNS_DP_MTPH_STATUS			0x0226C
+#define CDNS_DP_MTPH_ACT_STATUS			BIT(0)
+
+#define CDNS_DP_LANE_EN				0x02300
+#define CDNS_DP_LANE_EN_LANES(x)		GENMASK((x) - 1, 0)
+
+#define CDNS_DP_ENHNCD				0x02304
+
+#define CDNS_DPTX_STREAM(s)			(0x03000 + (s) * 0x80)
+#define CDNS_DP_MSA_HORIZONTAL_0(s)		(CDNS_DPTX_STREAM(s) + 0x00)
+#define CDNS_DP_MSAH0_H_TOTAL(x)		(x)
+#define CDNS_DP_MSAH0_HSYNC_START(x)		((x) << 16)
+
+#define CDNS_DP_MSA_HORIZONTAL_1(s)		(CDNS_DPTX_STREAM(s) + 0x04)
+#define CDNS_DP_MSAH1_HSYNC_WIDTH(x)		(x)
+#define CDNS_DP_MSAH1_HSYNC_POL_LOW		BIT(15)
+#define CDNS_DP_MSAH1_HDISP_WIDTH(x)		((x) << 16)
+
+#define CDNS_DP_MSA_VERTICAL_0(s)		(CDNS_DPTX_STREAM(s) + 0x08)
+#define CDNS_DP_MSAV0_V_TOTAL(x)		(x)
+#define CDNS_DP_MSAV0_VSYNC_START(x)		((x) << 16)
+
+#define CDNS_DP_MSA_VERTICAL_1(s)		(CDNS_DPTX_STREAM(s) + 0x0c)
+#define CDNS_DP_MSAV1_VSYNC_WIDTH(x)		(x)
+#define CDNS_DP_MSAV1_VSYNC_POL_LOW		BIT(15)
+#define CDNS_DP_MSAV1_VDISP_WIDTH(x)		((x) << 16)
+
+#define CDNS_DP_MSA_MISC(s)			(CDNS_DPTX_STREAM(s) + 0x10)
+#define CDNS_DP_STREAM_CONFIG(s)		(CDNS_DPTX_STREAM(s) + 0x14)
+#define CDNS_DP_STREAM_CONFIG_2(s)		(CDNS_DPTX_STREAM(s) + 0x2c)
+#define CDNS_DP_SC2_TU_VS_DIFF(x)		((x) << 8)
+
+#define CDNS_DP_HORIZONTAL(s)			(CDNS_DPTX_STREAM(s) + 0x30)
+#define CDNS_DP_H_HSYNC_WIDTH(x)		(x)
+#define CDNS_DP_H_H_TOTAL(x)			((x) << 16)
+
+#define CDNS_DP_VERTICAL_0(s)			(CDNS_DPTX_STREAM(s) + 0x34)
+#define CDNS_DP_V0_VHEIGHT(x)			(x)
+#define CDNS_DP_V0_VSTART(x)			((x) << 16)
+
+#define CDNS_DP_VERTICAL_1(s)			(CDNS_DPTX_STREAM(s) + 0x38)
+#define CDNS_DP_V1_VTOTAL(x)			(x)
+#define CDNS_DP_V1_VTOTAL_EVEN			BIT(16)
+
+#define CDNS_DP_MST_SLOT_ALLOCATE(s)		(CDNS_DPTX_STREAM(s) + 0x44)
+#define CDNS_DP_S_ALLOC_START_SLOT(x)		(x)
+#define CDNS_DP_S_ALLOC_END_SLOT(x)		((x) << 8)
+
+#define CDNS_DP_RATE_GOVERNING(s)		(CDNS_DPTX_STREAM(s) + 0x48)
+#define CDNS_DP_RG_TARG_AV_SLOTS_Y(x)		(x)
+#define CDNS_DP_RG_TARG_AV_SLOTS_X(x)		((x) << 4)
+#define CDNS_DP_RG_ENABLE			BIT(10)
+
+#define CDNS_DP_FRAMER_PXL_REPR(s)		(CDNS_DPTX_STREAM(s) + 0x4c)
+#define CDNS_DP_FRAMER_6_BPC			BIT(0)
+#define CDNS_DP_FRAMER_8_BPC			BIT(1)
+#define CDNS_DP_FRAMER_10_BPC			BIT(2)
+#define CDNS_DP_FRAMER_12_BPC			BIT(3)
+#define CDNS_DP_FRAMER_16_BPC			BIT(4)
+#define CDNS_DP_FRAMER_PXL_FORMAT		0x8
+#define CDNS_DP_FRAMER_RGB			BIT(0)
+#define CDNS_DP_FRAMER_YCBCR444			BIT(1)
+#define CDNS_DP_FRAMER_YCBCR422			BIT(2)
+#define CDNS_DP_FRAMER_YCBCR420			BIT(3)
+#define CDNS_DP_FRAMER_Y_ONLY			BIT(4)
+
+#define CDNS_DP_FRAMER_SP(s)			(CDNS_DPTX_STREAM(s) + 0x50)
+#define CDNS_DP_FRAMER_VSYNC_POL_LOW		BIT(0)
+#define CDNS_DP_FRAMER_HSYNC_POL_LOW		BIT(1)
+#define CDNS_DP_FRAMER_INTERLACE		BIT(2)
+
+#define CDNS_DP_LINE_THRESH(s)			(CDNS_DPTX_STREAM(s) + 0x64)
+#define CDNS_DP_ACTIVE_LINE_THRESH(x)		(x)
+
+#define CDNS_DP_VB_ID(s)			(CDNS_DPTX_STREAM(s) + 0x68)
+#define CDNS_DP_VB_ID_INTERLACED		BIT(2)
+#define CDNS_DP_VB_ID_COMPRESSED		BIT(6)
+
+#define CDNS_DP_FRONT_BACK_PORCH(s)		(CDNS_DPTX_STREAM(s) + 0x78)
+#define CDNS_DP_BACK_PORCH(x)			(x)
+#define CDNS_DP_FRONT_PORCH(x)			((x) << 16)
+
+#define CDNS_DP_BYTE_COUNT(s)			(CDNS_DPTX_STREAM(s) + 0x7c)
+#define CDNS_DP_BYTE_COUNT_BYTES_IN_CHUNK_SHIFT	16
+
+/* mailbox */
+#define MAILBOX_RETRY_US			1000
+#define MAILBOX_TIMEOUT_US			2000000
+
+#define MB_OPCODE_ID				0
+#define MB_MODULE_ID				1
+#define MB_SIZE_MSB_ID				2
+#define MB_SIZE_LSB_ID				3
+#define MB_DATA_ID				4
+
+#define MB_MODULE_ID_DP_TX			0x01
+#define MB_MODULE_ID_HDCP_TX			0x07
+#define MB_MODULE_ID_HDCP_RX			0x08
+#define MB_MODULE_ID_HDCP_GENERAL		0x09
+#define MB_MODULE_ID_GENERAL			0x0a
+
+/* firmware and opcodes */
+#define FW_NAME					"cadence/mhdp8546.bin"
+#define CDNS_MHDP_IMEM				0x10000
+
+#define GENERAL_MAIN_CONTROL			0x01
+#define GENERAL_TEST_ECHO			0x02
+#define GENERAL_BUS_SETTINGS			0x03
+#define GENERAL_TEST_ACCESS			0x04
+#define GENERAL_REGISTER_READ			0x07
+
+#define DPTX_SET_POWER_MNG			0x00
+#define DPTX_GET_EDID				0x02
+#define DPTX_READ_DPCD				0x03
+#define DPTX_WRITE_DPCD				0x04
+#define DPTX_ENABLE_EVENT			0x05
+#define DPTX_WRITE_REGISTER			0x06
+#define DPTX_READ_REGISTER			0x07
+#define DPTX_WRITE_FIELD			0x08
+#define DPTX_READ_EVENT				0x0a
+#define DPTX_GET_LAST_AUX_STAUS			0x0e
+#define DPTX_HPD_STATE				0x11
+#define DPTX_ADJUST_LT				0x12
+
+#define FW_STANDBY				0
+#define FW_ACTIVE				1
+
+/* HPD */
+#define DPTX_READ_EVENT_HPD_TO_HIGH             BIT(0)
+#define DPTX_READ_EVENT_HPD_TO_LOW              BIT(1)
+#define DPTX_READ_EVENT_HPD_PULSE               BIT(2)
+#define DPTX_READ_EVENT_HPD_STATE               BIT(3)
+
+/* general */
+#define CDNS_DP_TRAINING_PATTERN_4		0x7
+
+#define CDNS_KEEP_ALIVE_TIMEOUT			2000
+
+#define CDNS_VOLT_SWING(x)			((x) & GENMASK(1, 0))
+#define CDNS_FORCE_VOLT_SWING			BIT(2)
+
+#define CDNS_PRE_EMPHASIS(x)			((x) & GENMASK(1, 0))
+#define CDNS_FORCE_PRE_EMPHASIS			BIT(2)
+
+#define CDNS_SUPPORT_TPS(x)			BIT((x) - 1)
+
+#define CDNS_FAST_LINK_TRAINING			BIT(0)
+
+#define CDNS_LANE_MAPPING_TYPE_C_LANE_0(x)	((x) & GENMASK(1, 0))
+#define CDNS_LANE_MAPPING_TYPE_C_LANE_1(x)	((x) & GENMASK(3, 2))
+#define CDNS_LANE_MAPPING_TYPE_C_LANE_2(x)	((x) & GENMASK(5, 4))
+#define CDNS_LANE_MAPPING_TYPE_C_LANE_3(x)	((x) & GENMASK(7, 6))
+#define CDNS_LANE_MAPPING_NORMAL		0xe4
+#define CDNS_LANE_MAPPING_FLIPPED		0x1b
+
+#define CDNS_DP_MAX_NUM_LANES			4
+#define CDNS_DP_TEST_VSC_SDP			BIT(6) /* 1.3+ */
+#define CDNS_DP_TEST_COLOR_FORMAT_RAW_Y_ONLY	BIT(7)
+
+#define CDNS_MHDP_MAX_STREAMS			4
+
+#define DP_LINK_CAP_ENHANCED_FRAMING		BIT(0)
+
+struct cdns_mhdp_link {
+	unsigned char revision;
+	unsigned int rate;
+	unsigned int num_lanes;
+	unsigned long capabilities;
+};
+
+struct cdns_mhdp_host {
+	unsigned int link_rate;
+	u8 lanes_cnt;
+	u8 volt_swing;
+	u8 pre_emphasis;
+	u8 pattern_supp;
+	u8 lane_mapping;
+	bool fast_link;
+	bool enhanced;
+	bool scrambler;
+	bool ssc;
+};
+
+struct cdns_mhdp_sink {
+	unsigned int link_rate;
+	u8 lanes_cnt;
+	u8 pattern_supp;
+	bool fast_link;
+	bool enhanced;
+	bool ssc;
+};
+
+struct cdns_mhdp_display_fmt {
+	u32 color_format;
+	u32 bpc;
+	bool y_only;
+};
+
+/*
+ * These enums present MHDP hw initialization state
+ * Legal state transitions are:
+ * MHDP_HW_READY <-> MHDP_HW_STOPPED
+ */
+enum mhdp_hw_state {
+	MHDP_HW_READY = 1,	/* HW ready, FW active */
+	MHDP_HW_STOPPED		/* Driver removal FW to be stopped */
+};
+
+struct cdns_mhdp_device;
+
+struct mhdp_platform_ops {
+	int (*init)(struct cdns_mhdp_device *mhdp);
+	void (*exit)(struct cdns_mhdp_device *mhdp);
+	void (*enable)(struct cdns_mhdp_device *mhdp);
+	void (*disable)(struct cdns_mhdp_device *mhdp);
+};
+
+struct cdns_mhdp_bridge_state {
+	struct drm_bridge_state base;
+	struct drm_display_mode *current_mode;
+};
+
+struct cdns_mhdp_platform_info {
+	const struct drm_bridge_timings *timings;
+	const struct mhdp_platform_ops *ops;
+};
+
+#define to_cdns_mhdp_bridge_state(s) \
+		container_of(s, struct cdns_mhdp_bridge_state, base)
+
+struct cdns_mhdp_device {
+	void __iomem *regs;
+	void __iomem *j721e_regs;
+
+	struct device *dev;
+	struct clk *clk;
+	struct phy *phy;
+
+	const struct cdns_mhdp_platform_info *info;
+
+	/* This is to protect mailbox communications with the firmware */
+	struct mutex mbox_mutex;
+
+	/*
+	 * "link_mutex" protects the access to all the link parameters
+	 * including the link training process. Link training will be
+	 * invoked both from threaded interrupt handler and from atomic
+	 * callbacks when link_up is not set. So this mutex protects
+	 * flags such as link_up, bridge_enabled, link.num_lanes,
+	 * link.rate etc.
+	 */
+	struct mutex link_mutex;
+
+	struct drm_connector connector;
+	struct drm_bridge bridge;
+
+	struct cdns_mhdp_link link;
+	struct drm_dp_aux aux;
+
+	struct cdns_mhdp_host host;
+	struct cdns_mhdp_sink sink;
+	struct cdns_mhdp_display_fmt display_fmt;
+	u8 stream_id;
+
+	bool link_up;
+	bool plugged;
+
+	/*
+	 * "start_lock" protects the access to bridge_attached and
+	 * hw_state data members that control the delayed firmware
+	 * loading and attaching the bridge. They are accessed from
+	 * both the DRM core and cdns_mhdp_fw_cb(). In most cases just
+	 * protecting the data members is enough, but the irq mask
+	 * setting needs to be protected when enabling the FW.
+	 */
+	spinlock_t start_lock;
+	bool bridge_attached;
+	bool bridge_enabled;
+	enum mhdp_hw_state hw_state;
+	wait_queue_head_t fw_load_wq;
+
+	/* Work struct to schedule a uevent on link train failure */
+	struct work_struct modeset_retry_work;
+};
+
+#define connector_to_mhdp(x) container_of(x, struct cdns_mhdp_device, connector)
+#define bridge_to_mhdp(x) container_of(x, struct cdns_mhdp_device, bridge)
+
+#endif
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.c
new file mode 100644
index 000000000000..dfe1b59514f7
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * TI j721e Cadence MHDP8546 DP wrapper
+ *
+ * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/
+ * Author: Jyri Sarha <jsarha@ti.com>
+ */
+
+#include <linux/io.h>
+#include <linux/platform_device.h>
+
+#include "cdns-mhdp8546-j721e.h"
+
+#define	REVISION			0x00
+#define	DPTX_IPCFG			0x04
+#define	ECC_MEM_CFG			0x08
+#define	DPTX_DSC_CFG			0x0c
+#define	DPTX_SRC_CFG			0x10
+#define	DPTX_VIF_SECURE_MODE_CFG	0x14
+#define	DPTX_VIF_CONN_STATUS		0x18
+#define	PHY_CLK_STATUS			0x1c
+
+#define DPTX_SRC_AIF_EN			BIT(16)
+#define DPTX_SRC_VIF_3_IN30B		BIT(11)
+#define DPTX_SRC_VIF_2_IN30B		BIT(10)
+#define DPTX_SRC_VIF_1_IN30B		BIT(9)
+#define DPTX_SRC_VIF_0_IN30B		BIT(8)
+#define DPTX_SRC_VIF_3_SEL_DPI5		BIT(7)
+#define DPTX_SRC_VIF_3_SEL_DPI3		0
+#define DPTX_SRC_VIF_2_SEL_DPI4		BIT(6)
+#define DPTX_SRC_VIF_2_SEL_DPI2		0
+#define DPTX_SRC_VIF_1_SEL_DPI3		BIT(5)
+#define DPTX_SRC_VIF_1_SEL_DPI1		0
+#define DPTX_SRC_VIF_0_SEL_DPI2		BIT(4)
+#define DPTX_SRC_VIF_0_SEL_DPI0		0
+#define DPTX_SRC_VIF_3_EN		BIT(3)
+#define DPTX_SRC_VIF_2_EN		BIT(2)
+#define DPTX_SRC_VIF_1_EN		BIT(1)
+#define DPTX_SRC_VIF_0_EN		BIT(0)
+
+/* TODO turn DPTX_IPCFG fw_mem_clk_en at pm_runtime_suspend. */
+
+static int cdns_mhdp_j721e_init(struct cdns_mhdp_device *mhdp)
+{
+	struct platform_device *pdev = to_platform_device(mhdp->dev);
+
+	mhdp->j721e_regs = devm_platform_ioremap_resource(pdev, 1);
+	return PTR_ERR_OR_ZERO(mhdp->j721e_regs);
+}
+
+static void cdns_mhdp_j721e_enable(struct cdns_mhdp_device *mhdp)
+{
+	/*
+	 * Enable VIF_0 and select DPI2 as its input. DSS0 DPI0 is connected
+	 * to eDP DPI2. This is the only supported SST configuration on
+	 * J721E.
+	 */
+	writel(DPTX_SRC_VIF_0_EN | DPTX_SRC_VIF_0_SEL_DPI2,
+	       mhdp->j721e_regs + DPTX_SRC_CFG);
+}
+
+static void cdns_mhdp_j721e_disable(struct cdns_mhdp_device *mhdp)
+{
+	/* Put everything to defaults  */
+	writel(0, mhdp->j721e_regs + DPTX_DSC_CFG);
+}
+
+const struct mhdp_platform_ops mhdp_ti_j721e_ops = {
+	.init = cdns_mhdp_j721e_init,
+	.enable = cdns_mhdp_j721e_enable,
+	.disable = cdns_mhdp_j721e_disable,
+};
+
+const struct drm_bridge_timings mhdp_ti_j721e_bridge_timings = {
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE |
+			   DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE |
+			   DRM_BUS_FLAG_DE_HIGH,
+};
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.h b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.h
new file mode 100644
index 000000000000..97d20d115a24
--- /dev/null
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-j721e.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * TI j721e Cadence MHDP8546 DP wrapper
+ *
+ * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/
+ * Author: Jyri Sarha <jsarha@ti.com>
+ */
+
+#ifndef CDNS_MHDP8546_J721E_H
+#define CDNS_MHDP8546_J721E_H
+
+#include "cdns-mhdp8546-core.h"
+
+struct mhdp_platform_ops;
+
+extern const struct mhdp_platform_ops mhdp_ti_j721e_ops;
+extern const struct drm_bridge_timings mhdp_ti_j721e_bridge_timings;
+
+#endif /* !CDNS_MHDP8546_J721E_H */
diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c
new file mode 100644
index 000000000000..d734d9402c35
--- /dev/null
+++ b/drivers/gpu/drm/bridge/lontium-lt9611.c
@@ -0,0 +1,1230 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2018, The Linux Foundation. All rights reserved.
+ * Copyright (c) 2019-2020. Linaro Limited.
+ */
+
+#include <linux/gpio/consumer.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/of_graph.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
+
+#include <sound/hdmi-codec.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+
+#define EDID_SEG_SIZE	256
+#define EDID_LEN	32
+#define EDID_LOOP	8
+#define KEY_DDC_ACCS_DONE 0x02
+#define DDC_NO_ACK	0x50
+
+#define LT9611_4LANES	0
+
+struct lt9611 {
+	struct device *dev;
+	struct drm_bridge bridge;
+	struct drm_connector connector;
+
+	struct regmap *regmap;
+
+	struct device_node *dsi0_node;
+	struct device_node *dsi1_node;
+	struct mipi_dsi_device *dsi0;
+	struct mipi_dsi_device *dsi1;
+	struct platform_device *audio_pdev;
+
+	bool ac_mode;
+
+	struct gpio_desc *reset_gpio;
+	struct gpio_desc *enable_gpio;
+
+	bool power_on;
+	bool sleep;
+
+	struct regulator_bulk_data supplies[2];
+
+	struct i2c_client *client;
+
+	enum drm_connector_status status;
+
+	u8 edid_buf[EDID_SEG_SIZE];
+	u32 vic;
+};
+
+#define LT9611_PAGE_CONTROL	0xff
+
+static const struct regmap_range_cfg lt9611_ranges[] = {
+	{
+		.name = "register_range",
+		.range_min =  0,
+		.range_max = 0x85ff,
+		.selector_reg = LT9611_PAGE_CONTROL,
+		.selector_mask = 0xff,
+		.selector_shift = 0,
+		.window_start = 0,
+		.window_len = 0x100,
+	},
+};
+
+static const struct regmap_config lt9611_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = 0xffff,
+	.ranges = lt9611_ranges,
+	.num_ranges = ARRAY_SIZE(lt9611_ranges),
+};
+
+struct lt9611_mode {
+	u16 hdisplay;
+	u16 vdisplay;
+	u8 vrefresh;
+	u8 lanes;
+	u8 intfs;
+};
+
+static struct lt9611_mode lt9611_modes[] = {
+	{ 3840, 2160, 30, 4, 2 }, /* 3840x2160 24bit 30Hz 4Lane 2ports */
+	{ 1920, 1080, 60, 4, 1 }, /* 1080P 24bit 60Hz 4lane 1port */
+	{ 1920, 1080, 30, 3, 1 }, /* 1080P 24bit 30Hz 3lane 1port */
+	{ 1920, 1080, 24, 3, 1 },
+	{ 720, 480, 60, 4, 1 },
+	{ 720, 576, 50, 2, 1 },
+	{ 640, 480, 60, 2, 1 },
+};
+
+static struct lt9611 *bridge_to_lt9611(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct lt9611, bridge);
+}
+
+static struct lt9611 *connector_to_lt9611(struct drm_connector *connector)
+{
+	return container_of(connector, struct lt9611, connector);
+}
+
+static int lt9611_mipi_input_analog(struct lt9611 *lt9611)
+{
+	const struct reg_sequence reg_cfg[] = {
+		{ 0x8106, 0x40 }, /* port A rx current */
+		{ 0x810a, 0xfe }, /* port A ldo voltage set */
+		{ 0x810b, 0xbf }, /* enable port A lprx */
+		{ 0x8111, 0x40 }, /* port B rx current */
+		{ 0x8115, 0xfe }, /* port B ldo voltage set */
+		{ 0x8116, 0xbf }, /* enable port B lprx */
+
+		{ 0x811c, 0x03 }, /* PortA clk lane no-LP mode */
+		{ 0x8120, 0x03 }, /* PortB clk lane with-LP mode */
+	};
+
+	return regmap_multi_reg_write(lt9611->regmap, reg_cfg, ARRAY_SIZE(reg_cfg));
+}
+
+static int lt9611_mipi_input_digital(struct lt9611 *lt9611,
+				     const struct drm_display_mode *mode)
+{
+	struct reg_sequence reg_cfg[] = {
+		{ 0x8300, LT9611_4LANES },
+		{ 0x830a, 0x00 },
+		{ 0x824f, 0x80 },
+		{ 0x8250, 0x10 },
+		{ 0x8302, 0x0a },
+		{ 0x8306, 0x0a },
+	};
+
+	if (mode->hdisplay == 3840)
+		reg_cfg[1].def = 0x03;
+
+	return regmap_multi_reg_write(lt9611->regmap, reg_cfg, ARRAY_SIZE(reg_cfg));
+}
+
+static void lt9611_mipi_video_setup(struct lt9611 *lt9611,
+				    const struct drm_display_mode *mode)
+{
+	u32 h_total, hactive, hsync_len, hfront_porch, hsync_porch;
+	u32 v_total, vactive, vsync_len, vfront_porch, vsync_porch;
+
+	h_total = mode->htotal;
+	v_total = mode->vtotal;
+
+	hactive = mode->hdisplay;
+	hsync_len = mode->hsync_end - mode->hsync_start;
+	hfront_porch = mode->hsync_start - mode->hdisplay;
+	hsync_porch = hsync_len + mode->htotal - mode->hsync_end;
+
+	vactive = mode->vdisplay;
+	vsync_len = mode->vsync_end - mode->vsync_start;
+	vfront_porch = mode->vsync_start - mode->vdisplay;
+	vsync_porch = vsync_len + mode->vtotal - mode->vsync_end;
+
+	regmap_write(lt9611->regmap, 0x830d, (u8)(v_total / 256));
+	regmap_write(lt9611->regmap, 0x830e, (u8)(v_total % 256));
+
+	regmap_write(lt9611->regmap, 0x830f, (u8)(vactive / 256));
+	regmap_write(lt9611->regmap, 0x8310, (u8)(vactive % 256));
+
+	regmap_write(lt9611->regmap, 0x8311, (u8)(h_total / 256));
+	regmap_write(lt9611->regmap, 0x8312, (u8)(h_total % 256));
+
+	regmap_write(lt9611->regmap, 0x8313, (u8)(hactive / 256));
+	regmap_write(lt9611->regmap, 0x8314, (u8)(hactive % 256));
+
+	regmap_write(lt9611->regmap, 0x8315, (u8)(vsync_len % 256));
+	regmap_write(lt9611->regmap, 0x8316, (u8)(hsync_len % 256));
+
+	regmap_write(lt9611->regmap, 0x8317, (u8)(vfront_porch % 256));
+
+	regmap_write(lt9611->regmap, 0x8318, (u8)(vsync_porch % 256));
+
+	regmap_write(lt9611->regmap, 0x8319, (u8)(hfront_porch % 256));
+
+	regmap_write(lt9611->regmap, 0x831a, (u8)(hsync_porch / 256));
+	regmap_write(lt9611->regmap, 0x831b, (u8)(hsync_porch % 256));
+}
+
+static void lt9611_pcr_setup(struct lt9611 *lt9611, const struct drm_display_mode *mode)
+{
+	const struct reg_sequence reg_cfg[] = {
+		{ 0x830b, 0x01 },
+		{ 0x830c, 0x10 },
+		{ 0x8348, 0x00 },
+		{ 0x8349, 0x81 },
+
+		/* stage 1 */
+		{ 0x8321, 0x4a },
+		{ 0x8324, 0x71 },
+		{ 0x8325, 0x30 },
+		{ 0x832a, 0x01 },
+
+		/* stage 2 */
+		{ 0x834a, 0x40 },
+		{ 0x831d, 0x10 },
+
+		/* MK limit */
+		{ 0x832d, 0x38 },
+		{ 0x8331, 0x08 },
+	};
+	const struct reg_sequence reg_cfg2[] = {
+		{ 0x830b, 0x03 },
+		{ 0x830c, 0xd0 },
+		{ 0x8348, 0x03 },
+		{ 0x8349, 0xe0 },
+		{ 0x8324, 0x72 },
+		{ 0x8325, 0x00 },
+		{ 0x832a, 0x01 },
+		{ 0x834a, 0x10 },
+		{ 0x831d, 0x10 },
+		{ 0x8326, 0x37 },
+	};
+
+	regmap_multi_reg_write(lt9611->regmap, reg_cfg, ARRAY_SIZE(reg_cfg));
+
+	switch (mode->hdisplay) {
+	case 640:
+		regmap_write(lt9611->regmap, 0x8326, 0x14);
+		break;
+	case 1920:
+		regmap_write(lt9611->regmap, 0x8326, 0x37);
+		break;
+	case 3840:
+		regmap_multi_reg_write(lt9611->regmap, reg_cfg2, ARRAY_SIZE(reg_cfg2));
+		break;
+	}
+
+	/* pcr rst */
+	regmap_write(lt9611->regmap, 0x8011, 0x5a);
+	regmap_write(lt9611->regmap, 0x8011, 0xfa);
+}
+
+static int lt9611_pll_setup(struct lt9611 *lt9611, const struct drm_display_mode *mode)
+{
+	unsigned int pclk = mode->clock;
+	const struct reg_sequence reg_cfg[] = {
+		/* txpll init */
+		{ 0x8123, 0x40 },
+		{ 0x8124, 0x64 },
+		{ 0x8125, 0x80 },
+		{ 0x8126, 0x55 },
+		{ 0x812c, 0x37 },
+		{ 0x812f, 0x01 },
+		{ 0x8126, 0x55 },
+		{ 0x8127, 0x66 },
+		{ 0x8128, 0x88 },
+	};
+
+	regmap_multi_reg_write(lt9611->regmap, reg_cfg, ARRAY_SIZE(reg_cfg));
+
+	if (pclk > 150000)
+		regmap_write(lt9611->regmap, 0x812d, 0x88);
+	else if (pclk > 70000)
+		regmap_write(lt9611->regmap, 0x812d, 0x99);
+	else
+		regmap_write(lt9611->regmap, 0x812d, 0xaa);
+
+	/*
+	 * first divide pclk by 2 first
+	 *  - write divide by 64k to 19:16 bits which means shift by 17
+	 *  - write divide by 256 to 15:8 bits which means shift by 9
+	 *  - write remainder to 7:0 bits, which means shift by 1
+	 */
+	regmap_write(lt9611->regmap, 0x82e3, pclk >> 17); /* pclk[19:16] */
+	regmap_write(lt9611->regmap, 0x82e4, pclk >> 9);  /* pclk[15:8]  */
+	regmap_write(lt9611->regmap, 0x82e5, pclk >> 1);  /* pclk[7:0]   */
+
+	regmap_write(lt9611->regmap, 0x82de, 0x20);
+	regmap_write(lt9611->regmap, 0x82de, 0xe0);
+
+	regmap_write(lt9611->regmap, 0x8016, 0xf1);
+	regmap_write(lt9611->regmap, 0x8016, 0xf3);
+
+	return 0;
+}
+
+static int lt9611_read_video_check(struct lt9611 *lt9611, unsigned int reg)
+{
+	unsigned int temp, temp2;
+	int ret;
+
+	ret = regmap_read(lt9611->regmap, reg, &temp);
+	if (ret)
+		return ret;
+	temp <<= 8;
+	ret = regmap_read(lt9611->regmap, reg + 1, &temp2);
+	if (ret)
+		return ret;
+
+	return (temp + temp2);
+}
+
+static int lt9611_video_check(struct lt9611 *lt9611)
+{
+	u32 v_total, vactive, hactive_a, hactive_b, h_total_sysclk;
+	int temp;
+
+	/* top module video check */
+
+	/* vactive */
+	temp = lt9611_read_video_check(lt9611, 0x8282);
+	if (temp < 0)
+		goto end;
+	vactive = temp;
+
+	/* v_total */
+	temp = lt9611_read_video_check(lt9611, 0x826c);
+	if (temp < 0)
+		goto end;
+	v_total = temp;
+
+	/* h_total_sysclk */
+	temp = lt9611_read_video_check(lt9611, 0x8286);
+	if (temp < 0)
+		goto end;
+	h_total_sysclk = temp;
+
+	/* hactive_a */
+	temp = lt9611_read_video_check(lt9611, 0x8382);
+	if (temp < 0)
+		goto end;
+	hactive_a = temp / 3;
+
+	/* hactive_b */
+	temp = lt9611_read_video_check(lt9611, 0x8386);
+	if (temp < 0)
+		goto end;
+	hactive_b = temp / 3;
+
+	dev_info(lt9611->dev,
+		 "video check: hactive_a=%d, hactive_b=%d, vactive=%d, v_total=%d, h_total_sysclk=%d\n",
+		 hactive_a, hactive_b, vactive, v_total, h_total_sysclk);
+
+	return 0;
+
+end:
+	dev_err(lt9611->dev, "read video check error\n");
+	return temp;
+}
+
+static void lt9611_hdmi_tx_digital(struct lt9611 *lt9611)
+{
+	regmap_write(lt9611->regmap, 0x8443, 0x46 - lt9611->vic);
+	regmap_write(lt9611->regmap, 0x8447, lt9611->vic);
+	regmap_write(lt9611->regmap, 0x843d, 0x0a); /* UD1 infoframe */
+
+	regmap_write(lt9611->regmap, 0x82d6, 0x8c);
+	regmap_write(lt9611->regmap, 0x82d7, 0x04);
+}
+
+static void lt9611_hdmi_tx_phy(struct lt9611 *lt9611)
+{
+	struct reg_sequence reg_cfg[] = {
+		{ 0x8130, 0x6a },
+		{ 0x8131, 0x44 }, /* HDMI DC mode */
+		{ 0x8132, 0x4a },
+		{ 0x8133, 0x0b },
+		{ 0x8134, 0x00 },
+		{ 0x8135, 0x00 },
+		{ 0x8136, 0x00 },
+		{ 0x8137, 0x44 },
+		{ 0x813f, 0x0f },
+		{ 0x8140, 0xa0 },
+		{ 0x8141, 0xa0 },
+		{ 0x8142, 0xa0 },
+		{ 0x8143, 0xa0 },
+		{ 0x8144, 0x0a },
+	};
+
+	/* HDMI AC mode */
+	if (lt9611->ac_mode)
+		reg_cfg[2].def = 0x73;
+
+	regmap_multi_reg_write(lt9611->regmap, reg_cfg, ARRAY_SIZE(reg_cfg));
+}
+
+static irqreturn_t lt9611_irq_thread_handler(int irq, void *dev_id)
+{
+	struct lt9611 *lt9611 = dev_id;
+	unsigned int irq_flag0 = 0;
+	unsigned int irq_flag3 = 0;
+
+	regmap_read(lt9611->regmap, 0x820f, &irq_flag3);
+	regmap_read(lt9611->regmap, 0x820c, &irq_flag0);
+
+	/* hpd changed low */
+	if (irq_flag3 & 0x80) {
+		dev_info(lt9611->dev, "hdmi cable disconnected\n");
+
+		regmap_write(lt9611->regmap, 0x8207, 0xbf);
+		regmap_write(lt9611->regmap, 0x8207, 0x3f);
+	}
+
+	/* hpd changed high */
+	if (irq_flag3 & 0x40) {
+		dev_info(lt9611->dev, "hdmi cable connected\n");
+
+		regmap_write(lt9611->regmap, 0x8207, 0x7f);
+		regmap_write(lt9611->regmap, 0x8207, 0x3f);
+	}
+
+	if (irq_flag3 & 0xc0 && lt9611->bridge.dev)
+		drm_kms_helper_hotplug_event(lt9611->bridge.dev);
+
+	/* video input changed */
+	if (irq_flag0 & 0x01) {
+		dev_info(lt9611->dev, "video input changed\n");
+		regmap_write(lt9611->regmap, 0x829e, 0xff);
+		regmap_write(lt9611->regmap, 0x829e, 0xf7);
+		regmap_write(lt9611->regmap, 0x8204, 0xff);
+		regmap_write(lt9611->regmap, 0x8204, 0xfe);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void lt9611_enable_hpd_interrupts(struct lt9611 *lt9611)
+{
+	unsigned int val;
+
+	regmap_read(lt9611->regmap, 0x8203, &val);
+
+	val &= ~0xc0;
+	regmap_write(lt9611->regmap, 0x8203, val);
+	regmap_write(lt9611->regmap, 0x8207, 0xff); /* clear */
+	regmap_write(lt9611->regmap, 0x8207, 0x3f);
+}
+
+static void lt9611_sleep_setup(struct lt9611 *lt9611)
+{
+	const struct reg_sequence sleep_setup[] = {
+		{ 0x8024, 0x76 },
+		{ 0x8023, 0x01 },
+		{ 0x8157, 0x03 }, /* set addr pin as output */
+		{ 0x8149, 0x0b },
+		{ 0x8151, 0x30 }, /* disable IRQ */
+		{ 0x8102, 0x48 }, /* MIPI Rx power down */
+		{ 0x8123, 0x80 },
+		{ 0x8130, 0x00 },
+		{ 0x8100, 0x01 }, /* bandgap power down */
+		{ 0x8101, 0x00 }, /* system clk power down */
+	};
+
+	regmap_multi_reg_write(lt9611->regmap,
+			       sleep_setup, ARRAY_SIZE(sleep_setup));
+	lt9611->sleep = true;
+}
+
+static int lt9611_power_on(struct lt9611 *lt9611)
+{
+	int ret;
+	const struct reg_sequence seq[] = {
+		/* LT9611_System_Init */
+		{ 0x8101, 0x18 }, /* sel xtal clock */
+
+		/* timer for frequency meter */
+		{ 0x821b, 0x69 }, /* timer 2 */
+		{ 0x821c, 0x78 },
+		{ 0x82cb, 0x69 }, /* timer 1 */
+		{ 0x82cc, 0x78 },
+
+		/* irq init */
+		{ 0x8251, 0x01 },
+		{ 0x8258, 0x0a }, /* hpd irq */
+		{ 0x8259, 0x80 }, /* hpd debounce width */
+		{ 0x829e, 0xf7 }, /* video check irq */
+
+		/* power consumption for work */
+		{ 0x8004, 0xf0 },
+		{ 0x8006, 0xf0 },
+		{ 0x800a, 0x80 },
+		{ 0x800b, 0x40 },
+		{ 0x800d, 0xef },
+		{ 0x8011, 0xfa },
+	};
+
+	if (lt9611->power_on)
+		return 0;
+
+	ret = regmap_multi_reg_write(lt9611->regmap, seq, ARRAY_SIZE(seq));
+	if (!ret)
+		lt9611->power_on = true;
+
+	return ret;
+}
+
+static int lt9611_power_off(struct lt9611 *lt9611)
+{
+	int ret;
+
+	ret = regmap_write(lt9611->regmap, 0x8130, 0x6a);
+	if (!ret)
+		lt9611->power_on = false;
+
+	return ret;
+}
+
+static void lt9611_reset(struct lt9611 *lt9611)
+{
+	gpiod_set_value_cansleep(lt9611->reset_gpio, 1);
+	msleep(20);
+
+	gpiod_set_value_cansleep(lt9611->reset_gpio, 0);
+	msleep(20);
+
+	gpiod_set_value_cansleep(lt9611->reset_gpio, 1);
+	msleep(100);
+}
+
+static void lt9611_assert_5v(struct lt9611 *lt9611)
+{
+	if (!lt9611->enable_gpio)
+		return;
+
+	gpiod_set_value_cansleep(lt9611->enable_gpio, 1);
+	msleep(20);
+}
+
+static int lt9611_regulator_init(struct lt9611 *lt9611)
+{
+	int ret;
+
+	lt9611->supplies[0].supply = "vdd";
+	lt9611->supplies[1].supply = "vcc";
+
+	ret = devm_regulator_bulk_get(lt9611->dev, 2, lt9611->supplies);
+	if (ret < 0)
+		return ret;
+
+	return regulator_set_load(lt9611->supplies[0].consumer, 300000);
+}
+
+static int lt9611_regulator_enable(struct lt9611 *lt9611)
+{
+	int ret;
+
+	ret = regulator_enable(lt9611->supplies[0].consumer);
+	if (ret < 0)
+		return ret;
+
+	usleep_range(1000, 10000);
+
+	ret = regulator_enable(lt9611->supplies[1].consumer);
+	if (ret < 0) {
+		regulator_disable(lt9611->supplies[0].consumer);
+		return ret;
+	}
+
+	return 0;
+}
+
+static struct lt9611_mode *lt9611_find_mode(const struct drm_display_mode *mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(lt9611_modes); i++) {
+		if (lt9611_modes[i].hdisplay == mode->hdisplay &&
+		    lt9611_modes[i].vdisplay == mode->vdisplay &&
+		    lt9611_modes[i].vrefresh == drm_mode_vrefresh(mode)) {
+			return &lt9611_modes[i];
+		}
+	}
+
+	return NULL;
+}
+
+/* connector funcs */
+static enum drm_connector_status
+lt9611_connector_detect(struct drm_connector *connector, bool force)
+{
+	struct lt9611 *lt9611 = connector_to_lt9611(connector);
+	unsigned int reg_val = 0;
+	int connected = 0;
+
+	regmap_read(lt9611->regmap, 0x825e, &reg_val);
+	connected  = (reg_val & BIT(2));
+
+	lt9611->status = connected ?  connector_status_connected :
+				connector_status_disconnected;
+
+	return lt9611->status;
+}
+
+static int lt9611_read_edid(struct lt9611 *lt9611)
+{
+	unsigned int temp;
+	int ret = 0;
+	int i, j;
+
+	/* memset to clear old buffer, if any */
+	memset(lt9611->edid_buf, 0, sizeof(lt9611->edid_buf));
+
+	regmap_write(lt9611->regmap, 0x8503, 0xc9);
+
+	/* 0xA0 is EDID device address */
+	regmap_write(lt9611->regmap, 0x8504, 0xa0);
+	/* 0x00 is EDID offset address */
+	regmap_write(lt9611->regmap, 0x8505, 0x00);
+
+	/* length for read */
+	regmap_write(lt9611->regmap, 0x8506, EDID_LEN);
+	regmap_write(lt9611->regmap, 0x8514, 0x7f);
+
+	for (i = 0; i < EDID_LOOP; i++) {
+		/* offset address */
+		regmap_write(lt9611->regmap, 0x8505, i * EDID_LEN);
+		regmap_write(lt9611->regmap, 0x8507, 0x36);
+		regmap_write(lt9611->regmap, 0x8507, 0x31);
+		regmap_write(lt9611->regmap, 0x8507, 0x37);
+		usleep_range(5000, 10000);
+
+		regmap_read(lt9611->regmap, 0x8540, &temp);
+
+		if (temp & KEY_DDC_ACCS_DONE) {
+			for (j = 0; j < EDID_LEN; j++) {
+				regmap_read(lt9611->regmap, 0x8583, &temp);
+				lt9611->edid_buf[i * EDID_LEN + j] = temp;
+			}
+
+		} else if (temp & DDC_NO_ACK) { /* DDC No Ack or Abitration lost */
+			dev_err(lt9611->dev, "read edid failed: no ack\n");
+			ret = -EIO;
+			goto end;
+
+		} else {
+			dev_err(lt9611->dev, "read edid failed: access not done\n");
+			ret = -EIO;
+			goto end;
+		}
+	}
+
+end:
+	regmap_write(lt9611->regmap, 0x8507, 0x1f);
+	return ret;
+}
+
+static int
+lt9611_get_edid_block(void *data, u8 *buf, unsigned int block, size_t len)
+{
+	struct lt9611 *lt9611 = data;
+	int ret;
+
+	if (len > 128)
+		return -EINVAL;
+
+	/* supports up to 1 extension block */
+	/* TODO: add support for more extension blocks */
+	if (block > 1)
+		return -EINVAL;
+
+	if (block == 0) {
+		ret = lt9611_read_edid(lt9611);
+		if (ret) {
+			dev_err(lt9611->dev, "edid read failed\n");
+			return ret;
+		}
+	}
+
+	block %= 2;
+	memcpy(buf, lt9611->edid_buf + (block * 128), len);
+
+	return 0;
+}
+
+static int lt9611_connector_get_modes(struct drm_connector *connector)
+{
+	struct lt9611 *lt9611 = connector_to_lt9611(connector);
+	unsigned int count;
+	struct edid *edid;
+
+	lt9611_power_on(lt9611);
+	edid = drm_do_get_edid(connector, lt9611_get_edid_block, lt9611);
+	drm_connector_update_edid_property(connector, edid);
+	count = drm_add_edid_modes(connector, edid);
+	kfree(edid);
+
+	return count;
+}
+
+static enum drm_mode_status
+lt9611_connector_mode_valid(struct drm_connector *connector,
+			    struct drm_display_mode *mode)
+{
+	struct lt9611_mode *lt9611_mode = lt9611_find_mode(mode);
+
+	return lt9611_mode ? MODE_OK : MODE_BAD;
+}
+
+/* bridge funcs */
+static void lt9611_bridge_enable(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	if (lt9611_power_on(lt9611)) {
+		dev_err(lt9611->dev, "power on failed\n");
+		return;
+	}
+
+	lt9611_mipi_input_analog(lt9611);
+	lt9611_hdmi_tx_digital(lt9611);
+	lt9611_hdmi_tx_phy(lt9611);
+
+	msleep(500);
+
+	lt9611_video_check(lt9611);
+
+	/* Enable HDMI output */
+	regmap_write(lt9611->regmap, 0x8130, 0xea);
+}
+
+static void lt9611_bridge_disable(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+	int ret;
+
+	/* Disable HDMI output */
+	ret = regmap_write(lt9611->regmap, 0x8130, 0x6a);
+	if (ret) {
+		dev_err(lt9611->dev, "video on failed\n");
+		return;
+	}
+
+	if (lt9611_power_off(lt9611)) {
+		dev_err(lt9611->dev, "power on failed\n");
+		return;
+	}
+}
+
+static struct
+drm_connector_helper_funcs lt9611_bridge_connector_helper_funcs = {
+	.get_modes = lt9611_connector_get_modes,
+	.mode_valid = lt9611_connector_mode_valid,
+};
+
+static const struct drm_connector_funcs lt9611_bridge_connector_funcs = {
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.detect = lt9611_connector_detect,
+	.destroy = drm_connector_cleanup,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static struct mipi_dsi_device *lt9611_attach_dsi(struct lt9611 *lt9611,
+						 struct device_node *dsi_node)
+{
+	const struct mipi_dsi_device_info info = { "lt9611", 0, NULL };
+	struct mipi_dsi_device *dsi;
+	struct mipi_dsi_host *host;
+	int ret;
+
+	host = of_find_mipi_dsi_host_by_node(dsi_node);
+	if (!host) {
+		dev_err(lt9611->dev, "failed to find dsi host\n");
+		return ERR_PTR(-EPROBE_DEFER);
+	}
+
+	dsi = mipi_dsi_device_register_full(host, &info);
+	if (IS_ERR(dsi)) {
+		dev_err(lt9611->dev, "failed to create dsi device\n");
+		return dsi;
+	}
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
+			  MIPI_DSI_MODE_VIDEO_HSE;
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		dev_err(lt9611->dev, "failed to attach dsi to host\n");
+		mipi_dsi_device_unregister(dsi);
+		return ERR_PTR(ret);
+	}
+
+	return dsi;
+}
+
+static void lt9611_bridge_detach(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	if (lt9611->dsi1) {
+		mipi_dsi_detach(lt9611->dsi1);
+		mipi_dsi_device_unregister(lt9611->dsi1);
+	}
+
+	mipi_dsi_detach(lt9611->dsi0);
+	mipi_dsi_device_unregister(lt9611->dsi0);
+}
+
+static int lt9611_connector_init(struct drm_bridge *bridge, struct lt9611 *lt9611)
+{
+	int ret;
+
+	ret = drm_connector_init(bridge->dev, &lt9611->connector,
+				 &lt9611_bridge_connector_funcs,
+				 DRM_MODE_CONNECTOR_HDMIA);
+	if (ret) {
+		DRM_ERROR("Failed to initialize connector with drm\n");
+		return ret;
+	}
+
+	drm_connector_helper_add(&lt9611->connector,
+				 &lt9611_bridge_connector_helper_funcs);
+	drm_connector_attach_encoder(&lt9611->connector, bridge->encoder);
+
+	if (!bridge->encoder) {
+		DRM_ERROR("Parent encoder object not found");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int lt9611_bridge_attach(struct drm_bridge *bridge,
+				enum drm_bridge_attach_flags flags)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+	int ret;
+
+	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) {
+		ret = lt9611_connector_init(bridge, lt9611);
+		if (ret < 0)
+			return ret;
+	}
+
+	/* Attach primary DSI */
+	lt9611->dsi0 = lt9611_attach_dsi(lt9611, lt9611->dsi0_node);
+	if (IS_ERR(lt9611->dsi0))
+		return PTR_ERR(lt9611->dsi0);
+
+	/* Attach secondary DSI, if specified */
+	if (lt9611->dsi1_node) {
+		lt9611->dsi1 = lt9611_attach_dsi(lt9611, lt9611->dsi1_node);
+		if (IS_ERR(lt9611->dsi1)) {
+			ret = PTR_ERR(lt9611->dsi1);
+			goto err_unregister_dsi0;
+		}
+	}
+
+	return 0;
+
+err_unregister_dsi0:
+	lt9611_bridge_detach(bridge);
+	drm_connector_cleanup(&lt9611->connector);
+	mipi_dsi_device_unregister(lt9611->dsi0);
+
+	return ret;
+}
+
+static enum drm_mode_status lt9611_bridge_mode_valid(struct drm_bridge *bridge,
+						     const struct drm_display_info *info,
+						     const struct drm_display_mode *mode)
+{
+	struct lt9611_mode *lt9611_mode = lt9611_find_mode(mode);
+
+	return lt9611_mode ? MODE_OK : MODE_BAD;
+}
+
+static void lt9611_bridge_pre_enable(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	if (!lt9611->sleep)
+		return;
+
+	lt9611_reset(lt9611);
+	regmap_write(lt9611->regmap, 0x80ee, 0x01);
+
+	lt9611->sleep = false;
+}
+
+static void lt9611_bridge_post_disable(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	lt9611_sleep_setup(lt9611);
+}
+
+static void lt9611_bridge_mode_set(struct drm_bridge *bridge,
+				   const struct drm_display_mode *mode,
+				   const struct drm_display_mode *adj_mode)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+	struct hdmi_avi_infoframe avi_frame;
+	int ret;
+
+	lt9611_bridge_pre_enable(bridge);
+
+	lt9611_mipi_input_digital(lt9611, mode);
+	lt9611_pll_setup(lt9611, mode);
+	lt9611_mipi_video_setup(lt9611, mode);
+	lt9611_pcr_setup(lt9611, mode);
+
+	ret = drm_hdmi_avi_infoframe_from_display_mode(&avi_frame,
+						       &lt9611->connector,
+						       mode);
+	if (!ret)
+		lt9611->vic = avi_frame.video_code;
+}
+
+static enum drm_connector_status lt9611_bridge_detect(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+	unsigned int reg_val = 0;
+	int connected;
+
+	regmap_read(lt9611->regmap, 0x825e, &reg_val);
+	connected  = reg_val & BIT(2);
+
+	lt9611->status = connected ?  connector_status_connected :
+				connector_status_disconnected;
+
+	return lt9611->status;
+}
+
+static struct edid *lt9611_bridge_get_edid(struct drm_bridge *bridge,
+					   struct drm_connector *connector)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	lt9611_power_on(lt9611);
+	return drm_do_get_edid(connector, lt9611_get_edid_block, lt9611);
+}
+
+static void lt9611_bridge_hpd_enable(struct drm_bridge *bridge)
+{
+	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
+
+	lt9611_enable_hpd_interrupts(lt9611);
+}
+
+static const struct drm_bridge_funcs lt9611_bridge_funcs = {
+	.attach = lt9611_bridge_attach,
+	.detach = lt9611_bridge_detach,
+	.mode_valid = lt9611_bridge_mode_valid,
+	.enable = lt9611_bridge_enable,
+	.disable = lt9611_bridge_disable,
+	.post_disable = lt9611_bridge_post_disable,
+	.mode_set = lt9611_bridge_mode_set,
+	.detect = lt9611_bridge_detect,
+	.get_edid = lt9611_bridge_get_edid,
+	.hpd_enable = lt9611_bridge_hpd_enable,
+};
+
+static int lt9611_parse_dt(struct device *dev,
+			   struct lt9611 *lt9611)
+{
+	lt9611->dsi0_node = of_graph_get_remote_node(dev->of_node, 0, -1);
+	if (!lt9611->dsi0_node) {
+		dev_err(lt9611->dev, "failed to get remote node for primary dsi\n");
+		return -ENODEV;
+	}
+
+	lt9611->dsi1_node = of_graph_get_remote_node(dev->of_node, 1, -1);
+
+	lt9611->ac_mode = of_property_read_bool(dev->of_node, "lt,ac-mode");
+
+	return 0;
+}
+
+static int lt9611_gpio_init(struct lt9611 *lt9611)
+{
+	struct device *dev = lt9611->dev;
+
+	lt9611->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(lt9611->reset_gpio)) {
+		dev_err(dev, "failed to acquire reset gpio\n");
+		return PTR_ERR(lt9611->reset_gpio);
+	}
+
+	lt9611->enable_gpio = devm_gpiod_get_optional(dev, "enable",
+						      GPIOD_OUT_LOW);
+	if (IS_ERR(lt9611->enable_gpio)) {
+		dev_err(dev, "failed to acquire enable gpio\n");
+		return PTR_ERR(lt9611->enable_gpio);
+	}
+
+	return 0;
+}
+
+static int lt9611_read_device_rev(struct lt9611 *lt9611)
+{
+	unsigned int rev;
+	int ret;
+
+	regmap_write(lt9611->regmap, 0x80ee, 0x01);
+	ret = regmap_read(lt9611->regmap, 0x8002, &rev);
+	if (ret)
+		dev_err(lt9611->dev, "failed to read revision: %d\n", ret);
+	else
+		dev_info(lt9611->dev, "LT9611 revision: 0x%x\n", rev);
+
+	return ret;
+}
+
+static int lt9611_hdmi_hw_params(struct device *dev, void *data,
+				 struct hdmi_codec_daifmt *fmt,
+				 struct hdmi_codec_params *hparms)
+{
+	struct lt9611 *lt9611 = data;
+
+	if (hparms->sample_rate == 48000)
+		regmap_write(lt9611->regmap, 0x840f, 0x2b);
+	else if (hparms->sample_rate == 96000)
+		regmap_write(lt9611->regmap, 0x840f, 0xab);
+	else
+		return -EINVAL;
+
+	regmap_write(lt9611->regmap, 0x8435, 0x00);
+	regmap_write(lt9611->regmap, 0x8436, 0x18);
+	regmap_write(lt9611->regmap, 0x8437, 0x00);
+
+	return 0;
+}
+
+static int lt9611_audio_startup(struct device *dev, void *data)
+{
+	struct lt9611 *lt9611 = data;
+
+	regmap_write(lt9611->regmap, 0x82d6, 0x8c);
+	regmap_write(lt9611->regmap, 0x82d7, 0x04);
+
+	regmap_write(lt9611->regmap, 0x8406, 0x08);
+	regmap_write(lt9611->regmap, 0x8407, 0x10);
+
+	regmap_write(lt9611->regmap, 0x8434, 0xd5);
+
+	return 0;
+}
+
+static void lt9611_audio_shutdown(struct device *dev, void *data)
+{
+	struct lt9611 *lt9611 = data;
+
+	regmap_write(lt9611->regmap, 0x8406, 0x00);
+	regmap_write(lt9611->regmap, 0x8407, 0x00);
+}
+
+static int lt9611_hdmi_i2s_get_dai_id(struct snd_soc_component *component,
+				      struct device_node *endpoint)
+{
+	struct of_endpoint of_ep;
+	int ret;
+
+	ret = of_graph_parse_endpoint(endpoint, &of_ep);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * HDMI sound should be located as reg = <2>
+	 * Then, it is sound port 0
+	 */
+	if (of_ep.port == 2)
+		return 0;
+
+	return -EINVAL;
+}
+
+static const struct hdmi_codec_ops lt9611_codec_ops = {
+	.hw_params	= lt9611_hdmi_hw_params,
+	.audio_shutdown = lt9611_audio_shutdown,
+	.audio_startup	= lt9611_audio_startup,
+	.get_dai_id	= lt9611_hdmi_i2s_get_dai_id,
+};
+
+static struct hdmi_codec_pdata codec_data = {
+	.ops = &lt9611_codec_ops,
+	.max_i2s_channels = 8,
+	.i2s = 1,
+};
+
+static int lt9611_audio_init(struct device *dev, struct lt9611 *lt9611)
+{
+	codec_data.data = lt9611;
+	lt9611->audio_pdev =
+		platform_device_register_data(dev, HDMI_CODEC_DRV_NAME,
+					      PLATFORM_DEVID_AUTO,
+					      &codec_data, sizeof(codec_data));
+
+	return PTR_ERR_OR_ZERO(lt9611->audio_pdev);
+}
+
+static void lt9611_audio_exit(struct lt9611 *lt9611)
+{
+	if (lt9611->audio_pdev) {
+		platform_device_unregister(lt9611->audio_pdev);
+		lt9611->audio_pdev = NULL;
+	}
+}
+
+static int lt9611_probe(struct i2c_client *client,
+			const struct i2c_device_id *id)
+{
+	struct lt9611 *lt9611;
+	struct device *dev = &client->dev;
+	int ret;
+
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) {
+		dev_err(dev, "device doesn't support I2C\n");
+		return -ENODEV;
+	}
+
+	lt9611 = devm_kzalloc(dev, sizeof(*lt9611), GFP_KERNEL);
+	if (!lt9611)
+		return -ENOMEM;
+
+	lt9611->dev = &client->dev;
+	lt9611->client = client;
+	lt9611->sleep = false;
+
+	lt9611->regmap = devm_regmap_init_i2c(client, &lt9611_regmap_config);
+	if (IS_ERR(lt9611->regmap)) {
+		dev_err(lt9611->dev, "regmap i2c init failed\n");
+		return PTR_ERR(lt9611->regmap);
+	}
+
+	ret = lt9611_parse_dt(&client->dev, lt9611);
+	if (ret) {
+		dev_err(dev, "failed to parse device tree\n");
+		return ret;
+	}
+
+	ret = lt9611_gpio_init(lt9611);
+	if (ret < 0)
+		goto err_of_put;
+
+	ret = lt9611_regulator_init(lt9611);
+	if (ret < 0)
+		goto err_of_put;
+
+	lt9611_assert_5v(lt9611);
+
+	ret = lt9611_regulator_enable(lt9611);
+	if (ret)
+		goto err_of_put;
+
+	lt9611_reset(lt9611);
+
+	ret = lt9611_read_device_rev(lt9611);
+	if (ret) {
+		dev_err(dev, "failed to read chip rev\n");
+		goto err_disable_regulators;
+	}
+
+	ret = devm_request_threaded_irq(dev, client->irq, NULL,
+					lt9611_irq_thread_handler,
+					IRQF_ONESHOT, "lt9611", lt9611);
+	if (ret) {
+		dev_err(dev, "failed to request irq\n");
+		goto err_disable_regulators;
+	}
+
+	i2c_set_clientdata(client, lt9611);
+
+	lt9611->bridge.funcs = &lt9611_bridge_funcs;
+	lt9611->bridge.of_node = client->dev.of_node;
+	lt9611->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID |
+			     DRM_BRIDGE_OP_HPD | DRM_BRIDGE_OP_MODES;
+	lt9611->bridge.type = DRM_MODE_CONNECTOR_HDMIA;
+
+	drm_bridge_add(&lt9611->bridge);
+
+	lt9611_enable_hpd_interrupts(lt9611);
+
+	return lt9611_audio_init(dev, lt9611);
+
+err_disable_regulators:
+	regulator_bulk_disable(ARRAY_SIZE(lt9611->supplies), lt9611->supplies);
+
+err_of_put:
+	of_node_put(lt9611->dsi1_node);
+	of_node_put(lt9611->dsi0_node);
+
+	return ret;
+}
+
+static int lt9611_remove(struct i2c_client *client)
+{
+	struct lt9611 *lt9611 = i2c_get_clientdata(client);
+
+	disable_irq(client->irq);
+	lt9611_audio_exit(lt9611);
+	drm_bridge_remove(&lt9611->bridge);
+
+	regulator_bulk_disable(ARRAY_SIZE(lt9611->supplies), lt9611->supplies);
+
+	of_node_put(lt9611->dsi1_node);
+	of_node_put(lt9611->dsi0_node);
+
+	return 0;
+}
+
+static struct i2c_device_id lt9611_id[] = {
+	{ "lontium,lt9611", 0 },
+	{}
+};
+
+static const struct of_device_id lt9611_match_table[] = {
+	{ .compatible = "lontium,lt9611" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, lt9611_match_table);
+
+static struct i2c_driver lt9611_driver = {
+	.driver = {
+		.name = "lt9611",
+		.of_match_table = lt9611_match_table,
+	},
+	.probe = lt9611_probe,
+	.remove = lt9611_remove,
+	.id_table = lt9611_id,
+};
+module_i2c_driver(lt9611_driver);
+
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/bridge/lvds-codec.c b/drivers/gpu/drm/bridge/lvds-codec.c
index f19d9f7a5db2..f52ccffc1bd1 100644
--- a/drivers/gpu/drm/bridge/lvds-codec.c
+++ b/drivers/gpu/drm/bridge/lvds-codec.c
@@ -10,13 +10,16 @@
 #include <linux/of_device.h>
 #include <linux/of_graph.h>
 #include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
 
 #include <drm/drm_bridge.h>
 #include <drm/drm_panel.h>
 
 struct lvds_codec {
+	struct device *dev;
 	struct drm_bridge bridge;
 	struct drm_bridge *panel_bridge;
+	struct regulator *vcc;
 	struct gpio_desc *powerdown_gpio;
 	u32 connector_type;
 };
@@ -38,6 +41,14 @@ static int lvds_codec_attach(struct drm_bridge *bridge,
 static void lvds_codec_enable(struct drm_bridge *bridge)
 {
 	struct lvds_codec *lvds_codec = to_lvds_codec(bridge);
+	int ret;
+
+	ret = regulator_enable(lvds_codec->vcc);
+	if (ret) {
+		dev_err(lvds_codec->dev,
+			"Failed to enable regulator \"vcc\": %d\n", ret);
+		return;
+	}
 
 	if (lvds_codec->powerdown_gpio)
 		gpiod_set_value_cansleep(lvds_codec->powerdown_gpio, 0);
@@ -46,9 +57,15 @@ static void lvds_codec_enable(struct drm_bridge *bridge)
 static void lvds_codec_disable(struct drm_bridge *bridge)
 {
 	struct lvds_codec *lvds_codec = to_lvds_codec(bridge);
+	int ret;
 
 	if (lvds_codec->powerdown_gpio)
 		gpiod_set_value_cansleep(lvds_codec->powerdown_gpio, 1);
+
+	ret = regulator_disable(lvds_codec->vcc);
+	if (ret)
+		dev_err(lvds_codec->dev,
+			"Failed to disable regulator \"vcc\": %d\n", ret);
 }
 
 static const struct drm_bridge_funcs funcs = {
@@ -63,12 +80,24 @@ static int lvds_codec_probe(struct platform_device *pdev)
 	struct device_node *panel_node;
 	struct drm_panel *panel;
 	struct lvds_codec *lvds_codec;
+	int ret;
 
 	lvds_codec = devm_kzalloc(dev, sizeof(*lvds_codec), GFP_KERNEL);
 	if (!lvds_codec)
 		return -ENOMEM;
 
+	lvds_codec->dev = &pdev->dev;
 	lvds_codec->connector_type = (uintptr_t)of_device_get_match_data(dev);
+
+	lvds_codec->vcc = devm_regulator_get(lvds_codec->dev, "power");
+	if (IS_ERR(lvds_codec->vcc)) {
+		ret = PTR_ERR(lvds_codec->vcc);
+		if (ret != -EPROBE_DEFER)
+			dev_err(lvds_codec->dev,
+				"Unable to get \"vcc\" supply: %d\n", ret);
+		return ret;
+	}
+
 	lvds_codec->powerdown_gpio = devm_gpiod_get_optional(dev, "powerdown",
 							     GPIOD_OUT_HIGH);
 	if (IS_ERR(lvds_codec->powerdown_gpio))
diff --git a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
index 6200f12a37e6..d2808c4a6fb1 100644
--- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
+++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
@@ -61,7 +61,6 @@ struct ge_b850v3_lvds {
 	struct drm_bridge bridge;
 	struct i2c_client *stdp4028_i2c;
 	struct i2c_client *stdp2690_i2c;
-	struct edid *edid;
 };
 
 static struct ge_b850v3_lvds *ge_b850v3_lvds_ptr;
@@ -131,22 +130,26 @@ err:
 	return NULL;
 }
 
-static int ge_b850v3_lvds_get_modes(struct drm_connector *connector)
+static struct edid *ge_b850v3_lvds_get_edid(struct drm_bridge *bridge,
+					    struct drm_connector *connector)
 {
 	struct i2c_client *client;
-	int num_modes = 0;
 
 	client = ge_b850v3_lvds_ptr->stdp2690_i2c;
 
-	kfree(ge_b850v3_lvds_ptr->edid);
-	ge_b850v3_lvds_ptr->edid = (struct edid *)stdp2690_get_edid(client);
+	return (struct edid *)stdp2690_get_edid(client);
+}
 
-	if (ge_b850v3_lvds_ptr->edid) {
-		drm_connector_update_edid_property(connector,
-						      ge_b850v3_lvds_ptr->edid);
-		num_modes = drm_add_edid_modes(connector,
-					       ge_b850v3_lvds_ptr->edid);
-	}
+static int ge_b850v3_lvds_get_modes(struct drm_connector *connector)
+{
+	struct edid *edid;
+	int num_modes;
+
+	edid = ge_b850v3_lvds_get_edid(&ge_b850v3_lvds_ptr->bridge, connector);
+
+	drm_connector_update_edid_property(connector, edid);
+	num_modes = drm_add_edid_modes(connector, edid);
+	kfree(edid);
 
 	return num_modes;
 }
@@ -163,8 +166,7 @@ drm_connector_helper_funcs ge_b850v3_lvds_connector_helper_funcs = {
 	.mode_valid = ge_b850v3_lvds_mode_valid,
 };
 
-static enum drm_connector_status ge_b850v3_lvds_detect(
-		struct drm_connector *connector, bool force)
+static enum drm_connector_status ge_b850v3_lvds_bridge_detect(struct drm_bridge *bridge)
 {
 	struct i2c_client *stdp4028_i2c =
 			ge_b850v3_lvds_ptr->stdp4028_i2c;
@@ -182,6 +184,12 @@ static enum drm_connector_status ge_b850v3_lvds_detect(
 	return connector_status_unknown;
 }
 
+static enum drm_connector_status ge_b850v3_lvds_detect(struct drm_connector *connector,
+						       bool force)
+{
+	return ge_b850v3_lvds_bridge_detect(&ge_b850v3_lvds_ptr->bridge);
+}
+
 static const struct drm_connector_funcs ge_b850v3_lvds_connector_funcs = {
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.detect = ge_b850v3_lvds_detect,
@@ -191,34 +199,11 @@ static const struct drm_connector_funcs ge_b850v3_lvds_connector_funcs = {
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
-static irqreturn_t ge_b850v3_lvds_irq_handler(int irq, void *dev_id)
-{
-	struct i2c_client *stdp4028_i2c
-			= ge_b850v3_lvds_ptr->stdp4028_i2c;
-
-	i2c_smbus_write_word_data(stdp4028_i2c,
-				  STDP4028_DPTX_IRQ_STS_REG,
-				  STDP4028_DPTX_IRQ_CLEAR);
-
-	if (ge_b850v3_lvds_ptr->connector.dev)
-		drm_kms_helper_hotplug_event(ge_b850v3_lvds_ptr->connector.dev);
-
-	return IRQ_HANDLED;
-}
-
-static int ge_b850v3_lvds_attach(struct drm_bridge *bridge,
-				 enum drm_bridge_attach_flags flags)
+static int ge_b850v3_lvds_create_connector(struct drm_bridge *bridge)
 {
 	struct drm_connector *connector = &ge_b850v3_lvds_ptr->connector;
-	struct i2c_client *stdp4028_i2c
-			= ge_b850v3_lvds_ptr->stdp4028_i2c;
 	int ret;
 
-	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
-		DRM_ERROR("Fix bridge driver to make connector optional!");
-		return -EINVAL;
-	}
-
 	if (!bridge->encoder) {
 		DRM_ERROR("Parent encoder object not found");
 		return -ENODEV;
@@ -237,9 +222,29 @@ static int ge_b850v3_lvds_attach(struct drm_bridge *bridge,
 		return ret;
 	}
 
-	ret = drm_connector_attach_encoder(connector, bridge->encoder);
-	if (ret)
-		return ret;
+	return drm_connector_attach_encoder(connector, bridge->encoder);
+}
+
+static irqreturn_t ge_b850v3_lvds_irq_handler(int irq, void *dev_id)
+{
+	struct i2c_client *stdp4028_i2c
+			= ge_b850v3_lvds_ptr->stdp4028_i2c;
+
+	i2c_smbus_write_word_data(stdp4028_i2c,
+				  STDP4028_DPTX_IRQ_STS_REG,
+				  STDP4028_DPTX_IRQ_CLEAR);
+
+	if (ge_b850v3_lvds_ptr->bridge.dev)
+		drm_kms_helper_hotplug_event(ge_b850v3_lvds_ptr->bridge.dev);
+
+	return IRQ_HANDLED;
+}
+
+static int ge_b850v3_lvds_attach(struct drm_bridge *bridge,
+				 enum drm_bridge_attach_flags flags)
+{
+	struct i2c_client *stdp4028_i2c
+			= ge_b850v3_lvds_ptr->stdp4028_i2c;
 
 	/* Configures the bridge to re-enable interrupts after each ack. */
 	i2c_smbus_write_word_data(stdp4028_i2c,
@@ -251,11 +256,16 @@ static int ge_b850v3_lvds_attach(struct drm_bridge *bridge,
 				  STDP4028_DPTX_IRQ_EN_REG,
 				  STDP4028_DPTX_IRQ_CONFIG);
 
-	return 0;
+	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
+		return 0;
+
+	return ge_b850v3_lvds_create_connector(bridge);
 }
 
 static const struct drm_bridge_funcs ge_b850v3_lvds_funcs = {
 	.attach = ge_b850v3_lvds_attach,
+	.detect = ge_b850v3_lvds_bridge_detect,
+	.get_edid = ge_b850v3_lvds_get_edid,
 };
 
 static int ge_b850v3_lvds_init(struct device *dev)
@@ -291,8 +301,6 @@ static void ge_b850v3_lvds_remove(void)
 
 	drm_bridge_remove(&ge_b850v3_lvds_ptr->bridge);
 
-	kfree(ge_b850v3_lvds_ptr->edid);
-
 	ge_b850v3_lvds_ptr = NULL;
 out:
 	mutex_unlock(&ge_b850v3_lvds_dev_mutex);
@@ -302,14 +310,21 @@ static int stdp4028_ge_b850v3_fw_probe(struct i2c_client *stdp4028_i2c,
 				       const struct i2c_device_id *id)
 {
 	struct device *dev = &stdp4028_i2c->dev;
+	int ret;
+
+	ret = ge_b850v3_lvds_init(dev);
 
-	ge_b850v3_lvds_init(dev);
+	if (ret)
+		return ret;
 
 	ge_b850v3_lvds_ptr->stdp4028_i2c = stdp4028_i2c;
 	i2c_set_clientdata(stdp4028_i2c, ge_b850v3_lvds_ptr);
 
 	/* drm bridge initialization */
 	ge_b850v3_lvds_ptr->bridge.funcs = &ge_b850v3_lvds_funcs;
+	ge_b850v3_lvds_ptr->bridge.ops = DRM_BRIDGE_OP_DETECT |
+					 DRM_BRIDGE_OP_EDID;
+	ge_b850v3_lvds_ptr->bridge.type = DRM_MODE_CONNECTOR_DisplayPort;
 	ge_b850v3_lvds_ptr->bridge.of_node = dev->of_node;
 	drm_bridge_add(&ge_b850v3_lvds_ptr->bridge);
 
@@ -361,8 +376,12 @@ static int stdp2690_ge_b850v3_fw_probe(struct i2c_client *stdp2690_i2c,
 				       const struct i2c_device_id *id)
 {
 	struct device *dev = &stdp2690_i2c->dev;
+	int ret;
+
+	ret = ge_b850v3_lvds_init(dev);
 
-	ge_b850v3_lvds_init(dev);
+	if (ret)
+		return ret;
 
 	ge_b850v3_lvds_ptr->stdp2690_i2c = stdp2690_i2c;
 	i2c_set_clientdata(stdp2690_i2c, ge_b850v3_lvds_ptr);
diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c
index 438e566ce0a4..e941c1132598 100644
--- a/drivers/gpu/drm/bridge/nxp-ptn3460.c
+++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c
@@ -29,8 +29,7 @@ struct ptn3460_bridge {
 	struct drm_connector connector;
 	struct i2c_client *client;
 	struct drm_bridge bridge;
-	struct edid *edid;
-	struct drm_panel *panel;
+	struct drm_bridge *panel_bridge;
 	struct gpio_desc *gpio_pd_n;
 	struct gpio_desc *gpio_rst_n;
 	u32 edid_emulation;
@@ -127,11 +126,6 @@ static void ptn3460_pre_enable(struct drm_bridge *bridge)
 	usleep_range(10, 20);
 	gpiod_set_value(ptn_bridge->gpio_rst_n, 1);
 
-	if (drm_panel_prepare(ptn_bridge->panel)) {
-		DRM_ERROR("failed to prepare panel\n");
-		return;
-	}
-
 	/*
 	 * There's a bug in the PTN chip where it falsely asserts hotplug before
 	 * it is fully functional. We're forced to wait for the maximum start up
@@ -146,16 +140,6 @@ static void ptn3460_pre_enable(struct drm_bridge *bridge)
 	ptn_bridge->enabled = true;
 }
 
-static void ptn3460_enable(struct drm_bridge *bridge)
-{
-	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
-
-	if (drm_panel_enable(ptn_bridge->panel)) {
-		DRM_ERROR("failed to enable panel\n");
-		return;
-	}
-}
-
 static void ptn3460_disable(struct drm_bridge *bridge)
 {
 	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
@@ -165,36 +149,18 @@ static void ptn3460_disable(struct drm_bridge *bridge)
 
 	ptn_bridge->enabled = false;
 
-	if (drm_panel_disable(ptn_bridge->panel)) {
-		DRM_ERROR("failed to disable panel\n");
-		return;
-	}
-
 	gpiod_set_value(ptn_bridge->gpio_rst_n, 1);
 	gpiod_set_value(ptn_bridge->gpio_pd_n, 0);
 }
 
-static void ptn3460_post_disable(struct drm_bridge *bridge)
-{
-	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
 
-	if (drm_panel_unprepare(ptn_bridge->panel)) {
-		DRM_ERROR("failed to unprepare panel\n");
-		return;
-	}
-}
-
-static int ptn3460_get_modes(struct drm_connector *connector)
+static struct edid *ptn3460_get_edid(struct drm_bridge *bridge,
+				     struct drm_connector *connector)
 {
-	struct ptn3460_bridge *ptn_bridge;
-	u8 *edid;
-	int ret, num_modes = 0;
+	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
 	bool power_off;
-
-	ptn_bridge = connector_to_ptn3460(connector);
-
-	if (ptn_bridge->edid)
-		return drm_add_edid_modes(connector, ptn_bridge->edid);
+	u8 *edid;
+	int ret;
 
 	power_off = !ptn_bridge->enabled;
 	ptn3460_pre_enable(&ptn_bridge->bridge);
@@ -202,30 +168,40 @@ static int ptn3460_get_modes(struct drm_connector *connector)
 	edid = kmalloc(EDID_LENGTH, GFP_KERNEL);
 	if (!edid) {
 		DRM_ERROR("Failed to allocate EDID\n");
-		return 0;
+		goto out;
 	}
 
 	ret = ptn3460_read_bytes(ptn_bridge, PTN3460_EDID_ADDR, edid,
-			EDID_LENGTH);
+				 EDID_LENGTH);
 	if (ret) {
 		kfree(edid);
+		edid = NULL;
 		goto out;
 	}
 
-	ptn_bridge->edid = (struct edid *)edid;
-	drm_connector_update_edid_property(connector, ptn_bridge->edid);
-
-	num_modes = drm_add_edid_modes(connector, ptn_bridge->edid);
-
 out:
 	if (power_off)
 		ptn3460_disable(&ptn_bridge->bridge);
 
+	return (struct edid *)edid;
+}
+
+static int ptn3460_connector_get_modes(struct drm_connector *connector)
+{
+	struct ptn3460_bridge *ptn_bridge = connector_to_ptn3460(connector);
+	struct edid *edid;
+	int num_modes;
+
+	edid = ptn3460_get_edid(&ptn_bridge->bridge, connector);
+	drm_connector_update_edid_property(connector, edid);
+	num_modes = drm_add_edid_modes(connector, edid);
+	kfree(edid);
+
 	return num_modes;
 }
 
 static const struct drm_connector_helper_funcs ptn3460_connector_helper_funcs = {
-	.get_modes = ptn3460_get_modes,
+	.get_modes = ptn3460_connector_get_modes,
 };
 
 static const struct drm_connector_funcs ptn3460_connector_funcs = {
@@ -242,10 +218,14 @@ static int ptn3460_bridge_attach(struct drm_bridge *bridge,
 	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
 	int ret;
 
-	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
-		DRM_ERROR("Fix bridge driver to make connector optional!");
-		return -EINVAL;
-	}
+	/* Let this driver create connector if requested */
+	ret = drm_bridge_attach(bridge->encoder, ptn_bridge->panel_bridge,
+				bridge, flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	if (ret < 0)
+		return ret;
+
+	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
+		return 0;
 
 	if (!bridge->encoder) {
 		DRM_ERROR("Parent encoder object not found");
@@ -265,9 +245,6 @@ static int ptn3460_bridge_attach(struct drm_bridge *bridge,
 	drm_connector_attach_encoder(&ptn_bridge->connector,
 							bridge->encoder);
 
-	if (ptn_bridge->panel)
-		drm_panel_attach(ptn_bridge->panel, &ptn_bridge->connector);
-
 	drm_helper_hpd_irq_event(ptn_bridge->connector.dev);
 
 	return ret;
@@ -275,10 +252,9 @@ static int ptn3460_bridge_attach(struct drm_bridge *bridge,
 
 static const struct drm_bridge_funcs ptn3460_bridge_funcs = {
 	.pre_enable = ptn3460_pre_enable,
-	.enable = ptn3460_enable,
 	.disable = ptn3460_disable,
-	.post_disable = ptn3460_post_disable,
 	.attach = ptn3460_bridge_attach,
+	.get_edid = ptn3460_get_edid,
 };
 
 static int ptn3460_probe(struct i2c_client *client,
@@ -286,6 +262,8 @@ static int ptn3460_probe(struct i2c_client *client,
 {
 	struct device *dev = &client->dev;
 	struct ptn3460_bridge *ptn_bridge;
+	struct drm_bridge *panel_bridge;
+	struct drm_panel *panel;
 	int ret;
 
 	ptn_bridge = devm_kzalloc(dev, sizeof(*ptn_bridge), GFP_KERNEL);
@@ -293,10 +271,15 @@ static int ptn3460_probe(struct i2c_client *client,
 		return -ENOMEM;
 	}
 
-	ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0, &ptn_bridge->panel, NULL);
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0, &panel, NULL);
 	if (ret)
 		return ret;
 
+	panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+	if (IS_ERR(panel_bridge))
+		return PTR_ERR(panel_bridge);
+
+	ptn_bridge->panel_bridge = panel_bridge;
 	ptn_bridge->client = client;
 
 	ptn_bridge->gpio_pd_n = devm_gpiod_get(&client->dev, "powerdown",
@@ -327,6 +310,8 @@ static int ptn3460_probe(struct i2c_client *client,
 	}
 
 	ptn_bridge->bridge.funcs = &ptn3460_bridge_funcs;
+	ptn_bridge->bridge.ops = DRM_BRIDGE_OP_EDID;
+	ptn_bridge->bridge.type = DRM_MODE_CONNECTOR_LVDS;
 	ptn_bridge->bridge.of_node = dev->of_node;
 	drm_bridge_add(&ptn_bridge->bridge);
 
diff --git a/drivers/gpu/drm/bridge/panel.c b/drivers/gpu/drm/bridge/panel.c
index 1e63ed6b18aa..0ddc37551194 100644
--- a/drivers/gpu/drm/bridge/panel.c
+++ b/drivers/gpu/drm/bridge/panel.c
@@ -82,18 +82,11 @@ static int panel_bridge_attach(struct drm_bridge *bridge,
 	drm_connector_attach_encoder(&panel_bridge->connector,
 					  bridge->encoder);
 
-	ret = drm_panel_attach(panel_bridge->panel, &panel_bridge->connector);
-	if (ret < 0)
-		return ret;
-
 	return 0;
 }
 
 static void panel_bridge_detach(struct drm_bridge *bridge)
 {
-	struct panel_bridge *panel_bridge = drm_bridge_to_panel_bridge(bridge);
-
-	drm_panel_detach(panel_bridge->panel);
 }
 
 static void panel_bridge_pre_enable(struct drm_bridge *bridge)
diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c
index d789ea2a7fb9..614b19f0f1b7 100644
--- a/drivers/gpu/drm/bridge/parade-ps8622.c
+++ b/drivers/gpu/drm/bridge/parade-ps8622.c
@@ -42,10 +42,9 @@
 #endif
 
 struct ps8622_bridge {
-	struct drm_connector connector;
 	struct i2c_client *client;
 	struct drm_bridge bridge;
-	struct drm_panel *panel;
+	struct drm_bridge *panel_bridge;
 	struct regulator *v12;
 	struct backlight_device *bl;
 
@@ -64,12 +63,6 @@ static inline struct ps8622_bridge *
 	return container_of(bridge, struct ps8622_bridge, bridge);
 }
 
-static inline struct ps8622_bridge *
-		connector_to_ps8622(struct drm_connector *connector)
-{
-	return container_of(connector, struct ps8622_bridge, connector);
-}
-
 static int ps8622_set(struct i2c_client *client, u8 page, u8 reg, u8 val)
 {
 	int ret;
@@ -365,11 +358,6 @@ static void ps8622_pre_enable(struct drm_bridge *bridge)
 			DRM_ERROR("fails to enable ps8622->v12");
 	}
 
-	if (drm_panel_prepare(ps8622->panel)) {
-		DRM_ERROR("failed to prepare panel\n");
-		return;
-	}
-
 	gpiod_set_value(ps8622->gpio_slp, 1);
 
 	/*
@@ -399,24 +387,9 @@ static void ps8622_pre_enable(struct drm_bridge *bridge)
 	ps8622->enabled = true;
 }
 
-static void ps8622_enable(struct drm_bridge *bridge)
-{
-	struct ps8622_bridge *ps8622 = bridge_to_ps8622(bridge);
-
-	if (drm_panel_enable(ps8622->panel)) {
-		DRM_ERROR("failed to enable panel\n");
-		return;
-	}
-}
-
 static void ps8622_disable(struct drm_bridge *bridge)
 {
-	struct ps8622_bridge *ps8622 = bridge_to_ps8622(bridge);
-
-	if (drm_panel_disable(ps8622->panel)) {
-		DRM_ERROR("failed to disable panel\n");
-		return;
-	}
+	/* Delay after panel is disabled */
 	msleep(PS8622_PWMO_END_T12_MS);
 }
 
@@ -436,11 +409,6 @@ static void ps8622_post_disable(struct drm_bridge *bridge)
 	 */
 	gpiod_set_value(ps8622->gpio_slp, 0);
 
-	if (drm_panel_unprepare(ps8622->panel)) {
-		DRM_ERROR("failed to unprepare panel\n");
-		return;
-	}
-
 	if (ps8622->v12)
 		regulator_disable(ps8622->v12);
 
@@ -455,67 +423,17 @@ static void ps8622_post_disable(struct drm_bridge *bridge)
 	msleep(PS8622_POWER_OFF_T17_MS);
 }
 
-static int ps8622_get_modes(struct drm_connector *connector)
-{
-	struct ps8622_bridge *ps8622;
-
-	ps8622 = connector_to_ps8622(connector);
-
-	return drm_panel_get_modes(ps8622->panel, connector);
-}
-
-static const struct drm_connector_helper_funcs ps8622_connector_helper_funcs = {
-	.get_modes = ps8622_get_modes,
-};
-
-static const struct drm_connector_funcs ps8622_connector_funcs = {
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = drm_connector_cleanup,
-	.reset = drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
 static int ps8622_attach(struct drm_bridge *bridge,
 			 enum drm_bridge_attach_flags flags)
 {
 	struct ps8622_bridge *ps8622 = bridge_to_ps8622(bridge);
-	int ret;
-
-	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
-		DRM_ERROR("Fix bridge driver to make connector optional!");
-		return -EINVAL;
-	}
 
-	if (!bridge->encoder) {
-		DRM_ERROR("Parent encoder object not found");
-		return -ENODEV;
-	}
-
-	ps8622->connector.polled = DRM_CONNECTOR_POLL_HPD;
-	ret = drm_connector_init(bridge->dev, &ps8622->connector,
-			&ps8622_connector_funcs, DRM_MODE_CONNECTOR_LVDS);
-	if (ret) {
-		DRM_ERROR("Failed to initialize connector with drm\n");
-		return ret;
-	}
-	drm_connector_helper_add(&ps8622->connector,
-					&ps8622_connector_helper_funcs);
-	drm_connector_register(&ps8622->connector);
-	drm_connector_attach_encoder(&ps8622->connector,
-							bridge->encoder);
-
-	if (ps8622->panel)
-		drm_panel_attach(ps8622->panel, &ps8622->connector);
-
-	drm_helper_hpd_irq_event(ps8622->connector.dev);
-
-	return ret;
+	return drm_bridge_attach(ps8622->bridge.encoder, ps8622->panel_bridge,
+				 &ps8622->bridge, flags);
 }
 
 static const struct drm_bridge_funcs ps8622_bridge_funcs = {
 	.pre_enable = ps8622_pre_enable,
-	.enable = ps8622_enable,
 	.disable = ps8622_disable,
 	.post_disable = ps8622_post_disable,
 	.attach = ps8622_attach,
@@ -533,16 +451,23 @@ static int ps8622_probe(struct i2c_client *client,
 {
 	struct device *dev = &client->dev;
 	struct ps8622_bridge *ps8622;
+	struct drm_bridge *panel_bridge;
+	struct drm_panel *panel;
 	int ret;
 
 	ps8622 = devm_kzalloc(dev, sizeof(*ps8622), GFP_KERNEL);
 	if (!ps8622)
 		return -ENOMEM;
 
-	ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0, &ps8622->panel, NULL);
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0, &panel, NULL);
 	if (ret)
 		return ret;
 
+	panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+	if (IS_ERR(panel_bridge))
+		return PTR_ERR(panel_bridge);
+
+	ps8622->panel_bridge = panel_bridge;
 	ps8622->client = client;
 
 	ps8622->v12 = devm_regulator_get(dev, "vdd12");
@@ -595,6 +520,7 @@ static int ps8622_probe(struct i2c_client *client,
 	}
 
 	ps8622->bridge.funcs = &ps8622_bridge_funcs;
+	ps8622->bridge.type = DRM_MODE_CONNECTOR_LVDS;
 	ps8622->bridge.of_node = dev->of_node;
 	drm_bridge_add(&ps8622->bridge);
 
diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c
index 4b099196afeb..7bd0affa057a 100644
--- a/drivers/gpu/drm/bridge/parade-ps8640.c
+++ b/drivers/gpu/drm/bridge/parade-ps8640.c
@@ -65,6 +65,7 @@ struct ps8640 {
 	struct regulator_bulk_data supplies[2];
 	struct gpio_desc *gpio_reset;
 	struct gpio_desc *gpio_powerdown;
+	bool powered;
 };
 
 static inline struct ps8640 *bridge_to_ps8640(struct drm_bridge *e)
@@ -82,19 +83,24 @@ static int ps8640_bridge_vdo_control(struct ps8640 *ps_bridge,
 	ret = i2c_smbus_write_i2c_block_data(client, PAGE3_SET_ADD,
 					     sizeof(vdo_ctrl_buf),
 					     vdo_ctrl_buf);
-	if (ret < 0)
+	if (ret < 0) {
+		DRM_ERROR("failed to %sable VDO: %d\n",
+			  ctrl == ENABLE ? "en" : "dis", ret);
 		return ret;
+	}
 
 	return 0;
 }
 
-static void ps8640_pre_enable(struct drm_bridge *bridge)
+static void ps8640_bridge_poweron(struct ps8640 *ps_bridge)
 {
-	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
 	struct i2c_client *client = ps_bridge->page[PAGE2_TOP_CNTL];
 	unsigned long timeout;
 	int ret, status;
 
+	if (ps_bridge->powered)
+		return;
+
 	ret = regulator_bulk_enable(ARRAY_SIZE(ps_bridge->supplies),
 				    ps_bridge->supplies);
 	if (ret < 0) {
@@ -149,12 +155,6 @@ static void ps8640_pre_enable(struct drm_bridge *bridge)
 		goto err_regulators_disable;
 	}
 
-	ret = ps8640_bridge_vdo_control(ps_bridge, ENABLE);
-	if (ret) {
-		DRM_ERROR("failed to enable VDO: %d\n", ret);
-		goto err_regulators_disable;
-	}
-
 	/* Switch access edp panel's edid through i2c */
 	ret = i2c_smbus_write_byte_data(client, PAGE2_I2C_BYPASS,
 					I2C_BYPASS_EN);
@@ -163,6 +163,8 @@ static void ps8640_pre_enable(struct drm_bridge *bridge)
 		goto err_regulators_disable;
 	}
 
+	ps_bridge->powered = true;
+
 	return;
 
 err_regulators_disable:
@@ -170,14 +172,12 @@ err_regulators_disable:
 			       ps_bridge->supplies);
 }
 
-static void ps8640_post_disable(struct drm_bridge *bridge)
+static void ps8640_bridge_poweroff(struct ps8640 *ps_bridge)
 {
-	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
 	int ret;
 
-	ret = ps8640_bridge_vdo_control(ps_bridge, DISABLE);
-	if (ret < 0)
-		DRM_ERROR("failed to disable VDO: %d\n", ret);
+	if (!ps_bridge->powered)
+		return;
 
 	gpiod_set_value(ps_bridge->gpio_reset, 1);
 	gpiod_set_value(ps_bridge->gpio_powerdown, 1);
@@ -185,6 +185,28 @@ static void ps8640_post_disable(struct drm_bridge *bridge)
 				     ps_bridge->supplies);
 	if (ret < 0)
 		DRM_ERROR("cannot disable regulators %d\n", ret);
+
+	ps_bridge->powered = false;
+}
+
+static void ps8640_pre_enable(struct drm_bridge *bridge)
+{
+	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
+	int ret;
+
+	ps8640_bridge_poweron(ps_bridge);
+
+	ret = ps8640_bridge_vdo_control(ps_bridge, ENABLE);
+	if (ret < 0)
+		ps8640_bridge_poweroff(ps_bridge);
+}
+
+static void ps8640_post_disable(struct drm_bridge *bridge)
+{
+	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
+
+	ps8640_bridge_vdo_control(ps_bridge, DISABLE);
+	ps8640_bridge_poweroff(ps_bridge);
 }
 
 static int ps8640_bridge_attach(struct drm_bridge *bridge,
@@ -200,6 +222,10 @@ static int ps8640_bridge_attach(struct drm_bridge *bridge,
 						   .channel = 0,
 						   .node = NULL,
 						 };
+
+	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
+		return -EINVAL;
+
 	/* port@0 is ps8640 dsi input port */
 	in_ep = of_graph_get_endpoint_by_regs(dev->of_node, 0, -1);
 	if (!in_ep)
@@ -242,8 +268,43 @@ err_dsi_attach:
 	return ret;
 }
 
+static struct edid *ps8640_bridge_get_edid(struct drm_bridge *bridge,
+					   struct drm_connector *connector)
+{
+	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
+	bool poweroff = !ps_bridge->powered;
+	struct edid *edid;
+
+	/*
+	 * When we end calling get_edid() triggered by an ioctl, i.e
+	 *
+	 *   drm_mode_getconnector (ioctl)
+	 *     -> drm_helper_probe_single_connector_modes
+	 *        -> drm_bridge_connector_get_modes
+	 *           -> ps8640_bridge_get_edid
+	 *
+	 * We need to make sure that what we need is enabled before reading
+	 * EDID, for this chip, we need to do a full poweron, otherwise it will
+	 * fail.
+	 */
+	drm_bridge_chain_pre_enable(bridge);
+
+	edid = drm_get_edid(connector,
+			    ps_bridge->page[PAGE0_DP_CNTL]->adapter);
+
+	/*
+	 * If we call the get_edid() function without having enabled the chip
+	 * before, return the chip to its original power state.
+	 */
+	if (poweroff)
+		drm_bridge_chain_post_disable(bridge);
+
+	return edid;
+}
+
 static const struct drm_bridge_funcs ps8640_bridge_funcs = {
 	.attach = ps8640_bridge_attach,
+	.get_edid = ps8640_bridge_get_edid,
 	.post_disable = ps8640_post_disable,
 	.pre_enable = ps8640_pre_enable,
 };
@@ -294,6 +355,8 @@ static int ps8640_probe(struct i2c_client *client)
 
 	ps_bridge->bridge.funcs = &ps8640_bridge_funcs;
 	ps_bridge->bridge.of_node = dev->of_node;
+	ps_bridge->bridge.ops = DRM_BRIDGE_OP_EDID;
+	ps_bridge->bridge.type = DRM_MODE_CONNECTOR_eDP;
 
 	ps_bridge->page[PAGE0_DP_CNTL] = client;
 
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
index d580b2aa4ce9..6b268f9445b3 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
@@ -89,7 +89,9 @@
 #define VID_MODE_TYPE_NON_BURST_SYNC_EVENTS	0x1
 #define VID_MODE_TYPE_BURST			0x2
 #define VID_MODE_TYPE_MASK			0x3
+#define ENABLE_LOW_POWER_CMD		BIT(15)
 #define VID_MODE_VPG_ENABLE		BIT(16)
+#define VID_MODE_VPG_MODE		BIT(20)
 #define VID_MODE_VPG_HORIZONTAL		BIT(24)
 
 #define DSI_VID_PKT_SIZE		0x3c
@@ -220,6 +222,21 @@
 #define PHY_STATUS_TIMEOUT_US		10000
 #define CMD_PKT_STATUS_TIMEOUT_US	20000
 
+#ifdef CONFIG_DEBUG_FS
+#define VPG_DEFS(name, dsi) \
+	((void __force *)&((*dsi).vpg_defs.name))
+
+#define REGISTER(name, mask, dsi) \
+	{ #name, VPG_DEFS(name, dsi), mask, dsi }
+
+struct debugfs_entries {
+	const char				*name;
+	bool					*reg;
+	u32					mask;
+	struct dw_mipi_dsi			*dsi;
+};
+#endif /* CONFIG_DEBUG_FS */
+
 struct dw_mipi_dsi {
 	struct drm_bridge bridge;
 	struct mipi_dsi_host dsi_host;
@@ -237,9 +254,12 @@ struct dw_mipi_dsi {
 
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *debugfs;
-
-	bool vpg;
-	bool vpg_horizontal;
+	struct debugfs_entries *debugfs_vpg;
+	struct {
+		bool vpg;
+		bool vpg_horizontal;
+		bool vpg_ber_pattern;
+	} vpg_defs;
 #endif /* CONFIG_DEBUG_FS */
 
 	struct dw_mipi_dsi *master; /* dual-dsi master ptr */
@@ -360,13 +380,28 @@ static void dw_mipi_message_config(struct dw_mipi_dsi *dsi,
 	bool lpm = msg->flags & MIPI_DSI_MSG_USE_LPM;
 	u32 val = 0;
 
+	/*
+	 * TODO dw drv improvements
+	 * largest packet sizes during hfp or during vsa/vpb/vfp
+	 * should be computed according to byte lane, lane number and only
+	 * if sending lp cmds in high speed is enable (PHY_TXREQUESTCLKHS)
+	 */
+	dsi_write(dsi, DSI_DPI_LP_CMD_TIM, OUTVACT_LPCMD_TIME(16)
+		  | INVACT_LPCMD_TIME(4));
+
 	if (msg->flags & MIPI_DSI_MSG_REQ_ACK)
 		val |= ACK_RQST_EN;
 	if (lpm)
 		val |= CMD_MODE_ALL_LP;
 
-	dsi_write(dsi, DSI_LPCLK_CTRL, lpm ? 0 : PHY_TXREQUESTCLKHS);
 	dsi_write(dsi, DSI_CMD_MODE_CFG, val);
+
+	val = dsi_read(dsi, DSI_VID_MODE_CFG);
+	if (lpm)
+		val |= ENABLE_LOW_POWER_CMD;
+	else
+		val &= ~ENABLE_LOW_POWER_CMD;
+	dsi_write(dsi, DSI_VID_MODE_CFG, val);
 }
 
 static int dw_mipi_dsi_gen_pkt_hdr_write(struct dw_mipi_dsi *dsi, u32 hdr_val)
@@ -529,9 +564,11 @@ static void dw_mipi_dsi_video_mode_config(struct dw_mipi_dsi *dsi)
 		val |= VID_MODE_TYPE_NON_BURST_SYNC_EVENTS;
 
 #ifdef CONFIG_DEBUG_FS
-	if (dsi->vpg) {
+	if (dsi->vpg_defs.vpg) {
 		val |= VID_MODE_VPG_ENABLE;
-		val |= dsi->vpg_horizontal ? VID_MODE_VPG_HORIZONTAL : 0;
+		val |= dsi->vpg_defs.vpg_horizontal ?
+		       VID_MODE_VPG_HORIZONTAL : 0;
+		val |= dsi->vpg_defs.vpg_ber_pattern ? VID_MODE_VPG_MODE : 0;
 	}
 #endif /* CONFIG_DEBUG_FS */
 
@@ -541,16 +578,22 @@ static void dw_mipi_dsi_video_mode_config(struct dw_mipi_dsi *dsi)
 static void dw_mipi_dsi_set_mode(struct dw_mipi_dsi *dsi,
 				 unsigned long mode_flags)
 {
+	u32 val;
+
 	dsi_write(dsi, DSI_PWR_UP, RESET);
 
 	if (mode_flags & MIPI_DSI_MODE_VIDEO) {
 		dsi_write(dsi, DSI_MODE_CFG, ENABLE_VIDEO_MODE);
 		dw_mipi_dsi_video_mode_config(dsi);
-		dsi_write(dsi, DSI_LPCLK_CTRL, PHY_TXREQUESTCLKHS);
 	} else {
 		dsi_write(dsi, DSI_MODE_CFG, ENABLE_CMD_MODE);
 	}
 
+	val = PHY_TXREQUESTCLKHS;
+	if (dsi->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)
+		val |= AUTO_CLKLANE_CTRL;
+	dsi_write(dsi, DSI_LPCLK_CTRL, val);
+
 	dsi_write(dsi, DSI_PWR_UP, POWERUP);
 }
 
@@ -562,15 +605,30 @@ static void dw_mipi_dsi_disable(struct dw_mipi_dsi *dsi)
 
 static void dw_mipi_dsi_init(struct dw_mipi_dsi *dsi)
 {
+	const struct dw_mipi_dsi_phy_ops *phy_ops = dsi->plat_data->phy_ops;
+	unsigned int esc_rate; /* in MHz */
+	u32 esc_clk_division;
+	int ret;
+
 	/*
 	 * The maximum permitted escape clock is 20MHz and it is derived from
-	 * lanebyteclk, which is running at "lane_mbps / 8".  Thus we want:
-	 *
-	 *     (lane_mbps >> 3) / esc_clk_division < 20
+	 * lanebyteclk, which is running at "lane_mbps / 8".
+	 */
+	if (phy_ops->get_esc_clk_rate) {
+		ret = phy_ops->get_esc_clk_rate(dsi->plat_data->priv_data,
+						&esc_rate);
+		if (ret)
+			DRM_DEBUG_DRIVER("Phy get_esc_clk_rate() failed\n");
+	} else
+		esc_rate = 20; /* Default to 20MHz */
+
+	/*
+	 * We want :
+	 *     (lane_mbps >> 3) / esc_clk_division < X
 	 * which is:
-	 *     (lane_mbps >> 3) / 20 > esc_clk_division
+	 *     (lane_mbps >> 3) / X > esc_clk_division
 	 */
-	u32 esc_clk_division = (dsi->lane_mbps >> 3) / 20 + 1;
+	esc_clk_division = (dsi->lane_mbps >> 3) / esc_rate + 1;
 
 	dsi_write(dsi, DSI_PWR_UP, RESET);
 
@@ -611,14 +669,6 @@ static void dw_mipi_dsi_dpi_config(struct dw_mipi_dsi *dsi,
 	dsi_write(dsi, DSI_DPI_VCID, DPI_VCID(dsi->channel));
 	dsi_write(dsi, DSI_DPI_COLOR_CODING, color);
 	dsi_write(dsi, DSI_DPI_CFG_POL, val);
-	/*
-	 * TODO dw drv improvements
-	 * largest packet sizes during hfp or during vsa/vpb/vfp
-	 * should be computed according to byte lane, lane number and only
-	 * if sending lp cmds in high speed is enable (PHY_TXREQUESTCLKHS)
-	 */
-	dsi_write(dsi, DSI_DPI_LP_CMD_TIM, OUTVACT_LPCMD_TIME(4)
-		  | INVACT_LPCMD_TIME(4));
 }
 
 static void dw_mipi_dsi_packet_handler_config(struct dw_mipi_dsi *dsi)
@@ -964,6 +1014,66 @@ static const struct drm_bridge_funcs dw_mipi_dsi_bridge_funcs = {
 
 #ifdef CONFIG_DEBUG_FS
 
+static int dw_mipi_dsi_debugfs_write(void *data, u64 val)
+{
+	struct debugfs_entries *vpg = data;
+	struct dw_mipi_dsi *dsi;
+	u32 mode_cfg;
+
+	if (!vpg)
+		return -ENODEV;
+
+	dsi = vpg->dsi;
+
+	*vpg->reg = (bool)val;
+
+	mode_cfg = dsi_read(dsi, DSI_VID_MODE_CFG);
+
+	if (*vpg->reg)
+		mode_cfg |= vpg->mask;
+	else
+		mode_cfg &= ~vpg->mask;
+
+	dsi_write(dsi, DSI_VID_MODE_CFG, mode_cfg);
+
+	return 0;
+}
+
+static int dw_mipi_dsi_debugfs_show(void *data, u64 *val)
+{
+	struct debugfs_entries *vpg = data;
+
+	if (!vpg)
+		return -ENODEV;
+
+	*val = *vpg->reg;
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(fops_x32, dw_mipi_dsi_debugfs_show,
+			 dw_mipi_dsi_debugfs_write, "%llu\n");
+
+static void debugfs_create_files(void *data)
+{
+	struct dw_mipi_dsi *dsi = data;
+	struct debugfs_entries debugfs[] = {
+		REGISTER(vpg, VID_MODE_VPG_ENABLE, dsi),
+		REGISTER(vpg_horizontal, VID_MODE_VPG_HORIZONTAL, dsi),
+		REGISTER(vpg_ber_pattern, VID_MODE_VPG_MODE, dsi),
+	};
+	int i;
+
+	dsi->debugfs_vpg = kmemdup(debugfs, sizeof(debugfs), GFP_KERNEL);
+	if (!dsi->debugfs_vpg)
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(debugfs); i++)
+		debugfs_create_file(dsi->debugfs_vpg[i].name, 0644,
+				    dsi->debugfs, &dsi->debugfs_vpg[i],
+				    &fops_x32);
+}
+
 static void dw_mipi_dsi_debugfs_init(struct dw_mipi_dsi *dsi)
 {
 	dsi->debugfs = debugfs_create_dir(dev_name(dsi->dev), NULL);
@@ -972,14 +1082,13 @@ static void dw_mipi_dsi_debugfs_init(struct dw_mipi_dsi *dsi)
 		return;
 	}
 
-	debugfs_create_bool("vpg", 0660, dsi->debugfs, &dsi->vpg);
-	debugfs_create_bool("vpg_horizontal", 0660, dsi->debugfs,
-			    &dsi->vpg_horizontal);
+	debugfs_create_files(dsi);
 }
 
 static void dw_mipi_dsi_debugfs_remove(struct dw_mipi_dsi *dsi)
 {
 	debugfs_remove_recursive(dsi->debugfs);
+	kfree(dsi->debugfs_vpg);
 }
 
 #else
diff --git a/drivers/gpu/drm/bridge/tc358762.c b/drivers/gpu/drm/bridge/tc358762.c
new file mode 100644
index 000000000000..1bfdfc6affaf
--- /dev/null
+++ b/drivers/gpu/drm/bridge/tc358762.c
@@ -0,0 +1,280 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020 Marek Vasut <marex@denx.de>
+ *
+ * Based on tc358764.c by
+ *  Andrzej Hajda <a.hajda@samsung.com>
+ *  Maciej Purski <m.purski@samsung.com>
+ *
+ * Based on rpi_touchscreen.c by
+ *  Eric Anholt <eric@anholt.net>
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/of_graph.h>
+#include <linux/regulator/consumer.h>
+
+#include <video/mipi_display.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+
+/* PPI layer registers */
+#define PPI_STARTPPI		0x0104 /* START control bit */
+#define PPI_LPTXTIMECNT		0x0114 /* LPTX timing signal */
+#define PPI_D0S_ATMR		0x0144
+#define PPI_D1S_ATMR		0x0148
+#define PPI_D0S_CLRSIPOCOUNT	0x0164 /* Assertion timer for Lane 0 */
+#define PPI_D1S_CLRSIPOCOUNT	0x0168 /* Assertion timer for Lane 1 */
+#define PPI_START_FUNCTION	1
+
+/* DSI layer registers */
+#define DSI_STARTDSI		0x0204 /* START control bit of DSI-TX */
+#define DSI_LANEENABLE		0x0210 /* Enables each lane */
+#define DSI_RX_START		1
+
+/* LCDC/DPI Host Registers */
+#define LCDCTRL			0x0420
+
+/* SPI Master Registers */
+#define SPICMR			0x0450
+#define SPITCR			0x0454
+
+/* System Controller Registers */
+#define SYSCTRL			0x0464
+
+/* System registers */
+#define LPX_PERIOD		3
+
+/* Lane enable PPI and DSI register bits */
+#define LANEENABLE_CLEN		BIT(0)
+#define LANEENABLE_L0EN		BIT(1)
+#define LANEENABLE_L1EN		BIT(2)
+
+struct tc358762 {
+	struct device *dev;
+	struct drm_bridge bridge;
+	struct drm_connector connector;
+	struct regulator *regulator;
+	struct drm_bridge *panel_bridge;
+	bool pre_enabled;
+	int error;
+};
+
+static int tc358762_clear_error(struct tc358762 *ctx)
+{
+	int ret = ctx->error;
+
+	ctx->error = 0;
+	return ret;
+}
+
+static void tc358762_write(struct tc358762 *ctx, u16 addr, u32 val)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+	ssize_t ret;
+	u8 data[6];
+
+	if (ctx->error)
+		return;
+
+	data[0] = addr;
+	data[1] = addr >> 8;
+	data[2] = val;
+	data[3] = val >> 8;
+	data[4] = val >> 16;
+	data[5] = val >> 24;
+
+	ret = mipi_dsi_generic_write(dsi, data, sizeof(data));
+	if (ret < 0)
+		ctx->error = ret;
+}
+
+static inline struct tc358762 *bridge_to_tc358762(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct tc358762, bridge);
+}
+
+static int tc358762_init(struct tc358762 *ctx)
+{
+	tc358762_write(ctx, DSI_LANEENABLE,
+		       LANEENABLE_L0EN | LANEENABLE_CLEN);
+	tc358762_write(ctx, PPI_D0S_CLRSIPOCOUNT, 5);
+	tc358762_write(ctx, PPI_D1S_CLRSIPOCOUNT, 5);
+	tc358762_write(ctx, PPI_D0S_ATMR, 0);
+	tc358762_write(ctx, PPI_D1S_ATMR, 0);
+	tc358762_write(ctx, PPI_LPTXTIMECNT, LPX_PERIOD);
+
+	tc358762_write(ctx, SPICMR, 0x00);
+	tc358762_write(ctx, LCDCTRL, 0x00100150);
+	tc358762_write(ctx, SYSCTRL, 0x040f);
+	msleep(100);
+
+	tc358762_write(ctx, PPI_STARTPPI, PPI_START_FUNCTION);
+	tc358762_write(ctx, DSI_STARTDSI, DSI_RX_START);
+
+	msleep(100);
+
+	return tc358762_clear_error(ctx);
+}
+
+static void tc358762_post_disable(struct drm_bridge *bridge)
+{
+	struct tc358762 *ctx = bridge_to_tc358762(bridge);
+	int ret;
+
+	/*
+	 * The post_disable hook might be called multiple times.
+	 * We want to avoid regulator imbalance below.
+	 */
+	if (!ctx->pre_enabled)
+		return;
+
+	ctx->pre_enabled = false;
+
+	ret = regulator_disable(ctx->regulator);
+	if (ret < 0)
+		dev_err(ctx->dev, "error disabling regulators (%d)\n", ret);
+}
+
+static void tc358762_pre_enable(struct drm_bridge *bridge)
+{
+	struct tc358762 *ctx = bridge_to_tc358762(bridge);
+	int ret;
+
+	ret = regulator_enable(ctx->regulator);
+	if (ret < 0)
+		dev_err(ctx->dev, "error enabling regulators (%d)\n", ret);
+
+	ret = tc358762_init(ctx);
+	if (ret < 0)
+		dev_err(ctx->dev, "error initializing bridge (%d)\n", ret);
+
+	ctx->pre_enabled = true;
+}
+
+static int tc358762_attach(struct drm_bridge *bridge,
+			   enum drm_bridge_attach_flags flags)
+{
+	struct tc358762 *ctx = bridge_to_tc358762(bridge);
+
+	return drm_bridge_attach(bridge->encoder, ctx->panel_bridge,
+				 bridge, flags);
+}
+
+static const struct drm_bridge_funcs tc358762_bridge_funcs = {
+	.post_disable = tc358762_post_disable,
+	.pre_enable = tc358762_pre_enable,
+	.attach = tc358762_attach,
+};
+
+static int tc358762_parse_dt(struct tc358762 *ctx)
+{
+	struct drm_bridge *panel_bridge;
+	struct device *dev = ctx->dev;
+	struct drm_panel *panel;
+	int ret;
+
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 0, &panel, NULL);
+	if (ret)
+		return ret;
+
+	panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+
+	if (IS_ERR(panel_bridge))
+		return PTR_ERR(panel_bridge);
+
+	ctx->panel_bridge = panel_bridge;
+
+	return 0;
+}
+
+static int tc358762_configure_regulators(struct tc358762 *ctx)
+{
+	ctx->regulator = devm_regulator_get(ctx->dev, "vddc");
+	if (IS_ERR(ctx->regulator))
+		return PTR_ERR(ctx->regulator);
+
+	return 0;
+}
+
+static int tc358762_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct tc358762 *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(dev, sizeof(struct tc358762), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	ctx->dev = dev;
+	ctx->pre_enabled = false;
+
+	/* TODO: Find out how to get dual-lane mode working */
+	dsi->lanes = 1;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
+			  MIPI_DSI_MODE_LPM;
+
+	ret = tc358762_parse_dt(ctx);
+	if (ret < 0)
+		return ret;
+
+	ret = tc358762_configure_regulators(ctx);
+	if (ret < 0)
+		return ret;
+
+	ctx->bridge.funcs = &tc358762_bridge_funcs;
+	ctx->bridge.type = DRM_MODE_CONNECTOR_DPI;
+	ctx->bridge.of_node = dev->of_node;
+
+	drm_bridge_add(&ctx->bridge);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		drm_bridge_remove(&ctx->bridge);
+		dev_err(dev, "failed to attach dsi\n");
+	}
+
+	return ret;
+}
+
+static int tc358762_remove(struct mipi_dsi_device *dsi)
+{
+	struct tc358762 *ctx = mipi_dsi_get_drvdata(dsi);
+
+	mipi_dsi_detach(dsi);
+	drm_bridge_remove(&ctx->bridge);
+
+	return 0;
+}
+
+static const struct of_device_id tc358762_of_match[] = {
+	{ .compatible = "toshiba,tc358762" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, tc358762_of_match);
+
+static struct mipi_dsi_driver tc358762_driver = {
+	.probe = tc358762_probe,
+	.remove = tc358762_remove,
+	.driver = {
+		.name = "tc358762",
+		.of_match_table = tc358762_of_match,
+	},
+};
+module_mipi_dsi_driver(tc358762_driver);
+
+MODULE_AUTHOR("Marek Vasut <marex@denx.de>");
+MODULE_DESCRIPTION("MIPI-DSI based Driver for TC358762 DSI/DPI Bridge");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/bridge/tc358764.c b/drivers/gpu/drm/bridge/tc358764.c
index 5ac1430fab04..d89394bc5aa4 100644
--- a/drivers/gpu/drm/bridge/tc358764.c
+++ b/drivers/gpu/drm/bridge/tc358764.c
@@ -153,10 +153,9 @@ static const char * const tc358764_supplies[] = {
 struct tc358764 {
 	struct device *dev;
 	struct drm_bridge bridge;
-	struct drm_connector connector;
 	struct regulator_bulk_data supplies[ARRAY_SIZE(tc358764_supplies)];
 	struct gpio_desc *gpio_reset;
-	struct drm_panel *panel;
+	struct drm_bridge *panel_bridge;
 	int error;
 };
 
@@ -210,12 +209,6 @@ static inline struct tc358764 *bridge_to_tc358764(struct drm_bridge *bridge)
 	return container_of(bridge, struct tc358764, bridge);
 }
 
-static inline
-struct tc358764 *connector_to_tc358764(struct drm_connector *connector)
-{
-	return container_of(connector, struct tc358764, connector);
-}
-
 static int tc358764_init(struct tc358764 *ctx)
 {
 	u32 v = 0;
@@ -278,43 +271,11 @@ static void tc358764_reset(struct tc358764 *ctx)
 	usleep_range(1000, 2000);
 }
 
-static int tc358764_get_modes(struct drm_connector *connector)
-{
-	struct tc358764 *ctx = connector_to_tc358764(connector);
-
-	return drm_panel_get_modes(ctx->panel, connector);
-}
-
-static const
-struct drm_connector_helper_funcs tc358764_connector_helper_funcs = {
-	.get_modes = tc358764_get_modes,
-};
-
-static const struct drm_connector_funcs tc358764_connector_funcs = {
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = drm_connector_cleanup,
-	.reset = drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
-static void tc358764_disable(struct drm_bridge *bridge)
-{
-	struct tc358764 *ctx = bridge_to_tc358764(bridge);
-	int ret = drm_panel_disable(bridge_to_tc358764(bridge)->panel);
-
-	if (ret < 0)
-		dev_err(ctx->dev, "error disabling panel (%d)\n", ret);
-}
-
 static void tc358764_post_disable(struct drm_bridge *bridge)
 {
 	struct tc358764 *ctx = bridge_to_tc358764(bridge);
 	int ret;
 
-	ret = drm_panel_unprepare(ctx->panel);
-	if (ret < 0)
-		dev_err(ctx->dev, "error unpreparing panel (%d)\n", ret);
 	tc358764_reset(ctx);
 	usleep_range(10000, 15000);
 	ret = regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies);
@@ -335,73 +296,28 @@ static void tc358764_pre_enable(struct drm_bridge *bridge)
 	ret = tc358764_init(ctx);
 	if (ret < 0)
 		dev_err(ctx->dev, "error initializing bridge (%d)\n", ret);
-	ret = drm_panel_prepare(ctx->panel);
-	if (ret < 0)
-		dev_err(ctx->dev, "error preparing panel (%d)\n", ret);
-}
-
-static void tc358764_enable(struct drm_bridge *bridge)
-{
-	struct tc358764 *ctx = bridge_to_tc358764(bridge);
-	int ret = drm_panel_enable(ctx->panel);
-
-	if (ret < 0)
-		dev_err(ctx->dev, "error enabling panel (%d)\n", ret);
 }
 
 static int tc358764_attach(struct drm_bridge *bridge,
 			   enum drm_bridge_attach_flags flags)
 {
 	struct tc358764 *ctx = bridge_to_tc358764(bridge);
-	struct drm_device *drm = bridge->dev;
-	int ret;
-
-	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
-		DRM_ERROR("Fix bridge driver to make connector optional!");
-		return -EINVAL;
-	}
-
-	ctx->connector.polled = DRM_CONNECTOR_POLL_HPD;
-	ret = drm_connector_init(drm, &ctx->connector,
-				 &tc358764_connector_funcs,
-				 DRM_MODE_CONNECTOR_LVDS);
-	if (ret) {
-		DRM_ERROR("Failed to initialize connector\n");
-		return ret;
-	}
-
-	drm_connector_helper_add(&ctx->connector,
-				 &tc358764_connector_helper_funcs);
-	drm_connector_attach_encoder(&ctx->connector, bridge->encoder);
-	drm_panel_attach(ctx->panel, &ctx->connector);
-	ctx->connector.funcs->reset(&ctx->connector);
-	drm_connector_register(&ctx->connector);
-
-	return 0;
-}
-
-static void tc358764_detach(struct drm_bridge *bridge)
-{
-	struct tc358764 *ctx = bridge_to_tc358764(bridge);
 
-	drm_connector_unregister(&ctx->connector);
-	drm_panel_detach(ctx->panel);
-	ctx->panel = NULL;
-	drm_connector_put(&ctx->connector);
+	return drm_bridge_attach(bridge->encoder, ctx->panel_bridge,
+				 bridge, flags);
 }
 
 static const struct drm_bridge_funcs tc358764_bridge_funcs = {
-	.disable = tc358764_disable,
 	.post_disable = tc358764_post_disable,
-	.enable = tc358764_enable,
 	.pre_enable = tc358764_pre_enable,
 	.attach = tc358764_attach,
-	.detach = tc358764_detach,
 };
 
 static int tc358764_parse_dt(struct tc358764 *ctx)
 {
+	struct drm_bridge *panel_bridge;
 	struct device *dev = ctx->dev;
+	struct drm_panel *panel;
 	int ret;
 
 	ctx->gpio_reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
@@ -410,12 +326,16 @@ static int tc358764_parse_dt(struct tc358764 *ctx)
 		return PTR_ERR(ctx->gpio_reset);
 	}
 
-	ret = drm_of_find_panel_or_bridge(ctx->dev->of_node, 1, 0, &ctx->panel,
-					  NULL);
-	if (ret && ret != -EPROBE_DEFER)
-		dev_err(dev, "cannot find panel (%d)\n", ret);
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 0, &panel, NULL);
+	if (ret)
+		return ret;
 
-	return ret;
+	panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+	if (IS_ERR(panel_bridge))
+		return PTR_ERR(panel_bridge);
+
+	ctx->panel_bridge = panel_bridge;
+	return 0;
 }
 
 static int tc358764_configure_regulators(struct tc358764 *ctx)
@@ -461,6 +381,7 @@ static int tc358764_probe(struct mipi_dsi_device *dsi)
 		return ret;
 
 	ctx->bridge.funcs = &tc358764_bridge_funcs;
+	ctx->bridge.type = DRM_MODE_CONNECTOR_LVDS;
 	ctx->bridge.of_node = dev->of_node;
 
 	drm_bridge_add(&ctx->bridge);
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index c2777b226c75..34a3e4e9f717 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -244,14 +244,12 @@ struct tc_data {
 	struct drm_dp_aux	aux;
 
 	struct drm_bridge	bridge;
+	struct drm_bridge	*panel_bridge;
 	struct drm_connector	connector;
-	struct drm_panel	*panel;
 
 	/* link settings */
 	struct tc_edp_link	link;
 
-	/* display edid */
-	struct edid		*edid;
 	/* current mode */
 	struct drm_display_mode	mode;
 
@@ -1236,13 +1234,6 @@ static int tc_stream_disable(struct tc_data *tc)
 	return 0;
 }
 
-static void tc_bridge_pre_enable(struct drm_bridge *bridge)
-{
-	struct tc_data *tc = bridge_to_tc(bridge);
-
-	drm_panel_prepare(tc->panel);
-}
-
 static void tc_bridge_enable(struct drm_bridge *bridge)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
@@ -1266,8 +1257,6 @@ static void tc_bridge_enable(struct drm_bridge *bridge)
 		tc_main_link_disable(tc);
 		return;
 	}
-
-	drm_panel_enable(tc->panel);
 }
 
 static void tc_bridge_disable(struct drm_bridge *bridge)
@@ -1275,8 +1264,6 @@ static void tc_bridge_disable(struct drm_bridge *bridge)
 	struct tc_data *tc = bridge_to_tc(bridge);
 	int ret;
 
-	drm_panel_disable(tc->panel);
-
 	ret = tc_stream_disable(tc);
 	if (ret < 0)
 		dev_err(tc->dev, "main link stream stop error: %d\n", ret);
@@ -1286,13 +1273,6 @@ static void tc_bridge_disable(struct drm_bridge *bridge)
 		dev_err(tc->dev, "main link disable error: %d\n", ret);
 }
 
-static void tc_bridge_post_disable(struct drm_bridge *bridge)
-{
-	struct tc_data *tc = bridge_to_tc(bridge);
-
-	drm_panel_unprepare(tc->panel);
-}
-
 static bool tc_bridge_mode_fixup(struct drm_bridge *bridge,
 				 const struct drm_display_mode *mode,
 				 struct drm_display_mode *adj)
@@ -1335,11 +1315,19 @@ static void tc_bridge_mode_set(struct drm_bridge *bridge,
 	tc->mode = *mode;
 }
 
+static struct edid *tc_get_edid(struct drm_bridge *bridge,
+				struct drm_connector *connector)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+
+	return drm_get_edid(connector, &tc->aux.ddc);
+}
+
 static int tc_connector_get_modes(struct drm_connector *connector)
 {
 	struct tc_data *tc = connector_to_tc(connector);
+	int num_modes;
 	struct edid *edid;
-	int count;
 	int ret;
 
 	ret = tc_get_display_props(tc);
@@ -1348,42 +1336,30 @@ static int tc_connector_get_modes(struct drm_connector *connector)
 		return 0;
 	}
 
-	count = drm_panel_get_modes(tc->panel, connector);
-	if (count > 0)
-		return count;
-
-	edid = drm_get_edid(connector, &tc->aux.ddc);
-
-	kfree(tc->edid);
-	tc->edid = edid;
-	if (!edid)
-		return 0;
+	if (tc->panel_bridge) {
+		num_modes = drm_bridge_get_modes(tc->panel_bridge, connector);
+		if (num_modes > 0)
+			return num_modes;
+	}
 
-	drm_connector_update_edid_property(connector, edid);
-	count = drm_add_edid_modes(connector, edid);
+	edid = tc_get_edid(&tc->bridge, connector);
+	num_modes = drm_add_edid_modes(connector, edid);
+	kfree(edid);
 
-	return count;
+	return num_modes;
 }
 
 static const struct drm_connector_helper_funcs tc_connector_helper_funcs = {
 	.get_modes = tc_connector_get_modes,
 };
 
-static enum drm_connector_status tc_connector_detect(struct drm_connector *connector,
-						     bool force)
+static enum drm_connector_status tc_bridge_detect(struct drm_bridge *bridge)
 {
-	struct tc_data *tc = connector_to_tc(connector);
+	struct tc_data *tc = bridge_to_tc(bridge);
 	bool conn;
 	u32 val;
 	int ret;
 
-	if (tc->hpd_pin < 0) {
-		if (tc->panel)
-			return connector_status_connected;
-		else
-			return connector_status_unknown;
-	}
-
 	ret = regmap_read(tc->regmap, GPIOI, &val);
 	if (ret)
 		return connector_status_unknown;
@@ -1396,6 +1372,20 @@ static enum drm_connector_status tc_connector_detect(struct drm_connector *conne
 		return connector_status_disconnected;
 }
 
+static enum drm_connector_status
+tc_connector_detect(struct drm_connector *connector, bool force)
+{
+	struct tc_data *tc = connector_to_tc(connector);
+
+	if (tc->hpd_pin >= 0)
+		return tc_bridge_detect(&tc->bridge);
+
+	if (tc->panel_bridge)
+		return connector_status_connected;
+	else
+		return connector_status_unknown;
+}
+
 static const struct drm_connector_funcs tc_connector_funcs = {
 	.detect = tc_connector_detect,
 	.fill_modes = drm_helper_probe_single_connector_modes,
@@ -1413,16 +1403,20 @@ static int tc_bridge_attach(struct drm_bridge *bridge,
 	struct drm_device *drm = bridge->dev;
 	int ret;
 
-	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
-		DRM_ERROR("Fix bridge driver to make connector optional!");
-		return -EINVAL;
+	if (tc->panel_bridge) {
+		/* If a connector is required then this driver shall create it */
+		ret = drm_bridge_attach(tc->bridge.encoder, tc->panel_bridge,
+					&tc->bridge, flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+		if (ret)
+			return ret;
 	}
 
+	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
+		return 0;
+
 	/* Create DP/eDP connector */
 	drm_connector_helper_add(&tc->connector, &tc_connector_helper_funcs);
-	ret = drm_connector_init(drm, &tc->connector, &tc_connector_funcs,
-				 tc->panel ? DRM_MODE_CONNECTOR_eDP :
-				 DRM_MODE_CONNECTOR_DisplayPort);
+	ret = drm_connector_init(drm, &tc->connector, &tc_connector_funcs, tc->bridge.type);
 	if (ret)
 		return ret;
 
@@ -1435,9 +1429,6 @@ static int tc_bridge_attach(struct drm_bridge *bridge,
 					       DRM_CONNECTOR_POLL_DISCONNECT;
 	}
 
-	if (tc->panel)
-		drm_panel_attach(tc->panel, &tc->connector);
-
 	drm_display_info_set_bus_formats(&tc->connector.display_info,
 					 &bus_format, 1);
 	tc->connector.display_info.bus_flags =
@@ -1453,11 +1444,11 @@ static const struct drm_bridge_funcs tc_bridge_funcs = {
 	.attach = tc_bridge_attach,
 	.mode_valid = tc_mode_valid,
 	.mode_set = tc_bridge_mode_set,
-	.pre_enable = tc_bridge_pre_enable,
 	.enable = tc_bridge_enable,
 	.disable = tc_bridge_disable,
-	.post_disable = tc_bridge_post_disable,
 	.mode_fixup = tc_bridge_mode_fixup,
+	.detect = tc_bridge_detect,
+	.get_edid = tc_get_edid,
 };
 
 static bool tc_readable_reg(struct device *dev, unsigned int reg)
@@ -1547,6 +1538,7 @@ static irqreturn_t tc_irq_handler(int irq, void *arg)
 static int tc_probe(struct i2c_client *client, const struct i2c_device_id *id)
 {
 	struct device *dev = &client->dev;
+	struct drm_panel *panel;
 	struct tc_data *tc;
 	int ret;
 
@@ -1557,10 +1549,23 @@ static int tc_probe(struct i2c_client *client, const struct i2c_device_id *id)
 	tc->dev = dev;
 
 	/* port@2 is the output port */
-	ret = drm_of_find_panel_or_bridge(dev->of_node, 2, 0, &tc->panel, NULL);
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 2, 0, &panel, NULL);
 	if (ret && ret != -ENODEV)
 		return ret;
 
+	if (panel) {
+		struct drm_bridge *panel_bridge;
+
+		panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+		if (IS_ERR(panel_bridge))
+			return PTR_ERR(panel_bridge);
+
+		tc->panel_bridge = panel_bridge;
+		tc->bridge.type = DRM_MODE_CONNECTOR_eDP;
+	} else {
+		tc->bridge.type = DRM_MODE_CONNECTOR_DisplayPort;
+	}
+
 	/* Shut down GPIO is optional */
 	tc->sd_gpio = devm_gpiod_get_optional(dev, "shutdown", GPIOD_OUT_HIGH);
 	if (IS_ERR(tc->sd_gpio))
@@ -1680,6 +1685,10 @@ static int tc_probe(struct i2c_client *client, const struct i2c_device_id *id)
 		return ret;
 
 	tc->bridge.funcs = &tc_bridge_funcs;
+	if (tc->hpd_pin >= 0)
+		tc->bridge.ops |= DRM_BRIDGE_OP_DETECT;
+	tc->bridge.ops |= DRM_BRIDGE_OP_EDID;
+
 	tc->bridge.of_node = dev->of_node;
 	drm_bridge_add(&tc->bridge);
 
diff --git a/drivers/gpu/drm/bridge/tc358775.c b/drivers/gpu/drm/bridge/tc358775.c
new file mode 100644
index 000000000000..2272adcc5b4a
--- /dev/null
+++ b/drivers/gpu/drm/bridge/tc358775.c
@@ -0,0 +1,749 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * TC358775 DSI to LVDS bridge driver
+ *
+ * Copyright (C) 2020 SMART Wireless Computing
+ * Author: Vinay Simha BN <simhavcs@gmail.com>
+ *
+ */
+/* #define DEBUG */
+#include <linux/bitfield.h>
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/gpio/consumer.h>
+#include <linux/i2c.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/regulator/consumer.h>
+#include <linux/slab.h>
+
+#include <asm/unaligned.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
+
+#define FLD_VAL(val, start, end) FIELD_PREP(GENMASK(start, end), val)
+
+/* Registers */
+
+/* DSI D-PHY Layer Registers */
+#define D0W_DPHYCONTTX  0x0004  /* Data Lane 0 DPHY Tx Control */
+#define CLW_DPHYCONTRX  0x0020  /* Clock Lane DPHY Rx Control */
+#define D0W_DPHYCONTRX  0x0024  /* Data Lane 0 DPHY Rx Control */
+#define D1W_DPHYCONTRX  0x0028  /* Data Lane 1 DPHY Rx Control */
+#define D2W_DPHYCONTRX  0x002C  /* Data Lane 2 DPHY Rx Control */
+#define D3W_DPHYCONTRX  0x0030  /* Data Lane 3 DPHY Rx Control */
+#define COM_DPHYCONTRX  0x0038  /* DPHY Rx Common Control */
+#define CLW_CNTRL       0x0040  /* Clock Lane Control */
+#define D0W_CNTRL       0x0044  /* Data Lane 0 Control */
+#define D1W_CNTRL       0x0048  /* Data Lane 1 Control */
+#define D2W_CNTRL       0x004C  /* Data Lane 2 Control */
+#define D3W_CNTRL       0x0050  /* Data Lane 3 Control */
+#define DFTMODE_CNTRL   0x0054  /* DFT Mode Control */
+
+/* DSI PPI Layer Registers */
+#define PPI_STARTPPI    0x0104  /* START control bit of PPI-TX function. */
+#define PPI_START_FUNCTION      1
+
+#define PPI_BUSYPPI     0x0108
+#define PPI_LINEINITCNT 0x0110  /* Line Initialization Wait Counter  */
+#define PPI_LPTXTIMECNT 0x0114
+#define PPI_LANEENABLE  0x0134  /* Enables each lane at the PPI layer. */
+#define PPI_TX_RX_TA    0x013C  /* DSI Bus Turn Around timing parameters */
+
+/* Analog timer function enable */
+#define PPI_CLS_ATMR    0x0140  /* Delay for Clock Lane in LPRX  */
+#define PPI_D0S_ATMR    0x0144  /* Delay for Data Lane 0 in LPRX */
+#define PPI_D1S_ATMR    0x0148  /* Delay for Data Lane 1 in LPRX */
+#define PPI_D2S_ATMR    0x014C  /* Delay for Data Lane 2 in LPRX */
+#define PPI_D3S_ATMR    0x0150  /* Delay for Data Lane 3 in LPRX */
+
+#define PPI_D0S_CLRSIPOCOUNT    0x0164  /* For lane 0 */
+#define PPI_D1S_CLRSIPOCOUNT    0x0168  /* For lane 1 */
+#define PPI_D2S_CLRSIPOCOUNT    0x016C  /* For lane 2 */
+#define PPI_D3S_CLRSIPOCOUNT    0x0170  /* For lane 3 */
+
+#define CLS_PRE         0x0180  /* Digital Counter inside of PHY IO */
+#define D0S_PRE         0x0184  /* Digital Counter inside of PHY IO */
+#define D1S_PRE         0x0188  /* Digital Counter inside of PHY IO */
+#define D2S_PRE         0x018C  /* Digital Counter inside of PHY IO */
+#define D3S_PRE         0x0190  /* Digital Counter inside of PHY IO */
+#define CLS_PREP        0x01A0  /* Digital Counter inside of PHY IO */
+#define D0S_PREP        0x01A4  /* Digital Counter inside of PHY IO */
+#define D1S_PREP        0x01A8  /* Digital Counter inside of PHY IO */
+#define D2S_PREP        0x01AC  /* Digital Counter inside of PHY IO */
+#define D3S_PREP        0x01B0  /* Digital Counter inside of PHY IO */
+#define CLS_ZERO        0x01C0  /* Digital Counter inside of PHY IO */
+#define D0S_ZERO        0x01C4  /* Digital Counter inside of PHY IO */
+#define D1S_ZERO        0x01C8  /* Digital Counter inside of PHY IO */
+#define D2S_ZERO        0x01CC  /* Digital Counter inside of PHY IO */
+#define D3S_ZERO        0x01D0  /* Digital Counter inside of PHY IO */
+
+#define PPI_CLRFLG      0x01E0  /* PRE Counters has reached set values */
+#define PPI_CLRSIPO     0x01E4  /* Clear SIPO values, Slave mode use only. */
+#define HSTIMEOUT       0x01F0  /* HS Rx Time Out Counter */
+#define HSTIMEOUTENABLE 0x01F4  /* Enable HS Rx Time Out Counter */
+#define DSI_STARTDSI    0x0204  /* START control bit of DSI-TX function */
+#define DSI_RX_START	1
+
+#define DSI_BUSYDSI     0x0208
+#define DSI_LANEENABLE  0x0210  /* Enables each lane at the Protocol layer. */
+#define DSI_LANESTATUS0 0x0214  /* Displays lane is in HS RX mode. */
+#define DSI_LANESTATUS1 0x0218  /* Displays lane is in ULPS or STOP state */
+
+#define DSI_INTSTATUS   0x0220  /* Interrupt Status */
+#define DSI_INTMASK     0x0224  /* Interrupt Mask */
+#define DSI_INTCLR      0x0228  /* Interrupt Clear */
+#define DSI_LPTXTO      0x0230  /* Low Power Tx Time Out Counter */
+
+#define DSIERRCNT       0x0300  /* DSI Error Count */
+#define APLCTRL         0x0400  /* Application Layer Control */
+#define RDPKTLN         0x0404  /* Command Read Packet Length */
+
+#define VPCTRL          0x0450  /* Video Path Control */
+#define HTIM1           0x0454  /* Horizontal Timing Control 1 */
+#define HTIM2           0x0458  /* Horizontal Timing Control 2 */
+#define VTIM1           0x045C  /* Vertical Timing Control 1 */
+#define VTIM2           0x0460  /* Vertical Timing Control 2 */
+#define VFUEN           0x0464  /* Video Frame Timing Update Enable */
+#define VFUEN_EN	BIT(0)  /* Upload Enable */
+
+/* Mux Input Select for LVDS LINK Input */
+#define LV_MX0003        0x0480  /* Bit 0 to 3 */
+#define LV_MX0407        0x0484  /* Bit 4 to 7 */
+#define LV_MX0811        0x0488  /* Bit 8 to 11 */
+#define LV_MX1215        0x048C  /* Bit 12 to 15 */
+#define LV_MX1619        0x0490  /* Bit 16 to 19 */
+#define LV_MX2023        0x0494  /* Bit 20 to 23 */
+#define LV_MX2427        0x0498  /* Bit 24 to 27 */
+#define LV_MX(b0, b1, b2, b3)	(FLD_VAL(b0, 4, 0) | FLD_VAL(b1, 12, 8) | \
+				FLD_VAL(b2, 20, 16) | FLD_VAL(b3, 28, 24))
+
+/* Input bit numbers used in mux registers */
+enum {
+	LVI_R0,
+	LVI_R1,
+	LVI_R2,
+	LVI_R3,
+	LVI_R4,
+	LVI_R5,
+	LVI_R6,
+	LVI_R7,
+	LVI_G0,
+	LVI_G1,
+	LVI_G2,
+	LVI_G3,
+	LVI_G4,
+	LVI_G5,
+	LVI_G6,
+	LVI_G7,
+	LVI_B0,
+	LVI_B1,
+	LVI_B2,
+	LVI_B3,
+	LVI_B4,
+	LVI_B5,
+	LVI_B6,
+	LVI_B7,
+	LVI_HS,
+	LVI_VS,
+	LVI_DE,
+	LVI_L0
+};
+
+#define LVCFG           0x049C  /* LVDS Configuration  */
+#define LVPHY0          0x04A0  /* LVDS PHY 0 */
+#define LV_PHY0_RST(v)          FLD_VAL(v, 22, 22) /* PHY reset */
+#define LV_PHY0_IS(v)           FLD_VAL(v, 15, 14)
+#define LV_PHY0_ND(v)           FLD_VAL(v, 4, 0) /* Frequency range select */
+#define LV_PHY0_PRBS_ON(v)      FLD_VAL(v, 20, 16) /* Clock/Data Flag pins */
+
+#define LVPHY1          0x04A4  /* LVDS PHY 1 */
+#define SYSSTAT         0x0500  /* System Status  */
+#define SYSRST          0x0504  /* System Reset  */
+
+#define SYS_RST_I2CS	BIT(0) /* Reset I2C-Slave controller */
+#define SYS_RST_I2CM	BIT(1) /* Reset I2C-Master controller */
+#define SYS_RST_LCD	BIT(2) /* Reset LCD controller */
+#define SYS_RST_BM	BIT(3) /* Reset Bus Management controller */
+#define SYS_RST_DSIRX	BIT(4) /* Reset DSI-RX and App controller */
+#define SYS_RST_REG	BIT(5) /* Reset Register module */
+
+/* GPIO Registers */
+#define GPIOC           0x0520  /* GPIO Control  */
+#define GPIOO           0x0524  /* GPIO Output  */
+#define GPIOI           0x0528  /* GPIO Input  */
+
+/* I2C Registers */
+#define I2CTIMCTRL      0x0540  /* I2C IF Timing and Enable Control */
+#define I2CMADDR        0x0544  /* I2C Master Addressing */
+#define WDATAQ          0x0548  /* Write Data Queue */
+#define RDATAQ          0x054C  /* Read Data Queue */
+
+/* Chip ID and Revision ID Register */
+#define IDREG           0x0580
+
+#define LPX_PERIOD		4
+#define TTA_GET			0x40000
+#define TTA_SURE		6
+#define SINGLE_LINK		1
+#define DUAL_LINK		2
+
+#define TC358775XBG_ID  0x00007500
+
+/* Debug Registers */
+#define DEBUG00         0x05A0  /* Debug */
+#define DEBUG01         0x05A4  /* LVDS Data */
+
+#define DSI_CLEN_BIT		BIT(0)
+#define DIVIDE_BY_3		3 /* PCLK=DCLK/3 */
+#define DIVIDE_BY_6		6 /* PCLK=DCLK/6 */
+#define LVCFG_LVEN_BIT		BIT(0)
+
+#define L0EN BIT(1)
+
+#define TC358775_VPCTRL_VSDELAY__MASK	0x3FF00000
+#define TC358775_VPCTRL_VSDELAY__SHIFT	20
+static inline u32 TC358775_VPCTRL_VSDELAY(uint32_t val)
+{
+	return ((val) << TC358775_VPCTRL_VSDELAY__SHIFT) &
+			TC358775_VPCTRL_VSDELAY__MASK;
+}
+
+#define TC358775_VPCTRL_OPXLFMT__MASK	0x00000100
+#define TC358775_VPCTRL_OPXLFMT__SHIFT	8
+static inline u32 TC358775_VPCTRL_OPXLFMT(uint32_t val)
+{
+	return ((val) << TC358775_VPCTRL_OPXLFMT__SHIFT) &
+			TC358775_VPCTRL_OPXLFMT__MASK;
+}
+
+#define TC358775_VPCTRL_MSF__MASK	0x00000001
+#define TC358775_VPCTRL_MSF__SHIFT	0
+static inline u32 TC358775_VPCTRL_MSF(uint32_t val)
+{
+	return ((val) << TC358775_VPCTRL_MSF__SHIFT) &
+			TC358775_VPCTRL_MSF__MASK;
+}
+
+#define TC358775_LVCFG_PCLKDIV__MASK	0x000000f0
+#define TC358775_LVCFG_PCLKDIV__SHIFT	4
+static inline u32 TC358775_LVCFG_PCLKDIV(uint32_t val)
+{
+	return ((val) << TC358775_LVCFG_PCLKDIV__SHIFT) &
+			TC358775_LVCFG_PCLKDIV__MASK;
+}
+
+#define TC358775_LVCFG_LVDLINK__MASK                         0x00000002
+#define TC358775_LVCFG_LVDLINK__SHIFT                        0
+static inline u32 TC358775_LVCFG_LVDLINK(uint32_t val)
+{
+	return ((val) << TC358775_LVCFG_LVDLINK__SHIFT) &
+			TC358775_LVCFG_LVDLINK__MASK;
+}
+
+enum tc358775_ports {
+	TC358775_DSI_IN,
+	TC358775_LVDS_OUT0,
+	TC358775_LVDS_OUT1,
+};
+
+struct tc_data {
+	struct i2c_client	*i2c;
+	struct device		*dev;
+
+	struct drm_bridge	bridge;
+	struct drm_bridge	*panel_bridge;
+
+	struct device_node *host_node;
+	struct mipi_dsi_device *dsi;
+	u8 num_dsi_lanes;
+
+	struct regulator	*vdd;
+	struct regulator	*vddio;
+	struct gpio_desc	*reset_gpio;
+	struct gpio_desc	*stby_gpio;
+	u8			lvds_link; /* single-link or dual-link */
+	u8			bpc;
+};
+
+static inline struct tc_data *bridge_to_tc(struct drm_bridge *b)
+{
+	return container_of(b, struct tc_data, bridge);
+}
+
+static void tc_bridge_pre_enable(struct drm_bridge *bridge)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+	struct device *dev = &tc->dsi->dev;
+	int ret;
+
+	ret = regulator_enable(tc->vddio);
+	if (ret < 0)
+		dev_err(dev, "regulator vddio enable failed, %d\n", ret);
+	usleep_range(10000, 11000);
+
+	ret = regulator_enable(tc->vdd);
+	if (ret < 0)
+		dev_err(dev, "regulator vdd enable failed, %d\n", ret);
+	usleep_range(10000, 11000);
+
+	gpiod_set_value(tc->stby_gpio, 0);
+	usleep_range(10000, 11000);
+
+	gpiod_set_value(tc->reset_gpio, 0);
+	usleep_range(10, 20);
+}
+
+static void tc_bridge_post_disable(struct drm_bridge *bridge)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+	struct device *dev = &tc->dsi->dev;
+	int ret;
+
+	gpiod_set_value(tc->reset_gpio, 1);
+	usleep_range(10, 20);
+
+	gpiod_set_value(tc->stby_gpio, 1);
+	usleep_range(10000, 11000);
+
+	ret = regulator_disable(tc->vdd);
+	if (ret < 0)
+		dev_err(dev, "regulator vdd disable failed, %d\n", ret);
+	usleep_range(10000, 11000);
+
+	ret = regulator_disable(tc->vddio);
+	if (ret < 0)
+		dev_err(dev, "regulator vddio disable failed, %d\n", ret);
+	usleep_range(10000, 11000);
+}
+
+static void d2l_read(struct i2c_client *i2c, u16 addr, u32 *val)
+{
+	int ret;
+	u8 buf_addr[2];
+
+	put_unaligned_be16(addr, buf_addr);
+	ret = i2c_master_send(i2c, buf_addr, sizeof(buf_addr));
+	if (ret < 0)
+		goto fail;
+
+	ret = i2c_master_recv(i2c, (u8 *)val, sizeof(*val));
+	if (ret < 0)
+		goto fail;
+
+	pr_debug("d2l: I2C : addr:%04x value:%08x\n", addr, *val);
+
+fail:
+	dev_err(&i2c->dev, "Error %d reading from subaddress 0x%x\n",
+		ret, addr);
+}
+
+static void d2l_write(struct i2c_client *i2c, u16 addr, u32 val)
+{
+	u8 data[6];
+	int ret;
+
+	put_unaligned_be16(addr, data);
+	put_unaligned_le32(val, data + 2);
+
+	ret = i2c_master_send(i2c, data, ARRAY_SIZE(data));
+	if (ret < 0)
+		dev_err(&i2c->dev, "Error %d writing to subaddress 0x%x\n",
+			ret, addr);
+}
+
+/* helper function to access bus_formats */
+static struct drm_connector *get_connector(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct drm_connector *connector;
+
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
+		if (connector->encoder == encoder)
+			return connector;
+
+	return NULL;
+}
+
+static void tc_bridge_enable(struct drm_bridge *bridge)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+	u32 hback_porch, hsync_len, hfront_porch, hactive, htime1, htime2;
+	u32 vback_porch, vsync_len, vfront_porch, vactive, vtime1, vtime2;
+	u32 val = 0;
+	u16 dsiclk, clkdiv, byteclk, t1, t2, t3, vsdelay;
+	struct drm_display_mode *mode;
+	struct drm_connector *connector = get_connector(bridge->encoder);
+
+	mode = &bridge->encoder->crtc->state->adjusted_mode;
+
+	hback_porch = mode->htotal - mode->hsync_end;
+	hsync_len  = mode->hsync_end - mode->hsync_start;
+	vback_porch = mode->vtotal - mode->vsync_end;
+	vsync_len  = mode->vsync_end - mode->vsync_start;
+
+	htime1 = (hback_porch << 16) + hsync_len;
+	vtime1 = (vback_porch << 16) + vsync_len;
+
+	hfront_porch = mode->hsync_start - mode->hdisplay;
+	hactive = mode->hdisplay;
+	vfront_porch = mode->vsync_start - mode->vdisplay;
+	vactive = mode->vdisplay;
+
+	htime2 = (hfront_porch << 16) + hactive;
+	vtime2 = (vfront_porch << 16) + vactive;
+
+	d2l_read(tc->i2c, IDREG, &val);
+
+	dev_info(tc->dev, "DSI2LVDS Chip ID.%02x Revision ID. %02x **\n",
+		 (val >> 8) & 0xFF, val & 0xFF);
+
+	d2l_write(tc->i2c, SYSRST, SYS_RST_REG | SYS_RST_DSIRX | SYS_RST_BM |
+		  SYS_RST_LCD | SYS_RST_I2CM | SYS_RST_I2CS);
+	usleep_range(30000, 40000);
+
+	d2l_write(tc->i2c, PPI_TX_RX_TA, TTA_GET | TTA_SURE);
+	d2l_write(tc->i2c, PPI_LPTXTIMECNT, LPX_PERIOD);
+	d2l_write(tc->i2c, PPI_D0S_CLRSIPOCOUNT, 3);
+	d2l_write(tc->i2c, PPI_D1S_CLRSIPOCOUNT, 3);
+	d2l_write(tc->i2c, PPI_D2S_CLRSIPOCOUNT, 3);
+	d2l_write(tc->i2c, PPI_D3S_CLRSIPOCOUNT, 3);
+
+	val = ((L0EN << tc->num_dsi_lanes) - L0EN) | DSI_CLEN_BIT;
+	d2l_write(tc->i2c, PPI_LANEENABLE, val);
+	d2l_write(tc->i2c, DSI_LANEENABLE, val);
+
+	d2l_write(tc->i2c, PPI_STARTPPI, PPI_START_FUNCTION);
+	d2l_write(tc->i2c, DSI_STARTDSI, DSI_RX_START);
+
+	if (tc->bpc == 8)
+		val = TC358775_VPCTRL_OPXLFMT(1);
+	else /* bpc = 6; */
+		val = TC358775_VPCTRL_MSF(1);
+
+	dsiclk = mode->crtc_clock * 3 * tc->bpc / tc->num_dsi_lanes / 1000;
+	clkdiv = dsiclk / DIVIDE_BY_3 * tc->lvds_link;
+	byteclk = dsiclk / 4;
+	t1 = hactive * (tc->bpc * 3 / 8) / tc->num_dsi_lanes;
+	t2 = ((100000 / clkdiv)) * (hactive + hback_porch + hsync_len + hfront_porch) / 1000;
+	t3 = ((t2 * byteclk) / 100) - (hactive * (tc->bpc * 3 / 8) /
+		tc->num_dsi_lanes);
+
+	vsdelay = (clkdiv * (t1 + t3) / byteclk) - hback_porch - hsync_len - hactive;
+
+	val |= TC358775_VPCTRL_VSDELAY(vsdelay);
+	d2l_write(tc->i2c, VPCTRL, val);
+
+	d2l_write(tc->i2c, HTIM1, htime1);
+	d2l_write(tc->i2c, VTIM1, vtime1);
+	d2l_write(tc->i2c, HTIM2, htime2);
+	d2l_write(tc->i2c, VTIM2, vtime2);
+
+	d2l_write(tc->i2c, VFUEN, VFUEN_EN);
+	d2l_write(tc->i2c, SYSRST, SYS_RST_LCD);
+	d2l_write(tc->i2c, LVPHY0, LV_PHY0_PRBS_ON(4) | LV_PHY0_ND(6));
+
+	dev_dbg(tc->dev, "bus_formats %04x bpc %d\n",
+		connector->display_info.bus_formats[0],
+		tc->bpc);
+	/*
+	 * Default hardware register settings of tc358775 configured
+	 * with MEDIA_BUS_FMT_RGB888_1X7X4_JEIDA jeida-24 format
+	 */
+	if (connector->display_info.bus_formats[0] ==
+		MEDIA_BUS_FMT_RGB888_1X7X4_SPWG) {
+		/* VESA-24 */
+		d2l_write(tc->i2c, LV_MX0003, LV_MX(LVI_R0, LVI_R1, LVI_R2, LVI_R3));
+		d2l_write(tc->i2c, LV_MX0407, LV_MX(LVI_R4, LVI_R7, LVI_R5, LVI_G0));
+		d2l_write(tc->i2c, LV_MX0811, LV_MX(LVI_G1, LVI_G2, LVI_G6, LVI_G7));
+		d2l_write(tc->i2c, LV_MX1215, LV_MX(LVI_G3, LVI_G4, LVI_G5, LVI_B0));
+		d2l_write(tc->i2c, LV_MX1619, LV_MX(LVI_B6, LVI_B7, LVI_B1, LVI_B2));
+		d2l_write(tc->i2c, LV_MX2023, LV_MX(LVI_B3, LVI_B4, LVI_B5, LVI_L0));
+		d2l_write(tc->i2c, LV_MX2427, LV_MX(LVI_HS, LVI_VS, LVI_DE, LVI_R6));
+	} else { /*  MEDIA_BUS_FMT_RGB666_1X7X3_SPWG - JEIDA-18 */
+		d2l_write(tc->i2c, LV_MX0003, LV_MX(LVI_R0, LVI_R1, LVI_R2, LVI_R3));
+		d2l_write(tc->i2c, LV_MX0407, LV_MX(LVI_R4, LVI_L0, LVI_R5, LVI_G0));
+		d2l_write(tc->i2c, LV_MX0811, LV_MX(LVI_G1, LVI_G2, LVI_L0, LVI_L0));
+		d2l_write(tc->i2c, LV_MX1215, LV_MX(LVI_G3, LVI_G4, LVI_G5, LVI_B0));
+		d2l_write(tc->i2c, LV_MX1619, LV_MX(LVI_L0, LVI_L0, LVI_B1, LVI_B2));
+		d2l_write(tc->i2c, LV_MX2023, LV_MX(LVI_B3, LVI_B4, LVI_B5, LVI_L0));
+		d2l_write(tc->i2c, LV_MX2427, LV_MX(LVI_HS, LVI_VS, LVI_DE, LVI_L0));
+	}
+
+	d2l_write(tc->i2c, VFUEN, VFUEN_EN);
+
+	val = LVCFG_LVEN_BIT;
+	if (tc->lvds_link == DUAL_LINK) {
+		val |= TC358775_LVCFG_LVDLINK(1);
+		val |= TC358775_LVCFG_PCLKDIV(DIVIDE_BY_6);
+	} else {
+		val |= TC358775_LVCFG_PCLKDIV(DIVIDE_BY_3);
+	}
+	d2l_write(tc->i2c, LVCFG, val);
+}
+
+static enum drm_mode_status
+tc_mode_valid(struct drm_bridge *bridge,
+	      const struct drm_display_info *info,
+	      const struct drm_display_mode *mode)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+
+	/*
+	 * Maximum pixel clock speed 135MHz for single-link
+	 * 270MHz for dual-link
+	 */
+	if ((mode->clock > 135000 && tc->lvds_link == SINGLE_LINK) ||
+	    (mode->clock > 270000 && tc->lvds_link == DUAL_LINK))
+		return MODE_CLOCK_HIGH;
+
+	switch (info->bus_formats[0]) {
+	case MEDIA_BUS_FMT_RGB888_1X7X4_SPWG:
+	case MEDIA_BUS_FMT_RGB888_1X7X4_JEIDA:
+		/* RGB888 */
+		tc->bpc = 8;
+		break;
+	case MEDIA_BUS_FMT_RGB666_1X7X3_SPWG:
+		/* RGB666 */
+		tc->bpc = 6;
+		break;
+	default:
+		dev_warn(tc->dev,
+			 "unsupported LVDS bus format 0x%04x\n",
+			 info->bus_formats[0]);
+		return MODE_NOMODE;
+	}
+
+	return MODE_OK;
+}
+
+static int tc358775_parse_dt(struct device_node *np, struct tc_data *tc)
+{
+	struct device_node *endpoint;
+	struct device_node *parent;
+	struct device_node *remote;
+	struct property *prop;
+	int len = 0;
+
+	/*
+	 * To get the data-lanes of dsi, we need to access the dsi0_out of port1
+	 *  of dsi0 endpoint from bridge port0 of d2l_in
+	 */
+	endpoint = of_graph_get_endpoint_by_regs(tc->dev->of_node,
+						 TC358775_DSI_IN, -1);
+	if (endpoint) {
+		/* dsi0_out node */
+		parent = of_graph_get_remote_port_parent(endpoint);
+		of_node_put(endpoint);
+		if (parent) {
+			/* dsi0 port 1 */
+			endpoint = of_graph_get_endpoint_by_regs(parent, 1, -1);
+			of_node_put(parent);
+			if (endpoint) {
+				prop = of_find_property(endpoint, "data-lanes",
+							&len);
+				of_node_put(endpoint);
+				if (!prop) {
+					dev_err(tc->dev,
+						"failed to find data lane\n");
+					return -EPROBE_DEFER;
+				}
+			}
+		}
+	}
+
+	tc->num_dsi_lanes = len / sizeof(u32);
+
+	if (tc->num_dsi_lanes < 1 || tc->num_dsi_lanes > 4)
+		return -EINVAL;
+
+	tc->host_node = of_graph_get_remote_node(np, 0, 0);
+	if (!tc->host_node)
+		return -ENODEV;
+
+	of_node_put(tc->host_node);
+
+	tc->lvds_link = SINGLE_LINK;
+	endpoint = of_graph_get_endpoint_by_regs(tc->dev->of_node,
+						 TC358775_LVDS_OUT1, -1);
+	if (endpoint) {
+		remote = of_graph_get_remote_port_parent(endpoint);
+		of_node_put(endpoint);
+
+		if (remote) {
+			if (of_device_is_available(remote))
+				tc->lvds_link = DUAL_LINK;
+			of_node_put(remote);
+		}
+	}
+
+	dev_dbg(tc->dev, "no.of dsi lanes: %d\n", tc->num_dsi_lanes);
+	dev_dbg(tc->dev, "operating in %d-link mode\n",	tc->lvds_link);
+
+	return 0;
+}
+
+static int tc_bridge_attach(struct drm_bridge *bridge,
+			    enum drm_bridge_attach_flags flags)
+{
+	struct tc_data *tc = bridge_to_tc(bridge);
+	struct device *dev = &tc->i2c->dev;
+	struct mipi_dsi_host *host;
+	struct mipi_dsi_device *dsi;
+	int ret;
+
+	const struct mipi_dsi_device_info info = { .type = "tc358775",
+							.channel = 0,
+							.node = NULL,
+						};
+
+	host = of_find_mipi_dsi_host_by_node(tc->host_node);
+	if (!host) {
+		dev_err(dev, "failed to find dsi host\n");
+		return -EPROBE_DEFER;
+	}
+
+	dsi = mipi_dsi_device_register_full(host, &info);
+	if (IS_ERR(dsi)) {
+		dev_err(dev, "failed to create dsi device\n");
+		ret = PTR_ERR(dsi);
+		goto err_dsi_device;
+	}
+
+	tc->dsi = dsi;
+
+	dsi->lanes = tc->num_dsi_lanes;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO;
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		dev_err(dev, "failed to attach dsi to host\n");
+		goto err_dsi_attach;
+	}
+
+	/* Attach the panel-bridge to the dsi bridge */
+	return drm_bridge_attach(bridge->encoder, tc->panel_bridge,
+				 &tc->bridge, flags);
+err_dsi_attach:
+	mipi_dsi_device_unregister(dsi);
+err_dsi_device:
+	return ret;
+}
+
+static const struct drm_bridge_funcs tc_bridge_funcs = {
+	.attach = tc_bridge_attach,
+	.pre_enable = tc_bridge_pre_enable,
+	.enable = tc_bridge_enable,
+	.mode_valid = tc_mode_valid,
+	.post_disable = tc_bridge_post_disable,
+};
+
+static int tc_probe(struct i2c_client *client, const struct i2c_device_id *id)
+{
+	struct device *dev = &client->dev;
+	struct drm_panel *panel;
+	struct tc_data *tc;
+	int ret;
+
+	tc = devm_kzalloc(dev, sizeof(*tc), GFP_KERNEL);
+	if (!tc)
+		return -ENOMEM;
+
+	tc->dev = dev;
+	tc->i2c = client;
+
+	ret = drm_of_find_panel_or_bridge(dev->of_node, TC358775_LVDS_OUT0,
+					  0, &panel, NULL);
+	if (ret < 0)
+		return ret;
+	if (!panel)
+		return -ENODEV;
+
+	tc->panel_bridge = devm_drm_panel_bridge_add(dev, panel);
+	if (IS_ERR(tc->panel_bridge))
+		return PTR_ERR(tc->panel_bridge);
+
+	ret = tc358775_parse_dt(dev->of_node, tc);
+	if (ret)
+		return ret;
+
+	tc->vddio = devm_regulator_get(dev, "vddio-supply");
+	if (IS_ERR(tc->vddio)) {
+		ret = PTR_ERR(tc->vddio);
+		dev_err(dev, "vddio-supply not found\n");
+		return ret;
+	}
+
+	tc->vdd = devm_regulator_get(dev, "vdd-supply");
+	if (IS_ERR(tc->vdd)) {
+		ret = PTR_ERR(tc->vdd);
+		dev_err(dev, "vdd-supply not found\n");
+		return ret;
+	}
+
+	tc->stby_gpio = devm_gpiod_get(dev, "stby", GPIOD_OUT_HIGH);
+	if (IS_ERR(tc->stby_gpio)) {
+		ret = PTR_ERR(tc->stby_gpio);
+		dev_err(dev, "cannot get stby-gpio %d\n", ret);
+		return ret;
+	}
+
+	tc->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(tc->reset_gpio)) {
+		ret = PTR_ERR(tc->reset_gpio);
+		dev_err(dev, "cannot get reset-gpios %d\n", ret);
+		return ret;
+	}
+
+	tc->bridge.funcs = &tc_bridge_funcs;
+	tc->bridge.of_node = dev->of_node;
+	drm_bridge_add(&tc->bridge);
+
+	i2c_set_clientdata(client, tc);
+
+	return 0;
+}
+
+static int tc_remove(struct i2c_client *client)
+{
+	struct tc_data *tc = i2c_get_clientdata(client);
+
+	drm_bridge_remove(&tc->bridge);
+
+	return 0;
+}
+
+static const struct i2c_device_id tc358775_i2c_ids[] = {
+	{ "tc358775", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, tc358775_i2c_ids);
+
+static const struct of_device_id tc358775_of_ids[] = {
+	{ .compatible = "toshiba,tc358775", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, tc358775_of_ids);
+
+static struct i2c_driver tc358775_driver = {
+	.driver = {
+		.name = "tc358775",
+		.of_match_table = tc358775_of_ids,
+	},
+	.id_table = tc358775_i2c_ids,
+	.probe = tc_probe,
+	.remove	= tc_remove,
+};
+module_i2c_driver(tc358775_driver);
+
+MODULE_AUTHOR("Vinay Simha BN <simhavcs@gmail.com>");
+MODULE_DESCRIPTION("TC358775 DSI/LVDS bridge driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 5b6e19ecbc84..ecdf9b01340f 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -394,9 +394,6 @@ static int ti_sn_bridge_attach(struct drm_bridge *bridge,
 	}
 	pdata->dsi = dsi;
 
-	/* attach panel to bridge */
-	drm_panel_attach(pdata->panel, &pdata->connector);
-
 	return 0;
 
 err_dsi_attach:
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 9e1ad493e689..f9170b4b22e7 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1115,9 +1115,7 @@ disable_outputs(struct drm_device *dev, struct drm_atomic_state *old_state)
  * @old_state: atomic state object with old state structures
  *
  * This function updates all the various legacy modeset state pointers in
- * connectors, encoders and CRTCs. It also updates the timestamping constants
- * used for precise vblank timestamps by calling
- * drm_calc_timestamping_constants().
+ * connectors, encoders and CRTCs.
  *
  * Drivers can use this for building their own atomic commit if they don't have
  * a pure helper-based modeset implementation.
@@ -1186,13 +1184,30 @@ drm_atomic_helper_update_legacy_modeset_state(struct drm_device *dev,
 			crtc->x = new_plane_state->src_x >> 16;
 			crtc->y = new_plane_state->src_y >> 16;
 		}
+	}
+}
+EXPORT_SYMBOL(drm_atomic_helper_update_legacy_modeset_state);
 
+/**
+ * drm_atomic_helper_calc_timestamping_constants - update vblank timestamping constants
+ * @state: atomic state object
+ *
+ * Updates the timestamping constants used for precise vblank timestamps
+ * by calling drm_calc_timestamping_constants() for all enabled crtcs in @state.
+ */
+void drm_atomic_helper_calc_timestamping_constants(struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *new_crtc_state;
+	struct drm_crtc *crtc;
+	int i;
+
+	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
 		if (new_crtc_state->enable)
 			drm_calc_timestamping_constants(crtc,
 							&new_crtc_state->adjusted_mode);
 	}
 }
-EXPORT_SYMBOL(drm_atomic_helper_update_legacy_modeset_state);
+EXPORT_SYMBOL(drm_atomic_helper_calc_timestamping_constants);
 
 static void
 crtc_set_mode(struct drm_device *dev, struct drm_atomic_state *old_state)
@@ -1276,6 +1291,7 @@ void drm_atomic_helper_commit_modeset_disables(struct drm_device *dev,
 	disable_outputs(dev, old_state);
 
 	drm_atomic_helper_update_legacy_modeset_state(dev, old_state);
+	drm_atomic_helper_calc_timestamping_constants(old_state);
 
 	crtc_set_mode(dev, old_state);
 }
diff --git a/drivers/gpu/drm/drm_bridge_connector.c b/drivers/gpu/drm/drm_bridge_connector.c
index c6994fe673f3..a58cbde59c34 100644
--- a/drivers/gpu/drm/drm_bridge_connector.c
+++ b/drivers/gpu/drm/drm_bridge_connector.c
@@ -187,6 +187,7 @@ drm_bridge_connector_detect(struct drm_connector *connector, bool force)
 		case DRM_MODE_CONNECTOR_DPI:
 		case DRM_MODE_CONNECTOR_LVDS:
 		case DRM_MODE_CONNECTOR_DSI:
+		case DRM_MODE_CONNECTOR_eDP:
 			status = connector_status_connected;
 			break;
 		default:
diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c
index 03e01b000f7a..0fe3c496002a 100644
--- a/drivers/gpu/drm/drm_cache.c
+++ b/drivers/gpu/drm/drm_cache.c
@@ -127,7 +127,7 @@ drm_clflush_sg(struct sg_table *st)
 		struct sg_page_iter sg_iter;
 
 		mb(); /*CLFLUSH is ordered only by using memory barriers*/
-		for_each_sg_page(st->sgl, &sg_iter, st->nents, 0)
+		for_each_sgtable_page(st, &sg_iter, 0)
 			drm_clflush_page(sg_page_iter_page(&sg_iter));
 		mb(); /*Make sure that all cache line entry is flushed*/
 
diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index 00e40a26a800..717c4e7271b0 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -850,7 +850,7 @@ static const struct drm_prop_enum_list drm_dvi_i_select_enum_list[] = {
 DRM_ENUM_NAME_FN(drm_get_dvi_i_select_name, drm_dvi_i_select_enum_list)
 
 static const struct drm_prop_enum_list drm_dvi_i_subconnector_enum_list[] = {
-	{ DRM_MODE_SUBCONNECTOR_Unknown,   "Unknown"   }, /* DVI-I and TV-out */
+	{ DRM_MODE_SUBCONNECTOR_Unknown,   "Unknown"   }, /* DVI-I, TV-out and DP */
 	{ DRM_MODE_SUBCONNECTOR_DVID,      "DVI-D"     }, /* DVI-I  */
 	{ DRM_MODE_SUBCONNECTOR_DVIA,      "DVI-A"     }, /* DVI-I  */
 };
@@ -867,7 +867,7 @@ static const struct drm_prop_enum_list drm_tv_select_enum_list[] = {
 DRM_ENUM_NAME_FN(drm_get_tv_select_name, drm_tv_select_enum_list)
 
 static const struct drm_prop_enum_list drm_tv_subconnector_enum_list[] = {
-	{ DRM_MODE_SUBCONNECTOR_Unknown,   "Unknown"   }, /* DVI-I and TV-out */
+	{ DRM_MODE_SUBCONNECTOR_Unknown,   "Unknown"   }, /* DVI-I, TV-out and DP */
 	{ DRM_MODE_SUBCONNECTOR_Composite, "Composite" }, /* TV-out */
 	{ DRM_MODE_SUBCONNECTOR_SVIDEO,    "SVIDEO"    }, /* TV-out */
 	{ DRM_MODE_SUBCONNECTOR_Component, "Component" }, /* TV-out */
@@ -876,6 +876,19 @@ static const struct drm_prop_enum_list drm_tv_subconnector_enum_list[] = {
 DRM_ENUM_NAME_FN(drm_get_tv_subconnector_name,
 		 drm_tv_subconnector_enum_list)
 
+static const struct drm_prop_enum_list drm_dp_subconnector_enum_list[] = {
+	{ DRM_MODE_SUBCONNECTOR_Unknown,     "Unknown"   }, /* DVI-I, TV-out and DP */
+	{ DRM_MODE_SUBCONNECTOR_VGA,	     "VGA"       }, /* DP */
+	{ DRM_MODE_SUBCONNECTOR_DVID,	     "DVI-D"     }, /* DP */
+	{ DRM_MODE_SUBCONNECTOR_HDMIA,	     "HDMI"      }, /* DP */
+	{ DRM_MODE_SUBCONNECTOR_DisplayPort, "DP"        }, /* DP */
+	{ DRM_MODE_SUBCONNECTOR_Wireless,    "Wireless"  }, /* DP */
+	{ DRM_MODE_SUBCONNECTOR_Native,	     "Native"    }, /* DP */
+};
+
+DRM_ENUM_NAME_FN(drm_get_dp_subconnector_name,
+		 drm_dp_subconnector_enum_list)
+
 static const struct drm_prop_enum_list hdmi_colorspaces[] = {
 	/* For Default case, driver will set the colorspace */
 	{ DRM_MODE_COLORIMETRY_DEFAULT, "Default" },
@@ -1217,6 +1230,14 @@ static const struct drm_prop_enum_list dp_colorspaces[] = {
  *	can also expose this property to external outputs, in which case they
  *	must support "None", which should be the default (since external screens
  *	have a built-in scaler).
+ *
+ * subconnector:
+ *	This property is used by DVI-I, TVout and DisplayPort to indicate different
+ *	connector subtypes. Enum values more or less match with those from main
+ *	connector types.
+ *	For DVI-I and TVout there is also a matching property "select subconnector"
+ *	allowing to switch between signal types.
+ *	DP subconnector corresponds to a downstream port.
  */
 
 int drm_connector_create_standard_properties(struct drm_device *dev)
@@ -1306,6 +1327,30 @@ int drm_mode_create_dvi_i_properties(struct drm_device *dev)
 EXPORT_SYMBOL(drm_mode_create_dvi_i_properties);
 
 /**
+ * drm_connector_attach_dp_subconnector_property - create subconnector property for DP
+ * @connector: drm_connector to attach property
+ *
+ * Called by a driver when DP connector is created.
+ */
+void drm_connector_attach_dp_subconnector_property(struct drm_connector *connector)
+{
+	struct drm_mode_config *mode_config = &connector->dev->mode_config;
+
+	if (!mode_config->dp_subconnector_property)
+		mode_config->dp_subconnector_property =
+			drm_property_create_enum(connector->dev,
+				DRM_MODE_PROP_IMMUTABLE,
+				"subconnector",
+				drm_dp_subconnector_enum_list,
+				ARRAY_SIZE(drm_dp_subconnector_enum_list));
+
+	drm_object_attach_property(&connector->base,
+				   mode_config->dp_subconnector_property,
+				   DRM_MODE_SUBCONNECTOR_Unknown);
+}
+EXPORT_SYMBOL(drm_connector_attach_dp_subconnector_property);
+
+/**
  * DOC: HDMI connector properties
  *
  * content type (HDMI specific):
@@ -2244,7 +2289,7 @@ static struct drm_encoder *drm_connector_get_encoder(struct drm_connector *conne
 
 static bool
 drm_mode_expose_to_userspace(const struct drm_display_mode *mode,
-			     const struct list_head *export_list,
+			     const struct list_head *modes,
 			     const struct drm_file *file_priv)
 {
 	/*
@@ -2260,15 +2305,17 @@ drm_mode_expose_to_userspace(const struct drm_display_mode *mode,
 	 * while preparing the list of user-modes.
 	 */
 	if (!file_priv->aspect_ratio_allowed) {
-		struct drm_display_mode *mode_itr;
+		const struct drm_display_mode *mode_itr;
 
-		list_for_each_entry(mode_itr, export_list, export_head)
-			if (drm_mode_match(mode_itr, mode,
+		list_for_each_entry(mode_itr, modes, head) {
+			if (mode_itr->expose_to_userspace &&
+			    drm_mode_match(mode_itr, mode,
 					   DRM_MODE_MATCH_TIMINGS |
 					   DRM_MODE_MATCH_CLOCK |
 					   DRM_MODE_MATCH_FLAGS |
 					   DRM_MODE_MATCH_3D_FLAGS))
 				return false;
+		}
 	}
 
 	return true;
@@ -2288,7 +2335,6 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	struct drm_mode_modeinfo u_mode;
 	struct drm_mode_modeinfo __user *mode_ptr;
 	uint32_t __user *encoder_ptr;
-	LIST_HEAD(export_list);
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EOPNOTSUPP;
@@ -2332,25 +2378,30 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	out_resp->connection = connector->status;
 
 	/* delayed so we get modes regardless of pre-fill_modes state */
-	list_for_each_entry(mode, &connector->modes, head)
-		if (drm_mode_expose_to_userspace(mode, &export_list,
+	list_for_each_entry(mode, &connector->modes, head) {
+		WARN_ON(mode->expose_to_userspace);
+
+		if (drm_mode_expose_to_userspace(mode, &connector->modes,
 						 file_priv)) {
-			list_add_tail(&mode->export_head, &export_list);
+			mode->expose_to_userspace = true;
 			mode_count++;
 		}
+	}
 
 	/*
 	 * This ioctl is called twice, once to determine how much space is
 	 * needed, and the 2nd time to fill it.
-	 * The modes that need to be exposed to the user are maintained in the
-	 * 'export_list'. When the ioctl is called first time to determine the,
-	 * space, the export_list gets filled, to find the no.of modes. In the
-	 * 2nd time, the user modes are filled, one by one from the export_list.
 	 */
 	if ((out_resp->count_modes >= mode_count) && mode_count) {
 		copied = 0;
 		mode_ptr = (struct drm_mode_modeinfo __user *)(unsigned long)out_resp->modes_ptr;
-		list_for_each_entry(mode, &export_list, export_head) {
+		list_for_each_entry(mode, &connector->modes, head) {
+			if (!mode->expose_to_userspace)
+				continue;
+
+			/* Clear the tag for the next time around */
+			mode->expose_to_userspace = false;
+
 			drm_mode_convert_to_umode(&u_mode, mode);
 			/*
 			 * Reset aspect ratio flags of user-mode, if modes with
@@ -2361,13 +2412,26 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 			if (copy_to_user(mode_ptr + copied,
 					 &u_mode, sizeof(u_mode))) {
 				ret = -EFAULT;
+
+				/*
+				 * Clear the tag for the rest of
+				 * the modes for the next time around.
+				 */
+				list_for_each_entry_continue(mode, &connector->modes, head)
+					mode->expose_to_userspace = false;
+
 				mutex_unlock(&dev->mode_config.mutex);
 
 				goto out;
 			}
 			copied++;
 		}
+	} else {
+		/* Clear the tag for the next time around */
+		list_for_each_entry(mode, &connector->modes, head)
+			mode->expose_to_userspace = false;
 	}
+
 	out_resp->count_modes = mode_count;
 	mutex_unlock(&dev->mode_config.mutex);
 
diff --git a/drivers/gpu/drm/drm_debugfs_crc.c b/drivers/gpu/drm/drm_debugfs_crc.c
index 5d67a41f7c3a..3dd70d813f69 100644
--- a/drivers/gpu/drm/drm_debugfs_crc.c
+++ b/drivers/gpu/drm/drm_debugfs_crc.c
@@ -144,8 +144,10 @@ static ssize_t crc_control_write(struct file *file, const char __user *ubuf,
 		source[len - 1] = '\0';
 
 	ret = crtc->funcs->verify_crc_source(crtc, source, &values_cnt);
-	if (ret)
+	if (ret) {
+		kfree(source);
 		return ret;
+	}
 
 	spin_lock_irq(&crc->lock);
 
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 092c8c985911..90807a6b415c 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -363,6 +363,66 @@ int drm_dp_dpcd_read_link_status(struct drm_dp_aux *aux,
 }
 EXPORT_SYMBOL(drm_dp_dpcd_read_link_status);
 
+static bool is_edid_digital_input_dp(const struct edid *edid)
+{
+	return edid && edid->revision >= 4 &&
+		edid->input & DRM_EDID_INPUT_DIGITAL &&
+		(edid->input & DRM_EDID_DIGITAL_TYPE_MASK) == DRM_EDID_DIGITAL_TYPE_DP;
+}
+
+/**
+ * drm_dp_downstream_is_type() - is the downstream facing port of certain type?
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: port capabilities
+ *
+ * Caveat: Only works with DPCD 1.1+ port caps.
+ *
+ * Returns: whether the downstream facing port matches the type.
+ */
+bool drm_dp_downstream_is_type(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+			       const u8 port_cap[4], u8 type)
+{
+	return drm_dp_is_branch(dpcd) &&
+		dpcd[DP_DPCD_REV] >= 0x11 &&
+		(port_cap[0] & DP_DS_PORT_TYPE_MASK) == type;
+}
+EXPORT_SYMBOL(drm_dp_downstream_is_type);
+
+/**
+ * drm_dp_downstream_is_tmds() - is the downstream facing port TMDS?
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: port capabilities
+ * @edid: EDID
+ *
+ * Returns: whether the downstream facing port is TMDS (HDMI/DVI).
+ */
+bool drm_dp_downstream_is_tmds(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+			       const u8 port_cap[4],
+			       const struct edid *edid)
+{
+	if (dpcd[DP_DPCD_REV] < 0x11) {
+		switch (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_TYPE_MASK) {
+		case DP_DWN_STRM_PORT_TYPE_TMDS:
+			return true;
+		default:
+			return false;
+		}
+	}
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_DP_DUALMODE:
+		if (is_edid_digital_input_dp(edid))
+			return false;
+		fallthrough;
+	case DP_DS_PORT_TYPE_DVI:
+	case DP_DS_PORT_TYPE_HDMI:
+		return true;
+	default:
+		return false;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_is_tmds);
+
 /**
  * drm_dp_send_real_edid_checksum() - send back real edid checksum value
  * @aux: DisplayPort AUX channel
@@ -423,66 +483,313 @@ bool drm_dp_send_real_edid_checksum(struct drm_dp_aux *aux,
 }
 EXPORT_SYMBOL(drm_dp_send_real_edid_checksum);
 
+static u8 drm_dp_downstream_port_count(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	u8 port_count = dpcd[DP_DOWN_STREAM_PORT_COUNT] & DP_PORT_COUNT_MASK;
+
+	if (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE && port_count > 4)
+		port_count = 4;
+
+	return port_count;
+}
+
+static int drm_dp_read_extended_dpcd_caps(struct drm_dp_aux *aux,
+					  u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	u8 dpcd_ext[6];
+	int ret;
+
+	/*
+	 * Prior to DP1.3 the bit represented by
+	 * DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT was reserved.
+	 * If it is set DP_DPCD_REV at 0000h could be at a value less than
+	 * the true capability of the panel. The only way to check is to
+	 * then compare 0000h and 2200h.
+	 */
+	if (!(dpcd[DP_TRAINING_AUX_RD_INTERVAL] &
+	      DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT))
+		return 0;
+
+	ret = drm_dp_dpcd_read(aux, DP_DP13_DPCD_REV, &dpcd_ext,
+			       sizeof(dpcd_ext));
+	if (ret < 0)
+		return ret;
+	if (ret != sizeof(dpcd_ext))
+		return -EIO;
+
+	if (dpcd[DP_DPCD_REV] > dpcd_ext[DP_DPCD_REV]) {
+		DRM_DEBUG_KMS("%s: Extended DPCD rev less than base DPCD rev (%d > %d)\n",
+			      aux->name, dpcd[DP_DPCD_REV],
+			      dpcd_ext[DP_DPCD_REV]);
+		return 0;
+	}
+
+	if (!memcmp(dpcd, dpcd_ext, sizeof(dpcd_ext)))
+		return 0;
+
+	DRM_DEBUG_KMS("%s: Base DPCD: %*ph\n",
+		      aux->name, DP_RECEIVER_CAP_SIZE, dpcd);
+
+	memcpy(dpcd, dpcd_ext, sizeof(dpcd_ext));
+
+	return 0;
+}
+
+/**
+ * drm_dp_read_dpcd_caps() - read DPCD caps and extended DPCD caps if
+ * available
+ * @aux: DisplayPort AUX channel
+ * @dpcd: Buffer to store the resulting DPCD in
+ *
+ * Attempts to read the base DPCD caps for @aux. Additionally, this function
+ * checks for and reads the extended DPRX caps (%DP_DP13_DPCD_REV) if
+ * present.
+ *
+ * Returns: %0 if the DPCD was read successfully, negative error code
+ * otherwise.
+ */
+int drm_dp_read_dpcd_caps(struct drm_dp_aux *aux,
+			  u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	int ret;
+
+	ret = drm_dp_dpcd_read(aux, DP_DPCD_REV, dpcd, DP_RECEIVER_CAP_SIZE);
+	if (ret < 0)
+		return ret;
+	if (ret != DP_RECEIVER_CAP_SIZE || dpcd[DP_DPCD_REV] == 0)
+		return -EIO;
+
+	ret = drm_dp_read_extended_dpcd_caps(aux, dpcd);
+	if (ret < 0)
+		return ret;
+
+	DRM_DEBUG_KMS("%s: DPCD: %*ph\n",
+		      aux->name, DP_RECEIVER_CAP_SIZE, dpcd);
+
+	return ret;
+}
+EXPORT_SYMBOL(drm_dp_read_dpcd_caps);
+
 /**
- * drm_dp_downstream_max_clock() - extract branch device max
- *                                 pixel rate for legacy VGA
- *                                 converter or max TMDS clock
- *                                 rate for others
+ * drm_dp_read_downstream_info() - read DPCD downstream port info if available
+ * @aux: DisplayPort AUX channel
+ * @dpcd: A cached copy of the port's DPCD
+ * @downstream_ports: buffer to store the downstream port info in
+ *
+ * See also:
+ * drm_dp_downstream_max_clock()
+ * drm_dp_downstream_max_bpc()
+ *
+ * Returns: 0 if either the downstream port info was read successfully or
+ * there was no downstream info to read, or a negative error code otherwise.
+ */
+int drm_dp_read_downstream_info(struct drm_dp_aux *aux,
+				const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				u8 downstream_ports[DP_MAX_DOWNSTREAM_PORTS])
+{
+	int ret;
+	u8 len;
+
+	memset(downstream_ports, 0, DP_MAX_DOWNSTREAM_PORTS);
+
+	/* No downstream info to read */
+	if (!drm_dp_is_branch(dpcd) ||
+	    dpcd[DP_DPCD_REV] < DP_DPCD_REV_10 ||
+	    !(dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_PRESENT))
+		return 0;
+
+	len = drm_dp_downstream_port_count(dpcd);
+	if (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE)
+		len *= 4;
+
+	ret = drm_dp_dpcd_read(aux, DP_DOWNSTREAM_PORT_0, downstream_ports, len);
+	if (ret < 0)
+		return ret;
+	if (ret != len)
+		return -EIO;
+
+	DRM_DEBUG_KMS("%s: DPCD DFP: %*ph\n",
+		      aux->name, len, downstream_ports);
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dp_read_downstream_info);
+
+/**
+ * drm_dp_downstream_max_dotclock() - extract downstream facing port max dot clock
  * @dpcd: DisplayPort configuration data
  * @port_cap: port capabilities
  *
- * Returns max clock in kHz on success or 0 if max clock not defined
+ * Returns: Downstream facing port max dot clock in kHz on success,
+ * or 0 if max clock not defined
  */
-int drm_dp_downstream_max_clock(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
-				const u8 port_cap[4])
+int drm_dp_downstream_max_dotclock(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				   const u8 port_cap[4])
 {
-	int type = port_cap[0] & DP_DS_PORT_TYPE_MASK;
-	bool detailed_cap_info = dpcd[DP_DOWNSTREAMPORT_PRESENT] &
-		DP_DETAILED_CAP_INFO_AVAILABLE;
+	if (!drm_dp_is_branch(dpcd))
+		return 0;
 
-	if (!detailed_cap_info)
+	if (dpcd[DP_DPCD_REV] < 0x11)
 		return 0;
 
-	switch (type) {
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
 	case DP_DS_PORT_TYPE_VGA:
-		return port_cap[1] * 8 * 1000;
-	case DP_DS_PORT_TYPE_DVI:
-	case DP_DS_PORT_TYPE_HDMI:
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return 0;
+		return port_cap[1] * 8000;
+	default:
+		return 0;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_max_dotclock);
+
+/**
+ * drm_dp_downstream_max_tmds_clock() - extract downstream facing port max TMDS clock
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: port capabilities
+ * @edid: EDID
+ *
+ * Returns: HDMI/DVI downstream facing port max TMDS clock in kHz on success,
+ * or 0 if max TMDS clock not defined
+ */
+int drm_dp_downstream_max_tmds_clock(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				     const u8 port_cap[4],
+				     const struct edid *edid)
+{
+	if (!drm_dp_is_branch(dpcd))
+		return 0;
+
+	if (dpcd[DP_DPCD_REV] < 0x11) {
+		switch (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_TYPE_MASK) {
+		case DP_DWN_STRM_PORT_TYPE_TMDS:
+			return 165000;
+		default:
+			return 0;
+		}
+	}
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
 	case DP_DS_PORT_TYPE_DP_DUALMODE:
+		if (is_edid_digital_input_dp(edid))
+			return 0;
+		/*
+		 * It's left up to the driver to check the
+		 * DP dual mode adapter's max TMDS clock.
+		 *
+		 * Unfortunatley it looks like branch devices
+		 * may not fordward that the DP dual mode i2c
+		 * access so we just usually get i2c nak :(
+		 */
+		fallthrough;
+	case DP_DS_PORT_TYPE_HDMI:
+		 /*
+		  * We should perhaps assume 165 MHz when detailed cap
+		  * info is not available. But looks like many typical
+		  * branch devices fall into that category and so we'd
+		  * probably end up with users complaining that they can't
+		  * get high resolution modes with their favorite dongle.
+		  *
+		  * So let's limit to 300 MHz instead since DPCD 1.4
+		  * HDMI 2.0 DFPs are required to have the detailed cap
+		  * info. So it's more likely we're dealing with a HDMI 1.4
+		  * compatible* device here.
+		  */
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return 300000;
+		return port_cap[1] * 2500;
+	case DP_DS_PORT_TYPE_DVI:
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return 165000;
+		/* FIXME what to do about DVI dual link? */
 		return port_cap[1] * 2500;
 	default:
 		return 0;
 	}
 }
-EXPORT_SYMBOL(drm_dp_downstream_max_clock);
+EXPORT_SYMBOL(drm_dp_downstream_max_tmds_clock);
 
 /**
- * drm_dp_downstream_max_bpc() - extract branch device max
- *                               bits per component
+ * drm_dp_downstream_min_tmds_clock() - extract downstream facing port min TMDS clock
  * @dpcd: DisplayPort configuration data
  * @port_cap: port capabilities
+ * @edid: EDID
  *
- * Returns max bpc on success or 0 if max bpc not defined
+ * Returns: HDMI/DVI downstream facing port min TMDS clock in kHz on success,
+ * or 0 if max TMDS clock not defined
  */
-int drm_dp_downstream_max_bpc(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
-			      const u8 port_cap[4])
+int drm_dp_downstream_min_tmds_clock(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				     const u8 port_cap[4],
+				     const struct edid *edid)
 {
-	int type = port_cap[0] & DP_DS_PORT_TYPE_MASK;
-	bool detailed_cap_info = dpcd[DP_DOWNSTREAMPORT_PRESENT] &
-		DP_DETAILED_CAP_INFO_AVAILABLE;
-	int bpc;
-
-	if (!detailed_cap_info)
+	if (!drm_dp_is_branch(dpcd))
 		return 0;
 
-	switch (type) {
-	case DP_DS_PORT_TYPE_VGA:
+	if (dpcd[DP_DPCD_REV] < 0x11) {
+		switch (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_TYPE_MASK) {
+		case DP_DWN_STRM_PORT_TYPE_TMDS:
+			return 25000;
+		default:
+			return 0;
+		}
+	}
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_DP_DUALMODE:
+		if (is_edid_digital_input_dp(edid))
+			return 0;
+		fallthrough;
 	case DP_DS_PORT_TYPE_DVI:
 	case DP_DS_PORT_TYPE_HDMI:
+		/*
+		 * Unclear whether the protocol converter could
+		 * utilize pixel replication. Assume it won't.
+		 */
+		return 25000;
+	default:
+		return 0;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_min_tmds_clock);
+
+/**
+ * drm_dp_downstream_max_bpc() - extract downstream facing port max
+ *                               bits per component
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: downstream facing port capabilities
+ * @edid: EDID
+ *
+ * Returns: Max bpc on success or 0 if max bpc not defined
+ */
+int drm_dp_downstream_max_bpc(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+			      const u8 port_cap[4],
+			      const struct edid *edid)
+{
+	if (!drm_dp_is_branch(dpcd))
+		return 0;
+
+	if (dpcd[DP_DPCD_REV] < 0x11) {
+		switch (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_TYPE_MASK) {
+		case DP_DWN_STRM_PORT_TYPE_DP:
+			return 0;
+		default:
+			return 8;
+		}
+	}
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_DP:
+		return 0;
 	case DP_DS_PORT_TYPE_DP_DUALMODE:
-		bpc = port_cap[2] & DP_DS_MAX_BPC_MASK;
+		if (is_edid_digital_input_dp(edid))
+			return 0;
+		fallthrough;
+	case DP_DS_PORT_TYPE_HDMI:
+	case DP_DS_PORT_TYPE_DVI:
+	case DP_DS_PORT_TYPE_VGA:
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return 8;
 
-		switch (bpc) {
+		switch (port_cap[2] & DP_DS_MAX_BPC_MASK) {
 		case DP_DS_8BPC:
 			return 8;
 		case DP_DS_10BPC:
@@ -491,15 +798,131 @@ int drm_dp_downstream_max_bpc(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
 			return 12;
 		case DP_DS_16BPC:
 			return 16;
+		default:
+			return 8;
 		}
-		fallthrough;
+		break;
 	default:
-		return 0;
+		return 8;
 	}
 }
 EXPORT_SYMBOL(drm_dp_downstream_max_bpc);
 
 /**
+ * drm_dp_downstream_420_passthrough() - determine downstream facing port
+ *                                       YCbCr 4:2:0 pass-through capability
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: downstream facing port capabilities
+ *
+ * Returns: whether the downstream facing port can pass through YCbCr 4:2:0
+ */
+bool drm_dp_downstream_420_passthrough(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				       const u8 port_cap[4])
+{
+	if (!drm_dp_is_branch(dpcd))
+		return false;
+
+	if (dpcd[DP_DPCD_REV] < 0x13)
+		return false;
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_DP:
+		return true;
+	case DP_DS_PORT_TYPE_HDMI:
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return false;
+
+		return port_cap[3] & DP_DS_HDMI_YCBCR420_PASS_THROUGH;
+	default:
+		return false;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_420_passthrough);
+
+/**
+ * drm_dp_downstream_444_to_420_conversion() - determine downstream facing port
+ *                                             YCbCr 4:4:4->4:2:0 conversion capability
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: downstream facing port capabilities
+ *
+ * Returns: whether the downstream facing port can convert YCbCr 4:4:4 to 4:2:0
+ */
+bool drm_dp_downstream_444_to_420_conversion(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+					     const u8 port_cap[4])
+{
+	if (!drm_dp_is_branch(dpcd))
+		return false;
+
+	if (dpcd[DP_DPCD_REV] < 0x13)
+		return false;
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_HDMI:
+		if ((dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE) == 0)
+			return false;
+
+		return port_cap[3] & DP_DS_HDMI_YCBCR444_TO_420_CONV;
+	default:
+		return false;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_444_to_420_conversion);
+
+/**
+ * drm_dp_downstream_mode() - return a mode for downstream facing port
+ * @dpcd: DisplayPort configuration data
+ * @port_cap: port capabilities
+ *
+ * Provides a suitable mode for downstream facing ports without EDID.
+ *
+ * Returns: A new drm_display_mode on success or NULL on failure
+ */
+struct drm_display_mode *
+drm_dp_downstream_mode(struct drm_device *dev,
+		       const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+		       const u8 port_cap[4])
+
+{
+	u8 vic;
+
+	if (!drm_dp_is_branch(dpcd))
+		return NULL;
+
+	if (dpcd[DP_DPCD_REV] < 0x11)
+		return NULL;
+
+	switch (port_cap[0] & DP_DS_PORT_TYPE_MASK) {
+	case DP_DS_PORT_TYPE_NON_EDID:
+		switch (port_cap[0] & DP_DS_NON_EDID_MASK) {
+		case DP_DS_NON_EDID_720x480i_60:
+			vic = 6;
+			break;
+		case DP_DS_NON_EDID_720x480i_50:
+			vic = 21;
+			break;
+		case DP_DS_NON_EDID_1920x1080i_60:
+			vic = 5;
+			break;
+		case DP_DS_NON_EDID_1920x1080i_50:
+			vic = 20;
+			break;
+		case DP_DS_NON_EDID_1280x720_60:
+			vic = 4;
+			break;
+		case DP_DS_NON_EDID_1280x720_50:
+			vic = 19;
+			break;
+		default:
+			return NULL;
+		}
+		return drm_display_mode_from_cea_vic(dev, vic);
+	default:
+		return NULL;
+	}
+}
+EXPORT_SYMBOL(drm_dp_downstream_mode);
+
+/**
  * drm_dp_downstream_id() - identify branch device
  * @aux: DisplayPort AUX channel
  * @id: DisplayPort branch device id
@@ -517,12 +940,15 @@ EXPORT_SYMBOL(drm_dp_downstream_id);
  * @m: pointer for debugfs file
  * @dpcd: DisplayPort configuration data
  * @port_cap: port capabilities
+ * @edid: EDID
  * @aux: DisplayPort AUX channel
  *
  */
 void drm_dp_downstream_debug(struct seq_file *m,
 			     const u8 dpcd[DP_RECEIVER_CAP_SIZE],
-			     const u8 port_cap[4], struct drm_dp_aux *aux)
+			     const u8 port_cap[4],
+			     const struct edid *edid,
+			     struct drm_dp_aux *aux)
 {
 	bool detailed_cap_info = dpcd[DP_DOWNSTREAMPORT_PRESENT] &
 				 DP_DETAILED_CAP_INFO_AVAILABLE;
@@ -580,16 +1006,19 @@ void drm_dp_downstream_debug(struct seq_file *m,
 		seq_printf(m, "\t\tSW: %d.%d\n", rev[0], rev[1]);
 
 	if (detailed_cap_info) {
-		clk = drm_dp_downstream_max_clock(dpcd, port_cap);
+		clk = drm_dp_downstream_max_dotclock(dpcd, port_cap);
+		if (clk > 0)
+			seq_printf(m, "\t\tMax dot clock: %d kHz\n", clk);
 
-		if (clk > 0) {
-			if (type == DP_DS_PORT_TYPE_VGA)
-				seq_printf(m, "\t\tMax dot clock: %d kHz\n", clk);
-			else
-				seq_printf(m, "\t\tMax TMDS clock: %d kHz\n", clk);
-		}
+		clk = drm_dp_downstream_max_tmds_clock(dpcd, port_cap, edid);
+		if (clk > 0)
+			seq_printf(m, "\t\tMax TMDS clock: %d kHz\n", clk);
 
-		bpc = drm_dp_downstream_max_bpc(dpcd, port_cap);
+		clk = drm_dp_downstream_min_tmds_clock(dpcd, port_cap, edid);
+		if (clk > 0)
+			seq_printf(m, "\t\tMin TMDS clock: %d kHz\n", clk);
+
+		bpc = drm_dp_downstream_max_bpc(dpcd, port_cap, edid);
 
 		if (bpc > 0)
 			seq_printf(m, "\t\tMax bpc: %d\n", bpc);
@@ -597,6 +1026,125 @@ void drm_dp_downstream_debug(struct seq_file *m,
 }
 EXPORT_SYMBOL(drm_dp_downstream_debug);
 
+/**
+ * drm_dp_subconnector_type() - get DP branch device type
+ *
+ */
+enum drm_mode_subconnector
+drm_dp_subconnector_type(const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+			 const u8 port_cap[4])
+{
+	int type;
+	if (!drm_dp_is_branch(dpcd))
+		return DRM_MODE_SUBCONNECTOR_Native;
+	/* DP 1.0 approach */
+	if (dpcd[DP_DPCD_REV] == DP_DPCD_REV_10) {
+		type = dpcd[DP_DOWNSTREAMPORT_PRESENT] &
+		       DP_DWN_STRM_PORT_TYPE_MASK;
+
+		switch (type) {
+		case DP_DWN_STRM_PORT_TYPE_TMDS:
+			/* Can be HDMI or DVI-D, DVI-D is a safer option */
+			return DRM_MODE_SUBCONNECTOR_DVID;
+		case DP_DWN_STRM_PORT_TYPE_ANALOG:
+			/* Can be VGA or DVI-A, VGA is more popular */
+			return DRM_MODE_SUBCONNECTOR_VGA;
+		case DP_DWN_STRM_PORT_TYPE_DP:
+			return DRM_MODE_SUBCONNECTOR_DisplayPort;
+		case DP_DWN_STRM_PORT_TYPE_OTHER:
+		default:
+			return DRM_MODE_SUBCONNECTOR_Unknown;
+		}
+	}
+	type = port_cap[0] & DP_DS_PORT_TYPE_MASK;
+
+	switch (type) {
+	case DP_DS_PORT_TYPE_DP:
+	case DP_DS_PORT_TYPE_DP_DUALMODE:
+		return DRM_MODE_SUBCONNECTOR_DisplayPort;
+	case DP_DS_PORT_TYPE_VGA:
+		return DRM_MODE_SUBCONNECTOR_VGA;
+	case DP_DS_PORT_TYPE_DVI:
+		return DRM_MODE_SUBCONNECTOR_DVID;
+	case DP_DS_PORT_TYPE_HDMI:
+		return DRM_MODE_SUBCONNECTOR_HDMIA;
+	case DP_DS_PORT_TYPE_WIRELESS:
+		return DRM_MODE_SUBCONNECTOR_Wireless;
+	case DP_DS_PORT_TYPE_NON_EDID:
+	default:
+		return DRM_MODE_SUBCONNECTOR_Unknown;
+	}
+}
+EXPORT_SYMBOL(drm_dp_subconnector_type);
+
+/**
+ * drm_mode_set_dp_subconnector_property - set subconnector for DP connector
+ *
+ * Called by a driver on every detect event.
+ */
+void drm_dp_set_subconnector_property(struct drm_connector *connector,
+				      enum drm_connector_status status,
+				      const u8 *dpcd,
+				      const u8 port_cap[4])
+{
+	enum drm_mode_subconnector subconnector = DRM_MODE_SUBCONNECTOR_Unknown;
+
+	if (status == connector_status_connected)
+		subconnector = drm_dp_subconnector_type(dpcd, port_cap);
+	drm_object_property_set_value(&connector->base,
+			connector->dev->mode_config.dp_subconnector_property,
+			subconnector);
+}
+EXPORT_SYMBOL(drm_dp_set_subconnector_property);
+
+/**
+ * drm_dp_read_sink_count_cap() - Check whether a given connector has a valid sink
+ * count
+ * @connector: The DRM connector to check
+ * @dpcd: A cached copy of the connector's DPCD RX capabilities
+ * @desc: A cached copy of the connector's DP descriptor
+ *
+ * See also: drm_dp_read_sink_count()
+ *
+ * Returns: %True if the (e)DP connector has a valid sink count that should
+ * be probed, %false otherwise.
+ */
+bool drm_dp_read_sink_count_cap(struct drm_connector *connector,
+				const u8 dpcd[DP_RECEIVER_CAP_SIZE],
+				const struct drm_dp_desc *desc)
+{
+	/* Some eDP panels don't set a valid value for the sink count */
+	return connector->connector_type != DRM_MODE_CONNECTOR_eDP &&
+		dpcd[DP_DPCD_REV] >= DP_DPCD_REV_11 &&
+		dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_PRESENT &&
+		!drm_dp_has_quirk(desc, 0, DP_DPCD_QUIRK_NO_SINK_COUNT);
+}
+EXPORT_SYMBOL(drm_dp_read_sink_count_cap);
+
+/**
+ * drm_dp_read_sink_count() - Retrieve the sink count for a given sink
+ * @aux: The DP AUX channel to use
+ *
+ * See also: drm_dp_read_sink_count_cap()
+ *
+ * Returns: The current sink count reported by @aux, or a negative error code
+ * otherwise.
+ */
+int drm_dp_read_sink_count(struct drm_dp_aux *aux)
+{
+	u8 count;
+	int ret;
+
+	ret = drm_dp_dpcd_readb(aux, DP_SINK_COUNT, &count);
+	if (ret < 0)
+		return ret;
+	if (ret != 1)
+		return -EIO;
+
+	return DP_GET_SINK_COUNT(count);
+}
+EXPORT_SYMBOL(drm_dp_read_sink_count);
+
 /*
  * I2C-over-AUX implementation
  */
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index 67dd72ea200e..e87542533640 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -20,11 +20,13 @@
  * OF THIS SOFTWARE.
  */
 
+#include <linux/bitfield.h>
 #include <linux/delay.h>
 #include <linux/errno.h>
 #include <linux/i2c.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/random.h>
 #include <linux/sched.h>
 #include <linux/seq_file.h>
 #include <linux/iopoll.h>
@@ -423,6 +425,22 @@ drm_dp_encode_sideband_req(const struct drm_dp_sideband_msg_req_body *req,
 		memcpy(&buf[idx], req->u.i2c_write.bytes, req->u.i2c_write.num_bytes);
 		idx += req->u.i2c_write.num_bytes;
 		break;
+	case DP_QUERY_STREAM_ENC_STATUS: {
+		const struct drm_dp_query_stream_enc_status *msg;
+
+		msg = &req->u.enc_status;
+		buf[idx] = msg->stream_id;
+		idx++;
+		memcpy(&buf[idx], msg->client_id, sizeof(msg->client_id));
+		idx += sizeof(msg->client_id);
+		buf[idx] = 0;
+		buf[idx] |= FIELD_PREP(GENMASK(1, 0), msg->stream_event);
+		buf[idx] |= msg->valid_stream_event ? BIT(2) : 0;
+		buf[idx] |= FIELD_PREP(GENMASK(4, 3), msg->stream_behavior);
+		buf[idx] |= msg->valid_stream_behavior ? BIT(5) : 0;
+		idx++;
+		}
+		break;
 	}
 	raw->cur_len = idx;
 }
@@ -551,6 +569,20 @@ drm_dp_decode_sideband_req(const struct drm_dp_sideband_msg_tx *raw,
 				return -ENOMEM;
 		}
 		break;
+	case DP_QUERY_STREAM_ENC_STATUS:
+		req->u.enc_status.stream_id = buf[idx++];
+		for (i = 0; i < sizeof(req->u.enc_status.client_id); i++)
+			req->u.enc_status.client_id[i] = buf[idx++];
+
+		req->u.enc_status.stream_event = FIELD_GET(GENMASK(1, 0),
+							   buf[idx]);
+		req->u.enc_status.valid_stream_event = FIELD_GET(BIT(2),
+								 buf[idx]);
+		req->u.enc_status.stream_behavior = FIELD_GET(GENMASK(4, 3),
+							      buf[idx]);
+		req->u.enc_status.valid_stream_behavior = FIELD_GET(BIT(5),
+								    buf[idx]);
+		break;
 	}
 
 	return 0;
@@ -629,6 +661,16 @@ drm_dp_dump_sideband_msg_req_body(const struct drm_dp_sideband_msg_req_body *req
 		  req->u.i2c_write.num_bytes, req->u.i2c_write.num_bytes,
 		  req->u.i2c_write.bytes);
 		break;
+	case DP_QUERY_STREAM_ENC_STATUS:
+		P("stream_id=%u client_id=%*ph stream_event=%x "
+		  "valid_event=%d stream_behavior=%x valid_behavior=%d",
+		  req->u.enc_status.stream_id,
+		  (int)ARRAY_SIZE(req->u.enc_status.client_id),
+		  req->u.enc_status.client_id, req->u.enc_status.stream_event,
+		  req->u.enc_status.valid_stream_event,
+		  req->u.enc_status.stream_behavior,
+		  req->u.enc_status.valid_stream_behavior);
+		break;
 	default:
 		P("???\n");
 		break;
@@ -936,6 +978,42 @@ static bool drm_dp_sideband_parse_power_updown_phy_ack(struct drm_dp_sideband_ms
 	return true;
 }
 
+static bool
+drm_dp_sideband_parse_query_stream_enc_status(
+				struct drm_dp_sideband_msg_rx *raw,
+				struct drm_dp_sideband_msg_reply_body *repmsg)
+{
+	struct drm_dp_query_stream_enc_status_ack_reply *reply;
+
+	reply = &repmsg->u.enc_status;
+
+	reply->stream_id = raw->msg[3];
+
+	reply->reply_signed = raw->msg[2] & BIT(0);
+
+	/*
+	 * NOTE: It's my impression from reading the spec that the below parsing
+	 * is correct. However I noticed while testing with an HDCP 1.4 display
+	 * through an HDCP 2.2 hub that only bit 3 was set. In that case, I
+	 * would expect both bits to be set. So keep the parsing following the
+	 * spec, but beware reality might not match the spec (at least for some
+	 * configurations).
+	 */
+	reply->hdcp_1x_device_present = raw->msg[2] & BIT(4);
+	reply->hdcp_2x_device_present = raw->msg[2] & BIT(3);
+
+	reply->query_capable_device_present = raw->msg[2] & BIT(5);
+	reply->legacy_device_present = raw->msg[2] & BIT(6);
+	reply->unauthorizable_device_present = raw->msg[2] & BIT(7);
+
+	reply->auth_completed = !!(raw->msg[1] & BIT(3));
+	reply->encryption_enabled = !!(raw->msg[1] & BIT(4));
+	reply->repeater_present = !!(raw->msg[1] & BIT(5));
+	reply->state = (raw->msg[1] & GENMASK(7, 6)) >> 6;
+
+	return true;
+}
+
 static bool drm_dp_sideband_parse_reply(struct drm_dp_sideband_msg_rx *raw,
 					struct drm_dp_sideband_msg_reply_body *msg)
 {
@@ -961,6 +1039,8 @@ static bool drm_dp_sideband_parse_reply(struct drm_dp_sideband_msg_rx *raw,
 		return drm_dp_sideband_parse_remote_dpcd_write(raw, msg);
 	case DP_REMOTE_I2C_READ:
 		return drm_dp_sideband_parse_remote_i2c_read_ack(raw, msg);
+	case DP_REMOTE_I2C_WRITE:
+		return true; /* since there's nothing to parse */
 	case DP_ENUM_PATH_RESOURCES:
 		return drm_dp_sideband_parse_enum_path_resources_ack(raw, msg);
 	case DP_ALLOCATE_PAYLOAD:
@@ -970,6 +1050,8 @@ static bool drm_dp_sideband_parse_reply(struct drm_dp_sideband_msg_rx *raw,
 		return drm_dp_sideband_parse_power_updown_phy_ack(raw, msg);
 	case DP_CLEAR_PAYLOAD_ID_TABLE:
 		return true; /* since there's nothing to parse */
+	case DP_QUERY_STREAM_ENC_STATUS:
+		return drm_dp_sideband_parse_query_stream_enc_status(raw, msg);
 	default:
 		DRM_ERROR("Got unknown reply 0x%02x (%s)\n", msg->req_type,
 			  drm_dp_mst_req_type_str(msg->req_type));
@@ -1121,6 +1203,25 @@ static void build_power_updown_phy(struct drm_dp_sideband_msg_tx *msg,
 	msg->path_msg = true;
 }
 
+static int
+build_query_stream_enc_status(struct drm_dp_sideband_msg_tx *msg, u8 stream_id,
+			      u8 *q_id)
+{
+	struct drm_dp_sideband_msg_req_body req;
+
+	req.req_type = DP_QUERY_STREAM_ENC_STATUS;
+	req.u.enc_status.stream_id = stream_id;
+	memcpy(req.u.enc_status.client_id, q_id,
+	       sizeof(req.u.enc_status.client_id));
+	req.u.enc_status.stream_event = 0;
+	req.u.enc_status.valid_stream_event = false;
+	req.u.enc_status.stream_behavior = 0;
+	req.u.enc_status.valid_stream_behavior = false;
+
+	drm_dp_encode_sideband_req(&req, msg);
+	return 0;
+}
+
 static int drm_dp_mst_assign_payload_id(struct drm_dp_mst_topology_mgr *mgr,
 					struct drm_dp_vcpi *vcpi)
 {
@@ -3153,6 +3254,57 @@ int drm_dp_send_power_updown_phy(struct drm_dp_mst_topology_mgr *mgr,
 }
 EXPORT_SYMBOL(drm_dp_send_power_updown_phy);
 
+int drm_dp_send_query_stream_enc_status(struct drm_dp_mst_topology_mgr *mgr,
+		struct drm_dp_mst_port *port,
+		struct drm_dp_query_stream_enc_status_ack_reply *status)
+{
+	struct drm_dp_sideband_msg_tx *txmsg;
+	u8 nonce[7];
+	int len, ret;
+
+	txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL);
+	if (!txmsg)
+		return -ENOMEM;
+
+	port = drm_dp_mst_topology_get_port_validated(mgr, port);
+	if (!port) {
+		ret = -EINVAL;
+		goto out_get_port;
+	}
+
+	get_random_bytes(nonce, sizeof(nonce));
+
+	/*
+	 * "Source device targets the QUERY_STREAM_ENCRYPTION_STATUS message
+	 *  transaction at the MST Branch device directly connected to the
+	 *  Source"
+	 */
+	txmsg->dst = mgr->mst_primary;
+
+	len = build_query_stream_enc_status(txmsg, port->vcpi.vcpi, nonce);
+
+	drm_dp_queue_down_tx(mgr, txmsg);
+
+	ret = drm_dp_mst_wait_tx_reply(mgr->mst_primary, txmsg);
+	if (ret < 0) {
+		goto out;
+	} else if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) {
+		drm_dbg_kms(mgr->dev, "query encryption status nak received\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	ret = 0;
+	memcpy(status, &txmsg->reply.u.enc_status, sizeof(*status));
+
+out:
+	drm_dp_mst_topology_put_port(port);
+out_get_port:
+	kfree(txmsg);
+	return ret;
+}
+EXPORT_SYMBOL(drm_dp_send_query_stream_enc_status);
+
 static int drm_dp_create_payload_step1(struct drm_dp_mst_topology_mgr *mgr,
 				       int id,
 				       struct drm_dp_payload *payload)
@@ -3487,6 +3639,28 @@ static int drm_dp_get_vc_payload_bw(u8 dp_link_bw, u8  dp_link_count)
 }
 
 /**
+ * drm_dp_read_mst_cap() - check whether or not a sink supports MST
+ * @aux: The DP AUX channel to use
+ * @dpcd: A cached copy of the DPCD capabilities for this sink
+ *
+ * Returns: %True if the sink supports MST, %false otherwise
+ */
+bool drm_dp_read_mst_cap(struct drm_dp_aux *aux,
+			 const u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	u8 mstm_cap;
+
+	if (dpcd[DP_DPCD_REV] < DP_DPCD_REV_12)
+		return false;
+
+	if (drm_dp_dpcd_readb(aux, DP_MSTM_CAP, &mstm_cap) != 1)
+		return false;
+
+	return mstm_cap & DP_MST_CAP;
+}
+EXPORT_SYMBOL(drm_dp_read_mst_cap);
+
+/**
  * drm_dp_mst_topology_mgr_set_mst() - Set the MST state for a topology manager
  * @mgr: manager to set state for
  * @mst_state: true to enable MST on this connector - false to disable.
@@ -5327,29 +5501,29 @@ static bool remote_i2c_read_ok(const struct i2c_msg msgs[], int num)
 		msgs[num - 1].len <= 0xff;
 }
 
-/* I2C device */
-static int drm_dp_mst_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
-			       int num)
+static bool remote_i2c_write_ok(const struct i2c_msg msgs[], int num)
+{
+	int i;
+
+	for (i = 0; i < num - 1; i++) {
+		if (msgs[i].flags & I2C_M_RD || !(msgs[i].flags & I2C_M_STOP) ||
+		    msgs[i].len > 0xff)
+			return false;
+	}
+
+	return !(msgs[num - 1].flags & I2C_M_RD) && msgs[num - 1].len <= 0xff;
+}
+
+static int drm_dp_mst_i2c_read(struct drm_dp_mst_branch *mstb,
+			       struct drm_dp_mst_port *port,
+			       struct i2c_msg *msgs, int num)
 {
-	struct drm_dp_aux *aux = adapter->algo_data;
-	struct drm_dp_mst_port *port = container_of(aux, struct drm_dp_mst_port, aux);
-	struct drm_dp_mst_branch *mstb;
 	struct drm_dp_mst_topology_mgr *mgr = port->mgr;
 	unsigned int i;
 	struct drm_dp_sideband_msg_req_body msg;
 	struct drm_dp_sideband_msg_tx *txmsg = NULL;
 	int ret;
 
-	mstb = drm_dp_mst_topology_get_mstb_validated(mgr, port->parent);
-	if (!mstb)
-		return -EREMOTEIO;
-
-	if (!remote_i2c_read_ok(msgs, num)) {
-		DRM_DEBUG_KMS("Unsupported I2C transaction for MST device\n");
-		ret = -EIO;
-		goto out;
-	}
-
 	memset(&msg, 0, sizeof(msg));
 	msg.req_type = DP_REMOTE_I2C_READ;
 	msg.u.i2c_read.num_transactions = num - 1;
@@ -5390,6 +5564,78 @@ static int drm_dp_mst_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs
 	}
 out:
 	kfree(txmsg);
+	return ret;
+}
+
+static int drm_dp_mst_i2c_write(struct drm_dp_mst_branch *mstb,
+				struct drm_dp_mst_port *port,
+				struct i2c_msg *msgs, int num)
+{
+	struct drm_dp_mst_topology_mgr *mgr = port->mgr;
+	unsigned int i;
+	struct drm_dp_sideband_msg_req_body msg;
+	struct drm_dp_sideband_msg_tx *txmsg = NULL;
+	int ret;
+
+	txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL);
+	if (!txmsg) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	for (i = 0; i < num; i++) {
+		memset(&msg, 0, sizeof(msg));
+		msg.req_type = DP_REMOTE_I2C_WRITE;
+		msg.u.i2c_write.port_number = port->port_num;
+		msg.u.i2c_write.write_i2c_device_id = msgs[i].addr;
+		msg.u.i2c_write.num_bytes = msgs[i].len;
+		msg.u.i2c_write.bytes = msgs[i].buf;
+
+		memset(txmsg, 0, sizeof(*txmsg));
+		txmsg->dst = mstb;
+
+		drm_dp_encode_sideband_req(&msg, txmsg);
+		drm_dp_queue_down_tx(mgr, txmsg);
+
+		ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
+		if (ret > 0) {
+			if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) {
+				ret = -EREMOTEIO;
+				goto out;
+			}
+		} else {
+			goto out;
+		}
+	}
+	ret = num;
+out:
+	kfree(txmsg);
+	return ret;
+}
+
+/* I2C device */
+static int drm_dp_mst_i2c_xfer(struct i2c_adapter *adapter,
+			       struct i2c_msg *msgs, int num)
+{
+	struct drm_dp_aux *aux = adapter->algo_data;
+	struct drm_dp_mst_port *port =
+		container_of(aux, struct drm_dp_mst_port, aux);
+	struct drm_dp_mst_branch *mstb;
+	struct drm_dp_mst_topology_mgr *mgr = port->mgr;
+	int ret;
+
+	mstb = drm_dp_mst_topology_get_mstb_validated(mgr, port->parent);
+	if (!mstb)
+		return -EREMOTEIO;
+
+	if (remote_i2c_read_ok(msgs, num)) {
+		ret = drm_dp_mst_i2c_read(mstb, port, msgs, num);
+	} else if (remote_i2c_write_ok(msgs, num)) {
+		ret = drm_dp_mst_i2c_write(mstb, port, msgs, num);
+	} else {
+		DRM_DEBUG_KMS("Unsupported I2C transaction for MST device\n");
+		ret = -EIO;
+	}
+
 	drm_dp_mst_topology_put_mstb(mstb);
 	return ret;
 }
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 13068fdf4331..cd162d406078 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -240,13 +240,13 @@ void drm_minor_release(struct drm_minor *minor)
  * DOC: driver instance overview
  *
  * A device instance for a drm driver is represented by &struct drm_device. This
- * is initialized with drm_dev_init(), usually from bus-specific ->probe()
- * callbacks implemented by the driver. The driver then needs to initialize all
- * the various subsystems for the drm device like memory management, vblank
- * handling, modesetting support and intial output configuration plus obviously
- * initialize all the corresponding hardware bits. Finally when everything is up
- * and running and ready for userspace the device instance can be published
- * using drm_dev_register().
+ * is allocated and initialized with devm_drm_dev_alloc(), usually from
+ * bus-specific ->probe() callbacks implemented by the driver. The driver then
+ * needs to initialize all the various subsystems for the drm device like memory
+ * management, vblank handling, modesetting support and initial output
+ * configuration plus obviously initialize all the corresponding hardware bits.
+ * Finally when everything is up and running and ready for userspace the device
+ * instance can be published using drm_dev_register().
  *
  * There is also deprecated support for initalizing device instances using
  * bus-specific helpers and the &drm_driver.load callback. But due to
@@ -274,7 +274,7 @@ void drm_minor_release(struct drm_minor *minor)
  *
  * The following example shows a typical structure of a DRM display driver.
  * The example focus on the probe() function and the other functions that is
- * almost always present and serves as a demonstration of devm_drm_dev_init().
+ * almost always present and serves as a demonstration of devm_drm_dev_alloc().
  *
  * .. code-block:: c
  *
@@ -294,22 +294,12 @@ void drm_minor_release(struct drm_minor *minor)
  *		struct drm_device *drm;
  *		int ret;
  *
- *		// devm_kzalloc() can't be used here because the drm_device '
- *		// lifetime can exceed the device lifetime if driver unbind
- *		// happens when userspace still has open file descriptors.
- *		priv = kzalloc(sizeof(*priv), GFP_KERNEL);
- *		if (!priv)
- *			return -ENOMEM;
- *
+ *		priv = devm_drm_dev_alloc(&pdev->dev, &driver_drm_driver,
+ *					  struct driver_device, drm);
+ *		if (IS_ERR(priv))
+ *			return PTR_ERR(priv);
  *		drm = &priv->drm;
  *
- *		ret = devm_drm_dev_init(&pdev->dev, drm, &driver_drm_driver);
- *		if (ret) {
- *			kfree(priv);
- *			return ret;
- *		}
- *		drmm_add_final_kfree(drm, priv);
- *
  *		ret = drmm_mode_config_init(drm);
  *		if (ret)
  *			return ret;
@@ -550,9 +540,9 @@ static void drm_fs_inode_free(struct inode *inode)
  * following guidelines apply:
  *
  *  - The entire device initialization procedure should be run from the
- *    &component_master_ops.master_bind callback, starting with drm_dev_init(),
- *    then binding all components with component_bind_all() and finishing with
- *    drm_dev_register().
+ *    &component_master_ops.master_bind callback, starting with
+ *    devm_drm_dev_alloc(), then binding all components with
+ *    component_bind_all() and finishing with drm_dev_register().
  *
  *  - The opaque pointer passed to all components through component_bind_all()
  *    should point at &struct drm_device of the device instance, not some driver
@@ -583,43 +573,9 @@ static void drm_dev_init_release(struct drm_device *dev, void *res)
 	drm_legacy_destroy_members(dev);
 }
 
-/**
- * drm_dev_init - Initialise new DRM device
- * @dev: DRM device
- * @driver: DRM driver
- * @parent: Parent device object
- *
- * Initialize a new DRM device. No device registration is done.
- * Call drm_dev_register() to advertice the device to user space and register it
- * with other core subsystems. This should be done last in the device
- * initialization sequence to make sure userspace can't access an inconsistent
- * state.
- *
- * The initial ref-count of the object is 1. Use drm_dev_get() and
- * drm_dev_put() to take and drop further ref-counts.
- *
- * It is recommended that drivers embed &struct drm_device into their own device
- * structure.
- *
- * Drivers that do not want to allocate their own device struct
- * embedding &struct drm_device can call drm_dev_alloc() instead. For drivers
- * that do embed &struct drm_device it must be placed first in the overall
- * structure, and the overall structure must be allocated using kmalloc(): The
- * drm core's release function unconditionally calls kfree() on the @dev pointer
- * when the final reference is released. To override this behaviour, and so
- * allow embedding of the drm_device inside the driver's device struct at an
- * arbitrary offset, you must supply a &drm_driver.release callback and control
- * the finalization explicitly.
- *
- * Note that drivers must call drmm_add_final_kfree() after this function has
- * completed successfully.
- *
- * RETURNS:
- * 0 on success, or error code on failure.
- */
-int drm_dev_init(struct drm_device *dev,
-		 struct drm_driver *driver,
-		 struct device *parent)
+static int drm_dev_init(struct drm_device *dev,
+			struct drm_driver *driver,
+			struct device *parent)
 {
 	int ret;
 
@@ -699,31 +655,15 @@ err:
 
 	return ret;
 }
-EXPORT_SYMBOL(drm_dev_init);
 
 static void devm_drm_dev_init_release(void *data)
 {
 	drm_dev_put(data);
 }
 
-/**
- * devm_drm_dev_init - Resource managed drm_dev_init()
- * @parent: Parent device object
- * @dev: DRM device
- * @driver: DRM driver
- *
- * Managed drm_dev_init(). The DRM device initialized with this function is
- * automatically put on driver detach using drm_dev_put().
- *
- * Note that drivers must call drmm_add_final_kfree() after this function has
- * completed successfully.
- *
- * RETURNS:
- * 0 on success, or error code on failure.
- */
-int devm_drm_dev_init(struct device *parent,
-		      struct drm_device *dev,
-		      struct drm_driver *driver)
+static int devm_drm_dev_init(struct device *parent,
+			     struct drm_device *dev,
+			     struct drm_driver *driver)
 {
 	int ret;
 
@@ -737,7 +677,6 @@ int devm_drm_dev_init(struct device *parent,
 
 	return ret;
 }
-EXPORT_SYMBOL(devm_drm_dev_init);
 
 void *__devm_drm_dev_alloc(struct device *parent, struct drm_driver *driver,
 			   size_t size, size_t offset)
@@ -767,19 +706,9 @@ EXPORT_SYMBOL(__devm_drm_dev_alloc);
  * @driver: DRM driver to allocate device for
  * @parent: Parent device object
  *
- * Allocate and initialize a new DRM device. No device registration is done.
- * Call drm_dev_register() to advertice the device to user space and register it
- * with other core subsystems. This should be done last in the device
- * initialization sequence to make sure userspace can't access an inconsistent
- * state.
- *
- * The initial ref-count of the object is 1. Use drm_dev_get() and
- * drm_dev_put() to take and drop further ref-counts.
- *
- * Note that for purely virtual devices @parent can be NULL.
- *
- * Drivers that wish to subclass or embed &struct drm_device into their
- * own struct should look at using drm_dev_init() instead.
+ * This is the deprecated version of devm_drm_dev_alloc(), which does not support
+ * subclassing through embedding the struct &drm_device in a driver private
+ * structure, and which does not support automatic cleanup through devres.
  *
  * RETURNS:
  * Pointer to new DRM device, or ERR_PTR on failure.
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 6840f0530a38..a82f37d44258 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -3738,6 +3738,34 @@ drm_add_cmdb_modes(struct drm_connector *connector, u8 svd)
 	bitmap_set(hdmi->y420_cmdb_modes, vic, 1);
 }
 
+/**
+ * drm_display_mode_from_cea_vic() - return a mode for CEA VIC
+ * @dev: DRM device
+ * @vic: CEA VIC of the mode
+ *
+ * Creates a new mode matching the specified CEA VIC.
+ *
+ * Returns: A new drm_display_mode on success or NULL on failure
+ */
+struct drm_display_mode *
+drm_display_mode_from_cea_vic(struct drm_device *dev,
+			      u8 video_code)
+{
+	const struct drm_display_mode *cea_mode;
+	struct drm_display_mode *newmode;
+
+	cea_mode = cea_mode_for_vic(video_code);
+	if (!cea_mode)
+		return NULL;
+
+	newmode = drm_mode_duplicate(dev, cea_mode);
+	if (!newmode)
+		return NULL;
+
+	return newmode;
+}
+EXPORT_SYMBOL(drm_display_mode_from_cea_vic);
+
 static int
 do_cea_modes(struct drm_connector *connector, const u8 *db, u8 len)
 {
diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
index df656366a530..2f5b0c2bb0fe 100644
--- a/drivers/gpu/drm/drm_framebuffer.c
+++ b/drivers/gpu/drm/drm_framebuffer.c
@@ -176,8 +176,7 @@ static int framebuffer_check(struct drm_device *dev,
 	int i;
 
 	/* check if the format is supported at all */
-	info = __drm_format_info(r->pixel_format);
-	if (!info) {
+	if (!__drm_format_info(r->pixel_format)) {
 		struct drm_format_name_buf format_name;
 
 		DRM_DEBUG_KMS("bad framebuffer format %s\n",
@@ -186,9 +185,6 @@ static int framebuffer_check(struct drm_device *dev,
 		return -EINVAL;
 	}
 
-	/* now let the driver pick its own format info */
-	info = drm_get_format_info(dev, r);
-
 	if (r->width == 0) {
 		DRM_DEBUG_KMS("bad framebuffer width %u\n", r->width);
 		return -EINVAL;
@@ -199,6 +195,9 @@ static int framebuffer_check(struct drm_device *dev,
 		return -EINVAL;
 	}
 
+	/* now let the driver pick its own format info */
+	info = drm_get_format_info(dev, r);
+
 	for (i = 0; i < info->num_planes; i++) {
 		unsigned int width = fb_plane_width(r->width, info, i);
 		unsigned int height = fb_plane_height(r->height, info, i);
diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c
index 822edeadbab3..59b9ca207b42 100644
--- a/drivers/gpu/drm/drm_gem_cma_helper.c
+++ b/drivers/gpu/drm/drm_gem_cma_helper.c
@@ -471,26 +471,9 @@ drm_gem_cma_prime_import_sg_table(struct drm_device *dev,
 {
 	struct drm_gem_cma_object *cma_obj;
 
-	if (sgt->nents != 1) {
-		/* check if the entries in the sg_table are contiguous */
-		dma_addr_t next_addr = sg_dma_address(sgt->sgl);
-		struct scatterlist *s;
-		unsigned int i;
-
-		for_each_sg(sgt->sgl, s, sgt->nents, i) {
-			/*
-			 * sg_dma_address(s) is only valid for entries
-			 * that have sg_dma_len(s) != 0
-			 */
-			if (!sg_dma_len(s))
-				continue;
-
-			if (sg_dma_address(s) != next_addr)
-				return ERR_PTR(-EINVAL);
-
-			next_addr = sg_dma_address(s) + sg_dma_len(s);
-		}
-	}
+	/* check if the entries in the sg_table are contiguous */
+	if (drm_prime_get_contiguous_size(sgt) < attach->dmabuf->size)
+		return ERR_PTR(-EINVAL);
 
 	/* Create a CMA GEM buffer. */
 	cma_obj = __drm_gem_cma_create(dev, attach->dmabuf->size);
diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c
index 4b7cfbac4daa..d77c9f8ff26c 100644
--- a/drivers/gpu/drm/drm_gem_shmem_helper.c
+++ b/drivers/gpu/drm/drm_gem_shmem_helper.c
@@ -126,8 +126,8 @@ void drm_gem_shmem_free_object(struct drm_gem_object *obj)
 		drm_prime_gem_destroy(obj, shmem->sgt);
 	} else {
 		if (shmem->sgt) {
-			dma_unmap_sg(obj->dev->dev, shmem->sgt->sgl,
-				     shmem->sgt->nents, DMA_BIDIRECTIONAL);
+			dma_unmap_sgtable(obj->dev->dev, shmem->sgt,
+					  DMA_BIDIRECTIONAL, 0);
 			sg_free_table(shmem->sgt);
 			kfree(shmem->sgt);
 		}
@@ -424,8 +424,7 @@ void drm_gem_shmem_purge_locked(struct drm_gem_object *obj)
 
 	WARN_ON(!drm_gem_shmem_is_purgeable(shmem));
 
-	dma_unmap_sg(obj->dev->dev, shmem->sgt->sgl,
-		     shmem->sgt->nents, DMA_BIDIRECTIONAL);
+	dma_unmap_sgtable(obj->dev->dev, shmem->sgt, DMA_BIDIRECTIONAL, 0);
 	sg_free_table(shmem->sgt);
 	kfree(shmem->sgt);
 	shmem->sgt = NULL;
@@ -656,7 +655,7 @@ struct sg_table *drm_gem_shmem_get_sg_table(struct drm_gem_object *obj)
 
 	WARN_ON(shmem->base.import_attach);
 
-	return drm_prime_pages_to_sg(shmem->pages, obj->size >> PAGE_SHIFT);
+	return drm_prime_pages_to_sg(obj->dev, shmem->pages, obj->size >> PAGE_SHIFT);
 }
 EXPORT_SYMBOL_GPL(drm_gem_shmem_get_sg_table);
 
@@ -697,12 +696,17 @@ struct sg_table *drm_gem_shmem_get_pages_sgt(struct drm_gem_object *obj)
 		goto err_put_pages;
 	}
 	/* Map the pages for use by the h/w. */
-	dma_map_sg(obj->dev->dev, sgt->sgl, sgt->nents, DMA_BIDIRECTIONAL);
+	ret = dma_map_sgtable(obj->dev->dev, sgt, DMA_BIDIRECTIONAL, 0);
+	if (ret)
+		goto err_free_sgt;
 
 	shmem->sgt = sgt;
 
 	return sgt;
 
+err_free_sgt:
+	sg_free_table(sgt);
+	kfree(sgt);
 err_put_pages:
 	drm_gem_shmem_put_pages(shmem);
 	return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/drm_gem_ttm_helper.c b/drivers/gpu/drm/drm_gem_ttm_helper.c
index 892b2288a104..0e4fb9ba43ad 100644
--- a/drivers/gpu/drm/drm_gem_ttm_helper.c
+++ b/drivers/gpu/drm/drm_gem_ttm_helper.c
@@ -43,12 +43,9 @@ void drm_gem_ttm_print_info(struct drm_printer *p, unsigned int indent,
 	drm_print_bits(p, bo->mem.placement, plname, ARRAY_SIZE(plname));
 	drm_printf(p, "\n");
 
-	if (bo->mem.bus.is_iomem) {
-		drm_printf_indent(p, indent, "bus.base=%lx\n",
-				  (unsigned long)bo->mem.bus.base);
+	if (bo->mem.bus.is_iomem)
 		drm_printf_indent(p, indent, "bus.offset=%lx\n",
 				  (unsigned long)bo->mem.bus.offset);
-	}
 }
 EXPORT_SYMBOL(drm_gem_ttm_print_info);
 
diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c
index 3296ed3df358..50cad0e4a92e 100644
--- a/drivers/gpu/drm/drm_gem_vram_helper.c
+++ b/drivers/gpu/drm/drm_gem_vram_helper.c
@@ -97,8 +97,8 @@ static const struct drm_gem_object_funcs drm_gem_vram_object_funcs;
  * hardware's draing engine.
  *
  * To access a buffer object's memory from the DRM driver, call
- * drm_gem_vram_kmap(). It (optionally) maps the buffer into kernel address
- * space and returns the memory address. Use drm_gem_vram_kunmap() to
+ * drm_gem_vram_vmap(). It maps the buffer into kernel address
+ * space and returns the memory address. Use drm_gem_vram_vunmap() to
  * release the mapping.
  */
 
@@ -135,28 +135,28 @@ static void ttm_buffer_object_destroy(struct ttm_buffer_object *bo)
 static void drm_gem_vram_placement(struct drm_gem_vram_object *gbo,
 				   unsigned long pl_flag)
 {
+	u32 invariant_flags = 0;
 	unsigned int i;
 	unsigned int c = 0;
-	u32 invariant_flags = pl_flag & TTM_PL_FLAG_TOPDOWN;
+
+	if (pl_flag & DRM_GEM_VRAM_PL_FLAG_TOPDOWN)
+		pl_flag = TTM_PL_FLAG_TOPDOWN;
 
 	gbo->placement.placement = gbo->placements;
 	gbo->placement.busy_placement = gbo->placements;
 
-	if (pl_flag & TTM_PL_FLAG_VRAM)
+	if (pl_flag & DRM_GEM_VRAM_PL_FLAG_VRAM) {
+		gbo->placements[c].mem_type = TTM_PL_VRAM;
 		gbo->placements[c++].flags = TTM_PL_FLAG_WC |
 					     TTM_PL_FLAG_UNCACHED |
-					     TTM_PL_FLAG_VRAM |
-					     invariant_flags;
-
-	if (pl_flag & TTM_PL_FLAG_SYSTEM)
-		gbo->placements[c++].flags = TTM_PL_MASK_CACHING |
-					     TTM_PL_FLAG_SYSTEM |
 					     invariant_flags;
+	}
 
-	if (!c)
+	if (pl_flag & DRM_GEM_VRAM_PL_FLAG_SYSTEM || !c) {
+		gbo->placements[c].mem_type = TTM_PL_SYSTEM;
 		gbo->placements[c++].flags = TTM_PL_MASK_CACHING |
-					     TTM_PL_FLAG_SYSTEM |
 					     invariant_flags;
+	}
 
 	gbo->placement.num_placement = c;
 	gbo->placement.num_busy_placement = c;
@@ -167,6 +167,10 @@ static void drm_gem_vram_placement(struct drm_gem_vram_object *gbo,
 	}
 }
 
+/*
+ * Note that on error, drm_gem_vram_init will free the buffer object.
+ */
+
 static int drm_gem_vram_init(struct drm_device *dev,
 			     struct drm_gem_vram_object *gbo,
 			     size_t size, unsigned long pg_align)
@@ -176,32 +180,37 @@ static int drm_gem_vram_init(struct drm_device *dev,
 	int ret;
 	size_t acc_size;
 
-	if (WARN_ONCE(!vmm, "VRAM MM not initialized"))
+	if (WARN_ONCE(!vmm, "VRAM MM not initialized")) {
+		kfree(gbo);
 		return -EINVAL;
+	}
 	bdev = &vmm->bdev;
 
 	gbo->bo.base.funcs = &drm_gem_vram_object_funcs;
 
 	ret = drm_gem_object_init(dev, &gbo->bo.base, size);
-	if (ret)
+	if (ret) {
+		kfree(gbo);
 		return ret;
+	}
 
 	acc_size = ttm_bo_dma_acc_size(bdev, size, sizeof(*gbo));
 
 	gbo->bo.bdev = bdev;
-	drm_gem_vram_placement(gbo, TTM_PL_FLAG_VRAM | TTM_PL_FLAG_SYSTEM);
+	drm_gem_vram_placement(gbo, DRM_GEM_VRAM_PL_FLAG_VRAM |
+			       DRM_GEM_VRAM_PL_FLAG_SYSTEM);
 
 	ret = ttm_bo_init(bdev, &gbo->bo, size, ttm_bo_type_device,
 			  &gbo->placement, pg_align, false, acc_size,
 			  NULL, NULL, ttm_buffer_object_destroy);
 	if (ret)
-		goto err_drm_gem_object_release;
+		/*
+		 * A failing ttm_bo_init will call ttm_buffer_object_destroy
+		 * to release gbo->bo.base and kfree gbo.
+		 */
+		return ret;
 
 	return 0;
-
-err_drm_gem_object_release:
-	drm_gem_object_release(&gbo->bo.base);
-	return ret;
 }
 
 /**
@@ -235,13 +244,9 @@ struct drm_gem_vram_object *drm_gem_vram_create(struct drm_device *dev,
 
 	ret = drm_gem_vram_init(dev, gbo, size, pg_align);
 	if (ret < 0)
-		goto err_kfree;
+		return ERR_PTR(ret);
 
 	return gbo;
-
-err_kfree:
-	kfree(gbo);
-	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL(drm_gem_vram_create);
 
@@ -436,39 +441,6 @@ out:
 	return kmap->virtual;
 }
 
-/**
- * drm_gem_vram_kmap() - Maps a GEM VRAM object into kernel address space
- * @gbo:	the GEM VRAM object
- * @map:	establish a mapping if necessary
- * @is_iomem:	returns true if the mapped memory is I/O memory, or false \
-	otherwise; can be NULL
- *
- * This function maps the buffer object into the kernel's address space
- * or returns the current mapping. If the parameter map is false, the
- * function only queries the current mapping, but does not establish a
- * new one.
- *
- * Returns:
- * The buffers virtual address if mapped, or
- * NULL if not mapped, or
- * an ERR_PTR()-encoded error code otherwise.
- */
-void *drm_gem_vram_kmap(struct drm_gem_vram_object *gbo, bool map,
-			bool *is_iomem)
-{
-	int ret;
-	void *virtual;
-
-	ret = ttm_bo_reserve(&gbo->bo, true, false, NULL);
-	if (ret)
-		return ERR_PTR(ret);
-	virtual = drm_gem_vram_kmap_locked(gbo, map, is_iomem);
-	ttm_bo_unreserve(&gbo->bo);
-
-	return virtual;
-}
-EXPORT_SYMBOL(drm_gem_vram_kmap);
-
 static void drm_gem_vram_kunmap_locked(struct drm_gem_vram_object *gbo)
 {
 	if (WARN_ON_ONCE(!gbo->kmap_use_count))
@@ -485,22 +457,6 @@ static void drm_gem_vram_kunmap_locked(struct drm_gem_vram_object *gbo)
 }
 
 /**
- * drm_gem_vram_kunmap() - Unmaps a GEM VRAM object
- * @gbo:	the GEM VRAM object
- */
-void drm_gem_vram_kunmap(struct drm_gem_vram_object *gbo)
-{
-	int ret;
-
-	ret = ttm_bo_reserve(&gbo->bo, false, false, NULL);
-	if (WARN_ONCE(ret, "ttm_bo_reserve_failed(): ret=%d\n", ret))
-		return;
-	drm_gem_vram_kunmap_locked(gbo);
-	ttm_bo_unreserve(&gbo->bo);
-}
-EXPORT_SYMBOL(drm_gem_vram_kunmap);
-
-/**
  * drm_gem_vram_vmap() - Pins and maps a GEM VRAM object into kernel address
  *                       space
  * @gbo:	The GEM VRAM object to map
@@ -511,9 +467,6 @@ EXPORT_SYMBOL(drm_gem_vram_kunmap);
  * permanently. Call drm_gem_vram_vunmap() with the returned address to
  * unmap and unpin the GEM VRAM object.
  *
- * If you have special requirements for the pinning or mapping operations,
- * call drm_gem_vram_pin() and drm_gem_vram_kmap() directly.
- *
  * Returns:
  * The buffer's virtual address on success, or
  * an ERR_PTR()-encoded error code otherwise.
@@ -647,13 +600,13 @@ static bool drm_is_gem_vram(struct ttm_buffer_object *bo)
 static void drm_gem_vram_bo_driver_evict_flags(struct drm_gem_vram_object *gbo,
 					       struct ttm_placement *pl)
 {
-	drm_gem_vram_placement(gbo, TTM_PL_FLAG_SYSTEM);
+	drm_gem_vram_placement(gbo, DRM_GEM_VRAM_PL_FLAG_SYSTEM);
 	*pl = gbo->placement;
 }
 
 static void drm_gem_vram_bo_driver_move_notify(struct drm_gem_vram_object *gbo,
 					       bool evict,
-					       struct ttm_mem_reg *new_mem)
+					       struct ttm_resource *new_mem)
 {
 	struct ttm_bo_kmap_obj *kmap = &gbo->kmap;
 
@@ -967,16 +920,13 @@ static const struct drm_gem_object_funcs drm_gem_vram_object_funcs = {
  * TTM TT
  */
 
-static void backend_func_destroy(struct ttm_tt *tt)
+static void bo_driver_ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *tt)
 {
+	ttm_tt_destroy_common(bdev, tt);
 	ttm_tt_fini(tt);
 	kfree(tt);
 }
 
-static struct ttm_backend_func backend_func = {
-	.destroy = backend_func_destroy
-};
-
 /*
  * TTM BO device
  */
@@ -991,8 +941,6 @@ static struct ttm_tt *bo_driver_ttm_tt_create(struct ttm_buffer_object *bo,
 	if (!tt)
 		return NULL;
 
-	tt->func = &backend_func;
-
 	ret = ttm_tt_init(tt, bo, page_flags);
 	if (ret < 0)
 		goto err_ttm_tt_init;
@@ -1004,28 +952,6 @@ err_ttm_tt_init:
 	return NULL;
 }
 
-static int bo_driver_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-				   struct ttm_mem_type_manager *man)
-{
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		man->flags = 0;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_VRAM:
-		man->func = &ttm_bo_manager_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED;
-		man->available_caching = TTM_PL_FLAG_UNCACHED |
-					 TTM_PL_FLAG_WC;
-		man->default_caching = TTM_PL_FLAG_WC;
-		break;
-	default:
-		return -EINVAL;
-	}
-	return 0;
-}
-
 static void bo_driver_evict_flags(struct ttm_buffer_object *bo,
 				  struct ttm_placement *placement)
 {
@@ -1042,7 +968,7 @@ static void bo_driver_evict_flags(struct ttm_buffer_object *bo,
 
 static void bo_driver_move_notify(struct ttm_buffer_object *bo,
 				  bool evict,
-				  struct ttm_mem_reg *new_mem)
+				  struct ttm_resource *new_mem)
 {
 	struct drm_gem_vram_object *gbo;
 
@@ -1056,22 +982,15 @@ static void bo_driver_move_notify(struct ttm_buffer_object *bo,
 }
 
 static int bo_driver_io_mem_reserve(struct ttm_bo_device *bdev,
-				    struct ttm_mem_reg *mem)
+				    struct ttm_resource *mem)
 {
 	struct drm_vram_mm *vmm = drm_vram_mm_of_bdev(bdev);
 
-	mem->bus.addr = NULL;
-	mem->bus.size = mem->num_pages << PAGE_SHIFT;
-
 	switch (mem->mem_type) {
 	case TTM_PL_SYSTEM:	/* nothing to do */
-		mem->bus.offset = 0;
-		mem->bus.base = 0;
-		mem->bus.is_iomem = false;
 		break;
 	case TTM_PL_VRAM:
-		mem->bus.offset = mem->start << PAGE_SHIFT;
-		mem->bus.base = vmm->vram_base;
+		mem->bus.offset = (mem->start << PAGE_SHIFT) + vmm->vram_base;
 		mem->bus.is_iomem = true;
 		break;
 	default:
@@ -1083,9 +1002,7 @@ static int bo_driver_io_mem_reserve(struct ttm_bo_device *bdev,
 
 static struct ttm_bo_driver bo_driver = {
 	.ttm_tt_create = bo_driver_ttm_tt_create,
-	.ttm_tt_populate = ttm_pool_populate,
-	.ttm_tt_unpopulate = ttm_pool_unpopulate,
-	.init_mem_type = bo_driver_init_mem_type,
+	.ttm_tt_destroy = bo_driver_ttm_tt_destroy,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = bo_driver_evict_flags,
 	.move_notify = bo_driver_move_notify,
@@ -1100,12 +1017,10 @@ static int drm_vram_mm_debugfs(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_vram_mm *vmm = node->minor->dev->vram_mm;
-	struct drm_mm *mm = vmm->bdev.man[TTM_PL_VRAM].priv;
+	struct ttm_resource_manager *man = ttm_manager_type(&vmm->bdev, TTM_PL_VRAM);
 	struct drm_printer p = drm_seq_file_printer(m);
 
-	spin_lock(&ttm_bo_glob.lru_lock);
-	drm_mm_print(mm, &p);
-	spin_unlock(&ttm_bo_glob.lru_lock);
+	ttm_resource_manager_debug(man, &p);
 	return 0;
 }
 
@@ -1142,7 +1057,8 @@ static int drm_vram_mm_init(struct drm_vram_mm *vmm, struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	ret = ttm_bo_init_mm(&vmm->bdev, TTM_PL_VRAM, vram_size >> PAGE_SHIFT);
+	ret = ttm_range_man_init(&vmm->bdev, TTM_PL_VRAM,
+				 false, vram_size >> PAGE_SHIFT);
 	if (ret)
 		return ret;
 
@@ -1151,6 +1067,7 @@ static int drm_vram_mm_init(struct drm_vram_mm *vmm, struct drm_device *dev,
 
 static void drm_vram_mm_cleanup(struct drm_vram_mm *vmm)
 {
+	ttm_range_man_fini(&vmm->bdev, TTM_PL_VRAM);
 	ttm_bo_device_release(&vmm->bdev);
 }
 
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index 8e01caaf95cc..b65865c630b0 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -95,6 +95,7 @@ void drm_minor_release(struct drm_minor *minor);
 
 /* drm_managed.c */
 void drm_managed_release(struct drm_device *dev);
+void drmm_add_final_kfree(struct drm_device *dev, void *container);
 
 /* drm_vblank.c */
 static inline bool drm_vblank_passed(u64 seq, u64 ref)
diff --git a/drivers/gpu/drm/drm_managed.c b/drivers/gpu/drm/drm_managed.c
index 1e1356560c2e..37d7db6223be 100644
--- a/drivers/gpu/drm/drm_managed.c
+++ b/drivers/gpu/drm/drm_managed.c
@@ -27,7 +27,7 @@
  * be done directly with drmm_kmalloc() and the related functions. Everything
  * will be released on the final drm_dev_put() in reverse order of how the
  * release actions have been added and memory has been allocated since driver
- * loading started with drm_dev_init().
+ * loading started with devm_drm_dev_alloc().
  *
  * Note that release actions and managed memory can also be added and removed
  * during the lifetime of the driver, all the functions are fully concurrent
@@ -125,18 +125,6 @@ static void add_dr(struct drm_device *dev, struct drmres *dr)
 		       dr, dr->node.name, (unsigned long) dr->node.size);
 }
 
-/**
- * drmm_add_final_kfree - add release action for the final kfree()
- * @dev: DRM device
- * @container: pointer to the kmalloc allocation containing @dev
- *
- * Since the allocation containing the struct &drm_device must be allocated
- * before it can be initialized with drm_dev_init() there's no way to allocate
- * that memory with drmm_kmalloc(). To side-step this chicken-egg problem the
- * pointer for this final kfree() must be specified by calling this function. It
- * will be released in the final drm_dev_put() for @dev, after all other release
- * actions installed through drmm_add_action() have been processed.
- */
 void drmm_add_final_kfree(struct drm_device *dev, void *container)
 {
 	WARN_ON(dev->managed.final_kfree);
@@ -144,7 +132,6 @@ void drmm_add_final_kfree(struct drm_device *dev, void *container)
 	WARN_ON(dev + 1 > (struct drm_device *) (container + ksize(container)));
 	dev->managed.final_kfree = container;
 }
-EXPORT_SYMBOL(drmm_add_final_kfree);
 
 int __drmm_add_action(struct drm_device *dev,
 		      drmres_release_t action,
diff --git a/drivers/gpu/drm/drm_panel.c b/drivers/gpu/drm/drm_panel.c
index 8c7bac85a793..f634371c717a 100644
--- a/drivers/gpu/drm/drm_panel.c
+++ b/drivers/gpu/drm/drm_panel.c
@@ -70,16 +70,12 @@ EXPORT_SYMBOL(drm_panel_init);
  *
  * Add a panel to the global registry so that it can be looked up by display
  * drivers.
- *
- * Return: 0 on success or a negative error code on failure.
  */
-int drm_panel_add(struct drm_panel *panel)
+void drm_panel_add(struct drm_panel *panel)
 {
 	mutex_lock(&panel_lock);
 	list_add_tail(&panel->list, &panel_list);
 	mutex_unlock(&panel_lock);
-
-	return 0;
 }
 EXPORT_SYMBOL(drm_panel_add);
 
@@ -98,42 +94,6 @@ void drm_panel_remove(struct drm_panel *panel)
 EXPORT_SYMBOL(drm_panel_remove);
 
 /**
- * drm_panel_attach - attach a panel to a connector
- * @panel: DRM panel
- * @connector: DRM connector
- *
- * After obtaining a pointer to a DRM panel a display driver calls this
- * function to attach a panel to a connector.
- *
- * An error is returned if the panel is already attached to another connector.
- *
- * When unloading, the driver should detach from the panel by calling
- * drm_panel_detach().
- *
- * Return: 0 on success or a negative error code on failure.
- */
-int drm_panel_attach(struct drm_panel *panel, struct drm_connector *connector)
-{
-	return 0;
-}
-EXPORT_SYMBOL(drm_panel_attach);
-
-/**
- * drm_panel_detach - detach a panel from a connector
- * @panel: DRM panel
- *
- * Detaches a panel from the connector it is attached to. If a panel is not
- * attached to any connector this is effectively a no-op.
- *
- * This function should not be called by the panel device itself. It
- * is only for the drm device that called drm_panel_attach().
- */
-void drm_panel_detach(struct drm_panel *panel)
-{
-}
-EXPORT_SYMBOL(drm_panel_detach);
-
-/**
  * drm_panel_prepare - power on a panel
  * @panel: DRM panel
  *
@@ -300,6 +260,49 @@ struct drm_panel *of_drm_find_panel(const struct device_node *np)
 	return ERR_PTR(-EPROBE_DEFER);
 }
 EXPORT_SYMBOL(of_drm_find_panel);
+
+/**
+ * of_drm_get_panel_orientation - look up the orientation of the panel through
+ * the "rotation" binding from a device tree node
+ * @np: device tree node of the panel
+ * @orientation: orientation enum to be filled in
+ *
+ * Looks up the rotation of a panel in the device tree. The orientation of the
+ * panel is expressed as a property name "rotation" in the device tree. The
+ * rotation in the device tree is counter clockwise.
+ *
+ * Return: 0 when a valid rotation value (0, 90, 180, or 270) is read or the
+ * rotation property doesn't exist. Return a negative error code on failure.
+ */
+int of_drm_get_panel_orientation(const struct device_node *np,
+				 enum drm_panel_orientation *orientation)
+{
+	int rotation, ret;
+
+	ret = of_property_read_u32(np, "rotation", &rotation);
+	if (ret == -EINVAL) {
+		/* Don't return an error if there's no rotation property. */
+		*orientation = DRM_MODE_PANEL_ORIENTATION_UNKNOWN;
+		return 0;
+	}
+
+	if (ret < 0)
+		return ret;
+
+	if (rotation == 0)
+		*orientation = DRM_MODE_PANEL_ORIENTATION_NORMAL;
+	else if (rotation == 90)
+		*orientation = DRM_MODE_PANEL_ORIENTATION_RIGHT_UP;
+	else if (rotation == 180)
+		*orientation = DRM_MODE_PANEL_ORIENTATION_BOTTOM_UP;
+	else if (rotation == 270)
+		*orientation = DRM_MODE_PANEL_ORIENTATION_LEFT_UP;
+	else
+		return -EINVAL;
+
+	return 0;
+}
+EXPORT_SYMBOL(of_drm_get_panel_orientation);
 #endif
 
 #if IS_REACHABLE(CONFIG_BACKLIGHT_CLASS_DEVICE)
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 1693aa7c14b5..11fe9ff76fd5 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -617,6 +617,7 @@ struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach,
 {
 	struct drm_gem_object *obj = attach->dmabuf->priv;
 	struct sg_table *sgt;
+	int ret;
 
 	if (WARN_ON(dir == DMA_NONE))
 		return ERR_PTR(-EINVAL);
@@ -626,11 +627,12 @@ struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach,
 	else
 		sgt = obj->dev->driver->gem_prime_get_sg_table(obj);
 
-	if (!dma_map_sg_attrs(attach->dev, sgt->sgl, sgt->nents, dir,
-			      DMA_ATTR_SKIP_CPU_SYNC)) {
+	ret = dma_map_sgtable(attach->dev, sgt, dir,
+			      DMA_ATTR_SKIP_CPU_SYNC);
+	if (ret) {
 		sg_free_table(sgt);
 		kfree(sgt);
-		sgt = ERR_PTR(-ENOMEM);
+		sgt = ERR_PTR(ret);
 	}
 
 	return sgt;
@@ -652,8 +654,7 @@ void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach,
 	if (!sgt)
 		return;
 
-	dma_unmap_sg_attrs(attach->dev, sgt->sgl, sgt->nents, dir,
-			   DMA_ATTR_SKIP_CPU_SYNC);
+	dma_unmap_sgtable(attach->dev, sgt, dir, DMA_ATTR_SKIP_CPU_SYNC);
 	sg_free_table(sgt);
 	kfree(sgt);
 }
@@ -802,9 +803,11 @@ static const struct dma_buf_ops drm_gem_prime_dmabuf_ops =  {
  *
  * This is useful for implementing &drm_gem_object_funcs.get_sg_table.
  */
-struct sg_table *drm_prime_pages_to_sg(struct page **pages, unsigned int nr_pages)
+struct sg_table *drm_prime_pages_to_sg(struct drm_device *dev,
+				       struct page **pages, unsigned int nr_pages)
 {
 	struct sg_table *sg = NULL;
+	size_t max_segment = 0;
 	int ret;
 
 	sg = kmalloc(sizeof(struct sg_table), GFP_KERNEL);
@@ -813,8 +816,13 @@ struct sg_table *drm_prime_pages_to_sg(struct page **pages, unsigned int nr_page
 		goto out;
 	}
 
-	ret = sg_alloc_table_from_pages(sg, pages, nr_pages, 0,
-				nr_pages << PAGE_SHIFT, GFP_KERNEL);
+	if (dev)
+		max_segment = dma_max_mapping_size(dev->dev);
+	if (max_segment == 0 || max_segment > SCATTERLIST_MAX_SEGMENT)
+		max_segment = SCATTERLIST_MAX_SEGMENT;
+	ret = __sg_alloc_table_from_pages(sg, pages, nr_pages, 0,
+					  nr_pages << PAGE_SHIFT,
+					  max_segment, GFP_KERNEL);
 	if (ret)
 		goto out;
 
@@ -826,6 +834,37 @@ out:
 EXPORT_SYMBOL(drm_prime_pages_to_sg);
 
 /**
+ * drm_prime_get_contiguous_size - returns the contiguous size of the buffer
+ * @sgt: sg_table describing the buffer to check
+ *
+ * This helper calculates the contiguous size in the DMA address space
+ * of the the buffer described by the provided sg_table.
+ *
+ * This is useful for implementing
+ * &drm_gem_object_funcs.gem_prime_import_sg_table.
+ */
+unsigned long drm_prime_get_contiguous_size(struct sg_table *sgt)
+{
+	dma_addr_t expected = sg_dma_address(sgt->sgl);
+	struct scatterlist *sg;
+	unsigned long size = 0;
+	int i;
+
+	for_each_sgtable_dma_sg(sgt, sg, i) {
+		unsigned int len = sg_dma_len(sg);
+
+		if (!len)
+			break;
+		if (sg_dma_address(sg) != expected)
+			break;
+		expected += len;
+		size += len;
+	}
+	return size;
+}
+EXPORT_SYMBOL(drm_prime_get_contiguous_size);
+
+/**
  * drm_gem_prime_export - helper library implementation of the export callback
  * @obj: GEM object to export
  * @flags: flags like DRM_CLOEXEC and DRM_RDWR
@@ -959,45 +998,26 @@ EXPORT_SYMBOL(drm_gem_prime_import);
 int drm_prime_sg_to_page_addr_arrays(struct sg_table *sgt, struct page **pages,
 				     dma_addr_t *addrs, int max_entries)
 {
-	unsigned count;
-	struct scatterlist *sg;
-	struct page *page;
-	u32 page_len, page_index;
-	dma_addr_t addr;
-	u32 dma_len, dma_index;
-
-	/*
-	 * Scatterlist elements contains both pages and DMA addresses, but
-	 * one shoud not assume 1:1 relation between them. The sg->length is
-	 * the size of the physical memory chunk described by the sg->page,
-	 * while sg_dma_len(sg) is the size of the DMA (IO virtual) chunk
-	 * described by the sg_dma_address(sg).
-	 */
-	page_index = 0;
-	dma_index = 0;
-	for_each_sg(sgt->sgl, sg, sgt->nents, count) {
-		page_len = sg->length;
-		page = sg_page(sg);
-		dma_len = sg_dma_len(sg);
-		addr = sg_dma_address(sg);
-
-		while (pages && page_len > 0) {
-			if (WARN_ON(page_index >= max_entries))
+	struct sg_dma_page_iter dma_iter;
+	struct sg_page_iter page_iter;
+	struct page **p = pages;
+	dma_addr_t *a = addrs;
+
+	if (pages) {
+		for_each_sgtable_page(sgt, &page_iter, 0) {
+			if (WARN_ON(p - pages >= max_entries))
 				return -1;
-			pages[page_index] = page;
-			page++;
-			page_len -= PAGE_SIZE;
-			page_index++;
+			*p++ = sg_page_iter_page(&page_iter);
 		}
-		while (addrs && dma_len > 0) {
-			if (WARN_ON(dma_index >= max_entries))
+	}
+	if (addrs) {
+		for_each_sgtable_dma_page(sgt, &dma_iter, 0) {
+			if (WARN_ON(a - addrs >= max_entries))
 				return -1;
-			addrs[dma_index] = addr;
-			addr += PAGE_SIZE;
-			dma_len -= PAGE_SIZE;
-			dma_index++;
+			*a++ = sg_page_iter_dma_address(&dma_iter);
 		}
 	}
+
 	return 0;
 }
 EXPORT_SYMBOL(drm_prime_sg_to_page_addr_arrays);
diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
index 3bf73971daf3..6e74e6745eca 100644
--- a/drivers/gpu/drm/drm_syncobj.c
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -297,7 +297,7 @@ void drm_syncobj_add_point(struct drm_syncobj *syncobj,
 	prev = drm_syncobj_fence_get(syncobj);
 	/* You are adding an unorder point to timeline, which could cause payload returned from query_ioctl is 0! */
 	if (prev && prev->seqno >= point)
-		DRM_ERROR("You are adding an unorder point to timeline!\n");
+		DRM_DEBUG("You are adding an unorder point to timeline!\n");
 	dma_fence_chain_init(chain, prev, fence, point);
 	rcu_assign_pointer(syncobj->fence, &chain->base);
 
diff --git a/drivers/gpu/drm/drm_vblank.c b/drivers/gpu/drm/drm_vblank.c
index b18e1efbbae1..f135b79593dd 100644
--- a/drivers/gpu/drm/drm_vblank.c
+++ b/drivers/gpu/drm/drm_vblank.c
@@ -674,7 +674,7 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants);
  *
  * Note that atomic drivers must call drm_calc_timestamping_constants() before
  * enabling a CRTC. The atomic helpers already take care of that in
- * drm_atomic_helper_update_legacy_modeset_state().
+ * drm_atomic_helper_calc_timestamping_constants().
  *
  * Returns:
  *
@@ -819,7 +819,7 @@ EXPORT_SYMBOL(drm_crtc_vblank_helper_get_vblank_timestamp_internal);
  *
  * Note that atomic drivers must call drm_calc_timestamping_constants() before
  * enabling a CRTC. The atomic helpers already take care of that in
- * drm_atomic_helper_update_legacy_modeset_state().
+ * drm_atomic_helper_calc_timestamping_constants().
  *
  * Returns:
  *
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
index f06e19e7be04..d1533bdc1335 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
@@ -27,7 +27,7 @@ static void etnaviv_gem_scatter_map(struct etnaviv_gem_object *etnaviv_obj)
 	 * because display controller, GPU, etc. are not coherent.
 	 */
 	if (etnaviv_obj->flags & ETNA_BO_CACHE_MASK)
-		dma_map_sg(dev->dev, sgt->sgl, sgt->nents, DMA_BIDIRECTIONAL);
+		dma_map_sgtable(dev->dev, sgt, DMA_BIDIRECTIONAL, 0);
 }
 
 static void etnaviv_gem_scatterlist_unmap(struct etnaviv_gem_object *etnaviv_obj)
@@ -51,7 +51,7 @@ static void etnaviv_gem_scatterlist_unmap(struct etnaviv_gem_object *etnaviv_obj
 	 * discard those writes.
 	 */
 	if (etnaviv_obj->flags & ETNA_BO_CACHE_MASK)
-		dma_unmap_sg(dev->dev, sgt->sgl, sgt->nents, DMA_BIDIRECTIONAL);
+		dma_unmap_sgtable(dev->dev, sgt, DMA_BIDIRECTIONAL, 0);
 }
 
 /* called with etnaviv_obj->lock held */
@@ -103,7 +103,8 @@ struct page **etnaviv_gem_get_pages(struct etnaviv_gem_object *etnaviv_obj)
 		int npages = etnaviv_obj->base.size >> PAGE_SHIFT;
 		struct sg_table *sgt;
 
-		sgt = drm_prime_pages_to_sg(etnaviv_obj->pages, npages);
+		sgt = drm_prime_pages_to_sg(etnaviv_obj->base.dev,
+					    etnaviv_obj->pages, npages);
 		if (IS_ERR(sgt)) {
 			dev_err(dev->dev, "failed to allocate sgt: %ld\n",
 				PTR_ERR(sgt));
@@ -404,9 +405,8 @@ int etnaviv_gem_cpu_prep(struct drm_gem_object *obj, u32 op,
 	}
 
 	if (etnaviv_obj->flags & ETNA_BO_CACHED) {
-		dma_sync_sg_for_cpu(dev->dev, etnaviv_obj->sgt->sgl,
-				    etnaviv_obj->sgt->nents,
-				    etnaviv_op_to_dma_dir(op));
+		dma_sync_sgtable_for_cpu(dev->dev, etnaviv_obj->sgt,
+					 etnaviv_op_to_dma_dir(op));
 		etnaviv_obj->last_cpu_prep_op = op;
 	}
 
@@ -421,8 +421,7 @@ int etnaviv_gem_cpu_fini(struct drm_gem_object *obj)
 	if (etnaviv_obj->flags & ETNA_BO_CACHED) {
 		/* fini without a prep is almost certainly a userspace error */
 		WARN_ON(etnaviv_obj->last_cpu_prep_op == 0);
-		dma_sync_sg_for_device(dev->dev, etnaviv_obj->sgt->sgl,
-			etnaviv_obj->sgt->nents,
+		dma_sync_sgtable_for_device(dev->dev, etnaviv_obj->sgt,
 			etnaviv_op_to_dma_dir(etnaviv_obj->last_cpu_prep_op));
 		etnaviv_obj->last_cpu_prep_op = 0;
 	}
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
index 6d9e5c3c4dd5..4aa3426a9ba4 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
@@ -19,7 +19,7 @@ struct sg_table *etnaviv_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	if (WARN_ON(!etnaviv_obj->pages))  /* should have already pinned! */
 		return ERR_PTR(-EINVAL);
 
-	return drm_prime_pages_to_sg(etnaviv_obj->pages, npages);
+	return drm_prime_pages_to_sg(obj->dev, etnaviv_obj->pages, npages);
 }
 
 void *etnaviv_gem_prime_vmap(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
index 3607d348c298..15d9fa3879e5 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
@@ -73,13 +73,13 @@ static int etnaviv_iommu_map(struct etnaviv_iommu_context *context, u32 iova,
 			     struct sg_table *sgt, unsigned len, int prot)
 {	struct scatterlist *sg;
 	unsigned int da = iova;
-	unsigned int i, j;
+	unsigned int i;
 	int ret;
 
 	if (!context || !sgt)
 		return -EINVAL;
 
-	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
+	for_each_sgtable_dma_sg(sgt, sg, i) {
 		u32 pa = sg_dma_address(sg) - sg->offset;
 		size_t bytes = sg_dma_len(sg) + sg->offset;
 
@@ -95,14 +95,7 @@ static int etnaviv_iommu_map(struct etnaviv_iommu_context *context, u32 iova,
 	return 0;
 
 fail:
-	da = iova;
-
-	for_each_sg(sgt->sgl, sg, i, j) {
-		size_t bytes = sg_dma_len(sg) + sg->offset;
-
-		etnaviv_context_unmap(context, da, bytes);
-		da += bytes;
-	}
+	etnaviv_context_unmap(context, iova, da - iova);
 	return ret;
 }
 
@@ -113,7 +106,7 @@ static void etnaviv_iommu_unmap(struct etnaviv_iommu_context *context, u32 iova,
 	unsigned int da = iova;
 	int i;
 
-	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
+	for_each_sgtable_dma_sg(sgt, sg, i) {
 		size_t bytes = sg_dma_len(sg) + sg->offset;
 
 		etnaviv_context_unmap(context, da, bytes);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c
index 58b89ec11b0e..5887f7f52f96 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dma.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c
@@ -31,23 +31,6 @@
 #define EXYNOS_DEV_ADDR_START	0x20000000
 #define EXYNOS_DEV_ADDR_SIZE	0x40000000
 
-static inline int configure_dma_max_seg_size(struct device *dev)
-{
-	if (!dev->dma_parms)
-		dev->dma_parms = kzalloc(sizeof(*dev->dma_parms), GFP_KERNEL);
-	if (!dev->dma_parms)
-		return -ENOMEM;
-
-	dma_set_max_seg_size(dev, DMA_BIT_MASK(32));
-	return 0;
-}
-
-static inline void clear_dma_max_seg_size(struct device *dev)
-{
-	kfree(dev->dma_parms);
-	dev->dma_parms = NULL;
-}
-
 /*
  * drm_iommu_attach_device- attach device to iommu mapping
  *
@@ -69,10 +52,7 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev,
 		return -EINVAL;
 	}
 
-	ret = configure_dma_max_seg_size(subdrv_dev);
-	if (ret)
-		return ret;
-
+	dma_set_max_seg_size(subdrv_dev, DMA_BIT_MASK(32));
 	if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) {
 		/*
 		 * Keep the original DMA mapping of the sub-device and
@@ -89,9 +69,6 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev,
 		ret = iommu_attach_device(priv->mapping, subdrv_dev);
 	}
 
-	if (ret)
-		clear_dma_max_seg_size(subdrv_dev);
-
 	return ret;
 }
 
@@ -114,8 +91,6 @@ static void drm_iommu_detach_device(struct drm_device *drm_dev,
 		arm_iommu_attach_device(subdrv_dev, *dma_priv);
 	} else if (IS_ENABLED(CONFIG_IOMMU_DMA))
 		iommu_detach_device(priv->mapping, subdrv_dev);
-
-	clear_dma_max_seg_size(subdrv_dev);
 }
 
 int exynos_drm_register_dma(struct drm_device *drm, struct device *dev,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 7ba5354e7d94..741323a2e6c3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -42,11 +42,6 @@ static inline struct exynos_dpi *encoder_to_dpi(struct drm_encoder *e)
 static enum drm_connector_status
 exynos_dpi_detect(struct drm_connector *connector, bool force)
 {
-	struct exynos_dpi *ctx = connector_to_dpi(connector);
-
-	if (ctx->panel)
-		drm_panel_attach(ctx->panel, &ctx->connector);
-
 	return connector_status_connected;
 }
 
@@ -249,8 +244,5 @@ int exynos_dpi_remove(struct drm_encoder *encoder)
 
 	exynos_dpi_disable(&ctx->encoder);
 
-	if (ctx->panel)
-		drm_panel_detach(ctx->panel);
-
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index b38e9b592b8a..5b9666fc7af1 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -1551,12 +1551,10 @@ static int exynos_dsi_host_attach(struct mipi_dsi_host *host,
 		}
 
 		dsi->panel = of_drm_find_panel(device->dev.of_node);
-		if (IS_ERR(dsi->panel)) {
+		if (IS_ERR(dsi->panel))
 			dsi->panel = NULL;
-		} else {
-			drm_panel_attach(dsi->panel, &dsi->connector);
+		else
 			dsi->connector.status = connector_status_connected;
-		}
 	}
 
 	/*
@@ -1596,7 +1594,6 @@ static int exynos_dsi_host_detach(struct mipi_dsi_host *host,
 	if (dsi->panel) {
 		mutex_lock(&drm->mode_config.mutex);
 		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);
@@ -1763,11 +1760,8 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 	dsi->supplies[1].supply = "vddio";
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(dsi->supplies),
 				      dsi->supplies);
-	if (ret) {
-		if (ret != -EPROBE_DEFER)
-			dev_info(dev, "failed to get regulators: %d\n", ret);
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to get regulators\n");
 
 	dsi->clks = devm_kcalloc(dev,
 			dsi->driver_data->num_clks, sizeof(*dsi->clks),
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index 03be31427181..967a5cdc120e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -395,8 +395,8 @@ static void g2d_userptr_put_dma_addr(struct g2d_data *g2d,
 		return;
 
 out:
-	dma_unmap_sg(to_dma_dev(g2d->drm_dev), g2d_userptr->sgt->sgl,
-			g2d_userptr->sgt->nents, DMA_BIDIRECTIONAL);
+	dma_unmap_sgtable(to_dma_dev(g2d->drm_dev), g2d_userptr->sgt,
+			  DMA_BIDIRECTIONAL, 0);
 
 	pages = frame_vector_pages(g2d_userptr->vec);
 	if (!IS_ERR(pages)) {
@@ -511,10 +511,10 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d,
 
 	g2d_userptr->sgt = sgt;
 
-	if (!dma_map_sg(to_dma_dev(g2d->drm_dev), sgt->sgl, sgt->nents,
-				DMA_BIDIRECTIONAL)) {
+	ret = dma_map_sgtable(to_dma_dev(g2d->drm_dev), sgt,
+			      DMA_BIDIRECTIONAL, 0);
+	if (ret) {
 		DRM_DEV_ERROR(g2d->dev, "failed to map sgt with dma region.\n");
-		ret = -ENOMEM;
 		goto err_sg_free_table;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index efa476858db5..1716a023bca0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -431,27 +431,10 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev,
 {
 	struct exynos_drm_gem *exynos_gem;
 
-	if (sgt->nents < 1)
+	/* check if the entries in the sg_table are contiguous */
+	if (drm_prime_get_contiguous_size(sgt) < attach->dmabuf->size) {
+		DRM_ERROR("buffer chunks must be mapped contiguously");
 		return ERR_PTR(-EINVAL);
-
-	/*
-	 * Check if the provided buffer has been mapped as contiguous
-	 * into DMA address space.
-	 */
-	if (sgt->nents > 1) {
-		dma_addr_t next_addr = sg_dma_address(sgt->sgl);
-		struct scatterlist *s;
-		unsigned int i;
-
-		for_each_sg(sgt->sgl, s, sgt->nents, i) {
-			if (!sg_dma_len(s))
-				break;
-			if (sg_dma_address(s) != next_addr) {
-				DRM_ERROR("buffer chunks must be mapped contiguously");
-				return ERR_PTR(-EINVAL);
-			}
-			next_addr = sg_dma_address(s) + sg_dma_len(s);
-		}
 	}
 
 	exynos_gem = exynos_drm_gem_init(dev, attach->dmabuf->size);
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index c5ba32fca5f3..dc01c188c0e0 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1797,11 +1797,8 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 		hdata->regul_bulk[i].supply = supply[i];
 
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), hdata->regul_bulk);
-	if (ret) {
-		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to get regulators\n");
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to get regulators\n");
 
 	hdata->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en");
 
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c
index 9b0c4736c21a..4d4a715b429d 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c
@@ -40,10 +40,7 @@ int fsl_dcu_drm_encoder_create(struct fsl_dcu_drm_device *fsl_dev,
 
 static void fsl_dcu_drm_connector_destroy(struct drm_connector *connector)
 {
-	struct fsl_dcu_drm_connector *fsl_con = to_fsl_dcu_connector(connector);
-
 	drm_connector_unregister(connector);
-	drm_panel_detach(fsl_con->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -101,12 +98,6 @@ static int fsl_dcu_attach_panel(struct fsl_dcu_drm_device *fsl_dev,
 	if (ret < 0)
 		goto err_sysfs;
 
-	ret = drm_panel_attach(panel, connector);
-	if (ret) {
-		dev_err(fsl_dev->dev, "failed to attach panel\n");
-		goto err_sysfs;
-	}
-
 	return 0;
 
 err_sysfs:
diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c
index f41cbb753bb4..720a767118c9 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_dp.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c
@@ -2078,7 +2078,7 @@ cdv_intel_dp_init(struct drm_device *dev, struct psb_intel_mode_device *mode_dev
 					       intel_dp->dpcd,
 					       sizeof(intel_dp->dpcd));
 		cdv_intel_edp_panel_vdd_off(gma_encoder);
-		if (ret == 0) {
+		if (ret <= 0) {
 			/* if this fails, presume the device is a ghost */
 			DRM_INFO("failed to retrieve link info, disabling eDP\n");
 			drm_encoder_cleanup(encoder);
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index da02d7e8a8f5..54d9876b5305 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -164,7 +164,7 @@ static int psbfb_mmap(struct fb_info *info, struct vm_area_struct *vma)
 	return 0;
 }
 
-static struct fb_ops psbfb_ops = {
+static const struct fb_ops psbfb_ops = {
 	.owner = THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_setcolreg = psbfb_setcolreg,
@@ -175,7 +175,7 @@ static struct fb_ops psbfb_ops = {
 	.fb_sync = psbfb_sync,
 };
 
-static struct fb_ops psbfb_roll_ops = {
+static const struct fb_ops psbfb_roll_ops = {
 	.owner = THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_setcolreg = psbfb_setcolreg,
@@ -186,7 +186,7 @@ static struct fb_ops psbfb_roll_ops = {
 	.fb_mmap = psbfb_mmap,
 };
 
-static struct fb_ops psbfb_unaccel_ops = {
+static const struct fb_ops psbfb_unaccel_ops = {
 	.owner = THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_setcolreg = psbfb_setcolreg,
diff --git a/drivers/gpu/drm/gma500/mdfld_device.c b/drivers/gpu/drm/gma500/mdfld_device.c
index b718efccdcf2..be9cf6b1e3b3 100644
--- a/drivers/gpu/drm/gma500/mdfld_device.c
+++ b/drivers/gpu/drm/gma500/mdfld_device.c
@@ -6,6 +6,7 @@
  **************************************************************************/
 
 #include <linux/delay.h>
+#include <linux/gpio/machine.h>
 
 #include <asm/intel_scu_ipc.h>
 
@@ -505,12 +506,31 @@ static const struct psb_offset mdfld_regmap[3] = {
 	},
 };
 
+/*
+ * The GPIO lines for resetting DSI pipe 0 and 2 are available in the
+ * PCI device 0000:00:0c.0 on the Medfield.
+ */
+static struct gpiod_lookup_table mdfld_dsi_pipe_gpio_table = {
+	.table  = {
+		GPIO_LOOKUP("0000:00:0c.0", 128, "dsi-pipe0-reset",
+			    GPIO_ACTIVE_HIGH),
+		GPIO_LOOKUP("0000:00:0c.0", 34, "dsi-pipe2-reset",
+			    GPIO_ACTIVE_HIGH),
+		{ },
+	},
+};
+
 static int mdfld_chip_setup(struct drm_device *dev)
 {
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	if (pci_enable_msi(dev->pdev))
 		dev_warn(dev->dev, "Enabling MSI failed!\n");
 	dev_priv->regmap = mdfld_regmap;
+
+	/* Associate the GPIO lines with the DRM device */
+	mdfld_dsi_pipe_gpio_table.dev_id = dev_name(dev->dev);
+	gpiod_add_lookup_table(&mdfld_dsi_pipe_gpio_table);
+
 	return mid_chip_setup(dev);
 }
 
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
index c976a9dd9240..ae1223f631a7 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
@@ -955,7 +955,7 @@ struct mdfld_dsi_encoder *mdfld_dsi_dpi_init(struct drm_device *dev,
 
 		/* panel hard-reset */
 		if (p_funcs->reset) {
-			ret = p_funcs->reset(pipe);
+			ret = p_funcs->reset(dev, pipe);
 			if (ret) {
 				DRM_ERROR("Panel %d hard-reset failed\n", pipe);
 				return NULL;
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_output.c b/drivers/gpu/drm/gma500/mdfld_dsi_output.c
index f350ac1ead18..4aab76613bd9 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_output.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_output.c
@@ -28,6 +28,7 @@
 #include <linux/delay.h>
 #include <linux/moduleparam.h>
 #include <linux/pm_runtime.h>
+#include <linux/gpio/consumer.h>
 
 #include <asm/intel_scu_ipc.h>
 
@@ -366,7 +367,7 @@ static enum drm_mode_status mdfld_dsi_connector_mode_valid(struct drm_connector
 	/**
 	 * FIXME: current DC has no fitting unit, reject any mode setting
 	 * request
-	 * Will figure out a way to do up-scaling(pannel fitting) later.
+	 * Will figure out a way to do up-scaling(panel fitting) later.
 	 **/
 	if (fixed_mode) {
 		if (mode->hdisplay != fixed_mode->hdisplay)
@@ -432,42 +433,42 @@ static int mdfld_dsi_get_default_config(struct drm_device *dev,
 	return 0;
 }
 
-int mdfld_dsi_panel_reset(int pipe)
+int mdfld_dsi_panel_reset(struct drm_device *ddev, int pipe)
 {
-	unsigned gpio;
-	int ret = 0;
-
+	struct device *dev = ddev->dev;
+	struct gpio_desc *gpiod;
+
+	/*
+	 * Raise the GPIO reset line for the corresponding pipe to HIGH,
+	 * this is probably because it is active low so this takes the
+	 * respective pipe out of reset. (We have no code to put it back
+	 * into reset in this driver.)
+	 */
 	switch (pipe) {
 	case 0:
-		gpio = 128;
+		gpiod = gpiod_get(dev, "dsi-pipe0-reset", GPIOD_OUT_HIGH);
+		if (IS_ERR(gpiod))
+			return PTR_ERR(gpiod);
 		break;
 	case 2:
-		gpio = 34;
+		gpiod = gpiod_get(dev, "dsi-pipe2-reset", GPIOD_OUT_HIGH);
+		if (IS_ERR(gpiod))
+			return PTR_ERR(gpiod);
 		break;
 	default:
-		DRM_ERROR("Invalid output\n");
+		DRM_DEV_ERROR(dev, "Invalid output pipe\n");
 		return -EINVAL;
 	}
+	gpiod_put(gpiod);
 
-	ret = gpio_request(gpio, "gfx");
-	if (ret) {
-		DRM_ERROR("gpio_rqueset failed\n");
-		return ret;
-	}
-
-	ret = gpio_direction_output(gpio, 1);
-	if (ret) {
-		DRM_ERROR("gpio_direction_output failed\n");
-		goto gpio_error;
-	}
+	/* Flush posted writes on the device */
+	gpiod = gpiod_get(dev, "dsi-pipe0-reset", GPIOD_ASIS);
+	if (IS_ERR(gpiod))
+		return PTR_ERR(gpiod);
+	gpiod_get_value(gpiod);
+	gpiod_put(gpiod);
 
-	gpio_get_value(128);
-
-gpio_error:
-	if (gpio_is_valid(gpio))
-		gpio_free(gpio);
-
-	return ret;
+	return 0;
 }
 
 /*
@@ -531,7 +532,7 @@ void mdfld_dsi_output_init(struct drm_device *dev,
 	dsi_config->connector = dsi_connector;
 
 	if (!dsi_config->fixed_mode) {
-		DRM_ERROR("No pannel fixed mode was found\n");
+		DRM_ERROR("No panel fixed mode was found\n");
 		goto dsi_init_err0;
 	}
 
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_output.h b/drivers/gpu/drm/gma500/mdfld_dsi_output.h
index 0cccfe400a98..5c0db3c2903f 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_output.h
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_output.h
@@ -372,6 +372,6 @@ extern void mdfld_dsi_controller_init(struct mdfld_dsi_config *dsi_config,
 
 extern int mdfld_dsi_get_power_mode(struct mdfld_dsi_config *dsi_config,
 					u32 *mode, bool hs);
-extern int mdfld_dsi_panel_reset(int pipe);
+extern int mdfld_dsi_panel_reset(struct drm_device *dev, int pipe);
 
 #endif /*__MDFLD_DSI_OUTPUT_H__*/
diff --git a/drivers/gpu/drm/gma500/mdfld_output.h b/drivers/gpu/drm/gma500/mdfld_output.h
index 17a944d70add..37a516cc56be 100644
--- a/drivers/gpu/drm/gma500/mdfld_output.h
+++ b/drivers/gpu/drm/gma500/mdfld_output.h
@@ -54,7 +54,7 @@ struct panel_funcs {
 	const struct drm_encoder_helper_funcs *encoder_helper_funcs;
 	struct drm_display_mode * (*get_config_mode)(struct drm_device *);
 	int (*get_panel_info)(struct drm_device *, int, struct panel_info *);
-	int (*reset)(int pipe);
+	int (*reset)(struct drm_device *, int);
 	void (*drv_ic_init)(struct mdfld_dsi_config *dsi_config, int pipe);
 };
 
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index 3dd5718c3e31..5340225d6997 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -13,7 +13,6 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_vblank.h>
-#include <linux/gpio.h>
 #include "gma_display.h"
 
 /*
diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
index 06e44f47e73e..907f966d6f22 100644
--- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c
+++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
@@ -125,7 +125,7 @@ struct psb_intel_sdvo {
 	bool is_lvds;
 
 	/**
-	 * This is sdvo fixed pannel mode pointer
+	 * This is sdvo fixed panel mode pointer
 	 */
 	struct drm_display_mode *sdvo_lvds_fixed_mode;
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/Kconfig b/drivers/gpu/drm/hisilicon/hibmc/Kconfig
index dfc5aef62f7b..43943e980203 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/Kconfig
+++ b/drivers/gpu/drm/hisilicon/hibmc/Kconfig
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0-only
 config DRM_HISI_HIBMC
 	tristate "DRM Support for Hisilicon Hibmc"
-	depends on DRM && PCI && MMU && ARM64
+	depends on DRM && PCI && ARM64
 	select DRM_KMS_HELPER
 	select DRM_VRAM_HELPER
 	select DRM_TTM
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
index cc70e836522f..4d57ec688f82 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
@@ -17,9 +17,6 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_vram_helper.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_print.h>
-#include <drm/drm_probe_helper.h>
 #include <drm/drm_vblank.h>
 
 #include "hibmc_drm_drv.h"
@@ -74,12 +71,12 @@ static int hibmc_plane_atomic_check(struct drm_plane *plane,
 		return PTR_ERR(crtc_state);
 
 	if (src_w != state->crtc_w || src_h != state->crtc_h) {
-		DRM_DEBUG_ATOMIC("scale not support\n");
+		drm_dbg_atomic(plane->dev, "scale not support\n");
 		return -EINVAL;
 	}
 
 	if (state->crtc_x < 0 || state->crtc_y < 0) {
-		DRM_DEBUG_ATOMIC("crtc_x/y of drm_plane state is invalid\n");
+		drm_dbg_atomic(plane->dev, "crtc_x/y of drm_plane state is invalid\n");
 		return -EINVAL;
 	}
 
@@ -90,12 +87,12 @@ static int hibmc_plane_atomic_check(struct drm_plane *plane,
 	    crtc_state->adjusted_mode.hdisplay ||
 	    state->crtc_y + state->crtc_h >
 	    crtc_state->adjusted_mode.vdisplay) {
-		DRM_DEBUG_ATOMIC("visible portion of plane is invalid\n");
+		drm_dbg_atomic(plane->dev, "visible portion of plane is invalid\n");
 		return -EINVAL;
 	}
 
 	if (state->fb->pitches[0] % 128 != 0) {
-		DRM_DEBUG_ATOMIC("wrong stride with 128-byte aligned\n");
+		drm_dbg_atomic(plane->dev, "wrong stride with 128-byte aligned\n");
 		return -EINVAL;
 	}
 	return 0;
@@ -160,37 +157,6 @@ static const struct drm_plane_helper_funcs hibmc_plane_helper_funcs = {
 	.atomic_update = hibmc_plane_atomic_update,
 };
 
-static struct drm_plane *hibmc_plane_init(struct hibmc_drm_private *priv)
-{
-	struct drm_device *dev = priv->dev;
-	struct drm_plane *plane;
-	int ret = 0;
-
-	plane = devm_kzalloc(dev->dev, sizeof(*plane), GFP_KERNEL);
-	if (!plane) {
-		DRM_ERROR("failed to alloc memory when init plane\n");
-		return ERR_PTR(-ENOMEM);
-	}
-	/*
-	 * plane init
-	 * TODO: Now only support primary plane, overlay planes
-	 * need to do.
-	 */
-	ret = drm_universal_plane_init(dev, plane, 1, &hibmc_plane_funcs,
-				       channel_formats1,
-				       ARRAY_SIZE(channel_formats1),
-				       NULL,
-				       DRM_PLANE_TYPE_PRIMARY,
-				       NULL);
-	if (ret) {
-		DRM_ERROR("failed to init plane: %d\n", ret);
-		return ERR_PTR(ret);
-	}
-
-	drm_plane_helper_add(plane, &hibmc_plane_helper_funcs);
-	return plane;
-}
-
 static void hibmc_crtc_dpms(struct drm_crtc *crtc, int dpms)
 {
 	struct hibmc_drm_private *priv = crtc->dev->dev_private;
@@ -537,32 +503,34 @@ static const struct drm_crtc_helper_funcs hibmc_crtc_helper_funcs = {
 int hibmc_de_init(struct hibmc_drm_private *priv)
 {
 	struct drm_device *dev = priv->dev;
-	struct drm_crtc *crtc;
-	struct drm_plane *plane;
+	struct drm_crtc *crtc = &priv->crtc;
+	struct drm_plane *plane = &priv->primary_plane;
 	int ret;
 
-	plane = hibmc_plane_init(priv);
-	if (IS_ERR(plane)) {
-		DRM_ERROR("failed to create plane: %ld\n", PTR_ERR(plane));
-		return PTR_ERR(plane);
-	}
+	ret = drm_universal_plane_init(dev, plane, 1, &hibmc_plane_funcs,
+				       channel_formats1,
+				       ARRAY_SIZE(channel_formats1),
+				       NULL,
+				       DRM_PLANE_TYPE_PRIMARY,
+				       NULL);
 
-	crtc = devm_kzalloc(dev->dev, sizeof(*crtc), GFP_KERNEL);
-	if (!crtc) {
-		DRM_ERROR("failed to alloc memory when init crtc\n");
-		return -ENOMEM;
+	if (ret) {
+		drm_err(dev, "failed to init plane: %d\n", ret);
+		return ret;
 	}
 
+	drm_plane_helper_add(plane, &hibmc_plane_helper_funcs);
+
 	ret = drm_crtc_init_with_planes(dev, crtc, plane,
 					NULL, &hibmc_crtc_funcs, NULL);
 	if (ret) {
-		DRM_ERROR("failed to init crtc: %d\n", ret);
+		drm_err(dev, "failed to init crtc: %d\n", ret);
 		return ret;
 	}
 
 	ret = drm_mode_crtc_set_gamma_size(crtc, 256);
 	if (ret) {
-		DRM_ERROR("failed to set gamma size: %d\n", ret);
+		drm_err(dev, "failed to set gamma size: %d\n", ret);
 		return ret;
 	}
 	drm_crtc_helper_add(crtc, &hibmc_crtc_helper_funcs);
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index eea13e60187b..085d1b2fa8c0 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -11,18 +11,14 @@
  *	Jianhua Li <lijianhua@huawei.com>
  */
 
-#include <linux/console.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_vram_helper.h>
 #include <drm/drm_irq.h>
 #include <drm/drm_managed.h>
-#include <drm/drm_print.h>
-#include <drm/drm_probe_helper.h>
 #include <drm/drm_vblank.h>
 
 #include "hibmc_drm_drv.h"
@@ -102,13 +98,13 @@ static int hibmc_kms_init(struct hibmc_drm_private *priv)
 
 	ret = hibmc_de_init(priv);
 	if (ret) {
-		DRM_ERROR("failed to init de: %d\n", ret);
+		drm_err(priv->dev, "failed to init de: %d\n", ret);
 		return ret;
 	}
 
 	ret = hibmc_vdac_init(priv);
 	if (ret) {
-		DRM_ERROR("failed to init vdac: %d\n", ret);
+		drm_err(priv->dev, "failed to init vdac: %d\n", ret);
 		return ret;
 	}
 
@@ -216,7 +212,7 @@ static int hibmc_hw_map(struct hibmc_drm_private *priv)
 	iosize = pci_resource_len(pdev, 1);
 	priv->mmio = devm_ioremap(dev->dev, ioaddr, iosize);
 	if (!priv->mmio) {
-		DRM_ERROR("Cannot map mmio region\n");
+		drm_err(dev, "Cannot map mmio region\n");
 		return -ENOMEM;
 	}
 
@@ -224,7 +220,7 @@ static int hibmc_hw_map(struct hibmc_drm_private *priv)
 	size = pci_resource_len(pdev, 0);
 	priv->fb_map = devm_ioremap(dev->dev, addr, size);
 	if (!priv->fb_map) {
-		DRM_ERROR("Cannot map framebuffer\n");
+		drm_err(dev, "Cannot map framebuffer\n");
 		return -ENOMEM;
 	}
 	priv->fb_base = addr;
@@ -254,9 +250,8 @@ static int hibmc_unload(struct drm_device *dev)
 
 	if (dev->irq_enabled)
 		drm_irq_uninstall(dev);
-	if (priv->msi_enabled)
-		pci_disable_msi(dev->pdev);
 
+	pci_disable_msi(dev->pdev);
 	hibmc_kms_fini(priv);
 	hibmc_mm_fini(priv);
 	dev->dev_private = NULL;
@@ -270,7 +265,7 @@ static int hibmc_load(struct drm_device *dev)
 
 	priv = drmm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv) {
-		DRM_ERROR("no memory to allocate for hibmc_drm_private\n");
+		drm_err(dev, "no memory to allocate for hibmc_drm_private\n");
 		return -ENOMEM;
 	}
 	dev->dev_private = priv;
@@ -290,19 +285,17 @@ static int hibmc_load(struct drm_device *dev)
 
 	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
 	if (ret) {
-		DRM_ERROR("failed to initialize vblank: %d\n", ret);
+		drm_err(dev, "failed to initialize vblank: %d\n", ret);
 		goto err;
 	}
 
-	priv->msi_enabled = 0;
 	ret = pci_enable_msi(dev->pdev);
 	if (ret) {
-		DRM_WARN("enabling MSI failed: %d\n", ret);
+		drm_warn(dev, "enabling MSI failed: %d\n", ret);
 	} else {
-		priv->msi_enabled = 1;
 		ret = drm_irq_install(dev, dev->pdev->irq);
 		if (ret)
-			DRM_WARN("install irq failed: %d\n", ret);
+			drm_warn(dev, "install irq failed: %d\n", ret);
 	}
 
 	/* reset all the states of crtc/plane/encoder/connector */
@@ -312,7 +305,7 @@ static int hibmc_load(struct drm_device *dev)
 
 err:
 	hibmc_unload(dev);
-	DRM_ERROR("failed to initialize drm driver: %d\n", ret);
+	drm_err(dev, "failed to initialize drm driver: %d\n", ret);
 	return ret;
 }
 
@@ -338,19 +331,19 @@ static int hibmc_pci_probe(struct pci_dev *pdev,
 
 	ret = pci_enable_device(pdev);
 	if (ret) {
-		DRM_ERROR("failed to enable pci device: %d\n", ret);
+		drm_err(dev, "failed to enable pci device: %d\n", ret);
 		goto err_free;
 	}
 
 	ret = hibmc_load(dev);
 	if (ret) {
-		DRM_ERROR("failed to load hibmc: %d\n", ret);
+		drm_err(dev, "failed to load hibmc: %d\n", ret);
 		goto err_disable;
 	}
 
 	ret = drm_dev_register(dev, 0);
 	if (ret) {
-		DRM_ERROR("failed to register drv for userspace access: %d\n",
+		drm_err(dev, "failed to register drv for userspace access: %d\n",
 			  ret);
 		goto err_unload;
 	}
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
index 609768748de6..197485e2fe0b 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
@@ -25,10 +25,11 @@ struct hibmc_drm_private {
 	void __iomem   *fb_map;
 	unsigned long  fb_base;
 	unsigned long  fb_size;
-	bool msi_enabled;
 
 	/* drm */
 	struct drm_device  *dev;
+	struct drm_plane primary_plane;
+	struct drm_crtc crtc;
 	struct drm_encoder encoder;
 	struct drm_connector connector;
 	bool mode_config_initialized;
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index 2ca69c38491a..376a05ddbc2f 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -11,10 +11,8 @@
  *	Jianhua Li <lijianhua@huawei.com>
  */
 
-#include <drm/drm_gem_vram_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_print.h>
 
 #include "hibmc_drm_drv.h"
@@ -87,7 +85,7 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 	ret = drm_encoder_init(dev, encoder, &hibmc_encoder_funcs,
 			       DRM_MODE_ENCODER_DAC, NULL);
 	if (ret) {
-		DRM_ERROR("failed to init encoder: %d\n", ret);
+		drm_err(dev, "failed to init encoder: %d\n", ret);
 		return ret;
 	}
 
@@ -96,7 +94,7 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 	ret = drm_connector_init(dev, connector, &hibmc_connector_funcs,
 				 DRM_MODE_CONNECTOR_VGA);
 	if (ret) {
-		DRM_ERROR("failed to init connector: %d\n", ret);
+		drm_err(dev, "failed to init connector: %d\n", ret);
 		return ret;
 	}
 	drm_connector_helper_add(connector, &hibmc_connector_helper_funcs);
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
index 322bd542e89d..602ece11bb4a 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
@@ -32,7 +32,7 @@ int hibmc_mm_init(struct hibmc_drm_private *hibmc)
 				       hibmc->fb_size);
 	if (IS_ERR(vmm)) {
 		ret = PTR_ERR(vmm);
-		DRM_ERROR("Error initializing VRAM MM; %d\n", ret);
+		drm_err(dev, "Error initializing VRAM MM; %d\n", ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index 303c2d483c6e..88250860f8e4 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -853,11 +853,11 @@ static void i810_dma_quiescent(struct drm_device *dev)
 	i810_wait_ring(dev, dev_priv->ring.Size - 8);
 }
 
-static int i810_flush_queue(struct drm_device *dev)
+static void i810_flush_queue(struct drm_device *dev)
 {
 	drm_i810_private_t *dev_priv = dev->dev_private;
 	struct drm_device_dma *dma = dev->dma;
-	int i, ret = 0;
+	int i;
 	RING_LOCALS;
 
 	i810_kernel_lost_context(dev);
@@ -882,7 +882,7 @@ static int i810_flush_queue(struct drm_device *dev)
 			DRM_DEBUG("still on client\n");
 	}
 
-	return ret;
+	return;
 }
 
 /* Must be called with the lock held */
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index bda4c0e408f8..e5574e506a5c 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -234,6 +234,7 @@ i915-y += \
 	display/intel_ddi.o \
 	display/intel_dp.o \
 	display/intel_dp_aux_backlight.o \
+	display/intel_dp_hdcp.o \
 	display/intel_dp_link_training.o \
 	display/intel_dp_mst.o \
 	display/intel_dsi.o \
diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index f4053dd6bde9..520715b7d5b5 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -1646,6 +1646,7 @@ static const struct drm_encoder_funcs gen11_dsi_encoder_funcs = {
 };
 
 static const struct drm_connector_funcs gen11_dsi_connector_funcs = {
+	.detect = intel_panel_detect,
 	.late_register = intel_connector_register,
 	.early_unregister = intel_connector_unregister,
 	.destroy = intel_connector_destroy,
diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
index 630f49b7aa01..86be032bcf96 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic.c
@@ -527,8 +527,6 @@ void intel_atomic_state_clear(struct drm_atomic_state *s)
 	intel_atomic_clear_global_state(state);
 
 	state->dpll_set = state->modeset = false;
-	state->global_state_changed = false;
-	state->active_pipes = 0;
 }
 
 struct intel_crtc_state *
@@ -542,40 +540,3 @@ intel_atomic_get_crtc_state(struct drm_atomic_state *state,
 
 	return to_intel_crtc_state(crtc_state);
 }
-
-int _intel_atomic_lock_global_state(struct intel_atomic_state *state)
-{
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
-	struct intel_crtc *crtc;
-
-	state->global_state_changed = true;
-
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
-		int ret;
-
-		ret = drm_modeset_lock(&crtc->base.mutex,
-				       state->base.acquire_ctx);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-int _intel_atomic_serialize_global_state(struct intel_atomic_state *state)
-{
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
-	struct intel_crtc *crtc;
-
-	state->global_state_changed = true;
-
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
-		struct intel_crtc_state *crtc_state;
-
-		crtc_state = intel_atomic_get_crtc_state(&state->base, crtc);
-		if (IS_ERR(crtc_state))
-			return PTR_ERR(crtc_state);
-	}
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/i915/display/intel_atomic.h b/drivers/gpu/drm/i915/display/intel_atomic.h
index 11146292b06f..285de07011dc 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.h
+++ b/drivers/gpu/drm/i915/display/intel_atomic.h
@@ -56,8 +56,4 @@ int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
 			       struct intel_crtc *intel_crtc,
 			       struct intel_crtc_state *crtc_state);
 
-int _intel_atomic_lock_global_state(struct intel_atomic_state *state);
-
-int _intel_atomic_serialize_global_state(struct intel_atomic_state *state);
-
 #endif /* __INTEL_ATOMIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c
index ad4aa66fd676..f7de55707746 100644
--- a/drivers/gpu/drm/i915/display/intel_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_audio.c
@@ -958,13 +958,8 @@ static int glk_force_audio_cdclk_commit(struct intel_atomic_state *state,
 	if (IS_ERR(cdclk_state))
 		return PTR_ERR(cdclk_state);
 
-	cdclk_state->force_min_cdclk_changed = true;
 	cdclk_state->force_min_cdclk = enable ? 2 * 96000 : 0;
 
-	ret = intel_atomic_lock_global_state(&cdclk_state->base);
-	if (ret)
-		return ret;
-
 	return drm_atomic_commit(&state->base);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index a0a41ec5c341..4716484af62d 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -1656,6 +1656,8 @@ static enum port dvo_port_to_port(struct drm_i915_private *dev_priv,
 		[PORT_E] = { DVO_PORT_HDMIE, DVO_PORT_DPE, DVO_PORT_CRT },
 		[PORT_F] = { DVO_PORT_HDMIF, DVO_PORT_DPF, -1 },
 		[PORT_G] = { DVO_PORT_HDMIG, DVO_PORT_DPG, -1 },
+		[PORT_H] = { DVO_PORT_HDMIH, DVO_PORT_DPH, -1 },
+		[PORT_I] = { DVO_PORT_HDMII, DVO_PORT_DPI, -1 },
 	};
 	/*
 	 * Bspec lists the ports as A, B, C, D - however internally in our
@@ -2133,7 +2135,7 @@ void intel_bios_init(struct drm_i915_private *dev_priv)
 
 	INIT_LIST_HEAD(&dev_priv->vbt.display_devices);
 
-	if (!HAS_DISPLAY(dev_priv) || !INTEL_DISPLAY_ENABLED(dev_priv)) {
+	if (!HAS_DISPLAY(dev_priv)) {
 		drm_dbg_kms(&dev_priv->drm,
 			    "Skipping VBT init due to disabled display.\n");
 		return;
@@ -2650,6 +2652,12 @@ enum aux_ch intel_bios_port_aux_ch(struct drm_i915_private *dev_priv,
 	case DP_AUX_G:
 		aux_ch = AUX_CH_G;
 		break;
+	case DP_AUX_H:
+		aux_ch = AUX_CH_H;
+		break;
+	case DP_AUX_I:
+		aux_ch = AUX_CH_I;
+		break;
 	default:
 		MISSING_CASE(info->alternate_aux_channel);
 		aux_ch = AUX_CH_A;
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 91a8161e7c05..cb93f6cf6d37 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -2426,7 +2426,6 @@ static struct intel_global_state *intel_cdclk_duplicate_state(struct intel_globa
 	if (!cdclk_state)
 		return NULL;
 
-	cdclk_state->force_min_cdclk_changed = false;
 	cdclk_state->pipe = INVALID_PIPE;
 
 	return &cdclk_state->base;
@@ -2501,6 +2500,7 @@ int intel_modeset_calc_cdclk(struct intel_atomic_state *state)
 		if (ret)
 			return ret;
 	} else if (old_cdclk_state->active_pipes != new_cdclk_state->active_pipes ||
+		   old_cdclk_state->force_min_cdclk != new_cdclk_state->force_min_cdclk ||
 		   intel_cdclk_changed(&old_cdclk_state->logical,
 				       &new_cdclk_state->logical)) {
 		ret = intel_atomic_lock_global_state(&new_cdclk_state->base);
@@ -2677,7 +2677,7 @@ void intel_update_cdclk(struct drm_i915_private *dev_priv)
 	 */
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 		intel_de_write(dev_priv, GMBUSFREQ_VLV,
-		               DIV_ROUND_UP(dev_priv->cdclk.hw.cdclk, 1000));
+			       DIV_ROUND_UP(dev_priv->cdclk.hw.cdclk, 1000));
 }
 
 static int cnp_rawclk(struct drm_i915_private *dev_priv)
@@ -2903,9 +2903,10 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv)
 		dev_priv->display.get_cdclk = i85x_get_cdclk;
 	else if (IS_I845G(dev_priv))
 		dev_priv->display.get_cdclk = fixed_200mhz_get_cdclk;
-	else { /* 830 */
-		drm_WARN(&dev_priv->drm, !IS_I830(dev_priv),
-			 "Unknown platform. Assuming 133 MHz CDCLK\n");
+	else if (IS_I830(dev_priv))
+		dev_priv->display.get_cdclk = fixed_133mhz_get_cdclk;
+
+	if (drm_WARN(&dev_priv->drm, !dev_priv->display.get_cdclk,
+		     "Unknown platform. Assuming 133 MHz CDCLK\n"))
 		dev_priv->display.get_cdclk = fixed_133mhz_get_cdclk;
-	}
 }
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.h b/drivers/gpu/drm/i915/display/intel_cdclk.h
index 5731806e4cee..b34eb00fb327 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.h
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.h
@@ -17,8 +17,8 @@ struct intel_atomic_state;
 struct intel_crtc_state;
 
 struct intel_cdclk_vals {
-	u16 refclk;
 	u32 cdclk;
+	u16 refclk;
 	u8 divider;	/* CD2X divider * 2 */
 	u8 ratio;
 };
@@ -49,7 +49,6 @@ struct intel_cdclk_state {
 
 	/* forced minimum cdclk for glk+ audio w/a */
 	int force_min_cdclk;
-	bool force_min_cdclk_changed;
 
 	/* bitmask of active pipes */
 	u8 active_pipes;
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 5b4510ce5693..4934edd51cb0 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -833,6 +833,9 @@ intel_crt_detect(struct drm_connector *connector,
 		    connector->base.id, connector->name,
 		    force);
 
+	if (!INTEL_DISPLAY_ENABLED(dev_priv))
+		return connector_status_disconnected;
+
 	if (dev_priv->params.load_detect_test) {
 		wakeref = intel_display_power_get(dev_priv,
 						  intel_encoder->power_domain);
diff --git a/drivers/gpu/drm/i915/display/intel_csr.c b/drivers/gpu/drm/i915/display/intel_csr.c
index f22a7645c249..d5db16764619 100644
--- a/drivers/gpu/drm/i915/display/intel_csr.c
+++ b/drivers/gpu/drm/i915/display/intel_csr.c
@@ -40,12 +40,12 @@
 
 #define GEN12_CSR_MAX_FW_SIZE		ICL_CSR_MAX_FW_SIZE
 
-#define RKL_CSR_PATH			"i915/rkl_dmc_ver2_01.bin"
-#define RKL_CSR_VERSION_REQUIRED	CSR_VERSION(2, 1)
+#define RKL_CSR_PATH			"i915/rkl_dmc_ver2_02.bin"
+#define RKL_CSR_VERSION_REQUIRED	CSR_VERSION(2, 2)
 MODULE_FIRMWARE(RKL_CSR_PATH);
 
-#define TGL_CSR_PATH			"i915/tgl_dmc_ver2_06.bin"
-#define TGL_CSR_VERSION_REQUIRED	CSR_VERSION(2, 6)
+#define TGL_CSR_PATH			"i915/tgl_dmc_ver2_08.bin"
+#define TGL_CSR_VERSION_REQUIRED	CSR_VERSION(2, 8)
 #define TGL_CSR_MAX_FW_SIZE		0x6000
 MODULE_FIRMWARE(TGL_CSR_PATH);
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index a49ff3a1a63c..4d06178cd76c 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -572,13 +572,13 @@ static const struct cnl_ddi_buf_trans ehl_combo_phy_ddi_translations_dp[] = {
 						/* NT mV Trans mV db    */
 	{ 0xA, 0x33, 0x3F, 0x00, 0x00 },	/* 350   350      0.0   */
 	{ 0xA, 0x47, 0x36, 0x00, 0x09 },	/* 350   500      3.1   */
-	{ 0xC, 0x64, 0x30, 0x00, 0x0F },	/* 350   700      6.0   */
-	{ 0x6, 0x7F, 0x2C, 0x00, 0x13 },	/* 350   900      8.2   */
+	{ 0xC, 0x64, 0x34, 0x00, 0x0B },	/* 350   700      6.0   */
+	{ 0x6, 0x7F, 0x30, 0x00, 0x0F },	/* 350   900      8.2   */
 	{ 0xA, 0x46, 0x3F, 0x00, 0x00 },	/* 500   500      0.0   */
-	{ 0xC, 0x64, 0x36, 0x00, 0x09 },	/* 500   700      2.9   */
-	{ 0x6, 0x7F, 0x30, 0x00, 0x0F },	/* 500   900      5.1   */
+	{ 0xC, 0x64, 0x38, 0x00, 0x07 },	/* 500   700      2.9   */
+	{ 0x6, 0x7F, 0x32, 0x00, 0x0D },	/* 500   900      5.1   */
 	{ 0xC, 0x61, 0x3F, 0x00, 0x00 },	/* 650   700      0.6   */
-	{ 0x6, 0x7F, 0x37, 0x00, 0x08 },	/* 600   900      3.5   */
+	{ 0x6, 0x7F, 0x38, 0x00, 0x07 },	/* 600   900      3.5   */
 	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 900   900      0.0   */
 };
 
@@ -706,6 +706,42 @@ static const struct cnl_ddi_buf_trans tgl_combo_phy_ddi_translations_dp_hbr2[] =
 	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 900   900      0.0   */
 };
 
+static const struct cnl_ddi_buf_trans tgl_uy_combo_phy_ddi_translations_dp_hbr2[] = {
+						/* NT mV Trans mV db    */
+	{ 0xA, 0x35, 0x3F, 0x00, 0x00 },	/* 350   350      0.0   */
+	{ 0xA, 0x4F, 0x36, 0x00, 0x09 },	/* 350   500      3.1   */
+	{ 0xC, 0x60, 0x32, 0x00, 0x0D },	/* 350   700      6.0   */
+	{ 0xC, 0x7F, 0x2D, 0x00, 0x12 },	/* 350   900      8.2   */
+	{ 0xC, 0x47, 0x3F, 0x00, 0x00 },	/* 500   500      0.0   */
+	{ 0xC, 0x6F, 0x36, 0x00, 0x09 },	/* 500   700      2.9   */
+	{ 0x6, 0x7D, 0x32, 0x00, 0x0D },	/* 500   900      5.1   */
+	{ 0x6, 0x60, 0x3C, 0x00, 0x03 },	/* 650   700      0.6   */
+	{ 0x6, 0x7F, 0x34, 0x00, 0x0B },	/* 600   900      3.5   */
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 900   900      0.0   */
+};
+
+/*
+ * Cloned the HOBL entry to comply with the voltage and pre-emphasis entries
+ * that DisplayPort specification requires
+ */
+static const struct cnl_ddi_buf_trans tgl_combo_phy_ddi_translations_edp_hbr2_hobl[] = {
+						/* VS	pre-emp	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 0	0	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 0	1	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 0	2	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 0	3	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 1	0	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 1	1	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 1	2	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 2	0	*/
+	{ 0x6, 0x7F, 0x3F, 0x00, 0x00 },	/* 2	1	*/
+};
+
+static bool is_hobl_buf_trans(const struct cnl_ddi_buf_trans *table)
+{
+	return table == tgl_combo_phy_ddi_translations_edp_hbr2_hobl;
+}
+
 static const struct ddi_buf_trans *
 bdw_get_buf_trans_edp(struct intel_encoder *encoder, int *n_entries)
 {
@@ -1038,27 +1074,74 @@ static const struct cnl_ddi_buf_trans *
 ehl_get_combo_buf_trans(struct intel_encoder *encoder, int type, int rate,
 			int *n_entries)
 {
-	if (type != INTEL_OUTPUT_HDMI && type != INTEL_OUTPUT_EDP) {
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+
+	switch (type) {
+	case INTEL_OUTPUT_HDMI:
+		*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_hdmi);
+		return icl_combo_phy_ddi_translations_hdmi;
+	case INTEL_OUTPUT_EDP:
+		if (dev_priv->vbt.edp.low_vswing) {
+			if (rate > 540000) {
+				*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_edp_hbr3);
+				return icl_combo_phy_ddi_translations_edp_hbr3;
+			} else {
+				*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_edp_hbr2);
+				return icl_combo_phy_ddi_translations_edp_hbr2;
+			}
+		}
+		/* fall through */
+	default:
+		/* All combo DP and eDP ports that do not support low_vswing */
 		*n_entries = ARRAY_SIZE(ehl_combo_phy_ddi_translations_dp);
 		return ehl_combo_phy_ddi_translations_dp;
 	}
-
-	return icl_get_combo_buf_trans(encoder, type, rate, n_entries);
 }
 
 static const struct cnl_ddi_buf_trans *
 tgl_get_combo_buf_trans(struct intel_encoder *encoder, int type, int rate,
 			int *n_entries)
 {
-	if (type == INTEL_OUTPUT_HDMI || type == INTEL_OUTPUT_EDP) {
-		return icl_get_combo_buf_trans(encoder, type, rate, n_entries);
-	} else if (rate > 270000) {
-		*n_entries = ARRAY_SIZE(tgl_combo_phy_ddi_translations_dp_hbr2);
-		return tgl_combo_phy_ddi_translations_dp_hbr2;
-	}
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+
+	switch (type) {
+	case INTEL_OUTPUT_HDMI:
+		*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_hdmi);
+		return icl_combo_phy_ddi_translations_hdmi;
+	case INTEL_OUTPUT_EDP:
+		if (dev_priv->vbt.edp.hobl) {
+			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+			if (!intel_dp->hobl_failed && rate <= 540000) {
+				/* Same table applies to TGL, RKL and DG1 */
+				*n_entries = ARRAY_SIZE(tgl_combo_phy_ddi_translations_edp_hbr2_hobl);
+				return tgl_combo_phy_ddi_translations_edp_hbr2_hobl;
+			}
+		}
+
+		if (rate > 540000) {
+			*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_edp_hbr3);
+			return icl_combo_phy_ddi_translations_edp_hbr3;
+		} else if (dev_priv->vbt.edp.low_vswing) {
+			*n_entries = ARRAY_SIZE(icl_combo_phy_ddi_translations_edp_hbr2);
+			return icl_combo_phy_ddi_translations_edp_hbr2;
+		}
+		/* fall through */
+	default:
+		/* All combo DP and eDP ports that do not support low_vswing */
+		if (rate > 270000) {
+			if (IS_TGL_U(dev_priv) || IS_TGL_Y(dev_priv)) {
+				*n_entries = ARRAY_SIZE(tgl_uy_combo_phy_ddi_translations_dp_hbr2);
+				return tgl_uy_combo_phy_ddi_translations_dp_hbr2;
+			}
+
+			*n_entries = ARRAY_SIZE(tgl_combo_phy_ddi_translations_dp_hbr2);
+			return tgl_combo_phy_ddi_translations_dp_hbr2;
+		}
 
-	*n_entries = ARRAY_SIZE(tgl_combo_phy_ddi_translations_dp_hbr);
-	return tgl_combo_phy_ddi_translations_dp_hbr;
+		*n_entries = ARRAY_SIZE(tgl_combo_phy_ddi_translations_dp_hbr);
+		return tgl_combo_phy_ddi_translations_dp_hbr;
+	}
 }
 
 static const struct tgl_dkl_phy_ddi_buf_trans *
@@ -1738,6 +1821,8 @@ void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state
 
 	ctl = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(cpu_transcoder));
 
+	drm_WARN_ON(crtc->base.dev, ctl & TRANS_DDI_HDCP_SIGNALLING);
+
 	ctl &= ~TRANS_DDI_FUNC_ENABLE;
 
 	if (IS_GEN_RANGE(dev_priv, 8, 10))
@@ -1765,12 +1850,12 @@ void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state
 }
 
 int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder,
+				     enum transcoder cpu_transcoder,
 				     bool enable)
 {
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	intel_wakeref_t wakeref;
-	enum pipe pipe = 0;
 	int ret = 0;
 	u32 tmp;
 
@@ -1779,19 +1864,12 @@ int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder,
 	if (drm_WARN_ON(dev, !wakeref))
 		return -ENXIO;
 
-	if (drm_WARN_ON(dev,
-			!intel_encoder->get_hw_state(intel_encoder, &pipe))) {
-		ret = -EIO;
-		goto out;
-	}
-
-	tmp = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(pipe));
+	tmp = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(cpu_transcoder));
 	if (enable)
 		tmp |= TRANS_DDI_HDCP_SIGNALLING;
 	else
 		tmp &= ~TRANS_DDI_HDCP_SIGNALLING;
-	intel_de_write(dev_priv, TRANS_DDI_FUNC_CTL(pipe), tmp);
-out:
+	intel_de_write(dev_priv, TRANS_DDI_FUNC_CTL(cpu_transcoder), tmp);
 	intel_display_power_put(dev_priv, intel_encoder->power_domain, wakeref);
 	return ret;
 }
@@ -2392,6 +2470,15 @@ static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 		level = n_entries - 1;
 	}
 
+	if (type == INTEL_OUTPUT_EDP) {
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+		val = EDP4K2K_MODE_OVRD_EN | EDP4K2K_MODE_OVRD_OPTIMIZED;
+		intel_dp->hobl_active = is_hobl_buf_trans(ddi_translations);
+		intel_de_rmw(dev_priv, ICL_PORT_CL_DW10(phy), val,
+			     intel_dp->hobl_active ? val : 0);
+	}
+
 	/* Set PORT_TX_DW5 */
 	val = intel_de_read(dev_priv, ICL_PORT_TX_DW5_LN0(phy));
 	val &= ~(SCALING_MODE_SEL_MASK | RTERM_SELECT_MASK |
@@ -2802,7 +2889,9 @@ hsw_set_signal_levels(struct intel_dp *intel_dp)
 static u32 icl_dpclka_cfgcr0_clk_off(struct drm_i915_private *dev_priv,
 				     enum phy phy)
 {
-	if (intel_phy_is_combo(dev_priv, phy)) {
+	if (IS_ROCKETLAKE(dev_priv)) {
+		return RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy);
+	} else if (intel_phy_is_combo(dev_priv, phy)) {
 		return ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy);
 	} else if (intel_phy_is_tc(dev_priv, phy)) {
 		enum tc_port tc_port = intel_port_to_tc(dev_priv,
@@ -2829,6 +2918,16 @@ static void icl_map_plls_to_ports(struct intel_encoder *encoder,
 		    (val & icl_dpclka_cfgcr0_clk_off(dev_priv, phy)) == 0);
 
 	if (intel_phy_is_combo(dev_priv, phy)) {
+		u32 mask, sel;
+
+		if (IS_ROCKETLAKE(dev_priv)) {
+			mask = RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
+			sel = RKL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy);
+		} else {
+			mask = ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
+			sel = ICL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy);
+		}
+
 		/*
 		 * Even though this register references DDIs, note that we
 		 * want to pass the PHY rather than the port (DDI).  For
@@ -2839,8 +2938,8 @@ static void icl_map_plls_to_ports(struct intel_encoder *encoder,
 		 *   Clock Select chooses the PLL for both DDIA and DDID and
 		 *   drives port A in all cases."
 		 */
-		val &= ~ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
-		val |= ICL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy);
+		val &= ~mask;
+		val |= sel;
 		intel_de_write(dev_priv, ICL_DPCLKA_CFGCR0, val);
 		intel_de_posting_read(dev_priv, ICL_DPCLKA_CFGCR0);
 	}
@@ -3371,6 +3470,7 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state,
 	intel_ddi_init_dp_buf_reg(encoder);
 	if (!is_mst)
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+	intel_dp_configure_protocol_converter(intel_dp);
 	intel_dp_sink_set_decompression_state(intel_dp, crtc_state,
 					      true);
 	intel_dp_sink_set_fec_ready(intel_dp, crtc_state);
@@ -3482,19 +3582,17 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state,
 		intel_ddi_pre_enable_hdmi(state, encoder, crtc_state,
 					  conn_state);
 	} else {
-		struct intel_lspcon *lspcon =
-				enc_to_intel_lspcon(encoder);
+		struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 
 		intel_ddi_pre_enable_dp(state, encoder, crtc_state,
 					conn_state);
-		if (lspcon->active) {
-			struct intel_digital_port *dig_port =
-					enc_to_dig_port(encoder);
 
+		/* FIXME precompute everything properly */
+		/* FIXME how do we turn infoframes off again? */
+		if (dig_port->lspcon.active && dig_port->dp.has_hdmi_sink)
 			dig_port->set_infoframes(encoder,
 						 crtc_state->has_infoframe,
 						 crtc_state, conn_state);
-		}
 	}
 }
 
@@ -3938,18 +4036,19 @@ static void intel_ddi_update_pipe_dp(struct intel_atomic_state *state,
 
 	intel_psr_update(intel_dp, crtc_state, conn_state);
 	intel_dp_set_infoframes(encoder, true, crtc_state, conn_state);
-	intel_edp_drrs_enable(intel_dp, crtc_state);
+	intel_edp_drrs_update(intel_dp, crtc_state);
 
 	intel_panel_update_backlight(state, encoder, crtc_state, conn_state);
 }
 
-static void intel_ddi_update_pipe(struct intel_atomic_state *state,
-				  struct intel_encoder *encoder,
-				  const struct intel_crtc_state *crtc_state,
-				  const struct drm_connector_state *conn_state)
+void intel_ddi_update_pipe(struct intel_atomic_state *state,
+			   struct intel_encoder *encoder,
+			   const struct intel_crtc_state *crtc_state,
+			   const struct drm_connector_state *conn_state)
 {
 
-	if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
+	if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) &&
+	    !intel_encoder_is_mst(encoder))
 		intel_ddi_update_pipe_dp(state, encoder, crtc_state,
 					 conn_state);
 
@@ -4037,8 +4136,7 @@ static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp)
 			intel_wait_ddi_buf_idle(dev_priv, port);
 	}
 
-	dp_tp_ctl = DP_TP_CTL_ENABLE |
-		    DP_TP_CTL_LINK_TRAIN_PAT1 | DP_TP_CTL_SCRAMBLE_DISABLE;
+	dp_tp_ctl = DP_TP_CTL_ENABLE | DP_TP_CTL_LINK_TRAIN_PAT1;
 	if (intel_dp->link_mst)
 		dp_tp_ctl |= DP_TP_CTL_MODE_MST;
 	else {
@@ -4061,16 +4159,10 @@ static void intel_ddi_set_link_train(struct intel_dp *intel_dp,
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
 	u8 train_pat_mask = drm_dp_training_pattern_mask(intel_dp->dpcd);
-	enum port port = dp_to_dig_port(intel_dp)->base.port;
 	u32 temp;
 
 	temp = intel_de_read(dev_priv, intel_dp->regs.dp_tp_ctl);
 
-	if (dp_train_pat & DP_LINK_SCRAMBLING_DISABLE)
-		temp |= DP_TP_CTL_SCRAMBLE_DISABLE;
-	else
-		temp &= ~DP_TP_CTL_SCRAMBLE_DISABLE;
-
 	temp &= ~DP_TP_CTL_LINK_TRAIN_MASK;
 	switch (dp_train_pat & train_pat_mask) {
 	case DP_TRAINING_PATTERN_DISABLE:
@@ -4091,9 +4183,6 @@ static void intel_ddi_set_link_train(struct intel_dp *intel_dp,
 	}
 
 	intel_de_write(dev_priv, intel_dp->regs.dp_tp_ctl, temp);
-
-	intel_de_write(dev_priv, DDI_BUF_CTL(port), intel_dp->DP);
-	intel_de_posting_read(dev_priv, DDI_BUF_CTL(port));
 }
 
 static void intel_ddi_set_idle_link_train(struct intel_dp *intel_dp)
@@ -4878,6 +4967,64 @@ intel_ddi_max_lanes(struct intel_digital_port *dig_port)
 	return max_lanes;
 }
 
+static bool hti_uses_phy(struct drm_i915_private *i915, enum phy phy)
+{
+	return i915->hti_state & HDPORT_ENABLED &&
+		(i915->hti_state & HDPORT_PHY_USED_DP(phy) ||
+		 i915->hti_state & HDPORT_PHY_USED_HDMI(phy));
+}
+
+static enum hpd_pin tgl_hpd_pin(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	if (port >= PORT_D)
+		return HPD_PORT_TC1 + port - PORT_D;
+	else
+		return HPD_PORT_A + port - PORT_A;
+}
+
+static enum hpd_pin rkl_hpd_pin(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	if (HAS_PCH_TGP(dev_priv))
+		return tgl_hpd_pin(dev_priv, port);
+
+	if (port >= PORT_D)
+		return HPD_PORT_C + port - PORT_D;
+	else
+		return HPD_PORT_A + port - PORT_A;
+}
+
+static enum hpd_pin icl_hpd_pin(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	if (port >= PORT_C)
+		return HPD_PORT_TC1 + port - PORT_C;
+	else
+		return HPD_PORT_A + port - PORT_A;
+}
+
+static enum hpd_pin ehl_hpd_pin(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	if (port == PORT_D)
+		return HPD_PORT_A;
+
+	if (HAS_PCH_MCC(dev_priv))
+		return icl_hpd_pin(dev_priv, port);
+
+	return HPD_PORT_A + port - PORT_A;
+}
+
+static enum hpd_pin cnl_hpd_pin(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	if (port == PORT_F)
+		return HPD_PORT_E;
+
+	return HPD_PORT_A + port - PORT_A;
+}
+
 void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 {
 	struct intel_digital_port *dig_port;
@@ -4885,6 +5032,18 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	bool init_hdmi, init_dp, init_lspcon = false;
 	enum phy phy = intel_port_to_phy(dev_priv, port);
 
+	/*
+	 * On platforms with HTI (aka HDPORT), if it's enabled at boot it may
+	 * have taken over some of the PHYs and made them unavailable to the
+	 * driver.  In that case we should skip initializing the corresponding
+	 * outputs.
+	 */
+	if (hti_uses_phy(dev_priv, phy)) {
+		drm_dbg_kms(&dev_priv->drm, "PORT %c / PHY %c reserved by HTI\n",
+			    port_name(port), phy_name(phy));
+		return;
+	}
+
 	init_hdmi = intel_bios_port_supports_dvi(dev_priv, port) ||
 		intel_bios_port_supports_hdmi(dev_priv, port);
 	init_dp = intel_bios_port_supports_dp(dev_priv, port);
@@ -4918,6 +5077,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_ddi_funcs,
 			 DRM_MODE_ENCODER_TMDS, "DDI %c", port_name(port));
 
+	mutex_init(&dig_port->hdcp_mutex);
+	dig_port->num_hdcp_streams = 0;
+
 	encoder->hotplug = intel_ddi_hotplug;
 	encoder->compute_output_type = intel_ddi_compute_output_type;
 	encoder->compute_config = intel_ddi_compute_config;
@@ -4939,6 +5101,19 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	encoder->cloneable = 0;
 	encoder->pipe_mask = ~0;
 
+	if (IS_ROCKETLAKE(dev_priv))
+		encoder->hpd_pin = rkl_hpd_pin(dev_priv, port);
+	else if (INTEL_GEN(dev_priv) >= 12)
+		encoder->hpd_pin = tgl_hpd_pin(dev_priv, port);
+	else if (IS_ELKHARTLAKE(dev_priv))
+		encoder->hpd_pin = ehl_hpd_pin(dev_priv, port);
+	else if (IS_GEN(dev_priv, 11))
+		encoder->hpd_pin = icl_hpd_pin(dev_priv, port);
+	else if (IS_GEN(dev_priv, 10))
+		encoder->hpd_pin = cnl_hpd_pin(dev_priv, port);
+	else
+		encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port);
+
 	if (INTEL_GEN(dev_priv) >= 11)
 		dig_port->saved_port_bits =
 			intel_de_read(dev_priv, DDI_BUF_CTL(port))
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h
index 077e9dbbe367..f5fb62fc9400 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.h
+++ b/drivers/gpu/drm/i915/display/intel_ddi.h
@@ -16,6 +16,7 @@ struct intel_crtc_state;
 struct intel_dp;
 struct intel_dpll_hw_state;
 struct intel_encoder;
+enum transcoder;
 
 void intel_ddi_fdi_post_disable(struct intel_atomic_state *state,
 				struct intel_encoder *intel_encoder,
@@ -43,6 +44,7 @@ void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv,
 u32 bxt_signal_levels(struct intel_dp *intel_dp);
 u32 ddi_signal_levels(struct intel_dp *intel_dp);
 int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder,
+				     enum transcoder cpu_transcoder,
 				     bool enable);
 void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index b18c5ac2934d..631b4338224e 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -47,6 +47,7 @@
 #include "display/intel_ddi.h"
 #include "display/intel_dp.h"
 #include "display/intel_dp_mst.h"
+#include "display/intel_dpll_mgr.h"
 #include "display/intel_dsi.h"
 #include "display/intel_dvo.h"
 #include "display/intel_gmbus.h"
@@ -66,6 +67,7 @@
 #include "intel_bw.h"
 #include "intel_cdclk.h"
 #include "intel_color.h"
+#include "intel_csr.h"
 #include "intel_display_types.h"
 #include "intel_dp_link_training.h"
 #include "intel_fbc.h"
@@ -2310,7 +2312,7 @@ err:
 
 void intel_unpin_fb_vma(struct i915_vma *vma, unsigned long flags)
 {
-	i915_gem_object_lock(vma->obj);
+	i915_gem_object_lock(vma->obj, NULL);
 	if (flags & PLANE_HAS_FENCE)
 		i915_vma_unpin_fence(vma);
 	i915_gem_object_unpin_from_display_plane(vma);
@@ -3450,7 +3452,7 @@ initial_plane_vma(struct drm_i915_private *i915,
 	if (IS_ERR(vma))
 		goto err_obj;
 
-	if (i915_ggtt_pin(vma, 0, PIN_MAPPABLE | PIN_OFFSET_FIXED | base))
+	if (i915_ggtt_pin(vma, NULL, 0, PIN_MAPPABLE | PIN_OFFSET_FIXED | base))
 		goto err_obj;
 
 	if (i915_gem_object_is_tiled(obj) &&
@@ -3761,6 +3763,44 @@ static int glk_max_plane_width(const struct drm_framebuffer *fb,
 	}
 }
 
+static int icl_min_plane_width(const struct drm_framebuffer *fb)
+{
+	/* Wa_14011264657, Wa_14011050563: gen11+ */
+	switch (fb->format->format) {
+	case DRM_FORMAT_C8:
+		return 18;
+	case DRM_FORMAT_RGB565:
+		return 10;
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_ABGR8888:
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_XVYU2101010:
+	case DRM_FORMAT_Y212:
+	case DRM_FORMAT_Y216:
+		return 6;
+	case DRM_FORMAT_NV12:
+		return 20;
+	case DRM_FORMAT_P010:
+	case DRM_FORMAT_P012:
+	case DRM_FORMAT_P016:
+		return 12;
+	case DRM_FORMAT_XRGB16161616F:
+	case DRM_FORMAT_XBGR16161616F:
+	case DRM_FORMAT_ARGB16161616F:
+	case DRM_FORMAT_ABGR16161616F:
+	case DRM_FORMAT_XVYU12_16161616:
+	case DRM_FORMAT_XVYU16161616:
+		return 4;
+	default:
+		return 1;
+	}
+}
+
 static int icl_max_plane_width(const struct drm_framebuffer *fb,
 			       int color_plane,
 			       unsigned int rotation)
@@ -3843,29 +3883,31 @@ static int skl_check_main_surface(struct intel_plane_state *plane_state)
 	int y = plane_state->uapi.src.y1 >> 16;
 	int w = drm_rect_width(&plane_state->uapi.src) >> 16;
 	int h = drm_rect_height(&plane_state->uapi.src) >> 16;
-	int max_width;
-	int max_height;
-	u32 alignment;
-	u32 offset;
+	int max_width, min_width, max_height;
+	u32 alignment, offset;
 	int aux_plane = intel_main_to_aux_plane(fb, 0);
 	u32 aux_offset = plane_state->color_plane[aux_plane].offset;
 
-	if (INTEL_GEN(dev_priv) >= 11)
+	if (INTEL_GEN(dev_priv) >= 11) {
 		max_width = icl_max_plane_width(fb, 0, rotation);
-	else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
+		min_width = icl_min_plane_width(fb);
+	} else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) {
 		max_width = glk_max_plane_width(fb, 0, rotation);
-	else
+		min_width = 1;
+	} else {
 		max_width = skl_max_plane_width(fb, 0, rotation);
+		min_width = 1;
+	}
 
 	if (INTEL_GEN(dev_priv) >= 11)
 		max_height = icl_max_plane_height();
 	else
 		max_height = skl_max_plane_height();
 
-	if (w > max_width || h > max_height) {
+	if (w > max_width || w < min_width || h > max_height) {
 		drm_dbg_kms(&dev_priv->drm,
-			    "requested Y/RGB source size %dx%d too big (limit %dx%d)\n",
-			    w, h, max_width, max_height);
+			    "requested Y/RGB source size %dx%d outside limits (min: %dx1 max: %dx%d)\n",
+			    w, h, min_width, max_width, max_height);
 		return -EINVAL;
 	}
 
@@ -7290,6 +7332,10 @@ enum intel_display_power_domain intel_port_to_power_domain(enum port port)
 		return POWER_DOMAIN_PORT_DDI_F_LANES;
 	case PORT_G:
 		return POWER_DOMAIN_PORT_DDI_G_LANES;
+	case PORT_H:
+		return POWER_DOMAIN_PORT_DDI_H_LANES;
+	case PORT_I:
+		return POWER_DOMAIN_PORT_DDI_I_LANES;
 	default:
 		MISSING_CASE(port);
 		return POWER_DOMAIN_PORT_OTHER;
@@ -7315,6 +7361,10 @@ intel_aux_power_domain(struct intel_digital_port *dig_port)
 			return POWER_DOMAIN_AUX_F_TBT;
 		case AUX_CH_G:
 			return POWER_DOMAIN_AUX_G_TBT;
+		case AUX_CH_H:
+			return POWER_DOMAIN_AUX_H_TBT;
+		case AUX_CH_I:
+			return POWER_DOMAIN_AUX_I_TBT;
 		default:
 			MISSING_CASE(dig_port->aux_ch);
 			return POWER_DOMAIN_AUX_C_TBT;
@@ -7346,6 +7396,10 @@ intel_legacy_aux_to_power_domain(enum aux_ch aux_ch)
 		return POWER_DOMAIN_AUX_F;
 	case AUX_CH_G:
 		return POWER_DOMAIN_AUX_G;
+	case AUX_CH_H:
+		return POWER_DOMAIN_AUX_H;
+	case AUX_CH_I:
+		return POWER_DOMAIN_AUX_I;
 	default:
 		MISSING_CASE(aux_ch);
 		return POWER_DOMAIN_AUX_A;
@@ -8114,7 +8168,7 @@ static void compute_m_n(unsigned int m, unsigned int n,
 	 * which the devices expect also in synchronous clock mode.
 	 */
 	if (constant_n)
-		*ret_n = 0x8000;
+		*ret_n = DP_LINK_CONSTANT_N_VALUE;
 	else
 		*ret_n = min_t(unsigned int, roundup_pow_of_two(n), DATA_LINK_N_MAX);
 
@@ -10802,9 +10856,18 @@ static void icl_get_ddi_pll(struct drm_i915_private *dev_priv, enum port port,
 	u32 temp;
 
 	if (intel_phy_is_combo(dev_priv, phy)) {
-		temp = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0) &
-			ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
-		id = temp >> ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy);
+		u32 mask, shift;
+
+		if (IS_ROCKETLAKE(dev_priv)) {
+			mask = RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
+			shift = RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy);
+		} else {
+			mask = ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy);
+			shift = ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy);
+		}
+
+		temp = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0) & mask;
+		id = temp >> shift;
 		port_dpll_id = ICL_PORT_DPLL_DEFAULT;
 	} else if (intel_phy_is_tc(dev_priv, phy)) {
 		u32 clk_sel = intel_de_read(dev_priv, DDI_CLK_SEL(port)) & DDI_CLK_SEL_MASK;
@@ -12760,6 +12823,9 @@ static int intel_crtc_atomic_check(struct intel_atomic_state *state,
 
 	}
 
+	if (!mode_changed)
+		intel_psr2_sel_fetch_update(state, crtc);
+
 	return 0;
 }
 
@@ -13418,12 +13484,6 @@ encoder_retry:
 		    "hw max bpp: %i, pipe bpp: %i, dithering: %i\n",
 		    base_bpp, pipe_config->pipe_bpp, pipe_config->dither);
 
-	/*
-	 * Make drm_calc_timestamping_constants in
-	 * drm_atomic_helper_update_legacy_modeset_state() happy
-	 */
-	pipe_config->uapi.adjusted_mode = pipe_config->hw.adjusted_mode;
-
 	return 0;
 }
 
@@ -14244,7 +14304,6 @@ verify_crtc_state(struct intel_crtc *crtc,
 	struct intel_encoder *encoder;
 	struct intel_crtc_state *pipe_config = old_crtc_state;
 	struct drm_atomic_state *state = old_crtc_state->uapi.state;
-	bool active;
 
 	__drm_atomic_helper_crtc_destroy_state(&old_crtc_state->uapi);
 	intel_crtc_free_hw_state(old_crtc_state);
@@ -14254,16 +14313,19 @@ verify_crtc_state(struct intel_crtc *crtc,
 	drm_dbg_kms(&dev_priv->drm, "[CRTC:%d:%s]\n", crtc->base.base.id,
 		    crtc->base.name);
 
-	active = dev_priv->display.get_pipe_config(crtc, pipe_config);
+	pipe_config->hw.enable = new_crtc_state->hw.enable;
+
+	pipe_config->hw.active =
+		dev_priv->display.get_pipe_config(crtc, pipe_config);
 
 	/* we keep both pipes enabled on 830 */
-	if (IS_I830(dev_priv))
-		active = new_crtc_state->hw.active;
+	if (IS_I830(dev_priv) && pipe_config->hw.active)
+		pipe_config->hw.active = new_crtc_state->hw.active;
 
-	I915_STATE_WARN(new_crtc_state->hw.active != active,
+	I915_STATE_WARN(new_crtc_state->hw.active != pipe_config->hw.active,
 			"crtc active state doesn't match with hw state "
 			"(expected %i, found %i)\n",
-			new_crtc_state->hw.active, active);
+			new_crtc_state->hw.active, pipe_config->hw.active);
 
 	I915_STATE_WARN(crtc->active != new_crtc_state->hw.active,
 			"transitional active state does not match atomic hw state "
@@ -14272,6 +14334,7 @@ verify_crtc_state(struct intel_crtc *crtc,
 
 	for_each_encoder_on_crtc(dev, &crtc->base, encoder) {
 		enum pipe pipe;
+		bool active;
 
 		active = encoder->get_hw_state(encoder, &pipe);
 		I915_STATE_WARN(active != new_crtc_state->hw.active,
@@ -14583,16 +14646,8 @@ u8 intel_calc_active_pipes(struct intel_atomic_state *state,
 static int intel_modeset_checks(struct intel_atomic_state *state)
 {
 	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
-	int ret;
 
 	state->modeset = true;
-	state->active_pipes = intel_calc_active_pipes(state, dev_priv->active_pipes);
-
-	if (state->active_pipes != dev_priv->active_pipes) {
-		ret = _intel_atomic_lock_global_state(state);
-		if (ret)
-			return ret;
-	}
 
 	if (IS_HASWELL(dev_priv))
 		return hsw_mode_set_planes_workaround(state);
@@ -14736,7 +14791,8 @@ static int intel_atomic_check_cdclk(struct intel_atomic_state *state,
 				    bool *need_cdclk_calc)
 {
 	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
-	struct intel_cdclk_state *new_cdclk_state;
+	const struct intel_cdclk_state *old_cdclk_state;
+	const struct intel_cdclk_state *new_cdclk_state;
 	struct intel_plane_state *plane_state;
 	struct intel_bw_state *new_bw_state;
 	struct intel_plane *plane;
@@ -14755,9 +14811,11 @@ static int intel_atomic_check_cdclk(struct intel_atomic_state *state,
 			return ret;
 	}
 
+	old_cdclk_state = intel_atomic_get_old_cdclk_state(state);
 	new_cdclk_state = intel_atomic_get_new_cdclk_state(state);
 
-	if (new_cdclk_state && new_cdclk_state->force_min_cdclk_changed)
+	if (new_cdclk_state &&
+	    old_cdclk_state->force_min_cdclk != new_cdclk_state->force_min_cdclk)
 		*need_cdclk_calc = true;
 
 	ret = dev_priv->display.bw_calc_min_cdclk(state);
@@ -15134,6 +15192,8 @@ static void commit_pipe_config(struct intel_atomic_state *state,
 
 		if (new_crtc_state->update_pipe)
 			intel_pipe_fastset(old_crtc_state, new_crtc_state);
+
+		intel_psr2_program_trans_man_trk_ctl(new_crtc_state);
 	}
 
 	if (dev_priv->display.atomic_update_watermarks)
@@ -15702,14 +15762,6 @@ static void intel_atomic_track_fbs(struct intel_atomic_state *state)
 					plane->frontbuffer_bit);
 }
 
-static void assert_global_state_locked(struct drm_i915_private *dev_priv)
-{
-	struct intel_crtc *crtc;
-
-	for_each_intel_crtc(&dev_priv->drm, crtc)
-		drm_modeset_lock_assert_held(&crtc->base.mutex);
-}
-
 static int intel_atomic_commit(struct drm_device *dev,
 			       struct drm_atomic_state *_state,
 			       bool nonblock)
@@ -15785,12 +15837,6 @@ static int intel_atomic_commit(struct drm_device *dev,
 	intel_shared_dpll_swap_state(state);
 	intel_atomic_track_fbs(state);
 
-	if (state->global_state_changed) {
-		assert_global_state_locked(dev_priv);
-
-		dev_priv->active_pipes = state->active_pipes;
-	}
-
 	drm_atomic_state_get(&state->base);
 	INIT_WORK(&state->base.commit_work, intel_atomic_commit_work);
 
@@ -16837,7 +16883,7 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv)
 
 	intel_pps_init(dev_priv);
 
-	if (!HAS_DISPLAY(dev_priv) || !INTEL_DISPLAY_ENABLED(dev_priv))
+	if (!HAS_DISPLAY(dev_priv))
 		return;
 
 	if (IS_ROCKETLAKE(dev_priv)) {
@@ -17137,7 +17183,7 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb,
 	if (!intel_fb->frontbuffer)
 		return -ENOMEM;
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	tiling = i915_gem_object_get_tiling(obj);
 	stride = i915_gem_object_get_stride(obj);
 	i915_gem_object_unlock(obj);
@@ -17823,6 +17869,27 @@ int intel_modeset_init_noirq(struct drm_i915_private *i915)
 {
 	int ret;
 
+	if (i915_inject_probe_failure(i915))
+		return -ENODEV;
+
+	if (HAS_DISPLAY(i915)) {
+		ret = drm_vblank_init(&i915->drm,
+				      INTEL_NUM_PIPES(i915));
+		if (ret)
+			return ret;
+	}
+
+	intel_bios_init(i915);
+
+	ret = intel_vga_register(i915);
+	if (ret)
+		goto cleanup_bios;
+
+	/* FIXME: completely on the wrong abstraction layer */
+	intel_power_domains_init_hw(i915, false);
+
+	intel_csr_ucode_init(i915);
+
 	i915->modeset_wq = alloc_ordered_workqueue("i915_modeset", 0);
 	i915->flip_wq = alloc_workqueue("i915_flip", WQ_HIGHPRI |
 					WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE);
@@ -17831,15 +17898,15 @@ int intel_modeset_init_noirq(struct drm_i915_private *i915)
 
 	ret = intel_cdclk_init(i915);
 	if (ret)
-		return ret;
+		goto cleanup_vga_client_pw_domain_csr;
 
 	ret = intel_dbuf_init(i915);
 	if (ret)
-		return ret;
+		goto cleanup_vga_client_pw_domain_csr;
 
 	ret = intel_bw_init(i915);
 	if (ret)
-		return ret;
+		goto cleanup_vga_client_pw_domain_csr;
 
 	init_llist_head(&i915->atomic_helper.free_list);
 	INIT_WORK(&i915->atomic_helper.free_work,
@@ -17850,10 +17917,19 @@ int intel_modeset_init_noirq(struct drm_i915_private *i915)
 	intel_fbc_init(i915);
 
 	return 0;
+
+cleanup_vga_client_pw_domain_csr:
+	intel_csr_ucode_fini(i915);
+	intel_power_domains_driver_remove(i915);
+	intel_vga_unregister(i915);
+cleanup_bios:
+	intel_bios_driver_remove(i915);
+
+	return ret;
 }
 
-/* part #2: call after irq install */
-int intel_modeset_init(struct drm_i915_private *i915)
+/* part #2: call after irq install, but before gem init */
+int intel_modeset_init_nogem(struct drm_i915_private *i915)
 {
 	struct drm_device *dev = &i915->drm;
 	enum pipe pipe;
@@ -17870,7 +17946,7 @@ int intel_modeset_init(struct drm_i915_private *i915)
 		    INTEL_NUM_PIPES(i915),
 		    INTEL_NUM_PIPES(i915) > 1 ? "s" : "");
 
-	if (HAS_DISPLAY(i915) && INTEL_DISPLAY_ENABLED(i915)) {
+	if (HAS_DISPLAY(i915)) {
 		for_each_pipe(i915, pipe) {
 			ret = intel_crtc_init(i915, pipe);
 			if (ret) {
@@ -17892,6 +17968,13 @@ int intel_modeset_init(struct drm_i915_private *i915)
 	if (i915->max_cdclk_freq == 0)
 		intel_update_max_cdclk(i915);
 
+	/*
+	 * If the platform has HTI, we need to find out whether it has reserved
+	 * any display resources before we create our display outputs.
+	 */
+	if (INTEL_INFO(i915)->display.has_hti)
+		i915->hti_state = intel_de_read(i915, HDPORT_STATE);
+
 	/* Just disable it once at startup */
 	intel_vga_disable(i915);
 	intel_setup_outputs(i915);
@@ -17945,6 +18028,30 @@ int intel_modeset_init(struct drm_i915_private *i915)
 	return 0;
 }
 
+/* part #3: call after gem init */
+int intel_modeset_init(struct drm_i915_private *i915)
+{
+	int ret;
+
+	intel_overlay_setup(i915);
+
+	if (!HAS_DISPLAY(i915))
+		return 0;
+
+	ret = intel_fbdev_init(&i915->drm);
+	if (ret)
+		return ret;
+
+	/* Only enable hotplug handling once the fbdev is fully set up. */
+	intel_hpd_init(i915);
+
+	intel_init_ipc(i915);
+
+	intel_psr_set_force_mode_changed(i915->psr.dp);
+
+	return 0;
+}
+
 void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 {
 	struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
@@ -18829,6 +18936,18 @@ void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
 	intel_fbc_cleanup_cfb(i915);
 }
 
+/* part #3: call after gem init */
+void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915)
+{
+	intel_csr_ucode_fini(i915);
+
+	intel_power_domains_driver_remove(i915);
+
+	intel_vga_unregister(i915);
+
+	intel_bios_driver_remove(i915);
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
 
 struct intel_display_error_state {
@@ -18889,7 +19008,7 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv)
 
 	BUILD_BUG_ON(ARRAY_SIZE(transcoders) != ARRAY_SIZE(error->transcoder));
 
-	if (!HAS_DISPLAY(dev_priv) || !INTEL_DISPLAY_ENABLED(dev_priv))
+	if (!HAS_DISPLAY(dev_priv))
 		return NULL;
 
 	error = kzalloc(sizeof(*error), GFP_ATOMIC);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index e890c8fb779b..d10b7c8cde3f 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -272,8 +272,6 @@ enum dpio_phy {
 	DPIO_PHY2,
 };
 
-#define I915_NUM_PHYS_VLV 2
-
 enum aux_ch {
 	AUX_CH_A,
 	AUX_CH_B,
@@ -282,6 +280,8 @@ enum aux_ch {
 	AUX_CH_E, /* ICL+ */
 	AUX_CH_F,
 	AUX_CH_G,
+	AUX_CH_H,
+	AUX_CH_I,
 };
 
 #define aux_ch_name(a) ((a) + 'A')
@@ -629,9 +629,11 @@ intel_format_info_is_yuv_semiplanar(const struct drm_format_info *info,
 /* modesetting */
 void intel_modeset_init_hw(struct drm_i915_private *i915);
 int intel_modeset_init_noirq(struct drm_i915_private *i915);
+int intel_modeset_init_nogem(struct drm_i915_private *i915);
 int intel_modeset_init(struct drm_i915_private *i915);
 void intel_modeset_driver_remove(struct drm_i915_private *i915);
 void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915);
+void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_resume(struct drm_device *dev);
 void intel_init_pch_refclk(struct drm_i915_private *dev_priv);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index 5a5cfe25085b..0bf31f9a8af5 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -417,6 +417,9 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
 			su_blocks = su_blocks >> PSR2_SU_STATUS_SHIFT(frame);
 			seq_printf(m, "%d\t%d\n", frame, su_blocks);
 		}
+
+		seq_printf(m, "PSR2 selective fetch: %s\n",
+			   enableddisabled(psr->psr2_sel_fetch_enabled));
 	}
 
 unlock:
@@ -598,6 +601,11 @@ static void intel_hdcp_info(struct seq_file *m,
 {
 	bool hdcp_cap, hdcp2_cap;
 
+	if (!intel_connector->hdcp.shim) {
+		seq_puts(m, "No Connector Support");
+		goto out;
+	}
+
 	hdcp_cap = intel_hdcp_capable(intel_connector);
 	hdcp2_cap = intel_hdcp2_capable(intel_connector);
 
@@ -609,6 +617,7 @@ static void intel_hdcp_info(struct seq_file *m,
 	if (!hdcp_cap && !hdcp2_cap)
 		seq_puts(m, "None");
 
+out:
 	seq_puts(m, "\n");
 }
 
@@ -617,6 +626,7 @@ static void intel_dp_info(struct seq_file *m,
 {
 	struct intel_encoder *intel_encoder = intel_attached_encoder(intel_connector);
 	struct intel_dp *intel_dp = enc_to_intel_dp(intel_encoder);
+	const struct drm_property_blob *edid = intel_connector->base.edid_blob_ptr;
 
 	seq_printf(m, "\tDPCD rev: %x\n", intel_dp->dpcd[DP_DPCD_REV]);
 	seq_printf(m, "\taudio support: %s\n", yesno(intel_dp->has_audio));
@@ -624,11 +634,7 @@ static void intel_dp_info(struct seq_file *m,
 		intel_panel_info(m, &intel_connector->panel);
 
 	drm_dp_downstream_debug(m, intel_dp->dpcd, intel_dp->downstream_ports,
-				&intel_dp->aux);
-	if (intel_connector->hdcp.shim) {
-		seq_puts(m, "\tHDCP version: ");
-		intel_hdcp_info(m, intel_connector);
-	}
+				edid ? edid->data : NULL, &intel_dp->aux);
 }
 
 static void intel_dp_mst_info(struct seq_file *m,
@@ -646,10 +652,6 @@ static void intel_hdmi_info(struct seq_file *m,
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(intel_encoder);
 
 	seq_printf(m, "\taudio support: %s\n", yesno(intel_hdmi->has_audio));
-	if (intel_connector->hdcp.shim) {
-		seq_puts(m, "\tHDCP version: ");
-		intel_hdcp_info(m, intel_connector);
-	}
 }
 
 static void intel_lvds_info(struct seq_file *m,
@@ -705,6 +707,9 @@ static void intel_connector_info(struct seq_file *m,
 		break;
 	}
 
+	seq_puts(m, "\tHDCP version: ");
+	intel_hdcp_info(m, intel_connector);
+
 	seq_printf(m, "\tmodes:\n");
 	list_for_each_entry(mode, &connector->modes, head)
 		intel_seq_print_mode(m, 2, mode);
@@ -1066,10 +1071,18 @@ static void drrs_status_per_crtc(struct seq_file *m,
 
 	drm_connector_list_iter_begin(dev, &conn_iter);
 	drm_for_each_connector_iter(connector, &conn_iter) {
+		bool supported = false;
+
 		if (connector->state->crtc != &intel_crtc->base)
 			continue;
 
 		seq_printf(m, "%s:\n", connector->name);
+
+		if (connector->connector_type == DRM_MODE_CONNECTOR_eDP &&
+		    drrs->type == SEAMLESS_DRRS_SUPPORT)
+			supported = true;
+
+		seq_printf(m, "\tDRRS Supported: %s\n", yesno(supported));
 	}
 	drm_connector_list_iter_end(&conn_iter);
 
@@ -1080,7 +1093,7 @@ static void drrs_status_per_crtc(struct seq_file *m,
 
 		mutex_lock(&drrs->mutex);
 		/* DRRS Supported */
-		seq_puts(m, "\tDRRS Supported: Yes\n");
+		seq_puts(m, "\tDRRS Enabled: Yes\n");
 
 		/* disable_drrs() will make drrs->dp NULL */
 		if (!drrs->dp) {
@@ -1115,7 +1128,7 @@ static void drrs_status_per_crtc(struct seq_file *m,
 		mutex_unlock(&drrs->mutex);
 	} else {
 		/* DRRS not supported. Print the VBT parameter*/
-		seq_puts(m, "\tDRRS Supported : No");
+		seq_puts(m, "\tDRRS Enabled : No");
 	}
 	seq_puts(m, "\n");
 }
@@ -2026,10 +2039,6 @@ static int i915_hdcp_sink_capability_show(struct seq_file *m, void *data)
 	if (connector->status != connector_status_connected)
 		return -ENODEV;
 
-	/* HDCP is supported by connector */
-	if (!intel_connector->hdcp.shim)
-		return -EINVAL;
-
 	seq_printf(m, "%s:%d HDCP version: ", connector->name,
 		   connector->base.id);
 	intel_hdcp_info(m, intel_connector);
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index e0fcb89c736b..7277e58b01f1 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -3927,12 +3927,13 @@ tgl_tc_cold_request(struct drm_i915_private *i915, bool block)
 	int ret;
 
 	while (1) {
-		u32 low_val = 0, high_val;
+		u32 low_val;
+		u32 high_val = 0;
 
 		if (block)
-			high_val = TGL_PCODE_EXIT_TCCOLD_DATA_H_BLOCK_REQ;
+			low_val = TGL_PCODE_EXIT_TCCOLD_DATA_L_BLOCK_REQ;
 		else
-			high_val = TGL_PCODE_EXIT_TCCOLD_DATA_H_UNBLOCK_REQ;
+			low_val = TGL_PCODE_EXIT_TCCOLD_DATA_L_UNBLOCK_REQ;
 
 		/*
 		 * Spec states that we should timeout the request after 200us
@@ -3951,8 +3952,7 @@ tgl_tc_cold_request(struct drm_i915_private *i915, bool block)
 		if (++tries == 3)
 			break;
 
-		if (ret == -EAGAIN)
-			msleep(1);
+		msleep(1);
 	}
 
 	if (ret)
@@ -5263,7 +5263,7 @@ static void tgl_bw_buddy_init(struct drm_i915_private *dev_priv)
 	unsigned long abox_mask = INTEL_INFO(dev_priv)->abox_mask;
 	int config, i;
 
-	if (IS_TGL_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_B0))
+	if (IS_TGL_DISP_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_B0))
 		/* Wa_1409767108: tgl */
 		table = wa_1409767108_buddy_page_masks;
 	else
@@ -5302,6 +5302,12 @@ static void icl_display_core_init(struct drm_i915_private *dev_priv,
 
 	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
 
+	/* Wa_14011294188:ehl,jsl,tgl,rkl */
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_JSP &&
+	    INTEL_PCH_TYPE(dev_priv) < PCH_DG1)
+		intel_de_rmw(dev_priv, SOUTH_DSPCLK_GATE_D, 0,
+			     PCH_DPMGUNIT_CLOCK_GATE_DISABLE);
+
 	/* 1. Enable PCH reset handshake. */
 	intel_pch_reset_handshake(dev_priv, !HAS_PCH_NOP(dev_priv));
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index e8f809161c75..3d4bf9b6a0a2 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -28,6 +28,7 @@
 
 #include <linux/async.h>
 #include <linux/i2c.h>
+#include <linux/pwm.h>
 #include <linux/sched/clock.h>
 
 #include <drm/drm_atomic.h>
@@ -223,6 +224,7 @@ struct intel_panel {
 		bool util_pin_active_low;	/* bxt+ */
 		u8 controller;		/* bxt+ only */
 		struct pwm_device *pwm;
+		struct pwm_state pwm_state;
 
 		/* DPCD backlight */
 		u8 pwmgen_bit_count;
@@ -314,10 +316,12 @@ struct intel_hdcp_shim {
 
 	/* Enables HDCP signalling on the port */
 	int (*toggle_signalling)(struct intel_digital_port *dig_port,
+				 enum transcoder cpu_transcoder,
 				 bool enable);
 
 	/* Ensures the link is still protected */
-	bool (*check_link)(struct intel_digital_port *dig_port);
+	bool (*check_link)(struct intel_digital_port *dig_port,
+			   struct intel_connector *connector);
 
 	/* Detects panel's hdcp capability. This is optional for HDMI. */
 	int (*hdcp_capable)(struct intel_digital_port *dig_port,
@@ -479,8 +483,6 @@ struct intel_atomic_state {
 
 	bool dpll_set, modeset;
 
-	u8 active_pipes;
-
 	struct intel_shared_dpll_state shared_dpll[I915_NUM_PLLS];
 
 	/*
@@ -491,11 +493,6 @@ struct intel_atomic_state {
 
 	bool rps_interactive;
 
-	/*
-	 * active_pipes
-	 */
-	bool global_state_changed;
-
 	struct i915_sw_fence commit_ready;
 
 	struct llist_node freed;
@@ -931,6 +928,7 @@ struct intel_crtc_state {
 
 	bool has_psr;
 	bool has_psr2;
+	bool enable_psr2_sel_fetch;
 	u32 dc3co_exitline;
 
 	/*
@@ -1073,6 +1071,8 @@ struct intel_crtc_state {
 
 	/* For DSB related info */
 	struct intel_dsb *dsb;
+
+	u32 psr2_man_track_ctl;
 };
 
 enum intel_pipe_crc_source {
@@ -1272,6 +1272,7 @@ struct intel_dp {
 	u8 sink_count;
 	bool link_mst;
 	bool link_trained;
+	bool has_hdmi_sink;
 	bool has_audio;
 	bool reset_link_params;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
@@ -1373,8 +1374,19 @@ struct intel_dp {
 	/* Displayport compliance testing */
 	struct intel_dp_compliance compliance;
 
+	/* Downstream facing port caps */
+	struct {
+		int min_tmds_clock, max_tmds_clock;
+		int max_dotclock;
+		u8 max_bpc;
+		bool ycbcr_444_to_420;
+	} dfp;
+
 	/* Display stream compression testing */
 	bool force_dsc_en;
+
+	bool hobl_failed;
+	bool hobl_active;
 };
 
 enum lspcon_vendor {
@@ -1409,6 +1421,11 @@ struct intel_digital_port {
 	enum phy_fia tc_phy_fia;
 	u8 tc_phy_fia_idx;
 
+	/* protects num_hdcp_streams reference count */
+	struct mutex hdcp_mutex;
+	/* the number of pipes using HDCP signalling out of this port */
+	unsigned int num_hdcp_streams;
+
 	void (*write_infoframe)(struct intel_encoder *encoder,
 				const struct intel_crtc_state *crtc_state,
 				unsigned int type,
@@ -1519,6 +1536,18 @@ static inline bool intel_encoder_is_dig_port(struct intel_encoder *encoder)
 	}
 }
 
+static inline bool intel_encoder_is_mst(struct intel_encoder *encoder)
+{
+	return encoder->type == INTEL_OUTPUT_DP_MST;
+}
+
+static inline struct intel_dp_mst_encoder *
+enc_to_mst(struct intel_encoder *encoder)
+{
+	return container_of(&encoder->base, struct intel_dp_mst_encoder,
+			    base.base);
+}
+
 static inline struct intel_digital_port *
 enc_to_dig_port(struct intel_encoder *encoder)
 {
@@ -1527,6 +1556,8 @@ enc_to_dig_port(struct intel_encoder *encoder)
 	if (intel_encoder_is_dig_port(intel_encoder))
 		return container_of(&encoder->base, struct intel_digital_port,
 				    base.base);
+	else if (intel_encoder_is_mst(intel_encoder))
+		return enc_to_mst(encoder)->primary;
 	else
 		return NULL;
 }
@@ -1537,13 +1568,6 @@ intel_attached_dig_port(struct intel_connector *connector)
 	return enc_to_dig_port(intel_attached_encoder(connector));
 }
 
-static inline struct intel_dp_mst_encoder *
-enc_to_mst(struct intel_encoder *encoder)
-{
-	return container_of(&encoder->base, struct intel_dp_mst_encoder,
-			    base.base);
-}
-
 static inline struct intel_dp *enc_to_intel_dp(struct intel_encoder *encoder)
 {
 	return &enc_to_dig_port(encoder)->dp;
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index d6295eb20b63..bf1e9cf1c0f3 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -38,7 +38,6 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
-#include <drm/drm_hdcp.h>
 #include <drm/drm_probe_helper.h>
 
 #include "i915_debugfs.h"
@@ -248,29 +247,6 @@ intel_dp_max_data_rate(int max_link_clock, int max_lanes)
 	return max_link_clock * max_lanes;
 }
 
