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

Doing a backmerge to be able to merge topic/mei-hdcp-2019-02-19 PR.

Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

779 files changed, 17829 insertions, 23868 deletions

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index ce8d1d384319..1ac55c65eac0 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -51,7 +51,7 @@ obj-$(CONFIG_DRM_DEBUG_SELFTEST) += selftests/
 obj-$(CONFIG_DRM)	+= drm.o
 obj-$(CONFIG_DRM_MIPI_DSI) += drm_mipi_dsi.o
 obj-$(CONFIG_DRM_PANEL_ORIENTATION_QUIRKS) += drm_panel_orientation_quirks.o
-obj-$(CONFIG_DRM_ARM)	+= arm/
+obj-y			+= arm/
 obj-$(CONFIG_DRM_TTM)	+= ttm/
 obj-$(CONFIG_DRM_SCHED)	+= scheduler/
 obj-$(CONFIG_DRM_TDFX)	+= tdfx/
@@ -81,7 +81,7 @@ obj-$(CONFIG_DRM_UDL) += udl/
 obj-$(CONFIG_DRM_AST) += ast/
 obj-$(CONFIG_DRM_ARMADA) += armada/
 obj-$(CONFIG_DRM_ATMEL_HLCDC)	+= atmel-hlcdc/
-obj-$(CONFIG_DRM_RCAR_DU) += rcar-du/
+obj-y			+= rcar-du/
 obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/
 obj-y			+= omapdrm/
 obj-$(CONFIG_DRM_SUN4I) += sun4i/
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index f76bcb9c45e4..466da5954a68 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -57,7 +57,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 
 # add asic specific block
 amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \
-	ci_smc.o ci_dpm.o dce_v8_0.o gfx_v7_0.o cik_sdma.o uvd_v4_2.o vce_v2_0.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
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index bcef6ea4bcf9..9efa681d0878 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -542,6 +542,11 @@ struct amdgpu_asic_funcs {
 	bool (*need_full_reset)(struct amdgpu_device *adev);
 	/* initialize doorbell layout for specific asic*/
 	void (*init_doorbell_index)(struct amdgpu_device *adev);
+	/* PCIe bandwidth usage */
+	void (*get_pcie_usage)(struct amdgpu_device *adev, uint64_t *count0,
+			       uint64_t *count1);
+	/* do we need to reset the asic at init time (e.g., kexec) */
+	bool (*need_reset_on_init)(struct amdgpu_device *adev);
 };
 
 /*
@@ -634,7 +639,7 @@ struct amdgpu_nbio_funcs {
 	void (*hdp_flush)(struct amdgpu_device *adev, struct amdgpu_ring *ring);
 	u32 (*get_memsize)(struct amdgpu_device *adev);
 	void (*sdma_doorbell_range)(struct amdgpu_device *adev, int instance,
-				    bool use_doorbell, int doorbell_index);
+			bool use_doorbell, int doorbell_index, int doorbell_size);
 	void (*enable_doorbell_aperture)(struct amdgpu_device *adev,
 					 bool enable);
 	void (*enable_doorbell_selfring_aperture)(struct amdgpu_device *adev,
@@ -1042,6 +1047,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define amdgpu_asic_invalidate_hdp(adev, r) (adev)->asic_funcs->invalidate_hdp((adev), (r))
 #define amdgpu_asic_need_full_reset(adev) (adev)->asic_funcs->need_full_reset((adev))
 #define amdgpu_asic_init_doorbell_index(adev) (adev)->asic_funcs->init_doorbell_index((adev))
+#define amdgpu_asic_get_pcie_usage(adev, cnt0, cnt1) ((adev)->asic_funcs->get_pcie_usage((adev), (cnt0), (cnt1)))
+#define amdgpu_asic_need_reset_on_init(adev) (adev)->asic_funcs->need_reset_on_init((adev))
 
 /* Common functions */
 bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 2dfaf158ef07..e957e42c539a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -28,8 +28,6 @@
 #include <linux/module.h>
 #include <linux/dma-buf.h>
 
-const struct kgd2kfd_calls *kgd2kfd;
-
 static const unsigned int compute_vmid_bitmap = 0xFF00;
 
 /* Total memory size in system memory and all GPU VRAM. Used to
@@ -47,12 +45,9 @@ int amdgpu_amdkfd_init(void)
 	amdgpu_amdkfd_total_mem_size *= si.mem_unit;
 
 #ifdef CONFIG_HSA_AMD
-	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
-	if (ret)
-		kgd2kfd = NULL;
+	ret = kgd2kfd_init();
 	amdgpu_amdkfd_gpuvm_init_mem_limits();
 #else
-	kgd2kfd = NULL;
 	ret = -ENOENT;
 #endif
 
@@ -61,17 +56,13 @@ int amdgpu_amdkfd_init(void)
 
 void amdgpu_amdkfd_fini(void)
 {
-	if (kgd2kfd)
-		kgd2kfd->exit();
+	kgd2kfd_exit();
 }
 
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 {
 	const struct kfd2kgd_calls *kfd2kgd;
 
-	if (!kgd2kfd)
-		return;
-
 	switch (adev->asic_type) {
 #ifdef CONFIG_DRM_AMDGPU_CIK
 	case CHIP_KAVERI:
@@ -98,8 +89,8 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 		return;
 	}
 
-	adev->kfd.dev = kgd2kfd->probe((struct kgd_dev *)adev,
-				       adev->pdev, kfd2kgd);
+	adev->kfd.dev = kgd2kfd_probe((struct kgd_dev *)adev,
+				      adev->pdev, kfd2kgd);
 
 	if (adev->kfd.dev)
 		amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size;
@@ -182,7 +173,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 				&gpu_resources.doorbell_start_offset);
 
 		if (adev->asic_type < CHIP_VEGA10) {
-			kgd2kfd->device_init(adev->kfd.dev, &gpu_resources);
+			kgd2kfd_device_init(adev->kfd.dev, &gpu_resources);
 			return;
 		}
 
@@ -197,13 +188,13 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 			 * can use each doorbell assignment twice.
 			 */
 			gpu_resources.sdma_doorbell[0][i] =
-				adev->doorbell_index.sdma_engine0 + (i >> 1);
+				adev->doorbell_index.sdma_engine[0] + (i >> 1);
 			gpu_resources.sdma_doorbell[0][i+1] =
-				adev->doorbell_index.sdma_engine0 + 0x200 + (i >> 1);
+				adev->doorbell_index.sdma_engine[0] + 0x200 + (i >> 1);
 			gpu_resources.sdma_doorbell[1][i] =
-				adev->doorbell_index.sdma_engine1 + (i >> 1);
+				adev->doorbell_index.sdma_engine[1] + (i >> 1);
 			gpu_resources.sdma_doorbell[1][i+1] =
-				adev->doorbell_index.sdma_engine1 + 0x200 + (i >> 1);
+				adev->doorbell_index.sdma_engine[1] + 0x200 + (i >> 1);
 		}
 		/* Doorbells 0x0e0-0ff and 0x2e0-2ff are reserved for
 		 * SDMA, IH and VCN. So don't use them for the CP.
@@ -211,14 +202,14 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 		gpu_resources.reserved_doorbell_mask = 0x1e0;
 		gpu_resources.reserved_doorbell_val  = 0x0e0;
 
-		kgd2kfd->device_init(adev->kfd.dev, &gpu_resources);
+		kgd2kfd_device_init(adev->kfd.dev, &gpu_resources);
 	}
 }
 
 void amdgpu_amdkfd_device_fini(struct amdgpu_device *adev)
 {
 	if (adev->kfd.dev) {
-		kgd2kfd->device_exit(adev->kfd.dev);
+		kgd2kfd_device_exit(adev->kfd.dev);
 		adev->kfd.dev = NULL;
 	}
 }
@@ -227,13 +218,13 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 		const void *ih_ring_entry)
 {
 	if (adev->kfd.dev)
-		kgd2kfd->interrupt(adev->kfd.dev, ih_ring_entry);
+		kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry);
 }
 
 void amdgpu_amdkfd_suspend(struct amdgpu_device *adev)
 {
 	if (adev->kfd.dev)
-		kgd2kfd->suspend(adev->kfd.dev);
+		kgd2kfd_suspend(adev->kfd.dev);
 }
 
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
@@ -241,7 +232,7 @@ int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
 	int r = 0;
 
 	if (adev->kfd.dev)
-		r = kgd2kfd->resume(adev->kfd.dev);
+		r = kgd2kfd_resume(adev->kfd.dev);
 
 	return r;
 }
@@ -251,7 +242,7 @@ int amdgpu_amdkfd_pre_reset(struct amdgpu_device *adev)
 	int r = 0;
 
 	if (adev->kfd.dev)
-		r = kgd2kfd->pre_reset(adev->kfd.dev);
+		r = kgd2kfd_pre_reset(adev->kfd.dev);
 
 	return r;
 }
@@ -261,7 +252,7 @@ int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev)
 	int r = 0;
 
 	if (adev->kfd.dev)
-		r = kgd2kfd->post_reset(adev->kfd.dev);
+		r = kgd2kfd_post_reset(adev->kfd.dev);
 
 	return r;
 }
@@ -619,4 +610,47 @@ struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void)
 {
 	return NULL;
 }
+
+struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd, struct pci_dev *pdev,
+			      const struct kfd2kgd_calls *f2g)
+{
+	return NULL;
+}
+
+bool kgd2kfd_device_init(struct kfd_dev *kfd,
+			 const struct kgd2kfd_shared_resources *gpu_resources)
+{
+	return false;
+}
+
+void kgd2kfd_device_exit(struct kfd_dev *kfd)
+{
+}
+
+void kgd2kfd_exit(void)
+{
+}
+
+void kgd2kfd_suspend(struct kfd_dev *kfd)
+{
+}
+
+int kgd2kfd_resume(struct kfd_dev *kfd)
+{
+	return 0;
+}
+
+int kgd2kfd_pre_reset(struct kfd_dev *kfd)
+{
+	return 0;
+}
+
+int kgd2kfd_post_reset(struct kfd_dev *kfd)
+{
+	return 0;
+}
+
+void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
+{
+}
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 70429f7aa9a8..0b31a1859023 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -33,7 +33,6 @@
 #include "amdgpu_sync.h"
 #include "amdgpu_vm.h"
 
-extern const struct kgd2kfd_calls *kgd2kfd;
 extern uint64_t amdgpu_amdkfd_total_mem_size;
 
 struct amdgpu_device;
@@ -214,4 +213,22 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd,
 void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
 void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo);
 
+/* KGD2KFD callbacks */
+int kgd2kfd_init(void);
+void kgd2kfd_exit(void);
+struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd, struct pci_dev *pdev,
+			      const struct kfd2kgd_calls *f2g);
+bool kgd2kfd_device_init(struct kfd_dev *kfd,
+			 const struct kgd2kfd_shared_resources *gpu_resources);
+void kgd2kfd_device_exit(struct kfd_dev *kfd);
+void kgd2kfd_suspend(struct kfd_dev *kfd);
+int kgd2kfd_resume(struct kfd_dev *kfd);
+int kgd2kfd_pre_reset(struct kfd_dev *kfd);
+int kgd2kfd_post_reset(struct kfd_dev *kfd);
+void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
+int kgd2kfd_quiesce_mm(struct mm_struct *mm);
+int kgd2kfd_resume_mm(struct mm_struct *mm);
+int kgd2kfd_schedule_evict_and_restore_process(struct mm_struct *mm,
+					       struct dma_fence *fence);
+
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
index 574c1181ae9a..3107b9575929 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
@@ -122,7 +122,7 @@ static bool amdkfd_fence_enable_signaling(struct dma_fence *f)
 	if (dma_fence_is_signaled(f))
 		return true;
 
-	if (!kgd2kfd->schedule_evict_and_restore_process(fence->mm, f))
+	if (!kgd2kfd_schedule_evict_and_restore_process(fence->mm, f))
 		return true;
 
 	return false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index be1ab43473c6..d7b10d79f1de 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1790,7 +1790,7 @@ int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem,
 	evicted_bos = atomic_inc_return(&process_info->evicted_bos);
 	if (evicted_bos == 1) {
 		/* First eviction, stop the queues */
-		r = kgd2kfd->quiesce_mm(mm);
+		r = kgd2kfd_quiesce_mm(mm);
 		if (r)
 			pr_err("Failed to quiesce KFD\n");
 		schedule_delayed_work(&process_info->restore_userptr_work,
@@ -2082,7 +2082,7 @@ static void amdgpu_amdkfd_restore_userptr_worker(struct work_struct *work)
 	    evicted_bos)
 		goto unlock_out;
 	evicted_bos = 0;
-	if (kgd2kfd->resume_mm(mm)) {
+	if (kgd2kfd_resume_mm(mm)) {
 		pr_err("%s: Failed to resume KFD\n", __func__);
 		/* No recovery from this failure. Probably the CP is
 		 * hanging. No point trying again.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index a028661d9e20..92b11de19581 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -576,6 +576,7 @@ static const struct amdgpu_px_quirk amdgpu_px_quirk_list[] = {
 	{ 0x1002, 0x6900, 0x1028, 0x0812, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0x1002, 0x6900, 0x1028, 0x0813, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0x1002, 0x6900, 0x1025, 0x125A, AMDGPU_PX_QUIRK_FORCE_ATPX },
+	{ 0x1002, 0x6900, 0x17AA, 0x3806, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0, 0, 0, 0, 0 },
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 69ad6ec0a4f3..bf04c12bd324 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -25,8 +25,8 @@
  */
 #include <drm/drmP.h>
 #include <drm/drm_edid.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "atom.h"
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 1c49b8266d69..52a5e4fdc95b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -214,6 +214,7 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs
 		case AMDGPU_CHUNK_ID_DEPENDENCIES:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
+		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
 			break;
 
 		default:
@@ -1090,6 +1091,15 @@ static int amdgpu_cs_process_fence_dep(struct amdgpu_cs_parser *p,
 
 		fence = amdgpu_ctx_get_fence(ctx, entity,
 					     deps[i].handle);
+
+		if (chunk->chunk_id == AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES) {
+			struct drm_sched_fence *s_fence = to_drm_sched_fence(fence);
+			struct dma_fence *old = fence;
+
+			fence = dma_fence_get(&s_fence->scheduled);
+			dma_fence_put(old);
+		}
+
 		if (IS_ERR(fence)) {
 			r = PTR_ERR(fence);
 			amdgpu_ctx_put(ctx);
@@ -1177,7 +1187,8 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 
 		chunk = &p->chunks[i];
 
-		if (chunk->chunk_id == AMDGPU_CHUNK_ID_DEPENDENCIES) {
+		if (chunk->chunk_id == AMDGPU_CHUNK_ID_DEPENDENCIES ||
+		    chunk->chunk_id == AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES) {
 			r = amdgpu_cs_process_fence_dep(p, chunk);
 			if (r)
 				return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 28bccceaa363..4f8fb4ecde34 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -30,8 +30,8 @@
 #include <linux/console.h>
 #include <linux/slab.h>
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include <linux/vgaarb.h>
 #include <linux/vga_switcheroo.h>
@@ -1645,7 +1645,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 		if (r) {
 			DRM_ERROR("sw_init of IP block <%s> failed %d\n",
 				  adev->ip_blocks[i].version->funcs->name, r);
-			return r;
+			goto init_failed;
 		}
 		adev->ip_blocks[i].status.sw = true;
 
@@ -1654,17 +1654,17 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			r = amdgpu_device_vram_scratch_init(adev);
 			if (r) {
 				DRM_ERROR("amdgpu_vram_scratch_init failed %d\n", r);
-				return r;
+				goto init_failed;
 			}
 			r = adev->ip_blocks[i].version->funcs->hw_init((void *)adev);
 			if (r) {
 				DRM_ERROR("hw_init %d failed %d\n", i, r);
-				return r;
+				goto init_failed;
 			}
 			r = amdgpu_device_wb_init(adev);
 			if (r) {
 				DRM_ERROR("amdgpu_device_wb_init failed %d\n", r);
-				return r;
+				goto init_failed;
 			}
 			adev->ip_blocks[i].status.hw = true;
 
@@ -1675,7 +1675,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 								AMDGPU_CSA_SIZE);
 				if (r) {
 					DRM_ERROR("allocate CSA failed %d\n", r);
-					return r;
+					goto init_failed;
 				}
 			}
 		}
@@ -1683,28 +1683,32 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 
 	r = amdgpu_ucode_create_bo(adev); /* create ucode bo when sw_init complete*/
 	if (r)
-		return r;
+		goto init_failed;
 
 	r = amdgpu_device_ip_hw_init_phase1(adev);
 	if (r)
-		return r;
+		goto init_failed;
 
 	r = amdgpu_device_fw_loading(adev);
 	if (r)
-		return r;
+		goto init_failed;
 
 	r = amdgpu_device_ip_hw_init_phase2(adev);
 	if (r)
-		return r;
+		goto init_failed;
 
 	if (adev->gmc.xgmi.num_physical_nodes > 1)
 		amdgpu_xgmi_add_device(adev);
 	amdgpu_amdkfd_device_init(adev);
 
-	if (amdgpu_sriov_vf(adev))
+init_failed:
+	if (amdgpu_sriov_vf(adev)) {
+		if (!r)
+			amdgpu_virt_init_data_exchange(adev);
 		amdgpu_virt_release_full_gpu(adev, true);
+	}
 
-	return 0;
+	return r;
 }
 
 /**
@@ -2131,7 +2135,7 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
 				continue;
 
 			r = block->version->funcs->hw_init(adev);
-			DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
+			DRM_INFO("RE-INIT-early: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
 			if (r)
 				return r;
 		}
@@ -2165,7 +2169,7 @@ static int amdgpu_device_ip_reinit_late_sriov(struct amdgpu_device *adev)
 				continue;
 
 			r = block->version->funcs->hw_init(adev);
-			DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
+			DRM_INFO("RE-INIT-late: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
 			if (r)
 				return r;
 		}
@@ -2546,6 +2550,17 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	/* detect if we are with an SRIOV vbios */
 	amdgpu_device_detect_sriov_bios(adev);
 
+	/* check if we need to reset the asic
+	 *  E.g., driver was not cleanly unloaded previously, etc.
+	 */
+	if (!amdgpu_sriov_vf(adev) && amdgpu_asic_need_reset_on_init(adev)) {
+		r = amdgpu_asic_reset(adev);
+		if (r) {
+			dev_err(adev->dev, "asic reset on init failed\n");
+			goto failed;
+		}
+	}
+
 	/* Post card if necessary */
 	if (amdgpu_device_need_post(adev)) {
 		if (!adev->bios) {
@@ -2610,6 +2625,8 @@ fence_driver_init:
 		}
 		dev_err(adev->dev, "amdgpu_device_ip_init failed\n");
 		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_AMDGPU_INIT_FAIL, 0, 0);
+		if (amdgpu_virt_request_full_gpu(adev, false))
+			amdgpu_virt_release_full_gpu(adev, false);
 		goto failed;
 	}
 
@@ -2632,9 +2649,6 @@ fence_driver_init:
 		goto failed;
 	}
 
-	if (amdgpu_sriov_vf(adev))
-		amdgpu_virt_init_data_exchange(adev);
-
 	amdgpu_fbdev_init(adev);
 
 	r = amdgpu_pm_sysfs_init(adev);
@@ -2798,7 +2812,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon)
 			struct drm_framebuffer *fb = crtc->primary->fb;
 			struct amdgpu_bo *robj;
 
-			if (amdgpu_crtc->cursor_bo) {
+			if (amdgpu_crtc->cursor_bo && !adev->enable_virtual_display) {
 				struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
 				r = amdgpu_bo_reserve(aobj, true);
 				if (r == 0) {
@@ -2906,7 +2920,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool resume, bool fbcon)
 		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 			struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 
-			if (amdgpu_crtc->cursor_bo) {
+			if (amdgpu_crtc->cursor_bo && !adev->enable_virtual_display) {
 				struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
 				r = amdgpu_bo_reserve(aobj, true);
 				if (r == 0) {
@@ -3226,6 +3240,7 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
 	r = amdgpu_ib_ring_tests(adev);
 
 error:
+	amdgpu_virt_init_data_exchange(adev);
 	amdgpu_virt_release_full_gpu(adev, true);
 	if (!r && adev->virt.gim_feature & AMDGIM_FEATURE_GIM_FLR_VRAMLOST) {
 		atomic_inc(&adev->vram_lost_counter);
@@ -3298,17 +3313,15 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 		if (!ring || !ring->sched.thread)
 			continue;
 
-		kthread_park(ring->sched.thread);
-
-		if (job && job->base.sched != &ring->sched)
-			continue;
-
-		drm_sched_hw_job_reset(&ring->sched, job ? &job->base : NULL);
+		drm_sched_stop(&ring->sched);
 
 		/* after all hw jobs are reset, hw fence is meaningless, so force_completion */
 		amdgpu_fence_driver_force_completion(ring);
 	}
 
+	if(job)
+		drm_sched_increase_karma(&job->base);
+
 
 
 	if (!amdgpu_sriov_vf(adev)) {
@@ -3454,14 +3467,10 @@ static void amdgpu_device_post_asic_reset(struct amdgpu_device *adev,
 		if (!ring || !ring->sched.thread)
 			continue;
 
-		/* only need recovery sched of the given job's ring
-		 * or all rings (in the case @job is NULL)
-		 * after above amdgpu_reset accomplished
-		 */
-		if ((!job || job->base.sched == &ring->sched) && !adev->asic_reset_res)
-			drm_sched_job_recovery(&ring->sched);
+		if (!adev->asic_reset_res)
+			drm_sched_resubmit_jobs(&ring->sched);
 
-		kthread_unpark(ring->sched.thread);
+		drm_sched_start(&ring->sched, !adev->asic_reset_res);
 	}
 
 	if (!amdgpu_device_has_dc_support(adev)) {
@@ -3521,9 +3530,9 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	 * by different nodes. No point also since the one node already executing
 	 * reset will also reset all the other nodes in the hive.
 	 */
-	hive = amdgpu_get_xgmi_hive(adev);
+	hive = amdgpu_get_xgmi_hive(adev, 0);
 	if (hive && adev->gmc.xgmi.num_physical_nodes > 1 &&
-	    !mutex_trylock(&hive->hive_lock))
+	    !mutex_trylock(&hive->reset_lock))
 		return 0;
 
 	/* Start with adev pre asic reset first for soft reset check.*/
@@ -3602,13 +3611,45 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 	}
 
 	if (hive && adev->gmc.xgmi.num_physical_nodes > 1)
-		mutex_unlock(&hive->hive_lock);
+		mutex_unlock(&hive->reset_lock);
 
 	if (r)
 		dev_info(adev->dev, "GPU reset end with ret = %d\n", r);
 	return r;
 }
 
+static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev,
+						  enum pci_bus_speed *speed,
+						  enum pcie_link_width *width)
+{
+	struct pci_dev *pdev = adev->pdev;
+	enum pci_bus_speed cur_speed;
+	enum pcie_link_width cur_width;
+
+	*speed = PCI_SPEED_UNKNOWN;
+	*width = PCIE_LNK_WIDTH_UNKNOWN;
+
+	while (pdev) {
+		cur_speed = pcie_get_speed_cap(pdev);
+		cur_width = pcie_get_width_cap(pdev);
+
+		if (cur_speed != PCI_SPEED_UNKNOWN) {
+			if (*speed == PCI_SPEED_UNKNOWN)
+				*speed = cur_speed;
+			else if (cur_speed < *speed)
+				*speed = cur_speed;
+		}
+
+		if (cur_width != PCIE_LNK_WIDTH_UNKNOWN) {
+			if (*width == PCIE_LNK_WIDTH_UNKNOWN)
+				*width = cur_width;
+			else if (cur_width < *width)
+				*width = cur_width;
+		}
+		pdev = pci_upstream_bridge(pdev);
+	}
+}
+
 /**
  * amdgpu_device_get_pcie_info - fence pcie info about the PCIE slot
  *
@@ -3621,8 +3662,8 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 {
 	struct pci_dev *pdev;
-	enum pci_bus_speed speed_cap;
-	enum pcie_link_width link_width;
+	enum pci_bus_speed speed_cap, platform_speed_cap;
+	enum pcie_link_width platform_link_width;
 
 	if (amdgpu_pcie_gen_cap)
 		adev->pm.pcie_gen_mask = amdgpu_pcie_gen_cap;
@@ -3639,6 +3680,12 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 		return;
 	}
 
+	if (adev->pm.pcie_gen_mask && adev->pm.pcie_mlw_mask)
+		return;
+
+	amdgpu_device_get_min_pci_speed_width(adev, &platform_speed_cap,
+					      &platform_link_width);
+
 	if (adev->pm.pcie_gen_mask == 0) {
 		/* asic caps */
 		pdev = adev->pdev;
@@ -3664,22 +3711,20 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 				adev->pm.pcie_gen_mask |= CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1;
 		}
 		/* platform caps */
-		pdev = adev->ddev->pdev->bus->self;
-		speed_cap = pcie_get_speed_cap(pdev);
-		if (speed_cap == PCI_SPEED_UNKNOWN) {
+		if (platform_speed_cap == PCI_SPEED_UNKNOWN) {
 			adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
 						   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2);
 		} else {
-			if (speed_cap == PCIE_SPEED_16_0GT)
+			if (platform_speed_cap == PCIE_SPEED_16_0GT)
 				adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4);
-			else if (speed_cap == PCIE_SPEED_8_0GT)
+			else if (platform_speed_cap == PCIE_SPEED_8_0GT)
 				adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3);
-			else if (speed_cap == PCIE_SPEED_5_0GT)
+			else if (platform_speed_cap == PCIE_SPEED_5_0GT)
 				adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
 							   CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2);
 			else
@@ -3688,12 +3733,10 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 		}
 	}
 	if (adev->pm.pcie_mlw_mask == 0) {
-		pdev = adev->ddev->pdev->bus->self;
-		link_width = pcie_get_width_cap(pdev);
-		if (link_width == PCIE_LNK_WIDTH_UNKNOWN) {
+		if (platform_link_width == PCIE_LNK_WIDTH_UNKNOWN) {
 			adev->pm.pcie_mlw_mask |= AMDGPU_DEFAULT_PCIE_MLW_MASK;
 		} else {
-			switch (link_width) {
+			switch (platform_link_width) {
 			case PCIE_LNK_X32:
 				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X32 |
 							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index 15ce7e681d67..b083b219b1a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -188,10 +188,12 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc,
 		goto cleanup;
 	}
 