-static int
-intel_dp_downstream_max_dotclock(struct intel_dp *intel_dp)
-{
-	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	struct intel_encoder *encoder = &dig_port->base;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	int max_dotclk = dev_priv->max_dotclk_freq;
-	int ds_max_dotclk;
-
-	int type = intel_dp->downstream_ports[0] & DP_DS_PORT_TYPE_MASK;
-
-	if (type != DP_DS_PORT_TYPE_VGA)
-		return max_dotclk;
-
-	ds_max_dotclk = drm_dp_downstream_max_clock(intel_dp->dpcd,
-						    intel_dp->downstream_ports);
-
-	if (ds_max_dotclk != 0)
-		max_dotclk = min(max_dotclk, ds_max_dotclk);
-
-	return max_dotclk;
-}
-
 static int cnl_max_source_rate(struct intel_dp *intel_dp)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
@@ -636,6 +612,34 @@ static bool intel_dp_hdisplay_bad(struct drm_i915_private *dev_priv,
 }
 
 static enum drm_mode_status
+intel_dp_mode_valid_downstream(struct intel_connector *connector,
+			       const struct drm_display_mode *mode,
+			       int target_clock)
+{
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	const struct drm_display_info *info = &connector->base.display_info;
+	int tmds_clock;
+
+	if (intel_dp->dfp.max_dotclock &&
+	    target_clock > intel_dp->dfp.max_dotclock)
+		return MODE_CLOCK_HIGH;
+
+	/* Assume 8bpc for the DP++/HDMI/DVI TMDS clock check */
+	tmds_clock = target_clock;
+	if (drm_mode_is_420_only(info, mode))
+		tmds_clock /= 2;
+
+	if (intel_dp->dfp.min_tmds_clock &&
+	    tmds_clock < intel_dp->dfp.min_tmds_clock)
+		return MODE_CLOCK_LOW;
+	if (intel_dp->dfp.max_tmds_clock &&
+	    tmds_clock > intel_dp->dfp.max_tmds_clock)
+		return MODE_CLOCK_HIGH;
+
+	return MODE_OK;
+}
+
+static enum drm_mode_status
 intel_dp_mode_valid(struct drm_connector *connector,
 		    struct drm_display_mode *mode)
 {
@@ -645,15 +649,14 @@ intel_dp_mode_valid(struct drm_connector *connector,
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	int target_clock = mode->clock;
 	int max_rate, mode_rate, max_lanes, max_link_clock;
-	int max_dotclk;
+	int max_dotclk = dev_priv->max_dotclk_freq;
 	u16 dsc_max_output_bpp = 0;
 	u8 dsc_slice_count = 0;
+	enum drm_mode_status status;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
 
-	max_dotclk = intel_dp_downstream_max_dotclock(intel_dp);
-
 	if (intel_dp_is_edp(intel_dp) && fixed_mode) {
 		if (mode->hdisplay > fixed_mode->hdisplay)
 			return MODE_PANEL;
@@ -709,6 +712,11 @@ intel_dp_mode_valid(struct drm_connector *connector,
 	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
 		return MODE_H_ILLEGAL;
 
+	status = intel_dp_mode_valid_downstream(intel_connector,
+						mode, target_clock);
+	if (status != MODE_OK)
+		return status;
+
 	return intel_mode_valid_max_plane_size(dev_priv, mode);
 }
 
@@ -1563,6 +1571,20 @@ intel_dp_aux_header(u8 txbuf[HEADER_SIZE],
 	txbuf[3] = msg->size - 1;
 }
 
+static u32 intel_dp_aux_xfer_flags(const struct drm_dp_aux_msg *msg)
+{
+	/*
+	 * If we're trying to send the HDCP Aksv, we need to set a the Aksv
+	 * select bit to inform the hardware to send the Aksv after our header
+	 * since we can't access that data from software.
+	 */
+	if ((msg->request & ~DP_AUX_I2C_MOT) == DP_AUX_NATIVE_WRITE &&
+	    msg->address == DP_AUX_HDCP_AKSV)
+		return DP_AUX_CH_CTL_AUX_AKSV_SELECT;
+
+	return 0;
+}
+
 static ssize_t
 intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 {
@@ -1570,6 +1592,7 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	u8 txbuf[20], rxbuf[20];
 	size_t txsize, rxsize;
+	u32 flags = intel_dp_aux_xfer_flags(msg);
 	int ret;
 
 	intel_dp_aux_header(txbuf, msg);
@@ -1590,7 +1613,7 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 			memcpy(txbuf + HEADER_SIZE, msg->buffer, msg->size);
 
 		ret = intel_dp_aux_xfer(intel_dp, txbuf, txsize,
-					rxbuf, rxsize, 0);
+					rxbuf, rxsize, flags);
 		if (ret > 0) {
 			msg->reply = rxbuf[0] >> 4;
 
@@ -1613,7 +1636,7 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 			return -E2BIG;
 
 		ret = intel_dp_aux_xfer(intel_dp, txbuf, txsize,
-					rxbuf, rxsize, 0);
+					rxbuf, rxsize, flags);
 		if (ret > 0) {
 			msg->reply = rxbuf[0] >> 4;
 			/*
@@ -1954,19 +1977,72 @@ static bool intel_dp_supports_dsc(struct intel_dp *intel_dp,
 		drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd);
 }
 
-static int intel_dp_compute_bpp(struct intel_dp *intel_dp,
-				struct intel_crtc_state *pipe_config)
+static bool intel_dp_hdmi_ycbcr420(struct intel_dp *intel_dp,
+				   const struct intel_crtc_state *crtc_state)
+{
+	return crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
+		(crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 &&
+		 intel_dp->dfp.ycbcr_444_to_420);
+}
+
+static int intel_dp_hdmi_tmds_clock(struct intel_dp *intel_dp,
+				    const struct intel_crtc_state *crtc_state, int bpc)
+{
+	int clock = crtc_state->hw.adjusted_mode.crtc_clock * bpc / 8;
+
+	if (intel_dp_hdmi_ycbcr420(intel_dp, crtc_state))
+		clock /= 2;
+
+	return clock;
+}
+
+static bool intel_dp_hdmi_tmds_clock_valid(struct intel_dp *intel_dp,
+					   const struct intel_crtc_state *crtc_state, int bpc)
+{
+	int tmds_clock = intel_dp_hdmi_tmds_clock(intel_dp, crtc_state, bpc);
+
+	if (intel_dp->dfp.min_tmds_clock &&
+	    tmds_clock < intel_dp->dfp.min_tmds_clock)
+		return false;
+
+	if (intel_dp->dfp.max_tmds_clock &&
+	    tmds_clock > intel_dp->dfp.max_tmds_clock)
+		return false;
+
+	return true;
+}
+
+static bool intel_dp_hdmi_deep_color_possible(struct intel_dp *intel_dp,
+					      const struct intel_crtc_state *crtc_state,
+					      int bpc)
+{
+
+	return intel_hdmi_deep_color_possible(crtc_state, bpc,
+					      intel_dp->has_hdmi_sink,
+					      intel_dp_hdmi_ycbcr420(intel_dp, crtc_state)) &&
+		intel_dp_hdmi_tmds_clock_valid(intel_dp, crtc_state, bpc);
+}
+
+static int intel_dp_max_bpp(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
 	struct intel_connector *intel_connector = intel_dp->attached_connector;
 	int bpp, bpc;
 
-	bpp = pipe_config->pipe_bpp;
-	bpc = drm_dp_downstream_max_bpc(intel_dp->dpcd, intel_dp->downstream_ports);
+	bpc = crtc_state->pipe_bpp / 3;
+
+	if (intel_dp->dfp.max_bpc)
+		bpc = min_t(int, bpc, intel_dp->dfp.max_bpc);
 
-	if (bpc > 0)
-		bpp = min(bpp, 3*bpc);
+	if (intel_dp->dfp.min_tmds_clock) {
+		for (; bpc >= 10; bpc -= 2) {
+			if (intel_dp_hdmi_deep_color_possible(intel_dp, crtc_state, bpc))
+				break;
+		}
+	}
 
+	bpp = bpc * 3;
 	if (intel_dp_is_edp(intel_dp)) {
 		/* Get bpp from vbt only for panels that dont have bpp in edid */
 		if (intel_connector->base.display_info.bpc == 0 &&
@@ -2288,7 +2364,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 	limits.max_lane_count = intel_dp_max_lane_count(intel_dp);
 
 	limits.min_bpp = intel_dp_min_bpp(pipe_config);
-	limits.max_bpp = intel_dp_compute_bpp(intel_dp, pipe_config);
+	limits.max_bpp = intel_dp_max_bpp(intel_dp, pipe_config);
 
 	if (intel_dp_is_edp(intel_dp)) {
 		/*
@@ -2363,10 +2439,16 @@ intel_dp_ycbcr420_config(struct intel_dp *intel_dp,
 	const struct drm_display_mode *adjusted_mode =
 		&crtc_state->hw.adjusted_mode;
 
-	if (!drm_mode_is_420_only(info, adjusted_mode) ||
-	    !intel_dp_get_colorimetry_status(intel_dp) ||
-	    !connector->ycbcr_420_allowed)
+	if (!connector->ycbcr_420_allowed)
+		return 0;
+
+	if (!drm_mode_is_420_only(info, adjusted_mode))
+		return 0;
+
+	if (intel_dp->dfp.ycbcr_444_to_420) {
+		crtc_state->output_format = INTEL_OUTPUT_FORMAT_YCBCR444;
 		return 0;
+	}
 
 	crtc_state->output_format = INTEL_OUTPUT_FORMAT_YCBCR420;
 
@@ -2575,6 +2657,34 @@ intel_dp_compute_hdr_metadata_infoframe_sdp(struct intel_dp *intel_dp,
 		intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GAMUT_METADATA);
 }
 
+static void
+intel_dp_drrs_compute_config(struct intel_dp *intel_dp,
+			     struct intel_crtc_state *pipe_config,
+			     int output_bpp, bool constant_n)
+{
+	struct intel_connector *intel_connector = intel_dp->attached_connector;
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+
+	/*
+	 * DRRS and PSR can't be enable together, so giving preference to PSR
+	 * as it allows more power-savings by complete shutting down display,
+	 * so to guarantee this, intel_dp_drrs_compute_config() must be called
+	 * after intel_psr_compute_config().
+	 */
+	if (pipe_config->has_psr)
+		return;
+
+	if (!intel_connector->panel.downclock_mode ||
+	    dev_priv->drrs.type != SEAMLESS_DRRS_SUPPORT)
+		return;
+
+	pipe_config->has_drrs = true;
+	intel_link_compute_m_n(output_bpp, pipe_config->lane_count,
+			       intel_connector->panel.downclock_mode->clock,
+			       pipe_config->port_clock, &pipe_config->dp_m2_n2,
+			       constant_n, pipe_config->fec_enable);
+}
+
 int
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_state *pipe_config,
@@ -2605,7 +2715,6 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	if (ret)
 		return ret;
 
-	pipe_config->has_drrs = false;
 	if (!intel_dp_port_has_audio(dev_priv, port))
 		pipe_config->has_audio = false;
 	else if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
@@ -2657,21 +2766,12 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 			       &pipe_config->dp_m_n,
 			       constant_n, pipe_config->fec_enable);
 
-	if (intel_connector->panel.downclock_mode != NULL &&
-		dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) {
-			pipe_config->has_drrs = true;
-			intel_link_compute_m_n(output_bpp,
-					       pipe_config->lane_count,
-					       intel_connector->panel.downclock_mode->clock,
-					       pipe_config->port_clock,
-					       &pipe_config->dp_m2_n2,
-					       constant_n, pipe_config->fec_enable);
-	}
-
 	if (!HAS_DDI(dev_priv))
 		intel_dp_set_clock(encoder, pipe_config);
 
 	intel_psr_compute_config(intel_dp, pipe_config);
+	intel_dp_drrs_compute_config(intel_dp, pipe_config, output_bpp,
+				     constant_n);
 	intel_dp_compute_vsc_sdp(intel_dp, pipe_config, conn_state);
 	intel_dp_compute_hdr_metadata_infoframe_sdp(intel_dp, pipe_config, conn_state);
 
@@ -3752,6 +3852,43 @@ static void intel_dp_enable_port(struct intel_dp *intel_dp,
 	intel_de_posting_read(dev_priv, intel_dp->output_reg);
 }
 
+void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp)
+{
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+	u8 tmp;
+
+	if (intel_dp->dpcd[DP_DPCD_REV] < 0x13)
+		return;
+
+	if (!drm_dp_is_branch(intel_dp->dpcd))
+		return;
+
+	tmp = intel_dp->has_hdmi_sink ?
+		DP_HDMI_DVI_OUTPUT_CONFIG : 0;
+
+	if (drm_dp_dpcd_writeb(&intel_dp->aux,
+			       DP_PROTOCOL_CONVERTER_CONTROL_0, tmp) != 1)
+		drm_dbg_kms(&i915->drm, "Failed to set protocol converter HDMI mode to %s\n",
+			    enableddisabled(intel_dp->has_hdmi_sink));
+
+	tmp = intel_dp->dfp.ycbcr_444_to_420 ?
+		DP_CONVERSION_TO_YCBCR420_ENABLE : 0;
+
+	if (drm_dp_dpcd_writeb(&intel_dp->aux,
+			       DP_PROTOCOL_CONVERTER_CONTROL_1, tmp) != 1)
+		drm_dbg_kms(&i915->drm,
+			    "Failed to set protocol converter YCbCr 4:2:0 conversion mode to %s\n",
+			    enableddisabled(intel_dp->dfp.ycbcr_444_to_420));
+
+	tmp = 0;
+
+	if (drm_dp_dpcd_writeb(&intel_dp->aux,
+			       DP_PROTOCOL_CONVERTER_CONTROL_2, tmp) <= 0)
+		drm_dbg_kms(&i915->drm,
+			    "Failed to set protocol converter YCbCr 4:2:2 conversion mode to %s\n",
+			    enableddisabled(false));
+}
+
 static void intel_enable_dp(struct intel_atomic_state *state,
 			    struct intel_encoder *encoder,
 			    const struct intel_crtc_state *pipe_config,
@@ -3789,6 +3926,7 @@ static void intel_enable_dp(struct intel_atomic_state *state,
 	}
 
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+	intel_dp_configure_protocol_converter(intel_dp);
 	intel_dp_start_link_train(intel_dp);
 	intel_dp_stop_link_train(intel_dp);
 
@@ -4449,62 +4587,6 @@ intel_dp_link_down(struct intel_encoder *encoder,
 	}
 }
 
-static void
-intel_dp_extended_receiver_capabilities(struct intel_dp *intel_dp)
-{
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	u8 dpcd_ext[6];
-
-	/*
-	 * Prior to DP1.3 the bit represented by
-	 * DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT was reserved.
-	 * if it is set DP_DPCD_REV at 0000h could be at a value less than
-	 * the true capability of the panel. The only way to check is to
-	 * then compare 0000h and 2200h.
-	 */
-	if (!(intel_dp->dpcd[DP_TRAINING_AUX_RD_INTERVAL] &
-	      DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT))
-		return;
-
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_DP13_DPCD_REV,
-			     &dpcd_ext, sizeof(dpcd_ext)) != sizeof(dpcd_ext)) {
-		drm_err(&i915->drm,
-			"DPCD failed read at extended capabilities\n");
-		return;
-	}
-
-	if (intel_dp->dpcd[DP_DPCD_REV] > dpcd_ext[DP_DPCD_REV]) {
-		drm_dbg_kms(&i915->drm,
-			    "DPCD extended DPCD rev less than base DPCD rev\n");
-		return;
-	}
-
-	if (!memcmp(intel_dp->dpcd, dpcd_ext, sizeof(dpcd_ext)))
-		return;
-
-	drm_dbg_kms(&i915->drm, "Base DPCD: %*ph\n",
-		    (int)sizeof(intel_dp->dpcd), intel_dp->dpcd);
-
-	memcpy(intel_dp->dpcd, dpcd_ext, sizeof(dpcd_ext));
-}
-
-bool
-intel_dp_read_dpcd(struct intel_dp *intel_dp)
-{
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-
-	if (drm_dp_dpcd_read(&intel_dp->aux, 0x000, intel_dp->dpcd,
-			     sizeof(intel_dp->dpcd)) < 0)
-		return false; /* aux transfer failed */
-
-	intel_dp_extended_receiver_capabilities(intel_dp);
-
-	drm_dbg_kms(&i915->drm, "DPCD: %*ph\n", (int)sizeof(intel_dp->dpcd),
-		    intel_dp->dpcd);
-
-	return intel_dp->dpcd[DP_DPCD_REV] != 0;
-}
-
 bool intel_dp_get_colorimetry_status(struct intel_dp *intel_dp)
 {
 	u8 dprx = 0;
@@ -4563,7 +4645,7 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp)
 	/* this function is meant to be called only once */
 	drm_WARN_ON(&dev_priv->drm, intel_dp->dpcd[DP_DPCD_REV] != 0);
 
-	if (!intel_dp_read_dpcd(intel_dp))
+	if (drm_dp_read_dpcd_caps(&intel_dp->aux, intel_dp->dpcd) != 0)
 		return false;
 
 	drm_dp_read_desc(&intel_dp->aux, &intel_dp->desc,
@@ -4634,11 +4716,23 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp)
 	return true;
 }
 
+static bool
+intel_dp_has_sink_count(struct intel_dp *intel_dp)
+{
+	if (!intel_dp->attached_connector)
+		return false;
+
+	return drm_dp_read_sink_count_cap(&intel_dp->attached_connector->base,
+					  intel_dp->dpcd,
+					  &intel_dp->desc);
+}
 
 static bool
 intel_dp_get_dpcd(struct intel_dp *intel_dp)
 {
-	if (!intel_dp_read_dpcd(intel_dp))
+	int ret;
+
+	if (drm_dp_read_dpcd_caps(&intel_dp->aux, intel_dp->dpcd))
 		return false;
 
 	/*
@@ -4653,18 +4747,9 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 		intel_dp_set_common_rates(intel_dp);
 	}
 
-	/*
-	 * Some eDP panels do not set a valid value for sink count, that is why
-	 * it don't care about read it here and in intel_edp_init_dpcd().
-	 */
-	if (!intel_dp_is_edp(intel_dp) &&
-	    !drm_dp_has_quirk(&intel_dp->desc, 0,
-			      DP_DPCD_QUIRK_NO_SINK_COUNT)) {
-		u8 count;
-		ssize_t r;
-
-		r = drm_dp_dpcd_readb(&intel_dp->aux, DP_SINK_COUNT, &count);
-		if (r < 1)
+	if (intel_dp_has_sink_count(intel_dp)) {
+		ret = drm_dp_read_sink_count(&intel_dp->aux);
+		if (ret < 0)
 			return false;
 
 		/*
@@ -4672,7 +4757,7 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 		 * a member variable in intel_dp will track any changes
 		 * between short pulse interrupts.
 		 */
-		intel_dp->sink_count = DP_GET_SINK_COUNT(count);
+		intel_dp->sink_count = ret;
 
 		/*
 		 * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that
@@ -4685,32 +4770,8 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 			return false;
 	}
 
-	if (!drm_dp_is_branch(intel_dp->dpcd))
-		return true; /* native DP sink */
-
-	if (intel_dp->dpcd[DP_DPCD_REV] == 0x10)
-		return true; /* no per-port downstream info */
-
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_DOWNSTREAM_PORT_0,
-			     intel_dp->downstream_ports,
-			     DP_MAX_DOWNSTREAM_PORTS) < 0)
-		return false; /* downstream port status fetch failed */
-
-	return true;
-}
-
-static bool
-intel_dp_sink_can_mst(struct intel_dp *intel_dp)
-{
-	u8 mstm_cap;
-
-	if (intel_dp->dpcd[DP_DPCD_REV] < 0x12)
-		return false;
-
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_MSTM_CAP, &mstm_cap) != 1)
-		return false;
-
-	return mstm_cap & DP_MST_CAP;
+	return drm_dp_read_downstream_info(&intel_dp->aux, intel_dp->dpcd,
+					   intel_dp->downstream_ports) == 0;
 }
 
 static bool
@@ -4720,7 +4781,7 @@ intel_dp_can_mst(struct intel_dp *intel_dp)
 
 	return i915->params.enable_dp_mst &&
 		intel_dp->can_mst &&
-		intel_dp_sink_can_mst(intel_dp);
+		drm_dp_read_mst_cap(&intel_dp->aux, intel_dp->dpcd);
 }
 
 static void
@@ -4729,7 +4790,7 @@ intel_dp_configure_mst(struct intel_dp *intel_dp)
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	struct intel_encoder *encoder =
 		&dp_to_dig_port(intel_dp)->base;
-	bool sink_can_mst = intel_dp_sink_can_mst(intel_dp);
+	bool sink_can_mst = drm_dp_read_mst_cap(&intel_dp->aux, intel_dp->dpcd);
 
 	drm_dbg_kms(&i915->drm,
 		    "[ENCODER:%d:%s] MST support: port: %s, sink: %s, modparam: %s\n",
@@ -5963,9 +6024,8 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 		return connector_status_connected;
 
 	/* If we're HPD-aware, SINK_COUNT changes dynamically */
-	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
+	if (intel_dp_has_sink_count(intel_dp) &&
 	    intel_dp->downstream_ports[0] & DP_DS_PORT_HPD) {
-
 		return intel_dp->sink_count ?
 		connector_status_connected : connector_status_disconnected;
 	}
@@ -6106,16 +6166,103 @@ intel_dp_get_edid(struct intel_dp *intel_dp)
 }
 
 static void
+intel_dp_update_dfp(struct intel_dp *intel_dp,
+		    const struct edid *edid)
+{
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+	struct intel_connector *connector = intel_dp->attached_connector;
+
+	intel_dp->dfp.max_bpc =
+		drm_dp_downstream_max_bpc(intel_dp->dpcd,
+					  intel_dp->downstream_ports, edid);
+
+	intel_dp->dfp.max_dotclock =
+		drm_dp_downstream_max_dotclock(intel_dp->dpcd,
+					       intel_dp->downstream_ports);
+
+	intel_dp->dfp.min_tmds_clock =
+		drm_dp_downstream_min_tmds_clock(intel_dp->dpcd,
+						 intel_dp->downstream_ports,
+						 edid);
+	intel_dp->dfp.max_tmds_clock =
+		drm_dp_downstream_max_tmds_clock(intel_dp->dpcd,
+						 intel_dp->downstream_ports,
+						 edid);
+
+	drm_dbg_kms(&i915->drm,
+		    "[CONNECTOR:%d:%s] DFP max bpc %d, max dotclock %d, TMDS clock %d-%d\n",
+		    connector->base.base.id, connector->base.name,
+		    intel_dp->dfp.max_bpc,
+		    intel_dp->dfp.max_dotclock,
+		    intel_dp->dfp.min_tmds_clock,
+		    intel_dp->dfp.max_tmds_clock);
+}
+
+static void
+intel_dp_update_420(struct intel_dp *intel_dp)
+{
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+	struct intel_connector *connector = intel_dp->attached_connector;
+	bool is_branch, ycbcr_420_passthrough, ycbcr_444_to_420;
+
+	/* No YCbCr output support on gmch platforms */
+	if (HAS_GMCH(i915))
+		return;
+
+	/*
+	 * ILK doesn't seem capable of DP YCbCr output. The
+	 * displayed image is severly corrupted. SNB+ is fine.
+	 */
+	if (IS_GEN(i915, 5))
+		return;
+
+	is_branch = drm_dp_is_branch(intel_dp->dpcd);
+	ycbcr_420_passthrough =
+		drm_dp_downstream_420_passthrough(intel_dp->dpcd,
+						  intel_dp->downstream_ports);
+	ycbcr_444_to_420 =
+		drm_dp_downstream_444_to_420_conversion(intel_dp->dpcd,
+							intel_dp->downstream_ports);
+
+	if (INTEL_GEN(i915) >= 11) {
+		/* Prefer 4:2:0 passthrough over 4:4:4->4:2:0 conversion */
+		intel_dp->dfp.ycbcr_444_to_420 =
+			ycbcr_444_to_420 && !ycbcr_420_passthrough;
+
+		connector->base.ycbcr_420_allowed =
+			!is_branch || ycbcr_444_to_420 || ycbcr_420_passthrough;
+	} else {
+		/* 4:4:4->4:2:0 conversion is the only way */
+		intel_dp->dfp.ycbcr_444_to_420 = ycbcr_444_to_420;
+
+		connector->base.ycbcr_420_allowed = ycbcr_444_to_420;
+	}
+
+	drm_dbg_kms(&i915->drm,
+		    "[CONNECTOR:%d:%s] YCbCr 4:2:0 allowed? %s, YCbCr 4:4:4->4:2:0 conversion? %s\n",
+		    connector->base.base.id, connector->base.name,
+		    yesno(connector->base.ycbcr_420_allowed),
+		    yesno(intel_dp->dfp.ycbcr_444_to_420));
+}
+
+static void
 intel_dp_set_edid(struct intel_dp *intel_dp)
 {
-	struct intel_connector *intel_connector = intel_dp->attached_connector;
+	struct intel_connector *connector = intel_dp->attached_connector;
 	struct edid *edid;
 
 	intel_dp_unset_edid(intel_dp);
 	edid = intel_dp_get_edid(intel_dp);
-	intel_connector->detect_edid = edid;
+	connector->detect_edid = edid;
+
+	intel_dp_update_dfp(intel_dp, edid);
+	intel_dp_update_420(intel_dp);
+
+	if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) {
+		intel_dp->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
+		intel_dp->has_audio = drm_detect_monitor_audio(edid);
+	}
 
-	intel_dp->has_audio = drm_detect_monitor_audio(edid);
 	drm_dp_cec_set_edid(&intel_dp->aux, edid);
 	intel_dp->edid_quirks = drm_dp_get_edid_quirks(edid);
 }
@@ -6123,14 +6270,23 @@ intel_dp_set_edid(struct intel_dp *intel_dp)
 static void
 intel_dp_unset_edid(struct intel_dp *intel_dp)
 {
-	struct intel_connector *intel_connector = intel_dp->attached_connector;
+	struct intel_connector *connector = intel_dp->attached_connector;
 
 	drm_dp_cec_unset_edid(&intel_dp->aux);
-	kfree(intel_connector->detect_edid);
-	intel_connector->detect_edid = NULL;
+	kfree(connector->detect_edid);
+	connector->detect_edid = NULL;
 
+	intel_dp->has_hdmi_sink = false;
 	intel_dp->has_audio = false;
 	intel_dp->edid_quirks = 0;
+
+	intel_dp->dfp.max_bpc = 0;
+	intel_dp->dfp.max_dotclock = 0;
+	intel_dp->dfp.min_tmds_clock = 0;
+	intel_dp->dfp.max_tmds_clock = 0;
+
+	intel_dp->dfp.ycbcr_444_to_420 = false;
+	connector->base.ycbcr_420_allowed = false;
 }
 
 static int
@@ -6149,6 +6305,9 @@ intel_dp_detect(struct drm_connector *connector,
 	drm_WARN_ON(&dev_priv->drm,
 		    !drm_modeset_is_locked(&dev_priv->drm.mode_config.connection_mutex));
 
+	if (!INTEL_DISPLAY_ENABLED(dev_priv))
+		return connector_status_disconnected;
+
 	/* Can't disconnect eDP */
 	if (intel_dp_is_edp(intel_dp))
 		status = edp_detect(intel_dp);
@@ -6243,6 +6402,11 @@ out:
 	 */
 	intel_display_power_flush_work(dev_priv);
 
+	if (!intel_dp_is_edp(intel_dp))
+		drm_dp_set_subconnector_property(connector,
+						 status,
+						 intel_dp->dpcd,
+						 intel_dp->downstream_ports);
 	return status;
 }
 
@@ -6284,7 +6448,7 @@ static int intel_dp_get_modes(struct drm_connector *connector)
 	}
 
 	/* if eDP has no EDID, fall back to fixed mode */
-	if (intel_dp_is_edp(intel_attached_dp(to_intel_connector(connector))) &&
+	if (intel_dp_is_edp(intel_attached_dp(intel_connector)) &&
 	    intel_connector->panel.fixed_mode) {
 		struct drm_display_mode *mode;
 
@@ -6296,6 +6460,19 @@ static int intel_dp_get_modes(struct drm_connector *connector)
 		}
 	}
 
+	if (!edid) {
+		struct intel_dp *intel_dp = intel_attached_dp(intel_connector);
+		struct drm_display_mode *mode;
+
+		mode = drm_dp_downstream_mode(connector->dev,
+					      intel_dp->dpcd,
+					      intel_dp->downstream_ports);
+		if (mode) {
+			drm_mode_probed_add(connector, mode);
+			return 1;
+		}
+	}
+
 	return 0;
 }
 
@@ -6381,628 +6558,6 @@ void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder)
 		edp_panel_vdd_off_sync(intel_dp);
 }
 
-static void intel_dp_hdcp_wait_for_cp_irq(struct intel_hdcp *hdcp, int timeout)
-{
-	long ret;
-
-#define C (hdcp->cp_irq_count_cached != atomic_read(&hdcp->cp_irq_count))
-	ret = wait_event_interruptible_timeout(hdcp->cp_irq_queue, C,
-					       msecs_to_jiffies(timeout));
-
-	if (!ret)
-		DRM_DEBUG_KMS("Timedout at waiting for CP_IRQ\n");
-}
-
-static
-int intel_dp_hdcp_write_an_aksv(struct intel_digital_port *dig_port,
-				u8 *an)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	struct intel_dp *intel_dp = enc_to_intel_dp(to_intel_encoder(&dig_port->base.base));
-	static const struct drm_dp_aux_msg msg = {
-		.request = DP_AUX_NATIVE_WRITE,
-		.address = DP_AUX_HDCP_AKSV,
-		.size = DRM_HDCP_KSV_LEN,
-	};
-	u8 txbuf[HEADER_SIZE + DRM_HDCP_KSV_LEN] = {}, rxbuf[2], reply = 0;
-	ssize_t dpcd_ret;
-	int ret;
-
-	/* Output An first, that's easy */
-	dpcd_ret = drm_dp_dpcd_write(&dig_port->dp.aux, DP_AUX_HDCP_AN,
-				     an, DRM_HDCP_AN_LEN);
-	if (dpcd_ret != DRM_HDCP_AN_LEN) {
-		drm_dbg_kms(&i915->drm,
-			    "Failed to write An over DP/AUX (%zd)\n",
-			    dpcd_ret);
-		return dpcd_ret >= 0 ? -EIO : dpcd_ret;
-	}
-
-	/*
-	 * Since Aksv is Oh-So-Secret, we can't access it in software. So in
-	 * order to get it on the wire, we need to create the AUX header as if
-	 * we were writing the data, and then tickle the hardware to output the
-	 * data once the header is sent out.
-	 */
-	intel_dp_aux_header(txbuf, &msg);
-
-	ret = intel_dp_aux_xfer(intel_dp, txbuf, HEADER_SIZE + msg.size,
-				rxbuf, sizeof(rxbuf),
-				DP_AUX_CH_CTL_AUX_AKSV_SELECT);
-	if (ret < 0) {
-		drm_dbg_kms(&i915->drm,
-			    "Write Aksv over DP/AUX failed (%d)\n", ret);
-		return ret;
-	} else if (ret == 0) {
-		drm_dbg_kms(&i915->drm, "Aksv write over DP/AUX was empty\n");
-		return -EIO;
-	}
-
-	reply = (rxbuf[0] >> 4) & DP_AUX_NATIVE_REPLY_MASK;
-	if (reply != DP_AUX_NATIVE_REPLY_ACK) {
-		drm_dbg_kms(&i915->drm,
-			    "Aksv write: no DP_AUX_NATIVE_REPLY_ACK %x\n",
-			    reply);
-		return -EIO;
-	}
-	return 0;
-}
-
-static int intel_dp_hdcp_read_bksv(struct intel_digital_port *dig_port,
-				   u8 *bksv)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BKSV, bksv,
-			       DRM_HDCP_KSV_LEN);
-	if (ret != DRM_HDCP_KSV_LEN) {
-		drm_dbg_kms(&i915->drm,
-			    "Read Bksv from DP/AUX failed (%zd)\n", ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-	return 0;
-}
-
-static int intel_dp_hdcp_read_bstatus(struct intel_digital_port *dig_port,
-				      u8 *bstatus)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	/*
-	 * For some reason the HDMI and DP HDCP specs call this register
-	 * definition by different names. In the HDMI spec, it's called BSTATUS,
-	 * but in DP it's called BINFO.
-	 */
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BINFO,
-			       bstatus, DRM_HDCP_BSTATUS_LEN);
-	if (ret != DRM_HDCP_BSTATUS_LEN) {
-		drm_dbg_kms(&i915->drm,
-			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-	return 0;
-}
-
-static
-int intel_dp_hdcp_read_bcaps(struct intel_digital_port *dig_port,
-			     u8 *bcaps)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BCAPS,
-			       bcaps, 1);
-	if (ret != 1) {
-		drm_dbg_kms(&i915->drm,
-			    "Read bcaps from DP/AUX failed (%zd)\n", ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-
-	return 0;
-}
-
-static
-int intel_dp_hdcp_repeater_present(struct intel_digital_port *dig_port,
-				   bool *repeater_present)
-{
-	ssize_t ret;
-	u8 bcaps;
-
-	ret = intel_dp_hdcp_read_bcaps(dig_port, &bcaps);
-	if (ret)
-		return ret;
-
-	*repeater_present = bcaps & DP_BCAPS_REPEATER_PRESENT;
-	return 0;
-}
-
-static
-int intel_dp_hdcp_read_ri_prime(struct intel_digital_port *dig_port,
-				u8 *ri_prime)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_RI_PRIME,
-			       ri_prime, DRM_HDCP_RI_LEN);
-	if (ret != DRM_HDCP_RI_LEN) {
-		drm_dbg_kms(&i915->drm, "Read Ri' from DP/AUX failed (%zd)\n",
-			    ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-	return 0;
-}
-
-static
-int intel_dp_hdcp_read_ksv_ready(struct intel_digital_port *dig_port,
-				 bool *ksv_ready)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-	u8 bstatus;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BSTATUS,
-			       &bstatus, 1);
-	if (ret != 1) {
-		drm_dbg_kms(&i915->drm,
-			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-	*ksv_ready = bstatus & DP_BSTATUS_READY;
-	return 0;
-}
-
-static
-int intel_dp_hdcp_read_ksv_fifo(struct intel_digital_port *dig_port,
-				int num_downstream, u8 *ksv_fifo)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-	int i;
-
-	/* KSV list is read via 15 byte window (3 entries @ 5 bytes each) */
-	for (i = 0; i < num_downstream; i += 3) {
-		size_t len = min(num_downstream - i, 3) * DRM_HDCP_KSV_LEN;
-		ret = drm_dp_dpcd_read(&dig_port->dp.aux,
-				       DP_AUX_HDCP_KSV_FIFO,
-				       ksv_fifo + i * DRM_HDCP_KSV_LEN,
-				       len);
-		if (ret != len) {
-			drm_dbg_kms(&i915->drm,
-				    "Read ksv[%d] from DP/AUX failed (%zd)\n",
-				    i, ret);
-			return ret >= 0 ? -EIO : ret;
-		}
-	}
-	return 0;
-}
-
-static
-int intel_dp_hdcp_read_v_prime_part(struct intel_digital_port *dig_port,
-				    int i, u32 *part)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	if (i >= DRM_HDCP_V_PRIME_NUM_PARTS)
-		return -EINVAL;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
-			       DP_AUX_HDCP_V_PRIME(i), part,
-			       DRM_HDCP_V_PRIME_PART_LEN);
-	if (ret != DRM_HDCP_V_PRIME_PART_LEN) {
-		drm_dbg_kms(&i915->drm,
-			    "Read v'[%d] from DP/AUX failed (%zd)\n", i, ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-	return 0;
-}
-
-static
-int intel_dp_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
-				    bool enable)
-{
-	/* Not used for single stream DisplayPort setups */
-	return 0;
-}
-
-static
-bool intel_dp_hdcp_check_link(struct intel_digital_port *dig_port)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-	u8 bstatus;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BSTATUS,
-			       &bstatus, 1);
-	if (ret != 1) {
-		drm_dbg_kms(&i915->drm,
-			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
-		return false;
-	}
-
-	return !(bstatus & (DP_BSTATUS_LINK_FAILURE | DP_BSTATUS_REAUTH_REQ));
-}
-
-static
-int intel_dp_hdcp_capable(struct intel_digital_port *dig_port,
-			  bool *hdcp_capable)
-{
-	ssize_t ret;
-	u8 bcaps;
-
-	ret = intel_dp_hdcp_read_bcaps(dig_port, &bcaps);
-	if (ret)
-		return ret;
-
-	*hdcp_capable = bcaps & DP_BCAPS_HDCP_CAPABLE;
-	return 0;
-}
-
-struct hdcp2_dp_errata_stream_type {
-	u8	msg_id;
-	u8	stream_type;
-} __packed;
-
-struct hdcp2_dp_msg_data {
-	u8 msg_id;
-	u32 offset;
-	bool msg_detectable;
-	u32 timeout;
-	u32 timeout2; /* Added for non_paired situation */
-};
-
-static const struct hdcp2_dp_msg_data hdcp2_dp_msg_data[] = {
-	{ HDCP_2_2_AKE_INIT, DP_HDCP_2_2_AKE_INIT_OFFSET, false, 0, 0 },
-	{ HDCP_2_2_AKE_SEND_CERT, DP_HDCP_2_2_AKE_SEND_CERT_OFFSET,
-	  false, HDCP_2_2_CERT_TIMEOUT_MS, 0 },
-	{ HDCP_2_2_AKE_NO_STORED_KM, DP_HDCP_2_2_AKE_NO_STORED_KM_OFFSET,
-	  false, 0, 0 },
-	{ HDCP_2_2_AKE_STORED_KM, DP_HDCP_2_2_AKE_STORED_KM_OFFSET,
-	  false, 0, 0 },
-	{ HDCP_2_2_AKE_SEND_HPRIME, DP_HDCP_2_2_AKE_SEND_HPRIME_OFFSET,
-	  true, HDCP_2_2_HPRIME_PAIRED_TIMEOUT_MS,
-	  HDCP_2_2_HPRIME_NO_PAIRED_TIMEOUT_MS },
-	{ HDCP_2_2_AKE_SEND_PAIRING_INFO,
-	  DP_HDCP_2_2_AKE_SEND_PAIRING_INFO_OFFSET, true,
-	  HDCP_2_2_PAIRING_TIMEOUT_MS, 0 },
-	{ HDCP_2_2_LC_INIT, DP_HDCP_2_2_LC_INIT_OFFSET, false, 0, 0 },
-	{ HDCP_2_2_LC_SEND_LPRIME, DP_HDCP_2_2_LC_SEND_LPRIME_OFFSET,
-	  false, HDCP_2_2_DP_LPRIME_TIMEOUT_MS, 0 },
-	{ HDCP_2_2_SKE_SEND_EKS, DP_HDCP_2_2_SKE_SEND_EKS_OFFSET, false,
-	  0, 0 },
-	{ HDCP_2_2_REP_SEND_RECVID_LIST,
-	  DP_HDCP_2_2_REP_SEND_RECVID_LIST_OFFSET, true,
-	  HDCP_2_2_RECVID_LIST_TIMEOUT_MS, 0 },
-	{ HDCP_2_2_REP_SEND_ACK, DP_HDCP_2_2_REP_SEND_ACK_OFFSET, false,
-	  0, 0 },
-	{ HDCP_2_2_REP_STREAM_MANAGE,
-	  DP_HDCP_2_2_REP_STREAM_MANAGE_OFFSET, false,
-	  0, 0 },
-	{ HDCP_2_2_REP_STREAM_READY, DP_HDCP_2_2_REP_STREAM_READY_OFFSET,
-	  false, HDCP_2_2_STREAM_READY_TIMEOUT_MS, 0 },
-/* local define to shovel this through the write_2_2 interface */
-#define HDCP_2_2_ERRATA_DP_STREAM_TYPE	50
-	{ HDCP_2_2_ERRATA_DP_STREAM_TYPE,
-	  DP_HDCP_2_2_REG_STREAM_TYPE_OFFSET, false,
-	  0, 0 },
-};
-
-static int
-intel_dp_hdcp2_read_rx_status(struct intel_digital_port *dig_port,
-			      u8 *rx_status)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	ssize_t ret;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
-			       DP_HDCP_2_2_REG_RXSTATUS_OFFSET, rx_status,
-			       HDCP_2_2_DP_RXSTATUS_LEN);
-	if (ret != HDCP_2_2_DP_RXSTATUS_LEN) {
-		drm_dbg_kms(&i915->drm,
-			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
-		return ret >= 0 ? -EIO : ret;
-	}
-
-	return 0;
-}
-
-static
-int hdcp2_detect_msg_availability(struct intel_digital_port *dig_port,
-				  u8 msg_id, bool *msg_ready)
-{
-	u8 rx_status;
-	int ret;
-
-	*msg_ready = false;
-	ret = intel_dp_hdcp2_read_rx_status(dig_port, &rx_status);
-	if (ret < 0)
-		return ret;
-
-	switch (msg_id) {
-	case HDCP_2_2_AKE_SEND_HPRIME:
-		if (HDCP_2_2_DP_RXSTATUS_H_PRIME(rx_status))
-			*msg_ready = true;
-		break;
-	case HDCP_2_2_AKE_SEND_PAIRING_INFO:
-		if (HDCP_2_2_DP_RXSTATUS_PAIRING(rx_status))
-			*msg_ready = true;
-		break;
-	case HDCP_2_2_REP_SEND_RECVID_LIST:
-		if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
-			*msg_ready = true;
-		break;
-	default:
-		DRM_ERROR("Unidentified msg_id: %d\n", msg_id);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static ssize_t
-intel_dp_hdcp2_wait_for_msg(struct intel_digital_port *dig_port,
-			    const struct hdcp2_dp_msg_data *hdcp2_msg_data)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	struct intel_dp *dp = &dig_port->dp;
-	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
-	u8 msg_id = hdcp2_msg_data->msg_id;
-	int ret, timeout;
-	bool msg_ready = false;
-
-	if (msg_id == HDCP_2_2_AKE_SEND_HPRIME && !hdcp->is_paired)
-		timeout = hdcp2_msg_data->timeout2;
-	else
-		timeout = hdcp2_msg_data->timeout;
-
-	/*
-	 * There is no way to detect the CERT, LPRIME and STREAM_READY
-	 * availability. So Wait for timeout and read the msg.
-	 */
-	if (!hdcp2_msg_data->msg_detectable) {
-		mdelay(timeout);
-		ret = 0;
-	} else {
-		/*
-		 * As we want to check the msg availability at timeout, Ignoring
-		 * the timeout at wait for CP_IRQ.
-		 */
-		intel_dp_hdcp_wait_for_cp_irq(hdcp, timeout);
-		ret = hdcp2_detect_msg_availability(dig_port,
-						    msg_id, &msg_ready);
-		if (!msg_ready)
-			ret = -ETIMEDOUT;
-	}
-
-	if (ret)
-		drm_dbg_kms(&i915->drm,
-			    "msg_id %d, ret %d, timeout(mSec): %d\n",
-			    hdcp2_msg_data->msg_id, ret, timeout);
-
-	return ret;
-}
-
-static const struct hdcp2_dp_msg_data *get_hdcp2_dp_msg_data(u8 msg_id)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdcp2_dp_msg_data); i++)
-		if (hdcp2_dp_msg_data[i].msg_id == msg_id)
-			return &hdcp2_dp_msg_data[i];
-
-	return NULL;
-}
-
-static
-int intel_dp_hdcp2_write_msg(struct intel_digital_port *dig_port,
-			     void *buf, size_t size)
-{
-	struct intel_dp *dp = &dig_port->dp;
-	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
-	unsigned int offset;
-	u8 *byte = buf;
-	ssize_t ret, bytes_to_write, len;
-	const struct hdcp2_dp_msg_data *hdcp2_msg_data;
-
-	hdcp2_msg_data = get_hdcp2_dp_msg_data(*byte);
-	if (!hdcp2_msg_data)
-		return -EINVAL;
-
-	offset = hdcp2_msg_data->offset;
-
-	/* No msg_id in DP HDCP2.2 msgs */
-	bytes_to_write = size - 1;
-	byte++;
-
-	hdcp->cp_irq_count_cached = atomic_read(&hdcp->cp_irq_count);
-
-	while (bytes_to_write) {
-		len = bytes_to_write > DP_AUX_MAX_PAYLOAD_BYTES ?
-				DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_write;
-
-		ret = drm_dp_dpcd_write(&dig_port->dp.aux,
-					offset, (void *)byte, len);
-		if (ret < 0)
-			return ret;
-
-		bytes_to_write -= ret;
-		byte += ret;
-		offset += ret;
-	}
-
-	return size;
-}
-
-static
-ssize_t get_receiver_id_list_size(struct intel_digital_port *dig_port)
-{
-	u8 rx_info[HDCP_2_2_RXINFO_LEN];
-	u32 dev_cnt;
-	ssize_t ret;
-
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
-			       DP_HDCP_2_2_REG_RXINFO_OFFSET,
-			       (void *)rx_info, HDCP_2_2_RXINFO_LEN);
-	if (ret != HDCP_2_2_RXINFO_LEN)
-		return ret >= 0 ? -EIO : ret;
-
-	dev_cnt = (HDCP_2_2_DEV_COUNT_HI(rx_info[0]) << 4 |
-		   HDCP_2_2_DEV_COUNT_LO(rx_info[1]));
-
-	if (dev_cnt > HDCP_2_2_MAX_DEVICE_COUNT)
-		dev_cnt = HDCP_2_2_MAX_DEVICE_COUNT;
-
-	ret = sizeof(struct hdcp2_rep_send_receiverid_list) -
-		HDCP_2_2_RECEIVER_IDS_MAX_LEN +
-		(dev_cnt * HDCP_2_2_RECEIVER_ID_LEN);
-
-	return ret;
-}
-
-static
-int intel_dp_hdcp2_read_msg(struct intel_digital_port *dig_port,
-			    u8 msg_id, void *buf, size_t size)
-{
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	unsigned int offset;
-	u8 *byte = buf;
-	ssize_t ret, bytes_to_recv, len;
-	const struct hdcp2_dp_msg_data *hdcp2_msg_data;
-
-	hdcp2_msg_data = get_hdcp2_dp_msg_data(msg_id);
-	if (!hdcp2_msg_data)
-		return -EINVAL;
-	offset = hdcp2_msg_data->offset;
-
-	ret = intel_dp_hdcp2_wait_for_msg(dig_port, hdcp2_msg_data);
-	if (ret < 0)
-		return ret;
-
-	if (msg_id == HDCP_2_2_REP_SEND_RECVID_LIST) {
-		ret = get_receiver_id_list_size(dig_port);
-		if (ret < 0)
-			return ret;
-
-		size = ret;
-	}
-	bytes_to_recv = size - 1;
-
-	/* DP adaptation msgs has no msg_id */
-	byte++;
-
-	while (bytes_to_recv) {
-		len = bytes_to_recv > DP_AUX_MAX_PAYLOAD_BYTES ?
-		      DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_recv;
-
-		ret = drm_dp_dpcd_read(&dig_port->dp.aux, offset,
-				       (void *)byte, len);
-		if (ret < 0) {
-			drm_dbg_kms(&i915->drm, "msg_id %d, ret %zd\n",
-				    msg_id, ret);
-			return ret;
-		}
-
-		bytes_to_recv -= ret;
-		byte += ret;
-		offset += ret;
-	}
-	byte = buf;
-	*byte = msg_id;
-
-	return size;
-}
-
-static
-int intel_dp_hdcp2_config_stream_type(struct intel_digital_port *dig_port,
-				      bool is_repeater, u8 content_type)
-{
-	int ret;
-	struct hdcp2_dp_errata_stream_type stream_type_msg;
-
-	if (is_repeater)
-		return 0;
-
-	/*
-	 * Errata for DP: As Stream type is used for encryption, Receiver
-	 * should be communicated with stream type for the decryption of the
-	 * content.
-	 * Repeater will be communicated with stream type as a part of it's
-	 * auth later in time.
-	 */
-	stream_type_msg.msg_id = HDCP_2_2_ERRATA_DP_STREAM_TYPE;
-	stream_type_msg.stream_type = content_type;
-
-	ret =  intel_dp_hdcp2_write_msg(dig_port, &stream_type_msg,
-					sizeof(stream_type_msg));
-
-	return ret < 0 ? ret : 0;
-
-}
-
-static
-int intel_dp_hdcp2_check_link(struct intel_digital_port *dig_port)
-{
-	u8 rx_status;
-	int ret;
-
-	ret = intel_dp_hdcp2_read_rx_status(dig_port, &rx_status);
-	if (ret)
-		return ret;
-
-	if (HDCP_2_2_DP_RXSTATUS_REAUTH_REQ(rx_status))
-		ret = HDCP_REAUTH_REQUEST;
-	else if (HDCP_2_2_DP_RXSTATUS_LINK_FAILED(rx_status))
-		ret = HDCP_LINK_INTEGRITY_FAILURE;
-	else if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
-		ret = HDCP_TOPOLOGY_CHANGE;
-
-	return ret;
-}
-
-static
-int intel_dp_hdcp2_capable(struct intel_digital_port *dig_port,
-			   bool *capable)
-{
-	u8 rx_caps[3];
-	int ret;
-
-	*capable = false;
-	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
-			       DP_HDCP_2_2_REG_RX_CAPS_OFFSET,
-			       rx_caps, HDCP_2_2_RXCAPS_LEN);
-	if (ret != HDCP_2_2_RXCAPS_LEN)
-		return ret >= 0 ? -EIO : ret;
-
-	if (rx_caps[0] == HDCP_2_2_RX_CAPS_VERSION_VAL &&
-	    HDCP_2_2_DP_HDCP_CAPABLE(rx_caps[2]))
-		*capable = true;
-
-	return 0;
-}
-
-static const struct intel_hdcp_shim intel_dp_hdcp_shim = {
-	.write_an_aksv = intel_dp_hdcp_write_an_aksv,
-	.read_bksv = intel_dp_hdcp_read_bksv,
-	.read_bstatus = intel_dp_hdcp_read_bstatus,
-	.repeater_present = intel_dp_hdcp_repeater_present,
-	.read_ri_prime = intel_dp_hdcp_read_ri_prime,
-	.read_ksv_ready = intel_dp_hdcp_read_ksv_ready,
-	.read_ksv_fifo = intel_dp_hdcp_read_ksv_fifo,
-	.read_v_prime_part = intel_dp_hdcp_read_v_prime_part,
-	.toggle_signalling = intel_dp_hdcp_toggle_signalling,
-	.check_link = intel_dp_hdcp_check_link,
-	.hdcp_capable = intel_dp_hdcp_capable,
-	.write_2_2_msg = intel_dp_hdcp2_write_msg,
-	.read_2_2_msg = intel_dp_hdcp2_read_msg,
-	.config_stream_type = intel_dp_hdcp2_config_stream_type,
-	.check_2_2_link = intel_dp_hdcp2_check_link,
-	.hdcp_2_2_capable = intel_dp_hdcp2_capable,
-	.protocol = HDCP_PROTOCOL_DP,
-};
-
 static void intel_edp_panel_vdd_sanitize(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
@@ -7312,6 +6867,9 @@ intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connect
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	enum port port = dp_to_dig_port(intel_dp)->base.port;
 
+	if (!intel_dp_is_edp(intel_dp))
+		drm_connector_attach_dp_subconnector_property(connector);
+
 	if (!IS_G4X(dev_priv) && port != PORT_A)
 		intel_attach_force_audio_property(connector);
 
@@ -7710,6 +7268,15 @@ static void intel_dp_set_drrs_state(struct drm_i915_private *dev_priv,
 		    refresh_rate);
 }
 
+static void
+intel_edp_drrs_enable_locked(struct intel_dp *intel_dp)
+{
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+
+	dev_priv->drrs.busy_frontbuffer_bits = 0;
+	dev_priv->drrs.dp = intel_dp;
+}
+
 /**
  * intel_edp_drrs_enable - init drrs struct if supported
  * @intel_dp: DP struct
@@ -7722,31 +7289,40 @@ void intel_edp_drrs_enable(struct intel_dp *intel_dp,
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
 
-	if (!crtc_state->has_drrs) {
-		drm_dbg_kms(&dev_priv->drm, "Panel doesn't support DRRS\n");
+	if (!crtc_state->has_drrs)
 		return;
-	}
 
-	if (dev_priv->psr.enabled) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "PSR enabled. Not enabling DRRS.\n");
-		return;
-	}
+	drm_dbg_kms(&dev_priv->drm, "Enabling DRRS\n");
 
 	mutex_lock(&dev_priv->drrs.mutex);
+
 	if (dev_priv->drrs.dp) {
-		drm_dbg_kms(&dev_priv->drm, "DRRS already enabled\n");
+		drm_warn(&dev_priv->drm, "DRRS already enabled\n");
 		goto unlock;
 	}
 
-	dev_priv->drrs.busy_frontbuffer_bits = 0;
-
-	dev_priv->drrs.dp = intel_dp;
+	intel_edp_drrs_enable_locked(intel_dp);
 
 unlock:
 	mutex_unlock(&dev_priv->drrs.mutex);
 }
 
+static void
+intel_edp_drrs_disable_locked(struct intel_dp *intel_dp,
+			      const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+
+	if (dev_priv->drrs.refresh_rate_type == DRRS_LOW_RR) {
+		int refresh;
+
+		refresh = drm_mode_vrefresh(intel_dp->attached_connector->panel.fixed_mode);
+		intel_dp_set_drrs_state(dev_priv, crtc_state, refresh);
+	}
+
+	dev_priv->drrs.dp = NULL;
+}
+
 /**
  * intel_edp_drrs_disable - Disable DRRS
  * @intel_dp: DP struct
@@ -7767,16 +7343,45 @@ void intel_edp_drrs_disable(struct intel_dp *intel_dp,
 		return;
 	}
 
-	if (dev_priv->drrs.refresh_rate_type == DRRS_LOW_RR)
-		intel_dp_set_drrs_state(dev_priv, old_crtc_state,
-			drm_mode_vrefresh(intel_dp->attached_connector->panel.fixed_mode));
-
-	dev_priv->drrs.dp = NULL;
+	intel_edp_drrs_disable_locked(intel_dp, old_crtc_state);
 	mutex_unlock(&dev_priv->drrs.mutex);
 
 	cancel_delayed_work_sync(&dev_priv->drrs.work);
 }
 
+/**
+ * intel_edp_drrs_update - Update DRRS state
+ * @intel_dp: Intel DP
+ * @crtc_state: new CRTC state
+ *
+ * This function will update DRRS states, disabling or enabling DRRS when
+ * executing fastsets. For full modeset, intel_edp_drrs_disable() and
+ * intel_edp_drrs_enable() should be called instead.
+ */
+void
+intel_edp_drrs_update(struct intel_dp *intel_dp,
+		      const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+
+	if (dev_priv->drrs.type != SEAMLESS_DRRS_SUPPORT)
+		return;
+
+	mutex_lock(&dev_priv->drrs.mutex);
+
+	/* New state matches current one? */
+	if (crtc_state->has_drrs == !!dev_priv->drrs.dp)
+		goto unlock;
+
+	if (crtc_state->has_drrs)
+		intel_edp_drrs_enable_locked(intel_dp);
+	else
+		intel_edp_drrs_disable_locked(intel_dp, crtc_state);
+
+unlock:
+	mutex_unlock(&dev_priv->drrs.mutex);
+}
+
 static void intel_edp_drrs_downclock_work(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
@@ -8208,10 +7813,6 @@ intel_dp_init_connector(struct intel_digital_port *dig_port,
 		connector->interlace_allowed = true;
 	connector->doublescan_allowed = 0;
 
-	if (INTEL_GEN(dev_priv) >= 11)
-		connector->ycbcr_420_allowed = true;
-
-	intel_encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port);
 	intel_connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	intel_dp_aux_init(intel_dp);
@@ -8236,7 +7837,7 @@ intel_dp_init_connector(struct intel_digital_port *dig_port,
 	intel_dp_add_properties(intel_dp, connector);
 
 	if (is_hdcp_supported(dev_priv, port) && !intel_dp_is_edp(intel_dp)) {
-		int ret = intel_hdcp_init(intel_connector, &intel_dp_hdcp_shim);
+		int ret = intel_dp_init_hdcp(dig_port, intel_connector);
 		if (ret)
 			drm_dbg_kms(&dev_priv->drm,
 				    "HDCP init failed, skipping.\n");
@@ -8280,6 +7881,8 @@ bool intel_dp_init(struct drm_i915_private *dev_priv,
 	intel_encoder = &dig_port->base;
 	encoder = &intel_encoder->base;
 
+	mutex_init(&dig_port->hdcp_mutex);
+
 	if (drm_encoder_init(&dev_priv->drm, &intel_encoder->base,
 			     &intel_dp_enc_funcs, DRM_MODE_ENCODER_TMDS,
 			     "DP %c", port_name(port)))
@@ -8354,6 +7957,7 @@ bool intel_dp_init(struct drm_i915_private *dev_priv,
 	}
 	intel_encoder->cloneable = 0;
 	intel_encoder->port = port;
+	intel_encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port);
 
 	dig_port->hpd_pulse = intel_dp_hpd_pulse;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index b901ab850cbd..08a1c0aa8b94 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -17,6 +17,7 @@ struct drm_encoder;
 struct drm_i915_private;
 struct drm_modeset_acquire_ctx;
 struct drm_dp_vsc_sdp;
+struct intel_atomic_state;
 struct intel_connector;
 struct intel_crtc_state;
 struct intel_digital_port;
@@ -50,6 +51,7 @@ int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp,
 int intel_dp_retrain_link(struct intel_encoder *encoder,
 			  struct drm_modeset_acquire_ctx *ctx);
 void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
+void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp);
 void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp,
 					   const struct intel_crtc_state *crtc_state,
 					   bool enable);
@@ -81,6 +83,8 @@ void intel_edp_drrs_enable(struct intel_dp *intel_dp,
 			   const struct intel_crtc_state *crtc_state);
 void intel_edp_drrs_disable(struct intel_dp *intel_dp,
 			    const struct intel_crtc_state *crtc_state);
+void intel_edp_drrs_update(struct intel_dp *intel_dp,
+			   const struct intel_crtc_state *crtc_state);
 void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv,
 			       unsigned int frontbuffer_bits);
 void intel_edp_drrs_flush(struct drm_i915_private *dev_priv,
@@ -99,7 +103,6 @@ bool intel_dp_source_supports_hbr3(struct intel_dp *intel_dp);
 bool
 intel_dp_get_link_status(struct intel_dp *intel_dp, u8 *link_status);
 
-bool intel_dp_read_dpcd(struct intel_dp *intel_dp);
 bool intel_dp_get_colorimetry_status(struct intel_dp *intel_dp);
 int intel_dp_link_required(int pixel_clock, int bpp);
 int intel_dp_max_data_rate(int max_link_clock, int max_lanes);
@@ -128,4 +131,12 @@ static inline unsigned int intel_dp_unused_lane_mask(int lane_count)
 
 u32 intel_dp_mode_to_fec_clock(u32 mode_clock);
 
+void intel_ddi_update_pipe(struct intel_atomic_state *state,
+			   struct intel_encoder *encoder,
+			   const struct intel_crtc_state *crtc_state,
+			   const struct drm_connector_state *conn_state);
+
+int intel_dp_init_hdcp(struct intel_digital_port *dig_port,
+		       struct intel_connector *intel_connector);
+
 #endif /* __INTEL_DP_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_dp_hdcp.c b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c
new file mode 100644
index 000000000000..03424d20e9f7
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c
@@ -0,0 +1,703 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright (C) 2020 Google, Inc.
+ *
+ * Authors:
+ * Sean Paul <seanpaul@chromium.org>
+ */
+
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_dp_mst_helper.h>
+#include <drm/drm_hdcp.h>
+#include <drm/drm_print.h>
+
+#include "intel_display_types.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
+#include "intel_hdcp.h"
+
+static void intel_dp_hdcp_wait_for_cp_irq(struct intel_hdcp *hdcp, int timeout)
+{
+	long ret;
+
+#define C (hdcp->cp_irq_count_cached != atomic_read(&hdcp->cp_irq_count))
+	ret = wait_event_interruptible_timeout(hdcp->cp_irq_queue, C,
+					       msecs_to_jiffies(timeout));
+
+	if (!ret)
+		DRM_DEBUG_KMS("Timedout at waiting for CP_IRQ\n");
+}
+
+static
+int intel_dp_hdcp_write_an_aksv(struct intel_digital_port *dig_port,
+				u8 *an)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	u8 aksv[DRM_HDCP_KSV_LEN] = {};
+	ssize_t dpcd_ret;
+
+	/* Output An first, that's easy */
+	dpcd_ret = drm_dp_dpcd_write(&dig_port->dp.aux, DP_AUX_HDCP_AN,
+				     an, DRM_HDCP_AN_LEN);
+	if (dpcd_ret != DRM_HDCP_AN_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Failed to write An over DP/AUX (%zd)\n",
+			    dpcd_ret);
+		return dpcd_ret >= 0 ? -EIO : dpcd_ret;
+	}
+
+	/*
+	 * Since Aksv is Oh-So-Secret, we can't access it in software. So we
+	 * send an empty buffer of the correct length through the DP helpers. On
+	 * the other side, in the transfer hook, we'll generate a flag based on
+	 * the destination address which will tickle the hardware to output the
+	 * Aksv on our behalf after the header is sent.
+	 */
+	dpcd_ret = drm_dp_dpcd_write(&dig_port->dp.aux, DP_AUX_HDCP_AKSV,
+				     aksv, DRM_HDCP_KSV_LEN);
+	if (dpcd_ret != DRM_HDCP_KSV_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Failed to write Aksv over DP/AUX (%zd)\n",
+			    dpcd_ret);
+		return dpcd_ret >= 0 ? -EIO : dpcd_ret;
+	}
+	return 0;
+}
+
+static int intel_dp_hdcp_read_bksv(struct intel_digital_port *dig_port,
+				   u8 *bksv)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BKSV, bksv,
+			       DRM_HDCP_KSV_LEN);
+	if (ret != DRM_HDCP_KSV_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Read Bksv from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+	return 0;
+}
+
+static int intel_dp_hdcp_read_bstatus(struct intel_digital_port *dig_port,
+				      u8 *bstatus)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	/*
+	 * For some reason the HDMI and DP HDCP specs call this register
+	 * definition by different names. In the HDMI spec, it's called BSTATUS,
+	 * but in DP it's called BINFO.
+	 */
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BINFO,
+			       bstatus, DRM_HDCP_BSTATUS_LEN);
+	if (ret != DRM_HDCP_BSTATUS_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+	return 0;
+}
+
+static
+int intel_dp_hdcp_read_bcaps(struct intel_digital_port *dig_port,
+			     u8 *bcaps)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BCAPS,
+			       bcaps, 1);
+	if (ret != 1) {
+		drm_dbg_kms(&i915->drm,
+			    "Read bcaps from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+
+	return 0;
+}
+
+static
+int intel_dp_hdcp_repeater_present(struct intel_digital_port *dig_port,
+				   bool *repeater_present)
+{
+	ssize_t ret;
+	u8 bcaps;
+
+	ret = intel_dp_hdcp_read_bcaps(dig_port, &bcaps);
+	if (ret)
+		return ret;
+
+	*repeater_present = bcaps & DP_BCAPS_REPEATER_PRESENT;
+	return 0;
+}
+
+static
+int intel_dp_hdcp_read_ri_prime(struct intel_digital_port *dig_port,
+				u8 *ri_prime)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_RI_PRIME,
+			       ri_prime, DRM_HDCP_RI_LEN);
+	if (ret != DRM_HDCP_RI_LEN) {
+		drm_dbg_kms(&i915->drm, "Read Ri' from DP/AUX failed (%zd)\n",
+			    ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+	return 0;
+}
+
+static
+int intel_dp_hdcp_read_ksv_ready(struct intel_digital_port *dig_port,
+				 bool *ksv_ready)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+	u8 bstatus;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BSTATUS,
+			       &bstatus, 1);
+	if (ret != 1) {
+		drm_dbg_kms(&i915->drm,
+			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+	*ksv_ready = bstatus & DP_BSTATUS_READY;
+	return 0;
+}
+
+static
+int intel_dp_hdcp_read_ksv_fifo(struct intel_digital_port *dig_port,
+				int num_downstream, u8 *ksv_fifo)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+	int i;
+
+	/* KSV list is read via 15 byte window (3 entries @ 5 bytes each) */
+	for (i = 0; i < num_downstream; i += 3) {
+		size_t len = min(num_downstream - i, 3) * DRM_HDCP_KSV_LEN;
+		ret = drm_dp_dpcd_read(&dig_port->dp.aux,
+				       DP_AUX_HDCP_KSV_FIFO,
+				       ksv_fifo + i * DRM_HDCP_KSV_LEN,
+				       len);
+		if (ret != len) {
+			drm_dbg_kms(&i915->drm,
+				    "Read ksv[%d] from DP/AUX failed (%zd)\n",
+				    i, ret);
+			return ret >= 0 ? -EIO : ret;
+		}
+	}
+	return 0;
+}
+
+static
+int intel_dp_hdcp_read_v_prime_part(struct intel_digital_port *dig_port,
+				    int i, u32 *part)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	if (i >= DRM_HDCP_V_PRIME_NUM_PARTS)
+		return -EINVAL;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
+			       DP_AUX_HDCP_V_PRIME(i), part,
+			       DRM_HDCP_V_PRIME_PART_LEN);
+	if (ret != DRM_HDCP_V_PRIME_PART_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Read v'[%d] from DP/AUX failed (%zd)\n", i, ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+	return 0;
+}
+
+static
+int intel_dp_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
+				    enum transcoder cpu_transcoder,
+				    bool enable)
+{
+	/* Not used for single stream DisplayPort setups */
+	return 0;
+}
+
+static
+bool intel_dp_hdcp_check_link(struct intel_digital_port *dig_port,
+			      struct intel_connector *connector)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+	u8 bstatus;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux, DP_AUX_HDCP_BSTATUS,
+			       &bstatus, 1);
+	if (ret != 1) {
+		drm_dbg_kms(&i915->drm,
+			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
+		return false;
+	}
+
+	return !(bstatus & (DP_BSTATUS_LINK_FAILURE | DP_BSTATUS_REAUTH_REQ));
+}
+
+static
+int intel_dp_hdcp_capable(struct intel_digital_port *dig_port,
+			  bool *hdcp_capable)
+{
+	ssize_t ret;
+	u8 bcaps;
+
+	ret = intel_dp_hdcp_read_bcaps(dig_port, &bcaps);
+	if (ret)
+		return ret;
+
+	*hdcp_capable = bcaps & DP_BCAPS_HDCP_CAPABLE;
+	return 0;
+}
+
+struct hdcp2_dp_errata_stream_type {
+	u8	msg_id;
+	u8	stream_type;
+} __packed;
+
+struct hdcp2_dp_msg_data {
+	u8 msg_id;
+	u32 offset;
+	bool msg_detectable;
+	u32 timeout;
+	u32 timeout2; /* Added for non_paired situation */
+};
+
+static const struct hdcp2_dp_msg_data hdcp2_dp_msg_data[] = {
+	{ HDCP_2_2_AKE_INIT, DP_HDCP_2_2_AKE_INIT_OFFSET, false, 0, 0 },
+	{ HDCP_2_2_AKE_SEND_CERT, DP_HDCP_2_2_AKE_SEND_CERT_OFFSET,
+	  false, HDCP_2_2_CERT_TIMEOUT_MS, 0 },
+	{ HDCP_2_2_AKE_NO_STORED_KM, DP_HDCP_2_2_AKE_NO_STORED_KM_OFFSET,
+	  false, 0, 0 },
+	{ HDCP_2_2_AKE_STORED_KM, DP_HDCP_2_2_AKE_STORED_KM_OFFSET,
+	  false, 0, 0 },
+	{ HDCP_2_2_AKE_SEND_HPRIME, DP_HDCP_2_2_AKE_SEND_HPRIME_OFFSET,
+	  true, HDCP_2_2_HPRIME_PAIRED_TIMEOUT_MS,
+	  HDCP_2_2_HPRIME_NO_PAIRED_TIMEOUT_MS },
+	{ HDCP_2_2_AKE_SEND_PAIRING_INFO,
+	  DP_HDCP_2_2_AKE_SEND_PAIRING_INFO_OFFSET, true,
+	  HDCP_2_2_PAIRING_TIMEOUT_MS, 0 },
+	{ HDCP_2_2_LC_INIT, DP_HDCP_2_2_LC_INIT_OFFSET, false, 0, 0 },
+	{ HDCP_2_2_LC_SEND_LPRIME, DP_HDCP_2_2_LC_SEND_LPRIME_OFFSET,
+	  false, HDCP_2_2_DP_LPRIME_TIMEOUT_MS, 0 },
+	{ HDCP_2_2_SKE_SEND_EKS, DP_HDCP_2_2_SKE_SEND_EKS_OFFSET, false,
+	  0, 0 },
+	{ HDCP_2_2_REP_SEND_RECVID_LIST,
+	  DP_HDCP_2_2_REP_SEND_RECVID_LIST_OFFSET, true,
+	  HDCP_2_2_RECVID_LIST_TIMEOUT_MS, 0 },
+	{ HDCP_2_2_REP_SEND_ACK, DP_HDCP_2_2_REP_SEND_ACK_OFFSET, false,
+	  0, 0 },
+	{ HDCP_2_2_REP_STREAM_MANAGE,
+	  DP_HDCP_2_2_REP_STREAM_MANAGE_OFFSET, false,
+	  0, 0 },
+	{ HDCP_2_2_REP_STREAM_READY, DP_HDCP_2_2_REP_STREAM_READY_OFFSET,
+	  false, HDCP_2_2_STREAM_READY_TIMEOUT_MS, 0 },
+/* local define to shovel this through the write_2_2 interface */
+#define HDCP_2_2_ERRATA_DP_STREAM_TYPE	50
+	{ HDCP_2_2_ERRATA_DP_STREAM_TYPE,
+	  DP_HDCP_2_2_REG_STREAM_TYPE_OFFSET, false,
+	  0, 0 },
+};
+
+static int
+intel_dp_hdcp2_read_rx_status(struct intel_digital_port *dig_port,
+			      u8 *rx_status)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RXSTATUS_OFFSET, rx_status,
+			       HDCP_2_2_DP_RXSTATUS_LEN);
+	if (ret != HDCP_2_2_DP_RXSTATUS_LEN) {
+		drm_dbg_kms(&i915->drm,
+			    "Read bstatus from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+
+	return 0;
+}
+
+static
+int hdcp2_detect_msg_availability(struct intel_digital_port *dig_port,
+				  u8 msg_id, bool *msg_ready)
+{
+	u8 rx_status;
+	int ret;
+
+	*msg_ready = false;
+	ret = intel_dp_hdcp2_read_rx_status(dig_port, &rx_status);
+	if (ret < 0)
+		return ret;
+
+	switch (msg_id) {
+	case HDCP_2_2_AKE_SEND_HPRIME:
+		if (HDCP_2_2_DP_RXSTATUS_H_PRIME(rx_status))
+			*msg_ready = true;
+		break;
+	case HDCP_2_2_AKE_SEND_PAIRING_INFO:
+		if (HDCP_2_2_DP_RXSTATUS_PAIRING(rx_status))
+			*msg_ready = true;
+		break;
+	case HDCP_2_2_REP_SEND_RECVID_LIST:
+		if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
+			*msg_ready = true;
+		break;
+	default:
+		DRM_ERROR("Unidentified msg_id: %d\n", msg_id);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static ssize_t
+intel_dp_hdcp2_wait_for_msg(struct intel_digital_port *dig_port,
+			    const struct hdcp2_dp_msg_data *hdcp2_msg_data)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_dp *dp = &dig_port->dp;
+	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
+	u8 msg_id = hdcp2_msg_data->msg_id;
+	int ret, timeout;
+	bool msg_ready = false;
+
+	if (msg_id == HDCP_2_2_AKE_SEND_HPRIME && !hdcp->is_paired)
+		timeout = hdcp2_msg_data->timeout2;
+	else
+		timeout = hdcp2_msg_data->timeout;
+
+	/*
+	 * There is no way to detect the CERT, LPRIME and STREAM_READY
+	 * availability. So Wait for timeout and read the msg.
+	 */
+	if (!hdcp2_msg_data->msg_detectable) {
+		mdelay(timeout);
+		ret = 0;
+	} else {
+		/*
+		 * As we want to check the msg availability at timeout, Ignoring
+		 * the timeout at wait for CP_IRQ.
+		 */
+		intel_dp_hdcp_wait_for_cp_irq(hdcp, timeout);
+		ret = hdcp2_detect_msg_availability(dig_port,
+						    msg_id, &msg_ready);
+		if (!msg_ready)
+			ret = -ETIMEDOUT;
+	}
+
+	if (ret)
+		drm_dbg_kms(&i915->drm,
+			    "msg_id %d, ret %d, timeout(mSec): %d\n",
+			    hdcp2_msg_data->msg_id, ret, timeout);
+
+	return ret;
+}
+
+static const struct hdcp2_dp_msg_data *get_hdcp2_dp_msg_data(u8 msg_id)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdcp2_dp_msg_data); i++)
+		if (hdcp2_dp_msg_data[i].msg_id == msg_id)
+			return &hdcp2_dp_msg_data[i];
+
+	return NULL;
+}
+
+static
+int intel_dp_hdcp2_write_msg(struct intel_digital_port *dig_port,
+			     void *buf, size_t size)
+{
+	struct intel_dp *dp = &dig_port->dp;
+	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
+	unsigned int offset;
+	u8 *byte = buf;
+	ssize_t ret, bytes_to_write, len;
+	const struct hdcp2_dp_msg_data *hdcp2_msg_data;
+
+	hdcp2_msg_data = get_hdcp2_dp_msg_data(*byte);
+	if (!hdcp2_msg_data)
+		return -EINVAL;
+
+	offset = hdcp2_msg_data->offset;
+
+	/* No msg_id in DP HDCP2.2 msgs */
+	bytes_to_write = size - 1;
+	byte++;
+
+	hdcp->cp_irq_count_cached = atomic_read(&hdcp->cp_irq_count);
+
+	while (bytes_to_write) {
+		len = bytes_to_write > DP_AUX_MAX_PAYLOAD_BYTES ?
+				DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_write;
+
+		ret = drm_dp_dpcd_write(&dig_port->dp.aux,
+					offset, (void *)byte, len);
+		if (ret < 0)
+			return ret;
+
+		bytes_to_write -= ret;
+		byte += ret;
+		offset += ret;
+	}
+
+	return size;
+}
+
+static
+ssize_t get_receiver_id_list_size(struct intel_digital_port *dig_port)
+{
+	u8 rx_info[HDCP_2_2_RXINFO_LEN];
+	u32 dev_cnt;
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RXINFO_OFFSET,
+			       (void *)rx_info, HDCP_2_2_RXINFO_LEN);
+	if (ret != HDCP_2_2_RXINFO_LEN)
+		return ret >= 0 ? -EIO : ret;
+
+	dev_cnt = (HDCP_2_2_DEV_COUNT_HI(rx_info[0]) << 4 |
+		   HDCP_2_2_DEV_COUNT_LO(rx_info[1]));
+
+	if (dev_cnt > HDCP_2_2_MAX_DEVICE_COUNT)
+		dev_cnt = HDCP_2_2_MAX_DEVICE_COUNT;
+
+	ret = sizeof(struct hdcp2_rep_send_receiverid_list) -
+		HDCP_2_2_RECEIVER_IDS_MAX_LEN +
+		(dev_cnt * HDCP_2_2_RECEIVER_ID_LEN);
+
+	return ret;
+}
+
+static
+int intel_dp_hdcp2_read_msg(struct intel_digital_port *dig_port,
+			    u8 msg_id, void *buf, size_t size)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	unsigned int offset;
+	u8 *byte = buf;
+	ssize_t ret, bytes_to_recv, len;
+	const struct hdcp2_dp_msg_data *hdcp2_msg_data;
+
+	hdcp2_msg_data = get_hdcp2_dp_msg_data(msg_id);
+	if (!hdcp2_msg_data)
+		return -EINVAL;
+	offset = hdcp2_msg_data->offset;
+
+	ret = intel_dp_hdcp2_wait_for_msg(dig_port, hdcp2_msg_data);
+	if (ret < 0)
+		return ret;
+
+	if (msg_id == HDCP_2_2_REP_SEND_RECVID_LIST) {
+		ret = get_receiver_id_list_size(dig_port);
+		if (ret < 0)
+			return ret;
+
+		size = ret;
+	}
+	bytes_to_recv = size - 1;
+
+	/* DP adaptation msgs has no msg_id */
+	byte++;
+
+	while (bytes_to_recv) {
+		len = bytes_to_recv > DP_AUX_MAX_PAYLOAD_BYTES ?
+		      DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_recv;
+
+		ret = drm_dp_dpcd_read(&dig_port->dp.aux, offset,
+				       (void *)byte, len);
+		if (ret < 0) {
+			drm_dbg_kms(&i915->drm, "msg_id %d, ret %zd\n",
+				    msg_id, ret);
+			return ret;
+		}
+
+		bytes_to_recv -= ret;
+		byte += ret;
+		offset += ret;
+	}
+	byte = buf;
+	*byte = msg_id;
+
+	return size;
+}
+
+static
+int intel_dp_hdcp2_config_stream_type(struct intel_digital_port *dig_port,
+				      bool is_repeater, u8 content_type)
+{
+	int ret;
+	struct hdcp2_dp_errata_stream_type stream_type_msg;
+
+	if (is_repeater)
+		return 0;
+
+	/*
+	 * Errata for DP: As Stream type is used for encryption, Receiver
+	 * should be communicated with stream type for the decryption of the
+	 * content.
+	 * Repeater will be communicated with stream type as a part of it's
+	 * auth later in time.
+	 */
+	stream_type_msg.msg_id = HDCP_2_2_ERRATA_DP_STREAM_TYPE;
+	stream_type_msg.stream_type = content_type;
+
+	ret =  intel_dp_hdcp2_write_msg(dig_port, &stream_type_msg,
+					sizeof(stream_type_msg));
+
+	return ret < 0 ? ret : 0;
+
+}
+
+static
+int intel_dp_hdcp2_check_link(struct intel_digital_port *dig_port)
+{
+	u8 rx_status;
+	int ret;
+
+	ret = intel_dp_hdcp2_read_rx_status(dig_port, &rx_status);
+	if (ret)
+		return ret;
+
+	if (HDCP_2_2_DP_RXSTATUS_REAUTH_REQ(rx_status))
+		ret = HDCP_REAUTH_REQUEST;
+	else if (HDCP_2_2_DP_RXSTATUS_LINK_FAILED(rx_status))
+		ret = HDCP_LINK_INTEGRITY_FAILURE;
+	else if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
+		ret = HDCP_TOPOLOGY_CHANGE;
+
+	return ret;
+}
+
+static
+int intel_dp_hdcp2_capable(struct intel_digital_port *dig_port,
+			   bool *capable)
+{
+	u8 rx_caps[3];
+	int ret;
+
+	*capable = false;
+	ret = drm_dp_dpcd_read(&dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RX_CAPS_OFFSET,
+			       rx_caps, HDCP_2_2_RXCAPS_LEN);
+	if (ret != HDCP_2_2_RXCAPS_LEN)
+		return ret >= 0 ? -EIO : ret;
+
+	if (rx_caps[0] == HDCP_2_2_RX_CAPS_VERSION_VAL &&
+	    HDCP_2_2_DP_HDCP_CAPABLE(rx_caps[2]))
+		*capable = true;
+
+	return 0;
+}
+
+static const struct intel_hdcp_shim intel_dp_hdcp_shim = {
+	.write_an_aksv = intel_dp_hdcp_write_an_aksv,
+	.read_bksv = intel_dp_hdcp_read_bksv,
+	.read_bstatus = intel_dp_hdcp_read_bstatus,
+	.repeater_present = intel_dp_hdcp_repeater_present,
+	.read_ri_prime = intel_dp_hdcp_read_ri_prime,
+	.read_ksv_ready = intel_dp_hdcp_read_ksv_ready,
+	.read_ksv_fifo = intel_dp_hdcp_read_ksv_fifo,
+	.read_v_prime_part = intel_dp_hdcp_read_v_prime_part,
+	.toggle_signalling = intel_dp_hdcp_toggle_signalling,
+	.check_link = intel_dp_hdcp_check_link,
+	.hdcp_capable = intel_dp_hdcp_capable,
+	.write_2_2_msg = intel_dp_hdcp2_write_msg,
+	.read_2_2_msg = intel_dp_hdcp2_read_msg,
+	.config_stream_type = intel_dp_hdcp2_config_stream_type,
+	.check_2_2_link = intel_dp_hdcp2_check_link,
+	.hdcp_2_2_capable = intel_dp_hdcp2_capable,
+	.protocol = HDCP_PROTOCOL_DP,
+};
+
+static int
+intel_dp_mst_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
+				    enum transcoder cpu_transcoder,
+				    bool enable)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	int ret;
+
+	if (!enable)
+		usleep_range(6, 60); /* Bspec says >= 6us */
+
+	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base,
+					       cpu_transcoder, enable);
+	if (ret)
+		drm_dbg_kms(&i915->drm, "%s HDCP signalling failed (%d)\n",
+			      enable ? "Enable" : "Disable", ret);
+	return ret;
+}
+
+static
+bool intel_dp_mst_hdcp_check_link(struct intel_digital_port *dig_port,
+				  struct intel_connector *connector)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_dp *intel_dp = &dig_port->dp;
+	struct drm_dp_query_stream_enc_status_ack_reply reply;
+	int ret;
+
+	if (!intel_dp_hdcp_check_link(dig_port, connector))
+		return false;
+
+	ret = drm_dp_send_query_stream_enc_status(&intel_dp->mst_mgr,
+						  connector->port, &reply);
+	if (ret) {
+		drm_dbg_kms(&i915->drm,
+			    "[CONNECTOR:%d:%s] failed QSES ret=%d\n",
+			    connector->base.base.id, connector->base.name, ret);
+		return false;
+	}
+
+	return reply.auth_completed && reply.encryption_enabled;
+}
+
+static const struct intel_hdcp_shim intel_dp_mst_hdcp_shim = {
+	.write_an_aksv = intel_dp_hdcp_write_an_aksv,
+	.read_bksv = intel_dp_hdcp_read_bksv,
+	.read_bstatus = intel_dp_hdcp_read_bstatus,
+	.repeater_present = intel_dp_hdcp_repeater_present,
+	.read_ri_prime = intel_dp_hdcp_read_ri_prime,
+	.read_ksv_ready = intel_dp_hdcp_read_ksv_ready,
+	.read_ksv_fifo = intel_dp_hdcp_read_ksv_fifo,
+	.read_v_prime_part = intel_dp_hdcp_read_v_prime_part,
+	.toggle_signalling = intel_dp_mst_hdcp_toggle_signalling,
+	.check_link = intel_dp_mst_hdcp_check_link,
+	.hdcp_capable = intel_dp_hdcp_capable,
+
+	.protocol = HDCP_PROTOCOL_DP,
+};
+
+int intel_dp_init_hdcp(struct intel_digital_port *dig_port,
+		       struct intel_connector *intel_connector)
+{
+	struct drm_device *dev = intel_connector->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_encoder *intel_encoder = &dig_port->base;
+	enum port port = intel_encoder->port;
+	struct intel_dp *intel_dp = &dig_port->dp;
+
+	if (!is_hdcp_supported(dev_priv, port))
+		return 0;
+
+	if (intel_connector->mst_port)
+		return intel_hdcp_init(intel_connector, port,
+				       &intel_dp_mst_hdcp_shim);
+	else if (!intel_dp_is_edp(intel_dp))
+		return intel_hdcp_init(intel_connector, port,
+				       &intel_dp_hdcp_shim);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index a23ed7290843..f2c8b56be9ea 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -410,10 +410,17 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		    intel_connector->base.base.id,
 		    intel_connector->base.name,
 		    intel_dp->link_rate, intel_dp->lane_count);
-	if (!intel_dp_get_link_train_fallback_values(intel_dp,
-						     intel_dp->link_rate,
-						     intel_dp->lane_count))
-		/* Schedule a Hotplug Uevent to userspace to start modeset */
-		schedule_work(&intel_connector->modeset_retry_work);
-	return;
+
+	if (intel_dp->hobl_active) {
+		drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
+			    "Link Training failed with HOBL active, not enabling it from now on");
+		intel_dp->hobl_failed = true;
+	} else if (intel_dp_get_link_train_fallback_values(intel_dp,
+							   intel_dp->link_rate,
+							   intel_dp->lane_count)) {
+		return;
+	}
+
+	/* Schedule a Hotplug Uevent to userspace to start modeset */
+	schedule_work(&intel_connector->modeset_retry_work);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index a2d91a499700..64d885539e94 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -37,6 +37,7 @@
 #include "intel_dp.h"
 #include "intel_dp_mst.h"
 #include "intel_dpio_phy.h"
+#include "intel_hdcp.h"
 
 static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder,
 					    struct intel_crtc_state *crtc_state,
@@ -352,6 +353,8 @@ static void intel_mst_disable_dp(struct intel_atomic_state *state,
 	drm_dbg_kms(&i915->drm, "active links %d\n",
 		    intel_dp->active_mst_links);
 
+	intel_hdcp_disable(intel_mst->connector);
+
 	drm_dp_mst_reset_vcpi_slots(&intel_dp->mst_mgr, connector->port);
 
 	ret = drm_dp_update_payload_part1(&intel_dp->mst_mgr);
@@ -556,6 +559,13 @@ static void intel_mst_enable_dp(struct intel_atomic_state *state,
 
 	if (pipe_config->has_audio)
 		intel_audio_codec_enable(encoder, pipe_config, conn_state);
+
+	/* Enable hdcp if it's desired */
+	if (conn_state->content_protection ==
+	    DRM_MODE_CONTENT_PROTECTION_DESIRED)
+		intel_hdcp_enable(to_intel_connector(conn_state->connector),
+				  pipe_config->cpu_transcoder,
+				  (u8)conn_state->hdcp_content_type);
 }
 
 static bool intel_dp_mst_enc_get_hw_state(struct intel_encoder *encoder,
@@ -709,9 +719,13 @@ static int
 intel_dp_mst_detect(struct drm_connector *connector,
 		    struct drm_modeset_acquire_ctx *ctx, bool force)
 {
+	struct drm_i915_private *i915 = to_i915(connector->dev);
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_dp *intel_dp = intel_connector->mst_port;
 
+	if (!INTEL_DISPLAY_ENABLED(i915))
+		return connector_status_disconnected;
+
 	if (drm_connector_is_unregistered(connector))
 		return connector_status_disconnected;
 
@@ -799,6 +813,14 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo
 	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
 
+
+	/* TODO: Figure out how to make HDCP work on GEN12+ */
+	if (INTEL_GEN(dev_priv) < 12) {
+		ret = intel_dp_init_hdcp(dig_port, intel_connector);
+		if (ret)
+			DRM_DEBUG_KMS("HDCP init failed, skipping.\n");
+	}
+
 	/*
 	 * Reuse the prop from the SST connector because we're
 	 * not allowed to create new props after device registration.
@@ -865,6 +887,7 @@ intel_dp_create_fake_mst_encoder(struct intel_digital_port *dig_port, enum pipe
 	intel_encoder->compute_config_late = intel_dp_mst_compute_config_late;
 	intel_encoder->disable = intel_mst_disable_dp;
 	intel_encoder->post_disable = intel_mst_post_disable_dp;
+	intel_encoder->update_pipe = intel_ddi_update_pipe;
 	intel_encoder->pre_pll_enable = intel_mst_pre_pll_enable_dp;
 	intel_encoder->pre_enable = intel_mst_pre_enable_dp;
 	intel_encoder->enable = intel_mst_enable_dp;
diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index afa7a378b31d..e08684e34078 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -147,6 +147,18 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv,
 			pll->info->name, onoff(state), onoff(cur_state));
 }
 
+static i915_reg_t
+intel_combo_pll_enable_reg(struct drm_i915_private *i915,
+			   struct intel_shared_dpll *pll)
+{
+
+	if (IS_ELKHARTLAKE(i915) && (pll->info->id == DPLL_ID_EHL_DPLL4))
+		return MG_PLL_ENABLE(0);
+
+	return CNL_DPLL_ENABLE(pll->info->id);
+
+
+}
 /**
  * intel_prepare_shared_dpll - call a dpll's prepare hook
  * @crtc_state: CRTC, and its state, which has a shared dpll
@@ -3475,6 +3487,14 @@ static void icl_update_active_dpll(struct intel_atomic_state *state,
 	icl_set_active_port_dpll(crtc_state, port_dpll_id);
 }
 
+static u32 intel_get_hti_plls(struct drm_i915_private *i915)
+{
+	if (!(i915->hti_state & HDPORT_ENABLED))
+		return 0;
+
+	return REG_FIELD_GET(HDPORT_DPLL_USED_MASK, i915->hti_state);
+}
+
 static bool icl_get_combo_phy_dpll(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc,
 				   struct intel_encoder *encoder)
@@ -3504,13 +3524,22 @@ static bool icl_get_combo_phy_dpll(struct intel_atomic_state *state,
 
 	icl_calc_dpll_state(dev_priv, &pll_params, &port_dpll->hw_state);
 
-	if (IS_ELKHARTLAKE(dev_priv) && port != PORT_A)
+	if (IS_ROCKETLAKE(dev_priv)) {
 		dpll_mask =
 			BIT(DPLL_ID_EHL_DPLL4) |
 			BIT(DPLL_ID_ICL_DPLL1) |
 			BIT(DPLL_ID_ICL_DPLL0);
-	else
+	} else if (IS_ELKHARTLAKE(dev_priv) && port != PORT_A) {
+		dpll_mask =
+			BIT(DPLL_ID_EHL_DPLL4) |
+			BIT(DPLL_ID_ICL_DPLL1) |
+			BIT(DPLL_ID_ICL_DPLL0);
+	} else {
 		dpll_mask = BIT(DPLL_ID_ICL_DPLL1) | BIT(DPLL_ID_ICL_DPLL0);
+	}
+
+	/* Eliminate DPLLs from consideration if reserved by HTI */
+	dpll_mask &= ~intel_get_hti_plls(dev_priv);
 
 	port_dpll->pll = intel_find_shared_dpll(state, crtc,
 						&port_dpll->hw_state,
@@ -3791,7 +3820,12 @@ static bool icl_pll_get_hw_state(struct drm_i915_private *dev_priv,
 	if (!(val & PLL_ENABLE))
 		goto out;
 
-	if (INTEL_GEN(dev_priv) >= 12) {
+	if (IS_ROCKETLAKE(dev_priv)) {
+		hw_state->cfgcr0 = intel_de_read(dev_priv,
+						 RKL_DPLL_CFGCR0(id));
+		hw_state->cfgcr1 = intel_de_read(dev_priv,
+						 RKL_DPLL_CFGCR1(id));
+	} else if (INTEL_GEN(dev_priv) >= 12) {
 		hw_state->cfgcr0 = intel_de_read(dev_priv,
 						 TGL_DPLL_CFGCR0(id));
 		hw_state->cfgcr1 = intel_de_read(dev_priv,
@@ -3820,12 +3854,7 @@ static bool combo_pll_get_hw_state(struct drm_i915_private *dev_priv,
 				   struct intel_shared_dpll *pll,
 				   struct intel_dpll_hw_state *hw_state)
 {
-	i915_reg_t enable_reg = CNL_DPLL_ENABLE(pll->info->id);
-
-	if (IS_ELKHARTLAKE(dev_priv) &&
-	    pll->info->id == DPLL_ID_EHL_DPLL4) {
-		enable_reg = MG_PLL_ENABLE(0);
-	}
+	i915_reg_t enable_reg = intel_combo_pll_enable_reg(dev_priv, pll);
 
 	return icl_pll_get_hw_state(dev_priv, pll, hw_state, enable_reg);
 }
@@ -3844,7 +3873,10 @@ static void icl_dpll_write(struct drm_i915_private *dev_priv,
 	const enum intel_dpll_id id = pll->info->id;
 	i915_reg_t cfgcr0_reg, cfgcr1_reg;
 
-	if (INTEL_GEN(dev_priv) >= 12) {
+	if (IS_ROCKETLAKE(dev_priv)) {
+		cfgcr0_reg = RKL_DPLL_CFGCR0(id);
+		cfgcr1_reg = RKL_DPLL_CFGCR1(id);
+	} else if (INTEL_GEN(dev_priv) >= 12) {
 		cfgcr0_reg = TGL_DPLL_CFGCR0(id);
 		cfgcr1_reg = TGL_DPLL_CFGCR1(id);
 	} else {
@@ -4020,11 +4052,10 @@ static void icl_pll_enable(struct drm_i915_private *dev_priv,
 static void combo_pll_enable(struct drm_i915_private *dev_priv,
 			     struct intel_shared_dpll *pll)
 {
-	i915_reg_t enable_reg = CNL_DPLL_ENABLE(pll->info->id);
+	i915_reg_t enable_reg = intel_combo_pll_enable_reg(dev_priv, pll);
 
 	if (IS_ELKHARTLAKE(dev_priv) &&
 	    pll->info->id == DPLL_ID_EHL_DPLL4) {
-		enable_reg = MG_PLL_ENABLE(0);
 
 		/*
 		 * We need to disable DC states when this DPLL is enabled.
@@ -4132,19 +4163,14 @@ static void icl_pll_disable(struct drm_i915_private *dev_priv,
 static void combo_pll_disable(struct drm_i915_private *dev_priv,
 			      struct intel_shared_dpll *pll)
 {
-	i915_reg_t enable_reg = CNL_DPLL_ENABLE(pll->info->id);
+	i915_reg_t enable_reg = intel_combo_pll_enable_reg(dev_priv, pll);
 
-	if (IS_ELKHARTLAKE(dev_priv) &&
-	    pll->info->id == DPLL_ID_EHL_DPLL4) {
-		enable_reg = MG_PLL_ENABLE(0);
-		icl_pll_disable(dev_priv, pll, enable_reg);
+	icl_pll_disable(dev_priv, pll, enable_reg);
 
+	if (IS_ELKHARTLAKE(dev_priv) &&
+	    pll->info->id == DPLL_ID_EHL_DPLL4)
 		intel_display_power_put(dev_priv, POWER_DOMAIN_DPLL_DC_OFF,
 					pll->wakeref);
-		return;
-	}
-
-	icl_pll_disable(dev_priv, pll, enable_reg);
 }
 
 static void tbt_pll_disable(struct drm_i915_private *dev_priv,
@@ -4276,6 +4302,21 @@ static const struct intel_dpll_mgr tgl_pll_mgr = {
 	.dump_hw_state = icl_dump_hw_state,
 };
 
+static const struct dpll_info rkl_plls[] = {
+	{ "DPLL 0", &combo_pll_funcs, DPLL_ID_ICL_DPLL0, 0 },
+	{ "DPLL 1", &combo_pll_funcs, DPLL_ID_ICL_DPLL1, 0 },
+	{ "DPLL 4", &combo_pll_funcs, DPLL_ID_EHL_DPLL4, 0 },
+	{ },
+};
+
+static const struct intel_dpll_mgr rkl_pll_mgr = {
+	.dpll_info = rkl_plls,
+	.get_dplls = icl_get_dplls,
+	.put_dplls = icl_put_dplls,
+	.update_ref_clks = icl_update_dpll_ref_clks,
+	.dump_hw_state = icl_dump_hw_state,
+};
+
 /**
  * intel_shared_dpll_init - Initialize shared DPLLs
  * @dev: drm device
@@ -4289,7 +4330,9 @@ void intel_shared_dpll_init(struct drm_device *dev)
 	const struct dpll_info *dpll_info;
 	int i;
 
-	if (INTEL_GEN(dev_priv) >= 12)
+	if (IS_ROCKETLAKE(dev_priv))
+		dpll_mgr = &rkl_pll_mgr;
+	else if (INTEL_GEN(dev_priv) >= 12)
 		dpll_mgr = &tgl_pll_mgr;
 	else if (IS_ELKHARTLAKE(dev_priv))
 		dpll_mgr = &ehl_pll_mgr;
diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c
index 307ed8ae9a19..237dbb1ba0ee 100644
--- a/drivers/gpu/drm/i915/display/intel_dvo.c
+++ b/drivers/gpu/drm/i915/display/intel_dvo.c
@@ -313,9 +313,15 @@ static void intel_dvo_pre_enable(struct intel_atomic_state *state,
 static enum drm_connector_status
 intel_dvo_detect(struct drm_connector *connector, bool force)
 {
+	struct drm_i915_private *i915 = to_i915(connector->dev);
 	struct intel_dvo *intel_dvo = intel_attached_dvo(to_intel_connector(connector));
+
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
+
+	if (!INTEL_DISPLAY_ENABLED(i915))
+		return connector_status_disconnected;
+
 	return intel_dvo->dev.dev_ops->detect(&intel_dvo->dev);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 24c3a0f212c6..135f5e8a4d70 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -424,6 +424,14 @@ static void intel_fbc_deactivate(struct drm_i915_private *dev_priv,
 	fbc->no_fbc_reason = reason;
 }
 
+static u64 intel_fbc_cfb_base_max(struct drm_i915_private *i915)
+{
+	if (INTEL_GEN(i915) >= 5 || IS_G4X(i915))
+		return BIT_ULL(28);
+	else
+		return BIT_ULL(32);
+}
+
 static int find_compression_threshold(struct drm_i915_private *dev_priv,
 				      struct drm_mm_node *node,
 				      unsigned int size,
@@ -442,6 +450,8 @@ static int find_compression_threshold(struct drm_i915_private *dev_priv,
 	else
 		end = U64_MAX;
 
+	end = min(end, intel_fbc_cfb_base_max(dev_priv));
+
 	/* HACK: This code depends on what we will do in *_enable_fbc. If that
 	 * code changes, this code needs to change as well.
 	 *
@@ -1416,6 +1426,13 @@ static int intel_sanitize_fbc_option(struct drm_i915_private *dev_priv)
 	if (!HAS_FBC(dev_priv))
 		return 0;
 
+	/*
+	 * Fbc is causing random underruns in CI execution on TGL platforms.
+	 * Disabling the same while the problem is being debugged and analyzed.
+	 */
+	if (IS_TIGERLAKE(dev_priv))
+		return 0;
+
 	if (IS_BROADWELL(dev_priv) || INTEL_GEN(dev_priv) >= 9)
 		return 1;
 
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index bd39eb6a21b8..842c04e63214 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -451,8 +451,7 @@ int intel_fbdev_init(struct drm_device *dev)
 	struct intel_fbdev *ifbdev;
 	int ret;
 
-	if (drm_WARN_ON(dev, !HAS_DISPLAY(dev_priv) ||
-			!INTEL_DISPLAY_ENABLED(dev_priv)))
+	if (drm_WARN_ON(dev, !HAS_DISPLAY(dev_priv)))
 		return -ENODEV;
 
 	ifbdev = kzalloc(sizeof(struct intel_fbdev), GFP_KERNEL);
diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
index 2979ed2588eb..d898b370d7a4 100644
--- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c
+++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
@@ -232,6 +232,8 @@ static void frontbuffer_release(struct kref *ref)
 	RCU_INIT_POINTER(obj->frontbuffer, NULL);
 	spin_unlock(&to_i915(obj->base.dev)->fb_tracking.lock);
 
+	i915_active_fini(&front->write);
+
 	i915_gem_object_put(obj);
 	kfree_rcu(front, rcu);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index a8d119b6b45c..e6b8d6dfb598 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -834,7 +834,7 @@ int intel_gmbus_setup(struct drm_i915_private *dev_priv)
 	unsigned int pin;
 	int ret;
 
-	if (!HAS_DISPLAY(dev_priv) || !INTEL_DISPLAY_ENABLED(dev_priv))
+	if (!HAS_DISPLAY(dev_priv))
 		return 0;
 
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 1a0d49af2a08..5492076d1ae0 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -148,9 +148,8 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
 
 static bool hdcp_key_loadable(struct drm_i915_private *dev_priv)
 {
-	struct i915_power_domains *power_domains = &dev_priv->power_domains;
-	struct i915_power_well *power_well;
 	enum i915_power_well_id id;
+	intel_wakeref_t wakeref;
 	bool enabled = false;
 
 	/*
@@ -162,17 +161,9 @@ static bool hdcp_key_loadable(struct drm_i915_private *dev_priv)
 	else
 		id = SKL_DISP_PW_1;
 
-	mutex_lock(&power_domains->lock);
-
 	/* PG1 (power well #1) needs to be enabled */
-	for_each_power_well(dev_priv, power_well) {
-		if (power_well->desc->id == id) {
-			enabled = power_well->desc->ops->is_enabled(dev_priv,
-								    power_well);
-			break;
-		}
-	}
-	mutex_unlock(&power_domains->lock);
+	with_intel_runtime_pm(&dev_priv->runtime_pm, wakeref)
+		enabled = intel_display_power_well_is_enabled(dev_priv, id);
 
 	/*
 	 * Another req for hdcp key loadability is enabled state of pll for
@@ -713,7 +704,7 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 		intel_de_write(dev_priv, HDCP_REP_CTL,
 			       intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, port));
 
-	ret = shim->toggle_signalling(dig_port, true);
+	ret = shim->toggle_signalling(dig_port, cpu_transcoder, true);
 	if (ret)
 		return ret;
 
@@ -801,6 +792,19 @@ static int _intel_hdcp_disable(struct intel_connector *connector)
 	drm_dbg_kms(&dev_priv->drm, "[%s:%d] HDCP is being disabled...\n",
 		    connector->base.name, connector->base.base.id);
 
+	/*
+	 * If there are other connectors on this port using HDCP, don't disable
+	 * it. Instead, toggle the HDCP signalling off on that particular
+	 * connector/pipe and exit.
+	 */
+	if (dig_port->num_hdcp_streams > 0) {
+		ret = hdcp->shim->toggle_signalling(dig_port,
+						    cpu_transcoder, false);
+		if (ret)
+			DRM_ERROR("Failed to disable HDCP signalling\n");
+		return ret;
+	}
+
 	hdcp->hdcp_encrypted = false;
 	intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port), 0);
 	if (intel_de_wait_for_clear(dev_priv,
@@ -816,7 +820,7 @@ static int _intel_hdcp_disable(struct intel_connector *connector)
 	intel_de_write(dev_priv, HDCP_REP_CTL,
 		       intel_de_read(dev_priv, HDCP_REP_CTL) & ~repeater_ctl);
 
-	ret = hdcp->shim->toggle_signalling(dig_port, false);
+	ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder, false);
 	if (ret) {
 		drm_err(&dev_priv->drm, "Failed to disable HDCP signalling\n");
 		return ret;
@@ -876,6 +880,34 @@ static struct intel_connector *intel_hdcp_to_connector(struct intel_hdcp *hdcp)
 	return container_of(hdcp, struct intel_connector, hdcp);
 }
 
+static void intel_hdcp_update_value(struct intel_connector *connector,
+				    u64 value, bool update_property)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+
+	drm_WARN_ON(connector->base.dev, !mutex_is_locked(&hdcp->mutex));
+
+	if (hdcp->value == value)
+		return;
+
+	drm_WARN_ON(dev, !mutex_is_locked(&dig_port->hdcp_mutex));
+
+	if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED) {
+		if (!drm_WARN_ON(dev, dig_port->num_hdcp_streams == 0))
+			dig_port->num_hdcp_streams--;
+	} else if (value == DRM_MODE_CONTENT_PROTECTION_ENABLED) {
+		dig_port->num_hdcp_streams++;
+	}
+
+	hdcp->value = value;
+	if (update_property) {
+		drm_connector_get(&connector->base);
+		schedule_work(&hdcp->prop_work);
+	}
+}
+
 /* Implements Part 3 of the HDCP authorization procedure */
 static int intel_hdcp_check_link(struct intel_connector *connector)
 {
@@ -887,6 +919,8 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 	int ret = 0;
 
 	mutex_lock(&hdcp->mutex);
+	mutex_lock(&dig_port->hdcp_mutex);
+
 	cpu_transcoder = hdcp->cpu_transcoder;
 
 	/* Check_link valid only when HDCP1.4 is enabled */
@@ -903,15 +937,16 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 			connector->base.name, connector->base.base.id,
 			intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)));
 		ret = -ENXIO;
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_DESIRED,
+					true);
 		goto out;
 	}
 
-	if (hdcp->shim->check_link(dig_port)) {
+	if (hdcp->shim->check_link(dig_port, connector)) {
 		if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
-			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-			schedule_work(&hdcp->prop_work);
+			intel_hdcp_update_value(connector,
+				DRM_MODE_CONTENT_PROTECTION_ENABLED, true);
 		}
 		goto out;
 	}
@@ -923,20 +958,23 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 	ret = _intel_hdcp_disable(connector);
 	if (ret) {
 		drm_err(&dev_priv->drm, "Failed to disable hdcp (%d)\n", ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_DESIRED,
+					true);
 		goto out;
 	}
 
 	ret = _intel_hdcp_enable(connector);
 	if (ret) {
 		drm_err(&dev_priv->drm, "Failed to enable hdcp (%d)\n", ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_DESIRED,
+					true);
 		goto out;
 	}
 
 out:
+	mutex_unlock(&dig_port->hdcp_mutex);
 	mutex_unlock(&hdcp->mutex);
 	return ret;
 }
@@ -962,6 +1000,8 @@ static void intel_hdcp_prop_work(struct work_struct *work)
 
 	mutex_unlock(&hdcp->mutex);
 	drm_modeset_unlock(&dev_priv->drm.mode_config.connection_mutex);
+
+	drm_connector_put(&connector->base);
 }
 
 bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port)
@@ -1600,7 +1640,8 @@ static int hdcp2_enable_encryption(struct intel_connector *connector)
 		    intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
 		    LINK_ENCRYPTION_STATUS);
 	if (hdcp->shim->toggle_signalling) {
-		ret = hdcp->shim->toggle_signalling(dig_port, true);
+		ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder,
+						    true);
 		if (ret) {
 			drm_err(&dev_priv->drm,
 				"Failed to enable HDCP signalling. %d\n",
@@ -1650,7 +1691,8 @@ static int hdcp2_disable_encryption(struct intel_connector *connector)
 		drm_dbg_kms(&dev_priv->drm, "Disable Encryption Timedout");
 
 	if (hdcp->shim->toggle_signalling) {
-		ret = hdcp->shim->toggle_signalling(dig_port, false);
+		ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder,
+						    false);
 		if (ret) {
 			drm_err(&dev_priv->drm,
 				"Failed to disable HDCP signalling. %d\n",
@@ -1766,16 +1808,18 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 			"HDCP2.2 link stopped the encryption, %x\n",
 			intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)));
 		ret = -ENXIO;
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_DESIRED,
+					true);
 		goto out;
 	}
 
 	ret = hdcp->shim->check_2_2_link(dig_port);
 	if (ret == HDCP_LINK_PROTECTED) {
 		if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
-			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-			schedule_work(&hdcp->prop_work);
+			intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_ENABLED,
+					true);
 		}
 		goto out;
 	}
@@ -1788,8 +1832,9 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 			    "HDCP2.2 Downstream topology change\n");
 		ret = hdcp2_authenticate_repeater_topology(connector);
 		if (!ret) {
-			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-			schedule_work(&hdcp->prop_work);
+			intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_ENABLED,
+					true);
 			goto out;
 		}
 		drm_dbg_kms(&dev_priv->drm,
@@ -1807,8 +1852,8 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 		drm_err(&dev_priv->drm,
 			"[%s:%d] Failed to disable hdcp2.2 (%d)\n",
 			connector->base.name, connector->base.base.id, ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+				DRM_MODE_CONTENT_PROTECTION_DESIRED, true);
 		goto out;
 	}
 
@@ -1818,8 +1863,9 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 			    "[%s:%d] Failed to enable hdcp2.2 (%d)\n",
 			    connector->base.name, connector->base.base.id,
 			    ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_DESIRED,
+					true);
 		goto out;
 	}
 
@@ -1835,6 +1881,9 @@ static void intel_hdcp_check_work(struct work_struct *work)
 					       check_work);
 	struct intel_connector *connector = intel_hdcp_to_connector(hdcp);
 
+	if (drm_connector_is_unregistered(&connector->base))
+		return;
+
 	if (!intel_hdcp2_check_link(connector))
 		schedule_delayed_work(&hdcp->check_work,
 				      DRM_HDCP2_CHECK_PERIOD_MS);
@@ -1896,6 +1945,7 @@ static enum mei_fw_tc intel_get_mei_fw_tc(enum transcoder cpu_transcoder)
 }
 
 static int initialize_hdcp_port_data(struct intel_connector *connector,
+				     enum port port,
 				     const struct intel_hdcp_shim *shim)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
@@ -1903,8 +1953,7 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 	struct hdcp_port_data *data = &hdcp->port_data;
 
 	if (INTEL_GEN(dev_priv) < 12)
-		data->fw_ddi =
-			intel_get_mei_fw_ddi_index(intel_attached_encoder(connector)->port);
+		data->fw_ddi = intel_get_mei_fw_ddi_index(port);
 	else
 		/*
 		 * As per ME FW API expectation, for GEN 12+, fw_ddi is filled
@@ -1974,14 +2023,14 @@ void intel_hdcp_component_init(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void intel_hdcp2_init(struct intel_connector *connector,
+static void intel_hdcp2_init(struct intel_connector *connector, enum port port,
 			     const struct intel_hdcp_shim *shim)
 {
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	int ret;
 
-	ret = initialize_hdcp_port_data(connector, shim);
+	ret = initialize_hdcp_port_data(connector, port, shim);
 	if (ret) {
 		drm_dbg_kms(&i915->drm, "Mei hdcp data init failed\n");
 		return;
@@ -1991,6 +2040,7 @@ static void intel_hdcp2_init(struct intel_connector *connector,
 }
 
 int intel_hdcp_init(struct intel_connector *connector,
+		    enum port port,
 		    const struct intel_hdcp_shim *shim)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
@@ -2000,8 +2050,8 @@ int intel_hdcp_init(struct intel_connector *connector,
 	if (!shim)
 		return -EINVAL;
 
-	if (is_hdcp2_supported(dev_priv))
-		intel_hdcp2_init(connector, shim);
+	if (is_hdcp2_supported(dev_priv) && !connector->mst_port)
+		intel_hdcp2_init(connector, port, shim);
 
 	ret =
 	drm_connector_attach_content_protection_property(&connector->base,
@@ -2025,6 +2075,7 @@ int intel_hdcp_enable(struct intel_connector *connector,
 		      enum transcoder cpu_transcoder, u8 content_type)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS;
 	int ret = -EINVAL;
@@ -2033,14 +2084,14 @@ int intel_hdcp_enable(struct intel_connector *connector,
 		return -ENOENT;
 
 	mutex_lock(&hdcp->mutex);
+	mutex_lock(&dig_port->hdcp_mutex);
 	drm_WARN_ON(&dev_priv->drm,
 		    hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED);
 	hdcp->content_type = content_type;
+	hdcp->cpu_transcoder = cpu_transcoder;
 
-	if (INTEL_GEN(dev_priv) >= 12) {
-		hdcp->cpu_transcoder = cpu_transcoder;
+	if (INTEL_GEN(dev_priv) >= 12)
 		hdcp->port_data.fw_tc = intel_get_mei_fw_tc(cpu_transcoder);
-	}
 
 	/*
 	 * Considering that HDCP2.2 is more secure than HDCP1.4, If the setup
@@ -2063,16 +2114,19 @@ int intel_hdcp_enable(struct intel_connector *connector,
 
 	if (!ret) {
 		schedule_delayed_work(&hdcp->check_work, check_link_interval);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-		schedule_work(&hdcp->prop_work);
+		intel_hdcp_update_value(connector,
+					DRM_MODE_CONTENT_PROTECTION_ENABLED,
+					true);
 	}
 
+	mutex_unlock(&dig_port->hdcp_mutex);
 	mutex_unlock(&hdcp->mutex);
 	return ret;
 }
 
 int intel_hdcp_disable(struct intel_connector *connector)
 {
+	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	int ret = 0;
 
@@ -2080,15 +2134,20 @@ int intel_hdcp_disable(struct intel_connector *connector)
 		return -ENOENT;
 
 	mutex_lock(&hdcp->mutex);
+	mutex_lock(&dig_port->hdcp_mutex);
 
-	if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_UNDESIRED;
-		if (hdcp->hdcp2_encrypted)
-			ret = _intel_hdcp2_disable(connector);
-		else if (hdcp->hdcp_encrypted)
-			ret = _intel_hdcp_disable(connector);
-	}
+	if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_UNDESIRED)
+		goto out;
+
+	intel_hdcp_update_value(connector,
+				DRM_MODE_CONTENT_PROTECTION_UNDESIRED, false);
+	if (hdcp->hdcp2_encrypted)
+		ret = _intel_hdcp2_disable(connector);
+	else if (hdcp->hdcp_encrypted)
+		ret = _intel_hdcp_disable(connector);
 
+out:
+	mutex_unlock(&dig_port->hdcp_mutex);
 	mutex_unlock(&hdcp->mutex);
 	cancel_delayed_work_sync(&hdcp->check_work);
 	return ret;
@@ -2102,11 +2161,15 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 	struct intel_connector *connector =
 				to_intel_connector(conn_state->connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
-	bool content_protection_type_changed =
+	bool content_protection_type_changed, desired_and_not_enabled = false;
+
+	if (!connector->hdcp.shim)
+		return;
+
+	content_protection_type_changed =
 		(conn_state->hdcp_content_type != hdcp->content_type &&
 		 conn_state->content_protection !=
 		 DRM_MODE_CONTENT_PROTECTION_UNDESIRED);
-	bool desired_and_not_enabled = false;
 
 	/*
 	 * During the HDCP encryption session if Type change is requested,
@@ -2159,12 +2222,39 @@ void intel_hdcp_component_fini(struct drm_i915_private *dev_priv)
 
 void intel_hdcp_cleanup(struct intel_connector *connector)
 {
-	if (!connector->hdcp.shim)
+	struct intel_hdcp *hdcp = &connector->hdcp;
+
+	if (!hdcp->shim)
 		return;
 
-	mutex_lock(&connector->hdcp.mutex);
-	kfree(connector->hdcp.port_data.streams);
-	mutex_unlock(&connector->hdcp.mutex);
+	/*
+	 * If the connector is registered, it's possible userspace could kick
+	 * off another HDCP enable, which would re-spawn the workers.
+	 */
+	drm_WARN_ON(connector->base.dev,
+		connector->base.registration_state == DRM_CONNECTOR_REGISTERED);
+
+	/*
+	 * Now that the connector is not registered, check_work won't be run,
+	 * but cancel any outstanding instances of it
+	 */
+	cancel_delayed_work_sync(&hdcp->check_work);
+
+	/*
+	 * We don't cancel prop_work in the same way as check_work since it
+	 * requires connection_mutex which could be held while calling this
+	 * function. Instead, we rely on the connector references grabbed before
+	 * scheduling prop_work to ensure the connector is alive when prop_work
+	 * is run. So if we're in the destroy path (which is where this
+	 * function should be called), we're "guaranteed" that prop_work is not
+	 * active (tl;dr This Should Never Happen).
+	 */
+	drm_WARN_ON(connector->base.dev, work_pending(&hdcp->prop_work));
+
+	mutex_lock(&hdcp->mutex);
+	kfree(hdcp->port_data.streams);
+	hdcp->shim = NULL;
+	mutex_unlock(&hdcp->mutex);
 }
 
 void intel_hdcp_atomic_check(struct drm_connector *connector,
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h
index 86bbaec120cc..1bbf5b67ed0a 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.h
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.h
@@ -22,7 +22,7 @@ enum transcoder;
 void intel_hdcp_atomic_check(struct drm_connector *connector,
 			     struct drm_connector_state *old_state,
 			     struct drm_connector_state *new_state);
-int intel_hdcp_init(struct intel_connector *connector,
+int intel_hdcp_init(struct intel_connector *connector, enum port port,
 		    const struct intel_hdcp_shim *hdcp_shim);
 int intel_hdcp_enable(struct intel_connector *connector,
 		      enum transcoder cpu_transcoder, u8 content_type);
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index de2ce5632b94..3f2008d845c2 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -1477,7 +1477,8 @@ int intel_hdmi_hdcp_read_v_prime_part(struct intel_digital_port *dig_port,
 	return ret;
 }
 
-static int kbl_repositioning_enc_en_signal(struct intel_connector *connector)
+static int kbl_repositioning_enc_en_signal(struct intel_connector *connector,
+					   enum transcoder cpu_transcoder)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
@@ -1494,13 +1495,15 @@ static int kbl_repositioning_enc_en_signal(struct intel_connector *connector)
 		usleep_range(25, 50);
 	}
 
-	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, false);
+	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, cpu_transcoder,
+					       false);
 	if (ret) {
 		drm_err(&dev_priv->drm,
 			"Disable HDCP signalling failed (%d)\n", ret);
 		return ret;
 	}
-	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, true);
+	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, cpu_transcoder,
+					       true);
 	if (ret) {
 		drm_err(&dev_priv->drm,
 			"Enable HDCP signalling failed (%d)\n", ret);
@@ -1512,6 +1515,7 @@ static int kbl_repositioning_enc_en_signal(struct intel_connector *connector)
 
 static
 int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
+				      enum transcoder cpu_transcoder,
 				      bool enable)
 {
 	struct intel_hdmi *hdmi = &dig_port->hdmi;
@@ -1522,7 +1526,8 @@ int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
 	if (!enable)
 		usleep_range(6, 60); /* Bspec says >= 6us */
 
-	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, enable);
+	ret = intel_ddi_toggle_hdcp_signalling(&dig_port->base, cpu_transcoder,
+					       enable);
 	if (ret) {
 		drm_err(&dev_priv->drm, "%s HDCP signalling failed (%d)\n",
 			enable ? "Enable" : "Disable", ret);
@@ -1534,17 +1539,17 @@ int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
 	 * opportunity and enc_en signalling in KABYLAKE.
 	 */
 	if (IS_KABYLAKE(dev_priv) && enable)
-		return kbl_repositioning_enc_en_signal(connector);
+		return kbl_repositioning_enc_en_signal(connector,
+						       cpu_transcoder);
 
 	return 0;
 }
 
 static
-bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port)
+bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port,
+				     struct intel_connector *connector)
 {
 	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
-	struct intel_connector *connector =
-		dig_port->hdmi.attached_connector;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder;
 	int ret;
@@ -1572,13 +1577,14 @@ bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port)
 }
 
 static
-bool intel_hdmi_hdcp_check_link(struct intel_digital_port *dig_port)
+bool intel_hdmi_hdcp_check_link(struct intel_digital_port *dig_port,
+				struct intel_connector *connector)
 {
 	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	int retry;
 
 	for (retry = 0; retry < 3; retry++)
-		if (intel_hdmi_hdcp_check_link_once(dig_port))
+		if (intel_hdmi_hdcp_check_link_once(dig_port, connector))
 			return true;
 
 	drm_err(&i915->drm, "Link check failed\n");
@@ -2271,35 +2277,18 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 	return intel_mode_valid_max_plane_size(dev_priv, mode);
 }
 
-static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
-				     int bpc)
+bool intel_hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
+				    int bpc, bool has_hdmi_sink, bool ycbcr420_output)
 {
-	struct drm_i915_private *dev_priv =
-		to_i915(crtc_state->uapi.crtc->dev);
 	struct drm_atomic_state *state = crtc_state->uapi.state;
 	struct drm_connector_state *connector_state;
 	struct drm_connector *connector;
-	const struct drm_display_mode *adjusted_mode =
-		&crtc_state->hw.adjusted_mode;
 	int i;
 
-	if (HAS_GMCH(dev_priv))
-		return false;
-
-	if (bpc == 10 && INTEL_GEN(dev_priv) < 11)
-		return false;
-
 	if (crtc_state->pipe_bpp < bpc * 3)
 		return false;
 
-	if (!crtc_state->has_hdmi_sink)
-		return false;
-
-	/*
-	 * HDMI deep color affects the clocks, so it's only possible
-	 * when not cloning with other encoder types.
-	 */
-	if (crtc_state->output_types != 1 << INTEL_OUTPUT_HDMI)
+	if (!has_hdmi_sink)
 		return false;
 
 	for_each_new_connector_in_state(state, connector, connector_state, i) {
@@ -2308,7 +2297,7 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
 		if (connector_state->crtc != crtc_state->uapi.crtc)
 			continue;
 
-		if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420) {
+		if (ycbcr420_output) {
 			const struct drm_hdmi_info *hdmi = &info->hdmi;
 
 			if (bpc == 12 && !(hdmi->y420_dc_modes &
@@ -2327,6 +2316,30 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
 		}
 	}
 
+	return true;
+}
+
+static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
+				     int bpc)
+{
+	struct drm_i915_private *dev_priv =
+		to_i915(crtc_state->uapi.crtc->dev);
+	const struct drm_display_mode *adjusted_mode =
+		&crtc_state->hw.adjusted_mode;
+
+	if (HAS_GMCH(dev_priv))
+		return false;
+
+	if (bpc == 10 && INTEL_GEN(dev_priv) < 11)
+		return false;
+
+	/*
+	 * HDMI deep color affects the clocks, so it's only possible
+	 * when not cloning with other encoder types.
+	 */
+	if (crtc_state->output_types != BIT(INTEL_OUTPUT_HDMI))
+		return false;
+
 	/* Display Wa_1405510057:icl,ehl */
 	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 &&
 	    bpc == 10 && IS_GEN(dev_priv, 11) &&
@@ -2334,7 +2347,10 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
 	     adjusted_mode->crtc_hblank_start) % 8 == 2)
 		return false;
 
-	return true;
+	return intel_hdmi_deep_color_possible(crtc_state, bpc,
+					      crtc_state->has_hdmi_sink,
+					      crtc_state->output_format ==
+					      INTEL_OUTPUT_FORMAT_YCBCR420);
 }
 
 static int
@@ -2459,6 +2475,23 @@ bool intel_hdmi_limited_color_range(const struct intel_crtc_state *crtc_state,
 	}
 }
 
+static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
+				 const struct intel_crtc_state *crtc_state,
+				 const struct drm_connector_state *conn_state)
+{
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+	const struct intel_digital_connector_state *intel_conn_state =
+		to_intel_digital_connector_state(conn_state);
+
+	if (!crtc_state->has_hdmi_sink)
+		return false;
+
+	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
+		return intel_hdmi->has_audio;
+	else
+		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
+}
+
 int intel_hdmi_compute_config(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config,
 			      struct drm_connector_state *conn_state)
@@ -2468,8 +2501,6 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 	struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
 	struct drm_connector *connector = conn_state->connector;
 	struct drm_scdc *scdc = &connector->display_info.hdmi.scdc;
-	struct intel_digital_connector_state *intel_conn_state =
-		to_intel_digital_connector_state(conn_state);
 	int ret;
 
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
@@ -2495,13 +2526,8 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 	if (HAS_PCH_SPLIT(dev_priv) && !HAS_DDI(dev_priv))
 		pipe_config->has_pch_encoder = true;
 
-	if (pipe_config->has_hdmi_sink) {
-		if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-			pipe_config->has_audio = intel_hdmi->has_audio;
-		else
-			pipe_config->has_audio =
-				intel_conn_state->force_audio == HDMI_AUDIO_ON;
-	}
+	pipe_config->has_audio =
+		intel_hdmi_has_audio(encoder, pipe_config, conn_state);
 
 	ret = intel_hdmi_compute_clock(encoder, pipe_config);
 	if (ret)
@@ -2667,6 +2693,9 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s]\n",
 		    connector->base.id, connector->name);
 
+	if (!INTEL_DISPLAY_ENABLED(dev_priv))
+		return connector_status_disconnected;
+
 	wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
 	if (INTEL_GEN(dev_priv) >= 11 &&
@@ -3250,7 +3279,6 @@ void intel_hdmi_init_connector(struct intel_digital_port *dig_port,
 	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
 		connector->ycbcr_420_allowed = true;
 
-	intel_encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port);
 	intel_connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	if (HAS_DDI(dev_priv))
@@ -3264,7 +3292,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *dig_port,
 	intel_hdmi->attached_connector = intel_connector;
 
 	if (is_hdcp_supported(dev_priv, port)) {
-		int ret = intel_hdcp_init(intel_connector,
+		int ret = intel_hdcp_init(intel_connector, port,
 					  &intel_hdmi_hdcp_shim);
 		if (ret)
 			drm_dbg_kms(&dev_priv->drm,
@@ -3335,6 +3363,8 @@ void intel_hdmi_init(struct drm_i915_private *dev_priv,
 
 	intel_encoder = &dig_port->base;
 
+	mutex_init(&dig_port->hdcp_mutex);
+
 	drm_encoder_init(&dev_priv->drm, &intel_encoder->base,
 			 &intel_hdmi_enc_funcs, DRM_MODE_ENCODER_TMDS,
 			 "HDMI %c", port_name(port));
@@ -3382,6 +3412,7 @@ void intel_hdmi_init(struct drm_i915_private *dev_priv,
 		intel_encoder->pipe_mask = ~0;
 	}
 	intel_encoder->cloneable = 1 << INTEL_OUTPUT_ANALOG;
+	intel_encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port);
 	/*
 	 * BSpec is unclear about HDMI+HDMI cloning on g4x, but it seems
 	 * to work on real hardware. And since g4x can send infoframes to
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h
index 5b348dcab77a..15eb0ccde76e 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.h
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.h
@@ -48,5 +48,7 @@ void intel_read_infoframe(struct intel_encoder *encoder,
 			  union hdmi_infoframe *frame);
 bool intel_hdmi_limited_color_range(const struct intel_crtc_state *crtc_state,
 				    const struct drm_connector_state *conn_state);
+bool intel_hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, int bpc,
+				    bool has_hdmi_sink, bool ycbcr420_output);
 
 #endif /* __INTEL_HDMI_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index 3f1d7b804a66..5c58c1ed6493 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -81,33 +81,12 @@
  *
  * It is only valid and used by digital port encoder.
  *
- * Return pin that is associatade with @port and HDP_NONE if no pin is
- * hard associated with that @port.
+ * Return pin that is associatade with @port.
  */
 enum hpd_pin intel_hpd_pin_default(struct drm_i915_private *dev_priv,
 				   enum port port)
 {
-	enum phy phy = intel_port_to_phy(dev_priv, port);
-
-	/*
-	 * RKL + TGP PCH is a special case; we effectively choose the hpd_pin
-	 * based on the DDI rather than the PHY (i.e., the last two outputs
-	 * shold be HPD_PORT_{D,E} rather than {C,D}.  Note that this differs
-	 * from the behavior of both TGL+TGP and RKL+CMP.
-	 */
-	if (IS_ROCKETLAKE(dev_priv) && HAS_PCH_TGP(dev_priv))
-		return HPD_PORT_A + port - PORT_A;
-
-	switch (phy) {
-	case PHY_F:
-		return IS_CNL_WITH_PORT_F(dev_priv) ? HPD_PORT_E : HPD_PORT_F;
-	case PHY_A ... PHY_E:
-	case PHY_G ... PHY_I:
-		return HPD_PORT_A + phy - PHY_A;
-	default:
-		MISSING_CASE(phy);
-		return HPD_NONE;
-	}
+	return HPD_PORT_A + port - PORT_A;
 }
 
 #define HPD_STORM_DETECT_PERIOD		1000
@@ -503,7 +482,6 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	 * only the one of them (DP) will have ->hpd_pulse().
 	 */
 	for_each_intel_encoder(&dev_priv->drm, encoder) {
-		bool has_hpd_pulse = intel_encoder_has_hpd_pulse(encoder);
 		enum port port = encoder->port;
 		bool long_hpd;
 
@@ -511,7 +489,7 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 		if (!(BIT(pin) & pin_mask))
 			continue;
 
-		if (!has_hpd_pulse)
+		if (!intel_encoder_has_hpd_pulse(encoder))
 			continue;
 
 		long_hpd = long_mask & BIT(pin);
diff --git a/drivers/gpu/drm/i915/display/intel_lspcon.c b/drivers/gpu/drm/i915/display/intel_lspcon.c
index b781bf469644..dc1b35559afd 100644
--- a/drivers/gpu/drm/i915/display/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/display/intel_lspcon.c
@@ -571,7 +571,7 @@ bool lspcon_init(struct intel_digital_port *dig_port)
 		return false;
 	}
 
-	if (!intel_dp_read_dpcd(dp)) {
+	if (drm_dp_read_dpcd_caps(&dp->aux, dp->dpcd) != 0) {
 		DRM_ERROR("LSPCON DPCD read failed\n");
 		return false;
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 1888611244db..e65c2de522c3 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -456,12 +456,6 @@ static int intel_lvds_compute_config(struct intel_encoder *intel_encoder,
 	return 0;
 }
 
-static enum drm_connector_status
-intel_lvds_detect(struct drm_connector *connector, bool force)
-{
-	return connector_status_connected;
-}
-
 /*
  * Return the list of DDC modes if available, or the BIOS fixed mode otherwise.
  */
@@ -490,7 +484,7 @@ static const struct drm_connector_helper_funcs intel_lvds_connector_helper_funcs
 };
 
 static const struct drm_connector_funcs intel_lvds_connector_funcs = {
-	.detect = intel_lvds_detect,
+	.detect = intel_panel_detect,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.atomic_get_property = intel_digital_connector_atomic_get_property,
 	.atomic_set_property = intel_digital_connector_atomic_set_property,
diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c
index 4072d7062efd..9f23bac0d792 100644
--- a/drivers/gpu/drm/i915/display/intel_panel.c
+++ b/drivers/gpu/drm/i915/display/intel_panel.c
@@ -40,8 +40,6 @@
 #include "intel_dsi_dcs_backlight.h"
 #include "intel_panel.h"
 
-#define CRC_PMIC_PWM_PERIOD_NS	21333
-
 void
 intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
 		       struct drm_display_mode *adjusted_mode)
@@ -594,10 +592,10 @@ static u32 bxt_get_backlight(struct intel_connector *connector)
 static u32 pwm_get_backlight(struct intel_connector *connector)
 {
 	struct intel_panel *panel = &connector->panel;
-	int duty_ns;
+	struct pwm_state state;
 
-	duty_ns = pwm_get_duty_cycle(panel->backlight.pwm);
-	return DIV_ROUND_UP(duty_ns * 100, CRC_PMIC_PWM_PERIOD_NS);
+	pwm_get_state(panel->backlight.pwm, &state);
+	return pwm_get_relative_duty_cycle(&state, 100);
 }
 
 static void lpt_set_backlight(const struct drm_connector_state *conn_state, u32 level)
@@ -671,9 +669,9 @@ static void bxt_set_backlight(const struct drm_connector_state *conn_state, u32
 static void pwm_set_backlight(const struct drm_connector_state *conn_state, u32 level)
 {
 	struct intel_panel *panel = &to_intel_connector(conn_state->connector)->panel;
-	int duty_ns = DIV_ROUND_UP(level * CRC_PMIC_PWM_PERIOD_NS, 100);
 
-	pwm_config(panel->backlight.pwm, duty_ns, CRC_PMIC_PWM_PERIOD_NS);
+	pwm_set_relative_duty_cycle(&panel->backlight.pwm_state, level, 100);
+	pwm_apply_state(panel->backlight.pwm, &panel->backlight.pwm_state);
 }
 
 static void
@@ -842,10 +840,8 @@ static void pwm_disable_backlight(const struct drm_connector_state *old_conn_sta
 	struct intel_connector *connector = to_intel_connector(old_conn_state->connector);
 	struct intel_panel *panel = &connector->panel;
 
-	/* Disable the backlight */
-	intel_panel_actually_set_backlight(old_conn_state, 0);
-	usleep_range(2000, 3000);
-	pwm_disable(panel->backlight.pwm);
+	panel->backlight.pwm_state.enabled = false;
+	pwm_apply_state(panel->backlight.pwm, &panel->backlight.pwm_state);
 }
 
 void intel_panel_disable_backlight(const struct drm_connector_state *old_conn_state)
@@ -1177,9 +1173,12 @@ static void pwm_enable_backlight(const struct intel_crtc_state *crtc_state,
 {
 	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	struct intel_panel *panel = &connector->panel;
+	int level = panel->backlight.level;
 
-	pwm_enable(panel->backlight.pwm);
-	intel_panel_actually_set_backlight(conn_state, panel->backlight.level);
+	level = intel_panel_compute_brightness(connector, level);
+	pwm_set_relative_duty_cycle(&panel->backlight.pwm_state, level, 100);
+	panel->backlight.pwm_state.enabled = true;
+	pwm_apply_state(panel->backlight.pwm, &panel->backlight.pwm_state);
 }
 
 static void __intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,
@@ -1543,18 +1542,9 @@ static u32 vlv_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 	return DIV_ROUND_CLOSEST(clock, pwm_freq_hz * mul);
 }
 
-static u32 get_backlight_max_vbt(struct intel_connector *connector)
+static u16 get_vbt_pwm_freq(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
-	struct intel_panel *panel = &connector->panel;
 	u16 pwm_freq_hz = dev_priv->vbt.backlight.pwm_freq_hz;
-	u32 pwm;
-
-	if (!panel->backlight.hz_to_pwm) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "backlight frequency conversion not supported\n");
-		return 0;
-	}
 
 	if (pwm_freq_hz) {
 		drm_dbg_kms(&dev_priv->drm,
@@ -1567,6 +1557,22 @@ static u32 get_backlight_max_vbt(struct intel_connector *connector)
 			    pwm_freq_hz);
 	}
 
+	return pwm_freq_hz;
+}
+
+static u32 get_backlight_max_vbt(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_panel *panel = &connector->panel;
+	u16 pwm_freq_hz = get_vbt_pwm_freq(dev_priv);
+	u32 pwm;
+
+	if (!panel->backlight.hz_to_pwm) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "backlight frequency conversion not supported\n");
+		return 0;
+	}
+
 	pwm = panel->backlight.hz_to_pwm(connector, pwm_freq_hz);
 	if (!pwm) {
 		drm_dbg_kms(&dev_priv->drm,
@@ -1891,8 +1897,7 @@ static int pwm_setup_backlight(struct intel_connector *connector,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_panel *panel = &connector->panel;
 	const char *desc;
-	u32 level, ns;
-	int retval;
+	u32 level;
 
 	/* Get the right PWM chip for DSI backlight according to VBT */
 	if (dev_priv->vbt.dsi.config->pwm_blc == PPS_BLC_PMIC) {
@@ -1910,30 +1915,28 @@ static int pwm_setup_backlight(struct intel_connector *connector,
 		return -ENODEV;
 	}
 
-	/*
-	 * FIXME: pwm_apply_args() should be removed when switching to
-	 * the atomic PWM API.
-	 */
-	pwm_apply_args(panel->backlight.pwm);
-
-	panel->backlight.min = 0; /* 0% */
 	panel->backlight.max = 100; /* 100% */
-	level = intel_panel_compute_brightness(connector, 100);
-	ns = DIV_ROUND_UP(level * CRC_PMIC_PWM_PERIOD_NS, 100);
+	panel->backlight.min = get_backlight_min_vbt(connector);
 
-	retval = pwm_config(panel->backlight.pwm, ns, CRC_PMIC_PWM_PERIOD_NS);
-	if (retval < 0) {
-		drm_err(&dev_priv->drm, "Failed to configure the pwm chip\n");
-		pwm_put(panel->backlight.pwm);
-		panel->backlight.pwm = NULL;
-		return retval;
-	}
+	if (pwm_is_enabled(panel->backlight.pwm)) {
+		/* PWM is already enabled, use existing settings */
+		pwm_get_state(panel->backlight.pwm, &panel->backlight.pwm_state);
+
+		level = pwm_get_relative_duty_cycle(&panel->backlight.pwm_state,
+						    100);
+		level = intel_panel_compute_brightness(connector, level);
+		panel->backlight.level = clamp(level, panel->backlight.min,
+					       panel->backlight.max);
+		panel->backlight.enabled = true;
 
-	level = DIV_ROUND_UP_ULL(pwm_get_duty_cycle(panel->backlight.pwm) * 100,
-			     CRC_PMIC_PWM_PERIOD_NS);
-	panel->backlight.level =
-		intel_panel_compute_brightness(connector, level);
-	panel->backlight.enabled = panel->backlight.level != 0;
+		drm_dbg_kms(&dev_priv->drm, "PWM already enabled at freq %ld, VBT freq %d, level %d\n",
+			    NSEC_PER_SEC / (unsigned long)panel->backlight.pwm_state.period,
+			    get_vbt_pwm_freq(dev_priv), level);
+	} else {
+		/* Set period from VBT frequency, leave other settings at 0. */
+		panel->backlight.pwm_state.period =
+			NSEC_PER_SEC / get_vbt_pwm_freq(dev_priv);
+	}
 
 	drm_info(&dev_priv->drm, "Using %s PWM for LCD backlight control\n",
 		 desc);
@@ -2092,6 +2095,17 @@ intel_panel_init_backlight_funcs(struct intel_panel *panel)
 	}
 }
 
+enum drm_connector_status
+intel_panel_detect(struct drm_connector *connector, bool force)
+{
+	struct drm_i915_private *i915 = to_i915(connector->dev);
+
+	if (!INTEL_DISPLAY_ENABLED(i915))
+		return connector_status_disconnected;
+
+	return connector_status_connected;
+}
+
 int intel_panel_init(struct intel_panel *panel,
 		     struct drm_display_mode *fixed_mode,
 		     struct drm_display_mode *downclock_mode)
diff --git a/drivers/gpu/drm/i915/display/intel_panel.h b/drivers/gpu/drm/i915/display/intel_panel.h
index 968b95281cb4..5b813fe90557 100644
--- a/drivers/gpu/drm/i915/display/intel_panel.h
+++ b/drivers/gpu/drm/i915/display/intel_panel.h
@@ -23,6 +23,8 @@ int intel_panel_init(struct intel_panel *panel,
 		     struct drm_display_mode *fixed_mode,
 		     struct drm_display_mode *downclock_mode);
 void intel_panel_fini(struct intel_panel *panel);
+enum drm_connector_status
+intel_panel_detect(struct drm_connector *connector, bool force);
 void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
 			    struct drm_display_mode *adjusted_mode);
 int intel_pch_panel_fitting(struct intel_crtc_state *crtc_state,
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index bf9e320c547d..8a9d0bdde1bf 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -553,6 +553,22 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 		val |= EDP_PSR2_FAST_WAKE(7);
 	}
 
+	if (dev_priv->psr.psr2_sel_fetch_enabled) {
+		/* WA 1408330847 */
+		if (IS_TGL_DISP_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_A0) ||
+		    IS_RKL_REVID(dev_priv, RKL_REVID_A0, RKL_REVID_A0))
+			intel_de_rmw(dev_priv, CHICKEN_PAR1_1,
+				     DIS_RAM_BYPASS_PSR2_MAN_TRACK,
+				     DIS_RAM_BYPASS_PSR2_MAN_TRACK);
+
+		intel_de_write(dev_priv,
+			       PSR2_MAN_TRK_CTL(dev_priv->psr.transcoder),
+			       PSR2_MAN_TRK_CTL_ENABLE);
+	} else if (HAS_PSR2_SEL_FETCH(dev_priv)) {
+		intel_de_write(dev_priv,
+			       PSR2_MAN_TRK_CTL(dev_priv->psr.transcoder), 0);
+	}
+
 	/*
 	 * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is
 	 * recommending keep this bit unset while PSR2 is enabled.
@@ -663,6 +679,38 @@ tgl_dc3co_exitline_compute_config(struct intel_dp *intel_dp,
 	crtc_state->dc3co_exitline = crtc_vdisplay - exit_scanlines;
 }
 
+static bool intel_psr2_sel_fetch_config_valid(struct intel_dp *intel_dp,
+					      struct intel_crtc_state *crtc_state)
+{
+	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	struct intel_plane_state *plane_state;
+	struct intel_plane *plane;
+	int i;
+
+	if (!dev_priv->params.enable_psr2_sel_fetch) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "PSR2 sel fetch not enabled, disabled by parameter\n");
+		return false;
+	}
+
+	if (crtc_state->uapi.async_flip) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "PSR2 sel fetch not enabled, async flip enabled\n");
+		return false;
+	}
+
+	for_each_new_intel_plane_in_state(state, plane, plane_state, i) {
+		if (plane_state->uapi.rotation != DRM_MODE_ROTATE_0) {
+			drm_dbg_kms(&dev_priv->drm,
+				    "PSR2 sel fetch not enabled, plane rotated\n");
+			return false;
+		}
+	}
+
+	return crtc_state->enable_psr2_sel_fetch = true;
+}
+
 static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 				    struct intel_crtc_state *crtc_state)
 {
@@ -732,22 +780,17 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 		return false;
 	}
 
-	/*
-	 * Some platforms lack PSR2 HW tracking and instead require manual
-	 * tracking by software.  In this case, the driver is required to track
-	 * the areas that need updates and program hardware to send selective
-	 * updates.
-	 *
-	 * So until the software tracking is implemented, PSR2 needs to be
-	 * disabled for platforms without PSR2 HW tracking.
-	 */
-	if (!HAS_PSR_HW_TRACKING(dev_priv)) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "No PSR2 HW tracking in the platform\n");
-		return false;
+	if (HAS_PSR2_SEL_FETCH(dev_priv)) {
+		if (!intel_psr2_sel_fetch_config_valid(intel_dp, crtc_state) &&
+		    !HAS_PSR_HW_TRACKING(dev_priv)) {
+			drm_dbg_kms(&dev_priv->drm,
+				    "PSR2 not enabled, selective fetch not valid and no HW tracking available\n");
+			return false;
+		}
 	}
 
-	if (crtc_hdisplay > psr_max_h || crtc_vdisplay > psr_max_v) {
+	if (!crtc_state->enable_psr2_sel_fetch &&
+	    (crtc_hdisplay > psr_max_h || crtc_vdisplay > psr_max_v)) {
 		drm_dbg_kms(&dev_priv->drm,
 			    "PSR2 not enabled, resolution %dx%d > max supported %dx%d\n",
 			    crtc_hdisplay, crtc_vdisplay,
@@ -898,6 +941,11 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 		val |= EXITLINE_ENABLE;
 		intel_de_write(dev_priv, EXITLINE(cpu_transcoder), val);
 	}
+
+	if (HAS_PSR_HW_TRACKING(dev_priv))
+		intel_de_rmw(dev_priv, CHICKEN_PAR1_1, IGNORE_PSR2_HW_TRACKING,
+			     dev_priv->psr.psr2_sel_fetch_enabled ?
+			     IGNORE_PSR2_HW_TRACKING : 0);
 }
 
 static void intel_psr_enable_locked(struct drm_i915_private *dev_priv,
@@ -919,6 +967,7 @@ static void intel_psr_enable_locked(struct drm_i915_private *dev_priv,
 	/* DC5/DC6 requires at least 6 idle frames */
 	val = usecs_to_jiffies(intel_get_frame_time_us(crtc_state) * 6);
 	dev_priv->psr.dc3co_exit_delay = val;
+	dev_priv->psr.psr2_sel_fetch_enabled = crtc_state->enable_psr2_sel_fetch;
 
 	/*
 	 * If a PSR error happened and the driver is reloaded, the EDP_PSR_IIR
@@ -1058,6 +1107,13 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 				    psr_status_mask, 2000))
 		drm_err(&dev_priv->drm, "Timed out waiting PSR idle state\n");
 
+	/* WA 1408330847 */
+	if (dev_priv->psr.psr2_sel_fetch_enabled &&
+	    (IS_TGL_DISP_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_A0) ||
+	     IS_RKL_REVID(dev_priv, RKL_REVID_A0, RKL_REVID_A0)))
+		intel_de_rmw(dev_priv, CHICKEN_PAR1_1,
+			     DIS_RAM_BYPASS_PSR2_MAN_TRACK, 0);
+
 	/* Disable PSR on Sink */
 	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, 0);
 
@@ -1115,6 +1171,32 @@ static void psr_force_hw_tracking_exit(struct drm_i915_private *dev_priv)
 		intel_psr_exit(dev_priv);
 }
 
+void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct i915_psr *psr = &dev_priv->psr;
+
+	if (!HAS_PSR2_SEL_FETCH(dev_priv) ||
+	    !crtc_state->enable_psr2_sel_fetch)
+		return;
+
+	intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(psr->transcoder),
+		       crtc_state->psr2_man_track_ctl);
+}
+
+void intel_psr2_sel_fetch_update(struct intel_atomic_state *state,
+				 struct intel_crtc *crtc)
+{
+	struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc);
+
+	if (!crtc_state->enable_psr2_sel_fetch)
+		return;
+
+	crtc_state->psr2_man_track_ctl = PSR2_MAN_TRK_CTL_ENABLE |
+					 PSR2_MAN_TRK_CTL_SF_SINGLE_FULL_FRAME;
+}
+
 /**
  * intel_psr_update - Update PSR state
  * @intel_dp: Intel DP
diff --git a/drivers/gpu/drm/i915/display/intel_psr.h b/drivers/gpu/drm/i915/display/intel_psr.h
index b4515186d5f4..6a83c8e682e6 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.h
+++ b/drivers/gpu/drm/i915/display/intel_psr.h
@@ -13,6 +13,8 @@ struct drm_connector_state;
 struct drm_i915_private;
 struct intel_crtc_state;
 struct intel_dp;
+struct intel_crtc;
+struct intel_atomic_state;
 
 #define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support)
 void intel_psr_init_dpcd(struct intel_dp *intel_dp);
@@ -43,5 +45,8 @@ void intel_psr_atomic_check(struct drm_connector *connector,
 			    struct drm_connector_state *old_state,
 			    struct drm_connector_state *new_state);
 void intel_psr_set_force_mode_changed(struct intel_dp *intel_dp);
+void intel_psr2_sel_fetch_update(struct intel_atomic_state *state,
+				 struct intel_crtc *crtc);
+void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_state);
 
 #endif /* __INTEL_PSR_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 5e9fb349c829..4eaa4aa86ecd 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -2084,14 +2084,18 @@ intel_sdvo_connector_matches_edid(struct intel_sdvo_connector *sdvo,
 static enum drm_connector_status
 intel_sdvo_detect(struct drm_connector *connector, bool force)
 {
-	u16 response;
+	struct drm_i915_private *i915 = to_i915(connector->dev);
 	struct intel_sdvo *intel_sdvo = intel_attached_sdvo(to_intel_connector(connector));
 	struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector);
 	enum drm_connector_status ret;
+	u16 response;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
+	if (!INTEL_DISPLAY_ENABLED(i915))
+		return connector_status_disconnected;
+
 	if (!intel_sdvo_get_value(intel_sdvo,
 				  SDVO_CMD_GET_ATTACHED_DISPLAYS,
 				  &response, 2))
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c
index c89f5f7ccb06..63040cb0d4e1 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite.c
@@ -1626,8 +1626,7 @@ static int g4x_sprite_min_cdclk(const struct intel_crtc_state *crtc_state,
 	hscale = drm_rect_calc_hscale(&plane_state->uapi.src,
 				      &plane_state->uapi.dst,
 				      0, INT_MAX);
-	if (hscale < 0x10000)
-		return pixel_rate;
+	hscale = max(hscale, 0x10000u);
 
 	/* Decimation steps at 2x,4x,8x,16x */
 	decimate = ilog2(hscale >> 16);
@@ -1640,8 +1639,8 @@ static int g4x_sprite_min_cdclk(const struct intel_crtc_state *crtc_state,
 	limit -= decimate;
 
 	/* -10% for RGB */
-	if (fb->format->cpp[0] >= 4)
-		limit--; /* -10% for RGB */
+	if (!fb->format->is_yuv)
+		limit--;
 
 	/*
 	 * We should also do -10% if sprite scaling is enabled
@@ -2843,8 +2842,9 @@ static bool skl_plane_format_mod_supported(struct drm_plane *_plane,
 static bool gen12_plane_supports_mc_ccs(struct drm_i915_private *dev_priv,
 					enum plane_id plane_id)
 {
-	/* Wa_14010477008:tgl[a0..c0] */
-	if (IS_TGL_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_C0))
+	/* Wa_14010477008:tgl[a0..c0],rkl[all] */
+	if (IS_ROCKETLAKE(dev_priv) ||
+	    IS_TGL_DISP_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_C0))
 		return false;
 
 	return plane_id < PLANE_SPRITE4;
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index 777032d9697b..7a7b99b015a5 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -1706,6 +1706,9 @@ intel_tv_detect(struct drm_connector *connector,
 	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] force=%d\n",
 		    connector->base.id, connector->name, force);
 
+	if (!INTEL_DISPLAY_ENABLED(i915))
+		return connector_status_disconnected;
+
 	if (force) {
 		struct intel_load_detect_pipe tmp;
 		int ret;
diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
index 6faabd4f6d49..54bcc6a6947c 100644
--- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
@@ -293,8 +293,12 @@ struct bdb_general_features {
 #define DVO_PORT_HDMIE		12				/* 193 */
 #define DVO_PORT_DPF		13				/* N/A */
 #define DVO_PORT_HDMIF		14				/* N/A */
-#define DVO_PORT_DPG		15
-#define DVO_PORT_HDMIG		16
+#define DVO_PORT_DPG		15				/* 217 */
+#define DVO_PORT_HDMIG		16				/* 217 */
+#define DVO_PORT_DPH		17				/* 217 */
+#define DVO_PORT_HDMIH		18				/* 217 */
+#define DVO_PORT_DPI		19				/* 217 */
+#define DVO_PORT_HDMII		20				/* 217 */
 #define DVO_PORT_MIPIA		21				/* 171 */
 #define DVO_PORT_MIPIB		22				/* 171 */
 #define DVO_PORT_MIPIC		23				/* 171 */
@@ -330,6 +334,8 @@ enum vbt_gmbus_ddi {
 #define DP_AUX_E 0x50
 #define DP_AUX_F 0x60
 #define DP_AUX_G 0x70
+#define DP_AUX_H 0x80
+#define DP_AUX_I 0x90
 
 #define VBT_DP_MAX_LINK_RATE_HBR3	0
 #define VBT_DP_MAX_LINK_RATE_HBR2	1
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index 052e0b31a2da..5e5522923b1e 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -1585,6 +1585,7 @@ static const struct drm_connector_helper_funcs intel_dsi_connector_helper_funcs
 };
 
 static const struct drm_connector_funcs intel_dsi_connector_funcs = {
+	.detect = intel_panel_detect,
 	.late_register = intel_connector_register,
 	.early_unregister = intel_connector_unregister,
 	.destroy = intel_connector_destroy,
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index d0a514301575..4070b00c3690 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -483,7 +483,7 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 
 	if (dsi_ratio < dsi_ratio_min || dsi_ratio > dsi_ratio_max) {
 		drm_err(&dev_priv->drm,
-			"Cant get a suitable ratio from DSI PLL ratios\n");
+			"Can't get a suitable ratio from DSI PLL ratios\n");
 		return -ECHRNG;
 	} else
 		drm_dbg_kms(&dev_priv->drm, "DSI PLL calculation is Done!!\n");
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
index 278664f831e7..272cf3ea68d5 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
@@ -32,12 +32,13 @@ static void vma_clear_pages(struct i915_vma *vma)
 	vma->pages = NULL;
 }
 
-static int vma_bind(struct i915_address_space *vm,
-		    struct i915_vma *vma,
-		    enum i915_cache_level cache_level,
-		    u32 flags)
+static void vma_bind(struct i915_address_space *vm,
+		     struct i915_vm_pt_stash *stash,
+		     struct i915_vma *vma,
+		     enum i915_cache_level cache_level,
+		     u32 flags)
 {
-	return vm->vma_ops.bind_vma(vm, vma, cache_level, flags);
+	vm->vma_ops.bind_vma(vm, stash, vma, cache_level, flags);
 }
 
 static void vma_unbind(struct i915_address_space *vm, struct i915_vma *vma)
@@ -157,6 +158,7 @@ static void clear_pages_worker(struct work_struct *work)
 	struct clear_pages_work *w = container_of(work, typeof(*w), work);
 	struct drm_i915_gem_object *obj = w->sleeve->vma->obj;
 	struct i915_vma *vma = w->sleeve->vma;
+	struct i915_gem_ww_ctx ww;
 	struct i915_request *rq;
 	struct i915_vma *batch;
 	int err = w->dma.error;
@@ -172,17 +174,20 @@ static void clear_pages_worker(struct work_struct *work)
 	obj->read_domains = I915_GEM_GPU_DOMAINS;
 	obj->write_domain = 0;
 
-	err = i915_vma_pin(vma, 0, 0, PIN_USER);
-	if (unlikely(err))
+	i915_gem_ww_ctx_init(&ww, false);
+	intel_engine_pm_get(w->ce->engine);
+retry:
+	err = intel_context_pin_ww(w->ce, &ww);
+	if (err)
 		goto out_signal;
 
-	batch = intel_emit_vma_fill_blt(w->ce, vma, w->value);
+	batch = intel_emit_vma_fill_blt(w->ce, vma, &ww, w->value);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
-		goto out_unpin;
+		goto out_ctx;
 	}
 
-	rq = intel_context_create_request(w->ce);
+	rq = i915_request_create(w->ce);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto out_batch;
@@ -224,9 +229,19 @@ out_request:
 	i915_request_add(rq);
 out_batch:
 	intel_emit_vma_release(w->ce, batch);
-out_unpin:
-	i915_vma_unpin(vma);
+out_ctx:
+	intel_context_unpin(w->ce);
 out_signal:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+
+	i915_vma_unpin(w->sleeve->vma);
+	intel_engine_pm_put(w->ce->engine);
+
 	if (unlikely(err)) {
 		dma_fence_set_error(&w->dma, err);
 		dma_fence_signal(&w->dma);
@@ -234,6 +249,44 @@ out_signal:
 	}
 }
 
+static int pin_wait_clear_pages_work(struct clear_pages_work *w,
+				     struct intel_context *ce)
+{
+	struct i915_vma *vma = w->sleeve->vma;
+	struct i915_gem_ww_ctx ww;
+	int err;
+
+	i915_gem_ww_ctx_init(&ww, false);
+retry:
+	err = i915_gem_object_lock(vma->obj, &ww);
+	if (err)
+		goto out;
+
+	err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_USER);
+	if (unlikely(err))
+		goto out;
+
+	err = i915_sw_fence_await_reservation(&w->wait,
+					      vma->obj->base.resv, NULL,
+					      true, 0, I915_FENCE_GFP);
+	if (err)
+		goto err_unpin_vma;
+
+	dma_resv_add_excl_fence(vma->obj->base.resv, &w->dma);
+
+err_unpin_vma:
+	if (err)
+		i915_vma_unpin(vma);
+out:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	return err;
+}
+
 static int __i915_sw_fence_call
 clear_pages_work_notify(struct i915_sw_fence *fence,
 			enum i915_sw_fence_notify state)
@@ -287,17 +340,9 @@ int i915_gem_schedule_fill_pages_blt(struct drm_i915_gem_object *obj,
 	dma_fence_init(&work->dma, &clear_pages_work_ops, &fence_lock, 0, 0);
 	i915_sw_fence_init(&work->wait, clear_pages_work_notify);
 
-	i915_gem_object_lock(obj);
-	err = i915_sw_fence_await_reservation(&work->wait,
-					      obj->base.resv, NULL, true, 0,
-					      I915_FENCE_GFP);
-	if (err < 0) {
+	err = pin_wait_clear_pages_work(work, ce);
+	if (err < 0)
 		dma_fence_set_error(&work->dma, err);
-	} else {
-		dma_resv_add_excl_fence(obj->base.resv, &work->dma);
-		err = 0;
-	}
-	i915_gem_object_unlock(obj);
 
 	dma_fence_get(&work->dma);
 	i915_sw_fence_commit(&work->wait);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index ef755dd5e68f..4fd38101bb56 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -390,24 +390,6 @@ __context_engines_static(const struct i915_gem_context *ctx)
 	return rcu_dereference_protected(ctx->engines, true);
 }
 
-static bool __reset_engine(struct intel_engine_cs *engine)
-{
-	struct intel_gt *gt = engine->gt;
-	bool success = false;
-
-	if (!intel_has_reset_engine(gt))
-		return false;
-
-	if (!test_and_set_bit(I915_RESET_ENGINE + engine->id,
-			      &gt->reset.flags)) {
-		success = intel_engine_reset(engine, NULL) == 0;
-		clear_and_wake_up_bit(I915_RESET_ENGINE + engine->id,
-				      &gt->reset.flags);
-	}
-
-	return success;
-}
-
 static void __reset_context(struct i915_gem_context *ctx,
 			    struct intel_engine_cs *engine)
 {
@@ -431,12 +413,7 @@ static bool __cancel_engine(struct intel_engine_cs *engine)
 	 * kill the banned context, we fallback to doing a local reset
 	 * instead.
 	 */
-	if (IS_ACTIVE(CONFIG_DRM_I915_PREEMPT_TIMEOUT) &&
-	    !intel_engine_pulse(engine))
-		return true;
-
-	/* If we are unable to send a pulse, try resetting this engine. */
-	return __reset_engine(engine);
+	return intel_engine_pulse(engine) == 0;
 }
 
 static bool
@@ -460,8 +437,8 @@ __active_engine(struct i915_request *rq, struct intel_engine_cs **active)
 		spin_lock(&locked->active.lock);
 	}
 
-	if (!i915_request_completed(rq)) {
-		if (i915_request_is_active(rq) && rq->fence.error != -EIO)
+	if (i915_request_is_active(rq)) {
+		if (!i915_request_completed(rq))
 			*active = locked;
 		ret = true;
 	}
@@ -479,13 +456,26 @@ static struct intel_engine_cs *active_engine(struct intel_context *ce)
 	if (!ce->timeline)
 		return NULL;
 
+	/*
+	 * rq->link is only SLAB_TYPESAFE_BY_RCU, we need to hold a reference
+	 * to the request to prevent it being transferred to a new timeline
+	 * (and onto a new timeline->requests list).
+	 */
 	rcu_read_lock();
-	list_for_each_entry_rcu(rq, &ce->timeline->requests, link) {
-		if (i915_request_is_active(rq) && i915_request_completed(rq))
-			continue;
+	list_for_each_entry_reverse(rq, &ce->timeline->requests, link) {
+		bool found;
+
+		/* timeline is already completed upto this point? */
+		if (!i915_request_get_rcu(rq))
+			break;
 
 		/* Check with the backend if the request is inflight */
-		if (__active_engine(rq, &engine))
+		found = true;
+		if (likely(rcu_access_pointer(rq->timeline) == ce->timeline))
+			found = __active_engine(rq, &engine);
+
+		i915_request_put(rq);
+		if (found)
 			break;
 	}
 	rcu_read_unlock();
@@ -493,7 +483,7 @@ static struct intel_engine_cs *active_engine(struct intel_context *ce)
 	return engine;
 }
 
-static void kill_engines(struct i915_gem_engines *engines)
+static void kill_engines(struct i915_gem_engines *engines, bool ban)
 {
 	struct i915_gem_engines_iter it;
 	struct intel_context *ce;
@@ -508,7 +498,7 @@ static void kill_engines(struct i915_gem_engines *engines)
 	for_each_gem_engine(ce, engines, it) {
 		struct intel_engine_cs *engine;
 
-		if (intel_context_set_banned(ce))
+		if (ban && intel_context_set_banned(ce))
 			continue;
 
 		/*
@@ -521,7 +511,7 @@ static void kill_engines(struct i915_gem_engines *engines)
 		engine = active_engine(ce);
 
 		/* First attempt to gracefully cancel the context */
-		if (engine && !__cancel_engine(engine))
+		if (engine && !__cancel_engine(engine) && ban)
 			/*
 			 * If we are unable to send a preemptive pulse to bump
 			 * the context from the GPU, we have to resort to a full
@@ -531,8 +521,10 @@ static void kill_engines(struct i915_gem_engines *engines)
 	}
 }
 
-static void kill_stale_engines(struct i915_gem_context *ctx)
+static void kill_context(struct i915_gem_context *ctx)
 {
+	bool ban = (!i915_gem_context_is_persistent(ctx) ||
+		    !ctx->i915->params.enable_hangcheck);
 	struct i915_gem_engines *pos, *next;
 
 	spin_lock_irq(&ctx->stale.lock);
@@ -545,7 +537,7 @@ static void kill_stale_engines(struct i915_gem_context *ctx)
 
 		spin_unlock_irq(&ctx->stale.lock);
 
-		kill_engines(pos);
+		kill_engines(pos, ban);
 
 		spin_lock_irq(&ctx->stale.lock);
 		GEM_BUG_ON(i915_sw_fence_signaled(&pos->fence));
@@ -557,11 +549,6 @@ static void kill_stale_engines(struct i915_gem_context *ctx)
 	spin_unlock_irq(&ctx->stale.lock);
 }
 
-static void kill_context(struct i915_gem_context *ctx)
-{
-	kill_stale_engines(ctx);
-}
-
 static void engines_idle_release(struct i915_gem_context *ctx,
 				 struct i915_gem_engines *engines)
 {
@@ -596,7 +583,7 @@ static void engines_idle_release(struct i915_gem_context *ctx,
 
 kill:
 	if (list_empty(&engines->link)) /* raced, already closed */
-		kill_engines(engines);
+		kill_engines(engines, true);
 
 	i915_sw_fence_commit(&engines->fence);
 }
@@ -654,9 +641,7 @@ static void context_close(struct i915_gem_context *ctx)
 	 * case we opt to forcibly kill off all remaining requests on
 	 * context close.
 	 */
-	if (!i915_gem_context_is_persistent(ctx) ||
-	    !ctx->i915->params.enable_hangcheck)
-		kill_context(ctx);
+	kill_context(ctx);
 
 	i915_gem_context_put(ctx);
 }
@@ -892,7 +877,7 @@ i915_gem_create_context(struct drm_i915_private *i915, unsigned int flags)
 	if (flags & I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE) {
 		struct intel_timeline *timeline;
 
-		timeline = intel_timeline_create(&i915->gt, NULL);
+		timeline = intel_timeline_create(&i915->gt);
 		if (IS_ERR(timeline)) {
 			context_close(ctx);
 			return ERR_CAST(timeline);
@@ -1106,6 +1091,7 @@ I915_SELFTEST_DECLARE(static intel_engine_mask_t context_barrier_inject_fault);
 static int context_barrier_task(struct i915_gem_context *ctx,
 				intel_engine_mask_t engines,
 				bool (*skip)(struct intel_context *ce, void *data),
+				int (*pin)(struct intel_context *ce, struct i915_gem_ww_ctx *ww, void *data),
 				int (*emit)(struct i915_request *rq, void *data),
 				void (*task)(void *data),
 				void *data)
@@ -1113,6 +1099,7 @@ static int context_barrier_task(struct i915_gem_context *ctx,
 	struct context_barrier_task *cb;
 	struct i915_gem_engines_iter it;
 	struct i915_gem_engines *e;
+	struct i915_gem_ww_ctx ww;
 	struct intel_context *ce;
 	int err = 0;
 
@@ -1150,10 +1137,21 @@ static int context_barrier_task(struct i915_gem_context *ctx,
 		if (skip && skip(ce, data))
 			continue;
 
-		rq = intel_context_create_request(ce);
+		i915_gem_ww_ctx_init(&ww, true);
+retry:
+		err = intel_context_pin_ww(ce, &ww);
+		if (err)
+			goto err;
+
+		if (pin)
+			err = pin(ce, &ww, data);
+		if (err)
+			goto err_unpin;
+
+		rq = i915_request_create(ce);
 		if (IS_ERR(rq)) {
 			err = PTR_ERR(rq);
-			break;
+			goto err_unpin;
 		}
 
 		err = 0;
@@ -1163,6 +1161,16 @@ static int context_barrier_task(struct i915_gem_context *ctx,
 			err = i915_active_add_request(&cb->base, rq);
 
 		i915_request_add(rq);
+err_unpin:
+		intel_context_unpin(ce);
+err:
+		if (err == -EDEADLK) {
+			err = i915_gem_ww_ctx_backoff(&ww);
+			if (!err)
+				goto retry;
+		}
+		i915_gem_ww_ctx_fini(&ww);
+
 		if (err)
 			break;
 	}
@@ -1218,6 +1226,17 @@ static void set_ppgtt_barrier(void *data)
 	i915_vm_close(old);
 }
 
+static int pin_ppgtt_update(struct intel_context *ce, struct i915_gem_ww_ctx *ww, void *data)
+{
+	struct i915_address_space *vm = ce->vm;
+
+	if (!HAS_LOGICAL_RING_CONTEXTS(vm->i915))
+		/* ppGTT is not part of the legacy context image */
+		return gen6_ppgtt_pin(i915_vm_to_ppgtt(vm), ww);
+
+	return 0;
+}
+
 static int emit_ppgtt_update(struct i915_request *rq, void *data)
 {
 	struct i915_address_space *vm = rq->context->vm;
@@ -1274,20 +1293,10 @@ static int emit_ppgtt_update(struct i915_request *rq, void *data)
 
 static bool skip_ppgtt_update(struct intel_context *ce, void *data)
 {
-	if (!test_bit(CONTEXT_ALLOC_BIT, &ce->flags))
-		return true;
-
 	if (HAS_LOGICAL_RING_CONTEXTS(ce->engine->i915))
-		return false;
-
-	if (!atomic_read(&ce->pin_count))
-		return true;
-
-	/* ppGTT is not part of the legacy context image */
-	if (gen6_ppgtt_pin(i915_vm_to_ppgtt(ce->vm)))
-		return true;
-
-	return false;
+		return !ce->state;
+	else
+		return !atomic_read(&ce->pin_count);
 }
 
 static int set_ppgtt(struct drm_i915_file_private *file_priv,
@@ -1338,6 +1347,7 @@ static int set_ppgtt(struct drm_i915_file_private *file_priv,
 	 */
 	err = context_barrier_task(ctx, ALL_ENGINES,
 				   skip_ppgtt_update,
+				   pin_ppgtt_update,
 				   emit_ppgtt_update,
 				   set_ppgtt_barrier,
 				   old);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
index 2679380159fc..8dd295dbe241 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
@@ -48,12 +48,9 @@ static struct sg_table *i915_gem_map_dma_buf(struct dma_buf_attachment *attachme
 		src = sg_next(src);
 	}
 
-	if (!dma_map_sg_attrs(attachment->dev,
-			      st->sgl, st->nents, dir,
-			      DMA_ATTR_SKIP_CPU_SYNC)) {
-		ret = -ENOMEM;
+	ret = dma_map_sgtable(attachment->dev, st, dir, DMA_ATTR_SKIP_CPU_SYNC);
+	if (ret)
 		goto err_free_sg;
-	}
 
 	return st;
 
@@ -73,9 +70,7 @@ static void i915_gem_unmap_dma_buf(struct dma_buf_attachment *attachment,
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(attachment->dmabuf);
 
-	dma_unmap_sg_attrs(attachment->dev,
-			   sg->sgl, sg->nents, dir,
-			   DMA_ATTR_SKIP_CPU_SYNC);
+	dma_unmap_sgtable(attachment->dev, sg, dir, DMA_ATTR_SKIP_CPU_SYNC);
 	sg_free_table(sg);
 	kfree(sg);
 
@@ -128,7 +123,7 @@ static int i915_gem_begin_cpu_access(struct dma_buf *dma_buf, enum dma_data_dire
 	if (err)
 		return err;
 
-	err = i915_gem_object_lock_interruptible(obj);
+	err = i915_gem_object_lock_interruptible(obj, NULL);
 	if (err)
 		goto out;
 
@@ -149,7 +144,7 @@ static int i915_gem_end_cpu_access(struct dma_buf *dma_buf, enum dma_data_direct
 	if (err)
 		return err;
 
-	err = i915_gem_object_lock_interruptible(obj);
+	err = i915_gem_object_lock_interruptible(obj, NULL);
 	if (err)
 		goto out;
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index 7f76fc68f498..7c90a63c273d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -32,11 +32,17 @@ void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj)
 	if (!i915_gem_object_is_framebuffer(obj))
 		return;
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	__i915_gem_object_flush_for_display(obj);
 	i915_gem_object_unlock(obj);
 }
 
+void i915_gem_object_flush_if_display_locked(struct drm_i915_gem_object *obj)
+{
+	if (i915_gem_object_is_framebuffer(obj))
+		__i915_gem_object_flush_for_display(obj);
+}
+
 /**
  * Moves a single object to the WC read, and possibly write domain.
  * @obj: object to act on
@@ -197,18 +203,12 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	if (ret)
 		return ret;
 
-	ret = i915_gem_object_lock_interruptible(obj);
-	if (ret)
-		return ret;
-
 	/* Always invalidate stale cachelines */
 	if (obj->cache_level != cache_level) {
 		i915_gem_object_set_cache_coherency(obj, cache_level);
 		obj->cache_dirty = true;
 	}
 
-	i915_gem_object_unlock(obj);
-
 	/* The cache-level will be applied when each vma is rebound. */
 	return i915_gem_object_unbind(obj,
 				      I915_GEM_OBJECT_UNBIND_ACTIVE |
@@ -293,7 +293,12 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 		goto out;
 	}
 
+	ret = i915_gem_object_lock_interruptible(obj, NULL);
+	if (ret)
+		goto out;
+
 	ret = i915_gem_object_set_cache_level(obj, level);
+	i915_gem_object_unlock(obj);
 
 out:
 	i915_gem_object_put(obj);
@@ -313,6 +318,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     unsigned int flags)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+	struct i915_gem_ww_ctx ww;
 	struct i915_vma *vma;
 	int ret;
 
@@ -320,6 +326,11 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	if (HAS_LMEM(i915) && !i915_gem_object_is_lmem(obj))
 		return ERR_PTR(-EINVAL);
 
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	ret = i915_gem_object_lock(obj, &ww);
+	if (ret)
+		goto err;
 	/*
 	 * The display engine is not coherent with the LLC cache on gen6.  As
 	 * a result, we make sure that the pinning that is about to occur is
@@ -334,7 +345,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 					      HAS_WT(i915) ?
 					      I915_CACHE_WT : I915_CACHE_NONE);
 	if (ret)
-		return ERR_PTR(ret);
+		goto err;
 
 	/*
 	 * As the user may map the buffer once pinned in the display plane
@@ -347,18 +358,31 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	vma = ERR_PTR(-ENOSPC);
 	if ((flags & PIN_MAPPABLE) == 0 &&
 	    (!view || view->type == I915_GGTT_VIEW_NORMAL))
-		vma = i915_gem_object_ggtt_pin(obj, view, 0, alignment,
-					       flags |
-					       PIN_MAPPABLE |
-					       PIN_NONBLOCK);
-	if (IS_ERR(vma))
-		vma = i915_gem_object_ggtt_pin(obj, view, 0, alignment, flags);
-	if (IS_ERR(vma))
-		return vma;
+		vma = i915_gem_object_ggtt_pin_ww(obj, &ww, view, 0, alignment,
+						  flags | PIN_MAPPABLE |
+						  PIN_NONBLOCK);
+	if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK))
+		vma = i915_gem_object_ggtt_pin_ww(obj, &ww, view, 0,
+						  alignment, flags);
+	if (IS_ERR(vma)) {
+		ret = PTR_ERR(vma);
+		goto err;
+	}
 
 	vma->display_alignment = max_t(u64, vma->display_alignment, alignment);
 
-	i915_gem_object_flush_if_display(obj);
+	i915_gem_object_flush_if_display_locked(obj);
+
+err:
+	if (ret == -EDEADLK) {
+		ret = i915_gem_ww_ctx_backoff(&ww);
+		if (!ret)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+
+	if (ret)
+		return ERR_PTR(ret);
 
 	return vma;
 }
@@ -536,7 +560,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	if (err)
 		goto out;
 
-	err = i915_gem_object_lock_interruptible(obj);
+	err = i915_gem_object_lock_interruptible(obj, NULL);
 	if (err)
 		goto out_unpin;
 
@@ -576,19 +600,17 @@ int i915_gem_object_prepare_read(struct drm_i915_gem_object *obj,
 	if (!i915_gem_object_has_struct_page(obj))
 		return -ENODEV;
 
-	ret = i915_gem_object_lock_interruptible(obj);
-	if (ret)
-		return ret;
+	assert_object_held(obj);
 
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE,
 				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
-		goto err_unlock;
+		return ret;
 
 	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
-		goto err_unlock;
+		return ret;
 
 	if (obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ ||
 	    !static_cpu_has(X86_FEATURE_CLFLUSH)) {
@@ -616,8 +638,6 @@ out:
 
 err_unpin:
 	i915_gem_object_unpin_pages(obj);
-err_unlock:
-	i915_gem_object_unlock(obj);
 	return ret;
 }
 
@@ -630,20 +650,18 @@ int i915_gem_object_prepare_write(struct drm_i915_gem_object *obj,
 	if (!i915_gem_object_has_struct_page(obj))
 		return -ENODEV;
 
-	ret = i915_gem_object_lock_interruptible(obj);
-	if (ret)
-		return ret;
+	assert_object_held(obj);
 
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_ALL,
 				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
-		goto err_unlock;
+		return ret;
 
 	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
-		goto err_unlock;
+		return ret;
 
 	if (obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE ||
 	    !static_cpu_has(X86_FEATURE_CLFLUSH)) {
@@ -680,7 +698,5 @@ out:
 
 err_unpin:
 	i915_gem_object_unpin_pages(obj);
-err_unlock:
-	i915_gem_object_unlock(obj);
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 446e76e95c38..4b09bcd70cf4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -26,6 +26,7 @@
 #include "i915_gem_ioctls.h"
 #include "i915_sw_fence_work.h"
 #include "i915_trace.h"
+#include "i915_user_extensions.h"
 
 struct eb_vma {
 	struct i915_vma *vma;
@@ -40,11 +41,6 @@ struct eb_vma {
 	u32 handle;
 };
 
-struct eb_vma_array {
-	struct kref kref;
-	struct eb_vma vma[];
-};
-
 enum {
 	FORCE_CPU_RELOC = 1,
 	FORCE_GTT_RELOC,
@@ -57,9 +53,11 @@ enum {
 #define __EXEC_OBJECT_NEEDS_MAP		BIT(29)
 #define __EXEC_OBJECT_NEEDS_BIAS	BIT(28)
 #define __EXEC_OBJECT_INTERNAL_FLAGS	(~0u << 28) /* all of the above */
+#define __EXEC_OBJECT_RESERVED (__EXEC_OBJECT_HAS_PIN | __EXEC_OBJECT_HAS_FENCE)
 
 #define __EXEC_HAS_RELOC	BIT(31)
-#define __EXEC_INTERNAL_FLAGS	(~0u << 31)
+#define __EXEC_ENGINE_PINNED	BIT(30)
+#define __EXEC_INTERNAL_FLAGS	(~0u << 30)
 #define UPDATE			PIN_OFFSET_FIXED
 
 #define BATCH_OFFSET_BIAS (256*1024)
@@ -229,6 +227,13 @@ enum {
  * the batchbuffer in trusted mode, otherwise the ioctl is rejected.
  */
 
+struct eb_fence {
+	struct drm_syncobj *syncobj; /* Use with ptr_mask_bits() */
+	struct dma_fence *dma_fence;
+	u64 value;
+	struct dma_fence_chain *chain_fence;
+};
+
 struct i915_execbuffer {
 	struct drm_i915_private *i915; /** i915 backpointer */
 	struct drm_file *file; /** per-file lookup tables and limits */
@@ -253,6 +258,8 @@ struct i915_execbuffer {
 	/** list of vma that have execobj.relocation_count */
 	struct list_head relocs;
 
+	struct i915_gem_ww_ctx ww;
+
 	/**
 	 * Track the most recently used object for relocations, as we
 	 * frequently have to perform multiple relocations within the same
@@ -268,19 +275,22 @@ struct i915_execbuffer {
 		bool has_fence : 1;
 		bool needs_unfenced : 1;
 
-		struct i915_vma *target;
 		struct i915_request *rq;
-		struct i915_vma *rq_vma;
 		u32 *rq_cmd;
 		unsigned int rq_size;
+		struct intel_gt_buffer_pool_node *pool;
 	} reloc_cache;
 
+	struct intel_gt_buffer_pool_node *reloc_pool; /** relocation pool for -EDEADLK handling */
+	struct intel_context *reloc_context;
+
 	u64 invalid_flags; /** Set of execobj.flags that are invalid */
 	u32 context_flags; /** Set of execobj.flags to insert from the ctx */
 
 	u32 batch_start_offset; /** Location within object of batch */
 	u32 batch_len; /** Length of batch within object */
 	u32 batch_flags; /** Flags composed for emit_bb_start() */
+	struct intel_gt_buffer_pool_node *batch_pool; /** pool node for batch buffer */
 
 	/**
 	 * Indicate either the size of the hastable used to resolve
@@ -289,9 +299,16 @@ struct i915_execbuffer {
 	 */
 	int lut_size;
 	struct hlist_head *buckets; /** ht for relocation handles */
-	struct eb_vma_array *array;
+
+	struct eb_fence *fences;
+	unsigned long num_fences;
 };
 
+static int eb_parse(struct i915_execbuffer *eb);
+static struct i915_request *eb_pin_engine(struct i915_execbuffer *eb,
+					  bool throttle);
+static void eb_unpin_engine(struct i915_execbuffer *eb);
+
 static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb)
 {
 	return intel_engine_requires_cmd_parser(eb->engine) ||
@@ -299,62 +316,8 @@ static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb)
 		 eb->args->batch_len);
 }
 
-static struct eb_vma_array *eb_vma_array_create(unsigned int count)
-{
-	struct eb_vma_array *arr;
-
-	arr = kvmalloc(struct_size(arr, vma, count), GFP_KERNEL | __GFP_NOWARN);
-	if (!arr)
-		return NULL;
-
-	kref_init(&arr->kref);
-	arr->vma[0].vma = NULL;
-
-	return arr;
-}
-
-static inline void eb_unreserve_vma(struct eb_vma *ev)
-{
-	struct i915_vma *vma = ev->vma;
-
-	if (unlikely(ev->flags & __EXEC_OBJECT_HAS_FENCE))
-		__i915_vma_unpin_fence(vma);
-
-	if (ev->flags & __EXEC_OBJECT_HAS_PIN)
-		__i915_vma_unpin(vma);
-
-	ev->flags &= ~(__EXEC_OBJECT_HAS_PIN |
-		       __EXEC_OBJECT_HAS_FENCE);
-}
-
-static void eb_vma_array_destroy(struct kref *kref)
-{
-	struct eb_vma_array *arr = container_of(kref, typeof(*arr), kref);
-	struct eb_vma *ev = arr->vma;
-
-	while (ev->vma) {
-		eb_unreserve_vma(ev);
-		i915_vma_put(ev->vma);
-		ev++;
-	}
-
-	kvfree(arr);
-}
-
-static void eb_vma_array_put(struct eb_vma_array *arr)
-{
-	kref_put(&arr->kref, eb_vma_array_destroy);
-}
-
 static int eb_create(struct i915_execbuffer *eb)
 {
-	/* Allocate an extra slot for use by the command parser + sentinel */
-	eb->array = eb_vma_array_create(eb->buffer_count + 2);
-	if (!eb->array)
-		return -ENOMEM;
-
-	eb->vma = eb->array->vma;
-
 	if (!(eb->args->flags & I915_EXEC_HANDLE_LUT)) {
 		unsigned int size = 1 + ilog2(eb->buffer_count);
 
@@ -388,10 +351,8 @@ static int eb_create(struct i915_execbuffer *eb)
 				break;
 		} while (--size);
 
-		if (unlikely(!size)) {
-			eb_vma_array_put(eb->array);
+		if (unlikely(!size))
 			return -ENOMEM;
-		}
 
 		eb->lut_size = size;
 	} else {
@@ -475,16 +436,17 @@ eb_pin_vma(struct i915_execbuffer *eb,
 		pin_flags |= PIN_GLOBAL;
 
 	/* Attempt to reuse the current location if available */
-	if (unlikely(i915_vma_pin(vma, 0, 0, pin_flags))) {
+	/* TODO: Add -EDEADLK handling here */
+	if (unlikely(i915_vma_pin_ww(vma, &eb->ww, 0, 0, pin_flags))) {
 		if (entry->flags & EXEC_OBJECT_PINNED)
 			return false;
 
 		/* Failing that pick any _free_ space if suitable */
-		if (unlikely(i915_vma_pin(vma,
-					  entry->pad_to_size,
-					  entry->alignment,
-					  eb_pin_flags(entry, ev->flags) |
-					  PIN_USER | PIN_NOEVICT)))
+		if (unlikely(i915_vma_pin_ww(vma, &eb->ww,
+					     entry->pad_to_size,
+					     entry->alignment,
+					     eb_pin_flags(entry, ev->flags) |
+					     PIN_USER | PIN_NOEVICT)))
 			return false;
 	}
 
@@ -502,6 +464,19 @@ eb_pin_vma(struct i915_execbuffer *eb,
 	return !eb_vma_misplaced(entry, vma, ev->flags);
 }
 
+static inline void
+eb_unreserve_vma(struct eb_vma *ev)
+{
+	if (!(ev->flags & __EXEC_OBJECT_HAS_PIN))
+		return;
+
+	if (unlikely(ev->flags & __EXEC_OBJECT_HAS_FENCE))
+		__i915_vma_unpin_fence(ev->vma);
+
+	__i915_vma_unpin(ev->vma);
+	ev->flags &= ~__EXEC_OBJECT_RESERVED;
+}
+
 static int
 eb_validate_vma(struct i915_execbuffer *eb,
 		struct drm_i915_gem_exec_object2 *entry,
@@ -593,16 +568,6 @@ eb_add_vma(struct i915_execbuffer *eb,
 
 		eb->batch = ev;
 	}
-
-	if (eb_pin_vma(eb, entry, ev)) {
-		if (entry->offset != vma->node.start) {
-			entry->offset = vma->node.start | UPDATE;
-			eb->args->flags |= __EXEC_HAS_RELOC;
-		}
-	} else {
-		eb_unreserve_vma(ev);
-		list_add_tail(&ev->bind_link, &eb->unbound);
-	}
 }
 
 static inline int use_cpu_reloc(const struct reloc_cache *cache,
@@ -622,7 +587,7 @@ static inline int use_cpu_reloc(const struct reloc_cache *cache,
 		obj->cache_level != I915_CACHE_NONE);
 }
 
-static int eb_reserve_vma(const struct i915_execbuffer *eb,
+static int eb_reserve_vma(struct i915_execbuffer *eb,
 			  struct eb_vma *ev,
 			  u64 pin_flags)
 {
@@ -637,7 +602,7 @@ static int eb_reserve_vma(const struct i915_execbuffer *eb,
 			return err;
 	}
 
-	err = i915_vma_pin(vma,
+	err = i915_vma_pin_ww(vma, &eb->ww,
 			   entry->pad_to_size, entry->alignment,
 			   eb_pin_flags(entry, ev->flags) | pin_flags);
 	if (err)
@@ -687,10 +652,6 @@ static int eb_reserve(struct i915_execbuffer *eb)
 	 * This avoid unnecessary unbinding of later objects in order to make
 	 * room for the earlier objects *unless* we need to defragment.
 	 */
-
-	if (mutex_lock_interruptible(&eb->i915->drm.struct_mutex))
-		return -EINTR;
-
 	pass = 0;
 	do {
 		list_for_each_entry(ev, &eb->unbound, bind_link) {
@@ -698,8 +659,8 @@ static int eb_reserve(struct i915_execbuffer *eb)
 			if (err)
 				break;
 		}
-		if (!(err == -ENOSPC || err == -EAGAIN))
-			break;
+		if (err != -ENOSPC)
+			return err;
 
 		/* Resort *all* the objects into priority order */
 		INIT_LIST_HEAD(&eb->unbound);
@@ -729,13 +690,6 @@ static int eb_reserve(struct i915_execbuffer *eb)
 		}
 		list_splice_tail(&last, &eb->unbound);
 
-		if (err == -EAGAIN) {
-			mutex_unlock(&eb->i915->drm.struct_mutex);
-			flush_workqueue(eb->i915->mm.userptr_wq);
-			mutex_lock(&eb->i915->drm.struct_mutex);
-			continue;
-		}
-
 		switch (pass++) {
 		case 0:
 			break;
@@ -746,20 +700,15 @@ static int eb_reserve(struct i915_execbuffer *eb)
 			err = i915_gem_evict_vm(eb->context->vm);
 			mutex_unlock(&eb->context->vm->mutex);
 			if (err)
-				goto unlock;
+				return err;
 			break;
 
 		default:
-			err = -ENOSPC;
-			goto unlock;
+			return -ENOSPC;
 		}
 
 		pin_flags = PIN_USER;
 	} while (1);
-
-unlock:
-	mutex_unlock(&eb->i915->drm.struct_mutex);
-	return err;
 }
 
 static unsigned int eb_batch_index(const struct i915_execbuffer *eb)
@@ -882,12 +831,12 @@ static struct i915_vma *eb_lookup_vma(struct i915_execbuffer *eb, u32 handle)
 
 static int eb_lookup_vmas(struct i915_execbuffer *eb)
 {
+	struct drm_i915_private *i915 = eb->i915;
 	unsigned int batch = eb_batch_index(eb);
 	unsigned int i;
 	int err = 0;
 
 	INIT_LIST_HEAD(&eb->relocs);
-	INIT_LIST_HEAD(&eb->unbound);
 
 	for (i = 0; i < eb->buffer_count; i++) {
 		struct i915_vma *vma;
@@ -895,22 +844,83 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb)
 		vma = eb_lookup_vma(eb, eb->exec[i].handle);
 		if (IS_ERR(vma)) {
 			err = PTR_ERR(vma);
-			break;
+			goto err;
 		}
 
 		err = eb_validate_vma(eb, &eb->exec[i], vma);
 		if (unlikely(err)) {
 			i915_vma_put(vma);
-			break;
+			goto err;
 		}
 
 		eb_add_vma(eb, i, batch, vma);
 	}
 
+	if (unlikely(eb->batch->flags & EXEC_OBJECT_WRITE)) {
+		drm_dbg(&i915->drm,
+			"Attempting to use self-modifying batch buffer\n");
+		return -EINVAL;
+	}
+
+	if (range_overflows_t(u64,
+			      eb->batch_start_offset, eb->batch_len,
+			      eb->batch->vma->size)) {
+		drm_dbg(&i915->drm, "Attempting to use out-of-bounds batch\n");
+		return -EINVAL;
+	}
+
+	if (eb->batch_len == 0)
+		eb->batch_len = eb->batch->vma->size - eb->batch_start_offset;
+
+	return 0;
+
+err:
 	eb->vma[i].vma = NULL;
 	return err;
 }
 
+static int eb_validate_vmas(struct i915_execbuffer *eb)
+{
+	unsigned int i;
+	int err;
+
+	INIT_LIST_HEAD(&eb->unbound);
+
+	for (i = 0; i < eb->buffer_count; i++) {
+		struct drm_i915_gem_exec_object2 *entry = &eb->exec[i];
+		struct eb_vma *ev = &eb->vma[i];
+		struct i915_vma *vma = ev->vma;
+
+		err = i915_gem_object_lock(vma->obj, &eb->ww);
+		if (err)
+			return err;
+
+		if (eb_pin_vma(eb, entry, ev)) {
+			if (entry->offset != vma->node.start) {
+				entry->offset = vma->node.start | UPDATE;
+				eb->args->flags |= __EXEC_HAS_RELOC;
+			}
+		} else {
+			eb_unreserve_vma(ev);
+
+			list_add_tail(&ev->bind_link, &eb->unbound);
+			if (drm_mm_node_allocated(&vma->node)) {
+				err = i915_vma_unbind(vma);
+				if (err)
+					return err;
+			}
+		}
+
+		GEM_BUG_ON(drm_mm_node_allocated(&vma->node) &&
+			   eb_vma_misplaced(&eb->exec[i], vma, ev->flags));
+	}
+
+	if (!list_empty(&eb->unbound))
+		return eb_reserve(eb);
+
+	return 0;
+}
+
 static struct eb_vma *
 eb_get_vma(const struct i915_execbuffer *eb, unsigned long handle)
 {
@@ -931,13 +941,31 @@ eb_get_vma(const struct i915_execbuffer *eb, unsigned long handle)
 	}
 }
 
+static void eb_release_vmas(struct i915_execbuffer *eb, bool final)
+{
+	const unsigned int count = eb->buffer_count;
+	unsigned int i;
+
+	for (i = 0; i < count; i++) {
+		struct eb_vma *ev = &eb->vma[i];
+		struct i915_vma *vma = ev->vma;
+
+		if (!vma)
+			break;
+
+		eb_unreserve_vma(ev);
+
+		if (final)
+			i915_vma_put(vma);
+	}
+
+	eb_unpin_engine(eb);
+}
+
 static void eb_destroy(const struct i915_execbuffer *eb)
 {
 	GEM_BUG_ON(eb->reloc_cache.rq);
 
-	if (eb->array)
-		eb_vma_array_put(eb->array);
-
 	if (eb->lut_size > 0)
 		kfree(eb->buckets);
 }
@@ -949,6 +977,14 @@ relocation_target(const struct drm_i915_gem_relocation_entry *reloc,
 	return gen8_canonical_addr((int)reloc->delta + target->node.start);
 }
 
+static void reloc_cache_clear(struct reloc_cache *cache)
+{
+	cache->rq = NULL;
+	cache->rq_cmd = NULL;
+	cache->pool = NULL;
+	cache->rq_size = 0;
+}
+
 static void reloc_cache_init(struct reloc_cache *cache,
 			     struct drm_i915_private *i915)
 {
@@ -961,8 +997,7 @@ static void reloc_cache_init(struct reloc_cache *cache,
 	cache->has_fence = cache->gen < 4;
 	cache->needs_unfenced = INTEL_INFO(i915)->unfenced_needs_alignment;
 	cache->node.flags = 0;
-	cache->rq = NULL;
-	cache->target = NULL;
+	reloc_cache_clear(cache);
 }
 
 static inline void *unmask_page(unsigned long p)
@@ -984,132 +1019,60 @@ static inline struct i915_ggtt *cache_to_ggtt(struct reloc_cache *cache)
 	return &i915->ggtt;
 }
 
-#define RELOC_TAIL 4
-
-static int reloc_gpu_chain(struct reloc_cache *cache)
+static void reloc_cache_put_pool(struct i915_execbuffer *eb, struct reloc_cache *cache)
 {
-	struct intel_gt_buffer_pool_node *pool;
-	struct i915_request *rq = cache->rq;
-	struct i915_vma *batch;
-	u32 *cmd;
-	int err;
-
-	pool = intel_gt_get_buffer_pool(rq->engine->gt, PAGE_SIZE);
-	if (IS_ERR(pool))
-		return PTR_ERR(pool);
-
-	batch = i915_vma_instance(pool->obj, rq->context->vm, NULL);
-	if (IS_ERR(batch)) {
-		err = PTR_ERR(batch);
-		goto out_pool;
-	}
-
-	err = i915_vma_pin(batch, 0, 0, PIN_USER | PIN_NONBLOCK);
-	if (err)
-		goto out_pool;
-
-	GEM_BUG_ON(cache->rq_size + RELOC_TAIL > PAGE_SIZE  / sizeof(u32));
-	cmd = cache->rq_cmd + cache->rq_size;
-	*cmd++ = MI_ARB_CHECK;
-	if (cache->gen >= 8)
-		*cmd++ = MI_BATCH_BUFFER_START_GEN8;
-	else if (cache->gen >= 6)
-		*cmd++ = MI_BATCH_BUFFER_START;
-	else
-		*cmd++ = MI_BATCH_BUFFER_START | MI_BATCH_GTT;
-	*cmd++ = lower_32_bits(batch->node.start);
-	*cmd++ = upper_32_bits(batch->node.start); /* Always 0 for gen<8 */
-	i915_gem_object_flush_map(cache->rq_vma->obj);
-	i915_gem_object_unpin_map(cache->rq_vma->obj);
-	cache->rq_vma = NULL;
-
-	err = intel_gt_buffer_pool_mark_active(pool, rq);
-	if (err == 0) {
-		i915_vma_lock(batch);
-		err = i915_request_await_object(rq, batch->obj, false);
-		if (err == 0)
-			err = i915_vma_move_to_active(batch, rq, 0);
-		i915_vma_unlock(batch);
-	}
-	i915_vma_unpin(batch);
-	if (err)
-		goto out_pool;
-
-	cmd = i915_gem_object_pin_map(batch->obj,
-				      cache->has_llc ?
-				      I915_MAP_FORCE_WB :
-				      I915_MAP_FORCE_WC);
-	if (IS_ERR(cmd)) {
-		err = PTR_ERR(cmd);
-		goto out_pool;
-	}
-
-	/* Return with batch mapping (cmd) still pinned */
-	cache->rq_cmd = cmd;
-	cache->rq_size = 0;
-	cache->rq_vma = batch;
-
-out_pool:
-	intel_gt_buffer_pool_put(pool);
-	return err;
-}
+	if (!cache->pool)
+		return;
 
-static unsigned int reloc_bb_flags(const struct reloc_cache *cache)
-{
-	return cache->gen > 5 ? 0 : I915_DISPATCH_SECURE;
+	/*
+	 * This is a bit nasty, normally we keep objects locked until the end
+	 * of execbuffer, but we already submit this, and have to unlock before
+	 * dropping the reference. Fortunately we can only hold 1 pool node at
+	 * a time, so this should be harmless.
+	 */
+	i915_gem_ww_unlock_single(cache->pool->obj);
+	intel_gt_buffer_pool_put(cache->pool);
+	cache->pool = NULL;
 }
 
-static int reloc_gpu_flush(struct reloc_cache *cache)
+static void reloc_gpu_flush(struct i915_execbuffer *eb, struct reloc_cache *cache)
 {
-	struct i915_request *rq;
-	int err;
+	struct drm_i915_gem_object *obj = cache->rq->batch->obj;
 
-	rq = fetch_and_zero(&cache->rq);
-	if (!rq)
-		return 0;
+	GEM_BUG_ON(cache->rq_size >= obj->base.size / sizeof(u32));
+	cache->rq_cmd[cache->rq_size] = MI_BATCH_BUFFER_END;
 
-	if (cache->rq_vma) {
-		struct drm_i915_gem_object *obj = cache->rq_vma->obj;
+	__i915_gem_object_flush_map(obj, 0, sizeof(u32) * (cache->rq_size + 1));
+	i915_gem_object_unpin_map(obj);
 
-		GEM_BUG_ON(cache->rq_size >= obj->base.size / sizeof(u32));
-		cache->rq_cmd[cache->rq_size++] = MI_BATCH_BUFFER_END;
+	intel_gt_chipset_flush(cache->rq->engine->gt);
 
-		__i915_gem_object_flush_map(obj,
-					    0, sizeof(u32) * cache->rq_size);
-		i915_gem_object_unpin_map(obj);
-	}
+	i915_request_add(cache->rq);
+	reloc_cache_put_pool(eb, cache);
+	reloc_cache_clear(cache);
 
-	err = 0;
-	if (rq->engine->emit_init_breadcrumb)
-		err = rq->engine->emit_init_breadcrumb(rq);
-	if (!err)
-		err = rq->engine->emit_bb_start(rq,
-						rq->batch->node.start,
-						PAGE_SIZE,
-						reloc_bb_flags(cache));
-	if (err)
-		i915_request_set_error_once(rq, err);
-
-	intel_gt_chipset_flush(rq->engine->gt);
-	i915_request_add(rq);
-
-	return err;
+	eb->reloc_pool = NULL;
 }
 
-static void reloc_cache_reset(struct reloc_cache *cache)
+static void reloc_cache_reset(struct reloc_cache *cache, struct i915_execbuffer *eb)
 {
 	void *vaddr;
 
+	if (cache->rq)
+		reloc_gpu_flush(eb, cache);
+
 	if (!cache->vaddr)
 		return;
 
 	vaddr = unmask_page(cache->vaddr);
 	if (cache->vaddr & KMAP) {
+		struct drm_i915_gem_object *obj =
+			(struct drm_i915_gem_object *)cache->node.mm;
 		if (cache->vaddr & CLFLUSH_AFTER)
 			mb();
 
 		kunmap_atomic(vaddr);
-		i915_gem_object_finish_access((struct drm_i915_gem_object *)cache->node.mm);
+		i915_gem_object_finish_access(obj);
 	} else {
 		struct i915_ggtt *ggtt = cache_to_ggtt(cache);
 
@@ -1134,9 +1097,10 @@ static void reloc_cache_reset(struct reloc_cache *cache)
 
 static void *reloc_kmap(struct drm_i915_gem_object *obj,
 			struct reloc_cache *cache,
-			unsigned long page)
+			unsigned long pageno)
 {
 	void *vaddr;
+	struct page *page;
 
 	if (cache->vaddr) {
 		kunmap_atomic(unmask_page(cache->vaddr));
@@ -1157,17 +1121,22 @@ static void *reloc_kmap(struct drm_i915_gem_object *obj,
 			mb();
 	}
 
-	vaddr = kmap_atomic(i915_gem_object_get_dirty_page(obj, page));
+	page = i915_gem_object_get_page(obj, pageno);
+	if (!obj->mm.dirty)
+		set_page_dirty(page);
+
+	vaddr = kmap_atomic(page);
 	cache->vaddr = unmask_flags(cache->vaddr) | (unsigned long)vaddr;
-	cache->page = page;
+	cache->page = pageno;
 
 	return vaddr;
 }
 
 static void *reloc_iomap(struct drm_i915_gem_object *obj,
-			 struct reloc_cache *cache,
+			 struct i915_execbuffer *eb,
 			 unsigned long page)
 {
+	struct reloc_cache *cache = &eb->reloc_cache;
 	struct i915_ggtt *ggtt = cache_to_ggtt(cache);
 	unsigned long offset;
 	void *vaddr;
@@ -1185,16 +1154,17 @@ static void *reloc_iomap(struct drm_i915_gem_object *obj,
 		if (use_cpu_reloc(cache, obj))
 			return NULL;
 
-		i915_gem_object_lock(obj);
 		err = i915_gem_object_set_to_gtt_domain(obj, true);
-		i915_gem_object_unlock(obj);
 		if (err)
 			return ERR_PTR(err);
 
-		vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
-					       PIN_MAPPABLE |
-					       PIN_NONBLOCK /* NOWARN */ |
-					       PIN_NOEVICT);
+		vma = i915_gem_object_ggtt_pin_ww(obj, &eb->ww, NULL, 0, 0,
+						  PIN_MAPPABLE |
+						  PIN_NONBLOCK /* NOWARN */ |
+						  PIN_NOEVICT);
+		if (vma == ERR_PTR(-EDEADLK))
+			return vma;
+
 		if (IS_ERR(vma)) {
 			memset(&cache->node, 0, sizeof(cache->node));
 			mutex_lock(&ggtt->vm.mutex);
@@ -1230,9 +1200,10 @@ static void *reloc_iomap(struct drm_i915_gem_object *obj,
 }
 
 static void *reloc_vaddr(struct drm_i915_gem_object *obj,
-			 struct reloc_cache *cache,
+			 struct i915_execbuffer *eb,
 			 unsigned long page)
 {
+	struct reloc_cache *cache = &eb->reloc_cache;
 	void *vaddr;
 
 	if (cache->page == page) {
@@ -1240,7 +1211,7 @@ static void *reloc_vaddr(struct drm_i915_gem_object *obj,
 	} else {
 		vaddr = NULL;
 		if ((cache->vaddr & KMAP) == 0)
-			vaddr = reloc_iomap(obj, cache, page);
+			vaddr = reloc_iomap(obj, eb, page);
 		if (!vaddr)
 			vaddr = reloc_kmap(obj, cache, page);
 	}
@@ -1276,7 +1247,7 @@ static int reloc_move_to_gpu(struct i915_request *rq, struct i915_vma *vma)
 	struct drm_i915_gem_object *obj = vma->obj;
 	int err;
 
-	i915_vma_lock(vma);
+	assert_vma_held(vma);
 
 	if (obj->cache_dirty & ~obj->cache_coherent)
 		i915_gem_clflush_object(obj, 0);
@@ -1286,25 +1257,31 @@ static int reloc_move_to_gpu(struct i915_request *rq, struct i915_vma *vma)
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
 
-	i915_vma_unlock(vma);
-
 	return err;
 }
 
 static int __reloc_gpu_alloc(struct i915_execbuffer *eb,
 			     struct intel_engine_cs *engine,
+			     struct i915_vma *vma,
 			     unsigned int len)
 {
 	struct reloc_cache *cache = &eb->reloc_cache;
-	struct intel_gt_buffer_pool_node *pool;
+	struct intel_gt_buffer_pool_node *pool = eb->reloc_pool;
 	struct i915_request *rq;
 	struct i915_vma *batch;
 	u32 *cmd;
 	int err;
 
-	pool = intel_gt_get_buffer_pool(engine->gt, PAGE_SIZE);
-	if (IS_ERR(pool))
-		return PTR_ERR(pool);
+	if (!pool) {
+		pool = intel_gt_get_buffer_pool(engine->gt, PAGE_SIZE);
+		if (IS_ERR(pool))
+			return PTR_ERR(pool);
+	}
+	eb->reloc_pool = NULL;
+
+	err = i915_gem_object_lock(pool->obj, &eb->ww);
+	if (err)
+		goto err_pool;
 
 	cmd = i915_gem_object_pin_map(pool->obj,
 				      cache->has_llc ?
@@ -1312,35 +1289,42 @@ static int __reloc_gpu_alloc(struct i915_execbuffer *eb,
 				      I915_MAP_FORCE_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
-		goto out_pool;
+		goto err_pool;
 	}
 
-	batch = i915_vma_instance(pool->obj, eb->context->vm, NULL);
+	batch = i915_vma_instance(pool->obj, vma->vm, NULL);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
 		goto err_unmap;
 	}
 
-	err = i915_vma_pin(batch, 0, 0, PIN_USER | PIN_NONBLOCK);
+	err = i915_vma_pin_ww(batch, &eb->ww, 0, 0, PIN_USER | PIN_NONBLOCK);
 	if (err)
 		goto err_unmap;
 
 	if (engine == eb->context->engine) {
 		rq = i915_request_create(eb->context);
 	} else {
-		struct intel_context *ce;
+		struct intel_context *ce = eb->reloc_context;
 
-		ce = intel_context_create(engine);
-		if (IS_ERR(ce)) {
-			err = PTR_ERR(ce);
-			goto err_unpin;
+		if (!ce) {
+			ce = intel_context_create(engine);
+			if (IS_ERR(ce)) {
+				err = PTR_ERR(ce);
+				goto err_unpin;
+			}
+
+			i915_vm_put(ce->vm);
+			ce->vm = i915_vm_get(eb->context->vm);
+			eb->reloc_context = ce;
 		}
 
-		i915_vm_put(ce->vm);
-		ce->vm = i915_vm_get(eb->context->vm);
+		err = intel_context_pin_ww(ce, &eb->ww);
+		if (err)
+			goto err_unpin;
 
-		rq = intel_context_create_request(ce);
-		intel_context_put(ce);
+		rq = i915_request_create(ce);
+		intel_context_unpin(ce);
 	}
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
@@ -1351,11 +1335,20 @@ static int __reloc_gpu_alloc(struct i915_execbuffer *eb,
 	if (err)
 		goto err_request;
 
-	i915_vma_lock(batch);
+	err = reloc_move_to_gpu(rq, vma);
+	if (err)
+		goto err_request;
+
+	err = eb->engine->emit_bb_start(rq,
+					batch->node.start, PAGE_SIZE,
+					cache->gen > 5 ? 0 : I915_DISPATCH_SECURE);
+	if (err)
+		goto skip_request;
+
+	assert_vma_held(batch);
 	err = i915_request_await_object(rq, batch->obj, false);
 	if (err == 0)
 		err = i915_vma_move_to_active(batch, rq, 0);
-	i915_vma_unlock(batch);
 	if (err)
 		goto skip_request;
 
@@ -1365,10 +1358,10 @@ static int __reloc_gpu_alloc(struct i915_execbuffer *eb,
 	cache->rq = rq;
 	cache->rq_cmd = cmd;
 	cache->rq_size = 0;
-	cache->rq_vma = batch;
+	cache->pool = pool;
 
 	/* Return with batch mapping (cmd) still pinned */
-	goto out_pool;
+	return 0;
 
 skip_request:
 	i915_request_set_error_once(rq, err);
@@ -1378,8 +1371,8 @@ err_unpin:
 	i915_vma_unpin(batch);
 err_unmap:
 	i915_gem_object_unpin_map(pool->obj);
-out_pool:
-	intel_gt_buffer_pool_put(pool);
+err_pool:
+	eb->reloc_pool = pool;
 	return err;
 }
 
@@ -1394,9 +1387,12 @@ static u32 *reloc_gpu(struct i915_execbuffer *eb,
 {
 	struct reloc_cache *cache = &eb->reloc_cache;
 	u32 *cmd;
-	int err;
+
+	if (cache->rq_size > PAGE_SIZE/sizeof(u32) - (len + 1))
+		reloc_gpu_flush(eb, cache);
 
 	if (unlikely(!cache->rq)) {
+		int err;
 		struct intel_engine_cs *engine = eb->engine;
 
 		if (!reloc_can_use_engine(engine)) {
@@ -1405,31 +1401,11 @@ static u32 *reloc_gpu(struct i915_execbuffer *eb,
 				return ERR_PTR(-ENODEV);
 		}
 
-		err = __reloc_gpu_alloc(eb, engine, len);
+		err = __reloc_gpu_alloc(eb, engine, vma, len);
 		if (unlikely(err))
 			return ERR_PTR(err);
 	}
 
-	if (vma != cache->target) {
-		err = reloc_move_to_gpu(cache->rq, vma);
-		if (unlikely(err)) {
-			i915_request_set_error_once(cache->rq, err);
-			return ERR_PTR(err);
-		}
-
-		cache->target = vma;
-	}
-
-	if (unlikely(cache->rq_size + len >
-		     PAGE_SIZE / sizeof(u32) - RELOC_TAIL)) {
-		err = reloc_gpu_chain(cache);
-		if (unlikely(err)) {
-			i915_request_set_error_once(cache->rq, err);
-			return ERR_PTR(err);
-		}
-	}
-
-	GEM_BUG_ON(cache->rq_size + len >= PAGE_SIZE  / sizeof(u32));
 	cmd = cache->rq_cmd + cache->rq_size;
 	cache->rq_size += len;
 
@@ -1461,7 +1437,7 @@ static unsigned long vma_phys_addr(struct i915_vma *vma, u32 offset)
 	return addr + offset_in_page(offset);
 }
 
-static bool __reloc_entry_gpu(struct i915_execbuffer *eb,
+static int __reloc_entry_gpu(struct i915_execbuffer *eb,
 			      struct i915_vma *vma,
 			      u64 offset,
 			      u64 target_addr)
@@ -1479,7 +1455,9 @@ static bool __reloc_entry_gpu(struct i915_execbuffer *eb,
 		len = 3;
 
 	batch = reloc_gpu(eb, vma, len);
-	if (IS_ERR(batch))
+	if (batch == ERR_PTR(-EDEADLK))
+		return -EDEADLK;
+	else if (IS_ERR(batch))
 		return false;
 
 	addr = gen8_canonical_addr(vma->node.start + offset);
@@ -1532,7 +1510,7 @@ static bool __reloc_entry_gpu(struct i915_execbuffer *eb,
 	return true;
 }
 
-static bool reloc_entry_gpu(struct i915_execbuffer *eb,
+static int reloc_entry_gpu(struct i915_execbuffer *eb,
 			    struct i915_vma *vma,
 			    u64 offset,
 			    u64 target_addr)
@@ -1554,14 +1532,17 @@ relocate_entry(struct i915_vma *vma,
 {
 	u64 target_addr = relocation_target(reloc, target);
 	u64 offset = reloc->offset;
+	int reloc_gpu = reloc_entry_gpu(eb, vma, offset, target_addr);
 
-	if (!reloc_entry_gpu(eb, vma, offset, target_addr)) {
+	if (reloc_gpu < 0)
+		return reloc_gpu;
+
+	if (!reloc_gpu) {
 		bool wide = eb->reloc_cache.use_64bit_reloc;
 		void *vaddr;
 
 repeat:
-		vaddr = reloc_vaddr(vma->obj,
-				    &eb->reloc_cache,
+		vaddr = reloc_vaddr(vma->obj, eb,
 				    offset >> PAGE_SHIFT);
 		if (IS_ERR(vaddr))
 			return PTR_ERR(vaddr);
@@ -1712,7 +1693,9 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev)
 		 * we would try to acquire the struct mutex again. Obviously
 		 * this is bad and so lockdep complains vehemently.
 		 */
-		copied = __copy_from_user(r, urelocs, count * sizeof(r[0]));
+		pagefault_disable();
+		copied = __copy_from_user_inatomic(r, urelocs, count * sizeof(r[0]));
+		pagefault_enable();
 		if (unlikely(copied)) {
 			remain = -EFAULT;
 			goto out;
@@ -1756,74 +1739,400 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev)
 		urelocs += ARRAY_SIZE(stack);
 	} while (remain);
 out:
-	reloc_cache_reset(&eb->reloc_cache);
+	reloc_cache_reset(&eb->reloc_cache, eb);
 	return remain;
 }
 
-static int eb_relocate(struct i915_execbuffer *eb)
+static int
+eb_relocate_vma_slow(struct i915_execbuffer *eb, struct eb_vma *ev)
 {
+	const struct drm_i915_gem_exec_object2 *entry = ev->exec;
+	struct drm_i915_gem_relocation_entry *relocs =
+		u64_to_ptr(typeof(*relocs), entry->relocs_ptr);
+	unsigned int i;
 	int err;
 
-	err = eb_lookup_vmas(eb);
-	if (err)
-		return err;
+	for (i = 0; i < entry->relocation_count; i++) {
+		u64 offset = eb_relocate_entry(eb, ev, &relocs[i]);
+
+		if ((s64)offset < 0) {
+			err = (int)offset;
+			goto err;
+		}
+	}
+	err = 0;
+err:
+	reloc_cache_reset(&eb->reloc_cache, eb);
+	return err;
+}
+
+static int check_relocations(const struct drm_i915_gem_exec_object2 *entry)
+{
+	const char __user *addr, *end;
+	unsigned long size;
+	char __maybe_unused c;
+
+	size = entry->relocation_count;
+	if (size == 0)
+		return 0;
 
-	if (!list_empty(&eb->unbound)) {
-		err = eb_reserve(eb);
+	if (size > N_RELOC(ULONG_MAX))
+		return -EINVAL;
+
+	addr = u64_to_user_ptr(entry->relocs_ptr);
+	size *= sizeof(struct drm_i915_gem_relocation_entry);
+	if (!access_ok(addr, size))
+		return -EFAULT;
+
+	end = addr + size;
+	for (; addr < end; addr += PAGE_SIZE) {
+		int err = __get_user(c, addr);
 		if (err)
 			return err;
 	}
+	return __get_user(c, end - 1);
+}
 
-	/* The objects are in their final locations, apply the relocations. */
-	if (eb->args->flags & __EXEC_HAS_RELOC) {
-		struct eb_vma *ev;
-		int flush;
+static int eb_copy_relocations(const struct i915_execbuffer *eb)
+{
+	struct drm_i915_gem_relocation_entry *relocs;
+	const unsigned int count = eb->buffer_count;
+	unsigned int i;
+	int err;
 
-		list_for_each_entry(ev, &eb->relocs, reloc_link) {
+	for (i = 0; i < count; i++) {
+		const unsigned int nreloc = eb->exec[i].relocation_count;
+		struct drm_i915_gem_relocation_entry __user *urelocs;
+		unsigned long size;
+		unsigned long copied;
+
+		if (nreloc == 0)
+			continue;
+
+		err = check_relocations(&eb->exec[i]);
+		if (err)
+			goto err;
+
+		urelocs = u64_to_user_ptr(eb->exec[i].relocs_ptr);
+		size = nreloc * sizeof(*relocs);
+
+		relocs = kvmalloc_array(size, 1, GFP_KERNEL);
+		if (!relocs) {
+			err = -ENOMEM;
+			goto err;
+		}
+
+		/* copy_from_user is limited to < 4GiB */
+		copied = 0;
+		do {
+			unsigned int len =
+				min_t(u64, BIT_ULL(31), size - copied);
+
+			if (__copy_from_user((char *)relocs + copied,
+					     (char __user *)urelocs + copied,
+					     len))
+				goto end;
+
+			copied += len;
+		} while (copied < size);
+
+		/*
+		 * As we do not update the known relocation offsets after
+		 * relocating (due to the complexities in lock handling),
+		 * we need to mark them as invalid now so that we force the
+		 * relocation processing next time. Just in case the target
+		 * object is evicted and then rebound into its old
+		 * presumed_offset before the next execbuffer - if that
+		 * happened we would make the mistake of assuming that the
+		 * relocations were valid.
+		 */
+		if (!user_access_begin(urelocs, size))
+			goto end;
+
+		for (copied = 0; copied < nreloc; copied++)
+			unsafe_put_user(-1,
+					&urelocs[copied].presumed_offset,
+					end_user);
+		user_access_end();
+
+		eb->exec[i].relocs_ptr = (uintptr_t)relocs;
+	}
+
+	return 0;
+
+end_user:
+	user_access_end();
+end:
+	kvfree(relocs);
+	err = -EFAULT;
+err:
+	while (i--) {
+		relocs = u64_to_ptr(typeof(*relocs), eb->exec[i].relocs_ptr);
+		if (eb->exec[i].relocation_count)
+			kvfree(relocs);
+	}
+	return err;
+}
+
+static int eb_prefault_relocations(const struct i915_execbuffer *eb)
+{
+	const unsigned int count = eb->buffer_count;
+	unsigned int i;
+
+	for (i = 0; i < count; i++) {
+		int err;
+
+		err = check_relocations(&eb->exec[i]);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static noinline int eb_relocate_parse_slow(struct i915_execbuffer *eb,
+					   struct i915_request *rq)
+{
+	bool have_copy = false;
+	struct eb_vma *ev;
+	int err = 0;
+
+repeat:
+	if (signal_pending(current)) {
+		err = -ERESTARTSYS;
+		goto out;
+	}
+
+	/* We may process another execbuffer during the unlock... */
+	eb_release_vmas(eb, false);
+	i915_gem_ww_ctx_fini(&eb->ww);
+
+	if (rq) {
+		/* nonblocking is always false */
+		if (i915_request_wait(rq, I915_WAIT_INTERRUPTIBLE,
+				      MAX_SCHEDULE_TIMEOUT) < 0) {
+			i915_request_put(rq);
+			rq = NULL;
+
+			err = -EINTR;
+			goto err_relock;
+		}
+
+		i915_request_put(rq);
+		rq = NULL;
+	}
+
+	/*
+	 * We take 3 passes through the slowpatch.
+	 *
+	 * 1 - we try to just prefault all the user relocation entries and
+	 * then attempt to reuse the atomic pagefault disabled fast path again.
+	 *
+	 * 2 - we copy the user entries to a local buffer here outside of the
+	 * local and allow ourselves to wait upon any rendering before
+	 * relocations
+	 *
+	 * 3 - we already have a local copy of the relocation entries, but
+	 * were interrupted (EAGAIN) whilst waiting for the objects, try again.
+	 */
+	if (!err) {
+		err = eb_prefault_relocations(eb);
+	} else if (!have_copy) {
+		err = eb_copy_relocations(eb);
+		have_copy = err == 0;
+	} else {
+		cond_resched();
+		err = 0;
+	}
+
+	if (!err)
+		flush_workqueue(eb->i915->mm.userptr_wq);
+
+err_relock:
+	i915_gem_ww_ctx_init(&eb->ww, true);
+	if (err)
+		goto out;
+
+	/* reacquire the objects */
+repeat_validate:
+	rq = eb_pin_engine(eb, false);
+	if (IS_ERR(rq)) {
+		err = PTR_ERR(rq);
+		rq = NULL;
+		goto err;
+	}
+
+	/* We didn't throttle, should be NULL */
+	GEM_WARN_ON(rq);
+
+	err = eb_validate_vmas(eb);
+	if (err)
+		goto err;
+
+	GEM_BUG_ON(!eb->batch);
+
+	list_for_each_entry(ev, &eb->relocs, reloc_link) {
+		if (!have_copy) {
+			pagefault_disable();
 			err = eb_relocate_vma(eb, ev);
+			pagefault_enable();
+			if (err)
+				break;
+		} else {
+			err = eb_relocate_vma_slow(eb, ev);
 			if (err)
 				break;
 		}
+	}
+
+	if (err == -EDEADLK)
+		goto err;
 
-		flush = reloc_gpu_flush(&eb->reloc_cache);
+	if (err && !have_copy)
+		goto repeat;
+
+	if (err)
+		goto err;
+
+	/* as last step, parse the command buffer */
+	err = eb_parse(eb);
+	if (err)
+		goto err;
+
+	/*
+	 * Leave the user relocations as are, this is the painfully slow path,
+	 * and we want to avoid the complication of dropping the lock whilst
+	 * having buffers reserved in the aperture and so causing spurious
+	 * ENOSPC for random operations.
+	 */
+
+err:
+	if (err == -EDEADLK) {
+		eb_release_vmas(eb, false);
+		err = i915_gem_ww_ctx_backoff(&eb->ww);
 		if (!err)
-			err = flush;
+			goto repeat_validate;
+	}
+
+	if (err == -EAGAIN)
+		goto repeat;
+
+out:
+	if (have_copy) {
+		const unsigned int count = eb->buffer_count;
+		unsigned int i;
+
+		for (i = 0; i < count; i++) {
+			const struct drm_i915_gem_exec_object2 *entry =
+				&eb->exec[i];
+			struct drm_i915_gem_relocation_entry *relocs;
+
+			if (!entry->relocation_count)
+				continue;
+
+			relocs = u64_to_ptr(typeof(*relocs), entry->relocs_ptr);
+			kvfree(relocs);
+		}
 	}
 
+	if (rq)
+		i915_request_put(rq);
+
 	return err;
 }
 
-static int eb_move_to_gpu(struct i915_execbuffer *eb)
+static int eb_relocate_parse(struct i915_execbuffer *eb)
 {
-	const unsigned int count = eb->buffer_count;
-	struct ww_acquire_ctx acquire;
-	unsigned int i;
-	int err = 0;
+	int err;
+	struct i915_request *rq = NULL;
+	bool throttle = true;
 
-	ww_acquire_init(&acquire, &reservation_ww_class);
+retry:
+	rq = eb_pin_engine(eb, throttle);
+	if (IS_ERR(rq)) {
+		err = PTR_ERR(rq);
+		rq = NULL;
+		if (err != -EDEADLK)
+			return err;
 
-	for (i = 0; i < count; i++) {
-		struct eb_vma *ev = &eb->vma[i];
-		struct i915_vma *vma = ev->vma;
+		goto err;
+	}
 
-		err = ww_mutex_lock_interruptible(&vma->resv->lock, &acquire);
-		if (err == -EDEADLK) {
-			GEM_BUG_ON(i == 0);
-			do {
-				int j = i - 1;
+	if (rq) {
+		bool nonblock = eb->file->filp->f_flags & O_NONBLOCK;
 
-				ww_mutex_unlock(&eb->vma[j].vma->resv->lock);
+		/* Need to drop all locks now for throttling, take slowpath */
+		err = i915_request_wait(rq, I915_WAIT_INTERRUPTIBLE, 0);
+		if (err == -ETIME) {
+			if (nonblock) {
+				err = -EWOULDBLOCK;
+				i915_request_put(rq);
+				goto err;
+			}
+			goto slow;
+		}
+		i915_request_put(rq);
+		rq = NULL;
+	}
 
-				swap(eb->vma[i],  eb->vma[j]);
-			} while (--i);
+	/* only throttle once, even if we didn't need to throttle */
+	throttle = false;
 
-			err = ww_mutex_lock_slow_interruptible(&vma->resv->lock,
-							       &acquire);
+	err = eb_validate_vmas(eb);
+	if (err == -EAGAIN)
+		goto slow;
+	else if (err)
+		goto err;
+
+	/* The objects are in their final locations, apply the relocations. */
+	if (eb->args->flags & __EXEC_HAS_RELOC) {
+		struct eb_vma *ev;
+
+		list_for_each_entry(ev, &eb->relocs, reloc_link) {
+			err = eb_relocate_vma(eb, ev);
+			if (err)
+				break;
 		}
-		if (err)
-			break;
+
+		if (err == -EDEADLK)
+			goto err;
+		else if (err)
+			goto slow;
+	}
+
+	if (!err)
+		err = eb_parse(eb);
+
+err:
+	if (err == -EDEADLK) {
+		eb_release_vmas(eb, false);
+		err = i915_gem_ww_ctx_backoff(&eb->ww);
+		if (!err)
+			goto retry;
 	}
-	ww_acquire_done(&acquire);
+
+	return err;
+
+slow:
+	err = eb_relocate_parse_slow(eb, rq);
+	if (err)
+		/*
+		 * If the user expects the execobject.offset and
+		 * reloc.presumed_offset to be an exact match,
+		 * as for using NO_RELOC, then we cannot update
+		 * the execobject.offset until we have completed
+		 * relocation.
+		 */
+		eb->args->flags &= ~__EXEC_HAS_RELOC;
+
+	return err;
+}
+
+static int eb_move_to_gpu(struct i915_execbuffer *eb)
+{
+	const unsigned int count = eb->buffer_count;
+	unsigned int i = count;
+	int err = 0;
 
 	while (i--) {
 		struct eb_vma *ev = &eb->vma[i];
@@ -1868,13 +2177,7 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb)
 
 		if (err == 0)
 			err = i915_vma_move_to_active(vma, eb->request, flags);
-
-		i915_vma_unlock(vma);
-		eb_unreserve_vma(ev);
 	}
-	ww_acquire_fini(&acquire);
-
-	eb_vma_array_put(fetch_and_zero(&eb->array));
 
 	if (unlikely(err))
 		goto err_skip;
@@ -1894,7 +2197,8 @@ static int i915_gem_check_execbuffer(struct drm_i915_gem_execbuffer2 *exec)
 		return -EINVAL;
 
 	/* Kernel clipping was a DRI1 misfeature */
-	if (!(exec->flags & I915_EXEC_FENCE_ARRAY)) {
+	if (!(exec->flags & (I915_EXEC_FENCE_ARRAY |
+			     I915_EXEC_USE_EXTENSIONS))) {
 		if (exec->num_cliprects || exec->cliprects_ptr)
 			return -EINVAL;
 	}
@@ -1938,7 +2242,8 @@ static int i915_reset_gen7_sol_offsets(struct i915_request *rq)
 }
 
 static struct i915_vma *
-shadow_batch_pin(struct drm_i915_gem_object *obj,
+shadow_batch_pin(struct i915_execbuffer *eb,
+		 struct drm_i915_gem_object *obj,
 		 struct i915_address_space *vm,
 		 unsigned int flags)
 {
@@ -1949,7 +2254,7 @@ shadow_batch_pin(struct drm_i915_gem_object *obj,
 	if (IS_ERR(vma))
 		return vma;
 
-	err = i915_vma_pin(vma, 0, 0, flags);
+	err = i915_vma_pin_ww(vma, &eb->ww, 0, 0, flags);
 	if (err)
 		return ERR_PTR(err);
 
@@ -1962,8 +2267,8 @@ struct eb_parse_work {
 	struct i915_vma *batch;
 	struct i915_vma *shadow;
 	struct i915_vma *trampoline;
-	unsigned int batch_offset;
-	unsigned int batch_length;
+	unsigned long batch_offset;
+	unsigned long batch_length;
 };
 
 static int __eb_parse(struct dma_fence_work *work)
@@ -2001,7 +2306,7 @@ __parser_mark_active(struct i915_vma *vma,
 {
 	struct intel_gt_buffer_pool_node *node = vma->private;
 
-	return i915_active_ref(&node->active, tl, fence);
+	return i915_active_ref(&node->active, tl->fence_context, fence);
 }
 
 static int
@@ -2033,6 +2338,9 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb,
 	struct eb_parse_work *pw;
 	int err;
 
+	GEM_BUG_ON(overflows_type(eb->batch_start_offset, pw->batch_offset));
+	GEM_BUG_ON(overflows_type(eb->batch_len, pw->batch_length));
+
 	pw = kzalloc(sizeof(*pw), GFP_KERNEL);
 	if (!pw)
 		return -ENOMEM;
@@ -2065,36 +2373,26 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb,
 	if (err)
 		goto err_commit;
 
-	err = dma_resv_lock_interruptible(pw->batch->resv, NULL);
-	if (err)
-		goto err_commit;
-
 	err = dma_resv_reserve_shared(pw->batch->resv, 1);
 	if (err)
-		goto err_commit_unlock;
+		goto err_commit;
 
 	/* Wait for all writes (and relocs) into the batch to complete */
 	err = i915_sw_fence_await_reservation(&pw->base.chain,
 					      pw->batch->resv, NULL, false,
 					      0, I915_FENCE_GFP);
 	if (err < 0)
-		goto err_commit_unlock;
+		goto err_commit;
 
 	/* Keep the batch alive and unwritten as we parse */
 	dma_resv_add_shared_fence(pw->batch->resv, &pw->base.dma);
 
-	dma_resv_unlock(pw->batch->resv);
-
 	/* Force execution to wait for completion of the parser */
-	dma_resv_lock(shadow->resv, NULL);
 	dma_resv_add_excl_fence(shadow->resv, &pw->base.dma);
-	dma_resv_unlock(shadow->resv);
 
 	dma_fence_work_commit_imm(&pw->base);
 	return 0;
 
-err_commit_unlock:
-	dma_resv_unlock(pw->batch->resv);
 err_commit:
 	i915_sw_fence_set_error_once(&pw->base.chain, err);
 	dma_fence_work_commit_imm(&pw->base);
@@ -2109,16 +2407,33 @@ err_free:
 	return err;
 }
 
+static struct i915_vma *eb_dispatch_secure(struct i915_execbuffer *eb, struct i915_vma *vma)
+{
+	/*
+	 * snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure
+	 * batch" bit. Hence we need to pin secure batches into the global gtt.
+	 * hsw should have this fixed, but bdw mucks it up again. */
+	if (eb->batch_flags & I915_DISPATCH_SECURE)
+		return i915_gem_object_ggtt_pin_ww(vma->obj, &eb->ww, NULL, 0, 0, 0);
+
+	return NULL;
+}
+
 static int eb_parse(struct i915_execbuffer *eb)
 {
 	struct drm_i915_private *i915 = eb->i915;
-	struct intel_gt_buffer_pool_node *pool;
-	struct i915_vma *shadow, *trampoline;
+	struct intel_gt_buffer_pool_node *pool = eb->batch_pool;
+	struct i915_vma *shadow, *trampoline, *batch;
 	unsigned int len;
 	int err;
 
-	if (!eb_use_cmdparser(eb))
-		return 0;
+	if (!eb_use_cmdparser(eb)) {
+		batch = eb_dispatch_secure(eb, eb->batch->vma);
+		if (IS_ERR(batch))
+			return PTR_ERR(batch);
+
+		goto secure_batch;
+	}
 
 	len = eb->batch_len;
 	if (!CMDPARSER_USES_GGTT(eb->i915)) {
@@ -2135,11 +2450,18 @@ static int eb_parse(struct i915_execbuffer *eb)
 		len += I915_CMD_PARSER_TRAMPOLINE_SIZE;
 	}
 
-	pool = intel_gt_get_buffer_pool(eb->engine->gt, len);
-	if (IS_ERR(pool))
-		return PTR_ERR(pool);
+	if (!pool) {
+		pool = intel_gt_get_buffer_pool(eb->engine->gt, len);
+		if (IS_ERR(pool))
+			return PTR_ERR(pool);
+		eb->batch_pool = pool;
+	}
+
+	err = i915_gem_object_lock(pool->obj, &eb->ww);
+	if (err)
+		goto err;
 
-	shadow = shadow_batch_pin(pool->obj, eb->context->vm, PIN_USER);
+	shadow = shadow_batch_pin(eb, pool->obj, eb->context->vm, PIN_USER);
 	if (IS_ERR(shadow)) {
 		err = PTR_ERR(shadow);
 		goto err;
@@ -2151,7 +2473,7 @@ static int eb_parse(struct i915_execbuffer *eb)
 	if (CMDPARSER_USES_GGTT(eb->i915)) {
 		trampoline = shadow;
 
-		shadow = shadow_batch_pin(pool->obj,
+		shadow = shadow_batch_pin(eb, pool->obj,
 					  &eb->engine->gt->ggtt->vm,
 					  PIN_GLOBAL);
 		if (IS_ERR(shadow)) {
@@ -2164,42 +2486,43 @@ static int eb_parse(struct i915_execbuffer *eb)
 		eb->batch_flags |= I915_DISPATCH_SECURE;
 	}
 
+	batch = eb_dispatch_secure(eb, shadow);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto err_trampoline;
+	}
+
 	err = eb_parse_pipeline(eb, shadow, trampoline);
 	if (err)
-		goto err_trampoline;
+		goto err_unpin_batch;
 
-	eb->vma[eb->buffer_count].vma = i915_vma_get(shadow);
-	eb->vma[eb->buffer_count].flags = __EXEC_OBJECT_HAS_PIN;
 	eb->batch = &eb->vma[eb->buffer_count++];
-	eb->vma[eb->buffer_count].vma = NULL;
+	eb->batch->vma = i915_vma_get(shadow);
+	eb->batch->flags = __EXEC_OBJECT_HAS_PIN;
 
 	eb->trampoline = trampoline;
 	eb->batch_start_offset = 0;
 
+secure_batch:
+	if (batch) {
+		eb->batch = &eb->vma[eb->buffer_count++];
+		eb->batch->flags = __EXEC_OBJECT_HAS_PIN;
+		eb->batch->vma = i915_vma_get(batch);
+	}
 	return 0;
 
+err_unpin_batch:
+	if (batch)
+		i915_vma_unpin(batch);
 err_trampoline:
 	if (trampoline)
 		i915_vma_unpin(trampoline);
 err_shadow:
 	i915_vma_unpin(shadow);
 err:
-	intel_gt_buffer_pool_put(pool);
 	return err;
 }
 
-static void
-add_to_client(struct i915_request *rq, struct drm_file *file)
-{
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-
-	rq->file_priv = file_priv;
-
-	spin_lock(&file_priv->mm.lock);
-	list_add_tail(&rq->client_link, &file_priv->mm.request_list);
-	spin_unlock(&file_priv->mm.lock);
-}
-
 static int eb_submit(struct i915_execbuffer *eb, struct i915_vma *batch)
 {
 	int err;
@@ -2281,7 +2604,7 @@ static const enum intel_engine_id user_ring_map[] = {
 	[I915_EXEC_VEBOX]	= VECS0
 };
 
-static struct i915_request *eb_throttle(struct intel_context *ce)
+static struct i915_request *eb_throttle(struct i915_execbuffer *eb, struct intel_context *ce)
 {
 	struct intel_ring *ring = ce->ring;
 	struct intel_timeline *tl = ce->timeline;
@@ -2315,31 +2638,26 @@ static struct i915_request *eb_throttle(struct intel_context *ce)
 	return i915_request_get(rq);
 }
 
-static int __eb_pin_engine(struct i915_execbuffer *eb, struct intel_context *ce)
+static struct i915_request *eb_pin_engine(struct i915_execbuffer *eb, bool throttle)
 {
+	struct intel_context *ce = eb->context;
 	struct intel_timeline *tl;
-	struct i915_request *rq;
+	struct i915_request *rq = NULL;
 	int err;
 
-	/*
-	 * ABI: Before userspace accesses the GPU (e.g. execbuffer), report
-	 * EIO if the GPU is already wedged.
-	 */
-	err = intel_gt_terminally_wedged(ce->engine->gt);
-	if (err)
-		return err;
+	GEM_BUG_ON(eb->args->flags & __EXEC_ENGINE_PINNED);
 
 	if (unlikely(intel_context_is_banned(ce)))
-		return -EIO;
+		return ERR_PTR(-EIO);
 
 	/*
 	 * Pinning the contexts may generate requests in order to acquire
 	 * GGTT space, so do this first before we reserve a seqno for
 	 * ourselves.
 	 */
-	err = intel_context_pin(ce);
+	err = intel_context_pin_ww(ce, &eb->ww);
 	if (err)
-		return err;
+		return ERR_PTR(err);
 
 	/*
 	 * Take a local wakeref for preparing to dispatch the execbuf as
@@ -2351,45 +2669,17 @@ static int __eb_pin_engine(struct i915_execbuffer *eb, struct intel_context *ce)
 	 */
 	tl = intel_context_timeline_lock(ce);
 	if (IS_ERR(tl)) {
-		err = PTR_ERR(tl);
-		goto err_unpin;
+		intel_context_unpin(ce);
+		return ERR_CAST(tl);
 	}
 
 	intel_context_enter(ce);
-	rq = eb_throttle(ce);
-
+	if (throttle)
+		rq = eb_throttle(eb, ce);
 	intel_context_timeline_unlock(tl);
 
-	if (rq) {
-		bool nonblock = eb->file->filp->f_flags & O_NONBLOCK;
-		long timeout;
-
-		timeout = MAX_SCHEDULE_TIMEOUT;
-		if (nonblock)
-			timeout = 0;
-
-		timeout = i915_request_wait(rq,
-					    I915_WAIT_INTERRUPTIBLE,
-					    timeout);
-		i915_request_put(rq);
-
-		if (timeout < 0) {
-			err = nonblock ? -EWOULDBLOCK : timeout;
-			goto err_exit;
-		}
-	}
-
-	eb->engine = ce->engine;
-	eb->context = ce;
-	return 0;
-
-err_exit:
-	mutex_lock(&tl->mutex);
-	intel_context_exit(ce);
-	intel_context_timeline_unlock(tl);
-err_unpin:
-	intel_context_unpin(ce);
-	return err;
+	eb->args->flags |= __EXEC_ENGINE_PINNED;
+	return rq;
 }
 
 static void eb_unpin_engine(struct i915_execbuffer *eb)
@@ -2397,6 +2687,11 @@ static void eb_unpin_engine(struct i915_execbuffer *eb)
 	struct intel_context *ce = eb->context;
 	struct intel_timeline *tl = ce->timeline;
 
+	if (!(eb->args->flags & __EXEC_ENGINE_PINNED))
+		return;
+
+	eb->args->flags &= ~__EXEC_ENGINE_PINNED;
+
 	mutex_lock(&tl->mutex);
 	intel_context_exit(ce);
 	mutex_unlock(&tl->mutex);
@@ -2405,11 +2700,10 @@ static void eb_unpin_engine(struct i915_execbuffer *eb)
 }
 
 static unsigned int
-eb_select_legacy_ring(struct i915_execbuffer *eb,
-		      struct drm_file *file,
-		      struct drm_i915_gem_execbuffer2 *args)
+eb_select_legacy_ring(struct i915_execbuffer *eb)
 {
 	struct drm_i915_private *i915 = eb->i915;
+	struct drm_i915_gem_execbuffer2 *args = eb->args;
 	unsigned int user_ring_id = args->flags & I915_EXEC_RING_MASK;
 
 	if (user_ring_id != I915_EXEC_BSD &&
@@ -2424,7 +2718,7 @@ eb_select_legacy_ring(struct i915_execbuffer *eb,
 		unsigned int bsd_idx = args->flags & I915_EXEC_BSD_MASK;
 
 		if (bsd_idx == I915_EXEC_BSD_DEFAULT) {
-			bsd_idx = gen8_dispatch_bsd_engine(i915, file);
+			bsd_idx = gen8_dispatch_bsd_engine(i915, eb->file);
 		} else if (bsd_idx >= I915_EXEC_BSD_RING1 &&
 			   bsd_idx <= I915_EXEC_BSD_RING2) {
 			bsd_idx >>= I915_EXEC_BSD_SHIFT;
@@ -2449,131 +2743,297 @@ eb_select_legacy_ring(struct i915_execbuffer *eb,
 }
 
 static int
-eb_pin_engine(struct i915_execbuffer *eb,
-	      struct drm_file *file,
-	      struct drm_i915_gem_execbuffer2 *args)
+eb_select_engine(struct i915_execbuffer *eb)
 {
 	struct intel_context *ce;
 	unsigned int idx;
 	int err;
 
 	if (i915_gem_context_user_engines(eb->gem_context))
-		idx = args->flags & I915_EXEC_RING_MASK;
+		idx = eb->args->flags & I915_EXEC_RING_MASK;
 	else
-		idx = eb_select_legacy_ring(eb, file, args);
+		idx = eb_select_legacy_ring(eb);
 
 	ce = i915_gem_context_get_engine(eb->gem_context, idx);
 	if (IS_ERR(ce))
 		return PTR_ERR(ce);
 
-	err = __eb_pin_engine(eb, ce);
-	intel_context_put(ce);
+	intel_gt_pm_get(ce->engine->gt);
+
+	if (!test_bit(CONTEXT_ALLOC_BIT, &ce->flags)) {
+		err = intel_context_alloc_state(ce);
+		if (err)
+			goto err;
+	}
+
+	/*
+	 * ABI: Before userspace accesses the GPU (e.g. execbuffer), report
+	 * EIO if the GPU is already wedged.
+	 */
+	err = intel_gt_terminally_wedged(ce->engine->gt);
+	if (err)
+		goto err;
+
+	eb->context = ce;
+	eb->engine = ce->engine;
 
+	/*
+	 * Make sure engine pool stays alive even if we call intel_context_put
+	 * during ww handling. The pool is destroyed when last pm reference
+	 * is dropped, which breaks our -EDEADLK handling.
+	 */
 	return err;
+
+err:
+	intel_gt_pm_put(ce->engine->gt);
+	intel_context_put(ce);
+	return err;
+}
+
+static void
+eb_put_engine(struct i915_execbuffer *eb)
+{
+	intel_gt_pm_put(eb->engine->gt);
+	intel_context_put(eb->context);
 }
 
 static void
-__free_fence_array(struct drm_syncobj **fences, unsigned int n)
+__free_fence_array(struct eb_fence *fences, unsigned int n)
 {
-	while (n--)
-		drm_syncobj_put(ptr_mask_bits(fences[n], 2));
+	while (n--) {
+		drm_syncobj_put(ptr_mask_bits(fences[n].syncobj, 2));
+		dma_fence_put(fences[n].dma_fence);
+		kfree(fences[n].chain_fence);
+	}
 	kvfree(fences);
 }
 
-static struct drm_syncobj **
-get_fence_array(struct drm_i915_gem_execbuffer2 *args,
-		struct drm_file *file)
+static int
+add_timeline_fence_array(struct i915_execbuffer *eb,
+			 const struct drm_i915_gem_execbuffer_ext_timeline_fences *timeline_fences)
 {
-	const unsigned long nfences = args->num_cliprects;
-	struct drm_i915_gem_exec_fence __user *user;
-	struct drm_syncobj **fences;
-	unsigned long n;
-	int err;
+	struct drm_i915_gem_exec_fence __user *user_fences;
+	u64 __user *user_values;
+	struct eb_fence *f;
+	u64 nfences;
+	int err = 0;
 
-	if (!(args->flags & I915_EXEC_FENCE_ARRAY))
-		return NULL;
+	nfences = timeline_fences->fence_count;
+	if (!nfences)
+		return 0;
 
 	/* Check multiplication overflow for access_ok() and kvmalloc_array() */
 	BUILD_BUG_ON(sizeof(size_t) > sizeof(unsigned long));
 	if (nfences > min_t(unsigned long,
-			    ULONG_MAX / sizeof(*user),
-			    SIZE_MAX / sizeof(*fences)))
-		return ERR_PTR(-EINVAL);
+			    ULONG_MAX / sizeof(*user_fences),
+			    SIZE_MAX / sizeof(*f)) - eb->num_fences)
+		return -EINVAL;
 
-	user = u64_to_user_ptr(args->cliprects_ptr);
-	if (!access_ok(user, nfences * sizeof(*user)))
-		return ERR_PTR(-EFAULT);
+	user_fences = u64_to_user_ptr(timeline_fences->handles_ptr);
+	if (!access_ok(user_fences, nfences * sizeof(*user_fences)))
+		return -EFAULT;
 
-	fences = kvmalloc_array(nfences, sizeof(*fences),
-				__GFP_NOWARN | GFP_KERNEL);
-	if (!fences)
-		return ERR_PTR(-ENOMEM);
+	user_values = u64_to_user_ptr(timeline_fences->values_ptr);
+	if (!access_ok(user_values, nfences * sizeof(*user_values)))
+		return -EFAULT;
+
+	f = krealloc(eb->fences,
+		     (eb->num_fences + nfences) * sizeof(*f),
+		     __GFP_NOWARN | GFP_KERNEL);
+	if (!f)
+		return -ENOMEM;
 
-	for (n = 0; n < nfences; n++) {
-		struct drm_i915_gem_exec_fence fence;
+	eb->fences = f;
+	f += eb->num_fences;
+
+	BUILD_BUG_ON(~(ARCH_KMALLOC_MINALIGN - 1) &
+		     ~__I915_EXEC_FENCE_UNKNOWN_FLAGS);
+
+	while (nfences--) {
+		struct drm_i915_gem_exec_fence user_fence;
 		struct drm_syncobj *syncobj;
+		struct dma_fence *fence = NULL;
+		u64 point;
 
-		if (__copy_from_user(&fence, user++, sizeof(fence))) {
-			err = -EFAULT;
-			goto err;
+		if (__copy_from_user(&user_fence,
+				     user_fences++,
+				     sizeof(user_fence)))
+			return -EFAULT;
+
+		if (user_fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS)
+			return -EINVAL;
+
+		if (__get_user(point, user_values++))
+			return -EFAULT;
+
+		syncobj = drm_syncobj_find(eb->file, user_fence.handle);
+		if (!syncobj) {
+			DRM_DEBUG("Invalid syncobj handle provided\n");
+			return -ENOENT;
 		}
 
-		if (fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS) {
-			err = -EINVAL;
-			goto err;
+		fence = drm_syncobj_fence_get(syncobj);
+
+		if (!fence && user_fence.flags &&
+		    !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) {
+			DRM_DEBUG("Syncobj handle has no fence\n");
+			drm_syncobj_put(syncobj);
+			return -EINVAL;
 		}
 
-		syncobj = drm_syncobj_find(file, fence.handle);
+		if (fence)
+			err = dma_fence_chain_find_seqno(&fence, point);
+
+		if (err && !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) {
+			DRM_DEBUG("Syncobj handle missing requested point %llu\n", point);
+			dma_fence_put(fence);
+			drm_syncobj_put(syncobj);
+			return err;
+		}
+
+		/*
+		 * A point might have been signaled already and
+		 * garbage collected from the timeline. In this case
+		 * just ignore the point and carry on.
+		 */
+		if (!fence && !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) {
+			drm_syncobj_put(syncobj);
+			continue;
+		}
+
+		/*
+		 * For timeline syncobjs we need to preallocate chains for
+		 * later signaling.
+		 */
+		if (point != 0 && user_fence.flags & I915_EXEC_FENCE_SIGNAL) {
+			/*
+			 * Waiting and signaling the same point (when point !=
+			 * 0) would break the timeline.
+			 */
+			if (user_fence.flags & I915_EXEC_FENCE_WAIT) {
+				DRM_DEBUG("Trying to wait & signal the same timeline point.\n");
+				dma_fence_put(fence);
+				drm_syncobj_put(syncobj);
+				return -EINVAL;
+			}
+
+			f->chain_fence =
+				kmalloc(sizeof(*f->chain_fence),
+					GFP_KERNEL);
+			if (!f->chain_fence) {
+				drm_syncobj_put(syncobj);
+				dma_fence_put(fence);
+				return -ENOMEM;
+			}
+		} else {
+			f->chain_fence = NULL;
+		}
+
+		f->syncobj = ptr_pack_bits(syncobj, user_fence.flags, 2);
+		f->dma_fence = fence;
+		f->value = point;
+		f++;
+		eb->num_fences++;
+	}
+
+	return 0;
+}
+
+static int add_fence_array(struct i915_execbuffer *eb)
+{
+	struct drm_i915_gem_execbuffer2 *args = eb->args;
+	struct drm_i915_gem_exec_fence __user *user;
+	unsigned long num_fences = args->num_cliprects;
+	struct eb_fence *f;
+
+	if (!(args->flags & I915_EXEC_FENCE_ARRAY))
+		return 0;
+
+	if (!num_fences)
+		return 0;
+
+	/* Check multiplication overflow for access_ok() and kvmalloc_array() */
+	BUILD_BUG_ON(sizeof(size_t) > sizeof(unsigned long));
+	if (num_fences > min_t(unsigned long,
+			       ULONG_MAX / sizeof(*user),
+			       SIZE_MAX / sizeof(*f) - eb->num_fences))
+		return -EINVAL;
+
+	user = u64_to_user_ptr(args->cliprects_ptr);
+	if (!access_ok(user, num_fences * sizeof(*user)))
+		return -EFAULT;
+
+	f = krealloc(eb->fences,
+		     (eb->num_fences + num_fences) * sizeof(*f),
+		     __GFP_NOWARN | GFP_KERNEL);
+	if (!f)
+		return -ENOMEM;
+
+	eb->fences = f;
+	f += eb->num_fences;
+	while (num_fences--) {
+		struct drm_i915_gem_exec_fence user_fence;
+		struct drm_syncobj *syncobj;
+		struct dma_fence *fence = NULL;
+
+		if (__copy_from_user(&user_fence, user++, sizeof(user_fence)))
+			return -EFAULT;
+
+		if (user_fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS)
+			return -EINVAL;
+
+		syncobj = drm_syncobj_find(eb->file, user_fence.handle);
 		if (!syncobj) {
 			DRM_DEBUG("Invalid syncobj handle provided\n");
-			err = -ENOENT;
-			goto err;
+			return -ENOENT;
+		}
+
+		if (user_fence.flags & I915_EXEC_FENCE_WAIT) {
+			fence = drm_syncobj_fence_get(syncobj);
+			if (!fence) {
+				DRM_DEBUG("Syncobj handle has no fence\n");
+				drm_syncobj_put(syncobj);
+				return -EINVAL;
+			}
 		}
 
 		BUILD_BUG_ON(~(ARCH_KMALLOC_MINALIGN - 1) &
 			     ~__I915_EXEC_FENCE_UNKNOWN_FLAGS);
 
-		fences[n] = ptr_pack_bits(syncobj, fence.flags, 2);
+		f->syncobj = ptr_pack_bits(syncobj, user_fence.flags, 2);
+		f->dma_fence = fence;
+		f->value = 0;
+		f->chain_fence = NULL;
+		f++;
+		eb->num_fences++;
 	}
 
-	return fences;
-
-err:
-	__free_fence_array(fences, n);
-	return ERR_PTR(err);
+	return 0;
 }
 
-static void
-put_fence_array(struct drm_i915_gem_execbuffer2 *args,
-		struct drm_syncobj **fences)
+static void put_fence_array(struct eb_fence *fences, int num_fences)
 {
 	if (fences)
-		__free_fence_array(fences, args->num_cliprects);
+		__free_fence_array(fences, num_fences);
 }
 
 static int
-await_fence_array(struct i915_execbuffer *eb,
-		  struct drm_syncobj **fences)
+await_fence_array(struct i915_execbuffer *eb)
 {
-	const unsigned int nfences = eb->args->num_cliprects;
 	unsigned int n;
 	int err;
 
-	for (n = 0; n < nfences; n++) {
+	for (n = 0; n < eb->num_fences; n++) {
 		struct drm_syncobj *syncobj;
-		struct dma_fence *fence;
 		unsigned int flags;
 
-		syncobj = ptr_unpack_bits(fences[n], &flags, 2);
-		if (!(flags & I915_EXEC_FENCE_WAIT))
-			continue;
+		syncobj = ptr_unpack_bits(eb->fences[n].syncobj, &flags, 2);
 
-		fence = drm_syncobj_fence_get(syncobj);
-		if (!fence)
-			return -EINVAL;
+		if (!eb->fences[n].dma_fence)
+			continue;
 
-		err = i915_request_await_dma_fence(eb->request, fence);
-		dma_fence_put(fence);
+		err = i915_request_await_dma_fence(eb->request,
+						   eb->fences[n].dma_fence);
 		if (err < 0)
 			return err;
 	}
@@ -2581,26 +3041,47 @@ await_fence_array(struct i915_execbuffer *eb,
 	return 0;
 }
 
-static void
-signal_fence_array(struct i915_execbuffer *eb,
-		   struct drm_syncobj **fences)
+static void signal_fence_array(const struct i915_execbuffer *eb)
 {
-	const unsigned int nfences = eb->args->num_cliprects;
 	struct dma_fence * const fence = &eb->request->fence;
 	unsigned int n;
 
-	for (n = 0; n < nfences; n++) {
+	for (n = 0; n < eb->num_fences; n++) {
 		struct drm_syncobj *syncobj;
 		unsigned int flags;
 
-		syncobj = ptr_unpack_bits(fences[n], &flags, 2);
+		syncobj = ptr_unpack_bits(eb->fences[n].syncobj, &flags, 2);
 		if (!(flags & I915_EXEC_FENCE_SIGNAL))
 			continue;
 
-		drm_syncobj_replace_fence(syncobj, fence);
+		if (eb->fences[n].chain_fence) {
+			drm_syncobj_add_point(syncobj,
+					      eb->fences[n].chain_fence,
+					      fence,
+					      eb->fences[n].value);
+			/*
+			 * The chain's ownership is transferred to the
+			 * timeline.
+			 */
+			eb->fences[n].chain_fence = NULL;
+		} else {
+			drm_syncobj_replace_fence(syncobj, fence);
+		}
 	}
 }
 
+static int
+parse_timeline_fences(struct i915_user_extension __user *ext, void *data)
+{
+	struct i915_execbuffer *eb = data;
+	struct drm_i915_gem_execbuffer_ext_timeline_fences timeline_fences;
+
+	if (copy_from_user(&timeline_fences, ext, sizeof(timeline_fences)))
+		return -EFAULT;
+
+	return add_timeline_fence_array(eb, &timeline_fences);
+}
+
 static void retire_requests(struct intel_timeline *tl, struct i915_request *end)
 {
 	struct i915_request *rq, *rn;
@@ -2642,12 +3123,37 @@ static void eb_request_add(struct i915_execbuffer *eb)
 	mutex_unlock(&tl->mutex);
 }
 
+static const i915_user_extension_fn execbuf_extensions[] = {
+	[DRM_I915_GEM_EXECBUFFER_EXT_TIMELINE_FENCES] = parse_timeline_fences,
+};
+
+static int
+parse_execbuf2_extensions(struct drm_i915_gem_execbuffer2 *args,
+			  struct i915_execbuffer *eb)
+{
+	if (!(args->flags & I915_EXEC_USE_EXTENSIONS))
+		return 0;
+
+	/* The execbuf2 extension mechanism reuses cliprects_ptr. So we cannot
+	 * have another flag also using it at the same time.
+	 */
+	if (eb->args->flags & I915_EXEC_FENCE_ARRAY)
+		return -EINVAL;
+
+	if (args->num_cliprects != 0)
+		return -EINVAL;
+
+	return i915_user_extensions(u64_to_user_ptr(args->cliprects_ptr),
+				    execbuf_extensions,
+				    ARRAY_SIZE(execbuf_extensions),
+				    eb);
+}
+
 static int
 i915_gem_do_execbuffer(struct drm_device *dev,
 		       struct drm_file *file,
 		       struct drm_i915_gem_execbuffer2 *args,
-		       struct drm_i915_gem_exec_object2 *exec,
-		       struct drm_syncobj **fences)
+		       struct drm_i915_gem_exec_object2 *exec)
 {
 	struct drm_i915_private *i915 = to_i915(dev);
 	struct i915_execbuffer eb;
@@ -2668,6 +3174,10 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 		args->flags |= __EXEC_HAS_RELOC;
 
 	eb.exec = exec;
+	eb.vma = (struct eb_vma *)(exec + args->buffer_count + 1);
+	eb.vma[0].vma = NULL;
+	eb.reloc_pool = eb.batch_pool = NULL;
+	eb.reloc_context = NULL;
 
 	eb.invalid_flags = __EXEC_OBJECT_UNKNOWN_FLAGS;
 	reloc_cache_init(&eb.reloc_cache, eb.i915);
@@ -2677,6 +3187,9 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	eb.batch_len = args->batch_len;
 	eb.trampoline = NULL;
 
+	eb.fences = NULL;
+	eb.num_fences = 0;
+
 	eb.batch_flags = 0;
 	if (args->flags & I915_EXEC_SECURE) {
 		if (INTEL_GEN(i915) >= 11)
@@ -2694,14 +3207,24 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		eb.batch_flags |= I915_DISPATCH_PINNED;
 
+	err = parse_execbuf2_extensions(args, &eb);
+	if (err)
+		goto err_ext;
+
+	err = add_fence_array(&eb);
+	if (err)
+		goto err_ext;
+
 #define IN_FENCES (I915_EXEC_FENCE_IN | I915_EXEC_FENCE_SUBMIT)
 	if (args->flags & IN_FENCES) {
 		if ((args->flags & IN_FENCES) == IN_FENCES)
 			return -EINVAL;
 
 		in_fence = sync_file_get_fence(lower_32_bits(args->rsvd2));
-		if (!in_fence)
-			return -EINVAL;
+		if (!in_fence) {
+			err = -EINVAL;
+			goto err_ext;
+		}
 	}
 #undef IN_FENCES
 
@@ -2723,11 +3246,19 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	if (unlikely(err))
 		goto err_destroy;
 
-	err = eb_pin_engine(&eb, file, args);
+	err = eb_select_engine(&eb);
 	if (unlikely(err))
 		goto err_context;
 
-	err = eb_relocate(&eb);
+	err = eb_lookup_vmas(&eb);
+	if (err) {
+		eb_release_vmas(&eb, true);
+		goto err_engine;
+	}
+
+	i915_gem_ww_ctx_init(&eb.ww, true);
+
+	err = eb_relocate_parse(&eb);
 	if (err) {
 		/*
 		 * If the user expects the execobject.offset and
@@ -2740,54 +3271,9 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 		goto err_vma;
 	}
 
-	if (unlikely(eb.batch->flags & EXEC_OBJECT_WRITE)) {
-		drm_dbg(&i915->drm,
-			"Attempting to use self-modifying batch buffer\n");
-		err = -EINVAL;
-		goto err_vma;
-	}
-
-	if (range_overflows_t(u64,
-			      eb.batch_start_offset, eb.batch_len,
-			      eb.batch->vma->size)) {
-		drm_dbg(&i915->drm, "Attempting to use out-of-bounds batch\n");
-		err = -EINVAL;
-		goto err_vma;
-	}
-
-	if (eb.batch_len == 0)
-		eb.batch_len = eb.batch->vma->size - eb.batch_start_offset;
-
-	err = eb_parse(&eb);
-	if (err)
-		goto err_vma;
+	ww_acquire_done(&eb.ww.ctx);
 
-	/*
-	 * snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure
-	 * batch" bit. Hence we need to pin secure batches into the global gtt.
-	 * hsw should have this fixed, but bdw mucks it up again. */
 	batch = eb.batch->vma;
-	if (eb.batch_flags & I915_DISPATCH_SECURE) {
-		struct i915_vma *vma;
-
-		/*
-		 * So on first glance it looks freaky that we pin the batch here
-		 * outside of the reservation loop. But:
-		 * - The batch is already pinned into the relevant ppgtt, so we
-		 *   already have the backing storage fully allocated.
-		 * - No other BO uses the global gtt (well contexts, but meh),
-		 *   so we don't really have issues with multiple objects not
-		 *   fitting due to fragmentation.
-		 * So this is actually safe.
-		 */
-		vma = i915_gem_object_ggtt_pin(batch->obj, NULL, 0, 0, 0);
-		if (IS_ERR(vma)) {
-			err = PTR_ERR(vma);
-			goto err_parse;
-		}
-
-		batch = vma;
-	}
 
 	/* All GPU relocation batches must be submitted prior to the user rq */
 	GEM_BUG_ON(eb.reloc_cache.rq);
@@ -2796,7 +3282,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	eb.request = i915_request_create(eb.context);
 	if (IS_ERR(eb.request)) {
 		err = PTR_ERR(eb.request);
-		goto err_batch_unpin;
+		goto err_vma;
 	}
 
 	if (in_fence) {
@@ -2811,8 +3297,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 			goto err_request;
 	}
 
-	if (fences) {
-		err = await_fence_array(&eb, fences);
+	if (eb.fences) {
+		err = await_fence_array(&eb);
 		if (err)
 			goto err_request;
 	}
@@ -2833,18 +3319,17 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	 * to explicitly hold another reference here.
 	 */
 	eb.request->batch = batch;
-	if (batch->private)
-		intel_gt_buffer_pool_mark_active(batch->private, eb.request);
+	if (eb.batch_pool)
+		intel_gt_buffer_pool_mark_active(eb.batch_pool, eb.request);
 
 	trace_i915_request_queue(eb.request, eb.batch_flags);
 	err = eb_submit(&eb, batch);
 err_request:
-	add_to_client(eb.request, file);
 	i915_request_get(eb.request);
 	eb_request_add(&eb);
 
-	if (fences)
-		signal_fence_array(&eb, fences);
+	if (eb.fences)
+		signal_fence_array(&eb);
 
 	if (out_fence) {
 		if (err == 0) {
@@ -2858,16 +3343,21 @@ err_request:
 	}
 	i915_request_put(eb.request);
 
-err_batch_unpin:
-	if (eb.batch_flags & I915_DISPATCH_SECURE)
-		i915_vma_unpin(batch);
-err_parse:
-	if (batch->private)
-		intel_gt_buffer_pool_put(batch->private);
 err_vma:
+	eb_release_vmas(&eb, true);
 	if (eb.trampoline)
 		i915_vma_unpin(eb.trampoline);
-	eb_unpin_engine(&eb);
+	WARN_ON(err == -EDEADLK);
+	i915_gem_ww_ctx_fini(&eb.ww);
+
+	if (eb.batch_pool)
+		intel_gt_buffer_pool_put(eb.batch_pool);
+	if (eb.reloc_pool)
+		intel_gt_buffer_pool_put(eb.reloc_pool);
+	if (eb.reloc_context)
+		intel_context_put(eb.reloc_context);
+err_engine:
+	eb_put_engine(&eb);
 err_context:
 	i915_gem_context_put(eb.gem_context);
 err_destroy:
@@ -2877,12 +3367,14 @@ err_out_fence:
 		put_unused_fd(out_fence_fd);
 err_in_fence:
 	dma_fence_put(in_fence);
+err_ext:
+	put_fence_array(eb.fences, eb.num_fences);
 	return err;
 }
 
 static size_t eb_element_size(void)
 {
-	return sizeof(struct drm_i915_gem_exec_object2);
+	return sizeof(struct drm_i915_gem_exec_object2) + sizeof(struct eb_vma);
 }
 
 static bool check_buffer_count(size_t count)
@@ -2938,7 +3430,9 @@ i915_gem_execbuffer_ioctl(struct drm_device *dev, void *data,
 	/* Copy in the exec list from userland */
 	exec_list = kvmalloc_array(count, sizeof(*exec_list),
 				   __GFP_NOWARN | GFP_KERNEL);
-	exec2_list = kvmalloc_array(count, eb_element_size(),
+
+	/* Allocate extra slots for use by the command parser */
+	exec2_list = kvmalloc_array(count + 2, eb_element_size(),
 				    __GFP_NOWARN | GFP_KERNEL);
 	if (exec_list == NULL || exec2_list == NULL) {
 		drm_dbg(&i915->drm,
@@ -2971,7 +3465,7 @@ i915_gem_execbuffer_ioctl(struct drm_device *dev, void *data,
 			exec2_list[i].flags = 0;
 	}
 
-	err = i915_gem_do_execbuffer(dev, file, &exec2, exec2_list, NULL);
+	err = i915_gem_do_execbuffer(dev, file, &exec2, exec2_list);
 	if (exec2.flags & __EXEC_HAS_RELOC) {
 		struct drm_i915_gem_exec_object __user *user_exec_list =
 			u64_to_user_ptr(args->buffers_ptr);
@@ -3003,7 +3497,6 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data,
 	struct drm_i915_private *i915 = to_i915(dev);
 	struct drm_i915_gem_execbuffer2 *args = data;
 	struct drm_i915_gem_exec_object2 *exec2_list;
-	struct drm_syncobj **fences = NULL;
 	const size_t count = args->buffer_count;
 	int err;
 
@@ -3016,7 +3509,8 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data,
 	if (err)
 		return err;
 
-	exec2_list = kvmalloc_array(count, eb_element_size(),
+	/* Allocate extra slots for use by the command parser */
+	exec2_list = kvmalloc_array(count + 2, eb_element_size(),
 				    __GFP_NOWARN | GFP_KERNEL);
 	if (exec2_list == NULL) {
 		drm_dbg(&i915->drm, "Failed to allocate exec list for %zd buffers\n",
@@ -3031,15 +3525,7 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data,
 		return -EFAULT;
 	}
 
-	if (args->flags & I915_EXEC_FENCE_ARRAY) {
-		fences = get_fence_array(args, file);
-		if (IS_ERR(fences)) {
-			kvfree(exec2_list);
-			return PTR_ERR(fences);
-		}
-	}
-
-	err = i915_gem_do_execbuffer(dev, file, args, exec2_list, fences);
+	err = i915_gem_do_execbuffer(dev, file, args, exec2_list);
 
 	/*
 	 * Now that we have begun execution of the batchbuffer, we ignore
@@ -3080,7 +3566,6 @@ end:;
 	}
 
 	args->flags &= ~__I915_EXEC_UNKNOWN_FLAGS;
-	put_fence_array(args, fences);
 	kvfree(exec2_list);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index 753f82d87a31..3d69e51f3e4d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -283,37 +283,46 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf)
 	struct intel_runtime_pm *rpm = &i915->runtime_pm;
 	struct i915_ggtt *ggtt = &i915->ggtt;
 	bool write = area->vm_flags & VM_WRITE;
+	struct i915_gem_ww_ctx ww;
 	intel_wakeref_t wakeref;
 	struct i915_vma *vma;
 	pgoff_t page_offset;
 	int srcu;
 	int ret;
 
-	/* Sanity check that we allow writing into this object */
-	if (i915_gem_object_is_readonly(obj) && write)
-		return VM_FAULT_SIGBUS;
-
 	/* We don't use vmf->pgoff since that has the fake offset */
 	page_offset = (vmf->address - area->vm_start) >> PAGE_SHIFT;
 
 	trace_i915_gem_object_fault(obj, page_offset, true, write);
 
-	ret = i915_gem_object_pin_pages(obj);
+	wakeref = intel_runtime_pm_get(rpm);
+
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	ret = i915_gem_object_lock(obj, &ww);
 	if (ret)
-		goto err;
+		goto err_rpm;
 
-	wakeref = intel_runtime_pm_get(rpm);
+	/* Sanity check that we allow writing into this object */
+	if (i915_gem_object_is_readonly(obj) && write) {
+		ret = -EFAULT;
+		goto err_rpm;
+	}
 
-	ret = intel_gt_reset_trylock(ggtt->vm.gt, &srcu);
+	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
 		goto err_rpm;
 
+	ret = intel_gt_reset_trylock(ggtt->vm.gt, &srcu);
+	if (ret)
+		goto err_pages;
+
 	/* Now pin it into the GTT as needed */
-	vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
-				       PIN_MAPPABLE |
-				       PIN_NONBLOCK /* NOWARN */ |
-				       PIN_NOEVICT);
-	if (IS_ERR(vma)) {
+	vma = i915_gem_object_ggtt_pin_ww(obj, &ww, NULL, 0, 0,
+					  PIN_MAPPABLE |
+					  PIN_NONBLOCK /* NOWARN */ |
+					  PIN_NOEVICT);
+	if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK)) {
 		/* Use a partial view if it is bigger than available space */
 		struct i915_ggtt_view view =
 			compute_partial_view(obj, page_offset, MIN_CHUNK_PAGES);
@@ -328,11 +337,11 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf)
 		 * all hope that the hardware is able to track future writes.
 		 */
 
-		vma = i915_gem_object_ggtt_pin(obj, &view, 0, 0, flags);
-		if (IS_ERR(vma)) {
+		vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags);
+		if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK)) {
 			flags = PIN_MAPPABLE;
 			view.type = I915_GGTT_VIEW_PARTIAL;
-			vma = i915_gem_object_ggtt_pin(obj, &view, 0, 0, flags);
+			vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags);
 		}
 
 		/* The entire mappable GGTT is pinned? Unexpected! */
@@ -389,10 +398,16 @@ err_unpin:
 	__i915_vma_unpin(vma);
 err_reset:
 	intel_gt_reset_unlock(ggtt->vm.gt, srcu);
+err_pages:
+	i915_gem_object_unpin_pages(obj);
 err_rpm:
+	if (ret == -EDEADLK) {
+		ret = i915_gem_ww_ctx_backoff(&ww);
+		if (!ret)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
 	intel_runtime_pm_put(rpm, wakeref);
-	i915_gem_object_unpin_pages(obj);
-err:
 	return i915_error_to_vmf_fault(ret);
 }
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 9cf4ad78ece6..d46db8d8f38e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -110,20 +110,44 @@ i915_gem_object_put(struct drm_i915_gem_object *obj)
 
 #define assert_object_held(obj) dma_resv_assert_held((obj)->base.resv)
 
-static inline void i915_gem_object_lock(struct drm_i915_gem_object *obj)
+static inline int __i915_gem_object_lock(struct drm_i915_gem_object *obj,
+					 struct i915_gem_ww_ctx *ww,
+					 bool intr)
 {
-	dma_resv_lock(obj->base.resv, NULL);
+	int ret;
+
+	if (intr)
+		ret = dma_resv_lock_interruptible(obj->base.resv, ww ? &ww->ctx : NULL);
+	else
+		ret = dma_resv_lock(obj->base.resv, ww ? &ww->ctx : NULL);
+
+	if (!ret && ww)
+		list_add_tail(&obj->obj_link, &ww->obj_list);
+	if (ret == -EALREADY)
+		ret = 0;
+
+	if (ret == -EDEADLK)
+		ww->contended = obj;
+
+	return ret;
 }
 
-static inline bool i915_gem_object_trylock(struct drm_i915_gem_object *obj)
+static inline int i915_gem_object_lock(struct drm_i915_gem_object *obj,
+				       struct i915_gem_ww_ctx *ww)
 {
-	return dma_resv_trylock(obj->base.resv);
+	return __i915_gem_object_lock(obj, ww, ww && ww->intr);
 }
 
-static inline int
-i915_gem_object_lock_interruptible(struct drm_i915_gem_object *obj)
+static inline int i915_gem_object_lock_interruptible(struct drm_i915_gem_object *obj,
+						     struct i915_gem_ww_ctx *ww)
 {
-	return dma_resv_lock_interruptible(obj->base.resv, NULL);
+	WARN_ON(ww && !ww->intr);
+	return __i915_gem_object_lock(obj, ww, true);
+}
+
+static inline bool i915_gem_object_trylock(struct drm_i915_gem_object *obj)
+{
+	return dma_resv_trylock(obj->base.resv);
 }
 
 static inline void i915_gem_object_unlock(struct drm_i915_gem_object *obj)
@@ -412,7 +436,6 @@ static inline void
 i915_gem_object_finish_access(struct drm_i915_gem_object *obj)
 {
 	i915_gem_object_unpin_pages(obj);
-	i915_gem_object_unlock(obj);
 }
 
 static inline struct intel_engine_cs *
@@ -435,6 +458,7 @@ i915_gem_object_last_write_engine(struct drm_i915_gem_object *obj)
 void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 					 unsigned int cache_level);
 void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj);
+void i915_gem_object_flush_if_display_locked(struct drm_i915_gem_object *obj);
 
 int __must_check
 i915_gem_object_set_to_wc_domain(struct drm_i915_gem_object *obj, bool write);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
index bfdb32d46877..aee7ad3cc3c6 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
@@ -14,6 +14,7 @@
 
 struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 					 struct i915_vma *vma,
+					 struct i915_gem_ww_ctx *ww,
 					 u32 value)
 {
 	struct drm_i915_private *i915 = ce->vm->i915;
@@ -39,10 +40,24 @@ struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 		goto out_pm;
 	}
 
+	err = i915_gem_object_lock(pool->obj, ww);
+	if (err)
+		goto out_put;
+
+	batch = i915_vma_instance(pool->obj, ce->vm, NULL);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_put;
+	}
+
+	err = i915_vma_pin_ww(batch, ww, 0, 0, PIN_USER);
+	if (unlikely(err))
+		goto out_put;
+
 	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
-		goto out_put;
+		goto out_unpin;
 	}
 
 	rem = vma->size;
@@ -84,19 +99,11 @@ struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 
 	intel_gt_chipset_flush(ce->vm->gt);
 
-	batch = i915_vma_instance(pool->obj, ce->vm, NULL);
-	if (IS_ERR(batch)) {
-		err = PTR_ERR(batch);
-		goto out_put;
-	}
-
-	err = i915_vma_pin(batch, 0, 0, PIN_USER);
-	if (unlikely(err))
-		goto out_put;
-
 	batch->private = pool;
 	return batch;
 
+out_unpin:
+	i915_vma_unpin(batch);
 out_put:
 	intel_gt_buffer_pool_put(pool);
 out_pm:
@@ -108,11 +115,9 @@ int intel_emit_vma_mark_active(struct i915_vma *vma, struct i915_request *rq)
 {
 	int err;
 
-	i915_vma_lock(vma);
 	err = i915_request_await_object(rq, vma->obj, false);
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, 0);
-	i915_vma_unlock(vma);
 	if (unlikely(err))
 		return err;
 
@@ -141,6 +146,7 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 			     struct intel_context *ce,
 			     u32 value)
 {
+	struct i915_gem_ww_ctx ww;
 	struct i915_request *rq;
 	struct i915_vma *batch;
 	struct i915_vma *vma;
@@ -150,17 +156,28 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
-	err = i915_vma_pin(vma, 0, 0, PIN_USER);
-	if (unlikely(err))
-		return err;
+	i915_gem_ww_ctx_init(&ww, true);
+	intel_engine_pm_get(ce->engine);
+retry:
+	err = i915_gem_object_lock(obj, &ww);
+	if (err)
+		goto out;
 
-	batch = intel_emit_vma_fill_blt(ce, vma, value);
+	err = intel_context_pin_ww(ce, &ww);
+	if (err)
+		goto out;
+
+	err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_USER);
+	if (err)
+		goto out_ctx;
+
+	batch = intel_emit_vma_fill_blt(ce, vma, &ww, value);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
-		goto out_unpin;
+		goto out_vma;
 	}
 
-	rq = intel_context_create_request(ce);
+	rq = i915_request_create(ce);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto out_batch;
@@ -170,11 +187,9 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 	if (unlikely(err))
 		goto out_request;
 
-	i915_vma_lock(vma);
 	err = move_obj_to_gpu(vma->obj, rq, true);
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
 	if (unlikely(err))
 		goto out_request;
 
@@ -193,8 +208,18 @@ out_request:
 	i915_request_add(rq);
 out_batch:
 	intel_emit_vma_release(ce, batch);
-out_unpin:
+out_vma:
 	i915_vma_unpin(vma);
+out_ctx:
+	intel_context_unpin(ce);
+out:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	intel_engine_pm_put(ce->engine);
 	return err;
 }
 
@@ -210,6 +235,7 @@ static bool wa_1209644611_applies(struct drm_i915_private *i915, u32 size)
 }
 
 struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
+					 struct i915_gem_ww_ctx *ww,
 					 struct i915_vma *src,
 					 struct i915_vma *dst)
 {
@@ -236,10 +262,24 @@ struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
 		goto out_pm;
 	}
 
+	err = i915_gem_object_lock(pool->obj, ww);
+	if (err)
+		goto out_put;
+
+	batch = i915_vma_instance(pool->obj, ce->vm, NULL);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_put;
+	}
+
+	err = i915_vma_pin_ww(batch, ww, 0, 0, PIN_USER);
+	if (unlikely(err))
+		goto out_put;
+
 	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
-		goto out_put;
+		goto out_unpin;
 	}
 
 	rem = src->size;
@@ -296,20 +336,11 @@ struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
 	i915_gem_object_unpin_map(pool->obj);
 
 	intel_gt_chipset_flush(ce->vm->gt);
-
-	batch = i915_vma_instance(pool->obj, ce->vm, NULL);
-	if (IS_ERR(batch)) {
-		err = PTR_ERR(batch);
-		goto out_put;
-	}
-
-	err = i915_vma_pin(batch, 0, 0, PIN_USER);
-	if (unlikely(err))
-		goto out_put;
-
 	batch->private = pool;
 	return batch;
 
+out_unpin:
+	i915_vma_unpin(batch);
 out_put:
 	intel_gt_buffer_pool_put(pool);
 out_pm:
@@ -321,10 +352,9 @@ int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
 			     struct drm_i915_gem_object *dst,
 			     struct intel_context *ce)
 {
-	struct drm_gem_object *objs[] = { &src->base, &dst->base };
 	struct i915_address_space *vm = ce->vm;
 	struct i915_vma *vma[2], *batch;
-	struct ww_acquire_ctx acquire;
+	struct i915_gem_ww_ctx ww;
 	struct i915_request *rq;
 	int err, i;
 
@@ -332,25 +362,36 @@ int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
 	if (IS_ERR(vma[0]))
 		return PTR_ERR(vma[0]);
 
-	err = i915_vma_pin(vma[0], 0, 0, PIN_USER);
-	if (unlikely(err))
-		return err;
-
 	vma[1] = i915_vma_instance(dst, vm, NULL);
 	if (IS_ERR(vma[1]))
-		goto out_unpin_src;
+		return PTR_ERR(vma[1]);
 
-	err = i915_vma_pin(vma[1], 0, 0, PIN_USER);
+	i915_gem_ww_ctx_init(&ww, true);
+	intel_engine_pm_get(ce->engine);
+retry:
+	err = i915_gem_object_lock(src, &ww);
+	if (!err)
+		err = i915_gem_object_lock(dst, &ww);
+	if (!err)
+		err = intel_context_pin_ww(ce, &ww);
+	if (err)
+		goto out;
+
+	err = i915_vma_pin_ww(vma[0], &ww, 0, 0, PIN_USER);
+	if (err)
+		goto out_ctx;
+
+	err = i915_vma_pin_ww(vma[1], &ww, 0, 0, PIN_USER);
 	if (unlikely(err))
 		goto out_unpin_src;
 
-	batch = intel_emit_vma_copy_blt(ce, vma[0], vma[1]);
+	batch = intel_emit_vma_copy_blt(ce, &ww, vma[0], vma[1]);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
 		goto out_unpin_dst;
 	}
 
-	rq = intel_context_create_request(ce);
+	rq = i915_request_create(ce);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto out_batch;
@@ -360,14 +401,10 @@ int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
 	if (unlikely(err))
 		goto out_request;
 
-	err = drm_gem_lock_reservations(objs, ARRAY_SIZE(objs), &acquire);
-	if (unlikely(err))
-		goto out_request;
-
 	for (i = 0; i < ARRAY_SIZE(vma); i++) {
 		err = move_obj_to_gpu(vma[i]->obj, rq, i);
 		if (unlikely(err))
-			goto out_unlock;
+			goto out_request;
 	}
 
 	for (i = 0; i < ARRAY_SIZE(vma); i++) {
@@ -375,20 +412,19 @@ int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
 
 		err = i915_vma_move_to_active(vma[i], rq, flags);
 		if (unlikely(err))
-			goto out_unlock;
+			goto out_request;
 	}
 
 	if (rq->engine->emit_init_breadcrumb) {
 		err = rq->engine->emit_init_breadcrumb(rq);
 		if (unlikely(err))
-			goto out_unlock;
+			goto out_request;
 	}
 
 	err = rq->engine->emit_bb_start(rq,
 					batch->node.start, batch->node.size,
 					0);
-out_unlock:
-	drm_gem_unlock_reservations(objs, ARRAY_SIZE(objs), &acquire);
+
 out_request:
 	if (unlikely(err))
 		i915_request_set_error_once(rq, err);
@@ -400,6 +436,16 @@ out_unpin_dst:
 	i915_vma_unpin(vma[1]);
 out_unpin_src:
 	i915_vma_unpin(vma[0]);
+out_ctx:
+	intel_context_unpin(ce);
+out:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	intel_engine_pm_put(ce->engine);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.h b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.h
index 8bcd336a90dc..2409fdcccf0e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.h
@@ -13,12 +13,15 @@
 #include "i915_vma.h"
 
 struct drm_i915_gem_object;
+struct i915_gem_ww_ctx;
 
 struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 					 struct i915_vma *vma,
+					 struct i915_gem_ww_ctx *ww,
 					 u32 value);
 
 struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
+					 struct i915_gem_ww_ctx *ww,
 					 struct i915_vma *src,
 					 struct i915_vma *dst);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 5335f799b548..b5c15557cc87 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -123,6 +123,15 @@ struct drm_i915_gem_object {
 	struct list_head lut_list;
 	spinlock_t lut_lock; /* guards lut_list */
 
+	/**
+	 * @obj_link: Link into @i915_gem_ww_ctx.obj_list
+	 *
+	 * When we lock this object through i915_gem_object_lock() with a
+	 * context, we add it to the list to ensure we can unlock everything
+	 * when i915_gem_ww_ctx_backoff() or i915_gem_ww_ctx_fini() are called.
+	 */
+	struct list_head obj_link;
+
 	/** Stolen memory for this object, instead of being backed by shmem. */
 	struct drm_mm_node *stolen;
 	union {
@@ -282,6 +291,7 @@ struct drm_i915_gem_object {
 		} userptr;
 
 		unsigned long scratch;
+		u64 encode;
 
 		void *gvt_info;
 	};
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index e8a083743e09..d6eeefab3d01 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -254,9 +254,35 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj,
 	if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC)
 		return NULL;
 
+	if (GEM_WARN_ON(type == I915_MAP_WC &&
+			!static_cpu_has(X86_FEATURE_PAT)))
+		return NULL;
+
 	/* A single page can always be kmapped */
-	if (n_pte == 1 && type == I915_MAP_WB)
-		return kmap(sg_page(sgt->sgl));
+	if (n_pte == 1 && type == I915_MAP_WB) {
+		struct page *page = sg_page(sgt->sgl);
+
+		/*
+		 * On 32b, highmem using a finite set of indirect PTE (i.e.
+		 * vmap) to provide virtual mappings of the high pages.
+		 * As these are finite, map_new_virtual() must wait for some
+		 * other kmap() to finish when it runs out. If we map a large
+		 * number of objects, there is no method for it to tell us
+		 * to release the mappings, and we deadlock.
+		 *
+		 * However, if we make an explicit vmap of the page, that
+		 * uses a larger vmalloc arena, and also has the ability
+		 * to tell us to release unwanted mappings. Most importantly,
+		 * it will fail and propagate an error instead of waiting
+		 * forever.
+		 *
+		 * So if the page is beyond the 32b boundary, make an explicit
+		 * vmap. On 64b, this check will be optimised away as we can
+		 * directly kmap any page on the system.
+		 */
+		if (!PageHighMem(page))
+			return kmap(page);
+	}
 
 	mem = stack;
 	if (n_pte > ARRAY_SIZE(stack)) {
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
index 3d215164dd5a..40d3e40500fa 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
@@ -84,7 +84,7 @@ void i915_gem_suspend_late(struct drm_i915_private *i915)
 
 			spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 
-			i915_gem_object_lock(obj);
+			i915_gem_object_lock(obj, NULL);
 			drm_WARN_ON(&i915->drm,
 			    i915_gem_object_set_to_gtt_domain(obj, false));
 			i915_gem_object_unlock(obj);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
index 540ef0551789..1929d6cf4150 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
@@ -9,6 +9,7 @@
 #include <drm/drm_file.h>
 
 #include "i915_drv.h"
+#include "i915_gem_context.h"
 #include "i915_gem_ioctls.h"
 #include "i915_gem_object.h"
 
@@ -35,9 +36,10 @@ int
 i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file)
 {
+	const unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
-	struct i915_request *request, *target = NULL;
+	struct i915_gem_context *ctx;
+	unsigned long idx;
 	long ret;
 
 	/* ABI: return -EIO if already wedged */
@@ -45,27 +47,54 @@ i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	spin_lock(&file_priv->mm.lock);
-	list_for_each_entry(request, &file_priv->mm.request_list, client_link) {
-		if (time_after_eq(request->emitted_jiffies, recent_enough))
-			break;
+	rcu_read_lock();
+	xa_for_each(&file_priv->context_xa, idx, ctx) {
+		struct i915_gem_engines_iter it;
+		struct intel_context *ce;
 
-		if (target && xchg(&target->file_priv, NULL))
-			list_del(&target->client_link);
+		if (!kref_get_unless_zero(&ctx->ref))
+			continue;
+		rcu_read_unlock();
 
-		target = request;
-	}
-	if (target)
-		i915_request_get(target);
-	spin_unlock(&file_priv->mm.lock);
+		for_each_gem_engine(ce,
+				    i915_gem_context_lock_engines(ctx),
+				    it) {
+			struct i915_request *rq, *target = NULL;
+
+			if (!ce->timeline)
+				continue;
+
+			mutex_lock(&ce->timeline->mutex);
+			list_for_each_entry_reverse(rq,
+						    &ce->timeline->requests,
+						    link) {
+				if (i915_request_completed(rq))
+					break;
 
-	if (!target)
-		return 0;
+				if (time_after(rq->emitted_jiffies,
+					       recent_enough))
+					continue;
 
-	ret = i915_request_wait(target,
-				I915_WAIT_INTERRUPTIBLE,
-				MAX_SCHEDULE_TIMEOUT);
-	i915_request_put(target);
+				target = i915_request_get(rq);
+				break;
+			}
+			mutex_unlock(&ce->timeline->mutex);
+			if (!target)
+				continue;
+
+			ret = i915_request_wait(target,
+						I915_WAIT_INTERRUPTIBLE,
+						MAX_SCHEDULE_TIMEOUT);
+			i915_request_put(target);
+			if (ret < 0)
+				break;
+		}
+		i915_gem_context_unlock_engines(ctx);
+		i915_gem_context_put(ctx);
+
+		rcu_read_lock();
+	}
+	rcu_read_unlock();
 
 	return ret < 0 ? ret : 0;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
index ff72ee2fd9cd..ffcaee74a249 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
@@ -249,7 +249,7 @@ i915_gem_object_set_tiling(struct drm_i915_gem_object *obj,
 	 * whilst executing a fenced command for an untiled object.
 	 */
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	if (i915_gem_object_is_framebuffer(obj)) {
 		i915_gem_object_unlock(obj);
 		return -EBUSY;
diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
index 8291ede6902c..1f35e71429b4 100644
--- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
@@ -393,7 +393,7 @@ static int igt_mock_exhaust_device_supported_pages(void *arg)
 	 */
 
 	for (i = 1; i < BIT(ARRAY_SIZE(page_sizes)); i++) {
-		unsigned int combination = 0;
+		unsigned int combination = SZ_4K; /* Required for ppGTT */
 
 		for (j = 0; j < ARRAY_SIZE(page_sizes); j++) {
 			if (i & BIT(j))
@@ -947,7 +947,7 @@ static int gpu_write(struct intel_context *ce,
 {
 	int err;
 
-	i915_gem_object_lock(vma->obj);
+	i915_gem_object_lock(vma->obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(vma->obj, true);
 	i915_gem_object_unlock(vma->obj);
 	if (err)
@@ -964,9 +964,10 @@ __cpu_check_shmem(struct drm_i915_gem_object *obj, u32 dword, u32 val)
 	unsigned long n;
 	int err;
 
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_prepare_read(obj, &needs_flush);
 	if (err)
-		return err;
+		goto err_unlock;
 
 	for (n = 0; n < obj->base.size >> PAGE_SHIFT; ++n) {
 		u32 *ptr = kmap_atomic(i915_gem_object_get_page(obj, n));
@@ -986,6 +987,8 @@ __cpu_check_shmem(struct drm_i915_gem_object *obj, u32 dword, u32 val)
 	}
 
 	i915_gem_object_finish_access(obj);
+err_unlock:
+	i915_gem_object_unlock(obj);
 
 	return err;
 }
@@ -1614,7 +1617,7 @@ int i915_gem_huge_page_mock_selftests(void)
 out_put:
 	i915_vm_put(&ppgtt->vm);
 out_unlock:
-	drm_dev_put(&dev_priv->drm);
+	mock_destroy_device(dev_priv);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
index 299c29e9ad86..4e36d4897ea6 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
@@ -75,7 +75,7 @@ static int __igt_client_fill(struct intel_engine_cs *engine)
 		if (err)
 			goto err_unpin;
 
-		i915_gem_object_lock(obj);
+		i915_gem_object_lock(obj, NULL);
 		err = i915_gem_object_set_to_cpu_domain(obj, false);
 		i915_gem_object_unlock(obj);
 		if (err)
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
index 87d7d8aa080f..7049a6bbc03d 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
@@ -27,9 +27,10 @@ static int cpu_set(struct context *ctx, unsigned long offset, u32 v)
 	u32 *cpu;
 	int err;
 
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_prepare_write(ctx->obj, &needs_clflush);
 	if (err)
-		return err;
+		goto out;
 
 	page = i915_gem_object_get_page(ctx->obj, offset >> PAGE_SHIFT);
 	map = kmap_atomic(page);
@@ -46,7 +47,9 @@ static int cpu_set(struct context *ctx, unsigned long offset, u32 v)
 	kunmap_atomic(map);
 	i915_gem_object_finish_access(ctx->obj);
 
-	return 0;
+out:
+	i915_gem_object_unlock(ctx->obj);
+	return err;
 }
 
 static int cpu_get(struct context *ctx, unsigned long offset, u32 *v)
@@ -57,9 +60,10 @@ static int cpu_get(struct context *ctx, unsigned long offset, u32 *v)
 	u32 *cpu;
 	int err;
 
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_prepare_read(ctx->obj, &needs_clflush);
 	if (err)
-		return err;
+		goto out;
 
 	page = i915_gem_object_get_page(ctx->obj, offset >> PAGE_SHIFT);
 	map = kmap_atomic(page);
@@ -73,7 +77,9 @@ static int cpu_get(struct context *ctx, unsigned long offset, u32 *v)
 	kunmap_atomic(map);
 	i915_gem_object_finish_access(ctx->obj);
 
-	return 0;
+out:
+	i915_gem_object_unlock(ctx->obj);
+	return err;
 }
 
 static int gtt_set(struct context *ctx, unsigned long offset, u32 v)
@@ -82,7 +88,7 @@ static int gtt_set(struct context *ctx, unsigned long offset, u32 v)
 	u32 __iomem *map;
 	int err = 0;
 
-	i915_gem_object_lock(ctx->obj);
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(ctx->obj, true);
 	i915_gem_object_unlock(ctx->obj);
 	if (err)
@@ -115,7 +121,7 @@ static int gtt_get(struct context *ctx, unsigned long offset, u32 *v)
 	u32 __iomem *map;
 	int err = 0;
 
-	i915_gem_object_lock(ctx->obj);
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(ctx->obj, false);
 	i915_gem_object_unlock(ctx->obj);
 	if (err)
@@ -147,7 +153,7 @@ static int wc_set(struct context *ctx, unsigned long offset, u32 v)
 	u32 *map;
 	int err;
 
-	i915_gem_object_lock(ctx->obj);
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_set_to_wc_domain(ctx->obj, true);
 	i915_gem_object_unlock(ctx->obj);
 	if (err)
@@ -170,7 +176,7 @@ static int wc_get(struct context *ctx, unsigned long offset, u32 *v)
 	u32 *map;
 	int err;
 
-	i915_gem_object_lock(ctx->obj);
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_set_to_wc_domain(ctx->obj, false);
 	i915_gem_object_unlock(ctx->obj);
 	if (err)
@@ -193,27 +199,27 @@ static int gpu_set(struct context *ctx, unsigned long offset, u32 v)
 	u32 *cs;
 	int err;
 
-	i915_gem_object_lock(ctx->obj);
+	i915_gem_object_lock(ctx->obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(ctx->obj, true);
-	i915_gem_object_unlock(ctx->obj);
 	if (err)
-		return err;
+		goto out_unlock;
 
 	vma = i915_gem_object_ggtt_pin(ctx->obj, NULL, 0, 0, 0);
-	if (IS_ERR(vma))
-		return PTR_ERR(vma);
+	if (IS_ERR(vma)) {
+		err = PTR_ERR(vma);
+		goto out_unlock;
+	}
 
 	rq = intel_engine_create_kernel_request(ctx->engine);
 	if (IS_ERR(rq)) {
-		i915_vma_unpin(vma);
-		return PTR_ERR(rq);
+		err = PTR_ERR(rq);
+		goto out_unpin;
 	}
 
 	cs = intel_ring_begin(rq, 4);
 	if (IS_ERR(cs)) {
-		i915_request_add(rq);
-		i915_vma_unpin(vma);
-		return PTR_ERR(cs);
+		err = PTR_ERR(cs);
+		goto out_rq;
 	}
 
 	if (INTEL_GEN(ctx->engine->i915) >= 8) {
@@ -234,14 +240,16 @@ static int gpu_set(struct context *ctx, unsigned long offset, u32 v)
 	}
 	intel_ring_advance(rq, cs);
 
-	i915_vma_lock(vma);
 	err = i915_request_await_object(rq, vma->obj, true);
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
-	i915_vma_unpin(vma);
 
+out_rq:
 	i915_request_add(rq);
+out_unpin:
+	i915_vma_unpin(vma);
+out_unlock:
+	i915_gem_object_unlock(ctx->obj);
 
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
index 7ffc3c751432..d3f87dc4eda3 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
@@ -461,9 +461,10 @@ static int cpu_fill(struct drm_i915_gem_object *obj, u32 value)
 	unsigned int n, m, need_flush;
 	int err;
 
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_prepare_write(obj, &need_flush);
 	if (err)
-		return err;
+		goto out;
 
 	for (n = 0; n < real_page_count(obj); n++) {
 		u32 *map;
@@ -479,7 +480,9 @@ static int cpu_fill(struct drm_i915_gem_object *obj, u32 value)
 	i915_gem_object_finish_access(obj);
 	obj->read_domains = I915_GEM_DOMAIN_GTT | I915_GEM_DOMAIN_CPU;
 	obj->write_domain = 0;
-	return 0;
+out:
+	i915_gem_object_unlock(obj);
+	return err;
 }
 
 static noinline int cpu_check(struct drm_i915_gem_object *obj,
@@ -488,9 +491,10 @@ static noinline int cpu_check(struct drm_i915_gem_object *obj,
 	unsigned int n, m, needs_flush;
 	int err;
 
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_prepare_read(obj, &needs_flush);
 	if (err)
-		return err;
+		goto out_unlock;
 
 	for (n = 0; n < real_page_count(obj); n++) {
 		u32 *map;
@@ -527,6 +531,8 @@ out_unmap:
 	}
 
 	i915_gem_object_finish_access(obj);
+out_unlock:
+	i915_gem_object_unlock(obj);
 	return err;
 }
 
@@ -887,24 +893,15 @@ out_file:
 	return err;
 }
 
-static struct i915_vma *rpcs_query_batch(struct i915_vma *vma)
+static int rpcs_query_batch(struct drm_i915_gem_object *rpcs, struct i915_vma *vma)
 {
-	struct drm_i915_gem_object *obj;
 	u32 *cmd;
-	int err;
 
-	if (INTEL_GEN(vma->vm->i915) < 8)
-		return ERR_PTR(-EINVAL);
+	GEM_BUG_ON(INTEL_GEN(vma->vm->i915) < 8);
 
-	obj = i915_gem_object_create_internal(vma->vm->i915, PAGE_SIZE);
-	if (IS_ERR(obj))
-		return ERR_CAST(obj);
-
-	cmd = i915_gem_object_pin_map(obj, I915_MAP_WB);
-	if (IS_ERR(cmd)) {
-		err = PTR_ERR(cmd);
-		goto err;
-	}
+	cmd = i915_gem_object_pin_map(rpcs, I915_MAP_WB);
+	if (IS_ERR(cmd))
+		return PTR_ERR(cmd);
 
 	*cmd++ = MI_STORE_REGISTER_MEM_GEN8;
 	*cmd++ = i915_mmio_reg_offset(GEN8_R_PWR_CLK_STATE);
@@ -912,26 +909,12 @@ static struct i915_vma *rpcs_query_batch(struct i915_vma *vma)
 	*cmd++ = upper_32_bits(vma->node.start);
 	*cmd = MI_BATCH_BUFFER_END;
 
-	__i915_gem_object_flush_map(obj, 0, 64);
-	i915_gem_object_unpin_map(obj);
+	__i915_gem_object_flush_map(rpcs, 0, 64);
+	i915_gem_object_unpin_map(rpcs);
 
 	intel_gt_chipset_flush(vma->vm->gt);
 
-	vma = i915_vma_instance(obj, vma->vm, NULL);
-	if (IS_ERR(vma)) {
-		err = PTR_ERR(vma);
-		goto err;
-	}
-
-	err = i915_vma_pin(vma, 0, 0, PIN_USER);
-	if (err)
-		goto err;
-
-	return vma;
-
-err:
-	i915_gem_object_put(obj);
-	return ERR_PTR(err);
+	return 0;
 }
 
 static int
@@ -939,52 +922,68 @@ emit_rpcs_query(struct drm_i915_gem_object *obj,
 		struct intel_context *ce,
 		struct i915_request **rq_out)
 {
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 	struct i915_request *rq;
+	struct i915_gem_ww_ctx ww;
 	struct i915_vma *batch;
 	struct i915_vma *vma;
+	struct drm_i915_gem_object *rpcs;
 	int err;
 
 	GEM_BUG_ON(!intel_engine_can_store_dword(ce->engine));
 
+	if (INTEL_GEN(i915) < 8)
+		return -EINVAL;
+
 	vma = i915_vma_instance(obj, ce->vm, NULL);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
-	i915_gem_object_lock(obj);
-	err = i915_gem_object_set_to_gtt_domain(obj, false);
-	i915_gem_object_unlock(obj);
-	if (err)
-		return err;
-
-	err = i915_vma_pin(vma, 0, 0, PIN_USER);
-	if (err)
-		return err;
+	rpcs = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	if (IS_ERR(rpcs))
+		return PTR_ERR(rpcs);
 
-	batch = rpcs_query_batch(vma);
+	batch = i915_vma_instance(rpcs, ce->vm, NULL);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
-		goto err_vma;
+		goto err_put;
 	}
 
+	i915_gem_ww_ctx_init(&ww, false);
+retry:
+	err = i915_gem_object_lock(obj, &ww);
+	if (!err)
+		err = i915_gem_object_lock(rpcs, &ww);
+	if (!err)
+		err = i915_gem_object_set_to_gtt_domain(obj, false);
+	if (!err)
+		err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_USER);
+	if (err)
+		goto err_put;
+
+	err = i915_vma_pin_ww(batch, &ww, 0, 0, PIN_USER);
+	if (err)
+		goto err_vma;
+
+	err = rpcs_query_batch(rpcs, vma);
+	if (err)
+		goto err_batch;
+
 	rq = i915_request_create(ce);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto err_batch;
 	}
 
-	i915_vma_lock(batch);
 	err = i915_request_await_object(rq, batch->obj, false);
 	if (err == 0)
 		err = i915_vma_move_to_active(batch, rq, 0);
-	i915_vma_unlock(batch);
 	if (err)
 		goto skip_request;
 
-	i915_vma_lock(vma);
 	err = i915_request_await_object(rq, vma->obj, true);
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
 	if (err)
 		goto skip_request;
 
@@ -1000,23 +999,24 @@ emit_rpcs_query(struct drm_i915_gem_object *obj,
 	if (err)
 		goto skip_request;
 
-	i915_vma_unpin_and_release(&batch, 0);
-	i915_vma_unpin(vma);
-
 	*rq_out = i915_request_get(rq);
 
-	i915_request_add(rq);
-
-	return 0;
-
 skip_request:
-	i915_request_set_error_once(rq, err);
+	if (err)
+		i915_request_set_error_once(rq, err);
 	i915_request_add(rq);
 err_batch:
-	i915_vma_unpin_and_release(&batch, 0);
+	i915_vma_unpin(batch);
 err_vma:
 	i915_vma_unpin(vma);
-
+err_put:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	i915_gem_object_put(rpcs);
 	return err;
 }
 
@@ -1709,7 +1709,7 @@ static int read_from_scratch(struct i915_gem_context *ctx,
 
 	i915_request_add(rq);
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_set_to_cpu_domain(obj, false);
 	i915_gem_object_unlock(obj);
 	if (err)
@@ -1748,7 +1748,7 @@ static int check_scratch_page(struct i915_gem_context *ctx, u32 *out)
 	if (!vm)
 		return -ENODEV;
 
-	page = vm->scratch[0].base.page;
+	page = __px_page(vm->scratch[0]);
 	if (!page) {
 		pr_err("No scratch page!\n");
 		return -EINVAL;
@@ -1914,8 +1914,8 @@ static int mock_context_barrier(void *arg)
 		return -ENOMEM;
 
 	counter = 0;
-	err = context_barrier_task(ctx, 0,
-				   NULL, NULL, mock_barrier_task, &counter);
+	err = context_barrier_task(ctx, 0, NULL, NULL, NULL,
+				   mock_barrier_task, &counter);
 	if (err) {
 		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
 		goto out;
@@ -1927,11 +1927,8 @@ static int mock_context_barrier(void *arg)
 	}
 
 	counter = 0;
-	err = context_barrier_task(ctx, ALL_ENGINES,
-				   skip_unused_engines,
-				   NULL,
-				   mock_barrier_task,
-				   &counter);
+	err = context_barrier_task(ctx, ALL_ENGINES, skip_unused_engines,
+				   NULL, NULL, mock_barrier_task, &counter);
 	if (err) {
 		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
 		goto out;
@@ -1951,8 +1948,8 @@ static int mock_context_barrier(void *arg)
 
 	counter = 0;
 	context_barrier_inject_fault = BIT(RCS0);
-	err = context_barrier_task(ctx, ALL_ENGINES,
-				   NULL, NULL, mock_barrier_task, &counter);
+	err = context_barrier_task(ctx, ALL_ENGINES, NULL, NULL, NULL,
+				   mock_barrier_task, &counter);
 	context_barrier_inject_fault = 0;
 	if (err == -ENXIO)
 		err = 0;
@@ -1966,11 +1963,8 @@ static int mock_context_barrier(void *arg)
 		goto out;
 
 	counter = 0;
-	err = context_barrier_task(ctx, ALL_ENGINES,
-				   skip_unused_engines,
-				   NULL,
-				   mock_barrier_task,
-				   &counter);
+	err = context_barrier_task(ctx, ALL_ENGINES, skip_unused_engines,
+				   NULL, NULL, mock_barrier_task, &counter);
 	if (err) {
 		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
 		goto out;
@@ -2003,7 +1997,7 @@ int i915_gem_context_mock_selftests(void)
 
 	err = i915_subtests(tests, i915);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c
index 2a52b92586b9..0845ce1ae37c 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c
@@ -272,7 +272,7 @@ int i915_gem_dmabuf_mock_selftests(void)
 
 	err = i915_subtests(tests, i915);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_execbuffer.c
index a49016f8ee0d..e1d50a5a1477 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_execbuffer.c
@@ -32,46 +32,39 @@ static int __igt_gpu_reloc(struct i915_execbuffer *eb,
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
-	err = i915_vma_pin(vma, 0, 0, PIN_USER | PIN_HIGH);
+	err = i915_gem_object_lock(obj, &eb->ww);
+	if (err)
+		return err;
+
+	err = i915_vma_pin_ww(vma, &eb->ww, 0, 0, PIN_USER | PIN_HIGH);
 	if (err)
 		return err;
 
 	/* 8-Byte aligned */
-	if (!__reloc_entry_gpu(eb, vma,
-			       offsets[0] * sizeof(u32),
-			       0)) {
-		err = -EIO;
-		goto unpin_vma;
-	}
+	err = __reloc_entry_gpu(eb, vma, offsets[0] * sizeof(u32), 0);
+	if (err <= 0)
+		goto reloc_err;
 
 	/* !8-Byte aligned */
-	if (!__reloc_entry_gpu(eb, vma,
-			       offsets[1] * sizeof(u32),
-			       1)) {
-		err = -EIO;
-		goto unpin_vma;
-	}
+	err = __reloc_entry_gpu(eb, vma, offsets[1] * sizeof(u32), 1);
+	if (err <= 0)
+		goto reloc_err;
 
 	/* Skip to the end of the cmd page */
-	i = PAGE_SIZE / sizeof(u32) - RELOC_TAIL - 1;
+	i = PAGE_SIZE / sizeof(u32) - 1;
 	i -= eb->reloc_cache.rq_size;
 	memset32(eb->reloc_cache.rq_cmd + eb->reloc_cache.rq_size,
 		 MI_NOOP, i);
 	eb->reloc_cache.rq_size += i;
 
-	/* Force batch chaining */
-	if (!__reloc_entry_gpu(eb, vma,
-			       offsets[2] * sizeof(u32),
-			       2)) {
-		err = -EIO;
-		goto unpin_vma;
-	}
+	/* Force next batch */
+	err = __reloc_entry_gpu(eb, vma, offsets[2] * sizeof(u32), 2);
+	if (err <= 0)
+		goto reloc_err;
 
 	GEM_BUG_ON(!eb->reloc_cache.rq);
 	rq = i915_request_get(eb->reloc_cache.rq);
-	err = reloc_gpu_flush(&eb->reloc_cache);
-	if (err)
-		goto put_rq;
+	reloc_gpu_flush(eb, &eb->reloc_cache);
 	GEM_BUG_ON(eb->reloc_cache.rq);
 
 	err = i915_gem_object_wait(obj, I915_WAIT_INTERRUPTIBLE, HZ / 2);
@@ -103,6 +96,11 @@ put_rq:
 unpin_vma:
 	i915_vma_unpin(vma);
 	return err;
+
+reloc_err:
+	if (!err)
+		err = -EIO;
+	goto unpin_vma;
 }
 
 static int igt_gpu_reloc(void *arg)
@@ -124,6 +122,8 @@ static int igt_gpu_reloc(void *arg)
 		goto err_scratch;
 	}
 
+	intel_gt_pm_get(&eb.i915->gt);
+
 	for_each_uabi_engine(eb.engine, eb.i915) {
 		reloc_cache_init(&eb.reloc_cache, eb.i915);
 		memset(map, POISON_INUSE, 4096);
@@ -134,15 +134,29 @@ static int igt_gpu_reloc(void *arg)
 			err = PTR_ERR(eb.context);
 			goto err_pm;
 		}
+		eb.reloc_pool = NULL;
+		eb.reloc_context = NULL;
 
-		err = intel_context_pin(eb.context);
-		if (err)
-			goto err_put;
+		i915_gem_ww_ctx_init(&eb.ww, false);
+retry:
+		err = intel_context_pin_ww(eb.context, &eb.ww);
+		if (!err) {
+			err = __igt_gpu_reloc(&eb, scratch);
+
+			intel_context_unpin(eb.context);
+		}
+		if (err == -EDEADLK) {
+			err = i915_gem_ww_ctx_backoff(&eb.ww);
+			if (!err)
+				goto retry;
+		}
+		i915_gem_ww_ctx_fini(&eb.ww);
 
-		err = __igt_gpu_reloc(&eb, scratch);
+		if (eb.reloc_pool)
+			intel_gt_buffer_pool_put(eb.reloc_pool);
+		if (eb.reloc_context)
+			intel_context_put(eb.reloc_context);
 
-		intel_context_unpin(eb.context);
-err_put:
 		intel_context_put(eb.context);
 err_pm:
 		intel_engine_pm_put(eb.engine);
@@ -153,6 +167,7 @@ err_pm:
 	if (igt_flush_test(eb.i915))
 		err = -EIO;
 
+	intel_gt_pm_put(&eb.i915->gt);
 err_scratch:
 	i915_gem_object_put(scratch);
 	return err;
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index 9c7402ce5bf9..d27d87a678c8 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -103,7 +103,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj,
 	GEM_BUG_ON(i915_gem_object_get_tiling(obj) != tile->tiling);
 	GEM_BUG_ON(i915_gem_object_get_stride(obj) != tile->stride);
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(obj, true);
 	i915_gem_object_unlock(obj);
 	if (err) {
@@ -188,7 +188,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj,
 	GEM_BUG_ON(i915_gem_object_get_tiling(obj) != tile->tiling);
 	GEM_BUG_ON(i915_gem_object_get_stride(obj) != tile->stride);
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(obj, true);
 	i915_gem_object_unlock(obj);
 	if (err) {
@@ -528,31 +528,42 @@ static int make_obj_busy(struct drm_i915_gem_object *obj)
 	for_each_uabi_engine(engine, i915) {
 		struct i915_request *rq;
 		struct i915_vma *vma;
+		struct i915_gem_ww_ctx ww;
 		int err;
 
 		vma = i915_vma_instance(obj, &engine->gt->ggtt->vm, NULL);
 		if (IS_ERR(vma))
 			return PTR_ERR(vma);
 
-		err = i915_vma_pin(vma, 0, 0, PIN_USER);
+		i915_gem_ww_ctx_init(&ww, false);
+retry:
+		err = i915_gem_object_lock(obj, &ww);
+		if (!err)
+			err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_USER);
 		if (err)
-			return err;
+			goto err;
 
 		rq = intel_engine_create_kernel_request(engine);
 		if (IS_ERR(rq)) {
-			i915_vma_unpin(vma);
-			return PTR_ERR(rq);
+			err = PTR_ERR(rq);
+			goto err_unpin;
 		}
 
-		i915_vma_lock(vma);
 		err = i915_request_await_object(rq, vma->obj, true);
 		if (err == 0)
 			err = i915_vma_move_to_active(vma, rq,
 						      EXEC_OBJECT_WRITE);
-		i915_vma_unlock(vma);
 
 		i915_request_add(rq);
+err_unpin:
 		i915_vma_unpin(vma);
+err:
+		if (err == -EDEADLK) {
+			err = i915_gem_ww_ctx_backoff(&ww);
+			if (!err)
+				goto retry;
+		}
+		i915_gem_ww_ctx_fini(&ww);
 		if (err)
 			return err;
 	}
@@ -1123,6 +1134,7 @@ static int __igt_mmap_gpu(struct drm_i915_private *i915,
 	for_each_uabi_engine(engine, i915) {
 		struct i915_request *rq;
 		struct i915_vma *vma;
+		struct i915_gem_ww_ctx ww;
 
 		vma = i915_vma_instance(obj, engine->kernel_context->vm, NULL);
 		if (IS_ERR(vma)) {
@@ -1130,9 +1142,13 @@ static int __igt_mmap_gpu(struct drm_i915_private *i915,
 			goto out_unmap;
 		}
 
-		err = i915_vma_pin(vma, 0, 0, PIN_USER);
+		i915_gem_ww_ctx_init(&ww, false);
+retry:
+		err = i915_gem_object_lock(obj, &ww);
+		if (!err)
+			err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_USER);
 		if (err)
-			goto out_unmap;
+			goto out_ww;
 
 		rq = i915_request_create(engine->kernel_context);
 		if (IS_ERR(rq)) {
@@ -1140,11 +1156,9 @@ static int __igt_mmap_gpu(struct drm_i915_private *i915,
 			goto out_unpin;
 		}
 
-		i915_vma_lock(vma);
 		err = i915_request_await_object(rq, vma->obj, false);
 		if (err == 0)
 			err = i915_vma_move_to_active(vma, rq, 0);
-		i915_vma_unlock(vma);
 
 		err = engine->emit_bb_start(rq, vma->node.start, 0, 0);
 		i915_request_get(rq);
@@ -1166,6 +1180,13 @@ static int __igt_mmap_gpu(struct drm_i915_private *i915,
 
 out_unpin:
 		i915_vma_unpin(vma);
+out_ww:
+		if (err == -EDEADLK) {
+			err = i915_gem_ww_ctx_backoff(&ww);
+			if (!err)
+				goto retry;
+		}
+		i915_gem_ww_ctx_fini(&ww);
 		if (err)
 			goto out_unmap;
 	}
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c
index faa5b6d91795..bf853c40ec65 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c
@@ -85,7 +85,7 @@ int i915_gem_object_mock_selftests(void)
 
 	err = i915_subtests(tests, i915);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_phys.c
index 34932871b3a5..8cee68c6a6dc 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_phys.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_phys.c
@@ -44,7 +44,7 @@ static int mock_phys_object(void *arg)
 	}
 
 	/* Make the object dirty so that put_pages must do copy back the data */
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_set_to_gtt_domain(obj, true);
 	i915_gem_object_unlock(obj);
 	if (err) {
@@ -73,6 +73,6 @@ int i915_gem_phys_mock_selftests(void)
 
 	err = i915_subtests(tests, i915);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c
index debaf7b18ab5..be30b27e2926 100644
--- a/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c
+++ b/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c
@@ -28,10 +28,9 @@ static struct sg_table *mock_map_dma_buf(struct dma_buf_attachment *attachment,
 		sg = sg_next(sg);
 	}
 
-	if (!dma_map_sg(attachment->dev, st->sgl, st->nents, dir)) {
-		err = -ENOMEM;
+	err = dma_map_sgtable(attachment->dev, st, dir, 0);
+	if (err)
 		goto err_st;
-	}
 
 	return st;
 
@@ -46,7 +45,7 @@ static void mock_unmap_dma_buf(struct dma_buf_attachment *attachment,
 			       struct sg_table *st,
 			       enum dma_data_direction dir)
 {
-	dma_unmap_sg(attachment->dev, st->sgl, st->nents, dir);
+	dma_unmap_sgtable(attachment->dev, st, dir, 0);
 	sg_free_table(st);
 	kfree(st);
 }
diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
index cdc0b9c54305..fd0d24d28763 100644
--- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
@@ -16,8 +16,10 @@ static inline void gen6_write_pde(const struct gen6_ppgtt *ppgtt,
 				  const unsigned int pde,
 				  const struct i915_page_table *pt)
 {
+	dma_addr_t addr = pt ? px_dma(pt) : px_dma(ppgtt->base.vm.scratch[1]);
+
 	/* Caller needs to make sure the write completes if necessary */
-	iowrite32(GEN6_PDE_ADDR_ENCODE(px_dma(pt)) | GEN6_PDE_VALID,
+	iowrite32(GEN6_PDE_ADDR_ENCODE(addr) | GEN6_PDE_VALID,
 		  ppgtt->pd_addr + pde);
 }
 
@@ -79,7 +81,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 {
 	struct gen6_ppgtt * const ppgtt = to_gen6_ppgtt(i915_vm_to_ppgtt(vm));
 	const unsigned int first_entry = start / I915_GTT_PAGE_SIZE;
-	const gen6_pte_t scratch_pte = vm->scratch[0].encode;
+	const gen6_pte_t scratch_pte = vm->scratch[0]->encode;
 	unsigned int pde = first_entry / GEN6_PTES;
 	unsigned int pte = first_entry % GEN6_PTES;
 	unsigned int num_entries = length / I915_GTT_PAGE_SIZE;
@@ -90,8 +92,6 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 		const unsigned int count = min(num_entries, GEN6_PTES - pte);
 		gen6_pte_t *vaddr;
 
-		GEM_BUG_ON(px_base(pt) == px_base(&vm->scratch[1]));
-
 		num_entries -= count;
 
 		GEM_BUG_ON(count > atomic_read(&pt->used));
@@ -127,7 +127,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 	struct sgt_dma iter = sgt_dma(vma);
 	gen6_pte_t *vaddr;
 
-	GEM_BUG_ON(pd->entry[act_pt] == &vm->scratch[1]);
+	GEM_BUG_ON(!pd->entry[act_pt]);
 
 	vaddr = kmap_atomic_px(i915_pt_entry(pd, act_pt));
 	do {
@@ -177,39 +177,36 @@ static void gen6_flush_pd(struct gen6_ppgtt *ppgtt, u64 start, u64 end)
 	mutex_unlock(&ppgtt->flush);
 }
 
-static int gen6_alloc_va_range(struct i915_address_space *vm,
-			       u64 start, u64 length)
+static void gen6_alloc_va_range(struct i915_address_space *vm,
+				struct i915_vm_pt_stash *stash,
+				u64 start, u64 length)
 {
 	struct gen6_ppgtt *ppgtt = to_gen6_ppgtt(i915_vm_to_ppgtt(vm));
 	struct i915_page_directory * const pd = ppgtt->base.pd;
-	struct i915_page_table *pt, *alloc = NULL;
+	struct i915_page_table *pt;
 	bool flush = false;
 	u64 from = start;
 	unsigned int pde;
-	int ret = 0;
 
 	spin_lock(&pd->lock);
 	gen6_for_each_pde(pt, pd, start, length, pde) {
 		const unsigned int count = gen6_pte_count(start, length);
 
-		if (px_base(pt) == px_base(&vm->scratch[1])) {
+		if (!pt) {
 			spin_unlock(&pd->lock);
 
-			pt = fetch_and_zero(&alloc);
-			if (!pt)
-				pt = alloc_pt(vm);
-			if (IS_ERR(pt)) {
-				ret = PTR_ERR(pt);
-				goto unwind_out;
-			}
+			pt = stash->pt[0];
+			__i915_gem_object_pin_pages(pt->base);
+			i915_gem_object_make_unshrinkable(pt->base);
 
-			fill32_px(pt, vm->scratch[0].encode);
+			fill32_px(pt, vm->scratch[0]->encode);
 
 			spin_lock(&pd->lock);
-			if (pd->entry[pde] == &vm->scratch[1]) {
+			if (!pd->entry[pde]) {
+				stash->pt[0] = pt->stash;
+				atomic_set(&pt->used, 0);
 				pd->entry[pde] = pt;
 			} else {
-				alloc = pt;
 				pt = pd->entry[pde];
 			}
 
@@ -226,38 +223,32 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,
 		with_intel_runtime_pm(&vm->i915->runtime_pm, wakeref)
 			gen6_flush_pd(ppgtt, from, start);
 	}
-
-	goto out;
-
-unwind_out:
-	gen6_ppgtt_clear_range(vm, from, start - from);
-out:
-	if (alloc)
-		free_px(vm, alloc);
-	return ret;
 }
 
 static int gen6_ppgtt_init_scratch(struct gen6_ppgtt *ppgtt)
 {
 	struct i915_address_space * const vm = &ppgtt->base.vm;
-	struct i915_page_directory * const pd = ppgtt->base.pd;
 	int ret;
 
-	ret = setup_scratch_page(vm, __GFP_HIGHMEM);
+	ret = setup_scratch_page(vm);
 	if (ret)
 		return ret;
 
-	vm->scratch[0].encode =
-		vm->pte_encode(px_dma(&vm->scratch[0]),
+	vm->scratch[0]->encode =
+		vm->pte_encode(px_dma(vm->scratch[0]),
 			       I915_CACHE_NONE, PTE_READ_ONLY);
 
-	if (unlikely(setup_page_dma(vm, px_base(&vm->scratch[1])))) {
-		cleanup_scratch_page(vm);
-		return -ENOMEM;
+	vm->scratch[1] = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
+	if (IS_ERR(vm->scratch[1]))
+		return PTR_ERR(vm->scratch[1]);
+
+	ret = pin_pt_dma(vm, vm->scratch[1]);
+	if (ret) {
+		i915_gem_object_put(vm->scratch[1]);
+		return ret;
 	}
 
-	fill32_px(&vm->scratch[1], vm->scratch[0].encode);
-	memset_p(pd->entry, &vm->scratch[1], I915_PDES);
+	fill32_px(vm->scratch[1], vm->scratch[0]->encode);
 
 	return 0;
 }
@@ -265,14 +256,12 @@ static int gen6_ppgtt_init_scratch(struct gen6_ppgtt *ppgtt)
 static void gen6_ppgtt_free_pd(struct gen6_ppgtt *ppgtt)
 {
 	struct i915_page_directory * const pd = ppgtt->base.pd;
-	struct i915_page_dma * const scratch =
-		px_base(&ppgtt->base.vm.scratch[1]);
 	struct i915_page_table *pt;
 	u32 pde;
 
 	gen6_for_all_pdes(pt, pd, pde)
-		if (px_base(pt) != scratch)
-			free_px(&ppgtt->base.vm, pt);
+		if (pt)
+			free_pt(&ppgtt->base.vm, pt);
 }
 
 static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
@@ -286,7 +275,8 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 
 	mutex_destroy(&ppgtt->flush);
 	mutex_destroy(&ppgtt->pin_mutex);
-	kfree(ppgtt->base.pd);
+
+	free_pd(&ppgtt->base.vm, ppgtt->base.pd);
 }
 
 static int pd_vma_set_pages(struct i915_vma *vma)
@@ -302,28 +292,26 @@ static void pd_vma_clear_pages(struct i915_vma *vma)
 	vma->pages = NULL;
 }
 
-static int pd_vma_bind(struct i915_address_space *vm,
-		       struct i915_vma *vma,
-		       enum i915_cache_level cache_level,
-		       u32 unused)
+static void pd_vma_bind(struct i915_address_space *vm,
+			struct i915_vm_pt_stash *stash,
+			struct i915_vma *vma,
+			enum i915_cache_level cache_level,
+			u32 unused)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 	struct gen6_ppgtt *ppgtt = vma->private;
 	u32 ggtt_offset = i915_ggtt_offset(vma) / I915_GTT_PAGE_SIZE;
 
-	px_base(ppgtt->base.pd)->ggtt_offset = ggtt_offset * sizeof(gen6_pte_t);
+	ppgtt->pp_dir = ggtt_offset * sizeof(gen6_pte_t) << 10;
 	ppgtt->pd_addr = (gen6_pte_t __iomem *)ggtt->gsm + ggtt_offset;
 
 	gen6_flush_pd(ppgtt, 0, ppgtt->base.vm.total);
-	return 0;
 }
 
 static void pd_vma_unbind(struct i915_address_space *vm, struct i915_vma *vma)
 {
 	struct gen6_ppgtt *ppgtt = vma->private;
 	struct i915_page_directory * const pd = ppgtt->base.pd;
-	struct i915_page_dma * const scratch =
-		px_base(&ppgtt->base.vm.scratch[1]);
 	struct i915_page_table *pt;
 	unsigned int pde;
 
@@ -332,11 +320,11 @@ static void pd_vma_unbind(struct i915_address_space *vm, struct i915_vma *vma)
 
 	/* Free all no longer used page tables */
 	gen6_for_all_pdes(pt, ppgtt->base.pd, pde) {
-		if (px_base(pt) == scratch || atomic_read(&pt->used))
+		if (!pt || atomic_read(&pt->used))
 			continue;
 
-		free_px(&ppgtt->base.vm, pt);
-		pd->entry[pde] = scratch;
+		free_pt(&ppgtt->base.vm, pt);
+		pd->entry[pde] = NULL;
 	}
 
 	ppgtt->scan_for_unused_pt = false;
@@ -380,7 +368,7 @@ static struct i915_vma *pd_vma_create(struct gen6_ppgtt *ppgtt, int size)
 	return vma;
 }
 
-int gen6_ppgtt_pin(struct i915_ppgtt *base)
+int gen6_ppgtt_pin(struct i915_ppgtt *base, struct i915_gem_ww_ctx *ww)
 {
 	struct gen6_ppgtt *ppgtt = to_gen6_ppgtt(base);
 	int err;
@@ -406,7 +394,7 @@ int gen6_ppgtt_pin(struct i915_ppgtt *base)
 	 */
 	err = 0;
 	if (!atomic_read(&ppgtt->pin_count))
-		err = i915_ggtt_pin(ppgtt->vma, GEN6_PD_ALIGN, PIN_HIGH);
+		err = i915_ggtt_pin(ppgtt->vma, ww, GEN6_PD_ALIGN, PIN_HIGH);
 	if (!err)
 		atomic_inc(&ppgtt->pin_count);
 	mutex_unlock(&ppgtt->pin_mutex);
@@ -448,6 +436,7 @@ struct i915_ppgtt *gen6_ppgtt_create(struct intel_gt *gt)
 	mutex_init(&ppgtt->pin_mutex);
 
 	ppgtt_init(&ppgtt->base, gt);
+	ppgtt->base.vm.pd_shift = ilog2(SZ_4K * SZ_4K / sizeof(gen6_pte_t));
 	ppgtt->base.vm.top = 1;
 
 	ppgtt->base.vm.bind_async_flags = I915_VMA_LOCAL_BIND;
@@ -456,9 +445,10 @@ struct i915_ppgtt *gen6_ppgtt_create(struct intel_gt *gt)
 	ppgtt->base.vm.insert_entries = gen6_ppgtt_insert_entries;
 	ppgtt->base.vm.cleanup = gen6_ppgtt_cleanup;
 
+	ppgtt->base.vm.alloc_pt_dma = alloc_pt_dma;
 	ppgtt->base.vm.pte_encode = ggtt->vm.pte_encode;
 
-	ppgtt->base.pd = __alloc_pd(sizeof(*ppgtt->base.pd));
+	ppgtt->base.pd = __alloc_pd(I915_PDES);
 	if (!ppgtt->base.pd) {
 		err = -ENOMEM;
 		goto err_free;
@@ -479,7 +469,7 @@ struct i915_ppgtt *gen6_ppgtt_create(struct intel_gt *gt)
 err_scratch:
 	free_scratch(&ppgtt->base.vm);
 err_pd:
-	kfree(ppgtt->base.pd);
+	free_pd(&ppgtt->base.vm, ppgtt->base.pd);
 err_free:
 	mutex_destroy(&ppgtt->pin_mutex);
 	kfree(ppgtt);
diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.h b/drivers/gpu/drm/i915/gt/gen6_ppgtt.h
index 72e481806c96..3357228f3304 100644
--- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.h
+++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.h
@@ -8,12 +8,15 @@
 
 #include "intel_gtt.h"
 
+struct i915_gem_ww_ctx;
+
 struct gen6_ppgtt {
 	struct i915_ppgtt base;
 
 	struct mutex flush;
 	struct i915_vma *vma;
 	gen6_pte_t __iomem *pd_addr;
+	u32 pp_dir;
 
 	atomic_t pin_count;
 	struct mutex pin_mutex;
@@ -66,7 +69,7 @@ static inline struct gen6_ppgtt *to_gen6_ppgtt(struct i915_ppgtt *base)
 		     (pt = i915_pt_entry(pd, iter), true);		\
 	     ++iter)
 
-int gen6_ppgtt_pin(struct i915_ppgtt *base);
+int gen6_ppgtt_pin(struct i915_ppgtt *base, struct i915_gem_ww_ctx *ww);
 void gen6_ppgtt_unpin(struct i915_ppgtt *base);
 void gen6_ppgtt_unpin_all(struct i915_ppgtt *base);
 void gen6_ppgtt_enable(struct intel_gt *gt);
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 699125928272..eb64f474a78c 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -181,7 +181,7 @@ static void __gen8_ppgtt_cleanup(struct i915_address_space *vm,
 		} while (pde++, --count);
 	}
 
-	free_px(vm, pd);
+	free_px(vm, &pd->pt, lvl);
 }
 
 static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
@@ -199,7 +199,7 @@ static u64 __gen8_ppgtt_clear(struct i915_address_space * const vm,
 			      struct i915_page_directory * const pd,
 			      u64 start, const u64 end, int lvl)
 {
-	const struct i915_page_scratch * const scratch = &vm->scratch[lvl];
+	const struct drm_i915_gem_object * const scratch = vm->scratch[lvl];
 	unsigned int idx, len;
 
 	GEM_BUG_ON(end > vm->total >> GEN8_PTE_SHIFT);
@@ -239,7 +239,7 @@ static u64 __gen8_ppgtt_clear(struct i915_address_space * const vm,
 
 			vaddr = kmap_atomic_px(pt);
 			memset64(vaddr + gen8_pd_index(start, 0),
-				 vm->scratch[0].encode,
+				 vm->scratch[0]->encode,
 				 count);
 			kunmap_atomic(vaddr);
 
@@ -248,7 +248,7 @@ static u64 __gen8_ppgtt_clear(struct i915_address_space * const vm,
 		}
 
 		if (release_pd_entry(pd, idx, pt, scratch))
-			free_px(vm, pt);
+			free_px(vm, pt, lvl);
 	} while (idx++, --len);
 
 	return start;
@@ -269,14 +269,12 @@ static void gen8_ppgtt_clear(struct i915_address_space *vm,
 			   start, start + length, vm->top);
 }
 
-static int __gen8_ppgtt_alloc(struct i915_address_space * const vm,
-			      struct i915_page_directory * const pd,
-			      u64 * const start, const u64 end, int lvl)
+static void __gen8_ppgtt_alloc(struct i915_address_space * const vm,
+			       struct i915_vm_pt_stash *stash,
+			       struct i915_page_directory * const pd,
+			       u64 * const start, const u64 end, int lvl)
 {
-	const struct i915_page_scratch * const scratch = &vm->scratch[lvl];
-	struct i915_page_table *alloc = NULL;
 	unsigned int idx, len;
-	int ret = 0;
 
 	GEM_BUG_ON(end > vm->total >> GEN8_PTE_SHIFT);
 
@@ -297,49 +295,31 @@ static int __gen8_ppgtt_alloc(struct i915_address_space * const vm,
 			DBG("%s(%p):{ lvl:%d, idx:%d } allocating new tree\n",
 			    __func__, vm, lvl + 1, idx);
 
-			pt = fetch_and_zero(&alloc);
-			if (lvl) {
-				if (!pt) {
-					pt = &alloc_pd(vm)->pt;
-					if (IS_ERR(pt)) {
-						ret = PTR_ERR(pt);
-						goto out;
-					}
-				}
-
-				fill_px(pt, vm->scratch[lvl].encode);
-			} else {
-				if (!pt) {
-					pt = alloc_pt(vm);
-					if (IS_ERR(pt)) {
-						ret = PTR_ERR(pt);
-						goto out;
-					}
-				}
-
-				if (intel_vgpu_active(vm->i915) ||
-				    gen8_pt_count(*start, end) < I915_PDES)
-					fill_px(pt, vm->scratch[lvl].encode);
-			}
+			pt = stash->pt[!!lvl];
+			__i915_gem_object_pin_pages(pt->base);
+			i915_gem_object_make_unshrinkable(pt->base);
+
+			if (lvl ||
+			    gen8_pt_count(*start, end) < I915_PDES ||
+			    intel_vgpu_active(vm->i915))
+				fill_px(pt, vm->scratch[lvl]->encode);
 
 			spin_lock(&pd->lock);
-			if (likely(!pd->entry[idx]))
+			if (likely(!pd->entry[idx])) {
+				stash->pt[!!lvl] = pt->stash;
+				atomic_set(&pt->used, 0);
 				set_pd_entry(pd, idx, pt);
-			else
-				alloc = pt, pt = pd->entry[idx];
+			} else {
+				pt = pd->entry[idx];
+			}
 		}
 
 		if (lvl) {
 			atomic_inc(&pt->used);
 			spin_unlock(&pd->lock);
 
-			ret = __gen8_ppgtt_alloc(vm, as_pd(pt),
-						 start, end, lvl);
-			if (unlikely(ret)) {
-				if (release_pd_entry(pd, idx, pt, scratch))
-					free_px(vm, pt);
-				goto out;
-			}
+			__gen8_ppgtt_alloc(vm, stash,
+					   as_pd(pt), start, end, lvl);
 
 			spin_lock(&pd->lock);
 			atomic_dec(&pt->used);
@@ -359,18 +339,12 @@ static int __gen8_ppgtt_alloc(struct i915_address_space * const vm,
 		}
 	} while (idx++, --len);
 	spin_unlock(&pd->lock);
-out:
-	if (alloc)
-		free_px(vm, alloc);
-	return ret;
 }
 
-static int gen8_ppgtt_alloc(struct i915_address_space *vm,
-			    u64 start, u64 length)
+static void gen8_ppgtt_alloc(struct i915_address_space *vm,
+			     struct i915_vm_pt_stash *stash,
+			     u64 start, u64 length)
 {
-	u64 from;
-	int err;
-
 	GEM_BUG_ON(!IS_ALIGNED(start, BIT_ULL(GEN8_PTE_SHIFT)));
 	GEM_BUG_ON(!IS_ALIGNED(length, BIT_ULL(GEN8_PTE_SHIFT)));
 	GEM_BUG_ON(range_overflows(start, length, vm->total));
@@ -378,25 +352,9 @@ static int gen8_ppgtt_alloc(struct i915_address_space *vm,
 	start >>= GEN8_PTE_SHIFT;
 	length >>= GEN8_PTE_SHIFT;
 	GEM_BUG_ON(length == 0);
-	from = start;
 
-	err = __gen8_ppgtt_alloc(vm, i915_vm_to_ppgtt(vm)->pd,
-				 &start, start + length, vm->top);
-	if (unlikely(err && from != start))
-		__gen8_ppgtt_clear(vm, i915_vm_to_ppgtt(vm)->pd,
-				   from, start, vm->top);
-
-	return err;
-}
-
-static __always_inline void
-write_pte(gen8_pte_t *pte, const gen8_pte_t val)
-{
-	/* Magic delays? Or can we refine these to flush all in one pass? */
-	*pte = val;
-	wmb(); /* cpu to cache */
-	clflush(pte); /* cache to memory */
-	wmb(); /* visible to all */
+	__gen8_ppgtt_alloc(vm, stash, i915_vm_to_ppgtt(vm)->pd,
+			   &start, start + length, vm->top);
 }
 
 static __always_inline u64
@@ -415,8 +373,7 @@ gen8_ppgtt_insert_pte(struct i915_ppgtt *ppgtt,
 	vaddr = kmap_atomic_px(i915_pt_entry(pd, gen8_pd_index(idx, 1)));
 	do {
 		GEM_BUG_ON(iter->sg->length < I915_GTT_PAGE_SIZE);
-		write_pte(&vaddr[gen8_pd_index(idx, 0)],
-			  pte_encode | iter->dma);
+		vaddr[gen8_pd_index(idx, 0)] = pte_encode | iter->dma;
 
 		iter->dma += I915_GTT_PAGE_SIZE;
 		if (iter->dma >= iter->max) {
@@ -439,10 +396,12 @@ gen8_ppgtt_insert_pte(struct i915_ppgtt *ppgtt,
 				pd = pdp->entry[gen8_pd_index(idx, 2)];
 			}
 
+			clflush_cache_range(vaddr, PAGE_SIZE);
 			kunmap_atomic(vaddr);
 			vaddr = kmap_atomic_px(i915_pt_entry(pd, gen8_pd_index(idx, 1)));
 		}
 	} while (1);
+	clflush_cache_range(vaddr, PAGE_SIZE);
 	kunmap_atomic(vaddr);
 
 	return idx;
@@ -498,7 +457,7 @@ static void gen8_ppgtt_insert_huge(struct i915_vma *vma,
 
 		do {
 			GEM_BUG_ON(iter->sg->length < page_size);
-			write_pte(&vaddr[index++], encode | iter->dma);
+			vaddr[index++] = encode | iter->dma;
 
 			start += page_size;
 			iter->dma += page_size;
@@ -523,6 +482,7 @@ static void gen8_ppgtt_insert_huge(struct i915_vma *vma,
 			}
 		} while (rem >= page_size && index < I915_PDES);
 
+		clflush_cache_range(vaddr, PAGE_SIZE);
 		kunmap_atomic(vaddr);
 
 		/*
@@ -554,7 +514,7 @@ static void gen8_ppgtt_insert_huge(struct i915_vma *vma,
 			if (I915_SELFTEST_ONLY(vma->vm->scrub_64K)) {
 				u16 i;
 
-				encode = vma->vm->scratch[0].encode;
+				encode = vma->vm->scratch[0]->encode;
 				vaddr = kmap_atomic_px(i915_pt_entry(pd, maybe_64K));
 
 				for (i = 1; i < index; i += 16)
@@ -608,27 +568,37 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 		GEM_BUG_ON(!clone->has_read_only);
 
 		vm->scratch_order = clone->scratch_order;
-		memcpy(vm->scratch, clone->scratch, sizeof(vm->scratch));
-		px_dma(&vm->scratch[0]) = 0; /* no xfer of ownership */
+		for (i = 0; i <= vm->top; i++)
+			vm->scratch[i] = i915_gem_object_get(clone->scratch[i]);
+
 		return 0;
 	}
 
-	ret = setup_scratch_page(vm, __GFP_HIGHMEM);
+	ret = setup_scratch_page(vm);
 	if (ret)
 		return ret;
 
-	vm->scratch[0].encode =
-		gen8_pte_encode(px_dma(&vm->scratch[0]),
+	vm->scratch[0]->encode =
+		gen8_pte_encode(px_dma(vm->scratch[0]),
 				I915_CACHE_LLC, vm->has_read_only);
 
 	for (i = 1; i <= vm->top; i++) {
-		if (unlikely(setup_page_dma(vm, px_base(&vm->scratch[i]))))
+		struct drm_i915_gem_object *obj;
+
+		obj = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
+		if (IS_ERR(obj))
+			goto free_scratch;
+
+		ret = pin_pt_dma(vm, obj);
+		if (ret) {
+			i915_gem_object_put(obj);
 			goto free_scratch;
+		}
 
-		fill_px(&vm->scratch[i], vm->scratch[i - 1].encode);
-		vm->scratch[i].encode =
-			gen8_pde_encode(px_dma(&vm->scratch[i]),
-					I915_CACHE_LLC);
+		fill_px(obj, vm->scratch[i - 1]->encode);
+		obj->encode = gen8_pde_encode(px_dma(obj), I915_CACHE_LLC);
+
+		vm->scratch[i] = obj;
 	}
 
 	return 0;
@@ -649,12 +619,20 @@ static int gen8_preallocate_top_level_pdp(struct i915_ppgtt *ppgtt)
 
 	for (idx = 0; idx < GEN8_3LVL_PDPES; idx++) {
 		struct i915_page_directory *pde;
+		int err;
 
 		pde = alloc_pd(vm);
 		if (IS_ERR(pde))
 			return PTR_ERR(pde);
 
-		fill_px(pde, vm->scratch[1].encode);
+		err = pin_pt_dma(vm, pde->pt.base);
+		if (err) {
+			i915_gem_object_put(pde->pt.base);
+			free_pd(vm, pde);
+			return err;
+		}
+
+		fill_px(pde, vm->scratch[1]->encode);
 		set_pd_entry(pd, idx, pde);
 		atomic_inc(px_used(pde)); /* keep pinned */
 	}
@@ -668,21 +646,32 @@ gen8_alloc_top_pd(struct i915_address_space *vm)
 {
 	const unsigned int count = gen8_pd_top_count(vm);
 	struct i915_page_directory *pd;
+	int err;
 
-	GEM_BUG_ON(count > ARRAY_SIZE(pd->entry));
+	GEM_BUG_ON(count > I915_PDES);
 
-	pd = __alloc_pd(offsetof(typeof(*pd), entry[count]));
+	pd = __alloc_pd(count);
 	if (unlikely(!pd))
 		return ERR_PTR(-ENOMEM);
 
-	if (unlikely(setup_page_dma(vm, px_base(pd)))) {
-		kfree(pd);
-		return ERR_PTR(-ENOMEM);
+	pd->pt.base = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
+	if (IS_ERR(pd->pt.base)) {
+		err = PTR_ERR(pd->pt.base);
+		pd->pt.base = NULL;
+		goto err_pd;
 	}
 
-	fill_page_dma(px_base(pd), vm->scratch[vm->top].encode, count);
+	err = pin_pt_dma(vm, pd->pt.base);
+	if (err)
+		goto err_pd;
+
+	fill_page_dma(px_base(pd), vm->scratch[vm->top]->encode, count);
 	atomic_inc(px_used(pd)); /* mark as pinned */
 	return pd;
+
+err_pd:
+	free_pd(vm, pd);
+	return ERR_PTR(err);
 }
 
 /*
@@ -703,6 +692,7 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt)
 
 	ppgtt_init(ppgtt, gt);
 	ppgtt->vm.top = i915_vm_is_4lvl(&ppgtt->vm) ? 3 : 2;
+	ppgtt->vm.pd_shift = ilog2(SZ_4K * SZ_4K / sizeof(gen8_pte_t));
 
 	/*
 	 * From bdw, there is hw support for read-only pages in the PPGTT.
@@ -714,12 +704,7 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt)
 	 */
 	ppgtt->vm.has_read_only = !IS_GEN_RANGE(gt->i915, 11, 12);
 
-	/*
-	 * There are only few exceptions for gen >=6. chv and bxt.
-	 * And we are not sure about the latter so play safe for now.
-	 */
-	if (IS_CHERRYVIEW(gt->i915) || IS_BROXTON(gt->i915))
-		ppgtt->vm.pt_kmap_wc = true;
+	ppgtt->vm.alloc_pt_dma = alloc_pt_dma;
 
 	err = gen8_init_scratch(&ppgtt->vm);
 	if (err)
diff --git a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
index 91786310c114..d8b206e53660 100644
--- a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
+++ b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
@@ -28,6 +28,8 @@
 
 #include "i915_drv.h"
 #include "i915_trace.h"
+#include "intel_breadcrumbs.h"
+#include "intel_context.h"
 #include "intel_gt_pm.h"
 #include "intel_gt_requests.h"
 
@@ -53,33 +55,65 @@ static void irq_disable(struct intel_engine_cs *engine)
 	spin_unlock(&engine->gt->irq_lock);
 }
 
-static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b)
+static void __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)
 {
-	struct intel_engine_cs *engine =
-		container_of(b, struct intel_engine_cs, breadcrumbs);
+	lockdep_assert_held(&b->irq_lock);
+
+	if (!b->irq_engine || b->irq_armed)
+		return;
+
+	if (!intel_gt_pm_get_if_awake(b->irq_engine->gt))
+		return;
+
+	/*
+	 * The breadcrumb irq will be disarmed on the interrupt after the
+	 * waiters are signaled. This gives us a single interrupt window in
+	 * which we can add a new waiter and avoid the cost of re-enabling
+	 * the irq.
+	 */
+	WRITE_ONCE(b->irq_armed, true);
+
+	/*
+	 * Since we are waiting on a request, the GPU should be busy
+	 * and should have its own rpm reference. This is tracked
+	 * by i915->gt.awake, we can forgo holding our own wakref
+	 * for the interrupt as before i915->gt.awake is released (when
+	 * the driver is idle) we disarm the breadcrumbs.
+	 */
 
+	if (!b->irq_enabled++)
+		irq_enable(b->irq_engine);
+}
+
+static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b)
+{
 	lockdep_assert_held(&b->irq_lock);
 
+	if (!b->irq_engine || !b->irq_armed)
+		return;
+
 	GEM_BUG_ON(!b->irq_enabled);
 	if (!--b->irq_enabled)
-		irq_disable(engine);
+		irq_disable(b->irq_engine);
 
 	WRITE_ONCE(b->irq_armed, false);
-	intel_gt_pm_put_async(engine->gt);
+	intel_gt_pm_put_async(b->irq_engine->gt);
 }
 
-void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine)
+static void add_signaling_context(struct intel_breadcrumbs *b,
+				  struct intel_context *ce)
 {
-	struct intel_breadcrumbs *b = &engine->breadcrumbs;
-	unsigned long flags;
-
-	if (!READ_ONCE(b->irq_armed))
-		return;
+	intel_context_get(ce);
+	list_add_tail(&ce->signal_link, &b->signalers);
+	if (list_is_first(&ce->signal_link, &b->signalers))
+		__intel_breadcrumbs_arm_irq(b);
+}
 
-	spin_lock_irqsave(&b->irq_lock, flags);
-	if (b->irq_armed)
-		__intel_breadcrumbs_disarm_irq(b);
-	spin_unlock_irqrestore(&b->irq_lock, flags);
+static void remove_signaling_context(struct intel_breadcrumbs *b,
+				     struct intel_context *ce)
+{
+	list_del(&ce->signal_link);
+	intel_context_put(ce);
 }
 
 static inline bool __request_completed(const struct i915_request *rq)
@@ -90,6 +124,9 @@ static inline bool __request_completed(const struct i915_request *rq)
 __maybe_unused static bool
 check_signal_order(struct intel_context *ce, struct i915_request *rq)
 {
+	if (rq->context != ce)
+		return false;
+
 	if (!list_is_last(&rq->signal_link, &ce->signals) &&
 	    i915_seqno_passed(rq->fence.seqno,
 			      list_next_entry(rq, signal_link)->fence.seqno))
@@ -133,25 +170,21 @@ __dma_fence_signal__notify(struct dma_fence *fence,
 
 static void add_retire(struct intel_breadcrumbs *b, struct intel_timeline *tl)
 {
-	struct intel_engine_cs *engine =
-		container_of(b, struct intel_engine_cs, breadcrumbs);
-
-	if (unlikely(intel_engine_is_virtual(engine)))
-		engine = intel_virtual_engine_get_sibling(engine, 0);
-
-	intel_engine_add_retire(engine, tl);
+	if (b->irq_engine)
+		intel_engine_add_retire(b->irq_engine, tl);
 }
 
-static void __signal_request(struct i915_request *rq, struct list_head *signals)
+static bool __signal_request(struct i915_request *rq, struct list_head *signals)
 {
-	GEM_BUG_ON(!test_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags));
 	clear_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags);
 
-	if (!__dma_fence_signal(&rq->fence))
-		return;
+	if (!__dma_fence_signal(&rq->fence)) {
+		i915_request_put(rq);
+		return false;
+	}
 
-	i915_request_get(rq);
 	list_add_tail(&rq->signal_link, signals);
+	return true;
 }
 
 static void signal_irq_work(struct irq_work *work)
@@ -164,7 +197,7 @@ static void signal_irq_work(struct irq_work *work)
 
 	spin_lock(&b->irq_lock);
 
-	if (b->irq_armed && list_empty(&b->signalers))
+	if (list_empty(&b->signalers))
 		__intel_breadcrumbs_disarm_irq(b);
 
 	list_splice_init(&b->signaled_requests, &signal);
@@ -197,8 +230,8 @@ static void signal_irq_work(struct irq_work *work)
 			/* Advance the list to the first incomplete request */
 			__list_del_many(&ce->signals, pos);
 			if (&ce->signals == pos) { /* now empty */
-				list_del_init(&ce->signal_link);
 				add_retire(b, ce->timeline);
+				remove_signaling_context(b, ce);
 			}
 		}
 	}
@@ -220,116 +253,89 @@ static void signal_irq_work(struct irq_work *work)
 	}
 }
 
-static bool __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)
+struct intel_breadcrumbs *
+intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)
 {
-	struct intel_engine_cs *engine =
-		container_of(b, struct intel_engine_cs, breadcrumbs);
-
-	lockdep_assert_held(&b->irq_lock);
-	if (b->irq_armed)
-		return true;
-
-	if (!intel_gt_pm_get_if_awake(engine->gt))
-		return false;
-
-	/*
-	 * The breadcrumb irq will be disarmed on the interrupt after the
-	 * waiters are signaled. This gives us a single interrupt window in
-	 * which we can add a new waiter and avoid the cost of re-enabling
-	 * the irq.
-	 */
-	WRITE_ONCE(b->irq_armed, true);
-
-	/*
-	 * Since we are waiting on a request, the GPU should be busy
-	 * and should have its own rpm reference. This is tracked
-	 * by i915->gt.awake, we can forgo holding our own wakref
-	 * for the interrupt as before i915->gt.awake is released (when
-	 * the driver is idle) we disarm the breadcrumbs.
-	 */
-
-	if (!b->irq_enabled++)
-		irq_enable(engine);
+	struct intel_breadcrumbs *b;
 
-	return true;
-}
-
-void intel_engine_init_breadcrumbs(struct intel_engine_cs *engine)
-{
-	struct intel_breadcrumbs *b = &engine->breadcrumbs;
+	b = kzalloc(sizeof(*b), GFP_KERNEL);
+	if (!b)
+		return NULL;
 
 	spin_lock_init(&b->irq_lock);
 	INIT_LIST_HEAD(&b->signalers);
 	INIT_LIST_HEAD(&b->signaled_requests);
 
 	init_irq_work(&b->irq_work, signal_irq_work);
+
+	b->irq_engine = irq_engine;
+
+	return b;
 }
 
-void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine)
+void intel_breadcrumbs_reset(struct intel_breadcrumbs *b)
 {
-	struct intel_breadcrumbs *b = &engine->breadcrumbs;
 	unsigned long flags;
 
+	if (!b->irq_engine)
+		return;
+
 	spin_lock_irqsave(&b->irq_lock, flags);
 
 	if (b->irq_enabled)
-		irq_enable(engine);
+		irq_enable(b->irq_engine);
 	else
-		irq_disable(engine);
+		irq_disable(b->irq_engine);
 
 	spin_unlock_irqrestore(&b->irq_lock, flags);
 }
 
-void intel_engine_transfer_stale_breadcrumbs(struct intel_engine_cs *engine,
-					     struct intel_context *ce)
+void intel_breadcrumbs_park(struct intel_breadcrumbs *b)
 {
-	struct intel_breadcrumbs *b = &engine->breadcrumbs;
 	unsigned long flags;
 
-	spin_lock_irqsave(&b->irq_lock, flags);
-	if (!list_empty(&ce->signals)) {
-		struct i915_request *rq, *next;
-
-		/* Queue for executing the signal callbacks in the irq_work */
-		list_for_each_entry_safe(rq, next, &ce->signals, signal_link) {
-			GEM_BUG_ON(rq->engine != engine);
-			GEM_BUG_ON(!__request_completed(rq));
-
-			__signal_request(rq, &b->signaled_requests);
-		}
+	if (!READ_ONCE(b->irq_armed))
+		return;
 
-		INIT_LIST_HEAD(&ce->signals);
-		list_del_init(&ce->signal_link);
+	spin_lock_irqsave(&b->irq_lock, flags);
+	__intel_breadcrumbs_disarm_irq(b);
+	spin_unlock_irqrestore(&b->irq_lock, flags);
 
+	if (!list_empty(&b->signalers))
 		irq_work_queue(&b->irq_work);
-	}
-	spin_unlock_irqrestore(&b->irq_lock, flags);
 }
 
-void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine)
+void intel_breadcrumbs_free(struct intel_breadcrumbs *b)
 {
+	kfree(b);
 }
 
-bool i915_request_enable_breadcrumb(struct i915_request *rq)
+static void insert_breadcrumb(struct i915_request *rq,
+			      struct intel_breadcrumbs *b)
 {
-	lockdep_assert_held(&rq->lock);
-
-	if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &rq->fence.flags))
-		return true;
+	struct intel_context *ce = rq->context;
+	struct list_head *pos;
 
-	if (test_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags)) {
-		struct intel_breadcrumbs *b = &rq->engine->breadcrumbs;
-		struct intel_context *ce = rq->context;
-		struct list_head *pos;
-
-		spin_lock(&b->irq_lock);
+	if (test_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags))
+		return;
 
-		if (test_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags))
-			goto unlock;
+	i915_request_get(rq);
 
-		if (!__intel_breadcrumbs_arm_irq(b))
-			goto unlock;
+	/*
+	 * If the request is already completed, we can transfer it
+	 * straight onto a signaled list, and queue the irq worker for
+	 * its signal completion.
+	 */
+	if (__request_completed(rq)) {
+		if (__signal_request(rq, &b->signaled_requests))
+			irq_work_queue(&b->irq_work);
+		return;
+	}
 
+	if (list_empty(&ce->signals)) {
+		add_signaling_context(b, ce);
+		pos = &ce->signals;
+	} else {
 		/*
 		 * We keep the seqno in retirement order, so we can break
 		 * inside intel_engine_signal_breadcrumbs as soon as we've
@@ -351,24 +357,75 @@ bool i915_request_enable_breadcrumb(struct i915_request *rq)
 			if (i915_seqno_passed(rq->fence.seqno, it->fence.seqno))
 				break;
 		}
-		list_add(&rq->signal_link, pos);
-		if (pos == &ce->signals) /* catch transitions from empty list */
-			list_move_tail(&ce->signal_link, &b->signalers);
-		GEM_BUG_ON(!check_signal_order(ce, rq));
+	}
+	list_add(&rq->signal_link, pos);
+	GEM_BUG_ON(!check_signal_order(ce, rq));
+	set_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags);
+
+	/* Check after attaching to irq, interrupt may have already fired. */
+	if (__request_completed(rq))
+		irq_work_queue(&b->irq_work);
+}
 
-		set_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags);
-unlock:
+bool i915_request_enable_breadcrumb(struct i915_request *rq)
+{
+	struct intel_breadcrumbs *b;
+
+	/* Serialises with i915_request_retire() using rq->lock */
+	if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &rq->fence.flags))
+		return true;
+
+	/*
+	 * Peek at i915_request_submit()/i915_request_unsubmit() status.
+	 *
+	 * If the request is not yet active (and not signaled), we will
+	 * attach the breadcrumb later.
+	 */
+	if (!test_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags))
+		return true;
+
+	/*
+	 * rq->engine is locked by rq->engine->active.lock. That however
+	 * is not known until after rq->engine has been dereferenced and
+	 * the lock acquired. Hence we acquire the lock and then validate
+	 * that rq->engine still matches the lock we hold for it.
+	 *
+	 * Here, we are using the breadcrumb lock as a proxy for the
+	 * rq->engine->active.lock, and we know that since the breadcrumb
+	 * will be serialised within i915_request_submit/i915_request_unsubmit,
+	 * the engine cannot change while active as long as we hold the
+	 * breadcrumb lock on that engine.
+	 *
+	 * From the dma_fence_enable_signaling() path, we are outside of the
+	 * request submit/unsubmit path, and so we must be more careful to
+	 * acquire the right lock.
+	 */
+	b = READ_ONCE(rq->engine)->breadcrumbs;
+	spin_lock(&b->irq_lock);
+	while (unlikely(b != READ_ONCE(rq->engine)->breadcrumbs)) {
 		spin_unlock(&b->irq_lock);
+		b = READ_ONCE(rq->engine)->breadcrumbs;
+		spin_lock(&b->irq_lock);
 	}
 
-	return !__request_completed(rq);
+	/*
+	 * Now that we are finally serialised with request submit/unsubmit,
+	 * [with b->irq_lock] and with i915_request_retire() [via checking
+	 * SIGNALED with rq->lock] confirm the request is indeed active. If
+	 * it is no longer active, the breadcrumb will be attached upon
+	 * i915_request_submit().
+	 */
+	if (test_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags))
+		insert_breadcrumb(rq, b);
+
+	spin_unlock(&b->irq_lock);
+
+	return true;
 }
 
 void i915_request_cancel_breadcrumb(struct i915_request *rq)
 {
-	struct intel_breadcrumbs *b = &rq->engine->breadcrumbs;
-
-	lockdep_assert_held(&rq->lock);
+	struct intel_breadcrumbs *b = rq->engine->breadcrumbs;
 
 	/*
 	 * We must wait for b->irq_lock so that we know the interrupt handler
@@ -382,23 +439,19 @@ void i915_request_cancel_breadcrumb(struct i915_request *rq)
 
 		list_del(&rq->signal_link);
 		if (list_empty(&ce->signals))
-			list_del_init(&ce->signal_link);
+			remove_signaling_context(b, ce);
 
 		clear_bit(I915_FENCE_FLAG_SIGNAL, &rq->fence.flags);
+		i915_request_put(rq);
 	}
 	spin_unlock(&b->irq_lock);
 }
 
-void intel_engine_print_breadcrumbs(struct intel_engine_cs *engine,
-				    struct drm_printer *p)
+static void print_signals(struct intel_breadcrumbs *b, struct drm_printer *p)
 {
-	struct intel_breadcrumbs *b = &engine->breadcrumbs;
 	struct intel_context *ce;
 	struct i915_request *rq;
 
-	if (list_empty(&b->signalers))
-		return;
-
 	drm_printf(p, "Signals:\n");
 
 	spin_lock_irq(&b->irq_lock);
@@ -414,3 +467,17 @@ void intel_engine_print_breadcrumbs(struct intel_engine_cs *engine,
 	}
 	spin_unlock_irq(&b->irq_lock);
 }
+
+void intel_engine_print_breadcrumbs(struct intel_engine_cs *engine,
+				    struct drm_printer *p)
+{
+	struct intel_breadcrumbs *b;
+
+	b = engine->breadcrumbs;
+	if (!b)
+		return;
+
+	drm_printf(p, "IRQ: %s\n", enableddisabled(b->irq_armed));
+	if (!list_empty(&b->signalers))
+		print_signals(b, p);
+}
diff --git a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.h b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.h
new file mode 100644
index 000000000000..ed3d1deabfbd
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_BREADCRUMBS__
+#define __INTEL_BREADCRUMBS__
+
+#include <linux/irq_work.h>
+
+#include "intel_engine_types.h"
+
+struct drm_printer;
+struct i915_request;
+struct intel_breadcrumbs;
+
+struct intel_breadcrumbs *
+intel_breadcrumbs_create(struct intel_engine_cs *irq_engine);
+void intel_breadcrumbs_free(struct intel_breadcrumbs *b);
+
+void intel_breadcrumbs_reset(struct intel_breadcrumbs *b);
+void intel_breadcrumbs_park(struct intel_breadcrumbs *b);
+
+static inline void
+intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine)
+{
+	irq_work_queue(&engine->breadcrumbs->irq_work);
+}
+
+void intel_engine_print_breadcrumbs(struct intel_engine_cs *engine,
+				    struct drm_printer *p);
+
+bool i915_request_enable_breadcrumb(struct i915_request *request);
+void i915_request_cancel_breadcrumb(struct i915_request *request);
+
+#endif /* __INTEL_BREADCRUMBS__ */
diff --git a/drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h b/drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h
new file mode 100644
index 000000000000..8e53b9942695
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_BREADCRUMBS_TYPES__
+#define __INTEL_BREADCRUMBS_TYPES__
+
+#include <linux/irq_work.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+
+/*
+ * Rather than have every client wait upon all user interrupts,
+ * with the herd waking after every interrupt and each doing the
+ * heavyweight seqno dance, we delegate the task (of being the
+ * bottom-half of the user interrupt) to the first client. After
+ * every interrupt, we wake up one client, who does the heavyweight
+ * coherent seqno read and either goes back to sleep (if incomplete),
+ * or wakes up all the completed clients in parallel, before then
+ * transferring the bottom-half status to the next client in the queue.
+ *
+ * Compared to walking the entire list of waiters in a single dedicated
+ * bottom-half, we reduce the latency of the first waiter by avoiding
+ * a context switch, but incur additional coherent seqno reads when
+ * following the chain of request breadcrumbs. Since it is most likely
+ * that we have a single client waiting on each seqno, then reducing
+ * the overhead of waking that client is much preferred.
+ */
+struct intel_breadcrumbs {
+	spinlock_t irq_lock; /* protects the lists used in hardirq context */
+
+	/* Not all breadcrumbs are attached to physical HW */
+	struct intel_engine_cs *irq_engine;
+
+	struct list_head signalers;
+	struct list_head signaled_requests;
+
+	struct irq_work irq_work; /* for use from inside irq_lock */
+
+	unsigned int irq_enabled;
+
+	bool irq_armed;
+};
+
+#endif /* __INTEL_BREADCRUMBS_TYPES__ */
diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c
index 52db2bde44a3..92a3f25c4006 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.c
+++ b/drivers/gpu/drm/i915/gt/intel_context.c
@@ -93,57 +93,210 @@ static void intel_context_active_release(struct intel_context *ce)
 	i915_active_release(&ce->active);
 }
 
-int __intel_context_do_pin(struct intel_context *ce)
+static int __context_pin_state(struct i915_vma *vma, struct i915_gem_ww_ctx *ww)
+{
+	unsigned int bias = i915_ggtt_pin_bias(vma) | PIN_OFFSET_BIAS;
+	int err;
+
+	err = i915_ggtt_pin(vma, ww, 0, bias | PIN_HIGH);
+	if (err)
+		return err;
+
+	err = i915_active_acquire(&vma->active);
+	if (err)
+		goto err_unpin;
+
+	/*
+	 * And mark it as a globally pinned object to let the shrinker know
+	 * it cannot reclaim the object until we release it.
+	 */
+	i915_vma_make_unshrinkable(vma);
+	vma->obj->mm.dirty = true;
+
+	return 0;
+
+err_unpin:
+	i915_vma_unpin(vma);
+	return err;
+}
+
+static void __context_unpin_state(struct i915_vma *vma)
+{
+	i915_vma_make_shrinkable(vma);
+	i915_active_release(&vma->active);
+	__i915_vma_unpin(vma);
+}
+
+static int __ring_active(struct intel_ring *ring,
+			 struct i915_gem_ww_ctx *ww)
+{
+	int err;
+
+	err = intel_ring_pin(ring, ww);
+	if (err)
+		return err;
+
+	err = i915_active_acquire(&ring->vma->active);
+	if (err)
+		goto err_pin;
+
+	return 0;
+
+err_pin:
+	intel_ring_unpin(ring);
+	return err;
+}
+
+static void __ring_retire(struct intel_ring *ring)
+{
+	i915_active_release(&ring->vma->active);
+	intel_ring_unpin(ring);
+}
+
+static int intel_context_pre_pin(struct intel_context *ce,
+				 struct i915_gem_ww_ctx *ww)
 {
 	int err;
 
+	CE_TRACE(ce, "active\n");
+
+	err = __ring_active(ce->ring, ww);
+	if (err)
+		return err;
+
+	err = intel_timeline_pin(ce->timeline, ww);
+	if (err)
+		goto err_ring;
+
+	if (!ce->state)
+		return 0;
+
+	err = __context_pin_state(ce->state, ww);
+	if (err)
+		goto err_timeline;
+
+
+	return 0;
+
+err_timeline:
+	intel_timeline_unpin(ce->timeline);
+err_ring:
+	__ring_retire(ce->ring);
+	return err;
+}
+
+static void intel_context_post_unpin(struct intel_context *ce)
+{
+	if (ce->state)
+		__context_unpin_state(ce->state);
+
+	intel_timeline_unpin(ce->timeline);
+	__ring_retire(ce->ring);
+}
+
+int __intel_context_do_pin_ww(struct intel_context *ce,
+			      struct i915_gem_ww_ctx *ww)
+{
+	bool handoff = false;
+	void *vaddr;
+	int err = 0;
+
 	if (unlikely(!test_bit(CONTEXT_ALLOC_BIT, &ce->flags))) {
 		err = intel_context_alloc_state(ce);
 		if (err)
 			return err;
 	}
 
-	err = i915_active_acquire(&ce->active);
+	/*
+	 * We always pin the context/ring/timeline here, to ensure a pin
+	 * refcount for __intel_context_active(), which prevent a lock
+	 * inversion of ce->pin_mutex vs dma_resv_lock().
+	 */
+
+	err = i915_gem_object_lock(ce->timeline->hwsp_ggtt->obj, ww);
+	if (!err && ce->ring->vma->obj)
+		err = i915_gem_object_lock(ce->ring->vma->obj, ww);
+	if (!err && ce->state)
+		err = i915_gem_object_lock(ce->state->obj, ww);
+	if (!err)
+		err = intel_context_pre_pin(ce, ww);
 	if (err)
 		return err;
 
-	if (mutex_lock_interruptible(&ce->pin_mutex)) {
-		err = -EINTR;
-		goto out_release;
-	}
+	err = i915_active_acquire(&ce->active);
+	if (err)
+		goto err_ctx_unpin;
+
+	err = ce->ops->pre_pin(ce, ww, &vaddr);
+	if (err)
+		goto err_release;
+
+	err = mutex_lock_interruptible(&ce->pin_mutex);
+	if (err)
+		goto err_post_unpin;
 
 	if (unlikely(intel_context_is_closed(ce))) {
 		err = -ENOENT;
-		goto out_unlock;
+		goto err_unlock;
 	}
 
 	if (likely(!atomic_add_unless(&ce->pin_count, 1, 0))) {
 		err = intel_context_active_acquire(ce);
 		if (unlikely(err))
-			goto out_unlock;
+			goto err_unlock;
 
-		err = ce->ops->pin(ce);
-		if (unlikely(err))
-			goto err_active;
+		err = ce->ops->pin(ce, vaddr);
+		if (err) {
+			intel_context_active_release(ce);
+			goto err_unlock;
+		}
 
 		CE_TRACE(ce, "pin ring:{start:%08x, head:%04x, tail:%04x}\n",
 			 i915_ggtt_offset(ce->ring->vma),
 			 ce->ring->head, ce->ring->tail);
 
+		handoff = true;
 		smp_mb__before_atomic(); /* flush pin before it is visible */
 		atomic_inc(&ce->pin_count);
 	}
 
 	GEM_BUG_ON(!intel_context_is_pinned(ce)); /* no overflow! */
-	GEM_BUG_ON(i915_active_is_idle(&ce->active));
-	goto out_unlock;
 
-err_active:
-	intel_context_active_release(ce);
-out_unlock:
+err_unlock:
 	mutex_unlock(&ce->pin_mutex);
-out_release:
+err_post_unpin:
+	if (!handoff)
+		ce->ops->post_unpin(ce);
+err_release:
 	i915_active_release(&ce->active);
+err_ctx_unpin:
+	intel_context_post_unpin(ce);
+
+	/*
+	 * Unlock the hwsp_ggtt object since it's shared.
+	 * In principle we can unlock all the global state locked above
+	 * since it's pinned and doesn't need fencing, and will
+	 * thus remain resident until it is explicitly unpinned.
+	 */
+	i915_gem_ww_unlock_single(ce->timeline->hwsp_ggtt->obj);
+
+	return err;
+}
+
+int __intel_context_do_pin(struct intel_context *ce)
+{
+	struct i915_gem_ww_ctx ww;
+	int err;
+
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	err = __intel_context_do_pin_ww(ce, &ww);
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
 	return err;
 }
 
@@ -154,6 +307,7 @@ void intel_context_unpin(struct intel_context *ce)
 
 	CE_TRACE(ce, "unpin\n");
 	ce->ops->unpin(ce);
+	ce->ops->post_unpin(ce);
 
 	/*
 	 * Once released, we may asynchronously drop the active reference.
@@ -166,65 +320,6 @@ void intel_context_unpin(struct intel_context *ce)
 	intel_context_put(ce);
 }
 
-static int __context_pin_state(struct i915_vma *vma)
-{
-	unsigned int bias = i915_ggtt_pin_bias(vma) | PIN_OFFSET_BIAS;
-	int err;
-
-	err = i915_ggtt_pin(vma, 0, bias | PIN_HIGH);
-	if (err)
-		return err;
-
-	err = i915_active_acquire(&vma->active);
-	if (err)
-		goto err_unpin;
-
-	/*
-	 * And mark it as a globally pinned object to let the shrinker know
-	 * it cannot reclaim the object until we release it.
-	 */
-	i915_vma_make_unshrinkable(vma);
-	vma->obj->mm.dirty = true;
-
-	return 0;
-
-err_unpin:
-	i915_vma_unpin(vma);
-	return err;
-}
-
-static void __context_unpin_state(struct i915_vma *vma)
-{
-	i915_vma_make_shrinkable(vma);
-	i915_active_release(&vma->active);
-	__i915_vma_unpin(vma);
-}
-
-static int __ring_active(struct intel_ring *ring)
-{
-	int err;
-
-	err = intel_ring_pin(ring);
-	if (err)
-		return err;
-
-	err = i915_active_acquire(&ring->vma->active);
-	if (err)
-		goto err_pin;
-
-	return 0;
-
-err_pin:
-	intel_ring_unpin(ring);
-	return err;
-}
-
-static void __ring_retire(struct intel_ring *ring)
-{
-	i915_active_release(&ring->vma->active);
-	intel_ring_unpin(ring);
-}
-
 __i915_active_call
 static void __intel_context_retire(struct i915_active *active)
 {
@@ -235,48 +330,29 @@ static void __intel_context_retire(struct i915_active *active)
 		 intel_context_get_avg_runtime_ns(ce));
 
 	set_bit(CONTEXT_VALID_BIT, &ce->flags);
-	if (ce->state)
-		__context_unpin_state(ce->state);
-
-	intel_timeline_unpin(ce->timeline);
-	__ring_retire(ce->ring);
-
+	intel_context_post_unpin(ce);
 	intel_context_put(ce);
 }
 
 static int __intel_context_active(struct i915_active *active)
 {
 	struct intel_context *ce = container_of(active, typeof(*ce), active);
-	int err;
-
-	CE_TRACE(ce, "active\n");
 
 	intel_context_get(ce);
 
-	err = __ring_active(ce->ring);
-	if (err)
-		goto err_put;
+	/* everything should already be activated by intel_context_pre_pin() */
+	GEM_WARN_ON(!i915_active_acquire_if_busy(&ce->ring->vma->active));
+	__intel_ring_pin(ce->ring);
 
-	err = intel_timeline_pin(ce->timeline);
-	if (err)
-		goto err_ring;
+	__intel_timeline_pin(ce->timeline);
 
-	if (!ce->state)
-		return 0;
-
-	err = __context_pin_state(ce->state);
-	if (err)
-		goto err_timeline;
+	if (ce->state) {
+		GEM_WARN_ON(!i915_active_acquire_if_busy(&ce->state->active));
+		__i915_vma_pin(ce->state);
+		i915_vma_make_unshrinkable(ce->state);
+	}
 
 	return 0;
-
-err_timeline:
-	intel_timeline_unpin(ce->timeline);
-err_ring:
-	__ring_retire(ce->ring);
-err_put:
-	intel_context_put(ce);
-	return err;
 }
 
 void
@@ -382,15 +458,38 @@ int intel_context_prepare_remote_request(struct intel_context *ce,
 
 struct i915_request *intel_context_create_request(struct intel_context *ce)
 {
+	struct i915_gem_ww_ctx ww;
 	struct i915_request *rq;
 	int err;
 
-	err = intel_context_pin(ce);
-	if (unlikely(err))
-		return ERR_PTR(err);
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	err = intel_context_pin_ww(ce, &ww);
+	if (!err) {
+		rq = i915_request_create(ce);
+		intel_context_unpin(ce);
+	} else if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+		rq = ERR_PTR(err);
+	} else {
+		rq = ERR_PTR(err);
+	}
+
+	i915_gem_ww_ctx_fini(&ww);
 
-	rq = i915_request_create(ce);
-	intel_context_unpin(ce);
+	if (IS_ERR(rq))
+		return rq;
+
+	/*
+	 * timeline->mutex should be the inner lock, but is used as outer lock.
+	 * Hack around this to shut up lockdep in selftests..
+	 */
+	lockdep_unpin_lock(&ce->timeline->mutex, rq->cookie);
+	mutex_release(&ce->timeline->mutex.dep_map, _RET_IP_);
+	mutex_acquire(&ce->timeline->mutex.dep_map, SINGLE_DEPTH_NESTING, 0, _RET_IP_);
+	rq->cookie = lockdep_pin_lock(&ce->timeline->mutex);
 
 	return rq;
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h
index 07be021882cc..fda2eba81e22 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.h
+++ b/drivers/gpu/drm/i915/gt/intel_context.h
@@ -25,6 +25,8 @@
 		     ##__VA_ARGS__);					\
 } while (0)
 
+struct i915_gem_ww_ctx;
+
 void intel_context_init(struct intel_context *ce,
 			struct intel_engine_cs *engine);
 void intel_context_fini(struct intel_context *ce);
@@ -81,6 +83,8 @@ static inline void intel_context_unlock_pinned(struct intel_context *ce)
 }
 
 int __intel_context_do_pin(struct intel_context *ce);
+int __intel_context_do_pin_ww(struct intel_context *ce,
+			      struct i915_gem_ww_ctx *ww);
 
 static inline bool intel_context_pin_if_active(struct intel_context *ce)
 {
@@ -95,6 +99,15 @@ static inline int intel_context_pin(struct intel_context *ce)
 	return __intel_context_do_pin(ce);
 }
 
+static inline int intel_context_pin_ww(struct intel_context *ce,
+				       struct i915_gem_ww_ctx *ww)
+{
+	if (likely(intel_context_pin_if_active(ce)))
+		return 0;
+
+	return __intel_context_do_pin_ww(ce, ww);
+}
+
 static inline void __intel_context_pin(struct intel_context *ce)
 {
 	GEM_BUG_ON(!intel_context_is_pinned(ce));
diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
index 4954b0df4864..552cb57a2e8c 100644
--- a/drivers/gpu/drm/i915/gt/intel_context_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
@@ -23,6 +23,7 @@
 DECLARE_EWMA(runtime, 3, 8);
 
 struct i915_gem_context;
+struct i915_gem_ww_ctx;
 struct i915_vma;
 struct intel_context;
 struct intel_ring;
@@ -30,8 +31,10 @@ struct intel_ring;
 struct intel_context_ops {
 	int (*alloc)(struct intel_context *ce);
 
-	int (*pin)(struct intel_context *ce);
+	int (*pre_pin)(struct intel_context *ce, struct i915_gem_ww_ctx *ww, void **vaddr);
+	int (*pin)(struct intel_context *ce, void *vaddr);
 	void (*unpin)(struct intel_context *ce);
+	void (*post_unpin)(struct intel_context *ce);
 
 	void (*enter)(struct intel_context *ce);
 	void (*exit)(struct intel_context *ce);
diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h
index a9249a23903a..7c3a1012e702 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine.h
@@ -223,26 +223,6 @@ void intel_engine_get_instdone(const struct intel_engine_cs *engine,
 
 void intel_engine_init_execlists(struct intel_engine_cs *engine);
 
-void intel_engine_init_breadcrumbs(struct intel_engine_cs *engine);
-void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
-
-void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine);
-
-static inline void
-intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine)
-{
-	irq_work_queue(&engine->breadcrumbs.irq_work);
-}
-
-void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine);
-void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
-
-void intel_engine_transfer_stale_breadcrumbs(struct intel_engine_cs *engine,
-					     struct intel_context *ce);
-
-void intel_engine_print_breadcrumbs(struct intel_engine_cs *engine,
-				    struct drm_printer *p);
-
 static inline u32 *__gen8_emit_pipe_control(u32 *batch, u32 flags0, u32 flags1, u32 offset)
 {
 	memset(batch, 0, 6 * sizeof(u32));
@@ -357,4 +337,13 @@ intel_engine_has_preempt_reset(const struct intel_engine_cs *engine)
 	return intel_engine_has_preemption(engine);
 }
 
+static inline bool
+intel_engine_has_heartbeat(const struct intel_engine_cs *engine)
+{
+	if (!IS_ACTIVE(CONFIG_DRM_I915_HEARTBEAT_INTERVAL))
+		return false;
+
+	return READ_ONCE(engine->props.heartbeat_interval_ms);
+}
+
 #endif /* _INTEL_RINGBUFFER_H_ */
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 26087dd79782..5bfb5f7ed02c 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -28,6 +28,7 @@
 
 #include "i915_drv.h"
 
+#include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_engine.h"
 #include "intel_engine_pm.h"
@@ -634,7 +635,7 @@ static int pin_ggtt_status_page(struct intel_engine_cs *engine,
 	else
 		flags = PIN_HIGH;
 
-	return i915_ggtt_pin(vma, 0, flags);
+	return i915_ggtt_pin(vma, NULL, 0, flags);
 }
 
 static int init_status_page(struct intel_engine_cs *engine)
@@ -700,8 +701,13 @@ static int engine_setup_common(struct intel_engine_cs *engine)
 	if (err)
 		return err;
 
+	engine->breadcrumbs = intel_breadcrumbs_create(engine);
+	if (!engine->breadcrumbs) {
+		err = -ENOMEM;
+		goto err_status;
+	}
+
 	intel_engine_init_active(engine, ENGINE_PHYSICAL);
-	intel_engine_init_breadcrumbs(engine);
 	intel_engine_init_execlists(engine);
 	intel_engine_init_cmd_parser(engine);
 	intel_engine_init__pm(engine);
@@ -716,6 +722,10 @@ static int engine_setup_common(struct intel_engine_cs *engine)
 	intel_engine_init_ctx_wa(engine);
 
 	return 0;
+
+err_status:
+	cleanup_status_page(engine);
+	return err;
 }
 
 struct measure_breadcrumb {
@@ -785,9 +795,11 @@ intel_engine_init_active(struct intel_engine_cs *engine, unsigned int subclass)
 }
 
 static struct intel_context *
-create_kernel_context(struct intel_engine_cs *engine)
+create_pinned_context(struct intel_engine_cs *engine,
+		      unsigned int hwsp,
+		      struct lock_class_key *key,
+		      const char *name)
 {
-	static struct lock_class_key kernel;
 	struct intel_context *ce;
 	int err;
 
@@ -796,6 +808,7 @@ create_kernel_context(struct intel_engine_cs *engine)
 		return ce;
 
 	__set_bit(CONTEXT_BARRIER_BIT, &ce->flags);
+	ce->timeline = page_pack_bits(NULL, hwsp);
 
 	err = intel_context_pin(ce); /* perma-pin so it is always available */
 	if (err) {
@@ -809,11 +822,20 @@ create_kernel_context(struct intel_engine_cs *engine)
 	 * should we need to inject GPU operations during their request
 	 * construction.
 	 */
-	lockdep_set_class(&ce->timeline->mutex, &kernel);
+	lockdep_set_class_and_name(&ce->timeline->mutex, key, name);
 
 	return ce;
 }
 
+static struct intel_context *
+create_kernel_context(struct intel_engine_cs *engine)
+{
+	static struct lock_class_key kernel;
+
+	return create_pinned_context(engine, I915_GEM_HWS_SEQNO_ADDR,
+				     &kernel, "kernel_context");
+}
+
 /**
  * intel_engines_init_common - initialize cengine state which might require hw access
  * @engine: Engine to initialize.
@@ -902,9 +924,9 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)
 	tasklet_kill(&engine->execlists.tasklet); /* flush the callback */
 
 	cleanup_status_page(engine);
+	intel_breadcrumbs_free(engine->breadcrumbs);
 
 	intel_engine_fini_retire(engine);
-	intel_engine_fini_breadcrumbs(engine);
 	intel_engine_cleanup_cmd_parser(engine);
 
 	if (engine->default_state)
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c b/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c
index 8ffdf676c0a0..5067d0524d4b 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c
@@ -177,36 +177,82 @@ void intel_engine_init_heartbeat(struct intel_engine_cs *engine)
 	INIT_DELAYED_WORK(&engine->heartbeat.work, heartbeat);
 }
 
+static int __intel_engine_pulse(struct intel_engine_cs *engine)
+{
+	struct i915_sched_attr attr = { .priority = I915_PRIORITY_BARRIER };
+	struct intel_context *ce = engine->kernel_context;
+	struct i915_request *rq;
+
+	lockdep_assert_held(&ce->timeline->mutex);
+	GEM_BUG_ON(!intel_engine_has_preemption(engine));
+	GEM_BUG_ON(!intel_engine_pm_is_awake(engine));
+
+	intel_context_enter(ce);
+	rq = __i915_request_create(ce, GFP_NOWAIT | __GFP_NOWARN);
+	intel_context_exit(ce);
+	if (IS_ERR(rq))
+		return PTR_ERR(rq);
+
+	__set_bit(I915_FENCE_FLAG_SENTINEL, &rq->fence.flags);
+	idle_pulse(engine, rq);
+
+	__i915_request_commit(rq);
+	__i915_request_queue(rq, &attr);
+	GEM_BUG_ON(rq->sched.attr.priority < I915_PRIORITY_BARRIER);
+
+	return 0;
+}
+
+static unsigned long set_heartbeat(struct intel_engine_cs *engine,
+				   unsigned long delay)
+{
+	unsigned long old;
+
+	old = xchg(&engine->props.heartbeat_interval_ms, delay);
+	if (delay)
+		intel_engine_unpark_heartbeat(engine);
+	else
+		intel_engine_park_heartbeat(engine);
+
+	return old;
+}
+
 int intel_engine_set_heartbeat(struct intel_engine_cs *engine,
 			       unsigned long delay)
 {
-	int err;
+	struct intel_context *ce = engine->kernel_context;
+	int err = 0;
 
-	/* Send one last pulse before to cleanup persistent hogs */
-	if (!delay && IS_ACTIVE(CONFIG_DRM_I915_PREEMPT_TIMEOUT)) {
-		err = intel_engine_pulse(engine);
-		if (err)
-			return err;
-	}
+	if (!delay && !intel_engine_has_preempt_reset(engine))
+		return -ENODEV;
+
+	intel_engine_pm_get(engine);
+
+	err = mutex_lock_interruptible(&ce->timeline->mutex);
+	if (err)
+		goto out_rpm;
 
-	WRITE_ONCE(engine->props.heartbeat_interval_ms, delay);
+	if (delay != engine->props.heartbeat_interval_ms) {
+		unsigned long saved = set_heartbeat(engine, delay);
 
-	if (intel_engine_pm_get_if_awake(engine)) {
-		if (delay)
-			intel_engine_unpark_heartbeat(engine);
-		else
-			intel_engine_park_heartbeat(engine);
-		intel_engine_pm_put(engine);
+		/* recheck current execution */
+		if (intel_engine_has_preemption(engine)) {
+			err = __intel_engine_pulse(engine);
+			if (err)
+				set_heartbeat(engine, saved);
+		}
 	}
 
-	return 0;
+	mutex_unlock(&ce->timeline->mutex);
+
+out_rpm:
+	intel_engine_pm_put(engine);
+	return err;
 }
 
 int intel_engine_pulse(struct intel_engine_cs *engine)
 {
-	struct i915_sched_attr attr = { .priority = I915_PRIORITY_BARRIER };
 	struct intel_context *ce = engine->kernel_context;
-	struct i915_request *rq;
 	int err;
 
 	if (!intel_engine_has_preemption(engine))
@@ -215,30 +261,12 @@ int intel_engine_pulse(struct intel_engine_cs *engine)
 	if (!intel_engine_pm_get_if_awake(engine))
 		return 0;
 
-	if (mutex_lock_interruptible(&ce->timeline->mutex)) {
-		err = -EINTR;
-		goto out_rpm;
-	}
-
-	intel_context_enter(ce);
-	rq = __i915_request_create(ce, GFP_NOWAIT | __GFP_NOWARN);
-	intel_context_exit(ce);
-	if (IS_ERR(rq)) {
-		err = PTR_ERR(rq);
-		goto out_unlock;
+	err = -EINTR;
+	if (!mutex_lock_interruptible(&ce->timeline->mutex)) {
+		err = __intel_engine_pulse(engine);
+		mutex_unlock(&ce->timeline->mutex);
 	}
 
-	__set_bit(I915_FENCE_FLAG_SENTINEL, &rq->fence.flags);
-	idle_pulse(engine, rq);
-
-	__i915_request_commit(rq);
-	__i915_request_queue(rq, &attr);
-	GEM_BUG_ON(rq->sched.attr.priority < I915_PRIORITY_BARRIER);
-	err = 0;
-
-out_unlock:
-	mutex_unlock(&ce->timeline->mutex);
-out_rpm:
 	intel_engine_pm_put(engine);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
index 8ec3eecf3e39..f7b2e07e2229 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
@@ -6,6 +6,7 @@
 
 #include "i915_drv.h"
 
+#include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_engine.h"
 #include "intel_engine_heartbeat.h"
@@ -247,7 +248,7 @@ static int __engine_park(struct intel_wakeref *wf)
 	call_idle_barriers(engine); /* cleanup after wedging */
 
 	intel_engine_park_heartbeat(engine);
-	intel_engine_disarm_breadcrumbs(engine);
+	intel_breadcrumbs_park(engine->breadcrumbs);
 
 	/* Must be reset upon idling, or we may miss the busy wakeup. */
 	GEM_BUG_ON(engine->execlists.queue_priority_hint != INT_MIN);
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 8de92fd7d392..c400aaa2287b 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -22,6 +22,7 @@
 #include "i915_pmu.h"
 #include "i915_priolist_types.h"
 #include "i915_selftest.h"
+#include "intel_breadcrumbs_types.h"
 #include "intel_sseu.h"
 #include "intel_timeline_types.h"
 #include "intel_uncore.h"
@@ -373,34 +374,8 @@ struct intel_engine_cs {
 	 */
 	struct ewma__engine_latency latency;
 
-	/* Rather than have every client wait upon all user interrupts,
-	 * with the herd waking after every interrupt and each doing the
-	 * heavyweight seqno dance, we delegate the task (of being the
-	 * bottom-half of the user interrupt) to the first client. After
-	 * every interrupt, we wake up one client, who does the heavyweight
-	 * coherent seqno read and either goes back to sleep (if incomplete),
-	 * or wakes up all the completed clients in parallel, before then
-	 * transferring the bottom-half status to the next client in the queue.
-	 *
-	 * Compared to walking the entire list of waiters in a single dedicated
-	 * bottom-half, we reduce the latency of the first waiter by avoiding
-	 * a context switch, but incur additional coherent seqno reads when
-	 * following the chain of request breadcrumbs. Since it is most likely
-	 * that we have a single client waiting on each seqno, then reducing
-	 * the overhead of waking that client is much preferred.
-	 */
-	struct intel_breadcrumbs {
-		spinlock_t irq_lock;
-		struct list_head signalers;
-
-		struct list_head signaled_requests;
-
-		struct irq_work irq_work; /* for use from inside irq_lock */
-
-		unsigned int irq_enabled;
-
-		bool irq_armed;
-	} breadcrumbs;
+	/* Keep track of all the seqno used, a trail of breadcrumbs */
+	struct intel_breadcrumbs *breadcrumbs;
 
 	struct intel_engine_pmu {
 		/**
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index 99e28d9021e8..81c05f551b9c 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -78,8 +78,6 @@ int i915_ggtt_init_hw(struct drm_i915_private *i915)
 {
 	int ret;
 
-	stash_init(&i915->mm.wc_stash);
-
 	/*
 	 * Note that we use page colouring to enforce a guard page at the
 	 * end of the address space. This is required as the CS may prefetch
@@ -232,7 +230,7 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 
 	/* Fill the allocated but "unused" space beyond the end of the buffer */
 	while (gte < end)
-		gen8_set_pte(gte++, vm->scratch[0].encode);
+		gen8_set_pte(gte++, vm->scratch[0]->encode);
 
 	/*
 	 * We want to flush the TLBs only after we're certain all the PTE
@@ -283,7 +281,7 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 
 	/* Fill the allocated but "unused" space beyond the end of the buffer */
 	while (gte < end)
-		iowrite32(vm->scratch[0].encode, gte++);
+		iowrite32(vm->scratch[0]->encode, gte++);
 
 	/*
 	 * We want to flush the TLBs only after we're certain all the PTE
@@ -303,7 +301,7 @@ static void gen8_ggtt_clear_range(struct i915_address_space *vm,
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 	unsigned int first_entry = start / I915_GTT_PAGE_SIZE;
 	unsigned int num_entries = length / I915_GTT_PAGE_SIZE;
-	const gen8_pte_t scratch_pte = vm->scratch[0].encode;
+	const gen8_pte_t scratch_pte = vm->scratch[0]->encode;
 	gen8_pte_t __iomem *gtt_base =
 		(gen8_pte_t __iomem *)ggtt->gsm + first_entry;
 	const int max_entries = ggtt_total_entries(ggtt) - first_entry;
@@ -401,7 +399,7 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 		 first_entry, num_entries, max_entries))
 		num_entries = max_entries;
 
-	scratch_pte = vm->scratch[0].encode;
+	scratch_pte = vm->scratch[0]->encode;
 	for (i = 0; i < num_entries; i++)
 		iowrite32(scratch_pte, &gtt_base[i]);
 }
@@ -436,16 +434,17 @@ static void i915_ggtt_clear_range(struct i915_address_space *vm,
 	intel_gtt_clear_range(start >> PAGE_SHIFT, length >> PAGE_SHIFT);
 }
 
-static int ggtt_bind_vma(struct i915_address_space *vm,
-			 struct i915_vma *vma,
-			 enum i915_cache_level cache_level,
-			 u32 flags)
+static void ggtt_bind_vma(struct i915_address_space *vm,
+			  struct i915_vm_pt_stash *stash,
+			  struct i915_vma *vma,
+			  enum i915_cache_level cache_level,
+			  u32 flags)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
 	u32 pte_flags;
 
 	if (i915_vma_is_bound(vma, ~flags & I915_VMA_BIND_MASK))
-		return 0;
+		return;
 
 	/* Applicable to VLV (gen8+ do not support RO in the GGTT) */
 	pte_flags = 0;
@@ -454,8 +453,6 @@ static int ggtt_bind_vma(struct i915_address_space *vm,
 
 	vm->insert_entries(vm, vma, cache_level, pte_flags);
 	vma->page_sizes.gtt = I915_GTT_PAGE_SIZE;
-
-	return 0;
 }
 
 static void ggtt_unbind_vma(struct i915_address_space *vm, struct i915_vma *vma)
@@ -568,31 +565,25 @@ err:
 	return ret;
 }
 
-static int aliasing_gtt_bind_vma(struct i915_address_space *vm,
-				 struct i915_vma *vma,
-				 enum i915_cache_level cache_level,
-				 u32 flags)
+static void aliasing_gtt_bind_vma(struct i915_address_space *vm,
+				  struct i915_vm_pt_stash *stash,
+				  struct i915_vma *vma,
+				  enum i915_cache_level cache_level,
+				  u32 flags)
 {
 	u32 pte_flags;
-	int ret;
 
 	/* Currently applicable only to VLV */
 	pte_flags = 0;
 	if (i915_gem_object_is_readonly(vma->obj))
 		pte_flags |= PTE_READ_ONLY;
 
-	if (flags & I915_VMA_LOCAL_BIND) {
-		struct i915_ppgtt *alias = i915_vm_to_ggtt(vm)->alias;
-
-		ret = ppgtt_bind_vma(&alias->vm, vma, cache_level, flags);
-		if (ret)
-			return ret;
-	}
+	if (flags & I915_VMA_LOCAL_BIND)
+		ppgtt_bind_vma(&i915_vm_to_ggtt(vm)->alias->vm,
+			       stash, vma, cache_level, flags);
 
 	if (flags & I915_VMA_GLOBAL_BIND)
 		vm->insert_entries(vm, vma, cache_level, pte_flags);
-
-	return 0;
 }
 
 static void aliasing_gtt_unbind_vma(struct i915_address_space *vm,
@@ -607,6 +598,7 @@ static void aliasing_gtt_unbind_vma(struct i915_address_space *vm,
 
 static int init_aliasing_ppgtt(struct i915_ggtt *ggtt)
 {
+	struct i915_vm_pt_stash stash = {};
 	struct i915_ppgtt *ppgtt;
 	int err;
 
@@ -619,15 +611,21 @@ static int init_aliasing_ppgtt(struct i915_ggtt *ggtt)
 		goto err_ppgtt;
 	}
 
+	err = i915_vm_alloc_pt_stash(&ppgtt->vm, &stash, ggtt->vm.total);
+	if (err)
+		goto err_ppgtt;
+
+	err = i915_vm_pin_pt_stash(&ppgtt->vm, &stash);
+	if (err)
+		goto err_stash;
+
 	/*
 	 * Note we only pre-allocate as far as the end of the global
 	 * GTT. On 48b / 4-level page-tables, the difference is very,
 	 * very significant! We have to preallocate as GVT/vgpu does
 	 * not like the page directory disappearing.
 	 */
-	err = ppgtt->vm.allocate_va_range(&ppgtt->vm, 0, ggtt->vm.total);
-	if (err)
-		goto err_ppgtt;
+	ppgtt->vm.allocate_va_range(&ppgtt->vm, &stash, 0, ggtt->vm.total);
 
 	ggtt->alias = ppgtt;
 	ggtt->vm.bind_async_flags |= ppgtt->vm.bind_async_flags;
@@ -638,8 +636,11 @@ static int init_aliasing_ppgtt(struct i915_ggtt *ggtt)
 	GEM_BUG_ON(ggtt->vm.vma_ops.unbind_vma != ggtt_unbind_vma);
 	ggtt->vm.vma_ops.unbind_vma = aliasing_gtt_unbind_vma;
 
+	i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 	return 0;
 
+err_stash:
+	i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 err_ppgtt:
 	i915_vm_put(&ppgtt->vm);
 	return err;
@@ -715,18 +716,11 @@ static void ggtt_cleanup_hw(struct i915_ggtt *ggtt)
 void i915_ggtt_driver_release(struct drm_i915_private *i915)
 {
 	struct i915_ggtt *ggtt = &i915->ggtt;
-	struct pagevec *pvec;
 
 	fini_aliasing_ppgtt(ggtt);
 
 	intel_ggtt_fini_fences(ggtt);
 	ggtt_cleanup_hw(ggtt);
-
-	pvec = &i915->mm.wc_stash.pvec;
-	if (pvec->nr) {
-		set_pages_array_wb(pvec->pages, pvec->nr);
-		__pagevec_release(pvec);
-	}
 }
 
 static unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
@@ -789,7 +783,7 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size)
 		return -ENOMEM;
 	}
 
-	ret = setup_scratch_page(&ggtt->vm, GFP_DMA32);
+	ret = setup_scratch_page(&ggtt->vm);
 	if (ret) {
 		drm_err(&i915->drm, "Scratch setup failed\n");
 		/* iounmap will also get called at remove, but meh */
@@ -797,8 +791,8 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size)
 		return ret;
 	}
 
-	ggtt->vm.scratch[0].encode =
-		ggtt->vm.pte_encode(px_dma(&ggtt->vm.scratch[0]),
+	ggtt->vm.scratch[0]->encode =
+		ggtt->vm.pte_encode(px_dma(ggtt->vm.scratch[0]),
 				    I915_CACHE_NONE, 0);
 
 	return 0;
@@ -824,7 +818,7 @@ static void gen6_gmch_remove(struct i915_address_space *vm)
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 
 	iounmap(ggtt->gsm);
-	cleanup_scratch_page(vm);
+	free_scratch(vm);
 }
 
 static struct resource pci_resource(struct pci_dev *pdev, int bar)
@@ -852,6 +846,8 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	else
 		size = gen8_get_total_gtt_size(snb_gmch_ctl);
 
+	ggtt->vm.alloc_pt_dma = alloc_pt_dma;
+
 	ggtt->vm.total = (size / sizeof(gen8_pte_t)) * I915_GTT_PAGE_SIZE;
 	ggtt->vm.cleanup = gen6_gmch_remove;
 	ggtt->vm.insert_page = gen8_ggtt_insert_page;
@@ -1000,6 +996,8 @@ static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 	size = gen6_get_total_gtt_size(snb_gmch_ctl);
 	ggtt->vm.total = (size / sizeof(gen6_pte_t)) * I915_GTT_PAGE_SIZE;
 
+	ggtt->vm.alloc_pt_dma = alloc_pt_dma;
+
 	ggtt->vm.clear_range = nop_clear_range;
 	if (!HAS_FULL_PPGTT(i915) || intel_scanout_needs_vtd_wa(i915))
 		ggtt->vm.clear_range = gen6_ggtt_clear_range;
@@ -1050,6 +1048,8 @@ static int i915_gmch_probe(struct i915_ggtt *ggtt)
 	ggtt->gmadr =
 		(struct resource)DEFINE_RES_MEM(gmadr_base, ggtt->mappable_end);
 
+	ggtt->vm.alloc_pt_dma = alloc_pt_dma;
+
 	ggtt->do_idle_maps = needs_idle_maps(i915);
 	ggtt->vm.insert_page = i915_ggtt_insert_page;
 	ggtt->vm.insert_entries = i915_ggtt_insert_entries;
@@ -1165,11 +1165,6 @@ void i915_ggtt_disable_guc(struct i915_ggtt *ggtt)
 	ggtt->invalidate(ggtt);
 }
 
-static unsigned int clear_bind(struct i915_vma *vma)
-{
-	return atomic_fetch_and(~I915_VMA_BIND_MASK, &vma->flags);
-}
-
 void i915_ggtt_resume(struct i915_ggtt *ggtt)
 {
 	struct i915_vma *vma;
@@ -1187,11 +1182,13 @@ void i915_ggtt_resume(struct i915_ggtt *ggtt)
 	/* clflush objects bound into the GGTT and rebind them. */
 	list_for_each_entry(vma, &ggtt->vm.bound_list, vm_link) {
 		struct drm_i915_gem_object *obj = vma->obj;
-		unsigned int was_bound = clear_bind(vma);
+		unsigned int was_bound =
+			atomic_read(&vma->flags) & I915_VMA_BIND_MASK;
 
-		WARN_ON(i915_vma_bind(vma,
-				      obj ? obj->cache_level : 0,
-				      was_bound, NULL));
+		GEM_BUG_ON(!was_bound);
+		vma->ops->bind_vma(&ggtt->vm, NULL, vma,
+				   obj ? obj->cache_level : 0,
+				   was_bound);
 		if (obj) { /* only used during resume => exclusive access */
 			flush |= fetch_and_zero(&obj->write_domain);
 			obj->read_domains |= I915_GEM_DOMAIN_GTT;
diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
index e0755f1a904b..39b428c5049c 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt.c
@@ -356,7 +356,7 @@ static int intel_gt_init_scratch(struct intel_gt *gt, unsigned int size)
 		goto err_unref;
 	}
 
-	ret = i915_ggtt_pin(vma, 0, PIN_HIGH);
+	ret = i915_ggtt_pin(vma, NULL, 0, PIN_HIGH);
 	if (ret)
 		goto err_unref;
 
@@ -406,21 +406,20 @@ static int __engines_record_defaults(struct intel_gt *gt)
 		/* We must be able to switch to something! */
 		GEM_BUG_ON(!engine->kernel_context);
 
-		err = intel_renderstate_init(&so, engine);
-		if (err)
-			goto out;
-
 		ce = intel_context_create(engine);
 		if (IS_ERR(ce)) {
 			err = PTR_ERR(ce);
 			goto out;
 		}
 
-		rq = intel_context_create_request(ce);
+		err = intel_renderstate_init(&so, ce);
+		if (err)
+			goto err;
+
+		rq = i915_request_create(ce);
 		if (IS_ERR(rq)) {
 			err = PTR_ERR(rq);
-			intel_context_put(ce);
-			goto out;
+			goto err_fini;
 		}
 
 		err = intel_engine_emit_ctx_wa(rq);
@@ -434,9 +433,13 @@ static int __engines_record_defaults(struct intel_gt *gt)
 err_rq:
 		requests[id] = i915_request_get(rq);
 		i915_request_add(rq);
-		intel_renderstate_fini(&so);
-		if (err)
+err_fini:
+		intel_renderstate_fini(&so, ce);
+err:
+		if (err) {
+			intel_context_put(ce);
 			goto out;
+		}
 	}
 
 	/* Flush the default context image to memory, and enable powersaving. */
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
index 418ae184cecf..104cb30e8c13 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
@@ -35,39 +35,65 @@ static void node_free(struct intel_gt_buffer_pool_node *node)
 {
 	i915_gem_object_put(node->obj);
 	i915_active_fini(&node->active);
-	kfree(node);
+	kfree_rcu(node, rcu);
 }
 
-static void pool_free_work(struct work_struct *wrk)
+static bool pool_free_older_than(struct intel_gt_buffer_pool *pool, long keep)
 {
-	struct intel_gt_buffer_pool *pool =
-		container_of(wrk, typeof(*pool), work.work);
-	struct intel_gt_buffer_pool_node *node, *next;
-	unsigned long old = jiffies - HZ;
+	struct intel_gt_buffer_pool_node *node, *stale = NULL;
 	bool active = false;
-	LIST_HEAD(stale);
 	int n;
 
 	/* Free buffers that have not been used in the past second */
-	spin_lock_irq(&pool->lock);
 	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++) {
 		struct list_head *list = &pool->cache_list[n];
 
-		/* Most recent at head; oldest at tail */
-		list_for_each_entry_safe_reverse(node, next, list, link) {
-			if (time_before(node->age, old))
-				break;
+		if (list_empty(list))
+			continue;
+
+		if (spin_trylock_irq(&pool->lock)) {
+			struct list_head *pos;
+
+			/* Most recent at head; oldest at tail */
+			list_for_each_prev(pos, list) {
+				unsigned long age;
+
+				node = list_entry(pos, typeof(*node), link);
+
+				age = READ_ONCE(node->age);
+				if (!age || jiffies - age < keep)
+					break;
+
+				/* Check we are the first to claim this node */
+				if (!xchg(&node->age, 0))
+					break;
 
-			list_move(&node->link, &stale);
+				node->free = stale;
+				stale = node;
+			}
+			if (!list_is_last(pos, list))
+				__list_del_many(pos, list);
+
+			spin_unlock_irq(&pool->lock);
 		}
+
 		active |= !list_empty(list);
 	}
-	spin_unlock_irq(&pool->lock);
 
-	list_for_each_entry_safe(node, next, &stale, link)
+	while ((node = stale)) {
+		stale = stale->free;
 		node_free(node);
+	}
+
+	return active;
+}
+
+static void pool_free_work(struct work_struct *wrk)
+{
+	struct intel_gt_buffer_pool *pool =
+		container_of(wrk, typeof(*pool), work.work);
 
-	if (active)
+	if (pool_free_older_than(pool, HZ))
 		schedule_delayed_work(&pool->work,
 				      round_jiffies_up_relative(HZ));
 }
@@ -108,9 +134,10 @@ static void pool_retire(struct i915_active *ref)
 	/* Return this object to the shrinker pool */
 	i915_gem_object_make_purgeable(node->obj);
 
+	GEM_BUG_ON(node->age);
 	spin_lock_irqsave(&pool->lock, flags);
-	node->age = jiffies;
-	list_add(&node->link, list);
+	list_add_rcu(&node->link, list);
+	WRITE_ONCE(node->age, jiffies ?: 1); /* 0 reserved for active nodes */
 	spin_unlock_irqrestore(&pool->lock, flags);
 
 	schedule_delayed_work(&pool->work,
@@ -129,6 +156,7 @@ node_create(struct intel_gt_buffer_pool *pool, size_t sz)
 	if (!node)
 		return ERR_PTR(-ENOMEM);
 
+	node->age = 0;
 	node->pool = pool;
 	i915_active_init(&node->active, pool_active, pool_retire);
 
@@ -151,20 +179,30 @@ intel_gt_get_buffer_pool(struct intel_gt *gt, size_t size)
 	struct intel_gt_buffer_pool *pool = &gt->buffer_pool;
 	struct intel_gt_buffer_pool_node *node;
 	struct list_head *list;
-	unsigned long flags;
 	int ret;
 
 	size = PAGE_ALIGN(size);
 	list = bucket_for_size(pool, size);
 
-	spin_lock_irqsave(&pool->lock, flags);
-	list_for_each_entry(node, list, link) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(node, list, link) {
+		unsigned long age;
+
 		if (node->obj->base.size < size)
 			continue;
-		list_del(&node->link);
-		break;
+
+		age = READ_ONCE(node->age);
+		if (!age)
+			continue;
+
+		if (cmpxchg(&node->age, age, 0) == age) {
+			spin_lock_irq(&pool->lock);
+			list_del_rcu(&node->link);
+			spin_unlock_irq(&pool->lock);
+			break;
+		}
 	}
-	spin_unlock_irqrestore(&pool->lock, flags);
+	rcu_read_unlock();
 
 	if (&node->link == list) {
 		node = node_create(pool, size);
@@ -192,28 +230,13 @@ void intel_gt_init_buffer_pool(struct intel_gt *gt)
 	INIT_DELAYED_WORK(&pool->work, pool_free_work);
 }
 
-static void pool_free_imm(struct intel_gt_buffer_pool *pool)
-{
-	int n;
-
-	spin_lock_irq(&pool->lock);
-	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++) {
-		struct intel_gt_buffer_pool_node *node, *next;
-		struct list_head *list = &pool->cache_list[n];
-
-		list_for_each_entry_safe(node, next, list, link)
-			node_free(node);
-		INIT_LIST_HEAD(list);
-	}
-	spin_unlock_irq(&pool->lock);
-}
-
 void intel_gt_flush_buffer_pool(struct intel_gt *gt)
 {
 	struct intel_gt_buffer_pool *pool = &gt->buffer_pool;
 
 	do {
-		pool_free_imm(pool);
+		while (pool_free_older_than(pool, 0))
+			;
 	} while (cancel_delayed_work_sync(&pool->work));
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool_types.h b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool_types.h
index e28bdda771ed..bcf1658c9633 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool_types.h
@@ -25,7 +25,11 @@ struct intel_gt_buffer_pool_node {
 	struct i915_active active;
 	struct drm_i915_gem_object *obj;
 	struct list_head link;
-	struct intel_gt_buffer_pool *pool;
+	union {
+		struct intel_gt_buffer_pool *pool;
+		struct intel_gt_buffer_pool_node *free;
+		struct rcu_head rcu;
+	};
 	unsigned long age;
 };
 
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index b05da68e52f4..257063a57101 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -8,6 +8,7 @@
 
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "intel_breadcrumbs.h"
 #include "intel_gt.h"
 #include "intel_gt_irq.h"
 #include "intel_uncore.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c
index 2a72cce63fd9..3f1114b58b01 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.c
@@ -11,160 +11,24 @@
 #include "intel_gt.h"
 #include "intel_gtt.h"
 
-void stash_init(struct pagestash *stash)
+struct drm_i915_gem_object *alloc_pt_dma(struct i915_address_space *vm, int sz)
 {
-	pagevec_init(&stash->pvec);
-	spin_lock_init(&stash->lock);
-}
-
-static struct page *stash_pop_page(struct pagestash *stash)
-{
-	struct page *page = NULL;
-
-	spin_lock(&stash->lock);
-	if (likely(stash->pvec.nr))
-		page = stash->pvec.pages[--stash->pvec.nr];
-	spin_unlock(&stash->lock);
-
-	return page;
-}
-
-static void stash_push_pagevec(struct pagestash *stash, struct pagevec *pvec)
-{
-	unsigned int nr;
-
-	spin_lock_nested(&stash->lock, SINGLE_DEPTH_NESTING);
-
-	nr = min_t(typeof(nr), pvec->nr, pagevec_space(&stash->pvec));
-	memcpy(stash->pvec.pages + stash->pvec.nr,
-	       pvec->pages + pvec->nr - nr,
-	       sizeof(pvec->pages[0]) * nr);
-	stash->pvec.nr += nr;
-
-	spin_unlock(&stash->lock);
-
-	pvec->nr -= nr;
-}
-
-static struct page *vm_alloc_page(struct i915_address_space *vm, gfp_t gfp)
-{
-	struct pagevec stack;
-	struct page *page;
-
 	if (I915_SELFTEST_ONLY(should_fail(&vm->fault_attr, 1)))
 		i915_gem_shrink_all(vm->i915);
 
-	page = stash_pop_page(&vm->free_pages);
-	if (page)
-		return page;
-
-	if (!vm->pt_kmap_wc)
-		return alloc_page(gfp);
-
-	/* Look in our global stash of WC pages... */
-	page = stash_pop_page(&vm->i915->mm.wc_stash);
-	if (page)
-		return page;
-
-	/*
-	 * Otherwise batch allocate pages to amortize cost of set_pages_wc.
-	 *
-	 * We have to be careful as page allocation may trigger the shrinker
-	 * (via direct reclaim) which will fill up the WC stash underneath us.
-	 * So we add our WB pages into a temporary pvec on the stack and merge
-	 * them into the WC stash after all the allocations are complete.
-	 */
-	pagevec_init(&stack);
-	do {
-		struct page *page;
-
-		page = alloc_page(gfp);
-		if (unlikely(!page))
-			break;
-
-		stack.pages[stack.nr++] = page;
-	} while (pagevec_space(&stack));
-
-	if (stack.nr && !set_pages_array_wc(stack.pages, stack.nr)) {
-		page = stack.pages[--stack.nr];
-
-		/* Merge spare WC pages to the global stash */
-		if (stack.nr)
-			stash_push_pagevec(&vm->i915->mm.wc_stash, &stack);
-
-		/* Push any surplus WC pages onto the local VM stash */
-		if (stack.nr)
-			stash_push_pagevec(&vm->free_pages, &stack);
-	}
-
-	/* Return unwanted leftovers */
-	if (unlikely(stack.nr)) {
-		WARN_ON_ONCE(set_pages_array_wb(stack.pages, stack.nr));
-		__pagevec_release(&stack);
-	}
-
-	return page;
+	return i915_gem_object_create_internal(vm->i915, sz);
 }
 
-static void vm_free_pages_release(struct i915_address_space *vm,
-				  bool immediate)
+int pin_pt_dma(struct i915_address_space *vm, struct drm_i915_gem_object *obj)
 {
-	struct pagevec *pvec = &vm->free_pages.pvec;
-	struct pagevec stack;
-
-	lockdep_assert_held(&vm->free_pages.lock);
-	GEM_BUG_ON(!pagevec_count(pvec));
-
-	if (vm->pt_kmap_wc) {
-		/*
-		 * When we use WC, first fill up the global stash and then
-		 * only if full immediately free the overflow.
-		 */
-		stash_push_pagevec(&vm->i915->mm.wc_stash, pvec);
-
-		/*
-		 * As we have made some room in the VM's free_pages,
-		 * we can wait for it to fill again. Unless we are
-		 * inside i915_address_space_fini() and must
-		 * immediately release the pages!
-		 */
-		if (pvec->nr <= (immediate ? 0 : PAGEVEC_SIZE - 1))
-			return;
+	int err;
 
-		/*
-		 * We have to drop the lock to allow ourselves to sleep,
-		 * so take a copy of the pvec and clear the stash for
-		 * others to use it as we sleep.
-		 */
-		stack = *pvec;
-		pagevec_reinit(pvec);
-		spin_unlock(&vm->free_pages.lock);
-
-		pvec = &stack;
-		set_pages_array_wb(pvec->pages, pvec->nr);
-
-		spin_lock(&vm->free_pages.lock);
-	}
+	err = i915_gem_object_pin_pages(obj);
+	if (err)
+		return err;
 
-	__pagevec_release(pvec);
-}
-
-static void vm_free_page(struct i915_address_space *vm, struct page *page)
-{
-	/*
-	 * On !llc, we need to change the pages back to WB. We only do so
-	 * in bulk, so we rarely need to change the page attributes here,
-	 * but doing so requires a stop_machine() from deep inside arch/x86/mm.
-	 * To make detection of the possible sleep more likely, use an
-	 * unconditional might_sleep() for everybody.
-	 */
-	might_sleep();
-	spin_lock(&vm->free_pages.lock);
-	while (!pagevec_space(&vm->free_pages.pvec))
-		vm_free_pages_release(vm, false);
-	GEM_BUG_ON(pagevec_count(&vm->free_pages.pvec) >= PAGEVEC_SIZE);
-	pagevec_add(&vm->free_pages.pvec, page);
-	spin_unlock(&vm->free_pages.lock);
+	i915_gem_object_make_unshrinkable(obj);
+	return 0;
 }
 
 void __i915_vm_close(struct i915_address_space *vm)
@@ -194,14 +58,7 @@ void __i915_vm_close(struct i915_address_space *vm)
 
 void i915_address_space_fini(struct i915_address_space *vm)
 {
-	spin_lock(&vm->free_pages.lock);
-	if (pagevec_count(&vm->free_pages.pvec))
-		vm_free_pages_release(vm, true);
-	GEM_BUG_ON(pagevec_count(&vm->free_pages.pvec));
-	spin_unlock(&vm->free_pages.lock);
-
 	drm_mm_takedown(&vm->mm);
-
 	mutex_destroy(&vm->mutex);
 }
 
@@ -246,8 +103,6 @@ void i915_address_space_init(struct i915_address_space *vm, int subclass)
 	drm_mm_init(&vm->mm, 0, vm->total);
 	vm->mm.head_node.color = I915_COLOR_UNEVICTABLE;
 
-	stash_init(&vm->free_pages);
-
 	INIT_LIST_HEAD(&vm->bound_list);
 }
 
@@ -264,64 +119,50 @@ void clear_pages(struct i915_vma *vma)
 	memset(&vma->page_sizes, 0, sizeof(vma->page_sizes));
 }
 
-static int __setup_page_dma(struct i915_address_space *vm,
-			    struct i915_page_dma *p,
-			    gfp_t gfp)
-{
-	p->page = vm_alloc_page(vm, gfp | I915_GFP_ALLOW_FAIL);
-	if (unlikely(!p->page))
-		return -ENOMEM;
-
-	p->daddr = dma_map_page_attrs(vm->dma,
-				      p->page, 0, PAGE_SIZE,
-				      PCI_DMA_BIDIRECTIONAL,
-				      DMA_ATTR_SKIP_CPU_SYNC |
-				      DMA_ATTR_NO_WARN);
-	if (unlikely(dma_mapping_error(vm->dma, p->daddr))) {
-		vm_free_page(vm, p->page);
-		return -ENOMEM;
-	}
-
-	return 0;
-}
-
-int setup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p)
+dma_addr_t __px_dma(struct drm_i915_gem_object *p)
 {
-	return __setup_page_dma(vm, p, __GFP_HIGHMEM);
+	GEM_BUG_ON(!i915_gem_object_has_pages(p));
+	return sg_dma_address(p->mm.pages->sgl);
 }
 
-void cleanup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p)
+struct page *__px_page(struct drm_i915_gem_object *p)
 {
-	dma_unmap_page(vm->dma, p->daddr, PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
-	vm_free_page(vm, p->page);
+	GEM_BUG_ON(!i915_gem_object_has_pages(p));
+	return sg_page(p->mm.pages->sgl);
 }
 
 void
-fill_page_dma(const struct i915_page_dma *p, const u64 val, unsigned int count)
+fill_page_dma(struct drm_i915_gem_object *p, const u64 val, unsigned int count)
 {
-	kunmap_atomic(memset64(kmap_atomic(p->page), val, count));
+	struct page *page = __px_page(p);
+	void *vaddr;
+
+	vaddr = kmap(page);
+	memset64(vaddr, val, count);
+	clflush_cache_range(vaddr, PAGE_SIZE);
+	kunmap(page);
 }
 
-static void poison_scratch_page(struct page *page, unsigned long size)
+static void poison_scratch_page(struct drm_i915_gem_object *scratch)
 {
-	if (!IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))
-		return;
+	struct sgt_iter sgt;
+	struct page *page;
+	u8 val;
 
-	GEM_BUG_ON(!IS_ALIGNED(size, PAGE_SIZE));
+	val = 0;
+	if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))
+		val = POISON_FREE;
 
-	do {
+	for_each_sgt_page(page, sgt, scratch->mm.pages) {
 		void *vaddr;
 
 		vaddr = kmap(page);
-		memset(vaddr, POISON_FREE, PAGE_SIZE);
+		memset(vaddr, val, PAGE_SIZE);
 		kunmap(page);
-
-		page = pfn_to_page(page_to_pfn(page) + 1);
-		size -= PAGE_SIZE;
-	} while (size);
+	}
 }
 
-int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
+int setup_scratch_page(struct i915_address_space *vm)
 {
 	unsigned long size;
 
@@ -338,21 +179,27 @@ int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 	 */
 	size = I915_GTT_PAGE_SIZE_4K;
 	if (i915_vm_is_4lvl(vm) &&
-	    HAS_PAGE_SIZES(vm->i915, I915_GTT_PAGE_SIZE_64K)) {
+	    HAS_PAGE_SIZES(vm->i915, I915_GTT_PAGE_SIZE_64K))
 		size = I915_GTT_PAGE_SIZE_64K;
-		gfp |= __GFP_NOWARN;
-	}
-	gfp |= __GFP_ZERO | __GFP_RETRY_MAYFAIL;
 
 	do {
-		unsigned int order = get_order(size);
-		struct page *page;
-		dma_addr_t addr;
+		struct drm_i915_gem_object *obj;
 
-		page = alloc_pages(gfp, order);
-		if (unlikely(!page))
+		obj = vm->alloc_pt_dma(vm, size);
+		if (IS_ERR(obj))
 			goto skip;
 
+		if (pin_pt_dma(vm, obj))
+			goto skip_obj;
+
+		/* We need a single contiguous page for our scratch */
+		if (obj->mm.page_sizes.sg < size)
+			goto skip_obj;
+
+		/* And it needs to be correspondingly aligned */
+		if (__px_dma(obj) & (size - 1))
+			goto skip_obj;
+
 		/*
 		 * Use a non-zero scratch page for debugging.
 		 *
@@ -362,61 +209,28 @@ int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 		 * should it ever be accidentally used, the effect should be
 		 * fairly benign.
 		 */
-		poison_scratch_page(page, size);
-
-		addr = dma_map_page_attrs(vm->dma,
-					  page, 0, size,
-					  PCI_DMA_BIDIRECTIONAL,
-					  DMA_ATTR_SKIP_CPU_SYNC |
-					  DMA_ATTR_NO_WARN);
-		if (unlikely(dma_mapping_error(vm->dma, addr)))
-			goto free_page;
-
-		if (unlikely(!IS_ALIGNED(addr, size)))
-			goto unmap_page;
-
-		vm->scratch[0].base.page = page;
-		vm->scratch[0].base.daddr = addr;
-		vm->scratch_order = order;
+		poison_scratch_page(obj);
+
+		vm->scratch[0] = obj;
+		vm->scratch_order = get_order(size);
 		return 0;
 
-unmap_page:
-		dma_unmap_page(vm->dma, addr, size, PCI_DMA_BIDIRECTIONAL);
-free_page:
-		__free_pages(page, order);
+skip_obj:
+		i915_gem_object_put(obj);
 skip:
 		if (size == I915_GTT_PAGE_SIZE_4K)
 			return -ENOMEM;
 
 		size = I915_GTT_PAGE_SIZE_4K;
-		gfp &= ~__GFP_NOWARN;
 	} while (1);
 }
 
-void cleanup_scratch_page(struct i915_address_space *vm)
-{
-	struct i915_page_dma *p = px_base(&vm->scratch[0]);
-	unsigned int order = vm->scratch_order;
-
-	dma_unmap_page(vm->dma, p->daddr, BIT(order) << PAGE_SHIFT,
-		       PCI_DMA_BIDIRECTIONAL);
-	__free_pages(p->page, order);
-}
-
 void free_scratch(struct i915_address_space *vm)
 {
 	int i;
 
-	if (!px_dma(&vm->scratch[0])) /* set to 0 on clones */
-		return;
-
-	for (i = 1; i <= vm->top; i++) {
-		if (!px_dma(&vm->scratch[i]))
-			break;
-		cleanup_page_dma(vm, px_base(&vm->scratch[i]));
-	}
-
-	cleanup_scratch_page(vm);
+	for (i = 0; i <= vm->top; i++)
+		i915_gem_object_put(vm->scratch[i]);
 }
 
 void gtt_write_workarounds(struct intel_gt *gt)
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index f2b75078e05f..c13c650ced22 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -134,38 +134,29 @@ typedef u64 gen8_pte_t;
 #define GEN8_PDE_IPS_64K BIT(11)
 #define GEN8_PDE_PS_2M   BIT(7)
 
+enum i915_cache_level;
+
+struct drm_i915_file_private;
+struct drm_i915_gem_object;
 struct i915_fence_reg;
+struct i915_vma;
+struct intel_gt;
 
 #define for_each_sgt_daddr(__dp, __iter, __sgt) \
 	__for_each_sgt_daddr(__dp, __iter, __sgt, I915_GTT_PAGE_SIZE)
 
-struct i915_page_dma {
-	struct page *page;
+struct i915_page_table {
+	struct drm_i915_gem_object *base;
 	union {
-		dma_addr_t daddr;
-
-		/*
-		 * For gen6/gen7 only. This is the offset in the GGTT
-		 * where the page directory entries for PPGTT begin
-		 */
-		u32 ggtt_offset;
+		atomic_t used;
+		struct i915_page_table *stash;
 	};
 };
 
-struct i915_page_scratch {
-	struct i915_page_dma base;
-	u64 encode;
-};
-
-struct i915_page_table {
-	struct i915_page_dma base;
-	atomic_t used;
-};
-
 struct i915_page_directory {
 	struct i915_page_table pt;
 	spinlock_t lock;
-	void *entry[512];
+	void **entry;
 };
 
 #define __px_choose_expr(x, type, expr, other) \
@@ -176,12 +167,14 @@ struct i915_page_directory {
 	other)
 
 #define px_base(px) \
-	__px_choose_expr(px, struct i915_page_dma *, __x, \
-	__px_choose_expr(px, struct i915_page_scratch *, &__x->base, \
-	__px_choose_expr(px, struct i915_page_table *, &__x->base, \
-	__px_choose_expr(px, struct i915_page_directory *, &__x->pt.base, \
-	(void)0))))
-#define px_dma(px) (px_base(px)->daddr)
+	__px_choose_expr(px, struct drm_i915_gem_object *, __x, \
+	__px_choose_expr(px, struct i915_page_table *, __x->base, \
+	__px_choose_expr(px, struct i915_page_directory *, __x->pt.base, \
+	(void)0)))
+
+struct page *__px_page(struct drm_i915_gem_object *p);
+dma_addr_t __px_dma(struct drm_i915_gem_object *p);
+#define px_dma(px) (__px_dma(px_base(px)))
 
 #define px_pt(px) \
 	__px_choose_expr(px, struct i915_page_table *, __x, \
@@ -189,19 +182,18 @@ struct i915_page_directory {
 	(void)0))
 #define px_used(px) (&px_pt(px)->used)
 
-enum i915_cache_level;
-
-struct drm_i915_file_private;
-struct drm_i915_gem_object;
-struct i915_vma;
-struct intel_gt;
+struct i915_vm_pt_stash {
+	/* preallocated chains of page tables/directories */
+	struct i915_page_table *pt[2];
+};
 
 struct i915_vma_ops {
 	/* Map an object into an address space with the given cache flags. */
-	int (*bind_vma)(struct i915_address_space *vm,
-			struct i915_vma *vma,
-			enum i915_cache_level cache_level,
-			u32 flags);
+	void (*bind_vma)(struct i915_address_space *vm,
+			 struct i915_vm_pt_stash *stash,
+			 struct i915_vma *vma,
+			 enum i915_cache_level cache_level,
+			 u32 flags);
 	/*
 	 * Unmap an object from an address space. This usually consists of
 	 * setting the valid PTE entries to a reserved scratch page.
@@ -213,13 +205,6 @@ struct i915_vma_ops {
 	void (*clear_pages)(struct i915_vma *vma);
 };
 
-struct pagestash {
-	spinlock_t lock;
-	struct pagevec pvec;
-};
-
-void stash_init(struct pagestash *stash);
-
 struct i915_address_space {
 	struct kref ref;
 	struct rcu_work rcu;
@@ -256,33 +241,33 @@ struct i915_address_space {
 #define VM_CLASS_GGTT 0
 #define VM_CLASS_PPGTT 1
 
-	struct i915_page_scratch scratch[4];
-	unsigned int scratch_order;
-	unsigned int top;
-
+	struct drm_i915_gem_object *scratch[4];
 	/**
 	 * List of vma currently bound.
 	 */
 	struct list_head bound_list;
 
-	struct pagestash free_pages;
-
 	/* Global GTT */
 	bool is_ggtt:1;
 
-	/* Some systems require uncached updates of the page directories */
-	bool pt_kmap_wc:1;
-
 	/* Some systems support read-only mappings for GGTT and/or PPGTT */
 	bool has_read_only:1;
 
+	u8 top;
+	u8 pd_shift;
+	u8 scratch_order;
+
+	struct drm_i915_gem_object *
+		(*alloc_pt_dma)(struct i915_address_space *vm, int sz);
+
 	u64 (*pte_encode)(dma_addr_t addr,
 			  enum i915_cache_level level,
 			  u32 flags); /* Create a valid PTE */
 #define PTE_READ_ONLY	BIT(0)
 
-	int (*allocate_va_range)(struct i915_address_space *vm,
-				 u64 start, u64 length);
+	void (*allocate_va_range)(struct i915_address_space *vm,
+				  struct i915_vm_pt_stash *stash,
+				  u64 start, u64 length);
 	void (*clear_range)(struct i915_address_space *vm,
 			    u64 start, u64 length);
 	void (*insert_page)(struct i915_address_space *vm,
@@ -490,9 +475,9 @@ i915_pd_entry(const struct i915_page_directory * const pdp,
 static inline dma_addr_t
 i915_page_dir_dma_addr(const struct i915_ppgtt *ppgtt, const unsigned int n)
 {
-	struct i915_page_dma *pt = ppgtt->pd->entry[n];
+	struct i915_page_table *pt = ppgtt->pd->entry[n];
 
-	return px_dma(pt ?: px_base(&ppgtt->vm.scratch[ppgtt->vm.top]));
+	return __px_dma(pt ? px_base(pt) : ppgtt->vm.scratch[ppgtt->vm.top]);
 }
 
 void ppgtt_init(struct i915_ppgtt *ppgtt, struct intel_gt *gt);
@@ -517,13 +502,10 @@ struct i915_ppgtt *i915_ppgtt_create(struct intel_gt *gt);
 void i915_ggtt_suspend(struct i915_ggtt *gtt);
 void i915_ggtt_resume(struct i915_ggtt *ggtt);
 
-int setup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p);
-void cleanup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p);
-
-#define kmap_atomic_px(px) kmap_atomic(px_base(px)->page)
+#define kmap_atomic_px(px) kmap_atomic(__px_page(px_base(px)))
 
 void
-fill_page_dma(const struct i915_page_dma *p, const u64 val, unsigned int count);
+fill_page_dma(struct drm_i915_gem_object *p, const u64 val, unsigned int count);
 
 #define fill_px(px, v) fill_page_dma(px_base(px), (v), PAGE_SIZE / sizeof(u64))
 #define fill32_px(px, v) do {						\
@@ -531,47 +513,51 @@ fill_page_dma(const struct i915_page_dma *p, const u64 val, unsigned int count);
 	fill_px((px), v__ << 32 | v__);					\
 } while (0)
 
-int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp);
-void cleanup_scratch_page(struct i915_address_space *vm);
+int setup_scratch_page(struct i915_address_space *vm);
 void free_scratch(struct i915_address_space *vm);
 
+struct drm_i915_gem_object *alloc_pt_dma(struct i915_address_space *vm, int sz);
 struct i915_page_table *alloc_pt(struct i915_address_space *vm);
 struct i915_page_directory *alloc_pd(struct i915_address_space *vm);
-struct i915_page_directory *__alloc_pd(size_t sz);
+struct i915_page_directory *__alloc_pd(int npde);
 
-void free_pd(struct i915_address_space *vm, struct i915_page_dma *pd);
+int pin_pt_dma(struct i915_address_space *vm, struct drm_i915_gem_object *obj);
 
-#define free_px(vm, px) free_pd(vm, px_base(px))
+void free_px(struct i915_address_space *vm,
+	     struct i915_page_table *pt, int lvl);
+#define free_pt(vm, px) free_px(vm, px, 0)
+#define free_pd(vm, px) free_px(vm, px_pt(px), 1)
 
 void
 __set_pd_entry(struct i915_page_directory * const pd,
 	       const unsigned short idx,
-	       struct i915_page_dma * const to,
+	       struct i915_page_table *pt,
 	       u64 (*encode)(const dma_addr_t, const enum i915_cache_level));
 
 #define set_pd_entry(pd, idx, to) \
-	__set_pd_entry((pd), (idx), px_base(to), gen8_pde_encode)
+	__set_pd_entry((pd), (idx), px_pt(to), gen8_pde_encode)
 
 void
 clear_pd_entry(struct i915_page_directory * const pd,
 	       const unsigned short idx,
-	       const struct i915_page_scratch * const scratch);
+	       const struct drm_i915_gem_object * const scratch);
 
 bool
 release_pd_entry(struct i915_page_directory * const pd,
 		 const unsigned short idx,
 		 struct i915_page_table * const pt,
-		 const struct i915_page_scratch * const scratch);
+		 const struct drm_i915_gem_object * const scratch);
 void gen6_ggtt_invalidate(struct i915_ggtt *ggtt);
 
 int ggtt_set_pages(struct i915_vma *vma);
 int ppgtt_set_pages(struct i915_vma *vma);
 void clear_pages(struct i915_vma *vma);
 
-int ppgtt_bind_vma(struct i915_address_space *vm,
-		   struct i915_vma *vma,
-		   enum i915_cache_level cache_level,
-		   u32 flags);
+void ppgtt_bind_vma(struct i915_address_space *vm,
+		    struct i915_vm_pt_stash *stash,
+		    struct i915_vma *vma,
+		    enum i915_cache_level cache_level,
+		    u32 flags);
 void ppgtt_unbind_vma(struct i915_address_space *vm,
 		      struct i915_vma *vma);
 
@@ -579,6 +565,14 @@ void gtt_write_workarounds(struct intel_gt *gt);
 
 void setup_private_pat(struct intel_uncore *uncore);
 
+int i915_vm_alloc_pt_stash(struct i915_address_space *vm,
+			   struct i915_vm_pt_stash *stash,
+			   u64 size);
+int i915_vm_pin_pt_stash(struct i915_address_space *vm,
+			 struct i915_vm_pt_stash *stash);
+void i915_vm_free_pt_stash(struct i915_address_space *vm,
+			   struct i915_vm_pt_stash *stash);
+
 static inline struct sgt_dma {
 	struct scatterlist *sg;
 	dma_addr_t dma, max;
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 9eeaca957a7e..0412a44f25f2 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -137,6 +137,7 @@
 #include "i915_perf.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
+#include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_engine_pm.h"
 #include "intel_gt.h"
@@ -1148,20 +1149,6 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine)
 		} else {
 			struct intel_engine_cs *owner = rq->context->engine;
 
-			/*
-			 * Decouple the virtual breadcrumb before moving it
-			 * back to the virtual engine -- we don't want the
-			 * request to complete in the background and try
-			 * and cancel the breadcrumb on the virtual engine
-			 * (instead of the old engine where it is linked)!
-			 */
-			if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
-				     &rq->fence.flags)) {
-				spin_lock_nested(&rq->lock,
-						 SINGLE_DEPTH_NESTING);
-				i915_request_cancel_breadcrumb(rq);
-				spin_unlock(&rq->lock);
-			}
 			WRITE_ONCE(rq->engine, owner);
 			owner->submit_request(rq);
 			active = NULL;
@@ -1819,16 +1806,31 @@ static bool virtual_matches(const struct virtual_engine *ve,
 	return true;
 }
 
-static void virtual_xfer_breadcrumbs(struct virtual_engine *ve)
+static void virtual_xfer_context(struct virtual_engine *ve,
+				 struct intel_engine_cs *engine)
 {
+	unsigned int n;
+
+	if (likely(engine == ve->siblings[0]))
+		return;
+
+	GEM_BUG_ON(READ_ONCE(ve->context.inflight));
+	if (!intel_engine_has_relative_mmio(engine))
+		virtual_update_register_offsets(ve->context.lrc_reg_state,
+						engine);
+
 	/*
-	 * All the outstanding signals on ve->siblings[0] must have
-	 * been completed, just pending the interrupt handler. As those
-	 * signals still refer to the old sibling (via rq->engine), we must
-	 * transfer those to the old irq_worker to keep our locking
-	 * consistent.
+	 * Move the bound engine to the top of the list for
+	 * future execution. We then kick this tasklet first
+	 * before checking others, so that we preferentially
+	 * reuse this set of bound registers.
 	 */
-	intel_engine_transfer_stale_breadcrumbs(ve->siblings[0], &ve->context);
+	for (n = 1; n < ve->num_siblings; n++) {
+		if (ve->siblings[n] == engine) {
+			swap(ve->siblings[n], ve->siblings[0]);
+			break;
+		}
+	}
 }
 
 #define for_each_waiter(p__, rq__) \
@@ -2279,38 +2281,23 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 			GEM_BUG_ON(!(rq->execution_mask & engine->mask));
 			WRITE_ONCE(rq->engine, engine);
 
-			if (engine != ve->siblings[0]) {
-				u32 *regs = ve->context.lrc_reg_state;
-				unsigned int n;
-
-				GEM_BUG_ON(READ_ONCE(ve->context.inflight));
-
-				if (!intel_engine_has_relative_mmio(engine))
-					virtual_update_register_offsets(regs,
-									engine);
-
-				if (!list_empty(&ve->context.signals))
-					virtual_xfer_breadcrumbs(ve);
-
+			if (__i915_request_submit(rq)) {
 				/*
-				 * Move the bound engine to the top of the list
-				 * for future execution. We then kick this
-				 * tasklet first before checking others, so that
-				 * we preferentially reuse this set of bound
-				 * registers.
+				 * Only after we confirm that we will submit
+				 * this request (i.e. it has not already
+				 * completed), do we want to update the context.
+				 *
+				 * This serves two purposes. It avoids
+				 * unnecessary work if we are resubmitting an
+				 * already completed request after timeslicing.
+				 * But more importantly, it prevents us altering
+				 * ve->siblings[] on an idle context, where
+				 * we may be using ve->siblings[] in
+				 * virtual_context_enter / virtual_context_exit.
 				 */
-				for (n = 1; n < ve->num_siblings; n++) {
-					if (ve->siblings[n] == engine) {
-						swap(ve->siblings[n],
-						     ve->siblings[0]);
-						break;
-					}
-				}
-
+				virtual_xfer_context(ve, engine);
 				GEM_BUG_ON(ve->siblings[0] != engine);
-			}
 
-			if (__i915_request_submit(rq)) {
 				submit = true;
 				last = rq;
 			}
@@ -3316,7 +3303,10 @@ static void execlists_context_unpin(struct intel_context *ce)
 {
 	check_redzone((void *)ce->lrc_reg_state - LRC_STATE_OFFSET,
 		      ce->engine);
+}
 
+static void execlists_context_post_unpin(struct intel_context *ce)
+{
 	i915_gem_object_unpin_map(ce->state->obj);
 }
 
@@ -3478,20 +3468,24 @@ __execlists_update_reg_state(const struct intel_context *ce,
 }
 
 static int
-__execlists_context_pin(struct intel_context *ce,
-			struct intel_engine_cs *engine)
+execlists_context_pre_pin(struct intel_context *ce,
+			  struct i915_gem_ww_ctx *ww, void **vaddr)
 {
-	void *vaddr;
-
 	GEM_BUG_ON(!ce->state);
 	GEM_BUG_ON(!i915_vma_is_pinned(ce->state));
 
-	vaddr = i915_gem_object_pin_map(ce->state->obj,
-					i915_coherent_map_type(engine->i915) |
+	*vaddr = i915_gem_object_pin_map(ce->state->obj,
+					i915_coherent_map_type(ce->engine->i915) |
 					I915_MAP_OVERRIDE);
-	if (IS_ERR(vaddr))
-		return PTR_ERR(vaddr);
 
+	return PTR_ERR_OR_ZERO(*vaddr);
+}
+
+static int
+__execlists_context_pin(struct intel_context *ce,
+			struct intel_engine_cs *engine,
+			void *vaddr)
+{
 	ce->lrc.lrca = lrc_descriptor(ce, engine) | CTX_DESC_FORCE_RESTORE;
 	ce->lrc_reg_state = vaddr + LRC_STATE_OFFSET;
 	__execlists_update_reg_state(ce, engine, ce->ring->tail);
@@ -3499,9 +3493,9 @@ __execlists_context_pin(struct intel_context *ce,
 	return 0;
 }
 
-static int execlists_context_pin(struct intel_context *ce)
+static int execlists_context_pin(struct intel_context *ce, void *vaddr)
 {
-	return __execlists_context_pin(ce, ce->engine);
+	return __execlists_context_pin(ce, ce->engine, vaddr);
 }
 
 static int execlists_context_alloc(struct intel_context *ce)
@@ -3527,8 +3521,10 @@ static void execlists_context_reset(struct intel_context *ce)
 static const struct intel_context_ops execlists_context_ops = {
 	.alloc = execlists_context_alloc,
 
+	.pre_pin = execlists_context_pre_pin,
 	.pin = execlists_context_pin,
 	.unpin = execlists_context_unpin,
+	.post_unpin = execlists_context_post_unpin,
 
 	.enter = intel_context_enter_engine,
 	.exit = intel_context_exit_engine,
@@ -3892,7 +3888,7 @@ static int lrc_setup_wa_ctx(struct intel_engine_cs *engine)
 		goto err;
 	}
 
-	err = i915_ggtt_pin(vma, 0, PIN_HIGH);
+	err = i915_ggtt_pin(vma, NULL, 0, PIN_HIGH);
 	if (err)
 		goto err;
 
@@ -4133,7 +4129,7 @@ static int execlists_resume(struct intel_engine_cs *engine)
 {
 	intel_mocs_init_engine(engine);
 
-	intel_engine_reset_breadcrumbs(engine);
+	intel_breadcrumbs_reset(engine->breadcrumbs);
 
 	if (GEM_SHOW_DEBUG() && unexpected_starting_state(engine)) {
 		struct drm_printer p = drm_debug_printer(__func__);
@@ -4562,7 +4558,7 @@ static int gen8_emit_flush_render(struct i915_request *request,
 			vf_flush_wa = true;
 
 		/* WaForGAMHang:kbl */
-		if (IS_KBL_REVID(request->engine->i915, 0, KBL_REVID_B0))
+		if (IS_KBL_GT_REVID(request->engine->i915, 0, KBL_REVID_B0))
 			dc_flush_wa = true;
 	}
 
@@ -4764,14 +4760,21 @@ static int gen12_emit_flush(struct i915_request *request, u32 mode)
 	intel_engine_mask_t aux_inv = 0;
 	u32 cmd, *cs;
 
+	cmd = 4;
+	if (mode & EMIT_INVALIDATE)
+		cmd += 2;
 	if (mode & EMIT_INVALIDATE)
 		aux_inv = request->engine->mask & ~BIT(BCS0);
+	if (aux_inv)
+		cmd += 2 * hweight8(aux_inv) + 2;
 
-	cs = intel_ring_begin(request,
-			      4 + (aux_inv ? 2 * hweight8(aux_inv) + 2 : 0));
+	cs = intel_ring_begin(request, cmd);
 	if (IS_ERR(cs))
 		return PTR_ERR(cs);
 
+	if (mode & EMIT_INVALIDATE)
+		*cs++ = preparser_disable(true);
+
 	cmd = MI_FLUSH_DW + 1;
 
 	/* We always require a command barrier so that subsequent
@@ -4804,6 +4807,10 @@ static int gen12_emit_flush(struct i915_request *request, u32 mode)
 		}
 		*cs++ = MI_NOOP;
 	}
+
+	if (mode & EMIT_INVALIDATE)
+		*cs++ = preparser_disable(false);
+
 	intel_ring_advance(request, cs);
 
 	return 0;
@@ -5302,6 +5309,14 @@ populate_lr_context(struct intel_context *ce,
 	return 0;
 }
 
+static struct intel_timeline *pinned_timeline(struct intel_context *ce)
+{
+	struct intel_timeline *tl = fetch_and_zero(&ce->timeline);
+
+	return intel_timeline_create_from_engine(ce->engine,
+						 page_unmask_bits(tl));
+}
+
 static int __execlists_context_alloc(struct intel_context *ce,
 				     struct intel_engine_cs *engine)
 {
@@ -5332,19 +5347,17 @@ static int __execlists_context_alloc(struct intel_context *ce,
 		goto error_deref_obj;
 	}
 
-	if (!ce->timeline) {
+	if (!page_mask_bits(ce->timeline)) {
 		struct intel_timeline *tl;
-		struct i915_vma *hwsp;
 
 		/*
 		 * Use the static global HWSP for the kernel context, and
 		 * a dynamically allocated cacheline for everyone else.
 		 */
-		hwsp = NULL;
-		if (unlikely(intel_context_is_barrier(ce)))
-			hwsp = engine->status_page.vma;
-
-		tl = intel_timeline_create(engine->gt, hwsp);
+		if (unlikely(ce->timeline))
+			tl = pinned_timeline(ce);
+		else
+			tl = intel_timeline_create(engine->gt);
 		if (IS_ERR(tl)) {
 			ret = PTR_ERR(tl);
 			goto error_deref_obj;
@@ -5450,12 +5463,12 @@ static int virtual_context_alloc(struct intel_context *ce)
 	return __execlists_context_alloc(ce, ve->siblings[0]);
 }
 
-static int virtual_context_pin(struct intel_context *ce)
+static int virtual_context_pin(struct intel_context *ce, void *vaddr)
 {
 	struct virtual_engine *ve = container_of(ce, typeof(*ve), context);
 
 	/* Note: we must use a real engine class for setting up reg state */
-	return __execlists_context_pin(ce, ve->siblings[0]);
+	return __execlists_context_pin(ce, ve->siblings[0], vaddr);
 }
 
 static void virtual_context_enter(struct intel_context *ce)
@@ -5483,8 +5496,10 @@ static void virtual_context_exit(struct intel_context *ce)
 static const struct intel_context_ops virtual_context_ops = {
 	.alloc = virtual_context_alloc,
 
+	.pre_pin = execlists_context_pre_pin,
 	.pin = virtual_context_pin,
 	.unpin = execlists_context_unpin,
+	.post_unpin = execlists_context_post_unpin,
 
 	.enter = virtual_context_enter,
 	.exit = virtual_context_exit,
@@ -5718,9 +5733,7 @@ intel_execlists_create_virtual(struct intel_engine_cs **siblings,
 	snprintf(ve->base.name, sizeof(ve->base.name), "virtual");
 
 	intel_engine_init_active(&ve->base, ENGINE_VIRTUAL);
-	intel_engine_init_breadcrumbs(&ve->base);
 	intel_engine_init_execlists(&ve->base);
-	ve->base.breadcrumbs.irq_armed = true; /* fake HW, used for irq_work */
 
 	ve->base.cops = &virtual_context_ops;
 	ve->base.request_alloc = execlists_request_alloc;
@@ -5737,6 +5750,12 @@ intel_execlists_create_virtual(struct intel_engine_cs **siblings,
 
 	intel_context_init(&ve->context, &ve->base);
 
+	ve->base.breadcrumbs = intel_breadcrumbs_create(NULL);
+	if (!ve->base.breadcrumbs) {
+		err = -ENOMEM;
+		goto err_put;
+	}
+
 	for (n = 0; n < count; n++) {
 		struct intel_engine_cs *sibling = siblings[n];
 
diff --git a/drivers/gpu/drm/i915/gt/intel_ppgtt.c b/drivers/gpu/drm/i915/gt/intel_ppgtt.c
index f0862e924d11..46d9aceda64c 100644
--- a/drivers/gpu/drm/i915/gt/intel_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ppgtt.c
@@ -18,7 +18,8 @@ struct i915_page_table *alloc_pt(struct i915_address_space *vm)
 	if (unlikely(!pt))
 		return ERR_PTR(-ENOMEM);
 
-	if (unlikely(setup_page_dma(vm, &pt->base))) {
+	pt->base = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
+	if (IS_ERR(pt->base)) {
 		kfree(pt);
 		return ERR_PTR(-ENOMEM);
 	}
@@ -27,14 +28,20 @@ struct i915_page_table *alloc_pt(struct i915_address_space *vm)
 	return pt;
 }
 
-struct i915_page_directory *__alloc_pd(size_t sz)
+struct i915_page_directory *__alloc_pd(int count)
 {
 	struct i915_page_directory *pd;
 
-	pd = kzalloc(sz, I915_GFP_ALLOW_FAIL);
+	pd = kzalloc(sizeof(*pd), I915_GFP_ALLOW_FAIL);
 	if (unlikely(!pd))
 		return NULL;
 
+	pd->entry = kcalloc(count, sizeof(*pd->entry), I915_GFP_ALLOW_FAIL);
+	if (unlikely(!pd->entry)) {
+		kfree(pd);
+		return NULL;
+	}
+
 	spin_lock_init(&pd->lock);
 	return pd;
 }
@@ -43,11 +50,13 @@ struct i915_page_directory *alloc_pd(struct i915_address_space *vm)
 {
 	struct i915_page_directory *pd;
 
-	pd = __alloc_pd(sizeof(*pd));
+	pd = __alloc_pd(I915_PDES);
 	if (unlikely(!pd))
 		return ERR_PTR(-ENOMEM);
 
-	if (unlikely(setup_page_dma(vm, px_base(pd)))) {
+	pd->pt.base = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
+	if (IS_ERR(pd->pt.base)) {
+		kfree(pd->entry);
 		kfree(pd);
 		return ERR_PTR(-ENOMEM);
 	}
@@ -55,41 +64,52 @@ struct i915_page_directory *alloc_pd(struct i915_address_space *vm)
 	return pd;
 }
 
-void free_pd(struct i915_address_space *vm, struct i915_page_dma *pd)
+void free_px(struct i915_address_space *vm, struct i915_page_table *pt, int lvl)
 {
-	cleanup_page_dma(vm, pd);
-	kfree(pd);
+	BUILD_BUG_ON(offsetof(struct i915_page_directory, pt));
+
+	if (lvl) {
+		struct i915_page_directory *pd =
+			container_of(pt, typeof(*pd), pt);
+		kfree(pd->entry);
+	}
+
+	if (pt->base)
+		i915_gem_object_put(pt->base);
+
+	kfree(pt);
 }
 
 static inline void
-write_dma_entry(struct i915_page_dma * const pdma,
+write_dma_entry(struct drm_i915_gem_object * const pdma,
 		const unsigned short idx,
 		const u64 encoded_entry)
 {
-	u64 * const vaddr = kmap_atomic(pdma->page);
+	u64 * const vaddr = kmap_atomic(__px_page(pdma));
 
 	vaddr[idx] = encoded_entry;
+	clflush_cache_range(&vaddr[idx], sizeof(u64));
 	kunmap_atomic(vaddr);
 }
 
 void
 __set_pd_entry(struct i915_page_directory * const pd,
 	       const unsigned short idx,
-	       struct i915_page_dma * const to,
+	       struct i915_page_table * const to,
 	       u64 (*encode)(const dma_addr_t, const enum i915_cache_level))
 {
 	/* Each thread pre-pins the pd, and we may have a thread per pde. */
-	GEM_BUG_ON(atomic_read(px_used(pd)) > NALLOC * ARRAY_SIZE(pd->entry));
+	GEM_BUG_ON(atomic_read(px_used(pd)) > NALLOC * I915_PDES);
 
 	atomic_inc(px_used(pd));
 	pd->entry[idx] = to;
-	write_dma_entry(px_base(pd), idx, encode(to->daddr, I915_CACHE_LLC));
+	write_dma_entry(px_base(pd), idx, encode(px_dma(to), I915_CACHE_LLC));
 }
 
 void
 clear_pd_entry(struct i915_page_directory * const pd,
 	       const unsigned short idx,
-	       const struct i915_page_scratch * const scratch)
+	       const struct drm_i915_gem_object * const scratch)
 {
 	GEM_BUG_ON(atomic_read(px_used(pd)) == 0);
 
@@ -102,7 +122,7 @@ bool
 release_pd_entry(struct i915_page_directory * const pd,
 		 const unsigned short idx,
 		 struct i915_page_table * const pt,
-		 const struct i915_page_scratch * const scratch)
+		 const struct drm_i915_gem_object * const scratch)
 {
 	bool free = false;
 
@@ -155,19 +175,16 @@ struct i915_ppgtt *i915_ppgtt_create(struct intel_gt *gt)
 	return ppgtt;
 }
 
-int ppgtt_bind_vma(struct i915_address_space *vm,
-		   struct i915_vma *vma,
-		   enum i915_cache_level cache_level,
-		   u32 flags)
+void ppgtt_bind_vma(struct i915_address_space *vm,
+		    struct i915_vm_pt_stash *stash,
+		    struct i915_vma *vma,
+		    enum i915_cache_level cache_level,
+		    u32 flags)
 {
 	u32 pte_flags;
-	int err;
 
 	if (!test_bit(I915_VMA_ALLOC_BIT, __i915_vma_flags(vma))) {
-		err = vm->allocate_va_range(vm, vma->node.start, vma->size);
-		if (err)
-			return err;
-
+		vm->allocate_va_range(vm, stash, vma->node.start, vma->size);
 		set_bit(I915_VMA_ALLOC_BIT, __i915_vma_flags(vma));
 	}
 
@@ -178,8 +195,6 @@ int ppgtt_bind_vma(struct i915_address_space *vm,
 
 	vm->insert_entries(vm, vma, cache_level, pte_flags);
 	wmb();
-
-	return 0;
 }
 
 void ppgtt_unbind_vma(struct i915_address_space *vm, struct i915_vma *vma)
@@ -188,12 +203,93 @@ void ppgtt_unbind_vma(struct i915_address_space *vm, struct i915_vma *vma)
 		vm->clear_range(vm, vma->node.start, vma->size);
 }
 
+static unsigned long pd_count(u64 size, int shift)
+{
+	/* Beware later misalignment */
+	return (size + 2 * (BIT_ULL(shift) - 1)) >> shift;
+}
+
+int i915_vm_alloc_pt_stash(struct i915_address_space *vm,
+			   struct i915_vm_pt_stash *stash,
+			   u64 size)
+{
+	unsigned long count;
+	int shift, n;
+
+	shift = vm->pd_shift;
+	if (!shift)
+		return 0;
+
+	count = pd_count(size, shift);
+	while (count--) {
+		struct i915_page_table *pt;
+
+		pt = alloc_pt(vm);
+		if (IS_ERR(pt)) {
+			i915_vm_free_pt_stash(vm, stash);
+			return PTR_ERR(pt);
+		}
+
+		pt->stash = stash->pt[0];
+		stash->pt[0] = pt;
+	}
+
+	for (n = 1; n < vm->top; n++) {
+		shift += ilog2(I915_PDES); /* Each PD holds 512 entries */
+		count = pd_count(size, shift);
+		while (count--) {
+			struct i915_page_directory *pd;
+
+			pd = alloc_pd(vm);
+			if (IS_ERR(pd)) {
+				i915_vm_free_pt_stash(vm, stash);
+				return PTR_ERR(pd);
+			}
+
+			pd->pt.stash = stash->pt[1];
+			stash->pt[1] = &pd->pt;
+		}
+	}
+
+	return 0;
+}
+
+int i915_vm_pin_pt_stash(struct i915_address_space *vm,
+			 struct i915_vm_pt_stash *stash)
+{
+	struct i915_page_table *pt;
+	int n, err;
+
+	for (n = 0; n < ARRAY_SIZE(stash->pt); n++) {
+		for (pt = stash->pt[n]; pt; pt = pt->stash) {
+			err = pin_pt_dma(vm, pt->base);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+void i915_vm_free_pt_stash(struct i915_address_space *vm,
+			   struct i915_vm_pt_stash *stash)
+{
+	struct i915_page_table *pt;
+	int n;
+
+	for (n = 0; n < ARRAY_SIZE(stash->pt); n++) {
+		while ((pt = stash->pt[n])) {
+			stash->pt[n] = pt->stash;
+			free_px(vm, pt, n);
+		}
+	}
+}
+
 int ppgtt_set_pages(struct i915_vma *vma)
 {
 	GEM_BUG_ON(vma->pages);
 
 	vma->pages = vma->obj->mm.pages;
-
 	vma->page_sizes = vma->obj->mm.page_sizes;
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/gt/intel_renderstate.c b/drivers/gpu/drm/i915/gt/intel_renderstate.c
index 1bfad589c63b..ea2a77c7b469 100644
--- a/drivers/gpu/drm/i915/gt/intel_renderstate.c
+++ b/drivers/gpu/drm/i915/gt/intel_renderstate.c
@@ -27,6 +27,7 @@
 
 #include "i915_drv.h"
 #include "intel_renderstate.h"
+#include "gt/intel_context.h"
 #include "intel_ring.h"
 
 static const struct intel_renderstate_rodata *
@@ -157,33 +158,47 @@ out:
 #undef OUT_BATCH
 
 int intel_renderstate_init(struct intel_renderstate *so,
-			   struct intel_engine_cs *engine)
+			   struct intel_context *ce)
 {
-	struct drm_i915_gem_object *obj;
+	struct intel_engine_cs *engine = ce->engine;
+	struct drm_i915_gem_object *obj = NULL;
 	int err;
 
 	memset(so, 0, sizeof(*so));
 
 	so->rodata = render_state_get_rodata(engine);
-	if (!so->rodata)
-		return 0;
+	if (so->rodata) {
+		if (so->rodata->batch_items * 4 > PAGE_SIZE)
+			return -EINVAL;
+
+		obj = i915_gem_object_create_internal(engine->i915, PAGE_SIZE);
+		if (IS_ERR(obj))
+			return PTR_ERR(obj);
+
+		so->vma = i915_vma_instance(obj, &engine->gt->ggtt->vm, NULL);
+		if (IS_ERR(so->vma)) {
+			err = PTR_ERR(so->vma);
+			goto err_obj;
+		}
+	}
 
-	if (so->rodata->batch_items * 4 > PAGE_SIZE)
-		return -EINVAL;
+	i915_gem_ww_ctx_init(&so->ww, true);
+retry:
+	err = intel_context_pin_ww(ce, &so->ww);
+	if (err)
+		goto err_fini;
 
-	obj = i915_gem_object_create_internal(engine->i915, PAGE_SIZE);
-	if (IS_ERR(obj))
-		return PTR_ERR(obj);
+	/* return early if there's nothing to setup */
+	if (!err && !so->rodata)
+		return 0;
 
-	so->vma = i915_vma_instance(obj, &engine->gt->ggtt->vm, NULL);
-	if (IS_ERR(so->vma)) {
-		err = PTR_ERR(so->vma);
-		goto err_obj;
-	}
+	err = i915_gem_object_lock(so->vma->obj, &so->ww);
+	if (err)
+		goto err_context;
 
 	err = i915_vma_pin(so->vma, 0, 0, PIN_GLOBAL | PIN_HIGH);
 	if (err)
-		goto err_obj;
+		goto err_context;
 
 	err = render_state_setup(so, engine->i915);
 	if (err)
@@ -193,8 +208,18 @@ int intel_renderstate_init(struct intel_renderstate *so,
 
 err_unpin:
 	i915_vma_unpin(so->vma);
+err_context:
+	intel_context_unpin(ce);
+err_fini:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&so->ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&so->ww);
 err_obj:
-	i915_gem_object_put(obj);
+	if (obj)
+		i915_gem_object_put(obj);
 	so->vma = NULL;
 	return err;
 }
@@ -208,11 +233,9 @@ int intel_renderstate_emit(struct intel_renderstate *so,
 	if (!so->vma)
 		return 0;
 
-	i915_vma_lock(so->vma);
 	err = i915_request_await_object(rq, so->vma->obj, false);
 	if (err == 0)
 		err = i915_vma_move_to_active(so->vma, rq, 0);
-	i915_vma_unlock(so->vma);
 	if (err)
 		return err;
 
@@ -233,7 +256,17 @@ int intel_renderstate_emit(struct intel_renderstate *so,
 	return 0;
 }
 
-void intel_renderstate_fini(struct intel_renderstate *so)
+void intel_renderstate_fini(struct intel_renderstate *so,
+			    struct intel_context *ce)
 {
-	i915_vma_unpin_and_release(&so->vma, 0);
+	if (so->vma) {
+		i915_vma_unpin(so->vma);
+		i915_vma_close(so->vma);
+	}
+
+	intel_context_unpin(ce);
+	i915_gem_ww_ctx_fini(&so->ww);
+
+	if (so->vma)
+		i915_gem_object_put(so->vma->obj);
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_renderstate.h b/drivers/gpu/drm/i915/gt/intel_renderstate.h
index 5700be69a05a..713aa1e86c80 100644
--- a/drivers/gpu/drm/i915/gt/intel_renderstate.h
+++ b/drivers/gpu/drm/i915/gt/intel_renderstate.h
@@ -25,9 +25,10 @@
 #define _INTEL_RENDERSTATE_H_
 
 #include <linux/types.h>
+#include "i915_gem.h"
 
 struct i915_request;
-struct intel_engine_cs;
+struct intel_context;
 struct i915_vma;
 
 struct intel_renderstate_rodata {
@@ -49,6 +50,7 @@ extern const struct intel_renderstate_rodata gen8_null_state;
 extern const struct intel_renderstate_rodata gen9_null_state;
 
 struct intel_renderstate {
+	struct i915_gem_ww_ctx ww;
 	const struct intel_renderstate_rodata *rodata;
 	struct i915_vma *vma;
 	u32 batch_offset;
@@ -58,9 +60,10 @@ struct intel_renderstate {
 };
 
 int intel_renderstate_init(struct intel_renderstate *so,
-			   struct intel_engine_cs *engine);
+			   struct intel_context *ce);
 int intel_renderstate_emit(struct intel_renderstate *so,
 			   struct i915_request *rq);
-void intel_renderstate_fini(struct intel_renderstate *so);
+void intel_renderstate_fini(struct intel_renderstate *so,
+			    struct intel_context *ce);
 
 #endif /* _INTEL_RENDERSTATE_H_ */
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 46a5ceffc22f..ac36b67fb46b 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -15,6 +15,7 @@
 #include "i915_drv.h"
 #include "i915_gpu_error.h"
 #include "i915_irq.h"
+#include "intel_breadcrumbs.h"
 #include "intel_engine_pm.h"
 #include "intel_gt.h"
 #include "intel_gt_pm.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c
index bdb324167ef3..4034a4bac7f0 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring.c
+++ b/drivers/gpu/drm/i915/gt/intel_ring.c
@@ -21,7 +21,13 @@ unsigned int intel_ring_update_space(struct intel_ring *ring)
 	return space;
 }
 
-int intel_ring_pin(struct intel_ring *ring)
+void __intel_ring_pin(struct intel_ring *ring)
+{
+	GEM_BUG_ON(!atomic_read(&ring->pin_count));
+	atomic_inc(&ring->pin_count);
+}
+
+int intel_ring_pin(struct intel_ring *ring, struct i915_gem_ww_ctx *ww)
 {
 	struct i915_vma *vma = ring->vma;
 	unsigned int flags;
@@ -39,7 +45,7 @@ int intel_ring_pin(struct intel_ring *ring)
 	else
 		flags |= PIN_HIGH;
 
-	ret = i915_ggtt_pin(vma, 0, flags);
+	ret = i915_ggtt_pin(vma, ww, 0, flags);
 	if (unlikely(ret))
 		goto err_unpin;
 
diff --git a/drivers/gpu/drm/i915/gt/intel_ring.h b/drivers/gpu/drm/i915/gt/intel_ring.h
index cc0ebca65167..1700579bdc93 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring.h
+++ b/drivers/gpu/drm/i915/gt/intel_ring.h
@@ -21,7 +21,8 @@ int intel_ring_cacheline_align(struct i915_request *rq);
 
 unsigned int intel_ring_update_space(struct intel_ring *ring);
 
-int intel_ring_pin(struct intel_ring *ring);
+void __intel_ring_pin(struct intel_ring *ring);
+int intel_ring_pin(struct intel_ring *ring, struct i915_gem_ww_ctx *ww);
 void intel_ring_unpin(struct intel_ring *ring);
 void intel_ring_reset(struct intel_ring *ring, u32 tail);
 
diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
index 898593ca4889..16b48e72c369 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
@@ -32,6 +32,7 @@
 #include "gen6_ppgtt.h"
 #include "gen7_renderclear.h"
 #include "i915_drv.h"
+#include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_gt.h"
 #include "intel_reset.h"
@@ -201,16 +202,18 @@ static struct i915_address_space *vm_alias(struct i915_address_space *vm)
 	return vm;
 }
 
+static u32 pp_dir(struct i915_address_space *vm)
+{
+	return to_gen6_ppgtt(i915_vm_to_ppgtt(vm))->pp_dir;
+}
+
 static void set_pp_dir(struct intel_engine_cs *engine)
 {
 	struct i915_address_space *vm = vm_alias(engine->gt->vm);
 
 	if (vm) {
-		struct i915_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
-
 		ENGINE_WRITE(engine, RING_PP_DIR_DCLV, PP_DIR_DCLV_2G);
-		ENGINE_WRITE(engine, RING_PP_DIR_BASE,
-			     px_base(ppgtt->pd)->ggtt_offset << 10);
+		ENGINE_WRITE(engine, RING_PP_DIR_BASE, pp_dir(vm));
 	}
 }
 
@@ -255,7 +258,7 @@ static int xcs_resume(struct intel_engine_cs *engine)
 	else
 		ring_setup_status_page(engine);
 
-	intel_engine_reset_breadcrumbs(engine);
+	intel_breadcrumbs_reset(engine->breadcrumbs);
 
 	/* Enforce ordering by reading HEAD register back */
 	ENGINE_POSTING_READ(engine, RING_HEAD);
@@ -474,14 +477,16 @@ static void ring_context_destroy(struct kref *ref)
 	intel_context_free(ce);
 }
 
-static int __context_pin_ppgtt(struct intel_context *ce)
+static int ring_context_pre_pin(struct intel_context *ce,
+				struct i915_gem_ww_ctx *ww,
+				void **unused)
 {
 	struct i915_address_space *vm;
 	int err = 0;
 
 	vm = vm_alias(ce->vm);
 	if (vm)
-		err = gen6_ppgtt_pin(i915_vm_to_ppgtt((vm)));
+		err = gen6_ppgtt_pin(i915_vm_to_ppgtt((vm)), ww);
 
 	return err;
 }
@@ -497,6 +502,10 @@ static void __context_unpin_ppgtt(struct intel_context *ce)
 
 static void ring_context_unpin(struct intel_context *ce)
 {
+}
+
+static void ring_context_post_unpin(struct intel_context *ce)
+{
 	__context_unpin_ppgtt(ce);
 }
 
@@ -584,9 +593,9 @@ static int ring_context_alloc(struct intel_context *ce)
 	return 0;
 }
 
-static int ring_context_pin(struct intel_context *ce)
+static int ring_context_pin(struct intel_context *ce, void *unused)
 {
-	return __context_pin_ppgtt(ce);
+	return 0;
 }
 
 static void ring_context_reset(struct intel_context *ce)
@@ -597,8 +606,10 @@ static void ring_context_reset(struct intel_context *ce)
 static const struct intel_context_ops ring_context_ops = {
 	.alloc = ring_context_alloc,
 
+	.pre_pin = ring_context_pre_pin,
 	.pin = ring_context_pin,
 	.unpin = ring_context_unpin,
+	.post_unpin = ring_context_post_unpin,
 
 	.enter = intel_context_enter_engine,
 	.exit = intel_context_exit_engine,
@@ -608,7 +619,7 @@ static const struct intel_context_ops ring_context_ops = {
 };
 
 static int load_pd_dir(struct i915_request *rq,
-		       const struct i915_ppgtt *ppgtt,
+		       struct i915_address_space *vm,
 		       u32 valid)
 {
 	const struct intel_engine_cs * const engine = rq->engine;
@@ -624,7 +635,7 @@ static int load_pd_dir(struct i915_request *rq,
 
 	*cs++ = MI_LOAD_REGISTER_IMM(1);
 	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine->mmio_base));
-	*cs++ = px_base(ppgtt->pd)->ggtt_offset << 10;
+	*cs++ = pp_dir(vm);
 
 	/* Stall until the page table load is complete? */
 	*cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT;
@@ -826,7 +837,7 @@ static int switch_mm(struct i915_request *rq, struct i915_address_space *vm)
 	 * post-sync op, this extra pass appears vital before a
 	 * mm switch!
 	 */
-	ret = load_pd_dir(rq, i915_vm_to_ppgtt(vm), PP_DIR_DCLV_2G);
+	ret = load_pd_dir(rq, vm, PP_DIR_DCLV_2G);
 	if (ret)
 		return ret;
 
@@ -1250,14 +1261,15 @@ int intel_ring_submission_setup(struct intel_engine_cs *engine)
 		return -ENODEV;
 	}
 
-	timeline = intel_timeline_create(engine->gt, engine->status_page.vma);
+	timeline = intel_timeline_create_from_engine(engine,
+						     I915_GEM_HWS_SEQNO_ADDR);
 	if (IS_ERR(timeline)) {
 		err = PTR_ERR(timeline);
 		goto err;
 	}
 	GEM_BUG_ON(timeline->has_initial_breadcrumb);
 
-	err = intel_timeline_pin(timeline);
+	err = intel_timeline_pin(timeline, NULL);
 	if (err)
 		goto err_timeline;
 
@@ -1267,7 +1279,7 @@ int intel_ring_submission_setup(struct intel_engine_cs *engine)
 		goto err_timeline_unpin;
 	}
 
-	err = intel_ring_pin(ring);
+	err = intel_ring_pin(ring, NULL);
 	if (err)
 		goto err_ring;
 
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 97ba14ad52e4..e6a00eea0631 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -7,6 +7,7 @@
 #include <drm/i915_drm.h>
 
 #include "i915_drv.h"
+#include "intel_breadcrumbs.h"
 #include "intel_gt.h"
 #include "intel_gt_clock_utils.h"
 #include "intel_gt_irq.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.c b/drivers/gpu/drm/i915/gt/intel_timeline.c
index 46d20f5f3ddc..a2f74cefe4c3 100644
--- a/drivers/gpu/drm/i915/gt/intel_timeline.c
+++ b/drivers/gpu/drm/i915/gt/intel_timeline.c
@@ -215,7 +215,8 @@ static void cacheline_free(struct intel_timeline_cacheline *cl)
 
 static int intel_timeline_init(struct intel_timeline *timeline,
 			       struct intel_gt *gt,
-			       struct i915_vma *hwsp)
+			       struct i915_vma *hwsp,
+			       unsigned int offset)
 {
 	void *vaddr;
 
@@ -246,8 +247,7 @@ static int intel_timeline_init(struct intel_timeline *timeline,
 
 		vaddr = page_mask_bits(cl->vaddr);
 	} else {
-		timeline->hwsp_offset = I915_GEM_HWS_SEQNO_ADDR;
-
+		timeline->hwsp_offset = offset;
 		vaddr = i915_gem_object_pin_map(hwsp->obj, I915_MAP_WB);
 		if (IS_ERR(vaddr))
 			return PTR_ERR(vaddr);
@@ -297,7 +297,9 @@ static void intel_timeline_fini(struct intel_timeline *timeline)
 }
 
 struct intel_timeline *
-intel_timeline_create(struct intel_gt *gt, struct i915_vma *global_hwsp)
+__intel_timeline_create(struct intel_gt *gt,
+			struct i915_vma *global_hwsp,
+			unsigned int offset)
 {
 	struct intel_timeline *timeline;
 	int err;
@@ -306,7 +308,7 @@ intel_timeline_create(struct intel_gt *gt, struct i915_vma *global_hwsp)
 	if (!timeline)
 		return ERR_PTR(-ENOMEM);
 
-	err = intel_timeline_init(timeline, gt, global_hwsp);
+	err = intel_timeline_init(timeline, gt, global_hwsp, offset);
 	if (err) {
 		kfree(timeline);
 		return ERR_PTR(err);
@@ -315,14 +317,20 @@ intel_timeline_create(struct intel_gt *gt, struct i915_vma *global_hwsp)
 	return timeline;
 }
 
-int intel_timeline_pin(struct intel_timeline *tl)
+void __intel_timeline_pin(struct intel_timeline *tl)
+{
+	GEM_BUG_ON(!atomic_read(&tl->pin_count));
+	atomic_inc(&tl->pin_count);
+}
+
+int intel_timeline_pin(struct intel_timeline *tl, struct i915_gem_ww_ctx *ww)
 {
 	int err;
 
 	if (atomic_add_unless(&tl->pin_count, 1, 0))
 		return 0;
 
-	err = i915_ggtt_pin(tl->hwsp_ggtt, 0, PIN_HIGH);
+	err = i915_ggtt_pin(tl->hwsp_ggtt, ww, 0, PIN_HIGH);
 	if (err)
 		return err;
 
@@ -465,7 +473,7 @@ __intel_timeline_get_seqno(struct intel_timeline *tl,
 		goto err_rollback;
 	}
 
-	err = i915_ggtt_pin(vma, 0, PIN_HIGH);
+	err = i915_ggtt_pin(vma, NULL, 0, PIN_HIGH);
 	if (err) {
 		__idle_hwsp_free(vma->private, cacheline);
 		goto err_rollback;
@@ -484,7 +492,9 @@ __intel_timeline_get_seqno(struct intel_timeline *tl,
 	 * free it after the current request is retired, which ensures that
 	 * all writes into the cacheline from previous requests are complete.
 	 */
-	err = i915_active_ref(&tl->hwsp_cacheline->active, tl, &rq->fence);
+	err = i915_active_ref(&tl->hwsp_cacheline->active,
+			      tl->fence_context,
+			      &rq->fence);
 	if (err)
 		goto err_cacheline;
 
diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.h b/drivers/gpu/drm/i915/gt/intel_timeline.h
index 4298b9ac7327..9882cd911d8e 100644
--- a/drivers/gpu/drm/i915/gt/intel_timeline.h
+++ b/drivers/gpu/drm/i915/gt/intel_timeline.h
@@ -29,10 +29,27 @@
 
 #include "i915_active.h"
 #include "i915_syncmap.h"
-#include "gt/intel_timeline_types.h"
+#include "intel_timeline_types.h"
 
 struct intel_timeline *
-intel_timeline_create(struct intel_gt *gt, struct i915_vma *global_hwsp);
+__intel_timeline_create(struct intel_gt *gt,
+			struct i915_vma *global_hwsp,
+			unsigned int offset);
+
+static inline struct intel_timeline *
+intel_timeline_create(struct intel_gt *gt)
+{
+	return __intel_timeline_create(gt, NULL, 0);
+}
+
+static inline struct intel_timeline *
+intel_timeline_create_from_engine(struct intel_engine_cs *engine,
+				  unsigned int offset)
+{
+	return __intel_timeline_create(engine->gt,
+				       engine->status_page.vma,
+				       offset);
+}
 
 static inline struct intel_timeline *
 intel_timeline_get(struct intel_timeline *timeline)
@@ -71,7 +88,8 @@ static inline bool intel_timeline_sync_is_later(struct intel_timeline *tl,
 	return __intel_timeline_sync_is_later(tl, fence->context, fence->seqno);
 }
 
-int intel_timeline_pin(struct intel_timeline *tl);
+void __intel_timeline_pin(struct intel_timeline *tl);
+int intel_timeline_pin(struct intel_timeline *tl, struct i915_gem_ww_ctx *ww);
 void intel_timeline_enter(struct intel_timeline *tl);
 int intel_timeline_get_seqno(struct intel_timeline *tl,
 			     struct i915_request *rq,
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 5726cd0a37e0..6c580d0d9ea8 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -52,6 +52,37 @@
  * - Public functions to init or apply the given workaround type.
  */
 
+/*
+ * KBL revision ID ordering is bizarre; higher revision ID's map to lower
+ * steppings in some cases.  So rather than test against the revision ID
+ * directly, let's map that into our own range of increasing ID's that we
+ * can test against in a regular manner.
+ */
+
+const struct i915_rev_steppings kbl_revids[] = {
+	[0] = { .gt_stepping = KBL_REVID_A0, .disp_stepping = KBL_REVID_A0 },
+	[1] = { .gt_stepping = KBL_REVID_B0, .disp_stepping = KBL_REVID_B0 },
+	[2] = { .gt_stepping = KBL_REVID_C0, .disp_stepping = KBL_REVID_B0 },
+	[3] = { .gt_stepping = KBL_REVID_D0, .disp_stepping = KBL_REVID_B0 },
+	[4] = { .gt_stepping = KBL_REVID_F0, .disp_stepping = KBL_REVID_C0 },
+	[5] = { .gt_stepping = KBL_REVID_C0, .disp_stepping = KBL_REVID_B1 },
+	[6] = { .gt_stepping = KBL_REVID_D1, .disp_stepping = KBL_REVID_B1 },
+	[7] = { .gt_stepping = KBL_REVID_G0, .disp_stepping = KBL_REVID_C0 },
+};
+
+const struct i915_rev_steppings tgl_uy_revids[] = {
+	[0] = { .gt_stepping = TGL_REVID_A0, .disp_stepping = TGL_REVID_A0 },
+	[1] = { .gt_stepping = TGL_REVID_B0, .disp_stepping = TGL_REVID_C0 },
+	[2] = { .gt_stepping = TGL_REVID_B1, .disp_stepping = TGL_REVID_C0 },
+	[3] = { .gt_stepping = TGL_REVID_C0, .disp_stepping = TGL_REVID_D0 },
+};
+
+/* Same GT stepping between tgl_uy_revids and tgl_revids don't mean the same HW */
+const struct i915_rev_steppings tgl_revids[] = {
+	[0] = { .gt_stepping = TGL_REVID_A0, .disp_stepping = TGL_REVID_B0 },
+	[1] = { .gt_stepping = TGL_REVID_B0, .disp_stepping = TGL_REVID_D0 },
+};
+
 static void wa_init_start(struct i915_wa_list *wal, const char *name, const char *engine_name)
 {
 	wal->name = name;
@@ -470,7 +501,7 @@ static void kbl_ctx_workarounds_init(struct intel_engine_cs *engine,
 	gen9_ctx_workarounds_init(engine, wal);
 
 	/* WaToEnableHwFixForPushConstHWBug:kbl */
-	if (IS_KBL_REVID(i915, KBL_REVID_C0, REVID_FOREVER))
+	if (IS_KBL_GT_REVID(i915, KBL_REVID_C0, REVID_FOREVER))
 		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
 				  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
 
@@ -596,8 +627,8 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine,
 	wa_masked_en(wal, GEN9_ROW_CHICKEN4, GEN11_DIS_PICK_2ND_EU);
 }
 
-static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine,
-				     struct i915_wa_list *wal)
+static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine,
+				       struct i915_wa_list *wal)
 {
 	/*
 	 * Wa_1409142259:tgl
@@ -607,12 +638,28 @@ static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine,
 	 * Wa_1409207793:tgl
 	 * Wa_1409178076:tgl
 	 * Wa_1408979724:tgl
+	 * Wa_14010443199:rkl
+	 * Wa_14010698770:rkl
 	 */
 	WA_SET_BIT_MASKED(GEN11_COMMON_SLICE_CHICKEN3,
 			  GEN12_DISABLE_CPS_AWARE_COLOR_PIPE);
 
+	/* WaDisableGPGPUMidThreadPreemption:gen12 */
+	WA_SET_FIELD_MASKED(GEN8_CS_CHICKEN1,
+			    GEN9_PREEMPT_GPGPU_LEVEL_MASK,
+			    GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL);
+}
+
+static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine,
+				     struct i915_wa_list *wal)
+{
+	gen12_ctx_workarounds_init(engine, wal);
+
 	/*
-	 * Wa_1604555607:gen12 and Wa_1608008084:gen12
+	 * Wa_1604555607:tgl,rkl
+	 *
+	 * Note that the implementation of this workaround is further modified
+	 * according to the FF_MODE2 guidance given by Wa_1608008084:gen12.
 	 * FF_MODE2 register will return the wrong value when read. The default
 	 * value for this register is zero for all fields and there are no bit
 	 * masks. So instead of doing a RMW we should just write the GS Timer
@@ -623,11 +670,6 @@ static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine,
 	       FF_MODE2_GS_TIMER_MASK | FF_MODE2_TDS_TIMER_MASK,
 	       FF_MODE2_GS_TIMER_224  | FF_MODE2_TDS_TIMER_128,
 	       0);
-
-	/* WaDisableGPGPUMidThreadPreemption:tgl */
-	WA_SET_FIELD_MASKED(GEN8_CS_CHICKEN1,
-			    GEN9_PREEMPT_GPGPU_LEVEL_MASK,
-			    GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL);
 }
 
 static void
@@ -642,8 +684,10 @@ __intel_engine_init_ctx_wa(struct intel_engine_cs *engine,
 
 	wa_init_start(wal, name, engine->name);
 
-	if (IS_GEN(i915, 12))
+	if (IS_ROCKETLAKE(i915) || IS_TIGERLAKE(i915))
 		tgl_ctx_workarounds_init(engine, wal);
+	else if (IS_GEN(i915, 12))
+		gen12_ctx_workarounds_init(engine, wal);
 	else if (IS_GEN(i915, 11))
 		icl_ctx_workarounds_init(engine, wal);
 	else if (IS_CANNONLAKE(i915))
@@ -995,7 +1039,7 @@ kbl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 	gen9_gt_workarounds_init(i915, wal);
 
 	/* WaDisableDynamicCreditSharing:kbl */
-	if (IS_KBL_REVID(i915, 0, KBL_REVID_B0))
+	if (IS_KBL_GT_REVID(i915, 0, KBL_REVID_B0))
 		wa_write_or(wal,
 			    GAMT_CHKN_BIT_REG,
 			    GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING);
@@ -1176,18 +1220,25 @@ icl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 }
 
 static void
-tgl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
+gen12_gt_workarounds_init(struct drm_i915_private *i915,
+			  struct i915_wa_list *wal)
 {
 	wa_init_mcr(i915, wal);
+}
+
+static void
+tgl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
+{
+	gen12_gt_workarounds_init(i915, wal);
 
 	/* Wa_1409420604:tgl */
-	if (IS_TGL_REVID(i915, TGL_REVID_A0, TGL_REVID_A0))
+	if (IS_TGL_UY_GT_REVID(i915, TGL_REVID_A0, TGL_REVID_A0))
 		wa_write_or(wal,
 			    SUBSLICE_UNIT_LEVEL_CLKGATE2,
 			    CPSSUNIT_CLKGATE_DIS);
 
 	/* Wa_1607087056:tgl also know as BUG:1409180338 */
-	if (IS_TGL_REVID(i915, TGL_REVID_A0, TGL_REVID_A0))
+	if (IS_TGL_UY_GT_REVID(i915, TGL_REVID_A0, TGL_REVID_A0))
 		wa_write_or(wal,
 			    SLICE_UNIT_LEVEL_CLKGATE,
 			    L3_CLKGATE_DIS | L3_CR2X_CLKGATE_DIS);
@@ -1196,8 +1247,10 @@ tgl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 static void
 gt_init_workarounds(struct drm_i915_private *i915, struct i915_wa_list *wal)
 {
-	if (IS_GEN(i915, 12))
+	if (IS_TIGERLAKE(i915))
 		tgl_gt_workarounds_init(i915, wal);
+	else if (IS_GEN(i915, 12))
+		gen12_gt_workarounds_init(i915, wal);
 	else if (IS_GEN(i915, 11))
 		icl_gt_workarounds_init(i915, wal);
 	else if (IS_CANNONLAKE(i915))
@@ -1620,7 +1673,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 {
 	struct drm_i915_private *i915 = engine->i915;
 
-	if (IS_TGL_REVID(i915, TGL_REVID_A0, TGL_REVID_A0)) {
+	if (IS_TGL_UY_GT_REVID(i915, TGL_REVID_A0, TGL_REVID_A0)) {
 		/*
 		 * Wa_1607138336:tgl
 		 * Wa_1607063988:tgl
@@ -1630,18 +1683,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 			    GEN12_DISABLE_POSH_BUSY_FF_DOP_CG);
 
 		/*
-		 * Wa_1607030317:tgl
-		 * Wa_1607186500:tgl
-		 * Wa_1607297627:tgl there is 3 entries for this WA on BSpec, 2
-		 * of then says it is fixed on B0 the other one says it is
-		 * permanent
-		 */
-		wa_masked_en(wal,
-			     GEN6_RC_SLEEP_PSMI_CONTROL,
-			     GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE |
-			     GEN8_RC_SEMA_IDLE_MSG_DISABLE);
-
-		/*
 		 * Wa_1606679103:tgl
 		 * (see also Wa_1606682166:icl)
 		 */
@@ -1654,22 +1695,17 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 			    VSUNIT_CLKGATE_DIS_TGL);
 	}
 
-	if (IS_TIGERLAKE(i915)) {
-		/* Wa_1606931601:tgl */
+	if (IS_ROCKETLAKE(i915) || IS_TIGERLAKE(i915)) {
+		/* Wa_1606931601:tgl,rkl */
 		wa_masked_en(wal, GEN7_ROW_CHICKEN2, GEN12_DISABLE_EARLY_READ);
 
-		/* Wa_1409804808:tgl */
+		/* Wa_1409804808:tgl,rkl */
 		wa_masked_en(wal, GEN7_ROW_CHICKEN2,
 			     GEN12_PUSH_CONST_DEREF_HOLD_DIS);
 
-		/* Wa_1606700617:tgl */
-		wa_masked_en(wal,
-			     GEN9_CS_DEBUG_MODE1,
-			     FF_DOP_CLOCK_GATE_DISABLE);
-
 		/*
 		 * Wa_1409085225:tgl
-		 * Wa_14010229206:tgl
+		 * Wa_14010229206:tgl,rkl
 		 */
 		wa_masked_en(wal, GEN9_ROW_CHICKEN4, GEN12_DISABLE_TDL_PUSH);
 
@@ -1677,9 +1713,37 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 		 * Wa_1407928979:tgl A*
 		 * Wa_18011464164:tgl B0+
 		 * Wa_22010931296:tgl B0+
+		 * Wa_14010919138:rkl,tgl
 		 */
 		wa_write_or(wal, GEN7_FF_THREAD_MODE,
 			    GEN12_FF_TESSELATION_DOP_GATE_DISABLE);
+
+		/*
+		 * Wa_1607030317:tgl
+		 * Wa_1607186500:tgl
+		 * Wa_1607297627:tgl,rkl there are multiple entries for this
+		 * WA in the BSpec; some indicate this is an A0-only WA,
+		 * others indicate it applies to all steppings.
+		 */
+		wa_masked_en(wal,
+			     GEN6_RC_SLEEP_PSMI_CONTROL,
+			     GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE |
+			     GEN8_RC_SEMA_IDLE_MSG_DISABLE);
+
+		/*
+		 * Wa_1606700617:tgl
+		 * Wa_22010271021:tgl,rkl
+		 */
+		wa_masked_en(wal,
+			     GEN9_CS_DEBUG_MODE1,
+			     FF_DOP_CLOCK_GATE_DISABLE);
+	}
+
+	if (IS_GEN(i915, 12)) {
+		/* Wa_1406941453:gen12 */
+		wa_masked_en(wal,
+			     GEN10_SAMPLER_MODE,
+			     ENABLE_SMALLPL);
 	}
 
 	if (IS_GEN(i915, 11)) {
@@ -1898,7 +1962,7 @@ xcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 	struct drm_i915_private *i915 = engine->i915;
 
 	/* WaKBLVECSSemaphoreWaitPoll:kbl */
-	if (IS_KBL_REVID(i915, KBL_REVID_A0, KBL_REVID_E0)) {
+	if (IS_KBL_GT_REVID(i915, KBL_REVID_A0, KBL_REVID_E0)) {
 		wa_write(wal,
 			 RING_SEMA_WAIT_POLL(engine->mmio_base),
 			 1);
@@ -2045,6 +2109,7 @@ static int engine_wa_list_verify(struct intel_context *ce,
 	const struct i915_wa *wa;
 	struct i915_request *rq;
 	struct i915_vma *vma;
+	struct i915_gem_ww_ctx ww;
 	unsigned int i;
 	u32 *results;
 	int err;
@@ -2057,29 +2122,34 @@ static int engine_wa_list_verify(struct intel_context *ce,
 		return PTR_ERR(vma);
 
 	intel_engine_pm_get(ce->engine);
-	rq = intel_context_create_request(ce);
-	intel_engine_pm_put(ce->engine);
+	i915_gem_ww_ctx_init(&ww, false);
+retry:
+	err = i915_gem_object_lock(vma->obj, &ww);
+	if (err == 0)
+		err = intel_context_pin_ww(ce, &ww);
+	if (err)
+		goto err_pm;
+
+	rq = i915_request_create(ce);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
-		goto err_vma;
+		goto err_unpin;
 	}
 
-	i915_vma_lock(vma);
 	err = i915_request_await_object(rq, vma->obj, true);
 	if (err == 0)
 		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
-	if (err) {
-		i915_request_add(rq);
-		goto err_vma;
-	}
-
-	err = wa_list_srm(rq, wal, vma);
-	if (err)
-		goto err_vma;
+	if (err == 0)
+		err = wa_list_srm(rq, wal, vma);
 
 	i915_request_get(rq);
+	if (err)
+		i915_request_set_error_once(rq, err);
 	i915_request_add(rq);
+
+	if (err)
+		goto err_rq;
+
 	if (i915_request_wait(rq, 0, HZ / 5) < 0) {
 		err = -ETIME;
 		goto err_rq;
@@ -2104,7 +2174,16 @@ static int engine_wa_list_verify(struct intel_context *ce,
 
 err_rq:
 	i915_request_put(rq);
-err_vma:
+err_unpin:
+	intel_context_unpin(ce);
+err_pm:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	intel_engine_pm_put(ce->engine);
 	i915_vma_unpin(vma);
 	i915_vma_put(vma);
 	return err;
diff --git a/drivers/gpu/drm/i915/gt/mock_engine.c b/drivers/gpu/drm/i915/gt/mock_engine.c
index b8dd3cbc8696..dfd1cfb8a7ec 100644
--- a/drivers/gpu/drm/i915/gt/mock_engine.c
+++ b/drivers/gpu/drm/i915/gt/mock_engine.c
@@ -131,6 +131,10 @@ static void mock_context_unpin(struct intel_context *ce)
 {
 }
 
+static void mock_context_post_unpin(struct intel_context *ce)
+{
+}
+
 static void mock_context_destroy(struct kref *ref)
 {
 	struct intel_context *ce = container_of(ref, typeof(*ce), ref);
@@ -152,8 +156,7 @@ static int mock_context_alloc(struct intel_context *ce)
 	if (!ce->ring)
 		return -ENOMEM;
 
-	GEM_BUG_ON(ce->timeline);
-	ce->timeline = intel_timeline_create(ce->engine->gt, NULL);
+	ce->timeline = intel_timeline_create(ce->engine->gt);
 	if (IS_ERR(ce->timeline)) {
 		kfree(ce->engine);
 		return PTR_ERR(ce->timeline);
@@ -164,7 +167,13 @@ static int mock_context_alloc(struct intel_context *ce)
 	return 0;
 }
 
-static int mock_context_pin(struct intel_context *ce)
+static int mock_context_pre_pin(struct intel_context *ce,
+				struct i915_gem_ww_ctx *ww, void **unused)
+{
+	return 0;
+}
+
+static int mock_context_pin(struct intel_context *ce, void *unused)
 {
 	return 0;
 }
@@ -176,8 +185,10 @@ static void mock_context_reset(struct intel_context *ce)
 static const struct intel_context_ops mock_context_ops = {
 	.alloc = mock_context_alloc,
 
+	.pre_pin = mock_context_pre_pin,
 	.pin = mock_context_pin,
 	.unpin = mock_context_unpin,
+	.post_unpin = mock_context_post_unpin,
 
 	.enter = intel_context_enter_engine,
 	.exit = intel_context_exit_engine,
@@ -261,11 +272,12 @@ static void mock_engine_release(struct intel_engine_cs *engine)
 
 	GEM_BUG_ON(timer_pending(&mock->hw_delay));
 
+	intel_breadcrumbs_free(engine->breadcrumbs);
+
 	intel_context_unpin(engine->kernel_context);
 	intel_context_put(engine->kernel_context);
 
 	intel_engine_fini_retire(engine);
-	intel_engine_fini_breadcrumbs(engine);
 }
 
 struct intel_engine_cs *mock_engine(struct drm_i915_private *i915,
@@ -323,20 +335,26 @@ int mock_engine_init(struct intel_engine_cs *engine)
 	struct intel_context *ce;
 
 	intel_engine_init_active(engine, ENGINE_MOCK);
-	intel_engine_init_breadcrumbs(engine);
 	intel_engine_init_execlists(engine);
 	intel_engine_init__pm(engine);
 	intel_engine_init_retire(engine);
 
+	engine->breadcrumbs = intel_breadcrumbs_create(NULL);
+	if (!engine->breadcrumbs)
+		return -ENOMEM;
+
 	ce = create_kernel_context(engine);
 	if (IS_ERR(ce))
 		goto err_breadcrumbs;
 
+	/* We insist the kernel context is using the status_page */
+	engine->status_page.vma = ce->timeline->hwsp_ggtt;
+
 	engine->kernel_context = ce;
 	return 0;
 
 err_breadcrumbs:
-	intel_engine_fini_breadcrumbs(engine);
+	intel_breadcrumbs_free(engine->breadcrumbs);
 	return -ENOMEM;
 }
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_context.c b/drivers/gpu/drm/i915/gt/selftest_context.c
index 52af1cee9a94..1f4020e906a8 100644
--- a/drivers/gpu/drm/i915/gt/selftest_context.c
+++ b/drivers/gpu/drm/i915/gt/selftest_context.c
@@ -68,6 +68,8 @@ static int context_sync(struct intel_context *ce)
 	} while (!err);
 	mutex_unlock(&tl->mutex);
 
+	/* Wait for all barriers to complete (remote CPU) before we check */
+	i915_active_unlock_wait(&ce->active);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
index 73243ba59c7d..e73854dd2fe0 100644
--- a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
+++ b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
@@ -47,7 +47,10 @@ static int pulse_active(struct i915_active *active)
 
 static void pulse_free(struct kref *kref)
 {
-	kfree(container_of(kref, struct pulse, kref));
+	struct pulse *p = container_of(kref, typeof(*p), kref);
+
+	i915_active_fini(&p->active);
+	kfree(p);
 }
 
 static void pulse_put(struct pulse *p)
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 3fc5de961280..95d41c01d0e0 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -2729,7 +2729,7 @@ static int create_gang(struct intel_engine_cs *engine,
 	i915_gem_object_put(obj);
 	intel_context_put(ce);
 
-	rq->client_link.next = &(*prev)->client_link;
+	rq->mock.link.next = &(*prev)->mock.link;
 	*prev = rq;
 	return 0;
 
@@ -2970,8 +2970,7 @@ static int live_preempt_gang(void *arg)
 		}
 
 		while (rq) { /* wait for each rq from highest to lowest prio */
-			struct i915_request *n =
-				list_next_entry(rq, client_link);
+			struct i915_request *n = list_next_entry(rq, mock.link);
 
 			if (err == 0 && i915_request_wait(rq, 0, HZ / 5) < 0) {
 				struct drm_printer p =
@@ -3090,7 +3089,7 @@ static struct i915_vma *create_global(struct intel_gt *gt, size_t sz)
 		return vma;
 	}
 
-	err = i915_ggtt_pin(vma, 0, 0);
+	err = i915_ggtt_pin(vma, NULL, 0, 0);
 	if (err) {
 		i915_vma_put(vma);
 		return ERR_PTR(err);
@@ -4997,6 +4996,7 @@ static int __live_lrc_state(struct intel_engine_cs *engine,
 {
 	struct intel_context *ce;
 	struct i915_request *rq;
+	struct i915_gem_ww_ctx ww;
 	enum {
 		RING_START_IDX = 0,
 		RING_TAIL_IDX,
@@ -5011,7 +5011,11 @@ static int __live_lrc_state(struct intel_engine_cs *engine,
 	if (IS_ERR(ce))
 		return PTR_ERR(ce);
 
-	err = intel_context_pin(ce);
+	i915_gem_ww_ctx_init(&ww, false);
+retry:
+	err = i915_gem_object_lock(scratch->obj, &ww);
+	if (!err)
+		err = intel_context_pin_ww(ce, &ww);
 	if (err)
 		goto err_put;
 
@@ -5040,11 +5044,9 @@ static int __live_lrc_state(struct intel_engine_cs *engine,
 	*cs++ = i915_ggtt_offset(scratch) + RING_TAIL_IDX * sizeof(u32);
 	*cs++ = 0;
 
-	i915_vma_lock(scratch);
 	err = i915_request_await_object(rq, scratch->obj, true);
 	if (!err)
 		err = i915_vma_move_to_active(scratch, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(scratch);
 
 	i915_request_get(rq);
 	i915_request_add(rq);
@@ -5081,6 +5083,12 @@ err_rq:
 err_unpin:
 	intel_context_unpin(ce);
 err_put:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
 	intel_context_put(ce);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gt/selftest_rps.c b/drivers/gpu/drm/i915/gt/selftest_rps.c
index 8624f5d2a1f3..3540ba9bd459 100644
--- a/drivers/gpu/drm/i915/gt/selftest_rps.c
+++ b/drivers/gpu/drm/i915/gt/selftest_rps.c
@@ -77,20 +77,20 @@ create_spin_counter(struct intel_engine_cs *engine,
 
 	vma = i915_vma_instance(obj, vm, NULL);
 	if (IS_ERR(vma)) {
-		i915_gem_object_put(obj);
-		return vma;
+		err = PTR_ERR(vma);
+		goto err_put;
 	}
 
 	err = i915_vma_pin(vma, 0, 0, PIN_USER);
-	if (err) {
-		i915_vma_put(vma);
-		return ERR_PTR(err);
-	}
+	if (err)
+		goto err_unlock;
+
+	i915_vma_lock(vma);
 
 	base = i915_gem_object_pin_map(obj, I915_MAP_WC);
 	if (IS_ERR(base)) {
-		i915_gem_object_put(obj);
-		return ERR_CAST(base);
+		err = PTR_ERR(base);
+		goto err_unpin;
 	}
 	cs = base;
 
@@ -134,6 +134,14 @@ create_spin_counter(struct intel_engine_cs *engine,
 	*cancel = base + loop;
 	*counter = srm ? memset32(base + end, 0, 1) : NULL;
 	return vma;
+
+err_unpin:
+	i915_vma_unpin(vma);
+err_unlock:
+	i915_vma_unlock(vma);
+err_put:
+	i915_gem_object_put(obj);
+	return ERR_PTR(err);
 }
 
 static u8 wait_for_freq(struct intel_rps *rps, u8 freq, int timeout_ms)
@@ -639,7 +647,6 @@ int live_rps_frequency_cs(void *arg)
 			goto err_vma;
 		}
 
-		i915_vma_lock(vma);
 		err = i915_request_await_object(rq, vma->obj, false);
 		if (!err)
 			err = i915_vma_move_to_active(vma, rq, 0);
@@ -647,7 +654,6 @@ int live_rps_frequency_cs(void *arg)
 			err = rq->engine->emit_bb_start(rq,
 							vma->node.start,
 							PAGE_SIZE, 0);
-		i915_vma_unlock(vma);
 		i915_request_add(rq);
 		if (err)
 			goto err_vma;
@@ -700,7 +706,7 @@ int live_rps_frequency_cs(void *arg)
 				f = act; /* may skip ahead [pcu granularity] */
 			}
 
-			err = -EINVAL;
+			err = -EINTR; /* ignore error, continue on with test */
 		}
 
 err_vma:
@@ -708,6 +714,7 @@ err_vma:
 		i915_gem_object_flush_map(vma->obj);
 		i915_gem_object_unpin_map(vma->obj);
 		i915_vma_unpin(vma);
+		i915_vma_unlock(vma);
 		i915_vma_put(vma);
 
 		st_engine_heartbeat_enable(engine);
@@ -781,7 +788,6 @@ int live_rps_frequency_srm(void *arg)
 			goto err_vma;
 		}
 
-		i915_vma_lock(vma);
 		err = i915_request_await_object(rq, vma->obj, false);
 		if (!err)
 			err = i915_vma_move_to_active(vma, rq, 0);
@@ -789,7 +795,6 @@ int live_rps_frequency_srm(void *arg)
 			err = rq->engine->emit_bb_start(rq,
 							vma->node.start,
 							PAGE_SIZE, 0);
-		i915_vma_unlock(vma);
 		i915_request_add(rq);
 		if (err)
 			goto err_vma;
@@ -841,7 +846,7 @@ int live_rps_frequency_srm(void *arg)
 				f = act; /* may skip ahead [pcu granularity] */
 			}
 
-			err = -EINVAL;
+			err = -EINTR; /* ignore error, continue on with test */
 		}
 
 err_vma:
@@ -849,6 +854,7 @@ err_vma:
 		i915_gem_object_flush_map(vma->obj);
 		i915_gem_object_unpin_map(vma->obj);
 		i915_vma_unpin(vma);
+		i915_vma_unlock(vma);
 		i915_vma_put(vma);
 
 		st_engine_heartbeat_enable(engine);
diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c
index fb5b7d3498a6..19c2cb166e7c 100644
--- a/drivers/gpu/drm/i915/gt/selftest_timeline.c
+++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c
@@ -72,7 +72,7 @@ static int __mock_hwsp_timeline(struct mock_hwsp_freelist *state,
 		unsigned long cacheline;
 		int err;
 
-		tl = intel_timeline_create(state->gt, NULL);
+		tl = intel_timeline_create(state->gt);
 		if (IS_ERR(tl))
 			return PTR_ERR(tl);
 
@@ -158,7 +158,7 @@ out:
 		__mock_hwsp_record(&state, na, NULL);
 	kfree(state.history);
 err_put:
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
@@ -455,7 +455,7 @@ tl_write(struct intel_timeline *tl, struct intel_engine_cs *engine, u32 value)
 	struct i915_request *rq;
 	int err;
 
-	err = intel_timeline_pin(tl);
+	err = intel_timeline_pin(tl, NULL);
 	if (err) {
 		rq = ERR_PTR(err);
 		goto out;
@@ -487,11 +487,11 @@ checked_intel_timeline_create(struct intel_gt *gt)
 {
 	struct intel_timeline *tl;
 
-	tl = intel_timeline_create(gt, NULL);
+	tl = intel_timeline_create(gt);
 	if (IS_ERR(tl))
 		return tl;
 
-	if (*tl->hwsp_seqno != tl->seqno) {
+	if (READ_ONCE(*tl->hwsp_seqno) != tl->seqno) {
 		pr_err("Timeline created with incorrect breadcrumb, found %x, expected %x\n",
 		       *tl->hwsp_seqno, tl->seqno);
 		intel_timeline_put(tl);
@@ -561,9 +561,9 @@ static int live_hwsp_engine(void *arg)
 	for (n = 0; n < count; n++) {
 		struct intel_timeline *tl = timelines[n];
 
-		if (!err && *tl->hwsp_seqno != n) {
-			pr_err("Invalid seqno stored in timeline %lu @ %x, found 0x%x\n",
-			       n, tl->hwsp_offset, *tl->hwsp_seqno);
+		if (!err && READ_ONCE(*tl->hwsp_seqno) != n) {
+			GEM_TRACE_ERR("Invalid seqno:%lu stored in timeline %llu @ %x, found 0x%x\n",
+				      n, tl->fence_context, tl->hwsp_offset, *tl->hwsp_seqno);
 			GEM_TRACE_DUMP();
 			err = -EINVAL;
 		}
@@ -633,9 +633,9 @@ out:
 	for (n = 0; n < count; n++) {
 		struct intel_timeline *tl = timelines[n];
 
-		if (!err && *tl->hwsp_seqno != n) {
-			pr_err("Invalid seqno stored in timeline %lu @ %x, found 0x%x\n",
-			       n, tl->hwsp_offset, *tl->hwsp_seqno);
+		if (!err && READ_ONCE(*tl->hwsp_seqno) != n) {
+			GEM_TRACE_ERR("Invalid seqno:%lu stored in timeline %llu @ %x, found 0x%x\n",
+				      n, tl->fence_context, tl->hwsp_offset, *tl->hwsp_seqno);
 			GEM_TRACE_DUMP();
 			err = -EINVAL;
 		}
@@ -660,14 +660,14 @@ static int live_hwsp_wrap(void *arg)
 	 * foreign GPU references.
 	 */
 
-	tl = intel_timeline_create(gt, NULL);
+	tl = intel_timeline_create(gt);
 	if (IS_ERR(tl))
 		return PTR_ERR(tl);
 
 	if (!tl->has_initial_breadcrumb || !tl->hwsp_cacheline)
 		goto out_free;
 
-	err = intel_timeline_pin(tl);
+	err = intel_timeline_pin(tl, NULL);
 	if (err)
 		goto out_free;
 
@@ -733,7 +733,8 @@ static int live_hwsp_wrap(void *arg)
 			goto out;
 		}
 
-		if (*hwsp_seqno[0] != seqno[0] || *hwsp_seqno[1] != seqno[1]) {
+		if (READ_ONCE(*hwsp_seqno[0]) != seqno[0] ||
+		    READ_ONCE(*hwsp_seqno[1]) != seqno[1]) {
 			pr_err("Bad timeline values: found (%x, %x), expected (%x, %x)\n",
 			       *hwsp_seqno[0], *hwsp_seqno[1],
 			       seqno[0], seqno[1]);
@@ -966,9 +967,10 @@ static int live_hwsp_recycle(void *arg)
 				break;
 			}
 
-			if (*tl->hwsp_seqno != count) {
-				pr_err("Invalid seqno stored in timeline %lu @ tl->hwsp_offset, found 0x%x\n",
-				       count, *tl->hwsp_seqno);
+			if (READ_ONCE(*tl->hwsp_seqno) != count) {
+				GEM_TRACE_ERR("Invalid seqno:%lu stored in timeline %llu @ %x found 0x%x\n",
+					      count, tl->fence_context,
+					      tl->hwsp_offset, *tl->hwsp_seqno);
 				GEM_TRACE_DUMP();
 				err = -EINVAL;
 			}
diff --git a/drivers/gpu/drm/i915/gt/selftest_workarounds.c b/drivers/gpu/drm/i915/gt/selftest_workarounds.c
index febc9e6692ba..61a0532d0f3d 100644
--- a/drivers/gpu/drm/i915/gt/selftest_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/selftest_workarounds.c
@@ -214,7 +214,7 @@ static int check_whitelist(struct i915_gem_context *ctx,
 		return PTR_ERR(results);
 
 	err = 0;
-	i915_gem_object_lock(results);
+	i915_gem_object_lock(results, NULL);
 	intel_wedge_on_timeout(&wedge, engine->gt, HZ / 5) /* safety net! */
 		err = i915_gem_object_set_to_cpu_domain(results, false);
 	i915_gem_object_unlock(results);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
index 861657897c0f..942c7c187adb 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
@@ -677,7 +677,7 @@ struct i915_vma *intel_guc_allocate_vma(struct intel_guc *guc, u32 size)
 		goto err;
 
 	flags = PIN_OFFSET_BIAS | i915_ggtt_pin_bias(vma);
-	ret = i915_ggtt_pin(vma, 0, flags);
+	ret = i915_ggtt_pin(vma, NULL, 0, flags);
 	if (ret) {
 		vma = ERR_PTR(ret);
 		goto err;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 59b27aba15c6..80e8b6c3bc8c 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -51,8 +51,8 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw,
  * Note that RKL uses the same firmware as TGL.
  */
 #define INTEL_UC_FIRMWARE_DEFS(fw_def, guc_def, huc_def) \
-	fw_def(ROCKETLAKE,  0, guc_def(tgl, 35, 2, 0), huc_def(tgl,  7, 0, 12)) \
-	fw_def(TIGERLAKE,   0, guc_def(tgl, 35, 2, 0), huc_def(tgl,  7, 0, 12)) \
+	fw_def(ROCKETLAKE,  0, guc_def(tgl, 35, 2, 0), huc_def(tgl,  7, 5, 0)) \
+	fw_def(TIGERLAKE,   0, guc_def(tgl, 35, 2, 0), huc_def(tgl,  7, 5, 0)) \
 	fw_def(ELKHARTLAKE, 0, guc_def(ehl, 33, 0, 4), huc_def(ehl,  9, 0, 0)) \
 	fw_def(ICELAKE,     0, guc_def(icl, 33, 0, 0), huc_def(icl,  9, 0, 0)) \
 	fw_def(COMETLAKE,   5, guc_def(cml, 33, 0, 0), huc_def(cml,  4, 0, 0)) \
diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index f1940939260a..16b582cb97ed 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -936,7 +936,7 @@ static int cmd_reg_handler(struct parser_exec_state *s,
 		return -EFAULT;
 	}
 
-	if (!intel_gvt_mmio_is_cmd_access(gvt, offset)) {
+	if (!intel_gvt_mmio_is_cmd_accessible(gvt, offset)) {
 		gvt_vgpu_err("%s access to non-render register (%x)\n",
 				cmd, offset);
 		return -EBADRQC;
@@ -976,7 +976,7 @@ static int cmd_reg_handler(struct parser_exec_state *s,
 	 * inhibit context will restore with correct values
 	 */
 	if (IS_GEN(s->engine->i915, 9) &&
-	    intel_gvt_mmio_is_in_ctx(gvt, offset) &&
+	    intel_gvt_mmio_is_sr_in_ctx(gvt, offset) &&
 	    !strncmp(cmd, "lri", 3)) {
 		intel_gvt_hypervisor_read_gpa(s->vgpu,
 			s->workload->ring_context_gpa + 12, &ctx_sr_ctl, 4);
@@ -992,8 +992,6 @@ static int cmd_reg_handler(struct parser_exec_state *s,
 		}
 	}
 
-	/* TODO: Update the global mask if this MMIO is a masked-MMIO */
-	intel_gvt_mmio_set_cmd_accessed(gvt, offset);
 	return 0;
 }
 
@@ -1923,6 +1921,7 @@ static int perform_bb_shadow(struct parser_exec_state *s)
 	if (ret)
 		goto err_unmap;
 
+	i915_gem_object_unlock(bb->obj);
 	INIT_LIST_HEAD(&bb->list);
 	list_add(&bb->list, &s->workload->shadow_bb);
 
@@ -2982,7 +2981,7 @@ static int shadow_indirect_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 		goto put_obj;
 	}
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	ret = i915_gem_object_set_to_cpu_domain(obj, false);
 	i915_gem_object_unlock(obj);
 	if (ret) {
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index ff7f2515a6fe..9831361f181e 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -256,11 +256,11 @@ struct intel_gvt_mmio {
 /* This reg has been accessed by a VM */
 #define F_ACCESSED	(1 << 4)
 /* This reg has been accessed through GPU commands */
-#define F_CMD_ACCESSED	(1 << 5)
-/* This reg could be accessed by unaligned address */
 #define F_UNALIGN	(1 << 6)
-/* This reg is saved/restored in context */
-#define F_IN_CTX	(1 << 7)
+/* This reg is in GVT's mmio save-restor list and in hardware
+ * logical context image
+ */
+#define F_SR_IN_CTX	(1 << 7)
 
 	struct gvt_mmio_block *mmio_block;
 	unsigned int num_mmio_block;
@@ -597,39 +597,42 @@ static inline void intel_gvt_mmio_set_accessed(
 }
 
 /**
- * intel_gvt_mmio_is_cmd_accessed - mark a MMIO could be accessed by command
+ * intel_gvt_mmio_is_cmd_accessible - if a MMIO could be accessed by command
  * @gvt: a GVT device
  * @offset: register offset
  *
+ * Returns:
+ * True if an MMIO is able to be accessed by GPU commands
  */
-static inline bool intel_gvt_mmio_is_cmd_access(
+static inline bool intel_gvt_mmio_is_cmd_accessible(
 			struct intel_gvt *gvt, unsigned int offset)
 {
 	return gvt->mmio.mmio_attribute[offset >> 2] & F_CMD_ACCESS;
 }
 
 /**
- * intel_gvt_mmio_is_unalign - mark a MMIO could be accessed unaligned
+ * intel_gvt_mmio_set_cmd_accessible -
+ *				mark a MMIO could be accessible by command
  * @gvt: a GVT device
  * @offset: register offset
  *
  */
-static inline bool intel_gvt_mmio_is_unalign(
+static inline void intel_gvt_mmio_set_cmd_accessible(
 			struct intel_gvt *gvt, unsigned int offset)
 {
-	return gvt->mmio.mmio_attribute[offset >> 2] & F_UNALIGN;
+	gvt->mmio.mmio_attribute[offset >> 2] |= F_CMD_ACCESS;
 }
 
 /**
- * intel_gvt_mmio_set_cmd_accessed - mark a MMIO has been accessed by command
+ * intel_gvt_mmio_is_unalign - mark a MMIO could be accessed unaligned
  * @gvt: a GVT device
  * @offset: register offset
  *
  */
-static inline void intel_gvt_mmio_set_cmd_accessed(
+static inline bool intel_gvt_mmio_is_unalign(
 			struct intel_gvt *gvt, unsigned int offset)
 {
-	gvt->mmio.mmio_attribute[offset >> 2] |= F_CMD_ACCESSED;
+	return gvt->mmio.mmio_attribute[offset >> 2] & F_UNALIGN;
 }
 
 /**
@@ -648,30 +651,33 @@ static inline bool intel_gvt_mmio_has_mode_mask(
 }
 
 /**
- * intel_gvt_mmio_is_in_ctx - check if a MMIO has in-ctx mask
+ * intel_gvt_mmio_is_sr_in_ctx -
+ *		check if an MMIO has F_SR_IN_CTX mask
  * @gvt: a GVT device
  * @offset: register offset
  *
  * Returns:
- * True if a MMIO has a in-context mask, false if it isn't.
+ * True if an MMIO has an F_SR_IN_CTX  mask, false if it isn't.
  *
  */
-static inline bool intel_gvt_mmio_is_in_ctx(
+static inline bool intel_gvt_mmio_is_sr_in_ctx(
 			struct intel_gvt *gvt, unsigned int offset)
 {
-	return gvt->mmio.mmio_attribute[offset >> 2] & F_IN_CTX;
+	return gvt->mmio.mmio_attribute[offset >> 2] & F_SR_IN_CTX;
 }
 
 /**
- * intel_gvt_mmio_set_in_ctx - mask a MMIO in logical context
+ * intel_gvt_mmio_set_sr_in_ctx -
+ *		mask an MMIO in GVT's mmio save-restore list and also
+ *		in hardware logical context image
  * @gvt: a GVT device
  * @offset: register offset
  *
  */
-static inline void intel_gvt_mmio_set_in_ctx(
+static inline void intel_gvt_mmio_set_sr_in_ctx(
 			struct intel_gvt *gvt, unsigned int offset)
 {
-	gvt->mmio.mmio_attribute[offset >> 2] |= F_IN_CTX;
+	gvt->mmio.mmio_attribute[offset >> 2] |= F_SR_IN_CTX;
 }
 
 void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu);
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 05f3bc98d242..3be37e6fe33d 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1892,7 +1892,7 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	struct drm_i915_private *dev_priv = gvt->gt->i915;
 	int ret;
 
-	MMIO_RING_DFH(RING_IMR, D_ALL, F_CMD_ACCESS, NULL,
+	MMIO_RING_DFH(RING_IMR, D_ALL, 0, NULL,
 		intel_vgpu_reg_imr_handler);
 
 	MMIO_DFH(SDEIMR, D_ALL, 0, NULL, intel_vgpu_reg_imr_handler);
@@ -1900,7 +1900,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(SDEIIR, D_ALL, 0, NULL, intel_vgpu_reg_iir_handler);
 	MMIO_D(SDEISR, D_ALL);
 
-	MMIO_RING_DFH(RING_HWSTAM, D_ALL, F_CMD_ACCESS, NULL, NULL);
+	MMIO_RING_DFH(RING_HWSTAM, D_ALL, 0, NULL, NULL);
+
 
 	MMIO_DH(GEN8_GAMW_ECO_DEV_RW_IA, D_BDW_PLUS, NULL,
 		gamw_echo_dev_rw_ia_write);
@@ -1927,11 +1928,11 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_GM_RDR(_MMIO(0x12198), D_ALL, NULL, NULL);
 	MMIO_D(GEN7_CXT_SIZE, D_ALL);
 
-	MMIO_RING_DFH(RING_TAIL, D_ALL, F_CMD_ACCESS, NULL, NULL);
-	MMIO_RING_DFH(RING_HEAD, D_ALL, F_CMD_ACCESS, NULL, NULL);
-	MMIO_RING_DFH(RING_CTL, D_ALL, F_CMD_ACCESS, NULL, NULL);
-	MMIO_RING_DFH(RING_ACTHD, D_ALL, F_CMD_ACCESS, mmio_read_from_hw, NULL);
-	MMIO_RING_GM_RDR(RING_START, D_ALL, NULL, NULL);
+	MMIO_RING_DFH(RING_TAIL, D_ALL, 0, NULL, NULL);
+	MMIO_RING_DFH(RING_HEAD, D_ALL, 0, NULL, NULL);
+	MMIO_RING_DFH(RING_CTL, D_ALL, 0, NULL, NULL);
+	MMIO_RING_DFH(RING_ACTHD, D_ALL, 0, mmio_read_from_hw, NULL);
+	MMIO_RING_GM(RING_START, D_ALL, NULL, NULL);
 
 	/* RING MODE */
 #define RING_REG(base) _MMIO((base) + 0x29c)
@@ -2686,7 +2687,7 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(_MMIO(0x4094), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 
 	MMIO_DFH(ARB_MODE, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
-	MMIO_RING_GM_RDR(RING_BBADDR, D_ALL, NULL, NULL);
+	MMIO_RING_GM(RING_BBADDR, D_ALL, NULL, NULL);
 	MMIO_DFH(_MMIO(0x2220), D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0x12220), D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0x22220), D_ALL, F_CMD_ACCESS, NULL, NULL);
@@ -2771,7 +2772,7 @@ static int init_bdw_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(GEN8_MASTER_IRQ, D_BDW_PLUS, NULL,
 		intel_vgpu_reg_master_irq_handler);
 
-	MMIO_RING_DFH(RING_ACTHD_UDW, D_BDW_PLUS, F_CMD_ACCESS,
+	MMIO_RING_DFH(RING_ACTHD_UDW, D_BDW_PLUS, 0,
 		mmio_read_from_hw, NULL);
 
 #define RING_REG(base) _MMIO((base) + 0xd0)
@@ -2785,7 +2786,7 @@ static int init_bdw_mmio_info(struct intel_gvt *gvt)
 #undef RING_REG
 
 #define RING_REG(base) _MMIO((base) + 0x234)
-	MMIO_RING_F(RING_REG, 8, F_RO | F_CMD_ACCESS, 0, ~0, D_BDW_PLUS,
+	MMIO_RING_F(RING_REG, 8, F_RO, 0, ~0, D_BDW_PLUS,
 		NULL, NULL);
 #undef RING_REG
 
@@ -2820,7 +2821,7 @@ static int init_bdw_mmio_info(struct intel_gvt *gvt)
 	MMIO_RING_F(RING_REG, 32, F_CMD_ACCESS, 0, 0, D_BDW_PLUS, NULL, NULL);
 #undef RING_REG
 
-	MMIO_RING_GM_RDR(RING_HWS_PGA, D_BDW_PLUS, NULL, hws_pga_write);
+	MMIO_RING_GM(RING_HWS_PGA, D_BDW_PLUS, NULL, hws_pga_write);
 
 	MMIO_DFH(HDC_CHICKEN0, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 
@@ -2921,7 +2922,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(GEN9_MEDIA_PG_IDLE_HYSTERESIS, D_SKL_PLUS);
 	MMIO_D(GEN9_RENDER_PG_IDLE_HYSTERESIS, D_SKL_PLUS);
 	MMIO_DFH(GEN9_GAMT_ECO_REG_RW_IA, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DH(MMCD_MISC_CTRL, D_SKL_PLUS, NULL, NULL);
+	MMIO_DFH(MMCD_MISC_CTRL, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DH(CHICKEN_PAR1_1, D_SKL_PLUS, NULL, NULL);
 	MMIO_D(DC_STATE_EN, D_SKL_PLUS);
 	MMIO_D(DC_STATE_DEBUG, D_SKL_PLUS);
@@ -3137,7 +3138,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(GEN9_WM_CHICKEN3, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
 		 NULL, NULL);
 
-	MMIO_D(GAMT_CHKN_BIT_REG, D_KBL | D_CFL);
+	MMIO_DFH(GAMT_CHKN_BIT_REG, D_KBL | D_CFL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_D(GEN9_CTX_PREEMPT_REG, D_SKL_PLUS);
 
 	return 0;
@@ -3357,7 +3358,10 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt)
 	gvt->mmio.mmio_attribute = NULL;
 }
 
-/* Special MMIO blocks. */
+/* Special MMIO blocks. registers in MMIO block ranges should not be command
+ * accessible (should have no F_CMD_ACCESS flag).
+ * otherwise, need to update cmd_reg_handler in cmd_parser.c
+ */
 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},
diff --git a/drivers/gpu/drm/i915/gvt/mmio.c b/drivers/gpu/drm/i915/gvt/mmio.c
index 291993615af9..b6811f6a230d 100644
--- a/drivers/gpu/drm/i915/gvt/mmio.c
+++ b/drivers/gpu/drm/i915/gvt/mmio.c
@@ -251,6 +251,9 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr)
 		/* set the bit 0:2(Core C-State ) to C0 */
 		vgpu_vreg_t(vgpu, GEN6_GT_CORE_STATUS) = 0;
 
+		/* uc reset hw expect GS_MIA_IN_RESET */
+		vgpu_vreg_t(vgpu, GUC_STATUS) |= GS_MIA_IN_RESET;
+
 		if (IS_BROXTON(vgpu->gvt->gt->i915)) {
 			vgpu_vreg_t(vgpu, BXT_P_CR_GT_DISP_PWRON) &=
 				    ~(BIT(0) | BIT(1));
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 86a60bdf0818..afe574d6b3b5 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -595,7 +595,7 @@ void intel_gvt_init_engine_mmio_context(struct intel_gvt *gvt)
 	     i915_mmio_reg_valid(mmio->reg); mmio++) {
 		if (mmio->in_context) {
 			gvt->engine_mmio_list.ctx_mmio_count[mmio->id]++;
-			intel_gvt_mmio_set_in_ctx(gvt, mmio->reg.reg);
+			intel_gvt_mmio_set_sr_in_ctx(gvt, mmio->reg.reg);
 		}
 	}
 }
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 3c3b9842bbbd..1570eb8aa978 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -403,6 +403,14 @@ static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 	wa_ctx->indirect_ctx.shadow_va = NULL;
 }
 
+static void set_dma_address(struct i915_page_directory *pd, dma_addr_t addr)
+{
+	struct scatterlist *sg = pd->pt.base->mm.pages->sgl;
+
+	/* This is not a good idea */
+	sg->dma_address = addr;
+}
+
 static void set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 					  struct intel_context *ce)
 {
@@ -411,7 +419,7 @@ static void set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 	int i = 0;
 
 	if (mm->ppgtt_mm.root_entry_type == GTT_TYPE_PPGTT_ROOT_L4_ENTRY) {
-		px_dma(ppgtt->pd) = mm->ppgtt_mm.shadow_pdps[0];
+		set_dma_address(ppgtt->pd, mm->ppgtt_mm.shadow_pdps[0]);
 	} else {
 		for (i = 0; i < GVT_RING_CTX_NR_PDPS; i++) {
 			struct i915_page_directory * const pd =
@@ -421,7 +429,8 @@ static void set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 			   shadow ppgtt. */
 			if (!pd)
 				break;
-			px_dma(pd) = mm->ppgtt_mm.shadow_pdps[i];
+
+			set_dma_address(pd, mm->ppgtt_mm.shadow_pdps[i]);
 		}
 	}
 }
@@ -1240,13 +1249,13 @@ i915_context_ppgtt_root_restore(struct intel_vgpu_submission *s,
 	int i;
 
 	if (i915_vm_is_4lvl(&ppgtt->vm)) {
-		px_dma(ppgtt->pd) = s->i915_context_pml4;
+		set_dma_address(ppgtt->pd, s->i915_context_pml4);
 	} else {
 		for (i = 0; i < GEN8_3LVL_PDPES; i++) {
 			struct i915_page_directory * const pd =
 				i915_pd_entry(ppgtt->pd, i);
 
-			px_dma(pd) = s->i915_context_pdps[i];
+			set_dma_address(pd, s->i915_context_pdps[i]);
 		}
 	}
 }
diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index 839bd53df6e9..10a865f3dc09 100644
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -28,12 +28,14 @@ static struct i915_global_active {
 } global;
 
 struct active_node {
+	struct rb_node node;
 	struct i915_active_fence base;
 	struct i915_active *ref;
-	struct rb_node node;
 	u64 timeline;
 };
 
+#define fetch_node(x) rb_entry(READ_ONCE(x), typeof(struct active_node), node)
+
 static inline struct active_node *
 node_from_active(struct i915_active_fence *active)
 {
@@ -128,8 +130,8 @@ static inline void debug_active_assert(struct i915_active *ref) { }
 static void
 __active_retire(struct i915_active *ref)
 {
+	struct rb_root root = RB_ROOT;
 	struct active_node *it, *n;
-	struct rb_root root;
 	unsigned long flags;
 
 	GEM_BUG_ON(i915_active_is_idle(ref));
@@ -141,9 +143,25 @@ __active_retire(struct i915_active *ref)
 	GEM_BUG_ON(rcu_access_pointer(ref->excl.fence));
 	debug_active_deactivate(ref);
 
-	root = ref->tree;
-	ref->tree = RB_ROOT;
-	ref->cache = NULL;
+	/* Even if we have not used the cache, we may still have a barrier */
+	if (!ref->cache)
+		ref->cache = fetch_node(ref->tree.rb_node);
+
+	/* Keep the MRU cached node for reuse */
+	if (ref->cache) {
+		/* Discard all other nodes in the tree */
+		rb_erase(&ref->cache->node, &ref->tree);
+		root = ref->tree;
+
+		/* Rebuild the tree with only the cached node */
+		rb_link_node(&ref->cache->node, NULL, &ref->tree.rb_node);
+		rb_insert_color(&ref->cache->node, &ref->tree);
+		GEM_BUG_ON(ref->tree.rb_node != &ref->cache->node);
+
+		/* Make the cached node available for reuse with any timeline */
+		if (IS_ENABLED(CONFIG_64BIT))
+			ref->cache->timeline = 0; /* needs cmpxchg(u64) */
+	}
 
 	spin_unlock_irqrestore(&ref->tree_lock, flags);
 
@@ -154,6 +172,7 @@ __active_retire(struct i915_active *ref)
 	/* ... except if you wait on it, you must manage your own references! */
 	wake_up_var(ref);
 
+	/* Finally free the discarded timeline tree  */
 	rbtree_postorder_for_each_entry_safe(it, n, &root, node) {
 		GEM_BUG_ON(i915_active_fence_isset(&it->base));
 		kmem_cache_free(global.slab_cache, it);
@@ -216,12 +235,11 @@ excl_retire(struct dma_fence *fence, struct dma_fence_cb *cb)
 		active_retire(container_of(cb, struct i915_active, excl.cb));
 }
 
-static struct i915_active_fence *
-active_instance(struct i915_active *ref, struct intel_timeline *tl)
+static struct active_node *__active_lookup(struct i915_active *ref, u64 idx)
 {
-	struct active_node *node, *prealloc;
-	struct rb_node **p, *parent;
-	u64 idx = tl->fence_context;
+	struct active_node *it;
+
+	GEM_BUG_ON(idx == 0); /* 0 is the unordered timeline, rsvd for cache */
 
 	/*
 	 * We track the most recently used timeline to skip a rbtree search
@@ -230,8 +248,59 @@ active_instance(struct i915_active *ref, struct intel_timeline *tl)
 	 * after the previous activity has been retired, or if it matches the
 	 * current timeline.
 	 */
-	node = READ_ONCE(ref->cache);
-	if (node && node->timeline == idx)
+	it = READ_ONCE(ref->cache);
+	if (it) {
+		u64 cached = READ_ONCE(it->timeline);
+
+		/* Once claimed, this slot will only belong to this idx */
+		if (cached == idx)
+			return it;
+
+#ifdef CONFIG_64BIT /* for cmpxchg(u64) */
+		/*
+		 * An unclaimed cache [.timeline=0] can only be claimed once.
+		 *
+		 * If the value is already non-zero, some other thread has
+		 * claimed the cache and we know that is does not match our
+		 * idx. If, and only if, the timeline is currently zero is it
+		 * worth competing to claim it atomically for ourselves (for
+		 * only the winner of that race will cmpxchg return the old
+		 * value of 0).
+		 */
+		if (!cached && !cmpxchg(&it->timeline, 0, idx))
+			return it;
+#endif
+	}
+
+	BUILD_BUG_ON(offsetof(typeof(*it), node));
+
+	/* While active, the tree can only be built; not destroyed */
+	GEM_BUG_ON(i915_active_is_idle(ref));
+
+	it = fetch_node(ref->tree.rb_node);
+	while (it) {
+		if (it->timeline < idx) {
+			it = fetch_node(it->node.rb_right);
+		} else if (it->timeline > idx) {
+			it = fetch_node(it->node.rb_left);
+		} else {
+			WRITE_ONCE(ref->cache, it);
+			break;
+		}
+	}
+
+	/* NB: If the tree rotated beneath us, we may miss our target. */
+	return it;
+}
+
+static struct i915_active_fence *
+active_instance(struct i915_active *ref, u64 idx)
+{
+	struct active_node *node, *prealloc;
+	struct rb_node **p, *parent;
+
+	node = __active_lookup(ref, idx);
+	if (likely(node))
 		return &node->base;
 
 	/* Preallocate a replacement, just in case */
@@ -268,10 +337,9 @@ active_instance(struct i915_active *ref, struct intel_timeline *tl)
 	rb_insert_color(&node->node, &ref->tree);
 
 out:
-	ref->cache = node;
+	WRITE_ONCE(ref->cache, node);
 	spin_unlock_irq(&ref->tree_lock);
 
-	BUILD_BUG_ON(offsetof(typeof(*node), base));
 	return &node->base;
 }
 
@@ -353,69 +421,116 @@ __active_del_barrier(struct i915_active *ref, struct active_node *node)
 	return ____active_del_barrier(ref, node, barrier_to_engine(node));
 }
 
-int i915_active_ref(struct i915_active *ref,
-		    struct intel_timeline *tl,
-		    struct dma_fence *fence)
+static bool
+replace_barrier(struct i915_active *ref, struct i915_active_fence *active)
+{
+	if (!is_barrier(active)) /* proto-node used by our idle barrier? */
+		return false;
+
+	/*
+	 * This request is on the kernel_context timeline, and so
+	 * we can use it to substitute for the pending idle-barrer
+	 * request that we want to emit on the kernel_context.
+	 */
+	__active_del_barrier(ref, node_from_active(active));
+	return true;
+}
+
+int i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence)
 {
 	struct i915_active_fence *active;
 	int err;
 
-	lockdep_assert_held(&tl->mutex);
-
 	/* Prevent reaping in case we malloc/wait while building the tree */
 	err = i915_active_acquire(ref);
 	if (err)
 		return err;
 
-	active = active_instance(ref, tl);
+	active = active_instance(ref, idx);
 	if (!active) {
 		err = -ENOMEM;
 		goto out;
 	}
 
-	if (is_barrier(active)) { /* proto-node used by our idle barrier */
-		/*
-		 * This request is on the kernel_context timeline, and so
-		 * we can use it to substitute for the pending idle-barrer
-		 * request that we want to emit on the kernel_context.
-		 */
-		__active_del_barrier(ref, node_from_active(active));
+	if (replace_barrier(ref, active)) {
 		RCU_INIT_POINTER(active->fence, NULL);
 		atomic_dec(&ref->count);
 	}
 	if (!__i915_active_fence_set(active, fence))
-		atomic_inc(&ref->count);
+		__i915_active_acquire(ref);
 
 out:
 	i915_active_release(ref);
 	return err;
 }
 
-struct dma_fence *
-i915_active_set_exclusive(struct i915_active *ref, struct dma_fence *f)
+static struct dma_fence *
+__i915_active_set_fence(struct i915_active *ref,
+			struct i915_active_fence *active,
+			struct dma_fence *fence)
 {
 	struct dma_fence *prev;
 
-	/* We expect the caller to manage the exclusive timeline ordering */
-	GEM_BUG_ON(i915_active_is_idle(ref));
+	if (replace_barrier(ref, active)) {
+		RCU_INIT_POINTER(active->fence, fence);
+		return NULL;
+	}
 
 	rcu_read_lock();
-	prev = __i915_active_fence_set(&ref->excl, f);
+	prev = __i915_active_fence_set(active, fence);
 	if (prev)
 		prev = dma_fence_get_rcu(prev);
 	else
-		atomic_inc(&ref->count);
+		__i915_active_acquire(ref);
 	rcu_read_unlock();
 
 	return prev;
 }
 
+static struct i915_active_fence *
+__active_fence(struct i915_active *ref, u64 idx)
+{
+	struct active_node *it;
+
+	it = __active_lookup(ref, idx);
+	if (unlikely(!it)) { /* Contention with parallel tree builders! */
+		spin_lock_irq(&ref->tree_lock);
+		it = __active_lookup(ref, idx);
+		spin_unlock_irq(&ref->tree_lock);
+	}
+	GEM_BUG_ON(!it); /* slot must be preallocated */
+
+	return &it->base;
+}
+
+struct dma_fence *
+__i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence)
+{
+	/* Only valid while active, see i915_active_acquire_for_context() */
+	return __i915_active_set_fence(ref, __active_fence(ref, idx), fence);
+}
+
+struct dma_fence *
+i915_active_set_exclusive(struct i915_active *ref, struct dma_fence *f)
+{
+	/* We expect the caller to manage the exclusive timeline ordering */
+	return __i915_active_set_fence(ref, &ref->excl, f);
+}
+
 bool i915_active_acquire_if_busy(struct i915_active *ref)
 {
 	debug_active_assert(ref);
 	return atomic_add_unless(&ref->count, 1, 0);
 }
 
+static void __i915_active_activate(struct i915_active *ref)
+{
+	spin_lock_irq(&ref->tree_lock); /* __active_retire() */
+	if (!atomic_fetch_inc(&ref->count))
+		debug_active_activate(ref);
+	spin_unlock_irq(&ref->tree_lock);
+}
+
 int i915_active_acquire(struct i915_active *ref)
 {
 	int err;
@@ -423,19 +538,19 @@ int i915_active_acquire(struct i915_active *ref)
 	if (i915_active_acquire_if_busy(ref))
 		return 0;
 
+	if (!ref->active) {
+		__i915_active_activate(ref);
+		return 0;
+	}
+
 	err = mutex_lock_interruptible(&ref->mutex);
 	if (err)
 		return err;
 
 	if (likely(!i915_active_acquire_if_busy(ref))) {
-		if (ref->active)
-			err = ref->active(ref);
-		if (!err) {
-			spin_lock_irq(&ref->tree_lock); /* __active_retire() */
-			debug_active_activate(ref);
-			atomic_inc(&ref->count);
-			spin_unlock_irq(&ref->tree_lock);
-		}
+		err = ref->active(ref);
+		if (!err)
+			__i915_active_activate(ref);
 	}
 
 	mutex_unlock(&ref->mutex);
@@ -443,6 +558,24 @@ int i915_active_acquire(struct i915_active *ref)
 	return err;
 }
 
+int i915_active_acquire_for_context(struct i915_active *ref, u64 idx)
+{
+	struct i915_active_fence *active;
+	int err;
+
+	err = i915_active_acquire(ref);
+	if (err)
+		return err;
+
+	active = active_instance(ref, idx);
+	if (!active) {
+		i915_active_release(ref);
+		return -ENOMEM;
+	}
+
+	return 0; /* return with active ref */
+}
+
 void i915_active_release(struct i915_active *ref)
 {
 	debug_active_assert(ref);
@@ -651,16 +784,16 @@ int i915_sw_fence_await_active(struct i915_sw_fence *fence,
 	return await_active(ref, flags, sw_await_fence, fence, fence);
 }
 
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
 void i915_active_fini(struct i915_active *ref)
 {
 	debug_active_fini(ref);
 	GEM_BUG_ON(atomic_read(&ref->count));
 	GEM_BUG_ON(work_pending(&ref->work));
-	GEM_BUG_ON(!RB_EMPTY_ROOT(&ref->tree));
 	mutex_destroy(&ref->mutex);
+
+	if (ref->cache)
+		kmem_cache_free(global.slab_cache, ref->cache);
 }
-#endif
 
 static inline bool is_idle_barrier(struct active_node *node, u64 idx)
 {
@@ -674,7 +807,6 @@ static struct active_node *reuse_idle_barrier(struct i915_active *ref, u64 idx)
 	if (RB_EMPTY_ROOT(&ref->tree))
 		return NULL;
 
-	spin_lock_irq(&ref->tree_lock);
 	GEM_BUG_ON(i915_active_is_idle(ref));
 
 	/*
@@ -700,9 +832,9 @@ static struct active_node *reuse_idle_barrier(struct i915_active *ref, u64 idx)
 
 		prev = p;
 		if (node->timeline < idx)
-			p = p->rb_right;
+			p = READ_ONCE(p->rb_right);
 		else
-			p = p->rb_left;
+			p = READ_ONCE(p->rb_left);
 	}
 
 	/*
@@ -739,14 +871,13 @@ static struct active_node *reuse_idle_barrier(struct i915_active *ref, u64 idx)
 			goto match;
 	}
 
-	spin_unlock_irq(&ref->tree_lock);
-
 	return NULL;
 
 match:
+	spin_lock_irq(&ref->tree_lock);
 	rb_erase(p, &ref->tree); /* Hide from waits and sibling allocations */
 	if (p == &ref->cache->node)
-		ref->cache = NULL;
+		WRITE_ONCE(ref->cache, NULL);
 	spin_unlock_irq(&ref->tree_lock);
 
 	return rb_entry(p, struct active_node, node);
@@ -758,7 +889,6 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
 	intel_engine_mask_t tmp, mask = engine->mask;
 	struct llist_node *first = NULL, *last = NULL;
 	struct intel_gt *gt = engine->gt;
-	int err;
 
 	GEM_BUG_ON(i915_active_is_idle(ref));
 
@@ -778,13 +908,13 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
 		struct llist_node *prev = first;
 		struct active_node *node;
 
+		rcu_read_lock();
 		node = reuse_idle_barrier(ref, idx);
+		rcu_read_unlock();
 		if (!node) {
 			node = kmem_cache_alloc(global.slab_cache, GFP_KERNEL);
-			if (!node) {
-				err = ENOMEM;
+			if (!node)
 				goto unwind;
-			}
 
 			RCU_INIT_POINTER(node->base.fence, NULL);
 			node->base.cb.func = node_retire;
@@ -804,7 +934,7 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
 			 */
 			RCU_INIT_POINTER(node->base.fence, ERR_PTR(-EAGAIN));
 			node->base.cb.node.prev = (void *)engine;
-			atomic_inc(&ref->count);
+			__i915_active_acquire(ref);
 		}
 		GEM_BUG_ON(rcu_access_pointer(node->base.fence) != ERR_PTR(-EAGAIN));
 
@@ -832,7 +962,7 @@ unwind:
 
 		kmem_cache_free(global.slab_cache, node);
 	}
-	return err;
+	return -ENOMEM;
 }
 
 void i915_active_acquire_barrier(struct i915_active *ref)
diff --git a/drivers/gpu/drm/i915/i915_active.h b/drivers/gpu/drm/i915/i915_active.h
index cf4058150966..fb165d3f01cf 100644
--- a/drivers/gpu/drm/i915/i915_active.h
+++ b/drivers/gpu/drm/i915/i915_active.h
@@ -163,14 +163,16 @@ void __i915_active_init(struct i915_active *ref,
 	__i915_active_init(ref, active, retire, &__mkey, &__wkey);	\
 } while (0)
 
-int i915_active_ref(struct i915_active *ref,
-		    struct intel_timeline *tl,
-		    struct dma_fence *fence);
+struct dma_fence *
+__i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence);
+int i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence);
 
 static inline int
 i915_active_add_request(struct i915_active *ref, struct i915_request *rq)
 {
-	return i915_active_ref(ref, i915_request_timeline(rq), &rq->fence);
+	return i915_active_ref(ref,
+			       i915_request_timeline(rq)->fence_context,
+			       &rq->fence);
 }
 
 struct dma_fence *
@@ -198,7 +200,9 @@ int i915_request_await_active(struct i915_request *rq,
 #define I915_ACTIVE_AWAIT_BARRIER BIT(2)
 
 int i915_active_acquire(struct i915_active *ref);
+int i915_active_acquire_for_context(struct i915_active *ref, u64 idx);
 bool i915_active_acquire_if_busy(struct i915_active *ref);
+
 void i915_active_release(struct i915_active *ref);
 
 static inline void __i915_active_acquire(struct i915_active *ref)
@@ -213,11 +217,7 @@ i915_active_is_idle(const struct i915_active *ref)
 	return !atomic_read(&ref->count);
 }
 
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
 void i915_active_fini(struct i915_active *ref);
-#else
-static inline void i915_active_fini(struct i915_active *ref) { }
-#endif
 
 int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
 					    struct intel_engine_cs *engine);
@@ -231,4 +231,19 @@ struct i915_active *i915_active_create(void);
 struct i915_active *i915_active_get(struct i915_active *ref);
 void i915_active_put(struct i915_active *ref);
 
+static inline int __i915_request_await_exclusive(struct i915_request *rq,
+						 struct i915_active *active)
+{
+	struct dma_fence *fence;
+	int err = 0;
+
+	fence = i915_active_fence_get(&active->excl);
+	if (fence) {
+		err = i915_request_await_dma_fence(rq, fence);
+		dma_fence_put(fence);
+	}
+
+	return err;
+}
+
 #endif /* _I915_ACTIVE_H_ */
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 5ac4a999f05a..e88970256e8e 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1136,7 +1136,7 @@ find_reg(const struct intel_engine_cs *engine, u32 addr)
 /* Returns a vmap'd pointer to dst_obj, which the caller must unmap */
 static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
 		       struct drm_i915_gem_object *src_obj,
-		       u32 offset, u32 length)
+		       unsigned long offset, unsigned long length)
 {
 	bool needs_clflush;
 	void *dst, *src;
@@ -1166,8 +1166,8 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
 		}
 	}
 	if (IS_ERR(src)) {
+		unsigned long x, n;
 		void *ptr;
-		int x, n;
 
 		/*
 		 * We can avoid clflushing partial cachelines before the write
@@ -1184,7 +1184,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
 		ptr = dst;
 		x = offset_in_page(offset);
 		for (n = offset >> PAGE_SHIFT; length; n++) {
-			int len = min_t(int, length, PAGE_SIZE - x);
+			int len = min(length, PAGE_SIZE - x);
 
 			src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
 			if (needs_clflush)
@@ -1414,8 +1414,8 @@ static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
  */
 int intel_engine_cmd_parser(struct intel_engine_cs *engine,
 			    struct i915_vma *batch,
-			    u32 batch_offset,
-			    u32 batch_length,
+			    unsigned long batch_offset,
+			    unsigned long batch_length,
 			    struct i915_vma *shadow,
 			    bool trampoline)
 {
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 784219962193..ea469168cd44 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -326,6 +326,7 @@ static void print_context_stats(struct seq_file *m,
 		}
 		i915_gem_context_unlock_engines(ctx);
 
+		mutex_lock(&ctx->mutex);
 		if (!IS_ERR_OR_NULL(ctx->file_priv)) {
 			struct file_stats stats = {
 				.vm = rcu_access_pointer(ctx->vm),
@@ -346,6 +347,7 @@ static void print_context_stats(struct seq_file *m,
 
 			print_file_stats(m, name, stats);
 		}
+		mutex_unlock(&ctx->mutex);
 
 		spin_lock(&i915->gem.contexts.lock);
 		list_safe_reset_next(ctx, cn, link);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 5fd5af4bc855..acc32066cec3 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -58,7 +58,6 @@
 #include "display/intel_hotplug.h"
 #include "display/intel_overlay.h"
 #include "display/intel_pipe_crc.h"
-#include "display/intel_psr.h"
 #include "display/intel_sprite.h"
 #include "display/intel_vga.h"
 
@@ -216,125 +215,6 @@ intel_teardown_mchbar(struct drm_i915_private *dev_priv)
 		release_resource(&dev_priv->mch_res);
 }
 
-/* part #1: call before irq install */
-static int i915_driver_modeset_probe_noirq(struct drm_i915_private *i915)
-{
-	int ret;
-
-	if (i915_inject_probe_failure(i915))
-		return -ENODEV;
-
-	if (HAS_DISPLAY(i915) && INTEL_DISPLAY_ENABLED(i915)) {
-		ret = drm_vblank_init(&i915->drm,
-				      INTEL_NUM_PIPES(i915));
-		if (ret)
-			return ret;
-	}
-
-	intel_bios_init(i915);
-
-	ret = intel_vga_register(i915);
-	if (ret)
-		goto cleanup_bios;
-
-	intel_power_domains_init_hw(i915, false);
-
-	intel_csr_ucode_init(i915);
-
-	ret = intel_modeset_init_noirq(i915);
-	if (ret)
-		goto cleanup_vga_client_pw_domain_csr;
-
-	return 0;
-
-cleanup_vga_client_pw_domain_csr:
-	intel_csr_ucode_fini(i915);
-	intel_power_domains_driver_remove(i915);
-	intel_vga_unregister(i915);
-cleanup_bios:
-	intel_bios_driver_remove(i915);
-	return ret;
-}
-
-/* part #2: call after irq install */
-static int i915_driver_modeset_probe(struct drm_i915_private *i915)
-{
-	int ret;
-
-	/* Important: The output setup functions called by modeset_init need
-	 * working irqs for e.g. gmbus and dp aux transfers. */
-	ret = intel_modeset_init(i915);
-	if (ret)
-		goto out;
-
-	ret = i915_gem_init(i915);
-	if (ret)
-		goto cleanup_modeset;
-
-	intel_overlay_setup(i915);
-
-	if (!HAS_DISPLAY(i915) || !INTEL_DISPLAY_ENABLED(i915))
-		return 0;
-
-	ret = intel_fbdev_init(&i915->drm);
-	if (ret)
-		goto cleanup_gem;
-
-	/* Only enable hotplug handling once the fbdev is fully set up. */
-	intel_hpd_init(i915);
-
-	intel_init_ipc(i915);
-
-	intel_psr_set_force_mode_changed(i915->psr.dp);
-
-	return 0;
-
-cleanup_gem:
-	i915_gem_suspend(i915);
-	i915_gem_driver_remove(i915);
-	i915_gem_driver_release(i915);
-cleanup_modeset:
-	/* FIXME */
-	intel_modeset_driver_remove(i915);
-	intel_irq_uninstall(i915);
-	intel_modeset_driver_remove_noirq(i915);
-out:
-	return ret;
-}
-
-/* part #1: call before irq uninstall */
-static void i915_driver_modeset_remove(struct drm_i915_private *i915)
-{
-	intel_modeset_driver_remove(i915);
-}
-
-/* part #2: call after irq uninstall */
-static void i915_driver_modeset_remove_noirq(struct drm_i915_private *i915)
-{
-	intel_csr_ucode_fini(i915);
-
-	intel_power_domains_driver_remove(i915);
-
-	intel_vga_unregister(i915);
-
-	intel_bios_driver_remove(i915);
-}
-
-static void intel_init_dpio(struct drm_i915_private *dev_priv)
-{
-	/*
-	 * IOSF_PORT_DPIO is used for VLV x2 PHY (DP/HDMI B and C),
-	 * CHV x1 PHY (DP/HDMI D)
-	 * IOSF_PORT_DPIO_2 is used for CHV x2 PHY (DP/HDMI B and C)
-	 */
-	if (IS_CHERRYVIEW(dev_priv)) {
-		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO_2;
-		DPIO_PHY_IOSF_PORT(DPIO_PHY1) = IOSF_PORT_DPIO;
-	} else if (IS_VALLEYVIEW(dev_priv)) {
-		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
-	}
-}
-
 static int i915_workqueues_init(struct drm_i915_private *dev_priv)
 {
 	/*
@@ -392,7 +272,7 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv)
 	pre |= IS_HSW_EARLY_SDV(dev_priv);
 	pre |= IS_SKL_REVID(dev_priv, 0, SKL_REVID_F0);
 	pre |= IS_BXT_REVID(dev_priv, 0, BXT_REVID_B_LAST);
-	pre |= IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0);
+	pre |= IS_KBL_GT_REVID(dev_priv, 0, KBL_REVID_A0);
 	pre |= IS_GLK_REVID(dev_priv, 0, GLK_REVID_A2);
 
 	if (pre) {
@@ -463,7 +343,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 	intel_detect_pch(dev_priv);
 
 	intel_pm_setup(dev_priv);
-	intel_init_dpio(dev_priv);
 	ret = intel_power_domains_init(dev_priv);
 	if (ret < 0)
 		goto err_gem;
@@ -798,7 +677,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv)
 		drm_err(&dev_priv->drm,
 			"Failed to register driver for userspace access!\n");
 
-	if (HAS_DISPLAY(dev_priv) && INTEL_DISPLAY_ENABLED(dev_priv)) {
+	if (HAS_DISPLAY(dev_priv)) {
 		/* Must be done after probing outputs */
 		intel_opregion_register(dev_priv);
 		acpi_video_register();
@@ -821,7 +700,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv)
 	 * We need to coordinate the hotplugs with the asynchronous fbdev
 	 * configuration, for which we use the fbdev->async_cookie.
 	 */
-	if (HAS_DISPLAY(dev_priv) && INTEL_DISPLAY_ENABLED(dev_priv))
+	if (HAS_DISPLAY(dev_priv))
 		drm_kms_helper_poll_init(dev);
 
 	intel_power_domains_enable(dev_priv);
@@ -988,7 +867,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret < 0)
 		goto out_cleanup_mmio;
 
-	ret = i915_driver_modeset_probe_noirq(i915);
+	ret = intel_modeset_init_noirq(i915);
 	if (ret < 0)
 		goto out_cleanup_hw;
 
@@ -996,10 +875,18 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto out_cleanup_modeset;
 
-	ret = i915_driver_modeset_probe(i915);
-	if (ret < 0)
+	ret = intel_modeset_init_nogem(i915);
+	if (ret)
 		goto out_cleanup_irq;
 
+	ret = i915_gem_init(i915);
+	if (ret)
+		goto out_cleanup_modeset2;
+
+	ret = intel_modeset_init(i915);
+	if (ret)
+		goto out_cleanup_gem;
+
 	i915_driver_register(i915);
 
 	enable_rpm_wakeref_asserts(&i915->runtime_pm);
@@ -1010,10 +897,20 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	return 0;
 
+out_cleanup_gem:
+	i915_gem_suspend(i915);
+	i915_gem_driver_remove(i915);
+	i915_gem_driver_release(i915);
+out_cleanup_modeset2:
+	/* FIXME clean up the error path */
+	intel_modeset_driver_remove(i915);
+	intel_irq_uninstall(i915);
+	intel_modeset_driver_remove_noirq(i915);
+	goto out_cleanup_modeset;
 out_cleanup_irq:
 	intel_irq_uninstall(i915);
 out_cleanup_modeset:
-	i915_driver_modeset_remove_noirq(i915);
+	intel_modeset_driver_remove_nogem(i915);
 out_cleanup_hw:
 	i915_driver_hw_remove(i915);
 	intel_memory_regions_driver_release(i915);
@@ -1045,7 +942,7 @@ void i915_driver_remove(struct drm_i915_private *i915)
 
 	intel_gvt_driver_remove(i915);
 
-	i915_driver_modeset_remove(i915);
+	intel_modeset_driver_remove(i915);
 
 	intel_irq_uninstall(i915);
 
@@ -1054,7 +951,7 @@ void i915_driver_remove(struct drm_i915_private *i915)
 	i915_reset_error_state(i915);
 	i915_gem_driver_remove(i915);
 
-	i915_driver_modeset_remove_noirq(i915);
+	intel_modeset_driver_remove_nogem(i915);
 
 	i915_driver_hw_remove(i915);
 
@@ -1075,6 +972,7 @@ static void i915_driver_release(struct drm_device *dev)
 
 	intel_memory_regions_driver_release(dev_priv);
 	i915_ggtt_driver_release(dev_priv);
+	i915_gem_drain_freed_objects(dev_priv);
 
 	i915_driver_mmio_release(dev_priv);
 
@@ -1119,7 +1017,6 @@ static void i915_driver_postclose(struct drm_device *dev, struct drm_file *file)
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
 	i915_gem_context_close(file);
-	i915_gem_release(dev, file);
 
 	kfree_rcu(file_priv, rcu);
 
@@ -1846,7 +1743,8 @@ static struct drm_driver driver = {
 	 */
 	.driver_features =
 	    DRIVER_GEM |
-	    DRIVER_RENDER | DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_SYNCOBJ,
+	    DRIVER_RENDER | DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_SYNCOBJ |
+	    DRIVER_SYNCOBJ_TIMELINE,
 	.release = i915_driver_release,
 	.open = i915_driver_open,
 	.lastclose = i915_driver_lastclose,
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e4f7f6518945..eef9a821c49c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -108,18 +108,11 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20200715"
-#define DRIVER_TIMESTAMP	1594811881
+#define DRIVER_DATE		"20200917"
+#define DRIVER_TIMESTAMP	1600375437
 
 struct drm_i915_gem_object;
 
-/*
- * The code assumes that the hpd_pins below have consecutive values and
- * starting with HPD_PORT_A, the HPD pin associated with any port can be
- * retrieved by adding the corresponding port (or phy) enum value to
- * HPD_PORT_A in most cases. For example:
- * HPD_PORT_C = HPD_PORT_A + PHY_C - PHY_A
- */
 enum hpd_pin {
 	HPD_NONE = 0,
 	HPD_TV = HPD_NONE,     /* TV is known to be unreliable */
@@ -131,10 +124,12 @@ enum hpd_pin {
 	HPD_PORT_C,
 	HPD_PORT_D,
 	HPD_PORT_E,
-	HPD_PORT_F,
-	HPD_PORT_G,
-	HPD_PORT_H,
-	HPD_PORT_I,
+	HPD_PORT_TC1,
+	HPD_PORT_TC2,
+	HPD_PORT_TC3,
+	HPD_PORT_TC4,
+	HPD_PORT_TC5,
+	HPD_PORT_TC6,
 
 	HPD_NUM_PINS
 };
@@ -203,11 +198,6 @@ struct drm_i915_file_private {
 		struct rcu_head rcu;
 	};
 
-	struct {
-		spinlock_t lock;
-		struct list_head request_list;
-	} mm;
-
 	struct xarray context_xa;
 	struct xarray vm_xa;
 
@@ -506,6 +496,7 @@ struct i915_psr {
 	bool link_standby;
 	bool colorimetry_support;
 	bool psr2_enabled;
+	bool psr2_sel_fetch_enabled;
 	u8 sink_sync_latency;
 	ktime_t last_entry_attempt;
 	ktime_t last_exit;
@@ -541,13 +532,9 @@ struct intel_gmbus {
 
 struct i915_suspend_saved_registers {
 	u32 saveDSPARB;
-	u32 saveFBC_CONTROL;
-	u32 saveCACHE_MODE_0;
-	u32 saveMI_ARB_STATE;
 	u32 saveSWF0[16];
 	u32 saveSWF1[16];
 	u32 saveSWF3[3];
-	u32 savePCH_PORT_HOTPLUG;
 	u16 saveGCDGMBUS;
 };
 
@@ -592,11 +579,6 @@ struct i915_gem_mm {
 	atomic_t free_count;
 
 	/**
-	 * Small stash of WC pages
-	 */
-	struct pagestash wc_stash;
-
-	/**
 	 * tmpfs instance used for shmem backed objects
 	 */
 	struct vfsmount *gemfs;
@@ -1029,8 +1011,6 @@ struct drm_i915_private {
 	 */
 	u8 active_pipes;
 
-	int dpio_phy_iosf_port[I915_NUM_PHYS_VLV];
-
 	struct i915_wa_list gt_wa_list;
 
 	struct i915_frontbuffer_tracking fb_tracking;
@@ -1045,6 +1025,14 @@ struct drm_i915_private {
 	struct intel_l3_parity l3_parity;
 
 	/*
+	 * HTI (aka HDPORT) state read during initial hw readout.  Most
+	 * platforms don't have HTI, so this will just stay 0.  Those that do
+	 * will use this later to figure out which PLLs and PHYs are unavailable
+	 * for driver usage.
+	 */
+	u32 hti_state;
+
+	/*
 	 * edram size in MB.
 	 * Cannot be determined by PCIID. You must always read a register.
 	 */
@@ -1489,6 +1477,12 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define IS_ICL_WITH_PORT_F(dev_priv) \
 	IS_SUBPLATFORM(dev_priv, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF)
 
+#define IS_TGL_U(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_TIGERLAKE, INTEL_SUBPLATFORM_ULT)
+
+#define IS_TGL_Y(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_TIGERLAKE, INTEL_SUBPLATFORM_ULX)
+
 #define SKL_REVID_A0		0x0
 #define SKL_REVID_B0		0x1
 #define SKL_REVID_C0		0x2
@@ -1509,14 +1503,34 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define IS_BXT_REVID(dev_priv, since, until) \
 	(IS_BROXTON(dev_priv) && IS_REVID(dev_priv, since, until))
 
-#define KBL_REVID_A0		0x0
-#define KBL_REVID_B0		0x1
-#define KBL_REVID_C0		0x2
-#define KBL_REVID_D0		0x3
-#define KBL_REVID_E0		0x4
+enum {
+	KBL_REVID_A0,
+	KBL_REVID_B0,
+	KBL_REVID_B1,
+	KBL_REVID_C0,
+	KBL_REVID_D0,
+	KBL_REVID_D1,
+	KBL_REVID_E0,
+	KBL_REVID_F0,
+	KBL_REVID_G0,
+};
+
+struct i915_rev_steppings {
+	u8 gt_stepping;
+	u8 disp_stepping;
+};
+
+/* Defined in intel_workarounds.c */
+extern const struct i915_rev_steppings kbl_revids[];
 
-#define IS_KBL_REVID(dev_priv, since, until) \
-	(IS_KABYLAKE(dev_priv) && IS_REVID(dev_priv, since, until))
+#define IS_KBL_GT_REVID(dev_priv, since, until) \
+	(IS_KABYLAKE(dev_priv) && \
+	 kbl_revids[INTEL_REVID(dev_priv)].gt_stepping >= since && \
+	 kbl_revids[INTEL_REVID(dev_priv)].gt_stepping <= until)
+#define IS_KBL_DISP_REVID(dev_priv, since, until) \
+	(IS_KABYLAKE(dev_priv) && \
+	 kbl_revids[INTEL_REVID(dev_priv)].disp_stepping >= since && \
+	 kbl_revids[INTEL_REVID(dev_priv)].disp_stepping <= until)
 
 #define GLK_REVID_A0		0x0
 #define GLK_REVID_A1		0x1
@@ -1547,12 +1561,41 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define IS_EHL_REVID(p, since, until) \
 	(IS_ELKHARTLAKE(p) && IS_REVID(p, since, until))
 
-#define TGL_REVID_A0		0x0
-#define TGL_REVID_B0		0x1
-#define TGL_REVID_C0		0x2
+enum {
+	TGL_REVID_A0,
+	TGL_REVID_B0,
+	TGL_REVID_B1,
+	TGL_REVID_C0,
+	TGL_REVID_D0,
+};
+
+extern const struct i915_rev_steppings tgl_uy_revids[];
+extern const struct i915_rev_steppings tgl_revids[];
+
+static inline const struct i915_rev_steppings *
+tgl_revids_get(struct drm_i915_private *dev_priv)
+{
+	if (IS_TGL_U(dev_priv) || IS_TGL_Y(dev_priv))
+		return tgl_uy_revids;
+	else
+		return tgl_revids;
+}
+
+#define IS_TGL_DISP_REVID(p, since, until) \
+	(IS_TIGERLAKE(p) && \
+	 tgl_revids_get(p)->disp_stepping >= (since) && \
+	 tgl_revids_get(p)->disp_stepping <= (until))
+
+#define IS_TGL_UY_GT_REVID(p, since, until) \
+	((IS_TGL_U(p) || IS_TGL_Y(p)) && \
+	 tgl_uy_revids->gt_stepping >= (since) && \
+	 tgl_uy_revids->gt_stepping <= (until))
 
-#define IS_TGL_REVID(p, since, until) \
-	(IS_TIGERLAKE(p) && IS_REVID(p, since, until))
+#define IS_TGL_GT_REVID(p, since, until) \
+	(IS_TIGERLAKE(p) && \
+	 !(IS_TGL_U(p) || IS_TGL_Y(p)) && \
+	 tgl_revids->gt_stepping >= (since) && \
+	 tgl_revids->gt_stepping <= (until))
 
 #define RKL_REVID_A0		0x0
 #define RKL_REVID_B0		0x1
@@ -1665,6 +1708,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define HAS_PSR(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_psr)
 #define HAS_PSR_HW_TRACKING(dev_priv) \
 	(INTEL_INFO(dev_priv)->display.has_psr_hw_tracking)
+#define HAS_PSR2_SEL_FETCH(dev_priv)	 (INTEL_GEN(dev_priv) >= 12)
 #define HAS_TRANSCODER(dev_priv, trans)	 ((INTEL_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
 
 #define HAS_RC6(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6)
@@ -1790,11 +1834,18 @@ static inline void i915_gem_drain_workqueue(struct drm_i915_private *i915)
 }
 
 struct i915_vma * __must_check
+i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
+			    struct i915_gem_ww_ctx *ww,
+			    const struct i915_ggtt_view *view,
+			    u64 size, u64 alignment, u64 flags);
+
+static inline struct i915_vma * __must_check
 i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 			 const struct i915_ggtt_view *view,
-			 u64 size,
-			 u64 alignment,
-			 u64 flags);
+			 u64 size, u64 alignment, u64 flags)
+{
+	return i915_gem_object_ggtt_pin_ww(obj, NULL, view, size, alignment, flags);
+}
 
 int i915_gem_object_unbind(struct drm_i915_gem_object *obj,
 			   unsigned long flags);
@@ -1831,7 +1882,6 @@ void i915_gem_suspend_late(struct drm_i915_private *dev_priv);
 void i915_gem_resume(struct drm_i915_private *dev_priv);
 
 int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file);
-void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 
 int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				    enum i915_cache_level cache_level);
@@ -1899,8 +1949,8 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine);
 void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine);
 int intel_engine_cmd_parser(struct intel_engine_cs *engine,
 			    struct i915_vma *batch,
-			    u32 batch_offset,
-			    u32 batch_length,
+			    unsigned long batch_offset,
+			    unsigned long batch_length,
 			    struct i915_vma *shadow,
 			    bool trampoline);
 #define I915_CMD_PARSER_TRAMPOLINE_SIZE 8
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 9aa3066cb75d..bb0c12975f38 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -335,12 +335,20 @@ i915_gem_shmem_pread(struct drm_i915_gem_object *obj,
 	u64 remain;
 	int ret;
 
-	ret = i915_gem_object_prepare_read(obj, &needs_clflush);
+	ret = i915_gem_object_lock_interruptible(obj, NULL);
 	if (ret)
 		return ret;
 
+	ret = i915_gem_object_prepare_read(obj, &needs_clflush);
+	if (ret) {
+		i915_gem_object_unlock(obj);
+		return ret;
+	}
+
 	fence = i915_gem_object_lock_fence(obj);
 	i915_gem_object_finish_access(obj);
+	i915_gem_object_unlock(obj);
+
 	if (!fence)
 		return -ENOMEM;
 
@@ -420,7 +428,7 @@ i915_gem_gtt_pread(struct drm_i915_gem_object *obj,
 		GEM_BUG_ON(!drm_mm_node_allocated(&node));
 	}
 
-	ret = i915_gem_object_lock_interruptible(obj);
+	ret = i915_gem_object_lock_interruptible(obj, NULL);
 	if (ret)
 		goto out_unpin;
 
@@ -619,7 +627,7 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj,
 		GEM_BUG_ON(!drm_mm_node_allocated(&node));
 	}
 
-	ret = i915_gem_object_lock_interruptible(obj);
+	ret = i915_gem_object_lock_interruptible(obj, NULL);
 	if (ret)
 		goto out_unpin;
 
@@ -734,12 +742,20 @@ i915_gem_shmem_pwrite(struct drm_i915_gem_object *obj,
 	u64 remain;
 	int ret;
 
-	ret = i915_gem_object_prepare_write(obj, &needs_clflush);
+	ret = i915_gem_object_lock_interruptible(obj, NULL);
 	if (ret)
 		return ret;
 
+	ret = i915_gem_object_prepare_write(obj, &needs_clflush);
+	if (ret) {
+		i915_gem_object_unlock(obj);
+		return ret;
+	}
+
 	fence = i915_gem_object_lock_fence(obj);
 	i915_gem_object_finish_access(obj);
+	i915_gem_object_unlock(obj);
+
 	if (!fence)
 		return -ENOMEM;
 
@@ -946,11 +962,10 @@ static void discard_ggtt_vma(struct i915_vma *vma)
 }
 
 struct i915_vma *
-i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
-			 const struct i915_ggtt_view *view,
-			 u64 size,
-			 u64 alignment,
-			 u64 flags)
+i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
+			    struct i915_gem_ww_ctx *ww,
+			    const struct i915_ggtt_view *view,
+			    u64 size, u64 alignment, u64 flags)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 	struct i915_ggtt *ggtt = &i915->ggtt;
@@ -1016,7 +1031,7 @@ new_vma:
 			return ERR_PTR(ret);
 	}
 
-	ret = i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL);
+	ret = i915_vma_pin_ww(vma, ww, size, alignment, flags | PIN_GLOBAL);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -1290,7 +1305,7 @@ int i915_gem_freeze_late(struct drm_i915_private *i915)
 	i915_gem_drain_freed_objects(i915);
 
 	list_for_each_entry(obj, &i915->mm.shrink_list, mm.link) {
-		i915_gem_object_lock(obj);
+		i915_gem_object_lock(obj, NULL);
 		drm_WARN_ON(&i915->drm,
 			    i915_gem_object_set_to_cpu_domain(obj, true));
 		i915_gem_object_unlock(obj);
@@ -1301,21 +1316,6 @@ int i915_gem_freeze_late(struct drm_i915_private *i915)
 	return 0;
 }
 
-void i915_gem_release(struct drm_device *dev, struct drm_file *file)
-{
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_request *request;
-
-	/* Clean up our request list when the client is going away, so that
-	 * later retire_requests won't dereference our soon-to-be-gone
-	 * file_priv.
-	 */
-	spin_lock(&file_priv->mm.lock);
-	list_for_each_entry(request, &file_priv->mm.request_list, client_link)
-		request->file_priv = NULL;
-	spin_unlock(&file_priv->mm.lock);
-}
-
 int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv;
@@ -1331,9 +1331,6 @@ int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file)
 	file_priv->dev_priv = i915;
 	file_priv->file = file;
 
-	spin_lock_init(&file_priv->mm.lock);
-	INIT_LIST_HEAD(&file_priv->mm.request_list);
-
 	file_priv->bsd_engine = -1;
 	file_priv->hang_timestamp = jiffies;
 
@@ -1344,6 +1341,58 @@ int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file)
 	return ret;
 }
 
+void i915_gem_ww_ctx_init(struct i915_gem_ww_ctx *ww, bool intr)
+{
+	ww_acquire_init(&ww->ctx, &reservation_ww_class);
+	INIT_LIST_HEAD(&ww->obj_list);
+	ww->intr = intr;
+	ww->contended = NULL;
+}
+
+static void i915_gem_ww_ctx_unlock_all(struct i915_gem_ww_ctx *ww)
+{
+	struct drm_i915_gem_object *obj;
+
+	while ((obj = list_first_entry_or_null(&ww->obj_list, struct drm_i915_gem_object, obj_link))) {
+		list_del(&obj->obj_link);
+		i915_gem_object_unlock(obj);
+	}
+}
+
+void i915_gem_ww_unlock_single(struct drm_i915_gem_object *obj)
+{
+	list_del(&obj->obj_link);
+	i915_gem_object_unlock(obj);
+}
+
+void i915_gem_ww_ctx_fini(struct i915_gem_ww_ctx *ww)
+{
+	i915_gem_ww_ctx_unlock_all(ww);
+	WARN_ON(ww->contended);
+	ww_acquire_fini(&ww->ctx);
+}
+
+int __must_check i915_gem_ww_ctx_backoff(struct i915_gem_ww_ctx *ww)
+{
+	int ret = 0;
+
+	if (WARN_ON(!ww->contended))
+		return -EINVAL;
+
+	i915_gem_ww_ctx_unlock_all(ww);
+	if (ww->intr)
+		ret = dma_resv_lock_slow_interruptible(ww->contended->base.resv, &ww->ctx);
+	else
+		dma_resv_lock_slow(ww->contended->base.resv, &ww->ctx);
+
+	if (!ret)
+		list_add_tail(&ww->contended->obj_link, &ww->obj_list);
+
+	ww->contended = NULL;
+
+	return ret;
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/mock_gem_device.c"
 #include "selftests/i915_gem.c"
diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h
index f333e88a2b6e..a4cad3f154ca 100644
--- a/drivers/gpu/drm/i915/i915_gem.h
+++ b/drivers/gpu/drm/i915/i915_gem.h
@@ -116,4 +116,16 @@ static inline bool __tasklet_is_scheduled(struct tasklet_struct *t)
 	return test_bit(TASKLET_STATE_SCHED, &t->state);
 }
 
+struct i915_gem_ww_ctx {
+	struct ww_acquire_ctx ctx;
+	struct list_head obj_list;
+	bool intr;
+	struct drm_i915_gem_object *contended;
+};
+
+void i915_gem_ww_ctx_init(struct i915_gem_ww_ctx *ctx, bool intr);
+void i915_gem_ww_ctx_fini(struct i915_gem_ww_ctx *ctx);
+int __must_check i915_gem_ww_ctx_backoff(struct i915_gem_ww_ctx *ctx);
+void i915_gem_ww_unlock_single(struct drm_i915_gem_object *obj);
+
 #endif /* __I915_GEM_H__ */
diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c
index 421613219ae9..f96032c60a12 100644
--- a/drivers/gpu/drm/i915/i915_getparam.c
+++ b/drivers/gpu/drm/i915/i915_getparam.c
@@ -132,6 +132,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data,
 	case I915_PARAM_HAS_EXEC_BATCH_FIRST:
 	case I915_PARAM_HAS_EXEC_FENCE_ARRAY:
 	case I915_PARAM_HAS_EXEC_SUBMIT_FENCE:
+	case I915_PARAM_HAS_EXEC_TIMELINE_FENCES:
 		/* For the time being all of these are always true;
 		 * if some supported hardware does not have one of these
 		 * features this value needs to be provided from
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 3e6cbb0d1150..a635ec8d0b94 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -311,6 +311,8 @@ static int compress_page(struct i915_vma_compress *c,
 
 		if (zlib_deflate(zstream, Z_NO_FLUSH) != Z_OK)
 			return -EIO;
+
+		cond_resched();
 	} while (zstream->avail_in);
 
 	/* Fallback to uncompressed if we increase size? */
@@ -397,6 +399,7 @@ static int compress_page(struct i915_vma_compress *c,
 	if (!(wc && i915_memcpy_from_wc(ptr, src, PAGE_SIZE)))
 		memcpy(ptr, src, PAGE_SIZE);
 	dst->pages[dst->page_count++] = ptr;
+	cond_resched();
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 1fa67700d8f4..759f523c6a6b 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -41,6 +41,7 @@
 #include "display/intel_lpe_audio.h"
 #include "display/intel_psr.h"
 
+#include "gt/intel_breadcrumbs.h"
 #include "gt/intel_gt.h"
 #include "gt/intel_gt_irq.h"
 #include "gt/intel_gt_pm_irq.h"
@@ -131,40 +132,24 @@ static const u32 hpd_bxt[HPD_NUM_PINS] = {
 };
 
 static const u32 hpd_gen11[HPD_NUM_PINS] = {
-	[HPD_PORT_C] = GEN11_TC1_HOTPLUG | GEN11_TBT1_HOTPLUG,
-	[HPD_PORT_D] = GEN11_TC2_HOTPLUG | GEN11_TBT2_HOTPLUG,
-	[HPD_PORT_E] = GEN11_TC3_HOTPLUG | GEN11_TBT3_HOTPLUG,
-	[HPD_PORT_F] = GEN11_TC4_HOTPLUG | GEN11_TBT4_HOTPLUG,
-};
-
-static const u32 hpd_gen12[HPD_NUM_PINS] = {
-	[HPD_PORT_D] = GEN11_TC1_HOTPLUG | GEN11_TBT1_HOTPLUG,
-	[HPD_PORT_E] = GEN11_TC2_HOTPLUG | GEN11_TBT2_HOTPLUG,
-	[HPD_PORT_F] = GEN11_TC3_HOTPLUG | GEN11_TBT3_HOTPLUG,
-	[HPD_PORT_G] = GEN11_TC4_HOTPLUG | GEN11_TBT4_HOTPLUG,
-	[HPD_PORT_H] = GEN12_TC5_HOTPLUG | GEN12_TBT5_HOTPLUG,
-	[HPD_PORT_I] = GEN12_TC6_HOTPLUG | GEN12_TBT6_HOTPLUG,
+	[HPD_PORT_TC1] = GEN11_TC_HOTPLUG(PORT_TC1) | GEN11_TBT_HOTPLUG(PORT_TC1),
+	[HPD_PORT_TC2] = GEN11_TC_HOTPLUG(PORT_TC2) | GEN11_TBT_HOTPLUG(PORT_TC2),
+	[HPD_PORT_TC3] = GEN11_TC_HOTPLUG(PORT_TC3) | GEN11_TBT_HOTPLUG(PORT_TC3),
+	[HPD_PORT_TC4] = GEN11_TC_HOTPLUG(PORT_TC4) | GEN11_TBT_HOTPLUG(PORT_TC4),
+	[HPD_PORT_TC5] = GEN11_TC_HOTPLUG(PORT_TC5) | GEN11_TBT_HOTPLUG(PORT_TC5),
+	[HPD_PORT_TC6] = GEN11_TC_HOTPLUG(PORT_TC6) | GEN11_TBT_HOTPLUG(PORT_TC6),
 };
 
 static const u32 hpd_icp[HPD_NUM_PINS] = {
 	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(PORT_A),
 	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(PORT_B),
-	[HPD_PORT_C] = SDE_TC_HOTPLUG_ICP(PORT_TC1),
-	[HPD_PORT_D] = SDE_TC_HOTPLUG_ICP(PORT_TC2),
-	[HPD_PORT_E] = SDE_TC_HOTPLUG_ICP(PORT_TC3),
-	[HPD_PORT_F] = SDE_TC_HOTPLUG_ICP(PORT_TC4),
-};
-
-static const u32 hpd_tgp[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(PORT_A),
-	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(PORT_B),
 	[HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(PORT_C),
-	[HPD_PORT_D] = SDE_TC_HOTPLUG_ICP(PORT_TC1),
-	[HPD_PORT_E] = SDE_TC_HOTPLUG_ICP(PORT_TC2),
-	[HPD_PORT_F] = SDE_TC_HOTPLUG_ICP(PORT_TC3),
-	[HPD_PORT_G] = SDE_TC_HOTPLUG_ICP(PORT_TC4),
-	[HPD_PORT_H] = SDE_TC_HOTPLUG_ICP(PORT_TC5),
-	[HPD_PORT_I] = SDE_TC_HOTPLUG_ICP(PORT_TC6),
+	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(PORT_TC1),
+	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(PORT_TC2),
+	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(PORT_TC3),
+	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(PORT_TC4),
+	[HPD_PORT_TC5] = SDE_TC_HOTPLUG_ICP(PORT_TC5),
+	[HPD_PORT_TC6] = SDE_TC_HOTPLUG_ICP(PORT_TC6),
 };
 
 static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
@@ -180,9 +165,7 @@ static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
 		return;
 	}
 
-	if (INTEL_GEN(dev_priv) >= 12)
-		hpd->hpd = hpd_gen12;
-	else if (INTEL_GEN(dev_priv) >= 11)
+	if (INTEL_GEN(dev_priv) >= 11)
 		hpd->hpd = hpd_gen11;
 	else if (IS_GEN9_LP(dev_priv))
 		hpd->hpd = hpd_bxt;
@@ -196,9 +179,8 @@ static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
 	if (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv))
 		return;
 
-	if (HAS_PCH_TGP(dev_priv) || HAS_PCH_JSP(dev_priv))
-		hpd->pch_hpd = hpd_tgp;
-	else if (HAS_PCH_ICP(dev_priv) || HAS_PCH_MCC(dev_priv))
+	if (HAS_PCH_TGP(dev_priv) || HAS_PCH_JSP(dev_priv) ||
+	    HAS_PCH_ICP(dev_priv) || HAS_PCH_MCC(dev_priv))
 		hpd->pch_hpd = hpd_icp;
 	else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv))
 		hpd->pch_hpd = hpd_spt;
@@ -1048,33 +1030,17 @@ out:
 static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
 {
 	switch (pin) {
-	case HPD_PORT_C:
-		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC1);
-	case HPD_PORT_D:
-		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC2);
-	case HPD_PORT_E:
-		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC3);
-	case HPD_PORT_F:
-		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC4);
-	default:
-		return false;
-	}
-}
-
-static bool gen12_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_D:
+	case HPD_PORT_TC1:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC1);
-	case HPD_PORT_E:
+	case HPD_PORT_TC2:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC2);
-	case HPD_PORT_F:
+	case HPD_PORT_TC3:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC3);
-	case HPD_PORT_G:
+	case HPD_PORT_TC4:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC4);
-	case HPD_PORT_H:
+	case HPD_PORT_TC5:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC5);
-	case HPD_PORT_I:
+	case HPD_PORT_TC6:
 		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(PORT_TC6);
 	default:
 		return false;
@@ -1112,33 +1078,17 @@ static bool icp_ddi_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
 static bool icp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
 {
 	switch (pin) {
-	case HPD_PORT_C:
-		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC1);
-	case HPD_PORT_D:
-		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC2);
-	case HPD_PORT_E:
-		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC3);
-	case HPD_PORT_F:
-		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC4);
-	default:
-		return false;
-	}
-}
-
-static bool tgp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_D:
+	case HPD_PORT_TC1:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC1);
-	case HPD_PORT_E:
+	case HPD_PORT_TC2:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC2);
-	case HPD_PORT_F:
+	case HPD_PORT_TC3:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC3);
-	case HPD_PORT_G:
+	case HPD_PORT_TC4:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC4);
-	case HPD_PORT_H:
+	case HPD_PORT_TC5:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC5);
-	case HPD_PORT_I:
+	case HPD_PORT_TC6:
 		return val & ICP_TC_HPD_LONG_DETECT(PORT_TC6);
 	default:
 		return false;
@@ -1892,19 +1842,16 @@ static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 {
 	u32 ddi_hotplug_trigger, tc_hotplug_trigger;
 	u32 pin_mask = 0, long_mask = 0;
-	bool (*tc_port_hotplug_long_detect)(enum hpd_pin pin, u32 val);
 
 	if (HAS_PCH_TGP(dev_priv)) {
 		ddi_hotplug_trigger = pch_iir & SDE_DDI_MASK_TGP;
 		tc_hotplug_trigger = pch_iir & SDE_TC_MASK_TGP;
-		tc_port_hotplug_long_detect = tgp_tc_port_hotplug_long_detect;
 	} else if (HAS_PCH_JSP(dev_priv)) {
 		ddi_hotplug_trigger = pch_iir & SDE_DDI_MASK_TGP;
 		tc_hotplug_trigger = 0;
 	} else if (HAS_PCH_MCC(dev_priv)) {
 		ddi_hotplug_trigger = pch_iir & SDE_DDI_MASK_ICP;
 		tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_ICP(PORT_TC1);
-		tc_port_hotplug_long_detect = icp_tc_port_hotplug_long_detect;
 	} else {
 		drm_WARN(&dev_priv->drm, !HAS_PCH_ICP(dev_priv),
 			 "Unrecognized PCH type 0x%x\n",
@@ -1912,7 +1859,6 @@ static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 
 		ddi_hotplug_trigger = pch_iir & SDE_DDI_MASK_ICP;
 		tc_hotplug_trigger = pch_iir & SDE_TC_MASK_ICP;
-		tc_port_hotplug_long_detect = icp_tc_port_hotplug_long_detect;
 	}
 
 	if (ddi_hotplug_trigger) {
@@ -1936,7 +1882,7 @@ static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
 				   tc_hotplug_trigger, dig_hotplug_reg,
 				   dev_priv->hotplug.pch_hpd,
-				   tc_port_hotplug_long_detect);
+				   icp_tc_port_hotplug_long_detect);
 	}
 
 	if (pin_mask)
@@ -2184,12 +2130,6 @@ static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 	u32 pin_mask = 0, long_mask = 0;
 	u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK;
 	u32 trigger_tbt = iir & GEN11_DE_TBT_HOTPLUG_MASK;
-	long_pulse_detect_func long_pulse_detect;
-
-	if (INTEL_GEN(dev_priv) >= 12)
-		long_pulse_detect = gen12_port_hotplug_long_detect;
-	else
-		long_pulse_detect = gen11_port_hotplug_long_detect;
 
 	if (trigger_tc) {
 		u32 dig_hotplug_reg;
@@ -2200,7 +2140,7 @@ static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
 				   trigger_tc, dig_hotplug_reg,
 				   dev_priv->hotplug.hpd,
-				   long_pulse_detect);
+				   gen11_port_hotplug_long_detect);
 	}
 
 	if (trigger_tbt) {
@@ -2212,7 +2152,7 @@ static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
 				   trigger_tbt, dig_hotplug_reg,
 				   dev_priv->hotplug.hpd,
-				   long_pulse_detect);
+				   gen11_port_hotplug_long_detect);
 	}
 
 	if (pin_mask)
@@ -3047,6 +2987,18 @@ static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv,
 	return enabled_irqs;
 }
 
+static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv,
+				  const u32 hpd[HPD_NUM_PINS])
+{
+	struct intel_encoder *encoder;
+	u32 hotplug_irqs = 0;
+
+	for_each_intel_encoder(&dev_priv->drm, encoder)
+		hotplug_irqs |= hpd[encoder->hpd_pin];
+
+	return hotplug_irqs;
+}
+
 static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug;
@@ -3076,50 +3028,50 @@ static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	if (HAS_PCH_IBX(dev_priv))
-		hotplug_irqs = SDE_HOTPLUG_MASK;
-	else
-		hotplug_irqs = SDE_HOTPLUG_MASK_CPT;
-
 	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
 
 	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
 
 	ibx_hpd_detection_setup(dev_priv);
 }
 
-static void icp_hpd_detection_setup(struct drm_i915_private *dev_priv,
-				    u32 ddi_hotplug_enable_mask,
-				    u32 tc_hotplug_enable_mask)
+static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv,
+					u32 enable_mask)
 {
 	u32 hotplug;
 
 	hotplug = I915_READ(SHOTPLUG_CTL_DDI);
-	hotplug |= ddi_hotplug_enable_mask;
+	hotplug |= enable_mask;
 	I915_WRITE(SHOTPLUG_CTL_DDI, hotplug);
+}
 
-	if (tc_hotplug_enable_mask) {
-		hotplug = I915_READ(SHOTPLUG_CTL_TC);
-		hotplug |= tc_hotplug_enable_mask;
-		I915_WRITE(SHOTPLUG_CTL_TC, hotplug);
-	}
+static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv,
+				       u32 enable_mask)
+{
+	u32 hotplug;
+
+	hotplug = I915_READ(SHOTPLUG_CTL_TC);
+	hotplug |= enable_mask;
+	I915_WRITE(SHOTPLUG_CTL_TC, hotplug);
 }
 
 static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv,
-			      u32 sde_ddi_mask, u32 sde_tc_mask,
 			      u32 ddi_enable_mask, u32 tc_enable_mask)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	hotplug_irqs = sde_ddi_mask | sde_tc_mask;
 	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
 
 	if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP)
 		I915_WRITE(SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
 
 	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
 
-	icp_hpd_detection_setup(dev_priv, ddi_enable_mask, tc_enable_mask);
+	icp_ddi_hpd_detection_setup(dev_priv, ddi_enable_mask);
+	if (tc_enable_mask)
+		icp_tc_hpd_detection_setup(dev_priv, tc_enable_mask);
 }
 
 /*
@@ -3129,7 +3081,6 @@ static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv,
 static void mcc_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	icp_hpd_irq_setup(dev_priv,
-			  SDE_DDI_MASK_ICP, SDE_TC_HOTPLUG_ICP(PORT_TC1),
 			  ICP_DDI_HPD_ENABLE_MASK, ICP_TC_HPD_ENABLE(PORT_TC1));
 }
 
@@ -3141,7 +3092,6 @@ static void mcc_hpd_irq_setup(struct drm_i915_private *dev_priv)
 static void jsp_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	icp_hpd_irq_setup(dev_priv,
-			  SDE_DDI_MASK_TGP, 0,
 			  TGP_DDI_HPD_ENABLE_MASK, 0);
 }
 
@@ -3153,14 +3103,18 @@ static void gen11_hpd_detection_setup(struct drm_i915_private *dev_priv)
 	hotplug |= GEN11_HOTPLUG_CTL_ENABLE(PORT_TC1) |
 		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC2) |
 		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC3) |
-		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC4);
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC4) |
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC5) |
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC6);
 	I915_WRITE(GEN11_TC_HOTPLUG_CTL, hotplug);
 
 	hotplug = I915_READ(GEN11_TBT_HOTPLUG_CTL);
 	hotplug |= GEN11_HOTPLUG_CTL_ENABLE(PORT_TC1) |
 		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC2) |
 		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC3) |
-		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC4);
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC4) |
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC5) |
+		   GEN11_HOTPLUG_CTL_ENABLE(PORT_TC6);
 	I915_WRITE(GEN11_TBT_HOTPLUG_CTL, hotplug);
 }
 
@@ -3170,7 +3124,7 @@ static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	u32 val;
 
 	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
-	hotplug_irqs = GEN11_DE_TC_HOTPLUG_MASK | GEN11_DE_TBT_HOTPLUG_MASK;
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.hpd);
 
 	val = I915_READ(GEN11_DE_HPD_IMR);
 	val &= ~hotplug_irqs;
@@ -3181,10 +3135,10 @@ static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	gen11_hpd_detection_setup(dev_priv);
 
 	if (INTEL_PCH_TYPE(dev_priv) >= PCH_TGP)
-		icp_hpd_irq_setup(dev_priv, SDE_DDI_MASK_TGP, SDE_TC_MASK_TGP,
+		icp_hpd_irq_setup(dev_priv,
 				  TGP_DDI_HPD_ENABLE_MASK, TGP_TC_HPD_ENABLE_MASK);
 	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		icp_hpd_irq_setup(dev_priv, SDE_DDI_MASK_ICP, SDE_TC_MASK_ICP,
+		icp_hpd_irq_setup(dev_priv,
 				  ICP_DDI_HPD_ENABLE_MASK, ICP_TC_HPD_ENABLE_MASK);
 }
 
@@ -3220,8 +3174,8 @@ static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
 		I915_WRITE(SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
 
-	hotplug_irqs = SDE_HOTPLUG_MASK_SPT;
 	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.pch_hpd);
 
 	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
 
@@ -3248,22 +3202,13 @@ static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	if (INTEL_GEN(dev_priv) >= 8) {
-		hotplug_irqs = GEN8_PORT_DP_A_HOTPLUG;
-		enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.hpd);
 
+	if (INTEL_GEN(dev_priv) >= 8)
 		bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
-	} else if (INTEL_GEN(dev_priv) >= 7) {
-		hotplug_irqs = DE_DP_A_HOTPLUG_IVB;
-		enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
-
-		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
-	} else {
-		hotplug_irqs = DE_DP_A_HOTPLUG;
-		enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
-
+	else
 		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
-	}
 
 	ilk_hpd_detection_setup(dev_priv);
 
@@ -3312,7 +3257,7 @@ static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	u32 hotplug_irqs, enabled_irqs;
 
 	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->hotplug.hpd);
-	hotplug_irqs = BXT_DE_PORT_HOTPLUG_MASK;
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->hotplug.hpd);
 
 	bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
 
@@ -3533,17 +3478,18 @@ static void icp_irq_postinstall(struct drm_i915_private *dev_priv)
 	gen3_assert_iir_is_zero(&dev_priv->uncore, SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 
-	if (HAS_PCH_TGP(dev_priv))
-		icp_hpd_detection_setup(dev_priv, TGP_DDI_HPD_ENABLE_MASK,
-					TGP_TC_HPD_ENABLE_MASK);
-	else if (HAS_PCH_JSP(dev_priv))
-		icp_hpd_detection_setup(dev_priv, TGP_DDI_HPD_ENABLE_MASK, 0);
-	else if (HAS_PCH_MCC(dev_priv))
-		icp_hpd_detection_setup(dev_priv, ICP_DDI_HPD_ENABLE_MASK,
-					ICP_TC_HPD_ENABLE(PORT_TC1));
-	else
-		icp_hpd_detection_setup(dev_priv, ICP_DDI_HPD_ENABLE_MASK,
-					ICP_TC_HPD_ENABLE_MASK);
+	if (HAS_PCH_TGP(dev_priv)) {
+		icp_ddi_hpd_detection_setup(dev_priv, TGP_DDI_HPD_ENABLE_MASK);
+		icp_tc_hpd_detection_setup(dev_priv, TGP_TC_HPD_ENABLE_MASK);
+	} else if (HAS_PCH_JSP(dev_priv)) {
+		icp_ddi_hpd_detection_setup(dev_priv, TGP_DDI_HPD_ENABLE_MASK);
+	} else if (HAS_PCH_MCC(dev_priv)) {
+		icp_ddi_hpd_detection_setup(dev_priv, ICP_DDI_HPD_ENABLE_MASK);
+		icp_tc_hpd_detection_setup(dev_priv, ICP_TC_HPD_ENABLE(PORT_TC1));
+	} else {
+		icp_ddi_hpd_detection_setup(dev_priv, ICP_DDI_HPD_ENABLE_MASK);
+		icp_tc_hpd_detection_setup(dev_priv, ICP_TC_HPD_ENABLE_MASK);
+	}
 }
 
 static void gen11_irq_postinstall(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 8d8db9ff0a48..7f139ea4a90b 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -102,6 +102,11 @@ i915_param_named(psr_safest_params, bool, 0400,
 	"is helpful to detect if PSR issues are related to bad values set in "
 	" VBT. (0=use VBT parameters, 1=use safest parameters)");
 
+i915_param_named_unsafe(enable_psr2_sel_fetch, bool, 0400,
+	"Enable PSR2 selective fetch "
+	"(0=disabled, 1=enabled) "
+	"Default: 0");
+
 i915_param_named_unsafe(force_probe, charp, 0400,
 	"Force probe the driver for specified devices. "
 	"See CONFIG_DRM_I915_FORCE_PROBE for details.");
diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h
index 53fb5ba8fbed..330c03e2b4f7 100644
--- a/drivers/gpu/drm/i915/i915_params.h
+++ b/drivers/gpu/drm/i915/i915_params.h
@@ -54,6 +54,7 @@ struct drm_printer;
 	param(int, enable_fbc, -1, 0600) \
 	param(int, enable_psr, -1, 0600) \
 	param(bool, psr_safest_params, false, 0600) \
+	param(bool, enable_psr2_sel_fetch, false, 0600) \
 	param(int, disable_power_well, -1, 0400) \
 	param(int, enable_ips, 1, 0600) \
 	param(int, invert_brightness, 0, 0600) \
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 2338f92ce490..366ddfc8df6b 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -890,6 +890,7 @@ static const struct intel_device_info rkl_info = {
 	.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 		BIT(TRANSCODER_C),
 	.require_force_probe = 1,
+	.display.has_hti = 1,
 	.display.has_psr_hw_tracking = 0,
 	.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0),
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index c6f6370283cf..e94976976571 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -1195,24 +1195,39 @@ static struct intel_context *oa_pin_context(struct i915_perf_stream *stream)
 	struct i915_gem_engines_iter it;
 	struct i915_gem_context *ctx = stream->ctx;
 	struct intel_context *ce;
-	int err;
+	struct i915_gem_ww_ctx ww;
+	int err = -ENODEV;
 
 	for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) {
 		if (ce->engine != stream->engine) /* first match! */
 			continue;
 
-		/*
-		 * As the ID is the gtt offset of the context's vma we
-		 * pin the vma to ensure the ID remains fixed.
-		 */
-		err = intel_context_pin(ce);
-		if (err == 0) {
-			stream->pinned_ctx = ce;
-			break;
-		}
+		err = 0;
+		break;
 	}
 	i915_gem_context_unlock_engines(ctx);
 
+	if (err)
+		return ERR_PTR(err);
+
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	/*
+	 * As the ID is the gtt offset of the context's vma we
+	 * pin the vma to ensure the ID remains fixed.
+	 */
+	err = intel_context_pin_ww(ce, &ww);
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+
+	if (err)
+		return ERR_PTR(err);
+
+	stream->pinned_ctx = ce;
 	return stream->pinned_ctx;
 }
 
@@ -1923,15 +1938,22 @@ emit_oa_config(struct i915_perf_stream *stream,
 {
 	struct i915_request *rq;
 	struct i915_vma *vma;
+	struct i915_gem_ww_ctx ww;
 	int err;
 
 	vma = get_oa_vma(stream, oa_config);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
-	err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH);
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	err = i915_gem_object_lock(vma->obj, &ww);
+	if (err)
+		goto err;
+
+	err = i915_vma_pin_ww(vma, &ww, 0, 0, PIN_GLOBAL | PIN_HIGH);
 	if (err)
-		goto err_vma_put;
+		goto err;
 
 	intel_engine_pm_get(ce->engine);
 	rq = i915_request_create(ce);
@@ -1953,11 +1975,9 @@ emit_oa_config(struct i915_perf_stream *stream,
 			goto err_add_request;
 	}
 
-	i915_vma_lock(vma);
 	err = i915_request_await_object(rq, vma->obj, 0);
 	if (!err)
 		err = i915_vma_move_to_active(vma, rq, 0);
-	i915_vma_unlock(vma);
 	if (err)
 		goto err_add_request;
 
@@ -1971,7 +1991,14 @@ err_add_request:
 	i915_request_add(rq);
 err_vma_unpin:
 	i915_vma_unpin(vma);
-err_vma_put:
+err:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+
+	i915_gem_ww_ctx_fini(&ww);
 	i915_vma_put(vma);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 4e796ff4d7d0..d805d4da6181 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1382,7 +1382,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define  DPIO_CMNRST			(1 << 0)
 
 #define DPIO_PHY(pipe)			((pipe) >> 1)
-#define DPIO_PHY_IOSF_PORT(phy)		(dev_priv->dpio_phy_iosf_port[phy])
 
 /*
  * Per pipe/PLL DPIO regs
@@ -1898,6 +1897,8 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define  PWR_DOWN_LN_3_1_0		(0xb << 4)
 #define  PWR_DOWN_LN_MASK		(0xf << 4)
 #define  PWR_DOWN_LN_SHIFT		4
+#define  EDP4K2K_MODE_OVRD_EN		(1 << 3)
+#define  EDP4K2K_MODE_OVRD_OPTIMIZED	(1 << 2)
 
 #define ICL_PORT_CL_DW12(phy)		_MMIO(_ICL_PORT_CL_DW(12, phy))
 #define   ICL_LANE_ENABLE_AUX		(1 << 0)
@@ -2919,6 +2920,12 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define MBUS_BBOX_CTL_S1		_MMIO(0x45040)
 #define MBUS_BBOX_CTL_S2		_MMIO(0x45044)
 
+#define HDPORT_STATE			_MMIO(0x45050)
+#define   HDPORT_DPLL_USED_MASK		REG_GENMASK(14, 12)
+#define   HDPORT_PHY_USED_DP(phy)	REG_BIT(2 * (phy) + 2)
+#define   HDPORT_PHY_USED_HDMI(phy)	REG_BIT(2 * (phy) + 1)
+#define   HDPORT_ENABLED		REG_BIT(0)
+
 /* Make render/texture TLB fetches lower priorty than associated data
  *   fetches. This is not turned on by default
  */
@@ -7752,32 +7759,20 @@ enum {
 #define GEN11_DE_HPD_IMR		_MMIO(0x44474)
 #define GEN11_DE_HPD_IIR		_MMIO(0x44478)
 #define GEN11_DE_HPD_IER		_MMIO(0x4447c)
-#define  GEN12_TC6_HOTPLUG			(1 << 21)
-#define  GEN12_TC5_HOTPLUG			(1 << 20)
-#define  GEN11_TC4_HOTPLUG			(1 << 19)
-#define  GEN11_TC3_HOTPLUG			(1 << 18)
-#define  GEN11_TC2_HOTPLUG			(1 << 17)
-#define  GEN11_TC1_HOTPLUG			(1 << 16)
 #define  GEN11_TC_HOTPLUG(tc_port)		(1 << ((tc_port) + 16))
-#define  GEN11_DE_TC_HOTPLUG_MASK		(GEN12_TC6_HOTPLUG | \
-						 GEN12_TC5_HOTPLUG | \
-						 GEN11_TC4_HOTPLUG | \
-						 GEN11_TC3_HOTPLUG | \
-						 GEN11_TC2_HOTPLUG | \
-						 GEN11_TC1_HOTPLUG)
-#define  GEN12_TBT6_HOTPLUG			(1 << 5)
-#define  GEN12_TBT5_HOTPLUG			(1 << 4)
-#define  GEN11_TBT4_HOTPLUG			(1 << 3)
-#define  GEN11_TBT3_HOTPLUG			(1 << 2)
-#define  GEN11_TBT2_HOTPLUG			(1 << 1)
-#define  GEN11_TBT1_HOTPLUG			(1 << 0)
+#define  GEN11_DE_TC_HOTPLUG_MASK		(GEN11_TC_HOTPLUG(PORT_TC6) | \
+						 GEN11_TC_HOTPLUG(PORT_TC5) | \
+						 GEN11_TC_HOTPLUG(PORT_TC4) | \
+						 GEN11_TC_HOTPLUG(PORT_TC3) | \
+						 GEN11_TC_HOTPLUG(PORT_TC2) | \
+						 GEN11_TC_HOTPLUG(PORT_TC1))
 #define  GEN11_TBT_HOTPLUG(tc_port)		(1 << (tc_port))
-#define  GEN11_DE_TBT_HOTPLUG_MASK		(GEN12_TBT6_HOTPLUG | \
-						 GEN12_TBT5_HOTPLUG | \
-						 GEN11_TBT4_HOTPLUG | \
-						 GEN11_TBT3_HOTPLUG | \
-						 GEN11_TBT2_HOTPLUG | \
-						 GEN11_TBT1_HOTPLUG)
+#define  GEN11_DE_TBT_HOTPLUG_MASK		(GEN11_TBT_HOTPLUG(PORT_TC6) | \
+						 GEN11_TBT_HOTPLUG(PORT_TC5) | \
+						 GEN11_TBT_HOTPLUG(PORT_TC4) | \
+						 GEN11_TBT_HOTPLUG(PORT_TC3) | \
+						 GEN11_TBT_HOTPLUG(PORT_TC2) | \
+						 GEN11_TBT_HOTPLUG(PORT_TC1))
 
 #define GEN11_TBT_HOTPLUG_CTL				_MMIO(0x44030)
 #define GEN11_TC_HOTPLUG_CTL				_MMIO(0x44038)
@@ -7870,6 +7865,7 @@ enum {
 # define CHICKEN3_DGMG_DONE_FIX_DISABLE		(1 << 2)
 
 #define CHICKEN_PAR1_1			_MMIO(0x42080)
+#define  DIS_RAM_BYPASS_PSR2_MAN_TRACK	(1 << 16)
 #define  SKL_DE_COMPRESSED_HASH_MODE	(1 << 15)
 #define  DPA_MASK_VBLANK_SRD		(1 << 15)
 #define  FORCE_ARB_IDLE_PLANES		(1 << 14)
@@ -8711,6 +8707,7 @@ enum {
 #define  PCH_GMBUSUNIT_CLOCK_GATE_DISABLE (1 << 31)
 #define  PCH_DPLUNIT_CLOCK_GATE_DISABLE (1 << 30)
 #define  PCH_DPLSUNIT_CLOCK_GATE_DISABLE (1 << 29)
+#define  PCH_DPMGUNIT_CLOCK_GATE_DISABLE (1 << 15)
 #define  PCH_CPUNIT_CLOCK_GATE_DISABLE (1 << 14)
 #define  CNP_PWM_CGE_GATING_DISABLE (1 << 13)
 #define  PCH_LP_PARTITION_LEVEL_DISABLE  (1 << 12)
@@ -9217,8 +9214,8 @@ enum {
 #define   DISPLAY_IPS_CONTROL			0x19
 #define   TGL_PCODE_TCCOLD			0x26
 #define     TGL_PCODE_EXIT_TCCOLD_DATA_L_EXIT_FAILED	REG_BIT(0)
-#define     TGL_PCODE_EXIT_TCCOLD_DATA_H_BLOCK_REQ	0
-#define     TGL_PCODE_EXIT_TCCOLD_DATA_H_UNBLOCK_REQ	REG_BIT(0)
+#define     TGL_PCODE_EXIT_TCCOLD_DATA_L_BLOCK_REQ	0
+#define     TGL_PCODE_EXIT_TCCOLD_DATA_L_UNBLOCK_REQ	REG_BIT(0)
             /* See also IPS_CTL */
 #define     IPS_PCODE_CONTROL			(1 << 30)
 #define   HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL	0x1A
@@ -9305,6 +9302,7 @@ enum {
 #define   GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC	(1 << 7)
 
 #define GEN10_SAMPLER_MODE		_MMIO(0xE18C)
+#define   ENABLE_SMALLPL			REG_BIT(15)
 #define   GEN11_SAMPLER_ENABLE_HEADLESS_MSG	REG_BIT(5)
 
 /* IVYBRIDGE DPF */
@@ -10277,12 +10275,18 @@ enum skl_power_gate {
 
 #define ICL_DPCLKA_CFGCR0			_MMIO(0x164280)
 #define  ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)	(1 << _PICK(phy, 10, 11, 24))
+#define  RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)	REG_BIT((phy) + 10)
 #define  ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port)	(1 << ((tc_port) < PORT_TC4 ? \
 						       (tc_port) + 12 : \
 						       (tc_port) - PORT_TC4 + 21))
 #define  ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)	((phy) * 2)
 #define  ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy)	(3 << ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy))
 #define  ICL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll, phy)	((pll) << ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy))
+#define  RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)	_PICK(phy, 0, 2, 4, 27)
+#define  RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy) \
+	(3 << RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy))
+#define  RKL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll, phy) \
+	((pll) << RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy))
 
 /* CNL PLL */
 #define DPLL0_ENABLE		0x46010
@@ -10503,19 +10507,21 @@ enum skl_power_gate {
 
 #define _TGL_DPLL0_CFGCR0		0x164284
 #define _TGL_DPLL1_CFGCR0		0x16428C
-/* TODO: add DPLL4 */
 #define _TGL_TBTPLL_CFGCR0		0x16429C
 #define TGL_DPLL_CFGCR0(pll)		_MMIO_PLL3(pll, _TGL_DPLL0_CFGCR0, \
 						  _TGL_DPLL1_CFGCR0, \
 						  _TGL_TBTPLL_CFGCR0)
+#define RKL_DPLL_CFGCR0(pll)		_MMIO_PLL(pll, _TGL_DPLL0_CFGCR0, \
+						  _TGL_DPLL1_CFGCR0)
 
 #define _TGL_DPLL0_CFGCR1		0x164288
 #define _TGL_DPLL1_CFGCR1		0x164290
-/* TODO: add DPLL4 */
 #define _TGL_TBTPLL_CFGCR1		0x1642A0
 #define TGL_DPLL_CFGCR1(pll)		_MMIO_PLL3(pll, _TGL_DPLL0_CFGCR1, \
 						   _TGL_DPLL1_CFGCR1, \
 						   _TGL_TBTPLL_CFGCR1)
+#define RKL_DPLL_CFGCR1(pll)		_MMIO_PLL(pll, _TGL_DPLL0_CFGCR1, \
+						  _TGL_DPLL1_CFGCR1)
 
 #define _DKL_PHY1_BASE			0x168000
 #define _DKL_PHY2_BASE			0x169000
@@ -12336,4 +12342,10 @@ enum skl_power_gate {
 #define   DSB_ENABLE			(1 << 31)
 #define   DSB_STATUS			(1 << 0)
 
+#define TGL_ROOT_DEVICE_ID		0x9A00
+#define TGL_ROOT_DEVICE_MASK		0xFF00
+#define TGL_ROOT_DEVICE_SKU_MASK	0xF
+#define TGL_ROOT_DEVICE_SKU_ULX		0x2
+#define TGL_ROOT_DEVICE_SKU_ULT		0x4
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index 781a6783affe..0e813819b041 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -31,6 +31,7 @@
 #include <linux/sched/signal.h>
 
 #include "gem/i915_gem_context.h"
+#include "gt/intel_breadcrumbs.h"
 #include "gt/intel_context.h"
 #include "gt/intel_ring.h"
 #include "gt/intel_rps.h"
@@ -186,48 +187,34 @@ static void irq_execute_cb_hook(struct irq_work *wrk)
 	irq_execute_cb(wrk);
 }
 
-static void __notify_execute_cb(struct i915_request *rq)
+static __always_inline void
+__notify_execute_cb(struct i915_request *rq, bool (*fn)(struct irq_work *wrk))
 {
 	struct execute_cb *cb, *cn;
 
-	lockdep_assert_held(&rq->lock);
-
-	GEM_BUG_ON(!i915_request_is_active(rq));
 	if (llist_empty(&rq->execute_cb))
 		return;
 
-	llist_for_each_entry_safe(cb, cn, rq->execute_cb.first, work.llnode)
-		irq_work_queue(&cb->work);
-
-	/*
-	 * XXX Rollback on __i915_request_unsubmit()
-	 *
-	 * In the future, perhaps when we have an active time-slicing scheduler,
-	 * it will be interesting to unsubmit parallel execution and remove
-	 * busywaits from the GPU until their master is restarted. This is
-	 * quite hairy, we have to carefully rollback the fence and do a
-	 * preempt-to-idle cycle on the target engine, all the while the
-	 * master execute_cb may refire.
-	 */
-	init_llist_head(&rq->execute_cb);
+	llist_for_each_entry_safe(cb, cn,
+				  llist_del_all(&rq->execute_cb),
+				  work.llnode)
+		fn(&cb->work);
 }
 
-static inline void
-remove_from_client(struct i915_request *request)
+static void __notify_execute_cb_irq(struct i915_request *rq)
 {
-	struct drm_i915_file_private *file_priv;
+	__notify_execute_cb(rq, irq_work_queue);
+}
 
-	if (!READ_ONCE(request->file_priv))
-		return;
+static bool irq_work_imm(struct irq_work *wrk)
+{
+	wrk->func(wrk);
+	return false;
+}
 
-	rcu_read_lock();
-	file_priv = xchg(&request->file_priv, NULL);
-	if (file_priv) {
-		spin_lock(&file_priv->mm.lock);
-		list_del(&request->client_link);
-		spin_unlock(&file_priv->mm.lock);
-	}
-	rcu_read_unlock();
+static void __notify_execute_cb_imm(struct i915_request *rq)
+{
+	__notify_execute_cb(rq, irq_work_imm);
 }
 
 static void free_capture_list(struct i915_request *request)
@@ -274,9 +261,16 @@ static void remove_from_engine(struct i915_request *rq)
 		locked = engine;
 	}
 	list_del_init(&rq->sched.link);
+
 	clear_bit(I915_FENCE_FLAG_PQUEUE, &rq->fence.flags);
 	clear_bit(I915_FENCE_FLAG_HOLD, &rq->fence.flags);
+
+	/* Prevent further __await_execution() registering a cb, then flush */
+	set_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags);
+
 	spin_unlock_irq(&locked->active.lock);
+
+	__notify_execute_cb_imm(rq);
 }
 
 bool i915_request_retire(struct i915_request *rq)
@@ -288,6 +282,7 @@ bool i915_request_retire(struct i915_request *rq)
 
 	GEM_BUG_ON(!i915_sw_fence_signaled(&rq->submit));
 	trace_i915_request_retire(rq);
+	i915_request_mark_complete(rq);
 
 	/*
 	 * We know the GPU must have read the request to have
@@ -305,32 +300,30 @@ bool i915_request_retire(struct i915_request *rq)
 		__i915_request_fill(rq, POISON_FREE);
 	rq->ring->head = rq->postfix;
 
+	if (!i915_request_signaled(rq)) {
+		spin_lock_irq(&rq->lock);
+		dma_fence_signal_locked(&rq->fence);
+		spin_unlock_irq(&rq->lock);
+	}
+
+	if (i915_request_has_waitboost(rq)) {
+		GEM_BUG_ON(!atomic_read(&rq->engine->gt->rps.num_waiters));
+		atomic_dec(&rq->engine->gt->rps.num_waiters);
+	}
+
 	/*
 	 * We only loosely track inflight requests across preemption,
 	 * and so we may find ourselves attempting to retire a _completed_
 	 * request that we have removed from the HW and put back on a run
 	 * queue.
+	 *
+	 * As we set I915_FENCE_FLAG_ACTIVE on the request, this should be
+	 * after removing the breadcrumb and signaling it, so that we do not
+	 * inadvertently attach the breadcrumb to a completed request.
 	 */
 	remove_from_engine(rq);
-
-	spin_lock_irq(&rq->lock);
-	i915_request_mark_complete(rq);
-	if (!i915_request_signaled(rq))
-		dma_fence_signal_locked(&rq->fence);
-	if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &rq->fence.flags))
-		i915_request_cancel_breadcrumb(rq);
-	if (i915_request_has_waitboost(rq)) {
-		GEM_BUG_ON(!atomic_read(&rq->engine->gt->rps.num_waiters));
-		atomic_dec(&rq->engine->gt->rps.num_waiters);
-	}
-	if (!test_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags)) {
-		set_bit(I915_FENCE_FLAG_ACTIVE, &rq->fence.flags);
-		__notify_execute_cb(rq);
-	}
 	GEM_BUG_ON(!llist_empty(&rq->execute_cb));
-	spin_unlock_irq(&rq->lock);
 
-	remove_from_client(rq);
 	__list_del_entry(&rq->link); /* poison neither prev/next (RCU walks) */
 
 	intel_context_exit(rq->context);
@@ -357,12 +350,6 @@ void i915_request_retire_upto(struct i915_request *rq)
 	} while (i915_request_retire(tmp) && tmp != rq);
 }
 
-static void __llist_add(struct llist_node *node, struct llist_head *head)
-{
-	node->next = head->first;
-	head->first = node;
-}
-
 static struct i915_request * const *
 __engine_active(struct intel_engine_cs *engine)
 {
@@ -460,18 +447,24 @@ __await_execution(struct i915_request *rq,
 		cb->work.func = irq_execute_cb_hook;
 	}
 
-	spin_lock_irq(&signal->lock);
-	if (i915_request_is_active(signal) || __request_in_flight(signal)) {
-		if (hook) {
-			hook(rq, &signal->fence);
-			i915_request_put(signal);
-		}
-		i915_sw_fence_complete(cb->fence);
-		kmem_cache_free(global.slab_execute_cbs, cb);
-	} else {
-		__llist_add(&cb->work.llnode, &signal->execute_cb);
+	/*
+	 * Register the callback first, then see if the signaler is already
+	 * active. This ensures that if we race with the
+	 * __notify_execute_cb from i915_request_submit() and we are not
+	 * included in that list, we get a second bite of the cherry and
+	 * execute it ourselves. After this point, a future
+	 * i915_request_submit() will notify us.
+	 *
+	 * In i915_request_retire() we set the ACTIVE bit on a completed
+	 * request (then flush the execute_cb). So by registering the
+	 * callback first, then checking the ACTIVE bit, we serialise with
+	 * the completed/retired request.
+	 */
+	if (llist_add(&cb->work.llnode, &signal->execute_cb)) {
+		if (i915_request_is_active(signal) ||
+		    __request_in_flight(signal))
+			__notify_execute_cb_imm(signal);
 	}
-	spin_unlock_irq(&signal->lock);
 
 	return 0;
 }
@@ -549,8 +542,13 @@ bool __i915_request_submit(struct i915_request *request)
 	if (i915_request_completed(request))
 		goto xfer;
 
+	if (unlikely(intel_context_is_closed(request->context) &&
+		     !intel_engine_has_heartbeat(engine)))
+		intel_context_set_banned(request->context);
+
 	if (unlikely(intel_context_is_banned(request->context)))
 		i915_request_set_error_once(request, -EIO);
+
 	if (unlikely(fatal_error(request->fence.error)))
 		__i915_request_skip(request);
 
@@ -587,19 +585,21 @@ xfer:
 		clear_bit(I915_FENCE_FLAG_PQUEUE, &request->fence.flags);
 	}
 
-	/* We may be recursing from the signal callback of another i915 fence */
-	if (!i915_request_signaled(request)) {
-		spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING);
-
-		__notify_execute_cb(request);
-		if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
-			     &request->fence.flags) &&
-		    !i915_request_enable_breadcrumb(request))
-			intel_engine_signal_breadcrumbs(engine);
+	/*
+	 * XXX Rollback bonded-execution on __i915_request_unsubmit()?
+	 *
+	 * In the future, perhaps when we have an active time-slicing scheduler,
+	 * it will be interesting to unsubmit parallel execution and remove
+	 * busywaits from the GPU until their master is restarted. This is
+	 * quite hairy, we have to carefully rollback the fence and do a
+	 * preempt-to-idle cycle on the target engine, all the while the
+	 * master execute_cb may refire.
+	 */
+	__notify_execute_cb_irq(request);
 
-		spin_unlock(&request->lock);
-		GEM_BUG_ON(!llist_empty(&request->execute_cb));
-	}
+	/* We may be recursing from the signal callback of another i915 fence */
+	if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags))
+		i915_request_enable_breadcrumb(request);
 
 	return result;
 }
@@ -621,27 +621,27 @@ void __i915_request_unsubmit(struct i915_request *request)
 {
 	struct intel_engine_cs *engine = request->engine;
 
+	/*
+	 * Only unwind in reverse order, required so that the per-context list
+	 * is kept in seqno/ring order.
+	 */
 	RQ_TRACE(request, "\n");
 
 	GEM_BUG_ON(!irqs_disabled());
 	lockdep_assert_held(&engine->active.lock);
 
 	/*
-	 * Only unwind in reverse order, required so that the per-context list
-	 * is kept in seqno/ring order.
+	 * Before we remove this breadcrumb from the signal list, we have
+	 * to ensure that a concurrent dma_fence_enable_signaling() does not
+	 * attach itself. We first mark the request as no longer active and
+	 * make sure that is visible to other cores, and then remove the
+	 * breadcrumb if attached.
 	 */
-
-	/* We may be recursing from the signal callback of another i915 fence */
-	spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING);
-
+	GEM_BUG_ON(!test_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags));
+	clear_bit_unlock(I915_FENCE_FLAG_ACTIVE, &request->fence.flags);
 	if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags))
 		i915_request_cancel_breadcrumb(request);
 
-	GEM_BUG_ON(!test_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags));
-	clear_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags);
-
-	spin_unlock(&request->lock);
-
 	/* We've already spun, don't charge on resubmitting. */
 	if (request->sched.semaphores && i915_request_started(request))
 		request->sched.semaphores = 0;
@@ -778,7 +778,6 @@ static void __i915_request_ctor(void *arg)
 
 	dma_fence_init(&rq->fence, &i915_fence_ops, &rq->lock, 0, 0);
 
-	rq->file_priv = NULL;
 	rq->capture_list = NULL;
 
 	init_llist_head(&rq->execute_cb);
@@ -868,7 +867,6 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp)
 
 	/* No zalloc, everything must be cleared after use */
 	rq->batch = NULL;
-	GEM_BUG_ON(rq->file_priv);
 	GEM_BUG_ON(rq->capture_list);
 	GEM_BUG_ON(!llist_empty(&rq->execute_cb));
 
@@ -1661,7 +1659,7 @@ static bool busywait_stop(unsigned long timeout, unsigned int cpu)
 	return this_cpu != cpu;
 }
 
-static bool __i915_spin_request(const struct i915_request * const rq, int state)
+static bool __i915_spin_request(struct i915_request * const rq, int state)
 {
 	unsigned long timeout_ns;
 	unsigned int cpu;
@@ -1694,7 +1692,7 @@ static bool __i915_spin_request(const struct i915_request * const rq, int state)
 	timeout_ns = READ_ONCE(rq->engine->props.max_busywait_duration_ns);
 	timeout_ns += local_clock_ns(&cpu);
 	do {
-		if (i915_request_completed(rq))
+		if (dma_fence_is_signaled(&rq->fence))
 			return true;
 
 		if (signal_pending_state(state, current))
@@ -1718,7 +1716,7 @@ static void request_wait_wake(struct dma_fence *fence, struct dma_fence_cb *cb)
 {
 	struct request_wait *wait = container_of(cb, typeof(*wait), cb);
 
-	wake_up_process(wait->tsk);
+	wake_up_process(fetch_and_zero(&wait->tsk));
 }
 
 /**
@@ -1787,10 +1785,8 @@ long i915_request_wait(struct i915_request *rq,
 	 * duration, which we currently lack.
 	 */
 	if (IS_ACTIVE(CONFIG_DRM_I915_MAX_REQUEST_BUSYWAIT) &&
-	    __i915_spin_request(rq, state)) {
-		dma_fence_signal(&rq->fence);
+	    __i915_spin_request(rq, state))
 		goto out;
-	}
 
 	/*
 	 * This client is about to stall waiting for the GPU. In many cases
@@ -1804,25 +1800,36 @@ long i915_request_wait(struct i915_request *rq,
 	 * but at a cost of spending more power processing the workload
 	 * (bad for battery).
 	 */
-	if (flags & I915_WAIT_PRIORITY) {
-		if (!i915_request_started(rq) &&
-		    INTEL_GEN(rq->engine->i915) >= 6)
-			intel_rps_boost(rq);
-	}
+	if (flags & I915_WAIT_PRIORITY && !i915_request_started(rq))
+		intel_rps_boost(rq);
 
 	wait.tsk = current;
 	if (dma_fence_add_callback(&rq->fence, &wait.cb, request_wait_wake))
 		goto out;
 
+	/*
+	 * Flush the submission tasklet, but only if it may help this request.
+	 *
+	 * We sometimes experience some latency between the HW interrupts and
+	 * tasklet execution (mostly due to ksoftirqd latency, but it can also
+	 * be due to lazy CS events), so lets run the tasklet manually if there
+	 * is a chance it may submit this request. If the request is not ready
+	 * to run, as it is waiting for other fences to be signaled, flushing
+	 * the tasklet is busy work without any advantage for this client.
+	 *
+	 * If the HW is being lazy, this is the last chance before we go to
+	 * sleep to catch any pending events. We will check periodically in
+	 * the heartbeat to flush the submission tasklets as a last resort
+	 * for unhappy HW.
+	 */
+	if (i915_request_is_ready(rq))
+		intel_engine_flush_submission(rq->engine);
+
 	for (;;) {
 		set_current_state(state);
 
-		if (i915_request_completed(rq)) {
-			dma_fence_signal(&rq->fence);
+		if (dma_fence_is_signaled(&rq->fence))
 			break;
-		}
-
-		intel_engine_flush_submission(rq->engine);
 
 		if (signal_pending_state(state, current)) {
 			timeout = -ERESTARTSYS;
@@ -1838,7 +1845,9 @@ long i915_request_wait(struct i915_request *rq,
 	}
 	__set_current_state(TASK_RUNNING);
 
-	dma_fence_remove_callback(&rq->fence, &wait.cb);
+	if (READ_ONCE(wait.tsk))
+		dma_fence_remove_callback(&rq->fence, &wait.cb);
+	GEM_BUG_ON(!list_empty(&wait.cb.node));
 
 out:
 	mutex_release(&rq->engine->gt->reset.mutex.dep_map, _THIS_IP_);
diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index 590762820761..16b721080195 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -284,10 +284,6 @@ struct i915_request {
 	/** timeline->request entry for this request */
 	struct list_head link;
 
-	struct drm_i915_file_private *file_priv;
-	/** file_priv list entry for this request */
-	struct list_head client_link;
-
 	I915_SELFTEST_DECLARE(struct {
 		struct list_head link;
 		unsigned long delay;
@@ -365,10 +361,6 @@ void i915_request_submit(struct i915_request *request);
 void __i915_request_unsubmit(struct i915_request *request);
 void i915_request_unsubmit(struct i915_request *request);
 
-/* Note: part of the intel_breadcrumbs family */
-bool i915_request_enable_breadcrumb(struct i915_request *request);
-void i915_request_cancel_breadcrumb(struct i915_request *request);
-
 long i915_request_wait(struct i915_request *rq,
 		       unsigned int flags,
 		       long timeout)
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index ed2be3489f8e..7b64e7137270 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -34,17 +34,25 @@
 
 static void i915_save_display(struct drm_i915_private *dev_priv)
 {
+	struct pci_dev *pdev = dev_priv->drm.pdev;
+
 	/* Display arbitration control */
 	if (INTEL_GEN(dev_priv) <= 4)
 		dev_priv->regfile.saveDSPARB = I915_READ(DSPARB);
 
-	/* save FBC interval */
-	if (HAS_FBC(dev_priv) && INTEL_GEN(dev_priv) <= 4 && !IS_G4X(dev_priv))
-		dev_priv->regfile.saveFBC_CONTROL = I915_READ(FBC_CONTROL);
+	if (IS_GEN(dev_priv, 4))
+		pci_read_config_word(pdev, GCDGMBUS,
+				     &dev_priv->regfile.saveGCDGMBUS);
 }
 
 static void i915_restore_display(struct drm_i915_private *dev_priv)
 {
+	struct pci_dev *pdev = dev_priv->drm.pdev;
+
+	if (IS_GEN(dev_priv, 4))
+		pci_write_config_word(pdev, GCDGMBUS,
+				      dev_priv->regfile.saveGCDGMBUS);
+
 	/* Display arbitration */
 	if (INTEL_GEN(dev_priv) <= 4)
 		I915_WRITE(DSPARB, dev_priv->regfile.saveDSPARB);
@@ -52,31 +60,17 @@ static void i915_restore_display(struct drm_i915_private *dev_priv)
 	/* only restore FBC info on the platform that supports FBC*/
 	intel_fbc_global_disable(dev_priv);
 
-	/* restore FBC interval */
-	if (HAS_FBC(dev_priv) && INTEL_GEN(dev_priv) <= 4 && !IS_G4X(dev_priv))
-		I915_WRITE(FBC_CONTROL, dev_priv->regfile.saveFBC_CONTROL);
-
 	intel_vga_redisable(dev_priv);
+
+	intel_gmbus_reset(dev_priv);
 }
 
 int i915_save_state(struct drm_i915_private *dev_priv)
 {
-	struct pci_dev *pdev = dev_priv->drm.pdev;
 	int i;
 
 	i915_save_display(dev_priv);
 
-	if (IS_GEN(dev_priv, 4))
-		pci_read_config_word(pdev, GCDGMBUS,
-				     &dev_priv->regfile.saveGCDGMBUS);
-
-	/* Cache mode state */
-	if (INTEL_GEN(dev_priv) < 7)
-		dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
-
-	/* Memory Arbitration state */
-	dev_priv->regfile.saveMI_ARB_STATE = I915_READ(MI_ARB_STATE);
-
 	/* Scratch space */
 	if (IS_GEN(dev_priv, 2) && IS_MOBILE(dev_priv)) {
 		for (i = 0; i < 7; i++) {
@@ -102,22 +96,10 @@ int i915_save_state(struct drm_i915_private *dev_priv)
 
 int i915_restore_state(struct drm_i915_private *dev_priv)
 {
-	struct pci_dev *pdev = dev_priv->drm.pdev;
 	int i;
 
-	if (IS_GEN(dev_priv, 4))
-		pci_write_config_word(pdev, GCDGMBUS,
-				      dev_priv->regfile.saveGCDGMBUS);
 	i915_restore_display(dev_priv);
 
-	/* Cache mode state */
-	if (INTEL_GEN(dev_priv) < 7)
-		I915_WRITE(CACHE_MODE_0, dev_priv->regfile.saveCACHE_MODE_0 |
-			   0xffff0000);
-
-	/* Memory arbitration state */
-	I915_WRITE(MI_ARB_STATE, dev_priv->regfile.saveMI_ARB_STATE | 0xffff0000);
-
 	/* Scratch space */
 	if (IS_GEN(dev_priv, 2) && IS_MOBILE(dev_priv)) {
 		for (i = 0; i < 7; i++) {
@@ -138,7 +120,5 @@ int i915_restore_state(struct drm_i915_private *dev_priv)
 			I915_WRITE(SWF3(i), dev_priv->regfile.saveSWF3[i]);
 	}
 
-	intel_gmbus_reset(dev_priv);
-
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index bc64f773dcdb..ffb5287e055a 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -291,6 +291,8 @@ i915_vma_instance(struct drm_i915_gem_object *obj,
 
 struct i915_vma_work {
 	struct dma_fence_work base;
+	struct i915_address_space *vm;
+	struct i915_vm_pt_stash stash;
 	struct i915_vma *vma;
 	struct drm_i915_gem_object *pinned;
 	struct i915_sw_dma_fence_cb cb;
@@ -302,13 +304,10 @@ static int __vma_bind(struct dma_fence_work *work)
 {
 	struct i915_vma_work *vw = container_of(work, typeof(*vw), base);
 	struct i915_vma *vma = vw->vma;
-	int err;
-
-	err = vma->ops->bind_vma(vma->vm, vma, vw->cache_level, vw->flags);
-	if (err)
-		atomic_or(I915_VMA_ERROR, &vma->flags);
 
-	return err;
+	vma->ops->bind_vma(vw->vm, &vw->stash,
+			   vma, vw->cache_level, vw->flags);
+	return 0;
 }
 
 static void __vma_release(struct dma_fence_work *work)
@@ -317,6 +316,9 @@ static void __vma_release(struct dma_fence_work *work)
 
 	if (vw->pinned)
 		__i915_gem_object_unpin_pages(vw->pinned);
+
+	i915_vm_free_pt_stash(vw->vm, &vw->stash);
+	i915_vm_put(vw->vm);
 }
 
 static const struct dma_fence_work_ops bind_ops = {
@@ -376,7 +378,6 @@ int i915_vma_bind(struct i915_vma *vma,
 {
 	u32 bind_flags;
 	u32 vma_flags;
-	int ret;
 
 	GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
 	GEM_BUG_ON(vma->size > vma->node.size);
@@ -433,9 +434,7 @@ int i915_vma_bind(struct i915_vma *vma,
 			work->pinned = vma->obj;
 		}
 	} else {
-		ret = vma->ops->bind_vma(vma->vm, vma, cache_level, bind_flags);
-		if (ret)
-			return ret;
+		vma->ops->bind_vma(vma->vm, NULL, vma, cache_level, bind_flags);
 	}
 
 	atomic_or(bind_flags, &vma->flags);
@@ -853,13 +852,19 @@ static void vma_unbind_pages(struct i915_vma *vma)
 	__vma_put_pages(vma, count | count << I915_VMA_PAGES_BIAS);
 }
 
-int i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
+int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
+		    u64 size, u64 alignment, u64 flags)
 {
 	struct i915_vma_work *work = NULL;
 	intel_wakeref_t wakeref = 0;
 	unsigned int bound;
 	int err;
 
+#ifdef CONFIG_PROVE_LOCKING
+	if (debug_locks && lockdep_is_held(&vma->vm->i915->drm.struct_mutex))
+		WARN_ON(!ww);
+#endif
+
 	BUILD_BUG_ON(PIN_GLOBAL != I915_VMA_GLOBAL_BIND);
 	BUILD_BUG_ON(PIN_USER != I915_VMA_LOCAL_BIND);
 
@@ -873,16 +878,32 @@ int i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 	if (err)
 		return err;
 
+	if (flags & PIN_GLOBAL)
+		wakeref = intel_runtime_pm_get(&vma->vm->i915->runtime_pm);
+
 	if (flags & vma->vm->bind_async_flags) {
 		work = i915_vma_work();
 		if (!work) {
 			err = -ENOMEM;
-			goto err_pages;
+			goto err_rpm;
 		}
-	}
 
-	if (flags & PIN_GLOBAL)
-		wakeref = intel_runtime_pm_get(&vma->vm->i915->runtime_pm);
+		work->vm = i915_vm_get(vma->vm);
+
+		/* Allocate enough page directories to used PTE */
+		if (vma->vm->allocate_va_range) {
+			err = i915_vm_alloc_pt_stash(vma->vm,
+						     &work->stash,
+						     vma->size);
+			if (err)
+				goto err_fence;
+
+			err = i915_vm_pin_pt_stash(vma->vm,
+						   &work->stash);
+			if (err)
+				goto err_fence;
+		}
+	}
 
 	/*
 	 * Differentiate between user/kernel vma inside the aliasing-ppgtt.
@@ -971,9 +992,9 @@ err_unlock:
 err_fence:
 	if (work)
 		dma_fence_work_commit_imm(&work->base);
+err_rpm:
 	if (wakeref)
 		intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref);
-err_pages:
 	vma_put_pages(vma);
 	return err;
 }
@@ -989,7 +1010,8 @@ static void flush_idle_contexts(struct intel_gt *gt)
 	intel_gt_wait_for_idle(gt, MAX_SCHEDULE_TIMEOUT);
 }
 
-int i915_ggtt_pin(struct i915_vma *vma, u32 align, unsigned int flags)
+int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
+		  u32 align, unsigned int flags)
 {
 	struct i915_address_space *vm = vma->vm;
 	int err;
@@ -997,7 +1019,7 @@ int i915_ggtt_pin(struct i915_vma *vma, u32 align, unsigned int flags)
 	GEM_BUG_ON(!i915_vma_is_ggtt(vma));
 
 	do {
-		err = i915_vma_pin(vma, 0, align, flags | PIN_GLOBAL);
+		err = i915_vma_pin_ww(vma, ww, 0, align, flags | PIN_GLOBAL);
 		if (err != -ENOSPC) {
 			if (!err) {
 				err = i915_vma_wait_for_bind(vma);
@@ -1167,6 +1189,12 @@ void i915_vma_revoke_mmap(struct i915_vma *vma)
 		list_del(&vma->obj->userfault_link);
 }
 
+static int
+__i915_request_await_bind(struct i915_request *rq, struct i915_vma *vma)
+{
+	return __i915_request_await_exclusive(rq, &vma->active);
+}
+
 int __i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq)
 {
 	int err;
@@ -1174,8 +1202,7 @@ int __i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq)
 	GEM_BUG_ON(!i915_vma_is_pinned(vma));
 
 	/* Wait for the vma to be bound before we start! */
-	err = i915_request_await_active(rq, &vma->active,
-					I915_ACTIVE_AWAIT_EXCL);
+	err = __i915_request_await_bind(rq, vma);
 	if (err)
 		return err;
 
diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
index d0d01f909548..5b3a3c653454 100644
--- a/drivers/gpu/drm/i915/i915_vma.h
+++ b/drivers/gpu/drm/i915/i915_vma.h
@@ -237,8 +237,17 @@ static inline void i915_vma_unlock(struct i915_vma *vma)
 }
 
 int __must_check
-i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags);
-int i915_ggtt_pin(struct i915_vma *vma, u32 align, unsigned int flags);
+i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
+		u64 size, u64 alignment, u64 flags);
+
+static inline int __must_check
+i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
+{
+	return i915_vma_pin_ww(vma, NULL, size, alignment, flags);
+}
+
+int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
+		  u32 align, unsigned int flags);
 
 static inline int i915_vma_pin_count(const struct i915_vma *vma)
 {
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 40c590db3c76..adc836f15fde 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -346,6 +346,25 @@ void intel_device_info_subplatform_init(struct drm_i915_private *i915)
 		mask = BIT(INTEL_SUBPLATFORM_PORTF);
 	}
 
+	if (IS_TIGERLAKE(i915)) {
+		struct pci_dev *root, *pdev = i915->drm.pdev;
+
+		root = list_first_entry(&pdev->bus->devices, typeof(*root), bus_list);
+
+		drm_WARN_ON(&i915->drm, mask);
+		drm_WARN_ON(&i915->drm, (root->device & TGL_ROOT_DEVICE_MASK) !=
+			    TGL_ROOT_DEVICE_ID);
+
+		switch (root->device & TGL_ROOT_DEVICE_SKU_MASK) {
+		case TGL_ROOT_DEVICE_SKU_ULX:
+			mask = BIT(INTEL_SUBPLATFORM_ULX);
+			break;
+		case TGL_ROOT_DEVICE_SKU_ULT:
+			mask = BIT(INTEL_SUBPLATFORM_ULT);
+			break;
+		}
+	}
+
 	GEM_BUG_ON(mask & ~INTEL_SUBPLATFORM_BITS);
 
 	RUNTIME_INFO(i915)->platform_mask[pi] |= mask;
@@ -497,6 +516,14 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 					    S32_MAX),
 				USEC_PER_SEC));
 	}
+
+	if (!HAS_DISPLAY(dev_priv)) {
+		dev_priv->drm.driver_features &= ~(DRIVER_MODESET |
+						   DRIVER_ATOMIC);
+		memset(&info->display, 0, sizeof(info->display));
+		memset(runtime->num_sprites, 0, sizeof(runtime->num_sprites));
+		memset(runtime->num_scalers, 0, sizeof(runtime->num_scalers));
+	}
 }
 
 void intel_driver_caps_print(const struct intel_driver_caps *caps,
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index fd2385457ab6..6a3d607218aa 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -146,6 +146,7 @@ enum intel_ppgtt_type {
 	func(has_gmch); \
 	func(has_hdcp); \
 	func(has_hotplug); \
+	func(has_hti); \
 	func(has_ipc); \
 	func(has_modular_fia); \
 	func(has_overlay); \
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index cfabbe0481ab..34e0d22d456b 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -100,12 +100,6 @@ static void gen9_init_clock_gating(struct drm_i915_private *dev_priv)
 	 */
 	I915_WRITE(DISP_ARB_CTL, I915_READ(DISP_ARB_CTL) |
 		   DISP_FBC_MEMORY_WAKE);
-
-	if (IS_SKYLAKE(dev_priv)) {
-		/* WaDisableDopClockGating */
-		I915_WRITE(GEN7_MISCCPCTL, I915_READ(GEN7_MISCCPCTL)
-			   & ~GEN7_DOP_CLOCK_GATE_ENABLE);
-	}
 }
 
 static void bxt_init_clock_gating(struct drm_i915_private *dev_priv)
@@ -7142,7 +7136,7 @@ static void tgl_init_clock_gating(struct drm_i915_private *dev_priv)
 		   I915_READ(POWERGATE_ENABLE) | vd_pg_enable);
 
 	/* Wa_1409825376:tgl (pre-prod)*/
-	if (IS_TGL_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_A0))
+	if (IS_TGL_DISP_REVID(dev_priv, TGL_REVID_A0, TGL_REVID_B1))
 		I915_WRITE(GEN9_CLKGATE_DIS_3, I915_READ(GEN9_CLKGATE_DIS_3) |
 			   TGL_VRH_GATING_DIS);
 
@@ -7223,12 +7217,12 @@ static void kbl_init_clock_gating(struct drm_i915_private *dev_priv)
 	gen9_init_clock_gating(dev_priv);
 
 	/* WaDisableSDEUnitClockGating:kbl */
-	if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0))
+	if (IS_KBL_GT_REVID(dev_priv, 0, KBL_REVID_B0))
 		I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 			   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
 
 	/* WaDisableGamClockGating:kbl */
-	if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0))
+	if (IS_KBL_GT_REVID(dev_priv, 0, KBL_REVID_B0))
 		I915_WRITE(GEN6_UCGCTL1, I915_READ(GEN6_UCGCTL1) |
 			   GEN6_GAMUNIT_CLOCK_GATE_DISABLE);
 
@@ -7251,6 +7245,10 @@ static void skl_init_clock_gating(struct drm_i915_private *dev_priv)
 {
 	gen9_init_clock_gating(dev_priv);
 
+	/* WaDisableDopClockGating:skl */
+	I915_WRITE(GEN7_MISCCPCTL, I915_READ(GEN7_MISCCPCTL) &
+		   ~GEN7_DOP_CLOCK_GATE_ENABLE);
+
 	/* WAC6entrylatency:skl */
 	I915_WRITE(FBC_LLC_READ_CTRL, I915_READ(FBC_LLC_READ_CTRL) |
 		   FBC_LLC_FULLY_OPEN);
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 916ccd1c0e96..5b3279262123 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -231,9 +231,21 @@ void vlv_ccu_write(struct drm_i915_private *i915, u32 reg, u32 val)
 			SB_CRWRDA_NP, reg, &val);
 }
 
+static u32 vlv_dpio_phy_iosf_port(struct drm_i915_private *i915, enum dpio_phy phy)
+{
+	/*
+	 * IOSF_PORT_DPIO: VLV x2 PHY (DP/HDMI B and C), CHV x1 PHY (DP/HDMI D)
+	 * IOSF_PORT_DPIO_2: CHV x2 PHY (DP/HDMI B and C)
+	 */
+	if (IS_CHERRYVIEW(i915))
+		return phy == DPIO_PHY0 ? IOSF_PORT_DPIO_2 : IOSF_PORT_DPIO;
+	else
+		return IOSF_PORT_DPIO;
+}
+
 u32 vlv_dpio_read(struct drm_i915_private *i915, enum pipe pipe, int reg)
 {
-	int port = i915->dpio_phy_iosf_port[DPIO_PHY(pipe)];
+	u32 port = vlv_dpio_phy_iosf_port(i915, DPIO_PHY(pipe));
 	u32 val = 0;
 
 	vlv_sideband_rw(i915, DPIO_DEVFN, port, SB_MRD_NP, reg, &val);
@@ -252,7 +264,7 @@ u32 vlv_dpio_read(struct drm_i915_private *i915, enum pipe pipe, int reg)
 void vlv_dpio_write(struct drm_i915_private *i915,
 		    enum pipe pipe, int reg, u32 val)
 {
-	int port = i915->dpio_phy_iosf_port[DPIO_PHY(pipe)];
+	u32 port = vlv_dpio_phy_iosf_port(i915, DPIO_PHY(pipe));
 
 	vlv_sideband_rw(i915, DPIO_DEVFN, port, SB_MWR_NP, reg, &val);
 }
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 8d5a933e6af6..263ffcb832b7 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1993,13 +1993,14 @@ int __intel_wait_for_register_fw(struct intel_uncore *uncore,
 				 unsigned int slow_timeout_ms,
 				 u32 *out_value)
 {
-	u32 reg_value;
+	u32 reg_value = 0;
 #define done (((reg_value = intel_uncore_read_fw(uncore, reg)) & mask) == value)
 	int ret;
 
 	/* Catch any overuse of this function */
 	might_sleep_if(slow_timeout_ms);
 	GEM_BUG_ON(fast_timeout_us > 20000);
+	GEM_BUG_ON(!fast_timeout_us && !slow_timeout_ms);
 
 	ret = -ETIMEDOUT;
 	if (fast_timeout_us && fast_timeout_us <= 20000)
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem.c b/drivers/gpu/drm/i915/selftests/i915_gem.c
index 88d400b9df88..23a6132c5f4e 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem.c
@@ -199,11 +199,52 @@ out:
 	return err;
 }
 
+static int igt_gem_ww_ctx(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct drm_i915_gem_object *obj, *obj2;
+	struct i915_gem_ww_ctx ww;
+	int err = 0;
+
+	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	if (IS_ERR(obj))
+		return PTR_ERR(obj);
+
+	obj2 = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	if (IS_ERR(obj)) {
+		err = PTR_ERR(obj);
+		goto put1;
+	}
+
+	i915_gem_ww_ctx_init(&ww, true);
+retry:
+	/* Lock the objects, twice for good measure (-EALREADY handling) */
+	err = i915_gem_object_lock(obj, &ww);
+	if (!err)
+		err = i915_gem_object_lock_interruptible(obj, &ww);
+	if (!err)
+		err = i915_gem_object_lock_interruptible(obj2, &ww);
+	if (!err)
+		err = i915_gem_object_lock(obj2, &ww);
+
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err)
+			goto retry;
+	}
+	i915_gem_ww_ctx_fini(&ww);
+	i915_gem_object_put(obj2);
+put1:
+	i915_gem_object_put(obj);
+	return err;
+}
+
 int i915_gem_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(igt_gem_suspend),
 		SUBTEST(igt_gem_hibernate),
+		SUBTEST(igt_gem_ww_ctx),
 	};
 
 	if (intel_gt_is_wedged(&i915->gt))
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
index 028baae9631f..f88473d396f4 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
@@ -536,7 +536,7 @@ int i915_gem_evict_mock_selftests(void)
 	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
 		err = i915_subtests(tests, &i915->gt);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 0016ffc7d914..c53a222e3dec 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -172,35 +172,45 @@ static int igt_ppgtt_alloc(void *arg)
 
 	/* Check we can allocate the entire range */
 	for (size = 4096; size <= limit; size <<= 2) {
-		err = ppgtt->vm.allocate_va_range(&ppgtt->vm, 0, size);
+		struct i915_vm_pt_stash stash = {};
+
+		err = i915_vm_alloc_pt_stash(&ppgtt->vm, &stash, size);
+		if (err)
+			goto err_ppgtt_cleanup;
+
+		err = i915_vm_pin_pt_stash(&ppgtt->vm, &stash);
 		if (err) {
-			if (err == -ENOMEM) {
-				pr_info("[1] Ran out of memory for va_range [0 + %llx] [bit %d]\n",
-					size, ilog2(size));
-				err = 0; /* virtual space too large! */
-			}
+			i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 			goto err_ppgtt_cleanup;
 		}
 
+		ppgtt->vm.allocate_va_range(&ppgtt->vm, &stash, 0, size);
 		cond_resched();
 
 		ppgtt->vm.clear_range(&ppgtt->vm, 0, size);
+
+		i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 	}
 
 	/* Check we can incrementally allocate the entire range */
 	for (last = 0, size = 4096; size <= limit; last = size, size <<= 2) {
-		err = ppgtt->vm.allocate_va_range(&ppgtt->vm,
-						  last, size - last);
+		struct i915_vm_pt_stash stash = {};
+
+		err = i915_vm_alloc_pt_stash(&ppgtt->vm, &stash, size - last);
+		if (err)
+			goto err_ppgtt_cleanup;
+
+		err = i915_vm_pin_pt_stash(&ppgtt->vm, &stash);
 		if (err) {
-			if (err == -ENOMEM) {
-				pr_info("[2] Ran out of memory for va_range [%llx + %llx] [bit %d]\n",
-					last, size - last, ilog2(size));
-				err = 0; /* virtual space too large! */
-			}
+			i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 			goto err_ppgtt_cleanup;
 		}
 
+		ppgtt->vm.allocate_va_range(&ppgtt->vm, &stash,
+					    last, size - last);
 		cond_resched();
+
+		i915_vm_free_pt_stash(&ppgtt->vm, &stash);
 	}
 
 err_ppgtt_cleanup:
@@ -284,9 +294,23 @@ static int lowlevel_hole(struct i915_address_space *vm,
 				break;
 			}
 
-			if (vm->allocate_va_range &&
-			    vm->allocate_va_range(vm, addr, BIT_ULL(size)))
-				break;
+			if (vm->allocate_va_range) {
+				struct i915_vm_pt_stash stash = {};
+
+				if (i915_vm_alloc_pt_stash(vm, &stash,
+							   BIT_ULL(size)))
+					break;
+
+				if (i915_vm_pin_pt_stash(vm, &stash)) {
+					i915_vm_free_pt_stash(vm, &stash);
+					break;
+				}
+
+				vm->allocate_va_range(vm, &stash,
+						      addr, BIT_ULL(size));
+
+				i915_vm_free_pt_stash(vm, &stash);
+			}
 
 			mock_vma->pages = obj->mm.pages;
 			mock_vma->node.size = BIT_ULL(size);
@@ -1703,7 +1727,7 @@ int i915_gem_gtt_mock_selftests(void)
 	mock_fini_ggtt(ggtt);
 	kfree(ggtt);
 out_put:
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
@@ -1881,6 +1905,7 @@ static int igt_cs_tlb(void *arg)
 			continue;
 
 		while (!__igt_timeout(end_time, NULL)) {
+			struct i915_vm_pt_stash stash = {};
 			struct i915_request *rq;
 			u64 offset;
 
@@ -1888,10 +1913,6 @@ static int igt_cs_tlb(void *arg)
 						   0, vm->total - PAGE_SIZE,
 						   chunk_size, PAGE_SIZE);
 
-			err = vm->allocate_va_range(vm, offset, chunk_size);
-			if (err)
-				goto end;
-
 			memset32(result, STACK_MAGIC, PAGE_SIZE / sizeof(u32));
 
 			vma = i915_vma_instance(bbe, vm, NULL);
@@ -1904,6 +1925,20 @@ static int igt_cs_tlb(void *arg)
 			if (err)
 				goto end;
 
+			err = i915_vm_alloc_pt_stash(vm, &stash, chunk_size);
+			if (err)
+				goto end;
+
+			err = i915_vm_pin_pt_stash(vm, &stash);
+			if (err) {
+				i915_vm_free_pt_stash(vm, &stash);
+				goto end;
+			}
+
+			vm->allocate_va_range(vm, &stash, offset, chunk_size);
+
+			i915_vm_free_pt_stash(vm, &stash);
+
 			/* Prime the TLB with the dummy pages */
 			for (i = 0; i < count; i++) {
 				vma->node.start = offset + i * PAGE_SIZE;
diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c
index c2d001d9c0ec..debbac660519 100644
--- a/drivers/gpu/drm/i915/selftests/i915_perf.c
+++ b/drivers/gpu/drm/i915/selftests/i915_perf.c
@@ -307,7 +307,7 @@ static int live_noa_gpr(void *arg)
 	}
 
 	/* Poison the ce->vm so we detect writes not to the GGTT gt->scratch */
-	scratch = kmap(ce->vm->scratch[0].base.page);
+	scratch = kmap(__px_page(ce->vm->scratch[0]));
 	memset(scratch, POISON_FREE, PAGE_SIZE);
 
 	rq = intel_context_create_request(ce);
@@ -405,7 +405,7 @@ static int live_noa_gpr(void *arg)
 out_rq:
 	i915_request_put(rq);
 out_ce:
-	kunmap(ce->vm->scratch[0].base.page);
+	kunmap(__px_page(ce->vm->scratch[0]));
 	intel_context_put(ce);
 out:
 	stream_destroy(stream);
diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c
index 57dd6f5122ee..64bbb8288249 100644
--- a/drivers/gpu/drm/i915/selftests/i915_request.c
+++ b/drivers/gpu/drm/i915/selftests/i915_request.c
@@ -331,7 +331,7 @@ static int __igt_breadcrumbs_smoketest(void *arg)
 		if (!wait) {
 			i915_sw_fence_commit(submit);
 			heap_fence_put(submit);
-			err = ENOMEM;
+			err = -ENOMEM;
 			break;
 		}
 
@@ -527,7 +527,7 @@ int i915_request_mock_selftests(void)
 	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
 		err = i915_subtests(tests, i915);
 
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 
 	return err;
 }
@@ -862,6 +862,8 @@ static int live_all_engines(void *arg)
 		goto out_free;
 	}
 
+	i915_vma_lock(batch);
+
 	idx = 0;
 	for_each_uabi_engine(engine, i915) {
 		request[idx] = intel_engine_create_kernel_request(engine);
@@ -872,11 +874,9 @@ static int live_all_engines(void *arg)
 			goto out_request;
 		}
 
-		i915_vma_lock(batch);
 		err = i915_request_await_object(request[idx], batch->obj, 0);
 		if (err == 0)
 			err = i915_vma_move_to_active(batch, request[idx], 0);
-		i915_vma_unlock(batch);
 		GEM_BUG_ON(err);
 
 		err = engine->emit_bb_start(request[idx],
@@ -891,6 +891,8 @@ static int live_all_engines(void *arg)
 		idx++;
 	}
 
+	i915_vma_unlock(batch);
+
 	idx = 0;
 	for_each_uabi_engine(engine, i915) {
 		if (i915_request_completed(request[idx])) {
@@ -981,12 +983,13 @@ static int live_sequential_engines(void *arg)
 			goto out_free;
 		}
 
+		i915_vma_lock(batch);
 		request[idx] = intel_engine_create_kernel_request(engine);
 		if (IS_ERR(request[idx])) {
 			err = PTR_ERR(request[idx]);
 			pr_err("%s: Request allocation failed for %s with err=%d\n",
 			       __func__, engine->name, err);
-			goto out_request;
+			goto out_unlock;
 		}
 
 		if (prev) {
@@ -996,16 +999,14 @@ static int live_sequential_engines(void *arg)
 				i915_request_add(request[idx]);
 				pr_err("%s: Request await failed for %s with err=%d\n",
 				       __func__, engine->name, err);
-				goto out_request;
+				goto out_unlock;
 			}
 		}
 
-		i915_vma_lock(batch);
 		err = i915_request_await_object(request[idx],
 						batch->obj, false);
 		if (err == 0)
 			err = i915_vma_move_to_active(batch, request[idx], 0);
-		i915_vma_unlock(batch);
 		GEM_BUG_ON(err);
 
 		err = engine->emit_bb_start(request[idx],
@@ -1020,6 +1021,11 @@ static int live_sequential_engines(void *arg)
 
 		prev = request[idx];
 		idx++;
+
+out_unlock:
+		i915_vma_unlock(batch);
+		if (err)
+			goto out_request;
 	}
 
 	idx = 0;
diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c
index af89c7fc8f59..1b6125e4c1ac 100644
--- a/drivers/gpu/drm/i915/selftests/i915_vma.c
+++ b/drivers/gpu/drm/i915/selftests/i915_vma.c
@@ -841,7 +841,7 @@ int i915_vma_mock_selftests(void)
 	mock_fini_ggtt(ggtt);
 	kfree(ggtt);
 out_put:
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
@@ -892,7 +892,7 @@ static int igt_vma_remapped_gtt(void *arg)
 			unsigned int x, y;
 			int err;
 
-			i915_gem_object_lock(obj);
+			i915_gem_object_lock(obj, NULL);
 			err = i915_gem_object_set_to_gtt_domain(obj, true);
 			i915_gem_object_unlock(obj);
 			if (err)
diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
index 6e80d99048e4..334b0648e253 100644
--- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
@@ -509,7 +509,7 @@ static int igt_lmem_write_cpu(void *arg)
 	if (err)
 		goto out_unpin;
 
-	i915_gem_object_lock(obj);
+	i915_gem_object_lock(obj, NULL);
 	err = i915_gem_object_set_to_wc_domain(obj, true);
 	i915_gem_object_unlock(obj);
 	if (err)
@@ -522,9 +522,9 @@ static int igt_lmem_write_cpu(void *arg)
 		goto out_unpin;
 	}
 
-	/* We want to throw in a random width/align */
-	bytes[0] = igt_random_offset(&prng, 0, PAGE_SIZE, sizeof(u32),
-				     sizeof(u32));
+	/* A random multiple of u32, picked between [64, PAGE_SIZE - 64] */
+	bytes[0] = igt_random_offset(&prng, 64, PAGE_SIZE - 64, 0, sizeof(u32));
+	GEM_BUG_ON(!IS_ALIGNED(bytes[0], sizeof(u32)));
 
 	i = 0;
 	do {
@@ -791,7 +791,7 @@ int intel_memory_region_mock_selftests(void)
 
 	intel_memory_region_put(mem);
 out_unref:
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
index 397c313a8b69..b6c42fd872ad 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
@@ -79,8 +79,6 @@ static void mock_device_release(struct drm_device *dev)
 
 out:
 	i915_params_free(&i915->params);
-	put_device(&i915->drm.pdev->dev);
-	i915->drm.pdev = NULL;
 }
 
 static struct drm_driver mock_driver = {
@@ -123,17 +121,10 @@ struct drm_i915_private *mock_gem_device(void)
 #endif
 	struct drm_i915_private *i915;
 	struct pci_dev *pdev;
-	int err;
 
 	pdev = kzalloc(sizeof(*pdev), GFP_KERNEL);
 	if (!pdev)
 		return NULL;
-	i915 = kzalloc(sizeof(*i915), GFP_KERNEL);
-	if (!i915) {
-		kfree(pdev);
-		return NULL;
-	}
-
 	device_initialize(&pdev->dev);
 	pdev->class = PCI_BASE_CLASS_DISPLAY << 16;
 	pdev->dev.release = release_dev;
@@ -144,8 +135,23 @@ struct drm_i915_private *mock_gem_device(void)
 	/* HACK to disable iommu for the fake device; force identity mapping */
 	pdev->dev.iommu = &fake_iommu;
 #endif
+	if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) {
+		put_device(&pdev->dev);
+		return NULL;
+	}
+
+	i915 = devm_drm_dev_alloc(&pdev->dev, &mock_driver,
+				  struct drm_i915_private, drm);
+	if (IS_ERR(i915)) {
+		pr_err("Failed to allocate mock GEM device: err=%ld\n", PTR_ERR(i915));
+		devres_release_group(&pdev->dev, NULL);
+		put_device(&pdev->dev);
+
+		return NULL;
+	}
 
 	pci_set_drvdata(pdev, i915);
+	i915->drm.pdev = pdev;
 
 	dev_pm_domain_set(&pdev->dev, &pm_domain);
 	pm_runtime_enable(&pdev->dev);
@@ -153,16 +159,6 @@ struct drm_i915_private *mock_gem_device(void)
 	if (pm_runtime_enabled(&pdev->dev))
 		WARN_ON(pm_runtime_get_sync(&pdev->dev));
 
-	err = drm_dev_init(&i915->drm, &mock_driver, &pdev->dev);
-	if (err) {
-		pr_err("Failed to initialise mock GEM device: err=%d\n", err);
-		put_device(&pdev->dev);
-		kfree(i915);
-
-		return NULL;
-	}
-	i915->drm.pdev = pdev;
-	drmm_add_final_kfree(&i915->drm, i915);
 
 	i915_params_copy(&i915->params, &i915_modparams);
 
@@ -222,7 +218,15 @@ err_drv:
 	intel_gt_driver_late_release(&i915->gt);
 	intel_memory_regions_driver_release(i915);
 	drm_mode_config_cleanup(&i915->drm);
-	drm_dev_put(&i915->drm);
+	mock_destroy_device(i915);
 
 	return NULL;
 }
+
+void mock_destroy_device(struct drm_i915_private *i915)
+{
+	struct device *dev = i915->drm.dev;
+
+	devres_release_group(dev, NULL);
+	put_device(dev);
+}
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.h b/drivers/gpu/drm/i915/selftests/mock_gem_device.h
index b5dc4e394555..953cfe4fab34 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.h
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.h
@@ -7,4 +7,6 @@ struct drm_i915_private;
 struct drm_i915_private *mock_gem_device(void);
 void mock_device_flush(struct drm_i915_private *i915);
 
+void mock_destroy_device(struct drm_i915_private *i915);
+
 #endif /* !__MOCK_GEM_DEVICE_H__ */
diff --git a/drivers/gpu/drm/i915/selftests/mock_gtt.c b/drivers/gpu/drm/i915/selftests/mock_gtt.c
index b173086411ef..7270fc8ca801 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gtt.c
@@ -38,14 +38,14 @@ static void mock_insert_entries(struct i915_address_space *vm,
 {
 }
 
-static int mock_bind_ppgtt(struct i915_address_space *vm,
-			   struct i915_vma *vma,
-			   enum i915_cache_level cache_level,
-			   u32 flags)
+static void mock_bind_ppgtt(struct i915_address_space *vm,
+			    struct i915_vm_pt_stash *stash,
+			    struct i915_vma *vma,
+			    enum i915_cache_level cache_level,
+			    u32 flags)
 {
 	GEM_BUG_ON(flags & I915_VMA_GLOBAL_BIND);
 	set_bit(I915_VMA_LOCAL_BIND_BIT, __i915_vma_flags(vma));
-	return 0;
 }
 
 static void mock_unbind_ppgtt(struct i915_address_space *vm,
@@ -74,9 +74,12 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name)
 	ppgtt->vm.i915 = i915;
 	ppgtt->vm.total = round_down(U64_MAX, PAGE_SIZE);
 	ppgtt->vm.file = ERR_PTR(-ENODEV);
+	ppgtt->vm.dma = &i915->drm.pdev->dev;
 
 	i915_address_space_init(&ppgtt->vm, VM_CLASS_PPGTT);
 
+	ppgtt->vm.alloc_pt_dma = alloc_pt_dma;
+
 	ppgtt->vm.clear_range = mock_clear_range;
 	ppgtt->vm.insert_page = mock_insert_page;
 	ppgtt->vm.insert_entries = mock_insert_entries;
@@ -90,13 +93,12 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name)
 	return ppgtt;
 }
 
-static int mock_bind_ggtt(struct i915_address_space *vm,
-			  struct i915_vma *vma,
-			  enum i915_cache_level cache_level,
-			  u32 flags)
+static void mock_bind_ggtt(struct i915_address_space *vm,
+			   struct i915_vm_pt_stash *stash,
+			   struct i915_vma *vma,
+			   enum i915_cache_level cache_level,
+			   u32 flags)
 {
-	atomic_or(I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND, &vma->flags);
-	return 0;
 }
 
 static void mock_unbind_ggtt(struct i915_address_space *vm,
@@ -116,6 +118,8 @@ void mock_init_ggtt(struct drm_i915_private *i915, struct i915_ggtt *ggtt)
 	ggtt->mappable_end = resource_size(&ggtt->gmadr);
 	ggtt->vm.total = 4096 * PAGE_SIZE;
 
+	ggtt->vm.alloc_pt_dma = alloc_pt_dma;
+
 	ggtt->vm.clear_range = mock_clear_range;
 	ggtt->vm.insert_page = mock_insert_page;
 	ggtt->vm.insert_entries = mock_insert_entries;
diff --git a/drivers/gpu/drm/imx/Kconfig b/drivers/gpu/drm/imx/Kconfig
index 207bf7409dfb..6231048aa5aa 100644
--- a/drivers/gpu/drm/imx/Kconfig
+++ b/drivers/gpu/drm/imx/Kconfig
@@ -39,3 +39,5 @@ config DRM_IMX_HDMI
 	depends on DRM_IMX
 	help
 	  Choose this if you want to use HDMI on i.MX6.
+
+source "drivers/gpu/drm/imx/dcss/Kconfig"
diff --git a/drivers/gpu/drm/imx/Makefile b/drivers/gpu/drm/imx/Makefile
index 21cdcc2faabc..b644deffe948 100644
--- a/drivers/gpu/drm/imx/Makefile
+++ b/drivers/gpu/drm/imx/Makefile
@@ -9,3 +9,4 @@ obj-$(CONFIG_DRM_IMX_TVE) += imx-tve.o
 obj-$(CONFIG_DRM_IMX_LDB) += imx-ldb.o
 