-	r = amdgpu_bo_pin(new_abo, amdgpu_display_supported_domains(adev));
-	if (unlikely(r != 0)) {
-		DRM_ERROR("failed to pin new abo buffer before flip\n");
-		goto unreserve;
+	if (!adev->enable_virtual_display) {
+		r = amdgpu_bo_pin(new_abo, amdgpu_display_supported_domains(adev));
+		if (unlikely(r != 0)) {
+			DRM_ERROR("failed to pin new abo buffer before flip\n");
+			goto unreserve;
+		}
 	}
 
 	r = amdgpu_ttm_alloc_gart(&new_abo->tbo);
@@ -211,7 +213,8 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc,
 	amdgpu_bo_get_tiling_flags(new_abo, &tiling_flags);
 	amdgpu_bo_unreserve(new_abo);
 
-	work->base = amdgpu_bo_gpu_offset(new_abo);
+	if (!adev->enable_virtual_display)
+		work->base = amdgpu_bo_gpu_offset(new_abo);
 	work->target_vblank = target - (uint32_t)drm_crtc_vblank_count(crtc) +
 		amdgpu_get_vblank_counter_kms(dev, work->crtc_id);
 
@@ -242,9 +245,10 @@ pflip_cleanup:
 		goto cleanup;
 	}
 unpin:
-	if (unlikely(amdgpu_bo_unpin(new_abo) != 0)) {
-		DRM_ERROR("failed to unpin new abo in error path\n");
-	}
+	if (!adev->enable_virtual_display)
+		if (unlikely(amdgpu_bo_unpin(new_abo) != 0))
+			DRM_ERROR("failed to unpin new abo in error path\n");
+
 unreserve:
 	amdgpu_bo_unreserve(new_abo);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index be620b29f4aa..1cfec06f81d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -51,14 +51,7 @@ struct amdgpu_doorbell_index {
 	uint32_t userqueue_start;
 	uint32_t userqueue_end;
 	uint32_t gfx_ring0;
-	uint32_t sdma_engine0;
-	uint32_t sdma_engine1;
-	uint32_t sdma_engine2;
-	uint32_t sdma_engine3;
-	uint32_t sdma_engine4;
-	uint32_t sdma_engine5;
-	uint32_t sdma_engine6;
-	uint32_t sdma_engine7;
+	uint32_t sdma_engine[8];
 	uint32_t ih;
 	union {
 		struct {
@@ -79,6 +72,8 @@ struct amdgpu_doorbell_index {
 		} uvd_vce;
 	};
 	uint32_t max_assignment;
+	/* Per engine SDMA doorbell size in dword */
+	uint32_t sdma_doorbell_range;
 };
 
 typedef enum _AMDGPU_DOORBELL_ASSIGNMENT
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
index f972cd156795..2f61e9edb1c1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
@@ -364,6 +364,14 @@ enum amdgpu_pcie_gen {
 		((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)))
+
+#define amdgpu_dpm_set_ppfeature_status(adev, ppfeatures) \
+		((adev)->powerplay.pp_funcs->set_ppfeature_status(\
+			(adev)->powerplay.pp_handle, (ppfeatures)))
+
 struct amdgpu_dpm {
 	struct amdgpu_ps        *ps;
 	/* number of valid power states */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index c806f984bcc5..7f3aa7b7e1d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -32,7 +32,7 @@
 #include <linux/module.h>
 #include <linux/pm_runtime.h>
 #include <linux/vga_switcheroo.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "amdgpu.h"
 #include "amdgpu_irq.h"
@@ -71,9 +71,11 @@
  * - 3.25.0 - Add support for sensor query info (stable pstate sclk/mclk).
  * - 3.26.0 - GFX9: Process AMDGPU_IB_FLAG_TC_WB_NOT_INVALIDATE.
  * - 3.27.0 - Add new chunk to to AMDGPU_CS to enable BO_LIST creation.
+ * - 3.28.0 - Add AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES
+ * - 3.29.0 - Add AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	27
+#define KMS_DRIVER_MINOR	29
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;
@@ -1189,7 +1191,7 @@ amdgpu_get_crtc_scanout_position(struct drm_device *dev, unsigned int pipe,
 static struct drm_driver kms_driver = {
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_ATOMIC |
-	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
+	    DRIVER_GEM |
 	    DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ,
 	.load = amdgpu_driver_load_kms,
 	.open = amdgpu_driver_open_kms,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
index ecbcefe49a98..f89f5734d985 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
@@ -37,6 +37,8 @@ struct amdgpu_gds {
 	struct amdgpu_gds_asic_info	mem;
 	struct amdgpu_gds_asic_info	gws;
 	struct amdgpu_gds_asic_info	oa;
+	uint32_t			gds_compute_max_wave_id;
+
 	/* At present, GDS, GWS and OA resources for gfx (graphics)
 	 * is always pre-allocated and available for graphics operation.
 	 * Such resource is shared between all gfx clients.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index f4f00217546e..d21dd2f369da 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -54,10 +54,6 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 
 	memset(&bp, 0, sizeof(bp));
 	*obj = NULL;
-	/* At least align on page size */
-	if (alignment < PAGE_SIZE) {
-		alignment = PAGE_SIZE;
-	}
 
 	bp.size = size;
 	bp.byte_align = alignment;
@@ -244,9 +240,6 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 			return -EINVAL;
 		}
 		flags |= AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
-		/* GDS allocations must be DW aligned */
-		if (args->in.domains & AMDGPU_GEM_DOMAIN_GDS)
-			size = ALIGN(size, 4);
 	}
 
 	if (flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index c48207b377bc..0b8ef2d27d6b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -202,12 +202,12 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 			amdgpu_asic_flush_hdp(adev, ring);
 	}
 
+	if (need_ctx_switch)
+		status |= AMDGPU_HAVE_CTX_SWITCH;
+
 	skip_preamble = ring->current_ctx == fence_ctx;
 	if (job && ring->funcs->emit_cntxcntl) {
-		if (need_ctx_switch)
-			status |= AMDGPU_HAVE_CTX_SWITCH;
 		status |= job->preamble_status;
-
 		amdgpu_ring_emit_cntxcntl(ring, status);
 	}
 
@@ -221,8 +221,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 			!amdgpu_sriov_vf(adev)) /* for SRIOV preemption, Preamble CE ib must be inserted anyway */
 			continue;
 
-		amdgpu_ring_emit_ib(ring, job, ib, need_ctx_switch);
-		need_ctx_switch = false;
+		amdgpu_ring_emit_ib(ring, job, ib, status);
+		status &= ~AMDGPU_HAVE_CTX_SWITCH;
 	}
 
 	if (ring->funcs->emit_tmz)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index 8af67f649660..d0a5db777b6d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -52,6 +52,8 @@ int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 	ih->use_bus_addr = use_bus_addr;
 
 	if (use_bus_addr) {
+		dma_addr_t dma_addr;
+
 		if (ih->ring)
 			return 0;
 
@@ -59,21 +61,26 @@ int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 		 * add them to the end of the ring allocation.
 		 */
 		ih->ring = dma_alloc_coherent(adev->dev, ih->ring_size + 8,
-					      &ih->rb_dma_addr, GFP_KERNEL);
+					      &dma_addr, GFP_KERNEL);
 		if (ih->ring == NULL)
 			return -ENOMEM;
 
 		memset((void *)ih->ring, 0, ih->ring_size + 8);
-		ih->wptr_offs = (ih->ring_size / 4) + 0;
-		ih->rptr_offs = (ih->ring_size / 4) + 1;
+		ih->gpu_addr = dma_addr;
+		ih->wptr_addr = dma_addr + ih->ring_size;
+		ih->wptr_cpu = &ih->ring[ih->ring_size / 4];
+		ih->rptr_addr = dma_addr + ih->ring_size + 4;
+		ih->rptr_cpu = &ih->ring[(ih->ring_size / 4) + 1];
 	} else {
-		r = amdgpu_device_wb_get(adev, &ih->wptr_offs);
+		unsigned wptr_offs, rptr_offs;
+
+		r = amdgpu_device_wb_get(adev, &wptr_offs);
 		if (r)
 			return r;
 
-		r = amdgpu_device_wb_get(adev, &ih->rptr_offs);
+		r = amdgpu_device_wb_get(adev, &rptr_offs);
 		if (r) {
-			amdgpu_device_wb_free(adev, ih->wptr_offs);
+			amdgpu_device_wb_free(adev, wptr_offs);
 			return r;
 		}
 
@@ -82,10 +89,15 @@ int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 					    &ih->ring_obj, &ih->gpu_addr,
 					    (void **)&ih->ring);
 		if (r) {
-			amdgpu_device_wb_free(adev, ih->rptr_offs);
-			amdgpu_device_wb_free(adev, ih->wptr_offs);
+			amdgpu_device_wb_free(adev, rptr_offs);
+			amdgpu_device_wb_free(adev, wptr_offs);
 			return r;
 		}
+
+		ih->wptr_addr = adev->wb.gpu_addr + wptr_offs * 4;
+		ih->wptr_cpu = &adev->wb.wb[wptr_offs];
+		ih->rptr_addr = adev->wb.gpu_addr + rptr_offs * 4;
+		ih->rptr_cpu = &adev->wb.wb[rptr_offs];
 	}
 	return 0;
 }
@@ -109,13 +121,13 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih)
 		 * add them to the end of the ring allocation.
 		 */
 		dma_free_coherent(adev->dev, ih->ring_size + 8,
-				  (void *)ih->ring, ih->rb_dma_addr);
+				  (void *)ih->ring, ih->gpu_addr);
 		ih->ring = NULL;
 	} else {
 		amdgpu_bo_free_kernel(&ih->ring_obj, &ih->gpu_addr,
 				      (void **)&ih->ring);
-		amdgpu_device_wb_free(adev, ih->wptr_offs);
-		amdgpu_device_wb_free(adev, ih->rptr_offs);
+		amdgpu_device_wb_free(adev, (ih->wptr_addr - ih->gpu_addr) / 4);
+		amdgpu_device_wb_free(adev, (ih->rptr_addr - ih->gpu_addr) / 4);
 	}
 }
 
@@ -137,7 +149,7 @@ int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 	if (!ih->enabled || adev->shutdown)
 		return IRQ_NONE;
 
-	wptr = amdgpu_ih_get_wptr(adev);
+	wptr = amdgpu_ih_get_wptr(adev, ih);
 
 restart_ih:
 	/* is somebody else already processing irqs? */
@@ -154,11 +166,11 @@ restart_ih:
 		ih->rptr &= ih->ptr_mask;
 	}
 
-	amdgpu_ih_set_rptr(adev);
+	amdgpu_ih_set_rptr(adev, ih);
 	atomic_set(&ih->lock, 0);
 
 	/* make sure wptr hasn't changed while processing */
-	wptr = amdgpu_ih_get_wptr(adev);
+	wptr = amdgpu_ih_get_wptr(adev, ih);
 	if (wptr != ih->rptr)
 		goto restart_ih;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
index f877bb78d10a..1ccb1831382a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
@@ -31,34 +31,40 @@ struct amdgpu_iv_entry;
  * R6xx+ IH ring
  */
 struct amdgpu_ih_ring {
-	struct amdgpu_bo	*ring_obj;
-	volatile uint32_t	*ring;
-	unsigned		rptr;
 	unsigned		ring_size;
-	uint64_t		gpu_addr;
 	uint32_t		ptr_mask;
-	atomic_t		lock;
-	bool                    enabled;
-	unsigned		wptr_offs;
-	unsigned		rptr_offs;
 	u32			doorbell_index;
 	bool			use_doorbell;
 	bool			use_bus_addr;
-	dma_addr_t		rb_dma_addr; /* only used when use_bus_addr = true */
+
+	struct amdgpu_bo	*ring_obj;
+	volatile uint32_t	*ring;
+	uint64_t		gpu_addr;
+
+	uint64_t		wptr_addr;
+	volatile uint32_t	*wptr_cpu;
+
+	uint64_t		rptr_addr;
+	volatile uint32_t	*rptr_cpu;
+
+	bool                    enabled;
+	unsigned		rptr;
+	atomic_t		lock;
 };
 
 /* provided by the ih block */
 struct amdgpu_ih_funcs {
 	/* ring read/write ptr handling, called from interrupt context */
-	u32 (*get_wptr)(struct amdgpu_device *adev);
-	void (*decode_iv)(struct amdgpu_device *adev,
+	u32 (*get_wptr)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih);
+	void (*decode_iv)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 			  struct amdgpu_iv_entry *entry);
-	void (*set_rptr)(struct amdgpu_device *adev);
+	void (*set_rptr)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih);
 };
 
-#define amdgpu_ih_get_wptr(adev) (adev)->irq.ih_funcs->get_wptr((adev))
-#define amdgpu_ih_decode_iv(adev, iv) (adev)->irq.ih_funcs->decode_iv((adev), (iv))
-#define amdgpu_ih_set_rptr(adev) (adev)->irq.ih_funcs->set_rptr((adev))
+#define amdgpu_ih_get_wptr(adev, ih) (adev)->irq.ih_funcs->get_wptr((adev), (ih))
+#define amdgpu_ih_decode_iv(adev, iv) \
+	(adev)->irq.ih_funcs->decode_iv((adev), (ih), (iv))
+#define amdgpu_ih_set_rptr(adev, ih) (adev)->irq.ih_funcs->set_rptr((adev), (ih))
 
 int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 			unsigned ring_size, bool use_bus_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index b7968f426862..8bfb3dab46f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -148,6 +148,8 @@ static void amdgpu_irq_callback(struct amdgpu_device *adev,
 	entry.iv_entry = (const uint32_t *)&ih->ring[ring_index];
 	amdgpu_ih_decode_iv(adev, &entry);
 
+	trace_amdgpu_iv(ih - &adev->irq.ih, &entry);
+
 	amdgpu_irq_dispatch(adev, &entry);
 }
 
@@ -175,6 +177,36 @@ irqreturn_t amdgpu_irq_handler(int irq, void *arg)
 }
 
 /**
+ * amdgpu_irq_handle_ih1 - kick of processing for IH1
+ *
+ * @work: work structure in struct amdgpu_irq
+ *
+ * Kick of processing IH ring 1.
+ */
+static void amdgpu_irq_handle_ih1(struct work_struct *work)
+{
+	struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
+						  irq.ih1_work);
+
+	amdgpu_ih_process(adev, &adev->irq.ih1, amdgpu_irq_callback);
+}
+
+/**
+ * amdgpu_irq_handle_ih2 - kick of processing for IH2
+ *
+ * @work: work structure in struct amdgpu_irq
+ *
+ * Kick of processing IH ring 2.
+ */
+static void amdgpu_irq_handle_ih2(struct work_struct *work)
+{
+	struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
+						  irq.ih2_work);
+
+	amdgpu_ih_process(adev, &adev->irq.ih2, amdgpu_irq_callback);
+}
+
+/**
  * amdgpu_msi_ok - check whether MSI functionality is enabled
  *
  * @adev: amdgpu device pointer (unused)
@@ -238,6 +270,9 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 				amdgpu_hotplug_work_func);
 	}
 
+	INIT_WORK(&adev->irq.ih1_work, amdgpu_irq_handle_ih1);
+	INIT_WORK(&adev->irq.ih2_work, amdgpu_irq_handle_ih2);
+
 	adev->irq.installed = true;
 	r = drm_irq_install(adev->ddev, adev->ddev->pdev->irq);
 	if (r) {
@@ -367,8 +402,6 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
 	bool handled = false;
 	int r;
 
-	trace_amdgpu_iv(entry);
-
 	if (client_id >= AMDGPU_IRQ_CLIENTID_MAX) {
 		DRM_DEBUG("Invalid client_id in IV: %d\n", client_id);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
index f6ce171cb8aa..c27decfda494 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
@@ -87,9 +87,11 @@ struct amdgpu_irq {
 	/* status, etc. */
 	bool				msi_enabled; /* msi enabled */
 
-	/* interrupt ring */
-	struct amdgpu_ih_ring		ih;
-	const struct amdgpu_ih_funcs	*ih_funcs;
+	/* interrupt rings */
+	struct amdgpu_ih_ring		ih, ih1, ih2;
+	const struct amdgpu_ih_funcs    *ih_funcs;
+	struct work_struct		ih1_work, ih2_work;
+	struct amdgpu_irq_src		self_irq;
 
 	/* gen irq stuff */
 	struct irq_domain		*domain; /* GPU irq controller domain */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index aadd0fa42e43..698fd8a2f775 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -38,6 +38,7 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/i2c.h>
 #include <linux/i2c-algo-bit.h>
 #include <linux/hrtimer.h>
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 728e15e5d68a..fd9c4beeaaa4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -426,12 +426,20 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 	size_t acc_size;
 	int r;
 
-	page_align = roundup(bp->byte_align, PAGE_SIZE) >> PAGE_SHIFT;
-	if (bp->domain & (AMDGPU_GEM_DOMAIN_GDS | AMDGPU_GEM_DOMAIN_GWS |
-			  AMDGPU_GEM_DOMAIN_OA))
+	/* Note that GDS/GWS/OA allocates 1 page per byte/resource. */
+	if (bp->domain & (AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) {
+		/* GWS and OA don't need any alignment. */
+		page_align = bp->byte_align;
 		size <<= PAGE_SHIFT;
-	else
+	} else if (bp->domain & AMDGPU_GEM_DOMAIN_GDS) {
+		/* Both size and alignment must be a multiple of 4. */
+		page_align = ALIGN(bp->byte_align, 4);
+		size = ALIGN(size, 4) << PAGE_SHIFT;
+	} else {
+		/* Memory should be aligned at least to a page size. */
+		page_align = ALIGN(bp->byte_align, PAGE_SIZE) >> PAGE_SHIFT;
 		size = ALIGN(size, PAGE_SIZE);
+	}
 
 	if (!amdgpu_bo_validate_size(adev, size, bp->domain))
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index 1f61ed95727c..a7adb7b6bd98 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -626,11 +626,71 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev,
 }
 
 /**
- * DOC: pp_dpm_sclk pp_dpm_mclk pp_dpm_pcie
+ * DOC: ppfeatures
+ *
+ * The amdgpu driver provides a sysfs API for adjusting what powerplay
+ * features to be enabled. The file ppfeatures is used for this. And
+ * this is only available for Vega10 and later dGPUs.
+ *
+ * Reading back the file will show you the followings:
+ * - Current ppfeature masks
+ * - List of the all supported powerplay features with their naming,
+ *   bitmasks and enablement status('Y'/'N' means "enabled"/"disabled").
+ *
+ * To manually enable or disable a specific feature, just set or clear
+ * the corresponding bit from original ppfeature masks and input the
+ * new ppfeature masks.
+ */
+static ssize_t amdgpu_set_ppfeature_status(struct device *dev,
+		struct device_attribute *attr,
+		const char *buf,
+		size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	uint64_t featuremask;
+	int ret;
+
+	ret = kstrtou64(buf, 0, &featuremask);
+	if (ret)
+		return -EINVAL;
+
+	pr_debug("featuremask = 0x%llx\n", featuremask);
+
+	if (adev->powerplay.pp_funcs->set_ppfeature_status) {
+		ret = amdgpu_dpm_set_ppfeature_status(adev, featuremask);
+		if (ret)
+			return -EINVAL;
+	}
+
+	return count;
+}
+
+static ssize_t amdgpu_get_ppfeature_status(struct device *dev,
+		struct device_attribute *attr,
+		char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	if (adev->powerplay.pp_funcs->get_ppfeature_status)
+		return amdgpu_dpm_get_ppfeature_status(adev, buf);
+
+	return snprintf(buf, PAGE_SIZE, "\n");
+}
+
+/**
+ * DOC: pp_dpm_sclk pp_dpm_mclk pp_dpm_socclk pp_dpm_fclk pp_dpm_dcefclk
+ * pp_dpm_pcie
  *
  * The amdgpu driver provides a sysfs API for adjusting what power levels
  * are enabled for a given power state.  The files pp_dpm_sclk, pp_dpm_mclk,
- * and pp_dpm_pcie are used for this.
+ * pp_dpm_socclk, pp_dpm_fclk, pp_dpm_dcefclk and pp_dpm_pcie are used for
+ * this.
+ *
+ * pp_dpm_socclk and pp_dpm_dcefclk interfaces are only available for
+ * Vega10 and later ASICs.
+ * pp_dpm_fclk interface is only available for Vega20 and later ASICs.
  *
  * Reading back the files will show you the available power levels within
  * the power state and the clock information for those levels.
@@ -640,6 +700,8 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev,
  * Secondly,Enter a new value for each level by inputing a string that
  * contains " echo xx xx xx > pp_dpm_sclk/mclk/pcie"
  * E.g., echo 4 5 6 to > pp_dpm_sclk will enable sclk levels 4, 5, and 6.
+ *
+ * NOTE: change to the dcefclk max dpm level is not supported now
  */
 
 static ssize_t amdgpu_get_pp_dpm_sclk(struct device *dev,
@@ -750,6 +812,114 @@ static ssize_t amdgpu_set_pp_dpm_mclk(struct device *dev,
 	return count;
 }
 
+static ssize_t amdgpu_get_pp_dpm_socclk(struct device *dev,
+		struct device_attribute *attr,
+		char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	if (adev->powerplay.pp_funcs->print_clock_levels)
+		return amdgpu_dpm_print_clock_levels(adev, PP_SOCCLK, buf);
+	else
+		return snprintf(buf, PAGE_SIZE, "\n");
+}
+
+static ssize_t amdgpu_set_pp_dpm_socclk(struct device *dev,
+		struct device_attribute *attr,
+		const char *buf,
+		size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	int ret;
+	uint32_t mask = 0;
+
+	ret = amdgpu_read_mask(buf, count, &mask);
+	if (ret)
+		return ret;
+
+	if (adev->powerplay.pp_funcs->force_clock_level)
+		ret = amdgpu_dpm_force_clock_level(adev, PP_SOCCLK, mask);
+
+	if (ret)
+		return -EINVAL;
+
+	return count;
+}
+
+static ssize_t amdgpu_get_pp_dpm_fclk(struct device *dev,
+		struct device_attribute *attr,
+		char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	if (adev->powerplay.pp_funcs->print_clock_levels)
+		return amdgpu_dpm_print_clock_levels(adev, PP_FCLK, buf);
+	else
+		return snprintf(buf, PAGE_SIZE, "\n");
+}
+
+static ssize_t amdgpu_set_pp_dpm_fclk(struct device *dev,
+		struct device_attribute *attr,
+		const char *buf,
+		size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	int ret;
+	uint32_t mask = 0;
+
+	ret = amdgpu_read_mask(buf, count, &mask);
+	if (ret)
+		return ret;
+
+	if (adev->powerplay.pp_funcs->force_clock_level)
+		ret = amdgpu_dpm_force_clock_level(adev, PP_FCLK, mask);
+
+	if (ret)
+		return -EINVAL;
+
+	return count;
+}
+
+static ssize_t amdgpu_get_pp_dpm_dcefclk(struct device *dev,
+		struct device_attribute *attr,
+		char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	if (adev->powerplay.pp_funcs->print_clock_levels)
+		return amdgpu_dpm_print_clock_levels(adev, PP_DCEFCLK, buf);
+	else
+		return snprintf(buf, PAGE_SIZE, "\n");
+}
+
+static ssize_t amdgpu_set_pp_dpm_dcefclk(struct device *dev,
+		struct device_attribute *attr,
+		const char *buf,
+		size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	int ret;
+	uint32_t mask = 0;
+
+	ret = amdgpu_read_mask(buf, count, &mask);
+	if (ret)
+		return ret;
+
+	if (adev->powerplay.pp_funcs->force_clock_level)
+		ret = amdgpu_dpm_force_clock_level(adev, PP_DCEFCLK, mask);
+
+	if (ret)
+		return -EINVAL;
+
+	return count;
+}
+
 static ssize_t amdgpu_get_pp_dpm_pcie(struct device *dev,
 		struct device_attribute *attr,
 		char *buf)
@@ -990,6 +1160,31 @@ static ssize_t amdgpu_get_busy_percent(struct device *dev,
 	return snprintf(buf, PAGE_SIZE, "%d\n", value);
 }
 
+/**
+ * DOC: pcie_bw
+ *
+ * The amdgpu driver provides a sysfs API for estimating how much data
+ * has been received and sent by the GPU in the last second through PCIe.
+ * The file pcie_bw is used for this.
+ * The Perf counters count the number of received and sent messages and return
+ * those values, as well as the maximum payload size of a PCIe packet (mps).
+ * Note that it is not possible to easily and quickly obtain the size of each
+ * packet transmitted, so we output the max payload size (mps) to allow for
+ * quick estimation of the PCIe bandwidth usage
+ */
+static ssize_t amdgpu_get_pcie_bw(struct device *dev,
+		struct device_attribute *attr,
+		char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	uint64_t count0, count1;
+
+	amdgpu_asic_get_pcie_usage(adev, &count0, &count1);
+	return snprintf(buf, PAGE_SIZE,	"%llu %llu %i\n",
+			count0, count1, pcie_get_mps(adev->pdev));
+}
+
 static DEVICE_ATTR(power_dpm_state, S_IRUGO | S_IWUSR, amdgpu_get_dpm_state, amdgpu_set_dpm_state);
 static DEVICE_ATTR(power_dpm_force_performance_level, S_IRUGO | S_IWUSR,
 		   amdgpu_get_dpm_forced_performance_level,
@@ -1008,6 +1203,15 @@ static DEVICE_ATTR(pp_dpm_sclk, S_IRUGO | S_IWUSR,
 static DEVICE_ATTR(pp_dpm_mclk, S_IRUGO | S_IWUSR,
 		amdgpu_get_pp_dpm_mclk,
 		amdgpu_set_pp_dpm_mclk);
+static DEVICE_ATTR(pp_dpm_socclk, S_IRUGO | S_IWUSR,
+		amdgpu_get_pp_dpm_socclk,
+		amdgpu_set_pp_dpm_socclk);
+static DEVICE_ATTR(pp_dpm_fclk, S_IRUGO | S_IWUSR,
+		amdgpu_get_pp_dpm_fclk,
+		amdgpu_set_pp_dpm_fclk);
+static DEVICE_ATTR(pp_dpm_dcefclk, S_IRUGO | S_IWUSR,
+		amdgpu_get_pp_dpm_dcefclk,
+		amdgpu_set_pp_dpm_dcefclk);
 static DEVICE_ATTR(pp_dpm_pcie, S_IRUGO | S_IWUSR,
 		amdgpu_get_pp_dpm_pcie,
 		amdgpu_set_pp_dpm_pcie);
@@ -1025,6 +1229,10 @@ static DEVICE_ATTR(pp_od_clk_voltage, S_IRUGO | S_IWUSR,
 		amdgpu_set_pp_od_clk_voltage);
 static DEVICE_ATTR(gpu_busy_percent, S_IRUGO,
 		amdgpu_get_busy_percent, NULL);
+static DEVICE_ATTR(pcie_bw, S_IRUGO, amdgpu_get_pcie_bw, NULL);
+static DEVICE_ATTR(ppfeatures, S_IRUGO | S_IWUSR,
+		amdgpu_get_ppfeature_status,
+		amdgpu_set_ppfeature_status);
 
 static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 				      struct device_attribute *attr,
@@ -1516,6 +1724,75 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 	return count;
 }
 