 obj-$(CONFIG_DRM_IMX_HDMI) += dw_hdmi-imx.o
+obj-$(CONFIG_DRM_IMX_DCSS) += dcss/
diff --git a/drivers/gpu/drm/imx/dcss/Kconfig b/drivers/gpu/drm/imx/dcss/Kconfig
new file mode 100644
index 000000000000..2b17a964ff05
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/Kconfig
@@ -0,0 +1,9 @@
+config DRM_IMX_DCSS
+	tristate "i.MX8MQ DCSS"
+	select IMX_IRQSTEER
+	select DRM_KMS_CMA_HELPER
+	select VIDEOMODE_HELPERS
+	depends on DRM && ARCH_MXC && ARM64
+	help
+	  Choose this if you have a NXP i.MX8MQ based system and want to use the
+	  Display Controller Subsystem. This option enables DCSS support.
diff --git a/drivers/gpu/drm/imx/dcss/Makefile b/drivers/gpu/drm/imx/dcss/Makefile
new file mode 100644
index 000000000000..8c7c8da42792
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/Makefile
@@ -0,0 +1,6 @@
+imx-dcss-objs := dcss-drv.o dcss-dev.o dcss-blkctl.o dcss-ctxld.o dcss-dtg.o \
+				 dcss-ss.o dcss-dpr.o dcss-scaler.o dcss-kms.o dcss-crtc.o \
+				 dcss-plane.o
+
+obj-$(CONFIG_DRM_IMX_DCSS) += imx-dcss.o
+
diff --git a/drivers/gpu/drm/imx/dcss/dcss-blkctl.c b/drivers/gpu/drm/imx/dcss/dcss-blkctl.c
new file mode 100644
index 000000000000..c9b54bb2692d
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-blkctl.c
@@ -0,0 +1,70 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/device.h>
+#include <linux/of.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_BLKCTL_RESET_CTRL		0x00
+#define   B_CLK_RESETN			BIT(0)
+#define   APB_CLK_RESETN		BIT(1)
+#define   P_CLK_RESETN			BIT(2)
+#define   RTR_CLK_RESETN		BIT(4)
+#define DCSS_BLKCTL_CONTROL0		0x10
+#define   HDMI_MIPI_CLK_SEL		BIT(0)
+#define   DISPMIX_REFCLK_SEL_POS	4
+#define   DISPMIX_REFCLK_SEL_MASK	GENMASK(5, 4)
+#define   DISPMIX_PIXCLK_SEL		BIT(8)
+#define   HDMI_SRC_SECURE_EN		BIT(16)
+
+struct dcss_blkctl {
+	struct dcss_dev *dcss;
+	void __iomem *base_reg;
+};
+
+void dcss_blkctl_cfg(struct dcss_blkctl *blkctl)
+{
+	if (blkctl->dcss->hdmi_output)
+		dcss_writel(0, blkctl->base_reg + DCSS_BLKCTL_CONTROL0);
+	else
+		dcss_writel(DISPMIX_PIXCLK_SEL,
+			    blkctl->base_reg + DCSS_BLKCTL_CONTROL0);
+
+	dcss_set(B_CLK_RESETN | APB_CLK_RESETN | P_CLK_RESETN | RTR_CLK_RESETN,
+		 blkctl->base_reg + DCSS_BLKCTL_RESET_CTRL);
+}
+
+int dcss_blkctl_init(struct dcss_dev *dcss, unsigned long blkctl_base)
+{
+	struct dcss_blkctl *blkctl;
+
+	blkctl = kzalloc(sizeof(*blkctl), GFP_KERNEL);
+	if (!blkctl)
+		return -ENOMEM;
+
+	blkctl->base_reg = ioremap(blkctl_base, SZ_4K);
+	if (!blkctl->base_reg) {
+		dev_err(dcss->dev, "unable to remap BLK CTRL base\n");
+		kfree(blkctl);
+		return -ENOMEM;
+	}
+
+	dcss->blkctl = blkctl;
+	blkctl->dcss = dcss;
+
+	dcss_blkctl_cfg(blkctl);
+
+	return 0;
+}
+
+void dcss_blkctl_exit(struct dcss_blkctl *blkctl)
+{
+	if (blkctl->base_reg)
+		iounmap(blkctl->base_reg);
+
+	kfree(blkctl);
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-crtc.c b/drivers/gpu/drm/imx/dcss/dcss-crtc.c
new file mode 100644
index 000000000000..36abff0890b2
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-crtc.c
@@ -0,0 +1,219 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_vblank.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+
+#include "dcss-dev.h"
+#include "dcss-kms.h"
+
+static int dcss_enable_vblank(struct drm_crtc *crtc)
+{
+	struct dcss_crtc *dcss_crtc = container_of(crtc, struct dcss_crtc,
+						   base);
+	struct dcss_dev *dcss = crtc->dev->dev_private;
+
+	dcss_dtg_vblank_irq_enable(dcss->dtg, true);
+
+	dcss_dtg_ctxld_kick_irq_enable(dcss->dtg, true);
+
+	enable_irq(dcss_crtc->irq);
+
+	return 0;
+}
+
+static void dcss_disable_vblank(struct drm_crtc *crtc)
+{
+	struct dcss_crtc *dcss_crtc = container_of(crtc, struct dcss_crtc,
+						   base);
+	struct dcss_dev *dcss = dcss_crtc->base.dev->dev_private;
+
+	disable_irq_nosync(dcss_crtc->irq);
+
+	dcss_dtg_vblank_irq_enable(dcss->dtg, false);
+
+	if (dcss_crtc->disable_ctxld_kick_irq)
+		dcss_dtg_ctxld_kick_irq_enable(dcss->dtg, false);
+}
+
+static const struct drm_crtc_funcs dcss_crtc_funcs = {
+	.set_config = drm_atomic_helper_set_config,
+	.destroy = drm_crtc_cleanup,
+	.page_flip = drm_atomic_helper_page_flip,
+	.reset = drm_atomic_helper_crtc_reset,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+	.enable_vblank = dcss_enable_vblank,
+	.disable_vblank = dcss_disable_vblank,
+};
+
+static void dcss_crtc_atomic_begin(struct drm_crtc *crtc,
+				   struct drm_crtc_state *old_crtc_state)
+{
+	drm_crtc_vblank_on(crtc);
+}
+
+static void dcss_crtc_atomic_flush(struct drm_crtc *crtc,
+				   struct drm_crtc_state *old_crtc_state)
+{
+	struct dcss_crtc *dcss_crtc = container_of(crtc, struct dcss_crtc,
+						   base);
+	struct dcss_dev *dcss = dcss_crtc->base.dev->dev_private;
+
+	spin_lock_irq(&crtc->dev->event_lock);
+	if (crtc->state->event) {
+		WARN_ON(drm_crtc_vblank_get(crtc));
+		drm_crtc_arm_vblank_event(crtc, crtc->state->event);
+		crtc->state->event = NULL;
+	}
+	spin_unlock_irq(&crtc->dev->event_lock);
+
+	if (dcss_dtg_is_enabled(dcss->dtg))
+		dcss_ctxld_enable(dcss->ctxld);
+}
+
+static void dcss_crtc_atomic_enable(struct drm_crtc *crtc,
+				    struct drm_crtc_state *old_crtc_state)
+{
+	struct dcss_crtc *dcss_crtc = container_of(crtc, struct dcss_crtc,
+						   base);
+	struct dcss_dev *dcss = dcss_crtc->base.dev->dev_private;
+	struct drm_display_mode *mode = &crtc->state->adjusted_mode;
+	struct drm_display_mode *old_mode = &old_crtc_state->adjusted_mode;
+	struct videomode vm;
+
+	drm_display_mode_to_videomode(mode, &vm);
+
+	pm_runtime_get_sync(dcss->dev);
+
+	vm.pixelclock = mode->crtc_clock * 1000;
+
+	dcss_ss_subsam_set(dcss->ss);
+	dcss_dtg_css_set(dcss->dtg);
+
+	if (!drm_mode_equal(mode, old_mode) || !old_crtc_state->active) {
+		dcss_dtg_sync_set(dcss->dtg, &vm);
+		dcss_ss_sync_set(dcss->ss, &vm,
+				 mode->flags & DRM_MODE_FLAG_PHSYNC,
+				 mode->flags & DRM_MODE_FLAG_PVSYNC);
+	}
+
+	dcss_enable_dtg_and_ss(dcss);
+
+	dcss_ctxld_enable(dcss->ctxld);
+
+	/* Allow CTXLD kick interrupt to be disabled when VBLANK is disabled. */
+	dcss_crtc->disable_ctxld_kick_irq = true;
+}
+
+static void dcss_crtc_atomic_disable(struct drm_crtc *crtc,
+				     struct drm_crtc_state *old_crtc_state)
+{
+	struct dcss_crtc *dcss_crtc = container_of(crtc, struct dcss_crtc,
+						   base);
+	struct dcss_dev *dcss = dcss_crtc->base.dev->dev_private;
+	struct drm_display_mode *mode = &crtc->state->adjusted_mode;
+	struct drm_display_mode *old_mode = &old_crtc_state->adjusted_mode;
+
+	drm_atomic_helper_disable_planes_on_crtc(old_crtc_state, false);
+
+	spin_lock_irq(&crtc->dev->event_lock);
+	if (crtc->state->event) {
+		drm_crtc_send_vblank_event(crtc, crtc->state->event);
+		crtc->state->event = NULL;
+	}
+	spin_unlock_irq(&crtc->dev->event_lock);
+
+	dcss_dtg_ctxld_kick_irq_enable(dcss->dtg, true);
+
+	reinit_completion(&dcss->disable_completion);
+
+	dcss_disable_dtg_and_ss(dcss);
+
+	dcss_ctxld_enable(dcss->ctxld);
+
+	if (!drm_mode_equal(mode, old_mode) || !crtc->state->active)
+		if (!wait_for_completion_timeout(&dcss->disable_completion,
+						 msecs_to_jiffies(100)))
+			dev_err(dcss->dev, "Shutting off DTG timed out.\n");
+
+	/*
+	 * Do not shut off CTXLD kick interrupt when shutting VBLANK off. It
+	 * will be needed to commit the last changes, before going to suspend.
+	 */
+	dcss_crtc->disable_ctxld_kick_irq = false;
+
+	drm_crtc_vblank_off(crtc);
+
+	pm_runtime_mark_last_busy(dcss->dev);
+	pm_runtime_put_autosuspend(dcss->dev);
+}
+
+static const struct drm_crtc_helper_funcs dcss_helper_funcs = {
+	.atomic_begin = dcss_crtc_atomic_begin,
+	.atomic_flush = dcss_crtc_atomic_flush,
+	.atomic_enable = dcss_crtc_atomic_enable,
+	.atomic_disable = dcss_crtc_atomic_disable,
+};
+
+static irqreturn_t dcss_crtc_irq_handler(int irq, void *dev_id)
+{
+	struct dcss_crtc *dcss_crtc = dev_id;
+	struct dcss_dev *dcss = dcss_crtc->base.dev->dev_private;
+
+	if (!dcss_dtg_vblank_irq_valid(dcss->dtg))
+		return IRQ_NONE;
+
+	if (dcss_ctxld_is_flushed(dcss->ctxld))
+		drm_crtc_handle_vblank(&dcss_crtc->base);
+
+	dcss_dtg_vblank_irq_clear(dcss->dtg);
+
+	return IRQ_HANDLED;
+}
+
+int dcss_crtc_init(struct dcss_crtc *crtc, struct drm_device *drm)
+{
+	struct dcss_dev *dcss = drm->dev_private;
+	struct platform_device *pdev = to_platform_device(dcss->dev);
+	int ret;
+
+	crtc->plane[0] = dcss_plane_init(drm, drm_crtc_mask(&crtc->base),
+					 DRM_PLANE_TYPE_PRIMARY, 0);
+	if (IS_ERR(crtc->plane[0]))
+		return PTR_ERR(crtc->plane[0]);
+
+	crtc->base.port = dcss->of_port;
+
+	drm_crtc_helper_add(&crtc->base, &dcss_helper_funcs);
+	ret = drm_crtc_init_with_planes(drm, &crtc->base, &crtc->plane[0]->base,
+					NULL, &dcss_crtc_funcs, NULL);
+	if (ret) {
+		dev_err(dcss->dev, "failed to init crtc\n");
+		return ret;
+	}
+
+	crtc->irq = platform_get_irq_byname(pdev, "vblank");
+	if (crtc->irq < 0)
+		return crtc->irq;
+
+	ret = request_irq(crtc->irq, dcss_crtc_irq_handler,
+			  0, "dcss_drm", crtc);
+	if (ret) {
+		dev_err(dcss->dev, "irq request failed with %d.\n", ret);
+		return ret;
+	}
+
+	disable_irq(crtc->irq);
+
+	return 0;
+}
+
+void dcss_crtc_deinit(struct dcss_crtc *crtc, struct drm_device *drm)
+{
+	free_irq(crtc->irq, crtc);
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-ctxld.c b/drivers/gpu/drm/imx/dcss/dcss-ctxld.c
new file mode 100644
index 000000000000..3a84cb3209c4
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-ctxld.c
@@ -0,0 +1,424 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_CTXLD_CONTROL_STATUS	0x0
+#define   CTXLD_ENABLE			BIT(0)
+#define   ARB_SEL			BIT(1)
+#define   RD_ERR_EN			BIT(2)
+#define   DB_COMP_EN			BIT(3)
+#define   SB_HP_COMP_EN			BIT(4)
+#define   SB_LP_COMP_EN			BIT(5)
+#define   DB_PEND_SB_REC_EN		BIT(6)
+#define   SB_PEND_DISP_ACTIVE_EN	BIT(7)
+#define   AHB_ERR_EN			BIT(8)
+#define   RD_ERR			BIT(16)
+#define   DB_COMP			BIT(17)
+#define   SB_HP_COMP			BIT(18)
+#define   SB_LP_COMP			BIT(19)
+#define   DB_PEND_SB_REC		BIT(20)
+#define   SB_PEND_DISP_ACTIVE		BIT(21)
+#define   AHB_ERR			BIT(22)
+#define DCSS_CTXLD_DB_BASE_ADDR		0x10
+#define DCSS_CTXLD_DB_COUNT		0x14
+#define DCSS_CTXLD_SB_BASE_ADDR		0x18
+#define DCSS_CTXLD_SB_COUNT		0x1C
+#define   SB_HP_COUNT_POS		0
+#define   SB_HP_COUNT_MASK		0xffff
+#define   SB_LP_COUNT_POS		16
+#define   SB_LP_COUNT_MASK		0xffff0000
+#define DCSS_AHB_ERR_ADDR		0x20
+
+#define CTXLD_IRQ_COMPLETION		(DB_COMP | SB_HP_COMP | SB_LP_COMP)
+#define CTXLD_IRQ_ERROR			(RD_ERR | DB_PEND_SB_REC | AHB_ERR)
+
+/* The following sizes are in context loader entries, 8 bytes each. */
+#define CTXLD_DB_CTX_ENTRIES		1024	/* max 65536 */
+#define CTXLD_SB_LP_CTX_ENTRIES		10240	/* max 65536 */
+#define CTXLD_SB_HP_CTX_ENTRIES		20000	/* max 65536 */
+#define CTXLD_SB_CTX_ENTRIES		(CTXLD_SB_LP_CTX_ENTRIES + \
+					 CTXLD_SB_HP_CTX_ENTRIES)
+
+/* Sizes, in entries, of the DB, SB_HP and SB_LP context regions. */
+static u16 dcss_ctxld_ctx_size[3] = {
+	CTXLD_DB_CTX_ENTRIES,
+	CTXLD_SB_HP_CTX_ENTRIES,
+	CTXLD_SB_LP_CTX_ENTRIES
+};
+
+/* this represents an entry in the context loader map */
+struct dcss_ctxld_item {
+	u32 val;
+	u32 ofs;
+};
+
+#define CTX_ITEM_SIZE			sizeof(struct dcss_ctxld_item)
+
+struct dcss_ctxld {
+	struct device *dev;
+	void __iomem *ctxld_reg;
+	int irq;
+	bool irq_en;
+
+	struct dcss_ctxld_item *db[2];
+	struct dcss_ctxld_item *sb_hp[2];
+	struct dcss_ctxld_item *sb_lp[2];
+
+	dma_addr_t db_paddr[2];
+	dma_addr_t sb_paddr[2];
+
+	u16 ctx_size[2][3]; /* holds the sizes of DB, SB_HP and SB_LP ctx */
+	u8 current_ctx;
+
+	bool in_use;
+	bool armed;
+
+	spinlock_t lock; /* protects concurent access to private data */
+};
+
+static irqreturn_t dcss_ctxld_irq_handler(int irq, void *data)
+{
+	struct dcss_ctxld *ctxld = data;
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(ctxld->dev);
+	u32 irq_status;
+
+	irq_status = dcss_readl(ctxld->ctxld_reg + DCSS_CTXLD_CONTROL_STATUS);
+
+	if (irq_status & CTXLD_IRQ_COMPLETION &&
+	    !(irq_status & CTXLD_ENABLE) && ctxld->in_use) {
+		ctxld->in_use = false;
+
+		if (dcss && dcss->disable_callback)
+			dcss->disable_callback(dcss);
+	} else if (irq_status & CTXLD_IRQ_ERROR) {
+		/*
+		 * Except for throwing an error message and clearing the status
+		 * register, there's not much we can do here.
+		 */
+		dev_err(ctxld->dev, "ctxld: error encountered: %08x\n",
+			irq_status);
+		dev_err(ctxld->dev, "ctxld: db=%d, sb_hp=%d, sb_lp=%d\n",
+			ctxld->ctx_size[ctxld->current_ctx ^ 1][CTX_DB],
+			ctxld->ctx_size[ctxld->current_ctx ^ 1][CTX_SB_HP],
+			ctxld->ctx_size[ctxld->current_ctx ^ 1][CTX_SB_LP]);
+	}
+
+	dcss_clr(irq_status & (CTXLD_IRQ_ERROR | CTXLD_IRQ_COMPLETION),
+		 ctxld->ctxld_reg + DCSS_CTXLD_CONTROL_STATUS);
+
+	return IRQ_HANDLED;
+}
+
+static int dcss_ctxld_irq_config(struct dcss_ctxld *ctxld,
+				 struct platform_device *pdev)
+{
+	int ret;
+
+	ctxld->irq = platform_get_irq_byname(pdev, "ctxld");
+	if (ctxld->irq < 0)
+		return ctxld->irq;
+
+	ret = request_irq(ctxld->irq, dcss_ctxld_irq_handler,
+			  0, "dcss_ctxld", ctxld);
+	if (ret) {
+		dev_err(ctxld->dev, "ctxld: irq request failed.\n");
+		return ret;
+	}
+
+	ctxld->irq_en = true;
+
+	return 0;
+}
+
+static void dcss_ctxld_hw_cfg(struct dcss_ctxld *ctxld)
+{
+	dcss_writel(RD_ERR_EN | SB_HP_COMP_EN |
+		    DB_PEND_SB_REC_EN | AHB_ERR_EN | RD_ERR | AHB_ERR,
+		    ctxld->ctxld_reg + DCSS_CTXLD_CONTROL_STATUS);
+}
+
+static void dcss_ctxld_free_ctx(struct dcss_ctxld *ctxld)
+{
+	struct dcss_ctxld_item *ctx;
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		if (ctxld->db[i]) {
+			dma_free_coherent(ctxld->dev,
+					  CTXLD_DB_CTX_ENTRIES * sizeof(*ctx),
+					  ctxld->db[i], ctxld->db_paddr[i]);
+			ctxld->db[i] = NULL;
+			ctxld->db_paddr[i] = 0;
+		}
+
+		if (ctxld->sb_hp[i]) {
+			dma_free_coherent(ctxld->dev,
+					  CTXLD_SB_CTX_ENTRIES * sizeof(*ctx),
+					  ctxld->sb_hp[i], ctxld->sb_paddr[i]);
+			ctxld->sb_hp[i] = NULL;
+			ctxld->sb_paddr[i] = 0;
+		}
+	}
+}
+
+static int dcss_ctxld_alloc_ctx(struct dcss_ctxld *ctxld)
+{
+	struct dcss_ctxld_item *ctx;
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		ctx = dma_alloc_coherent(ctxld->dev,
+					 CTXLD_DB_CTX_ENTRIES * sizeof(*ctx),
+					 &ctxld->db_paddr[i], GFP_KERNEL);
+		if (!ctx)
+			return -ENOMEM;
+
+		ctxld->db[i] = ctx;
+
+		ctx = dma_alloc_coherent(ctxld->dev,
+					 CTXLD_SB_CTX_ENTRIES * sizeof(*ctx),
+					 &ctxld->sb_paddr[i], GFP_KERNEL);
+		if (!ctx)
+			return -ENOMEM;
+
+		ctxld->sb_hp[i] = ctx;
+		ctxld->sb_lp[i] = ctx + CTXLD_SB_HP_CTX_ENTRIES;
+	}
+
+	return 0;
+}
+
+int dcss_ctxld_init(struct dcss_dev *dcss, unsigned long ctxld_base)
+{
+	struct dcss_ctxld *ctxld;
+	int ret;
+
+	ctxld = kzalloc(sizeof(*ctxld), GFP_KERNEL);
+	if (!ctxld)
+		return -ENOMEM;
+
+	dcss->ctxld = ctxld;
+	ctxld->dev = dcss->dev;
+
+	spin_lock_init(&ctxld->lock);
+
+	ret = dcss_ctxld_alloc_ctx(ctxld);
+	if (ret) {
+		dev_err(dcss->dev, "ctxld: cannot allocate context memory.\n");
+		goto err;
+	}
+
+	ctxld->ctxld_reg = ioremap(ctxld_base, SZ_4K);
+	if (!ctxld->ctxld_reg) {
+		dev_err(dcss->dev, "ctxld: unable to remap ctxld base\n");
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	ret = dcss_ctxld_irq_config(ctxld, to_platform_device(dcss->dev));
+	if (ret)
+		goto err_irq;
+
+	dcss_ctxld_hw_cfg(ctxld);
+
+	return 0;
+
+err_irq:
+	iounmap(ctxld->ctxld_reg);
+
+err:
+	dcss_ctxld_free_ctx(ctxld);
+	kfree(ctxld);
+
+	return ret;
+}
+
+void dcss_ctxld_exit(struct dcss_ctxld *ctxld)
+{
+	free_irq(ctxld->irq, ctxld);
+
+	if (ctxld->ctxld_reg)
+		iounmap(ctxld->ctxld_reg);
+
+	dcss_ctxld_free_ctx(ctxld);
+	kfree(ctxld);
+}
+
+static int dcss_ctxld_enable_locked(struct dcss_ctxld *ctxld)
+{
+	int curr_ctx = ctxld->current_ctx;
+	u32 db_base, sb_base, sb_count;
+	u32 sb_hp_cnt, sb_lp_cnt, db_cnt;
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(ctxld->dev);
+
+	if (!dcss)
+		return 0;
+
+	dcss_dpr_write_sysctrl(dcss->dpr);
+
+	dcss_scaler_write_sclctrl(dcss->scaler);
+
+	sb_hp_cnt = ctxld->ctx_size[curr_ctx][CTX_SB_HP];
+	sb_lp_cnt = ctxld->ctx_size[curr_ctx][CTX_SB_LP];
+	db_cnt = ctxld->ctx_size[curr_ctx][CTX_DB];
+
+	/* make sure SB_LP context area comes after SB_HP */
+	if (sb_lp_cnt &&
+	    ctxld->sb_lp[curr_ctx] != ctxld->sb_hp[curr_ctx] + sb_hp_cnt) {
+		struct dcss_ctxld_item *sb_lp_adjusted;
+
+		sb_lp_adjusted = ctxld->sb_hp[curr_ctx] + sb_hp_cnt;
+
+		memcpy(sb_lp_adjusted, ctxld->sb_lp[curr_ctx],
+		       sb_lp_cnt * CTX_ITEM_SIZE);
+	}
+
+	db_base = db_cnt ? ctxld->db_paddr[curr_ctx] : 0;
+
+	dcss_writel(db_base, ctxld->ctxld_reg + DCSS_CTXLD_DB_BASE_ADDR);
+	dcss_writel(db_cnt, ctxld->ctxld_reg + DCSS_CTXLD_DB_COUNT);
+
+	if (sb_hp_cnt)
+		sb_count = ((sb_hp_cnt << SB_HP_COUNT_POS) & SB_HP_COUNT_MASK) |
+			   ((sb_lp_cnt << SB_LP_COUNT_POS) & SB_LP_COUNT_MASK);
+	else
+		sb_count = (sb_lp_cnt << SB_HP_COUNT_POS) & SB_HP_COUNT_MASK;
+
+	sb_base = sb_count ? ctxld->sb_paddr[curr_ctx] : 0;
+
+	dcss_writel(sb_base, ctxld->ctxld_reg + DCSS_CTXLD_SB_BASE_ADDR);
+	dcss_writel(sb_count, ctxld->ctxld_reg + DCSS_CTXLD_SB_COUNT);
+
+	/* enable the context loader */
+	dcss_set(CTXLD_ENABLE, ctxld->ctxld_reg + DCSS_CTXLD_CONTROL_STATUS);
+
+	ctxld->in_use = true;
+
+	/*
+	 * Toggle the current context to the alternate one so that any updates
+	 * in the modules' settings take place there.
+	 */
+	ctxld->current_ctx ^= 1;
+
+	ctxld->ctx_size[ctxld->current_ctx][CTX_DB] = 0;
+	ctxld->ctx_size[ctxld->current_ctx][CTX_SB_HP] = 0;
+	ctxld->ctx_size[ctxld->current_ctx][CTX_SB_LP] = 0;
+
+	return 0;
+}
+
+int dcss_ctxld_enable(struct dcss_ctxld *ctxld)
+{
+	spin_lock_irq(&ctxld->lock);
+	ctxld->armed = true;
+	spin_unlock_irq(&ctxld->lock);
+
+	return 0;
+}
+
+void dcss_ctxld_kick(struct dcss_ctxld *ctxld)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&ctxld->lock, flags);
+	if (ctxld->armed && !ctxld->in_use) {
+		ctxld->armed = false;
+		dcss_ctxld_enable_locked(ctxld);
+	}
+	spin_unlock_irqrestore(&ctxld->lock, flags);
+}
+
+void dcss_ctxld_write_irqsafe(struct dcss_ctxld *ctxld, u32 ctx_id, u32 val,
+			      u32 reg_ofs)
+{
+	int curr_ctx = ctxld->current_ctx;
+	struct dcss_ctxld_item *ctx[] = {
+		[CTX_DB] = ctxld->db[curr_ctx],
+		[CTX_SB_HP] = ctxld->sb_hp[curr_ctx],
+		[CTX_SB_LP] = ctxld->sb_lp[curr_ctx]
+	};
+	int item_idx = ctxld->ctx_size[curr_ctx][ctx_id];
+
+	if (item_idx + 1 > dcss_ctxld_ctx_size[ctx_id]) {
+		WARN_ON(1);
+		return;
+	}
+
+	ctx[ctx_id][item_idx].val = val;
+	ctx[ctx_id][item_idx].ofs = reg_ofs;
+	ctxld->ctx_size[curr_ctx][ctx_id] += 1;
+}
+
+void dcss_ctxld_write(struct dcss_ctxld *ctxld, u32 ctx_id,
+		      u32 val, u32 reg_ofs)
+{
+	spin_lock_irq(&ctxld->lock);
+	dcss_ctxld_write_irqsafe(ctxld, ctx_id, val, reg_ofs);
+	spin_unlock_irq(&ctxld->lock);
+}
+
+bool dcss_ctxld_is_flushed(struct dcss_ctxld *ctxld)
+{
+	return ctxld->ctx_size[ctxld->current_ctx][CTX_DB] == 0 &&
+		ctxld->ctx_size[ctxld->current_ctx][CTX_SB_HP] == 0 &&
+		ctxld->ctx_size[ctxld->current_ctx][CTX_SB_LP] == 0;
+}
+
+int dcss_ctxld_resume(struct dcss_ctxld *ctxld)
+{
+	dcss_ctxld_hw_cfg(ctxld);
+
+	if (!ctxld->irq_en) {
+		enable_irq(ctxld->irq);
+		ctxld->irq_en = true;
+	}
+
+	return 0;
+}
+
+int dcss_ctxld_suspend(struct dcss_ctxld *ctxld)
+{
+	int ret = 0;
+	unsigned long timeout = jiffies + msecs_to_jiffies(500);
+
+	if (!dcss_ctxld_is_flushed(ctxld)) {
+		dcss_ctxld_kick(ctxld);
+
+		while (!time_after(jiffies, timeout) && ctxld->in_use)
+			msleep(20);
+
+		if (time_after(jiffies, timeout))
+			return -ETIMEDOUT;
+	}
+
+	spin_lock_irq(&ctxld->lock);
+
+	if (ctxld->irq_en) {
+		disable_irq_nosync(ctxld->irq);
+		ctxld->irq_en = false;
+	}
+
+	/* reset context region and sizes */
+	ctxld->current_ctx = 0;
+	ctxld->ctx_size[0][CTX_DB] = 0;
+	ctxld->ctx_size[0][CTX_SB_HP] = 0;
+	ctxld->ctx_size[0][CTX_SB_LP] = 0;
+
+	spin_unlock_irq(&ctxld->lock);
+
+	return ret;
+}
+
+void dcss_ctxld_assert_locked(struct dcss_ctxld *ctxld)
+{
+	lockdep_assert_held(&ctxld->lock);
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-dev.c b/drivers/gpu/drm/imx/dcss/dcss-dev.c
new file mode 100644
index 000000000000..c849533ca83e
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-dev.c
@@ -0,0 +1,325 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/clk.h>
+#include <linux/of_device.h>
+#include <linux/of_graph.h>
+#include <linux/pm_runtime.h>
+#include <linux/slab.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/drm_device.h>
+#include <drm/drm_modeset_helper.h>
+
+#include "dcss-dev.h"
+#include "dcss-kms.h"
+
+static void dcss_clocks_enable(struct dcss_dev *dcss)
+{
+	clk_prepare_enable(dcss->axi_clk);
+	clk_prepare_enable(dcss->apb_clk);
+	clk_prepare_enable(dcss->rtrm_clk);
+	clk_prepare_enable(dcss->dtrc_clk);
+	clk_prepare_enable(dcss->pix_clk);
+}
+
+static void dcss_clocks_disable(struct dcss_dev *dcss)
+{
+	clk_disable_unprepare(dcss->pix_clk);
+	clk_disable_unprepare(dcss->dtrc_clk);
+	clk_disable_unprepare(dcss->rtrm_clk);
+	clk_disable_unprepare(dcss->apb_clk);
+	clk_disable_unprepare(dcss->axi_clk);
+}
+
+static void dcss_disable_dtg_and_ss_cb(void *data)
+{
+	struct dcss_dev *dcss = data;
+
+	dcss->disable_callback = NULL;
+
+	dcss_ss_shutoff(dcss->ss);
+	dcss_dtg_shutoff(dcss->dtg);
+
+	complete(&dcss->disable_completion);
+}
+
+void dcss_disable_dtg_and_ss(struct dcss_dev *dcss)
+{
+	dcss->disable_callback = dcss_disable_dtg_and_ss_cb;
+}
+
+void dcss_enable_dtg_and_ss(struct dcss_dev *dcss)
+{
+	if (dcss->disable_callback)
+		dcss->disable_callback = NULL;
+
+	dcss_dtg_enable(dcss->dtg);
+	dcss_ss_enable(dcss->ss);
+}
+
+static int dcss_submodules_init(struct dcss_dev *dcss)
+{
+	int ret = 0;
+	u32 base_addr = dcss->start_addr;
+	const struct dcss_type_data *devtype = dcss->devtype;
+
+	dcss_clocks_enable(dcss);
+
+	ret = dcss_blkctl_init(dcss, base_addr + devtype->blkctl_ofs);
+	if (ret)
+		return ret;
+
+	ret = dcss_ctxld_init(dcss, base_addr + devtype->ctxld_ofs);
+	if (ret)
+		goto ctxld_err;
+
+	ret = dcss_dtg_init(dcss, base_addr + devtype->dtg_ofs);
+	if (ret)
+		goto dtg_err;
+
+	ret = dcss_ss_init(dcss, base_addr + devtype->ss_ofs);
+	if (ret)
+		goto ss_err;
+
+	ret = dcss_dpr_init(dcss, base_addr + devtype->dpr_ofs);
+	if (ret)
+		goto dpr_err;
+
+	ret = dcss_scaler_init(dcss, base_addr + devtype->scaler_ofs);
+	if (ret)
+		goto scaler_err;
+
+	dcss_clocks_disable(dcss);
+
+	return 0;
+
+scaler_err:
+	dcss_dpr_exit(dcss->dpr);
+
+dpr_err:
+	dcss_ss_exit(dcss->ss);
+
+ss_err:
+	dcss_dtg_exit(dcss->dtg);
+
+dtg_err:
+	dcss_ctxld_exit(dcss->ctxld);
+
+ctxld_err:
+	dcss_blkctl_exit(dcss->blkctl);
+
+	dcss_clocks_disable(dcss);
+
+	return ret;
+}
+
+static void dcss_submodules_stop(struct dcss_dev *dcss)
+{
+	dcss_clocks_enable(dcss);
+	dcss_scaler_exit(dcss->scaler);
+	dcss_dpr_exit(dcss->dpr);
+	dcss_ss_exit(dcss->ss);
+	dcss_dtg_exit(dcss->dtg);
+	dcss_ctxld_exit(dcss->ctxld);
+	dcss_blkctl_exit(dcss->blkctl);
+	dcss_clocks_disable(dcss);
+}
+
+static int dcss_clks_init(struct dcss_dev *dcss)
+{
+	int i;
+	struct {
+		const char *id;
+		struct clk **clk;
+	} clks[] = {
+		{"apb",   &dcss->apb_clk},
+		{"axi",   &dcss->axi_clk},
+		{"pix",   &dcss->pix_clk},
+		{"rtrm",  &dcss->rtrm_clk},
+		{"dtrc",  &dcss->dtrc_clk},
+	};
+
+	for (i = 0; i < ARRAY_SIZE(clks); i++) {
+		*clks[i].clk = devm_clk_get(dcss->dev, clks[i].id);
+		if (IS_ERR(*clks[i].clk)) {
+			dev_err(dcss->dev, "failed to get %s clock\n",
+				clks[i].id);
+			return PTR_ERR(*clks[i].clk);
+		}
+	}
+
+	return 0;
+}
+
+static void dcss_clks_release(struct dcss_dev *dcss)
+{
+	devm_clk_put(dcss->dev, dcss->dtrc_clk);
+	devm_clk_put(dcss->dev, dcss->rtrm_clk);
+	devm_clk_put(dcss->dev, dcss->pix_clk);
+	devm_clk_put(dcss->dev, dcss->axi_clk);
+	devm_clk_put(dcss->dev, dcss->apb_clk);
+}
+
+struct dcss_dev *dcss_dev_create(struct device *dev, bool hdmi_output)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	int ret;
+	struct resource *res;
+	struct dcss_dev *dcss;
+	const struct dcss_type_data *devtype;
+
+	devtype = of_device_get_match_data(dev);
+	if (!devtype) {
+		dev_err(dev, "no device match found\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(dev, "cannot get memory resource\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	dcss = kzalloc(sizeof(*dcss), GFP_KERNEL);
+	if (!dcss)
+		return ERR_PTR(-ENOMEM);
+
+	dcss->dev = dev;
+	dcss->devtype = devtype;
+	dcss->hdmi_output = hdmi_output;
+
+	ret = dcss_clks_init(dcss);
+	if (ret) {
+		dev_err(dev, "clocks initialization failed\n");
+		goto err;
+	}
+
+	dcss->of_port = of_graph_get_port_by_id(dev->of_node, 0);
+	if (!dcss->of_port) {
+		dev_err(dev, "no port@0 node in %s\n", dev->of_node->full_name);
+		ret = -ENODEV;
+		goto clks_err;
+	}
+
+	dcss->start_addr = res->start;
+
+	ret = dcss_submodules_init(dcss);
+	if (ret) {
+		dev_err(dev, "submodules initialization failed\n");
+		goto clks_err;
+	}
+
+	init_completion(&dcss->disable_completion);
+
+	pm_runtime_set_autosuspend_delay(dev, 100);
+	pm_runtime_use_autosuspend(dev);
+	pm_runtime_set_suspended(dev);
+	pm_runtime_allow(dev);
+	pm_runtime_enable(dev);
+
+	return dcss;
+
+clks_err:
+	dcss_clks_release(dcss);
+
+err:
+	kfree(dcss);
+
+	return ERR_PTR(ret);
+}
+
+void dcss_dev_destroy(struct dcss_dev *dcss)
+{
+	if (!pm_runtime_suspended(dcss->dev)) {
+		dcss_ctxld_suspend(dcss->ctxld);
+		dcss_clocks_disable(dcss);
+	}
+
+	pm_runtime_disable(dcss->dev);
+
+	dcss_submodules_stop(dcss);
+
+	dcss_clks_release(dcss);
+
+	kfree(dcss);
+}
+
+#ifdef CONFIG_PM_SLEEP
+int dcss_dev_suspend(struct device *dev)
+{
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(dev);
+	struct drm_device *ddev = dcss_drv_dev_to_drm(dev);
+	struct dcss_kms_dev *kms = container_of(ddev, struct dcss_kms_dev, base);
+	int ret;
+
+	drm_bridge_connector_disable_hpd(kms->connector);
+
+	drm_mode_config_helper_suspend(ddev);
+
+	if (pm_runtime_suspended(dev))
+		return 0;
+
+	ret = dcss_ctxld_suspend(dcss->ctxld);
+	if (ret)
+		return ret;
+
+	dcss_clocks_disable(dcss);
+
+	return 0;
+}
+
+int dcss_dev_resume(struct device *dev)
+{
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(dev);
+	struct drm_device *ddev = dcss_drv_dev_to_drm(dev);
+	struct dcss_kms_dev *kms = container_of(ddev, struct dcss_kms_dev, base);
+
+	if (pm_runtime_suspended(dev)) {
+		drm_mode_config_helper_resume(ddev);
+		return 0;
+	}
+
+	dcss_clocks_enable(dcss);
+
+	dcss_blkctl_cfg(dcss->blkctl);
+
+	dcss_ctxld_resume(dcss->ctxld);
+
+	drm_mode_config_helper_resume(ddev);
+
+	drm_bridge_connector_enable_hpd(kms->connector);
+
+	return 0;
+}
+#endif /* CONFIG_PM_SLEEP */
+
+#ifdef CONFIG_PM
+int dcss_dev_runtime_suspend(struct device *dev)
+{
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(dev);
+	int ret;
+
+	ret = dcss_ctxld_suspend(dcss->ctxld);
+	if (ret)
+		return ret;
+
+	dcss_clocks_disable(dcss);
+
+	return 0;
+}
+
+int dcss_dev_runtime_resume(struct device *dev)
+{
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(dev);
+
+	dcss_clocks_enable(dcss);
+
+	dcss_blkctl_cfg(dcss->blkctl);
+
+	dcss_ctxld_resume(dcss->ctxld);
+
+	return 0;
+}
+#endif /* CONFIG_PM */
diff --git a/drivers/gpu/drm/imx/dcss/dcss-dev.h b/drivers/gpu/drm/imx/dcss/dcss-dev.h
new file mode 100644
index 000000000000..c642ae17837f
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-dev.h
@@ -0,0 +1,177 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2019 NXP.
+ */
+
+#ifndef __DCSS_PRV_H__
+#define __DCSS_PRV_H__
+
+#include <drm/drm_fourcc.h>
+#include <linux/io.h>
+#include <video/videomode.h>
+
+#define SET			0x04
+#define CLR			0x08
+#define TGL			0x0C
+
+#define dcss_writel(v, c)	writel((v), (c))
+#define dcss_readl(c)		readl(c)
+#define dcss_set(v, c)		writel((v), (c) + SET)
+#define dcss_clr(v, c)		writel((v), (c) + CLR)
+#define dcss_toggle(v, c)	writel((v), (c) + TGL)
+
+static inline void dcss_update(u32 v, u32 m, void __iomem *c)
+{
+	writel((readl(c) & ~(m)) | (v), (c));
+}
+
+#define DCSS_DBG_REG(reg)	{.name = #reg, .ofs = reg}
+
+enum {
+	DCSS_IMX8MQ = 0,
+};
+
+struct dcss_type_data {
+	const char *name;
+	u32 blkctl_ofs;
+	u32 ctxld_ofs;
+	u32 rdsrc_ofs;
+	u32 wrscl_ofs;
+	u32 dtg_ofs;
+	u32 scaler_ofs;
+	u32 ss_ofs;
+	u32 dpr_ofs;
+	u32 dtrc_ofs;
+	u32 dec400d_ofs;
+	u32 hdr10_ofs;
+};
+
+struct dcss_debug_reg {
+	char *name;
+	u32 ofs;
+};
+
+enum dcss_ctxld_ctx_type {
+	CTX_DB,
+	CTX_SB_HP, /* high-priority */
+	CTX_SB_LP, /* low-priority  */
+};
+
+struct dcss_dev {
+	struct device *dev;
+	const struct dcss_type_data *devtype;
+	struct device_node *of_port;
+
+	u32 start_addr;
+
+	struct dcss_blkctl *blkctl;
+	struct dcss_ctxld *ctxld;
+	struct dcss_dpr *dpr;
+	struct dcss_dtg *dtg;
+	struct dcss_ss *ss;
+	struct dcss_hdr10 *hdr10;
+	struct dcss_scaler *scaler;
+	struct dcss_dtrc *dtrc;
+	struct dcss_dec400d *dec400d;
+	struct dcss_wrscl *wrscl;
+	struct dcss_rdsrc *rdsrc;
+
+	struct clk *apb_clk;
+	struct clk *axi_clk;
+	struct clk *pix_clk;
+	struct clk *rtrm_clk;
+	struct clk *dtrc_clk;
+	struct clk *pll_src_clk;
+	struct clk *pll_phy_ref_clk;
+
+	bool hdmi_output;
+
+	void (*disable_callback)(void *data);
+	struct completion disable_completion;
+};
+
+struct dcss_dev *dcss_drv_dev_to_dcss(struct device *dev);
+struct drm_device *dcss_drv_dev_to_drm(struct device *dev);
+struct dcss_dev *dcss_dev_create(struct device *dev, bool hdmi_output);
+void dcss_dev_destroy(struct dcss_dev *dcss);
+int dcss_dev_runtime_suspend(struct device *dev);
+int dcss_dev_runtime_resume(struct device *dev);
+int dcss_dev_suspend(struct device *dev);
+int dcss_dev_resume(struct device *dev);
+void dcss_enable_dtg_and_ss(struct dcss_dev *dcss);
+void dcss_disable_dtg_and_ss(struct dcss_dev *dcss);
+
+/* BLKCTL */
+int dcss_blkctl_init(struct dcss_dev *dcss, unsigned long blkctl_base);
+void dcss_blkctl_cfg(struct dcss_blkctl *blkctl);
+void dcss_blkctl_exit(struct dcss_blkctl *blkctl);
+
+/* CTXLD */
+int dcss_ctxld_init(struct dcss_dev *dcss, unsigned long ctxld_base);
+void dcss_ctxld_exit(struct dcss_ctxld *ctxld);
+void dcss_ctxld_write(struct dcss_ctxld *ctxld, u32 ctx_id,
+		      u32 val, u32 reg_idx);
+int dcss_ctxld_resume(struct dcss_ctxld *dcss_ctxld);
+int dcss_ctxld_suspend(struct dcss_ctxld *dcss_ctxld);
+void dcss_ctxld_write_irqsafe(struct dcss_ctxld *ctlxd, u32 ctx_id, u32 val,
+			      u32 reg_ofs);
+void dcss_ctxld_kick(struct dcss_ctxld *ctxld);
+bool dcss_ctxld_is_flushed(struct dcss_ctxld *ctxld);
+int dcss_ctxld_enable(struct dcss_ctxld *ctxld);
+void dcss_ctxld_register_completion(struct dcss_ctxld *ctxld,
+				    struct completion *dis_completion);
+void dcss_ctxld_assert_locked(struct dcss_ctxld *ctxld);
+
+/* DPR */
+int dcss_dpr_init(struct dcss_dev *dcss, unsigned long dpr_base);
+void dcss_dpr_exit(struct dcss_dpr *dpr);
+void dcss_dpr_write_sysctrl(struct dcss_dpr *dpr);
+void dcss_dpr_set_res(struct dcss_dpr *dpr, int ch_num, u32 xres, u32 yres);
+void dcss_dpr_addr_set(struct dcss_dpr *dpr, int ch_num, u32 luma_base_addr,
+		       u32 chroma_base_addr, u16 pitch);
+void dcss_dpr_enable(struct dcss_dpr *dpr, int ch_num, bool en);
+void dcss_dpr_format_set(struct dcss_dpr *dpr, int ch_num,
+			 const struct drm_format_info *format, u64 modifier);
+void dcss_dpr_set_rotation(struct dcss_dpr *dpr, int ch_num, u32 rotation);
+
+/* DTG */
+int dcss_dtg_init(struct dcss_dev *dcss, unsigned long dtg_base);
+void dcss_dtg_exit(struct dcss_dtg *dtg);
+bool dcss_dtg_vblank_irq_valid(struct dcss_dtg *dtg);
+void dcss_dtg_vblank_irq_enable(struct dcss_dtg *dtg, bool en);
+void dcss_dtg_vblank_irq_clear(struct dcss_dtg *dtg);
+void dcss_dtg_sync_set(struct dcss_dtg *dtg, struct videomode *vm);
+void dcss_dtg_css_set(struct dcss_dtg *dtg);
+void dcss_dtg_enable(struct dcss_dtg *dtg);
+void dcss_dtg_shutoff(struct dcss_dtg *dtg);
+bool dcss_dtg_is_enabled(struct dcss_dtg *dtg);
+void dcss_dtg_ctxld_kick_irq_enable(struct dcss_dtg *dtg, bool en);
+bool dcss_dtg_global_alpha_changed(struct dcss_dtg *dtg, int ch_num, int alpha);
+void dcss_dtg_plane_alpha_set(struct dcss_dtg *dtg, int ch_num,
+			      const struct drm_format_info *format, int alpha);
+void dcss_dtg_plane_pos_set(struct dcss_dtg *dtg, int ch_num,
+			    int px, int py, int pw, int ph);
+void dcss_dtg_ch_enable(struct dcss_dtg *dtg, int ch_num, bool en);
+
+/* SUBSAM */
+int dcss_ss_init(struct dcss_dev *dcss, unsigned long subsam_base);
+void dcss_ss_exit(struct dcss_ss *ss);
+void dcss_ss_enable(struct dcss_ss *ss);
+void dcss_ss_shutoff(struct dcss_ss *ss);
+void dcss_ss_subsam_set(struct dcss_ss *ss);
+void dcss_ss_sync_set(struct dcss_ss *ss, struct videomode *vm,
+		      bool phsync, bool pvsync);
+
+/* SCALER */
+int dcss_scaler_init(struct dcss_dev *dcss, unsigned long scaler_base);
+void dcss_scaler_exit(struct dcss_scaler *scl);
+void dcss_scaler_setup(struct dcss_scaler *scl, int ch_num,
+		       const struct drm_format_info *format,
+		       int src_xres, int src_yres, int dst_xres, int dst_yres,
+		       u32 vrefresh_hz);
+void dcss_scaler_ch_enable(struct dcss_scaler *scl, int ch_num, bool en);
+int dcss_scaler_get_min_max_ratios(struct dcss_scaler *scl, int ch_num,
+				   int *min, int *max);
+void dcss_scaler_write_sclctrl(struct dcss_scaler *scl);
+
+#endif /* __DCSS_PRV_H__ */
diff --git a/drivers/gpu/drm/imx/dcss/dcss-dpr.c b/drivers/gpu/drm/imx/dcss/dcss-dpr.c
new file mode 100644
index 000000000000..df9dab949bf2
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-dpr.c
@@ -0,0 +1,562 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_DPR_SYSTEM_CTRL0			0x000
+#define   RUN_EN				BIT(0)
+#define   SOFT_RESET				BIT(1)
+#define   REPEAT_EN				BIT(2)
+#define   SHADOW_LOAD_EN			BIT(3)
+#define   SW_SHADOW_LOAD_SEL			BIT(4)
+#define   BCMD2AXI_MSTR_ID_CTRL			BIT(16)
+#define DCSS_DPR_IRQ_MASK			0x020
+#define DCSS_DPR_IRQ_MASK_STATUS		0x030
+#define DCSS_DPR_IRQ_NONMASK_STATUS		0x040
+#define   IRQ_DPR_CTRL_DONE			BIT(0)
+#define   IRQ_DPR_RUN				BIT(1)
+#define   IRQ_DPR_SHADOW_LOADED			BIT(2)
+#define   IRQ_AXI_READ_ERR			BIT(3)
+#define   DPR2RTR_YRGB_FIFO_OVFL		BIT(4)
+#define   DPR2RTR_UV_FIFO_OVFL			BIT(5)
+#define   DPR2RTR_FIFO_LD_BUF_RDY_YRGB_ERR	BIT(6)
+#define   DPR2RTR_FIFO_LD_BUF_RDY_UV_ERR	BIT(7)
+#define DCSS_DPR_MODE_CTRL0			0x050
+#define   RTR_3BUF_EN				BIT(0)
+#define   RTR_4LINE_BUF_EN			BIT(1)
+#define   TILE_TYPE_POS				2
+#define   TILE_TYPE_MASK			GENMASK(4, 2)
+#define   YUV_EN				BIT(6)
+#define   COMP_2PLANE_EN			BIT(7)
+#define   PIX_SIZE_POS				8
+#define   PIX_SIZE_MASK				GENMASK(9, 8)
+#define   PIX_LUMA_UV_SWAP			BIT(10)
+#define   PIX_UV_SWAP				BIT(11)
+#define   B_COMP_SEL_POS			12
+#define   B_COMP_SEL_MASK			GENMASK(13, 12)
+#define   G_COMP_SEL_POS			14
+#define   G_COMP_SEL_MASK			GENMASK(15, 14)
+#define   R_COMP_SEL_POS			16
+#define   R_COMP_SEL_MASK			GENMASK(17, 16)
+#define   A_COMP_SEL_POS			18
+#define   A_COMP_SEL_MASK			GENMASK(19, 18)
+#define DCSS_DPR_FRAME_CTRL0			0x070
+#define   HFLIP_EN				BIT(0)
+#define   VFLIP_EN				BIT(1)
+#define   ROT_ENC_POS				2
+#define   ROT_ENC_MASK				GENMASK(3, 2)
+#define   ROT_FLIP_ORDER_EN			BIT(4)
+#define   PITCH_POS				16
+#define   PITCH_MASK				GENMASK(31, 16)
+#define DCSS_DPR_FRAME_1P_CTRL0			0x090
+#define DCSS_DPR_FRAME_1P_PIX_X_CTRL		0x0A0
+#define DCSS_DPR_FRAME_1P_PIX_Y_CTRL		0x0B0
+#define DCSS_DPR_FRAME_1P_BASE_ADDR		0x0C0
+#define DCSS_DPR_FRAME_2P_CTRL0			0x0E0
+#define DCSS_DPR_FRAME_2P_PIX_X_CTRL		0x0F0
+#define DCSS_DPR_FRAME_2P_PIX_Y_CTRL		0x100
+#define DCSS_DPR_FRAME_2P_BASE_ADDR		0x110
+#define DCSS_DPR_STATUS_CTRL0			0x130
+#define   STATUS_MUX_SEL_MASK			GENMASK(2, 0)
+#define   STATUS_SRC_SEL_POS			16
+#define   STATUS_SRC_SEL_MASK			GENMASK(18, 16)
+#define DCSS_DPR_STATUS_CTRL1			0x140
+#define DCSS_DPR_RTRAM_CTRL0			0x200
+#define   NUM_ROWS_ACTIVE			BIT(0)
+#define   THRES_HIGH_POS			1
+#define   THRES_HIGH_MASK			GENMASK(3, 1)
+#define   THRES_LOW_POS				4
+#define   THRES_LOW_MASK			GENMASK(6, 4)
+#define   ABORT_SEL				BIT(7)
+
+enum dcss_tile_type {
+	TILE_LINEAR = 0,
+	TILE_GPU_STANDARD,
+	TILE_GPU_SUPER,
+	TILE_VPU_YUV420,
+	TILE_VPU_VP9,
+};
+
+enum dcss_pix_size {
+	PIX_SIZE_8,
+	PIX_SIZE_16,
+	PIX_SIZE_32,
+};
+
+struct dcss_dpr_ch {
+	struct dcss_dpr *dpr;
+	void __iomem *base_reg;
+	u32 base_ofs;
+
+	struct drm_format_info format;
+	enum dcss_pix_size pix_size;
+	enum dcss_tile_type tile;
+	bool rtram_4line_en;
+	bool rtram_3buf_en;
+
+	u32 frame_ctrl;
+	u32 mode_ctrl;
+	u32 sys_ctrl;
+	u32 rtram_ctrl;
+
+	bool sys_ctrl_chgd;
+
+	int ch_num;
+	int irq;
+};
+
+struct dcss_dpr {
+	struct device *dev;
+	struct dcss_ctxld *ctxld;
+	u32  ctx_id;
+
+	struct dcss_dpr_ch ch[3];
+};
+
+static void dcss_dpr_write(struct dcss_dpr_ch *ch, u32 val, u32 ofs)
+{
+	struct dcss_dpr *dpr = ch->dpr;
+
+	dcss_ctxld_write(dpr->ctxld, dpr->ctx_id, val, ch->base_ofs + ofs);
+}
+
+static int dcss_dpr_ch_init_all(struct dcss_dpr *dpr, unsigned long dpr_base)
+{
+	struct dcss_dpr_ch *ch;
+	int i;
+
+	for (i = 0; i < 3; i++) {
+		ch = &dpr->ch[i];
+
+		ch->base_ofs = dpr_base + i * 0x1000;
+
+		ch->base_reg = ioremap(ch->base_ofs, SZ_4K);
+		if (!ch->base_reg) {
+			dev_err(dpr->dev, "dpr: unable to remap ch %d base\n",
+				i);
+			return -ENOMEM;
+		}
+
+		ch->dpr = dpr;
+		ch->ch_num = i;
+
+		dcss_writel(0xff, ch->base_reg + DCSS_DPR_IRQ_MASK);
+	}
+
+	return 0;
+}
+
+int dcss_dpr_init(struct dcss_dev *dcss, unsigned long dpr_base)
+{
+	struct dcss_dpr *dpr;
+
+	dpr = kzalloc(sizeof(*dpr), GFP_KERNEL);
+	if (!dpr)
+		return -ENOMEM;
+
+	dcss->dpr = dpr;
+	dpr->dev = dcss->dev;
+	dpr->ctxld = dcss->ctxld;
+	dpr->ctx_id = CTX_SB_HP;
+
+	if (dcss_dpr_ch_init_all(dpr, dpr_base)) {
+		int i;
+
+		for (i = 0; i < 3; i++) {
+			if (dpr->ch[i].base_reg)
+				iounmap(dpr->ch[i].base_reg);
+		}
+
+		kfree(dpr);
+
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void dcss_dpr_exit(struct dcss_dpr *dpr)
+{
+	int ch_no;
+
+	/* stop DPR on all channels */
+	for (ch_no = 0; ch_no < 3; ch_no++) {
+		struct dcss_dpr_ch *ch = &dpr->ch[ch_no];
+
+		dcss_writel(0, ch->base_reg + DCSS_DPR_SYSTEM_CTRL0);
+
+		if (ch->base_reg)
+			iounmap(ch->base_reg);
+	}
+
+	kfree(dpr);
+}
+
+static u32 dcss_dpr_x_pix_wide_adjust(struct dcss_dpr_ch *ch, u32 pix_wide,
+				      u32 pix_format)
+{
+	u8 pix_in_64byte_map[3][5] = {
+		/* LIN, GPU_STD, GPU_SUP, VPU_YUV420, VPU_VP9 */
+		{   64,       8,       8,          8,     16}, /* PIX_SIZE_8  */
+		{   32,       8,       8,          8,      8}, /* PIX_SIZE_16 */
+		{   16,       4,       4,          8,      8}, /* PIX_SIZE_32 */
+	};
+	u32 offset;
+	u32 div_64byte_mod, pix_in_64byte;
+
+	pix_in_64byte = pix_in_64byte_map[ch->pix_size][ch->tile];
+
+	div_64byte_mod = pix_wide % pix_in_64byte;
+	offset = (div_64byte_mod == 0) ? 0 : (pix_in_64byte - div_64byte_mod);
+
+	return pix_wide + offset;
+}
+
+static u32 dcss_dpr_y_pix_high_adjust(struct dcss_dpr_ch *ch, u32 pix_high,
+				      u32 pix_format)
+{
+	u8 num_rows_buf = ch->rtram_4line_en ? 4 : 8;
+	u32 offset, pix_y_mod;
+
+	pix_y_mod = pix_high % num_rows_buf;
+	offset = pix_y_mod ? (num_rows_buf - pix_y_mod) : 0;
+
+	return pix_high + offset;
+}
+
+void dcss_dpr_set_res(struct dcss_dpr *dpr, int ch_num, u32 xres, u32 yres)
+{
+	struct dcss_dpr_ch *ch = &dpr->ch[ch_num];
+	u32 pix_format = ch->format.format;
+	u32 gap = DCSS_DPR_FRAME_2P_BASE_ADDR - DCSS_DPR_FRAME_1P_BASE_ADDR;
+	int plane, max_planes = 1;
+	u32 pix_x_wide, pix_y_high;
+
+	if (pix_format == DRM_FORMAT_NV12 ||
+	    pix_format == DRM_FORMAT_NV21)
+		max_planes = 2;
+
+	for (plane = 0; plane < max_planes; plane++) {
+		yres = plane == 1 ? yres >> 1 : yres;
+
+		pix_x_wide = dcss_dpr_x_pix_wide_adjust(ch, xres, pix_format);
+		pix_y_high = dcss_dpr_y_pix_high_adjust(ch, yres, pix_format);
+
+		dcss_dpr_write(ch, pix_x_wide,
+			       DCSS_DPR_FRAME_1P_PIX_X_CTRL + plane * gap);
+		dcss_dpr_write(ch, pix_y_high,
+			       DCSS_DPR_FRAME_1P_PIX_Y_CTRL + plane * gap);
+
+		dcss_dpr_write(ch, 2, DCSS_DPR_FRAME_1P_CTRL0 + plane * gap);
+	}
+}
+
+void dcss_dpr_addr_set(struct dcss_dpr *dpr, int ch_num, u32 luma_base_addr,
+		       u32 chroma_base_addr, u16 pitch)
+{
+	struct dcss_dpr_ch *ch = &dpr->ch[ch_num];
+
+	dcss_dpr_write(ch, luma_base_addr, DCSS_DPR_FRAME_1P_BASE_ADDR);
+
+	dcss_dpr_write(ch, chroma_base_addr, DCSS_DPR_FRAME_2P_BASE_ADDR);
+
+	ch->frame_ctrl &= ~PITCH_MASK;
+	ch->frame_ctrl |= (((u32)pitch << PITCH_POS) & PITCH_MASK);
+}
+
+static void dcss_dpr_argb_comp_sel(struct dcss_dpr_ch *ch, int a_sel, int r_sel,
+				   int g_sel, int b_sel)
+{
+	u32 sel;
+
+	sel = ((a_sel << A_COMP_SEL_POS) & A_COMP_SEL_MASK) |
+	      ((r_sel << R_COMP_SEL_POS) & R_COMP_SEL_MASK) |
+	      ((g_sel << G_COMP_SEL_POS) & G_COMP_SEL_MASK) |
+	      ((b_sel << B_COMP_SEL_POS) & B_COMP_SEL_MASK);
+
+	ch->mode_ctrl &= ~(A_COMP_SEL_MASK | R_COMP_SEL_MASK |
+			   G_COMP_SEL_MASK | B_COMP_SEL_MASK);
+	ch->mode_ctrl |= sel;
+}
+
+static void dcss_dpr_pix_size_set(struct dcss_dpr_ch *ch,
+				  const struct drm_format_info *format)
+{
+	u32 val;
+
+	switch (format->format) {
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+		val = PIX_SIZE_8;
+		break;
+
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_VYUY:
+	case DRM_FORMAT_YUYV:
+	case DRM_FORMAT_YVYU:
+		val = PIX_SIZE_16;
+		break;
+
+	default:
+		val = PIX_SIZE_32;
+		break;
+	}
+
+	ch->pix_size = val;
+
+	ch->mode_ctrl &= ~PIX_SIZE_MASK;
+	ch->mode_ctrl |= ((val << PIX_SIZE_POS) & PIX_SIZE_MASK);
+}
+
+static void dcss_dpr_uv_swap(struct dcss_dpr_ch *ch, bool swap)
+{
+	ch->mode_ctrl &= ~PIX_UV_SWAP;
+	ch->mode_ctrl |= (swap ? PIX_UV_SWAP : 0);
+}
+
+static void dcss_dpr_y_uv_swap(struct dcss_dpr_ch *ch, bool swap)
+{
+	ch->mode_ctrl &= ~PIX_LUMA_UV_SWAP;
+	ch->mode_ctrl |= (swap ? PIX_LUMA_UV_SWAP : 0);
+}
+
+static void dcss_dpr_2plane_en(struct dcss_dpr_ch *ch, bool en)
+{
+	ch->mode_ctrl &= ~COMP_2PLANE_EN;
+	ch->mode_ctrl |= (en ? COMP_2PLANE_EN : 0);
+}
+
+static void dcss_dpr_yuv_en(struct dcss_dpr_ch *ch, bool en)
+{
+	ch->mode_ctrl &= ~YUV_EN;
+	ch->mode_ctrl |= (en ? YUV_EN : 0);
+}
+
+void dcss_dpr_enable(struct dcss_dpr *dpr, int ch_num, bool en)
+{
+	struct dcss_dpr_ch *ch = &dpr->ch[ch_num];
+	u32 sys_ctrl;
+
+	sys_ctrl = (en ? REPEAT_EN | RUN_EN : 0);
+
+	if (en) {
+		dcss_dpr_write(ch, ch->mode_ctrl, DCSS_DPR_MODE_CTRL0);
+		dcss_dpr_write(ch, ch->frame_ctrl, DCSS_DPR_FRAME_CTRL0);
+		dcss_dpr_write(ch, ch->rtram_ctrl, DCSS_DPR_RTRAM_CTRL0);
+	}
+
+	if (ch->sys_ctrl != sys_ctrl)
+		ch->sys_ctrl_chgd = true;
+
+	ch->sys_ctrl = sys_ctrl;
+}
+
+struct rgb_comp_sel {
+	u32 drm_format;
+	int a_sel;
+	int r_sel;
+	int g_sel;
+	int b_sel;
+};
+
+static struct rgb_comp_sel comp_sel_map[] = {
+	{DRM_FORMAT_ARGB8888, 3, 2, 1, 0},
+	{DRM_FORMAT_XRGB8888, 3, 2, 1, 0},
+	{DRM_FORMAT_ABGR8888, 3, 0, 1, 2},
+	{DRM_FORMAT_XBGR8888, 3, 0, 1, 2},
+	{DRM_FORMAT_RGBA8888, 0, 3, 2, 1},
+	{DRM_FORMAT_RGBX8888, 0, 3, 2, 1},
+	{DRM_FORMAT_BGRA8888, 0, 1, 2, 3},
+	{DRM_FORMAT_BGRX8888, 0, 1, 2, 3},
+};
+
+static int to_comp_sel(u32 pix_fmt, int *a_sel, int *r_sel, int *g_sel,
+		       int *b_sel)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(comp_sel_map); i++) {
+		if (comp_sel_map[i].drm_format == pix_fmt) {
+			*a_sel = comp_sel_map[i].a_sel;
+			*r_sel = comp_sel_map[i].r_sel;
+			*g_sel = comp_sel_map[i].g_sel;
+			*b_sel = comp_sel_map[i].b_sel;
+
+			return 0;
+		}
+	}
+
+	return -1;
+}
+
+static void dcss_dpr_rtram_set(struct dcss_dpr_ch *ch, u32 pix_format)
+{
+	u32 val, mask;
+
+	switch (pix_format) {
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV12:
+		ch->rtram_3buf_en = true;
+		ch->rtram_4line_en = false;
+		break;
+
+	default:
+		ch->rtram_3buf_en = true;
+		ch->rtram_4line_en = true;
+		break;
+	}
+
+	val = (ch->rtram_4line_en ? RTR_4LINE_BUF_EN : 0);
+	val |= (ch->rtram_3buf_en ? RTR_3BUF_EN : 0);
+	mask = RTR_4LINE_BUF_EN | RTR_3BUF_EN;
+
+	ch->mode_ctrl &= ~mask;
+	ch->mode_ctrl |= (val & mask);
+
+	val = (ch->rtram_4line_en ? 0 : NUM_ROWS_ACTIVE);
+	val |= (3 << THRES_LOW_POS) & THRES_LOW_MASK;
+	val |= (4 << THRES_HIGH_POS) & THRES_HIGH_MASK;
+	mask = THRES_LOW_MASK | THRES_HIGH_MASK | NUM_ROWS_ACTIVE;
+
+	ch->rtram_ctrl &= ~mask;
+	ch->rtram_ctrl |= (val & mask);
+}
+
+static void dcss_dpr_setup_components(struct dcss_dpr_ch *ch,
+				      const struct drm_format_info *format)
+{
+	int a_sel, r_sel, g_sel, b_sel;
+	bool uv_swap, y_uv_swap;
+
+	switch (format->format) {
+	case DRM_FORMAT_YVYU:
+		uv_swap = true;
+		y_uv_swap = true;
+		break;
+
+	case DRM_FORMAT_VYUY:
+	case DRM_FORMAT_NV21:
+		uv_swap = true;
+		y_uv_swap = false;
+		break;
+
+	case DRM_FORMAT_YUYV:
+		uv_swap = false;
+		y_uv_swap = true;
+		break;
+
+	default:
+		uv_swap = false;
+		y_uv_swap = false;
+		break;
+	}
+
+	dcss_dpr_uv_swap(ch, uv_swap);
+
+	dcss_dpr_y_uv_swap(ch, y_uv_swap);
+
+	if (!format->is_yuv) {
+		if (!to_comp_sel(format->format, &a_sel, &r_sel,
+				 &g_sel, &b_sel)) {
+			dcss_dpr_argb_comp_sel(ch, a_sel, r_sel, g_sel, b_sel);
+		} else {
+			dcss_dpr_argb_comp_sel(ch, 3, 2, 1, 0);
+		}
+	} else {
+		dcss_dpr_argb_comp_sel(ch, 0, 0, 0, 0);
+	}
+}
+
+static void dcss_dpr_tile_set(struct dcss_dpr_ch *ch, uint64_t modifier)
+{
+	switch (ch->ch_num) {
+	case 0:
+		switch (modifier) {
+		case DRM_FORMAT_MOD_LINEAR:
+			ch->tile = TILE_LINEAR;
+			break;
+		case DRM_FORMAT_MOD_VIVANTE_TILED:
+			ch->tile = TILE_GPU_STANDARD;
+			break;
+		case DRM_FORMAT_MOD_VIVANTE_SUPER_TILED:
+			ch->tile = TILE_GPU_SUPER;
+			break;
+		default:
+			WARN_ON(1);
+			break;
+		}
+		break;
+	case 1:
+	case 2:
+		ch->tile = TILE_LINEAR;
+		break;
+	default:
+		WARN_ON(1);
+		return;
+	}
+
+	ch->mode_ctrl &= ~TILE_TYPE_MASK;
+	ch->mode_ctrl |= ((ch->tile << TILE_TYPE_POS) & TILE_TYPE_MASK);
+}
+
+void dcss_dpr_format_set(struct dcss_dpr *dpr, int ch_num,
+			 const struct drm_format_info *format, u64 modifier)
+{
+	struct dcss_dpr_ch *ch = &dpr->ch[ch_num];
+
+	ch->format = *format;
+
+	dcss_dpr_yuv_en(ch, format->is_yuv);
+
+	dcss_dpr_pix_size_set(ch, format);
+
+	dcss_dpr_setup_components(ch, format);
+
+	dcss_dpr_2plane_en(ch, format->num_planes == 2);
+
+	dcss_dpr_rtram_set(ch, format->format);
+
+	dcss_dpr_tile_set(ch, modifier);
+}
+
+/* This function will be called from interrupt context. */
+void dcss_dpr_write_sysctrl(struct dcss_dpr *dpr)
+{
+	int chnum;
+
+	dcss_ctxld_assert_locked(dpr->ctxld);
+
+	for (chnum = 0; chnum < 3; chnum++) {
+		struct dcss_dpr_ch *ch = &dpr->ch[chnum];
+
+		if (ch->sys_ctrl_chgd) {
+			dcss_ctxld_write_irqsafe(dpr->ctxld, dpr->ctx_id,
+						 ch->sys_ctrl,
+						 ch->base_ofs +
+						 DCSS_DPR_SYSTEM_CTRL0);
+			ch->sys_ctrl_chgd = false;
+		}
+	}
+}
+
+void dcss_dpr_set_rotation(struct dcss_dpr *dpr, int ch_num, u32 rotation)
+{
+	struct dcss_dpr_ch *ch = &dpr->ch[ch_num];
+
+	ch->frame_ctrl &= ~(HFLIP_EN | VFLIP_EN | ROT_ENC_MASK);
+
+	ch->frame_ctrl |= rotation & DRM_MODE_REFLECT_X ? HFLIP_EN : 0;
+	ch->frame_ctrl |= rotation & DRM_MODE_REFLECT_Y ? VFLIP_EN : 0;
+
+	if (rotation & DRM_MODE_ROTATE_90)
+		ch->frame_ctrl |= 1 << ROT_ENC_POS;
+	else if (rotation & DRM_MODE_ROTATE_180)
+		ch->frame_ctrl |= 2 << ROT_ENC_POS;
+	else if (rotation & DRM_MODE_ROTATE_270)
+		ch->frame_ctrl |= 3 << ROT_ENC_POS;
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-drv.c b/drivers/gpu/drm/imx/dcss/dcss-drv.c
new file mode 100644
index 000000000000..8dc2f85c514b
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-drv.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <drm/drm_of.h>
+
+#include "dcss-dev.h"
+#include "dcss-kms.h"
+
+struct dcss_drv {
+	struct dcss_dev *dcss;
+	struct dcss_kms_dev *kms;
+};
+
+struct dcss_dev *dcss_drv_dev_to_dcss(struct device *dev)
+{
+	struct dcss_drv *mdrv = dev_get_drvdata(dev);
+
+	return mdrv ? mdrv->dcss : NULL;
+}
+
+struct drm_device *dcss_drv_dev_to_drm(struct device *dev)
+{
+	struct dcss_drv *mdrv = dev_get_drvdata(dev);
+
+	return mdrv ? &mdrv->kms->base : NULL;
+}
+
+static int dcss_drv_platform_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *remote;
+	struct dcss_drv *mdrv;
+	int err = 0;
+	bool hdmi_output = true;
+
+	if (!dev->of_node)
+		return -ENODEV;
+
+	remote = of_graph_get_remote_node(dev->of_node, 0, 0);
+	if (!remote)
+		return -ENODEV;
+
+	hdmi_output = !of_device_is_compatible(remote, "fsl,imx8mq-nwl-dsi");
+
+	of_node_put(remote);
+
+	mdrv = kzalloc(sizeof(*mdrv), GFP_KERNEL);
+	if (!mdrv)
+		return -ENOMEM;
+
+	mdrv->dcss = dcss_dev_create(dev, hdmi_output);
+	if (IS_ERR(mdrv->dcss)) {
+		err = PTR_ERR(mdrv->dcss);
+		goto err;
+	}
+
+	dev_set_drvdata(dev, mdrv);
+
+	mdrv->kms = dcss_kms_attach(mdrv->dcss);
+	if (IS_ERR(mdrv->kms)) {
+		err = PTR_ERR(mdrv->kms);
+		goto dcss_shutoff;
+	}
+
+	return 0;
+
+dcss_shutoff:
+	dcss_dev_destroy(mdrv->dcss);
+
+	dev_set_drvdata(dev, NULL);
+
+err:
+	kfree(mdrv);
+	return err;
+}
+
+static int dcss_drv_platform_remove(struct platform_device *pdev)
+{
+	struct dcss_drv *mdrv = dev_get_drvdata(&pdev->dev);
+
+	if (!mdrv)
+		return 0;
+
+	dcss_kms_detach(mdrv->kms);
+	dcss_dev_destroy(mdrv->dcss);
+
+	dev_set_drvdata(&pdev->dev, NULL);
+
+	kfree(mdrv);
+
+	return 0;
+}
+
+static struct dcss_type_data dcss_types[] = {
+	[DCSS_IMX8MQ] = {
+		.name = "DCSS_IMX8MQ",
+		.blkctl_ofs = 0x2F000,
+		.ctxld_ofs = 0x23000,
+		.dtg_ofs = 0x20000,
+		.scaler_ofs = 0x1C000,
+		.ss_ofs = 0x1B000,
+		.dpr_ofs = 0x18000,
+	},
+};
+
+static const struct of_device_id dcss_of_match[] = {
+	{ .compatible = "nxp,imx8mq-dcss", .data = &dcss_types[DCSS_IMX8MQ], },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, dcss_of_match);
+
+static const struct dev_pm_ops dcss_dev_pm = {
+	SET_SYSTEM_SLEEP_PM_OPS(dcss_dev_suspend, dcss_dev_resume)
+	SET_RUNTIME_PM_OPS(dcss_dev_runtime_suspend,
+			   dcss_dev_runtime_resume, NULL)
+};
+
+static struct platform_driver dcss_platform_driver = {
+	.probe	= dcss_drv_platform_probe,
+	.remove	= dcss_drv_platform_remove,
+	.driver	= {
+		.name = "imx-dcss",
+		.of_match_table	= dcss_of_match,
+		.pm = &dcss_dev_pm,
+	},
+};
+
+module_platform_driver(dcss_platform_driver);
+
+MODULE_AUTHOR("Laurentiu Palcu <laurentiu.palcu@nxp.com>");
+MODULE_DESCRIPTION("DCSS driver for i.MX8MQ");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/imx/dcss/dcss-dtg.c b/drivers/gpu/drm/imx/dcss/dcss-dtg.c
new file mode 100644
index 000000000000..30de00540f63
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-dtg.c
@@ -0,0 +1,409 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_DTG_TC_CONTROL_STATUS			0x00
+#define   CH3_EN					BIT(0)
+#define   CH2_EN					BIT(1)
+#define   CH1_EN					BIT(2)
+#define   OVL_DATA_MODE					BIT(3)
+#define   BLENDER_VIDEO_ALPHA_SEL			BIT(7)
+#define   DTG_START					BIT(8)
+#define   DBY_MODE_EN					BIT(9)
+#define   CH1_ALPHA_SEL					BIT(10)
+#define   CSS_PIX_COMP_SWAP_POS				12
+#define   CSS_PIX_COMP_SWAP_MASK			GENMASK(14, 12)
+#define   DEFAULT_FG_ALPHA_POS				24
+#define   DEFAULT_FG_ALPHA_MASK				GENMASK(31, 24)
+#define DCSS_DTG_TC_DTG					0x04
+#define DCSS_DTG_TC_DISP_TOP				0x08
+#define DCSS_DTG_TC_DISP_BOT				0x0C
+#define DCSS_DTG_TC_CH1_TOP				0x10
+#define DCSS_DTG_TC_CH1_BOT				0x14
+#define DCSS_DTG_TC_CH2_TOP				0x18
+#define DCSS_DTG_TC_CH2_BOT				0x1C
+#define DCSS_DTG_TC_CH3_TOP				0x20
+#define DCSS_DTG_TC_CH3_BOT				0x24
+#define   TC_X_POS					0
+#define   TC_X_MASK					GENMASK(12, 0)
+#define   TC_Y_POS					16
+#define   TC_Y_MASK					GENMASK(28, 16)
+#define DCSS_DTG_TC_CTXLD				0x28
+#define   TC_CTXLD_DB_Y_POS				0
+#define   TC_CTXLD_DB_Y_MASK				GENMASK(12, 0)
+#define   TC_CTXLD_SB_Y_POS				16
+#define   TC_CTXLD_SB_Y_MASK				GENMASK(28, 16)
+#define DCSS_DTG_TC_CH1_BKRND				0x2C
+#define DCSS_DTG_TC_CH2_BKRND				0x30
+#define   BKRND_R_Y_COMP_POS				20
+#define   BKRND_R_Y_COMP_MASK				GENMASK(29, 20)
+#define   BKRND_G_U_COMP_POS				10
+#define   BKRND_G_U_COMP_MASK				GENMASK(19, 10)
+#define   BKRND_B_V_COMP_POS				0
+#define   BKRND_B_V_COMP_MASK				GENMASK(9, 0)
+#define DCSS_DTG_BLENDER_DBY_RANGEINV			0x38
+#define DCSS_DTG_BLENDER_DBY_RANGEMIN			0x3C
+#define DCSS_DTG_BLENDER_DBY_BDP			0x40
+#define DCSS_DTG_BLENDER_BKRND_I			0x44
+#define DCSS_DTG_BLENDER_BKRND_P			0x48
+#define DCSS_DTG_BLENDER_BKRND_T			0x4C
+#define DCSS_DTG_LINE0_INT				0x50
+#define DCSS_DTG_LINE1_INT				0x54
+#define DCSS_DTG_BG_ALPHA_DEFAULT			0x58
+#define DCSS_DTG_INT_STATUS				0x5C
+#define DCSS_DTG_INT_CONTROL				0x60
+#define DCSS_DTG_TC_CH3_BKRND				0x64
+#define DCSS_DTG_INT_MASK				0x68
+#define   LINE0_IRQ					BIT(0)
+#define   LINE1_IRQ					BIT(1)
+#define   LINE2_IRQ					BIT(2)
+#define   LINE3_IRQ					BIT(3)
+#define DCSS_DTG_LINE2_INT				0x6C
+#define DCSS_DTG_LINE3_INT				0x70
+#define DCSS_DTG_DBY_OL					0x74
+#define DCSS_DTG_DBY_BL					0x78
+#define DCSS_DTG_DBY_EL					0x7C
+
+struct dcss_dtg {
+	struct device *dev;
+	struct dcss_ctxld *ctxld;
+	void __iomem *base_reg;
+	u32 base_ofs;
+
+	u32 ctx_id;
+
+	bool in_use;
+
+	u32 dis_ulc_x;
+	u32 dis_ulc_y;
+
+	u32 control_status;
+	u32 alpha;
+	u32 alpha_cfg;
+
+	int ctxld_kick_irq;
+	bool ctxld_kick_irq_en;
+};
+
+static void dcss_dtg_write(struct dcss_dtg *dtg, u32 val, u32 ofs)
+{
+	if (!dtg->in_use)
+		dcss_writel(val, dtg->base_reg + ofs);
+
+	dcss_ctxld_write(dtg->ctxld, dtg->ctx_id,
+			 val, dtg->base_ofs + ofs);
+}
+
+static irqreturn_t dcss_dtg_irq_handler(int irq, void *data)
+{
+	struct dcss_dtg *dtg = data;
+	u32 status;
+
+	status = dcss_readl(dtg->base_reg + DCSS_DTG_INT_STATUS);
+
+	if (!(status & LINE0_IRQ))
+		return IRQ_NONE;
+
+	dcss_ctxld_kick(dtg->ctxld);
+
+	dcss_writel(status & LINE0_IRQ, dtg->base_reg + DCSS_DTG_INT_CONTROL);
+
+	return IRQ_HANDLED;
+}
+
+static int dcss_dtg_irq_config(struct dcss_dtg *dtg,
+			       struct platform_device *pdev)
+{
+	int ret;
+
+	dtg->ctxld_kick_irq = platform_get_irq_byname(pdev, "ctxld_kick");
+	if (dtg->ctxld_kick_irq < 0)
+		return dtg->ctxld_kick_irq;
+
+	dcss_update(0, LINE0_IRQ | LINE1_IRQ,
+		    dtg->base_reg + DCSS_DTG_INT_MASK);
+
+	ret = request_irq(dtg->ctxld_kick_irq, dcss_dtg_irq_handler,
+			  0, "dcss_ctxld_kick", dtg);
+	if (ret) {
+		dev_err(dtg->dev, "dtg: irq request failed.\n");
+		return ret;
+	}
+
+	disable_irq(dtg->ctxld_kick_irq);
+
+	dtg->ctxld_kick_irq_en = false;
+
+	return 0;
+}
+
+int dcss_dtg_init(struct dcss_dev *dcss, unsigned long dtg_base)
+{
+	int ret = 0;
+	struct dcss_dtg *dtg;
+
+	dtg = kzalloc(sizeof(*dtg), GFP_KERNEL);
+	if (!dtg)
+		return -ENOMEM;
+
+	dcss->dtg = dtg;
+	dtg->dev = dcss->dev;
+	dtg->ctxld = dcss->ctxld;
+
+	dtg->base_reg = ioremap(dtg_base, SZ_4K);
+	if (!dtg->base_reg) {
+		dev_err(dcss->dev, "dtg: unable to remap dtg base\n");
+		ret = -ENOMEM;
+		goto err_ioremap;
+	}
+
+	dtg->base_ofs = dtg_base;
+	dtg->ctx_id = CTX_DB;
+
+	dtg->alpha = 255;
+
+	dtg->control_status |= OVL_DATA_MODE | BLENDER_VIDEO_ALPHA_SEL |
+		((dtg->alpha << DEFAULT_FG_ALPHA_POS) & DEFAULT_FG_ALPHA_MASK);
+
+	ret = dcss_dtg_irq_config(dtg, to_platform_device(dcss->dev));
+	if (ret)
+		goto err_irq;
+
+	return 0;
+
+err_irq:
+	iounmap(dtg->base_reg);
+
+err_ioremap:
+	kfree(dtg);
+
+	return ret;
+}
+
+void dcss_dtg_exit(struct dcss_dtg *dtg)
+{
+	free_irq(dtg->ctxld_kick_irq, dtg);
+
+	if (dtg->base_reg)
+		iounmap(dtg->base_reg);
+
+	kfree(dtg);
+}
+
+void dcss_dtg_sync_set(struct dcss_dtg *dtg, struct videomode *vm)
+{
+	struct dcss_dev *dcss = dcss_drv_dev_to_dcss(dtg->dev);
+	u16 dtg_lrc_x, dtg_lrc_y;
+	u16 dis_ulc_x, dis_ulc_y;
+	u16 dis_lrc_x, dis_lrc_y;
+	u32 sb_ctxld_trig, db_ctxld_trig;
+	u32 pixclock = vm->pixelclock;
+	u32 actual_clk;
+
+	dtg_lrc_x = vm->hfront_porch + vm->hback_porch + vm->hsync_len +
+		    vm->hactive - 1;
+	dtg_lrc_y = vm->vfront_porch + vm->vback_porch + vm->vsync_len +
+		    vm->vactive - 1;
+	dis_ulc_x = vm->hsync_len + vm->hback_porch - 1;
+	dis_ulc_y = vm->vsync_len + vm->vfront_porch + vm->vback_porch - 1;
+	dis_lrc_x = vm->hsync_len + vm->hback_porch + vm->hactive - 1;
+	dis_lrc_y = vm->vsync_len + vm->vfront_porch + vm->vback_porch +
+		    vm->vactive - 1;
+
+	clk_disable_unprepare(dcss->pix_clk);
+	clk_set_rate(dcss->pix_clk, vm->pixelclock);
+	clk_prepare_enable(dcss->pix_clk);
+
+	actual_clk = clk_get_rate(dcss->pix_clk);
+	if (pixclock != actual_clk) {
+		dev_info(dtg->dev,
+			 "Pixel clock set to %u kHz instead of %u kHz.\n",
+			 (actual_clk / 1000), (pixclock / 1000));
+	}
+
+	dcss_dtg_write(dtg, ((dtg_lrc_y << TC_Y_POS) | dtg_lrc_x),
+		       DCSS_DTG_TC_DTG);
+	dcss_dtg_write(dtg, ((dis_ulc_y << TC_Y_POS) | dis_ulc_x),
+		       DCSS_DTG_TC_DISP_TOP);
+	dcss_dtg_write(dtg, ((dis_lrc_y << TC_Y_POS) | dis_lrc_x),
+		       DCSS_DTG_TC_DISP_BOT);
+
+	dtg->dis_ulc_x = dis_ulc_x;
+	dtg->dis_ulc_y = dis_ulc_y;
+
+	sb_ctxld_trig = ((0 * dis_lrc_y / 100) << TC_CTXLD_SB_Y_POS) &
+							TC_CTXLD_SB_Y_MASK;
+	db_ctxld_trig = ((99 * dis_lrc_y / 100) << TC_CTXLD_DB_Y_POS) &
+							TC_CTXLD_DB_Y_MASK;
+
+	dcss_dtg_write(dtg, sb_ctxld_trig | db_ctxld_trig, DCSS_DTG_TC_CTXLD);
+
+	/* vblank trigger */
+	dcss_dtg_write(dtg, 0, DCSS_DTG_LINE1_INT);
+
+	/* CTXLD trigger */
+	dcss_dtg_write(dtg, ((90 * dis_lrc_y) / 100) << 16, DCSS_DTG_LINE0_INT);
+}
+
+void dcss_dtg_plane_pos_set(struct dcss_dtg *dtg, int ch_num,
+			    int px, int py, int pw, int ph)
+{
+	u16 p_ulc_x, p_ulc_y;
+	u16 p_lrc_x, p_lrc_y;
+
+	p_ulc_x = dtg->dis_ulc_x + px;
+	p_ulc_y = dtg->dis_ulc_y + py;
+	p_lrc_x = p_ulc_x + pw;
+	p_lrc_y = p_ulc_y + ph;
+
+	if (!px && !py && !pw && !ph) {
+		dcss_dtg_write(dtg, 0, DCSS_DTG_TC_CH1_TOP + 0x8 * ch_num);
+		dcss_dtg_write(dtg, 0, DCSS_DTG_TC_CH1_BOT + 0x8 * ch_num);
+	} else {
+		dcss_dtg_write(dtg, ((p_ulc_y << TC_Y_POS) | p_ulc_x),
+			       DCSS_DTG_TC_CH1_TOP + 0x8 * ch_num);
+		dcss_dtg_write(dtg, ((p_lrc_y << TC_Y_POS) | p_lrc_x),
+			       DCSS_DTG_TC_CH1_BOT + 0x8 * ch_num);
+	}
+}
+
+bool dcss_dtg_global_alpha_changed(struct dcss_dtg *dtg, int ch_num, int alpha)
+{
+	if (ch_num)
+		return false;
+
+	return alpha != dtg->alpha;
+}
+
+void dcss_dtg_plane_alpha_set(struct dcss_dtg *dtg, int ch_num,
+			      const struct drm_format_info *format, int alpha)
+{
+	/* we care about alpha only when channel 0 is concerned */
+	if (ch_num)
+		return;
+
+	/*
+	 * Use global alpha if pixel format does not have alpha channel or the
+	 * user explicitly chose to use global alpha (i.e. alpha is not OPAQUE).
+	 */
+	if (!format->has_alpha || alpha != 255)
+		dtg->alpha_cfg = (alpha << DEFAULT_FG_ALPHA_POS) & DEFAULT_FG_ALPHA_MASK;
+	else /* use per-pixel alpha otherwise */
+		dtg->alpha_cfg = CH1_ALPHA_SEL;
+
+	dtg->alpha = alpha;
+}
+
+void dcss_dtg_css_set(struct dcss_dtg *dtg)
+{
+	dtg->control_status |=
+			(0x5 << CSS_PIX_COMP_SWAP_POS) & CSS_PIX_COMP_SWAP_MASK;
+}
+
+void dcss_dtg_enable(struct dcss_dtg *dtg)
+{
+	dtg->control_status |= DTG_START;
+
+	dtg->control_status &= ~(CH1_ALPHA_SEL | DEFAULT_FG_ALPHA_MASK);
+	dtg->control_status |= dtg->alpha_cfg;
+
+	dcss_dtg_write(dtg, dtg->control_status, DCSS_DTG_TC_CONTROL_STATUS);
+
+	dtg->in_use = true;
+}
+
+void dcss_dtg_shutoff(struct dcss_dtg *dtg)
+{
+	dtg->control_status &= ~DTG_START;
+
+	dcss_writel(dtg->control_status,
+		    dtg->base_reg + DCSS_DTG_TC_CONTROL_STATUS);
+
+	dtg->in_use = false;
+}
+
+bool dcss_dtg_is_enabled(struct dcss_dtg *dtg)
+{
+	return dtg->in_use;
+}
+
+void dcss_dtg_ch_enable(struct dcss_dtg *dtg, int ch_num, bool en)
+{
+	u32 ch_en_map[] = {CH1_EN, CH2_EN, CH3_EN};
+	u32 control_status;
+
+	control_status = dtg->control_status & ~ch_en_map[ch_num];
+	control_status |= en ? ch_en_map[ch_num] : 0;
+
+	control_status &= ~(CH1_ALPHA_SEL | DEFAULT_FG_ALPHA_MASK);
+	control_status |= dtg->alpha_cfg;
+
+	if (dtg->control_status != control_status)
+		dcss_dtg_write(dtg, control_status, DCSS_DTG_TC_CONTROL_STATUS);
+
+	dtg->control_status = control_status;
+}
+
+void dcss_dtg_vblank_irq_enable(struct dcss_dtg *dtg, bool en)
+{
+	u32 status;
+	u32 mask = en ? LINE1_IRQ : 0;
+
+	if (en) {
+		status = dcss_readl(dtg->base_reg + DCSS_DTG_INT_STATUS);
+		dcss_writel(status & LINE1_IRQ,
+			    dtg->base_reg + DCSS_DTG_INT_CONTROL);
+	}
+
+	dcss_update(mask, LINE1_IRQ, dtg->base_reg + DCSS_DTG_INT_MASK);
+}
+
+void dcss_dtg_ctxld_kick_irq_enable(struct dcss_dtg *dtg, bool en)
+{
+	u32 status;
+	u32 mask = en ? LINE0_IRQ : 0;
+
+	if (en) {
+		status = dcss_readl(dtg->base_reg + DCSS_DTG_INT_STATUS);
+
+		if (!dtg->ctxld_kick_irq_en) {
+			dcss_writel(status & LINE0_IRQ,
+				    dtg->base_reg + DCSS_DTG_INT_CONTROL);
+			enable_irq(dtg->ctxld_kick_irq);
+			dtg->ctxld_kick_irq_en = true;
+			dcss_update(mask, LINE0_IRQ,
+				    dtg->base_reg + DCSS_DTG_INT_MASK);
+		}
+
+		return;
+	}
+
+	if (!dtg->ctxld_kick_irq_en)
+		return;
+
+	disable_irq_nosync(dtg->ctxld_kick_irq);
+	dtg->ctxld_kick_irq_en = false;
+
+	dcss_update(mask, LINE0_IRQ, dtg->base_reg + DCSS_DTG_INT_MASK);
+}
+
+void dcss_dtg_vblank_irq_clear(struct dcss_dtg *dtg)
+{
+	dcss_update(LINE1_IRQ, LINE1_IRQ, dtg->base_reg + DCSS_DTG_INT_CONTROL);
+}
+
+bool dcss_dtg_vblank_irq_valid(struct dcss_dtg *dtg)
+{
+	return !!(dcss_readl(dtg->base_reg + DCSS_DTG_INT_STATUS) & LINE1_IRQ);
+}
+
diff --git a/drivers/gpu/drm/imx/dcss/dcss-kms.c b/drivers/gpu/drm/imx/dcss/dcss-kms.c
new file mode 100644
index 000000000000..135a62366ab8
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-kms.c
@@ -0,0 +1,198 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_vblank.h>
+
+#include "dcss-dev.h"
+#include "dcss-kms.h"
+
+DEFINE_DRM_GEM_CMA_FOPS(dcss_cma_fops);
+
+static const struct drm_mode_config_funcs dcss_drm_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create,
+	.output_poll_changed = drm_fb_helper_output_poll_changed,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
+static struct drm_driver dcss_kms_driver = {
+	.driver_features	= DRIVER_MODESET | DRIVER_GEM | DRIVER_ATOMIC,
+	.gem_free_object_unlocked = drm_gem_cma_free_object,
+	.gem_vm_ops		= &drm_gem_cma_vm_ops,
+	.dumb_create		= drm_gem_cma_dumb_create,
+
+	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
+	.gem_prime_import	= drm_gem_prime_import,
+	.gem_prime_export	= drm_gem_prime_export,
+	.gem_prime_get_sg_table	= drm_gem_cma_prime_get_sg_table,
+	.gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table,
+	.gem_prime_vmap		= drm_gem_cma_prime_vmap,
+	.gem_prime_vunmap	= drm_gem_cma_prime_vunmap,
+	.gem_prime_mmap		= drm_gem_cma_prime_mmap,
+	.fops			= &dcss_cma_fops,
+	.name			= "imx-dcss",
+	.desc			= "i.MX8MQ Display Subsystem",
+	.date			= "20190917",
+	.major			= 1,
+	.minor			= 0,
+	.patchlevel		= 0,
+};
+
+static const struct drm_mode_config_helper_funcs dcss_mode_config_helpers = {
+	.atomic_commit_tail = drm_atomic_helper_commit_tail_rpm,
+};
+
+static void dcss_kms_mode_config_init(struct dcss_kms_dev *kms)
+{
+	struct drm_mode_config *config = &kms->base.mode_config;
+
+	drm_mode_config_init(&kms->base);
+
+	config->min_width = 1;
+	config->min_height = 1;
+	config->max_width = 4096;
+	config->max_height = 4096;
+	config->allow_fb_modifiers = true;
+	config->normalize_zpos = true;
+
+	config->funcs = &dcss_drm_mode_config_funcs;
+	config->helper_private = &dcss_mode_config_helpers;
+}
+
+static const struct drm_encoder_funcs dcss_kms_simple_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+static int dcss_kms_bridge_connector_init(struct dcss_kms_dev *kms)
+{
+	struct drm_device *ddev = &kms->base;
+	struct drm_encoder *encoder = &kms->encoder;
+	struct drm_crtc *crtc = (struct drm_crtc *)&kms->crtc;
+	struct drm_panel *panel;
+	struct drm_bridge *bridge;
+	int ret;
+
+	ret = drm_of_find_panel_or_bridge(ddev->dev->of_node, 0, 0,
+					  &panel, &bridge);
+	if (ret)
+		return ret;
+
+	if (!bridge) {
+		dev_err(ddev->dev, "No bridge found %d.\n", ret);
+		return -ENODEV;
+	}
+
+	encoder->possible_crtcs = drm_crtc_mask(crtc);
+
+	ret = drm_encoder_init(&kms->base, encoder,
+			       &dcss_kms_simple_encoder_funcs,
+			       DRM_MODE_ENCODER_NONE, NULL);
+	if (ret) {
+		dev_err(ddev->dev, "Failed initializing encoder %d.\n", ret);
+		return ret;
+	}
+
+	ret = drm_bridge_attach(encoder, bridge, NULL,
+				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	if (ret < 0) {
+		dev_err(ddev->dev, "Unable to attach bridge %pOF\n",
+			bridge->of_node);
+		return ret;
+	}
+
+	kms->connector = drm_bridge_connector_init(ddev, encoder);
+	if (IS_ERR(kms->connector)) {
+		dev_err(ddev->dev, "Unable to create bridge connector.\n");
+		return PTR_ERR(kms->connector);
+	}
+
+	drm_connector_attach_encoder(kms->connector, encoder);
+
+	return 0;
+}
+
+struct dcss_kms_dev *dcss_kms_attach(struct dcss_dev *dcss)
+{
+	struct dcss_kms_dev *kms;
+	struct drm_device *drm;
+	struct dcss_crtc *crtc;
+	int ret;
+
+	kms = devm_drm_dev_alloc(dcss->dev, &dcss_kms_driver,
+				 struct dcss_kms_dev, base);
+	if (IS_ERR(kms))
+		return kms;
+
+	drm = &kms->base;
+	crtc = &kms->crtc;
+
+	drm->dev_private = dcss;
+
+	dcss_kms_mode_config_init(kms);
+
+	ret = drm_vblank_init(drm, 1);
+	if (ret)
+		goto cleanup_mode_config;
+
+	drm->irq_enabled = true;
+
+	ret = dcss_kms_bridge_connector_init(kms);
+	if (ret)
+		goto cleanup_mode_config;
+
+	ret = dcss_crtc_init(crtc, drm);
+	if (ret)
+		goto cleanup_mode_config;
+
+	drm_mode_config_reset(drm);
+
+	drm_kms_helper_poll_init(drm);
+
+	drm_bridge_connector_enable_hpd(kms->connector);
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		goto cleanup_crtc;
+
+	drm_fbdev_generic_setup(drm, 32);
+
+	return kms;
+
+cleanup_crtc:
+	drm_bridge_connector_disable_hpd(kms->connector);
+	drm_kms_helper_poll_fini(drm);
+	dcss_crtc_deinit(crtc, drm);
+
+cleanup_mode_config:
+	drm_mode_config_cleanup(drm);
+	drm->dev_private = NULL;
+
+	return ERR_PTR(ret);
+}
+
+void dcss_kms_detach(struct dcss_kms_dev *kms)
+{
+	struct drm_device *drm = &kms->base;
+
+	drm_dev_unregister(drm);
+	drm_bridge_connector_disable_hpd(kms->connector);
+	drm_kms_helper_poll_fini(drm);
+	drm_atomic_helper_shutdown(drm);
+	drm_crtc_vblank_off(&kms->crtc.base);
+	drm->irq_enabled = false;
+	drm_mode_config_cleanup(drm);
+	dcss_crtc_deinit(&kms->crtc, drm);
+	drm->dev_private = NULL;
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-kms.h b/drivers/gpu/drm/imx/dcss/dcss-kms.h
new file mode 100644
index 000000000000..dfe5dd99eea3
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-kms.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2019 NXP.
+ */
+
+#ifndef _DCSS_KMS_H_
+#define _DCSS_KMS_H_
+
+#include <drm/drm_encoder.h>
+
+struct dcss_plane {
+	struct drm_plane base;
+
+	int ch_num;
+};
+
+struct dcss_crtc {
+	struct drm_crtc		base;
+	struct drm_crtc_state	*state;
+
+	struct dcss_plane	*plane[3];
+
+	int			irq;
+
+	bool disable_ctxld_kick_irq;
+};
+
+struct dcss_kms_dev {
+	struct drm_device base;
+	struct dcss_crtc crtc;
+	struct drm_encoder encoder;
+	struct drm_connector *connector;
+};
+
+struct dcss_kms_dev *dcss_kms_attach(struct dcss_dev *dcss);
+void dcss_kms_detach(struct dcss_kms_dev *kms);
+int dcss_crtc_init(struct dcss_crtc *crtc, struct drm_device *drm);
+void dcss_crtc_deinit(struct dcss_crtc *crtc, struct drm_device *drm);
+struct dcss_plane *dcss_plane_init(struct drm_device *drm,
+				   unsigned int possible_crtcs,
+				   enum drm_plane_type type,
+				   unsigned int zpos);
+
+#endif
diff --git a/drivers/gpu/drm/imx/dcss/dcss-plane.c b/drivers/gpu/drm/imx/dcss/dcss-plane.c
new file mode 100644
index 000000000000..961d671f171b
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-plane.c
@@ -0,0 +1,405 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+
+#include "dcss-dev.h"
+#include "dcss-kms.h"
+
+static const u32 dcss_common_formats[] = {
+	/* RGB */
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_RGBX8888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_BGRX8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_RGBX1010102,
+	DRM_FORMAT_BGRX1010102,
+	DRM_FORMAT_ARGB2101010,
+	DRM_FORMAT_ABGR2101010,
+	DRM_FORMAT_RGBA1010102,
+	DRM_FORMAT_BGRA1010102,
+};
+
+static const u64 dcss_video_format_modifiers[] = {
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID,
+};
+
+static const u64 dcss_graphics_format_modifiers[] = {
+	DRM_FORMAT_MOD_VIVANTE_TILED,
+	DRM_FORMAT_MOD_VIVANTE_SUPER_TILED,
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID,
+};
+
+static inline struct dcss_plane *to_dcss_plane(struct drm_plane *p)
+{
+	return container_of(p, struct dcss_plane, base);
+}
+
+static inline bool dcss_plane_fb_is_linear(const struct drm_framebuffer *fb)
+{
+	return ((fb->flags & DRM_MODE_FB_MODIFIERS) == 0) ||
+	       ((fb->flags & DRM_MODE_FB_MODIFIERS) != 0 &&
+		fb->modifier == DRM_FORMAT_MOD_LINEAR);
+}
+
+static void dcss_plane_destroy(struct drm_plane *plane)
+{
+	struct dcss_plane *dcss_plane = container_of(plane, struct dcss_plane,
+						     base);
+
+	drm_plane_cleanup(plane);
+	kfree(dcss_plane);
+}
+
+static bool dcss_plane_format_mod_supported(struct drm_plane *plane,
+					    u32 format,
+					    u64 modifier)
+{
+	switch (plane->type) {
+	case DRM_PLANE_TYPE_PRIMARY:
+		switch (format) {
+		case DRM_FORMAT_ARGB8888:
+		case DRM_FORMAT_XRGB8888:
+		case DRM_FORMAT_ARGB2101010:
+			return modifier == DRM_FORMAT_MOD_LINEAR ||
+			       modifier == DRM_FORMAT_MOD_VIVANTE_TILED ||
+			       modifier == DRM_FORMAT_MOD_VIVANTE_SUPER_TILED;
+		default:
+			return modifier == DRM_FORMAT_MOD_LINEAR;
+		}
+		break;
+	case DRM_PLANE_TYPE_OVERLAY:
+		return modifier == DRM_FORMAT_MOD_LINEAR;
+	default:
+		return false;
+	}
+}
+
+static const struct drm_plane_funcs dcss_plane_funcs = {
+	.update_plane		= drm_atomic_helper_update_plane,
+	.disable_plane		= drm_atomic_helper_disable_plane,
+	.destroy		= dcss_plane_destroy,
+	.reset			= drm_atomic_helper_plane_reset,
+	.atomic_duplicate_state	= drm_atomic_helper_plane_duplicate_state,
+	.atomic_destroy_state	= drm_atomic_helper_plane_destroy_state,
+	.format_mod_supported	= dcss_plane_format_mod_supported,
+};
+
+static bool dcss_plane_can_rotate(const struct drm_format_info *format,
+				  bool mod_present, u64 modifier,
+				  unsigned int rotation)
+{
+	bool linear_format = !mod_present ||
+			     (mod_present && modifier == DRM_FORMAT_MOD_LINEAR);
+	u32 supported_rotation = DRM_MODE_ROTATE_0;
+
+	if (!format->is_yuv && linear_format)
+		supported_rotation = DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_180 |
+				     DRM_MODE_REFLECT_MASK;
+	else if (!format->is_yuv &&
+		 modifier == DRM_FORMAT_MOD_VIVANTE_TILED)
+		supported_rotation = DRM_MODE_ROTATE_MASK |
+				     DRM_MODE_REFLECT_MASK;
+	else if (format->is_yuv && linear_format &&
+		 (format->format == DRM_FORMAT_NV12 ||
+		  format->format == DRM_FORMAT_NV21))
+		supported_rotation = DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_180 |
+				     DRM_MODE_REFLECT_MASK;
+
+	return !!(rotation & supported_rotation);
+}
+
+static bool dcss_plane_is_source_size_allowed(u16 src_w, u16 src_h, u32 pix_fmt)
+{
+	if (src_w < 64 &&
+	    (pix_fmt == DRM_FORMAT_NV12 || pix_fmt == DRM_FORMAT_NV21))
+		return false;
+	else if (src_w < 32 &&
+		 (pix_fmt == DRM_FORMAT_UYVY || pix_fmt == DRM_FORMAT_VYUY ||
+		  pix_fmt == DRM_FORMAT_YUYV || pix_fmt == DRM_FORMAT_YVYU))
+		return false;
+
+	return src_w >= 16 && src_h >= 8;
+}
+
+static int dcss_plane_atomic_check(struct drm_plane *plane,
+				   struct drm_plane_state *state)
+{
+	struct dcss_plane *dcss_plane = to_dcss_plane(plane);
+	struct dcss_dev *dcss = plane->dev->dev_private;
+	struct drm_framebuffer *fb = state->fb;
+	bool is_primary_plane = plane->type == DRM_PLANE_TYPE_PRIMARY;
+	struct drm_gem_cma_object *cma_obj;
+	struct drm_crtc_state *crtc_state;
+	int hdisplay, vdisplay;
+	int min, max;
+	int ret;
+
+	if (!fb || !state->crtc)
+		return 0;
+
+	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
+	WARN_ON(!cma_obj);
+
+	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
+							state->crtc);
+
+	hdisplay = crtc_state->adjusted_mode.hdisplay;
+	vdisplay = crtc_state->adjusted_mode.vdisplay;
+
+	if (!dcss_plane_is_source_size_allowed(state->src_w >> 16,
+					       state->src_h >> 16,
+					       fb->format->format)) {
+		DRM_DEBUG_KMS("Source plane size is not allowed!\n");
+		return -EINVAL;
+	}
+
+	dcss_scaler_get_min_max_ratios(dcss->scaler, dcss_plane->ch_num,
+				       &min, &max);
+
+	ret = drm_atomic_helper_check_plane_state(state, crtc_state,
+						  min, max, !is_primary_plane,
+						  false);
+	if (ret)
+		return ret;
+
+	if (!state->visible)
+		return 0;
+
+	if (!dcss_plane_can_rotate(fb->format,
+				   !!(fb->flags & DRM_MODE_FB_MODIFIERS),
+				   fb->modifier,
+				   state->rotation)) {
+		DRM_DEBUG_KMS("requested rotation is not allowed!\n");
+		return -EINVAL;
+	}
+
+	if ((state->crtc_x < 0 || state->crtc_y < 0 ||
+	     state->crtc_x + state->crtc_w > hdisplay ||
+	     state->crtc_y + state->crtc_h > vdisplay) &&
+	    !dcss_plane_fb_is_linear(fb)) {
+		DRM_DEBUG_KMS("requested cropping operation is not allowed!\n");
+		return -EINVAL;
+	}
+
+	if ((fb->flags & DRM_MODE_FB_MODIFIERS) &&
+	    !plane->funcs->format_mod_supported(plane,
+				fb->format->format,
+				fb->modifier)) {
+		DRM_DEBUG_KMS("Invalid modifier: %llx", fb->modifier);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void dcss_plane_atomic_set_base(struct dcss_plane *dcss_plane)
+{
+	struct drm_plane *plane = &dcss_plane->base;
+	struct drm_plane_state *state = plane->state;
+	struct dcss_dev *dcss = plane->dev->dev_private;
+	struct drm_framebuffer *fb = state->fb;
+	const struct drm_format_info *format = fb->format;
+	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
+	unsigned long p1_ba = 0, p2_ba = 0;
+
+	if (!format->is_yuv ||
+	    format->format == DRM_FORMAT_NV12 ||
+	    format->format == DRM_FORMAT_NV21)
+		p1_ba = cma_obj->paddr + fb->offsets[0] +
+			fb->pitches[0] * (state->src.y1 >> 16) +
+			format->char_per_block[0] * (state->src.x1 >> 16);
+	else if (format->format == DRM_FORMAT_UYVY ||
+		 format->format == DRM_FORMAT_VYUY ||
+		 format->format == DRM_FORMAT_YUYV ||
+		 format->format == DRM_FORMAT_YVYU)
+		p1_ba = cma_obj->paddr + fb->offsets[0] +
+			fb->pitches[0] * (state->src.y1 >> 16) +
+			2 * format->char_per_block[0] * (state->src.x1 >> 17);
+
+	if (format->format == DRM_FORMAT_NV12 ||
+	    format->format == DRM_FORMAT_NV21)
+		p2_ba = cma_obj->paddr + fb->offsets[1] +
+			(((fb->pitches[1] >> 1) * (state->src.y1 >> 17) +
+			(state->src.x1 >> 17)) << 1);
+
+	dcss_dpr_addr_set(dcss->dpr, dcss_plane->ch_num, p1_ba, p2_ba,
+			  fb->pitches[0]);
+}
+
+static bool dcss_plane_needs_setup(struct drm_plane_state *state,
+				   struct drm_plane_state *old_state)
+{
+	struct drm_framebuffer *fb = state->fb;
+	struct drm_framebuffer *old_fb = old_state->fb;
+
+	return state->crtc_x != old_state->crtc_x ||
+	       state->crtc_y != old_state->crtc_y ||
+	       state->crtc_w != old_state->crtc_w ||
+	       state->crtc_h != old_state->crtc_h ||
+	       state->src_x  != old_state->src_x  ||
+	       state->src_y  != old_state->src_y  ||
+	       state->src_w  != old_state->src_w  ||
+	       state->src_h  != old_state->src_h  ||
+	       fb->format->format != old_fb->format->format ||
+	       fb->modifier  != old_fb->modifier ||
+	       state->rotation != old_state->rotation;
+}
+
+static void dcss_plane_atomic_update(struct drm_plane *plane,
+				     struct drm_plane_state *old_state)
+{
+	struct drm_plane_state *state = plane->state;
+	struct dcss_plane *dcss_plane = to_dcss_plane(plane);
+	struct dcss_dev *dcss = plane->dev->dev_private;
+	struct drm_framebuffer *fb = state->fb;
+	u32 pixel_format;
+	struct drm_crtc_state *crtc_state;
+	bool modifiers_present;
+	u32 src_w, src_h, dst_w, dst_h;
+	struct drm_rect src, dst;
+	bool enable = true;
+
+	if (!fb || !state->crtc || !state->visible)
+		return;
+
+	pixel_format = state->fb->format->format;
+	crtc_state = state->crtc->state;
+	modifiers_present = !!(fb->flags & DRM_MODE_FB_MODIFIERS);
+
+	if (old_state->fb && !drm_atomic_crtc_needs_modeset(crtc_state) &&
+	    !dcss_plane_needs_setup(state, old_state)) {
+		dcss_plane_atomic_set_base(dcss_plane);
+		return;
+	}
+
+	src = plane->state->src;
+	dst = plane->state->dst;
+
+	/*
+	 * The width and height after clipping.
+	 */
+	src_w = drm_rect_width(&src) >> 16;
+	src_h = drm_rect_height(&src) >> 16;
+	dst_w = drm_rect_width(&dst);
+	dst_h = drm_rect_height(&dst);
+
+	if (plane->type == DRM_PLANE_TYPE_OVERLAY &&
+	    modifiers_present && fb->modifier == DRM_FORMAT_MOD_LINEAR)
+		modifiers_present = false;
+
+	dcss_dpr_format_set(dcss->dpr, dcss_plane->ch_num, state->fb->format,
+			    modifiers_present ? fb->modifier :
+						DRM_FORMAT_MOD_LINEAR);
+	dcss_dpr_set_res(dcss->dpr, dcss_plane->ch_num, src_w, src_h);
+	dcss_dpr_set_rotation(dcss->dpr, dcss_plane->ch_num,
+			      state->rotation);
+
+	dcss_plane_atomic_set_base(dcss_plane);
+
+	dcss_scaler_setup(dcss->scaler, dcss_plane->ch_num,
+			  state->fb->format, src_w, src_h,
+			  dst_w, dst_h,
+			  drm_mode_vrefresh(&crtc_state->mode));
+
+	dcss_dtg_plane_pos_set(dcss->dtg, dcss_plane->ch_num,
+			       dst.x1, dst.y1, dst_w, dst_h);
+	dcss_dtg_plane_alpha_set(dcss->dtg, dcss_plane->ch_num,
+				 fb->format, state->alpha >> 8);
+
+	if (!dcss_plane->ch_num && (state->alpha >> 8) == 0)
+		enable = false;
+
+	dcss_dpr_enable(dcss->dpr, dcss_plane->ch_num, enable);
+	dcss_scaler_ch_enable(dcss->scaler, dcss_plane->ch_num, enable);
+
+	if (!enable)
+		dcss_dtg_plane_pos_set(dcss->dtg, dcss_plane->ch_num,
+				       0, 0, 0, 0);
+
+	dcss_dtg_ch_enable(dcss->dtg, dcss_plane->ch_num, enable);
+}
+
+static void dcss_plane_atomic_disable(struct drm_plane *plane,
+				      struct drm_plane_state *old_state)
+{
+	struct dcss_plane *dcss_plane = to_dcss_plane(plane);
+	struct dcss_dev *dcss = plane->dev->dev_private;
+
+	dcss_dpr_enable(dcss->dpr, dcss_plane->ch_num, false);
+	dcss_scaler_ch_enable(dcss->scaler, dcss_plane->ch_num, false);
+	dcss_dtg_plane_pos_set(dcss->dtg, dcss_plane->ch_num, 0, 0, 0, 0);
+	dcss_dtg_ch_enable(dcss->dtg, dcss_plane->ch_num, false);
+}
+
+static const struct drm_plane_helper_funcs dcss_plane_helper_funcs = {
+	.prepare_fb = drm_gem_fb_prepare_fb,
+	.atomic_check = dcss_plane_atomic_check,
+	.atomic_update = dcss_plane_atomic_update,
+	.atomic_disable = dcss_plane_atomic_disable,
+};
+
+struct dcss_plane *dcss_plane_init(struct drm_device *drm,
+				   unsigned int possible_crtcs,
+				   enum drm_plane_type type,
+				   unsigned int zpos)
+{
+	struct dcss_plane *dcss_plane;
+	const u64 *format_modifiers = dcss_video_format_modifiers;
+	int ret;
+
+	if (zpos > 2)
+		return ERR_PTR(-EINVAL);
+
+	dcss_plane = kzalloc(sizeof(*dcss_plane), GFP_KERNEL);
+	if (!dcss_plane) {
+		DRM_ERROR("failed to allocate plane\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	if (type == DRM_PLANE_TYPE_PRIMARY)
+		format_modifiers = dcss_graphics_format_modifiers;
+
+	ret = drm_universal_plane_init(drm, &dcss_plane->base, possible_crtcs,
+				       &dcss_plane_funcs, dcss_common_formats,
+				       ARRAY_SIZE(dcss_common_formats),
+				       format_modifiers, type, NULL);
+	if (ret) {
+		DRM_ERROR("failed to initialize plane\n");
+		kfree(dcss_plane);
+		return ERR_PTR(ret);
+	}
+
+	drm_plane_helper_add(&dcss_plane->base, &dcss_plane_helper_funcs);
+
+	ret = drm_plane_create_zpos_immutable_property(&dcss_plane->base, zpos);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_plane_create_rotation_property(&dcss_plane->base,
+					   DRM_MODE_ROTATE_0,
+					   DRM_MODE_ROTATE_0   |
+					   DRM_MODE_ROTATE_90  |
+					   DRM_MODE_ROTATE_180 |
+					   DRM_MODE_ROTATE_270 |
+					   DRM_MODE_REFLECT_X  |
+					   DRM_MODE_REFLECT_Y);
+
+	dcss_plane->ch_num = zpos;
+
+	return dcss_plane;
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-scaler.c b/drivers/gpu/drm/imx/dcss/dcss-scaler.c
new file mode 100644
index 000000000000..cd21905de580
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-scaler.c
@@ -0,0 +1,826 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ *
+ * Scaling algorithms were contributed by Dzung Hoang <dzung.hoang@nxp.com>
+ */
+
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_SCALER_CTRL			0x00
+#define   SCALER_EN				BIT(0)
+#define   REPEAT_EN				BIT(4)
+#define   SCALE2MEM_EN				BIT(8)
+#define   MEM2OFIFO_EN				BIT(12)
+#define DCSS_SCALER_OFIFO_CTRL			0x04
+#define   OFIFO_LOW_THRES_POS			0
+#define   OFIFO_LOW_THRES_MASK			GENMASK(9, 0)
+#define   OFIFO_HIGH_THRES_POS			16
+#define   OFIFO_HIGH_THRES_MASK			GENMASK(25, 16)
+#define   UNDERRUN_DETECT_CLR			BIT(26)
+#define   LOW_THRES_DETECT_CLR			BIT(27)
+#define   HIGH_THRES_DETECT_CLR			BIT(28)
+#define   UNDERRUN_DETECT_EN			BIT(29)
+#define   LOW_THRES_DETECT_EN			BIT(30)
+#define   HIGH_THRES_DETECT_EN			BIT(31)
+#define DCSS_SCALER_SDATA_CTRL			0x08
+#define   YUV_EN				BIT(0)
+#define   RTRAM_8LINES				BIT(1)
+#define   Y_UV_BYTE_SWAP			BIT(4)
+#define   A2R10G10B10_FORMAT_POS		8
+#define   A2R10G10B10_FORMAT_MASK		GENMASK(11, 8)
+#define DCSS_SCALER_BIT_DEPTH			0x0C
+#define   LUM_BIT_DEPTH_POS			0
+#define   LUM_BIT_DEPTH_MASK			GENMASK(1, 0)
+#define   CHR_BIT_DEPTH_POS			4
+#define   CHR_BIT_DEPTH_MASK			GENMASK(5, 4)
+#define DCSS_SCALER_SRC_FORMAT			0x10
+#define DCSS_SCALER_DST_FORMAT			0x14
+#define   FORMAT_MASK				GENMASK(1, 0)
+#define DCSS_SCALER_SRC_LUM_RES			0x18
+#define DCSS_SCALER_SRC_CHR_RES			0x1C
+#define DCSS_SCALER_DST_LUM_RES			0x20
+#define DCSS_SCALER_DST_CHR_RES			0x24
+#define   WIDTH_POS				0
+#define   WIDTH_MASK				GENMASK(11, 0)
+#define   HEIGHT_POS				16
+#define   HEIGHT_MASK				GENMASK(27, 16)
+#define DCSS_SCALER_V_LUM_START			0x48
+#define   V_START_MASK				GENMASK(15, 0)
+#define DCSS_SCALER_V_LUM_INC			0x4C
+#define   V_INC_MASK				GENMASK(15, 0)
+#define DCSS_SCALER_H_LUM_START			0x50
+#define   H_START_MASK				GENMASK(18, 0)
+#define DCSS_SCALER_H_LUM_INC			0x54
+#define   H_INC_MASK				GENMASK(15, 0)
+#define DCSS_SCALER_V_CHR_START			0x58
+#define DCSS_SCALER_V_CHR_INC			0x5C
+#define DCSS_SCALER_H_CHR_START			0x60
+#define DCSS_SCALER_H_CHR_INC			0x64
+#define DCSS_SCALER_COEF_VLUM			0x80
+#define DCSS_SCALER_COEF_HLUM			0x140
+#define DCSS_SCALER_COEF_VCHR			0x200
+#define DCSS_SCALER_COEF_HCHR			0x300
+
+struct dcss_scaler_ch {
+	void __iomem *base_reg;
+	u32 base_ofs;
+	struct dcss_scaler *scl;
+
+	u32 sdata_ctrl;
+	u32 scaler_ctrl;
+
+	bool scaler_ctrl_chgd;
+
+	u32 c_vstart;
+	u32 c_hstart;
+};
+
+struct dcss_scaler {
+	struct device *dev;
+
+	struct dcss_ctxld *ctxld;
+	u32 ctx_id;
+
+	struct dcss_scaler_ch ch[3];
+};
+
+/* scaler coefficients generator */
+#define PSC_FRAC_BITS 30
+#define PSC_FRAC_SCALE BIT(PSC_FRAC_BITS)
+#define PSC_BITS_FOR_PHASE 4
+#define PSC_NUM_PHASES 16
+#define PSC_STORED_PHASES (PSC_NUM_PHASES / 2 + 1)
+#define PSC_NUM_TAPS 7
+#define PSC_NUM_TAPS_RGBA 5
+#define PSC_COEFF_PRECISION 10
+#define PSC_PHASE_FRACTION_BITS 13
+#define PSC_PHASE_MASK (PSC_NUM_PHASES - 1)
+#define PSC_Q_FRACTION 19
+#define PSC_Q_ROUND_OFFSET (1 << (PSC_Q_FRACTION - 1))
+
+/**
+ * mult_q() - Performs fixed-point multiplication.
+ * @A: multiplier
+ * @B: multiplicand
+ */
+static int mult_q(int A, int B)
+{
+	int result;
+	s64 temp;
+
+	temp = (int64_t)A * (int64_t)B;
+	temp += PSC_Q_ROUND_OFFSET;
+	result = (int)(temp >> PSC_Q_FRACTION);
+	return result;
+}
+
+/**
+ * div_q() - Performs fixed-point division.
+ * @A: dividend
+ * @B: divisor
+ */
+static int div_q(int A, int B)
+{
+	int result;
+	s64 temp;
+
+	temp = (int64_t)A << PSC_Q_FRACTION;
+	if ((temp >= 0 && B >= 0) || (temp < 0 && B < 0))
+		temp += B / 2;
+	else
+		temp -= B / 2;
+
+	result = (int)(temp / B);
+	return result;
+}
+
+/**
+ * exp_approx_q() - Compute approximation to exp(x) function using Taylor
+ *		    series.
+ * @x: fixed-point argument of exp function
+ */
+static int exp_approx_q(int x)
+{
+	int sum = 1 << PSC_Q_FRACTION;
+	int term = 1 << PSC_Q_FRACTION;
+
+	term = mult_q(term, div_q(x, 1 << PSC_Q_FRACTION));
+	sum += term;
+	term = mult_q(term, div_q(x, 2 << PSC_Q_FRACTION));
+	sum += term;
+	term = mult_q(term, div_q(x, 3 << PSC_Q_FRACTION));
+	sum += term;
+	term = mult_q(term, div_q(x, 4 << PSC_Q_FRACTION));
+	sum += term;
+
+	return sum;
+}
+
+/**
+ * dcss_scaler_gaussian_filter() - Generate gaussian prototype filter.
+ * @fc_q: fixed-point cutoff frequency normalized to range [0, 1]
+ * @use_5_taps: indicates whether to use 5 taps or 7 taps
+ * @coef: output filter coefficients
+ */
+static void dcss_scaler_gaussian_filter(int fc_q, bool use_5_taps,
+					bool phase0_identity,
+					int coef[][PSC_NUM_TAPS])
+{
+	int sigma_q, g0_q, g1_q, g2_q;
+	int tap_cnt1, tap_cnt2, tap_idx, phase_cnt;
+	int mid;
+	int phase;
+	int i;
+	int taps;
+
+	if (use_5_taps)
+		for (phase = 0; phase < PSC_STORED_PHASES; phase++) {
+			coef[phase][0] = 0;
+			coef[phase][PSC_NUM_TAPS - 1] = 0;
+		}
+
+	/* seed coefficient scanner */
+	taps = use_5_taps ? PSC_NUM_TAPS_RGBA : PSC_NUM_TAPS;
+	mid = (PSC_NUM_PHASES * taps) / 2 - 1;
+	phase_cnt = (PSC_NUM_PHASES * (PSC_NUM_TAPS + 1)) / 2;
+	tap_cnt1 = (PSC_NUM_PHASES * PSC_NUM_TAPS) / 2;
+	tap_cnt2 = (PSC_NUM_PHASES * PSC_NUM_TAPS) / 2;
+
+	/* seed gaussian filter generator */
+	sigma_q = div_q(PSC_Q_ROUND_OFFSET, fc_q);
+	g0_q = 1 << PSC_Q_FRACTION;
+	g1_q = exp_approx_q(div_q(-PSC_Q_ROUND_OFFSET,
+				  mult_q(sigma_q, sigma_q)));
+	g2_q = mult_q(g1_q, g1_q);
+	coef[phase_cnt & PSC_PHASE_MASK][tap_cnt1 >> PSC_BITS_FOR_PHASE] = g0_q;
+
+	for (i = 0; i < mid; i++) {
+		phase_cnt++;
+		tap_cnt1--;
+		tap_cnt2++;
+
+		g0_q = mult_q(g0_q, g1_q);
+		g1_q = mult_q(g1_q, g2_q);
+
+		if ((phase_cnt & PSC_PHASE_MASK) <= 8) {
+			tap_idx = tap_cnt1 >> PSC_BITS_FOR_PHASE;
+			coef[phase_cnt & PSC_PHASE_MASK][tap_idx] = g0_q;
+		}
+		if (((-phase_cnt) & PSC_PHASE_MASK) <= 8) {
+			tap_idx = tap_cnt2 >> PSC_BITS_FOR_PHASE;
+			coef[(-phase_cnt) & PSC_PHASE_MASK][tap_idx] = g0_q;
+		}
+	}
+
+	phase_cnt++;
+	tap_cnt1--;
+	coef[phase_cnt & PSC_PHASE_MASK][tap_cnt1 >> PSC_BITS_FOR_PHASE] = 0;
+
+	/* override phase 0 with identity filter if specified */
+	if (phase0_identity)
+		for (i = 0; i < PSC_NUM_TAPS; i++)
+			coef[0][i] = i == (PSC_NUM_TAPS >> 1) ?
+						(1 << PSC_COEFF_PRECISION) : 0;
+
+	/* normalize coef */
+	for (phase = 0; phase < PSC_STORED_PHASES; phase++) {
+		int sum = 0;
+		s64 ll_temp;
+
+		for (i = 0; i < PSC_NUM_TAPS; i++)
+			sum += coef[phase][i];
+		for (i = 0; i < PSC_NUM_TAPS; i++) {
+			ll_temp = coef[phase][i];
+			ll_temp <<= PSC_COEFF_PRECISION;
+			ll_temp += sum >> 1;
+			ll_temp /= sum;
+			coef[phase][i] = (int)ll_temp;
+		}
+	}
+}
+
+/**
+ * dcss_scaler_filter_design() - Compute filter coefficients using
+ *				 Gaussian filter.
+ * @src_length: length of input
+ * @dst_length: length of output
+ * @use_5_taps: 0 for 7 taps per phase, 1 for 5 taps
+ * @coef: output coefficients
+ */
+static void dcss_scaler_filter_design(int src_length, int dst_length,
+				      bool use_5_taps, bool phase0_identity,
+				      int coef[][PSC_NUM_TAPS])
+{
+	int fc_q;
+
+	/* compute cutoff frequency */
+	if (dst_length >= src_length)
+		fc_q = div_q(1, PSC_NUM_PHASES);
+	else
+		fc_q = div_q(dst_length, src_length * PSC_NUM_PHASES);
+
+	/* compute gaussian filter coefficients */
+	dcss_scaler_gaussian_filter(fc_q, use_5_taps, phase0_identity, coef);
+}
+
+static void dcss_scaler_write(struct dcss_scaler_ch *ch, u32 val, u32 ofs)
+{
+	struct dcss_scaler *scl = ch->scl;
+
+	dcss_ctxld_write(scl->ctxld, scl->ctx_id, val, ch->base_ofs + ofs);
+}
+
+static int dcss_scaler_ch_init_all(struct dcss_scaler *scl,
+				   unsigned long scaler_base)
+{
+	struct dcss_scaler_ch *ch;
+	int i;
+
+	for (i = 0; i < 3; i++) {
+		ch = &scl->ch[i];
+
+		ch->base_ofs = scaler_base + i * 0x400;
+
+		ch->base_reg = ioremap(ch->base_ofs, SZ_4K);
+		if (!ch->base_reg) {
+			dev_err(scl->dev, "scaler: unable to remap ch base\n");
+			return -ENOMEM;
+		}
+
+		ch->scl = scl;
+	}
+
+	return 0;
+}
+
+int dcss_scaler_init(struct dcss_dev *dcss, unsigned long scaler_base)
+{
+	struct dcss_scaler *scaler;
+
+	scaler = kzalloc(sizeof(*scaler), GFP_KERNEL);
+	if (!scaler)
+		return -ENOMEM;
+
+	dcss->scaler = scaler;
+	scaler->dev = dcss->dev;
+	scaler->ctxld = dcss->ctxld;
+	scaler->ctx_id = CTX_SB_HP;
+
+	if (dcss_scaler_ch_init_all(scaler, scaler_base)) {
+		int i;
+
+		for (i = 0; i < 3; i++) {
+			if (scaler->ch[i].base_reg)
+				iounmap(scaler->ch[i].base_reg);
+		}
+
+		kfree(scaler);
+
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void dcss_scaler_exit(struct dcss_scaler *scl)
+{
+	int ch_no;
+
+	for (ch_no = 0; ch_no < 3; ch_no++) {
+		struct dcss_scaler_ch *ch = &scl->ch[ch_no];
+
+		dcss_writel(0, ch->base_reg + DCSS_SCALER_CTRL);
+
+		if (ch->base_reg)
+			iounmap(ch->base_reg);
+	}
+
+	kfree(scl);
+}
+
+void dcss_scaler_ch_enable(struct dcss_scaler *scl, int ch_num, bool en)
+{
+	struct dcss_scaler_ch *ch = &scl->ch[ch_num];
+	u32 scaler_ctrl;
+
+	scaler_ctrl = en ? SCALER_EN | REPEAT_EN : 0;
+
+	if (en)
+		dcss_scaler_write(ch, ch->sdata_ctrl, DCSS_SCALER_SDATA_CTRL);
+
+	if (ch->scaler_ctrl != scaler_ctrl)
+		ch->scaler_ctrl_chgd = true;
+
+	ch->scaler_ctrl = scaler_ctrl;
+}
+
+static void dcss_scaler_yuv_enable(struct dcss_scaler_ch *ch, bool en)
+{
+	ch->sdata_ctrl &= ~YUV_EN;
+	ch->sdata_ctrl |= en ? YUV_EN : 0;
+}
+
+static void dcss_scaler_rtr_8lines_enable(struct dcss_scaler_ch *ch, bool en)
+{
+	ch->sdata_ctrl &= ~RTRAM_8LINES;
+	ch->sdata_ctrl |= en ? RTRAM_8LINES : 0;
+}
+
+static void dcss_scaler_bit_depth_set(struct dcss_scaler_ch *ch, int depth)
+{
+	u32 val;
+
+	val = depth == 30 ? 2 : 0;
+
+	dcss_scaler_write(ch,
+			  ((val << CHR_BIT_DEPTH_POS) & CHR_BIT_DEPTH_MASK) |
+			  ((val << LUM_BIT_DEPTH_POS) & LUM_BIT_DEPTH_MASK),
+			  DCSS_SCALER_BIT_DEPTH);
+}
+
+enum buffer_format {
+	BUF_FMT_YUV420,
+	BUF_FMT_YUV422,
+	BUF_FMT_ARGB8888_YUV444,
+};
+
+enum chroma_location {
+	PSC_LOC_HORZ_0_VERT_1_OVER_4 = 0,
+	PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_4 = 1,
+	PSC_LOC_HORZ_0_VERT_0 = 2,
+	PSC_LOC_HORZ_1_OVER_4_VERT_0 = 3,
+	PSC_LOC_HORZ_0_VERT_1_OVER_2 = 4,
+	PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_2 = 5
+};
+
+static void dcss_scaler_format_set(struct dcss_scaler_ch *ch,
+				   enum buffer_format src_fmt,
+				   enum buffer_format dst_fmt)
+{
+	dcss_scaler_write(ch, src_fmt, DCSS_SCALER_SRC_FORMAT);
+	dcss_scaler_write(ch, dst_fmt, DCSS_SCALER_DST_FORMAT);
+}
+
+static void dcss_scaler_res_set(struct dcss_scaler_ch *ch,
+				int src_xres, int src_yres,
+				int dst_xres, int dst_yres,
+				u32 pix_format, enum buffer_format dst_format)
+{
+	u32 lsrc_xres, lsrc_yres, csrc_xres, csrc_yres;
+	u32 ldst_xres, ldst_yres, cdst_xres, cdst_yres;
+	bool src_is_444 = true;
+
+	lsrc_xres = src_xres;
+	csrc_xres = src_xres;
+	lsrc_yres = src_yres;
+	csrc_yres = src_yres;
+	ldst_xres = dst_xres;
+	cdst_xres = dst_xres;
+	ldst_yres = dst_yres;
+	cdst_yres = dst_yres;
+
+	if (pix_format == DRM_FORMAT_UYVY || pix_format == DRM_FORMAT_VYUY ||
+	    pix_format == DRM_FORMAT_YUYV || pix_format == DRM_FORMAT_YVYU) {
+		csrc_xres >>= 1;
+		src_is_444 = false;
+	} else if (pix_format == DRM_FORMAT_NV12 ||
+		   pix_format == DRM_FORMAT_NV21) {
+		csrc_xres >>= 1;
+		csrc_yres >>= 1;
+		src_is_444 = false;
+	}
+
+	if (dst_format == BUF_FMT_YUV422)
+		cdst_xres >>= 1;
+
+	/* for 4:4:4 to 4:2:2 conversion, source height should be 1 less */
+	if (src_is_444 && dst_format == BUF_FMT_YUV422) {
+		lsrc_yres--;
+		csrc_yres--;
+	}
+
+	dcss_scaler_write(ch, (((lsrc_yres - 1) << HEIGHT_POS) & HEIGHT_MASK) |
+			       (((lsrc_xres - 1) << WIDTH_POS) & WIDTH_MASK),
+			  DCSS_SCALER_SRC_LUM_RES);
+	dcss_scaler_write(ch, (((csrc_yres - 1) << HEIGHT_POS) & HEIGHT_MASK) |
+			       (((csrc_xres - 1) << WIDTH_POS) & WIDTH_MASK),
+			  DCSS_SCALER_SRC_CHR_RES);
+	dcss_scaler_write(ch, (((ldst_yres - 1) << HEIGHT_POS) & HEIGHT_MASK) |
+			       (((ldst_xres - 1) << WIDTH_POS) & WIDTH_MASK),
+			  DCSS_SCALER_DST_LUM_RES);
+	dcss_scaler_write(ch, (((cdst_yres - 1) << HEIGHT_POS) & HEIGHT_MASK) |
+			       (((cdst_xres - 1) << WIDTH_POS) & WIDTH_MASK),
+			  DCSS_SCALER_DST_CHR_RES);
+}
+
+#define downscale_fp(factor, fp_pos)		((factor) << (fp_pos))
+#define upscale_fp(factor, fp_pos)		((1 << (fp_pos)) / (factor))
+
+struct dcss_scaler_factors {
+	int downscale;
+	int upscale;
+};
+
+static const struct dcss_scaler_factors dcss_scaler_factors[] = {
+	{3, 8}, {5, 8}, {5, 8},
+};
+
+static void dcss_scaler_fractions_set(struct dcss_scaler_ch *ch,
+				      int src_xres, int src_yres,
+				      int dst_xres, int dst_yres,
+				      u32 src_format, u32 dst_format,
+				      enum chroma_location src_chroma_loc)
+{
+	int src_c_xres, src_c_yres, dst_c_xres, dst_c_yres;
+	u32 l_vinc, l_hinc, c_vinc, c_hinc;
+	u32 c_vstart, c_hstart;
+
+	src_c_xres = src_xres;
+	src_c_yres = src_yres;
+	dst_c_xres = dst_xres;
+	dst_c_yres = dst_yres;
+
+	c_vstart = 0;
+	c_hstart = 0;
+
+	/* adjustments for source chroma location */
+	if (src_format == BUF_FMT_YUV420) {
+		/* vertical input chroma position adjustment */
+		switch (src_chroma_loc) {
+		case PSC_LOC_HORZ_0_VERT_1_OVER_4:
+		case PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_4:
+			/*
+			 * move chroma up to first luma line
+			 * (1/4 chroma input line spacing)
+			 */
+			c_vstart -= (1 << (PSC_PHASE_FRACTION_BITS - 2));
+			break;
+		case PSC_LOC_HORZ_0_VERT_1_OVER_2:
+		case PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_2:
+			/*
+			 * move chroma up to first luma line
+			 * (1/2 chroma input line spacing)
+			 */
+			c_vstart -= (1 << (PSC_PHASE_FRACTION_BITS - 1));
+			break;
+		default:
+			break;
+		}
+		/* horizontal input chroma position adjustment */
+		switch (src_chroma_loc) {
+		case PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_4:
+		case PSC_LOC_HORZ_1_OVER_4_VERT_0:
+		case PSC_LOC_HORZ_1_OVER_4_VERT_1_OVER_2:
+			/* move chroma left 1/4 chroma input sample spacing */
+			c_hstart -= (1 << (PSC_PHASE_FRACTION_BITS - 2));
+			break;
+		default:
+			break;
+		}
+	}
+
+	/* adjustments to chroma resolution */
+	if (src_format == BUF_FMT_YUV420) {
+		src_c_xres >>= 1;
+		src_c_yres >>= 1;
+	} else if (src_format == BUF_FMT_YUV422) {
+		src_c_xres >>= 1;
+	}
+
+	if (dst_format == BUF_FMT_YUV422)
+		dst_c_xres >>= 1;
+
+	l_vinc = ((src_yres << 13) + (dst_yres >> 1)) / dst_yres;
+	c_vinc = ((src_c_yres << 13) + (dst_c_yres >> 1)) / dst_c_yres;
+	l_hinc = ((src_xres << 13) + (dst_xres >> 1)) / dst_xres;
+	c_hinc = ((src_c_xres << 13) + (dst_c_xres >> 1)) / dst_c_xres;
+
+	/* save chroma start phase */
+	ch->c_vstart = c_vstart;
+	ch->c_hstart = c_hstart;
+
+	dcss_scaler_write(ch, 0, DCSS_SCALER_V_LUM_START);
+	dcss_scaler_write(ch, l_vinc, DCSS_SCALER_V_LUM_INC);
+
+	dcss_scaler_write(ch, 0, DCSS_SCALER_H_LUM_START);
+	dcss_scaler_write(ch, l_hinc, DCSS_SCALER_H_LUM_INC);
+
+	dcss_scaler_write(ch, c_vstart, DCSS_SCALER_V_CHR_START);
+	dcss_scaler_write(ch, c_vinc, DCSS_SCALER_V_CHR_INC);
+
+	dcss_scaler_write(ch, c_hstart, DCSS_SCALER_H_CHR_START);
+	dcss_scaler_write(ch, c_hinc, DCSS_SCALER_H_CHR_INC);
+}
+
+int dcss_scaler_get_min_max_ratios(struct dcss_scaler *scl, int ch_num,
+				   int *min, int *max)
+{
+	*min = upscale_fp(dcss_scaler_factors[ch_num].upscale, 16);
+	*max = downscale_fp(dcss_scaler_factors[ch_num].downscale, 16);
+
+	return 0;
+}
+
+static void dcss_scaler_program_5_coef_set(struct dcss_scaler_ch *ch,
+					   int base_addr,
+					   int coef[][PSC_NUM_TAPS])
+{
+	int i, phase;
+
+	for (i = 0; i < PSC_STORED_PHASES; i++) {
+		dcss_scaler_write(ch, ((coef[i][1] & 0xfff) << 16 |
+				       (coef[i][2] & 0xfff) << 4  |
+				       (coef[i][3] & 0xf00) >> 8),
+				  base_addr + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[i][3] & 0x0ff) << 20 |
+				       (coef[i][4] & 0xfff) << 8  |
+				       (coef[i][5] & 0xff0) >> 4),
+				  base_addr + 0x40 + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[i][5] & 0x00f) << 24),
+				  base_addr + 0x80 + i * sizeof(u32));
+	}
+
+	/* reverse both phase and tap orderings */
+	for (phase = (PSC_NUM_PHASES >> 1) - 1;
+			i < PSC_NUM_PHASES; i++, phase--) {
+		dcss_scaler_write(ch, ((coef[phase][5] & 0xfff) << 16 |
+				       (coef[phase][4] & 0xfff) << 4  |
+				       (coef[phase][3] & 0xf00) >> 8),
+				  base_addr + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[phase][3] & 0x0ff) << 20 |
+				       (coef[phase][2] & 0xfff) << 8  |
+				       (coef[phase][1] & 0xff0) >> 4),
+				  base_addr + 0x40 + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[phase][1] & 0x00f) << 24),
+				  base_addr + 0x80 + i * sizeof(u32));
+	}
+}
+
+static void dcss_scaler_program_7_coef_set(struct dcss_scaler_ch *ch,
+					   int base_addr,
+					   int coef[][PSC_NUM_TAPS])
+{
+	int i, phase;
+
+	for (i = 0; i < PSC_STORED_PHASES; i++) {
+		dcss_scaler_write(ch, ((coef[i][0] & 0xfff) << 16 |
+				       (coef[i][1] & 0xfff) << 4  |
+				       (coef[i][2] & 0xf00) >> 8),
+				  base_addr + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[i][2] & 0x0ff) << 20 |
+				       (coef[i][3] & 0xfff) << 8  |
+				       (coef[i][4] & 0xff0) >> 4),
+				  base_addr + 0x40 + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[i][4] & 0x00f) << 24 |
+				       (coef[i][5] & 0xfff) << 12 |
+				       (coef[i][6] & 0xfff)),
+				  base_addr + 0x80 + i * sizeof(u32));
+	}
+
+	/* reverse both phase and tap orderings */
+	for (phase = (PSC_NUM_PHASES >> 1) - 1;
+			i < PSC_NUM_PHASES; i++, phase--) {
+		dcss_scaler_write(ch, ((coef[phase][6] & 0xfff) << 16 |
+				       (coef[phase][5] & 0xfff) << 4  |
+				       (coef[phase][4] & 0xf00) >> 8),
+				  base_addr + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[phase][4] & 0x0ff) << 20 |
+				       (coef[phase][3] & 0xfff) << 8  |
+				       (coef[phase][2] & 0xff0) >> 4),
+				  base_addr + 0x40 + i * sizeof(u32));
+		dcss_scaler_write(ch, ((coef[phase][2] & 0x00f) << 24 |
+				       (coef[phase][1] & 0xfff) << 12 |
+				       (coef[phase][0] & 0xfff)),
+				  base_addr + 0x80 + i * sizeof(u32));
+	}
+}
+
+static void dcss_scaler_yuv_coef_set(struct dcss_scaler_ch *ch,
+				     enum buffer_format src_format,
+				     enum buffer_format dst_format,
+				     bool use_5_taps,
+				     int src_xres, int src_yres, int dst_xres,
+				     int dst_yres)
+{
+	int coef[PSC_STORED_PHASES][PSC_NUM_TAPS];
+	bool program_5_taps = use_5_taps ||
+			      (dst_format == BUF_FMT_YUV422 &&
+			       src_format == BUF_FMT_ARGB8888_YUV444);
+
+	/* horizontal luma */
+	dcss_scaler_filter_design(src_xres, dst_xres, false,
+				  src_xres == dst_xres, coef);
+	dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_HLUM, coef);
+
+	/* vertical luma */
+	dcss_scaler_filter_design(src_yres, dst_yres, program_5_taps,
+				  src_yres == dst_yres, coef);
+
+	if (program_5_taps)
+		dcss_scaler_program_5_coef_set(ch, DCSS_SCALER_COEF_VLUM, coef);
+	else
+		dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_VLUM, coef);
+
+	/* adjust chroma resolution */
+	if (src_format != BUF_FMT_ARGB8888_YUV444)
+		src_xres >>= 1;
+	if (src_format == BUF_FMT_YUV420)
+		src_yres >>= 1;
+	if (dst_format != BUF_FMT_ARGB8888_YUV444)
+		dst_xres >>= 1;
+	if (dst_format == BUF_FMT_YUV420) /* should not happen */
+		dst_yres >>= 1;
+
+	/* horizontal chroma */
+	dcss_scaler_filter_design(src_xres, dst_xres, false,
+				  (src_xres == dst_xres) && (ch->c_hstart == 0),
+				  coef);
+
+	dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_HCHR, coef);
+
+	/* vertical chroma */
+	dcss_scaler_filter_design(src_yres, dst_yres, program_5_taps,
+				  (src_yres == dst_yres) && (ch->c_vstart == 0),
+				  coef);
+	if (program_5_taps)
+		dcss_scaler_program_5_coef_set(ch, DCSS_SCALER_COEF_VCHR, coef);
+	else
+		dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_VCHR, coef);
+}
+
+static void dcss_scaler_rgb_coef_set(struct dcss_scaler_ch *ch,
+				     int src_xres, int src_yres, int dst_xres,
+				     int dst_yres)
+{
+	int coef[PSC_STORED_PHASES][PSC_NUM_TAPS];
+
+	/* horizontal RGB */
+	dcss_scaler_filter_design(src_xres, dst_xres, false,
+				  src_xres == dst_xres, coef);
+	dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_HLUM, coef);
+
+	/* vertical RGB */
+	dcss_scaler_filter_design(src_yres, dst_yres, false,
+				  src_yres == dst_yres, coef);
+	dcss_scaler_program_7_coef_set(ch, DCSS_SCALER_COEF_VLUM, coef);
+}
+
+static void dcss_scaler_set_rgb10_order(struct dcss_scaler_ch *ch,
+					const struct drm_format_info *format)
+{
+	u32 a2r10g10b10_format;
+
+	if (format->is_yuv)
+		return;
+
+	ch->sdata_ctrl &= ~A2R10G10B10_FORMAT_MASK;
+
+	if (format->depth != 30)
+		return;
+
+	switch (format->format) {
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_XRGB2101010:
+		a2r10g10b10_format = 0;
+		break;
+
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_XBGR2101010:
+		a2r10g10b10_format = 5;
+		break;
+
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_RGBX1010102:
+		a2r10g10b10_format = 6;
+		break;
+
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_BGRX1010102:
+		a2r10g10b10_format = 11;
+		break;
+
+	default:
+		a2r10g10b10_format = 0;
+		break;
+	}
+
+	ch->sdata_ctrl |= a2r10g10b10_format << A2R10G10B10_FORMAT_POS;
+}
+
+void dcss_scaler_setup(struct dcss_scaler *scl, int ch_num,
+		       const struct drm_format_info *format,
+		       int src_xres, int src_yres, int dst_xres, int dst_yres,
+		       u32 vrefresh_hz)
+{
+	struct dcss_scaler_ch *ch = &scl->ch[ch_num];
+	unsigned int pixel_depth = 0;
+	bool rtr_8line_en = false;
+	bool use_5_taps = false;
+	enum buffer_format src_format = BUF_FMT_ARGB8888_YUV444;
+	enum buffer_format dst_format = BUF_FMT_ARGB8888_YUV444;
+	u32 pix_format = format->format;
+
+	if (format->is_yuv) {
+		dcss_scaler_yuv_enable(ch, true);
+
+		if (pix_format == DRM_FORMAT_NV12 ||
+		    pix_format == DRM_FORMAT_NV21) {
+			rtr_8line_en = true;
+			src_format = BUF_FMT_YUV420;
+		} else if (pix_format == DRM_FORMAT_UYVY ||
+			   pix_format == DRM_FORMAT_VYUY ||
+			   pix_format == DRM_FORMAT_YUYV ||
+			   pix_format == DRM_FORMAT_YVYU) {
+			src_format = BUF_FMT_YUV422;
+		}
+
+		use_5_taps = !rtr_8line_en;
+	} else {
+		dcss_scaler_yuv_enable(ch, false);
+
+		pixel_depth = format->depth;
+	}
+
+	dcss_scaler_fractions_set(ch, src_xres, src_yres, dst_xres,
+				  dst_yres, src_format, dst_format,
+				  PSC_LOC_HORZ_0_VERT_1_OVER_4);
+
+	if (format->is_yuv)
+		dcss_scaler_yuv_coef_set(ch, src_format, dst_format,
+					 use_5_taps, src_xres, src_yres,
+					 dst_xres, dst_yres);
+	else
+		dcss_scaler_rgb_coef_set(ch, src_xres, src_yres,
+					 dst_xres, dst_yres);
+
+	dcss_scaler_rtr_8lines_enable(ch, rtr_8line_en);
+	dcss_scaler_bit_depth_set(ch, pixel_depth);
+	dcss_scaler_set_rgb10_order(ch, format);
+	dcss_scaler_format_set(ch, src_format, dst_format);
+	dcss_scaler_res_set(ch, src_xres, src_yres, dst_xres, dst_yres,
+			    pix_format, dst_format);
+}
+
+/* This function will be called from interrupt context. */
+void dcss_scaler_write_sclctrl(struct dcss_scaler *scl)
+{
+	int chnum;
+
+	dcss_ctxld_assert_locked(scl->ctxld);
+
+	for (chnum = 0; chnum < 3; chnum++) {
+		struct dcss_scaler_ch *ch = &scl->ch[chnum];
+
+		if (ch->scaler_ctrl_chgd) {
+			dcss_ctxld_write_irqsafe(scl->ctxld, scl->ctx_id,
+						 ch->scaler_ctrl,
+						 ch->base_ofs +
+						 DCSS_SCALER_CTRL);
+			ch->scaler_ctrl_chgd = false;
+		}
+	}
+}
diff --git a/drivers/gpu/drm/imx/dcss/dcss-ss.c b/drivers/gpu/drm/imx/dcss/dcss-ss.c
new file mode 100644
index 000000000000..8ddf08da911b
--- /dev/null
+++ b/drivers/gpu/drm/imx/dcss/dcss-ss.c
@@ -0,0 +1,180 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 NXP.
+ */
+
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include "dcss-dev.h"
+
+#define DCSS_SS_SYS_CTRL			0x00
+#define   RUN_EN				BIT(0)
+#define DCSS_SS_DISPLAY				0x10
+#define   LRC_X_POS				0
+#define   LRC_X_MASK				GENMASK(12, 0)
+#define   LRC_Y_POS				16
+#define   LRC_Y_MASK				GENMASK(28, 16)
+#define DCSS_SS_HSYNC				0x20
+#define DCSS_SS_VSYNC				0x30
+#define   SYNC_START_POS			0
+#define   SYNC_START_MASK			GENMASK(12, 0)
+#define   SYNC_END_POS				16
+#define   SYNC_END_MASK				GENMASK(28, 16)
+#define   SYNC_POL				BIT(31)
+#define DCSS_SS_DE_ULC				0x40
+#define   ULC_X_POS				0
+#define   ULC_X_MASK				GENMASK(12, 0)
+#define   ULC_Y_POS				16
+#define   ULC_Y_MASK				GENMASK(28, 16)
+#define   ULC_POL				BIT(31)
+#define DCSS_SS_DE_LRC				0x50
+#define DCSS_SS_MODE				0x60
+#define   PIPE_MODE_POS				0
+#define   PIPE_MODE_MASK			GENMASK(1, 0)
+#define DCSS_SS_COEFF				0x70
+#define   HORIZ_A_POS				0
+#define   HORIZ_A_MASK				GENMASK(3, 0)
+#define   HORIZ_B_POS				4
+#define   HORIZ_B_MASK				GENMASK(7, 4)
+#define   HORIZ_C_POS				8
+#define   HORIZ_C_MASK				GENMASK(11, 8)
+#define   HORIZ_H_NORM_POS			12
+#define   HORIZ_H_NORM_MASK			GENMASK(14, 12)
+#define   VERT_A_POS				16
+#define   VERT_A_MASK				GENMASK(19, 16)
+#define   VERT_B_POS				20
+#define   VERT_B_MASK				GENMASK(23, 20)
+#define   VERT_C_POS				24
+#define   VERT_C_MASK				GENMASK(27, 24)
+#define   VERT_H_NORM_POS			28
+#define   VERT_H_NORM_MASK			GENMASK(30, 28)
+#define DCSS_SS_CLIP_CB				0x80
+#define DCSS_SS_CLIP_CR				0x90
+#define   CLIP_MIN_POS				0
+#define   CLIP_MIN_MASK				GENMASK(9, 0)
+#define   CLIP_MAX_POS				0
+#define   CLIP_MAX_MASK				GENMASK(23, 16)
+#define DCSS_SS_INTER_MODE			0xA0
+#define   INT_EN				BIT(0)
+#define   VSYNC_SHIFT				BIT(1)
+
+struct dcss_ss {
+	struct device *dev;
+	void __iomem *base_reg;
+	u32 base_ofs;
+
+	struct dcss_ctxld *ctxld;
+	u32 ctx_id;
+
+	bool in_use;
+};
+
+static void dcss_ss_write(struct dcss_ss *ss, u32 val, u32 ofs)
+{
+	if (!ss->in_use)
+		dcss_writel(val, ss->base_reg + ofs);
+
+	dcss_ctxld_write(ss->ctxld, ss->ctx_id, val,
+			 ss->base_ofs + ofs);
+}
+
+int dcss_ss_init(struct dcss_dev *dcss, unsigned long ss_base)
+{
+	struct dcss_ss *ss;
+
+	ss = kzalloc(sizeof(*ss), GFP_KERNEL);
+	if (!ss)
+		return -ENOMEM;
+
+	dcss->ss = ss;
+	ss->dev = dcss->dev;
+	ss->ctxld = dcss->ctxld;
+
+	ss->base_reg = ioremap(ss_base, SZ_4K);
+	if (!ss->base_reg) {
+		dev_err(dcss->dev, "ss: unable to remap ss base\n");
+		kfree(ss);
+		return -ENOMEM;
+	}
+
+	ss->base_ofs = ss_base;
+	ss->ctx_id = CTX_SB_HP;
+
+	return 0;
+}
+
+void dcss_ss_exit(struct dcss_ss *ss)
+{
+	/* stop SS */
+	dcss_writel(0, ss->base_reg + DCSS_SS_SYS_CTRL);
+
+	if (ss->base_reg)
+		iounmap(ss->base_reg);
+
+	kfree(ss);
+}
+
+void dcss_ss_subsam_set(struct dcss_ss *ss)
+{
+	dcss_ss_write(ss, 0x41614161, DCSS_SS_COEFF);
+	dcss_ss_write(ss, 0, DCSS_SS_MODE);
+	dcss_ss_write(ss, 0x03ff0000, DCSS_SS_CLIP_CB);
+	dcss_ss_write(ss, 0x03ff0000, DCSS_SS_CLIP_CR);
+}
+
+void dcss_ss_sync_set(struct dcss_ss *ss, struct videomode *vm,
+		      bool phsync, bool pvsync)
+{
+	u16 lrc_x, lrc_y;
+	u16 hsync_start, hsync_end;
+	u16 vsync_start, vsync_end;
+	u16 de_ulc_x, de_ulc_y;
+	u16 de_lrc_x, de_lrc_y;
+
+	lrc_x = vm->hfront_porch + vm->hback_porch + vm->hsync_len +
+		vm->hactive - 1;
+	lrc_y = vm->vfront_porch + vm->vback_porch + vm->vsync_len +
+		vm->vactive - 1;
+
+	dcss_ss_write(ss, (lrc_y << LRC_Y_POS) | lrc_x, DCSS_SS_DISPLAY);
+
+	hsync_start = vm->hfront_porch + vm->hback_porch + vm->hsync_len +
+		      vm->hactive - 1;
+	hsync_end = vm->hsync_len - 1;
+
+	dcss_ss_write(ss, (phsync ? SYNC_POL : 0) |
+		      ((u32)hsync_end << SYNC_END_POS) | hsync_start,
+		      DCSS_SS_HSYNC);
+
+	vsync_start = vm->vfront_porch - 1;
+	vsync_end = vm->vfront_porch + vm->vsync_len - 1;
+
+	dcss_ss_write(ss, (pvsync ? SYNC_POL : 0) |
+		      ((u32)vsync_end << SYNC_END_POS) | vsync_start,
+		      DCSS_SS_VSYNC);
+
+	de_ulc_x = vm->hsync_len + vm->hback_porch - 1;
+	de_ulc_y = vm->vsync_len + vm->vfront_porch + vm->vback_porch;
+
+	dcss_ss_write(ss, SYNC_POL | ((u32)de_ulc_y << ULC_Y_POS) | de_ulc_x,
+		      DCSS_SS_DE_ULC);
+
+	de_lrc_x = vm->hsync_len + vm->hback_porch + vm->hactive - 1;
+	de_lrc_y = vm->vsync_len + vm->vfront_porch + vm->vback_porch +
+		   vm->vactive - 1;
+
+	dcss_ss_write(ss, (de_lrc_y << LRC_Y_POS) | de_lrc_x, DCSS_SS_DE_LRC);
+}
+
+void dcss_ss_enable(struct dcss_ss *ss)
+{
+	dcss_ss_write(ss, RUN_EN, DCSS_SS_SYS_CTRL);
+	ss->in_use = true;
+}
+
+void dcss_ss_shutoff(struct dcss_ss *ss)
+{
+	dcss_writel(0, ss->base_reg + DCSS_SS_SYS_CTRL);
+	ss->in_use = false;
+}
diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c
index 8791d60be92e..af757d1e21fe 100644
--- a/drivers/gpu/drm/imx/imx-ldb.c
+++ b/drivers/gpu/drm/imx/imx-ldb.c
@@ -455,13 +455,6 @@ static int imx_ldb_register(struct drm_device *drm,
 		drm_connector_attach_encoder(&imx_ldb_ch->connector, encoder);
 	}
 
-	if (imx_ldb_ch->panel) {
-		ret = drm_panel_attach(imx_ldb_ch->panel,
-				       &imx_ldb_ch->connector);
-		if (ret)
-			return ret;
-	}
-
 	return 0;
 }
 
@@ -702,9 +695,6 @@ static void imx_ldb_unbind(struct device *dev, struct device *master,
 	for (i = 0; i < 2; i++) {
 		struct imx_ldb_channel *channel = &imx_ldb->channel[i];
 
-		if (channel->panel)
-			drm_panel_detach(channel->panel);
-
 		kfree(channel->edid);
 		i2c_put_adapter(channel->ddc);
 	}
diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c
index a831b5bd1613..8232f512b9ed 100644
--- a/drivers/gpu/drm/imx/parallel-display.c
+++ b/drivers/gpu/drm/imx/parallel-display.c
@@ -289,9 +289,6 @@ static int imx_pd_register(struct drm_device *drm,
 				   DRM_MODE_CONNECTOR_DPI);
 	}
 
-	if (imxpd->panel)
-		drm_panel_attach(imxpd->panel, &imxpd->connector);
-
 	if (imxpd->next_bridge) {
 		ret = drm_bridge_attach(encoder, imxpd->next_bridge,
 					&imxpd->bridge, 0);
@@ -357,9 +354,6 @@ static void imx_pd_unbind(struct device *dev, struct device *master,
 {
 	struct imx_parallel_display *imxpd = dev_get_drvdata(dev);
 
-	if (imxpd->panel)
-		drm_panel_detach(imxpd->panel);
-
 	kfree(imxpd->edid);
 }
 
diff --git a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
index b7074161ccf0..a3d1617d7c67 100644
--- a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
+++ b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
@@ -199,26 +199,20 @@ static int ingenic_drm_crtc_atomic_check(struct drm_crtc *crtc,
 {
 	struct ingenic_drm *priv = drm_crtc_get_priv(crtc);
 	struct drm_plane_state *f1_state, *f0_state, *ipu_state = NULL;
-	long rate;
-
-	if (!drm_atomic_crtc_needs_modeset(state))
-		return 0;
-
-	if (state->mode.hdisplay > priv->soc_info->max_width ||
-	    state->mode.vdisplay > priv->soc_info->max_height)
-		return -EINVAL;
 
-	rate = clk_round_rate(priv->pix_clk,
-			      state->adjusted_mode.clock * 1000);
-	if (rate < 0)
-		return rate;
-
-	if (priv->soc_info->has_osd) {
+	if (drm_atomic_crtc_needs_modeset(state) && priv->soc_info->has_osd) {
 		f1_state = drm_atomic_get_plane_state(state->state, &priv->f1);
+		if (IS_ERR(f1_state))
+			return PTR_ERR(f1_state);
+
 		f0_state = drm_atomic_get_plane_state(state->state, &priv->f0);
+		if (IS_ERR(f0_state))
+			return PTR_ERR(f0_state);
 
 		if (IS_ENABLED(CONFIG_DRM_INGENIC_IPU) && priv->ipu_plane) {
 			ipu_state = drm_atomic_get_plane_state(state->state, priv->ipu_plane);
+			if (IS_ERR(ipu_state))
+				return PTR_ERR(ipu_state);
 
 			/* IPU and F1 planes cannot be enabled at the same time. */
 			if (f1_state->fb && ipu_state->fb) {
@@ -235,6 +229,24 @@ static int ingenic_drm_crtc_atomic_check(struct drm_crtc *crtc,
 	return 0;
 }
 
+static enum drm_mode_status
+ingenic_drm_crtc_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *mode)
+{
+	struct ingenic_drm *priv = drm_crtc_get_priv(crtc);
+	long rate;
+
+	if (mode->hdisplay > priv->soc_info->max_width)
+		return MODE_BAD_HVALUE;
+	if (mode->vdisplay > priv->soc_info->max_height)
+		return MODE_BAD_VVALUE;
+
+	rate = clk_round_rate(priv->pix_clk, mode->clock * 1000);
+	if (rate < 0)
+		return MODE_CLOCK_RANGE;
+
+	return MODE_OK;
+}
+
 static void ingenic_drm_crtc_atomic_begin(struct drm_crtc *crtc,
 					  struct drm_crtc_state *oldstate)
 {
@@ -648,6 +660,7 @@ static const struct drm_crtc_helper_funcs ingenic_drm_crtc_helper_funcs = {
 	.atomic_begin		= ingenic_drm_crtc_atomic_begin,
 	.atomic_flush		= ingenic_drm_crtc_atomic_flush,
 	.atomic_check		= ingenic_drm_crtc_atomic_check,
+	.mode_valid		= ingenic_drm_crtc_mode_valid,
 };
 
 static const struct drm_encoder_helper_funcs ingenic_drm_encoder_helper_funcs = {
diff --git a/drivers/gpu/drm/ingenic/ingenic-ipu.c b/drivers/gpu/drm/ingenic/ingenic-ipu.c
index 7a0a8bd865d3..fc8c6e970ee3 100644
--- a/drivers/gpu/drm/ingenic/ingenic-ipu.c
+++ b/drivers/gpu/drm/ingenic/ingenic-ipu.c
@@ -35,6 +35,7 @@ struct soc_info {
 	const u32 *formats;
 	size_t num_formats;
 	bool has_bicubic;
+	bool manual_restart;
 
 	void (*set_coefs)(struct ingenic_ipu *ipu, unsigned int reg,
 			  unsigned int sharpness, bool downscale,
@@ -48,6 +49,7 @@ struct ingenic_ipu {
 	struct regmap *map;
 	struct clk *clk;
 	const struct soc_info *soc_info;
+	bool clk_enabled;
 
 	unsigned int num_w, num_h, denom_w, denom_h;
 
@@ -287,12 +289,23 @@ static void ingenic_ipu_plane_atomic_update(struct drm_plane *plane,
 	const struct drm_format_info *finfo;
 	u32 ctrl, stride = 0, coef_index = 0, format = 0;
 	bool needs_modeset, upscaling_w, upscaling_h;
+	int err;
 
 	if (!state || !state->fb)
 		return;
 
 	finfo = drm_format_info(state->fb->format->format);
 
+	if (!ipu->clk_enabled) {
+		err = clk_enable(ipu->clk);
+		if (err) {
+			dev_err(ipu->dev, "Unable to enable clock: %d\n", err);
+			return;
+		}
+
+		ipu->clk_enabled = true;
+	}
+
 	/* Reset all the registers if needed */
 	needs_modeset = drm_atomic_crtc_needs_modeset(state->crtc->state);
 	if (needs_modeset) {
@@ -577,6 +590,11 @@ static void ingenic_ipu_plane_atomic_disable(struct drm_plane *plane,
 	regmap_clear_bits(ipu->map, JZ_REG_IPU_CTRL, JZ_IPU_CTRL_CHIP_EN);
 
 	ingenic_drm_plane_disable(ipu->master, plane);
+
+	if (ipu->clk_enabled) {
+		clk_disable(ipu->clk);
+		ipu->clk_enabled = false;
+	}
 }
 
 static const struct drm_plane_helper_funcs ingenic_ipu_plane_helper_funcs = {
@@ -645,7 +663,8 @@ static irqreturn_t ingenic_ipu_irq_handler(int irq, void *arg)
 	unsigned int dummy;
 
 	/* dummy read allows CPU to reconfigure IPU */
-	regmap_read(ipu->map, JZ_REG_IPU_STATUS, &dummy);
+	if (ipu->soc_info->manual_restart)
+		regmap_read(ipu->map, JZ_REG_IPU_STATUS, &dummy);
 
 	/* ACK interrupt */
 	regmap_write(ipu->map, JZ_REG_IPU_STATUS, 0);
@@ -656,7 +675,8 @@ static irqreturn_t ingenic_ipu_irq_handler(int irq, void *arg)
 	regmap_write(ipu->map, JZ_REG_IPU_V_ADDR, ipu->addr_v);
 
 	/* Run IPU for the new frame */
-	regmap_set_bits(ipu->map, JZ_REG_IPU_CTRL, JZ_IPU_CTRL_RUN);
+	if (ipu->soc_info->manual_restart)
+		regmap_set_bits(ipu->map, JZ_REG_IPU_CTRL, JZ_IPU_CTRL_RUN);
 
 	drm_crtc_handle_vblank(crtc);
 
@@ -758,9 +778,9 @@ static int ingenic_ipu_bind(struct device *dev, struct device *master, void *d)
 	drm_object_attach_property(&plane->base, ipu->sharpness_prop,
 				   ipu->sharpness);
 
-	err = clk_prepare_enable(ipu->clk);
+	err = clk_prepare(ipu->clk);
 	if (err) {
-		dev_err(dev, "Unable to enable clock\n");
+		dev_err(dev, "Unable to prepare clock\n");
 		return err;
 	}
 
@@ -772,7 +792,7 @@ static void ingenic_ipu_unbind(struct device *dev,
 {
 	struct ingenic_ipu *ipu = dev_get_drvdata(dev);
 
-	clk_disable_unprepare(ipu->clk);
+	clk_unprepare(ipu->clk);
 }
 
 static const struct component_ops ingenic_ipu_ops = {
@@ -792,10 +812,16 @@ static int ingenic_ipu_remove(struct platform_device *pdev)
 }
 
 static const u32 jz4725b_ipu_formats[] = {
+	/*
+	 * While officially supported, packed YUV 4:2:2 formats can cause
+	 * random hardware crashes on JZ4725B under certain circumstances.
+	 * It seems to happen with some specific resize ratios.
+	 * Until a proper workaround or fix is found, disable these formats.
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YVYU,
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_VYUY,
+	*/
 	DRM_FORMAT_YUV411,
 	DRM_FORMAT_YUV420,
 	DRM_FORMAT_YUV422,
@@ -806,6 +832,7 @@ static const struct soc_info jz4725b_soc_info = {
 	.formats	= jz4725b_ipu_formats,
 	.num_formats	= ARRAY_SIZE(jz4725b_ipu_formats),
 	.has_bicubic	= false,
+	.manual_restart	= true,
 	.set_coefs	= jz4725b_set_coefs,
 };
 
@@ -831,6 +858,7 @@ static const struct soc_info jz4760_soc_info = {
 	.formats	= jz4760_ipu_formats,
 	.num_formats	= ARRAY_SIZE(jz4760_ipu_formats),
 	.has_bicubic	= true,
+	.manual_restart	= false,
 	.set_coefs	= jz4760_set_coefs,
 };
 
diff --git a/drivers/gpu/drm/lima/lima_gem.c b/drivers/gpu/drm/lima/lima_gem.c
index 155f2b4b4030..11223fe348df 100644
--- a/drivers/gpu/drm/lima/lima_gem.c
+++ b/drivers/gpu/drm/lima/lima_gem.c
@@ -69,8 +69,7 @@ int lima_heap_alloc(struct lima_bo *bo, struct lima_vm *vm)
 		return ret;
 
 	if (bo->base.sgt) {
-		dma_unmap_sg(dev, bo->base.sgt->sgl,
-			     bo->base.sgt->nents, DMA_BIDIRECTIONAL);
+		dma_unmap_sgtable(dev, bo->base.sgt, DMA_BIDIRECTIONAL, 0);
 		sg_free_table(bo->base.sgt);
 	} else {
 		bo->base.sgt = kmalloc(sizeof(*bo->base.sgt), GFP_KERNEL);
@@ -80,7 +79,13 @@ int lima_heap_alloc(struct lima_bo *bo, struct lima_vm *vm)
 		}
 	}
 
-	dma_map_sg(dev, sgt.sgl, sgt.nents, DMA_BIDIRECTIONAL);
+	ret = dma_map_sgtable(dev, &sgt, DMA_BIDIRECTIONAL, 0);
+	if (ret) {
+		sg_free_table(&sgt);
+		kfree(bo->base.sgt);
+		bo->base.sgt = NULL;
+		return ret;
+	}
 
 	*bo->base.sgt = sgt;
 
diff --git a/drivers/gpu/drm/lima/lima_vm.c b/drivers/gpu/drm/lima/lima_vm.c
index 5b92fb82674a..2b2739adc7f5 100644
--- a/drivers/gpu/drm/lima/lima_vm.c
+++ b/drivers/gpu/drm/lima/lima_vm.c
@@ -124,7 +124,7 @@ int lima_vm_bo_add(struct lima_vm *vm, struct lima_bo *bo, bool create)
 	if (err)
 		goto err_out1;
 
-	for_each_sg_dma_page(bo->base.sgt->sgl, &sg_iter, bo->base.sgt->nents, 0) {
+	for_each_sgtable_dma_page(bo->base.sgt, &sg_iter, 0) {
 		err = lima_vm_map_page(vm, sg_page_iter_dma_address(&sg_iter),
 				       bo_va->node.start + offset);
 		if (err)
@@ -298,8 +298,7 @@ int lima_vm_map_bo(struct lima_vm *vm, struct lima_bo *bo, int pageoff)
 	mutex_lock(&vm->lock);
 
 	base = bo_va->node.start + (pageoff << PAGE_SHIFT);
-	for_each_sg_dma_page(bo->base.sgt->sgl, &sg_iter,
-			     bo->base.sgt->nents, pageoff) {
+	for_each_sgtable_dma_page(bo->base.sgt, &sg_iter, pageoff) {
 		err = lima_vm_map_page(vm, sg_page_iter_dma_address(&sg_iter),
 				       base + offset);
 		if (err)
diff --git a/drivers/gpu/drm/mcde/mcde_display.c b/drivers/gpu/drm/mcde/mcde_display.c
index 4d2290f88edb..c271e5bf042e 100644
--- a/drivers/gpu/drm/mcde/mcde_display.c
+++ b/drivers/gpu/drm/mcde/mcde_display.c
@@ -7,6 +7,7 @@
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/dma-buf.h>
+#include <linux/regulator/consumer.h>
 
 #include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
@@ -89,7 +90,7 @@ void mcde_display_irq(struct mcde *mcde)
 		 * the update function is called, then we disable the
 		 * flow on the channel once we get the TE IRQ.
 		 */
-		if (mcde->oneshot_mode) {
+		if (mcde->flow_mode == MCDE_COMMAND_ONESHOT_FLOW) {
 			spin_lock(&mcde->flow_lock);
 			if (--mcde->flow_active == 0) {
 				dev_dbg(mcde->dev, "TE0 IRQ\n");
@@ -333,7 +334,7 @@ static void mcde_configure_overlay(struct mcde *mcde, enum mcde_overlay ovl,
 				   enum mcde_extsrc src,
 				   enum mcde_channel ch,
 				   const struct drm_display_mode *mode,
-				   u32 format)
+				   u32 format, int cpp)
 {
 	u32 val;
 	u32 conf1;
@@ -342,6 +343,7 @@ static void mcde_configure_overlay(struct mcde *mcde, enum mcde_overlay ovl,
 	u32 ljinc;
 	u32 cr;
 	u32 comp;
+	u32 pixel_fetcher_watermark;
 
 	switch (ovl) {
 	case MCDE_OVERLAY_0:
@@ -426,8 +428,33 @@ static void mcde_configure_overlay(struct mcde *mcde, enum mcde_overlay ovl,
 			format);
 		break;
 	}
-	/* The default watermark level for overlay 0 is 48 */
-	val |= 48 << MCDE_OVLXCONF2_PIXELFETCHERWATERMARKLEVEL_SHIFT;
+
+	/*
+	 * Pixel fetch watermark level is max 0x1FFF pixels.
+	 * Two basic rules should be followed:
+	 * 1. The value should be at least 256 bits.
+	 * 2. The sum of all active overlays pixelfetch watermark level
+	 *    multiplied with bits per pixel, should be lower than the
+	 *    size of input_fifo_size in bits.
+	 * 3. The value should be a multiple of a line (256 bits).
+	 */
+	switch (cpp) {
+	case 2:
+		pixel_fetcher_watermark = 128;
+		break;
+	case 3:
+		pixel_fetcher_watermark = 96;
+		break;
+	case 4:
+		pixel_fetcher_watermark = 48;
+		break;
+	default:
+		pixel_fetcher_watermark = 48;
+		break;
+	}
+	dev_dbg(mcde->dev, "pixel fetcher watermark level %d pixels\n",
+		pixel_fetcher_watermark);
+	val |= pixel_fetcher_watermark << MCDE_OVLXCONF2_PIXELFETCHERWATERMARKLEVEL_SHIFT;
 	writel(val, mcde->regs + conf2);
 
 	/* Number of bytes to fetch per line */
@@ -498,19 +525,47 @@ static void mcde_configure_channel(struct mcde *mcde, enum mcde_channel ch,
 	}
 
 	/* Set up channel 0 sync (based on chnl_update_registers()) */
-	if (mcde->video_mode || mcde->te_sync)
+	switch (mcde->flow_mode) {
+	case MCDE_COMMAND_ONESHOT_FLOW:
+		/* Oneshot is achieved with software sync */
+		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SOFTWARE
+			<< MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SHIFT;
+		break;
+	case MCDE_COMMAND_TE_FLOW:
 		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_HARDWARE
 			<< MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SHIFT;
-	else
-		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SOFTWARE
+		val |= MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_TE0
+			<< MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_SHIFT;
+		break;
+	case MCDE_COMMAND_BTA_TE_FLOW:
+		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_HARDWARE
+			<< MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SHIFT;
+		/*
+		 * TODO:
+		 * The vendor driver uses the formatter as sync source
+		 * for BTA TE mode. Test to use TE if you have a panel
+		 * that uses this mode.
+		 */
+		val |= MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_FORMATTER
+			<< MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_SHIFT;
+		break;
+	case MCDE_VIDEO_TE_FLOW:
+		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_HARDWARE
 			<< MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SHIFT;
-
-	if (mcde->te_sync)
 		val |= MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_TE0
 			<< MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_SHIFT;
-	else
+		break;
+	case MCDE_VIDEO_FORMATTER_FLOW:
+		val = MCDE_CHNLXSYNCHMOD_SRC_SYNCH_HARDWARE
+			<< MCDE_CHNLXSYNCHMOD_SRC_SYNCH_SHIFT;
 		val |= MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_FORMATTER
 			<< MCDE_CHNLXSYNCHMOD_OUT_SYNCH_SRC_SHIFT;
+		break;
+	default:
+		dev_err(mcde->dev, "unknown flow mode %d\n",
+			mcde->flow_mode);
+		break;
+	}
 
 	writel(val, mcde->regs + sync);
 
@@ -825,6 +880,14 @@ static void mcde_display_enable(struct drm_simple_display_pipe *pipe,
 	u32 formatter_frame;
 	u32 pkt_div;
 	u32 val;
+	int ret;
+
+	/* This powers up the entire MCDE block and the DSI hardware */
+	ret = regulator_enable(mcde->epod);
+	if (ret) {
+		dev_err(drm->dev, "can't re-enable EPOD regulator\n");
+		return;
+	}
 
 	dev_info(drm->dev, "enable MCDE, %d x %d format %s\n",
 		 mode->hdisplay, mode->vdisplay,
@@ -835,6 +898,26 @@ static void mcde_display_enable(struct drm_simple_display_pipe *pipe,
 		return;
 	}
 
+	/* Set up the main control, watermark level at 7 */
+	val = 7 << MCDE_CONF0_IFIFOCTRLWTRMRKLVL_SHIFT;
+	/* 24 bits DPI: connect LSB Ch B to D[0:7] */
+	val |= 3 << MCDE_CONF0_OUTMUX0_SHIFT;
+	/* TV out: connect LSB Ch B to D[8:15] */
+	val |= 3 << MCDE_CONF0_OUTMUX1_SHIFT;
+	/* Don't care about this muxing */
+	val |= 0 << MCDE_CONF0_OUTMUX2_SHIFT;
+	/* 24 bits DPI: connect MID Ch B to D[24:31] */
+	val |= 4 << MCDE_CONF0_OUTMUX3_SHIFT;
+	/* 5: 24 bits DPI: connect MSB Ch B to D[32:39] */
+	val |= 5 << MCDE_CONF0_OUTMUX4_SHIFT;
+	/* Syncmux bits zero: DPI channel A and B on output pins A and B resp */
+	writel(val, mcde->regs + MCDE_CONF0);
+
+	/* Clear any pending interrupts */
+	mcde_display_disable_irqs(mcde);
+	writel(0, mcde->regs + MCDE_IMSCERR);
+	writel(0xFFFFFFFF, mcde->regs + MCDE_RISERR);
+
 	dev_info(drm->dev, "output in %s mode, format %dbpp\n",
 		 (mcde->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO) ?
 		 "VIDEO" : "CMD",
@@ -904,7 +987,7 @@ static void mcde_display_enable(struct drm_simple_display_pipe *pipe,
 	 * channel 0
 	 */
 	mcde_configure_overlay(mcde, MCDE_OVERLAY_0, MCDE_EXTSRC_0,
-			       MCDE_CHANNEL_0, mode, format);
+			       MCDE_CHANNEL_0, mode, format, cpp);
 
 	/*
 	 * Configure pixel-per-line and line-per-frame for channel 0 and then
@@ -916,11 +999,25 @@ static void mcde_display_enable(struct drm_simple_display_pipe *pipe,
 	mcde_configure_fifo(mcde, MCDE_FIFO_A, MCDE_DSI_FORMATTER_0,
 			    fifo_wtrmrk);
 
+	/*
+	 * This brings up the DSI bridge which is tightly connected
+	 * to the MCDE DSI formatter.
+	 *
+	 * FIXME: if we want to use another formatter, such as DPI,
+	 * we need to be more elaborate here and select the appropriate
+	 * bridge.
+	 */
+	mcde_dsi_enable(mcde->bridge);
+
 	/* Configure the DSI formatter 0 for the DSI panel output */
 	mcde_configure_dsi_formatter(mcde, MCDE_DSI_FORMATTER_0,
 				     formatter_frame, pkt_size);
 
-	if (mcde->te_sync) {
+	switch (mcde->flow_mode) {
+	case MCDE_COMMAND_TE_FLOW:
+	case MCDE_COMMAND_BTA_TE_FLOW:
+	case MCDE_VIDEO_TE_FLOW:
+		/* We are using TE in some comination */
 		if (mode->flags & DRM_MODE_FLAG_NVSYNC)
 			val = MCDE_VSCRC_VSPOL;
 		else
@@ -930,16 +1027,31 @@ static void mcde_display_enable(struct drm_simple_display_pipe *pipe,
 		val = readl(mcde->regs + MCDE_CRC);
 		val |= MCDE_CRC_SYCEN0;
 		writel(val, mcde->regs + MCDE_CRC);
+		break;
+	default:
+		/* No TE capture */
+		break;
 	}
 
 	drm_crtc_vblank_on(crtc);
 
-	if (mcde->video_mode)
-		/*
-		 * Keep FIFO permanently enabled in video mode,
-		 * otherwise MCDE will stop feeding data to the panel.
-		 */
+	/*
+	 * If we're using oneshot mode we don't start the flow
+	 * until each time the display is given an update, and
+	 * then we disable it immediately after. For all other
+	 * modes (command or video) we start the FIFO flow
+	 * right here. This is necessary for the hardware to
+	 * behave right.
+	 */
+	if (mcde->flow_mode != MCDE_COMMAND_ONESHOT_FLOW) {
 		mcde_enable_fifo(mcde, MCDE_FIFO_A);
+		dev_dbg(mcde->dev, "started MCDE video FIFO flow\n");
+	}
+
+	/* Enable MCDE with automatic clock gating */
+	val = readl(mcde->regs + MCDE_CR);
+	val |= MCDE_CR_MCDEEN | MCDE_CR_AUTOCLKG_EN;
+	writel(val, mcde->regs + MCDE_CR);
 
 	dev_info(drm->dev, "MCDE display is enabled\n");
 }
@@ -950,12 +1062,16 @@ static void mcde_display_disable(struct drm_simple_display_pipe *pipe)
 	struct drm_device *drm = crtc->dev;
 	struct mcde *mcde = to_mcde(drm);
 	struct drm_pending_vblank_event *event;
+	int ret;
 
 	drm_crtc_vblank_off(crtc);
 
 	/* Disable FIFO A flow */
 	mcde_disable_fifo(mcde, MCDE_FIFO_A, true);
 
+	/* This disables the DSI bridge */
+	mcde_dsi_disable(mcde->bridge);
+
 	event = crtc->state->event;
 	if (event) {
 		crtc->state->event = NULL;
@@ -965,43 +1081,47 @@ static void mcde_display_disable(struct drm_simple_display_pipe *pipe)
 		spin_unlock_irq(&crtc->dev->event_lock);
 	}
 
+	ret = regulator_disable(mcde->epod);
+	if (ret)
+		dev_err(drm->dev, "can't disable EPOD regulator\n");
+	/* Make sure we are powered down (before we may power up again) */
+	usleep_range(50000, 70000);
+
 	dev_info(drm->dev, "MCDE display is disabled\n");
 }
 
-static void mcde_display_send_one_frame(struct mcde *mcde)
+static void mcde_start_flow(struct mcde *mcde)
 {
-	/* Request a TE ACK */
-	if (mcde->te_sync)
+	/* Request a TE ACK only in TE+BTA mode */
+	if (mcde->flow_mode == MCDE_COMMAND_BTA_TE_FLOW)
 		mcde_dsi_te_request(mcde->mdsi);
 
 	/* Enable FIFO A flow */
 	mcde_enable_fifo(mcde, MCDE_FIFO_A);
 
-	if (mcde->te_sync) {
+	/*
+	 * If oneshot mode is enabled, the flow will be disabled
+	 * when the TE0 IRQ arrives in the interrupt handler. Otherwise
+	 * updates are continuously streamed to the display after this
+	 * point.
+	 */
+
+	if (mcde->flow_mode == MCDE_COMMAND_ONESHOT_FLOW) {
+		/* Trigger a software sync out on channel 0 */
+		writel(MCDE_CHNLXSYNCHSW_SW_TRIG,
+		       mcde->regs + MCDE_CHNL0SYNCHSW);
+
 		/*
-		 * If oneshot mode is enabled, the flow will be disabled
-		 * when the TE0 IRQ arrives in the interrupt handler. Otherwise
-		 * updates are continuously streamed to the display after this
-		 * point.
+		 * Disable FIFO A flow again: since we are using TE sync we
+		 * need to wait for the FIFO to drain before we continue
+		 * so repeated calls to this function will not cause a mess
+		 * in the hardware by pushing updates will updates are going
+		 * on already.
 		 */
-		dev_dbg(mcde->dev, "sent TE0 framebuffer update\n");
-		return;
+		mcde_disable_fifo(mcde, MCDE_FIFO_A, true);
 	}
 
-	/* Trigger a software sync out on channel 0 */
-	writel(MCDE_CHNLXSYNCHSW_SW_TRIG,
-	       mcde->regs + MCDE_CHNL0SYNCHSW);
-
-	/*
-	 * Disable FIFO A flow again: since we are using TE sync we
-	 * need to wait for the FIFO to drain before we continue
-	 * so repeated calls to this function will not cause a mess
-	 * in the hardware by pushing updates will updates are going
-	 * on already.
-	 */
-	mcde_disable_fifo(mcde, MCDE_FIFO_A, true);
-
-	dev_dbg(mcde->dev, "sent SW framebuffer update\n");
+	dev_dbg(mcde->dev, "started MCDE FIFO flow\n");
 }
 
 static void mcde_set_extsrc(struct mcde *mcde, u32 buffer_address)
@@ -1060,15 +1180,13 @@ static void mcde_display_update(struct drm_simple_display_pipe *pipe,
 	 */
 	if (fb) {
 		mcde_set_extsrc(mcde, drm_fb_cma_get_gem_addr(fb, pstate, 0));
-		if (!mcde->video_mode) {
-			/*
-			 * Send a single frame using software sync if the flow
-			 * is not active yet.
-			 */
-			if (mcde->flow_active == 0)
-				mcde_display_send_one_frame(mcde);
-		}
-		dev_info_once(mcde->dev, "sent first display update\n");
+		dev_info_once(mcde->dev, "first update of display contents\n");
+		/*
+		 * Usually the flow is already active, unless we are in
+		 * oneshot mode, then we need to kick the flow right here.
+		 */
+		if (mcde->flow_active == 0)
+			mcde_start_flow(mcde);
 	} else {
 		/*
 		 * If an update is receieved before the MCDE is enabled
diff --git a/drivers/gpu/drm/mcde/mcde_drm.h b/drivers/gpu/drm/mcde/mcde_drm.h
index 679c2c4e6d9d..8253e2f9993e 100644
--- a/drivers/gpu/drm/mcde/mcde_drm.h
+++ b/drivers/gpu/drm/mcde/mcde_drm.h
@@ -9,6 +9,61 @@
 #ifndef _MCDE_DRM_H_
 #define _MCDE_DRM_H_
 
+/* Shared basic registers */
+#define MCDE_CR 0x00000000
+#define MCDE_CR_IFIFOEMPTYLINECOUNT_V422_SHIFT 0
+#define MCDE_CR_IFIFOEMPTYLINECOUNT_V422_MASK 0x0000003F
+#define MCDE_CR_IFIFOCTRLEN BIT(15)
+#define MCDE_CR_UFRECOVERY_MODE_V422 BIT(16)
+#define MCDE_CR_WRAP_MODE_V422_SHIFT BIT(17)
+#define MCDE_CR_AUTOCLKG_EN BIT(30)
+#define MCDE_CR_MCDEEN BIT(31)
+
+#define MCDE_CONF0 0x00000004
+#define MCDE_CONF0_SYNCMUX0 BIT(0)
+#define MCDE_CONF0_SYNCMUX1 BIT(1)
+#define MCDE_CONF0_SYNCMUX2 BIT(2)
+#define MCDE_CONF0_SYNCMUX3 BIT(3)
+#define MCDE_CONF0_SYNCMUX4 BIT(4)
+#define MCDE_CONF0_SYNCMUX5 BIT(5)
+#define MCDE_CONF0_SYNCMUX6 BIT(6)
+#define MCDE_CONF0_SYNCMUX7 BIT(7)
+#define MCDE_CONF0_IFIFOCTRLWTRMRKLVL_SHIFT 12
+#define MCDE_CONF0_IFIFOCTRLWTRMRKLVL_MASK 0x00007000
+#define MCDE_CONF0_OUTMUX0_SHIFT 16
+#define MCDE_CONF0_OUTMUX0_MASK 0x00070000
+#define MCDE_CONF0_OUTMUX1_SHIFT 19
+#define MCDE_CONF0_OUTMUX1_MASK 0x00380000
+#define MCDE_CONF0_OUTMUX2_SHIFT 22
+#define MCDE_CONF0_OUTMUX2_MASK 0x01C00000
+#define MCDE_CONF0_OUTMUX3_SHIFT 25
+#define MCDE_CONF0_OUTMUX3_MASK 0x0E000000
+#define MCDE_CONF0_OUTMUX4_SHIFT 28
+#define MCDE_CONF0_OUTMUX4_MASK 0x70000000
+
+#define MCDE_SSP 0x00000008
+#define MCDE_AIS 0x00000100
+#define MCDE_IMSCERR 0x00000110
+#define MCDE_RISERR 0x00000120
+#define MCDE_MISERR 0x00000130
+#define MCDE_SISERR 0x00000140
+
+enum mcde_flow_mode {
+	/* One-shot mode: flow stops after one frame */
+	MCDE_COMMAND_ONESHOT_FLOW,
+	/* Command mode with tearing effect (TE) IRQ sync */
+	MCDE_COMMAND_TE_FLOW,
+	/*
+	 * Command mode with bus turn-around (BTA) and tearing effect
+	 * (TE) IRQ sync.
+	 */
+	MCDE_COMMAND_BTA_TE_FLOW,
+	/* Video mode with tearing effect (TE) sync IRQ */
+	MCDE_VIDEO_TE_FLOW,
+	/* Video mode with the formatter itself as sync source */
+	MCDE_VIDEO_FORMATTER_FLOW,
+};
+
 struct mcde {
 	struct drm_device drm;
 	struct device *dev;
@@ -18,9 +73,7 @@ struct mcde {
 	struct drm_simple_display_pipe pipe;
 	struct mipi_dsi_device *mdsi;
 	s16 stride;
-	bool te_sync;
-	bool video_mode;
-	bool oneshot_mode;
+	enum mcde_flow_mode flow_mode;
 	unsigned int flow_active;
 	spinlock_t flow_lock; /* Locks the channel flow control */
 
@@ -36,8 +89,16 @@ struct mcde {
 
 #define to_mcde(dev) container_of(dev, struct mcde, drm)
 
+static inline bool mcde_flow_is_video(struct mcde *mcde)
+{
+	return (mcde->flow_mode == MCDE_VIDEO_TE_FLOW ||
+		mcde->flow_mode == MCDE_VIDEO_FORMATTER_FLOW);
+}
+
 bool mcde_dsi_irq(struct mipi_dsi_device *mdsi);
 void mcde_dsi_te_request(struct mipi_dsi_device *mdsi);
+void mcde_dsi_enable(struct drm_bridge *bridge);
+void mcde_dsi_disable(struct drm_bridge *bridge);
 extern struct platform_driver mcde_dsi_driver;
 
 void mcde_display_irq(struct mcde *mcde);
diff --git a/drivers/gpu/drm/mcde/mcde_drv.c b/drivers/gpu/drm/mcde/mcde_drv.c
index 82137ab76cfc..c592957ed07f 100644
--- a/drivers/gpu/drm/mcde/mcde_drv.c
+++ b/drivers/gpu/drm/mcde/mcde_drv.c
@@ -63,6 +63,7 @@
 #include <linux/platform_device.h>
 #include <linux/regulator/consumer.h>
 #include <linux/slab.h>
+#include <linux/delay.h>
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
@@ -82,44 +83,6 @@
 
 #define DRIVER_DESC	"DRM module for MCDE"
 
-#define MCDE_CR 0x00000000
-#define MCDE_CR_IFIFOEMPTYLINECOUNT_V422_SHIFT 0
-#define MCDE_CR_IFIFOEMPTYLINECOUNT_V422_MASK 0x0000003F
-#define MCDE_CR_IFIFOCTRLEN BIT(15)
-#define MCDE_CR_UFRECOVERY_MODE_V422 BIT(16)
-#define MCDE_CR_WRAP_MODE_V422_SHIFT BIT(17)
-#define MCDE_CR_AUTOCLKG_EN BIT(30)
-#define MCDE_CR_MCDEEN BIT(31)
-
-#define MCDE_CONF0 0x00000004
-#define MCDE_CONF0_SYNCMUX0 BIT(0)
-#define MCDE_CONF0_SYNCMUX1 BIT(1)
-#define MCDE_CONF0_SYNCMUX2 BIT(2)
-#define MCDE_CONF0_SYNCMUX3 BIT(3)
-#define MCDE_CONF0_SYNCMUX4 BIT(4)
-#define MCDE_CONF0_SYNCMUX5 BIT(5)
-#define MCDE_CONF0_SYNCMUX6 BIT(6)
-#define MCDE_CONF0_SYNCMUX7 BIT(7)
-#define MCDE_CONF0_IFIFOCTRLWTRMRKLVL_SHIFT 12
-#define MCDE_CONF0_IFIFOCTRLWTRMRKLVL_MASK 0x00007000
-#define MCDE_CONF0_OUTMUX0_SHIFT 16
-#define MCDE_CONF0_OUTMUX0_MASK 0x00070000
-#define MCDE_CONF0_OUTMUX1_SHIFT 19
-#define MCDE_CONF0_OUTMUX1_MASK 0x00380000
-#define MCDE_CONF0_OUTMUX2_SHIFT 22
-#define MCDE_CONF0_OUTMUX2_MASK 0x01C00000
-#define MCDE_CONF0_OUTMUX3_SHIFT 25
-#define MCDE_CONF0_OUTMUX3_MASK 0x0E000000
-#define MCDE_CONF0_OUTMUX4_SHIFT 28
-#define MCDE_CONF0_OUTMUX4_MASK 0x70000000
-
-#define MCDE_SSP 0x00000008
-#define MCDE_AIS 0x00000100
-#define MCDE_IMSCERR 0x00000110
-#define MCDE_RISERR 0x00000120
-#define MCDE_MISERR 0x00000130
-#define MCDE_SISERR 0x00000140
-
 #define MCDE_PID 0x000001FC
 #define MCDE_PID_METALFIX_VERSION_SHIFT 0
 #define MCDE_PID_METALFIX_VERSION_MASK 0x000000FF
@@ -293,7 +256,6 @@ static int mcde_probe(struct platform_device *pdev)
 	struct component_match *match = NULL;
 	struct resource *res;
 	u32 pid;
-	u32 val;
 	int irq;
 	int ret;
 	int i;
@@ -305,9 +267,6 @@ static int mcde_probe(struct platform_device *pdev)
 	mcde->dev = dev;
 	platform_set_drvdata(pdev, drm);
 
-	/* Enable continuous updates: this is what Linux' framebuffer expects */
-	mcde->oneshot_mode = false;
-
 	/* First obtain and turn on the main power */
 	mcde->epod = devm_regulator_get(dev, "epod");
 	if (IS_ERR(mcde->epod)) {
@@ -405,27 +364,7 @@ static int mcde_probe(struct platform_device *pdev)
 		goto clk_disable;
 	}
 
-	/* Set up the main control, watermark level at 7 */
-	val = 7 << MCDE_CONF0_IFIFOCTRLWTRMRKLVL_SHIFT;
-	/* 24 bits DPI: connect LSB Ch B to D[0:7] */
-	val |= 3 << MCDE_CONF0_OUTMUX0_SHIFT;
-	/* TV out: connect LSB Ch B to D[8:15] */
-	val |= 3 << MCDE_CONF0_OUTMUX1_SHIFT;
-	/* Don't care about this muxing */
-	val |= 0 << MCDE_CONF0_OUTMUX2_SHIFT;
-	/* 24 bits DPI: connect MID Ch B to D[24:31] */
-	val |= 4 << MCDE_CONF0_OUTMUX3_SHIFT;
-	/* 5: 24 bits DPI: connect MSB Ch B to D[32:39] */
-	val |= 5 << MCDE_CONF0_OUTMUX4_SHIFT;
-	/* Syncmux bits zero: DPI channel A and B on output pins A and B resp */
-	writel(val, mcde->regs + MCDE_CONF0);
-
-	/* Enable automatic clock gating */
-	val = readl(mcde->regs + MCDE_CR);
-	val |= MCDE_CR_MCDEEN | MCDE_CR_AUTOCLKG_EN;
-	writel(val, mcde->regs + MCDE_CR);
-
-	/* Clear any pending interrupts */
+	/* Disable and clear any pending interrupts */
 	mcde_display_disable_irqs(mcde);
 	writel(0, mcde->regs + MCDE_IMSCERR);
 	writel(0xFFFFFFFF, mcde->regs + MCDE_RISERR);
@@ -455,12 +394,28 @@ static int mcde_probe(struct platform_device *pdev)
 		ret = PTR_ERR(match);
 		goto clk_disable;
 	}
+
+	/*
+	 * Perform an invasive reset of the MCDE and all blocks by
+	 * cutting the power to the subsystem, then bring it back up
+	 * later when we enable the display as a result of
+	 * component_master_add_with_match().
+	 */
+	ret = regulator_disable(mcde->epod);
+	if (ret) {
+		dev_err(dev, "can't disable EPOD regulator\n");
+		return ret;
+	}
+	/* Wait 50 ms so we are sure we cut the power */
+	usleep_range(50000, 70000);
+
 	ret = component_master_add_with_match(&pdev->dev, &mcde_drm_comp_ops,
 					      match);
 	if (ret) {
 		dev_err(dev, "failed to add component master\n");
 		goto clk_disable;
 	}
+
 	return 0;
 
 clk_disable:
diff --git a/drivers/gpu/drm/mcde/mcde_dsi.c b/drivers/gpu/drm/mcde/mcde_dsi.c
index 981923caa7e6..2314c8122992 100644
--- a/drivers/gpu/drm/mcde/mcde_dsi.c
+++ b/drivers/gpu/drm/mcde/mcde_dsi.c
@@ -43,6 +43,7 @@ struct mcde_dsi {
 	struct drm_bridge *bridge_out;
 	struct mipi_dsi_host dsi_host;
 	struct mipi_dsi_device *mdsi;
+	const struct drm_display_mode *mode;
 	struct clk *hs_clk;
 	struct clk *lp_clk;
 	unsigned long hs_freq;
@@ -148,9 +149,22 @@ static void mcde_dsi_attach_to_mcde(struct mcde_dsi *d)
 {
 	d->mcde->mdsi = d->mdsi;
 
-	d->mcde->video_mode = !!(d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO);
-	/* Enable use of the TE signal for all command mode panels */
-	d->mcde->te_sync = !d->mcde->video_mode;
+	/*
+	 * Select the way the DSI data flow is pushing to the display:
+	 * currently we just support video or command mode depending
+	 * on the type of display. Video mode defaults to using the
+	 * formatter itself for synchronization (stateless video panel).
+	 *
+	 * FIXME: add flags to struct mipi_dsi_device .flags to indicate
+	 * displays that require BTA (bus turn around) so we can handle
+	 * such displays as well. Figure out how to properly handle
+	 * single frame on-demand updates with DRM for command mode
+	 * displays (MCDE_COMMAND_ONESHOT_FLOW).
+	 */
+	if (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO)
+		d->mcde->flow_mode = MCDE_VIDEO_FORMATTER_FLOW;
+	else
+		d->mcde->flow_mode = MCDE_COMMAND_TE_FLOW;
 }
 
 static int mcde_dsi_host_attach(struct mipi_dsi_host *host,
@@ -194,79 +208,16 @@ static int mcde_dsi_host_detach(struct mipi_dsi_host *host,
 	 (type == MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM) || \
 	 (type == MIPI_DSI_DCS_READ))
 
-static ssize_t mcde_dsi_host_transfer(struct mipi_dsi_host *host,
-				      const struct mipi_dsi_msg *msg)
+static int mcde_dsi_execute_transfer(struct mcde_dsi *d,
+				     const struct mipi_dsi_msg *msg)
 {
-	struct mcde_dsi *d = host_to_mcde_dsi(host);
 	const u32 loop_delay_us = 10; /* us */
-	const u8 *tx = msg->tx_buf;
 	u32 loop_counter;
 	size_t txlen = msg->tx_len;
 	size_t rxlen = msg->rx_len;
+	int i;
 	u32 val;
 	int ret;
-	int i;
-
-	if (txlen > 16) {
-		dev_err(d->dev,
-			"dunno how to write more than 16 bytes yet\n");
-		return -EIO;
-	}
-	if (rxlen > 4) {
-		dev_err(d->dev,
-			"dunno how to read more than 4 bytes yet\n");
-		return -EIO;
-	}
-
-	dev_dbg(d->dev,
-		"message to channel %d, write %zd bytes read %zd bytes\n",
-		msg->channel, txlen, rxlen);
-
-	/* Command "nature" */
-	if (MCDE_DSI_HOST_IS_READ(msg->type))
-		/* MCTL_MAIN_DATA_CTL already set up */
-		val = DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_NAT_READ;
-	else
-		val = DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_NAT_WRITE;
-	/*
-	 * More than 2 bytes will not fit in a single packet, so it's
-	 * time to set the "long not short" bit. One byte is used by
-	 * the MIPI DCS command leaving just one byte for the payload
-	 * in a short package.
-	 */
-	if (mipi_dsi_packet_format_is_long(msg->type))
-		val |= DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_LONGNOTSHORT;
-	val |= 0 << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_ID_SHIFT;
-	val |= txlen << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_SIZE_SHIFT;
-	val |= DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_LP_EN;
-	val |= msg->type << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_HEAD_SHIFT;
-	writel(val, d->regs + DSI_DIRECT_CMD_MAIN_SETTINGS);
-
-	/* MIPI DCS command is part of the data */
-	if (txlen > 0) {
-		val = 0;
-		for (i = 0; i < 4 && i < txlen; i++)
-			val |= tx[i] << (i * 8);
-	}
-	writel(val, d->regs + DSI_DIRECT_CMD_WRDAT0);
-	if (txlen > 4) {
-		val = 0;
-		for (i = 0; i < 4 && (i + 4) < txlen; i++)
-			val |= tx[i + 4] << (i * 8);
-		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT1);
-	}
-	if (txlen > 8) {
-		val = 0;
-		for (i = 0; i < 4 && (i + 8) < txlen; i++)
-			val |= tx[i + 8] << (i * 8);
-		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT2);
-	}
-	if (txlen > 12) {
-		val = 0;
-		for (i = 0; i < 4 && (i + 12) < txlen; i++)
-			val |= tx[i + 12] << (i * 8);
-		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT3);
-	}
 
 	writel(~0, d->regs + DSI_DIRECT_CMD_STS_CLR);
 	writel(~0, d->regs + DSI_CMD_MODE_STS_CLR);
@@ -283,6 +234,7 @@ static ssize_t mcde_dsi_host_transfer(struct mipi_dsi_host *host,
 			usleep_range(loop_delay_us, (loop_delay_us * 3) / 2);
 		if (!loop_counter) {
 			dev_err(d->dev, "DSI read timeout!\n");
+			/* Set exit code and retry */
 			return -ETIME;
 		}
 	} else {
@@ -293,6 +245,7 @@ static ssize_t mcde_dsi_host_transfer(struct mipi_dsi_host *host,
 			usleep_range(loop_delay_us, (loop_delay_us * 3) / 2);
 
 		if (!loop_counter) {
+			/* Set exit code and retry */
 			dev_err(d->dev, "DSI write timeout!\n");
 			return -ETIME;
 		}
@@ -334,6 +287,93 @@ static ssize_t mcde_dsi_host_transfer(struct mipi_dsi_host *host,
 		ret = rdsz;
 	}
 
+	/* Successful transmission */
+	return ret;
+}
+
+static ssize_t mcde_dsi_host_transfer(struct mipi_dsi_host *host,
+				      const struct mipi_dsi_msg *msg)
+{
+	struct mcde_dsi *d = host_to_mcde_dsi(host);
+	const u8 *tx = msg->tx_buf;
+	size_t txlen = msg->tx_len;
+	size_t rxlen = msg->rx_len;
+	unsigned int retries = 0;
+	u32 val;
+	int ret;
+	int i;
+
+	if (txlen > 16) {
+		dev_err(d->dev,
+			"dunno how to write more than 16 bytes yet\n");
+		return -EIO;
+	}
+	if (rxlen > 4) {
+		dev_err(d->dev,
+			"dunno how to read more than 4 bytes yet\n");
+		return -EIO;
+	}
+
+	dev_dbg(d->dev,
+		"message to channel %d, write %zd bytes read %zd bytes\n",
+		msg->channel, txlen, rxlen);
+
+	/* Command "nature" */
+	if (MCDE_DSI_HOST_IS_READ(msg->type))
+		/* MCTL_MAIN_DATA_CTL already set up */
+		val = DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_NAT_READ;
+	else
+		val = DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_NAT_WRITE;
+	/*
+	 * More than 2 bytes will not fit in a single packet, so it's
+	 * time to set the "long not short" bit. One byte is used by
+	 * the MIPI DCS command leaving just one byte for the payload
+	 * in a short package.
+	 */
+	if (mipi_dsi_packet_format_is_long(msg->type))
+		val |= DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_LONGNOTSHORT;
+	val |= 0 << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_ID_SHIFT;
+	val |= txlen << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_SIZE_SHIFT;
+	val |= DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_LP_EN;
+	val |= msg->type << DSI_DIRECT_CMD_MAIN_SETTINGS_CMD_HEAD_SHIFT;
+	writel(val, d->regs + DSI_DIRECT_CMD_MAIN_SETTINGS);
+
+	/* MIPI DCS command is part of the data */
+	if (txlen > 0) {
+		val = 0;
+		for (i = 0; i < 4 && i < txlen; i++)
+			val |= tx[i] << (i * 8);
+	}
+	writel(val, d->regs + DSI_DIRECT_CMD_WRDAT0);
+	if (txlen > 4) {
+		val = 0;
+		for (i = 0; i < 4 && (i + 4) < txlen; i++)
+			val |= tx[i + 4] << (i * 8);
+		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT1);
+	}
+	if (txlen > 8) {
+		val = 0;
+		for (i = 0; i < 4 && (i + 8) < txlen; i++)
+			val |= tx[i + 8] << (i * 8);
+		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT2);
+	}
+	if (txlen > 12) {
+		val = 0;
+		for (i = 0; i < 4 && (i + 12) < txlen; i++)
+			val |= tx[i + 12] << (i * 8);
+		writel(val, d->regs + DSI_DIRECT_CMD_WRDAT3);
+	}
+
+	while (retries < 3) {
+		ret = mcde_dsi_execute_transfer(d, msg);
+		if (ret >= 0)
+			break;
+		retries++;
+	}
+	if (ret < 0 && retries)
+		dev_err(d->dev, "gave up after %d retries\n", retries);
+
+	/* Clear any errors */
 	writel(~0, d->regs + DSI_DIRECT_CMD_STS_CLR);
 	writel(~0, d->regs + DSI_CMD_MODE_STS_CLR);
 
@@ -799,10 +839,11 @@ static void mcde_dsi_start(struct mcde_dsi *d)
 	/* Command mode, clear IF1 ID */
 	val = readl(d->regs + DSI_CMD_MODE_CTL);
 	/*
-	 * If we enable low-power mode here, with
-	 * val |= DSI_CMD_MODE_CTL_IF1_LP_EN
+	 * If we enable low-power mode here,
 	 * then display updates become really slow.
 	 */
+	if (d->mdsi->mode_flags & MIPI_DSI_MODE_LPM)
+		val |= DSI_CMD_MODE_CTL_IF1_LP_EN;
 	val &= ~DSI_CMD_MODE_CTL_IF1_ID_MASK;
 	writel(val, d->regs + DSI_CMD_MODE_CTL);
 
@@ -811,23 +852,11 @@ static void mcde_dsi_start(struct mcde_dsi *d)
 	dev_info(d->dev, "DSI link enabled\n");
 }
 
-
-static void mcde_dsi_bridge_enable(struct drm_bridge *bridge)
-{
-	struct mcde_dsi *d = bridge_to_mcde_dsi(bridge);
-	u32 val;
-
-	if (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
-		/* Enable video mode */
-		val = readl(d->regs + DSI_MCTL_MAIN_DATA_CTL);
-		val |= DSI_MCTL_MAIN_DATA_CTL_VID_EN;
-		writel(val, d->regs + DSI_MCTL_MAIN_DATA_CTL);
-	}
-
-	dev_info(d->dev, "enable DSI master\n");
-};
-
-static void mcde_dsi_bridge_pre_enable(struct drm_bridge *bridge)
+/*
+ * Notice that this is called from inside the display controller
+ * and not from the bridge callbacks.
+ */
+void mcde_dsi_enable(struct drm_bridge *bridge)
 {
 	struct mcde_dsi *d = bridge_to_mcde_dsi(bridge);
 	unsigned long hs_freq, lp_freq;
@@ -871,7 +900,25 @@ static void mcde_dsi_bridge_pre_enable(struct drm_bridge *bridge)
 		dev_info(d->dev, "DSI HS clock rate %lu Hz\n",
 			 d->hs_freq);
 
+	/* Assert RESET through the PRCMU, active low */
+	/* FIXME: which DSI block? */
+	regmap_update_bits(d->prcmu, PRCM_DSI_SW_RESET,
+			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN, 0);
+
+	usleep_range(100, 200);
+
+	/* De-assert RESET again */
+	regmap_update_bits(d->prcmu, PRCM_DSI_SW_RESET,
+			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN,
+			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN);
+
+	/* Start up the hardware */
+	mcde_dsi_start(d);
+
 	if (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
+		/* Set up the video mode from the DRM mode */
+		mcde_dsi_setup_video_mode(d, d->mode);
+
 		/* Put IF1 into video mode */
 		val = readl(d->regs + DSI_MCTL_MAIN_DATA_CTL);
 		val |= DSI_MCTL_MAIN_DATA_CTL_IF1_MODE;
@@ -887,17 +934,25 @@ static void mcde_dsi_bridge_pre_enable(struct drm_bridge *bridge)
 		val |= DSI_VID_MODE_STS_CTL_ERR_MISSING_VSYNC;
 		val |= DSI_VID_MODE_STS_CTL_ERR_MISSING_DATA;
 		writel(val, d->regs + DSI_VID_MODE_STS_CTL);
+
+		/* Enable video mode */
+		val = readl(d->regs + DSI_MCTL_MAIN_DATA_CTL);
+		val |= DSI_MCTL_MAIN_DATA_CTL_VID_EN;
+		writel(val, d->regs + DSI_MCTL_MAIN_DATA_CTL);
 	} else {
 		/* Command mode, clear IF1 ID */
 		val = readl(d->regs + DSI_CMD_MODE_CTL);
 		/*
-		 * If we enable low-power mode here with
-		 * val |= DSI_CMD_MODE_CTL_IF1_LP_EN
+		 * If we enable low-power mode here
 		 * the display updates become really slow.
 		 */
+		if (d->mdsi->mode_flags & MIPI_DSI_MODE_LPM)
+			val |= DSI_CMD_MODE_CTL_IF1_LP_EN;
 		val &= ~DSI_CMD_MODE_CTL_IF1_ID_MASK;
 		writel(val, d->regs + DSI_CMD_MODE_CTL);
 	}
+
+	dev_info(d->dev, "enabled MCDE DSI master\n");
 }
 
 static void mcde_dsi_bridge_mode_set(struct drm_bridge *bridge,
@@ -911,13 +966,12 @@ static void mcde_dsi_bridge_mode_set(struct drm_bridge *bridge,
 		return;
 	}
 
+	d->mode = mode;
+
 	dev_info(d->dev, "set DSI master to %dx%d %u Hz %s mode\n",
 		 mode->hdisplay, mode->vdisplay, mode->clock * 1000,
 		 (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO) ? "VIDEO" : "CMD"
 		);
-
-	if (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO)
-		mcde_dsi_setup_video_mode(d, mode);
 }
 
 static void mcde_dsi_wait_for_command_mode_stop(struct mcde_dsi *d)
@@ -961,14 +1015,15 @@ static void mcde_dsi_wait_for_video_mode_stop(struct mcde_dsi *d)
 	}
 }
 
-static void mcde_dsi_bridge_disable(struct drm_bridge *bridge)
+/*
+ * Notice that this is called from inside the display controller
+ * and not from the bridge callbacks.
+ */
+void mcde_dsi_disable(struct drm_bridge *bridge)
 {
 	struct mcde_dsi *d = bridge_to_mcde_dsi(bridge);
 	u32 val;
 
-	/* Disable all error interrupts */
-	writel(0, d->regs + DSI_VID_MODE_STS_CTL);
-
 	if (d->mdsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
 		/* Stop video mode */
 		val = readl(d->regs + DSI_MCTL_MAIN_DATA_CTL);
@@ -980,7 +1035,14 @@ static void mcde_dsi_bridge_disable(struct drm_bridge *bridge)
 		mcde_dsi_wait_for_command_mode_stop(d);
 	}
 
-	/* Stop clocks */
+	/*
+	 * Stop clocks and terminate any DSI traffic here so the panel can
+	 * send commands to shut down the display using DSI direct write until
+	 * this point.
+	 */
+
+	/* Disable all error interrupts */
+	writel(0, d->regs + DSI_VID_MODE_STS_CTL);
 	clk_disable_unprepare(d->hs_clk);
 	clk_disable_unprepare(d->lp_clk);
 }
@@ -1010,9 +1072,6 @@ static int mcde_dsi_bridge_attach(struct drm_bridge *bridge,
 static const struct drm_bridge_funcs mcde_dsi_bridge_funcs = {
 	.attach = mcde_dsi_bridge_attach,
 	.mode_set = mcde_dsi_bridge_mode_set,
-	.disable = mcde_dsi_bridge_disable,
-	.enable = mcde_dsi_bridge_enable,
-	.pre_enable = mcde_dsi_bridge_pre_enable,
 };
 
 static int mcde_dsi_bind(struct device *dev, struct device *master,
@@ -1048,21 +1107,6 @@ static int mcde_dsi_bind(struct device *dev, struct device *master,
 		return PTR_ERR(d->lp_clk);
 	}
 
-	/* Assert RESET through the PRCMU, active low */
-	/* FIXME: which DSI block? */
-	regmap_update_bits(d->prcmu, PRCM_DSI_SW_RESET,
-			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN, 0);
-
-	usleep_range(100, 200);
-
-	/* De-assert RESET again */
-	regmap_update_bits(d->prcmu, PRCM_DSI_SW_RESET,
-			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN,
-			   PRCM_DSI_SW_RESET_DSI0_SW_RESETN);
-
-	/* Start up the hardware */
-	mcde_dsi_start(d);
-
 	/* Look for a panel as a child to this node */
 	for_each_available_child_of_node(dev->of_node, child) {
 		panel = of_drm_find_panel(child);
diff --git a/drivers/gpu/drm/mediatek/Kconfig b/drivers/gpu/drm/mediatek/Kconfig
index aa74aac3cbcc..65cd03a4be29 100644
--- a/drivers/gpu/drm/mediatek/Kconfig
+++ b/drivers/gpu/drm/mediatek/Kconfig
@@ -24,6 +24,6 @@ config DRM_MEDIATEK_HDMI
 	tristate "DRM HDMI Support for Mediatek SoCs"
 	depends on DRM_MEDIATEK
 	select SND_SOC_HDMI_CODEC if SND_SOC
-	select GENERIC_PHY
+	select PHY_MTK_HDMI
 	help
 	  DRM/KMS HDMI driver for Mediatek SoCs
diff --git a/drivers/gpu/drm/mediatek/Makefile b/drivers/gpu/drm/mediatek/Makefile
index b7a82ed5788f..77b0fd86063d 100644
--- a/drivers/gpu/drm/mediatek/Makefile
+++ b/drivers/gpu/drm/mediatek/Makefile
@@ -19,9 +19,6 @@ obj-$(CONFIG_DRM_MEDIATEK) += mediatek-drm.o
 
 mediatek-drm-hdmi-objs := mtk_cec.o \
 			  mtk_hdmi.o \
-			  mtk_hdmi_ddc.o \
-			  mtk_mt2701_hdmi_phy.o \
-			  mtk_mt8173_hdmi_phy.o \
-			  mtk_hdmi_phy.o
+			  mtk_hdmi_ddc.o
 
 obj-$(CONFIG_DRM_MEDIATEK_HDMI) += mediatek-drm-hdmi.o
diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c
index d4f0fb7ad312..cf11c4850b40 100644
--- a/drivers/gpu/drm/mediatek/mtk_dpi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dpi.c
@@ -64,7 +64,8 @@ enum mtk_dpi_out_color_format {
 struct mtk_dpi {
 	struct mtk_ddp_comp ddp_comp;
 	struct drm_encoder encoder;
-	struct drm_bridge *bridge;
+	struct drm_bridge bridge;
+	struct drm_bridge *next_bridge;
 	void __iomem *regs;
 	struct device *dev;
 	struct clk *engine_clk;
@@ -83,9 +84,9 @@ struct mtk_dpi {
 	int refcount;
 };
 
-static inline struct mtk_dpi *mtk_dpi_from_encoder(struct drm_encoder *e)
+static inline struct mtk_dpi *bridge_to_dpi(struct drm_bridge *b)
 {
-	return container_of(e, struct mtk_dpi, encoder);
+	return container_of(b, struct mtk_dpi, bridge);
 }
 
 enum mtk_dpi_polarity {
@@ -521,50 +522,53 @@ static int mtk_dpi_set_display_mode(struct mtk_dpi *dpi,
 	return 0;
 }
 
-static bool mtk_dpi_encoder_mode_fixup(struct drm_encoder *encoder,
-				       const struct drm_display_mode *mode,
-				       struct drm_display_mode *adjusted_mode)
+static void mtk_dpi_encoder_destroy(struct drm_encoder *encoder)
 {
-	return true;
+	drm_encoder_cleanup(encoder);
 }
 
-static void mtk_dpi_encoder_mode_set(struct drm_encoder *encoder,
-				     struct drm_display_mode *mode,
-				     struct drm_display_mode *adjusted_mode)
+static const struct drm_encoder_funcs mtk_dpi_encoder_funcs = {
+	.destroy = mtk_dpi_encoder_destroy,
+};
+
+static int mtk_dpi_bridge_attach(struct drm_bridge *bridge,
+				 enum drm_bridge_attach_flags flags)
 {
-	struct mtk_dpi *dpi = mtk_dpi_from_encoder(encoder);
+	struct mtk_dpi *dpi = bridge_to_dpi(bridge);
+
+	return drm_bridge_attach(bridge->encoder, dpi->next_bridge,
+				 &dpi->bridge, flags);
+}
+
+static void mtk_dpi_bridge_mode_set(struct drm_bridge *bridge,
+				const struct drm_display_mode *mode,
+				const struct drm_display_mode *adjusted_mode)
+{
+	struct mtk_dpi *dpi = bridge_to_dpi(bridge);
 
 	drm_mode_copy(&dpi->mode, adjusted_mode);
 }
 
-static void mtk_dpi_encoder_disable(struct drm_encoder *encoder)
+static void mtk_dpi_bridge_disable(struct drm_bridge *bridge)
 {
-	struct mtk_dpi *dpi = mtk_dpi_from_encoder(encoder);
+	struct mtk_dpi *dpi = bridge_to_dpi(bridge);
 
 	mtk_dpi_power_off(dpi);
 }
 
-static void mtk_dpi_encoder_enable(struct drm_encoder *encoder)
+static void mtk_dpi_bridge_enable(struct drm_bridge *bridge)
 {
-	struct mtk_dpi *dpi = mtk_dpi_from_encoder(encoder);
+	struct mtk_dpi *dpi = bridge_to_dpi(bridge);
 
 	mtk_dpi_power_on(dpi);
 	mtk_dpi_set_display_mode(dpi, &dpi->mode);
 }
 
-static int mtk_dpi_atomic_check(struct drm_encoder *encoder,
-				struct drm_crtc_state *crtc_state,
-				struct drm_connector_state *conn_state)
-{
-	return 0;
-}
-
-static const struct drm_encoder_helper_funcs mtk_dpi_encoder_helper_funcs = {
-	.mode_fixup = mtk_dpi_encoder_mode_fixup,
-	.mode_set = mtk_dpi_encoder_mode_set,
-	.disable = mtk_dpi_encoder_disable,
-	.enable = mtk_dpi_encoder_enable,
-	.atomic_check = mtk_dpi_atomic_check,
+static const struct drm_bridge_funcs mtk_dpi_bridge_funcs = {
+	.attach = mtk_dpi_bridge_attach,
+	.mode_set = mtk_dpi_bridge_mode_set,
+	.disable = mtk_dpi_bridge_disable,
+	.enable = mtk_dpi_bridge_enable,
 };
 
 static void mtk_dpi_start(struct mtk_ddp_comp *comp)
@@ -605,12 +609,10 @@ static int mtk_dpi_bind(struct device *dev, struct device *master, void *data)
 		dev_err(dev, "Failed to initialize decoder: %d\n", ret);
 		goto err_unregister;
 	}
-	drm_encoder_helper_add(&dpi->encoder, &mtk_dpi_encoder_helper_funcs);
 
-	/* Currently DPI0 is fixed to be driven by OVL1 */
-	dpi->encoder.possible_crtcs = BIT(1);
+	dpi->encoder.possible_crtcs = mtk_drm_find_possible_crtc_by_comp(drm_dev, dpi->ddp_comp);
 
-	ret = drm_bridge_attach(&dpi->encoder, dpi->bridge, NULL, 0);
+	ret = drm_bridge_attach(&dpi->encoder, &dpi->bridge, NULL, 0);
 	if (ret) {
 		dev_err(dev, "Failed to attach bridge: %d\n", ret);
 		goto err_cleanup;
@@ -770,11 +772,11 @@ static int mtk_dpi_probe(struct platform_device *pdev)
 	}
 
 	ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0,
-					  NULL, &dpi->bridge);
+					  NULL, &dpi->next_bridge);
 	if (ret)
 		return ret;
 
-	dev_info(dev, "Found bridge node: %pOF\n", dpi->bridge->of_node);
+	dev_info(dev, "Found bridge node: %pOF\n", dpi->next_bridge->of_node);
 
 	comp_id = mtk_ddp_comp_get_id(dev->of_node, MTK_DPI);
 	if (comp_id < 0) {
@@ -791,8 +793,15 @@ static int mtk_dpi_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, dpi);
 
+	dpi->bridge.funcs = &mtk_dpi_bridge_funcs;
+	dpi->bridge.of_node = dev->of_node;
+	dpi->bridge.type = DRM_MODE_CONNECTOR_DPI;
+
+	drm_bridge_add(&dpi->bridge);
+
 	ret = component_add(dev, &mtk_dpi_component_ops);
 	if (ret) {
+		drm_bridge_remove(&dpi->bridge);
 		dev_err(dev, "Failed to add component: %d\n", ret);
 		return ret;
 	}
@@ -802,7 +811,10 @@ static int mtk_dpi_probe(struct platform_device *pdev)
 
 static int mtk_dpi_remove(struct platform_device *pdev)
 {
+	struct mtk_dpi *dpi = platform_get_drvdata(pdev);
+
 	component_del(&pdev->dev, &mtk_dpi_component_ops);
+	drm_bridge_remove(&dpi->bridge);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
index 526648885b97..8eba44be3a8a 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
@@ -13,6 +13,8 @@
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
 #include <linux/soc/mediatek/mtk-cmdq.h>
+#include <drm/drm_print.h>
+
 #include "mtk_drm_drv.h"
 #include "mtk_drm_plane.h"
 #include "mtk_drm_ddp_comp.h"
@@ -412,6 +414,22 @@ static const struct mtk_ddp_comp_match mtk_ddp_matches[DDP_COMPONENT_ID_MAX] = {
 	[DDP_COMPONENT_WDMA1]	= { MTK_DISP_WDMA,	1, NULL },
 };
 
+static bool mtk_drm_find_comp_in_ddp(struct mtk_ddp_comp ddp_comp,
+				     const enum mtk_ddp_comp_id *path,
+				     unsigned int path_len)
+{
+	unsigned int i;
+
+	if (path == NULL)
+		return false;
+
+	for (i = 0U; i < path_len; i++)
+		if (ddp_comp.id == path[i])
+			return true;
+
+	return false;
+}
+
 int mtk_ddp_comp_get_id(struct device_node *node,
 			enum mtk_ddp_comp_type comp_type)
 {
@@ -427,6 +445,26 @@ int mtk_ddp_comp_get_id(struct device_node *node,
 	return -EINVAL;
 }
 
+unsigned int mtk_drm_find_possible_crtc_by_comp(struct drm_device *drm,
+						struct mtk_ddp_comp ddp_comp)
+{
+	struct mtk_drm_private *private = drm->dev_private;
+	unsigned int ret = 0;
+
+	if (mtk_drm_find_comp_in_ddp(ddp_comp, private->data->main_path, private->data->main_len))
+		ret = BIT(0);
+	else if (mtk_drm_find_comp_in_ddp(ddp_comp, private->data->ext_path,
+					  private->data->ext_len))
+		ret = BIT(1);
+	else if (mtk_drm_find_comp_in_ddp(ddp_comp, private->data->third_path,
+					  private->data->third_len))
+		ret = BIT(2);
+	else
+		DRM_INFO("Failed to find comp in ddp table\n");
+
+	return ret;
+}
+
 int mtk_ddp_comp_init(struct device *dev, struct device_node *node,
 		      struct mtk_ddp_comp *comp, enum mtk_ddp_comp_id comp_id,
 		      const struct mtk_ddp_comp_funcs *funcs)
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.h b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.h
index debe36395fe7..1d9e00b69462 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.h
+++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.h
@@ -202,6 +202,8 @@ static inline void mtk_ddp_ctm_set(struct mtk_ddp_comp *comp,
 
 int mtk_ddp_comp_get_id(struct device_node *node,
 			enum mtk_ddp_comp_type comp_type);
+unsigned int mtk_drm_find_possible_crtc_by_comp(struct drm_device *drm,
+						struct mtk_ddp_comp ddp_comp);
 int mtk_ddp_comp_init(struct device *dev, struct device_node *comp_node,
 		      struct mtk_ddp_comp *comp, enum mtk_ddp_comp_id comp_id,
 		      const struct mtk_ddp_comp_funcs *funcs);
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
index 2d982740b1a4..59c85c63b7cc 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
@@ -73,6 +73,19 @@ static const enum mtk_ddp_comp_id mt2701_mtk_ddp_ext[] = {
 	DDP_COMPONENT_DPI0,
 };
 
+static const enum mtk_ddp_comp_id mt7623_mtk_ddp_main[] = {
+	DDP_COMPONENT_OVL0,
+	DDP_COMPONENT_RDMA0,
+	DDP_COMPONENT_COLOR0,
+	DDP_COMPONENT_BLS,
+	DDP_COMPONENT_DPI0,
+};
+
+static const enum mtk_ddp_comp_id mt7623_mtk_ddp_ext[] = {
+	DDP_COMPONENT_RDMA1,
+	DDP_COMPONENT_DSI0,
+};
+
 static const enum mtk_ddp_comp_id mt2712_mtk_ddp_main[] = {
 	DDP_COMPONENT_OVL0,
 	DDP_COMPONENT_COLOR0,
@@ -126,6 +139,14 @@ static const struct mtk_mmsys_driver_data mt2701_mmsys_driver_data = {
 	.shadow_register = true,
 };
 
+static const struct mtk_mmsys_driver_data mt7623_mmsys_driver_data = {
+	.main_path = mt7623_mtk_ddp_main,
+	.main_len = ARRAY_SIZE(mt7623_mtk_ddp_main),
+	.ext_path = mt7623_mtk_ddp_ext,
+	.ext_len = ARRAY_SIZE(mt7623_mtk_ddp_ext),
+	.shadow_register = true,
+};
+
 static const struct mtk_mmsys_driver_data mt2712_mmsys_driver_data = {
 	.main_path = mt2712_mtk_ddp_main,
 	.main_len = ARRAY_SIZE(mt2712_mtk_ddp_main),
@@ -424,6 +445,8 @@ static const struct of_device_id mtk_ddp_comp_dt_ids[] = {
 static const struct of_device_id mtk_drm_of_ids[] = {
 	{ .compatible = "mediatek,mt2701-mmsys",
 	  .data = &mt2701_mmsys_driver_data},
+	{ .compatible = "mediatek,mt7623-mmsys",
+	  .data = &mt7623_mmsys_driver_data},
 	{ .compatible = "mediatek,mt2712-mmsys",
 	  .data = &mt2712_mmsys_driver_data},
 	{ .compatible = "mediatek,mt8173-mmsys",
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.c b/drivers/gpu/drm/mediatek/mtk_drm_gem.c
index 6190cc3b7b0d..0583e557ad37 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_gem.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.c
@@ -212,46 +212,28 @@ struct drm_gem_object *mtk_gem_prime_import_sg_table(struct drm_device *dev,
 			struct dma_buf_attachment *attach, struct sg_table *sg)
 {
 	struct mtk_drm_gem_obj *mtk_gem;
-	int ret;
-	struct scatterlist *s;
-	unsigned int i;
-	dma_addr_t expected;
 
-	mtk_gem = mtk_drm_gem_init(dev, attach->dmabuf->size);
+	/* check if the entries in the sg_table are contiguous */
+	if (drm_prime_get_contiguous_size(sg) < attach->dmabuf->size) {
+		DRM_ERROR("sg_table is not contiguous");
+		return ERR_PTR(-EINVAL);
+	}
 
+	mtk_gem = mtk_drm_gem_init(dev, attach->dmabuf->size);
 	if (IS_ERR(mtk_gem))
 		return ERR_CAST(mtk_gem);
 
-	expected = sg_dma_address(sg->sgl);
-	for_each_sg(sg->sgl, s, sg->nents, i) {
-		if (!sg_dma_len(s))
-			break;
-
-		if (sg_dma_address(s) != expected) {
-			DRM_ERROR("sg_table is not contiguous");
-			ret = -EINVAL;
-			goto err_gem_free;
-		}
-		expected = sg_dma_address(s) + sg_dma_len(s);
-	}
-
 	mtk_gem->dma_addr = sg_dma_address(sg->sgl);
 	mtk_gem->sg = sg;
 
 	return &mtk_gem->base;
-
-err_gem_free:
-	kfree(mtk_gem);
-	return ERR_PTR(ret);
 }
 
 void *mtk_drm_gem_prime_vmap(struct drm_gem_object *obj)
 {
 	struct mtk_drm_gem_obj *mtk_gem = to_mtk_gem_obj(obj);
 	struct sg_table *sgt;
-	struct sg_page_iter iter;
 	unsigned int npages;
-	unsigned int i = 0;
 
 	if (mtk_gem->kvaddr)
 		return mtk_gem->kvaddr;
@@ -265,11 +247,8 @@ void *mtk_drm_gem_prime_vmap(struct drm_gem_object *obj)
 	if (!mtk_gem->pages)
 		goto out;
 
-	for_each_sg_page(sgt->sgl, &iter, sgt->orig_nents, 0) {
-		mtk_gem->pages[i++] = sg_page_iter_page(&iter);
-		if (i > npages)
-			break;
-	}
+	drm_prime_sg_to_page_addr_arrays(sgt, mtk_gem->pages, NULL, npages);
+
 	mtk_gem->kvaddr = vmap(mtk_gem->pages, npages, VM_MAP,
 			       pgprot_writecombine(PAGE_KERNEL));
 
diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c
index 80b7a082e874..4a188a942c38 100644
--- a/drivers/gpu/drm/mediatek/mtk_dsi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dsi.c
@@ -969,11 +969,7 @@ static int mtk_dsi_encoder_init(struct drm_device *drm, struct mtk_dsi *dsi)
 		return ret;
 	}
 
-	/*
-	 * Currently display data paths are statically assigned to a crtc each.
-	 * crtc 0 is OVL0 -> COLOR0 -> AAL -> OD -> RDMA0 -> UFOE -> DSI0
-	 */
-	dsi->encoder.possible_crtcs = 1;
+	dsi->encoder.possible_crtcs = mtk_drm_find_possible_crtc_by_comp(drm, dsi->ddp_comp);
 
 	ret = drm_bridge_attach(&dsi->encoder, &dsi->bridge, NULL,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c
index a97725680d4e..97a1ff529a1d 100644
--- a/drivers/gpu/drm/mediatek/mtk_hdmi.c
+++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c
@@ -12,6 +12,7 @@
 #include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/mfd/syscon.h>
+#include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/of_platform.h>
 #include <linux/of.h>
@@ -145,11 +146,16 @@ struct hdmi_audio_param {
 	struct hdmi_codec_params codec_params;
 };
 
+struct mtk_hdmi_conf {
+	bool tz_disabled;
+};
+
 struct mtk_hdmi {
 	struct drm_bridge bridge;
 	struct drm_bridge *next_bridge;
 	struct drm_connector conn;
 	struct device *dev;
+	const struct mtk_hdmi_conf *conf;
 	struct phy *phy;
 	struct device *cec_dev;
 	struct i2c_adapter *ddc_adpt;
@@ -234,7 +240,6 @@ static void mtk_hdmi_hw_vid_black(struct mtk_hdmi *hdmi, bool black)
 static void mtk_hdmi_hw_make_reg_writable(struct mtk_hdmi *hdmi, bool enable)
 {
 	struct arm_smccc_res res;
-	struct mtk_hdmi_phy *hdmi_phy = phy_get_drvdata(hdmi->phy);
 
 	/*
 	 * MT8173 HDMI hardware has an output control bit to enable/disable HDMI
@@ -242,7 +247,7 @@ static void mtk_hdmi_hw_make_reg_writable(struct mtk_hdmi *hdmi, bool enable)
 	 * The ARM trusted firmware provides an API for the HDMI driver to set
 	 * this control bit to enable HDMI output in supervisor mode.
 	 */
-	if (hdmi_phy->conf && hdmi_phy->conf->tz_disabled)
+	if (hdmi->conf && hdmi->conf->tz_disabled)
 		regmap_update_bits(hdmi->sys_regmap,
 				   hdmi->sys_offset + HDMI_SYS_CFG20,
 				   0x80008005, enable ? 0x80000005 : 0x8000);
@@ -1733,6 +1738,7 @@ static int mtk_drm_hdmi_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	hdmi->dev = dev;
+	hdmi->conf = of_device_get_match_data(dev);
 
 	ret = mtk_hdmi_dt_parse_pdata(hdmi, pdev);
 	if (ret)
@@ -1813,8 +1819,16 @@ static int mtk_hdmi_resume(struct device *dev)
 static SIMPLE_DEV_PM_OPS(mtk_hdmi_pm_ops,
 			 mtk_hdmi_suspend, mtk_hdmi_resume);
 
+static const struct mtk_hdmi_conf mtk_hdmi_conf_mt2701 = {
+	.tz_disabled = true,
+};
+
 static const struct of_device_id mtk_drm_hdmi_of_ids[] = {
-	{ .compatible = "mediatek,mt8173-hdmi", },
+	{ .compatible = "mediatek,mt2701-hdmi",
+	  .data = &mtk_hdmi_conf_mt2701,
+	},
+	{ .compatible = "mediatek,mt8173-hdmi",
+	},
 	{}
 };
 
@@ -1829,7 +1843,6 @@ static struct platform_driver mtk_hdmi_driver = {
 };
 
 static struct platform_driver * const mtk_hdmi_drivers[] = {
-	&mtk_hdmi_phy_driver,
 	&mtk_hdmi_ddc_driver,
 	&mtk_cec_driver,
 	&mtk_hdmi_driver,
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.h b/drivers/gpu/drm/mediatek/mtk_hdmi.h
index bb3653de6bd1..472bf141c92b 100644
--- a/drivers/gpu/drm/mediatek/mtk_hdmi.h
+++ b/drivers/gpu/drm/mediatek/mtk_hdmi.h
@@ -5,7 +5,6 @@
  */
 #ifndef _MTK_HDMI_CTRL_H
 #define _MTK_HDMI_CTRL_H
-#include "mtk_hdmi_phy.h"
 
 struct platform_driver;
 
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c
deleted file mode 100644
index 5223498502c4..000000000000
--- a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c
+++ /dev/null
@@ -1,210 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2018 MediaTek Inc.
- * Author: Jie Qiu <jie.qiu@mediatek.com>
- */
-
-#include "mtk_hdmi_phy.h"
-
-static int mtk_hdmi_phy_power_on(struct phy *phy);
-static int mtk_hdmi_phy_power_off(struct phy *phy);
-
-static const struct phy_ops mtk_hdmi_phy_dev_ops = {
-	.power_on = mtk_hdmi_phy_power_on,
-	.power_off = mtk_hdmi_phy_power_off,
-	.owner = THIS_MODULE,
-};
-
-void mtk_hdmi_phy_clear_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-			     u32 bits)
-{
-	void __iomem *reg = hdmi_phy->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp &= ~bits;
-	writel(tmp, reg);
-}
-
-void mtk_hdmi_phy_set_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-			   u32 bits)
-{
-	void __iomem *reg = hdmi_phy->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp |= bits;
-	writel(tmp, reg);
-}
-
-void mtk_hdmi_phy_mask(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-		       u32 val, u32 mask)
-{
-	void __iomem *reg = hdmi_phy->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp = (tmp & ~mask) | (val & mask);
-	writel(tmp, reg);
-}
-
-inline struct mtk_hdmi_phy *to_mtk_hdmi_phy(struct clk_hw *hw)
-{
-	return container_of(hw, struct mtk_hdmi_phy, pll_hw);
-}
-
-static int mtk_hdmi_phy_power_on(struct phy *phy)
-{
-	struct mtk_hdmi_phy *hdmi_phy = phy_get_drvdata(phy);
-	int ret;
-
-	ret = clk_prepare_enable(hdmi_phy->pll);
-	if (ret < 0)
-		return ret;
-
-	hdmi_phy->conf->hdmi_phy_enable_tmds(hdmi_phy);
-	return 0;
-}
-
-static int mtk_hdmi_phy_power_off(struct phy *phy)
-{
-	struct mtk_hdmi_phy *hdmi_phy = phy_get_drvdata(phy);
-
-	hdmi_phy->conf->hdmi_phy_disable_tmds(hdmi_phy);
-	clk_disable_unprepare(hdmi_phy->pll);
-
-	return 0;
-}
-
-static const struct phy_ops *
-mtk_hdmi_phy_dev_get_ops(const struct mtk_hdmi_phy *hdmi_phy)
-{
-	if (hdmi_phy && hdmi_phy->conf &&
-	    hdmi_phy->conf->hdmi_phy_enable_tmds &&
-	    hdmi_phy->conf->hdmi_phy_disable_tmds)
-		return &mtk_hdmi_phy_dev_ops;
-
-	dev_err(hdmi_phy->dev, "Failed to get dev ops of phy\n");
-		return NULL;
-}
-
-static void mtk_hdmi_phy_clk_get_data(struct mtk_hdmi_phy *hdmi_phy,
-				      struct clk_init_data *clk_init)
-{
-	clk_init->flags = hdmi_phy->conf->flags;
-	clk_init->ops = hdmi_phy->conf->hdmi_phy_clk_ops;
-}
-
-static int mtk_hdmi_phy_probe(struct platform_device *pdev)
-{
-	struct device *dev = &pdev->dev;
-	struct mtk_hdmi_phy *hdmi_phy;
-	struct resource *mem;
-	struct clk *ref_clk;
-	const char *ref_clk_name;
-	struct clk_init_data clk_init = {
-		.num_parents = 1,
-		.parent_names = (const char * const *)&ref_clk_name,
-	};
-
-	struct phy *phy;
-	struct phy_provider *phy_provider;
-	int ret;
-
-	hdmi_phy = devm_kzalloc(dev, sizeof(*hdmi_phy), GFP_KERNEL);
-	if (!hdmi_phy)
-		return -ENOMEM;
-
-	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	hdmi_phy->regs = devm_ioremap_resource(dev, mem);
-	if (IS_ERR(hdmi_phy->regs)) {
-		ret = PTR_ERR(hdmi_phy->regs);
-		dev_err(dev, "Failed to get memory resource: %d\n", ret);
-		return ret;
-	}
-
-	ref_clk = devm_clk_get(dev, "pll_ref");
-	if (IS_ERR(ref_clk)) {
-		ret = PTR_ERR(ref_clk);
-		dev_err(&pdev->dev, "Failed to get PLL reference clock: %d\n",
-			ret);
-		return ret;
-	}
-	ref_clk_name = __clk_get_name(ref_clk);
-
-	ret = of_property_read_string(dev->of_node, "clock-output-names",
-				      &clk_init.name);
-	if (ret < 0) {
-		dev_err(dev, "Failed to read clock-output-names: %d\n", ret);
-		return ret;
-	}
-
-	hdmi_phy->dev = dev;
-	hdmi_phy->conf =
-		(struct mtk_hdmi_phy_conf *)of_device_get_match_data(dev);
-	mtk_hdmi_phy_clk_get_data(hdmi_phy, &clk_init);
-	hdmi_phy->pll_hw.init = &clk_init;
-	hdmi_phy->pll = devm_clk_register(dev, &hdmi_phy->pll_hw);
-	if (IS_ERR(hdmi_phy->pll)) {
-		ret = PTR_ERR(hdmi_phy->pll);
-		dev_err(dev, "Failed to register PLL: %d\n", ret);
-		return ret;
-	}
-
-	ret = of_property_read_u32(dev->of_node, "mediatek,ibias",
-				   &hdmi_phy->ibias);
-	if (ret < 0) {
-		dev_err(&pdev->dev, "Failed to get ibias: %d\n", ret);
-		return ret;
-	}
-
-	ret = of_property_read_u32(dev->of_node, "mediatek,ibias_up",
-				   &hdmi_phy->ibias_up);
-	if (ret < 0) {
-		dev_err(&pdev->dev, "Failed to get ibias up: %d\n", ret);
-		return ret;
-	}
-
-	dev_info(dev, "Using default TX DRV impedance: 4.2k/36\n");
-	hdmi_phy->drv_imp_clk = 0x30;
-	hdmi_phy->drv_imp_d2 = 0x30;
-	hdmi_phy->drv_imp_d1 = 0x30;
-	hdmi_phy->drv_imp_d0 = 0x30;
-
-	phy = devm_phy_create(dev, NULL, mtk_hdmi_phy_dev_get_ops(hdmi_phy));
-	if (IS_ERR(phy)) {
-		dev_err(dev, "Failed to create HDMI PHY\n");
-		return PTR_ERR(phy);
-	}
-	phy_set_drvdata(phy, hdmi_phy);
-
-	phy_provider = devm_of_phy_provider_register(dev, of_phy_simple_xlate);
-	if (IS_ERR(phy_provider)) {
-		dev_err(dev, "Failed to register HDMI PHY\n");
-		return PTR_ERR(phy_provider);
-	}
-
-	return of_clk_add_provider(dev->of_node, of_clk_src_simple_get,
-				   hdmi_phy->pll);
-}
-
-static const struct of_device_id mtk_hdmi_phy_match[] = {
-	{ .compatible = "mediatek,mt2701-hdmi-phy",
-	  .data = &mtk_hdmi_phy_2701_conf,
-	},
-	{ .compatible = "mediatek,mt8173-hdmi-phy",
-	  .data = &mtk_hdmi_phy_8173_conf,
-	},
-	{},
-};
-
-struct platform_driver mtk_hdmi_phy_driver = {
-	.probe = mtk_hdmi_phy_probe,
-	.driver = {
-		.name = "mediatek-hdmi-phy",
-		.of_match_table = mtk_hdmi_phy_match,
-	},
-};
-
-MODULE_DESCRIPTION("MediaTek HDMI PHY Driver");
-MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h
deleted file mode 100644
index 2d8b3182470d..000000000000
--- a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h
+++ /dev/null
@@ -1,57 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Copyright (c) 2018 MediaTek Inc.
- * Author: Chunhui Dai <chunhui.dai@mediatek.com>
- */
-
-#ifndef _MTK_HDMI_PHY_H
-#define _MTK_HDMI_PHY_H
-#include <linux/clk.h>
-#include <linux/clk-provider.h>
-#include <linux/delay.h>
-#include <linux/io.h>
-#include <linux/mfd/syscon.h>
-#include <linux/module.h>
-#include <linux/of_device.h>
-#include <linux/phy/phy.h>
-#include <linux/platform_device.h>
-#include <linux/types.h>
-
-struct mtk_hdmi_phy;
-
-struct mtk_hdmi_phy_conf {
-	bool tz_disabled;
-	unsigned long flags;
-	const struct clk_ops *hdmi_phy_clk_ops;
-	void (*hdmi_phy_enable_tmds)(struct mtk_hdmi_phy *hdmi_phy);
-	void (*hdmi_phy_disable_tmds)(struct mtk_hdmi_phy *hdmi_phy);
-};
-
-struct mtk_hdmi_phy {
-	void __iomem *regs;
-	struct device *dev;
-	struct mtk_hdmi_phy_conf *conf;
-	struct clk *pll;
-	struct clk_hw pll_hw;
-	unsigned long pll_rate;
-	unsigned char drv_imp_clk;
-	unsigned char drv_imp_d2;
-	unsigned char drv_imp_d1;
-	unsigned char drv_imp_d0;
-	unsigned int ibias;
-	unsigned int ibias_up;
-};
-
-void mtk_hdmi_phy_clear_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-			     u32 bits);
-void mtk_hdmi_phy_set_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-			   u32 bits);
-void mtk_hdmi_phy_mask(struct mtk_hdmi_phy *hdmi_phy, u32 offset,
-		       u32 val, u32 mask);
-struct mtk_hdmi_phy *to_mtk_hdmi_phy(struct clk_hw *hw);
-
-extern struct platform_driver mtk_hdmi_phy_driver;
-extern struct mtk_hdmi_phy_conf mtk_hdmi_phy_8173_conf;
-extern struct mtk_hdmi_phy_conf mtk_hdmi_phy_2701_conf;
-
-#endif /* _MTK_HDMI_PHY_H */
diff --git a/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c
deleted file mode 100644
index d3cc4022e988..000000000000
--- a/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c
+++ /dev/null
@@ -1,249 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2018 MediaTek Inc.
- * Author: Chunhui Dai <chunhui.dai@mediatek.com>
- */
-
-#include "mtk_hdmi_phy.h"
-
-#define HDMI_CON0	0x00
-#define RG_HDMITX_DRV_IBIAS		0
-#define RG_HDMITX_DRV_IBIAS_MASK	(0x3f << 0)
-#define RG_HDMITX_EN_SER		12
-#define RG_HDMITX_EN_SER_MASK		(0x0f << 12)
-#define RG_HDMITX_EN_SLDO		16
-#define RG_HDMITX_EN_SLDO_MASK		(0x0f << 16)
-#define RG_HDMITX_EN_PRED		20
-#define RG_HDMITX_EN_PRED_MASK		(0x0f << 20)
-#define RG_HDMITX_EN_IMP		24
-#define RG_HDMITX_EN_IMP_MASK		(0x0f << 24)
-#define RG_HDMITX_EN_DRV		28
-#define RG_HDMITX_EN_DRV_MASK		(0x0f << 28)
-
-#define HDMI_CON1	0x04
-#define RG_HDMITX_PRED_IBIAS		18
-#define RG_HDMITX_PRED_IBIAS_MASK	(0x0f << 18)
-#define RG_HDMITX_PRED_IMP		(0x01 << 22)
-#define RG_HDMITX_DRV_IMP		26
-#define RG_HDMITX_DRV_IMP_MASK		(0x3f << 26)
-
-#define HDMI_CON2	0x08
-#define RG_HDMITX_EN_TX_CKLDO		(0x01 << 0)
-#define RG_HDMITX_EN_TX_POSDIV		(0x01 << 1)
-#define RG_HDMITX_TX_POSDIV		3
-#define RG_HDMITX_TX_POSDIV_MASK	(0x03 << 3)
-#define RG_HDMITX_EN_MBIAS		(0x01 << 6)
-#define RG_HDMITX_MBIAS_LPF_EN		(0x01 << 7)
-
-#define HDMI_CON4	0x10
-#define RG_HDMITX_RESERVE_MASK		(0xffffffff << 0)
-
-#define HDMI_CON6	0x18
-#define RG_HTPLL_BR			0
-#define RG_HTPLL_BR_MASK		(0x03 << 0)
-#define RG_HTPLL_BC			2
-#define RG_HTPLL_BC_MASK		(0x03 << 2)
-#define RG_HTPLL_BP			4
-#define RG_HTPLL_BP_MASK		(0x0f << 4)
-#define RG_HTPLL_IR			8
-#define RG_HTPLL_IR_MASK		(0x0f << 8)
-#define RG_HTPLL_IC			12
-#define RG_HTPLL_IC_MASK		(0x0f << 12)
-#define RG_HTPLL_POSDIV			16
-#define RG_HTPLL_POSDIV_MASK		(0x03 << 16)
-#define RG_HTPLL_PREDIV			18
-#define RG_HTPLL_PREDIV_MASK		(0x03 << 18)
-#define RG_HTPLL_FBKSEL			20
-#define RG_HTPLL_FBKSEL_MASK		(0x03 << 20)
-#define RG_HTPLL_RLH_EN			(0x01 << 22)
-#define RG_HTPLL_FBKDIV			24
-#define RG_HTPLL_FBKDIV_MASK		(0x7f << 24)
-#define RG_HTPLL_EN			(0x01 << 31)
-
-#define HDMI_CON7	0x1c
-#define RG_HTPLL_AUTOK_EN		(0x01 << 23)
-#define RG_HTPLL_DIVEN			28
-#define RG_HTPLL_DIVEN_MASK		(0x07 << 28)
-
-static int mtk_hdmi_pll_prepare(struct clk_hw *hw)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON7, RG_HTPLL_AUTOK_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_RLH_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_MBIAS);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_CKLDO);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK);
-	usleep_range(80, 100);
-	return 0;
-}
-
-static void mtk_hdmi_pll_unprepare(struct clk_hw *hw)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_CKLDO);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_EN);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_MBIAS);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_RLH_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON7, RG_HTPLL_AUTOK_EN);
-	usleep_range(80, 100);
-}
-
-static long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate,
-				    unsigned long *parent_rate)
-{
-	return rate;
-}
-
-static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate,
-				 unsigned long parent_rate)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-	u32 pos_div;
-
-	if (rate <= 64000000)
-		pos_div = 3;
-	else if (rate <= 128000000)
-		pos_div = 2;
-	else
-		pos_div = 1;
-
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_PREDIV_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x1 << RG_HTPLL_IC),
-			  RG_HTPLL_IC_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x1 << RG_HTPLL_IR),
-			  RG_HTPLL_IR_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON2, (pos_div << RG_HDMITX_TX_POSDIV),
-			  RG_HDMITX_TX_POSDIV_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (1 << RG_HTPLL_FBKSEL),
-			  RG_HTPLL_FBKSEL_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (19 << RG_HTPLL_FBKDIV),
-			  RG_HTPLL_FBKDIV_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON7, (0x2 << RG_HTPLL_DIVEN),
-			  RG_HTPLL_DIVEN_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0xc << RG_HTPLL_BP),
-			  RG_HTPLL_BP_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x2 << RG_HTPLL_BC),
-			  RG_HTPLL_BC_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x1 << RG_HTPLL_BR),
-			  RG_HTPLL_BR_MASK);
-
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PRED_IMP);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON1, (0x3 << RG_HDMITX_PRED_IBIAS),
-			  RG_HDMITX_PRED_IBIAS_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_IMP_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON1, (0x28 << RG_HDMITX_DRV_IMP),
-			  RG_HDMITX_DRV_IMP_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON4, 0x28, RG_HDMITX_RESERVE_MASK);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON0, (0xa << RG_HDMITX_DRV_IBIAS),
-			  RG_HDMITX_DRV_IBIAS_MASK);
-	return 0;
-}
-
-static unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw,
-					      unsigned long parent_rate)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-	unsigned long out_rate, val;
-
-	val = (readl(hdmi_phy->regs + HDMI_CON6)
-	       & RG_HTPLL_PREDIV_MASK) >> RG_HTPLL_PREDIV;
-	switch (val) {
-	case 0x00:
-		out_rate = parent_rate;
-		break;
-	case 0x01:
-		out_rate = parent_rate / 2;
-		break;
-	default:
-		out_rate = parent_rate / 4;
-		break;
-	}
-
-	val = (readl(hdmi_phy->regs + HDMI_CON6)
-	       & RG_HTPLL_FBKDIV_MASK) >> RG_HTPLL_FBKDIV;
-	out_rate *= (val + 1) * 2;
-	val = (readl(hdmi_phy->regs + HDMI_CON2)
-	       & RG_HDMITX_TX_POSDIV_MASK);
-	out_rate >>= (val >> RG_HDMITX_TX_POSDIV);
-
-	if (readl(hdmi_phy->regs + HDMI_CON2) & RG_HDMITX_EN_TX_POSDIV)
-		out_rate /= 5;
-
-	return out_rate;
-}
-
-static const struct clk_ops mtk_hdmi_phy_pll_ops = {
-	.prepare = mtk_hdmi_pll_prepare,
-	.unprepare = mtk_hdmi_pll_unprepare,
-	.set_rate = mtk_hdmi_pll_set_rate,
-	.round_rate = mtk_hdmi_pll_round_rate,
-	.recalc_rate = mtk_hdmi_pll_recalc_rate,
-};
-
-static void mtk_hdmi_phy_enable_tmds(struct mtk_hdmi_phy *hdmi_phy)
-{
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON7, RG_HTPLL_AUTOK_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_RLH_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_MBIAS);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_CKLDO);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK);
-	usleep_range(80, 100);
-}
-
-static void mtk_hdmi_phy_disable_tmds(struct mtk_hdmi_phy *hdmi_phy)
-{
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_CKLDO);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_EN);
-	usleep_range(80, 100);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_MBIAS);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_RLH_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON7, RG_HTPLL_AUTOK_EN);
-	usleep_range(80, 100);
-}
-
-struct mtk_hdmi_phy_conf mtk_hdmi_phy_2701_conf = {
-	.tz_disabled = true,
-	.flags = CLK_SET_RATE_GATE,
-	.hdmi_phy_clk_ops = &mtk_hdmi_phy_pll_ops,
-	.hdmi_phy_enable_tmds = mtk_hdmi_phy_enable_tmds,
-	.hdmi_phy_disable_tmds = mtk_hdmi_phy_disable_tmds,
-};
-
-MODULE_AUTHOR("Chunhui Dai <chunhui.dai@mediatek.com>");
-MODULE_DESCRIPTION("MediaTek HDMI PHY Driver");
-MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c
deleted file mode 100644
index 827b93786fac..000000000000
--- a/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c
+++ /dev/null
@@ -1,282 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Copyright (c) 2014 MediaTek Inc.
- * Author: Jie Qiu <jie.qiu@mediatek.com>
- */
-
-#include "mtk_hdmi_phy.h"
-
-#define HDMI_CON0		0x00
-#define RG_HDMITX_PLL_EN		BIT(31)
-#define RG_HDMITX_PLL_FBKDIV		(0x7f << 24)
-#define PLL_FBKDIV_SHIFT		24
-#define RG_HDMITX_PLL_FBKSEL		(0x3 << 22)
-#define PLL_FBKSEL_SHIFT		22
-#define RG_HDMITX_PLL_PREDIV		(0x3 << 20)
-#define PREDIV_SHIFT			20
-#define RG_HDMITX_PLL_POSDIV		(0x3 << 18)
-#define POSDIV_SHIFT			18
-#define RG_HDMITX_PLL_RST_DLY		(0x3 << 16)
-#define RG_HDMITX_PLL_IR		(0xf << 12)
-#define PLL_IR_SHIFT			12
-#define RG_HDMITX_PLL_IC		(0xf << 8)
-#define PLL_IC_SHIFT			8
-#define RG_HDMITX_PLL_BP		(0xf << 4)
-#define PLL_BP_SHIFT			4
-#define RG_HDMITX_PLL_BR		(0x3 << 2)
-#define PLL_BR_SHIFT			2
-#define RG_HDMITX_PLL_BC		(0x3 << 0)
-#define PLL_BC_SHIFT			0
-#define HDMI_CON1		0x04
-#define RG_HDMITX_PLL_DIVEN		(0x7 << 29)
-#define PLL_DIVEN_SHIFT			29
-#define RG_HDMITX_PLL_AUTOK_EN		BIT(28)
-#define RG_HDMITX_PLL_AUTOK_KF		(0x3 << 26)
-#define RG_HDMITX_PLL_AUTOK_KS		(0x3 << 24)
-#define RG_HDMITX_PLL_AUTOK_LOAD	BIT(23)
-#define RG_HDMITX_PLL_BAND		(0x3f << 16)
-#define RG_HDMITX_PLL_REF_SEL		BIT(15)
-#define RG_HDMITX_PLL_BIAS_EN		BIT(14)
-#define RG_HDMITX_PLL_BIAS_LPF_EN	BIT(13)
-#define RG_HDMITX_PLL_TXDIV_EN		BIT(12)
-#define RG_HDMITX_PLL_TXDIV		(0x3 << 10)
-#define PLL_TXDIV_SHIFT			10
-#define RG_HDMITX_PLL_LVROD_EN		BIT(9)
-#define RG_HDMITX_PLL_MONVC_EN		BIT(8)
-#define RG_HDMITX_PLL_MONCK_EN		BIT(7)
-#define RG_HDMITX_PLL_MONREF_EN		BIT(6)
-#define RG_HDMITX_PLL_TST_EN		BIT(5)
-#define RG_HDMITX_PLL_TST_CK_EN		BIT(4)
-#define RG_HDMITX_PLL_TST_SEL		(0xf << 0)
-#define HDMI_CON2		0x08
-#define RGS_HDMITX_PLL_AUTOK_BAND	(0x7f << 8)
-#define RGS_HDMITX_PLL_AUTOK_FAIL	BIT(1)
-#define RG_HDMITX_EN_TX_CKLDO		BIT(0)
-#define HDMI_CON3		0x0c
-#define RG_HDMITX_SER_EN		(0xf << 28)
-#define RG_HDMITX_PRD_EN		(0xf << 24)
-#define RG_HDMITX_PRD_IMP_EN		(0xf << 20)
-#define RG_HDMITX_DRV_EN		(0xf << 16)
-#define RG_HDMITX_DRV_IMP_EN		(0xf << 12)
-#define DRV_IMP_EN_SHIFT		12
-#define RG_HDMITX_MHLCK_FORCE		BIT(10)
-#define RG_HDMITX_MHLCK_PPIX_EN		BIT(9)
-#define RG_HDMITX_MHLCK_EN		BIT(8)
-#define RG_HDMITX_SER_DIN_SEL		(0xf << 4)
-#define RG_HDMITX_SER_5T1_BIST_EN	BIT(3)
-#define RG_HDMITX_SER_BIST_TOG		BIT(2)
-#define RG_HDMITX_SER_DIN_TOG		BIT(1)
-#define RG_HDMITX_SER_CLKDIG_INV	BIT(0)
-#define HDMI_CON4		0x10
-#define RG_HDMITX_PRD_IBIAS_CLK		(0xf << 24)
-#define RG_HDMITX_PRD_IBIAS_D2		(0xf << 16)
-#define RG_HDMITX_PRD_IBIAS_D1		(0xf << 8)
-#define RG_HDMITX_PRD_IBIAS_D0		(0xf << 0)
-#define PRD_IBIAS_CLK_SHIFT		24
-#define PRD_IBIAS_D2_SHIFT		16
-#define PRD_IBIAS_D1_SHIFT		8
-#define PRD_IBIAS_D0_SHIFT		0
-#define HDMI_CON5		0x14
-#define RG_HDMITX_DRV_IBIAS_CLK		(0x3f << 24)
-#define RG_HDMITX_DRV_IBIAS_D2		(0x3f << 16)
-#define RG_HDMITX_DRV_IBIAS_D1		(0x3f << 8)
-#define RG_HDMITX_DRV_IBIAS_D0		(0x3f << 0)
-#define DRV_IBIAS_CLK_SHIFT		24
-#define DRV_IBIAS_D2_SHIFT		16
-#define DRV_IBIAS_D1_SHIFT		8
-#define DRV_IBIAS_D0_SHIFT		0
-#define HDMI_CON6		0x18
-#define RG_HDMITX_DRV_IMP_CLK		(0x3f << 24)
-#define RG_HDMITX_DRV_IMP_D2		(0x3f << 16)
-#define RG_HDMITX_DRV_IMP_D1		(0x3f << 8)
-#define RG_HDMITX_DRV_IMP_D0		(0x3f << 0)
-#define DRV_IMP_CLK_SHIFT		24
-#define DRV_IMP_D2_SHIFT		16
-#define DRV_IMP_D1_SHIFT		8
-#define DRV_IMP_D0_SHIFT		0
-#define HDMI_CON7		0x1c
-#define RG_HDMITX_MHLCK_DRV_IBIAS	(0x1f << 27)
-#define RG_HDMITX_SER_DIN		(0x3ff << 16)
-#define RG_HDMITX_CHLDC_TST		(0xf << 12)
-#define RG_HDMITX_CHLCK_TST		(0xf << 8)
-#define RG_HDMITX_RESERVE		(0xff << 0)
-#define HDMI_CON8		0x20
-#define RGS_HDMITX_2T1_LEV		(0xf << 16)
-#define RGS_HDMITX_2T1_EDG		(0xf << 12)
-#define RGS_HDMITX_5T1_LEV		(0xf << 8)
-#define RGS_HDMITX_5T1_EDG		(0xf << 4)
-#define RGS_HDMITX_PLUG_TST		BIT(0)
-
-static int mtk_hdmi_pll_prepare(struct clk_hw *hw)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_AUTOK_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_PLL_POSDIV);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON3, RG_HDMITX_MHLCK_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_BIAS_EN);
-	usleep_range(100, 150);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_PLL_EN);
-	usleep_range(100, 150);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_BIAS_LPF_EN);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_TXDIV_EN);
-
-	return 0;
-}
-
-static void mtk_hdmi_pll_unprepare(struct clk_hw *hw)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_TXDIV_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_BIAS_LPF_EN);
-	usleep_range(100, 150);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_PLL_EN);
-	usleep_range(100, 150);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_BIAS_EN);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_PLL_POSDIV);
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON1, RG_HDMITX_PLL_AUTOK_EN);
-	usleep_range(100, 150);
-}
-
-static long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate,
-				    unsigned long *parent_rate)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	hdmi_phy->pll_rate = rate;
-	if (rate <= 74250000)
-		*parent_rate = rate;
-	else
-		*parent_rate = rate / 2;
-
-	return rate;
-}
-
-static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate,
-				 unsigned long parent_rate)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-	unsigned int pre_div;
-	unsigned int div;
-	unsigned int pre_ibias;
-	unsigned int hdmi_ibias;
-	unsigned int imp_en;
-
-	dev_dbg(hdmi_phy->dev, "%s: %lu Hz, parent: %lu Hz\n", __func__,
-		rate, parent_rate);
-
-	if (rate <= 27000000) {
-		pre_div = 0;
-		div = 3;
-	} else if (rate <= 74250000) {
-		pre_div = 1;
-		div = 2;
-	} else {
-		pre_div = 1;
-		div = 1;
-	}
-
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON0,
-			  (pre_div << PREDIV_SHIFT), RG_HDMITX_PLL_PREDIV);
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_PLL_POSDIV);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON0,
-			  (0x1 << PLL_IC_SHIFT) | (0x1 << PLL_IR_SHIFT),
-			  RG_HDMITX_PLL_IC | RG_HDMITX_PLL_IR);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON1,
-			  (div << PLL_TXDIV_SHIFT), RG_HDMITX_PLL_TXDIV);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON0,
-			  (0x1 << PLL_FBKSEL_SHIFT) | (19 << PLL_FBKDIV_SHIFT),
-			  RG_HDMITX_PLL_FBKSEL | RG_HDMITX_PLL_FBKDIV);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON1,
-			  (0x2 << PLL_DIVEN_SHIFT), RG_HDMITX_PLL_DIVEN);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON0,
-			  (0xc << PLL_BP_SHIFT) | (0x2 << PLL_BC_SHIFT) |
-			  (0x1 << PLL_BR_SHIFT),
-			  RG_HDMITX_PLL_BP | RG_HDMITX_PLL_BC |
-			  RG_HDMITX_PLL_BR);
-	if (rate < 165000000) {
-		mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON3,
-					RG_HDMITX_PRD_IMP_EN);
-		pre_ibias = 0x3;
-		imp_en = 0x0;
-		hdmi_ibias = hdmi_phy->ibias;
-	} else {
-		mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON3,
-				      RG_HDMITX_PRD_IMP_EN);
-		pre_ibias = 0x6;
-		imp_en = 0xf;
-		hdmi_ibias = hdmi_phy->ibias_up;
-	}
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON4,
-			  (pre_ibias << PRD_IBIAS_CLK_SHIFT) |
-			  (pre_ibias << PRD_IBIAS_D2_SHIFT) |
-			  (pre_ibias << PRD_IBIAS_D1_SHIFT) |
-			  (pre_ibias << PRD_IBIAS_D0_SHIFT),
-			  RG_HDMITX_PRD_IBIAS_CLK |
-			  RG_HDMITX_PRD_IBIAS_D2 |
-			  RG_HDMITX_PRD_IBIAS_D1 |
-			  RG_HDMITX_PRD_IBIAS_D0);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON3,
-			  (imp_en << DRV_IMP_EN_SHIFT),
-			  RG_HDMITX_DRV_IMP_EN);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6,
-			  (hdmi_phy->drv_imp_clk << DRV_IMP_CLK_SHIFT) |
-			  (hdmi_phy->drv_imp_d2 << DRV_IMP_D2_SHIFT) |
-			  (hdmi_phy->drv_imp_d1 << DRV_IMP_D1_SHIFT) |
-			  (hdmi_phy->drv_imp_d0 << DRV_IMP_D0_SHIFT),
-			  RG_HDMITX_DRV_IMP_CLK | RG_HDMITX_DRV_IMP_D2 |
-			  RG_HDMITX_DRV_IMP_D1 | RG_HDMITX_DRV_IMP_D0);
-	mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON5,
-			  (hdmi_ibias << DRV_IBIAS_CLK_SHIFT) |
-			  (hdmi_ibias << DRV_IBIAS_D2_SHIFT) |
-			  (hdmi_ibias << DRV_IBIAS_D1_SHIFT) |
-			  (hdmi_ibias << DRV_IBIAS_D0_SHIFT),
-			  RG_HDMITX_DRV_IBIAS_CLK |
-			  RG_HDMITX_DRV_IBIAS_D2 |
-			  RG_HDMITX_DRV_IBIAS_D1 |
-			  RG_HDMITX_DRV_IBIAS_D0);
-	return 0;
-}
-
-static unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw,
-					      unsigned long parent_rate)
-{
-	struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw);
-
-	return hdmi_phy->pll_rate;
-}
-
-static const struct clk_ops mtk_hdmi_phy_pll_ops = {
-	.prepare = mtk_hdmi_pll_prepare,
-	.unprepare = mtk_hdmi_pll_unprepare,
-	.set_rate = mtk_hdmi_pll_set_rate,
-	.round_rate = mtk_hdmi_pll_round_rate,
-	.recalc_rate = mtk_hdmi_pll_recalc_rate,
-};
-
-static void mtk_hdmi_phy_enable_tmds(struct mtk_hdmi_phy *hdmi_phy)
-{
-	mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON3,
-			      RG_HDMITX_SER_EN | RG_HDMITX_PRD_EN |
-			      RG_HDMITX_DRV_EN);
-	usleep_range(100, 150);
-}
-
-static void mtk_hdmi_phy_disable_tmds(struct mtk_hdmi_phy *hdmi_phy)
-{
-	mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON3,
-				RG_HDMITX_DRV_EN | RG_HDMITX_PRD_EN |
-				RG_HDMITX_SER_EN);
-}
-
-struct mtk_hdmi_phy_conf mtk_hdmi_phy_8173_conf = {
-	.flags = CLK_SET_RATE_PARENT | CLK_SET_RATE_GATE,
-	.hdmi_phy_clk_ops = &mtk_hdmi_phy_pll_ops,
-	.hdmi_phy_enable_tmds = mtk_hdmi_phy_enable_tmds,
-	.hdmi_phy_disable_tmds = mtk_hdmi_phy_disable_tmds,
-};
-
-MODULE_AUTHOR("Jie Qiu <jie.qiu@mediatek.com>");
-MODULE_DESCRIPTION("MediaTek MT8173 HDMI PHY Driver");
-MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/mgag200/Kconfig b/drivers/gpu/drm/mgag200/Kconfig
index 93be766715c9..eec59658a938 100644
--- a/drivers/gpu/drm/mgag200/Kconfig
+++ b/drivers/gpu/drm/mgag200/Kconfig
@@ -1,13 +1,11 @@
 # SPDX-License-Identifier: GPL-2.0-only
 config DRM_MGAG200
-	tristate "Kernel modesetting driver for MGA G200 server engines"
+	tristate "Matrox G200"
 	depends on DRM && PCI && MMU
 	select DRM_GEM_SHMEM_HELPER
 	select DRM_KMS_HELPER
 	help
-	 This is a KMS driver for the MGA G200 server chips, it
-	 does not support the original MGA G200 or any of the desktop
-	 chips. It requires 0.3.0 of the modesetting userspace driver,
-	 and a version of mga driver that will fail on KMS enabled
-	 devices.
-
+	 This is a KMS driver for Matrox G200 chips. It supports the original
+	 MGA G200 desktop chips and the server variants. It requires 0.3.0
+	 of the modesetting userspace driver, and a version of mga driver
+	 that will fail on KMS enabled devices.
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.c b/drivers/gpu/drm/mgag200/mgag200_drv.c
index e19660f4a637..771b26aeee19 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.c
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.c
@@ -9,6 +9,7 @@
 #include <linux/console.h>
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/vmalloc.h>
 
 #include <drm/drm_drv.h>
 #include <drm/drm_file.h>
@@ -36,6 +37,7 @@ static struct drm_driver mgag200_driver = {
 	.major = DRIVER_MAJOR,
 	.minor = DRIVER_MINOR,
 	.patchlevel = DRIVER_PATCHLEVEL,
+	.gem_create_object = drm_gem_shmem_create_object_cached,
 	DRM_GEM_SHMEM_DRIVER_OPS,
 };
 
@@ -43,18 +45,66 @@ static struct drm_driver mgag200_driver = {
  * DRM device
  */
 
-static int mgag200_device_init(struct mga_device *mdev, unsigned long flags)
+static bool mgag200_has_sgram(struct mga_device *mdev)
 {
 	struct drm_device *dev = &mdev->base;
-	int ret, option;
+	u32 option;
+	int ret;
 
-	mdev->flags = mgag200_flags_from_driver_data(flags);
-	mdev->type = mgag200_type_from_driver_data(flags);
+	ret = pci_read_config_dword(dev->pdev, PCI_MGA_OPTION, &option);
+	if (drm_WARN(dev, ret, "failed to read PCI config dword: %d\n", ret))
+		return false;
 
-	pci_read_config_dword(dev->pdev, PCI_MGA_OPTION, &option);
-	mdev->has_sdram = !(option & (1 << 14));
+	return !!(option & PCI_MGA_OPTION_HARDPWMSK);
+}
 
-	/* BAR 0 is the framebuffer, BAR 1 contains registers */
+static int mgag200_regs_init(struct mga_device *mdev)
+{
+	struct drm_device *dev = &mdev->base;
+	u32 option, option2;
+	u8 crtcext3;
+
+	switch (mdev->type) {
+	case G200_PCI:
+	case G200_AGP:
+		if (mgag200_has_sgram(mdev))
+			option = 0x4049cd21;
+		else
+			option = 0x40499121;
+		option2 = 0x00008000;
+		break;
+	case G200_SE_A:
+	case G200_SE_B:
+		option = 0x40049120;
+		if (mgag200_has_sgram(mdev))
+			option |= PCI_MGA_OPTION_HARDPWMSK;
+		option2 = 0x00008000;
+		break;
+	case G200_WB:
+	case G200_EW3:
+		option = 0x41049120;
+		option2 = 0x0000b000;
+		break;
+	case G200_EV:
+		option = 0x00000120;
+		option2 = 0x0000b000;
+		break;
+	case G200_EH:
+	case G200_EH3:
+		option = 0x00000120;
+		option2 = 0x0000b000;
+		break;
+	default:
+		option = 0;
+		option2 = 0;
+	}
+
+	if (option)
+		pci_write_config_dword(dev->pdev, PCI_MGA_OPTION, option);
+	if (option2)
+		pci_write_config_dword(dev->pdev, PCI_MGA_OPTION2, option2);
+
+	/* BAR 1 contains registers */
 	mdev->rmmio_base = pci_resource_start(dev->pdev, 1);
 	mdev->rmmio_size = pci_resource_len(dev->pdev, 1);
 
@@ -68,12 +118,163 @@ static int mgag200_device_init(struct mga_device *mdev, unsigned long flags)
 	if (mdev->rmmio == NULL)
 		return -ENOMEM;
 
-	/* stash G200 SE model number for later use */
-	if (IS_G200_SE(mdev)) {
-		mdev->unique_rev_id = RREG32(0x1e24);
-		drm_dbg(dev, "G200 SE unique revision id is 0x%x\n",
-			mdev->unique_rev_id);
+	RREG_ECRT(0x03, crtcext3);
+	crtcext3 |= MGAREG_CRTCEXT3_MGAMODE;
+	WREG_ECRT(0x03, crtcext3);
+
+	return 0;
+}
+
+static void mgag200_g200_interpret_bios(struct mga_device *mdev,
+					const unsigned char *bios,
+					size_t size)
+{
+	static const char matrox[] = {'M', 'A', 'T', 'R', 'O', 'X'};
+	static const unsigned int expected_length[6] = {
+		0, 64, 64, 64, 128, 128
+	};
+	struct drm_device *dev = &mdev->base;
+	const unsigned char *pins;
+	unsigned int pins_len, version;
+	int offset;
+	int tmp;
+
+	/* Test for MATROX string. */
+	if (size < 45 + sizeof(matrox))
+		return;
+	if (memcmp(&bios[45], matrox, sizeof(matrox)) != 0)
+		return;
+
+	/* Get the PInS offset. */
+	if (size < MGA_BIOS_OFFSET + 2)
+		return;
+	offset = (bios[MGA_BIOS_OFFSET + 1] << 8) | bios[MGA_BIOS_OFFSET];
+
+	/* Get PInS data structure. */
+
+	if (size < offset + 6)
+		return;
+	pins = bios + offset;
+	if (pins[0] == 0x2e && pins[1] == 0x41) {
+		version = pins[5];
+		pins_len = pins[2];
+	} else {
+		version = 1;
+		pins_len = pins[0] + (pins[1] << 8);
+	}
+
+	if (version < 1 || version > 5) {
+		drm_warn(dev, "Unknown BIOS PInS version: %d\n", version);
+		return;
+	}
+	if (pins_len != expected_length[version]) {
+		drm_warn(dev, "Unexpected BIOS PInS size: %d expected: %d\n",
+			 pins_len, expected_length[version]);
+		return;
 	}
+	if (size < offset + pins_len)
+		return;
+
+	drm_dbg_kms(dev, "MATROX BIOS PInS version %d size: %d found\n",
+		    version, pins_len);
+
+	/* Extract the clock values */
+
+	switch (version) {
+	case 1:
+		tmp = pins[24] + (pins[25] << 8);
+		if (tmp)
+			mdev->model.g200.pclk_max = tmp * 10;
+		break;
+	case 2:
+		if (pins[41] != 0xff)
+			mdev->model.g200.pclk_max = (pins[41] + 100) * 1000;
+		break;
+	case 3:
+		if (pins[36] != 0xff)
+			mdev->model.g200.pclk_max = (pins[36] + 100) * 1000;
+		if (pins[52] & 0x20)
+			mdev->model.g200.ref_clk = 14318;
+		break;
+	case 4:
+		if (pins[39] != 0xff)
+			mdev->model.g200.pclk_max = pins[39] * 4 * 1000;
+		if (pins[92] & 0x01)
+			mdev->model.g200.ref_clk = 14318;
+		break;
+	case 5:
+		tmp = pins[4] ? 8000 : 6000;
+		if (pins[123] != 0xff)
+			mdev->model.g200.pclk_min = pins[123] * tmp;
+		if (pins[38] != 0xff)
+			mdev->model.g200.pclk_max = pins[38] * tmp;
+		if (pins[110] & 0x01)
+			mdev->model.g200.ref_clk = 14318;
+		break;
+	default:
+		break;
+	}
+}
+
+static void mgag200_g200_init_refclk(struct mga_device *mdev)
+{
+	struct drm_device *dev = &mdev->base;
+	unsigned char __iomem *rom;
+	unsigned char *bios;
+	size_t size;
+
+	mdev->model.g200.pclk_min = 50000;
+	mdev->model.g200.pclk_max = 230000;
+	mdev->model.g200.ref_clk = 27050;
+
+	rom = pci_map_rom(dev->pdev, &size);
+	if (!rom)
+		return;
+
+	bios = vmalloc(size);
+	if (!bios)
+		goto out;
+	memcpy_fromio(bios, rom, size);
+
+	if (size != 0 && bios[0] == 0x55 && bios[1] == 0xaa)
+		mgag200_g200_interpret_bios(mdev, bios, size);
+
+	drm_dbg_kms(dev, "pclk_min: %ld pclk_max: %ld ref_clk: %ld\n",
+		    mdev->model.g200.pclk_min, mdev->model.g200.pclk_max,
+		    mdev->model.g200.ref_clk);
+
+	vfree(bios);
+out:
+	pci_unmap_rom(dev->pdev, rom);
+}
+
+static void mgag200_g200se_init_unique_id(struct mga_device *mdev)
+{
+	struct drm_device *dev = &mdev->base;
+
+	/* stash G200 SE model number for later use */
+	mdev->model.g200se.unique_rev_id = RREG32(0x1e24);
+
+	drm_dbg(dev, "G200 SE unique revision id is 0x%x\n",
+		mdev->model.g200se.unique_rev_id);
+}
+
+static int mgag200_device_init(struct mga_device *mdev, unsigned long flags)
+{
+	struct drm_device *dev = &mdev->base;
+	int ret;
+
+	mdev->flags = mgag200_flags_from_driver_data(flags);
+	mdev->type = mgag200_type_from_driver_data(flags);
+
+	ret = mgag200_regs_init(mdev);
+	if (ret)
+		return ret;
+
+	if (mdev->type == G200_PCI || mdev->type == G200_AGP)
+		mgag200_g200_init_refclk(mdev);
+	else if (IS_G200_SE(mdev))
+		mgag200_g200se_init_unique_id(mdev);
 
 	ret = mgag200_mm_init(mdev);
 	if (ret)
@@ -116,6 +317,8 @@ mgag200_device_create(struct pci_dev *pdev, unsigned long flags)
  */
 
 static const struct pci_device_id mgag200_pciidlist[] = {
+	{ PCI_VENDOR_ID_MATROX, 0x520, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_PCI },
+	{ PCI_VENDOR_ID_MATROX, 0x521, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_AGP },
 	{ PCI_VENDOR_ID_MATROX, 0x522, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		G200_SE_A | MGAG200_FLAG_HW_BUG_NO_STARTADD},
 	{ PCI_VENDOR_ID_MATROX, 0x524, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_B },
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h
index 3817520bfefc..749a075fe9e4 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.h
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.h
@@ -38,6 +38,8 @@
 #define RREG32(reg) ioread32(((void __iomem *)mdev->rmmio) + (reg))
 #define WREG32(reg, v) iowrite32(v, ((void __iomem *)mdev->rmmio) + (reg))
 
+#define MGA_BIOS_OFFSET		0x7ffc
+
 #define ATTR_INDEX 0x1fc0
 #define ATTR_DATA 0x1fc1
 
@@ -129,6 +131,8 @@ struct mga_mc {
 };
 
 enum mga_type {
+	G200_PCI,
+	G200_AGP,
 	G200_SE_A,
 	G200_SE_B,
 	G200_WB,
@@ -161,14 +165,23 @@ struct mga_device {
 	size_t				vram_fb_available;
 
 	enum mga_type			type;
-	int				has_sdram;
 
 	int bpp_shifts[4];
 
 	int fb_mtrr;
 
-	/* SE model number stored in reg 0x1e24 */
-	u32 unique_rev_id;
+	union {
+		struct {
+			long ref_clk;
+			long pclk_min;
+			long pclk_max;
+		} g200;
+		struct {
+			/* SE model number stored in reg 0x1e24 */
+			u32 unique_rev_id;
+		} g200se;
+	} model;
+
 
 	struct mga_connector connector;
 	struct drm_simple_display_pipe display_pipe;
diff --git a/drivers/gpu/drm/mgag200/mgag200_mm.c b/drivers/gpu/drm/mgag200/mgag200_mm.c
index 7b69392bcb89..641f1aa992be 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mm.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mm.c
@@ -90,9 +90,17 @@ static void mgag200_mm_release(struct drm_device *dev, void *ptr)
 int mgag200_mm_init(struct mga_device *mdev)
 {
 	struct drm_device *dev = &mdev->base;
+	u8 misc;
 	resource_size_t start, len;
 	int ret;
 
+	WREG_ECRT(0x04, 0x00);
+
+	misc = RREG8(MGA_MISC_IN);
+	misc |= MGAREG_MISC_RAMMAPEN |
+		MGAREG_MISC_HIGH_PG_SEL;
+	WREG8(MGA_MISC_OUT, misc);
+
 	/* BAR 0 is VRAM */
 	start = pci_resource_start(dev->pdev, 0);
 	len = pci_resource_len(dev->pdev, 0);
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index e0d037a7413c..38672f9e5c4f 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -9,7 +9,6 @@
  */
 
 #include <linux/delay.h>
-#include <linux/pci.h>
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_atomic_state_helper.h>
@@ -109,10 +108,82 @@ static inline void mga_wait_busy(struct mga_device *mdev)
 	} while ((status & 0x01) && time_before(jiffies, timeout));
 }
 
+/*
+ * PLL setup
+ */
+
+static int mgag200_g200_set_plls(struct mga_device *mdev, long clock)
+{
+	struct drm_device *dev = &mdev->base;
+	const int post_div_max = 7;
+	const int in_div_min = 1;
+	const int in_div_max = 6;
+	const int feed_div_min = 7;
+	const int feed_div_max = 127;
+	u8 testm, testn;
+	u8 n = 0, m = 0, p, s;
+	long f_vco;
+	long computed;
+	long delta, tmp_delta;
+	long ref_clk = mdev->model.g200.ref_clk;
+	long p_clk_min = mdev->model.g200.pclk_min;
+	long p_clk_max =  mdev->model.g200.pclk_max;
+
+	if (clock > p_clk_max) {
+		drm_err(dev, "Pixel Clock %ld too high\n", clock);
+		return 1;
+	}
+
+	if (clock < p_clk_min >> 3)
+		clock = p_clk_min >> 3;
+
+	f_vco = clock;
+	for (p = 0;
+	     p <= post_div_max && f_vco < p_clk_min;
+	     p = (p << 1) + 1, f_vco <<= 1)
+		;
+
+	delta = clock;
+
+	for (testm = in_div_min; testm <= in_div_max; testm++) {
+		for (testn = feed_div_min; testn <= feed_div_max; testn++) {
+			computed = ref_clk * (testn + 1) / (testm + 1);
+			if (computed < f_vco)
+				tmp_delta = f_vco - computed;
+			else
+				tmp_delta = computed - f_vco;
+			if (tmp_delta < delta) {
+				delta = tmp_delta;
+				m = testm;
+				n = testn;
+			}
+		}
+	}
+	f_vco = ref_clk * (n + 1) / (m + 1);
+	if (f_vco < 100000)
+		s = 0;
+	else if (f_vco < 140000)
+		s = 1;
+	else if (f_vco < 180000)
+		s = 2;
+	else
+		s = 3;
+
+	drm_dbg_kms(dev, "clock: %ld vco: %ld m: %d n: %d p: %d s: %d\n",
+		    clock, f_vco, m, n, p, s);
+
+	WREG_DAC(MGA1064_PIX_PLLC_M, m);
+	WREG_DAC(MGA1064_PIX_PLLC_N, n);
+	WREG_DAC(MGA1064_PIX_PLLC_P, (p | (s << 3)));
+
+	return 0;
+}
+
 #define P_ARRAY_SIZE 9
 
 static int mga_g200se_set_plls(struct mga_device *mdev, long clock)
 {
+	u32 unique_rev_id = mdev->model.g200se.unique_rev_id;
 	unsigned int vcomax, vcomin, pllreffreq;
 	unsigned int delta, tmpdelta, permitteddelta;
 	unsigned int testp, testm, testn;
@@ -122,7 +193,7 @@ static int mga_g200se_set_plls(struct mga_device *mdev, long clock)
 	unsigned int fvv;
 	unsigned int i;
 
-	if (mdev->unique_rev_id <= 0x03) {
+	if (unique_rev_id <= 0x03) {
 
 		m = n = p = 0;
 		vcomax = 320000;
@@ -220,7 +291,7 @@ static int mga_g200se_set_plls(struct mga_device *mdev, long clock)
 	WREG_DAC(MGA1064_PIX_PLLC_N, n);
 	WREG_DAC(MGA1064_PIX_PLLC_P, p);
 
-	if (mdev->unique_rev_id >= 0x04) {
+	if (unique_rev_id >= 0x04) {
 		WREG_DAC(0x1a, 0x09);
 		msleep(20);
 		WREG_DAC(0x1a, 0x01);
@@ -717,6 +788,9 @@ static int mgag200_crtc_set_plls(struct mga_device *mdev, long clock)
 	u8 misc;
 
 	switch(mdev->type) {
+	case G200_PCI:
+	case G200_AGP:
+		return mgag200_g200_set_plls(mdev, clock);
 	case G200_SE_A:
 	case G200_SE_B:
 		return mga_g200se_set_plls(mdev, clock);
@@ -877,45 +951,6 @@ static void mgag200_set_startadd(struct mga_device *mdev,
 	WREG_ECRT(0x00, crtcext0);
 }
 
-static void mgag200_set_pci_regs(struct mga_device *mdev)
-{
-	uint32_t option = 0, option2 = 0;
-	struct drm_device *dev = &mdev->base;
-
-	switch (mdev->type) {
-	case G200_SE_A:
-	case G200_SE_B:
-		if (mdev->has_sdram)
-			option = 0x40049120;
-		else
-			option = 0x4004d120;
-		option2 = 0x00008000;
-		break;
-	case G200_WB:
-	case G200_EW3:
-		option = 0x41049120;
-		option2 = 0x0000b000;
-		break;
-	case G200_EV:
-		option = 0x00000120;
-		option2 = 0x0000b000;
-		break;
-	case G200_EH:
-	case G200_EH3:
-		option = 0x00000120;
-		option2 = 0x0000b000;
-		break;
-	case G200_ER:
-		break;
-	}
-
-	if (option)
-		pci_write_config_dword(dev->pdev, PCI_MGA_OPTION, option);
-
-	if (option2)
-		pci_write_config_dword(dev->pdev, PCI_MGA_OPTION2, option2);
-}
-
 static void mgag200_set_dac_regs(struct mga_device *mdev)
 {
 	size_t i;
@@ -933,6 +968,12 @@ static void mgag200_set_dac_regs(struct mga_device *mdev)
 	};
 
 	switch (mdev->type) {
+	case G200_PCI:
+	case G200_AGP:
+		dacvalue[MGA1064_SYS_PLL_M] = 0x04;
+		dacvalue[MGA1064_SYS_PLL_N] = 0x2D;
+		dacvalue[MGA1064_SYS_PLL_P] = 0x19;
+		break;
 	case G200_SE_A:
 	case G200_SE_B:
 		dacvalue[MGA1064_VREF_CTL] = 0x03;
@@ -986,9 +1027,8 @@ static void mgag200_set_dac_regs(struct mga_device *mdev)
 
 static void mgag200_init_regs(struct mga_device *mdev)
 {
-	u8 crtc11, crtcext3, crtcext4, misc;
+	u8 crtc11, misc;
 
-	mgag200_set_pci_regs(mdev);
 	mgag200_set_dac_regs(mdev);
 
 	WREG_SEQ(2, 0x0f);
@@ -1002,14 +1042,6 @@ static void mgag200_init_regs(struct mga_device *mdev)
 	WREG_CRT(14, 0);
 	WREG_CRT(15, 0);
 
-	RREG_ECRT(0x03, crtcext3);
-
-	crtcext3 |= BIT(7); /* enable MGA mode */
-	crtcext4 = 0x00;
-
-	WREG_ECRT(0x03, crtcext3);
-	WREG_ECRT(0x04, crtcext4);
-
 	RREG_CRT(0x11, crtc11);
 	crtc11 &= ~(MGAREG_CRTC11_CRTCPROTECT |
 		    MGAREG_CRTC11_VINTEN |
@@ -1023,9 +1055,7 @@ static void mgag200_init_regs(struct mga_device *mdev)
 		WREG_ECRT(0x34, 0x5);
 
 	misc = RREG8(MGA_MISC_IN);
-	misc |= MGAREG_MISC_IOADSEL |
-		MGAREG_MISC_RAMMAPEN |
-		MGAREG_MISC_HIGH_PG_SEL;
+	misc |= MGAREG_MISC_IOADSEL;
 	WREG8(MGA_MISC_OUT, misc);
 }
 
@@ -1234,12 +1264,13 @@ static void mgag200_g200se_set_hiprilvl(struct mga_device *mdev,
 					const struct drm_display_mode *mode,
 					const struct drm_framebuffer *fb)
 {
+	u32 unique_rev_id = mdev->model.g200se.unique_rev_id;
 	unsigned int hiprilvl;
 	u8 crtcext6;
 
-	if  (mdev->unique_rev_id >= 0x04) {
+	if  (unique_rev_id >= 0x04) {
 		hiprilvl = 0;
-	} else if (mdev->unique_rev_id >= 0x02) {
+	} else if (unique_rev_id >= 0x02) {
 		unsigned int bpp;
 		unsigned long mb;
 
@@ -1264,7 +1295,7 @@ static void mgag200_g200se_set_hiprilvl(struct mga_device *mdev,
 		else
 			hiprilvl = 5;
 
-	} else if (mdev->unique_rev_id >= 0x01) {
+	} else if (unique_rev_id >= 0x01) {
 		hiprilvl = 3;
 	} else {
 		hiprilvl = 4;
@@ -1388,7 +1419,9 @@ static enum drm_mode_status mga_vga_mode_valid(struct drm_connector *connector,
 	int bpp = 32;
 
 	if (IS_G200_SE(mdev)) {
-		if (mdev->unique_rev_id == 0x01) {
+		u32 unique_rev_id = mdev->model.g200se.unique_rev_id;
+
+		if (unique_rev_id == 0x01) {
 			if (mode->hdisplay > 1600)
 				return MODE_VIRTUAL_X;
 			if (mode->vdisplay > 1200)
@@ -1396,7 +1429,7 @@ static enum drm_mode_status mga_vga_mode_valid(struct drm_connector *connector,
 			if (mga_vga_calculate_mode_bandwidth(mode, bpp)
 				> (24400 * 1024))
 				return MODE_BANDWIDTH;
-		} else if (mdev->unique_rev_id == 0x02) {
+		} else if (unique_rev_id == 0x02) {
 			if (mode->hdisplay > 1920)
 				return MODE_VIRTUAL_X;
 			if (mode->vdisplay > 1200)
diff --git a/drivers/gpu/drm/mgag200/mgag200_reg.h b/drivers/gpu/drm/mgag200/mgag200_reg.h
index c3b7bcad52ed..977be0565c06 100644
--- a/drivers/gpu/drm/mgag200/mgag200_reg.h
+++ b/drivers/gpu/drm/mgag200/mgag200_reg.h
@@ -256,6 +256,8 @@
 #define MGAREG_CRTCEXT1_VSYNCOFF	BIT(5)
 #define MGAREG_CRTCEXT1_HSYNCOFF	BIT(4)
 
+#define MGAREG_CRTCEXT3_MGAMODE		BIT(7)
+
 /* Cursor X and Y position */
 #define MGA_CURPOSXL 0x3c0c
 #define MGA_CURPOSXH 0x3c0d
@@ -282,6 +284,8 @@
 #define PCI_MGA_OPTION2		0x50
 #define PCI_MGA_OPTION3		0x54
 
+#define PCI_MGA_OPTION_HARDPWMSK	BIT(14)
+
 #define RAMDAC_OFFSET		0x3c00
 
 /* TVP3026 direct registers */
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 6deaa7d01654..e5816b498494 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -6,8 +6,8 @@ config DRM_MSM
 	depends on ARCH_QCOM || SOC_IMX5 || (ARM && COMPILE_TEST)
 	depends on OF && COMMON_CLK
 	depends on MMU
-	depends on INTERCONNECT || !INTERCONNECT
 	depends on QCOM_OCMEM || QCOM_OCMEM=n
+	select IOMMU_IO_PGTABLE
 	select QCOM_MDT_LOADER if ARCH_QCOM
 	select REGULATOR
 	select DRM_KMS_HELPER
@@ -57,6 +57,15 @@ config DRM_MSM_HDMI_HDCP
 	help
 	  Choose this option to enable HDCP state machine
 
+config DRM_MSM_DP
+	bool "Enable DisplayPort support in MSM DRM driver"
+	depends on DRM_MSM
+	default y
+	help
+	  Compile in support for DP driver in MSM DRM driver. DP external
+	  display support is enabled through this config option. It can
+	  be primary or secondary display on device.
+
 config DRM_MSM_DSI
 	bool "Enable DSI support in MSM DRM driver"
 	depends on DRM_MSM
@@ -110,3 +119,11 @@ config DRM_MSM_DSI_10NM_PHY
 	default y
 	help
 	  Choose this option if DSI PHY on SDM845 is used on the platform.
+
+config DRM_MSM_DSI_7NM_PHY
+	bool "Enable DSI 7nm PHY driver in MSM DRM (used by SM8150/SM8250)"
+	depends on DRM_MSM_DSI
+	default y
+	help
+	  Choose this option if DSI PHY on SM8150/SM8250 is used on the
+	  platform.
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 42f8aae28b31..340682cd0f32 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -2,6 +2,7 @@
 ccflags-y := -I $(srctree)/$(src)
 ccflags-y += -I $(srctree)/$(src)/disp/dpu1
 ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(srctree)/$(src)/dsi
+ccflags-$(CONFIG_DRM_MSM_DP) += -I $(srctree)/$(src)/dp
 
 msm-y := \
 	adreno/adreno_device.o \
@@ -95,10 +96,23 @@ msm-y := \
 	msm_gpu_tracepoints.o \
 	msm_gpummu.o
 
-msm-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o
+msm-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o \
+	dp/dp_debug.o
 
 msm-$(CONFIG_DRM_MSM_GPU_STATE)	+= adreno/a6xx_gpu_state.o
 
+msm-$(CONFIG_DRM_MSM_DP)+= dp/dp_aux.o \
+	dp/dp_catalog.o \
+	dp/dp_ctrl.o \
+	dp/dp_display.o \
+	dp/dp_drm.o \
+	dp/dp_hpd.o \
+	dp/dp_link.o \
+	dp/dp_panel.o \
+	dp/dp_parser.o \
+	dp/dp_power.o \
+	dp/dp_audio.o
+
 msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o
 msm-$(CONFIG_COMMON_CLK) += disp/mdp4/mdp4_lvds_pll.o
 msm-$(CONFIG_COMMON_CLK) += hdmi/hdmi_pll_8960.o
@@ -119,6 +133,7 @@ msm-$(CONFIG_DRM_MSM_DSI_20NM_PHY) += dsi/phy/dsi_phy_20nm.o
 msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/phy/dsi_phy_28nm_8960.o
 msm-$(CONFIG_DRM_MSM_DSI_14NM_PHY) += dsi/phy/dsi_phy_14nm.o
 msm-$(CONFIG_DRM_MSM_DSI_10NM_PHY) += dsi/phy/dsi_phy_10nm.o
+msm-$(CONFIG_DRM_MSM_DSI_7NM_PHY) += dsi/phy/dsi_phy_7nm.o
 
 ifeq ($(CONFIG_DRM_MSM_DSI_PLL),y)
 msm-y += dsi/pll/dsi_pll.o
@@ -126,6 +141,7 @@ msm-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/pll/dsi_pll_28nm.o
 msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/pll/dsi_pll_28nm_8960.o
 msm-$(CONFIG_DRM_MSM_DSI_14NM_PHY) += dsi/pll/dsi_pll_14nm.o
 msm-$(CONFIG_DRM_MSM_DSI_10NM_PHY) += dsi/pll/dsi_pll_10nm.o
+msm-$(CONFIG_DRM_MSM_DSI_7NM_PHY) += dsi/pll/dsi_pll_7nm.o
 endif
 
 obj-$(CONFIG_DRM_MSM)	+= msm.o
diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c
index 48fa49f69d6d..7e82c41a85f1 100644
--- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c
@@ -10,6 +10,48 @@ extern bool hang_debug;
 static void a2xx_dump(struct msm_gpu *gpu);
 static bool a2xx_idle(struct msm_gpu *gpu);
 
+static void a2xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
+{
+	struct msm_drm_private *priv = gpu->dev->dev_private;
+	struct msm_ringbuffer *ring = submit->ring;
+	unsigned int i;
+
+	for (i = 0; i < submit->nr_cmds; i++) {
+		switch (submit->cmd[i].type) {
+		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
+			/* ignore IB-targets */
+			break;
+		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
+			/* ignore if there has not been a ctx switch: */
+			if (priv->lastctx == submit->queue->ctx)
+				break;
+			fallthrough;
+		case MSM_SUBMIT_CMD_BUF:
+			OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFD, 2);
+			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
+			OUT_RING(ring, submit->cmd[i].size);
+			OUT_PKT2(ring);
+			break;
+		}
+	}
+
+	OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1);
+	OUT_RING(ring, submit->seqno);
+
+	/* wait for idle before cache flush/interrupt */
+	OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1);
+	OUT_RING(ring, 0x00000000);
+
+	OUT_PKT3(ring, CP_EVENT_WRITE, 3);
+	OUT_RING(ring, CACHE_FLUSH_TS);
+	OUT_RING(ring, rbmemptr(ring, fence));
+	OUT_RING(ring, submit->seqno);
+	OUT_PKT3(ring, CP_INTERRUPT, 1);
+	OUT_RING(ring, 0x80000000);
+
+	adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR);
+}
+
 static bool a2xx_me_init(struct msm_gpu *gpu)
 {
 	struct msm_ringbuffer *ring = gpu->rb[0];
@@ -53,7 +95,7 @@ static bool a2xx_me_init(struct msm_gpu *gpu)
 	OUT_PKT3(ring, CP_SET_PROTECTED_MODE, 1);
 	OUT_RING(ring, 1);
 
-	gpu->funcs->flush(gpu, ring);
+	adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR);
 	return a2xx_idle(gpu);
 }
 
@@ -421,16 +463,11 @@ a2xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev)
 	return aspace;
 }
 
-/* Register offset defines for A2XX - copy of A3XX */
-static const unsigned int a2xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_AXXX_CP_RB_BASE),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_BASE_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR, REG_AXXX_CP_RB_RPTR_ADDR),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_RPTR_ADDR_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_AXXX_CP_RB_RPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_AXXX_CP_RB_WPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_AXXX_CP_RB_CNTL),
-};
+static u32 a2xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	ring->memptrs->rptr = gpu_read(gpu, REG_AXXX_CP_RB_RPTR);
+	return ring->memptrs->rptr;
+}
 
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
@@ -439,8 +476,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.pm_suspend = msm_gpu_pm_suspend,
 		.pm_resume = msm_gpu_pm_resume,
 		.recover = a2xx_recover,
-		.submit = adreno_submit,
-		.flush = adreno_flush,
+		.submit = a2xx_submit,
 		.active_ring = adreno_active_ring,
 		.irq = a2xx_irq,
 		.destroy = a2xx_destroy,
@@ -450,6 +486,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.gpu_state_get = a2xx_gpu_state_get,
 		.gpu_state_put = adreno_gpu_state_put,
 		.create_address_space = a2xx_create_address_space,
+		.get_rptr = a2xx_get_rptr,
 	},
 };
 
@@ -491,8 +528,6 @@ struct msm_gpu *a2xx_gpu_init(struct drm_device *dev)
 	else
 		adreno_gpu->registers = a220_registers;
 
-	adreno_gpu->reg_offsets = a2xx_register_offsets;
-
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
 	if (ret)
 		goto fail;
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
index f6471145a7a6..f29c77d9cd42 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
@@ -28,6 +28,61 @@ extern bool hang_debug;
 static void a3xx_dump(struct msm_gpu *gpu);
 static bool a3xx_idle(struct msm_gpu *gpu);
 
+static void a3xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
+{
+	struct msm_drm_private *priv = gpu->dev->dev_private;
+	struct msm_ringbuffer *ring = submit->ring;
+	unsigned int i;
+
+	for (i = 0; i < submit->nr_cmds; i++) {
+		switch (submit->cmd[i].type) {
+		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
+			/* ignore IB-targets */
+			break;
+		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
+			/* ignore if there has not been a ctx switch: */
+			if (priv->lastctx == submit->queue->ctx)
+				break;
+			fallthrough;
+		case MSM_SUBMIT_CMD_BUF:
+			OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFD, 2);
+			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
+			OUT_RING(ring, submit->cmd[i].size);
+			OUT_PKT2(ring);
+			break;
+		}
+	}
+
+	OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1);
+	OUT_RING(ring, submit->seqno);
+
+	/* Flush HLSQ lazy updates to make sure there is nothing
+	 * pending for indirect loads after the timestamp has
+	 * passed:
+	 */
+	OUT_PKT3(ring, CP_EVENT_WRITE, 1);
+	OUT_RING(ring, HLSQ_FLUSH);
+
+	/* wait for idle before cache flush/interrupt */
+	OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1);
+	OUT_RING(ring, 0x00000000);
+
+	/* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */
+	OUT_PKT3(ring, CP_EVENT_WRITE, 3);
+	OUT_RING(ring, CACHE_FLUSH_TS | BIT(31));
+	OUT_RING(ring, rbmemptr(ring, fence));
+	OUT_RING(ring, submit->seqno);
+
+#if 0
+	/* Dummy set-constant to trigger context rollover */
+	OUT_PKT3(ring, CP_SET_CONSTANT, 2);
+	OUT_RING(ring, CP_REG(REG_A3XX_HLSQ_CL_KERNEL_GROUP_X_REG));
+	OUT_RING(ring, 0x00000000);
+#endif
+
+	adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR);
+}
+
 static bool a3xx_me_init(struct msm_gpu *gpu)
 {
 	struct msm_ringbuffer *ring = gpu->rb[0];
@@ -51,7 +106,7 @@ static bool a3xx_me_init(struct msm_gpu *gpu)
 	OUT_RING(ring, 0x00000000);
 	OUT_RING(ring, 0x00000000);
 
-	gpu->funcs->flush(gpu, ring);
+	adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR);
 	return a3xx_idle(gpu);
 }
 
@@ -423,16 +478,11 @@ static struct msm_gpu_state *a3xx_gpu_state_get(struct msm_gpu *gpu)
 	return state;
 }
 
-/* Register offset defines for A3XX */
-static const unsigned int a3xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_AXXX_CP_RB_BASE),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_BASE_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR, REG_AXXX_CP_RB_RPTR_ADDR),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_RPTR_ADDR_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_AXXX_CP_RB_RPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_AXXX_CP_RB_WPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_AXXX_CP_RB_CNTL),
-};
+static u32 a3xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	ring->memptrs->rptr = gpu_read(gpu, REG_AXXX_CP_RB_RPTR);
+	return ring->memptrs->rptr;
+}
 
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
@@ -441,8 +491,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.pm_suspend = msm_gpu_pm_suspend,
 		.pm_resume = msm_gpu_pm_resume,
 		.recover = a3xx_recover,
-		.submit = adreno_submit,
-		.flush = adreno_flush,
+		.submit = a3xx_submit,
 		.active_ring = adreno_active_ring,
 		.irq = a3xx_irq,
 		.destroy = a3xx_destroy,
@@ -452,6 +501,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.gpu_state_get = a3xx_gpu_state_get,
 		.gpu_state_put = adreno_gpu_state_put,
 		.create_address_space = adreno_iommu_create_address_space,
+		.get_rptr = a3xx_get_rptr,
 	},
 };
 
@@ -490,7 +540,6 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 	gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs);
 
 	adreno_gpu->registers = a3xx_registers;
-	adreno_gpu->reg_offsets = a3xx_register_offsets;
 
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
 	if (ret)
diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
index 954753600625..2b93b33b05e4 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
@@ -22,6 +22,54 @@ extern bool hang_debug;
 static void a4xx_dump(struct msm_gpu *gpu);
 static bool a4xx_idle(struct msm_gpu *gpu);
 
+static void a4xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
+{
+	struct msm_drm_private *priv = gpu->dev->dev_private;
+	struct msm_ringbuffer *ring = submit->ring;
+	unsigned int i;
+
+	for (i = 0; i < submit->nr_cmds; i++) {
+		switch (submit->cmd[i].type) {
+		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
+			/* ignore IB-targets */
+			break;
+		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
+			/* ignore if there has not been a ctx switch: */
+			if (priv->lastctx == submit->queue->ctx)
+				break;
+			fallthrough;
+		case MSM_SUBMIT_CMD_BUF:
+			OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFE, 2);
+			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
+			OUT_RING(ring, submit->cmd[i].size);
+			OUT_PKT2(ring);
+			break;
+		}
+	}
+
+	OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1);
+	OUT_RING(ring, submit->seqno);
+
+	/* Flush HLSQ lazy updates to make sure there is nothing
+	 * pending for indirect loads after the timestamp has
+	 * passed:
+	 */
+	OUT_PKT3(ring, CP_EVENT_WRITE, 1);
+	OUT_RING(ring, HLSQ_FLUSH);
+
+	/* wait for idle before cache flush/interrupt */
+	OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1);
+	OUT_RING(ring, 0x00000000);
+
+	/* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */
+	OUT_PKT3(ring, CP_EVENT_WRITE, 3);
+	OUT_RING(ring, CACHE_FLUSH_TS | BIT(31));
+	OUT_RING(ring, rbmemptr(ring, fence));
+	OUT_RING(ring, submit->seqno);
+
+	adreno_flush(gpu, ring, REG_A4XX_CP_RB_WPTR);
+}
+
 /*
  * a4xx_enable_hwcg() - Program the clock control registers
  * @device: The adreno device pointer
@@ -129,7 +177,7 @@ static bool a4xx_me_init(struct msm_gpu *gpu)
 	OUT_RING(ring, 0x00000000);
 	OUT_RING(ring, 0x00000000);
 
-	gpu->funcs->flush(gpu, ring);
+	adreno_flush(gpu, ring, REG_A4XX_CP_RB_WPTR);
 	return a4xx_idle(gpu);
 }
 
@@ -515,17 +563,6 @@ static struct msm_gpu_state *a4xx_gpu_state_get(struct msm_gpu *gpu)
 	return state;
 }
 
-/* Register offset defines for A4XX, in order of enum adreno_regs */
-static const unsigned int a4xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_A4XX_CP_RB_BASE),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_BASE_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR, REG_A4XX_CP_RB_RPTR_ADDR),
-	REG_ADRENO_SKIP(REG_ADRENO_CP_RB_RPTR_ADDR_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_A4XX_CP_RB_RPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_A4XX_CP_RB_WPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A4XX_CP_RB_CNTL),
-};
-
 static void a4xx_dump(struct msm_gpu *gpu)
 {
 	printk("status:   %08x\n",
@@ -576,6 +613,12 @@ static int a4xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 	return 0;
 }
 
+static u32 a4xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	ring->memptrs->rptr = gpu_read(gpu, REG_A4XX_CP_RB_RPTR);
+	return ring->memptrs->rptr;
+}
+
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
 		.get_param = adreno_get_param,
@@ -583,8 +626,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.pm_suspend = a4xx_pm_suspend,
 		.pm_resume = a4xx_pm_resume,
 		.recover = a4xx_recover,
-		.submit = adreno_submit,
-		.flush = adreno_flush,
+		.submit = a4xx_submit,
 		.active_ring = adreno_active_ring,
 		.irq = a4xx_irq,
 		.destroy = a4xx_destroy,
@@ -594,6 +636,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.gpu_state_get = a4xx_gpu_state_get,
 		.gpu_state_put = adreno_gpu_state_put,
 		.create_address_space = adreno_iommu_create_address_space,
+		.get_rptr = a4xx_get_rptr,
 	},
 	.get_timestamp = a4xx_get_timestamp,
 };
@@ -631,15 +674,12 @@ struct msm_gpu *a4xx_gpu_init(struct drm_device *dev)
 
 	adreno_gpu->registers = adreno_is_a405(adreno_gpu) ? a405_registers :
 							     a4xx_registers;
-	adreno_gpu->reg_offsets = a4xx_register_offsets;
 
 	/* if needed, allocate gmem: */
-	if (adreno_is_a4xx(adreno_gpu)) {
-		ret = adreno_gpu_ocmem_init(dev->dev, adreno_gpu,
-					    &a4xx_gpu->ocmem);
-		if (ret)
-			goto fail;
-	}
+	ret = adreno_gpu_ocmem_init(dev->dev, adreno_gpu,
+				    &a4xx_gpu->ocmem);
+	if (ret)
+		goto fail;
 
 	if (!gpu->aspace) {
 		/* TODO we think it is possible to configure the GPU to
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
index 68eddac7771c..fc2c905b6c9e 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
@@ -11,7 +11,7 @@
 
 #include "a5xx_gpu.h"
 
-static int pfp_print(struct msm_gpu *gpu, struct drm_printer *p)
+static void pfp_print(struct msm_gpu *gpu, struct drm_printer *p)
 {
 	int i;
 
@@ -22,11 +22,9 @@ static int pfp_print(struct msm_gpu *gpu, struct drm_printer *p)
 		drm_printf(p, "  %02x: %08x\n", i,
 			gpu_read(gpu, REG_A5XX_CP_PFP_STAT_DATA));
 	}
-
-	return 0;
 }
 
-static int me_print(struct msm_gpu *gpu, struct drm_printer *p)
+static void me_print(struct msm_gpu *gpu, struct drm_printer *p)
 {
 	int i;
 
@@ -37,11 +35,9 @@ static int me_print(struct msm_gpu *gpu, struct drm_printer *p)
 		drm_printf(p, "  %02x: %08x\n", i,
 			gpu_read(gpu, REG_A5XX_CP_ME_STAT_DATA));
 	}
-
-	return 0;
 }
 
-static int meq_print(struct msm_gpu *gpu, struct drm_printer *p)
+static void meq_print(struct msm_gpu *gpu, struct drm_printer *p)
 {
 	int i;
 
@@ -52,11 +48,9 @@ static int meq_print(struct msm_gpu *gpu, struct drm_printer *p)
 		drm_printf(p, "  %02x: %08x\n", i,
 			gpu_read(gpu, REG_A5XX_CP_MEQ_DBG_DATA));
 	}
-
-	return 0;
 }
 
-static int roq_print(struct msm_gpu *gpu, struct drm_printer *p)
+static void roq_print(struct msm_gpu *gpu, struct drm_printer *p)
 {
 	int i;
 
@@ -71,8 +65,6 @@ static int roq_print(struct msm_gpu *gpu, struct drm_printer *p)
 		drm_printf(p, "  %02x: %08x %08x %08x %08x\n", i,
 			val[0], val[1], val[2], val[3]);
 	}
-
-	return 0;
 }
 
 static int show(struct seq_file *m, void *arg)
@@ -81,10 +73,11 @@ static int show(struct seq_file *m, void *arg)
 	struct drm_device *dev = node->minor->dev;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_printer p = drm_seq_file_printer(m);
-	int (*show)(struct msm_gpu *gpu, struct drm_printer *p) =
+	void (*show)(struct msm_gpu *gpu, struct drm_printer *p) =
 		node->info_ent->data;
 
-	return show(priv->gpu, &p);
+	show(priv->gpu, &p);
+	return 0;
 }
 
 #define ENT(n) { .name = #n, .show = show, .data = n ##_print }
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
index 91726da82ed6..d6804a802355 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
@@ -18,13 +18,24 @@ static void a5xx_dump(struct msm_gpu *gpu);
 
 #define GPU_PAS_ID 13
 
-static void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring,
+		bool sync)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu);
 	uint32_t wptr;
 	unsigned long flags;
 
+	/*
+	 * Most flush operations need to issue a WHERE_AM_I opcode to sync up
+	 * the rptr shadow
+	 */
+	if (a5xx_gpu->has_whereami && sync) {
+		OUT_PKT7(ring, CP_WHERE_AM_I, 2);
+		OUT_RING(ring, lower_32_bits(shadowptr(a5xx_gpu, ring)));
+		OUT_RING(ring, upper_32_bits(shadowptr(a5xx_gpu, ring)));
+	}
+
 	spin_lock_irqsave(&ring->lock, flags);
 
 	/* Copy the shadow to the actual register */
@@ -43,8 +54,7 @@ static void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 		gpu_write(gpu, REG_A5XX_CP_RB_WPTR, wptr);
 }
 
-static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-	struct msm_file_private *ctx)
+static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 {
 	struct msm_drm_private *priv = gpu->dev->dev_private;
 	struct msm_ringbuffer *ring = submit->ring;
@@ -57,7 +67,7 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit
 		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
 			break;
 		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
-			if (priv->lastctx == ctx)
+			if (priv->lastctx == submit->queue->ctx)
 				break;
 			fallthrough;
 		case MSM_SUBMIT_CMD_BUF:
@@ -91,7 +101,7 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit
 		}
 	}
 
-	a5xx_flush(gpu, ring);
+	a5xx_flush(gpu, ring, true);
 	a5xx_preempt_trigger(gpu);
 
 	/* we might not necessarily have a cmd from userspace to
@@ -103,8 +113,7 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit
 	msm_gpu_retire(gpu);
 }
 
-static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-	struct msm_file_private *ctx)
+static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu);
@@ -114,7 +123,7 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 
 	if (IS_ENABLED(CONFIG_DRM_MSM_GPU_SUDO) && submit->in_rb) {
 		priv->lastctx = NULL;
-		a5xx_submit_in_rb(gpu, submit, ctx);
+		a5xx_submit_in_rb(gpu, submit);
 		return;
 	}
 
@@ -148,7 +157,7 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
 			break;
 		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
-			if (priv->lastctx == ctx)
+			if (priv->lastctx == submit->queue->ctx)
 				break;
 			fallthrough;
 		case MSM_SUBMIT_CMD_BUF:
@@ -206,7 +215,8 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 	/* Set bit 0 to trigger an interrupt on preempt complete */
 	OUT_RING(ring, 0x01);
 
-	a5xx_flush(gpu, ring);
+	/* A WHERE_AM_I packet is not needed after a YIELD */
+	a5xx_flush(gpu, ring, false);
 
 	/* Check to see if we need to start preemption */
 	a5xx_preempt_trigger(gpu);
@@ -365,7 +375,7 @@ static int a5xx_me_init(struct msm_gpu *gpu)
 	OUT_RING(ring, 0x00000000);
 	OUT_RING(ring, 0x00000000);
 
-	gpu->funcs->flush(gpu, ring);
+	a5xx_flush(gpu, ring, true);
 	return a5xx_idle(gpu, ring) ? 0 : -EINVAL;
 }
 
@@ -407,11 +417,31 @@ static int a5xx_preempt_start(struct msm_gpu *gpu)
 	OUT_RING(ring, 0x01);
 	OUT_RING(ring, 0x01);
 
-	gpu->funcs->flush(gpu, ring);
+	/* The WHERE_AMI_I packet is not needed after a YIELD is issued */
+	a5xx_flush(gpu, ring, false);
 
 	return a5xx_idle(gpu, ring) ? 0 : -EINVAL;
 }
 
+static void a5xx_ucode_check_version(struct a5xx_gpu *a5xx_gpu,
+		struct drm_gem_object *obj)
+{
+	u32 *buf = msm_gem_get_vaddr_active(obj);
+
+	if (IS_ERR(buf))
+		return;
+
+	/*
+	 * If the lowest nibble is 0xa that is an indication that this microcode
+	 * has been patched. The actual version is in dword [3] but we only care
+	 * about the patchlevel which is the lowest nibble of dword [3]
+	 */
+	if (((buf[0] & 0xf) == 0xa) && (buf[2] & 0xf) >= 1)
+		a5xx_gpu->has_whereami = true;
+
+	msm_gem_put_vaddr(obj);
+}
+
 static int a5xx_ucode_init(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -447,6 +477,7 @@ static int a5xx_ucode_init(struct msm_gpu *gpu)
 		}
 
 		msm_gem_object_set_name(a5xx_gpu->pfp_bo, "pfpfw");
+		a5xx_ucode_check_version(a5xx_gpu, a5xx_gpu->pfp_bo);
 	}
 
 	gpu_write64(gpu, REG_A5XX_CP_ME_INSTR_BASE_LO,
@@ -506,6 +537,7 @@ static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 static int a5xx_hw_init(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu);
 	int ret;
 
 	gpu_write(gpu, REG_A5XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003);
@@ -714,9 +746,36 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 	gpu_write64(gpu, REG_A5XX_CP_RB_BASE, REG_A5XX_CP_RB_BASE_HI,
 		gpu->rb[0]->iova);
 
+	/*
+	 * If the microcode supports the WHERE_AM_I opcode then we can use that
+	 * in lieu of the RPTR shadow and enable preemption. Otherwise, we
+	 * can't safely use the RPTR shadow or preemption. In either case, the
+	 * RPTR shadow should be disabled in hardware.
+	 */
 	gpu_write(gpu, REG_A5XX_CP_RB_CNTL,
 		MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE);
 
+	/* Disable preemption if WHERE_AM_I isn't available */
+	if (!a5xx_gpu->has_whereami && gpu->nr_rings > 1) {
+		a5xx_preempt_fini(gpu);
+		gpu->nr_rings = 1;
+	} else {
+		/* Create a privileged buffer for the RPTR shadow */
+		if (!a5xx_gpu->shadow_bo) {
+			a5xx_gpu->shadow = msm_gem_kernel_new(gpu->dev,
+				sizeof(u32) * gpu->nr_rings,
+				MSM_BO_UNCACHED | MSM_BO_MAP_PRIV,
+				gpu->aspace, &a5xx_gpu->shadow_bo,
+				&a5xx_gpu->shadow_iova);
+
+			if (IS_ERR(a5xx_gpu->shadow))
+				return PTR_ERR(a5xx_gpu->shadow);
+		}
+
+		gpu_write64(gpu, REG_A5XX_CP_RB_RPTR_ADDR,
+			REG_A5XX_CP_RB_RPTR_ADDR_HI, shadowptr(a5xx_gpu, gpu->rb[0]));
+	}
+
 	a5xx_preempt_hw_init(gpu);
 
 	/* Disable the interrupts through the initial bringup stage */
@@ -740,7 +799,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 		OUT_PKT7(gpu->rb[0], CP_EVENT_WRITE, 1);
 		OUT_RING(gpu->rb[0], CP_EVENT_WRITE_0_EVENT(STAT_EVENT));
 
-		gpu->funcs->flush(gpu, gpu->rb[0]);
+		a5xx_flush(gpu, gpu->rb[0], true);
 		if (!a5xx_idle(gpu, gpu->rb[0]))
 			return -EINVAL;
 	}
@@ -758,7 +817,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu)
 		OUT_PKT7(gpu->rb[0], CP_SET_SECURE_MODE, 1);
 		OUT_RING(gpu->rb[0], 0x00000000);
 
-		gpu->funcs->flush(gpu, gpu->rb[0]);
+		a5xx_flush(gpu, gpu->rb[0], true);
 		if (!a5xx_idle(gpu, gpu->rb[0]))
 			return -EINVAL;
 	} else if (ret == -ENODEV) {
@@ -825,6 +884,11 @@ static void a5xx_destroy(struct msm_gpu *gpu)
 		drm_gem_object_put(a5xx_gpu->gpmu_bo);
 	}
 
+	if (a5xx_gpu->shadow_bo) {
+		msm_gem_unpin_iova(a5xx_gpu->shadow_bo, gpu->aspace);
+		drm_gem_object_put(a5xx_gpu->shadow_bo);
+	}
+
 	adreno_gpu_cleanup(adreno_gpu);
 	kfree(a5xx_gpu);
 }
@@ -1057,17 +1121,6 @@ static irqreturn_t a5xx_irq(struct msm_gpu *gpu)
 	return IRQ_HANDLED;
 }
 
-static const u32 a5xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_A5XX_CP_RB_BASE),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE_HI, REG_A5XX_CP_RB_BASE_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR, REG_A5XX_CP_RB_RPTR_ADDR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR_HI,
-		REG_A5XX_CP_RB_RPTR_ADDR_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_A5XX_CP_RB_RPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_A5XX_CP_RB_WPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A5XX_CP_RB_CNTL),
-};
-
 static const u32 a5xx_registers[] = {
 	0x0000, 0x0002, 0x0004, 0x0020, 0x0022, 0x0026, 0x0029, 0x002B,
 	0x002E, 0x0035, 0x0038, 0x0042, 0x0044, 0x0044, 0x0047, 0x0095,
@@ -1432,6 +1485,17 @@ static unsigned long a5xx_gpu_busy(struct msm_gpu *gpu)
 	return (unsigned long)busy_time;
 }
 
+static uint32_t a5xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu);
+
+	if (a5xx_gpu->has_whereami)
+		return a5xx_gpu->shadow[ring->id];
+
+	return ring->memptrs->rptr = gpu_read(gpu, REG_A5XX_CP_RB_RPTR);
+}
+
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
 		.get_param = adreno_get_param,
@@ -1440,7 +1504,6 @@ static const struct adreno_gpu_funcs funcs = {
 		.pm_resume = a5xx_pm_resume,
 		.recover = a5xx_recover,
 		.submit = a5xx_submit,
-		.flush = a5xx_flush,
 		.active_ring = a5xx_active_ring,
 		.irq = a5xx_irq,
 		.destroy = a5xx_destroy,
@@ -1454,6 +1517,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.gpu_state_get = a5xx_gpu_state_get,
 		.gpu_state_put = a5xx_gpu_state_put,
 		.create_address_space = adreno_iommu_create_address_space,
+		.get_rptr = a5xx_get_rptr,
 	},
 	.get_timestamp = a5xx_get_timestamp,
 };
@@ -1512,14 +1576,12 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev)
 	gpu = &adreno_gpu->base;
 
 	adreno_gpu->registers = a5xx_registers;
-	adreno_gpu->reg_offsets = a5xx_register_offsets;
 
 	a5xx_gpu->lm_leakage = 0x4E001A;
 
 	check_speed_bin(&pdev->dev);
 
-	/* Restricting nr_rings to 1 to temporarily disable preemption */
-	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
+	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4);
 	if (ret) {
 		a5xx_destroy(&(a5xx_gpu->base.base));
 		return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.h b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
index 1e5b1a15a70f..c7187bcc5e90 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h
@@ -37,6 +37,13 @@ struct a5xx_gpu {
 
 	atomic_t preempt_state;
 	struct timer_list preempt_timer;
+
+	struct drm_gem_object *shadow_bo;
+	uint64_t shadow_iova;
+	uint32_t *shadow;
+
+	/* True if the microcode supports the WHERE_AM_I opcode */
+	bool has_whereami;
 };
 
 #define to_a5xx_gpu(x) container_of(x, struct a5xx_gpu, base)
@@ -141,6 +148,9 @@ static inline int spin_usecs(struct msm_gpu *gpu, uint32_t usecs,
 	return -ETIMEDOUT;
 }
 
+#define shadowptr(a5xx_gpu, ring) ((a5xx_gpu)->shadow_iova + \
+		((ring)->id * sizeof(uint32_t)))
+
 bool a5xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
 void a5xx_set_hwcg(struct msm_gpu *gpu, bool state);
 
@@ -150,6 +160,8 @@ void a5xx_preempt_trigger(struct msm_gpu *gpu);
 void a5xx_preempt_irq(struct msm_gpu *gpu);
 void a5xx_preempt_fini(struct msm_gpu *gpu);
 
+void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, bool sync);
+
 /* Return true if we are in a preempt state */
 static inline bool a5xx_in_preempt(struct a5xx_gpu *a5xx_gpu)
 {
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c
index 321a8061fd32..f176a6f3eff6 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_power.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c
@@ -240,7 +240,7 @@ static int a5xx_gpmu_init(struct msm_gpu *gpu)
 	OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1);
 	OUT_RING(ring, 1);
 
-	gpu->funcs->flush(gpu, ring);
+	a5xx_flush(gpu, ring, true);
 
 	if (!a5xx_idle(gpu, ring)) {
 		DRM_ERROR("%s: Unable to load GPMU firmware. GPMU will not be active\n",
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
index 9f3fe177b00e..7e04509c4e1f 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c
@@ -259,8 +259,9 @@ static int preempt_init_ring(struct a5xx_gpu *a5xx_gpu,
 	ptr->magic = A5XX_PREEMPT_RECORD_MAGIC;
 	ptr->info = 0;
 	ptr->data = 0;
-	ptr->cntl = MSM_GPU_RB_CNTL_DEFAULT;
-	ptr->rptr_addr = rbmemptr(ring, rptr);
+	ptr->cntl = MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE;
+
+	ptr->rptr_addr = shadowptr(a5xx_gpu, ring);
 	ptr->counter = counters_iova;
 
 	return 0;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index e1c7bcd1b1eb..491fee410daf 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -11,6 +11,7 @@
 #include "a6xx_gpu.h"
 #include "a6xx_gmu.xml.h"
 #include "msm_gem.h"
+#include "msm_gpu_trace.h"
 #include "msm_mmu.h"
 
 static void a6xx_gmu_fault(struct a6xx_gmu *gmu)
@@ -124,6 +125,8 @@ void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp)
 	gmu->current_perf_index = perf_index;
 	gmu->freq = gmu->gpu_freqs[perf_index];
 
+	trace_msm_gmu_freq_change(gmu->freq, perf_index);
+
 	/*
 	 * This can get called from devfreq while the hardware is idle. Don't
 	 * bring up the power if it isn't already active
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 66a95e22b7b3..948f3656c20c 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -51,9 +51,20 @@ bool a6xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 
 static void a6xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 {
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 	uint32_t wptr;
 	unsigned long flags;
 
+	/* Expanded APRIV doesn't need to issue the WHERE_AM_I opcode */
+	if (a6xx_gpu->has_whereami && !adreno_gpu->base.hw_apriv) {
+		struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+
+		OUT_PKT7(ring, CP_WHERE_AM_I, 2);
+		OUT_RING(ring, lower_32_bits(shadowptr(a6xx_gpu, ring)));
+		OUT_RING(ring, upper_32_bits(shadowptr(a6xx_gpu, ring)));
+	}
+
 	spin_lock_irqsave(&ring->lock, flags);
 
 	/* Copy the shadow to the actual register */
@@ -81,8 +92,50 @@ static void get_stats_counter(struct msm_ringbuffer *ring, u32 counter,
 	OUT_RING(ring, upper_32_bits(iova));
 }
 
-static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-	struct msm_file_private *ctx)
+static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu,
+		struct msm_ringbuffer *ring, struct msm_file_private *ctx)
+{
+	phys_addr_t ttbr;
+	u32 asid;
+	u64 memptr = rbmemptr(ring, ttbr0);
+
+	if (ctx == a6xx_gpu->cur_ctx)
+		return;
+
+	if (msm_iommu_pagetable_params(ctx->aspace->mmu, &ttbr, &asid))
+		return;
+
+	/* Execute the table update */
+	OUT_PKT7(ring, CP_SMMU_TABLE_UPDATE, 4);
+	OUT_RING(ring, CP_SMMU_TABLE_UPDATE_0_TTBR0_LO(lower_32_bits(ttbr)));
+
+	OUT_RING(ring,
+		CP_SMMU_TABLE_UPDATE_1_TTBR0_HI(upper_32_bits(ttbr)) |
+		CP_SMMU_TABLE_UPDATE_1_ASID(asid));
+	OUT_RING(ring, CP_SMMU_TABLE_UPDATE_2_CONTEXTIDR(0));
+	OUT_RING(ring, CP_SMMU_TABLE_UPDATE_3_CONTEXTBANK(0));
+
+	/*
+	 * Write the new TTBR0 to the memstore. This is good for debugging.
+	 */
+	OUT_PKT7(ring, CP_MEM_WRITE, 4);
+	OUT_RING(ring, CP_MEM_WRITE_0_ADDR_LO(lower_32_bits(memptr)));
+	OUT_RING(ring, CP_MEM_WRITE_1_ADDR_HI(upper_32_bits(memptr)));
+	OUT_RING(ring, lower_32_bits(ttbr));
+	OUT_RING(ring, (asid << 16) | upper_32_bits(ttbr));
+
+	/*
+	 * And finally, trigger a uche flush to be sure there isn't anything
+	 * lingering in that part of the GPU
+	 */
+
+	OUT_PKT7(ring, CP_EVENT_WRITE, 1);
+	OUT_RING(ring, 0x31);
+
+	a6xx_gpu->cur_ctx = ctx;
+}
+
+static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 {
 	unsigned int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT;
 	struct msm_drm_private *priv = gpu->dev->dev_private;
@@ -91,6 +144,8 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 	struct msm_ringbuffer *ring = submit->ring;
 	unsigned int i;
 
+	a6xx_set_pagetable(a6xx_gpu, ring, submit->queue->ctx);
+
 	get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP_0_LO,
 		rbmemptr_stats(ring, index, cpcycles_start));
 
@@ -115,7 +170,7 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
 			break;
 		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
-			if (priv->lastctx == ctx)
+			if (priv->lastctx == submit->queue->ctx)
 				break;
 			fallthrough;
 		case MSM_SUBMIT_CMD_BUF:
@@ -464,6 +519,30 @@ static int a6xx_cp_init(struct msm_gpu *gpu)
 	return a6xx_idle(gpu, ring) ? 0 : -EINVAL;
 }
 
+static void a6xx_ucode_check_version(struct a6xx_gpu *a6xx_gpu,
+		struct drm_gem_object *obj)
+{
+	u32 *buf = msm_gem_get_vaddr_active(obj);
+
+	if (IS_ERR(buf))
+		return;
+
+	/*
+	 * If the lowest nibble is 0xa that is an indication that this microcode
+	 * has been patched. The actual version is in dword [3] but we only care
+	 * about the patchlevel which is the lowest nibble of dword [3]
+	 *
+	 * Otherwise check that the firmware is greater than or equal to 1.90
+	 * which was the first version that had this fix built in
+	 */
+	if (((buf[0] & 0xf) == 0xa) && (buf[2] & 0xf) >= 1)
+		a6xx_gpu->has_whereami = true;
+	else if ((buf[0] & 0xfff) > 0x190)
+		a6xx_gpu->has_whereami = true;
+
+	msm_gem_put_vaddr(obj);
+}
+
 static int a6xx_ucode_init(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -484,6 +563,7 @@ static int a6xx_ucode_init(struct msm_gpu *gpu)
 		}
 
 		msm_gem_object_set_name(a6xx_gpu->sqe_bo, "sqefw");
+		a6xx_ucode_check_version(a6xx_gpu, a6xx_gpu->sqe_bo);
 	}
 
 	gpu_write64(gpu, REG_A6XX_CP_SQE_INSTR_BASE_LO,
@@ -699,12 +779,43 @@ static int a6xx_hw_init(struct msm_gpu *gpu)
 	gpu_write64(gpu, REG_A6XX_CP_RB_BASE, REG_A6XX_CP_RB_BASE_HI,
 		gpu->rb[0]->iova);
 
-	gpu_write(gpu, REG_A6XX_CP_RB_CNTL,
-		MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE);
+	/* Targets that support extended APRIV can use the RPTR shadow from
+	 * hardware but all the other ones need to disable the feature. Targets
+	 * that support the WHERE_AM_I opcode can use that instead
+	 */
+	if (adreno_gpu->base.hw_apriv)
+		gpu_write(gpu, REG_A6XX_CP_RB_CNTL, MSM_GPU_RB_CNTL_DEFAULT);
+	else
+		gpu_write(gpu, REG_A6XX_CP_RB_CNTL,
+			MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE);
+
+	/*
+	 * Expanded APRIV and targets that support WHERE_AM_I both need a
+	 * privileged buffer to store the RPTR shadow
+	 */
+
+	if (adreno_gpu->base.hw_apriv || a6xx_gpu->has_whereami) {
+		if (!a6xx_gpu->shadow_bo) {
+			a6xx_gpu->shadow = msm_gem_kernel_new_locked(gpu->dev,
+				sizeof(u32) * gpu->nr_rings,
+				MSM_BO_UNCACHED | MSM_BO_MAP_PRIV,
+				gpu->aspace, &a6xx_gpu->shadow_bo,
+				&a6xx_gpu->shadow_iova);
+
+			if (IS_ERR(a6xx_gpu->shadow))
+				return PTR_ERR(a6xx_gpu->shadow);
+		}
+
+		gpu_write64(gpu, REG_A6XX_CP_RB_RPTR_ADDR_LO,
+			REG_A6XX_CP_RB_RPTR_ADDR_HI,
+			shadowptr(a6xx_gpu, gpu->rb[0]));
+	}
 
 	/* Always come up on rb 0 */
 	a6xx_gpu->cur_ring = gpu->rb[0];
 
+	a6xx_gpu->cur_ctx = NULL;
+
 	/* Enable the SQE_to start the CP engine */
 	gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 1);
 
@@ -911,18 +1022,6 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 	return IRQ_HANDLED;
 }
 
-static const u32 a6xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_A6XX_CP_RB_BASE),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE_HI, REG_A6XX_CP_RB_BASE_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR,
-		REG_A6XX_CP_RB_RPTR_ADDR_LO),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR_HI,
-		REG_A6XX_CP_RB_RPTR_ADDR_HI),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_A6XX_CP_RB_RPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_A6XX_CP_RB_WPTR),
-	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A6XX_CP_RB_CNTL),
-};
-
 static int a6xx_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -931,6 +1030,8 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
 
 	gpu->needs_hw_init = true;
 
+	trace_msm_gpu_resume(0);
+
 	ret = a6xx_gmu_resume(a6xx_gpu);
 	if (ret)
 		return ret;
@@ -945,6 +1046,8 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 
+	trace_msm_gpu_suspend(0);
+
 	devfreq_suspend_device(gpu->devfreq.devfreq);
 
 	return a6xx_gmu_stop(a6xx_gpu);
@@ -983,6 +1086,11 @@ static void a6xx_destroy(struct msm_gpu *gpu)
 		drm_gem_object_put(a6xx_gpu->sqe_bo);
 	}
 
+	if (a6xx_gpu->shadow_bo) {
+		msm_gem_unpin_iova(a6xx_gpu->shadow_bo, gpu->aspace);
+		drm_gem_object_put(a6xx_gpu->shadow_bo);
+	}
+
 	a6xx_gmu_remove(a6xx_gpu);
 
 	adreno_gpu_cleanup(adreno_gpu);
@@ -1017,6 +1125,31 @@ static unsigned long a6xx_gpu_busy(struct msm_gpu *gpu)
 	return (unsigned long)busy_time;
 }
 
+static struct msm_gem_address_space *
+a6xx_create_private_address_space(struct msm_gpu *gpu)
+{
+	struct msm_mmu *mmu;
+
+	mmu = msm_iommu_pagetable_create(gpu->aspace->mmu);
+
+	if (IS_ERR(mmu))
+		return ERR_CAST(mmu);
+
+	return msm_gem_address_space_create(mmu,
+		"gpu", 0x100000000ULL, 0x1ffffffffULL);
+}
+
+static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+
+	if (adreno_gpu->base.hw_apriv || a6xx_gpu->has_whereami)
+		return a6xx_gpu->shadow[ring->id];
+
+	return ring->memptrs->rptr = gpu_read(gpu, REG_A6XX_CP_RB_RPTR);
+}
+
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
 		.get_param = adreno_get_param,
@@ -1025,7 +1158,6 @@ static const struct adreno_gpu_funcs funcs = {
 		.pm_resume = a6xx_pm_resume,
 		.recover = a6xx_recover,
 		.submit = a6xx_submit,
-		.flush = a6xx_flush,
 		.active_ring = a6xx_active_ring,
 		.irq = a6xx_irq,
 		.destroy = a6xx_destroy,
@@ -1040,6 +1172,8 @@ static const struct adreno_gpu_funcs funcs = {
 		.gpu_state_put = a6xx_gpu_state_put,
 #endif
 		.create_address_space = adreno_iommu_create_address_space,
+		.create_private_address_space = a6xx_create_private_address_space,
+		.get_rptr = a6xx_get_rptr,
 	},
 	.get_timestamp = a6xx_get_timestamp,
 };
@@ -1048,6 +1182,8 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct platform_device *pdev = priv->gpu_pdev;
+	struct adreno_platform_config *config = pdev->dev.platform_data;
+	const struct adreno_info *info;
 	struct device_node *node;
 	struct a6xx_gpu *a6xx_gpu;
 	struct adreno_gpu *adreno_gpu;
@@ -1062,9 +1198,15 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	gpu = &adreno_gpu->base;
 
 	adreno_gpu->registers = NULL;
-	adreno_gpu->reg_offsets = a6xx_register_offsets;
 
-	if (adreno_is_a650(adreno_gpu))
+	/*
+	 * We need to know the platform type before calling into adreno_gpu_init
+	 * so that the hw_apriv flag can be correctly set. Snoop into the info
+	 * and grab the revision number
+	 */
+	info = adreno_info(config->rev);
+
+	if (info && info->revn == 650)
 		adreno_gpu->base.hw_apriv = true;
 
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 03ba60d5b07f..3eeebf6a754b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -19,8 +19,15 @@ struct a6xx_gpu {
 	uint64_t sqe_iova;
 
 	struct msm_ringbuffer *cur_ring;
+	struct msm_file_private *cur_ctx;
 
 	struct a6xx_gmu gmu;
+
+	struct drm_gem_object *shadow_bo;
+	uint64_t shadow_iova;
+	uint32_t *shadow;
+
+	bool has_whereami;
 };
 
 #define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base)
@@ -50,6 +57,9 @@ static inline bool a6xx_has_gbif(struct adreno_gpu *gpu)
 	return true;
 }
 
+#define shadowptr(_a6xx_gpu, _ring) ((_a6xx_gpu)->shadow_iova + \
+		((_ring)->id * sizeof(uint32_t)))
+
 int a6xx_gmu_resume(struct a6xx_gpu *gpu);
 int a6xx_gmu_stop(struct a6xx_gpu *gpu);
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
index b12f5b4a1bea..e9ede19193b0 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
@@ -875,7 +875,7 @@ static void a6xx_get_indexed_registers(struct msm_gpu *gpu,
 	int i;
 
 	a6xx_state->indexed_regs = state_kcalloc(a6xx_state, count,
-		sizeof(a6xx_state->indexed_regs));
+		sizeof(*a6xx_state->indexed_regs));
 	if (!a6xx_state->indexed_regs)
 		return;
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 9eeb46bf2a5d..58e03b20e1c7 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -282,7 +282,7 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev)
 	int ret;
 
 	if (pdev)
-		gpu = platform_get_drvdata(pdev);
+		gpu = dev_to_gpu(&pdev->dev);
 
 	if (!gpu) {
 		dev_err_once(dev->dev, "no GPU device was found\n");
@@ -417,15 +417,13 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
 		return PTR_ERR(gpu);
 	}
 
-	dev_set_drvdata(dev, gpu);
-
 	return 0;
 }
 
 static void adreno_unbind(struct device *dev, struct device *master,
 		void *data)
 {
-	struct msm_gpu *gpu = dev_get_drvdata(dev);
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 
 	pm_runtime_force_suspend(dev);
 	gpu->funcs->destroy(gpu);
@@ -490,16 +488,14 @@ static const struct of_device_id dt_match[] = {
 #ifdef CONFIG_PM
 static int adreno_resume(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct msm_gpu *gpu = platform_get_drvdata(pdev);
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 
 	return gpu->funcs->pm_resume(gpu);
 }
 
 static int adreno_suspend(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct msm_gpu *gpu = platform_get_drvdata(pdev);
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 
 	return gpu->funcs->pm_suspend(gpu);
 }
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 862dd35b27d3..458b5b26d3c2 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -189,12 +189,27 @@ struct msm_gem_address_space *
 adreno_iommu_create_address_space(struct msm_gpu *gpu,
 		struct platform_device *pdev)
 {
-	struct iommu_domain *iommu = iommu_domain_alloc(&platform_bus_type);
-	struct msm_mmu *mmu = msm_iommu_new(&pdev->dev, iommu);
+	struct iommu_domain *iommu;
+	struct msm_mmu *mmu;
 	struct msm_gem_address_space *aspace;
+	u64 start, size;
 
-	aspace = msm_gem_address_space_create(mmu, "gpu", SZ_16M,
-		0xffffffff - SZ_16M);
+	iommu = iommu_domain_alloc(&platform_bus_type);
+	if (!iommu)
+		return NULL;
+
+	mmu = msm_iommu_new(&pdev->dev, iommu);
+
+	/*
+	 * Use the aperture start or SZ_16M, whichever is greater. This will
+	 * ensure that we align with the allocated pagetable range while still
+	 * allowing room in the lower 32 bits for GMEM and whatnot
+	 */
+	start = max_t(u64, SZ_16M, iommu->geometry.aperture_start);
+	size = iommu->geometry.aperture_end - start + 1;
+
+	aspace = msm_gem_address_space_create(mmu, "gpu",
+		start & GENMASK_ULL(48, 0), size);
 
 	if (IS_ERR(aspace) && !IS_ERR(mmu))
 		mmu->funcs->destroy(mmu);
@@ -407,8 +422,9 @@ int adreno_hw_init(struct msm_gpu *gpu)
 static uint32_t get_rptr(struct adreno_gpu *adreno_gpu,
 		struct msm_ringbuffer *ring)
 {
-	return ring->memptrs->rptr = adreno_gpu_read(
-		adreno_gpu, REG_ADRENO_CP_RB_RPTR);
+	struct msm_gpu *gpu = &adreno_gpu->base;
+
+	return gpu->funcs->get_rptr(gpu, ring);
 }
 
 struct msm_ringbuffer *adreno_active_ring(struct msm_gpu *gpu)
@@ -434,81 +450,8 @@ void adreno_recover(struct msm_gpu *gpu)
 	}
 }
 
-void adreno_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-		struct msm_file_private *ctx)
-{
-	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	struct msm_drm_private *priv = gpu->dev->dev_private;
-	struct msm_ringbuffer *ring = submit->ring;
-	unsigned i;
-
-	for (i = 0; i < submit->nr_cmds; i++) {
-		switch (submit->cmd[i].type) {
-		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
-			/* ignore IB-targets */
-			break;
-		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
-			/* ignore if there has not been a ctx switch: */
-			if (priv->lastctx == ctx)
-				break;
-			fallthrough;
-		case MSM_SUBMIT_CMD_BUF:
-			OUT_PKT3(ring, adreno_is_a4xx(adreno_gpu) ?
-				CP_INDIRECT_BUFFER_PFE : CP_INDIRECT_BUFFER_PFD, 2);
-			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
-			OUT_RING(ring, submit->cmd[i].size);
-			OUT_PKT2(ring);
-			break;
-		}
-	}
-
-	OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1);
-	OUT_RING(ring, submit->seqno);
-
-	if (adreno_is_a3xx(adreno_gpu) || adreno_is_a4xx(adreno_gpu)) {
-		/* Flush HLSQ lazy updates to make sure there is nothing
-		 * pending for indirect loads after the timestamp has
-		 * passed:
-		 */
-		OUT_PKT3(ring, CP_EVENT_WRITE, 1);
-		OUT_RING(ring, HLSQ_FLUSH);
-	}
-
-	/* wait for idle before cache flush/interrupt */
-	OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1);
-	OUT_RING(ring, 0x00000000);
-
-	if (!adreno_is_a2xx(adreno_gpu)) {
-		/* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */
-		OUT_PKT3(ring, CP_EVENT_WRITE, 3);
-		OUT_RING(ring, CACHE_FLUSH_TS | BIT(31));
-		OUT_RING(ring, rbmemptr(ring, fence));
-		OUT_RING(ring, submit->seqno);
-	} else {
-		/* BIT(31) means something else on a2xx */
-		OUT_PKT3(ring, CP_EVENT_WRITE, 3);
-		OUT_RING(ring, CACHE_FLUSH_TS);
-		OUT_RING(ring, rbmemptr(ring, fence));
-		OUT_RING(ring, submit->seqno);
-		OUT_PKT3(ring, CP_INTERRUPT, 1);
-		OUT_RING(ring, 0x80000000);
-	}
-
-#if 0
-	if (adreno_is_a3xx(adreno_gpu)) {
-		/* Dummy set-constant to trigger context rollover */
-		OUT_PKT3(ring, CP_SET_CONSTANT, 2);
-		OUT_RING(ring, CP_REG(REG_A3XX_HLSQ_CL_KERNEL_GROUP_X_REG));
-		OUT_RING(ring, 0x00000000);
-	}
-#endif
-
-	gpu->funcs->flush(gpu, ring);
-}
-
-void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, u32 reg)
 {
-	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	uint32_t wptr;
 
 	/* Copy the shadow to the actual register */
@@ -524,7 +467,7 @@ void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	/* ensure writes to ringbuffer have hit system memory: */
 	mb();
 
-	adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_WPTR, wptr);
+	gpu_write(gpu, reg, wptr);
 }
 
 bool adreno_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index e55abae365b5..c3775f79525a 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -17,29 +17,8 @@
 #include "adreno_common.xml.h"
 #include "adreno_pm4.xml.h"
 
-#define REG_ADRENO_DEFINE(_offset, _reg) [_offset] = (_reg) + 1
-#define REG_SKIP ~0
-#define REG_ADRENO_SKIP(_offset) [_offset] = REG_SKIP
-
 extern bool snapshot_debugbus;
 
-/**
- * adreno_regs: List of registers that are used in across all
- * 3D devices. Each device type has different offset value for the same
- * register, so an array of register offsets are declared for every device
- * and are indexed by the enumeration values defined in this enum
- */
-enum adreno_regs {
-	REG_ADRENO_CP_RB_BASE,
-	REG_ADRENO_CP_RB_BASE_HI,
-	REG_ADRENO_CP_RB_RPTR_ADDR,
-	REG_ADRENO_CP_RB_RPTR_ADDR_HI,
-	REG_ADRENO_CP_RB_RPTR,
-	REG_ADRENO_CP_RB_WPTR,
-	REG_ADRENO_CP_RB_CNTL,
-	REG_ADRENO_REGISTER_MAX,
-};
-
 enum {
 	ADRENO_FW_PM4 = 0,
 	ADRENO_FW_SQE = 0, /* a6xx */
@@ -176,11 +155,6 @@ static inline bool adreno_is_a225(struct adreno_gpu *gpu)
 	return gpu->revn == 225;
 }
 
-static inline bool adreno_is_a3xx(struct adreno_gpu *gpu)
-{
-	return (gpu->revn >= 300) && (gpu->revn < 400);
-}
-
 static inline bool adreno_is_a305(struct adreno_gpu *gpu)
 {
 	return gpu->revn == 305;
@@ -207,11 +181,6 @@ static inline bool adreno_is_a330v2(struct adreno_gpu *gpu)
 	return adreno_is_a330(gpu) && (gpu->rev.patchid > 0);
 }
 
-static inline bool adreno_is_a4xx(struct adreno_gpu *gpu)
-{
-	return (gpu->revn >= 400) && (gpu->revn < 500);
-}
-
 static inline int adreno_is_a405(struct adreno_gpu *gpu)
 {
 	return gpu->revn == 405;
@@ -269,9 +238,7 @@ struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu,
 		const struct firmware *fw, u64 *iova);
 int adreno_hw_init(struct msm_gpu *gpu);
 void adreno_recover(struct msm_gpu *gpu);
-void adreno_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-		struct msm_file_private *ctx);
-void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
+void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, u32 reg);
 bool adreno_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
 #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP)
 void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state,
@@ -365,59 +332,12 @@ OUT_PKT7(struct msm_ringbuffer *ring, uint8_t opcode, uint16_t cnt)
 		((opcode & 0x7F) << 16) | (PM4_PARITY(opcode) << 23));
 }
 
-/*
- * adreno_reg_check() - Checks the validity of a register enum
- * @gpu:		Pointer to struct adreno_gpu
- * @offset_name:	The register enum that is checked
- */
-static inline bool adreno_reg_check(struct adreno_gpu *gpu,
-		enum adreno_regs offset_name)
-{
-	BUG_ON(offset_name >= REG_ADRENO_REGISTER_MAX || !gpu->reg_offsets[offset_name]);
-
-	/*
-	 * REG_SKIP is a special value that tell us that the register in
-	 * question isn't implemented on target but don't trigger a BUG(). This
-	 * is used to cleanly implement adreno_gpu_write64() and
-	 * adreno_gpu_read64() in a generic fashion
-	 */
-	if (gpu->reg_offsets[offset_name] == REG_SKIP)
-		return false;
-
-	return true;
-}
-
-static inline u32 adreno_gpu_read(struct adreno_gpu *gpu,
-		enum adreno_regs offset_name)
-{
-	u32 reg = gpu->reg_offsets[offset_name];
-	u32 val = 0;
-	if(adreno_reg_check(gpu,offset_name))
-		val = gpu_read(&gpu->base, reg - 1);
-	return val;
-}
-
-static inline void adreno_gpu_write(struct adreno_gpu *gpu,
-		enum adreno_regs offset_name, u32 data)
-{
-	u32 reg = gpu->reg_offsets[offset_name];
-	if(adreno_reg_check(gpu, offset_name))
-		gpu_write(&gpu->base, reg - 1, data);
-}
-
 struct msm_gpu *a2xx_gpu_init(struct drm_device *dev);
 struct msm_gpu *a3xx_gpu_init(struct drm_device *dev);
 struct msm_gpu *a4xx_gpu_init(struct drm_device *dev);
 struct msm_gpu *a5xx_gpu_init(struct drm_device *dev);
 struct msm_gpu *a6xx_gpu_init(struct drm_device *dev);
 
-static inline void adreno_gpu_write64(struct adreno_gpu *gpu,
-		enum adreno_regs lo, enum adreno_regs hi, u64 data)
-{
-	adreno_gpu_write(gpu, lo, lower_32_bits(data));
-	adreno_gpu_write(gpu, hi, upper_32_bits(data));
-}
-
 static inline uint32_t get_wptr(struct msm_ringbuffer *ring)
 {
 	return (ring->cur - ring->start) % (MSM_GPU_RINGBUFFER_SZ >> 2);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
index 3931eecadaff..59bb8c1ffce6 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
@@ -298,6 +298,7 @@ enum adreno_pm4_type3_packets {
 	CP_SET_BIN_DATA5_OFFSET = 46,
 	CP_SET_CTXSWITCH_IB = 85,
 	CP_REG_WRITE = 109,
+	CP_WHERE_AM_I = 98,
 };
 
 enum adreno_state_block {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c
index f1bc6a1af7a7..84ea09d9692f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c
@@ -288,19 +288,6 @@ static void dpu_disable_all_irqs(struct dpu_kms *dpu_kms)
 }
 
 #ifdef CONFIG_DEBUG_FS
-#define DEFINE_DPU_DEBUGFS_SEQ_FOPS(__prefix)				\
-static int __prefix ## _open(struct inode *inode, struct file *file)	\
-{									\
-	return single_open(file, __prefix ## _show, inode->i_private);	\
-}									\
-static const struct file_operations __prefix ## _fops = {		\
-	.owner = THIS_MODULE,						\
-	.open = __prefix ## _open,					\
-	.release = single_release,					\
-	.read = seq_read,						\
-	.llseek = seq_lseek,						\
-}
-
 static int dpu_debugfs_core_irq_show(struct seq_file *s, void *v)
 {
 	struct dpu_irq *irq_obj = s->private;
@@ -328,7 +315,7 @@ static int dpu_debugfs_core_irq_show(struct seq_file *s, void *v)
 	return 0;
 }
 
-DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_debugfs_core_irq);
+DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_core_irq);
 
 void dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms,
 		struct dentry *parent)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
index b36919d95362..393858ef8a83 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
@@ -30,6 +30,74 @@ enum dpu_perf_mode {
 	DPU_PERF_MODE_MAX
 };
 
+/**
+ * @_dpu_core_perf_calc_bw() - to calculate BW per crtc
+ * @kms -  pointer to the dpu_kms
+ * @crtc - pointer to a crtc
+ * Return: returns aggregated BW for all planes in crtc.
+ */
+static u64 _dpu_core_perf_calc_bw(struct dpu_kms *kms,
+		struct drm_crtc *crtc)
+{
+	struct drm_plane *plane;
+	struct dpu_plane_state *pstate;
+	u64 crtc_plane_bw = 0;
+	u32 bw_factor;
+
+	drm_atomic_crtc_for_each_plane(plane, crtc) {
+		pstate = to_dpu_plane_state(plane->state);
+		if (!pstate)
+			continue;
+
+		crtc_plane_bw += pstate->plane_fetch_bw;
+	}
+
+	bw_factor = kms->catalog->perf.bw_inefficiency_factor;
+	if (bw_factor) {
+		crtc_plane_bw *= bw_factor;
+		do_div(crtc_plane_bw, 100);
+	}
+
+	return crtc_plane_bw;
+}
+
+/**
+ * _dpu_core_perf_calc_clk() - to calculate clock per crtc
+ * @kms -  pointer to the dpu_kms
+ * @crtc - pointer to a crtc
+ * @state - pointer to a crtc state
+ * Return: returns max clk for all planes in crtc.
+ */
+static u64 _dpu_core_perf_calc_clk(struct dpu_kms *kms,
+		struct drm_crtc *crtc, struct drm_crtc_state *state)
+{
+	struct drm_plane *plane;
+	struct dpu_plane_state *pstate;
+	struct drm_display_mode *mode;
+	u64 crtc_clk;
+	u32 clk_factor;
+
+	mode = &state->adjusted_mode;
+
+	crtc_clk = mode->vtotal * mode->hdisplay * drm_mode_vrefresh(mode);
+
+	drm_atomic_crtc_for_each_plane(plane, crtc) {
+		pstate = to_dpu_plane_state(plane->state);
+		if (!pstate)
+			continue;
+
+		crtc_clk = max(pstate->plane_clk, crtc_clk);
+	}
+
+	clk_factor = kms->catalog->perf.clk_inefficiency_factor;
+	if (clk_factor) {
+		crtc_clk *= clk_factor;
+		do_div(crtc_clk, 100);
+	}
+
+	return crtc_clk;
+}
+
 static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc)
 {
 	struct msm_drm_private *priv;
@@ -52,12 +120,7 @@ static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms,
 	dpu_cstate = to_dpu_crtc_state(state);
 	memset(perf, 0, sizeof(struct dpu_core_perf_params));
 
-	if (!dpu_cstate->bw_control) {
-		perf->bw_ctl = kms->catalog->perf.max_bw_high *
-					1000ULL;
-		perf->max_per_pipe_ib = perf->bw_ctl;
-		perf->core_clk_rate = kms->perf.max_core_clk_rate;
-	} else if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) {
+	if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) {
 		perf->bw_ctl = 0;
 		perf->max_per_pipe_ib = 0;
 		perf->core_clk_rate = 0;
@@ -65,6 +128,10 @@ static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms,
 		perf->bw_ctl = kms->perf.fix_core_ab_vote;
 		perf->max_per_pipe_ib = kms->perf.fix_core_ib_vote;
 		perf->core_clk_rate = kms->perf.fix_core_clk_rate;
+	} else {
+		perf->bw_ctl = _dpu_core_perf_calc_bw(kms, crtc);
+		perf->max_per_pipe_ib = kms->catalog->perf.min_dram_ib;
+		perf->core_clk_rate = _dpu_core_perf_calc_clk(kms, crtc, state);
 	}
 
 	DPU_DEBUG(
@@ -116,11 +183,7 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc,
 			DPU_DEBUG("crtc:%d bw:%llu ctrl:%d\n",
 				tmp_crtc->base.id, tmp_cstate->new_perf.bw_ctl,
 				tmp_cstate->bw_control);
-			/*
-			 * For bw check only use the bw if the
-			 * atomic property has been already set
-			 */
-			if (tmp_cstate->bw_control)
+
 				bw_sum_of_intfs += tmp_cstate->new_perf.bw_ctl;
 		}
 
@@ -132,9 +195,7 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc,
 
 		DPU_DEBUG("final threshold bw limit = %d\n", threshold);
 
-		if (!dpu_cstate->bw_control) {
-			DPU_DEBUG("bypass bandwidth check\n");
-		} else if (!threshold) {
+		if (!threshold) {
 			DPU_ERROR("no bandwidth limits specified\n");
 			return -E2BIG;
 		} else if (bw > threshold) {
@@ -155,7 +216,11 @@ static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms,
 					= dpu_crtc_get_client_type(crtc);
 	struct drm_crtc *tmp_crtc;
 	struct dpu_crtc_state *dpu_cstate;
-	int ret = 0;
+	int i, ret = 0;
+	u64 avg_bw;
+
+	if (!kms->num_paths)
+		return -EINVAL;
 
 	drm_for_each_crtc(tmp_crtc, crtc->dev) {
 		if (tmp_crtc->enabled &&
@@ -166,10 +231,20 @@ static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms,
 			perf.max_per_pipe_ib = max(perf.max_per_pipe_ib,
 					dpu_cstate->new_perf.max_per_pipe_ib);
 
-			DPU_DEBUG("crtc=%d bw=%llu\n", tmp_crtc->base.id,
-					dpu_cstate->new_perf.bw_ctl);
+			perf.bw_ctl += dpu_cstate->new_perf.bw_ctl;
+
+			DPU_DEBUG("crtc=%d bw=%llu paths:%d\n",
+				  tmp_crtc->base.id,
+				  dpu_cstate->new_perf.bw_ctl, kms->num_paths);
 		}
 	}
+
+	avg_bw = perf.bw_ctl;
+	do_div(avg_bw, (kms->num_paths * 1000)); /*Bps_to_icc*/
+
+	for (i = 0; i < kms->num_paths; i++)
+		icc_set_bw(kms->path[i], avg_bw, perf.max_per_pipe_ib);
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index c2729f71e2fa..f56414a06ec4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -265,11 +265,6 @@ enum dpu_intf_mode dpu_crtc_get_intf_mode(struct drm_crtc *crtc)
 {
 	struct drm_encoder *encoder;
 
-	if (!crtc) {
-		DPU_ERROR("invalid crtc\n");
-		return INTF_MODE_NONE;
-	}
-
 	/*
 	 * TODO: This function is called from dpu debugfs and as part of atomic
 	 * check. When called from debugfs, the crtc->mutex must be held to
@@ -297,7 +292,6 @@ void dpu_crtc_vblank_callback(struct drm_crtc *crtc)
 		dpu_crtc->vblank_cb_time = ktime_get();
 	else
 		dpu_crtc->vblank_cb_count++;
-	_dpu_crtc_complete_flip(crtc);
 	drm_crtc_handle_vblank(crtc);
 	trace_dpu_crtc_vblank_cb(DRMID(crtc));
 }
@@ -402,6 +396,7 @@ static void dpu_crtc_frame_event_cb(void *data, u32 event)
 void dpu_crtc_complete_commit(struct drm_crtc *crtc)
 {
 	trace_dpu_crtc_complete_commit(DRMID(crtc));
+	_dpu_crtc_complete_flip(crtc);
 }
 
 static void _dpu_crtc_setup_lm_bounds(struct drm_crtc *crtc,
@@ -421,8 +416,6 @@ static void _dpu_crtc_setup_lm_bounds(struct drm_crtc *crtc,
 
 		trace_dpu_crtc_setup_lm_bounds(DRMID(crtc), i, r);
 	}
-
-	drm_mode_debug_printmodeline(adj_mode);
 }
 
 static void _dpu_crtc_get_pcc_coeff(struct drm_crtc_state *state,
@@ -457,7 +450,6 @@ static void _dpu_crtc_setup_cp_blocks(struct drm_crtc *crtc)
 	struct dpu_crtc_mixer *mixer = cstate->mixers;
 	struct dpu_hw_pcc_cfg cfg;
 	struct dpu_hw_ctl *ctl;
-	struct dpu_hw_mixer *lm;
 	struct dpu_hw_dspp *dspp;
 	int i;
 
@@ -467,7 +459,6 @@ static void _dpu_crtc_setup_cp_blocks(struct drm_crtc *crtc)
 
 	for (i = 0; i < cstate->num_mixers; i++) {
 		ctl = mixer[i].lm_ctl;
-		lm = mixer[i].hw_lm;
 		dspp = mixer[i].hw_dspp;
 
 		if (!dspp || !dspp->ops.setup_pcc)
@@ -496,16 +487,8 @@ static void _dpu_crtc_setup_cp_blocks(struct drm_crtc *crtc)
 static void dpu_crtc_atomic_begin(struct drm_crtc *crtc,
 		struct drm_crtc_state *old_state)
 {
-	struct dpu_crtc *dpu_crtc;
-	struct dpu_crtc_state *cstate;
+	struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state);
 	struct drm_encoder *encoder;
-	struct drm_device *dev;
-	unsigned long flags;
-
-	if (!crtc) {
-		DPU_ERROR("invalid crtc\n");
-		return;
-	}
 
 	if (!crtc->state->enable) {
 		DPU_DEBUG("crtc%d -> enable %d, skip atomic_begin\n",
@@ -515,21 +498,8 @@ static void dpu_crtc_atomic_begin(struct drm_crtc *crtc,
 
 	DPU_DEBUG("crtc%d\n", crtc->base.id);
 
-	dpu_crtc = to_dpu_crtc(crtc);
-	cstate = to_dpu_crtc_state(crtc->state);
-	dev = crtc->dev;
-
 	_dpu_crtc_setup_lm_bounds(crtc, crtc->state);
 
-	if (dpu_crtc->event) {
-		WARN_ON(dpu_crtc->event);
-	} else {
-		spin_lock_irqsave(&dev->event_lock, flags);
-		dpu_crtc->event = crtc->state->event;
-		crtc->state->event = NULL;
-		spin_unlock_irqrestore(&dev->event_lock, flags);
-	}
-
 	/* encoder will trigger pending mask now */
 	drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask)
 		dpu_encoder_trigger_kickoff_pending(encoder);
@@ -583,14 +553,11 @@ static void dpu_crtc_atomic_flush(struct drm_crtc *crtc,
 		return;
 	}
 
-	if (dpu_crtc->event) {
-		DPU_DEBUG("already received dpu_crtc->event\n");
-	} else {
-		spin_lock_irqsave(&dev->event_lock, flags);
-		dpu_crtc->event = crtc->state->event;
-		crtc->state->event = NULL;
-		spin_unlock_irqrestore(&dev->event_lock, flags);
-	}
+	WARN_ON(dpu_crtc->event);
+	spin_lock_irqsave(&dev->event_lock, flags);
+	dpu_crtc->event = crtc->state->event;
+	crtc->state->event = NULL;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
 
 	/*
 	 * If no mixers has been allocated in dpu_crtc_atomic_check(),
@@ -635,14 +602,7 @@ static void dpu_crtc_atomic_flush(struct drm_crtc *crtc,
 static void dpu_crtc_destroy_state(struct drm_crtc *crtc,
 		struct drm_crtc_state *state)
 {
-	struct dpu_crtc_state *cstate;
-
-	if (!crtc || !state) {
-		DPU_ERROR("invalid argument(s)\n");
-		return;
-	}
-
-	cstate = to_dpu_crtc_state(state);
+	struct dpu_crtc_state *cstate = to_dpu_crtc_state(state);
 
 	DPU_DEBUG("crtc%d\n", crtc->base.id);
 
@@ -731,14 +691,8 @@ static void dpu_crtc_reset(struct drm_crtc *crtc)
  */
 static struct drm_crtc_state *dpu_crtc_duplicate_state(struct drm_crtc *crtc)
 {
-	struct dpu_crtc_state *cstate, *old_cstate;
+	struct dpu_crtc_state *cstate, *old_cstate = to_dpu_crtc_state(crtc->state);
 
-	if (!crtc || !crtc->state) {
-		DPU_ERROR("invalid argument(s)\n");
-		return NULL;
-	}
-
-	old_cstate = to_dpu_crtc_state(crtc->state);
 	cstate = kmemdup(old_cstate, sizeof(*old_cstate), GFP_KERNEL);
 	if (!cstate) {
 		DPU_ERROR("failed to allocate state\n");
@@ -754,19 +708,12 @@ static struct drm_crtc_state *dpu_crtc_duplicate_state(struct drm_crtc *crtc)
 static void dpu_crtc_disable(struct drm_crtc *crtc,
 			     struct drm_crtc_state *old_crtc_state)
 {
-	struct dpu_crtc *dpu_crtc;
-	struct dpu_crtc_state *cstate;
+	struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc);
+	struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state);
 	struct drm_encoder *encoder;
 	unsigned long flags;
 	bool release_bandwidth = false;
 
-	if (!crtc || !crtc->state) {
-		DPU_ERROR("invalid crtc\n");
-		return;
-	}
-	dpu_crtc = to_dpu_crtc(crtc);
-	cstate = to_dpu_crtc_state(crtc->state);
-
 	DRM_DEBUG_KMS("crtc%d\n", crtc->base.id);
 
 	/* Disable/save vblank irq handling */
@@ -825,19 +772,13 @@ static void dpu_crtc_disable(struct drm_crtc *crtc,
 static void dpu_crtc_enable(struct drm_crtc *crtc,
 		struct drm_crtc_state *old_crtc_state)
 {
-	struct dpu_crtc *dpu_crtc;
+	struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc);
 	struct drm_encoder *encoder;
 	bool request_bandwidth = false;
 
-	if (!crtc) {
-		DPU_ERROR("invalid crtc\n");
-		return;
-	}
-
 	pm_runtime_get_sync(crtc->dev->dev);
 
 	DRM_DEBUG_KMS("crtc%d\n", crtc->base.id);
-	dpu_crtc = to_dpu_crtc(crtc);
 
 	drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) {
 		/* in video mode, we hold an extra bandwidth reference
@@ -873,15 +814,15 @@ struct plane_state {
 static int dpu_crtc_atomic_check(struct drm_crtc *crtc,
 		struct drm_crtc_state *state)
 {
-	struct dpu_crtc *dpu_crtc;
+	struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc);
+	struct dpu_crtc_state *cstate = to_dpu_crtc_state(state);
 	struct plane_state *pstates;
-	struct dpu_crtc_state *cstate;
 
 	const struct drm_plane_state *pstate;
 	struct drm_plane *plane;
 	struct drm_display_mode *mode;
 
-	int cnt = 0, rc = 0, mixer_width, i, z_pos;
+	int cnt = 0, rc = 0, mixer_width = 0, i, z_pos;
 
 	struct dpu_multirect_plane_states multirect_plane[DPU_STAGE_MAX * 2];
 	int multirect_count = 0;
@@ -889,16 +830,8 @@ static int dpu_crtc_atomic_check(struct drm_crtc *crtc,
 	int left_zpos_cnt = 0, right_zpos_cnt = 0;
 	struct drm_rect crtc_rect = { 0 };
 
-	if (!crtc) {
-		DPU_ERROR("invalid crtc\n");
-		return -EINVAL;
-	}
-
 	pstates = kzalloc(sizeof(*pstates) * DPU_STAGE_MAX * 4, GFP_KERNEL);
 
-	dpu_crtc = to_dpu_crtc(crtc);
-	cstate = to_dpu_crtc_state(state);
-
 	if (!state->enable || !state->active) {
 		DPU_DEBUG("crtc%d -> enable %d, active %d, skip atomic_check\n",
 				crtc->base.id, state->enable, state->active);
@@ -914,9 +847,11 @@ static int dpu_crtc_atomic_check(struct drm_crtc *crtc,
 
 	memset(pipe_staged, 0, sizeof(pipe_staged));
 
-	mixer_width = mode->hdisplay / cstate->num_mixers;
+	if (cstate->num_mixers) {
+		mixer_width = mode->hdisplay / cstate->num_mixers;
 
-	_dpu_crtc_setup_lm_bounds(crtc, state);
+		_dpu_crtc_setup_lm_bounds(crtc, state);
+	}
 
 	crtc_rect.x2 = mode->hdisplay;
 	crtc_rect.y2 = mode->vdisplay;
@@ -1242,23 +1177,7 @@ static int _dpu_debugfs_status_show(struct seq_file *s, void *data)
 	return 0;
 }
 
-static int _dpu_debugfs_status_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, _dpu_debugfs_status_show, inode->i_private);
-}
-
-#define DEFINE_DPU_DEBUGFS_SEQ_FOPS(__prefix)                          \
-static int __prefix ## _open(struct inode *inode, struct file *file)	\
-{									\
-	return single_open(file, __prefix ## _show, inode->i_private);	\
-}									\
-static const struct file_operations __prefix ## _fops = {		\
-	.owner = THIS_MODULE,						\
-	.open = __prefix ## _open,					\
-	.release = single_release,					\
-	.read = seq_read,						\
-	.llseek = seq_lseek,						\
-}
+DEFINE_SHOW_ATTRIBUTE(_dpu_debugfs_status);
 
 static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v)
 {
@@ -1275,25 +1194,18 @@ static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v)
 
 	return 0;
 }
-DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_crtc_debugfs_state);
+DEFINE_SHOW_ATTRIBUTE(dpu_crtc_debugfs_state);
 
 static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc)
 {
 	struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc);
 
-	static const struct file_operations debugfs_status_fops = {
-		.open =		_dpu_debugfs_status_open,
-		.read =		seq_read,
-		.llseek =	seq_lseek,
-		.release =	single_release,
-	};
-
 	dpu_crtc->debugfs_root = debugfs_create_dir(dpu_crtc->name,
 			crtc->dev->primary->debugfs_root);
 
 	debugfs_create_file("status", 0400,
 			dpu_crtc->debugfs_root,
-			dpu_crtc, &debugfs_status_fops);
+			dpu_crtc, &_dpu_debugfs_status_fops);
 	debugfs_create_file("state", 0600,
 			dpu_crtc->debugfs_root,
 			&dpu_crtc->base,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index bd6def436c65..f7f5c258b553 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1001,6 +1001,9 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 
 	trace_dpu_enc_mode_set(DRMID(drm_enc));
 
+	if (drm_enc->encoder_type == DRM_MODE_ENCODER_TMDS && priv->dp)
+		msm_dp_display_mode_set(priv->dp, drm_enc, mode, adj_mode);
+
 	list_for_each_entry(conn_iter, connector_list, head)
 		if (conn_iter->encoder == drm_enc)
 			conn = conn_iter;
@@ -1109,6 +1112,13 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc)
 		return;
 	}
 
+
+	if (dpu_enc->disp_info.intf_type == DRM_MODE_CONNECTOR_DisplayPort &&
+		dpu_enc->cur_master->hw_mdptop &&
+		dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select)
+		dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select(
+			dpu_enc->cur_master->hw_mdptop);
+
 	_dpu_encoder_update_vsync_source(dpu_enc, &dpu_enc->disp_info);
 
 	if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_DSI &&
@@ -1146,6 +1156,7 @@ static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc)
 {
 	struct dpu_encoder_virt *dpu_enc = NULL;
 	int ret = 0;
+	struct msm_drm_private *priv;
 	struct drm_display_mode *cur_mode = NULL;
 
 	if (!drm_enc) {
@@ -1156,6 +1167,7 @@ static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc)
 
 	mutex_lock(&dpu_enc->enc_lock);
 	cur_mode = &dpu_enc->base.crtc->state->adjusted_mode;
+	priv = drm_enc->dev->dev_private;
 
 	trace_dpu_enc_enable(DRMID(drm_enc), cur_mode->hdisplay,
 			     cur_mode->vdisplay);
@@ -1176,6 +1188,15 @@ static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc)
 
 	_dpu_encoder_virt_enable_helper(drm_enc);
 
+	if (drm_enc->encoder_type == DRM_MODE_ENCODER_TMDS && priv->dp) {
+		ret = msm_dp_display_enable(priv->dp,
+						drm_enc);
+		if (ret) {
+			DPU_ERROR_ENC(dpu_enc, "dp display enable failed: %d\n",
+				ret);
+			goto out;
+		}
+	}
 	dpu_enc->enabled = true;
 
 out:
@@ -1211,6 +1232,11 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc)
 	/* wait for idle */
 	dpu_encoder_wait_for_event(drm_enc, MSM_ENC_TX_COMPLETE);
 
+	if (drm_enc->encoder_type == DRM_MODE_ENCODER_TMDS && priv->dp) {
+		if (msm_dp_display_pre_disable(priv->dp, drm_enc))
+			DPU_ERROR_ENC(dpu_enc, "dp display push idle failed\n");
+	}
+
 	dpu_encoder_resource_control(drm_enc, DPU_ENC_RC_EVENT_PRE_STOP);
 
 	for (i = 0; i < dpu_enc->num_phys_encs; i++) {
@@ -1220,6 +1246,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc)
 			phys->ops.disable(phys);
 	}
 
+
 	/* after phys waits for frame-done, should be no more frames pending */
 	if (atomic_xchg(&dpu_enc->frame_done_timeout_ms, 0)) {
 		DPU_ERROR("enc%d timeout pending\n", drm_enc->base.id);
@@ -1234,6 +1261,11 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc)
 
 	DPU_DEBUG_ENC(dpu_enc, "encoder disabled\n");
 
+	if (drm_enc->encoder_type == DRM_MODE_ENCODER_TMDS && priv->dp) {
+		if (msm_dp_display_disable(priv->dp, drm_enc))
+			DPU_ERROR_ENC(dpu_enc, "dp display disable failed\n");
+	}
+
 	mutex_unlock(&dpu_enc->enc_lock);
 }
 
@@ -1880,24 +1912,13 @@ static int _dpu_encoder_status_show(struct seq_file *s, void *data)
 	return 0;
 }
 
-static int _dpu_encoder_debugfs_status_open(struct inode *inode,
-		struct file *file)
-{
-	return single_open(file, _dpu_encoder_status_show, inode->i_private);
-}
+DEFINE_SHOW_ATTRIBUTE(_dpu_encoder_status);
 
 static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc)
 {
 	struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc);
 	int i;
 
-	static const struct file_operations debugfs_status_fops = {
-		.open =		_dpu_encoder_debugfs_status_open,
-		.read =		seq_read,
-		.llseek =	seq_lseek,
-		.release =	single_release,
-	};
-
 	char name[DPU_NAME_SIZE];
 
 	if (!drm_enc->dev) {
@@ -1913,7 +1934,7 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc)
 
 	/* don't error check these */
 	debugfs_create_file("status", 0600,
-		dpu_enc->debugfs_root, dpu_enc, &debugfs_status_fops);
+		dpu_enc->debugfs_root, dpu_enc, &_dpu_encoder_status_fops);
 
 	for (i = 0; i < dpu_enc->num_phys_encs; i++)
 		if (dpu_enc->phys_encs[i]->ops.late_register)
@@ -2008,7 +2029,7 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 {
 	int ret = 0;
 	int i = 0;
-	enum dpu_intf_type intf_type;
+	enum dpu_intf_type intf_type = INTF_NONE;
 	struct dpu_enc_phys_init_params phys_params;
 
 	if (!dpu_enc) {
@@ -2030,9 +2051,9 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 	case DRM_MODE_ENCODER_DSI:
 		intf_type = INTF_DSI;
 		break;
-	default:
-		DPU_ERROR_ENC(dpu_enc, "unsupported display interface type\n");
-		return -EINVAL;
+	case DRM_MODE_ENCODER_TMDS:
+		intf_type = INTF_DP;
+		break;
 	}
 
 	WARN_ON(disp_info->num_of_h_tiles < 1);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index b5a49050d131..805e059b50b7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -100,6 +100,14 @@ static void drm_mode_to_intf_timing_params(
 	 * display_v_end -= mode->hsync_start - mode->hdisplay;
 	 * }
 	 */
+	/* for DP/EDP, Shift timings to align it to bottom right */
+	if ((phys_enc->hw_intf->cap->type == INTF_DP) ||
+		(phys_enc->hw_intf->cap->type == INTF_EDP)) {
+		timing->h_back_porch += timing->h_front_porch;
+		timing->h_front_porch = 0;
+		timing->v_back_porch += timing->v_front_porch;
+		timing->v_front_porch = 0;
+	}
 }
 
 static u32 get_horizontal_total(const struct intf_timing_params *timing)
@@ -298,7 +306,6 @@ static void dpu_encoder_phys_vid_vblank_irq(void *arg, int irq_idx)
 	struct dpu_hw_ctl *hw_ctl;
 	unsigned long lock_flags;
 	u32 flush_register = 0;
-	int new_cnt = -1, old_cnt = -1;
 
 	hw_ctl = phys_enc->hw_ctl;
 
@@ -308,7 +315,7 @@ static void dpu_encoder_phys_vid_vblank_irq(void *arg, int irq_idx)
 		phys_enc->parent_ops->handle_vblank_virt(phys_enc->parent,
 				phys_enc);
 
-	old_cnt  = atomic_read(&phys_enc->pending_kickoff_cnt);
+	atomic_read(&phys_enc->pending_kickoff_cnt);
 
 	/*
 	 * only decrement the pending flush count if we've actually flushed
@@ -320,8 +327,7 @@ static void dpu_encoder_phys_vid_vblank_irq(void *arg, int irq_idx)
 		flush_register = hw_ctl->ops.get_flush_register(hw_ctl);
 
 	if (!(flush_register & hw_ctl->ops.get_pending_flush(hw_ctl)))
-		new_cnt = atomic_add_unless(&phys_enc->pending_kickoff_cnt,
-				-1, 0);
+		atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0);
 	spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags);
 
 	/* Signal any waiting atomic commit thread */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 97d122eee96d..60b304b72b7c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -684,7 +684,8 @@ static const struct dpu_perf_cfg sc7180_perf_data = {
 	.max_bw_high = 6800000,
 	.min_core_ib = 2400000,
 	.min_llcc_ib = 800000,
-	.min_dram_ib = 800000,
+	.min_dram_ib = 1600000,
+	.min_prefill_lines = 24,
 	.danger_lut_tbl = {0xff, 0xffff, 0x0},
 	.qos_lut_tbl = {
 		{.nentry = ARRAY_SIZE(sc7180_qos_linear),
@@ -701,6 +702,8 @@ static const struct dpu_perf_cfg sc7180_perf_data = {
 		{.rd_enable = 1, .wr_enable = 1},
 		{.rd_enable = 1, .wr_enable = 0}
 	},
+	.clk_inefficiency_factor = 105,
+	.bw_inefficiency_factor = 120,
 };
 
 static const struct dpu_perf_cfg sm8150_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 1b7a9213a756..3544af1a45c5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -659,6 +659,8 @@ struct dpu_perf_cdp_cfg {
  * @downscaling_prefill_lines  downscaling latency in lines
  * @amortizable_theshold minimum y position for traffic shaping prefill
  * @min_prefill_lines  minimum pipeline latency in lines
+ * @clk_inefficiency_factor DPU src clock inefficiency factor
+ * @bw_inefficiency_factor DPU axi bus bw inefficiency factor
  * @safe_lut_tbl: LUT tables for safe signals
  * @danger_lut_tbl: LUT tables for danger signals
  * @qos_lut_tbl: LUT tables for QoS signals
@@ -683,6 +685,8 @@ struct dpu_perf_cfg {
 	u32 downscaling_prefill_lines;
 	u32 amortizable_threshold;
 	u32 min_prefill_lines;
+	u32 clk_inefficiency_factor;
+	u32 bw_inefficiency_factor;
 	u32 safe_lut_tbl[DPU_QOS_LUT_USAGE_MAX];
 	u32 danger_lut_tbl[DPU_QOS_LUT_USAGE_MAX];
 	struct dpu_qos_lut_tbl qos_lut_tbl[DPU_QOS_LUT_USAGE_MAX];
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index c0a4d4e16d82..d93c44f6996d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -85,30 +85,17 @@ static int _dpu_danger_signal_status(struct seq_file *s,
 	return 0;
 }
 
-#define DEFINE_DPU_DEBUGFS_SEQ_FOPS(__prefix)				\
-static int __prefix ## _open(struct inode *inode, struct file *file)	\
-{									\
-	return single_open(file, __prefix ## _show, inode->i_private);	\
-}									\
-static const struct file_operations __prefix ## _fops = {		\
-	.owner = THIS_MODULE,						\
-	.open = __prefix ## _open,					\
-	.release = single_release,					\
-	.read = seq_read,						\
-	.llseek = seq_lseek,						\
-}
-
 static int dpu_debugfs_danger_stats_show(struct seq_file *s, void *v)
 {
 	return _dpu_danger_signal_status(s, true);
 }
-DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_debugfs_danger_stats);
+DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_danger_stats);
 
 static int dpu_debugfs_safe_stats_show(struct seq_file *s, void *v)
 {
 	return _dpu_danger_signal_status(s, false);
 }
-DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_debugfs_safe_stats);
+DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_safe_stats);
 
 static void dpu_debugfs_danger_init(struct dpu_kms *dpu_kms,
 		struct dentry *parent)
@@ -195,10 +182,15 @@ static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor)
 	struct dpu_kms *dpu_kms = to_dpu_kms(kms);
 	void *p = dpu_hw_util_get_log_mask_ptr();
 	struct dentry *entry;
+	struct drm_device *dev;
+	struct msm_drm_private *priv;
 
 	if (!p)
 		return -EINVAL;
 
+	dev = dpu_kms->dev;
+	priv = dev->dev_private;
+
 	entry = debugfs_create_dir("debug", minor->debugfs_root);
 
 	debugfs_create_x32(DPU_DEBUGFS_HWMASKNAME, 0600, entry, p);
@@ -207,6 +199,9 @@ static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor)
 	dpu_debugfs_vbif_init(dpu_kms, entry);
 	dpu_debugfs_core_irq_init(dpu_kms, entry);
 
+	if (priv->dp)
+		msm_dp_debugfs_init(priv->dp, minor);
+
 	return dpu_core_perf_debugfs_init(dpu_kms, entry);
 }
 #endif
@@ -290,6 +285,28 @@ static int dpu_kms_global_obj_init(struct dpu_kms *dpu_kms)
 	return 0;
 }
 
+static int dpu_kms_parse_data_bus_icc_path(struct dpu_kms *dpu_kms)
+{
+	struct icc_path *path0;
+	struct icc_path *path1;
+	struct drm_device *dev = dpu_kms->dev;
+
+	path0 = of_icc_get(dev->dev, "mdp0-mem");
+	path1 = of_icc_get(dev->dev, "mdp1-mem");
+
+	if (IS_ERR_OR_NULL(path0))
+		return PTR_ERR_OR_ZERO(path0);
+
+	dpu_kms->path[0] = path0;
+	dpu_kms->num_paths = 1;
+
+	if (!IS_ERR_OR_NULL(path1)) {
+		dpu_kms->path[1] = path1;
+		dpu_kms->num_paths++;
+	}
+	return 0;
+}
+
 static int dpu_kms_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc)
 {
 	return dpu_crtc_vblank(crtc, true);
@@ -479,6 +496,33 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 	return rc;
 }
 
+static int _dpu_kms_initialize_displayport(struct drm_device *dev,
+					    struct msm_drm_private *priv,
+					    struct dpu_kms *dpu_kms)
+{
+	struct drm_encoder *encoder = NULL;
+	int rc = 0;
+
+	if (!priv->dp)
+		return rc;
+
+	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS);
+	if (IS_ERR(encoder)) {
+		DPU_ERROR("encoder init failed for dsi display\n");
+		return PTR_ERR(encoder);
+	}
+
+	rc = msm_dp_modeset_init(priv->dp, dev, encoder);
+	if (rc) {
+		DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc);
+		drm_encoder_cleanup(encoder);
+		return rc;
+	}
+
+	priv->encoders[priv->num_encoders++] = encoder;
+	return rc;
+}
+
 /**
  * _dpu_kms_setup_displays - create encoders, bridges and connectors
  *                           for underlying displays
@@ -491,12 +535,21 @@ static int _dpu_kms_setup_displays(struct drm_device *dev,
 				    struct msm_drm_private *priv,
 				    struct dpu_kms *dpu_kms)
 {
-	/**
-	 * Extend this function to initialize other
-	 * types of displays
-	 */
+	int rc = 0;
 
-	return _dpu_kms_initialize_dsi(dev, priv, dpu_kms);
+	rc = _dpu_kms_initialize_dsi(dev, priv, dpu_kms);
+	if (rc) {
+		DPU_ERROR("initialize_dsi failed, rc = %d\n", rc);
+		return rc;
+	}
+
+	rc = _dpu_kms_initialize_displayport(dev, priv, dpu_kms);
+	if (rc) {
+		DPU_ERROR("initialize_DP failed, rc = %d\n", rc);
+		return rc;
+	}
+
+	return rc;
 }
 
 static void _dpu_kms_drm_obj_destroy(struct dpu_kms *dpu_kms)
@@ -681,13 +734,20 @@ static void _dpu_kms_set_encoder_mode(struct msm_kms *kms,
 	info.capabilities = cmd_mode ? MSM_DISPLAY_CAP_CMD_MODE :
 			MSM_DISPLAY_CAP_VID_MODE;
 
-	/* TODO: No support for DSI swap */
-	for (i = 0; i < ARRAY_SIZE(priv->dsi); i++) {
-		if (priv->dsi[i]) {
-			info.h_tile_instance[info.num_of_h_tiles] = i;
-			info.num_of_h_tiles++;
+	switch (info.intf_type) {
+	case DRM_MODE_ENCODER_DSI:
+		/* TODO: No support for DSI swap */
+		for (i = 0; i < ARRAY_SIZE(priv->dsi); i++) {
+			if (priv->dsi[i]) {
+				info.h_tile_instance[info.num_of_h_tiles] = i;
+				info.num_of_h_tiles++;
+			}
 		}
-	}
+		break;
+	case DRM_MODE_ENCODER_TMDS:
+		info.num_of_h_tiles = 1;
+		break;
+	};
 
 	rc = dpu_encoder_setup(encoder->dev, encoder, &info);
 	if (rc)
@@ -709,6 +769,23 @@ static void dpu_irq_preinstall(struct msm_kms *kms)
 	dpu_core_irq_preinstall(dpu_kms);
 }
 
+static int dpu_irq_postinstall(struct msm_kms *kms)
+{
+	struct msm_drm_private *priv;
+	struct dpu_kms *dpu_kms = to_dpu_kms(kms);
+
+	if (!dpu_kms || !dpu_kms->dev)
+		return -EINVAL;
+
+	priv = dpu_kms->dev->dev_private;
+	if (!priv)
+		return -EINVAL;
+
+	msm_dp_irq_postinstall(priv->dp);
+
+	return 0;
+}
+
 static void dpu_irq_uninstall(struct msm_kms *kms)
 {
 	struct dpu_kms *dpu_kms = to_dpu_kms(kms);
@@ -719,6 +796,7 @@ static void dpu_irq_uninstall(struct msm_kms *kms)
 static const struct msm_kms_funcs kms_funcs = {
 	.hw_init         = dpu_kms_hw_init,
 	.irq_preinstall  = dpu_irq_preinstall,
+	.irq_postinstall = dpu_irq_postinstall,
 	.irq_uninstall   = dpu_irq_uninstall,
 	.irq             = dpu_irq,
 	.enable_commit   = dpu_kms_enable_commit,
@@ -952,6 +1030,9 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 
 	dpu_vbif_init_memtypes(dpu_kms);
 
+	if (of_device_is_compatible(dev->dev->of_node, "qcom,sc7180-mdss"))
+		dpu_kms_parse_data_bus_icc_path(dpu_kms);
+
 	pm_runtime_put_sync(&dpu_kms->pdev->dev);
 
 	return 0;
@@ -1079,7 +1160,7 @@ static int dpu_dev_remove(struct platform_device *pdev)
 
 static int __maybe_unused dpu_runtime_suspend(struct device *dev)
 {
-	int rc = -1;
+	int i, rc = -1;
 	struct platform_device *pdev = to_platform_device(dev);
 	struct dpu_kms *dpu_kms = platform_get_drvdata(pdev);
 	struct dss_module_power *mp = &dpu_kms->mp;
@@ -1090,6 +1171,9 @@ static int __maybe_unused dpu_runtime_suspend(struct device *dev)
 	if (rc)
 		DPU_ERROR("clock disable failed rc:%d\n", rc);
 
+	for (i = 0; i < dpu_kms->num_paths; i++)
+		icc_set_bw(dpu_kms->path[i], 0, 0);
+
 	return rc;
 }
 
@@ -1101,8 +1185,15 @@ static int __maybe_unused dpu_runtime_resume(struct device *dev)
 	struct drm_encoder *encoder;
 	struct drm_device *ddev;
 	struct dss_module_power *mp = &dpu_kms->mp;
+	int i;
 
 	ddev = dpu_kms->dev;
+
+	/* Min vote of BW is required before turning on AXI clk */
+	for (i = 0; i < dpu_kms->num_paths; i++)
+		icc_set_bw(dpu_kms->path[i], 0,
+			dpu_kms->catalog->perf.min_dram_ib);
+
 	rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, true);
 	if (rc) {
 		DPU_ERROR("clock enable failed rc:%d\n", rc);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index e140cd633071..1c0e4c0c9ffb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -8,6 +8,8 @@
 #ifndef __DPU_KMS_H__
 #define __DPU_KMS_H__
 
+#include <linux/interconnect.h>
+
 #include <drm/drm_drv.h>
 
 #include "msm_drv.h"
@@ -140,6 +142,8 @@ struct dpu_kms {
 	 * when disabled.
 	 */
 	atomic_t bandwidth_ref;
+	struct icc_path *path[2];
+	u32 num_paths;
 };
 
 struct vsync_info {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
index 7d3fdbb00e7e..cd4078807db1 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
@@ -8,7 +8,6 @@
 #include <linux/irqdesc.h>
 #include <linux/irqchip/chained_irq.h>
 #include "dpu_kms.h"
-#include <linux/interconnect.h>
 
 #define to_dpu_mdss(x) container_of(x, struct dpu_mdss, base)
 
@@ -277,9 +276,11 @@ int dpu_mdss_init(struct drm_device *dev)
 
 	DRM_DEBUG("mapped mdss address space @%pK\n", dpu_mdss->mmio);
 
-	ret = dpu_mdss_parse_data_bus_icc_path(dev, dpu_mdss);
-	if (ret)
-		return ret;
+	if (!of_device_is_compatible(dev->dev->of_node, "qcom,sc7180-mdss")) {
+		ret = dpu_mdss_parse_data_bus_icc_path(dev, dpu_mdss);
+		if (ret)
+			return ret;
+	}
 
 	mp = &dpu_mdss->mp;
 	ret = msm_dss_parse_clock(pdev, mp);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 29e373d2e7b5..7ea90d25a3b6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -132,6 +132,86 @@ static struct dpu_kms *_dpu_plane_get_kms(struct drm_plane *plane)
 }
 
 /**
+ * _dpu_plane_calc_bw - calculate bandwidth required for a plane
+ * @Plane: Pointer to drm plane.
+ * Result: Updates calculated bandwidth in the plane state.
+ * BW Equation: src_w * src_h * bpp * fps * (v_total / v_dest)
+ * Prefill BW Equation: line src bytes * line_time
+ */
+static void _dpu_plane_calc_bw(struct drm_plane *plane,
+	struct drm_framebuffer *fb)
+{
+	struct dpu_plane *pdpu = to_dpu_plane(plane);
+	struct dpu_plane_state *pstate;
+	struct drm_display_mode *mode;
+	const struct dpu_format *fmt = NULL;
+	struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane);
+	int src_width, src_height, dst_height, fps;
+	u64 plane_prefill_bw;
+	u64 plane_bw;
+	u32 hw_latency_lines;
+	u64 scale_factor;
+	int vbp, vpw;
+
+	pstate = to_dpu_plane_state(plane->state);
+	mode = &plane->state->crtc->mode;
+
+	fmt = dpu_get_dpu_format_ext(fb->format->format, fb->modifier);
+
+	src_width = drm_rect_width(&pdpu->pipe_cfg.src_rect);
+	src_height = drm_rect_height(&pdpu->pipe_cfg.src_rect);
+	dst_height = drm_rect_height(&pdpu->pipe_cfg.dst_rect);
+	fps = drm_mode_vrefresh(mode);
+	vbp = mode->vtotal - mode->vsync_end;
+	vpw = mode->vsync_end - mode->vsync_start;
+	hw_latency_lines =  dpu_kms->catalog->perf.min_prefill_lines;
+	scale_factor = src_height > dst_height ?
+		mult_frac(src_height, 1, dst_height) : 1;
+
+	plane_bw =
+		src_width * mode->vtotal * fps * fmt->bpp *
+		scale_factor;
+
+	plane_prefill_bw =
+		src_width * hw_latency_lines * fps * fmt->bpp *
+		scale_factor * mode->vtotal;
+
+	do_div(plane_prefill_bw, (vbp+vpw));
+
+	pstate->plane_fetch_bw = max(plane_bw, plane_prefill_bw);
+}
+
+/**
+ * _dpu_plane_calc_clk - calculate clock required for a plane
+ * @Plane: Pointer to drm plane.
+ * Result: Updates calculated clock in the plane state.
+ * Clock equation: dst_w * v_total * fps * (src_h / dst_h)
+ */
+static void _dpu_plane_calc_clk(struct drm_plane *plane)
+{
+	struct dpu_plane *pdpu = to_dpu_plane(plane);
+	struct dpu_plane_state *pstate;
+	struct drm_display_mode *mode;
+	int dst_width, src_height, dst_height, fps;
+
+	pstate = to_dpu_plane_state(plane->state);
+	mode = &plane->state->crtc->mode;
+
+	src_height = drm_rect_height(&pdpu->pipe_cfg.src_rect);
+	dst_width = drm_rect_width(&pdpu->pipe_cfg.dst_rect);
+	dst_height = drm_rect_height(&pdpu->pipe_cfg.dst_rect);
+	fps = drm_mode_vrefresh(mode);
+
+	pstate->plane_clk =
+		dst_width * mode->vtotal * fps;
+
+	if (src_height > dst_height) {
+		pstate->plane_clk *= src_height;
+		do_div(pstate->plane_clk, dst_height);
+	}
+}
+
+/**
  * _dpu_plane_calc_fill_level - calculate fill level of the given source format
  * @plane:		Pointer to drm plane
  * @fmt:		Pointer to source buffer format
@@ -1102,6 +1182,10 @@ static void dpu_plane_sspp_atomic_update(struct drm_plane *plane)
 	}
 
 	_dpu_plane_set_qos_remap(plane);
+
+	_dpu_plane_calc_bw(plane, fb);
+
+	_dpu_plane_calc_clk(plane);
 }
 
 static void _dpu_plane_atomic_disable(struct drm_plane *plane)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
index 456949713e90..ca83b8753d59 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
@@ -25,6 +25,8 @@
  * @scaler3_cfg: configuration data for scaler3
  * @pixel_ext: configuration data for pixel extensions
  * @cdp_cfg:	CDP configuration
+ * @plane_fetch_bw: calculated BW per plane
+ * @plane_clk: calculated clk per plane
  */
 struct dpu_plane_state {
 	struct drm_plane_state base;
@@ -39,6 +41,8 @@ struct dpu_plane_state {
 	struct dpu_hw_pixel_ext pixel_ext;
 
 	struct dpu_hw_pipe_cdp_cfg cdp_cfg;
+	u64 plane_fetch_bw;
+	u64 plane_clk;
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c
index 5d8956055286..88645dbc3785 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c
@@ -25,54 +25,9 @@ static struct mdp4_kms *get_kms(struct drm_encoder *encoder)
 	return to_mdp4_kms(to_mdp_kms(priv->kms));
 }
 
-#ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-#include <mach/board.h>
-/* not ironically named at all.. no, really.. */
-static void bs_init(struct mdp4_dtv_encoder *mdp4_dtv_encoder)
-{
-	struct drm_device *dev = mdp4_dtv_encoder->base.dev;
-	struct lcdc_platform_data *dtv_pdata = mdp4_find_pdata("dtv.0");
-
-	if (!dtv_pdata) {
-		DRM_DEV_ERROR(dev->dev, "could not find dtv pdata\n");
-		return;
-	}
-
-	if (dtv_pdata->bus_scale_table) {
-		mdp4_dtv_encoder->bsc = msm_bus_scale_register_client(
-				dtv_pdata->bus_scale_table);
-		DBG("bus scale client: %08x", mdp4_dtv_encoder->bsc);
-		DBG("lcdc_power_save: %p", dtv_pdata->lcdc_power_save);
-		if (dtv_pdata->lcdc_power_save)
-			dtv_pdata->lcdc_power_save(1);
-	}
-}
-
-static void bs_fini(struct mdp4_dtv_encoder *mdp4_dtv_encoder)
-{
-	if (mdp4_dtv_encoder->bsc) {
-		msm_bus_scale_unregister_client(mdp4_dtv_encoder->bsc);
-		mdp4_dtv_encoder->bsc = 0;
-	}
-}
-
-static void bs_set(struct mdp4_dtv_encoder *mdp4_dtv_encoder, int idx)
-{
-	if (mdp4_dtv_encoder->bsc) {
-		DBG("set bus scaling: %d", idx);
-		msm_bus_scale_client_update_request(mdp4_dtv_encoder->bsc, idx);
-	}
-}
-#else
-static void bs_init(struct mdp4_dtv_encoder *mdp4_dtv_encoder) {}
-static void bs_fini(struct mdp4_dtv_encoder *mdp4_dtv_encoder) {}
-static void bs_set(struct mdp4_dtv_encoder *mdp4_dtv_encoder, int idx) {}
-#endif
-
 static void mdp4_dtv_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct mdp4_dtv_encoder *mdp4_dtv_encoder = to_mdp4_dtv_encoder(encoder);
-	bs_fini(mdp4_dtv_encoder);
 	drm_encoder_cleanup(encoder);
 	kfree(mdp4_dtv_encoder);
 }
@@ -162,8 +117,6 @@ static void mdp4_dtv_encoder_disable(struct drm_encoder *encoder)
 	clk_disable_unprepare(mdp4_dtv_encoder->hdmi_clk);
 	clk_disable_unprepare(mdp4_dtv_encoder->mdp_clk);
 
-	bs_set(mdp4_dtv_encoder, 0);
-
 	mdp4_dtv_encoder->enabled = false;
 }
 
@@ -185,8 +138,6 @@ static void mdp4_dtv_encoder_enable(struct drm_encoder *encoder)
 			MDP4_DMA_CONFIG_PACK(0x21));
 	mdp4_crtc_set_intf(encoder->crtc, INTF_LCDC_DTV, 1);
 
-	bs_set(mdp4_dtv_encoder, 1);
-
 	DBG("setting mdp_clk=%lu", pc);
 
 	ret = clk_set_rate(mdp4_dtv_encoder->mdp_clk, pc);
@@ -252,8 +203,6 @@ struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev)
 		goto fail;
 	}
 
-	bs_init(mdp4_dtv_encoder);
-
 	return encoder;
 
 fail:
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
index 18933bd81c77..e8ee92ab7956 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
@@ -222,17 +222,4 @@ static inline struct clk *mpd4_lvds_pll_init(struct drm_device *dev)
 }
 #endif
 
-#ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-/* bus scaling data is associated with extra pointless platform devices,
- * "dtv", etc.. this is a bit of a hack, but we need a way for encoders
- * to find their pdata to make the bus-scaling stuff work.
- */
-static inline void *mdp4_find_pdata(const char *devname)
-{
-	struct device *dev;
-	dev = bus_find_device_by_name(&platform_bus_type, NULL, devname);
-	return dev ? dev->platform_data : NULL;
-}
-#endif
-
 #endif /* __MDP4_KMS_H__ */
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
index 871f3514ef69..10eb3e5b218e 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
@@ -30,51 +30,10 @@ static struct mdp4_kms *get_kms(struct drm_encoder *encoder)
 	return to_mdp4_kms(to_mdp_kms(priv->kms));
 }
 
-#ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-#include <mach/board.h>
-static void bs_init(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder)
-{
-	struct drm_device *dev = mdp4_lcdc_encoder->base.dev;
-	struct lcdc_platform_data *lcdc_pdata = mdp4_find_pdata("lvds.0");
-
-	if (!lcdc_pdata) {
-		DRM_DEV_ERROR(dev->dev, "could not find lvds pdata\n");
-		return;
-	}
-
-	if (lcdc_pdata->bus_scale_table) {
-		mdp4_lcdc_encoder->bsc = msm_bus_scale_register_client(
-				lcdc_pdata->bus_scale_table);
-		DBG("lvds : bus scale client: %08x", mdp4_lcdc_encoder->bsc);
-	}
-}
-
-static void bs_fini(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder)
-{
-	if (mdp4_lcdc_encoder->bsc) {
-		msm_bus_scale_unregister_client(mdp4_lcdc_encoder->bsc);
-		mdp4_lcdc_encoder->bsc = 0;
-	}
-}
-
-static void bs_set(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder, int idx)
-{
-	if (mdp4_lcdc_encoder->bsc) {
-		DBG("set bus scaling: %d", idx);
-		msm_bus_scale_client_update_request(mdp4_lcdc_encoder->bsc, idx);
-	}
-}
-#else
-static void bs_init(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder) {}
-static void bs_fini(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder) {}
-static void bs_set(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder, int idx) {}
-#endif
-
 static void mdp4_lcdc_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder =
 			to_mdp4_lcdc_encoder(encoder);
-	bs_fini(mdp4_lcdc_encoder);
 	drm_encoder_cleanup(encoder);
 	kfree(mdp4_lcdc_encoder);
 }
@@ -348,8 +307,6 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder)
 			DRM_DEV_ERROR(dev->dev, "failed to disable regulator: %d\n", ret);
 	}
 
-	bs_set(mdp4_lcdc_encoder, 0);
-
 	mdp4_lcdc_encoder->enabled = false;
 }
 
@@ -382,8 +339,6 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 	mdp4_crtc_set_config(encoder->crtc, config);
 	mdp4_crtc_set_intf(encoder->crtc, INTF_LCDC_DTV, 0);
 
-	bs_set(mdp4_lcdc_encoder, 1);
-
 	for (i = 0; i < ARRAY_SIZE(mdp4_lcdc_encoder->regs); i++) {
 		ret = regulator_enable(mdp4_lcdc_encoder->regs[i]);
 		if (ret)
@@ -480,8 +435,6 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
 	}
 	mdp4_lcdc_encoder->regs[2] = reg;
 
-	bs_init(mdp4_lcdc_encoder);
-
 	return encoder;
 
 fail:
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c
index c7df71e2fafc..7288041dd86a 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c
@@ -50,14 +50,9 @@ static int mdp4_lvds_connector_get_modes(struct drm_connector *connector)
 	struct drm_panel *panel = mdp4_lvds_connector->panel;
 	int ret = 0;
 
-	if (panel) {
-		drm_panel_attach(panel, connector);
-
+	if (panel)
 		ret = drm_panel_get_modes(panel, connector);
 
-		drm_panel_detach(panel);
-	}
-
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
index eeef41fcd4e1..ff2c1d583c79 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
@@ -14,27 +14,6 @@ static struct mdp5_kms *get_kms(struct drm_encoder *encoder)
 	return to_mdp5_kms(to_mdp_kms(priv->kms));
 }
 
-#ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-#include <mach/board.h>
-#include <linux/msm-bus.h>
-#include <linux/msm-bus-board.h>
-
-static void bs_set(struct mdp5_encoder *mdp5_cmd_enc, int idx)
-{
-	if (mdp5_cmd_enc->bsc) {
-		DBG("set bus scaling: %d", idx);
-		/* HACK: scaling down, and then immediately back up
-		 * seems to leave things broken (underflow).. so
-		 * never disable:
-		 */
-		idx = 1;
-		msm_bus_scale_client_update_request(mdp5_cmd_enc->bsc, idx);
-	}
-}
-#else
-static void bs_set(struct mdp5_encoder *mdp5_cmd_enc, int idx) {}
-#endif
-
 #define VSYNC_CLK_RATE 19200000
 static int pingpong_tearcheck_setup(struct drm_encoder *encoder,
 				    struct drm_display_mode *mode)
@@ -146,8 +125,6 @@ void mdp5_cmd_encoder_disable(struct drm_encoder *encoder)
 	mdp5_ctl_set_encoder_state(ctl, pipeline, false);
 	mdp5_ctl_commit(ctl, pipeline, mdp_ctl_flush_mask_encoder(intf), true);
 
-	bs_set(mdp5_cmd_enc, 0);
-
 	mdp5_cmd_enc->enabled = false;
 }
 
@@ -161,7 +138,6 @@ void mdp5_cmd_encoder_enable(struct drm_encoder *encoder)
 	if (WARN_ON(mdp5_cmd_enc->enabled))
 		return;
 
-	bs_set(mdp5_cmd_enc, 1);
 	if (pingpong_tearcheck_enable(encoder))
 		return;
 
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
index f48827283c2b..79d67c495780 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
@@ -16,72 +16,9 @@ static struct mdp5_kms *get_kms(struct drm_encoder *encoder)
 	return to_mdp5_kms(to_mdp_kms(priv->kms));
 }
 
-#ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-#include <mach/board.h>
-#include <mach/msm_bus.h>
-#include <mach/msm_bus_board.h>
-#define MDP_BUS_VECTOR_ENTRY(ab_val, ib_val)		\
-	{						\
-		.src = MSM_BUS_MASTER_MDP_PORT0,	\
-		.dst = MSM_BUS_SLAVE_EBI_CH0,		\
-		.ab = (ab_val),				\
-		.ib = (ib_val),				\
-	}
-
-static struct msm_bus_vectors mdp_bus_vectors[] = {
-	MDP_BUS_VECTOR_ENTRY(0, 0),
-	MDP_BUS_VECTOR_ENTRY(2000000000, 2000000000),
-};
-static struct msm_bus_paths mdp_bus_usecases[] = { {
-		.num_paths = 1,
-		.vectors = &mdp_bus_vectors[0],
-}, {
-		.num_paths = 1,
-		.vectors = &mdp_bus_vectors[1],
-} };
-static struct msm_bus_scale_pdata mdp_bus_scale_table = {
-	.usecase = mdp_bus_usecases,
-	.num_usecases = ARRAY_SIZE(mdp_bus_usecases),
-	.name = "mdss_mdp",
-};
-
-static void bs_init(struct mdp5_encoder *mdp5_encoder)
-{
-	mdp5_encoder->bsc = msm_bus_scale_register_client(
-			&mdp_bus_scale_table);
-	DBG("bus scale client: %08x", mdp5_encoder->bsc);
-}
-
-static void bs_fini(struct mdp5_encoder *mdp5_encoder)
-{
-	if (mdp5_encoder->bsc) {
-		msm_bus_scale_unregister_client(mdp5_encoder->bsc);
-		mdp5_encoder->bsc = 0;
-	}
-}
-
-static void bs_set(struct mdp5_encoder *mdp5_encoder, int idx)
-{
-	if (mdp5_encoder->bsc) {
-		DBG("set bus scaling: %d", idx);
-		/* HACK: scaling down, and then immediately back up
-		 * seems to leave things broken (underflow).. so
-		 * never disable:
-		 */
-		idx = 1;
-		msm_bus_scale_client_update_request(mdp5_encoder->bsc, idx);
-	}
-}
-#else
-static void bs_init(struct mdp5_encoder *mdp5_encoder) {}
-static void bs_fini(struct mdp5_encoder *mdp5_encoder) {}
-static void bs_set(struct mdp5_encoder *mdp5_encoder, int idx) {}
-#endif
-
 static void mdp5_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder);
-	bs_fini(mdp5_encoder);
 	drm_encoder_cleanup(encoder);
 	kfree(mdp5_encoder);
 }
@@ -222,8 +159,6 @@ static void mdp5_vid_encoder_disable(struct drm_encoder *encoder)
 	 */
 	mdp_irq_wait(&mdp5_kms->base, intf2vblank(mixer, intf));
 
-	bs_set(mdp5_encoder, 0);
-
 	mdp5_encoder->enabled = false;
 }
 
@@ -240,7 +175,6 @@ static void mdp5_vid_encoder_enable(struct drm_encoder *encoder)
 	if (WARN_ON(mdp5_encoder->enabled))
 		return;
 
-	bs_set(mdp5_encoder, 1);
 	spin_lock_irqsave(&mdp5_encoder->intf_lock, flags);
 	mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(intfn), 1);
 	spin_unlock_irqrestore(&mdp5_encoder->intf_lock, flags);
@@ -426,8 +360,6 @@ struct drm_encoder *mdp5_encoder_init(struct drm_device *dev,
 
 	drm_encoder_helper_add(encoder, &mdp5_encoder_helper_funcs);
 
-	bs_init(mdp5_encoder);
-
 	return encoder;
 
 fail:
diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c
new file mode 100644
index 000000000000..82a8673ab8da
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_audio.c
@@ -0,0 +1,638 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2016-2020, The Linux Foundation. All rights reserved.
+ */
+
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <linux/of_platform.h>
+
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_edid.h>
+
+#include "dp_catalog.h"
+#include "dp_audio.h"
+#include "dp_panel.h"
+#include "dp_display.h"
+
+#define HEADER_BYTE_2_BIT	 0
+#define PARITY_BYTE_2_BIT	 8
+#define HEADER_BYTE_1_BIT	16
+#define PARITY_BYTE_1_BIT	24
+#define HEADER_BYTE_3_BIT	16
+#define PARITY_BYTE_3_BIT	24
+
+struct dp_audio_private {
+	struct platform_device *audio_pdev;
+	struct platform_device *pdev;
+	struct dp_catalog *catalog;
+	struct dp_panel *panel;
+
+	bool engine_on;
+	u32 channels;
+
+	struct dp_audio dp_audio;
+};
+
+static u8 dp_audio_get_g0_value(u8 data)
+{
+	u8 c[4];
+	u8 g[4];
+	u8 ret_data = 0;
+	u8 i;
+
+	for (i = 0; i < 4; i++)
+		c[i] = (data >> i) & 0x01;
+
+	g[0] = c[3];
+	g[1] = c[0] ^ c[3];
+	g[2] = c[1];
+	g[3] = c[2];
+
+	for (i = 0; i < 4; i++)
+		ret_data = ((g[i] & 0x01) << i) | ret_data;
+
+	return ret_data;
+}
+
+static u8 dp_audio_get_g1_value(u8 data)
+{
+	u8 c[4];
+	u8 g[4];
+	u8 ret_data = 0;
+	u8 i;
+
+	for (i = 0; i < 4; i++)
+		c[i] = (data >> i) & 0x01;
+
+	g[0] = c[0] ^ c[3];
+	g[1] = c[0] ^ c[1] ^ c[3];
+	g[2] = c[1] ^ c[2];
+	g[3] = c[2] ^ c[3];
+
+	for (i = 0; i < 4; i++)
+		ret_data = ((g[i] & 0x01) << i) | ret_data;
+
+	return ret_data;
+}
+
+static u8 dp_audio_calculate_parity(u32 data)
+{
+	u8 x0 = 0;
+	u8 x1 = 0;
+	u8 ci = 0;
+	u8 iData = 0;
+	u8 i = 0;
+	u8 parity_byte;
+	u8 num_byte = (data & 0xFF00) > 0 ? 8 : 2;
+
+	for (i = 0; i < num_byte; i++) {
+		iData = (data >> i*4) & 0xF;
+
+		ci = iData ^ x1;
+		x1 = x0 ^ dp_audio_get_g1_value(ci);
+		x0 = dp_audio_get_g0_value(ci);
+	}
+
+	parity_byte = x1 | (x0 << 4);
+
+	return parity_byte;
+}
+
+static u32 dp_audio_get_header(struct dp_catalog *catalog,
+		enum dp_catalog_audio_sdp_type sdp,
+		enum dp_catalog_audio_header_type header)
+{
+	catalog->sdp_type = sdp;
+	catalog->sdp_header = header;
+	dp_catalog_audio_get_header(catalog);
+
+	return catalog->audio_data;
+}
+
+static void dp_audio_set_header(struct dp_catalog *catalog,
+		u32 data,
+		enum dp_catalog_audio_sdp_type sdp,
+		enum dp_catalog_audio_header_type header)
+{
+	catalog->sdp_type = sdp;
+	catalog->sdp_header = header;
+	catalog->audio_data = data;
+	dp_catalog_audio_set_header(catalog);
+}
+
+static void dp_audio_stream_sdp(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 value, new_value;
+	u8 parity_byte;
+
+	/* Config header and parity byte 1 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_1);
+
+	new_value = 0x02;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_1_BIT)
+			| (parity_byte << PARITY_BYTE_1_BIT));
+	DRM_DEBUG_DP("Header Byte 1: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_1);
+
+	/* Config header and parity byte 2 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_2);
+	new_value = value;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_2_BIT)
+			| (parity_byte << PARITY_BYTE_2_BIT));
+	DRM_DEBUG_DP("Header Byte 2: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_2);
+
+	/* Config header and parity byte 3 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_3);
+
+	new_value = audio->channels - 1;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_3_BIT)
+			| (parity_byte << PARITY_BYTE_3_BIT));
+	DRM_DEBUG_DP("Header Byte 3: value = 0x%x, parity_byte = 0x%x\n",
+		value, parity_byte);
+
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_STREAM, DP_AUDIO_SDP_HEADER_3);
+}
+
+static void dp_audio_timestamp_sdp(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 value, new_value;
+	u8 parity_byte;
+
+	/* Config header and parity byte 1 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_1);
+
+	new_value = 0x1;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_1_BIT)
+			| (parity_byte << PARITY_BYTE_1_BIT));
+	DRM_DEBUG_DP("Header Byte 1: value = 0x%x, parity_byte = 0x%x\n",
+		value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_1);
+
+	/* Config header and parity byte 2 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_2);
+
+	new_value = 0x17;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_2_BIT)
+			| (parity_byte << PARITY_BYTE_2_BIT));
+	DRM_DEBUG_DP("Header Byte 2: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_2);
+
+	/* Config header and parity byte 3 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_3);
+
+	new_value = (0x0 | (0x11 << 2));
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_3_BIT)
+			| (parity_byte << PARITY_BYTE_3_BIT));
+	DRM_DEBUG_DP("Header Byte 3: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_TIMESTAMP, DP_AUDIO_SDP_HEADER_3);
+}
+
+static void dp_audio_infoframe_sdp(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 value, new_value;
+	u8 parity_byte;
+
+	/* Config header and parity byte 1 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_1);
+
+	new_value = 0x84;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_1_BIT)
+			| (parity_byte << PARITY_BYTE_1_BIT));
+	DRM_DEBUG_DP("Header Byte 1: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_1);
+
+	/* Config header and parity byte 2 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_2);
+
+	new_value = 0x1b;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_2_BIT)
+			| (parity_byte << PARITY_BYTE_2_BIT));
+	DRM_DEBUG_DP("Header Byte 2: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_2);
+
+	/* Config header and parity byte 3 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_3);
+
+	new_value = (0x0 | (0x11 << 2));
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_3_BIT)
+			| (parity_byte << PARITY_BYTE_3_BIT));
+	DRM_DEBUG_DP("Header Byte 3: value = 0x%x, parity_byte = 0x%x\n",
+			new_value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_INFOFRAME, DP_AUDIO_SDP_HEADER_3);
+}
+
+static void dp_audio_copy_management_sdp(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 value, new_value;
+	u8 parity_byte;
+
+	/* Config header and parity byte 1 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_1);
+
+	new_value = 0x05;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_1_BIT)
+			| (parity_byte << PARITY_BYTE_1_BIT));
+	DRM_DEBUG_DP("Header Byte 1: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_1);
+
+	/* Config header and parity byte 2 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_2);
+
+	new_value = 0x0F;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_2_BIT)
+			| (parity_byte << PARITY_BYTE_2_BIT));
+	DRM_DEBUG_DP("Header Byte 2: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_2);
+
+	/* Config header and parity byte 3 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_3);
+
+	new_value = 0x0;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_3_BIT)
+			| (parity_byte << PARITY_BYTE_3_BIT));
+	DRM_DEBUG_DP("Header Byte 3: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_COPYMANAGEMENT, DP_AUDIO_SDP_HEADER_3);
+}
+
+static void dp_audio_isrc_sdp(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 value, new_value;
+	u8 parity_byte;
+
+	/* Config header and parity byte 1 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_ISRC, DP_AUDIO_SDP_HEADER_1);
+
+	new_value = 0x06;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_1_BIT)
+			| (parity_byte << PARITY_BYTE_1_BIT));
+	DRM_DEBUG_DP("Header Byte 1: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_ISRC, DP_AUDIO_SDP_HEADER_1);
+
+	/* Config header and parity byte 2 */
+	value = dp_audio_get_header(catalog,
+			DP_AUDIO_SDP_ISRC, DP_AUDIO_SDP_HEADER_2);
+
+	new_value = 0x0F;
+	parity_byte = dp_audio_calculate_parity(new_value);
+	value |= ((new_value << HEADER_BYTE_2_BIT)
+			| (parity_byte << PARITY_BYTE_2_BIT));
+	DRM_DEBUG_DP("Header Byte 2: value = 0x%x, parity_byte = 0x%x\n",
+			value, parity_byte);
+	dp_audio_set_header(catalog, value,
+		DP_AUDIO_SDP_ISRC, DP_AUDIO_SDP_HEADER_2);
+}
+
+static void dp_audio_setup_sdp(struct dp_audio_private *audio)
+{
+	dp_catalog_audio_config_sdp(audio->catalog);
+
+	dp_audio_stream_sdp(audio);
+	dp_audio_timestamp_sdp(audio);
+	dp_audio_infoframe_sdp(audio);
+	dp_audio_copy_management_sdp(audio);
+	dp_audio_isrc_sdp(audio);
+}
+
+static void dp_audio_setup_acr(struct dp_audio_private *audio)
+{
+	u32 select = 0;
+	struct dp_catalog *catalog = audio->catalog;
+
+	switch (audio->dp_audio.bw_code) {
+	case DP_LINK_BW_1_62:
+		select = 0;
+		break;
+	case DP_LINK_BW_2_7:
+		select = 1;
+		break;
+	case DP_LINK_BW_5_4:
+		select = 2;
+		break;
+	case DP_LINK_BW_8_1:
+		select = 3;
+		break;
+	default:
+		DRM_DEBUG_DP("Unknown link rate\n");
+		select = 0;
+		break;
+	}
+
+	catalog->audio_data = select;
+	dp_catalog_audio_config_acr(catalog);
+}
+
+static void dp_audio_safe_to_exit_level(struct dp_audio_private *audio)
+{
+	struct dp_catalog *catalog = audio->catalog;
+	u32 safe_to_exit_level = 0;
+
+	switch (audio->dp_audio.lane_count) {
+	case 1:
+		safe_to_exit_level = 14;
+		break;
+	case 2:
+		safe_to_exit_level = 8;
+		break;
+	case 4:
+		safe_to_exit_level = 5;
+		break;
+	default:
+		DRM_DEBUG_DP("setting the default safe_to_exit_level = %u\n",
+				safe_to_exit_level);
+		safe_to_exit_level = 14;
+		break;
+	}
+
+	catalog->audio_data = safe_to_exit_level;
+	dp_catalog_audio_sfe_level(catalog);
+}
+
+static void dp_audio_enable(struct dp_audio_private *audio, bool enable)
+{
+	struct dp_catalog *catalog = audio->catalog;
+
+	catalog->audio_data = enable;
+	dp_catalog_audio_enable(catalog);
+
+	audio->engine_on = enable;
+}
+
+static struct dp_audio_private *dp_audio_get_data(struct platform_device *pdev)
+{
+	struct dp_audio *dp_audio;
+	struct msm_dp *dp_display;
+
+	if (!pdev) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	dp_display = platform_get_drvdata(pdev);
+	if (!dp_display) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	dp_audio = dp_display->dp_audio;
+
+	if (!dp_audio) {
+		DRM_ERROR("invalid dp_audio data\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	return container_of(dp_audio, struct dp_audio_private, dp_audio);
+}
+
+static int dp_audio_hook_plugged_cb(struct device *dev, void *data,
+		hdmi_codec_plugged_cb fn,
+		struct device *codec_dev)
+{
+
+	struct platform_device *pdev;
+	struct msm_dp *dp_display;
+
+	pdev = to_platform_device(dev);
+	if (!pdev) {
+		pr_err("invalid input\n");
+		return -ENODEV;
+	}
+
+	dp_display = platform_get_drvdata(pdev);
+	if (!dp_display) {
+		pr_err("invalid input\n");
+		return -ENODEV;
+	}
+
+	return dp_display_set_plugged_cb(dp_display, fn, codec_dev);
+}
+
+static int dp_audio_get_eld(struct device *dev,
+	void *data, uint8_t *buf, size_t len)
+{
+	struct platform_device *pdev;
+	struct msm_dp *dp_display;
+
+	pdev = to_platform_device(dev);
+
+	if (!pdev) {
+		DRM_ERROR("invalid input\n");
+		return -ENODEV;
+	}
+
+	dp_display = platform_get_drvdata(pdev);
+	if (!dp_display) {
+		DRM_ERROR("invalid input\n");
+		return -ENODEV;
+	}
+
+	memcpy(buf, dp_display->connector->eld,
+		min(sizeof(dp_display->connector->eld), len));
+
+	return 0;
+}
+
+int dp_audio_hw_params(struct device *dev,
+	void *data,
+	struct hdmi_codec_daifmt *daifmt,
+	struct hdmi_codec_params *params)
+{
+	int rc = 0;
+	struct dp_audio_private *audio;
+	struct platform_device *pdev;
+	struct msm_dp *dp_display;
+
+	pdev = to_platform_device(dev);
+	dp_display = platform_get_drvdata(pdev);
+
+	/*
+	 * there could be cases where sound card can be opened even
+	 * before OR even when DP is not connected . This can cause
+	 * unclocked access as the audio subsystem relies on the DP
+	 * driver to maintain the correct state of clocks. To protect
+	 * such cases check for connection status and bail out if not
+	 * connected.
+	 */
+	if (!dp_display->power_on) {
+		rc = -EINVAL;
+		goto end;
+	}
+
+	audio = dp_audio_get_data(pdev);
+	if (IS_ERR(audio)) {
+		rc = PTR_ERR(audio);
+		goto end;
+	}
+
+	audio->channels = params->channels;
+
+	dp_audio_setup_sdp(audio);
+	dp_audio_setup_acr(audio);
+	dp_audio_safe_to_exit_level(audio);
+	dp_audio_enable(audio, true);
+	dp_display->audio_enabled = true;
+
+end:
+	return rc;
+}
+
+static void dp_audio_shutdown(struct device *dev, void *data)
+{
+	struct dp_audio_private *audio;
+	struct platform_device *pdev;
+	struct msm_dp *dp_display;
+
+	pdev = to_platform_device(dev);
+	dp_display = platform_get_drvdata(pdev);
+	audio = dp_audio_get_data(pdev);
+	if (IS_ERR(audio)) {
+		DRM_ERROR("failed to get audio data\n");
+		return;
+	}
+
+	/*
+	 * if audio was not enabled there is no need
+	 * to execute the shutdown and we can bail out early.
+	 * This also makes sure that we dont cause an unclocked
+	 * access when audio subsystem calls this without DP being
+	 * connected. is_connected cannot be used here as its set
+	 * to false earlier than this call
+	 */
+	if (!dp_display->audio_enabled)
+		return;
+
+	dp_audio_enable(audio, false);
+	/* signal the dp display to safely shutdown clocks */
+	dp_display_signal_audio_complete(dp_display);
+}
+
+static const struct hdmi_codec_ops dp_audio_codec_ops = {
+	.hw_params = dp_audio_hw_params,
+	.audio_shutdown = dp_audio_shutdown,
+	.get_eld = dp_audio_get_eld,
+	.hook_plugged_cb = dp_audio_hook_plugged_cb,
+};
+
+static struct hdmi_codec_pdata codec_data = {
+	.ops = &dp_audio_codec_ops,
+	.max_i2s_channels = 8,
+	.i2s = 1,
+};
+
+int dp_register_audio_driver(struct device *dev,
+		struct dp_audio *dp_audio)
+{
+	struct dp_audio_private *audio_priv;
+
+	audio_priv = container_of(dp_audio,
+			struct dp_audio_private, dp_audio);
+
+	audio_priv->audio_pdev = platform_device_register_data(dev,
+						HDMI_CODEC_DRV_NAME,
+						PLATFORM_DEVID_AUTO,
+						&codec_data,
+						sizeof(codec_data));
+	return PTR_ERR_OR_ZERO(audio_priv->audio_pdev);
+}
+
+struct dp_audio *dp_audio_get(struct platform_device *pdev,
+			struct dp_panel *panel,
+			struct dp_catalog *catalog)
+{
+	int rc = 0;
+	struct dp_audio_private *audio;
+	struct dp_audio *dp_audio;
+
+	if (!pdev || !panel || !catalog) {
+		DRM_ERROR("invalid input\n");
+		rc = -EINVAL;
+		goto error;
+	}
+
+	audio = devm_kzalloc(&pdev->dev, sizeof(*audio), GFP_KERNEL);
+	if (!audio) {
+		rc = -ENOMEM;
+		goto error;
+	}
+
+	audio->pdev = pdev;
+	audio->panel = panel;
+	audio->catalog = catalog;
+
+	dp_audio = &audio->dp_audio;
+
+	dp_catalog_audio_init(catalog);
+
+	return dp_audio;
+error:
+	return ERR_PTR(rc);
+}
+
+void dp_audio_put(struct dp_audio *dp_audio)
+{
+	struct dp_audio_private *audio;
+
+	if (!dp_audio)
+		return;
+
+	audio = container_of(dp_audio, struct dp_audio_private, dp_audio);
+
+	devm_kfree(&audio->pdev->dev, audio);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_audio.h b/drivers/gpu/drm/msm/dp/dp_audio.h
new file mode 100644
index 000000000000..84e5f4a5d26b
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_audio.h
@@ -0,0 +1,72 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_AUDIO_H_
+#define _DP_AUDIO_H_
+
+#include <linux/platform_device.h>
+
+#include "dp_panel.h"
+#include "dp_catalog.h"
+#include <sound/hdmi-codec.h>
+
+/**
+ * struct dp_audio
+ * @lane_count: number of lanes configured in current session
+ * @bw_code: link rate's bandwidth code for current session
+ */
+struct dp_audio {
+	u32 lane_count;
+	u32 bw_code;
+};
+
+/**
+ * dp_audio_get()
+ *
+ * Creates and instance of dp audio.
+ *
+ * @pdev: caller's platform device instance.
+ * @panel: an instance of dp_panel module.
+ * @catalog: an instance of dp_catalog module.
+ *
+ * Returns the error code in case of failure, otherwize
+ * an instance of newly created dp_module.
+ */
+struct dp_audio *dp_audio_get(struct platform_device *pdev,
+			struct dp_panel *panel,
+			struct dp_catalog *catalog);
+
+/**
+ * dp_register_audio_driver()
+ *
+ * Registers DP device with hdmi_codec interface.
+ *
+ * @dev: DP device instance.
+ * @dp_audio: an instance of dp_audio module.
+ *
+ *
+ * Returns the error code in case of failure, otherwise
+ * zero on success.
+ */
+int dp_register_audio_driver(struct device *dev,
+		struct dp_audio *dp_audio);
+
+/**
+ * dp_audio_put()
+ *
+ * Cleans the dp_audio instance.
+ *
+ * @dp_audio: an instance of dp_audio.
+ */
+void dp_audio_put(struct dp_audio *dp_audio);
+
+int dp_audio_hw_params(struct device *dev,
+	void *data,
+	struct hdmi_codec_daifmt *daifmt,
+	struct hdmi_codec_params *params);
+
+#endif /* _DP_AUDIO_H_ */
+
+
diff --git a/drivers/gpu/drm/msm/dp/dp_aux.c b/drivers/gpu/drm/msm/dp/dp_aux.c
new file mode 100644
index 000000000000..19b35ae3e927
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_aux.c
@@ -0,0 +1,535 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/delay.h>
+#include <drm/drm_print.h>
+
+#include "dp_reg.h"
+#include "dp_aux.h"
+
+#define DP_AUX_ENUM_STR(x)		#x
+
+struct dp_aux_private {
+	struct device *dev;
+	struct dp_catalog *catalog;
+
+	struct mutex mutex;
+	struct completion comp;
+
+	u32 aux_error_num;
+	u32 retry_cnt;
+	bool cmd_busy;
+	bool native;
+	bool read;
+	bool no_send_addr;
+	bool no_send_stop;
+	u32 offset;
+	u32 segment;
+	u32 isr;
+
+	struct drm_dp_aux dp_aux;
+};
+
+static const char *dp_aux_get_error(u32 aux_error)
+{
+	switch (aux_error) {
+	case DP_AUX_ERR_NONE:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_NONE);
+	case DP_AUX_ERR_ADDR:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_ADDR);
+	case DP_AUX_ERR_TOUT:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_TOUT);
+	case DP_AUX_ERR_NACK:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_NACK);
+	case DP_AUX_ERR_DEFER:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_DEFER);
+	case DP_AUX_ERR_NACK_DEFER:
+		return DP_AUX_ENUM_STR(DP_AUX_ERR_NACK_DEFER);
+	default:
+		return "unknown";
+	}
+}
+
+static u32 dp_aux_write(struct dp_aux_private *aux,
+			struct drm_dp_aux_msg *msg)
+{
+	u32 data[4], reg, len;
+	u8 *msgdata = msg->buffer;
+	int const AUX_CMD_FIFO_LEN = 128;
+	int i = 0;
+
+	if (aux->read)
+		len = 4;
+	else
+		len = msg->size + 4;
+
+	/*
+	 * cmd fifo only has depth of 144 bytes
+	 * limit buf length to 128 bytes here
+	 */
+	if (len > AUX_CMD_FIFO_LEN) {
+		DRM_ERROR("buf size greater than allowed size of 128 bytes\n");
+		return 0;
+	}
+
+	/* Pack cmd and write to HW */
+	data[0] = (msg->address >> 16) & 0xf; /* addr[19:16] */
+	if (aux->read)
+		data[0] |=  BIT(4); /* R/W */
+
+	data[1] = (msg->address >> 8) & 0xff;	/* addr[15:8] */
+	data[2] = msg->address & 0xff;		/* addr[7:0] */
+	data[3] = (msg->size - 1) & 0xff;	/* len[7:0] */
+
+	for (i = 0; i < len; i++) {
+		reg = (i < 4) ? data[i] : msgdata[i - 4];
+		/* index = 0, write */
+		reg = (((reg) << DP_AUX_DATA_OFFSET)
+		       & DP_AUX_DATA_MASK) | DP_AUX_DATA_WRITE;
+		if (i == 0)
+			reg |= DP_AUX_DATA_INDEX_WRITE;
+		aux->catalog->aux_data = reg;
+		dp_catalog_aux_write_data(aux->catalog);
+	}
+
+	dp_catalog_aux_clear_trans(aux->catalog, false);
+	dp_catalog_aux_clear_hw_interrupts(aux->catalog);
+
+	reg = 0; /* Transaction number == 1 */
+	if (!aux->native) { /* i2c */
+		reg |= DP_AUX_TRANS_CTRL_I2C;
+
+		if (aux->no_send_addr)
+			reg |= DP_AUX_TRANS_CTRL_NO_SEND_ADDR;
+
+		if (aux->no_send_stop)
+			reg |= DP_AUX_TRANS_CTRL_NO_SEND_STOP;
+	}
+
+	reg |= DP_AUX_TRANS_CTRL_GO;
+	aux->catalog->aux_data = reg;
+	dp_catalog_aux_write_trans(aux->catalog);
+
+	return len;
+}
+
+static int dp_aux_cmd_fifo_tx(struct dp_aux_private *aux,
+			      struct drm_dp_aux_msg *msg)
+{
+	u32 ret, len, timeout;
+	int aux_timeout_ms = HZ/4;
+
+	reinit_completion(&aux->comp);
+
+	len = dp_aux_write(aux, msg);
+	if (len == 0) {
+		DRM_ERROR("DP AUX write failed\n");
+		return -EINVAL;
+	}
+
+	timeout = wait_for_completion_timeout(&aux->comp, aux_timeout_ms);
+	if (!timeout) {
+		DRM_ERROR("aux %s timeout\n", (aux->read ? "read" : "write"));
+		return -ETIMEDOUT;
+	}
+
+	if (aux->aux_error_num == DP_AUX_ERR_NONE) {
+		ret = len;
+	} else {
+		DRM_ERROR_RATELIMITED("aux err: %s\n",
+			dp_aux_get_error(aux->aux_error_num));
+
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static void dp_aux_cmd_fifo_rx(struct dp_aux_private *aux,
+		struct drm_dp_aux_msg *msg)
+{
+	u32 data;
+	u8 *dp;
+	u32 i, actual_i;
+	u32 len = msg->size;
+
+	dp_catalog_aux_clear_trans(aux->catalog, true);
+
+	data = DP_AUX_DATA_INDEX_WRITE; /* INDEX_WRITE */
+	data |= DP_AUX_DATA_READ;  /* read */
+
+	aux->catalog->aux_data = data;
+	dp_catalog_aux_write_data(aux->catalog);
+
+	dp = msg->buffer;
+
+	/* discard first byte */
+	data = dp_catalog_aux_read_data(aux->catalog);
+
+	for (i = 0; i < len; i++) {
+		data = dp_catalog_aux_read_data(aux->catalog);
+		*dp++ = (u8)((data >> DP_AUX_DATA_OFFSET) & 0xff);
+
+		actual_i = (data >> DP_AUX_DATA_INDEX_OFFSET) & 0xFF;
+		if (i != actual_i)
+			DRM_ERROR("Index mismatch: expected %d, found %d\n",
+				i, actual_i);
+	}
+}
+
+static void dp_aux_native_handler(struct dp_aux_private *aux)
+{
+	u32 isr = aux->isr;
+
+	if (isr & DP_INTR_AUX_I2C_DONE)
+		aux->aux_error_num = DP_AUX_ERR_NONE;
+	else if (isr & DP_INTR_WRONG_ADDR)
+		aux->aux_error_num = DP_AUX_ERR_ADDR;
+	else if (isr & DP_INTR_TIMEOUT)
+		aux->aux_error_num = DP_AUX_ERR_TOUT;
+	if (isr & DP_INTR_NACK_DEFER)
+		aux->aux_error_num = DP_AUX_ERR_NACK;
+	if (isr & DP_INTR_AUX_ERROR) {
+		aux->aux_error_num = DP_AUX_ERR_PHY;
+		dp_catalog_aux_clear_hw_interrupts(aux->catalog);
+	}
+
+	complete(&aux->comp);
+}
+
+static void dp_aux_i2c_handler(struct dp_aux_private *aux)
+{
+	u32 isr = aux->isr;
+
+	if (isr & DP_INTR_AUX_I2C_DONE) {
+		if (isr & (DP_INTR_I2C_NACK | DP_INTR_I2C_DEFER))
+			aux->aux_error_num = DP_AUX_ERR_NACK;
+		else
+			aux->aux_error_num = DP_AUX_ERR_NONE;
+	} else {
+		if (isr & DP_INTR_WRONG_ADDR)
+			aux->aux_error_num = DP_AUX_ERR_ADDR;
+		else if (isr & DP_INTR_TIMEOUT)
+			aux->aux_error_num = DP_AUX_ERR_TOUT;
+		if (isr & DP_INTR_NACK_DEFER)
+			aux->aux_error_num = DP_AUX_ERR_NACK_DEFER;
+		if (isr & DP_INTR_I2C_NACK)
+			aux->aux_error_num = DP_AUX_ERR_NACK;
+		if (isr & DP_INTR_I2C_DEFER)
+			aux->aux_error_num = DP_AUX_ERR_DEFER;
+		if (isr & DP_INTR_AUX_ERROR) {
+			aux->aux_error_num = DP_AUX_ERR_PHY;
+			dp_catalog_aux_clear_hw_interrupts(aux->catalog);
+		}
+	}
+
+	complete(&aux->comp);
+}
+
+static void dp_aux_update_offset_and_segment(struct dp_aux_private *aux,
+					     struct drm_dp_aux_msg *input_msg)
+{
+	u32 edid_address = 0x50;
+	u32 segment_address = 0x30;
+	bool i2c_read = input_msg->request &
+		(DP_AUX_I2C_READ & DP_AUX_NATIVE_READ);
+	u8 *data;
+
+	if (aux->native || i2c_read || ((input_msg->address != edid_address) &&
+		(input_msg->address != segment_address)))
+		return;
+
+
+	data = input_msg->buffer;
+	if (input_msg->address == segment_address)
+		aux->segment = *data;
+	else
+		aux->offset = *data;
+}
+
+/**
+ * dp_aux_transfer_helper() - helper function for EDID read transactions
+ *
+ * @aux: DP AUX private structure
+ * @input_msg: input message from DRM upstream APIs
+ * @send_seg: send the segment to sink
+ *
+ * return: void
+ *
+ * This helper function is used to fix EDID reads for non-compliant
+ * sinks that do not handle the i2c middle-of-transaction flag correctly.
+ */
+static void dp_aux_transfer_helper(struct dp_aux_private *aux,
+				   struct drm_dp_aux_msg *input_msg,
+				   bool send_seg)
+{
+	struct drm_dp_aux_msg helper_msg;
+	u32 message_size = 0x10;
+	u32 segment_address = 0x30;
+	u32 const edid_block_length = 0x80;
+	bool i2c_mot = input_msg->request & DP_AUX_I2C_MOT;
+	bool i2c_read = input_msg->request &
+		(DP_AUX_I2C_READ & DP_AUX_NATIVE_READ);
+
+	if (!i2c_mot || !i2c_read || (input_msg->size == 0))
+		return;
+
+	/*
+	 * Sending the segment value and EDID offset will be performed
+	 * from the DRM upstream EDID driver for each block. Avoid
+	 * duplicate AUX transactions related to this while reading the
+	 * first 16 bytes of each block.
+	 */
+	if (!(aux->offset % edid_block_length) || !send_seg)
+		goto end;
+
+	aux->read = false;
+	aux->cmd_busy = true;
+	aux->no_send_addr = true;
+	aux->no_send_stop = true;
+
+	/*
+	 * Send the segment address for every i2c read in which the
+	 * middle-of-tranaction flag is set. This is required to support EDID
+	 * reads of more than 2 blocks as the segment address is reset to 0
+	 * since we are overriding the middle-of-transaction flag for read
+	 * transactions.
+	 */
+
+	if (aux->segment) {
+		memset(&helper_msg, 0, sizeof(helper_msg));
+		helper_msg.address = segment_address;
+		helper_msg.buffer = &aux->segment;
+		helper_msg.size = 1;
+		dp_aux_cmd_fifo_tx(aux, &helper_msg);
+	}
+
+	/*
+	 * Send the offset address for every i2c read in which the
+	 * middle-of-transaction flag is set. This will ensure that the sink
+	 * will update its read pointer and return the correct portion of the
+	 * EDID buffer in the subsequent i2c read trasntion triggered in the
+	 * native AUX transfer function.
+	 */
+	memset(&helper_msg, 0, sizeof(helper_msg));
+	helper_msg.address = input_msg->address;
+	helper_msg.buffer = &aux->offset;
+	helper_msg.size = 1;
+	dp_aux_cmd_fifo_tx(aux, &helper_msg);
+
+end:
+	aux->offset += message_size;
+	if (aux->offset == 0x80 || aux->offset == 0x100)
+		aux->segment = 0x0; /* reset segment at end of block */
+}
+
+/*
+ * This function does the real job to process an AUX transaction.
+ * It will call aux_reset() function to reset the AUX channel,
+ * if the waiting is timeout.
+ */
+static ssize_t dp_aux_transfer(struct drm_dp_aux *dp_aux,
+			       struct drm_dp_aux_msg *msg)
+{
+	ssize_t ret;
+	int const aux_cmd_native_max = 16;
+	int const aux_cmd_i2c_max = 128;
+	int const retry_count = 5;
+	struct dp_aux_private *aux = container_of(dp_aux,
+		struct dp_aux_private, dp_aux);
+
+	mutex_lock(&aux->mutex);
+
+	aux->native = msg->request & (DP_AUX_NATIVE_WRITE & DP_AUX_NATIVE_READ);
+
+	/* Ignore address only message */
+	if ((msg->size == 0) || (msg->buffer == NULL)) {
+		msg->reply = aux->native ?
+			DP_AUX_NATIVE_REPLY_ACK : DP_AUX_I2C_REPLY_ACK;
+		ret = msg->size;
+		goto unlock_exit;
+	}
+
+	/* msg sanity check */
+	if ((aux->native && (msg->size > aux_cmd_native_max)) ||
+		(msg->size > aux_cmd_i2c_max)) {
+		DRM_ERROR("%s: invalid msg: size(%zu), request(%x)\n",
+			__func__, msg->size, msg->request);
+		ret = -EINVAL;
+		goto unlock_exit;
+	}
+
+	dp_aux_update_offset_and_segment(aux, msg);
+	dp_aux_transfer_helper(aux, msg, true);
+
+	aux->read = msg->request & (DP_AUX_I2C_READ & DP_AUX_NATIVE_READ);
+	aux->cmd_busy = true;
+
+	if (aux->read) {
+		aux->no_send_addr = true;
+		aux->no_send_stop = false;
+	} else {
+		aux->no_send_addr = true;
+		aux->no_send_stop = true;
+	}
+
+	ret = dp_aux_cmd_fifo_tx(aux, msg);
+
+	if (ret < 0) {
+		if (aux->native) {
+			aux->retry_cnt++;
+			if (!(aux->retry_cnt % retry_count))
+				dp_catalog_aux_update_cfg(aux->catalog);
+			dp_catalog_aux_reset(aux->catalog);
+		}
+		usleep_range(400, 500); /* at least 400us to next try */
+		goto unlock_exit;
+	}
+
+	if (aux->aux_error_num == DP_AUX_ERR_NONE) {
+		if (aux->read)
+			dp_aux_cmd_fifo_rx(aux, msg);
+
+		msg->reply = aux->native ?
+			DP_AUX_NATIVE_REPLY_ACK : DP_AUX_I2C_REPLY_ACK;
+	} else {
+		/* Reply defer to retry */
+		msg->reply = aux->native ?
+			DP_AUX_NATIVE_REPLY_DEFER : DP_AUX_I2C_REPLY_DEFER;
+	}
+
+	/* Return requested size for success or retry */
+	ret = msg->size;
+	aux->retry_cnt = 0;
+
+unlock_exit:
+	aux->cmd_busy = false;
+	mutex_unlock(&aux->mutex);
+	return ret;
+}
+
+void dp_aux_isr(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+
+	if (!dp_aux) {
+		DRM_ERROR("invalid input\n");
+		return;
+	}
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	aux->isr = dp_catalog_aux_get_irq(aux->catalog);
+
+	if (!aux->cmd_busy)
+		return;
+
+	if (aux->native)
+		dp_aux_native_handler(aux);
+	else
+		dp_aux_i2c_handler(aux);
+}
+
+void dp_aux_reconfig(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	dp_catalog_aux_update_cfg(aux->catalog);
+	dp_catalog_aux_reset(aux->catalog);
+}
+
+void dp_aux_init(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+
+	if (!dp_aux) {
+		DRM_ERROR("invalid input\n");
+		return;
+	}
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	dp_catalog_aux_enable(aux->catalog, true);
+	aux->retry_cnt = 0;
+}
+
+void dp_aux_deinit(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	dp_catalog_aux_enable(aux->catalog, false);
+}
+
+int dp_aux_register(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+	int ret;
+
+	if (!dp_aux) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	aux->dp_aux.name = "dpu_dp_aux";
+	aux->dp_aux.dev = aux->dev;
+	aux->dp_aux.transfer = dp_aux_transfer;
+	ret = drm_dp_aux_register(&aux->dp_aux);
+	if (ret) {
+		DRM_ERROR("%s: failed to register drm aux: %d\n", __func__,
+				ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+void dp_aux_unregister(struct drm_dp_aux *dp_aux)
+{
+	drm_dp_aux_unregister(dp_aux);
+}
+
+struct drm_dp_aux *dp_aux_get(struct device *dev, struct dp_catalog *catalog)
+{
+	struct dp_aux_private *aux;
+
+	if (!catalog) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	aux = devm_kzalloc(dev, sizeof(*aux), GFP_KERNEL);
+	if (!aux)
+		return ERR_PTR(-ENOMEM);
+
+	init_completion(&aux->comp);
+	aux->cmd_busy = false;
+	mutex_init(&aux->mutex);
+
+	aux->dev = dev;
+	aux->catalog = catalog;
+	aux->retry_cnt = 0;
+
+	return &aux->dp_aux;
+}
+
+void dp_aux_put(struct drm_dp_aux *dp_aux)
+{
+	struct dp_aux_private *aux;
+
+	if (!dp_aux)
+		return;
+
+	aux = container_of(dp_aux, struct dp_aux_private, dp_aux);
+
+	mutex_destroy(&aux->mutex);
+
+	devm_kfree(aux->dev, aux);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_aux.h b/drivers/gpu/drm/msm/dp/dp_aux.h
new file mode 100644
index 000000000000..f8b8ba919465
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_aux.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_AUX_H_
+#define _DP_AUX_H_
+
+#include "dp_catalog.h"
+#include <drm/drm_dp_helper.h>
+
+#define DP_AUX_ERR_NONE		0
+#define DP_AUX_ERR_ADDR		-1
+#define DP_AUX_ERR_TOUT		-2
+#define DP_AUX_ERR_NACK		-3
+#define DP_AUX_ERR_DEFER	-4
+#define DP_AUX_ERR_NACK_DEFER	-5
+#define DP_AUX_ERR_PHY		-6
+
+int dp_aux_register(struct drm_dp_aux *dp_aux);
+void dp_aux_unregister(struct drm_dp_aux *dp_aux);
+void dp_aux_isr(struct drm_dp_aux *dp_aux);
+void dp_aux_init(struct drm_dp_aux *dp_aux);
+void dp_aux_deinit(struct drm_dp_aux *dp_aux);
+void dp_aux_reconfig(struct drm_dp_aux *dp_aux);
+
+struct drm_dp_aux *dp_aux_get(struct device *dev, struct dp_catalog *catalog);
+void dp_aux_put(struct drm_dp_aux *aux);
+
+#endif /*__DP_AUX_H_*/
diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.c b/drivers/gpu/drm/msm/dp/dp_catalog.c
new file mode 100644
index 000000000000..b15b4ce4ba35
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_catalog.c
@@ -0,0 +1,1019 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <linux/rational.h>
+#include <linux/delay.h>
+#include <linux/iopoll.h>
+#include <linux/phy/phy.h>
+#include <linux/phy/phy-dp.h>
+#include <linux/rational.h>
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_print.h>
+
+#include "dp_catalog.h"
+#include "dp_reg.h"
+
+#define POLLING_SLEEP_US			1000
+#define POLLING_TIMEOUT_US			10000
+
+#define SCRAMBLER_RESET_COUNT_VALUE		0xFC
+
+#define DP_INTERRUPT_STATUS_ACK_SHIFT	1
+#define DP_INTERRUPT_STATUS_MASK_SHIFT	2
+
+#define MSM_DP_CONTROLLER_AHB_OFFSET	0x0000
+#define MSM_DP_CONTROLLER_AHB_SIZE	0x0200
+#define MSM_DP_CONTROLLER_AUX_OFFSET	0x0200
+#define MSM_DP_CONTROLLER_AUX_SIZE	0x0200
+#define MSM_DP_CONTROLLER_LINK_OFFSET	0x0400
+#define MSM_DP_CONTROLLER_LINK_SIZE	0x0C00
+#define MSM_DP_CONTROLLER_P0_OFFSET	0x1000
+#define MSM_DP_CONTROLLER_P0_SIZE	0x0400
+
+#define DP_INTERRUPT_STATUS1 \
+	(DP_INTR_AUX_I2C_DONE| \
+	DP_INTR_WRONG_ADDR | DP_INTR_TIMEOUT | \
+	DP_INTR_NACK_DEFER | DP_INTR_WRONG_DATA_CNT | \
+	DP_INTR_I2C_NACK | DP_INTR_I2C_DEFER | \
+	DP_INTR_PLL_UNLOCKED | DP_INTR_AUX_ERROR)
+
+#define DP_INTERRUPT_STATUS1_ACK \
+	(DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_ACK_SHIFT)
+#define DP_INTERRUPT_STATUS1_MASK \
+	(DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_MASK_SHIFT)
+
+#define DP_INTERRUPT_STATUS2 \
+	(DP_INTR_READY_FOR_VIDEO | DP_INTR_IDLE_PATTERN_SENT | \
+	DP_INTR_FRAME_END | DP_INTR_CRC_UPDATED)
+
+#define DP_INTERRUPT_STATUS2_ACK \
+	(DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_ACK_SHIFT)
+#define DP_INTERRUPT_STATUS2_MASK \
+	(DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_MASK_SHIFT)
+
+struct dp_catalog_private {
+	struct device *dev;
+	struct dp_io *io;
+	u32 (*audio_map)[DP_AUDIO_SDP_HEADER_MAX];
+	struct dp_catalog dp_catalog;
+	u8 aux_lut_cfg_index[PHY_AUX_CFG_MAX];
+};
+
+static inline u32 dp_read_aux(struct dp_catalog_private *catalog, u32 offset)
+{
+	offset += MSM_DP_CONTROLLER_AUX_OFFSET;
+	return readl_relaxed(catalog->io->dp_controller.base + offset);
+}
+
+static inline void dp_write_aux(struct dp_catalog_private *catalog,
+			       u32 offset, u32 data)
+{
+	offset += MSM_DP_CONTROLLER_AUX_OFFSET;
+	/*
+	 * To make sure aux reg writes happens before any other operation,
+	 * this function uses writel() instread of writel_relaxed()
+	 */
+	writel(data, catalog->io->dp_controller.base + offset);
+}
+
+static inline u32 dp_read_ahb(struct dp_catalog_private *catalog, u32 offset)
+{
+	offset += MSM_DP_CONTROLLER_AHB_OFFSET;
+	return readl_relaxed(catalog->io->dp_controller.base + offset);
+}
+
+static inline void dp_write_ahb(struct dp_catalog_private *catalog,
+			       u32 offset, u32 data)
+{
+	offset += MSM_DP_CONTROLLER_AHB_OFFSET;
+	/*
+	 * To make sure phy reg writes happens before any other operation,
+	 * this function uses writel() instread of writel_relaxed()
+	 */
+	writel(data, catalog->io->dp_controller.base + offset);
+}
+
+static inline void dp_write_p0(struct dp_catalog_private *catalog,
+			       u32 offset, u32 data)
+{
+	offset += MSM_DP_CONTROLLER_P0_OFFSET;
+	/*
+	 * To make sure interface reg writes happens before any other operation,
+	 * this function uses writel() instread of writel_relaxed()
+	 */
+	writel(data, catalog->io->dp_controller.base + offset);
+}
+
+static inline u32 dp_read_p0(struct dp_catalog_private *catalog,
+			       u32 offset)
+{
+	offset += MSM_DP_CONTROLLER_P0_OFFSET;
+	/*
+	 * To make sure interface reg writes happens before any other operation,
+	 * this function uses writel() instread of writel_relaxed()
+	 */
+	return readl_relaxed(catalog->io->dp_controller.base + offset);
+}
+
+static inline u32 dp_read_link(struct dp_catalog_private *catalog, u32 offset)
+{
+	offset += MSM_DP_CONTROLLER_LINK_OFFSET;
+	return readl_relaxed(catalog->io->dp_controller.base + offset);
+}
+
+static inline void dp_write_link(struct dp_catalog_private *catalog,
+			       u32 offset, u32 data)
+{
+	offset += MSM_DP_CONTROLLER_LINK_OFFSET;
+	/*
+	 * To make sure link reg writes happens before any other operation,
+	 * this function uses writel() instread of writel_relaxed()
+	 */
+	writel(data, catalog->io->dp_controller.base + offset);
+}
+
+/* aux related catalog functions */
+u32 dp_catalog_aux_read_data(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	return dp_read_aux(catalog, REG_DP_AUX_DATA);
+}
+
+int dp_catalog_aux_write_data(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_aux(catalog, REG_DP_AUX_DATA, dp_catalog->aux_data);
+	return 0;
+}
+
+int dp_catalog_aux_write_trans(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, dp_catalog->aux_data);
+	return 0;
+}
+
+int dp_catalog_aux_clear_trans(struct dp_catalog *dp_catalog, bool read)
+{
+	u32 data;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	if (read) {
+		data = dp_read_aux(catalog, REG_DP_AUX_TRANS_CTRL);
+		data &= ~DP_AUX_TRANS_CTRL_GO;
+		dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, data);
+	} else {
+		dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, 0);
+	}
+	return 0;
+}
+
+int dp_catalog_aux_clear_hw_interrupts(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_read_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_STATUS);
+	dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x1f);
+	dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x9f);
+	dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0);
+	return 0;
+}
+
+void dp_catalog_aux_reset(struct dp_catalog *dp_catalog)
+{
+	u32 aux_ctrl;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	aux_ctrl = dp_read_aux(catalog, REG_DP_AUX_CTRL);
+
+	aux_ctrl |= DP_AUX_CTRL_RESET;
+	dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl);
+	usleep_range(1000, 1100); /* h/w recommended delay */
+
+	aux_ctrl &= ~DP_AUX_CTRL_RESET;
+	dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl);
+}
+
+void dp_catalog_aux_enable(struct dp_catalog *dp_catalog, bool enable)
+{
+	u32 aux_ctrl;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	aux_ctrl = dp_read_aux(catalog, REG_DP_AUX_CTRL);
+
+	if (enable) {
+		dp_write_aux(catalog, REG_DP_TIMEOUT_COUNT, 0xffff);
+		dp_write_aux(catalog, REG_DP_AUX_LIMITS, 0xffff);
+		aux_ctrl |= DP_AUX_CTRL_ENABLE;
+	} else {
+		aux_ctrl &= ~DP_AUX_CTRL_ENABLE;
+	}
+
+	dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl);
+}
+
+void dp_catalog_aux_update_cfg(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	struct dp_io *dp_io = catalog->io;
+	struct phy *phy = dp_io->phy;
+
+	phy_calibrate(phy);
+}
+
+static void dump_regs(void __iomem *base, int len)
+{
+	int i;
+	u32 x0, x4, x8, xc;
+	u32 addr_off = 0;
+
+	len = DIV_ROUND_UP(len, 16);
+	for (i = 0; i < len; i++) {
+		x0 = readl_relaxed(base + addr_off);
+		x4 = readl_relaxed(base + addr_off + 0x04);
+		x8 = readl_relaxed(base + addr_off + 0x08);
+		xc = readl_relaxed(base + addr_off + 0x0c);
+
+		pr_info("%08x: %08x %08x %08x %08x", addr_off, x0, x4, x8, xc);
+		addr_off += 16;
+	}
+}
+
+void dp_catalog_dump_regs(struct dp_catalog *dp_catalog)
+{
+	u32 offset, len;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	pr_info("AHB regs\n");
+	offset = MSM_DP_CONTROLLER_AHB_OFFSET;
+	len = MSM_DP_CONTROLLER_AHB_SIZE;
+	dump_regs(catalog->io->dp_controller.base + offset, len);
+
+	pr_info("AUXCLK regs\n");
+	offset = MSM_DP_CONTROLLER_AUX_OFFSET;
+	len = MSM_DP_CONTROLLER_AUX_SIZE;
+	dump_regs(catalog->io->dp_controller.base + offset, len);
+
+	pr_info("LCLK regs\n");
+	offset = MSM_DP_CONTROLLER_LINK_OFFSET;
+	len = MSM_DP_CONTROLLER_LINK_SIZE;
+	dump_regs(catalog->io->dp_controller.base + offset, len);
+
+	pr_info("P0CLK regs\n");
+	offset = MSM_DP_CONTROLLER_P0_OFFSET;
+	len = MSM_DP_CONTROLLER_P0_SIZE;
+	dump_regs(catalog->io->dp_controller.base + offset, len);
+}
+
+int dp_catalog_aux_get_irq(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	u32 intr, intr_ack;
+
+	intr = dp_read_ahb(catalog, REG_DP_INTR_STATUS);
+	intr &= ~DP_INTERRUPT_STATUS1_MASK;
+	intr_ack = (intr & DP_INTERRUPT_STATUS1)
+			<< DP_INTERRUPT_STATUS_ACK_SHIFT;
+	dp_write_ahb(catalog, REG_DP_INTR_STATUS, intr_ack |
+			DP_INTERRUPT_STATUS1_MASK);
+
+	return intr;
+
+}
+
+/* controller related catalog functions */
+void dp_catalog_ctrl_update_transfer_unit(struct dp_catalog *dp_catalog,
+				u32 dp_tu, u32 valid_boundary,
+				u32 valid_boundary2)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_link(catalog, REG_DP_VALID_BOUNDARY, valid_boundary);
+	dp_write_link(catalog, REG_DP_TU, dp_tu);
+	dp_write_link(catalog, REG_DP_VALID_BOUNDARY_2, valid_boundary2);
+}
+
+void dp_catalog_ctrl_state_ctrl(struct dp_catalog *dp_catalog, u32 state)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_link(catalog, REG_DP_STATE_CTRL, state);
+}
+
+void dp_catalog_ctrl_config_ctrl(struct dp_catalog *dp_catalog, u32 cfg)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	DRM_DEBUG_DP("DP_CONFIGURATION_CTRL=0x%x\n", cfg);
+
+	dp_write_link(catalog, REG_DP_CONFIGURATION_CTRL, cfg);
+}
+
+void dp_catalog_ctrl_lane_mapping(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	u32 ln_0 = 0, ln_1 = 1, ln_2 = 2, ln_3 = 3; /* One-to-One mapping */
+	u32 ln_mapping;
+
+	ln_mapping = ln_0 << LANE0_MAPPING_SHIFT;
+	ln_mapping |= ln_1 << LANE1_MAPPING_SHIFT;
+	ln_mapping |= ln_2 << LANE2_MAPPING_SHIFT;
+	ln_mapping |= ln_3 << LANE3_MAPPING_SHIFT;
+
+	dp_write_link(catalog, REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING,
+			ln_mapping);
+}
+
+void dp_catalog_ctrl_mainlink_ctrl(struct dp_catalog *dp_catalog,
+						bool enable)
+{
+	u32 mainlink_ctrl;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	if (enable) {
+		/*
+		 * To make sure link reg writes happens before other operation,
+		 * dp_write_link() function uses writel()
+		 */
+		mainlink_ctrl = dp_read_link(catalog, REG_DP_MAINLINK_CTRL);
+
+		mainlink_ctrl &= ~(DP_MAINLINK_CTRL_RESET |
+						DP_MAINLINK_CTRL_ENABLE);
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl);
+
+		mainlink_ctrl |= DP_MAINLINK_CTRL_RESET;
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl);
+
+		mainlink_ctrl &= ~DP_MAINLINK_CTRL_RESET;
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl);
+
+		mainlink_ctrl |= (DP_MAINLINK_CTRL_ENABLE |
+					DP_MAINLINK_FB_BOUNDARY_SEL);
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl);
+	} else {
+		mainlink_ctrl = dp_read_link(catalog, REG_DP_MAINLINK_CTRL);
+		mainlink_ctrl &= ~DP_MAINLINK_CTRL_ENABLE;
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl);
+	}
+}
+
+void dp_catalog_ctrl_config_misc(struct dp_catalog *dp_catalog,
+					u32 colorimetry_cfg,
+					u32 test_bits_depth)
+{
+	u32 misc_val;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	misc_val = dp_read_link(catalog, REG_DP_MISC1_MISC0);
+
+	/* clear bpp bits */
+	misc_val &= ~(0x07 << DP_MISC0_TEST_BITS_DEPTH_SHIFT);
+	misc_val |= colorimetry_cfg << DP_MISC0_COLORIMETRY_CFG_SHIFT;
+	misc_val |= test_bits_depth << DP_MISC0_TEST_BITS_DEPTH_SHIFT;
+	/* Configure clock to synchronous mode */
+	misc_val |= DP_MISC0_SYNCHRONOUS_CLK;
+
+	DRM_DEBUG_DP("misc settings = 0x%x\n", misc_val);
+	dp_write_link(catalog, REG_DP_MISC1_MISC0, misc_val);
+}
+
+void dp_catalog_ctrl_config_msa(struct dp_catalog *dp_catalog,
+					u32 rate, u32 stream_rate_khz,
+					bool fixed_nvid)
+{
+	u32 pixel_m, pixel_n;
+	u32 mvid, nvid, pixel_div = 0, dispcc_input_rate;
+	u32 const nvid_fixed = DP_LINK_CONSTANT_N_VALUE;
+	u32 const link_rate_hbr2 = 540000;
+	u32 const link_rate_hbr3 = 810000;
+	unsigned long den, num;
+
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	if (rate == link_rate_hbr3)
+		pixel_div = 6;
+	else if (rate == 1620000 || rate == 270000)
+		pixel_div = 2;
+	else if (rate == link_rate_hbr2)
+		pixel_div = 4;
+	else
+		DRM_ERROR("Invalid pixel mux divider\n");
+
+	dispcc_input_rate = (rate * 10) / pixel_div;
+
+	rational_best_approximation(dispcc_input_rate, stream_rate_khz,
+			(unsigned long)(1 << 16) - 1,
+			(unsigned long)(1 << 16) - 1, &den, &num);
+
+	den = ~(den - num);
+	den = den & 0xFFFF;
+	pixel_m = num;
+	pixel_n = den;
+
+	mvid = (pixel_m & 0xFFFF) * 5;
+	nvid = (0xFFFF & (~pixel_n)) + (pixel_m & 0xFFFF);
+
+	if (nvid < nvid_fixed) {
+		u32 temp;
+
+		temp = (nvid_fixed / nvid) * nvid;
+		mvid = (nvid_fixed / nvid) * mvid;
+		nvid = temp;
+	}
+
+	if (link_rate_hbr2 == rate)
+		nvid *= 2;
+
+	if (link_rate_hbr3 == rate)
+		nvid *= 3;
+
+	DRM_DEBUG_DP("mvid=0x%x, nvid=0x%x\n", mvid, nvid);
+	dp_write_link(catalog, REG_DP_SOFTWARE_MVID, mvid);
+	dp_write_link(catalog, REG_DP_SOFTWARE_NVID, nvid);
+	dp_write_p0(catalog, MMSS_DP_DSC_DTO, 0x0);
+}
+
+int dp_catalog_ctrl_set_pattern(struct dp_catalog *dp_catalog,
+					u32 pattern)
+{
+	int bit, ret;
+	u32 data;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	bit = BIT(pattern - 1);
+	DRM_DEBUG_DP("hw: bit=%d train=%d\n", bit, pattern);
+	dp_catalog_ctrl_state_ctrl(dp_catalog, bit);
+
+	bit = BIT(pattern - 1) << DP_MAINLINK_READY_LINK_TRAINING_SHIFT;
+
+	/* Poll for mainlink ready status */
+	ret = readx_poll_timeout(readl, catalog->io->dp_controller.base +
+					MSM_DP_CONTROLLER_LINK_OFFSET +
+					REG_DP_MAINLINK_READY,
+					data, data & bit,
+					POLLING_SLEEP_US, POLLING_TIMEOUT_US);
+	if (ret < 0) {
+		DRM_ERROR("set pattern for link_train=%d failed\n", pattern);
+		return ret;
+	}
+	return 0;
+}
+
+void dp_catalog_ctrl_reset(struct dp_catalog *dp_catalog)
+{
+	u32 sw_reset;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	sw_reset = dp_read_ahb(catalog, REG_DP_SW_RESET);
+
+	sw_reset |= DP_SW_RESET;
+	dp_write_ahb(catalog, REG_DP_SW_RESET, sw_reset);
+	usleep_range(1000, 1100); /* h/w recommended delay */
+
+	sw_reset &= ~DP_SW_RESET;
+	dp_write_ahb(catalog, REG_DP_SW_RESET, sw_reset);
+}
+
+bool dp_catalog_ctrl_mainlink_ready(struct dp_catalog *dp_catalog)
+{
+	u32 data;
+	int ret;
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	/* Poll for mainlink ready status */
+	ret = readl_poll_timeout(catalog->io->dp_controller.base +
+				MSM_DP_CONTROLLER_LINK_OFFSET +
+				REG_DP_MAINLINK_READY,
+				data, data & DP_MAINLINK_READY_FOR_VIDEO,
+				POLLING_SLEEP_US, POLLING_TIMEOUT_US);
+	if (ret < 0) {
+		DRM_ERROR("mainlink not ready\n");
+		return false;
+	}
+
+	return true;
+}
+
+void dp_catalog_ctrl_enable_irq(struct dp_catalog *dp_catalog,
+						bool enable)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	if (enable) {
+		dp_write_ahb(catalog, REG_DP_INTR_STATUS,
+				DP_INTERRUPT_STATUS1_MASK);
+		dp_write_ahb(catalog, REG_DP_INTR_STATUS2,
+				DP_INTERRUPT_STATUS2_MASK);
+	} else {
+		dp_write_ahb(catalog, REG_DP_INTR_STATUS, 0x00);
+		dp_write_ahb(catalog, REG_DP_INTR_STATUS2, 0x00);
+	}
+}
+
+void dp_catalog_hpd_config_intr(struct dp_catalog *dp_catalog,
+			u32 intr_mask, bool en)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	u32 config = dp_read_aux(catalog, REG_DP_DP_HPD_INT_MASK);
+
+	config = (en ? config | intr_mask : config & ~intr_mask);
+
+	dp_write_aux(catalog, REG_DP_DP_HPD_INT_MASK,
+				config & DP_DP_HPD_INT_MASK);
+}
+
+void dp_catalog_ctrl_hpd_config(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	u32 reftimer = dp_read_aux(catalog, REG_DP_DP_HPD_REFTIMER);
+
+	/* enable HPD interrupts */
+	dp_catalog_hpd_config_intr(dp_catalog,
+		DP_DP_HPD_PLUG_INT_MASK | DP_DP_IRQ_HPD_INT_MASK
+		| DP_DP_HPD_UNPLUG_INT_MASK | DP_DP_HPD_REPLUG_INT_MASK, true);
+
+	/* Configure REFTIMER and enable it */
+	reftimer |= DP_DP_HPD_REFTIMER_ENABLE;
+	dp_write_aux(catalog, REG_DP_DP_HPD_REFTIMER, reftimer);
+
+	/* Enable HPD */
+	dp_write_aux(catalog, REG_DP_DP_HPD_CTRL, DP_DP_HPD_CTRL_HPD_EN);
+}
+
+u32 dp_catalog_hpd_get_intr_status(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	int isr = 0;
+
+	isr = dp_read_aux(catalog, REG_DP_DP_HPD_INT_STATUS);
+	dp_write_aux(catalog, REG_DP_DP_HPD_INT_ACK,
+				 (isr & DP_DP_HPD_INT_MASK));
+
+	return isr;
+}
+
+int dp_catalog_ctrl_get_interrupt(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	u32 intr, intr_ack;
+
+	intr = dp_read_ahb(catalog, REG_DP_INTR_STATUS2);
+	intr &= ~DP_INTERRUPT_STATUS2_MASK;
+	intr_ack = (intr & DP_INTERRUPT_STATUS2)
+			<< DP_INTERRUPT_STATUS_ACK_SHIFT;
+	dp_write_ahb(catalog, REG_DP_INTR_STATUS2,
+			intr_ack | DP_INTERRUPT_STATUS2_MASK);
+
+	return intr;
+}
+
+void dp_catalog_ctrl_phy_reset(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_ahb(catalog, REG_DP_PHY_CTRL,
+			DP_PHY_CTRL_SW_RESET | DP_PHY_CTRL_SW_RESET_PLL);
+	usleep_range(1000, 1100); /* h/w recommended delay */
+	dp_write_ahb(catalog, REG_DP_PHY_CTRL, 0x0);
+}
+
+int dp_catalog_ctrl_update_vx_px(struct dp_catalog *dp_catalog,
+		u8 v_level, u8 p_level)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	struct dp_io *dp_io = catalog->io;
+	struct phy *phy = dp_io->phy;
+	struct phy_configure_opts_dp *opts_dp = &dp_io->phy_opts.dp;
+
+	/* TODO: Update for all lanes instead of just first one */
+	opts_dp->voltage[0] = v_level;
+	opts_dp->pre[0] = p_level;
+	opts_dp->set_voltages = 1;
+	phy_configure(phy, &dp_io->phy_opts);
+	opts_dp->set_voltages = 0;
+
+	return 0;
+}
+
+void dp_catalog_ctrl_send_phy_pattern(struct dp_catalog *dp_catalog,
+			u32 pattern)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	u32 value = 0x0;
+
+	/* Make sure to clear the current pattern before starting a new one */
+	dp_write_link(catalog, REG_DP_STATE_CTRL, 0x0);
+
+	switch (pattern) {
+	case DP_PHY_TEST_PATTERN_D10_2:
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+				DP_STATE_CTRL_LINK_TRAINING_PATTERN1);
+		break;
+	case DP_PHY_TEST_PATTERN_ERROR_COUNT:
+		value &= ~(1 << 16);
+		dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET,
+					value);
+		value |= SCRAMBLER_RESET_COUNT_VALUE;
+		dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET,
+					value);
+		dp_write_link(catalog, REG_DP_MAINLINK_LEVELS,
+					DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2);
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+					DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE);
+		break;
+	case DP_PHY_TEST_PATTERN_PRBS7:
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+				DP_STATE_CTRL_LINK_PRBS7);
+		break;
+	case DP_PHY_TEST_PATTERN_80BIT_CUSTOM:
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+				DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN);
+		/* 00111110000011111000001111100000 */
+		dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0,
+				0x3E0F83E0);
+		/* 00001111100000111110000011111000 */
+		dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1,
+				0x0F83E0F8);
+		/* 1111100000111110 */
+		dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2,
+				0x0000F83E);
+		break;
+	case DP_PHY_TEST_PATTERN_CP2520:
+		value = dp_read_link(catalog, REG_DP_MAINLINK_CTRL);
+		value &= ~DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER;
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, value);
+
+		value = DP_HBR2_ERM_PATTERN;
+		dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET,
+				value);
+		value |= SCRAMBLER_RESET_COUNT_VALUE;
+		dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET,
+					value);
+		dp_write_link(catalog, REG_DP_MAINLINK_LEVELS,
+					DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2);
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+					DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE);
+		value = dp_read_link(catalog, REG_DP_MAINLINK_CTRL);
+		value |= DP_MAINLINK_CTRL_ENABLE;
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL, value);
+		break;
+	case DP_PHY_TEST_PATTERN_SEL_MASK:
+		dp_write_link(catalog, REG_DP_MAINLINK_CTRL,
+				DP_MAINLINK_CTRL_ENABLE);
+		dp_write_link(catalog, REG_DP_STATE_CTRL,
+				DP_STATE_CTRL_LINK_TRAINING_PATTERN4);
+		break;
+	default:
+		DRM_DEBUG_DP("No valid test pattern requested:0x%x\n", pattern);
+		break;
+	}
+}
+
+u32 dp_catalog_ctrl_read_phy_pattern(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	return dp_read_link(catalog, REG_DP_MAINLINK_READY);
+}
+
+/* panel related catalog functions */
+int dp_catalog_panel_timing_cfg(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_link(catalog, REG_DP_TOTAL_HOR_VER,
+				dp_catalog->total);
+	dp_write_link(catalog, REG_DP_START_HOR_VER_FROM_SYNC,
+				dp_catalog->sync_start);
+	dp_write_link(catalog, REG_DP_HSYNC_VSYNC_WIDTH_POLARITY,
+				dp_catalog->width_blanking);
+	dp_write_link(catalog, REG_DP_ACTIVE_HOR_VER, dp_catalog->dp_active);
+	return 0;
+}
+
+void dp_catalog_panel_tpg_enable(struct dp_catalog *dp_catalog,
+				struct drm_display_mode *drm_mode)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+	u32 hsync_period, vsync_period;
+	u32 display_v_start, display_v_end;
+	u32 hsync_start_x, hsync_end_x;
+	u32 v_sync_width;
+	u32 hsync_ctl;
+	u32 display_hctl;
+
+	/* TPG config parameters*/
+	hsync_period = drm_mode->htotal;
+	vsync_period = drm_mode->vtotal;
+
+	display_v_start = ((drm_mode->vtotal - drm_mode->vsync_start) *
+					hsync_period);
+	display_v_end = ((vsync_period - (drm_mode->vsync_start -
+					drm_mode->vdisplay))
+					* hsync_period) - 1;
+
+	display_v_start += drm_mode->htotal - drm_mode->hsync_start;
+	display_v_end -= (drm_mode->hsync_start - drm_mode->hdisplay);
+
+	hsync_start_x = drm_mode->htotal - drm_mode->hsync_start;
+	hsync_end_x = hsync_period - (drm_mode->hsync_start -
+					drm_mode->hdisplay) - 1;
+
+	v_sync_width = drm_mode->vsync_end - drm_mode->vsync_start;
+
+	hsync_ctl = (hsync_period << 16) |
+			(drm_mode->hsync_end - drm_mode->hsync_start);
+	display_hctl = (hsync_end_x << 16) | hsync_start_x;
+
+
+	dp_write_p0(catalog, MMSS_DP_INTF_CONFIG, 0x0);
+	dp_write_p0(catalog, MMSS_DP_INTF_HSYNC_CTL, hsync_ctl);
+	dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PERIOD_F0, vsync_period *
+			hsync_period);
+	dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0, v_sync_width *
+			hsync_period);
+	dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PERIOD_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_HCTL, display_hctl);
+	dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_HCTL, 0);
+	dp_write_p0(catalog, MMSS_INTF_DISPLAY_V_START_F0, display_v_start);
+	dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_V_END_F0, display_v_end);
+	dp_write_p0(catalog, MMSS_INTF_DISPLAY_V_START_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_V_END_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_START_F0, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_END_F0, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_START_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_END_F1, 0);
+	dp_write_p0(catalog, MMSS_DP_INTF_POLARITY_CTL, 0);
+
+	dp_write_p0(catalog, MMSS_DP_TPG_MAIN_CONTROL,
+				DP_TPG_CHECKERED_RECT_PATTERN);
+	dp_write_p0(catalog, MMSS_DP_TPG_VIDEO_CONFIG,
+				DP_TPG_VIDEO_CONFIG_BPP_8BIT |
+				DP_TPG_VIDEO_CONFIG_RGB);
+	dp_write_p0(catalog, MMSS_DP_BIST_ENABLE,
+				DP_BIST_ENABLE_DPBIST_EN);
+	dp_write_p0(catalog, MMSS_DP_TIMING_ENGINE_EN,
+				DP_TIMING_ENGINE_EN_EN);
+	DRM_DEBUG_DP("%s: enabled tpg\n", __func__);
+}
+
+void dp_catalog_panel_tpg_disable(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog = container_of(dp_catalog,
+				struct dp_catalog_private, dp_catalog);
+
+	dp_write_p0(catalog, MMSS_DP_TPG_MAIN_CONTROL, 0x0);
+	dp_write_p0(catalog, MMSS_DP_BIST_ENABLE, 0x0);
+	dp_write_p0(catalog, MMSS_DP_TIMING_ENGINE_EN, 0x0);
+}
+
+struct dp_catalog *dp_catalog_get(struct device *dev, struct dp_io *io)
+{
+	struct dp_catalog_private *catalog;
+
+	if (!io) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	catalog  = devm_kzalloc(dev, sizeof(*catalog), GFP_KERNEL);
+	if (!catalog)
+		return ERR_PTR(-ENOMEM);
+
+	catalog->dev = dev;
+	catalog->io = io;
+
+	return &catalog->dp_catalog;
+}
+
+void dp_catalog_audio_get_header(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	u32 (*sdp_map)[DP_AUDIO_SDP_HEADER_MAX];
+	enum dp_catalog_audio_sdp_type sdp;
+	enum dp_catalog_audio_header_type header;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	sdp_map = catalog->audio_map;
+	sdp     = dp_catalog->sdp_type;
+	header  = dp_catalog->sdp_header;
+
+	dp_catalog->audio_data = dp_read_link(catalog,
+			sdp_map[sdp][header]);
+}
+
+void dp_catalog_audio_set_header(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	u32 (*sdp_map)[DP_AUDIO_SDP_HEADER_MAX];
+	enum dp_catalog_audio_sdp_type sdp;
+	enum dp_catalog_audio_header_type header;
+	u32 data;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	sdp_map = catalog->audio_map;
+	sdp     = dp_catalog->sdp_type;
+	header  = dp_catalog->sdp_header;
+	data    = dp_catalog->audio_data;
+
+	dp_write_link(catalog, sdp_map[sdp][header], data);
+}
+
+void dp_catalog_audio_config_acr(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	u32 acr_ctrl, select;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	select = dp_catalog->audio_data;
+	acr_ctrl = select << 4 | BIT(31) | BIT(8) | BIT(14);
+
+	DRM_DEBUG_DP("select = 0x%x, acr_ctrl = 0x%x\n", select, acr_ctrl);
+
+	dp_write_link(catalog, MMSS_DP_AUDIO_ACR_CTRL, acr_ctrl);
+}
+
+void dp_catalog_audio_enable(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	bool enable;
+	u32 audio_ctrl;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	enable = !!dp_catalog->audio_data;
+	audio_ctrl = dp_read_link(catalog, MMSS_DP_AUDIO_CFG);
+
+	if (enable)
+		audio_ctrl |= BIT(0);
+	else
+		audio_ctrl &= ~BIT(0);
+
+	DRM_DEBUG_DP("dp_audio_cfg = 0x%x\n", audio_ctrl);
+
+	dp_write_link(catalog, MMSS_DP_AUDIO_CFG, audio_ctrl);
+	/* make sure audio engine is disabled */
+	wmb();
+}
+
+void dp_catalog_audio_config_sdp(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	u32 sdp_cfg = 0;
+	u32 sdp_cfg2 = 0;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	sdp_cfg = dp_read_link(catalog, MMSS_DP_SDP_CFG);
+	/* AUDIO_TIMESTAMP_SDP_EN */
+	sdp_cfg |= BIT(1);
+	/* AUDIO_STREAM_SDP_EN */
+	sdp_cfg |= BIT(2);
+	/* AUDIO_COPY_MANAGEMENT_SDP_EN */
+	sdp_cfg |= BIT(5);
+	/* AUDIO_ISRC_SDP_EN  */
+	sdp_cfg |= BIT(6);
+	/* AUDIO_INFOFRAME_SDP_EN  */
+	sdp_cfg |= BIT(20);
+
+	DRM_DEBUG_DP("sdp_cfg = 0x%x\n", sdp_cfg);
+
+	dp_write_link(catalog, MMSS_DP_SDP_CFG, sdp_cfg);
+
+	sdp_cfg2 = dp_read_link(catalog, MMSS_DP_SDP_CFG2);
+	/* IFRM_REGSRC -> Do not use reg values */
+	sdp_cfg2 &= ~BIT(0);
+	/* AUDIO_STREAM_HB3_REGSRC-> Do not use reg values */
+	sdp_cfg2 &= ~BIT(1);
+
+	DRM_DEBUG_DP("sdp_cfg2 = 0x%x\n", sdp_cfg2);
+
+	dp_write_link(catalog, MMSS_DP_SDP_CFG2, sdp_cfg2);
+}
+
+void dp_catalog_audio_init(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+
+	static u32 sdp_map[][DP_AUDIO_SDP_HEADER_MAX] = {
+		{
+			MMSS_DP_AUDIO_STREAM_0,
+			MMSS_DP_AUDIO_STREAM_1,
+			MMSS_DP_AUDIO_STREAM_1,
+		},
+		{
+			MMSS_DP_AUDIO_TIMESTAMP_0,
+			MMSS_DP_AUDIO_TIMESTAMP_1,
+			MMSS_DP_AUDIO_TIMESTAMP_1,
+		},
+		{
+			MMSS_DP_AUDIO_INFOFRAME_0,
+			MMSS_DP_AUDIO_INFOFRAME_1,
+			MMSS_DP_AUDIO_INFOFRAME_1,
+		},
+		{
+			MMSS_DP_AUDIO_COPYMANAGEMENT_0,
+			MMSS_DP_AUDIO_COPYMANAGEMENT_1,
+			MMSS_DP_AUDIO_COPYMANAGEMENT_1,
+		},
+		{
+			MMSS_DP_AUDIO_ISRC_0,
+			MMSS_DP_AUDIO_ISRC_1,
+			MMSS_DP_AUDIO_ISRC_1,
+		},
+	};
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	catalog->audio_map = sdp_map;
+}
+
+void dp_catalog_audio_sfe_level(struct dp_catalog *dp_catalog)
+{
+	struct dp_catalog_private *catalog;
+	u32 mainlink_levels, safe_to_exit_level;
+
+	if (!dp_catalog)
+		return;
+
+	catalog = container_of(dp_catalog,
+		struct dp_catalog_private, dp_catalog);
+
+	safe_to_exit_level = dp_catalog->audio_data;
+	mainlink_levels = dp_read_link(catalog, REG_DP_MAINLINK_LEVELS);
+	mainlink_levels &= 0xFE0;
+	mainlink_levels |= safe_to_exit_level;
+
+	DRM_DEBUG_DP("mainlink_level = 0x%x, safe_to_exit_level = 0x%x\n",
+			 mainlink_levels, safe_to_exit_level);
+
+	dp_write_link(catalog, REG_DP_MAINLINK_LEVELS, mainlink_levels);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.h b/drivers/gpu/drm/msm/dp/dp_catalog.h
new file mode 100644
index 000000000000..4b7666f1fe6f
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_catalog.h
@@ -0,0 +1,131 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_CATALOG_H_
+#define _DP_CATALOG_H_
+
+#include <drm/drm_modes.h>
+
+#include "dp_parser.h"
+
+/* interrupts */
+#define DP_INTR_HPD		BIT(0)
+#define DP_INTR_AUX_I2C_DONE	BIT(3)
+#define DP_INTR_WRONG_ADDR	BIT(6)
+#define DP_INTR_TIMEOUT		BIT(9)
+#define DP_INTR_NACK_DEFER	BIT(12)
+#define DP_INTR_WRONG_DATA_CNT	BIT(15)
+#define DP_INTR_I2C_NACK	BIT(18)
+#define DP_INTR_I2C_DEFER	BIT(21)
+#define DP_INTR_PLL_UNLOCKED	BIT(24)
+#define DP_INTR_AUX_ERROR	BIT(27)
+
+#define DP_INTR_READY_FOR_VIDEO		BIT(0)
+#define DP_INTR_IDLE_PATTERN_SENT	BIT(3)
+#define DP_INTR_FRAME_END		BIT(6)
+#define DP_INTR_CRC_UPDATED		BIT(9)
+
+#define DP_AUX_CFG_MAX_VALUE_CNT 3
+
+/* PHY AUX config registers */
+enum dp_phy_aux_config_type {
+	PHY_AUX_CFG0,
+	PHY_AUX_CFG1,
+	PHY_AUX_CFG2,
+	PHY_AUX_CFG3,
+	PHY_AUX_CFG4,
+	PHY_AUX_CFG5,
+	PHY_AUX_CFG6,
+	PHY_AUX_CFG7,
+	PHY_AUX_CFG8,
+	PHY_AUX_CFG9,
+	PHY_AUX_CFG_MAX,
+};
+
+enum dp_catalog_audio_sdp_type {
+	DP_AUDIO_SDP_STREAM,
+	DP_AUDIO_SDP_TIMESTAMP,
+	DP_AUDIO_SDP_INFOFRAME,
+	DP_AUDIO_SDP_COPYMANAGEMENT,
+	DP_AUDIO_SDP_ISRC,
+	DP_AUDIO_SDP_MAX,
+};
+
+enum dp_catalog_audio_header_type {
+	DP_AUDIO_SDP_HEADER_1,
+	DP_AUDIO_SDP_HEADER_2,
+	DP_AUDIO_SDP_HEADER_3,
+	DP_AUDIO_SDP_HEADER_MAX,
+};
+
+struct dp_catalog {
+	u32 aux_data;
+	u32 total;
+	u32 sync_start;
+	u32 width_blanking;
+	u32 dp_active;
+	enum dp_catalog_audio_sdp_type sdp_type;
+	enum dp_catalog_audio_header_type sdp_header;
+	u32 audio_data;
+};
+
+/* AUX APIs */
+u32 dp_catalog_aux_read_data(struct dp_catalog *dp_catalog);
+int dp_catalog_aux_write_data(struct dp_catalog *dp_catalog);
+int dp_catalog_aux_write_trans(struct dp_catalog *dp_catalog);
+int dp_catalog_aux_clear_trans(struct dp_catalog *dp_catalog, bool read);
+int dp_catalog_aux_clear_hw_interrupts(struct dp_catalog *dp_catalog);
+void dp_catalog_aux_reset(struct dp_catalog *dp_catalog);
+void dp_catalog_aux_enable(struct dp_catalog *dp_catalog, bool enable);
+void dp_catalog_aux_update_cfg(struct dp_catalog *dp_catalog);
+int dp_catalog_aux_get_irq(struct dp_catalog *dp_catalog);
+
+/* DP Controller APIs */
+void dp_catalog_ctrl_state_ctrl(struct dp_catalog *dp_catalog, u32 state);
+void dp_catalog_ctrl_config_ctrl(struct dp_catalog *dp_catalog, u32 config);
+void dp_catalog_ctrl_lane_mapping(struct dp_catalog *dp_catalog);
+void dp_catalog_ctrl_mainlink_ctrl(struct dp_catalog *dp_catalog, bool enable);
+void dp_catalog_ctrl_config_misc(struct dp_catalog *dp_catalog, u32 cc, u32 tb);
+void dp_catalog_ctrl_config_msa(struct dp_catalog *dp_catalog, u32 rate,
+				u32 stream_rate_khz, bool fixed_nvid);
+int dp_catalog_ctrl_set_pattern(struct dp_catalog *dp_catalog, u32 pattern);
+void dp_catalog_ctrl_reset(struct dp_catalog *dp_catalog);
+bool dp_catalog_ctrl_mainlink_ready(struct dp_catalog *dp_catalog);
+void dp_catalog_ctrl_enable_irq(struct dp_catalog *dp_catalog, bool enable);
+void dp_catalog_hpd_config_intr(struct dp_catalog *dp_catalog,
+			u32 intr_mask, bool en);
+void dp_catalog_ctrl_hpd_config(struct dp_catalog *dp_catalog);
+u32 dp_catalog_hpd_get_intr_status(struct dp_catalog *dp_catalog);
+void dp_catalog_ctrl_phy_reset(struct dp_catalog *dp_catalog);
+int dp_catalog_ctrl_update_vx_px(struct dp_catalog *dp_catalog, u8 v_level,
+				u8 p_level);
+int dp_catalog_ctrl_get_interrupt(struct dp_catalog *dp_catalog);
+void dp_catalog_ctrl_update_transfer_unit(struct dp_catalog *dp_catalog,
+				u32 dp_tu, u32 valid_boundary,
+				u32 valid_boundary2);
+void dp_catalog_ctrl_send_phy_pattern(struct dp_catalog *dp_catalog,
+				u32 pattern);
+u32 dp_catalog_ctrl_read_phy_pattern(struct dp_catalog *dp_catalog);
+
+/* DP Panel APIs */
+int dp_catalog_panel_timing_cfg(struct dp_catalog *dp_catalog);
+void dp_catalog_dump_regs(struct dp_catalog *dp_catalog);
+void dp_catalog_panel_tpg_enable(struct dp_catalog *dp_catalog,
+				struct drm_display_mode *drm_mode);
+void dp_catalog_panel_tpg_disable(struct dp_catalog *dp_catalog);
+
+struct dp_catalog *dp_catalog_get(struct device *dev, struct dp_io *io);
+
+/* DP Audio APIs */
+void dp_catalog_audio_get_header(struct dp_catalog *catalog);
+void dp_catalog_audio_set_header(struct dp_catalog *catalog);
+void dp_catalog_audio_config_acr(struct dp_catalog *catalog);
+void dp_catalog_audio_enable(struct dp_catalog *catalog);
+void dp_catalog_audio_enable(struct dp_catalog *catalog);
+void dp_catalog_audio_config_sdp(struct dp_catalog *catalog);
+void dp_catalog_audio_init(struct dp_catalog *catalog);
+void dp_catalog_audio_sfe_level(struct dp_catalog *catalog);
+
+#endif /* _DP_CATALOG_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c
new file mode 100644
index 000000000000..2e3e1917351f
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c
@@ -0,0 +1,1869 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <linux/types.h>
+#include <linux/completion.h>
+#include <linux/delay.h>
+#include <linux/phy/phy.h>
+#include <linux/phy/phy-dp.h>
+#include <drm/drm_fixed.h>
+#include <drm/drm_dp_helper.h>
+#include <drm/drm_print.h>
+
+#include "dp_reg.h"
+#include "dp_ctrl.h"
+#include "dp_link.h"
+
+#define DP_KHZ_TO_HZ 1000
+#define IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES	(30 * HZ / 1000) /* 30 ms */
+#define WAIT_FOR_VIDEO_READY_TIMEOUT_JIFFIES (HZ / 2)
+
+#define DP_CTRL_INTR_READY_FOR_VIDEO     BIT(0)
+#define DP_CTRL_INTR_IDLE_PATTERN_SENT  BIT(3)
+
+#define MR_LINK_TRAINING1  0x8
+#define MR_LINK_SYMBOL_ERM 0x80
+#define MR_LINK_PRBS7 0x100
+#define MR_LINK_CUSTOM80 0x200
+#define MR_LINK_TRAINING4  0x40
+
+enum {
+	DP_TRAINING_NONE,
+	DP_TRAINING_1,
+	DP_TRAINING_2,
+};
+
+struct dp_tu_calc_input {
+	u64 lclk;        /* 162, 270, 540 and 810 */
+	u64 pclk_khz;    /* in KHz */
+	u64 hactive;     /* active h-width */
+	u64 hporch;      /* bp + fp + pulse */
+	int nlanes;      /* no.of.lanes */
+	int bpp;         /* bits */
+	int pixel_enc;   /* 444, 420, 422 */
+	int dsc_en;     /* dsc on/off */
+	int async_en;   /* async mode */
+	int fec_en;     /* fec */
+	int compress_ratio; /* 2:1 = 200, 3:1 = 300, 3.75:1 = 375 */
+	int num_of_dsc_slices; /* number of slices per line */
+};
+
+struct dp_vc_tu_mapping_table {
+	u32 vic;
+	u8 lanes;
+	u8 lrate; /* DP_LINK_RATE -> 162(6), 270(10), 540(20), 810 (30) */
+	u8 bpp;
+	u8 valid_boundary_link;
+	u16 delay_start_link;
+	bool boundary_moderation_en;
+	u8 valid_lower_boundary_link;
+	u8 upper_boundary_count;
+	u8 lower_boundary_count;
+	u8 tu_size_minus1;
+};
+
+struct dp_ctrl_private {
+	struct dp_ctrl dp_ctrl;
+	struct device *dev;
+	struct drm_dp_aux *aux;
+	struct dp_panel *panel;
+	struct dp_link *link;
+	struct dp_power *power;
+	struct dp_parser *parser;
+	struct dp_catalog *catalog;
+
+	struct completion idle_comp;
+	struct completion video_comp;
+};
+
+struct dp_cr_status {
+	u8 lane_0_1;
+	u8 lane_2_3;
+};
+
+#define DP_LANE0_1_CR_DONE	0x11
+
+static int dp_aux_link_configure(struct drm_dp_aux *aux,
+					struct dp_link_info *link)
+{
+	u8 values[2];
+	int err;
+
+	values[0] = drm_dp_link_rate_to_bw_code(link->rate);
+	values[1] = link->num_lanes;
+
+	if (link->capabilities & DP_LINK_CAP_ENHANCED_FRAMING)
+		values[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+
+	err = drm_dp_dpcd_write(aux, DP_LINK_BW_SET, values, sizeof(values));
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+void dp_ctrl_push_idle(struct dp_ctrl *dp_ctrl)
+{
+	struct dp_ctrl_private *ctrl;
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+
+	reinit_completion(&ctrl->idle_comp);
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_PUSH_IDLE);
+
+	if (!wait_for_completion_timeout(&ctrl->idle_comp,
+			IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES))
+		pr_warn("PUSH_IDLE pattern timedout\n");
+
+	pr_debug("mainlink off done\n");
+}
+
+static void dp_ctrl_config_ctrl(struct dp_ctrl_private *ctrl)
+{
+	u32 config = 0, tbd;
+	u8 *dpcd = ctrl->panel->dpcd;
+
+	/* Default-> LSCLK DIV: 1/4 LCLK  */
+	config |= (2 << DP_CONFIGURATION_CTRL_LSCLK_DIV_SHIFT);
+
+	/* Scrambler reset enable */
+	if (dpcd[DP_EDP_CONFIGURATION_CAP] & DP_ALTERNATE_SCRAMBLER_RESET_CAP)
+		config |= DP_CONFIGURATION_CTRL_ASSR;
+
+	tbd = dp_link_get_test_bits_depth(ctrl->link,
+			ctrl->panel->dp_mode.bpp);
+
+	if (tbd == DP_TEST_BIT_DEPTH_UNKNOWN) {
+		pr_debug("BIT_DEPTH not set. Configure default\n");
+		tbd = DP_TEST_BIT_DEPTH_8;
+	}
+
+	config |= tbd << DP_CONFIGURATION_CTRL_BPC_SHIFT;
+
+	/* Num of Lanes */
+	config |= ((ctrl->link->link_params.num_lanes - 1)
+			<< DP_CONFIGURATION_CTRL_NUM_OF_LANES_SHIFT);
+
+	if (drm_dp_enhanced_frame_cap(dpcd))
+		config |= DP_CONFIGURATION_CTRL_ENHANCED_FRAMING;
+
+	config |= DP_CONFIGURATION_CTRL_P_INTERLACED; /* progressive video */
+
+	/* sync clock & static Mvid */
+	config |= DP_CONFIGURATION_CTRL_STATIC_DYNAMIC_CN;
+	config |= DP_CONFIGURATION_CTRL_SYNC_ASYNC_CLK;
+
+	dp_catalog_ctrl_config_ctrl(ctrl->catalog, config);
+}
+
+static void dp_ctrl_configure_source_params(struct dp_ctrl_private *ctrl)
+{
+	u32 cc, tb;
+
+	dp_catalog_ctrl_lane_mapping(ctrl->catalog);
+	dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, true);
+
+	dp_ctrl_config_ctrl(ctrl);
+
+	tb = dp_link_get_test_bits_depth(ctrl->link,
+		ctrl->panel->dp_mode.bpp);
+	cc = dp_link_get_colorimetry_config(ctrl->link);
+	dp_catalog_ctrl_config_misc(ctrl->catalog, cc, tb);
+	dp_panel_timing_cfg(ctrl->panel);
+}
+
+/*
+ * The structure and few functions present below are IP/Hardware
+ * specific implementation. Most of the implementation will not
+ * have coding comments
+ */
+struct tu_algo_data {
+	s64 lclk_fp;
+	s64 pclk_fp;
+	s64 lwidth;
+	s64 lwidth_fp;
+	s64 hbp_relative_to_pclk;
+	s64 hbp_relative_to_pclk_fp;
+	int nlanes;
+	int bpp;
+	int pixelEnc;
+	int dsc_en;
+	int async_en;
+	int bpc;
+
+	uint delay_start_link_extra_pixclk;
+	int extra_buffer_margin;
+	s64 ratio_fp;
+	s64 original_ratio_fp;
+
+	s64 err_fp;
+	s64 n_err_fp;
+	s64 n_n_err_fp;
+	int tu_size;
+	int tu_size_desired;
+	int tu_size_minus1;
+
+	int valid_boundary_link;
+	s64 resulting_valid_fp;
+	s64 total_valid_fp;
+	s64 effective_valid_fp;
+	s64 effective_valid_recorded_fp;
+	int n_tus;
+	int n_tus_per_lane;
+	int paired_tus;
+	int remainder_tus;
+	int remainder_tus_upper;
+	int remainder_tus_lower;
+	int extra_bytes;
+	int filler_size;
+	int delay_start_link;
+
+	int extra_pclk_cycles;
+	int extra_pclk_cycles_in_link_clk;
+	s64 ratio_by_tu_fp;
+	s64 average_valid2_fp;
+	int new_valid_boundary_link;
+	int remainder_symbols_exist;
+	int n_symbols;
+	s64 n_remainder_symbols_per_lane_fp;
+	s64 last_partial_tu_fp;
+	s64 TU_ratio_err_fp;
+
+	int n_tus_incl_last_incomplete_tu;
+	int extra_pclk_cycles_tmp;
+	int extra_pclk_cycles_in_link_clk_tmp;
+	int extra_required_bytes_new_tmp;
+	int filler_size_tmp;
+	int lower_filler_size_tmp;
+	int delay_start_link_tmp;
+
+	bool boundary_moderation_en;
+	int boundary_mod_lower_err;
+	int upper_boundary_count;
+	int lower_boundary_count;
+	int i_upper_boundary_count;
+	int i_lower_boundary_count;
+	int valid_lower_boundary_link;
+	int even_distribution_BF;
+	int even_distribution_legacy;
+	int even_distribution;
+	int min_hblank_violated;
+	s64 delay_start_time_fp;
+	s64 hbp_time_fp;
+	s64 hactive_time_fp;
+	s64 diff_abs_fp;
+
+	s64 ratio;
+};
+
+static int _tu_param_compare(s64 a, s64 b)
+{
+	u32 a_sign;
+	u32 b_sign;
+	s64 a_temp, b_temp, minus_1;
+
+	if (a == b)
+		return 0;
+
+	minus_1 = drm_fixp_from_fraction(-1, 1);
+
+	a_sign = (a >> 32) & 0x80000000 ? 1 : 0;
+
+	b_sign = (b >> 32) & 0x80000000 ? 1 : 0;
+
+	if (a_sign > b_sign)
+		return 2;
+	else if (b_sign > a_sign)
+		return 1;
+
+	if (!a_sign && !b_sign) { /* positive */
+		if (a > b)
+			return 1;
+		else
+			return 2;
+	} else { /* negative */
+		a_temp = drm_fixp_mul(a, minus_1);
+		b_temp = drm_fixp_mul(b, minus_1);
+
+		if (a_temp > b_temp)
+			return 2;
+		else
+			return 1;
+	}
+}
+
+static void dp_panel_update_tu_timings(struct dp_tu_calc_input *in,
+					struct tu_algo_data *tu)
+{
+	int nlanes = in->nlanes;
+	int dsc_num_slices = in->num_of_dsc_slices;
+	int dsc_num_bytes  = 0;
+	int numerator;
+	s64 pclk_dsc_fp;
+	s64 dwidth_dsc_fp;
+	s64 hbp_dsc_fp;
+
+	int tot_num_eoc_symbols = 0;
+	int tot_num_hor_bytes   = 0;
+	int tot_num_dummy_bytes = 0;
+	int dwidth_dsc_bytes    = 0;
+	int  eoc_bytes           = 0;
+
+	s64 temp1_fp, temp2_fp, temp3_fp;
+
+	tu->lclk_fp              = drm_fixp_from_fraction(in->lclk, 1);
+	tu->pclk_fp              = drm_fixp_from_fraction(in->pclk_khz, 1000);
+	tu->lwidth               = in->hactive;
+	tu->hbp_relative_to_pclk = in->hporch;
+	tu->nlanes               = in->nlanes;
+	tu->bpp                  = in->bpp;
+	tu->pixelEnc             = in->pixel_enc;
+	tu->dsc_en               = in->dsc_en;
+	tu->async_en             = in->async_en;
+	tu->lwidth_fp            = drm_fixp_from_fraction(in->hactive, 1);
+	tu->hbp_relative_to_pclk_fp = drm_fixp_from_fraction(in->hporch, 1);
+
+	if (tu->pixelEnc == 420) {
+		temp1_fp = drm_fixp_from_fraction(2, 1);
+		tu->pclk_fp = drm_fixp_div(tu->pclk_fp, temp1_fp);
+		tu->lwidth_fp = drm_fixp_div(tu->lwidth_fp, temp1_fp);
+		tu->hbp_relative_to_pclk_fp =
+				drm_fixp_div(tu->hbp_relative_to_pclk_fp, 2);
+	}
+
+	if (tu->pixelEnc == 422) {
+		switch (tu->bpp) {
+		case 24:
+			tu->bpp = 16;
+			tu->bpc = 8;
+			break;
+		case 30:
+			tu->bpp = 20;
+			tu->bpc = 10;
+			break;
+		default:
+			tu->bpp = 16;
+			tu->bpc = 8;
+			break;
+		}
+	} else {
+		tu->bpc = tu->bpp/3;
+	}
+
+	if (!in->dsc_en)
+		goto fec_check;
+
+	temp1_fp = drm_fixp_from_fraction(in->compress_ratio, 100);
+	temp2_fp = drm_fixp_from_fraction(in->bpp, 1);
+	temp3_fp = drm_fixp_div(temp2_fp, temp1_fp);
+	temp2_fp = drm_fixp_mul(tu->lwidth_fp, temp3_fp);
+
+	temp1_fp = drm_fixp_from_fraction(8, 1);
+	temp3_fp = drm_fixp_div(temp2_fp, temp1_fp);
+
+	numerator = drm_fixp2int(temp3_fp);
+
+	dsc_num_bytes  = numerator / dsc_num_slices;
+	eoc_bytes           = dsc_num_bytes % nlanes;
+	tot_num_eoc_symbols = nlanes * dsc_num_slices;
+	tot_num_hor_bytes   = dsc_num_bytes * dsc_num_slices;
+	tot_num_dummy_bytes = (nlanes - eoc_bytes) * dsc_num_slices;
+
+	if (dsc_num_bytes == 0)
+		pr_info("incorrect no of bytes per slice=%d\n", dsc_num_bytes);
+
+	dwidth_dsc_bytes = (tot_num_hor_bytes +
+				tot_num_eoc_symbols +
+				(eoc_bytes == 0 ? 0 : tot_num_dummy_bytes));
+
+	dwidth_dsc_fp = drm_fixp_from_fraction(dwidth_dsc_bytes, 3);
+
+	temp2_fp = drm_fixp_mul(tu->pclk_fp, dwidth_dsc_fp);
+	temp1_fp = drm_fixp_div(temp2_fp, tu->lwidth_fp);
+	pclk_dsc_fp = temp1_fp;
+
+	temp1_fp = drm_fixp_div(pclk_dsc_fp, tu->pclk_fp);
+	temp2_fp = drm_fixp_mul(tu->hbp_relative_to_pclk_fp, temp1_fp);
+	hbp_dsc_fp = temp2_fp;
+
+	/* output */
+	tu->pclk_fp = pclk_dsc_fp;
+	tu->lwidth_fp = dwidth_dsc_fp;
+	tu->hbp_relative_to_pclk_fp = hbp_dsc_fp;
+
+fec_check:
+	if (in->fec_en) {
+		temp1_fp = drm_fixp_from_fraction(976, 1000); /* 0.976 */
+		tu->lclk_fp = drm_fixp_mul(tu->lclk_fp, temp1_fp);
+	}
+}
+
+static void _tu_valid_boundary_calc(struct tu_algo_data *tu)
+{
+	s64 temp1_fp, temp2_fp, temp, temp1, temp2;
+	int compare_result_1, compare_result_2, compare_result_3;
+
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp);
+
+	tu->new_valid_boundary_link = drm_fixp2int_ceil(temp2_fp);
+
+	temp = (tu->i_upper_boundary_count *
+				tu->new_valid_boundary_link +
+				tu->i_lower_boundary_count *
+				(tu->new_valid_boundary_link-1));
+	tu->average_valid2_fp = drm_fixp_from_fraction(temp,
+					(tu->i_upper_boundary_count +
+					tu->i_lower_boundary_count));
+
+	temp1_fp = drm_fixp_from_fraction(tu->bpp, 8);
+	temp2_fp = tu->lwidth_fp;
+	temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+	temp2_fp = drm_fixp_div(temp1_fp, tu->average_valid2_fp);
+	tu->n_tus = drm_fixp2int(temp2_fp);
+	if ((temp2_fp & 0xFFFFFFFF) > 0xFFFFF000)
+		tu->n_tus += 1;
+
+	temp1_fp = drm_fixp_from_fraction(tu->n_tus, 1);
+	temp2_fp = drm_fixp_mul(temp1_fp, tu->average_valid2_fp);
+	temp1_fp = drm_fixp_from_fraction(tu->n_symbols, 1);
+	temp2_fp = temp1_fp - temp2_fp;
+	temp1_fp = drm_fixp_from_fraction(tu->nlanes, 1);
+	temp2_fp = drm_fixp_div(temp2_fp, temp1_fp);
+	tu->n_remainder_symbols_per_lane_fp = temp2_fp;
+
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	tu->last_partial_tu_fp =
+			drm_fixp_div(tu->n_remainder_symbols_per_lane_fp,
+					temp1_fp);
+
+	if (tu->n_remainder_symbols_per_lane_fp != 0)
+		tu->remainder_symbols_exist = 1;
+	else
+		tu->remainder_symbols_exist = 0;
+
+	temp1_fp = drm_fixp_from_fraction(tu->n_tus, tu->nlanes);
+	tu->n_tus_per_lane = drm_fixp2int(temp1_fp);
+
+	tu->paired_tus = (int)((tu->n_tus_per_lane) /
+					(tu->i_upper_boundary_count +
+					 tu->i_lower_boundary_count));
+
+	tu->remainder_tus = tu->n_tus_per_lane - tu->paired_tus *
+						(tu->i_upper_boundary_count +
+						tu->i_lower_boundary_count);
+
+	if ((tu->remainder_tus - tu->i_upper_boundary_count) > 0) {
+		tu->remainder_tus_upper = tu->i_upper_boundary_count;
+		tu->remainder_tus_lower = tu->remainder_tus -
+						tu->i_upper_boundary_count;
+	} else {
+		tu->remainder_tus_upper = tu->remainder_tus;
+		tu->remainder_tus_lower = 0;
+	}
+
+	temp = tu->paired_tus * (tu->i_upper_boundary_count *
+				tu->new_valid_boundary_link +
+				tu->i_lower_boundary_count *
+				(tu->new_valid_boundary_link - 1)) +
+				(tu->remainder_tus_upper *
+				 tu->new_valid_boundary_link) +
+				(tu->remainder_tus_lower *
+				(tu->new_valid_boundary_link - 1));
+	tu->total_valid_fp = drm_fixp_from_fraction(temp, 1);
+
+	if (tu->remainder_symbols_exist) {
+		temp1_fp = tu->total_valid_fp +
+				tu->n_remainder_symbols_per_lane_fp;
+		temp2_fp = drm_fixp_from_fraction(tu->n_tus_per_lane, 1);
+		temp2_fp = temp2_fp + tu->last_partial_tu_fp;
+		temp1_fp = drm_fixp_div(temp1_fp, temp2_fp);
+	} else {
+		temp2_fp = drm_fixp_from_fraction(tu->n_tus_per_lane, 1);
+		temp1_fp = drm_fixp_div(tu->total_valid_fp, temp2_fp);
+	}
+	tu->effective_valid_fp = temp1_fp;
+
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp);
+	tu->n_n_err_fp = tu->effective_valid_fp - temp2_fp;
+
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp);
+	tu->n_err_fp = tu->average_valid2_fp - temp2_fp;
+
+	tu->even_distribution = tu->n_tus % tu->nlanes == 0 ? 1 : 0;
+
+	temp1_fp = drm_fixp_from_fraction(tu->bpp, 8);
+	temp2_fp = tu->lwidth_fp;
+	temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+	temp2_fp = drm_fixp_div(temp1_fp, tu->average_valid2_fp);
+
+	if (temp2_fp)
+		tu->n_tus_incl_last_incomplete_tu = drm_fixp2int_ceil(temp2_fp);
+	else
+		tu->n_tus_incl_last_incomplete_tu = 0;
+
+	temp1 = 0;
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp);
+	temp1_fp = tu->average_valid2_fp - temp2_fp;
+	temp2_fp = drm_fixp_from_fraction(tu->n_tus_incl_last_incomplete_tu, 1);
+	temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+
+	if (temp1_fp)
+		temp1 = drm_fixp2int_ceil(temp1_fp);
+
+	temp = tu->i_upper_boundary_count * tu->nlanes;
+	temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1);
+	temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp);
+	temp1_fp = drm_fixp_from_fraction(tu->new_valid_boundary_link, 1);
+	temp2_fp = temp1_fp - temp2_fp;
+	temp1_fp = drm_fixp_from_fraction(temp, 1);
+	temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp);
+
+	if (temp2_fp)
+		temp2 = drm_fixp2int_ceil(temp2_fp);
+	else
+		temp2 = 0;
+	tu->extra_required_bytes_new_tmp = (int)(temp1 + temp2);
+
+	temp1_fp = drm_fixp_from_fraction(8, tu->bpp);
+	temp2_fp = drm_fixp_from_fraction(
+	tu->extra_required_bytes_new_tmp, 1);
+	temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+
+	if (temp1_fp)
+		tu->extra_pclk_cycles_tmp = drm_fixp2int_ceil(temp1_fp);
+	else
+		tu->extra_pclk_cycles_tmp = 0;
+
+	temp1_fp = drm_fixp_from_fraction(tu->extra_pclk_cycles_tmp, 1);
+	temp2_fp = drm_fixp_div(tu->lclk_fp, tu->pclk_fp);
+	temp1_fp = drm_fixp_mul(temp1_fp, temp2_fp);
+
+	if (temp1_fp)
+		tu->extra_pclk_cycles_in_link_clk_tmp =
+						drm_fixp2int_ceil(temp1_fp);
+	else
+		tu->extra_pclk_cycles_in_link_clk_tmp = 0;
+
+	tu->filler_size_tmp = tu->tu_size - tu->new_valid_boundary_link;
+
+	tu->lower_filler_size_tmp = tu->filler_size_tmp + 1;
+
+	tu->delay_start_link_tmp = tu->extra_pclk_cycles_in_link_clk_tmp +
+					tu->lower_filler_size_tmp +
+					tu->extra_buffer_margin;
+
+	temp1_fp = drm_fixp_from_fraction(tu->delay_start_link_tmp, 1);
+	tu->delay_start_time_fp = drm_fixp_div(temp1_fp, tu->lclk_fp);
+
+	compare_result_1 = _tu_param_compare(tu->n_n_err_fp, tu->diff_abs_fp);
+	if (compare_result_1 == 2)
+		compare_result_1 = 1;
+	else
+		compare_result_1 = 0;
+
+	compare_result_2 = _tu_param_compare(tu->n_n_err_fp, tu->err_fp);
+	if (compare_result_2 == 2)
+		compare_result_2 = 1;
+	else
+		compare_result_2 = 0;
+
+	compare_result_3 = _tu_param_compare(tu->hbp_time_fp,
+					tu->delay_start_time_fp);
+	if (compare_result_3 == 2)
+		compare_result_3 = 0;
+	else
+		compare_result_3 = 1;
+
+	if (((tu->even_distribution == 1) ||
+			((tu->even_distribution_BF == 0) &&
+			(tu->even_distribution_legacy == 0))) &&
+			tu->n_err_fp >= 0 && tu->n_n_err_fp >= 0 &&
+			compare_result_2 &&
+			(compare_result_1 || (tu->min_hblank_violated == 1)) &&
+			(tu->new_valid_boundary_link - 1) > 0 &&
+			compare_result_3 &&
+			(tu->delay_start_link_tmp <= 1023)) {
+		tu->upper_boundary_count = tu->i_upper_boundary_count;
+		tu->lower_boundary_count = tu->i_lower_boundary_count;
+		tu->err_fp = tu->n_n_err_fp;
+		tu->boundary_moderation_en = true;
+		tu->tu_size_desired = tu->tu_size;
+		tu->valid_boundary_link = tu->new_valid_boundary_link;
+		tu->effective_valid_recorded_fp = tu->effective_valid_fp;
+		tu->even_distribution_BF = 1;
+		tu->delay_start_link = tu->delay_start_link_tmp;
+	} else if (tu->boundary_mod_lower_err == 0) {
+		compare_result_1 = _tu_param_compare(tu->n_n_err_fp,
+							tu->diff_abs_fp);
+		if (compare_result_1 == 2)
+			tu->boundary_mod_lower_err = 1;
+	}
+}
+
+static void _dp_ctrl_calc_tu(struct dp_tu_calc_input *in,
+				   struct dp_vc_tu_mapping_table *tu_table)
+{
+	struct tu_algo_data tu;
+	int compare_result_1, compare_result_2;
+	u64 temp = 0;
+	s64 temp_fp = 0, temp1_fp = 0, temp2_fp = 0;
+
+	s64 LCLK_FAST_SKEW_fp = drm_fixp_from_fraction(6, 10000); /* 0.0006 */
+	s64 const_p49_fp = drm_fixp_from_fraction(49, 100); /* 0.49 */
+	s64 const_p56_fp = drm_fixp_from_fraction(56, 100); /* 0.56 */
+	s64 RATIO_SCALE_fp = drm_fixp_from_fraction(1001, 1000);
+
+	u8 DP_BRUTE_FORCE = 1;
+	s64 BRUTE_FORCE_THRESHOLD_fp = drm_fixp_from_fraction(1, 10); /* 0.1 */
+	uint EXTRA_PIXCLK_CYCLE_DELAY = 4;
+	uint HBLANK_MARGIN = 4;
+
+	memset(&tu, 0, sizeof(tu));
+
+	dp_panel_update_tu_timings(in, &tu);
+
+	tu.err_fp = drm_fixp_from_fraction(1000, 1); /* 1000 */
+
+	temp1_fp = drm_fixp_from_fraction(4, 1);
+	temp2_fp = drm_fixp_mul(temp1_fp, tu.lclk_fp);
+	temp_fp = drm_fixp_div(temp2_fp, tu.pclk_fp);
+	tu.extra_buffer_margin = drm_fixp2int_ceil(temp_fp);
+
+	temp1_fp = drm_fixp_from_fraction(tu.bpp, 8);
+	temp2_fp = drm_fixp_mul(tu.pclk_fp, temp1_fp);
+	temp1_fp = drm_fixp_from_fraction(tu.nlanes, 1);
+	temp2_fp = drm_fixp_div(temp2_fp, temp1_fp);
+	tu.ratio_fp = drm_fixp_div(temp2_fp, tu.lclk_fp);
+
+	tu.original_ratio_fp = tu.ratio_fp;
+	tu.boundary_moderation_en = false;
+	tu.upper_boundary_count = 0;
+	tu.lower_boundary_count = 0;
+	tu.i_upper_boundary_count = 0;
+	tu.i_lower_boundary_count = 0;
+	tu.valid_lower_boundary_link = 0;
+	tu.even_distribution_BF = 0;
+	tu.even_distribution_legacy = 0;
+	tu.even_distribution = 0;
+	tu.delay_start_time_fp = 0;
+
+	tu.err_fp = drm_fixp_from_fraction(1000, 1);
+	tu.n_err_fp = 0;
+	tu.n_n_err_fp = 0;
+
+	tu.ratio = drm_fixp2int(tu.ratio_fp);
+	temp1_fp = drm_fixp_from_fraction(tu.nlanes, 1);
+	div64_u64_rem(tu.lwidth_fp, temp1_fp, &temp2_fp);
+	if (temp2_fp != 0 &&
+			!tu.ratio && tu.dsc_en == 0) {
+		tu.ratio_fp = drm_fixp_mul(tu.ratio_fp, RATIO_SCALE_fp);
+		tu.ratio = drm_fixp2int(tu.ratio_fp);
+		if (tu.ratio)
+			tu.ratio_fp = drm_fixp_from_fraction(1, 1);
+	}
+
+	if (tu.ratio > 1)
+		tu.ratio = 1;
+
+	if (tu.ratio == 1)
+		goto tu_size_calc;
+
+	compare_result_1 = _tu_param_compare(tu.ratio_fp, const_p49_fp);
+	if (!compare_result_1 || compare_result_1 == 1)
+		compare_result_1 = 1;
+	else
+		compare_result_1 = 0;
+
+	compare_result_2 = _tu_param_compare(tu.ratio_fp, const_p56_fp);
+	if (!compare_result_2 || compare_result_2 == 2)
+		compare_result_2 = 1;
+	else
+		compare_result_2 = 0;
+
+	if (tu.dsc_en && compare_result_1 && compare_result_2) {
+		HBLANK_MARGIN += 4;
+		DRM_DEBUG_DP("Info: increase HBLANK_MARGIN to %d\n",
+				HBLANK_MARGIN);
+	}
+
+tu_size_calc:
+	for (tu.tu_size = 32; tu.tu_size <= 64; tu.tu_size++) {
+		temp1_fp = drm_fixp_from_fraction(tu.tu_size, 1);
+		temp2_fp = drm_fixp_mul(tu.ratio_fp, temp1_fp);
+		temp = drm_fixp2int_ceil(temp2_fp);
+		temp1_fp = drm_fixp_from_fraction(temp, 1);
+		tu.n_err_fp = temp1_fp - temp2_fp;
+
+		if (tu.n_err_fp < tu.err_fp) {
+			tu.err_fp = tu.n_err_fp;
+			tu.tu_size_desired = tu.tu_size;
+		}
+	}
+
+	tu.tu_size_minus1 = tu.tu_size_desired - 1;
+
+	temp1_fp = drm_fixp_from_fraction(tu.tu_size_desired, 1);
+	temp2_fp = drm_fixp_mul(tu.ratio_fp, temp1_fp);
+	tu.valid_boundary_link = drm_fixp2int_ceil(temp2_fp);
+
+	temp1_fp = drm_fixp_from_fraction(tu.bpp, 8);
+	temp2_fp = tu.lwidth_fp;
+	temp2_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+
+	temp1_fp = drm_fixp_from_fraction(tu.valid_boundary_link, 1);
+	temp2_fp = drm_fixp_div(temp2_fp, temp1_fp);
+	tu.n_tus = drm_fixp2int(temp2_fp);
+	if ((temp2_fp & 0xFFFFFFFF) > 0xFFFFF000)
+		tu.n_tus += 1;
+
+	tu.even_distribution_legacy = tu.n_tus % tu.nlanes == 0 ? 1 : 0;
+	DRM_DEBUG_DP("Info: n_sym = %d, num_of_tus = %d\n",
+		tu.valid_boundary_link, tu.n_tus);
+
+	temp1_fp = drm_fixp_from_fraction(tu.tu_size_desired, 1);
+	temp2_fp = drm_fixp_mul(tu.original_ratio_fp, temp1_fp);
+	temp1_fp = drm_fixp_from_fraction(tu.valid_boundary_link, 1);
+	temp2_fp = temp1_fp - temp2_fp;
+	temp1_fp = drm_fixp_from_fraction(tu.n_tus + 1, 1);
+	temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp);
+
+	temp = drm_fixp2int(temp2_fp);
+	if (temp && temp2_fp)
+		tu.extra_bytes = drm_fixp2int_ceil(temp2_fp);
+	else
+		tu.extra_bytes = 0;
+
+	temp1_fp = drm_fixp_from_fraction(tu.extra_bytes, 1);
+	temp2_fp = drm_fixp_from_fraction(8, tu.bpp);
+	temp1_fp = drm_fixp_mul(temp1_fp, temp2_fp);
+
+	if (temp && temp1_fp)
+		tu.extra_pclk_cycles = drm_fixp2int_ceil(temp1_fp);
+	else
+		tu.extra_pclk_cycles = drm_fixp2int(temp1_fp);
+
+	temp1_fp = drm_fixp_div(tu.lclk_fp, tu.pclk_fp);
+	temp2_fp = drm_fixp_from_fraction(tu.extra_pclk_cycles, 1);
+	temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+
+	if (temp1_fp)
+		tu.extra_pclk_cycles_in_link_clk = drm_fixp2int_ceil(temp1_fp);
+	else
+		tu.extra_pclk_cycles_in_link_clk = drm_fixp2int(temp1_fp);
+
+	tu.filler_size = tu.tu_size_desired - tu.valid_boundary_link;
+
+	temp1_fp = drm_fixp_from_fraction(tu.tu_size_desired, 1);
+	tu.ratio_by_tu_fp = drm_fixp_mul(tu.ratio_fp, temp1_fp);
+
+	tu.delay_start_link = tu.extra_pclk_cycles_in_link_clk +
+				tu.filler_size + tu.extra_buffer_margin;
+
+	tu.resulting_valid_fp =
+			drm_fixp_from_fraction(tu.valid_boundary_link, 1);
+
+	temp1_fp = drm_fixp_from_fraction(tu.tu_size_desired, 1);
+	temp2_fp = drm_fixp_div(tu.resulting_valid_fp, temp1_fp);
+	tu.TU_ratio_err_fp = temp2_fp - tu.original_ratio_fp;
+
+	temp1_fp = drm_fixp_from_fraction(HBLANK_MARGIN, 1);
+	temp1_fp = tu.hbp_relative_to_pclk_fp - temp1_fp;
+	tu.hbp_time_fp = drm_fixp_div(temp1_fp, tu.pclk_fp);
+
+	temp1_fp = drm_fixp_from_fraction(tu.delay_start_link, 1);
+	tu.delay_start_time_fp = drm_fixp_div(temp1_fp, tu.lclk_fp);
+
+	compare_result_1 = _tu_param_compare(tu.hbp_time_fp,
+					tu.delay_start_time_fp);
+	if (compare_result_1 == 2) /* if (hbp_time_fp < delay_start_time_fp) */
+		tu.min_hblank_violated = 1;
+
+	tu.hactive_time_fp = drm_fixp_div(tu.lwidth_fp, tu.pclk_fp);
+
+	compare_result_2 = _tu_param_compare(tu.hactive_time_fp,
+						tu.delay_start_time_fp);
+	if (compare_result_2 == 2)
+		tu.min_hblank_violated = 1;
+
+	tu.delay_start_time_fp = 0;
+
+	/* brute force */
+
+	tu.delay_start_link_extra_pixclk = EXTRA_PIXCLK_CYCLE_DELAY;
+	tu.diff_abs_fp = tu.resulting_valid_fp - tu.ratio_by_tu_fp;
+
+	temp = drm_fixp2int(tu.diff_abs_fp);
+	if (!temp && tu.diff_abs_fp <= 0xffff)
+		tu.diff_abs_fp = 0;
+
+	/* if(diff_abs < 0) diff_abs *= -1 */
+	if (tu.diff_abs_fp < 0)
+		tu.diff_abs_fp = drm_fixp_mul(tu.diff_abs_fp, -1);
+
+	tu.boundary_mod_lower_err = 0;
+	if ((tu.diff_abs_fp != 0 &&
+			((tu.diff_abs_fp > BRUTE_FORCE_THRESHOLD_fp) ||
+			 (tu.even_distribution_legacy == 0) ||
+			 (DP_BRUTE_FORCE == 1))) ||
+			(tu.min_hblank_violated == 1)) {
+		do {
+			tu.err_fp = drm_fixp_from_fraction(1000, 1);
+
+			temp1_fp = drm_fixp_div(tu.lclk_fp, tu.pclk_fp);
+			temp2_fp = drm_fixp_from_fraction(
+					tu.delay_start_link_extra_pixclk, 1);
+			temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp);
+
+			if (temp1_fp)
+				tu.extra_buffer_margin =
+					drm_fixp2int_ceil(temp1_fp);
+			else
+				tu.extra_buffer_margin = 0;
+
+			temp1_fp = drm_fixp_from_fraction(tu.bpp, 8);
+			temp1_fp = drm_fixp_mul(tu.lwidth_fp, temp1_fp);
+
+			if (temp1_fp)
+				tu.n_symbols = drm_fixp2int_ceil(temp1_fp);
+			else
+				tu.n_symbols = 0;
+
+			for (tu.tu_size = 32; tu.tu_size <= 64; tu.tu_size++) {
+				for (tu.i_upper_boundary_count = 1;
+					tu.i_upper_boundary_count <= 15;
+					tu.i_upper_boundary_count++) {
+					for (tu.i_lower_boundary_count = 1;
+						tu.i_lower_boundary_count <= 15;
+						tu.i_lower_boundary_count++) {
+						_tu_valid_boundary_calc(&tu);
+					}
+				}
+			}
+			tu.delay_start_link_extra_pixclk--;
+		} while (tu.boundary_moderation_en != true &&
+			tu.boundary_mod_lower_err == 1 &&
+			tu.delay_start_link_extra_pixclk != 0);
+
+		if (tu.boundary_moderation_en == true) {
+			temp1_fp = drm_fixp_from_fraction(
+					(tu.upper_boundary_count *
+					tu.valid_boundary_link +
+					tu.lower_boundary_count *
+					(tu.valid_boundary_link - 1)), 1);
+			temp2_fp = drm_fixp_from_fraction(
+					(tu.upper_boundary_count +
+					tu.lower_boundary_count), 1);
+			tu.resulting_valid_fp =
+					drm_fixp_div(temp1_fp, temp2_fp);
+
+			temp1_fp = drm_fixp_from_fraction(
+					tu.tu_size_desired, 1);
+			tu.ratio_by_tu_fp =
+				drm_fixp_mul(tu.original_ratio_fp, temp1_fp);
+
+			tu.valid_lower_boundary_link =
+				tu.valid_boundary_link - 1;
+
+			temp1_fp = drm_fixp_from_fraction(tu.bpp, 8);
+			temp1_fp = drm_fixp_mul(tu.lwidth_fp, temp1_fp);
+			temp2_fp = drm_fixp_div(temp1_fp,
+						tu.resulting_valid_fp);
+			tu.n_tus = drm_fixp2int(temp2_fp);
+
+			tu.tu_size_minus1 = tu.tu_size_desired - 1;
+			tu.even_distribution_BF = 1;
+
+			temp1_fp =
+				drm_fixp_from_fraction(tu.tu_size_desired, 1);
+			temp2_fp =
+				drm_fixp_div(tu.resulting_valid_fp, temp1_fp);
+			tu.TU_ratio_err_fp = temp2_fp - tu.original_ratio_fp;
+		}
+	}
+
+	temp2_fp = drm_fixp_mul(LCLK_FAST_SKEW_fp, tu.lwidth_fp);
+
+	if (temp2_fp)
+		temp = drm_fixp2int_ceil(temp2_fp);
+	else
+		temp = 0;
+
+	temp1_fp = drm_fixp_from_fraction(tu.nlanes, 1);
+	temp2_fp = drm_fixp_mul(tu.original_ratio_fp, temp1_fp);
+	temp1_fp = drm_fixp_from_fraction(tu.bpp, 8);
+	temp2_fp = drm_fixp_div(temp1_fp, temp2_fp);
+	temp1_fp = drm_fixp_from_fraction(temp, 1);
+	temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp);
+	temp = drm_fixp2int(temp2_fp);
+
+	if (tu.async_en)
+		tu.delay_start_link += (int)temp;
+
+	temp1_fp = drm_fixp_from_fraction(tu.delay_start_link, 1);
+	tu.delay_start_time_fp = drm_fixp_div(temp1_fp, tu.lclk_fp);
+
+	/* OUTPUTS */
+	tu_table->valid_boundary_link       = tu.valid_boundary_link;
+	tu_table->delay_start_link          = tu.delay_start_link;
+	tu_table->boundary_moderation_en    = tu.boundary_moderation_en;
+	tu_table->valid_lower_boundary_link = tu.valid_lower_boundary_link;
+	tu_table->upper_boundary_count      = tu.upper_boundary_count;
+	tu_table->lower_boundary_count      = tu.lower_boundary_count;
+	tu_table->tu_size_minus1            = tu.tu_size_minus1;
+
+	DRM_DEBUG_DP("TU: valid_boundary_link: %d\n",
+				tu_table->valid_boundary_link);
+	DRM_DEBUG_DP("TU: delay_start_link: %d\n",
+				tu_table->delay_start_link);
+	DRM_DEBUG_DP("TU: boundary_moderation_en: %d\n",
+			tu_table->boundary_moderation_en);
+	DRM_DEBUG_DP("TU: valid_lower_boundary_link: %d\n",
+			tu_table->valid_lower_boundary_link);
+	DRM_DEBUG_DP("TU: upper_boundary_count: %d\n",
+			tu_table->upper_boundary_count);
+	DRM_DEBUG_DP("TU: lower_boundary_count: %d\n",
+			tu_table->lower_boundary_count);
+	DRM_DEBUG_DP("TU: tu_size_minus1: %d\n", tu_table->tu_size_minus1);
+}
+
+static void dp_ctrl_calc_tu_parameters(struct dp_ctrl_private *ctrl,
+		struct dp_vc_tu_mapping_table *tu_table)
+{
+	struct dp_tu_calc_input in;
+	struct drm_display_mode *drm_mode;
+
+	drm_mode = &ctrl->panel->dp_mode.drm_mode;
+
+	in.lclk = ctrl->link->link_params.rate / 1000;
+	in.pclk_khz = drm_mode->clock;
+	in.hactive = drm_mode->hdisplay;
+	in.hporch = drm_mode->htotal - drm_mode->hdisplay;
+	in.nlanes = ctrl->link->link_params.num_lanes;
+	in.bpp = ctrl->panel->dp_mode.bpp;
+	in.pixel_enc = 444;
+	in.dsc_en = 0;
+	in.async_en = 0;
+	in.fec_en = 0;
+	in.num_of_dsc_slices = 0;
+	in.compress_ratio = 100;
+
+	_dp_ctrl_calc_tu(&in, tu_table);
+}
+
+static void dp_ctrl_setup_tr_unit(struct dp_ctrl_private *ctrl)
+{
+	u32 dp_tu = 0x0;
+	u32 valid_boundary = 0x0;
+	u32 valid_boundary2 = 0x0;
+	struct dp_vc_tu_mapping_table tu_calc_table;
+
+	dp_ctrl_calc_tu_parameters(ctrl, &tu_calc_table);
+
+	dp_tu |= tu_calc_table.tu_size_minus1;
+	valid_boundary |= tu_calc_table.valid_boundary_link;
+	valid_boundary |= (tu_calc_table.delay_start_link << 16);
+
+	valid_boundary2 |= (tu_calc_table.valid_lower_boundary_link << 1);
+	valid_boundary2 |= (tu_calc_table.upper_boundary_count << 16);
+	valid_boundary2 |= (tu_calc_table.lower_boundary_count << 20);
+
+	if (tu_calc_table.boundary_moderation_en)
+		valid_boundary2 |= BIT(0);
+
+	pr_debug("dp_tu=0x%x, valid_boundary=0x%x, valid_boundary2=0x%x\n",
+			dp_tu, valid_boundary, valid_boundary2);
+
+	dp_catalog_ctrl_update_transfer_unit(ctrl->catalog,
+				dp_tu, valid_boundary, valid_boundary2);
+}
+
+static int dp_ctrl_wait4video_ready(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+
+	if (!wait_for_completion_timeout(&ctrl->video_comp,
+				WAIT_FOR_VIDEO_READY_TIMEOUT_JIFFIES)) {
+		DRM_ERROR("wait4video timedout\n");
+		ret = -ETIMEDOUT;
+	}
+	return ret;
+}
+
+static int dp_ctrl_update_vx_px(struct dp_ctrl_private *ctrl)
+{
+	struct dp_link *link = ctrl->link;
+	int ret = 0, lane, lane_cnt;
+	u8 buf[4];
+	u32 max_level_reached = 0;
+	u32 voltage_swing_level = link->phy_params.v_level;
+	u32 pre_emphasis_level = link->phy_params.p_level;
+
+	ret = dp_catalog_ctrl_update_vx_px(ctrl->catalog,
+		voltage_swing_level, pre_emphasis_level);
+
+	if (ret)
+		return ret;
+
+	if (voltage_swing_level >= DP_TRAIN_VOLTAGE_SWING_MAX) {
+		DRM_DEBUG_DP("max. voltage swing level reached %d\n",
+				voltage_swing_level);
+		max_level_reached |= DP_TRAIN_MAX_SWING_REACHED;
+	}
+
+	if (pre_emphasis_level >= DP_TRAIN_PRE_EMPHASIS_MAX) {
+		DRM_DEBUG_DP("max. pre-emphasis level reached %d\n",
+				pre_emphasis_level);
+		max_level_reached  |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+	}
+
+	pre_emphasis_level <<= DP_TRAIN_PRE_EMPHASIS_SHIFT;
+
+	lane_cnt = ctrl->link->link_params.num_lanes;
+	for (lane = 0; lane < lane_cnt; lane++)
+		buf[lane] = voltage_swing_level | pre_emphasis_level
+				| max_level_reached;
+
+	DRM_DEBUG_DP("sink: p|v=0x%x\n", voltage_swing_level
+					| pre_emphasis_level);
+	ret = drm_dp_dpcd_write(ctrl->aux, DP_TRAINING_LANE0_SET,
+					buf, lane_cnt);
+	if (ret == lane_cnt)
+		ret = 0;
+
+	return ret;
+}
+
+static bool dp_ctrl_train_pattern_set(struct dp_ctrl_private *ctrl,
+		u8 pattern)
+{
+	u8 buf;
+	int ret = 0;
+
+	DRM_DEBUG_DP("sink: pattern=%x\n", pattern);
+
+	buf = pattern;
+
+	if (pattern && pattern != DP_TRAINING_PATTERN_4)
+		buf |= DP_LINK_SCRAMBLING_DISABLE;
+
+	ret = drm_dp_dpcd_writeb(ctrl->aux, DP_TRAINING_PATTERN_SET, buf);
+	return ret == 1;
+}
+
+static int dp_ctrl_read_link_status(struct dp_ctrl_private *ctrl,
+				    u8 *link_status)
+{
+	int len = 0;
+	u32 const offset = DP_LANE_ALIGN_STATUS_UPDATED - DP_LANE0_1_STATUS;
+	u32 link_status_read_max_retries = 100;
+
+	while (--link_status_read_max_retries) {
+		len = drm_dp_dpcd_read_link_status(ctrl->aux,
+			link_status);
+		if (len != DP_LINK_STATUS_SIZE) {
+			DRM_ERROR("DP link status read failed, err: %d\n", len);
+			return len;
+		}
+
+		if (!(link_status[offset] & DP_LINK_STATUS_UPDATED))
+			return 0;
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int dp_ctrl_link_train_1(struct dp_ctrl_private *ctrl,
+		struct dp_cr_status *cr, int *training_step)
+{
+	int tries, old_v_level, ret = 0;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+	int const maximum_retries = 4;
+
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
+
+	*training_step = DP_TRAINING_1;
+
+	ret = dp_catalog_ctrl_set_pattern(ctrl->catalog, DP_TRAINING_PATTERN_1);
+	if (ret)
+		return ret;
+	dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_1 |
+		DP_LINK_SCRAMBLING_DISABLE);
+
+	ret = dp_ctrl_update_vx_px(ctrl);
+	if (ret)
+		return ret;
+
+	tries = 0;
+	old_v_level = ctrl->link->phy_params.v_level;
+	for (tries = 0; tries < maximum_retries; tries++) {
+		drm_dp_link_train_clock_recovery_delay(ctrl->panel->dpcd);
+
+		ret = dp_ctrl_read_link_status(ctrl, link_status);
+		if (ret)
+			return ret;
+
+		cr->lane_0_1 = link_status[0];
+		cr->lane_2_3 = link_status[1];
+
+		if (drm_dp_clock_recovery_ok(link_status,
+			ctrl->link->link_params.num_lanes)) {
+			return 0;
+		}
+
+		if (ctrl->link->phy_params.v_level >=
+			DP_TRAIN_VOLTAGE_SWING_MAX) {
+			DRM_ERROR_RATELIMITED("max v_level reached\n");
+			return -EAGAIN;
+		}
+
+		if (old_v_level != ctrl->link->phy_params.v_level) {
+			tries = 0;
+			old_v_level = ctrl->link->phy_params.v_level;
+		}
+
+		DRM_DEBUG_DP("clock recovery not done, adjusting vx px\n");
+
+		dp_link_adjust_levels(ctrl->link, link_status);
+		ret = dp_ctrl_update_vx_px(ctrl);
+		if (ret)
+			return ret;
+	}
+
+	DRM_ERROR("max tries reached\n");
+	return -ETIMEDOUT;
+}
+
+static int dp_ctrl_link_rate_down_shift(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+
+	switch (ctrl->link->link_params.rate) {
+	case 810000:
+		ctrl->link->link_params.rate = 540000;
+		break;
+	case 540000:
+		ctrl->link->link_params.rate = 270000;
+		break;
+	case 270000:
+		ctrl->link->link_params.rate = 162000;
+		break;
+	case 162000:
+	default:
+		ret = -EINVAL;
+		break;
+	};
+
+	if (!ret)
+		DRM_DEBUG_DP("new rate=0x%x\n", ctrl->link->link_params.rate);
+
+	return ret;
+}
+
+static int dp_ctrl_link_lane_down_shift(struct dp_ctrl_private *ctrl)
+{
+
+	if (ctrl->link->link_params.num_lanes == 1)
+		return -1;
+
+	ctrl->link->link_params.num_lanes /= 2;
+	ctrl->link->link_params.rate = ctrl->panel->link_info.rate;
+
+	ctrl->link->phy_params.p_level = 0;
+	ctrl->link->phy_params.v_level = 0;
+
+	return 0;
+}
+
+static void dp_ctrl_clear_training_pattern(struct dp_ctrl_private *ctrl)
+{
+	dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_DISABLE);
+	drm_dp_link_train_channel_eq_delay(ctrl->panel->dpcd);
+}
+
+static int dp_ctrl_link_train_2(struct dp_ctrl_private *ctrl,
+		struct dp_cr_status *cr, int *training_step)
+{
+	int tries = 0, ret = 0;
+	char pattern;
+	int const maximum_retries = 5;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
+
+	*training_step = DP_TRAINING_2;
+
+	if (drm_dp_tps3_supported(ctrl->panel->dpcd))
+		pattern = DP_TRAINING_PATTERN_3;
+	else
+		pattern = DP_TRAINING_PATTERN_2;
+
+	ret = dp_ctrl_update_vx_px(ctrl);
+	if (ret)
+		return ret;
+
+	ret = dp_catalog_ctrl_set_pattern(ctrl->catalog, pattern);
+	if (ret)
+		return ret;
+
+	dp_ctrl_train_pattern_set(ctrl, pattern | DP_RECOVERED_CLOCK_OUT_EN);
+
+	for (tries = 0; tries <= maximum_retries; tries++) {
+		drm_dp_link_train_channel_eq_delay(ctrl->panel->dpcd);
+
+		ret = dp_ctrl_read_link_status(ctrl, link_status);
+		if (ret)
+			return ret;
+		cr->lane_0_1 = link_status[0];
+		cr->lane_2_3 = link_status[1];
+
+		if (drm_dp_channel_eq_ok(link_status,
+			ctrl->link->link_params.num_lanes)) {
+			return 0;
+		}
+
+		dp_link_adjust_levels(ctrl->link, link_status);
+		ret = dp_ctrl_update_vx_px(ctrl);
+		if (ret)
+			return ret;
+
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int dp_ctrl_reinitialize_mainlink(struct dp_ctrl_private *ctrl);
+
+static int dp_ctrl_link_train(struct dp_ctrl_private *ctrl,
+		struct dp_cr_status *cr, int *training_step)
+{
+	int ret = 0;
+	u8 encoding = DP_SET_ANSI_8B10B;
+	struct dp_link_info link_info = {0};
+
+	dp_ctrl_config_ctrl(ctrl);
+
+	link_info.num_lanes = ctrl->link->link_params.num_lanes;
+	link_info.rate = ctrl->link->link_params.rate;
+	link_info.capabilities = DP_LINK_CAP_ENHANCED_FRAMING;
+
+	dp_aux_link_configure(ctrl->aux, &link_info);
+	drm_dp_dpcd_write(ctrl->aux, DP_MAIN_LINK_CHANNEL_CODING_SET,
+				&encoding, 1);
+
+	ret = dp_ctrl_link_train_1(ctrl, cr, training_step);
+	if (ret) {
+		DRM_ERROR("link training #1 failed. ret=%d\n", ret);
+		goto end;
+	}
+
+	/* print success info as this is a result of user initiated action */
+	DRM_DEBUG_DP("link training #1 successful\n");
+
+	ret = dp_ctrl_link_train_2(ctrl, cr, training_step);
+	if (ret) {
+		DRM_ERROR("link training #2 failed. ret=%d\n", ret);
+		goto end;
+	}
+
+	/* print success info as this is a result of user initiated action */
+	DRM_DEBUG_DP("link training #2 successful\n");
+
+end:
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
+
+	return ret;
+}
+
+static int dp_ctrl_setup_main_link(struct dp_ctrl_private *ctrl,
+		struct dp_cr_status *cr, int *training_step)
+{
+	int ret = 0;
+
+	dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, true);
+
+	if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN)
+		return ret;
+
+	/*
+	 * As part of previous calls, DP controller state might have
+	 * transitioned to PUSH_IDLE. In order to start transmitting
+	 * a link training pattern, we have to first do soft reset.
+	 */
+	dp_catalog_ctrl_reset(ctrl->catalog);
+
+	ret = dp_ctrl_link_train(ctrl, cr, training_step);
+
+	return ret;
+}
+
+static void dp_ctrl_set_clock_rate(struct dp_ctrl_private *ctrl,
+			enum dp_pm_type module, char *name, unsigned long rate)
+{
+	u32 num = ctrl->parser->mp[module].num_clk;
+	struct dss_clk *cfg = ctrl->parser->mp[module].clk_config;
+
+	while (num && strcmp(cfg->clk_name, name)) {
+		num--;
+		cfg++;
+	}
+
+	DRM_DEBUG_DP("setting rate=%lu on clk=%s\n", rate, name);
+
+	if (num)
+		cfg->rate = rate;
+	else
+		DRM_ERROR("%s clock doesn't exit to set rate %lu\n",
+				name, rate);
+}
+
+static int dp_ctrl_enable_mainlink_clocks(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+	struct dp_io *dp_io = &ctrl->parser->io;
+	struct phy *phy = dp_io->phy;
+	struct phy_configure_opts_dp *opts_dp = &dp_io->phy_opts.dp;
+
+	opts_dp->lanes = ctrl->link->link_params.num_lanes;
+	opts_dp->link_rate = ctrl->link->link_params.rate / 100;
+	dp_ctrl_set_clock_rate(ctrl, DP_CTRL_PM, "ctrl_link",
+					ctrl->link->link_params.rate * 1000);
+
+	phy_configure(phy, &dp_io->phy_opts);
+	phy_power_on(phy);
+
+	ret = dp_power_clk_enable(ctrl->power, DP_CTRL_PM, true);
+	if (ret)
+		DRM_ERROR("Unable to start link clocks. ret=%d\n", ret);
+
+	DRM_DEBUG_DP("link rate=%d pixel_clk=%d\n",
+		ctrl->link->link_params.rate, ctrl->dp_ctrl.pixel_rate);
+
+	return ret;
+}
+
+static int dp_ctrl_enable_stream_clocks(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+
+	dp_ctrl_set_clock_rate(ctrl, DP_STREAM_PM, "stream_pixel",
+					ctrl->dp_ctrl.pixel_rate * 1000);
+
+	ret = dp_power_clk_enable(ctrl->power, DP_STREAM_PM, true);
+	if (ret)
+		DRM_ERROR("Unabled to start pixel clocks. ret=%d\n", ret);
+
+	DRM_DEBUG_DP("link rate=%d pixel_clk=%d\n",
+			ctrl->link->link_params.rate, ctrl->dp_ctrl.pixel_rate);
+
+	return ret;
+}
+
+int dp_ctrl_host_init(struct dp_ctrl *dp_ctrl, bool flip)
+{
+	struct dp_ctrl_private *ctrl;
+	struct dp_io *dp_io;
+	struct phy *phy;
+
+	if (!dp_ctrl) {
+		DRM_ERROR("Invalid input data\n");
+		return -EINVAL;
+	}
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+	dp_io = &ctrl->parser->io;
+	phy = dp_io->phy;
+
+	ctrl->dp_ctrl.orientation = flip;
+
+	dp_catalog_ctrl_phy_reset(ctrl->catalog);
+	phy_init(phy);
+	dp_catalog_ctrl_enable_irq(ctrl->catalog, true);
+
+	return 0;
+}
+
+/**
+ * dp_ctrl_host_deinit() - Uninitialize DP controller
+ * @dp_ctrl: Display Port Driver data
+ *
+ * Perform required steps to uninitialize DP controller
+ * and its resources.
+ */
+void dp_ctrl_host_deinit(struct dp_ctrl *dp_ctrl)
+{
+	struct dp_ctrl_private *ctrl;
+
+	if (!dp_ctrl) {
+		DRM_ERROR("Invalid input data\n");
+		return;
+	}
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+
+	dp_catalog_ctrl_enable_irq(ctrl->catalog, false);
+
+	DRM_DEBUG_DP("Host deinitialized successfully\n");
+}
+
+static bool dp_ctrl_use_fixed_nvid(struct dp_ctrl_private *ctrl)
+{
+	u8 *dpcd = ctrl->panel->dpcd;
+	u32 edid_quirks = 0;
+
+	edid_quirks = drm_dp_get_edid_quirks(ctrl->panel->edid);
+	/*
+	 * For better interop experience, used a fixed NVID=0x8000
+	 * whenever connected to a VGA dongle downstream.
+	 */
+	if (drm_dp_is_branch(dpcd))
+		return (drm_dp_has_quirk(&ctrl->panel->desc, edid_quirks,
+				DP_DPCD_QUIRK_CONSTANT_N));
+
+	return false;
+}
+
+static int dp_ctrl_reinitialize_mainlink(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+	struct dp_io *dp_io = &ctrl->parser->io;
+	struct phy *phy = dp_io->phy;
+	struct phy_configure_opts_dp *opts_dp = &dp_io->phy_opts.dp;
+
+	dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false);
+	opts_dp->lanes = ctrl->link->link_params.num_lanes;
+	phy_configure(phy, &dp_io->phy_opts);
+	/*
+	 * Disable and re-enable the mainlink clock since the
+	 * link clock might have been adjusted as part of the
+	 * link maintenance.
+	 */
+	ret = dp_power_clk_enable(ctrl->power, DP_CTRL_PM, false);
+	if (ret) {
+		DRM_ERROR("Failed to disable clocks. ret=%d\n", ret);
+		return ret;
+	}
+	phy_power_off(phy);
+	/* hw recommended delay before re-enabling clocks */
+	msleep(20);
+
+	ret = dp_ctrl_enable_mainlink_clocks(ctrl);
+	if (ret) {
+		DRM_ERROR("Failed to enable mainlink clks. ret=%d\n", ret);
+		return ret;
+	}
+
+	return ret;
+}
+
+static int dp_ctrl_link_maintenance(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+	struct dp_cr_status cr;
+	int training_step = DP_TRAINING_NONE;
+
+	dp_ctrl_push_idle(&ctrl->dp_ctrl);
+	dp_catalog_ctrl_reset(ctrl->catalog);
+
+	ctrl->dp_ctrl.pixel_rate = ctrl->panel->dp_mode.drm_mode.clock;
+
+	ret = dp_ctrl_setup_main_link(ctrl, &cr, &training_step);
+	if (ret)
+		goto end;
+
+	dp_ctrl_clear_training_pattern(ctrl);
+
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO);
+
+	ret = dp_ctrl_wait4video_ready(ctrl);
+end:
+	return ret;
+}
+
+static int dp_ctrl_process_phy_test_request(struct dp_ctrl_private *ctrl)
+{
+	int ret = 0;
+
+	if (!ctrl->link->phy_params.phy_test_pattern_sel) {
+		DRM_DEBUG_DP("no test pattern selected by sink\n");
+		return ret;
+	}
+
+	/*
+	 * The global reset will need DP link related clocks to be
+	 * running. Add the global reset just before disabling the
+	 * link clocks and core clocks.
+	 */
+	ret = dp_ctrl_off(&ctrl->dp_ctrl);
+	if (ret) {
+		DRM_ERROR("failed to disable DP controller\n");
+		return ret;
+	}
+
+	ret = dp_ctrl_on_link(&ctrl->dp_ctrl);
+	if (!ret)
+		ret = dp_ctrl_on_stream(&ctrl->dp_ctrl);
+	else
+		DRM_ERROR("failed to enable DP link controller\n");
+
+	return ret;
+}
+
+static bool dp_ctrl_send_phy_test_pattern(struct dp_ctrl_private *ctrl)
+{
+	bool success = false;
+	u32 pattern_sent = 0x0;
+	u32 pattern_requested = ctrl->link->phy_params.phy_test_pattern_sel;
+
+	DRM_DEBUG_DP("request: 0x%x\n", pattern_requested);
+
+	if (dp_catalog_ctrl_update_vx_px(ctrl->catalog,
+			ctrl->link->phy_params.v_level,
+			ctrl->link->phy_params.p_level)) {
+		DRM_ERROR("Failed to set v/p levels\n");
+		return false;
+	}
+	dp_catalog_ctrl_send_phy_pattern(ctrl->catalog, pattern_requested);
+	dp_ctrl_update_vx_px(ctrl);
+	dp_link_send_test_response(ctrl->link);
+
+	pattern_sent = dp_catalog_ctrl_read_phy_pattern(ctrl->catalog);
+
+	switch (pattern_sent) {
+	case MR_LINK_TRAINING1:
+		success = (pattern_requested ==
+				DP_PHY_TEST_PATTERN_D10_2);
+		break;
+	case MR_LINK_SYMBOL_ERM:
+		success = ((pattern_requested ==
+			DP_PHY_TEST_PATTERN_ERROR_COUNT) ||
+				(pattern_requested ==
+				DP_PHY_TEST_PATTERN_CP2520));
+		break;
+	case MR_LINK_PRBS7:
+		success = (pattern_requested ==
+				DP_PHY_TEST_PATTERN_PRBS7);
+		break;
+	case MR_LINK_CUSTOM80:
+		success = (pattern_requested ==
+				DP_PHY_TEST_PATTERN_80BIT_CUSTOM);
+		break;
+	case MR_LINK_TRAINING4:
+		success = (pattern_requested ==
+				DP_PHY_TEST_PATTERN_SEL_MASK);
+		break;
+	default:
+		success = false;
+	}
+
+	DRM_DEBUG_DP("%s: test->0x%x\n", success ? "success" : "failed",
+						pattern_requested);
+	return success;
+}
+
+void dp_ctrl_handle_sink_request(struct dp_ctrl *dp_ctrl)
+{
+	struct dp_ctrl_private *ctrl;
+	u32 sink_request = 0x0;
+
+	if (!dp_ctrl) {
+		DRM_ERROR("invalid input\n");
+		return;
+	}
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+	sink_request = ctrl->link->sink_request;
+
+	if (sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) {
+		DRM_DEBUG_DP("PHY_TEST_PATTERN request\n");
+		if (dp_ctrl_process_phy_test_request(ctrl)) {
+			DRM_ERROR("process phy_test_req failed\n");
+			return;
+		}
+	}
+
+	if (sink_request & DP_LINK_STATUS_UPDATED) {
+		if (dp_ctrl_link_maintenance(ctrl)) {
+			DRM_ERROR("LM failed: TEST_LINK_TRAINING\n");
+			return;
+		}
+	}
+
+	if (sink_request & DP_TEST_LINK_TRAINING) {
+		dp_link_send_test_response(ctrl->link);
+		if (dp_ctrl_link_maintenance(ctrl)) {
+			DRM_ERROR("LM failed: TEST_LINK_TRAINING\n");
+			return;
+		}
+	}
+}
+
+int dp_ctrl_on_link(struct dp_ctrl *dp_ctrl)
+{
+	int rc = 0;
+	struct dp_ctrl_private *ctrl;
+	u32 rate = 0;
+	int link_train_max_retries = 5;
+	u32 const phy_cts_pixel_clk_khz = 148500;
+	struct dp_cr_status cr;
+	unsigned int training_step;
+
+	if (!dp_ctrl)
+		return -EINVAL;
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+
+	rate = ctrl->panel->link_info.rate;
+
+	dp_power_clk_enable(ctrl->power, DP_CORE_PM, true);
+
+	if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) {
+		DRM_DEBUG_DP("using phy test link parameters\n");
+		if (!ctrl->panel->dp_mode.drm_mode.clock)
+			ctrl->dp_ctrl.pixel_rate = phy_cts_pixel_clk_khz;
+	} else {
+		ctrl->link->link_params.rate = rate;
+		ctrl->link->link_params.num_lanes =
+			ctrl->panel->link_info.num_lanes;
+		ctrl->dp_ctrl.pixel_rate = ctrl->panel->dp_mode.drm_mode.clock;
+	}
+
+	DRM_DEBUG_DP("rate=%d, num_lanes=%d, pixel_rate=%d\n",
+		ctrl->link->link_params.rate,
+		ctrl->link->link_params.num_lanes, ctrl->dp_ctrl.pixel_rate);
+
+	rc = dp_ctrl_enable_mainlink_clocks(ctrl);
+	if (rc)
+		return rc;
+
+	ctrl->link->phy_params.p_level = 0;
+	ctrl->link->phy_params.v_level = 0;
+
+	while (--link_train_max_retries &&
+		!atomic_read(&ctrl->dp_ctrl.aborted)) {
+		rc = dp_ctrl_reinitialize_mainlink(ctrl);
+		if (rc) {
+			DRM_ERROR("Failed to reinitialize mainlink. rc=%d\n",
+					rc);
+			break;
+		}
+
+		training_step = DP_TRAINING_NONE;
+		rc = dp_ctrl_setup_main_link(ctrl, &cr, &training_step);
+		if (rc == 0) {
+			/* training completed successfully */
+			break;
+		} else if (training_step == DP_TRAINING_1) {
+			/* link train_1 failed */
+			rc = dp_ctrl_link_rate_down_shift(ctrl);
+			if (rc < 0) { /* already in RBR = 1.6G */
+				if (cr.lane_0_1 & DP_LANE0_1_CR_DONE) {
+					/*
+					 * some lanes are ready,
+					 * reduce lane number
+					 */
+					rc = dp_ctrl_link_lane_down_shift(ctrl);
+					if (rc < 0) { /* lane == 1 already */
+						/* end with failure */
+						break;
+					}
+				} else {
+					/* end with failure */
+					break; /* lane == 1 already */
+				}
+			}
+		} else if (training_step == DP_TRAINING_2) {
+			/* link train_2 failed, lower lane rate */
+			rc = dp_ctrl_link_lane_down_shift(ctrl);
+			if (rc < 0) {
+				/* end with failure */
+				break; /* lane == 1 already */
+			}
+		}
+	}
+
+	if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN)
+		return rc;
+
+	/* stop txing train pattern */
+	dp_ctrl_clear_training_pattern(ctrl);
+
+	/*
+	 * keep transmitting idle pattern until video ready
+	 * to avoid main link from loss of sync
+	 */
+	if (rc == 0)  /* link train successfully */
+		dp_ctrl_push_idle(dp_ctrl);
+
+	return rc;
+}
+
+int dp_ctrl_on_stream(struct dp_ctrl *dp_ctrl)
+{
+	u32 rate = 0;
+	int ret = 0;
+	bool mainlink_ready = false;
+	struct dp_ctrl_private *ctrl;
+
+	if (!dp_ctrl)
+		return -EINVAL;
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+
+	rate = ctrl->panel->link_info.rate;
+
+	ctrl->link->link_params.rate = rate;
+	ctrl->link->link_params.num_lanes = ctrl->panel->link_info.num_lanes;
+	ctrl->dp_ctrl.pixel_rate = ctrl->panel->dp_mode.drm_mode.clock;
+
+	DRM_DEBUG_DP("rate=%d, num_lanes=%d, pixel_rate=%d\n",
+		ctrl->link->link_params.rate,
+		ctrl->link->link_params.num_lanes, ctrl->dp_ctrl.pixel_rate);
+
+	if (!dp_power_clk_status(ctrl->power, DP_CTRL_PM)) { /* link clk is off */
+		ret = dp_ctrl_enable_mainlink_clocks(ctrl);
+		if (ret) {
+			DRM_ERROR("Failed to start link clocks. ret=%d\n", ret);
+			goto end;
+		}
+	}
+
+	ret = dp_ctrl_enable_stream_clocks(ctrl);
+	if (ret) {
+		DRM_ERROR("Failed to start pixel clocks. ret=%d\n", ret);
+		goto end;
+	}
+
+	if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) {
+		dp_ctrl_send_phy_test_pattern(ctrl);
+		return 0;
+	}
+
+	/*
+	 * Set up transfer unit values and set controller state to send
+	 * video.
+	 */
+	dp_ctrl_configure_source_params(ctrl);
+
+	dp_catalog_ctrl_config_msa(ctrl->catalog,
+		ctrl->link->link_params.rate,
+		ctrl->dp_ctrl.pixel_rate, dp_ctrl_use_fixed_nvid(ctrl));
+
+	reinit_completion(&ctrl->video_comp);
+
+	dp_ctrl_setup_tr_unit(ctrl);
+
+	dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO);
+
+	ret = dp_ctrl_wait4video_ready(ctrl);
+	if (ret)
+		return ret;
+
+	mainlink_ready = dp_catalog_ctrl_mainlink_ready(ctrl->catalog);
+	DRM_DEBUG_DP("mainlink %s\n", mainlink_ready ? "READY" : "NOT READY");
+
+end:
+	return ret;
+}
+
+int dp_ctrl_off(struct dp_ctrl *dp_ctrl)
+{
+	struct dp_ctrl_private *ctrl;
+	struct dp_io *dp_io;
+	struct phy *phy;
+	int ret = 0;
+
+	if (!dp_ctrl)
+		return -EINVAL;
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+	dp_io = &ctrl->parser->io;
+	phy = dp_io->phy;
+
+	dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false);
+
+	dp_catalog_ctrl_reset(ctrl->catalog);
+
+	ret = dp_power_clk_enable(ctrl->power, DP_STREAM_PM, false);
+	if (ret)
+		DRM_ERROR("Failed to disable pixel clocks. ret=%d\n", ret);
+
+	ret = dp_power_clk_enable(ctrl->power, DP_CTRL_PM, false);
+	if (ret) {
+		DRM_ERROR("Failed to disable link clocks. ret=%d\n", ret);
+	}
+
+	phy_power_off(phy);
+	phy_exit(phy);
+
+	DRM_DEBUG_DP("DP off done\n");
+	return ret;
+}
+
+void dp_ctrl_isr(struct dp_ctrl *dp_ctrl)
+{
+	struct dp_ctrl_private *ctrl;
+	u32 isr;
+
+	if (!dp_ctrl)
+		return;
+
+	ctrl = container_of(dp_ctrl, struct dp_ctrl_private, dp_ctrl);
+
+	isr = dp_catalog_ctrl_get_interrupt(ctrl->catalog);
+
+	if (isr & DP_CTRL_INTR_READY_FOR_VIDEO) {
+		DRM_DEBUG_DP("dp_video_ready\n");
+		complete(&ctrl->video_comp);
+	}
+
+	if (isr & DP_CTRL_INTR_IDLE_PATTERN_SENT) {
+		DRM_DEBUG_DP("idle_patterns_sent\n");
+		complete(&ctrl->idle_comp);
+	}
+}
+
+struct dp_ctrl *dp_ctrl_get(struct device *dev, struct dp_link *link,
+			struct dp_panel *panel,	struct drm_dp_aux *aux,
+			struct dp_power *power, struct dp_catalog *catalog,
+			struct dp_parser *parser)
+{
+	struct dp_ctrl_private *ctrl;
+
+	if (!dev || !panel || !aux ||
+	    !link || !catalog) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	ctrl = devm_kzalloc(dev, sizeof(*ctrl), GFP_KERNEL);
+	if (!ctrl) {
+		DRM_ERROR("Mem allocation failure\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	init_completion(&ctrl->idle_comp);
+	init_completion(&ctrl->video_comp);
+
+	/* in parameters */
+	ctrl->parser   = parser;
+	ctrl->panel    = panel;
+	ctrl->power    = power;
+	ctrl->aux      = aux;
+	ctrl->link     = link;
+	ctrl->catalog  = catalog;
+	ctrl->dev      = dev;
+
+	return &ctrl->dp_ctrl;
+}
+
+void dp_ctrl_put(struct dp_ctrl *dp_ctrl)
+{
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.h b/drivers/gpu/drm/msm/dp/dp_ctrl.h
new file mode 100644
index 000000000000..f60ba93c8678
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_ctrl.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_CTRL_H_
+#define _DP_CTRL_H_
+
+#include "dp_aux.h"
+#include "dp_panel.h"
+#include "dp_link.h"
+#include "dp_parser.h"
+#include "dp_power.h"
+#include "dp_catalog.h"
+
+struct dp_ctrl {
+	bool orientation;
+	atomic_t aborted;
+	u32 pixel_rate;
+};
+
+int dp_ctrl_host_init(struct dp_ctrl *dp_ctrl, bool flip);
+void dp_ctrl_host_deinit(struct dp_ctrl *dp_ctrl);
+int dp_ctrl_on_link(struct dp_ctrl *dp_ctrl);
+int dp_ctrl_on_stream(struct dp_ctrl *dp_ctrl);
+int dp_ctrl_off(struct dp_ctrl *dp_ctrl);
+void dp_ctrl_push_idle(struct dp_ctrl *dp_ctrl);
+void dp_ctrl_isr(struct dp_ctrl *dp_ctrl);
+void dp_ctrl_handle_sink_request(struct dp_ctrl *dp_ctrl);
+struct dp_ctrl *dp_ctrl_get(struct device *dev, struct dp_link *link,
+			struct dp_panel *panel,	struct drm_dp_aux *aux,
+			struct dp_power *power, struct dp_catalog *catalog,
+			struct dp_parser *parser);
+void dp_ctrl_put(struct dp_ctrl *dp_ctrl);
+
+#endif /* _DP_CTRL_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_debug.c b/drivers/gpu/drm/msm/dp/dp_debug.c
new file mode 100644
index 000000000000..84670bcdcfea
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_debug.c
@@ -0,0 +1,485 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)"[drm-dp] %s: " fmt, __func__
+
+#include <linux/debugfs.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_file.h>
+
+#include "dp_parser.h"
+#include "dp_catalog.h"
+#include "dp_aux.h"
+#include "dp_ctrl.h"
+#include "dp_debug.h"
+#include "dp_display.h"
+
+#define DEBUG_NAME "msm_dp"
+
+struct dp_debug_private {
+	struct dentry *root;
+
+	struct dp_usbpd *usbpd;
+	struct dp_link *link;
+	struct dp_panel *panel;
+	struct drm_connector **connector;
+	struct device *dev;
+	struct drm_device *drm_dev;
+
+	struct dp_debug dp_debug;
+};
+
+static int dp_debug_check_buffer_overflow(int rc, int *max_size, int *len)
+{
+	if (rc >= *max_size) {
+		DRM_ERROR("buffer overflow\n");
+		return -EINVAL;
+	}
+	*len += rc;
+	*max_size = SZ_4K - *len;
+
+	return 0;
+}
+
+static ssize_t dp_debug_read_info(struct file *file, char __user *user_buff,
+		size_t count, loff_t *ppos)
+{
+	struct dp_debug_private *debug = file->private_data;
+	char *buf;
+	u32 len = 0, rc = 0;
+	u64 lclk = 0;
+	u32 max_size = SZ_4K;
+	u32 link_params_rate;
+	struct drm_display_mode *drm_mode;
+
+	if (!debug)
+		return -ENODEV;
+
+	if (*ppos)
+		return 0;
+
+	buf = kzalloc(SZ_4K, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	drm_mode = &debug->panel->dp_mode.drm_mode;
+
+	rc = snprintf(buf + len, max_size, "\tname = %s\n", DEBUG_NAME);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\tdp_panel\n\t\tmax_pclk_khz = %d\n",
+			debug->panel->max_pclk_khz);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\tdrm_dp_link\n\t\trate = %u\n",
+			debug->panel->link_info.rate);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			 "\t\tnum_lanes = %u\n",
+			debug->panel->link_info.num_lanes);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tcapabilities = %lu\n",
+			debug->panel->link_info.capabilities);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\tdp_panel_info:\n\t\tactive = %dx%d\n",
+			drm_mode->hdisplay,
+			drm_mode->vdisplay);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tback_porch = %dx%d\n",
+			drm_mode->htotal - drm_mode->hsync_end,
+			drm_mode->vtotal - drm_mode->vsync_end);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tfront_porch = %dx%d\n",
+			drm_mode->hsync_start - drm_mode->hdisplay,
+			drm_mode->vsync_start - drm_mode->vdisplay);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tsync_width = %dx%d\n",
+			drm_mode->hsync_end - drm_mode->hsync_start,
+			drm_mode->vsync_end - drm_mode->vsync_start);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tactive_low = %dx%d\n",
+			debug->panel->dp_mode.h_active_low,
+			debug->panel->dp_mode.v_active_low);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\th_skew = %d\n",
+			drm_mode->hskew);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\trefresh rate = %d\n",
+			drm_mode_vrefresh(drm_mode));
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tpixel clock khz = %d\n",
+			drm_mode->clock);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tbpp = %d\n",
+			debug->panel->dp_mode.bpp);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	/* Link Information */
+	rc = snprintf(buf + len, max_size,
+			"\tdp_link:\n\t\ttest_requested = %d\n",
+			debug->link->sink_request);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tnum_lanes = %d\n",
+			debug->link->link_params.num_lanes);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	link_params_rate = debug->link->link_params.rate;
+	rc = snprintf(buf + len, max_size,
+			"\t\tbw_code = %d\n",
+			drm_dp_link_rate_to_bw_code(link_params_rate));
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	lclk = debug->link->link_params.rate * 1000;
+	rc = snprintf(buf + len, max_size,
+			"\t\tlclk = %lld\n", lclk);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tv_level = %d\n",
+			debug->link->phy_params.v_level);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	rc = snprintf(buf + len, max_size,
+			"\t\tp_level = %d\n",
+			debug->link->phy_params.p_level);
+	if (dp_debug_check_buffer_overflow(rc, &max_size, &len))
+		goto error;
+
+	if (copy_to_user(user_buff, buf, len))
+		goto error;
+
+	*ppos += len;
+
+	kfree(buf);
+	return len;
+ error:
+	kfree(buf);
+	return -EINVAL;
+}
+
+static int dp_test_data_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev;
+	struct dp_debug_private *debug;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+	u32 bpc;
+
+	debug = m->private;
+	dev = debug->drm_dev;
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+
+		if (connector->connector_type !=
+			DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected) {
+			bpc = debug->link->test_video.test_bit_depth;
+			seq_printf(m, "hdisplay: %d\n",
+					debug->link->test_video.test_h_width);
+			seq_printf(m, "vdisplay: %d\n",
+					debug->link->test_video.test_v_height);
+					seq_printf(m, "bpc: %u\n",
+					dp_link_bit_depth_to_bpc(bpc));
+		} else
+			seq_puts(m, "0");
+	}
+
+	drm_connector_list_iter_end(&conn_iter);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(dp_test_data);
+
+static int dp_test_type_show(struct seq_file *m, void *data)
+{
+	struct dp_debug_private *debug = m->private;
+	struct drm_device *dev = debug->drm_dev;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+
+		if (connector->connector_type !=
+			DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected)
+			seq_printf(m, "%02x", DP_TEST_LINK_VIDEO_PATTERN);
+		else
+			seq_puts(m, "0");
+	}
+	drm_connector_list_iter_end(&conn_iter);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(dp_test_type);
+
+static ssize_t dp_test_active_write(struct file *file,
+		const char __user *ubuf,
+		size_t len, loff_t *offp)
+{
+	char *input_buffer;
+	int status = 0;
+	struct dp_debug_private *debug;
+	struct drm_device *dev;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+	int val = 0;
+
+	debug = ((struct seq_file *)file->private_data)->private;
+	dev = debug->drm_dev;
+
+	if (len == 0)
+		return 0;
+
+	input_buffer = memdup_user_nul(ubuf, len);
+	if (IS_ERR(input_buffer))
+		return PTR_ERR(input_buffer);
+
+	DRM_DEBUG_DRIVER("Copied %d bytes from user\n", (unsigned int)len);
+
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+		if (connector->connector_type !=
+			DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected) {
+			status = kstrtoint(input_buffer, 10, &val);
+			if (status < 0)
+				break;
+			DRM_DEBUG_DRIVER("Got %d for test active\n", val);
+			/* To prevent erroneous activation of the compliance
+			 * testing code, only accept an actual value of 1 here
+			 */
+			if (val == 1)
+				debug->panel->video_test = true;
+			else
+				debug->panel->video_test = false;
+		}
+	}
+	drm_connector_list_iter_end(&conn_iter);
+	kfree(input_buffer);
+	if (status < 0)
+		return status;
+
+	*offp += len;
+	return len;
+}
+
+static int dp_test_active_show(struct seq_file *m, void *data)
+{
+	struct dp_debug_private *debug = m->private;
+	struct drm_device *dev = debug->drm_dev;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+		if (connector->connector_type !=
+			DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected) {
+			if (debug->panel->video_test)
+				seq_puts(m, "1");
+			else
+				seq_puts(m, "0");
+		} else
+			seq_puts(m, "0");
+	}
+	drm_connector_list_iter_end(&conn_iter);
+
+	return 0;
+}
+
+static int dp_test_active_open(struct inode *inode,
+		struct file *file)
+{
+	return single_open(file, dp_test_active_show,
+			inode->i_private);
+}
+
+static const struct file_operations dp_debug_fops = {
+	.open = simple_open,
+	.read = dp_debug_read_info,
+};
+
+static const struct file_operations test_active_fops = {
+	.owner = THIS_MODULE,
+	.open = dp_test_active_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+	.write = dp_test_active_write
+};
+
+static int dp_debug_init(struct dp_debug *dp_debug, struct drm_minor *minor)
+{
+	int rc = 0;
+	struct dp_debug_private *debug = container_of(dp_debug,
+			struct dp_debug_private, dp_debug);
+	struct dentry *file;
+	struct dentry *test_active;
+	struct dentry *test_data, *test_type;
+
+	file = debugfs_create_file("dp_debug", 0444, minor->debugfs_root,
+			debug, &dp_debug_fops);
+	if (IS_ERR_OR_NULL(file)) {
+		rc = PTR_ERR(file);
+		DRM_ERROR("[%s] debugfs create file failed, rc=%d\n",
+				  DEBUG_NAME, rc);
+	}
+
+	test_active = debugfs_create_file("msm_dp_test_active", 0444,
+			minor->debugfs_root,
+			debug, &test_active_fops);
+	if (IS_ERR_OR_NULL(test_active)) {
+		rc = PTR_ERR(test_active);
+		DRM_ERROR("[%s] debugfs test_active failed, rc=%d\n",
+				  DEBUG_NAME, rc);
+	}
+
+	test_data = debugfs_create_file("msm_dp_test_data", 0444,
+			minor->debugfs_root,
+			debug, &dp_test_data_fops);
+	if (IS_ERR_OR_NULL(test_data)) {
+		rc = PTR_ERR(test_data);
+		DRM_ERROR("[%s] debugfs test_data failed, rc=%d\n",
+				  DEBUG_NAME, rc);
+	}
+
+	test_type = debugfs_create_file("msm_dp_test_type", 0444,
+			minor->debugfs_root,
+			debug, &dp_test_type_fops);
+	if (IS_ERR_OR_NULL(test_type)) {
+		rc = PTR_ERR(test_type);
+		DRM_ERROR("[%s] debugfs test_type failed, rc=%d\n",
+				  DEBUG_NAME, rc);
+	}
+
+	debug->root = minor->debugfs_root;
+
+	return rc;
+}
+
+struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
+		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct drm_connector **connector, struct drm_minor *minor)
+{
+	int rc = 0;
+	struct dp_debug_private *debug;
+	struct dp_debug *dp_debug;
+
+	if (!dev || !panel || !usbpd || !link) {
+		DRM_ERROR("invalid input\n");
+		rc = -EINVAL;
+		goto error;
+	}
+
+	debug = devm_kzalloc(dev, sizeof(*debug), GFP_KERNEL);
+	if (!debug) {
+		rc = -ENOMEM;
+		goto error;
+	}
+
+	debug->dp_debug.debug_en = false;
+	debug->usbpd = usbpd;
+	debug->link = link;
+	debug->panel = panel;
+	debug->dev = dev;
+	debug->drm_dev = minor->dev;
+	debug->connector = connector;
+
+	dp_debug = &debug->dp_debug;
+	dp_debug->vdisplay = 0;
+	dp_debug->hdisplay = 0;
+	dp_debug->vrefresh = 0;
+
+	rc = dp_debug_init(dp_debug, minor);
+	if (rc) {
+		devm_kfree(dev, debug);
+		goto error;
+	}
+
+	return dp_debug;
+ error:
+	return ERR_PTR(rc);
+}
+
+static int dp_debug_deinit(struct dp_debug *dp_debug)
+{
+	struct dp_debug_private *debug;
+
+	if (!dp_debug)
+		return -EINVAL;
+
+	debug = container_of(dp_debug, struct dp_debug_private, dp_debug);
+
+	debugfs_remove_recursive(debug->root);
+
+	return 0;
+}
+
+void dp_debug_put(struct dp_debug *dp_debug)
+{
+	struct dp_debug_private *debug;
+
+	if (!dp_debug)
+		return;
+
+	debug = container_of(dp_debug, struct dp_debug_private, dp_debug);
+
+	dp_debug_deinit(dp_debug);
+
+	devm_kfree(debug->dev, debug);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_debug.h b/drivers/gpu/drm/msm/dp/dp_debug.h
new file mode 100644
index 000000000000..7eaedfbb149c
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_debug.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_DEBUG_H_
+#define _DP_DEBUG_H_
+
+#include "dp_panel.h"
+#include "dp_link.h"
+
+/**
+ * struct dp_debug
+ * @debug_en: specifies whether debug mode enabled
+ * @vdisplay: used to filter out vdisplay value
+ * @hdisplay: used to filter out hdisplay value
+ * @vrefresh: used to filter out vrefresh value
+ * @tpg_state: specifies whether tpg feature is enabled
+ */
+struct dp_debug {
+	bool debug_en;
+	int aspect_ratio;
+	int vdisplay;
+	int hdisplay;
+	int vrefresh;
+};
+
+#if defined(CONFIG_DEBUG_FS)
+
+/**
+ * dp_debug_get() - configure and get the DisplayPlot debug module data
+ *
+ * @dev: device instance of the caller
+ * @panel: instance of panel module
+ * @usbpd: instance of usbpd module
+ * @link: instance of link module
+ * @connector: double pointer to display connector
+ * @minor: pointer to drm minor number after device registration
+ * return: pointer to allocated debug module data
+ *
+ * This function sets up the debug module and provides a way
+ * for debugfs input to be communicated with existing modules
+ */
+struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
+		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct drm_connector **connector,
+		struct drm_minor *minor);
+
+/**
+ * dp_debug_put()
+ *
+ * Cleans up dp_debug instance
+ *
+ * @dp_debug: instance of dp_debug
+ */
+void dp_debug_put(struct dp_debug *dp_debug);
+
+#else
+
+static inline
+struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
+		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct drm_connector **connector, struct drm_minor *minor)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline void dp_debug_put(struct dp_debug *dp_debug)
+{
+}
+
+#endif /* defined(CONFIG_DEBUG_FS) */
+
+#endif /* _DP_DEBUG_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
new file mode 100644
index 000000000000..e175aa3fd3a9
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -0,0 +1,1463 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/debugfs.h>
+#include <linux/component.h>
+#include <linux/of_irq.h>
+#include <linux/delay.h>
+
+#include "msm_drv.h"
+#include "msm_kms.h"
+#include "dp_hpd.h"
+#include "dp_parser.h"
+#include "dp_power.h"
+#include "dp_catalog.h"
+#include "dp_aux.h"
+#include "dp_reg.h"
+#include "dp_link.h"
+#include "dp_panel.h"
+#include "dp_ctrl.h"
+#include "dp_display.h"
+#include "dp_drm.h"
+#include "dp_audio.h"
+#include "dp_debug.h"
+
+static struct msm_dp *g_dp_display;
+#define HPD_STRING_SIZE 30
+
+enum {
+	ISR_DISCONNECTED,
+	ISR_CONNECT_PENDING,
+	ISR_CONNECTED,
+	ISR_HPD_REPLUG_COUNT,
+	ISR_IRQ_HPD_PULSE_COUNT,
+	ISR_HPD_LO_GLITH_COUNT,
+};
+
+/* event thread connection state */
+enum {
+	ST_DISCONNECTED,
+	ST_CONNECT_PENDING,
+	ST_CONNECTED,
+	ST_DISCONNECT_PENDING,
+	ST_SUSPEND_PENDING,
+	ST_SUSPENDED,
+};
+
+enum {
+	EV_NO_EVENT,
+	/* hpd events */
+	EV_HPD_INIT_SETUP,
+	EV_HPD_PLUG_INT,
+	EV_IRQ_HPD_INT,
+	EV_HPD_REPLUG_INT,
+	EV_HPD_UNPLUG_INT,
+	EV_USER_NOTIFICATION,
+	EV_CONNECT_PENDING_TIMEOUT,
+	EV_DISCONNECT_PENDING_TIMEOUT,
+};
+
+#define EVENT_TIMEOUT	(HZ/10)	/* 100ms */
+#define DP_EVENT_Q_MAX	8
+
+#define DP_TIMEOUT_5_SECOND	(5000/EVENT_TIMEOUT)
+#define DP_TIMEOUT_NONE		0
+
+#define WAIT_FOR_RESUME_TIMEOUT_JIFFIES (HZ / 2)
+
+struct dp_event {
+	u32 event_id;
+	u32 data;
+	u32 delay;
+};
+
+struct dp_display_private {
+	char *name;
+	int irq;
+
+	/* state variables */
+	bool core_initialized;
+	bool hpd_irq_on;
+	bool audio_supported;
+
+	struct platform_device *pdev;
+	struct dentry *root;
+
+	struct dp_usbpd   *usbpd;
+	struct dp_parser  *parser;
+	struct dp_power   *power;
+	struct dp_catalog *catalog;
+	struct drm_dp_aux *aux;
+	struct dp_link    *link;
+	struct dp_panel   *panel;
+	struct dp_ctrl    *ctrl;
+	struct dp_debug   *debug;
+
+	struct dp_usbpd_cb usbpd_cb;
+	struct dp_display_mode dp_mode;
+	struct msm_dp dp_display;
+
+	/* wait for audio signaling */
+	struct completion audio_comp;
+
+	/* event related only access by event thread */
+	struct mutex event_mutex;
+	wait_queue_head_t event_q;
+	atomic_t hpd_state;
+	u32 event_pndx;
+	u32 event_gndx;
+	struct dp_event event_list[DP_EVENT_Q_MAX];
+	spinlock_t event_lock;
+
+	struct completion resume_comp;
+
+	struct dp_audio *audio;
+};
+
+static const struct of_device_id dp_dt_match[] = {
+	{.compatible = "qcom,sc7180-dp"},
+	{}
+};
+
+static int dp_add_event(struct dp_display_private *dp_priv, u32 event,
+						u32 data, u32 delay)
+{
+	unsigned long flag;
+	struct dp_event *todo;
+	int pndx;
+
+	spin_lock_irqsave(&dp_priv->event_lock, flag);
+	pndx = dp_priv->event_pndx + 1;
+	pndx %= DP_EVENT_Q_MAX;
+	if (pndx == dp_priv->event_gndx) {
+		pr_err("event_q is full: pndx=%d gndx=%d\n",
+			dp_priv->event_pndx, dp_priv->event_gndx);
+		spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+		return -EPERM;
+	}
+	todo = &dp_priv->event_list[dp_priv->event_pndx++];
+	dp_priv->event_pndx %= DP_EVENT_Q_MAX;
+	todo->event_id = event;
+	todo->data = data;
+	todo->delay = delay;
+	wake_up(&dp_priv->event_q);
+	spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+
+	return 0;
+}
+
+static int dp_del_event(struct dp_display_private *dp_priv, u32 event)
+{
+	unsigned long flag;
+	struct dp_event *todo;
+	u32	gndx;
+
+	spin_lock_irqsave(&dp_priv->event_lock, flag);
+	if (dp_priv->event_pndx == dp_priv->event_gndx) {
+		spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+		return -ENOENT;
+	}
+
+	gndx = dp_priv->event_gndx;
+	while (dp_priv->event_pndx != gndx) {
+		todo = &dp_priv->event_list[gndx];
+		if (todo->event_id == event) {
+			todo->event_id = EV_NO_EVENT;	/* deleted */
+			todo->delay = 0;
+		}
+		gndx++;
+		gndx %= DP_EVENT_Q_MAX;
+	}
+	spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+
+	return 0;
+}
+
+void dp_display_signal_audio_complete(struct msm_dp *dp_display)
+{
+	struct dp_display_private *dp;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	complete_all(&dp->audio_comp);
+}
+
+static int dp_display_bind(struct device *dev, struct device *master,
+			   void *data)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+	struct drm_device *drm;
+	struct msm_drm_private *priv;
+
+	drm = dev_get_drvdata(master);
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+	if (!dp) {
+		DRM_ERROR("DP driver bind failed. Invalid driver data\n");
+		return -EINVAL;
+	}
+
+	dp->dp_display.drm_dev = drm;
+	priv = drm->dev_private;
+	priv->dp = &(dp->dp_display);
+
+	rc = dp->parser->parse(dp->parser);
+	if (rc) {
+		DRM_ERROR("device tree parsing failed\n");
+		goto end;
+	}
+
+	rc = dp_aux_register(dp->aux);
+	if (rc) {
+		DRM_ERROR("DRM DP AUX register failed\n");
+		goto end;
+	}
+
+	rc = dp_power_client_init(dp->power);
+	if (rc) {
+		DRM_ERROR("Power client create failed\n");
+		goto end;
+	}
+
+	rc = dp_register_audio_driver(dev, dp->audio);
+	if (rc)
+		DRM_ERROR("Audio registration Dp failed\n");
+
+end:
+	return rc;
+}
+
+static void dp_display_unbind(struct device *dev, struct device *master,
+			      void *data)
+{
+	struct dp_display_private *dp;
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct msm_drm_private *priv = drm->dev_private;
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+	if (!dp) {
+		DRM_ERROR("Invalid DP driver data\n");
+		return;
+	}
+
+	dp_power_client_deinit(dp->power);
+	dp_aux_unregister(dp->aux);
+	priv->dp = NULL;
+}
+
+static const struct component_ops dp_display_comp_ops = {
+	.bind = dp_display_bind,
+	.unbind = dp_display_unbind,
+};
+
+static bool dp_display_is_ds_bridge(struct dp_panel *panel)
+{
+	return (panel->dpcd[DP_DOWNSTREAMPORT_PRESENT] &
+		DP_DWN_STRM_PORT_PRESENT);
+}
+
+static bool dp_display_is_sink_count_zero(struct dp_display_private *dp)
+{
+	return dp_display_is_ds_bridge(dp->panel) &&
+		(dp->link->sink_count == 0);
+}
+
+static void dp_display_send_hpd_event(struct msm_dp *dp_display)
+{
+	struct dp_display_private *dp;
+	struct drm_connector *connector;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	connector = dp->dp_display.connector;
+	drm_helper_hpd_irq_event(connector->dev);
+}
+
+static int dp_display_send_hpd_notification(struct dp_display_private *dp,
+					    bool hpd)
+{
+	static bool encoder_mode_set;
+	struct msm_drm_private *priv = dp->dp_display.drm_dev->dev_private;
+	struct msm_kms *kms = priv->kms;
+
+	if ((hpd && dp->dp_display.is_connected) ||
+			(!hpd && !dp->dp_display.is_connected)) {
+		DRM_DEBUG_DP("HPD already %s\n", (hpd ? "on" : "off"));
+		return 0;
+	}
+
+	/* reset video pattern flag on disconnect */
+	if (!hpd)
+		dp->panel->video_test = false;
+
+	dp->dp_display.is_connected = hpd;
+
+	if (dp->dp_display.is_connected && dp->dp_display.encoder
+				&& !encoder_mode_set
+				&& kms->funcs->set_encoder_mode) {
+		kms->funcs->set_encoder_mode(kms,
+				dp->dp_display.encoder, false);
+		DRM_DEBUG_DP("set_encoder_mode() Completed\n");
+		encoder_mode_set = true;
+	}
+
+	dp_display_send_hpd_event(&dp->dp_display);
+
+	return 0;
+}
+
+static int dp_display_process_hpd_high(struct dp_display_private *dp)
+{
+	int rc = 0;
+	struct edid *edid;
+
+	dp->panel->max_dp_lanes = dp->parser->max_dp_lanes;
+
+	rc = dp_panel_read_sink_caps(dp->panel, dp->dp_display.connector);
+	if (rc)
+		goto end;
+
+	dp_link_process_request(dp->link);
+
+	edid = dp->panel->edid;
+
+	dp->audio_supported = drm_detect_monitor_audio(edid);
+	dp_panel_handle_sink_request(dp->panel);
+
+	dp->dp_display.max_pclk_khz = DP_MAX_PIXEL_CLK_KHZ;
+	dp->dp_display.max_dp_lanes = dp->parser->max_dp_lanes;
+
+	rc = dp_ctrl_on_link(dp->ctrl);
+	if (rc) {
+		DRM_ERROR("failed to complete DP link training\n");
+		goto end;
+	}
+
+	dp_add_event(dp, EV_USER_NOTIFICATION, true, 0);
+
+
+end:
+	return rc;
+}
+
+static void dp_display_host_init(struct dp_display_private *dp)
+{
+	bool flip = false;
+
+	if (dp->core_initialized) {
+		DRM_DEBUG_DP("DP core already initialized\n");
+		return;
+	}
+
+	if (dp->usbpd->orientation == ORIENTATION_CC2)
+		flip = true;
+
+	dp_power_init(dp->power, flip);
+	dp_ctrl_host_init(dp->ctrl, flip);
+	dp_aux_init(dp->aux);
+	dp->core_initialized = true;
+}
+
+static int dp_display_usbpd_configure_cb(struct device *dev)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+
+	if (!dev) {
+		DRM_ERROR("invalid dev\n");
+		rc = -EINVAL;
+		goto end;
+	}
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+	if (!dp) {
+		DRM_ERROR("no driver data found\n");
+		rc = -ENODEV;
+		goto end;
+	}
+
+	dp_display_host_init(dp);
+
+	/*
+	 * set sink to normal operation mode -- D0
+	 * before dpcd read
+	 */
+	dp_link_psm_config(dp->link, &dp->panel->link_info, false);
+	rc = dp_display_process_hpd_high(dp);
+end:
+	return rc;
+}
+
+static int dp_display_usbpd_disconnect_cb(struct device *dev)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+
+	if (!dev) {
+		DRM_ERROR("invalid dev\n");
+		rc = -EINVAL;
+		return rc;
+	}
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+	if (!dp) {
+		DRM_ERROR("no driver data found\n");
+		rc = -ENODEV;
+		return rc;
+	}
+
+	dp_add_event(dp, EV_USER_NOTIFICATION, false, 0);
+
+	return rc;
+}
+
+static void dp_display_handle_video_request(struct dp_display_private *dp)
+{
+	if (dp->link->sink_request & DP_TEST_LINK_VIDEO_PATTERN) {
+		dp->panel->video_test = true;
+		dp_link_send_test_response(dp->link);
+	}
+}
+
+static int dp_display_handle_irq_hpd(struct dp_display_private *dp)
+{
+	u32 sink_request;
+
+	sink_request = dp->link->sink_request;
+
+	if (sink_request & DS_PORT_STATUS_CHANGED) {
+		dp_add_event(dp, EV_USER_NOTIFICATION, false, 0);
+		if (dp_display_is_sink_count_zero(dp)) {
+			DRM_DEBUG_DP("sink count is zero, nothing to do\n");
+			return 0;
+		}
+
+		return dp_display_process_hpd_high(dp);
+	}
+
+	dp_ctrl_handle_sink_request(dp->ctrl);
+
+	if (dp->link->sink_request & DP_TEST_LINK_VIDEO_PATTERN)
+		dp_display_handle_video_request(dp);
+
+	return 0;
+}
+
+static int dp_display_usbpd_attention_cb(struct device *dev)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+
+	if (!dev) {
+		DRM_ERROR("invalid dev\n");
+		return -EINVAL;
+	}
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+	if (!dp) {
+		DRM_ERROR("no driver data found\n");
+		return -ENODEV;
+	}
+
+	/* check for any test request issued by sink */
+	rc = dp_link_process_request(dp->link);
+	if (!rc)
+		dp_display_handle_irq_hpd(dp);
+
+	return rc;
+}
+
+static int dp_hpd_plug_handle(struct dp_display_private *dp, u32 data)
+{
+	struct dp_usbpd *hpd = dp->usbpd;
+	u32 state;
+	u32 tout = DP_TIMEOUT_5_SECOND;
+	int ret;
+
+	if (!hpd)
+		return 0;
+
+	mutex_lock(&dp->event_mutex);
+
+	state =  atomic_read(&dp->hpd_state);
+	if (state == ST_SUSPEND_PENDING) {
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	if (state == ST_CONNECT_PENDING || state == ST_CONNECTED) {
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	if (state == ST_DISCONNECT_PENDING) {
+		/* wait until ST_DISCONNECTED */
+		dp_add_event(dp, EV_HPD_PLUG_INT, 0, 1); /* delay = 1 */
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	if (state == ST_SUSPENDED)
+		tout = DP_TIMEOUT_NONE;
+
+	atomic_set(&dp->hpd_state, ST_CONNECT_PENDING);
+
+	hpd->hpd_high = 1;
+
+	ret = dp_display_usbpd_configure_cb(&dp->pdev->dev);
+	if (ret) {	/* failed */
+		hpd->hpd_high = 0;
+		atomic_set(&dp->hpd_state, ST_DISCONNECTED);
+	}
+
+	/* start sanity checking */
+	dp_add_event(dp, EV_CONNECT_PENDING_TIMEOUT, 0, tout);
+
+	mutex_unlock(&dp->event_mutex);
+
+	/* uevent will complete connection part */
+	return 0;
+};
+
+static int dp_display_enable(struct dp_display_private *dp, u32 data);
+static int dp_display_disable(struct dp_display_private *dp, u32 data);
+
+static int dp_connect_pending_timeout(struct dp_display_private *dp, u32 data)
+{
+	u32 state;
+
+	mutex_lock(&dp->event_mutex);
+
+	state =  atomic_read(&dp->hpd_state);
+	if (state == ST_CONNECT_PENDING) {
+		dp_display_enable(dp, 0);
+		atomic_set(&dp->hpd_state, ST_CONNECTED);
+	}
+
+	mutex_unlock(&dp->event_mutex);
+
+	return 0;
+}
+
+static void dp_display_handle_plugged_change(struct msm_dp *dp_display,
+		bool plugged)
+{
+	if (dp_display->plugged_cb && dp_display->codec_dev)
+		dp_display->plugged_cb(dp_display->codec_dev, plugged);
+}
+
+static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data)
+{
+	struct dp_usbpd *hpd = dp->usbpd;
+	u32 state;
+
+	if (!hpd)
+		return 0;
+
+	mutex_lock(&dp->event_mutex);
+
+	state = atomic_read(&dp->hpd_state);
+	if (state == ST_SUSPEND_PENDING) {
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	if (state == ST_DISCONNECT_PENDING || state == ST_DISCONNECTED) {
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	if (state == ST_CONNECT_PENDING) {
+		/* wait until CONNECTED */
+		dp_add_event(dp, EV_HPD_UNPLUG_INT, 0, 1); /* delay = 1 */
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	atomic_set(&dp->hpd_state, ST_DISCONNECT_PENDING);
+
+	/* disable HPD plug interrupt until disconnect is done */
+	dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_PLUG_INT_MASK
+				| DP_DP_IRQ_HPD_INT_MASK, false);
+
+	hpd->hpd_high = 0;
+
+	/*
+	 * We don't need separate work for disconnect as
+	 * connect/attention interrupts are disabled
+	 */
+	dp_display_usbpd_disconnect_cb(&dp->pdev->dev);
+
+	/* start sanity checking */
+	dp_add_event(dp, EV_DISCONNECT_PENDING_TIMEOUT, 0, DP_TIMEOUT_5_SECOND);
+
+	/* signal the disconnect event early to ensure proper teardown */
+	dp_display_handle_plugged_change(g_dp_display, false);
+	reinit_completion(&dp->audio_comp);
+
+	dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_PLUG_INT_MASK |
+					DP_DP_IRQ_HPD_INT_MASK, true);
+
+	/* uevent will complete disconnection part */
+	mutex_unlock(&dp->event_mutex);
+	return 0;
+}
+
+static int dp_disconnect_pending_timeout(struct dp_display_private *dp, u32 data)
+{
+	u32 state;
+
+	mutex_lock(&dp->event_mutex);
+
+	state =  atomic_read(&dp->hpd_state);
+	if (state == ST_DISCONNECT_PENDING) {
+		dp_display_disable(dp, 0);
+		atomic_set(&dp->hpd_state, ST_DISCONNECTED);
+	}
+
+	mutex_unlock(&dp->event_mutex);
+
+	return 0;
+}
+
+static int dp_irq_hpd_handle(struct dp_display_private *dp, u32 data)
+{
+	u32 state;
+
+	mutex_lock(&dp->event_mutex);
+
+	/* irq_hpd can happen at either connected or disconnected state */
+	state =  atomic_read(&dp->hpd_state);
+	if (state == ST_SUSPEND_PENDING) {
+		mutex_unlock(&dp->event_mutex);
+		return 0;
+	}
+
+	dp_display_usbpd_attention_cb(&dp->pdev->dev);
+
+	mutex_unlock(&dp->event_mutex);
+
+	return 0;
+}
+
+static void dp_display_deinit_sub_modules(struct dp_display_private *dp)
+{
+	dp_debug_put(dp->debug);
+	dp_ctrl_put(dp->ctrl);
+	dp_panel_put(dp->panel);
+	dp_aux_put(dp->aux);
+	dp_audio_put(dp->audio);
+}
+
+static int dp_init_sub_modules(struct dp_display_private *dp)
+{
+	int rc = 0;
+	struct device *dev = &dp->pdev->dev;
+	struct dp_usbpd_cb *cb = &dp->usbpd_cb;
+	struct dp_panel_in panel_in = {
+		.dev = dev,
+	};
+
+	/* Callback APIs used for cable status change event */
+	cb->configure  = dp_display_usbpd_configure_cb;
+	cb->disconnect = dp_display_usbpd_disconnect_cb;
+	cb->attention  = dp_display_usbpd_attention_cb;
+
+	dp->usbpd = dp_hpd_get(dev, cb);
+	if (IS_ERR(dp->usbpd)) {
+		rc = PTR_ERR(dp->usbpd);
+		DRM_ERROR("failed to initialize hpd, rc = %d\n", rc);
+		dp->usbpd = NULL;
+		goto error;
+	}
+
+	dp->parser = dp_parser_get(dp->pdev);
+	if (IS_ERR(dp->parser)) {
+		rc = PTR_ERR(dp->parser);
+		DRM_ERROR("failed to initialize parser, rc = %d\n", rc);
+		dp->parser = NULL;
+		goto error;
+	}
+
+	dp->catalog = dp_catalog_get(dev, &dp->parser->io);
+	if (IS_ERR(dp->catalog)) {
+		rc = PTR_ERR(dp->catalog);
+		DRM_ERROR("failed to initialize catalog, rc = %d\n", rc);
+		dp->catalog = NULL;
+		goto error;
+	}
+
+	dp->power = dp_power_get(dp->parser);
+	if (IS_ERR(dp->power)) {
+		rc = PTR_ERR(dp->power);
+		DRM_ERROR("failed to initialize power, rc = %d\n", rc);
+		dp->power = NULL;
+		goto error;
+	}
+
+	dp->aux = dp_aux_get(dev, dp->catalog);
+	if (IS_ERR(dp->aux)) {
+		rc = PTR_ERR(dp->aux);
+		DRM_ERROR("failed to initialize aux, rc = %d\n", rc);
+		dp->aux = NULL;
+		goto error;
+	}
+
+	dp->link = dp_link_get(dev, dp->aux);
+	if (IS_ERR(dp->link)) {
+		rc = PTR_ERR(dp->link);
+		DRM_ERROR("failed to initialize link, rc = %d\n", rc);
+		dp->link = NULL;
+		goto error_link;
+	}
+
+	panel_in.aux = dp->aux;
+	panel_in.catalog = dp->catalog;
+	panel_in.link = dp->link;
+
+	dp->panel = dp_panel_get(&panel_in);
+	if (IS_ERR(dp->panel)) {
+		rc = PTR_ERR(dp->panel);
+		DRM_ERROR("failed to initialize panel, rc = %d\n", rc);
+		dp->panel = NULL;
+		goto error_link;
+	}
+
+	dp->ctrl = dp_ctrl_get(dev, dp->link, dp->panel, dp->aux,
+			       dp->power, dp->catalog, dp->parser);
+	if (IS_ERR(dp->ctrl)) {
+		rc = PTR_ERR(dp->ctrl);
+		DRM_ERROR("failed to initialize ctrl, rc = %d\n", rc);
+		dp->ctrl = NULL;
+		goto error_ctrl;
+	}
+
+	dp->audio = dp_audio_get(dp->pdev, dp->panel, dp->catalog);
+	if (IS_ERR(dp->audio)) {
+		rc = PTR_ERR(dp->audio);
+		pr_err("failed to initialize audio, rc = %d\n", rc);
+		dp->audio = NULL;
+		goto error_audio;
+	}
+
+	return rc;
+
+error_audio:
+	dp_ctrl_put(dp->ctrl);
+error_ctrl:
+	dp_panel_put(dp->panel);
+error_link:
+	dp_aux_put(dp->aux);
+error:
+	return rc;
+}
+
+static int dp_display_set_mode(struct msm_dp *dp_display,
+			       struct dp_display_mode *mode)
+{
+	struct dp_display_private *dp;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	dp->panel->dp_mode.drm_mode = mode->drm_mode;
+	dp->panel->dp_mode.bpp = mode->bpp;
+	dp->panel->dp_mode.capabilities = mode->capabilities;
+	dp_panel_init_panel_info(dp->panel);
+	return 0;
+}
+
+static int dp_display_prepare(struct msm_dp *dp)
+{
+	return 0;
+}
+
+static int dp_display_enable(struct dp_display_private *dp, u32 data)
+{
+	int rc = 0;
+	struct msm_dp *dp_display;
+
+	dp_display = g_dp_display;
+
+	if (dp_display->power_on) {
+		DRM_DEBUG_DP("Link already setup, return\n");
+		return 0;
+	}
+
+	rc = dp_ctrl_on_stream(dp->ctrl);
+	if (!rc)
+		dp_display->power_on = true;
+
+	/* complete resume_comp regardless it is armed or not */
+	complete(&dp->resume_comp);
+	return rc;
+}
+
+static int dp_display_post_enable(struct msm_dp *dp_display)
+{
+	struct dp_display_private *dp;
+	u32 rate;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	rate = dp->link->link_params.rate;
+
+	if (dp->audio_supported) {
+		dp->audio->bw_code = drm_dp_link_rate_to_bw_code(rate);
+		dp->audio->lane_count = dp->link->link_params.num_lanes;
+	}
+
+	/* signal the connect event late to synchronize video and display */
+	dp_display_handle_plugged_change(dp_display, true);
+	return 0;
+}
+
+static int dp_display_disable(struct dp_display_private *dp, u32 data)
+{
+	struct msm_dp *dp_display;
+
+	dp_display = g_dp_display;
+
+	if (!dp_display->power_on)
+		return -EINVAL;
+
+	/* wait only if audio was enabled */
+	if (dp_display->audio_enabled) {
+		if (!wait_for_completion_timeout(&dp->audio_comp,
+				HZ * 5))
+			DRM_ERROR("audio comp timeout\n");
+	}
+
+	dp_display->audio_enabled = false;
+
+	dp_ctrl_off(dp->ctrl);
+
+	dp->core_initialized = false;
+
+	dp_display->power_on = false;
+
+	return 0;
+}
+
+static int dp_display_unprepare(struct msm_dp *dp)
+{
+	return 0;
+}
+
+int dp_display_set_plugged_cb(struct msm_dp *dp_display,
+		hdmi_codec_plugged_cb fn, struct device *codec_dev)
+{
+	bool plugged;
+
+	dp_display->plugged_cb = fn;
+	dp_display->codec_dev = codec_dev;
+	plugged = dp_display->is_connected;
+	dp_display_handle_plugged_change(dp_display, plugged);
+
+	return 0;
+}
+
+int dp_display_validate_mode(struct msm_dp *dp, u32 mode_pclk_khz)
+{
+	const u32 num_components = 3, default_bpp = 24;
+	struct dp_display_private *dp_display;
+	struct dp_link_info *link_info;
+	u32 mode_rate_khz = 0, supported_rate_khz = 0, mode_bpp = 0;
+
+	if (!dp || !mode_pclk_khz || !dp->connector) {
+		DRM_ERROR("invalid params\n");
+		return -EINVAL;
+	}
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+	link_info = &dp_display->panel->link_info;
+
+	mode_bpp = dp->connector->display_info.bpc * num_components;
+	if (!mode_bpp)
+		mode_bpp = default_bpp;
+
+	mode_bpp = dp_panel_get_mode_bpp(dp_display->panel,
+			mode_bpp, mode_pclk_khz);
+
+	mode_rate_khz = mode_pclk_khz * mode_bpp;
+	supported_rate_khz = link_info->num_lanes * link_info->rate * 8;
+
+	if (mode_rate_khz > supported_rate_khz)
+		return MODE_BAD;
+
+	return MODE_OK;
+}
+
+int dp_display_get_modes(struct msm_dp *dp,
+				struct dp_display_mode *dp_mode)
+{
+	struct dp_display_private *dp_display;
+	int ret = 0;
+
+	if (!dp) {
+		DRM_ERROR("invalid params\n");
+		return 0;
+	}
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	ret = dp_panel_get_modes(dp_display->panel,
+		dp->connector, dp_mode);
+	if (dp_mode->drm_mode.clock)
+		dp->max_pclk_khz = dp_mode->drm_mode.clock;
+	return ret;
+}
+
+bool dp_display_check_video_test(struct msm_dp *dp)
+{
+	struct dp_display_private *dp_display;
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	return dp_display->panel->video_test;
+}
+
+int dp_display_get_test_bpp(struct msm_dp *dp)
+{
+	struct dp_display_private *dp_display;
+
+	if (!dp) {
+		DRM_ERROR("invalid params\n");
+		return 0;
+	}
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	return dp_link_bit_depth_to_bpp(
+		dp_display->link->test_video.test_bit_depth);
+}
+
+static void dp_display_config_hpd(struct dp_display_private *dp)
+{
+
+	dp_display_host_init(dp);
+	dp_catalog_ctrl_hpd_config(dp->catalog);
+
+	/* Enable interrupt first time
+	 * we are leaving dp clocks on during disconnect
+	 * and never disable interrupt
+	 */
+	enable_irq(dp->irq);
+}
+
+static int hpd_event_thread(void *data)
+{
+	struct dp_display_private *dp_priv;
+	unsigned long flag;
+	struct dp_event *todo;
+	int timeout_mode = 0;
+
+	dp_priv = (struct dp_display_private *)data;
+
+	while (1) {
+		if (timeout_mode) {
+			wait_event_timeout(dp_priv->event_q,
+				(dp_priv->event_pndx == dp_priv->event_gndx),
+						EVENT_TIMEOUT);
+		} else {
+			wait_event_interruptible(dp_priv->event_q,
+				(dp_priv->event_pndx != dp_priv->event_gndx));
+		}
+		spin_lock_irqsave(&dp_priv->event_lock, flag);
+		todo = &dp_priv->event_list[dp_priv->event_gndx];
+		if (todo->delay) {
+			struct dp_event *todo_next;
+
+			dp_priv->event_gndx++;
+			dp_priv->event_gndx %= DP_EVENT_Q_MAX;
+
+			/* re enter delay event into q */
+			todo_next = &dp_priv->event_list[dp_priv->event_pndx++];
+			dp_priv->event_pndx %= DP_EVENT_Q_MAX;
+			todo_next->event_id = todo->event_id;
+			todo_next->data = todo->data;
+			todo_next->delay = todo->delay - 1;
+
+			/* clean up older event */
+			todo->event_id = EV_NO_EVENT;
+			todo->delay = 0;
+
+			/* switch to timeout mode */
+			timeout_mode = 1;
+			spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+			continue;
+		}
+
+		/* timeout with no events in q */
+		if (dp_priv->event_pndx == dp_priv->event_gndx) {
+			spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+			continue;
+		}
+
+		dp_priv->event_gndx++;
+		dp_priv->event_gndx %= DP_EVENT_Q_MAX;
+		timeout_mode = 0;
+		spin_unlock_irqrestore(&dp_priv->event_lock, flag);
+
+		switch (todo->event_id) {
+		case EV_HPD_INIT_SETUP:
+			dp_display_config_hpd(dp_priv);
+			break;
+		case EV_HPD_PLUG_INT:
+			dp_hpd_plug_handle(dp_priv, todo->data);
+			break;
+		case EV_HPD_UNPLUG_INT:
+			dp_hpd_unplug_handle(dp_priv, todo->data);
+			break;
+		case EV_IRQ_HPD_INT:
+			dp_irq_hpd_handle(dp_priv, todo->data);
+			break;
+		case EV_HPD_REPLUG_INT:
+			/* do nothing */
+			break;
+		case EV_USER_NOTIFICATION:
+			dp_display_send_hpd_notification(dp_priv,
+						todo->data);
+			break;
+		case EV_CONNECT_PENDING_TIMEOUT:
+			dp_connect_pending_timeout(dp_priv,
+						todo->data);
+			break;
+		case EV_DISCONNECT_PENDING_TIMEOUT:
+			dp_disconnect_pending_timeout(dp_priv,
+						todo->data);
+			break;
+		default:
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static void dp_hpd_event_setup(struct dp_display_private *dp_priv)
+{
+	init_waitqueue_head(&dp_priv->event_q);
+	spin_lock_init(&dp_priv->event_lock);
+
+	kthread_run(hpd_event_thread, dp_priv, "dp_hpd_handler");
+}
+
+static irqreturn_t dp_display_irq_handler(int irq, void *dev_id)
+{
+	struct dp_display_private *dp = dev_id;
+	irqreturn_t ret = IRQ_HANDLED;
+	u32 hpd_isr_status;
+
+	if (!dp) {
+		DRM_ERROR("invalid data\n");
+		return IRQ_NONE;
+	}
+
+	hpd_isr_status = dp_catalog_hpd_get_intr_status(dp->catalog);
+
+	if (hpd_isr_status & 0x0F) {
+		/* hpd related interrupts */
+		if (hpd_isr_status & DP_DP_HPD_PLUG_INT_MASK ||
+			hpd_isr_status & DP_DP_HPD_REPLUG_INT_MASK) {
+			dp_add_event(dp, EV_HPD_PLUG_INT, 0, 0);
+		}
+
+		if (hpd_isr_status & DP_DP_IRQ_HPD_INT_MASK) {
+			/* delete connect pending event first */
+			dp_del_event(dp, EV_CONNECT_PENDING_TIMEOUT);
+			dp_add_event(dp, EV_IRQ_HPD_INT, 0, 0);
+		}
+
+		if (hpd_isr_status & DP_DP_HPD_REPLUG_INT_MASK)
+			dp_add_event(dp, EV_HPD_REPLUG_INT, 0, 0);
+
+		if (hpd_isr_status & DP_DP_HPD_UNPLUG_INT_MASK)
+			dp_add_event(dp, EV_HPD_UNPLUG_INT, 0, 0);
+	}
+
+	/* DP controller isr */
+	dp_ctrl_isr(dp->ctrl);
+
+	/* DP aux isr */
+	dp_aux_isr(dp->aux);
+
+	return ret;
+}
+
+int dp_display_request_irq(struct msm_dp *dp_display)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+
+	if (!dp_display) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	dp->irq = irq_of_parse_and_map(dp->pdev->dev.of_node, 0);
+	if (dp->irq < 0) {
+		rc = dp->irq;
+		DRM_ERROR("failed to get irq: %d\n", rc);
+		return rc;
+	}
+
+	rc = devm_request_irq(&dp->pdev->dev, dp->irq,
+			dp_display_irq_handler,
+			IRQF_TRIGGER_HIGH, "dp_display_isr", dp);
+	if (rc < 0) {
+		DRM_ERROR("failed to request IRQ%u: %d\n",
+				dp->irq, rc);
+		return rc;
+	}
+	disable_irq(dp->irq);
+
+	return 0;
+}
+
+static int dp_display_probe(struct platform_device *pdev)
+{
+	int rc = 0;
+	struct dp_display_private *dp;
+
+	if (!pdev || !pdev->dev.of_node) {
+		DRM_ERROR("pdev not found\n");
+		return -ENODEV;
+	}
+
+	dp = devm_kzalloc(&pdev->dev, sizeof(*dp), GFP_KERNEL);
+	if (!dp)
+		return -ENOMEM;
+
+	dp->pdev = pdev;
+	dp->name = "drm_dp";
+
+	rc = dp_init_sub_modules(dp);
+	if (rc) {
+		DRM_ERROR("init sub module failed\n");
+		return -EPROBE_DEFER;
+	}
+
+	mutex_init(&dp->event_mutex);
+
+	init_completion(&dp->resume_comp);
+
+	g_dp_display = &dp->dp_display;
+
+	/* Store DP audio handle inside DP display */
+	g_dp_display->dp_audio = dp->audio;
+
+	init_completion(&dp->audio_comp);
+
+	platform_set_drvdata(pdev, g_dp_display);
+
+	rc = component_add(&pdev->dev, &dp_display_comp_ops);
+	if (rc) {
+		DRM_ERROR("component add failed, rc=%d\n", rc);
+		dp_display_deinit_sub_modules(dp);
+	}
+
+	return rc;
+}
+
+static int dp_display_remove(struct platform_device *pdev)
+{
+	struct dp_display_private *dp;
+
+	dp = container_of(g_dp_display,
+			struct dp_display_private, dp_display);
+
+	dp_display_deinit_sub_modules(dp);
+
+	component_del(&pdev->dev, &dp_display_comp_ops);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static int dp_pm_resume(struct device *dev)
+{
+	return 0;
+}
+
+static int dp_pm_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct dp_display_private *dp = platform_get_drvdata(pdev);
+
+	if (!dp) {
+		DRM_ERROR("DP driver bind failed. Invalid driver data\n");
+		return -EINVAL;
+	}
+
+	atomic_set(&dp->hpd_state, ST_SUSPENDED);
+
+	return 0;
+}
+
+static int dp_pm_prepare(struct device *dev)
+{
+	return 0;
+}
+
+static void dp_pm_complete(struct device *dev)
+{
+
+}
+
+static const struct dev_pm_ops dp_pm_ops = {
+	.suspend = dp_pm_suspend,
+	.resume =  dp_pm_resume,
+	.prepare = dp_pm_prepare,
+	.complete = dp_pm_complete,
+};
+
+static struct platform_driver dp_display_driver = {
+	.probe  = dp_display_probe,
+	.remove = dp_display_remove,
+	.driver = {
+		.name = "msm-dp-display",
+		.of_match_table = dp_dt_match,
+		.suppress_bind_attrs = true,
+		.pm = &dp_pm_ops,
+	},
+};
+
+int __init msm_dp_register(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&dp_display_driver);
+	if (ret)
+		DRM_ERROR("Dp display driver register failed");
+
+	return ret;
+}
+
+void __exit msm_dp_unregister(void)
+{
+	platform_driver_unregister(&dp_display_driver);
+}
+
+void msm_dp_irq_postinstall(struct msm_dp *dp_display)
+{
+	struct dp_display_private *dp;
+
+	if (!dp_display)
+		return;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+
+	dp_hpd_event_setup(dp);
+
+	dp_add_event(dp, EV_HPD_INIT_SETUP, 0, 100);
+}
+
+void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor)
+{
+	struct dp_display_private *dp;
+	struct device *dev;
+	int rc;
+
+	dp = container_of(dp_display, struct dp_display_private, dp_display);
+	dev = &dp->pdev->dev;
+
+	dp->debug = dp_debug_get(dev, dp->panel, dp->usbpd,
+					dp->link, &dp->dp_display.connector,
+					minor);
+	if (IS_ERR(dp->debug)) {
+		rc = PTR_ERR(dp->debug);
+		DRM_ERROR("failed to initialize debug, rc = %d\n", rc);
+		dp->debug = NULL;
+	}
+}
+
+int msm_dp_modeset_init(struct msm_dp *dp_display, struct drm_device *dev,
+			struct drm_encoder *encoder)
+{
+	struct msm_drm_private *priv;
+	int ret;
+
+	if (WARN_ON(!encoder) || WARN_ON(!dp_display) || WARN_ON(!dev))
+		return -EINVAL;
+
+	priv = dev->dev_private;
+	dp_display->drm_dev = dev;
+
+	ret = dp_display_request_irq(dp_display);
+	if (ret) {
+		DRM_ERROR("request_irq failed, ret=%d\n", ret);
+		return ret;
+	}
+
+	dp_display->encoder = encoder;
+
+	dp_display->connector = dp_drm_connector_init(dp_display);
+	if (IS_ERR(dp_display->connector)) {
+		ret = PTR_ERR(dp_display->connector);
+		DRM_DEV_ERROR(dev->dev,
+			"failed to create dp connector: %d\n", ret);
+		dp_display->connector = NULL;
+		return ret;
+	}
+
+	priv->connectors[priv->num_connectors++] = dp_display->connector;
+	return 0;
+}
+
+static int dp_display_wait4resume_done(struct dp_display_private *dp)
+{
+	int ret = 0;
+
+	reinit_completion(&dp->resume_comp);
+	if (!wait_for_completion_timeout(&dp->resume_comp,
+				WAIT_FOR_RESUME_TIMEOUT_JIFFIES)) {
+		DRM_ERROR("wait4resume_done timedout\n");
+		ret = -ETIMEDOUT;
+	}
+	return ret;
+}
+
+int msm_dp_display_enable(struct msm_dp *dp, struct drm_encoder *encoder)
+{
+	int rc = 0;
+	struct dp_display_private *dp_display;
+	u32 state;
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+	if (!dp_display->dp_mode.drm_mode.clock) {
+		DRM_ERROR("invalid params\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&dp_display->event_mutex);
+
+	rc = dp_display_set_mode(dp, &dp_display->dp_mode);
+	if (rc) {
+		DRM_ERROR("Failed to perform a mode set, rc=%d\n", rc);
+		mutex_unlock(&dp_display->event_mutex);
+		return rc;
+	}
+
+	rc = dp_display_prepare(dp);
+	if (rc) {
+		DRM_ERROR("DP display prepare failed, rc=%d\n", rc);
+		mutex_unlock(&dp_display->event_mutex);
+		return rc;
+	}
+
+	state =  atomic_read(&dp_display->hpd_state);
+	if (state == ST_SUSPENDED) {
+		/* start link training */
+		dp_add_event(dp_display, EV_HPD_PLUG_INT, 0, 0);
+		mutex_unlock(&dp_display->event_mutex);
+
+		/* wait until dp interface is up */
+		goto resume_done;
+	}
+
+	dp_display_enable(dp_display, 0);
+
+	rc = dp_display_post_enable(dp);
+	if (rc) {
+		DRM_ERROR("DP display post enable failed, rc=%d\n", rc);
+		dp_display_disable(dp_display, 0);
+		dp_display_unprepare(dp);
+	}
+
+	dp_del_event(dp_display, EV_CONNECT_PENDING_TIMEOUT);
+
+	if (state == ST_SUSPEND_PENDING)
+		dp_add_event(dp_display, EV_IRQ_HPD_INT, 0, 0);
+
+	/* completed connection */
+	atomic_set(&dp_display->hpd_state, ST_CONNECTED);
+
+	mutex_unlock(&dp_display->event_mutex);
+
+	return rc;
+
+resume_done:
+	dp_display_wait4resume_done(dp_display);
+	return rc;
+}
+
+int msm_dp_display_pre_disable(struct msm_dp *dp, struct drm_encoder *encoder)
+{
+	struct dp_display_private *dp_display;
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	dp_ctrl_push_idle(dp_display->ctrl);
+
+	return 0;
+}
+
+int msm_dp_display_disable(struct msm_dp *dp, struct drm_encoder *encoder)
+{
+	int rc = 0;
+	u32 state;
+	struct dp_display_private *dp_display;
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	mutex_lock(&dp_display->event_mutex);
+
+	dp_display_disable(dp_display, 0);
+
+	rc = dp_display_unprepare(dp);
+	if (rc)
+		DRM_ERROR("DP display unprepare failed, rc=%d\n", rc);
+
+	dp_del_event(dp_display, EV_DISCONNECT_PENDING_TIMEOUT);
+
+	state =  atomic_read(&dp_display->hpd_state);
+	if (state == ST_DISCONNECT_PENDING) {
+		/* completed disconnection */
+		atomic_set(&dp_display->hpd_state, ST_DISCONNECTED);
+	} else {
+		atomic_set(&dp_display->hpd_state, ST_SUSPEND_PENDING);
+	}
+
+	mutex_unlock(&dp_display->event_mutex);
+	return rc;
+}
+
+void msm_dp_display_mode_set(struct msm_dp *dp, struct drm_encoder *encoder,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct dp_display_private *dp_display;
+
+	dp_display = container_of(dp, struct dp_display_private, dp_display);
+
+	memset(&dp_display->dp_mode, 0x0, sizeof(struct dp_display_mode));
+
+	if (dp_display_check_video_test(dp))
+		dp_display->dp_mode.bpp = dp_display_get_test_bpp(dp);
+	else /* Default num_components per pixel = 3 */
+		dp_display->dp_mode.bpp = dp->connector->display_info.bpc * 3;
+
+	if (!dp_display->dp_mode.bpp)
+		dp_display->dp_mode.bpp = 24; /* Default bpp */
+
+	drm_mode_copy(&dp_display->dp_mode.drm_mode, adjusted_mode);
+
+	dp_display->dp_mode.v_active_low =
+		!!(dp_display->dp_mode.drm_mode.flags & DRM_MODE_FLAG_NVSYNC);
+
+	dp_display->dp_mode.h_active_low =
+		!!(dp_display->dp_mode.drm_mode.flags & DRM_MODE_FLAG_NHSYNC);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_display.h b/drivers/gpu/drm/msm/dp/dp_display.h
new file mode 100644
index 000000000000..6092ba1ed85e
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_display.h
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_DISPLAY_H_
+#define _DP_DISPLAY_H_
+
+#include "dp_panel.h"
+#include <sound/hdmi-codec.h>
+
+struct msm_dp {
+	struct drm_device *drm_dev;
+	struct device *codec_dev;
+	struct drm_connector *connector;
+	struct drm_encoder *encoder;
+	bool is_connected;
+	bool audio_enabled;
+	bool power_on;
+
+	hdmi_codec_plugged_cb plugged_cb;
+
+	u32 max_pclk_khz;
+
+	u32 max_dp_lanes;
+	struct dp_audio *dp_audio;
+};
+
+int dp_display_set_plugged_cb(struct msm_dp *dp_display,
+		hdmi_codec_plugged_cb fn, struct device *codec_dev);
+int dp_display_validate_mode(struct msm_dp *dp_display, u32 mode_pclk_khz);
+int dp_display_get_modes(struct msm_dp *dp_display,
+		struct dp_display_mode *dp_mode);
+int dp_display_request_irq(struct msm_dp *dp_display);
+bool dp_display_check_video_test(struct msm_dp *dp_display);
+int dp_display_get_test_bpp(struct msm_dp *dp_display);
+void dp_display_signal_audio_complete(struct msm_dp *dp_display);
+
+#endif /* _DP_DISPLAY_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_drm.c b/drivers/gpu/drm/msm/dp/dp_drm.c
new file mode 100644
index 000000000000..764f4b81017e
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_drm.c
@@ -0,0 +1,164 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_crtc.h>
+
+#include "msm_drv.h"
+#include "msm_kms.h"
+#include "dp_drm.h"
+
+struct dp_connector {
+	struct drm_connector base;
+	struct msm_dp *dp_display;
+};
+#define to_dp_connector(x) container_of(x, struct dp_connector, base)
+
+/**
+ * dp_connector_detect - callback to determine if connector is connected
+ * @conn: Pointer to drm connector structure
+ * @force: Force detect setting from drm framework
+ * Returns: Connector 'is connected' status
+ */
+static enum drm_connector_status dp_connector_detect(struct drm_connector *conn,
+		bool force)
+{
+	struct msm_dp *dp;
+
+	dp = to_dp_connector(conn)->dp_display;
+
+	DRM_DEBUG_DP("is_connected = %s\n",
+		(dp->is_connected) ? "true" : "false");
+
+	return (dp->is_connected) ? connector_status_connected :
+					connector_status_disconnected;
+}
+
+/**
+ * dp_connector_get_modes - callback to add drm modes via drm_mode_probed_add()
+ * @connector: Pointer to drm connector structure
+ * Returns: Number of modes added
+ */
+static int dp_connector_get_modes(struct drm_connector *connector)
+{
+	int rc = 0;
+	struct msm_dp *dp;
+	struct dp_display_mode *dp_mode = NULL;
+	struct drm_display_mode *m, drm_mode;
+
+	if (!connector)
+		return 0;
+
+	dp = to_dp_connector(connector)->dp_display;
+
+	dp_mode = kzalloc(sizeof(*dp_mode),  GFP_KERNEL);
+	if (!dp_mode)
+		return 0;
+
+	/* pluggable case assumes EDID is read when HPD */
+	if (dp->is_connected) {
+		/*
+		 *The get_modes() function might return one mode that is stored
+		 * in dp_mode when compliance test is in progress. If not, the
+		 * return value is equal to the total number of modes supported
+		 * by the sink
+		 */
+		rc = dp_display_get_modes(dp, dp_mode);
+		if (rc <= 0) {
+			DRM_ERROR("failed to get DP sink modes, rc=%d\n", rc);
+			kfree(dp_mode);
+			return rc;
+		}
+		if (dp_mode->drm_mode.clock) { /* valid DP mode */
+			memset(&drm_mode, 0x0, sizeof(drm_mode));
+			drm_mode_copy(&drm_mode, &dp_mode->drm_mode);
+			m = drm_mode_duplicate(connector->dev, &drm_mode);
+			if (!m) {
+				DRM_ERROR("failed to add mode %ux%u\n",
+				       drm_mode.hdisplay,
+				       drm_mode.vdisplay);
+				kfree(dp_mode);
+				return 0;
+			}
+			drm_mode_probed_add(connector, m);
+		}
+	} else {
+		DRM_DEBUG_DP("No sink connected\n");
+	}
+	kfree(dp_mode);
+	return rc;
+}
+
+/**
+ * dp_connector_mode_valid - callback to determine if specified mode is valid
+ * @connector: Pointer to drm connector structure
+ * @mode: Pointer to drm mode structure
+ * Returns: Validity status for specified mode
+ */
+static enum drm_mode_status dp_connector_mode_valid(
+		struct drm_connector *connector,
+		struct drm_display_mode *mode)
+{
+	struct msm_dp *dp_disp;
+
+	dp_disp = to_dp_connector(connector)->dp_display;
+
+	if ((dp_disp->max_pclk_khz <= 0) ||
+			(dp_disp->max_pclk_khz > DP_MAX_PIXEL_CLK_KHZ) ||
+			(mode->clock > dp_disp->max_pclk_khz))
+		return MODE_BAD;
+
+	return dp_display_validate_mode(dp_disp, mode->clock);
+}
+
+static const struct drm_connector_funcs dp_connector_funcs = {
+	.detect = dp_connector_detect,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static const struct drm_connector_helper_funcs dp_connector_helper_funcs = {
+	.get_modes = dp_connector_get_modes,
+	.mode_valid = dp_connector_mode_valid,
+};
+
+/* connector initialization */
+struct drm_connector *dp_drm_connector_init(struct msm_dp *dp_display)
+{
+	struct drm_connector *connector = NULL;
+	struct dp_connector *dp_connector;
+	int ret;
+
+	dp_connector = devm_kzalloc(dp_display->drm_dev->dev,
+					sizeof(*dp_connector),
+					GFP_KERNEL);
+	if (!dp_connector)
+		return ERR_PTR(-ENOMEM);
+
+	dp_connector->dp_display = dp_display;
+
+	connector = &dp_connector->base;
+
+	ret = drm_connector_init(dp_display->drm_dev, connector,
+			&dp_connector_funcs,
+			DRM_MODE_CONNECTOR_DisplayPort);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_connector_helper_add(connector, &dp_connector_helper_funcs);
+
+	/*
+	 * Enable HPD to let hpd event is handled when cable is connected.
+	 */
+	connector->polled = DRM_CONNECTOR_POLL_HPD;
+
+	drm_connector_attach_encoder(connector, dp_display->encoder);
+
+	return connector;
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_drm.h b/drivers/gpu/drm/msm/dp/dp_drm.h
new file mode 100644
index 000000000000..c27bfceefdf0
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_drm.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_DRM_H_
+#define _DP_DRM_H_
+
+#include <linux/types.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_crtc_helper.h>
+
+#include "msm_drv.h"
+#include "dp_display.h"
+
+struct drm_connector *dp_drm_connector_init(struct msm_dp *dp_display);
+
+#endif /* _DP_DRM_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.c b/drivers/gpu/drm/msm/dp/dp_hpd.c
new file mode 100644
index 000000000000..5b8fe32022b5
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_hpd.c
@@ -0,0 +1,69 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <linux/slab.h>
+#include <linux/device.h>
+
+#include "dp_hpd.h"
+
+/* DP specific VDM commands */
+#define DP_USBPD_VDM_STATUS	0x10
+#define DP_USBPD_VDM_CONFIGURE	0x11
+
+/* USBPD-TypeC specific Macros */
+#define VDM_VERSION		0x0
+#define USB_C_DP_SID		0xFF01
+
+struct dp_hpd_private {
+	struct device *dev;
+	struct dp_usbpd_cb *dp_cb;
+	struct dp_usbpd dp_usbpd;
+};
+
+int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd)
+{
+	int rc = 0;
+	struct dp_hpd_private *hpd_priv;
+
+	hpd_priv = container_of(dp_usbpd, struct dp_hpd_private,
+					dp_usbpd);
+
+	dp_usbpd->hpd_high = hpd;
+
+	if (!hpd_priv->dp_cb && !hpd_priv->dp_cb->configure
+				&& !hpd_priv->dp_cb->disconnect) {
+		pr_err("hpd dp_cb not initialized\n");
+		return -EINVAL;
+	}
+	if (hpd)
+		hpd_priv->dp_cb->configure(hpd_priv->dev);
+	else
+		hpd_priv->dp_cb->disconnect(hpd_priv->dev);
+
+	return rc;
+}
+
+struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb)
+{
+	struct dp_hpd_private *dp_hpd;
+
+	if (!cb) {
+		pr_err("invalid cb data\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	dp_hpd = devm_kzalloc(dev, sizeof(*dp_hpd), GFP_KERNEL);
+	if (!dp_hpd)
+		return ERR_PTR(-ENOMEM);
+
+	dp_hpd->dev = dev;
+	dp_hpd->dp_cb = cb;
+
+	dp_hpd->dp_usbpd.connect = dp_hpd_connect;
+
+	return &dp_hpd->dp_usbpd;
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.h b/drivers/gpu/drm/msm/dp/dp_hpd.h
new file mode 100644
index 000000000000..5bc5bb64680f
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_hpd.h
@@ -0,0 +1,80 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_HPD_H_
+#define _DP_HPD_H_
+
+//#include <linux/usb/usbpd.h>
+
+#include <linux/types.h>
+#include <linux/device.h>
+
+enum plug_orientation {
+	ORIENTATION_NONE,
+	ORIENTATION_CC1,
+	ORIENTATION_CC2,
+};
+
+/**
+ * struct dp_usbpd - DisplayPort status
+ *
+ * @orientation: plug orientation configuration
+ * @low_pow_st: low power state
+ * @adaptor_dp_en: adaptor functionality enabled
+ * @multi_func: multi-function preferred
+ * @usb_config_req: request to switch to usb
+ * @exit_dp_mode: request exit from displayport mode
+ * @hpd_high: Hot Plug Detect signal is high.
+ * @hpd_irq: Change in the status since last message
+ * @alt_mode_cfg_done: bool to specify alt mode status
+ * @debug_en: bool to specify debug mode
+ * @connect: simulate disconnect or connect for debug mode
+ */
+struct dp_usbpd {
+	enum plug_orientation orientation;
+	bool low_pow_st;
+	bool adaptor_dp_en;
+	bool multi_func;
+	bool usb_config_req;
+	bool exit_dp_mode;
+	bool hpd_high;
+	bool hpd_irq;
+	bool alt_mode_cfg_done;
+	bool debug_en;
+
+	int (*connect)(struct dp_usbpd *dp_usbpd, bool hpd);
+};
+
+/**
+ * struct dp_usbpd_cb - callback functions provided by the client
+ *
+ * @configure: called by usbpd module when PD communication has
+ * been completed and the usb peripheral has been configured on
+ * dp mode.
+ * @disconnect: notify the cable disconnect issued by usb.
+ * @attention: notify any attention message issued by usb.
+ */
+struct dp_usbpd_cb {
+	int (*configure)(struct device *dev);
+	int (*disconnect)(struct device *dev);
+	int (*attention)(struct device *dev);
+};
+
+/**
+ * dp_hpd_get() - setup hpd module
+ *
+ * @dev: device instance of the caller
+ * @cb: struct containing callback function pointers.
+ *
+ * This function allows the client to initialize the usbpd
+ * module. The module will communicate with HPD module.
+ */
+struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb);
+
+int dp_hpd_register(struct dp_usbpd *dp_usbpd);
+void dp_hpd_unregister(struct dp_usbpd *dp_usbpd);
+int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd);
+
+#endif /* _DP_HPD_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_link.c b/drivers/gpu/drm/msm/dp/dp_link.c
new file mode 100644
index 000000000000..c811da515fb3
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_link.c
@@ -0,0 +1,1210 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <drm/drm_print.h>
+
+#include "dp_link.h"
+#include "dp_panel.h"
+
+#define DP_TEST_REQUEST_MASK		0x7F
+
+enum audio_sample_rate {
+	AUDIO_SAMPLE_RATE_32_KHZ	= 0x00,
+	AUDIO_SAMPLE_RATE_44_1_KHZ	= 0x01,
+	AUDIO_SAMPLE_RATE_48_KHZ	= 0x02,
+	AUDIO_SAMPLE_RATE_88_2_KHZ	= 0x03,
+	AUDIO_SAMPLE_RATE_96_KHZ	= 0x04,
+	AUDIO_SAMPLE_RATE_176_4_KHZ	= 0x05,
+	AUDIO_SAMPLE_RATE_192_KHZ	= 0x06,
+};
+
+enum audio_pattern_type {
+	AUDIO_TEST_PATTERN_OPERATOR_DEFINED	= 0x00,
+	AUDIO_TEST_PATTERN_SAWTOOTH		= 0x01,
+};
+
+struct dp_link_request {
+	u32 test_requested;
+	u32 test_link_rate;
+	u32 test_lane_count;
+};
+
+struct dp_link_private {
+	u32 prev_sink_count;
+	struct device *dev;
+	struct drm_dp_aux *aux;
+	struct dp_link dp_link;
+
+	struct dp_link_request request;
+	struct mutex psm_mutex;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+};
+
+static int dp_aux_link_power_up(struct drm_dp_aux *aux,
+					struct dp_link_info *link)
+{
+	u8 value;
+	int err;
+
+	if (link->revision < 0x11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D0;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	usleep_range(1000, 2000);
+
+	return 0;
+}
+
+static int dp_aux_link_power_down(struct drm_dp_aux *aux,
+					struct dp_link_info *link)
+{
+	u8 value;
+	int err;
+
+	if (link->revision < 0x11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D3;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static int dp_link_get_period(struct dp_link_private *link, int const addr)
+{
+	int ret = 0;
+	u8 data;
+	u32 const max_audio_period = 0xA;
+
+	/* TEST_AUDIO_PERIOD_CH_XX */
+	if (drm_dp_dpcd_readb(link->aux, addr, &data) < 0) {
+		DRM_ERROR("failed to read test_audio_period (0x%x)\n", addr);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	/* Period - Bits 3:0 */
+	data = data & 0xF;
+	if ((int)data > max_audio_period) {
+		DRM_ERROR("invalid test_audio_period_ch_1 = 0x%x\n", data);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	ret = data;
+exit:
+	return ret;
+}
+
+static int dp_link_parse_audio_channel_period(struct dp_link_private *link)
+{
+	int ret = 0;
+	struct dp_link_test_audio *req = &link->dp_link.test_audio;
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH1);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_1 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_1 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH2);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_2 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_2 = 0x%x\n", ret);
+
+	/* TEST_AUDIO_PERIOD_CH_3 (Byte 0x275) */
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH3);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_3 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_3 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH4);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_4 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_4 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH5);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_5 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_5 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH6);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_6 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_6 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH7);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_7 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_7 = 0x%x\n", ret);
+
+	ret = dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH8);
+	if (ret == -EINVAL)
+		goto exit;
+
+	req->test_audio_period_ch_8 = ret;
+	DRM_DEBUG_DP("test_audio_period_ch_8 = 0x%x\n", ret);
+exit:
+	return ret;
+}
+
+static int dp_link_parse_audio_pattern_type(struct dp_link_private *link)
+{
+	int ret = 0;
+	u8 data;
+	ssize_t rlen;
+	int const max_audio_pattern_type = 0x1;
+
+	rlen = drm_dp_dpcd_readb(link->aux,
+				DP_TEST_AUDIO_PATTERN_TYPE, &data);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read link audio mode. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	/* Audio Pattern Type - Bits 7:0 */
+	if ((int)data > max_audio_pattern_type) {
+		DRM_ERROR("invalid audio pattern type = 0x%x\n", data);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	link->dp_link.test_audio.test_audio_pattern_type = data;
+	DRM_DEBUG_DP("audio pattern type = 0x%x\n", data);
+exit:
+	return ret;
+}
+
+static int dp_link_parse_audio_mode(struct dp_link_private *link)
+{
+	int ret = 0;
+	u8 data;
+	ssize_t rlen;
+	int const max_audio_sampling_rate = 0x6;
+	int const max_audio_channel_count = 0x8;
+	int sampling_rate = 0x0;
+	int channel_count = 0x0;
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_AUDIO_MODE, &data);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read link audio mode. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	/* Sampling Rate - Bits 3:0 */
+	sampling_rate = data & 0xF;
+	if (sampling_rate > max_audio_sampling_rate) {
+		DRM_ERROR("sampling rate (0x%x) greater than max (0x%x)\n",
+				sampling_rate, max_audio_sampling_rate);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	/* Channel Count - Bits 7:4 */
+	channel_count = ((data & 0xF0) >> 4) + 1;
+	if (channel_count > max_audio_channel_count) {
+		DRM_ERROR("channel_count (0x%x) greater than max (0x%x)\n",
+				channel_count, max_audio_channel_count);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	link->dp_link.test_audio.test_audio_sampling_rate = sampling_rate;
+	link->dp_link.test_audio.test_audio_channel_count = channel_count;
+	DRM_DEBUG_DP("sampling_rate = 0x%x, channel_count = 0x%x\n",
+					sampling_rate, channel_count);
+exit:
+	return ret;
+}
+
+static int dp_link_parse_audio_pattern_params(struct dp_link_private *link)
+{
+	int ret = 0;
+
+	ret = dp_link_parse_audio_mode(link);
+	if (ret)
+		goto exit;
+
+	ret = dp_link_parse_audio_pattern_type(link);
+	if (ret)
+		goto exit;
+
+	ret = dp_link_parse_audio_channel_period(link);
+
+exit:
+	return ret;
+}
+
+static bool dp_link_is_video_pattern_valid(u32 pattern)
+{
+	switch (pattern) {
+	case DP_NO_TEST_PATTERN:
+	case DP_COLOR_RAMP:
+	case DP_BLACK_AND_WHITE_VERTICAL_LINES:
+	case DP_COLOR_SQUARE:
+		return true;
+	default:
+		return false;
+	}
+}
+
+/**
+ * dp_link_is_bit_depth_valid() - validates the bit depth requested
+ * @tbd: bit depth requested by the sink
+ *
+ * Returns true if the requested bit depth is supported.
+ */
+static bool dp_link_is_bit_depth_valid(u32 tbd)
+{
+	/* DP_TEST_VIDEO_PATTERN_NONE is treated as invalid */
+	switch (tbd) {
+	case DP_TEST_BIT_DEPTH_6:
+	case DP_TEST_BIT_DEPTH_8:
+	case DP_TEST_BIT_DEPTH_10:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static int dp_link_parse_timing_params1(struct dp_link_private *link,
+					int addr, int len, u32 *val)
+{
+	u8 bp[2];
+	int rlen;
+
+	if (len != 2)
+		return -EINVAL;
+
+	/* Read the requested video link pattern (Byte 0x221). */
+	rlen = drm_dp_dpcd_read(link->aux, addr, bp, len);
+	if (rlen < len) {
+		DRM_ERROR("failed to read 0x%x\n", addr);
+		return -EINVAL;
+	}
+
+	*val = bp[1] | (bp[0] << 8);
+
+	return 0;
+}
+
+static int dp_link_parse_timing_params2(struct dp_link_private *link,
+					int addr, int len,
+					u32 *val1, u32 *val2)
+{
+	u8 bp[2];
+	int rlen;
+
+	if (len != 2)
+		return -EINVAL;
+
+	/* Read the requested video link pattern (Byte 0x221). */
+	rlen = drm_dp_dpcd_read(link->aux, addr, bp, len);
+	if (rlen < len) {
+		DRM_ERROR("failed to read 0x%x\n", addr);
+		return -EINVAL;
+	}
+
+	*val1 = (bp[0] & BIT(7)) >> 7;
+	*val2 = bp[1] | ((bp[0] & 0x7F) << 8);
+
+	return 0;
+}
+
+static int dp_link_parse_timing_params3(struct dp_link_private *link,
+					int addr, u32 *val)
+{
+	u8 bp;
+	u32 len = 1;
+	int rlen;
+
+	rlen = drm_dp_dpcd_read(link->aux, addr, &bp, len);
+	if (rlen < 1) {
+		DRM_ERROR("failed to read 0x%x\n", addr);
+		return -EINVAL;
+	}
+	*val = bp;
+
+	return 0;
+}
+
+/**
+ * dp_parse_video_pattern_params() - parses video pattern parameters from DPCD
+ * @link: Display Port Driver data
+ *
+ * Returns 0 if it successfully parses the video link pattern and the link
+ * bit depth requested by the sink and, and if the values parsed are valid.
+ */
+static int dp_link_parse_video_pattern_params(struct dp_link_private *link)
+{
+	int ret = 0;
+	ssize_t rlen;
+	u8 bp;
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_PATTERN, &bp);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read link video pattern. rlen=%zd\n",
+			rlen);
+		return rlen;
+	}
+
+	if (!dp_link_is_video_pattern_valid(bp)) {
+		DRM_ERROR("invalid link video pattern = 0x%x\n", bp);
+		ret = -EINVAL;
+		return ret;
+	}
+
+	link->dp_link.test_video.test_video_pattern = bp;
+
+	/* Read the requested color bit depth and dynamic range (Byte 0x232) */
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_MISC0, &bp);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read link bit depth. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	/* Dynamic Range */
+	link->dp_link.test_video.test_dyn_range =
+			(bp & DP_TEST_DYNAMIC_RANGE_CEA);
+
+	/* Color bit depth */
+	bp &= DP_TEST_BIT_DEPTH_MASK;
+	if (!dp_link_is_bit_depth_valid(bp)) {
+		DRM_ERROR("invalid link bit depth = 0x%x\n", bp);
+		ret = -EINVAL;
+		return ret;
+	}
+
+	link->dp_link.test_video.test_bit_depth = bp;
+
+	/* resolution timing params */
+	ret = dp_link_parse_timing_params1(link, DP_TEST_H_TOTAL_HI, 2,
+			&link->dp_link.test_video.test_h_total);
+	if (ret) {
+		DRM_ERROR("failed to parse test_htotal(DP_TEST_H_TOTAL_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params1(link, DP_TEST_V_TOTAL_HI, 2,
+			&link->dp_link.test_video.test_v_total);
+	if (ret) {
+		DRM_ERROR("failed to parse test_v_total(DP_TEST_V_TOTAL_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params1(link, DP_TEST_H_START_HI, 2,
+			&link->dp_link.test_video.test_h_start);
+	if (ret) {
+		DRM_ERROR("failed to parse test_h_start(DP_TEST_H_START_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params1(link, DP_TEST_V_START_HI, 2,
+			&link->dp_link.test_video.test_v_start);
+	if (ret) {
+		DRM_ERROR("failed to parse test_v_start(DP_TEST_V_START_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params2(link, DP_TEST_HSYNC_HI, 2,
+			&link->dp_link.test_video.test_hsync_pol,
+			&link->dp_link.test_video.test_hsync_width);
+	if (ret) {
+		DRM_ERROR("failed to parse (DP_TEST_HSYNC_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params2(link, DP_TEST_VSYNC_HI, 2,
+			&link->dp_link.test_video.test_vsync_pol,
+			&link->dp_link.test_video.test_vsync_width);
+	if (ret) {
+		DRM_ERROR("failed to parse (DP_TEST_VSYNC_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params1(link, DP_TEST_H_WIDTH_HI, 2,
+			&link->dp_link.test_video.test_h_width);
+	if (ret) {
+		DRM_ERROR("failed to parse test_h_width(DP_TEST_H_WIDTH_HI)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params1(link, DP_TEST_V_HEIGHT_HI, 2,
+			&link->dp_link.test_video.test_v_height);
+	if (ret) {
+		DRM_ERROR("failed to parse test_v_height\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params3(link, DP_TEST_MISC1,
+		&link->dp_link.test_video.test_rr_d);
+	link->dp_link.test_video.test_rr_d &= DP_TEST_REFRESH_DENOMINATOR;
+	if (ret) {
+		DRM_ERROR("failed to parse test_rr_d (DP_TEST_MISC1)\n");
+		return ret;
+	}
+
+	ret = dp_link_parse_timing_params3(link, DP_TEST_REFRESH_RATE_NUMERATOR,
+		&link->dp_link.test_video.test_rr_n);
+	if (ret) {
+		DRM_ERROR("failed to parse test_rr_n\n");
+		return ret;
+	}
+
+	DRM_DEBUG_DP("link video pattern = 0x%x\n"
+		"link dynamic range = 0x%x\n"
+		"link bit depth = 0x%x\n"
+		"TEST_H_TOTAL = %d, TEST_V_TOTAL = %d\n"
+		"TEST_H_START = %d, TEST_V_START = %d\n"
+		"TEST_HSYNC_POL = %d\n"
+		"TEST_HSYNC_WIDTH = %d\n"
+		"TEST_VSYNC_POL = %d\n"
+		"TEST_VSYNC_WIDTH = %d\n"
+		"TEST_H_WIDTH = %d\n"
+		"TEST_V_HEIGHT = %d\n"
+		"TEST_REFRESH_DENOMINATOR = %d\n"
+		 "TEST_REFRESH_NUMERATOR = %d\n",
+		link->dp_link.test_video.test_video_pattern,
+		link->dp_link.test_video.test_dyn_range,
+		link->dp_link.test_video.test_bit_depth,
+		link->dp_link.test_video.test_h_total,
+		link->dp_link.test_video.test_v_total,
+		link->dp_link.test_video.test_h_start,
+		link->dp_link.test_video.test_v_start,
+		link->dp_link.test_video.test_hsync_pol,
+		link->dp_link.test_video.test_hsync_width,
+		link->dp_link.test_video.test_vsync_pol,
+		link->dp_link.test_video.test_vsync_width,
+		link->dp_link.test_video.test_h_width,
+		link->dp_link.test_video.test_v_height,
+		link->dp_link.test_video.test_rr_d,
+		link->dp_link.test_video.test_rr_n);
+
+	return ret;
+}
+
+/**
+ * dp_link_parse_link_training_params() - parses link training parameters from
+ * DPCD
+ * @link: Display Port Driver data
+ *
+ * Returns 0 if it successfully parses the link rate (Byte 0x219) and lane
+ * count (Byte 0x220), and if these values parse are valid.
+ */
+static int dp_link_parse_link_training_params(struct dp_link_private *link)
+{
+	u8 bp;
+	ssize_t rlen;
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_LINK_RATE,	&bp);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read link rate. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	if (!is_link_rate_valid(bp)) {
+		DRM_ERROR("invalid link rate = 0x%x\n", bp);
+		return -EINVAL;
+	}
+
+	link->request.test_link_rate = bp;
+	DRM_DEBUG_DP("link rate = 0x%x\n", link->request.test_link_rate);
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_LANE_COUNT, &bp);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read lane count. rlen=%zd\n", rlen);
+		return rlen;
+	}
+	bp &= DP_MAX_LANE_COUNT_MASK;
+
+	if (!is_lane_count_valid(bp)) {
+		DRM_ERROR("invalid lane count = 0x%x\n", bp);
+		return -EINVAL;
+	}
+
+	link->request.test_lane_count = bp;
+	DRM_DEBUG_DP("lane count = 0x%x\n", link->request.test_lane_count);
+	return 0;
+}
+
+/**
+ * dp_parse_phy_test_params() - parses the phy link parameters
+ * @link: Display Port Driver data
+ *
+ * Parses the DPCD (Byte 0x248) for the DP PHY link pattern that is being
+ * requested.
+ */
+static int dp_link_parse_phy_test_params(struct dp_link_private *link)
+{
+	u8 data;
+	ssize_t rlen;
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_PHY_TEST_PATTERN,
+					&data);
+	if (rlen < 0) {
+		DRM_ERROR("failed to read phy link pattern. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	link->dp_link.phy_params.phy_test_pattern_sel = data & 0x07;
+
+	DRM_DEBUG_DP("phy_test_pattern_sel = 0x%x\n", data);
+
+	switch (data) {
+	case DP_PHY_TEST_PATTERN_SEL_MASK:
+	case DP_PHY_TEST_PATTERN_NONE:
+	case DP_PHY_TEST_PATTERN_D10_2:
+	case DP_PHY_TEST_PATTERN_ERROR_COUNT:
+	case DP_PHY_TEST_PATTERN_PRBS7:
+	case DP_PHY_TEST_PATTERN_80BIT_CUSTOM:
+	case DP_PHY_TEST_PATTERN_CP2520:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+/**
+ * dp_link_is_video_audio_test_requested() - checks for audio/video link request
+ * @link: link requested by the sink
+ *
+ * Returns true if the requested link is a permitted audio/video link.
+ */
+static bool dp_link_is_video_audio_test_requested(u32 link)
+{
+	u8 video_audio_test = (DP_TEST_LINK_VIDEO_PATTERN |
+				DP_TEST_LINK_AUDIO_PATTERN |
+				DP_TEST_LINK_AUDIO_DISABLED_VIDEO);
+
+	return ((link & video_audio_test) &&
+		!(link & ~video_audio_test));
+}
+
+/**
+ * dp_link_parse_request() - parses link request parameters from sink
+ * @link: Display Port Driver data
+ *
+ * Parses the DPCD to check if an automated link is requested (Byte 0x201),
+ * and what type of link automation is being requested (Byte 0x218).
+ */
+static int dp_link_parse_request(struct dp_link_private *link)
+{
+	int ret = 0;
+	u8 data;
+	ssize_t rlen;
+
+	/**
+	 * Read the device service IRQ vector (Byte 0x201) to determine
+	 * whether an automated link has been requested by the sink.
+	 */
+	rlen = drm_dp_dpcd_readb(link->aux,
+				DP_DEVICE_SERVICE_IRQ_VECTOR, &data);
+	if (rlen < 0) {
+		DRM_ERROR("aux read failed. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	DRM_DEBUG_DP("device service irq vector = 0x%x\n", data);
+
+	if (!(data & DP_AUTOMATED_TEST_REQUEST)) {
+		DRM_DEBUG_DP("no test requested\n");
+		return 0;
+	}
+
+	/**
+	 * Read the link request byte (Byte 0x218) to determine what type
+	 * of automated link has been requested by the sink.
+	 */
+	rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_REQUEST, &data);
+	if (rlen < 0) {
+		DRM_ERROR("aux read failed. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	if (!data || (data == DP_TEST_LINK_FAUX_PATTERN)) {
+		DRM_DEBUG_DP("link 0x%x not supported\n", data);
+		goto end;
+	}
+
+	DRM_DEBUG_DP("Test:(0x%x) requested\n", data);
+	link->request.test_requested = data;
+	if (link->request.test_requested == DP_TEST_LINK_PHY_TEST_PATTERN) {
+		ret = dp_link_parse_phy_test_params(link);
+		if (ret)
+			goto end;
+		ret = dp_link_parse_link_training_params(link);
+		if (ret)
+			goto end;
+	}
+
+	if (link->request.test_requested == DP_TEST_LINK_TRAINING) {
+		ret = dp_link_parse_link_training_params(link);
+		if (ret)
+			goto end;
+	}
+
+	if (dp_link_is_video_audio_test_requested(
+			link->request.test_requested)) {
+		ret = dp_link_parse_video_pattern_params(link);
+		if (ret)
+			goto end;
+
+		ret = dp_link_parse_audio_pattern_params(link);
+	}
+end:
+	/*
+	 * Send a DP_TEST_ACK if all link parameters are valid, otherwise send
+	 * a DP_TEST_NAK.
+	 */
+	if (ret) {
+		link->dp_link.test_response = DP_TEST_NAK;
+	} else {
+		if (link->request.test_requested != DP_TEST_LINK_EDID_READ)
+			link->dp_link.test_response = DP_TEST_ACK;
+		else
+			link->dp_link.test_response =
+				DP_TEST_EDID_CHECKSUM_WRITE;
+	}
+
+	return ret;
+}
+
+/**
+ * dp_link_parse_sink_count() - parses the sink count
+ * @dp_link: pointer to link module data
+ *
+ * Parses the DPCD to check if there is an update to the sink count
+ * (Byte 0x200), and whether all the sink devices connected have Content
+ * Protection enabled.
+ */
+static int dp_link_parse_sink_count(struct dp_link *dp_link)
+{
+	ssize_t rlen;
+	bool cp_ready;
+
+	struct dp_link_private *link = container_of(dp_link,
+			struct dp_link_private, dp_link);
+
+	rlen = drm_dp_dpcd_readb(link->aux, DP_SINK_COUNT,
+				 &link->dp_link.sink_count);
+	if (rlen < 0) {
+		DRM_ERROR("sink count read failed. rlen=%zd\n", rlen);
+		return rlen;
+	}
+
+	cp_ready = link->dp_link.sink_count & DP_SINK_CP_READY;
+
+	link->dp_link.sink_count =
+		DP_GET_SINK_COUNT(link->dp_link.sink_count);
+
+	DRM_DEBUG_DP("sink_count = 0x%x, cp_ready = 0x%x\n",
+		link->dp_link.sink_count, cp_ready);
+	return 0;
+}
+
+static void dp_link_parse_sink_status_field(struct dp_link_private *link)
+{
+	int len = 0;
+
+	link->prev_sink_count = link->dp_link.sink_count;
+	dp_link_parse_sink_count(&link->dp_link);
+
+	len = drm_dp_dpcd_read_link_status(link->aux,
+		link->link_status);
+	if (len < DP_LINK_STATUS_SIZE)
+		DRM_ERROR("DP link status read failed\n");
+	dp_link_parse_request(link);
+}
+
+/**
+ * dp_link_process_link_training_request() - processes new training requests
+ * @link: Display Port link data
+ *
+ * This function will handle new link training requests that are initiated by
+ * the sink. In particular, it will update the requested lane count and link
+ * rate, and then trigger the link retraining procedure.
+ *
+ * The function will return 0 if a link training request has been processed,
+ * otherwise it will return -EINVAL.
+ */
+static int dp_link_process_link_training_request(struct dp_link_private *link)
+{
+	if (link->request.test_requested != DP_TEST_LINK_TRAINING)
+		return -EINVAL;
+
+	DRM_DEBUG_DP("Test:0x%x link rate = 0x%x, lane count = 0x%x\n",
+			DP_TEST_LINK_TRAINING,
+			link->request.test_link_rate,
+			link->request.test_lane_count);
+
+	link->dp_link.link_params.num_lanes = link->request.test_lane_count;
+	link->dp_link.link_params.rate = link->request.test_link_rate;
+
+	return 0;
+}
+
+bool dp_link_send_test_response(struct dp_link *dp_link)
+{
+	struct dp_link_private *link = NULL;
+	int ret = 0;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid input\n");
+		return false;
+	}
+
+	link = container_of(dp_link, struct dp_link_private, dp_link);
+
+	ret = drm_dp_dpcd_writeb(link->aux, DP_TEST_RESPONSE,
+			dp_link->test_response);
+
+	return ret == 1;
+}
+
+int dp_link_psm_config(struct dp_link *dp_link,
+			      struct dp_link_info *link_info, bool enable)
+{
+	struct dp_link_private *link = NULL;
+	int ret = 0;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid params\n");
+		return -EINVAL;
+	}
+
+	link = container_of(dp_link, struct dp_link_private, dp_link);
+
+	mutex_lock(&link->psm_mutex);
+	if (enable)
+		ret = dp_aux_link_power_down(link->aux, link_info);
+	else
+		ret = dp_aux_link_power_up(link->aux, link_info);
+
+	if (ret)
+		DRM_ERROR("Failed to %s low power mode\n", enable ?
+							"enter" : "exit");
+	else
+		dp_link->psm_enabled = enable;
+
+	mutex_unlock(&link->psm_mutex);
+	return ret;
+}
+
+bool dp_link_send_edid_checksum(struct dp_link *dp_link, u8 checksum)
+{
+	struct dp_link_private *link = NULL;
+	int ret = 0;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid input\n");
+		return false;
+	}
+
+	link = container_of(dp_link, struct dp_link_private, dp_link);
+
+	ret = drm_dp_dpcd_writeb(link->aux, DP_TEST_EDID_CHECKSUM,
+						checksum);
+	return ret == 1;
+}
+
+static int dp_link_parse_vx_px(struct dp_link_private *link)
+{
+	int ret = 0;
+
+	DRM_DEBUG_DP("vx: 0=%d, 1=%d, 2=%d, 3=%d\n",
+		drm_dp_get_adjust_request_voltage(link->link_status, 0),
+		drm_dp_get_adjust_request_voltage(link->link_status, 1),
+		drm_dp_get_adjust_request_voltage(link->link_status, 2),
+		drm_dp_get_adjust_request_voltage(link->link_status, 3));
+
+	DRM_DEBUG_DP("px: 0=%d, 1=%d, 2=%d, 3=%d\n",
+		drm_dp_get_adjust_request_pre_emphasis(link->link_status, 0),
+		drm_dp_get_adjust_request_pre_emphasis(link->link_status, 1),
+		drm_dp_get_adjust_request_pre_emphasis(link->link_status, 2),
+		drm_dp_get_adjust_request_pre_emphasis(link->link_status, 3));
+
+	/**
+	 * Update the voltage and pre-emphasis levels as per DPCD request
+	 * vector.
+	 */
+	DRM_DEBUG_DP("Current: v_level = 0x%x, p_level = 0x%x\n",
+			link->dp_link.phy_params.v_level,
+			link->dp_link.phy_params.p_level);
+	link->dp_link.phy_params.v_level =
+		drm_dp_get_adjust_request_voltage(link->link_status, 0);
+	link->dp_link.phy_params.p_level =
+		drm_dp_get_adjust_request_pre_emphasis(link->link_status, 0);
+	DRM_DEBUG_DP("Requested: v_level = 0x%x, p_level = 0x%x\n",
+			link->dp_link.phy_params.v_level,
+			link->dp_link.phy_params.p_level);
+
+	return ret;
+}
+
+/**
+ * dp_link_process_phy_test_pattern_request() - process new phy link requests
+ * @link: Display Port Driver data
+ *
+ * This function will handle new phy link pattern requests that are initiated
+ * by the sink. The function will return 0 if a phy link pattern has been
+ * processed, otherwise it will return -EINVAL.
+ */
+static int dp_link_process_phy_test_pattern_request(
+		struct dp_link_private *link)
+{
+	int ret = 0;
+
+	if (!(link->request.test_requested & DP_TEST_LINK_PHY_TEST_PATTERN)) {
+		DRM_DEBUG_DP("no phy test\n");
+		return -EINVAL;
+	}
+
+	if (!is_link_rate_valid(link->request.test_link_rate) ||
+		!is_lane_count_valid(link->request.test_lane_count)) {
+		DRM_ERROR("Invalid: link rate = 0x%x,lane count = 0x%x\n",
+				link->request.test_link_rate,
+				link->request.test_lane_count);
+		return -EINVAL;
+	}
+
+	DRM_DEBUG_DP("Current: rate = 0x%x, lane count = 0x%x\n",
+			link->dp_link.link_params.rate,
+			link->dp_link.link_params.num_lanes);
+
+	DRM_DEBUG_DP("Requested: rate = 0x%x, lane count = 0x%x\n",
+			link->request.test_link_rate,
+			link->request.test_lane_count);
+
+	link->dp_link.link_params.num_lanes = link->request.test_lane_count;
+	link->dp_link.link_params.rate = link->request.test_link_rate;
+
+	ret = dp_link_parse_vx_px(link);
+
+	if (ret)
+		DRM_ERROR("parse_vx_px failed. ret=%d\n", ret);
+
+	return ret;
+}
+
+static u8 get_link_status(const u8 link_status[DP_LINK_STATUS_SIZE], int r)
+{
+	return link_status[r - DP_LANE0_1_STATUS];
+}
+
+/**
+ * dp_link_process_link_status_update() - processes link status updates
+ * @link: Display Port link module data
+ *
+ * This function will check for changes in the link status, e.g. clock
+ * recovery done on all lanes, and trigger link training if there is a
+ * failure/error on the link.
+ *
+ * The function will return 0 if the a link status update has been processed,
+ * otherwise it will return -EINVAL.
+ */
+static int dp_link_process_link_status_update(struct dp_link_private *link)
+{
+	if (!(get_link_status(link->link_status,
+				DP_LANE_ALIGN_STATUS_UPDATED) &
+				DP_LINK_STATUS_UPDATED) ||
+			(drm_dp_clock_recovery_ok(link->link_status,
+					link->dp_link.link_params.num_lanes) &&
+			drm_dp_channel_eq_ok(link->link_status,
+					link->dp_link.link_params.num_lanes)))
+		return -EINVAL;
+
+	DRM_DEBUG_DP("channel_eq_done = %d, clock_recovery_done = %d\n",
+			drm_dp_clock_recovery_ok(link->link_status,
+			link->dp_link.link_params.num_lanes),
+			drm_dp_clock_recovery_ok(link->link_status,
+			link->dp_link.link_params.num_lanes));
+
+	return 0;
+}
+
+/**
+ * dp_link_process_downstream_port_status_change() - process port status changes
+ * @link: Display Port Driver data
+ *
+ * This function will handle downstream port updates that are initiated by
+ * the sink. If the downstream port status has changed, the EDID is read via
+ * AUX.
+ *
+ * The function will return 0 if a downstream port update has been
+ * processed, otherwise it will return -EINVAL.
+ */
+static int dp_link_process_ds_port_status_change(struct dp_link_private *link)
+{
+	if (get_link_status(link->link_status, DP_LANE_ALIGN_STATUS_UPDATED) &
+					DP_DOWNSTREAM_PORT_STATUS_CHANGED)
+		goto reset;
+
+	if (link->prev_sink_count == link->dp_link.sink_count)
+		return -EINVAL;
+
+reset:
+	/* reset prev_sink_count */
+	link->prev_sink_count = link->dp_link.sink_count;
+
+	return 0;
+}
+
+static bool dp_link_is_video_pattern_requested(struct dp_link_private *link)
+{
+	return (link->request.test_requested & DP_TEST_LINK_VIDEO_PATTERN)
+		&& !(link->request.test_requested &
+		DP_TEST_LINK_AUDIO_DISABLED_VIDEO);
+}
+
+static bool dp_link_is_audio_pattern_requested(struct dp_link_private *link)
+{
+	return (link->request.test_requested & DP_TEST_LINK_AUDIO_PATTERN);
+}
+
+static void dp_link_reset_data(struct dp_link_private *link)
+{
+	link->request = (const struct dp_link_request){ 0 };
+	link->dp_link.test_video = (const struct dp_link_test_video){ 0 };
+	link->dp_link.test_video.test_bit_depth = DP_TEST_BIT_DEPTH_UNKNOWN;
+	link->dp_link.test_audio = (const struct dp_link_test_audio){ 0 };
+	link->dp_link.phy_params.phy_test_pattern_sel = 0;
+	link->dp_link.sink_request = 0;
+	link->dp_link.test_response = 0;
+}
+
+/**
+ * dp_link_process_request() - handle HPD IRQ transition to HIGH
+ * @dp_link: pointer to link module data
+ *
+ * This function will handle the HPD IRQ state transitions from LOW to HIGH
+ * (including cases when there are back to back HPD IRQ HIGH) indicating
+ * the start of a new link training request or sink status update.
+ */
+int dp_link_process_request(struct dp_link *dp_link)
+{
+	int ret = 0;
+	struct dp_link_private *link;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	link = container_of(dp_link, struct dp_link_private, dp_link);
+
+	dp_link_reset_data(link);
+
+	dp_link_parse_sink_status_field(link);
+
+	if (link->request.test_requested == DP_TEST_LINK_EDID_READ) {
+		dp_link->sink_request |= DP_TEST_LINK_EDID_READ;
+		return ret;
+	}
+
+	ret = dp_link_process_ds_port_status_change(link);
+	if (!ret) {
+		dp_link->sink_request |= DS_PORT_STATUS_CHANGED;
+		return ret;
+	}
+
+	ret = dp_link_process_link_training_request(link);
+	if (!ret) {
+		dp_link->sink_request |= DP_TEST_LINK_TRAINING;
+		return ret;
+	}
+
+	ret = dp_link_process_phy_test_pattern_request(link);
+	if (!ret) {
+		dp_link->sink_request |= DP_TEST_LINK_PHY_TEST_PATTERN;
+		return ret;
+	}
+
+	ret = dp_link_process_link_status_update(link);
+	if (!ret) {
+		dp_link->sink_request |= DP_LINK_STATUS_UPDATED;
+		return ret;
+	}
+
+	if (dp_link_is_video_pattern_requested(link)) {
+		ret = 0;
+		dp_link->sink_request |= DP_TEST_LINK_VIDEO_PATTERN;
+	}
+
+	if (dp_link_is_audio_pattern_requested(link)) {
+		dp_link->sink_request |= DP_TEST_LINK_AUDIO_PATTERN;
+		return -EINVAL;
+	}
+
+	return ret;
+}
+
+int dp_link_get_colorimetry_config(struct dp_link *dp_link)
+{
+	u32 cc;
+	struct dp_link_private *link;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	link = container_of(dp_link, struct dp_link_private, dp_link);
+
+	/*
+	 * Unless a video pattern CTS test is ongoing, use RGB_VESA
+	 * Only RGB_VESA and RGB_CEA supported for now
+	 */
+	if (dp_link_is_video_pattern_requested(link))
+		cc = link->dp_link.test_video.test_dyn_range;
+	else
+		cc = DP_TEST_DYNAMIC_RANGE_VESA;
+
+	return cc;
+}
+
+int dp_link_adjust_levels(struct dp_link *dp_link, u8 *link_status)
+{
+	int i;
+	int v_max = 0, p_max = 0;
+
+	if (!dp_link) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	/* use the max level across lanes */
+	for (i = 0; i < dp_link->link_params.num_lanes; i++) {
+		u8 data_v = drm_dp_get_adjust_request_voltage(link_status, i);
+		u8 data_p = drm_dp_get_adjust_request_pre_emphasis(link_status,
+									 i);
+		DRM_DEBUG_DP("lane=%d req_vol_swing=%d req_pre_emphasis=%d\n",
+				i, data_v, data_p);
+		if (v_max < data_v)
+			v_max = data_v;
+		if (p_max < data_p)
+			p_max = data_p;
+	}
+
+	dp_link->phy_params.v_level = v_max >> DP_TRAIN_VOLTAGE_SWING_SHIFT;
+	dp_link->phy_params.p_level = p_max >> DP_TRAIN_PRE_EMPHASIS_SHIFT;
+
+	/**
+	 * Adjust the voltage swing and pre-emphasis level combination to within
+	 * the allowable range.
+	 */
+	if (dp_link->phy_params.v_level > DP_TRAIN_VOLTAGE_SWING_MAX) {
+		DRM_DEBUG_DP("Requested vSwingLevel=%d, change to %d\n",
+			dp_link->phy_params.v_level,
+			DP_TRAIN_VOLTAGE_SWING_MAX);
+		dp_link->phy_params.v_level = DP_TRAIN_VOLTAGE_SWING_MAX;
+	}
+
+	if (dp_link->phy_params.p_level > DP_TRAIN_PRE_EMPHASIS_MAX) {
+		DRM_DEBUG_DP("Requested preEmphasisLevel=%d, change to %d\n",
+			dp_link->phy_params.p_level,
+			DP_TRAIN_PRE_EMPHASIS_MAX);
+		dp_link->phy_params.p_level = DP_TRAIN_PRE_EMPHASIS_MAX;
+	}
+
+	if ((dp_link->phy_params.p_level > DP_TRAIN_PRE_EMPHASIS_LVL_1)
+		&& (dp_link->phy_params.v_level ==
+			DP_TRAIN_VOLTAGE_SWING_LVL_2)) {
+		DRM_DEBUG_DP("Requested preEmphasisLevel=%d, change to %d\n",
+			dp_link->phy_params.p_level,
+			DP_TRAIN_PRE_EMPHASIS_LVL_1);
+		dp_link->phy_params.p_level = DP_TRAIN_PRE_EMPHASIS_LVL_1;
+	}
+
+	DRM_DEBUG_DP("adjusted: v_level=%d, p_level=%d\n",
+		dp_link->phy_params.v_level, dp_link->phy_params.p_level);
+
+	return 0;
+}
+
+u32 dp_link_get_test_bits_depth(struct dp_link *dp_link, u32 bpp)
+{
+	u32 tbd;
+
+	/*
+	 * Few simplistic rules and assumptions made here:
+	 *    1. Test bit depth is bit depth per color component
+	 *    2. Assume 3 color components
+	 */
+	switch (bpp) {
+	case 18:
+		tbd = DP_TEST_BIT_DEPTH_6;
+		break;
+	case 24:
+		tbd = DP_TEST_BIT_DEPTH_8;
+		break;
+	case 30:
+		tbd = DP_TEST_BIT_DEPTH_10;
+		break;
+	default:
+		tbd = DP_TEST_BIT_DEPTH_UNKNOWN;
+		break;
+	}
+
+	if (tbd != DP_TEST_BIT_DEPTH_UNKNOWN)
+		tbd = (tbd >> DP_TEST_BIT_DEPTH_SHIFT);
+
+	return tbd;
+}
+
+struct dp_link *dp_link_get(struct device *dev, struct drm_dp_aux *aux)
+{
+	struct dp_link_private *link;
+	struct dp_link *dp_link;
+
+	if (!dev || !aux) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	link = devm_kzalloc(dev, sizeof(*link), GFP_KERNEL);
+	if (!link)
+		return ERR_PTR(-ENOMEM);
+
+	link->dev   = dev;
+	link->aux   = aux;
+
+	mutex_init(&link->psm_mutex);
+	dp_link = &link->dp_link;
+
+	return dp_link;
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_link.h b/drivers/gpu/drm/msm/dp/dp_link.h
new file mode 100644
index 000000000000..49811b6221e5
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_link.h
@@ -0,0 +1,155 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_LINK_H_
+#define _DP_LINK_H_
+
+#include "dp_aux.h"
+
+#define DS_PORT_STATUS_CHANGED 0x200
+#define DP_TEST_BIT_DEPTH_UNKNOWN 0xFFFFFFFF
+#define DP_LINK_CAP_ENHANCED_FRAMING (1 << 0)
+
+struct dp_link_info {
+	unsigned char revision;
+	unsigned int rate;
+	unsigned int num_lanes;
+	unsigned long capabilities;
+};
+
+enum dp_link_voltage_level {
+	DP_TRAIN_VOLTAGE_SWING_LVL_0	= 0,
+	DP_TRAIN_VOLTAGE_SWING_LVL_1	= 1,
+	DP_TRAIN_VOLTAGE_SWING_LVL_2	= 2,
+	DP_TRAIN_VOLTAGE_SWING_MAX	= DP_TRAIN_VOLTAGE_SWING_LVL_2,
+};
+
+enum dp_link_preemaphasis_level {
+	DP_TRAIN_PRE_EMPHASIS_LVL_0	= 0,
+	DP_TRAIN_PRE_EMPHASIS_LVL_1	= 1,
+	DP_TRAIN_PRE_EMPHASIS_LVL_2	= 2,
+	DP_TRAIN_PRE_EMPHASIS_MAX	= DP_TRAIN_PRE_EMPHASIS_LVL_2,
+};
+
+struct dp_link_test_video {
+	u32 test_video_pattern;
+	u32 test_bit_depth;
+	u32 test_dyn_range;
+	u32 test_h_total;
+	u32 test_v_total;
+	u32 test_h_start;
+	u32 test_v_start;
+	u32 test_hsync_pol;
+	u32 test_hsync_width;
+	u32 test_vsync_pol;
+	u32 test_vsync_width;
+	u32 test_h_width;
+	u32 test_v_height;
+	u32 test_rr_d;
+	u32 test_rr_n;
+};
+
+struct dp_link_test_audio {
+	u32 test_audio_sampling_rate;
+	u32 test_audio_channel_count;
+	u32 test_audio_pattern_type;
+	u32 test_audio_period_ch_1;
+	u32 test_audio_period_ch_2;
+	u32 test_audio_period_ch_3;
+	u32 test_audio_period_ch_4;
+	u32 test_audio_period_ch_5;
+	u32 test_audio_period_ch_6;
+	u32 test_audio_period_ch_7;
+	u32 test_audio_period_ch_8;
+};
+
+struct dp_link_phy_params {
+	u32 phy_test_pattern_sel;
+	u8 v_level;
+	u8 p_level;
+};
+
+struct dp_link {
+	u32 sink_request;
+	u32 test_response;
+	bool psm_enabled;
+
+	u8 sink_count;
+	struct dp_link_test_video test_video;
+	struct dp_link_test_audio test_audio;
+	struct dp_link_phy_params phy_params;
+	struct dp_link_info link_params;
+};
+
+/**
+ * mdss_dp_test_bit_depth_to_bpp() - convert test bit depth to bpp
+ * @tbd: test bit depth
+ *
+ * Returns the bits per pixel (bpp) to be used corresponding to the
+ * git bit depth value. This function assumes that bit depth has
+ * already been validated.
+ */
+static inline u32 dp_link_bit_depth_to_bpp(u32 tbd)
+{
+	/*
+	 * Few simplistic rules and assumptions made here:
+	 *    1. Bit depth is per color component
+	 *    2. If bit depth is unknown return 0
+	 *    3. Assume 3 color components
+	 */
+	switch (tbd) {
+	case DP_TEST_BIT_DEPTH_6:
+		return 18;
+	case DP_TEST_BIT_DEPTH_8:
+		return 24;
+	case DP_TEST_BIT_DEPTH_10:
+		return 30;
+	case DP_TEST_BIT_DEPTH_UNKNOWN:
+	default:
+		return 0;
+	}
+}
+
+/**
+ * dp_test_bit_depth_to_bpc() - convert test bit depth to bpc
+ * @tbd: test bit depth
+ *
+ * Returns the bits per comp (bpc) to be used corresponding to the
+ * bit depth value. This function assumes that bit depth has
+ * already been validated.
+ */
+static inline u32 dp_link_bit_depth_to_bpc(u32 tbd)
+{
+	switch (tbd) {
+	case DP_TEST_BIT_DEPTH_6:
+		return 6;
+	case DP_TEST_BIT_DEPTH_8:
+		return 8;
+	case DP_TEST_BIT_DEPTH_10:
+		return 10;
+	case DP_TEST_BIT_DEPTH_UNKNOWN:
+	default:
+		return 0;
+	}
+}
+
+u32 dp_link_get_test_bits_depth(struct dp_link *dp_link, u32 bpp);
+int dp_link_process_request(struct dp_link *dp_link);
+int dp_link_get_colorimetry_config(struct dp_link *dp_link);
+int dp_link_adjust_levels(struct dp_link *dp_link, u8 *link_status);
+bool dp_link_send_test_response(struct dp_link *dp_link);
+int dp_link_psm_config(struct dp_link *dp_link,
+		struct dp_link_info *link_info, bool enable);
+bool dp_link_send_edid_checksum(struct dp_link *dp_link, u8 checksum);
+
+/**
+ * dp_link_get() - get the functionalities of dp test module
+ *
+ *
+ * return: a pointer to dp_link struct
+ */
+struct dp_link *dp_link_get(struct device *dev, struct drm_dp_aux *aux);
+
+#endif /* _DP_LINK_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_panel.c b/drivers/gpu/drm/msm/dp/dp_panel.c
new file mode 100644
index 000000000000..18cec4fc5e0b
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_panel.c
@@ -0,0 +1,463 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#include "dp_panel.h"
+
+#include <drm/drm_connector.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_print.h>
+
+struct dp_panel_private {
+	struct device *dev;
+	struct dp_panel dp_panel;
+	struct drm_dp_aux *aux;
+	struct dp_link *link;
+	struct dp_catalog *catalog;
+	bool panel_on;
+	bool aux_cfg_update_done;
+};
+
+static int dp_panel_read_dpcd(struct dp_panel *dp_panel)
+{
+	int rc = 0;
+	size_t len;
+	ssize_t rlen;
+	struct dp_panel_private *panel;
+	struct dp_link_info *link_info;
+	u8 *dpcd, major = 0, minor = 0, temp;
+	u32 offset = DP_DPCD_REV;
+
+	dpcd = dp_panel->dpcd;
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+	link_info = &dp_panel->link_info;
+
+	rlen = drm_dp_dpcd_read(panel->aux, offset,
+			dpcd, (DP_RECEIVER_CAP_SIZE + 1));
+	if (rlen < (DP_RECEIVER_CAP_SIZE + 1)) {
+		DRM_ERROR("dpcd read failed, rlen=%zd\n", rlen);
+		if (rlen == -ETIMEDOUT)
+			rc = rlen;
+		else
+			rc = -EINVAL;
+
+		goto end;
+	}
+
+	temp = dpcd[DP_TRAINING_AUX_RD_INTERVAL];
+
+	/* check for EXTENDED_RECEIVER_CAPABILITY_FIELD_PRESENT */
+	if (temp & BIT(7)) {
+		DRM_DEBUG_DP("using EXTENDED_RECEIVER_CAPABILITY_FIELD\n");
+		offset = DPRX_EXTENDED_DPCD_FIELD;
+	}
+
+	rlen = drm_dp_dpcd_read(panel->aux, offset,
+		dpcd, (DP_RECEIVER_CAP_SIZE + 1));
+	if (rlen < (DP_RECEIVER_CAP_SIZE + 1)) {
+		DRM_ERROR("dpcd read failed, rlen=%zd\n", rlen);
+		if (rlen == -ETIMEDOUT)
+			rc = rlen;
+		else
+			rc = -EINVAL;
+
+		goto end;
+	}
+
+	link_info->revision = dpcd[DP_DPCD_REV];
+	major = (link_info->revision >> 4) & 0x0f;
+	minor = link_info->revision & 0x0f;
+
+	link_info->rate = drm_dp_bw_code_to_link_rate(dpcd[DP_MAX_LINK_RATE]);
+	link_info->num_lanes = dpcd[DP_MAX_LANE_COUNT] & DP_MAX_LANE_COUNT_MASK;
+
+	if (link_info->num_lanes > dp_panel->max_dp_lanes)
+		link_info->num_lanes = dp_panel->max_dp_lanes;
+
+	/* Limit support upto HBR2 until HBR3 support is added */
+	if (link_info->rate >= (drm_dp_bw_code_to_link_rate(DP_LINK_BW_5_4)))
+		link_info->rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_5_4);
+
+	DRM_DEBUG_DP("version: %d.%d\n", major, minor);
+	DRM_DEBUG_DP("link_rate=%d\n", link_info->rate);
+	DRM_DEBUG_DP("lane_count=%d\n", link_info->num_lanes);
+
+	if (drm_dp_enhanced_frame_cap(dpcd))
+		link_info->capabilities |= DP_LINK_CAP_ENHANCED_FRAMING;
+
+	dp_panel->dfp_present = dpcd[DP_DOWNSTREAMPORT_PRESENT];
+	dp_panel->dfp_present &= DP_DWN_STRM_PORT_PRESENT;
+
+	if (dp_panel->dfp_present && (dpcd[DP_DPCD_REV] > 0x10)) {
+		dp_panel->ds_port_cnt = dpcd[DP_DOWN_STREAM_PORT_COUNT];
+		dp_panel->ds_port_cnt &= DP_PORT_COUNT_MASK;
+		len = DP_DOWNSTREAM_PORTS * DP_DOWNSTREAM_CAP_SIZE;
+
+		rlen = drm_dp_dpcd_read(panel->aux,
+			DP_DOWNSTREAM_PORT_0, dp_panel->ds_cap_info, len);
+		if (rlen < len) {
+			DRM_ERROR("ds port status failed, rlen=%zd\n", rlen);
+			rc = -EINVAL;
+			goto end;
+		}
+	}
+
+end:
+	return rc;
+}
+
+static u32 dp_panel_get_supported_bpp(struct dp_panel *dp_panel,
+		u32 mode_edid_bpp, u32 mode_pclk_khz)
+{
+	struct dp_link_info *link_info;
+	const u32 max_supported_bpp = 30, min_supported_bpp = 18;
+	u32 bpp = 0, data_rate_khz = 0;
+
+	bpp = min_t(u32, mode_edid_bpp, max_supported_bpp);
+
+	link_info = &dp_panel->link_info;
+	data_rate_khz = link_info->num_lanes * link_info->rate * 8;
+
+	while (bpp > min_supported_bpp) {
+		if (mode_pclk_khz * bpp <= data_rate_khz)
+			break;
+		bpp -= 6;
+	}
+
+	return bpp;
+}
+
+static int dp_panel_update_modes(struct drm_connector *connector,
+	struct edid *edid)
+{
+	int rc = 0;
+
+	if (edid) {
+		rc = drm_connector_update_edid_property(connector, edid);
+		if (rc) {
+			DRM_ERROR("failed to update edid property %d\n", rc);
+			return rc;
+		}
+		rc = drm_add_edid_modes(connector, edid);
+		DRM_DEBUG_DP("%s -", __func__);
+		return rc;
+	}
+
+	rc = drm_connector_update_edid_property(connector, NULL);
+	if (rc)
+		DRM_ERROR("failed to update edid property %d\n", rc);
+
+	return rc;
+}
+
+int dp_panel_read_sink_caps(struct dp_panel *dp_panel,
+	struct drm_connector *connector)
+{
+	int rc = 0, bw_code;
+	int rlen, count;
+	struct dp_panel_private *panel;
+
+	if (!dp_panel || !connector) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+
+	rc = dp_panel_read_dpcd(dp_panel);
+	bw_code = drm_dp_link_rate_to_bw_code(dp_panel->link_info.rate);
+	if (rc || !is_link_rate_valid(bw_code) ||
+			!is_lane_count_valid(dp_panel->link_info.num_lanes) ||
+			(bw_code > dp_panel->max_bw_code)) {
+		DRM_ERROR("read dpcd failed %d\n", rc);
+		return rc;
+	}
+
+	if (dp_panel->dfp_present) {
+		rlen = drm_dp_dpcd_read(panel->aux, DP_SINK_COUNT,
+				&count, 1);
+		if (rlen == 1) {
+			count = DP_GET_SINK_COUNT(count);
+			if (!count) {
+				DRM_ERROR("no downstream ports connected\n");
+				panel->link->sink_count = 0;
+				rc = -ENOTCONN;
+				goto end;
+			}
+		}
+	}
+
+	kfree(dp_panel->edid);
+	dp_panel->edid = NULL;
+
+	dp_panel->edid = drm_get_edid(connector,
+					      &panel->aux->ddc);
+	if (!dp_panel->edid) {
+		DRM_ERROR("panel edid read failed\n");
+
+		/* fail safe edid */
+		mutex_lock(&connector->dev->mode_config.mutex);
+		if (drm_add_modes_noedid(connector, 640, 480))
+			drm_set_preferred_mode(connector, 640, 480);
+		mutex_unlock(&connector->dev->mode_config.mutex);
+	}
+
+	if (panel->aux_cfg_update_done) {
+		DRM_DEBUG_DP("read DPCD with updated AUX config\n");
+		rc = dp_panel_read_dpcd(dp_panel);
+		bw_code = drm_dp_link_rate_to_bw_code(dp_panel->link_info.rate);
+		if (rc || !is_link_rate_valid(bw_code) ||
+			!is_lane_count_valid(dp_panel->link_info.num_lanes)
+			|| (bw_code > dp_panel->max_bw_code)) {
+			DRM_ERROR("read dpcd failed %d\n", rc);
+			return rc;
+		}
+		panel->aux_cfg_update_done = false;
+	}
+end:
+	return rc;
+}
+
+u32 dp_panel_get_mode_bpp(struct dp_panel *dp_panel,
+		u32 mode_edid_bpp, u32 mode_pclk_khz)
+{
+	struct dp_panel_private *panel;
+	u32 bpp = mode_edid_bpp;
+
+	if (!dp_panel || !mode_edid_bpp || !mode_pclk_khz) {
+		DRM_ERROR("invalid input\n");
+		return 0;
+	}
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+
+	if (dp_panel->video_test)
+		bpp = dp_link_bit_depth_to_bpp(
+				panel->link->test_video.test_bit_depth);
+	else
+		bpp = dp_panel_get_supported_bpp(dp_panel, mode_edid_bpp,
+				mode_pclk_khz);
+
+	return bpp;
+}
+
+int dp_panel_get_modes(struct dp_panel *dp_panel,
+	struct drm_connector *connector, struct dp_display_mode *mode)
+{
+	if (!dp_panel) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	if (dp_panel->edid)
+		return dp_panel_update_modes(connector, dp_panel->edid);
+
+	return 0;
+}
+
+static u8 dp_panel_get_edid_checksum(struct edid *edid)
+{
+	struct edid *last_block;
+	u8 *raw_edid;
+	bool is_edid_corrupt;
+
+	if (!edid) {
+		DRM_ERROR("invalid edid input\n");
+		return 0;
+	}
+
+	raw_edid = (u8 *)edid;
+	raw_edid += (edid->extensions * EDID_LENGTH);
+	last_block = (struct edid *)raw_edid;
+
+	/* block type extension */
+	drm_edid_block_valid(raw_edid, 1, false, &is_edid_corrupt);
+	if (!is_edid_corrupt)
+		return last_block->checksum;
+
+	DRM_ERROR("Invalid block, no checksum\n");
+	return 0;
+}
+
+void dp_panel_handle_sink_request(struct dp_panel *dp_panel)
+{
+	struct dp_panel_private *panel;
+
+	if (!dp_panel) {
+		DRM_ERROR("invalid input\n");
+		return;
+	}
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+
+	if (panel->link->sink_request & DP_TEST_LINK_EDID_READ) {
+		u8 checksum = dp_panel_get_edid_checksum(dp_panel->edid);
+
+		dp_link_send_edid_checksum(panel->link, checksum);
+		dp_link_send_test_response(panel->link);
+	}
+}
+
+void dp_panel_tpg_config(struct dp_panel *dp_panel, bool enable)
+{
+	struct dp_catalog *catalog;
+	struct dp_panel_private *panel;
+
+	if (!dp_panel) {
+		DRM_ERROR("invalid input\n");
+		return;
+	}
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+	catalog = panel->catalog;
+
+	if (!panel->panel_on) {
+		DRM_DEBUG_DP("DP panel not enabled, handle TPG on next on\n");
+		return;
+	}
+
+	if (!enable) {
+		dp_catalog_panel_tpg_disable(catalog);
+		return;
+	}
+
+	DRM_DEBUG_DP("%s: calling catalog tpg_enable\n", __func__);
+	dp_catalog_panel_tpg_enable(catalog, &panel->dp_panel.dp_mode.drm_mode);
+}
+
+void dp_panel_dump_regs(struct dp_panel *dp_panel)
+{
+	struct dp_catalog *catalog;
+	struct dp_panel_private *panel;
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+	catalog = panel->catalog;
+
+	dp_catalog_dump_regs(catalog);
+}
+
+int dp_panel_timing_cfg(struct dp_panel *dp_panel)
+{
+	int rc = 0;
+	u32 data, total_ver, total_hor;
+	struct dp_catalog *catalog;
+	struct dp_panel_private *panel;
+	struct drm_display_mode *drm_mode;
+
+	panel = container_of(dp_panel, struct dp_panel_private, dp_panel);
+	catalog = panel->catalog;
+	drm_mode = &panel->dp_panel.dp_mode.drm_mode;
+
+	DRM_DEBUG_DP("width=%d hporch= %d %d %d\n",
+		drm_mode->hdisplay, drm_mode->htotal - drm_mode->hsync_end,
+		drm_mode->hsync_start - drm_mode->hdisplay,
+		drm_mode->hsync_end - drm_mode->hsync_start);
+
+	DRM_DEBUG_DP("height=%d vporch= %d %d %d\n",
+		drm_mode->vdisplay, drm_mode->vtotal - drm_mode->vsync_end,
+		drm_mode->vsync_start - drm_mode->vdisplay,
+		drm_mode->vsync_end - drm_mode->vsync_start);
+
+	total_hor = drm_mode->htotal;
+
+	total_ver = drm_mode->vtotal;
+
+	data = total_ver;
+	data <<= 16;
+	data |= total_hor;
+
+	catalog->total = data;
+
+	data = (drm_mode->vtotal - drm_mode->vsync_start);
+	data <<= 16;
+	data |= (drm_mode->htotal - drm_mode->hsync_start);
+
+	catalog->sync_start = data;
+
+	data = drm_mode->vsync_end - drm_mode->vsync_start;
+	data <<= 16;
+	data |= (panel->dp_panel.dp_mode.v_active_low << 31);
+	data |= drm_mode->hsync_end - drm_mode->hsync_start;
+	data |= (panel->dp_panel.dp_mode.h_active_low << 15);
+
+	catalog->width_blanking = data;
+
+	data = drm_mode->vdisplay;
+	data <<= 16;
+	data |= drm_mode->hdisplay;
+
+	catalog->dp_active = data;
+
+	dp_catalog_panel_timing_cfg(catalog);
+	panel->panel_on = true;
+
+	return rc;
+}
+
+int dp_panel_init_panel_info(struct dp_panel *dp_panel)
+{
+	int rc = 0;
+	struct drm_display_mode *drm_mode;
+
+	drm_mode = &dp_panel->dp_mode.drm_mode;
+
+	/*
+	 * print resolution info as this is a result
+	 * of user initiated action of cable connection
+	 */
+	DRM_DEBUG_DP("SET NEW RESOLUTION:\n");
+	DRM_DEBUG_DP("%dx%d@%dfps\n", drm_mode->hdisplay,
+		drm_mode->vdisplay, drm_mode_vrefresh(drm_mode));
+	DRM_DEBUG_DP("h_porches(back|front|width) = (%d|%d|%d)\n",
+			drm_mode->htotal - drm_mode->hsync_end,
+			drm_mode->hsync_start - drm_mode->hdisplay,
+			drm_mode->hsync_end - drm_mode->hsync_start);
+	DRM_DEBUG_DP("v_porches(back|front|width) = (%d|%d|%d)\n",
+			drm_mode->vtotal - drm_mode->vsync_end,
+			drm_mode->vsync_start - drm_mode->vdisplay,
+			drm_mode->vsync_end - drm_mode->vsync_start);
+	DRM_DEBUG_DP("pixel clock (KHz)=(%d)\n", drm_mode->clock);
+	DRM_DEBUG_DP("bpp = %d\n", dp_panel->dp_mode.bpp);
+
+	dp_panel->dp_mode.bpp = max_t(u32, 18,
+					min_t(u32, dp_panel->dp_mode.bpp, 30));
+	DRM_DEBUG_DP("updated bpp = %d\n", dp_panel->dp_mode.bpp);
+
+	return rc;
+}
+
+struct dp_panel *dp_panel_get(struct dp_panel_in *in)
+{
+	struct dp_panel_private *panel;
+	struct dp_panel *dp_panel;
+
+	if (!in->dev || !in->catalog || !in->aux || !in->link) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	panel = devm_kzalloc(in->dev, sizeof(*panel), GFP_KERNEL);
+	if (!panel)
+		return ERR_PTR(-ENOMEM);
+
+	panel->dev = in->dev;
+	panel->aux = in->aux;
+	panel->catalog = in->catalog;
+	panel->link = in->link;
+
+	dp_panel = &panel->dp_panel;
+	dp_panel->max_bw_code = DP_LINK_BW_8_1;
+	panel->aux_cfg_update_done = false;
+
+	return dp_panel;
+}
+
+void dp_panel_put(struct dp_panel *dp_panel)
+{
+	if (!dp_panel)
+		return;
+
+	kfree(dp_panel->edid);
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_panel.h b/drivers/gpu/drm/msm/dp/dp_panel.h
new file mode 100644
index 000000000000..9023e5bb4b8b
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_panel.h
@@ -0,0 +1,100 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_PANEL_H_
+#define _DP_PANEL_H_
+
+#include <drm/msm_drm.h>
+
+#include "dp_aux.h"
+#include "dp_link.h"
+#include "dp_hpd.h"
+
+struct edid;
+
+#define DPRX_EXTENDED_DPCD_FIELD	0x2200
+
+#define DP_DOWNSTREAM_PORTS		4
+#define DP_DOWNSTREAM_CAP_SIZE		4
+
+struct dp_display_mode {
+	struct drm_display_mode drm_mode;
+	u32 capabilities;
+	u32 bpp;
+	u32 h_active_low;
+	u32 v_active_low;
+};
+
+struct dp_panel_in {
+	struct device *dev;
+	struct drm_dp_aux *aux;
+	struct dp_link *link;
+	struct dp_catalog *catalog;
+};
+
+struct dp_panel {
+	/* dpcd raw data */
+	u8 dpcd[DP_RECEIVER_CAP_SIZE + 1];
+	u8 ds_cap_info[DP_DOWNSTREAM_PORTS * DP_DOWNSTREAM_CAP_SIZE];
+	u32 ds_port_cnt;
+	u32 dfp_present;
+
+	struct dp_link_info link_info;
+	struct drm_dp_desc desc;
+	struct edid *edid;
+	struct drm_connector *connector;
+	struct dp_display_mode dp_mode;
+	bool video_test;
+
+	u32 vic;
+	u32 max_pclk_khz;
+	u32 max_dp_lanes;
+
+	u32 max_bw_code;
+};
+
+int dp_panel_init_panel_info(struct dp_panel *dp_panel);
+int dp_panel_deinit(struct dp_panel *dp_panel);
+int dp_panel_timing_cfg(struct dp_panel *dp_panel);
+void dp_panel_dump_regs(struct dp_panel *dp_panel);
+int dp_panel_read_sink_caps(struct dp_panel *dp_panel,
+		struct drm_connector *connector);
+u32 dp_panel_get_mode_bpp(struct dp_panel *dp_panel, u32 mode_max_bpp,
+			u32 mode_pclk_khz);
+int dp_panel_get_modes(struct dp_panel *dp_panel,
+		struct drm_connector *connector, struct dp_display_mode *mode);
+void dp_panel_handle_sink_request(struct dp_panel *dp_panel);
+void dp_panel_tpg_config(struct dp_panel *dp_panel, bool enable);
+
+/**
+ * is_link_rate_valid() - validates the link rate
+ * @lane_rate: link rate requested by the sink
+ *
+ * Returns true if the requested link rate is supported.
+ */
+static inline bool is_link_rate_valid(u32 bw_code)
+{
+	return (bw_code == DP_LINK_BW_1_62 ||
+		bw_code == DP_LINK_BW_2_7 ||
+		bw_code == DP_LINK_BW_5_4 ||
+		bw_code == DP_LINK_BW_8_1);
+}
+
+/**
+ * dp_link_is_lane_count_valid() - validates the lane count
+ * @lane_count: lane count requested by the sink
+ *
+ * Returns true if the requested lane count is supported.
+ */
+static inline bool is_lane_count_valid(u32 lane_count)
+{
+	return (lane_count == 1 ||
+		lane_count == 2 ||
+		lane_count == 4);
+}
+
+struct dp_panel *dp_panel_get(struct dp_panel_in *in);
+void dp_panel_put(struct dp_panel *dp_panel);
+#endif /* _DP_PANEL_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_parser.c b/drivers/gpu/drm/msm/dp/dp_parser.c
new file mode 100644
index 000000000000..0519dd3ac3c3
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_parser.c
@@ -0,0 +1,293 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/of_gpio.h>
+#include <linux/phy/phy.h>
+
+#include <drm/drm_print.h>
+
+#include "dp_parser.h"
+#include "dp_reg.h"
+
+static const struct dp_regulator_cfg sdm845_dp_reg_cfg = {
+	.num = 2,
+	.regs = {
+		{"vdda-1p2", 21800, 4 },	/* 1.2 V */
+		{"vdda-0p9", 36000, 32 },	/* 0.9 V */
+	},
+};
+
+static int msm_dss_ioremap(struct platform_device *pdev,
+				struct dss_io_data *io_data)
+{
+	struct resource *res = NULL;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		DRM_ERROR("%pS->%s: msm_dss_get_res failed\n",
+			__builtin_return_address(0), __func__);
+		return -ENODEV;
+	}
+
+	io_data->len = (u32)resource_size(res);
+	io_data->base = ioremap(res->start, io_data->len);
+	if (!io_data->base) {
+		DRM_ERROR("%pS->%s: ioremap failed\n",
+			__builtin_return_address(0), __func__);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static void msm_dss_iounmap(struct dss_io_data *io_data)
+{
+	if (io_data->base) {
+		iounmap(io_data->base);
+		io_data->base = NULL;
+	}
+	io_data->len = 0;
+}
+
+static void dp_parser_unmap_io_resources(struct dp_parser *parser)
+{
+	struct dp_io *io = &parser->io;
+
+	msm_dss_iounmap(&io->dp_controller);
+}
+
+static int dp_parser_ctrl_res(struct dp_parser *parser)
+{
+	int rc = 0;
+	struct platform_device *pdev = parser->pdev;
+	struct dp_io *io = &parser->io;
+
+	rc = msm_dss_ioremap(pdev, &io->dp_controller);
+	if (rc) {
+		DRM_ERROR("unable to remap dp io resources, rc=%d\n", rc);
+		goto err;
+	}
+
+	io->phy = devm_phy_get(&pdev->dev, "dp");
+	if (IS_ERR(io->phy)) {
+		rc = PTR_ERR(io->phy);
+		goto err;
+	}
+
+	return 0;
+err:
+	dp_parser_unmap_io_resources(parser);
+	return rc;
+}
+
+static int dp_parser_misc(struct dp_parser *parser)
+{
+	struct device_node *of_node = parser->pdev->dev.of_node;
+	int len = 0;
+	const char *data_lane_property = "data-lanes";
+
+	len = of_property_count_elems_of_size(of_node,
+			 data_lane_property, sizeof(u32));
+	if (len < 0) {
+		DRM_WARN("Invalid property %s, default max DP lanes = %d\n",
+				data_lane_property, DP_MAX_NUM_DP_LANES);
+		len = DP_MAX_NUM_DP_LANES;
+	}
+
+	parser->max_dp_lanes = len;
+	return 0;
+}
+
+static inline bool dp_parser_check_prefix(const char *clk_prefix,
+						const char *clk_name)
+{
+	return !strncmp(clk_prefix, clk_name, strlen(clk_prefix));
+}
+
+static int dp_parser_init_clk_data(struct dp_parser *parser)
+{
+	int num_clk, i, rc;
+	int core_clk_count = 0, ctrl_clk_count = 0, stream_clk_count = 0;
+	const char *clk_name;
+	struct device *dev = &parser->pdev->dev;
+	struct dss_module_power *core_power = &parser->mp[DP_CORE_PM];
+	struct dss_module_power *ctrl_power = &parser->mp[DP_CTRL_PM];
+	struct dss_module_power *stream_power = &parser->mp[DP_STREAM_PM];
+
+	num_clk = of_property_count_strings(dev->of_node, "clock-names");
+	if (num_clk <= 0) {
+		DRM_ERROR("no clocks are defined\n");
+		return -EINVAL;
+	}
+
+	for (i = 0; i < num_clk; i++) {
+		rc = of_property_read_string_index(dev->of_node,
+				"clock-names", i, &clk_name);
+		if (rc < 0)
+			return rc;
+
+		if (dp_parser_check_prefix("core", clk_name))
+			core_clk_count++;
+
+		if (dp_parser_check_prefix("ctrl", clk_name))
+			ctrl_clk_count++;
+
+		if (dp_parser_check_prefix("stream", clk_name))
+			stream_clk_count++;
+	}
+
+	/* Initialize the CORE power module */
+	if (core_clk_count == 0) {
+		DRM_ERROR("no core clocks are defined\n");
+		return -EINVAL;
+	}
+
+	core_power->num_clk = core_clk_count;
+	core_power->clk_config = devm_kzalloc(dev,
+			sizeof(struct dss_clk) * core_power->num_clk,
+			GFP_KERNEL);
+	if (!core_power->clk_config)
+		return -EINVAL;
+
+	/* Initialize the CTRL power module */
+	if (ctrl_clk_count == 0) {
+		DRM_ERROR("no ctrl clocks are defined\n");
+		return -EINVAL;
+	}
+
+	ctrl_power->num_clk = ctrl_clk_count;
+	ctrl_power->clk_config = devm_kzalloc(dev,
+			sizeof(struct dss_clk) * ctrl_power->num_clk,
+			GFP_KERNEL);
+	if (!ctrl_power->clk_config) {
+		ctrl_power->num_clk = 0;
+		return -EINVAL;
+	}
+
+	/* Initialize the STREAM power module */
+	if (stream_clk_count == 0) {
+		DRM_ERROR("no stream (pixel) clocks are defined\n");
+		return -EINVAL;
+	}
+
+	stream_power->num_clk = stream_clk_count;
+	stream_power->clk_config = devm_kzalloc(dev,
+			sizeof(struct dss_clk) * stream_power->num_clk,
+			GFP_KERNEL);
+	if (!stream_power->clk_config) {
+		stream_power->num_clk = 0;
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int dp_parser_clock(struct dp_parser *parser)
+{
+	int rc = 0, i = 0;
+	int num_clk = 0;
+	int core_clk_index = 0, ctrl_clk_index = 0, stream_clk_index = 0;
+	int core_clk_count = 0, ctrl_clk_count = 0, stream_clk_count = 0;
+	const char *clk_name;
+	struct device *dev = &parser->pdev->dev;
+	struct dss_module_power *core_power = &parser->mp[DP_CORE_PM];
+	struct dss_module_power *ctrl_power = &parser->mp[DP_CTRL_PM];
+	struct dss_module_power *stream_power = &parser->mp[DP_STREAM_PM];
+
+	rc =  dp_parser_init_clk_data(parser);
+	if (rc) {
+		DRM_ERROR("failed to initialize power data %d\n", rc);
+		return -EINVAL;
+	}
+
+	core_clk_count = core_power->num_clk;
+	ctrl_clk_count = ctrl_power->num_clk;
+	stream_clk_count = stream_power->num_clk;
+
+	num_clk = core_clk_count + ctrl_clk_count + stream_clk_count;
+
+	for (i = 0; i < num_clk; i++) {
+		rc = of_property_read_string_index(dev->of_node, "clock-names",
+				i, &clk_name);
+		if (rc) {
+			DRM_ERROR("error reading clock-names %d\n", rc);
+			return rc;
+		}
+		if (dp_parser_check_prefix("core", clk_name) &&
+				core_clk_index < core_clk_count) {
+			struct dss_clk *clk =
+				&core_power->clk_config[core_clk_index];
+			strlcpy(clk->clk_name, clk_name, sizeof(clk->clk_name));
+			clk->type = DSS_CLK_AHB;
+			core_clk_index++;
+		} else if (dp_parser_check_prefix("stream", clk_name) &&
+				stream_clk_index < stream_clk_count) {
+			struct dss_clk *clk =
+				&stream_power->clk_config[stream_clk_index];
+			strlcpy(clk->clk_name, clk_name, sizeof(clk->clk_name));
+			clk->type = DSS_CLK_PCLK;
+			stream_clk_index++;
+		} else if (dp_parser_check_prefix("ctrl", clk_name) &&
+			   ctrl_clk_index < ctrl_clk_count) {
+			struct dss_clk *clk =
+				&ctrl_power->clk_config[ctrl_clk_index];
+			strlcpy(clk->clk_name, clk_name, sizeof(clk->clk_name));
+			ctrl_clk_index++;
+			if (dp_parser_check_prefix("ctrl_link", clk_name) ||
+			    dp_parser_check_prefix("stream_pixel", clk_name))
+				clk->type = DSS_CLK_PCLK;
+			else
+				clk->type = DSS_CLK_AHB;
+		}
+	}
+
+	DRM_DEBUG_DP("clock parsing successful\n");
+
+	return 0;
+}
+
+static int dp_parser_parse(struct dp_parser *parser)
+{
+	int rc = 0;
+
+	if (!parser) {
+		DRM_ERROR("invalid input\n");
+		return -EINVAL;
+	}
+
+	rc = dp_parser_ctrl_res(parser);
+	if (rc)
+		return rc;
+
+	rc = dp_parser_misc(parser);
+	if (rc)
+		return rc;
+
+	rc = dp_parser_clock(parser);
+	if (rc)
+		return rc;
+
+	/* Map the corresponding regulator information according to
+	 * version. Currently, since we only have one supported platform,
+	 * mapping the regulator directly.
+	 */
+	parser->regulator_cfg = &sdm845_dp_reg_cfg;
+
+	return 0;
+}
+
+struct dp_parser *dp_parser_get(struct platform_device *pdev)
+{
+	struct dp_parser *parser;
+
+	parser = devm_kzalloc(&pdev->dev, sizeof(*parser), GFP_KERNEL);
+	if (!parser)
+		return ERR_PTR(-ENOMEM);
+
+	parser->parse = dp_parser_parse;
+	parser->pdev = pdev;
+
+	return parser;
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_parser.h b/drivers/gpu/drm/msm/dp/dp_parser.h
new file mode 100644
index 000000000000..34b49628bbaf
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_parser.h
@@ -0,0 +1,136 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_PARSER_H_
+#define _DP_PARSER_H_
+
+#include <linux/platform_device.h>
+#include <linux/phy/phy.h>
+#include <linux/phy/phy-dp.h>
+
+#include "dpu_io_util.h"
+#include "msm_drv.h"
+
+#define DP_LABEL "MDSS DP DISPLAY"
+#define DP_MAX_PIXEL_CLK_KHZ	675000
+#define DP_MAX_NUM_DP_LANES	4
+
+enum dp_pm_type {
+	DP_CORE_PM,
+	DP_CTRL_PM,
+	DP_STREAM_PM,
+	DP_PHY_PM,
+	DP_MAX_PM
+};
+
+struct dss_io_data {
+	u32 len;
+	void __iomem *base;
+};
+
+static inline const char *dp_parser_pm_name(enum dp_pm_type module)
+{
+	switch (module) {
+	case DP_CORE_PM:	return "DP_CORE_PM";
+	case DP_CTRL_PM:	return "DP_CTRL_PM";
+	case DP_STREAM_PM:	return "DP_STREAM_PM";
+	case DP_PHY_PM:		return "DP_PHY_PM";
+	default:		return "???";
+	}
+}
+
+/**
+ * struct dp_display_data  - display related device tree data.
+ *
+ * @ctrl_node: referece to controller device
+ * @phy_node:  reference to phy device
+ * @is_active: is the controller currently active
+ * @name: name of the display
+ * @display_type: type of the display
+ */
+struct dp_display_data {
+	struct device_node *ctrl_node;
+	struct device_node *phy_node;
+	bool is_active;
+	const char *name;
+	const char *display_type;
+};
+
+/**
+ * struct dp_ctrl_resource - controller's IO related data
+ *
+ * @dp_controller: Display Port controller mapped memory address
+ * @phy_io: phy's mapped memory address
+ */
+struct dp_io {
+	struct dss_io_data dp_controller;
+	struct phy *phy;
+	union phy_configure_opts phy_opts;
+};
+
+/**
+ * struct dp_pinctrl - DP's pin control
+ *
+ * @pin: pin-controller's instance
+ * @state_active: active state pin control
+ * @state_hpd_active: hpd active state pin control
+ * @state_suspend: suspend state pin control
+ */
+struct dp_pinctrl {
+	struct pinctrl *pin;
+	struct pinctrl_state *state_active;
+	struct pinctrl_state *state_hpd_active;
+	struct pinctrl_state *state_suspend;
+};
+
+#define DP_DEV_REGULATOR_MAX	4
+
+/* Regulators for DP devices */
+struct dp_reg_entry {
+	char name[32];
+	int enable_load;
+	int disable_load;
+};
+
+struct dp_regulator_cfg {
+	int num;
+	struct dp_reg_entry regs[DP_DEV_REGULATOR_MAX];
+};
+
+/**
+ * struct dp_parser - DP parser's data exposed to clients
+ *
+ * @pdev: platform data of the client
+ * @mp: gpio, regulator and clock related data
+ * @pinctrl: pin-control related data
+ * @disp_data: controller's display related data
+ * @parse: function to be called by client to parse device tree.
+ */
+struct dp_parser {
+	struct platform_device *pdev;
+	struct dss_module_power mp[DP_MAX_PM];
+	struct dp_pinctrl pinctrl;
+	struct dp_io io;
+	struct dp_display_data disp_data;
+	const struct dp_regulator_cfg *regulator_cfg;
+	u32 max_dp_lanes;
+
+	int (*parse)(struct dp_parser *parser);
+};
+
+/**
+ * dp_parser_get() - get the DP's device tree parser module
+ *
+ * @pdev: platform data of the client
+ * return: pointer to dp_parser structure.
+ *
+ * This function provides client capability to parse the
+ * device tree and populate the data structures. The data
+ * related to clock, regulators, pin-control and other
+ * can be parsed using this module.
+ */
+struct dp_parser *dp_parser_get(struct platform_device *pdev);
+
+#endif
diff --git a/drivers/gpu/drm/msm/dp/dp_power.c b/drivers/gpu/drm/msm/dp/dp_power.c
new file mode 100644
index 000000000000..17c1fc6a2d44
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_power.c
@@ -0,0 +1,372 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/regulator/consumer.h>
+#include "dp_power.h"
+#include "msm_drv.h"
+
+struct dp_power_private {
+	struct dp_parser *parser;
+	struct platform_device *pdev;
+	struct clk *link_clk_src;
+	struct clk *pixel_provider;
+	struct clk *link_provider;
+	struct regulator_bulk_data supplies[DP_DEV_REGULATOR_MAX];
+
+	struct dp_power dp_power;
+};
+
+static void dp_power_regulator_disable(struct dp_power_private *power)
+{
+	struct regulator_bulk_data *s = power->supplies;
+	const struct dp_reg_entry *regs = power->parser->regulator_cfg->regs;
+	int num = power->parser->regulator_cfg->num;
+	int i;
+
+	DBG("");
+	for (i = num - 1; i >= 0; i--)
+		if (regs[i].disable_load >= 0)
+			regulator_set_load(s[i].consumer,
+					   regs[i].disable_load);
+
+	regulator_bulk_disable(num, s);
+}
+
+static int dp_power_regulator_enable(struct dp_power_private *power)
+{
+	struct regulator_bulk_data *s = power->supplies;
+	const struct dp_reg_entry *regs = power->parser->regulator_cfg->regs;
+	int num = power->parser->regulator_cfg->num;
+	int ret, i;
+
+	DBG("");
+	for (i = 0; i < num; i++) {
+		if (regs[i].enable_load >= 0) {
+			ret = regulator_set_load(s[i].consumer,
+						 regs[i].enable_load);
+			if (ret < 0) {
+				pr_err("regulator %d set op mode failed, %d\n",
+					i, ret);
+				goto fail;
+			}
+		}
+	}
+
+	ret = regulator_bulk_enable(num, s);
+	if (ret < 0) {
+		pr_err("regulator enable failed, %d\n", ret);
+		goto fail;
+	}
+
+	return 0;
+
+fail:
+	for (i--; i >= 0; i--)
+		regulator_set_load(s[i].consumer, regs[i].disable_load);
+	return ret;
+}
+
+static int dp_power_regulator_init(struct dp_power_private *power)
+{
+	struct regulator_bulk_data *s = power->supplies;
+	const struct dp_reg_entry *regs = power->parser->regulator_cfg->regs;
+	struct platform_device *pdev = power->pdev;
+	int num = power->parser->regulator_cfg->num;
+	int i, ret;
+
+	for (i = 0; i < num; i++)
+		s[i].supply = regs[i].name;
+
+	ret = devm_regulator_bulk_get(&pdev->dev, num, s);
+	if (ret < 0) {
+		pr_err("%s: failed to init regulator, ret=%d\n",
+						__func__, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int dp_power_clk_init(struct dp_power_private *power)
+{
+	int rc = 0;
+	struct dss_module_power *core, *ctrl, *stream;
+	struct device *dev = &power->pdev->dev;
+
+	core = &power->parser->mp[DP_CORE_PM];
+	ctrl = &power->parser->mp[DP_CTRL_PM];
+	stream = &power->parser->mp[DP_STREAM_PM];
+
+	rc = msm_dss_get_clk(dev, core->clk_config, core->num_clk);
+	if (rc) {
+		DRM_ERROR("failed to get %s clk. err=%d\n",
+			dp_parser_pm_name(DP_CORE_PM), rc);
+		return rc;
+	}
+
+	rc = msm_dss_get_clk(dev, ctrl->clk_config, ctrl->num_clk);
+	if (rc) {
+		DRM_ERROR("failed to get %s clk. err=%d\n",
+			dp_parser_pm_name(DP_CTRL_PM), rc);
+		msm_dss_put_clk(core->clk_config, core->num_clk);
+		return -ENODEV;
+	}
+
+	rc = msm_dss_get_clk(dev, stream->clk_config, stream->num_clk);
+	if (rc) {
+		DRM_ERROR("failed to get %s clk. err=%d\n",
+			dp_parser_pm_name(DP_CTRL_PM), rc);
+		msm_dss_put_clk(core->clk_config, core->num_clk);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int dp_power_clk_deinit(struct dp_power_private *power)
+{
+	struct dss_module_power *core, *ctrl, *stream;
+
+	core = &power->parser->mp[DP_CORE_PM];
+	ctrl = &power->parser->mp[DP_CTRL_PM];
+	stream = &power->parser->mp[DP_STREAM_PM];
+
+	if (!core || !ctrl || !stream) {
+		DRM_ERROR("invalid power_data\n");
+		return -EINVAL;
+	}
+
+	msm_dss_put_clk(ctrl->clk_config, ctrl->num_clk);
+	msm_dss_put_clk(core->clk_config, core->num_clk);
+	msm_dss_put_clk(stream->clk_config, stream->num_clk);
+	return 0;
+}
+
+static int dp_power_clk_set_rate(struct dp_power_private *power,
+		enum dp_pm_type module, bool enable)
+{
+	int rc = 0;
+	struct dss_module_power *mp = &power->parser->mp[module];
+
+	if (enable) {
+		rc = msm_dss_clk_set_rate(mp->clk_config, mp->num_clk);
+		if (rc) {
+			DRM_ERROR("failed to set clks rate.\n");
+			return rc;
+		}
+	}
+
+	rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, enable);
+	if (rc) {
+		DRM_ERROR("failed to %d clks, err: %d\n", enable, rc);
+		return rc;
+	}
+
+	return 0;
+}
+
+int dp_power_clk_status(struct dp_power *dp_power, enum dp_pm_type pm_type)
+{
+	if (pm_type == DP_CORE_PM)
+		return dp_power->core_clks_on;
+
+	if (pm_type == DP_CTRL_PM)
+		return dp_power->link_clks_on;
+
+	if (pm_type == DP_STREAM_PM)
+		return dp_power->stream_clks_on;
+
+	return 0;
+}
+
+int dp_power_clk_enable(struct dp_power *dp_power,
+		enum dp_pm_type pm_type, bool enable)
+{
+	int rc = 0;
+	struct dp_power_private *power;
+
+	power = container_of(dp_power, struct dp_power_private, dp_power);
+
+	if (pm_type != DP_CORE_PM && pm_type != DP_CTRL_PM &&
+			pm_type != DP_STREAM_PM) {
+		DRM_ERROR("unsupported power module: %s\n",
+				dp_parser_pm_name(pm_type));
+		return -EINVAL;
+	}
+
+	if (enable) {
+		if (pm_type == DP_CORE_PM && dp_power->core_clks_on) {
+			DRM_DEBUG_DP("core clks already enabled\n");
+			return 0;
+		}
+
+		if (pm_type == DP_CTRL_PM && dp_power->link_clks_on) {
+			DRM_DEBUG_DP("links clks already enabled\n");
+			return 0;
+		}
+
+		if (pm_type == DP_STREAM_PM && dp_power->stream_clks_on) {
+			DRM_DEBUG_DP("pixel clks already enabled\n");
+			return 0;
+		}
+
+		if ((pm_type == DP_CTRL_PM) && (!dp_power->core_clks_on)) {
+			DRM_DEBUG_DP("Enable core clks before link clks\n");
+
+			rc = dp_power_clk_set_rate(power, DP_CORE_PM, enable);
+			if (rc) {
+				DRM_ERROR("fail to enable clks: %s. err=%d\n",
+					dp_parser_pm_name(DP_CORE_PM), rc);
+				return rc;
+			}
+			dp_power->core_clks_on = true;
+		}
+	}
+
+	rc = dp_power_clk_set_rate(power, pm_type, enable);
+	if (rc) {
+		DRM_ERROR("failed to '%s' clks for: %s. err=%d\n",
+			enable ? "enable" : "disable",
+			dp_parser_pm_name(pm_type), rc);
+			return rc;
+	}
+
+	if (pm_type == DP_CORE_PM)
+		dp_power->core_clks_on = enable;
+	else if (pm_type == DP_STREAM_PM)
+		dp_power->stream_clks_on = enable;
+	else
+		dp_power->link_clks_on = enable;
+
+	DRM_DEBUG_DP("%s clocks for %s\n",
+			enable ? "enable" : "disable",
+			dp_parser_pm_name(pm_type));
+	DRM_DEBUG_DP("strem_clks:%s link_clks:%s core_clks:%s\n",
+		dp_power->stream_clks_on ? "on" : "off",
+		dp_power->link_clks_on ? "on" : "off",
+		dp_power->core_clks_on ? "on" : "off");
+
+	return 0;
+}
+
+int dp_power_client_init(struct dp_power *dp_power)
+{
+	int rc = 0;
+	struct dp_power_private *power;
+
+	if (!dp_power) {
+		DRM_ERROR("invalid power data\n");
+		return -EINVAL;
+	}
+
+	power = container_of(dp_power, struct dp_power_private, dp_power);
+
+	pm_runtime_enable(&power->pdev->dev);
+
+	rc = dp_power_regulator_init(power);
+	if (rc) {
+		DRM_ERROR("failed to init regulators %d\n", rc);
+		goto error;
+	}
+
+	rc = dp_power_clk_init(power);
+	if (rc) {
+		DRM_ERROR("failed to init clocks %d\n", rc);
+		goto error;
+	}
+	return 0;
+
+error:
+	pm_runtime_disable(&power->pdev->dev);
+	return rc;
+}
+
+void dp_power_client_deinit(struct dp_power *dp_power)
+{
+	struct dp_power_private *power;
+
+	if (!dp_power) {
+		DRM_ERROR("invalid power data\n");
+		return;
+	}
+
+	power = container_of(dp_power, struct dp_power_private, dp_power);
+
+	dp_power_clk_deinit(power);
+	pm_runtime_disable(&power->pdev->dev);
+
+}
+
+int dp_power_init(struct dp_power *dp_power, bool flip)
+{
+	int rc = 0;
+	struct dp_power_private *power = NULL;
+
+	if (!dp_power) {
+		DRM_ERROR("invalid power data\n");
+		return -EINVAL;
+	}
+
+	power = container_of(dp_power, struct dp_power_private, dp_power);
+
+	pm_runtime_get_sync(&power->pdev->dev);
+	rc = dp_power_regulator_enable(power);
+	if (rc) {
+		DRM_ERROR("failed to enable regulators, %d\n", rc);
+		goto exit;
+	}
+
+	rc = dp_power_clk_enable(dp_power, DP_CORE_PM, true);
+	if (rc) {
+		DRM_ERROR("failed to enable DP core clocks, %d\n", rc);
+		goto err_clk;
+	}
+
+	return 0;
+
+err_clk:
+	dp_power_regulator_disable(power);
+exit:
+	pm_runtime_put_sync(&power->pdev->dev);
+	return rc;
+}
+
+int dp_power_deinit(struct dp_power *dp_power)
+{
+	struct dp_power_private *power;
+
+	power = container_of(dp_power, struct dp_power_private, dp_power);
+
+	dp_power_clk_enable(dp_power, DP_CORE_PM, false);
+	dp_power_regulator_disable(power);
+	pm_runtime_put_sync(&power->pdev->dev);
+	return 0;
+}
+
+struct dp_power *dp_power_get(struct dp_parser *parser)
+{
+	struct dp_power_private *power;
+	struct dp_power *dp_power;
+
+	if (!parser) {
+		DRM_ERROR("invalid input\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	power = devm_kzalloc(&parser->pdev->dev, sizeof(*power), GFP_KERNEL);
+	if (!power)
+		return ERR_PTR(-ENOMEM);
+
+	power->parser = parser;
+	power->pdev = parser->pdev;
+
+	dp_power = &power->dp_power;
+
+	return dp_power;
+}
diff --git a/drivers/gpu/drm/msm/dp/dp_power.h b/drivers/gpu/drm/msm/dp/dp_power.h
new file mode 100644
index 000000000000..76743d755833
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_power.h
@@ -0,0 +1,107 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_POWER_H_
+#define _DP_POWER_H_
+
+#include "dp_parser.h"
+
+/**
+ * sruct dp_power - DisplayPort's power related data
+ *
+ * @init: initializes the regulators/core clocks/GPIOs/pinctrl
+ * @deinit: turns off the regulators/core clocks/GPIOs/pinctrl
+ * @clk_enable: enable/disable the DP clocks
+ * @set_pixel_clk_parent: set the parent of DP pixel clock
+ */
+struct dp_power {
+	bool core_clks_on;
+	bool link_clks_on;
+	bool stream_clks_on;
+};
+
+/**
+ * dp_power_init() - enable power supplies for display controller
+ *
+ * @power: instance of power module
+ * @flip: bool for flipping gpio direction
+ * return: 0 if success or error if failure.
+ *
+ * This API will turn on the regulators and configures gpio's
+ * aux/hpd.
+ */
+int dp_power_init(struct dp_power *power, bool flip);
+
+/**
+ * dp_power_deinit() - turn off regulators and gpios.
+ *
+ * @power: instance of power module
+ * return: 0 for success
+ *
+ * This API turns off power and regulators.
+ */
+int dp_power_deinit(struct dp_power *power);
+
+/**
+ * dp_power_clk_status() - display controller clocks status
+ *
+ * @power: instance of power module
+ * @pm_type: type of pm, core/ctrl/phy
+ * return: status of power clocks
+ *
+ * This API return status of DP clocks
+ */
+
+int dp_power_clk_status(struct dp_power *dp_power, enum dp_pm_type pm_type);
+
+/**
+ * dp_power_clk_enable() - enable display controller clocks
+ *
+ * @power: instance of power module
+ * @pm_type: type of pm, core/ctrl/phy
+ * @enable: enables or disables
+ * return: pointer to allocated power module data
+ *
+ * This API will call setrate and enable for DP clocks
+ */
+
+int dp_power_clk_enable(struct dp_power *power, enum dp_pm_type pm_type,
+				bool enable);
+
+/**
+ * dp_power_client_init() - initialize clock and regulator modules
+ *
+ * @power: instance of power module
+ * return: 0 for success, error for failure.
+ *
+ * This API will configure the DisplayPort's clocks and regulator
+ * modules.
+ */
+int dp_power_client_init(struct dp_power *power);
+
+/**
+ * dp_power_clinet_deinit() - de-initialize clock and regulator modules
+ *
+ * @power: instance of power module
+ * return: 0 for success, error for failure.
+ *
+ * This API will de-initialize the DisplayPort's clocks and regulator
+ * modueles.
+ */
+void dp_power_client_deinit(struct dp_power *power);
+
+/**
+ * dp_power_get() - configure and get the DisplayPort power module data
+ *
+ * @parser: instance of parser module
+ * return: pointer to allocated power module data
+ *
+ * This API will configure the DisplayPort's power module and provides
+ * methods to be called by the client to configure the power related
+ * modueles.
+ */
+struct dp_power *dp_power_get(struct dp_parser *parser);
+
+#endif /* _DP_POWER_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_reg.h b/drivers/gpu/drm/msm/dp/dp_reg.h
new file mode 100644
index 000000000000..43042ff90a19
--- /dev/null
+++ b/drivers/gpu/drm/msm/dp/dp_reg.h
@@ -0,0 +1,306 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DP_REG_H_
+#define _DP_REG_H_
+
+/* DP_TX Registers */
+#define REG_DP_HW_VERSION			(0x00000000)
+
+#define REG_DP_SW_RESET				(0x00000010)
+#define DP_SW_RESET				(0x00000001)
+
+#define REG_DP_PHY_CTRL				(0x00000014)
+#define DP_PHY_CTRL_SW_RESET_PLL		(0x00000001)
+#define DP_PHY_CTRL_SW_RESET			(0x00000004)
+
+#define REG_DP_CLK_CTRL				(0x00000018)
+#define REG_DP_CLK_ACTIVE			(0x0000001C)
+#define REG_DP_INTR_STATUS			(0x00000020)
+#define REG_DP_INTR_STATUS2			(0x00000024)
+#define REG_DP_INTR_STATUS3			(0x00000028)
+
+#define REG_DP_DP_HPD_CTRL			(0x00000000)
+#define DP_DP_HPD_CTRL_HPD_EN			(0x00000001)
+
+#define REG_DP_DP_HPD_INT_STATUS		(0x00000004)
+
+#define REG_DP_DP_HPD_INT_ACK			(0x00000008)
+#define DP_DP_HPD_PLUG_INT_ACK			(0x00000001)
+#define DP_DP_IRQ_HPD_INT_ACK			(0x00000002)
+#define DP_DP_HPD_REPLUG_INT_ACK		(0x00000004)
+#define DP_DP_HPD_UNPLUG_INT_ACK		(0x00000008)
+
+#define REG_DP_DP_HPD_INT_MASK			(0x0000000C)
+#define DP_DP_HPD_PLUG_INT_MASK			(0x00000001)
+#define DP_DP_IRQ_HPD_INT_MASK			(0x00000002)
+#define DP_DP_HPD_REPLUG_INT_MASK		(0x00000004)
+#define DP_DP_HPD_UNPLUG_INT_MASK		(0x00000008)
+#define DP_DP_HPD_INT_MASK			(DP_DP_HPD_PLUG_INT_MASK | \
+						DP_DP_IRQ_HPD_INT_MASK | \
+						DP_DP_HPD_REPLUG_INT_MASK | \
+						DP_DP_HPD_UNPLUG_INT_MASK)
+#define DP_DP_HPD_STATE_STATUS_CONNECTED	(0x40000000)
+#define DP_DP_HPD_STATE_STATUS_PENDING		(0x20000000)
+#define DP_DP_HPD_STATE_STATUS_DISCONNECTED	(0x00000000)
+#define DP_DP_HPD_STATE_STATUS_MASK		(0xE0000000)
+
+#define REG_DP_DP_HPD_REFTIMER			(0x00000018)
+#define DP_DP_HPD_REFTIMER_ENABLE		(1 << 16)
+
+#define REG_DP_DP_HPD_EVENT_TIME_0		(0x0000001C)
+#define REG_DP_DP_HPD_EVENT_TIME_1		(0x00000020)
+#define DP_DP_HPD_EVENT_TIME_0_VAL		(0x3E800FA)
+#define DP_DP_HPD_EVENT_TIME_1_VAL		(0x1F407D0)
+
+#define REG_DP_AUX_CTRL				(0x00000030)
+#define DP_AUX_CTRL_ENABLE			(0x00000001)
+#define DP_AUX_CTRL_RESET			(0x00000002)
+
+#define REG_DP_AUX_DATA				(0x00000034)
+#define DP_AUX_DATA_READ			(0x00000001)
+#define DP_AUX_DATA_WRITE			(0x00000000)
+#define DP_AUX_DATA_OFFSET			(0x00000008)
+#define DP_AUX_DATA_INDEX_OFFSET		(0x00000010)
+#define DP_AUX_DATA_MASK			(0x0000ff00)
+#define DP_AUX_DATA_INDEX_WRITE			(0x80000000)
+
+#define REG_DP_AUX_TRANS_CTRL			(0x00000038)
+#define DP_AUX_TRANS_CTRL_I2C			(0x00000100)
+#define DP_AUX_TRANS_CTRL_GO			(0x00000200)
+#define DP_AUX_TRANS_CTRL_NO_SEND_ADDR		(0x00000400)
+#define DP_AUX_TRANS_CTRL_NO_SEND_STOP		(0x00000800)
+
+#define REG_DP_TIMEOUT_COUNT			(0x0000003C)
+#define REG_DP_AUX_LIMITS			(0x00000040)
+#define REG_DP_AUX_STATUS			(0x00000044)
+
+#define DP_DPCD_CP_IRQ				(0x201)
+#define DP_DPCD_RXSTATUS			(0x69493)
+
+#define DP_INTERRUPT_TRANS_NUM			(0x000000A0)
+
+#define REG_DP_MAINLINK_CTRL			(0x00000000)
+#define DP_MAINLINK_CTRL_ENABLE			(0x00000001)
+#define DP_MAINLINK_CTRL_RESET			(0x00000002)
+#define DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER	(0x00000010)
+#define DP_MAINLINK_FB_BOUNDARY_SEL		(0x02000000)
+
+#define REG_DP_STATE_CTRL			(0x00000004)
+#define DP_STATE_CTRL_LINK_TRAINING_PATTERN1	(0x00000001)
+#define DP_STATE_CTRL_LINK_TRAINING_PATTERN2	(0x00000002)
+#define DP_STATE_CTRL_LINK_TRAINING_PATTERN3	(0x00000004)
+#define DP_STATE_CTRL_LINK_TRAINING_PATTERN4	(0x00000008)
+#define DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE	(0x00000010)
+#define DP_STATE_CTRL_LINK_PRBS7		(0x00000020)
+#define DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN	(0x00000040)
+#define DP_STATE_CTRL_SEND_VIDEO		(0x00000080)
+#define DP_STATE_CTRL_PUSH_IDLE			(0x00000100)
+
+#define REG_DP_CONFIGURATION_CTRL		(0x00000008)
+#define DP_CONFIGURATION_CTRL_SYNC_ASYNC_CLK	(0x00000001)
+#define DP_CONFIGURATION_CTRL_STATIC_DYNAMIC_CN (0x00000002)
+#define DP_CONFIGURATION_CTRL_P_INTERLACED	(0x00000004)
+#define DP_CONFIGURATION_CTRL_INTERLACED_BTF	(0x00000008)
+#define DP_CONFIGURATION_CTRL_NUM_OF_LANES	(0x00000010)
+#define DP_CONFIGURATION_CTRL_ENHANCED_FRAMING	(0x00000040)
+#define DP_CONFIGURATION_CTRL_SEND_VSC		(0x00000080)
+#define DP_CONFIGURATION_CTRL_BPC		(0x00000100)
+#define DP_CONFIGURATION_CTRL_ASSR		(0x00000400)
+#define DP_CONFIGURATION_CTRL_RGB_YUV		(0x00000800)
+#define DP_CONFIGURATION_CTRL_LSCLK_DIV		(0x00002000)
+#define DP_CONFIGURATION_CTRL_NUM_OF_LANES_SHIFT	(0x04)
+#define DP_CONFIGURATION_CTRL_BPC_SHIFT		(0x08)
+#define DP_CONFIGURATION_CTRL_LSCLK_DIV_SHIFT	(0x0D)
+
+#define REG_DP_SOFTWARE_MVID			(0x00000010)
+#define REG_DP_SOFTWARE_NVID			(0x00000018)
+#define REG_DP_TOTAL_HOR_VER			(0x0000001C)
+#define REG_DP_START_HOR_VER_FROM_SYNC		(0x00000020)
+#define REG_DP_HSYNC_VSYNC_WIDTH_POLARITY	(0x00000024)
+#define REG_DP_ACTIVE_HOR_VER			(0x00000028)
+
+#define REG_DP_MISC1_MISC0			(0x0000002C)
+#define DP_MISC0_SYNCHRONOUS_CLK		(0x00000001)
+#define DP_MISC0_COLORIMETRY_CFG_SHIFT		(0x00000001)
+#define DP_MISC0_TEST_BITS_DEPTH_SHIFT		(0x00000005)
+
+#define REG_DP_VALID_BOUNDARY			(0x00000030)
+#define REG_DP_VALID_BOUNDARY_2			(0x00000034)
+
+#define REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING	(0x00000038)
+#define LANE0_MAPPING_SHIFT			(0x00000000)
+#define LANE1_MAPPING_SHIFT			(0x00000002)
+#define LANE2_MAPPING_SHIFT			(0x00000004)
+#define LANE3_MAPPING_SHIFT			(0x00000006)
+
+#define REG_DP_MAINLINK_READY			(0x00000040)
+#define DP_MAINLINK_READY_FOR_VIDEO		(0x00000001)
+#define DP_MAINLINK_READY_LINK_TRAINING_SHIFT	(0x00000003)
+
+#define REG_DP_MAINLINK_LEVELS			(0x00000044)
+#define DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2	(0x00000002)
+
+
+#define REG_DP_TU				(0x0000004C)
+
+#define REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET	(0x00000054)
+#define DP_HBR2_ERM_PATTERN			(0x00010000)
+
+#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0	(0x000000C0)
+#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1	(0x000000C4)
+#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2	(0x000000C8)
+
+#define MMSS_DP_MISC1_MISC0			(0x0000002C)
+#define MMSS_DP_AUDIO_TIMING_GEN		(0x00000080)
+#define MMSS_DP_AUDIO_TIMING_RBR_32		(0x00000084)
+#define MMSS_DP_AUDIO_TIMING_HBR_32		(0x00000088)
+#define MMSS_DP_AUDIO_TIMING_RBR_44		(0x0000008C)
+#define MMSS_DP_AUDIO_TIMING_HBR_44		(0x00000090)
+#define MMSS_DP_AUDIO_TIMING_RBR_48		(0x00000094)
+#define MMSS_DP_AUDIO_TIMING_HBR_48		(0x00000098)
+
+#define MMSS_DP_PSR_CRC_RG			(0x00000154)
+#define MMSS_DP_PSR_CRC_B			(0x00000158)
+
+#define REG_DP_COMPRESSION_MODE_CTRL		(0x00000180)
+
+#define MMSS_DP_AUDIO_CFG			(0x00000200)
+#define MMSS_DP_AUDIO_STATUS			(0x00000204)
+#define MMSS_DP_AUDIO_PKT_CTRL			(0x00000208)
+#define MMSS_DP_AUDIO_PKT_CTRL2			(0x0000020C)
+#define MMSS_DP_AUDIO_ACR_CTRL			(0x00000210)
+#define MMSS_DP_AUDIO_CTRL_RESET		(0x00000214)
+
+#define MMSS_DP_SDP_CFG				(0x00000228)
+#define MMSS_DP_SDP_CFG2			(0x0000022C)
+#define MMSS_DP_AUDIO_TIMESTAMP_0		(0x00000230)
+#define MMSS_DP_AUDIO_TIMESTAMP_1		(0x00000234)
+
+#define MMSS_DP_AUDIO_STREAM_0			(0x00000240)
+#define MMSS_DP_AUDIO_STREAM_1			(0x00000244)
+
+#define MMSS_DP_EXTENSION_0			(0x00000250)
+#define MMSS_DP_EXTENSION_1			(0x00000254)
+#define MMSS_DP_EXTENSION_2			(0x00000258)
+#define MMSS_DP_EXTENSION_3			(0x0000025C)
+#define MMSS_DP_EXTENSION_4			(0x00000260)
+#define MMSS_DP_EXTENSION_5			(0x00000264)
+#define MMSS_DP_EXTENSION_6			(0x00000268)
+#define MMSS_DP_EXTENSION_7			(0x0000026C)
+#define MMSS_DP_EXTENSION_8			(0x00000270)
+#define MMSS_DP_EXTENSION_9			(0x00000274)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_0		(0x00000278)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_1		(0x0000027C)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_2		(0x00000280)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_3		(0x00000284)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_4		(0x00000288)
+#define MMSS_DP_AUDIO_COPYMANAGEMENT_5		(0x0000028C)
+#define MMSS_DP_AUDIO_ISRC_0			(0x00000290)
+#define MMSS_DP_AUDIO_ISRC_1			(0x00000294)
+#define MMSS_DP_AUDIO_ISRC_2			(0x00000298)
+#define MMSS_DP_AUDIO_ISRC_3			(0x0000029C)
+#define MMSS_DP_AUDIO_ISRC_4			(0x000002A0)
+#define MMSS_DP_AUDIO_ISRC_5			(0x000002A4)
+#define MMSS_DP_AUDIO_INFOFRAME_0		(0x000002A8)
+#define MMSS_DP_AUDIO_INFOFRAME_1		(0x000002AC)
+#define MMSS_DP_AUDIO_INFOFRAME_2		(0x000002B0)
+
+#define MMSS_DP_GENERIC0_0			(0x00000300)
+#define MMSS_DP_GENERIC0_1			(0x00000304)
+#define MMSS_DP_GENERIC0_2			(0x00000308)
+#define MMSS_DP_GENERIC0_3			(0x0000030C)
+#define MMSS_DP_GENERIC0_4			(0x00000310)
+#define MMSS_DP_GENERIC0_5			(0x00000314)
+#define MMSS_DP_GENERIC0_6			(0x00000318)
+#define MMSS_DP_GENERIC0_7			(0x0000031C)
+#define MMSS_DP_GENERIC0_8			(0x00000320)
+#define MMSS_DP_GENERIC0_9			(0x00000324)
+#define MMSS_DP_GENERIC1_0			(0x00000328)
+#define MMSS_DP_GENERIC1_1			(0x0000032C)
+#define MMSS_DP_GENERIC1_2			(0x00000330)
+#define MMSS_DP_GENERIC1_3			(0x00000334)
+#define MMSS_DP_GENERIC1_4			(0x00000338)
+#define MMSS_DP_GENERIC1_5			(0x0000033C)
+#define MMSS_DP_GENERIC1_6			(0x00000340)
+#define MMSS_DP_GENERIC1_7			(0x00000344)
+#define MMSS_DP_GENERIC1_8			(0x00000348)
+#define MMSS_DP_GENERIC1_9			(0x0000034C)
+
+#define MMSS_DP_VSCEXT_0			(0x000002D0)
+#define MMSS_DP_VSCEXT_1			(0x000002D4)
+#define MMSS_DP_VSCEXT_2			(0x000002D8)
+#define MMSS_DP_VSCEXT_3			(0x000002DC)
+#define MMSS_DP_VSCEXT_4			(0x000002E0)
+#define MMSS_DP_VSCEXT_5			(0x000002E4)
+#define MMSS_DP_VSCEXT_6			(0x000002E8)
+#define MMSS_DP_VSCEXT_7			(0x000002EC)
+#define MMSS_DP_VSCEXT_8			(0x000002F0)
+#define MMSS_DP_VSCEXT_9			(0x000002F4)
+
+#define MMSS_DP_BIST_ENABLE			(0x00000000)
+#define DP_BIST_ENABLE_DPBIST_EN		(0x00000001)
+
+#define MMSS_DP_TIMING_ENGINE_EN		(0x00000010)
+#define DP_TIMING_ENGINE_EN_EN			(0x00000001)
+
+#define MMSS_DP_INTF_CONFIG			(0x00000014)
+#define MMSS_DP_INTF_HSYNC_CTL			(0x00000018)
+#define MMSS_DP_INTF_VSYNC_PERIOD_F0		(0x0000001C)
+#define MMSS_DP_INTF_VSYNC_PERIOD_F1		(0x00000020)
+#define MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0	(0x00000024)
+#define MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1	(0x00000028)
+#define MMSS_INTF_DISPLAY_V_START_F0		(0x0000002C)
+#define MMSS_INTF_DISPLAY_V_START_F1		(0x00000030)
+#define MMSS_DP_INTF_DISPLAY_V_END_F0		(0x00000034)
+#define MMSS_DP_INTF_DISPLAY_V_END_F1		(0x00000038)
+#define MMSS_DP_INTF_ACTIVE_V_START_F0		(0x0000003C)
+#define MMSS_DP_INTF_ACTIVE_V_START_F1		(0x00000040)
+#define MMSS_DP_INTF_ACTIVE_V_END_F0		(0x00000044)
+#define MMSS_DP_INTF_ACTIVE_V_END_F1		(0x00000048)
+#define MMSS_DP_INTF_DISPLAY_HCTL		(0x0000004C)
+#define MMSS_DP_INTF_ACTIVE_HCTL		(0x00000050)
+#define MMSS_DP_INTF_POLARITY_CTL		(0x00000058)
+
+#define MMSS_DP_TPG_MAIN_CONTROL		(0x00000060)
+#define MMSS_DP_DSC_DTO				(0x0000007C)
+#define DP_TPG_CHECKERED_RECT_PATTERN		(0x00000100)
+
+#define MMSS_DP_TPG_VIDEO_CONFIG		(0x00000064)
+#define DP_TPG_VIDEO_CONFIG_BPP_8BIT		(0x00000001)
+#define DP_TPG_VIDEO_CONFIG_RGB			(0x00000004)
+
+#define MMSS_DP_ASYNC_FIFO_CONFIG		(0x00000088)
+
+#define REG_DP_PHY_AUX_INTERRUPT_CLEAR          (0x0000004C)
+#define REG_DP_PHY_AUX_BIST_CFG			(0x00000050)
+#define REG_DP_PHY_AUX_INTERRUPT_STATUS         (0x000000BC)
+
+/* DP HDCP 1.3 registers */
+#define DP_HDCP_CTRL                                   (0x0A0)
+#define DP_HDCP_STATUS                                 (0x0A4)
+#define DP_HDCP_SW_UPPER_AKSV                          (0x098)
+#define DP_HDCP_SW_LOWER_AKSV                          (0x09C)
+#define DP_HDCP_ENTROPY_CTRL0                          (0x350)
+#define DP_HDCP_ENTROPY_CTRL1                          (0x35C)
+#define DP_HDCP_SHA_STATUS                             (0x0C8)
+#define DP_HDCP_RCVPORT_DATA2_0                        (0x0B0)
+#define DP_HDCP_RCVPORT_DATA3                          (0x0A4)
+#define DP_HDCP_RCVPORT_DATA4                          (0x0A8)
+#define DP_HDCP_RCVPORT_DATA5                          (0x0C0)
+#define DP_HDCP_RCVPORT_DATA6                          (0x0C4)
+
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_SHA_CTRL           (0x024)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_SHA_DATA           (0x028)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA0      (0x004)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA1      (0x008)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA7      (0x00C)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA8      (0x010)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA9      (0x014)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA10     (0x018)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA11     (0x01C)
+#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA12     (0x020)
+
+#endif /* _DP_REG_H_ */
diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h
index 4de771d6f0be..78ef5d4ed922 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.h
@@ -30,6 +30,8 @@ enum msm_dsi_phy_type {
 	MSM_DSI_PHY_28NM_8960,
 	MSM_DSI_PHY_14NM,
 	MSM_DSI_PHY_10NM,
+	MSM_DSI_PHY_7NM,
+	MSM_DSI_PHY_7NM_V4_1,
 	MSM_DSI_PHY_MAX
 };
 
diff --git a/drivers/gpu/drm/msm/dsi/dsi.xml.h b/drivers/gpu/drm/msm/dsi/dsi.xml.h
index 8e536e060070..50eb4d1b8fdd 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.xml.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.xml.h
@@ -1886,5 +1886,428 @@ static inline uint32_t REG_DSI_10nm_PHY_LN_TX_DCTRL(uint32_t i0) { return 0x0000
 
 #define REG_DSI_10nm_PHY_PLL_COMMON_STATUS_ONE			0x000001a0
 
+#define REG_DSI_7nm_PHY_CMN_REVISION_ID0			0x00000000
+
+#define REG_DSI_7nm_PHY_CMN_REVISION_ID1			0x00000004
+
+#define REG_DSI_7nm_PHY_CMN_REVISION_ID2			0x00000008
+
+#define REG_DSI_7nm_PHY_CMN_REVISION_ID3			0x0000000c
+
+#define REG_DSI_7nm_PHY_CMN_CLK_CFG0				0x00000010
+
+#define REG_DSI_7nm_PHY_CMN_CLK_CFG1				0x00000014
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_CTRL				0x00000018
+
+#define REG_DSI_7nm_PHY_CMN_RBUF_CTRL				0x0000001c
+
+#define REG_DSI_7nm_PHY_CMN_VREG_CTRL_0				0x00000020
+
+#define REG_DSI_7nm_PHY_CMN_CTRL_0				0x00000024
+
+#define REG_DSI_7nm_PHY_CMN_CTRL_1				0x00000028
+
+#define REG_DSI_7nm_PHY_CMN_CTRL_2				0x0000002c
+
+#define REG_DSI_7nm_PHY_CMN_CTRL_3				0x00000030
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CFG0				0x00000034
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CFG1				0x00000038
+
+#define REG_DSI_7nm_PHY_CMN_PLL_CNTRL				0x0000003c
+
+#define REG_DSI_7nm_PHY_CMN_DPHY_SOT				0x00000040
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CTRL0				0x000000a0
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CTRL1				0x000000a4
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CTRL2				0x000000a8
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CTRL3				0x000000ac
+
+#define REG_DSI_7nm_PHY_CMN_LANE_CTRL4				0x000000b0
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_0			0x000000b4
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_1			0x000000b8
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_2			0x000000bc
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_3			0x000000c0
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_4			0x000000c4
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_5			0x000000c8
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_6			0x000000cc
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_7			0x000000d0
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_8			0x000000d4
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_9			0x000000d8
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_10			0x000000dc
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_11			0x000000e0
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_12			0x000000e4
+
+#define REG_DSI_7nm_PHY_CMN_TIMING_CTRL_13			0x000000e8
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_HSTX_STR_CTRL_0		0x000000ec
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_HSTX_STR_CTRL_1		0x000000f0
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_RESCODE_OFFSET_TOP_CTRL	0x000000f4
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_RESCODE_OFFSET_BOT_CTRL	0x000000f8
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_RESCODE_OFFSET_MID_CTRL	0x000000fc
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_LPTX_STR_CTRL			0x00000100
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_PEMPH_CTRL_0			0x00000104
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_PEMPH_CTRL_1			0x00000108
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_STR_SWI_CAL_SEL_CTRL		0x0000010c
+
+#define REG_DSI_7nm_PHY_CMN_VREG_CTRL_1				0x00000110
+
+#define REG_DSI_7nm_PHY_CMN_CTRL_4				0x00000114
+
+#define REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE4			0x00000128
+
+#define REG_DSI_7nm_PHY_CMN_PHY_STATUS				0x00000140
+
+#define REG_DSI_7nm_PHY_CMN_LANE_STATUS0			0x00000148
+
+#define REG_DSI_7nm_PHY_CMN_LANE_STATUS1			0x0000014c
+
+static inline uint32_t REG_DSI_7nm_PHY_LN(uint32_t i0) { return 0x00000000 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_CFG0(uint32_t i0) { return 0x00000000 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_CFG1(uint32_t i0) { return 0x00000004 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_CFG2(uint32_t i0) { return 0x00000008 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_TEST_DATAPATH(uint32_t i0) { return 0x0000000c + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_PIN_SWAP(uint32_t i0) { return 0x00000010 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_LPRX_CTRL(uint32_t i0) { return 0x00000014 + 0x80*i0; }
+
+static inline uint32_t REG_DSI_7nm_PHY_LN_TX_DCTRL(uint32_t i0) { return 0x00000018 + 0x80*i0; }
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_ONE			0x00000000
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_TWO			0x00000004
+
+#define REG_DSI_7nm_PHY_PLL_INT_LOOP_SETTINGS			0x00000008
+
+#define REG_DSI_7nm_PHY_PLL_INT_LOOP_SETTINGS_TWO		0x0000000c
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_THREE		0x00000010
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FOUR		0x00000014
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FIVE		0x00000018
+
+#define REG_DSI_7nm_PHY_PLL_INT_LOOP_CONTROLS			0x0000001c
+
+#define REG_DSI_7nm_PHY_PLL_DSM_DIVIDER				0x00000020
+
+#define REG_DSI_7nm_PHY_PLL_FEEDBACK_DIVIDER			0x00000024
+
+#define REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES			0x00000028
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_UPDATE_CONTROL_OVERRIDES	0x0000002c
+
+#define REG_DSI_7nm_PHY_PLL_CMODE				0x00000030
+
+#define REG_DSI_7nm_PHY_PLL_PSM_CTRL				0x00000034
+
+#define REG_DSI_7nm_PHY_PLL_RSM_CTRL				0x00000038
+
+#define REG_DSI_7nm_PHY_PLL_VCO_TUNE_MAP			0x0000003c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_CNTRL				0x00000040
+
+#define REG_DSI_7nm_PHY_PLL_CALIBRATION_SETTINGS		0x00000044
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_TIMER_LOW		0x00000048
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_TIMER_HIGH		0x0000004c
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_SETTINGS		0x00000050
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_MIN			0x00000054
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_MAX			0x00000058
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_PFILT			0x0000005c
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_IFILT			0x00000060
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_TWO		0x00000064
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_THREE		0x00000068
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_FOUR		0x0000006c
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_ICODE_HIGH			0x00000070
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_ICODE_LOW			0x00000074
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DETECT_SETTINGS_ONE		0x00000078
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DETECT_THRESH			0x0000007c
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DET_REFCLK_HIGH		0x00000080
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DET_REFCLK_LOW			0x00000084
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DET_PLLCLK_HIGH		0x00000088
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_DET_PLLCLK_LOW			0x0000008c
+
+#define REG_DSI_7nm_PHY_PLL_PFILT				0x00000090
+
+#define REG_DSI_7nm_PHY_PLL_IFILT				0x00000094
+
+#define REG_DSI_7nm_PHY_PLL_PLL_GAIN				0x00000098
+
+#define REG_DSI_7nm_PHY_PLL_ICODE_LOW				0x0000009c
+
+#define REG_DSI_7nm_PHY_PLL_ICODE_HIGH				0x000000a0
+
+#define REG_DSI_7nm_PHY_PLL_LOCKDET				0x000000a4
+
+#define REG_DSI_7nm_PHY_PLL_OUTDIV				0x000000a8
+
+#define REG_DSI_7nm_PHY_PLL_FASTLOCK_CONTROL			0x000000ac
+
+#define REG_DSI_7nm_PHY_PLL_PASS_OUT_OVERRIDE_ONE		0x000000b0
+
+#define REG_DSI_7nm_PHY_PLL_PASS_OUT_OVERRIDE_TWO		0x000000b4
+
+#define REG_DSI_7nm_PHY_PLL_CORE_OVERRIDE			0x000000b8
+
+#define REG_DSI_7nm_PHY_PLL_CORE_INPUT_OVERRIDE			0x000000bc
+
+#define REG_DSI_7nm_PHY_PLL_RATE_CHANGE				0x000000c0
+
+#define REG_DSI_7nm_PHY_PLL_PLL_DIGITAL_TIMERS			0x000000c4
+
+#define REG_DSI_7nm_PHY_PLL_PLL_DIGITAL_TIMERS_TWO		0x000000c8
+
+#define REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START			0x000000cc
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_LOW			0x000000d0
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_MID			0x000000d4
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_HIGH			0x000000d8
+
+#define REG_DSI_7nm_PHY_PLL_DEC_FRAC_MUXES			0x000000dc
+
+#define REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START_1			0x000000e0
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_LOW_1		0x000000e4
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_MID_1		0x000000e8
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_HIGH_1		0x000000ec
+
+#define REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START_2			0x000000f0
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_LOW_2		0x000000f4
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_MID_2		0x000000f8
+
+#define REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_HIGH_2		0x000000fc
+
+#define REG_DSI_7nm_PHY_PLL_MASH_CONTROL			0x00000100
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_LOW			0x00000104
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_HIGH			0x00000108
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_LOW			0x0000010c
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_HIGH			0x00000110
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_LOW			0x00000114
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_HIGH			0x00000118
+
+#define REG_DSI_7nm_PHY_PLL_SSC_MUX_CONTROL			0x0000011c
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_LOW_1			0x00000120
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_HIGH_1			0x00000124
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_LOW_1			0x00000128
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_HIGH_1			0x0000012c
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_LOW_1			0x00000130
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_HIGH_1			0x00000134
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_LOW_2			0x00000138
+
+#define REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_HIGH_2			0x0000013c
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_LOW_2			0x00000140
+
+#define REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_HIGH_2			0x00000144
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_LOW_2			0x00000148
+
+#define REG_DSI_7nm_PHY_PLL_SSC_ADJPER_HIGH_2			0x0000014c
+
+#define REG_DSI_7nm_PHY_PLL_SSC_CONTROL				0x00000150
+
+#define REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE			0x00000154
+
+#define REG_DSI_7nm_PHY_PLL_PLL_LOCKDET_RATE_1			0x00000158
+
+#define REG_DSI_7nm_PHY_PLL_PLL_LOCKDET_RATE_2			0x0000015c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_PROP_GAIN_RATE_1		0x00000160
+
+#define REG_DSI_7nm_PHY_PLL_PLL_PROP_GAIN_RATE_2		0x00000164
+
+#define REG_DSI_7nm_PHY_PLL_PLL_BAND_SEL_RATE_1			0x00000168
+
+#define REG_DSI_7nm_PHY_PLL_PLL_BAND_SEL_RATE_2			0x0000016c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_1		0x00000170
+
+#define REG_DSI_7nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_2		0x00000174
+
+#define REG_DSI_7nm_PHY_PLL_PLL_FL_INT_GAIN_PFILT_BAND_1	0x00000178
+
+#define REG_DSI_7nm_PHY_PLL_PLL_FL_INT_GAIN_PFILT_BAND_2	0x0000017c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_FASTLOCK_EN_BAND		0x00000180
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_MID		0x00000184
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_HIGH		0x00000188
+
+#define REG_DSI_7nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_MUX		0x0000018c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_LOCK_OVERRIDE			0x00000190
+
+#define REG_DSI_7nm_PHY_PLL_PLL_LOCK_DELAY			0x00000194
+
+#define REG_DSI_7nm_PHY_PLL_PLL_LOCK_MIN_DELAY			0x00000198
+
+#define REG_DSI_7nm_PHY_PLL_CLOCK_INVERTERS			0x0000019c
+
+#define REG_DSI_7nm_PHY_PLL_SPARE_AND_JPC_OVERRIDES		0x000001a0
+
+#define REG_DSI_7nm_PHY_PLL_BIAS_CONTROL_1			0x000001a4
+
+#define REG_DSI_7nm_PHY_PLL_BIAS_CONTROL_2			0x000001a8
+
+#define REG_DSI_7nm_PHY_PLL_ALOG_OBSV_BUS_CTRL_1		0x000001ac
+
+#define REG_DSI_7nm_PHY_PLL_COMMON_STATUS_ONE			0x000001b0
+
+#define REG_DSI_7nm_PHY_PLL_COMMON_STATUS_TWO			0x000001b4
+
+#define REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL			0x000001b8
+
+#define REG_DSI_7nm_PHY_PLL_ICODE_ACCUM_STATUS_LOW		0x000001bc
+
+#define REG_DSI_7nm_PHY_PLL_ICODE_ACCUM_STATUS_HIGH		0x000001c0
+
+#define REG_DSI_7nm_PHY_PLL_FD_OUT_LOW				0x000001c4
+
+#define REG_DSI_7nm_PHY_PLL_FD_OUT_HIGH				0x000001c8
+
+#define REG_DSI_7nm_PHY_PLL_ALOG_OBSV_BUS_STATUS_1		0x000001cc
+
+#define REG_DSI_7nm_PHY_PLL_PLL_MISC_CONFIG			0x000001d0
+
+#define REG_DSI_7nm_PHY_PLL_FLL_CONFIG				0x000001d4
+
+#define REG_DSI_7nm_PHY_PLL_FLL_FREQ_ACQ_TIME			0x000001d8
+
+#define REG_DSI_7nm_PHY_PLL_FLL_CODE0				0x000001dc
+
+#define REG_DSI_7nm_PHY_PLL_FLL_CODE1				0x000001e0
+
+#define REG_DSI_7nm_PHY_PLL_FLL_GAIN0				0x000001e4
+
+#define REG_DSI_7nm_PHY_PLL_FLL_GAIN1				0x000001e8
+
+#define REG_DSI_7nm_PHY_PLL_SW_RESET				0x000001ec
+
+#define REG_DSI_7nm_PHY_PLL_FAST_PWRUP				0x000001f0
+
+#define REG_DSI_7nm_PHY_PLL_LOCKTIME0				0x000001f4
+
+#define REG_DSI_7nm_PHY_PLL_LOCKTIME1				0x000001f8
+
+#define REG_DSI_7nm_PHY_PLL_DEBUG_BUS_SEL			0x000001fc
+
+#define REG_DSI_7nm_PHY_PLL_DEBUG_BUS0				0x00000200
+
+#define REG_DSI_7nm_PHY_PLL_DEBUG_BUS1				0x00000204
+
+#define REG_DSI_7nm_PHY_PLL_DEBUG_BUS2				0x00000208
+
+#define REG_DSI_7nm_PHY_PLL_DEBUG_BUS3				0x0000020c
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_FLL_CONTROL_OVERRIDES	0x00000210
+
+#define REG_DSI_7nm_PHY_PLL_VCO_CONFIG				0x00000214
+
+#define REG_DSI_7nm_PHY_PLL_VCO_CAL_CODE1_MODE0_STATUS		0x00000218
+
+#define REG_DSI_7nm_PHY_PLL_VCO_CAL_CODE1_MODE1_STATUS		0x0000021c
+
+#define REG_DSI_7nm_PHY_PLL_RESET_SM_STATUS			0x00000220
+
+#define REG_DSI_7nm_PHY_PLL_TDC_OFFSET				0x00000224
+
+#define REG_DSI_7nm_PHY_PLL_PS3_PWRDOWN_CONTROLS		0x00000228
+
+#define REG_DSI_7nm_PHY_PLL_PS4_PWRDOWN_CONTROLS		0x0000022c
+
+#define REG_DSI_7nm_PHY_PLL_PLL_RST_CONTROLS			0x00000230
+
+#define REG_DSI_7nm_PHY_PLL_GEAR_BAND_SELECT_CONTROLS		0x00000234
+
+#define REG_DSI_7nm_PHY_PLL_PSM_CLK_CONTROLS			0x00000238
+
+#define REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES_2			0x0000023c
+
+#define REG_DSI_7nm_PHY_PLL_VCO_CONFIG_1			0x00000240
+
+#define REG_DSI_7nm_PHY_PLL_VCO_CONFIG_2			0x00000244
+
+#define REG_DSI_7nm_PHY_PLL_CLOCK_INVERTERS_1			0x00000248
+
+#define REG_DSI_7nm_PHY_PLL_CLOCK_INVERTERS_2			0x0000024c
+
+#define REG_DSI_7nm_PHY_PLL_CMODE_1				0x00000250
+
+#define REG_DSI_7nm_PHY_PLL_CMODE_2				0x00000254
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FIVE_1		0x00000258
+
+#define REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FIVE_2		0x0000025c
+
+#define REG_DSI_7nm_PHY_PLL_PERF_OPTIMIZE			0x00000260
 
 #endif /* DSI_XML */
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
index f892f2cbe8bb..b2ff68a15791 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
@@ -265,9 +265,12 @@ static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = {
 		&msm8998_dsi_cfg, &msm_dsi_6g_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_2_1,
 		&sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops},
+	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_3_0,
+		&sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops},
+	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_4_0,
+		&sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_4_1,
 		&sc7180_dsi_cfg, &msm_dsi_6g_v2_host_ops},
-
 };
 
 const struct msm_dsi_cfg_handler *msm_dsi_cfg_get(u32 major, u32 minor)
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
index efd469d1db45..ade9b609c7d9 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
@@ -21,6 +21,8 @@
 #define MSM_DSI_6G_VER_MINOR_V2_1_0	0x20010000
 #define MSM_DSI_6G_VER_MINOR_V2_2_0	0x20000000
 #define MSM_DSI_6G_VER_MINOR_V2_2_1	0x20020001
+#define MSM_DSI_6G_VER_MINOR_V2_3_0	0x20030000
+#define MSM_DSI_6G_VER_MINOR_V2_4_0	0x20040000
 #define MSM_DSI_6G_VER_MINOR_V2_4_1	0x20040001
 
 #define MSM_DSI_V2_VER_MINOR_8064	0x0
diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 4b363bd7ddff..1d28dfba2c9b 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -328,7 +328,6 @@ static int dsi_mgr_connector_get_modes(struct drm_connector *connector)
 	 * In dual DSI mode, we have one connector that can be
 	 * attached to the drm_panel.
 	 */
-	drm_panel_attach(panel, connector);
 	num = drm_panel_get_modes(panel, connector);
 	if (!num)
 		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 009f5b843dd1..e8c1a727179c 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -364,6 +364,102 @@ int msm_dsi_dphy_timing_calc_v3(struct msm_dsi_dphy_timing *timing,
 	return 0;
 }
 
+int msm_dsi_dphy_timing_calc_v4(struct msm_dsi_dphy_timing *timing,
+	struct msm_dsi_phy_clk_request *clk_req)
+{
+	const unsigned long bit_rate = clk_req->bitclk_rate;
+	const unsigned long esc_rate = clk_req->escclk_rate;
+	s32 ui, ui_x8;
+	s32 tmax, tmin;
+	s32 pcnt_clk_prep = 50;
+	s32 pcnt_clk_zero = 2;
+	s32 pcnt_clk_trail = 30;
+	s32 pcnt_hs_prep = 50;
+	s32 pcnt_hs_zero = 10;
+	s32 pcnt_hs_trail = 30;
+	s32 pcnt_hs_exit = 10;
+	s32 coeff = 1000; /* Precision, should avoid overflow */
+	s32 hb_en;
+	s32 temp;
+
+	if (!bit_rate || !esc_rate)
+		return -EINVAL;
+
+	hb_en = 0;
+
+	ui = mult_frac(NSEC_PER_MSEC, coeff, bit_rate / 1000);
+	ui_x8 = ui << 3;
+
+	/* TODO: verify these calculations against latest downstream driver
+	 * everything except clk_post/clk_pre uses calculations from v3 based
+	 * on the downstream driver having the same calculations for v3 and v4
+	 */
+
+	temp = S_DIV_ROUND_UP(38 * coeff, ui_x8);
+	tmin = max_t(s32, temp, 0);
+	temp = (95 * coeff) / ui_x8;
+	tmax = max_t(s32, temp, 0);
+	timing->clk_prepare = linear_inter(tmax, tmin, pcnt_clk_prep, 0, false);
+
+	temp = 300 * coeff - (timing->clk_prepare << 3) * ui;
+	tmin = S_DIV_ROUND_UP(temp, ui_x8) - 1;
+	tmax = (tmin > 255) ? 511 : 255;
+	timing->clk_zero = linear_inter(tmax, tmin, pcnt_clk_zero, 0, false);
+
+	tmin = DIV_ROUND_UP(60 * coeff + 3 * ui, ui_x8);
+	temp = 105 * coeff + 12 * ui - 20 * coeff;
+	tmax = (temp + 3 * ui) / ui_x8;
+	timing->clk_trail = linear_inter(tmax, tmin, pcnt_clk_trail, 0, false);
+
+	temp = S_DIV_ROUND_UP(40 * coeff + 4 * ui, ui_x8);
+	tmin = max_t(s32, temp, 0);
+	temp = (85 * coeff + 6 * ui) / ui_x8;
+	tmax = max_t(s32, temp, 0);
+	timing->hs_prepare = linear_inter(tmax, tmin, pcnt_hs_prep, 0, false);
+
+	temp = 145 * coeff + 10 * ui - (timing->hs_prepare << 3) * ui;
+	tmin = S_DIV_ROUND_UP(temp, ui_x8) - 1;
+	tmax = 255;
+	timing->hs_zero = linear_inter(tmax, tmin, pcnt_hs_zero, 0, false);
+
+	tmin = DIV_ROUND_UP(60 * coeff + 4 * ui, ui_x8) - 1;
+	temp = 105 * coeff + 12 * ui - 20 * coeff;
+	tmax = (temp / ui_x8) - 1;
+	timing->hs_trail = linear_inter(tmax, tmin, pcnt_hs_trail, 0, false);
+
+	temp = 50 * coeff + ((hb_en << 2) - 8) * ui;
+	timing->hs_rqst = S_DIV_ROUND_UP(temp, ui_x8);
+
+	tmin = DIV_ROUND_UP(100 * coeff, ui_x8) - 1;
+	tmax = 255;
+	timing->hs_exit = linear_inter(tmax, tmin, pcnt_hs_exit, 0, false);
+
+	/* recommended min
+	 * = roundup((mipi_min_ns + t_hs_trail_ns)/(16*bit_clk_ns), 0) - 1
+	 */
+	temp = 60 * coeff + 52 * ui + + (timing->hs_trail + 1) * ui_x8;
+	tmin = DIV_ROUND_UP(temp, 16 * ui) - 1;
+	tmax = 255;
+	timing->shared_timings.clk_post = linear_inter(tmax, tmin, 5, 0, false);
+
+	/* recommended min
+	 * val1 = (tlpx_ns + clk_prepare_ns + clk_zero_ns + hs_rqst_ns)
+	 * val2 = (16 * bit_clk_ns)
+	 * final = roundup(val1/val2, 0) - 1
+	 */
+	temp = 52 * coeff + (timing->clk_prepare + timing->clk_zero + 1) * ui_x8 + 54 * coeff;
+	tmin = DIV_ROUND_UP(temp, 16 * ui) - 1;
+	tmax = 255;
+	timing->shared_timings.clk_pre = DIV_ROUND_UP((tmax - tmin) * 125, 10000) + tmin;
+
+	DBG("%d, %d, %d, %d, %d, %d, %d, %d, %d, %d",
+		timing->shared_timings.clk_pre, timing->shared_timings.clk_post,
+		timing->clk_zero, timing->clk_trail, timing->clk_prepare, timing->hs_exit,
+		timing->hs_zero, timing->hs_prepare, timing->hs_trail, timing->hs_rqst);
+
+	return 0;
+}
+
 void msm_dsi_phy_set_src_pll(struct msm_dsi_phy *phy, int pll_id, u32 reg,
 				u32 bit_mask)
 {
@@ -508,6 +604,12 @@ static const struct of_device_id dsi_phy_dt_match[] = {
 	{ .compatible = "qcom,dsi-phy-10nm-8998",
 	  .data = &dsi_phy_10nm_8998_cfgs },
 #endif
+#ifdef CONFIG_DRM_MSM_DSI_7NM_PHY
+	{ .compatible = "qcom,dsi-phy-7nm",
+	  .data = &dsi_phy_7nm_cfgs },
+	{ .compatible = "qcom,dsi-phy-7nm-8150",
+	  .data = &dsi_phy_7nm_8150_cfgs },
+#endif
 	{}
 };
 
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
index ef8672d7b123..d2bd74b6f357 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
@@ -48,10 +48,10 @@ extern const struct msm_dsi_phy_cfg dsi_phy_14nm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_14nm_660_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_10nm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_10nm_8998_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_7nm_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_7nm_8150_cfgs;
 
 struct msm_dsi_dphy_timing {
-	u32 clk_pre;
-	u32 clk_post;
 	u32 clk_zero;
 	u32 clk_trail;
 	u32 clk_prepare;
@@ -102,6 +102,8 @@ int msm_dsi_dphy_timing_calc_v2(struct msm_dsi_dphy_timing *timing,
 				struct msm_dsi_phy_clk_request *clk_req);
 int msm_dsi_dphy_timing_calc_v3(struct msm_dsi_dphy_timing *timing,
 				struct msm_dsi_phy_clk_request *clk_req);
+int msm_dsi_dphy_timing_calc_v4(struct msm_dsi_dphy_timing *timing,
+				struct msm_dsi_phy_clk_request *clk_req);
 void msm_dsi_phy_set_src_pll(struct msm_dsi_phy *phy, int pll_id, u32 reg,
 				u32 bit_mask);
 int msm_dsi_phy_init_common(struct msm_dsi_phy *phy);
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
new file mode 100644
index 000000000000..255b5f5ab2ce
--- /dev/null
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -0,0 +1,255 @@
+/*
+ * SPDX-License-Identifier: GPL-2.0
+ * Copyright (c) 2018, The Linux Foundation
+ */
+
+#include <linux/iopoll.h>
+
+#include "dsi_phy.h"
+#include "dsi.xml.h"
+
+static int dsi_phy_hw_v4_0_is_pll_on(struct msm_dsi_phy *phy)
+{
+	void __iomem *base = phy->base;
+	u32 data = 0;
+
+	data = dsi_phy_read(base + REG_DSI_7nm_PHY_CMN_PLL_CNTRL);
+	mb(); /* make sure read happened */
+
+	return (data & BIT(0));
+}
+
+static void dsi_phy_hw_v4_0_config_lpcdrx(struct msm_dsi_phy *phy, bool enable)
+{
+	void __iomem *lane_base = phy->lane_base;
+	int phy_lane_0 = 0;	/* TODO: Support all lane swap configs */
+
+	/*
+	 * LPRX and CDRX need to enabled only for physical data lane
+	 * corresponding to the logical data lane 0
+	 */
+	if (enable)
+		dsi_phy_write(lane_base +
+			      REG_DSI_7nm_PHY_LN_LPRX_CTRL(phy_lane_0), 0x3);
+	else
+		dsi_phy_write(lane_base +
+			      REG_DSI_7nm_PHY_LN_LPRX_CTRL(phy_lane_0), 0);
+}
+
+static void dsi_phy_hw_v4_0_lane_settings(struct msm_dsi_phy *phy)
+{
+	int i;
+	const u8 tx_dctrl_0[] = { 0x00, 0x00, 0x00, 0x04, 0x01 };
+	const u8 tx_dctrl_1[] = { 0x40, 0x40, 0x40, 0x46, 0x41 };
+	const u8 *tx_dctrl = tx_dctrl_0;
+	void __iomem *lane_base = phy->lane_base;
+
+	if (phy->cfg->type == MSM_DSI_PHY_7NM_V4_1)
+		tx_dctrl = tx_dctrl_1;
+
+	/* Strength ctrl settings */
+	for (i = 0; i < 5; i++) {
+		/*
+		 * Disable LPRX and CDRX for all lanes. And later on, it will
+		 * be only enabled for the physical data lane corresponding
+		 * to the logical data lane 0
+		 */
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_LPRX_CTRL(i), 0);
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_PIN_SWAP(i), 0x0);
+	}
+
+	dsi_phy_hw_v4_0_config_lpcdrx(phy, true);
+
+	/* other settings */
+	for (i = 0; i < 5; i++) {
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_CFG0(i), 0x0);
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_CFG1(i), 0x0);
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_CFG2(i), i == 4 ? 0x8a : 0xa);
+		dsi_phy_write(lane_base + REG_DSI_7nm_PHY_LN_TX_DCTRL(i), tx_dctrl[i]);
+	}
+}
+
+static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id,
+			      struct msm_dsi_phy_clk_request *clk_req)
+{
+	int ret;
+	u32 status;
+	u32 const delay_us = 5;
+	u32 const timeout_us = 1000;
+	struct msm_dsi_dphy_timing *timing = &phy->timing;
+	void __iomem *base = phy->base;
+	bool less_than_1500_mhz;
+	u32 vreg_ctrl_0, glbl_str_swi_cal_sel_ctrl, glbl_hstx_str_ctrl_0;
+	u32 glbl_rescode_top_ctrl, glbl_rescode_bot_ctrl;
+	u32 data;
+
+	DBG("");
+
+	if (msm_dsi_dphy_timing_calc_v4(timing, clk_req)) {
+		DRM_DEV_ERROR(&phy->pdev->dev,
+			"%s: D-PHY timing calculation failed\n", __func__);
+		return -EINVAL;
+	}
+
+	if (dsi_phy_hw_v4_0_is_pll_on(phy))
+		pr_warn("PLL turned on before configuring PHY\n");
+
+	/* wait for REFGEN READY */
+	ret = readl_poll_timeout_atomic(base + REG_DSI_7nm_PHY_CMN_PHY_STATUS,
+					status, (status & BIT(0)),
+					delay_us, timeout_us);
+	if (ret) {
+		pr_err("Ref gen not ready. Aborting\n");
+		return -EINVAL;
+	}
+
+	/* TODO: CPHY enable path (this is for DPHY only) */
+
+	/* Alter PHY configurations if data rate less than 1.5GHZ*/
+	less_than_1500_mhz = (clk_req->bitclk_rate <= 1500000000);
+
+	if (phy->cfg->type == MSM_DSI_PHY_7NM_V4_1) {
+		vreg_ctrl_0 = less_than_1500_mhz ? 0x53 : 0x52;
+		glbl_rescode_top_ctrl = less_than_1500_mhz ? 0x3d :  0x00;
+		glbl_rescode_bot_ctrl = less_than_1500_mhz ? 0x39 :  0x3c;
+		glbl_str_swi_cal_sel_ctrl = 0x00;
+		glbl_hstx_str_ctrl_0 = 0x88;
+	} else {
+		vreg_ctrl_0 = less_than_1500_mhz ? 0x5B : 0x59;
+		glbl_str_swi_cal_sel_ctrl = less_than_1500_mhz ? 0x03 : 0x00;
+		glbl_hstx_str_ctrl_0 = less_than_1500_mhz ? 0x66 : 0x88;
+		glbl_rescode_top_ctrl = 0x03;
+		glbl_rescode_bot_ctrl = 0x3c;
+	}
+
+	/* de-assert digital and pll power down */
+	data = BIT(6) | BIT(5);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_CTRL_0, data);
+
+	/* Assert PLL core reset */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_PLL_CNTRL, 0x00);
+
+	/* turn off resync FIFO */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_RBUF_CTRL, 0x00);
+
+	/* program CMN_CTRL_4 for minor_ver 2 chipsets*/
+	data = dsi_phy_read(base + REG_DSI_7nm_PHY_CMN_REVISION_ID0);
+	data = data & (0xf0);
+	if (data == 0x20)
+		dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_CTRL_4, 0x04);
+
+	/* Configure PHY lane swap (TODO: we need to calculate this) */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_LANE_CFG0, 0x21);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_LANE_CFG1, 0x84);
+
+	/* Enable LDO */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_VREG_CTRL_0, vreg_ctrl_0);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_VREG_CTRL_1, 0x5c);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_CTRL_3, 0x00);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_STR_SWI_CAL_SEL_CTRL,
+		      glbl_str_swi_cal_sel_ctrl);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_HSTX_STR_CTRL_0,
+		      glbl_hstx_str_ctrl_0);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_PEMPH_CTRL_0, 0x00);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_RESCODE_OFFSET_TOP_CTRL,
+		      glbl_rescode_top_ctrl);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_RESCODE_OFFSET_BOT_CTRL,
+		      glbl_rescode_bot_ctrl);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_GLBL_LPTX_STR_CTRL, 0x55);
+
+	/* Remove power down from all blocks */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_CTRL_0, 0x7f);
+
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_LANE_CTRL0, 0x1f);
+
+	/* Select full-rate mode */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_CTRL_2, 0x40);
+
+	ret = msm_dsi_pll_set_usecase(phy->pll, phy->usecase);
+	if (ret) {
+		DRM_DEV_ERROR(&phy->pdev->dev, "%s: set pll usecase failed, %d\n",
+			__func__, ret);
+		return ret;
+	}
+
+	/* DSI PHY timings */
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_0, 0x00);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_1, timing->clk_zero);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_2, timing->clk_prepare);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_3, timing->clk_trail);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_4, timing->hs_exit);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_5, timing->hs_zero);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_6, timing->hs_prepare);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_7, timing->hs_trail);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_8, timing->hs_rqst);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_9, 0x02);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_10, 0x04);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_11, 0x00);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_12,
+		      timing->shared_timings.clk_pre);
+	dsi_phy_write(base + REG_DSI_7nm_PHY_CMN_TIMING_CTRL_13,
+		      timing->shared_timings.clk_post);
+
+	/* DSI lane settings */
+	dsi_phy_hw_v4_0_lane_settings(phy);
+
+	DBG("DSI%d PHY enabled", phy->id);
+
+	return 0;
+}
+
+static void dsi_7nm_phy_disable(struct msm_dsi_phy *phy)
+{
+	/* TODO */
+}
+
+static int dsi_7nm_phy_init(struct msm_dsi_phy *phy)
+{
+	struct platform_device *pdev = phy->pdev;
+
+	phy->lane_base = msm_ioremap(pdev, "dsi_phy_lane",
+				     "DSI_PHY_LANE");
+	if (IS_ERR(phy->lane_base)) {
+		DRM_DEV_ERROR(&pdev->dev, "%s: failed to map phy lane base\n",
+			__func__);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+const struct msm_dsi_phy_cfg dsi_phy_7nm_cfgs = {
+	.type = MSM_DSI_PHY_7NM_V4_1,
+	.src_pll_truthtable = { {false, false}, {true, false} },
+	.reg_cfg = {
+		.num = 1,
+		.regs = {
+			{"vdds", 36000, 32},
+		},
+	},
+	.ops = {
+		.enable = dsi_7nm_phy_enable,
+		.disable = dsi_7nm_phy_disable,
+		.init = dsi_7nm_phy_init,
+	},
+	.io_start = { 0xae94400, 0xae96400 },
+	.num_dsi_phy = 2,
+};
+
+const struct msm_dsi_phy_cfg dsi_phy_7nm_8150_cfgs = {
+	.type = MSM_DSI_PHY_7NM,
+	.src_pll_truthtable = { {false, false}, {true, false} },
+	.reg_cfg = {
+		.num = 1,
+		.regs = {
+			{"vdds", 36000, 32},
+		},
+	},
+	.ops = {
+		.enable = dsi_7nm_phy_enable,
+		.disable = dsi_7nm_phy_disable,
+		.init = dsi_7nm_phy_init,
+	},
+	.io_start = { 0xae94400, 0xae96400 },
+	.num_dsi_phy = 2,
+};
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
index 4a4aa3c61d71..a45fe95aff49 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
@@ -161,6 +161,10 @@ struct msm_dsi_pll *msm_dsi_pll_init(struct platform_device *pdev,
 	case MSM_DSI_PHY_10NM:
 		pll = msm_dsi_pll_10nm_init(pdev, id);
 		break;
+	case MSM_DSI_PHY_7NM:
+	case MSM_DSI_PHY_7NM_V4_1:
+		pll = msm_dsi_pll_7nm_init(pdev, id);
+		break;
 	default:
 		pll = ERR_PTR(-ENXIO);
 		break;
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
index c6a3623f905d..3405982a092c 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
@@ -116,5 +116,15 @@ msm_dsi_pll_10nm_init(struct platform_device *pdev, int id)
 	return ERR_PTR(-ENODEV);
 }
 #endif
+#ifdef CONFIG_DRM_MSM_DSI_7NM_PHY
+struct msm_dsi_pll *msm_dsi_pll_7nm_init(struct platform_device *pdev, int id);
+#else
+static inline struct msm_dsi_pll *
+msm_dsi_pll_7nm_init(struct platform_device *pdev, int id)
+{
+	return ERR_PTR(-ENODEV);
+}
+#endif
+
 #endif /* __DSI_PLL_H__ */
 