+static ssize_t amdgpu_hwmon_show_sclk(struct device *dev,
+				      struct device_attribute *attr,
+				      char *buf)
+{
+	struct amdgpu_device *adev = dev_get_drvdata(dev);
+	struct drm_device *ddev = adev->ddev;
+	uint32_t sclk;
+	int r, size = sizeof(sclk);
+
+	/* Can't get voltage when the card is off */
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
+	/* sanity check PP is enabled */
+	if (!(adev->powerplay.pp_funcs &&
+	      adev->powerplay.pp_funcs->read_sensor))
+	      return -EINVAL;
+
+	/* get the sclk */
+	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_SCLK,
+				   (void *)&sclk, &size);
+	if (r)
+		return r;
+
+	return snprintf(buf, PAGE_SIZE, "%d\n", sclk * 10 * 1000);
+}
+
+static ssize_t amdgpu_hwmon_show_sclk_label(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "sclk\n");
+}
+
+static ssize_t amdgpu_hwmon_show_mclk(struct device *dev,
+				      struct device_attribute *attr,
+				      char *buf)
+{
+	struct amdgpu_device *adev = dev_get_drvdata(dev);
+	struct drm_device *ddev = adev->ddev;
+	uint32_t mclk;
+	int r, size = sizeof(mclk);
+
+	/* Can't get voltage when the card is off */
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
+	/* sanity check PP is enabled */
+	if (!(adev->powerplay.pp_funcs &&
+	      adev->powerplay.pp_funcs->read_sensor))
+	      return -EINVAL;
+
+	/* get the sclk */
+	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_MCLK,
+				   (void *)&mclk, &size);
+	if (r)
+		return r;
+
+	return snprintf(buf, PAGE_SIZE, "%d\n", mclk * 10 * 1000);
+}
+
+static ssize_t amdgpu_hwmon_show_mclk_label(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "mclk\n");
+}
 
 /**
  * DOC: hwmon
@@ -1532,6 +1809,10 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
  *
  * - GPU fan
  *
+ * - GPU gfx/compute engine clock
+ *
+ * - GPU memory clock (dGPU only)
+ *
  * hwmon interfaces for GPU temperature:
  *
  * - temp1_input: the on die GPU temperature in millidegrees Celsius
@@ -1576,6 +1857,12 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
  *
  * - fan[1-*]_enable: Enable or disable the sensors.1: Enable 0: Disable
  *
+ * hwmon interfaces for GPU clocks:
+ *
+ * - freq1_input: the gfx/compute clock in hertz
+ *
+ * - freq2_input: the memory clock in hertz
+ *
  * You can use hwmon tools like sensors to view this information on your system.
  *
  */
@@ -1600,6 +1887,10 @@ static SENSOR_DEVICE_ATTR(power1_average, S_IRUGO, amdgpu_hwmon_show_power_avg,
 static SENSOR_DEVICE_ATTR(power1_cap_max, S_IRUGO, amdgpu_hwmon_show_power_cap_max, NULL, 0);
 static SENSOR_DEVICE_ATTR(power1_cap_min, S_IRUGO, amdgpu_hwmon_show_power_cap_min, NULL, 0);
 static SENSOR_DEVICE_ATTR(power1_cap, S_IRUGO | S_IWUSR, amdgpu_hwmon_show_power_cap, amdgpu_hwmon_set_power_cap, 0);
+static SENSOR_DEVICE_ATTR(freq1_input, S_IRUGO, amdgpu_hwmon_show_sclk, NULL, 0);
+static SENSOR_DEVICE_ATTR(freq1_label, S_IRUGO, amdgpu_hwmon_show_sclk_label, NULL, 0);
+static SENSOR_DEVICE_ATTR(freq2_input, S_IRUGO, amdgpu_hwmon_show_mclk, NULL, 0);
+static SENSOR_DEVICE_ATTR(freq2_label, S_IRUGO, amdgpu_hwmon_show_mclk_label, NULL, 0);
 
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
@@ -1622,6 +1913,10 @@ static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_power1_cap_max.dev_attr.attr,
 	&sensor_dev_attr_power1_cap_min.dev_attr.attr,
 	&sensor_dev_attr_power1_cap.dev_attr.attr,
+	&sensor_dev_attr_freq1_input.dev_attr.attr,
+	&sensor_dev_attr_freq1_label.dev_attr.attr,
+	&sensor_dev_attr_freq2_input.dev_attr.attr,
+	&sensor_dev_attr_freq2_label.dev_attr.attr,
 	NULL
 };
 
@@ -1686,7 +1981,8 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 		effective_mode &= ~S_IWUSR;
 
 	if ((adev->flags & AMD_IS_APU) &&
-	    (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
+	    (attr == &sensor_dev_attr_power1_average.dev_attr.attr ||
+	     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;
@@ -1712,6 +2008,12 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_in1_label.dev_attr.attr))
 		return 0;
 
+	/* no mclk on APUs */
+	if ((adev->flags & AMD_IS_APU) &&
+	    (attr == &sensor_dev_attr_freq2_input.dev_attr.attr ||
+	     attr == &sensor_dev_attr_freq2_label.dev_attr.attr))
+		return 0;
+
 	return effective_mode;
 }
 
@@ -2008,6 +2310,7 @@ void amdgpu_pm_print_power_states(struct amdgpu_device *adev)
 
 int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 {
+	struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
 	int ret;
 
 	if (adev->pm.sysfs_initialized)
@@ -2069,6 +2372,25 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 		DRM_ERROR("failed to create device file pp_dpm_mclk\n");
 		return ret;
 	}
+	if (adev->asic_type >= CHIP_VEGA10) {
+		ret = device_create_file(adev->dev, &dev_attr_pp_dpm_socclk);
+		if (ret) {
+			DRM_ERROR("failed to create device file pp_dpm_socclk\n");
+			return ret;
+		}
+		ret = device_create_file(adev->dev, &dev_attr_pp_dpm_dcefclk);
+		if (ret) {
+			DRM_ERROR("failed to create device file pp_dpm_dcefclk\n");
+			return ret;
+		}
+	}
+	if (adev->asic_type >= CHIP_VEGA20) {
+		ret = device_create_file(adev->dev, &dev_attr_pp_dpm_fclk);
+		if (ret) {
+			DRM_ERROR("failed to create device file pp_dpm_fclk\n");
+			return ret;
+		}
+	}
 	ret = device_create_file(adev->dev, &dev_attr_pp_dpm_pcie);
 	if (ret) {
 		DRM_ERROR("failed to create device file pp_dpm_pcie\n");
@@ -2091,12 +2413,14 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 				"pp_power_profile_mode\n");
 		return ret;
 	}
-	ret = device_create_file(adev->dev,
-			&dev_attr_pp_od_clk_voltage);
-	if (ret) {
-		DRM_ERROR("failed to create device file	"
-				"pp_od_clk_voltage\n");
-		return ret;
+	if (hwmgr->od_enabled) {
+		ret = device_create_file(adev->dev,
+				&dev_attr_pp_od_clk_voltage);
+		if (ret) {
+			DRM_ERROR("failed to create device file	"
+					"pp_od_clk_voltage\n");
+			return ret;
+		}
 	}
 	ret = device_create_file(adev->dev,
 			&dev_attr_gpu_busy_percent);
@@ -2105,12 +2429,31 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 				"gpu_busy_level\n");
 		return ret;
 	}
+	/* PCIe Perf counters won't work on APU nodes */
+	if (!(adev->flags & AMD_IS_APU)) {
+		ret = device_create_file(adev->dev, &dev_attr_pcie_bw);
+		if (ret) {
+			DRM_ERROR("failed to create device file pcie_bw\n");
+			return ret;
+		}
+	}
 	ret = amdgpu_debugfs_pm_init(adev);
 	if (ret) {
 		DRM_ERROR("Failed to register debugfs file for dpm!\n");
 		return ret;
 	}
 
+	if ((adev->asic_type >= CHIP_VEGA10) &&
+	    !(adev->flags & AMD_IS_APU)) {
+		ret = device_create_file(adev->dev,
+				&dev_attr_ppfeatures);
+		if (ret) {
+			DRM_ERROR("failed to create device file	"
+					"ppfeatures\n");
+			return ret;
+		}
+	}
+
 	adev->pm.sysfs_initialized = true;
 
 	return 0;
@@ -2118,6 +2461,8 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 
 void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev)
 {
+	struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
+
 	if (adev->pm.dpm_enabled == 0)
 		return;
 
@@ -2133,14 +2478,26 @@ void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev)
 
 	device_remove_file(adev->dev, &dev_attr_pp_dpm_sclk);
 	device_remove_file(adev->dev, &dev_attr_pp_dpm_mclk);
+	if (adev->asic_type >= CHIP_VEGA10) {
+		device_remove_file(adev->dev, &dev_attr_pp_dpm_socclk);
+		device_remove_file(adev->dev, &dev_attr_pp_dpm_dcefclk);
+	}
 	device_remove_file(adev->dev, &dev_attr_pp_dpm_pcie);
+	if (adev->asic_type >= CHIP_VEGA20)
+		device_remove_file(adev->dev, &dev_attr_pp_dpm_fclk);
 	device_remove_file(adev->dev, &dev_attr_pp_sclk_od);
 	device_remove_file(adev->dev, &dev_attr_pp_mclk_od);
 	device_remove_file(adev->dev,
 			&dev_attr_pp_power_profile_mode);
-	device_remove_file(adev->dev,
-			&dev_attr_pp_od_clk_voltage);
+	if (hwmgr->od_enabled)
+		device_remove_file(adev->dev,
+				&dev_attr_pp_od_clk_voltage);
 	device_remove_file(adev->dev, &dev_attr_gpu_busy_percent);
+	if (!(adev->flags & AMD_IS_APU))
+		device_remove_file(adev->dev, &dev_attr_pcie_bw);
+	if ((adev->asic_type >= CHIP_VEGA10) &&
+	    !(adev->flags & AMD_IS_APU))
+		device_remove_file(adev->dev, &dev_attr_ppfeatures);
 }
 
 void amdgpu_pm_compute_clocks(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
index 71913a18d142..a38e0fb4a6fe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c
@@ -38,6 +38,7 @@
 #include "amdgpu_gem.h"
 #include <drm/amdgpu_drm.h>
 #include <linux/dma-buf.h>
+#include <linux/dma-fence-array.h>
 
 /**
  * amdgpu_gem_prime_get_sg_table - &drm_driver.gem_prime_get_sg_table
@@ -187,6 +188,48 @@ error:
 	return ERR_PTR(ret);
 }
 
+static int
+__reservation_object_make_exclusive(struct reservation_object *obj)
+{
+	struct dma_fence **fences;
+	unsigned int count;
+	int r;
+
+	if (!reservation_object_get_list(obj)) /* no shared fences to convert */
+		return 0;
+
+	r = reservation_object_get_fences_rcu(obj, NULL, &count, &fences);
+	if (r)
+		return r;
+
+	if (count == 0) {
+		/* Now that was unexpected. */
+	} else if (count == 1) {
+		reservation_object_add_excl_fence(obj, fences[0]);
+		dma_fence_put(fences[0]);
+		kfree(fences);
+	} else {
+		struct dma_fence_array *array;
+
+		array = dma_fence_array_create(count, fences,
+					       dma_fence_context_alloc(1), 0,
+					       false);
+		if (!array)
+			goto err_fences_put;
+
+		reservation_object_add_excl_fence(obj, &array->base);
+		dma_fence_put(&array->base);
+	}
+
+	return 0;
+
+err_fences_put:
+	while (count--)
+		dma_fence_put(fences[count]);
+	kfree(fences);
+	return -ENOMEM;
+}
+
 /**
  * amdgpu_gem_map_attach - &dma_buf_ops.attach implementation
  * @dma_buf: Shared DMA buffer
@@ -218,16 +261,16 @@ static int amdgpu_gem_map_attach(struct dma_buf *dma_buf,
 
 	if (attach->dev->driver != adev->dev->driver) {
 		/*
-		 * Wait for all shared fences to complete before we switch to future
-		 * use of exclusive fence on this prime shared bo.
+		 * We only create shared fences for internal use, but importers
+		 * of the dmabuf rely on exclusive fences for implicitly
+		 * tracking write hazards. As any of the current fences may
+		 * correspond to a write, we need to convert all existing
+		 * fences on the reservation object into a single exclusive
+		 * fence.
 		 */
-		r = reservation_object_wait_timeout_rcu(bo->tbo.resv,
-							true, false,
-							MAX_SCHEDULE_TIMEOUT);
-		if (unlikely(r < 0)) {
-			DRM_DEBUG_PRIME("Fence wait failed: %li\n", r);
+		r = __reservation_object_make_exclusive(bo->tbo.resv);
+		if (r)
 			goto error_unreserve;
-		}
 	}
 
 	/* pin buffer into GTT */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 8fab0d637ee5..3091488cd8cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -67,9 +67,6 @@ static int psp_sw_init(void *handle)
 
 	psp->adev = adev;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	ret = psp_init_microcode(psp);
 	if (ret) {
 		DRM_ERROR("Failed to load psp firmware!\n");
@@ -83,15 +80,14 @@ static int psp_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	release_firmware(adev->psp.sos_fw);
 	adev->psp.sos_fw = NULL;
 	release_firmware(adev->psp.asd_fw);
 	adev->psp.asd_fw = NULL;
-	release_firmware(adev->psp.ta_fw);
-	adev->psp.ta_fw = NULL;
+	if (adev->psp.ta_fw) {
+		release_firmware(adev->psp.ta_fw);
+		adev->psp.ta_fw = NULL;
+	}
 	return 0;
 }
 
@@ -140,13 +136,24 @@ psp_cmd_submit_buf(struct psp_context *psp,
 	while (*((unsigned int *)psp->fence_buf) != index)
 		msleep(1);
 
-	/* the status field must be 0 after FW is loaded */
-	if (ucode && psp->cmd_buf_mem->resp.status) {
-		DRM_ERROR("failed loading with status (%d) and ucode id (%d)\n",
-			  psp->cmd_buf_mem->resp.status, ucode->ucode_id);
-		return -EINVAL;
+	/* In some cases, psp response status is not 0 even there is no
+	 * problem while the command is submitted. Some version of PSP FW
+	 * doesn't write 0 to that field.
+	 * So here we would like to only print a warning instead of an error
+	 * during psp initialization to avoid breaking hw_init and it doesn't
+	 * return -EINVAL.
+	 */
+	if (psp->cmd_buf_mem->resp.status) {
+		if (ucode)
+			DRM_WARN("failed to load ucode id (%d) ",
+				  ucode->ucode_id);
+		DRM_WARN("psp command failed and response status is (%d)\n",
+			  psp->cmd_buf_mem->resp.status);
 	}
 
+	/* get xGMI session id from response buffer */
+	cmd->resp.session_id = psp->cmd_buf_mem->resp.session_id;
+
 	if (ucode) {
 		ucode->tmr_mc_addr_lo = psp->cmd_buf_mem->resp.fw_addr_lo;
 		ucode->tmr_mc_addr_hi = psp->cmd_buf_mem->resp.fw_addr_hi;
@@ -435,6 +442,9 @@ static int psp_xgmi_initialize(struct psp_context *psp)
 	struct ta_xgmi_shared_memory *xgmi_cmd;
 	int ret;
 
+	if (!psp->adev->psp.ta_fw)
+		return -ENOENT;
+
 	if (!psp->xgmi_context.initialized) {
 		ret = psp_xgmi_init_shared_buf(psp);
 		if (ret)
@@ -495,6 +505,98 @@ static int psp_hw_start(struct psp_context *psp)
 	return 0;
 }
 
+static int psp_get_fw_type(struct amdgpu_firmware_info *ucode,
+			   enum psp_gfx_fw_type *type)
+{
+	switch (ucode->ucode_id) {
+	case AMDGPU_UCODE_ID_SDMA0:
+		*type = GFX_FW_TYPE_SDMA0;
+		break;
+	case AMDGPU_UCODE_ID_SDMA1:
+		*type = GFX_FW_TYPE_SDMA1;
+		break;
+	case AMDGPU_UCODE_ID_CP_CE:
+		*type = GFX_FW_TYPE_CP_CE;
+		break;
+	case AMDGPU_UCODE_ID_CP_PFP:
+		*type = GFX_FW_TYPE_CP_PFP;
+		break;
+	case AMDGPU_UCODE_ID_CP_ME:
+		*type = GFX_FW_TYPE_CP_ME;
+		break;
+	case AMDGPU_UCODE_ID_CP_MEC1:
+		*type = GFX_FW_TYPE_CP_MEC;
+		break;
+	case AMDGPU_UCODE_ID_CP_MEC1_JT:
+		*type = GFX_FW_TYPE_CP_MEC_ME1;
+		break;
+	case AMDGPU_UCODE_ID_CP_MEC2:
+		*type = GFX_FW_TYPE_CP_MEC;
+		break;
+	case AMDGPU_UCODE_ID_CP_MEC2_JT:
+		*type = GFX_FW_TYPE_CP_MEC_ME2;
+		break;
+	case AMDGPU_UCODE_ID_RLC_G:
+		*type = GFX_FW_TYPE_RLC_G;
+		break;
+	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL:
+		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_SRM_CNTL;
+		break;
+	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM:
+		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_GPM_MEM;
+		break;
+	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM:
+		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_SRM_MEM;
+		break;
+	case AMDGPU_UCODE_ID_SMC:
+		*type = GFX_FW_TYPE_SMU;
+		break;
+	case AMDGPU_UCODE_ID_UVD:
+		*type = GFX_FW_TYPE_UVD;
+		break;
+	case AMDGPU_UCODE_ID_UVD1:
+		*type = GFX_FW_TYPE_UVD1;
+		break;
+	case AMDGPU_UCODE_ID_VCE:
+		*type = GFX_FW_TYPE_VCE;
+		break;
+	case AMDGPU_UCODE_ID_VCN:
+		*type = GFX_FW_TYPE_VCN;
+		break;
+	case AMDGPU_UCODE_ID_DMCU_ERAM:
+		*type = GFX_FW_TYPE_DMCU_ERAM;
+		break;
+	case AMDGPU_UCODE_ID_DMCU_INTV:
+		*type = GFX_FW_TYPE_DMCU_ISR;
+		break;
+	case AMDGPU_UCODE_ID_MAXIMUM:
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode,
+				       struct psp_gfx_cmd_resp *cmd)
+{
+	int ret;
+	uint64_t fw_mem_mc_addr = ucode->mc_addr;
+
+	memset(cmd, 0, sizeof(struct psp_gfx_cmd_resp));
+
+	cmd->cmd_id = GFX_CMD_ID_LOAD_IP_FW;
+	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_lo = lower_32_bits(fw_mem_mc_addr);
+	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_hi = upper_32_bits(fw_mem_mc_addr);
+	cmd->cmd.cmd_load_ip_fw.fw_size = ucode->ucode_size;
+
+	ret = psp_get_fw_type(ucode, &cmd->cmd.cmd_load_ip_fw.fw_type);
+	if (ret)
+		DRM_ERROR("Unknown firmware type\n");
+
+	return ret;
+}
+
 static int psp_np_fw_load(struct psp_context *psp)
 {
 	int i, ret;
@@ -516,7 +618,7 @@ static int psp_np_fw_load(struct psp_context *psp)
 			/*skip ucode loading in SRIOV VF */
 			continue;
 
-		ret = psp_prep_cmd_buf(ucode, psp->cmd);
+		ret = psp_prep_load_ip_fw_cmd_buf(ucode, psp->cmd);
 		if (ret)
 			return ret;
 
@@ -541,7 +643,7 @@ static int psp_load_fw(struct amdgpu_device *adev)
 	struct psp_context *psp = &adev->psp;
 
 	if (amdgpu_sriov_vf(adev) && adev->in_gpu_reset) {
-		psp_ring_destroy(psp, PSP_RING_TYPE__KM);
+		psp_ring_stop(psp, PSP_RING_TYPE__KM); /* should not destroy ring, only stop */
 		goto skip_memalloc;
 	}
 
@@ -618,10 +720,6 @@ static int psp_hw_init(void *handle)
 	int ret;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	mutex_lock(&adev->firmware.mutex);
 	/*
 	 * This sequence is just used on hw_init only once, no need on
@@ -651,9 +749,6 @@ static int psp_hw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct psp_context *psp = &adev->psp;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	if (adev->gmc.xgmi.num_physical_nodes > 1 &&
 	    psp->xgmi_context.initialized == 1)
                 psp_xgmi_terminate(psp);
@@ -682,9 +777,6 @@ static int psp_suspend(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct psp_context *psp = &adev->psp;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	if (adev->gmc.xgmi.num_physical_nodes > 1 &&
 	    psp->xgmi_context.initialized == 1) {
 		ret = psp_xgmi_terminate(psp);
@@ -709,9 +801,6 @@ static int psp_resume(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct psp_context *psp = &adev->psp;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
-		return 0;
-
 	DRM_INFO("PSP is resuming...\n");
 
 	mutex_lock(&adev->firmware.mutex);
@@ -747,11 +836,6 @@ static bool psp_check_fw_loading_status(struct amdgpu_device *adev,
 {
 	struct amdgpu_firmware_info *ucode = NULL;
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-		DRM_INFO("firmware is not loaded by PSP\n");
-		return true;
-	}
-
 	if (!adev->firmware.fw_size)
 		return false;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 3ee573b4016e..2ef98cc755d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -65,8 +65,6 @@ struct psp_funcs
 	int (*init_microcode)(struct psp_context *psp);
 	int (*bootloader_load_sysdrv)(struct psp_context *psp);
 	int (*bootloader_load_sos)(struct psp_context *psp);
-	int (*prep_cmd_buf)(struct amdgpu_firmware_info *ucode,
-			    struct psp_gfx_cmd_resp *cmd);
 	int (*ring_init)(struct psp_context *psp, enum psp_ring_type ring_type);
 	int (*ring_create)(struct psp_context *psp,
 			   enum psp_ring_type ring_type);
@@ -176,7 +174,6 @@ struct psp_xgmi_topology_info {
 	struct psp_xgmi_node_info	nodes[AMDGPU_XGMI_MAX_CONNECTED_NODES];
 };
 
-#define psp_prep_cmd_buf(ucode, type) (psp)->funcs->prep_cmd_buf((ucode), (type))
 #define psp_ring_init(psp, type) (psp)->funcs->ring_init((psp), (type))
 #define psp_ring_create(psp, type) (psp)->funcs->ring_create((psp), (type))
 #define psp_ring_stop(psp, type) (psp)->funcs->ring_stop((psp), (type))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index d87e828a084b..d7fae2676269 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -131,7 +131,7 @@ struct amdgpu_ring_funcs {
 	void (*emit_ib)(struct amdgpu_ring *ring,
 			struct amdgpu_job *job,
 			struct amdgpu_ib *ib,
-			bool ctx_switch);
+			uint32_t flags);
 	void (*emit_fence)(struct amdgpu_ring *ring, uint64_t addr,
 			   uint64_t seq, unsigned flags);
 	void (*emit_pipeline_sync)(struct amdgpu_ring *ring);
@@ -229,7 +229,7 @@ struct amdgpu_ring {
 #define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r))
 #define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
 #define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
-#define amdgpu_ring_emit_ib(r, job, ib, c) ((r)->funcs->emit_ib((r), (job), (ib), (c)))
+#define amdgpu_ring_emit_ib(r, job, ib, flags) ((r)->funcs->emit_ib((r), (job), (ib), (flags)))
 #define amdgpu_ring_emit_pipeline_sync(r) (r)->funcs->emit_pipeline_sync((r))
 #define amdgpu_ring_emit_vm_flush(r, vmid, addr) (r)->funcs->emit_vm_flush((r), (vmid), (addr))
 #define amdgpu_ring_emit_fence(r, addr, seq, flags) (r)->funcs->emit_fence((r), (addr), (seq), (flags))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index 626abca770a0..d3ca2424b5fe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -76,9 +76,10 @@ TRACE_EVENT(amdgpu_mm_wreg,
 );
 
 TRACE_EVENT(amdgpu_iv,
-	    TP_PROTO(struct amdgpu_iv_entry *iv),
-	    TP_ARGS(iv),
+	    TP_PROTO(unsigned ih, struct amdgpu_iv_entry *iv),
+	    TP_ARGS(ih, iv),
 	    TP_STRUCT__entry(
+			     __field(unsigned, ih)
 			     __field(unsigned, client_id)
 			     __field(unsigned, src_id)
 			     __field(unsigned, ring_id)
@@ -90,6 +91,7 @@ TRACE_EVENT(amdgpu_iv,
 			     __array(unsigned, src_data, 4)
 			    ),
 	    TP_fast_assign(
+			   __entry->ih = ih;
 			   __entry->client_id = iv->client_id;
 			   __entry->src_id = iv->src_id;
 			   __entry->ring_id = iv->ring_id;
@@ -103,8 +105,9 @@ TRACE_EVENT(amdgpu_iv,
 			   __entry->src_data[2] = iv->src_data[2];
 			   __entry->src_data[3] = iv->src_data[3];
 			   ),
-	    TP_printk("client_id:%u src_id:%u ring:%u vmid:%u timestamp: %llu pasid:%u src_data: %08x %08x %08x %08x",
-		      __entry->client_id, __entry->src_id,
+	    TP_printk("ih:%u client_id:%u src_id:%u ring:%u vmid:%u "
+		      "timestamp: %llu pasid:%u src_data: %08x %08x %08x %08x",
+		      __entry->ih, __entry->client_id, __entry->src_id,
 		      __entry->ring_id, __entry->vmid,
 		      __entry->timestamp, __entry->pasid,
 		      __entry->src_data[0], __entry->src_data[1],
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index c91ec3101d00..73e71e61dc99 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1546,7 +1546,8 @@ static struct ttm_bo_driver amdgpu_bo_driver = {
 	.io_mem_reserve = &amdgpu_ttm_io_mem_reserve,
 	.io_mem_free = &amdgpu_ttm_io_mem_free,
 	.io_mem_pfn = amdgpu_ttm_io_mem_pfn,
-	.access_memory = &amdgpu_ttm_access_memory
+	.access_memory = &amdgpu_ttm_access_memory,
+	.del_from_lru_notify = &amdgpu_vm_del_from_lru_notify
 };
 