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_7nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_7nm.c
new file mode 100644
index 000000000000..de0dfb815125
--- /dev/null
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_7nm.c
@@ -0,0 +1,904 @@
+/*
+ * SPDX-License-Identifier: GPL-2.0
+ * Copyright (c) 2018, The Linux Foundation
+ */
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/iopoll.h>
+
+#include "dsi_pll.h"
+#include "dsi.xml.h"
+
+/*
+ * DSI PLL 7nm - clock diagram (eg: DSI0): TODO: updated CPHY diagram
+ *
+ *           dsi0_pll_out_div_clk  dsi0_pll_bit_clk
+ *                              |                |
+ *                              |                |
+ *                 +---------+  |  +----------+  |  +----+
+ *  dsi0vco_clk ---| out_div |--o--| divl_3_0 |--o--| /8 |-- dsi0_phy_pll_out_byteclk
+ *                 +---------+  |  +----------+  |  +----+
+ *                              |                |
+ *                              |                |         dsi0_pll_by_2_bit_clk
+ *                              |                |          |
+ *                              |                |  +----+  |  |\  dsi0_pclk_mux
+ *                              |                |--| /2 |--o--| \   |
+ *                              |                |  +----+     |  \  |  +---------+
+ *                              |                --------------|  |--o--| div_7_4 |-- dsi0_phy_pll_out_dsiclk
+ *                              |------------------------------|  /     +---------+
+ *                              |          +-----+             | /
+ *                              -----------| /4? |--o----------|/
+ *                                         +-----+  |           |
+ *                                                  |           |dsiclk_sel
+ *                                                  |
+ *                                                  dsi0_pll_post_out_div_clk
+ */
+
+#define DSI_BYTE_PLL_CLK		0
+#define DSI_PIXEL_PLL_CLK		1
+#define NUM_PROVIDED_CLKS		2
+
+#define VCO_REF_CLK_RATE		19200000
+
+struct dsi_pll_regs {
+	u32 pll_prop_gain_rate;
+	u32 pll_lockdet_rate;
+	u32 decimal_div_start;
+	u32 frac_div_start_low;
+	u32 frac_div_start_mid;
+	u32 frac_div_start_high;
+	u32 pll_clock_inverters;
+	u32 ssc_stepsize_low;
+	u32 ssc_stepsize_high;
+	u32 ssc_div_per_low;
+	u32 ssc_div_per_high;
+	u32 ssc_adjper_low;
+	u32 ssc_adjper_high;
+	u32 ssc_control;
+};
+
+struct dsi_pll_config {
+	u32 ref_freq;
+	bool div_override;
+	u32 output_div;
+	bool ignore_frac;
+	bool disable_prescaler;
+	bool enable_ssc;
+	bool ssc_center;
+	u32 dec_bits;
+	u32 frac_bits;
+	u32 lock_timer;
+	u32 ssc_freq;
+	u32 ssc_offset;
+	u32 ssc_adj_per;
+	u32 thresh_cycles;
+	u32 refclk_cycles;
+};
+
+struct pll_7nm_cached_state {
+	unsigned long vco_rate;
+	u8 bit_clk_div;
+	u8 pix_clk_div;
+	u8 pll_out_div;
+	u8 pll_mux;
+};
+
+struct dsi_pll_7nm {
+	struct msm_dsi_pll base;
+
+	int id;
+	struct platform_device *pdev;
+
+	void __iomem *phy_cmn_mmio;
+	void __iomem *mmio;
+
+	u64 vco_ref_clk_rate;
+	u64 vco_current_rate;
+
+	/* protects REG_DSI_7nm_PHY_CMN_CLK_CFG0 register */
+	spinlock_t postdiv_lock;
+
+	int vco_delay;
+	struct dsi_pll_config pll_configuration;
+	struct dsi_pll_regs reg_setup;
+
+	/* private clocks: */
+	struct clk_hw *out_div_clk_hw;
+	struct clk_hw *bit_clk_hw;
+	struct clk_hw *byte_clk_hw;
+	struct clk_hw *by_2_bit_clk_hw;
+	struct clk_hw *post_out_div_clk_hw;
+	struct clk_hw *pclk_mux_hw;
+	struct clk_hw *out_dsiclk_hw;
+
+	/* clock-provider: */
+	struct clk_hw_onecell_data *hw_data;
+
+	struct pll_7nm_cached_state cached_state;
+
+	enum msm_dsi_phy_usecase uc;
+	struct dsi_pll_7nm *slave;
+};
+
+#define to_pll_7nm(x)	container_of(x, struct dsi_pll_7nm, base)
+
+/*
+ * Global list of private DSI PLL struct pointers. We need this for Dual DSI
+ * mode, where the master PLL's clk_ops needs access the slave's private data
+ */
+static struct dsi_pll_7nm *pll_7nm_list[DSI_MAX];
+
+static void dsi_pll_setup_config(struct dsi_pll_7nm *pll)
+{
+	struct dsi_pll_config *config = &pll->pll_configuration;
+
+	config->ref_freq = pll->vco_ref_clk_rate;
+	config->output_div = 1;
+	config->dec_bits = 8;
+	config->frac_bits = 18;
+	config->lock_timer = 64;
+	config->ssc_freq = 31500;
+	config->ssc_offset = 4800;
+	config->ssc_adj_per = 2;
+	config->thresh_cycles = 32;
+	config->refclk_cycles = 256;
+
+	config->div_override = false;
+	config->ignore_frac = false;
+	config->disable_prescaler = false;
+
+	/* TODO: ssc enable */
+	config->enable_ssc = false;
+	config->ssc_center = 0;
+}
+
+static void dsi_pll_calc_dec_frac(struct dsi_pll_7nm *pll)
+{
+	struct dsi_pll_config *config = &pll->pll_configuration;
+	struct dsi_pll_regs *regs = &pll->reg_setup;
+	u64 fref = pll->vco_ref_clk_rate;
+	u64 pll_freq;
+	u64 divider;
+	u64 dec, dec_multiple;
+	u32 frac;
+	u64 multiplier;
+
+	pll_freq = pll->vco_current_rate;
+
+	if (config->disable_prescaler)
+		divider = fref;
+	else
+		divider = fref * 2;
+
+	multiplier = 1 << config->frac_bits;
+	dec_multiple = div_u64(pll_freq * multiplier, divider);
+	div_u64_rem(dec_multiple, multiplier, &frac);
+
+	dec = div_u64(dec_multiple, multiplier);
+
+	if (pll->base.type != MSM_DSI_PHY_7NM_V4_1)
+		regs->pll_clock_inverters = 0x28;
+	else if (pll_freq <= 1000000000ULL)
+		regs->pll_clock_inverters = 0xa0;
+	else if (pll_freq <= 2500000000ULL)
+		regs->pll_clock_inverters = 0x20;
+	else if (pll_freq <= 3020000000ULL)
+		regs->pll_clock_inverters = 0x00;
+	else
+		regs->pll_clock_inverters = 0x40;
+
+	regs->pll_lockdet_rate = config->lock_timer;
+	regs->decimal_div_start = dec;
+	regs->frac_div_start_low = (frac & 0xff);
+	regs->frac_div_start_mid = (frac & 0xff00) >> 8;
+	regs->frac_div_start_high = (frac & 0x30000) >> 16;
+}
+
+#define SSC_CENTER		BIT(0)
+#define SSC_EN			BIT(1)
+
+static void dsi_pll_calc_ssc(struct dsi_pll_7nm *pll)
+{
+	struct dsi_pll_config *config = &pll->pll_configuration;
+	struct dsi_pll_regs *regs = &pll->reg_setup;
+	u32 ssc_per;
+	u32 ssc_mod;
+	u64 ssc_step_size;
+	u64 frac;
+
+	if (!config->enable_ssc) {
+		DBG("SSC not enabled\n");
+		return;
+	}
+
+	ssc_per = DIV_ROUND_CLOSEST(config->ref_freq, config->ssc_freq) / 2 - 1;
+	ssc_mod = (ssc_per + 1) % (config->ssc_adj_per + 1);
+	ssc_per -= ssc_mod;
+
+	frac = regs->frac_div_start_low |
+			(regs->frac_div_start_mid << 8) |
+			(regs->frac_div_start_high << 16);
+	ssc_step_size = regs->decimal_div_start;
+	ssc_step_size *= (1 << config->frac_bits);
+	ssc_step_size += frac;
+	ssc_step_size *= config->ssc_offset;
+	ssc_step_size *= (config->ssc_adj_per + 1);
+	ssc_step_size = div_u64(ssc_step_size, (ssc_per + 1));
+	ssc_step_size = DIV_ROUND_CLOSEST_ULL(ssc_step_size, 1000000);
+
+	regs->ssc_div_per_low = ssc_per & 0xFF;
+	regs->ssc_div_per_high = (ssc_per & 0xFF00) >> 8;
+	regs->ssc_stepsize_low = (u32)(ssc_step_size & 0xFF);
+	regs->ssc_stepsize_high = (u32)((ssc_step_size & 0xFF00) >> 8);
+	regs->ssc_adjper_low = config->ssc_adj_per & 0xFF;
+	regs->ssc_adjper_high = (config->ssc_adj_per & 0xFF00) >> 8;
+
+	regs->ssc_control = config->ssc_center ? SSC_CENTER : 0;
+
+	pr_debug("SCC: Dec:%d, frac:%llu, frac_bits:%d\n",
+		 regs->decimal_div_start, frac, config->frac_bits);
+	pr_debug("SSC: div_per:0x%X, stepsize:0x%X, adjper:0x%X\n",
+		 ssc_per, (u32)ssc_step_size, config->ssc_adj_per);
+}
+
+static void dsi_pll_ssc_commit(struct dsi_pll_7nm *pll)
+{
+	void __iomem *base = pll->mmio;
+	struct dsi_pll_regs *regs = &pll->reg_setup;
+
+	if (pll->pll_configuration.enable_ssc) {
+		pr_debug("SSC is enabled\n");
+
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_LOW_1,
+			  regs->ssc_stepsize_low);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_STEPSIZE_HIGH_1,
+			  regs->ssc_stepsize_high);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_LOW_1,
+			  regs->ssc_div_per_low);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_DIV_PER_HIGH_1,
+			  regs->ssc_div_per_high);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_ADJPER_LOW_1,
+			  regs->ssc_adjper_low);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_ADJPER_HIGH_1,
+			  regs->ssc_adjper_high);
+		pll_write(base + REG_DSI_7nm_PHY_PLL_SSC_CONTROL,
+			  SSC_EN | regs->ssc_control);
+	}
+}
+
+static void dsi_pll_config_hzindep_reg(struct dsi_pll_7nm *pll)
+{
+	void __iomem *base = pll->mmio;
+	u8 analog_controls_five_1 = 0x01, vco_config_1 = 0x00;
+
+	if (pll->base.type == MSM_DSI_PHY_7NM_V4_1) {
+		if (pll->vco_current_rate >= 3100000000ULL)
+			analog_controls_five_1 = 0x03;
+
+		if (pll->vco_current_rate < 1520000000ULL)
+			vco_config_1 = 0x08;
+		else if (pll->vco_current_rate < 2990000000ULL)
+			vco_config_1 = 0x01;
+	}
+
+	pll_write(base + REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FIVE_1,
+		  analog_controls_five_1);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_VCO_CONFIG_1, vco_config_1);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_FIVE, 0x01);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_TWO, 0x03);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_ANALOG_CONTROLS_THREE, 0x00);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_DSM_DIVIDER, 0x00);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_FEEDBACK_DIVIDER, 0x4e);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_CALIBRATION_SETTINGS, 0x40);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_THREE, 0xba);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_FREQ_DETECT_SETTINGS_ONE, 0x0c);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_OUTDIV, 0x00);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_CORE_OVERRIDE, 0x00);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_DIGITAL_TIMERS_TWO, 0x08);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_PROP_GAIN_RATE_1, 0x0a);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_BAND_SEL_RATE_1, 0xc0);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_1, 0x84);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_1, 0x82);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_FL_INT_GAIN_PFILT_BAND_1, 0x4c);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_LOCK_OVERRIDE, 0x80);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PFILT, 0x29);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PFILT, 0x2f);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_IFILT, 0x2a);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_IFILT,
+		  pll->base.type == MSM_DSI_PHY_7NM_V4_1 ? 0x3f : 0x22);
+
+	if (pll->base.type == MSM_DSI_PHY_7NM_V4_1) {
+		pll_write(base + REG_DSI_7nm_PHY_PLL_PERF_OPTIMIZE, 0x22);
+		if (pll->slave)
+			pll_write(pll->slave->mmio + REG_DSI_7nm_PHY_PLL_PERF_OPTIMIZE, 0x22);
+	}
+}
+
+static void dsi_pll_commit(struct dsi_pll_7nm *pll)
+{
+	void __iomem *base = pll->mmio;
+	struct dsi_pll_regs *reg = &pll->reg_setup;
+
+	pll_write(base + REG_DSI_7nm_PHY_PLL_CORE_INPUT_OVERRIDE, 0x12);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START_1, reg->decimal_div_start);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_LOW_1, reg->frac_div_start_low);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_MID_1, reg->frac_div_start_mid);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_HIGH_1, reg->frac_div_start_high);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_LOCKDET_RATE_1, 0x40);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_PLL_LOCK_DELAY, 0x06);
+	pll_write(base + REG_DSI_7nm_PHY_PLL_CMODE_1, 0x10); /* TODO: 0x00 for CPHY */
+	pll_write(base + REG_DSI_7nm_PHY_PLL_CLOCK_INVERTERS, reg->pll_clock_inverters);
+}
+
+static int dsi_pll_7nm_vco_set_rate(struct clk_hw *hw, unsigned long rate,
+				     unsigned long parent_rate)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+
+	DBG("DSI PLL%d rate=%lu, parent's=%lu", pll_7nm->id, rate,
+	    parent_rate);
+
+	pll_7nm->vco_current_rate = rate;
+	pll_7nm->vco_ref_clk_rate = VCO_REF_CLK_RATE;
+
+	dsi_pll_setup_config(pll_7nm);
+
+	dsi_pll_calc_dec_frac(pll_7nm);
+
+	dsi_pll_calc_ssc(pll_7nm);
+
+	dsi_pll_commit(pll_7nm);
+
+	dsi_pll_config_hzindep_reg(pll_7nm);
+
+	dsi_pll_ssc_commit(pll_7nm);
+
+	/* flush, ensure all register writes are done*/
+	wmb();
+
+	return 0;
+}
+
+static int dsi_pll_7nm_lock_status(struct dsi_pll_7nm *pll)
+{
+	int rc;
+	u32 status = 0;
+	u32 const delay_us = 100;
+	u32 const timeout_us = 5000;
+
+	rc = readl_poll_timeout_atomic(pll->mmio +
+				       REG_DSI_7nm_PHY_PLL_COMMON_STATUS_ONE,
+				       status,
+				       ((status & BIT(0)) > 0),
+				       delay_us,
+				       timeout_us);
+	if (rc)
+		pr_err("DSI PLL(%d) lock failed, status=0x%08x\n",
+		       pll->id, status);
+
+	return rc;
+}
+
+static void dsi_pll_disable_pll_bias(struct dsi_pll_7nm *pll)
+{
+	u32 data = pll_read(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CTRL_0);
+
+	pll_write(pll->mmio + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES, 0);
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CTRL_0, data & ~BIT(5));
+	ndelay(250);
+}
+
+static void dsi_pll_enable_pll_bias(struct dsi_pll_7nm *pll)
+{
+	u32 data = pll_read(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CTRL_0);
+
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CTRL_0, data | BIT(5));
+	pll_write(pll->mmio + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES, 0xc0);
+	ndelay(250);
+}
+
+static void dsi_pll_disable_global_clk(struct dsi_pll_7nm *pll)
+{
+	u32 data;
+
+	data = pll_read(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CLK_CFG1);
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CLK_CFG1, data & ~BIT(5));
+}
+
+static void dsi_pll_enable_global_clk(struct dsi_pll_7nm *pll)
+{
+	u32 data;
+
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CTRL_3, 0x04);
+
+	data = pll_read(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CLK_CFG1);
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_CLK_CFG1,
+		  data | BIT(5) | BIT(4));
+}
+
+static void dsi_pll_phy_dig_reset(struct dsi_pll_7nm *pll)
+{
+	/*
+	 * Reset the PHY digital domain. This would be needed when
+	 * coming out of a CX or analog rail power collapse while
+	 * ensuring that the pads maintain LP00 or LP11 state
+	 */
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE4, BIT(0));
+	wmb(); /* Ensure that the reset is deasserted */
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE4, 0x0);
+	wmb(); /* Ensure that the reset is deasserted */
+}
+
+static int dsi_pll_7nm_vco_prepare(struct clk_hw *hw)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	int rc;
+
+	dsi_pll_enable_pll_bias(pll_7nm);
+	if (pll_7nm->slave)
+		dsi_pll_enable_pll_bias(pll_7nm->slave);
+
+	/* Start PLL */
+	pll_write(pll_7nm->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_PLL_CNTRL, 0x01);
+
+	/*
+	 * ensure all PLL configurations are written prior to checking
+	 * for PLL lock.
+	 */
+	wmb();
+
+	/* Check for PLL lock */
+	rc = dsi_pll_7nm_lock_status(pll_7nm);
+	if (rc) {
+		pr_err("PLL(%d) lock failed\n", pll_7nm->id);
+		goto error;
+	}
+
+	pll->pll_on = true;
+
+	/*
+	 * assert power on reset for PHY digital in case the PLL is
+	 * enabled after CX of analog domain power collapse. This needs
+	 * to be done before enabling the global clk.
+	 */
+	dsi_pll_phy_dig_reset(pll_7nm);
+	if (pll_7nm->slave)
+		dsi_pll_phy_dig_reset(pll_7nm->slave);
+
+	dsi_pll_enable_global_clk(pll_7nm);
+	if (pll_7nm->slave)
+		dsi_pll_enable_global_clk(pll_7nm->slave);
+
+error:
+	return rc;
+}
+
+static void dsi_pll_disable_sub(struct dsi_pll_7nm *pll)
+{
+	pll_write(pll->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_RBUF_CTRL, 0);
+	dsi_pll_disable_pll_bias(pll);
+}
+
+static void dsi_pll_7nm_vco_unprepare(struct clk_hw *hw)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+
+	/*
+	 * To avoid any stray glitches while abruptly powering down the PLL
+	 * make sure to gate the clock using the clock enable bit before
+	 * powering down the PLL
+	 */
+	dsi_pll_disable_global_clk(pll_7nm);
+	pll_write(pll_7nm->phy_cmn_mmio + REG_DSI_7nm_PHY_CMN_PLL_CNTRL, 0);
+	dsi_pll_disable_sub(pll_7nm);
+	if (pll_7nm->slave) {
+		dsi_pll_disable_global_clk(pll_7nm->slave);
+		dsi_pll_disable_sub(pll_7nm->slave);
+	}
+	/* flush, ensure all register writes are done */
+	wmb();
+	pll->pll_on = false;
+}
+
+static unsigned long dsi_pll_7nm_vco_recalc_rate(struct clk_hw *hw,
+						  unsigned long parent_rate)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	void __iomem *base = pll_7nm->mmio;
+	u64 ref_clk = pll_7nm->vco_ref_clk_rate;
+	u64 vco_rate = 0x0;
+	u64 multiplier;
+	u32 frac;
+	u32 dec;
+	u64 pll_freq, tmp64;
+
+	dec = pll_read(base + REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START_1);
+	dec &= 0xff;
+
+	frac = pll_read(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_LOW_1);
+	frac |= ((pll_read(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_MID_1) &
+		  0xff) << 8);
+	frac |= ((pll_read(base + REG_DSI_7nm_PHY_PLL_FRAC_DIV_START_HIGH_1) &
+		  0x3) << 16);
+
+	/*
+	 * TODO:
+	 *	1. Assumes prescaler is disabled
+	 *	2. Multiplier is 2^18. it should be 2^(num_of_frac_bits)
+	 */
+	multiplier = 1 << 18;
+	pll_freq = dec * (ref_clk * 2);
+	tmp64 = (ref_clk * 2 * frac);
+	pll_freq += div_u64(tmp64, multiplier);
+
+	vco_rate = pll_freq;
+
+	DBG("DSI PLL%d returning vco rate = %lu, dec = %x, frac = %x",
+	    pll_7nm->id, (unsigned long)vco_rate, dec, frac);
+
+	return (unsigned long)vco_rate;
+}
+
+static const struct clk_ops clk_ops_dsi_pll_7nm_vco = {
+	.round_rate = msm_dsi_pll_helper_clk_round_rate,
+	.set_rate = dsi_pll_7nm_vco_set_rate,
+	.recalc_rate = dsi_pll_7nm_vco_recalc_rate,
+	.prepare = dsi_pll_7nm_vco_prepare,
+	.unprepare = dsi_pll_7nm_vco_unprepare,
+};
+
+/*
+ * PLL Callbacks
+ */
+
+static void dsi_pll_7nm_save_state(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	struct pll_7nm_cached_state *cached = &pll_7nm->cached_state;
+	void __iomem *phy_base = pll_7nm->phy_cmn_mmio;
+	u32 cmn_clk_cfg0, cmn_clk_cfg1;
+
+	cached->pll_out_div = pll_read(pll_7nm->mmio +
+				       REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE);
+	cached->pll_out_div &= 0x3;
+
+	cmn_clk_cfg0 = pll_read(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG0);
+	cached->bit_clk_div = cmn_clk_cfg0 & 0xf;
+	cached->pix_clk_div = (cmn_clk_cfg0 & 0xf0) >> 4;
+
+	cmn_clk_cfg1 = pll_read(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG1);
+	cached->pll_mux = cmn_clk_cfg1 & 0x3;
+
+	DBG("DSI PLL%d outdiv %x bit_clk_div %x pix_clk_div %x pll_mux %x",
+	    pll_7nm->id, cached->pll_out_div, cached->bit_clk_div,
+	    cached->pix_clk_div, cached->pll_mux);
+}
+
+static int dsi_pll_7nm_restore_state(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	struct pll_7nm_cached_state *cached = &pll_7nm->cached_state;
+	void __iomem *phy_base = pll_7nm->phy_cmn_mmio;
+	u32 val;
+
+	val = pll_read(pll_7nm->mmio + REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE);
+	val &= ~0x3;
+	val |= cached->pll_out_div;
+	pll_write(pll_7nm->mmio + REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE, val);
+
+	pll_write(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG0,
+		  cached->bit_clk_div | (cached->pix_clk_div << 4));
+
+	val = pll_read(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG1);
+	val &= ~0x3;
+	val |= cached->pll_mux;
+	pll_write(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG1, val);
+
+	DBG("DSI PLL%d", pll_7nm->id);
+
+	return 0;
+}
+
+static int dsi_pll_7nm_set_usecase(struct msm_dsi_pll *pll,
+				    enum msm_dsi_phy_usecase uc)
+{
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	void __iomem *base = pll_7nm->phy_cmn_mmio;
+	u32 data = 0x0;	/* internal PLL */
+
+	DBG("DSI PLL%d", pll_7nm->id);
+
+	switch (uc) {
+	case MSM_DSI_PHY_STANDALONE:
+		break;
+	case MSM_DSI_PHY_MASTER:
+		pll_7nm->slave = pll_7nm_list[(pll_7nm->id + 1) % DSI_MAX];
+		break;
+	case MSM_DSI_PHY_SLAVE:
+		data = 0x1; /* external PLL */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* set PLL src */
+	pll_write(base + REG_DSI_7nm_PHY_CMN_CLK_CFG1, (data << 2));
+
+	pll_7nm->uc = uc;
+
+	return 0;
+}
+
+static int dsi_pll_7nm_get_provider(struct msm_dsi_pll *pll,
+				     struct clk **byte_clk_provider,
+				     struct clk **pixel_clk_provider)
+{
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	struct clk_hw_onecell_data *hw_data = pll_7nm->hw_data;
+
+	DBG("DSI PLL%d", pll_7nm->id);
+
+	if (byte_clk_provider)
+		*byte_clk_provider = hw_data->hws[DSI_BYTE_PLL_CLK]->clk;
+	if (pixel_clk_provider)
+		*pixel_clk_provider = hw_data->hws[DSI_PIXEL_PLL_CLK]->clk;
+
+	return 0;
+}
+
+static void dsi_pll_7nm_destroy(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(pll);
+	struct device *dev = &pll_7nm->pdev->dev;
+
+	DBG("DSI PLL%d", pll_7nm->id);
+	of_clk_del_provider(dev->of_node);
+
+	clk_hw_unregister_divider(pll_7nm->out_dsiclk_hw);
+	clk_hw_unregister_mux(pll_7nm->pclk_mux_hw);
+	clk_hw_unregister_fixed_factor(pll_7nm->post_out_div_clk_hw);
+	clk_hw_unregister_fixed_factor(pll_7nm->by_2_bit_clk_hw);
+	clk_hw_unregister_fixed_factor(pll_7nm->byte_clk_hw);
+	clk_hw_unregister_divider(pll_7nm->bit_clk_hw);
+	clk_hw_unregister_divider(pll_7nm->out_div_clk_hw);
+	clk_hw_unregister(&pll_7nm->base.clk_hw);
+}
+
+/*
+ * The post dividers and mux clocks are created using the standard divider and
+ * mux API. Unlike the 14nm PHY, the slave PLL doesn't need its dividers/mux
+ * state to follow the master PLL's divider/mux state. Therefore, we don't
+ * require special clock ops that also configure the slave PLL registers
+ */
+static int pll_7nm_register(struct dsi_pll_7nm *pll_7nm)
+{
+	char clk_name[32], parent[32], vco_name[32];
+	char parent2[32], parent3[32], parent4[32];
+	struct clk_init_data vco_init = {
+		.parent_names = (const char *[]){ "bi_tcxo" },
+		.num_parents = 1,
+		.name = vco_name,
+		.flags = CLK_IGNORE_UNUSED,
+		.ops = &clk_ops_dsi_pll_7nm_vco,
+	};
+	struct device *dev = &pll_7nm->pdev->dev;
+	struct clk_hw_onecell_data *hw_data;
+	struct clk_hw *hw;
+	int ret;
+
+	DBG("DSI%d", pll_7nm->id);
+
+	hw_data = devm_kzalloc(dev, sizeof(*hw_data) +
+			       NUM_PROVIDED_CLKS * sizeof(struct clk_hw *),
+			       GFP_KERNEL);
+	if (!hw_data)
+		return -ENOMEM;
+
+	snprintf(vco_name, 32, "dsi%dvco_clk", pll_7nm->id);
+	pll_7nm->base.clk_hw.init = &vco_init;
+
+	ret = clk_hw_register(dev, &pll_7nm->base.clk_hw);
+	if (ret)
+		return ret;
+
+	snprintf(clk_name, 32, "dsi%d_pll_out_div_clk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%dvco_clk", pll_7nm->id);
+
+	hw = clk_hw_register_divider(dev, clk_name,
+				     parent, CLK_SET_RATE_PARENT,
+				     pll_7nm->mmio +
+				     REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE,
+				     0, 2, CLK_DIVIDER_POWER_OF_TWO, NULL);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_base_clk_hw;
+	}
+
+	pll_7nm->out_div_clk_hw = hw;
+
+	snprintf(clk_name, 32, "dsi%d_pll_bit_clk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pll_out_div_clk", pll_7nm->id);
+
+	/* BIT CLK: DIV_CTRL_3_0 */
+	hw = clk_hw_register_divider(dev, clk_name, parent,
+				     CLK_SET_RATE_PARENT,
+				     pll_7nm->phy_cmn_mmio +
+				     REG_DSI_7nm_PHY_CMN_CLK_CFG0,
+				     0, 4, CLK_DIVIDER_ONE_BASED,
+				     &pll_7nm->postdiv_lock);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_out_div_clk_hw;
+	}
+
+	pll_7nm->bit_clk_hw = hw;
+
+	snprintf(clk_name, 32, "dsi%d_phy_pll_out_byteclk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pll_bit_clk", pll_7nm->id);
+
+	/* DSI Byte clock = VCO_CLK / OUT_DIV / BIT_DIV / 8 */
+	hw = clk_hw_register_fixed_factor(dev, clk_name, parent,
+					  CLK_SET_RATE_PARENT, 1, 8);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_bit_clk_hw;
+	}
+
+	pll_7nm->byte_clk_hw = hw;
+	hw_data->hws[DSI_BYTE_PLL_CLK] = hw;
+
+	snprintf(clk_name, 32, "dsi%d_pll_by_2_bit_clk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pll_bit_clk", pll_7nm->id);
+
+	hw = clk_hw_register_fixed_factor(dev, clk_name, parent,
+					  0, 1, 2);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_byte_clk_hw;
+	}
+
+	pll_7nm->by_2_bit_clk_hw = hw;
+
+	snprintf(clk_name, 32, "dsi%d_pll_post_out_div_clk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pll_out_div_clk", pll_7nm->id);
+
+	hw = clk_hw_register_fixed_factor(dev, clk_name, parent,
+					  0, 1, 4);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_by_2_bit_clk_hw;
+	}
+
+	pll_7nm->post_out_div_clk_hw = hw;
+
+	snprintf(clk_name, 32, "dsi%d_pclk_mux", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pll_bit_clk", pll_7nm->id);
+	snprintf(parent2, 32, "dsi%d_pll_by_2_bit_clk", pll_7nm->id);
+	snprintf(parent3, 32, "dsi%d_pll_out_div_clk", pll_7nm->id);
+	snprintf(parent4, 32, "dsi%d_pll_post_out_div_clk", pll_7nm->id);
+
+	hw = clk_hw_register_mux(dev, clk_name,
+				 ((const char *[]){
+				 parent, parent2, parent3, parent4
+				 }), 4, 0, pll_7nm->phy_cmn_mmio +
+				 REG_DSI_7nm_PHY_CMN_CLK_CFG1,
+				 0, 2, 0, NULL);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_post_out_div_clk_hw;
+	}
+
+	pll_7nm->pclk_mux_hw = hw;
+
+	snprintf(clk_name, 32, "dsi%d_phy_pll_out_dsiclk", pll_7nm->id);
+	snprintf(parent, 32, "dsi%d_pclk_mux", pll_7nm->id);
+
+	/* PIX CLK DIV : DIV_CTRL_7_4*/
+	hw = clk_hw_register_divider(dev, clk_name, parent,
+				     0, pll_7nm->phy_cmn_mmio +
+					REG_DSI_7nm_PHY_CMN_CLK_CFG0,
+				     4, 4, CLK_DIVIDER_ONE_BASED,
+				     &pll_7nm->postdiv_lock);
+	if (IS_ERR(hw)) {
+		ret = PTR_ERR(hw);
+		goto err_pclk_mux_hw;
+	}
+
+	pll_7nm->out_dsiclk_hw = hw;
+	hw_data->hws[DSI_PIXEL_PLL_CLK] = hw;
+
+	hw_data->num = NUM_PROVIDED_CLKS;
+	pll_7nm->hw_data = hw_data;
+
+	ret = of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get,
+				     pll_7nm->hw_data);
+	if (ret) {
+		DRM_DEV_ERROR(dev, "failed to register clk provider: %d\n", ret);
+		goto err_dsiclk_hw;
+	}
+
+	return 0;
+
+err_dsiclk_hw:
+	clk_hw_unregister_divider(pll_7nm->out_dsiclk_hw);
+err_pclk_mux_hw:
+	clk_hw_unregister_mux(pll_7nm->pclk_mux_hw);
+err_post_out_div_clk_hw:
+	clk_hw_unregister_fixed_factor(pll_7nm->post_out_div_clk_hw);
+err_by_2_bit_clk_hw:
+	clk_hw_unregister_fixed_factor(pll_7nm->by_2_bit_clk_hw);
+err_byte_clk_hw:
+	clk_hw_unregister_fixed_factor(pll_7nm->byte_clk_hw);
+err_bit_clk_hw:
+	clk_hw_unregister_divider(pll_7nm->bit_clk_hw);
+err_out_div_clk_hw:
+	clk_hw_unregister_divider(pll_7nm->out_div_clk_hw);
+err_base_clk_hw:
+	clk_hw_unregister(&pll_7nm->base.clk_hw);
+
+	return ret;
+}
+
+struct msm_dsi_pll *msm_dsi_pll_7nm_init(struct platform_device *pdev, int id)
+{
+	struct dsi_pll_7nm *pll_7nm;
+	struct msm_dsi_pll *pll;
+	int ret;
+
+	pll_7nm = devm_kzalloc(&pdev->dev, sizeof(*pll_7nm), GFP_KERNEL);
+	if (!pll_7nm)
+		return ERR_PTR(-ENOMEM);
+
+	DBG("DSI PLL%d", id);
+
+	pll_7nm->pdev = pdev;
+	pll_7nm->id = id;
+	pll_7nm_list[id] = pll_7nm;
+
+	pll_7nm->phy_cmn_mmio = msm_ioremap(pdev, "dsi_phy", "DSI_PHY");
+	if (IS_ERR_OR_NULL(pll_7nm->phy_cmn_mmio)) {
+		DRM_DEV_ERROR(&pdev->dev, "failed to map CMN PHY base\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	pll_7nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL");
+	if (IS_ERR_OR_NULL(pll_7nm->mmio)) {
+		DRM_DEV_ERROR(&pdev->dev, "failed to map PLL base\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	spin_lock_init(&pll_7nm->postdiv_lock);
+
+	pll = &pll_7nm->base;
+	pll->min_rate = 1000000000UL;
+	pll->max_rate = 3500000000UL;
+	if (pll->type == MSM_DSI_PHY_7NM_V4_1) {
+		pll->min_rate = 600000000UL;
+		pll->max_rate = (unsigned long)5000000000ULL;
+		/* workaround for max rate overflowing on 32-bit builds: */
+		pll->max_rate = max(pll->max_rate, 0xffffffffUL);
+	}
+	pll->get_provider = dsi_pll_7nm_get_provider;
+	pll->destroy = dsi_pll_7nm_destroy;
+	pll->save_state = dsi_pll_7nm_save_state;
+	pll->restore_state = dsi_pll_7nm_restore_state;
+	pll->set_usecase = dsi_pll_7nm_set_usecase;
+
+	pll_7nm->vco_delay = 1;
+
+	ret = pll_7nm_register(pll_7nm);
+	if (ret) {
+		DRM_DEV_ERROR(&pdev->dev, "failed to register PLL: %d\n", ret);
+		return ERR_PTR(ret);
+	}
+
+	/* TODO: Remove this when we have proper display handover support */
+	msm_dsi_pll_save_state(pll);
+
+	return pll;
+}
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 79333842f70a..49685571dc0e 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -453,15 +453,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv)
 	if (ret)
 		goto err_msm_uninit;
 
-	if (!dev->dma_parms) {
-		dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms),
-					      GFP_KERNEL);
-		if (!dev->dma_parms) {
-			ret = -ENOMEM;
-			goto err_msm_uninit;
-		}
-	}
-	dma_set_max_seg_size(dev, DMA_BIT_MASK(32));
+	dma_set_max_seg_size(dev, UINT_MAX);
 
 	msm_gem_shrinker_init(ddev);
 
@@ -594,9 +586,10 @@ static int context_init(struct drm_device *dev, struct drm_file *file)
 	if (!ctx)
 		return -ENOMEM;
 
+	kref_init(&ctx->ref);
 	msm_submitqueue_init(dev, ctx);
 
-	ctx->aspace = priv->gpu ? priv->gpu->aspace : NULL;
+	ctx->aspace = msm_gpu_create_private_address_space(priv->gpu, current);
 	file->driver_priv = ctx;
 
 	return 0;
@@ -615,7 +608,7 @@ static int msm_open(struct drm_device *dev, struct drm_file *file)
 static void context_close(struct msm_file_private *ctx)
 {
 	msm_submitqueue_close(ctx);
-	kfree(ctx);
+	msm_file_private_put(ctx);
 }
 
 static void msm_postclose(struct drm_device *dev, struct drm_file *file)
@@ -779,18 +772,19 @@ static int msm_ioctl_gem_cpu_fini(struct drm_device *dev, void *data,
 }
 
 static int msm_ioctl_gem_info_iova(struct drm_device *dev,
-		struct drm_gem_object *obj, uint64_t *iova)
+		struct drm_file *file, struct drm_gem_object *obj,
+		uint64_t *iova)
 {
-	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_file_private *ctx = file->driver_priv;
 
-	if (!priv->gpu)
+	if (!ctx->aspace)
 		return -EINVAL;
 
 	/*
 	 * Don't pin the memory here - just get an address so that userspace can
 	 * be productive
 	 */
-	return msm_gem_get_iova(obj, priv->gpu->aspace, iova);
+	return msm_gem_get_iova(obj, ctx->aspace, iova);
 }
 
 static int msm_ioctl_gem_info(struct drm_device *dev, void *data,
@@ -829,7 +823,7 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data,
 		args->value = msm_gem_mmap_offset(obj);
 		break;
 	case MSM_INFO_GET_IOVA:
-		ret = msm_ioctl_gem_info_iova(dev, obj, &args->value);
+		ret = msm_ioctl_gem_info_iova(dev, file, obj, &args->value);
 		break;
 	case MSM_INFO_SET_NAME:
 		/* length check should leave room for terminating null: */
@@ -1358,6 +1352,7 @@ static int __init msm_drm_register(void)
 	msm_dsi_register();
 	msm_edp_register();
 	msm_hdmi_register();
+	msm_dp_register();
 	adreno_register();
 	return platform_driver_register(&msm_platform_driver);
 }
@@ -1366,6 +1361,7 @@ static void __exit msm_drm_unregister(void)
 {
 	DBG("fini");
 	platform_driver_unregister(&msm_platform_driver);
+	msm_dp_unregister();
 	msm_hdmi_unregister();
 	adreno_unregister();
 	msm_edp_unregister();
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index af259b0573ea..b9dd8f8f4887 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -57,6 +57,7 @@ struct msm_file_private {
 	struct list_head submitqueues;
 	int queueid;
 	struct msm_gem_address_space *aspace;
+	struct kref ref;
 };
 
 enum msm_mdp_plane_property {
@@ -159,6 +160,8 @@ struct msm_drm_private {
 	/* DSI is shared by mdp4 and mdp5 */
 	struct msm_dsi *dsi[2];
 
+	struct msm_dp *dp;
+
 	/* when we have more than one 'msm_gpu' these need to be an array: */
 	struct msm_gpu *gpu;
 	struct msm_file_private *lastctx;
@@ -248,6 +251,10 @@ int msm_gem_map_vma(struct msm_gem_address_space *aspace,
 void msm_gem_close_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma);
 
+
+struct msm_gem_address_space *
+msm_gem_address_space_get(struct msm_gem_address_space *aspace);
+
 void msm_gem_address_space_put(struct msm_gem_address_space *aspace);
 
 struct msm_gem_address_space *
@@ -302,9 +309,8 @@ void msm_gem_put_vaddr(struct drm_gem_object *obj);
 int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv);
 int msm_gem_sync_object(struct drm_gem_object *obj,
 		struct msm_fence_context *fctx, bool exclusive);
-void msm_gem_move_to_active(struct drm_gem_object *obj,
-		struct msm_gpu *gpu, bool exclusive, struct dma_fence *fence);
-void msm_gem_move_to_inactive(struct drm_gem_object *obj);
+void msm_gem_active_get(struct drm_gem_object *obj, struct msm_gpu *gpu);
+void msm_gem_active_put(struct drm_gem_object *obj);
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout);
 int msm_gem_cpu_fini(struct drm_gem_object *obj);
 void msm_gem_free_object(struct drm_gem_object *obj);
@@ -378,6 +384,63 @@ static inline int msm_dsi_modeset_init(struct msm_dsi *msm_dsi,
 }
 #endif
 
+#ifdef CONFIG_DRM_MSM_DP
+int __init msm_dp_register(void);
+void __exit msm_dp_unregister(void);
+int msm_dp_modeset_init(struct msm_dp *dp_display, struct drm_device *dev,
+			 struct drm_encoder *encoder);
+int msm_dp_display_enable(struct msm_dp *dp, struct drm_encoder *encoder);
+int msm_dp_display_disable(struct msm_dp *dp, struct drm_encoder *encoder);
+int msm_dp_display_pre_disable(struct msm_dp *dp, struct drm_encoder *encoder);
+void msm_dp_display_mode_set(struct msm_dp *dp, struct drm_encoder *encoder,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode);
+void msm_dp_irq_postinstall(struct msm_dp *dp_display);
+
+void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor);
+
+#else
+static inline int __init msm_dp_register(void)
+{
+	return -EINVAL;
+}
+static inline void __exit msm_dp_unregister(void)
+{
+}
+static inline int msm_dp_modeset_init(struct msm_dp *dp_display,
+				       struct drm_device *dev,
+				       struct drm_encoder *encoder)
+{
+	return -EINVAL;
+}
+static inline int msm_dp_display_enable(struct msm_dp *dp,
+					struct drm_encoder *encoder)
+{
+	return -EINVAL;
+}
+static inline int msm_dp_display_disable(struct msm_dp *dp,
+					struct drm_encoder *encoder)
+{
+	return -EINVAL;
+}
+static inline void msm_dp_display_mode_set(struct msm_dp *dp,
+				struct drm_encoder *encoder,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+}
+
+static inline void msm_dp_irq_postinstall(struct msm_dp *dp_display)
+{
+}
+
+static inline void msm_dp_debugfs_init(struct msm_dp *dp_display,
+		struct drm_minor *minor)
+{
+}
+
+#endif
+
 void __init msm_mdp_register(void);
 void __exit msm_mdp_unregister(void);
 void __init msm_dpu_register(void);
@@ -398,8 +461,9 @@ void msm_perf_debugfs_cleanup(struct msm_drm_private *priv);
 #else
 static inline int msm_debugfs_late_init(struct drm_device *dev) { return 0; }
 __printf(3, 4)
-static inline void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit,
-		const char *fmt, ...) {}
+static inline void msm_rd_dump_submit(struct msm_rd_state *rd,
+			struct msm_gem_submit *submit,
+			const char *fmt, ...) {}
 static inline void msm_rd_debugfs_cleanup(struct msm_drm_private *priv) {}
 static inline void msm_perf_debugfs_cleanup(struct msm_drm_private *priv) {}
 #endif
@@ -419,7 +483,8 @@ struct msm_gpu_submitqueue;
 int msm_submitqueue_init(struct drm_device *drm, struct msm_file_private *ctx);
 struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_file_private *ctx,
 		u32 id);
-int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx,
+int msm_submitqueue_create(struct drm_device *drm,
+		struct msm_file_private *ctx,
 		u32 prio, u32 flags, u32 *id);
 int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx,
 		struct drm_msm_submitqueue_query *args);
@@ -428,6 +493,26 @@ void msm_submitqueue_close(struct msm_file_private *ctx);
 
 void msm_submitqueue_destroy(struct kref *kref);
 
+static inline void __msm_file_private_destroy(struct kref *kref)
+{
+	struct msm_file_private *ctx = container_of(kref,
+		struct msm_file_private, ref);
+
+	msm_gem_address_space_put(ctx->aspace);
+	kfree(ctx);
+}
+
+static inline void msm_file_private_put(struct msm_file_private *ctx)
+{
+	kref_put(&ctx->ref, __msm_file_private_destroy);
+}
+
+static inline struct msm_file_private *msm_file_private_get(
+	struct msm_file_private *ctx)
+{
+	kref_get(&ctx->ref);
+	return ctx;
+}
 
 #define DBG(fmt, ...) DRM_DEBUG_DRIVER(fmt"\n", ##__VA_ARGS__)
 #define VERB(fmt, ...) if (0) DRM_DEBUG_DRIVER(fmt"\n", ##__VA_ARGS__)
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index b2f49152b4d4..ec602113be78 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -52,26 +52,14 @@ static void sync_for_device(struct msm_gem_object *msm_obj)
 {
 	struct device *dev = msm_obj->base.dev->dev;
 
-	if (get_dma_ops(dev) && IS_ENABLED(CONFIG_ARM64)) {
-		dma_sync_sg_for_device(dev, msm_obj->sgt->sgl,
-			msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
-	} else {
-		dma_map_sg(dev, msm_obj->sgt->sgl,
-			msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
-	}
+	dma_map_sgtable(dev, msm_obj->sgt, DMA_BIDIRECTIONAL, 0);
 }
 
 static void sync_for_cpu(struct msm_gem_object *msm_obj)
 {
 	struct device *dev = msm_obj->base.dev->dev;
 
-	if (get_dma_ops(dev) && IS_ENABLED(CONFIG_ARM64)) {
-		dma_sync_sg_for_cpu(dev, msm_obj->sgt->sgl,
-			msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
-	} else {
-		dma_unmap_sg(dev, msm_obj->sgt->sgl,
-			msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
-	}
+	dma_unmap_sgtable(dev, msm_obj->sgt, DMA_BIDIRECTIONAL, 0);
 }
 
 /* allocate pages from VRAM carveout, used when no IOMMU: */
@@ -126,7 +114,7 @@ static struct page **get_pages(struct drm_gem_object *obj)
 
 		msm_obj->pages = p;
 
-		msm_obj->sgt = drm_prime_pages_to_sg(p, npages);
+		msm_obj->sgt = drm_prime_pages_to_sg(obj->dev, p, npages);
 		if (IS_ERR(msm_obj->sgt)) {
 			void *ptr = ERR_CAST(msm_obj->sgt);
 
@@ -753,31 +741,31 @@ int msm_gem_sync_object(struct drm_gem_object *obj,
 	return 0;
 }
 
-void msm_gem_move_to_active(struct drm_gem_object *obj,
-		struct msm_gpu *gpu, bool exclusive, struct dma_fence *fence)
+void msm_gem_active_get(struct drm_gem_object *obj, struct msm_gpu *gpu)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex));
 	WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED);
-	msm_obj->gpu = gpu;
-	if (exclusive)
-		dma_resv_add_excl_fence(obj->resv, fence);
-	else
-		dma_resv_add_shared_fence(obj->resv, fence);
-	list_del_init(&msm_obj->mm_list);
-	list_add_tail(&msm_obj->mm_list, &gpu->active_list);
+
+	if (!atomic_fetch_inc(&msm_obj->active_count)) {
+		msm_obj->gpu = gpu;
+		list_del_init(&msm_obj->mm_list);
+		list_add_tail(&msm_obj->mm_list, &gpu->active_list);
+	}
 }
 
-void msm_gem_move_to_inactive(struct drm_gem_object *obj)
+void msm_gem_active_put(struct drm_gem_object *obj)
 {
-	struct drm_device *dev = obj->dev;
-	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	struct msm_drm_private *priv = obj->dev->dev_private;
 
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+	WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex));
 
-	msm_obj->gpu = NULL;
-	list_del_init(&msm_obj->mm_list);
-	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);
+	if (!atomic_dec_return(&msm_obj->active_count)) {
+		msm_obj->gpu = NULL;
+		list_del_init(&msm_obj->mm_list);
+		list_add_tail(&msm_obj->mm_list, &priv->inactive_list);
+	}
 }
 
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
@@ -852,11 +840,28 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
 
 		seq_puts(m, "      vmas:");
 
-		list_for_each_entry(vma, &msm_obj->vmas, list)
-			seq_printf(m, " [%s: %08llx,%s,inuse=%d]",
-				vma->aspace != NULL ? vma->aspace->name : NULL,
-				vma->iova, vma->mapped ? "mapped" : "unmapped",
+		list_for_each_entry(vma, &msm_obj->vmas, list) {
+			const char *name, *comm;
+			if (vma->aspace) {
+				struct msm_gem_address_space *aspace = vma->aspace;
+				struct task_struct *task =
+					get_pid_task(aspace->pid, PIDTYPE_PID);
+				if (task) {
+					comm = kstrdup(task->comm, GFP_KERNEL);
+				} else {
+					comm = NULL;
+				}
+				name = aspace->name;
+			} else {
+				name = comm = NULL;
+			}
+			seq_printf(m, " [%s%s%s: aspace=%p, %08llx,%s,inuse=%d]",
+				name, comm ? ":" : "", comm ? comm : "",
+				vma->aspace, vma->iova,
+				vma->mapped ? "mapped" : "unmapped",
 				vma->inuse);
+			kfree(comm);
+		}
 
 		seq_puts(m, "\n");
 	}
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 972490b14ba5..a1bf741b9b89 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -24,6 +24,11 @@ struct msm_gem_address_space {
 	spinlock_t lock; /* Protects drm_mm node allocation/removal */
 	struct msm_mmu *mmu;
 	struct kref kref;
+
+	/* For address spaces associated with a specific process, this
+	 * will be non-NULL:
+	 */
+	struct pid *pid;
 };
 
 struct msm_gem_vma {
@@ -83,12 +88,14 @@ struct msm_gem_object {
 	struct mutex lock; /* Protects resources associated with bo */
 
 	char name[32]; /* Identifier to print for the debugfs files */
+
+	atomic_t active_count;
 };
 #define to_msm_bo(x) container_of(x, struct msm_gem_object, base)
 
 static inline bool is_active(struct msm_gem_object *msm_obj)
 {
-	return msm_obj->gpu != NULL;
+	return atomic_read(&msm_obj->active_count);
 }
 
 static inline bool is_purgeable(struct msm_gem_object *msm_obj)
@@ -142,6 +149,7 @@ struct msm_gem_submit {
 	bool valid;         /* true if no cmdstream patching needed */
 	bool in_rb;         /* "sudo" mode, copy cmds into RB */
 	struct msm_ringbuffer *ring;
+	struct msm_file_private *ctx;
 	unsigned int nr_cmds;
 	unsigned int nr_bos;
 	u32 ident;	   /* A "identifier" for the submit for logging */
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index d7c8948427fe..515ef80816a0 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -19,7 +19,7 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	if (WARN_ON(!msm_obj->pages))  /* should have already pinned! */
 		return NULL;
 
-	return drm_prime_pages_to_sg(msm_obj->pages, npages);
+	return drm_prime_pages_to_sg(obj->dev, msm_obj->pages, npages);
 }
 
 void *msm_gem_prime_vmap(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c
index 722d61668a97..482576d7a39a 100644
--- a/drivers/gpu/drm/msm/msm_gem_shrinker.c
+++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c
@@ -6,6 +6,7 @@
 
 #include "msm_drv.h"
 #include "msm_gem.h"
+#include "msm_gpu_trace.h"
 
 static bool msm_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
 {
@@ -87,7 +88,7 @@ msm_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc)
 		mutex_unlock(&dev->struct_mutex);
 
 	if (freed > 0)
-		pr_info_ratelimited("Purging %lu bytes\n", freed << PAGE_SHIFT);
+		trace_msm_gem_purge(freed << PAGE_SHIFT);
 
 	return freed;
 }
@@ -123,7 +124,7 @@ msm_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr)
 	*(unsigned long *)ptr += unmapped;
 
 	if (unmapped > 0)
-		pr_info_ratelimited("Purging %u vmaps\n", unmapped);
+		trace_msm_gem_purge_vmaps(unmapped);
 
 	return NOTIFY_DONE;
 }
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 8cb9aa15ff90..aa5c60a7132d 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -27,7 +27,7 @@
 #define BO_PINNED   0x2000
 
 static struct msm_gem_submit *submit_create(struct drm_device *dev,
-		struct msm_gpu *gpu, struct msm_gem_address_space *aspace,
+		struct msm_gpu *gpu,
 		struct msm_gpu_submitqueue *queue, uint32_t nr_bos,
 		uint32_t nr_cmds)
 {
@@ -43,7 +43,7 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 		return NULL;
 
 	submit->dev = dev;
-	submit->aspace = aspace;
+	submit->aspace = queue->ctx->aspace;
 	submit->gpu = gpu;
 	submit->fence = NULL;
 	submit->cmd = (void *)&submit->bos[nr_bos];
@@ -677,7 +677,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		}
 	}
 
-	submit = submit_create(dev, gpu, ctx->aspace, queue, args->nr_bos,
+	submit = submit_create(dev, gpu, queue, args->nr_bos,
 		args->nr_cmds);
 	if (!submit) {
 		ret = -ENOMEM;
@@ -785,7 +785,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		}
 	}
 
-	msm_gpu_submit(gpu, submit, ctx);
+	msm_gpu_submit(gpu, submit);
 
 	args->fence = submit->fence->seqno;
 
diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 5f6a11211b64..f914ddbaea89 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -17,6 +17,7 @@ msm_gem_address_space_destroy(struct kref *kref)
 	drm_mm_takedown(&aspace->mm);
 	if (aspace->mmu)
 		aspace->mmu->funcs->destroy(aspace->mmu);
+	put_pid(aspace->pid);
 	kfree(aspace);
 }
 
@@ -27,6 +28,15 @@ void msm_gem_address_space_put(struct msm_gem_address_space *aspace)
 		kref_put(&aspace->kref, msm_gem_address_space_destroy);
 }
 
+struct msm_gem_address_space *
+msm_gem_address_space_get(struct msm_gem_address_space *aspace)
+{
+	if (!IS_ERR_OR_NULL(aspace))
+		kref_get(&aspace->kref);
+
+	return aspace;
+}
+
 /* Actually unmap memory for the vma */
 void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma)
@@ -78,8 +88,10 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace,
 		ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt,
 				size, prot);
 
-	if (ret)
+	if (ret) {
 		vma->mapped = false;
+		vma->inuse--;
+	}
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 57ddc9438351..55d16489d0f3 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -24,7 +24,7 @@
 static int msm_devfreq_target(struct device *dev, unsigned long *freq,
 		u32 flags)
 {
-	struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev));
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 	struct dev_pm_opp *opp;
 
 	opp = devfreq_recommended_opp(dev, freq, flags);
@@ -32,6 +32,8 @@ static int msm_devfreq_target(struct device *dev, unsigned long *freq,
 	if (IS_ERR(opp))
 		return PTR_ERR(opp);
 
+	trace_msm_gpu_freq_change(dev_pm_opp_get_freq(opp));
+
 	if (gpu->funcs->gpu_set_freq)
 		gpu->funcs->gpu_set_freq(gpu, opp);
 	else
@@ -45,7 +47,7 @@ static int msm_devfreq_target(struct device *dev, unsigned long *freq,
 static int msm_devfreq_get_dev_status(struct device *dev,
 		struct devfreq_dev_status *status)
 {
-	struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev));
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 	ktime_t time;
 
 	if (gpu->funcs->gpu_get_freq)
@@ -64,7 +66,7 @@ static int msm_devfreq_get_dev_status(struct device *dev,
 
 static int msm_devfreq_get_cur_freq(struct device *dev, unsigned long *freq)
 {
-	struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev));
+	struct msm_gpu *gpu = dev_to_gpu(dev);
 
 	if (gpu->funcs->gpu_get_freq)
 		*freq = gpu->funcs->gpu_get_freq(gpu);
@@ -200,6 +202,7 @@ int msm_gpu_pm_resume(struct msm_gpu *gpu)
 	int ret;
 
 	DBG("%s", gpu->name);
+	trace_msm_gpu_resume(0);
 
 	ret = enable_pwrrail(gpu);
 	if (ret)
@@ -225,6 +228,7 @@ int msm_gpu_pm_suspend(struct msm_gpu *gpu)
 	int ret;
 
 	DBG("%s", gpu->name);
+	trace_msm_gpu_suspend(0);
 
 	devfreq_suspend_device(gpu->devfreq.devfreq);
 
@@ -520,7 +524,7 @@ static void recover_worker(struct work_struct *work)
 			struct msm_ringbuffer *ring = gpu->rb[i];
 
 			list_for_each_entry(submit, &ring->submits, node)
-				gpu->funcs->submit(gpu, submit, NULL);
+				gpu->funcs->submit(gpu, submit);
 		}
 	}
 
@@ -694,8 +698,8 @@ static void retire_submit(struct msm_gpu *gpu, struct msm_ringbuffer *ring,
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
-		/* move to inactive: */
-		msm_gem_move_to_inactive(&msm_obj->base);
+
+		msm_gem_active_put(&msm_obj->base);
 		msm_gem_unpin_iova(&msm_obj->base, submit->aspace);
 		drm_gem_object_put_locked(&msm_obj->base);
 	}
@@ -747,8 +751,7 @@ void msm_gpu_retire(struct msm_gpu *gpu)
 }
 
 /* add bo's to gpu's ring, and kick gpu: */
-void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-		struct msm_file_private *ctx)
+void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 {
 	struct drm_device *dev = gpu->dev;
 	struct msm_drm_private *priv = dev->dev_private;
@@ -771,6 +774,7 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
+		struct drm_gem_object *drm_obj = &msm_obj->base;
 		uint64_t iova;
 
 		/* can't happen yet.. but when we add 2d support we'll have
@@ -783,13 +787,15 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		msm_gem_get_and_pin_iova(&msm_obj->base, submit->aspace, &iova);
 
 		if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE)
-			msm_gem_move_to_active(&msm_obj->base, gpu, true, submit->fence);
+			dma_resv_add_excl_fence(drm_obj->resv, submit->fence);
 		else if (submit->bos[i].flags & MSM_SUBMIT_BO_READ)
-			msm_gem_move_to_active(&msm_obj->base, gpu, false, submit->fence);
+			dma_resv_add_shared_fence(drm_obj->resv, submit->fence);
+
+		msm_gem_active_get(drm_obj, gpu);
 	}
 
-	gpu->funcs->submit(gpu, submit, ctx);
-	priv->lastctx = ctx;
+	gpu->funcs->submit(gpu, submit);
+	priv->lastctx = submit->queue->ctx;
 
 	hangcheck_timer_reset(gpu);
 }
@@ -824,6 +830,30 @@ static int get_clocks(struct platform_device *pdev, struct msm_gpu *gpu)
 	return 0;
 }
 
+/* Return a new address space for a msm_drm_private instance */
+struct msm_gem_address_space *
+msm_gpu_create_private_address_space(struct msm_gpu *gpu, struct task_struct *task)
+{
+	struct msm_gem_address_space *aspace = NULL;
+	if (!gpu)
+		return NULL;
+
+	/*
+	 * If the target doesn't support private address spaces then return
+	 * the global one
+	 */
+	if (gpu->funcs->create_private_address_space) {
+		aspace = gpu->funcs->create_private_address_space(gpu);
+		if (!IS_ERR(aspace))
+			aspace->pid = get_pid(task_pid(task));
+	}
+
+	if (IS_ERR_OR_NULL(aspace))
+		aspace = msm_gem_address_space_get(gpu->aspace);
+
+	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)
@@ -892,7 +922,7 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 		gpu->gpu_cx = NULL;
 
 	gpu->pdev = pdev;
-	platform_set_drvdata(pdev, gpu);
+	platform_set_drvdata(pdev, &gpu->adreno_smmu);
 
 	msm_devfreq_init(gpu);
 
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index 37cffac4cbe3..6c9e1fdc1a76 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -7,6 +7,7 @@
 #ifndef __MSM_GPU_H__
 #define __MSM_GPU_H__
 
+#include <linux/adreno-smmu-priv.h>
 #include <linux/clk.h>
 #include <linux/interconnect.h>
 #include <linux/pm_opp.h>
@@ -45,8 +46,7 @@ struct msm_gpu_funcs {
 	int (*hw_init)(struct msm_gpu *gpu);
 	int (*pm_suspend)(struct msm_gpu *gpu);
 	int (*pm_resume)(struct msm_gpu *gpu);
-	void (*submit)(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-			struct msm_file_private *ctx);
+	void (*submit)(struct msm_gpu *gpu, struct msm_gem_submit *submit);
 	void (*flush)(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
 	irqreturn_t (*irq)(struct msm_gpu *irq);
 	struct msm_ringbuffer *(*active_ring)(struct msm_gpu *gpu);
@@ -66,6 +66,9 @@ struct msm_gpu_funcs {
 	void (*gpu_set_freq)(struct msm_gpu *gpu, struct dev_pm_opp *opp);
 	struct msm_gem_address_space *(*create_address_space)
 		(struct msm_gpu *gpu, struct platform_device *pdev);
+	struct msm_gem_address_space *(*create_private_address_space)
+		(struct msm_gpu *gpu);
+	uint32_t (*get_rptr)(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
 };
 
 struct msm_gpu {
@@ -74,6 +77,8 @@ struct msm_gpu {
 	struct platform_device *pdev;
 	const struct msm_gpu_funcs *funcs;
 
+	struct adreno_smmu_priv adreno_smmu;
+
 	/* performance counters (hw & sw): */
 	spinlock_t perf_lock;
 	bool perfcntr_active;
@@ -144,6 +149,12 @@ struct msm_gpu {
 	bool hw_apriv;
 };
 
+static inline struct msm_gpu *dev_to_gpu(struct device *dev)
+{
+	struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(dev);
+	return container_of(adreno_smmu, struct msm_gpu, adreno_smmu);
+}
+
 /* It turns out that all targets use the same ringbuffer size */
 #define MSM_GPU_RINGBUFFER_SZ SZ_32K
 #define MSM_GPU_RINGBUFFER_BLKSIZE 32
@@ -184,6 +195,7 @@ struct msm_gpu_submitqueue {
 	u32 flags;
 	u32 prio;
 	int faults;
+	struct msm_file_private *ctx;
 	struct list_head node;
 	struct kref ref;
 };
@@ -283,13 +295,15 @@ int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
 		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs);
 
 void msm_gpu_retire(struct msm_gpu *gpu);
-void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
-		struct msm_file_private *ctx);
+void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit);
 
 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 msm_gem_address_space *
+msm_gpu_create_private_address_space(struct msm_gpu *gpu, struct task_struct *task);
+
 void msm_gpu_cleanup(struct msm_gpu *gpu);
 
 struct msm_gpu *adreno_load_gpu(struct drm_device *dev);
diff --git a/drivers/gpu/drm/msm/msm_gpu_trace.h b/drivers/gpu/drm/msm/msm_gpu_trace.h
index 122b84789238..03e0c2536b94 100644
--- a/drivers/gpu/drm/msm/msm_gpu_trace.h
+++ b/drivers/gpu/drm/msm/msm_gpu_trace.h
@@ -83,6 +83,89 @@ TRACE_EVENT(msm_gpu_submit_retired,
 		    __entry->start_ticks, __entry->end_ticks)
 );
 
+
+TRACE_EVENT(msm_gpu_freq_change,
+		TP_PROTO(u32 freq),
+		TP_ARGS(freq),
+		TP_STRUCT__entry(
+			__field(u32, freq)
+			),
+		TP_fast_assign(
+			/* trace freq in MHz to match intel_gpu_freq_change, to make life easier
+			 * for userspace
+			 */
+			__entry->freq = DIV_ROUND_UP(freq, 1000000);
+			),
+		TP_printk("new_freq=%u", __entry->freq)
+);
+
+
+TRACE_EVENT(msm_gmu_freq_change,
+		TP_PROTO(u32 freq, u32 perf_index),
+		TP_ARGS(freq, perf_index),
+		TP_STRUCT__entry(
+			__field(u32, freq)
+			__field(u32, perf_index)
+			),
+		TP_fast_assign(
+			__entry->freq = freq;
+			__entry->perf_index = perf_index;
+			),
+		TP_printk("freq=%u, perf_index=%u", __entry->freq, __entry->perf_index)
+);
+
+
+TRACE_EVENT(msm_gem_purge,
+		TP_PROTO(u32 bytes),
+		TP_ARGS(bytes),
+		TP_STRUCT__entry(
+			__field(u32, bytes)
+			),
+		TP_fast_assign(
+			__entry->bytes = bytes;
+			),
+		TP_printk("Purging %u bytes", __entry->bytes)
+);
+
+
+TRACE_EVENT(msm_gem_purge_vmaps,
+		TP_PROTO(u32 unmapped),
+		TP_ARGS(unmapped),
+		TP_STRUCT__entry(
+			__field(u32, unmapped)
+			),
+		TP_fast_assign(
+			__entry->unmapped = unmapped;
+			),
+		TP_printk("Purging %u vmaps", __entry->unmapped)
+);
+
+
+TRACE_EVENT(msm_gpu_suspend,
+		TP_PROTO(int dummy),
+		TP_ARGS(dummy),
+		TP_STRUCT__entry(
+			__field(u32, dummy)
+			),
+		TP_fast_assign(
+			__entry->dummy = dummy;
+			),
+		TP_printk("%u", __entry->dummy)
+);
+
+
+TRACE_EVENT(msm_gpu_resume,
+		TP_PROTO(int dummy),
+		TP_ARGS(dummy),
+		TP_STRUCT__entry(
+			__field(u32, dummy)
+			),
+		TP_fast_assign(
+			__entry->dummy = dummy;
+			),
+		TP_printk("%u", __entry->dummy)
+);
+
 #endif
 
 #undef TRACE_INCLUDE_PATH
diff --git a/drivers/gpu/drm/msm/msm_gpummu.c b/drivers/gpu/drm/msm/msm_gpummu.c
index 310a31b05faa..379496186c7f 100644
--- a/drivers/gpu/drm/msm/msm_gpummu.c
+++ b/drivers/gpu/drm/msm/msm_gpummu.c
@@ -30,21 +30,20 @@ static int msm_gpummu_map(struct msm_mmu *mmu, uint64_t iova,
 {
 	struct msm_gpummu *gpummu = to_msm_gpummu(mmu);
 	unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE;
-	struct scatterlist *sg;
+	struct sg_dma_page_iter dma_iter;
 	unsigned prot_bits = 0;
-	unsigned i, j;
 
 	if (prot & IOMMU_WRITE)
 		prot_bits |= 1;
 	if (prot & IOMMU_READ)
 		prot_bits |= 2;
 
-	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
-		dma_addr_t addr = sg->dma_address;
-		for (j = 0; j < sg->length / GPUMMU_PAGE_SIZE; j++, idx++) {
-			gpummu->table[idx] = addr | prot_bits;
-			addr += GPUMMU_PAGE_SIZE;
-		}
+	for_each_sgtable_dma_page(sgt, &dma_iter, 0) {
+		dma_addr_t addr = sg_page_iter_dma_address(&dma_iter);
+		int i;
+
+		for (i = 0; i < PAGE_SIZE; i += GPUMMU_PAGE_SIZE)
+			gpummu->table[idx++] = (addr + i) | prot_bits;
 	}
 
 	/* we can improve by deferring flush for multiple map() */
@@ -102,7 +101,7 @@ struct msm_mmu *msm_gpummu_new(struct device *dev, struct msm_gpu *gpu)
 	}
 
 	gpummu->gpu = gpu;
-	msm_mmu_init(&gpummu->base, dev, &funcs);
+	msm_mmu_init(&gpummu->base, dev, &funcs, MSM_MMU_GPUMMU);
 
 	return &gpummu->base;
 }
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 3a381a9674c9..22ac7c692a81 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -4,15 +4,210 @@
  * Author: Rob Clark <robdclark@gmail.com>
  */
 
+#include <linux/adreno-smmu-priv.h>
+#include <linux/io-pgtable.h>
 #include "msm_drv.h"
 #include "msm_mmu.h"
 
 struct msm_iommu {
 	struct msm_mmu base;
 	struct iommu_domain *domain;
+	atomic_t pagetables;
 };
+
 #define to_msm_iommu(x) container_of(x, struct msm_iommu, base)
 
+struct msm_iommu_pagetable {
+	struct msm_mmu base;
+	struct msm_mmu *parent;
+	struct io_pgtable_ops *pgtbl_ops;
+	phys_addr_t ttbr;
+	u32 asid;
+};
+static struct msm_iommu_pagetable *to_pagetable(struct msm_mmu *mmu)
+{
+	return container_of(mmu, struct msm_iommu_pagetable, base);
+}
+
+static int msm_iommu_pagetable_unmap(struct msm_mmu *mmu, u64 iova,
+		size_t size)
+{
+	struct msm_iommu_pagetable *pagetable = to_pagetable(mmu);
+	struct io_pgtable_ops *ops = pagetable->pgtbl_ops;
+	size_t unmapped = 0;
+
+	/* Unmap the block one page at a time */
+	while (size) {
+		unmapped += ops->unmap(ops, iova, 4096, NULL);
+		iova += 4096;
+		size -= 4096;
+	}
+
+	iommu_flush_iotlb_all(to_msm_iommu(pagetable->parent)->domain);
+
+	return (unmapped == size) ? 0 : -EINVAL;
+}
+
+static int msm_iommu_pagetable_map(struct msm_mmu *mmu, u64 iova,
+		struct sg_table *sgt, size_t len, int prot)
+{
+	struct msm_iommu_pagetable *pagetable = to_pagetable(mmu);
+	struct io_pgtable_ops *ops = pagetable->pgtbl_ops;
+	struct scatterlist *sg;
+	size_t mapped = 0;
+	u64 addr = iova;
+	unsigned int i;
+
+	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
+		size_t size = sg->length;
+		phys_addr_t phys = sg_phys(sg);
+
+		/* Map the block one page at a time */
+		while (size) {
+			if (ops->map(ops, addr, phys, 4096, prot, GFP_KERNEL)) {
+				msm_iommu_pagetable_unmap(mmu, iova, mapped);
+				return -EINVAL;
+			}
+
+			phys += 4096;
+			addr += 4096;
+			size -= 4096;
+			mapped += 4096;
+		}
+	}
+
+	return 0;
+}
+
+static void msm_iommu_pagetable_destroy(struct msm_mmu *mmu)
+{
+	struct msm_iommu_pagetable *pagetable = to_pagetable(mmu);
+	struct msm_iommu *iommu = to_msm_iommu(pagetable->parent);
+	struct adreno_smmu_priv *adreno_smmu =
+		dev_get_drvdata(pagetable->parent->dev);
+
+	/*
+	 * If this is the last attached pagetable for the parent,
+	 * disable TTBR0 in the arm-smmu driver
+	 */
+	if (atomic_dec_return(&iommu->pagetables) == 0)
+		adreno_smmu->set_ttbr0_cfg(adreno_smmu->cookie, NULL);
+
+	free_io_pgtable_ops(pagetable->pgtbl_ops);
+	kfree(pagetable);
+}
+
+int msm_iommu_pagetable_params(struct msm_mmu *mmu,
+		phys_addr_t *ttbr, int *asid)
+{
+	struct msm_iommu_pagetable *pagetable;
+
+	if (mmu->type != MSM_MMU_IOMMU_PAGETABLE)
+		return -EINVAL;
+
+	pagetable = to_pagetable(mmu);
+
+	if (ttbr)
+		*ttbr = pagetable->ttbr;
+
+	if (asid)
+		*asid = pagetable->asid;
+
+	return 0;
+}
+
+static const struct msm_mmu_funcs pagetable_funcs = {
+		.map = msm_iommu_pagetable_map,
+		.unmap = msm_iommu_pagetable_unmap,
+		.destroy = msm_iommu_pagetable_destroy,
+};
+
+static void msm_iommu_tlb_flush_all(void *cookie)
+{
+}
+
+static void msm_iommu_tlb_flush_walk(unsigned long iova, size_t size,
+		size_t granule, void *cookie)
+{
+}
+
+static void msm_iommu_tlb_add_page(struct iommu_iotlb_gather *gather,
+		unsigned long iova, size_t granule, void *cookie)
+{
+}
+
+static const struct iommu_flush_ops null_tlb_ops = {
+	.tlb_flush_all = msm_iommu_tlb_flush_all,
+	.tlb_flush_walk = msm_iommu_tlb_flush_walk,
+	.tlb_flush_leaf = msm_iommu_tlb_flush_walk,
+	.tlb_add_page = msm_iommu_tlb_add_page,
+};
+
+struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent)
+{
+	struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(parent->dev);
+	struct msm_iommu *iommu = to_msm_iommu(parent);
+	struct msm_iommu_pagetable *pagetable;
+	const struct io_pgtable_cfg *ttbr1_cfg = NULL;
+	struct io_pgtable_cfg ttbr0_cfg;
+	int ret;
+
+	/* Get the pagetable configuration from the domain */
+	if (adreno_smmu->cookie)
+		ttbr1_cfg = adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie);
+	if (!ttbr1_cfg)
+		return ERR_PTR(-ENODEV);
+
+	pagetable = kzalloc(sizeof(*pagetable), GFP_KERNEL);
+	if (!pagetable)
+		return ERR_PTR(-ENOMEM);
+
+	msm_mmu_init(&pagetable->base, parent->dev, &pagetable_funcs,
+		MSM_MMU_IOMMU_PAGETABLE);
+
+	/* Clone the TTBR1 cfg as starting point for TTBR0 cfg: */
+	ttbr0_cfg = *ttbr1_cfg;
+
+	/* The incoming cfg will have the TTBR1 quirk enabled */
+	ttbr0_cfg.quirks &= ~IO_PGTABLE_QUIRK_ARM_TTBR1;
+	ttbr0_cfg.tlb = &null_tlb_ops;
+
+	pagetable->pgtbl_ops = alloc_io_pgtable_ops(ARM_64_LPAE_S1,
+		&ttbr0_cfg, iommu->domain);
+
+	if (!pagetable->pgtbl_ops) {
+		kfree(pagetable);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	/*
+	 * If this is the first pagetable that we've allocated, send it back to
+	 * the arm-smmu driver as a trigger to set up TTBR0
+	 */
+	if (atomic_inc_return(&iommu->pagetables) == 1) {
+		ret = adreno_smmu->set_ttbr0_cfg(adreno_smmu->cookie, &ttbr0_cfg);
+		if (ret) {
+			free_io_pgtable_ops(pagetable->pgtbl_ops);
+			kfree(pagetable);
+			return ERR_PTR(ret);
+		}
+	}
+
+	/* Needed later for TLB flush */
+	pagetable->parent = parent;
+	pagetable->ttbr = ttbr0_cfg.arm_lpae_s1_cfg.ttbr;
+
+	/*
+	 * TODO we would like each set of page tables to have a unique ASID
+	 * to optimize TLB invalidation.  But iommu_flush_iotlb_all() will
+	 * end up flushing the ASID used for TTBR1 pagetables, which is not
+	 * what we want.  So for now just use the same ASID as TTBR1.
+	 */
+	pagetable->asid = 0;
+
+	return &pagetable->base;
+}
+
 static int msm_fault_handler(struct iommu_domain *domain, struct device *dev,
 		unsigned long iova, int flags, void *arg)
 {
@@ -36,7 +231,11 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	size_t ret;
 
-	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
+	/* The arm-smmu driver expects the addresses to be sign extended */
+	if (iova & BIT_ULL(48))
+		iova |= GENMASK_ULL(63, 49);
+
+	ret = iommu_map_sgtable(iommu->domain, iova, sgt, prot);
 	WARN_ON(!ret);
 
 	return (ret == len) ? 0 : -EINVAL;
@@ -46,6 +245,9 @@ static int msm_iommu_unmap(struct msm_mmu *mmu, uint64_t iova, size_t len)
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 
+	if (iova & BIT_ULL(48))
+		iova |= GENMASK_ULL(63, 49);
+
 	iommu_unmap(iommu->domain, iova, len);
 
 	return 0;
@@ -78,9 +280,11 @@ struct msm_mmu *msm_iommu_new(struct device *dev, struct iommu_domain *domain)
 		return ERR_PTR(-ENOMEM);
 
 	iommu->domain = domain;
-	msm_mmu_init(&iommu->base, dev, &funcs);
+	msm_mmu_init(&iommu->base, dev, &funcs, MSM_MMU_IOMMU);
 	iommu_set_fault_handler(domain, msm_fault_handler, iommu);
 
+	atomic_set(&iommu->pagetables, 0);
+
 	ret = iommu_attach_device(iommu->domain, dev);
 	if (ret) {
 		kfree(iommu);
diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
index 3a534ee59bf6..61ade89d9e48 100644
--- a/drivers/gpu/drm/msm/msm_mmu.h
+++ b/drivers/gpu/drm/msm/msm_mmu.h
@@ -17,18 +17,26 @@ struct msm_mmu_funcs {
 	void (*destroy)(struct msm_mmu *mmu);
 };
 
+enum msm_mmu_type {
+	MSM_MMU_GPUMMU,
+	MSM_MMU_IOMMU,
+	MSM_MMU_IOMMU_PAGETABLE,
+};
+
 struct msm_mmu {
 	const struct msm_mmu_funcs *funcs;
 	struct device *dev;
 	int (*handler)(void *arg, unsigned long iova, int flags);
 	void *arg;
+	enum msm_mmu_type type;
 };
 
 static inline void msm_mmu_init(struct msm_mmu *mmu, struct device *dev,
-		const struct msm_mmu_funcs *funcs)
+		const struct msm_mmu_funcs *funcs, enum msm_mmu_type type)
 {
 	mmu->dev = dev;
 	mmu->funcs = funcs;
+	mmu->type = type;
 }
 
 struct msm_mmu *msm_iommu_new(struct device *dev, struct iommu_domain *domain);
@@ -41,7 +49,13 @@ static inline void msm_mmu_set_fault_handler(struct msm_mmu *mmu, void *arg,
 	mmu->handler = handler;
 }
 
+struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent);
+
 void msm_gpummu_params(struct msm_mmu *mmu, dma_addr_t *pt_base,
 		dma_addr_t *tran_error);
 
+
+int msm_iommu_pagetable_params(struct msm_mmu *mmu, phys_addr_t *ttbr,
+		int *asid);
+
 #endif /* __MSM_MMU_H__ */
diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.h b/drivers/gpu/drm/msm/msm_ringbuffer.h
index 7764373d0ed2..0987d6bf848c 100644
--- a/drivers/gpu/drm/msm/msm_ringbuffer.h
+++ b/drivers/gpu/drm/msm/msm_ringbuffer.h
@@ -31,6 +31,7 @@ struct msm_rbmemptrs {
 	volatile uint32_t fence;
 
 	volatile struct msm_gpu_submit_stats stats[MSM_GPU_SUBMIT_STATS_COUNT];
+	volatile u64 ttbr0;
 };
 
 struct msm_ringbuffer {
diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c
index a1d94be7883a..c3d206105d28 100644
--- a/drivers/gpu/drm/msm/msm_submitqueue.c
+++ b/drivers/gpu/drm/msm/msm_submitqueue.c
@@ -12,6 +12,8 @@ void msm_submitqueue_destroy(struct kref *kref)
 	struct msm_gpu_submitqueue *queue = container_of(kref,
 		struct msm_gpu_submitqueue, ref);
 
+	msm_file_private_put(queue->ctx);
+
 	kfree(queue);
 }
 
@@ -49,8 +51,10 @@ void msm_submitqueue_close(struct msm_file_private *ctx)
 	 * No lock needed in close and there won't
 	 * be any more user ioctls coming our way
 	 */
-	list_for_each_entry_safe(entry, tmp, &ctx->submitqueues, node)
+	list_for_each_entry_safe(entry, tmp, &ctx->submitqueues, node) {
+		list_del(&entry->node);
 		msm_submitqueue_put(entry);
+	}
 }
 
 int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx,
@@ -81,6 +85,7 @@ int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx,
 
 	write_lock(&ctx->queuelock);
 
+	queue->ctx = msm_file_private_get(ctx);
 	queue->id = ctx->queueid++;
 
 	if (id)
diff --git a/drivers/gpu/drm/mxsfb/Kconfig b/drivers/gpu/drm/mxsfb/Kconfig
index 0dca8f27169e..0143d539f8f8 100644
--- a/drivers/gpu/drm/mxsfb/Kconfig
+++ b/drivers/gpu/drm/mxsfb/Kconfig
@@ -5,7 +5,7 @@ config DRM_MXS
 	  Choose this option to select drivers for MXS FB devices
 
 config DRM_MXSFB
-	tristate "i.MX23/i.MX28/i.MX6SX MXSFB LCD controller"
+	tristate "i.MX (e)LCDIF LCD controller"
 	depends on DRM && OF
 	depends on COMMON_CLK
 	select DRM_MXS
@@ -13,8 +13,10 @@ config DRM_MXSFB
 	select DRM_KMS_FB_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_PANEL
+	select DRM_PANEL_BRIDGE
 	help
-	  Choose this option if you have an i.MX23/i.MX28/i.MX6SX MXSFB
-	  LCD controller.
+	  Choose this option if you have an LCDIF or eLCDIF LCD controller.
+	  Those devices are found in various i.MX SoC (including i.MX23,
+	  i.MX28, i.MX6SX, i.MX7 and i.MX8M).
 
 	  If M is selected the module will be called mxsfb.
diff --git a/drivers/gpu/drm/mxsfb/Makefile b/drivers/gpu/drm/mxsfb/Makefile
index ff6e358088fa..26d153896d72 100644
--- a/drivers/gpu/drm/mxsfb/Makefile
+++ b/drivers/gpu/drm/mxsfb/Makefile
@@ -1,3 +1,3 @@
 # SPDX-License-Identifier: GPL-2.0-only
-mxsfb-y := mxsfb_drv.o mxsfb_crtc.o mxsfb_out.o
+mxsfb-y := mxsfb_drv.o mxsfb_kms.o
 obj-$(CONFIG_DRM_MXSFB)	+= mxsfb.o
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c b/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
deleted file mode 100644
index b69ace8bf526..000000000000
--- a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
+++ /dev/null
@@ -1,343 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * Copyright (C) 2016 Marek Vasut <marex@denx.de>
- *
- * This code is based on drivers/video/fbdev/mxsfb.c :
- * Copyright (C) 2010 Juergen Beisert, Pengutronix
- * Copyright (C) 2008-2009 Freescale Semiconductor, Inc. All Rights Reserved.
- * Copyright (C) 2008 Embedded Alley Solutions, Inc All Rights Reserved.
- */
-
-#include <linux/clk.h>
-#include <linux/iopoll.h>
-#include <linux/of_graph.h>
-#include <linux/platform_data/simplefb.h>
-
-#include <video/videomode.h>
-
-#include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc.h>
-#include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_gem_cma_helper.h>
-#include <drm/drm_of.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
-#include <drm/drm_vblank.h>
-
-#include "mxsfb_drv.h"
-#include "mxsfb_regs.h"
-
-#define MXS_SET_ADDR		0x4
-#define MXS_CLR_ADDR		0x8
-#define MODULE_CLKGATE		BIT(30)
-#define MODULE_SFTRST		BIT(31)
-/* 1 second delay should be plenty of time for block reset */
-#define RESET_TIMEOUT		1000000
-
-static u32 set_hsync_pulse_width(struct mxsfb_drm_private *mxsfb, u32 val)
-{
-	return (val & mxsfb->devdata->hs_wdth_mask) <<
-		mxsfb->devdata->hs_wdth_shift;
-}
-
-/* Setup the MXSFB registers for decoding the pixels out of the framebuffer */
-static int mxsfb_set_pixel_fmt(struct mxsfb_drm_private *mxsfb)
-{
-	struct drm_crtc *crtc = &mxsfb->pipe.crtc;
-	struct drm_device *drm = crtc->dev;
-	const u32 format = crtc->primary->state->fb->format->format;
-	u32 ctrl, ctrl1;
-
-	ctrl = CTRL_BYPASS_COUNT | CTRL_MASTER;
-
-	/*
-	 * WARNING: The bus width, CTRL_SET_BUS_WIDTH(), is configured to
-	 * match the selected mode here. This differs from the original
-	 * MXSFB driver, which had the option to configure the bus width
-	 * to arbitrary value. This limitation should not pose an issue.
-	 */
-
-	/* CTRL1 contains IRQ config and status bits, preserve those. */
-	ctrl1 = readl(mxsfb->base + LCDC_CTRL1);
-	ctrl1 &= CTRL1_CUR_FRAME_DONE_IRQ_EN | CTRL1_CUR_FRAME_DONE_IRQ;
-
-	switch (format) {
-	case DRM_FORMAT_RGB565:
-		dev_dbg(drm->dev, "Setting up RGB565 mode\n");
-		ctrl |= CTRL_SET_WORD_LENGTH(0);
-		ctrl1 |= CTRL1_SET_BYTE_PACKAGING(0xf);
-		break;
-	case DRM_FORMAT_XRGB8888:
-		dev_dbg(drm->dev, "Setting up XRGB8888 mode\n");
-		ctrl |= CTRL_SET_WORD_LENGTH(3);
-		/* Do not use packed pixels = one pixel per word instead. */
-		ctrl1 |= CTRL1_SET_BYTE_PACKAGING(0x7);
-		break;
-	default:
-		dev_err(drm->dev, "Unhandled pixel format %08x\n", format);
-		return -EINVAL;
-	}
-
-	writel(ctrl1, mxsfb->base + LCDC_CTRL1);
-	writel(ctrl, mxsfb->base + LCDC_CTRL);
-
-	return 0;
-}
-
-static void mxsfb_set_bus_fmt(struct mxsfb_drm_private *mxsfb)
-{
-	struct drm_crtc *crtc = &mxsfb->pipe.crtc;
-	struct drm_device *drm = crtc->dev;
-	u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24;
-	u32 reg;
-
-	reg = readl(mxsfb->base + LCDC_CTRL);
-
-	if (mxsfb->connector->display_info.num_bus_formats)
-		bus_format = mxsfb->connector->display_info.bus_formats[0];
-
-	DRM_DEV_DEBUG_DRIVER(drm->dev, "Using bus_format: 0x%08X\n",
-			     bus_format);
-
-	reg &= ~CTRL_BUS_WIDTH_MASK;
-	switch (bus_format) {
-	case MEDIA_BUS_FMT_RGB565_1X16:
-		reg |= CTRL_SET_BUS_WIDTH(STMLCDIF_16BIT);
-		break;
-	case MEDIA_BUS_FMT_RGB666_1X18:
-		reg |= CTRL_SET_BUS_WIDTH(STMLCDIF_18BIT);
-		break;
-	case MEDIA_BUS_FMT_RGB888_1X24:
-		reg |= CTRL_SET_BUS_WIDTH(STMLCDIF_24BIT);
-		break;
-	default:
-		dev_err(drm->dev, "Unknown media bus format %d\n", bus_format);
-		break;
-	}
-	writel(reg, mxsfb->base + LCDC_CTRL);
-}
-
-static void mxsfb_enable_controller(struct mxsfb_drm_private *mxsfb)
-{
-	u32 reg;
-
-	if (mxsfb->clk_disp_axi)
-		clk_prepare_enable(mxsfb->clk_disp_axi);
-	clk_prepare_enable(mxsfb->clk);
-
-	/* If it was disabled, re-enable the mode again */
-	writel(CTRL_DOTCLK_MODE, mxsfb->base + LCDC_CTRL + REG_SET);
-
-	/* Enable the SYNC signals first, then the DMA engine */
-	reg = readl(mxsfb->base + LCDC_VDCTRL4);
-	reg |= VDCTRL4_SYNC_SIGNALS_ON;
-	writel(reg, mxsfb->base + LCDC_VDCTRL4);
-
-	writel(CTRL_RUN, mxsfb->base + LCDC_CTRL + REG_SET);
-}
-
-static void mxsfb_disable_controller(struct mxsfb_drm_private *mxsfb)
-{
-	u32 reg;
-
-	/*
-	 * Even if we disable the controller here, it will still continue
-	 * until its FIFOs are running out of data
-	 */
-	writel(CTRL_DOTCLK_MODE, mxsfb->base + LCDC_CTRL + REG_CLR);
-
-	readl_poll_timeout(mxsfb->base + LCDC_CTRL, reg, !(reg & CTRL_RUN),
-			   0, 1000);
-
-	reg = readl(mxsfb->base + LCDC_VDCTRL4);
-	reg &= ~VDCTRL4_SYNC_SIGNALS_ON;
-	writel(reg, mxsfb->base + LCDC_VDCTRL4);
-
-	clk_disable_unprepare(mxsfb->clk);
-	if (mxsfb->clk_disp_axi)
-		clk_disable_unprepare(mxsfb->clk_disp_axi);
-}
-
-/*
- * Clear the bit and poll it cleared.  This is usually called with
- * a reset address and mask being either SFTRST(bit 31) or CLKGATE
- * (bit 30).
- */
-static int clear_poll_bit(void __iomem *addr, u32 mask)
-{
-	u32 reg;
-
-	writel(mask, addr + MXS_CLR_ADDR);
-	return readl_poll_timeout(addr, reg, !(reg & mask), 0, RESET_TIMEOUT);
-}
-
-static int mxsfb_reset_block(void __iomem *reset_addr)
-{
-	int ret;
-
-	ret = clear_poll_bit(reset_addr, MODULE_SFTRST);
-	if (ret)
-		return ret;
-
-	writel(MODULE_CLKGATE, reset_addr + MXS_CLR_ADDR);
-
-	ret = clear_poll_bit(reset_addr, MODULE_SFTRST);
-	if (ret)
-		return ret;
-
-	return clear_poll_bit(reset_addr, MODULE_CLKGATE);
-}
-
-static dma_addr_t mxsfb_get_fb_paddr(struct mxsfb_drm_private *mxsfb)
-{
-	struct drm_framebuffer *fb = mxsfb->pipe.plane.state->fb;
-	struct drm_gem_cma_object *gem;
-
-	if (!fb)
-		return 0;
-
-	gem = drm_fb_cma_get_gem_obj(fb, 0);
-	if (!gem)
-		return 0;
-
-	return gem->paddr;
-}
-
-static void mxsfb_crtc_mode_set_nofb(struct mxsfb_drm_private *mxsfb)
-{
-	struct drm_device *drm = mxsfb->pipe.crtc.dev;
-	struct drm_display_mode *m = &mxsfb->pipe.crtc.state->adjusted_mode;
-	u32 bus_flags = mxsfb->connector->display_info.bus_flags;
-	u32 vdctrl0, vsync_pulse_len, hsync_pulse_len;
-	int err;
-
-	/*
-	 * It seems, you can't re-program the controller if it is still
-	 * running. This may lead to shifted pictures (FIFO issue?), so
-	 * first stop the controller and drain its FIFOs.
-	 */
-
-	/* Mandatory eLCDIF reset as per the Reference Manual */
-	err = mxsfb_reset_block(mxsfb->base);
-	if (err)
-		return;
-
-	/* Clear the FIFOs */
-	writel(CTRL1_FIFO_CLEAR, mxsfb->base + LCDC_CTRL1 + REG_SET);
-
-	err = mxsfb_set_pixel_fmt(mxsfb);
-	if (err)
-		return;
-
-	clk_set_rate(mxsfb->clk, m->crtc_clock * 1000);
-
-	if (mxsfb->bridge && mxsfb->bridge->timings)
-		bus_flags = mxsfb->bridge->timings->input_bus_flags;
-
-	DRM_DEV_DEBUG_DRIVER(drm->dev, "Pixel clock: %dkHz (actual: %dkHz)\n",
-			     m->crtc_clock,
-			     (int)(clk_get_rate(mxsfb->clk) / 1000));
-	DRM_DEV_DEBUG_DRIVER(drm->dev, "Connector bus_flags: 0x%08X\n",
-			     bus_flags);
-	DRM_DEV_DEBUG_DRIVER(drm->dev, "Mode flags: 0x%08X\n", m->flags);
-
-	writel(TRANSFER_COUNT_SET_VCOUNT(m->crtc_vdisplay) |
-	       TRANSFER_COUNT_SET_HCOUNT(m->crtc_hdisplay),
-	       mxsfb->base + mxsfb->devdata->transfer_count);
-
-	vsync_pulse_len = m->crtc_vsync_end - m->crtc_vsync_start;
-
-	vdctrl0 = VDCTRL0_ENABLE_PRESENT |	/* Always in DOTCLOCK mode */
-		  VDCTRL0_VSYNC_PERIOD_UNIT |
-		  VDCTRL0_VSYNC_PULSE_WIDTH_UNIT |
-		  VDCTRL0_SET_VSYNC_PULSE_WIDTH(vsync_pulse_len);
-	if (m->flags & DRM_MODE_FLAG_PHSYNC)
-		vdctrl0 |= VDCTRL0_HSYNC_ACT_HIGH;
-	if (m->flags & DRM_MODE_FLAG_PVSYNC)
-		vdctrl0 |= VDCTRL0_VSYNC_ACT_HIGH;
-	/* Make sure Data Enable is high active by default */
-	if (!(bus_flags & DRM_BUS_FLAG_DE_LOW))
-		vdctrl0 |= VDCTRL0_ENABLE_ACT_HIGH;
-	/*
-	 * DRM_BUS_FLAG_PIXDATA_DRIVE_ defines are controller centric,
-	 * controllers VDCTRL0_DOTCLK is display centric.
-	 * Drive on positive edge       -> display samples on falling edge
-	 * DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE -> VDCTRL0_DOTCLK_ACT_FALLING
-	 */
-	if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE)
-		vdctrl0 |= VDCTRL0_DOTCLK_ACT_FALLING;
-
-	writel(vdctrl0, mxsfb->base + LCDC_VDCTRL0);
-
-	mxsfb_set_bus_fmt(mxsfb);
-
-	/* Frame length in lines. */
-	writel(m->crtc_vtotal, mxsfb->base + LCDC_VDCTRL1);
-
-	/* Line length in units of clocks or pixels. */
-	hsync_pulse_len = m->crtc_hsync_end - m->crtc_hsync_start;
-	writel(set_hsync_pulse_width(mxsfb, hsync_pulse_len) |
-	       VDCTRL2_SET_HSYNC_PERIOD(m->crtc_htotal),
-	       mxsfb->base + LCDC_VDCTRL2);
-
-	writel(SET_HOR_WAIT_CNT(m->crtc_htotal - m->crtc_hsync_start) |
-	       SET_VERT_WAIT_CNT(m->crtc_vtotal - m->crtc_vsync_start),
-	       mxsfb->base + LCDC_VDCTRL3);
-
-	writel(SET_DOTCLK_H_VALID_DATA_CNT(m->hdisplay),
-	       mxsfb->base + LCDC_VDCTRL4);
-}
-
-void mxsfb_crtc_enable(struct mxsfb_drm_private *mxsfb)
-{
-	dma_addr_t paddr;
-
-	mxsfb_enable_axi_clk(mxsfb);
-	mxsfb_crtc_mode_set_nofb(mxsfb);
-
-	/* Write cur_buf as well to avoid an initial corrupt frame */
-	paddr = mxsfb_get_fb_paddr(mxsfb);
-	if (paddr) {
-		writel(paddr, mxsfb->base + mxsfb->devdata->cur_buf);
-		writel(paddr, mxsfb->base + mxsfb->devdata->next_buf);
-	}
-
-	mxsfb_enable_controller(mxsfb);
-}
-
-void mxsfb_crtc_disable(struct mxsfb_drm_private *mxsfb)
-{
-	mxsfb_disable_controller(mxsfb);
-	mxsfb_disable_axi_clk(mxsfb);
-}
-
-void mxsfb_plane_atomic_update(struct mxsfb_drm_private *mxsfb,
-			       struct drm_plane_state *state)
-{
-	struct drm_simple_display_pipe *pipe = &mxsfb->pipe;
-	struct drm_crtc *crtc = &pipe->crtc;
-	struct drm_pending_vblank_event *event;
-	dma_addr_t paddr;
-
-	spin_lock_irq(&crtc->dev->event_lock);
-	event = crtc->state->event;
-	if (event) {
-		crtc->state->event = NULL;
-
-		if (drm_crtc_vblank_get(crtc) == 0) {
-			drm_crtc_arm_vblank_event(crtc, event);
-		} else {
-			drm_crtc_send_vblank_event(crtc, event);
-		}
-	}
-	spin_unlock_irq(&crtc->dev->event_lock);
-
-	paddr = mxsfb_get_fb_paddr(mxsfb);
-	if (paddr) {
-		mxsfb_enable_axi_clk(mxsfb);
-		writel(paddr, mxsfb->base + mxsfb->devdata->next_buf);
-		mxsfb_disable_axi_clk(mxsfb);
-	}
-}
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.c b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
index 508764fccd27..35122aef037b 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_drv.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
@@ -9,30 +9,25 @@
  */
 
 #include <linux/clk.h>
-#include <linux/component.h>
 #include <linux/dma-mapping.h>
-#include <linux/list.h>
+#include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of_device.h>
-#include <linux/of_graph.h>
-#include <linux/of_reserved_mem.h>
+#include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
-#include <linux/dma-resv.h>
-#include <linux/spinlock.h>
 
-#include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_connector.h>
 #include <drm/drm_drv.h>
-#include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_irq.h>
+#include <drm/drm_mode_config.h>
 #include <drm/drm_of.h>
-#include <drm/drm_panel.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
 #include <drm/drm_vblank.h>
 
 #include "mxsfb_drv.h"
@@ -41,6 +36,11 @@
 enum mxsfb_devtype {
 	MXSFB_V3,
 	MXSFB_V4,
+	/*
+	 * Starting at i.MX6 the hardware version register is gone, use the
+	 * i.MX family number as the version.
+	 */
+	MXSFB_V6,
 };
 
 static const struct mxsfb_devdata mxsfb_devdata[] = {
@@ -48,38 +48,28 @@ static const struct mxsfb_devdata mxsfb_devdata[] = {
 		.transfer_count	= LCDC_V3_TRANSFER_COUNT,
 		.cur_buf	= LCDC_V3_CUR_BUF,
 		.next_buf	= LCDC_V3_NEXT_BUF,
-		.debug0		= LCDC_V3_DEBUG0,
 		.hs_wdth_mask	= 0xff,
 		.hs_wdth_shift	= 24,
-		.ipversion	= 3,
+		.has_overlay	= false,
 	},
 	[MXSFB_V4] = {
 		.transfer_count	= LCDC_V4_TRANSFER_COUNT,
 		.cur_buf	= LCDC_V4_CUR_BUF,
 		.next_buf	= LCDC_V4_NEXT_BUF,
-		.debug0		= LCDC_V4_DEBUG0,
 		.hs_wdth_mask	= 0x3fff,
 		.hs_wdth_shift	= 18,
-		.ipversion	= 4,
+		.has_overlay	= false,
+	},
+	[MXSFB_V6] = {
+		.transfer_count	= LCDC_V4_TRANSFER_COUNT,
+		.cur_buf	= LCDC_V4_CUR_BUF,
+		.next_buf	= LCDC_V4_NEXT_BUF,
+		.hs_wdth_mask	= 0x3fff,
+		.hs_wdth_shift	= 18,
+		.has_overlay	= true,
 	},
 };
 
-static const uint32_t mxsfb_formats[] = {
-	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_RGB565
-};
-
-static const uint64_t mxsfb_modifiers[] = {
-	DRM_FORMAT_MOD_LINEAR,
-	DRM_FORMAT_MOD_INVALID
-};
-
-static struct mxsfb_drm_private *
-drm_pipe_to_mxsfb_drm_private(struct drm_simple_display_pipe *pipe)
-{
-	return container_of(pipe, struct mxsfb_drm_private, pipe);
-}
-
 void mxsfb_enable_axi_clk(struct mxsfb_drm_private *mxsfb)
 {
 	if (mxsfb->clk_axi)
@@ -92,8 +82,26 @@ void mxsfb_disable_axi_clk(struct mxsfb_drm_private *mxsfb)
 		clk_disable_unprepare(mxsfb->clk_axi);
 }
 
+static struct drm_framebuffer *
+mxsfb_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;
+
+	info = drm_get_format_info(dev, mode_cmd);
+	if (!info)
+		return ERR_PTR(-EINVAL);
+
+	if (mode_cmd->width * info->cpp[0] != mode_cmd->pitches[0]) {
+		dev_dbg(dev->dev, "Invalid pitch: fb width must match pitch\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	return drm_gem_fb_create(dev, file_priv, mode_cmd);
+}
+
 static const struct drm_mode_config_funcs mxsfb_mode_config_funcs = {
-	.fb_create		= drm_gem_fb_create,
+	.fb_create		= mxsfb_fb_create,
 	.atomic_check		= drm_atomic_helper_check,
 	.atomic_commit		= drm_atomic_helper_commit,
 };
@@ -102,101 +110,51 @@ static const struct drm_mode_config_helper_funcs mxsfb_mode_config_helpers = {
 	.atomic_commit_tail = drm_atomic_helper_commit_tail_rpm,
 };
 
-static void mxsfb_pipe_enable(struct drm_simple_display_pipe *pipe,
-			      struct drm_crtc_state *crtc_state,
-			      struct drm_plane_state *plane_state)
+static int mxsfb_attach_bridge(struct mxsfb_drm_private *mxsfb)
 {
-	struct drm_connector *connector;
-	struct mxsfb_drm_private *mxsfb = drm_pipe_to_mxsfb_drm_private(pipe);
-	struct drm_device *drm = pipe->plane.dev;
-
-	if (!mxsfb->connector) {
-		list_for_each_entry(connector,
-				    &drm->mode_config.connector_list,
-				    head)
-			if (connector->encoder == &mxsfb->pipe.encoder) {
-				mxsfb->connector = connector;
-				break;
-			}
-	}
-
-	if (!mxsfb->connector) {
-		dev_warn(drm->dev, "No connector attached, using default\n");
-		mxsfb->connector = &mxsfb->panel_connector;
-	}
-
-	pm_runtime_get_sync(drm->dev);
-	drm_panel_prepare(mxsfb->panel);
-	mxsfb_crtc_enable(mxsfb);
-	drm_panel_enable(mxsfb->panel);
-}
+	struct drm_device *drm = mxsfb->drm;
+	struct drm_connector_list_iter iter;
+	struct drm_panel *panel;
+	struct drm_bridge *bridge;
+	int ret;
 
-static void mxsfb_pipe_disable(struct drm_simple_display_pipe *pipe)
-{
-	struct mxsfb_drm_private *mxsfb = drm_pipe_to_mxsfb_drm_private(pipe);
-	struct drm_device *drm = pipe->plane.dev;
-	struct drm_crtc *crtc = &pipe->crtc;
-	struct drm_pending_vblank_event *event;
-
-	drm_panel_disable(mxsfb->panel);
-	mxsfb_crtc_disable(mxsfb);
-	drm_panel_unprepare(mxsfb->panel);
-	pm_runtime_put_sync(drm->dev);
+	ret = drm_of_find_panel_or_bridge(drm->dev->of_node, 0, 0, &panel,
+					  &bridge);
+	if (ret)
+		return ret;
 
-	spin_lock_irq(&drm->event_lock);
-	event = crtc->state->event;
-	if (event) {
-		crtc->state->event = NULL;
-		drm_crtc_send_vblank_event(crtc, event);
+	if (panel) {
+		bridge = devm_drm_panel_bridge_add_typed(drm->dev, panel,
+							 DRM_MODE_CONNECTOR_DPI);
+		if (IS_ERR(bridge))
+			return PTR_ERR(bridge);
 	}
-	spin_unlock_irq(&drm->event_lock);
-
-	if (mxsfb->connector != &mxsfb->panel_connector)
-		mxsfb->connector = NULL;
-}
 
-static void mxsfb_pipe_update(struct drm_simple_display_pipe *pipe,
-			      struct drm_plane_state *plane_state)
-{
-	struct mxsfb_drm_private *mxsfb = drm_pipe_to_mxsfb_drm_private(pipe);
+	if (!bridge)
+		return -ENODEV;
 
-	mxsfb_plane_atomic_update(mxsfb, plane_state);
-}
+	ret = drm_bridge_attach(&mxsfb->encoder, bridge, NULL, 0);
+	if (ret) {
+		DRM_DEV_ERROR(drm->dev,
+			      "failed to attach bridge: %d\n", ret);
+		return ret;
+	}
 
-static int mxsfb_pipe_enable_vblank(struct drm_simple_display_pipe *pipe)
-{
-	struct mxsfb_drm_private *mxsfb = drm_pipe_to_mxsfb_drm_private(pipe);
+	mxsfb->bridge = bridge;
 
-	/* Clear and enable VBLANK IRQ */
-	mxsfb_enable_axi_clk(mxsfb);
-	writel(CTRL1_CUR_FRAME_DONE_IRQ, mxsfb->base + LCDC_CTRL1 + REG_CLR);
-	writel(CTRL1_CUR_FRAME_DONE_IRQ_EN, mxsfb->base + LCDC_CTRL1 + REG_SET);
-	mxsfb_disable_axi_clk(mxsfb);
+	/*
+	 * Get hold of the connector. This is a bit of a hack, until the bridge
+	 * API gives us bus flags and formats.
+	 */
+	drm_connector_list_iter_begin(drm, &iter);
+	mxsfb->connector = drm_connector_list_iter_next(&iter);
+	drm_connector_list_iter_end(&iter);
 
 	return 0;
 }
 
-static void mxsfb_pipe_disable_vblank(struct drm_simple_display_pipe *pipe)
-{
-	struct mxsfb_drm_private *mxsfb = drm_pipe_to_mxsfb_drm_private(pipe);
-
-	/* Disable and clear VBLANK IRQ */
-	mxsfb_enable_axi_clk(mxsfb);
-	writel(CTRL1_CUR_FRAME_DONE_IRQ_EN, mxsfb->base + LCDC_CTRL1 + REG_CLR);
-	writel(CTRL1_CUR_FRAME_DONE_IRQ, mxsfb->base + LCDC_CTRL1 + REG_CLR);
-	mxsfb_disable_axi_clk(mxsfb);
-}
-
-static struct drm_simple_display_pipe_funcs mxsfb_funcs = {
-	.enable		= mxsfb_pipe_enable,
-	.disable	= mxsfb_pipe_disable,
-	.update		= mxsfb_pipe_update,
-	.prepare_fb	= drm_gem_fb_simple_display_pipe_prepare_fb,
-	.enable_vblank	= mxsfb_pipe_enable_vblank,
-	.disable_vblank	= mxsfb_pipe_disable_vblank,
-};
-
-static int mxsfb_load(struct drm_device *drm)
+static int mxsfb_load(struct drm_device *drm,
+		      const struct mxsfb_devdata *devdata)
 {
 	struct platform_device *pdev = to_platform_device(drm->dev);
 	struct mxsfb_drm_private *mxsfb;
@@ -207,8 +165,9 @@ static int mxsfb_load(struct drm_device *drm)
 	if (!mxsfb)
 		return -ENOMEM;
 
+	mxsfb->drm = drm;
 	drm->dev_private = mxsfb;
-	mxsfb->devdata = &mxsfb_devdata[pdev->id_entry->driver_data];
+	mxsfb->devdata = devdata;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	mxsfb->base = devm_ioremap_resource(drm->dev, res);
@@ -233,50 +192,28 @@ static int mxsfb_load(struct drm_device *drm)
 
 	pm_runtime_enable(drm->dev);
 
-	ret = drm_vblank_init(drm, drm->mode_config.num_crtc);
-	if (ret < 0) {
-		dev_err(drm->dev, "Failed to initialise vblank\n");
-		goto err_vblank;
-	}
-
 	/* Modeset init */
 	drm_mode_config_init(drm);
 
-	ret = mxsfb_create_output(drm);
+	ret = mxsfb_kms_init(mxsfb);
 	if (ret < 0) {
-		dev_err(drm->dev, "Failed to create outputs\n");
+		dev_err(drm->dev, "Failed to initialize KMS pipeline\n");
 		goto err_vblank;
 	}
 
-	ret = drm_simple_display_pipe_init(drm, &mxsfb->pipe, &mxsfb_funcs,
-			mxsfb_formats, ARRAY_SIZE(mxsfb_formats),
-			mxsfb_modifiers, mxsfb->connector);
+	ret = drm_vblank_init(drm, drm->mode_config.num_crtc);
 	if (ret < 0) {
-		dev_err(drm->dev, "Cannot setup simple display pipe\n");
+		dev_err(drm->dev, "Failed to initialise vblank\n");
 		goto err_vblank;
 	}
 
-	/*
-	 * Attach panel only if there is one.
-	 * If there is no panel attach, it must be a bridge. In this case, we
-	 * need a reference to its connector for a proper initialization.
-	 * We will do this check in pipe->enable(), since the connector won't
-	 * be attached to an encoder until then.
-	 */
+	/* Start with vertical blanking interrupt reporting disabled. */
+	drm_crtc_vblank_off(&mxsfb->crtc);
 
-	if (mxsfb->panel) {
-		ret = drm_panel_attach(mxsfb->panel, mxsfb->connector);
-		if (ret) {
-			dev_err(drm->dev, "Cannot connect panel: %d\n", ret);
-			goto err_vblank;
-		}
-	} else if (mxsfb->bridge) {
-		ret = drm_simple_display_pipe_attach_bridge(&mxsfb->pipe,
-							    mxsfb->bridge);
-		if (ret) {
-			dev_err(drm->dev, "Cannot connect bridge: %d\n", ret);
-			goto err_vblank;
-		}
+	ret = mxsfb_attach_bridge(mxsfb);
+	if (ret) {
+		dev_err(drm->dev, "Cannot connect bridge: %d\n", ret);
+		goto err_vblank;
 	}
 
 	drm->mode_config.min_width	= MXSFB_MIN_XRES;
@@ -294,7 +231,7 @@ static int mxsfb_load(struct drm_device *drm)
 
 	if (ret < 0) {
 		dev_err(drm->dev, "Failed to install IRQ handler\n");
-		goto err_irq;
+		goto err_vblank;
 	}
 
 	drm_kms_helper_poll_init(drm);
@@ -305,8 +242,6 @@ static int mxsfb_load(struct drm_device *drm)
 
 	return 0;
 
-err_irq:
-	drm_panel_detach(mxsfb->panel);
 err_vblank:
 	pm_runtime_disable(drm->dev);
 
@@ -327,11 +262,13 @@ static void mxsfb_unload(struct drm_device *drm)
 	pm_runtime_disable(drm->dev);
 }
 
-static void mxsfb_irq_preinstall(struct drm_device *drm)
+static void mxsfb_irq_disable(struct drm_device *drm)
 {
 	struct mxsfb_drm_private *mxsfb = drm->dev_private;
 
-	mxsfb_pipe_disable_vblank(&mxsfb->pipe);
+	mxsfb_enable_axi_clk(mxsfb);
+	mxsfb->crtc.funcs->disable_vblank(&mxsfb->crtc);
+	mxsfb_disable_axi_clk(mxsfb);
 }
 
 static irqreturn_t mxsfb_irq_handler(int irq, void *data)
@@ -340,17 +277,13 @@ static irqreturn_t mxsfb_irq_handler(int irq, void *data)
 	struct mxsfb_drm_private *mxsfb = drm->dev_private;
 	u32 reg;
 
-	mxsfb_enable_axi_clk(mxsfb);
-
 	reg = readl(mxsfb->base + LCDC_CTRL1);
 
 	if (reg & CTRL1_CUR_FRAME_DONE_IRQ)
-		drm_crtc_handle_vblank(&mxsfb->pipe.crtc);
+		drm_crtc_handle_vblank(&mxsfb->crtc);
 
 	writel(CTRL1_CUR_FRAME_DONE_IRQ, mxsfb->base + LCDC_CTRL1 + REG_CLR);
 
-	mxsfb_disable_axi_clk(mxsfb);
-
 	return IRQ_HANDLED;
 }
 
@@ -359,8 +292,8 @@ DEFINE_DRM_GEM_CMA_FOPS(fops);
 static struct drm_driver mxsfb_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC,
 	.irq_handler		= mxsfb_irq_handler,
-	.irq_preinstall		= mxsfb_irq_preinstall,
-	.irq_uninstall		= mxsfb_irq_preinstall,
+	.irq_preinstall		= mxsfb_irq_disable,
+	.irq_uninstall		= mxsfb_irq_disable,
 	DRM_GEM_CMA_DRIVER_OPS,
 	.fops	= &fops,
 	.name	= "mxsfb-drm",
@@ -370,18 +303,10 @@ static struct drm_driver mxsfb_driver = {
 	.minor	= 0,
 };
 
-static const struct platform_device_id mxsfb_devtype[] = {
-	{ .name = "imx23-fb", .driver_data = MXSFB_V3, },
-	{ .name = "imx28-fb", .driver_data = MXSFB_V4, },
-	{ .name = "imx6sx-fb", .driver_data = MXSFB_V4, },
-	{ /* sentinel */ }
-};
-MODULE_DEVICE_TABLE(platform, mxsfb_devtype);
-
 static const struct of_device_id mxsfb_dt_ids[] = {
-	{ .compatible = "fsl,imx23-lcdif", .data = &mxsfb_devtype[0], },
-	{ .compatible = "fsl,imx28-lcdif", .data = &mxsfb_devtype[1], },
-	{ .compatible = "fsl,imx6sx-lcdif", .data = &mxsfb_devtype[2], },
+	{ .compatible = "fsl,imx23-lcdif", .data = &mxsfb_devdata[MXSFB_V3], },
+	{ .compatible = "fsl,imx28-lcdif", .data = &mxsfb_devdata[MXSFB_V4], },
+	{ .compatible = "fsl,imx6sx-lcdif", .data = &mxsfb_devdata[MXSFB_V6], },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, mxsfb_dt_ids);
@@ -396,14 +321,11 @@ static int mxsfb_probe(struct platform_device *pdev)
 	if (!pdev->dev.of_node)
 		return -ENODEV;
 
-	if (of_id)
-		pdev->id_entry = of_id->data;
-
 	drm = drm_dev_alloc(&mxsfb_driver, &pdev->dev);
 	if (IS_ERR(drm))
 		return PTR_ERR(drm);
 
-	ret = mxsfb_load(drm);
+	ret = mxsfb_load(drm, of_id->data);
 	if (ret)
 		goto err_free;
 
@@ -457,7 +379,6 @@ static const struct dev_pm_ops mxsfb_pm_ops = {
 static struct platform_driver mxsfb_platform_driver = {
 	.probe		= mxsfb_probe,
 	.remove		= mxsfb_remove,
-	.id_table	= mxsfb_devtype,
 	.driver	= {
 		.name		= "mxsfb",
 		.of_match_table	= mxsfb_dt_ids,
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.h b/drivers/gpu/drm/mxsfb/mxsfb_drv.h
index 0b65b5194a9c..399d23e91ed1 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_drv.h
+++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.h
@@ -8,14 +8,20 @@
 #ifndef __MXSFB_DRV_H__
 #define __MXSFB_DRV_H__
 
+#include <drm/drm_crtc.h>
+#include <drm/drm_device.h>
+#include <drm/drm_encoder.h>
+#include <drm/drm_plane.h>
+
+struct clk;
+
 struct mxsfb_devdata {
-	unsigned int	 transfer_count;
-	unsigned int	 cur_buf;
-	unsigned int	 next_buf;
-	unsigned int	 debug0;
-	unsigned int	 hs_wdth_mask;
-	unsigned int	 hs_wdth_shift;
-	unsigned int	 ipversion;
+	unsigned int	transfer_count;
+	unsigned int	cur_buf;
+	unsigned int	next_buf;
+	unsigned int	hs_wdth_mask;
+	unsigned int	hs_wdth_shift;
+	bool		has_overlay;
 };
 
 struct mxsfb_drm_private {
@@ -26,22 +32,26 @@ struct mxsfb_drm_private {
 	struct clk			*clk_axi;
 	struct clk			*clk_disp_axi;
 
-	struct drm_simple_display_pipe	pipe;
-	struct drm_connector		panel_connector;
+	struct drm_device		*drm;
+	struct {
+		struct drm_plane	primary;
+		struct drm_plane	overlay;
+	} planes;
+	struct drm_crtc			crtc;
+	struct drm_encoder		encoder;
 	struct drm_connector		*connector;
-	struct drm_panel		*panel;
 	struct drm_bridge		*bridge;
 };
 
-int mxsfb_setup_crtc(struct drm_device *dev);
-int mxsfb_create_output(struct drm_device *dev);
+static inline struct mxsfb_drm_private *
+to_mxsfb_drm_private(struct drm_device *drm)
+{
+	return drm->dev_private;
+}
 
 void mxsfb_enable_axi_clk(struct mxsfb_drm_private *mxsfb);
 void mxsfb_disable_axi_clk(struct mxsfb_drm_private *mxsfb);
 
-void mxsfb_crtc_enable(struct mxsfb_drm_private *mxsfb);
-void mxsfb_crtc_disable(struct mxsfb_drm_private *mxsfb);
-void mxsfb_plane_atomic_update(struct mxsfb_drm_private *mxsfb,
-			       struct drm_plane_state *state);
+int mxsfb_kms_init(struct mxsfb_drm_private *mxsfb);
 
 #endif /* __MXSFB_DRV_H__ */
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_kms.c b/drivers/gpu/drm/mxsfb/mxsfb_kms.c
new file mode 100644
index 000000000000..b721b8b262ce
--- /dev/null
+++ b/drivers/gpu/drm/mxsfb/mxsfb_kms.c
@@ -0,0 +1,571 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2016 Marek Vasut <marex@denx.de>
+ *
+ * This code is based on drivers/video/fbdev/mxsfb.c :
+ * Copyright (C) 2010 Juergen Beisert, Pengutronix
+ * Copyright (C) 2008-2009 Freescale Semiconductor, Inc. All Rights Reserved.
+ * Copyright (C) 2008 Embedded Alley Solutions, Inc All Rights Reserved.
+ */
+
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/pm_runtime.h>
+#include <linux/spinlock.h>
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_encoder.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_plane.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_vblank.h>
+
+#include "mxsfb_drv.h"
+#include "mxsfb_regs.h"
+
+/* 1 second delay should be plenty of time for block reset */
+#define RESET_TIMEOUT		1000000
+
+/* -----------------------------------------------------------------------------
+ * CRTC
+ */
+
+static u32 set_hsync_pulse_width(struct mxsfb_drm_private *mxsfb, u32 val)
+{
+	return (val & mxsfb->devdata->hs_wdth_mask) <<
+		mxsfb->devdata->hs_wdth_shift;
+}
+
+/*
+ * Setup the MXSFB registers for decoding the pixels out of the framebuffer and
+ * outputting them on the bus.
+ */
+static void mxsfb_set_formats(struct mxsfb_drm_private *mxsfb)
+{
+	struct drm_device *drm = mxsfb->drm;
+	const u32 format = mxsfb->crtc.primary->state->fb->format->format;
+	u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	u32 ctrl, ctrl1;
+
+	if (mxsfb->connector->display_info.num_bus_formats)
+		bus_format = mxsfb->connector->display_info.bus_formats[0];
+
+	DRM_DEV_DEBUG_DRIVER(drm->dev, "Using bus_format: 0x%08X\n",
+			     bus_format);
+
+	ctrl = CTRL_BYPASS_COUNT | CTRL_MASTER;
+
+	/* CTRL1 contains IRQ config and status bits, preserve those. */
+	ctrl1 = readl(mxsfb->base + LCDC_CTRL1);
+	ctrl1 &= CTRL1_CUR_FRAME_DONE_IRQ_EN | CTRL1_CUR_FRAME_DONE_IRQ;
+
+	switch (format) {
+	case DRM_FORMAT_RGB565:
+		dev_dbg(drm->dev, "Setting up RGB565 mode\n");
+		ctrl |= CTRL_WORD_LENGTH_16;
+		ctrl1 |= CTRL1_SET_BYTE_PACKAGING(0xf);
+		break;
+	case DRM_FORMAT_XRGB8888:
+		dev_dbg(drm->dev, "Setting up XRGB8888 mode\n");
+		ctrl |= CTRL_WORD_LENGTH_24;
+		/* Do not use packed pixels = one pixel per word instead. */
+		ctrl1 |= CTRL1_SET_BYTE_PACKAGING(0x7);
+		break;
+	}
+
+	switch (bus_format) {
+	case MEDIA_BUS_FMT_RGB565_1X16:
+		ctrl |= CTRL_BUS_WIDTH_16;
+		break;
+	case MEDIA_BUS_FMT_RGB666_1X18:
+		ctrl |= CTRL_BUS_WIDTH_18;
+		break;
+	case MEDIA_BUS_FMT_RGB888_1X24:
+		ctrl |= CTRL_BUS_WIDTH_24;
+		break;
+	default:
+		dev_err(drm->dev, "Unknown media bus format %d\n", bus_format);
+		break;
+	}
+
+	writel(ctrl1, mxsfb->base + LCDC_CTRL1);
+	writel(ctrl, mxsfb->base + LCDC_CTRL);
+}
+
+static void mxsfb_enable_controller(struct mxsfb_drm_private *mxsfb)
+{
+	u32 reg;
+
+	if (mxsfb->clk_disp_axi)
+		clk_prepare_enable(mxsfb->clk_disp_axi);
+	clk_prepare_enable(mxsfb->clk);
+
+	/* If it was disabled, re-enable the mode again */
+	writel(CTRL_DOTCLK_MODE, mxsfb->base + LCDC_CTRL + REG_SET);
+
+	/* Enable the SYNC signals first, then the DMA engine */
+	reg = readl(mxsfb->base + LCDC_VDCTRL4);
+	reg |= VDCTRL4_SYNC_SIGNALS_ON;
+	writel(reg, mxsfb->base + LCDC_VDCTRL4);
+
+	writel(CTRL_RUN, mxsfb->base + LCDC_CTRL + REG_SET);
+}
+
+static void mxsfb_disable_controller(struct mxsfb_drm_private *mxsfb)
+{
+	u32 reg;
+
+	/*
+	 * Even if we disable the controller here, it will still continue
+	 * until its FIFOs are running out of data
+	 */
+	writel(CTRL_DOTCLK_MODE, mxsfb->base + LCDC_CTRL + REG_CLR);
+
+	readl_poll_timeout(mxsfb->base + LCDC_CTRL, reg, !(reg & CTRL_RUN),
+			   0, 1000);
+
+	reg = readl(mxsfb->base + LCDC_VDCTRL4);
+	reg &= ~VDCTRL4_SYNC_SIGNALS_ON;
+	writel(reg, mxsfb->base + LCDC_VDCTRL4);
+
+	clk_disable_unprepare(mxsfb->clk);
+	if (mxsfb->clk_disp_axi)
+		clk_disable_unprepare(mxsfb->clk_disp_axi);
+}
+
+/*
+ * Clear the bit and poll it cleared.  This is usually called with
+ * a reset address and mask being either SFTRST(bit 31) or CLKGATE
+ * (bit 30).
+ */
+static int clear_poll_bit(void __iomem *addr, u32 mask)
+{
+	u32 reg;
+
+	writel(mask, addr + REG_CLR);
+	return readl_poll_timeout(addr, reg, !(reg & mask), 0, RESET_TIMEOUT);
+}
+
+static int mxsfb_reset_block(struct mxsfb_drm_private *mxsfb)
+{
+	int ret;
+
+	ret = clear_poll_bit(mxsfb->base + LCDC_CTRL, CTRL_SFTRST);
+	if (ret)
+		return ret;
+
+	writel(CTRL_CLKGATE, mxsfb->base + LCDC_CTRL + REG_CLR);
+
+	ret = clear_poll_bit(mxsfb->base + LCDC_CTRL, CTRL_SFTRST);
+	if (ret)
+		return ret;
+
+	return clear_poll_bit(mxsfb->base + LCDC_CTRL, CTRL_CLKGATE);
+}
+
+static dma_addr_t mxsfb_get_fb_paddr(struct drm_plane *plane)
+{
+	struct drm_framebuffer *fb = plane->state->fb;
+	struct drm_gem_cma_object *gem;
+
+	if (!fb)
+		return 0;
+
+	gem = drm_fb_cma_get_gem_obj(fb, 0);
+	if (!gem)
+		return 0;
+
+	return gem->paddr;
+}
+
+static void mxsfb_crtc_mode_set_nofb(struct mxsfb_drm_private *mxsfb)
+{
+	struct drm_device *drm = mxsfb->crtc.dev;
+	struct drm_display_mode *m = &mxsfb->crtc.state->adjusted_mode;
+	u32 bus_flags = mxsfb->connector->display_info.bus_flags;
+	u32 vdctrl0, vsync_pulse_len, hsync_pulse_len;
+	int err;
+
+	/*
+	 * It seems, you can't re-program the controller if it is still
+	 * running. This may lead to shifted pictures (FIFO issue?), so
+	 * first stop the controller and drain its FIFOs.
+	 */
+
+	/* Mandatory eLCDIF reset as per the Reference Manual */
+	err = mxsfb_reset_block(mxsfb);
+	if (err)
+		return;
+
+	/* Clear the FIFOs */
+	writel(CTRL1_FIFO_CLEAR, mxsfb->base + LCDC_CTRL1 + REG_SET);
+
+	if (mxsfb->devdata->has_overlay)
+		writel(0, mxsfb->base + LCDC_AS_CTRL);
+
+	mxsfb_set_formats(mxsfb);
+
+	clk_set_rate(mxsfb->clk, m->crtc_clock * 1000);
+
+	if (mxsfb->bridge && mxsfb->bridge->timings)
+		bus_flags = mxsfb->bridge->timings->input_bus_flags;
+
+	DRM_DEV_DEBUG_DRIVER(drm->dev, "Pixel clock: %dkHz (actual: %dkHz)\n",
+			     m->crtc_clock,
+			     (int)(clk_get_rate(mxsfb->clk) / 1000));
+	DRM_DEV_DEBUG_DRIVER(drm->dev, "Connector bus_flags: 0x%08X\n",
+			     bus_flags);
+	DRM_DEV_DEBUG_DRIVER(drm->dev, "Mode flags: 0x%08X\n", m->flags);
+
+	writel(TRANSFER_COUNT_SET_VCOUNT(m->crtc_vdisplay) |
+	       TRANSFER_COUNT_SET_HCOUNT(m->crtc_hdisplay),
+	       mxsfb->base + mxsfb->devdata->transfer_count);
+
+	vsync_pulse_len = m->crtc_vsync_end - m->crtc_vsync_start;
+
+	vdctrl0 = VDCTRL0_ENABLE_PRESENT |	/* Always in DOTCLOCK mode */
+		  VDCTRL0_VSYNC_PERIOD_UNIT |
+		  VDCTRL0_VSYNC_PULSE_WIDTH_UNIT |
+		  VDCTRL0_SET_VSYNC_PULSE_WIDTH(vsync_pulse_len);
+	if (m->flags & DRM_MODE_FLAG_PHSYNC)
+		vdctrl0 |= VDCTRL0_HSYNC_ACT_HIGH;
+	if (m->flags & DRM_MODE_FLAG_PVSYNC)
+		vdctrl0 |= VDCTRL0_VSYNC_ACT_HIGH;
+	/* Make sure Data Enable is high active by default */
+	if (!(bus_flags & DRM_BUS_FLAG_DE_LOW))
+		vdctrl0 |= VDCTRL0_ENABLE_ACT_HIGH;
+	/*
+	 * DRM_BUS_FLAG_PIXDATA_DRIVE_ defines are controller centric,
+	 * controllers VDCTRL0_DOTCLK is display centric.
+	 * Drive on positive edge       -> display samples on falling edge
+	 * DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE -> VDCTRL0_DOTCLK_ACT_FALLING
+	 */
+	if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE)
+		vdctrl0 |= VDCTRL0_DOTCLK_ACT_FALLING;
+
+	writel(vdctrl0, mxsfb->base + LCDC_VDCTRL0);
+
+	/* Frame length in lines. */
+	writel(m->crtc_vtotal, mxsfb->base + LCDC_VDCTRL1);
+
+	/* Line length in units of clocks or pixels. */
+	hsync_pulse_len = m->crtc_hsync_end - m->crtc_hsync_start;
+	writel(set_hsync_pulse_width(mxsfb, hsync_pulse_len) |
+	       VDCTRL2_SET_HSYNC_PERIOD(m->crtc_htotal),
+	       mxsfb->base + LCDC_VDCTRL2);
+
+	writel(SET_HOR_WAIT_CNT(m->crtc_htotal - m->crtc_hsync_start) |
+	       SET_VERT_WAIT_CNT(m->crtc_vtotal - m->crtc_vsync_start),
+	       mxsfb->base + LCDC_VDCTRL3);
+
+	writel(SET_DOTCLK_H_VALID_DATA_CNT(m->hdisplay),
+	       mxsfb->base + LCDC_VDCTRL4);
+}
+
+static int mxsfb_crtc_atomic_check(struct drm_crtc *crtc,
+				   struct drm_crtc_state *state)
+{
+	bool has_primary = state->plane_mask &
+			   drm_plane_mask(crtc->primary);
+
+	/* The primary plane has to be enabled when the CRTC is active. */
+	if (state->active && !has_primary)
+		return -EINVAL;
+
+	/* TODO: Is this needed ? */
+	return drm_atomic_add_affected_planes(state->state, crtc);
+}
+
+static void mxsfb_crtc_atomic_flush(struct drm_crtc *crtc,
+				    struct drm_crtc_state *old_state)
+{
+	struct drm_pending_vblank_event *event;
+
+	event = crtc->state->event;
+	crtc->state->event = NULL;
+
+	if (!event)
+		return;
+
+	spin_lock_irq(&crtc->dev->event_lock);
+	if (drm_crtc_vblank_get(crtc) == 0)
+		drm_crtc_arm_vblank_event(crtc, event);
+	else
+		drm_crtc_send_vblank_event(crtc, event);
+	spin_unlock_irq(&crtc->dev->event_lock);
+}
+
+static void mxsfb_crtc_atomic_enable(struct drm_crtc *crtc,
+				     struct drm_crtc_state *old_state)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(crtc->dev);
+	struct drm_device *drm = mxsfb->drm;
+	dma_addr_t paddr;
+
+	pm_runtime_get_sync(drm->dev);
+	mxsfb_enable_axi_clk(mxsfb);
+
+	drm_crtc_vblank_on(crtc);
+
+	mxsfb_crtc_mode_set_nofb(mxsfb);
+
+	/* Write cur_buf as well to avoid an initial corrupt frame */
+	paddr = mxsfb_get_fb_paddr(crtc->primary);
+	if (paddr) {
+		writel(paddr, mxsfb->base + mxsfb->devdata->cur_buf);
+		writel(paddr, mxsfb->base + mxsfb->devdata->next_buf);
+	}
+
+	mxsfb_enable_controller(mxsfb);
+}
+
+static void mxsfb_crtc_atomic_disable(struct drm_crtc *crtc,
+				      struct drm_crtc_state *old_state)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(crtc->dev);
+	struct drm_device *drm = mxsfb->drm;
+	struct drm_pending_vblank_event *event;
+
+	mxsfb_disable_controller(mxsfb);
+
+	spin_lock_irq(&drm->event_lock);
+	event = crtc->state->event;
+	if (event) {
+		crtc->state->event = NULL;
+		drm_crtc_send_vblank_event(crtc, event);
+	}
+	spin_unlock_irq(&drm->event_lock);
+
+	drm_crtc_vblank_off(crtc);
+
+	mxsfb_disable_axi_clk(mxsfb);
+	pm_runtime_put_sync(drm->dev);
+}
+
+static int mxsfb_crtc_enable_vblank(struct drm_crtc *crtc)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(crtc->dev);
+
+	/* Clear and enable VBLANK IRQ */
+	writel(CTRL1_CUR_FRAME_DONE_IRQ, mxsfb->base + LCDC_CTRL1 + REG_CLR);
+	writel(CTRL1_CUR_FRAME_DONE_IRQ_EN, mxsfb->base + LCDC_CTRL1 + REG_SET);
+
+	return 0;
+}
+
+static void mxsfb_crtc_disable_vblank(struct drm_crtc *crtc)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(crtc->dev);
+
+	/* Disable and clear VBLANK IRQ */
+	writel(CTRL1_CUR_FRAME_DONE_IRQ_EN, mxsfb->base + LCDC_CTRL1 + REG_CLR);
+	writel(CTRL1_CUR_FRAME_DONE_IRQ, mxsfb->base + LCDC_CTRL1 + REG_CLR);
+}
+
+static const struct drm_crtc_helper_funcs mxsfb_crtc_helper_funcs = {
+	.atomic_check = mxsfb_crtc_atomic_check,
+	.atomic_flush = mxsfb_crtc_atomic_flush,
+	.atomic_enable = mxsfb_crtc_atomic_enable,
+	.atomic_disable = mxsfb_crtc_atomic_disable,
+};
+
+static const struct drm_crtc_funcs mxsfb_crtc_funcs = {
+	.reset = drm_atomic_helper_crtc_reset,
+	.destroy = drm_crtc_cleanup,
+	.set_config = drm_atomic_helper_set_config,
+	.page_flip = drm_atomic_helper_page_flip,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+	.enable_vblank = mxsfb_crtc_enable_vblank,
+	.disable_vblank = mxsfb_crtc_disable_vblank,
+};
+
+/* -----------------------------------------------------------------------------
+ * Encoder
+ */
+
+static const struct drm_encoder_funcs mxsfb_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+/* -----------------------------------------------------------------------------
+ * Planes
+ */
+
+static int mxsfb_plane_atomic_check(struct drm_plane *plane,
+				    struct drm_plane_state *plane_state)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(plane->dev);
+	struct drm_crtc_state *crtc_state;
+
+	crtc_state = drm_atomic_get_new_crtc_state(plane_state->state,
+						   &mxsfb->crtc);
+
+	return drm_atomic_helper_check_plane_state(plane_state, crtc_state,
+						   DRM_PLANE_HELPER_NO_SCALING,
+						   DRM_PLANE_HELPER_NO_SCALING,
+						   false, true);
+}
+
+static void mxsfb_plane_primary_atomic_update(struct drm_plane *plane,
+					      struct drm_plane_state *old_pstate)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(plane->dev);
+	dma_addr_t paddr;
+
+	paddr = mxsfb_get_fb_paddr(plane);
+	if (paddr)
+		writel(paddr, mxsfb->base + mxsfb->devdata->next_buf);
+}
+
+static void mxsfb_plane_overlay_atomic_update(struct drm_plane *plane,
+					      struct drm_plane_state *old_pstate)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(plane->dev);
+	struct drm_plane_state *state = plane->state;
+	dma_addr_t paddr;
+	u32 ctrl;
+
+	paddr = mxsfb_get_fb_paddr(plane);
+	if (!paddr) {
+		writel(0, mxsfb->base + LCDC_AS_CTRL);
+		return;
+	}
+
+	/*
+	 * HACK: The hardware seems to output 64 bytes of data of unknown
+	 * origin, and then to proceed with the framebuffer. Until the reason
+	 * is understood, live with the 16 initial invalid pixels on the first
+	 * line and start 64 bytes within the framebuffer.
+	 */
+	paddr += 64;
+
+	writel(paddr, mxsfb->base + LCDC_AS_NEXT_BUF);
+
+	/*
+	 * If the plane was previously disabled, write LCDC_AS_BUF as well to
+	 * provide the first buffer.
+	 */
+	if (!old_pstate->fb)
+		writel(paddr, mxsfb->base + LCDC_AS_BUF);
+
+	ctrl = AS_CTRL_AS_ENABLE | AS_CTRL_ALPHA(255);
+
+	switch (state->fb->format->format) {
+	case DRM_FORMAT_XRGB4444:
+		ctrl |= AS_CTRL_FORMAT_RGB444 | AS_CTRL_ALPHA_CTRL_OVERRIDE;
+		break;
+	case DRM_FORMAT_ARGB4444:
+		ctrl |= AS_CTRL_FORMAT_ARGB4444 | AS_CTRL_ALPHA_CTRL_EMBEDDED;
+		break;
+	case DRM_FORMAT_XRGB1555:
+		ctrl |= AS_CTRL_FORMAT_RGB555 | AS_CTRL_ALPHA_CTRL_OVERRIDE;
+		break;
+	case DRM_FORMAT_ARGB1555:
+		ctrl |= AS_CTRL_FORMAT_ARGB1555 | AS_CTRL_ALPHA_CTRL_EMBEDDED;
+		break;
+	case DRM_FORMAT_RGB565:
+		ctrl |= AS_CTRL_FORMAT_RGB565 | AS_CTRL_ALPHA_CTRL_OVERRIDE;
+		break;
+	case DRM_FORMAT_XRGB8888:
+		ctrl |= AS_CTRL_FORMAT_RGB888 | AS_CTRL_ALPHA_CTRL_OVERRIDE;
+		break;
+	case DRM_FORMAT_ARGB8888:
+		ctrl |= AS_CTRL_FORMAT_ARGB8888 | AS_CTRL_ALPHA_CTRL_EMBEDDED;
+		break;
+	}
+
+	writel(ctrl, mxsfb->base + LCDC_AS_CTRL);
+}
+
+static const struct drm_plane_helper_funcs mxsfb_plane_primary_helper_funcs = {
+	.atomic_check = mxsfb_plane_atomic_check,
+	.atomic_update = mxsfb_plane_primary_atomic_update,
+};
+
+static const struct drm_plane_helper_funcs mxsfb_plane_overlay_helper_funcs = {
+	.atomic_check = mxsfb_plane_atomic_check,
+	.atomic_update = mxsfb_plane_overlay_atomic_update,
+};
+
+static const struct drm_plane_funcs mxsfb_plane_funcs = {
+	.update_plane		= drm_atomic_helper_update_plane,
+	.disable_plane		= drm_atomic_helper_disable_plane,
+	.destroy		= drm_plane_cleanup,
+	.reset			= drm_atomic_helper_plane_reset,
+	.atomic_duplicate_state	= drm_atomic_helper_plane_duplicate_state,
+	.atomic_destroy_state	= drm_atomic_helper_plane_destroy_state,
+};
+
+static const uint32_t mxsfb_primary_plane_formats[] = {
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+};
+
+static const uint32_t mxsfb_overlay_plane_formats[] = {
+	DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ARGB8888,
+};
+
+static const uint64_t mxsfb_modifiers[] = {
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID
+};
+
+/* -----------------------------------------------------------------------------
+ * Initialization
+ */
+
+int mxsfb_kms_init(struct mxsfb_drm_private *mxsfb)
+{
+	struct drm_encoder *encoder = &mxsfb->encoder;
+	struct drm_crtc *crtc = &mxsfb->crtc;
+	int ret;
+
+	drm_plane_helper_add(&mxsfb->planes.primary,
+			     &mxsfb_plane_primary_helper_funcs);
+	ret = drm_universal_plane_init(mxsfb->drm, &mxsfb->planes.primary, 1,
+				       &mxsfb_plane_funcs,
+				       mxsfb_primary_plane_formats,
+				       ARRAY_SIZE(mxsfb_primary_plane_formats),
+				       mxsfb_modifiers, DRM_PLANE_TYPE_PRIMARY,
+				       NULL);
+	if (ret)
+		return ret;
+
+	if (mxsfb->devdata->has_overlay) {
+		drm_plane_helper_add(&mxsfb->planes.overlay,
+				     &mxsfb_plane_overlay_helper_funcs);
+		ret = drm_universal_plane_init(mxsfb->drm,
+					       &mxsfb->planes.overlay, 1,
+					       &mxsfb_plane_funcs,
+					       mxsfb_overlay_plane_formats,
+					       ARRAY_SIZE(mxsfb_overlay_plane_formats),
+					       mxsfb_modifiers, DRM_PLANE_TYPE_OVERLAY,
+					       NULL);
+		if (ret)
+			return ret;
+	}
+
+	drm_crtc_helper_add(crtc, &mxsfb_crtc_helper_funcs);
+	ret = drm_crtc_init_with_planes(mxsfb->drm, crtc,
+					&mxsfb->planes.primary, NULL,
+					&mxsfb_crtc_funcs, NULL);
+	if (ret)
+		return ret;
+
+	encoder->possible_crtcs = drm_crtc_mask(crtc);
+	return drm_encoder_init(mxsfb->drm, encoder, &mxsfb_encoder_funcs,
+				DRM_MODE_ENCODER_NONE, NULL);
+}
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_out.c b/drivers/gpu/drm/mxsfb/mxsfb_out.c
deleted file mode 100644
index 9eca1605d11d..000000000000
--- a/drivers/gpu/drm/mxsfb/mxsfb_out.c
+++ /dev/null
@@ -1,99 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * Copyright (C) 2016 Marek Vasut <marex@denx.de>
- */
-
-#include <linux/of_graph.h>
-
-#include <drm/drm_atomic.h>
-#include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc.h>
-#include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_gem_cma_helper.h>
-#include <drm/drm_of.h>
-#include <drm/drm_panel.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
-
-#include "mxsfb_drv.h"
-
-static struct mxsfb_drm_private *
-drm_connector_to_mxsfb_drm_private(struct drm_connector *connector)
-{
-	return container_of(connector, struct mxsfb_drm_private,
-			    panel_connector);
-}
-
-static int mxsfb_panel_get_modes(struct drm_connector *connector)
-{
-	struct mxsfb_drm_private *mxsfb =
-			drm_connector_to_mxsfb_drm_private(connector);
-
-	if (mxsfb->panel)
-		return drm_panel_get_modes(mxsfb->panel, connector);
-
-	return 0;
-}
-
-static const struct
-drm_connector_helper_funcs mxsfb_panel_connector_helper_funcs = {
-	.get_modes = mxsfb_panel_get_modes,
-};
-
-static enum drm_connector_status
-mxsfb_panel_connector_detect(struct drm_connector *connector, bool force)
-{
-	struct mxsfb_drm_private *mxsfb =
-			drm_connector_to_mxsfb_drm_private(connector);
-
-	if (mxsfb->panel)
-		return connector_status_connected;
-
-	return connector_status_disconnected;
-}
-
-static void mxsfb_panel_connector_destroy(struct drm_connector *connector)
-{
-	struct mxsfb_drm_private *mxsfb =
-			drm_connector_to_mxsfb_drm_private(connector);
-
-	if (mxsfb->panel)
-		drm_panel_detach(mxsfb->panel);
-
-	drm_connector_unregister(connector);
-	drm_connector_cleanup(connector);
-}
-
-static const struct drm_connector_funcs mxsfb_panel_connector_funcs = {
-	.detect			= mxsfb_panel_connector_detect,
-	.fill_modes		= drm_helper_probe_single_connector_modes,
-	.destroy		= mxsfb_panel_connector_destroy,
-	.reset			= drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state	= drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
-};
-
-int mxsfb_create_output(struct drm_device *drm)
-{
-	struct mxsfb_drm_private *mxsfb = drm->dev_private;
-	int ret;
-
-	ret = drm_of_find_panel_or_bridge(drm->dev->of_node, 0, 0,
-					  &mxsfb->panel, &mxsfb->bridge);
-	if (ret)
-		return ret;
-
-	if (mxsfb->panel) {
-		mxsfb->connector = &mxsfb->panel_connector;
-		mxsfb->connector->dpms = DRM_MODE_DPMS_OFF;
-		mxsfb->connector->polled = 0;
-		drm_connector_helper_add(mxsfb->connector,
-					 &mxsfb_panel_connector_helper_funcs);
-		ret = drm_connector_init(drm, mxsfb->connector,
-					 &mxsfb_panel_connector_funcs,
-					 DRM_MODE_CONNECTOR_Unknown);
-	}
-
-	return ret;
-}
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_regs.h b/drivers/gpu/drm/mxsfb/mxsfb_regs.h
index 932d7ea08fd5..55d28a27f912 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_regs.h
+++ b/drivers/gpu/drm/mxsfb/mxsfb_regs.h
@@ -27,52 +27,61 @@
 #define LCDC_VDCTRL4			0xb0
 #define LCDC_V4_DEBUG0			0x1d0
 #define LCDC_V3_DEBUG0			0x1f0
-
-#define CTRL_SFTRST			(1 << 31)
-#define CTRL_CLKGATE			(1 << 30)
-#define CTRL_BYPASS_COUNT		(1 << 19)
-#define CTRL_VSYNC_MODE			(1 << 18)
-#define CTRL_DOTCLK_MODE		(1 << 17)
-#define CTRL_DATA_SELECT		(1 << 16)
-#define CTRL_SET_BUS_WIDTH(x)		(((x) & 0x3) << 10)
-#define CTRL_GET_BUS_WIDTH(x)		(((x) >> 10) & 0x3)
+#define LCDC_AS_CTRL			0x210
+#define LCDC_AS_BUF			0x220
+#define LCDC_AS_NEXT_BUF		0x230
+#define LCDC_AS_CLRKEYLOW		0x240
+#define LCDC_AS_CLRKEYHIGH		0x250
+
+#define CTRL_SFTRST			BIT(31)
+#define CTRL_CLKGATE			BIT(30)
+#define CTRL_BYPASS_COUNT		BIT(19)
+#define CTRL_VSYNC_MODE			BIT(18)
+#define CTRL_DOTCLK_MODE		BIT(17)
+#define CTRL_DATA_SELECT		BIT(16)
+#define CTRL_BUS_WIDTH_16		(0 << 10)
+#define CTRL_BUS_WIDTH_8		(1 << 10)
+#define CTRL_BUS_WIDTH_18		(2 << 10)
+#define CTRL_BUS_WIDTH_24		(3 << 10)
 #define CTRL_BUS_WIDTH_MASK		(0x3 << 10)
-#define CTRL_SET_WORD_LENGTH(x)		(((x) & 0x3) << 8)
-#define CTRL_GET_WORD_LENGTH(x)		(((x) >> 8) & 0x3)
-#define CTRL_MASTER			(1 << 5)
-#define CTRL_DF16			(1 << 3)
-#define CTRL_DF18			(1 << 2)
-#define CTRL_DF24			(1 << 1)
-#define CTRL_RUN			(1 << 0)
-
-#define CTRL1_FIFO_CLEAR		(1 << 21)
+#define CTRL_WORD_LENGTH_16		(0 << 8)
+#define CTRL_WORD_LENGTH_8		(1 << 8)
+#define CTRL_WORD_LENGTH_18		(2 << 8)
+#define CTRL_WORD_LENGTH_24		(3 << 8)
+#define CTRL_MASTER			BIT(5)
+#define CTRL_DF16			BIT(3)
+#define CTRL_DF18			BIT(2)
+#define CTRL_DF24			BIT(1)
+#define CTRL_RUN			BIT(0)
+
+#define CTRL1_FIFO_CLEAR		BIT(21)
 #define CTRL1_SET_BYTE_PACKAGING(x)	(((x) & 0xf) << 16)
 #define CTRL1_GET_BYTE_PACKAGING(x)	(((x) >> 16) & 0xf)
-#define CTRL1_CUR_FRAME_DONE_IRQ_EN	(1 << 13)
-#define CTRL1_CUR_FRAME_DONE_IRQ	(1 << 9)
+#define CTRL1_CUR_FRAME_DONE_IRQ_EN	BIT(13)
+#define CTRL1_CUR_FRAME_DONE_IRQ	BIT(9)
 
 #define TRANSFER_COUNT_SET_VCOUNT(x)	(((x) & 0xffff) << 16)
 #define TRANSFER_COUNT_GET_VCOUNT(x)	(((x) >> 16) & 0xffff)
 #define TRANSFER_COUNT_SET_HCOUNT(x)	((x) & 0xffff)
 #define TRANSFER_COUNT_GET_HCOUNT(x)	((x) & 0xffff)
 
-#define VDCTRL0_ENABLE_PRESENT		(1 << 28)
-#define VDCTRL0_VSYNC_ACT_HIGH		(1 << 27)
-#define VDCTRL0_HSYNC_ACT_HIGH		(1 << 26)
-#define VDCTRL0_DOTCLK_ACT_FALLING	(1 << 25)
-#define VDCTRL0_ENABLE_ACT_HIGH		(1 << 24)
-#define VDCTRL0_VSYNC_PERIOD_UNIT	(1 << 21)
-#define VDCTRL0_VSYNC_PULSE_WIDTH_UNIT	(1 << 20)
-#define VDCTRL0_HALF_LINE		(1 << 19)
-#define VDCTRL0_HALF_LINE_MODE		(1 << 18)
+#define VDCTRL0_ENABLE_PRESENT		BIT(28)
+#define VDCTRL0_VSYNC_ACT_HIGH		BIT(27)
+#define VDCTRL0_HSYNC_ACT_HIGH		BIT(26)
+#define VDCTRL0_DOTCLK_ACT_FALLING	BIT(25)
+#define VDCTRL0_ENABLE_ACT_HIGH		BIT(24)
+#define VDCTRL0_VSYNC_PERIOD_UNIT	BIT(21)
+#define VDCTRL0_VSYNC_PULSE_WIDTH_UNIT	BIT(20)
+#define VDCTRL0_HALF_LINE		BIT(19)
+#define VDCTRL0_HALF_LINE_MODE		BIT(18)
 #define VDCTRL0_SET_VSYNC_PULSE_WIDTH(x) ((x) & 0x3ffff)
 #define VDCTRL0_GET_VSYNC_PULSE_WIDTH(x) ((x) & 0x3ffff)
 
 #define VDCTRL2_SET_HSYNC_PERIOD(x)	((x) & 0x3ffff)
 #define VDCTRL2_GET_HSYNC_PERIOD(x)	((x) & 0x3ffff)
 
-#define VDCTRL3_MUX_SYNC_SIGNALS	(1 << 29)
-#define VDCTRL3_VSYNC_ONLY		(1 << 28)
+#define VDCTRL3_MUX_SYNC_SIGNALS	BIT(29)
+#define VDCTRL3_VSYNC_ONLY		BIT(28)
 #define SET_HOR_WAIT_CNT(x)		(((x) & 0xfff) << 16)
 #define GET_HOR_WAIT_CNT(x)		(((x) >> 16) & 0xfff)
 #define SET_VERT_WAIT_CNT(x)		((x) & 0xffff)
@@ -80,28 +89,32 @@
 
 #define VDCTRL4_SET_DOTCLK_DLY(x)	(((x) & 0x7) << 29) /* v4 only */
 #define VDCTRL4_GET_DOTCLK_DLY(x)	(((x) >> 29) & 0x7) /* v4 only */
-#define VDCTRL4_SYNC_SIGNALS_ON		(1 << 18)
+#define VDCTRL4_SYNC_SIGNALS_ON		BIT(18)
 #define SET_DOTCLK_H_VALID_DATA_CNT(x)	((x) & 0x3ffff)
 
-#define DEBUG0_HSYNC			(1 < 26)
-#define DEBUG0_VSYNC			(1 < 25)
+#define DEBUG0_HSYNC			BIT(26)
+#define DEBUG0_VSYNC			BIT(25)
+
+#define AS_CTRL_PS_DISABLE		BIT(23)
+#define AS_CTRL_ALPHA_INVERT		BIT(20)
+#define AS_CTRL_ALPHA(a)		(((a) & 0xff) << 8)
+#define AS_CTRL_FORMAT_RGB565		(0xe << 4)
+#define AS_CTRL_FORMAT_RGB444		(0xd << 4)
+#define AS_CTRL_FORMAT_RGB555		(0xc << 4)
+#define AS_CTRL_FORMAT_ARGB4444		(0x9 << 4)
+#define AS_CTRL_FORMAT_ARGB1555		(0x8 << 4)
+#define AS_CTRL_FORMAT_RGB888		(0x4 << 4)
+#define AS_CTRL_FORMAT_ARGB8888		(0x0 << 4)
+#define AS_CTRL_ENABLE_COLORKEY		BIT(3)
+#define AS_CTRL_ALPHA_CTRL_ROP		(3 << 1)
+#define AS_CTRL_ALPHA_CTRL_MULTIPLY	(2 << 1)
+#define AS_CTRL_ALPHA_CTRL_OVERRIDE	(1 << 1)
+#define AS_CTRL_ALPHA_CTRL_EMBEDDED	(0 << 1)
+#define AS_CTRL_AS_ENABLE		BIT(0)
 
 #define MXSFB_MIN_XRES			120
 #define MXSFB_MIN_YRES			120
 #define MXSFB_MAX_XRES			0xffff
 #define MXSFB_MAX_YRES			0xffff
 
-#define RED 0
-#define GREEN 1
-#define BLUE 2
-#define TRANSP 3
-
-#define STMLCDIF_8BIT  1 /* pixel data bus to the display is of 8 bit width */
-#define STMLCDIF_16BIT 0 /* pixel data bus to the display is of 16 bit width */
-#define STMLCDIF_18BIT 2 /* pixel data bus to the display is of 18 bit width */
-#define STMLCDIF_24BIT 3 /* pixel data bus to the display is of 24 bit width */
-
-#define MXSFB_SYNC_DATA_ENABLE_HIGH_ACT	(1 << 6)
-#define MXSFB_SYNC_DOTCLK_FALLING_ACT	(1 << 7) /* negative edge sampling */
-
 #endif /* __MXSFB_REGS_H__ */
diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
index 6416b6907aeb..f9e962fd94d0 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
@@ -615,7 +615,7 @@ nv_crtc_swap_fbs(struct drm_crtc *crtc, struct drm_framebuffer *old_fb)
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	int ret;
 
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, false);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 	if (ret == 0) {
 		if (disp->image[nv_crtc->index])
 			nouveau_bo_unpin(disp->image[nv_crtc->index]);
@@ -1172,7 +1172,7 @@ nv04_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		return -ENOMEM;
 
 	if (new_bo != old_bo) {
-		ret = nouveau_bo_pin(new_bo, TTM_PL_FLAG_VRAM, true);
+		ret = nouveau_bo_pin(new_bo, NOUVEAU_GEM_DOMAIN_VRAM, true);
 		if (ret)
 			goto fail_free;
 	}
@@ -1336,10 +1336,11 @@ nv04_crtc_create(struct drm_device *dev, int crtc_num)
 	drm_mode_crtc_set_gamma_size(&nv_crtc->base, 256);
 
 	ret = nouveau_bo_new(&nouveau_drm(dev)->client, 64*64*4, 0x100,
-			     TTM_PL_FLAG_VRAM, 0, 0x0000, NULL, NULL,
+			     NOUVEAU_GEM_DOMAIN_VRAM, 0, 0x0000, NULL, NULL,
 			     &nv_crtc->cursor.nvbo);
 	if (!ret) {
-		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo, TTM_PL_FLAG_VRAM, false);
+		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo,
+				     NOUVEAU_GEM_DOMAIN_VRAM, false);
 		if (!ret) {
 			ret = nouveau_bo_map(nv_crtc->cursor.nvbo);
 			if (ret)
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dac.c b/drivers/gpu/drm/nouveau/dispnv04/dac.c
index ffdd447d8706..22d10f328559 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dac.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dac.c
@@ -419,7 +419,7 @@ static void nv04_dac_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 nouveau_encoder_connector_get(nv_encoder)->base.name,
+		 nv04_encoder_get_connector(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
index f9f4482c79b5..42687ea2a4ca 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
@@ -184,7 +184,8 @@ static bool nv04_dfp_mode_fixup(struct drm_encoder *encoder,
 				struct drm_display_mode *adjusted_mode)
 {
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nouveau_connector *nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	struct nouveau_connector *nv_connector =
+		nv04_encoder_get_connector(nv_encoder);
 
 	if (!nv_connector->native_mode ||
 	    nv_connector->scaling_mode == DRM_MODE_SCALE_NONE ||
@@ -478,7 +479,7 @@ static void nv04_dfp_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 nouveau_encoder_connector_get(nv_encoder)->base.name,
+		 nv04_encoder_get_connector(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
@@ -591,7 +592,7 @@ static void nv04_dfp_restore(struct drm_encoder *encoder)
 
 	if (nv_encoder->dcb->type == DCB_OUTPUT_LVDS) {
 		struct nouveau_connector *connector =
-			nouveau_encoder_connector_get(nv_encoder);
+			nv04_encoder_get_connector(nv_encoder);
 
 		if (connector && connector->native_mode)
 			call_lvds_script(dev, nv_encoder->dcb, head,
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 900ab69df7e8..7739f46470d3 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -35,9 +35,28 @@
 
 #include <nvif/if0004.h>
 
+struct nouveau_connector *
+nv04_encoder_get_connector(struct nouveau_encoder *encoder)
+{
+	struct drm_device *dev = to_drm_encoder(encoder)->dev;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+	struct nouveau_connector *nv_connector = NULL;
+
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+		if (connector->encoder == to_drm_encoder(encoder))
+			nv_connector = nouveau_connector(connector);
+	}
+	drm_connector_list_iter_end(&conn_iter);
+
+	return nv_connector;
+}
+
 static void
-nv04_display_fini(struct drm_device *dev, bool suspend)
+nv04_display_fini(struct drm_device *dev, bool runtime, bool suspend)
 {
+	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nv04_display *disp = nv04_display(dev);
 	struct drm_crtc *crtc;
 
@@ -49,6 +68,9 @@ nv04_display_fini(struct drm_device *dev, bool suspend)
 	if (nv_two_heads(dev))
 		NVWriteCRTC(dev, 1, NV_PCRTC_INTR_EN_0, 0);
 
+	if (!runtime)
+		cancel_work_sync(&drm->hpd_work);
+
 	if (!suspend)
 		return;
 
@@ -112,7 +134,7 @@ nv04_display_init(struct drm_device *dev, bool resume, bool runtime)
 		if (!fb || !fb->obj[0])
 			continue;
 		nvbo = nouveau_gem_object(fb->obj[0]);
-		ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, true);
+		ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, true);
 		if (ret)
 			NV_ERROR(drm, "Could not pin framebuffer\n");
 	}
@@ -122,7 +144,8 @@ nv04_display_init(struct drm_device *dev, bool resume, bool runtime)
 		if (!nv_crtc->cursor.nvbo)
 			continue;
 
-		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo, TTM_PL_FLAG_VRAM, true);
+		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo,
+				     NOUVEAU_GEM_DOMAIN_VRAM, true);
 		if (!ret && nv_crtc->cursor.set_offset)
 			ret = nouveau_bo_map(nv_crtc->cursor.nvbo);
 		if (ret)
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.h b/drivers/gpu/drm/nouveau/dispnv04/disp.h
index 495d3284e876..5ace5e906949 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.h
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.h
@@ -6,6 +6,8 @@
 
 #include "nouveau_display.h"
 
+struct nouveau_encoder;
+
 enum nv04_fp_display_regs {
 	FP_DISPLAY_END,
 	FP_TOTAL,
@@ -93,6 +95,8 @@ nv04_display(struct drm_device *dev)
 
 /* nv04_display.c */
 int nv04_display_create(struct drm_device *);
+struct nouveau_connector *
+nv04_encoder_get_connector(struct nouveau_encoder *nv_encoder);
 
 /* nv04_crtc.c */
 int nv04_crtc_create(struct drm_device *, int index);
diff --git a/drivers/gpu/drm/nouveau/dispnv04/overlay.c b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
index 193ba9498f3d..37e63e98cd08 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/overlay.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
@@ -142,7 +142,7 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		return ret;
 
 	nvbo = nouveau_gem_object(fb->obj[0]);
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, false);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 	if (ret)
 		return ret;
 
@@ -387,7 +387,7 @@ nv04_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		return ret;
 
 	nvbo = nouveau_gem_object(fb->obj[0]);
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, false);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
index b701a4d8fe76..3ba7b59580d5 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
@@ -172,7 +172,7 @@ static void nv04_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 nouveau_encoder_connector_get(nv_encoder)->base.name,
+		 nv04_encoder_get_connector(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
index 3a9489ed6544..be28e7bd7490 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
@@ -599,7 +599,7 @@ static void nv17_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_INFO(drm, "Output %s is running on CRTC %d using output %c\n",
-		nouveau_encoder_connector_get(nv_encoder)->base.name,
+		nv04_encoder_get_connector(nv_encoder)->base.name,
 		nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 1ed242070001..b111fe24a06b 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -417,6 +417,40 @@ nv50_outp_atomic_check(struct drm_encoder *encoder,
 	return 0;
 }
 
+struct nouveau_connector *
+nv50_outp_get_new_connector(struct nouveau_encoder *outp,
+			    struct drm_atomic_state *state)
+{
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
+	struct drm_encoder *encoder = to_drm_encoder(outp);
+	int i;
+
+	for_each_new_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->best_encoder == encoder)
+			return nouveau_connector(connector);
+	}
+
+	return NULL;
+}
+
+struct nouveau_connector *
+nv50_outp_get_old_connector(struct nouveau_encoder *outp,
+			    struct drm_atomic_state *state)
+{
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
+	struct drm_encoder *encoder = to_drm_encoder(outp);
+	int i;
+
+	for_each_old_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->best_encoder == encoder)
+			return nouveau_connector(connector);
+	}
+
+	return NULL;
+}
+
 /******************************************************************************
  * DAC
  *****************************************************************************/
@@ -558,16 +592,31 @@ nv50_audio_component_get_eld(struct device *kdev, int port, int dev_id,
 	struct nouveau_drm *drm = nouveau_drm(drm_dev);
 	struct drm_encoder *encoder;
 	struct nouveau_encoder *nv_encoder;
-	struct nouveau_connector *nv_connector;
+	struct drm_connector *connector;
 	struct nouveau_crtc *nv_crtc;
+	struct drm_connector_list_iter conn_iter;
 	int ret = 0;
 
 	*enabled = false;
+
 	drm_for_each_encoder(encoder, drm->dev) {
+		struct nouveau_connector *nv_connector = NULL;
+
 		nv_encoder = nouveau_encoder(encoder);
-		nv_connector = nouveau_encoder_connector_get(nv_encoder);
+
+		drm_connector_list_iter_begin(drm_dev, &conn_iter);
+		drm_for_each_connector_iter(connector, &conn_iter) {
+			if (connector->state->best_encoder == encoder) {
+				nv_connector = nouveau_connector(connector);
+				break;
+			}
+		}
+		drm_connector_list_iter_end(&conn_iter);
+		if (!nv_connector)
+			continue;
+
 		nv_crtc = nouveau_crtc(encoder->crtc);
-		if (!nv_connector || !nv_crtc || nv_encoder->or != port ||
+		if (!nv_crtc || nv_encoder->or != port ||
 		    nv_crtc->index != dev_id)
 			continue;
 		*enabled = nv_encoder->audio;
@@ -578,6 +627,7 @@ nv50_audio_component_get_eld(struct device *kdev, int port, int dev_id,
 		}
 		break;
 	}
+
 	return ret;
 }
 
@@ -671,7 +721,8 @@ nv50_audio_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc)
 }
 
 static void
-nv50_audio_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
+nv50_audio_enable(struct drm_encoder *encoder, struct drm_atomic_state *state,
+		  struct drm_display_mode *mode)
 {
 	struct nouveau_drm *drm = nouveau_drm(encoder->dev);
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
@@ -692,7 +743,7 @@ nv50_audio_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
 				     (0x0100 << nv_crtc->index),
 	};
 
-	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	nv_connector = nv50_outp_get_new_connector(nv_encoder, state);
 	if (!drm_detect_monitor_audio(nv_connector->edid))
 		return;
 
@@ -729,7 +780,8 @@ nv50_hdmi_disable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc)
 }
 
 static void
-nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
+nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_atomic_state *state,
+		 struct drm_display_mode *mode)
 {
 	struct nouveau_drm *drm = nouveau_drm(encoder->dev);
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
@@ -758,7 +810,7 @@ nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
 	int ret;
 	int size;
 
-	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	nv_connector = nv50_outp_get_new_connector(nv_encoder, state);
 	if (!drm_detect_hdmi_monitor(nv_connector->edid))
 		return;
 
@@ -804,7 +856,7 @@ nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
 		+ args.pwr.vendor_infoframe_length;
 	nvif_mthd(&disp->disp->object, 0, &args, size);
 
-	nv50_audio_enable(encoder, mode);
+	nv50_audio_enable(encoder, state, mode);
 
 	/* If SCDC is supported by the downstream monitor, update
 	 * divider / scrambling settings to what we programmed above.
@@ -833,16 +885,6 @@ nv50_hdmi_enable(struct drm_encoder *encoder, struct drm_display_mode *mode)
 #define nv50_mstc(p) container_of((p), struct nv50_mstc, connector)
 #define nv50_msto(p) container_of((p), struct nv50_msto, encoder)
 
-struct nv50_mstm {
-	struct nouveau_encoder *outp;
-
-	struct drm_dp_mst_topology_mgr mgr;
-
-	bool modified;
-	bool disabled;
-	int links;
-};
-
 struct nv50_mstc {
 	struct nv50_mstm *mstm;
 	struct drm_dp_mst_port *port;
@@ -1222,7 +1264,10 @@ nv50_mstc_detect(struct drm_connector *connector,
 
 	ret = drm_dp_mst_detect_port(connector, ctx, mstc->port->mgr,
 				     mstc->port);
+	if (ret != connector_status_connected)
+		goto out;
 
+out:
 	pm_runtime_mark_last_busy(connector->dev->dev);
 	pm_runtime_put_autosuspend(connector->dev->dev);
 	return ret;
@@ -1371,41 +1416,51 @@ nv50_mstm = {
 	.add_connector = nv50_mstm_add_connector,
 };
 
-void
-nv50_mstm_service(struct nv50_mstm *mstm)
+bool
+nv50_mstm_service(struct nouveau_drm *drm,
+		  struct nouveau_connector *nv_connector,
+		  struct nv50_mstm *mstm)
 {
-	struct drm_dp_aux *aux = mstm ? mstm->mgr.aux : NULL;
-	bool handled = true;
-	int ret;
+	struct drm_dp_aux *aux = &nv_connector->aux;
+	bool handled = true, ret = true;
+	int rc;
 	u8 esi[8] = {};
 
-	if (!aux)
-		return;
-
 	while (handled) {
-		ret = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8);
-		if (ret != 8) {
-			drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false);
-			return;
+		rc = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8);
+		if (rc != 8) {
+			ret = false;
+			break;
 		}
 
 		drm_dp_mst_hpd_irq(&mstm->mgr, esi, &handled);
 		if (!handled)
 			break;
 
-		drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1], 3);
+		rc = drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1],
+				       3);
+		if (rc != 3) {
+			ret = false;
+			break;
+		}
 	}
+
+	if (!ret)
+		NV_DEBUG(drm, "Failed to handle ESI on %s: %d\n",
+			 nv_connector->base.name, rc);
+
+	return ret;
 }
 
 void
 nv50_mstm_remove(struct nv50_mstm *mstm)
 {
-	if (mstm)
-		drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false);
+	mstm->is_mst = false;
+	drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false);
 }
 
 static int
-nv50_mstm_enable(struct nv50_mstm *mstm, u8 dpcd, int state)
+nv50_mstm_enable(struct nv50_mstm *mstm, int state)
 {
 	struct nouveau_encoder *outp = mstm->outp;
 	struct {
@@ -1420,106 +1475,85 @@ nv50_mstm_enable(struct nv50_mstm *mstm, u8 dpcd, int state)
 	};
 	struct nouveau_drm *drm = nouveau_drm(outp->base.base.dev);
 	struct nvif_object *disp = &drm->display->disp.object;
-	int ret;
-
-	if (dpcd >= 0x12) {
-		/* Even if we're enabling MST, start with disabling the
-		 * branching unit to clear any sink-side MST topology state
-		 * that wasn't set by us
-		 */
-		ret = drm_dp_dpcd_writeb(mstm->mgr.aux, DP_MSTM_CTRL, 0);
-		if (ret < 0)
-			return ret;
-
-		if (state) {
-			/* Now, start initializing */
-			ret = drm_dp_dpcd_writeb(mstm->mgr.aux, DP_MSTM_CTRL,
-						 DP_MST_EN);
-			if (ret < 0)
-				return ret;
-		}
-	}
 
 	return nvif_mthd(disp, 0, &args, sizeof(args));
 }
 
 int
-nv50_mstm_detect(struct nv50_mstm *mstm, u8 dpcd[8], int allow)
+nv50_mstm_detect(struct nouveau_encoder *outp)
 {
+	struct nv50_mstm *mstm = outp->dp.mstm;
 	struct drm_dp_aux *aux;
 	int ret;
-	bool old_state, new_state;
-	u8 mstm_ctrl;
 
-	if (!mstm)
+	if (!mstm || !mstm->can_mst)
 		return 0;
 
-	mutex_lock(&mstm->mgr.lock);
-
-	old_state = mstm->mgr.mst_state;
-	new_state = old_state;
 	aux = mstm->mgr.aux;
 
-	if (old_state) {
-		/* Just check that the MST hub is still as we expect it */
-		ret = drm_dp_dpcd_readb(aux, DP_MSTM_CTRL, &mstm_ctrl);
-		if (ret < 0 || !(mstm_ctrl & DP_MST_EN)) {
-			DRM_DEBUG_KMS("Hub gone, disabling MST topology\n");
-			new_state = false;
-		}
-	} else if (dpcd[0] >= 0x12) {
-		ret = drm_dp_dpcd_readb(aux, DP_MSTM_CAP, &dpcd[1]);
-		if (ret < 0)
-			goto probe_error;
-
-		if (!(dpcd[1] & DP_MST_CAP))
-			dpcd[0] = 0x11;
-		else
-			new_state = allow;
-	}
-
-	if (new_state == old_state) {
-		mutex_unlock(&mstm->mgr.lock);
-		return new_state;
-	}
-
-	ret = nv50_mstm_enable(mstm, dpcd[0], new_state);
-	if (ret)
-		goto probe_error;
-
-	mutex_unlock(&mstm->mgr.lock);
+	/* Clear any leftover MST state we didn't set ourselves by first
+	 * disabling MST if it was already enabled
+	 */
+	ret = drm_dp_dpcd_writeb(aux, DP_MSTM_CTRL, 0);
+	if (ret < 0)
+		return ret;
 
-	ret = drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, new_state);
+	/* And start enabling */
+	ret = nv50_mstm_enable(mstm, true);
 	if (ret)
-		return nv50_mstm_enable(mstm, dpcd[0], 0);
+		return ret;
 
-	return new_state;
+	ret = drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, true);
+	if (ret) {
+		nv50_mstm_enable(mstm, false);
+		return ret;
+	}
 
-probe_error:
-	mutex_unlock(&mstm->mgr.lock);
-	return ret;
+	mstm->is_mst = true;
+	return 1;
 }
 
 static void
-nv50_mstm_fini(struct nv50_mstm *mstm)
+nv50_mstm_fini(struct nouveau_encoder *outp)
 {
-	if (mstm && mstm->mgr.mst_state)
+	struct nv50_mstm *mstm = outp->dp.mstm;
+
+	if (!mstm)
+		return;
+
+	/* Don't change the MST state of this connector until we've finished
+	 * resuming, since we can't safely grab hpd_irq_lock in our resume
+	 * path to protect mstm->is_mst without potentially deadlocking
+	 */
+	mutex_lock(&outp->dp.hpd_irq_lock);
+	mstm->suspended = true;
+	mutex_unlock(&outp->dp.hpd_irq_lock);
+
+	if (mstm->is_mst)
 		drm_dp_mst_topology_mgr_suspend(&mstm->mgr);
 }
 
 static void
-nv50_mstm_init(struct nv50_mstm *mstm, bool runtime)
+nv50_mstm_init(struct nouveau_encoder *outp, bool runtime)
 {
-	int ret;
+	struct nv50_mstm *mstm = outp->dp.mstm;
+	int ret = 0;
 
-	if (!mstm || !mstm->mgr.mst_state)
+	if (!mstm)
 		return;
 
-	ret = drm_dp_mst_topology_mgr_resume(&mstm->mgr, !runtime);
-	if (ret == -1) {
-		drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false);
-		drm_kms_helper_hotplug_event(mstm->mgr.dev);
+	if (mstm->is_mst) {
+		ret = drm_dp_mst_topology_mgr_resume(&mstm->mgr, !runtime);
+		if (ret == -1)
+			nv50_mstm_remove(mstm);
 	}
+
+	mutex_lock(&outp->dp.hpd_irq_lock);
+	mstm->suspended = false;
+	mutex_unlock(&outp->dp.hpd_irq_lock);
+
+	if (ret == -1)
+		drm_kms_helper_hotplug_event(mstm->mgr.dev);
 }
 
 static void
@@ -1541,17 +1575,6 @@ nv50_mstm_new(struct nouveau_encoder *outp, struct drm_dp_aux *aux, int aux_max,
 	struct drm_device *dev = outp->base.base.dev;
 	struct nv50_mstm *mstm;
 	int ret;
-	u8 dpcd;
-
-	/* This is a workaround for some monitors not functioning
-	 * correctly in MST mode on initial module load.  I think
-	 * some bad interaction with the VBIOS may be responsible.
-	 *
-	 * A good ol' off and on again seems to work here ;)
-	 */
-	ret = drm_dp_dpcd_readb(aux, DP_DPCD_REV, &dpcd);
-	if (ret >= 0 && dpcd >= 0x12)
-		drm_dp_dpcd_writeb(aux, DP_MSTM_CTRL, 0);
 
 	if (!(mstm = *pmstm = kzalloc(sizeof(*mstm), GFP_KERNEL)))
 		return -ENOMEM;
@@ -1590,23 +1613,27 @@ nv50_sor_update(struct nouveau_encoder *nv_encoder, u8 head,
 }
 
 static void
-nv50_sor_disable(struct drm_encoder *encoder)
+nv50_sor_disable(struct drm_encoder *encoder,
+		 struct drm_atomic_state *state)
 {
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
+	struct nouveau_connector *nv_connector =
+		nv50_outp_get_old_connector(nv_encoder, state);
 
 	nv_encoder->crtc = NULL;
 
 	if (nv_crtc) {
-		struct nvkm_i2c_aux *aux = nv_encoder->aux;
+		struct drm_dp_aux *aux = &nv_connector->aux;
 		u8 pwr;
 
-		if (aux) {
-			int ret = nvkm_rdaux(aux, DP_SET_POWER, &pwr, 1);
+		if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+			int ret = drm_dp_dpcd_readb(aux, DP_SET_POWER, &pwr);
+
 			if (ret == 0) {
 				pwr &= ~DP_SET_POWER_MASK;
 				pwr |=  DP_SET_POWER_D3;
-				nvkm_wraux(aux, DP_SET_POWER, &pwr, 1);
+				drm_dp_dpcd_writeb(aux, DP_SET_POWER, pwr);
 			}
 		}
 
@@ -1618,7 +1645,8 @@ nv50_sor_disable(struct drm_encoder *encoder)
 }
 
 static void
-nv50_sor_enable(struct drm_encoder *encoder)
+nv50_sor_enable(struct drm_encoder *encoder,
+		struct drm_atomic_state *state)
 {
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
@@ -1642,7 +1670,7 @@ nv50_sor_enable(struct drm_encoder *encoder)
 	u8 proto = NV507D_SOR_SET_CONTROL_PROTOCOL_CUSTOM;
 	u8 depth = NV837D_SOR_SET_CONTROL_PIXEL_DEPTH_DEFAULT;
 
-	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	nv_connector = nv50_outp_get_new_connector(nv_encoder, state);
 	nv_encoder->crtc = encoder->crtc;
 
 	if ((disp->disp->object.oclass == GT214_DISP ||
@@ -1669,7 +1697,7 @@ nv50_sor_enable(struct drm_encoder *encoder)
 			proto = NV507D_SOR_SET_CONTROL_PROTOCOL_SINGLE_TMDS_B;
 		}
 
-		nv50_hdmi_enable(&nv_encoder->base.base, mode);
+		nv50_hdmi_enable(&nv_encoder->base.base, state, mode);
 		break;
 	case DCB_OUTPUT_LVDS:
 		proto = NV507D_SOR_SET_CONTROL_PROTOCOL_LVDS_CUSTOM;
@@ -1710,7 +1738,7 @@ nv50_sor_enable(struct drm_encoder *encoder)
 		else
 			proto = NV887D_SOR_SET_CONTROL_PROTOCOL_DP_B;
 
-		nv50_audio_enable(encoder, mode);
+		nv50_audio_enable(encoder, state, mode);
 		break;
 	default:
 		BUG();
@@ -1723,8 +1751,8 @@ nv50_sor_enable(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs
 nv50_sor_help = {
 	.atomic_check = nv50_outp_atomic_check,
-	.enable = nv50_sor_enable,
-	.disable = nv50_sor_disable,
+	.atomic_enable = nv50_sor_enable,
+	.atomic_disable = nv50_sor_disable,
 };
 
 static void
@@ -1733,6 +1761,10 @@ nv50_sor_destroy(struct drm_encoder *encoder)
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	nv50_mstm_del(&nv_encoder->dp.mstm);
 	drm_encoder_cleanup(encoder);
+
+	if (nv_encoder->dcb->type == DCB_OUTPUT_DP)
+		mutex_destroy(&nv_encoder->dp.hpd_irq_lock);
+
 	kfree(encoder);
 }
 
@@ -1792,6 +1824,8 @@ nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe)
 		struct nvkm_i2c_aux *aux =
 			nvkm_i2c_aux_find(i2c, dcbe->i2c_index);
 
+		mutex_init(&nv_encoder->dp.hpd_irq_lock);
+
 		if (aux) {
 			if (disp->disp->object.oclass < GF110_DISP) {
 				/* HW has no support for address-only
@@ -2035,6 +2069,7 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 	drm_atomic_helper_wait_for_fences(dev, state, false);
 	drm_atomic_helper_wait_for_dependencies(state);
 	drm_atomic_helper_update_legacy_modeset_state(dev, state);
+	drm_atomic_helper_calc_timestamping_constants(state);
 
 	if (atom->lock_core)
 		mutex_lock(&disp->mutex);
@@ -2083,7 +2118,7 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 			  outp->clr.mask, outp->set.mask);
 
 		if (outp->clr.mask) {
-			help->disable(encoder);
+			help->atomic_disable(encoder, state);
 			interlock[NV50_DISP_INTERLOCK_CORE] |= 1;
 			if (outp->flush_disable) {
 				nv50_disp_atomic_commit_wndw(state, interlock);
@@ -2122,7 +2157,7 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 			  outp->set.mask, outp->clr.mask);
 
 		if (outp->set.mask) {
-			help->enable(encoder);
+			help->atomic_enable(encoder, state);
 			interlock[NV50_DISP_INTERLOCK_CORE] = 1;
 		}
 
@@ -2490,9 +2525,9 @@ nv50_disp_func = {
  *****************************************************************************/
 
 static void
-nv50_display_fini(struct drm_device *dev, bool suspend)
+nv50_display_fini(struct drm_device *dev, bool runtime, bool suspend)
 {
-	struct nouveau_encoder *nv_encoder;
+	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_encoder *encoder;
 	struct drm_plane *plane;
 
@@ -2504,11 +2539,12 @@ nv50_display_fini(struct drm_device *dev, bool suspend)
 	}
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) {
-			nv_encoder = nouveau_encoder(encoder);
-			nv50_mstm_fini(nv_encoder->dp.mstm);
-		}
+		if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST)
+			nv50_mstm_fini(nouveau_encoder(encoder));
 	}
+
+	if (!runtime)
+		cancel_work_sync(&drm->hpd_work);
 }
 
 static int
@@ -2525,7 +2561,7 @@ nv50_display_init(struct drm_device *dev, bool resume, bool runtime)
 		if (encoder->encoder_type != DRM_MODE_ENCODER_DPMST) {
 			struct nouveau_encoder *nv_encoder =
 				nouveau_encoder(encoder);
-			nv50_mstm_init(nv_encoder->dp.mstm, runtime);
+			nv50_mstm_init(nv_encoder, runtime);
 		}
 	}
 
@@ -2587,10 +2623,11 @@ nv50_display_create(struct drm_device *dev)
 	dev->mode_config.normalize_zpos = true;
 
 	/* small shared memory area we use for notifiers and semaphores */
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
+	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
+			     NOUVEAU_GEM_DOMAIN_VRAM,
 			     0, 0x0000, NULL, NULL, &disp->sync);
 	if (!ret) {
-		ret = nouveau_bo_pin(disp->sync, TTM_PL_FLAG_VRAM, true);
+		ret = nouveau_bo_pin(disp->sync, NOUVEAU_GEM_DOMAIN_VRAM, true);
 		if (!ret) {
 			ret = nouveau_bo_map(disp->sync);
 			if (ret)
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
index 447ecc9fec42..0356474ad6f6 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
@@ -542,7 +542,7 @@ nv50_wndw_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state)
 		return 0;
 
 	nvbo = nouveau_gem_object(fb->obj[0]);
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, true);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, true);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 21537ca1dd39..9a5be6f32424 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -328,7 +328,8 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 	ret = nouveau_gem_new(cli, PAGE_SIZE, 0, NOUVEAU_GEM_DOMAIN_GART,
 			      0, 0, &chan->ntfy);
 	if (ret == 0)
-		ret = nouveau_bo_pin(chan->ntfy, TTM_PL_FLAG_TT, false);
+		ret = nouveau_bo_pin(chan->ntfy, NOUVEAU_GEM_DOMAIN_GART,
+				     false);
 	if (ret)
 		goto done;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 7806278dce57..2ee75646ad6f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -44,6 +44,9 @@
 #include <nvif/if500b.h>
 #include <nvif/if900b.h>
 
+static int nouveau_ttm_tt_bind(struct ttm_bo_device *bdev, struct ttm_tt *ttm,
+			       struct ttm_resource *reg);
+
 /*
  * NV10-NV40 tiling helpers
  */
@@ -137,6 +140,7 @@ nouveau_bo_del_ttm(struct ttm_buffer_object *bo)
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 
 	WARN_ON(nvbo->pin_refcnt > 0);
+	nouveau_bo_del_io_reserve_lru(bo);
 	nv10_bo_put_tile_region(dev, nvbo->tile, NULL);
 
 	/*
@@ -158,8 +162,7 @@ roundup_64(u64 x, u32 y)
 }
 
 static void
-nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags,
-		       int *align, u64 *size)
+nouveau_bo_fixup_align(struct nouveau_bo *nvbo, int *align, u64 *size)
 {
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	struct nvif_device *device = &drm->client.device;
@@ -192,7 +195,7 @@ nouveau_bo_fixup_align(struct nouveau_bo *nvbo, u32 flags,
 }
 
 struct nouveau_bo *
-nouveau_bo_alloc(struct nouveau_cli *cli, u64 *size, int *align, u32 flags,
+nouveau_bo_alloc(struct nouveau_cli *cli, u64 *size, int *align, u32 domain,
 		 u32 tile_mode, u32 tile_flags)
 {
 	struct nouveau_drm *drm = cli->drm;
@@ -218,7 +221,7 @@ nouveau_bo_alloc(struct nouveau_cli *cli, u64 *size, int *align, u32 flags,
 	 * mapping, but is what NOUVEAU_GEM_DOMAIN_COHERENT gets translated
 	 * into in nouveau_gem_new().
 	 */
-	if (flags & TTM_PL_FLAG_UNCACHED) {
+	if (domain & NOUVEAU_GEM_DOMAIN_COHERENT) {
 		/* Determine if we can get a cache-coherent map, forcing
 		 * uncached mapping if we can't.
 		 */
@@ -258,9 +261,9 @@ nouveau_bo_alloc(struct nouveau_cli *cli, u64 *size, int *align, u32 flags,
 		 * Skip page sizes that can't support needed domains.
 		 */
 		if (cli->device.info.family > NV_DEVICE_INFO_V0_CURIE &&
-		    (flags & TTM_PL_FLAG_VRAM) && !vmm->page[i].vram)
+		    (domain & NOUVEAU_GEM_DOMAIN_VRAM) && !vmm->page[i].vram)
 			continue;
-		if ((flags & TTM_PL_FLAG_TT) &&
+		if ((domain & NOUVEAU_GEM_DOMAIN_GART) &&
 		    (!vmm->page[i].host || vmm->page[i].shift > PAGE_SHIFT))
 			continue;
 
@@ -287,13 +290,13 @@ nouveau_bo_alloc(struct nouveau_cli *cli, u64 *size, int *align, u32 flags,
 	}
 	nvbo->page = vmm->page[pi].shift;
 
-	nouveau_bo_fixup_align(nvbo, flags, align, size);
+	nouveau_bo_fixup_align(nvbo, align, size);
 
 	return nvbo;
 }
 
 int
-nouveau_bo_init(struct nouveau_bo *nvbo, u64 size, int align, u32 flags,
+nouveau_bo_init(struct nouveau_bo *nvbo, u64 size, int align, u32 domain,
 		struct sg_table *sg, struct dma_resv *robj)
 {
 	int type = sg ? ttm_bo_type_sg : ttm_bo_type_device;
@@ -303,7 +306,8 @@ nouveau_bo_init(struct nouveau_bo *nvbo, u64 size, int align, u32 flags,
 	acc_size = ttm_bo_dma_acc_size(nvbo->bo.bdev, size, sizeof(*nvbo));
 
 	nvbo->bo.mem.num_pages = size >> PAGE_SHIFT;
-	nouveau_bo_placement_set(nvbo, flags, 0);
+	nouveau_bo_placement_set(nvbo, domain, 0);
+	INIT_LIST_HEAD(&nvbo->io_reserve_lru);
 
 	ret = ttm_bo_init(nvbo->bo.bdev, &nvbo->bo, size, type,
 			  &nvbo->placement, align >> PAGE_SHIFT, false,
@@ -318,19 +322,19 @@ nouveau_bo_init(struct nouveau_bo *nvbo, u64 size, int align, u32 flags,
 
 int
 nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align,
-	       uint32_t flags, uint32_t tile_mode, uint32_t tile_flags,
+	       uint32_t domain, uint32_t tile_mode, uint32_t tile_flags,
 	       struct sg_table *sg, struct dma_resv *robj,
 	       struct nouveau_bo **pnvbo)
 {
 	struct nouveau_bo *nvbo;
 	int ret;
 
-	nvbo = nouveau_bo_alloc(cli, &size, &align, flags, tile_mode,
+	nvbo = nouveau_bo_alloc(cli, &size, &align, domain, tile_mode,
 				tile_flags);
 	if (IS_ERR(nvbo))
 		return PTR_ERR(nvbo);
 
-	ret = nouveau_bo_init(nvbo, size, align, flags, sg, robj);
+	ret = nouveau_bo_init(nvbo, size, align, domain, sg, robj);
 	if (ret)
 		return ret;
 
@@ -339,27 +343,49 @@ nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align,
 }
 
 static void
-set_placement_list(struct ttm_place *pl, unsigned *n, uint32_t type, uint32_t flags)
+set_placement_list(struct nouveau_drm *drm, struct ttm_place *pl, unsigned *n,
+		   uint32_t domain, uint32_t flags)
 {
 	*n = 0;
 
-	if (type & TTM_PL_FLAG_VRAM)
-		pl[(*n)++].flags = TTM_PL_FLAG_VRAM | flags;
-	if (type & TTM_PL_FLAG_TT)
-		pl[(*n)++].flags = TTM_PL_FLAG_TT | flags;
-	if (type & TTM_PL_FLAG_SYSTEM)
-		pl[(*n)++].flags = TTM_PL_FLAG_SYSTEM | flags;
+	if (domain & NOUVEAU_GEM_DOMAIN_VRAM) {
+		struct nvif_mmu *mmu = &drm->client.mmu;
+		const u8 type = mmu->type[drm->ttm.type_vram].type;
+
+		pl[*n].mem_type = TTM_PL_VRAM;
+		pl[*n].flags = flags & ~TTM_PL_FLAG_CACHED;
+
+		/* Some BARs do not support being ioremapped WC */
+		if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA &&
+		    type & NVIF_MEM_UNCACHED)
+			pl[*n].flags &= ~TTM_PL_FLAG_WC;
+
+		(*n)++;
+	}
+	if (domain & NOUVEAU_GEM_DOMAIN_GART) {
+		pl[*n].mem_type = TTM_PL_TT;
+		pl[*n].flags = flags;
+
+		if (drm->agp.bridge)
+			pl[*n].flags &= ~TTM_PL_FLAG_CACHED;
+
+		(*n)++;
+	}
+	if (domain & NOUVEAU_GEM_DOMAIN_CPU) {
+		pl[*n].mem_type = TTM_PL_SYSTEM;
+		pl[(*n)++].flags = flags;
+	}
 }
 
 static void
-set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
+set_placement_range(struct nouveau_bo *nvbo, uint32_t domain)
 {
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	u32 vram_pages = drm->client.device.info.ram_size >> PAGE_SHIFT;
 	unsigned i, fpfn, lpfn;
 
 	if (drm->client.device.info.family == NV_DEVICE_INFO_V0_CELSIUS &&
-	    nvbo->mode && (type & TTM_PL_FLAG_VRAM) &&
+	    nvbo->mode && (domain & NOUVEAU_GEM_DOMAIN_VRAM) &&
 	    nvbo->bo.mem.num_pages < vram_pages / 4) {
 		/*
 		 * Make sure that the color and depth buffers are handled
@@ -386,26 +412,28 @@ set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
 }
 
 void
-nouveau_bo_placement_set(struct nouveau_bo *nvbo, uint32_t type, uint32_t busy)
+nouveau_bo_placement_set(struct nouveau_bo *nvbo, uint32_t domain,
+			 uint32_t busy)
 {
+	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	struct ttm_placement *pl = &nvbo->placement;
 	uint32_t flags = (nvbo->force_coherent ? TTM_PL_FLAG_UNCACHED :
 						 TTM_PL_MASK_CACHING) |
 			 (nvbo->pin_refcnt ? TTM_PL_FLAG_NO_EVICT : 0);
 
 	pl->placement = nvbo->placements;
-	set_placement_list(nvbo->placements, &pl->num_placement,
-			   type, flags);
+	set_placement_list(drm, nvbo->placements, &pl->num_placement,
+			   domain, flags);
 
 	pl->busy_placement = nvbo->busy_placements;
-	set_placement_list(nvbo->busy_placements, &pl->num_busy_placement,
-			   type | busy, flags);
+	set_placement_list(drm, nvbo->busy_placements, &pl->num_busy_placement,
+			   domain | busy, flags);
 
-	set_placement_range(nvbo, type);
+	set_placement_range(nvbo, domain);
 }
 
 int
-nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype, bool contig)
+nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t domain, bool contig)
 {
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	struct ttm_buffer_object *bo = &nvbo->bo;
@@ -417,7 +445,7 @@ nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype, bool contig)
 		return ret;
 
 	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA &&
-	    memtype == TTM_PL_FLAG_VRAM && contig) {
+	    domain == NOUVEAU_GEM_DOMAIN_VRAM && contig) {
 		if (!nvbo->contig) {
 			nvbo->contig = true;
 			force = true;
@@ -426,10 +454,22 @@ nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype, bool contig)
 	}
 
 	if (nvbo->pin_refcnt) {
-		if (!(memtype & (1 << bo->mem.mem_type)) || evict) {
+		bool error = evict;
+
+		switch (bo->mem.mem_type) {
+		case TTM_PL_VRAM:
+			error |= !(domain & NOUVEAU_GEM_DOMAIN_VRAM);
+			break;
+		case TTM_PL_TT:
+			error |= !(domain & NOUVEAU_GEM_DOMAIN_GART);
+		default:
+			break;
+		}
+
+		if (error) {
 			NV_ERROR(drm, "bo %p pinned elsewhere: "
 				      "0x%08x vs 0x%08x\n", bo,
-				 1 << bo->mem.mem_type, memtype);
+				 bo->mem.mem_type, domain);
 			ret = -EBUSY;
 		}
 		nvbo->pin_refcnt++;
@@ -437,14 +477,14 @@ nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype, bool contig)
 	}
 
 	if (evict) {
-		nouveau_bo_placement_set(nvbo, TTM_PL_FLAG_TT, 0);
+		nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_GART, 0);
 		ret = nouveau_bo_validate(nvbo, false, false);
 		if (ret)
 			goto out;
 	}
 
 	nvbo->pin_refcnt++;
-	nouveau_bo_placement_set(nvbo, memtype, 0);
+	nouveau_bo_placement_set(nvbo, domain, 0);
 
 	/* drop pin_refcnt temporarily, so we don't trip the assertion
 	 * in nouveau_bo_move() that makes sure we're not trying to
@@ -490,7 +530,16 @@ nouveau_bo_unpin(struct nouveau_bo *nvbo)
 	if (ref)
 		goto out;
 
-	nouveau_bo_placement_set(nvbo, bo->mem.placement, 0);
+	switch (bo->mem.mem_type) {
+	case TTM_PL_VRAM:
+		nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, 0);
+		break;
+	case TTM_PL_TT:
+		nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_GART, 0);
+		break;
+	default:
+		break;
+	}
 
 	ret = nouveau_bo_validate(nvbo, false, false);
 	if (ret == 0) {
@@ -574,6 +623,26 @@ nouveau_bo_sync_for_cpu(struct nouveau_bo *nvbo)
 					PAGE_SIZE, DMA_FROM_DEVICE);
 }
 
+void nouveau_bo_add_io_reserve_lru(struct ttm_buffer_object *bo)
+{
+	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
+	struct nouveau_bo *nvbo = nouveau_bo(bo);
+
+	mutex_lock(&drm->ttm.io_reserve_mutex);
+	list_move_tail(&nvbo->io_reserve_lru, &drm->ttm.io_reserve_lru);
+	mutex_unlock(&drm->ttm.io_reserve_mutex);
+}
+
+void nouveau_bo_del_io_reserve_lru(struct ttm_buffer_object *bo)
+{
+	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
+	struct nouveau_bo *nvbo = nouveau_bo(bo);
+
+	mutex_lock(&drm->ttm.io_reserve_mutex);
+	list_del_init(&nvbo->io_reserve_lru);
+	mutex_unlock(&drm->ttm.io_reserve_mutex);
+}
+
 int
 nouveau_bo_validate(struct nouveau_bo *nvbo, bool interruptible,
 		    bool no_wait_gpu)
@@ -647,63 +716,33 @@ nouveau_ttm_tt_create(struct ttm_buffer_object *bo, uint32_t page_flags)
 }
 
 static int
-nouveau_bo_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-			 struct ttm_mem_type_manager *man)
+nouveau_ttm_tt_bind(struct ttm_bo_device *bdev, struct ttm_tt *ttm,
+		    struct ttm_resource *reg)
 {
+#if IS_ENABLED(CONFIG_AGP)
 	struct nouveau_drm *drm = nouveau_bdev(bdev);
-	struct nvif_mmu *mmu = &drm->client.mmu;
-
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		man->flags = 0;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_VRAM:
-		man->flags = TTM_MEMTYPE_FLAG_FIXED;
-		man->available_caching = TTM_PL_FLAG_UNCACHED |
-					 TTM_PL_FLAG_WC;
-		man->default_caching = TTM_PL_FLAG_WC;
-
-		if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA) {
-			/* Some BARs do not support being ioremapped WC */
-			const u8 type = mmu->type[drm->ttm.type_vram].type;
-			if (type & NVIF_MEM_UNCACHED) {
-				man->available_caching = TTM_PL_FLAG_UNCACHED;
-				man->default_caching = TTM_PL_FLAG_UNCACHED;
-			}
-
-			man->func = &nouveau_vram_manager;
-			man->use_io_reserve_lru = true;
-		} else {
-			man->func = &ttm_bo_manager_func;
-		}
-		break;
-	case TTM_PL_TT:
-		if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA)
-			man->func = &nouveau_gart_manager;
-		else
-		if (!drm->agp.bridge)
-			man->func = &nv04_gart_manager;
-		else
-			man->func = &ttm_bo_manager_func;
+#endif
+	if (!reg)
+		return -EINVAL;
+#if IS_ENABLED(CONFIG_AGP)
+	if (drm->agp.bridge)
+		return ttm_agp_bind(ttm, reg);
+#endif
+	return nouveau_sgdma_bind(bdev, ttm, reg);
+}
 
-		if (drm->agp.bridge) {
-			man->flags = 0;
-			man->available_caching = TTM_PL_FLAG_UNCACHED |
-				TTM_PL_FLAG_WC;
-			man->default_caching = TTM_PL_FLAG_WC;
-		} else {
-			man->flags = 0;
-			man->available_caching = TTM_PL_MASK_CACHING;
-			man->default_caching = TTM_PL_FLAG_CACHED;
-		}
+static void
+nouveau_ttm_tt_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
+{
+#if IS_ENABLED(CONFIG_AGP)
+	struct nouveau_drm *drm = nouveau_bdev(bdev);
 
-		break;
-	default:
-		return -EINVAL;
+	if (drm->agp.bridge) {
+		ttm_agp_unbind(ttm);
+		return;
 	}
-	return 0;
+#endif
+	nouveau_sgdma_unbind(bdev, ttm);
 }
 
 static void
@@ -713,11 +752,11 @@ nouveau_bo_evict_flags(struct ttm_buffer_object *bo, struct ttm_placement *pl)
 
 	switch (bo->mem.mem_type) {
 	case TTM_PL_VRAM:
-		nouveau_bo_placement_set(nvbo, TTM_PL_FLAG_TT,
-					 TTM_PL_FLAG_SYSTEM);
+		nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_GART,
+					 NOUVEAU_GEM_DOMAIN_CPU);
 		break;
 	default:
-		nouveau_bo_placement_set(nvbo, TTM_PL_FLAG_SYSTEM, 0);
+		nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_CPU, 0);
 		break;
 	}
 
@@ -726,7 +765,7 @@ nouveau_bo_evict_flags(struct ttm_buffer_object *bo, struct ttm_placement *pl)
 
 static int
 nouveau_bo_move_prep(struct nouveau_drm *drm, struct ttm_buffer_object *bo,
-		     struct ttm_mem_reg *reg)
+		     struct ttm_resource *reg)
 {
 	struct nouveau_mem *old_mem = nouveau_mem(&bo->mem);
 	struct nouveau_mem *new_mem = nouveau_mem(reg);
@@ -758,7 +797,7 @@ done:
 
 static int
 nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, bool intr,
-		     bool no_wait_gpu, struct ttm_mem_reg *new_reg)
+		     bool no_wait_gpu, struct ttm_resource *new_reg)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
 	struct nouveau_channel *chan = drm->ttm.chan;
@@ -768,7 +807,7 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, bool intr,
 
 	/* create temporary vmas for the transfer and attach them to the
 	 * old nvkm_mem node, these will get cleaned up after ttm has
-	 * destroyed the ttm_mem_reg
+	 * destroyed the ttm_resource
 	 */
 	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA) {
 		ret = nouveau_bo_move_prep(drm, bo, new_reg);
@@ -785,7 +824,7 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, bool intr,
 			if (ret == 0) {
 				ret = ttm_bo_move_accel_cleanup(bo,
 								&fence->base,
-								evict,
+								evict, false,
 								new_reg);
 				nouveau_fence_unref(&fence);
 			}
@@ -804,7 +843,7 @@ nouveau_bo_move_init(struct nouveau_drm *drm)
 		s32 oclass;
 		int (*exec)(struct nouveau_channel *,
 			    struct ttm_buffer_object *,
-			    struct ttm_mem_reg *, struct ttm_mem_reg *);
+			    struct ttm_resource *, struct ttm_resource *);
 		int (*init)(struct nouveau_channel *, u32 handle);
 	} _methods[] = {
 		{  "COPY", 4, 0xc5b5, nve0_bo_move_copy, nve0_bo_move_init },
@@ -865,16 +904,17 @@ nouveau_bo_move_init(struct nouveau_drm *drm)
 
 static int
 nouveau_bo_move_flipd(struct ttm_buffer_object *bo, bool evict, bool intr,
-		      bool no_wait_gpu, struct ttm_mem_reg *new_reg)
+		      bool no_wait_gpu, struct ttm_resource *new_reg)
 {
 	struct ttm_operation_ctx ctx = { intr, no_wait_gpu };
 	struct ttm_place placement_memtype = {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_TT | TTM_PL_MASK_CACHING
+		.mem_type = TTM_PL_TT,
+		.flags = TTM_PL_MASK_CACHING
 	};
 	struct ttm_placement placement;
-	struct ttm_mem_reg tmp_reg;
+	struct ttm_resource tmp_reg;
 	int ret;
 
 	placement.num_placement = placement.num_busy_placement = 1;
@@ -886,7 +926,11 @@ nouveau_bo_move_flipd(struct ttm_buffer_object *bo, bool evict, bool intr,
 	if (ret)
 		return ret;
 
-	ret = ttm_tt_bind(bo->ttm, &tmp_reg, &ctx);
+	ret = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
+	if (ret)
+		goto out;
+
+	ret = nouveau_ttm_tt_bind(bo->bdev, bo->ttm, &tmp_reg);
 	if (ret)
 		goto out;
 
@@ -896,22 +940,23 @@ nouveau_bo_move_flipd(struct ttm_buffer_object *bo, bool evict, bool intr,
 
 	ret = ttm_bo_move_ttm(bo, &ctx, new_reg);
 out:
-	ttm_bo_mem_put(bo, &tmp_reg);
+	ttm_resource_free(bo, &tmp_reg);
 	return ret;
 }
 
 static int
 nouveau_bo_move_flips(struct ttm_buffer_object *bo, bool evict, bool intr,
-		      bool no_wait_gpu, struct ttm_mem_reg *new_reg)
+		      bool no_wait_gpu, struct ttm_resource *new_reg)
 {
 	struct ttm_operation_ctx ctx = { intr, no_wait_gpu };
 	struct ttm_place placement_memtype = {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_TT | TTM_PL_MASK_CACHING
+		.mem_type = TTM_PL_TT,
+		.flags = TTM_PL_MASK_CACHING
 	};
 	struct ttm_placement placement;
-	struct ttm_mem_reg tmp_reg;
+	struct ttm_resource tmp_reg;
 	int ret;
 
 	placement.num_placement = placement.num_busy_placement = 1;
@@ -932,13 +977,13 @@ nouveau_bo_move_flips(struct ttm_buffer_object *bo, bool evict, bool intr,
 		goto out;
 
 out:
-	ttm_bo_mem_put(bo, &tmp_reg);
+	ttm_resource_free(bo, &tmp_reg);
 	return ret;
 }
 
 static void
 nouveau_bo_move_ntfy(struct ttm_buffer_object *bo, bool evict,
-		     struct ttm_mem_reg *new_reg)
+		     struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = new_reg ? nouveau_mem(new_reg) : NULL;
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
@@ -948,6 +993,8 @@ nouveau_bo_move_ntfy(struct ttm_buffer_object *bo, bool evict,
 	if (bo->destroy != nouveau_bo_del_ttm)
 		return;
 
+	nouveau_bo_del_io_reserve_lru(bo);
+
 	if (mem && new_reg->mem_type != TTM_PL_SYSTEM &&
 	    mem->mem.page == nvbo->page) {
 		list_for_each_entry(vma, &nvbo->vma_list, head) {
@@ -970,7 +1017,7 @@ nouveau_bo_move_ntfy(struct ttm_buffer_object *bo, bool evict,
 }
 
 static int
-nouveau_bo_vm_bind(struct ttm_buffer_object *bo, struct ttm_mem_reg *new_reg,
+nouveau_bo_vm_bind(struct ttm_buffer_object *bo, struct ttm_resource *new_reg,
 		   struct nouveau_drm_tile **new_tile)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
@@ -1006,11 +1053,11 @@ nouveau_bo_vm_cleanup(struct ttm_buffer_object *bo,
 static int
 nouveau_bo_move(struct ttm_buffer_object *bo, bool evict,
 		struct ttm_operation_ctx *ctx,
-		struct ttm_mem_reg *new_reg)
+		struct ttm_resource *new_reg)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
-	struct ttm_mem_reg *old_reg = &bo->mem;
+	struct ttm_resource *old_reg = &bo->mem;
 	struct nouveau_drm_tile *new_tile = NULL;
 	int ret = 0;
 
@@ -1029,9 +1076,7 @@ nouveau_bo_move(struct ttm_buffer_object *bo, bool evict,
 
 	/* Fake bo copy. */
 	if (old_reg->mem_type == TTM_PL_SYSTEM && !bo->ttm) {
-		BUG_ON(bo->mem.mm_node != NULL);
-		bo->mem = *new_reg;
-		new_reg->mm_node = NULL;
+		ttm_bo_move_null(bo, new_reg);
 		goto out;
 	}
 
@@ -1078,38 +1123,60 @@ nouveau_bo_verify_access(struct ttm_buffer_object *bo, struct file *filp)
 					  filp->private_data);
 }
 
+static void
+nouveau_ttm_io_mem_free_locked(struct nouveau_drm *drm,
+			       struct ttm_resource *reg)
+{
+	struct nouveau_mem *mem = nouveau_mem(reg);
+
+	if (drm->client.mem->oclass >= NVIF_CLASS_MEM_NV50) {
+		switch (reg->mem_type) {
+		case TTM_PL_TT:
+			if (mem->kind)
+				nvif_object_unmap_handle(&mem->mem.object);
+			break;
+		case TTM_PL_VRAM:
+			nvif_object_unmap_handle(&mem->mem.object);
+			break;
+		default:
+			break;
+		}
+	}
+}
+
 static int
-nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg)
+nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_resource *reg)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bdev);
 	struct nvkm_device *device = nvxx_device(&drm->client.device);
 	struct nouveau_mem *mem = nouveau_mem(reg);
+	int ret;
 
-	reg->bus.addr = NULL;
-	reg->bus.offset = 0;
-	reg->bus.size = reg->num_pages << PAGE_SHIFT;
-	reg->bus.base = 0;
-	reg->bus.is_iomem = false;
-
+	mutex_lock(&drm->ttm.io_reserve_mutex);
+retry:
 	switch (reg->mem_type) {
 	case TTM_PL_SYSTEM:
 		/* System memory */
-		return 0;
+		ret = 0;
+		goto out;
 	case TTM_PL_TT:
 #if IS_ENABLED(CONFIG_AGP)
 		if (drm->agp.bridge) {
-			reg->bus.offset = reg->start << PAGE_SHIFT;
-			reg->bus.base = drm->agp.base;
+			reg->bus.offset = (reg->start << PAGE_SHIFT) +
+				drm->agp.base;
 			reg->bus.is_iomem = !drm->agp.cma;
 		}
 #endif
-		if (drm->client.mem->oclass < NVIF_CLASS_MEM_NV50 || !mem->kind)
+		if (drm->client.mem->oclass < NVIF_CLASS_MEM_NV50 ||
+		    !mem->kind) {
 			/* untiled */
+			ret = 0;
 			break;
+		}
 		fallthrough;	/* tiled memory */
 	case TTM_PL_VRAM:
-		reg->bus.offset = reg->start << PAGE_SHIFT;
-		reg->bus.base = device->func->resource_addr(device, 1);
+		reg->bus.offset = (reg->start << PAGE_SHIFT) +
+			device->func->resource_addr(device, 1);
 		reg->bus.is_iomem = true;
 		if (drm->client.mem->oclass >= NVIF_CLASS_MEM_NV50) {
 			union {
@@ -1118,7 +1185,6 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg)
 			} args;
 			u64 handle, length;
 			u32 argc = 0;
-			int ret;
 
 			switch (mem->mem.object.oclass) {
 			case NVIF_CLASS_MEM_NV50:
@@ -1144,39 +1210,46 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg)
 						     &handle, &length);
 			if (ret != 1) {
 				if (WARN_ON(ret == 0))
-					return -EINVAL;
-				return ret;
+					ret = -EINVAL;
+				goto out;
 			}
 
-			reg->bus.base = 0;
 			reg->bus.offset = handle;
+			ret = 0;
 		}
 		break;
 	default:
-		return -EINVAL;
+		ret = -EINVAL;
 	}
-	return 0;
+
+out:
+	if (ret == -ENOSPC) {
+		struct nouveau_bo *nvbo;
+
+		nvbo = list_first_entry_or_null(&drm->ttm.io_reserve_lru,
+						typeof(*nvbo),
+						io_reserve_lru);
+		if (nvbo) {
+			list_del_init(&nvbo->io_reserve_lru);
+			drm_vma_node_unmap(&nvbo->bo.base.vma_node,
+					   bdev->dev_mapping);
+			nouveau_ttm_io_mem_free_locked(drm, &nvbo->bo.mem);
+			goto retry;
+		}
+
+	}
+	mutex_unlock(&drm->ttm.io_reserve_mutex);
+	return ret;
 }
 
 static void
-nouveau_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg)
+nouveau_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_resource *reg)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bdev);
-	struct nouveau_mem *mem = nouveau_mem(reg);
 
-	if (drm->client.mem->oclass >= NVIF_CLASS_MEM_NV50) {
-		switch (reg->mem_type) {
-		case TTM_PL_TT:
-			if (mem->kind)
-				nvif_object_unmap_handle(&mem->mem.object);
-			break;
-		case TTM_PL_VRAM:
-			nvif_object_unmap_handle(&mem->mem.object);
-			break;
-		default:
-			break;
-		}
-	}
+	mutex_lock(&drm->ttm.io_reserve_mutex);
+	nouveau_ttm_io_mem_free_locked(drm, reg);
+	mutex_unlock(&drm->ttm.io_reserve_mutex);
 }
 
 static int
@@ -1197,7 +1270,8 @@ nouveau_ttm_fault_reserve_notify(struct ttm_buffer_object *bo)
 			return 0;
 
 		if (bo->mem.mem_type == TTM_PL_SYSTEM) {
-			nouveau_bo_placement_set(nvbo, TTM_PL_TT, 0);
+			nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_GART,
+						 0);
 
 			ret = nouveau_bo_validate(nvbo, false, false);
 			if (ret)
@@ -1221,37 +1295,36 @@ nouveau_ttm_fault_reserve_notify(struct ttm_buffer_object *bo)
 		nvbo->busy_placements[i].lpfn = mappable;
 	}
 
-	nouveau_bo_placement_set(nvbo, TTM_PL_FLAG_VRAM, 0);
+	nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, 0);
 	return nouveau_bo_validate(nvbo, false, false);
 }
 
 static int
-nouveau_ttm_tt_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
+nouveau_ttm_tt_populate(struct ttm_bo_device *bdev,
+			struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 {
 	struct ttm_dma_tt *ttm_dma = (void *)ttm;
 	struct nouveau_drm *drm;
 	struct device *dev;
-	unsigned i;
-	int r;
 	bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG);
 
-	if (ttm->state != tt_unpopulated)
+	if (ttm_tt_is_populated(ttm))
 		return 0;
 
 	if (slave && ttm->sg) {
 		/* make userspace faulting work */
 		drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
 						 ttm_dma->dma_address, ttm->num_pages);
-		ttm->state = tt_unbound;
+		ttm_tt_set_populated(ttm);
 		return 0;
 	}
 
-	drm = nouveau_bdev(ttm->bdev);
+	drm = nouveau_bdev(bdev);
 	dev = drm->dev->dev;
 
 #if IS_ENABLED(CONFIG_AGP)
 	if (drm->agp.bridge) {
-		return ttm_agp_tt_populate(ttm, ctx);
+		return ttm_pool_populate(ttm, ctx);
 	}
 #endif
 
@@ -1260,51 +1333,27 @@ nouveau_ttm_tt_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 		return ttm_dma_populate((void *)ttm, dev, ctx);
 	}
 #endif
-
-	r = ttm_pool_populate(ttm, ctx);
-	if (r) {
-		return r;
-	}
-
-	for (i = 0; i < ttm->num_pages; i++) {
-		dma_addr_t addr;
-
-		addr = dma_map_page(dev, ttm->pages[i], 0, PAGE_SIZE,
-				    DMA_BIDIRECTIONAL);
-
-		if (dma_mapping_error(dev, addr)) {
-			while (i--) {
-				dma_unmap_page(dev, ttm_dma->dma_address[i],
-					       PAGE_SIZE, DMA_BIDIRECTIONAL);
-				ttm_dma->dma_address[i] = 0;
-			}
-			ttm_pool_unpopulate(ttm);
-			return -EFAULT;
-		}
-
-		ttm_dma->dma_address[i] = addr;
-	}
-	return 0;
+	return ttm_populate_and_map_pages(dev, ttm_dma, ctx);
 }
 
 static void
-nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm)
+nouveau_ttm_tt_unpopulate(struct ttm_bo_device *bdev,
+			  struct ttm_tt *ttm)
 {
 	struct ttm_dma_tt *ttm_dma = (void *)ttm;
 	struct nouveau_drm *drm;
 	struct device *dev;
-	unsigned i;
 	bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG);
 
 	if (slave)
 		return;
 
-	drm = nouveau_bdev(ttm->bdev);
+	drm = nouveau_bdev(bdev);
 	dev = drm->dev->dev;
 
 #if IS_ENABLED(CONFIG_AGP)
 	if (drm->agp.bridge) {
-		ttm_agp_tt_unpopulate(ttm);
+		ttm_pool_unpopulate(ttm);
 		return;
 	}
 #endif
@@ -1316,14 +1365,23 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	}
 #endif
 
-	for (i = 0; i < ttm->num_pages; i++) {
-		if (ttm_dma->dma_address[i]) {
-			dma_unmap_page(dev, ttm_dma->dma_address[i], PAGE_SIZE,
-				       DMA_BIDIRECTIONAL);
-		}
-	}
+	ttm_unmap_and_unpopulate_pages(dev, ttm_dma);
+}
 
-	ttm_pool_unpopulate(ttm);
+static void
+nouveau_ttm_tt_destroy(struct ttm_bo_device *bdev,
+		       struct ttm_tt *ttm)
+{
+#if IS_ENABLED(CONFIG_AGP)
+	struct nouveau_drm *drm = nouveau_bdev(bdev);
+	if (drm->agp.bridge) {
+		ttm_agp_unbind(ttm);
+		ttm_tt_destroy_common(bdev, ttm);
+		ttm_agp_destroy(ttm);
+		return;
+	}
+#endif
+	nouveau_sgdma_destroy(bdev, ttm);
 }
 
 void
@@ -1341,7 +1399,9 @@ struct ttm_bo_driver nouveau_bo_driver = {
 	.ttm_tt_create = &nouveau_ttm_tt_create,
 	.ttm_tt_populate = &nouveau_ttm_tt_populate,
 	.ttm_tt_unpopulate = &nouveau_ttm_tt_unpopulate,
-	.init_mem_type = nouveau_bo_init_mem_type,
+	.ttm_tt_bind = &nouveau_ttm_tt_bind,
+	.ttm_tt_unbind = &nouveau_ttm_tt_unbind,
+	.ttm_tt_destroy = &nouveau_ttm_tt_destroy,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = nouveau_bo_evict_flags,
 	.move_notify = nouveau_bo_move_ntfy,
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 52489ce7d029..2a23c8207436 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -18,6 +18,7 @@ struct nouveau_bo {
 	bool force_coherent;
 	struct ttm_bo_kmap_obj kmap;
 	struct list_head head;
+	struct list_head io_reserve_lru;
 
 	/* protected by ttm_bo_reserve() */
 	struct drm_file *reserved_by;
@@ -76,10 +77,10 @@ extern struct ttm_bo_driver nouveau_bo_driver;
 
 void nouveau_bo_move_init(struct nouveau_drm *);
 struct nouveau_bo *nouveau_bo_alloc(struct nouveau_cli *, u64 *size, int *align,
-				    u32 flags, u32 tile_mode, u32 tile_flags);
-int  nouveau_bo_init(struct nouveau_bo *, u64 size, int align, u32 flags,
+				    u32 domain, u32 tile_mode, u32 tile_flags);
+int  nouveau_bo_init(struct nouveau_bo *, u64 size, int align, u32 domain,
 		     struct sg_table *sg, struct dma_resv *robj);
-int  nouveau_bo_new(struct nouveau_cli *, u64 size, int align, u32 flags,
+int  nouveau_bo_new(struct nouveau_cli *, u64 size, int align, u32 domain,
 		    u32 tile_mode, u32 tile_flags, struct sg_table *sg,
 		    struct dma_resv *robj,
 		    struct nouveau_bo **);
@@ -96,6 +97,8 @@ int  nouveau_bo_validate(struct nouveau_bo *, bool interruptible,
 			 bool no_wait_gpu);
 void nouveau_bo_sync_for_device(struct nouveau_bo *nvbo);
 void nouveau_bo_sync_for_cpu(struct nouveau_bo *nvbo);
+void nouveau_bo_add_io_reserve_lru(struct ttm_buffer_object *bo);
+void nouveau_bo_del_io_reserve_lru(struct ttm_buffer_object *bo);
 
 /* TODO: submit equivalent to TTM generic API upstream? */
 static inline void __iomem *
@@ -119,13 +122,13 @@ nouveau_bo_unmap_unpin_unref(struct nouveau_bo **pnvbo)
 }
 
 static inline int
-nouveau_bo_new_pin_map(struct nouveau_cli *cli, u64 size, int align, u32 flags,
+nouveau_bo_new_pin_map(struct nouveau_cli *cli, u64 size, int align, u32 domain,
 		       struct nouveau_bo **pnvbo)
 {
-	int ret = nouveau_bo_new(cli, size, align, flags,
+	int ret = nouveau_bo_new(cli, size, align, domain,
 				 0, 0, NULL, NULL, pnvbo);
 	if (ret == 0) {
-		ret = nouveau_bo_pin(*pnvbo, flags, true);
+		ret = nouveau_bo_pin(*pnvbo, domain, true);
 		if (ret == 0) {
 			ret = nouveau_bo_map(*pnvbo);
 			if (ret == 0)
@@ -139,28 +142,28 @@ nouveau_bo_new_pin_map(struct nouveau_cli *cli, u64 size, int align, u32 flags,
 
 int nv04_bo_move_init(struct nouveau_channel *, u32);
 int nv04_bo_move_m2mf(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nv50_bo_move_init(struct nouveau_channel *, u32);
 int nv50_bo_move_m2mf(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nv84_bo_move_exec(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nva3_bo_move_copy(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nvc0_bo_move_init(struct nouveau_channel *, u32);
 int nvc0_bo_move_m2mf(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nvc0_bo_move_copy(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 int nve0_bo_move_init(struct nouveau_channel *, u32);
 int nve0_bo_move_copy(struct nouveau_channel *, struct ttm_buffer_object *,
-		      struct ttm_mem_reg *, struct ttm_mem_reg *);
+		      struct ttm_resource *, struct ttm_resource *);
 
 #define NVBO_WR32_(b,o,dr,f) nouveau_bo_wr32((b), (o)/4 + (dr), (f))
 #define NVBO_RD32_(b,o,dr)   nouveau_bo_rd32((b), (o)/4 + (dr))
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo0039.c b/drivers/gpu/drm/nouveau/nouveau_bo0039.c
index bf7ae2cecaf6..7390132129fe 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo0039.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo0039.c
@@ -36,7 +36,7 @@
 
 static inline uint32_t
 nouveau_bo_mem_ctxdma(struct ttm_buffer_object *bo,
-		      struct nouveau_channel *chan, struct ttm_mem_reg *reg)
+		      struct nouveau_channel *chan, struct ttm_resource *reg)
 {
 	if (reg->mem_type == TTM_PL_TT)
 		return NvDmaTT;
@@ -45,7 +45,7 @@ nouveau_bo_mem_ctxdma(struct ttm_buffer_object *bo,
 
 int
 nv04_bo_move_m2mf(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nvif_push *push = chan->chan.push;
 	u32 src_ctxdma = nouveau_bo_mem_ctxdma(bo, chan, old_reg);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo5039.c b/drivers/gpu/drm/nouveau/nouveau_bo5039.c
index f9b9b85abe44..4c75c7b3804c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo5039.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo5039.c
@@ -37,7 +37,7 @@
 
 int
 nv50_bo_move_m2mf(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
 	struct nvif_push *push = chan->chan.push;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo74c1.c b/drivers/gpu/drm/nouveau/nouveau_bo74c1.c
index 1b5fd78ddcba..ed6c09d67840 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo74c1.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo74c1.c
@@ -34,7 +34,7 @@
 
 int
 nv84_bo_move_exec(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
 	struct nvif_push *push = chan->chan.push;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo85b5.c b/drivers/gpu/drm/nouveau/nouveau_bo85b5.c
index f0df172b029e..dec29b2d8bb2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo85b5.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo85b5.c
@@ -38,7 +38,7 @@
 
 int
 nva3_bo_move_copy(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
 	struct nvif_push *push = chan->chan.push;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo9039.c b/drivers/gpu/drm/nouveau/nouveau_bo9039.c
index 52fefb37064c..776b04976cdf 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo9039.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo9039.c
@@ -36,7 +36,7 @@
 
 int
 nvc0_bo_move_m2mf(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nvif_push *push = chan->chan.push;
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo90b5.c b/drivers/gpu/drm/nouveau/nouveau_bo90b5.c
index 34b79d561c7f..8499f58213e3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo90b5.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo90b5.c
@@ -31,7 +31,7 @@
 
 int
 nvc0_bo_move_copy(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
 	struct nvif_push *push = chan->chan.push;
diff --git a/drivers/gpu/drm/nouveau/nouveau_boa0b5.c b/drivers/gpu/drm/nouveau/nouveau_boa0b5.c
index 394e29012e50..575212472e7a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_boa0b5.c
+++ b/drivers/gpu/drm/nouveau/nouveau_boa0b5.c
@@ -36,7 +36,7 @@
 
 int
 nve0_bo_move_copy(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
-		  struct ttm_mem_reg *old_reg, struct ttm_mem_reg *new_reg)
+		  struct ttm_resource *old_reg, struct ttm_resource *new_reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(old_reg);
 	struct nvif_push *push = chan->chan.push;
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index b80e4ebf14a6..8f099601d2f2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -163,9 +163,9 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 	atomic_set(&chan->killed, 0);
 
 	/* allocate memory for dma push buffer */
-	target = TTM_PL_FLAG_TT | TTM_PL_FLAG_UNCACHED;
+	target = NOUVEAU_GEM_DOMAIN_GART | NOUVEAU_GEM_DOMAIN_COHERENT;
 	if (nouveau_vram_pushbuf)
-		target = TTM_PL_FLAG_VRAM;
+		target = NOUVEAU_GEM_DOMAIN_VRAM;
 
 	ret = nouveau_bo_new(cli, size, 0, target, 0, 0, NULL, NULL,
 			    &chan->push.buffer);
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 7674025a4bfe..49dd0cbc332f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -391,20 +391,6 @@ find_encoder(struct drm_connector *connector, int type)
 	return NULL;
 }
 
-struct nouveau_connector *
-nouveau_encoder_connector_get(struct nouveau_encoder *encoder)
-{
-	struct drm_device *dev = to_drm_encoder(encoder)->dev;
-	struct drm_connector *drm_connector;
-
-	list_for_each_entry(drm_connector, &dev->mode_config.connector_list, head) {
-		if (drm_connector->encoder == to_drm_encoder(encoder))
-			return nouveau_connector(drm_connector);
-	}
-
-	return NULL;
-}
-
 static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
@@ -435,7 +421,8 @@ nouveau_connector_ddc_detect(struct drm_connector *connector)
 
 		switch (nv_encoder->dcb->type) {
 		case DCB_OUTPUT_DP:
-			ret = nouveau_dp_detect(nv_encoder);
+			ret = nouveau_dp_detect(nouveau_connector(connector),
+						nv_encoder);
 			if (ret == NOUVEAU_DP_MST)
 				return NULL;
 			else if (ret == NOUVEAU_DP_SST)
@@ -541,6 +528,17 @@ nouveau_connector_set_encoder(struct drm_connector *connector,
 	}
 }
 
+static void
+nouveau_connector_set_edid(struct nouveau_connector *nv_connector,
+			   struct edid *edid)
+{
+	struct edid *old_edid = nv_connector->edid;
+
+	drm_connector_update_edid_property(&nv_connector->base, edid);
+	kfree(old_edid);
+	nv_connector->edid = edid;
+}
+
 static enum drm_connector_status
 nouveau_connector_detect(struct drm_connector *connector, bool force)
 {
@@ -554,13 +552,6 @@ nouveau_connector_detect(struct drm_connector *connector, bool force)
 	int ret;
 	enum drm_connector_status conn_status = connector_status_disconnected;
 
-	/* Cleanup the previous EDID block. */
-	if (nv_connector->edid) {
-		drm_connector_update_edid_property(connector, NULL);
-		kfree(nv_connector->edid);
-		nv_connector->edid = NULL;
-	}
-
 	/* Outputs are only polled while runtime active, so resuming the
 	 * device here is unnecessary (and would deadlock upon runtime suspend
 	 * because it waits for polling to finish). We do however, want to
@@ -573,22 +564,23 @@ nouveau_connector_detect(struct drm_connector *connector, bool force)
 		ret = pm_runtime_get_sync(dev->dev);
 		if (ret < 0 && ret != -EACCES) {
 			pm_runtime_put_autosuspend(dev->dev);
+			nouveau_connector_set_edid(nv_connector, NULL);
 			return conn_status;
 		}
 	}
 
 	nv_encoder = nouveau_connector_ddc_detect(connector);
 	if (nv_encoder && (i2c = nv_encoder->i2c) != NULL) {
+		struct edid *new_edid;
+
 		if ((vga_switcheroo_handler_flags() &
 		     VGA_SWITCHEROO_CAN_SWITCH_DDC) &&
 		    nv_connector->type == DCB_CONNECTOR_LVDS)
-			nv_connector->edid = drm_get_edid_switcheroo(connector,
-								     i2c);
+			new_edid = drm_get_edid_switcheroo(connector, i2c);
 		else
-			nv_connector->edid = drm_get_edid(connector, i2c);
+			new_edid = drm_get_edid(connector, i2c);
 
-		drm_connector_update_edid_property(connector,
-							nv_connector->edid);
+		nouveau_connector_set_edid(nv_connector, new_edid);
 		if (!nv_connector->edid) {
 			NV_ERROR(drm, "DDC responded, but no EDID for %s\n",
 				 connector->name);
@@ -622,6 +614,8 @@ nouveau_connector_detect(struct drm_connector *connector, bool force)
 		conn_status = connector_status_connected;
 		drm_dp_cec_set_edid(&nv_connector->aux, nv_connector->edid);
 		goto out;
+	} else {
+		nouveau_connector_set_edid(nv_connector, NULL);
 	}
 
 	nv_encoder = nouveau_connector_of_detect(connector);
@@ -646,10 +640,11 @@ detect_analog:
 			conn_status = connector_status_connected;
 			goto out;
 		}
-
 	}
 
  out:
+	if (!nv_connector->edid)
+		drm_dp_cec_unset_edid(&nv_connector->aux);
 
 	pm_runtime_mark_last_busy(dev->dev);
 	pm_runtime_put_autosuspend(dev->dev);
@@ -664,18 +659,12 @@ nouveau_connector_detect_lvds(struct drm_connector *connector, bool force)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct nouveau_encoder *nv_encoder = NULL;
+	struct edid *edid = NULL;
 	enum drm_connector_status status = connector_status_disconnected;
 
-	/* Cleanup the previous EDID block. */
-	if (nv_connector->edid) {
-		drm_connector_update_edid_property(connector, NULL);
-		kfree(nv_connector->edid);
-		nv_connector->edid = NULL;
-	}
-
 	nv_encoder = find_encoder(connector, DCB_OUTPUT_LVDS);
 	if (!nv_encoder)
-		return connector_status_disconnected;
+		goto out;
 
 	/* Try retrieving EDID via DDC */
 	if (!drm->vbios.fp_no_ddc) {
@@ -694,7 +683,8 @@ nouveau_connector_detect_lvds(struct drm_connector *connector, bool force)
 	 * valid - it's not (rh#613284)
 	 */
 	if (nv_encoder->dcb->lvdsconf.use_acpi_for_edid) {
-		if ((nv_connector->edid = nouveau_acpi_edid(dev, connector))) {
+		edid = nouveau_acpi_edid(dev, connector);
+		if (edid) {
 			status = connector_status_connected;
 			goto out;
 		}
@@ -714,12 +704,10 @@ nouveau_connector_detect_lvds(struct drm_connector *connector, bool force)
 	 * stored for the panel stored in them.
 	 */
 	if (!drm->vbios.fp_no_ddc) {
-		struct edid *edid =
-			(struct edid *)nouveau_bios_embedded_edid(dev);
+		edid = (struct edid *)nouveau_bios_embedded_edid(dev);
 		if (edid) {
-			nv_connector->edid =
-					kmemdup(edid, EDID_LENGTH, GFP_KERNEL);
-			if (nv_connector->edid)
+			edid = kmemdup(edid, EDID_LENGTH, GFP_KERNEL);
+			if (edid)
 				status = connector_status_connected;
 		}
 	}
@@ -732,7 +720,7 @@ out:
 		status = connector_status_unknown;
 #endif
 
-	drm_connector_update_edid_property(connector, nv_connector->edid);
+	nouveau_connector_set_edid(nv_connector, edid);
 	nouveau_connector_set_encoder(connector, nv_encoder);
 	return status;
 }
@@ -1150,59 +1138,39 @@ nouveau_connector_funcs_lvds = {
 	.early_unregister = nouveau_connector_early_unregister,
 };
 
+void
+nouveau_connector_hpd(struct drm_connector *connector)
+{
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	u32 mask = drm_connector_mask(connector);
+
+	mutex_lock(&drm->hpd_lock);
+	if (!(drm->hpd_pending & mask)) {
+		drm->hpd_pending |= mask;
+		schedule_work(&drm->hpd_work);
+	}
+	mutex_unlock(&drm->hpd_lock);
+}
+
 static int
 nouveau_connector_hotplug(struct nvif_notify *notify)
 {
 	struct nouveau_connector *nv_connector =
 		container_of(notify, typeof(*nv_connector), hpd);
 	struct drm_connector *connector = &nv_connector->base;
-	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	struct drm_device *dev = connector->dev;
+	struct nouveau_drm *drm = nouveau_drm(dev);
 	const struct nvif_notify_conn_rep_v0 *rep = notify->data;
-	const char *name = connector->name;
-	struct nouveau_encoder *nv_encoder;
-	int ret;
 	bool plugged = (rep->mask != NVIF_NOTIFY_CONN_V0_UNPLUG);
 
 	if (rep->mask & NVIF_NOTIFY_CONN_V0_IRQ) {
-		NV_DEBUG(drm, "service %s\n", name);
-		drm_dp_cec_irq(&nv_connector->aux);
-		if ((nv_encoder = find_encoder(connector, DCB_OUTPUT_DP)))
-			nv50_mstm_service(nv_encoder->dp.mstm);
-
+		nouveau_dp_irq(drm, nv_connector);
 		return NVIF_NOTIFY_KEEP;
 	}
 
-	ret = pm_runtime_get(drm->dev->dev);
-	if (ret == 0) {
-		/* We can't block here if there's a pending PM request
-		 * running, as we'll deadlock nouveau_display_fini() when it
-		 * calls nvif_put() on our nvif_notify struct. So, simply
-		 * defer the hotplug event until the device finishes resuming
-		 */
-		NV_DEBUG(drm, "Deferring HPD on %s until runtime resume\n",
-			 name);
-		schedule_work(&drm->hpd_work);
-
-		pm_runtime_put_noidle(drm->dev->dev);
-		return NVIF_NOTIFY_KEEP;
-	} else if (ret != 1 && ret != -EACCES) {
-		NV_WARN(drm, "HPD on %s dropped due to RPM failure: %d\n",
-			name, ret);
-		return NVIF_NOTIFY_DROP;
-	}
-
-	if (!plugged)
-		drm_dp_cec_unset_edid(&nv_connector->aux);
-	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un", name);
-	if ((nv_encoder = find_encoder(connector, DCB_OUTPUT_DP))) {
-		if (!plugged)
-			nv50_mstm_remove(nv_encoder->dp.mstm);
-	}
-
-	drm_helper_hpd_irq_event(connector->dev);
+	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un", connector->name);
+	nouveau_connector_hpd(connector);
 
-	pm_runtime_mark_last_busy(drm->dev->dev);
-	pm_runtime_put_autosuspend(drm->dev->dev);
 	return NVIF_NOTIFY_KEEP;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index d6de5cb8e223..d0b859c4a80e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -187,6 +187,7 @@ nouveau_crtc_connector_get(struct nouveau_crtc *nv_crtc)
 
 struct drm_connector *
 nouveau_connector_create(struct drm_device *, const struct dcb_output *);
+void nouveau_connector_hpd(struct drm_connector *connector);
 
 extern int nouveau_tv_disable;
 extern int nouveau_ignorelid;
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 5f31b11ac2e7..bceb48a2dfca 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -457,16 +457,70 @@ static struct nouveau_drm_prop_enum_list dither_depth[] = {
 	}                                                                      \
 } while(0)
 
+void
+nouveau_display_hpd_resume(struct drm_device *dev)
+{
+	struct nouveau_drm *drm = nouveau_drm(dev);
+
+	mutex_lock(&drm->hpd_lock);
+	drm->hpd_pending = ~0;
+	mutex_unlock(&drm->hpd_lock);
+
+	schedule_work(&drm->hpd_work);
+}
+
 static void
 nouveau_display_hpd_work(struct work_struct *work)
 {
 	struct nouveau_drm *drm = container_of(work, typeof(*drm), hpd_work);
+	struct drm_device *dev = drm->dev;
+	struct drm_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+	u32 pending;
+	bool changed = false;
+
+	pm_runtime_get_sync(dev->dev);
 
-	pm_runtime_get_sync(drm->dev->dev);
+	mutex_lock(&drm->hpd_lock);
+	pending = drm->hpd_pending;
+	drm->hpd_pending = 0;
+	mutex_unlock(&drm->hpd_lock);
 
-	drm_helper_hpd_irq_event(drm->dev);
+	/* Nothing to do, exit early without updating the last busy counter */
+	if (!pending)
+		goto noop;
+
+	mutex_lock(&dev->mode_config.mutex);
+	drm_connector_list_iter_begin(dev, &conn_iter);
+
+	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
+		enum drm_connector_status old_status = connector->status;
+		u64 old_epoch_counter = connector->epoch_counter;
+
+		if (!(pending & drm_connector_mask(connector)))
+			continue;
+
+		connector->status = drm_helper_probe_detect(connector, NULL,
+							    false);
+		if (old_epoch_counter == connector->epoch_counter)
+			continue;
+
+		changed = true;
+		drm_dbg_kms(dev, "[CONNECTOR:%d:%s] status updated from %s to %s (epoch counter %llu->%llu)\n",
+			    connector->base.id, connector->name,
+			    drm_get_connector_status_name(old_status),
+			    drm_get_connector_status_name(connector->status),
+			    old_epoch_counter, connector->epoch_counter);
+	}
+
+	drm_connector_list_iter_end(&conn_iter);
+	mutex_unlock(&dev->mode_config.mutex);
+
+	if (changed)
+		drm_kms_helper_hotplug_event(dev);
 
 	pm_runtime_mark_last_busy(drm->dev->dev);
+noop:
 	pm_runtime_put_sync(drm->dev->dev);
 }
 
@@ -490,12 +544,11 @@ nouveau_display_acpi_ntfy(struct notifier_block *nb, unsigned long val,
 				 */
 				pm_runtime_put_autosuspend(drm->dev->dev);
 			} else if (ret == 0) {
-				/* This may be the only indication we receive
-				 * of a connector hotplug on a runtime
-				 * suspended GPU, schedule hpd_work to check.
+				/* We've started resuming the GPU already, so
+				 * it will handle scheduling a full reprobe
+				 * itself
 				 */
 				NV_DEBUG(drm, "ACPI requested connector reprobe\n");
-				schedule_work(&drm->hpd_work);
 				pm_runtime_put_noidle(drm->dev->dev);
 			} else {
 				NV_WARN(drm, "Dropped ACPI reprobe event due to RPM error: %d\n",
@@ -569,7 +622,7 @@ nouveau_display_fini(struct drm_device *dev, bool suspend, bool runtime)
 		cancel_work_sync(&drm->hpd_work);
 
 	drm_kms_helper_poll_disable(dev);
-	disp->fini(dev, suspend);
+	disp->fini(dev, runtime, suspend);
 }
 
 static void
@@ -686,6 +739,7 @@ nouveau_display_create(struct drm_device *dev)
 	}
 
 	INIT_WORK(&drm->hpd_work, nouveau_display_hpd_work);
+	mutex_init(&drm->hpd_lock);
 #ifdef CONFIG_ACPI
 	drm->acpi_nb.notifier_call = nouveau_display_acpi_ntfy;
 	register_acpi_notifier(&drm->acpi_nb);
@@ -705,9 +759,10 @@ void
 nouveau_display_destroy(struct drm_device *dev)
 {
 	struct nouveau_display *disp = nouveau_display(dev);
+	struct nouveau_drm *drm = nouveau_drm(dev);
 
 #ifdef CONFIG_ACPI
-	unregister_acpi_notifier(&nouveau_drm(dev)->acpi_nb);
+	unregister_acpi_notifier(&drm->acpi_nb);
 #endif
 
 	drm_kms_helper_poll_fini(dev);
@@ -719,6 +774,7 @@ nouveau_display_destroy(struct drm_device *dev)
 	nvif_disp_dtor(&disp->disp);
 
 	nouveau_drm(dev)->display = NULL;
+	mutex_destroy(&drm->hpd_lock);
 	kfree(disp);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index 6e0d900441d6..616c43427059 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -18,7 +18,7 @@ struct nouveau_display {
 	void *priv;
 	void (*dtor)(struct drm_device *);
 	int  (*init)(struct drm_device *, bool resume, bool runtime);
-	void (*fini)(struct drm_device *, bool suspend);
+	void (*fini)(struct drm_device *, bool suspend, bool runtime);
 
 	struct nvif_disp disp;
 
@@ -45,6 +45,7 @@ nouveau_display(struct drm_device *dev)
 int  nouveau_display_create(struct drm_device *dev);
 void nouveau_display_destroy(struct drm_device *dev);
 int  nouveau_display_init(struct drm_device *dev, bool resume, bool runtime);
+void nouveau_display_hpd_resume(struct drm_device *dev);
 void nouveau_display_fini(struct drm_device *dev, bool suspend, bool runtime);
 int  nouveau_display_suspend(struct drm_device *dev, bool runtime);
 void nouveau_display_resume(struct drm_device *dev, bool runtime);
diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
index a13c6215bba8..92987daa5e17 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
@@ -256,12 +256,12 @@ nouveau_dmem_chunk_alloc(struct nouveau_drm *drm, struct page **ppage)
 	chunk->pagemap.owner = drm->dev;
 
 	ret = nouveau_bo_new(&drm->client, DMEM_CHUNK_SIZE, 0,
-			     TTM_PL_FLAG_VRAM, 0, 0, NULL, NULL,
+			     NOUVEAU_GEM_DOMAIN_VRAM, 0, 0, NULL, NULL,
 			     &chunk->bo);
 	if (ret)
 		goto out_release;
 
-	ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+	ret = nouveau_bo_pin(chunk->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 	if (ret)
 		goto out_bo_free;
 
@@ -347,7 +347,7 @@ nouveau_dmem_resume(struct nouveau_drm *drm)
 
 	mutex_lock(&drm->dmem->mutex);
 	list_for_each_entry(chunk, &drm->dmem->chunks, list) {
-		ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+		ret = nouveau_bo_pin(chunk->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 		/* FIXME handle pin failure */
 		WARN_ON(ret);
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index 8a0f7994e1ae..7b640e05bd4c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -36,50 +36,123 @@ MODULE_PARM_DESC(mst, "Enable DisplayPort multi-stream (default: enabled)");
 static int nouveau_mst = 1;
 module_param_named(mst, nouveau_mst, int, 0400);
 
-static void
-nouveau_dp_probe_oui(struct drm_device *dev, struct nvkm_i2c_aux *aux, u8 *dpcd)
+static bool
+nouveau_dp_has_sink_count(struct drm_connector *connector,
+			  struct nouveau_encoder *outp)
 {
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u8 buf[3];
-
-	if (!(dpcd[DP_DOWN_STREAM_PORT_COUNT] & DP_OUI_SUPPORT))
-		return;
-
-	if (!nvkm_rdaux(aux, DP_SINK_OUI, buf, 3))
-		NV_DEBUG(drm, "Sink OUI: %02hx%02hx%02hx\n",
-			     buf[0], buf[1], buf[2]);
-
-	if (!nvkm_rdaux(aux, DP_BRANCH_OUI, buf, 3))
-		NV_DEBUG(drm, "Branch OUI: %02hx%02hx%02hx\n",
-			     buf[0], buf[1], buf[2]);
+	return drm_dp_read_sink_count_cap(connector, outp->dp.dpcd, &outp->dp.desc);
+}
 
+static enum drm_connector_status
+nouveau_dp_probe_dpcd(struct nouveau_connector *nv_connector,
+		      struct nouveau_encoder *outp)
+{
+	struct drm_connector *connector = &nv_connector->base;
+	struct drm_dp_aux *aux = &nv_connector->aux;
+	struct nv50_mstm *mstm = NULL;
+	enum drm_connector_status status = connector_status_disconnected;
+	int ret;
+	u8 *dpcd = outp->dp.dpcd;
+
+	ret = drm_dp_read_dpcd_caps(aux, dpcd);
+	if (ret < 0)
+		goto out;
+
+	ret = drm_dp_read_desc(aux, &outp->dp.desc, drm_dp_is_branch(dpcd));
+	if (ret < 0)
+		goto out;
+
+	if (nouveau_mst) {
+		mstm = outp->dp.mstm;
+		if (mstm)
+			mstm->can_mst = drm_dp_read_mst_cap(aux, dpcd);
+	}
+
+	if (nouveau_dp_has_sink_count(connector, outp)) {
+		ret = drm_dp_read_sink_count(aux);
+		if (ret < 0)
+			goto out;
+
+		outp->dp.sink_count = ret;
+
+		/*
+		 * Dongle connected, but no display. Don't bother reading
+		 * downstream port info
+		 */
+		if (!outp->dp.sink_count)
+			return connector_status_disconnected;
+	}
+
+	ret = drm_dp_read_downstream_info(aux, dpcd,
+					  outp->dp.downstream_ports);
+	if (ret < 0)
+		goto out;
+
+	status = connector_status_connected;
+out:
+	if (status != connector_status_connected) {
+		/* Clear any cached info */
+		outp->dp.sink_count = 0;
+	}
+	return status;
 }
 
 int
-nouveau_dp_detect(struct nouveau_encoder *nv_encoder)
+nouveau_dp_detect(struct nouveau_connector *nv_connector,
+		  struct nouveau_encoder *nv_encoder)
 {
 	struct drm_device *dev = nv_encoder->base.base.dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvkm_i2c_aux *aux;
-	u8 dpcd[8];
-	int ret;
-
-	aux = nv_encoder->aux;
-	if (!aux)
-		return -ENODEV;
-
-	ret = nvkm_rdaux(aux, DP_DPCD_REV, dpcd, sizeof(dpcd));
-	if (ret)
-		return ret;
+	struct drm_connector *connector = &nv_connector->base;
+	struct nv50_mstm *mstm = nv_encoder->dp.mstm;
+	enum drm_connector_status status;
+	u8 *dpcd = nv_encoder->dp.dpcd;
+	int ret = NOUVEAU_DP_NONE;
+
+	/* If we've already read the DPCD on an eDP device, we don't need to
+	 * reread it as it won't change
+	 */
+	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP &&
+	    dpcd[DP_DPCD_REV] != 0)
+		return NOUVEAU_DP_SST;
 
-	nv_encoder->dp.link_bw = 27000 * dpcd[1];
-	nv_encoder->dp.link_nr = dpcd[2] & DP_MAX_LANE_COUNT_MASK;
+	mutex_lock(&nv_encoder->dp.hpd_irq_lock);
+	if (mstm) {
+		/* If we're not ready to handle MST state changes yet, just
+		 * report the last status of the connector. We'll reprobe it
+		 * once we've resumed.
+		 */
+		if (mstm->suspended) {
+			if (mstm->is_mst)
+				ret = NOUVEAU_DP_MST;
+			else if (connector->status ==
+				 connector_status_connected)
+				ret = NOUVEAU_DP_SST;
+
+			goto out;
+		}
+	}
+
+	status = nouveau_dp_probe_dpcd(nv_connector, nv_encoder);
+	if (status == connector_status_disconnected)
+		goto out;
+
+	/* If we're in MST mode, we're done here */
+	if (mstm && mstm->can_mst && mstm->is_mst) {
+		ret = NOUVEAU_DP_MST;
+		goto out;
+	}
+
+	nv_encoder->dp.link_bw = 27000 * dpcd[DP_MAX_LINK_RATE];
+	nv_encoder->dp.link_nr =
+		dpcd[DP_MAX_LANE_COUNT] & DP_MAX_LANE_COUNT_MASK;
 
 	NV_DEBUG(drm, "display: %dx%d dpcd 0x%02x\n",
-		     nv_encoder->dp.link_nr, nv_encoder->dp.link_bw, dpcd[0]);
+		 nv_encoder->dp.link_nr, nv_encoder->dp.link_bw,
+		 dpcd[DP_DPCD_REV]);
 	NV_DEBUG(drm, "encoder: %dx%d\n",
-		     nv_encoder->dcb->dpconf.link_nr,
-		     nv_encoder->dcb->dpconf.link_bw);
+		 nv_encoder->dcb->dpconf.link_nr,
+		 nv_encoder->dcb->dpconf.link_bw);
 
 	if (nv_encoder->dcb->dpconf.link_nr < nv_encoder->dp.link_nr)
 		nv_encoder->dp.link_nr = nv_encoder->dcb->dpconf.link_nr;
@@ -87,23 +160,68 @@ nouveau_dp_detect(struct nouveau_encoder *nv_encoder)
 		nv_encoder->dp.link_bw = nv_encoder->dcb->dpconf.link_bw;
 
 	NV_DEBUG(drm, "maximum: %dx%d\n",
-		     nv_encoder->dp.link_nr, nv_encoder->dp.link_bw);
+		 nv_encoder->dp.link_nr, nv_encoder->dp.link_bw);
+
+	if (mstm && mstm->can_mst) {
+		ret = nv50_mstm_detect(nv_encoder);
+		if (ret == 1) {
+			ret = NOUVEAU_DP_MST;
+			goto out;
+		} else if (ret != 0) {
+			goto out;
+		}
+	}
+	ret = NOUVEAU_DP_SST;
+
+out:
+	if (mstm && !mstm->suspended && ret != NOUVEAU_DP_MST)
+		nv50_mstm_remove(mstm);
+
+	mutex_unlock(&nv_encoder->dp.hpd_irq_lock);
+	return ret;
+}
 
-	nouveau_dp_probe_oui(dev, aux, dpcd);
+void nouveau_dp_irq(struct nouveau_drm *drm,
+		    struct nouveau_connector *nv_connector)
+{
+	struct drm_connector *connector = &nv_connector->base;
+	struct nouveau_encoder *outp = find_encoder(connector, DCB_OUTPUT_DP);
+	struct nv50_mstm *mstm;
+	int ret;
+	bool send_hpd = false;
 
-	ret = nv50_mstm_detect(nv_encoder->dp.mstm, dpcd, nouveau_mst);
-	if (ret == 1)
-		return NOUVEAU_DP_MST;
-	if (ret == 0)
-		return NOUVEAU_DP_SST;
-	return ret;
+	if (!outp)
+		return;
+
+	mstm = outp->dp.mstm;
+	NV_DEBUG(drm, "service %s\n", connector->name);
+
+	mutex_lock(&outp->dp.hpd_irq_lock);
+
+	if (mstm && mstm->is_mst) {
+		if (!nv50_mstm_service(drm, nv_connector, mstm))
+			send_hpd = true;
+	} else {
+		drm_dp_cec_irq(&nv_connector->aux);
+
+		if (nouveau_dp_has_sink_count(connector, outp)) {
+			ret = drm_dp_read_sink_count(&nv_connector->aux);
+			if (ret != outp->dp.sink_count)
+				send_hpd = true;
+			if (ret >= 0)
+				outp->dp.sink_count = ret;
+		}
+	}
+
+	mutex_unlock(&outp->dp.hpd_irq_lock);
+
+	if (send_hpd)
+		nouveau_connector_hpd(connector);
 }
 
 /* TODO:
  * - Use the minimum possible BPC here, once we add support for the max bpc
  *   property.
- * - Validate the mode against downstream port caps (see
- *   drm_dp_downstream_max_clock())
  * - Validate against the DP caps advertised by the GPU (we don't check these
  *   yet)
  */
@@ -114,15 +232,19 @@ nv50_dp_mode_valid(struct drm_connector *connector,
 		   unsigned *out_clock)
 {
 	const unsigned min_clock = 25000;
-	unsigned max_clock, clock;
+	unsigned max_clock, ds_clock, clock;
 	enum drm_mode_status ret;
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE && !outp->caps.dp_interlace)
 		return MODE_NO_INTERLACE;
 
 	max_clock = outp->dp.link_nr * outp->dp.link_bw;
-	clock = mode->clock * (connector->display_info.bpc * 3) / 10;
+	ds_clock = drm_dp_downstream_max_dotclock(outp->dp.dpcd,
+						  outp->dp.downstream_ports);
+	if (ds_clock)
+		max_clock = min(max_clock, ds_clock);
 
+	clock = mode->clock * (connector->display_info.bpc * 3) / 10;
 	ret = nouveau_conn_mode_clock_valid(mode, min_clock, max_clock,
 					    &clock);
 	if (out_clock)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 22d246acc5e5..42fc5c813a9b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -953,7 +953,7 @@ nouveau_pmops_resume(struct device *dev)
 	ret = nouveau_do_resume(drm_dev, false);
 
 	/* Monitors may have been connected / disconnected during suspend */
-	schedule_work(&nouveau_drm(drm_dev)->hpd_work);
+	nouveau_display_hpd_resume(drm_dev);
 
 	return ret;
 }
@@ -1036,7 +1036,7 @@ nouveau_pmops_runtime_resume(struct device *dev)
 	drm_dev->switch_power_state = DRM_SWITCH_POWER_ON;
 
 	/* Monitors may have been connected / disconnected during suspend */
-	schedule_work(&nouveau_drm(drm_dev)->hpd_work);
+	nouveau_display_hpd_resume(drm_dev);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index ae76a5865a5a..b8025507a9e4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -157,13 +157,15 @@ struct nouveau_drm {
 		atomic_t validate_sequence;
 		int (*move)(struct nouveau_channel *,
 			    struct ttm_buffer_object *,
-			    struct ttm_mem_reg *, struct ttm_mem_reg *);
+			    struct ttm_resource *, struct ttm_resource *);
 		struct nouveau_channel *chan;
 		struct nvif_object copy;
 		int mtrr;
 		int type_vram;
 		int type_host[2];
 		int type_ncoh[2];
+		struct mutex io_reserve_mutex;
+		struct list_head io_reserve_lru;
 	} ttm;
 
 	/* GEM interface support */
@@ -198,6 +200,8 @@ struct nouveau_drm {
 	struct nvbios vbios;
 	struct nouveau_display *display;
 	struct work_struct hpd_work;
+	struct mutex hpd_lock;
+	u32 hpd_pending;
 	struct work_struct fbcon_work;
 	int fbcon_new_state;
 #ifdef CONFIG_ACPI
diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h
index a72c412ac8b1..21937f1c7dd9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_encoder.h
+++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h
@@ -33,6 +33,7 @@
 #include <drm/drm_dp_mst_helper.h>
 #include "dispnv04/disp.h"
 struct nv50_head_atom;
+struct nouveau_connector;
 
 #define NV_DPMS_CLEARED 0x80
 
@@ -64,6 +65,17 @@ struct nouveau_encoder {
 			struct nv50_mstm *mstm;
 			int link_nr;
 			int link_bw;
+
+			/* Protects DP state that needs to be accessed outside
+			 * connector reprobing contexts
+			 */
+			struct mutex hpd_irq_lock;
+
+			u8 dpcd[DP_RECEIVER_CAP_SIZE];
+			u8 downstream_ports[DP_MAX_DOWNSTREAM_PORTS];
+			struct drm_dp_desc desc;
+
+			u8 sink_count;
 		} dp;
 	};
 
@@ -77,6 +89,21 @@ struct nouveau_encoder {
 		       struct nv50_head_atom *, u8 proto, u8 depth);
 };
 
+struct nv50_mstm {
+	struct nouveau_encoder *outp;
+
+	struct drm_dp_mst_topology_mgr mgr;
+
+	/* Protected under nouveau_encoder->dp.hpd_irq_lock */
+	bool can_mst;
+	bool is_mst;
+	bool suspended;
+
+	bool modified;
+	bool disabled;
+	int links;
+};
+
 struct nouveau_encoder *
 find_encoder(struct drm_connector *connector, int type);
 
@@ -100,20 +127,29 @@ get_slave_funcs(struct drm_encoder *enc)
 
 /* nouveau_dp.c */
 enum nouveau_dp_status {
+	NOUVEAU_DP_NONE,
 	NOUVEAU_DP_SST,
 	NOUVEAU_DP_MST,
 };
 
-int nouveau_dp_detect(struct nouveau_encoder *);
+int nouveau_dp_detect(struct nouveau_connector *, struct nouveau_encoder *);
+void nouveau_dp_irq(struct nouveau_drm *drm,
+		    struct nouveau_connector *nv_connector);
 enum drm_mode_status nv50_dp_mode_valid(struct drm_connector *,
 					struct nouveau_encoder *,
 					const struct drm_display_mode *,
 					unsigned *clock);
 
 struct nouveau_connector *
-nouveau_encoder_connector_get(struct nouveau_encoder *encoder);
-
-int nv50_mstm_detect(struct nv50_mstm *, u8 dpcd[8], int allow);
-void nv50_mstm_remove(struct nv50_mstm *);
-void nv50_mstm_service(struct nv50_mstm *);
+nv50_outp_get_new_connector(struct nouveau_encoder *outp,
+			    struct drm_atomic_state *state);
+struct nouveau_connector *
+nv50_outp_get_old_connector(struct nouveau_encoder *outp,
+			    struct drm_atomic_state *state);
+
+int nv50_mstm_detect(struct nouveau_encoder *encoder);
+void nv50_mstm_remove(struct nv50_mstm *mstm);
+bool nv50_mstm_service(struct nouveau_drm *drm,
+		       struct nouveau_connector *nv_connector,
+		       struct nv50_mstm *mstm);
 #endif /* __NOUVEAU_ENCODER_H__ */
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index fad8030ec1f8..24ec5339efb4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -341,7 +341,7 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 	if (ret)
 		goto out_unref;
 
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_VRAM, false);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 	if (ret) {
 		NV_ERROR(drm, "failed to pin fb: %d\n", ret);
 		goto out_unref;
@@ -378,8 +378,7 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 			      FBINFO_HWACCEL_FILLRECT |
 			      FBINFO_HWACCEL_IMAGEBLIT;
 	info->fbops = &nouveau_fbcon_sw_ops;
-	info->fix.smem_start = nvbo->bo.mem.bus.base +
-			       nvbo->bo.mem.bus.offset;
+	info->fix.smem_start = nvbo->bo.mem.bus.offset;
 	info->fix.smem_len = nvbo->bo.mem.num_pages << PAGE_SHIFT;
 
 	info->screen_base = nvbo_kmap_obj_iovirtual(nvbo);
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 81f111ad3f4f..89adadf4706b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -176,20 +176,12 @@ nouveau_gem_new(struct nouveau_cli *cli, u64 size, int align, uint32_t domain,
 {
 	struct nouveau_drm *drm = cli->drm;
 	struct nouveau_bo *nvbo;
-	u32 flags = 0;
 	int ret;
 
-	if (domain & NOUVEAU_GEM_DOMAIN_VRAM)
-		flags |= TTM_PL_FLAG_VRAM;
-	if (domain & NOUVEAU_GEM_DOMAIN_GART)
-		flags |= TTM_PL_FLAG_TT;
-	if (!flags || domain & NOUVEAU_GEM_DOMAIN_CPU)
-		flags |= TTM_PL_FLAG_SYSTEM;
+	if (!(domain & (NOUVEAU_GEM_DOMAIN_VRAM | NOUVEAU_GEM_DOMAIN_GART)))
+		domain |= NOUVEAU_GEM_DOMAIN_CPU;
 
-	if (domain & NOUVEAU_GEM_DOMAIN_COHERENT)
-		flags |= TTM_PL_FLAG_UNCACHED;
-
-	nvbo = nouveau_bo_alloc(cli, &size, &align, flags, tile_mode,
+	nvbo = nouveau_bo_alloc(cli, &size, &align, domain, tile_mode,
 				tile_flags);
 	if (IS_ERR(nvbo))
 		return PTR_ERR(nvbo);
@@ -202,7 +194,7 @@ nouveau_gem_new(struct nouveau_cli *cli, u64 size, int align, uint32_t domain,
 		return ret;
 	}
 
-	ret = nouveau_bo_init(nvbo, size, align, flags, NULL, NULL);
+	ret = nouveau_bo_init(nvbo, size, align, domain, NULL, NULL);
 	if (ret) {
 		nouveau_bo_ref(NULL, &nvbo);
 		return ret;
@@ -296,32 +288,28 @@ nouveau_gem_set_domain(struct drm_gem_object *gem, uint32_t read_domains,
 	struct ttm_buffer_object *bo = &nvbo->bo;
 	uint32_t domains = valid_domains & nvbo->valid_domains &
 		(write_domains ? write_domains : read_domains);
-	uint32_t pref_flags = 0, valid_flags = 0;
+	uint32_t pref_domains = 0;;
 
 	if (!domains)
 		return -EINVAL;
 
-	if (valid_domains & NOUVEAU_GEM_DOMAIN_VRAM)
-		valid_flags |= TTM_PL_FLAG_VRAM;
-
-	if (valid_domains & NOUVEAU_GEM_DOMAIN_GART)
-		valid_flags |= TTM_PL_FLAG_TT;
+	valid_domains &= ~(NOUVEAU_GEM_DOMAIN_VRAM | NOUVEAU_GEM_DOMAIN_GART);
 
 	if ((domains & NOUVEAU_GEM_DOMAIN_VRAM) &&
 	    bo->mem.mem_type == TTM_PL_VRAM)
-		pref_flags |= TTM_PL_FLAG_VRAM;
+		pref_domains |= NOUVEAU_GEM_DOMAIN_VRAM;
 
 	else if ((domains & NOUVEAU_GEM_DOMAIN_GART) &&
 		 bo->mem.mem_type == TTM_PL_TT)
-		pref_flags |= TTM_PL_FLAG_TT;
+		pref_domains |= NOUVEAU_GEM_DOMAIN_GART;
 
 	else if (domains & NOUVEAU_GEM_DOMAIN_VRAM)
-		pref_flags |= TTM_PL_FLAG_VRAM;
+		pref_domains |= NOUVEAU_GEM_DOMAIN_VRAM;
 
 	else
-		pref_flags |= TTM_PL_FLAG_TT;
+		pref_domains |= NOUVEAU_GEM_DOMAIN_GART;
 
-	nouveau_bo_placement_set(nvbo, pref_flags, valid_flags);
+	nouveau_bo_placement_set(nvbo, pref_domains, valid_domains);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
index e5fae57fffbd..9dfcce1b9846 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.c
@@ -92,7 +92,7 @@ nouveau_mem_fini(struct nouveau_mem *mem)
 }
 
 int
-nouveau_mem_host(struct ttm_mem_reg *reg, struct ttm_dma_tt *tt)
+nouveau_mem_host(struct ttm_resource *reg, struct ttm_dma_tt *tt)
 {
 	struct nouveau_mem *mem = nouveau_mem(reg);
 	struct nouveau_cli *cli = mem->cli;
@@ -130,7 +130,7 @@ nouveau_mem_host(struct ttm_mem_reg *reg, struct ttm_dma_tt *tt)
 }
 
 int
-nouveau_mem_vram(struct ttm_mem_reg *reg, bool contig, u8 page)
+nouveau_mem_vram(struct ttm_resource *reg, bool contig, u8 page)
 {
 	struct nouveau_mem *mem = nouveau_mem(reg);
 	struct nouveau_cli *cli = mem->cli;
@@ -173,7 +173,7 @@ nouveau_mem_vram(struct ttm_mem_reg *reg, bool contig, u8 page)
 }
 
 void
-nouveau_mem_del(struct ttm_mem_reg *reg)
+nouveau_mem_del(struct ttm_resource *reg)
 {
 	struct nouveau_mem *mem = nouveau_mem(reg);
 	if (!mem)
@@ -185,7 +185,7 @@ nouveau_mem_del(struct ttm_mem_reg *reg)
 
 int
 nouveau_mem_new(struct nouveau_cli *cli, u8 kind, u8 comp,
-		struct ttm_mem_reg *reg)
+		struct ttm_resource *reg)
 {
 	struct nouveau_mem *mem;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.h b/drivers/gpu/drm/nouveau/nouveau_mem.h
index f6d039e73812..3fe1cfed57a1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.h
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.h
@@ -7,7 +7,7 @@ struct ttm_dma_tt;
 #include <nvif/vmm.h>
 
 static inline struct nouveau_mem *
-nouveau_mem(struct ttm_mem_reg *reg)
+nouveau_mem(struct ttm_resource *reg)
 {
 	return reg->mm_node;
 }
@@ -21,10 +21,10 @@ struct nouveau_mem {
 };
 
 int nouveau_mem_new(struct nouveau_cli *, u8 kind, u8 comp,
-		    struct ttm_mem_reg *);
-void nouveau_mem_del(struct ttm_mem_reg *);
-int nouveau_mem_vram(struct ttm_mem_reg *, bool contig, u8 page);
-int nouveau_mem_host(struct ttm_mem_reg *, struct ttm_dma_tt *);
+		    struct ttm_resource *);
+void nouveau_mem_del(struct ttm_resource *);
+int nouveau_mem_vram(struct ttm_resource *, bool contig, u8 page);
+int nouveau_mem_host(struct ttm_resource *, struct ttm_dma_tt *);
 void nouveau_mem_fini(struct nouveau_mem *);
 int nouveau_mem_map(struct nouveau_mem *, struct nvif_vmm *, struct nvif_vma *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_prime.c b/drivers/gpu/drm/nouveau/nouveau_prime.c
index bae6a3eccee0..b2ecb91f8ddc 100644
--- a/drivers/gpu/drm/nouveau/nouveau_prime.c
+++ b/drivers/gpu/drm/nouveau/nouveau_prime.c
@@ -32,7 +32,7 @@ struct sg_table *nouveau_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	struct nouveau_bo *nvbo = nouveau_gem_object(obj);
 	int npages = nvbo->bo.num_pages;
 
-	return drm_prime_pages_to_sg(nvbo->bo.ttm->pages, npages);
+	return drm_prime_pages_to_sg(obj->dev, nvbo->bo.ttm->pages, npages);
 }
 
 void *nouveau_gem_prime_vmap(struct drm_gem_object *obj)
@@ -64,14 +64,12 @@ struct drm_gem_object *nouveau_gem_prime_import_sg_table(struct drm_device *dev,
 	struct nouveau_bo *nvbo;
 	struct dma_resv *robj = attach->dmabuf->resv;
 	u64 size = attach->dmabuf->size;
-	u32 flags = 0;
 	int align = 0;
 	int ret;
 
-	flags = TTM_PL_FLAG_TT;
-
 	dma_resv_lock(robj, NULL);
-	nvbo = nouveau_bo_alloc(&drm->client, &size, &align, flags, 0, 0);
+	nvbo = nouveau_bo_alloc(&drm->client, &size, &align,
+				NOUVEAU_GEM_DOMAIN_GART, 0, 0);
 	if (IS_ERR(nvbo)) {
 		obj = ERR_CAST(nvbo);
 		goto unlock;
@@ -88,7 +86,8 @@ struct drm_gem_object *nouveau_gem_prime_import_sg_table(struct drm_device *dev,
 		goto unlock;
 	}
 
-	ret = nouveau_bo_init(nvbo, size, align, flags, sg, robj);
+	ret = nouveau_bo_init(nvbo, size, align, NOUVEAU_GEM_DOMAIN_GART,
+			      sg, robj);
 	if (ret) {
 		nouveau_bo_ref(NULL, &nvbo);
 		obj = ERR_PTR(ret);
@@ -108,7 +107,7 @@ int nouveau_gem_prime_pin(struct drm_gem_object *obj)
 	int ret;
 
 	/* pin buffer into GTT */
-	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_TT, false);
+	ret = nouveau_bo_pin(nvbo, NOUVEAU_GEM_DOMAIN_GART, false);
 	if (ret)
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
index c3ccf661b7a6..806d9ec310f5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
@@ -14,87 +14,65 @@ struct nouveau_sgdma_be {
 	struct nouveau_mem *mem;
 };
 
-static void
-nouveau_sgdma_destroy(struct ttm_tt *ttm)
+void
+nouveau_sgdma_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct nouveau_sgdma_be *nvbe = (struct nouveau_sgdma_be *)ttm;
 
 	if (ttm) {
+		nouveau_sgdma_unbind(bdev, ttm);
+		ttm_tt_destroy_common(bdev, ttm);
 		ttm_dma_tt_fini(&nvbe->ttm);
 		kfree(nvbe);
 	}
 }
 
-static int
-nv04_sgdma_bind(struct ttm_tt *ttm, struct ttm_mem_reg *reg)
+int
+nouveau_sgdma_bind(struct ttm_bo_device *bdev, struct ttm_tt *ttm, struct ttm_resource *reg)
 {
 	struct nouveau_sgdma_be *nvbe = (struct nouveau_sgdma_be *)ttm;
+	struct nouveau_drm *drm = nouveau_bdev(bdev);
 	struct nouveau_mem *mem = nouveau_mem(reg);
 	int ret;
 
+	if (nvbe->mem)
+		return 0;
+
 	ret = nouveau_mem_host(reg, &nvbe->ttm);
 	if (ret)
 		return ret;
 
-	ret = nouveau_mem_map(mem, &mem->cli->vmm.vmm, &mem->vma[0]);
-	if (ret) {
-		nouveau_mem_fini(mem);
-		return ret;
+	if (drm->client.device.info.family < NV_DEVICE_INFO_V0_TESLA) {
+		ret = nouveau_mem_map(mem, &mem->cli->vmm.vmm, &mem->vma[0]);
+		if (ret) {
+			nouveau_mem_fini(mem);
+			return ret;
+		}
 	}
 
 	nvbe->mem = mem;
 	return 0;
 }
 
-static void
-nv04_sgdma_unbind(struct ttm_tt *ttm)
+void
+nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct nouveau_sgdma_be *nvbe = (struct nouveau_sgdma_be *)ttm;
-	nouveau_mem_fini(nvbe->mem);
-}
-
-static struct ttm_backend_func nv04_sgdma_backend = {
-	.bind			= nv04_sgdma_bind,
-	.unbind			= nv04_sgdma_unbind,
-	.destroy		= nouveau_sgdma_destroy
-};
-
-static int
-nv50_sgdma_bind(struct ttm_tt *ttm, struct ttm_mem_reg *reg)
-{
-	struct nouveau_sgdma_be *nvbe = (struct nouveau_sgdma_be *)ttm;
-	struct nouveau_mem *mem = nouveau_mem(reg);
-	int ret;
-
-	ret = nouveau_mem_host(reg, &nvbe->ttm);
-	if (ret)
-		return ret;
-
-	nvbe->mem = mem;
-	return 0;
+	if (nvbe->mem) {
+		nouveau_mem_fini(nvbe->mem);
+		nvbe->mem = NULL;
+	}
 }
 
-static struct ttm_backend_func nv50_sgdma_backend = {
-	.bind			= nv50_sgdma_bind,
-	.unbind			= nv04_sgdma_unbind,
-	.destroy		= nouveau_sgdma_destroy
-};
-
 struct ttm_tt *
 nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo, uint32_t page_flags)
 {
-	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
 	struct nouveau_sgdma_be *nvbe;
 
 	nvbe = kzalloc(sizeof(*nvbe), GFP_KERNEL);
 	if (!nvbe)
 		return NULL;
 
-	if (drm->client.device.info.family < NV_DEVICE_INFO_V0_TESLA)
-		nvbe->ttm.ttm.func = &nv04_sgdma_backend;
-	else
-		nvbe->ttm.ttm.func = &nv50_sgdma_backend;
-
 	if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags)) {
 		kfree(nvbe);
 		return NULL;
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index e89ea052cf71..427341753441 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -31,35 +31,17 @@
 
 #include <core/tegra.h>
 
-static int
-nouveau_manager_init(struct ttm_mem_type_manager *man, unsigned long psize)
-{
-	return 0;
-}
-
-static int
-nouveau_manager_fini(struct ttm_mem_type_manager *man)
-{
-	return 0;
-}
-
 static void
-nouveau_manager_del(struct ttm_mem_type_manager *man, struct ttm_mem_reg *reg)
+nouveau_manager_del(struct ttm_resource_manager *man, struct ttm_resource *reg)
 {
 	nouveau_mem_del(reg);
 }
 
-static void
-nouveau_manager_debug(struct ttm_mem_type_manager *man,
-		      struct drm_printer *printer)
-{
-}
-
 static int
-nouveau_vram_manager_new(struct ttm_mem_type_manager *man,
+nouveau_vram_manager_new(struct ttm_resource_manager *man,
 			 struct ttm_buffer_object *bo,
 			 const struct ttm_place *place,
-			 struct ttm_mem_reg *reg)
+			 struct ttm_resource *reg)
 {
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
@@ -81,19 +63,16 @@ nouveau_vram_manager_new(struct ttm_mem_type_manager *man,
 	return 0;
 }
 
-const struct ttm_mem_type_manager_func nouveau_vram_manager = {
-	.init = nouveau_manager_init,
-	.takedown = nouveau_manager_fini,
-	.get_node = nouveau_vram_manager_new,
-	.put_node = nouveau_manager_del,
-	.debug = nouveau_manager_debug,
+const struct ttm_resource_manager_func nouveau_vram_manager = {
+	.alloc = nouveau_vram_manager_new,
+	.free = nouveau_manager_del,
 };
 
 static int
-nouveau_gart_manager_new(struct ttm_mem_type_manager *man,
+nouveau_gart_manager_new(struct ttm_resource_manager *man,
 			 struct ttm_buffer_object *bo,
 			 const struct ttm_place *place,
-			 struct ttm_mem_reg *reg)
+			 struct ttm_resource *reg)
 {
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
@@ -107,19 +86,16 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man,
 	return 0;
 }
 
-const struct ttm_mem_type_manager_func nouveau_gart_manager = {
-	.init = nouveau_manager_init,
-	.takedown = nouveau_manager_fini,
-	.get_node = nouveau_gart_manager_new,
-	.put_node = nouveau_manager_del,
-	.debug = nouveau_manager_debug
+const struct ttm_resource_manager_func nouveau_gart_manager = {
+	.alloc = nouveau_gart_manager_new,
+	.free = nouveau_manager_del,
 };
 
 static int
-nv04_gart_manager_new(struct ttm_mem_type_manager *man,
+nv04_gart_manager_new(struct ttm_resource_manager *man,
 		      struct ttm_buffer_object *bo,
 		      const struct ttm_place *place,
-		      struct ttm_mem_reg *reg)
+		      struct ttm_resource *reg)
 {
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
@@ -142,12 +118,41 @@ nv04_gart_manager_new(struct ttm_mem_type_manager *man,
 	return 0;
 }
 
-const struct ttm_mem_type_manager_func nv04_gart_manager = {
-	.init = nouveau_manager_init,
-	.takedown = nouveau_manager_fini,
-	.get_node = nv04_gart_manager_new,
-	.put_node = nouveau_manager_del,
-	.debug = nouveau_manager_debug
+const struct ttm_resource_manager_func nv04_gart_manager = {
+	.alloc = nv04_gart_manager_new,
+	.free = nouveau_manager_del,
+};
+
+static vm_fault_t nouveau_ttm_fault(struct vm_fault *vmf)
+{
+	struct vm_area_struct *vma = vmf->vma;
+	struct ttm_buffer_object *bo = vma->vm_private_data;
+	pgprot_t prot;
+	vm_fault_t ret;
+
+	ret = ttm_bo_vm_reserve(bo, vmf);
+	if (ret)
+		return ret;
+
+	nouveau_bo_del_io_reserve_lru(bo);
+
+	prot = vm_get_page_prot(vma->vm_flags);
+	ret = ttm_bo_vm_fault_reserved(vmf, prot, TTM_BO_VM_NUM_PREFAULT, 1);
+	if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT))
+		return ret;
+
+	nouveau_bo_add_io_reserve_lru(bo);
+
+	dma_resv_unlock(bo->base.resv);
+
+	return ret;
+}
+
+static struct vm_operations_struct nouveau_ttm_vm_ops = {
+	.fault = nouveau_ttm_fault,
+	.open = ttm_bo_vm_open,
+	.close = ttm_bo_vm_close,
+	.access = ttm_bo_vm_access
 };
 
 int
@@ -155,8 +160,14 @@ nouveau_ttm_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	struct drm_file *file_priv = filp->private_data;
 	struct nouveau_drm *drm = nouveau_drm(file_priv->minor->dev);
+	int ret;
 
-	return ttm_bo_mmap(filp, vma, &drm->ttm.bdev);
+	ret = ttm_bo_mmap(filp, vma, &drm->ttm.bdev);
+	if (ret)
+		return ret;
+
+	vma->vm_ops = &nouveau_ttm_vm_ops;
+	return 0;
 }
 
 static int
@@ -180,6 +191,87 @@ nouveau_ttm_init_host(struct nouveau_drm *drm, u8 kind)
 	return 0;
 }
 
+static int
+nouveau_ttm_init_vram(struct nouveau_drm *drm)
+{
+	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA) {
+		struct ttm_resource_manager *man = kzalloc(sizeof(*man), GFP_KERNEL);
+
+		if (!man)
+			return -ENOMEM;
+
+		man->func = &nouveau_vram_manager;
+
+		ttm_resource_manager_init(man,
+					  drm->gem.vram_available >> PAGE_SHIFT);
+		ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_VRAM, man);
+		ttm_resource_manager_set_used(man, true);
+		return 0;
+	} else {
+		return ttm_range_man_init(&drm->ttm.bdev, TTM_PL_VRAM, false,
+					  drm->gem.vram_available >> PAGE_SHIFT);
+	}
+}
+
+static void
+nouveau_ttm_fini_vram(struct nouveau_drm *drm)
+{
+	struct ttm_resource_manager *man = ttm_manager_type(&drm->ttm.bdev, TTM_PL_VRAM);
+
+	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA) {
+		ttm_resource_manager_set_used(man, false);
+		ttm_resource_manager_force_list_clean(&drm->ttm.bdev, man);
+		ttm_resource_manager_cleanup(man);
+		ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_VRAM, NULL);
+		kfree(man);
+	} else
+		ttm_range_man_fini(&drm->ttm.bdev, TTM_PL_VRAM);
+}
+
+static int
+nouveau_ttm_init_gtt(struct nouveau_drm *drm)
+{
+	struct ttm_resource_manager *man;
+	unsigned long size_pages = drm->gem.gart_available >> PAGE_SHIFT;
+	const struct ttm_resource_manager_func *func = NULL;
+
+	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_TESLA)
+		func = &nouveau_gart_manager;
+	else if (!drm->agp.bridge)
+		func = &nv04_gart_manager;
+	else
+		return ttm_range_man_init(&drm->ttm.bdev, TTM_PL_TT, true,
+					  size_pages);
+
+	man = kzalloc(sizeof(*man), GFP_KERNEL);
+	if (!man)
+		return -ENOMEM;
+
+	man->func = func;
+	man->use_tt = true;
+	ttm_resource_manager_init(man, size_pages);
+	ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_TT, man);
+	ttm_resource_manager_set_used(man, true);
+	return 0;
+}
+
+static void
+nouveau_ttm_fini_gtt(struct nouveau_drm *drm)
+{
+	struct ttm_resource_manager *man = ttm_manager_type(&drm->ttm.bdev, TTM_PL_TT);
+
+	if (drm->client.device.info.family < NV_DEVICE_INFO_V0_TESLA &&
+	    drm->agp.bridge)
+		ttm_range_man_fini(&drm->ttm.bdev, TTM_PL_TT);
+	else {
+		ttm_resource_manager_set_used(man, false);
+		ttm_resource_manager_force_list_clean(&drm->ttm.bdev, man);
+		ttm_resource_manager_cleanup(man);
+		ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_TT, NULL);
+		kfree(man);
+	}
+}
+
 int
 nouveau_ttm_init(struct nouveau_drm *drm)
 {
@@ -237,8 +329,7 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 	arch_io_reserve_memtype_wc(device->func->resource_addr(device, 1),
 				   device->func->resource_size(device, 1));
 
-	ret = ttm_bo_init_mm(&drm->ttm.bdev, TTM_PL_VRAM,
-			      drm->gem.vram_available >> PAGE_SHIFT);
+	ret = nouveau_ttm_init_vram(drm);
 	if (ret) {
 		NV_ERROR(drm, "VRAM mm init failed, %d\n", ret);
 		return ret;
@@ -254,13 +345,15 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 		drm->gem.gart_available = drm->agp.size;
 	}
 
-	ret = ttm_bo_init_mm(&drm->ttm.bdev, TTM_PL_TT,
-			      drm->gem.gart_available >> PAGE_SHIFT);
+	ret = nouveau_ttm_init_gtt(drm);
 	if (ret) {
 		NV_ERROR(drm, "GART mm init failed, %d\n", ret);
 		return ret;
 	}
 
+	mutex_init(&drm->ttm.io_reserve_mutex);
+	INIT_LIST_HEAD(&drm->ttm.io_reserve_lru);
+
 	NV_INFO(drm, "VRAM: %d MiB\n", (u32)(drm->gem.vram_available >> 20));
 	NV_INFO(drm, "GART: %d MiB\n", (u32)(drm->gem.gart_available >> 20));
 	return 0;
@@ -271,8 +364,8 @@ nouveau_ttm_fini(struct nouveau_drm *drm)
 {
 	struct nvkm_device *device = nvxx_device(&drm->client.device);
 
-	ttm_bo_clean_mm(&drm->ttm.bdev, TTM_PL_VRAM);
-	ttm_bo_clean_mm(&drm->ttm.bdev, TTM_PL_TT);
+	nouveau_ttm_fini_vram(drm);
+	nouveau_ttm_fini_gtt(drm);
 
 	ttm_bo_device_release(&drm->ttm.bdev);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.h b/drivers/gpu/drm/nouveau/nouveau_ttm.h
index 085280754b3e..69552049bb96 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.h
@@ -8,9 +8,9 @@ nouveau_bdev(struct ttm_bo_device *bd)
 	return container_of(bd, struct nouveau_drm, ttm.bdev);
 }
 
-extern const struct ttm_mem_type_manager_func nouveau_vram_manager;
-extern const struct ttm_mem_type_manager_func nouveau_gart_manager;
-extern const struct ttm_mem_type_manager_func nv04_gart_manager;
+extern const struct ttm_resource_manager_func nouveau_vram_manager;
+extern const struct ttm_resource_manager_func nouveau_gart_manager;
+extern const struct ttm_resource_manager_func nv04_gart_manager;
 
 struct ttm_tt *nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo,
 					u32 page_flags);
@@ -22,4 +22,7 @@ int  nouveau_ttm_mmap(struct file *, struct vm_area_struct *);
 int  nouveau_ttm_global_init(struct nouveau_drm *);
 void nouveau_ttm_global_release(struct nouveau_drm *);
 
+int nouveau_sgdma_bind(struct ttm_bo_device *bdev, struct ttm_tt *ttm, struct ttm_resource *reg);
+void nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm);
+void nouveau_sgdma_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nv17_fence.c b/drivers/gpu/drm/nouveau/nv17_fence.c
index cd1e87a528a4..1253fdec712d 100644
--- a/drivers/gpu/drm/nouveau/nv17_fence.c
+++ b/drivers/gpu/drm/nouveau/nv17_fence.c
@@ -78,7 +78,7 @@ nv17_fence_context_new(struct nouveau_channel *chan)
 {
 	struct nv10_fence_priv *priv = chan->drm->fence;
 	struct nv10_fence_chan *fctx;
-	struct ttm_mem_reg *reg = &priv->bo->bo.mem;
+	struct ttm_resource *reg = &priv->bo->bo.mem;
 	u32 start = reg->start * PAGE_SIZE;
 	u32 limit = start + reg->size - 1;
 	int ret = 0;
@@ -130,10 +130,11 @@ nv17_fence_create(struct nouveau_drm *drm)
 	priv->base.context_del = nv10_fence_context_del;
 	spin_lock_init(&priv->lock);
 
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
+	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
+			     NOUVEAU_GEM_DOMAIN_VRAM,
 			     0, 0x0000, NULL, NULL, &priv->bo);
 	if (!ret) {
-		ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM, false);
+		ret = nouveau_bo_pin(priv->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 		if (!ret) {
 			ret = nouveau_bo_map(priv->bo);
 			if (ret)
diff --git a/drivers/gpu/drm/nouveau/nv50_fence.c b/drivers/gpu/drm/nouveau/nv50_fence.c
index ebb740686b44..447238e3cbe7 100644
--- a/drivers/gpu/drm/nouveau/nv50_fence.c
+++ b/drivers/gpu/drm/nouveau/nv50_fence.c
@@ -37,7 +37,7 @@ nv50_fence_context_new(struct nouveau_channel *chan)
 {
 	struct nv10_fence_priv *priv = chan->drm->fence;
 	struct nv10_fence_chan *fctx;
-	struct ttm_mem_reg *reg = &priv->bo->bo.mem;
+	struct ttm_resource *reg = &priv->bo->bo.mem;
 	u32 start = reg->start * PAGE_SIZE;
 	u32 limit = start + reg->size - 1;
 	int ret;
@@ -81,10 +81,11 @@ nv50_fence_create(struct nouveau_drm *drm)
 	priv->base.context_del = nv10_fence_context_del;
 	spin_lock_init(&priv->lock);
 
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
+	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
+			     NOUVEAU_GEM_DOMAIN_VRAM,
 			     0, 0x0000, NULL, NULL, &priv->bo);
 	if (!ret) {
-		ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM, false);
+		ret = nouveau_bo_pin(priv->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
 		if (!ret) {
 			ret = nouveau_bo_map(priv->bo);
 			if (ret)
diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c
index 7ed36b3a6b7d..7c9c928c3196 100644
--- a/drivers/gpu/drm/nouveau/nv84_fence.c
+++ b/drivers/gpu/drm/nouveau/nv84_fence.c
@@ -209,12 +209,13 @@ nv84_fence_create(struct nouveau_drm *drm)
 	mutex_init(&priv->mutex);
 
 	/* Use VRAM if there is any ; otherwise fallback to system memory */
-	domain = drm->client.device.info.ram_size != 0 ? TTM_PL_FLAG_VRAM :
-			 /*
-			  * fences created in sysmem must be non-cached or we
-			  * will lose CPU/GPU coherency!
-			  */
-			 TTM_PL_FLAG_TT | TTM_PL_FLAG_UNCACHED;
+	domain = drm->client.device.info.ram_size != 0 ?
+		NOUVEAU_GEM_DOMAIN_VRAM :
+		 /*
+		  * fences created in sysmem must be non-cached or we
+		  * will lose CPU/GPU coherency!
+		  */
+		NOUVEAU_GEM_DOMAIN_GART | NOUVEAU_GEM_DOMAIN_COHERENT;
 	ret = nouveau_bo_new(&drm->client, 16 * drm->chan.nr, 0,
 			     domain, 0, 0, NULL, NULL, &priv->bo);
 	if (ret == 0) {
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index 2578c95570f6..a14fbf06cb30 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -20,7 +20,6 @@
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/clk.h>
-#include <linux/gpio.h>
 #include <linux/regulator/consumer.h>
 #include <linux/component.h>
 #include <linux/of.h>
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index 4d4c1fabd0a1..b738d9750686 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -24,7 +24,6 @@
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/clk.h>
-#include <linux/gpio.h>
 #include <linux/regulator/consumer.h>
 #include <linux/component.h>
 #include <linux/of.h>
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index bd12eae0cb31..5c027c81760f 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -781,7 +781,7 @@ static int venc_probe_of(struct venc_device *venc)
 		venc->type = OMAP_DSS_VENC_TYPE_SVIDEO;
 		break;
 	default:
-		dev_err(&venc->pdev->dev, "bad channel propert '%d'\n",
+		dev_err(&venc->pdev->dev, "bad channel property '%d'\n",
 			channels);
 		r = -EINVAL;
 		goto err;
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 4526967978b7..53d5e184ee77 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -349,13 +349,6 @@ static int omap_modeset_init(struct drm_device *dev)
 
 		drm_connector_attach_encoder(pipe->connector, encoder);
 
-		if (pipe->output->panel) {
-			ret = drm_panel_attach(pipe->output->panel,
-					       pipe->connector);
-			if (ret < 0)
-				return ret;
-		}
-
 		crtc = omap_crtc_init(dev, pipe, priv->planes[i]);
 		if (IS_ERR(crtc))
 			return PTR_ERR(crtc);
@@ -394,18 +387,8 @@ static int omap_modeset_init(struct drm_device *dev)
 
 static void omap_modeset_fini(struct drm_device *ddev)
 {
-	struct omap_drm_private *priv = ddev->dev_private;
-	unsigned int i;
-
 	omap_drm_irq_uninstall(ddev);
 
-	for (i = 0; i < priv->num_pipes; i++) {
-		struct omap_drm_pipeline *pipe = &priv->pipes[i];
-
-		if (pipe->output->panel)
-			drm_panel_detach(pipe->output->panel);
-	}
-
 	drm_mode_config_cleanup(ddev);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c
index d0d12d5dd76c..f67f223c6479 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem.c
@@ -1297,10 +1297,9 @@ struct drm_gem_object *omap_gem_new_dmabuf(struct drm_device *dev, size_t size,
 		omap_obj->dma_addr = sg_dma_address(sgt->sgl);
 	} else {
 		/* Create pages list from sgt */
-		struct sg_page_iter iter;
 		struct page **pages;
 		unsigned int npages;
-		unsigned int i = 0;
+		unsigned int ret;
 
 		npages = DIV_ROUND_UP(size, PAGE_SIZE);
 		pages = kcalloc(npages, sizeof(*pages), GFP_KERNEL);
@@ -1311,14 +1310,9 @@ struct drm_gem_object *omap_gem_new_dmabuf(struct drm_device *dev, size_t size,
 		}
 
 		omap_obj->pages = pages;
-
-		for_each_sg_page(sgt->sgl, &iter, sgt->orig_nents, 0) {
-			pages[i++] = sg_page_iter_page(&iter);
-			if (i > npages)
-				break;
-		}
-
-		if (WARN_ON(i != npages)) {
+		ret = drm_prime_sg_to_page_addr_arrays(sgt, pages, NULL,
+						       npages);
+		if (ret) {
 			omap_gem_free_object(obj);
 			obj = ERR_PTR(-ENOMEM);
 			goto done;
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index de2f2a452be5..b9dbedf8f15e 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -217,6 +217,17 @@ config DRM_PANEL_NOVATEK_NT39016
 	  Say Y here if you want to enable support for the panels built
 	  around the Novatek NT39016 display controller.
 
+config DRM_PANEL_MANTIX_MLAF057WE51
+	tristate "Mantix MLAF057WE51-X MIPI-DSI LCD panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for the Mantix
+	  MLAF057WE51-X MIPI DSI panel as e.g. used in the Librem 5. It
+	  has a resolution of 720x1440 pixels, a built in backlight and touch
+	  controller.
+
 config DRM_PANEL_OLIMEX_LCD_OLINUXINO
 	tristate "Olimex LCD-OLinuXino panel"
 	depends on OF
@@ -313,13 +324,30 @@ config DRM_PANEL_SAMSUNG_S6E63J0X03
 	select VIDEOMODE_HELPERS
 
 config DRM_PANEL_SAMSUNG_S6E63M0
-	tristate "Samsung S6E63M0 RGB/SPI panel"
+	tristate "Samsung S6E63M0 RGB panel"
 	depends on OF
-	depends on SPI
 	depends on BACKLIGHT_CLASS_DEVICE
 	help
 	  Say Y here if you want to enable support for Samsung S6E63M0
-	  AMOLED LCD panel.
+	  AMOLED LCD panel. This panel can be accessed using SPI or
+	  DSI.
+
+config DRM_PANEL_SAMSUNG_S6E63M0_SPI
+	tristate "Samsung S6E63M0 RGB SPI interface"
+	depends on SPI
+	depends on DRM_PANEL_SAMSUNG_S6E63M0
+	default DRM_PANEL_SAMSUNG_S6E63M0
+	help
+	  Say Y here if you want to be able to access the Samsung
+	  S6E63M0 panel using SPI.
+
+config DRM_PANEL_SAMSUNG_S6E63M0_DSI
+	tristate "Samsung S6E63M0 RGB DSI interface"
+	depends on DRM_MIPI_DSI
+	depends on DRM_PANEL_SAMSUNG_S6E63M0
+	help
+	  Say Y here if you want to be able to access the Samsung
+	  S6E63M0 panel using DSI.
 
 config DRM_PANEL_SAMSUNG_S6E88A0_AMS452EF01
 	tristate "Samsung AMS452EF01 panel with S6E88A0 DSI video mode controller"
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index e45ceac6286f..2ba560bca61d 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_DRM_PANEL_LG_LG4573) += panel-lg-lg4573.o
 obj-$(CONFIG_DRM_PANEL_NEC_NL8048HL11) += panel-nec-nl8048hl11.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT35510) += panel-novatek-nt35510.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT39016) += panel-novatek-nt39016.o
+obj-$(CONFIG_DRM_PANEL_MANTIX_MLAF057WE51) += panel-mantix-mlaf057we51.o
 obj-$(CONFIG_DRM_PANEL_OLIMEX_LCD_OLINUXINO) += panel-olimex-lcd-olinuxino.o
 obj-$(CONFIG_DRM_PANEL_ORISETECH_OTM8009A) += panel-orisetech-otm8009a.o
 obj-$(CONFIG_DRM_PANEL_OSD_OSD101T2587_53TS) += panel-osd-osd101t2587-53ts.o
@@ -33,6 +34,8 @@ obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6D16D0) += panel-samsung-s6d16d0.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E3HA2) += panel-samsung-s6e3ha2.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63J0X03) += panel-samsung-s6e63j0x03.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63M0) += panel-samsung-s6e63m0.o
+obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63M0_SPI) += panel-samsung-s6e63m0-spi.o
+obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63M0_DSI) += panel-samsung-s6e63m0-dsi.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E88A0_AMS452EF01) += panel-samsung-s6e88a0-ams452ef01.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0) += panel-samsung-s6e8aa0.o
 obj-$(CONFIG_DRM_PANEL_SEIKO_43WVF1G) += panel-seiko-43wvf1g.o
diff --git a/drivers/gpu/drm/panel/panel-arm-versatile.c b/drivers/gpu/drm/panel/panel-arm-versatile.c
index 47b37fef7ee8..abb0788843c6 100644
--- a/drivers/gpu/drm/panel/panel-arm-versatile.c
+++ b/drivers/gpu/drm/panel/panel-arm-versatile.c
@@ -349,7 +349,9 @@ static int versatile_panel_probe(struct platform_device *pdev)
 	drm_panel_init(&vpanel->panel, dev, &versatile_panel_drm_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&vpanel->panel);
+	drm_panel_add(&vpanel->panel);
+
+	return 0;
 }
 
 static const struct of_device_id versatile_panel_match[] = {
diff --git a/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c b/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
index 9a5b7644d756..e95bc9f60b3f 100644
--- a/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
+++ b/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
@@ -315,11 +315,7 @@ static int tm5p5_nt35596_probe(struct mipi_dsi_device *dsi)
 		return ret;
 	}
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0) {
-		dev_err(dev, "Failed to add panel: %d\n", ret);
-		return ret;
-	}
+	drm_panel_add(&ctx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
diff --git a/drivers/gpu/drm/panel/panel-boe-himax8279d.c b/drivers/gpu/drm/panel/panel-boe-himax8279d.c
index 7c27bd5e3486..42854bd37fd5 100644
--- a/drivers/gpu/drm/panel/panel-boe-himax8279d.c
+++ b/drivers/gpu/drm/panel/panel-boe-himax8279d.c
@@ -19,7 +19,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <video/mipi_display.h>
 
@@ -93,8 +92,7 @@ static int boe_panel_disable(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_off(pinfo->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display off: %d\n", err);
 		return err;
 	}
 
@@ -113,13 +111,11 @@ static int boe_panel_unprepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_off(pinfo->link);
 	if (err < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display off: %d\n", err);
 
 	err = mipi_dsi_dcs_enter_sleep_mode(pinfo->link);
 	if (err < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", err);
 
 	/* sleep_mode_delay: 1ms - 2ms */
 	usleep_range(1000, 2000);
@@ -163,15 +159,13 @@ static int boe_panel_prepare(struct drm_panel *panel)
 	/* send init code */
 	err = send_mipi_cmds(panel, pinfo->desc->on_cmds);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to send DCS Init Code: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to send DCS Init Code: %d\n", err);
 		goto poweroff;
 	}
 
 	err = mipi_dsi_dcs_exit_sleep_mode(pinfo->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to exit sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to exit sleep mode: %d\n", err);
 		goto poweroff;
 	}
 
@@ -180,8 +174,7 @@ static int boe_panel_prepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_on(pinfo->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display on: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display on: %d\n", err);
 		goto poweroff;
 	}
 
@@ -209,8 +202,7 @@ static int boe_panel_enable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(pinfo->link);
 	if (ret < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display on: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to set display on: %d\n", ret);
 		return ret;
 	}
 
@@ -228,8 +220,8 @@ static int boe_panel_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, m);
 	if (!mode) {
-		DRM_DEV_ERROR(pinfo->base.dev, "failed to add mode %ux%u@%u\n",
-			      m->hdisplay, m->vdisplay, drm_mode_vrefresh(m));
+		dev_err(pinfo->base.dev, "failed to add mode %ux%u@%u\n",
+			m->hdisplay, m->vdisplay, drm_mode_vrefresh(m));
 		return -ENOMEM;
 	}
 
@@ -865,8 +857,7 @@ static int panel_add(struct panel_info *pinfo)
 	if (IS_ERR(pinfo->pp18_gpio)) {
 		ret = PTR_ERR(pinfo->pp18_gpio);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to get pp18 gpio: %d\n",
-				      ret);
+			dev_err(dev, "failed to get pp18 gpio: %d\n", ret);
 		return ret;
 	}
 
@@ -874,8 +865,7 @@ static int panel_add(struct panel_info *pinfo)
 	if (IS_ERR(pinfo->pp33_gpio)) {
 		ret = PTR_ERR(pinfo->pp33_gpio);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to get pp33 gpio: %d\n",
-				      ret);
+			dev_err(dev, "failed to get pp33 gpio: %d\n", ret);
 		return ret;
 	}
 
@@ -883,8 +873,7 @@ static int panel_add(struct panel_info *pinfo)
 	if (IS_ERR(pinfo->enable_gpio)) {
 		ret = PTR_ERR(pinfo->enable_gpio);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to get enable gpio: %d\n",
-				      ret);
+			dev_err(dev, "failed to get enable gpio: %d\n", ret);
 		return ret;
 	}
 
@@ -895,7 +884,9 @@ static int panel_add(struct panel_info *pinfo)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&pinfo->base);
+	drm_panel_add(&pinfo->base);
+
+	return 0;
 }
 
 static int panel_probe(struct mipi_dsi_device *dsi)
@@ -935,18 +926,15 @@ static int panel_remove(struct mipi_dsi_device *dsi)
 
 	err = boe_panel_disable(&pinfo->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to disable panel: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to disable panel: %d\n", err);
 
 	err = boe_panel_unprepare(&pinfo->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to unprepare panel: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to unprepare panel: %d\n", err);
 
 	err = mipi_dsi_detach(dsi);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to detach from DSI host: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to detach from DSI host: %d\n", err);
 
 	drm_panel_remove(&pinfo->base);
 
diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index e320aa30b9ae..db9d0b86d542 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -11,6 +11,7 @@
 #include <linux/of_device.h>
 #include <linux/regulator/consumer.h>
 
+#include <drm/drm_connector.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
@@ -43,6 +44,7 @@ struct boe_panel {
 
 	const struct panel_desc *desc;
 
+	enum drm_panel_orientation orientation;
 	struct regulator *pp1800;
 	struct regulator *avee;
 	struct regulator *avdd;
@@ -740,6 +742,7 @@ static int boe_panel_get_modes(struct drm_panel *panel,
 	connector->display_info.width_mm = boe->desc->size.width_mm;
 	connector->display_info.height_mm = boe->desc->size.height_mm;
 	connector->display_info.bpc = boe->desc->bpc;
+	drm_connector_set_panel_orientation(connector, boe->orientation);
 
 	return 1;
 }
@@ -779,6 +782,11 @@ static int boe_panel_add(struct boe_panel *boe)
 
 	drm_panel_init(&boe->base, dev, &boe_panel_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
+	err = of_drm_get_panel_orientation(dev->of_node, &boe->orientation);
+	if (err < 0) {
+		dev_err(dev, "%pOF: failed to get orientation %d\n", dev->of_node, err);
+		return err;
+	}
 
 	err = drm_panel_of_backlight(&boe->base);
 	if (err)
@@ -787,7 +795,9 @@ static int boe_panel_add(struct boe_panel *boe)
 	boe->base.funcs = &boe_panel_funcs;
 	boe->base.dev = &boe->dsi->dev;
 
-	return drm_panel_add(&boe->base);
+	drm_panel_add(&boe->base);
+
+	return 0;
 }
 
 static int boe_panel_probe(struct mipi_dsi_device *dsi)
diff --git a/drivers/gpu/drm/panel/panel-elida-kd35t133.c b/drivers/gpu/drm/panel/panel-elida-kd35t133.c
index e9675514d77b..bc36aa3c1123 100644
--- a/drivers/gpu/drm/panel/panel-elida-kd35t133.c
+++ b/drivers/gpu/drm/panel/panel-elida-kd35t133.c
@@ -22,7 +22,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 /* Manufacturer specific Commands send via DSI */
 #define KD35T133_CMD_INTERFACEMODECTRL		0xb0
@@ -89,7 +88,7 @@ static int kd35t133_init_sequence(struct kd35t133 *ctx)
 			  0xa9, 0x51, 0x2c, 0x82);
 	mipi_dsi_dcs_write(dsi, MIPI_DCS_ENTER_INVERT_MODE, NULL, 0);
 
-	DRM_DEV_DEBUG_DRIVER(dev, "Panel init sequence done\n");
+	dev_dbg(dev, "Panel init sequence done\n");
 	return 0;
 }
 
@@ -104,13 +103,11 @@ static int kd35t133_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to set display off: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to enter sleep mode: %d\n", ret);
 		return ret;
 	}
 
@@ -131,18 +128,16 @@ static int kd35t133_prepare(struct drm_panel *panel)
 	if (ctx->prepared)
 		return 0;
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n");
+	dev_dbg(ctx->dev, "Resetting the panel\n");
 	ret = regulator_enable(ctx->vdd);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable vdd supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable vdd supply: %d\n", ret);
 		return ret;
 	}
 
 	ret = regulator_enable(ctx->iovcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable iovcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret);
 		goto disable_vdd;
 	}
 
@@ -156,7 +151,7 @@ static int kd35t133_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
+		dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -164,14 +159,13 @@ static int kd35t133_prepare(struct drm_panel *panel)
 
 	ret = kd35t133_init_sequence(ctx);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n",
-			      ret);
+		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
 		goto disable_iovcc;
 	}
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to set display on: %d\n", ret);
+		dev_err(ctx->dev, "Failed to set display on: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -210,9 +204,9 @@ static int kd35t133_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to add mode %ux%u@%u\n",
-			      default_mode.hdisplay, default_mode.vdisplay,
-			      drm_mode_vrefresh(&default_mode));
+		dev_err(ctx->dev, "Failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -244,7 +238,7 @@ static int kd35t133_probe(struct mipi_dsi_device *dsi)
 
 	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		dev_err(dev, "cannot get reset gpio\n");
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -252,9 +246,7 @@ static int kd35t133_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->vdd)) {
 		ret = PTR_ERR(ctx->vdd);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request vdd regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request vdd regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -262,9 +254,7 @@ static int kd35t133_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->iovcc)) {
 		ret = PTR_ERR(ctx->iovcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request iovcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request iovcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -288,7 +278,7 @@ static int kd35t133_probe(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "mipi_dsi_attach failed: %d\n", ret);
+		dev_err(dev, "mipi_dsi_attach failed: %d\n", ret);
 		drm_panel_remove(&ctx->panel);
 		return ret;
 	}
@@ -303,13 +293,11 @@ static void kd35t133_shutdown(struct mipi_dsi_device *dsi)
 
 	ret = drm_panel_unprepare(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
 
 	ret = drm_panel_disable(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
 }
 
 static int kd35t133_remove(struct mipi_dsi_device *dsi)
@@ -321,8 +309,7 @@ static int kd35t133_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
 
 	drm_panel_remove(&ctx->panel);
 
diff --git a/drivers/gpu/drm/panel/panel-feixin-k101-im2ba02.c b/drivers/gpu/drm/panel/panel-feixin-k101-im2ba02.c
index 54610651ecdb..2a602aee61c3 100644
--- a/drivers/gpu/drm/panel/panel-feixin-k101-im2ba02.c
+++ b/drivers/gpu/drm/panel/panel-feixin-k101-im2ba02.c
@@ -13,7 +13,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define K101_IM2BA02_INIT_CMD_LEN	2
 
@@ -374,13 +373,11 @@ static int k101_im2ba02_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(ctx->dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to set display off: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", ret);
 
 	msleep(200);
 
@@ -416,10 +413,10 @@ static int k101_im2ba02_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &k101_im2ba02_default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(&ctx->dsi->dev, "failed to add mode %ux%ux@%u\n",
-			      k101_im2ba02_default_mode.hdisplay,
-			      k101_im2ba02_default_mode.vdisplay,
-			      drm_mode_vrefresh(&k101_im2ba02_default_mode));
+		dev_err(&ctx->dsi->dev, "failed to add mode %ux%u@%u\n",
+			k101_im2ba02_default_mode.hdisplay,
+			k101_im2ba02_default_mode.vdisplay,
+			drm_mode_vrefresh(&k101_im2ba02_default_mode));
 		return -ENOMEM;
 	}
 
@@ -460,13 +457,13 @@ static int k101_im2ba02_dsi_probe(struct mipi_dsi_device *dsi)
 	ret = devm_regulator_bulk_get(&dsi->dev, ARRAY_SIZE(ctx->supplies),
 				      ctx->supplies);
 	if (ret < 0) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get regulators\n");
+		dev_err(&dsi->dev, "Couldn't get regulators\n");
 		return ret;
 	}
 
 	ctx->reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our reset GPIO\n");
 		return PTR_ERR(ctx->reset);
 	}
 
@@ -477,9 +474,7 @@ static int k101_im2ba02_dsi_probe(struct mipi_dsi_device *dsi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO;
 	dsi->format = MIPI_DSI_FMT_RGB888;
diff --git a/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c b/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c
index 19a6274b10f5..581661b506f8 100644
--- a/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c
+++ b/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c
@@ -7,7 +7,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <linux/gpio/consumer.h>
 #include <linux/delay.h>
@@ -118,13 +117,11 @@ static int feiyang_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(ctx->dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to set display off: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", ret);
 
 	/* T13 (backlight fall + video & logic signal fall) T13 >= 200ms */
 	msleep(200);
@@ -165,10 +162,10 @@ static int feiyang_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &feiyang_default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(&ctx->dsi->dev, "failed to add mode %ux%ux@%u\n",
-			      feiyang_default_mode.hdisplay,
-			      feiyang_default_mode.vdisplay,
-			      drm_mode_vrefresh(&feiyang_default_mode));
+		dev_err(&ctx->dsi->dev, "failed to add mode %ux%u@%u\n",
+			feiyang_default_mode.hdisplay,
+			feiyang_default_mode.vdisplay,
+			drm_mode_vrefresh(&feiyang_default_mode));
 		return -ENOMEM;
 	}
 
@@ -204,19 +201,19 @@ static int feiyang_dsi_probe(struct mipi_dsi_device *dsi)
 
 	ctx->dvdd = devm_regulator_get(&dsi->dev, "dvdd");
 	if (IS_ERR(ctx->dvdd)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get dvdd regulator\n");
+		dev_err(&dsi->dev, "Couldn't get dvdd regulator\n");
 		return PTR_ERR(ctx->dvdd);
 	}
 
 	ctx->avdd = devm_regulator_get(&dsi->dev, "avdd");
 	if (IS_ERR(ctx->avdd)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get avdd regulator\n");
+		dev_err(&dsi->dev, "Couldn't get avdd regulator\n");
 		return PTR_ERR(ctx->avdd);
 	}
 
 	ctx->reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our reset GPIO\n");
 		return PTR_ERR(ctx->reset);
 	}
 
@@ -224,9 +221,7 @@ static int feiyang_dsi_probe(struct mipi_dsi_device *dsi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_BURST;
 	dsi->format = MIPI_DSI_FMT_RGB888;
diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
index 67a64d1999f6..074e18559b9f 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
@@ -33,7 +33,6 @@
 
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define ILI9322_CHIP_ID			0x00
 #define ILI9322_CHIP_ID_MAGIC		0x96
@@ -683,7 +682,7 @@ static int ili9322_get_modes(struct drm_panel *panel,
 		break;
 	}
 	if (!mode) {
-		DRM_ERROR("bad mode or failed to add mode\n");
+		dev_err(panel->dev, "bad mode or failed to add mode\n");
 		return -EINVAL;
 	}
 	drm_mode_set_name(mode);
@@ -892,7 +891,9 @@ static int ili9322_probe(struct spi_device *spi)
 	drm_panel_init(&ili->panel, dev, &ili9322_drm_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&ili->panel);
+	drm_panel_add(&ili->panel);
+
+	return 0;
 }
 
 static int ili9322_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
index 3ed8635a6fbd..0145129d7c66 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
@@ -10,6 +10,7 @@
 #include <linux/fb.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/of_device.h>
 
 #include <linux/gpio/consumer.h>
 #include <linux/regulator/consumer.h>
@@ -20,14 +21,6 @@
 
 #include <video/mipi_display.h>
 
-struct ili9881c {
-	struct drm_panel	panel;
-	struct mipi_dsi_device	*dsi;
-
-	struct regulator	*power;
-	struct gpio_desc	*reset;
-};
-
 enum ili9881c_op {
 	ILI9881C_SWITCH_PAGE,
 	ILI9881C_COMMAND,
@@ -45,6 +38,21 @@ struct ili9881c_instr {
 	} arg;
 };
 
+struct ili9881c_desc {
+	const struct ili9881c_instr *init;
+	const size_t init_length;
+	const struct drm_display_mode *mode;
+};
+
+struct ili9881c {
+	struct drm_panel	panel;
+	struct mipi_dsi_device	*dsi;
+	const struct ili9881c_desc	*desc;
+
+	struct regulator	*power;
+	struct gpio_desc	*reset;
+};
+
 #define ILI9881C_SWITCH_PAGE_INSTR(_page)	\
 	{					\
 		.op = ILI9881C_SWITCH_PAGE,	\
@@ -64,7 +72,7 @@ struct ili9881c_instr {
 		},					\
 	}
 
-static const struct ili9881c_instr ili9881c_init[] = {
+static const struct ili9881c_instr lhr050h41_init[] = {
 	ILI9881C_SWITCH_PAGE_INSTR(3),
 	ILI9881C_COMMAND_INSTR(0x01, 0x00),
 	ILI9881C_COMMAND_INSTR(0x02, 0x00),
@@ -252,6 +260,199 @@ static const struct ili9881c_instr ili9881c_init[] = {
 	ILI9881C_COMMAND_INSTR(0xD3, 0x3F),
 };
 
+static const struct ili9881c_instr k101_im2byl02_init[] = {
+	ILI9881C_SWITCH_PAGE_INSTR(3),
+	ILI9881C_COMMAND_INSTR(0x01, 0x00),
+	ILI9881C_COMMAND_INSTR(0x02, 0x00),
+	ILI9881C_COMMAND_INSTR(0x03, 0x73),
+	ILI9881C_COMMAND_INSTR(0x04, 0x00),
+	ILI9881C_COMMAND_INSTR(0x05, 0x00),
+	ILI9881C_COMMAND_INSTR(0x06, 0x08),
+	ILI9881C_COMMAND_INSTR(0x07, 0x00),
+	ILI9881C_COMMAND_INSTR(0x08, 0x00),
+	ILI9881C_COMMAND_INSTR(0x09, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0A, 0x01),
+	ILI9881C_COMMAND_INSTR(0x0B, 0x01),
+	ILI9881C_COMMAND_INSTR(0x0C, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0D, 0x01),
+	ILI9881C_COMMAND_INSTR(0x0E, 0x01),
+	ILI9881C_COMMAND_INSTR(0x0F, 0x00),
+	ILI9881C_COMMAND_INSTR(0x10, 0x00),
+	ILI9881C_COMMAND_INSTR(0x11, 0x00),
+	ILI9881C_COMMAND_INSTR(0x12, 0x00),
+	ILI9881C_COMMAND_INSTR(0x13, 0x00),
+	ILI9881C_COMMAND_INSTR(0x14, 0x00),
+	ILI9881C_COMMAND_INSTR(0x15, 0x00),
+	ILI9881C_COMMAND_INSTR(0x16, 0x00),
+	ILI9881C_COMMAND_INSTR(0x17, 0x00),
+	ILI9881C_COMMAND_INSTR(0x18, 0x00),
+	ILI9881C_COMMAND_INSTR(0x19, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1A, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1B, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1C, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1D, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1E, 0x40),
+	ILI9881C_COMMAND_INSTR(0x1F, 0xC0),
+	ILI9881C_COMMAND_INSTR(0x20, 0x06),
+	ILI9881C_COMMAND_INSTR(0x21, 0x01),
+	ILI9881C_COMMAND_INSTR(0x22, 0x06),
+	ILI9881C_COMMAND_INSTR(0x23, 0x01),
+	ILI9881C_COMMAND_INSTR(0x24, 0x88),
+	ILI9881C_COMMAND_INSTR(0x25, 0x88),
+	ILI9881C_COMMAND_INSTR(0x26, 0x00),
+	ILI9881C_COMMAND_INSTR(0x27, 0x00),
+	ILI9881C_COMMAND_INSTR(0x28, 0x3B),
+	ILI9881C_COMMAND_INSTR(0x29, 0x03),
+	ILI9881C_COMMAND_INSTR(0x2A, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2B, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2C, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2D, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2E, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2F, 0x00),
+	ILI9881C_COMMAND_INSTR(0x30, 0x00),
+	ILI9881C_COMMAND_INSTR(0x31, 0x00),
+	ILI9881C_COMMAND_INSTR(0x32, 0x00),
+	ILI9881C_COMMAND_INSTR(0x33, 0x00),
+	ILI9881C_COMMAND_INSTR(0x34, 0x00), /* GPWR1/2 non overlap time 2.62us */
+	ILI9881C_COMMAND_INSTR(0x35, 0x00),
+	ILI9881C_COMMAND_INSTR(0x36, 0x00),
+	ILI9881C_COMMAND_INSTR(0x37, 0x00),
+	ILI9881C_COMMAND_INSTR(0x38, 0x00),
+	ILI9881C_COMMAND_INSTR(0x39, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3A, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3B, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3C, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3D, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3E, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3F, 0x00),
+	ILI9881C_COMMAND_INSTR(0x40, 0x00),
+	ILI9881C_COMMAND_INSTR(0x41, 0x00),
+	ILI9881C_COMMAND_INSTR(0x42, 0x00),
+	ILI9881C_COMMAND_INSTR(0x43, 0x00),
+	ILI9881C_COMMAND_INSTR(0x44, 0x00),
+	ILI9881C_COMMAND_INSTR(0x50, 0x01),
+	ILI9881C_COMMAND_INSTR(0x51, 0x23),
+	ILI9881C_COMMAND_INSTR(0x52, 0x45),
+	ILI9881C_COMMAND_INSTR(0x53, 0x67),
+	ILI9881C_COMMAND_INSTR(0x54, 0x89),
+	ILI9881C_COMMAND_INSTR(0x55, 0xAB),
+	ILI9881C_COMMAND_INSTR(0x56, 0x01),
+	ILI9881C_COMMAND_INSTR(0x57, 0x23),
+	ILI9881C_COMMAND_INSTR(0x58, 0x45),
+	ILI9881C_COMMAND_INSTR(0x59, 0x67),
+	ILI9881C_COMMAND_INSTR(0x5A, 0x89),
+	ILI9881C_COMMAND_INSTR(0x5B, 0xAB),
+	ILI9881C_COMMAND_INSTR(0x5C, 0xCD),
+	ILI9881C_COMMAND_INSTR(0x5D, 0xEF),
+	ILI9881C_COMMAND_INSTR(0x5E, 0x00),
+	ILI9881C_COMMAND_INSTR(0x5F, 0x01),
+	ILI9881C_COMMAND_INSTR(0x60, 0x01),
+	ILI9881C_COMMAND_INSTR(0x61, 0x06),
+	ILI9881C_COMMAND_INSTR(0x62, 0x06),
+	ILI9881C_COMMAND_INSTR(0x63, 0x07),
+	ILI9881C_COMMAND_INSTR(0x64, 0x07),
+	ILI9881C_COMMAND_INSTR(0x65, 0x00),
+	ILI9881C_COMMAND_INSTR(0x66, 0x00),
+	ILI9881C_COMMAND_INSTR(0x67, 0x02),
+	ILI9881C_COMMAND_INSTR(0x68, 0x02),
+	ILI9881C_COMMAND_INSTR(0x69, 0x05),
+	ILI9881C_COMMAND_INSTR(0x6A, 0x05),
+	ILI9881C_COMMAND_INSTR(0x6B, 0x02),
+	ILI9881C_COMMAND_INSTR(0x6C, 0x0D),
+	ILI9881C_COMMAND_INSTR(0x6D, 0x0D),
+	ILI9881C_COMMAND_INSTR(0x6E, 0x0C),
+	ILI9881C_COMMAND_INSTR(0x6F, 0x0C),
+	ILI9881C_COMMAND_INSTR(0x70, 0x0F),
+	ILI9881C_COMMAND_INSTR(0x71, 0x0F),
+	ILI9881C_COMMAND_INSTR(0x72, 0x0E),
+	ILI9881C_COMMAND_INSTR(0x73, 0x0E),
+	ILI9881C_COMMAND_INSTR(0x74, 0x02),
+	ILI9881C_COMMAND_INSTR(0x75, 0x01),
+	ILI9881C_COMMAND_INSTR(0x76, 0x01),
+	ILI9881C_COMMAND_INSTR(0x77, 0x06),
+	ILI9881C_COMMAND_INSTR(0x78, 0x06),
+	ILI9881C_COMMAND_INSTR(0x79, 0x07),
+	ILI9881C_COMMAND_INSTR(0x7A, 0x07),
+	ILI9881C_COMMAND_INSTR(0x7B, 0x00),
+	ILI9881C_COMMAND_INSTR(0x7C, 0x00),
+	ILI9881C_COMMAND_INSTR(0x7D, 0x02),
+	ILI9881C_COMMAND_INSTR(0x7E, 0x02),
+	ILI9881C_COMMAND_INSTR(0x7F, 0x05),
+	ILI9881C_COMMAND_INSTR(0x80, 0x05),
+	ILI9881C_COMMAND_INSTR(0x81, 0x02),
+	ILI9881C_COMMAND_INSTR(0x82, 0x0D),
+	ILI9881C_COMMAND_INSTR(0x83, 0x0D),
+	ILI9881C_COMMAND_INSTR(0x84, 0x0C),
+	ILI9881C_COMMAND_INSTR(0x85, 0x0C),
+	ILI9881C_COMMAND_INSTR(0x86, 0x0F),
+	ILI9881C_COMMAND_INSTR(0x87, 0x0F),
+	ILI9881C_COMMAND_INSTR(0x88, 0x0E),
+	ILI9881C_COMMAND_INSTR(0x89, 0x0E),
+	ILI9881C_COMMAND_INSTR(0x8A, 0x02),
+	ILI9881C_SWITCH_PAGE_INSTR(4),
+	ILI9881C_COMMAND_INSTR(0x3B, 0xC0), /* ILI4003D sel */
+	ILI9881C_COMMAND_INSTR(0x6C, 0x15), /* Set VCORE voltage = 1.5V */
+	ILI9881C_COMMAND_INSTR(0x6E, 0x2A), /* di_pwr_reg=0 for power mode 2A, VGH clamp 18V */
+	ILI9881C_COMMAND_INSTR(0x6F, 0x33), /* pumping ratio VGH=5x VGL=-3x */
+	ILI9881C_COMMAND_INSTR(0x8D, 0x1B), /* VGL clamp -10V */
+	ILI9881C_COMMAND_INSTR(0x87, 0xBA), /* ESD */
+	ILI9881C_COMMAND_INSTR(0x3A, 0x24), /* POWER SAVING */
+	ILI9881C_COMMAND_INSTR(0x26, 0x76),
+	ILI9881C_COMMAND_INSTR(0xB2, 0xD1),
+	ILI9881C_SWITCH_PAGE_INSTR(1),
+	ILI9881C_COMMAND_INSTR(0x22, 0x0A), /* BGR, SS */
+	ILI9881C_COMMAND_INSTR(0x31, 0x00), /* Zigzag type3 inversion */
+	ILI9881C_COMMAND_INSTR(0x40, 0x53), /* ILI4003D sel */
+	ILI9881C_COMMAND_INSTR(0x43, 0x66),
+	ILI9881C_COMMAND_INSTR(0x53, 0x4C),
+	ILI9881C_COMMAND_INSTR(0x50, 0x87),
+	ILI9881C_COMMAND_INSTR(0x51, 0x82),
+	ILI9881C_COMMAND_INSTR(0x60, 0x15),
+	ILI9881C_COMMAND_INSTR(0x61, 0x01),
+	ILI9881C_COMMAND_INSTR(0x62, 0x0C),
+	ILI9881C_COMMAND_INSTR(0x63, 0x00),
+	ILI9881C_COMMAND_INSTR(0xA0, 0x00),
+	ILI9881C_COMMAND_INSTR(0xA1, 0x13), /* VP251 */
+	ILI9881C_COMMAND_INSTR(0xA2, 0x23), /* VP247 */
+	ILI9881C_COMMAND_INSTR(0xA3, 0x14), /* VP243 */
+	ILI9881C_COMMAND_INSTR(0xA4, 0x16), /* VP239 */
+	ILI9881C_COMMAND_INSTR(0xA5, 0x29), /* VP231 */
+	ILI9881C_COMMAND_INSTR(0xA6, 0x1E), /* VP219 */
+	ILI9881C_COMMAND_INSTR(0xA7, 0x1D), /* VP203 */
+	ILI9881C_COMMAND_INSTR(0xA8, 0x86), /* VP175 */
+	ILI9881C_COMMAND_INSTR(0xA9, 0x1E), /* VP144 */
+	ILI9881C_COMMAND_INSTR(0xAA, 0x29), /* VP111 */
+	ILI9881C_COMMAND_INSTR(0xAB, 0x74), /* VP80 */
+	ILI9881C_COMMAND_INSTR(0xAC, 0x19), /* VP52 */
+	ILI9881C_COMMAND_INSTR(0xAD, 0x17), /* VP36 */
+	ILI9881C_COMMAND_INSTR(0xAE, 0x4B), /* VP24 */
+	ILI9881C_COMMAND_INSTR(0xAF, 0x20), /* VP16 */
+	ILI9881C_COMMAND_INSTR(0xB0, 0x26), /* VP12 */
+	ILI9881C_COMMAND_INSTR(0xB1, 0x4C), /* VP8 */
+	ILI9881C_COMMAND_INSTR(0xB2, 0x5D), /* VP4 */
+	ILI9881C_COMMAND_INSTR(0xB3, 0x3F), /* VP0 */
+	ILI9881C_COMMAND_INSTR(0xC0, 0x00), /* VN255 GAMMA N */
+	ILI9881C_COMMAND_INSTR(0xC1, 0x13), /* VN251 */
+	ILI9881C_COMMAND_INSTR(0xC2, 0x23), /* VN247 */
+	ILI9881C_COMMAND_INSTR(0xC3, 0x14), /* VN243 */
+	ILI9881C_COMMAND_INSTR(0xC4, 0x16), /* VN239 */
+	ILI9881C_COMMAND_INSTR(0xC5, 0x29), /* VN231 */
+	ILI9881C_COMMAND_INSTR(0xC6, 0x1E), /* VN219 */
+	ILI9881C_COMMAND_INSTR(0xC7, 0x1D), /* VN203 */
+	ILI9881C_COMMAND_INSTR(0xC8, 0x86), /* VN175 */
+	ILI9881C_COMMAND_INSTR(0xC9, 0x1E), /* VN144 */
+	ILI9881C_COMMAND_INSTR(0xCA, 0x29), /* VN111 */
+	ILI9881C_COMMAND_INSTR(0xCB, 0x74), /* VN80 */
+	ILI9881C_COMMAND_INSTR(0xCC, 0x19), /* VN52 */
+	ILI9881C_COMMAND_INSTR(0xCD, 0x17), /* VN36 */
+	ILI9881C_COMMAND_INSTR(0xCE, 0x4B), /* VN24 */
+	ILI9881C_COMMAND_INSTR(0xCF, 0x20), /* VN16 */
+	ILI9881C_COMMAND_INSTR(0xD0, 0x26), /* VN12 */
+	ILI9881C_COMMAND_INSTR(0xD1, 0x4C), /* VN8 */
+	ILI9881C_COMMAND_INSTR(0xD2, 0x5D), /* VN4 */
+	ILI9881C_COMMAND_INSTR(0xD3, 0x3F), /* VN0 */
+};
+
 static inline struct ili9881c *panel_to_ili9881c(struct drm_panel *panel)
 {
 	return container_of(panel, struct ili9881c, panel);
@@ -311,8 +512,8 @@ static int ili9881c_prepare(struct drm_panel *panel)
 	gpiod_set_value(ctx->reset, 0);
 	msleep(20);
 
-	for (i = 0; i < ARRAY_SIZE(ili9881c_init); i++) {
-		const struct ili9881c_instr *instr = &ili9881c_init[i];
+	for (i = 0; i < ctx->desc->init_length; i++) {
+		const struct ili9881c_instr *instr = &ctx->desc->init[i];
 
 		if (instr->op == ILI9881C_SWITCH_PAGE)
 			ret = ili9881c_switch_page(ctx, instr->arg.page);
@@ -368,7 +569,7 @@ static int ili9881c_unprepare(struct drm_panel *panel)
 	return 0;
 }
 
-static const struct drm_display_mode bananapi_default_mode = {
+static const struct drm_display_mode lhr050h41_default_mode = {
 	.clock		= 62000,
 
 	.hdisplay	= 720,
@@ -380,6 +581,26 @@ static const struct drm_display_mode bananapi_default_mode = {
 	.vsync_start	= 1280 + 10,
 	.vsync_end	= 1280 + 10 + 10,
 	.vtotal		= 1280 + 10 + 10 + 20,
+
+	.width_mm	= 62,
+	.height_mm	= 110,
+};
+
+static const struct drm_display_mode k101_im2byl02_default_mode = {
+	.clock		= 69700,
+
+	.hdisplay	= 800,
+	.hsync_start	= 800 + 6,
+	.hsync_end	= 800 + 6 + 15,
+	.htotal		= 800 + 6 + 15 + 16,
+
+	.vdisplay	= 1280,
+	.vsync_start	= 1280 + 8,
+	.vsync_end	= 1280 + 8 + 48,
+	.vtotal		= 1280 + 8 + 48 + 52,
+
+	.width_mm	= 135,
+	.height_mm	= 217,
 };
 
 static int ili9881c_get_modes(struct drm_panel *panel,
@@ -388,12 +609,12 @@ static int ili9881c_get_modes(struct drm_panel *panel,
 	struct ili9881c *ctx = panel_to_ili9881c(panel);
 	struct drm_display_mode *mode;
 
-	mode = drm_mode_duplicate(connector->dev, &bananapi_default_mode);
+	mode = drm_mode_duplicate(connector->dev, ctx->desc->mode);
 	if (!mode) {
 		dev_err(&ctx->dsi->dev, "failed to add mode %ux%ux@%u\n",
-			bananapi_default_mode.hdisplay,
-			bananapi_default_mode.vdisplay,
-			drm_mode_vrefresh(&bananapi_default_mode));
+			ctx->desc->mode->hdisplay,
+			ctx->desc->mode->vdisplay,
+			drm_mode_vrefresh(ctx->desc->mode));
 		return -ENOMEM;
 	}
 
@@ -402,8 +623,8 @@ static int ili9881c_get_modes(struct drm_panel *panel,
 	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
 	drm_mode_probed_add(connector, mode);
 
-	connector->display_info.width_mm = 62;
-	connector->display_info.height_mm = 110;
+	connector->display_info.width_mm = mode->width_mm;
+	connector->display_info.height_mm = mode->height_mm;
 
 	return 1;
 }
@@ -426,6 +647,7 @@ static int ili9881c_dsi_probe(struct mipi_dsi_device *dsi)
 		return -ENOMEM;
 	mipi_dsi_set_drvdata(dsi, ctx);
 	ctx->dsi = dsi;
+	ctx->desc = of_device_get_match_data(&dsi->dev);
 
 	drm_panel_init(&ctx->panel, &dsi->dev, &ili9881c_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
@@ -446,9 +668,7 @@ static int ili9881c_dsi_probe(struct mipi_dsi_device *dsi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
 	dsi->format = MIPI_DSI_FMT_RGB888;
@@ -467,8 +687,21 @@ static int ili9881c_dsi_remove(struct mipi_dsi_device *dsi)
 	return 0;
 }
 
+static const struct ili9881c_desc lhr050h41_desc = {
+	.init = lhr050h41_init,
+	.init_length = ARRAY_SIZE(lhr050h41_init),
+	.mode = &lhr050h41_default_mode,
+};
+
+static const struct ili9881c_desc k101_im2byl02_desc = {
+	.init = k101_im2byl02_init,
+	.init_length = ARRAY_SIZE(k101_im2byl02_init),
+	.mode = &k101_im2byl02_default_mode,
+};
+
 static const struct of_device_id ili9881c_of_match[] = {
-	{ .compatible = "bananapi,lhr050h41" },
+	{ .compatible = "bananapi,lhr050h41", .data = &lhr050h41_desc },
+	{ .compatible = "feixin,k101-im2byl02", .data = &k101_im2byl02_desc },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, ili9881c_of_match);
diff --git a/drivers/gpu/drm/panel/panel-innolux-p079zca.c b/drivers/gpu/drm/panel/panel-innolux-p079zca.c
index fdf030f4cf92..aea316225391 100644
--- a/drivers/gpu/drm/panel/panel-innolux-p079zca.c
+++ b/drivers/gpu/drm/panel/panel-innolux-p079zca.c
@@ -17,7 +17,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct panel_init_cmd {
 	size_t len;
@@ -85,13 +84,11 @@ static int innolux_panel_unprepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_off(innolux->link);
 	if (err < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display off: %d\n", err);
 
 	err = mipi_dsi_dcs_enter_sleep_mode(innolux->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", err);
 		return err;
 	}
 
@@ -147,8 +144,7 @@ static int innolux_panel_prepare(struct drm_panel *panel)
 			err = mipi_dsi_generic_write(innolux->link, cmd->data,
 						     cmd->len);
 			if (err < 0) {
-				dev_err(panel->dev,
-					"failed to write command %u\n", i);
+				dev_err(panel->dev, "failed to write command %u\n", i);
 				goto poweroff;
 			}
 
@@ -159,8 +155,7 @@ static int innolux_panel_prepare(struct drm_panel *panel)
 			 */
 			err = mipi_dsi_dcs_nop(innolux->link);
 			if (err < 0) {
-				dev_err(panel->dev,
-					"failed to send DCS nop: %d\n", err);
+				dev_err(panel->dev, "failed to send DCS nop: %d\n", err);
 				goto poweroff;
 			}
 		}
@@ -168,8 +163,7 @@ static int innolux_panel_prepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_exit_sleep_mode(innolux->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to exit sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to exit sleep mode: %d\n", err);
 		goto poweroff;
 	}
 
@@ -178,8 +172,7 @@ static int innolux_panel_prepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_on(innolux->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display on: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display on: %d\n", err);
 		goto poweroff;
 	}
 
@@ -398,8 +391,8 @@ static int innolux_panel_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, m);
 	if (!mode) {
-		DRM_DEV_ERROR(panel->dev, "failed to add mode %ux%ux@%u\n",
-			      m->hdisplay, m->vdisplay, drm_mode_vrefresh(m));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			m->hdisplay, m->vdisplay, drm_mode_vrefresh(m));
 		return -ENOMEM;
 	}
 
@@ -475,9 +468,7 @@ static int innolux_panel_add(struct mipi_dsi_device *dsi,
 	if (err)
 		return err;
 
-	err = drm_panel_add(&innolux->base);
-	if (err < 0)
-		return err;
+	drm_panel_add(&innolux->base);
 
 	mipi_dsi_set_drvdata(dsi, innolux);
 	innolux->link = dsi;
@@ -514,17 +505,15 @@ static int innolux_panel_remove(struct mipi_dsi_device *dsi)
 
 	err = drm_panel_unprepare(&innolux->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to unprepare panel: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to unprepare panel: %d\n", err);
 
 	err = drm_panel_disable(&innolux->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to disable panel: %d\n", err);
+		dev_err(&dsi->dev, "failed to disable panel: %d\n", err);
 
 	err = mipi_dsi_detach(dsi);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to detach from DSI host: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to detach from DSI host: %d\n", err);
 
 	innolux_panel_del(innolux);
 
diff --git a/drivers/gpu/drm/panel/panel-jdi-lt070me05000.c b/drivers/gpu/drm/panel/panel-jdi-lt070me05000.c
index 1e3fd6633981..733010b5e4f5 100644
--- a/drivers/gpu/drm/panel/panel-jdi-lt070me05000.c
+++ b/drivers/gpu/drm/panel/panel-jdi-lt070me05000.c
@@ -440,9 +440,9 @@ static int jdi_panel_add(struct jdi_panel *jdi)
 	drm_panel_init(&jdi->base, &jdi->dsi->dev, &jdi_panel_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_add(&jdi->base);
+	drm_panel_add(&jdi->base);
 
-	return ret;
+	return 0;
 }
 
 static void jdi_panel_del(struct jdi_panel *jdi)
diff --git a/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c b/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c
index 0d397af23afe..86e4213e8bb1 100644
--- a/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c
+++ b/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c
@@ -16,7 +16,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct kingdisplay_panel {
 	struct drm_panel base;
@@ -191,8 +190,7 @@ static int kingdisplay_panel_disable(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_off(kingdisplay->link);
 	if (err < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display off: %d\n", err);
 
 	kingdisplay->enabled = false;
 
@@ -209,8 +207,7 @@ static int kingdisplay_panel_unprepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_enter_sleep_mode(kingdisplay->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", err);
 		return err;
 	}
 
@@ -255,16 +252,14 @@ static int kingdisplay_panel_prepare(struct drm_panel *panel)
 		err = mipi_dsi_generic_write(kingdisplay->link, &init_code[i],
 					sizeof(struct kingdisplay_panel_cmd));
 		if (err < 0) {
-			DRM_DEV_ERROR(panel->dev, "failed write init cmds: %d\n",
-				      err);
+			dev_err(panel->dev, "failed write init cmds: %d\n", err);
 			goto poweroff;
 		}
 	}
 
 	err = mipi_dsi_dcs_exit_sleep_mode(kingdisplay->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to exit sleep mode: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to exit sleep mode: %d\n", err);
 		goto poweroff;
 	}
 
@@ -273,8 +268,7 @@ static int kingdisplay_panel_prepare(struct drm_panel *panel)
 
 	err = mipi_dsi_dcs_set_display_on(kingdisplay->link);
 	if (err < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display on: %d\n",
-			      err);
+		dev_err(panel->dev, "failed to set display on: %d\n", err);
 		goto poweroff;
 	}
 
@@ -290,8 +284,7 @@ poweroff:
 
 	regulator_err = regulator_disable(kingdisplay->supply);
 	if (regulator_err)
-		DRM_DEV_ERROR(panel->dev, "failed to disable regulator: %d\n",
-			      regulator_err);
+		dev_err(panel->dev, "failed to disable regulator: %d\n", regulator_err);
 
 	return err;
 }
@@ -327,9 +320,9 @@ static int kingdisplay_panel_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(panel->dev, "failed to add mode %ux%ux@%u\n",
-			      default_mode.hdisplay, default_mode.vdisplay,
-			      drm_mode_vrefresh(&default_mode));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -382,7 +375,9 @@ static int kingdisplay_panel_add(struct kingdisplay_panel *kingdisplay)
 	if (err)
 		return err;
 
-	return drm_panel_add(&kingdisplay->base);
+	drm_panel_add(&kingdisplay->base);
+
+	return 0;
 }
 
 static void kingdisplay_panel_del(struct kingdisplay_panel *kingdisplay)
@@ -421,17 +416,15 @@ static int kingdisplay_panel_remove(struct mipi_dsi_device *dsi)
 
 	err = drm_panel_unprepare(&kingdisplay->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to unprepare panel: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to unprepare panel: %d\n", err);
 
 	err = drm_panel_disable(&kingdisplay->base);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to disable panel: %d\n", err);
+		dev_err(&dsi->dev, "failed to disable panel: %d\n", err);
 
 	err = mipi_dsi_detach(dsi);
 	if (err < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to detach from DSI host: %d\n",
-			      err);
+		dev_err(&dsi->dev, "failed to detach from DSI host: %d\n", err);
 
 	kingdisplay_panel_del(kingdisplay);
 
diff --git a/drivers/gpu/drm/panel/panel-leadtek-ltk050h3146w.c b/drivers/gpu/drm/panel/panel-leadtek-ltk050h3146w.c
index eaa9da3ebbea..ed0d5f959037 100644
--- a/drivers/gpu/drm/panel/panel-leadtek-ltk050h3146w.c
+++ b/drivers/gpu/drm/panel/panel-leadtek-ltk050h3146w.c
@@ -17,7 +17,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct ltk050h3146w_cmd {
 	char cmd;
@@ -314,8 +313,7 @@ static int ltk050h3146w_init_sequence(struct ltk050h3146w *ctx)
 
 	ret = mipi_dsi_dcs_set_tear_on(dsi, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to set tear on: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to set tear on: %d\n", ret);
 		return ret;
 	}
 
@@ -360,8 +358,7 @@ static int ltk050h3146w_a2_write_page(struct ltk050h3146w *ctx, int page,
 
 	ret = ltk050h3146w_a2_select_page(ctx, page);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to select page %d: %d\n",
-			      page, ret);
+		dev_err(ctx->dev, "failed to select page %d: %d\n", page, ret);
 		return ret;
 	}
 
@@ -369,9 +366,7 @@ static int ltk050h3146w_a2_write_page(struct ltk050h3146w *ctx, int page,
 		ret = mipi_dsi_generic_write(dsi, &cmds[i],
 					     sizeof(struct ltk050h3146w_cmd));
 		if (ret < 0) {
-			DRM_DEV_ERROR(ctx->dev,
-				      "failed to write page %d init cmds: %d\n",
-				       page, ret);
+			dev_err(ctx->dev, "failed to write page %d init cmds: %d\n", page, ret);
 			return ret;
 		}
 	}
@@ -405,15 +400,14 @@ static int ltk050h3146w_a2_init_sequence(struct ltk050h3146w *ctx)
 
 	ret = ltk050h3146w_a2_select_page(ctx, 0);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to select page 0: %d\n", ret);
+		dev_err(ctx->dev, "failed to select page 0: %d\n", ret);
 		return ret;
 	}
 
 	/* vendor code called this without param, where there should be one */
 	ret = mipi_dsi_dcs_set_tear_on(dsi, 0);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to set tear on: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to set tear on: %d\n", ret);
 		return ret;
 	}
 
@@ -452,15 +446,13 @@ static int ltk050h3146w_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to set display off: %d\n", ret);
 		return ret;
 	}
 
 	mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to enter sleep mode: %d\n", ret);
 		return ret;
 	}
 
@@ -481,17 +473,15 @@ static int ltk050h3146w_prepare(struct drm_panel *panel)
 	if (ctx->prepared)
 		return 0;
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n");
+	dev_dbg(ctx->dev, "Resetting the panel\n");
 	ret = regulator_enable(ctx->vci);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable vci supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable vci supply: %d\n", ret);
 		return ret;
 	}
 	ret = regulator_enable(ctx->iovcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable iovcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret);
 		goto disable_vci;
 	}
 
@@ -502,14 +492,13 @@ static int ltk050h3146w_prepare(struct drm_panel *panel)
 
 	ret = ctx->panel_desc->init(ctx);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n",
-			      ret);
+		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
 		goto disable_iovcc;
 	}
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
+		dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -518,7 +507,7 @@ static int ltk050h3146w_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to set display on: %d\n", ret);
+		dev_err(ctx->dev, "Failed to set display on: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -577,7 +566,7 @@ static int ltk050h3146w_probe(struct mipi_dsi_device *dsi)
 
 	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		dev_err(dev, "cannot get reset gpio\n");
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -585,9 +574,7 @@ static int ltk050h3146w_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->vci)) {
 		ret = PTR_ERR(ctx->vci);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request vci regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request vci regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -595,9 +582,7 @@ static int ltk050h3146w_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->iovcc)) {
 		ret = PTR_ERR(ctx->iovcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request iovcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request iovcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -621,7 +606,7 @@ static int ltk050h3146w_probe(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "mipi_dsi_attach failed: %d\n", ret);
+		dev_err(dev, "mipi_dsi_attach failed: %d\n", ret);
 		drm_panel_remove(&ctx->panel);
 		return ret;
 	}
@@ -636,13 +621,11 @@ static void ltk050h3146w_shutdown(struct mipi_dsi_device *dsi)
 
 	ret = drm_panel_unprepare(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
 
 	ret = drm_panel_disable(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
 }
 
 static int ltk050h3146w_remove(struct mipi_dsi_device *dsi)
@@ -654,8 +637,7 @@ static int ltk050h3146w_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
 
 	drm_panel_remove(&ctx->panel);
 
diff --git a/drivers/gpu/drm/panel/panel-leadtek-ltk500hd1829.c b/drivers/gpu/drm/panel/panel-leadtek-ltk500hd1829.c
index 0f6a248c47a5..3c00e4f8f803 100644
--- a/drivers/gpu/drm/panel/panel-leadtek-ltk500hd1829.c
+++ b/drivers/gpu/drm/panel/panel-leadtek-ltk500hd1829.c
@@ -20,7 +20,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct ltk500hd1829 {
 	struct device *dev;
@@ -278,13 +277,11 @@ static int ltk500hd1829_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to set display off: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to enter sleep mode: %d\n", ret);
 	}
 
 	/* 120ms to enter sleep mode */
@@ -310,14 +307,12 @@ static int ltk500hd1829_prepare(struct drm_panel *panel)
 
 	ret = regulator_enable(ctx->vcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable vci supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable vci supply: %d\n", ret);
 		return ret;
 	}
 	ret = regulator_enable(ctx->iovcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable iovcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret);
 		goto disable_vcc;
 	}
 
@@ -333,16 +328,14 @@ static int ltk500hd1829_prepare(struct drm_panel *panel)
 		ret = mipi_dsi_generic_write(dsi, &init_code[i],
 					     sizeof(struct ltk500hd1829_cmd));
 		if (ret < 0) {
-			DRM_DEV_ERROR(panel->dev,
-				      "failed to write init cmds: %d\n", ret);
+			dev_err(panel->dev, "failed to write init cmds: %d\n", ret);
 			goto disable_iovcc;
 		}
 	}
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to exit sleep mode: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to exit sleep mode: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -351,8 +344,7 @@ static int ltk500hd1829_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(panel->dev, "failed to set display on: %d\n",
-			      ret);
+		dev_err(panel->dev, "failed to set display on: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -389,9 +381,9 @@ static int ltk500hd1829_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->dev, "failed to add mode %ux%ux@%u\n",
-			      default_mode.hdisplay, default_mode.vdisplay,
-			      drm_mode_vrefresh(&default_mode));
+		dev_err(ctx->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -423,7 +415,7 @@ static int ltk500hd1829_probe(struct mipi_dsi_device *dsi)
 
 	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		dev_err(dev, "cannot get reset gpio\n");
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -431,9 +423,7 @@ static int ltk500hd1829_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->vcc)) {
 		ret = PTR_ERR(ctx->vcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request vcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request vcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -441,9 +431,7 @@ static int ltk500hd1829_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->iovcc)) {
 		ret = PTR_ERR(ctx->iovcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request iovcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request iovcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -467,7 +455,7 @@ static int ltk500hd1829_probe(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "mipi_dsi_attach failed: %d\n", ret);
+		dev_err(dev, "mipi_dsi_attach failed: %d\n", ret);
 		drm_panel_remove(&ctx->panel);
 		return ret;
 	}
@@ -482,13 +470,11 @@ static void ltk500hd1829_shutdown(struct mipi_dsi_device *dsi)
 
 	ret = drm_panel_unprepare(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
 
 	ret = drm_panel_disable(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
 }
 
 static int ltk500hd1829_remove(struct mipi_dsi_device *dsi)
@@ -500,8 +486,7 @@ static int ltk500hd1829_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "failed to detach from DSI host: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "failed to detach from DSI host: %d\n", ret);
 
 	drm_panel_remove(&ctx->panel);
 
diff --git a/drivers/gpu/drm/panel/panel-lg-lb035q02.c b/drivers/gpu/drm/panel/panel-lg-lb035q02.c
index 14456b9cd5c0..f3183b68704f 100644
--- a/drivers/gpu/drm/panel/panel-lg-lb035q02.c
+++ b/drivers/gpu/drm/panel/panel-lg-lb035q02.c
@@ -198,7 +198,9 @@ static int lb035q02_probe(struct spi_device *spi)
 	drm_panel_init(&lcd->panel, &lcd->spi->dev, &lb035q02_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&lcd->panel);
+	drm_panel_add(&lcd->panel);
+
+	return 0;
 }
 
 static int lb035q02_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-lg-lg4573.c b/drivers/gpu/drm/panel/panel-lg-lg4573.c
index aedc485d0727..8e5160af1de5 100644
--- a/drivers/gpu/drm/panel/panel-lg-lg4573.c
+++ b/drivers/gpu/drm/panel/panel-lg-lg4573.c
@@ -261,7 +261,9 @@ static int lg4573_probe(struct spi_device *spi)
 	drm_panel_init(&ctx->panel, &spi->dev, &lg4573_drm_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&ctx->panel);
+	drm_panel_add(&ctx->panel);
+
+	return 0;
 }
 
 static int lg4573_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-lvds.c b/drivers/gpu/drm/panel/panel-lvds.c
index 5ce3f4a2b7a1..66c7d765b8f7 100644
--- a/drivers/gpu/drm/panel/panel-lvds.c
+++ b/drivers/gpu/drm/panel/panel-lvds.c
@@ -37,6 +37,8 @@ struct panel_lvds {
 
 	struct gpio_desc *enable_gpio;
 	struct gpio_desc *reset_gpio;
+
+	enum drm_panel_orientation orientation;
 };
 
 static inline struct panel_lvds *to_panel_lvds(struct drm_panel *panel)
@@ -99,6 +101,7 @@ static int panel_lvds_get_modes(struct drm_panel *panel,
 	connector->display_info.bus_flags = lvds->data_mirror
 					  ? DRM_BUS_FLAG_DATA_LSB_TO_MSB
 					  : DRM_BUS_FLAG_DATA_MSB_TO_LSB;
+	drm_connector_set_panel_orientation(connector, lvds->orientation);
 
 	return 1;
 }
@@ -116,6 +119,12 @@ static int panel_lvds_parse_dt(struct panel_lvds *lvds)
 	const char *mapping;
 	int ret;
 
+	ret = of_drm_get_panel_orientation(np, &lvds->orientation);
+	if (ret < 0) {
+		dev_err(lvds->dev, "%pOF: failed to get orientation %d\n", np, ret);
+		return ret;
+	}
+
 	ret = of_get_display_timing(np, "panel-timing", &timing);
 	if (ret < 0) {
 		dev_err(lvds->dev, "%pOF: problems parsing panel-timing (%d)\n",
@@ -227,9 +236,7 @@ static int panel_lvds_probe(struct platform_device *pdev)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&lvds->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&lvds->panel);
 
 	dev_set_drvdata(lvds->dev, lvds);
 	return 0;
diff --git a/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c
new file mode 100644
index 000000000000..3482e28e30fc
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c
@@ -0,0 +1,328 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Mantix MLAF057WE51 5.7" MIPI-DSI panel driver
+ *
+ * Copyright (C) Purism SPC 2020
+ */
+
+#include <linux/backlight.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/regulator/consumer.h>
+
+#include <video/mipi_display.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+
+#define DRV_NAME "panel-mantix-mlaf057we51"
+
+/* Manufacturer specific Commands send via DSI */
+#define MANTIX_CMD_OTP_STOP_RELOAD_MIPI 0x41
+#define MANTIX_CMD_INT_CANCEL           0x4C
+
+struct mantix {
+	struct device *dev;
+	struct drm_panel panel;
+	struct gpio_desc *reset_gpio;
+
+	struct regulator *avdd;
+	struct regulator *avee;
+	struct regulator *vddi;
+};
+
+static inline struct mantix *panel_to_mantix(struct drm_panel *panel)
+{
+	return container_of(panel, struct mantix, panel);
+}
+
+#define dsi_generic_write_seq(dsi, seq...) do {				\
+		static const u8 d[] = { seq };				\
+		int ret;						\
+		ret = mipi_dsi_generic_write(dsi, d, ARRAY_SIZE(d));	\
+		if (ret < 0)						\
+			return ret;					\
+	} while (0)
+
+static int mantix_init_sequence(struct mantix *ctx)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+	struct device *dev = ctx->dev;
+
+	/*
+	 * Init sequence was supplied by the panel vendor.
+	 */
+	dsi_generic_write_seq(dsi, MANTIX_CMD_OTP_STOP_RELOAD_MIPI, 0x5A);
+
+	dsi_generic_write_seq(dsi, MANTIX_CMD_INT_CANCEL, 0x03);
+	dsi_generic_write_seq(dsi, MANTIX_CMD_OTP_STOP_RELOAD_MIPI, 0x5A, 0x03);
+	dsi_generic_write_seq(dsi, 0x80, 0xA9, 0x00);
+
+	dsi_generic_write_seq(dsi, MANTIX_CMD_OTP_STOP_RELOAD_MIPI, 0x5A, 0x09);
+	dsi_generic_write_seq(dsi, 0x80, 0x64, 0x00, 0x64, 0x00, 0x00);
+	msleep(20);
+
+	dev_dbg(dev, "Panel init sequence done\n");
+	return 0;
+}
+
+static int mantix_enable(struct drm_panel *panel)
+{
+	struct mantix *ctx = panel_to_mantix(panel);
+	struct device *dev = ctx->dev;
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+	int ret;
+
+	ret = mantix_init_sequence(ctx);
+	if (ret < 0) {
+		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to exit sleep mode\n");
+		return ret;
+	}
+	msleep(20);
+
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret)
+		return ret;
+	usleep_range(10000, 12000);
+
+	ret = mipi_dsi_turn_on_peripheral(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to turn on peripheral\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int mantix_disable(struct drm_panel *panel)
+{
+	struct mantix *ctx = panel_to_mantix(panel);
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+	int ret;
+
+	ret = mipi_dsi_dcs_set_display_off(dsi);
+	if (ret < 0)
+		dev_err(ctx->dev, "Failed to turn off the display: %d\n", ret);
+
+	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
+	if (ret < 0)
+		dev_err(ctx->dev, "Failed to enter sleep mode: %d\n", ret);
+
+
+	return 0;
+}
+
+static int mantix_unprepare(struct drm_panel *panel)
+{
+	struct mantix *ctx = panel_to_mantix(panel);
+
+	regulator_disable(ctx->avee);
+	regulator_disable(ctx->avdd);
+	/* T11 */
+	usleep_range(5000, 6000);
+	regulator_disable(ctx->vddi);
+	/* T14 */
+	msleep(50);
+
+	return 0;
+}
+
+static int mantix_prepare(struct drm_panel *panel)
+{
+	struct mantix *ctx = panel_to_mantix(panel);
+	int ret;
+
+	/* Focaltech FT8006P, section 7.3.1 and 7.3.4 */
+	dev_dbg(ctx->dev, "Resetting the panel\n");
+	ret = regulator_enable(ctx->vddi);
+	if (ret < 0) {
+		dev_err(ctx->dev, "Failed to enable vddi supply: %d\n", ret);
+		return ret;
+	}
+
+	/* T1 + T2 */
+	usleep_range(8000, 10000);
+
+	ret = regulator_enable(ctx->avdd);
+	if (ret < 0) {
+		dev_err(ctx->dev, "Failed to enable avdd supply: %d\n", ret);
+		return ret;
+	}
+
+	/* T2d */
+	usleep_range(3500, 4000);
+	ret = regulator_enable(ctx->avee);
+	if (ret < 0) {
+		dev_err(ctx->dev, "Failed to enable avee supply: %d\n", ret);
+		return ret;
+	}
+
+	/* T3+T5 */
+	usleep_range(10000, 12000);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(5150, 7000);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+
+	/* T6 */
+	msleep(50);
+
+	return 0;
+}
+
+static const struct drm_display_mode default_mode = {
+	.hdisplay    = 720,
+	.hsync_start = 720 + 45,
+	.hsync_end   = 720 + 45 + 14,
+	.htotal	     = 720 + 45 + 14 + 25,
+	.vdisplay    = 1440,
+	.vsync_start = 1440 + 130,
+	.vsync_end   = 1440 + 130 + 8,
+	.vtotal	     = 1440 + 130 + 8 + 106,
+	.clock	     = 85298,
+	.flags	     = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+	.width_mm    = 65,
+	.height_mm   = 130,
+};
+
+static int mantix_get_modes(struct drm_panel *panel,
+			    struct drm_connector *connector)
+{
+	struct mantix *ctx = panel_to_mantix(panel);
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(connector->dev, &default_mode);
+	if (!mode) {
+		dev_err(ctx->dev, "Failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(mode));
+		return -ENOMEM;
+	}
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	connector->display_info.width_mm = mode->width_mm;
+	connector->display_info.height_mm = mode->height_mm;
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs mantix_drm_funcs = {
+	.disable   = mantix_disable,
+	.unprepare = mantix_unprepare,
+	.prepare   = mantix_prepare,
+	.enable	   = mantix_enable,
+	.get_modes = mantix_get_modes,
+};
+
+static int mantix_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct mantix *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->reset_gpio)) {
+		dev_err(dev, "cannot get reset gpio\n");
+		return PTR_ERR(ctx->reset_gpio);
+	}
+
+	mipi_dsi_set_drvdata(dsi, ctx);
+	ctx->dev = dev;
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO |
+		MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
+
+	ctx->avdd = devm_regulator_get(dev, "avdd");
+	if (IS_ERR(ctx->avdd))
+		return dev_err_probe(dev, PTR_ERR(ctx->avdd), "Failed to request avdd regulator\n");
+
+	ctx->avee = devm_regulator_get(dev, "avee");
+	if (IS_ERR(ctx->avee))
+		return dev_err_probe(dev, PTR_ERR(ctx->avee), "Failed to request avee regulator\n");
+
+	ctx->vddi = devm_regulator_get(dev, "vddi");
+	if (IS_ERR(ctx->vddi))
+		return dev_err_probe(dev, PTR_ERR(ctx->vddi), "Failed to request vddi regulator\n");
+
+	drm_panel_init(&ctx->panel, dev, &mantix_drm_funcs,
+		       DRM_MODE_CONNECTOR_DSI);
+
+	ret = drm_panel_of_backlight(&ctx->panel);
+	if (ret)
+		return ret;
+
+	drm_panel_add(&ctx->panel);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		dev_err(dev, "mipi_dsi_attach failed (%d). Is host ready?\n", ret);
+		drm_panel_remove(&ctx->panel);
+		return ret;
+	}
+
+	dev_info(dev, "%ux%u@%u %ubpp dsi %udl - ready\n",
+		 default_mode.hdisplay, default_mode.vdisplay,
+		 drm_mode_vrefresh(&default_mode),
+		 mipi_dsi_pixel_format_to_bpp(dsi->format), dsi->lanes);
+
+	return 0;
+}
+
+static void mantix_shutdown(struct mipi_dsi_device *dsi)
+{
+	struct mantix *ctx = mipi_dsi_get_drvdata(dsi);
+
+	drm_panel_unprepare(&ctx->panel);
+	drm_panel_disable(&ctx->panel);
+}
+
+static int mantix_remove(struct mipi_dsi_device *dsi)
+{
+	struct mantix *ctx = mipi_dsi_get_drvdata(dsi);
+
+	mantix_shutdown(dsi);
+
+	mipi_dsi_detach(dsi);
+	drm_panel_remove(&ctx->panel);
+
+	return 0;
+}
+
+static const struct of_device_id mantix_of_match[] = {
+	{ .compatible = "mantix,mlaf057we51-x" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, mantix_of_match);
+
+static struct mipi_dsi_driver mantix_driver = {
+	.probe	= mantix_probe,
+	.remove = mantix_remove,
+	.shutdown = mantix_shutdown,
+	.driver = {
+		.name = DRV_NAME,
+		.of_match_table = mantix_of_match,
+	},
+};
+module_mipi_dsi_driver(mantix_driver);
+
+MODULE_AUTHOR("Guido Günther <agx@sigxcpu.org>");
+MODULE_DESCRIPTION("DRM driver for Mantix MLAF057WE51-X MIPI DSI panel");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panel/panel-nec-nl8048hl11.c b/drivers/gpu/drm/panel/panel-nec-nl8048hl11.c
index f894971c1c7c..6e5ab1debc8b 100644
--- a/drivers/gpu/drm/panel/panel-nec-nl8048hl11.c
+++ b/drivers/gpu/drm/panel/panel-nec-nl8048hl11.c
@@ -207,7 +207,9 @@ static int nl8048_probe(struct spi_device *spi)
 	drm_panel_init(&lcd->panel, &lcd->spi->dev, &nl8048_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&lcd->panel);
+	drm_panel_add(&lcd->panel);
+
+	return 0;
 }
 
 static int nl8048_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35510.c b/drivers/gpu/drm/panel/panel-novatek-nt35510.c
index e98d54df00e7..b9a0e56f33e2 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt35510.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt35510.c
@@ -35,7 +35,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define MCS_CMD_MAUCCTR		0xF0 /* Manufacturer command enable */
 #define MCS_CMD_READ_ID1	0xDA
@@ -376,6 +375,10 @@ struct nt35510 {
 };
 
 /* Manufacturer command has strictly this byte sequence */
+static const u8 nt35510_mauc_mtp_read_param[] = { 0xAA, 0x55, 0x25, 0x01 };
+static const u8 nt35510_mauc_mtp_read_setting[] = { 0x01, 0x02, 0x00, 0x20,
+						    0x33, 0x13, 0x00, 0x40,
+						    0x00, 0x00, 0x23, 0x02 };
 static const u8 nt35510_mauc_select_page_0[] = { 0x55, 0xAA, 0x52, 0x08, 0x00 };
 static const u8 nt35510_mauc_select_page_1[] = { 0x55, 0xAA, 0x52, 0x08, 0x01 };
 static const u8 nt35510_vgh_on[] = { 0x01 };
@@ -400,9 +403,7 @@ static int nt35510_send_long(struct nt35510 *nt, struct mipi_dsi_device *dsi,
 		chunk = 15;
 	ret = mipi_dsi_dcs_write(dsi, cmd, seqp, chunk);
 	if (ret < 0) {
-		DRM_DEV_ERROR(nt->dev,
-			      "error sending DCS command seq cmd %02x\n",
-			      cmd);
+		dev_err(nt->dev, "error sending DCS command seq cmd %02x\n", cmd);
 		return ret;
 	}
 	cmdwritten += chunk;
@@ -414,16 +415,13 @@ static int nt35510_send_long(struct nt35510 *nt, struct mipi_dsi_device *dsi,
 			chunk = 15;
 		ret = mipi_dsi_generic_write(dsi, seqp, chunk);
 		if (ret < 0) {
-			DRM_DEV_ERROR(nt->dev,
-				      "error sending generic write seq %02x\n",
-				      cmd);
+			dev_err(nt->dev, "error sending generic write seq %02x\n", cmd);
 			return ret;
 		}
 		cmdwritten += chunk;
 		seqp += chunk;
 	}
-	DRM_DEV_DEBUG(nt->dev, "sent command %02x %02x bytes\n",
-		      cmd, cmdlen);
+	dev_dbg(nt->dev, "sent command %02x %02x bytes\n", cmd, cmdlen);
 	return 0;
 }
 
@@ -435,17 +433,17 @@ static int nt35510_read_id(struct nt35510 *nt)
 
 	ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID1, &id1, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(nt->dev, "could not read MTP ID1\n");
+		dev_err(nt->dev, "could not read MTP ID1\n");
 		return ret;
 	}
 	ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID2, &id2, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(nt->dev, "could not read MTP ID2\n");
+		dev_err(nt->dev, "could not read MTP ID2\n");
 		return ret;
 	}
 	ret = mipi_dsi_dcs_read(dsi, MCS_CMD_READ_ID3, &id3, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(nt->dev, "could not read MTP ID3\n");
+		dev_err(nt->dev, "could not read MTP ID3\n");
 		return ret;
 	}
 
@@ -454,9 +452,7 @@ static int nt35510_read_id(struct nt35510 *nt)
 	 * ID (e.g. Hydis 0x55), driver ID (e.g. NT35510 0xc0) and
 	 * version.
 	 */
-	DRM_DEV_INFO(nt->dev,
-		     "MTP ID manufacturer: %02x version: %02x driver: %02x\n",
-		     id1, id2, id3);
+	dev_info(nt->dev, "MTP ID manufacturer: %02x version: %02x driver: %02x\n", id1, id2, id3);
 
 	return 0;
 }
@@ -657,7 +653,7 @@ static int nt35510_set_brightness(struct backlight_device *bl)
 	u8 brightness = bl->props.brightness;
 	int ret;
 
-	DRM_DEV_DEBUG(nt->dev, "set brightness %d\n", brightness);
+	dev_dbg(nt->dev, "set brightness %d\n", brightness);
 	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_SET_DISPLAY_BRIGHTNESS,
 				 &brightness,
 				 sizeof(brightness));
@@ -698,6 +694,18 @@ static int nt35510_power_on(struct nt35510 *nt)
 		usleep_range(120000, 140000);
 	}
 
+	ret = nt35510_send_long(nt, dsi, MCS_CMD_MTP_READ_PARAM,
+				ARRAY_SIZE(nt35510_mauc_mtp_read_param),
+				nt35510_mauc_mtp_read_param);
+	if (ret)
+		return ret;
+
+	ret = nt35510_send_long(nt, dsi, MCS_CMD_MTP_READ_SETTING,
+				ARRAY_SIZE(nt35510_mauc_mtp_read_setting),
+				nt35510_mauc_mtp_read_setting);
+	if (ret)
+		return ret;
+
 	ret = nt35510_read_id(nt);
 	if (ret)
 		return ret;
@@ -780,8 +788,7 @@ static int nt35510_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(nt->dev, "failed to turn display off (%d)\n",
-			      ret);
+		dev_err(nt->dev, "failed to turn display off (%d)\n", ret);
 		return ret;
 	}
 	usleep_range(10000, 20000);
@@ -789,8 +796,7 @@ static int nt35510_unprepare(struct drm_panel *panel)
 	/* Enter sleep mode */
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(nt->dev, "failed to enter sleep mode (%d)\n",
-			      ret);
+		dev_err(nt->dev, "failed to enter sleep mode (%d)\n", ret);
 		return ret;
 	}
 
@@ -817,8 +823,7 @@ static int nt35510_prepare(struct drm_panel *panel)
 	/* Exit sleep mode */
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(nt->dev, "failed to exit sleep mode (%d)\n",
-			      ret);
+		dev_err(nt->dev, "failed to exit sleep mode (%d)\n", ret);
 		return ret;
 	}
 	/* Up to 120 ms */
@@ -826,8 +831,7 @@ static int nt35510_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(nt->dev, "failed to turn display on (%d)\n",
-			      ret);
+		dev_err(nt->dev, "failed to turn display on (%d)\n", ret);
 		return ret;
 	}
 	/* Some 10 ms */
@@ -848,7 +852,7 @@ static int nt35510_get_modes(struct drm_panel *panel,
 	info->height_mm = nt->conf->height_mm;
 	mode = drm_mode_duplicate(connector->dev, &nt->conf->mode);
 	if (!mode) {
-		DRM_ERROR("bad mode or failed to add mode\n");
+		dev_err(panel->dev, "bad mode or failed to add mode\n");
 		return -EINVAL;
 	}
 	drm_mode_set_name(mode);
@@ -947,7 +951,7 @@ static int nt35510_probe(struct mipi_dsi_device *dsi)
 		bl = devm_backlight_device_register(dev, "nt35510", dev, nt,
 						    &nt35510_bl_ops, NULL);
 		if (IS_ERR(bl)) {
-			DRM_DEV_ERROR(dev, "failed to register backlight device\n");
+			dev_err(dev, "failed to register backlight device\n");
 			return PTR_ERR(bl);
 		}
 		bl->props.max_brightness = 255;
@@ -956,9 +960,7 @@ static int nt35510_probe(struct mipi_dsi_device *dsi)
 		nt->panel.backlight = bl;
 	}
 