 /*
@@ -1755,7 +1756,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_kernel(adev, adev->gds.mem.gfx_partition_size,
-				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GDS,
+				    4, AMDGPU_GEM_DOMAIN_GDS,
 				    &adev->gds.gds_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
@@ -1768,7 +1769,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_kernel(adev, adev->gds.gws.gfx_partition_size,
-				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GWS,
+				    1, AMDGPU_GEM_DOMAIN_GWS,
 				    &adev->gds.gws_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
@@ -1781,7 +1782,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_kernel(adev, adev->gds.oa.gfx_partition_size,
-				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_OA,
+				    1, AMDGPU_GEM_DOMAIN_OA,
 				    &adev->gds.oa_gfx_bo, NULL, NULL);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 98a1b2ce2b9d..c021b114c8a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -1035,7 +1035,7 @@ out:
 void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring,
 				struct amdgpu_job *job,
 				struct amdgpu_ib *ib,
-				bool ctx_switch)
+				uint32_t flags)
 {
 	amdgpu_ring_write(ring, VCE_CMD_IB);
 	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
index 50293652af14..30ea54dd9117 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
@@ -66,7 +66,7 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp);
 int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx);
 int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx);
 void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
-				struct amdgpu_ib *ib, bool ctx_switch);
+				struct amdgpu_ib *ib, uint32_t flags);
 void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
 				unsigned flags);
 int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index e73d152659a2..7cd2336e29ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -107,14 +107,6 @@ struct amdgpu_pte_update_params {
 	 * DMA addresses to use for mapping, used during VM update by CPU
 	 */
 	dma_addr_t *pages_addr;
-
-	/**
-	 * @kptr:
-	 *
-	 * Kernel pointer of PD/PT BO that needs to be updated,
-	 * used during VM update by CPU
-	 */
-	void *kptr;
 };
 
 /**
@@ -623,6 +615,28 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
 	list_add(&entry->tv.head, validated);
 }
 
+void amdgpu_vm_del_from_lru_notify(struct ttm_buffer_object *bo)
+{
+	struct amdgpu_bo *abo;
+	struct amdgpu_vm_bo_base *bo_base;
+
+	if (!amdgpu_bo_is_amdgpu_bo(bo))
+		return;
+
+	if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT)
+		return;
+
+	abo = ttm_to_amdgpu_bo(bo);
+	if (!abo->parent)
+		return;
+	for (bo_base = abo->vm_bo; bo_base; bo_base = bo_base->next) {
+		struct amdgpu_vm *vm = bo_base->vm;
+
+		if (abo->tbo.resv == vm->root.base.bo->tbo.resv)
+			vm->bulk_moveable = false;
+	}
+
+}
 /**
  * amdgpu_vm_move_to_lru_tail - move all BOs to the end of LRU
  *
@@ -799,9 +813,16 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 		addr += ats_entries * 8;
 	}
 
-	if (entries)
+	if (entries) {
+		uint64_t value = 0;
+
+		/* Workaround for fault priority problem on GMC9 */
+		if (level == AMDGPU_VM_PTB && adev->asic_type >= CHIP_VEGA10)
+			value = AMDGPU_PTE_EXECUTABLE;
+
 		amdgpu_vm_set_pte_pde(adev, &job->ibs[0], addr, 0,
-				      entries, 0, 0);
+				      entries, 0, value);
+	}
 
 	amdgpu_ring_pad_ib(ring, &job->ibs[0]);
 
@@ -847,9 +868,6 @@ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	bp->size = amdgpu_vm_bo_size(adev, level);
 	bp->byte_align = AMDGPU_GPU_PAGE_SIZE;
 	bp->domain = AMDGPU_GEM_DOMAIN_VRAM;
-	if (bp->size <= PAGE_SIZE && adev->asic_type >= CHIP_VEGA10 &&
-	    adev->flags & AMD_IS_APU)
-		bp->domain |= AMDGPU_GEM_DOMAIN_GTT;
 	bp->domain = amdgpu_bo_get_preferred_pin_domain(adev, bp->domain);
 	bp->flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
 		AMDGPU_GEM_CREATE_CPU_GTT_USWC;
@@ -1506,20 +1524,27 @@ error:
 }
 
 /**
- * amdgpu_vm_update_huge - figure out parameters for PTE updates
+ * amdgpu_vm_update_flags - figure out flags for PTE updates
  *
  * Make sure to set the right flags for the PTEs at the desired level.
  */
-static void amdgpu_vm_update_huge(struct amdgpu_pte_update_params *params,
-				  struct amdgpu_bo *bo, unsigned level,
-				  uint64_t pe, uint64_t addr,
-				  unsigned count, uint32_t incr,
-				  uint64_t flags)
+static void amdgpu_vm_update_flags(struct amdgpu_pte_update_params *params,
+				   struct amdgpu_bo *bo, unsigned level,
+				   uint64_t pe, uint64_t addr,
+				   unsigned count, uint32_t incr,
+				   uint64_t flags)
 
 {
 	if (level != AMDGPU_VM_PTB) {
 		flags |= AMDGPU_PDE_PTE;
 		amdgpu_gmc_get_vm_pde(params->adev, level, &addr, &flags);
+
+	} else if (params->adev->asic_type >= CHIP_VEGA10 &&
+		   !(flags & AMDGPU_PTE_VALID) &&
+		   !(flags & AMDGPU_PTE_PRT)) {
+
+		/* Workaround for fault priority problem on GMC9 */
+		flags |= AMDGPU_PTE_EXECUTABLE;
 	}
 
 	amdgpu_vm_update_func(params, bo, pe, addr, count, incr, flags);
@@ -1676,9 +1701,9 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 			uint64_t upd_end = min(entry_end, frag_end);
 			unsigned nptes = (upd_end - frag_start) >> shift;
 
-			amdgpu_vm_update_huge(params, pt, cursor.level,
-					      pe_start, dst, nptes, incr,
-					      flags | AMDGPU_PTE_FRAG(frag));
+			amdgpu_vm_update_flags(params, pt, cursor.level,
+					       pe_start, dst, nptes, incr,
+					       flags | AMDGPU_PTE_FRAG(frag));
 
 			pe_start += nptes * 8;
 			dst += (uint64_t)nptes * AMDGPU_GPU_PAGE_SIZE << shift;
@@ -1756,13 +1781,20 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 		if (pages_addr)
 			params.src = ~0;
 
-		/* Wait for PT BOs to be free. PTs share the same resv. object
+		/* Wait for PT BOs to be idle. PTs share the same resv. object
 		 * as the root PD BO
 		 */
 		r = amdgpu_vm_wait_pd(adev, vm, owner);
 		if (unlikely(r))
 			return r;
 
+		/* Wait for any BO move to be completed */
+		if (exclusive) {
+			r = dma_fence_wait(exclusive, true);
+			if (unlikely(r))
+				return r;
+		}
+
 		params.func = amdgpu_vm_cpu_set_ptes;
 		params.pages_addr = pages_addr;
 		return amdgpu_vm_update_ptes(&params, start, last + 1,
@@ -1776,13 +1808,12 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 	/*
 	 * reserve space for two commands every (1 << BLOCK_SIZE)
 	 *  entries or 2k dwords (whatever is smaller)
-         *
-         * The second command is for the shadow pagetables.
 	 */
+	ncmds = ((nptes >> min(adev->vm_manager.block_size, 11u)) + 1);
+
+	/* The second command is for the shadow pagetables. */
 	if (vm->root.base.bo->shadow)
-		ncmds = ((nptes >> min(adev->vm_manager.block_size, 11u)) + 1) * 2;
-	else
-		ncmds = ((nptes >> min(adev->vm_manager.block_size, 11u)) + 1);
+		ncmds *= 2;
 
 	/* padding, etc. */
 	ndw = 64;
@@ -1801,10 +1832,11 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 		ndw += ncmds * 10;
 
 		/* extra commands for begin/end fragments */
+		ncmds = 2 * adev->vm_manager.fragment_size;
 		if (vm->root.base.bo->shadow)
-		        ndw += 2 * 10 * adev->vm_manager.fragment_size * 2;
-		else
-		        ndw += 2 * 10 * adev->vm_manager.fragment_size;
+			ncmds *= 2;
+
+		ndw += 10 * ncmds;
 
 		params.func = amdgpu_vm_do_set_ptes;
 	}
@@ -3006,7 +3038,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
-	WARN_ONCE((vm->use_cpu_for_update & !amdgpu_gmc_vram_full_visible(&adev->gmc)),
+	WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)),
 		  "CPU update of VM recommended only for large BAR system\n");
 	vm->last_update = NULL;
 
@@ -3136,7 +3168,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, uns
 	vm->pte_support_ats = pte_support_ats;
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
-	WARN_ONCE((vm->use_cpu_for_update & !amdgpu_gmc_vram_full_visible(&adev->gmc)),
+	WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)),
 		  "CPU update of VM recommended only for large BAR system\n");
 
 	if (vm->pasid) {
@@ -3366,14 +3398,15 @@ void amdgpu_vm_get_task_info(struct amdgpu_device *adev, unsigned int pasid,
 			 struct amdgpu_task_info *task_info)
 {
 	struct amdgpu_vm *vm;
+	unsigned long flags;
 
-	spin_lock(&adev->vm_manager.pasid_lock);
+	spin_lock_irqsave(&adev->vm_manager.pasid_lock, flags);
 
 	vm = idr_find(&adev->vm_manager.pasid_idr, pasid);
 	if (vm)
 		*task_info = vm->task_info;
 
-	spin_unlock(&adev->vm_manager.pasid_lock);
+	spin_unlock_irqrestore(&adev->vm_manager.pasid_lock, flags);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index e8dcfd59fc93..81ff8177f092 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -363,4 +363,6 @@ int amdgpu_vm_add_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key)
 
 void amdgpu_vm_clear_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key);
 
+void amdgpu_vm_del_from_lru_notify(struct ttm_buffer_object *bo);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 8a8bc60cb6b4..407dd16cc35c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -40,26 +40,40 @@ void *amdgpu_xgmi_hive_try_lock(struct amdgpu_hive_info *hive)
 	return &hive->device_list;
 }
 
-struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev)
+struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock)
 {
 	int i;
 	struct amdgpu_hive_info *tmp;
 
 	if (!adev->gmc.xgmi.hive_id)
 		return NULL;
+
+	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 (tmp->hive_id == adev->gmc.xgmi.hive_id) {
+			if (lock)
+				mutex_lock(&tmp->hive_lock);
+			mutex_unlock(&xgmi_mutex);
 			return tmp;
+		}
 	}
-	if (i >= AMDGPU_MAX_XGMI_HIVE)
+	if (i >= AMDGPU_MAX_XGMI_HIVE) {
+		mutex_unlock(&xgmi_mutex);
 		return NULL;
+	}
 
 	/* initialize new hive if not exist */
 	tmp = &xgmi_hives[hive_count++];
 	tmp->hive_id = adev->gmc.xgmi.hive_id;
 	INIT_LIST_HEAD(&tmp->device_list);
 	mutex_init(&tmp->hive_lock);
+	mutex_init(&tmp->reset_lock);
+	if (lock)
+		mutex_lock(&tmp->hive_lock);
+
+	mutex_unlock(&xgmi_mutex);
 
 	return tmp;
 }
@@ -77,10 +91,6 @@ int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_dev
 			"XGMI: Set topology failure on device %llx, hive %llx, ret %d",
 			adev->gmc.xgmi.node_id,
 			adev->gmc.xgmi.hive_id, ret);
-	else
-		dev_info(adev->dev, "XGMI: Set topology for node %d, hive 0x%llx.\n",
-			 adev->gmc.xgmi.physical_node_id,
-				 adev->gmc.xgmi.hive_id);
 
 	return ret;
 }
@@ -111,10 +121,14 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 		return ret;
 	}
 
-	mutex_lock(&xgmi_mutex);
-	hive = amdgpu_get_xgmi_hive(adev);
-	if (!hive)
+	hive = amdgpu_get_xgmi_hive(adev, 1);
+	if (!hive) {
+		ret = -EINVAL;
+		dev_err(adev->dev,
+			"XGMI: node 0x%llx, can not match hive 0x%llx in the hive list.\n",
+			adev->gmc.xgmi.node_id, adev->gmc.xgmi.hive_id);
 		goto exit;
+	}
 
 	hive_topology = &hive->topology_info;
 
@@ -142,8 +156,11 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 			break;
 	}
 
+	dev_info(adev->dev, "XGMI: Add node %d, hive 0x%llx.\n",
+		 adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id);
+
+	mutex_unlock(&hive->hive_lock);
 exit:
-	mutex_unlock(&xgmi_mutex);
 	return ret;
 }
 
@@ -154,15 +171,14 @@ void amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 	if (!adev->gmc.xgmi.supported)
 		return;
 
-	mutex_lock(&xgmi_mutex);
-
-	hive = amdgpu_get_xgmi_hive(adev);
+	hive = amdgpu_get_xgmi_hive(adev, 1);
 	if (!hive)
-		goto exit;
+		return;
 
-	if (!(hive->number_devices--))
+	if (!(hive->number_devices--)) {
 		mutex_destroy(&hive->hive_lock);
-
-exit:
-	mutex_unlock(&xgmi_mutex);
+		mutex_destroy(&hive->reset_lock);
+	} else {
+		mutex_unlock(&hive->hive_lock);
+	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index 6151eb9c8ad3..14bc60664159 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -29,10 +29,11 @@ struct amdgpu_hive_info {
 	struct list_head	device_list;
 	struct psp_xgmi_topology_info	topology_info;
 	int number_devices;
-	struct mutex hive_lock;
+	struct mutex hive_lock,
+		     reset_lock;
 };
 
-struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev);
+struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock);
 int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev);
 int amdgpu_xgmi_add_device(struct amdgpu_device *adev);
 void amdgpu_xgmi_remove_device(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
deleted file mode 100644
index 86e14c754dd4..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
+++ /dev/null
@@ -1,6844 +0,0 @@
-/*
- * Copyright 2013 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/firmware.h>
-#include <drm/drmP.h>
-#include "amdgpu.h"
-#include "amdgpu_pm.h"
-#include "amdgpu_ucode.h"
-#include "cikd.h"
-#include "amdgpu_dpm.h"
-#include "ci_dpm.h"
-#include "gfx_v7_0.h"
-#include "atom.h"
-#include "amd_pcie.h"
-#include <linux/seq_file.h>
-
-#include "smu/smu_7_0_1_d.h"
-#include "smu/smu_7_0_1_sh_mask.h"
-
-#include "dce/dce_8_0_d.h"
-#include "dce/dce_8_0_sh_mask.h"
-
-#include "bif/bif_4_1_d.h"
-#include "bif/bif_4_1_sh_mask.h"
-
-#include "gca/gfx_7_2_d.h"
-#include "gca/gfx_7_2_sh_mask.h"
-
-#include "gmc/gmc_7_1_d.h"
-#include "gmc/gmc_7_1_sh_mask.h"
-
-MODULE_FIRMWARE("amdgpu/bonaire_smc.bin");
-MODULE_FIRMWARE("amdgpu/bonaire_k_smc.bin");
-MODULE_FIRMWARE("amdgpu/hawaii_smc.bin");
-MODULE_FIRMWARE("amdgpu/hawaii_k_smc.bin");
-
-#define MC_CG_ARB_FREQ_F0           0x0a
-#define MC_CG_ARB_FREQ_F1           0x0b
-#define MC_CG_ARB_FREQ_F2           0x0c
-#define MC_CG_ARB_FREQ_F3           0x0d
-
-#define SMC_RAM_END 0x40000
-
-#define VOLTAGE_SCALE               4
-#define VOLTAGE_VID_OFFSET_SCALE1    625
-#define VOLTAGE_VID_OFFSET_SCALE2    100
-
-static const struct amd_pm_funcs ci_dpm_funcs;
-
-static const struct ci_pt_defaults defaults_hawaii_xt =
-{
-	1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0xB0000,
-	{ 0x2E,  0x00,  0x00,  0x88,  0x00,  0x00,  0x72,  0x60,  0x51,  0xA7,  0x79,  0x6B,  0x90,  0xBD,  0x79  },
-	{ 0x217, 0x217, 0x217, 0x242, 0x242, 0x242, 0x269, 0x269, 0x269, 0x2A1, 0x2A1, 0x2A1, 0x2C9, 0x2C9, 0x2C9 }
-};
-
-static const struct ci_pt_defaults defaults_hawaii_pro =
-{
-	1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0x65062,
-	{ 0x2E,  0x00,  0x00,  0x88,  0x00,  0x00,  0x72,  0x60,  0x51,  0xA7,  0x79,  0x6B,  0x90,  0xBD,  0x79  },
-	{ 0x217, 0x217, 0x217, 0x242, 0x242, 0x242, 0x269, 0x269, 0x269, 0x2A1, 0x2A1, 0x2A1, 0x2C9, 0x2C9, 0x2C9 }
-};
-
-static const struct ci_pt_defaults defaults_bonaire_xt =
-{
-	1, 0xF, 0xFD, 0x19, 5, 45, 0, 0xB0000,
-	{ 0x79,  0x253, 0x25D, 0xAE,  0x72,  0x80,  0x83,  0x86,  0x6F,  0xC8,  0xC9,  0xC9,  0x2F,  0x4D,  0x61  },
-	{ 0x17C, 0x172, 0x180, 0x1BC, 0x1B3, 0x1BD, 0x206, 0x200, 0x203, 0x25D, 0x25A, 0x255, 0x2C3, 0x2C5, 0x2B4 }
-};
-
-#if 0
-static const struct ci_pt_defaults defaults_bonaire_pro =
-{
-	1, 0xF, 0xFD, 0x19, 5, 45, 0, 0x65062,
-	{ 0x8C,  0x23F, 0x244, 0xA6,  0x83,  0x85,  0x86,  0x86,  0x83,  0xDB,  0xDB,  0xDA,  0x67,  0x60,  0x5F  },
-	{ 0x187, 0x193, 0x193, 0x1C7, 0x1D1, 0x1D1, 0x210, 0x219, 0x219, 0x266, 0x26C, 0x26C, 0x2C9, 0x2CB, 0x2CB }
-};
-#endif
-
-static const struct ci_pt_defaults defaults_saturn_xt =
-{
-	1, 0xF, 0xFD, 0x19, 5, 55, 0, 0x70000,
-	{ 0x8C,  0x247, 0x249, 0xA6,  0x80,  0x81,  0x8B,  0x89,  0x86,  0xC9,  0xCA,  0xC9,  0x4D,  0x4D,  0x4D  },
-	{ 0x187, 0x187, 0x187, 0x1C7, 0x1C7, 0x1C7, 0x210, 0x210, 0x210, 0x266, 0x266, 0x266, 0x2C9, 0x2C9, 0x2C9 }
-};
-
-#if 0
-static const struct ci_pt_defaults defaults_saturn_pro =
-{
-	1, 0xF, 0xFD, 0x19, 5, 55, 0, 0x30000,
-	{ 0x96,  0x21D, 0x23B, 0xA1,  0x85,  0x87,  0x83,  0x84,  0x81,  0xE6,  0xE6,  0xE6,  0x71,  0x6A,  0x6A  },
-	{ 0x193, 0x19E, 0x19E, 0x1D2, 0x1DC, 0x1DC, 0x21A, 0x223, 0x223, 0x26E, 0x27E, 0x274, 0x2CF, 0x2D2, 0x2D2 }
-};
-#endif
-
-static const struct ci_pt_config_reg didt_config_ci[] =
-{
-	{ 0x10, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x10, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x10, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x10, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x11, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x11, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x11, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x11, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x12, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x12, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x12, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x12, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x2, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x2, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x2, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x1, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x1, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x0, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x30, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x30, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x30, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x30, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x31, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x31, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x31, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x31, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x32, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x32, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x32, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x32, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x22, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x22, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x22, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x21, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x21, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x20, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x50, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x50, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x50, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x50, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x51, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x51, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x51, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x51, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x52, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x52, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x52, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x52, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x42, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x42, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x42, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x41, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x41, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x40, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x70, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x70, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x70, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x70, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x71, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x71, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x71, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x71, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x72, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x72, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x72, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x72, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x62, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x62, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x62, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x61, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x61, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0x60, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
-	{ 0xFFFFFFFF }
-};
-
-static u8 ci_get_memory_module_index(struct amdgpu_device *adev)
-{
-	return (u8) ((RREG32(mmBIOS_SCRATCH_4) >> 16) & 0xff);
-}
-
-#define MC_CG_ARB_FREQ_F0           0x0a
-#define MC_CG_ARB_FREQ_F1           0x0b
-#define MC_CG_ARB_FREQ_F2           0x0c
-#define MC_CG_ARB_FREQ_F3           0x0d
-
-static int ci_copy_and_switch_arb_sets(struct amdgpu_device *adev,
-				       u32 arb_freq_src, u32 arb_freq_dest)
-{
-	u32 mc_arb_dram_timing;
-	u32 mc_arb_dram_timing2;
-	u32 burst_time;
-	u32 mc_cg_config;
-
-	switch (arb_freq_src) {
-	case MC_CG_ARB_FREQ_F0:
-		mc_arb_dram_timing  = RREG32(mmMC_ARB_DRAM_TIMING);
-		mc_arb_dram_timing2 = RREG32(mmMC_ARB_DRAM_TIMING2);
-		burst_time = (RREG32(mmMC_ARB_BURST_TIME) & MC_ARB_BURST_TIME__STATE0_MASK) >>
-			 MC_ARB_BURST_TIME__STATE0__SHIFT;
-		break;
-	case MC_CG_ARB_FREQ_F1:
-		mc_arb_dram_timing  = RREG32(mmMC_ARB_DRAM_TIMING_1);
-		mc_arb_dram_timing2 = RREG32(mmMC_ARB_DRAM_TIMING2_1);
-		burst_time = (RREG32(mmMC_ARB_BURST_TIME) & MC_ARB_BURST_TIME__STATE1_MASK) >>
-			 MC_ARB_BURST_TIME__STATE1__SHIFT;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	switch (arb_freq_dest) {
-	case MC_CG_ARB_FREQ_F0:
-		WREG32(mmMC_ARB_DRAM_TIMING, mc_arb_dram_timing);
-		WREG32(mmMC_ARB_DRAM_TIMING2, mc_arb_dram_timing2);
-		WREG32_P(mmMC_ARB_BURST_TIME, (burst_time << MC_ARB_BURST_TIME__STATE0__SHIFT),
-			~MC_ARB_BURST_TIME__STATE0_MASK);
-		break;
-	case MC_CG_ARB_FREQ_F1:
-		WREG32(mmMC_ARB_DRAM_TIMING_1, mc_arb_dram_timing);
-		WREG32(mmMC_ARB_DRAM_TIMING2_1, mc_arb_dram_timing2);
-		WREG32_P(mmMC_ARB_BURST_TIME, (burst_time << MC_ARB_BURST_TIME__STATE1__SHIFT),
-			~MC_ARB_BURST_TIME__STATE1_MASK);
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	mc_cg_config = RREG32(mmMC_CG_CONFIG) | 0x0000000F;
-	WREG32(mmMC_CG_CONFIG, mc_cg_config);
-	WREG32_P(mmMC_ARB_CG, (arb_freq_dest) << MC_ARB_CG__CG_ARB_REQ__SHIFT,
-		~MC_ARB_CG__CG_ARB_REQ_MASK);
-
-	return 0;
-}
-
-static u8 ci_get_ddr3_mclk_frequency_ratio(u32 memory_clock)
-{
-	u8 mc_para_index;
-
-	if (memory_clock < 10000)
-		mc_para_index = 0;
-	else if (memory_clock >= 80000)
-		mc_para_index = 0x0f;
-	else
-		mc_para_index = (u8)((memory_clock - 10000) / 5000 + 1);
-	return mc_para_index;
-}
-
-static u8 ci_get_mclk_frequency_ratio(u32 memory_clock, bool strobe_mode)
-{
-	u8 mc_para_index;
-
-	if (strobe_mode) {
-		if (memory_clock < 12500)
-			mc_para_index = 0x00;
-		else if (memory_clock > 47500)
-			mc_para_index = 0x0f;
-		else
-			mc_para_index = (u8)((memory_clock - 10000) / 2500);
-	} else {
-		if (memory_clock < 65000)
-			mc_para_index = 0x00;
-		else if (memory_clock > 135000)
-			mc_para_index = 0x0f;
-		else
-			mc_para_index = (u8)((memory_clock - 60000) / 5000);
-	}
-	return mc_para_index;
-}
-
-static void ci_trim_voltage_table_to_fit_state_table(struct amdgpu_device *adev,
-						     u32 max_voltage_steps,
-						     struct atom_voltage_table *voltage_table)
-{
-	unsigned int i, diff;
-
-	if (voltage_table->count <= max_voltage_steps)
-		return;
-
-	diff = voltage_table->count - max_voltage_steps;
-
-	for (i = 0; i < max_voltage_steps; i++)
-		voltage_table->entries[i] = voltage_table->entries[i + diff];
-
-	voltage_table->count = max_voltage_steps;
-}
-
-static int ci_get_std_voltage_value_sidd(struct amdgpu_device *adev,
-					 struct atom_voltage_table_entry *voltage_table,
-					 u16 *std_voltage_hi_sidd, u16 *std_voltage_lo_sidd);
-static int ci_set_power_limit(struct amdgpu_device *adev, u32 n);
-static int ci_set_overdrive_target_tdp(struct amdgpu_device *adev,
-				       u32 target_tdp);
-static int ci_update_uvd_dpm(struct amdgpu_device *adev, bool gate);
-static void ci_dpm_set_irq_funcs(struct amdgpu_device *adev);
-
-static PPSMC_Result amdgpu_ci_send_msg_to_smc_with_parameter(struct amdgpu_device *adev,
-							     PPSMC_Msg msg, u32 parameter);
-static void ci_thermal_start_smc_fan_control(struct amdgpu_device *adev);
-static void ci_fan_ctrl_set_default_mode(struct amdgpu_device *adev);
-
-static struct ci_power_info *ci_get_pi(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = adev->pm.dpm.priv;
-
-	return pi;
-}
-
-static struct ci_ps *ci_get_ps(struct amdgpu_ps *rps)
-{
-	struct ci_ps *ps = rps->ps_priv;
-
-	return ps;
-}
-
-static void ci_initialize_powertune_defaults(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	switch (adev->pdev->device) {
-	case 0x6649:
-	case 0x6650:
-	case 0x6651:
-	case 0x6658:
-	case 0x665C:
-	case 0x665D:
-	default:
-		pi->powertune_defaults = &defaults_bonaire_xt;
-		break;
-	case 0x6640:
-	case 0x6641:
-	case 0x6646:
-	case 0x6647:
-		pi->powertune_defaults = &defaults_saturn_xt;
-		break;
-	case 0x67B8:
-	case 0x67B0:
-		pi->powertune_defaults = &defaults_hawaii_xt;
-		break;
-	case 0x67BA:
-	case 0x67B1:
-		pi->powertune_defaults = &defaults_hawaii_pro;
-		break;
-	case 0x67A0:
-	case 0x67A1:
-	case 0x67A2:
-	case 0x67A8:
-	case 0x67A9:
-	case 0x67AA:
-	case 0x67B9:
-	case 0x67BE:
-		pi->powertune_defaults = &defaults_bonaire_xt;
-		break;
-	}
-
-	pi->dte_tj_offset = 0;
-
-	pi->caps_power_containment = true;
-	pi->caps_cac = false;
-	pi->caps_sq_ramping = false;
-	pi->caps_db_ramping = false;
-	pi->caps_td_ramping = false;
-	pi->caps_tcp_ramping = false;
-
-	if (pi->caps_power_containment) {
-		pi->caps_cac = true;
-		if (adev->asic_type == CHIP_HAWAII)
-			pi->enable_bapm_feature = false;
-		else
-			pi->enable_bapm_feature = true;
-		pi->enable_tdc_limit_feature = true;
-		pi->enable_pkg_pwr_tracking_feature = true;
-	}
-}
-
-static u8 ci_convert_to_vid(u16 vddc)
-{
-	return (6200 - (vddc * VOLTAGE_SCALE)) / 25;
-}
-
-static int ci_populate_bapm_vddc_vid_sidd(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u8 *hi_vid = pi->smc_powertune_table.BapmVddCVidHiSidd;
-	u8 *lo_vid = pi->smc_powertune_table.BapmVddCVidLoSidd;
-	u8 *hi2_vid = pi->smc_powertune_table.BapmVddCVidHiSidd2;
-	u32 i;
-
-	if (adev->pm.dpm.dyn_state.cac_leakage_table.entries == NULL)
-		return -EINVAL;
-	if (adev->pm.dpm.dyn_state.cac_leakage_table.count > 8)
-		return -EINVAL;
-	if (adev->pm.dpm.dyn_state.cac_leakage_table.count !=
-	    adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count)
-		return -EINVAL;
-
-	for (i = 0; i < adev->pm.dpm.dyn_state.cac_leakage_table.count; i++) {
-		if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) {
-			lo_vid[i] = ci_convert_to_vid(adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc1);
-			hi_vid[i] = ci_convert_to_vid(adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc2);
-			hi2_vid[i] = ci_convert_to_vid(adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc3);
-		} else {
-			lo_vid[i] = ci_convert_to_vid(adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc);
-			hi_vid[i] = ci_convert_to_vid((u16)adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage);
-		}
-	}
-	return 0;
-}
-
-static int ci_populate_vddc_vid(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u8 *vid = pi->smc_powertune_table.VddCVid;
-	u32 i;
-
-	if (pi->vddc_voltage_table.count > 8)
-		return -EINVAL;
-
-	for (i = 0; i < pi->vddc_voltage_table.count; i++)
-		vid[i] = ci_convert_to_vid(pi->vddc_voltage_table.entries[i].value);
-
-	return 0;
-}
-
-static int ci_populate_svi_load_line(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
-
-	pi->smc_powertune_table.SviLoadLineEn = pt_defaults->svi_load_line_en;
-	pi->smc_powertune_table.SviLoadLineVddC = pt_defaults->svi_load_line_vddc;
-	pi->smc_powertune_table.SviLoadLineTrimVddC = 3;
-	pi->smc_powertune_table.SviLoadLineOffsetVddC = 0;
-
-	return 0;
-}
-
-static int ci_populate_tdc_limit(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
-	u16 tdc_limit;
-
-	tdc_limit = adev->pm.dpm.dyn_state.cac_tdp_table->tdc * 256;
-	pi->smc_powertune_table.TDC_VDDC_PkgLimit = cpu_to_be16(tdc_limit);
-	pi->smc_powertune_table.TDC_VDDC_ThrottleReleaseLimitPerc =
-		pt_defaults->tdc_vddc_throttle_release_limit_perc;
-	pi->smc_powertune_table.TDC_MAWt = pt_defaults->tdc_mawt;
-
-	return 0;
-}
-
-static int ci_populate_dw8(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
-	int ret;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, PmFuseTable) +
-				     offsetof(SMU7_Discrete_PmFuses, TdcWaterfallCtl),
-				     (u32 *)&pi->smc_powertune_table.TdcWaterfallCtl,
-				     pi->sram_end);
-	if (ret)
-		return -EINVAL;
-	else
-		pi->smc_powertune_table.TdcWaterfallCtl = pt_defaults->tdc_waterfall_ctl;
-
-	return 0;
-}
-
-static int ci_populate_fuzzy_fan(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if ((adev->pm.dpm.fan.fan_output_sensitivity & (1 << 15)) ||
-	    (adev->pm.dpm.fan.fan_output_sensitivity == 0))
-		adev->pm.dpm.fan.fan_output_sensitivity =
-			adev->pm.dpm.fan.default_fan_output_sensitivity;
-
-	pi->smc_powertune_table.FuzzyFan_PwmSetDelta =
-		cpu_to_be16(adev->pm.dpm.fan.fan_output_sensitivity);
-
-	return 0;
-}
-
-static int ci_min_max_v_gnbl_pm_lid_from_bapm_vddc(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u8 *hi_vid = pi->smc_powertune_table.BapmVddCVidHiSidd;
-	u8 *lo_vid = pi->smc_powertune_table.BapmVddCVidLoSidd;
-	int i, min, max;
-
-	min = max = hi_vid[0];
-	for (i = 0; i < 8; i++) {
-		if (0 != hi_vid[i]) {
-			if (min > hi_vid[i])
-				min = hi_vid[i];
-			if (max < hi_vid[i])
-				max = hi_vid[i];
-		}
-
-		if (0 != lo_vid[i]) {
-			if (min > lo_vid[i])
-				min = lo_vid[i];
-			if (max < lo_vid[i])
-				max = lo_vid[i];
-		}
-	}
-
-	if ((min == 0) || (max == 0))
-		return -EINVAL;
-	pi->smc_powertune_table.GnbLPMLMaxVid = (u8)max;
-	pi->smc_powertune_table.GnbLPMLMinVid = (u8)min;
-
-	return 0;
-}
-
-static int ci_populate_bapm_vddc_base_leakage_sidd(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u16 hi_sidd = pi->smc_powertune_table.BapmVddCBaseLeakageHiSidd;
-	u16 lo_sidd = pi->smc_powertune_table.BapmVddCBaseLeakageLoSidd;
-	struct amdgpu_cac_tdp_table *cac_tdp_table =
-		adev->pm.dpm.dyn_state.cac_tdp_table;
-
-	hi_sidd = cac_tdp_table->high_cac_leakage / 100 * 256;
-	lo_sidd = cac_tdp_table->low_cac_leakage / 100 * 256;
-
-	pi->smc_powertune_table.BapmVddCBaseLeakageHiSidd = cpu_to_be16(hi_sidd);
-	pi->smc_powertune_table.BapmVddCBaseLeakageLoSidd = cpu_to_be16(lo_sidd);
-
-	return 0;
-}
-
-static int ci_populate_bapm_parameters_in_dpm_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
-	SMU7_Discrete_DpmTable  *dpm_table = &pi->smc_state_table;
-	struct amdgpu_cac_tdp_table *cac_tdp_table =
-		adev->pm.dpm.dyn_state.cac_tdp_table;
-	struct amdgpu_ppm_table *ppm = adev->pm.dpm.dyn_state.ppm_table;
-	int i, j, k;
-	const u16 *def1;
-	const u16 *def2;
-
-	dpm_table->DefaultTdp = cac_tdp_table->tdp * 256;
-	dpm_table->TargetTdp = cac_tdp_table->configurable_tdp * 256;
-
-	dpm_table->DTETjOffset = (u8)pi->dte_tj_offset;
-	dpm_table->GpuTjMax =
-		(u8)(pi->thermal_temp_setting.temperature_high / 1000);
-	dpm_table->GpuTjHyst = 8;
-
-	dpm_table->DTEAmbientTempBase = pt_defaults->dte_ambient_temp_base;
-
-	if (ppm) {
-		dpm_table->PPM_PkgPwrLimit = cpu_to_be16((u16)ppm->dgpu_tdp * 256 / 1000);
-		dpm_table->PPM_TemperatureLimit = cpu_to_be16((u16)ppm->tj_max * 256);
-	} else {
-		dpm_table->PPM_PkgPwrLimit = cpu_to_be16(0);
-		dpm_table->PPM_TemperatureLimit = cpu_to_be16(0);
-	}
-
-	dpm_table->BAPM_TEMP_GRADIENT = cpu_to_be32(pt_defaults->bapm_temp_gradient);
-	def1 = pt_defaults->bapmti_r;
-	def2 = pt_defaults->bapmti_rc;
-
-	for (i = 0; i < SMU7_DTE_ITERATIONS; i++) {
-		for (j = 0; j < SMU7_DTE_SOURCES; j++) {
-			for (k = 0; k < SMU7_DTE_SINKS; k++) {
-				dpm_table->BAPMTI_R[i][j][k] = cpu_to_be16(*def1);
-				dpm_table->BAPMTI_RC[i][j][k] = cpu_to_be16(*def2);
-				def1++;
-				def2++;
-			}
-		}
-	}
-
-	return 0;
-}
-
-static int ci_populate_pm_base(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 pm_fuse_table_offset;
-	int ret;
-
-	if (pi->caps_power_containment) {
-		ret = amdgpu_ci_read_smc_sram_dword(adev,
-					     SMU7_FIRMWARE_HEADER_LOCATION +
-					     offsetof(SMU7_Firmware_Header, PmFuseTable),
-					     &pm_fuse_table_offset, pi->sram_end);
-		if (ret)
-			return ret;
-		ret = ci_populate_bapm_vddc_vid_sidd(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_vddc_vid(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_svi_load_line(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_tdc_limit(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_dw8(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_fuzzy_fan(adev);
-		if (ret)
-			return ret;
-		ret = ci_min_max_v_gnbl_pm_lid_from_bapm_vddc(adev);
-		if (ret)
-			return ret;
-		ret = ci_populate_bapm_vddc_base_leakage_sidd(adev);
-		if (ret)
-			return ret;
-		ret = amdgpu_ci_copy_bytes_to_smc(adev, pm_fuse_table_offset,
-					   (u8 *)&pi->smc_powertune_table,
-					   sizeof(SMU7_Discrete_PmFuses), pi->sram_end);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-static void ci_do_enable_didt(struct amdgpu_device *adev, const bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 data;
-
-	if (pi->caps_sq_ramping) {
-		data = RREG32_DIDT(ixDIDT_SQ_CTRL0);
-		if (enable)
-			data |= DIDT_SQ_CTRL0__DIDT_CTRL_EN_MASK;
-		else
-			data &= ~DIDT_SQ_CTRL0__DIDT_CTRL_EN_MASK;
-		WREG32_DIDT(ixDIDT_SQ_CTRL0, data);
-	}
-
-	if (pi->caps_db_ramping) {
-		data = RREG32_DIDT(ixDIDT_DB_CTRL0);
-		if (enable)
-			data |= DIDT_DB_CTRL0__DIDT_CTRL_EN_MASK;
-		else
-			data &= ~DIDT_DB_CTRL0__DIDT_CTRL_EN_MASK;
-		WREG32_DIDT(ixDIDT_DB_CTRL0, data);
-	}
-
-	if (pi->caps_td_ramping) {
-		data = RREG32_DIDT(ixDIDT_TD_CTRL0);
-		if (enable)
-			data |= DIDT_TD_CTRL0__DIDT_CTRL_EN_MASK;
-		else
-			data &= ~DIDT_TD_CTRL0__DIDT_CTRL_EN_MASK;
-		WREG32_DIDT(ixDIDT_TD_CTRL0, data);
-	}
-
-	if (pi->caps_tcp_ramping) {
-		data = RREG32_DIDT(ixDIDT_TCP_CTRL0);
-		if (enable)
-			data |= DIDT_TCP_CTRL0__DIDT_CTRL_EN_MASK;
-		else
-			data &= ~DIDT_TCP_CTRL0__DIDT_CTRL_EN_MASK;
-		WREG32_DIDT(ixDIDT_TCP_CTRL0, data);
-	}
-}
-
-static int ci_program_pt_config_registers(struct amdgpu_device *adev,
-					  const struct ci_pt_config_reg *cac_config_regs)
-{
-	const struct ci_pt_config_reg *config_regs = cac_config_regs;
-	u32 data;
-	u32 cache = 0;
-
-	if (config_regs == NULL)
-		return -EINVAL;
-
-	while (config_regs->offset != 0xFFFFFFFF) {
-		if (config_regs->type == CISLANDS_CONFIGREG_CACHE) {
-			cache |= ((config_regs->value << config_regs->shift) & config_regs->mask);
-		} else {
-			switch (config_regs->type) {
-			case CISLANDS_CONFIGREG_SMC_IND:
-				data = RREG32_SMC(config_regs->offset);
-				break;
-			case CISLANDS_CONFIGREG_DIDT_IND:
-				data = RREG32_DIDT(config_regs->offset);
-				break;
-			default:
-				data = RREG32(config_regs->offset);
-				break;
-			}
-
-			data &= ~config_regs->mask;
-			data |= ((config_regs->value << config_regs->shift) & config_regs->mask);
-			data |= cache;
-
-			switch (config_regs->type) {
-			case CISLANDS_CONFIGREG_SMC_IND:
-				WREG32_SMC(config_regs->offset, data);
-				break;
-			case CISLANDS_CONFIGREG_DIDT_IND:
-				WREG32_DIDT(config_regs->offset, data);
-				break;
-			default:
-				WREG32(config_regs->offset, data);
-				break;
-			}
-			cache = 0;
-		}
-		config_regs++;
-	}
-	return 0;
-}
-
-static int ci_enable_didt(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-
-	if (pi->caps_sq_ramping || pi->caps_db_ramping ||
-	    pi->caps_td_ramping || pi->caps_tcp_ramping) {
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
-
-		if (enable) {
-			ret = ci_program_pt_config_registers(adev, didt_config_ci);
-			if (ret) {
-				amdgpu_gfx_rlc_exit_safe_mode(adev);
-				return ret;
-			}
-		}
-
-		ci_do_enable_didt(adev, enable);
-
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
-	}
-
-	return 0;
-}
-
-static int ci_enable_power_containment(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-	int ret = 0;
-
-	if (enable) {
-		pi->power_containment_features = 0;
-		if (pi->caps_power_containment) {
-			if (pi->enable_bapm_feature) {
-				smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_EnableDTE);
-				if (smc_result != PPSMC_Result_OK)
-					ret = -EINVAL;
-				else
-					pi->power_containment_features |= POWERCONTAINMENT_FEATURE_BAPM;
-			}
-
-			if (pi->enable_tdc_limit_feature) {
-				smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_TDCLimitEnable);
-				if (smc_result != PPSMC_Result_OK)
-					ret = -EINVAL;
-				else
-					pi->power_containment_features |= POWERCONTAINMENT_FEATURE_TDCLimit;
-			}
-
-			if (pi->enable_pkg_pwr_tracking_feature) {
-				smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_PkgPwrLimitEnable);
-				if (smc_result != PPSMC_Result_OK) {
-					ret = -EINVAL;
-				} else {
-					struct amdgpu_cac_tdp_table *cac_tdp_table =
-						adev->pm.dpm.dyn_state.cac_tdp_table;
-					u32 default_pwr_limit =
-						(u32)(cac_tdp_table->maximum_power_delivery_limit * 256);
-
-					pi->power_containment_features |= POWERCONTAINMENT_FEATURE_PkgPwrLimit;
-
-					ci_set_power_limit(adev, default_pwr_limit);
-				}
-			}
-		}
-	} else {
-		if (pi->caps_power_containment && pi->power_containment_features) {
-			if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_TDCLimit)
-				amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_TDCLimitDisable);
-
-			if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_BAPM)
-				amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DisableDTE);
-
-			if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit)
-				amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_PkgPwrLimitDisable);
-			pi->power_containment_features = 0;
-		}
-	}
-
-	return ret;
-}
-
-static int ci_enable_smc_cac(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-	int ret = 0;
-
-	if (pi->caps_cac) {
-		if (enable) {
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_EnableCac);
-			if (smc_result != PPSMC_Result_OK) {
-				ret = -EINVAL;
-				pi->cac_enabled = false;
-			} else {
-				pi->cac_enabled = true;
-			}
-		} else if (pi->cac_enabled) {
-			amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DisableCac);
-			pi->cac_enabled = false;
-		}
-	}
-
-	return ret;
-}
-
-static int ci_enable_thermal_based_sclk_dpm(struct amdgpu_device *adev,
-					    bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result = PPSMC_Result_OK;
-
-	if (pi->thermal_sclk_dpm_enabled) {
-		if (enable)
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_ENABLE_THERMAL_DPM);
-		else
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DISABLE_THERMAL_DPM);
-	}
-
-	if (smc_result == PPSMC_Result_OK)
-		return 0;
-	else
-		return -EINVAL;
-}
-
-static int ci_power_control_set_level(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_cac_tdp_table *cac_tdp_table =
-		adev->pm.dpm.dyn_state.cac_tdp_table;
-	s32 adjust_percent;
-	s32 target_tdp;
-	int ret = 0;
-	bool adjust_polarity = false; /* ??? */
-
-	if (pi->caps_power_containment) {
-		adjust_percent = adjust_polarity ?
-			adev->pm.dpm.tdp_adjustment : (-1 * adev->pm.dpm.tdp_adjustment);
-		target_tdp = ((100 + adjust_percent) *
-			      (s32)cac_tdp_table->configurable_tdp) / 100;
-
-		ret = ci_set_overdrive_target_tdp(adev, (u32)target_tdp);
-	}
-
-	return ret;
-}
-
-static void ci_dpm_powergate_uvd(void *handle, bool gate)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	pi->uvd_power_gated = gate;
-
-	if (gate) {
-		/* stop the UVD block */
-		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_UVD,
-						       AMD_PG_STATE_GATE);
-		ci_update_uvd_dpm(adev, gate);
-	} else {
-		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_UVD,
-						       AMD_PG_STATE_UNGATE);
-		ci_update_uvd_dpm(adev, gate);
-	}
-}
-
-static bool ci_dpm_vblank_too_short(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	u32 vblank_time = amdgpu_dpm_get_vblank_time(adev);
-	u32 switch_limit = adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5 ? 450 : 300;
-
-	/* disable mclk switching if the refresh is >120Hz, even if the
-	 * blanking period would allow it
-	 */
-	if (amdgpu_dpm_get_vrefresh(adev) > 120)
-		return true;
-
-	if (vblank_time < switch_limit)
-		return true;
-	else
-		return false;
-
-}
-
-static void ci_apply_state_adjust_rules(struct amdgpu_device *adev,
-					struct amdgpu_ps *rps)
-{
-	struct ci_ps *ps = ci_get_ps(rps);
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_clock_and_voltage_limits *max_limits;
-	bool disable_mclk_switching;
-	u32 sclk, mclk;
-	int i;
-
-	if (rps->vce_active) {
-		rps->evclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].evclk;
-		rps->ecclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].ecclk;
-	} else {
-		rps->evclk = 0;
-		rps->ecclk = 0;
-	}
-
-	if ((adev->pm.dpm.new_active_crtc_count > 1) ||
-	    ci_dpm_vblank_too_short(adev))
-		disable_mclk_switching = true;
-	else
-		disable_mclk_switching = false;
-
-	if ((rps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY)
-		pi->battery_state = true;
-	else
-		pi->battery_state = false;
-
-	if (adev->pm.ac_power)
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-	else
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
-
-	if (adev->pm.ac_power == false) {
-		for (i = 0; i < ps->performance_level_count; i++) {
-			if (ps->performance_levels[i].mclk > max_limits->mclk)
-				ps->performance_levels[i].mclk = max_limits->mclk;
-			if (ps->performance_levels[i].sclk > max_limits->sclk)
-				ps->performance_levels[i].sclk = max_limits->sclk;
-		}
-	}
-
-	/* XXX validate the min clocks required for display */
-
-	if (disable_mclk_switching) {
-		mclk  = ps->performance_levels[ps->performance_level_count - 1].mclk;
-		sclk = ps->performance_levels[0].sclk;
-	} else {
-		mclk = ps->performance_levels[0].mclk;
-		sclk = ps->performance_levels[0].sclk;
-	}
-
-	if (adev->pm.pm_display_cfg.min_core_set_clock > sclk)
-		sclk = adev->pm.pm_display_cfg.min_core_set_clock;
-
-	if (adev->pm.pm_display_cfg.min_mem_set_clock > mclk)
-		mclk = adev->pm.pm_display_cfg.min_mem_set_clock;
-
-	if (rps->vce_active) {
-		if (sclk < adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].sclk)
-			sclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].sclk;
-		if (mclk < adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].mclk)
-			mclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].mclk;
-	}
-
-	ps->performance_levels[0].sclk = sclk;
-	ps->performance_levels[0].mclk = mclk;
-
-	if (ps->performance_levels[1].sclk < ps->performance_levels[0].sclk)
-		ps->performance_levels[1].sclk = ps->performance_levels[0].sclk;
-
-	if (disable_mclk_switching) {
-		if (ps->performance_levels[0].mclk < ps->performance_levels[1].mclk)
-			ps->performance_levels[0].mclk = ps->performance_levels[1].mclk;
-	} else {
-		if (ps->performance_levels[1].mclk < ps->performance_levels[0].mclk)
-			ps->performance_levels[1].mclk = ps->performance_levels[0].mclk;
-	}
-}
-
-static int ci_thermal_set_temperature_range(struct amdgpu_device *adev,
-					    int min_temp, int max_temp)
-{
-	int low_temp = 0 * 1000;
-	int high_temp = 255 * 1000;
-	u32 tmp;
-
-	if (low_temp < min_temp)
-		low_temp = min_temp;
-	if (high_temp > max_temp)
-		high_temp = max_temp;
-	if (high_temp < low_temp) {
-		DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp);
-		return -EINVAL;
-	}
-
-	tmp = RREG32_SMC(ixCG_THERMAL_INT);
-	tmp &= ~(CG_THERMAL_INT__DIG_THERM_INTH_MASK | CG_THERMAL_INT__DIG_THERM_INTL_MASK);
-	tmp |= ((high_temp / 1000) << CG_THERMAL_INT__DIG_THERM_INTH__SHIFT) |
-		((low_temp / 1000)) << CG_THERMAL_INT__DIG_THERM_INTL__SHIFT;
-	WREG32_SMC(ixCG_THERMAL_INT, tmp);
-
-#if 0
-	/* XXX: need to figure out how to handle this properly */
-	tmp = RREG32_SMC(ixCG_THERMAL_CTRL);
-	tmp &= DIG_THERM_DPM_MASK;
-	tmp |= DIG_THERM_DPM(high_temp / 1000);
-	WREG32_SMC(ixCG_THERMAL_CTRL, tmp);
-#endif
-
-	adev->pm.dpm.thermal.min_temp = low_temp;
-	adev->pm.dpm.thermal.max_temp = high_temp;
-	return 0;
-}
-
-static int ci_thermal_enable_alert(struct amdgpu_device *adev,
-				   bool enable)
-{
-	u32 thermal_int = RREG32_SMC(ixCG_THERMAL_INT);
-	PPSMC_Result result;
-
-	if (enable) {
-		thermal_int &= ~(CG_THERMAL_INT_CTRL__THERM_INTH_MASK_MASK |
-				 CG_THERMAL_INT_CTRL__THERM_INTL_MASK_MASK);
-		WREG32_SMC(ixCG_THERMAL_INT, thermal_int);
-		result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_Thermal_Cntl_Enable);
-		if (result != PPSMC_Result_OK) {
-			DRM_DEBUG_KMS("Could not enable thermal interrupts.\n");
-			return -EINVAL;
-		}
-	} else {
-		thermal_int |= CG_THERMAL_INT_CTRL__THERM_INTH_MASK_MASK |
-			CG_THERMAL_INT_CTRL__THERM_INTL_MASK_MASK;
-		WREG32_SMC(ixCG_THERMAL_INT, thermal_int);
-		result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_Thermal_Cntl_Disable);
-		if (result != PPSMC_Result_OK) {
-			DRM_DEBUG_KMS("Could not disable thermal interrupts.\n");
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-static void ci_fan_ctrl_set_static_mode(struct amdgpu_device *adev, u32 mode)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-
-	if (pi->fan_ctrl_is_in_default_mode) {
-		tmp = (RREG32_SMC(ixCG_FDO_CTRL2) & CG_FDO_CTRL2__FDO_PWM_MODE_MASK)
-			>> CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
-		pi->fan_ctrl_default_mode = tmp;
-		tmp = (RREG32_SMC(ixCG_FDO_CTRL2) & CG_FDO_CTRL2__TMIN_MASK)
-			>> CG_FDO_CTRL2__TMIN__SHIFT;
-		pi->t_min = tmp;
-		pi->fan_ctrl_is_in_default_mode = false;
-	}
-
-	tmp = RREG32_SMC(ixCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TMIN_MASK;
-	tmp |= 0 << CG_FDO_CTRL2__TMIN__SHIFT;
-	WREG32_SMC(ixCG_FDO_CTRL2, tmp);
-
-	tmp = RREG32_SMC(ixCG_FDO_CTRL2) & ~CG_FDO_CTRL2__FDO_PWM_MODE_MASK;
-	tmp |= mode << CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
-	WREG32_SMC(ixCG_FDO_CTRL2, tmp);
-}
-
-static int ci_thermal_setup_fan_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	SMU7_Discrete_FanTable fan_table = { FDO_MODE_HARDWARE };
-	u32 duty100;
-	u32 t_diff1, t_diff2, pwm_diff1, pwm_diff2;
-	u16 fdo_min, slope1, slope2;
-	u32 reference_clock, tmp;
-	int ret;
-	u64 tmp64;
-
-	if (!pi->fan_table_start) {
-		adev->pm.dpm.fan.ucode_fan_control = false;
-		return 0;
-	}
-
-	duty100 = (RREG32_SMC(ixCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK)
-		>> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
-
-	if (duty100 == 0) {
-		adev->pm.dpm.fan.ucode_fan_control = false;
-		return 0;
-	}
-
-	tmp64 = (u64)adev->pm.dpm.fan.pwm_min * duty100;
-	do_div(tmp64, 10000);
-	fdo_min = (u16)tmp64;
-
-	t_diff1 = adev->pm.dpm.fan.t_med - adev->pm.dpm.fan.t_min;
-	t_diff2 = adev->pm.dpm.fan.t_high - adev->pm.dpm.fan.t_med;
-
-	pwm_diff1 = adev->pm.dpm.fan.pwm_med - adev->pm.dpm.fan.pwm_min;
-	pwm_diff2 = adev->pm.dpm.fan.pwm_high - adev->pm.dpm.fan.pwm_med;
-
-	slope1 = (u16)((50 + ((16 * duty100 * pwm_diff1) / t_diff1)) / 100);
-	slope2 = (u16)((50 + ((16 * duty100 * pwm_diff2) / t_diff2)) / 100);
-
-	fan_table.TempMin = cpu_to_be16((50 + adev->pm.dpm.fan.t_min) / 100);
-	fan_table.TempMed = cpu_to_be16((50 + adev->pm.dpm.fan.t_med) / 100);
-	fan_table.TempMax = cpu_to_be16((50 + adev->pm.dpm.fan.t_max) / 100);
-
-	fan_table.Slope1 = cpu_to_be16(slope1);
-	fan_table.Slope2 = cpu_to_be16(slope2);
-
-	fan_table.FdoMin = cpu_to_be16(fdo_min);
-
-	fan_table.HystDown = cpu_to_be16(adev->pm.dpm.fan.t_hyst);
-
-	fan_table.HystUp = cpu_to_be16(1);
-
-	fan_table.HystSlope = cpu_to_be16(1);
-
-	fan_table.TempRespLim = cpu_to_be16(5);
-
-	reference_clock = amdgpu_asic_get_xclk(adev);
-
-	fan_table.RefreshPeriod = cpu_to_be32((adev->pm.dpm.fan.cycle_delay *
-					       reference_clock) / 1600);
-
-	fan_table.FdoMax = cpu_to_be16((u16)duty100);
-
-	tmp = (RREG32_SMC(ixCG_MULT_THERMAL_CTRL) & CG_MULT_THERMAL_CTRL__TEMP_SEL_MASK)
-		>> CG_MULT_THERMAL_CTRL__TEMP_SEL__SHIFT;
-	fan_table.TempSrc = (uint8_t)tmp;
-
-	ret = amdgpu_ci_copy_bytes_to_smc(adev,
-					  pi->fan_table_start,
-					  (u8 *)(&fan_table),
-					  sizeof(fan_table),
-					  pi->sram_end);
-
-	if (ret) {
-		DRM_ERROR("Failed to load fan table to the SMC.");
-		adev->pm.dpm.fan.ucode_fan_control = false;
-	}
-
-	return 0;
-}
-
-static int ci_fan_ctrl_start_smc_fan_control(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result ret;
-
-	if (pi->caps_od_fuzzy_fan_control_support) {
-		ret = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-							       PPSMC_StartFanControl,
-							       FAN_CONTROL_FUZZY);
-		if (ret != PPSMC_Result_OK)
-			return -EINVAL;
-		ret = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-							       PPSMC_MSG_SetFanPwmMax,
-							       adev->pm.dpm.fan.default_max_fan_pwm);
-		if (ret != PPSMC_Result_OK)
-			return -EINVAL;
-	} else {
-		ret = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-							       PPSMC_StartFanControl,
-							       FAN_CONTROL_TABLE);
-		if (ret != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	pi->fan_is_controlled_by_smc = true;
-	return 0;
-}
-
-
-static int ci_fan_ctrl_stop_smc_fan_control(struct amdgpu_device *adev)
-{
-	PPSMC_Result ret;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	ret = amdgpu_ci_send_msg_to_smc(adev, PPSMC_StopFanControl);
-	if (ret == PPSMC_Result_OK) {
-		pi->fan_is_controlled_by_smc = false;
-		return 0;
-	} else {
-		return -EINVAL;
-	}
-}
-
-static int ci_dpm_get_fan_speed_percent(void *handle,
-					u32 *speed)
-{
-	u32 duty, duty100;
-	u64 tmp64;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (adev->pm.no_fan)
-		return -ENOENT;
-
-	duty100 = (RREG32_SMC(ixCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK)
-		>> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
-	duty = (RREG32_SMC(ixCG_THERMAL_STATUS) & CG_THERMAL_STATUS__FDO_PWM_DUTY_MASK)
-		>> CG_THERMAL_STATUS__FDO_PWM_DUTY__SHIFT;
-
-	if (duty100 == 0)
-		return -EINVAL;
-
-	tmp64 = (u64)duty * 100;
-	do_div(tmp64, duty100);
-	*speed = (u32)tmp64;
-
-	if (*speed > 100)
-		*speed = 100;
-
-	return 0;
-}
-
-static int ci_dpm_set_fan_speed_percent(void *handle,
-					u32 speed)
-{
-	u32 tmp;
-	u32 duty, duty100;
-	u64 tmp64;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (adev->pm.no_fan)
-		return -ENOENT;
-
-	if (pi->fan_is_controlled_by_smc)
-		return -EINVAL;
-
-	if (speed > 100)
-		return -EINVAL;
-
-	duty100 = (RREG32_SMC(ixCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK)
-		>> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
-
-	if (duty100 == 0)
-		return -EINVAL;
-
-	tmp64 = (u64)speed * duty100;
-	do_div(tmp64, 100);
-	duty = (u32)tmp64;
-
-	tmp = RREG32_SMC(ixCG_FDO_CTRL0) & ~CG_FDO_CTRL0__FDO_STATIC_DUTY_MASK;
-	tmp |= duty << CG_FDO_CTRL0__FDO_STATIC_DUTY__SHIFT;
-	WREG32_SMC(ixCG_FDO_CTRL0, tmp);
-
-	return 0;
-}
-
-static void ci_dpm_set_fan_control_mode(void *handle, u32 mode)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	switch (mode) {
-	case AMD_FAN_CTRL_NONE:
-		if (adev->pm.dpm.fan.ucode_fan_control)
-			ci_fan_ctrl_stop_smc_fan_control(adev);
-		ci_dpm_set_fan_speed_percent(adev, 100);
-		break;
-	case AMD_FAN_CTRL_MANUAL:
-		if (adev->pm.dpm.fan.ucode_fan_control)
-			ci_fan_ctrl_stop_smc_fan_control(adev);
-		break;
-	case AMD_FAN_CTRL_AUTO:
-		if (adev->pm.dpm.fan.ucode_fan_control)
-			ci_thermal_start_smc_fan_control(adev);
-		break;
-	default:
-		break;
-	}
-}
-
-static u32 ci_dpm_get_fan_control_mode(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (pi->fan_is_controlled_by_smc)
-		return AMD_FAN_CTRL_AUTO;
-	else
-		return AMD_FAN_CTRL_MANUAL;
-}
-
-#if 0
-static int ci_fan_ctrl_get_fan_speed_rpm(struct amdgpu_device *adev,
-					 u32 *speed)
-{
-	u32 tach_period;
-	u32 xclk = amdgpu_asic_get_xclk(adev);
-
-	if (adev->pm.no_fan)
-		return -ENOENT;
-
-	if (adev->pm.fan_pulses_per_revolution == 0)
-		return -ENOENT;
-
-	tach_period = (RREG32_SMC(ixCG_TACH_STATUS) & CG_TACH_STATUS__TACH_PERIOD_MASK)
-		>> CG_TACH_STATUS__TACH_PERIOD__SHIFT;
-	if (tach_period == 0)
-		return -ENOENT;
-
-	*speed = 60 * xclk * 10000 / tach_period;
-
-	return 0;
-}
-
-static int ci_fan_ctrl_set_fan_speed_rpm(struct amdgpu_device *adev,
-					 u32 speed)
-{
-	u32 tach_period, tmp;
-	u32 xclk = amdgpu_asic_get_xclk(adev);
-
-	if (adev->pm.no_fan)
-		return -ENOENT;
-
-	if (adev->pm.fan_pulses_per_revolution == 0)
-		return -ENOENT;
-
-	if ((speed < adev->pm.fan_min_rpm) ||
-	    (speed > adev->pm.fan_max_rpm))
-		return -EINVAL;
-
-	if (adev->pm.dpm.fan.ucode_fan_control)
-		ci_fan_ctrl_stop_smc_fan_control(adev);
-
-	tach_period = 60 * xclk * 10000 / (8 * speed);
-	tmp = RREG32_SMC(ixCG_TACH_CTRL) & ~CG_TACH_CTRL__TARGET_PERIOD_MASK;
-	tmp |= tach_period << CG_TACH_CTRL__TARGET_PERIOD__SHIFT;
-	WREG32_SMC(CG_TACH_CTRL, tmp);
-
-	ci_fan_ctrl_set_static_mode(adev, FDO_PWM_MODE_STATIC_RPM);
-
-	return 0;
-}
-#endif
-
-static void ci_fan_ctrl_set_default_mode(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-
-	if (!pi->fan_ctrl_is_in_default_mode) {
-		tmp = RREG32_SMC(ixCG_FDO_CTRL2) & ~CG_FDO_CTRL2__FDO_PWM_MODE_MASK;
-		tmp |= pi->fan_ctrl_default_mode << CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
-		WREG32_SMC(ixCG_FDO_CTRL2, tmp);
-
-		tmp = RREG32_SMC(ixCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TMIN_MASK;
-		tmp |= pi->t_min << CG_FDO_CTRL2__TMIN__SHIFT;
-		WREG32_SMC(ixCG_FDO_CTRL2, tmp);
-		pi->fan_ctrl_is_in_default_mode = true;
-	}
-}
-
-static void ci_thermal_start_smc_fan_control(struct amdgpu_device *adev)
-{
-	if (adev->pm.dpm.fan.ucode_fan_control) {
-		ci_fan_ctrl_start_smc_fan_control(adev);
-		ci_fan_ctrl_set_static_mode(adev, FDO_PWM_MODE_STATIC);
-	}
-}
-
-static void ci_thermal_initialize(struct amdgpu_device *adev)
-{
-	u32 tmp;
-
-	if (adev->pm.fan_pulses_per_revolution) {
-		tmp = RREG32_SMC(ixCG_TACH_CTRL) & ~CG_TACH_CTRL__EDGE_PER_REV_MASK;
-		tmp |= (adev->pm.fan_pulses_per_revolution - 1)
-			<< CG_TACH_CTRL__EDGE_PER_REV__SHIFT;
-		WREG32_SMC(ixCG_TACH_CTRL, tmp);
-	}
-
-	tmp = RREG32_SMC(ixCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TACH_PWM_RESP_RATE_MASK;
-	tmp |= 0x28 << CG_FDO_CTRL2__TACH_PWM_RESP_RATE__SHIFT;
-	WREG32_SMC(ixCG_FDO_CTRL2, tmp);
-}
-
-static int ci_thermal_start_thermal_controller(struct amdgpu_device *adev)
-{
-	int ret;
-
-	ci_thermal_initialize(adev);
-	ret = ci_thermal_set_temperature_range(adev, CISLANDS_TEMP_RANGE_MIN, CISLANDS_TEMP_RANGE_MAX);
-	if (ret)
-		return ret;
-	ret = ci_thermal_enable_alert(adev, true);
-	if (ret)
-		return ret;
-	if (adev->pm.dpm.fan.ucode_fan_control) {
-		ret = ci_thermal_setup_fan_table(adev);
-		if (ret)
-			return ret;
-		ci_thermal_start_smc_fan_control(adev);
-	}
-
-	return 0;
-}
-
-static void ci_thermal_stop_thermal_controller(struct amdgpu_device *adev)
-{
-	if (!adev->pm.no_fan)
-		ci_fan_ctrl_set_default_mode(adev);
-}
-
-static int ci_read_smc_soft_register(struct amdgpu_device *adev,
-				     u16 reg_offset, u32 *value)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	return amdgpu_ci_read_smc_sram_dword(adev,
-				      pi->soft_regs_start + reg_offset,
-				      value, pi->sram_end);
-}
-
-static int ci_write_smc_soft_register(struct amdgpu_device *adev,
-				      u16 reg_offset, u32 value)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	return amdgpu_ci_write_smc_sram_dword(adev,
-				       pi->soft_regs_start + reg_offset,
-				       value, pi->sram_end);
-}
-
-static void ci_init_fps_limits(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	SMU7_Discrete_DpmTable *table = &pi->smc_state_table;
-
-	if (pi->caps_fps) {
-		u16 tmp;
-
-		tmp = 45;
-		table->FpsHighT = cpu_to_be16(tmp);
-
-		tmp = 30;
-		table->FpsLowT = cpu_to_be16(tmp);
-	}
-}
-
-static int ci_update_sclk_t(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret = 0;
-	u32 low_sclk_interrupt_t = 0;
-
-	if (pi->caps_sclk_throttle_low_notification) {
-		low_sclk_interrupt_t = cpu_to_be32(pi->low_sclk_interrupt_t);
-
-		ret = amdgpu_ci_copy_bytes_to_smc(adev,
-					   pi->dpm_table_start +
-					   offsetof(SMU7_Discrete_DpmTable, LowSclkInterruptT),
-					   (u8 *)&low_sclk_interrupt_t,
-					   sizeof(u32), pi->sram_end);
-
-	}
-
-	return ret;
-}
-
-static void ci_get_leakage_voltages(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u16 leakage_id, virtual_voltage_id;
-	u16 vddc, vddci;
-	int i;
-
-	pi->vddc_leakage.count = 0;
-	pi->vddci_leakage.count = 0;
-
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) {
-		for (i = 0; i < CISLANDS_MAX_LEAKAGE_COUNT; i++) {
-			virtual_voltage_id = ATOM_VIRTUAL_VOLTAGE_ID0 + i;
-			if (amdgpu_atombios_get_voltage_evv(adev, virtual_voltage_id, &vddc) != 0)
-				continue;
-			if (vddc != 0 && vddc != virtual_voltage_id) {
-				pi->vddc_leakage.actual_voltage[pi->vddc_leakage.count] = vddc;
-				pi->vddc_leakage.leakage_id[pi->vddc_leakage.count] = virtual_voltage_id;
-				pi->vddc_leakage.count++;
-			}
-		}
-	} else if (amdgpu_atombios_get_leakage_id_from_vbios(adev, &leakage_id) == 0) {
-		for (i = 0; i < CISLANDS_MAX_LEAKAGE_COUNT; i++) {
-			virtual_voltage_id = ATOM_VIRTUAL_VOLTAGE_ID0 + i;
-			if (amdgpu_atombios_get_leakage_vddc_based_on_leakage_params(adev, &vddc, &vddci,
-										     virtual_voltage_id,
-										     leakage_id) == 0) {
-				if (vddc != 0 && vddc != virtual_voltage_id) {
-					pi->vddc_leakage.actual_voltage[pi->vddc_leakage.count] = vddc;
-					pi->vddc_leakage.leakage_id[pi->vddc_leakage.count] = virtual_voltage_id;
-					pi->vddc_leakage.count++;
-				}
-				if (vddci != 0 && vddci != virtual_voltage_id) {
-					pi->vddci_leakage.actual_voltage[pi->vddci_leakage.count] = vddci;
-					pi->vddci_leakage.leakage_id[pi->vddci_leakage.count] = virtual_voltage_id;
-					pi->vddci_leakage.count++;
-				}
-			}
-		}
-	}
-}
-
-static void ci_set_dpm_event_sources(struct amdgpu_device *adev, u32 sources)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	bool want_thermal_protection;
-	enum amdgpu_dpm_event_src dpm_event_src;
-	u32 tmp;
-
-	switch (sources) {
-	case 0:
-	default:
-		want_thermal_protection = false;
-		break;
-	case (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL):
-		want_thermal_protection = true;
-		dpm_event_src = AMDGPU_DPM_EVENT_SRC_DIGITAL;
-		break;
-	case (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
-		want_thermal_protection = true;
-		dpm_event_src = AMDGPU_DPM_EVENT_SRC_EXTERNAL;
-		break;
-	case ((1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
-	      (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL)):
-		want_thermal_protection = true;
-		dpm_event_src = AMDGPU_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL;
-		break;
-	}
-
-	if (want_thermal_protection) {
-#if 0
-		/* XXX: need to figure out how to handle this properly */
-		tmp = RREG32_SMC(ixCG_THERMAL_CTRL);
-		tmp &= DPM_EVENT_SRC_MASK;
-		tmp |= DPM_EVENT_SRC(dpm_event_src);
-		WREG32_SMC(ixCG_THERMAL_CTRL, tmp);
-#endif
-
-		tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-		if (pi->thermal_protection)
-			tmp &= ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK;
-		else
-			tmp |= GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK;
-		WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-	} else {
-		tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-		tmp |= GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK;
-		WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-	}
-}
-
-static void ci_enable_auto_throttle_source(struct amdgpu_device *adev,
-					   enum amdgpu_dpm_auto_throttle_src source,
-					   bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (enable) {
-		if (!(pi->active_auto_throttle_sources & (1 << source))) {
-			pi->active_auto_throttle_sources |= 1 << source;
-			ci_set_dpm_event_sources(adev, pi->active_auto_throttle_sources);
-		}
-	} else {
-		if (pi->active_auto_throttle_sources & (1 << source)) {
-			pi->active_auto_throttle_sources &= ~(1 << source);
-			ci_set_dpm_event_sources(adev, pi->active_auto_throttle_sources);
-		}
-	}
-}
-
-static void ci_enable_vr_hot_gpio_interrupt(struct amdgpu_device *adev)
-{
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_REGULATOR_HOT)
-		amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_EnableVRHotGPIOInterrupt);
-}
-
-static int ci_unfreeze_sclk_mclk_dpm(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-
-	if (!pi->need_update_smu7_dpm_table)
-		return 0;
-
-	if ((!pi->sclk_dpm_key_disabled) &&
-	    (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK))) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_SCLKDPM_UnfreezeLevel);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	if ((!pi->mclk_dpm_key_disabled) &&
-	    (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MCLKDPM_UnfreezeLevel);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	pi->need_update_smu7_dpm_table = 0;
-	return 0;
-}
-
-static int ci_enable_sclk_mclk_dpm(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-
-	if (enable) {
-		if (!pi->sclk_dpm_key_disabled) {
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DPM_Enable);
-			if (smc_result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-
-		if (!pi->mclk_dpm_key_disabled) {
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MCLKDPM_Enable);
-			if (smc_result != PPSMC_Result_OK)
-				return -EINVAL;
-
-			WREG32_P(mmMC_SEQ_CNTL_3, MC_SEQ_CNTL_3__CAC_EN_MASK,