-	ret = drm_panel_add(&nt->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&nt->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0)
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt39016.c b/drivers/gpu/drm/panel/panel-novatek-nt39016.c
index 91df050ba3f6..f8151fe3ac9a 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt39016.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt39016.c
@@ -6,7 +6,6 @@
  * Copyright (C) 2019, Paul Cercueil <paul@crapouillou.net>
  */
 
-#include <linux/backlight.h>
 #include <linux/delay.h>
 #include <linux/device.h>
 #include <linux/gpio/consumer.h>
@@ -57,14 +56,11 @@ struct nt39016_panel_info {
 
 struct nt39016 {
 	struct drm_panel drm_panel;
-	struct device *dev;
 	struct regmap *map;
 	struct regulator *supply;
 	const struct nt39016_panel_info *panel_info;
 
 	struct gpio_desc *reset_gpio;
-
-	struct backlight_device *backlight;
 };
 
 static inline struct nt39016 *to_nt39016(struct drm_panel *panel)
@@ -127,7 +123,7 @@ static int nt39016_prepare(struct drm_panel *drm_panel)
 
 	err = regulator_enable(panel->supply);
 	if (err) {
-		dev_err(panel->dev, "Failed to enable power supply: %d", err);
+		dev_err(drm_panel->dev, "Failed to enable power supply: %d\n", err);
 		return err;
 	}
 
@@ -146,7 +142,7 @@ static int nt39016_prepare(struct drm_panel *drm_panel)
 	err = regmap_multi_reg_write(panel->map, nt39016_panel_regs,
 				     ARRAY_SIZE(nt39016_panel_regs));
 	if (err) {
-		dev_err(panel->dev, "Failed to init registers: %d", err);
+		dev_err(drm_panel->dev, "Failed to init registers: %d\n", err);
 		goto err_disable_regulator;
 	}
 
@@ -176,18 +172,16 @@ static int nt39016_enable(struct drm_panel *drm_panel)
 	ret = regmap_write(panel->map, NT39016_REG_SYSTEM,
 			   NT39016_SYSTEM_RESET_N | NT39016_SYSTEM_STANDBY);
 	if (ret) {
-		dev_err(panel->dev, "Unable to enable panel: %d", ret);
+		dev_err(drm_panel->dev, "Unable to enable panel: %d\n", ret);
 		return ret;
 	}
 
-	if (panel->backlight) {
+	if (drm_panel->backlight) {
 		/* Wait for the picture to be ready before enabling backlight */
 		msleep(150);
-
-		ret = backlight_enable(panel->backlight);
 	}
 
-	return ret;
+	return 0;
 }
 
 static int nt39016_disable(struct drm_panel *drm_panel)
@@ -195,12 +189,10 @@ static int nt39016_disable(struct drm_panel *drm_panel)
 	struct nt39016 *panel = to_nt39016(drm_panel);
 	int err;
 
-	backlight_disable(panel->backlight);
-
 	err = regmap_write(panel->map, NT39016_REG_SYSTEM,
 			   NT39016_SYSTEM_RESET_N);
 	if (err) {
-		dev_err(panel->dev, "Unable to disable panel: %d", err);
+		dev_err(drm_panel->dev, "Unable to disable panel: %d\n", err);
 		return err;
 	}
 
@@ -259,7 +251,6 @@ static int nt39016_probe(struct spi_device *spi)
 	if (!panel)
 		return -ENOMEM;
 
-	panel->dev = dev;
 	spi_set_drvdata(spi, panel);
 
 	panel->panel_info = of_device_get_match_data(dev);
@@ -268,13 +259,13 @@ static int nt39016_probe(struct spi_device *spi)
 
 	panel->supply = devm_regulator_get(dev, "power");
 	if (IS_ERR(panel->supply)) {
-		dev_err(dev, "Failed to get power supply");
+		dev_err(dev, "Failed to get power supply\n");
 		return PTR_ERR(panel->supply);
 	}
 
 	panel->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(panel->reset_gpio)) {
-		dev_err(dev, "Failed to get reset GPIO");
+		dev_err(dev, "Failed to get reset GPIO\n");
 		return PTR_ERR(panel->reset_gpio);
 	}
 
@@ -282,33 +273,28 @@ static int nt39016_probe(struct spi_device *spi)
 	spi->mode = SPI_MODE_3 | SPI_3WIRE;
 	err = spi_setup(spi);
 	if (err) {
-		dev_err(dev, "Failed to setup SPI");
+		dev_err(dev, "Failed to setup SPI\n");
 		return err;
 	}
 
 	panel->map = devm_regmap_init_spi(spi, &nt39016_regmap_config);
 	if (IS_ERR(panel->map)) {
-		dev_err(dev, "Failed to init regmap");
+		dev_err(dev, "Failed to init regmap\n");
 		return PTR_ERR(panel->map);
 	}
 
-	panel->backlight = devm_of_find_backlight(dev);
-	if (IS_ERR(panel->backlight)) {
-		err = PTR_ERR(panel->backlight);
-		if (err != -EPROBE_DEFER)
-			dev_err(dev, "Failed to get backlight handle");
-		return err;
-	}
-
 	drm_panel_init(&panel->drm_panel, dev, &nt39016_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	err = drm_panel_add(&panel->drm_panel);
-	if (err < 0) {
-		dev_err(dev, "Failed to register panel");
+	err = drm_panel_of_backlight(&panel->drm_panel);
+	if (err) {
+		if (err != -EPROBE_DEFER)
+			dev_err(dev, "Failed to get backlight handle\n");
 		return err;
 	}
 
+	drm_panel_add(&panel->drm_panel);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
index ecd76b5391d3..cb5cb27462df 100644
--- a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
+++ b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
@@ -283,7 +283,9 @@ static int lcd_olinuxino_probe(struct i2c_client *client,
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&lcd->panel);
+	drm_panel_add(&lcd->panel);
+
+	return 0;
 }
 
 static int lcd_olinuxino_remove(struct i2c_client *client)
diff --git a/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c b/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
index d956522f32ee..b6e377aa1131 100644
--- a/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
+++ b/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
@@ -17,7 +17,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define OTM8009A_BACKLIGHT_DEFAULT	240
 #define OTM8009A_BACKLIGHT_MAX		255
@@ -97,7 +96,7 @@ static void otm8009a_dcs_write_buf(struct otm8009a *ctx, const void *data,
 	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
 
 	if (mipi_dsi_dcs_write_buffer(dsi, data, len) < 0)
-		DRM_WARN("mipi dsi dcs write buffer failed\n");
+		dev_warn(ctx->dev, "mipi dsi dcs write buffer failed\n");
 }
 
 static void otm8009a_dcs_write_buf_hs(struct otm8009a *ctx, const void *data,
@@ -313,7 +312,7 @@ static int otm8009a_prepare(struct drm_panel *panel)
 
 	ret = regulator_enable(ctx->supply);
 	if (ret < 0) {
-		DRM_ERROR("failed to enable supply: %d\n", ret);
+		dev_err(panel->dev, "failed to enable supply: %d\n", ret);
 		return ret;
 	}
 
@@ -355,9 +354,9 @@ static int otm8009a_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_ERROR("failed to add mode %ux%ux@%u\n",
-			  default_mode.hdisplay, default_mode.vdisplay,
-			  drm_mode_vrefresh(&default_mode));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -390,7 +389,7 @@ static int otm8009a_backlight_update_status(struct backlight_device *bd)
 	u8 data[2];
 
 	if (!ctx->prepared) {
-		DRM_DEBUG("lcd not ready yet for setting its backlight!\n");
+		dev_dbg(&bd->dev, "lcd not ready yet for setting its backlight!\n");
 		return -ENXIO;
 	}
 
diff --git a/drivers/gpu/drm/panel/panel-osd-osd101t2587-53ts.c b/drivers/gpu/drm/panel/panel-osd-osd101t2587-53ts.c
index 83e5aa47f0d6..45b975dee587 100644
--- a/drivers/gpu/drm/panel/panel-osd-osd101t2587-53ts.c
+++ b/drivers/gpu/drm/panel/panel-osd-osd101t2587-53ts.c
@@ -164,7 +164,9 @@ static int osd101t2587_panel_add(struct osd101t2587_panel *osd101t2587)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&osd101t2587->base);
+	drm_panel_add(&osd101t2587->base);
+
+	return 0;
 }
 
 static int osd101t2587_panel_probe(struct mipi_dsi_device *dsi)
diff --git a/drivers/gpu/drm/panel/panel-panasonic-vvx10f034n00.c b/drivers/gpu/drm/panel/panel-panasonic-vvx10f034n00.c
index 627dfcf8adb4..3c20beeb1781 100644
--- a/drivers/gpu/drm/panel/panel-panasonic-vvx10f034n00.c
+++ b/drivers/gpu/drm/panel/panel-panasonic-vvx10f034n00.c
@@ -206,7 +206,9 @@ static int wuxga_nt_panel_add(struct wuxga_nt_panel *wuxga_nt)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&wuxga_nt->base);
+	drm_panel_add(&wuxga_nt->base);
+
+	return 0;
 }
 
 static void wuxga_nt_panel_del(struct wuxga_nt_panel *wuxga_nt)
diff --git a/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c b/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
index e50ee26474cf..5e9ccefb88f6 100644
--- a/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
+++ b/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
@@ -361,7 +361,7 @@ static int rpi_touchscreen_probe(struct i2c_client *i2c,
 	struct rpi_touchscreen *ts;
 	struct device_node *endpoint, *dsi_host_node;
 	struct mipi_dsi_host *host;
-	int ret, ver;
+	int ver;
 	struct mipi_dsi_device_info info = {
 		.type = RPI_DSI_DRIVER_NAME,
 		.channel = 0,
@@ -429,9 +429,7 @@ static int rpi_touchscreen_probe(struct i2c_client *i2c,
 	/* This appears last, as it's what will unblock the DSI host
 	 * driver's component bind function.
 	 */
-	ret = drm_panel_add(&ts->base);
-	if (ret)
-		return ret;
+	drm_panel_add(&ts->base);
 
 	return 0;
 
diff --git a/drivers/gpu/drm/panel/panel-raydium-rm67191.c b/drivers/gpu/drm/panel/panel-raydium-rm67191.c
index 57ff2b1f6361..572547d1aa83 100644
--- a/drivers/gpu/drm/panel/panel-raydium-rm67191.c
+++ b/drivers/gpu/drm/panel/panel-raydium-rm67191.c
@@ -19,7 +19,6 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 /* Panel specific color-format bits */
 #define COL_FMT_16BPP 0x55
@@ -329,7 +328,7 @@ static int rad_panel_enable(struct drm_panel *panel)
 
 	ret = rad_panel_push_cmd_list(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to send MCS (%d)\n", ret);
+		dev_err(dev, "Failed to send MCS (%d)\n", ret);
 		goto fail;
 	}
 
@@ -341,7 +340,7 @@ static int rad_panel_enable(struct drm_panel *panel)
 	/* Software reset */
 	ret = mipi_dsi_dcs_soft_reset(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to do Software Reset (%d)\n", ret);
+		dev_err(dev, "Failed to do Software Reset (%d)\n", ret);
 		goto fail;
 	}
 
@@ -350,33 +349,32 @@ static int rad_panel_enable(struct drm_panel *panel)
 	/* Set DSI mode */
 	ret = mipi_dsi_generic_write(dsi, (u8[]){ 0xC2, 0x0B }, 2);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set DSI mode (%d)\n", ret);
+		dev_err(dev, "Failed to set DSI mode (%d)\n", ret);
 		goto fail;
 	}
 	/* Set tear ON */
 	ret = mipi_dsi_dcs_set_tear_on(dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set tear ON (%d)\n", ret);
+		dev_err(dev, "Failed to set tear ON (%d)\n", ret);
 		goto fail;
 	}
 	/* Set tear scanline */
 	ret = mipi_dsi_dcs_set_tear_scanline(dsi, 0x380);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set tear scanline (%d)\n", ret);
+		dev_err(dev, "Failed to set tear scanline (%d)\n", ret);
 		goto fail;
 	}
 	/* Set pixel format */
 	ret = mipi_dsi_dcs_set_pixel_format(dsi, color_format);
-	DRM_DEV_DEBUG_DRIVER(dev, "Interface color format set to 0x%x\n",
-			     color_format);
+	dev_dbg(dev, "Interface color format set to 0x%x\n", color_format);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set pixel format (%d)\n", ret);
+		dev_err(dev, "Failed to set pixel format (%d)\n", ret);
 		goto fail;
 	}
 	/* Exit sleep mode */
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to exit sleep mode (%d)\n", ret);
+		dev_err(dev, "Failed to exit sleep mode (%d)\n", ret);
 		goto fail;
 	}
 
@@ -384,7 +382,7 @@ static int rad_panel_enable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set display ON (%d)\n", ret);
+		dev_err(dev, "Failed to set display ON (%d)\n", ret);
 		goto fail;
 	}
 
@@ -418,7 +416,7 @@ static int rad_panel_disable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to set display OFF (%d)\n", ret);
+		dev_err(dev, "Failed to set display OFF (%d)\n", ret);
 		return ret;
 	}
 
@@ -426,7 +424,7 @@ static int rad_panel_disable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "Failed to enter sleep mode (%d)\n", ret);
+		dev_err(dev, "Failed to enter sleep mode (%d)\n", ret);
 		return ret;
 	}
 
@@ -442,9 +440,9 @@ static int rad_panel_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(panel->dev, "failed to add mode %ux%ux@%u\n",
-			      default_mode.hdisplay, default_mode.vdisplay,
-			      drm_mode_vrefresh(&default_mode));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -554,8 +552,7 @@ static int rad_panel_probe(struct mipi_dsi_device *dsi)
 	panel->dsi = dsi;
 
 	dsi->format = MIPI_DSI_FMT_RGB888;
-	dsi->mode_flags =  MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO |
-			   MIPI_DSI_CLOCK_NON_CONTINUOUS;
+	dsi->mode_flags =  MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO;
 
 	ret = of_property_read_u32(np, "video-mode", &video_mode);
 	if (!ret) {
@@ -609,9 +606,7 @@ static int rad_panel_probe(struct mipi_dsi_device *dsi)
 		       DRM_MODE_CONNECTOR_DSI);
 	dev_set_drvdata(dev, panel);
 
-	ret = drm_panel_add(&panel->panel);
-	if (ret)
-		return ret;
+	drm_panel_add(&panel->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret)
@@ -628,8 +623,7 @@ static int rad_panel_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret)
-		DRM_DEV_ERROR(dev, "Failed to detach from host (%d)\n",
-			      ret);
+		dev_err(dev, "Failed to detach from host (%d)\n", ret);
 
 	drm_panel_remove(&rad->panel);
 
diff --git a/drivers/gpu/drm/panel/panel-raydium-rm68200.c b/drivers/gpu/drm/panel/panel-raydium-rm68200.c
index 81ae8be62d15..f908eeafb1af 100644
--- a/drivers/gpu/drm/panel/panel-raydium-rm68200.c
+++ b/drivers/gpu/drm/panel/panel-raydium-rm68200.c
@@ -17,7 +17,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 /*** Manufacturer Command Set ***/
 #define MCS_CMD_MODE_SW		0xFE /* CMD Mode Switch */
@@ -110,8 +109,7 @@ static void rm68200_dcs_write_buf(struct rm68200 *ctx, const void *data,
 
 	err = mipi_dsi_dcs_write_buffer(dsi, data, len);
 	if (err < 0)
-		DRM_ERROR_RATELIMITED("MIPI DSI DCS write buffer failed: %d\n",
-				      err);
+		dev_err_ratelimited(ctx->dev, "MIPI DSI DCS write buffer failed: %d\n", err);
 }
 
 static void rm68200_dcs_write_cmd(struct rm68200 *ctx, u8 cmd, u8 value)
@@ -121,7 +119,7 @@ static void rm68200_dcs_write_cmd(struct rm68200 *ctx, u8 cmd, u8 value)
 
 	err = mipi_dsi_dcs_write(dsi, cmd, &value, 1);
 	if (err < 0)
-		DRM_ERROR_RATELIMITED("MIPI DSI DCS write failed: %d\n", err);
+		dev_err_ratelimited(ctx->dev, "MIPI DSI DCS write failed: %d\n", err);
 }
 
 #define dcs_write_seq(ctx, seq...)				\
@@ -256,11 +254,11 @@ static int rm68200_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret)
-		DRM_WARN("failed to set display off: %d\n", ret);
+		dev_warn(panel->dev, "failed to set display off: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret)
-		DRM_WARN("failed to enter sleep mode: %d\n", ret);
+		dev_warn(panel->dev, "failed to enter sleep mode: %d\n", ret);
 
 	msleep(120);
 
@@ -287,7 +285,7 @@ static int rm68200_prepare(struct drm_panel *panel)
 
 	ret = regulator_enable(ctx->supply);
 	if (ret < 0) {
-		DRM_ERROR("failed to enable supply: %d\n", ret);
+		dev_err(ctx->dev, "failed to enable supply: %d\n", ret);
 		return ret;
 	}
 
@@ -336,9 +334,9 @@ static int rm68200_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_ERROR("failed to add mode %ux%ux@%u\n",
-			  default_mode.hdisplay, default_mode.vdisplay,
-			  drm_mode_vrefresh(&default_mode));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
diff --git a/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c b/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c
index a7b0b3e39e1a..535c8d1cca21 100644
--- a/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c
+++ b/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c
@@ -23,7 +23,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct rb070d30_panel {
 	struct drm_panel panel;
@@ -50,7 +49,7 @@ static int rb070d30_panel_prepare(struct drm_panel *panel)
 
 	ret = regulator_enable(ctx->supply);
 	if (ret < 0) {
-		DRM_DEV_ERROR(&ctx->dsi->dev, "Failed to enable supply: %d\n", ret);
+		dev_err(&ctx->dsi->dev, "Failed to enable supply: %d\n", ret);
 		return ret;
 	}
 
@@ -117,9 +116,8 @@ static int rb070d30_panel_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(&ctx->dsi->dev,
-			      "Failed to add mode " DRM_MODE_FMT "\n",
-			      DRM_MODE_ARG(&default_mode));
+		dev_err(&ctx->dsi->dev, "Failed to add mode " DRM_MODE_FMT "\n",
+			DRM_MODE_ARG(&default_mode));
 		return -EINVAL;
 	}
 
@@ -166,13 +164,13 @@ static int rb070d30_panel_dsi_probe(struct mipi_dsi_device *dsi)
 
 	ctx->gpios.reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->gpios.reset)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our reset GPIO\n");
 		return PTR_ERR(ctx->gpios.reset);
 	}
 
 	ctx->gpios.power = devm_gpiod_get(&dsi->dev, "power", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->gpios.power)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our power GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our power GPIO\n");
 		return PTR_ERR(ctx->gpios.power);
 	}
 
@@ -182,7 +180,7 @@ static int rb070d30_panel_dsi_probe(struct mipi_dsi_device *dsi)
 	 */
 	ctx->gpios.updn = devm_gpiod_get(&dsi->dev, "updn", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->gpios.updn)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our updn GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our updn GPIO\n");
 		return PTR_ERR(ctx->gpios.updn);
 	}
 
@@ -192,7 +190,7 @@ static int rb070d30_panel_dsi_probe(struct mipi_dsi_device *dsi)
 	 */
 	ctx->gpios.shlr = devm_gpiod_get(&dsi->dev, "shlr", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->gpios.shlr)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our shlr GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our shlr GPIO\n");
 		return PTR_ERR(ctx->gpios.shlr);
 	}
 
@@ -200,9 +198,7 @@ static int rb070d30_panel_dsi_probe(struct mipi_dsi_device *dsi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_LPM;
 	dsi->format = MIPI_DSI_FMT_RGB888;
diff --git a/drivers/gpu/drm/panel/panel-samsung-ld9040.c b/drivers/gpu/drm/panel/panel-samsung-ld9040.c
index 9bb2e8c7934a..f484147fc3a6 100644
--- a/drivers/gpu/drm/panel/panel-samsung-ld9040.c
+++ b/drivers/gpu/drm/panel/panel-samsung-ld9040.c
@@ -21,7 +21,6 @@
 
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 /* Manufacturer Command Set */
 #define MCS_MANPWR		0xb0
@@ -269,7 +268,7 @@ static int ld9040_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_create(connector->dev);
 	if (!mode) {
-		DRM_ERROR("failed to create a new display mode\n");
+		dev_err(panel->dev, "failed to create a new display mode\n");
 		return 0;
 	}
 
@@ -354,7 +353,9 @@ static int ld9040_probe(struct spi_device *spi)
 	drm_panel_init(&ctx->panel, dev, &ld9040_drm_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&ctx->panel);
+	drm_panel_add(&ctx->panel);
+
+	return 0;
 }
 
 static int ld9040_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c b/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
index f02645d396ac..4aac0d1573dd 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
@@ -7,7 +7,6 @@
 #include <drm/drm_modes.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <linux/gpio/consumer.h>
 #include <linux/regulator/consumer.h>
@@ -55,8 +54,7 @@ static int s6d16d0_unprepare(struct drm_panel *panel)
 	/* Enter sleep mode */
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to enter sleep mode (%d)\n",
-			      ret);
+		dev_err(s6->dev, "failed to enter sleep mode (%d)\n", ret);
 		return ret;
 	}
 
@@ -75,7 +73,7 @@ static int s6d16d0_prepare(struct drm_panel *panel)
 
 	ret = regulator_enable(s6->supply);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to enable supply (%d)\n", ret);
+		dev_err(s6->dev, "failed to enable supply (%d)\n", ret);
 		return ret;
 	}
 
@@ -90,15 +88,13 @@ static int s6d16d0_prepare(struct drm_panel *panel)
 	ret = mipi_dsi_dcs_set_tear_on(dsi,
 				       MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to enable vblank TE (%d)\n",
-			      ret);
+		dev_err(s6->dev, "failed to enable vblank TE (%d)\n", ret);
 		return ret;
 	}
 	/* Exit sleep mode and power on */
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to exit sleep mode (%d)\n",
-			      ret);
+		dev_err(s6->dev, "failed to exit sleep mode (%d)\n", ret);
 		return ret;
 	}
 
@@ -113,8 +109,7 @@ static int s6d16d0_enable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to turn display on (%d)\n",
-			      ret);
+		dev_err(s6->dev, "failed to turn display on (%d)\n", ret);
 		return ret;
 	}
 
@@ -129,8 +124,7 @@ static int s6d16d0_disable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(s6->dev, "failed to turn display off (%d)\n",
-			      ret);
+		dev_err(s6->dev, "failed to turn display off (%d)\n", ret);
 		return ret;
 	}
 
@@ -144,7 +138,7 @@ static int s6d16d0_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &samsung_s6d16d0_mode);
 	if (!mode) {
-		DRM_ERROR("bad mode or failed to add mode\n");
+		dev_err(panel->dev, "bad mode or failed to add mode\n");
 		return -EINVAL;
 	}
 	drm_mode_set_name(mode);
@@ -204,17 +198,14 @@ static int s6d16d0_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(s6->reset_gpio)) {
 		ret = PTR_ERR(s6->reset_gpio);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to request GPIO (%d)\n",
-				      ret);
+			dev_err(dev, "failed to request GPIO (%d)\n", ret);
 		return ret;
 	}
 
 	drm_panel_init(&s6->panel, dev, &s6d16d0_drm_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_add(&s6->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&s6->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0)
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e3ha2.c b/drivers/gpu/drm/panel/panel-samsung-s6e3ha2.c
index 80ef122e7466..1d1c79a18613 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e3ha2.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e3ha2.c
@@ -18,7 +18,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define S6E3HA2_MIN_BRIGHTNESS		0
 #define S6E3HA2_MAX_BRIGHTNESS		100
@@ -651,7 +650,7 @@ static int s6e3ha2_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, ctx->desc->mode);
 	if (!mode) {
-		DRM_ERROR("failed to add mode %ux%ux@%u\n",
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
 			ctx->desc->mode->hdisplay, ctx->desc->mode->vdisplay,
 			drm_mode_vrefresh(ctx->desc->mode));
 		return -ENOMEM;
@@ -733,9 +732,7 @@ static int s6e3ha2_probe(struct mipi_dsi_device *dsi)
 	drm_panel_init(&ctx->panel, dev, &s6e3ha2_drm_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		goto unregister_backlight;
+	drm_panel_add(&ctx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0)
@@ -745,8 +742,6 @@ static int s6e3ha2_probe(struct mipi_dsi_device *dsi)
 
 remove_panel:
 	drm_panel_remove(&ctx->panel);
-
-unregister_backlight:
 	backlight_device_unregister(ctx->bl_dev);
 
 	return ret;
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e63j0x03.c b/drivers/gpu/drm/panel/panel-samsung-s6e63j0x03.c
index 1247656d73bf..b962c817fb30 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e63j0x03.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e63j0x03.c
@@ -19,7 +19,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define MCS_LEVEL2_KEY		0xf0
 #define MCS_MTP_KEY		0xf1
@@ -406,7 +405,7 @@ static int s6e63j0x03_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_ERROR("failed to add mode %ux%ux@%u\n",
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
 			default_mode.hdisplay, default_mode.vdisplay,
 			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
@@ -479,9 +478,7 @@ static int s6e63j0x03_probe(struct mipi_dsi_device *dsi)
 	ctx->bl_dev->props.brightness = DEFAULT_BRIGHTNESS;
 	ctx->bl_dev->props.power = FB_BLANK_POWERDOWN;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		goto unregister_backlight;
+	drm_panel_add(&ctx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0)
@@ -491,8 +488,6 @@ static int s6e63j0x03_probe(struct mipi_dsi_device *dsi)
 
 remove_panel:
 	drm_panel_remove(&ctx->panel);
-
-unregister_backlight:
 	backlight_device_unregister(ctx->bl_dev);
 
 	return ret;
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e63m0-dsi.c b/drivers/gpu/drm/panel/panel-samsung-s6e63m0-dsi.c
new file mode 100644
index 000000000000..eec74c10ddda
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e63m0-dsi.c
@@ -0,0 +1,139 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DSI interface to the Samsung S6E63M0 panel.
+ * (C) 2019 Linus Walleij
+ */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/of_device.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_print.h>
+
+#include "panel-samsung-s6e63m0.h"
+
+#define MCS_GLOBAL_PARAM	0xb0
+#define S6E63M0_DSI_MAX_CHUNK	15 /* CMD + 15 bytes max */
+
+static int s6e63m0_dsi_dcs_read(struct device *dev, const u8 cmd, u8 *data)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+	int ret;
+
+	ret = mipi_dsi_dcs_read(dsi, cmd, data, 1);
+	if (ret < 0) {
+		dev_err(dev, "could not read DCS CMD %02x\n", cmd);
+		return ret;
+	}
+
+	dev_info(dev, "DSI read CMD %02x = %02x\n", cmd, *data);
+
+	return 0;
+}
+
+static int s6e63m0_dsi_dcs_write(struct device *dev, const u8 *data, size_t len)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+	const u8 *seqp = data;
+	u8 cmd;
+	u8 cmdwritten;
+	int remain;
+	int chunk;
+	int ret;
+
+	dev_info(dev, "DSI writing dcs seq: %*ph\n", (int)len, data);
+
+	/* Pick out and skip past the DCS command */
+	cmd = *seqp;
+	seqp++;
+	cmdwritten = 0;
+	remain = len - 1;
+	chunk = remain;
+
+	/* Send max S6E63M0_DSI_MAX_CHUNK bytes at a time */
+	if (chunk > S6E63M0_DSI_MAX_CHUNK)
+		chunk = S6E63M0_DSI_MAX_CHUNK;
+	ret = mipi_dsi_dcs_write(dsi, cmd, seqp, chunk);
+	if (ret < 0) {
+		dev_err(dev, "error sending DCS command seq cmd %02x\n", cmd);
+		return ret;
+	}
+	cmdwritten += chunk;
+	seqp += chunk;
+
+	while (cmdwritten < remain) {
+		chunk = remain - cmdwritten;
+		if (chunk > S6E63M0_DSI_MAX_CHUNK)
+			chunk = S6E63M0_DSI_MAX_CHUNK;
+		ret = mipi_dsi_dcs_write(dsi, MCS_GLOBAL_PARAM, &cmdwritten, 1);
+		if (ret < 0) {
+			dev_err(dev, "error sending CMD %02x global param %02x\n",
+				cmd, cmdwritten);
+			return ret;
+		}
+		ret = mipi_dsi_dcs_write(dsi, cmd, seqp, chunk);
+		if (ret < 0) {
+			dev_err(dev, "error sending CMD %02x chunk\n", cmd);
+			return ret;
+		}
+		cmdwritten += chunk;
+		seqp += chunk;
+	}
+	dev_info(dev, "sent command %02x %02x bytes\n", cmd, cmdwritten);
+
+	usleep_range(8000, 9000);
+
+	return 0;
+}
+
+static int s6e63m0_dsi_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	dsi->lanes = 2;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->hs_rate = 349440000;
+	dsi->lp_rate = 9600000;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO |
+		MIPI_DSI_MODE_EOT_PACKET |
+		MIPI_DSI_MODE_VIDEO_BURST;
+
+	ret = s6e63m0_probe(dev, s6e63m0_dsi_dcs_read, s6e63m0_dsi_dcs_write,
+			    true);
+	if (ret)
+		return ret;
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0)
+		s6e63m0_remove(dev);
+
+	return ret;
+}
+
+static int s6e63m0_dsi_remove(struct mipi_dsi_device *dsi)
+{
+	mipi_dsi_detach(dsi);
+	return s6e63m0_remove(&dsi->dev);
+}
+
+static const struct of_device_id s6e63m0_dsi_of_match[] = {
+	{ .compatible = "samsung,s6e63m0" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, s6e63m0_dsi_of_match);
+
+static struct mipi_dsi_driver s6e63m0_dsi_driver = {
+	.probe			= s6e63m0_dsi_probe,
+	.remove			= s6e63m0_dsi_remove,
+	.driver			= {
+		.name		= "panel-samsung-s6e63m0",
+		.of_match_table = s6e63m0_dsi_of_match,
+	},
+};
+module_mipi_dsi_driver(s6e63m0_dsi_driver);
+
+MODULE_AUTHOR("Linus Walleij <linusw@kernel.org>");
+MODULE_DESCRIPTION("s6e63m0 LCD DSI Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e63m0-spi.c b/drivers/gpu/drm/panel/panel-samsung-s6e63m0-spi.c
new file mode 100644
index 000000000000..d298d780220d
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e63m0-spi.c
@@ -0,0 +1,101 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/spi/spi.h>
+#include <linux/delay.h>
+
+#include <drm/drm_print.h>
+
+#include "panel-samsung-s6e63m0.h"
+
+#define DATA_MASK	0x100
+
+static int s6e63m0_spi_dcs_read(struct device *dev, const u8 cmd, u8 *data)
+{
+	/*
+	 * FIXME: implement reading DCS commands over SPI so we can
+	 * properly identify which physical panel is connected.
+	 */
+	*data = 0;
+
+	return 0;
+}
+
+static int s6e63m0_spi_write_word(struct device *dev, u16 data)
+{
+	struct spi_device *spi = to_spi_device(dev);
+	struct spi_transfer xfer = {
+		.len	= 2,
+		.tx_buf = &data,
+	};
+	struct spi_message msg;
+
+	spi_message_init(&msg);
+	spi_message_add_tail(&xfer, &msg);
+
+	return spi_sync(spi, &msg);
+}
+
+static int s6e63m0_spi_dcs_write(struct device *dev, const u8 *data, size_t len)
+{
+	int ret = 0;
+
+	dev_dbg(dev, "SPI writing dcs seq: %*ph\n", (int)len, data);
+	ret = s6e63m0_spi_write_word(dev, *data);
+
+	while (!ret && --len) {
+		++data;
+		ret = s6e63m0_spi_write_word(dev, *data | DATA_MASK);
+	}
+
+	if (ret) {
+		dev_err(dev, "SPI error %d writing dcs seq: %*ph\n", ret,
+			(int)len, data);
+	}
+
+	usleep_range(300, 310);
+
+	return ret;
+}
+
+static int s6e63m0_spi_probe(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	int ret;
+
+	spi->bits_per_word = 9;
+	spi->mode = SPI_MODE_3;
+	ret = spi_setup(spi);
+	if (ret < 0) {
+		dev_err(dev, "spi setup failed.\n");
+		return ret;
+	}
+	return s6e63m0_probe(dev, s6e63m0_spi_dcs_read, s6e63m0_spi_dcs_write,
+			     false);
+}
+
+static int s6e63m0_spi_remove(struct spi_device *spi)
+{
+	return s6e63m0_remove(&spi->dev);
+}
+
+static const struct of_device_id s6e63m0_spi_of_match[] = {
+	{ .compatible = "samsung,s6e63m0" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, s6e63m0_spi_of_match);
+
+static struct spi_driver s6e63m0_spi_driver = {
+	.probe			= s6e63m0_spi_probe,
+	.remove			= s6e63m0_spi_remove,
+	.driver			= {
+		.name		= "panel-samsung-s6e63m0",
+		.of_match_table = s6e63m0_spi_of_match,
+	},
+};
+module_spi_driver(s6e63m0_spi_driver);
+
+MODULE_AUTHOR("Paweł Chmiel <pawel.mikolaj.chmiel@gmail.com>");
+MODULE_DESCRIPTION("s6e63m0 LCD SPI Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e63m0.c b/drivers/gpu/drm/panel/panel-samsung-s6e63m0.c
index 64421347bfd4..3eee67e2d86a 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e63m0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e63m0.c
@@ -10,32 +10,40 @@
 
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <linux/backlight.h>
 #include <linux/delay.h>
 #include <linux/gpio/consumer.h>
 #include <linux/module.h>
 #include <linux/regulator/consumer.h>
-#include <linux/spi/spi.h>
 
 #include <video/mipi_display.h>
 
+#include "panel-samsung-s6e63m0.h"
+
 /* Manufacturer Command Set */
 #define MCS_ELVSS_ON                0xb1
 #define MCS_MIECTL1                0xc0
 #define MCS_BCMODE                              0xc1
+#define MCS_ERROR_CHECK		0xd5
+#define MCS_READ_ID1		0xda
+#define MCS_READ_ID2		0xdb
+#define MCS_READ_ID3		0xdc
+#define MCS_LEVEL_2_KEY		0xf0
+#define MCS_MTP_KEY		0xf1
 #define MCS_DISCTL   0xf2
 #define MCS_SRCCTL           0xf6
 #define MCS_IFCTL                       0xf7
 #define MCS_PANELCTL         0xF8
 #define MCS_PGAMMACTL                   0xfa
 
+#define S6E63M0_LCD_ID_VALUE_M2		0xA4
+#define S6E63M0_LCD_ID_VALUE_SM2	0xB4
+#define S6E63M0_LCD_ID_VALUE_SM2_1	0xB6
+
 #define NUM_GAMMA_LEVELS             11
 #define GAMMA_TABLE_COUNT           23
 
-#define DATA_MASK                                       0x100
-
 #define MAX_BRIGHTNESS              (NUM_GAMMA_LEVELS - 1)
 
 /* array of gamma tables for gamma value 2.2 */
@@ -88,8 +96,11 @@ static u8 const s6e63m0_gamma_22[NUM_GAMMA_LEVELS][GAMMA_TABLE_COUNT] = {
 
 struct s6e63m0 {
 	struct device *dev;
+	int (*dcs_read)(struct device *dev, const u8 cmd, u8 *val);
+	int (*dcs_write)(struct device *dev, const u8 *data, size_t len);
 	struct drm_panel panel;
 	struct backlight_device *bl_dev;
+	u8 lcd_type;
 
 	struct regulator_bulk_data supplies[2];
 	struct gpio_desc *reset_gpio;
@@ -135,43 +146,20 @@ static int s6e63m0_clear_error(struct s6e63m0 *ctx)
 	return ret;
 }
 
-static int s6e63m0_spi_write_word(struct s6e63m0 *ctx, u16 data)
+static void s6e63m0_dcs_read(struct s6e63m0 *ctx, const u8 cmd, u8 *data)
 {
-	struct spi_device *spi = to_spi_device(ctx->dev);
-	struct spi_transfer xfer = {
-		.len	= 2,
-		.tx_buf = &data,
-	};
-	struct spi_message msg;
-
-	spi_message_init(&msg);
-	spi_message_add_tail(&xfer, &msg);
+	if (ctx->error < 0)
+		return;
 
-	return spi_sync(spi, &msg);
+	ctx->error = ctx->dcs_read(ctx->dev, cmd, data);
 }
 
 static void s6e63m0_dcs_write(struct s6e63m0 *ctx, const u8 *data, size_t len)
 {
-	int ret = 0;
-
 	if (ctx->error < 0 || len == 0)
 		return;
 
-	DRM_DEV_DEBUG(ctx->dev, "writing dcs seq: %*ph\n", (int)len, data);
-	ret = s6e63m0_spi_write_word(ctx, *data);
-
-	while (!ret && --len) {
-		++data;
-		ret = s6e63m0_spi_write_word(ctx, *data | DATA_MASK);
-	}
-
-	if (ret) {
-		DRM_DEV_ERROR(ctx->dev, "error %d writing dcs seq: %*ph\n", ret,
-			      (int)len, data);
-		ctx->error = ret;
-	}
-
-	usleep_range(300, 310);
+	ctx->error = ctx->dcs_write(ctx->dev, data, len);
 }
 
 #define s6e63m0_dcs_write_seq_static(ctx, seq ...) \
@@ -180,6 +168,43 @@ static void s6e63m0_dcs_write(struct s6e63m0 *ctx, const u8 *data, size_t len)
 		s6e63m0_dcs_write(ctx, d, ARRAY_SIZE(d)); \
 	})
 
+static int s6e63m0_check_lcd_type(struct s6e63m0 *ctx)
+{
+	u8 id1, id2, id3;
+	int ret;
+
+	s6e63m0_dcs_read(ctx, MCS_READ_ID1, &id1);
+	s6e63m0_dcs_read(ctx, MCS_READ_ID2, &id2);
+	s6e63m0_dcs_read(ctx, MCS_READ_ID3, &id3);
+
+	ret = s6e63m0_clear_error(ctx);
+	if (ret) {
+		dev_err(ctx->dev, "error checking LCD type (%d)\n", ret);
+		ctx->lcd_type = 0x00;
+		return ret;
+	}
+
+	dev_info(ctx->dev, "MTP ID: %02x %02x %02x\n", id1, id2, id3);
+
+	/* We attempt to detect what panel is mounted on the controller */
+	switch (id2) {
+	case S6E63M0_LCD_ID_VALUE_M2:
+		dev_info(ctx->dev, "detected LCD panel AMS397GE MIPI M2\n");
+		break;
+	case S6E63M0_LCD_ID_VALUE_SM2:
+	case S6E63M0_LCD_ID_VALUE_SM2_1:
+		dev_info(ctx->dev, "detected LCD panel AMS397GE MIPI SM2\n");
+		break;
+	default:
+		dev_info(ctx->dev, "unknown LCD panel type %02x\n", id2);
+		break;
+	}
+
+	ctx->lcd_type = id2;
+
+	return 0;
+}
+
 static void s6e63m0_init(struct s6e63m0 *ctx)
 {
 	s6e63m0_dcs_write_seq_static(ctx, MCS_PANELCTL,
@@ -251,8 +276,6 @@ static void s6e63m0_init(struct s6e63m0 *ctx)
 
 	s6e63m0_dcs_write_seq_static(ctx, MCS_ELVSS_ON,
 				     0x0b);
-
-	s6e63m0_dcs_write_seq_static(ctx, MIPI_DCS_EXIT_SLEEP_MODE);
 }
 
 static int s6e63m0_power_on(struct s6e63m0 *ctx)
@@ -265,6 +288,9 @@ static int s6e63m0_power_on(struct s6e63m0 *ctx)
 
 	msleep(25);
 
+	/* Be sure to send a reset pulse */
+	gpiod_set_value(ctx->reset_gpio, 1);
+	msleep(5);
 	gpiod_set_value(ctx->reset_gpio, 0);
 	msleep(120);
 
@@ -294,8 +320,10 @@ static int s6e63m0_disable(struct drm_panel *panel)
 
 	backlight_disable(ctx->bl_dev);
 
+	s6e63m0_dcs_write_seq_static(ctx, MIPI_DCS_SET_DISPLAY_OFF);
+	msleep(10);
 	s6e63m0_dcs_write_seq_static(ctx, MIPI_DCS_ENTER_SLEEP_MODE);
-	msleep(200);
+	msleep(120);
 
 	ctx->enabled = false;
 
@@ -333,6 +361,15 @@ static int s6e63m0_prepare(struct drm_panel *panel)
 	if (ret < 0)
 		return ret;
 
+	/* Magic to unlock level 2 control of the display */
+	s6e63m0_dcs_write_seq_static(ctx, MCS_LEVEL_2_KEY, 0x5a, 0x5a);
+	/* Magic to unlock MTP reading */
+	s6e63m0_dcs_write_seq_static(ctx, MCS_MTP_KEY, 0x5a, 0x5a);
+
+	ret = s6e63m0_check_lcd_type(ctx);
+	if (ret < 0)
+		return ret;
+
 	s6e63m0_init(ctx);
 
 	ret = s6e63m0_clear_error(ctx);
@@ -352,7 +389,15 @@ static int s6e63m0_enable(struct drm_panel *panel)
 	if (ctx->enabled)
 		return 0;
 
+	s6e63m0_dcs_write_seq_static(ctx, MIPI_DCS_EXIT_SLEEP_MODE);
+	msleep(120);
 	s6e63m0_dcs_write_seq_static(ctx, MIPI_DCS_SET_DISPLAY_ON);
+	msleep(10);
+
+	s6e63m0_dcs_write_seq_static(ctx, MCS_ERROR_CHECK,
+				     0xE7, 0x14, 0x60, 0x17, 0x0A, 0x49, 0xC3,
+				     0x8F, 0x19, 0x64, 0x91, 0x84, 0x76, 0x20,
+				     0x0F, 0x00);
 
 	backlight_enable(ctx->bl_dev);
 
@@ -368,9 +413,9 @@ static int s6e63m0_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_ERROR("failed to add mode %ux%ux@%u\n",
-			  default_mode.hdisplay, default_mode.vdisplay,
-			  drm_mode_vrefresh(&default_mode));
+		dev_err(panel->dev, "failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -425,16 +470,17 @@ static int s6e63m0_backlight_register(struct s6e63m0 *ctx)
 						     &props);
 	if (IS_ERR(ctx->bl_dev)) {
 		ret = PTR_ERR(ctx->bl_dev);
-		DRM_DEV_ERROR(dev, "error registering backlight device (%d)\n",
-			      ret);
+		dev_err(dev, "error registering backlight device (%d)\n", ret);
 	}
 
 	return ret;
 }
 
-static int s6e63m0_probe(struct spi_device *spi)
+int s6e63m0_probe(struct device *dev,
+		  int (*dcs_read)(struct device *dev, const u8 cmd, u8 *val),
+		  int (*dcs_write)(struct device *dev, const u8 *data, size_t len),
+		  bool dsi_mode)
 {
-	struct device *dev = &spi->dev;
 	struct s6e63m0 *ctx;
 	int ret;
 
@@ -442,7 +488,9 @@ static int s6e63m0_probe(struct spi_device *spi)
 	if (!ctx)
 		return -ENOMEM;
 
-	spi_set_drvdata(spi, ctx);
+	ctx->dcs_read = dcs_read;
+	ctx->dcs_write = dcs_write;
+	dev_set_drvdata(dev, ctx);
 
 	ctx->dev = dev;
 	ctx->enabled = false;
@@ -453,59 +501,39 @@ static int s6e63m0_probe(struct spi_device *spi)
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(ctx->supplies),
 				      ctx->supplies);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "failed to get regulators: %d\n", ret);
+		dev_err(dev, "failed to get regulators: %d\n", ret);
 		return ret;
 	}
 
 	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset-gpios %ld\n",
-			      PTR_ERR(ctx->reset_gpio));
+		dev_err(dev, "cannot get reset-gpios %ld\n", PTR_ERR(ctx->reset_gpio));
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
-	spi->bits_per_word = 9;
-	spi->mode = SPI_MODE_3;
-	ret = spi_setup(spi);
-	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "spi setup failed.\n");
-		return ret;
-	}
-
 	drm_panel_init(&ctx->panel, dev, &s6e63m0_drm_funcs,
+		       dsi_mode ? DRM_MODE_CONNECTOR_DSI :
 		       DRM_MODE_CONNECTOR_DPI);
 
 	ret = s6e63m0_backlight_register(ctx);
 	if (ret < 0)
 		return ret;
 
-	return drm_panel_add(&ctx->panel);
+	drm_panel_add(&ctx->panel);
+
+	return 0;
 }
+EXPORT_SYMBOL_GPL(s6e63m0_probe);
 
-static int s6e63m0_remove(struct spi_device *spi)
+int s6e63m0_remove(struct device *dev)
 {
-	struct s6e63m0 *ctx = spi_get_drvdata(spi);
+	struct s6e63m0 *ctx = dev_get_drvdata(dev);
 
 	drm_panel_remove(&ctx->panel);
 
 	return 0;
 }
-
-static const struct of_device_id s6e63m0_of_match[] = {
-	{ .compatible = "samsung,s6e63m0" },
-	{ /* sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, s6e63m0_of_match);
-
-static struct spi_driver s6e63m0_driver = {
-	.probe			= s6e63m0_probe,
-	.remove			= s6e63m0_remove,
-	.driver			= {
-		.name		= "panel-samsung-s6e63m0",
-		.of_match_table = s6e63m0_of_match,
-	},
-};
-module_spi_driver(s6e63m0_driver);
+EXPORT_SYMBOL_GPL(s6e63m0_remove);
 
 MODULE_AUTHOR("Paweł Chmiel <pawel.mikolaj.chmiel@gmail.com>");
 MODULE_DESCRIPTION("s6e63m0 LCD Driver");
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e63m0.h b/drivers/gpu/drm/panel/panel-samsung-s6e63m0.h
new file mode 100644
index 000000000000..c669fec91763
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e63m0.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _PANEL_SAMSUNG_S6E63M0_H
+#define _PANEL_SAMSUNG_S6E63M0_H
+
+int s6e63m0_probe(struct device *dev,
+		  int (*dcs_read)(struct device *dev, const u8 cmd, u8 *val),
+		  int (*dcs_write)(struct device *dev, const u8 *data,
+				   size_t len),
+		  bool dsi_mode);
+int s6e63m0_remove(struct device *dev);
+
+#endif /* _PANEL_SAMSUNG_S6E63M0_H */
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c
index 485eabecfcc9..ea63799ff2a1 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c
@@ -242,11 +242,7 @@ static int s6e88a0_ams452ef01_probe(struct mipi_dsi_device *dsi)
 	drm_panel_init(&ctx->panel, dev, &s6e88a0_ams452ef01_panel_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0) {
-		dev_err(dev, "Failed to add panel: %d\n", ret);
-		return ret;
-	}
+	drm_panel_add(&ctx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c
index 8a028d2bd0d6..527371120266 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c
@@ -25,7 +25,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define LDI_MTP_LENGTH			24
 #define GAMMA_LEVEL_NUM			25
@@ -928,7 +927,7 @@ static int s6e8aa0_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_create(connector->dev);
 	if (!mode) {
-		DRM_ERROR("failed to create a new display mode\n");
+		dev_err(panel->dev, "failed to create a new display mode\n");
 		return 0;
 	}
 
@@ -1020,9 +1019,7 @@ static int s6e8aa0_probe(struct mipi_dsi_device *dsi)
 	drm_panel_init(&ctx->panel, dev, &s6e8aa0_drm_funcs,
 		       DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0)
diff --git a/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c b/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
index e417dc4921c2..0ee508576231 100644
--- a/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
+++ b/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
@@ -258,9 +258,7 @@ static int seiko_panel_probe(struct device *dev,
 	if (err)
 		return err;
 
-	err = drm_panel_add(&panel->base);
-	if (err < 0)
-		return err;
+	drm_panel_add(&panel->base);
 
 	dev_set_drvdata(dev, panel);
 
diff --git a/drivers/gpu/drm/panel/panel-sharp-lq101r1sx01.c b/drivers/gpu/drm/panel/panel-sharp-lq101r1sx01.c
index f07324b705b3..f8cd2a42ed13 100644
--- a/drivers/gpu/drm/panel/panel-sharp-lq101r1sx01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-lq101r1sx01.c
@@ -325,7 +325,9 @@ static int sharp_panel_add(struct sharp_panel *sharp)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&sharp->base);
+	drm_panel_add(&sharp->base);
+
+	return 0;
 }
 
 static void sharp_panel_del(struct sharp_panel *sharp)
diff --git a/drivers/gpu/drm/panel/panel-sharp-ls037v7dw01.c b/drivers/gpu/drm/panel/panel-sharp-ls037v7dw01.c
index d7bf13b9e1d6..94992f45113a 100644
--- a/drivers/gpu/drm/panel/panel-sharp-ls037v7dw01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-ls037v7dw01.c
@@ -187,7 +187,9 @@ static int ls037v7dw01_probe(struct platform_device *pdev)
 	drm_panel_init(&lcd->panel, &pdev->dev, &ls037v7dw01_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	return drm_panel_add(&lcd->panel);
+	drm_panel_add(&lcd->panel);
+
+	return 0;
 }
 
 static int ls037v7dw01_remove(struct platform_device *pdev)
diff --git a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
index b2e58935529c..16dbf0f353ed 100644
--- a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
@@ -261,7 +261,9 @@ static int sharp_nt_panel_add(struct sharp_nt_panel *sharp_nt)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&sharp_nt->base);
+	drm_panel_add(&sharp_nt->base);
+
+	return 0;
 }
 
 static void sharp_nt_panel_del(struct sharp_nt_panel *sharp_nt)
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index cb6550d37e85..2be358fb46f7 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -112,6 +112,8 @@ struct panel_simple {
 	struct gpio_desc *hpd_gpio;
 
 	struct drm_display_mode override_mode;
+
+	enum drm_panel_orientation orientation;
 };
 
 static inline struct panel_simple *to_panel_simple(struct drm_panel *panel)
@@ -371,6 +373,9 @@ static int panel_simple_get_modes(struct drm_panel *panel,
 	/* add hard-coded panel modes */
 	num += panel_simple_get_non_edid_modes(p, connector);
 
+	/* set up connector's "panel orientation" property */
+	drm_connector_set_panel_orientation(connector, p->orientation);
+
 	return num;
 }
 
@@ -500,6 +505,8 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 	struct panel_simple *panel;
 	struct display_timing dt;
 	struct device_node *ddc;
+	int connector_type;
+	u32 bus_flags;
 	int err;
 
 	panel = devm_kzalloc(dev, sizeof(*panel), GFP_KERNEL);
@@ -530,6 +537,12 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 		return err;
 	}
 
+	err = of_drm_get_panel_orientation(dev->of_node, &panel->orientation);
+	if (err) {
+		dev_err(dev, "%pOF: failed to get orientation %d\n", dev->of_node, err);
+		return err;
+	}
+
 	ddc = of_parse_phandle(dev->of_node, "ddc-i2c-bus", 0);
 	if (ddc) {
 		panel->ddc = of_find_i2c_adapter_by_node(ddc);
@@ -549,8 +562,14 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 			panel_simple_parse_panel_timing_node(dev, panel, &dt);
 	}
 
-	if (desc->connector_type == DRM_MODE_CONNECTOR_LVDS) {
-		/* Catch common mistakes for LVDS panels. */
+	connector_type = desc->connector_type;
+	/* Catch common mistakes for panels. */
+	switch (connector_type) {
+	case 0:
+		dev_warn(dev, "Specify missing connector_type\n");
+		connector_type = DRM_MODE_CONNECTOR_DPI;
+		break;
+	case DRM_MODE_CONNECTOR_LVDS:
 		WARN_ON(desc->bus_flags &
 			~(DRM_BUS_FLAG_DE_LOW |
 			  DRM_BUS_FLAG_DE_HIGH |
@@ -564,18 +583,48 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 		WARN_ON((desc->bus_format == MEDIA_BUS_FMT_RGB888_1X7X4_SPWG ||
 			 desc->bus_format == MEDIA_BUS_FMT_RGB888_1X7X4_JEIDA) &&
 			desc->bpc != 8);
+		break;
+	case DRM_MODE_CONNECTOR_eDP:
+		if (desc->bus_format == 0)
+			dev_warn(dev, "Specify missing bus_format\n");
+		if (desc->bpc != 6 && desc->bpc != 8)
+			dev_warn(dev, "Expected bpc in {6,8} but got: %u\n", desc->bpc);
+		break;
+	case DRM_MODE_CONNECTOR_DSI:
+		if (desc->bpc != 6 && desc->bpc != 8)
+			dev_warn(dev, "Expected bpc in {6,8} but got: %u\n", desc->bpc);
+		break;
+	case DRM_MODE_CONNECTOR_DPI:
+		bus_flags = DRM_BUS_FLAG_DE_LOW |
+			    DRM_BUS_FLAG_DE_HIGH |
+			    DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE |
+			    DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE |
+			    DRM_BUS_FLAG_DATA_MSB_TO_LSB |
+			    DRM_BUS_FLAG_DATA_LSB_TO_MSB |
+			    DRM_BUS_FLAG_SYNC_SAMPLE_POSEDGE |
+			    DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE;
+		if (desc->bus_flags & ~bus_flags)
+			dev_warn(dev, "Unexpected bus_flags(%d)\n", desc->bus_flags & ~bus_flags);
+		if (!(desc->bus_flags & bus_flags))
+			dev_warn(dev, "Specify missing bus_flags\n");
+		if (desc->bus_format == 0)
+			dev_warn(dev, "Specify missing bus_format\n");
+		if (desc->bpc != 6 && desc->bpc != 8)
+			dev_warn(dev, "Expected bpc in {6,8} but got: %u\n", desc->bpc);
+		break;
+	default:
+		dev_warn(dev, "Specify a valid connector_type: %d\n", desc->connector_type);
+		connector_type = DRM_MODE_CONNECTOR_DPI;
+		break;
 	}
 
-	drm_panel_init(&panel->base, dev, &panel_simple_funcs,
-		       desc->connector_type);
+	drm_panel_init(&panel->base, dev, &panel_simple_funcs, connector_type);
 
 	err = drm_panel_of_backlight(&panel->base);
 	if (err)
 		goto free_ddc;
 
-	err = drm_panel_add(&panel->base);
-	if (err < 0)
-		goto free_ddc;
+	drm_panel_add(&panel->base);
 
 	dev_set_drvdata(dev, panel);
 
@@ -610,6 +659,32 @@ static void panel_simple_shutdown(struct device *dev)
 	drm_panel_unprepare(&panel->base);
 }
 
+static const struct drm_display_mode ampire_am_1280800n3tzqw_t00h_mode = {
+	.clock = 71100,
+	.hdisplay = 1280,
+	.hsync_start = 1280 + 40,
+	.hsync_end = 1280 + 40 + 80,
+	.htotal = 1280 + 40 + 80 + 40,
+	.vdisplay = 800,
+	.vsync_start = 800 + 3,
+	.vsync_end = 800 + 3 + 10,
+	.vtotal = 800 + 3 + 10 + 10,
+	.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC,
+};
+
+static const struct panel_desc ampire_am_1280800n3tzqw_t00h = {
+	.modes = &ampire_am_1280800n3tzqw_t00h_mode,
+	.num_modes = 1,
+	.bpc = 6,
+	.size = {
+		.width = 217,
+		.height = 136,
+	},
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct drm_display_mode ampire_am_480272h3tmqw_t01h_mode = {
 	.clock = 9000,
 	.hdisplay = 480,
@@ -1191,10 +1266,14 @@ static const struct drm_display_mode boe_hv070wsa_mode = {
 static const struct panel_desc boe_hv070wsa = {
 	.modes = &boe_hv070wsa_mode,
 	.num_modes = 1,
+	.bpc = 8,
 	.size = {
 		.width = 154,
 		.height = 90,
 	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
 };
 
 static const struct drm_display_mode boe_nv101wxmn51_modes[] = {
@@ -1414,6 +1493,36 @@ static const struct panel_desc cdtech_s070wv95_ct16 = {
 	},
 };
 
+static const struct display_timing chefree_ch101olhlwh_002_timing = {
+	.pixelclock = { 68900000, 71100000, 73400000 },
+	.hactive = { 1280, 1280, 1280 },
+	.hfront_porch = { 65, 80, 95 },
+	.hback_porch = { 64, 79, 94 },
+	.hsync_len = { 1, 1, 1 },
+	.vactive = { 800, 800, 800 },
+	.vfront_porch = { 7, 11, 14 },
+	.vback_porch = { 7, 11, 14 },
+	.vsync_len = { 1, 1, 1 },
+	.flags = DISPLAY_FLAGS_DE_HIGH,
+};
+
+static const struct panel_desc chefree_ch101olhlwh_002 = {
+	.timings = &chefree_ch101olhlwh_002_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 217,
+		.height = 135,
+	},
+	.delay = {
+		.enable = 200,
+		.disable = 200,
+	},
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct drm_display_mode chunghwa_claa070wp03xg_mode = {
 	.clock = 66770,
 	.hdisplay = 800,
@@ -2258,6 +2367,34 @@ static const struct panel_desc ivo_m133nwf4_r0 = {
 	.connector_type = DRM_MODE_CONNECTOR_eDP,
 };
 
+static const struct drm_display_mode kingdisplay_kd116n21_30nv_a010_mode = {
+	.clock = 81000,
+	.hdisplay = 1366,
+	.hsync_start = 1366 + 40,
+	.hsync_end = 1366 + 40 + 32,
+	.htotal = 1366 + 40 + 32 + 62,
+	.vdisplay = 768,
+	.vsync_start = 768 + 5,
+	.vsync_end = 768 + 5 + 5,
+	.vtotal = 768 + 5 + 5 + 122,
+	.flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC,
+};
+
+static const struct panel_desc kingdisplay_kd116n21_30nv_a010 = {
+	.modes = &kingdisplay_kd116n21_30nv_a010_mode,
+	.num_modes = 1,
+	.bpc = 6,
+	.size = {
+		.width = 256,
+		.height = 144,
+	},
+	.delay = {
+		.hpd_absent_delay = 200,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
+	.connector_type = DRM_MODE_CONNECTOR_eDP,
+};
+
 static const struct display_timing koe_tx14d24vm1bpa_timing = {
 	.pixelclock = { 5580000, 5850000, 6200000 },
 	.hactive = { 320, 320, 320 },
@@ -2941,12 +3078,12 @@ static const struct drm_display_mode ortustech_com43h4m85ulc_mode  = {
 static const struct panel_desc ortustech_com43h4m85ulc = {
 	.modes = &ortustech_com43h4m85ulc_mode,
 	.num_modes = 1,
-	.bpc = 8,
+	.bpc = 6,
 	.size = {
 		.width = 56,
 		.height = 93,
 	},
-	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
 	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	.connector_type = DRM_MODE_CONNECTOR_DPI,
 };
@@ -3000,6 +3137,31 @@ static const struct panel_desc pda_91_00156_a0  = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
 };
 
+static const struct drm_display_mode powertip_ph800480t013_idf02_mode = {
+	.clock = 24750,
+	.hdisplay = 800,
+	.hsync_start = 800 + 54,
+	.hsync_end = 800 + 54 + 2,
+	.htotal = 800 + 54 + 2 + 44,
+	.vdisplay = 480,
+	.vsync_start = 480 + 49,
+	.vsync_end = 480 + 49 + 2,
+	.vtotal = 480 + 49 + 2 + 22,
+};
+
+static const struct panel_desc powertip_ph800480t013_idf02  = {
+	.modes = &powertip_ph800480t013_idf02_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH |
+		     DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE |
+		     DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE,
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.connector_type = DRM_MODE_CONNECTOR_DPI,
+};
 
 static const struct drm_display_mode qd43003c0_40_mode = {
 	.clock = 9000,
@@ -3301,22 +3463,36 @@ static const struct panel_desc sharp_lq123p1jx31 = {
 	},
 };
 
-static const struct display_timing sharp_ls020b1dd01d_timing = {
-	.pixelclock = { 2000000, 4200000, 5000000 },
-	.hactive = { 240, 240, 240 },
-	.hfront_porch = { 66, 66, 66 },
-	.hback_porch = { 1, 1, 1 },
-	.hsync_len = { 1, 1, 1 },
-	.vactive = { 160, 160, 160 },
-	.vfront_porch = { 52, 52, 52 },
-	.vback_porch = { 6, 6, 6 },
-	.vsync_len = { 10, 10, 10 },
-	.flags = DISPLAY_FLAGS_HSYNC_HIGH | DISPLAY_FLAGS_VSYNC_LOW,
+static const struct drm_display_mode sharp_ls020b1dd01d_modes[] = {
+	{ /* 50 Hz */
+		.clock = 3000,
+		.hdisplay = 240,
+		.hsync_start = 240 + 58,
+		.hsync_end = 240 + 58 + 1,
+		.htotal = 240 + 58 + 1 + 1,
+		.vdisplay = 160,
+		.vsync_start = 160 + 24,
+		.vsync_end = 160 + 24 + 10,
+		.vtotal = 160 + 24 + 10 + 6,
+		.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC,
+	},
+	{ /* 60 Hz */
+		.clock = 3000,
+		.hdisplay = 240,
+		.hsync_start = 240 + 8,
+		.hsync_end = 240 + 8 + 1,
+		.htotal = 240 + 8 + 1 + 1,
+		.vdisplay = 160,
+		.vsync_start = 160 + 24,
+		.vsync_end = 160 + 24 + 10,
+		.vtotal = 160 + 24 + 10 + 6,
+		.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC,
+	},
 };
 
 static const struct panel_desc sharp_ls020b1dd01d = {
-	.timings = &sharp_ls020b1dd01d_timing,
-	.num_timings = 1,
+	.modes = sharp_ls020b1dd01d_modes,
+	.num_modes = ARRAY_SIZE(sharp_ls020b1dd01d_modes),
 	.bpc = 6,
 	.size = {
 		.width = 42,
@@ -3725,6 +3901,9 @@ static const struct panel_desc arm_rtsm = {
 
 static const struct of_device_id platform_of_match[] = {
 	{
+		.compatible = "ampire,am-1280800n3tzqw-t00h",
+		.data = &ampire_am_1280800n3tzqw_t00h,
+	}, {
 		.compatible = "ampire,am-480272h3tmqw-t01h",
 		.data = &ampire_am_480272h3tmqw_t01h,
 	}, {
@@ -3821,6 +4000,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "cdtech,s070wv95-ct16",
 		.data = &cdtech_s070wv95_ct16,
 	}, {
+		.compatible = "chefree,ch101olhlwh-002",
+		.data = &chefree_ch101olhlwh_002,
+	}, {
 		.compatible = "chunghwa,claa070wp03xg",
 		.data = &chunghwa_claa070wp03xg,
 	}, {
@@ -3923,6 +4105,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "ivo,m133nwf4-r0",
 		.data = &ivo_m133nwf4_r0,
 	}, {
+		.compatible = "kingdisplay,kd116n21-30nv-a010",
+		.data = &kingdisplay_kd116n21_30nv_a010,
+	}, {
 		.compatible = "koe,tx14d24vm1bpa",
 		.data = &koe_tx14d24vm1bpa,
 	}, {
@@ -4013,6 +4198,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "pda,91-00156-a0",
 		.data = &pda_91_00156_a0,
 	}, {
+		.compatible = "powertip,ph800480t013-idf02",
+		.data = &powertip_ph800480t013_idf02,
+	}, {
 		.compatible = "qiaodian,qd43003c0-40",
 		.data = &qd43003c0_40,
 	}, {
diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7701.c b/drivers/gpu/drm/panel/panel-sitronix-st7701.c
index 692041ae4eb6..4d2a149b202c 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7701.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7701.c
@@ -7,7 +7,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <linux/gpio/consumer.h>
 #include <linux/delay.h>
@@ -269,10 +268,9 @@ static int st7701_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, desc_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(&st7701->dsi->dev,
-			      "failed to add mode %ux%ux@%u\n",
-			      desc_mode->hdisplay, desc_mode->vdisplay,
-			      drm_mode_vrefresh(desc_mode));
+		dev_err(&st7701->dsi->dev, "failed to add mode %ux%u@%u\n",
+			desc_mode->hdisplay, desc_mode->vdisplay,
+			drm_mode_vrefresh(desc_mode));
 		return -ENOMEM;
 	}
 
@@ -358,7 +356,7 @@ static int st7701_dsi_probe(struct mipi_dsi_device *dsi)
 
 	st7701->reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(st7701->reset)) {
-		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		dev_err(&dsi->dev, "Couldn't get our reset GPIO\n");
 		return PTR_ERR(st7701->reset);
 	}
 
@@ -380,9 +378,7 @@ static int st7701_dsi_probe(struct mipi_dsi_device *dsi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&st7701->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&st7701->panel);
 
 	mipi_dsi_set_drvdata(dsi, st7701);
 	st7701->dsi = dsi;
diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7703.c b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
index 8996ced2b721..c22e7c49e077 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7703.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
@@ -22,7 +22,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define DRV_NAME "panel-sitronix-st7703"
 
@@ -364,8 +363,7 @@ static int st7703_enable(struct drm_panel *panel)
 
 	ret = ctx->desc->init_sequence(ctx);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n",
-			      ret);
+		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
 		return ret;
 	}
 
@@ -373,7 +371,7 @@ static int st7703_enable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
+		dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
 		return ret;
 	}
 
@@ -384,7 +382,7 @@ static int st7703_enable(struct drm_panel *panel)
 	if (ret)
 		return ret;
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Panel init sequence done\n");
+	dev_dbg(ctx->dev, "Panel init sequence done\n");
 
 	return 0;
 }
@@ -397,13 +395,11 @@ static int st7703_disable(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to turn off the display: %d\n", ret);
+		dev_err(ctx->dev, "Failed to turn off the display: %d\n", ret);
 
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enter sleep mode: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enter sleep mode: %d\n", ret);
 
 	return 0;
 }
@@ -431,17 +427,15 @@ static int st7703_prepare(struct drm_panel *panel)
 	if (ctx->prepared)
 		return 0;
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n");
+	dev_dbg(ctx->dev, "Resetting the panel\n");
 	ret = regulator_enable(ctx->vcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable vcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable vcc supply: %d\n", ret);
 		return ret;
 	}
 	ret = regulator_enable(ctx->iovcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable iovcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret);
 		goto disable_vcc;
 	}
 
@@ -467,9 +461,9 @@ static int st7703_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, ctx->desc->mode);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to add mode %ux%u@%u\n",
-			      ctx->desc->mode->hdisplay, ctx->desc->mode->vdisplay,
-			      drm_mode_vrefresh(ctx->desc->mode));
+		dev_err(ctx->dev, "Failed to add mode %ux%u@%u\n",
+			ctx->desc->mode->hdisplay, ctx->desc->mode->vdisplay,
+			drm_mode_vrefresh(ctx->desc->mode));
 		return -ENOMEM;
 	}
 
@@ -496,7 +490,7 @@ static int allpixelson_set(void *data, u64 val)
 	struct st7703 *ctx = data;
 	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Setting all pixels on\n");
+	dev_dbg(ctx->dev, "Setting all pixels on\n");
 	dsi_generic_write_seq(dsi, ST7703_CMD_ALL_PIXEL_ON);
 	msleep(val * 1000);
 	/* Reset the panel to get video back */
@@ -537,7 +531,7 @@ static int st7703_probe(struct mipi_dsi_device *dsi)
 
 	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		dev_err(dev, "cannot get reset gpio\n");
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -554,18 +548,14 @@ static int st7703_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->vcc)) {
 		ret = PTR_ERR(ctx->vcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request vcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request vcc regulator: %d\n", ret);
 		return ret;
 	}
 	ctx->iovcc = devm_regulator_get(dev, "iovcc");
 	if (IS_ERR(ctx->iovcc)) {
 		ret = PTR_ERR(ctx->iovcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request iovcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request iovcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -580,17 +570,15 @@ static int st7703_probe(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev,
-			      "mipi_dsi_attach failed (%d). Is host ready?\n",
-			      ret);
+		dev_err(dev, "mipi_dsi_attach failed (%d). Is host ready?\n", ret);
 		drm_panel_remove(&ctx->panel);
 		return ret;
 	}
 
-	DRM_DEV_INFO(dev, "%ux%u@%u %ubpp dsi %udl - ready\n",
-		     ctx->desc->mode->hdisplay, ctx->desc->mode->vdisplay,
-		     drm_mode_vrefresh(ctx->desc->mode),
-		     mipi_dsi_pixel_format_to_bpp(dsi->format), dsi->lanes);
+	dev_info(dev, "%ux%u@%u %ubpp dsi %udl - ready\n",
+		 ctx->desc->mode->hdisplay, ctx->desc->mode->vdisplay,
+		 drm_mode_vrefresh(ctx->desc->mode),
+		 mipi_dsi_pixel_format_to_bpp(dsi->format), dsi->lanes);
 
 	st7703_debugfs_init(ctx);
 	return 0;
@@ -603,13 +591,11 @@ static void st7703_shutdown(struct mipi_dsi_device *dsi)
 
 	ret = drm_panel_unprepare(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
 
 	ret = drm_panel_disable(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
 }
 
 static int st7703_remove(struct mipi_dsi_device *dsi)
@@ -621,8 +607,7 @@ static int st7703_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
 
 	drm_panel_remove(&ctx->panel);
 
diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
index 3513ae40efa8..61e565524542 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
@@ -382,9 +382,7 @@ static int st7789v_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	ret = drm_panel_add(&ctx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&ctx->panel);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/panel/panel-sony-acx424akp.c b/drivers/gpu/drm/panel/panel-sony-acx424akp.c
index 97a1b4790d3c..065efae213f5 100644
--- a/drivers/gpu/drm/panel/panel-sony-acx424akp.c
+++ b/drivers/gpu/drm/panel/panel-sony-acx424akp.c
@@ -20,7 +20,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #define ACX424_DCS_READ_ID1		0xDA
 #define ACX424_DCS_READ_ID2		0xDB
@@ -110,13 +109,11 @@ static int acx424akp_set_brightness(struct backlight_device *bl)
 		      SCALE_FACTOR_NS_DIV_MHZ);
 
 	/* Set up PWM dutycycle ONE byte (differs from the standard) */
-	DRM_DEV_DEBUG(acx->dev, "calculated duty cycle %02x\n", pwm_ratio);
+	dev_dbg(acx->dev, "calculated duty cycle %02x\n", pwm_ratio);
 	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_SET_DISPLAY_BRIGHTNESS,
 				 &pwm_ratio, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to set display PWM ratio (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to set display PWM ratio (%d)\n", ret);
 		return ret;
 	}
 
@@ -132,40 +129,30 @@ static int acx424akp_set_brightness(struct backlight_device *bl)
 	par = 0xaa;
 	ret = mipi_dsi_dcs_write(dsi, 0xf3, &par, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to unlock CMD 2 (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to unlock CMD 2 (%d)\n", ret);
 		return ret;
 	}
 	par = 0x01;
 	ret = mipi_dsi_dcs_write(dsi, 0x00, &par, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to enter page 1 (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to enter page 1 (%d)\n", ret);
 		return ret;
 	}
 	par = 0x01;
 	ret = mipi_dsi_dcs_write(dsi, 0x7d, &par, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to disable MTP reload (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to disable MTP reload (%d)\n", ret);
 		return ret;
 	}
 	ret = mipi_dsi_dcs_write(dsi, 0x22, &pwm_div, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to set PWM divisor (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to set PWM divisor (%d)\n", ret);
 		return ret;
 	}
 	par = 0xaa;
 	ret = mipi_dsi_dcs_write(dsi, 0x7f, &par, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to lock CMD 2 (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to lock CMD 2 (%d)\n", ret);
 		return ret;
 	}
 
@@ -174,9 +161,7 @@ static int acx424akp_set_brightness(struct backlight_device *bl)
 	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY,
 				 &par, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to enable display backlight (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to enable display backlight (%d)\n", ret);
 		return ret;
 	}
 
@@ -196,22 +181,22 @@ static int acx424akp_read_id(struct acx424akp *acx)
 
 	ret = mipi_dsi_dcs_read(dsi, ACX424_DCS_READ_ID1, &vendor, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev, "could not vendor ID byte\n");
+		dev_err(acx->dev, "could not vendor ID byte\n");
 		return ret;
 	}
 	ret = mipi_dsi_dcs_read(dsi, ACX424_DCS_READ_ID2, &version, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev, "could not read device version byte\n");
+		dev_err(acx->dev, "could not read device version byte\n");
 		return ret;
 	}
 	ret = mipi_dsi_dcs_read(dsi, ACX424_DCS_READ_ID3, &panel, 1);
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev, "could not read panel ID byte\n");
+		dev_err(acx->dev, "could not read panel ID byte\n");
 		return ret;
 	}
 
 	if (vendor == 0x00) {
-		DRM_DEV_ERROR(acx->dev, "device vendor ID is zero\n");
+		dev_err(acx->dev, "device vendor ID is zero\n");
 		return -ENODEV;
 	}
 
@@ -220,14 +205,12 @@ static int acx424akp_read_id(struct acx424akp *acx)
 	case DISPLAY_SONY_ACX424AKP_ID1:
 	case DISPLAY_SONY_ACX424AKP_ID2:
 	case DISPLAY_SONY_ACX424AKP_ID3:
-		DRM_DEV_INFO(acx->dev,
-			     "MTP vendor: %02x, version: %02x, panel: %02x\n",
-			     vendor, version, panel);
+		dev_info(acx->dev, "MTP vendor: %02x, version: %02x, panel: %02x\n",
+			 vendor, version, panel);
 		break;
 	default:
-		DRM_DEV_INFO(acx->dev,
-			     "unknown vendor: %02x, version: %02x, panel: %02x\n",
-			     vendor, version, panel);
+		dev_info(acx->dev, "unknown vendor: %02x, version: %02x, panel: %02x\n",
+			 vendor, version, panel);
 		break;
 	}
 
@@ -240,7 +223,7 @@ static int acx424akp_power_on(struct acx424akp *acx)
 
 	ret = regulator_enable(acx->supply);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to enable supply (%d)\n", ret);
+		dev_err(acx->dev, "failed to enable supply (%d)\n", ret);
 		return ret;
 	}
 
@@ -276,7 +259,7 @@ static int acx424akp_prepare(struct drm_panel *panel)
 
 	ret = acx424akp_read_id(acx);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to read panel ID (%d)\n", ret);
+		dev_err(acx->dev, "failed to read panel ID (%d)\n", ret);
 		goto err_power_off;
 	}
 
@@ -284,8 +267,7 @@ static int acx424akp_prepare(struct drm_panel *panel)
 	ret = mipi_dsi_dcs_set_tear_on(dsi,
 				       MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to enable vblank TE (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to enable vblank TE (%d)\n", ret);
 		goto err_power_off;
 	}
 
@@ -302,23 +284,21 @@ static int acx424akp_prepare(struct drm_panel *panel)
 	ret = mipi_dsi_dcs_write(dsi, ACX424_DCS_SET_MDDI,
 				 &mddi, sizeof(mddi));
 	if (ret < 0) {
-		DRM_DEV_ERROR(acx->dev, "failed to set MDDI (%d)\n", ret);
+		dev_err(acx->dev, "failed to set MDDI (%d)\n", ret);
 		goto err_power_off;
 	}
 
 	/* Exit sleep mode */
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to exit sleep mode (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to exit sleep mode (%d)\n", ret);
 		goto err_power_off;
 	}
 	msleep(140);
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to turn display on (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to turn display on (%d)\n", ret);
 		goto err_power_off;
 	}
 	if (acx->video_mode) {
@@ -351,24 +331,20 @@ static int acx424akp_unprepare(struct drm_panel *panel)
 	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY,
 				 &par, 1);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev,
-			      "failed to disable display backlight (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to disable display backlight (%d)\n", ret);
 		return ret;
 	}
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to turn display off (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to turn display off (%d)\n", ret);
 		return ret;
 	}
 
 	/* Enter sleep mode */
 	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret) {
-		DRM_DEV_ERROR(acx->dev, "failed to enter sleep mode (%d)\n",
-			      ret);
+		dev_err(acx->dev, "failed to enter sleep mode (%d)\n", ret);
 		return ret;
 	}
 	msleep(85);
@@ -418,7 +394,7 @@ static int acx424akp_get_modes(struct drm_panel *panel,
 		mode = drm_mode_duplicate(connector->dev,
 					  &sony_acx424akp_cmd_mode);
 	if (!mode) {
-		DRM_ERROR("bad mode or failed to add mode\n");
+		dev_err(panel->dev, "bad mode or failed to add mode\n");
 		return -EINVAL;
 	}
 	drm_mode_set_name(mode);
@@ -486,8 +462,7 @@ static int acx424akp_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(acx->reset_gpio)) {
 		ret = PTR_ERR(acx->reset_gpio);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "failed to request GPIO (%d)\n",
-				      ret);
+			dev_err(dev, "failed to request GPIO (%d)\n", ret);
 		return ret;
 	}
 
@@ -497,16 +472,14 @@ static int acx424akp_probe(struct mipi_dsi_device *dsi)
 	acx->bl = devm_backlight_device_register(dev, "acx424akp", dev, acx,
 						 &acx424akp_bl_ops, NULL);
 	if (IS_ERR(acx->bl)) {
-		DRM_DEV_ERROR(dev, "failed to register backlight device\n");
+		dev_err(dev, "failed to register backlight device\n");
 		return PTR_ERR(acx->bl);
 	}
 	acx->bl->props.max_brightness = 1023;
 	acx->bl->props.brightness = 512;
 	acx->bl->props.power = FB_BLANK_POWERDOWN;
 
-	ret = drm_panel_add(&acx->panel);
-	if (ret < 0)
-		return ret;
+	drm_panel_add(&acx->panel);
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
diff --git a/drivers/gpu/drm/panel/panel-sony-acx565akm.c b/drivers/gpu/drm/panel/panel-sony-acx565akm.c
index fc6a7e451abe..e95fdfb16b6c 100644
--- a/drivers/gpu/drm/panel/panel-sony-acx565akm.c
+++ b/drivers/gpu/drm/panel/panel-sony-acx565akm.c
@@ -650,12 +650,7 @@ static int acx565akm_probe(struct spi_device *spi)
 	drm_panel_init(&lcd->panel, &lcd->spi->dev, &acx565akm_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	ret = drm_panel_add(&lcd->panel);
-	if (ret < 0) {
-		if (lcd->has_bc)
-			acx565akm_backlight_cleanup(lcd);
-		return ret;
-	}
+	drm_panel_add(&lcd->panel);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/panel/panel-tpo-td028ttec1.c b/drivers/gpu/drm/panel/panel-tpo-td028ttec1.c
index 58d683cc5215..037c14fd6bac 100644
--- a/drivers/gpu/drm/panel/panel-tpo-td028ttec1.c
+++ b/drivers/gpu/drm/panel/panel-tpo-td028ttec1.c
@@ -350,7 +350,9 @@ static int td028ttec1_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	return drm_panel_add(&lcd->panel);
+	drm_panel_add(&lcd->panel);
+
+	return 0;
 }
 
 static int td028ttec1_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-tpo-td043mtea1.c b/drivers/gpu/drm/panel/panel-tpo-td043mtea1.c
index 9b2a356c4d9a..49e6c9386258 100644
--- a/drivers/gpu/drm/panel/panel-tpo-td043mtea1.c
+++ b/drivers/gpu/drm/panel/panel-tpo-td043mtea1.c
@@ -460,11 +460,7 @@ static int td043mtea1_probe(struct spi_device *spi)
 	drm_panel_init(&lcd->panel, &lcd->spi->dev, &td043mtea1_funcs,
 		       DRM_MODE_CONNECTOR_DPI);
 
-	ret = drm_panel_add(&lcd->panel);
-	if (ret < 0) {
-		sysfs_remove_group(&spi->dev.kobj, &td043mtea1_attr_group);
-		return ret;
-	}
+	drm_panel_add(&lcd->panel);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/panel/panel-tpo-tpg110.c b/drivers/gpu/drm/panel/panel-tpo-tpg110.c
index c7a2f0ae5ba5..d57ed75a977c 100644
--- a/drivers/gpu/drm/panel/panel-tpo-tpg110.c
+++ b/drivers/gpu/drm/panel/panel-tpo-tpg110.c
@@ -12,7 +12,6 @@
  */
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <linux/bitops.h>
 #include <linux/delay.h>
@@ -238,7 +237,7 @@ static u8 tpg110_readwrite_reg(struct tpg110 *tpg, bool write,
 	spi_message_add_tail(&t[1], &m);
 	ret = spi_sync(tpg->spi, &m);
 	if (ret) {
-		DRM_DEV_ERROR(tpg->dev, "SPI message error %d\n", ret);
+		dev_err(tpg->dev, "SPI message error %d\n", ret);
 		return ret;
 	}
 	if (write)
@@ -265,18 +264,18 @@ static int tpg110_startup(struct tpg110 *tpg)
 	/* De-assert the reset signal */
 	gpiod_set_value_cansleep(tpg->grestb, 0);
 	usleep_range(1000, 2000);
-	DRM_DEV_DEBUG(tpg->dev, "de-asserted GRESTB\n");
+	dev_dbg(tpg->dev, "de-asserted GRESTB\n");
 
 	/* Test display communication */
 	tpg110_write_reg(tpg, TPG110_TEST, 0x55);
 	val = tpg110_read_reg(tpg, TPG110_TEST);
 	if (val != 0x55) {
-		DRM_DEV_ERROR(tpg->dev, "failed communication test\n");
+		dev_err(tpg->dev, "failed communication test\n");
 		return -ENODEV;
 	}
 
 	val = tpg110_read_reg(tpg, TPG110_CHIPID);
-	DRM_DEV_INFO(tpg->dev, "TPG110 chip ID: %d version: %d\n",
+	dev_info(tpg->dev, "TPG110 chip ID: %d version: %d\n",
 		 val >> 4, val & 0x0f);
 
 	/* Show display resolution */
@@ -284,27 +283,25 @@ static int tpg110_startup(struct tpg110 *tpg)
 	val &= TPG110_RES_MASK;
 	switch (val) {
 	case TPG110_RES_400X240_D:
-		DRM_DEV_INFO(tpg->dev,
-			 "IN 400x240 RGB -> OUT 800x480 RGB (dual scan)\n");
+		dev_info(tpg->dev, "IN 400x240 RGB -> OUT 800x480 RGB (dual scan)\n");
 		break;
 	case TPG110_RES_480X272_D:
-		DRM_DEV_INFO(tpg->dev,
-			 "IN 480x272 RGB -> OUT 800x480 RGB (dual scan)\n");
+		dev_info(tpg->dev, "IN 480x272 RGB -> OUT 800x480 RGB (dual scan)\n");
 		break;
 	case TPG110_RES_480X640:
-		DRM_DEV_INFO(tpg->dev, "480x640 RGB\n");
+		dev_info(tpg->dev, "480x640 RGB\n");
 		break;
 	case TPG110_RES_480X272:
-		DRM_DEV_INFO(tpg->dev, "480x272 RGB\n");
+		dev_info(tpg->dev, "480x272 RGB\n");
 		break;
 	case TPG110_RES_640X480:
-		DRM_DEV_INFO(tpg->dev, "640x480 RGB\n");
+		dev_info(tpg->dev, "640x480 RGB\n");
 		break;
 	case TPG110_RES_800X480:
-		DRM_DEV_INFO(tpg->dev, "800x480 RGB\n");
+		dev_info(tpg->dev, "800x480 RGB\n");
 		break;
 	default:
-		DRM_DEV_ERROR(tpg->dev, "ILLEGAL RESOLUTION 0x%02x\n", val);
+		dev_err(tpg->dev, "ILLEGAL RESOLUTION 0x%02x\n", val);
 		break;
 	}
 
@@ -322,13 +319,12 @@ static int tpg110_startup(struct tpg110 *tpg)
 		}
 	}
 	if (i == ARRAY_SIZE(tpg110_modes)) {
-		DRM_DEV_ERROR(tpg->dev, "unsupported mode (%02x) detected\n",
-			val);
+		dev_err(tpg->dev, "unsupported mode (%02x) detected\n", val);
 		return -ENODEV;
 	}
 
 	val = tpg110_read_reg(tpg, TPG110_CTRL2);
-	DRM_DEV_INFO(tpg->dev, "resolution and standby is controlled by %s\n",
+	dev_info(tpg->dev, "resolution and standby is controlled by %s\n",
 		 (val & TPG110_CTRL2_RES_PM_CTRL) ? "software" : "hardware");
 	/* Take control over resolution and standby */
 	val |= TPG110_CTRL2_RES_PM_CTRL;
@@ -414,15 +410,15 @@ static int tpg110_probe(struct spi_device *spi)
 	/* We get the physical display dimensions from the DT */
 	ret = of_property_read_u32(np, "width-mm", &tpg->width);
 	if (ret)
-		DRM_DEV_ERROR(dev, "no panel width specified\n");
+		dev_err(dev, "no panel width specified\n");
 	ret = of_property_read_u32(np, "height-mm", &tpg->height);
 	if (ret)
-		DRM_DEV_ERROR(dev, "no panel height specified\n");
+		dev_err(dev, "no panel height specified\n");
 
 	/* This asserts the GRESTB signal, putting the display into reset */
 	tpg->grestb = devm_gpiod_get(dev, "grestb", GPIOD_OUT_HIGH);
 	if (IS_ERR(tpg->grestb)) {
-		DRM_DEV_ERROR(dev, "no GRESTB GPIO\n");
+		dev_err(dev, "no GRESTB GPIO\n");
 		return -ENODEV;
 	}
 
@@ -430,7 +426,7 @@ static int tpg110_probe(struct spi_device *spi)
 	spi->mode |= SPI_3WIRE_HIZ;
 	ret = spi_setup(spi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "spi setup failed.\n");
+		dev_err(dev, "spi setup failed.\n");
 		return ret;
 	}
 	tpg->spi = spi;
@@ -448,7 +444,9 @@ static int tpg110_probe(struct spi_device *spi)
 
 	spi_set_drvdata(spi, tpg);
 
-	return drm_panel_add(&tpg->panel);
+	drm_panel_add(&tpg->panel);
+
+	return 0;
 }
 
 static int tpg110_remove(struct spi_device *spi)
diff --git a/drivers/gpu/drm/panel/panel-truly-nt35597.c b/drivers/gpu/drm/panel/panel-truly-nt35597.c
index 9b9c167b8dc8..b24b92d93ea5 100644
--- a/drivers/gpu/drm/panel/panel-truly-nt35597.c
+++ b/drivers/gpu/drm/panel/panel-truly-nt35597.c
@@ -17,7 +17,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 static const char * const regulator_names[] = {
 	"vdda",
@@ -231,9 +230,7 @@ static int truly_dcs_write(struct drm_panel *panel, u32 command)
 	for (i = 0; i < ARRAY_SIZE(ctx->dsi); i++) {
 		ret = mipi_dsi_dcs_write(ctx->dsi[i], command, NULL, 0);
 		if (ret < 0) {
-			DRM_DEV_ERROR(ctx->dev,
-				"cmd 0x%x failed for dsi = %d\n",
-				command, i);
+			dev_err(ctx->dev, "cmd 0x%x failed for dsi = %d\n", command, i);
 		}
 	}
 
@@ -250,8 +247,7 @@ static int truly_dcs_write_buf(struct drm_panel *panel,
 	for (i = 0; i < ARRAY_SIZE(ctx->dsi); i++) {
 		ret = mipi_dsi_dcs_write_buffer(ctx->dsi[i], buf, size);
 		if (ret < 0) {
-			DRM_DEV_ERROR(ctx->dev,
-				"failed to tx cmd [%d], err: %d\n", i, ret);
+			dev_err(ctx->dev, "failed to tx cmd [%d], err: %d\n", i, ret);
 			return ret;
 		}
 	}
@@ -300,16 +296,14 @@ static int truly_nt35597_power_off(struct truly_nt35597 *ctx)
 		ret = regulator_set_load(ctx->supplies[i].consumer,
 				regulator_disable_loads[i]);
 		if (ret) {
-			DRM_DEV_ERROR(ctx->dev,
-				"regulator_set_load failed %d\n", ret);
+			dev_err(ctx->dev, "regulator_set_load failed %d\n", ret);
 			return ret;
 		}
 	}
 
 	ret = regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies);
 	if (ret) {
-		DRM_DEV_ERROR(ctx->dev,
-			"regulator_bulk_disable failed %d\n", ret);
+		dev_err(ctx->dev, "regulator_bulk_disable failed %d\n", ret);
 	}
 	return ret;
 }
@@ -325,8 +319,7 @@ static int truly_nt35597_disable(struct drm_panel *panel)
 	if (ctx->backlight) {
 		ret = backlight_disable(ctx->backlight);
 		if (ret < 0)
-			DRM_DEV_ERROR(ctx->dev, "backlight disable failed %d\n",
-				ret);
+			dev_err(ctx->dev, "backlight disable failed %d\n", ret);
 	}
 
 	ctx->enabled = false;
@@ -346,9 +339,7 @@ static int truly_nt35597_unprepare(struct drm_panel *panel)
 
 	ret = truly_dcs_write(panel, MIPI_DCS_SET_DISPLAY_OFF);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			"set_display_off cmd failed ret = %d\n",
-			ret);
+		dev_err(ctx->dev, "set_display_off cmd failed ret = %d\n", ret);
 	}
 
 	/* 120ms delay required here as per DCS spec */
@@ -356,13 +347,12 @@ static int truly_nt35597_unprepare(struct drm_panel *panel)
 
 	ret = truly_dcs_write(panel, MIPI_DCS_ENTER_SLEEP_MODE);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			"enter_sleep cmd failed ret = %d\n", ret);
+		dev_err(ctx->dev, "enter_sleep cmd failed ret = %d\n", ret);
 	}
 
 	ret = truly_nt35597_power_off(ctx);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->dev, "power_off failed ret = %d\n", ret);
+		dev_err(ctx->dev, "power_off failed ret = %d\n", ret);
 
 	ctx->prepared = false;
 	return ret;
@@ -396,18 +386,14 @@ static int truly_nt35597_prepare(struct drm_panel *panel)
 				panel_on_cmds[i].size,
 					panel_on_cmds[i].commands);
 		if (ret < 0) {
-			DRM_DEV_ERROR(ctx->dev,
-				"cmd set tx failed i = %d ret = %d\n",
-					i, ret);
+			dev_err(ctx->dev, "cmd set tx failed i = %d ret = %d\n", i, ret);
 			goto power_off;
 		}
 	}
 
 	ret = truly_dcs_write(panel, MIPI_DCS_EXIT_SLEEP_MODE);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			"exit_sleep_mode cmd failed ret = %d\n",
-			ret);
+		dev_err(ctx->dev, "exit_sleep_mode cmd failed ret = %d\n", ret);
 		goto power_off;
 	}
 
@@ -416,8 +402,7 @@ static int truly_nt35597_prepare(struct drm_panel *panel)
 
 	ret = truly_dcs_write(panel, MIPI_DCS_SET_DISPLAY_ON);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			"set_display_on cmd failed ret = %d\n", ret);
+		dev_err(ctx->dev, "set_display_on cmd failed ret = %d\n", ret);
 		goto power_off;
 	}
 
@@ -430,7 +415,7 @@ static int truly_nt35597_prepare(struct drm_panel *panel)
 
 power_off:
 	if (truly_nt35597_power_off(ctx))
-		DRM_DEV_ERROR(ctx->dev, "power_off failed\n");
+		dev_err(ctx->dev, "power_off failed\n");
 	return ret;
 }
 
@@ -445,8 +430,7 @@ static int truly_nt35597_enable(struct drm_panel *panel)
 	if (ctx->backlight) {
 		ret = backlight_enable(ctx->backlight);
 		if (ret < 0)
-			DRM_DEV_ERROR(ctx->dev, "backlight enable failed %d\n",
-						  ret);
+			dev_err(ctx->dev, "backlight enable failed %d\n", ret);
 	}
 
 	ctx->enabled = true;
@@ -464,8 +448,7 @@ static int truly_nt35597_get_modes(struct drm_panel *panel,
 	config = ctx->config;
 	mode = drm_mode_create(connector->dev);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->dev,
-			"failed to create a new display mode\n");
+		dev_err(ctx->dev, "failed to create a new display mode\n");
 		return 0;
 	}
 
@@ -501,15 +484,13 @@ static int truly_nt35597_panel_add(struct truly_nt35597 *ctx)
 
 	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio %ld\n",
-			PTR_ERR(ctx->reset_gpio));
+		dev_err(dev, "cannot get reset gpio %ld\n", PTR_ERR(ctx->reset_gpio));
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
 	ctx->mode_gpio = devm_gpiod_get(dev, "mode", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->mode_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get mode gpio %ld\n",
-			PTR_ERR(ctx->mode_gpio));
+		dev_err(dev, "cannot get mode gpio %ld\n", PTR_ERR(ctx->mode_gpio));
 		return PTR_ERR(ctx->mode_gpio);
 	}
 
@@ -584,22 +565,21 @@ static int truly_nt35597_probe(struct mipi_dsi_device *dsi)
 
 	dsi1 = of_graph_get_remote_node(dsi->dev.of_node, 1, -1);
 	if (!dsi1) {
-		DRM_DEV_ERROR(dev,
-			"failed to get remote node for dsi1_device\n");
+		dev_err(dev, "failed to get remote node for dsi1_device\n");
 		return -ENODEV;
 	}
 
 	dsi1_host = of_find_mipi_dsi_host_by_node(dsi1);
 	of_node_put(dsi1);
 	if (!dsi1_host) {
-		DRM_DEV_ERROR(dev, "failed to find dsi host\n");
+		dev_err(dev, "failed to find dsi host\n");
 		return -EPROBE_DEFER;
 	}
 
 	/* register the second DSI device */
 	dsi1_device = mipi_dsi_device_register_full(dsi1_host, &info);
 	if (IS_ERR(dsi1_device)) {
-		DRM_DEV_ERROR(dev, "failed to create dsi device\n");
+		dev_err(dev, "failed to create dsi device\n");
 		return PTR_ERR(dsi1_device);
 	}
 
@@ -611,7 +591,7 @@ static int truly_nt35597_probe(struct mipi_dsi_device *dsi)
 
 	ret = truly_nt35597_panel_add(ctx);
 	if (ret) {
-		DRM_DEV_ERROR(dev, "failed to add panel\n");
+		dev_err(dev, "failed to add panel\n");
 		goto err_panel_add;
 	}
 
@@ -623,8 +603,7 @@ static int truly_nt35597_probe(struct mipi_dsi_device *dsi)
 			MIPI_DSI_CLOCK_NON_CONTINUOUS;
 		ret = mipi_dsi_attach(dsi_dev);
 		if (ret < 0) {
-			DRM_DEV_ERROR(dev,
-				"dsi attach failed i = %d\n", i);
+			dev_err(dev, "dsi attach failed i = %d\n", i);
 			goto err_dsi_attach;
 		}
 	}
diff --git a/drivers/gpu/drm/panel/panel-visionox-rm69299.c b/drivers/gpu/drm/panel/panel-visionox-rm69299.c
index a12976b497ce..eb43503ec97b 100644
--- a/drivers/gpu/drm/panel/panel-visionox-rm69299.c
+++ b/drivers/gpu/drm/panel/panel-visionox-rm69299.c
@@ -14,7 +14,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 struct visionox_rm69299 {
 	struct drm_panel panel;
@@ -69,16 +68,14 @@ static int visionox_rm69299_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_write(ctx->dsi, MIPI_DCS_SET_DISPLAY_OFF, NULL, 0);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "set_display_off cmd failed ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "set_display_off cmd failed ret = %d\n", ret);
 
 	/* 120ms delay required here as per DCS spec */
 	msleep(120);
 
 	ret = mipi_dsi_dcs_write(ctx->dsi, MIPI_DCS_ENTER_SLEEP_MODE, NULL, 0);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "enter_sleep cmd failed ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "enter_sleep cmd failed ret = %d\n", ret);
 	}
 
 	ret = visionox_rm69299_power_off(ctx);
@@ -103,36 +100,31 @@ static int visionox_rm69299_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, (u8[]) { 0xfe, 0x00 }, 2);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "cmd set tx 0 failed, ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "cmd set tx 0 failed, ret = %d\n", ret);
 		goto power_off;
 	}
 
 	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, (u8[]) { 0xc2, 0x08 }, 2);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "cmd set tx 1 failed, ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "cmd set tx 1 failed, ret = %d\n", ret);
 		goto power_off;
 	}
 
 	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, (u8[]) { 0x35, 0x00 }, 2);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "cmd set tx 2 failed, ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "cmd set tx 2 failed, ret = %d\n", ret);
 		goto power_off;
 	}
 
 	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, (u8[]) { 0x51, 0xff }, 2);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "cmd set tx 3 failed, ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "cmd set tx 3 failed, ret = %d\n", ret);
 		goto power_off;
 	}
 
 	ret = mipi_dsi_dcs_write(ctx->dsi, MIPI_DCS_EXIT_SLEEP_MODE, NULL, 0);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "exit_sleep_mode cmd failed ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "exit_sleep_mode cmd failed ret = %d\n", ret);
 		goto power_off;
 	}
 
@@ -141,8 +133,7 @@ static int visionox_rm69299_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_write(ctx->dsi, MIPI_DCS_SET_DISPLAY_ON, NULL, 0);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "set_display_on cmd failed ret = %d\n", ret);
+		dev_err(ctx->panel.dev, "set_display_on cmd failed ret = %d\n", ret);
 		goto power_off;
 	}
 
@@ -179,8 +170,7 @@ static int visionox_rm69299_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_create(connector->dev);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->panel.dev,
-			      "failed to create a new display mode\n");
+		dev_err(ctx->panel.dev, "failed to create a new display mode\n");
 		return 0;
 	}
 
@@ -225,8 +215,7 @@ static int visionox_rm69299_probe(struct mipi_dsi_device *dsi)
 	ctx->reset_gpio = devm_gpiod_get(ctx->panel.dev,
 					 "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio %ld\n",
-			      PTR_ERR(ctx->reset_gpio));
+		dev_err(dev, "cannot get reset gpio %ld\n", PTR_ERR(ctx->reset_gpio));
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -242,23 +231,19 @@ static int visionox_rm69299_probe(struct mipi_dsi_device *dsi)
 			  MIPI_DSI_CLOCK_NON_CONTINUOUS;
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "dsi attach failed ret = %d\n", ret);
+		dev_err(dev, "dsi attach failed ret = %d\n", ret);
 		goto err_dsi_attach;
 	}
 
 	ret = regulator_set_load(ctx->supplies[0].consumer, 32000);
 	if (ret) {
-		DRM_DEV_ERROR(dev,
-			      "regulator set load failed for vdda supply ret = %d\n",
-			      ret);
+		dev_err(dev, "regulator set load failed for vdda supply ret = %d\n", ret);
 		goto err_set_load;
 	}
 
 	ret = regulator_set_load(ctx->supplies[1].consumer, 13200);
 	if (ret) {
-		DRM_DEV_ERROR(dev,
-			      "regulator set load failed for vdd3p3 supply ret = %d\n",
-			      ret);
+		dev_err(dev, "regulator set load failed for vdd3p3 supply ret = %d\n", ret);
 		goto err_set_load;
 	}
 
diff --git a/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c b/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c
index 06341deb60ca..55172d63a922 100644
--- a/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c
+++ b/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c
@@ -12,7 +12,6 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_print.h>
 
 #include <video/display_timing.h>
 #include <video/mipi_display.h>
@@ -135,7 +134,7 @@ static int xpp055c272_init_sequence(struct xpp055c272 *ctx)
 
 	msleep(60);
 
-	DRM_DEV_DEBUG_DRIVER(dev, "Panel init sequence done\n");
+	dev_dbg(dev, "Panel init sequence done\n");
 	return 0;
 }
 
@@ -150,13 +149,11 @@ static int xpp055c272_unprepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_off(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(ctx->dev, "failed to set display off: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to set display off: %d\n", ret);
 
 	mipi_dsi_dcs_enter_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "failed to enter sleep mode: %d\n",
-			      ret);
+		dev_err(ctx->dev, "failed to enter sleep mode: %d\n", ret);
 		return ret;
 	}
 
@@ -177,17 +174,15 @@ static int xpp055c272_prepare(struct drm_panel *panel)
 	if (ctx->prepared)
 		return 0;
 
-	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n");
+	dev_dbg(ctx->dev, "Resetting the panel\n");
 	ret = regulator_enable(ctx->vci);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable vci supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable vci supply: %d\n", ret);
 		return ret;
 	}
 	ret = regulator_enable(ctx->iovcc);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev,
-			      "Failed to enable iovcc supply: %d\n", ret);
+		dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret);
 		goto disable_vci;
 	}
 
@@ -201,14 +196,13 @@ static int xpp055c272_prepare(struct drm_panel *panel)
 
 	ret = xpp055c272_init_sequence(ctx);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n",
-			      ret);
+		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
 		goto disable_iovcc;
 	}
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
+		dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -217,7 +211,7 @@ static int xpp055c272_prepare(struct drm_panel *panel)
 
 	ret = mipi_dsi_dcs_set_display_on(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to set display on: %d\n", ret);
+		dev_err(ctx->dev, "Failed to set display on: %d\n", ret);
 		goto disable_iovcc;
 	}
 
@@ -256,9 +250,9 @@ static int xpp055c272_get_modes(struct drm_panel *panel,
 
 	mode = drm_mode_duplicate(connector->dev, &default_mode);
 	if (!mode) {
-		DRM_DEV_ERROR(ctx->dev, "Failed to add mode %ux%u@%u\n",
-			      default_mode.hdisplay, default_mode.vdisplay,
-			      drm_mode_vrefresh(&default_mode));
+		dev_err(ctx->dev, "Failed to add mode %ux%u@%u\n",
+			default_mode.hdisplay, default_mode.vdisplay,
+			drm_mode_vrefresh(&default_mode));
 		return -ENOMEM;
 	}
 
@@ -290,7 +284,7 @@ static int xpp055c272_probe(struct mipi_dsi_device *dsi)
 
 	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ctx->reset_gpio)) {
-		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		dev_err(dev, "cannot get reset gpio\n");
 		return PTR_ERR(ctx->reset_gpio);
 	}
 
@@ -298,9 +292,7 @@ static int xpp055c272_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->vci)) {
 		ret = PTR_ERR(ctx->vci);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request vci regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request vci regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -308,9 +300,7 @@ static int xpp055c272_probe(struct mipi_dsi_device *dsi)
 	if (IS_ERR(ctx->iovcc)) {
 		ret = PTR_ERR(ctx->iovcc);
 		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev,
-				      "Failed to request iovcc regulator: %d\n",
-				      ret);
+			dev_err(dev, "Failed to request iovcc regulator: %d\n", ret);
 		return ret;
 	}
 
@@ -334,7 +324,7 @@ static int xpp055c272_probe(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_attach(dsi);
 	if (ret < 0) {
-		DRM_DEV_ERROR(dev, "mipi_dsi_attach failed: %d\n", ret);
+		dev_err(dev, "mipi_dsi_attach failed: %d\n", ret);
 		drm_panel_remove(&ctx->panel);
 		return ret;
 	}
@@ -349,13 +339,11 @@ static void xpp055c272_shutdown(struct mipi_dsi_device *dsi)
 
 	ret = drm_panel_unprepare(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
 
 	ret = drm_panel_disable(&ctx->panel);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
 }
 
 static int xpp055c272_remove(struct mipi_dsi_device *dsi)
@@ -367,8 +355,7 @@ static int xpp055c272_remove(struct mipi_dsi_device *dsi)
 
 	ret = mipi_dsi_detach(dsi);
 	if (ret < 0)
-		DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n",
-			      ret);
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
 
 	drm_panel_remove(&ctx->panel);
 
diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c
index 413987038fbf..8ab025d0035f 100644
--- a/drivers/gpu/drm/panfrost/panfrost_devfreq.c
+++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c
@@ -1,20 +1,29 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright 2019 Collabora ltd. */
+
+#include <linux/clk.h>
 #include <linux/devfreq.h>
 #include <linux/devfreq_cooling.h>
 #include <linux/platform_device.h>
 #include <linux/pm_opp.h>
-#include <linux/clk.h>
-#include <linux/regulator/consumer.h>
 
 #include "panfrost_device.h"
 #include "panfrost_devfreq.h"
-#include "panfrost_features.h"
-#include "panfrost_issues.h"
-#include "panfrost_gpu.h"
-#include "panfrost_regs.h"
 
-static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev);
+static void panfrost_devfreq_update_utilization(struct panfrost_devfreq *pfdevfreq)
+{
+	ktime_t now, last;
+
+	now = ktime_get();
+	last = pfdevfreq->time_last_update;
+
+	if (pfdevfreq->busy_count > 0)
+		pfdevfreq->busy_time += ktime_sub(now, last);
+	else
+		pfdevfreq->idle_time += ktime_sub(now, last);
+
+	pfdevfreq->time_last_update = now;
+}
 
 static int panfrost_devfreq_target(struct device *dev, unsigned long *freq,
 				   u32 flags)
@@ -34,30 +43,37 @@ static int panfrost_devfreq_target(struct device *dev, unsigned long *freq,
 	return 0;
 }
 
-static void panfrost_devfreq_reset(struct panfrost_device *pfdev)
+static void panfrost_devfreq_reset(struct panfrost_devfreq *pfdevfreq)
 {
-	pfdev->devfreq.busy_time = 0;
-	pfdev->devfreq.idle_time = 0;
-	pfdev->devfreq.time_last_update = ktime_get();
+	pfdevfreq->busy_time = 0;
+	pfdevfreq->idle_time = 0;
+	pfdevfreq->time_last_update = ktime_get();
 }
 
 static int panfrost_devfreq_get_dev_status(struct device *dev,
 					   struct devfreq_dev_status *status)
 {
 	struct panfrost_device *pfdev = dev_get_drvdata(dev);
-
-	panfrost_devfreq_update_utilization(pfdev);
+	struct panfrost_devfreq *pfdevfreq = &pfdev->pfdevfreq;
+	unsigned long irqflags;
 
 	status->current_frequency = clk_get_rate(pfdev->clock);
-	status->total_time = ktime_to_ns(ktime_add(pfdev->devfreq.busy_time,
-						   pfdev->devfreq.idle_time));
 
-	status->busy_time = ktime_to_ns(pfdev->devfreq.busy_time);
+	spin_lock_irqsave(&pfdevfreq->lock, irqflags);
+
+	panfrost_devfreq_update_utilization(pfdevfreq);
+
+	status->total_time = ktime_to_ns(ktime_add(pfdevfreq->busy_time,
+						   pfdevfreq->idle_time));
+
+	status->busy_time = ktime_to_ns(pfdevfreq->busy_time);
+
+	panfrost_devfreq_reset(pfdevfreq);
 
-	panfrost_devfreq_reset(pfdev);
+	spin_unlock_irqrestore(&pfdevfreq->lock, irqflags);
 
-	dev_dbg(pfdev->dev, "busy %lu total %lu %lu %% freq %lu MHz\n", status->busy_time,
-		status->total_time,
+	dev_dbg(pfdev->dev, "busy %lu total %lu %lu %% freq %lu MHz\n",
+		status->busy_time, status->total_time,
 		status->busy_time / (status->total_time / 100),
 		status->current_frequency / 1000 / 1000);
 
@@ -77,21 +93,43 @@ int panfrost_devfreq_init(struct panfrost_device *pfdev)
 	unsigned long cur_freq;
 	struct device *dev = &pfdev->pdev->dev;
 	struct devfreq *devfreq;
+	struct opp_table *opp_table;
 	struct thermal_cooling_device *cooling;
+	struct panfrost_devfreq *pfdevfreq = &pfdev->pfdevfreq;
+
+	opp_table = dev_pm_opp_set_regulators(dev, pfdev->comp->supply_names,
+					      pfdev->comp->num_supplies);
+	if (IS_ERR(opp_table)) {
+		ret = PTR_ERR(opp_table);
+		/* Continue if the optional regulator is missing */
+		if (ret != -ENODEV) {
+			DRM_DEV_ERROR(dev, "Couldn't set OPP regulators\n");
+			goto err_fini;
+		}
+	} else {
+		pfdevfreq->regulators_opp_table = opp_table;
+	}
 
 	ret = dev_pm_opp_of_add_table(dev);
-	if (ret == -ENODEV) /* Optional, continue without devfreq */
-		return 0;
-	else if (ret)
-		return ret;
+	if (ret) {
+		/* Optional, continue without devfreq */
+		if (ret == -ENODEV)
+			ret = 0;
+		goto err_fini;
+	}
+	pfdevfreq->opp_of_table_added = true;
+
+	spin_lock_init(&pfdevfreq->lock);
 
-	panfrost_devfreq_reset(pfdev);
+	panfrost_devfreq_reset(pfdevfreq);
 
 	cur_freq = clk_get_rate(pfdev->clock);
 
 	opp = devfreq_recommended_opp(dev, &cur_freq, 0);
-	if (IS_ERR(opp))
-		return PTR_ERR(opp);
+	if (IS_ERR(opp)) {
+		ret = PTR_ERR(opp);
+		goto err_fini;
+	}
 
 	panfrost_devfreq_profile.initial_freq = cur_freq;
 	dev_pm_opp_put(opp);
@@ -100,75 +138,94 @@ int panfrost_devfreq_init(struct panfrost_device *pfdev)
 					  DEVFREQ_GOV_SIMPLE_ONDEMAND, NULL);
 	if (IS_ERR(devfreq)) {
 		DRM_DEV_ERROR(dev, "Couldn't initialize GPU devfreq\n");
-		dev_pm_opp_of_remove_table(dev);
-		return PTR_ERR(devfreq);
+		ret = PTR_ERR(devfreq);
+		goto err_fini;
 	}
-	pfdev->devfreq.devfreq = devfreq;
+	pfdevfreq->devfreq = devfreq;
 
 	cooling = of_devfreq_cooling_register(dev->of_node, devfreq);
 	if (IS_ERR(cooling))
 		DRM_DEV_INFO(dev, "Failed to register cooling device\n");
 	else
-		pfdev->devfreq.cooling = cooling;
+		pfdevfreq->cooling = cooling;
 
 	return 0;
+
+err_fini:
+	panfrost_devfreq_fini(pfdev);
+	return ret;
 }
 
 void panfrost_devfreq_fini(struct panfrost_device *pfdev)
 {
-	if (pfdev->devfreq.cooling)
-		devfreq_cooling_unregister(pfdev->devfreq.cooling);
-	dev_pm_opp_of_remove_table(&pfdev->pdev->dev);
+	struct panfrost_devfreq *pfdevfreq = &pfdev->pfdevfreq;
+
+	if (pfdevfreq->cooling) {
+		devfreq_cooling_unregister(pfdevfreq->cooling);
+		pfdevfreq->cooling = NULL;
+	}
+
+	if (pfdevfreq->opp_of_table_added) {
+		dev_pm_opp_of_remove_table(&pfdev->pdev->dev);
+		pfdevfreq->opp_of_table_added = false;
+	}
+
+	if (pfdevfreq->regulators_opp_table) {
+		dev_pm_opp_put_regulators(pfdevfreq->regulators_opp_table);
+		pfdevfreq->regulators_opp_table = NULL;
+	}
 }
 
 void panfrost_devfreq_resume(struct panfrost_device *pfdev)
 {
-	if (!pfdev->devfreq.devfreq)
+	struct panfrost_devfreq *pfdevfreq = &pfdev->pfdevfreq;
+
+	if (!pfdevfreq->devfreq)
 		return;
 
-	panfrost_devfreq_reset(pfdev);
+	panfrost_devfreq_reset(pfdevfreq);
 
-	devfreq_resume_device(pfdev->devfreq.devfreq);
+	devfreq_resume_device(pfdevfreq->devfreq);
 }
 
 void panfrost_devfreq_suspend(struct panfrost_device *pfdev)
 {
-	if (!pfdev->devfreq.devfreq)
+	struct panfrost_devfreq *pfdevfreq = &pfdev->pfdevfreq;
+
+	if (!pfdevfreq->devfreq)
 		return;
 
-	devfreq_suspend_device(pfdev->devfreq.devfreq);
+	devfreq_suspend_device(pfdevfreq->devfreq);
 }
 
-static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev)
+void panfrost_devfreq_record_busy(struct panfrost_devfreq *pfdevfreq)
 {
-	ktime_t now;
-	ktime_t last;
+	unsigned long irqflags;
 
-	if (!pfdev->devfreq.devfreq)
+	if (!pfdevfreq->devfreq)
 		return;
 
-	now = ktime_get();
-	last = pfdev->devfreq.time_last_update;
+	spin_lock_irqsave(&pfdevfreq->lock, irqflags);
 
-	if (atomic_read(&pfdev->devfreq.busy_count) > 0)
-		pfdev->devfreq.busy_time += ktime_sub(now, last);
-	else
-		pfdev->devfreq.idle_time += ktime_sub(now, last);
+	panfrost_devfreq_update_utilization(pfdevfreq);
 
-	pfdev->devfreq.time_last_update = now;
-}
+	pfdevfreq->busy_count++;
 
-void panfrost_devfreq_record_busy(struct panfrost_device *pfdev)
-{
-	panfrost_devfreq_update_utilization(pfdev);
-	atomic_inc(&pfdev->devfreq.busy_count);
+	spin_unlock_irqrestore(&pfdevfreq->lock, irqflags);
 }
 
-void panfrost_devfreq_record_idle(struct panfrost_device *pfdev)
+void panfrost_devfreq_record_idle(struct panfrost_devfreq *pfdevfreq)
 {
-	int count;
+	unsigned long irqflags;
+
+	if (!pfdevfreq->devfreq)
+		return;
+
+	spin_lock_irqsave(&pfdevfreq->lock, irqflags);
+
+	panfrost_devfreq_update_utilization(pfdevfreq);
+
+	WARN_ON(--pfdevfreq->busy_count < 0);
 
-	panfrost_devfreq_update_utilization(pfdev);
-	count = atomic_dec_if_positive(&pfdev->devfreq.busy_count);
-	WARN_ON(count < 0);
+	spin_unlock_irqrestore(&pfdevfreq->lock, irqflags);
 }
diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.h b/drivers/gpu/drm/panfrost/panfrost_devfreq.h
index 0611beffc8d0..db6ea48e21f9 100644
--- a/drivers/gpu/drm/panfrost/panfrost_devfreq.h
+++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.h
@@ -4,13 +4,39 @@
 #ifndef __PANFROST_DEVFREQ_H__
 #define __PANFROST_DEVFREQ_H__
 
+#include <linux/spinlock.h>
+#include <linux/ktime.h>
+
+struct devfreq;
+struct opp_table;
+struct thermal_cooling_device;
+
+struct panfrost_device;
+
+struct panfrost_devfreq {
+	struct devfreq *devfreq;
+	struct opp_table *regulators_opp_table;
+	struct thermal_cooling_device *cooling;
+	bool opp_of_table_added;
+
+	ktime_t busy_time;
+	ktime_t idle_time;
+	ktime_t time_last_update;
+	int busy_count;
+	/*
+	 * Protect busy_time, idle_time, time_last_update and busy_count
+	 * because these can be updated concurrently between multiple jobs.
+	 */
+	spinlock_t lock;
+};
+
 int panfrost_devfreq_init(struct panfrost_device *pfdev);
 void panfrost_devfreq_fini(struct panfrost_device *pfdev);
 
 void panfrost_devfreq_resume(struct panfrost_device *pfdev);
 void panfrost_devfreq_suspend(struct panfrost_device *pfdev);
 
-void panfrost_devfreq_record_busy(struct panfrost_device *pfdev);
-void panfrost_devfreq_record_idle(struct panfrost_device *pfdev);
+void panfrost_devfreq_record_busy(struct panfrost_devfreq *devfreq);
+void panfrost_devfreq_record_idle(struct panfrost_devfreq *devfreq);
 
 #endif /* __PANFROST_DEVFREQ_H__ */
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c
index b172087eee6a..e6896733838a 100644
--- a/drivers/gpu/drm/panfrost/panfrost_device.c
+++ b/drivers/gpu/drm/panfrost/panfrost_device.c
@@ -90,9 +90,11 @@ static int panfrost_regulator_init(struct panfrost_device *pfdev)
 {
 	int ret, i;
 
-	if (WARN(pfdev->comp->num_supplies > ARRAY_SIZE(pfdev->regulators),
-			"Too many supplies in compatible structure.\n"))
-		return -EINVAL;
+	pfdev->regulators = devm_kcalloc(pfdev->dev, pfdev->comp->num_supplies,
+					 sizeof(*pfdev->regulators),
+					 GFP_KERNEL);
+	if (!pfdev->regulators)
+		return -ENOMEM;
 
 	for (i = 0; i < pfdev->comp->num_supplies; i++)
 		pfdev->regulators[i].supply = pfdev->comp->supply_names[i];
@@ -119,8 +121,10 @@ static int panfrost_regulator_init(struct panfrost_device *pfdev)
 
 static void panfrost_regulator_fini(struct panfrost_device *pfdev)
 {
-	regulator_bulk_disable(pfdev->comp->num_supplies,
-			pfdev->regulators);
+	if (!pfdev->regulators)
+		return;
+
+	regulator_bulk_disable(pfdev->comp->num_supplies, pfdev->regulators);
 }
 
 static void panfrost_pm_domain_fini(struct panfrost_device *pfdev)
@@ -214,58 +218,70 @@ int panfrost_device_init(struct panfrost_device *pfdev)
 		return err;
 	}
 
-	err = panfrost_regulator_init(pfdev);
-	if (err)
-		goto err_out0;
+	err = panfrost_devfreq_init(pfdev);
+	if (err) {
+		if (err != -EPROBE_DEFER)
+			dev_err(pfdev->dev, "devfreq init failed %d\n", err);
+		goto out_clk;
+	}
+
+	/* OPP will handle regulators */
+	if (!pfdev->pfdevfreq.opp_of_table_added) {
+		err = panfrost_regulator_init(pfdev);
+		if (err)
+			goto out_devfreq;
+	}
 
 	err = panfrost_reset_init(pfdev);
 	if (err) {
 		dev_err(pfdev->dev, "reset init failed %d\n", err);
-		goto err_out1;
+		goto out_regulator;
 	}
 
 	err = panfrost_pm_domain_init(pfdev);
 	if (err)
-		goto err_out2;
+		goto out_reset;
 
 	res = platform_get_resource(pfdev->pdev, IORESOURCE_MEM, 0);
 	pfdev->iomem = devm_ioremap_resource(pfdev->dev, res);
 	if (IS_ERR(pfdev->iomem)) {
 		dev_err(pfdev->dev, "failed to ioremap iomem\n");
 		err = PTR_ERR(pfdev->iomem);
-		goto err_out3;
+		goto out_pm_domain;
 	}
 
 	err = panfrost_gpu_init(pfdev);
 	if (err)
-		goto err_out3;
+		goto out_pm_domain;
 
 	err = panfrost_mmu_init(pfdev);
 	if (err)
-		goto err_out4;
+		goto out_gpu;
 
 	err = panfrost_job_init(pfdev);
 	if (err)
-		goto err_out5;
+		goto out_mmu;
 
 	err = panfrost_perfcnt_init(pfdev);
 	if (err)
-		goto err_out6;
+		goto out_job;
 
 	return 0;
-err_out6:
+out_job:
 	panfrost_job_fini(pfdev);
-err_out5:
+out_mmu:
 	panfrost_mmu_fini(pfdev);
-err_out4:
+out_gpu:
 	panfrost_gpu_fini(pfdev);
-err_out3:
+out_pm_domain:
 	panfrost_pm_domain_fini(pfdev);
-err_out2:
+out_reset:
 	panfrost_reset_fini(pfdev);
-err_out1:
+out_regulator:
 	panfrost_regulator_fini(pfdev);
-err_out0:
+out_devfreq:
+	panfrost_devfreq_fini(pfdev);
+out_clk:
 	panfrost_clk_fini(pfdev);
 	return err;
 }
@@ -278,6 +294,7 @@ void panfrost_device_fini(struct panfrost_device *pfdev)
 	panfrost_gpu_fini(pfdev);
 	panfrost_pm_domain_fini(pfdev);
 	panfrost_reset_fini(pfdev);
+	panfrost_devfreq_fini(pfdev);
 	panfrost_regulator_fini(pfdev);
 	panfrost_clk_fini(pfdev);
 }
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h
index c30c719a8059..2e9cbd1c4a58 100644
--- a/drivers/gpu/drm/panfrost/panfrost_device.h
+++ b/drivers/gpu/drm/panfrost/panfrost_device.h
@@ -13,6 +13,8 @@
 #include <drm/drm_mm.h>
 #include <drm/gpu_scheduler.h>
 
+#include "panfrost_devfreq.h"
+
 struct panfrost_device;
 struct panfrost_mmu;
 struct panfrost_job_slot;
@@ -20,7 +22,6 @@ struct panfrost_job;
 struct panfrost_perfcnt;
 
 #define NUM_JOB_SLOTS 3
-#define MAX_REGULATORS 2
 #define MAX_PM_DOMAINS 3
 
 struct panfrost_features {
@@ -69,6 +70,9 @@ struct panfrost_compatible {
 	int num_pm_domains;
 	/* Only required if num_pm_domains > 1. */
 	const char * const *pm_domain_names;
+
+	/* Vendor implementation quirks callback */
+	void (*vendor_quirk)(struct panfrost_device *pfdev);
 };
 
 struct panfrost_device {
@@ -79,7 +83,7 @@ struct panfrost_device {
 	void __iomem *iomem;
 	struct clk *clock;
 	struct clk *bus_clock;
-	struct regulator_bulk_data regulators[MAX_REGULATORS];
+	struct regulator_bulk_data *regulators;
 	struct reset_control *rstc;
 	/* pm_domains for devices with more than one. */
 	struct device *pm_domain_devs[MAX_PM_DOMAINS];
@@ -107,14 +111,7 @@ struct panfrost_device {
 	struct list_head shrinker_list;
 	struct shrinker shrinker;
 
-	struct {
-		struct devfreq *devfreq;
-		struct thermal_cooling_device *cooling;
-		ktime_t busy_time;
-		ktime_t idle_time;
-		ktime_t time_last_update;
-		atomic_t busy_count;
-	} devfreq;
+	struct panfrost_devfreq pfdevfreq;
 };
 
 struct panfrost_mmu {
diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
index ada51df9a7a3..37d4cb7a5491 100644
--- a/drivers/gpu/drm/panfrost/panfrost_drv.c
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -14,7 +14,6 @@
 #include <drm/drm_utils.h>
 
 #include "panfrost_device.h"
-#include "panfrost_devfreq.h"
 #include "panfrost_gem.h"
 #include "panfrost_mmu.h"
 #include "panfrost_job.h"
@@ -606,13 +605,6 @@ static int panfrost_probe(struct platform_device *pdev)
 		goto err_out0;
 	}
 
-	err = panfrost_devfreq_init(pfdev);
-	if (err) {
-		if (err != -EPROBE_DEFER)
-			dev_err(&pdev->dev, "Fatal error during devfreq init\n");
-		goto err_out1;
-	}
-
 	pm_runtime_set_active(pfdev->dev);
 	pm_runtime_mark_last_busy(pfdev->dev);
 	pm_runtime_enable(pfdev->dev);
@@ -625,16 +617,14 @@ static int panfrost_probe(struct platform_device *pdev)
 	 */
 	err = drm_dev_register(ddev, 0);
 	if (err < 0)
-		goto err_out2;
+		goto err_out1;
 
 	panfrost_gem_shrinker_init(ddev);
 
 	return 0;
 
-err_out2:
-	pm_runtime_disable(pfdev->dev);
-	panfrost_devfreq_fini(pfdev);
 err_out1:
+	pm_runtime_disable(pfdev->dev);
 	panfrost_device_fini(pfdev);
 err_out0:
 	drm_dev_put(ddev);
@@ -650,7 +640,6 @@ static int panfrost_remove(struct platform_device *pdev)
 	panfrost_gem_shrinker_cleanup(ddev);
 
 	pm_runtime_get_sync(pfdev->dev);
-	panfrost_devfreq_fini(pfdev);
 	panfrost_device_fini(pfdev);
 	pm_runtime_put_sync_suspend(pfdev->dev);
 	pm_runtime_disable(pfdev->dev);
@@ -667,7 +656,18 @@ static const struct panfrost_compatible default_data = {
 	.pm_domain_names = NULL,
 };
 
+static const struct panfrost_compatible amlogic_data = {
+	.num_supplies = ARRAY_SIZE(default_supplies),
+	.supply_names = default_supplies,
+	.vendor_quirk = panfrost_gpu_amlogic_quirk,
+};
+
 static const struct of_device_id dt_match[] = {
+	/* Set first to probe before the generic compatibles */
+	{ .compatible = "amlogic,meson-gxm-mali",
+	  .data = &amlogic_data, },
+	{ .compatible = "amlogic,meson-g12a-mali",
+	  .data = &amlogic_data, },
 	{ .compatible = "arm,mali-t604", .data = &default_data, },
 	{ .compatible = "arm,mali-t624", .data = &default_data, },
 	{ .compatible = "arm,mali-t628", .data = &default_data, },
@@ -677,6 +677,7 @@ static const struct of_device_id dt_match[] = {
 	{ .compatible = "arm,mali-t830", .data = &default_data, },
 	{ .compatible = "arm,mali-t860", .data = &default_data, },
 	{ .compatible = "arm,mali-t880", .data = &default_data, },
+	{ .compatible = "arm,mali-bifrost", .data = &default_data, },
 	{}
 };
 MODULE_DEVICE_TABLE(of, dt_match);
diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c
index 33355dd302f1..1a6cea0e0bd7 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gem.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gem.c
@@ -41,8 +41,8 @@ static void panfrost_gem_free_object(struct drm_gem_object *obj)
 
 		for (i = 0; i < n_sgt; i++) {
 			if (bo->sgts[i].sgl) {
-				dma_unmap_sg(pfdev->dev, bo->sgts[i].sgl,
-					     bo->sgts[i].nents, DMA_BIDIRECTIONAL);
+				dma_unmap_sgtable(pfdev->dev, &bo->sgts[i],
+						  DMA_BIDIRECTIONAL, 0);
 				sg_free_table(&bo->sgts[i]);
 			}
 		}
diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c
index f2c1ddc41a9b..2aae636f1cf5 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gpu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c
@@ -10,6 +10,7 @@
 #include <linux/io.h>
 #include <linux/iopoll.h>
 #include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 
 #include "panfrost_device.h"
 #include "panfrost_features.h"
@@ -75,6 +76,17 @@ int panfrost_gpu_soft_reset(struct panfrost_device *pfdev)
 	return 0;
 }
 
+void panfrost_gpu_amlogic_quirk(struct panfrost_device *pfdev)
+{
+	/*
+	 * The Amlogic integrated Mali-T820, Mali-G31 & Mali-G52 needs
+	 * these undocumented bits in GPU_PWR_OVERRIDE1 to be set in order
+	 * to operate correctly.
+	 */
+	gpu_write(pfdev, GPU_PWR_KEY, GPU_PWR_KEY_UNLOCK);
+	gpu_write(pfdev, GPU_PWR_OVERRIDE1, 0xfff | (0x20 << 16));
+}
+
 static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev)
 {
 	u32 quirks = 0;
@@ -135,6 +147,10 @@ static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev)
 
 	if (quirks)
 		gpu_write(pfdev, GPU_JM_CONFIG, quirks);
+
+	/* Here goes platform specific quirks */
+	if (pfdev->comp->vendor_quirk)
+		pfdev->comp->vendor_quirk(pfdev);
 }
 
 #define MAX_HW_REVS 6
@@ -304,16 +320,18 @@ void panfrost_gpu_power_on(struct panfrost_device *pfdev)
 	int ret;
 	u32 val;
 
+	panfrost_gpu_init_quirks(pfdev);
+
 	/* Just turn on everything for now */
 	gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present);
 	ret = readl_relaxed_poll_timeout(pfdev->iomem + L2_READY_LO,
-		val, val == pfdev->features.l2_present, 100, 1000);
+		val, val == pfdev->features.l2_present, 100, 20000);
 	if (ret)
 		dev_err(pfdev->dev, "error powering up gpu L2");
 
 	gpu_write(pfdev, SHADER_PWRON_LO, pfdev->features.shader_present);
 	ret = readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO,
-		val, val == pfdev->features.shader_present, 100, 1000);
+		val, val == pfdev->features.shader_present, 100, 20000);
 	if (ret)
 		dev_err(pfdev->dev, "error powering up gpu shader");
 
@@ -343,6 +361,7 @@ int panfrost_gpu_init(struct panfrost_device *pfdev)
 
 	dma_set_mask_and_coherent(pfdev->dev,
 		DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features)));
+	dma_set_max_seg_size(pfdev->dev, UINT_MAX);
 
 	irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "gpu");
 	if (irq <= 0)
@@ -355,7 +374,6 @@ int panfrost_gpu_init(struct panfrost_device *pfdev)
 		return err;
 	}
 
-	panfrost_gpu_init_quirks(pfdev);
 	panfrost_gpu_power_on(pfdev);
 
 	return 0;
@@ -368,7 +386,16 @@ void panfrost_gpu_fini(struct panfrost_device *pfdev)
 
 u32 panfrost_gpu_get_latest_flush_id(struct panfrost_device *pfdev)
 {
-	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION))
-		return gpu_read(pfdev, GPU_LATEST_FLUSH_ID);
+	u32 flush_id;
+
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION)) {
+		/* Flush reduction only makes sense when the GPU is kept powered on between jobs */
+		if (pm_runtime_get_if_in_use(pfdev->dev)) {
+			flush_id = gpu_read(pfdev, GPU_LATEST_FLUSH_ID);
+			pm_runtime_put(pfdev->dev);
+			return flush_id;
+		}
+	}
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.h b/drivers/gpu/drm/panfrost/panfrost_gpu.h
index 4112412087b2..468c51e7e46d 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gpu.h
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.h
@@ -16,4 +16,6 @@ int panfrost_gpu_soft_reset(struct panfrost_device *pfdev);
 void panfrost_gpu_power_on(struct panfrost_device *pfdev);
 void panfrost_gpu_power_off(struct panfrost_device *pfdev);
 
+void panfrost_gpu_amlogic_quirk(struct panfrost_device *pfdev);
+
 #endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c
index 360146f6f3d9..30e7b7196dab 100644
--- a/drivers/gpu/drm/panfrost/panfrost_job.c
+++ b/drivers/gpu/drm/panfrost/panfrost_job.c
@@ -145,7 +145,7 @@ static void panfrost_job_hw_submit(struct panfrost_job *job, int js)
 	u64 jc_head = job->jc;
 	int ret;
 
-	panfrost_devfreq_record_busy(pfdev);
+	panfrost_devfreq_record_busy(&pfdev->pfdevfreq);
 
 	ret = pm_runtime_get_sync(pfdev->dev);
 	if (ret < 0)
@@ -410,7 +410,7 @@ static void panfrost_job_timedout(struct drm_sched_job *sched_job)
 	for (i = 0; i < NUM_JOB_SLOTS; i++) {
 		if (pfdev->jobs[i]) {
 			pm_runtime_put_noidle(pfdev->dev);
-			panfrost_devfreq_record_idle(pfdev);
+			panfrost_devfreq_record_idle(&pfdev->pfdevfreq);
 			pfdev->jobs[i] = NULL;
 		}
 	}
@@ -478,7 +478,7 @@ static irqreturn_t panfrost_job_irq_handler(int irq, void *data)
 				pfdev->jobs[j] = NULL;
 
 				panfrost_mmu_as_put(pfdev, &job->file_priv->mmu);
-				panfrost_devfreq_record_idle(pfdev);
+				panfrost_devfreq_record_idle(&pfdev->pfdevfreq);
 
 				dma_fence_signal_locked(job->done_fence);
 				pm_runtime_put_autosuspend(pfdev->dev);
@@ -581,10 +581,6 @@ int panfrost_job_is_idle(struct panfrost_device *pfdev)
 	struct panfrost_job_slot *js = pfdev->js;
 	int i;
 
-	/* Check whether the hardware is idle */
-	if (atomic_read(&pfdev->devfreq.busy_count))
-		return false;
-
 	for (i = 0; i < NUM_JOB_SLOTS; i++) {
 		/* If there are any jobs in the HW queue, we're not idle */
 		if (atomic_read(&js->queue[i].sched.hw_rq_count))
diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
index e8f7b11352d2..776448c527ea 100644
--- a/drivers/gpu/drm/panfrost/panfrost_mmu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
@@ -253,7 +253,7 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu,
 	struct io_pgtable_ops *ops = mmu->pgtbl_ops;
 	u64 start_iova = iova;
 
-	for_each_sg(sgt->sgl, sgl, sgt->nents, count) {
+	for_each_sgtable_dma_sg(sgt, sgl, count) {
 		unsigned long paddr = sg_dma_address(sgl);
 		size_t len = sg_dma_len(sgl);
 
@@ -517,10 +517,9 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,
 	if (ret)
 		goto err_pages;
 
-	if (!dma_map_sg(pfdev->dev, sgt->sgl, sgt->nents, DMA_BIDIRECTIONAL)) {
-		ret = -EINVAL;
+	ret = dma_map_sgtable(pfdev->dev, sgt, DMA_BIDIRECTIONAL, 0);
+	if (ret)
 		goto err_map;
-	}
 
 	mmu_map_sg(pfdev, bomapping->mmu, addr,
 		   IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt);
diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
index ec4695cf3caf..fdbc8d949135 100644
--- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
+++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
@@ -83,11 +83,13 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
 
 	ret = pm_runtime_get_sync(pfdev->dev);
 	if (ret < 0)
-		return ret;
+		goto err_put_pm;
 
 	bo = drm_gem_shmem_create(pfdev->ddev, perfcnt->bosize);
-	if (IS_ERR(bo))
-		return PTR_ERR(bo);
+	if (IS_ERR(bo)) {
+		ret = PTR_ERR(bo);
+		goto err_put_pm;
+	}
 
 	/* Map the perfcnt buf in the address space attached to file_priv. */
 	ret = panfrost_gem_open(&bo->base, file_priv);
@@ -168,6 +170,8 @@ err_close_bo:
 	panfrost_gem_close(&bo->base, file_priv);
 err_put_bo:
 	drm_gem_object_put(&bo->base);
+err_put_pm:
+	pm_runtime_put(pfdev->dev);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/panfrost/panfrost_regs.h b/drivers/gpu/drm/panfrost/panfrost_regs.h
index ea38ac60581c..eddaa62ad8b0 100644
--- a/drivers/gpu/drm/panfrost/panfrost_regs.h
+++ b/drivers/gpu/drm/panfrost/panfrost_regs.h
@@ -51,6 +51,10 @@
 #define GPU_STATUS			0x34
 #define   GPU_STATUS_PRFCNT_ACTIVE	BIT(2)
 #define GPU_LATEST_FLUSH_ID		0x38
+#define GPU_PWR_KEY			0x50	/* (WO) Power manager key register */
+#define  GPU_PWR_KEY_UNLOCK		0x2968A819
+#define GPU_PWR_OVERRIDE0		0x54	/* (RW) Power manager override settings */
+#define GPU_PWR_OVERRIDE1		0x58	/* (RW) Power manager override settings */
 #define GPU_FAULT_STATUS		0x3C
 #define GPU_FAULT_ADDRESS_LO		0x40
 #define GPU_FAULT_ADDRESS_HI		0x44
diff --git a/drivers/gpu/drm/pl111/pl111_drv.c b/drivers/gpu/drm/pl111/pl111_drv.c
index 96e58fda75d8..46b0d1c4a16c 100644
--- a/drivers/gpu/drm/pl111/pl111_drv.c
+++ b/drivers/gpu/drm/pl111/pl111_drv.c
@@ -10,18 +10,11 @@
  */
 
 /**
- * DOC: ARM PrimeCell PL111 CLCD Driver
+ * DOC: ARM PrimeCell PL110 and PL111 CLCD Driver
  *
- * The PL111 is a simple LCD controller that can support TFT and STN
- * displays.  This driver exposes a standard KMS interface for them.
- *
- * This driver uses the same Device Tree binding as the fbdev CLCD
- * driver.  While the fbdev driver supports panels that may be
- * connected to the CLCD internally to the CLCD driver, in DRM the
- * panels get split out to drivers/gpu/drm/panels/.  This means that,
- * in converting from using fbdev to using DRM, you also need to write
- * a panel driver (which may be as simple as an entry in
- * panel-simple.c).
+ * The PL110/PL111 is a simple LCD controller that can support TFT
+ * and STN displays. This driver exposes a standard KMS interface
+ * for them.
  *
  * The driver currently doesn't expose the cursor.  The DRM API for
  * cursors requires support for 64x64 ARGB8888 cursor images, while
@@ -29,16 +22,13 @@
  * cursors.  While one could imagine trying to hack something together
  * to look at the ARGB8888 and program reasonable in monochrome, we
  * just don't expose the cursor at all instead, and leave cursor
- * support to the X11 software cursor layer.
+ * support to the application software cursor layer.
  *
  * TODO:
  *
  * - Fix race between setting plane base address and getting IRQ for
  *   vsync firing the pageflip completion.
  *
- * - Use the "max-memory-bandwidth" DT property to filter the
- *   supported formats.
- *
  * - Read back hardware state at boot to skip reprogramming the
  *   hardware when doing a no-op modeset.
  *
diff --git a/drivers/gpu/drm/qxl/qxl_cmd.c b/drivers/gpu/drm/qxl/qxl_cmd.c
index 798f9dd7ad75..54e3c3a97440 100644
--- a/drivers/gpu/drm/qxl/qxl_cmd.c
+++ b/drivers/gpu/drm/qxl/qxl_cmd.c
@@ -588,7 +588,7 @@ static int qxl_reap_surf(struct qxl_device *qdev, struct qxl_bo *surf, bool stal
 {
 	int ret;
 
-	ret = qxl_bo_reserve(surf, false);
+	ret = qxl_bo_reserve(surf);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 099dca48b0ff..6063f3a15329 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -26,6 +26,7 @@
 #include <linux/crc32.h>
 #include <linux/delay.h>
 
+#include <drm/drm_drv.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
@@ -162,7 +163,8 @@ static void qxl_update_offset_props(struct qxl_device *qdev)
 void qxl_display_read_client_monitors_config(struct qxl_device *qdev)
 {
 	struct drm_device *dev = &qdev->ddev;
-	int status, retries;
+	struct drm_modeset_acquire_ctx ctx;
+	int status, retries, ret;
 
 	for (retries = 0; retries < 10; retries++) {
 		status = qxl_display_copy_rom_client_monitors_config(qdev);
@@ -183,9 +185,9 @@ void qxl_display_read_client_monitors_config(struct qxl_device *qdev)
 		return;
 	}
 
-	drm_modeset_lock_all(dev);
+	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE, ret);
 	qxl_update_offset_props(qdev);
-	drm_modeset_unlock_all(dev);
+	DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
 	if (!drm_helper_hpd_irq_event(dev)) {
 		/* notify that the monitor configuration changed, to
 		   adjust at the arbitrary resolution */
@@ -403,18 +405,17 @@ static int qxl_framebuffer_surface_dirty(struct drm_framebuffer *fb,
 	struct qxl_device *qdev = to_qxl(fb->dev);
 	struct drm_clip_rect norect;
 	struct qxl_bo *qobj;
+	struct drm_modeset_acquire_ctx ctx;
 	bool is_primary;
-	int inc = 1;
+	int inc = 1, ret;
 
-	drm_modeset_lock_all(fb->dev);
+	DRM_MODESET_LOCK_ALL_BEGIN(fb->dev, ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE, ret);
 
 	qobj = gem_to_qxl_bo(fb->obj[0]);
 	/* if we aren't primary surface ignore this */
 	is_primary = qobj->shadow ? qobj->shadow->is_primary : qobj->is_primary;
-	if (!is_primary) {
-		drm_modeset_unlock_all(fb->dev);
-		return 0;
-	}
+	if (!is_primary)
+		goto out_lock_end;
 
 	if (!num_clips) {
 		num_clips = 1;
@@ -430,7 +431,8 @@ static int qxl_framebuffer_surface_dirty(struct drm_framebuffer *fb,
 	qxl_draw_dirty_fb(qdev, fb, qobj, flags, color,
 			  clips, num_clips, inc, 0);
 
-	drm_modeset_unlock_all(fb->dev);
+out_lock_end:
+	DRM_MODESET_LOCK_ALL_END(fb->dev, ctx, ret);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index 13872b882775..6e7f16f4cec7 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -96,7 +96,7 @@ qxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto disable_pci;
 
-	if (is_vga(pdev)) {
+	if (is_vga(pdev) && pdev->revision < 5) {
 		ret = vga_get_interruptible(pdev, VGA_RSRC_LEGACY_IO);
 		if (ret) {
 			DRM_ERROR("can't get legacy vga ioports\n");
@@ -127,7 +127,7 @@ modeset_cleanup:
 unload:
 	qxl_device_fini(qdev);
 put_vga:
-	if (is_vga(pdev))
+	if (is_vga(pdev) && pdev->revision < 5)
 		vga_put(pdev, VGA_RSRC_LEGACY_IO);
 disable_pci:
 	pci_disable_device(pdev);
@@ -155,7 +155,7 @@ qxl_pci_remove(struct pci_dev *pdev)
 
 	drm_dev_unregister(dev);
 	drm_atomic_helper_shutdown(dev);
-	if (is_vga(pdev))
+	if (is_vga(pdev) && pdev->revision < 5)
 		vga_put(pdev, VGA_RSRC_LEGACY_IO);
 }
 
diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h
index 9691449aefdb..aae90a9ee1db 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.h
+++ b/drivers/gpu/drm/qxl/qxl_drv.h
@@ -350,7 +350,7 @@ int qxl_mode_dumb_mmap(struct drm_file *filp,
 int qxl_ttm_init(struct qxl_device *qdev);
 void qxl_ttm_fini(struct qxl_device *qdev);
 int qxl_ttm_io_mem_reserve(struct ttm_bo_device *bdev,
-			   struct ttm_mem_reg *mem);
+			   struct ttm_resource *mem);
 
 /* qxl image */
 
diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c
index 8f605d5cc149..5cea6eea72ab 100644
--- a/drivers/gpu/drm/qxl/qxl_ioctl.c
+++ b/drivers/gpu/drm/qxl/qxl_ioctl.c
@@ -322,7 +322,7 @@ static int qxl_update_area_ioctl(struct drm_device *dev, void *data,
 
 	qobj = gem_to_qxl_bo(gobj);
 
-	ret = qxl_bo_reserve(qobj, false);
+	ret = qxl_bo_reserve(qobj);
 	if (ret)
 		goto out;
 
diff --git a/drivers/gpu/drm/qxl/qxl_object.c b/drivers/gpu/drm/qxl/qxl_object.c
index 80e7a17aaddd..2bc364412e8b 100644
--- a/drivers/gpu/drm/qxl/qxl_object.c
+++ b/drivers/gpu/drm/qxl/qxl_object.c
@@ -64,16 +64,24 @@ void qxl_ttm_placement_from_domain(struct qxl_bo *qbo, u32 domain, bool pinned)
 
 	qbo->placement.placement = qbo->placements;
 	qbo->placement.busy_placement = qbo->placements;
-	if (domain == QXL_GEM_DOMAIN_VRAM)
-		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_VRAM | pflag;
+	if (domain == QXL_GEM_DOMAIN_VRAM) {
+		qbo->placements[c].mem_type = TTM_PL_VRAM;
+		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | pflag;
+	}
 	if (domain == QXL_GEM_DOMAIN_SURFACE) {
-		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_PRIV | pflag;
-		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_VRAM | pflag;
+		qbo->placements[c].mem_type = TTM_PL_PRIV;
+		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | pflag;
+		qbo->placements[c].mem_type = TTM_PL_VRAM;
+		qbo->placements[c++].flags = TTM_PL_FLAG_CACHED | pflag;
+	}
+	if (domain == QXL_GEM_DOMAIN_CPU) {
+		qbo->placements[c].mem_type = TTM_PL_SYSTEM;
+		qbo->placements[c++].flags = TTM_PL_MASK_CACHING | pflag;
+	}
+	if (!c) {
+		qbo->placements[c].mem_type = TTM_PL_SYSTEM;
+		qbo->placements[c++].flags = TTM_PL_MASK_CACHING;
 	}
-	if (domain == QXL_GEM_DOMAIN_CPU)
-		qbo->placements[c++].flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM | pflag;
-	if (!c)
-		qbo->placements[c++].flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM;
 	qbo->placement.num_placement = c;
 	qbo->placement.num_busy_placement = c;
 	for (i = 0; i < c; ++i) {
@@ -167,6 +175,7 @@ int qxl_bo_kmap(struct qxl_bo *bo, void **ptr)
 void *qxl_bo_kmap_atomic_page(struct qxl_device *qdev,
 			      struct qxl_bo *bo, int page_offset)
 {
+	unsigned long offset;
 	void *rptr;
 	int ret;
 	struct io_mapping *map;
@@ -178,9 +187,8 @@ void *qxl_bo_kmap_atomic_page(struct qxl_device *qdev,
 	else
 		goto fallback;
 
-	ret = qxl_ttm_io_mem_reserve(bo->tbo.bdev, &bo->tbo.mem);
-
-	return io_mapping_map_atomic_wc(map, bo->tbo.mem.bus.offset + page_offset);
+	offset = bo->tbo.mem.start << PAGE_SHIFT;
+	return io_mapping_map_atomic_wc(map, offset + page_offset);
 fallback:
 	if (bo->kptr) {
 		rptr = bo->kptr + (page_offset * PAGE_SIZE);
@@ -284,7 +292,7 @@ int qxl_bo_pin(struct qxl_bo *bo)
 {
 	int r;
 
-	r = qxl_bo_reserve(bo, false);
+	r = qxl_bo_reserve(bo);
 	if (r)
 		return r;
 
@@ -302,7 +310,7 @@ int qxl_bo_unpin(struct qxl_bo *bo)
 {
 	int r;
 
-	r = qxl_bo_reserve(bo, false);
+	r = qxl_bo_reserve(bo);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/qxl/qxl_object.h b/drivers/gpu/drm/qxl/qxl_object.h
index 21fa81048f4f..6b434e5ef795 100644
--- a/drivers/gpu/drm/qxl/qxl_object.h
+++ b/drivers/gpu/drm/qxl/qxl_object.h
@@ -27,11 +27,11 @@
 
 #include "qxl_drv.h"
 
-static inline int qxl_bo_reserve(struct qxl_bo *bo, bool no_wait)
+static inline int qxl_bo_reserve(struct qxl_bo *bo)
 {
 	int r;
 
-	r = ttm_bo_reserve(&bo->tbo, true, no_wait, NULL);
+	r = ttm_bo_reserve(&bo->tbo, true, false, NULL);
 	if (unlikely(r != 0)) {
 		if (r != -ERESTARTSYS) {
 			struct drm_device *ddev = bo->tbo.base.dev;
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index bf9dc451583a..fd691fff8394 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -48,31 +48,6 @@ static struct qxl_device *qxl_get_qdev(struct ttm_bo_device *bdev)
 	return qdev;
 }
 
-static int qxl_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-			     struct ttm_mem_type_manager *man)
-{
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		/* System memory */
-		man->flags = 0;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_VRAM:
-	case TTM_PL_PRIV:
-		/* "On-card" video ram */
-		man->func = &ttm_bo_manager_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	default:
-		DRM_ERROR("Unsupported memory type %u\n", (unsigned int)type);
-		return -EINVAL;
-	}
-	return 0;
-}
-
 static void qxl_evict_flags(struct ttm_buffer_object *bo,
 				struct ttm_placement *placement)
 {
@@ -80,7 +55,8 @@ static void qxl_evict_flags(struct ttm_buffer_object *bo,
 	static const struct ttm_place placements = {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_MASK_CACHING
 	};
 
 	if (!qxl_ttm_bo_is_qxl_bo(bo)) {
@@ -96,29 +72,22 @@ static void qxl_evict_flags(struct ttm_buffer_object *bo,
 }
 
 int qxl_ttm_io_mem_reserve(struct ttm_bo_device *bdev,
-			   struct ttm_mem_reg *mem)
+			   struct ttm_resource *mem)
 {
 	struct qxl_device *qdev = qxl_get_qdev(bdev);
 
-	mem->bus.addr = NULL;
-	mem->bus.offset = 0;
-	mem->bus.size = mem->num_pages << PAGE_SHIFT;
-	mem->bus.base = 0;
-	mem->bus.is_iomem = false;
-
 	switch (mem->mem_type) {
 	case TTM_PL_SYSTEM:
 		/* system memory */
 		return 0;
 	case TTM_PL_VRAM:
 		mem->bus.is_iomem = true;
-		mem->bus.base = qdev->vram_base;
-		mem->bus.offset = mem->start << PAGE_SHIFT;
+		mem->bus.offset = (mem->start << PAGE_SHIFT) + qdev->vram_base;
 		break;
 	case TTM_PL_PRIV:
 		mem->bus.is_iomem = true;
-		mem->bus.base = qdev->surfaceram_base;
-		mem->bus.offset = mem->start << PAGE_SHIFT;
+		mem->bus.offset = (mem->start << PAGE_SHIFT) +
+			qdev->surfaceram_base;
 		break;
 	default:
 		return -EINVAL;
@@ -135,8 +104,9 @@ struct qxl_ttm_tt {
 	u64				offset;
 };
 
-static int qxl_ttm_backend_bind(struct ttm_tt *ttm,
-				struct ttm_mem_reg *bo_mem)
+static int qxl_ttm_backend_bind(struct ttm_bo_device *bdev,
+				struct ttm_tt *ttm,
+				struct ttm_resource *bo_mem)
 {
 	struct qxl_ttm_tt *gtt = (void *)ttm;
 
@@ -149,25 +119,22 @@ static int qxl_ttm_backend_bind(struct ttm_tt *ttm,
 	return -1;
 }
 
-static void qxl_ttm_backend_unbind(struct ttm_tt *ttm)
+static void qxl_ttm_backend_unbind(struct ttm_bo_device *bdev,
+				   struct ttm_tt *ttm)
 {
 	/* Not implemented */
 }
 
-static void qxl_ttm_backend_destroy(struct ttm_tt *ttm)
+static void qxl_ttm_backend_destroy(struct ttm_bo_device *bdev,
+				    struct ttm_tt *ttm)
 {
 	struct qxl_ttm_tt *gtt = (void *)ttm;
 
+	ttm_tt_destroy_common(bdev, ttm);
 	ttm_tt_fini(&gtt->ttm);
 	kfree(gtt);
 }
 
-static struct ttm_backend_func qxl_backend_func = {
-	.bind = &qxl_ttm_backend_bind,
-	.unbind = &qxl_ttm_backend_unbind,
-	.destroy = &qxl_ttm_backend_destroy,
-};
-
 static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo,
 					uint32_t page_flags)
 {
@@ -178,7 +145,6 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo,
 	gtt = kzalloc(sizeof(struct qxl_ttm_tt), GFP_KERNEL);
 	if (gtt == NULL)
 		return NULL;
-	gtt->ttm.func = &qxl_backend_func;
 	gtt->qdev = qdev;
 	if (ttm_tt_init(&gtt->ttm, bo, page_flags)) {
 		kfree(gtt);
@@ -187,21 +153,11 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo,
 	return &gtt->ttm;
 }
 
-static void qxl_move_null(struct ttm_buffer_object *bo,
-			     struct ttm_mem_reg *new_mem)
-{
-	struct ttm_mem_reg *old_mem = &bo->mem;
-
-	BUG_ON(old_mem->mm_node != NULL);
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
-}
-
 static int qxl_bo_move(struct ttm_buffer_object *bo, bool evict,
 		       struct ttm_operation_ctx *ctx,
-		       struct ttm_mem_reg *new_mem)
+		       struct ttm_resource *new_mem)
 {
-	struct ttm_mem_reg *old_mem = &bo->mem;
+	struct ttm_resource *old_mem = &bo->mem;
 	int ret;
 
 	ret = ttm_bo_wait(bo, ctx->interruptible, ctx->no_wait_gpu);
@@ -209,7 +165,7 @@ static int qxl_bo_move(struct ttm_buffer_object *bo, bool evict,
 		return ret;
 
 	if (old_mem->mem_type == TTM_PL_SYSTEM && bo->ttm == NULL) {
-		qxl_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 	return ttm_bo_move_memcpy(bo, ctx, new_mem);
@@ -217,7 +173,7 @@ static int qxl_bo_move(struct ttm_buffer_object *bo, bool evict,
 
 static void qxl_bo_move_notify(struct ttm_buffer_object *bo,
 			       bool evict,
-			       struct ttm_mem_reg *new_mem)
+			       struct ttm_resource *new_mem)
 {
 	struct qxl_bo *qbo;
 	struct qxl_device *qdev;
@@ -233,7 +189,9 @@ static void qxl_bo_move_notify(struct ttm_buffer_object *bo,
 
 static struct ttm_bo_driver qxl_bo_driver = {
 	.ttm_tt_create = &qxl_ttm_tt_create,
-	.init_mem_type = &qxl_init_mem_type,
+	.ttm_tt_bind = &qxl_ttm_backend_bind,
+	.ttm_tt_destroy = &qxl_ttm_backend_destroy,
+	.ttm_tt_unbind = &qxl_ttm_backend_unbind,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = &qxl_evict_flags,
 	.move = &qxl_bo_move,
@@ -241,6 +199,13 @@ static struct ttm_bo_driver qxl_bo_driver = {
 	.move_notify = &qxl_bo_move_notify,
 };
 
+static int qxl_ttm_init_mem_type(struct qxl_device *qdev,
+				 unsigned int type,
+				 uint64_t size)
+{
+	return ttm_range_man_init(&qdev->mman.bdev, type, false, size);
+}
+
 int qxl_ttm_init(struct qxl_device *qdev)
 {
 	int r;
@@ -258,14 +223,13 @@ int qxl_ttm_init(struct qxl_device *qdev)
 	}
 	/* NOTE: this includes the framebuffer (aka surface 0) */
 	num_io_pages = qdev->rom->ram_header_offset / PAGE_SIZE;
-	r = ttm_bo_init_mm(&qdev->mman.bdev, TTM_PL_VRAM,
-			   num_io_pages);
+	r = qxl_ttm_init_mem_type(qdev, TTM_PL_VRAM, num_io_pages);
 	if (r) {
 		DRM_ERROR("Failed initializing VRAM heap.\n");
 		return r;
 	}
-	r = ttm_bo_init_mm(&qdev->mman.bdev, TTM_PL_PRIV,
-			   qdev->surfaceram_size / PAGE_SIZE);
+	r = qxl_ttm_init_mem_type(qdev, TTM_PL_PRIV,
+				  qdev->surfaceram_size / PAGE_SIZE);
 	if (r) {
 		DRM_ERROR("Failed initializing Surfaces heap.\n");
 		return r;
@@ -281,8 +245,8 @@ int qxl_ttm_init(struct qxl_device *qdev)
 
 void qxl_ttm_fini(struct qxl_device *qdev)
 {
-	ttm_bo_clean_mm(&qdev->mman.bdev, TTM_PL_VRAM);
-	ttm_bo_clean_mm(&qdev->mman.bdev, TTM_PL_PRIV);
+	ttm_range_man_fini(&qdev->mman.bdev, TTM_PL_VRAM);
+	ttm_range_man_fini(&qdev->mman.bdev, TTM_PL_PRIV);
 	ttm_bo_device_release(&qdev->mman.bdev);
 	DRM_INFO("qxl: ttm finalized\n");
 }
@@ -293,12 +257,10 @@ void qxl_ttm_fini(struct qxl_device *qdev)
 static int qxl_mm_dump_table(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_mm *mm = (struct drm_mm *)node->info_ent->data;
+	struct ttm_resource_manager *man = (struct ttm_resource_manager *)node->info_ent->data;
 	struct drm_printer p = drm_seq_file_printer(m);
 
-	spin_lock(&ttm_bo_glob.lru_lock);
-	drm_mm_print(mm, &p);
-	spin_unlock(&ttm_bo_glob.lru_lock);
+	ttm_resource_manager_debug(man, &p);
 	return 0;
 }
 #endif
@@ -319,9 +281,9 @@ void qxl_ttm_debugfs_init(struct qxl_device *qdev)
 		qxl_mem_types_list[i].show = &qxl_mm_dump_table;
 		qxl_mem_types_list[i].driver_features = 0;
 		if (i == 0)
-			qxl_mem_types_list[i].data = qdev->mman.bdev.man[TTM_PL_VRAM].priv;
+			qxl_mem_types_list[i].data = ttm_manager_type(&qdev->mman.bdev, TTM_PL_VRAM);
 		else
-			qxl_mem_types_list[i].data = qdev->mman.bdev.man[TTM_PL_PRIV].priv;
+			qxl_mem_types_list[i].data = ttm_manager_type(&qdev->mman.bdev, TTM_PL_PRIV);
 
 	}
 	qxl_debugfs_add_files(qdev, qxl_mem_types_list, i);
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index b7c3fb2bfb54..a6d8de01194a 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -2815,10 +2815,12 @@ extern void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enabl
 extern void radeon_atom_set_clock_gating(struct radeon_device *rdev, int enable);
 extern void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain);
 extern bool radeon_ttm_bo_is_radeon_bo(struct ttm_buffer_object *bo);
-extern int radeon_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
+extern int radeon_ttm_tt_set_userptr(struct radeon_device *rdev,
+				     struct ttm_tt *ttm, uint64_t addr,
 				     uint32_t flags);
-extern bool radeon_ttm_tt_has_userptr(struct ttm_tt *ttm);
-extern bool radeon_ttm_tt_is_readonly(struct ttm_tt *ttm);
+extern bool radeon_ttm_tt_has_userptr(struct radeon_device *rdev, struct ttm_tt *ttm);
+extern bool radeon_ttm_tt_is_readonly(struct radeon_device *rdev, struct ttm_tt *ttm);
+bool radeon_ttm_tt_is_bound(struct ttm_bo_device *bdev, struct ttm_tt *ttm);
 extern void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64 base);
 extern void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
@@ -2857,7 +2859,7 @@ int radeon_vm_clear_invalids(struct radeon_device *rdev,
 			     struct radeon_vm *vm);
 int radeon_vm_bo_update(struct radeon_device *rdev,
 			struct radeon_bo_va *bo_va,
-			struct ttm_mem_reg *mem);
+			struct ttm_resource *mem);
 void radeon_vm_bo_invalidate(struct radeon_device *rdev,
 			     struct radeon_bo *bo);
 struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm,
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 33ae1b883268..21ce2f9502c0 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -160,7 +160,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 			p->relocs[i].allowed_domains = domain;
 		}
 
-		if (radeon_ttm_tt_has_userptr(p->relocs[i].robj->tbo.ttm)) {
+		if (radeon_ttm_tt_has_userptr(p->rdev, p->relocs[i].robj->tbo.ttm)) {
 			uint32_t domain = p->relocs[i].preferred_domains;
 			if (!(domain & RADEON_GEM_DOMAIN_GTT)) {
 				DRM_ERROR("Only RADEON_GEM_DOMAIN_GTT is "
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index f178ba321715..3808a753127b 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -72,8 +72,8 @@ int radeon_gart_table_ram_alloc(struct radeon_device *rdev)
 {
 	void *ptr;
 
-	ptr = pci_alloc_consistent(rdev->pdev, rdev->gart.table_size,
-				   &rdev->gart.table_addr);
+	ptr = dma_alloc_coherent(&rdev->pdev->dev, rdev->gart.table_size,
+				 &rdev->gart.table_addr, GFP_KERNEL);
 	if (ptr == NULL) {
 		return -ENOMEM;
 	}
@@ -85,7 +85,6 @@ int radeon_gart_table_ram_alloc(struct radeon_device *rdev)
 	}
 #endif
 	rdev->gart.ptr = ptr;
-	memset((void *)rdev->gart.ptr, 0, rdev->gart.table_size);
 	return 0;
 }
 
@@ -110,9 +109,8 @@ void radeon_gart_table_ram_free(struct radeon_device *rdev)
 			      rdev->gart.table_size >> PAGE_SHIFT);
 	}
 #endif
-	pci_free_consistent(rdev->pdev, rdev->gart.table_size,
-			    (void *)rdev->gart.ptr,
-			    rdev->gart.table_addr);
+	dma_free_coherent(&rdev->pdev->dev, rdev->gart.table_size,
+			  (void *)rdev->gart.ptr, rdev->gart.table_addr);
 	rdev->gart.ptr = NULL;
 	rdev->gart.table_addr = 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index 44157ada9b0e..e5c4271e64ed 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -224,9 +224,9 @@ int radeon_gem_info_ioctl(struct drm_device *dev, void *data,
 {
 	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_info *args = data;
-	struct ttm_mem_type_manager *man;
+	struct ttm_resource_manager *man;
 
-	man = &rdev->mman.bdev.man[TTM_PL_VRAM];
+	man = ttm_manager_type(&rdev->mman.bdev, TTM_PL_VRAM);
 
 	args->vram_size = (u64)man->size << PAGE_SHIFT;
 	args->vram_visible = rdev->mc.visible_vram_size;
@@ -331,7 +331,7 @@ int radeon_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		goto handle_lockup;
 
 	bo = gem_to_radeon_bo(gobj);
-	r = radeon_ttm_tt_set_userptr(bo->tbo.ttm, args->addr, args->flags);
+	r = radeon_ttm_tt_set_userptr(rdev, bo->tbo.ttm, args->addr, args->flags);
 	if (r)
 		goto release_object;
 
@@ -420,7 +420,7 @@ int radeon_mode_dumb_mmap(struct drm_file *filp,
 		return -ENOENT;
 	}
 	robj = gem_to_radeon_bo(gobj);
-	if (radeon_ttm_tt_has_userptr(robj->tbo.ttm)) {
+	if (radeon_ttm_tt_has_userptr(robj->rdev, robj->tbo.ttm)) {
 		drm_gem_object_put(gobj);
 		return -EPERM;
 	}
@@ -721,7 +721,7 @@ int radeon_gem_op_ioctl(struct drm_device *dev, void *data,
 	robj = gem_to_radeon_bo(gobj);
 
 	r = -EPERM;
-	if (radeon_ttm_tt_has_userptr(robj->tbo.ttm))
+	if (radeon_ttm_tt_has_userptr(robj->rdev, robj->tbo.ttm))
 		goto out;
 
 	r = radeon_bo_reserve(robj, false);
diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c
index f93829f08a4d..97b9b6dd6dd3 100644
--- a/drivers/gpu/drm/radeon/radeon_mn.c
+++ b/drivers/gpu/drm/radeon/radeon_mn.c
@@ -53,7 +53,7 @@ static bool radeon_mn_invalidate(struct mmu_interval_notifier *mn,
 	struct ttm_operation_ctx ctx = { false, false };
 	long r;
 
-	if (!bo->tbo.ttm || bo->tbo.ttm->state != tt_bound)
+	if (!bo->tbo.ttm || !radeon_ttm_tt_is_bound(bo->tbo.bdev, bo->tbo.ttm))
 		return true;
 
 	if (!mmu_notifier_range_blockable(range))
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index f3dee01250da..316e35d3f8a9 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -112,58 +112,58 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
 		    rbo->rdev->mc.visible_vram_size < rbo->rdev->mc.real_vram_size) {
 			rbo->placements[c].fpfn =
 				rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT;
+			rbo->placements[c].mem_type = TTM_PL_VRAM;
 			rbo->placements[c++].flags = TTM_PL_FLAG_WC |
-						     TTM_PL_FLAG_UNCACHED |
-						     TTM_PL_FLAG_VRAM;
+						     TTM_PL_FLAG_UNCACHED;
 		}
 
 		rbo->placements[c].fpfn = 0;
+		rbo->placements[c].mem_type = TTM_PL_VRAM;
 		rbo->placements[c++].flags = TTM_PL_FLAG_WC |
-					     TTM_PL_FLAG_UNCACHED |
-					     TTM_PL_FLAG_VRAM;
+					     TTM_PL_FLAG_UNCACHED;
 	}
 
 	if (domain & RADEON_GEM_DOMAIN_GTT) {
 		if (rbo->flags & RADEON_GEM_GTT_UC) {
 			rbo->placements[c].fpfn = 0;
-			rbo->placements[c++].flags = TTM_PL_FLAG_UNCACHED |
-				TTM_PL_FLAG_TT;
+			rbo->placements[c].mem_type = TTM_PL_TT;
+			rbo->placements[c++].flags = TTM_PL_FLAG_UNCACHED;
 
 		} else if ((rbo->flags & RADEON_GEM_GTT_WC) ||
 			   (rbo->rdev->flags & RADEON_IS_AGP)) {
 			rbo->placements[c].fpfn = 0;
+			rbo->placements[c].mem_type = TTM_PL_TT;
 			rbo->placements[c++].flags = TTM_PL_FLAG_WC |
-				TTM_PL_FLAG_UNCACHED |
-				TTM_PL_FLAG_TT;
+				TTM_PL_FLAG_UNCACHED;
 		} else {
 			rbo->placements[c].fpfn = 0;
-			rbo->placements[c++].flags = TTM_PL_FLAG_CACHED |
-						     TTM_PL_FLAG_TT;
+			rbo->placements[c].mem_type = TTM_PL_TT;
+			rbo->placements[c++].flags = TTM_PL_FLAG_CACHED;
 		}
 	}
 
 	if (domain & RADEON_GEM_DOMAIN_CPU) {
 		if (rbo->flags & RADEON_GEM_GTT_UC) {
 			rbo->placements[c].fpfn = 0;
-			rbo->placements[c++].flags = TTM_PL_FLAG_UNCACHED |
-				TTM_PL_FLAG_SYSTEM;
+			rbo->placements[c].mem_type = TTM_PL_SYSTEM;
+			rbo->placements[c++].flags = TTM_PL_FLAG_UNCACHED;
 
 		} else if ((rbo->flags & RADEON_GEM_GTT_WC) ||
 		    rbo->rdev->flags & RADEON_IS_AGP) {
 			rbo->placements[c].fpfn = 0;
+			rbo->placements[c].mem_type = TTM_PL_SYSTEM;
 			rbo->placements[c++].flags = TTM_PL_FLAG_WC |
-				TTM_PL_FLAG_UNCACHED |
-				TTM_PL_FLAG_SYSTEM;
+				TTM_PL_FLAG_UNCACHED;
 		} else {
 			rbo->placements[c].fpfn = 0;
-			rbo->placements[c++].flags = TTM_PL_FLAG_CACHED |
-						     TTM_PL_FLAG_SYSTEM;
+			rbo->placements[c].mem_type = TTM_PL_SYSTEM;
+			rbo->placements[c++].flags = TTM_PL_FLAG_CACHED;
 		}
 	}
 	if (!c) {
 		rbo->placements[c].fpfn = 0;
-		rbo->placements[c++].flags = TTM_PL_MASK_CACHING |
-					     TTM_PL_FLAG_SYSTEM;
+		rbo->placements[c].mem_type = TTM_PL_SYSTEM;
+		rbo->placements[c++].flags = TTM_PL_MASK_CACHING;
 	}
 
 	rbo->placement.num_placement = c;
@@ -171,7 +171,7 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
 
 	for (i = 0; i < c; ++i) {
 		if ((rbo->flags & RADEON_GEM_CPU_ACCESS) &&
-		    (rbo->placements[i].flags & TTM_PL_FLAG_VRAM) &&
+		    (rbo->placements[i].mem_type == TTM_PL_VRAM) &&
 		    !rbo->placements[i].fpfn)
 			rbo->placements[i].lpfn =
 				rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT;
@@ -331,7 +331,7 @@ int radeon_bo_pin_restricted(struct radeon_bo *bo, u32 domain, u64 max_offset,
 	struct ttm_operation_ctx ctx = { false, false };
 	int r, i;
 
-	if (radeon_ttm_tt_has_userptr(bo->tbo.ttm))
+	if (radeon_ttm_tt_has_userptr(bo->rdev, bo->tbo.ttm))
 		return -EPERM;
 
 	if (bo->pin_count) {
@@ -360,7 +360,7 @@ int radeon_bo_pin_restricted(struct radeon_bo *bo, u32 domain, u64 max_offset,
 	radeon_ttm_placement_from_domain(bo, domain);
 	for (i = 0; i < bo->placement.num_placement; i++) {
 		/* force to pin into visible video ram */
-		if ((bo->placements[i].flags & TTM_PL_FLAG_VRAM) &&
+		if ((bo->placements[i].mem_type == TTM_PL_VRAM) &&
 		    !(bo->flags & RADEON_GEM_NO_CPU_ACCESS) &&
 		    (!max_offset || max_offset > bo->rdev->mc.visible_vram_size))
 			bo->placements[i].lpfn =
@@ -775,7 +775,7 @@ int radeon_bo_check_tiling(struct radeon_bo *bo, bool has_moved,
 
 void radeon_bo_move_notify(struct ttm_buffer_object *bo,
 			   bool evict,
-			   struct ttm_mem_reg *new_mem)
+			   struct ttm_resource *new_mem)
 {
 	struct radeon_bo *rbo;
 
@@ -824,7 +824,7 @@ int radeon_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 	lpfn =	rdev->mc.visible_vram_size >> PAGE_SHIFT;
 	for (i = 0; i < rbo->placement.num_placement; i++) {
 		/* Force into visible VRAM */
-		if ((rbo->placements[i].flags & TTM_PL_FLAG_VRAM) &&
+		if ((rbo->placements[i].mem_type == TTM_PL_VRAM) &&
 		    (!rbo->placements[i].lpfn || rbo->placements[i].lpfn > lpfn))
 			rbo->placements[i].lpfn = lpfn;
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index 60275b822f79..44b47241ee42 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -165,7 +165,7 @@ extern int radeon_bo_check_tiling(struct radeon_bo *bo, bool has_moved,
 				bool force_drop);
 extern void radeon_bo_move_notify(struct ttm_buffer_object *bo,
 				  bool evict,
-				  struct ttm_mem_reg *new_mem);
+				  struct ttm_resource *new_mem);
 extern int radeon_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 extern int radeon_bo_get_surface_reg(struct radeon_bo *bo);
 extern void radeon_bo_fence(struct radeon_bo *bo, struct radeon_fence *fence,
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 8c5d6fda0d75..05c4196a8212 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -712,6 +712,31 @@ static SENSOR_DEVICE_ATTR(pwm1_enable, S_IRUGO | S_IWUSR, radeon_hwmon_get_pwm1_
 static SENSOR_DEVICE_ATTR(pwm1_min, S_IRUGO, radeon_hwmon_get_pwm1_min, NULL, 0);
 static SENSOR_DEVICE_ATTR(pwm1_max, S_IRUGO, radeon_hwmon_get_pwm1_max, NULL, 0);
 
+static ssize_t radeon_hwmon_show_sclk(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct radeon_device *rdev = dev_get_drvdata(dev);
+	struct drm_device *ddev = rdev->ddev;
+	u32 sclk = 0;
+
+	/* Can't get clock frequency when the card is off */
+	if ((rdev->flags & RADEON_IS_PX) &&
+	    (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
+	if (rdev->asic->dpm.get_current_sclk)
+		sclk = radeon_dpm_get_current_sclk(rdev);
+
+	/* Value returned by dpm is in 10 KHz units, need to convert it into Hz 
+	   for hwmon */
+	sclk *= 10000;
+
+	return snprintf(buf, PAGE_SIZE, "%u\n", sclk);
+}
+
+static SENSOR_DEVICE_ATTR(freq1_input, S_IRUGO, radeon_hwmon_show_sclk, NULL,
+			  0);
+
 
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
@@ -721,6 +746,7 @@ static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_pwm1_enable.dev_attr.attr,
 	&sensor_dev_attr_pwm1_min.dev_attr.attr,
 	&sensor_dev_attr_pwm1_max.dev_attr.attr,
+	&sensor_dev_attr_freq1_input.dev_attr.attr,
 	NULL
 };
 
@@ -738,7 +764,8 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_pwm1.dev_attr.attr ||
 	     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr ||
 	     attr == &sensor_dev_attr_pwm1_max.dev_attr.attr ||
-	     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr))
+	     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr ||
+	     attr == &sensor_dev_attr_freq1_input.dev_attr.attr))
 		return 0;
 
 	/* Skip fan attributes if fan is not present */
diff --git a/drivers/gpu/drm/radeon/radeon_prime.c b/drivers/gpu/drm/radeon/radeon_prime.c
index b906e8fbd5f3..b9de0e51c0be 100644
--- a/drivers/gpu/drm/radeon/radeon_prime.c
+++ b/drivers/gpu/drm/radeon/radeon_prime.c
@@ -36,7 +36,7 @@ struct sg_table *radeon_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	struct radeon_bo *bo = gem_to_radeon_bo(obj);
 	int npages = bo->tbo.num_pages;
 
-	return drm_prime_pages_to_sg(bo->tbo.ttm->pages, npages);
+	return drm_prime_pages_to_sg(obj->dev, bo->tbo.ttm->pages, npages);
 }
 
 void *radeon_gem_prime_vmap(struct drm_gem_object *obj)
@@ -121,7 +121,7 @@ struct dma_buf *radeon_gem_prime_export(struct drm_gem_object *gobj,
 					int flags)
 {
 	struct radeon_bo *bo = gem_to_radeon_bo(gobj);
-	if (radeon_ttm_tt_has_userptr(bo->tbo.ttm))
+	if (radeon_ttm_tt_has_userptr(bo->rdev, bo->tbo.ttm))
 		return ERR_PTR(-EPERM);
 	return drm_gem_prime_export(gobj, flags);
 }
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 004344dce140..36150b7f31a9 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -56,6 +56,10 @@
 static int radeon_ttm_debugfs_init(struct radeon_device *rdev);
 static void radeon_ttm_debugfs_fini(struct radeon_device *rdev);
 
+static int radeon_ttm_tt_bind(struct ttm_bo_device *bdev,
+			      struct ttm_tt *ttm,
+			      struct ttm_resource *bo_mem);
+
 struct radeon_device *radeon_get_rdev(struct ttm_bo_device *bdev)
 {
 	struct radeon_mman *mman;
@@ -66,53 +70,16 @@ struct radeon_device *radeon_get_rdev(struct ttm_bo_device *bdev)
 	return rdev;
 }
 
-static int radeon_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-				struct ttm_mem_type_manager *man)
+static int radeon_ttm_init_vram(struct radeon_device *rdev)
 {
-	struct radeon_device *rdev;
-
-	rdev = radeon_get_rdev(bdev);
+	return ttm_range_man_init(&rdev->mman.bdev, TTM_PL_VRAM,
+				  false, rdev->mc.real_vram_size >> PAGE_SHIFT);
+}
 
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		/* System memory */
-		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_TT:
-		man->func = &ttm_bo_manager_func;
-		man->available_caching = TTM_PL_MASK_CACHING;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
-#if IS_ENABLED(CONFIG_AGP)
-		if (rdev->flags & RADEON_IS_AGP) {
-			if (!rdev->ddev->agp) {
-				DRM_ERROR("AGP is not enabled for memory type %u\n",
-					  (unsigned)type);
-				return -EINVAL;
-			}
-			if (!rdev->ddev->agp->cant_use_aperture)
-				man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
-			man->available_caching = TTM_PL_FLAG_UNCACHED |
-						 TTM_PL_FLAG_WC;
-			man->default_caching = TTM_PL_FLAG_WC;
-		}
-#endif
-		break;
-	case TTM_PL_VRAM:
-		/* "On-card" video ram */
-		man->func = &ttm_bo_manager_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED |
-			     TTM_MEMTYPE_FLAG_MAPPABLE;
-		man->available_caching = TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_WC;
-		man->default_caching = TTM_PL_FLAG_WC;
-		break;
-	default:
-		DRM_ERROR("Unsupported memory type %u\n", (unsigned)type);
-		return -EINVAL;
-	}
-	return 0;
+static int radeon_ttm_init_gtt(struct radeon_device *rdev)
+{
+	return ttm_range_man_init(&rdev->mman.bdev, TTM_PL_TT,
+				  true, rdev->mc.gtt_size >> PAGE_SHIFT);
 }
 
 static void radeon_evict_flags(struct ttm_buffer_object *bo,
@@ -121,7 +88,8 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo,
 	static const struct ttm_place placements = {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_MASK_CACHING
 	};
 
 	struct radeon_bo *rbo;
@@ -152,7 +120,7 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo,
 							 RADEON_GEM_DOMAIN_GTT);
 			rbo->placement.num_busy_placement = 0;
 			for (i = 0; i < rbo->placement.num_placement; i++) {
-				if (rbo->placements[i].flags & TTM_PL_FLAG_VRAM) {
+				if (rbo->placements[i].mem_type == TTM_PL_VRAM) {
 					if (rbo->placements[i].fpfn < fpfn)
 						rbo->placements[i].fpfn = fpfn;
 				} else {
@@ -174,27 +142,18 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo,
 static int radeon_verify_access(struct ttm_buffer_object *bo, struct file *filp)
 {
 	struct radeon_bo *rbo = container_of(bo, struct radeon_bo, tbo);
+	struct radeon_device *rdev = radeon_get_rdev(bo->bdev);
 
-	if (radeon_ttm_tt_has_userptr(bo->ttm))
+	if (radeon_ttm_tt_has_userptr(rdev, bo->ttm))
 		return -EPERM;
 	return drm_vma_node_verify_access(&rbo->tbo.base.vma_node,
 					  filp->private_data);
 }
 
-static void radeon_move_null(struct ttm_buffer_object *bo,
-			     struct ttm_mem_reg *new_mem)
-{
-	struct ttm_mem_reg *old_mem = &bo->mem;
-
-	BUG_ON(old_mem->mm_node != NULL);
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
-}
-
 static int radeon_move_blit(struct ttm_buffer_object *bo,
 			bool evict, bool no_wait_gpu,
-			struct ttm_mem_reg *new_mem,
-			struct ttm_mem_reg *old_mem)
+			struct ttm_resource *new_mem,
+			struct ttm_resource *old_mem)
 {
 	struct radeon_device *rdev;
 	uint64_t old_start, new_start;
@@ -241,7 +200,7 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
 	if (IS_ERR(fence))
 		return PTR_ERR(fence);
 
-	r = ttm_bo_move_accel_cleanup(bo, &fence->base, evict, new_mem);
+	r = ttm_bo_move_accel_cleanup(bo, &fence->base, evict, false, new_mem);
 	radeon_fence_unref(&fence);
 	return r;
 }
@@ -249,11 +208,11 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
 static int radeon_move_vram_ram(struct ttm_buffer_object *bo,
 				bool evict, bool interruptible,
 				bool no_wait_gpu,
-				struct ttm_mem_reg *new_mem)
+				struct ttm_resource *new_mem)
 {
 	struct ttm_operation_ctx ctx = { interruptible, no_wait_gpu };
-	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg tmp_mem;
+	struct ttm_resource *old_mem = &bo->mem;
+	struct ttm_resource tmp_mem;
 	struct ttm_place placements;
 	struct ttm_placement placement;
 	int r;
@@ -266,7 +225,8 @@ static int radeon_move_vram_ram(struct ttm_buffer_object *bo,
 	placement.busy_placement = &placements;
 	placements.fpfn = 0;
 	placements.lpfn = 0;
-	placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
+	placements.mem_type = TTM_PL_TT;
+	placements.flags = TTM_PL_MASK_CACHING;
 	r = ttm_bo_mem_space(bo, &placement, &tmp_mem, &ctx);
 	if (unlikely(r)) {
 		return r;
@@ -277,7 +237,12 @@ static int radeon_move_vram_ram(struct ttm_buffer_object *bo,
 		goto out_cleanup;
 	}
 
-	r = ttm_tt_bind(bo->ttm, &tmp_mem, &ctx);
+	r = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
+	if (unlikely(r)) {
+		goto out_cleanup;
+	}
+
+	r = radeon_ttm_tt_bind(bo->bdev, bo->ttm, &tmp_mem);
 	if (unlikely(r)) {
 		goto out_cleanup;
 	}
@@ -287,18 +252,18 @@ static int radeon_move_vram_ram(struct ttm_buffer_object *bo,
 	}
 	r = ttm_bo_move_ttm(bo, &ctx, new_mem);
 out_cleanup:
-	ttm_bo_mem_put(bo, &tmp_mem);
+	ttm_resource_free(bo, &tmp_mem);
 	return r;
 }
 
 static int radeon_move_ram_vram(struct ttm_buffer_object *bo,
 				bool evict, bool interruptible,
 				bool no_wait_gpu,
-				struct ttm_mem_reg *new_mem)
+				struct ttm_resource *new_mem)
 {
 	struct ttm_operation_ctx ctx = { interruptible, no_wait_gpu };
-	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg tmp_mem;
+	struct ttm_resource *old_mem = &bo->mem;
+	struct ttm_resource tmp_mem;
 	struct ttm_placement placement;
 	struct ttm_place placements;
 	int r;
@@ -311,7 +276,8 @@ static int radeon_move_ram_vram(struct ttm_buffer_object *bo,
 	placement.busy_placement = &placements;
 	placements.fpfn = 0;
 	placements.lpfn = 0;
-	placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
+	placements.mem_type = TTM_PL_TT;
+	placements.flags = TTM_PL_MASK_CACHING;
 	r = ttm_bo_mem_space(bo, &placement, &tmp_mem, &ctx);
 	if (unlikely(r)) {
 		return r;
@@ -325,17 +291,17 @@ static int radeon_move_ram_vram(struct ttm_buffer_object *bo,
 		goto out_cleanup;
 	}
 out_cleanup:
-	ttm_bo_mem_put(bo, &tmp_mem);
+	ttm_resource_free(bo, &tmp_mem);
 	return r;
 }
 
 static int radeon_bo_move(struct ttm_buffer_object *bo, bool evict,
 			  struct ttm_operation_ctx *ctx,
-			  struct ttm_mem_reg *new_mem)
+			  struct ttm_resource *new_mem)
 {
 	struct radeon_device *rdev;
 	struct radeon_bo *rbo;
-	struct ttm_mem_reg *old_mem = &bo->mem;
+	struct ttm_resource *old_mem = &bo->mem;
 	int r;
 
 	r = ttm_bo_wait(bo, ctx->interruptible, ctx->no_wait_gpu);
@@ -349,7 +315,7 @@ static int radeon_bo_move(struct ttm_buffer_object *bo, bool evict,
 
 	rdev = radeon_get_rdev(bo->bdev);
 	if (old_mem->mem_type == TTM_PL_SYSTEM && bo->ttm == NULL) {
-		radeon_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 	if ((old_mem->mem_type == TTM_PL_TT &&
@@ -357,7 +323,7 @@ static int radeon_bo_move(struct ttm_buffer_object *bo, bool evict,
 	    (old_mem->mem_type == TTM_PL_SYSTEM &&
 	     new_mem->mem_type == TTM_PL_TT)) {
 		/* bind is enough */
-		radeon_move_null(bo, new_mem);
+		ttm_bo_move_null(bo, new_mem);
 		return 0;
 	}
 	if (!rdev->ring[radeon_copy_ring_index(rdev)].ready ||
@@ -392,18 +358,11 @@ memcpy:
 	return 0;
 }
 
-static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
+static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_resource *mem)
 {
-	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
 	struct radeon_device *rdev = radeon_get_rdev(bdev);
+	size_t bus_size = (size_t)mem->num_pages << PAGE_SHIFT;
 
-	mem->bus.addr = NULL;
-	mem->bus.offset = 0;
-	mem->bus.size = mem->num_pages << PAGE_SHIFT;
-	mem->bus.base = 0;
-	mem->bus.is_iomem = false;
-	if (!(man->flags & TTM_MEMTYPE_FLAG_MAPPABLE))
-		return -EINVAL;
 	switch (mem->mem_type) {
 	case TTM_PL_SYSTEM:
 		/* system memory */
@@ -412,8 +371,8 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 #if IS_ENABLED(CONFIG_AGP)
 		if (rdev->flags & RADEON_IS_AGP) {
 			/* RADEON_IS_AGP is set only if AGP is active */
-			mem->bus.offset = mem->start << PAGE_SHIFT;
-			mem->bus.base = rdev->mc.agp_base;
+			mem->bus.offset = (mem->start << PAGE_SHIFT) +
+				rdev->mc.agp_base;
 			mem->bus.is_iomem = !rdev->ddev->agp->cant_use_aperture;
 		}
 #endif
@@ -421,9 +380,9 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 	case TTM_PL_VRAM:
 		mem->bus.offset = mem->start << PAGE_SHIFT;
 		/* check if it's visible */
-		if ((mem->bus.offset + mem->bus.size) > rdev->mc.visible_vram_size)
+		if ((mem->bus.offset + bus_size) > rdev->mc.visible_vram_size)
 			return -EINVAL;
-		mem->bus.base = rdev->mc.aper_base;
+		mem->bus.offset += rdev->mc.aper_base;
 		mem->bus.is_iomem = true;
 #ifdef __alpha__
 		/*
@@ -432,12 +391,10 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 		 */
 		if (mem->placement & TTM_PL_FLAG_WC)
 			mem->bus.addr =
-				ioremap_wc(mem->bus.base + mem->bus.offset,
-					   mem->bus.size);
+				ioremap_wc(mem->bus.offset, bus_size);
 		else
 			mem->bus.addr =
-				ioremap(mem->bus.base + mem->bus.offset,
-						mem->bus.size);
+				ioremap(mem->bus.offset, bus_size);
 		if (!mem->bus.addr)
 			return -ENOMEM;
 
@@ -447,7 +404,7 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 		 * It then can be used to build PTEs for VRAM
 		 * access, as done in ttm_bo_vm_fault().
 		 */
-		mem->bus.base = (mem->bus.base & 0x0ffffffffUL) +
+		mem->bus.offset = (mem->bus.offset & 0x0ffffffffUL) +
 			rdev->ddev->hose->dense_mem_base;
 #endif
 		break;
@@ -462,18 +419,18 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
  */
 struct radeon_ttm_tt {
 	struct ttm_dma_tt		ttm;
-	struct radeon_device		*rdev;
 	u64				offset;
 
 	uint64_t			userptr;
 	struct mm_struct		*usermm;
 	uint32_t			userflags;
+	bool bound;
 };
 
 /* prepare the sg table with the user pages */
-static int radeon_ttm_tt_pin_userptr(struct ttm_tt *ttm)
+static int radeon_ttm_tt_pin_userptr(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
-	struct radeon_device *rdev = radeon_get_rdev(ttm->bdev);
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
 	struct radeon_ttm_tt *gtt = (void *)ttm;
 	unsigned pinned = 0;
 	int r;
@@ -532,9 +489,9 @@ release_pages:
 	return r;
 }
 
-static void radeon_ttm_tt_unpin_userptr(struct ttm_tt *ttm)
+static void radeon_ttm_tt_unpin_userptr(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
-	struct radeon_device *rdev = radeon_get_rdev(ttm->bdev);
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
 	struct radeon_ttm_tt *gtt = (void *)ttm;
 	struct sg_page_iter sg_iter;
 
@@ -561,16 +518,28 @@ static void radeon_ttm_tt_unpin_userptr(struct ttm_tt *ttm)
 	sg_free_table(ttm->sg);
 }
 
-static int radeon_ttm_backend_bind(struct ttm_tt *ttm,
-				   struct ttm_mem_reg *bo_mem)
+static bool radeon_ttm_backend_is_bound(struct ttm_tt *ttm)
 {
 	struct radeon_ttm_tt *gtt = (void*)ttm;
+
+	return (gtt->bound);
+}
+
+static int radeon_ttm_backend_bind(struct ttm_bo_device *bdev,
+				   struct ttm_tt *ttm,
+				   struct ttm_resource *bo_mem)
+{
+	struct radeon_ttm_tt *gtt = (void*)ttm;
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
 	uint32_t flags = RADEON_GART_PAGE_VALID | RADEON_GART_PAGE_READ |
 		RADEON_GART_PAGE_WRITE;
 	int r;
 
+	if (gtt->bound)
+		return 0;
+
 	if (gtt->userptr) {
-		radeon_ttm_tt_pin_userptr(ttm);
+		radeon_ttm_tt_pin_userptr(bdev, ttm);
 		flags &= ~RADEON_GART_PAGE_WRITE;
 	}
 
@@ -581,40 +550,43 @@ static int radeon_ttm_backend_bind(struct ttm_tt *ttm,
 	}
 	if (ttm->caching_state == tt_cached)
 		flags |= RADEON_GART_PAGE_SNOOP;
-	r = radeon_gart_bind(gtt->rdev, gtt->offset, ttm->num_pages,
+	r = radeon_gart_bind(rdev, gtt->offset, ttm->num_pages,
 			     ttm->pages, gtt->ttm.dma_address, flags);
 	if (r) {
 		DRM_ERROR("failed to bind %lu pages at 0x%08X\n",
 			  ttm->num_pages, (unsigned)gtt->offset);
 		return r;
 	}
+	gtt->bound = true;
 	return 0;
 }
 
-static void radeon_ttm_backend_unbind(struct ttm_tt *ttm)
+static void radeon_ttm_backend_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct radeon_ttm_tt *gtt = (void *)ttm;
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
 
-	radeon_gart_unbind(gtt->rdev, gtt->offset, ttm->num_pages);
+	if (!gtt->bound)
+		return;
+
+	radeon_gart_unbind(rdev, gtt->offset, ttm->num_pages);
 
 	if (gtt->userptr)
-		radeon_ttm_tt_unpin_userptr(ttm);
+		radeon_ttm_tt_unpin_userptr(bdev, ttm);
+	gtt->bound = false;
 }
 
-static void radeon_ttm_backend_destroy(struct ttm_tt *ttm)
+static void radeon_ttm_backend_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct radeon_ttm_tt *gtt = (void *)ttm;
 
+	radeon_ttm_backend_unbind(bdev, ttm);
+	ttm_tt_destroy_common(bdev, ttm);
+
 	ttm_dma_tt_fini(&gtt->ttm);
 	kfree(gtt);
 }
 
-static struct ttm_backend_func radeon_backend_func = {
-	.bind = &radeon_ttm_backend_bind,
-	.unbind = &radeon_ttm_backend_unbind,
-	.destroy = &radeon_ttm_backend_destroy,
-};
-
 static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo,
 					   uint32_t page_flags)
 {
@@ -633,8 +605,6 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo,
 	if (gtt == NULL) {
 		return NULL;
 	}
-	gtt->ttm.ttm.func = &radeon_backend_func;
-	gtt->rdev = rdev;
 	if (ttm_dma_tt_init(&gtt->ttm, bo, page_flags)) {
 		kfree(gtt);
 		return NULL;
@@ -642,18 +612,25 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo,
 	return &gtt->ttm.ttm;
 }
 
-static struct radeon_ttm_tt *radeon_ttm_tt_to_gtt(struct ttm_tt *ttm)
+static struct radeon_ttm_tt *radeon_ttm_tt_to_gtt(struct radeon_device *rdev,
+						  struct ttm_tt *ttm)
 {
-	if (!ttm || ttm->func != &radeon_backend_func)
+#if IS_ENABLED(CONFIG_AGP)
+	if (rdev->flags & RADEON_IS_AGP)
 		return NULL;
-	return (struct radeon_ttm_tt *)ttm;
+#endif
+
+	if (!ttm)
+		return NULL;
+	return container_of(ttm, struct radeon_ttm_tt, ttm.ttm);
 }
 
-static int radeon_ttm_tt_populate(struct ttm_tt *ttm,
-			struct ttm_operation_ctx *ctx)
+static int radeon_ttm_tt_populate(struct ttm_bo_device *bdev,
+				  struct ttm_tt *ttm,
+				  struct ttm_operation_ctx *ctx)
 {
-	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(ttm);
-	struct radeon_device *rdev;
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 	bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG);
 
 	if (gtt && gtt->userptr) {
@@ -662,21 +639,20 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm,
 			return -ENOMEM;
 
 		ttm->page_flags |= TTM_PAGE_FLAG_SG;
-		ttm->state = tt_unbound;
+		ttm_tt_set_populated(ttm);
 		return 0;
 	}
 
 	if (slave && ttm->sg) {
 		drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
 						 gtt->ttm.dma_address, ttm->num_pages);
-		ttm->state = tt_unbound;
+		ttm_tt_set_populated(ttm);
 		return 0;
 	}
 
-	rdev = radeon_get_rdev(ttm->bdev);
 #if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
-		return ttm_agp_tt_populate(ttm, ctx);
+		return ttm_pool_populate(ttm, ctx);
 	}
 #endif
 
@@ -689,10 +665,10 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm,
 	return ttm_populate_and_map_pages(rdev->dev, &gtt->ttm, ctx);
 }
 
-static void radeon_ttm_tt_unpopulate(struct ttm_tt *ttm)
+static void radeon_ttm_tt_unpopulate(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
-	struct radeon_device *rdev;
-	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(ttm);
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 	bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG);
 
 	if (gtt && gtt->userptr) {
@@ -704,10 +680,9 @@ static void radeon_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	if (slave)
 		return;
 
-	rdev = radeon_get_rdev(ttm->bdev);
 #if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
-		ttm_agp_tt_unpopulate(ttm);
+		ttm_pool_unpopulate(ttm);
 		return;
 	}
 #endif
@@ -722,10 +697,11 @@ static void radeon_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	ttm_unmap_and_unpopulate_pages(rdev->dev, &gtt->ttm);
 }
 
-int radeon_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
+int radeon_ttm_tt_set_userptr(struct radeon_device *rdev,
+			      struct ttm_tt *ttm, uint64_t addr,
 			      uint32_t flags)
 {
-	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(ttm);
+	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 
 	if (gtt == NULL)
 		return -EINVAL;
@@ -736,9 +712,69 @@ int radeon_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
 	return 0;
 }
 
-bool radeon_ttm_tt_has_userptr(struct ttm_tt *ttm)
+bool radeon_ttm_tt_is_bound(struct ttm_bo_device *bdev,
+			    struct ttm_tt *ttm)
+{
+#if IS_ENABLED(CONFIG_AGP)
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+	if (rdev->flags & RADEON_IS_AGP)
+		return ttm_agp_is_bound(ttm);
+#endif
+	return radeon_ttm_backend_is_bound(ttm);
+}
+
+static int radeon_ttm_tt_bind(struct ttm_bo_device *bdev,
+			      struct ttm_tt *ttm,
+			      struct ttm_resource *bo_mem)
+{
+#if IS_ENABLED(CONFIG_AGP)
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+#endif
+
+	if (!bo_mem)
+		return -EINVAL;
+#if IS_ENABLED(CONFIG_AGP)
+	if (rdev->flags & RADEON_IS_AGP)
+		return ttm_agp_bind(ttm, bo_mem);
+#endif
+
+	return radeon_ttm_backend_bind(bdev, ttm, bo_mem);
+}
+
+static void radeon_ttm_tt_unbind(struct ttm_bo_device *bdev,
+				 struct ttm_tt *ttm)
 {
-	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(ttm);
+#if IS_ENABLED(CONFIG_AGP)
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+
+	if (rdev->flags & RADEON_IS_AGP) {
+		ttm_agp_unbind(ttm);
+		return;
+	}
+#endif
+	radeon_ttm_backend_unbind(bdev, ttm);
+}
+
+static void radeon_ttm_tt_destroy(struct ttm_bo_device *bdev,
+				  struct ttm_tt *ttm)
+{
+#if IS_ENABLED(CONFIG_AGP)
+	struct radeon_device *rdev = radeon_get_rdev(bdev);
+
+	if (rdev->flags & RADEON_IS_AGP) {
+		ttm_agp_unbind(ttm);
+		ttm_tt_destroy_common(bdev, ttm);
+		ttm_agp_destroy(ttm);
+		return;
+	}
+#endif
+	radeon_ttm_backend_destroy(bdev, ttm);
+}
+
+bool radeon_ttm_tt_has_userptr(struct radeon_device *rdev,
+			       struct ttm_tt *ttm)
+{
+	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 
 	if (gtt == NULL)
 		return false;
@@ -746,9 +782,10 @@ bool radeon_ttm_tt_has_userptr(struct ttm_tt *ttm)
 	return !!gtt->userptr;
 }
 
-bool radeon_ttm_tt_is_readonly(struct ttm_tt *ttm)
+bool radeon_ttm_tt_is_readonly(struct radeon_device *rdev,
+			       struct ttm_tt *ttm)
 {
-	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(ttm);
+	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 
 	if (gtt == NULL)
 		return false;
@@ -760,7 +797,9 @@ static struct ttm_bo_driver radeon_bo_driver = {
 	.ttm_tt_create = &radeon_ttm_tt_create,
 	.ttm_tt_populate = &radeon_ttm_tt_populate,
 	.ttm_tt_unpopulate = &radeon_ttm_tt_unpopulate,
-	.init_mem_type = &radeon_init_mem_type,
+	.ttm_tt_bind = &radeon_ttm_tt_bind,
+	.ttm_tt_unbind = &radeon_ttm_tt_unbind,
+	.ttm_tt_destroy = &radeon_ttm_tt_destroy,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = &radeon_evict_flags,
 	.move = &radeon_bo_move,
@@ -785,8 +824,8 @@ int radeon_ttm_init(struct radeon_device *rdev)
 		return r;
 	}
 	rdev->mman.initialized = true;
-	r = ttm_bo_init_mm(&rdev->mman.bdev, TTM_PL_VRAM,
-				rdev->mc.real_vram_size >> PAGE_SHIFT);
+
+	r = radeon_ttm_init_vram(rdev);
 	if (r) {
 		DRM_ERROR("Failed initializing VRAM heap.\n");
 		return r;
@@ -811,8 +850,8 @@ int radeon_ttm_init(struct radeon_device *rdev)
 	}
 	DRM_INFO("radeon: %uM of VRAM memory ready\n",
 		 (unsigned) (rdev->mc.real_vram_size / (1024 * 1024)));
-	r = ttm_bo_init_mm(&rdev->mman.bdev, TTM_PL_TT,
-				rdev->mc.gtt_size >> PAGE_SHIFT);
+
+	r = radeon_ttm_init_gtt(rdev);
 	if (r) {
 		DRM_ERROR("Failed initializing GTT heap.\n");
 		return r;
@@ -843,8 +882,8 @@ void radeon_ttm_fini(struct radeon_device *rdev)
 		}
 		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);
+	ttm_range_man_fini(&rdev->mman.bdev, TTM_PL_VRAM);
+	ttm_range_man_fini(&rdev->mman.bdev, TTM_PL_TT);
 	ttm_bo_device_release(&rdev->mman.bdev);
 	radeon_gart_fini(rdev);
 	rdev->mman.initialized = false;
@@ -855,12 +894,12 @@ void radeon_ttm_fini(struct radeon_device *rdev)
  * isn't running */
 void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size)
 {
-	struct ttm_mem_type_manager *man;
+	struct ttm_resource_manager *man;
 
 	if (!rdev->mman.initialized)
 		return;
 
-	man = &rdev->mman.bdev.man[TTM_PL_VRAM];
+	man = ttm_manager_type(&rdev->mman.bdev, TTM_PL_VRAM);
 	/* this just adjusts TTM size idea, which sets lpfn to the correct value */
 	man->size = size >> PAGE_SHIFT;
 }
@@ -914,7 +953,7 @@ static int radeon_mm_dump_table(struct seq_file *m, void *data)
 	unsigned ttm_pl = *(int*)node->info_ent->data;
 	struct drm_device *dev = node->minor->dev;
 	struct radeon_device *rdev = dev->dev_private;
-	struct ttm_mem_type_manager *man = &rdev->mman.bdev.man[ttm_pl];
+	struct ttm_resource_manager *man = ttm_manager_type(&rdev->mman.bdev, ttm_pl);
 	struct drm_printer p = drm_seq_file_printer(m);
 
 	man->func->debug(man, &p);
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index f60fae0aed11..27b14eff532c 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -188,7 +188,7 @@ struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev,
 	    vm_id->last_id_use == rdev->vm_manager.active[vm_id->id])
 		return NULL;
 
-	/* we definately need to flush */
+	/* we definitely need to flush */
 	vm_id->pd_gpu_addr = ~0ll;
 
 	/* skip over VMID 0, since it is the system VM */
@@ -911,7 +911,7 @@ static void radeon_vm_fence_pts(struct radeon_vm *vm,
  */
 int radeon_vm_bo_update(struct radeon_device *rdev,
 			struct radeon_bo_va *bo_va,
-			struct ttm_mem_reg *mem)
+			struct ttm_resource *mem)
 {
 	struct radeon_vm *vm = bo_va->vm;
 	struct radeon_ib ib;
@@ -942,7 +942,7 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 	bo_va->flags &= ~RADEON_VM_PAGE_VALID;
 	bo_va->flags &= ~RADEON_VM_PAGE_SYSTEM;
 	bo_va->flags &= ~RADEON_VM_PAGE_SNOOPED;
-	if (bo_va->bo && radeon_ttm_tt_is_readonly(bo_va->bo->tbo.ttm))
+	if (bo_va->bo && radeon_ttm_tt_is_readonly(rdev, bo_va->bo->tbo.ttm))
 		bo_va->flags &= ~RADEON_VM_PAGE_WRITEABLE;
 
 	if (mem) {
diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c
index 800721153d51..58557c2263a7 100644
--- a/drivers/gpu/drm/radeon/uvd_v1_0.c
+++ b/drivers/gpu/drm/radeon/uvd_v1_0.c
@@ -117,7 +117,7 @@ int uvd_v1_0_resume(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	/* programm the VCPU memory controller bits 0-27 */
+	/* program the VCPU memory controller bits 0-27 */
 	addr = (rdev->uvd.gpu_addr >> 3) + 16;
 	size = RADEON_GPU_PAGE_ALIGN(rdev->uvd_fw->size) >> 3;
 	WREG32(UVD_VCPU_CACHE_OFFSET0, addr);
@@ -360,7 +360,7 @@ int uvd_v1_0_start(struct radeon_device *rdev)
 	/* Set the write pointer delay */
 	WREG32(UVD_RBC_RB_WPTR_CNTL, 0);
 
-	/* programm the 4GB memory segment for rptr and ring buffer */
+	/* program the 4GB memory segment for rptr and ring buffer */
 	WREG32(UVD_LMI_EXT40_ADDR, upper_32_bits(ring->gpu_addr) |
 				   (0x7 << 16) | (0x1 << 31));
 
diff --git a/drivers/gpu/drm/radeon/uvd_v2_2.c b/drivers/gpu/drm/radeon/uvd_v2_2.c
index 23b18edda20e..6266167886d9 100644
--- a/drivers/gpu/drm/radeon/uvd_v2_2.c
+++ b/drivers/gpu/drm/radeon/uvd_v2_2.c
@@ -109,7 +109,7 @@ int uvd_v2_2_resume(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	/* programm the VCPU memory controller bits 0-27 */
+	/* program the VCPU memory controller bits 0-27 */
 	addr = rdev->uvd.gpu_addr >> 3;
 	size = RADEON_GPU_PAGE_ALIGN(rdev->uvd_fw->size + 4) >> 3;
 	WREG32(UVD_VCPU_CACHE_OFFSET0, addr);
diff --git a/drivers/gpu/drm/radeon/uvd_v4_2.c b/drivers/gpu/drm/radeon/uvd_v4_2.c
index dc54fa4aaea8..f9e97fa63674 100644
--- a/drivers/gpu/drm/radeon/uvd_v4_2.c
+++ b/drivers/gpu/drm/radeon/uvd_v4_2.c
@@ -40,7 +40,7 @@ int uvd_v4_2_resume(struct radeon_device *rdev)
 	uint64_t addr;
 	uint32_t size;
 
-	/* programm the VCPU memory controller bits 0-27 */
+	/* program the VCPU memory controller bits 0-27 */
 
 	/* skip over the header of the new firmware format */
 	if (rdev->uvd.fw_header_present)
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index f65d1489dc50..b47e74421e34 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -22,11 +22,11 @@ config DRM_RCAR_CMM
 	  Enable support for R-Car Color Management Module (CMM).
 
 config DRM_RCAR_DW_HDMI
-	tristate "R-Car DU Gen3 HDMI Encoder Support"
+	tristate "R-Car Gen3 and RZ/G2 DU HDMI Encoder Support"
 	depends on DRM && OF
 	select DRM_DW_HDMI
 	help
-	  Enable support for R-Car Gen3 internal HDMI encoder.
+	  Enable support for R-Car Gen3 or RZ/G2 internal HDMI encoder.
 
 config DRM_RCAR_LVDS
 	tristate "R-Car DU LVDS Encoder Support"
@@ -49,3 +49,4 @@ config DRM_RCAR_VSP
 config DRM_RCAR_WRITEBACK
 	bool
 	default y if ARM64
+	depends on DRM_RCAR_DU
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index f53b0ec71085..447be991fa25 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -186,6 +186,35 @@ static const struct rcar_du_device_info rcar_du_r8a774c0_info = {
 	.lvds_clk_mask =  BIT(1) | BIT(0),
 };
 
+static const struct rcar_du_device_info rcar_du_r8a774e1_info = {
+	.gen = 3,
+	.features = RCAR_DU_FEATURE_CRTC_IRQ_CLOCK
+		  | RCAR_DU_FEATURE_VSP1_SOURCE
+		  | RCAR_DU_FEATURE_INTERLACED
+		  | RCAR_DU_FEATURE_TVM_SYNC,
+	.channels_mask = BIT(3) | BIT(1) | BIT(0),
+	.routes = {
+		/*
+		 * R8A774E1 has one RGB output, one LVDS output and one HDMI
+		 * output.
+		 */
+		[RCAR_DU_OUTPUT_DPAD0] = {
+			.possible_crtcs = BIT(2),
+			.port = 0,
+		},
+		[RCAR_DU_OUTPUT_HDMI0] = {
+			.possible_crtcs = BIT(1),
+			.port = 1,
+		},
+		[RCAR_DU_OUTPUT_LVDS0] = {
+			.possible_crtcs = BIT(0),
+			.port = 2,
+		},
+	},
+	.num_lvds = 1,
+	.dpll_mask =  BIT(1),
+};
+
 static const struct rcar_du_device_info rcar_du_r8a7779_info = {
 	.gen = 1,
 	.features = RCAR_DU_FEATURE_INTERLACED
@@ -216,8 +245,9 @@ static const struct rcar_du_device_info rcar_du_r8a7790_info = {
 	.channels_mask = BIT(2) | BIT(1) | BIT(0),
 	.routes = {
 		/*
-		 * R8A7790 has one RGB output, two LVDS outputs and one
-		 * (currently unsupported) TCON output.
+		 * R8A7742 and R8A7790 each have one RGB output and two LVDS
+		 * outputs. Additionally R8A7790 supports one TCON output
+		 * (currently unsupported by the driver).
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
 			.possible_crtcs = BIT(2) | BIT(1) | BIT(0),
@@ -443,6 +473,7 @@ static const struct rcar_du_device_info rcar_du_r8a7799x_info = {
 };
 
 static const struct of_device_id rcar_du_of_table[] = {
+	{ .compatible = "renesas,du-r8a7742", .data = &rcar_du_r8a7790_info },
 	{ .compatible = "renesas,du-r8a7743", .data = &rzg1_du_r8a7743_info },
 	{ .compatible = "renesas,du-r8a7744", .data = &rzg1_du_r8a7743_info },
 	{ .compatible = "renesas,du-r8a7745", .data = &rzg1_du_r8a7745_info },
@@ -450,6 +481,7 @@ static const struct of_device_id rcar_du_of_table[] = {
 	{ .compatible = "renesas,du-r8a774a1", .data = &rcar_du_r8a774a1_info },
 	{ .compatible = "renesas,du-r8a774b1", .data = &rcar_du_r8a774b1_info },
 	{ .compatible = "renesas,du-r8a774c0", .data = &rcar_du_r8a774c0_info },
+	{ .compatible = "renesas,du-r8a774e1", .data = &rcar_du_r8a774e1_info },
 	{ .compatible = "renesas,du-r8a7779", .data = &rcar_du_r8a7779_info },
 	{ .compatible = "renesas,du-r8a7790", .data = &rcar_du_r8a7790_info },
 	{ .compatible = "renesas,du-r8a7791", .data = &rcar_du_r8a7791_info },
@@ -458,6 +490,7 @@ static const struct of_device_id rcar_du_of_table[] = {
 	{ .compatible = "renesas,du-r8a7794", .data = &rcar_du_r8a7794_info },
 	{ .compatible = "renesas,du-r8a7795", .data = &rcar_du_r8a7795_info },
 	{ .compatible = "renesas,du-r8a7796", .data = &rcar_du_r8a7796_info },
+	{ .compatible = "renesas,du-r8a77961", .data = &rcar_du_r8a7796_info },
 	{ .compatible = "renesas,du-r8a77965", .data = &rcar_du_r8a77965_info },
 	{ .compatible = "renesas,du-r8a77970", .data = &rcar_du_r8a77970_info },
 	{ .compatible = "renesas,du-r8a77980", .data = &rcar_du_r8a77970_info },
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
index 482329102f19..72dda446355f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
@@ -40,6 +40,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_RGB565,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 		.pnmr = PnMR_SPIM_TP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -47,6 +48,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_ARGB555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 		.pnmr = PnMR_SPIM_ALP | PnMR_DDDF_ARGB,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -61,6 +63,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_XBGR32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 		.pnmr = PnMR_SPIM_TP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_RGB888,
 	}, {
@@ -68,6 +71,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_ABGR32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 		.pnmr = PnMR_SPIM_ALP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_ARGB8888,
 	}, {
@@ -75,6 +79,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_UYVY,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -82,6 +87,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_YUYV,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -89,6 +95,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_NV12M,
 		.bpp = 12,
 		.planes = 2,
+		.hsub = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -96,6 +103,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_NV21M,
 		.bpp = 12,
 		.planes = 2,
+		.hsub = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
@@ -103,6 +111,7 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_NV16M,
 		.bpp = 16,
 		.planes = 2,
+		.hsub = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	},
@@ -115,156 +124,187 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.v4l2 = V4L2_PIX_FMT_RGB332,
 		.bpp = 8,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_ARGB4444,
 		.v4l2 = V4L2_PIX_FMT_ARGB444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_XRGB4444,
 		.v4l2 = V4L2_PIX_FMT_XRGB444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBA4444,
 		.v4l2 = V4L2_PIX_FMT_RGBA444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBX4444,
 		.v4l2 = V4L2_PIX_FMT_RGBX444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_ABGR4444,
 		.v4l2 = V4L2_PIX_FMT_ABGR444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_XBGR4444,
 		.v4l2 = V4L2_PIX_FMT_XBGR444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRA4444,
 		.v4l2 = V4L2_PIX_FMT_BGRA444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRX4444,
 		.v4l2 = V4L2_PIX_FMT_BGRX444,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBA5551,
 		.v4l2 = V4L2_PIX_FMT_RGBA555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBX5551,
 		.v4l2 = V4L2_PIX_FMT_RGBX555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_ABGR1555,
 		.v4l2 = V4L2_PIX_FMT_ABGR555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_XBGR1555,
 		.v4l2 = V4L2_PIX_FMT_XBGR555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRA5551,
 		.v4l2 = V4L2_PIX_FMT_BGRA555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRX5551,
 		.v4l2 = V4L2_PIX_FMT_BGRX555,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGR888,
 		.v4l2 = V4L2_PIX_FMT_RGB24,
 		.bpp = 24,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGB888,
 		.v4l2 = V4L2_PIX_FMT_BGR24,
 		.bpp = 24,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBA8888,
 		.v4l2 = V4L2_PIX_FMT_BGRA32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGBX8888,
 		.v4l2 = V4L2_PIX_FMT_BGRX32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_ABGR8888,
 		.v4l2 = V4L2_PIX_FMT_RGBA32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_XBGR8888,
 		.v4l2 = V4L2_PIX_FMT_RGBX32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRA8888,
 		.v4l2 = V4L2_PIX_FMT_ARGB32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRX8888,
 		.v4l2 = V4L2_PIX_FMT_XRGB32,
 		.bpp = 32,
 		.planes = 1,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_YVYU,
 		.v4l2 = V4L2_PIX_FMT_YVYU,
 		.bpp = 16,
 		.planes = 1,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_NV61,
 		.v4l2 = V4L2_PIX_FMT_NV61M,
 		.bpp = 16,
 		.planes = 2,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YUV420,
 		.v4l2 = V4L2_PIX_FMT_YUV420M,
 		.bpp = 12,
 		.planes = 3,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YVU420,
 		.v4l2 = V4L2_PIX_FMT_YVU420M,
 		.bpp = 12,
 		.planes = 3,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YUV422,
 		.v4l2 = V4L2_PIX_FMT_YUV422M,
 		.bpp = 16,
 		.planes = 3,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YVU422,
 		.v4l2 = V4L2_PIX_FMT_YVU422M,
 		.bpp = 16,
 		.planes = 3,
+		.hsub = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YUV444,
 		.v4l2 = V4L2_PIX_FMT_YUV444M,
 		.bpp = 24,
 		.planes = 3,
+		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_YVU444,
 		.v4l2 = V4L2_PIX_FMT_YVU444M,
 		.bpp = 24,
 		.planes = 3,
+		.hsub = 1,
 	},
 };
 
@@ -311,6 +351,7 @@ rcar_du_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 {
 	struct rcar_du_device *rcdu = dev->dev_private;
 	const struct rcar_du_format_info *format;
+	unsigned int chroma_pitch;
 	unsigned int max_pitch;
 	unsigned int align;
 	unsigned int i;
@@ -353,10 +394,19 @@ rcar_du_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		return ERR_PTR(-EINVAL);
 	}
 
+	/*
+	 * Calculate the chroma plane(s) pitch using the horizontal subsampling
+	 * factor. For semi-planar formats, the U and V planes are combined, the
+	 * pitch must thus be doubled.
+	 */
+	chroma_pitch = mode_cmd->pitches[0] / format->hsub;
+	if (format->planes == 2)
+		chroma_pitch *= 2;
+
 	for (i = 1; i < format->planes; ++i) {
-		if (mode_cmd->pitches[i] != mode_cmd->pitches[0]) {
+		if (mode_cmd->pitches[i] != chroma_pitch) {
 			dev_dbg(dev->dev,
-				"luma and chroma pitches do not match\n");
+				"luma and chroma pitches are not compatible\n");
 			return ERR_PTR(-EINVAL);
 		}
 	}
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.h b/drivers/gpu/drm/rcar-du/rcar_du_kms.h
index 0346504d8c59..8f5fff176754 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.h
@@ -22,6 +22,7 @@ struct rcar_du_format_info {
 	u32 v4l2;
 	unsigned int bpp;
 	unsigned int planes;
+	unsigned int hsub;
 	unsigned int pnmr;
 	unsigned int edf;
 };
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
index f1a81c9b184d..f6a69aa116e6 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
@@ -13,6 +13,7 @@
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_managed.h>
 #include <drm/drm_plane_helper.h>
 #include <drm/drm_vblank.h>
 
@@ -197,9 +198,8 @@ int rcar_du_vsp_map_fb(struct rcar_du_vsp *vsp, struct drm_framebuffer *fb,
 			goto fail;
 
 		ret = vsp1_du_map_sg(vsp->vsp, sgt);
-		if (!ret) {
+		if (ret) {
 			sg_free_table(sgt);
-			ret = -ENOMEM;
 			goto fail;
 		}
 	}
@@ -279,7 +279,7 @@ static void rcar_du_vsp_plane_atomic_update(struct drm_plane *plane,
 
 	if (plane->state->visible)
 		rcar_du_vsp_plane_setup(rplane);
-	else
+	else if (old_state->crtc)
 		vsp1_du_atomic_update(rplane->vsp->vsp, crtc->vsp_pipe,
 				      rplane->index, NULL);
 }
@@ -341,6 +341,13 @@ static const struct drm_plane_funcs rcar_du_vsp_plane_funcs = {
 	.atomic_destroy_state = rcar_du_vsp_plane_atomic_destroy_state,
 };
 
+static void rcar_du_vsp_cleanup(struct drm_device *dev, void *res)
+{
+	struct rcar_du_vsp *vsp = res;
+
+	put_device(vsp->vsp);
+}
+
 int rcar_du_vsp_init(struct rcar_du_vsp *vsp, struct device_node *np,
 		     unsigned int crtcs)
 {
@@ -357,6 +364,10 @@ int rcar_du_vsp_init(struct rcar_du_vsp *vsp, struct device_node *np,
 
 	vsp->vsp = &pdev->dev;
 
+	ret = drmm_add_action(rcdu->ddev, rcar_du_vsp_cleanup, vsp);
+	if (ret < 0)
+		return ret;
+
 	ret = vsp1_du_init(vsp->vsp);
 	if (ret < 0)
 		return ret;
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c
index ab0d49618cf9..70dbbe44bb23 100644
--- a/drivers/gpu/drm/rcar-du/rcar_lvds.c
+++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c
@@ -677,15 +677,11 @@ static int rcar_lvds_attach(struct drm_bridge *bridge,
 	if (ret < 0)
 		return ret;
 
-	return drm_panel_attach(lvds->panel, connector);
+	return 0;
 }
 
 static void rcar_lvds_detach(struct drm_bridge *bridge)
 {
-	struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge);
-
-	if (lvds->panel)
-		drm_panel_detach(lvds->panel);
 }
 
 static const struct drm_bridge_funcs rcar_lvds_bridge_ops = {
@@ -982,11 +978,13 @@ static const struct rcar_lvds_device_info rcar_lvds_r8a77995_info = {
 };
 
 static const struct of_device_id rcar_lvds_of_table[] = {
+	{ .compatible = "renesas,r8a7742-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a7743-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a7744-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a774a1-lvds", .data = &rcar_lvds_gen3_info },
 	{ .compatible = "renesas,r8a774b1-lvds", .data = &rcar_lvds_gen3_info },
 	{ .compatible = "renesas,r8a774c0-lvds", .data = &rcar_lvds_r8a77990_info },
+	{ .compatible = "renesas,r8a774e1-lvds", .data = &rcar_lvds_gen3_info },
 	{ .compatible = "renesas,r8a7790-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a7791-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a7793-lvds", .data = &rcar_lvds_gen2_info },
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
index b9275ba7c5a5..62e5d0970525 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
@@ -36,8 +36,8 @@ static int rockchip_gem_iommu_map(struct rockchip_gem_object *rk_obj)
 
 	rk_obj->dma_addr = rk_obj->mm.start;
 
-	ret = iommu_map_sg(private->domain, rk_obj->dma_addr, rk_obj->sgt->sgl,
-			   rk_obj->sgt->nents, prot);
+	ret = iommu_map_sgtable(private->domain, rk_obj->dma_addr, rk_obj->sgt,
+				prot);
 	if (ret < rk_obj->base.size) {
 		DRM_ERROR("failed to map buffer: size=%zd request_size=%zd\n",
 			  ret, rk_obj->base.size);
@@ -85,7 +85,8 @@ static int rockchip_gem_get_pages(struct rockchip_gem_object *rk_obj)
 
 	rk_obj->num_pages = rk_obj->base.size >> PAGE_SHIFT;
 
-	rk_obj->sgt = drm_prime_pages_to_sg(rk_obj->pages, rk_obj->num_pages);
+	rk_obj->sgt = drm_prime_pages_to_sg(rk_obj->base.dev,
+					    rk_obj->pages, rk_obj->num_pages);
 	if (IS_ERR(rk_obj->sgt)) {
 		ret = PTR_ERR(rk_obj->sgt);
 		goto err_put_pages;
@@ -98,11 +99,10 @@ static int rockchip_gem_get_pages(struct rockchip_gem_object *rk_obj)
 	 * TODO: Replace this by drm_clflush_sg() once it can be implemented
 	 * without relying on symbols that are not exported.
 	 */
-	for_each_sg(rk_obj->sgt->sgl, s, rk_obj->sgt->nents, i)
+	for_each_sgtable_sg(rk_obj->sgt, s, i)
 		sg_dma_address(s) = sg_phys(s);
 
-	dma_sync_sg_for_device(drm->dev, rk_obj->sgt->sgl, rk_obj->sgt->nents,
-			       DMA_TO_DEVICE);
+	dma_sync_sgtable_for_device(drm->dev, rk_obj->sgt, DMA_TO_DEVICE);
 
 	return 0;
 
@@ -350,8 +350,8 @@ void rockchip_gem_free_object(struct drm_gem_object *obj)
 		if (private->domain) {
 			rockchip_gem_iommu_unmap(rk_obj);
 		} else {
-			dma_unmap_sg(drm->dev, rk_obj->sgt->sgl,
-				     rk_obj->sgt->nents, DMA_BIDIRECTIONAL);
+			dma_unmap_sgtable(drm->dev, rk_obj->sgt,
+					  DMA_BIDIRECTIONAL, 0);
 		}
 		drm_prime_gem_destroy(obj, rk_obj->sgt);
 	} else {
@@ -442,7 +442,7 @@ struct sg_table *rockchip_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	int ret;
 
 	if (rk_obj->pages)
-		return drm_prime_pages_to_sg(rk_obj->pages, rk_obj->num_pages);
+		return drm_prime_pages_to_sg(obj->dev, rk_obj->pages, rk_obj->num_pages);
 
 	sgt = kzalloc(sizeof(*sgt), GFP_KERNEL);
 	if (!sgt)
@@ -460,23 +460,6 @@ struct sg_table *rockchip_gem_prime_get_sg_table(struct drm_gem_object *obj)
 	return sgt;
 }
 
-static unsigned long rockchip_sg_get_contiguous_size(struct sg_table *sgt,
-						     int count)
-{
-	struct scatterlist *s;
-	dma_addr_t expected = sg_dma_address(sgt->sgl);
-	unsigned int i;
-	unsigned long size = 0;
-
-	for_each_sg(sgt->sgl, s, count, i) {
-		if (sg_dma_address(s) != expected)
-			break;
-		expected = sg_dma_address(s) + sg_dma_len(s);
-		size += sg_dma_len(s);
-	}
-	return size;
-}
-
 static int
 rockchip_gem_iommu_map_sg(struct drm_device *drm,
 			  struct dma_buf_attachment *attach,
@@ -493,15 +476,13 @@ rockchip_gem_dma_map_sg(struct drm_device *drm,
 			struct sg_table *sg,
 			struct rockchip_gem_object *rk_obj)
 {
-	int count = dma_map_sg(drm->dev, sg->sgl, sg->nents,
-			       DMA_BIDIRECTIONAL);
-	if (!count)
-		return -EINVAL;
+	int err = dma_map_sgtable(drm->dev, sg, DMA_BIDIRECTIONAL, 0);
+	if (err)
+		return err;
 
-	if (rockchip_sg_get_contiguous_size(sg, count) < attach->dmabuf->size) {
+	if (drm_prime_get_contiguous_size(sg) < attach->dmabuf->size) {
 		DRM_ERROR("failed to map sg_table to contiguous linear address.\n");
-		dma_unmap_sg(drm->dev, sg->sgl, sg->nents,
-			     DMA_BIDIRECTIONAL);
+		dma_unmap_sgtable(drm->dev, sg, DMA_BIDIRECTIONAL, 0);
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c
index 63f967902c2d..f292c6a6e20f 100644
--- a/drivers/gpu/drm/rockchip/rockchip_lvds.c
+++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c
@@ -634,13 +634,6 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master,
 				      "failed to attach encoder: %d\n", ret);
 			goto err_free_connector;
 		}
-
-		ret = drm_panel_attach(lvds->panel, connector);
-		if (ret < 0) {
-			DRM_DEV_ERROR(drm_dev->dev,
-				      "failed to attach panel: %d\n", ret);
-			goto err_free_connector;
-		}
 	} else {
 		ret = drm_bridge_attach(encoder, lvds->bridge, NULL, 0);
 		if (ret) {
@@ -676,8 +669,6 @@ static void rockchip_lvds_unbind(struct device *dev, struct device *master,
 
 	encoder_funcs = lvds->soc_data->helper_funcs;
 	encoder_funcs->disable(&lvds->encoder);
-	if (lvds->panel)
-		drm_panel_detach(lvds->panel);
 	pm_runtime_disable(dev);
 	drm_connector_cleanup(&lvds->connector);
 	drm_encoder_cleanup(&lvds->encoder);
diff --git a/drivers/gpu/drm/scheduler/sched_fence.c b/drivers/gpu/drm/scheduler/sched_fence.c
index 8b45c3a1b84e..69de2c76731f 100644
--- a/drivers/gpu/drm/scheduler/sched_fence.c
+++ b/drivers/gpu/drm/scheduler/sched_fence.c
@@ -101,7 +101,7 @@ static void drm_sched_fence_free(struct rcu_head *rcu)
 /**
  * drm_sched_fence_release_scheduled - callback that fence can be freed
  *
- * @fence: fence
+ * @f: fence
  *
  * This function is called when the reference count becomes zero.
  * It just RCU schedules freeing up the fence.
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 96f763d888af..9a0d77a68018 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -625,7 +625,7 @@ drm_sched_select_entity(struct drm_gpu_scheduler *sched)
 		return NULL;
 
 	/* Kernel run queue has higher priority than normal run queue*/
-	for (i = DRM_SCHED_PRIORITY_MAX - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
+	for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
 		entity = drm_sched_rq_select_entity(&sched->sched_rq[i]);
 		if (entity)
 			break;
@@ -852,7 +852,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
 	sched->name = name;
 	sched->timeout = timeout;
 	sched->hang_limit = hang_limit;
-	for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_MAX; i++)
+	for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_COUNT; i++)
 		drm_sched_rq_init(sched, &sched->sched_rq[i]);
 
 	init_waitqueue_head(&sched->wake_up_worker);
diff --git a/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c b/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c
index bd990d178765..1d696ec001cf 100644
--- a/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c
+++ b/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c
@@ -5,6 +5,8 @@
 
 #define PREFIX_STR "[drm_dp_mst_helper]"
 
+#include <linux/random.h>
+
 #include <drm/drm_dp_mst_helper.h>
 #include <drm/drm_print.h>
 
@@ -237,6 +239,21 @@ int igt_dp_mst_sideband_msg_req_decode(void *unused)
 	in.u.i2c_write.bytes = data;
 	DO_TEST();
 
+	in.req_type = DP_QUERY_STREAM_ENC_STATUS;
+	in.u.enc_status.stream_id = 1;
+	DO_TEST();
+	get_random_bytes(in.u.enc_status.client_id,
+			 sizeof(in.u.enc_status.client_id));
+	DO_TEST();
+	in.u.enc_status.stream_event = 3;
+	DO_TEST();
+	in.u.enc_status.valid_stream_event = 0;
+	DO_TEST();
+	in.u.enc_status.stream_behavior = 3;
+	DO_TEST();
+	in.u.enc_status.valid_stream_behavior = 1;
+	DO_TEST();
+
 #undef DO_TEST
 	return 0;
 }
diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index de4af7735c46..ddb4184f0726 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -389,8 +389,6 @@ sti_dvo_connector_detect(struct drm_connector *connector, bool force)
 		dvo->panel = of_drm_find_panel(dvo->panel_node);
 		if (IS_ERR(dvo->panel))
 			dvo->panel = NULL;
-		else
-			drm_panel_attach(dvo->panel, connector);
 	}
 
 	if (dvo->panel)
diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c
index ed5d86617802..2f26f85ef538 100644
--- a/drivers/gpu/drm/sun4i/sun4i_backend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_backend.c
@@ -768,7 +768,7 @@ static const struct sunxi_engine_ops sun4i_backend_engine_ops = {
 	.vblank_quirk			= sun4i_backend_vblank_quirk,
 };
 
-static struct regmap_config sun4i_backend_regmap_config = {
+static const struct regmap_config sun4i_backend_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun4i_framebuffer.c b/drivers/gpu/drm/sun4i/sun4i_framebuffer.c
index 1568f68f9a9e..6825ef46f43f 100644
--- a/drivers/gpu/drm/sun4i/sun4i_framebuffer.c
+++ b/drivers/gpu/drm/sun4i/sun4i_framebuffer.c
@@ -35,7 +35,7 @@ static const struct drm_mode_config_funcs sun4i_de_mode_config_funcs = {
 	.fb_create		= drm_gem_fb_create,
 };
 
-static struct drm_mode_config_helper_funcs sun4i_de_mode_config_helpers = {
+static const struct drm_mode_config_helper_funcs sun4i_de_mode_config_helpers = {
 	.atomic_commit_tail	= drm_atomic_helper_commit_tail_rpm,
 };
 
diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.c b/drivers/gpu/drm/sun4i/sun4i_frontend.c
index ec2a032e07b9..b51cc685c13a 100644
--- a/drivers/gpu/drm/sun4i/sun4i_frontend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_frontend.c
@@ -545,7 +545,7 @@ int sun4i_frontend_enable(struct sun4i_frontend *frontend)
 }
 EXPORT_SYMBOL(sun4i_frontend_enable);
 
-static struct regmap_config sun4i_frontend_regmap_config = {
+static const struct regmap_config sun4i_frontend_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun4i_lvds.c b/drivers/gpu/drm/sun4i/sun4i_lvds.c
index ffda3184aa12..ac570437172e 100644
--- a/drivers/gpu/drm/sun4i/sun4i_lvds.c
+++ b/drivers/gpu/drm/sun4i/sun4i_lvds.c
@@ -47,16 +47,13 @@ static int sun4i_lvds_get_modes(struct drm_connector *connector)
 	return drm_panel_get_modes(lvds->panel, connector);
 }
 
-static struct drm_connector_helper_funcs sun4i_lvds_con_helper_funcs = {
+static const struct drm_connector_helper_funcs sun4i_lvds_con_helper_funcs = {
 	.get_modes	= sun4i_lvds_get_modes,
 };
 
 static void
 sun4i_lvds_connector_destroy(struct drm_connector *connector)
 {
-	struct sun4i_lvds *lvds = drm_connector_to_sun4i_lvds(connector);
-
-	drm_panel_detach(lvds->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -141,12 +138,6 @@ int sun4i_lvds_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 
 		drm_connector_attach_encoder(&lvds->connector,
 						  &lvds->encoder);
-
-		ret = drm_panel_attach(lvds->panel, &lvds->connector);
-		if (ret) {
-			dev_err(drm->dev, "Couldn't attach our panel\n");
-			goto err_cleanup_connector;
-		}
 	}
 
 	if (bridge) {
diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c
index 5a7d43939ae6..e172426eb7e9 100644
--- a/drivers/gpu/drm/sun4i/sun4i_rgb.c
+++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c
@@ -138,16 +138,13 @@ out:
 	return MODE_OK;
 }
 
-static struct drm_connector_helper_funcs sun4i_rgb_con_helper_funcs = {
+static const struct drm_connector_helper_funcs sun4i_rgb_con_helper_funcs = {
 	.get_modes	= sun4i_rgb_get_modes,
 };
 
 static void
 sun4i_rgb_connector_destroy(struct drm_connector *connector)
 {
-	struct sun4i_rgb *rgb = drm_connector_to_sun4i_rgb(connector);
-
-	drm_panel_detach(rgb->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -183,7 +180,7 @@ static void sun4i_rgb_encoder_disable(struct drm_encoder *encoder)
 	}
 }
 
-static struct drm_encoder_helper_funcs sun4i_rgb_enc_helper_funcs = {
+static const struct drm_encoder_helper_funcs sun4i_rgb_enc_helper_funcs = {
 	.disable	= sun4i_rgb_encoder_disable,
 	.enable		= sun4i_rgb_encoder_enable,
 	.mode_valid	= sun4i_rgb_mode_valid,
@@ -233,12 +230,6 @@ int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 
 		drm_connector_attach_encoder(&rgb->connector,
 						  &rgb->encoder);
-
-		ret = drm_panel_attach(rgb->panel, &rgb->connector);
-		if (ret) {
-			dev_err(drm->dev, "Couldn't attach our panel\n");
-			goto err_cleanup_connector;
-		}
 	}
 
 	if (rgb->bridge) {
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index e40c542254f6..eaaf5d70e352 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -474,9 +474,7 @@ static void sun4i_tcon0_mode_set_lvds(struct sun4i_tcon *tcon,
 		     SUN4I_TCON0_BASIC2_V_TOTAL(mode->crtc_vtotal * 2) |
 		     SUN4I_TCON0_BASIC2_V_BACKPORCH(bp));
 
-	reg = SUN4I_TCON0_LVDS_IF_CLK_SEL_TCON0 |
-		SUN4I_TCON0_LVDS_IF_DATA_POL_NORMAL |
-		SUN4I_TCON0_LVDS_IF_CLK_POL_NORMAL;
+	reg = SUN4I_TCON0_LVDS_IF_CLK_SEL_TCON0;
 	if (sun4i_tcon_get_pixel_depth(encoder) == 24)
 		reg |= SUN4I_TCON0_LVDS_IF_BITWIDTH_24BITS;
 	else
@@ -825,7 +823,7 @@ static int sun4i_tcon_init_irq(struct device *dev,
 	return 0;
 }
 
-static struct regmap_config sun4i_tcon_regmap_config = {
+static const struct regmap_config sun4i_tcon_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun4i_tv.c b/drivers/gpu/drm/sun4i/sun4i_tv.c
index 63f4428ac3bf..cb91bc11a0c7 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tv.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tv.c
@@ -468,7 +468,7 @@ static void sun4i_tv_mode_set(struct drm_encoder *encoder,
 	regmap_write(tv->regs, SUN4I_TVE_SLAVE_REG, 0);
 }
 
-static struct drm_encoder_helper_funcs sun4i_tv_helper_funcs = {
+static const struct drm_encoder_helper_funcs sun4i_tv_helper_funcs = {
 	.disable	= sun4i_tv_disable,
 	.enable		= sun4i_tv_enable,
 	.mode_set	= sun4i_tv_mode_set,
@@ -504,7 +504,7 @@ static int sun4i_tv_comp_mode_valid(struct drm_connector *connector,
 	return MODE_OK;
 }
 
-static struct drm_connector_helper_funcs sun4i_tv_comp_connector_helper_funcs = {
+static const struct drm_connector_helper_funcs sun4i_tv_comp_connector_helper_funcs = {
 	.get_modes	= sun4i_tv_comp_get_modes,
 	.mode_valid	= sun4i_tv_comp_mode_valid,
 };
@@ -523,7 +523,7 @@ static const struct drm_connector_funcs sun4i_tv_comp_connector_funcs = {
 	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
 };
 
-static struct regmap_config sun4i_tv_regmap_config = {
+static const struct regmap_config sun4i_tv_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
index de8a11abd66a..4f5efcace68e 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
@@ -820,7 +820,7 @@ static int sun6i_dsi_get_modes(struct drm_connector *connector)
 	return drm_panel_get_modes(dsi->panel, connector);
 }
 
-static struct drm_connector_helper_funcs sun6i_dsi_connector_helper_funcs = {
+static const struct drm_connector_helper_funcs sun6i_dsi_connector_helper_funcs = {
 	.get_modes	= sun6i_dsi_get_modes,
 };
 
@@ -973,7 +973,6 @@ static int sun6i_dsi_attach(struct mipi_dsi_host *host,
 	dsi->panel = panel;
 	dsi->device = device;
 
-	drm_panel_attach(dsi->panel, &dsi->connector);
 	drm_kms_helper_hotplug_event(dsi->drm);
 
 	dev_info(host->dev, "Attached device %s\n", device->name);
@@ -985,12 +984,10 @@ static int sun6i_dsi_detach(struct mipi_dsi_host *host,
 			    struct mipi_dsi_device *device)
 {
 	struct sun6i_dsi *dsi = host_to_sun6i_dsi(host);
-	struct drm_panel *panel = dsi->panel;
 
 	dsi->panel = NULL;
 	dsi->device = NULL;
 
-	drm_panel_detach(panel);
 	drm_kms_helper_hotplug_event(dsi->drm);
 
 	return 0;
diff --git a/drivers/gpu/drm/sun4i/sun8i_hdmi_phy.c b/drivers/gpu/drm/sun4i/sun8i_hdmi_phy.c
index 156d00e5165b..35c2133724e2 100644
--- a/drivers/gpu/drm/sun4i/sun8i_hdmi_phy.c
+++ b/drivers/gpu/drm/sun4i/sun8i_hdmi_phy.c
@@ -534,7 +534,7 @@ void sun8i_hdmi_phy_set_ops(struct sun8i_hdmi_phy *phy,
 	}
 }
 
-static struct regmap_config sun8i_hdmi_phy_regmap_config = {
+static const struct regmap_config sun8i_hdmi_phy_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c
index c3304028e3dc..5b42cf25cc86 100644
--- a/drivers/gpu/drm/sun4i/sun8i_mixer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c
@@ -303,7 +303,7 @@ static const struct sunxi_engine_ops sun8i_engine_ops = {
 	.layers_init	= sun8i_layers_init,
 };
 
-static struct regmap_config sun8i_mixer_regmap_config = {
+static const struct regmap_config sun8i_mixer_regmap_config = {
 	.reg_bits	= 32,
 	.val_bits	= 32,
 	.reg_stride	= 4,
diff --git a/drivers/gpu/drm/sun4i/sun8i_ui_layer.c b/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
index 54f937a7d5e7..816ad4ce8996 100644
--- a/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
@@ -298,7 +298,7 @@ static void sun8i_ui_layer_atomic_update(struct drm_plane *plane,
 			      true, zpos, old_zpos);
 }
 
-static struct drm_plane_helper_funcs sun8i_ui_layer_helper_funcs = {
+static const struct drm_plane_helper_funcs sun8i_ui_layer_helper_funcs = {
 	.prepare_fb	= drm_gem_fb_prepare_fb,
 	.atomic_check	= sun8i_ui_layer_atomic_check,
 	.atomic_disable	= sun8i_ui_layer_atomic_disable,
diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
index c0147af6a840..76393fc976fe 100644
--- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
@@ -401,7 +401,7 @@ static void sun8i_vi_layer_atomic_update(struct drm_plane *plane,
 			      true, zpos, old_zpos);
 }
 
-static struct drm_plane_helper_funcs sun8i_vi_layer_helper_funcs = {
+static const struct drm_plane_helper_funcs sun8i_vi_layer_helper_funcs = {
 	.prepare_fb	= drm_gem_fb_prepare_fb,
 	.atomic_check	= sun8i_vi_layer_atomic_check,
 	.atomic_disable	= sun8i_vi_layer_atomic_disable,
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index b25443255be6..f38de08e0c95 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -12,6 +12,7 @@
 #include <linux/gpio/consumer.h>
 
 #include <drm/drm_atomic.h>
+#include <drm/drm_bridge.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
@@ -116,6 +117,7 @@ struct tegra_output {
 	struct device_node *of_node;
 	struct device *dev;
 
+	struct drm_bridge *bridge;
 	struct drm_panel *panel;
 	struct i2c_adapter *ddc;
 	const struct edid *edid;
diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c
index a7864e91fc0b..5691ef1b0e58 100644
--- a/drivers/gpu/drm/tegra/dsi.c
+++ b/drivers/gpu/drm/tegra/dsi.c
@@ -1498,10 +1498,8 @@ static int tegra_dsi_host_attach(struct mipi_dsi_host *host,
 		if (IS_ERR(output->panel))
 			output->panel = NULL;
 
-		if (output->panel && output->connector.dev) {
-			drm_panel_attach(output->panel, &output->connector);
+		if (output->panel && output->connector.dev)
 			drm_helper_hpd_irq_event(output->connector.dev);
-		}
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index 723df142a981..a2bac20ff19d 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -98,8 +98,8 @@ static struct sg_table *tegra_bo_pin(struct device *dev, struct host1x_bo *bo,
 		 * the SG table needs to be copied to avoid overwriting any
 		 * other potential users of the original SG table.
 		 */
-		err = sg_alloc_table_from_sg(sgt, obj->sgt->sgl, obj->sgt->nents,
-					     GFP_KERNEL);
+		err = sg_alloc_table_from_sg(sgt, obj->sgt->sgl,
+					     obj->sgt->orig_nents, GFP_KERNEL);
 		if (err < 0)
 			goto free;
 	} else {
@@ -196,8 +196,7 @@ static int tegra_bo_iommu_map(struct tegra_drm *tegra, struct tegra_bo *bo)
 
 	bo->iova = bo->mm->start;
 
-	bo->size = iommu_map_sg(tegra->domain, bo->iova, bo->sgt->sgl,
-				bo->sgt->nents, prot);
+	bo->size = iommu_map_sgtable(tegra->domain, bo->iova, bo->sgt, prot);
 	if (!bo->size) {
 		dev_err(tegra->drm->dev, "failed to map buffer\n");
 		err = -ENOMEM;
@@ -264,8 +263,7 @@ free:
 static void tegra_bo_free(struct drm_device *drm, struct tegra_bo *bo)
 {
 	if (bo->pages) {
-		dma_unmap_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-			     DMA_FROM_DEVICE);
+		dma_unmap_sgtable(drm->dev, bo->sgt, DMA_FROM_DEVICE, 0);
 		drm_gem_put_pages(&bo->gem, bo->pages, true, true);
 		sg_free_table(bo->sgt);
 		kfree(bo->sgt);
@@ -284,18 +282,15 @@ static int tegra_bo_get_pages(struct drm_device *drm, struct tegra_bo *bo)
 
 	bo->num_pages = bo->gem.size >> PAGE_SHIFT;
 
-	bo->sgt = drm_prime_pages_to_sg(bo->pages, bo->num_pages);
+	bo->sgt = drm_prime_pages_to_sg(bo->gem.dev, bo->pages, bo->num_pages);
 	if (IS_ERR(bo->sgt)) {
 		err = PTR_ERR(bo->sgt);
 		goto put_pages;
 	}
 
-	err = dma_map_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-			 DMA_FROM_DEVICE);
-	if (err == 0) {
-		err = -EFAULT;
+	err = dma_map_sgtable(drm->dev, bo->sgt, DMA_FROM_DEVICE, 0);
+	if (err)
 		goto free_sgt;
-	}
 
 	return 0;
 
@@ -571,7 +566,7 @@ tegra_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 			goto free;
 	}
 
-	if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0)
+	if (dma_map_sgtable(attach->dev, sgt, dir, 0))
 		goto free;
 
 	return sgt;
@@ -590,7 +585,7 @@ static void tegra_gem_prime_unmap_dma_buf(struct dma_buf_attachment *attach,
 	struct tegra_bo *bo = to_tegra_bo(gem);
 
 	if (bo->pages)
-		dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir);
+		dma_unmap_sgtable(attach->dev, sgt, dir, 0);
 
 	sg_free_table(sgt);
 	kfree(sgt);
@@ -609,8 +604,7 @@ static int tegra_gem_prime_begin_cpu_access(struct dma_buf *buf,
 	struct drm_device *drm = gem->dev;
 
 	if (bo->pages)
-		dma_sync_sg_for_cpu(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-				    DMA_FROM_DEVICE);
+		dma_sync_sgtable_for_cpu(drm->dev, bo->sgt, DMA_FROM_DEVICE);
 
 	return 0;
 }
@@ -623,8 +617,7 @@ static int tegra_gem_prime_end_cpu_access(struct dma_buf *buf,
 	struct drm_device *drm = gem->dev;
 
 	if (bo->pages)
-		dma_sync_sg_for_device(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-				       DMA_TO_DEVICE);
+		dma_sync_sgtable_for_device(drm->dev, bo->sgt, DMA_TO_DEVICE);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/tegra/output.c b/drivers/gpu/drm/tegra/output.c
index e36e5e7c2f69..5a4fd0dbf4cf 100644
--- a/drivers/gpu/drm/tegra/output.c
+++ b/drivers/gpu/drm/tegra/output.c
@@ -5,6 +5,7 @@
  */
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_of.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_simple_kms_helper.h>
 
@@ -99,27 +100,38 @@ int tegra_output_probe(struct tegra_output *output)
 	if (!output->of_node)
 		output->of_node = output->dev->of_node;
 
+	err = drm_of_find_panel_or_bridge(output->of_node, -1, -1,
+					  &output->panel, &output->bridge);
+	if (err && err != -ENODEV)
+		return err;
+
 	panel = of_parse_phandle(output->of_node, "nvidia,panel", 0);
 	if (panel) {
+		/*
+		 * Don't mix nvidia,panel phandle with the graph in a
+		 * device-tree.
+		 */
+		WARN_ON(output->panel || output->bridge);
+
 		output->panel = of_drm_find_panel(panel);
+		of_node_put(panel);
+
 		if (IS_ERR(output->panel))
 			return PTR_ERR(output->panel);
-
-		of_node_put(panel);
 	}
 
 	output->edid = of_get_property(output->of_node, "nvidia,edid", &size);
 
 	ddc = of_parse_phandle(output->of_node, "nvidia,ddc-i2c-bus", 0);
 	if (ddc) {
-		output->ddc = of_find_i2c_adapter_by_node(ddc);
+		output->ddc = of_get_i2c_adapter_by_node(ddc);
+		of_node_put(ddc);
+
 		if (!output->ddc) {
 			err = -EPROBE_DEFER;
 			of_node_put(ddc);
 			return err;
 		}
-
-		of_node_put(ddc);
 	}
 
 	output->hpd_gpio = devm_gpiod_get_from_of_node(output->dev,
@@ -173,19 +185,12 @@ void tegra_output_remove(struct tegra_output *output)
 		free_irq(output->hpd_irq, output);
 
 	if (output->ddc)
-		put_device(&output->ddc->dev);
+		i2c_put_adapter(output->ddc);
 }
 
 int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
 {
 	int connector_type;
-	int err;
-
-	if (output->panel) {
-		err = drm_panel_attach(output->panel, &output->connector);
-		if (err < 0)
-			return err;
-	}
 
 	/*
 	 * The connector is now registered and ready to receive hotplug events
@@ -220,9 +225,6 @@ void tegra_output_exit(struct tegra_output *output)
 	 */
 	if (output->hpd_gpio)
 		disable_irq(output->hpd_irq);
-
-	if (output->panel)
-		drm_panel_detach(output->panel);
 }
 
 void tegra_output_find_possible_crtcs(struct tegra_output *output,
diff --git a/drivers/gpu/drm/tegra/plane.c b/drivers/gpu/drm/tegra/plane.c
index 4cd0461cc508..539d14935728 100644
--- a/drivers/gpu/drm/tegra/plane.c
+++ b/drivers/gpu/drm/tegra/plane.c
@@ -131,12 +131,9 @@ static int tegra_dc_pin(struct tegra_dc *dc, struct tegra_plane_state *state)
 		}
 
 		if (sgt) {
-			err = dma_map_sg(dc->dev, sgt->sgl, sgt->nents,
-					 DMA_TO_DEVICE);
-			if (err == 0) {
-				err = -ENOMEM;
+			err = dma_map_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0);
+			if (err)
 				goto unpin;
-			}
 
 			/*
 			 * The display controller needs contiguous memory, so
@@ -144,7 +141,7 @@ static int tegra_dc_pin(struct tegra_dc *dc, struct tegra_plane_state *state)
 			 * map its SG table to a single contiguous chunk of
 			 * I/O virtual memory.
 			 */
-			if (err > 1) {
+			if (sgt->nents > 1) {
 				err = -EINVAL;
 				goto unpin;
 			}
@@ -166,8 +163,7 @@ unpin:
 		struct sg_table *sgt = state->sgt[i];
 
 		if (sgt)
-			dma_unmap_sg(dc->dev, sgt->sgl, sgt->nents,
-				     DMA_TO_DEVICE);
+			dma_unmap_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0);
 
 		host1x_bo_unpin(dc->dev, &bo->base, sgt);
 		state->iova[i] = DMA_MAPPING_ERROR;
@@ -186,8 +182,7 @@ static void tegra_dc_unpin(struct tegra_dc *dc, struct tegra_plane_state *state)
 		struct sg_table *sgt = state->sgt[i];
 
 		if (sgt)
-			dma_unmap_sg(dc->dev, sgt->sgl, sgt->nents,
-				     DMA_TO_DEVICE);
+			dma_unmap_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0);
 
 		host1x_bo_unpin(dc->dev, &bo->base, sgt);
 		state->iova[i] = DMA_MAPPING_ERROR;
diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index 0562a7eb793f..4142a56ca764 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -7,7 +7,7 @@
 #include <linux/clk.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_panel.h>
+#include <drm/drm_bridge_connector.h>
 #include <drm/drm_simple_kms_helper.h>
 
 #include "drm.h"
@@ -85,45 +85,13 @@ static void tegra_dc_write_regs(struct tegra_dc *dc,
 		tegra_dc_writel(dc, table[i].value, table[i].offset);
 }
 
-static const struct drm_connector_funcs tegra_rgb_connector_funcs = {
-	.reset = drm_atomic_helper_connector_reset,
-	.detect = tegra_output_connector_detect,
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = tegra_output_connector_destroy,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
-static enum drm_mode_status
-tegra_rgb_connector_mode_valid(struct drm_connector *connector,
-			       struct drm_display_mode *mode)
-{
-	/*
-	 * FIXME: For now, always assume that the mode is okay. There are
-	 * unresolved issues with clk_round_rate(), which doesn't always
-	 * reliably report whether a frequency can be set or not.
-	 */
-	return MODE_OK;
-}
-
-static const struct drm_connector_helper_funcs tegra_rgb_connector_helper_funcs = {
-	.get_modes = tegra_output_connector_get_modes,
-	.mode_valid = tegra_rgb_connector_mode_valid,
-};
-
 static void tegra_rgb_encoder_disable(struct drm_encoder *encoder)
 {
 	struct tegra_output *output = encoder_to_output(encoder);
 	struct tegra_rgb *rgb = to_rgb(output);
 
-	if (output->panel)
-		drm_panel_disable(output->panel);
-
 	tegra_dc_write_regs(rgb->dc, rgb_disable, ARRAY_SIZE(rgb_disable));
 	tegra_dc_commit(rgb->dc);
-
-	if (output->panel)
-		drm_panel_unprepare(output->panel);
 }
 
 static void tegra_rgb_encoder_enable(struct drm_encoder *encoder)
@@ -132,9 +100,6 @@ static void tegra_rgb_encoder_enable(struct drm_encoder *encoder)
 	struct tegra_rgb *rgb = to_rgb(output);
 	u32 value;
 
-	if (output->panel)
-		drm_panel_prepare(output->panel);
-
 	tegra_dc_write_regs(rgb->dc, rgb_enable, ARRAY_SIZE(rgb_enable));
 
 	value = DE_SELECT_ACTIVE | DE_CONTROL_NORMAL;
@@ -156,9 +121,6 @@ static void tegra_rgb_encoder_enable(struct drm_encoder *encoder)
 	tegra_dc_writel(rgb->dc, value, DC_DISP_SHIFT_CLOCK_OPTIONS);
 
 	tegra_dc_commit(rgb->dc);
-
-	if (output->panel)
-		drm_panel_enable(output->panel);
 }
 
 static int
@@ -267,24 +229,68 @@ int tegra_dc_rgb_remove(struct tegra_dc *dc)
 int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc)
 {
 	struct tegra_output *output = dc->rgb;
+	struct drm_connector *connector;
 	int err;
 
 	if (!dc->rgb)
 		return -ENODEV;
 
-	drm_connector_init(drm, &output->connector, &tegra_rgb_connector_funcs,
-			   DRM_MODE_CONNECTOR_LVDS);
-	drm_connector_helper_add(&output->connector,
-				 &tegra_rgb_connector_helper_funcs);
-	output->connector.dpms = DRM_MODE_DPMS_OFF;
-
 	drm_simple_encoder_init(drm, &output->encoder, DRM_MODE_ENCODER_LVDS);
 	drm_encoder_helper_add(&output->encoder,
 			       &tegra_rgb_encoder_helper_funcs);
 
-	drm_connector_attach_encoder(&output->connector,
-					  &output->encoder);
-	drm_connector_register(&output->connector);
+	/*
+	 * Wrap directly-connected panel into DRM bridge in order to let
+	 * DRM core to handle panel for us.
+	 */
+	if (output->panel) {
+		output->bridge = devm_drm_panel_bridge_add(output->dev,
+							   output->panel);
+		if (IS_ERR(output->bridge)) {
+			dev_err(output->dev,
+				"failed to wrap panel into bridge: %pe\n",
+				output->bridge);
+			return PTR_ERR(output->bridge);
+		}
+
+		output->panel = NULL;
+	}
+
+	/*
+	 * Tegra devices that have LVDS panel utilize LVDS encoder bridge
+	 * for converting up to 28 LCD LVTTL lanes into 5/4 LVDS lanes that
+	 * go to display panel's receiver.
+	 *
+	 * Encoder usually have a power-down control which needs to be enabled
+	 * in order to transmit data to the panel.  Historically devices that
+	 * use an older device-tree version didn't model the bridge, assuming
+	 * that encoder is turned ON by default, while today's DRM allows us
+	 * to model LVDS encoder properly.
+	 *
+	 * Newer device-trees utilize LVDS encoder bridge, which provides
+	 * us with a connector and handles the display panel.
+	 *
+	 * For older device-trees we wrapped panel into the panel-bridge.
+	 */
+	if (output->bridge) {
+		err = drm_bridge_attach(&output->encoder, output->bridge,
+					NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+		if (err) {
+			dev_err(output->dev, "failed to attach bridge: %d\n",
+				err);
+			return err;
+		}
+
+		connector = drm_bridge_connector_init(drm, &output->encoder);
+		if (IS_ERR(connector)) {
+			dev_err(output->dev,
+				"failed to initialize bridge connector: %pe\n",
+				connector);
+			return PTR_ERR(connector);
+		}
+
+		drm_connector_attach_encoder(connector, &output->encoder);
+	}
 
 	err = tegra_output_init(drm, output);
 	if (err < 0) {
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index 45b5258c77a2..e88a17c2937f 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -3728,7 +3728,12 @@ static int tegra_sor_probe(struct platform_device *pdev)
 		if (!sor->aux)
 			return -EPROBE_DEFER;
 
-		sor->output.ddc = &sor->aux->ddc;
+		if (get_device(&sor->aux->ddc.dev)) {
+			if (try_module_get(sor->aux->ddc.owner))
+				sor->output.ddc = &sor->aux->ddc;
+			else
+				put_device(&sor->aux->ddc.dev);
+		}
 	}
 
 	if (!sor->aux) {
diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index c3ece2c9d1c8..b669168ae7cb 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -19,6 +19,7 @@
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/regmap.h>
+#include <linux/sys_soc.h>
 
 #include <drm/drm_fourcc.h>
 #include <drm/drm_fb_cma_helper.h>
@@ -302,6 +303,8 @@ struct dispc_device {
 	u32 num_fourccs;
 
 	u32 memory_bandwidth_limit;
+
+	struct dispc_errata errata;
 };
 
 static void dispc_write(struct dispc_device *dispc, u16 reg, u32 val)
@@ -2641,6 +2644,19 @@ static int dispc_init_am65x_oldi_io_ctrl(struct device *dev,
 	return 0;
 }
 
+static void dispc_init_errata(struct dispc_device *dispc)
+{
+	static const struct soc_device_attribute am65x_sr10_soc_devices[] = {
+		{ .family = "AM65X", .revision = "SR1.0" },
+		{ /* sentinel */ }
+	};
+
+	if (soc_device_match(am65x_sr10_soc_devices)) {
+		dispc->errata.i2000 = true;
+		dev_info(dispc->dev, "WA for erratum i2000: YUV formats disabled\n");
+	}
+}
+
 int dispc_init(struct tidss_device *tidss)
 {
 	struct device *dev = tidss->dev;
@@ -2664,19 +2680,27 @@ int dispc_init(struct tidss_device *tidss)
 	if (!dispc)
 		return -ENOMEM;
 
+	dispc->tidss = tidss;
+	dispc->dev = dev;
+	dispc->feat = feat;
+
+	dispc_init_errata(dispc);
+
 	dispc->fourccs = devm_kcalloc(dev, ARRAY_SIZE(dispc_color_formats),
 				      sizeof(*dispc->fourccs), GFP_KERNEL);
 	if (!dispc->fourccs)
 		return -ENOMEM;
 
 	num_fourccs = 0;
-	for (i = 0; i < ARRAY_SIZE(dispc_color_formats); ++i)
+	for (i = 0; i < ARRAY_SIZE(dispc_color_formats); ++i) {
+		if (dispc->errata.i2000 &&
+		    dispc_fourcc_is_yuv(dispc_color_formats[i].fourcc)) {
+			continue;
+		}
 		dispc->fourccs[num_fourccs++] = dispc_color_formats[i].fourcc;
+	}
 
 	dispc->num_fourccs = num_fourccs;
-	dispc->tidss = tidss;
-	dispc->dev = dev;
-	dispc->feat = feat;
 
 	dispc_common_regmap = dispc->feat->common_regs;
 
diff --git a/drivers/gpu/drm/tidss/tidss_dispc.h b/drivers/gpu/drm/tidss/tidss_dispc.h
index 5984e0de2cd9..e49432f0abf5 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.h
+++ b/drivers/gpu/drm/tidss/tidss_dispc.h
@@ -46,6 +46,10 @@ struct dispc_features_scaling {
 	u32 xinc_max;
 };
 
+struct dispc_errata {
+	bool i2000; /* DSS Does Not Support YUV Pixel Data Formats */
+};
+
 enum dispc_vp_bus_type {
 	DISPC_VP_DPI,		/* DPI output */
 	DISPC_VP_OLDI,		/* OLDI (LVDS) output */
diff --git a/drivers/gpu/drm/ttm/Makefile b/drivers/gpu/drm/ttm/Makefile
index caea2a099496..90c0da88cc98 100644
--- a/drivers/gpu/drm/ttm/Makefile
+++ b/drivers/gpu/drm/ttm/Makefile
@@ -4,7 +4,8 @@
 
 ttm-y := ttm_memory.o ttm_tt.o ttm_bo.o \
 	ttm_bo_util.o ttm_bo_vm.o ttm_module.o \
-	ttm_execbuf_util.o ttm_page_alloc.o ttm_bo_manager.o
+	ttm_execbuf_util.o ttm_page_alloc.o ttm_range_manager.o \
+	ttm_resource.o
 ttm-$(CONFIG_AGP) += ttm_agp_backend.o
 ttm-$(CONFIG_DRM_TTM_DMA_PAGE_POOL) += ttm_page_alloc_dma.o
 
diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c
index 38f1351140e2..a98fd795b752 100644
--- a/drivers/gpu/drm/ttm/ttm_agp_backend.c
+++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c
@@ -48,7 +48,7 @@ struct ttm_agp_backend {
 	struct agp_bridge_data *bridge;
 };
 
-static int ttm_agp_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
+int ttm_agp_bind(struct ttm_tt *ttm, struct ttm_resource *bo_mem)
 {
 	struct ttm_agp_backend *agp_be = container_of(ttm, struct ttm_agp_backend, ttm);
 	struct page *dummy_read_page = ttm_bo_glob.dummy_read_page;
@@ -57,6 +57,9 @@ static int ttm_agp_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
 	int ret, cached = (bo_mem->placement & TTM_PL_FLAG_CACHED);
 	unsigned i;
 
+	if (agp_be->mem)
+		return 0;
+
 	mem = agp_allocate_memory(agp_be->bridge, ttm->num_pages, AGP_USER_MEMORY);
 	if (unlikely(mem == NULL))
 		return -ENOMEM;
@@ -81,8 +84,9 @@ static int ttm_agp_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
 
 	return ret;
 }
+EXPORT_SYMBOL(ttm_agp_bind);
 
-static void ttm_agp_unbind(struct ttm_tt *ttm)
+void ttm_agp_unbind(struct ttm_tt *ttm)
 {
 	struct ttm_agp_backend *agp_be = container_of(ttm, struct ttm_agp_backend, ttm);
 
@@ -95,8 +99,20 @@ static void ttm_agp_unbind(struct ttm_tt *ttm)
 		agp_be->mem = NULL;
 	}
 }
+EXPORT_SYMBOL(ttm_agp_unbind);
+
+bool ttm_agp_is_bound(struct ttm_tt *ttm)
+{
+	struct ttm_agp_backend *agp_be = container_of(ttm, struct ttm_agp_backend, ttm);
+
+	if (!ttm)
+		return false;
+
+	return (agp_be->mem != NULL);
+}
+EXPORT_SYMBOL(ttm_agp_is_bound);
 
-static void ttm_agp_destroy(struct ttm_tt *ttm)
+void ttm_agp_destroy(struct ttm_tt *ttm)
 {
 	struct ttm_agp_backend *agp_be = container_of(ttm, struct ttm_agp_backend, ttm);
 
@@ -105,12 +121,7 @@ static void ttm_agp_destroy(struct ttm_tt *ttm)
 	ttm_tt_fini(ttm);
 	kfree(agp_be);
 }
-
-static struct ttm_backend_func ttm_agp_func = {
-	.bind = ttm_agp_bind,
-	.unbind = ttm_agp_unbind,
-	.destroy = ttm_agp_destroy,
-};
+EXPORT_SYMBOL(ttm_agp_destroy);
 
 struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo,
 				 struct agp_bridge_data *bridge,
@@ -124,7 +135,6 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo,
 
 	agp_be->mem = NULL;
 	agp_be->bridge = bridge;
-	agp_be->ttm.func = &ttm_agp_func;
 
 	if (ttm_tt_init(&agp_be->ttm, bo, page_flags)) {
 		kfree(agp_be);
@@ -134,18 +144,3 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo,
 	return &agp_be->ttm;
 }
 EXPORT_SYMBOL(ttm_agp_tt_create);
-
-int ttm_agp_tt_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
-{
-	if (ttm->state != tt_unpopulated)
-		return 0;
-
-	return ttm_pool_populate(ttm, ctx);
-}
-EXPORT_SYMBOL(ttm_agp_tt_populate);
-
-void ttm_agp_tt_unpopulate(struct ttm_tt *ttm)
-{
-	ttm_pool_unpopulate(ttm);
-}
-EXPORT_SYMBOL(ttm_agp_tt_unpopulate);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index cc6a4e7551e3..70b3bee27850 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -64,51 +64,22 @@ static void ttm_bo_default_destroy(struct ttm_buffer_object *bo)
 	kfree(bo);
 }
 
-static inline int ttm_mem_type_from_place(const struct ttm_place *place,
-					  uint32_t *mem_type)
-{
-	int pos;
-
-	pos = ffs(place->flags & TTM_PL_MASK_MEM);
-	if (unlikely(!pos))
-		return -EINVAL;
-
-	*mem_type = pos - 1;
-	return 0;
-}
-
-static void ttm_mem_type_debug(struct ttm_bo_device *bdev, struct drm_printer *p,
-			       int mem_type)
-{
-	struct ttm_mem_type_manager *man = &bdev->man[mem_type];
-
-	drm_printf(p, "    has_type: %d\n", man->has_type);
-	drm_printf(p, "    use_type: %d\n", man->use_type);
-	drm_printf(p, "    flags: 0x%08X\n", man->flags);
-	drm_printf(p, "    size: %llu\n", man->size);
-	drm_printf(p, "    available_caching: 0x%08X\n", man->available_caching);
-	drm_printf(p, "    default_caching: 0x%08X\n", man->default_caching);
-	if (mem_type != TTM_PL_SYSTEM)
-		(*man->func->debug)(man, p);
-}
-
 static void ttm_bo_mem_space_debug(struct ttm_buffer_object *bo,
 					struct ttm_placement *placement)
 {
 	struct drm_printer p = drm_debug_printer(TTM_PFX);
-	int i, ret, mem_type;
+	struct ttm_resource_manager *man;
+	int i, mem_type;
 
 	drm_printf(&p, "No space for %p (%lu pages, %luK, %luM)\n",
 		   bo, bo->mem.num_pages, bo->mem.size >> 10,
 		   bo->mem.size >> 20);
 	for (i = 0; i < placement->num_placement; i++) {
-		ret = ttm_mem_type_from_place(&placement->placement[i],
-						&mem_type);
-		if (ret)
-			return;
+		mem_type = placement->placement[i].mem_type;
 		drm_printf(&p, "  placement[%d]=0x%08X (%d)\n",
 			   i, placement->placement[i].flags, mem_type);
-		ttm_mem_type_debug(bo->bdev, &p, mem_type);
+		man = ttm_manager_type(bo->bdev, mem_type);
+		ttm_resource_manager_debug(man, &p);
 	}
 }
 
@@ -138,17 +109,11 @@ static struct kobj_type ttm_bo_glob_kobj_type  = {
 	.default_attrs = ttm_bo_global_attrs
 };
 
-
-static inline uint32_t ttm_bo_type_flags(unsigned type)
-{
-	return 1 << (type);
-}
-
 static void ttm_bo_add_mem_to_lru(struct ttm_buffer_object *bo,
-				  struct ttm_mem_reg *mem)
+				  struct ttm_resource *mem)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man;
+	struct ttm_resource_manager *man;
 
 	if (!list_empty(&bo->lru))
 		return;
@@ -156,10 +121,10 @@ static void ttm_bo_add_mem_to_lru(struct ttm_buffer_object *bo,
 	if (mem->placement & TTM_PL_FLAG_NO_EVICT)
 		return;
 
-	man = &bdev->man[mem->mem_type];
+	man = ttm_manager_type(bdev, mem->mem_type);
 	list_add_tail(&bo->lru, &man->lru[bo->priority]);
 
-	if (!(man->flags & TTM_MEMTYPE_FLAG_FIXED) && bo->ttm &&
+	if (man->use_tt && bo->ttm &&
 	    !(bo->ttm->page_flags & (TTM_PAGE_FLAG_SG |
 				     TTM_PAGE_FLAG_SWAPPED))) {
 		list_add_tail(&bo->swap, &ttm_bo_glob.swap_lru[bo->priority]);
@@ -223,7 +188,7 @@ void ttm_bo_bulk_move_lru_tail(struct ttm_lru_bulk_move *bulk)
 
 	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) {
 		struct ttm_lru_bulk_move_pos *pos = &bulk->tt[i];
-		struct ttm_mem_type_manager *man;
+		struct ttm_resource_manager *man;
 
 		if (!pos->first)
 			continue;
@@ -231,14 +196,14 @@ void ttm_bo_bulk_move_lru_tail(struct ttm_lru_bulk_move *bulk)
 		dma_resv_assert_held(pos->first->base.resv);
 		dma_resv_assert_held(pos->last->base.resv);
 
-		man = &pos->first->bdev->man[TTM_PL_TT];
+		man = ttm_manager_type(pos->first->bdev, TTM_PL_TT);
 		list_bulk_move_tail(&man->lru[i], &pos->first->lru,
 				    &pos->last->lru);
 	}
 
 	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) {
 		struct ttm_lru_bulk_move_pos *pos = &bulk->vram[i];
-		struct ttm_mem_type_manager *man;
+		struct ttm_resource_manager *man;
 
 		if (!pos->first)
 			continue;
@@ -246,7 +211,7 @@ void ttm_bo_bulk_move_lru_tail(struct ttm_lru_bulk_move *bulk)
 		dma_resv_assert_held(pos->first->base.resv);
 		dma_resv_assert_held(pos->last->base.resv);
 
-		man = &pos->first->bdev->man[TTM_PL_VRAM];
+		man = ttm_manager_type(pos->first->bdev, TTM_PL_VRAM);
 		list_bulk_move_tail(&man->lru[i], &pos->first->lru,
 				    &pos->last->lru);
 	}
@@ -268,38 +233,38 @@ void ttm_bo_bulk_move_lru_tail(struct ttm_lru_bulk_move *bulk)
 EXPORT_SYMBOL(ttm_bo_bulk_move_lru_tail);
 
 static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
-				  struct ttm_mem_reg *mem, bool evict,
+				  struct ttm_resource *mem, bool evict,
 				  struct ttm_operation_ctx *ctx)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *old_man = &bdev->man[bo->mem.mem_type];
-	struct ttm_mem_type_manager *new_man = &bdev->man[mem->mem_type];
+	struct ttm_resource_manager *old_man = ttm_manager_type(bdev, bo->mem.mem_type);
+	struct ttm_resource_manager *new_man = ttm_manager_type(bdev, mem->mem_type);
 	int ret;
 
-	ret = ttm_mem_io_lock(old_man, true);
-	if (unlikely(ret != 0))
-		goto out_err;
-	ttm_bo_unmap_virtual_locked(bo);
-	ttm_mem_io_unlock(old_man);
+	ttm_bo_unmap_virtual(bo);
 
 	/*
 	 * Create and bind a ttm if required.
 	 */
 
-	if (!(new_man->flags & TTM_MEMTYPE_FLAG_FIXED)) {
-		if (bo->ttm == NULL) {
-			bool zero = !(old_man->flags & TTM_MEMTYPE_FLAG_FIXED);
-			ret = ttm_tt_create(bo, zero);
-			if (ret)
-				goto out_err;
-		}
+	if (new_man->use_tt) {
+		/* Zero init the new TTM structure if the old location should
+		 * have used one as well.
+		 */
+		ret = ttm_tt_create(bo, old_man->use_tt);
+		if (ret)
+			goto out_err;
 
 		ret = ttm_tt_set_placement_caching(bo->ttm, mem->placement);
 		if (ret)
 			goto out_err;
 
 		if (mem->mem_type != TTM_PL_SYSTEM) {
-			ret = ttm_tt_bind(bo->ttm, mem, ctx);
+			ret = ttm_tt_populate(bdev, bo->ttm, ctx);
+			if (ret)
+				goto out_err;
+
+			ret = ttm_bo_tt_bind(bo, mem);
 			if (ret)
 				goto out_err;
 		}
@@ -315,8 +280,7 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
 	if (bdev->driver->move_notify)
 		bdev->driver->move_notify(bo, evict, mem);
 
-	if (!(old_man->flags & TTM_MEMTYPE_FLAG_FIXED) &&
-	    !(new_man->flags & TTM_MEMTYPE_FLAG_FIXED))
+	if (old_man->use_tt && new_man->use_tt)
 		ret = ttm_bo_move_ttm(bo, ctx, mem);
 	else if (bdev->driver->move)
 		ret = bdev->driver->move(bo, evict, ctx, mem);
@@ -334,17 +298,13 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
 	}
 
 moved:
-	bo->evicted = false;
-
 	ctx->bytes_moved += bo->num_pages << PAGE_SHIFT;
 	return 0;
 
 out_err:
-	new_man = &bdev->man[bo->mem.mem_type];
-	if (new_man->flags & TTM_MEMTYPE_FLAG_FIXED) {
-		ttm_tt_destroy(bo->ttm);
-		bo->ttm = NULL;
-	}
+	new_man = ttm_manager_type(bdev, bo->mem.mem_type);
+	if (!new_man->use_tt)
+		ttm_bo_tt_destroy(bo);
 
 	return ret;
 }
@@ -362,9 +322,8 @@ static void ttm_bo_cleanup_memtype_use(struct ttm_buffer_object *bo)
 	if (bo->bdev->driver->move_notify)
 		bo->bdev->driver->move_notify(bo, false, NULL);
 
-	ttm_tt_destroy(bo->ttm);
-	bo->ttm = NULL;
-	ttm_bo_mem_put(bo, &bo->mem);
+	ttm_bo_tt_destroy(bo);
+	ttm_resource_free(bo, &bo->mem);
 }
 
 static int ttm_bo_individualize_resv(struct ttm_buffer_object *bo)
@@ -552,7 +511,6 @@ static void ttm_bo_release(struct kref *kref)
 	struct ttm_buffer_object *bo =
 	    container_of(kref, struct ttm_buffer_object, kref);
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type];
 	size_t acc_size = bo->acc_size;
 	int ret;
 
@@ -570,9 +528,7 @@ static void ttm_bo_release(struct kref *kref)
 			bo->bdev->driver->release_notify(bo);
 
 		drm_vma_offset_remove(bdev->vma_manager, &bo->base.vma_node);
-		ttm_mem_io_lock(man, false);
-		ttm_mem_io_free_vm(bo);
-		ttm_mem_io_unlock(man);
+		ttm_mem_io_free(bdev, &bo->mem);
 	}
 
 	if (!dma_resv_test_signaled_rcu(bo->base.resv, true) ||
@@ -643,7 +599,7 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo,
 			struct ttm_operation_ctx *ctx)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_reg evict_mem;
+	struct ttm_resource evict_mem;
 	struct ttm_placement placement;
 	int ret = 0;
 
@@ -654,17 +610,16 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo,
 	bdev->driver->evict_flags(bo, &placement);
 
 	if (!placement.num_placement && !placement.num_busy_placement) {
-		ret = ttm_bo_pipeline_gutting(bo);
-		if (ret)
-			return ret;
+		ttm_bo_wait(bo, false, false);
 
+		ttm_bo_cleanup_memtype_use(bo);
 		return ttm_tt_create(bo, false);
 	}
 
 	evict_mem = bo->mem;
 	evict_mem.mm_node = NULL;
-	evict_mem.bus.io_reserved_vm = false;
-	evict_mem.bus.io_reserved_count = 0;
+	evict_mem.bus.offset = 0;
+	evict_mem.bus.addr = NULL;
 
 	ret = ttm_bo_mem_space(bo, &placement, &evict_mem, ctx);
 	if (ret) {
@@ -680,10 +635,8 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo,
 	if (unlikely(ret)) {
 		if (ret != -ERESTARTSYS)
 			pr_err("Buffer eviction failed\n");
-		ttm_bo_mem_put(bo, &evict_mem);
-		goto out;
+		ttm_resource_free(bo, &evict_mem);
 	}
-	bo->evicted = true;
 out:
 	return ret;
 }
@@ -769,14 +722,13 @@ static int ttm_mem_evict_wait_busy(struct ttm_buffer_object *busy_bo,
 	return r == -EDEADLK ? -EBUSY : r;
 }
 
-static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
-			       uint32_t mem_type,
-			       const struct ttm_place *place,
-			       struct ttm_operation_ctx *ctx,
-			       struct ww_acquire_ctx *ticket)
+int ttm_mem_evict_first(struct ttm_bo_device *bdev,
+			struct ttm_resource_manager *man,
+			const struct ttm_place *place,
+			struct ttm_operation_ctx *ctx,
+			struct ww_acquire_ctx *ticket)
 {
 	struct ttm_buffer_object *bo = NULL, *busy_bo = NULL;
-	struct ttm_mem_type_manager *man = &bdev->man[mem_type];
 	bool locked = false;
 	unsigned i;
 	int ret;
@@ -842,38 +794,12 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
 	return ret;
 }
 
-static int ttm_bo_mem_get(struct ttm_buffer_object *bo,
-			  const struct ttm_place *place,
-			  struct ttm_mem_reg *mem)
-{
-	struct ttm_mem_type_manager *man = &bo->bdev->man[mem->mem_type];
-
-	mem->mm_node = NULL;
-	if (!man->func || !man->func->get_node)
-		return 0;
-
-	return man->func->get_node(man, bo, place, mem);
-}
-
-void ttm_bo_mem_put(struct ttm_buffer_object *bo, struct ttm_mem_reg *mem)
-{
-	struct ttm_mem_type_manager *man = &bo->bdev->man[mem->mem_type];
-
-	if (!man->func || !man->func->put_node)
-		return;
-
-	man->func->put_node(man, mem);
-	mem->mm_node = NULL;
-	mem->mem_type = TTM_PL_SYSTEM;
-}
-EXPORT_SYMBOL(ttm_bo_mem_put);
-
 /**
  * Add the last move fence to the BO and reserve a new shared slot.
  */
 static int ttm_bo_add_move_fence(struct ttm_buffer_object *bo,
-				 struct ttm_mem_type_manager *man,
-				 struct ttm_mem_reg *mem,
+				 struct ttm_resource_manager *man,
+				 struct ttm_resource *mem,
 				 bool no_wait_gpu)
 {
 	struct dma_fence *fence;
@@ -910,22 +836,22 @@ static int ttm_bo_add_move_fence(struct ttm_buffer_object *bo,
  */
 static int ttm_bo_mem_force_space(struct ttm_buffer_object *bo,
 				  const struct ttm_place *place,
-				  struct ttm_mem_reg *mem,
+				  struct ttm_resource *mem,
 				  struct ttm_operation_ctx *ctx)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
+	struct ttm_resource_manager *man = ttm_manager_type(bdev, mem->mem_type);
 	struct ww_acquire_ctx *ticket;
 	int ret;
 
 	ticket = dma_resv_locking_ctx(bo->base.resv);
 	do {
-		ret = ttm_bo_mem_get(bo, place, mem);
+		ret = ttm_resource_alloc(bo, place, mem);
 		if (likely(!ret))
 			break;
 		if (unlikely(ret != -ENOSPC))
 			return ret;
-		ret = ttm_mem_evict_first(bdev, mem->mem_type, place, ctx,
+		ret = ttm_mem_evict_first(bdev, man, place, ctx,
 					  ticket);
 		if (unlikely(ret != 0))
 			return ret;
@@ -934,7 +860,7 @@ static int ttm_bo_mem_force_space(struct ttm_buffer_object *bo,
 	return ttm_bo_add_move_fence(bo, man, mem, ctx->no_wait_gpu);
 }
 
-static uint32_t ttm_bo_select_caching(struct ttm_mem_type_manager *man,
+static uint32_t ttm_bo_select_caching(struct ttm_resource_manager *man,
 				      uint32_t cur_placement,
 				      uint32_t proposed_placement)
 {
@@ -947,8 +873,6 @@ static uint32_t ttm_bo_select_caching(struct ttm_mem_type_manager *man,
 
 	if ((cur_placement & caching) != 0)
 		result |= (cur_placement & caching);
-	else if ((man->default_caching & caching) != 0)
-		result |= man->default_caching;
 	else if ((TTM_PL_FLAG_CACHED & caching) != 0)
 		result |= TTM_PL_FLAG_CACHED;
 	else if ((TTM_PL_FLAG_WC & caching) != 0)
@@ -959,25 +883,6 @@ static uint32_t ttm_bo_select_caching(struct ttm_mem_type_manager *man,
 	return result;
 }
 
-static bool ttm_bo_mt_compatible(struct ttm_mem_type_manager *man,
-				 uint32_t mem_type,
-				 const struct ttm_place *place,
-				 uint32_t *masked_placement)
-{
-	uint32_t cur_flags = ttm_bo_type_flags(mem_type);
-
-	if ((cur_flags & place->flags & TTM_PL_MASK_MEM) == 0)
-		return false;
-
-	if ((place->flags & man->available_caching) == 0)
-		return false;
-
-	cur_flags |= (place->flags & man->available_caching);
-
-	*masked_placement = cur_flags;
-	return true;
-}
-
 /**
  * ttm_bo_mem_placement - check if placement is compatible
  * @bo: BO to find memory for
@@ -991,34 +896,22 @@ static bool ttm_bo_mt_compatible(struct ttm_mem_type_manager *man,
  */
 static int ttm_bo_mem_placement(struct ttm_buffer_object *bo,
 				const struct ttm_place *place,
-				struct ttm_mem_reg *mem,
+				struct ttm_resource *mem,
 				struct ttm_operation_ctx *ctx)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	uint32_t mem_type = TTM_PL_SYSTEM;
-	struct ttm_mem_type_manager *man;
+	struct ttm_resource_manager *man;
 	uint32_t cur_flags = 0;
-	int ret;
-
-	ret = ttm_mem_type_from_place(place, &mem_type);
-	if (ret)
-		return ret;
-
-	man = &bdev->man[mem_type];
-	if (!man->has_type || !man->use_type)
-		return -EBUSY;
 
-	if (!ttm_bo_mt_compatible(man, mem_type, place, &cur_flags))
+	man = ttm_manager_type(bdev, place->mem_type);
+	if (!man || !ttm_resource_manager_used(man))
 		return -EBUSY;
 
-	cur_flags = ttm_bo_select_caching(man, bo->mem.placement, cur_flags);
-	/*
-	 * Use the access and other non-mapping-related flag bits from
-	 * the memory placement flags to the current flags
-	 */
-	ttm_flag_masked(&cur_flags, place->flags, ~TTM_PL_MASK_MEMTYPE);
+	cur_flags = ttm_bo_select_caching(man, bo->mem.placement,
+					  place->flags);
+	cur_flags |= place->flags & ~TTM_PL_MASK_CACHING;
 
-	mem->mem_type = mem_type;
+	mem->mem_type = place->mem_type;
 	mem->placement = cur_flags;
 
 	spin_lock(&ttm_bo_glob.lru_lock);
@@ -1039,7 +932,7 @@ static int ttm_bo_mem_placement(struct ttm_buffer_object *bo,
  */
 int ttm_bo_mem_space(struct ttm_buffer_object *bo,
 			struct ttm_placement *placement,
-			struct ttm_mem_reg *mem,
+			struct ttm_resource *mem,
 			struct ttm_operation_ctx *ctx)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
@@ -1052,25 +945,23 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo,
 
 	for (i = 0; i < placement->num_placement; ++i) {
 		const struct ttm_place *place = &placement->placement[i];
-		struct ttm_mem_type_manager *man;
+		struct ttm_resource_manager *man;
 
 		ret = ttm_bo_mem_placement(bo, place, mem, ctx);
-		if (ret == -EBUSY)
-			continue;
 		if (ret)
-			goto error;
+			continue;
 
 		type_found = true;
-		ret = ttm_bo_mem_get(bo, place, mem);
+		ret = ttm_resource_alloc(bo, place, mem);
 		if (ret == -ENOSPC)
 			continue;
 		if (unlikely(ret))
 			goto error;
 
-		man = &bdev->man[mem->mem_type];
+		man = ttm_manager_type(bdev, mem->mem_type);
 		ret = ttm_bo_add_move_fence(bo, man, mem, ctx->no_wait_gpu);
 		if (unlikely(ret)) {
-			ttm_bo_mem_put(bo, mem);
+			ttm_resource_free(bo, mem);
 			if (ret == -EBUSY)
 				continue;
 
@@ -1083,10 +974,8 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo,
 		const struct ttm_place *place = &placement->busy_placement[i];
 
 		ret = ttm_bo_mem_placement(bo, place, mem, ctx);
-		if (ret == -EBUSY)
-			continue;
 		if (ret)
-			goto error;
+			continue;
 
 		type_found = true;
 		ret = ttm_bo_mem_force_space(bo, place, mem, ctx);
@@ -1105,9 +994,7 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo,
 
 error:
 	if (bo->mem.mem_type == TTM_PL_SYSTEM && !list_empty(&bo->lru)) {
-		spin_lock(&ttm_bo_glob.lru_lock);
-		ttm_bo_move_to_lru_tail(bo, NULL);
-		spin_unlock(&ttm_bo_glob.lru_lock);
+		ttm_bo_move_to_lru_tail_unlocked(bo);
 	}
 
 	return ret;
@@ -1119,15 +1006,15 @@ static int ttm_bo_move_buffer(struct ttm_buffer_object *bo,
 			      struct ttm_operation_ctx *ctx)
 {
 	int ret = 0;
-	struct ttm_mem_reg mem;
+	struct ttm_resource mem;
 
 	dma_resv_assert_held(bo->base.resv);
 
 	mem.num_pages = bo->num_pages;
 	mem.size = mem.num_pages << PAGE_SHIFT;
 	mem.page_alignment = bo->mem.page_alignment;
-	mem.bus.io_reserved_vm = false;
-	mem.bus.io_reserved_count = 0;
+	mem.bus.offset = 0;
+	mem.bus.addr = NULL;
 	mem.mm_node = NULL;
 
 	/*
@@ -1139,13 +1026,13 @@ static int ttm_bo_move_buffer(struct ttm_buffer_object *bo,
 	ret = ttm_bo_handle_move_mem(bo, &mem, false, ctx);
 out_unlock:
 	if (ret)
-		ttm_bo_mem_put(bo, &mem);
+		ttm_resource_free(bo, &mem);
 	return ret;
 }
 
 static bool ttm_bo_places_compat(const struct ttm_place *places,
 				 unsigned num_placement,
-				 struct ttm_mem_reg *mem,
+				 struct ttm_resource *mem,
 				 uint32_t *new_flags)
 {
 	unsigned i;
@@ -1159,7 +1046,7 @@ static bool ttm_bo_places_compat(const struct ttm_place *places,
 
 		*new_flags = heap->flags;
 		if ((*new_flags & mem->placement & TTM_PL_MASK_CACHING) &&
-		    (*new_flags & mem->placement & TTM_PL_MASK_MEM) &&
+		    (mem->mem_type == heap->mem_type) &&
 		    (!(*new_flags & TTM_PL_FLAG_CONTIGUOUS) ||
 		     (mem->placement & TTM_PL_FLAG_CONTIGUOUS)))
 			return true;
@@ -1168,7 +1055,7 @@ static bool ttm_bo_places_compat(const struct ttm_place *places,
 }
 
 bool ttm_bo_mem_compat(struct ttm_placement *placement,
-		       struct ttm_mem_reg *mem,
+		       struct ttm_resource *mem,
 		       uint32_t *new_flags)
 {
 	if (ttm_bo_places_compat(placement->placement, placement->num_placement,
@@ -1214,17 +1101,13 @@ int ttm_bo_validate(struct ttm_buffer_object *bo,
 		if (ret)
 			return ret;
 	} else {
-		/*
-		 * Use the access and other non-mapping-related flag bits from
-		 * the compatible memory placement flags to the active flags
-		 */
-		ttm_flag_masked(&bo->mem.placement, new_flags,
-				~TTM_PL_MASK_MEMTYPE);
+		bo->mem.placement &= TTM_PL_MASK_CACHING;
+		bo->mem.placement |= new_flags & ~TTM_PL_MASK_CACHING;
 	}
 	/*
 	 * We might need to add a TTM.
 	 */
-	if (bo->mem.mem_type == TTM_PL_SYSTEM && bo->ttm == NULL) {
+	if (bo->mem.mem_type == TTM_PL_SYSTEM) {
 		ret = ttm_tt_create(bo, true);
 		if (ret)
 			return ret;
@@ -1276,7 +1159,6 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev,
 	INIT_LIST_HEAD(&bo->lru);
 	INIT_LIST_HEAD(&bo->ddestroy);
 	INIT_LIST_HEAD(&bo->swap);
-	INIT_LIST_HEAD(&bo->io_reserve_lru);
 	bo->bdev = bdev;
 	bo->type = type;
 	bo->num_pages = num_pages;
@@ -1285,10 +1167,10 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev,
 	bo->mem.num_pages = bo->num_pages;
 	bo->mem.mm_node = NULL;
 	bo->mem.page_alignment = page_alignment;
-	bo->mem.bus.io_reserved_vm = false;
-	bo->mem.bus.io_reserved_count = 0;
+	bo->mem.bus.offset = 0;
+	bo->mem.bus.addr = NULL;
 	bo->moving = NULL;
-	bo->mem.placement = (TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED);
+	bo->mem.placement = TTM_PL_FLAG_CACHED;
 	bo->acc_size = acc_size;
 	bo->sg = sg;
 	if (resv) {
@@ -1335,9 +1217,7 @@ int ttm_bo_init_reserved(struct ttm_bo_device *bdev,
 		return ret;
 	}
 
-	spin_lock(&ttm_bo_glob.lru_lock);
-	ttm_bo_move_to_lru_tail(bo, NULL);
-	spin_unlock(&ttm_bo_glob.lru_lock);
+	ttm_bo_move_to_lru_tail_unlocked(bo);
 
 	return ret;
 }
@@ -1371,9 +1251,9 @@ int ttm_bo_init(struct ttm_bo_device *bdev,
 }
 EXPORT_SYMBOL(ttm_bo_init);
 
-size_t ttm_bo_acc_size(struct ttm_bo_device *bdev,
-		       unsigned long bo_size,
-		       unsigned struct_size)
+static size_t ttm_bo_acc_size(struct ttm_bo_device *bdev,
+			      unsigned long bo_size,
+			      unsigned struct_size)
 {
 	unsigned npages = (PAGE_ALIGN(bo_size)) >> PAGE_SHIFT;
 	size_t size = 0;
@@ -1383,7 +1263,6 @@ size_t ttm_bo_acc_size(struct ttm_bo_device *bdev,
 	size += ttm_round_pot(sizeof(struct ttm_tt));
 	return size;
 }
-EXPORT_SYMBOL(ttm_bo_acc_size);
 
 size_t ttm_bo_dma_acc_size(struct ttm_bo_device *bdev,
 			   unsigned long bo_size,
@@ -1426,144 +1305,24 @@ int ttm_bo_create(struct ttm_bo_device *bdev,
 }
 EXPORT_SYMBOL(ttm_bo_create);
 
-static int ttm_bo_force_list_clean(struct ttm_bo_device *bdev,
-				   unsigned mem_type)
-{
-	struct ttm_operation_ctx ctx = {
-		.interruptible = false,
-		.no_wait_gpu = false,
-		.flags = TTM_OPT_FLAG_FORCE_ALLOC
-	};
-	struct ttm_mem_type_manager *man = &bdev->man[mem_type];
-	struct ttm_bo_global *glob = &ttm_bo_glob;
-	struct dma_fence *fence;
-	int ret;
-	unsigned i;
-
-	/*
-	 * Can't use standard list traversal since we're unlocking.
-	 */
-
-	spin_lock(&glob->lru_lock);
-	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) {
-		while (!list_empty(&man->lru[i])) {
-			spin_unlock(&glob->lru_lock);
-			ret = ttm_mem_evict_first(bdev, mem_type, NULL, &ctx,
-						  NULL);
-			if (ret)
-				return ret;
-			spin_lock(&glob->lru_lock);
-		}
-	}
-	spin_unlock(&glob->lru_lock);
-
-	spin_lock(&man->move_lock);
-	fence = dma_fence_get(man->move);
-	spin_unlock(&man->move_lock);
-
-	if (fence) {
-		ret = dma_fence_wait(fence, false);
-		dma_fence_put(fence);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-int ttm_bo_clean_mm(struct ttm_bo_device *bdev, unsigned mem_type)
-{
-	struct ttm_mem_type_manager *man;
-	int ret = -EINVAL;
-
-	if (mem_type >= TTM_NUM_MEM_TYPES) {
-		pr_err("Illegal memory type %d\n", mem_type);
-		return ret;
-	}
-	man = &bdev->man[mem_type];
-
-	if (!man->has_type) {
-		pr_err("Trying to take down uninitialized memory manager type %u\n",
-		       mem_type);
-		return ret;
-	}
-
-	man->use_type = false;
-	man->has_type = false;
-
-	ret = 0;
-	if (mem_type > 0) {
-		ret = ttm_bo_force_list_clean(bdev, mem_type);
-		if (ret) {
-			pr_err("Cleanup eviction failed\n");
-			return ret;
-		}
-
-		ret = (*man->func->takedown)(man);
-	}
-
-	dma_fence_put(man->move);
-	man->move = NULL;
-
-	return ret;
-}
-EXPORT_SYMBOL(ttm_bo_clean_mm);
-
 int ttm_bo_evict_mm(struct ttm_bo_device *bdev, unsigned mem_type)
 {
-	struct ttm_mem_type_manager *man = &bdev->man[mem_type];
+	struct ttm_resource_manager *man = ttm_manager_type(bdev, mem_type);
 
 	if (mem_type == 0 || mem_type >= TTM_NUM_MEM_TYPES) {
 		pr_err("Illegal memory manager memory type %u\n", mem_type);
 		return -EINVAL;
 	}
 
-	if (!man->has_type) {
+	if (!man) {
 		pr_err("Memory type %u has not been initialized\n", mem_type);
 		return 0;
 	}
 
-	return ttm_bo_force_list_clean(bdev, mem_type);
+	return ttm_resource_manager_force_list_clean(bdev, man);
 }
 EXPORT_SYMBOL(ttm_bo_evict_mm);
 
-int ttm_bo_init_mm(struct ttm_bo_device *bdev, unsigned type,
-			unsigned long p_size)
-{
-	int ret;
-	struct ttm_mem_type_manager *man;
-	unsigned i;
-
-	BUG_ON(type >= TTM_NUM_MEM_TYPES);
-	man = &bdev->man[type];
-	BUG_ON(man->has_type);
-	man->use_io_reserve_lru = false;
-	mutex_init(&man->io_reserve_mutex);
-	spin_lock_init(&man->move_lock);
-	INIT_LIST_HEAD(&man->io_reserve_lru);
-
-	ret = bdev->driver->init_mem_type(bdev, type, man);
-	if (ret)
-		return ret;
-	man->bdev = bdev;
-
-	if (type != TTM_PL_SYSTEM) {
-		ret = (*man->func->init)(man, p_size);
-		if (ret)
-			return ret;
-	}
-	man->has_type = true;
-	man->use_type = true;
-	man->size = p_size;
-
-	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i)
-		INIT_LIST_HEAD(&man->lru[i]);
-	man->move = NULL;
-
-	return 0;
-}
-EXPORT_SYMBOL(ttm_bo_init_mm);
-
 static void ttm_bo_global_kobj_release(struct kobject *kobj)
 {
 	struct ttm_bo_global *glob =
@@ -1628,21 +1387,12 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev)
 {
 	struct ttm_bo_global *glob = &ttm_bo_glob;
 	int ret = 0;
-	unsigned i = TTM_NUM_MEM_TYPES;
-	struct ttm_mem_type_manager *man;
-
-	while (i--) {
-		man = &bdev->man[i];
-		if (man->has_type) {
-			man->use_type = false;
-			if ((i != TTM_PL_SYSTEM) && ttm_bo_clean_mm(bdev, i)) {
-				ret = -EBUSY;
-				pr_err("DRM memory manager type %d is not clean\n",
-				       i);
-			}
-			man->has_type = false;
-		}
-	}
+	unsigned i;
+	struct ttm_resource_manager *man;
+
+	man = ttm_manager_type(bdev, TTM_PL_SYSTEM);
+	ttm_resource_manager_set_used(man, false);
+	ttm_set_driver_manager(bdev, TTM_PL_SYSTEM, NULL);
 
 	mutex_lock(&ttm_global_mutex);
 	list_del(&bdev->device_list);
@@ -1655,7 +1405,7 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev)
 
 	spin_lock(&glob->lru_lock);
 	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i)
-		if (list_empty(&bdev->man[0].lru[0]))
+		if (list_empty(&man->lru[0]))
 			pr_debug("Swap list %d was clean\n", i);
 	spin_unlock(&glob->lru_lock);
 
@@ -1666,6 +1416,21 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev)
 }
 EXPORT_SYMBOL(ttm_bo_device_release);
 
+static void ttm_bo_init_sysman(struct ttm_bo_device *bdev)
+{
+	struct ttm_resource_manager *man = &bdev->sysman;
+
+	/*
+	 * Initialize the system memory buffer type.
+	 * Other types need to be driver / IOCTL initialized.
+	 */
+	man->use_tt = true;
+
+	ttm_resource_manager_init(man, 0);
+	ttm_set_driver_manager(bdev, TTM_PL_SYSTEM, man);
+	ttm_resource_manager_set_used(man, true);
+}
+
 int ttm_bo_device_init(struct ttm_bo_device *bdev,
 		       struct ttm_bo_driver *driver,
 		       struct address_space *mapping,
@@ -1684,15 +1449,7 @@ int ttm_bo_device_init(struct ttm_bo_device *bdev,
 
 	bdev->driver = driver;
 
-	memset(bdev->man, 0, sizeof(bdev->man));
-
-	/*
-	 * Initialize the system memory buffer type.
-	 * Other types need to be driver / IOCTL initialized.
-	 */
-	ret = ttm_bo_init_mm(bdev, TTM_PL_SYSTEM, 0);
-	if (unlikely(ret != 0))
-		goto out_no_sys;
+	ttm_bo_init_sysman(bdev);
 
 	bdev->vma_manager = vma_manager;
 	INIT_DELAYED_WORK(&bdev->wq, ttm_bo_delayed_workqueue);
@@ -1704,9 +1461,6 @@ int ttm_bo_device_init(struct ttm_bo_device *bdev,
 	mutex_unlock(&ttm_global_mutex);
 
 	return 0;
-out_no_sys:
-	ttm_bo_global_release();
-	return ret;
 }
 EXPORT_SYMBOL(ttm_bo_device_init);
 
@@ -1714,25 +1468,13 @@ EXPORT_SYMBOL(ttm_bo_device_init);
  * buffer object vm functions.
  */
 
-void ttm_bo_unmap_virtual_locked(struct ttm_buffer_object *bo)
-{
-	struct ttm_bo_device *bdev = bo->bdev;
-
-	drm_vma_node_unmap(&bo->base.vma_node, bdev->dev_mapping);
-	ttm_mem_io_free_vm(bo);
-}
-
 void ttm_bo_unmap_virtual(struct ttm_buffer_object *bo)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type];
 
-	ttm_mem_io_lock(man, false);
-	ttm_bo_unmap_virtual_locked(bo);
-	ttm_mem_io_unlock(man);
+	drm_vma_node_unmap(&bo->base.vma_node, bdev->dev_mapping);
+	ttm_mem_io_free(bdev, &bo->mem);
 }
-
-
 EXPORT_SYMBOL(ttm_bo_unmap_virtual);
 
 int ttm_bo_wait(struct ttm_buffer_object *bo,
@@ -1812,11 +1554,11 @@ int ttm_bo_swapout(struct ttm_bo_global *glob, struct ttm_operation_ctx *ctx)
 	if (bo->mem.mem_type != TTM_PL_SYSTEM ||
 	    bo->ttm->caching_state != tt_cached) {
 		struct ttm_operation_ctx ctx = { false, false };
-		struct ttm_mem_reg evict_mem;
+		struct ttm_resource evict_mem;
 
 		evict_mem = bo->mem;
 		evict_mem.mm_node = NULL;
-		evict_mem.placement = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED;
+		evict_mem.placement = TTM_PL_FLAG_CACHED;
 		evict_mem.mem_type = TTM_PL_SYSTEM;
 
 		ret = ttm_bo_handle_move_mem(bo, &evict_mem, true, &ctx);
@@ -1842,7 +1584,7 @@ int ttm_bo_swapout(struct ttm_bo_global *glob, struct ttm_operation_ctx *ctx)
 	if (bo->bdev->driver->swap_notify)
 		bo->bdev->driver->swap_notify(bo);
 
-	ret = ttm_tt_swapout(bo->ttm, bo->persistent_swap_storage);
+	ret = ttm_tt_swapout(bo->bdev, bo->ttm, bo->persistent_swap_storage);
 out:
 
 	/**
@@ -1867,3 +1609,22 @@ void ttm_bo_swapout_all(void)
 	while (ttm_bo_swapout(&ttm_bo_glob, &ctx) == 0);
 }
 EXPORT_SYMBOL(ttm_bo_swapout_all);
+
+void ttm_bo_tt_destroy(struct ttm_buffer_object *bo)
+{
+	if (bo->ttm == NULL)
+		return;
+
+	ttm_tt_destroy(bo->bdev, bo->ttm);
+	bo->ttm = NULL;
+}
+
+int ttm_bo_tt_bind(struct ttm_buffer_object *bo, struct ttm_resource *mem)
+{
+	return bo->bdev->driver->ttm_tt_bind(bo->bdev, bo->ttm, mem);
+}
+
+void ttm_bo_tt_unbind(struct ttm_buffer_object *bo)
+{
+	bo->bdev->driver->ttm_tt_unbind(bo->bdev, bo->ttm);
+}
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index e6c8bd254055..fb2a25f8408f 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -47,15 +47,15 @@ struct ttm_transfer_obj {
 
 void ttm_bo_free_old_node(struct ttm_buffer_object *bo)
 {
-	ttm_bo_mem_put(bo, &bo->mem);
+	ttm_resource_free(bo, &bo->mem);
 }
 
 int ttm_bo_move_ttm(struct ttm_buffer_object *bo,
 		   struct ttm_operation_ctx *ctx,
-		    struct ttm_mem_reg *new_mem)
+		    struct ttm_resource *new_mem)
 {
 	struct ttm_tt *ttm = bo->ttm;
-	struct ttm_mem_reg *old_mem = &bo->mem;
+	struct ttm_resource *old_mem = &bo->mem;
 	int ret;
 
 	if (old_mem->mem_type != TTM_PL_SYSTEM) {
@@ -67,10 +67,8 @@ int ttm_bo_move_ttm(struct ttm_buffer_object *bo,
 			return ret;
 		}
 
-		ttm_tt_unbind(ttm);
+		ttm_bo_tt_unbind(bo);
 		ttm_bo_free_old_node(bo);
-		ttm_flag_masked(&old_mem->placement, TTM_PL_FLAG_SYSTEM,
-				TTM_PL_MASK_MEM);
 		old_mem->mem_type = TTM_PL_SYSTEM;
 	}
 
@@ -79,146 +77,70 @@ int ttm_bo_move_ttm(struct ttm_buffer_object *bo,
 		return ret;
 
 	if (new_mem->mem_type != TTM_PL_SYSTEM) {
-		ret = ttm_tt_bind(ttm, new_mem, ctx);
+
+		ret = ttm_tt_populate(bo->bdev, ttm, ctx);
 		if (unlikely(ret != 0))
 			return ret;
-	}
 
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
+		ret = ttm_bo_tt_bind(bo, new_mem);
+		if (unlikely(ret != 0))
+			return ret;
+	}
 
+	ttm_bo_assign_mem(bo, new_mem);
 	return 0;
 }
 EXPORT_SYMBOL(ttm_bo_move_ttm);
 
-int ttm_mem_io_lock(struct ttm_mem_type_manager *man, bool interruptible)
-{
-	if (likely(!man->use_io_reserve_lru))
-		return 0;
-
-	if (interruptible)
-		return mutex_lock_interruptible(&man->io_reserve_mutex);
-
-	mutex_lock(&man->io_reserve_mutex);
-	return 0;
-}
-
-void ttm_mem_io_unlock(struct ttm_mem_type_manager *man)
-{
-	if (likely(!man->use_io_reserve_lru))
-		return;
-
-	mutex_unlock(&man->io_reserve_mutex);
-}
-
-static int ttm_mem_io_evict(struct ttm_mem_type_manager *man)
-{
-	struct ttm_buffer_object *bo;
-
-	bo = list_first_entry_or_null(&man->io_reserve_lru,
-				      struct ttm_buffer_object,
-				      io_reserve_lru);
-	if (!bo)
-		return -ENOSPC;
-
-	list_del_init(&bo->io_reserve_lru);
-	ttm_bo_unmap_virtual_locked(bo);
-	return 0;
-}
-
 int ttm_mem_io_reserve(struct ttm_bo_device *bdev,
-		       struct ttm_mem_reg *mem)
+		       struct ttm_resource *mem)
 {
-	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
-	int ret;
-
-	if (mem->bus.io_reserved_count++)
+	if (mem->bus.offset || mem->bus.addr)
 		return 0;
 
+	mem->bus.is_iomem = false;
 	if (!bdev->driver->io_mem_reserve)
 		return 0;
 
-retry:
-	ret = bdev->driver->io_mem_reserve(bdev, mem);
-	if (ret == -ENOSPC) {
-		ret = ttm_mem_io_evict(man);
-		if (ret == 0)
-			goto retry;
-	}
-	return ret;
+	return bdev->driver->io_mem_reserve(bdev, mem);
 }
 
 void ttm_mem_io_free(struct ttm_bo_device *bdev,
-		     struct ttm_mem_reg *mem)
+		     struct ttm_resource *mem)
 {
-	if (--mem->bus.io_reserved_count)
+	if (!mem->bus.offset && !mem->bus.addr)
 		return;
 
-	if (!bdev->driver->io_mem_free)
-		return;
+	if (bdev->driver->io_mem_free)
+		bdev->driver->io_mem_free(bdev, mem);
 
-	bdev->driver->io_mem_free(bdev, mem);
+	mem->bus.offset = 0;
+	mem->bus.addr = NULL;
 }
 
-int ttm_mem_io_reserve_vm(struct ttm_buffer_object *bo)
-{
-	struct ttm_mem_type_manager *man = &bo->bdev->man[bo->mem.mem_type];
-	struct ttm_mem_reg *mem = &bo->mem;
-	int ret;
-
-	if (mem->bus.io_reserved_vm)
-		return 0;
-
-	ret = ttm_mem_io_reserve(bo->bdev, mem);
-	if (unlikely(ret != 0))
-		return ret;
-	mem->bus.io_reserved_vm = true;
-	if (man->use_io_reserve_lru)
-		list_add_tail(&bo->io_reserve_lru,
-			      &man->io_reserve_lru);
-	return 0;
-}
-
-void ttm_mem_io_free_vm(struct ttm_buffer_object *bo)
-{
-	struct ttm_mem_reg *mem = &bo->mem;
-
-	if (!mem->bus.io_reserved_vm)
-		return;
-
-	mem->bus.io_reserved_vm = false;
-	list_del_init(&bo->io_reserve_lru);
-	ttm_mem_io_free(bo->bdev, mem);
-}
-
-static int ttm_mem_reg_ioremap(struct ttm_bo_device *bdev,
-			       struct ttm_mem_reg *mem,
+static int ttm_resource_ioremap(struct ttm_bo_device *bdev,
+			       struct ttm_resource *mem,
 			       void **virtual)
 {
-	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
 	int ret;
 	void *addr;
 
 	*virtual = NULL;
-	(void) ttm_mem_io_lock(man, false);
 	ret = ttm_mem_io_reserve(bdev, mem);
-	ttm_mem_io_unlock(man);
 	if (ret || !mem->bus.is_iomem)
 		return ret;
 
 	if (mem->bus.addr) {
 		addr = mem->bus.addr;
 	} else {
+		size_t bus_size = (size_t)mem->num_pages << PAGE_SHIFT;
+
 		if (mem->placement & TTM_PL_FLAG_WC)
-			addr = ioremap_wc(mem->bus.base + mem->bus.offset,
-					  mem->bus.size);
+			addr = ioremap_wc(mem->bus.offset, bus_size);
 		else
-			addr = ioremap(mem->bus.base + mem->bus.offset,
-				       mem->bus.size);
+			addr = ioremap(mem->bus.offset, bus_size);
 		if (!addr) {
-			(void) ttm_mem_io_lock(man, false);
 			ttm_mem_io_free(bdev, mem);
-			ttm_mem_io_unlock(man);
 			return -ENOMEM;
 		}
 	}
@@ -226,19 +148,13 @@ static int ttm_mem_reg_ioremap(struct ttm_bo_device *bdev,
 	return 0;
 }
 
-static void ttm_mem_reg_iounmap(struct ttm_bo_device *bdev,
-				struct ttm_mem_reg *mem,
+static void ttm_resource_iounmap(struct ttm_bo_device *bdev,
+				struct ttm_resource *mem,
 				void *virtual)
 {
-	struct ttm_mem_type_manager *man;
-
-	man = &bdev->man[mem->mem_type];
-
 	if (virtual && mem->bus.addr == NULL)
 		iounmap(virtual);
-	(void) ttm_mem_io_lock(man, false);
 	ttm_mem_io_free(bdev, mem);
-	ttm_mem_io_unlock(man);
 }
 
 static int ttm_copy_io_page(void *dst, void *src, unsigned long page)
@@ -300,13 +216,13 @@ static int ttm_copy_ttm_io_page(struct ttm_tt *ttm, void *dst,
 
 int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 		       struct ttm_operation_ctx *ctx,
-		       struct ttm_mem_reg *new_mem)
+		       struct ttm_resource *new_mem)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man = &bdev->man[new_mem->mem_type];
+	struct ttm_resource_manager *man = ttm_manager_type(bdev, new_mem->mem_type);
 	struct ttm_tt *ttm = bo->ttm;
-	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg old_copy = *old_mem;
+	struct ttm_resource *old_mem = &bo->mem;
+	struct ttm_resource old_copy = *old_mem;
 	void *old_iomap;
 	void *new_iomap;
 	int ret;
@@ -319,10 +235,10 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	if (ret)
 		return ret;
 
-	ret = ttm_mem_reg_ioremap(bdev, old_mem, &old_iomap);
+	ret = ttm_resource_ioremap(bdev, old_mem, &old_iomap);
 	if (ret)
 		return ret;
-	ret = ttm_mem_reg_ioremap(bdev, new_mem, &new_iomap);
+	ret = ttm_resource_ioremap(bdev, new_mem, &new_iomap);
 	if (ret)
 		goto out;
 
@@ -336,7 +252,7 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	 * Don't move nonexistent data. Clear destination instead.
 	 */
 	if (old_iomap == NULL &&
-	    (ttm == NULL || (ttm->state == tt_unpopulated &&
+	    (ttm == NULL || (!ttm_tt_is_populated(ttm) &&
 			     !(ttm->page_flags & TTM_PAGE_FLAG_SWAPPED)))) {
 		memset_io(new_iomap, 0, new_mem->num_pages*PAGE_SIZE);
 		goto out2;
@@ -346,7 +262,7 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	 * TTM might be null for moves within the same region.
 	 */
 	if (ttm) {
-		ret = ttm_tt_populate(ttm, ctx);
+		ret = ttm_tt_populate(bdev, ttm, ctx);
 		if (ret)
 			goto out1;
 	}
@@ -381,24 +297,22 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	mb();
 out2:
 	old_copy = *old_mem;
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
 
-	if (man->flags & TTM_MEMTYPE_FLAG_FIXED) {
-		ttm_tt_destroy(ttm);
-		bo->ttm = NULL;
-	}
+	ttm_bo_assign_mem(bo, new_mem);
+
+	if (!man->use_tt)
+		ttm_bo_tt_destroy(bo);
 
 out1:
-	ttm_mem_reg_iounmap(bdev, old_mem, new_iomap);
+	ttm_resource_iounmap(bdev, old_mem, new_iomap);
 out:
-	ttm_mem_reg_iounmap(bdev, &old_copy, old_iomap);
+	ttm_resource_iounmap(bdev, &old_copy, old_iomap);
 
 	/*
 	 * On error, keep the mm node!
 	 */
 	if (!ret)
-		ttm_bo_mem_put(bo, &old_copy);
+		ttm_resource_free(bo, &old_copy);
 	return ret;
 }
 EXPORT_SYMBOL(ttm_bo_move_memcpy);
@@ -452,7 +366,6 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
 	INIT_LIST_HEAD(&fbo->base.ddestroy);
 	INIT_LIST_HEAD(&fbo->base.lru);
 	INIT_LIST_HEAD(&fbo->base.swap);
-	INIT_LIST_HEAD(&fbo->base.io_reserve_lru);
 	fbo->base.moving = NULL;
 	drm_vma_node_reset(&fbo->base.base.vma_node);
 
@@ -502,7 +415,7 @@ static int ttm_bo_ioremap(struct ttm_buffer_object *bo,
 			  unsigned long size,
 			  struct ttm_bo_kmap_obj *map)
 {
-	struct ttm_mem_reg *mem = &bo->mem;
+	struct ttm_resource *mem = &bo->mem;
 
 	if (bo->mem.bus.addr) {
 		map->bo_kmap_type = ttm_bo_map_premapped;
@@ -510,12 +423,10 @@ static int ttm_bo_ioremap(struct ttm_buffer_object *bo,
 	} else {
 		map->bo_kmap_type = ttm_bo_map_iomap;
 		if (mem->placement & TTM_PL_FLAG_WC)
-			map->virtual = ioremap_wc(bo->mem.bus.base +
-						  bo->mem.bus.offset + offset,
+			map->virtual = ioremap_wc(bo->mem.bus.offset + offset,
 						  size);
 		else
-			map->virtual = ioremap(bo->mem.bus.base +
-					       bo->mem.bus.offset + offset,
+			map->virtual = ioremap(bo->mem.bus.offset + offset,
 					       size);
 	}
 	return (!map->virtual) ? -ENOMEM : 0;
@@ -526,7 +437,7 @@ static int ttm_bo_kmap_ttm(struct ttm_buffer_object *bo,
 			   unsigned long num_pages,
 			   struct ttm_bo_kmap_obj *map)
 {
-	struct ttm_mem_reg *mem = &bo->mem;
+	struct ttm_resource *mem = &bo->mem;
 	struct ttm_operation_ctx ctx = {
 		.interruptible = false,
 		.no_wait_gpu = false
@@ -537,7 +448,7 @@ static int ttm_bo_kmap_ttm(struct ttm_buffer_object *bo,
 
 	BUG_ON(!ttm);
 
-	ret = ttm_tt_populate(ttm, &ctx);
+	ret = ttm_tt_populate(bo->bdev, ttm, &ctx);
 	if (ret)
 		return ret;
 
@@ -567,8 +478,6 @@ int ttm_bo_kmap(struct ttm_buffer_object *bo,
 		unsigned long start_page, unsigned long num_pages,
 		struct ttm_bo_kmap_obj *map)
 {
-	struct ttm_mem_type_manager *man =
-		&bo->bdev->man[bo->mem.mem_type];
 	unsigned long offset, size;
 	int ret;
 
@@ -579,9 +488,7 @@ int ttm_bo_kmap(struct ttm_buffer_object *bo,
 	if (start_page > bo->num_pages)
 		return -EINVAL;
 
-	(void) ttm_mem_io_lock(man, false);
 	ret = ttm_mem_io_reserve(bo->bdev, &bo->mem);
-	ttm_mem_io_unlock(man);
 	if (ret)
 		return ret;
 	if (!bo->mem.bus.is_iomem) {
@@ -596,10 +503,6 @@ EXPORT_SYMBOL(ttm_bo_kmap);
 
 void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map)
 {
-	struct ttm_buffer_object *bo = map->bo;
-	struct ttm_mem_type_manager *man =
-		&bo->bdev->man[bo->mem.mem_type];
-
 	if (!map->virtual)
 		return;
 	switch (map->bo_kmap_type) {
@@ -617,167 +520,116 @@ void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map)
 	default:
 		BUG();
 	}
-	(void) ttm_mem_io_lock(man, false);
 	ttm_mem_io_free(map->bo->bdev, &map->bo->mem);
-	ttm_mem_io_unlock(man);
 	map->virtual = NULL;
 	map->page = NULL;
 }
 EXPORT_SYMBOL(ttm_bo_kunmap);
 
-int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
-			      struct dma_fence *fence,
-			      bool evict,
-			      struct ttm_mem_reg *new_mem)
+static int ttm_bo_wait_free_node(struct ttm_buffer_object *bo,
+				 bool dst_use_tt)
 {
-	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_type_manager *man = &bdev->man[new_mem->mem_type];
-	struct ttm_mem_reg *old_mem = &bo->mem;
 	int ret;
-	struct ttm_buffer_object *ghost_obj;
-
-	dma_resv_add_excl_fence(bo->base.resv, fence);
-	if (evict) {
-		ret = ttm_bo_wait(bo, false, false);
-		if (ret)
-			return ret;
-
-		if (man->flags & TTM_MEMTYPE_FLAG_FIXED) {
-			ttm_tt_destroy(bo->ttm);
-			bo->ttm = NULL;
-		}
-		ttm_bo_free_old_node(bo);
-	} else {
-		/**
-		 * This should help pipeline ordinary buffer moves.
-		 *
-		 * Hang old buffer memory on a new buffer object,
-		 * and leave it to be released when the GPU
-		 * operation has completed.
-		 */
-
-		dma_fence_put(bo->moving);
-		bo->moving = dma_fence_get(fence);
-
-		ret = ttm_buffer_object_transfer(bo, &ghost_obj);
-		if (ret)
-			return ret;
-
-		dma_resv_add_excl_fence(&ghost_obj->base._resv, fence);
-
-		/**
-		 * If we're not moving to fixed memory, the TTM object
-		 * needs to stay alive. Otherwhise hang it on the ghost
-		 * bo to be unbound and destroyed.
-		 */
-
-		if (!(man->flags & TTM_MEMTYPE_FLAG_FIXED))
-			ghost_obj->ttm = NULL;
-		else
-			bo->ttm = NULL;
-
-		dma_resv_unlock(&ghost_obj->base._resv);
-		ttm_bo_put(ghost_obj);
-	}
-
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
+	ret = ttm_bo_wait(bo, false, false);
+	if (ret)
+		return ret;
 
+	if (!dst_use_tt)
+		ttm_bo_tt_destroy(bo);
+	ttm_bo_free_old_node(bo);
 	return 0;
 }
-EXPORT_SYMBOL(ttm_bo_move_accel_cleanup);
 
-int ttm_bo_pipeline_move(struct ttm_buffer_object *bo,
-			 struct dma_fence *fence, bool evict,
-			 struct ttm_mem_reg *new_mem)
+static int ttm_bo_move_to_ghost(struct ttm_buffer_object *bo,
+				struct dma_fence *fence,
+				bool dst_use_tt)
 {
-	struct ttm_bo_device *bdev = bo->bdev;
-	struct ttm_mem_reg *old_mem = &bo->mem;
-
-	struct ttm_mem_type_manager *from = &bdev->man[old_mem->mem_type];
-	struct ttm_mem_type_manager *to = &bdev->man[new_mem->mem_type];
-
+	struct ttm_buffer_object *ghost_obj;
 	int ret;
 
-	dma_resv_add_excl_fence(bo->base.resv, fence);
-
-	if (!evict) {
-		struct ttm_buffer_object *ghost_obj;
-
-		/**
-		 * This should help pipeline ordinary buffer moves.
-		 *
-		 * Hang old buffer memory on a new buffer object,
-		 * and leave it to be released when the GPU
-		 * operation has completed.
-		 */
-
-		dma_fence_put(bo->moving);
-		bo->moving = dma_fence_get(fence);
+	/**
+	 * This should help pipeline ordinary buffer moves.
+	 *
+	 * Hang old buffer memory on a new buffer object,
+	 * and leave it to be released when the GPU
+	 * operation has completed.
+	 */
 
-		ret = ttm_buffer_object_transfer(bo, &ghost_obj);
-		if (ret)
-			return ret;
+	dma_fence_put(bo->moving);
+	bo->moving = dma_fence_get(fence);
 
-		dma_resv_add_excl_fence(&ghost_obj->base._resv, fence);
+	ret = ttm_buffer_object_transfer(bo, &ghost_obj);
+	if (ret)
+		return ret;
 
-		/**
-		 * If we're not moving to fixed memory, the TTM object
-		 * needs to stay alive. Otherwhise hang it on the ghost
-		 * bo to be unbound and destroyed.
-		 */
+	dma_resv_add_excl_fence(&ghost_obj->base._resv, fence);
 
-		if (!(to->flags & TTM_MEMTYPE_FLAG_FIXED))
-			ghost_obj->ttm = NULL;
-		else
-			bo->ttm = NULL;
+	/**
+	 * If we're not moving to fixed memory, the TTM object
+	 * needs to stay alive. Otherwhise hang it on the ghost
+	 * bo to be unbound and destroyed.
+	 */
 
-		dma_resv_unlock(&ghost_obj->base._resv);
-		ttm_bo_put(ghost_obj);
+	if (dst_use_tt)
+		ghost_obj->ttm = NULL;
+	else
+		bo->ttm = NULL;
 
-	} else if (from->flags & TTM_MEMTYPE_FLAG_FIXED) {
+	dma_resv_unlock(&ghost_obj->base._resv);
+	ttm_bo_put(ghost_obj);
+	return 0;
+}
 
-		/**
-		 * BO doesn't have a TTM we need to bind/unbind. Just remember
-		 * this eviction and free up the allocation
-		 */
+static void ttm_bo_move_pipeline_evict(struct ttm_buffer_object *bo,
+				       struct dma_fence *fence)
+{
+	struct ttm_bo_device *bdev = bo->bdev;
+	struct ttm_resource_manager *from = ttm_manager_type(bdev, bo->mem.mem_type);
 
-		spin_lock(&from->move_lock);
-		if (!from->move || dma_fence_is_later(fence, from->move)) {
-			dma_fence_put(from->move);
-			from->move = dma_fence_get(fence);
-		}
-		spin_unlock(&from->move_lock);
+	/**
+	 * BO doesn't have a TTM we need to bind/unbind. Just remember
+	 * this eviction and free up the allocation
+	 */
+	spin_lock(&from->move_lock);
+	if (!from->move || dma_fence_is_later(fence, from->move)) {
+		dma_fence_put(from->move);
+		from->move = dma_fence_get(fence);
+	}
+	spin_unlock(&from->move_lock);
 
-		ttm_bo_free_old_node(bo);
+	ttm_bo_free_old_node(bo);
 
-		dma_fence_put(bo->moving);
-		bo->moving = dma_fence_get(fence);
+	dma_fence_put(bo->moving);
+	bo->moving = dma_fence_get(fence);
+}
 
-	} else {
-		/**
-		 * Last resort, wait for the move to be completed.
-		 *
-		 * Should never happen in pratice.
-		 */
+int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
+			      struct dma_fence *fence,
+			      bool evict,
+			      bool pipeline,
+			      struct ttm_resource *new_mem)
+{
+	struct ttm_bo_device *bdev = bo->bdev;
+	struct ttm_resource_manager *from = ttm_manager_type(bdev, bo->mem.mem_type);
+	struct ttm_resource_manager *man = ttm_manager_type(bdev, new_mem->mem_type);
+	int ret = 0;
 
-		ret = ttm_bo_wait(bo, false, false);
-		if (ret)
-			return ret;
+	dma_resv_add_excl_fence(bo->base.resv, fence);
+	if (!evict)
+		ret = ttm_bo_move_to_ghost(bo, fence, man->use_tt);
+	else if (!from->use_tt && pipeline)
+		ttm_bo_move_pipeline_evict(bo, fence);
+	else
+		ret = ttm_bo_wait_free_node(bo, man->use_tt);
 
-		if (to->flags & TTM_MEMTYPE_FLAG_FIXED) {
-			ttm_tt_destroy(bo->ttm);
-			bo->ttm = NULL;
-		}
-		ttm_bo_free_old_node(bo);
-	}
+	if (ret)
+		return ret;
 
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
+	ttm_bo_assign_mem(bo, new_mem);
 
 	return 0;
 }
-EXPORT_SYMBOL(ttm_bo_pipeline_move);
+EXPORT_SYMBOL(ttm_bo_move_accel_cleanup);
 
 int ttm_bo_pipeline_gutting(struct ttm_buffer_object *bo)
 {
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index 4732dcc80e11..98a006fc30a5 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -101,8 +101,7 @@ static unsigned long ttm_bo_io_mem_pfn(struct ttm_buffer_object *bo,
 	if (bdev->driver->io_mem_pfn)
 		return bdev->driver->io_mem_pfn(bo, page_offset);
 
-	return ((bo->mem.bus.base + bo->mem.bus.offset) >> PAGE_SHIFT)
-		+ page_offset;
+	return (bo->mem.bus.offset >> PAGE_SHIFT) + page_offset;
 }
 
 /**
@@ -281,8 +280,6 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 	pgoff_t i;
 	vm_fault_t ret = VM_FAULT_NOPAGE;
 	unsigned long address = vmf->address;
-	struct ttm_mem_type_manager *man =
-		&bdev->man[bo->mem.mem_type];
 
 	/*
 	 * Refuse to fault imported pages. This should be handled
@@ -308,9 +305,7 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		}
 
 		if (bo->moving != moving) {
-			spin_lock(&ttm_bo_glob.lru_lock);
-			ttm_bo_move_to_lru_tail(bo, NULL);
-			spin_unlock(&ttm_bo_glob.lru_lock);
+			ttm_bo_move_to_lru_tail_unlocked(bo);
 		}
 		dma_fence_put(moving);
 	}
@@ -323,24 +318,17 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 	if (unlikely(ret != 0))
 		return ret;
 
-	err = ttm_mem_io_lock(man, true);
+	err = ttm_mem_io_reserve(bdev, &bo->mem);
 	if (unlikely(err != 0))
-		return VM_FAULT_NOPAGE;
-	err = ttm_mem_io_reserve_vm(bo);
-	if (unlikely(err != 0)) {
-		ret = VM_FAULT_SIGBUS;
-		goto out_io_unlock;
-	}
+		return VM_FAULT_SIGBUS;
 
 	page_offset = ((address - vma->vm_start) >> PAGE_SHIFT) +
 		vma->vm_pgoff - drm_vma_node_start(&bo->base.vma_node);
 	page_last = vma_pages(vma) + vma->vm_pgoff -
 		drm_vma_node_start(&bo->base.vma_node);
 
-	if (unlikely(page_offset >= bo->num_pages)) {
-		ret = VM_FAULT_SIGBUS;
-		goto out_io_unlock;
-	}
+	if (unlikely(page_offset >= bo->num_pages))
+		return VM_FAULT_SIGBUS;
 
 	prot = ttm_io_prot(bo->mem.placement, prot);
 	if (!bo->mem.bus.is_iomem) {
@@ -352,21 +340,17 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		};
 
 		ttm = bo->ttm;
-		if (ttm_tt_populate(bo->ttm, &ctx)) {
-			ret = VM_FAULT_OOM;
-			goto out_io_unlock;
-		}
+		if (ttm_tt_populate(bdev, bo->ttm, &ctx))
+			return VM_FAULT_OOM;
 	} else {
 		/* Iomem should not be marked encrypted */
 		prot = pgprot_decrypted(prot);
 	}
 
 	/* We don't prefault on huge faults. Yet. */
-	if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && fault_page_size != 1) {
-		ret = ttm_bo_vm_insert_huge(vmf, bo, page_offset,
-					    fault_page_size, prot);
-		goto out_io_unlock;
-	}
+	if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && fault_page_size != 1)
+		return ttm_bo_vm_insert_huge(vmf, bo, page_offset,
+					     fault_page_size, prot);
 
 	/*
 	 * Speculatively prefault a number of pages. Only error on
@@ -378,8 +362,7 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		} else {
 			page = ttm->pages[page_offset];
 			if (unlikely(!page && i == 0)) {
-				ret = VM_FAULT_OOM;
-				goto out_io_unlock;
+				return VM_FAULT_OOM;
 			} else if (unlikely(!page)) {
 				break;
 			}
@@ -406,7 +389,7 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		/* Never error on prefaulted PTEs */
 		if (unlikely((ret & VM_FAULT_ERROR))) {
 			if (i == 0)
-				goto out_io_unlock;
+				return VM_FAULT_NOPAGE;
 			else
 				break;
 		}
@@ -415,9 +398,6 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		if (unlikely(++page_offset >= page_last))
 			break;
 	}
-	ret = VM_FAULT_NOPAGE;
-out_io_unlock:
-	ttm_mem_io_unlock(man);
 	return ret;
 }
 EXPORT_SYMBOL(ttm_bo_vm_fault_reserved);
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index 1797f04c0534..8a8f1a6a83a6 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -93,7 +93,7 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket,
 	list_for_each_entry(entry, list, head) {
 		struct ttm_buffer_object *bo = entry->bo;
 
-		ret = __ttm_bo_reserve(bo, intr, (ticket == NULL), ticket);
+		ret = ttm_bo_reserve(bo, intr, (ticket == NULL), ticket);
 		if (ret == -EALREADY && dups) {
 			struct ttm_validate_buffer *safe = entry;
 			entry = list_prev_entry(entry, head);
@@ -119,13 +119,7 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket,
 		ttm_eu_backoff_reservation_reverse(list, entry);
 
 		if (ret == -EDEADLK) {
-			if (intr) {
-				ret = dma_resv_lock_slow_interruptible(bo->base.resv,
-										 ticket);
-			} else {
-				dma_resv_lock_slow(bo->base.resv, ticket);
-				ret = 0;
-			}
+			ret = ttm_bo_reserve_slowpath(bo, intr, ticket);
 		}
 
 		if (!ret && entry->num_shared)
@@ -133,8 +127,6 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket,
 								entry->num_shared);
 
 		if (unlikely(ret != 0)) {
-			if (ret == -EINTR)
-				ret = -ERESTARTSYS;
 			if (ticket) {
 				ww_acquire_done(ticket);
 				ww_acquire_fini(ticket);
diff --git a/drivers/gpu/drm/ttm/ttm_memory.c b/drivers/gpu/drm/ttm/ttm_memory.c
index acd63b70d814..89d50f38c0f2 100644
--- a/drivers/gpu/drm/ttm/ttm_memory.c
+++ b/drivers/gpu/drm/ttm/ttm_memory.c
@@ -259,7 +259,7 @@ static bool ttm_zones_above_swap_target(struct ttm_mem_global *glob,
 	return false;
 }
 
-/**
+/*
  * At this point we only support a single shrink callback.
  * Extend this if needed, perhaps using a linked list of callbacks.
  * Note that this function is reentrant:
@@ -554,7 +554,6 @@ ttm_check_under_lowerlimit(struct ttm_mem_global *glob,
 
 	return false;
 }
-EXPORT_SYMBOL(ttm_check_under_lowerlimit);
 
 static int ttm_mem_global_reserve(struct ttm_mem_global *glob,
 				  struct ttm_mem_zone *single_zone,
@@ -682,9 +681,3 @@ size_t ttm_round_pot(size_t size)
 	return 0;
 }
 EXPORT_SYMBOL(ttm_round_pot);
-
-uint64_t ttm_get_kernel_zone_memory_size(struct ttm_mem_global *glob)
-{
-	return glob->zone_kernel->max_mem;
-}
-EXPORT_SYMBOL(ttm_get_kernel_zone_memory_size);
diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c
index b40a4678c296..14660f723f71 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c
@@ -1044,7 +1044,7 @@ ttm_pool_unpopulate_helper(struct ttm_tt *ttm, unsigned mem_count_update)
 put_pages:
 	ttm_put_pages(ttm->pages, ttm->num_pages, ttm->page_flags,
 		      ttm->caching_state);
-	ttm->state = tt_unpopulated;
+	ttm_tt_set_unpopulated(ttm);
 }
 
 int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
@@ -1053,7 +1053,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 	unsigned i;
 	int ret;
 
-	if (ttm->state != tt_unpopulated)
+	if (ttm_tt_is_populated(ttm))
 		return 0;
 
 	if (ttm_check_under_lowerlimit(mem_glob, ttm->num_pages, ctx))
@@ -1083,7 +1083,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 		}
 	}
 
-	ttm->state = tt_unbound;
+	ttm_tt_set_populated(ttm);
 	return 0;
 }
 EXPORT_SYMBOL(ttm_pool_populate);
diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
index faefaaef7909..5e2df11685e7 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
@@ -894,7 +894,7 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev,
 	unsigned i;
 	int ret;
 
-	if (ttm->state != tt_unpopulated)
+	if (ttm_tt_is_populated(ttm))
 		return 0;
 
 	if (ttm_check_under_lowerlimit(mem_glob, num_pages, ctx))
@@ -982,7 +982,7 @@ skip_huge:
 		}
 	}
 
-	ttm->state = tt_unbound;
+	ttm_tt_set_populated(ttm);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(ttm_dma_populate);
@@ -1076,7 +1076,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev)
 	/* shrink pool if necessary (only on !is_cached pools)*/
 	if (npages)
 		ttm_dma_page_pool_free(pool, npages, false);
-	ttm->state = tt_unpopulated;
+	ttm_tt_set_unpopulated(ttm);
 }
 EXPORT_SYMBOL_GPL(ttm_dma_unpopulate);
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo_manager.c b/drivers/gpu/drm/ttm/ttm_range_manager.c
index facd3049c3aa..1da0e277c511 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_manager.c
+++ b/drivers/gpu/drm/ttm/ttm_range_manager.c
@@ -44,16 +44,22 @@
  */
 
 struct ttm_range_manager {
+	struct ttm_resource_manager manager;
 	struct drm_mm mm;
 	spinlock_t lock;
 };
 
-static int ttm_bo_man_get_node(struct ttm_mem_type_manager *man,
+static inline struct ttm_range_manager *to_range_manager(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct ttm_range_manager, manager);
+}
+
+static int ttm_range_man_alloc(struct ttm_resource_manager *man,
 			       struct ttm_buffer_object *bo,
 			       const struct ttm_place *place,
-			       struct ttm_mem_reg *mem)
+			       struct ttm_resource *mem)
 {
-	struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+	struct ttm_range_manager *rman = to_range_manager(man);
 	struct drm_mm *mm = &rman->mm;
 	struct drm_mm_node *node;
 	enum drm_mm_insert_mode mode;
@@ -89,10 +95,10 @@ static int ttm_bo_man_get_node(struct ttm_mem_type_manager *man,
 	return ret;
 }
 
-static void ttm_bo_man_put_node(struct ttm_mem_type_manager *man,
-				struct ttm_mem_reg *mem)
+static void ttm_range_man_free(struct ttm_resource_manager *man,
+			       struct ttm_resource *mem)
 {
-	struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+	struct ttm_range_manager *rman = to_range_manager(man);
 
 	if (mem->mm_node) {
 		spin_lock(&rman->lock);
@@ -104,53 +110,73 @@ static void ttm_bo_man_put_node(struct ttm_mem_type_manager *man,
 	}
 }
 
-static int ttm_bo_man_init(struct ttm_mem_type_manager *man,
-			   unsigned long p_size)
+static const struct ttm_resource_manager_func ttm_range_manager_func;
+
+int ttm_range_man_init(struct ttm_bo_device *bdev,
+		       unsigned type, bool use_tt,
+		       unsigned long p_size)
 {
+	struct ttm_resource_manager *man;
 	struct ttm_range_manager *rman;
 
 	rman = kzalloc(sizeof(*rman), GFP_KERNEL);
 	if (!rman)
 		return -ENOMEM;
 
+	man = &rman->manager;
+	man->use_tt = use_tt;
+
+	man->func = &ttm_range_manager_func;
+
+	ttm_resource_manager_init(man, p_size);
+
 	drm_mm_init(&rman->mm, 0, p_size);
 	spin_lock_init(&rman->lock);
-	man->priv = rman;
+
+	ttm_set_driver_manager(bdev, type, &rman->manager);
+	ttm_resource_manager_set_used(man, true);
 	return 0;
 }
+EXPORT_SYMBOL(ttm_range_man_init);
 
-static int ttm_bo_man_takedown(struct ttm_mem_type_manager *man)
+int ttm_range_man_fini(struct ttm_bo_device *bdev,
+		       unsigned type)
 {
-	struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+	struct ttm_resource_manager *man = ttm_manager_type(bdev, type);
+	struct ttm_range_manager *rman = to_range_manager(man);
 	struct drm_mm *mm = &rman->mm;
+	int ret;
+
+	ttm_resource_manager_set_used(man, false);
+
+	ret = ttm_resource_manager_force_list_clean(bdev, man);
+	if (ret)
+		return ret;
 
 	spin_lock(&rman->lock);
-	if (drm_mm_clean(mm)) {
-		drm_mm_takedown(mm);
-		spin_unlock(&rman->lock);
-		kfree(rman);
-		man->priv = NULL;
-		return 0;
-	}
+	drm_mm_clean(mm);
+	drm_mm_takedown(mm);
 	spin_unlock(&rman->lock);
-	return -EBUSY;
+
+	ttm_resource_manager_cleanup(man);
+	ttm_set_driver_manager(bdev, type, NULL);
+	kfree(rman);
+	return 0;
 }
+EXPORT_SYMBOL(ttm_range_man_fini);
 
-static void ttm_bo_man_debug(struct ttm_mem_type_manager *man,
-			     struct drm_printer *printer)
+static void ttm_range_man_debug(struct ttm_resource_manager *man,
+				struct drm_printer *printer)
 {
-	struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+	struct ttm_range_manager *rman = to_range_manager(man);
 
 	spin_lock(&rman->lock);
 	drm_mm_print(&rman->mm, printer);
 	spin_unlock(&rman->lock);
 }
 
-const struct ttm_mem_type_manager_func ttm_bo_manager_func = {
-	.init = ttm_bo_man_init,
-	.takedown = ttm_bo_man_takedown,
-	.get_node = ttm_bo_man_get_node,
-	.put_node = ttm_bo_man_put_node,
-	.debug = ttm_bo_man_debug
+static const struct ttm_resource_manager_func ttm_range_manager_func = {
+	.alloc = ttm_range_man_alloc,
+	.free = ttm_range_man_free,
+	.debug = ttm_range_man_debug
 };
-EXPORT_SYMBOL(ttm_bo_manager_func);
diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c
new file mode 100644
index 000000000000..b325b9264203
--- /dev/null
+++ b/drivers/gpu/drm/ttm/ttm_resource.c
@@ -0,0 +1,146 @@
+/*
+ * Copyright 2020 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"),
+ * 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: Christian König
+ */
+
+#include <drm/ttm/ttm_resource.h>
+#include <drm/ttm/ttm_bo_driver.h>
+
+int ttm_resource_alloc(struct ttm_buffer_object *bo,
+		       const struct ttm_place *place,
+		       struct ttm_resource *res)
+{
+	struct ttm_resource_manager *man =
+		ttm_manager_type(bo->bdev, res->mem_type);
+
+	res->mm_node = NULL;
+	if (!man->func || !man->func->alloc)
+		return 0;
+
+	return man->func->alloc(man, bo, place, res);
+}
+
+void ttm_resource_free(struct ttm_buffer_object *bo, struct ttm_resource *res)
+{
+	struct ttm_resource_manager *man =
+		ttm_manager_type(bo->bdev, res->mem_type);
+
+	if (man->func && man->func->free)
+		man->func->free(man, res);
+
+	res->mm_node = NULL;
+	res->mem_type = TTM_PL_SYSTEM;
+}
+EXPORT_SYMBOL(ttm_resource_free);
+
+/**
+ * ttm_resource_manager_init
+ *
+ * @man: memory manager object to init
+ * @p_size: size managed area in pages.
+ *
+ * Initialise core parts of a manager object.
+ */
+void ttm_resource_manager_init(struct ttm_resource_manager *man,
+			       unsigned long p_size)
+{
+	unsigned i;
+
+	spin_lock_init(&man->move_lock);
+	man->size = p_size;
+
+	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i)
+		INIT_LIST_HEAD(&man->lru[i]);
+	man->move = NULL;
+}
+EXPORT_SYMBOL(ttm_resource_manager_init);
+
+/*
+ * ttm_resource_manager_force_list_clean
+ *
+ * @bdev - device to use
+ * @man - manager to use
+ *
+ * Force all the objects out of a memory manager until clean.
+ * Part of memory manager cleanup sequence.
+ */
+int ttm_resource_manager_force_list_clean(struct ttm_bo_device *bdev,
+					  struct ttm_resource_manager *man)
+{
+	struct ttm_operation_ctx ctx = {
+		.interruptible = false,
+		.no_wait_gpu = false,
+		.flags = TTM_OPT_FLAG_FORCE_ALLOC
+	};
+	struct ttm_bo_global *glob = &ttm_bo_glob;
+	struct dma_fence *fence;
+	int ret;
+	unsigned i;
+
+	/*
+	 * Can't use standard list traversal since we're unlocking.
+	 */
+
+	spin_lock(&glob->lru_lock);
+	for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) {
+		while (!list_empty(&man->lru[i])) {
+			spin_unlock(&glob->lru_lock);
+			ret = ttm_mem_evict_first(bdev, man, NULL, &ctx,
+						  NULL);
+			if (ret)
+				return ret;
+			spin_lock(&glob->lru_lock);
+		}
+	}
+	spin_unlock(&glob->lru_lock);
+
+	spin_lock(&man->move_lock);
+	fence = dma_fence_get(man->move);
+	spin_unlock(&man->move_lock);
+
+	if (fence) {
+		ret = dma_fence_wait(fence, false);
+		dma_fence_put(fence);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(ttm_resource_manager_force_list_clean);
+
+/**
+ * ttm_resource_manager_debug
+ *
+ * @man: manager type to dump.
+ * @p: printer to use for debug.
+ */
+void ttm_resource_manager_debug(struct ttm_resource_manager *man,
+				struct drm_printer *p)
+{
+	drm_printf(p, "  use_type: %d\n", man->use_type);
+	drm_printf(p, "  use_tt: %d\n", man->use_tt);
+	drm_printf(p, "  size: %llu\n", man->size);
+	if (man->func && man->func->debug)
+		(*man->func->debug)(man, p);
+}
+EXPORT_SYMBOL(ttm_resource_manager_debug);
diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index 3437711ddb43..f43fa69a1e65 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -50,6 +50,9 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc)
 
 	dma_resv_assert_held(bo->base.resv);
 
+	if (bo->ttm)
+		return 0;
+
 	if (bdev->need_dma32)
 		page_flags |= TTM_PAGE_FLAG_DMA32;
 
@@ -67,7 +70,6 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc)
 		page_flags |= TTM_PAGE_FLAG_SG;
 		break;
 	default:
-		bo->ttm = NULL;
 		pr_err("Illegal buffer object type\n");
 		return -EINVAL;
 	}
@@ -154,7 +156,7 @@ static int ttm_tt_set_caching(struct ttm_tt *ttm,
 	if (ttm->caching_state == c_state)
 		return 0;
 
-	if (ttm->state == tt_unpopulated) {
+	if (!ttm_tt_is_populated(ttm)) {
 		/* Change caching but don't populate */
 		ttm->caching_state = c_state;
 		return 0;
@@ -205,33 +207,31 @@ int ttm_tt_set_placement_caching(struct ttm_tt *ttm, uint32_t placement)
 }
 EXPORT_SYMBOL(ttm_tt_set_placement_caching);
 
-void ttm_tt_destroy(struct ttm_tt *ttm)
+void ttm_tt_destroy_common(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
-	if (ttm == NULL)
-		return;
-
-	ttm_tt_unbind(ttm);
-
-	if (ttm->state == tt_unbound)
-		ttm_tt_unpopulate(ttm);
+	ttm_tt_unpopulate(bdev, ttm);
 
 	if (!(ttm->page_flags & TTM_PAGE_FLAG_PERSISTENT_SWAP) &&
 	    ttm->swap_storage)
 		fput(ttm->swap_storage);
 
 	ttm->swap_storage = NULL;
-	ttm->func->destroy(ttm);
+}
+EXPORT_SYMBOL(ttm_tt_destroy_common);
+
+void ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
+{
+	bdev->driver->ttm_tt_destroy(bdev, ttm);
 }
 
 static void ttm_tt_init_fields(struct ttm_tt *ttm,
 			       struct ttm_buffer_object *bo,
 			       uint32_t page_flags)
 {
-	ttm->bdev = bo->bdev;
 	ttm->num_pages = bo->num_pages;
 	ttm->caching_state = tt_cached;
 	ttm->page_flags = page_flags;
-	ttm->state = tt_unpopulated;
+	ttm_tt_set_unpopulated(ttm);
 	ttm->swap_storage = NULL;
 	ttm->sg = bo->sg;
 }
@@ -306,39 +306,6 @@ void ttm_dma_tt_fini(struct ttm_dma_tt *ttm_dma)
 }
 EXPORT_SYMBOL(ttm_dma_tt_fini);
 
-void ttm_tt_unbind(struct ttm_tt *ttm)
-{
-	if (ttm->state == tt_bound) {
-		ttm->func->unbind(ttm);
-		ttm->state = tt_unbound;
-	}
-}
-
-int ttm_tt_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem,
-		struct ttm_operation_ctx *ctx)
-{
-	int ret = 0;
-
-	if (!ttm)
-		return -EINVAL;
-
-	if (ttm->state == tt_bound)
-		return 0;
-
-	ret = ttm_tt_populate(ttm, ctx);
-	if (ret)
-		return ret;
-
-	ret = ttm->func->bind(ttm, bo_mem);
-	if (unlikely(ret != 0))
-		return ret;
-
-	ttm->state = tt_bound;
-
-	return 0;
-}
-EXPORT_SYMBOL(ttm_tt_bind);
-
 int ttm_tt_swapin(struct ttm_tt *ttm)
 {
 	struct address_space *swap_space;
@@ -381,7 +348,8 @@ out_err:
 	return ret;
 }
 
-int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage)
+int ttm_tt_swapout(struct ttm_bo_device *bdev,
+		   struct ttm_tt *ttm, struct file *persistent_swap_storage)
 {
 	struct address_space *swap_space;
 	struct file *swap_storage;
@@ -390,7 +358,6 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage)
 	int i;
 	int ret = -ENOMEM;
 
-	BUG_ON(ttm->state != tt_unbound && ttm->state != tt_unpopulated);
 	BUG_ON(ttm->caching_state != tt_cached);
 
 	if (!persistent_swap_storage) {
@@ -427,7 +394,7 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage)
 		put_page(to_page);
 	}
 
-	ttm_tt_unpopulate(ttm);
+	ttm_tt_unpopulate(bdev, ttm);
 	ttm->swap_storage = swap_storage;
 	ttm->page_flags |= TTM_PAGE_FLAG_SWAPPED;
 	if (persistent_swap_storage)
@@ -441,7 +408,7 @@ out_err:
 	return ret;
 }
 
-static void ttm_tt_add_mapping(struct ttm_tt *ttm)
+static void ttm_tt_add_mapping(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	pgoff_t i;
 
@@ -449,24 +416,29 @@ static void ttm_tt_add_mapping(struct ttm_tt *ttm)
 		return;
 
 	for (i = 0; i < ttm->num_pages; ++i)
-		ttm->pages[i]->mapping = ttm->bdev->dev_mapping;
+		ttm->pages[i]->mapping = bdev->dev_mapping;
 }
 
-int ttm_tt_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
+int ttm_tt_populate(struct ttm_bo_device *bdev,
+		    struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 {
 	int ret;
 
-	if (ttm->state != tt_unpopulated)
+	if (!ttm)
+		return -EINVAL;
+
+	if (ttm_tt_is_populated(ttm))
 		return 0;
 
-	if (ttm->bdev->driver->ttm_tt_populate)
-		ret = ttm->bdev->driver->ttm_tt_populate(ttm, ctx);
+	if (bdev->driver->ttm_tt_populate)
+		ret = bdev->driver->ttm_tt_populate(bdev, ttm, ctx);
 	else
 		ret = ttm_pool_populate(ttm, ctx);
 	if (!ret)
-		ttm_tt_add_mapping(ttm);
+		ttm_tt_add_mapping(bdev, ttm);
 	return ret;
 }
+EXPORT_SYMBOL(ttm_tt_populate);
 
 static void ttm_tt_clear_mapping(struct ttm_tt *ttm)
 {
@@ -482,14 +454,15 @@ static void ttm_tt_clear_mapping(struct ttm_tt *ttm)
 	}
 }
 
-void ttm_tt_unpopulate(struct ttm_tt *ttm)
+void ttm_tt_unpopulate(struct ttm_bo_device *bdev,
+		       struct ttm_tt *ttm)
 {
-	if (ttm->state == tt_unpopulated)
+	if (!ttm_tt_is_populated(ttm))
 		return;
 
 	ttm_tt_clear_mapping(ttm);
-	if (ttm->bdev->driver->ttm_tt_unpopulate)
-		ttm->bdev->driver->ttm_tt_unpopulate(ttm);
+	if (bdev->driver->ttm_tt_unpopulate)
+		bdev->driver->ttm_tt_unpopulate(bdev, ttm);
 	else
 		ttm_pool_unpopulate(ttm);
 }
diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index 82a7dfdd14c2..9f7c26193831 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -358,18 +358,7 @@ static struct platform_driver v3d_platform_driver = {
 	},
 };
 
-static int __init v3d_drm_register(void)
-{
-	return platform_driver_register(&v3d_platform_driver);
-}
-
-static void __exit v3d_drm_unregister(void)
-{
-	platform_driver_unregister(&v3d_platform_driver);
-}
-
-module_init(v3d_drm_register);
-module_exit(v3d_drm_unregister);
+module_platform_driver(v3d_platform_driver);
 
 MODULE_ALIAS("platform:v3d-drm");
 MODULE_DESCRIPTION("Broadcom V3D DRM Driver");
diff --git a/drivers/gpu/drm/v3d/v3d_mmu.c b/drivers/gpu/drm/v3d/v3d_mmu.c
index 3b81ea28c0bb..5a453532901f 100644
--- a/drivers/gpu/drm/v3d/v3d_mmu.c
+++ b/drivers/gpu/drm/v3d/v3d_mmu.c
@@ -90,18 +90,17 @@ void v3d_mmu_insert_ptes(struct v3d_bo *bo)
 	struct v3d_dev *v3d = to_v3d_dev(shmem_obj->base.dev);
 	u32 page = bo->node.start;
 	u32 page_prot = V3D_PTE_WRITEABLE | V3D_PTE_VALID;
-	unsigned int count;
-	struct scatterlist *sgl;
+	struct sg_dma_page_iter dma_iter;
 
-	for_each_sg(shmem_obj->sgt->sgl, sgl, shmem_obj->sgt->nents, count) {
-		u32 page_address = sg_dma_address(sgl) >> V3D_MMU_PAGE_SHIFT;
+	for_each_sgtable_dma_page(shmem_obj->sgt, &dma_iter, 0) {
+		dma_addr_t dma_addr = sg_page_iter_dma_address(&dma_iter);
+		u32 page_address = dma_addr >> V3D_MMU_PAGE_SHIFT;
 		u32 pte = page_prot | page_address;
 		u32 i;
 
-		BUG_ON(page_address + (sg_dma_len(sgl) >> V3D_MMU_PAGE_SHIFT) >=
+		BUG_ON(page_address + (PAGE_SIZE >> V3D_MMU_PAGE_SHIFT) >=
 		       BIT(24));
-
-		for (i = 0; i < sg_dma_len(sgl) >> V3D_MMU_PAGE_SHIFT; i++)
+		for (i = 0; i < PAGE_SIZE >> V3D_MMU_PAGE_SHIFT; i++)
 			v3d->pt[page++] = pte + i;
 	}
 
diff --git a/drivers/gpu/drm/vboxvideo/vbox_mode.c b/drivers/gpu/drm/vboxvideo/vbox_mode.c
index d9a5af62af89..4fcc0a542b8a 100644
--- a/drivers/gpu/drm/vboxvideo/vbox_mode.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_mode.c
@@ -397,11 +397,13 @@ static void vbox_cursor_atomic_update(struct drm_plane *plane,
 
 	vbox_crtc->cursor_enabled = true;
 
-	/* pinning is done in prepare/cleanup framebuffer */
-	src = drm_gem_vram_kmap(gbo, true, NULL);
+	src = drm_gem_vram_vmap(gbo);
 	if (IS_ERR(src)) {
+		/*
+		 * BUG: we should have pinned the BO in prepare_fb().
+		 */
 		mutex_unlock(&vbox->hw_mutex);
-		DRM_WARN("Could not kmap cursor bo, skipping update\n");
+		DRM_WARN("Could not map cursor bo, skipping update\n");
 		return;
 	}
 
@@ -414,7 +416,7 @@ static void vbox_cursor_atomic_update(struct drm_plane *plane,
 	data_size = width * height * 4 + mask_size;
 
 	copy_cursor_image(src, vbox->cursor_data, width, height, mask_size);
-	drm_gem_vram_kunmap(gbo);
+	drm_gem_vram_vunmap(gbo, src);
 
 	flags = VBOX_MOUSE_POINTER_VISIBLE | VBOX_MOUSE_POINTER_SHAPE |
 		VBOX_MOUSE_POINTER_ALPHA;
diff --git a/drivers/gpu/drm/vc4/Makefile b/drivers/gpu/drm/vc4/Makefile
index b303703bc7f3..d0163e18e9ca 100644
--- a/drivers/gpu/drm/vc4/Makefile
+++ b/drivers/gpu/drm/vc4/Makefile
@@ -12,6 +12,7 @@ vc4-y := \
 	vc4_kms.o \
 	vc4_gem.o \
 	vc4_hdmi.o \
+	vc4_hdmi_phy.o \
 	vc4_vec.o \
 	vc4_hvs.o \
 	vc4_irq.o \
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 6d8fa6118fc1..482219fb4db2 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -65,6 +65,20 @@ static const struct debugfs_reg32 crtc_regs[] = {
 	VC4_REG32(PV_HACT_ACT),
 };
 
+static unsigned int
+vc4_crtc_get_cob_allocation(struct vc4_dev *vc4, unsigned int channel)
+{
+	u32 dispbase = HVS_READ(SCALER_DISPBASEX(channel));
+	/* Top/base are supposed to be 4-pixel aligned, but the
+	 * Raspberry Pi firmware fills the low bits (which are
+	 * presumably ignored).
+	 */
+	u32 top = VC4_GET_FIELD(dispbase, SCALER_DISPBASEX_TOP) & ~3;
+	u32 base = VC4_GET_FIELD(dispbase, SCALER_DISPBASEX_BASE) & ~3;
+
+	return top - base + 4;
+}
+
 static bool vc4_crtc_get_scanout_position(struct drm_crtc *crtc,
 					  bool in_vblank_irq,
 					  int *vpos, int *hpos,
@@ -74,6 +88,8 @@ static bool vc4_crtc_get_scanout_position(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct vc4_crtc_state *vc4_crtc_state = to_vc4_crtc_state(crtc->state);
+	unsigned int cob_size;
 	u32 val;
 	int fifo_lines;
 	int vblank_lines;
@@ -89,7 +105,7 @@ static bool vc4_crtc_get_scanout_position(struct drm_crtc *crtc,
 	 * Read vertical scanline which is currently composed for our
 	 * pixelvalve by the HVS, and also the scaler status.
 	 */
-	val = HVS_READ(SCALER_DISPSTATX(vc4_crtc->channel));
+	val = HVS_READ(SCALER_DISPSTATX(vc4_crtc_state->assigned_channel));
 
 	/* Get optional system timestamp after query. */
 	if (etime)
@@ -109,8 +125,9 @@ static bool vc4_crtc_get_scanout_position(struct drm_crtc *crtc,
 			*hpos += mode->crtc_htotal / 2;
 	}
 
+	cob_size = vc4_crtc_get_cob_allocation(vc4, vc4_crtc_state->assigned_channel);
 	/* This is the offset we need for translating hvs -> pv scanout pos. */
-	fifo_lines = vc4_crtc->cob_size / mode->crtc_hdisplay;
+	fifo_lines = cob_size / mode->crtc_hdisplay;
 
 	if (fifo_lines > 0)
 		ret = true;
@@ -189,10 +206,22 @@ void vc4_crtc_destroy(struct drm_crtc *crtc)
 	drm_crtc_cleanup(crtc);
 }
 
-static u32 vc4_get_fifo_full_level(u32 format)
+static u32 vc4_get_fifo_full_level(struct vc4_crtc *vc4_crtc, u32 format)
 {
-	static const u32 fifo_len_bytes = 64;
+	const struct vc4_crtc_data *crtc_data = vc4_crtc_to_vc4_crtc_data(vc4_crtc);
+	const struct vc4_pv_data *pv_data = vc4_crtc_to_vc4_pv_data(vc4_crtc);
+	u32 fifo_len_bytes = pv_data->fifo_depth;
 
+	/*
+	 * Pixels are pulled from the HVS if the number of bytes is
+	 * lower than the FIFO full level.
+	 *
+	 * The latency of the pixel fetch mechanism is 6 pixels, so we
+	 * need to convert those 6 pixels in bytes, depending on the
+	 * format, and then subtract that from the length of the FIFO
+	 * to make sure we never end up in a situation where the FIFO
+	 * is full.
+	 */
 	switch (format) {
 	case PV_CONTROL_FORMAT_DSIV_16:
 	case PV_CONTROL_FORMAT_DSIC_16:
@@ -202,10 +231,30 @@ static u32 vc4_get_fifo_full_level(u32 format)
 	case PV_CONTROL_FORMAT_24:
 	case PV_CONTROL_FORMAT_DSIV_24:
 	default:
+		/*
+		 * For some reason, the pixelvalve4 doesn't work with
+		 * the usual formula and will only work with 32.
+		 */
+		if (crtc_data->hvs_output == 5)
+			return 32;
+
 		return fifo_len_bytes - 3 * HVS_FIFO_LATENCY_PIX;
 	}
 }
 
+static u32 vc4_crtc_get_fifo_full_level_bits(struct vc4_crtc *vc4_crtc,
+					     u32 format)
+{
+	u32 level = vc4_get_fifo_full_level(vc4_crtc, format);
+	u32 ret = 0;
+
+	ret |= VC4_SET_FIELD((level >> 6),
+			     PV5_CONTROL_FIFO_LEVEL_HIGH);
+
+	return ret | VC4_SET_FIELD(level & 0x3f,
+				   PV_CONTROL_FIFO_LEVEL);
+}
+
 /*
  * Returns the encoder attached to the CRTC.
  *
@@ -230,11 +279,23 @@ static struct drm_encoder *vc4_get_crtc_encoder(struct drm_crtc *crtc)
 	return NULL;
 }
 