-					~MC_SEQ_CNTL_3__CAC_EN_MASK);
-
-			WREG32_SMC(ixLCAC_MC0_CNTL, 0x05);
-			WREG32_SMC(ixLCAC_MC1_CNTL, 0x05);
-			WREG32_SMC(ixLCAC_CPL_CNTL, 0x100005);
-
-			udelay(10);
-
-			WREG32_SMC(ixLCAC_MC0_CNTL, 0x400005);
-			WREG32_SMC(ixLCAC_MC1_CNTL, 0x400005);
-			WREG32_SMC(ixLCAC_CPL_CNTL, 0x500005);
-		}
-	} else {
-		if (!pi->sclk_dpm_key_disabled) {
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DPM_Disable);
-			if (smc_result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-
-		if (!pi->mclk_dpm_key_disabled) {
-			smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MCLKDPM_Disable);
-			if (smc_result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-static int ci_start_dpm(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-	int ret;
-	u32 tmp;
-
-	tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-	tmp |= GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK;
-	WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-
-	tmp = RREG32_SMC(ixSCLK_PWRMGT_CNTL);
-	tmp |= SCLK_PWRMGT_CNTL__DYNAMIC_PM_EN_MASK;
-	WREG32_SMC(ixSCLK_PWRMGT_CNTL, tmp);
-
-	ci_write_smc_soft_register(adev, offsetof(SMU7_SoftRegisters, VoltageChangeTimeout), 0x1000);
-
-	WREG32_P(mmBIF_LNCNT_RESET, 0, ~BIF_LNCNT_RESET__RESET_LNCNT_EN_MASK);
-
-	smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_Voltage_Cntl_Enable);
-	if (smc_result != PPSMC_Result_OK)
-		return -EINVAL;
-
-	ret = ci_enable_sclk_mclk_dpm(adev, true);
-	if (ret)
-		return ret;
-
-	if (!pi->pcie_dpm_key_disabled) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_PCIeDPM_Enable);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_freeze_sclk_mclk_dpm(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-
-	if (!pi->need_update_smu7_dpm_table)
-		return 0;
-
-	if ((!pi->sclk_dpm_key_disabled) &&
-	    (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK))) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_SCLKDPM_FreezeLevel);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	if ((!pi->mclk_dpm_key_disabled) &&
-	    (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MCLKDPM_FreezeLevel);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_stop_dpm(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result smc_result;
-	int ret;
-	u32 tmp;
-
-	tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-	tmp &= ~GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK;
-	WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-
-	tmp = RREG32_SMC(ixSCLK_PWRMGT_CNTL);
-	tmp &= ~SCLK_PWRMGT_CNTL__DYNAMIC_PM_EN_MASK;
-	WREG32_SMC(ixSCLK_PWRMGT_CNTL, tmp);
-
-	if (!pi->pcie_dpm_key_disabled) {
-		smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_PCIeDPM_Disable);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	ret = ci_enable_sclk_mclk_dpm(adev, false);
-	if (ret)
-		return ret;
-
-	smc_result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_Voltage_Cntl_Disable);
-	if (smc_result != PPSMC_Result_OK)
-		return -EINVAL;
-
-	return 0;
-}
-
-static void ci_enable_sclk_control(struct amdgpu_device *adev, bool enable)
-{
-	u32 tmp = RREG32_SMC(ixSCLK_PWRMGT_CNTL);
-
-	if (enable)
-		tmp &= ~SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK;
-	else
-		tmp |= SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK;
-	WREG32_SMC(ixSCLK_PWRMGT_CNTL, tmp);
-}
-
-#if 0
-static int ci_notify_hw_of_power_source(struct amdgpu_device *adev,
-					bool ac_power)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_cac_tdp_table *cac_tdp_table =
-		adev->pm.dpm.dyn_state.cac_tdp_table;
-	u32 power_limit;
-
-	if (ac_power)
-		power_limit = (u32)(cac_tdp_table->maximum_power_delivery_limit * 256);
-	else
-		power_limit = (u32)(cac_tdp_table->battery_power_limit * 256);
-
-	ci_set_power_limit(adev, power_limit);
-
-	if (pi->caps_automatic_dc_transition) {
-		if (ac_power)
-			amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_RunningOnAC);
-		else
-			amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_Remove_DC_Clamp);
-	}
-
-	return 0;
-}
-#endif
-
-static PPSMC_Result amdgpu_ci_send_msg_to_smc_with_parameter(struct amdgpu_device *adev,
-						      PPSMC_Msg msg, u32 parameter)
-{
-	WREG32(mmSMC_MSG_ARG_0, parameter);
-	return amdgpu_ci_send_msg_to_smc(adev, msg);
-}
-
-static PPSMC_Result amdgpu_ci_send_msg_to_smc_return_parameter(struct amdgpu_device *adev,
-							PPSMC_Msg msg, u32 *parameter)
-{
-	PPSMC_Result smc_result;
-
-	smc_result = amdgpu_ci_send_msg_to_smc(adev, msg);
-
-	if ((smc_result == PPSMC_Result_OK) && parameter)
-		*parameter = RREG32(mmSMC_MSG_ARG_0);
-
-	return smc_result;
-}
-
-static int ci_dpm_force_state_sclk(struct amdgpu_device *adev, u32 n)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (!pi->sclk_dpm_key_disabled) {
-		PPSMC_Result smc_result =
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_SCLKDPM_SetEnabledMask, 1 << n);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_dpm_force_state_mclk(struct amdgpu_device *adev, u32 n)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (!pi->mclk_dpm_key_disabled) {
-		PPSMC_Result smc_result =
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_MCLKDPM_SetEnabledMask, 1 << n);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_dpm_force_state_pcie(struct amdgpu_device *adev, u32 n)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (!pi->pcie_dpm_key_disabled) {
-		PPSMC_Result smc_result =
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_PCIeDPM_ForceLevel, n);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_set_power_limit(struct amdgpu_device *adev, u32 n)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit) {
-		PPSMC_Result smc_result =
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_PkgPwrSetLimit, n);
-		if (smc_result != PPSMC_Result_OK)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_set_overdrive_target_tdp(struct amdgpu_device *adev,
-				       u32 target_tdp)
-{
-	PPSMC_Result smc_result =
-		amdgpu_ci_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_OverDriveSetTargetTdp, target_tdp);
-	if (smc_result != PPSMC_Result_OK)
-		return -EINVAL;
-	return 0;
-}
-
-#if 0
-static int ci_set_boot_state(struct amdgpu_device *adev)
-{
-	return ci_enable_sclk_mclk_dpm(adev, false);
-}
-#endif
-
-static u32 ci_get_average_sclk_freq(struct amdgpu_device *adev)
-{
-	u32 sclk_freq;
-	PPSMC_Result smc_result =
-		amdgpu_ci_send_msg_to_smc_return_parameter(adev,
-						    PPSMC_MSG_API_GetSclkFrequency,
-						    &sclk_freq);
-	if (smc_result != PPSMC_Result_OK)
-		sclk_freq = 0;
-
-	return sclk_freq;
-}
-
-static u32 ci_get_average_mclk_freq(struct amdgpu_device *adev)
-{
-	u32 mclk_freq;
-	PPSMC_Result smc_result =
-		amdgpu_ci_send_msg_to_smc_return_parameter(adev,
-						    PPSMC_MSG_API_GetMclkFrequency,
-						    &mclk_freq);
-	if (smc_result != PPSMC_Result_OK)
-		mclk_freq = 0;
-
-	return mclk_freq;
-}
-
-static void ci_dpm_start_smc(struct amdgpu_device *adev)
-{
-	int i;
-
-	amdgpu_ci_program_jump_on_start(adev);
-	amdgpu_ci_start_smc_clock(adev);
-	amdgpu_ci_start_smc(adev);
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32_SMC(ixFIRMWARE_FLAGS) & FIRMWARE_FLAGS__INTERRUPTS_ENABLED_MASK)
-			break;
-	}
-}
-
-static void ci_dpm_stop_smc(struct amdgpu_device *adev)
-{
-	amdgpu_ci_reset_smc(adev);
-	amdgpu_ci_stop_smc_clock(adev);
-}
-
-static int ci_process_firmware_header(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-	int ret;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, DpmTable),
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	pi->dpm_table_start = tmp;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, SoftRegisters),
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	pi->soft_regs_start = tmp;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, mcRegisterTable),
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	pi->mc_reg_table_start = tmp;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, FanTable),
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	pi->fan_table_start = tmp;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev,
-				     SMU7_FIRMWARE_HEADER_LOCATION +
-				     offsetof(SMU7_Firmware_Header, mcArbDramTimingTable),
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	pi->arb_table_start = tmp;
-
-	return 0;
-}
-
-static void ci_read_clock_registers(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	pi->clock_registers.cg_spll_func_cntl =
-		RREG32_SMC(ixCG_SPLL_FUNC_CNTL);
-	pi->clock_registers.cg_spll_func_cntl_2 =
-		RREG32_SMC(ixCG_SPLL_FUNC_CNTL_2);
-	pi->clock_registers.cg_spll_func_cntl_3 =
-		RREG32_SMC(ixCG_SPLL_FUNC_CNTL_3);
-	pi->clock_registers.cg_spll_func_cntl_4 =
-		RREG32_SMC(ixCG_SPLL_FUNC_CNTL_4);
-	pi->clock_registers.cg_spll_spread_spectrum =
-		RREG32_SMC(ixCG_SPLL_SPREAD_SPECTRUM);
-	pi->clock_registers.cg_spll_spread_spectrum_2 =
-		RREG32_SMC(ixCG_SPLL_SPREAD_SPECTRUM_2);
-	pi->clock_registers.dll_cntl = RREG32(mmDLL_CNTL);
-	pi->clock_registers.mclk_pwrmgt_cntl = RREG32(mmMCLK_PWRMGT_CNTL);
-	pi->clock_registers.mpll_ad_func_cntl = RREG32(mmMPLL_AD_FUNC_CNTL);
-	pi->clock_registers.mpll_dq_func_cntl = RREG32(mmMPLL_DQ_FUNC_CNTL);
-	pi->clock_registers.mpll_func_cntl = RREG32(mmMPLL_FUNC_CNTL);
-	pi->clock_registers.mpll_func_cntl_1 = RREG32(mmMPLL_FUNC_CNTL_1);
-	pi->clock_registers.mpll_func_cntl_2 = RREG32(mmMPLL_FUNC_CNTL_2);
-	pi->clock_registers.mpll_ss1 = RREG32(mmMPLL_SS1);
-	pi->clock_registers.mpll_ss2 = RREG32(mmMPLL_SS2);
-}
-
-static void ci_init_sclk_t(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	pi->low_sclk_interrupt_t = 0;
-}
-
-static void ci_enable_thermal_protection(struct amdgpu_device *adev,
-					 bool enable)
-{
-	u32 tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-
-	if (enable)
-		tmp &= ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK;
-	else
-		tmp |= GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK;
-	WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-}
-
-static void ci_enable_acpi_power_management(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-
-	tmp |= GENERAL_PWRMGT__STATIC_PM_EN_MASK;
-
-	WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-}
-
-#if 0
-static int ci_enter_ulp_state(struct amdgpu_device *adev)
-{
-
-	WREG32(mmSMC_MESSAGE_0, PPSMC_MSG_SwitchToMinimumPower);
-
-	udelay(25000);
-
-	return 0;
-}
-
-static int ci_exit_ulp_state(struct amdgpu_device *adev)
-{
-	int i;
-
-	WREG32(mmSMC_MESSAGE_0, PPSMC_MSG_ResumeFromMinimumPower);
-
-	udelay(7000);
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32(mmSMC_RESP_0) == 1)
-			break;
-		udelay(1000);
-	}
-
-	return 0;
-}
-#endif
-
-static int ci_notify_smc_display_change(struct amdgpu_device *adev,
-					bool has_display)
-{
-	PPSMC_Msg msg = has_display ? PPSMC_MSG_HasDisplay : PPSMC_MSG_NoDisplay;
-
-	return (amdgpu_ci_send_msg_to_smc(adev, msg) == PPSMC_Result_OK) ?  0 : -EINVAL;
-}
-
-static int ci_enable_ds_master_switch(struct amdgpu_device *adev,
-				      bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (enable) {
-		if (pi->caps_sclk_ds) {
-			if (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MASTER_DeepSleep_ON) != PPSMC_Result_OK)
-				return -EINVAL;
-		} else {
-			if (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MASTER_DeepSleep_OFF) != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	} else {
-		if (pi->caps_sclk_ds) {
-			if (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_MASTER_DeepSleep_OFF) != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-static void ci_program_display_gap(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixCG_DISPLAY_GAP_CNTL);
-	u32 pre_vbi_time_in_us;
-	u32 frame_time_in_us;
-	u32 ref_clock = adev->clock.spll.reference_freq;
-	u32 refresh_rate = amdgpu_dpm_get_vrefresh(adev);
-	u32 vblank_time = amdgpu_dpm_get_vblank_time(adev);
-
-	tmp &= ~CG_DISPLAY_GAP_CNTL__DISP_GAP_MASK;
-	if (adev->pm.dpm.new_active_crtc_count > 0)
-		tmp |= (AMDGPU_PM_DISPLAY_GAP_VBLANK_OR_WM << CG_DISPLAY_GAP_CNTL__DISP_GAP__SHIFT);
-	else
-		tmp |= (AMDGPU_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP_GAP__SHIFT);
-	WREG32_SMC(ixCG_DISPLAY_GAP_CNTL, tmp);
-
-	if (refresh_rate == 0)
-		refresh_rate = 60;
-	if (vblank_time == 0xffffffff)
-		vblank_time = 500;
-	frame_time_in_us = 1000000 / refresh_rate;
-	pre_vbi_time_in_us =
-		frame_time_in_us - 200 - vblank_time;
-	tmp = pre_vbi_time_in_us * (ref_clock / 100);
-
-	WREG32_SMC(ixCG_DISPLAY_GAP_CNTL2, tmp);
-	ci_write_smc_soft_register(adev, offsetof(SMU7_SoftRegisters, PreVBlankGap), 0x64);
-	ci_write_smc_soft_register(adev, offsetof(SMU7_SoftRegisters, VBlankTimeout), (frame_time_in_us - pre_vbi_time_in_us));
-
-
-	ci_notify_smc_display_change(adev, (adev->pm.dpm.new_active_crtc_count == 1));
-
-}
-
-static void ci_enable_spread_spectrum(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-
-	if (enable) {
-		if (pi->caps_sclk_ss_support) {
-			tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-			tmp |= GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK;
-			WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-		}
-	} else {
-		tmp = RREG32_SMC(ixCG_SPLL_SPREAD_SPECTRUM);
-		tmp &= ~CG_SPLL_SPREAD_SPECTRUM__SSEN_MASK;
-		WREG32_SMC(ixCG_SPLL_SPREAD_SPECTRUM, tmp);
-
-		tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-		tmp &= ~GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK;
-		WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-	}
-}
-
-static void ci_program_sstp(struct amdgpu_device *adev)
-{
-	WREG32_SMC(ixCG_STATIC_SCREEN_PARAMETER,
-	((CISLANDS_SSTU_DFLT << CG_STATIC_SCREEN_PARAMETER__STATIC_SCREEN_THRESHOLD_UNIT__SHIFT) |
-	 (CISLANDS_SST_DFLT << CG_STATIC_SCREEN_PARAMETER__STATIC_SCREEN_THRESHOLD__SHIFT)));
-}
-
-static void ci_enable_display_gap(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixCG_DISPLAY_GAP_CNTL);
-
-	tmp &= ~(CG_DISPLAY_GAP_CNTL__DISP_GAP_MASK |
-			CG_DISPLAY_GAP_CNTL__DISP_GAP_MCHG_MASK);
-	tmp |= ((AMDGPU_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP_GAP__SHIFT) |
-		(AMDGPU_PM_DISPLAY_GAP_VBLANK << CG_DISPLAY_GAP_CNTL__DISP_GAP_MCHG__SHIFT));
-
-	WREG32_SMC(ixCG_DISPLAY_GAP_CNTL, tmp);
-}
-
-static void ci_program_vc(struct amdgpu_device *adev)
-{
-	u32 tmp;
-
-	tmp = RREG32_SMC(ixSCLK_PWRMGT_CNTL);
-	tmp &= ~(SCLK_PWRMGT_CNTL__RESET_SCLK_CNT_MASK | SCLK_PWRMGT_CNTL__RESET_BUSY_CNT_MASK);
-	WREG32_SMC(ixSCLK_PWRMGT_CNTL, tmp);
-
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_0, CISLANDS_VRC_DFLT0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_1, CISLANDS_VRC_DFLT1);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_2, CISLANDS_VRC_DFLT2);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_3, CISLANDS_VRC_DFLT3);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_4, CISLANDS_VRC_DFLT4);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_5, CISLANDS_VRC_DFLT5);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_6, CISLANDS_VRC_DFLT6);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_7, CISLANDS_VRC_DFLT7);
-}
-
-static void ci_clear_vc(struct amdgpu_device *adev)
-{
-	u32 tmp;
-
-	tmp = RREG32_SMC(ixSCLK_PWRMGT_CNTL);
-	tmp |= (SCLK_PWRMGT_CNTL__RESET_SCLK_CNT_MASK | SCLK_PWRMGT_CNTL__RESET_BUSY_CNT_MASK);
-	WREG32_SMC(ixSCLK_PWRMGT_CNTL, tmp);
-
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_0, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_1, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_2, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_3, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_4, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_5, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_6, 0);
-	WREG32_SMC(ixCG_FREQ_TRAN_VOTING_7, 0);
-}
-
-static int ci_upload_firmware(struct amdgpu_device *adev)
-{
-	int i, ret;
-
-	if (amdgpu_ci_is_smc_running(adev)) {
-		DRM_INFO("smc is running, no need to load smc firmware\n");
-		return 0;
-	}
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32_SMC(ixRCU_UC_EVENTS) & RCU_UC_EVENTS__boot_seq_done_MASK)
-			break;
-	}
-	WREG32_SMC(ixSMC_SYSCON_MISC_CNTL, 1);
-
-	amdgpu_ci_stop_smc_clock(adev);
-	amdgpu_ci_reset_smc(adev);
-
-	ret = amdgpu_ci_load_smc_ucode(adev, SMC_RAM_END);
-
-	return ret;
-
-}
-
-static int ci_get_svi2_voltage_table(struct amdgpu_device *adev,
-				     struct amdgpu_clock_voltage_dependency_table *voltage_dependency_table,
-				     struct atom_voltage_table *voltage_table)
-{
-	u32 i;
-
-	if (voltage_dependency_table == NULL)
-		return -EINVAL;
-
-	voltage_table->mask_low = 0;
-	voltage_table->phase_delay = 0;
-
-	voltage_table->count = voltage_dependency_table->count;
-	for (i = 0; i < voltage_table->count; i++) {
-		voltage_table->entries[i].value = voltage_dependency_table->entries[i].v;
-		voltage_table->entries[i].smio_low = 0;
-	}
-
-	return 0;
-}
-
-static int ci_construct_voltage_tables(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-
-	if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
-		ret = amdgpu_atombios_get_voltage_table(adev, VOLTAGE_TYPE_VDDC,
-							VOLTAGE_OBJ_GPIO_LUT,
-							&pi->vddc_voltage_table);
-		if (ret)
-			return ret;
-	} else if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
-		ret = ci_get_svi2_voltage_table(adev,
-						&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
-						&pi->vddc_voltage_table);
-		if (ret)
-			return ret;
-	}
-
-	if (pi->vddc_voltage_table.count > SMU7_MAX_LEVELS_VDDC)
-		ci_trim_voltage_table_to_fit_state_table(adev, SMU7_MAX_LEVELS_VDDC,
-							 &pi->vddc_voltage_table);
-
-	if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
-		ret = amdgpu_atombios_get_voltage_table(adev, VOLTAGE_TYPE_VDDCI,
-							VOLTAGE_OBJ_GPIO_LUT,
-							&pi->vddci_voltage_table);
-		if (ret)
-			return ret;
-	} else if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
-		ret = ci_get_svi2_voltage_table(adev,
-						&adev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
-						&pi->vddci_voltage_table);
-		if (ret)
-			return ret;
-	}
-
-	if (pi->vddci_voltage_table.count > SMU7_MAX_LEVELS_VDDCI)
-		ci_trim_voltage_table_to_fit_state_table(adev, SMU7_MAX_LEVELS_VDDCI,
-							 &pi->vddci_voltage_table);
-
-	if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
-		ret = amdgpu_atombios_get_voltage_table(adev, VOLTAGE_TYPE_MVDDC,
-							VOLTAGE_OBJ_GPIO_LUT,
-							&pi->mvdd_voltage_table);
-		if (ret)
-			return ret;
-	} else if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
-		ret = ci_get_svi2_voltage_table(adev,
-						&adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
-						&pi->mvdd_voltage_table);
-		if (ret)
-			return ret;
-	}
-
-	if (pi->mvdd_voltage_table.count > SMU7_MAX_LEVELS_MVDD)
-		ci_trim_voltage_table_to_fit_state_table(adev, SMU7_MAX_LEVELS_MVDD,
-							 &pi->mvdd_voltage_table);
-
-	return 0;
-}
-
-static void ci_populate_smc_voltage_table(struct amdgpu_device *adev,
-					  struct atom_voltage_table_entry *voltage_table,
-					  SMU7_Discrete_VoltageLevel *smc_voltage_table)
-{
-	int ret;
-
-	ret = ci_get_std_voltage_value_sidd(adev, voltage_table,
-					    &smc_voltage_table->StdVoltageHiSidd,
-					    &smc_voltage_table->StdVoltageLoSidd);
-
-	if (ret) {
-		smc_voltage_table->StdVoltageHiSidd = voltage_table->value * VOLTAGE_SCALE;
-		smc_voltage_table->StdVoltageLoSidd = voltage_table->value * VOLTAGE_SCALE;
-	}
-
-	smc_voltage_table->Voltage = cpu_to_be16(voltage_table->value * VOLTAGE_SCALE);
-	smc_voltage_table->StdVoltageHiSidd =
-		cpu_to_be16(smc_voltage_table->StdVoltageHiSidd);
-	smc_voltage_table->StdVoltageLoSidd =
-		cpu_to_be16(smc_voltage_table->StdVoltageLoSidd);
-}
-
-static int ci_populate_smc_vddc_table(struct amdgpu_device *adev,
-				      SMU7_Discrete_DpmTable *table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	unsigned int count;
-
-	table->VddcLevelCount = pi->vddc_voltage_table.count;
-	for (count = 0; count < table->VddcLevelCount; count++) {
-		ci_populate_smc_voltage_table(adev,
-					      &pi->vddc_voltage_table.entries[count],
-					      &table->VddcLevel[count]);
-
-		if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
-			table->VddcLevel[count].Smio |=
-				pi->vddc_voltage_table.entries[count].smio_low;
-		else
-			table->VddcLevel[count].Smio = 0;
-	}
-	table->VddcLevelCount = cpu_to_be32(table->VddcLevelCount);
-
-	return 0;
-}
-
-static int ci_populate_smc_vddci_table(struct amdgpu_device *adev,
-				       SMU7_Discrete_DpmTable *table)
-{
-	unsigned int count;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	table->VddciLevelCount = pi->vddci_voltage_table.count;
-	for (count = 0; count < table->VddciLevelCount; count++) {
-		ci_populate_smc_voltage_table(adev,
-					      &pi->vddci_voltage_table.entries[count],
-					      &table->VddciLevel[count]);
-
-		if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
-			table->VddciLevel[count].Smio |=
-				pi->vddci_voltage_table.entries[count].smio_low;
-		else
-			table->VddciLevel[count].Smio = 0;
-	}
-	table->VddciLevelCount = cpu_to_be32(table->VddciLevelCount);
-
-	return 0;
-}
-
-static int ci_populate_smc_mvdd_table(struct amdgpu_device *adev,
-				      SMU7_Discrete_DpmTable *table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	unsigned int count;
-
-	table->MvddLevelCount = pi->mvdd_voltage_table.count;
-	for (count = 0; count < table->MvddLevelCount; count++) {
-		ci_populate_smc_voltage_table(adev,
-					      &pi->mvdd_voltage_table.entries[count],
-					      &table->MvddLevel[count]);
-
-		if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
-			table->MvddLevel[count].Smio |=
-				pi->mvdd_voltage_table.entries[count].smio_low;
-		else
-			table->MvddLevel[count].Smio = 0;
-	}
-	table->MvddLevelCount = cpu_to_be32(table->MvddLevelCount);
-
-	return 0;
-}
-
-static int ci_populate_smc_voltage_tables(struct amdgpu_device *adev,
-					  SMU7_Discrete_DpmTable *table)
-{
-	int ret;
-
-	ret = ci_populate_smc_vddc_table(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_vddci_table(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_mvdd_table(adev, table);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static int ci_populate_mvdd_value(struct amdgpu_device *adev, u32 mclk,
-				  SMU7_Discrete_VoltageLevel *voltage)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 i = 0;
-
-	if (pi->mvdd_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
-		for (i = 0; i < adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.count; i++) {
-			if (mclk <= adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries[i].clk) {
-				voltage->Voltage = pi->mvdd_voltage_table.entries[i].value;
-				break;
-			}
-		}
-
-		if (i >= adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.count)
-			return -EINVAL;
-	}
-
-	return -EINVAL;
-}
-
-static int ci_get_std_voltage_value_sidd(struct amdgpu_device *adev,
-					 struct atom_voltage_table_entry *voltage_table,
-					 u16 *std_voltage_hi_sidd, u16 *std_voltage_lo_sidd)
-{
-	u16 v_index, idx;
-	bool voltage_found = false;
-	*std_voltage_hi_sidd = voltage_table->value * VOLTAGE_SCALE;
-	*std_voltage_lo_sidd = voltage_table->value * VOLTAGE_SCALE;
-
-	if (adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries == NULL)
-		return -EINVAL;
-
-	if (adev->pm.dpm.dyn_state.cac_leakage_table.entries) {
-		for (v_index = 0; (u32)v_index < adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; v_index++) {
-			if (voltage_table->value ==
-			    adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[v_index].v) {
-				voltage_found = true;
-				if ((u32)v_index < adev->pm.dpm.dyn_state.cac_leakage_table.count)
-					idx = v_index;
-				else
-					idx = adev->pm.dpm.dyn_state.cac_leakage_table.count - 1;
-				*std_voltage_lo_sidd =
-					adev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].vddc * VOLTAGE_SCALE;
-				*std_voltage_hi_sidd =
-					adev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].leakage * VOLTAGE_SCALE;
-				break;
-			}
-		}
-
-		if (!voltage_found) {
-			for (v_index = 0; (u32)v_index < adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; v_index++) {
-				if (voltage_table->value <=
-				    adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[v_index].v) {
-					voltage_found = true;
-					if ((u32)v_index < adev->pm.dpm.dyn_state.cac_leakage_table.count)
-						idx = v_index;
-					else
-						idx = adev->pm.dpm.dyn_state.cac_leakage_table.count - 1;
-					*std_voltage_lo_sidd =
-						adev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].vddc * VOLTAGE_SCALE;