+static void vc4_crtc_pixelvalve_reset(struct drm_crtc *crtc)
+{
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+
+	/* The PV needs to be disabled before it can be flushed */
+	CRTC_WRITE(PV_CONTROL, CRTC_READ(PV_CONTROL) & ~PV_CONTROL_EN);
+	CRTC_WRITE(PV_CONTROL, CRTC_READ(PV_CONTROL) | PV_CONTROL_FIFO_CLR);
+}
+
 static void vc4_crtc_config_pv(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_encoder *encoder = vc4_get_crtc_encoder(crtc);
 	struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder);
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	const struct vc4_pv_data *pv_data = vc4_crtc_to_vc4_pv_data(vc4_crtc);
 	struct drm_crtc_state *state = crtc->state;
 	struct drm_display_mode *mode = &state->adjusted_mode;
 	bool interlace = mode->flags & DRM_MODE_FLAG_INTERLACE;
@@ -242,24 +303,29 @@ static void vc4_crtc_config_pv(struct drm_crtc *crtc)
 	bool is_dsi = (vc4_encoder->type == VC4_ENCODER_TYPE_DSI0 ||
 		       vc4_encoder->type == VC4_ENCODER_TYPE_DSI1);
 	u32 format = is_dsi ? PV_CONTROL_FORMAT_DSIV_24 : PV_CONTROL_FORMAT_24;
+	u8 ppc = pv_data->pixels_per_clock;
+	bool debug_dump_regs = false;
 
-	/* Reset the PV fifo. */
-	CRTC_WRITE(PV_CONTROL, 0);
-	CRTC_WRITE(PV_CONTROL, PV_CONTROL_FIFO_CLR | PV_CONTROL_EN);
-	CRTC_WRITE(PV_CONTROL, 0);
+	if (debug_dump_regs) {
+		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
+		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs before:\n",
+			 drm_crtc_index(crtc));
+		drm_print_regset32(&p, &vc4_crtc->regset);
+	}
+
+	vc4_crtc_pixelvalve_reset(crtc);
 
 	CRTC_WRITE(PV_HORZA,
-		   VC4_SET_FIELD((mode->htotal -
-				  mode->hsync_end) * pixel_rep,
+		   VC4_SET_FIELD((mode->htotal - mode->hsync_end) * pixel_rep / ppc,
 				 PV_HORZA_HBP) |
-		   VC4_SET_FIELD((mode->hsync_end -
-				  mode->hsync_start) * pixel_rep,
+		   VC4_SET_FIELD((mode->hsync_end - mode->hsync_start) * pixel_rep / ppc,
 				 PV_HORZA_HSYNC));
+
 	CRTC_WRITE(PV_HORZB,
-		   VC4_SET_FIELD((mode->hsync_start -
-				  mode->hdisplay) * pixel_rep,
+		   VC4_SET_FIELD((mode->hsync_start - mode->hdisplay) * pixel_rep / ppc,
 				 PV_HORZB_HFP) |
-		   VC4_SET_FIELD(mode->hdisplay * pixel_rep, PV_HORZB_HACTIVE));
+		   VC4_SET_FIELD(mode->hdisplay * pixel_rep / ppc,
+				 PV_HORZB_HACTIVE));
 
 	CRTC_WRITE(PV_VERTA,
 		   VC4_SET_FIELD(mode->crtc_vtotal - mode->crtc_vsync_end,
@@ -306,35 +372,20 @@ static void vc4_crtc_config_pv(struct drm_crtc *crtc)
 	if (is_dsi)
 		CRTC_WRITE(PV_HACT_ACT, mode->hdisplay * pixel_rep);
 
-	CRTC_WRITE(PV_CONTROL,
+	if (vc4->hvs->hvs5)
+		CRTC_WRITE(PV_MUX_CFG,
+			   VC4_SET_FIELD(PV_MUX_CFG_RGB_PIXEL_MUX_MODE_NO_SWAP,
+					 PV_MUX_CFG_RGB_PIXEL_MUX_MODE));
+
+	CRTC_WRITE(PV_CONTROL, PV_CONTROL_FIFO_CLR |
+		   vc4_crtc_get_fifo_full_level_bits(vc4_crtc, format) |
 		   VC4_SET_FIELD(format, PV_CONTROL_FORMAT) |
-		   VC4_SET_FIELD(vc4_get_fifo_full_level(format),
-				 PV_CONTROL_FIFO_LEVEL) |
 		   VC4_SET_FIELD(pixel_rep - 1, PV_CONTROL_PIXEL_REP) |
 		   PV_CONTROL_CLR_AT_START |
 		   PV_CONTROL_TRIGGER_UNDERFLOW |
 		   PV_CONTROL_WAIT_HSTART |
 		   VC4_SET_FIELD(vc4_encoder->clock_select,
-				 PV_CONTROL_CLK_SELECT) |
-		   PV_CONTROL_FIFO_CLR |
-		   PV_CONTROL_EN);
-}
-
-static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
-{
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
-	bool debug_dump_regs = false;
-
-	if (debug_dump_regs) {
-		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
-		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs before:\n",
-			 drm_crtc_index(crtc));
-		drm_print_regset32(&p, &vc4_crtc->regset);
-	}
-
-	vc4_crtc_config_pv(crtc);
-
-	vc4_hvs_mode_set_nofb(crtc);
+				 PV_CONTROL_CLK_SELECT));
 
 	if (debug_dump_regs) {
 		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
@@ -352,24 +403,86 @@ static void require_hvs_enabled(struct drm_device *dev)
 		     SCALER_DISPCTRL_ENABLE);
 }
 
+static int vc4_crtc_disable(struct drm_crtc *crtc, unsigned int channel)
+{
+	struct drm_encoder *encoder = vc4_get_crtc_encoder(crtc);
+	struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder);
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	int ret;
+
+	CRTC_WRITE(PV_V_CONTROL,
+		   CRTC_READ(PV_V_CONTROL) & ~PV_VCONTROL_VIDEN);
+	ret = wait_for(!(CRTC_READ(PV_V_CONTROL) & PV_VCONTROL_VIDEN), 1);
+	WARN_ONCE(ret, "Timeout waiting for !PV_VCONTROL_VIDEN\n");
+
+	/*
+	 * This delay is needed to avoid to get a pixel stuck in an
+	 * unflushable FIFO between the pixelvalve and the HDMI
+	 * controllers on the BCM2711.
+	 *
+	 * Timing is fairly sensitive here, so mdelay is the safest
+	 * approach.
+	 *
+	 * If it was to be reworked, the stuck pixel happens on a
+	 * BCM2711 when changing mode with a good probability, so a
+	 * script that changes mode on a regular basis should trigger
+	 * the bug after less than 10 attempts. It manifests itself with
+	 * every pixels being shifted by one to the right, and thus the
+	 * last pixel of a line actually being displayed as the first
+	 * pixel on the next line.
+	 */
+	mdelay(20);
+
+	if (vc4_encoder && vc4_encoder->post_crtc_disable)
+		vc4_encoder->post_crtc_disable(encoder);
+
+	vc4_crtc_pixelvalve_reset(crtc);
+	vc4_hvs_stop_channel(dev, channel);
+
+	if (vc4_encoder && vc4_encoder->post_crtc_powerdown)
+		vc4_encoder->post_crtc_powerdown(encoder);
+
+	return 0;
+}
+
+int vc4_crtc_disable_at_boot(struct drm_crtc *crtc)
+{
+	struct drm_device *drm = crtc->dev;
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	int channel;
+
+	if (!(of_device_is_compatible(vc4_crtc->pdev->dev.of_node,
+				      "brcm,bcm2711-pixelvalve2") ||
+	      of_device_is_compatible(vc4_crtc->pdev->dev.of_node,
+				      "brcm,bcm2711-pixelvalve4")))
+		return 0;
+
+	if (!(CRTC_READ(PV_CONTROL) & PV_CONTROL_EN))
+		return 0;
+
+	if (!(CRTC_READ(PV_V_CONTROL) & PV_VCONTROL_VIDEN))
+		return 0;
+
+	channel = vc4_hvs_get_fifo_from_output(drm, vc4_crtc->data->hvs_output);
+	if (channel < 0)
+		return 0;
+
+	return vc4_crtc_disable(crtc, channel);
+}
+
 static void vc4_crtc_atomic_disable(struct drm_crtc *crtc,
 				    struct drm_crtc_state *old_state)
 {
+	struct vc4_crtc_state *old_vc4_state = to_vc4_crtc_state(old_state);
 	struct drm_device *dev = crtc->dev;
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
-	int ret;
 
 	require_hvs_enabled(dev);
 
 	/* Disable vblank irq handling before crtc is disabled. */
 	drm_crtc_vblank_off(crtc);
 
-	CRTC_WRITE(PV_V_CONTROL,
-		   CRTC_READ(PV_V_CONTROL) & ~PV_VCONTROL_VIDEN);
-	ret = wait_for(!(CRTC_READ(PV_V_CONTROL) & PV_VCONTROL_VIDEN), 1);
-	WARN_ONCE(ret, "Timeout waiting for !PV_VCONTROL_VIDEN\n");
-
-	vc4_hvs_atomic_disable(crtc, old_state);
+	vc4_crtc_disable(crtc, old_vc4_state->assigned_channel);
 
 	/*
 	 * Make sure we issue a vblank event after disabling the CRTC if
@@ -390,6 +503,8 @@ static void vc4_crtc_atomic_enable(struct drm_crtc *crtc,
 {
 	struct drm_device *dev = crtc->dev;
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_encoder *encoder = vc4_get_crtc_encoder(crtc);
+	struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder);
 
 	require_hvs_enabled(dev);
 
@@ -400,11 +515,24 @@ static void vc4_crtc_atomic_enable(struct drm_crtc *crtc,
 
 	vc4_hvs_atomic_enable(crtc, old_state);
 
+	if (vc4_encoder->pre_crtc_configure)
+		vc4_encoder->pre_crtc_configure(encoder);
+
+	vc4_crtc_config_pv(crtc);
+
+	CRTC_WRITE(PV_CONTROL, CRTC_READ(PV_CONTROL) | PV_CONTROL_EN);
+
+	if (vc4_encoder->pre_crtc_enable)
+		vc4_encoder->pre_crtc_enable(encoder);
+
 	/* When feeding the transposer block the pixelvalve is unneeded and
 	 * should not be enabled.
 	 */
 	CRTC_WRITE(PV_V_CONTROL,
 		   CRTC_READ(PV_V_CONTROL) | PV_VCONTROL_VIDEN);
+
+	if (vc4_encoder->post_crtc_enable)
+		vc4_encoder->post_crtc_enable(encoder);
 }
 
 static enum drm_mode_status vc4_crtc_mode_valid(struct drm_crtc *crtc,
@@ -499,7 +627,7 @@ static void vc4_crtc_handle_page_flip(struct vc4_crtc *vc4_crtc)
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
-	u32 chan = vc4_crtc->channel;
+	u32 chan = vc4_state->assigned_channel;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev->event_lock, flags);
@@ -516,7 +644,7 @@ static void vc4_crtc_handle_page_flip(struct vc4_crtc *vc4_crtc)
 		 * the CRTC and encoder already reconfigured, leading to
 		 * underruns. This can be seen when reconfiguring the CRTC.
 		 */
-		vc4_hvs_unmask_underrun(dev, vc4_crtc->channel);
+		vc4_hvs_unmask_underrun(dev, chan);
 	}
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
@@ -698,6 +826,7 @@ struct drm_crtc_state *vc4_crtc_duplicate_state(struct drm_crtc *crtc)
 	old_vc4_state = to_vc4_crtc_state(crtc->state);
 	vc4_state->feed_txp = old_vc4_state->feed_txp;
 	vc4_state->margins = old_vc4_state->margins;
+	vc4_state->assigned_channel = old_vc4_state->assigned_channel;
 
 	__drm_atomic_helper_crtc_duplicate_state(crtc, &vc4_state->base);
 	return &vc4_state->base;
@@ -723,11 +852,19 @@ void vc4_crtc_destroy_state(struct drm_crtc *crtc,
 
 void vc4_crtc_reset(struct drm_crtc *crtc)
 {
+	struct vc4_crtc_state *vc4_crtc_state;
+
 	if (crtc->state)
 		vc4_crtc_destroy_state(crtc, crtc->state);
-	crtc->state = kzalloc(sizeof(struct vc4_crtc_state), GFP_KERNEL);
-	if (crtc->state)
-		__drm_atomic_helper_crtc_reset(crtc, crtc->state);
+
+	vc4_crtc_state = kzalloc(sizeof(*vc4_crtc_state), GFP_KERNEL);
+	if (!vc4_crtc_state) {
+		crtc->state = NULL;
+		return;
+	}
+
+	vc4_crtc_state->assigned_channel = VC4_HVS_CHANNEL_DISABLED;
+	__drm_atomic_helper_crtc_reset(crtc, &vc4_crtc_state->base);
 }
 
 static const struct drm_crtc_funcs vc4_crtc_funcs = {
@@ -747,7 +884,6 @@ static const struct drm_crtc_funcs vc4_crtc_funcs = {
 };
 
 static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
-	.mode_set_nofb = vc4_crtc_mode_set_nofb,
 	.mode_valid = vc4_crtc_mode_valid,
 	.atomic_check = vc4_crtc_atomic_check,
 	.atomic_flush = vc4_hvs_atomic_flush,
@@ -758,9 +894,12 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
 
 static const struct vc4_pv_data bcm2835_pv0_data = {
 	.base = {
-		.hvs_channel = 0,
+		.hvs_available_channels = BIT(0),
+		.hvs_output = 0,
 	},
 	.debugfs_name = "crtc0_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI0,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_DPI,
@@ -769,9 +908,12 @@ static const struct vc4_pv_data bcm2835_pv0_data = {
 
 static const struct vc4_pv_data bcm2835_pv1_data = {
 	.base = {
-		.hvs_channel = 2,
+		.hvs_available_channels = BIT(2),
+		.hvs_output = 2,
 	},
 	.debugfs_name = "crtc1_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI1,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_SMI,
@@ -780,19 +922,94 @@ static const struct vc4_pv_data bcm2835_pv1_data = {
 
 static const struct vc4_pv_data bcm2835_pv2_data = {
 	.base = {
-		.hvs_channel = 1,
+		.hvs_available_channels = BIT(1),
+		.hvs_output = 1,
 	},
 	.debugfs_name = "crtc2_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
 	.encoder_types = {
-		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_HDMI,
+		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_HDMI0,
 		[PV_CONTROL_CLK_SELECT_VEC] = VC4_ENCODER_TYPE_VEC,
 	},
 };
 
+static const struct vc4_pv_data bcm2711_pv0_data = {
+	.base = {
+		.hvs_available_channels = BIT(0),
+		.hvs_output = 0,
+	},
+	.debugfs_name = "crtc0_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
+	.encoder_types = {
+		[0] = VC4_ENCODER_TYPE_DSI0,
+		[1] = VC4_ENCODER_TYPE_DPI,
+	},
+};
+
+static const struct vc4_pv_data bcm2711_pv1_data = {
+	.base = {
+		.hvs_available_channels = BIT(0) | BIT(1) | BIT(2),
+		.hvs_output = 3,
+	},
+	.debugfs_name = "crtc1_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
+	.encoder_types = {
+		[0] = VC4_ENCODER_TYPE_DSI1,
+		[1] = VC4_ENCODER_TYPE_SMI,
+	},
+};
+
+static const struct vc4_pv_data bcm2711_pv2_data = {
+	.base = {
+		.hvs_available_channels = BIT(0) | BIT(1) | BIT(2),
+		.hvs_output = 4,
+	},
+	.debugfs_name = "crtc2_regs",
+	.fifo_depth = 256,
+	.pixels_per_clock = 2,
+	.encoder_types = {
+		[0] = VC4_ENCODER_TYPE_HDMI0,
+	},
+};
+
+static const struct vc4_pv_data bcm2711_pv3_data = {
+	.base = {
+		.hvs_available_channels = BIT(1),
+		.hvs_output = 1,
+	},
+	.debugfs_name = "crtc3_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 1,
+	.encoder_types = {
+		[0] = VC4_ENCODER_TYPE_VEC,
+	},
+};
+
+static const struct vc4_pv_data bcm2711_pv4_data = {
+	.base = {
+		.hvs_available_channels = BIT(0) | BIT(1) | BIT(2),
+		.hvs_output = 5,
+	},
+	.debugfs_name = "crtc4_regs",
+	.fifo_depth = 64,
+	.pixels_per_clock = 2,
+	.encoder_types = {
+		[0] = VC4_ENCODER_TYPE_HDMI1,
+	},
+};
+
 static const struct of_device_id vc4_crtc_dt_match[] = {
 	{ .compatible = "brcm,bcm2835-pixelvalve0", .data = &bcm2835_pv0_data },
 	{ .compatible = "brcm,bcm2835-pixelvalve1", .data = &bcm2835_pv1_data },
 	{ .compatible = "brcm,bcm2835-pixelvalve2", .data = &bcm2835_pv2_data },
+	{ .compatible = "brcm,bcm2711-pixelvalve0", .data = &bcm2711_pv0_data },
+	{ .compatible = "brcm,bcm2711-pixelvalve1", .data = &bcm2711_pv1_data },
+	{ .compatible = "brcm,bcm2711-pixelvalve2", .data = &bcm2711_pv2_data },
+	{ .compatible = "brcm,bcm2711-pixelvalve3", .data = &bcm2711_pv3_data },
+	{ .compatible = "brcm,bcm2711-pixelvalve4", .data = &bcm2711_pv4_data },
 	{}
 };
 
@@ -819,26 +1036,11 @@ static void vc4_set_crtc_possible_masks(struct drm_device *drm,
 	}
 }
 
-static void
-vc4_crtc_get_cob_allocation(struct vc4_crtc *vc4_crtc)
-{
-	struct drm_device *drm = vc4_crtc->base.dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
-	u32 dispbase = HVS_READ(SCALER_DISPBASEX(vc4_crtc->channel));
-	/* Top/base are supposed to be 4-pixel aligned, but the
-	 * Raspberry Pi firmware fills the low bits (which are
-	 * presumably ignored).
-	 */
-	u32 top = VC4_GET_FIELD(dispbase, SCALER_DISPBASEX_TOP) & ~3;
-	u32 base = VC4_GET_FIELD(dispbase, SCALER_DISPBASEX_BASE) & ~3;
-
-	vc4_crtc->cob_size = top - base + 4;
-}
-
 int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc,
 		  const struct drm_crtc_funcs *crtc_funcs,
 		  const struct drm_crtc_helper_funcs *crtc_helper_funcs)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(drm);
 	struct drm_crtc *crtc = &vc4_crtc->base;
 	struct drm_plane *primary_plane;
 	unsigned int i;
@@ -858,15 +1060,17 @@ int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc,
 	drm_crtc_init_with_planes(drm, crtc, primary_plane, NULL,
 				  crtc_funcs, NULL);
 	drm_crtc_helper_add(crtc, crtc_helper_funcs);
-	vc4_crtc->channel = vc4_crtc->data->hvs_channel;
-	drm_mode_crtc_set_gamma_size(crtc, ARRAY_SIZE(vc4_crtc->lut_r));
-	drm_crtc_enable_color_mgmt(crtc, 0, false, crtc->gamma_size);
 
-	/* We support CTM, but only for one CRTC at a time. It's therefore
-	 * implemented as private driver state in vc4_kms, not here.
-	 */
-	drm_crtc_enable_color_mgmt(crtc, 0, true, crtc->gamma_size);
-	vc4_crtc_get_cob_allocation(vc4_crtc);
+	if (!vc4->hvs->hvs5) {
+		drm_mode_crtc_set_gamma_size(crtc, ARRAY_SIZE(vc4_crtc->lut_r));
+
+		drm_crtc_enable_color_mgmt(crtc, 0, false, crtc->gamma_size);
+
+		/* We support CTM, but only for one CRTC at a time. It's therefore
+		 * implemented as private driver state in vc4_kms, not here.
+		 */
+		drm_crtc_enable_color_mgmt(crtc, 0, true, crtc->gamma_size);
+	}
 
 	for (i = 0; i < crtc->gamma_size; i++) {
 		vc4_crtc->lut_r[i] = i;
@@ -915,7 +1119,9 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
 	CRTC_WRITE(PV_INTEN, 0);
 	CRTC_WRITE(PV_INTSTAT, PV_INT_VFP_START);
 	ret = devm_request_irq(dev, platform_get_irq(pdev, 0),
-			       vc4_crtc_irq_handler, 0, "vc4 crtc", vc4_crtc);
+			       vc4_crtc_irq_handler,
+			       IRQF_SHARED,
+			       "vc4 crtc", vc4_crtc);
 	if (ret)
 		goto err_destroy_planes;
 
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index 38343d2fb4fb..f1a5fd5dab6f 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -252,6 +252,7 @@ static int vc4_drm_bind(struct device *dev)
 	struct drm_device *drm;
 	struct vc4_dev *vc4;
 	struct device_node *node;
+	struct drm_crtc *crtc;
 	int ret = 0;
 
 	dev->coherent_dma_mask = DMA_BIT_MASK(32);
@@ -298,6 +299,9 @@ static int vc4_drm_bind(struct device *dev)
 	if (ret < 0)
 		goto unbind_all;
 
+	drm_for_each_crtc(crtc, drm)
+		vc4_crtc_disable_at_boot(crtc);
+
 	ret = drm_dev_register(drm, 0);
 	if (ret < 0)
 		goto unbind_all;
@@ -368,6 +372,7 @@ static int vc4_platform_drm_remove(struct platform_device *pdev)
 }
 
 static const struct of_device_id vc4_of_match[] = {
+	{ .compatible = "brcm,bcm2711-vc5", },
 	{ .compatible = "brcm,bcm2835-vc4", },
 	{ .compatible = "brcm,cygnus-vc4", },
 	{},
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index fa19160c801f..90b911fd2a7f 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -73,7 +73,6 @@ struct vc4_perfmon {
 struct vc4_dev {
 	struct drm_device *dev;
 
-	struct vc4_hdmi *hdmi;
 	struct vc4_hvs *hvs;
 	struct vc4_v3d *v3d;
 	struct vc4_dpi *dpi;
@@ -201,6 +200,9 @@ struct vc4_dev {
 
 	int power_refcount;
 
+	/* Set to true when the load tracker is supported. */
+	bool load_tracker_available;
+
 	/* Set to true when the load tracker is active. */
 	bool load_tracker_enabled;
 
@@ -320,6 +322,8 @@ struct vc4_hvs {
 	void __iomem *regs;
 	u32 __iomem *dlist;
 
+	struct clk *core_clk;
+
 	/* Memory manager for CRTCs to allocate space in the display
 	 * list.  Units are dwords.
 	 */
@@ -329,7 +333,11 @@ struct vc4_hvs {
 	spinlock_t mm_lock;
 
 	struct drm_mm_node mitchell_netravali_filter;
+
 	struct debugfs_regset32 regset;
+
+	/* HVS version 5 flag, therefore requires updated dlist structures */
+	bool hvs5;
 };
 
 struct vc4_plane {
@@ -420,7 +428,8 @@ to_vc4_plane_state(struct drm_plane_state *state)
 
 enum vc4_encoder_type {
 	VC4_ENCODER_TYPE_NONE,
-	VC4_ENCODER_TYPE_HDMI,
+	VC4_ENCODER_TYPE_HDMI0,
+	VC4_ENCODER_TYPE_HDMI1,
 	VC4_ENCODER_TYPE_VEC,
 	VC4_ENCODER_TYPE_DSI0,
 	VC4_ENCODER_TYPE_DSI1,
@@ -432,6 +441,13 @@ struct vc4_encoder {
 	struct drm_encoder base;
 	enum vc4_encoder_type type;
 	u32 clock_select;
+
+	void (*pre_crtc_configure)(struct drm_encoder *encoder);
+	void (*pre_crtc_enable)(struct drm_encoder *encoder);
+	void (*post_crtc_enable)(struct drm_encoder *encoder);
+
+	void (*post_crtc_disable)(struct drm_encoder *encoder);
+	void (*post_crtc_powerdown)(struct drm_encoder *encoder);
 };
 
 static inline struct vc4_encoder *
@@ -441,13 +457,22 @@ to_vc4_encoder(struct drm_encoder *encoder)
 }
 
 struct vc4_crtc_data {
-	/* Which channel of the HVS this pixelvalve sources from. */
-	int hvs_channel;
+	/* Bitmask of channels (FIFOs) of the HVS that the output can source from */
+	unsigned int hvs_available_channels;
+
+	/* Which output of the HVS this pixelvalve sources from. */
+	int hvs_output;
 };
 
 struct vc4_pv_data {
 	struct vc4_crtc_data	base;
 
+	/* Depth of the PixelValve FIFO in bytes */
+	unsigned int fifo_depth;
+
+	/* Number of pixels output per clock period */
+	u8 pixels_per_clock;
+
 	enum vc4_encoder_type encoder_types[4];
 	const char *debugfs_name;
 
@@ -462,14 +487,9 @@ struct vc4_crtc {
 	/* Timestamp at start of vblank irq - unaffected by lock delays. */
 	ktime_t t_vblank;
 
-	/* Which HVS channel we're using for our CRTC. */
-	int channel;
-
 	u8 lut_r[256];
 	u8 lut_g[256];
 	u8 lut_b[256];
-	/* Size in pixels of the COB memory allocated to this CRTC. */
-	u32 cob_size;
 
 	struct drm_pending_vblank_event *event;
 
@@ -502,6 +522,7 @@ struct vc4_crtc_state {
 	struct drm_mm_node mm;
 	bool feed_txp;
 	bool txp_armed;
+	unsigned int assigned_channel;
 
 	struct {
 		unsigned int left;
@@ -511,6 +532,8 @@ struct vc4_crtc_state {
 	} margins;
 };
 
+#define VC4_HVS_CHANNEL_DISABLED ((unsigned int)-1)
+
 static inline struct vc4_crtc_state *
 to_vc4_crtc_state(struct drm_crtc_state *crtc_state)
 {
@@ -794,6 +817,7 @@ void vc4_bo_remove_from_purgeable_pool(struct vc4_bo *bo);
 
 /* vc4_crtc.c */
 extern struct platform_driver vc4_crtc_driver;
+int vc4_crtc_disable_at_boot(struct drm_crtc *crtc);
 int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc,
 		  const struct drm_crtc_funcs *crtc_funcs,
 		  const struct drm_crtc_helper_funcs *crtc_helper_funcs);
@@ -888,11 +912,12 @@ void vc4_irq_reset(struct drm_device *dev);
 
 /* vc4_hvs.c */
 extern struct platform_driver vc4_hvs_driver;
+void vc4_hvs_stop_channel(struct drm_device *dev, unsigned int output);
+int vc4_hvs_get_fifo_from_output(struct drm_device *dev, unsigned int output);
 int vc4_hvs_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state);
 void vc4_hvs_atomic_enable(struct drm_crtc *crtc, struct drm_crtc_state *old_state);
 void vc4_hvs_atomic_disable(struct drm_crtc *crtc, struct drm_crtc_state *old_state);
 void vc4_hvs_atomic_flush(struct drm_crtc *crtc, struct drm_crtc_state *state);
-void vc4_hvs_mode_set_nofb(struct drm_crtc *crtc);
 void vc4_hvs_dump_state(struct drm_device *dev);
 void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel);
 void vc4_hvs_mask_underrun(struct drm_device *dev, int channel);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 6339c6f0f571..e8f99e290655 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -43,177 +43,101 @@
 #include <linux/of_platform.h>
 #include <linux/pm_runtime.h>
 #include <linux/rational.h>
+#include <linux/reset.h>
 #include <sound/dmaengine_pcm.h>
 #include <sound/pcm_drm_eld.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "media/cec.h"
 #include "vc4_drv.h"
+#include "vc4_hdmi.h"
+#include "vc4_hdmi_regs.h"
 #include "vc4_regs.h"
 
-#define HSM_CLOCK_FREQ 163682864
-#define CEC_CLOCK_FREQ 40000
-#define CEC_CLOCK_DIV  (HSM_CLOCK_FREQ / CEC_CLOCK_FREQ)
-
-/* HDMI audio information */
-struct vc4_hdmi_audio {
-	struct snd_soc_card card;
-	struct snd_soc_dai_link link;
-	struct snd_soc_dai_link_component cpu;
-	struct snd_soc_dai_link_component codec;
-	struct snd_soc_dai_link_component platform;
-	int samplerate;
-	int channels;
-	struct snd_dmaengine_dai_dma_data dma_data;
-	struct snd_pcm_substream *substream;
-};
+#define VC5_HDMI_HORZA_HFP_SHIFT		16
+#define VC5_HDMI_HORZA_HFP_MASK			VC4_MASK(28, 16)
+#define VC5_HDMI_HORZA_VPOS			BIT(15)
+#define VC5_HDMI_HORZA_HPOS			BIT(14)
+#define VC5_HDMI_HORZA_HAP_SHIFT		0
+#define VC5_HDMI_HORZA_HAP_MASK			VC4_MASK(13, 0)
 
-/* General HDMI hardware state. */
-struct vc4_hdmi {
-	struct platform_device *pdev;
-
-	struct drm_encoder *encoder;
-	struct drm_connector *connector;
+#define VC5_HDMI_HORZB_HBP_SHIFT		16
+#define VC5_HDMI_HORZB_HBP_MASK			VC4_MASK(26, 16)
+#define VC5_HDMI_HORZB_HSP_SHIFT		0
+#define VC5_HDMI_HORZB_HSP_MASK			VC4_MASK(10, 0)
 
-	struct vc4_hdmi_audio audio;
+#define VC5_HDMI_VERTA_VSP_SHIFT		24
+#define VC5_HDMI_VERTA_VSP_MASK			VC4_MASK(28, 24)
+#define VC5_HDMI_VERTA_VFP_SHIFT		16
+#define VC5_HDMI_VERTA_VFP_MASK			VC4_MASK(22, 16)
+#define VC5_HDMI_VERTA_VAL_SHIFT		0
+#define VC5_HDMI_VERTA_VAL_MASK			VC4_MASK(12, 0)
 
-	struct i2c_adapter *ddc;
-	void __iomem *hdmicore_regs;
-	void __iomem *hd_regs;
-	int hpd_gpio;
-	bool hpd_active_low;
+#define VC5_HDMI_VERTB_VSPO_SHIFT		16
+#define VC5_HDMI_VERTB_VSPO_MASK		VC4_MASK(29, 16)
 
-	struct cec_adapter *cec_adap;
-	struct cec_msg cec_rx_msg;
-	bool cec_tx_ok;
-	bool cec_irq_was_rx;
+# define VC4_HD_M_SW_RST			BIT(2)
+# define VC4_HD_M_ENABLE			BIT(0)
 
-	struct clk *pixel_clock;
-	struct clk *hsm_clock;
+#define CEC_CLOCK_FREQ 40000
+#define VC4_HSM_MID_CLOCK 149985000
 
-	struct debugfs_regset32 hdmi_regset;
-	struct debugfs_regset32 hd_regset;
-};
+static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct vc4_hdmi *vc4_hdmi = node->info_ent->data;
+	struct drm_printer p = drm_seq_file_printer(m);
 
-#define HDMI_READ(offset) readl(vc4->hdmi->hdmicore_regs + offset)
-#define HDMI_WRITE(offset, val) writel(val, vc4->hdmi->hdmicore_regs + offset)
-#define HD_READ(offset) readl(vc4->hdmi->hd_regs + offset)
-#define HD_WRITE(offset, val) writel(val, vc4->hdmi->hd_regs + offset)
+	drm_print_regset32(&p, &vc4_hdmi->hdmi_regset);
+	drm_print_regset32(&p, &vc4_hdmi->hd_regset);
 
-/* VC4 HDMI encoder KMS struct */
-struct vc4_hdmi_encoder {
-	struct vc4_encoder base;
-	bool hdmi_monitor;
-	bool limited_rgb_range;
-};
+	return 0;
+}
 
-static inline struct vc4_hdmi_encoder *
-to_vc4_hdmi_encoder(struct drm_encoder *encoder)
+static void vc4_hdmi_reset(struct vc4_hdmi *vc4_hdmi)
 {
-	return container_of(encoder, struct vc4_hdmi_encoder, base.base);
-}
+	HDMI_WRITE(HDMI_M_CTL, VC4_HD_M_SW_RST);
+	udelay(1);
+	HDMI_WRITE(HDMI_M_CTL, 0);
 
-/* VC4 HDMI connector KMS struct */
-struct vc4_hdmi_connector {
-	struct drm_connector base;
+	HDMI_WRITE(HDMI_M_CTL, VC4_HD_M_ENABLE);
 
-	/* Since the connector is attached to just the one encoder,
-	 * this is the reference to it so we can do the best_encoder()
-	 * hook.
-	 */
-	struct drm_encoder *encoder;
-};
+	HDMI_WRITE(HDMI_SW_RESET_CONTROL,
+		   VC4_HDMI_SW_RESET_HDMI |
+		   VC4_HDMI_SW_RESET_FORMAT_DETECT);
 
-static inline struct vc4_hdmi_connector *
-to_vc4_hdmi_connector(struct drm_connector *connector)
-{
-	return container_of(connector, struct vc4_hdmi_connector, base);
+	HDMI_WRITE(HDMI_SW_RESET_CONTROL, 0);
 }
 
-static const struct debugfs_reg32 hdmi_regs[] = {
-	VC4_REG32(VC4_HDMI_CORE_REV),
-	VC4_REG32(VC4_HDMI_SW_RESET_CONTROL),
-	VC4_REG32(VC4_HDMI_HOTPLUG_INT),
-	VC4_REG32(VC4_HDMI_HOTPLUG),
-	VC4_REG32(VC4_HDMI_MAI_CHANNEL_MAP),
-	VC4_REG32(VC4_HDMI_MAI_CONFIG),
-	VC4_REG32(VC4_HDMI_MAI_FORMAT),
-	VC4_REG32(VC4_HDMI_AUDIO_PACKET_CONFIG),
-	VC4_REG32(VC4_HDMI_RAM_PACKET_CONFIG),
-	VC4_REG32(VC4_HDMI_HORZA),
-	VC4_REG32(VC4_HDMI_HORZB),
-	VC4_REG32(VC4_HDMI_FIFO_CTL),
-	VC4_REG32(VC4_HDMI_SCHEDULER_CONTROL),
-	VC4_REG32(VC4_HDMI_VERTA0),
-	VC4_REG32(VC4_HDMI_VERTA1),
-	VC4_REG32(VC4_HDMI_VERTB0),
-	VC4_REG32(VC4_HDMI_VERTB1),
-	VC4_REG32(VC4_HDMI_TX_PHY_RESET_CTL),
-	VC4_REG32(VC4_HDMI_TX_PHY_CTL0),
-
-	VC4_REG32(VC4_HDMI_CEC_CNTRL_1),
-	VC4_REG32(VC4_HDMI_CEC_CNTRL_2),
-	VC4_REG32(VC4_HDMI_CEC_CNTRL_3),
-	VC4_REG32(VC4_HDMI_CEC_CNTRL_4),
-	VC4_REG32(VC4_HDMI_CEC_CNTRL_5),
-	VC4_REG32(VC4_HDMI_CPU_STATUS),
-	VC4_REG32(VC4_HDMI_CPU_MASK_STATUS),
-
-	VC4_REG32(VC4_HDMI_CEC_RX_DATA_1),
-	VC4_REG32(VC4_HDMI_CEC_RX_DATA_2),
-	VC4_REG32(VC4_HDMI_CEC_RX_DATA_3),
-	VC4_REG32(VC4_HDMI_CEC_RX_DATA_4),
-	VC4_REG32(VC4_HDMI_CEC_TX_DATA_1),
-	VC4_REG32(VC4_HDMI_CEC_TX_DATA_2),
-	VC4_REG32(VC4_HDMI_CEC_TX_DATA_3),
-	VC4_REG32(VC4_HDMI_CEC_TX_DATA_4),
-};
-
-static const struct debugfs_reg32 hd_regs[] = {
-	VC4_REG32(VC4_HD_M_CTL),
-	VC4_REG32(VC4_HD_MAI_CTL),
-	VC4_REG32(VC4_HD_MAI_THR),
-	VC4_REG32(VC4_HD_MAI_FMT),
-	VC4_REG32(VC4_HD_MAI_SMP),
-	VC4_REG32(VC4_HD_VID_CTL),
-	VC4_REG32(VC4_HD_CSC_CTL),
-	VC4_REG32(VC4_HD_FRAME_COUNT),
-};
-
-static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
+static void vc5_hdmi_reset(struct vc4_hdmi *vc4_hdmi)
 {
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_hdmi *hdmi = vc4->hdmi;
-	struct drm_printer p = drm_seq_file_printer(m);
+	reset_control_reset(vc4_hdmi->reset);
 
-	drm_print_regset32(&p, &hdmi->hdmi_regset);
-	drm_print_regset32(&p, &hdmi->hd_regset);
+	HDMI_WRITE(HDMI_DVP_CTL, 0);
 
-	return 0;
+	HDMI_WRITE(HDMI_CLOCK_STOP,
+		   HDMI_READ(HDMI_CLOCK_STOP) | VC4_DVP_HT_CLOCK_STOP_PIXEL);
 }
 
 static enum drm_connector_status
 vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 {
-	struct drm_device *dev = connector->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
 
-	if (vc4->hdmi->hpd_gpio) {
-		if (gpio_get_value_cansleep(vc4->hdmi->hpd_gpio) ^
-		    vc4->hdmi->hpd_active_low)
+	if (vc4_hdmi->hpd_gpio) {
+		if (gpio_get_value_cansleep(vc4_hdmi->hpd_gpio) ^
+		    vc4_hdmi->hpd_active_low)
 			return connector_status_connected;
-		cec_phys_addr_invalidate(vc4->hdmi->cec_adap);
+		cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 		return connector_status_disconnected;
 	}
 
-	if (drm_probe_ddc(vc4->hdmi->ddc))
+	if (drm_probe_ddc(vc4_hdmi->ddc))
 		return connector_status_connected;
 
-	if (HDMI_READ(VC4_HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED)
+	if (HDMI_READ(HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED)
 		return connector_status_connected;
-	cec_phys_addr_invalidate(vc4->hdmi->cec_adap);
+	cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 	return connector_status_disconnected;
 }
 
@@ -225,17 +149,13 @@ static void vc4_hdmi_connector_destroy(struct drm_connector *connector)
 
 static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
 {
-	struct vc4_hdmi_connector *vc4_connector =
-		to_vc4_hdmi_connector(connector);
-	struct drm_encoder *encoder = vc4_connector->encoder;
-	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
-	struct drm_device *dev = connector->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
+	struct vc4_hdmi_encoder *vc4_encoder = &vc4_hdmi->encoder;
 	int ret = 0;
 	struct edid *edid;
 
-	edid = drm_get_edid(connector, vc4->hdmi->ddc);
-	cec_s_phys_addr_from_edid(vc4->hdmi->cec_adap, edid);
+	edid = drm_get_edid(connector, vc4_hdmi->ddc);
+	cec_s_phys_addr_from_edid(vc4_hdmi->cec_adap, edid);
 	if (!edid)
 		return -ENODEV;
 
@@ -267,32 +187,23 @@ static const struct drm_connector_helper_funcs vc4_hdmi_connector_helper_funcs =
 	.get_modes = vc4_hdmi_connector_get_modes,
 };
 
-static struct drm_connector *vc4_hdmi_connector_init(struct drm_device *dev,
-						     struct drm_encoder *encoder,
-						     struct i2c_adapter *ddc)
+static int vc4_hdmi_connector_init(struct drm_device *dev,
+				   struct vc4_hdmi *vc4_hdmi)
 {
-	struct drm_connector *connector;
-	struct vc4_hdmi_connector *hdmi_connector;
+	struct drm_connector *connector = &vc4_hdmi->connector;
+	struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base;
 	int ret;
 
-	hdmi_connector = devm_kzalloc(dev->dev, sizeof(*hdmi_connector),
-				      GFP_KERNEL);
-	if (!hdmi_connector)
-		return ERR_PTR(-ENOMEM);
-	connector = &hdmi_connector->base;
-
-	hdmi_connector->encoder = encoder;
-
 	drm_connector_init_with_ddc(dev, connector,
 				    &vc4_hdmi_connector_funcs,
 				    DRM_MODE_CONNECTOR_HDMIA,
-				    ddc);
+				    vc4_hdmi->ddc);
 	drm_connector_helper_add(connector, &vc4_hdmi_connector_helper_funcs);
 
 	/* Create and attach TV margin props to this connector. */
 	ret = drm_mode_create_tv_margin_properties(dev);
 	if (ret)
-		return ERR_PTR(ret);
+		return ret;
 
 	drm_connector_attach_tv_margin_properties(connector);
 
@@ -304,35 +215,37 @@ static struct drm_connector *vc4_hdmi_connector_init(struct drm_device *dev,
 
 	drm_connector_attach_encoder(connector, encoder);
 
-	return connector;
+	return 0;
 }
 
 static int vc4_hdmi_stop_packet(struct drm_encoder *encoder,
 				enum hdmi_infoframe_type type)
 {
-	struct drm_device *dev = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	u32 packet_id = type - 0x80;
 
-	HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG,
-		   HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) & ~BIT(packet_id));
+	HDMI_WRITE(HDMI_RAM_PACKET_CONFIG,
+		   HDMI_READ(HDMI_RAM_PACKET_CONFIG) & ~BIT(packet_id));
 
-	return wait_for(!(HDMI_READ(VC4_HDMI_RAM_PACKET_STATUS) &
+	return wait_for(!(HDMI_READ(HDMI_RAM_PACKET_STATUS) &
 			  BIT(packet_id)), 100);
 }
 
 static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder,
 				     union hdmi_infoframe *frame)
 {
-	struct drm_device *dev = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	u32 packet_id = frame->any.type - 0x80;
-	u32 packet_reg = VC4_HDMI_RAM_PACKET(packet_id);
+	const struct vc4_hdmi_register *ram_packet_start =
+		&vc4_hdmi->variant->registers[HDMI_RAM_PACKET_START];
+	u32 packet_reg = ram_packet_start->offset + VC4_HDMI_PACKET_STRIDE * packet_id;
+	void __iomem *base = __vc4_hdmi_get_field_base(vc4_hdmi,
+						       ram_packet_start->reg);
 	uint8_t buffer[VC4_HDMI_PACKET_STRIDE];
 	ssize_t len, i;
 	int ret;
 
-	WARN_ONCE(!(HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) &
+	WARN_ONCE(!(HDMI_READ(HDMI_RAM_PACKET_CONFIG) &
 		    VC4_HDMI_RAM_PACKET_ENABLE),
 		  "Packet RAM has to be on to store the packet.");
 
@@ -347,23 +260,23 @@ static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder,
 	}
 
 	for (i = 0; i < len; i += 7) {
-		HDMI_WRITE(packet_reg,
-			   buffer[i + 0] << 0 |
-			   buffer[i + 1] << 8 |
-			   buffer[i + 2] << 16);
+		writel(buffer[i + 0] << 0 |
+		       buffer[i + 1] << 8 |
+		       buffer[i + 2] << 16,
+		       base + packet_reg);
 		packet_reg += 4;
 
-		HDMI_WRITE(packet_reg,
-			   buffer[i + 3] << 0 |
-			   buffer[i + 4] << 8 |
-			   buffer[i + 5] << 16 |
-			   buffer[i + 6] << 24);
+		writel(buffer[i + 3] << 0 |
+		       buffer[i + 4] << 8 |
+		       buffer[i + 5] << 16 |
+		       buffer[i + 6] << 24,
+		       base + packet_reg);
 		packet_reg += 4;
 	}
 
-	HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG,
-		   HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) | BIT(packet_id));
-	ret = wait_for((HDMI_READ(VC4_HDMI_RAM_PACKET_STATUS) &
+	HDMI_WRITE(HDMI_RAM_PACKET_CONFIG,
+		   HDMI_READ(HDMI_RAM_PACKET_CONFIG) | BIT(packet_id));
+	ret = wait_for((HDMI_READ(HDMI_RAM_PACKET_STATUS) &
 			BIT(packet_id)), 100);
 	if (ret)
 		DRM_ERROR("Failed to wait for infoframe to start: %d\n", ret);
@@ -371,24 +284,24 @@ static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder,
 
 static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder)
 {
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
-	struct vc4_dev *vc4 = encoder->dev->dev_private;
-	struct vc4_hdmi *hdmi = vc4->hdmi;
-	struct drm_connector_state *cstate = hdmi->connector->state;
+	struct drm_connector *connector = &vc4_hdmi->connector;
+	struct drm_connector_state *cstate = connector->state;
 	struct drm_crtc *crtc = encoder->crtc;
 	const struct drm_display_mode *mode = &crtc->state->adjusted_mode;
 	union hdmi_infoframe frame;
 	int ret;
 
 	ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
-						       hdmi->connector, mode);
+						       connector, mode);
 	if (ret < 0) {
 		DRM_ERROR("couldn't fill AVI infoframe\n");
 		return;
 	}
 
 	drm_hdmi_avi_infoframe_quant_range(&frame.avi,
-					   hdmi->connector, mode,
+					   connector, mode,
 					   vc4_encoder->limited_rgb_range ?
 					   HDMI_QUANTIZATION_RANGE_LIMITED :
 					   HDMI_QUANTIZATION_RANGE_FULL);
@@ -416,9 +329,7 @@ static void vc4_hdmi_set_spd_infoframe(struct drm_encoder *encoder)
 
 static void vc4_hdmi_set_audio_infoframe(struct drm_encoder *encoder)
 {
-	struct drm_device *drm = encoder->dev;
-	struct vc4_dev *vc4 = drm->dev_private;
-	struct vc4_hdmi *hdmi = vc4->hdmi;
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	union hdmi_infoframe frame;
 	int ret;
 
@@ -427,45 +338,139 @@ static void vc4_hdmi_set_audio_infoframe(struct drm_encoder *encoder)
 	frame.audio.coding_type = HDMI_AUDIO_CODING_TYPE_STREAM;
 	frame.audio.sample_frequency = HDMI_AUDIO_SAMPLE_FREQUENCY_STREAM;
 	frame.audio.sample_size = HDMI_AUDIO_SAMPLE_SIZE_STREAM;
-	frame.audio.channels = hdmi->audio.channels;
+	frame.audio.channels = vc4_hdmi->audio.channels;
 
 	vc4_hdmi_write_infoframe(encoder, &frame);
 }
 
 static void vc4_hdmi_set_infoframes(struct drm_encoder *encoder)
 {
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
+
 	vc4_hdmi_set_avi_infoframe(encoder);
 	vc4_hdmi_set_spd_infoframe(encoder);
+	/*
+	 * If audio was streaming, then we need to reenabled the audio
+	 * infoframe here during encoder_enable.
+	 */
+	if (vc4_hdmi->audio.streaming)
+		vc4_hdmi_set_audio_infoframe(encoder);
 }
 
-static void vc4_hdmi_encoder_disable(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_post_crtc_disable(struct drm_encoder *encoder)
 {
-	struct drm_device *dev = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_hdmi *hdmi = vc4->hdmi;
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
+
+	HDMI_WRITE(HDMI_RAM_PACKET_CONFIG, 0);
+
+	HDMI_WRITE(HDMI_VID_CTL, HDMI_READ(HDMI_VID_CTL) |
+		   VC4_HD_VID_CTL_CLRRGB | VC4_HD_VID_CTL_CLRSYNC);
+
+	HDMI_WRITE(HDMI_VID_CTL,
+		   HDMI_READ(HDMI_VID_CTL) | VC4_HD_VID_CTL_BLANKPIX);
+}
+
+static void vc4_hdmi_encoder_post_crtc_powerdown(struct drm_encoder *encoder)
+{
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	int ret;
 
-	HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG, 0);
+	if (vc4_hdmi->variant->phy_disable)
+		vc4_hdmi->variant->phy_disable(vc4_hdmi);
 
-	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0xf << 16);
-	HD_WRITE(VC4_HD_VID_CTL,
-		 HD_READ(VC4_HD_VID_CTL) & ~VC4_HD_VID_CTL_ENABLE);
+	HDMI_WRITE(HDMI_VID_CTL,
+		   HDMI_READ(HDMI_VID_CTL) & ~VC4_HD_VID_CTL_ENABLE);
 
-	clk_disable_unprepare(hdmi->pixel_clock);
+	clk_disable_unprepare(vc4_hdmi->pixel_bvb_clock);
+	clk_disable_unprepare(vc4_hdmi->hsm_clock);
+	clk_disable_unprepare(vc4_hdmi->pixel_clock);
 
-	ret = pm_runtime_put(&hdmi->pdev->dev);
+	ret = pm_runtime_put(&vc4_hdmi->pdev->dev);
 	if (ret < 0)
 		DRM_ERROR("Failed to release power domain: %d\n", ret);
 }
 
-static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_disable(struct drm_encoder *encoder)
+{
+}
+
+static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable)
+{
+	u32 csc_ctl;
+
+	csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR,
+				VC4_HD_CSC_CTL_ORDER);
+
+	if (enable) {
+		/* CEA VICs other than #1 requre limited range RGB
+		 * output unless overridden by an AVI infoframe.
+		 * Apply a colorspace conversion to squash 0-255 down
+		 * to 16-235.  The matrix here is:
+		 *
+		 * [ 0      0      0.8594 16]
+		 * [ 0      0.8594 0      16]
+		 * [ 0.8594 0      0      16]
+		 * [ 0      0      0       1]
+		 */
+		csc_ctl |= VC4_HD_CSC_CTL_ENABLE;
+		csc_ctl |= VC4_HD_CSC_CTL_RGB2YCC;
+		csc_ctl |= VC4_SET_FIELD(VC4_HD_CSC_CTL_MODE_CUSTOM,
+					 VC4_HD_CSC_CTL_MODE);
+
+		HDMI_WRITE(HDMI_CSC_12_11, (0x000 << 16) | 0x000);
+		HDMI_WRITE(HDMI_CSC_14_13, (0x100 << 16) | 0x6e0);
+		HDMI_WRITE(HDMI_CSC_22_21, (0x6e0 << 16) | 0x000);
+		HDMI_WRITE(HDMI_CSC_24_23, (0x100 << 16) | 0x000);
+		HDMI_WRITE(HDMI_CSC_32_31, (0x000 << 16) | 0x6e0);
+		HDMI_WRITE(HDMI_CSC_34_33, (0x100 << 16) | 0x000);
+	}
+
+	/* The RGB order applies even when CSC is disabled. */
+	HDMI_WRITE(HDMI_CSC_CTL, csc_ctl);
+}
+
+static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable)
+{
+	u32 csc_ctl;
+
+	csc_ctl = 0x07;	/* RGB_CONVERT_MODE = custom matrix, || USE_RGB_TO_YCBCR */
+
+	if (enable) {
+		/* CEA VICs other than #1 requre limited range RGB
+		 * output unless overridden by an AVI infoframe.
+		 * Apply a colorspace conversion to squash 0-255 down
+		 * to 16-235.  The matrix here is:
+		 *
+		 * [ 0.8594 0      0      16]
+		 * [ 0      0.8594 0      16]
+		 * [ 0      0      0.8594 16]
+		 * [ 0      0      0       1]
+		 * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
+		 */
+		HDMI_WRITE(HDMI_CSC_12_11, (0x0000 << 16) | 0x1b80);
+		HDMI_WRITE(HDMI_CSC_14_13, (0x0400 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_22_21, (0x1b80 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_24_23, (0x0400 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_32_31, (0x0000 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_34_33, (0x0400 << 16) | 0x1b80);
+	} else {
+		/* Still use the matrix for full range, but make it unity.
+		 * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
+		 */
+		HDMI_WRITE(HDMI_CSC_12_11, (0x0000 << 16) | 0x2000);
+		HDMI_WRITE(HDMI_CSC_14_13, (0x0000 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_22_21, (0x2000 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_24_23, (0x0000 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_32_31, (0x0000 << 16) | 0x0000);
+		HDMI_WRITE(HDMI_CSC_34_33, (0x0000 << 16) | 0x2000);
+	}
+
+	HDMI_WRITE(HDMI_CSC_CTL, csc_ctl);
+}
+
+static void vc4_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
+				 struct drm_display_mode *mode)
 {
-	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
-	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
-	struct drm_device *dev = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_hdmi *hdmi = vc4->hdmi;
-	bool debug_dump_regs = false;
 	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
 	bool vsync_pos = mode->flags & DRM_MODE_FLAG_PVSYNC;
 	bool interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE;
@@ -483,213 +488,285 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
 					mode->crtc_vsync_end -
 					interlaced,
 					VC4_HDMI_VERTB_VBP));
-	u32 csc_ctl;
+
+	HDMI_WRITE(HDMI_HORZA,
+		   (vsync_pos ? VC4_HDMI_HORZA_VPOS : 0) |
+		   (hsync_pos ? VC4_HDMI_HORZA_HPOS : 0) |
+		   VC4_SET_FIELD(mode->hdisplay * pixel_rep,
+				 VC4_HDMI_HORZA_HAP));
+
+	HDMI_WRITE(HDMI_HORZB,
+		   VC4_SET_FIELD((mode->htotal -
+				  mode->hsync_end) * pixel_rep,
+				 VC4_HDMI_HORZB_HBP) |
+		   VC4_SET_FIELD((mode->hsync_end -
+				  mode->hsync_start) * pixel_rep,
+				 VC4_HDMI_HORZB_HSP) |
+		   VC4_SET_FIELD((mode->hsync_start -
+				  mode->hdisplay) * pixel_rep,
+				 VC4_HDMI_HORZB_HFP));
+
+	HDMI_WRITE(HDMI_VERTA0, verta);
+	HDMI_WRITE(HDMI_VERTA1, verta);
+
+	HDMI_WRITE(HDMI_VERTB0, vertb_even);
+	HDMI_WRITE(HDMI_VERTB1, vertb);
+}
+static void vc5_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
+				 struct drm_display_mode *mode)
+{
+	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
+	bool vsync_pos = mode->flags & DRM_MODE_FLAG_PVSYNC;
+	bool interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE;
+	u32 pixel_rep = (mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1;
+	u32 verta = (VC4_SET_FIELD(mode->crtc_vsync_end - mode->crtc_vsync_start,
+				   VC5_HDMI_VERTA_VSP) |
+		     VC4_SET_FIELD(mode->crtc_vsync_start - mode->crtc_vdisplay,
+				   VC5_HDMI_VERTA_VFP) |
+		     VC4_SET_FIELD(mode->crtc_vdisplay, VC5_HDMI_VERTA_VAL));
+	u32 vertb = (VC4_SET_FIELD(0, VC5_HDMI_VERTB_VSPO) |
+		     VC4_SET_FIELD(mode->crtc_vtotal - mode->crtc_vsync_end,
+				   VC4_HDMI_VERTB_VBP));
+	u32 vertb_even = (VC4_SET_FIELD(0, VC5_HDMI_VERTB_VSPO) |
+			  VC4_SET_FIELD(mode->crtc_vtotal -
+					mode->crtc_vsync_end -
+					interlaced,
+					VC4_HDMI_VERTB_VBP));
+
+	HDMI_WRITE(HDMI_VEC_INTERFACE_XBAR, 0x354021);
+	HDMI_WRITE(HDMI_HORZA,
+		   (vsync_pos ? VC5_HDMI_HORZA_VPOS : 0) |
+		   (hsync_pos ? VC5_HDMI_HORZA_HPOS : 0) |
+		   VC4_SET_FIELD(mode->hdisplay * pixel_rep,
+				 VC5_HDMI_HORZA_HAP) |
+		   VC4_SET_FIELD((mode->hsync_start -
+				  mode->hdisplay) * pixel_rep,
+				 VC5_HDMI_HORZA_HFP));
+
+	HDMI_WRITE(HDMI_HORZB,
+		   VC4_SET_FIELD((mode->htotal -
+				  mode->hsync_end) * pixel_rep,
+				 VC5_HDMI_HORZB_HBP) |
+		   VC4_SET_FIELD((mode->hsync_end -
+				  mode->hsync_start) * pixel_rep,
+				 VC5_HDMI_HORZB_HSP));
+
+	HDMI_WRITE(HDMI_VERTA0, verta);
+	HDMI_WRITE(HDMI_VERTA1, verta);
+
+	HDMI_WRITE(HDMI_VERTB0, vertb_even);
+	HDMI_WRITE(HDMI_VERTB1, vertb);
+
+	HDMI_WRITE(HDMI_CLOCK_STOP, 0);
+}
+
+static void vc4_hdmi_recenter_fifo(struct vc4_hdmi *vc4_hdmi)
+{
+	u32 drift;
+	int ret;
+
+	drift = HDMI_READ(HDMI_FIFO_CTL);
+	drift &= VC4_HDMI_FIFO_VALID_WRITE_MASK;
+
+	HDMI_WRITE(HDMI_FIFO_CTL,
+		   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
+	HDMI_WRITE(HDMI_FIFO_CTL,
+		   drift | VC4_HDMI_FIFO_CTL_RECENTER);
+	usleep_range(1000, 1100);
+	HDMI_WRITE(HDMI_FIFO_CTL,
+		   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
+	HDMI_WRITE(HDMI_FIFO_CTL,
+		   drift | VC4_HDMI_FIFO_CTL_RECENTER);
+
+	ret = wait_for(HDMI_READ(HDMI_FIFO_CTL) &
+		       VC4_HDMI_FIFO_CTL_RECENTER_DONE, 1);
+	WARN_ONCE(ret, "Timeout waiting for "
+		  "VC4_HDMI_FIFO_CTL_RECENTER_DONE");
+}
+
+static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
+{
+	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
+	unsigned long pixel_rate, hsm_rate;
 	int ret;
 
-	ret = pm_runtime_get_sync(&hdmi->pdev->dev);
+	ret = pm_runtime_get_sync(&vc4_hdmi->pdev->dev);
 	if (ret < 0) {
 		DRM_ERROR("Failed to retain power domain: %d\n", ret);
 		return;
 	}
 
-	ret = clk_set_rate(hdmi->pixel_clock,
-			   mode->clock * 1000 *
-			   ((mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1));
+	pixel_rate = mode->clock * 1000 * ((mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1);
+	ret = clk_set_rate(vc4_hdmi->pixel_clock, pixel_rate);
 	if (ret) {
 		DRM_ERROR("Failed to set pixel clock rate: %d\n", ret);
 		return;
 	}
 
-	ret = clk_prepare_enable(hdmi->pixel_clock);
+	ret = clk_prepare_enable(vc4_hdmi->pixel_clock);
 	if (ret) {
 		DRM_ERROR("Failed to turn on pixel clock: %d\n", ret);
 		return;
 	}
 
-	HDMI_WRITE(VC4_HDMI_SW_RESET_CONTROL,
-		   VC4_HDMI_SW_RESET_HDMI |
-		   VC4_HDMI_SW_RESET_FORMAT_DETECT);
-
-	HDMI_WRITE(VC4_HDMI_SW_RESET_CONTROL, 0);
-
-	/* PHY should be in reset, like
-	 * vc4_hdmi_encoder_disable() does.
+	/*
+	 * As stated in RPi's vc4 firmware "HDMI state machine (HSM) clock must
+	 * be faster than pixel clock, infinitesimally faster, tested in
+	 * simulation. Otherwise, exact value is unimportant for HDMI
+	 * operation." This conflicts with bcm2835's vc4 documentation, which
+	 * states HSM's clock has to be at least 108% of the pixel clock.
+	 *
+	 * Real life tests reveal that vc4's firmware statement holds up, and
+	 * users are able to use pixel clocks closer to HSM's, namely for
+	 * 1920x1200@60Hz. So it was decided to have leave a 1% margin between
+	 * both clocks. Which, for RPi0-3 implies a maximum pixel clock of
+	 * 162MHz.
+	 *
+	 * Additionally, the AXI clock needs to be at least 25% of
+	 * pixel clock, but HSM ends up being the limiting factor.
 	 */
-	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0xf << 16);
+	hsm_rate = max_t(unsigned long, 120000000, (pixel_rate / 100) * 101);
+	ret = clk_set_min_rate(vc4_hdmi->hsm_clock, hsm_rate);
+	if (ret) {
+		DRM_ERROR("Failed to set HSM clock rate: %d\n", ret);
+		return;
+	}
 
-	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0);
+	ret = clk_prepare_enable(vc4_hdmi->hsm_clock);
+	if (ret) {
+		DRM_ERROR("Failed to turn on HSM clock: %d\n", ret);
+		clk_disable_unprepare(vc4_hdmi->pixel_clock);
+		return;
+	}
 
-	if (debug_dump_regs) {
-		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+	/*
+	 * FIXME: When the pixel freq is 594MHz (4k60), this needs to be setup
+	 * at 300MHz.
+	 */
+	ret = clk_set_min_rate(vc4_hdmi->pixel_bvb_clock,
+			       (hsm_rate > VC4_HSM_MID_CLOCK ? 150000000 : 75000000));
+	if (ret) {
+		DRM_ERROR("Failed to set pixel bvb clock rate: %d\n", ret);
+		clk_disable_unprepare(vc4_hdmi->hsm_clock);
+		clk_disable_unprepare(vc4_hdmi->pixel_clock);
+		return;
+	}
 
-		dev_info(&hdmi->pdev->dev, "HDMI regs before:\n");
-		drm_print_regset32(&p, &hdmi->hdmi_regset);
-		drm_print_regset32(&p, &hdmi->hd_regset);
+	ret = clk_prepare_enable(vc4_hdmi->pixel_bvb_clock);
+	if (ret) {
+		DRM_ERROR("Failed to turn on pixel bvb clock: %d\n", ret);
+		clk_disable_unprepare(vc4_hdmi->hsm_clock);
+		clk_disable_unprepare(vc4_hdmi->pixel_clock);
+		return;
 	}
 
-	HD_WRITE(VC4_HD_VID_CTL, 0);
+	if (vc4_hdmi->variant->reset)
+		vc4_hdmi->variant->reset(vc4_hdmi);
 
-	HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
-		   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+	if (vc4_hdmi->variant->phy_init)
+		vc4_hdmi->variant->phy_init(vc4_hdmi, mode);
+
+	HDMI_WRITE(HDMI_SCHEDULER_CONTROL,
+		   HDMI_READ(HDMI_SCHEDULER_CONTROL) |
 		   VC4_HDMI_SCHEDULER_CONTROL_MANUAL_FORMAT |
 		   VC4_HDMI_SCHEDULER_CONTROL_IGNORE_VSYNC_PREDICTS);
 
-	HDMI_WRITE(VC4_HDMI_HORZA,
-		   (vsync_pos ? VC4_HDMI_HORZA_VPOS : 0) |
-		   (hsync_pos ? VC4_HDMI_HORZA_HPOS : 0) |
-		   VC4_SET_FIELD(mode->hdisplay * pixel_rep,
-				 VC4_HDMI_HORZA_HAP));
-
-	HDMI_WRITE(VC4_HDMI_HORZB,
-		   VC4_SET_FIELD((mode->htotal -
-				  mode->hsync_end) * pixel_rep,
-				 VC4_HDMI_HORZB_HBP) |
-		   VC4_SET_FIELD((mode->hsync_end -
-				  mode->hsync_start) * pixel_rep,
-				 VC4_HDMI_HORZB_HSP) |
-		   VC4_SET_FIELD((mode->hsync_start -
-				  mode->hdisplay) * pixel_rep,
-				 VC4_HDMI_HORZB_HFP));
-
-	HDMI_WRITE(VC4_HDMI_VERTA0, verta);
-	HDMI_WRITE(VC4_HDMI_VERTA1, verta);
-
-	HDMI_WRITE(VC4_HDMI_VERTB0, vertb_even);
-	HDMI_WRITE(VC4_HDMI_VERTB1, vertb);
-
-	HD_WRITE(VC4_HD_VID_CTL,
-		 (vsync_pos ? 0 : VC4_HD_VID_CTL_VSYNC_LOW) |
-		 (hsync_pos ? 0 : VC4_HD_VID_CTL_HSYNC_LOW));
+	if (vc4_hdmi->variant->set_timings)
+		vc4_hdmi->variant->set_timings(vc4_hdmi, mode);
+}
 
-	csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR,
-				VC4_HD_CSC_CTL_ORDER);
+static void vc4_hdmi_encoder_pre_crtc_enable(struct drm_encoder *encoder)
+{
+	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
+	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 
 	if (vc4_encoder->hdmi_monitor &&
-	    drm_default_rgb_quant_range(mode) ==
-	    HDMI_QUANTIZATION_RANGE_LIMITED) {
-		/* CEA VICs other than #1 requre limited range RGB
-		 * output unless overridden by an AVI infoframe.
-		 * Apply a colorspace conversion to squash 0-255 down
-		 * to 16-235.  The matrix here is:
-		 *
-		 * [ 0      0      0.8594 16]
-		 * [ 0      0.8594 0      16]
-		 * [ 0.8594 0      0      16]
-		 * [ 0      0      0       1]
-		 */
-		csc_ctl |= VC4_HD_CSC_CTL_ENABLE;
-		csc_ctl |= VC4_HD_CSC_CTL_RGB2YCC;
-		csc_ctl |= VC4_SET_FIELD(VC4_HD_CSC_CTL_MODE_CUSTOM,
-					 VC4_HD_CSC_CTL_MODE);
+	    drm_default_rgb_quant_range(mode) == HDMI_QUANTIZATION_RANGE_LIMITED) {
+		if (vc4_hdmi->variant->csc_setup)
+			vc4_hdmi->variant->csc_setup(vc4_hdmi, true);
 
-		HD_WRITE(VC4_HD_CSC_12_11, (0x000 << 16) | 0x000);
-		HD_WRITE(VC4_HD_CSC_14_13, (0x100 << 16) | 0x6e0);
-		HD_WRITE(VC4_HD_CSC_22_21, (0x6e0 << 16) | 0x000);
-		HD_WRITE(VC4_HD_CSC_24_23, (0x100 << 16) | 0x000);
-		HD_WRITE(VC4_HD_CSC_32_31, (0x000 << 16) | 0x6e0);
-		HD_WRITE(VC4_HD_CSC_34_33, (0x100 << 16) | 0x000);
 		vc4_encoder->limited_rgb_range = true;
 	} else {
+		if (vc4_hdmi->variant->csc_setup)
+			vc4_hdmi->variant->csc_setup(vc4_hdmi, false);
+
 		vc4_encoder->limited_rgb_range = false;
 	}
 
-	/* The RGB order applies even when CSC is disabled. */
-	HD_WRITE(VC4_HD_CSC_CTL, csc_ctl);
-
-	HDMI_WRITE(VC4_HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
+	HDMI_WRITE(HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
+}
 
-	if (debug_dump_regs) {
-		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+static void vc4_hdmi_encoder_post_crtc_enable(struct drm_encoder *encoder)
+{
+	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
+	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
+	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
+	bool vsync_pos = mode->flags & DRM_MODE_FLAG_PVSYNC;
+	int ret;
 
-		dev_info(&hdmi->pdev->dev, "HDMI regs after:\n");
-		drm_print_regset32(&p, &hdmi->hdmi_regset);
-		drm_print_regset32(&p, &hdmi->hd_regset);
-	}
+	HDMI_WRITE(HDMI_VID_CTL,
+		   VC4_HD_VID_CTL_ENABLE |
+		   VC4_HD_VID_CTL_UNDERFLOW_ENABLE |
+		   VC4_HD_VID_CTL_FRAME_COUNTER_RESET |
+		   (vsync_pos ? 0 : VC4_HD_VID_CTL_VSYNC_LOW) |
+		   (hsync_pos ? 0 : VC4_HD_VID_CTL_HSYNC_LOW));
 
-	HD_WRITE(VC4_HD_VID_CTL,
-		 HD_READ(VC4_HD_VID_CTL) |
-		 VC4_HD_VID_CTL_ENABLE |
-		 VC4_HD_VID_CTL_UNDERFLOW_ENABLE |
-		 VC4_HD_VID_CTL_FRAME_COUNTER_RESET);
+	HDMI_WRITE(HDMI_VID_CTL,
+		   HDMI_READ(HDMI_VID_CTL) & ~VC4_HD_VID_CTL_BLANKPIX);
 
 	if (vc4_encoder->hdmi_monitor) {
-		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
-			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+		HDMI_WRITE(HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(HDMI_SCHEDULER_CONTROL) |
 			   VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI);
 
-		ret = wait_for(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+		ret = wait_for(HDMI_READ(HDMI_SCHEDULER_CONTROL) &
 			       VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE, 1000);
 		WARN_ONCE(ret, "Timeout waiting for "
 			  "VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n");
 	} else {
-		HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG,
-			   HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) &
+		HDMI_WRITE(HDMI_RAM_PACKET_CONFIG,
+			   HDMI_READ(HDMI_RAM_PACKET_CONFIG) &
 			   ~(VC4_HDMI_RAM_PACKET_ENABLE));
-		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
-			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+		HDMI_WRITE(HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(HDMI_SCHEDULER_CONTROL) &
 			   ~VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI);
 
-		ret = wait_for(!(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+		ret = wait_for(!(HDMI_READ(HDMI_SCHEDULER_CONTROL) &
 				 VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE), 1000);
 		WARN_ONCE(ret, "Timeout waiting for "
 			  "!VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n");
 	}
 
 	if (vc4_encoder->hdmi_monitor) {
-		u32 drift;
-
-		WARN_ON(!(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+		WARN_ON(!(HDMI_READ(HDMI_SCHEDULER_CONTROL) &
 			  VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE));
-		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
-			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+		HDMI_WRITE(HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(HDMI_SCHEDULER_CONTROL) |
 			   VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT);
 
-		HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG,
+		HDMI_WRITE(HDMI_RAM_PACKET_CONFIG,
 			   VC4_HDMI_RAM_PACKET_ENABLE);
 
 		vc4_hdmi_set_infoframes(encoder);
-
-		drift = HDMI_READ(VC4_HDMI_FIFO_CTL);
-		drift &= VC4_HDMI_FIFO_VALID_WRITE_MASK;
-
-		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
-			   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
-		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
-			   drift | VC4_HDMI_FIFO_CTL_RECENTER);
-		usleep_range(1000, 1100);
-		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
-			   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
-		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
-			   drift | VC4_HDMI_FIFO_CTL_RECENTER);
-
-		ret = wait_for(HDMI_READ(VC4_HDMI_FIFO_CTL) &
-			       VC4_HDMI_FIFO_CTL_RECENTER_DONE, 1);
-		WARN_ONCE(ret, "Timeout waiting for "
-			  "VC4_HDMI_FIFO_CTL_RECENTER_DONE");
 	}
+
+	vc4_hdmi_recenter_fifo(vc4_hdmi);
+}
+
+static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
+{
 }
 
 static enum drm_mode_status
-vc4_hdmi_encoder_mode_valid(struct drm_encoder *crtc,
+vc4_hdmi_encoder_mode_valid(struct drm_encoder *encoder,
 			    const struct drm_display_mode *mode)
 {
-	/*
-	 * As stated in RPi's vc4 firmware "HDMI state machine (HSM) clock must
-	 * be faster than pixel clock, infinitesimally faster, tested in
-	 * simulation. Otherwise, exact value is unimportant for HDMI
-	 * operation." This conflicts with bcm2835's vc4 documentation, which
-	 * states HSM's clock has to be at least 108% of the pixel clock.
-	 *
-	 * Real life tests reveal that vc4's firmware statement holds up, and
-	 * users are able to use pixel clocks closer to HSM's, namely for
-	 * 1920x1200@60Hz. So it was decided to have leave a 1% margin between
-	 * both clocks. Which, for RPi0-3 implies a maximum pixel clock of
-	 * 162MHz.
-	 *
-	 * Additionally, the AXI clock needs to be at least 25% of
-	 * pixel clock, but HSM ends up being the limiting factor.
-	 */
-	if (mode->clock > HSM_CLOCK_FREQ / (1000 * 101 / 100))
+	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
+
+	if ((mode->clock * 1000) > vc4_hdmi->variant->max_pixel_clock)
 		return MODE_CLOCK_HIGH;
 
 	return MODE_OK;
@@ -701,34 +778,54 @@ static const struct drm_encoder_helper_funcs vc4_hdmi_encoder_helper_funcs = {
 	.enable = vc4_hdmi_encoder_enable,
 };
 
+static u32 vc4_hdmi_channel_map(struct vc4_hdmi *vc4_hdmi, u32 channel_mask)
+{
+	int i;
+	u32 channel_map = 0;
+
+	for (i = 0; i < 8; i++) {
+		if (channel_mask & BIT(i))
+			channel_map |= i << (3 * i);
+	}
+	return channel_map;
+}
+
+static u32 vc5_hdmi_channel_map(struct vc4_hdmi *vc4_hdmi, u32 channel_mask)
+{
+	int i;
+	u32 channel_map = 0;
+
+	for (i = 0; i < 8; i++) {
+		if (channel_mask & BIT(i))
+			channel_map |= i << (4 * i);
+	}
+	return channel_map;
+}
+
 /* HDMI audio codec callbacks */
-static void vc4_hdmi_audio_set_mai_clock(struct vc4_hdmi *hdmi)
+static void vc4_hdmi_audio_set_mai_clock(struct vc4_hdmi *vc4_hdmi)
 {
-	struct drm_device *drm = hdmi->encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
-	u32 hsm_clock = clk_get_rate(hdmi->hsm_clock);
+	u32 hsm_clock = clk_get_rate(vc4_hdmi->audio_clock);
 	unsigned long n, m;
 
-	rational_best_approximation(hsm_clock, hdmi->audio.samplerate,
+	rational_best_approximation(hsm_clock, vc4_hdmi->audio.samplerate,
 				    VC4_HD_MAI_SMP_N_MASK >>
 				    VC4_HD_MAI_SMP_N_SHIFT,
 				    (VC4_HD_MAI_SMP_M_MASK >>
 				     VC4_HD_MAI_SMP_M_SHIFT) + 1,
 				    &n, &m);
 
-	HD_WRITE(VC4_HD_MAI_SMP,
-		 VC4_SET_FIELD(n, VC4_HD_MAI_SMP_N) |
-		 VC4_SET_FIELD(m - 1, VC4_HD_MAI_SMP_M));
+	HDMI_WRITE(HDMI_MAI_SMP,
+		   VC4_SET_FIELD(n, VC4_HD_MAI_SMP_N) |
+		   VC4_SET_FIELD(m - 1, VC4_HD_MAI_SMP_M));
 }
 
-static void vc4_hdmi_set_n_cts(struct vc4_hdmi *hdmi)
+static void vc4_hdmi_set_n_cts(struct vc4_hdmi *vc4_hdmi)
 {
-	struct drm_encoder *encoder = hdmi->encoder;
+	struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base;
 	struct drm_crtc *crtc = encoder->crtc;
-	struct drm_device *drm = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
 	const struct drm_display_mode *mode = &crtc->state->adjusted_mode;
-	u32 samplerate = hdmi->audio.samplerate;
+	u32 samplerate = vc4_hdmi->audio.samplerate;
 	u32 n, cts;
 	u64 tmp;
 
@@ -737,7 +834,7 @@ static void vc4_hdmi_set_n_cts(struct vc4_hdmi *hdmi)
 	do_div(tmp, 128 * samplerate);
 	cts = tmp;
 
-	HDMI_WRITE(VC4_HDMI_CRP_CFG,
+	HDMI_WRITE(HDMI_CRP_CFG,
 		   VC4_HDMI_CRP_CFG_EXTERNAL_CTS_EN |
 		   VC4_SET_FIELD(n, VC4_HDMI_CRP_CFG_N));
 
@@ -746,8 +843,8 @@ static void vc4_hdmi_set_n_cts(struct vc4_hdmi *hdmi)
 	 * providing a CTS_1 value.  The two CTS values are alternated
 	 * between based on the period fields
 	 */
-	HDMI_WRITE(VC4_HDMI_CTS_0, cts);
-	HDMI_WRITE(VC4_HDMI_CTS_1, cts);
+	HDMI_WRITE(HDMI_CTS_0, cts);
+	HDMI_WRITE(HDMI_CTS_1, cts);
 }
 
 static inline struct vc4_hdmi *dai_to_hdmi(struct snd_soc_dai *dai)
@@ -760,26 +857,25 @@ static inline struct vc4_hdmi *dai_to_hdmi(struct snd_soc_dai *dai)
 static int vc4_hdmi_audio_startup(struct snd_pcm_substream *substream,
 				  struct snd_soc_dai *dai)
 {
-	struct vc4_hdmi *hdmi = dai_to_hdmi(dai);
-	struct drm_encoder *encoder = hdmi->encoder;
-	struct vc4_dev *vc4 = to_vc4_dev(encoder->dev);
+	struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai);
+	struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base;
+	struct drm_connector *connector = &vc4_hdmi->connector;
 	int ret;
 
-	if (hdmi->audio.substream && hdmi->audio.substream != substream)
+	if (vc4_hdmi->audio.substream && vc4_hdmi->audio.substream != substream)
 		return -EINVAL;
 
-	hdmi->audio.substream = substream;
+	vc4_hdmi->audio.substream = substream;
 
 	/*
 	 * If the HDMI encoder hasn't probed, or the encoder is
 	 * currently in DVI mode, treat the codec dai as missing.
 	 */
-	if (!encoder->crtc || !(HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) &
+	if (!encoder->crtc || !(HDMI_READ(HDMI_RAM_PACKET_CONFIG) &
 				VC4_HDMI_RAM_PACKET_ENABLE))
 		return -ENODEV;
 
-	ret = snd_pcm_hw_constraint_eld(substream->runtime,
-					hdmi->connector->eld);
+	ret = snd_pcm_hw_constraint_eld(substream->runtime, connector->eld);
 	if (ret)
 		return ret;
 
@@ -791,34 +887,33 @@ static int vc4_hdmi_audio_set_fmt(struct snd_soc_dai *dai, unsigned int fmt)
 	return 0;
 }
 
-static void vc4_hdmi_audio_reset(struct vc4_hdmi *hdmi)
+static void vc4_hdmi_audio_reset(struct vc4_hdmi *vc4_hdmi)
 {
-	struct drm_encoder *encoder = hdmi->encoder;
-	struct drm_device *drm = encoder->dev;
-	struct device *dev = &hdmi->pdev->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
+	struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base;
+	struct device *dev = &vc4_hdmi->pdev->dev;
 	int ret;
 
+	vc4_hdmi->audio.streaming = false;
 	ret = vc4_hdmi_stop_packet(encoder, HDMI_INFOFRAME_TYPE_AUDIO);
 	if (ret)
 		dev_err(dev, "Failed to stop audio infoframe: %d\n", ret);
 
-	HD_WRITE(VC4_HD_MAI_CTL, VC4_HD_MAI_CTL_RESET);
-	HD_WRITE(VC4_HD_MAI_CTL, VC4_HD_MAI_CTL_ERRORF);
-	HD_WRITE(VC4_HD_MAI_CTL, VC4_HD_MAI_CTL_FLUSH);
+	HDMI_WRITE(HDMI_MAI_CTL, VC4_HD_MAI_CTL_RESET);
+	HDMI_WRITE(HDMI_MAI_CTL, VC4_HD_MAI_CTL_ERRORF);
+	HDMI_WRITE(HDMI_MAI_CTL, VC4_HD_MAI_CTL_FLUSH);
 }
 
 static void vc4_hdmi_audio_shutdown(struct snd_pcm_substream *substream,
 				    struct snd_soc_dai *dai)
 {
-	struct vc4_hdmi *hdmi = dai_to_hdmi(dai);
+	struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai);
 
-	if (substream != hdmi->audio.substream)
+	if (substream != vc4_hdmi->audio.substream)
 		return;
 
-	vc4_hdmi_audio_reset(hdmi);
+	vc4_hdmi_audio_reset(vc4_hdmi);
 
-	hdmi->audio.substream = NULL;
+	vc4_hdmi->audio.substream = NULL;
 }
 
 /* HDMI audio codec callbacks */
@@ -826,72 +921,65 @@ static int vc4_hdmi_audio_hw_params(struct snd_pcm_substream *substream,
 				    struct snd_pcm_hw_params *params,
 				    struct snd_soc_dai *dai)
 {
-	struct vc4_hdmi *hdmi = dai_to_hdmi(dai);
-	struct drm_encoder *encoder = hdmi->encoder;
-	struct drm_device *drm = encoder->dev;
-	struct device *dev = &hdmi->pdev->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
+	struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai);
+	struct device *dev = &vc4_hdmi->pdev->dev;
 	u32 audio_packet_config, channel_mask;
-	u32 channel_map, i;
+	u32 channel_map;
 
-	if (substream != hdmi->audio.substream)
+	if (substream != vc4_hdmi->audio.substream)
 		return -EINVAL;
 
 	dev_dbg(dev, "%s: %u Hz, %d bit, %d channels\n", __func__,
 		params_rate(params), params_width(params),
 		params_channels(params));
 
-	hdmi->audio.channels = params_channels(params);
-	hdmi->audio.samplerate = params_rate(params);
+	vc4_hdmi->audio.channels = params_channels(params);
+	vc4_hdmi->audio.samplerate = params_rate(params);
 
-	HD_WRITE(VC4_HD_MAI_CTL,
-		 VC4_HD_MAI_CTL_RESET |
-		 VC4_HD_MAI_CTL_FLUSH |
-		 VC4_HD_MAI_CTL_DLATE |
-		 VC4_HD_MAI_CTL_ERRORE |
-		 VC4_HD_MAI_CTL_ERRORF);
+	HDMI_WRITE(HDMI_MAI_CTL,
+		   VC4_HD_MAI_CTL_RESET |
+		   VC4_HD_MAI_CTL_FLUSH |
+		   VC4_HD_MAI_CTL_DLATE |
+		   VC4_HD_MAI_CTL_ERRORE |
+		   VC4_HD_MAI_CTL_ERRORF);
 
-	vc4_hdmi_audio_set_mai_clock(hdmi);
+	vc4_hdmi_audio_set_mai_clock(vc4_hdmi);
 
+	/* The B frame identifier should match the value used by alsa-lib (8) */
 	audio_packet_config =
 		VC4_HDMI_AUDIO_PACKET_ZERO_DATA_ON_SAMPLE_FLAT |
 		VC4_HDMI_AUDIO_PACKET_ZERO_DATA_ON_INACTIVE_CHANNELS |
-		VC4_SET_FIELD(0xf, VC4_HDMI_AUDIO_PACKET_B_FRAME_IDENTIFIER);
+		VC4_SET_FIELD(0x8, VC4_HDMI_AUDIO_PACKET_B_FRAME_IDENTIFIER);
 
-	channel_mask = GENMASK(hdmi->audio.channels - 1, 0);
+	channel_mask = GENMASK(vc4_hdmi->audio.channels - 1, 0);
 	audio_packet_config |= VC4_SET_FIELD(channel_mask,
 					     VC4_HDMI_AUDIO_PACKET_CEA_MASK);
 
 	/* Set the MAI threshold.  This logic mimics the firmware's. */
-	if (hdmi->audio.samplerate > 96000) {
-		HD_WRITE(VC4_HD_MAI_THR,
-			 VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQHIGH) |
-			 VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQLOW));
-	} else if (hdmi->audio.samplerate > 48000) {
-		HD_WRITE(VC4_HD_MAI_THR,
-			 VC4_SET_FIELD(0x14, VC4_HD_MAI_THR_DREQHIGH) |
-			 VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQLOW));
+	if (vc4_hdmi->audio.samplerate > 96000) {
+		HDMI_WRITE(HDMI_MAI_THR,
+			   VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQHIGH) |
+			   VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQLOW));
+	} else if (vc4_hdmi->audio.samplerate > 48000) {
+		HDMI_WRITE(HDMI_MAI_THR,
+			   VC4_SET_FIELD(0x14, VC4_HD_MAI_THR_DREQHIGH) |
+			   VC4_SET_FIELD(0x12, VC4_HD_MAI_THR_DREQLOW));
 	} else {
-		HD_WRITE(VC4_HD_MAI_THR,
-			 VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_PANICHIGH) |
-			 VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_PANICLOW) |
-			 VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_DREQHIGH) |
-			 VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_DREQLOW));
+		HDMI_WRITE(HDMI_MAI_THR,
+			   VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_PANICHIGH) |
+			   VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_PANICLOW) |
+			   VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_DREQHIGH) |
+			   VC4_SET_FIELD(0x10, VC4_HD_MAI_THR_DREQLOW));
 	}
 
-	HDMI_WRITE(VC4_HDMI_MAI_CONFIG,
+	HDMI_WRITE(HDMI_MAI_CONFIG,
 		   VC4_HDMI_MAI_CONFIG_BIT_REVERSE |
 		   VC4_SET_FIELD(channel_mask, VC4_HDMI_MAI_CHANNEL_MASK));
 
-	channel_map = 0;
-	for (i = 0; i < 8; i++) {
-		if (channel_mask & BIT(i))
-			channel_map |= i << (3 * i);
-	}
-
-	HDMI_WRITE(VC4_HDMI_MAI_CHANNEL_MAP, channel_map);
-	HDMI_WRITE(VC4_HDMI_AUDIO_PACKET_CONFIG, audio_packet_config);
-	vc4_hdmi_set_n_cts(hdmi);
+	channel_map = vc4_hdmi->variant->channel_map(vc4_hdmi, channel_mask);
+	HDMI_WRITE(HDMI_MAI_CHANNEL_MAP, channel_map);
+	HDMI_WRITE(HDMI_AUDIO_PACKET_CONFIG, audio_packet_config);
+	vc4_hdmi_set_n_cts(vc4_hdmi);
 
 	return 0;
 }
@@ -899,30 +987,33 @@ static int vc4_hdmi_audio_hw_params(struct snd_pcm_substream *substream,
 static int vc4_hdmi_audio_trigger(struct snd_pcm_substream *substream, int cmd,
 				  struct snd_soc_dai *dai)
 {
-	struct vc4_hdmi *hdmi = dai_to_hdmi(dai);
-	struct drm_encoder *encoder = hdmi->encoder;
-	struct drm_device *drm = encoder->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
+	struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai);
+	struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base;
 
 	switch (cmd) {
 	case SNDRV_PCM_TRIGGER_START:
 		vc4_hdmi_set_audio_infoframe(encoder);
-		HDMI_WRITE(VC4_HDMI_TX_PHY_CTL0,
-			   HDMI_READ(VC4_HDMI_TX_PHY_CTL0) &
-			   ~VC4_HDMI_TX_PHY_RNG_PWRDN);
-		HD_WRITE(VC4_HD_MAI_CTL,
-			 VC4_SET_FIELD(hdmi->audio.channels,
-				       VC4_HD_MAI_CTL_CHNUM) |
-			 VC4_HD_MAI_CTL_ENABLE);
+		vc4_hdmi->audio.streaming = true;
+
+		if (vc4_hdmi->variant->phy_rng_enable)
+			vc4_hdmi->variant->phy_rng_enable(vc4_hdmi);
+
+		HDMI_WRITE(HDMI_MAI_CTL,
+			   VC4_SET_FIELD(vc4_hdmi->audio.channels,
+					 VC4_HD_MAI_CTL_CHNUM) |
+			   VC4_HD_MAI_CTL_ENABLE);
 		break;
 	case SNDRV_PCM_TRIGGER_STOP:
-		HD_WRITE(VC4_HD_MAI_CTL,
-			 VC4_HD_MAI_CTL_DLATE |
-			 VC4_HD_MAI_CTL_ERRORE |
-			 VC4_HD_MAI_CTL_ERRORF);
-		HDMI_WRITE(VC4_HDMI_TX_PHY_CTL0,
-			   HDMI_READ(VC4_HDMI_TX_PHY_CTL0) |
-			   VC4_HDMI_TX_PHY_RNG_PWRDN);
+		HDMI_WRITE(HDMI_MAI_CTL,
+			   VC4_HD_MAI_CTL_DLATE |
+			   VC4_HD_MAI_CTL_ERRORE |
+			   VC4_HD_MAI_CTL_ERRORF);
+
+		if (vc4_hdmi->variant->phy_rng_disable)
+			vc4_hdmi->variant->phy_rng_disable(vc4_hdmi);
+
+		vc4_hdmi->audio.streaming = false;
+
 		break;
 	default:
 		break;
@@ -943,10 +1034,11 @@ static int vc4_hdmi_audio_eld_ctl_info(struct snd_kcontrol *kcontrol,
 				       struct snd_ctl_elem_info *uinfo)
 {
 	struct snd_soc_component *component = snd_kcontrol_chip(kcontrol);
-	struct vc4_hdmi *hdmi = snd_component_to_hdmi(component);
+	struct vc4_hdmi *vc4_hdmi = snd_component_to_hdmi(component);
+	struct drm_connector *connector = &vc4_hdmi->connector;
 
 	uinfo->type = SNDRV_CTL_ELEM_TYPE_BYTES;
-	uinfo->count = sizeof(hdmi->connector->eld);
+	uinfo->count = sizeof(connector->eld);
 
 	return 0;
 }
@@ -955,10 +1047,11 @@ static int vc4_hdmi_audio_eld_ctl_get(struct snd_kcontrol *kcontrol,
 				      struct snd_ctl_elem_value *ucontrol)
 {
 	struct snd_soc_component *component = snd_kcontrol_chip(kcontrol);
-	struct vc4_hdmi *hdmi = snd_component_to_hdmi(component);
+	struct vc4_hdmi *vc4_hdmi = snd_component_to_hdmi(component);
+	struct drm_connector *connector = &vc4_hdmi->connector;
 
-	memcpy(ucontrol->value.bytes.data, hdmi->connector->eld,
-	       sizeof(hdmi->connector->eld));
+	memcpy(ucontrol->value.bytes.data, connector->eld,
+	       sizeof(connector->eld));
 
 	return 0;
 }
@@ -1023,9 +1116,9 @@ static const struct snd_soc_component_driver vc4_hdmi_audio_cpu_dai_comp = {
 
 static int vc4_hdmi_audio_cpu_dai_probe(struct snd_soc_dai *dai)
 {
-	struct vc4_hdmi *hdmi = dai_to_hdmi(dai);
+	struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai);
 
-	snd_soc_dai_init_dma_data(dai, &hdmi->audio.dma_data, NULL);
+	snd_soc_dai_init_dma_data(dai, &vc4_hdmi->audio.dma_data, NULL);
 
 	return 0;
 }
@@ -1051,12 +1144,15 @@ static const struct snd_dmaengine_pcm_config pcm_conf = {
 	.prepare_slave_config = snd_dmaengine_pcm_prepare_slave_config,
 };
 
-static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
+static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 {
-	struct snd_soc_dai_link *dai_link = &hdmi->audio.link;
-	struct snd_soc_card *card = &hdmi->audio.card;
-	struct device *dev = &hdmi->pdev->dev;
+	const struct vc4_hdmi_register *mai_data =
+		&vc4_hdmi->variant->registers[HDMI_MAI_DATA];
+	struct snd_soc_dai_link *dai_link = &vc4_hdmi->audio.link;
+	struct snd_soc_card *card = &vc4_hdmi->audio.card;
+	struct device *dev = &vc4_hdmi->pdev->dev;
 	const __be32 *addr;
+	int index;
 	int ret;
 
 	if (!of_find_property(dev->of_node, "dmas", NULL)) {
@@ -1065,6 +1161,11 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 		return 0;
 	}
 
+	if (mai_data->reg != VC4_HD) {
+		WARN_ONCE(true, "MAI isn't in the HD block\n");
+		return -EINVAL;
+	}
+
 	/*
 	 * Get the physical address of VC4_HD_MAI_DATA. We need to retrieve
 	 * the bus address specified in the DT, because the physical address
@@ -1072,10 +1173,16 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 	 * for DMA transfers.
 	 * This VC/MMU should probably be exposed to avoid this kind of hacks.
 	 */
-	addr = of_get_address(dev->of_node, 1, NULL, NULL);
-	hdmi->audio.dma_data.addr = be32_to_cpup(addr) + VC4_HD_MAI_DATA;
-	hdmi->audio.dma_data.addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
-	hdmi->audio.dma_data.maxburst = 2;
+	index = of_property_match_string(dev->of_node, "reg-names", "hd");
+	/* Before BCM2711, we don't have a named register range */
+	if (index < 0)
+		index = 1;
+
+	addr = of_get_address(dev->of_node, index, NULL, NULL);
+
+	vc4_hdmi->audio.dma_data.addr = be32_to_cpup(addr) + mai_data->offset;
+	vc4_hdmi->audio.dma_data.addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	vc4_hdmi->audio.dma_data.maxburst = 2;
 
 	ret = devm_snd_dmaengine_pcm_register(dev, &pcm_conf, 0);
 	if (ret) {
@@ -1098,9 +1205,9 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 		return ret;
 	}
 
-	dai_link->cpus		= &hdmi->audio.cpu;
-	dai_link->codecs	= &hdmi->audio.codec;
-	dai_link->platforms	= &hdmi->audio.platform;
+	dai_link->cpus		= &vc4_hdmi->audio.cpu;
+	dai_link->codecs	= &vc4_hdmi->audio.codec;
+	dai_link->platforms	= &vc4_hdmi->audio.platform;
 
 	dai_link->num_cpus	= 1;
 	dai_link->num_codecs	= 1;
@@ -1115,7 +1222,7 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 
 	card->dai_link = dai_link;
 	card->num_links = 1;
-	card->name = "vc4-hdmi";
+	card->name = vc4_hdmi->variant->card_name;
 	card->dev = dev;
 	card->owner = THIS_MODULE;
 
@@ -1126,7 +1233,7 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 	 * now stored in card->drvdata and should be retrieved with
 	 * snd_soc_card_get_drvdata() if needed.
 	 */
-	snd_soc_card_set_drvdata(card, hdmi);
+	snd_soc_card_set_drvdata(card, vc4_hdmi);
 	ret = devm_snd_soc_register_card(dev, card);
 	if (ret)
 		dev_err(dev, "Could not register sound card: %d\n", ret);
@@ -1138,35 +1245,35 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *hdmi)
 #ifdef CONFIG_DRM_VC4_HDMI_CEC
 static irqreturn_t vc4_cec_irq_handler_thread(int irq, void *priv)
 {
-	struct vc4_dev *vc4 = priv;
-	struct vc4_hdmi *hdmi = vc4->hdmi;
-
-	if (hdmi->cec_irq_was_rx) {
-		if (hdmi->cec_rx_msg.len)
-			cec_received_msg(hdmi->cec_adap, &hdmi->cec_rx_msg);
-	} else if (hdmi->cec_tx_ok) {
-		cec_transmit_done(hdmi->cec_adap, CEC_TX_STATUS_OK,
+	struct vc4_hdmi *vc4_hdmi = priv;
+
+	if (vc4_hdmi->cec_irq_was_rx) {
+		if (vc4_hdmi->cec_rx_msg.len)
+			cec_received_msg(vc4_hdmi->cec_adap,
+					 &vc4_hdmi->cec_rx_msg);
+	} else if (vc4_hdmi->cec_tx_ok) {
+		cec_transmit_done(vc4_hdmi->cec_adap, CEC_TX_STATUS_OK,
 				  0, 0, 0, 0);
 	} else {
 		/*
 		 * This CEC implementation makes 1 retry, so if we
 		 * get a NACK, then that means it made 2 attempts.
 		 */
-		cec_transmit_done(hdmi->cec_adap, CEC_TX_STATUS_NACK,
+		cec_transmit_done(vc4_hdmi->cec_adap, CEC_TX_STATUS_NACK,
 				  0, 2, 0, 0);
 	}
 	return IRQ_HANDLED;
 }
 
-static void vc4_cec_read_msg(struct vc4_dev *vc4, u32 cntrl1)
+static void vc4_cec_read_msg(struct vc4_hdmi *vc4_hdmi, u32 cntrl1)
 {
-	struct cec_msg *msg = &vc4->hdmi->cec_rx_msg;
+	struct cec_msg *msg = &vc4_hdmi->cec_rx_msg;
 	unsigned int i;
 
 	msg->len = 1 + ((cntrl1 & VC4_HDMI_CEC_REC_WRD_CNT_MASK) >>
 					VC4_HDMI_CEC_REC_WRD_CNT_SHIFT);
 	for (i = 0; i < msg->len; i += 4) {
-		u32 val = HDMI_READ(VC4_HDMI_CEC_RX_DATA_1 + i);
+		u32 val = HDMI_READ(HDMI_CEC_RX_DATA_1 + i);
 
 		msg->msg[i] = val & 0xff;
 		msg->msg[i + 1] = (val >> 8) & 0xff;
@@ -1177,38 +1284,37 @@ static void vc4_cec_read_msg(struct vc4_dev *vc4, u32 cntrl1)
 
 static irqreturn_t vc4_cec_irq_handler(int irq, void *priv)
 {
-	struct vc4_dev *vc4 = priv;
-	struct vc4_hdmi *hdmi = vc4->hdmi;
-	u32 stat = HDMI_READ(VC4_HDMI_CPU_STATUS);
+	struct vc4_hdmi *vc4_hdmi = priv;
+	u32 stat = HDMI_READ(HDMI_CEC_CPU_STATUS);
 	u32 cntrl1, cntrl5;
 
 	if (!(stat & VC4_HDMI_CPU_CEC))
 		return IRQ_NONE;
-	hdmi->cec_rx_msg.len = 0;
-	cntrl1 = HDMI_READ(VC4_HDMI_CEC_CNTRL_1);
-	cntrl5 = HDMI_READ(VC4_HDMI_CEC_CNTRL_5);
-	hdmi->cec_irq_was_rx = cntrl5 & VC4_HDMI_CEC_RX_CEC_INT;
-	if (hdmi->cec_irq_was_rx) {
-		vc4_cec_read_msg(vc4, cntrl1);
+	vc4_hdmi->cec_rx_msg.len = 0;
+	cntrl1 = HDMI_READ(HDMI_CEC_CNTRL_1);
+	cntrl5 = HDMI_READ(HDMI_CEC_CNTRL_5);
+	vc4_hdmi->cec_irq_was_rx = cntrl5 & VC4_HDMI_CEC_RX_CEC_INT;
+	if (vc4_hdmi->cec_irq_was_rx) {
+		vc4_cec_read_msg(vc4_hdmi, cntrl1);
 		cntrl1 |= VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1, cntrl1);
+		HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
 		cntrl1 &= ~VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
 	} else {
-		hdmi->cec_tx_ok = cntrl1 & VC4_HDMI_CEC_TX_STATUS_GOOD;
+		vc4_hdmi->cec_tx_ok = cntrl1 & VC4_HDMI_CEC_TX_STATUS_GOOD;
 		cntrl1 &= ~VC4_HDMI_CEC_START_XMIT_BEGIN;
 	}
-	HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1, cntrl1);
-	HDMI_WRITE(VC4_HDMI_CPU_CLEAR, VC4_HDMI_CPU_CEC);
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
+	HDMI_WRITE(HDMI_CEC_CPU_CLEAR, VC4_HDMI_CPU_CEC);
 
 	return IRQ_WAKE_THREAD;
 }
 
 static int vc4_hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable)
 {
-	struct vc4_dev *vc4 = cec_get_drvdata(adap);
+	struct vc4_hdmi *vc4_hdmi = cec_get_drvdata(adap);
 	/* clock period in microseconds */
 	const u32 usecs = 1000000 / CEC_CLOCK_FREQ;
-	u32 val = HDMI_READ(VC4_HDMI_CEC_CNTRL_5);
+	u32 val = HDMI_READ(HDMI_CEC_CNTRL_5);
 
 	val &= ~(VC4_HDMI_CEC_TX_SW_RESET | VC4_HDMI_CEC_RX_SW_RESET |
 		 VC4_HDMI_CEC_CNT_TO_4700_US_MASK |
@@ -1217,30 +1323,30 @@ static int vc4_hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable)
 	       ((4500 / usecs) << VC4_HDMI_CEC_CNT_TO_4500_US_SHIFT);
 
 	if (enable) {
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_5, val |
+		HDMI_WRITE(HDMI_CEC_CNTRL_5, val |
 			   VC4_HDMI_CEC_TX_SW_RESET | VC4_HDMI_CEC_RX_SW_RESET);
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_5, val);
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_2,
-			 ((1500 / usecs) << VC4_HDMI_CEC_CNT_TO_1500_US_SHIFT) |
-			 ((1300 / usecs) << VC4_HDMI_CEC_CNT_TO_1300_US_SHIFT) |
-			 ((800 / usecs) << VC4_HDMI_CEC_CNT_TO_800_US_SHIFT) |
-			 ((600 / usecs) << VC4_HDMI_CEC_CNT_TO_600_US_SHIFT) |
-			 ((400 / usecs) << VC4_HDMI_CEC_CNT_TO_400_US_SHIFT));
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_3,
-			 ((2750 / usecs) << VC4_HDMI_CEC_CNT_TO_2750_US_SHIFT) |
-			 ((2400 / usecs) << VC4_HDMI_CEC_CNT_TO_2400_US_SHIFT) |
-			 ((2050 / usecs) << VC4_HDMI_CEC_CNT_TO_2050_US_SHIFT) |
-			 ((1700 / usecs) << VC4_HDMI_CEC_CNT_TO_1700_US_SHIFT));
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_4,
-			 ((4300 / usecs) << VC4_HDMI_CEC_CNT_TO_4300_US_SHIFT) |
-			 ((3900 / usecs) << VC4_HDMI_CEC_CNT_TO_3900_US_SHIFT) |
-			 ((3600 / usecs) << VC4_HDMI_CEC_CNT_TO_3600_US_SHIFT) |
-			 ((3500 / usecs) << VC4_HDMI_CEC_CNT_TO_3500_US_SHIFT));
-
-		HDMI_WRITE(VC4_HDMI_CPU_MASK_CLEAR, VC4_HDMI_CPU_CEC);
+		HDMI_WRITE(HDMI_CEC_CNTRL_5, val);
+		HDMI_WRITE(HDMI_CEC_CNTRL_2,
+			   ((1500 / usecs) << VC4_HDMI_CEC_CNT_TO_1500_US_SHIFT) |
+			   ((1300 / usecs) << VC4_HDMI_CEC_CNT_TO_1300_US_SHIFT) |
+			   ((800 / usecs) << VC4_HDMI_CEC_CNT_TO_800_US_SHIFT) |
+			   ((600 / usecs) << VC4_HDMI_CEC_CNT_TO_600_US_SHIFT) |
+			   ((400 / usecs) << VC4_HDMI_CEC_CNT_TO_400_US_SHIFT));
+		HDMI_WRITE(HDMI_CEC_CNTRL_3,
+			   ((2750 / usecs) << VC4_HDMI_CEC_CNT_TO_2750_US_SHIFT) |
+			   ((2400 / usecs) << VC4_HDMI_CEC_CNT_TO_2400_US_SHIFT) |
+			   ((2050 / usecs) << VC4_HDMI_CEC_CNT_TO_2050_US_SHIFT) |
+			   ((1700 / usecs) << VC4_HDMI_CEC_CNT_TO_1700_US_SHIFT));
+		HDMI_WRITE(HDMI_CEC_CNTRL_4,
+			   ((4300 / usecs) << VC4_HDMI_CEC_CNT_TO_4300_US_SHIFT) |
+			   ((3900 / usecs) << VC4_HDMI_CEC_CNT_TO_3900_US_SHIFT) |
+			   ((3600 / usecs) << VC4_HDMI_CEC_CNT_TO_3600_US_SHIFT) |
+			   ((3500 / usecs) << VC4_HDMI_CEC_CNT_TO_3500_US_SHIFT));
+
+		HDMI_WRITE(HDMI_CEC_CPU_MASK_CLEAR, VC4_HDMI_CPU_CEC);
 	} else {
-		HDMI_WRITE(VC4_HDMI_CPU_MASK_SET, VC4_HDMI_CPU_CEC);
-		HDMI_WRITE(VC4_HDMI_CEC_CNTRL_5, val |
+		HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, VC4_HDMI_CPU_CEC);
+		HDMI_WRITE(HDMI_CEC_CNTRL_5, val |
 			   VC4_HDMI_CEC_TX_SW_RESET | VC4_HDMI_CEC_RX_SW_RESET);
 	}
 	return 0;
@@ -1248,10 +1354,10 @@ static int vc4_hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable)
 
 static int vc4_hdmi_cec_adap_log_addr(struct cec_adapter *adap, u8 log_addr)
 {
-	struct vc4_dev *vc4 = cec_get_drvdata(adap);
+	struct vc4_hdmi *vc4_hdmi = cec_get_drvdata(adap);
 
-	HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1,
-		   (HDMI_READ(VC4_HDMI_CEC_CNTRL_1) & ~VC4_HDMI_CEC_ADDR_MASK) |
+	HDMI_WRITE(HDMI_CEC_CNTRL_1,
+		   (HDMI_READ(HDMI_CEC_CNTRL_1) & ~VC4_HDMI_CEC_ADDR_MASK) |
 		   (log_addr & 0xf) << VC4_HDMI_CEC_ADDR_SHIFT);
 	return 0;
 }
@@ -1259,25 +1365,25 @@ static int vc4_hdmi_cec_adap_log_addr(struct cec_adapter *adap, u8 log_addr)
 static int vc4_hdmi_cec_adap_transmit(struct cec_adapter *adap, u8 attempts,
 				      u32 signal_free_time, struct cec_msg *msg)
 {
-	struct vc4_dev *vc4 = cec_get_drvdata(adap);
+	struct vc4_hdmi *vc4_hdmi = cec_get_drvdata(adap);
 	u32 val;
 	unsigned int i;
 
 	for (i = 0; i < msg->len; i += 4)
-		HDMI_WRITE(VC4_HDMI_CEC_TX_DATA_1 + i,
+		HDMI_WRITE(HDMI_CEC_TX_DATA_1 + i,
 			   (msg->msg[i]) |
 			   (msg->msg[i + 1] << 8) |
 			   (msg->msg[i + 2] << 16) |
 			   (msg->msg[i + 3] << 24));
 
-	val = HDMI_READ(VC4_HDMI_CEC_CNTRL_1);
+	val = HDMI_READ(HDMI_CEC_CNTRL_1);
 	val &= ~VC4_HDMI_CEC_START_XMIT_BEGIN;
-	HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1, val);
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, val);
 	val &= ~VC4_HDMI_CEC_MESSAGE_LENGTH_MASK;
 	val |= (msg->len - 1) << VC4_HDMI_CEC_MESSAGE_LENGTH_SHIFT;
 	val |= VC4_HDMI_CEC_START_XMIT_BEGIN;
 
-	HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1, val);
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, val);
 	return 0;
 }
 
@@ -1286,61 +1392,275 @@ static const struct cec_adap_ops vc4_hdmi_cec_adap_ops = {
 	.adap_log_addr = vc4_hdmi_cec_adap_log_addr,
 	.adap_transmit = vc4_hdmi_cec_adap_transmit,
 };
-#endif
 
-static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
+static int vc4_hdmi_cec_init(struct vc4_hdmi *vc4_hdmi)
 {
-#ifdef CONFIG_DRM_VC4_HDMI_CEC
 	struct cec_connector_info conn_info;
-#endif
-	struct platform_device *pdev = to_platform_device(dev);
-	struct drm_device *drm = dev_get_drvdata(master);
-	struct vc4_dev *vc4 = drm->dev_private;
-	struct vc4_hdmi *hdmi;
-	struct vc4_hdmi_encoder *vc4_hdmi_encoder;
-	struct device_node *ddc_node;
+	struct platform_device *pdev = vc4_hdmi->pdev;
 	u32 value;
 	int ret;
 
-	hdmi = devm_kzalloc(dev, sizeof(*hdmi), GFP_KERNEL);
-	if (!hdmi)
+	if (!vc4_hdmi->variant->cec_available)
+		return 0;
+
+	vc4_hdmi->cec_adap = cec_allocate_adapter(&vc4_hdmi_cec_adap_ops,
+						  vc4_hdmi, "vc4",
+						  CEC_CAP_DEFAULTS |
+						  CEC_CAP_CONNECTOR_INFO, 1);
+	ret = PTR_ERR_OR_ZERO(vc4_hdmi->cec_adap);
+	if (ret < 0)
+		return ret;
+
+	cec_fill_conn_info_from_drm(&conn_info, &vc4_hdmi->connector);
+	cec_s_conn_info(vc4_hdmi->cec_adap, &conn_info);
+
+	HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, 0xffffffff);
+	value = HDMI_READ(HDMI_CEC_CNTRL_1);
+	value &= ~VC4_HDMI_CEC_DIV_CLK_CNT_MASK;
+	/*
+	 * Set the logical address to Unregistered and set the clock
+	 * divider: the hsm_clock rate and this divider setting will
+	 * give a 40 kHz CEC clock.
+	 */
+	value |= VC4_HDMI_CEC_ADDR_MASK |
+		 (4091 << VC4_HDMI_CEC_DIV_CLK_CNT_SHIFT);
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, value);
+	ret = devm_request_threaded_irq(&pdev->dev, platform_get_irq(pdev, 0),
+					vc4_cec_irq_handler,
+					vc4_cec_irq_handler_thread, 0,
+					"vc4 hdmi cec", vc4_hdmi);
+	if (ret)
+		goto err_delete_cec_adap;
+
+	ret = cec_register_adapter(vc4_hdmi->cec_adap, &pdev->dev);
+	if (ret < 0)
+		goto err_delete_cec_adap;
+
+	return 0;
+
+err_delete_cec_adap:
+	cec_delete_adapter(vc4_hdmi->cec_adap);
+
+	return ret;
+}
+
+static void vc4_hdmi_cec_exit(struct vc4_hdmi *vc4_hdmi)
+{
+	cec_unregister_adapter(vc4_hdmi->cec_adap);
+}
+#else
+static int vc4_hdmi_cec_init(struct vc4_hdmi *vc4_hdmi)
+{
+	return 0;
+}
+
+static void vc4_hdmi_cec_exit(struct vc4_hdmi *vc4_hdmi) {};
+
+#endif
+
+static int vc4_hdmi_build_regset(struct vc4_hdmi *vc4_hdmi,
+				 struct debugfs_regset32 *regset,
+				 enum vc4_hdmi_regs reg)
+{
+	const struct vc4_hdmi_variant *variant = vc4_hdmi->variant;
+	struct debugfs_reg32 *regs, *new_regs;
+	unsigned int count = 0;
+	unsigned int i;
+
+	regs = kcalloc(variant->num_registers, sizeof(*regs),
+		       GFP_KERNEL);
+	if (!regs)
 		return -ENOMEM;
 
-	vc4_hdmi_encoder = devm_kzalloc(dev, sizeof(*vc4_hdmi_encoder),
-					GFP_KERNEL);
-	if (!vc4_hdmi_encoder)
+	for (i = 0; i < variant->num_registers; i++) {
+		const struct vc4_hdmi_register *field =	&variant->registers[i];
+
+		if (field->reg != reg)
+			continue;
+
+		regs[count].name = field->name;
+		regs[count].offset = field->offset;
+		count++;
+	}
+
+	new_regs = krealloc(regs, count * sizeof(*regs), GFP_KERNEL);
+	if (!new_regs)
 		return -ENOMEM;
-	vc4_hdmi_encoder->base.type = VC4_ENCODER_TYPE_HDMI;
-	hdmi->encoder = &vc4_hdmi_encoder->base.base;
-
-	hdmi->pdev = pdev;
-	hdmi->hdmicore_regs = vc4_ioremap_regs(pdev, 0);
-	if (IS_ERR(hdmi->hdmicore_regs))
-		return PTR_ERR(hdmi->hdmicore_regs);
-
-	hdmi->hd_regs = vc4_ioremap_regs(pdev, 1);
-	if (IS_ERR(hdmi->hd_regs))
-		return PTR_ERR(hdmi->hd_regs);
-
-	hdmi->hdmi_regset.base = hdmi->hdmicore_regs;
-	hdmi->hdmi_regset.regs = hdmi_regs;
-	hdmi->hdmi_regset.nregs = ARRAY_SIZE(hdmi_regs);
-	hdmi->hd_regset.base = hdmi->hd_regs;
-	hdmi->hd_regset.regs = hd_regs;
-	hdmi->hd_regset.nregs = ARRAY_SIZE(hd_regs);
-
-	hdmi->pixel_clock = devm_clk_get(dev, "pixel");
-	if (IS_ERR(hdmi->pixel_clock)) {
-		ret = PTR_ERR(hdmi->pixel_clock);
+
+	regset->base = __vc4_hdmi_get_field_base(vc4_hdmi, reg);
+	regset->regs = new_regs;
+	regset->nregs = count;
+
+	return 0;
+}
+
+static int vc4_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi)
+{
+	struct platform_device *pdev = vc4_hdmi->pdev;
+	struct device *dev = &pdev->dev;
+	int ret;
+
+	vc4_hdmi->hdmicore_regs = vc4_ioremap_regs(pdev, 0);
+	if (IS_ERR(vc4_hdmi->hdmicore_regs))
+		return PTR_ERR(vc4_hdmi->hdmicore_regs);
+
+	vc4_hdmi->hd_regs = vc4_ioremap_regs(pdev, 1);
+	if (IS_ERR(vc4_hdmi->hd_regs))
+		return PTR_ERR(vc4_hdmi->hd_regs);
+
+	ret = vc4_hdmi_build_regset(vc4_hdmi, &vc4_hdmi->hd_regset, VC4_HD);
+	if (ret)
+		return ret;
+
+	ret = vc4_hdmi_build_regset(vc4_hdmi, &vc4_hdmi->hdmi_regset, VC4_HDMI);
+	if (ret)
+		return ret;
+
+	vc4_hdmi->pixel_clock = devm_clk_get(dev, "pixel");
+	if (IS_ERR(vc4_hdmi->pixel_clock)) {
+		ret = PTR_ERR(vc4_hdmi->pixel_clock);
 		if (ret != -EPROBE_DEFER)
 			DRM_ERROR("Failed to get pixel clock\n");
 		return ret;
 	}
-	hdmi->hsm_clock = devm_clk_get(dev, "hdmi");
-	if (IS_ERR(hdmi->hsm_clock)) {
+
+	vc4_hdmi->hsm_clock = devm_clk_get(dev, "hdmi");
+	if (IS_ERR(vc4_hdmi->hsm_clock)) {
 		DRM_ERROR("Failed to get HDMI state machine clock\n");
-		return PTR_ERR(hdmi->hsm_clock);
+		return PTR_ERR(vc4_hdmi->hsm_clock);
 	}
+	vc4_hdmi->audio_clock = vc4_hdmi->hsm_clock;
+
+	return 0;
+}
+
+static int vc5_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi)
+{
+	struct platform_device *pdev = vc4_hdmi->pdev;
+	struct device *dev = &pdev->dev;
+	struct resource *res;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->hdmicore_regs = devm_ioremap(dev, res->start,
+					       resource_size(res));
+	if (!vc4_hdmi->hdmicore_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hd");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->hd_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->hd_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cec");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->cec_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->cec_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "csc");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->csc_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->csc_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "dvp");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->dvp_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->dvp_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "phy");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->phy_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->phy_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "packet");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->ram_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->ram_regs)
+		return -ENOMEM;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "rm");
+	if (!res)
+		return -ENODEV;
+
+	vc4_hdmi->rm_regs = devm_ioremap(dev, res->start, resource_size(res));
+	if (!vc4_hdmi->rm_regs)
+		return -ENOMEM;
+
+	vc4_hdmi->hsm_clock = devm_clk_get(dev, "hdmi");
+	if (IS_ERR(vc4_hdmi->hsm_clock)) {
+		DRM_ERROR("Failed to get HDMI state machine clock\n");
+		return PTR_ERR(vc4_hdmi->hsm_clock);
+	}
+
+	vc4_hdmi->pixel_bvb_clock = devm_clk_get(dev, "bvb");
+	if (IS_ERR(vc4_hdmi->pixel_bvb_clock)) {
+		DRM_ERROR("Failed to get pixel bvb clock\n");
+		return PTR_ERR(vc4_hdmi->pixel_bvb_clock);
+	}
+
+	vc4_hdmi->audio_clock = devm_clk_get(dev, "audio");
+	if (IS_ERR(vc4_hdmi->audio_clock)) {
+		DRM_ERROR("Failed to get audio clock\n");
+		return PTR_ERR(vc4_hdmi->audio_clock);
+	}
+
+	vc4_hdmi->reset = devm_reset_control_get(dev, NULL);
+	if (IS_ERR(vc4_hdmi->reset)) {
+		DRM_ERROR("Failed to get HDMI reset line\n");
+		return PTR_ERR(vc4_hdmi->reset);
+	}
+
+	return 0;
+}
+
+static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
+{
+	const struct vc4_hdmi_variant *variant = of_device_get_match_data(dev);
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_hdmi *vc4_hdmi;
+	struct drm_encoder *encoder;
+	struct device_node *ddc_node;
+	u32 value;
+	int ret;
+
+	vc4_hdmi = devm_kzalloc(dev, sizeof(*vc4_hdmi), GFP_KERNEL);
+	if (!vc4_hdmi)
+		return -ENOMEM;
+
+	dev_set_drvdata(dev, vc4_hdmi);
+	encoder = &vc4_hdmi->encoder.base.base;
+	vc4_hdmi->encoder.base.type = variant->encoder_type;
+	vc4_hdmi->encoder.base.pre_crtc_configure = vc4_hdmi_encoder_pre_crtc_configure;
+	vc4_hdmi->encoder.base.pre_crtc_enable = vc4_hdmi_encoder_pre_crtc_enable;
+	vc4_hdmi->encoder.base.post_crtc_enable = vc4_hdmi_encoder_post_crtc_enable;
+	vc4_hdmi->encoder.base.post_crtc_disable = vc4_hdmi_encoder_post_crtc_disable;
+	vc4_hdmi->encoder.base.post_crtc_powerdown = vc4_hdmi_encoder_post_crtc_powerdown;
+	vc4_hdmi->pdev = pdev;
+	vc4_hdmi->variant = variant;
+
+	ret = variant->init_resources(vc4_hdmi);
+	if (ret)
+		return ret;
 
 	ddc_node = of_parse_phandle(dev->of_node, "ddc", 0);
 	if (!ddc_node) {
@@ -1348,123 +1668,62 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 		return -ENODEV;
 	}
 
-	hdmi->ddc = of_find_i2c_adapter_by_node(ddc_node);
+	vc4_hdmi->ddc = of_find_i2c_adapter_by_node(ddc_node);
 	of_node_put(ddc_node);
-	if (!hdmi->ddc) {
+	if (!vc4_hdmi->ddc) {
 		DRM_DEBUG("Failed to get ddc i2c adapter by node\n");
 		return -EPROBE_DEFER;
 	}
 
-	/* This is the rate that is set by the firmware.  The number
-	 * needs to be a bit higher than the pixel clock rate
-	 * (generally 148.5Mhz).
-	 */
-	ret = clk_set_rate(hdmi->hsm_clock, HSM_CLOCK_FREQ);
-	if (ret) {
-		DRM_ERROR("Failed to set HSM clock rate: %d\n", ret);
-		goto err_put_i2c;
-	}
-
-	ret = clk_prepare_enable(hdmi->hsm_clock);
-	if (ret) {
-		DRM_ERROR("Failed to turn on HDMI state machine clock: %d\n",
-			  ret);
-		goto err_put_i2c;
-	}
-
 	/* Only use the GPIO HPD pin if present in the DT, otherwise
 	 * we'll use the HDMI core's register.
 	 */
 	if (of_find_property(dev->of_node, "hpd-gpios", &value)) {
 		enum of_gpio_flags hpd_gpio_flags;
 
-		hdmi->hpd_gpio = of_get_named_gpio_flags(dev->of_node,
-							 "hpd-gpios", 0,
-							 &hpd_gpio_flags);
-		if (hdmi->hpd_gpio < 0) {
-			ret = hdmi->hpd_gpio;
+		vc4_hdmi->hpd_gpio = of_get_named_gpio_flags(dev->of_node,
+							     "hpd-gpios", 0,
+							     &hpd_gpio_flags);
+		if (vc4_hdmi->hpd_gpio < 0) {
+			ret = vc4_hdmi->hpd_gpio;
 			goto err_unprepare_hsm;
 		}
 
-		hdmi->hpd_active_low = hpd_gpio_flags & OF_GPIO_ACTIVE_LOW;
+		vc4_hdmi->hpd_active_low = hpd_gpio_flags & OF_GPIO_ACTIVE_LOW;
 	}
 
-	vc4->hdmi = hdmi;
-
-	/* HDMI core must be enabled. */
-	if (!(HD_READ(VC4_HD_M_CTL) & VC4_HD_M_ENABLE)) {
-		HD_WRITE(VC4_HD_M_CTL, VC4_HD_M_SW_RST);
-		udelay(1);
-		HD_WRITE(VC4_HD_M_CTL, 0);
-
-		HD_WRITE(VC4_HD_M_CTL, VC4_HD_M_ENABLE);
-	}
 	pm_runtime_enable(dev);
 
-	drm_simple_encoder_init(drm, hdmi->encoder, DRM_MODE_ENCODER_TMDS);
-	drm_encoder_helper_add(hdmi->encoder, &vc4_hdmi_encoder_helper_funcs);
+	drm_simple_encoder_init(drm, encoder, DRM_MODE_ENCODER_TMDS);
+	drm_encoder_helper_add(encoder, &vc4_hdmi_encoder_helper_funcs);
 
-	hdmi->connector =
-		vc4_hdmi_connector_init(drm, hdmi->encoder, hdmi->ddc);
-	if (IS_ERR(hdmi->connector)) {
-		ret = PTR_ERR(hdmi->connector);
+	ret = vc4_hdmi_connector_init(drm, vc4_hdmi);
+	if (ret)
 		goto err_destroy_encoder;
-	}
-#ifdef CONFIG_DRM_VC4_HDMI_CEC
-	hdmi->cec_adap = cec_allocate_adapter(&vc4_hdmi_cec_adap_ops,
-					      vc4, "vc4",
-					      CEC_CAP_DEFAULTS |
-					      CEC_CAP_CONNECTOR_INFO, 1);
-	ret = PTR_ERR_OR_ZERO(hdmi->cec_adap);
-	if (ret < 0)
-		goto err_destroy_conn;
 
-	cec_fill_conn_info_from_drm(&conn_info, hdmi->connector);
-	cec_s_conn_info(hdmi->cec_adap, &conn_info);
-
-	HDMI_WRITE(VC4_HDMI_CPU_MASK_SET, 0xffffffff);
-	value = HDMI_READ(VC4_HDMI_CEC_CNTRL_1);
-	value &= ~VC4_HDMI_CEC_DIV_CLK_CNT_MASK;
-	/*
-	 * Set the logical address to Unregistered and set the clock
-	 * divider: the hsm_clock rate and this divider setting will
-	 * give a 40 kHz CEC clock.
-	 */
-	value |= VC4_HDMI_CEC_ADDR_MASK |
-		 (4091 << VC4_HDMI_CEC_DIV_CLK_CNT_SHIFT);
-	HDMI_WRITE(VC4_HDMI_CEC_CNTRL_1, value);
-	ret = devm_request_threaded_irq(dev, platform_get_irq(pdev, 0),
-					vc4_cec_irq_handler,
-					vc4_cec_irq_handler_thread, 0,
-					"vc4 hdmi cec", vc4);
+	ret = vc4_hdmi_cec_init(vc4_hdmi);
 	if (ret)
-		goto err_delete_cec_adap;
-	ret = cec_register_adapter(hdmi->cec_adap, dev);
-	if (ret < 0)
-		goto err_delete_cec_adap;
-#endif
+		goto err_destroy_conn;
 
-	ret = vc4_hdmi_audio_init(hdmi);
+	ret = vc4_hdmi_audio_init(vc4_hdmi);
 	if (ret)
-		goto err_destroy_encoder;
+		goto err_free_cec;
 
-	vc4_debugfs_add_file(drm, "hdmi_regs", vc4_hdmi_debugfs_regs, hdmi);
+	vc4_debugfs_add_file(drm, variant->debugfs_name,
+			     vc4_hdmi_debugfs_regs,
+			     vc4_hdmi);
 
 	return 0;
 
-#ifdef CONFIG_DRM_VC4_HDMI_CEC
-err_delete_cec_adap:
-	cec_delete_adapter(hdmi->cec_adap);
+err_free_cec:
+	vc4_hdmi_cec_exit(vc4_hdmi);
 err_destroy_conn:
-	vc4_hdmi_connector_destroy(hdmi->connector);
-#endif
+	vc4_hdmi_connector_destroy(&vc4_hdmi->connector);
 err_destroy_encoder:
-	drm_encoder_cleanup(hdmi->encoder);
+	drm_encoder_cleanup(encoder);
 err_unprepare_hsm:
-	clk_disable_unprepare(hdmi->hsm_clock);
 	pm_runtime_disable(dev);
-err_put_i2c:
-	put_device(&hdmi->ddc->dev);
+	put_device(&vc4_hdmi->ddc->dev);
 
 	return ret;
 }
@@ -1472,20 +1731,39 @@ err_put_i2c:
 static void vc4_hdmi_unbind(struct device *dev, struct device *master,
 			    void *data)
 {
-	struct drm_device *drm = dev_get_drvdata(master);
-	struct vc4_dev *vc4 = drm->dev_private;
-	struct vc4_hdmi *hdmi = vc4->hdmi;
+	struct vc4_hdmi *vc4_hdmi;
 
-	cec_unregister_adapter(hdmi->cec_adap);
-	vc4_hdmi_connector_destroy(hdmi->connector);
-	drm_encoder_cleanup(hdmi->encoder);
+	/*
+	 * ASoC makes it a bit hard to retrieve a pointer to the
+	 * vc4_hdmi structure. Registering the card will overwrite our
+	 * device drvdata with a pointer to the snd_soc_card structure,
+	 * which can then be used to retrieve whatever drvdata we want
+	 * to associate.
+	 *
+	 * However, that doesn't fly in the case where we wouldn't
+	 * register an ASoC card (because of an old DT that is missing
+	 * the dmas properties for example), then the card isn't
+	 * registered and the device drvdata wouldn't be set.
+	 *
+	 * We can deal with both cases by making sure a snd_soc_card
+	 * pointer and a vc4_hdmi structure are pointing to the same
+	 * memory address, so we can treat them indistinctly without any
+	 * issue.
+	 */
+	BUILD_BUG_ON(offsetof(struct vc4_hdmi_audio, card) != 0);
+	BUILD_BUG_ON(offsetof(struct vc4_hdmi, audio) != 0);
+	vc4_hdmi = dev_get_drvdata(dev);
 
-	clk_disable_unprepare(hdmi->hsm_clock);
-	pm_runtime_disable(dev);
+	kfree(vc4_hdmi->hdmi_regset.regs);
+	kfree(vc4_hdmi->hd_regset.regs);
 
-	put_device(&hdmi->ddc->dev);
+	vc4_hdmi_cec_exit(vc4_hdmi);
+	vc4_hdmi_connector_destroy(&vc4_hdmi->connector);
+	drm_encoder_cleanup(&vc4_hdmi->encoder.base.base);
+
+	pm_runtime_disable(dev);
 
-	vc4->hdmi = NULL;
+	put_device(&vc4_hdmi->ddc->dev);
 }
 
 static const struct component_ops vc4_hdmi_ops = {
@@ -1504,8 +1782,80 @@ static int vc4_hdmi_dev_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct vc4_hdmi_variant bcm2835_variant = {
+	.encoder_type		= VC4_ENCODER_TYPE_HDMI0,
+	.debugfs_name		= "hdmi_regs",
+	.card_name		= "vc4-hdmi",
+	.max_pixel_clock	= 162000000,
+	.cec_available		= true,
+	.registers		= vc4_hdmi_fields,
+	.num_registers		= ARRAY_SIZE(vc4_hdmi_fields),
+
+	.init_resources		= vc4_hdmi_init_resources,
+	.csc_setup		= vc4_hdmi_csc_setup,
+	.reset			= vc4_hdmi_reset,
+	.set_timings		= vc4_hdmi_set_timings,
+	.phy_init		= vc4_hdmi_phy_init,
+	.phy_disable		= vc4_hdmi_phy_disable,
+	.phy_rng_enable		= vc4_hdmi_phy_rng_enable,
+	.phy_rng_disable	= vc4_hdmi_phy_rng_disable,
+	.channel_map		= vc4_hdmi_channel_map,
+};
+
+static const struct vc4_hdmi_variant bcm2711_hdmi0_variant = {
+	.encoder_type		= VC4_ENCODER_TYPE_HDMI0,
+	.debugfs_name		= "hdmi0_regs",
+	.card_name		= "vc4-hdmi-0",
+	.max_pixel_clock	= 297000000,
+	.registers		= vc5_hdmi_hdmi0_fields,
+	.num_registers		= ARRAY_SIZE(vc5_hdmi_hdmi0_fields),
+	.phy_lane_mapping	= {
+		PHY_LANE_0,
+		PHY_LANE_1,
+		PHY_LANE_2,
+		PHY_LANE_CK,
+	},
+
+	.init_resources		= vc5_hdmi_init_resources,
+	.csc_setup		= vc5_hdmi_csc_setup,
+	.reset			= vc5_hdmi_reset,
+	.set_timings		= vc5_hdmi_set_timings,
+	.phy_init		= vc5_hdmi_phy_init,
+	.phy_disable		= vc5_hdmi_phy_disable,
+	.phy_rng_enable		= vc5_hdmi_phy_rng_enable,
+	.phy_rng_disable	= vc5_hdmi_phy_rng_disable,
+	.channel_map		= vc5_hdmi_channel_map,
+};
+
+static const struct vc4_hdmi_variant bcm2711_hdmi1_variant = {
+	.encoder_type		= VC4_ENCODER_TYPE_HDMI1,
+	.debugfs_name		= "hdmi1_regs",
+	.card_name		= "vc4-hdmi-1",
+	.max_pixel_clock	= 297000000,
+	.registers		= vc5_hdmi_hdmi1_fields,
+	.num_registers		= ARRAY_SIZE(vc5_hdmi_hdmi1_fields),
+	.phy_lane_mapping	= {
+		PHY_LANE_1,
+		PHY_LANE_0,
+		PHY_LANE_CK,
+		PHY_LANE_2,
+	},
+
+	.init_resources		= vc5_hdmi_init_resources,
+	.csc_setup		= vc5_hdmi_csc_setup,
+	.reset			= vc5_hdmi_reset,
+	.set_timings		= vc5_hdmi_set_timings,
+	.phy_init		= vc5_hdmi_phy_init,
+	.phy_disable		= vc5_hdmi_phy_disable,
+	.phy_rng_enable		= vc5_hdmi_phy_rng_enable,
+	.phy_rng_disable	= vc5_hdmi_phy_rng_disable,
+	.channel_map		= vc5_hdmi_channel_map,
+};
+
 static const struct of_device_id vc4_hdmi_dt_match[] = {
-	{ .compatible = "brcm,bcm2835-hdmi" },
+	{ .compatible = "brcm,bcm2835-hdmi", .data = &bcm2835_variant },
+	{ .compatible = "brcm,bcm2711-hdmi0", .data = &bcm2711_hdmi0_variant },
+	{ .compatible = "brcm,bcm2711-hdmi1", .data = &bcm2711_hdmi1_variant },
 	{}
 };
 
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
new file mode 100644
index 000000000000..63c6f8bddf1d
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
@@ -0,0 +1,184 @@
+#ifndef _VC4_HDMI_H_
+#define _VC4_HDMI_H_
+
+#include <drm/drm_connector.h>
+#include <media/cec.h>
+#include <sound/dmaengine_pcm.h>
+#include <sound/soc.h>
+
+#include "vc4_drv.h"
+
+/* VC4 HDMI encoder KMS struct */
+struct vc4_hdmi_encoder {
+	struct vc4_encoder base;
+	bool hdmi_monitor;
+	bool limited_rgb_range;
+};
+
+static inline struct vc4_hdmi_encoder *
+to_vc4_hdmi_encoder(struct drm_encoder *encoder)
+{
+	return container_of(encoder, struct vc4_hdmi_encoder, base.base);
+}
+
+struct drm_display_mode;
+
+struct vc4_hdmi;
+struct vc4_hdmi_register;
+
+enum vc4_hdmi_phy_channel {
+	PHY_LANE_0 = 0,
+	PHY_LANE_1,
+	PHY_LANE_2,
+	PHY_LANE_CK,
+};
+
+struct vc4_hdmi_variant {
+	/* Encoder Type for that controller */
+	enum vc4_encoder_type encoder_type;
+
+	/* ALSA card name */
+	const char *card_name;
+
+	/* Filename to expose the registers in debugfs */
+	const char *debugfs_name;
+
+	/* Set to true when the CEC support is available */
+	bool cec_available;
+
+	/* Maximum pixel clock supported by the controller (in Hz) */
+	unsigned long long max_pixel_clock;
+
+	/* List of the registers available on that variant */
+	const struct vc4_hdmi_register *registers;
+
+	/* Number of registers on that variant */
+	unsigned int num_registers;
+
+	/* BCM2711 Only.
+	 * The variants don't map the lane in the same order in the
+	 * PHY, so this is an array mapping the HDMI channel (index)
+	 * to the PHY lane (value).
+	 */
+	enum vc4_hdmi_phy_channel phy_lane_mapping[4];
+
+	/* Callback to get the resources (memory region, interrupts,
+	 * clocks, etc) for that variant.
+	 */
+	int (*init_resources)(struct vc4_hdmi *vc4_hdmi);
+
+	/* Callback to reset the HDMI block */
+	void (*reset)(struct vc4_hdmi *vc4_hdmi);
+
+	/* Callback to enable / disable the CSC */
+	void (*csc_setup)(struct vc4_hdmi *vc4_hdmi, bool enable);
+
+	/* Callback to configure the video timings in the HDMI block */
+	void (*set_timings)(struct vc4_hdmi *vc4_hdmi,
+			    struct drm_display_mode *mode);
+
+	/* Callback to initialize the PHY according to the mode */
+	void (*phy_init)(struct vc4_hdmi *vc4_hdmi,
+			 struct drm_display_mode *mode);
+
+	/* Callback to disable the PHY */
+	void (*phy_disable)(struct vc4_hdmi *vc4_hdmi);
+
+	/* Callback to enable the RNG in the PHY */
+	void (*phy_rng_enable)(struct vc4_hdmi *vc4_hdmi);
+
+	/* Callback to disable the RNG in the PHY */
+	void (*phy_rng_disable)(struct vc4_hdmi *vc4_hdmi);
+
+	/* Callback to get channel map */
+	u32 (*channel_map)(struct vc4_hdmi *vc4_hdmi, u32 channel_mask);
+};
+
+/* HDMI audio information */
+struct vc4_hdmi_audio {
+	struct snd_soc_card card;
+	struct snd_soc_dai_link link;
+	struct snd_soc_dai_link_component cpu;
+	struct snd_soc_dai_link_component codec;
+	struct snd_soc_dai_link_component platform;
+	int samplerate;
+	int channels;
+	struct snd_dmaengine_dai_dma_data dma_data;
+	struct snd_pcm_substream *substream;
+
+	bool streaming;
+};
+
+/* General HDMI hardware state. */
+struct vc4_hdmi {
+	struct vc4_hdmi_audio audio;
+
+	struct platform_device *pdev;
+	const struct vc4_hdmi_variant *variant;
+
+	struct vc4_hdmi_encoder encoder;
+	struct drm_connector connector;
+
+	struct i2c_adapter *ddc;
+	void __iomem *hdmicore_regs;
+	void __iomem *hd_regs;
+
+	/* VC5 Only */
+	void __iomem *cec_regs;
+	/* VC5 Only */
+	void __iomem *csc_regs;
+	/* VC5 Only */
+	void __iomem *dvp_regs;
+	/* VC5 Only */
+	void __iomem *phy_regs;
+	/* VC5 Only */
+	void __iomem *ram_regs;
+	/* VC5 Only */
+	void __iomem *rm_regs;
+
+	int hpd_gpio;
+	bool hpd_active_low;
+
+	struct cec_adapter *cec_adap;
+	struct cec_msg cec_rx_msg;
+	bool cec_tx_ok;
+	bool cec_irq_was_rx;
+
+	struct clk *pixel_clock;
+	struct clk *hsm_clock;
+	struct clk *audio_clock;
+	struct clk *pixel_bvb_clock;
+
+	struct reset_control *reset;
+
+	struct debugfs_regset32 hdmi_regset;
+	struct debugfs_regset32 hd_regset;
+};
+
+static inline struct vc4_hdmi *
+connector_to_vc4_hdmi(struct drm_connector *connector)
+{
+	return container_of(connector, struct vc4_hdmi, connector);
+}
+
+static inline struct vc4_hdmi *
+encoder_to_vc4_hdmi(struct drm_encoder *encoder)
+{
+	struct vc4_hdmi_encoder *_encoder = to_vc4_hdmi_encoder(encoder);
+
+	return container_of(_encoder, struct vc4_hdmi, encoder);
+}
+
+void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
+		       struct drm_display_mode *mode);
+void vc4_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
+void vc4_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi);
+void vc4_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi);
+
+void vc5_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
+		       struct drm_display_mode *mode);
+void vc5_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
+void vc5_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi);
+void vc5_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi);
+
+#endif /* _VC4_HDMI_H_ */
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi_phy.c b/drivers/gpu/drm/vc4/vc4_hdmi_phy.c
new file mode 100644
index 000000000000..057796b54c51
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_hdmi_phy.c
@@ -0,0 +1,521 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015 Broadcom
+ * Copyright (c) 2014 The Linux Foundation. All rights reserved.
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ */
+
+#include "vc4_hdmi.h"
+#include "vc4_regs.h"
+#include "vc4_hdmi_regs.h"
+
+#define VC4_HDMI_TX_PHY_RESET_CTL_PLL_RESETB	BIT(5)
+#define VC4_HDMI_TX_PHY_RESET_CTL_PLLDIV_RESETB	BIT(4)
+#define VC4_HDMI_TX_PHY_RESET_CTL_TX_CK_RESET	BIT(3)
+#define VC4_HDMI_TX_PHY_RESET_CTL_TX_2_RESET	BIT(2)
+#define VC4_HDMI_TX_PHY_RESET_CTL_TX_1_RESET	BIT(1)
+#define VC4_HDMI_TX_PHY_RESET_CTL_TX_0_RESET	BIT(0)
+
+#define VC4_HDMI_TX_PHY_POWERDOWN_CTL_RNDGEN_PWRDN	BIT(4)
+
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_PREEMP_SHIFT	29
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_PREEMP_MASK	VC4_MASK(31, 29)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_MAINDRV_SHIFT	24
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_MAINDRV_MASK	VC4_MASK(28, 24)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_PREEMP_SHIFT	21
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_PREEMP_MASK	VC4_MASK(23, 21)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_MAINDRV_SHIFT	16
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_MAINDRV_MASK	VC4_MASK(20, 16)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_PREEMP_SHIFT	13
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_PREEMP_MASK	VC4_MASK(15, 13)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_MAINDRV_SHIFT	8
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_MAINDRV_MASK	VC4_MASK(12, 8)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_PREEMP_SHIFT	5
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_PREEMP_MASK	VC4_MASK(7, 5)
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_MAINDRV_SHIFT	0
+#define VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_MAINDRV_MASK	VC4_MASK(4, 0)
+
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA2_SHIFT	15
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA2_MASK	VC4_MASK(19, 15)
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA1_SHIFT	10
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA1_MASK	VC4_MASK(14, 10)
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA0_SHIFT	5
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA0_MASK	VC4_MASK(9, 5)
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_CK_SHIFT		0
+#define VC4_HDMI_TX_PHY_CTL_1_RES_SEL_CK_MASK		VC4_MASK(4, 0)
+
+#define VC4_HDMI_TX_PHY_CTL_2_VCO_GAIN_SHIFT		16
+#define VC4_HDMI_TX_PHY_CTL_2_VCO_GAIN_MASK		VC4_MASK(19, 16)
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA2_SHIFT	12
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA2_MASK	VC4_MASK(15, 12)
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA1_SHIFT	8
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA1_MASK	VC4_MASK(11, 8)
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA0_SHIFT	4
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA0_MASK	VC4_MASK(7, 4)
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELCK_SHIFT	0
+#define VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELCK_MASK	VC4_MASK(3, 0)
+
+#define VC4_HDMI_TX_PHY_CTL_3_RP_SHIFT			17
+#define VC4_HDMI_TX_PHY_CTL_3_RP_MASK			VC4_MASK(19, 17)
+#define VC4_HDMI_TX_PHY_CTL_3_RZ_SHIFT			12
+#define VC4_HDMI_TX_PHY_CTL_3_RZ_MASK			VC4_MASK(16, 12)
+#define VC4_HDMI_TX_PHY_CTL_3_CP1_SHIFT			10
+#define VC4_HDMI_TX_PHY_CTL_3_CP1_MASK			VC4_MASK(11, 10)
+#define VC4_HDMI_TX_PHY_CTL_3_CP_SHIFT			8
+#define VC4_HDMI_TX_PHY_CTL_3_CP_MASK			VC4_MASK(9, 8)
+#define VC4_HDMI_TX_PHY_CTL_3_CZ_SHIFT			6
+#define VC4_HDMI_TX_PHY_CTL_3_CZ_MASK			VC4_MASK(7, 6)
+#define VC4_HDMI_TX_PHY_CTL_3_ICP_SHIFT			0
+#define VC4_HDMI_TX_PHY_CTL_3_ICP_MASK			VC4_MASK(5, 0)
+
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_MASH11_MODE		BIT(13)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VC_RANGE_EN		BIT(12)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_EMULATE_VC_LOW	BIT(11)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_EMULATE_VC_HIGH	BIT(10)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_SEL_SHIFT		9
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_SEL_MASK		VC4_MASK(9, 9)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_FB_DIV2		BIT(8)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_POST_DIV2		BIT(7)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_CONT_EN		BIT(6)
+#define VC4_HDMI_TX_PHY_PLL_CTL_0_ENA_VCO_CLK		BIT(5)
+
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_CPP_SHIFT			16
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_CPP_MASK			VC4_MASK(27, 16)
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_FREQ_DOUBLER_DELAY_SHIFT	14
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_FREQ_DOUBLER_DELAY_MASK	VC4_MASK(15, 14)
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_FREQ_DOUBLER_ENABLE		BIT(13)
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_POST_RST_SEL_SHIFT		11
+#define VC4_HDMI_TX_PHY_PLL_CTL_1_POST_RST_SEL_MASK		VC4_MASK(12, 11)
+
+#define VC4_HDMI_TX_PHY_CLK_DIV_VCO_SHIFT		8
+#define VC4_HDMI_TX_PHY_CLK_DIV_VCO_MASK		VC4_MASK(15, 8)
+
+#define VC4_HDMI_TX_PHY_PLL_CFG_PDIV_SHIFT		0
+#define VC4_HDMI_TX_PHY_PLL_CFG_PDIV_MASK		VC4_MASK(3, 0)
+
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TXCK_OUT_SEL_MASK	VC4_MASK(13, 12)
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TXCK_OUT_SEL_SHIFT	12
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX2_OUT_SEL_MASK	VC4_MASK(9, 8)
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX2_OUT_SEL_SHIFT	8
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX1_OUT_SEL_MASK	VC4_MASK(5, 4)
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX1_OUT_SEL_SHIFT	4
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX0_OUT_SEL_MASK	VC4_MASK(1, 0)
+#define VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX0_OUT_SEL_SHIFT	0
+
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1_MIN_LIMIT_MASK		VC4_MASK(27, 0)
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1_MIN_LIMIT_SHIFT	0
+
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2_MAX_LIMIT_MASK		VC4_MASK(27, 0)
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2_MAX_LIMIT_SHIFT	0
+
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_STABLE_THRESHOLD_MASK	VC4_MASK(31, 16)
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_STABLE_THRESHOLD_SHIFT	16
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_HOLD_THRESHOLD_MASK	VC4_MASK(15, 0)
+#define VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_HOLD_THRESHOLD_SHIFT	0
+
+#define VC4_HDMI_RM_CONTROL_EN_FREEZE_COUNTERS		BIT(19)
+#define VC4_HDMI_RM_CONTROL_EN_LOAD_INTEGRATOR		BIT(17)
+#define VC4_HDMI_RM_CONTROL_FREE_RUN			BIT(4)
+
+#define VC4_HDMI_RM_OFFSET_ONLY				BIT(31)
+#define VC4_HDMI_RM_OFFSET_OFFSET_SHIFT			0
+#define VC4_HDMI_RM_OFFSET_OFFSET_MASK			VC4_MASK(30, 0)
+
+#define VC4_HDMI_RM_FORMAT_SHIFT_SHIFT			24
+#define VC4_HDMI_RM_FORMAT_SHIFT_MASK			VC4_MASK(25, 24)
+
+#define OSCILLATOR_FREQUENCY	54000000
+
+void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi, struct drm_display_mode *mode)
+{
+	/* PHY should be in reset, like
+	 * vc4_hdmi_encoder_disable() does.
+	 */
+
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL, 0xf << 16);
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL, 0);
+}
+
+void vc4_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL, 0xf << 16);
+}
+
+void vc4_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_CTL_0,
+		   HDMI_READ(HDMI_TX_PHY_CTL_0) &
+		   ~VC4_HDMI_TX_PHY_RNG_PWRDN);
+}
+
+void vc4_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_CTL_0,
+		   HDMI_READ(HDMI_TX_PHY_CTL_0) |
+		   VC4_HDMI_TX_PHY_RNG_PWRDN);
+}
+
+static unsigned long long
+phy_get_vco_freq(unsigned long long clock, u8 *vco_sel, u8 *vco_div)
+{
+	unsigned long long vco_freq = clock;
+	unsigned int _vco_div = 0;
+	unsigned int _vco_sel = 0;
+
+	while (vco_freq < 3000000000ULL) {
+		_vco_div++;
+		vco_freq = clock * _vco_div * 10;
+	}
+
+	if (vco_freq > 4500000000ULL)
+		_vco_sel = 1;
+
+	*vco_sel = _vco_sel;
+	*vco_div = _vco_div;
+
+	return vco_freq;
+}
+
+static u8 phy_get_cp_current(unsigned long vco_freq)
+{
+	if (vco_freq < 3700000000ULL)
+		return 0x1c;
+
+	return 0x18;
+}
+
+static u32 phy_get_rm_offset(unsigned long long vco_freq)
+{
+	unsigned long long fref = OSCILLATOR_FREQUENCY;
+	u64 offset = 0;
+
+	/* RM offset is stored as 9.22 format */
+	offset = vco_freq * 2;
+	offset = offset << 22;
+	do_div(offset, fref);
+	offset >>= 2;
+
+	return offset;
+}
+
+static u8 phy_get_vco_gain(unsigned long long vco_freq)
+{
+	if (vco_freq < 3350000000ULL)
+		return 0xf;
+
+	if (vco_freq < 3700000000ULL)
+		return 0xc;
+
+	if (vco_freq < 4050000000ULL)
+		return 0x6;
+
+	if (vco_freq < 4800000000ULL)
+		return 0x5;
+
+	if (vco_freq < 5200000000ULL)
+		return 0x7;
+
+	return 0x2;
+}
+
+struct phy_lane_settings {
+	struct {
+		u8 preemphasis;
+		u8 main_driver;
+	} amplitude;
+
+	u8 res_sel_data;
+	u8 term_res_sel_data;
+};
+
+struct phy_settings {
+	unsigned long long min_rate;
+	unsigned long long max_rate;
+	struct phy_lane_settings channel[3];
+	struct phy_lane_settings clock;
+};
+
+static const struct phy_settings vc5_hdmi_phy_settings[] = {
+	{
+		0, 50000000,
+		{
+			{{0x0, 0x0A}, 0x12, 0x0},
+			{{0x0, 0x0A}, 0x12, 0x0},
+			{{0x0, 0x0A}, 0x12, 0x0}
+		},
+		{{0x0, 0x0A}, 0x18, 0x0},
+	},
+	{
+		50000001, 75000000,
+		{
+			{{0x0, 0x09}, 0x12, 0x0},
+			{{0x0, 0x09}, 0x12, 0x0},
+			{{0x0, 0x09}, 0x12, 0x0}
+		},
+		{{0x0, 0x0C}, 0x18, 0x3},
+	},
+	{
+		75000001,   165000000,
+		{
+			{{0x0, 0x09}, 0x12, 0x0},
+			{{0x0, 0x09}, 0x12, 0x0},
+			{{0x0, 0x09}, 0x12, 0x0}
+		},
+		{{0x0, 0x0C}, 0x18, 0x3},
+	},
+	{
+		165000001,  250000000,
+		{
+			{{0x0, 0x0F}, 0x12, 0x1},
+			{{0x0, 0x0F}, 0x12, 0x1},
+			{{0x0, 0x0F}, 0x12, 0x1}
+		},
+		{{0x0, 0x0C}, 0x18, 0x3},
+	},
+	{
+		250000001,  340000000,
+		{
+			{{0x2, 0x0D}, 0x12, 0x1},
+			{{0x2, 0x0D}, 0x12, 0x1},
+			{{0x2, 0x0D}, 0x12, 0x1}
+		},
+		{{0x0, 0x0C}, 0x18, 0xF},
+	},
+	{
+		340000001,  450000000,
+		{
+			{{0x0, 0x1B}, 0x12, 0xF},
+			{{0x0, 0x1B}, 0x12, 0xF},
+			{{0x0, 0x1B}, 0x12, 0xF}
+		},
+		{{0x0, 0x0A}, 0x12, 0xF},
+	},
+	{
+		450000001,  600000000,
+		{
+			{{0x0, 0x1C}, 0x12, 0xF},
+			{{0x0, 0x1C}, 0x12, 0xF},
+			{{0x0, 0x1C}, 0x12, 0xF}
+		},
+		{{0x0, 0x0B}, 0x13, 0xF},
+	},
+};
+
+static const struct phy_settings *phy_get_settings(unsigned long long tmds_rate)
+{
+	unsigned int count = ARRAY_SIZE(vc5_hdmi_phy_settings);
+	unsigned int i;
+
+	for (i = 0; i < count; i++) {
+		const struct phy_settings *s = &vc5_hdmi_phy_settings[i];
+
+		if (tmds_rate >= s->min_rate && tmds_rate <= s->max_rate)
+			return s;
+	}
+
+	/*
+	 * If the pixel clock exceeds our max setting, try the max
+	 * setting anyway.
+	 */
+	return &vc5_hdmi_phy_settings[count - 1];
+}
+
+static const struct phy_lane_settings *
+phy_get_channel_settings(enum vc4_hdmi_phy_channel chan,
+			 unsigned long long tmds_rate)
+{
+	const struct phy_settings *settings = phy_get_settings(tmds_rate);
+
+	if (chan == PHY_LANE_CK)
+		return &settings->clock;
+
+	return &settings->channel[chan];
+}
+
+static void vc5_hdmi_reset_phy(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL, 0x0f);
+	HDMI_WRITE(HDMI_TX_PHY_POWERDOWN_CTL, BIT(10));
+}
+
+void vc5_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi, struct drm_display_mode *mode)
+{
+	const struct phy_lane_settings *chan0_settings, *chan1_settings, *chan2_settings, *clock_settings;
+	const struct vc4_hdmi_variant *variant = vc4_hdmi->variant;
+	unsigned long long pixel_freq = mode->clock * 1000;
+	unsigned long long vco_freq;
+	unsigned char word_sel;
+	u8 vco_sel, vco_div;
+
+	vco_freq = phy_get_vco_freq(pixel_freq, &vco_sel, &vco_div);
+
+	vc5_hdmi_reset_phy(vc4_hdmi);
+
+	HDMI_WRITE(HDMI_TX_PHY_POWERDOWN_CTL,
+		   VC4_HDMI_TX_PHY_POWERDOWN_CTL_RNDGEN_PWRDN);
+
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL,
+		   HDMI_READ(HDMI_TX_PHY_RESET_CTL) &
+		   ~VC4_HDMI_TX_PHY_RESET_CTL_TX_0_RESET &
+		   ~VC4_HDMI_TX_PHY_RESET_CTL_TX_1_RESET &
+		   ~VC4_HDMI_TX_PHY_RESET_CTL_TX_2_RESET &
+		   ~VC4_HDMI_TX_PHY_RESET_CTL_TX_CK_RESET);
+
+	HDMI_WRITE(HDMI_RM_CONTROL,
+		   HDMI_READ(HDMI_RM_CONTROL) |
+		   VC4_HDMI_RM_CONTROL_EN_FREEZE_COUNTERS |
+		   VC4_HDMI_RM_CONTROL_EN_LOAD_INTEGRATOR |
+		   VC4_HDMI_RM_CONTROL_FREE_RUN);
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1,
+		   (HDMI_READ(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1) &
+		    ~VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1_MIN_LIMIT_MASK) |
+		   VC4_SET_FIELD(0, VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1_MIN_LIMIT));
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2,
+		   (HDMI_READ(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2) &
+		    ~VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2_MAX_LIMIT_MASK) |
+		   VC4_SET_FIELD(0, VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2_MAX_LIMIT));
+
+	HDMI_WRITE(HDMI_RM_OFFSET,
+		   VC4_SET_FIELD(phy_get_rm_offset(vco_freq),
+				 VC4_HDMI_RM_OFFSET_OFFSET) |
+		   VC4_HDMI_RM_OFFSET_ONLY);
+
+	HDMI_WRITE(HDMI_TX_PHY_CLK_DIV,
+		   VC4_SET_FIELD(vco_div, VC4_HDMI_TX_PHY_CLK_DIV_VCO));
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4,
+		   VC4_SET_FIELD(0xe147, VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_HOLD_THRESHOLD) |
+		   VC4_SET_FIELD(0xe14, VC4_HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4_STABLE_THRESHOLD));
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CTL_0,
+		   VC4_HDMI_TX_PHY_PLL_CTL_0_ENA_VCO_CLK |
+		   VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_CONT_EN |
+		   VC4_HDMI_TX_PHY_PLL_CTL_0_MASH11_MODE |
+		   VC4_SET_FIELD(vco_sel, VC4_HDMI_TX_PHY_PLL_CTL_0_VCO_SEL));
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CTL_1,
+		   HDMI_READ(HDMI_TX_PHY_PLL_CTL_1) |
+		   VC4_HDMI_TX_PHY_PLL_CTL_1_FREQ_DOUBLER_ENABLE |
+		   VC4_SET_FIELD(3, VC4_HDMI_TX_PHY_PLL_CTL_1_POST_RST_SEL) |
+		   VC4_SET_FIELD(1, VC4_HDMI_TX_PHY_PLL_CTL_1_FREQ_DOUBLER_DELAY) |
+		   VC4_SET_FIELD(0x8a, VC4_HDMI_TX_PHY_PLL_CTL_1_CPP));
+
+	HDMI_WRITE(HDMI_RM_FORMAT,
+		   HDMI_READ(HDMI_RM_FORMAT) |
+		   VC4_SET_FIELD(2, VC4_HDMI_RM_FORMAT_SHIFT));
+
+	HDMI_WRITE(HDMI_TX_PHY_PLL_CFG,
+		   HDMI_READ(HDMI_TX_PHY_PLL_CFG) |
+		   VC4_SET_FIELD(1, VC4_HDMI_TX_PHY_PLL_CFG_PDIV));
+
+	if (pixel_freq >= 340000000)
+		word_sel = 3;
+	else
+		word_sel = 0;
+	HDMI_WRITE(HDMI_TX_PHY_TMDS_CLK_WORD_SEL, word_sel);
+
+	HDMI_WRITE(HDMI_TX_PHY_CTL_3,
+		   VC4_SET_FIELD(phy_get_cp_current(vco_freq),
+				 VC4_HDMI_TX_PHY_CTL_3_ICP) |
+		   VC4_SET_FIELD(1, VC4_HDMI_TX_PHY_CTL_3_CP) |
+		   VC4_SET_FIELD(1, VC4_HDMI_TX_PHY_CTL_3_CP1) |
+		   VC4_SET_FIELD(3, VC4_HDMI_TX_PHY_CTL_3_CZ) |
+		   VC4_SET_FIELD(4, VC4_HDMI_TX_PHY_CTL_3_RP) |
+		   VC4_SET_FIELD(6, VC4_HDMI_TX_PHY_CTL_3_RZ));
+
+	chan0_settings =
+		phy_get_channel_settings(variant->phy_lane_mapping[PHY_LANE_0],
+					 pixel_freq);
+	chan1_settings =
+		phy_get_channel_settings(variant->phy_lane_mapping[PHY_LANE_1],
+					 pixel_freq);
+	chan2_settings =
+		phy_get_channel_settings(variant->phy_lane_mapping[PHY_LANE_2],
+					 pixel_freq);
+	clock_settings =
+		phy_get_channel_settings(variant->phy_lane_mapping[PHY_LANE_CK],
+					 pixel_freq);
+
+	HDMI_WRITE(HDMI_TX_PHY_CTL_0,
+		   VC4_SET_FIELD(chan0_settings->amplitude.preemphasis,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_PREEMP) |
+		   VC4_SET_FIELD(chan0_settings->amplitude.main_driver,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_0_MAINDRV) |
+		   VC4_SET_FIELD(chan1_settings->amplitude.preemphasis,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_PREEMP) |
+		   VC4_SET_FIELD(chan1_settings->amplitude.main_driver,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_1_MAINDRV) |
+		   VC4_SET_FIELD(chan2_settings->amplitude.preemphasis,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_PREEMP) |
+		   VC4_SET_FIELD(chan2_settings->amplitude.main_driver,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_2_MAINDRV) |
+		   VC4_SET_FIELD(clock_settings->amplitude.preemphasis,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_PREEMP) |
+		   VC4_SET_FIELD(clock_settings->amplitude.main_driver,
+				 VC4_HDMI_TX_PHY_CTL_0_PREEMP_CK_MAINDRV));
+
+	HDMI_WRITE(HDMI_TX_PHY_CTL_1,
+		   HDMI_READ(HDMI_TX_PHY_CTL_1) |
+		   VC4_SET_FIELD(chan0_settings->res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA0) |
+		   VC4_SET_FIELD(chan1_settings->res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA1) |
+		   VC4_SET_FIELD(chan2_settings->res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_1_RES_SEL_DATA2) |
+		   VC4_SET_FIELD(clock_settings->res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_1_RES_SEL_CK));
+
+	HDMI_WRITE(HDMI_TX_PHY_CTL_2,
+		   VC4_SET_FIELD(chan0_settings->term_res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA0) |
+		   VC4_SET_FIELD(chan1_settings->term_res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA1) |
+		   VC4_SET_FIELD(chan2_settings->term_res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELDATA2) |
+		   VC4_SET_FIELD(clock_settings->term_res_sel_data,
+				 VC4_HDMI_TX_PHY_CTL_2_TERM_RES_SELCK) |
+		   VC4_SET_FIELD(phy_get_vco_gain(vco_freq),
+				 VC4_HDMI_TX_PHY_CTL_2_VCO_GAIN));
+
+	HDMI_WRITE(HDMI_TX_PHY_CHANNEL_SWAP,
+		   VC4_SET_FIELD(variant->phy_lane_mapping[PHY_LANE_0],
+				 VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX0_OUT_SEL) |
+		   VC4_SET_FIELD(variant->phy_lane_mapping[PHY_LANE_1],
+				 VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX1_OUT_SEL) |
+		   VC4_SET_FIELD(variant->phy_lane_mapping[PHY_LANE_2],
+				 VC4_HDMI_TX_PHY_CHANNEL_SWAP_TX2_OUT_SEL) |
+		   VC4_SET_FIELD(variant->phy_lane_mapping[PHY_LANE_CK],
+				 VC4_HDMI_TX_PHY_CHANNEL_SWAP_TXCK_OUT_SEL));
+
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL,
+		   HDMI_READ(HDMI_TX_PHY_RESET_CTL) &
+		   ~(VC4_HDMI_TX_PHY_RESET_CTL_PLL_RESETB |
+		     VC4_HDMI_TX_PHY_RESET_CTL_PLLDIV_RESETB));
+
+	HDMI_WRITE(HDMI_TX_PHY_RESET_CTL,
+		   HDMI_READ(HDMI_TX_PHY_RESET_CTL) |
+		   VC4_HDMI_TX_PHY_RESET_CTL_PLL_RESETB |
+		   VC4_HDMI_TX_PHY_RESET_CTL_PLLDIV_RESETB);
+}
+
+void vc5_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi)
+{
+	vc5_hdmi_reset_phy(vc4_hdmi);
+}
+
+void vc5_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_POWERDOWN_CTL,
+		   HDMI_READ(HDMI_TX_PHY_POWERDOWN_CTL) &
+		   ~VC4_HDMI_TX_PHY_POWERDOWN_CTL_RNDGEN_PWRDN);
+}
+
+void vc5_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi)
+{
+	HDMI_WRITE(HDMI_TX_PHY_POWERDOWN_CTL,
+		   HDMI_READ(HDMI_TX_PHY_POWERDOWN_CTL) |
+		   VC4_HDMI_TX_PHY_POWERDOWN_CTL_RNDGEN_PWRDN);
+}
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi_regs.h b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h
new file mode 100644
index 000000000000..7c6b4818f245
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h
@@ -0,0 +1,442 @@
+#ifndef _VC4_HDMI_REGS_H_
+#define _VC4_HDMI_REGS_H_
+
+#include "vc4_hdmi.h"
+
+#define VC4_HDMI_PACKET_STRIDE			0x24
+
+enum vc4_hdmi_regs {
+	VC4_INVALID = 0,
+	VC4_HDMI,
+	VC4_HD,
+	VC5_CEC,
+	VC5_CSC,
+	VC5_DVP,
+	VC5_PHY,
+	VC5_RAM,
+	VC5_RM,
+};
+
+enum vc4_hdmi_field {
+	HDMI_AUDIO_PACKET_CONFIG,
+	HDMI_CEC_CNTRL_1,
+	HDMI_CEC_CNTRL_2,
+	HDMI_CEC_CNTRL_3,
+	HDMI_CEC_CNTRL_4,
+	HDMI_CEC_CNTRL_5,
+	HDMI_CEC_CPU_CLEAR,
+	HDMI_CEC_CPU_MASK_CLEAR,
+	HDMI_CEC_CPU_MASK_SET,
+	HDMI_CEC_CPU_MASK_STATUS,
+	HDMI_CEC_CPU_STATUS,
+
+	/*
+	 * Transmit data, first byte is low byte of the 32-bit reg.
+	 * MSB of each byte transmitted first.
+	 */
+	HDMI_CEC_RX_DATA_1,
+	HDMI_CEC_RX_DATA_2,
+	HDMI_CEC_RX_DATA_3,
+	HDMI_CEC_RX_DATA_4,
+	HDMI_CEC_TX_DATA_1,
+	HDMI_CEC_TX_DATA_2,
+	HDMI_CEC_TX_DATA_3,
+	HDMI_CEC_TX_DATA_4,
+	HDMI_CLOCK_STOP,
+	HDMI_CORE_REV,
+	HDMI_CRP_CFG,
+	HDMI_CSC_12_11,
+	HDMI_CSC_14_13,
+	HDMI_CSC_22_21,
+	HDMI_CSC_24_23,
+	HDMI_CSC_32_31,
+	HDMI_CSC_34_33,
+	HDMI_CSC_CTL,
+
+	/*
+	 * 20-bit fields containing CTS values to be transmitted if
+	 * !EXTERNAL_CTS_EN
+	 */
+	HDMI_CTS_0,
+	HDMI_CTS_1,
+	HDMI_DVP_CTL,
+	HDMI_FIFO_CTL,
+	HDMI_FRAME_COUNT,
+	HDMI_HORZA,
+	HDMI_HORZB,
+	HDMI_HOTPLUG,
+	HDMI_HOTPLUG_INT,
+
+	/*
+	 * 3 bits per field, where each field maps from that
+	 * corresponding MAI bus channel to the given HDMI channel.
+	 */
+	HDMI_MAI_CHANNEL_MAP,
+	HDMI_MAI_CONFIG,
+	HDMI_MAI_CTL,
+
+	/*
+	 * Register for DMAing in audio data to be transported over
+	 * the MAI bus to the Falcon core.
+	 */
+	HDMI_MAI_DATA,
+
+	/* Format header to be placed on the MAI data. Unused. */
+	HDMI_MAI_FMT,
+
+	/* Last received format word on the MAI bus. */
+	HDMI_MAI_FORMAT,
+	HDMI_MAI_SMP,
+	HDMI_MAI_THR,
+	HDMI_M_CTL,
+	HDMI_RAM_PACKET_CONFIG,
+	HDMI_RAM_PACKET_START,
+	HDMI_RAM_PACKET_STATUS,
+	HDMI_RM_CONTROL,
+	HDMI_RM_FORMAT,
+	HDMI_RM_OFFSET,
+	HDMI_SCHEDULER_CONTROL,
+	HDMI_SW_RESET_CONTROL,
+	HDMI_TX_PHY_CHANNEL_SWAP,
+	HDMI_TX_PHY_CLK_DIV,
+	HDMI_TX_PHY_CTL_0,
+	HDMI_TX_PHY_CTL_1,
+	HDMI_TX_PHY_CTL_2,
+	HDMI_TX_PHY_CTL_3,
+	HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1,
+	HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2,
+	HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4,
+	HDMI_TX_PHY_PLL_CFG,
+	HDMI_TX_PHY_PLL_CTL_0,
+	HDMI_TX_PHY_PLL_CTL_1,
+	HDMI_TX_PHY_POWERDOWN_CTL,
+	HDMI_TX_PHY_RESET_CTL,
+	HDMI_TX_PHY_TMDS_CLK_WORD_SEL,
+	HDMI_VEC_INTERFACE_XBAR,
+	HDMI_VERTA0,
+	HDMI_VERTA1,
+	HDMI_VERTB0,
+	HDMI_VERTB1,
+	HDMI_VID_CTL,
+};
+
+struct vc4_hdmi_register {
+	char *name;
+	enum vc4_hdmi_regs reg;
+	unsigned int offset;
+};
+
+#define _VC4_REG(_base, _reg, _offset)	\
+	[_reg] = {				\
+		.name = #_reg,			\
+		.reg = _base,			\
+		.offset = _offset,		\
+	}
+
+#define VC4_HD_REG(reg, offset)		_VC4_REG(VC4_HD, reg, offset)
+#define VC4_HDMI_REG(reg, offset)	_VC4_REG(VC4_HDMI, reg, offset)
+#define VC5_CEC_REG(reg, offset)	_VC4_REG(VC5_CEC, reg, offset)
+#define VC5_CSC_REG(reg, offset)	_VC4_REG(VC5_CSC, reg, offset)
+#define VC5_DVP_REG(reg, offset)	_VC4_REG(VC5_DVP, reg, offset)
+#define VC5_PHY_REG(reg, offset)	_VC4_REG(VC5_PHY, reg, offset)
+#define VC5_RAM_REG(reg, offset)	_VC4_REG(VC5_RAM, reg, offset)
+#define VC5_RM_REG(reg, offset)		_VC4_REG(VC5_RM, reg, offset)
+
+static const struct vc4_hdmi_register vc4_hdmi_fields[] = {
+	VC4_HD_REG(HDMI_M_CTL, 0x000c),
+	VC4_HD_REG(HDMI_MAI_CTL, 0x0014),
+	VC4_HD_REG(HDMI_MAI_THR, 0x0018),
+	VC4_HD_REG(HDMI_MAI_FMT, 0x001c),
+	VC4_HD_REG(HDMI_MAI_DATA, 0x0020),
+	VC4_HD_REG(HDMI_MAI_SMP, 0x002c),
+	VC4_HD_REG(HDMI_VID_CTL, 0x0038),
+	VC4_HD_REG(HDMI_CSC_CTL, 0x0040),
+	VC4_HD_REG(HDMI_CSC_12_11, 0x0044),
+	VC4_HD_REG(HDMI_CSC_14_13, 0x0048),
+	VC4_HD_REG(HDMI_CSC_22_21, 0x004c),
+	VC4_HD_REG(HDMI_CSC_24_23, 0x0050),
+	VC4_HD_REG(HDMI_CSC_32_31, 0x0054),
+	VC4_HD_REG(HDMI_CSC_34_33, 0x0058),
+	VC4_HD_REG(HDMI_FRAME_COUNT, 0x0068),
+
+	VC4_HDMI_REG(HDMI_CORE_REV, 0x0000),
+	VC4_HDMI_REG(HDMI_SW_RESET_CONTROL, 0x0004),
+	VC4_HDMI_REG(HDMI_HOTPLUG_INT, 0x0008),
+	VC4_HDMI_REG(HDMI_HOTPLUG, 0x000c),
+	VC4_HDMI_REG(HDMI_FIFO_CTL, 0x005c),
+	VC4_HDMI_REG(HDMI_MAI_CHANNEL_MAP, 0x0090),
+	VC4_HDMI_REG(HDMI_MAI_CONFIG, 0x0094),
+	VC4_HDMI_REG(HDMI_MAI_FORMAT, 0x0098),
+	VC4_HDMI_REG(HDMI_AUDIO_PACKET_CONFIG, 0x009c),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_CONFIG, 0x00a0),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_STATUS, 0x00a4),
+	VC4_HDMI_REG(HDMI_CRP_CFG, 0x00a8),
+	VC4_HDMI_REG(HDMI_CTS_0, 0x00ac),
+	VC4_HDMI_REG(HDMI_CTS_1, 0x00b0),
+	VC4_HDMI_REG(HDMI_SCHEDULER_CONTROL, 0x00c0),
+	VC4_HDMI_REG(HDMI_HORZA, 0x00c4),
+	VC4_HDMI_REG(HDMI_HORZB, 0x00c8),
+	VC4_HDMI_REG(HDMI_VERTA0, 0x00cc),
+	VC4_HDMI_REG(HDMI_VERTB0, 0x00d0),
+	VC4_HDMI_REG(HDMI_VERTA1, 0x00d4),
+	VC4_HDMI_REG(HDMI_VERTB1, 0x00d8),
+	VC4_HDMI_REG(HDMI_CEC_CNTRL_1, 0x00e8),
+	VC4_HDMI_REG(HDMI_CEC_CNTRL_2, 0x00ec),
+	VC4_HDMI_REG(HDMI_CEC_CNTRL_3, 0x00f0),
+	VC4_HDMI_REG(HDMI_CEC_CNTRL_4, 0x00f4),
+	VC4_HDMI_REG(HDMI_CEC_CNTRL_5, 0x00f8),
+	VC4_HDMI_REG(HDMI_CEC_TX_DATA_1, 0x00fc),
+	VC4_HDMI_REG(HDMI_CEC_TX_DATA_2, 0x0100),
+	VC4_HDMI_REG(HDMI_CEC_TX_DATA_3, 0x0104),
+	VC4_HDMI_REG(HDMI_CEC_TX_DATA_4, 0x0108),
+	VC4_HDMI_REG(HDMI_CEC_RX_DATA_1, 0x010c),
+	VC4_HDMI_REG(HDMI_CEC_RX_DATA_2, 0x0110),
+	VC4_HDMI_REG(HDMI_CEC_RX_DATA_3, 0x0114),
+	VC4_HDMI_REG(HDMI_CEC_RX_DATA_4, 0x0118),
+	VC4_HDMI_REG(HDMI_TX_PHY_RESET_CTL, 0x02c0),
+	VC4_HDMI_REG(HDMI_TX_PHY_CTL_0, 0x02c4),
+	VC4_HDMI_REG(HDMI_CEC_CPU_STATUS, 0x0340),
+	VC4_HDMI_REG(HDMI_CEC_CPU_CLEAR, 0x0348),
+	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_STATUS, 0x034c),
+	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_SET, 0x034c),
+	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_CLEAR, 0x0354),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_START, 0x0400),
+};
+
+static const struct vc4_hdmi_register vc5_hdmi_hdmi0_fields[] = {
+	VC4_HD_REG(HDMI_DVP_CTL, 0x0000),
+	VC4_HD_REG(HDMI_MAI_CTL, 0x0010),
+	VC4_HD_REG(HDMI_MAI_THR, 0x0014),
+	VC4_HD_REG(HDMI_MAI_FMT, 0x0018),
+	VC4_HD_REG(HDMI_MAI_DATA, 0x001c),
+	VC4_HD_REG(HDMI_MAI_SMP, 0x0020),
+	VC4_HD_REG(HDMI_VID_CTL, 0x0044),
+	VC4_HD_REG(HDMI_FRAME_COUNT, 0x0060),
+
+	VC4_HDMI_REG(HDMI_FIFO_CTL, 0x074),
+	VC4_HDMI_REG(HDMI_AUDIO_PACKET_CONFIG, 0x0b8),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_CONFIG, 0x0bc),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_STATUS, 0x0c4),
+	VC4_HDMI_REG(HDMI_CRP_CFG, 0x0c8),
+	VC4_HDMI_REG(HDMI_CTS_0, 0x0cc),
+	VC4_HDMI_REG(HDMI_CTS_1, 0x0d0),
+	VC4_HDMI_REG(HDMI_SCHEDULER_CONTROL, 0x0e0),
+	VC4_HDMI_REG(HDMI_HORZA, 0x0e4),
+	VC4_HDMI_REG(HDMI_HORZB, 0x0e8),
+	VC4_HDMI_REG(HDMI_VERTA0, 0x0ec),
+	VC4_HDMI_REG(HDMI_VERTB0, 0x0f0),
+	VC4_HDMI_REG(HDMI_VERTA1, 0x0f4),
+	VC4_HDMI_REG(HDMI_VERTB1, 0x0f8),
+	VC4_HDMI_REG(HDMI_MAI_CHANNEL_MAP, 0x09c),
+	VC4_HDMI_REG(HDMI_MAI_CONFIG, 0x0a0),
+	VC4_HDMI_REG(HDMI_HOTPLUG, 0x1a8),
+
+	VC5_DVP_REG(HDMI_CLOCK_STOP, 0x0bc),
+	VC5_DVP_REG(HDMI_VEC_INTERFACE_XBAR, 0x0f0),
+
+	VC5_PHY_REG(HDMI_TX_PHY_RESET_CTL, 0x000),
+	VC5_PHY_REG(HDMI_TX_PHY_POWERDOWN_CTL, 0x004),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_0, 0x008),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_1, 0x00c),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_2, 0x010),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_3, 0x014),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CTL_0, 0x01c),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CTL_1, 0x020),
+	VC5_PHY_REG(HDMI_TX_PHY_CLK_DIV, 0x028),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CFG, 0x034),
+	VC5_PHY_REG(HDMI_TX_PHY_TMDS_CLK_WORD_SEL, 0x044),
+	VC5_PHY_REG(HDMI_TX_PHY_CHANNEL_SWAP, 0x04c),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1, 0x050),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2, 0x054),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4, 0x05c),
+
+	VC5_RM_REG(HDMI_RM_CONTROL, 0x000),
+	VC5_RM_REG(HDMI_RM_OFFSET, 0x018),
+	VC5_RM_REG(HDMI_RM_FORMAT, 0x01c),
+
+	VC5_RAM_REG(HDMI_RAM_PACKET_START, 0x000),
+
+	VC5_CEC_REG(HDMI_CEC_CNTRL_1, 0x010),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_2, 0x014),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_3, 0x018),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_4, 0x01c),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_5, 0x020),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_1, 0x028),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_2, 0x02c),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_3, 0x030),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_4, 0x034),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_1, 0x038),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_2, 0x03c),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_3, 0x040),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_4, 0x044),
+
+	VC5_CSC_REG(HDMI_CSC_CTL, 0x000),
+	VC5_CSC_REG(HDMI_CSC_12_11, 0x004),
+	VC5_CSC_REG(HDMI_CSC_14_13, 0x008),
+	VC5_CSC_REG(HDMI_CSC_22_21, 0x00c),
+	VC5_CSC_REG(HDMI_CSC_24_23, 0x010),
+	VC5_CSC_REG(HDMI_CSC_32_31, 0x014),
+	VC5_CSC_REG(HDMI_CSC_34_33, 0x018),
+};
+
+static const struct vc4_hdmi_register vc5_hdmi_hdmi1_fields[] = {
+	VC4_HD_REG(HDMI_DVP_CTL, 0x0000),
+	VC4_HD_REG(HDMI_MAI_CTL, 0x0030),
+	VC4_HD_REG(HDMI_MAI_THR, 0x0034),
+	VC4_HD_REG(HDMI_MAI_FMT, 0x0038),
+	VC4_HD_REG(HDMI_MAI_DATA, 0x003c),
+	VC4_HD_REG(HDMI_MAI_SMP, 0x0040),
+	VC4_HD_REG(HDMI_VID_CTL, 0x0048),
+	VC4_HD_REG(HDMI_FRAME_COUNT, 0x0064),
+
+	VC4_HDMI_REG(HDMI_FIFO_CTL, 0x074),
+	VC4_HDMI_REG(HDMI_AUDIO_PACKET_CONFIG, 0x0b8),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_CONFIG, 0x0bc),
+	VC4_HDMI_REG(HDMI_RAM_PACKET_STATUS, 0x0c4),
+	VC4_HDMI_REG(HDMI_CRP_CFG, 0x0c8),
+	VC4_HDMI_REG(HDMI_CTS_0, 0x0cc),
+	VC4_HDMI_REG(HDMI_CTS_1, 0x0d0),
+	VC4_HDMI_REG(HDMI_SCHEDULER_CONTROL, 0x0e0),
+	VC4_HDMI_REG(HDMI_HORZA, 0x0e4),
+	VC4_HDMI_REG(HDMI_HORZB, 0x0e8),
+	VC4_HDMI_REG(HDMI_VERTA0, 0x0ec),
+	VC4_HDMI_REG(HDMI_VERTB0, 0x0f0),
+	VC4_HDMI_REG(HDMI_VERTA1, 0x0f4),
+	VC4_HDMI_REG(HDMI_VERTB1, 0x0f8),
+	VC4_HDMI_REG(HDMI_MAI_CHANNEL_MAP, 0x09c),
+	VC4_HDMI_REG(HDMI_MAI_CONFIG, 0x0a0),
+	VC4_HDMI_REG(HDMI_HOTPLUG, 0x1a8),
+
+	VC5_DVP_REG(HDMI_CLOCK_STOP, 0x0bc),
+	VC5_DVP_REG(HDMI_VEC_INTERFACE_XBAR, 0x0f0),
+
+	VC5_PHY_REG(HDMI_TX_PHY_RESET_CTL, 0x000),
+	VC5_PHY_REG(HDMI_TX_PHY_POWERDOWN_CTL, 0x004),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_0, 0x008),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_1, 0x00c),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_2, 0x010),
+	VC5_PHY_REG(HDMI_TX_PHY_CTL_3, 0x014),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CTL_0, 0x01c),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CTL_1, 0x020),
+	VC5_PHY_REG(HDMI_TX_PHY_CLK_DIV, 0x028),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CFG, 0x034),
+	VC5_PHY_REG(HDMI_TX_PHY_CHANNEL_SWAP, 0x04c),
+	VC5_PHY_REG(HDMI_TX_PHY_TMDS_CLK_WORD_SEL, 0x044),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_1, 0x050),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_2, 0x054),
+	VC5_PHY_REG(HDMI_TX_PHY_PLL_CALIBRATION_CONFIG_4, 0x05c),
+
+	VC5_RM_REG(HDMI_RM_CONTROL, 0x000),
+	VC5_RM_REG(HDMI_RM_OFFSET, 0x018),
+	VC5_RM_REG(HDMI_RM_FORMAT, 0x01c),
+
+	VC5_RAM_REG(HDMI_RAM_PACKET_START, 0x000),
+
+	VC5_CEC_REG(HDMI_CEC_CNTRL_1, 0x010),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_2, 0x014),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_3, 0x018),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_4, 0x01c),
+	VC5_CEC_REG(HDMI_CEC_CNTRL_5, 0x020),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_1, 0x028),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_2, 0x02c),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_3, 0x030),
+	VC5_CEC_REG(HDMI_CEC_TX_DATA_4, 0x034),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_1, 0x038),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_2, 0x03c),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_3, 0x040),
+	VC5_CEC_REG(HDMI_CEC_RX_DATA_4, 0x044),
+
+	VC5_CSC_REG(HDMI_CSC_CTL, 0x000),
+	VC5_CSC_REG(HDMI_CSC_12_11, 0x004),
+	VC5_CSC_REG(HDMI_CSC_14_13, 0x008),
+	VC5_CSC_REG(HDMI_CSC_22_21, 0x00c),
+	VC5_CSC_REG(HDMI_CSC_24_23, 0x010),
+	VC5_CSC_REG(HDMI_CSC_32_31, 0x014),
+	VC5_CSC_REG(HDMI_CSC_34_33, 0x018),
+};
+
+static inline
+void __iomem *__vc4_hdmi_get_field_base(struct vc4_hdmi *hdmi,
+					enum vc4_hdmi_regs reg)
+{
+	switch (reg) {
+	case VC4_HD:
+		return hdmi->hd_regs;
+
+	case VC4_HDMI:
+		return hdmi->hdmicore_regs;
+
+	case VC5_CSC:
+		return hdmi->csc_regs;
+
+	case VC5_CEC:
+		return hdmi->cec_regs;
+
+	case VC5_DVP:
+		return hdmi->dvp_regs;
+
+	case VC5_PHY:
+		return hdmi->phy_regs;
+
+	case VC5_RAM:
+		return hdmi->ram_regs;
+
+	case VC5_RM:
+		return hdmi->rm_regs;
+
+	default:
+		return NULL;
+	}
+
+	return NULL;
+}
+
+static inline u32 vc4_hdmi_read(struct vc4_hdmi *hdmi,
+				enum vc4_hdmi_field reg)
+{
+	const struct vc4_hdmi_register *field;
+	const struct vc4_hdmi_variant *variant = hdmi->variant;
+	void __iomem *base;
+
+	if (reg >= variant->num_registers) {
+		dev_warn(&hdmi->pdev->dev,
+			 "Invalid register ID %u\n", reg);
+		return 0;
+	}
+
+	field = &variant->registers[reg];
+	base = __vc4_hdmi_get_field_base(hdmi, field->reg);
+	if (!base) {
+		dev_warn(&hdmi->pdev->dev,
+			 "Unknown register ID %u\n", reg);
+		return 0;
+	}
+
+	return readl(base + field->offset);
+}
+#define HDMI_READ(reg)		vc4_hdmi_read(vc4_hdmi, reg)
+
+static inline void vc4_hdmi_write(struct vc4_hdmi *hdmi,
+				  enum vc4_hdmi_field reg,
+				  u32 value)
+{
+	const struct vc4_hdmi_register *field;
+	const struct vc4_hdmi_variant *variant = hdmi->variant;
+	void __iomem *base;
+
+	if (reg >= variant->num_registers) {
+		dev_warn(&hdmi->pdev->dev,
+			 "Invalid register ID %u\n", reg);
+		return;
+	}
+
+	field = &variant->registers[reg];
+	base = __vc4_hdmi_get_field_base(hdmi, field->reg);
+	if (!base)
+		return;
+
+	writel(value, base + field->offset);
+}
+#define HDMI_WRITE(reg, val)	vc4_hdmi_write(vc4_hdmi, reg, val)
+
+#endif /* _VC4_HDMI_REGS_H_ */
diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
index 2d2bf59c0503..4d0a833366ce 100644
--- a/drivers/gpu/drm/vc4/vc4_hvs.c
+++ b/drivers/gpu/drm/vc4/vc4_hvs.c
@@ -19,6 +19,8 @@
  * each CRTC.
  */
 
+#include <linux/bitfield.h>
+#include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/platform_device.h>
 
@@ -160,6 +162,7 @@ static void vc4_hvs_lut_load(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
 	u32 i;
 
 	/* The LUT memory is laid out with each HVS channel in order,
@@ -168,7 +171,7 @@ static void vc4_hvs_lut_load(struct drm_crtc *crtc)
 	 */
 	HVS_WRITE(SCALER_GAMADDR,
 		  SCALER_GAMADDR_AUTOINC |
-		  (vc4_crtc->channel * 3 * crtc->gamma_size));
+		  (vc4_state->assigned_channel * 3 * crtc->gamma_size));
 
 	for (i = 0; i < crtc->gamma_size; i++)
 		HVS_WRITE(SCALER_GAMDATA, vc4_crtc->lut_r[i]);
@@ -194,6 +197,135 @@ static void vc4_hvs_update_gamma_lut(struct drm_crtc *crtc)
 	vc4_hvs_lut_load(crtc);
 }
 
+int vc4_hvs_get_fifo_from_output(struct drm_device *dev, unsigned int output)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	u32 reg;
+	int ret;
+
+	if (!vc4->hvs->hvs5)
+		return output;
+
+	switch (output) {
+	case 0:
+		return 0;
+
+	case 1:
+		return 1;
+
+	case 2:
+		reg = HVS_READ(SCALER_DISPECTRL);
+		ret = FIELD_GET(SCALER_DISPECTRL_DSP2_MUX_MASK, reg);
+		if (ret == 0)
+			return 2;
+
+		return 0;
+
+	case 3:
+		reg = HVS_READ(SCALER_DISPCTRL);
+		ret = FIELD_GET(SCALER_DISPCTRL_DSP3_MUX_MASK, reg);
+		if (ret == 3)
+			return -EPIPE;
+
+		return ret;
+
+	case 4:
+		reg = HVS_READ(SCALER_DISPEOLN);
+		ret = FIELD_GET(SCALER_DISPEOLN_DSP4_MUX_MASK, reg);
+		if (ret == 3)
+			return -EPIPE;
+
+		return ret;
+
+	case 5:
+		reg = HVS_READ(SCALER_DISPDITHER);
+		ret = FIELD_GET(SCALER_DISPDITHER_DSP5_MUX_MASK, reg);
+		if (ret == 3)
+			return -EPIPE;
+
+		return ret;
+
+	default:
+		return -EPIPE;
+	}
+}
+
+static int vc4_hvs_init_channel(struct vc4_dev *vc4, struct drm_crtc *crtc,
+				struct drm_display_mode *mode, bool oneshot)
+{
+	struct vc4_crtc_state *vc4_crtc_state = to_vc4_crtc_state(crtc->state);
+	unsigned int chan = vc4_crtc_state->assigned_channel;
+	bool interlace = mode->flags & DRM_MODE_FLAG_INTERLACE;
+	u32 dispbkgndx;
+	u32 dispctrl;
+
+	HVS_WRITE(SCALER_DISPCTRLX(chan), 0);
+	HVS_WRITE(SCALER_DISPCTRLX(chan), SCALER_DISPCTRLX_RESET);
+	HVS_WRITE(SCALER_DISPCTRLX(chan), 0);
+
+	/* Turn on the scaler, which will wait for vstart to start
+	 * compositing.
+	 * When feeding the transposer, we should operate in oneshot
+	 * mode.
+	 */
+	dispctrl = SCALER_DISPCTRLX_ENABLE;
+
+	if (!vc4->hvs->hvs5)
+		dispctrl |= VC4_SET_FIELD(mode->hdisplay,
+					  SCALER_DISPCTRLX_WIDTH) |
+			    VC4_SET_FIELD(mode->vdisplay,
+					  SCALER_DISPCTRLX_HEIGHT) |
+			    (oneshot ? SCALER_DISPCTRLX_ONESHOT : 0);
+	else
+		dispctrl |= VC4_SET_FIELD(mode->hdisplay,
+					  SCALER5_DISPCTRLX_WIDTH) |
+			    VC4_SET_FIELD(mode->vdisplay,
+					  SCALER5_DISPCTRLX_HEIGHT) |
+			    (oneshot ? SCALER5_DISPCTRLX_ONESHOT : 0);
+
+	HVS_WRITE(SCALER_DISPCTRLX(chan), dispctrl);
+
+	dispbkgndx = HVS_READ(SCALER_DISPBKGNDX(chan));
+	dispbkgndx &= ~SCALER_DISPBKGND_GAMMA;
+	dispbkgndx &= ~SCALER_DISPBKGND_INTERLACE;
+
+	HVS_WRITE(SCALER_DISPBKGNDX(chan), dispbkgndx |
+		  SCALER_DISPBKGND_AUTOHS |
+		  ((!vc4->hvs->hvs5) ? SCALER_DISPBKGND_GAMMA : 0) |
+		  (interlace ? SCALER_DISPBKGND_INTERLACE : 0));
+
+	/* Reload the LUT, since the SRAMs would have been disabled if
+	 * all CRTCs had SCALER_DISPBKGND_GAMMA unset at once.
+	 */
+	vc4_hvs_lut_load(crtc);
+
+	return 0;
+}
+
+void vc4_hvs_stop_channel(struct drm_device *dev, unsigned int chan)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	if (HVS_READ(SCALER_DISPCTRLX(chan)) & SCALER_DISPCTRLX_ENABLE)
+		return;
+
+	HVS_WRITE(SCALER_DISPCTRLX(chan),
+		  HVS_READ(SCALER_DISPCTRLX(chan)) | SCALER_DISPCTRLX_RESET);
+	HVS_WRITE(SCALER_DISPCTRLX(chan),
+		  HVS_READ(SCALER_DISPCTRLX(chan)) & ~SCALER_DISPCTRLX_ENABLE);
+
+	/* Once we leave, the scaler should be disabled and its fifo empty. */
+	WARN_ON_ONCE(HVS_READ(SCALER_DISPCTRLX(chan)) & SCALER_DISPCTRLX_RESET);
+
+	WARN_ON_ONCE(VC4_GET_FIELD(HVS_READ(SCALER_DISPSTATX(chan)),
+				   SCALER_DISPSTATX_MODE) !=
+		     SCALER_DISPSTATX_MODE_DISABLED);
+
+	WARN_ON_ONCE((HVS_READ(SCALER_DISPSTATX(chan)) &
+		      (SCALER_DISPSTATX_FULL | SCALER_DISPSTATX_EMPTY)) !=
+		     SCALER_DISPSTATX_EMPTY);
+}
+
 int vc4_hvs_atomic_check(struct drm_crtc *crtc,
 			 struct drm_crtc_state *state)
 {
@@ -248,12 +380,12 @@ static void vc4_hvs_update_dlist(struct drm_crtc *crtc)
 			crtc->state->event = NULL;
 		}
 
-		HVS_WRITE(SCALER_DISPLISTX(vc4_crtc->channel),
+		HVS_WRITE(SCALER_DISPLISTX(vc4_state->assigned_channel),
 			  vc4_state->mm.start);
 
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 	} else {
-		HVS_WRITE(SCALER_DISPLISTX(vc4_crtc->channel),
+		HVS_WRITE(SCALER_DISPLISTX(vc4_state->assigned_channel),
 			  vc4_state->mm.start);
 	}
 }
@@ -263,59 +395,22 @@ void vc4_hvs_atomic_enable(struct drm_crtc *crtc,
 {
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
 	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
 	struct drm_display_mode *mode = &crtc->state->adjusted_mode;
 	bool oneshot = vc4_state->feed_txp;
-	u32 dispctrl;
 
 	vc4_hvs_update_dlist(crtc);
-
-	/* Turn on the scaler, which will wait for vstart to start
-	 * compositing.
-	 * When feeding the transposer, we should operate in oneshot
-	 * mode.
-	 */
-	dispctrl = SCALER_DISPCTRLX_ENABLE;
-	dispctrl |= VC4_SET_FIELD(mode->hdisplay,
-				  SCALER_DISPCTRLX_WIDTH) |
-		    VC4_SET_FIELD(mode->vdisplay,
-				  SCALER_DISPCTRLX_HEIGHT) |
-		    (oneshot ? SCALER_DISPCTRLX_ONESHOT : 0);
-
-	HVS_WRITE(SCALER_DISPCTRLX(vc4_crtc->channel), dispctrl);
+	vc4_hvs_init_channel(vc4, crtc, mode, oneshot);
 }
 
 void vc4_hvs_atomic_disable(struct drm_crtc *crtc,
 			    struct drm_crtc_state *old_state)
 {
 	struct drm_device *dev = crtc->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
-	u32 chan = vc4_crtc->channel;
+	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(old_state);
+	unsigned int chan = vc4_state->assigned_channel;
 
-	if (HVS_READ(SCALER_DISPCTRLX(chan)) &
-	    SCALER_DISPCTRLX_ENABLE) {
-		HVS_WRITE(SCALER_DISPCTRLX(chan),
-			  SCALER_DISPCTRLX_RESET);
-
-		/* While the docs say that reset is self-clearing, it
-		 * seems it doesn't actually.
-		 */
-		HVS_WRITE(SCALER_DISPCTRLX(chan), 0);
-	}
-
-	/* Once we leave, the scaler should be disabled and its fifo empty. */
-
-	WARN_ON_ONCE(HVS_READ(SCALER_DISPCTRLX(chan)) & SCALER_DISPCTRLX_RESET);
-
-	WARN_ON_ONCE(VC4_GET_FIELD(HVS_READ(SCALER_DISPSTATX(chan)),
-				   SCALER_DISPSTATX_MODE) !=
-		     SCALER_DISPSTATX_MODE_DISABLED);
-
-	WARN_ON_ONCE((HVS_READ(SCALER_DISPSTATX(chan)) &
-		      (SCALER_DISPSTATX_FULL | SCALER_DISPSTATX_EMPTY)) !=
-		     SCALER_DISPSTATX_EMPTY);
+	vc4_hvs_stop_channel(dev, chan);
 }
 
 void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
@@ -323,7 +418,6 @@ void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
 {
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
 	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
 	struct drm_plane *plane;
 	struct vc4_plane_state *vc4_plane_state;
@@ -365,8 +459,8 @@ void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
 		/* This sets a black background color fill, as is the case
 		 * with other DRM drivers.
 		 */
-		HVS_WRITE(SCALER_DISPBKGNDX(vc4_crtc->channel),
-			  HVS_READ(SCALER_DISPBKGNDX(vc4_crtc->channel)) |
+		HVS_WRITE(SCALER_DISPBKGNDX(vc4_state->assigned_channel),
+			  HVS_READ(SCALER_DISPBKGNDX(vc4_state->assigned_channel)) |
 			  SCALER_DISPBKGND_FILL);
 
 	/* Only update DISPLIST if the CRTC was already running and is not
@@ -380,7 +474,7 @@ void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
 		vc4_hvs_update_dlist(crtc);
 
 	if (crtc->state->color_mgmt_changed) {
-		u32 dispbkgndx = HVS_READ(SCALER_DISPBKGNDX(vc4_crtc->channel));
+		u32 dispbkgndx = HVS_READ(SCALER_DISPBKGNDX(vc4_state->assigned_channel));
 
 		if (crtc->state->gamma_lut) {
 			vc4_hvs_update_gamma_lut(crtc);
@@ -392,7 +486,7 @@ void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
 			 */
 			dispbkgndx &= ~SCALER_DISPBKGND_GAMMA;
 		}
-		HVS_WRITE(SCALER_DISPBKGNDX(vc4_crtc->channel), dispbkgndx);
+		HVS_WRITE(SCALER_DISPBKGNDX(vc4_state->assigned_channel), dispbkgndx);
 	}
 
 	if (debug_dump_regs) {
@@ -401,50 +495,6 @@ void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
 	}
 }
 
-void vc4_hvs_mode_set_nofb(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
-	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
-	struct drm_display_mode *mode = &crtc->state->adjusted_mode;
-	bool interlace = mode->flags & DRM_MODE_FLAG_INTERLACE;
-
-	if (vc4_crtc->data->hvs_channel == 2) {
-		u32 dispctrl;
-		u32 dsp3_mux;
-
-		/*
-		 * SCALER_DISPCTRL_DSP3 = X, where X < 2 means 'connect DSP3 to
-		 * FIFO X'.
-		 * SCALER_DISPCTRL_DSP3 = 3 means 'disable DSP 3'.
-		 *
-		 * DSP3 is connected to FIFO2 unless the transposer is
-		 * enabled. In this case, FIFO 2 is directly accessed by the
-		 * TXP IP, and we need to disable the FIFO2 -> pixelvalve1
-		 * route.
-		 */
-		if (vc4_state->feed_txp)
-			dsp3_mux = VC4_SET_FIELD(3, SCALER_DISPCTRL_DSP3_MUX);
-		else
-			dsp3_mux = VC4_SET_FIELD(2, SCALER_DISPCTRL_DSP3_MUX);
-
-		dispctrl = HVS_READ(SCALER_DISPCTRL) &
-			   ~SCALER_DISPCTRL_DSP3_MUX_MASK;
-		HVS_WRITE(SCALER_DISPCTRL, dispctrl | dsp3_mux);
-	}
-
-	HVS_WRITE(SCALER_DISPBKGNDX(vc4_crtc->channel),
-		  SCALER_DISPBKGND_AUTOHS |
-		  SCALER_DISPBKGND_GAMMA |
-		  (interlace ? SCALER_DISPBKGND_INTERLACE : 0));
-
-	/* Reload the LUT, since the SRAMs would have been disabled if
-	 * all CRTCs had SCALER_DISPBKGND_GAMMA unset at once.
-	 */
-	vc4_hvs_lut_load(crtc);
-}
-
 void vc4_hvs_mask_underrun(struct drm_device *dev, int channel)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
@@ -521,6 +571,9 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 
 	hvs->pdev = pdev;
 
+	if (of_device_is_compatible(pdev->dev.of_node, "brcm,bcm2711-hvs"))
+		hvs->hvs5 = true;
+
 	hvs->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(hvs->regs))
 		return PTR_ERR(hvs->regs);
@@ -529,7 +582,24 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	hvs->regset.regs = hvs_regs;
 	hvs->regset.nregs = ARRAY_SIZE(hvs_regs);
 
-	hvs->dlist = hvs->regs + SCALER_DLIST_START;
+	if (hvs->hvs5) {
+		hvs->core_clk = devm_clk_get(&pdev->dev, NULL);
+		if (IS_ERR(hvs->core_clk)) {
+			dev_err(&pdev->dev, "Couldn't get core clock\n");
+			return PTR_ERR(hvs->core_clk);
+		}
+
+		ret = clk_prepare_enable(hvs->core_clk);
+		if (ret) {
+			dev_err(&pdev->dev, "Couldn't enable the core clock\n");
+			return ret;
+		}
+	}
+
+	if (!hvs->hvs5)
+		hvs->dlist = hvs->regs + SCALER_DLIST_START;
+	else
+		hvs->dlist = hvs->regs + SCALER5_DLIST_START;
 
 	spin_lock_init(&hvs->mm_lock);
 
@@ -547,7 +617,12 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	 * between planes when they don't overlap on the screen, but
 	 * for now we just allocate globally.
 	 */
-	drm_mm_init(&hvs->lbm_mm, 0, 96 * 1024);
+	if (!hvs->hvs5)
+		/* 96kB */
+		drm_mm_init(&hvs->lbm_mm, 0, 96 * 1024);
+	else
+		/* 70k words */
+		drm_mm_init(&hvs->lbm_mm, 0, 70 * 2 * 1024);
 
 	/* Upload filter kernels.  We only have the one for now, so we
 	 * keep it around for the lifetime of the driver.
@@ -605,6 +680,7 @@ static void vc4_hvs_unbind(struct device *dev, struct device *master,
 {
 	struct drm_device *drm = dev_get_drvdata(master);
 	struct vc4_dev *vc4 = drm->dev_private;
+	struct vc4_hvs *hvs = vc4->hvs;
 
 	if (drm_mm_node_allocated(&vc4->hvs->mitchell_netravali_filter))
 		drm_mm_remove_node(&vc4->hvs->mitchell_netravali_filter);
@@ -612,6 +688,8 @@ static void vc4_hvs_unbind(struct device *dev, struct device *master,
 	drm_mm_takedown(&vc4->hvs->dlist_mm);
 	drm_mm_takedown(&vc4->hvs->lbm_mm);
 
+	clk_disable_unprepare(hvs->core_clk);
+
 	vc4->hvs = NULL;
 }
 
@@ -632,6 +710,7 @@ static int vc4_hvs_dev_remove(struct platform_device *pdev)
 }
 
 static const struct of_device_id vc4_hvs_dt_match[] = {
+	{ .compatible = "brcm,bcm2711-hvs" },
 	{ .compatible = "brcm,bcm2835-hvs" },
 	{}
 };
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index 08318e69061b..149825ff5df8 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -11,6 +11,8 @@
  * crtc, HDMI encoder).
  */
 
+#include <linux/clk.h>
+
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
@@ -144,22 +146,130 @@ vc4_ctm_commit(struct vc4_dev *vc4, struct drm_atomic_state *state)
 		  VC4_SET_FIELD(ctm_state->fifo, SCALER_OLEDOFFS_DISPFIFO));
 }
 
+static void vc4_hvs_pv_muxing_commit(struct vc4_dev *vc4,
+				     struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *crtc_state;
+	struct drm_crtc *crtc;
+	unsigned int i;
+
+	for_each_new_crtc_in_state(state, crtc, crtc_state, i) {
+		struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc_state);
+		u32 dispctrl;
+		u32 dsp3_mux;
+
+		if (!crtc_state->active)
+			continue;
+
+		if (vc4_state->assigned_channel != 2)
+			continue;
+
+		/*
+		 * SCALER_DISPCTRL_DSP3 = X, where X < 2 means 'connect DSP3 to
+		 * FIFO X'.
+		 * SCALER_DISPCTRL_DSP3 = 3 means 'disable DSP 3'.
+		 *
+		 * DSP3 is connected to FIFO2 unless the transposer is
+		 * enabled. In this case, FIFO 2 is directly accessed by the
+		 * TXP IP, and we need to disable the FIFO2 -> pixelvalve1
+		 * route.
+		 */
+		if (vc4_state->feed_txp)
+			dsp3_mux = VC4_SET_FIELD(3, SCALER_DISPCTRL_DSP3_MUX);
+		else
+			dsp3_mux = VC4_SET_FIELD(2, SCALER_DISPCTRL_DSP3_MUX);
+
+		dispctrl = HVS_READ(SCALER_DISPCTRL) &
+			   ~SCALER_DISPCTRL_DSP3_MUX_MASK;
+		HVS_WRITE(SCALER_DISPCTRL, dispctrl | dsp3_mux);
+	}
+}
+
+static void vc5_hvs_pv_muxing_commit(struct vc4_dev *vc4,
+				     struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *crtc_state;
+	struct drm_crtc *crtc;
+	unsigned char dsp2_mux = 0;
+	unsigned char dsp3_mux = 3;
+	unsigned char dsp4_mux = 3;
+	unsigned char dsp5_mux = 3;
+	unsigned int i;
+	u32 reg;
+
+	for_each_new_crtc_in_state(state, crtc, crtc_state, i) {
+		struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc_state);
+		struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+
+		if (!crtc_state->active)
+			continue;
+
+		switch (vc4_crtc->data->hvs_output) {
+		case 2:
+			dsp2_mux = (vc4_state->assigned_channel == 2) ? 0 : 1;
+			break;
+
+		case 3:
+			dsp3_mux = vc4_state->assigned_channel;
+			break;
+
+		case 4:
+			dsp4_mux = vc4_state->assigned_channel;
+			break;
+
+		case 5:
+			dsp5_mux = vc4_state->assigned_channel;
+			break;
+
+		default:
+			break;
+		}
+	}
+
+	reg = HVS_READ(SCALER_DISPECTRL);
+	HVS_WRITE(SCALER_DISPECTRL,
+		  (reg & ~SCALER_DISPECTRL_DSP2_MUX_MASK) |
+		  VC4_SET_FIELD(dsp2_mux, SCALER_DISPECTRL_DSP2_MUX));
+
+	reg = HVS_READ(SCALER_DISPCTRL);
+	HVS_WRITE(SCALER_DISPCTRL,
+		  (reg & ~SCALER_DISPCTRL_DSP3_MUX_MASK) |
+		  VC4_SET_FIELD(dsp3_mux, SCALER_DISPCTRL_DSP3_MUX));
+
+	reg = HVS_READ(SCALER_DISPEOLN);
+	HVS_WRITE(SCALER_DISPEOLN,
+		  (reg & ~SCALER_DISPEOLN_DSP4_MUX_MASK) |
+		  VC4_SET_FIELD(dsp4_mux, SCALER_DISPEOLN_DSP4_MUX));
+
+	reg = HVS_READ(SCALER_DISPDITHER);
+	HVS_WRITE(SCALER_DISPDITHER,
+		  (reg & ~SCALER_DISPDITHER_DSP5_MUX_MASK) |
+		  VC4_SET_FIELD(dsp5_mux, SCALER_DISPDITHER_DSP5_MUX));
+}
+
 static void
 vc4_atomic_complete_commit(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc *vc4_crtc;
+	struct vc4_hvs *hvs = vc4->hvs;
+	struct drm_crtc_state *new_crtc_state;
+	struct drm_crtc *crtc;
 	int i;
 
-	for (i = 0; i < dev->mode_config.num_crtc; i++) {
-		if (!state->crtcs[i].ptr || !state->crtcs[i].commit)
+	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
+		struct vc4_crtc_state *vc4_crtc_state;
+
+		if (!new_crtc_state->commit)
 			continue;
 
-		vc4_crtc = to_vc4_crtc(state->crtcs[i].ptr);
-		vc4_hvs_mask_underrun(dev, vc4_crtc->channel);
+		vc4_crtc_state = to_vc4_crtc_state(new_crtc_state);
+		vc4_hvs_mask_underrun(dev, vc4_crtc_state->assigned_channel);
 	}
 
+	if (vc4->hvs->hvs5)
+		clk_set_min_rate(hvs->core_clk, 500000000);
+
 	drm_atomic_helper_wait_for_fences(dev, state, false);
 
 	drm_atomic_helper_wait_for_dependencies(state);
@@ -168,6 +278,11 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 
 	vc4_ctm_commit(vc4, state);
 
+	if (vc4->hvs->hvs5)
+		vc5_hvs_pv_muxing_commit(vc4, state);
+	else
+		vc4_hvs_pv_muxing_commit(vc4, state);
+
 	drm_atomic_helper_commit_planes(dev, state, 0);
 
 	drm_atomic_helper_commit_modeset_enables(dev, state);
@@ -182,6 +297,9 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 
 	drm_atomic_helper_commit_cleanup_done(state);
 
+	if (vc4->hvs->hvs5)
+		clk_set_min_rate(hvs->core_clk, 0);
+
 	drm_atomic_state_put(state);
 
 	up(&vc4->async_modeset);
@@ -374,8 +492,11 @@ vc4_ctm_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 
 		/* CTM is being enabled or the matrix changed. */
 		if (new_crtc_state->ctm) {
+			struct vc4_crtc_state *vc4_crtc_state =
+				to_vc4_crtc_state(new_crtc_state);
+
 			/* fifo is 1-based since 0 disables CTM. */
-			int fifo = to_vc4_crtc(crtc)->channel + 1;
+			int fifo = vc4_crtc_state->assigned_channel + 1;
 
 			/* Check userland isn't trying to turn on CTM for more
 			 * than one CRTC at a time.
@@ -415,6 +536,9 @@ static int vc4_load_tracker_atomic_check(struct drm_atomic_state *state)
 	struct drm_plane *plane;
 	int i;
 
+	if (!vc4->load_tracker_available)
+		return 0;
+
 	priv_state = drm_atomic_get_private_obj_state(state,
 						      &vc4->load_tracker);
 	if (IS_ERR(priv_state))
@@ -485,10 +609,87 @@ static const struct drm_private_state_funcs vc4_load_tracker_state_funcs = {
 	.atomic_destroy_state = vc4_load_tracker_destroy_state,
 };
 
+#define NUM_OUTPUTS  6
+#define NUM_CHANNELS 3
+
 static int
 vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 {
-	int ret;
+	unsigned long unassigned_channels = GENMASK(NUM_CHANNELS - 1, 0);
+	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+	struct drm_crtc *crtc;
+	int i, ret;
+
+	/*
+	 * Since the HVS FIFOs are shared across all the pixelvalves and
+	 * the TXP (and thus all the CRTCs), we need to pull the current
+	 * state of all the enabled CRTCs so that an update to a single
+	 * CRTC still keeps the previous FIFOs enabled and assigned to
+	 * the same CRTCs, instead of evaluating only the CRTC being
+	 * modified.
+	 */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct drm_crtc_state *crtc_state;
+
+		if (!crtc->state->enable)
+			continue;
+
+		crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+	}
+
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		struct vc4_crtc_state *new_vc4_crtc_state =
+			to_vc4_crtc_state(new_crtc_state);
+		struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+		unsigned int matching_channels;
+
+		if (old_crtc_state->enable && !new_crtc_state->enable)
+			new_vc4_crtc_state->assigned_channel = VC4_HVS_CHANNEL_DISABLED;
+
+		if (!new_crtc_state->enable)
+			continue;
+
+		if (new_vc4_crtc_state->assigned_channel != VC4_HVS_CHANNEL_DISABLED) {
+			unassigned_channels &= ~BIT(new_vc4_crtc_state->assigned_channel);
+			continue;
+		}
+
+		/*
+		 * The problem we have to solve here is that we have
+		 * up to 7 encoders, connected to up to 6 CRTCs.
+		 *
+		 * Those CRTCs, depending on the instance, can be
+		 * routed to 1, 2 or 3 HVS FIFOs, and we need to set
+		 * the change the muxing between FIFOs and outputs in
+		 * the HVS accordingly.
+		 *
+		 * It would be pretty hard to come up with an
+		 * algorithm that would generically solve
+		 * this. However, the current routing trees we support
+		 * allow us to simplify a bit the problem.
+		 *
+		 * Indeed, with the current supported layouts, if we
+		 * try to assign in the ascending crtc index order the
+		 * FIFOs, we can't fall into the situation where an
+		 * earlier CRTC that had multiple routes is assigned
+		 * one that was the only option for a later CRTC.
+		 *
+		 * If the layout changes and doesn't give us that in
+		 * the future, we will need to have something smarter,
+		 * but it works so far.
+		 */
+		matching_channels = unassigned_channels & vc4_crtc->data->hvs_available_channels;
+		if (matching_channels) {
+			unsigned int channel = ffs(matching_channels) - 1;
+
+			new_vc4_crtc_state->assigned_channel = channel;
+			unassigned_channels &= ~BIT(channel);
+		} else {
+			return -EINVAL;
+		}
+	}
 
 	ret = vc4_ctm_atomic_check(dev, state);
 	if (ret < 0)
@@ -512,12 +713,18 @@ int vc4_kms_load(struct drm_device *dev)
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_ctm_state *ctm_state;
 	struct vc4_load_tracker_state *load_state;
+	bool is_vc5 = of_device_is_compatible(dev->dev->of_node,
+					      "brcm,bcm2711-vc5");
 	int ret;
 
-	/* Start with the load tracker enabled. Can be disabled through the
-	 * debugfs load_tracker file.
-	 */
-	vc4->load_tracker_enabled = true;
+	if (!is_vc5) {
+		vc4->load_tracker_available = true;
+
+		/* Start with the load tracker enabled. Can be
+		 * disabled through the debugfs load_tracker file.
+		 */
+		vc4->load_tracker_enabled = true;
+	}
 
 	sema_init(&vc4->async_modeset, 1);
 
@@ -531,8 +738,14 @@ int vc4_kms_load(struct drm_device *dev)
 		return ret;
 	}
 
-	dev->mode_config.max_width = 2048;
-	dev->mode_config.max_height = 2048;
+	if (is_vc5) {
+		dev->mode_config.max_width = 7680;
+		dev->mode_config.max_height = 7680;
+	} else {
+		dev->mode_config.max_width = 2048;
+		dev->mode_config.max_height = 2048;
+	}
+
 	dev->mode_config.funcs = &vc4_mode_funcs;
 	dev->mode_config.preferred_depth = 24;
 	dev->mode_config.async_page_flip = true;
@@ -547,14 +760,17 @@ int vc4_kms_load(struct drm_device *dev)
 	drm_atomic_private_obj_init(dev, &vc4->ctm_manager, &ctm_state->base,
 				    &vc4_ctm_state_funcs);
 
-	load_state = kzalloc(sizeof(*load_state), GFP_KERNEL);
-	if (!load_state) {
-		drm_atomic_private_obj_fini(&vc4->ctm_manager);
-		return -ENOMEM;
-	}
+	if (vc4->load_tracker_available) {
+		load_state = kzalloc(sizeof(*load_state), GFP_KERNEL);
+		if (!load_state) {
+			drm_atomic_private_obj_fini(&vc4->ctm_manager);
+			return -ENOMEM;
+		}
 
-	drm_atomic_private_obj_init(dev, &vc4->load_tracker, &load_state->base,
-				    &vc4_load_tracker_state_funcs);
+		drm_atomic_private_obj_init(dev, &vc4->load_tracker,
+					    &load_state->base,
+					    &vc4_load_tracker_state_funcs);
+	}
 
 	drm_mode_config_reset(dev);
 
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index d040d9f12c6d..6b39cc2ca18d 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -32,45 +32,60 @@ static const struct hvs_format {
 	u32 drm; /* DRM_FORMAT_* */
 	u32 hvs; /* HVS_FORMAT_* */
 	u32 pixel_order;
+	u32 pixel_order_hvs5;
 } hvs_formats[] = {
 	{
-		.drm = DRM_FORMAT_XRGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.drm = DRM_FORMAT_XRGB8888,
+		.hvs = HVS_PIXEL_FORMAT_RGBA8888,
 		.pixel_order = HVS_PIXEL_ORDER_ABGR,
+		.pixel_order_hvs5 = HVS_PIXEL_ORDER_ARGB,
 	},
 	{
-		.drm = DRM_FORMAT_ARGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.drm = DRM_FORMAT_ARGB8888,
+		.hvs = HVS_PIXEL_FORMAT_RGBA8888,
 		.pixel_order = HVS_PIXEL_ORDER_ABGR,
+		.pixel_order_hvs5 = HVS_PIXEL_ORDER_ARGB,
 	},
 	{
-		.drm = DRM_FORMAT_ABGR8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.drm = DRM_FORMAT_ABGR8888,
+		.hvs = HVS_PIXEL_FORMAT_RGBA8888,
 		.pixel_order = HVS_PIXEL_ORDER_ARGB,
+		.pixel_order_hvs5 = HVS_PIXEL_ORDER_ABGR,
 	},
 	{
-		.drm = DRM_FORMAT_XBGR8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.drm = DRM_FORMAT_XBGR8888,
+		.hvs = HVS_PIXEL_FORMAT_RGBA8888,
 		.pixel_order = HVS_PIXEL_ORDER_ARGB,
+		.pixel_order_hvs5 = HVS_PIXEL_ORDER_ABGR,
 	},
 	{
-		.drm = DRM_FORMAT_RGB565, .hvs = HVS_PIXEL_FORMAT_RGB565,
+		.drm = DRM_FORMAT_RGB565,
+		.hvs = HVS_PIXEL_FORMAT_RGB565,
 		.pixel_order = HVS_PIXEL_ORDER_XRGB,
 	},
 	{
-		.drm = DRM_FORMAT_BGR565, .hvs = HVS_PIXEL_FORMAT_RGB565,
+		.drm = DRM_FORMAT_BGR565,
+		.hvs = HVS_PIXEL_FORMAT_RGB565,
 		.pixel_order = HVS_PIXEL_ORDER_XBGR,
 	},
 	{
-		.drm = DRM_FORMAT_ARGB1555, .hvs = HVS_PIXEL_FORMAT_RGBA5551,
+		.drm = DRM_FORMAT_ARGB1555,
+		.hvs = HVS_PIXEL_FORMAT_RGBA5551,
 		.pixel_order = HVS_PIXEL_ORDER_ABGR,
 	},
 	{
-		.drm = DRM_FORMAT_XRGB1555, .hvs = HVS_PIXEL_FORMAT_RGBA5551,
+		.drm = DRM_FORMAT_XRGB1555,
+		.hvs = HVS_PIXEL_FORMAT_RGBA5551,
 		.pixel_order = HVS_PIXEL_ORDER_ABGR,
 	},
 	{
-		.drm = DRM_FORMAT_RGB888, .hvs = HVS_PIXEL_FORMAT_RGB888,
+		.drm = DRM_FORMAT_RGB888,
+		.hvs = HVS_PIXEL_FORMAT_RGB888,
 		.pixel_order = HVS_PIXEL_ORDER_XRGB,
 	},
 	{
-		.drm = DRM_FORMAT_BGR888, .hvs = HVS_PIXEL_FORMAT_RGB888,
+		.drm = DRM_FORMAT_BGR888,
+		.hvs = HVS_PIXEL_FORMAT_RGB888,
 		.pixel_order = HVS_PIXEL_ORDER_XBGR,
 	},
 	{
@@ -422,10 +437,7 @@ static void vc4_write_ppf(struct vc4_plane_state *vc4_state, u32 src, u32 dst)
 static u32 vc4_lbm_size(struct drm_plane_state *state)
 {
 	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
-	/* This is the worst case number.  One of the two sizes will
-	 * be used depending on the scaling configuration.
-	 */
-	u32 pix_per_line = max(vc4_state->src_w[0], (u32)vc4_state->crtc_w);
+	u32 pix_per_line;
 	u32 lbm;
 
 	/* LBM is not needed when there's no vertical scaling. */
@@ -433,6 +445,18 @@ static u32 vc4_lbm_size(struct drm_plane_state *state)
 	    vc4_state->y_scaling[1] == VC4_SCALING_NONE)
 		return 0;
 
+	/*
+	 * This can be further optimized in the RGB/YUV444 case if the PPF
+	 * decimation factor is between 0.5 and 1.0 by using crtc_w.
+	 *
+	 * It's not an issue though, since in that case since src_w[0] is going
+	 * to be greater than or equal to crtc_w.
+	 */
+	if (vc4_state->x_scaling[0] == VC4_SCALING_TPZ)
+		pix_per_line = vc4_state->crtc_w;
+	else
+		pix_per_line = vc4_state->src_w[0];
+
 	if (!vc4_state->is_yuv) {
 		if (vc4_state->y_scaling[0] == VC4_SCALING_TPZ)
 			lbm = pix_per_line * 8;
@@ -492,6 +516,11 @@ static void vc4_plane_calc_load(struct drm_plane_state *state)
 	struct vc4_plane_state *vc4_state;
 	struct drm_crtc_state *crtc_state;
 	unsigned int vscale_factor;
+	struct vc4_dev *vc4;
+
+	vc4 = to_vc4_dev(state->plane->dev);
+	if (!vc4->load_tracker_available)
+		return;
 
 	vc4_state = to_vc4_plane_state(state);
 	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
@@ -563,7 +592,9 @@ static int vc4_plane_allocate_lbm(struct drm_plane_state *state)
 		spin_lock_irqsave(&vc4->hvs->mm_lock, irqflags);
 		ret = drm_mm_insert_node_generic(&vc4->hvs->lbm_mm,
 						 &vc4_state->lbm,
-						 lbm_size, 32, 0, 0);
+						 lbm_size,
+						 vc4->hvs->hvs5 ? 64 : 32,
+						 0, 0);
 		spin_unlock_irqrestore(&vc4->hvs->mm_lock, irqflags);
 
 		if (ret)
@@ -776,35 +807,6 @@ static int vc4_plane_mode_set(struct drm_plane *plane,
 		return -EINVAL;
 	}
 
-	/* Control word */
-	vc4_dlist_write(vc4_state,
-			SCALER_CTL0_VALID |
-			(rotation & DRM_MODE_REFLECT_X ? SCALER_CTL0_HFLIP : 0) |
-			(rotation & DRM_MODE_REFLECT_Y ? SCALER_CTL0_VFLIP : 0) |
-			VC4_SET_FIELD(SCALER_CTL0_RGBA_EXPAND_ROUND, SCALER_CTL0_RGBA_EXPAND) |
-			(format->pixel_order << SCALER_CTL0_ORDER_SHIFT) |
-			(hvs_format << SCALER_CTL0_PIXEL_FORMAT_SHIFT) |
-			VC4_SET_FIELD(tiling, SCALER_CTL0_TILING) |
-			(vc4_state->is_unity ? SCALER_CTL0_UNITY : 0) |
-			VC4_SET_FIELD(scl0, SCALER_CTL0_SCL0) |
-			VC4_SET_FIELD(scl1, SCALER_CTL0_SCL1));
-
-	/* Position Word 0: Image Positions and Alpha Value */
-	vc4_state->pos0_offset = vc4_state->dlist_count;
-	vc4_dlist_write(vc4_state,
-			VC4_SET_FIELD(state->alpha >> 8, SCALER_POS0_FIXED_ALPHA) |
-			VC4_SET_FIELD(vc4_state->crtc_x, SCALER_POS0_START_X) |
-			VC4_SET_FIELD(vc4_state->crtc_y, SCALER_POS0_START_Y));
-
-	/* Position Word 1: Scaled Image Dimensions. */
-	if (!vc4_state->is_unity) {
-		vc4_dlist_write(vc4_state,
-				VC4_SET_FIELD(vc4_state->crtc_w,
-					      SCALER_POS1_SCL_WIDTH) |
-				VC4_SET_FIELD(vc4_state->crtc_h,
-					      SCALER_POS1_SCL_HEIGHT));
-	}
-
 	/* Don't waste cycles mixing with plane alpha if the set alpha
 	 * is opaque or there is no per-pixel alpha information.
 	 * In any case we use the alpha property value as the fixed alpha.
@@ -812,20 +814,120 @@ static int vc4_plane_mode_set(struct drm_plane *plane,
 	mix_plane_alpha = state->alpha != DRM_BLEND_ALPHA_OPAQUE &&
 			  fb->format->has_alpha;
 
-	/* Position Word 2: Source Image Size, Alpha */
-	vc4_state->pos2_offset = vc4_state->dlist_count;
-	vc4_dlist_write(vc4_state,
-			VC4_SET_FIELD(fb->format->has_alpha ?
-				      SCALER_POS2_ALPHA_MODE_PIPELINE :
-				      SCALER_POS2_ALPHA_MODE_FIXED,
-				      SCALER_POS2_ALPHA_MODE) |
-			(mix_plane_alpha ? SCALER_POS2_ALPHA_MIX : 0) |
-			(fb->format->has_alpha ? SCALER_POS2_ALPHA_PREMULT : 0) |
-			VC4_SET_FIELD(vc4_state->src_w[0], SCALER_POS2_WIDTH) |
-			VC4_SET_FIELD(vc4_state->src_h[0], SCALER_POS2_HEIGHT));
+	if (!vc4->hvs->hvs5) {
+	/* Control word */
+		vc4_dlist_write(vc4_state,
+				SCALER_CTL0_VALID |
+				(rotation & DRM_MODE_REFLECT_X ? SCALER_CTL0_HFLIP : 0) |
+				(rotation & DRM_MODE_REFLECT_Y ? SCALER_CTL0_VFLIP : 0) |
+				VC4_SET_FIELD(SCALER_CTL0_RGBA_EXPAND_ROUND, SCALER_CTL0_RGBA_EXPAND) |
+				(format->pixel_order << SCALER_CTL0_ORDER_SHIFT) |
+				(hvs_format << SCALER_CTL0_PIXEL_FORMAT_SHIFT) |
+				VC4_SET_FIELD(tiling, SCALER_CTL0_TILING) |
+				(vc4_state->is_unity ? SCALER_CTL0_UNITY : 0) |
+				VC4_SET_FIELD(scl0, SCALER_CTL0_SCL0) |
+				VC4_SET_FIELD(scl1, SCALER_CTL0_SCL1));
+
+		/* Position Word 0: Image Positions and Alpha Value */
+		vc4_state->pos0_offset = vc4_state->dlist_count;
+		vc4_dlist_write(vc4_state,
+				VC4_SET_FIELD(state->alpha >> 8, SCALER_POS0_FIXED_ALPHA) |
+				VC4_SET_FIELD(vc4_state->crtc_x, SCALER_POS0_START_X) |
+				VC4_SET_FIELD(vc4_state->crtc_y, SCALER_POS0_START_Y));
+
+		/* Position Word 1: Scaled Image Dimensions. */
+		if (!vc4_state->is_unity) {
+			vc4_dlist_write(vc4_state,
+					VC4_SET_FIELD(vc4_state->crtc_w,
+						      SCALER_POS1_SCL_WIDTH) |
+					VC4_SET_FIELD(vc4_state->crtc_h,
+						      SCALER_POS1_SCL_HEIGHT));
+		}
+
+		/* Position Word 2: Source Image Size, Alpha */
+		vc4_state->pos2_offset = vc4_state->dlist_count;
+		vc4_dlist_write(vc4_state,
+				VC4_SET_FIELD(fb->format->has_alpha ?
+					      SCALER_POS2_ALPHA_MODE_PIPELINE :
+					      SCALER_POS2_ALPHA_MODE_FIXED,
+					      SCALER_POS2_ALPHA_MODE) |
+				(mix_plane_alpha ? SCALER_POS2_ALPHA_MIX : 0) |
+				(fb->format->has_alpha ?
+						SCALER_POS2_ALPHA_PREMULT : 0) |
+				VC4_SET_FIELD(vc4_state->src_w[0],
+					      SCALER_POS2_WIDTH) |
+				VC4_SET_FIELD(vc4_state->src_h[0],
+					      SCALER_POS2_HEIGHT));
+
+		/* Position Word 3: Context.  Written by the HVS. */
+		vc4_dlist_write(vc4_state, 0xc0c0c0c0);
 
-	/* Position Word 3: Context.  Written by the HVS. */
-	vc4_dlist_write(vc4_state, 0xc0c0c0c0);
+	} else {
+		u32 hvs_pixel_order = format->pixel_order;
+
+		if (format->pixel_order_hvs5)
+			hvs_pixel_order = format->pixel_order_hvs5;
+
+		/* Control word */
+		vc4_dlist_write(vc4_state,
+				SCALER_CTL0_VALID |
+				(hvs_pixel_order << SCALER_CTL0_ORDER_SHIFT) |
+				(hvs_format << SCALER_CTL0_PIXEL_FORMAT_SHIFT) |
+				VC4_SET_FIELD(tiling, SCALER_CTL0_TILING) |
+				(vc4_state->is_unity ?
+						SCALER5_CTL0_UNITY : 0) |
+				VC4_SET_FIELD(scl0, SCALER_CTL0_SCL0) |
+				VC4_SET_FIELD(scl1, SCALER_CTL0_SCL1) |
+				SCALER5_CTL0_ALPHA_EXPAND |
+				SCALER5_CTL0_RGB_EXPAND);
+
+		/* Position Word 0: Image Positions and Alpha Value */
+		vc4_state->pos0_offset = vc4_state->dlist_count;
+		vc4_dlist_write(vc4_state,
+				(rotation & DRM_MODE_REFLECT_Y ?
+						SCALER5_POS0_VFLIP : 0) |
+				VC4_SET_FIELD(vc4_state->crtc_x,
+					      SCALER_POS0_START_X) |
+				(rotation & DRM_MODE_REFLECT_X ?
+					      SCALER5_POS0_HFLIP : 0) |
+				VC4_SET_FIELD(vc4_state->crtc_y,
+					      SCALER5_POS0_START_Y)
+			       );
+
+		/* Control Word 2 */
+		vc4_dlist_write(vc4_state,
+				VC4_SET_FIELD(state->alpha >> 4,
+					      SCALER5_CTL2_ALPHA) |
+				(fb->format->has_alpha ?
+					SCALER5_CTL2_ALPHA_PREMULT : 0) |
+				(mix_plane_alpha ?
+					SCALER5_CTL2_ALPHA_MIX : 0) |
+				VC4_SET_FIELD(fb->format->has_alpha ?
+				      SCALER5_CTL2_ALPHA_MODE_PIPELINE :
+				      SCALER5_CTL2_ALPHA_MODE_FIXED,
+				      SCALER5_CTL2_ALPHA_MODE)
+			       );
+
+		/* Position Word 1: Scaled Image Dimensions. */
+		if (!vc4_state->is_unity) {
+			vc4_dlist_write(vc4_state,
+					VC4_SET_FIELD(vc4_state->crtc_w,
+						      SCALER_POS1_SCL_WIDTH) |
+					VC4_SET_FIELD(vc4_state->crtc_h,
+						      SCALER_POS1_SCL_HEIGHT));
+		}
+
+		/* Position Word 2: Source Image Size */
+		vc4_state->pos2_offset = vc4_state->dlist_count;
+		vc4_dlist_write(vc4_state,
+				VC4_SET_FIELD(vc4_state->src_w[0],
+					      SCALER5_POS2_WIDTH) |
+				VC4_SET_FIELD(vc4_state->src_h[0],
+					      SCALER5_POS2_HEIGHT));
+
+		/* Position Word 3: Context.  Written by the HVS. */
+		vc4_dlist_write(vc4_state, 0xc0c0c0c0);
+	}
 
 
 	/* Pointer Word 0/1/2: RGB / Y / Cb / Cr Pointers
@@ -1203,6 +1305,10 @@ static bool vc4_format_mod_supported(struct drm_plane *plane,
 		default:
 			return false;
 		}
+	case DRM_FORMAT_RGBX1010102:
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
 	case DRM_FORMAT_YUV422:
 	case DRM_FORMAT_YVU422:
 	case DRM_FORMAT_YUV420:
@@ -1255,6 +1361,8 @@ struct drm_plane *vc4_plane_init(struct drm_device *dev,
 				       &vc4_plane_funcs,
 				       formats, ARRAY_SIZE(formats),
 				       modifiers, type, NULL);
+	if (ret)
+		return ERR_PTR(ret);
 
 	drm_plane_helper_add(plane, &vc4_plane_helper_funcs);
 
@@ -1283,7 +1391,7 @@ int vc4_plane_create_additional_planes(struct drm_device *drm)
 	 * modest number of planes to expose, that should hopefully
 	 * still cover any sane usecase.
 	 */
-	for (i = 0; i < 8; i++) {
+	for (i = 0; i < 16; i++) {
 		struct drm_plane *plane =
 			vc4_plane_init(drm, DRM_PLANE_TYPE_OVERLAY);
 
diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h
index 324462cc9cd4..be2c32a519b3 100644
--- a/drivers/gpu/drm/vc4/vc4_regs.h
+++ b/drivers/gpu/drm/vc4/vc4_regs.h
@@ -129,6 +129,8 @@
 #define V3D_ERRSTAT  0x00f20
 
 #define PV_CONTROL				0x00
+# define PV5_CONTROL_FIFO_LEVEL_HIGH_MASK	VC4_MASK(26, 25)
+# define PV5_CONTROL_FIFO_LEVEL_HIGH_SHIFT	25
 # define PV_CONTROL_FORMAT_MASK			VC4_MASK(23, 21)
 # define PV_CONTROL_FORMAT_SHIFT		21
 # define PV_CONTROL_FORMAT_24			0
@@ -208,6 +210,11 @@
 
 #define PV_HACT_ACT				0x30
 
+#define PV_MUX_CFG				0x34
+# define PV_MUX_CFG_RGB_PIXEL_MUX_MODE_MASK	VC4_MASK(5, 2)
+# define PV_MUX_CFG_RGB_PIXEL_MUX_MODE_SHIFT	2
+# define PV_MUX_CFG_RGB_PIXEL_MUX_MODE_NO_SWAP	8
+
 #define SCALER_CHANNELS_COUNT			3
 
 #define SCALER_DISPCTRL                         0x00000000
@@ -286,9 +293,19 @@
 
 #define SCALER_DISPID                           0x00000008
 #define SCALER_DISPECTRL                        0x0000000c
+# define SCALER_DISPECTRL_DSP2_MUX_SHIFT	31
+# define SCALER_DISPECTRL_DSP2_MUX_MASK		VC4_MASK(31, 31)
+
 #define SCALER_DISPPROF                         0x00000010
+
 #define SCALER_DISPDITHER                       0x00000014
+# define SCALER_DISPDITHER_DSP5_MUX_SHIFT	30
+# define SCALER_DISPDITHER_DSP5_MUX_MASK	VC4_MASK(31, 30)
+
 #define SCALER_DISPEOLN                         0x00000018
+# define SCALER_DISPEOLN_DSP4_MUX_SHIFT		30
+# define SCALER_DISPEOLN_DSP4_MUX_MASK		VC4_MASK(31, 30)
+
 #define SCALER_DISPLIST0                        0x00000020
 #define SCALER_DISPLIST1                        0x00000024
 #define SCALER_DISPLIST2                        0x00000028
@@ -327,6 +344,20 @@
 # define SCALER_DISPCTRLX_HEIGHT_MASK		VC4_MASK(11, 0)
 # define SCALER_DISPCTRLX_HEIGHT_SHIFT		0
 
+# define SCALER5_DISPCTRLX_WIDTH_MASK		VC4_MASK(28, 16)
+# define SCALER5_DISPCTRLX_WIDTH_SHIFT		16
+/* Generates a single frame when VSTART is seen and stops at the last
+ * pixel read from the FIFO.
+ */
+# define SCALER5_DISPCTRLX_ONESHOT		BIT(15)
+/* Processes a single context in the dlist and then task switch,
+ * instead of an entire line.
+ */
+# define SCALER5_DISPCTRLX_ONECTX_MASK		VC4_MASK(14, 13)
+# define SCALER5_DISPCTRLX_ONECTX_SHIFT		13
+# define SCALER5_DISPCTRLX_HEIGHT_MASK		VC4_MASK(12, 0)
+# define SCALER5_DISPCTRLX_HEIGHT_SHIFT		0
+
 #define SCALER_DISPBKGND0                       0x00000044
 # define SCALER_DISPBKGND_AUTOHS		BIT(31)
 # define SCALER_DISPBKGND_INTERLACE		BIT(30)
@@ -460,32 +491,18 @@
 #define SCALER_DLIST_START                      0x00002000
 #define SCALER_DLIST_SIZE                       0x00004000
 
-#define VC4_HDMI_CORE_REV			0x000
+#define SCALER5_DLIST_START			0x00004000
 
-#define VC4_HDMI_SW_RESET_CONTROL		0x004
 # define VC4_HDMI_SW_RESET_FORMAT_DETECT	BIT(1)
 # define VC4_HDMI_SW_RESET_HDMI			BIT(0)
 
-#define VC4_HDMI_HOTPLUG_INT			0x008
-
-#define VC4_HDMI_HOTPLUG			0x00c
 # define VC4_HDMI_HOTPLUG_CONNECTED		BIT(0)
 
-/* 3 bits per field, where each field maps from that corresponding MAI
- * bus channel to the given HDMI channel.
- */
-#define VC4_HDMI_MAI_CHANNEL_MAP		0x090
-
-#define VC4_HDMI_MAI_CONFIG			0x094
 # define VC4_HDMI_MAI_CONFIG_FORMAT_REVERSE		BIT(27)
 # define VC4_HDMI_MAI_CONFIG_BIT_REVERSE		BIT(26)
 # define VC4_HDMI_MAI_CHANNEL_MASK_MASK			VC4_MASK(15, 0)
 # define VC4_HDMI_MAI_CHANNEL_MASK_SHIFT		0
 
-/* Last received format word on the MAI bus. */
-#define VC4_HDMI_MAI_FORMAT			0x098
-
-#define VC4_HDMI_AUDIO_PACKET_CONFIG		0x09c
 # define VC4_HDMI_AUDIO_PACKET_ZERO_DATA_ON_SAMPLE_FLAT		BIT(29)
 # define VC4_HDMI_AUDIO_PACKET_ZERO_DATA_ON_INACTIVE_CHANNELS	BIT(24)
 # define VC4_HDMI_AUDIO_PACKET_FORCE_SAMPLE_PRESENT		BIT(19)
@@ -499,12 +516,8 @@
 # define VC4_HDMI_AUDIO_PACKET_CEA_MASK_MASK			VC4_MASK(7, 0)
 # define VC4_HDMI_AUDIO_PACKET_CEA_MASK_SHIFT			0
 
-#define VC4_HDMI_RAM_PACKET_CONFIG		0x0a0
 # define VC4_HDMI_RAM_PACKET_ENABLE		BIT(16)
 
-#define VC4_HDMI_RAM_PACKET_STATUS		0x0a4
-
-#define VC4_HDMI_CRP_CFG			0x0a8
 /* When set, the CTS_PERIOD counts based on MAI bus sync pulse instead
  * of pixel clock.
  */
@@ -518,23 +531,12 @@
 # define VC4_HDMI_CRP_CFG_N_MASK		VC4_MASK(19, 0)
 # define VC4_HDMI_CRP_CFG_N_SHIFT		0
 
-/* 20-bit fields containing CTS values to be transmitted if !EXTERNAL_CTS_EN */
-#define VC4_HDMI_CTS_0				0x0ac
-#define VC4_HDMI_CTS_1				0x0b0
-/* 20-bit fields containing number of clocks to send CTS0/1 before
- * switching to the other one.
- */
-#define VC4_HDMI_CTS_PERIOD_0			0x0b4
-#define VC4_HDMI_CTS_PERIOD_1			0x0b8
-
-#define VC4_HDMI_HORZA				0x0c4
 # define VC4_HDMI_HORZA_VPOS			BIT(14)
 # define VC4_HDMI_HORZA_HPOS			BIT(13)
 /* Horizontal active pixels (hdisplay). */
 # define VC4_HDMI_HORZA_HAP_MASK		VC4_MASK(12, 0)
 # define VC4_HDMI_HORZA_HAP_SHIFT		0
 
-#define VC4_HDMI_HORZB				0x0c8
 /* Horizontal pack porch (htotal - hsync_end). */
 # define VC4_HDMI_HORZB_HBP_MASK		VC4_MASK(29, 20)
 # define VC4_HDMI_HORZB_HBP_SHIFT		20
@@ -545,7 +547,6 @@
 # define VC4_HDMI_HORZB_HFP_MASK		VC4_MASK(9, 0)
 # define VC4_HDMI_HORZB_HFP_SHIFT		0
 
-#define VC4_HDMI_FIFO_CTL			0x05c
 # define VC4_HDMI_FIFO_CTL_RECENTER_DONE	BIT(14)
 # define VC4_HDMI_FIFO_CTL_USE_EMPTY		BIT(13)
 # define VC4_HDMI_FIFO_CTL_ON_VB		BIT(7)
@@ -558,15 +559,12 @@
 # define VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N	BIT(0)
 # define VC4_HDMI_FIFO_VALID_WRITE_MASK		0xefff
 
-#define VC4_HDMI_SCHEDULER_CONTROL		0x0c0
 # define VC4_HDMI_SCHEDULER_CONTROL_MANUAL_FORMAT BIT(15)
 # define VC4_HDMI_SCHEDULER_CONTROL_IGNORE_VSYNC_PREDICTS BIT(5)
 # define VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT	BIT(3)
 # define VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE	BIT(1)
 # define VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI	BIT(0)
 
-#define VC4_HDMI_VERTA0				0x0cc
-#define VC4_HDMI_VERTA1				0x0d4
 /* Vertical sync pulse (vsync_end - vsync_start). */
 # define VC4_HDMI_VERTA_VSP_MASK		VC4_MASK(24, 20)
 # define VC4_HDMI_VERTA_VSP_SHIFT		20
@@ -577,8 +575,6 @@
 # define VC4_HDMI_VERTA_VAL_MASK		VC4_MASK(12, 0)
 # define VC4_HDMI_VERTA_VAL_SHIFT		0
 
-#define VC4_HDMI_VERTB0				0x0d0
-#define VC4_HDMI_VERTB1				0x0d8
 /* Vertical sync pulse offset (for interlaced) */
 # define VC4_HDMI_VERTB_VSPO_MASK		VC4_MASK(21, 9)
 # define VC4_HDMI_VERTB_VSPO_SHIFT		9
@@ -586,7 +582,6 @@
 # define VC4_HDMI_VERTB_VBP_MASK		VC4_MASK(8, 0)
 # define VC4_HDMI_VERTB_VBP_SHIFT		0
 
-#define VC4_HDMI_CEC_CNTRL_1			0x0e8
 /* Set when the transmission has ended. */
 # define VC4_HDMI_CEC_TX_EOM			BIT(31)
 /* If set, transmission was acked on the 1st or 2nd attempt (only one
@@ -627,7 +622,6 @@
 /* Set these fields to how many bit clock cycles get to that many
  * microseconds.
  */
-#define VC4_HDMI_CEC_CNTRL_2			0x0ec
 # define VC4_HDMI_CEC_CNT_TO_1500_US_MASK	VC4_MASK(30, 24)
 # define VC4_HDMI_CEC_CNT_TO_1500_US_SHIFT	24
 # define VC4_HDMI_CEC_CNT_TO_1300_US_MASK	VC4_MASK(23, 17)
@@ -639,7 +633,6 @@
 # define VC4_HDMI_CEC_CNT_TO_400_US_MASK	VC4_MASK(4, 0)
 # define VC4_HDMI_CEC_CNT_TO_400_US_SHIFT	0
 
-#define VC4_HDMI_CEC_CNTRL_3			0x0f0
 # define VC4_HDMI_CEC_CNT_TO_2750_US_MASK	VC4_MASK(31, 24)
 # define VC4_HDMI_CEC_CNT_TO_2750_US_SHIFT	24
 # define VC4_HDMI_CEC_CNT_TO_2400_US_MASK	VC4_MASK(23, 16)
@@ -649,7 +642,6 @@
 # define VC4_HDMI_CEC_CNT_TO_1700_US_MASK	VC4_MASK(7, 0)
 # define VC4_HDMI_CEC_CNT_TO_1700_US_SHIFT	0
 
-#define VC4_HDMI_CEC_CNTRL_4			0x0f4
 # define VC4_HDMI_CEC_CNT_TO_4300_US_MASK	VC4_MASK(31, 24)
 # define VC4_HDMI_CEC_CNT_TO_4300_US_SHIFT	24
 # define VC4_HDMI_CEC_CNT_TO_3900_US_MASK	VC4_MASK(23, 16)
@@ -659,7 +651,6 @@
 # define VC4_HDMI_CEC_CNT_TO_3500_US_MASK	VC4_MASK(7, 0)
 # define VC4_HDMI_CEC_CNT_TO_3500_US_SHIFT	0
 
-#define VC4_HDMI_CEC_CNTRL_5			0x0f8
 # define VC4_HDMI_CEC_TX_SW_RESET		BIT(27)
 # define VC4_HDMI_CEC_RX_SW_RESET		BIT(26)
 # define VC4_HDMI_CEC_PAD_SW_RESET		BIT(25)
@@ -672,39 +663,11 @@
 # define VC4_HDMI_CEC_CNT_TO_4500_US_MASK	VC4_MASK(7, 0)
 # define VC4_HDMI_CEC_CNT_TO_4500_US_SHIFT	0
 
-/* Transmit data, first byte is low byte of the 32-bit reg.  MSB of
- * each byte transmitted first.
- */
-#define VC4_HDMI_CEC_TX_DATA_1			0x0fc
-#define VC4_HDMI_CEC_TX_DATA_2			0x100
-#define VC4_HDMI_CEC_TX_DATA_3			0x104
-#define VC4_HDMI_CEC_TX_DATA_4			0x108
-#define VC4_HDMI_CEC_RX_DATA_1			0x10c
-#define VC4_HDMI_CEC_RX_DATA_2			0x110
-#define VC4_HDMI_CEC_RX_DATA_3			0x114
-#define VC4_HDMI_CEC_RX_DATA_4			0x118
-
-#define VC4_HDMI_TX_PHY_RESET_CTL		0x2c0
-
-#define VC4_HDMI_TX_PHY_CTL0			0x2c4
 # define VC4_HDMI_TX_PHY_RNG_PWRDN		BIT(25)
 
-/* Interrupt status bits */
-#define VC4_HDMI_CPU_STATUS			0x340
-#define VC4_HDMI_CPU_SET			0x344
-#define VC4_HDMI_CPU_CLEAR			0x348
 # define VC4_HDMI_CPU_CEC			BIT(6)
 # define VC4_HDMI_CPU_HOTPLUG			BIT(0)
 
-#define VC4_HDMI_CPU_MASK_STATUS		0x34c
-#define VC4_HDMI_CPU_MASK_SET			0x350
-#define VC4_HDMI_CPU_MASK_CLEAR			0x354
-
-#define VC4_HDMI_GCP(x)				(0x400 + ((x) * 0x4))
-#define VC4_HDMI_RAM_PACKET(x)			(0x400 + ((x) * 0x24))
-#define VC4_HDMI_PACKET_STRIDE			0x24
-
-#define VC4_HD_M_CTL				0x00c
 /* Debug: Current receive value on the CEC pad. */
 # define VC4_HD_CECRXD				BIT(9)
 /* Debug: Override CEC output to 0. */
@@ -714,7 +677,6 @@
 # define VC4_HD_M_SW_RST			BIT(2)
 # define VC4_HD_M_ENABLE			BIT(0)
 
-#define VC4_HD_MAI_CTL				0x014
 /* Set when audio stream is received at a slower rate than the
  * sampling period, so MAI fifo goes empty.  Write 1 to clear.
  */
@@ -739,7 +701,6 @@
 /* Single-shot reset bit.  Read value is undefined. */
 # define VC4_HD_MAI_CTL_RESET			BIT(0)
 
-#define VC4_HD_MAI_THR				0x018
 # define VC4_HD_MAI_THR_PANICHIGH_MASK		VC4_MASK(29, 24)
 # define VC4_HD_MAI_THR_PANICHIGH_SHIFT		24
 # define VC4_HD_MAI_THR_PANICLOW_MASK		VC4_MASK(21, 16)
@@ -749,31 +710,23 @@
 # define VC4_HD_MAI_THR_DREQLOW_MASK		VC4_MASK(5, 0)
 # define VC4_HD_MAI_THR_DREQLOW_SHIFT		0
 
-/* Format header to be placed on the MAI data. Unused. */
-#define VC4_HD_MAI_FMT				0x01c
-
-/* Register for DMAing in audio data to be transported over the MAI
- * bus to the Falcon core.
- */
-#define VC4_HD_MAI_DATA				0x020
-
 /* Divider from HDMI HSM clock to MAI serial clock.  Sampling period
  * converges to N / (M + 1) cycles.
  */
-#define VC4_HD_MAI_SMP				0x02c
 # define VC4_HD_MAI_SMP_N_MASK			VC4_MASK(31, 8)
 # define VC4_HD_MAI_SMP_N_SHIFT			8
 # define VC4_HD_MAI_SMP_M_MASK			VC4_MASK(7, 0)
 # define VC4_HD_MAI_SMP_M_SHIFT			0
 
-#define VC4_HD_VID_CTL				0x038
 # define VC4_HD_VID_CTL_ENABLE			BIT(31)
 # define VC4_HD_VID_CTL_UNDERFLOW_ENABLE	BIT(30)
 # define VC4_HD_VID_CTL_FRAME_COUNTER_RESET	BIT(29)
 # define VC4_HD_VID_CTL_VSYNC_LOW		BIT(28)
 # define VC4_HD_VID_CTL_HSYNC_LOW		BIT(27)
+# define VC4_HD_VID_CTL_CLRSYNC			BIT(24)
+# define VC4_HD_VID_CTL_CLRRGB			BIT(23)
+# define VC4_HD_VID_CTL_BLANKPIX		BIT(18)
 
-#define VC4_HD_CSC_CTL				0x040
 # define VC4_HD_CSC_CTL_ORDER_MASK		VC4_MASK(7, 5)
 # define VC4_HD_CSC_CTL_ORDER_SHIFT		5
 # define VC4_HD_CSC_CTL_ORDER_RGB		0
@@ -791,14 +744,7 @@
 # define VC4_HD_CSC_CTL_RGB2YCC			BIT(1)
 # define VC4_HD_CSC_CTL_ENABLE			BIT(0)
 
-#define VC4_HD_CSC_12_11			0x044
-#define VC4_HD_CSC_14_13			0x048
-#define VC4_HD_CSC_22_21			0x04c
-#define VC4_HD_CSC_24_23			0x050
-#define VC4_HD_CSC_32_31			0x054
-#define VC4_HD_CSC_34_33			0x058
-
-#define VC4_HD_FRAME_COUNT			0x068
+# define VC4_DVP_HT_CLOCK_STOP_PIXEL		BIT(1)
 
 /* HVS display list information. */
 #define HVS_BOOTLOADER_DLIST_END                32
@@ -825,6 +771,8 @@ enum hvs_pixel_format {
 	HVS_PIXEL_FORMAT_PALETTE = 13,
 	HVS_PIXEL_FORMAT_YUV444_RGB = 14,
 	HVS_PIXEL_FORMAT_AYUV444_RGB = 15,
+	HVS_PIXEL_FORMAT_RGBA1010102 = 16,
+	HVS_PIXEL_FORMAT_YCBCR_10BIT = 17,
 };
 
 /* Note: the LSB is the rightmost character shown.  Only valid for
@@ -879,6 +827,10 @@ enum hvs_pixel_format {
 #define SCALER_CTL0_RGBA_EXPAND_MSB		2
 #define SCALER_CTL0_RGBA_EXPAND_ROUND		3
 
+#define SCALER5_CTL0_ALPHA_EXPAND		BIT(12)
+
+#define SCALER5_CTL0_RGB_EXPAND			BIT(11)
+
 #define SCALER_CTL0_SCL1_MASK			VC4_MASK(10, 8)
 #define SCALER_CTL0_SCL1_SHIFT			8
 
@@ -896,10 +848,13 @@ enum hvs_pixel_format {
 
 /* Set to indicate no scaling. */
 #define SCALER_CTL0_UNITY			BIT(4)
+#define SCALER5_CTL0_UNITY			BIT(15)
 
 #define SCALER_CTL0_PIXEL_FORMAT_MASK		VC4_MASK(3, 0)
 #define SCALER_CTL0_PIXEL_FORMAT_SHIFT		0
 
+#define SCALER5_CTL0_PIXEL_FORMAT_MASK		VC4_MASK(4, 0)
+
 #define SCALER_POS0_FIXED_ALPHA_MASK		VC4_MASK(31, 24)
 #define SCALER_POS0_FIXED_ALPHA_SHIFT		24
 
@@ -909,12 +864,48 @@ enum hvs_pixel_format {
 #define SCALER_POS0_START_X_MASK		VC4_MASK(11, 0)
 #define SCALER_POS0_START_X_SHIFT		0
 
+#define SCALER5_POS0_START_Y_MASK		VC4_MASK(27, 16)
+#define SCALER5_POS0_START_Y_SHIFT		16
+
+#define SCALER5_POS0_START_X_MASK		VC4_MASK(13, 0)
+#define SCALER5_POS0_START_X_SHIFT		0
+
+#define SCALER5_POS0_VFLIP			BIT(31)
+#define SCALER5_POS0_HFLIP			BIT(15)
+
+#define SCALER5_CTL2_ALPHA_MODE_MASK		VC4_MASK(31, 30)
+#define SCALER5_CTL2_ALPHA_MODE_SHIFT		30
+#define SCALER5_CTL2_ALPHA_MODE_PIPELINE		0
+#define SCALER5_CTL2_ALPHA_MODE_FIXED		1
+#define SCALER5_CTL2_ALPHA_MODE_FIXED_NONZERO	2
+#define SCALER5_CTL2_ALPHA_MODE_FIXED_OVER_0x07	3
+
+#define SCALER5_CTL2_ALPHA_PREMULT		BIT(29)
+
+#define SCALER5_CTL2_ALPHA_MIX			BIT(28)
+
+#define SCALER5_CTL2_ALPHA_LOC			BIT(25)
+
+#define SCALER5_CTL2_MAP_SEL_MASK		VC4_MASK(18, 17)
+#define SCALER5_CTL2_MAP_SEL_SHIFT		17
+
+#define SCALER5_CTL2_GAMMA			BIT(16)
+
+#define SCALER5_CTL2_ALPHA_MASK			VC4_MASK(15, 4)
+#define SCALER5_CTL2_ALPHA_SHIFT		4
+
 #define SCALER_POS1_SCL_HEIGHT_MASK		VC4_MASK(27, 16)
 #define SCALER_POS1_SCL_HEIGHT_SHIFT		16
 
 #define SCALER_POS1_SCL_WIDTH_MASK		VC4_MASK(11, 0)
 #define SCALER_POS1_SCL_WIDTH_SHIFT		0
 
+#define SCALER5_POS1_SCL_HEIGHT_MASK		VC4_MASK(28, 16)
+#define SCALER5_POS1_SCL_HEIGHT_SHIFT		16
+
+#define SCALER5_POS1_SCL_WIDTH_MASK		VC4_MASK(12, 0)
+#define SCALER5_POS1_SCL_WIDTH_SHIFT		0
+
 #define SCALER_POS2_ALPHA_MODE_MASK		VC4_MASK(31, 30)
 #define SCALER_POS2_ALPHA_MODE_SHIFT		30
 #define SCALER_POS2_ALPHA_MODE_PIPELINE		0
@@ -930,6 +921,12 @@ enum hvs_pixel_format {
 #define SCALER_POS2_WIDTH_MASK			VC4_MASK(11, 0)
 #define SCALER_POS2_WIDTH_SHIFT			0
 
+#define SCALER5_POS2_HEIGHT_MASK		VC4_MASK(28, 16)
+#define SCALER5_POS2_HEIGHT_SHIFT		16
+
+#define SCALER5_POS2_WIDTH_MASK			VC4_MASK(12, 0)
+#define SCALER5_POS2_WIDTH_SHIFT		0
+
 /* Color Space Conversion words.  Some values are S2.8 signed
  * integers, except that the 2 integer bits map as {0x0: 0, 0x1: 1,
  * 0x2: 2, 0x3: -1}
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index a7c3af0005a0..849dcafbfff1 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -436,7 +436,6 @@ static const struct drm_crtc_helper_funcs vc4_txp_crtc_helper_funcs = {
 	.atomic_flush	= vc4_hvs_atomic_flush,
 	.atomic_enable	= vc4_txp_atomic_enable,
 	.atomic_disable	= vc4_txp_atomic_disable,
-	.mode_set_nofb	= vc4_hvs_mode_set_nofb,
 };
 
 static irqreturn_t vc4_txp_interrupt(int irq, void *data)
@@ -452,7 +451,8 @@ static irqreturn_t vc4_txp_interrupt(int irq, void *data)
 }
 
 static const struct vc4_crtc_data vc4_txp_crtc_data = {
-	.hvs_channel = 2,
+	.hvs_available_channels = BIT(2),
+	.hvs_output = 2,
 };
 
 static int vc4_txp_bind(struct device *dev, struct device *master, void *data)
diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c
index a775feda1cc7..cb884c890065 100644
--- a/drivers/gpu/drm/vgem/vgem_drv.c
+++ b/drivers/gpu/drm/vgem/vgem_drv.c
@@ -321,7 +321,7 @@ static struct sg_table *vgem_prime_get_sg_table(struct drm_gem_object *obj)
 {
 	struct drm_vgem_gem_object *bo = to_vgem_bo(obj);
 
-	return drm_prime_pages_to_sg(bo->pages, bo->base.size >> PAGE_SHIFT);
+	return drm_prime_pages_to_sg(obj->dev, bo->pages, bo->base.size >> PAGE_SHIFT);
 }
 
 static struct drm_gem_object* vgem_prime_import(struct drm_device *dev,
@@ -401,16 +401,8 @@ static int vgem_prime_mmap(struct drm_gem_object *obj,
 	return 0;
 }
 
-static void vgem_release(struct drm_device *dev)
-{
-	struct vgem_device *vgem = container_of(dev, typeof(*vgem), drm);
-
-	platform_device_unregister(vgem->platform);
-}
-
 static struct drm_driver vgem_driver = {
 	.driver_features		= DRIVER_GEM | DRIVER_RENDER,
-	.release			= vgem_release,
 	.open				= vgem_open,
 	.postclose			= vgem_postclose,
 	.gem_free_object_unlocked	= vgem_gem_free_object,
@@ -442,48 +434,49 @@ static struct drm_driver vgem_driver = {
 static int __init vgem_init(void)
 {
 	int ret;
+	struct platform_device *pdev;
 
-	vgem_device = kzalloc(sizeof(*vgem_device), GFP_KERNEL);
-	if (!vgem_device)
-		return -ENOMEM;
+	pdev = platform_device_register_simple("vgem", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
 
-	vgem_device->platform =
-		platform_device_register_simple("vgem", -1, NULL, 0);
-	if (IS_ERR(vgem_device->platform)) {
-		ret = PTR_ERR(vgem_device->platform);
-		goto out_free;
+	if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) {
+		ret = -ENOMEM;
+		goto out_unregister;
 	}
 
-	dma_coerce_mask_and_coherent(&vgem_device->platform->dev,
+	dma_coerce_mask_and_coherent(&pdev->dev,
 				     DMA_BIT_MASK(64));
-	ret = drm_dev_init(&vgem_device->drm, &vgem_driver,
-			   &vgem_device->platform->dev);
-	if (ret)
-		goto out_unregister;
-	drmm_add_final_kfree(&vgem_device->drm, vgem_device);
+
+	vgem_device = devm_drm_dev_alloc(&pdev->dev, &vgem_driver,
+					 struct vgem_device, drm);
+	if (IS_ERR(vgem_device)) {
+		ret = PTR_ERR(vgem_device);
+		goto out_devres;
+	}
+	vgem_device->platform = pdev;
 
 	/* Final step: expose the device/driver to userspace */
 	ret = drm_dev_register(&vgem_device->drm, 0);
 	if (ret)
-		goto out_put;
+		goto out_devres;
 
 	return 0;
 
-out_put:
-	drm_dev_put(&vgem_device->drm);
-	return ret;
-
+out_devres:
+	devres_release_group(&pdev->dev, NULL);
 out_unregister:
-	platform_device_unregister(vgem_device->platform);
-out_free:
-	kfree(vgem_device);
+	platform_device_unregister(pdev);
 	return ret;
 }
 
 static void __exit vgem_exit(void)
 {
+	struct platform_device *pdev = vgem_device->platform;
+
 	drm_dev_unregister(&vgem_device->drm);
-	drm_dev_put(&vgem_device->drm);
+	devres_release_group(&pdev->dev, NULL);
+	platform_device_unregister(pdev);
 }
 
 module_init(vgem_init);
diff --git a/drivers/gpu/drm/virtio/Kconfig b/drivers/gpu/drm/virtio/Kconfig
index eff3047052d4..b925b8b1da16 100644
--- a/drivers/gpu/drm/virtio/Kconfig
+++ b/drivers/gpu/drm/virtio/Kconfig
@@ -1,9 +1,10 @@
 # SPDX-License-Identifier: GPL-2.0-only
 config DRM_VIRTIO_GPU
 	tristate "Virtio GPU driver"
-	depends on DRM && VIRTIO && MMU
+	depends on DRM && VIRTIO && VIRTIO_MENU && MMU
 	select DRM_KMS_HELPER
 	select DRM_GEM_SHMEM_HELPER
+	select VIRTIO_DMA_SHARED_BUFFER
 	help
 	   This is the virtual GPU driver for virtio.  It can be used with
 	   QEMU based VMMs (like KVM or Xen).
diff --git a/drivers/gpu/drm/virtio/virtgpu_debugfs.c b/drivers/gpu/drm/virtio/virtgpu_debugfs.c
index 3221520f61f0..d5b0c543bd6d 100644
--- a/drivers/gpu/drm/virtio/virtgpu_debugfs.c
+++ b/drivers/gpu/drm/virtio/virtgpu_debugfs.c
@@ -48,6 +48,7 @@ static int virtio_gpu_features(struct seq_file *m, void *data)
 	virtio_add_bool(m, "virgl", vgdev->has_virgl_3d);
 	virtio_add_bool(m, "edid", vgdev->has_edid);
 	virtio_add_bool(m, "indirect", vgdev->has_indirect);
+	virtio_add_bool(m, "resource uuid", vgdev->has_resource_assign_uuid);
 	virtio_add_int(m, "cap sets", vgdev->num_capsets);
 	virtio_add_int(m, "scanouts", vgdev->num_scanouts);
 	return 0;
diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index afd0f9200f90..f84b7e61311b 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -172,8 +172,6 @@ static int virtio_gpu_conn_get_modes(struct drm_connector *connector)
 	count = drm_add_modes_noedid(connector, XRES_MAX, YRES_MAX);
 
 	if (width == 0 || height == 0) {
-		width = XRES_DEF;
-		height = YRES_DEF;
 		drm_set_preferred_mode(connector, XRES_DEF, YRES_DEF);
 	} else {
 		DRM_DEBUG("add mode: %dx%d\n", width, height);
@@ -327,11 +325,14 @@ static const struct drm_mode_config_funcs virtio_gpu_mode_funcs = {
 	.atomic_commit = drm_atomic_helper_commit,
 };
 
-void virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev)
+int virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev)
 {
-	int i;
+	int i, ret;
+
+	ret = drmm_mode_config_init(vgdev->ddev);
+	if (ret)
+		return ret;
 
-	drm_mode_config_init(vgdev->ddev);
 	vgdev->ddev->mode_config.quirk_addfb_prefer_host_byte_order = true;
 	vgdev->ddev->mode_config.funcs = &virtio_gpu_mode_funcs;
 
@@ -345,6 +346,7 @@ void virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev)
 		vgdev_output_init(vgdev, i);
 
 	drm_mode_config_reset(vgdev->ddev);
+	return 0;
 }
 
 void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev)
@@ -353,5 +355,4 @@ void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev)
 
 	for (i = 0 ; i < vgdev->num_scanouts; ++i)
 		kfree(vgdev->outputs[i].edid);
-	drm_mode_config_cleanup(vgdev->ddev);
 }
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index ab4bed78e656..b039f493bda9 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -165,6 +165,7 @@ static unsigned int features[] = {
 	VIRTIO_GPU_F_VIRGL,
 #endif
 	VIRTIO_GPU_F_EDID,
+	VIRTIO_GPU_F_RESOURCE_UUID,
 };
 static struct virtio_driver virtio_gpu_driver = {
 	.feature_table = features,
@@ -202,6 +203,8 @@ static struct drm_driver driver = {
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_mmap = drm_gem_prime_mmap,
+	.gem_prime_export = virtgpu_gem_prime_export,
+	.gem_prime_import = virtgpu_gem_prime_import,
 	.gem_prime_import_sg_table = virtgpu_gem_prime_import_sg_table,
 
 	.gem_create_object = virtio_gpu_create_object,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index fbc04272db4f..55c34b4fc3e9 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -49,6 +49,10 @@
 #define DRIVER_MINOR 1
 #define DRIVER_PATCHLEVEL 0
 
+#define UUID_INITIALIZING 0
+#define UUID_INITIALIZED 1
+#define UUID_INITIALIZATION_FAILED 2
+
 struct virtio_gpu_object_params {
 	uint32_t format;
 	uint32_t width;
@@ -71,6 +75,9 @@ struct virtio_gpu_object {
 	uint32_t hw_res_handle;
 	bool dumb;
 	bool created;
+
+	int uuid_state;
+	uuid_t uuid;
 };
 #define gem_to_virtio_gpu_obj(gobj) \
 	container_of((gobj), struct virtio_gpu_object, base.base)
@@ -200,6 +207,7 @@ struct virtio_gpu_device {
 	bool has_virgl_3d;
 	bool has_edid;
 	bool has_indirect;
+	bool has_resource_assign_uuid;
 
 	struct work_struct config_changed_work;
 
@@ -210,6 +218,9 @@ struct virtio_gpu_device {
 	struct virtio_gpu_drv_capset *capsets;
 	uint32_t num_capsets;
 	struct list_head cap_cache;
+
+	/* protects resource state when exporting */
+	spinlock_t resource_export_lock;
 };
 
 struct virtio_gpu_fpriv {
@@ -336,8 +347,12 @@ void virtio_gpu_dequeue_fence_func(struct work_struct *work);
 
 void virtio_gpu_notify(struct virtio_gpu_device *vgdev);
 
+int
+virtio_gpu_cmd_resource_assign_uuid(struct virtio_gpu_device *vgdev,
+				    struct virtio_gpu_object_array *objs);
+
 /* virtgpu_display.c */
-void virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev);
+int virtio_gpu_modeset_init(struct virtio_gpu_device *vgdev);
 void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev);
 
 /* virtgpu_plane.c */
@@ -367,6 +382,12 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
 bool virtio_gpu_is_shmem(struct virtio_gpu_object *bo);
 
 /* virtgpu_prime.c */
+struct dma_buf *virtgpu_gem_prime_export(struct drm_gem_object *obj,
+					 int flags);
+struct drm_gem_object *virtgpu_gem_prime_import(struct drm_device *dev,
+						struct dma_buf *buf);
+int virtgpu_gem_prime_get_uuid(struct drm_gem_object *obj,
+			       uuid_t *uuid);
 struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
 	struct drm_device *dev, struct dma_buf_attachment *attach,
 	struct sg_table *sgt);
diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c
index 24ffacac99e4..c30c75ee83fc 100644
--- a/drivers/gpu/drm/virtio/virtgpu_gem.c
+++ b/drivers/gpu/drm/virtio/virtgpu_gem.c
@@ -154,9 +154,8 @@ void virtio_gpu_gem_object_close(struct drm_gem_object *obj,
 struct virtio_gpu_object_array *virtio_gpu_array_alloc(u32 nents)
 {
 	struct virtio_gpu_object_array *objs;
-	size_t size = sizeof(*objs) + sizeof(objs->objs[0]) * nents;
 
-	objs = kmalloc(size, GFP_KERNEL);
+	objs = kmalloc(struct_size(objs, objs, nents), GFP_KERNEL);
 	if (!objs)
 		return NULL;
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c b/drivers/gpu/drm/virtio/virtgpu_kms.c
index 4d944a0dff3e..eed57a931309 100644
--- a/drivers/gpu/drm/virtio/virtgpu_kms.c
+++ b/drivers/gpu/drm/virtio/virtgpu_kms.c
@@ -80,8 +80,10 @@ static void virtio_gpu_get_capsets(struct virtio_gpu_device *vgdev,
 					 vgdev->capsets[i].id > 0, 5 * HZ);
 		if (ret == 0) {
 			DRM_ERROR("timed out waiting for cap set %d\n", i);
+			spin_lock(&vgdev->display_info_lock);
 			kfree(vgdev->capsets);
 			vgdev->capsets = NULL;
+			spin_unlock(&vgdev->display_info_lock);
 			return;
 		}
 		DRM_INFO("cap set %d: id %d, max-version %d, max-size %d\n",
@@ -103,7 +105,7 @@ int virtio_gpu_init(struct drm_device *dev)
 	/* this will expand later */
 	struct virtqueue *vqs[2];
 	u32 num_scanouts, num_capsets;
-	int ret;
+	int ret = 0;
 
 	if (!virtio_has_feature(dev_to_virtio(dev->dev), VIRTIO_F_VERSION_1))
 		return -ENODEV;
@@ -118,6 +120,7 @@ int virtio_gpu_init(struct drm_device *dev)
 	vgdev->dev = dev->dev;
 
 	spin_lock_init(&vgdev->display_info_lock);
+	spin_lock_init(&vgdev->resource_export_lock);
 	ida_init(&vgdev->ctx_id_ida);
 	ida_init(&vgdev->resource_ida);
 	init_waitqueue_head(&vgdev->resp_wq);
@@ -146,6 +149,9 @@ int virtio_gpu_init(struct drm_device *dev)
 	if (virtio_has_feature(vgdev->vdev, VIRTIO_RING_F_INDIRECT_DESC)) {
 		vgdev->has_indirect = true;
 	}
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_RESOURCE_UUID)) {
+		vgdev->has_resource_assign_uuid = true;
+	}
 
 	DRM_INFO("features: %cvirgl %cedid\n",
 		 vgdev->has_virgl_3d ? '+' : '-',
@@ -180,7 +186,11 @@ int virtio_gpu_init(struct drm_device *dev)
 			num_capsets, &num_capsets);
 	DRM_INFO("number of cap sets: %d\n", num_capsets);
 
-	virtio_gpu_modeset_init(vgdev);
+	ret = virtio_gpu_modeset_init(vgdev);
+	if (ret) {
+		DRM_ERROR("modeset init failed\n");
+		goto err_scanouts;
+	}
 
 	virtio_device_ready(vgdev->vdev);
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
index 842f8b61aa89..00d6b95e259d 100644
--- a/drivers/gpu/drm/virtio/virtgpu_object.c
+++ b/drivers/gpu/drm/virtio/virtgpu_object.c
@@ -72,9 +72,8 @@ void virtio_gpu_cleanup_object(struct virtio_gpu_object *bo)
 
 		if (shmem->pages) {
 			if (shmem->mapped) {
-				dma_unmap_sg(vgdev->vdev->dev.parent,
-					     shmem->pages->sgl, shmem->mapped,
-					     DMA_TO_DEVICE);
+				dma_unmap_sgtable(vgdev->vdev->dev.parent,
+					     shmem->pages, DMA_TO_DEVICE, 0);
 				shmem->mapped = 0;
 			}
 
@@ -164,13 +163,13 @@ static int virtio_gpu_object_shmem_init(struct virtio_gpu_device *vgdev,
 	}
 
 	if (use_dma_api) {
-		shmem->mapped = dma_map_sg(vgdev->vdev->dev.parent,
-					   shmem->pages->sgl,
-					   shmem->pages->nents,
-					   DMA_TO_DEVICE);
-		*nents = shmem->mapped;
+		ret = dma_map_sgtable(vgdev->vdev->dev.parent,
+				      shmem->pages, DMA_TO_DEVICE, 0);
+		if (ret)
+			return ret;
+		*nents = shmem->mapped = shmem->pages->nents;
 	} else {
-		*nents = shmem->pages->nents;
+		*nents = shmem->pages->orig_nents;
 	}
 
 	*ents = kmalloc_array(*nents, sizeof(struct virtio_gpu_mem_entry),
@@ -180,13 +179,20 @@ static int virtio_gpu_object_shmem_init(struct virtio_gpu_device *vgdev,
 		return -ENOMEM;
 	}
 
-	for_each_sg(shmem->pages->sgl, sg, *nents, si) {
-		(*ents)[si].addr = cpu_to_le64(use_dma_api
-					       ? sg_dma_address(sg)
-					       : sg_phys(sg));
-		(*ents)[si].length = cpu_to_le32(sg->length);
-		(*ents)[si].padding = 0;
+	if (use_dma_api) {
+		for_each_sgtable_dma_sg(shmem->pages, sg, si) {
+			(*ents)[si].addr = cpu_to_le64(sg_dma_address(sg));
+			(*ents)[si].length = cpu_to_le32(sg_dma_len(sg));
+			(*ents)[si].padding = 0;
+		}
+	} else {
+		for_each_sgtable_sg(shmem->pages, sg, si) {
+			(*ents)[si].addr = cpu_to_le64(sg_phys(sg));
+			(*ents)[si].length = cpu_to_le32(sg->length);
+			(*ents)[si].padding = 0;
+		}
 	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c
index 050d24c39a8f..acd14ef73d56 100644
--- a/drivers/gpu/drm/virtio/virtgpu_prime.c
+++ b/drivers/gpu/drm/virtio/virtgpu_prime.c
@@ -23,12 +23,102 @@
  */
 
 #include <drm/drm_prime.h>
+#include <linux/virtio_dma_buf.h>
 
 #include "virtgpu_drv.h"
 
-/* Empty Implementations as there should not be any other driver for a virtual
- * device that might share buffers with virtgpu
- */
+static int virtgpu_virtio_get_uuid(struct dma_buf *buf,
+				   uuid_t *uuid)
+{
+	struct drm_gem_object *obj = buf->priv;
+	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
+	struct virtio_gpu_device *vgdev = obj->dev->dev_private;
+
+	wait_event(vgdev->resp_wq, bo->uuid_state != UUID_INITIALIZING);
+	if (bo->uuid_state != UUID_INITIALIZED)
+		return -ENODEV;
+
+	uuid_copy(uuid, &bo->uuid);
+
+	return 0;
+}
+
+const struct virtio_dma_buf_ops virtgpu_dmabuf_ops =  {
+	.ops = {
+		.cache_sgt_mapping = true,
+		.attach = virtio_dma_buf_attach,
+		.detach = drm_gem_map_detach,
+		.map_dma_buf = drm_gem_map_dma_buf,
+		.unmap_dma_buf = drm_gem_unmap_dma_buf,
+		.release = drm_gem_dmabuf_release,
+		.mmap = drm_gem_dmabuf_mmap,
+		.vmap = drm_gem_dmabuf_vmap,
+		.vunmap = drm_gem_dmabuf_vunmap,
+	},
+	.device_attach = drm_gem_map_attach,
+	.get_uuid = virtgpu_virtio_get_uuid,
+};
+
+struct dma_buf *virtgpu_gem_prime_export(struct drm_gem_object *obj,
+					 int flags)
+{
+	struct dma_buf *buf;
+	struct drm_device *dev = obj->dev;
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
+	struct virtio_gpu_object_array *objs;
+	int ret = 0;
+	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
+
+	if (vgdev->has_resource_assign_uuid) {
+		objs = virtio_gpu_array_alloc(1);
+		if (!objs)
+			return ERR_PTR(-ENOMEM);
+		virtio_gpu_array_add_obj(objs, &bo->base.base);
+
+		ret = virtio_gpu_cmd_resource_assign_uuid(vgdev, objs);
+		if (ret)
+			return ERR_PTR(ret);
+		virtio_gpu_notify(vgdev);
+	} else {
+		bo->uuid_state = UUID_INITIALIZATION_FAILED;
+	}
+
+	exp_info.ops = &virtgpu_dmabuf_ops.ops;
+	exp_info.size = obj->size;
+	exp_info.flags = flags;
+	exp_info.priv = obj;
+	exp_info.resv = obj->resv;
+
+	buf = virtio_dma_buf_export(&exp_info);
+	if (IS_ERR(buf))
+		return buf;
+
+	drm_dev_get(dev);
+	drm_gem_object_get(obj);
+
+	return buf;
+}
+
+struct drm_gem_object *virtgpu_gem_prime_import(struct drm_device *dev,
+						struct dma_buf *buf)
+{
+	struct drm_gem_object *obj;
+
+	if (buf->ops == &virtgpu_dmabuf_ops.ops) {
+		obj = buf->priv;
+		if (obj->dev == dev) {
+			/*
+			 * Importing dmabuf exported from our own gem increases
+			 * refcount on gem itself instead of f_count of dmabuf.
+			 */
+			drm_gem_object_get(obj);
+			return obj;
+		}
+	}
+
+	return drm_gem_prime_import(dev, buf);
+}
 
 struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
 	struct drm_device *dev, struct dma_buf_attachment *attach,
diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index 53af60d484a4..07945ca238e2 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -302,7 +302,7 @@ static struct sg_table *vmalloc_to_sgt(char *data, uint32_t size, int *sg_ents)
 		return NULL;
 	}
 
-	for_each_sg(sgt->sgl, sg, *sg_ents, i) {
+	for_each_sgtable_sg(sgt, sg, i) {
 		pg = vmalloc_to_page(data);
 		if (!pg) {
 			sg_free_table(sgt);
@@ -320,13 +320,13 @@ static struct sg_table *vmalloc_to_sgt(char *data, uint32_t size, int *sg_ents)
 	return sgt;
 }
 
-static void virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev,
-				      struct virtio_gpu_vbuffer *vbuf,
-				      struct virtio_gpu_fence *fence,
-				      int elemcnt,
-				      struct scatterlist **sgs,
-				      int outcnt,
-				      int incnt)
+static int virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev,
+				     struct virtio_gpu_vbuffer *vbuf,
+				     struct virtio_gpu_fence *fence,
+				     int elemcnt,
+				     struct scatterlist **sgs,
+				     int outcnt,
+				     int incnt)
 {
 	struct virtqueue *vq = vgdev->ctrlq.vq;
 	int ret, idx;
@@ -335,7 +335,7 @@ static void virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev,
 		if (fence && vbuf->objs)
 			virtio_gpu_array_unlock_resv(vbuf->objs);
 		free_vbuf(vgdev, vbuf);
-		return;
+		return -1;
 	}
 
 	if (vgdev->has_indirect)
@@ -373,15 +373,16 @@ again:
 	spin_unlock(&vgdev->ctrlq.qlock);
 
 	drm_dev_exit(idx);
+	return 0;
 }
 
-static void virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
-						struct virtio_gpu_vbuffer *vbuf,
-						struct virtio_gpu_fence *fence)
+static int virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
+					       struct virtio_gpu_vbuffer *vbuf,
+					       struct virtio_gpu_fence *fence)
 {
 	struct scatterlist *sgs[3], vcmd, vout, vresp;
 	struct sg_table *sgt = NULL;
-	int elemcnt = 0, outcnt = 0, incnt = 0;
+	int elemcnt = 0, outcnt = 0, incnt = 0, ret;
 
 	/* set up vcmd */
 	sg_init_one(&vcmd, vbuf->buf, vbuf->size);
@@ -398,7 +399,7 @@ static void virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
 			if (!sgt) {
 				if (fence && vbuf->objs)
 					virtio_gpu_array_unlock_resv(vbuf->objs);
-				return;
+				return -1;
 			}
 
 			elemcnt += sg_ents;
@@ -419,13 +420,14 @@ static void virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
 		incnt++;
 	}
 
-	virtio_gpu_queue_ctrl_sgs(vgdev, vbuf, fence, elemcnt, sgs, outcnt,
-				  incnt);
+	ret = virtio_gpu_queue_ctrl_sgs(vgdev, vbuf, fence, elemcnt, sgs, outcnt,
+					incnt);
 
 	if (sgt) {
 		sg_free_table(sgt);
 		kfree(sgt);
 	}
+	return ret;
 }
 
 void virtio_gpu_notify(struct virtio_gpu_device *vgdev)
@@ -444,10 +446,10 @@ void virtio_gpu_notify(struct virtio_gpu_device *vgdev)
 		virtqueue_notify(vgdev->ctrlq.vq);
 }
 
-static void virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
-					 struct virtio_gpu_vbuffer *vbuf)
+static int virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
+					struct virtio_gpu_vbuffer *vbuf)
 {
-	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, NULL);
+	return virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, NULL);
 }
 
 static void virtio_gpu_queue_cursor(struct virtio_gpu_device *vgdev,
@@ -534,6 +536,7 @@ void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev,
 {
 	struct virtio_gpu_resource_unref *cmd_p;
 	struct virtio_gpu_vbuffer *vbuf;
+	int ret;
 
 	cmd_p = virtio_gpu_alloc_cmd_cb(vgdev, &vbuf, sizeof(*cmd_p),
 					virtio_gpu_cmd_unref_cb);
@@ -543,7 +546,9 @@ void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev,
 	cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle);
 
 	vbuf->resp_cb_data = bo;
-	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	ret = virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	if (ret < 0)
+		virtio_gpu_cleanup_object(bo);
 }
 
 void virtio_gpu_cmd_set_scanout(struct virtio_gpu_device *vgdev,
@@ -603,9 +608,8 @@ void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev,
 	struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo);
 
 	if (use_dma_api)
-		dma_sync_sg_for_device(vgdev->vdev->dev.parent,
-				       shmem->pages->sgl, shmem->pages->nents,
-				       DMA_TO_DEVICE);
+		dma_sync_sgtable_for_device(vgdev->vdev->dev.parent,
+					    shmem->pages, DMA_TO_DEVICE);
 
 	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
 	memset(cmd_p, 0, sizeof(*cmd_p));
@@ -684,9 +688,13 @@ static void virtio_gpu_cmd_get_capset_info_cb(struct virtio_gpu_device *vgdev,
 	int i = le32_to_cpu(cmd->capset_index);
 
 	spin_lock(&vgdev->display_info_lock);
-	vgdev->capsets[i].id = le32_to_cpu(resp->capset_id);
-	vgdev->capsets[i].max_version = le32_to_cpu(resp->capset_max_version);
-	vgdev->capsets[i].max_size = le32_to_cpu(resp->capset_max_size);
+	if (vgdev->capsets) {
+		vgdev->capsets[i].id = le32_to_cpu(resp->capset_id);
+		vgdev->capsets[i].max_version = le32_to_cpu(resp->capset_max_version);
+		vgdev->capsets[i].max_size = le32_to_cpu(resp->capset_max_size);
+	} else {
+		DRM_ERROR("invalid capset memory.");
+	}
 	spin_unlock(&vgdev->display_info_lock);
 	wake_up(&vgdev->resp_wq);
 }
@@ -1019,9 +1027,8 @@ void virtio_gpu_cmd_transfer_to_host_3d(struct virtio_gpu_device *vgdev,
 	struct virtio_gpu_object_shmem *shmem = to_virtio_gpu_shmem(bo);
 
 	if (use_dma_api)
-		dma_sync_sg_for_device(vgdev->vdev->dev.parent,
-				       shmem->pages->sgl, shmem->pages->nents,
-				       DMA_TO_DEVICE);
+		dma_sync_sgtable_for_device(vgdev->vdev->dev.parent,
+					    shmem->pages, DMA_TO_DEVICE);
 
 	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
 	memset(cmd_p, 0, sizeof(*cmd_p));
@@ -1107,3 +1114,58 @@ void virtio_gpu_cursor_ping(struct virtio_gpu_device *vgdev,
 	memcpy(cur_p, &output->cursor, sizeof(output->cursor));
 	virtio_gpu_queue_cursor(vgdev, vbuf);
 }
+
+static void virtio_gpu_cmd_resource_uuid_cb(struct virtio_gpu_device *vgdev,
+					    struct virtio_gpu_vbuffer *vbuf)
+{
+	struct virtio_gpu_object *obj =
+		gem_to_virtio_gpu_obj(vbuf->objs->objs[0]);
+	struct virtio_gpu_resp_resource_uuid *resp =
+		(struct virtio_gpu_resp_resource_uuid *)vbuf->resp_buf;
+	uint32_t resp_type = le32_to_cpu(resp->hdr.type);
+
+	spin_lock(&vgdev->resource_export_lock);
+	WARN_ON(obj->uuid_state != UUID_INITIALIZING);
+
+	if (resp_type == VIRTIO_GPU_RESP_OK_RESOURCE_UUID &&
+	    obj->uuid_state == UUID_INITIALIZING) {
+		memcpy(&obj->uuid.b, resp->uuid, sizeof(obj->uuid.b));
+		obj->uuid_state = UUID_INITIALIZED;
+	} else {
+		obj->uuid_state = UUID_INITIALIZATION_FAILED;
+	}
+	spin_unlock(&vgdev->resource_export_lock);
+
+	wake_up_all(&vgdev->resp_wq);
+}
+
+int
+virtio_gpu_cmd_resource_assign_uuid(struct virtio_gpu_device *vgdev,
+				    struct virtio_gpu_object_array *objs)
+{
+	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(objs->objs[0]);
+	struct virtio_gpu_resource_assign_uuid *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+	struct virtio_gpu_resp_resource_uuid *resp_buf;
+
+	resp_buf = kzalloc(sizeof(*resp_buf), GFP_KERNEL);
+	if (!resp_buf) {
+		spin_lock(&vgdev->resource_export_lock);
+		bo->uuid_state = UUID_INITIALIZATION_FAILED;
+		spin_unlock(&vgdev->resource_export_lock);
+		virtio_gpu_array_put_free(objs);
+		return -ENOMEM;
+	}
+
+	cmd_p = virtio_gpu_alloc_cmd_resp
+		(vgdev, virtio_gpu_cmd_resource_uuid_cb, &vbuf, sizeof(*cmd_p),
+		 sizeof(struct virtio_gpu_resp_resource_uuid), resp_buf);
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_ASSIGN_UUID);
+	cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle);
+
+	vbuf->objs = objs;
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	return 0;
+}
diff --git a/drivers/gpu/drm/vkms/Makefile b/drivers/gpu/drm/vkms/Makefile
index 0b767d7efa24..333d3cead0e3 100644
--- a/drivers/gpu/drm/vkms/Makefile
+++ b/drivers/gpu/drm/vkms/Makefile
@@ -1,4 +1,11 @@
 # SPDX-License-Identifier: GPL-2.0-only
-vkms-y := vkms_drv.o vkms_plane.o vkms_output.o vkms_crtc.o vkms_gem.o vkms_composer.o
+vkms-y := \
+	vkms_drv.o \
+	vkms_plane.o \
+	vkms_output.o \
+	vkms_crtc.o \
+	vkms_gem.o \
+	vkms_composer.o \
+	vkms_writeback.o
 
 obj-$(CONFIG_DRM_VKMS) += vkms.o
diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index 4af2f19480f4..33c031f27c2c 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -9,60 +9,92 @@
 
 #include "vkms_drv.h"
 
+static u32 get_pixel_from_buffer(int x, int y, const u8 *buffer,
+				 const struct vkms_composer *composer)
+{
+	u32 pixel;
+	int src_offset = composer->offset + (y * composer->pitch)
+				      + (x * composer->cpp);
+
+	pixel = *(u32 *)&buffer[src_offset];
+
+	return pixel;
+}
+
 /**
  * compute_crc - Compute CRC value on output frame
  *
- * @vaddr_out: address to final framebuffer
+ * @vaddr: address to final framebuffer
  * @composer: framebuffer's metadata
  *
  * returns CRC value computed using crc32 on the visible portion of
  * the final framebuffer at vaddr_out
  */
-static uint32_t compute_crc(void *vaddr_out, struct vkms_composer *composer)
+static uint32_t compute_crc(const u8 *vaddr,
+			    const struct vkms_composer *composer)
 {
-	int i, j, src_offset;
+	int x, y;
+	u32 crc = 0, pixel = 0;
 	int x_src = composer->src.x1 >> 16;
 	int y_src = composer->src.y1 >> 16;
 	int h_src = drm_rect_height(&composer->src) >> 16;
 	int w_src = drm_rect_width(&composer->src) >> 16;
-	u32 crc = 0;
-
-	for (i = y_src; i < y_src + h_src; ++i) {
-		for (j = x_src; j < x_src + w_src; ++j) {
-			src_offset = composer->offset
-				     + (i * composer->pitch)
-				     + (j * composer->cpp);
-			/* XRGB format ignores Alpha channel */
-			memset(vaddr_out + src_offset + 24, 0,  8);
-			crc = crc32_le(crc, vaddr_out + src_offset,
-				       sizeof(u32));
+
+	for (y = y_src; y < y_src + h_src; ++y) {
+		for (x = x_src; x < x_src + w_src; ++x) {
+			pixel = get_pixel_from_buffer(x, y, vaddr, composer);
+			crc = crc32_le(crc, (void *)&pixel, sizeof(u32));
 		}
 	}
 
 	return crc;
 }
 
+static u8 blend_channel(u8 src, u8 dst, u8 alpha)
+{
+	u32 pre_blend;
+	u8 new_color;
+
+	pre_blend = (src * 255 + dst * (255 - alpha));
+
+	/* Faster div by 255 */
+	new_color = ((pre_blend + ((pre_blend + 257) >> 8)) >> 8);
+
+	return new_color;
+}
+
+static void alpha_blending(const u8 *argb_src, u8 *argb_dst)
+{
+	u8 alpha;
+
+	alpha = argb_src[3];
+	argb_dst[0] = blend_channel(argb_src[0], argb_dst[0], alpha);
+	argb_dst[1] = blend_channel(argb_src[1], argb_dst[1], alpha);
+	argb_dst[2] = blend_channel(argb_src[2], argb_dst[2], alpha);
+	/* Opaque primary */
+	argb_dst[3] = 0xFF;
+}
+
 /**
  * blend - blend value at vaddr_src with value at vaddr_dst
  * @vaddr_dst: destination address
  * @vaddr_src: source address
- * @dest_composer: destination framebuffer's metadata
+ * @dst_composer: destination framebuffer's metadata
  * @src_composer: source framebuffer's metadata
  *
- * Blend value at vaddr_src with value at vaddr_dst.
- * Currently, this function write value of vaddr_src on value
- * at vaddr_dst using buffer's metadata to locate the new values
- * from vaddr_src and their destination at vaddr_dst.
- *
- * TODO: Use the alpha value to blend vaddr_src with vaddr_dst
- *	 instead of overwriting it.
+ * Blend the vaddr_src value with the vaddr_dst value using the pre-multiplied
+ * alpha blending equation, since DRM currently assumes that the pixel color
+ * values have already been pre-multiplied with the alpha channel values. See
+ * more drm_plane_create_blend_mode_property(). This function uses buffer's
+ * metadata to locate the new composite values at vaddr_dst.
  */
 static void blend(void *vaddr_dst, void *vaddr_src,
-		  struct vkms_composer *dest_composer,
+		  struct vkms_composer *dst_composer,
 		  struct vkms_composer *src_composer)
 {
 	int i, j, j_dst, i_dst;
 	int offset_src, offset_dst;
+	u8 *pixel_dst, *pixel_src;
 
 	int x_src = src_composer->src.x1 >> 16;
 	int y_src = src_composer->src.y1 >> 16;
@@ -77,15 +109,16 @@ static void blend(void *vaddr_dst, void *vaddr_src,
 
 	for (i = y_src, i_dst = y_dst; i < y_limit; ++i) {
 		for (j = x_src, j_dst = x_dst; j < x_limit; ++j) {
-			offset_dst = dest_composer->offset
-				     + (i_dst * dest_composer->pitch)
-				     + (j_dst++ * dest_composer->cpp);
+			offset_dst = dst_composer->offset
+				     + (i_dst * dst_composer->pitch)
+				     + (j_dst++ * dst_composer->cpp);
 			offset_src = src_composer->offset
 				     + (i * src_composer->pitch)
 				     + (j * src_composer->cpp);
 
-			memcpy(vaddr_dst + offset_dst,
-			       vaddr_src + offset_src, sizeof(u32));
+			pixel_src = (u8 *)(vaddr_src + offset_src);
+			pixel_dst = (u8 *)(vaddr_dst + offset_dst);
+			alpha_blending(pixel_src, pixel_dst);
 		}
 		i_dst++;
 	}
@@ -108,35 +141,31 @@ static void compose_cursor(struct vkms_composer *cursor_composer,
 	      primary_composer, cursor_composer);
 }
 
-static uint32_t _vkms_get_crc(struct vkms_composer *primary_composer,
-			      struct vkms_composer *cursor_composer)
+static int compose_planes(void **vaddr_out,
+			  struct vkms_composer *primary_composer,
+			  struct vkms_composer *cursor_composer)
 {
 	struct drm_framebuffer *fb = &primary_composer->fb;
 	struct drm_gem_object *gem_obj = drm_gem_fb_get_obj(fb, 0);
 	struct vkms_gem_object *vkms_obj = drm_gem_to_vkms_gem(gem_obj);
-	void *vaddr_out = kzalloc(vkms_obj->gem.size, GFP_KERNEL);
-	u32 crc = 0;
 
-	if (!vaddr_out) {
-		DRM_ERROR("Failed to allocate memory for output frame.");
-		return 0;
+	if (!*vaddr_out) {
+		*vaddr_out = kzalloc(vkms_obj->gem.size, GFP_KERNEL);
+		if (!*vaddr_out) {
+			DRM_ERROR("Cannot allocate memory for output frame.");
+			return -ENOMEM;
+		}
 	}
 
-	if (WARN_ON(!vkms_obj->vaddr)) {
-		kfree(vaddr_out);
-		return crc;
-	}
+	if (WARN_ON(!vkms_obj->vaddr))
+		return -EINVAL;
 
-	memcpy(vaddr_out, vkms_obj->vaddr, vkms_obj->gem.size);
+	memcpy(*vaddr_out, vkms_obj->vaddr, vkms_obj->gem.size);
 
 	if (cursor_composer)
-		compose_cursor(cursor_composer, primary_composer, vaddr_out);
-
-	crc = compute_crc(vaddr_out, primary_composer);
+		compose_cursor(cursor_composer, primary_composer, *vaddr_out);
 
-	kfree(vaddr_out);
-
-	return crc;
+	return 0;
 }
 
 /**
@@ -157,14 +186,17 @@ void vkms_composer_worker(struct work_struct *work)
 	struct vkms_output *out = drm_crtc_to_vkms_output(crtc);
 	struct vkms_composer *primary_composer = NULL;
 	struct vkms_composer *cursor_composer = NULL;
+	bool crc_pending, wb_pending;
+	void *vaddr_out = NULL;
 	u32 crc32 = 0;
 	u64 frame_start, frame_end;
-	bool crc_pending;
+	int ret;
 
 	spin_lock_irq(&out->composer_lock);
 	frame_start = crtc_state->frame_start;
 	frame_end = crtc_state->frame_end;
 	crc_pending = crtc_state->crc_pending;
+	wb_pending = crtc_state->wb_pending;
 	crtc_state->frame_start = 0;
 	crtc_state->frame_end = 0;
 	crtc_state->crc_pending = false;
@@ -183,8 +215,29 @@ void vkms_composer_worker(struct work_struct *work)
 	if (crtc_state->num_active_planes == 2)
 		cursor_composer = crtc_state->active_planes[1]->composer;
 
-	if (primary_composer)
-		crc32 = _vkms_get_crc(primary_composer, cursor_composer);
+	if (!primary_composer)
+		return;
+
+	if (wb_pending)
+		vaddr_out = crtc_state->active_writeback;
+
+	ret = compose_planes(&vaddr_out, primary_composer, cursor_composer);
+	if (ret) {
+		if (ret == -EINVAL && !wb_pending)
+			kfree(vaddr_out);
+		return;
+	}
+
+	crc32 = compute_crc(vaddr_out, primary_composer);
+
+	if (wb_pending) {
+		drm_writeback_signal_completion(&out->wb_connector, 0);
+		spin_lock_irq(&out->composer_lock);
+		crtc_state->wb_pending = false;
+		spin_unlock_irq(&out->composer_lock);
+	} else {
+		kfree(vaddr_out);
+	}
 
 	/*
 	 * The worker can fall behind the vblank hrtimer, make sure we catch up.
@@ -233,6 +286,22 @@ int vkms_verify_crc_source(struct drm_crtc *crtc, const char *src_name,
 	return 0;
 }
 
+void vkms_set_composer(struct vkms_output *out, bool enabled)
+{
+	bool old_enabled;
+
+	if (enabled)
+		drm_crtc_vblank_get(&out->crtc);
+
+	spin_lock_irq(&out->lock);
+	old_enabled = out->composer_enabled;
+	out->composer_enabled = enabled;
+	spin_unlock_irq(&out->lock);
+
+	if (old_enabled)
+		drm_crtc_vblank_put(&out->crtc);
+}
+
 int vkms_set_crc_source(struct drm_crtc *crtc, const char *src_name)
 {
 	struct vkms_output *out = drm_crtc_to_vkms_output(crtc);
@@ -241,9 +310,7 @@ int vkms_set_crc_source(struct drm_crtc *crtc, const char *src_name)
 
 	ret = vkms_crc_parse_source(src_name, &enabled);
 
-	spin_lock_irq(&out->lock);
-	out->composer_enabled = enabled;
-	spin_unlock_irq(&out->lock);
+	vkms_set_composer(out, enabled);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c
index ac85e17428f8..09c012d54d58 100644
--- a/drivers/gpu/drm/vkms/vkms_crtc.c
+++ b/drivers/gpu/drm/vkms/vkms_crtc.c
@@ -86,6 +86,11 @@ static bool vkms_get_vblank_timestamp(struct drm_crtc *crtc,
 	struct vkms_output *output = &vkmsdev->output;
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 
+	if (!READ_ONCE(vblank->enabled)) {
+		*vblank_time = ktime_get();
+		return true;
+	}
+
 	*vblank_time = READ_ONCE(output->vblank_hrtimer.node.expires);
 
 	if (WARN_ON(*vblank_time == vblank->time))
diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c
index 57a8a397d5e8..cb0b6230c22c 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.c
+++ b/drivers/gpu/drm/vkms/vkms_drv.c
@@ -61,9 +61,6 @@ static void vkms_release(struct drm_device *dev)
 {
 	struct vkms_device *vkms = container_of(dev, struct vkms_device, drm);
 
-	platform_device_unregister(vkms->platform);
-	drm_atomic_helper_shutdown(&vkms->drm);
-	drm_mode_config_cleanup(&vkms->drm);
 	destroy_workqueue(vkms->output.composer_workq);
 }
 
@@ -144,30 +141,31 @@ static int vkms_modeset_init(struct vkms_device *vkmsdev)
 static int __init vkms_init(void)
 {
 	int ret;
+	struct platform_device *pdev;
 
-	vkms_device = kzalloc(sizeof(*vkms_device), GFP_KERNEL);
-	if (!vkms_device)
-		return -ENOMEM;
+	pdev = platform_device_register_simple(DRIVER_NAME, -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
 
-	vkms_device->platform =
-		platform_device_register_simple(DRIVER_NAME, -1, NULL, 0);
-	if (IS_ERR(vkms_device->platform)) {
-		ret = PTR_ERR(vkms_device->platform);
-		goto out_free;
+	if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) {
+		ret = -ENOMEM;
+		goto out_unregister;
 	}
 
-	ret = drm_dev_init(&vkms_device->drm, &vkms_driver,
-			   &vkms_device->platform->dev);
-	if (ret)
-		goto out_unregister;
-	drmm_add_final_kfree(&vkms_device->drm, vkms_device);
+	vkms_device = devm_drm_dev_alloc(&pdev->dev, &vkms_driver,
+					 struct vkms_device, drm);
+	if (IS_ERR(vkms_device)) {
+		ret = PTR_ERR(vkms_device);
+		goto out_devres;
+	}
+	vkms_device->platform = pdev;
 
 	ret = dma_coerce_mask_and_coherent(vkms_device->drm.dev,
 					   DMA_BIT_MASK(64));
 
 	if (ret) {
 		DRM_ERROR("Could not initialize DMA support\n");
-		goto out_put;
+		goto out_devres;
 	}
 
 	vkms_device->drm.irq_enabled = true;
@@ -175,39 +173,41 @@ static int __init vkms_init(void)
 	ret = drm_vblank_init(&vkms_device->drm, 1);
 	if (ret) {
 		DRM_ERROR("Failed to vblank\n");
-		goto out_put;
+		goto out_devres;
 	}
 
 	ret = vkms_modeset_init(vkms_device);
 	if (ret)
-		goto out_put;
+		goto out_devres;
 
 	ret = drm_dev_register(&vkms_device->drm, 0);
 	if (ret)
-		goto out_put;
+		goto out_devres;
 
 	return 0;
 
-out_put:
-	drm_dev_put(&vkms_device->drm);
-	return ret;
-
+out_devres:
+	devres_release_group(&pdev->dev, NULL);
 out_unregister:
-	platform_device_unregister(vkms_device->platform);
-out_free:
-	kfree(vkms_device);
+	platform_device_unregister(pdev);
 	return ret;
 }
 
 static void __exit vkms_exit(void)
 {
+	struct platform_device *pdev;
+
 	if (!vkms_device) {
 		DRM_INFO("vkms_device is NULL.\n");
 		return;
 	}
 
+	pdev = vkms_device->platform;
+
 	drm_dev_unregister(&vkms_device->drm);
-	drm_dev_put(&vkms_device->drm);
+	drm_atomic_helper_shutdown(&vkms_device->drm);
+	devres_release_group(&pdev->dev, NULL);
+	platform_device_unregister(pdev);
 }
 
 module_init(vkms_init);
diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h
index f4036bb0b9a8..380a8f27e156 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.h
+++ b/drivers/gpu/drm/vkms/vkms_drv.h
@@ -8,6 +8,7 @@
 #include <drm/drm.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_encoder.h>
+#include <drm/drm_writeback.h>
 
 #define XRES_MIN    20
 #define YRES_MIN    20
@@ -52,9 +53,11 @@ struct vkms_crtc_state {
 	int num_active_planes;
 	/* stack of active planes for crc computation, should be in z order */
 	struct vkms_plane_state **active_planes;
+	void *active_writeback;
 
-	/* below three are protected by vkms_output.composer_lock */
+	/* below four are protected by vkms_output.composer_lock */
 	bool crc_pending;
+	bool wb_pending;
 	u64 frame_start;
 	u64 frame_end;
 };
@@ -63,6 +66,7 @@ struct vkms_output {
 	struct drm_crtc crtc;
 	struct drm_encoder encoder;
 	struct drm_connector connector;
+	struct drm_writeback_connector wb_connector;
 	struct hrtimer vblank_hrtimer;
 	ktime_t period_ns;
 	struct drm_pending_vblank_event *event;
@@ -143,5 +147,9 @@ int vkms_verify_crc_source(struct drm_crtc *crtc, const char *source_name,
 
 /* Composer Support */
 void vkms_composer_worker(struct work_struct *work);
+void vkms_set_composer(struct vkms_output *out, bool enabled);
+
+/* Writeback */
+int vkms_enable_writeback_connector(struct vkms_device *vkmsdev);
 
 #endif /* _VKMS_DRV_H_ */
diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c
index 85afb77e97f0..4a1848b0318f 100644
--- a/drivers/gpu/drm/vkms/vkms_output.c
+++ b/drivers/gpu/drm/vkms/vkms_output.c
@@ -80,6 +80,10 @@ int vkms_output_init(struct vkms_device *vkmsdev, int index)
 		goto err_attach;
 	}
 
+	ret = vkms_enable_writeback_connector(vkmsdev);
+	if (ret)
+		DRM_ERROR("Failed to init writeback connector\n");
+
 	drm_mode_config_reset(dev);
 
 	return 0;
diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c
new file mode 100644
index 000000000000..094fa4aa061d
--- /dev/null
+++ b/drivers/gpu/drm/vkms/vkms_writeback.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0+
+
+#include "vkms_drv.h"
+#include <drm/drm_fourcc.h>
+#include <drm/drm_writeback.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+
+static const u32 vkms_wb_formats[] = {
+	DRM_FORMAT_XRGB8888,
+};
+
+static const struct drm_connector_funcs vkms_wb_connector_funcs = {
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static int vkms_wb_encoder_atomic_check(struct drm_encoder *encoder,
+					struct drm_crtc_state *crtc_state,
+					struct drm_connector_state *conn_state)
+{
+	struct drm_framebuffer *fb;
+	const struct drm_display_mode *mode = &crtc_state->mode;
+
+	if (!conn_state->writeback_job || !conn_state->writeback_job->fb)
+		return 0;
+
+	fb = conn_state->writeback_job->fb;
+	if (fb->width != mode->hdisplay || fb->height != mode->vdisplay) {
+		DRM_DEBUG_KMS("Invalid framebuffer size %ux%u\n",
+			      fb->width, fb->height);
+		return -EINVAL;
+	}
+
+	if (fb->format->format != vkms_wb_formats[0]) {
+		struct drm_format_name_buf format_name;
+
+		DRM_DEBUG_KMS("Invalid pixel format %s\n",
+			      drm_get_format_name(fb->format->format,
+						  &format_name));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const struct drm_encoder_helper_funcs vkms_wb_encoder_helper_funcs = {
+	.atomic_check = vkms_wb_encoder_atomic_check,
+};
+
+static int vkms_wb_connector_get_modes(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+
+	return drm_add_modes_noedid(connector, dev->mode_config.max_width,
+				    dev->mode_config.max_height);
+}
+
+static int vkms_wb_prepare_job(struct drm_writeback_connector *wb_connector,
+			       struct drm_writeback_job *job)
+{
+	struct vkms_gem_object *vkms_obj;
+	struct drm_gem_object *gem_obj;
+	int ret;
+
+	if (!job->fb)
+		return 0;
+
+	gem_obj = drm_gem_fb_get_obj(job->fb, 0);
+	ret = vkms_gem_vmap(gem_obj);
+	if (ret) {
+		DRM_ERROR("vmap failed: %d\n", ret);
+		return ret;
+	}
+
+	vkms_obj = drm_gem_to_vkms_gem(gem_obj);
+	job->priv = vkms_obj->vaddr;
+
+	return 0;
+}
+
+static void vkms_wb_cleanup_job(struct drm_writeback_connector *connector,
+				struct drm_writeback_job *job)
+{
+	struct drm_gem_object *gem_obj;
+	struct vkms_device *vkmsdev;
+
+	if (!job->fb)
+		return;
+
+	gem_obj = drm_gem_fb_get_obj(job->fb, 0);
+	vkms_gem_vunmap(gem_obj);
+
+	vkmsdev = drm_device_to_vkms_device(gem_obj->dev);
+	vkms_set_composer(&vkmsdev->output, false);
+}
+
+static void vkms_wb_atomic_commit(struct drm_connector *conn,
+				  struct drm_connector_state *state)
+{
+	struct vkms_device *vkmsdev = drm_device_to_vkms_device(conn->dev);
+	struct vkms_output *output = &vkmsdev->output;
+	struct drm_writeback_connector *wb_conn = &output->wb_connector;
+	struct drm_connector_state *conn_state = wb_conn->base.state;
+	struct vkms_crtc_state *crtc_state = output->composer_state;
+
+	if (!conn_state)
+		return;
+
+	vkms_set_composer(&vkmsdev->output, true);
+
+	spin_lock_irq(&output->composer_lock);
+	crtc_state->active_writeback = conn_state->writeback_job->priv;
+	crtc_state->wb_pending = true;
+	spin_unlock_irq(&output->composer_lock);
+	drm_writeback_queue_job(wb_conn, state);
+}
+
+static const struct drm_connector_helper_funcs vkms_wb_conn_helper_funcs = {
+	.get_modes = vkms_wb_connector_get_modes,
+	.prepare_writeback_job = vkms_wb_prepare_job,
+	.cleanup_writeback_job = vkms_wb_cleanup_job,
+	.atomic_commit = vkms_wb_atomic_commit,
+};
+
+int vkms_enable_writeback_connector(struct vkms_device *vkmsdev)
+{
+	struct drm_writeback_connector *wb = &vkmsdev->output.wb_connector;
+
+	vkmsdev->output.wb_connector.encoder.possible_crtcs = 1;
+	drm_connector_helper_add(&wb->base, &vkms_wb_conn_helper_funcs);
+
+	return drm_writeback_connector_init(&vkmsdev->drm, wb,
+					    &vkms_wb_connector_funcs,
+					    &vkms_wb_encoder_helper_funcs,
+					    vkms_wb_formats,
+					    ARRAY_SIZE(vkms_wb_formats));
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
index 1629427d5734..e8d66182cd7b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
@@ -464,14 +464,14 @@ int vmw_bo_cpu_blit(struct ttm_buffer_object *dst,
 	if (!(src->mem.placement & TTM_PL_FLAG_NO_EVICT))
 		dma_resv_assert_held(src->base.resv);
 
-	if (dst->ttm->state == tt_unpopulated) {
-		ret = dst->ttm->bdev->driver->ttm_tt_populate(dst->ttm, &ctx);
+	if (!ttm_tt_is_populated(dst->ttm)) {
+		ret = dst->bdev->driver->ttm_tt_populate(dst->bdev, dst->ttm, &ctx);
 		if (ret)
 			return ret;
 	}
 
-	if (src->ttm->state == tt_unpopulated) {
-		ret = src->ttm->bdev->driver->ttm_tt_populate(src->ttm, &ctx);
+	if (!ttm_tt_is_populated(src->ttm)) {
+		ret = src->bdev->driver->ttm_tt_populate(src->bdev, src->ttm, &ctx);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
index 1e59c019affa..813f1b148094 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
@@ -354,10 +354,12 @@ void vmw_bo_pin_reserved(struct vmw_buffer_object *vbo, bool pin)
 
 	pl.fpfn = 0;
 	pl.lpfn = 0;
-	pl.flags = TTM_PL_FLAG_VRAM | VMW_PL_FLAG_GMR | VMW_PL_FLAG_MOB
-		| TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED;
+	pl.mem_type = bo->mem.mem_type;
+	pl.flags = bo->mem.placement;
 	if (pin)
 		pl.flags |= TTM_PL_FLAG_NO_EVICT;
+	else
+		pl.flags &= ~TTM_PL_FLAG_NO_EVICT;
 
 	memset(&placement, 0, sizeof(placement));
 	placement.num_placement = 1;
@@ -1135,14 +1137,14 @@ void vmw_bo_swap_notify(struct ttm_buffer_object *bo)
  * vmw_bo_move_notify - TTM move_notify_callback
  *
  * @bo: The TTM buffer object about to move.
- * @mem: The struct ttm_mem_reg indicating to what memory
+ * @mem: The struct ttm_resource indicating to what memory
  *       region the move is taking place.
  *
  * Detaches cached maps and device bindings that require that the
  * buffer doesn't move.
  */
 void vmw_bo_move_notify(struct ttm_buffer_object *bo,
-			struct ttm_mem_reg *mem)
+			struct ttm_resource *mem)
 {
 	struct vmw_buffer_object *vbo;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index fb39826f72c1..31e3e5c9f362 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -620,6 +620,28 @@ static int vmw_dma_masks(struct vmw_private *dev_priv)
 	return ret;
 }
 
+static int vmw_vram_manager_init(struct vmw_private *dev_priv)
+{
+	int ret;
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	ret = vmw_thp_init(dev_priv);
+#else
+	ret = ttm_range_man_init(&dev_priv->bdev, TTM_PL_VRAM, false,
+				 dev_priv->vram_size >> PAGE_SHIFT);
+#endif
+	ttm_resource_manager_set_used(ttm_manager_type(&dev_priv->bdev, TTM_PL_VRAM), false);
+	return ret;
+}
+
+static void vmw_vram_manager_fini(struct vmw_private *dev_priv)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	vmw_thp_fini(dev_priv);
+#else
+	ttm_range_man_fini(&dev_priv->bdev, TTM_PL_VRAM);
+#endif
+}
+
 static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 {
 	struct vmw_private *dev_priv;
@@ -864,18 +886,23 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	 * Enable VRAM, but initially don't use it until SVGA is enabled and
 	 * unhidden.
 	 */
-	ret = ttm_bo_init_mm(&dev_priv->bdev, TTM_PL_VRAM,
-			     (dev_priv->vram_size >> PAGE_SHIFT));
+
+	ret = vmw_vram_manager_init(dev_priv);
 	if (unlikely(ret != 0)) {
 		DRM_ERROR("Failed initializing memory manager for VRAM.\n");
 		goto out_no_vram;
 	}
-	dev_priv->bdev.man[TTM_PL_VRAM].use_type = false;
 
+	/*
+	 * "Guest Memory Regions" is an aperture like feature with
+	 *  one slot per bo. There is an upper limit of the number of
+	 *  slots as well as the bo size.
+	 */
 	dev_priv->has_gmr = true;
+	/* TODO: This is most likely not correct */
 	if (((dev_priv->capabilities & (SVGA_CAP_GMR | SVGA_CAP_GMR2)) == 0) ||
-	    refuse_dma || ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_GMR,
-					 VMW_PL_GMR) != 0) {
+	    refuse_dma ||
+	    vmw_gmrid_man_init(dev_priv, VMW_PL_GMR) != 0) {
 		DRM_INFO("No GMR memory available. "
 			 "Graphics memory resources are very limited.\n");
 		dev_priv->has_gmr = false;
@@ -883,8 +910,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 
 	if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS && !refuse_dma) {
 		dev_priv->has_mob = true;
-		if (ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_MOB,
-				   VMW_PL_MOB) != 0) {
+
+		if (vmw_gmrid_man_init(dev_priv, VMW_PL_MOB) != 0) {
 			DRM_INFO("No MOB memory available. "
 				 "3D will be disabled.\n");
 			dev_priv->has_mob = false;
@@ -961,10 +988,10 @@ out_no_fifo:
 	vmw_kms_close(dev_priv);
 out_no_kms:
 	if (dev_priv->has_mob)
-		(void) ttm_bo_clean_mm(&dev_priv->bdev, VMW_PL_MOB);
+		vmw_gmrid_man_fini(dev_priv, VMW_PL_MOB);
 	if (dev_priv->has_gmr)
-		(void) ttm_bo_clean_mm(&dev_priv->bdev, VMW_PL_GMR);
-	(void)ttm_bo_clean_mm(&dev_priv->bdev, TTM_PL_VRAM);
+		vmw_gmrid_man_fini(dev_priv, VMW_PL_GMR);
+	vmw_vram_manager_fini(dev_priv);
 out_no_vram:
 	(void)ttm_bo_device_release(&dev_priv->bdev);
 out_no_bdev:
@@ -1012,12 +1039,12 @@ static void vmw_driver_unload(struct drm_device *dev)
 	vmw_overlay_close(dev_priv);
 
 	if (dev_priv->has_gmr)
-		(void)ttm_bo_clean_mm(&dev_priv->bdev, VMW_PL_GMR);
-	(void)ttm_bo_clean_mm(&dev_priv->bdev, TTM_PL_VRAM);
+		vmw_gmrid_man_fini(dev_priv, VMW_PL_GMR);
 
 	vmw_release_device_early(dev_priv);
 	if (dev_priv->has_mob)
-		(void) ttm_bo_clean_mm(&dev_priv->bdev, VMW_PL_MOB);
+		vmw_gmrid_man_fini(dev_priv, VMW_PL_MOB);
+	vmw_vram_manager_fini(dev_priv);
 	(void) ttm_bo_device_release(&dev_priv->bdev);
 	drm_vma_offset_manager_destroy(&dev_priv->vma_manager);
 	vmw_release_device_late(dev_priv);
@@ -1159,10 +1186,12 @@ static void vmw_master_drop(struct drm_device *dev,
  */
 static void __vmw_svga_enable(struct vmw_private *dev_priv)
 {
+	struct ttm_resource_manager *man = ttm_manager_type(&dev_priv->bdev, TTM_PL_VRAM);
+
 	spin_lock(&dev_priv->svga_lock);
-	if (!dev_priv->bdev.man[TTM_PL_VRAM].use_type) {
+	if (!ttm_resource_manager_used(man)) {
 		vmw_write(dev_priv, SVGA_REG_ENABLE, SVGA_REG_ENABLE);
-		dev_priv->bdev.man[TTM_PL_VRAM].use_type = true;
+		ttm_resource_manager_set_used(man, true);
 	}
 	spin_unlock(&dev_priv->svga_lock);
 }
@@ -1188,9 +1217,11 @@ void vmw_svga_enable(struct vmw_private *dev_priv)
  */
 static void __vmw_svga_disable(struct vmw_private *dev_priv)
 {
+	struct ttm_resource_manager *man = ttm_manager_type(&dev_priv->bdev, TTM_PL_VRAM);
+
 	spin_lock(&dev_priv->svga_lock);
-	if (dev_priv->bdev.man[TTM_PL_VRAM].use_type) {
-		dev_priv->bdev.man[TTM_PL_VRAM].use_type = false;
+	if (ttm_resource_manager_used(man)) {
+		ttm_resource_manager_set_used(man, false);
 		vmw_write(dev_priv, SVGA_REG_ENABLE,
 			  SVGA_REG_ENABLE_HIDE |
 			  SVGA_REG_ENABLE_ENABLE);
@@ -1207,6 +1238,7 @@ static void __vmw_svga_disable(struct vmw_private *dev_priv)
  */
 void vmw_svga_disable(struct vmw_private *dev_priv)
 {
+	struct ttm_resource_manager *man = ttm_manager_type(&dev_priv->bdev, TTM_PL_VRAM);
 	/*
 	 * Disabling SVGA will turn off device modesetting capabilities, so
 	 * notify KMS about that so that it doesn't cache atomic state that
@@ -1222,8 +1254,8 @@ void vmw_svga_disable(struct vmw_private *dev_priv)
 	vmw_kms_lost_device(dev_priv->dev);
 	ttm_write_lock(&dev_priv->reservation_sem, false);
 	spin_lock(&dev_priv->svga_lock);
-	if (dev_priv->bdev.man[TTM_PL_VRAM].use_type) {
-		dev_priv->bdev.man[TTM_PL_VRAM].use_type = false;
+	if (ttm_resource_manager_used(man)) {
+		ttm_resource_manager_set_used(man, false);
 		spin_unlock(&dev_priv->svga_lock);
 		if (ttm_bo_evict_mm(&dev_priv->bdev, TTM_PL_VRAM))
 			DRM_ERROR("Failed evicting VRAM buffers.\n");
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 3596f3923ea3..1523b51a7284 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -82,9 +82,7 @@
 			VMWGFX_NUM_GB_SCREEN_TARGET)
 
 #define VMW_PL_GMR (TTM_PL_PRIV + 0)
-#define VMW_PL_FLAG_GMR (TTM_PL_FLAG_PRIV << 0)
 #define VMW_PL_MOB (TTM_PL_PRIV + 1)
-#define VMW_PL_FLAG_MOB (TTM_PL_FLAG_PRIV << 1)
 
 #define VMW_RES_CONTEXT ttm_driver_type0
 #define VMW_RES_SURFACE ttm_driver_type1
@@ -793,7 +791,7 @@ extern void vmw_resource_unreserve(struct vmw_resource *res,
 				   struct vmw_buffer_object *new_backup,
 				   unsigned long new_backup_offset);
 extern void vmw_query_move_notify(struct ttm_buffer_object *bo,
-				  struct ttm_mem_reg *mem);
+				  struct ttm_resource *mem);
 extern int vmw_query_readback_all(struct vmw_buffer_object *dx_query_mob);
 extern void vmw_resource_evict_all(struct vmw_private *dev_priv);
 extern void vmw_resource_unbind_list(struct vmw_buffer_object *vbo);
@@ -878,7 +876,7 @@ extern void vmw_bo_fence_single(struct ttm_buffer_object *bo,
 extern void *vmw_bo_map_and_cache(struct vmw_buffer_object *vbo);
 extern void vmw_bo_unmap(struct vmw_buffer_object *vbo);
 extern void vmw_bo_move_notify(struct ttm_buffer_object *bo,
-			       struct ttm_mem_reg *mem);
+			       struct ttm_resource *mem);
 extern void vmw_bo_swap_notify(struct ttm_buffer_object *bo);
 extern struct vmw_buffer_object *
 vmw_user_bo_noref_lookup(struct ttm_object_file *tfile, u32 handle);
@@ -1019,10 +1017,12 @@ extern struct ttm_placement vmw_mob_placement;
 extern struct ttm_placement vmw_mob_ne_placement;
 extern struct ttm_placement vmw_nonfixed_placement;
 extern struct ttm_bo_driver vmw_bo_driver;
-extern int vmw_bo_map_dma(struct ttm_buffer_object *bo);
-extern void vmw_bo_unmap_dma(struct ttm_buffer_object *bo);
 extern const struct vmw_sg_table *
 vmw_bo_sg_table(struct ttm_buffer_object *bo);
+extern int vmw_bo_create_and_populate(struct vmw_private *dev_priv,
+				      unsigned long bo_size,
+				      struct ttm_buffer_object **bo_p);
+
 extern void vmw_piter_start(struct vmw_piter *viter,
 			    const struct vmw_sg_table *vsgt,
 			    unsigned long p_offs);
@@ -1219,7 +1219,8 @@ int vmw_overlay_num_free_overlays(struct vmw_private *dev_priv);
  * GMR Id manager
  */
 
-extern const struct ttm_mem_type_manager_func vmw_gmrid_manager_func;
+int vmw_gmrid_man_init(struct vmw_private *dev_priv, int type);
+void vmw_gmrid_man_fini(struct vmw_private *dev_priv, int type);
 
 /**
  * Prime - vmwgfx_prime.c
@@ -1518,9 +1519,8 @@ vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf,
 
 /* Transparent hugepage support - vmwgfx_thp.c */
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-extern const struct ttm_mem_type_manager_func vmw_thp_func;
-#else
-#define vmw_thp_func ttm_bo_manager_func
+extern int vmw_thp_init(struct vmw_private *dev_priv);
+void vmw_thp_fini(struct vmw_private *dev_priv);
 #endif
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
index f8bdd4ea294a..551042489036 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c
@@ -37,6 +37,7 @@
 #include <linux/kernel.h>
 
 struct vmwgfx_gmrid_man {
+	struct ttm_resource_manager manager;
 	spinlock_t lock;
 	struct ida gmr_ida;
 	uint32_t max_gmr_ids;
@@ -44,13 +45,17 @@ struct vmwgfx_gmrid_man {
 	uint32_t used_gmr_pages;
 };
 
-static int vmw_gmrid_man_get_node(struct ttm_mem_type_manager *man,
+static struct vmwgfx_gmrid_man *to_gmrid_manager(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct vmwgfx_gmrid_man, manager);
+}
+
+static int vmw_gmrid_man_get_node(struct ttm_resource_manager *man,
 				  struct ttm_buffer_object *bo,
 				  const struct ttm_place *place,
-				  struct ttm_mem_reg *mem)
+				  struct ttm_resource *mem)
 {
-	struct vmwgfx_gmrid_man *gman =
-		(struct vmwgfx_gmrid_man *)man->priv;
+	struct vmwgfx_gmrid_man *gman = to_gmrid_manager(man);
 	int id;
 
 	id = ida_alloc_max(&gman->gmr_ida, gman->max_gmr_ids - 1, GFP_KERNEL);
@@ -79,11 +84,10 @@ nospace:
 	return -ENOSPC;
 }
 
-static void vmw_gmrid_man_put_node(struct ttm_mem_type_manager *man,
-				   struct ttm_mem_reg *mem)
+static void vmw_gmrid_man_put_node(struct ttm_resource_manager *man,
+				   struct ttm_resource *mem)
 {
-	struct vmwgfx_gmrid_man *gman =
-		(struct vmwgfx_gmrid_man *)man->priv;
+	struct vmwgfx_gmrid_man *gman = to_gmrid_manager(man);
 
 	if (mem->mm_node) {
 		ida_free(&gman->gmr_ida, mem->start);
@@ -94,22 +98,28 @@ static void vmw_gmrid_man_put_node(struct ttm_mem_type_manager *man,
 	}
 }
 
-static int vmw_gmrid_man_init(struct ttm_mem_type_manager *man,
-			      unsigned long p_size)
+static const struct ttm_resource_manager_func vmw_gmrid_manager_func;
+
+int vmw_gmrid_man_init(struct vmw_private *dev_priv, int type)
 {
-	struct vmw_private *dev_priv =
-		container_of(man->bdev, struct vmw_private, bdev);
+	struct ttm_resource_manager *man;
 	struct vmwgfx_gmrid_man *gman =
 		kzalloc(sizeof(*gman), GFP_KERNEL);
 
 	if (unlikely(!gman))
 		return -ENOMEM;
 
+	man = &gman->manager;
+
+	man->func = &vmw_gmrid_manager_func;
+	/* TODO: This is most likely not correct */
+	man->use_tt = true;
+	ttm_resource_manager_init(man, 0);
 	spin_lock_init(&gman->lock);
 	gman->used_gmr_pages = 0;
 	ida_init(&gman->gmr_ida);
 
-	switch (p_size) {
+	switch (type) {
 	case VMW_PL_GMR:
 		gman->max_gmr_ids = dev_priv->max_gmr_ids;
 		gman->max_gmr_pages = dev_priv->max_gmr_pages;
@@ -121,32 +131,29 @@ static int vmw_gmrid_man_init(struct ttm_mem_type_manager *man,
 	default:
 		BUG();
 	}
-	man->priv = (void *) gman;
+	ttm_set_driver_manager(&dev_priv->bdev, type, &gman->manager);
+	ttm_resource_manager_set_used(man, true);
 	return 0;
 }
 
-static int vmw_gmrid_man_takedown(struct ttm_mem_type_manager *man)
+void vmw_gmrid_man_fini(struct vmw_private *dev_priv, int type)
 {
-	struct vmwgfx_gmrid_man *gman =
-		(struct vmwgfx_gmrid_man *)man->priv;
+	struct ttm_resource_manager *man = ttm_manager_type(&dev_priv->bdev, type);
+	struct vmwgfx_gmrid_man *gman = to_gmrid_manager(man);
 
-	if (gman) {
-		ida_destroy(&gman->gmr_ida);
-		kfree(gman);
-	}
-	return 0;
-}
+	ttm_resource_manager_set_used(man, false);
+
+	ttm_resource_manager_force_list_clean(&dev_priv->bdev, man);
+
+	ttm_resource_manager_cleanup(man);
+
+	ttm_set_driver_manager(&dev_priv->bdev, type, NULL);
+	ida_destroy(&gman->gmr_ida);
+	kfree(gman);
 
-static void vmw_gmrid_man_debug(struct ttm_mem_type_manager *man,
-				struct drm_printer *printer)
-{
-	drm_printf(printer, "No debug info available for the GMR id manager\n");
 }
 
-const struct ttm_mem_type_manager_func vmw_gmrid_manager_func = {
-	.init = vmw_gmrid_man_init,
-	.takedown = vmw_gmrid_man_takedown,
-	.get_node = vmw_gmrid_man_get_node,
-	.put_node = vmw_gmrid_man_put_node,
-	.debug = vmw_gmrid_man_debug
+static const struct ttm_resource_manager_func vmw_gmrid_manager_func = {
+	.alloc = vmw_gmrid_man_get_node,
+	.free = vmw_gmrid_man_put_node,
 };
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
index e8eb42933ca2..7f95ed6aa224 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
@@ -238,10 +238,6 @@ static int vmw_otable_batch_setup(struct vmw_private *dev_priv,
 	unsigned long offset;
 	unsigned long bo_size;
 	struct vmw_otable *otables = batch->otables;
-	struct ttm_operation_ctx ctx = {
-		.interruptible = false,
-		.no_wait_gpu = false
-	};
 	SVGAOTableType i;
 	int ret;
 
@@ -255,24 +251,9 @@ static int vmw_otable_batch_setup(struct vmw_private *dev_priv,
 		bo_size += otables[i].size;
 	}
 
-	ret = ttm_bo_create(&dev_priv->bdev, bo_size,
-			    ttm_bo_type_device,
-			    &vmw_sys_ne_placement,
-			    0, false, &batch->otable_bo);
-
-	if (unlikely(ret != 0))
-		goto out_no_bo;
-
-	ret = ttm_bo_reserve(batch->otable_bo, false, true, NULL);
-	BUG_ON(ret != 0);
-	ret = vmw_bo_driver.ttm_tt_populate(batch->otable_bo->ttm, &ctx);
-	if (unlikely(ret != 0))
-		goto out_unreserve;
-	ret = vmw_bo_map_dma(batch->otable_bo);
+	ret = vmw_bo_create_and_populate(dev_priv, bo_size, &batch->otable_bo);
 	if (unlikely(ret != 0))
-		goto out_unreserve;
-
-	ttm_bo_unreserve(batch->otable_bo);
+		return ret;
 
 	offset = 0;
 	for (i = 0; i < batch->num_otables; ++i) {
@@ -289,8 +270,6 @@ static int vmw_otable_batch_setup(struct vmw_private *dev_priv,
 
 	return 0;
 
-out_unreserve:
-	ttm_bo_unreserve(batch->otable_bo);
 out_no_setup:
 	for (i = 0; i < batch->num_otables; ++i) {
 		if (batch->otables[i].enabled)
@@ -300,7 +279,6 @@ out_no_setup:
 
 	ttm_bo_put(batch->otable_bo);
 	batch->otable_bo = NULL;
-out_no_bo:
 	return ret;
 }
 
@@ -432,41 +410,9 @@ struct vmw_mob *vmw_mob_create(unsigned long data_pages)
 static int vmw_mob_pt_populate(struct vmw_private *dev_priv,
 			       struct vmw_mob *mob)
 {
-	int ret;
-	struct ttm_operation_ctx ctx = {
-		.interruptible = false,
-		.no_wait_gpu = false
-	};
-
 	BUG_ON(mob->pt_bo != NULL);
 
-	ret = ttm_bo_create(&dev_priv->bdev, mob->num_pages * PAGE_SIZE,
-			    ttm_bo_type_device,
-			    &vmw_sys_ne_placement,
-			    0, false, &mob->pt_bo);
-	if (unlikely(ret != 0))
-		return ret;
-
-	ret = ttm_bo_reserve(mob->pt_bo, false, true, NULL);
-
-	BUG_ON(ret != 0);
-	ret = vmw_bo_driver.ttm_tt_populate(mob->pt_bo->ttm, &ctx);
-	if (unlikely(ret != 0))
-		goto out_unreserve;
-	ret = vmw_bo_map_dma(mob->pt_bo);
-	if (unlikely(ret != 0))
-		goto out_unreserve;
-
-	ttm_bo_unreserve(mob->pt_bo);
-
-	return 0;
-
-out_unreserve:
-	ttm_bo_unreserve(mob->pt_bo);
-	ttm_bo_put(mob->pt_bo);
-	mob->pt_bo = NULL;
-
-	return ret;
+	return vmw_bo_create_and_populate(dev_priv, mob->num_pages * PAGE_SIZE, &mob->pt_bo);
 }
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index c8441030637a..c0f156078dda 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -855,7 +855,7 @@ int vmw_query_readback_all(struct vmw_buffer_object *dx_query_mob)
  * states from the device.
  */
 void vmw_query_move_notify(struct ttm_buffer_object *bo,
-			   struct ttm_mem_reg *mem)
+			   struct ttm_resource *mem)
 {
 	struct vmw_buffer_object *dx_query_mob;
 	struct ttm_bo_device *bdev = bo->bdev;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c
index c8b9335bccd8..c8427998fa35 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c
@@ -16,14 +16,22 @@
  * @lock: Manager lock.
  */
 struct vmw_thp_manager {
+	struct ttm_resource_manager manager;
 	struct drm_mm mm;
 	spinlock_t lock;
 };
 
+static struct vmw_thp_manager *to_thp_manager(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct vmw_thp_manager, manager);
+}
+
+static const struct ttm_resource_manager_func vmw_thp_func;
+
 static int vmw_thp_insert_aligned(struct drm_mm *mm, struct drm_mm_node *node,
 				  unsigned long align_pages,
 				  const struct ttm_place *place,
-				  struct ttm_mem_reg *mem,
+				  struct ttm_resource *mem,
 				  unsigned long lpfn,
 				  enum drm_mm_insert_mode mode)
 {
@@ -38,12 +46,12 @@ static int vmw_thp_insert_aligned(struct drm_mm *mm, struct drm_mm_node *node,
 	return -ENOSPC;
 }
 
-static int vmw_thp_get_node(struct ttm_mem_type_manager *man,
+static int vmw_thp_get_node(struct ttm_resource_manager *man,
 			    struct ttm_buffer_object *bo,
 			    const struct ttm_place *place,
-			    struct ttm_mem_reg *mem)
+			    struct ttm_resource *mem)
 {
-	struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv;
+	struct vmw_thp_manager *rman = to_thp_manager(man);
 	struct drm_mm *mm = &rman->mm;
 	struct drm_mm_node *node;
 	unsigned long align_pages;
@@ -100,10 +108,10 @@ found_unlock:
 
 
 
-static void vmw_thp_put_node(struct ttm_mem_type_manager *man,
-			     struct ttm_mem_reg *mem)
+static void vmw_thp_put_node(struct ttm_resource_manager *man,
+			     struct ttm_resource *mem)
 {
-	struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv;
+	struct vmw_thp_manager *rman = to_thp_manager(man);
 
 	if (mem->mm_node) {
 		spin_lock(&rman->lock);
@@ -115,8 +123,7 @@ static void vmw_thp_put_node(struct ttm_mem_type_manager *man,
 	}
 }
 
-static int vmw_thp_init(struct ttm_mem_type_manager *man,
-			unsigned long p_size)
+int vmw_thp_init(struct vmw_private *dev_priv)
 {
 	struct vmw_thp_manager *rman;
 
@@ -124,43 +131,51 @@ static int vmw_thp_init(struct ttm_mem_type_manager *man,
 	if (!rman)
 		return -ENOMEM;
 
-	drm_mm_init(&rman->mm, 0, p_size);
+	ttm_resource_manager_init(&rman->manager,
+				  dev_priv->vram_size >> PAGE_SHIFT);
+
+	rman->manager.func = &vmw_thp_func;
+	drm_mm_init(&rman->mm, 0, rman->manager.size);
 	spin_lock_init(&rman->lock);
-	man->priv = rman;
+
+	ttm_set_driver_manager(&dev_priv->bdev, TTM_PL_VRAM, &rman->manager);
+	ttm_resource_manager_set_used(&rman->manager, true);
 	return 0;
 }
 
-static int vmw_thp_takedown(struct ttm_mem_type_manager *man)
+void vmw_thp_fini(struct vmw_private *dev_priv)
 {
-	struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv;
+	struct ttm_resource_manager *man = ttm_manager_type(&dev_priv->bdev, TTM_PL_VRAM);
+	struct vmw_thp_manager *rman = to_thp_manager(man);
 	struct drm_mm *mm = &rman->mm;
+	int ret;
+
+	ttm_resource_manager_set_used(man, false);
 
+	ret = ttm_resource_manager_force_list_clean(&dev_priv->bdev, man);
+	if (ret)
+		return;
 	spin_lock(&rman->lock);
-	if (drm_mm_clean(mm)) {
-		drm_mm_takedown(mm);
-		spin_unlock(&rman->lock);
-		kfree(rman);
-		man->priv = NULL;
-		return 0;
-	}
+	drm_mm_clean(mm);
+	drm_mm_takedown(mm);
 	spin_unlock(&rman->lock);
-	return -EBUSY;
+	ttm_resource_manager_cleanup(man);
+	ttm_set_driver_manager(&dev_priv->bdev, TTM_PL_VRAM, NULL);
+	kfree(rman);
 }
 
-static void vmw_thp_debug(struct ttm_mem_type_manager *man,
+static void vmw_thp_debug(struct ttm_resource_manager *man,
 			  struct drm_printer *printer)
 {
-	struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv;
+	struct vmw_thp_manager *rman = to_thp_manager(man);
 
 	spin_lock(&rman->lock);
 	drm_mm_print(&rman->mm, printer);
 	spin_unlock(&rman->lock);
 }
 
-const struct ttm_mem_type_manager_func vmw_thp_func = {
-	.init = vmw_thp_init,
-	.takedown = vmw_thp_takedown,
-	.get_node = vmw_thp_get_node,
-	.put_node = vmw_thp_put_node,
+static const struct ttm_resource_manager_func vmw_thp_func = {
+	.alloc = vmw_thp_get_node,
+	.free = vmw_thp_put_node,
 	.debug = vmw_thp_debug
 };
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
index ab524ab3b0b4..7f0310441da1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
@@ -33,49 +33,57 @@
 static const struct ttm_place vram_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
-	.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED
+	.mem_type = TTM_PL_VRAM,
+	.flags = TTM_PL_FLAG_CACHED
 };
 
 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
+	.mem_type = TTM_PL_VRAM,
+	.flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
 static const struct ttm_place sys_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
-	.flags = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED
+	.mem_type = TTM_PL_SYSTEM,
+	.flags = TTM_PL_FLAG_CACHED
 };
 
 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
+	.mem_type = TTM_PL_SYSTEM,
+	.flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
 static const struct ttm_place gmr_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
-	.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
+	.mem_type = VMW_PL_GMR,
+	.flags = TTM_PL_FLAG_CACHED
 };
 
 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
+	.mem_type = VMW_PL_GMR,
+	.flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
 static const struct ttm_place mob_placement_flags = {
 	.fpfn = 0,
 	.lpfn = 0,
-	.flags = VMW_PL_FLAG_MOB | TTM_PL_FLAG_CACHED
+	.mem_type = VMW_PL_MOB,
+	.flags = TTM_PL_FLAG_CACHED
 };
 
 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
+	.mem_type = VMW_PL_MOB,
+	.flags = TTM_PL_FLAG_CACHED | TTM_PL_FLAG_NO_EVICT
 };
 
 struct ttm_placement vmw_vram_placement = {
@@ -89,11 +97,13 @@ static const struct ttm_place vram_gmr_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED
+		.mem_type = TTM_PL_VRAM,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_GMR,
+		.flags = TTM_PL_FLAG_CACHED
 	}
 };
 
@@ -101,11 +111,13 @@ static const struct ttm_place gmr_vram_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_GMR,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED
+		.mem_type = TTM_PL_VRAM,
+		.flags = TTM_PL_FLAG_CACHED
 	}
 };
 
@@ -120,12 +132,14 @@ static const struct ttm_place vram_gmr_ne_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED |
+		.mem_type = TTM_PL_VRAM,
+		.flags = TTM_PL_FLAG_CACHED |
 			 TTM_PL_FLAG_NO_EVICT
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED |
+		.mem_type = VMW_PL_GMR,
+		.flags = TTM_PL_FLAG_CACHED |
 			 TTM_PL_FLAG_NO_EVICT
 	}
 };
@@ -169,19 +183,23 @@ static const struct ttm_place evictable_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_VRAM | TTM_PL_FLAG_CACHED
+		.mem_type = TTM_PL_VRAM,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_GMR,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_MOB | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_MOB,
+		.flags = TTM_PL_FLAG_CACHED
 	}
 };
 
@@ -189,15 +207,18 @@ static const struct ttm_place nonfixed_placement_flags[] = {
 	{
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_GMR | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_GMR,
+		.flags = TTM_PL_FLAG_CACHED
 	}, {
 		.fpfn = 0,
 		.lpfn = 0,
-		.flags = VMW_PL_FLAG_MOB | TTM_PL_FLAG_CACHED
+		.mem_type = VMW_PL_MOB,
+		.flags = TTM_PL_FLAG_CACHED
 	}
 };
 
@@ -246,6 +267,7 @@ struct vmw_ttm_tt {
 	struct vmw_sg_table vsgt;
 	uint64_t sg_alloc_size;
 	bool mapped;
+	bool bound;
 };
 
 const size_t vmw_tt_size = sizeof(struct vmw_ttm_tt);
@@ -362,8 +384,7 @@ static void vmw_ttm_unmap_from_dma(struct vmw_ttm_tt *vmw_tt)
 {
 	struct device *dev = vmw_tt->dev_priv->dev->dev;
 
-	dma_unmap_sg(dev, vmw_tt->sgt.sgl, vmw_tt->sgt.nents,
-		DMA_BIDIRECTIONAL);
+	dma_unmap_sgtable(dev, &vmw_tt->sgt, DMA_BIDIRECTIONAL, 0);
 	vmw_tt->sgt.nents = vmw_tt->sgt.orig_nents;
 }
 
@@ -383,16 +404,8 @@ static void vmw_ttm_unmap_from_dma(struct vmw_ttm_tt *vmw_tt)
 static int vmw_ttm_map_for_dma(struct vmw_ttm_tt *vmw_tt)
 {
 	struct device *dev = vmw_tt->dev_priv->dev->dev;
-	int ret;
-
-	ret = dma_map_sg(dev, vmw_tt->sgt.sgl, vmw_tt->sgt.orig_nents,
-			 DMA_BIDIRECTIONAL);
-	if (unlikely(ret == 0))
-		return -ENOMEM;
-
-	vmw_tt->sgt.nents = ret;
 
-	return 0;
+	return dma_map_sgtable(dev, &vmw_tt->sgt, DMA_BIDIRECTIONAL, 0);
 }
 
 /**
@@ -449,10 +462,10 @@ static int vmw_ttm_map_dma(struct vmw_ttm_tt *vmw_tt)
 		if (unlikely(ret != 0))
 			goto out_sg_alloc_fail;
 
-		if (vsgt->num_pages > vmw_tt->sgt.nents) {
+		if (vsgt->num_pages > vmw_tt->sgt.orig_nents) {
 			uint64_t over_alloc =
 				sgl_size * (vsgt->num_pages -
-					    vmw_tt->sgt.nents);
+					    vmw_tt->sgt.orig_nents);
 
 			ttm_mem_global_free(glob, over_alloc);
 			vmw_tt->sg_alloc_size -= over_alloc;
@@ -519,43 +532,6 @@ static void vmw_ttm_unmap_dma(struct vmw_ttm_tt *vmw_tt)
 	vmw_tt->mapped = false;
 }
 
-
-/**
- * vmw_bo_map_dma - Make sure buffer object pages are visible to the device
- *
- * @bo: Pointer to a struct ttm_buffer_object
- *
- * Wrapper around vmw_ttm_map_dma, that takes a TTM buffer object pointer
- * instead of a pointer to a struct vmw_ttm_backend as argument.
- * Note that the buffer object must be either pinned or reserved before
- * calling this function.
- */
-int vmw_bo_map_dma(struct ttm_buffer_object *bo)
-{
-	struct vmw_ttm_tt *vmw_tt =
-		container_of(bo->ttm, struct vmw_ttm_tt, dma_ttm.ttm);
-
-	return vmw_ttm_map_dma(vmw_tt);
-}
-
-
-/**
- * vmw_bo_unmap_dma - Make sure buffer object pages are visible to the device
- *
- * @bo: Pointer to a struct ttm_buffer_object
- *
- * Wrapper around vmw_ttm_unmap_dma, that takes a TTM buffer object pointer
- * instead of a pointer to a struct vmw_ttm_backend as argument.
- */
-void vmw_bo_unmap_dma(struct ttm_buffer_object *bo)
-{
-	struct vmw_ttm_tt *vmw_tt =
-		container_of(bo->ttm, struct vmw_ttm_tt, dma_ttm.ttm);
-
-	vmw_ttm_unmap_dma(vmw_tt);
-}
-
-
 /**
  * vmw_bo_sg_table - Return a struct vmw_sg_table object for a
  * TTM buffer object
@@ -576,11 +552,18 @@ const struct vmw_sg_table *vmw_bo_sg_table(struct ttm_buffer_object *bo)
 }
 
 
-static int vmw_ttm_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
+static int vmw_ttm_bind(struct ttm_bo_device *bdev,
+			struct ttm_tt *ttm, struct ttm_resource *bo_mem)
 {
 	struct vmw_ttm_tt *vmw_be =
 		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
-	int ret;
+	int ret = 0;
+
+	if (!bo_mem)
+		return -EINVAL;
+
+	if (vmw_be->bound)
+		return 0;
 
 	ret = vmw_ttm_map_dma(vmw_be);
 	if (unlikely(ret != 0))
@@ -591,8 +574,9 @@ static int vmw_ttm_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
 
 	switch (bo_mem->mem_type) {
 	case VMW_PL_GMR:
-		return vmw_gmr_bind(vmw_be->dev_priv, &vmw_be->vsgt,
+		ret = vmw_gmr_bind(vmw_be->dev_priv, &vmw_be->vsgt,
 				    ttm->num_pages, vmw_be->gmr_id);
+		break;
 	case VMW_PL_MOB:
 		if (unlikely(vmw_be->mob == NULL)) {
 			vmw_be->mob =
@@ -601,20 +585,26 @@ static int vmw_ttm_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
 				return -ENOMEM;
 		}
 
-		return vmw_mob_bind(vmw_be->dev_priv, vmw_be->mob,
+		ret = vmw_mob_bind(vmw_be->dev_priv, vmw_be->mob,
 				    &vmw_be->vsgt, ttm->num_pages,
 				    vmw_be->gmr_id);
+		break;
 	default:
 		BUG();
 	}
-	return 0;
+	vmw_be->bound = true;
+	return ret;
 }
 
-static void vmw_ttm_unbind(struct ttm_tt *ttm)
+static void vmw_ttm_unbind(struct ttm_bo_device *bdev,
+			   struct ttm_tt *ttm)
 {
 	struct vmw_ttm_tt *vmw_be =
 		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
 
+	if (!vmw_be->bound)
+		return;
+
 	switch (vmw_be->mem_type) {
 	case VMW_PL_GMR:
 		vmw_gmr_unbind(vmw_be->dev_priv, vmw_be->gmr_id);
@@ -628,14 +618,17 @@ static void vmw_ttm_unbind(struct ttm_tt *ttm)
 
 	if (vmw_be->dev_priv->map_mode == vmw_dma_map_bind)
 		vmw_ttm_unmap_dma(vmw_be);
+	vmw_be->bound = false;
 }
 
 
-static void vmw_ttm_destroy(struct ttm_tt *ttm)
+static void vmw_ttm_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm)
 {
 	struct vmw_ttm_tt *vmw_be =
 		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
 
+	vmw_ttm_unbind(bdev, ttm);
+	ttm_tt_destroy_common(bdev, ttm);
 	vmw_ttm_unmap_dma(vmw_be);
 	if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent)
 		ttm_dma_tt_fini(&vmw_be->dma_ttm);
@@ -649,7 +642,8 @@ static void vmw_ttm_destroy(struct ttm_tt *ttm)
 }
 
 
-static int vmw_ttm_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
+static int vmw_ttm_populate(struct ttm_bo_device *bdev,
+			    struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 {
 	struct vmw_ttm_tt *vmw_tt =
 		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
@@ -657,7 +651,7 @@ static int vmw_ttm_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 	struct ttm_mem_global *glob = vmw_mem_glob(dev_priv);
 	int ret;
 
-	if (ttm->state != tt_unpopulated)
+	if (ttm_tt_is_populated(ttm))
 		return 0;
 
 	if (dev_priv->map_mode == vmw_dma_alloc_coherent) {
@@ -677,7 +671,8 @@ static int vmw_ttm_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx)
 	return ret;
 }
 
-static void vmw_ttm_unpopulate(struct ttm_tt *ttm)
+static void vmw_ttm_unpopulate(struct ttm_bo_device *bdev,
+			       struct ttm_tt *ttm)
 {
 	struct vmw_ttm_tt *vmw_tt = container_of(ttm, struct vmw_ttm_tt,
 						 dma_ttm.ttm);
@@ -701,12 +696,6 @@ static void vmw_ttm_unpopulate(struct ttm_tt *ttm)
 		ttm_pool_unpopulate(ttm);
 }
 
-static struct ttm_backend_func vmw_ttm_func = {
-	.bind = vmw_ttm_bind,
-	.unbind = vmw_ttm_unbind,
-	.destroy = vmw_ttm_destroy,
-};
-
 static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo,
 					uint32_t page_flags)
 {
@@ -717,7 +706,6 @@ static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo,
 	if (!vmw_be)
 		return NULL;
 
-	vmw_be->dma_ttm.ttm.func = &vmw_ttm_func;
 	vmw_be->dev_priv = container_of(bo->bdev, struct vmw_private, bdev);
 	vmw_be->mob = NULL;
 
@@ -734,40 +722,6 @@ out_no_init:
 	return NULL;
 }
 
-static int vmw_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
-		      struct ttm_mem_type_manager *man)
-{
-	switch (type) {
-	case TTM_PL_SYSTEM:
-		/* System memory */
-		man->available_caching = TTM_PL_FLAG_CACHED;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case TTM_PL_VRAM:
-		/* "On-card" video ram */
-		man->func = &vmw_thp_func;
-		man->flags = TTM_MEMTYPE_FLAG_FIXED;
-		man->available_caching = TTM_PL_FLAG_CACHED;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	case VMW_PL_GMR:
-	case VMW_PL_MOB:
-		/*
-		 * "Guest Memory Regions" is an aperture like feature with
-		 *  one slot per bo. There is an upper limit of the number of
-		 *  slots as well as the bo size.
-		 */
-		man->func = &vmw_gmrid_manager_func;
-		man->available_caching = TTM_PL_FLAG_CACHED;
-		man->default_caching = TTM_PL_FLAG_CACHED;
-		break;
-	default:
-		DRM_ERROR("Unsupported memory type %u\n", (unsigned)type);
-		return -EINVAL;
-	}
-	return 0;
-}
-
 static void vmw_evict_flags(struct ttm_buffer_object *bo,
 		     struct ttm_placement *placement)
 {
@@ -782,24 +736,18 @@ static int vmw_verify_access(struct ttm_buffer_object *bo, struct file *filp)
 	return vmw_user_bo_verify_access(bo, tfile);
 }
 
-static int vmw_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
+static int vmw_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_resource *mem)
 {
 	struct vmw_private *dev_priv = container_of(bdev, struct vmw_private, bdev);
 
-	mem->bus.addr = NULL;
-	mem->bus.is_iomem = false;
-	mem->bus.offset = 0;
-	mem->bus.size = mem->num_pages << PAGE_SHIFT;
-	mem->bus.base = 0;
-
 	switch (mem->mem_type) {
 	case TTM_PL_SYSTEM:
 	case VMW_PL_GMR:
 	case VMW_PL_MOB:
 		return 0;
 	case TTM_PL_VRAM:
-		mem->bus.offset = mem->start << PAGE_SHIFT;
-		mem->bus.base = dev_priv->vram_start;
+		mem->bus.offset = (mem->start << PAGE_SHIFT) +
+			dev_priv->vram_start;
 		mem->bus.is_iomem = true;
 		break;
 	default:
@@ -812,7 +760,7 @@ static int vmw_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg
  * vmw_move_notify - TTM move_notify_callback
  *
  * @bo: The TTM buffer object about to move.
- * @mem: The struct ttm_mem_reg indicating to what memory
+ * @mem: The struct ttm_resource indicating to what memory
  *       region the move is taking place.
  *
  * Calls move_notify for all subsystems needing it.
@@ -820,7 +768,7 @@ static int vmw_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg
  */
 static void vmw_move_notify(struct ttm_buffer_object *bo,
 			    bool evict,
-			    struct ttm_mem_reg *mem)
+			    struct ttm_resource *mem)
 {
 	vmw_bo_move_notify(bo, mem);
 	vmw_query_move_notify(bo, mem);
@@ -843,7 +791,9 @@ struct ttm_bo_driver vmw_bo_driver = {
 	.ttm_tt_create = &vmw_ttm_tt_create,
 	.ttm_tt_populate = &vmw_ttm_populate,
 	.ttm_tt_unpopulate = &vmw_ttm_unpopulate,
-	.init_mem_type = vmw_init_mem_type,
+	.ttm_tt_bind = &vmw_ttm_bind,
+	.ttm_tt_unbind = &vmw_ttm_unbind,
+	.ttm_tt_destroy = &vmw_ttm_destroy,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = vmw_evict_flags,
 	.move = NULL,
@@ -852,3 +802,38 @@ struct ttm_bo_driver vmw_bo_driver = {
 	.swap_notify = vmw_swap_notify,
 	.io_mem_reserve = &vmw_ttm_io_mem_reserve,
 };
+
+int vmw_bo_create_and_populate(struct vmw_private *dev_priv,
+			       unsigned long bo_size,
+			       struct ttm_buffer_object **bo_p)
+{
+	struct ttm_operation_ctx ctx = {
+		.interruptible = false,
+		.no_wait_gpu = false
+	};
+	struct ttm_buffer_object *bo;
+	int ret;
+
+	ret = ttm_bo_create(&dev_priv->bdev, bo_size,
+			    ttm_bo_type_device,
+			    &vmw_sys_ne_placement,
+			    0, false, &bo);
+
+	if (unlikely(ret != 0))
+		return ret;
+
+	ret = ttm_bo_reserve(bo, false, true, NULL);
+	BUG_ON(ret != 0);
+	ret = vmw_ttm_populate(bo->bdev, bo->ttm, &ctx);
+	if (likely(ret == 0)) {
+		struct vmw_ttm_tt *vmw_tt =
+			container_of(bo->ttm, struct vmw_ttm_tt, dma_ttm.ttm);
+		ret = vmw_ttm_map_dma(vmw_tt);
+	}
+
+	ttm_bo_unreserve(bo);
+
+	if (likely(ret == 0))
+		*bo_p = bo;
+	return ret;
+}
diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c
index 534daf37c97e..2f464ef2d53e 100644
--- a/drivers/gpu/drm/xen/xen_drm_front_gem.c
+++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c
@@ -180,7 +180,8 @@ struct sg_table *xen_drm_front_gem_get_sg_table(struct drm_gem_object *gem_obj)
 	if (!xen_obj->pages)
 		return ERR_PTR(-ENOMEM);
 
-	return drm_prime_pages_to_sg(xen_obj->pages, xen_obj->num_pages);
+	return drm_prime_pages_to_sg(gem_obj->dev,
+				     xen_obj->pages, xen_obj->num_pages);
 }
 
 struct drm_gem_object *
@@ -217,7 +218,7 @@ xen_drm_front_gem_import_sg_table(struct drm_device *dev,
 		return ERR_PTR(ret);
 
 	DRM_DEBUG("Imported buffer of size %zu with nents %u\n",
-		  size, sgt->nents);
+		  size, sgt->orig_nents);
 
 	return &xen_obj->base;
 }
diff --git a/drivers/gpu/drm/xlnx/zynqmp_disp.c b/drivers/gpu/drm/xlnx/zynqmp_disp.c
index a455cfc1bee5..98bd48f13fd1 100644
--- a/drivers/gpu/drm/xlnx/zynqmp_disp.c
+++ b/drivers/gpu/drm/xlnx/zynqmp_disp.c
@@ -242,12 +242,6 @@ static const u32 scaling_factors_565[] = {
 	ZYNQMP_DISP_AV_BUF_5BIT_SF,
 };
 
-static const u32 scaling_factors_666[] = {
-	ZYNQMP_DISP_AV_BUF_6BIT_SF,
-	ZYNQMP_DISP_AV_BUF_6BIT_SF,
-	ZYNQMP_DISP_AV_BUF_6BIT_SF,
-};
-
 static const u32 scaling_factors_888[] = {
 	ZYNQMP_DISP_AV_BUF_8BIT_SF,
 	ZYNQMP_DISP_AV_BUF_8BIT_SF,
diff --git a/drivers/gpu/drm/xlnx/zynqmp_dpsub.c b/drivers/gpu/drm/xlnx/zynqmp_dpsub.c
index 26328c76305b..8e69303aad3f 100644
--- a/drivers/gpu/drm/xlnx/zynqmp_dpsub.c
+++ b/drivers/gpu/drm/xlnx/zynqmp_dpsub.c
@@ -111,7 +111,7 @@ static int zynqmp_dpsub_drm_init(struct zynqmp_dpsub *dpsub)
 	/* Initialize mode config, vblank and the KMS poll helper. */
 	ret = drmm_mode_config_init(drm);
 	if (ret < 0)
-		goto err_dev_put;
+		return ret;
 
 	drm->mode_config.funcs = &zynqmp_dpsub_mode_config_funcs;
 	drm->mode_config.min_width = 0;
@@ -121,7 +121,7 @@ static int zynqmp_dpsub_drm_init(struct zynqmp_dpsub *dpsub)
 
 	ret = drm_vblank_init(drm, 1);
 	if (ret)
-		goto err_dev_put;
+		return ret;
 
 	drm->irq_enabled = 1;
 
@@ -154,8 +154,6 @@ static int zynqmp_dpsub_drm_init(struct zynqmp_dpsub *dpsub)
 
 err_poll_fini:
 	drm_kms_helper_poll_fini(drm);
-err_dev_put:
-	drm_dev_put(drm);
 	return ret;
 }
 
@@ -208,27 +206,16 @@ static int zynqmp_dpsub_probe(struct platform_device *pdev)
 	int ret;
 
 	/* Allocate private data. */
-	dpsub = kzalloc(sizeof(*dpsub), GFP_KERNEL);
-	if (!dpsub)
-		return -ENOMEM;
+	dpsub = devm_drm_dev_alloc(&pdev->dev, &zynqmp_dpsub_drm_driver,
+				   struct zynqmp_dpsub, drm);
+	if (IS_ERR(dpsub))
+		return PTR_ERR(dpsub);
 
 	dpsub->dev = &pdev->dev;
 	platform_set_drvdata(pdev, dpsub);
 
 	dma_set_mask(dpsub->dev, DMA_BIT_MASK(ZYNQMP_DISP_MAX_DMA_BIT));
 
-	/*
-	 * Initialize the DRM device early, as the DRM core mandates usage of
-	 * the managed memory helpers tied to the DRM device.
-	 */
-	ret = drm_dev_init(&dpsub->drm, &zynqmp_dpsub_drm_driver, &pdev->dev);
-	if (ret < 0) {
-		kfree(dpsub);
-		return ret;
-	}
-
-	drmm_add_final_kfree(&dpsub->drm, dpsub);
-
 	/* Try the reserved memory. Proceed if there's none. */
 	of_reserved_mem_device_init(&pdev->dev);
 
@@ -286,8 +273,6 @@ static int zynqmp_dpsub_remove(struct platform_device *pdev)
 	clk_disable_unprepare(dpsub->apb_clk);
 	of_reserved_mem_device_release(&pdev->dev);
 
-	drm_dev_put(drm);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index 89b6c14b7392..82d0a60ba3f7 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -170,11 +170,9 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 				goto unpin;
 			}
 
-			err = dma_map_sg(dev, sgt->sgl, sgt->nents, dir);
-			if (!err) {
-				err = -ENOMEM;
+			err = dma_map_sgtable(dev, sgt, dir, 0);
+			if (err)
 				goto unpin;
-			}
 
 			job->unpins[job->num_unpins].dev = dev;
 			job->unpins[job->num_unpins].dir = dir;
@@ -228,7 +226,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		}
 
 		if (host->domain) {
-			for_each_sg(sgt->sgl, sg, sgt->nents, j)
+			for_each_sgtable_sg(sgt, sg, j)
 				gather_size += sg->length;
 			gather_size = iova_align(&host->iova, gather_size);
 
@@ -240,9 +238,9 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 				goto put;
 			}
 
-			err = iommu_map_sg(host->domain,
+			err = iommu_map_sgtable(host->domain,
 					iova_dma_addr(&host->iova, alloc),
-					sgt->sgl, sgt->nents, IOMMU_READ);
+					sgt, IOMMU_READ);
 			if (err == 0) {
 				__free_iova(&host->iova, alloc);
 				err = -EINVAL;
@@ -252,12 +250,9 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 			job->unpins[job->num_unpins].size = gather_size;
 			phys_addr = iova_dma_addr(&host->iova, alloc);
 		} else if (sgt) {
-			err = dma_map_sg(host->dev, sgt->sgl, sgt->nents,
-					 DMA_TO_DEVICE);
-			if (!err) {
-				err = -ENOMEM;
+			err = dma_map_sgtable(host->dev, sgt, DMA_TO_DEVICE, 0);
+			if (err)
 				goto put;
-			}
 
 			job->unpins[job->num_unpins].dir = DMA_TO_DEVICE;
 			job->unpins[job->num_unpins].dev = host->dev;
@@ -660,8 +655,7 @@ void host1x_job_unpin(struct host1x_job *job)
 		}
 
 		if (unpin->dev && sgt)
-			dma_unmap_sg(unpin->dev, sgt->sgl, sgt->nents,
-				     unpin->dir);
+			dma_unmap_sgtable(unpin->dev, sgt, unpin->dir, 0);
 
 		host1x_bo_unpin(dev, unpin->bo, sgt);
 		host1x_bo_put(unpin->bo);