-					*std_voltage_hi_sidd =
-						adev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].leakage * VOLTAGE_SCALE;
-					break;
-				}
-			}
-		}
-	}
-
-	return 0;
-}
-
-static void ci_populate_phase_value_based_on_sclk(struct amdgpu_device *adev,
-						  const struct amdgpu_phase_shedding_limits_table *limits,
-						  u32 sclk,
-						  u32 *phase_shedding)
-{
-	unsigned int i;
-
-	*phase_shedding = 1;
-
-	for (i = 0; i < limits->count; i++) {
-		if (sclk < limits->entries[i].sclk) {
-			*phase_shedding = i;
-			break;
-		}
-	}
-}
-
-static void ci_populate_phase_value_based_on_mclk(struct amdgpu_device *adev,
-						  const struct amdgpu_phase_shedding_limits_table *limits,
-						  u32 mclk,
-						  u32 *phase_shedding)
-{
-	unsigned int i;
-
-	*phase_shedding = 1;
-
-	for (i = 0; i < limits->count; i++) {
-		if (mclk < limits->entries[i].mclk) {
-			*phase_shedding = i;
-			break;
-		}
-	}
-}
-
-static int ci_init_arb_table_index(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-	int ret;
-
-	ret = amdgpu_ci_read_smc_sram_dword(adev, pi->arb_table_start,
-				     &tmp, pi->sram_end);
-	if (ret)
-		return ret;
-
-	tmp &= 0x00FFFFFF;
-	tmp |= MC_CG_ARB_FREQ_F1 << 24;
-
-	return amdgpu_ci_write_smc_sram_dword(adev, pi->arb_table_start,
-				       tmp, pi->sram_end);
-}
-
-static int ci_get_dependency_volt_by_clk(struct amdgpu_device *adev,
-					 struct amdgpu_clock_voltage_dependency_table *allowed_clock_voltage_table,
-					 u32 clock, u32 *voltage)
-{
-	u32 i = 0;
-
-	if (allowed_clock_voltage_table->count == 0)
-		return -EINVAL;
-
-	for (i = 0; i < allowed_clock_voltage_table->count; i++) {
-		if (allowed_clock_voltage_table->entries[i].clk >= clock) {
-			*voltage = allowed_clock_voltage_table->entries[i].v;
-			return 0;
-		}
-	}
-
-	*voltage = allowed_clock_voltage_table->entries[i-1].v;
-
-	return 0;
-}
-
-static u8 ci_get_sleep_divider_id_from_clock(u32 sclk, u32 min_sclk_in_sr)
-{
-	u32 i;
-	u32 tmp;
-	u32 min = max(min_sclk_in_sr, (u32)CISLAND_MINIMUM_ENGINE_CLOCK);
-
-	if (sclk < min)
-		return 0;
-
-	for (i = CISLAND_MAX_DEEPSLEEP_DIVIDER_ID;  ; i--) {
-		tmp = sclk >> i;
-		if (tmp >= min || i == 0)
-			break;
-	}
-
-	return (u8)i;
-}
-
-static int ci_initial_switch_from_arb_f0_to_f1(struct amdgpu_device *adev)
-{
-	return ci_copy_and_switch_arb_sets(adev, MC_CG_ARB_FREQ_F0, MC_CG_ARB_FREQ_F1);
-}
-
-static int ci_reset_to_default(struct amdgpu_device *adev)
-{
-	return (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_ResetToDefaults) == PPSMC_Result_OK) ?
-		0 : -EINVAL;
-}
-
-static int ci_force_switch_to_arb_f0(struct amdgpu_device *adev)
-{
-	u32 tmp;
-
-	tmp = (RREG32_SMC(ixSMC_SCRATCH9) & 0x0000ff00) >> 8;
-
-	if (tmp == MC_CG_ARB_FREQ_F0)
-		return 0;
-
-	return ci_copy_and_switch_arb_sets(adev, tmp, MC_CG_ARB_FREQ_F0);
-}
-
-static void ci_register_patching_mc_arb(struct amdgpu_device *adev,
-					const u32 engine_clock,
-					const u32 memory_clock,
-					u32 *dram_timimg2)
-{
-	bool patch;
-	u32 tmp, tmp2;
-
-	tmp = RREG32(mmMC_SEQ_MISC0);
-	patch = ((tmp & 0x0000f00) == 0x300) ? true : false;
-
-	if (patch &&
-	    ((adev->pdev->device == 0x67B0) ||
-	     (adev->pdev->device == 0x67B1))) {
-		if ((memory_clock > 100000) && (memory_clock <= 125000)) {
-			tmp2 = (((0x31 * engine_clock) / 125000) - 1) & 0xff;
-			*dram_timimg2 &= ~0x00ff0000;
-			*dram_timimg2 |= tmp2 << 16;
-		} else if ((memory_clock > 125000) && (memory_clock <= 137500)) {
-			tmp2 = (((0x36 * engine_clock) / 137500) - 1) & 0xff;
-			*dram_timimg2 &= ~0x00ff0000;
-			*dram_timimg2 |= tmp2 << 16;
-		}
-	}
-}
-
-static int ci_populate_memory_timing_parameters(struct amdgpu_device *adev,
-						u32 sclk,
-						u32 mclk,
-						SMU7_Discrete_MCArbDramTimingTableEntry *arb_regs)
-{
-	u32 dram_timing;
-	u32 dram_timing2;
-	u32 burst_time;
-
-	amdgpu_atombios_set_engine_dram_timings(adev, sclk, mclk);
-
-	dram_timing  = RREG32(mmMC_ARB_DRAM_TIMING);
-	dram_timing2 = RREG32(mmMC_ARB_DRAM_TIMING2);
-	burst_time = RREG32(mmMC_ARB_BURST_TIME) & MC_ARB_BURST_TIME__STATE0_MASK;
-
-	ci_register_patching_mc_arb(adev, sclk, mclk, &dram_timing2);
-
-	arb_regs->McArbDramTiming  = cpu_to_be32(dram_timing);
-	arb_regs->McArbDramTiming2 = cpu_to_be32(dram_timing2);
-	arb_regs->McArbBurstTime = (u8)burst_time;
-
-	return 0;
-}
-
-static int ci_do_program_memory_timing_parameters(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	SMU7_Discrete_MCArbDramTimingTable arb_regs;
-	u32 i, j;
-	int ret =  0;
-
-	memset(&arb_regs, 0, sizeof(SMU7_Discrete_MCArbDramTimingTable));
-
-	for (i = 0; i < pi->dpm_table.sclk_table.count; i++) {
-		for (j = 0; j < pi->dpm_table.mclk_table.count; j++) {
-			ret = ci_populate_memory_timing_parameters(adev,
-								   pi->dpm_table.sclk_table.dpm_levels[i].value,
-								   pi->dpm_table.mclk_table.dpm_levels[j].value,
-								   &arb_regs.entries[i][j]);
-			if (ret)
-				break;
-		}
-	}
-
-	if (ret == 0)
-		ret = amdgpu_ci_copy_bytes_to_smc(adev,
-					   pi->arb_table_start,
-					   (u8 *)&arb_regs,
-					   sizeof(SMU7_Discrete_MCArbDramTimingTable),
-					   pi->sram_end);
-
-	return ret;
-}
-
-static int ci_program_memory_timing_parameters(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (pi->need_update_smu7_dpm_table == 0)
-		return 0;
-
-	return ci_do_program_memory_timing_parameters(adev);
-}
-
-static void ci_populate_smc_initial_state(struct amdgpu_device *adev,
-					  struct amdgpu_ps *amdgpu_boot_state)
-{
-	struct ci_ps *boot_state = ci_get_ps(amdgpu_boot_state);
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 level = 0;
-
-	for (level = 0; level < adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; level++) {
-		if (adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[level].clk >=
-		    boot_state->performance_levels[0].sclk) {
-			pi->smc_state_table.GraphicsBootLevel = level;
-			break;
-		}
-	}
-
-	for (level = 0; level < adev->pm.dpm.dyn_state.vddc_dependency_on_mclk.count; level++) {
-		if (adev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries[level].clk >=
-		    boot_state->performance_levels[0].mclk) {
-			pi->smc_state_table.MemoryBootLevel = level;
-			break;
-		}
-	}
-}
-
-static u32 ci_get_dpm_level_enable_mask_value(struct ci_single_dpm_table *dpm_table)
-{
-	u32 i;
-	u32 mask_value = 0;
-
-	for (i = dpm_table->count; i > 0; i--) {
-		mask_value = mask_value << 1;
-		if (dpm_table->dpm_levels[i-1].enabled)
-			mask_value |= 0x1;
-		else
-			mask_value &= 0xFFFFFFFE;
-	}
-
-	return mask_value;
-}
-
-static void ci_populate_smc_link_level(struct amdgpu_device *adev,
-				       SMU7_Discrete_DpmTable *table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_dpm_table *dpm_table = &pi->dpm_table;
-	u32 i;
-
-	for (i = 0; i < dpm_table->pcie_speed_table.count; i++) {
-		table->LinkLevel[i].PcieGenSpeed =
-			(u8)dpm_table->pcie_speed_table.dpm_levels[i].value;
-		table->LinkLevel[i].PcieLaneCount =
-			amdgpu_encode_pci_lane_width(dpm_table->pcie_speed_table.dpm_levels[i].param1);
-		table->LinkLevel[i].EnabledForActivity = 1;
-		table->LinkLevel[i].DownT = cpu_to_be32(5);
-		table->LinkLevel[i].UpT = cpu_to_be32(30);
-	}
-
-	pi->smc_state_table.LinkLevelCount = (u8)dpm_table->pcie_speed_table.count;
-	pi->dpm_level_enable_mask.pcie_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&dpm_table->pcie_speed_table);
-}
-
-static int ci_populate_smc_uvd_level(struct amdgpu_device *adev,
-				     SMU7_Discrete_DpmTable *table)
-{
-	u32 count;
-	struct atom_clock_dividers dividers;
-	int ret = -EINVAL;
-
-	table->UvdLevelCount =
-		adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count;
-
-	for (count = 0; count < table->UvdLevelCount; count++) {
-		table->UvdLevel[count].VclkFrequency =
-			adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].vclk;
-		table->UvdLevel[count].DclkFrequency =
-			adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].dclk;
-		table->UvdLevel[count].MinVddc =
-			adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
-		table->UvdLevel[count].MinVddcPhases = 1;
-
-		ret = amdgpu_atombios_get_clock_dividers(adev,
-							 COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
-							 table->UvdLevel[count].VclkFrequency, false, &dividers);
-		if (ret)
-			return ret;
-
-		table->UvdLevel[count].VclkDivider = (u8)dividers.post_divider;
-
-		ret = amdgpu_atombios_get_clock_dividers(adev,
-							 COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
-							 table->UvdLevel[count].DclkFrequency, false, &dividers);
-		if (ret)
-			return ret;
-
-		table->UvdLevel[count].DclkDivider = (u8)dividers.post_divider;
-
-		table->UvdLevel[count].VclkFrequency = cpu_to_be32(table->UvdLevel[count].VclkFrequency);
-		table->UvdLevel[count].DclkFrequency = cpu_to_be32(table->UvdLevel[count].DclkFrequency);
-		table->UvdLevel[count].MinVddc = cpu_to_be16(table->UvdLevel[count].MinVddc);
-	}
-
-	return ret;
-}
-
-static int ci_populate_smc_vce_level(struct amdgpu_device *adev,
-				     SMU7_Discrete_DpmTable *table)
-{
-	u32 count;
-	struct atom_clock_dividers dividers;
-	int ret = -EINVAL;
-
-	table->VceLevelCount =
-		adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count;
-
-	for (count = 0; count < table->VceLevelCount; count++) {
-		table->VceLevel[count].Frequency =
-			adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[count].evclk;
-		table->VceLevel[count].MinVoltage =
-			(u16)adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
-		table->VceLevel[count].MinPhases = 1;
-
-		ret = amdgpu_atombios_get_clock_dividers(adev,
-							 COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
-							 table->VceLevel[count].Frequency, false, &dividers);
-		if (ret)
-			return ret;
-
-		table->VceLevel[count].Divider = (u8)dividers.post_divider;
-
-		table->VceLevel[count].Frequency = cpu_to_be32(table->VceLevel[count].Frequency);
-		table->VceLevel[count].MinVoltage = cpu_to_be16(table->VceLevel[count].MinVoltage);
-	}
-
-	return ret;
-
-}
-
-static int ci_populate_smc_acp_level(struct amdgpu_device *adev,
-				     SMU7_Discrete_DpmTable *table)
-{
-	u32 count;
-	struct atom_clock_dividers dividers;
-	int ret = -EINVAL;
-
-	table->AcpLevelCount = (u8)
-		(adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count);
-
-	for (count = 0; count < table->AcpLevelCount; count++) {
-		table->AcpLevel[count].Frequency =
-			adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[count].clk;
-		table->AcpLevel[count].MinVoltage =
-			adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[count].v;
-		table->AcpLevel[count].MinPhases = 1;
-
-		ret = amdgpu_atombios_get_clock_dividers(adev,
-							 COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
-							 table->AcpLevel[count].Frequency, false, &dividers);
-		if (ret)
-			return ret;
-
-		table->AcpLevel[count].Divider = (u8)dividers.post_divider;
-
-		table->AcpLevel[count].Frequency = cpu_to_be32(table->AcpLevel[count].Frequency);
-		table->AcpLevel[count].MinVoltage = cpu_to_be16(table->AcpLevel[count].MinVoltage);
-	}
-
-	return ret;
-}
-
-static int ci_populate_smc_samu_level(struct amdgpu_device *adev,
-				      SMU7_Discrete_DpmTable *table)
-{
-	u32 count;
-	struct atom_clock_dividers dividers;
-	int ret = -EINVAL;
-
-	table->SamuLevelCount =
-		adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count;
-
-	for (count = 0; count < table->SamuLevelCount; count++) {
-		table->SamuLevel[count].Frequency =
-			adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[count].clk;
-		table->SamuLevel[count].MinVoltage =
-			adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
-		table->SamuLevel[count].MinPhases = 1;
-
-		ret = amdgpu_atombios_get_clock_dividers(adev,
-							 COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
-							 table->SamuLevel[count].Frequency, false, &dividers);
-		if (ret)
-			return ret;
-
-		table->SamuLevel[count].Divider = (u8)dividers.post_divider;
-
-		table->SamuLevel[count].Frequency = cpu_to_be32(table->SamuLevel[count].Frequency);
-		table->SamuLevel[count].MinVoltage = cpu_to_be16(table->SamuLevel[count].MinVoltage);
-	}
-
-	return ret;
-}
-
-static int ci_calculate_mclk_params(struct amdgpu_device *adev,
-				    u32 memory_clock,
-				    SMU7_Discrete_MemoryLevel *mclk,
-				    bool strobe_mode,
-				    bool dll_state_on)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32  dll_cntl = pi->clock_registers.dll_cntl;
-	u32  mclk_pwrmgt_cntl = pi->clock_registers.mclk_pwrmgt_cntl;
-	u32  mpll_ad_func_cntl = pi->clock_registers.mpll_ad_func_cntl;
-	u32  mpll_dq_func_cntl = pi->clock_registers.mpll_dq_func_cntl;
-	u32  mpll_func_cntl = pi->clock_registers.mpll_func_cntl;
-	u32  mpll_func_cntl_1 = pi->clock_registers.mpll_func_cntl_1;
-	u32  mpll_func_cntl_2 = pi->clock_registers.mpll_func_cntl_2;
-	u32  mpll_ss1 = pi->clock_registers.mpll_ss1;
-	u32  mpll_ss2 = pi->clock_registers.mpll_ss2;
-	struct atom_mpll_param mpll_param;
-	int ret;
-
-	ret = amdgpu_atombios_get_memory_pll_dividers(adev, memory_clock, strobe_mode, &mpll_param);
-	if (ret)
-		return ret;
-
-	mpll_func_cntl &= ~MPLL_FUNC_CNTL__BWCTRL_MASK;
-	mpll_func_cntl |= (mpll_param.bwcntl << MPLL_FUNC_CNTL__BWCTRL__SHIFT);
-
-	mpll_func_cntl_1 &= ~(MPLL_FUNC_CNTL_1__CLKF_MASK | MPLL_FUNC_CNTL_1__CLKFRAC_MASK |
-			MPLL_FUNC_CNTL_1__VCO_MODE_MASK);
-	mpll_func_cntl_1 |= (mpll_param.clkf) << MPLL_FUNC_CNTL_1__CLKF__SHIFT |
-		(mpll_param.clkfrac << MPLL_FUNC_CNTL_1__CLKFRAC__SHIFT) |
-		(mpll_param.vco_mode << MPLL_FUNC_CNTL_1__VCO_MODE__SHIFT);
-
-	mpll_ad_func_cntl &= ~MPLL_AD_FUNC_CNTL__YCLK_POST_DIV_MASK;
-	mpll_ad_func_cntl |= (mpll_param.post_div << MPLL_AD_FUNC_CNTL__YCLK_POST_DIV__SHIFT);
-
-	if (adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5) {
-		mpll_dq_func_cntl &= ~(MPLL_DQ_FUNC_CNTL__YCLK_SEL_MASK |
-				MPLL_AD_FUNC_CNTL__YCLK_POST_DIV_MASK);
-		mpll_dq_func_cntl |= (mpll_param.yclk_sel << MPLL_DQ_FUNC_CNTL__YCLK_SEL__SHIFT) |
-				(mpll_param.post_div << MPLL_AD_FUNC_CNTL__YCLK_POST_DIV__SHIFT);
-	}
-
-	if (pi->caps_mclk_ss_support) {
-		struct amdgpu_atom_ss ss;
-		u32 freq_nom;
-		u32 tmp;
-		u32 reference_clock = adev->clock.mpll.reference_freq;
-
-		if (mpll_param.qdr == 1)
-			freq_nom = memory_clock * 4 * (1 << mpll_param.post_div);
-		else
-			freq_nom = memory_clock * 2 * (1 << mpll_param.post_div);
-
-		tmp = (freq_nom / reference_clock);
-		tmp = tmp * tmp;
-		if (amdgpu_atombios_get_asic_ss_info(adev, &ss,
-						     ASIC_INTERNAL_MEMORY_SS, freq_nom)) {
-			u32 clks = reference_clock * 5 / ss.rate;
-			u32 clkv = (u32)((((131 * ss.percentage * ss.rate) / 100) * tmp) / freq_nom);
-
-			mpll_ss1 &= ~MPLL_SS1__CLKV_MASK;
-			mpll_ss1 |= (clkv << MPLL_SS1__CLKV__SHIFT);
-
-			mpll_ss2 &= ~MPLL_SS2__CLKS_MASK;
-			mpll_ss2 |= (clks << MPLL_SS2__CLKS__SHIFT);
-		}
-	}
-
-	mclk_pwrmgt_cntl &= ~MCLK_PWRMGT_CNTL__DLL_SPEED_MASK;
-	mclk_pwrmgt_cntl |= (mpll_param.dll_speed << MCLK_PWRMGT_CNTL__DLL_SPEED__SHIFT);
-
-	if (dll_state_on)
-		mclk_pwrmgt_cntl |= MCLK_PWRMGT_CNTL__MRDCK0_PDNB_MASK |
-			MCLK_PWRMGT_CNTL__MRDCK1_PDNB_MASK;
-	else
-		mclk_pwrmgt_cntl &= ~(MCLK_PWRMGT_CNTL__MRDCK0_PDNB_MASK |
-			MCLK_PWRMGT_CNTL__MRDCK1_PDNB_MASK);
-
-	mclk->MclkFrequency = memory_clock;
-	mclk->MpllFuncCntl = mpll_func_cntl;
-	mclk->MpllFuncCntl_1 = mpll_func_cntl_1;
-	mclk->MpllFuncCntl_2 = mpll_func_cntl_2;
-	mclk->MpllAdFuncCntl = mpll_ad_func_cntl;
-	mclk->MpllDqFuncCntl = mpll_dq_func_cntl;
-	mclk->MclkPwrmgtCntl = mclk_pwrmgt_cntl;
-	mclk->DllCntl = dll_cntl;
-	mclk->MpllSs1 = mpll_ss1;
-	mclk->MpllSs2 = mpll_ss2;
-
-	return 0;
-}
-
-static int ci_populate_single_memory_level(struct amdgpu_device *adev,
-					   u32 memory_clock,
-					   SMU7_Discrete_MemoryLevel *memory_level)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-	bool dll_state_on;
-
-	if (adev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries) {
-		ret = ci_get_dependency_volt_by_clk(adev,
-						    &adev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
-						    memory_clock, &memory_level->MinVddc);
-		if (ret)
-			return ret;
-	}
-
-	if (adev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries) {
-		ret = ci_get_dependency_volt_by_clk(adev,
-						    &adev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
-						    memory_clock, &memory_level->MinVddci);
-		if (ret)
-			return ret;
-	}
-
-	if (adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries) {
-		ret = ci_get_dependency_volt_by_clk(adev,
-						    &adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
-						    memory_clock, &memory_level->MinMvdd);
-		if (ret)
-			return ret;
-	}
-
-	memory_level->MinVddcPhases = 1;
-
-	if (pi->vddc_phase_shed_control)
-		ci_populate_phase_value_based_on_mclk(adev,
-						      &adev->pm.dpm.dyn_state.phase_shedding_limits_table,
-						      memory_clock,
-						      &memory_level->MinVddcPhases);
-
-	memory_level->EnabledForActivity = 1;
-	memory_level->EnabledForThrottle = 1;
-	memory_level->UpH = 0;
-	memory_level->DownH = 100;
-	memory_level->VoltageDownH = 0;
-	memory_level->ActivityLevel = (u16)pi->mclk_activity_target;
-
-	memory_level->StutterEnable = false;
-	memory_level->StrobeEnable = false;
-	memory_level->EdcReadEnable = false;
-	memory_level->EdcWriteEnable = false;
-	memory_level->RttEnable = false;
-
-	memory_level->DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
-
-	if (pi->mclk_stutter_mode_threshold &&
-	    (memory_clock <= pi->mclk_stutter_mode_threshold) &&
-	    (!pi->uvd_enabled) &&
-	    (RREG32(mmDPG_PIPE_STUTTER_CONTROL) & DPG_PIPE_STUTTER_CONTROL__STUTTER_ENABLE_MASK) &&
-	    (adev->pm.dpm.new_active_crtc_count <= 2))
-		memory_level->StutterEnable = true;
-
-	if (pi->mclk_strobe_mode_threshold &&
-	    (memory_clock <= pi->mclk_strobe_mode_threshold))
-		memory_level->StrobeEnable = 1;
-
-	if (adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5) {
-		memory_level->StrobeRatio =
-			ci_get_mclk_frequency_ratio(memory_clock, memory_level->StrobeEnable);
-		if (pi->mclk_edc_enable_threshold &&
-		    (memory_clock > pi->mclk_edc_enable_threshold))
-			memory_level->EdcReadEnable = true;
-
-		if (pi->mclk_edc_wr_enable_threshold &&
-		    (memory_clock > pi->mclk_edc_wr_enable_threshold))
-			memory_level->EdcWriteEnable = true;
-
-		if (memory_level->StrobeEnable) {
-			if (ci_get_mclk_frequency_ratio(memory_clock, true) >=
-			    ((RREG32(mmMC_SEQ_MISC7) >> 16) & 0xf))
-				dll_state_on = ((RREG32(mmMC_SEQ_MISC5) >> 1) & 0x1) ? true : false;
-			else
-				dll_state_on = ((RREG32(mmMC_SEQ_MISC6) >> 1) & 0x1) ? true : false;
-		} else {
-			dll_state_on = pi->dll_default_on;
-		}
-	} else {
-		memory_level->StrobeRatio = ci_get_ddr3_mclk_frequency_ratio(memory_clock);
-		dll_state_on = ((RREG32(mmMC_SEQ_MISC5) >> 1) & 0x1) ? true : false;
-	}
-
-	ret = ci_calculate_mclk_params(adev, memory_clock, memory_level, memory_level->StrobeEnable, dll_state_on);
-	if (ret)
-		return ret;
-
-	memory_level->MinVddc = cpu_to_be32(memory_level->MinVddc * VOLTAGE_SCALE);
-	memory_level->MinVddcPhases = cpu_to_be32(memory_level->MinVddcPhases);
-	memory_level->MinVddci = cpu_to_be32(memory_level->MinVddci * VOLTAGE_SCALE);
-	memory_level->MinMvdd = cpu_to_be32(memory_level->MinMvdd * VOLTAGE_SCALE);
-
-	memory_level->MclkFrequency = cpu_to_be32(memory_level->MclkFrequency);
-	memory_level->ActivityLevel = cpu_to_be16(memory_level->ActivityLevel);
-	memory_level->MpllFuncCntl = cpu_to_be32(memory_level->MpllFuncCntl);
-	memory_level->MpllFuncCntl_1 = cpu_to_be32(memory_level->MpllFuncCntl_1);
-	memory_level->MpllFuncCntl_2 = cpu_to_be32(memory_level->MpllFuncCntl_2);
-	memory_level->MpllAdFuncCntl = cpu_to_be32(memory_level->MpllAdFuncCntl);
-	memory_level->MpllDqFuncCntl = cpu_to_be32(memory_level->MpllDqFuncCntl);
-	memory_level->MclkPwrmgtCntl = cpu_to_be32(memory_level->MclkPwrmgtCntl);
-	memory_level->DllCntl = cpu_to_be32(memory_level->DllCntl);
-	memory_level->MpllSs1 = cpu_to_be32(memory_level->MpllSs1);
-	memory_level->MpllSs2 = cpu_to_be32(memory_level->MpllSs2);
-
-	return 0;
-}
-
-static int ci_populate_smc_acpi_level(struct amdgpu_device *adev,
-				      SMU7_Discrete_DpmTable *table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct atom_clock_dividers dividers;
-	SMU7_Discrete_VoltageLevel voltage_level;
-	u32 spll_func_cntl = pi->clock_registers.cg_spll_func_cntl;
-	u32 spll_func_cntl_2 = pi->clock_registers.cg_spll_func_cntl_2;
-	u32 dll_cntl = pi->clock_registers.dll_cntl;
-	u32 mclk_pwrmgt_cntl = pi->clock_registers.mclk_pwrmgt_cntl;
-	int ret;
-
-	table->ACPILevel.Flags &= ~PPSMC_SWSTATE_FLAG_DC;
-
-	if (pi->acpi_vddc)
-		table->ACPILevel.MinVddc = cpu_to_be32(pi->acpi_vddc * VOLTAGE_SCALE);
-	else
-		table->ACPILevel.MinVddc = cpu_to_be32(pi->min_vddc_in_pp_table * VOLTAGE_SCALE);
-
-	table->ACPILevel.MinVddcPhases = pi->vddc_phase_shed_control ? 0 : 1;
-
-	table->ACPILevel.SclkFrequency = adev->clock.spll.reference_freq;
-
-	ret = amdgpu_atombios_get_clock_dividers(adev,
-						 COMPUTE_GPUCLK_INPUT_FLAG_SCLK,
-						 table->ACPILevel.SclkFrequency, false, &dividers);
-	if (ret)
-		return ret;
-
-	table->ACPILevel.SclkDid = (u8)dividers.post_divider;
-	table->ACPILevel.DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
-	table->ACPILevel.DeepSleepDivId = 0;
-
-	spll_func_cntl &= ~CG_SPLL_FUNC_CNTL__SPLL_PWRON_MASK;
-	spll_func_cntl |= CG_SPLL_FUNC_CNTL__SPLL_RESET_MASK;
-
-	spll_func_cntl_2 &= ~CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL_MASK;
-	spll_func_cntl_2 |= (4 << CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL__SHIFT);
-
-	table->ACPILevel.CgSpllFuncCntl = spll_func_cntl;
-	table->ACPILevel.CgSpllFuncCntl2 = spll_func_cntl_2;
-	table->ACPILevel.CgSpllFuncCntl3 = pi->clock_registers.cg_spll_func_cntl_3;
-	table->ACPILevel.CgSpllFuncCntl4 = pi->clock_registers.cg_spll_func_cntl_4;
-	table->ACPILevel.SpllSpreadSpectrum = pi->clock_registers.cg_spll_spread_spectrum;
-	table->ACPILevel.SpllSpreadSpectrum2 = pi->clock_registers.cg_spll_spread_spectrum_2;
-	table->ACPILevel.CcPwrDynRm = 0;
-	table->ACPILevel.CcPwrDynRm1 = 0;
-
-	table->ACPILevel.Flags = cpu_to_be32(table->ACPILevel.Flags);
-	table->ACPILevel.MinVddcPhases = cpu_to_be32(table->ACPILevel.MinVddcPhases);
-	table->ACPILevel.SclkFrequency = cpu_to_be32(table->ACPILevel.SclkFrequency);
-	table->ACPILevel.CgSpllFuncCntl = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl);
-	table->ACPILevel.CgSpllFuncCntl2 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl2);
-	table->ACPILevel.CgSpllFuncCntl3 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl3);
-	table->ACPILevel.CgSpllFuncCntl4 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl4);
-	table->ACPILevel.SpllSpreadSpectrum = cpu_to_be32(table->ACPILevel.SpllSpreadSpectrum);
-	table->ACPILevel.SpllSpreadSpectrum2 = cpu_to_be32(table->ACPILevel.SpllSpreadSpectrum2);
-	table->ACPILevel.CcPwrDynRm = cpu_to_be32(table->ACPILevel.CcPwrDynRm);
-	table->ACPILevel.CcPwrDynRm1 = cpu_to_be32(table->ACPILevel.CcPwrDynRm1);
-
-	table->MemoryACPILevel.MinVddc = table->ACPILevel.MinVddc;
-	table->MemoryACPILevel.MinVddcPhases = table->ACPILevel.MinVddcPhases;
-
-	if (pi->vddci_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
-		if (pi->acpi_vddci)
-			table->MemoryACPILevel.MinVddci =
-				cpu_to_be32(pi->acpi_vddci * VOLTAGE_SCALE);
-		else
-			table->MemoryACPILevel.MinVddci =
-				cpu_to_be32(pi->min_vddci_in_pp_table * VOLTAGE_SCALE);
-	}
-
-	if (ci_populate_mvdd_value(adev, 0, &voltage_level))
-		table->MemoryACPILevel.MinMvdd = 0;
-	else
-		table->MemoryACPILevel.MinMvdd =
-			cpu_to_be32(voltage_level.Voltage * VOLTAGE_SCALE);
-
-	mclk_pwrmgt_cntl |= MCLK_PWRMGT_CNTL__MRDCK0_RESET_MASK |
-		MCLK_PWRMGT_CNTL__MRDCK1_RESET_MASK;
-	mclk_pwrmgt_cntl &= ~(MCLK_PWRMGT_CNTL__MRDCK0_PDNB_MASK |
-			MCLK_PWRMGT_CNTL__MRDCK1_PDNB_MASK);
-
-	dll_cntl &= ~(DLL_CNTL__MRDCK0_BYPASS_MASK | DLL_CNTL__MRDCK1_BYPASS_MASK);
-
-	table->MemoryACPILevel.DllCntl = cpu_to_be32(dll_cntl);
-	table->MemoryACPILevel.MclkPwrmgtCntl = cpu_to_be32(mclk_pwrmgt_cntl);
-	table->MemoryACPILevel.MpllAdFuncCntl =
-		cpu_to_be32(pi->clock_registers.mpll_ad_func_cntl);
-	table->MemoryACPILevel.MpllDqFuncCntl =
-		cpu_to_be32(pi->clock_registers.mpll_dq_func_cntl);
-	table->MemoryACPILevel.MpllFuncCntl =
-		cpu_to_be32(pi->clock_registers.mpll_func_cntl);
-	table->MemoryACPILevel.MpllFuncCntl_1 =
-		cpu_to_be32(pi->clock_registers.mpll_func_cntl_1);
-	table->MemoryACPILevel.MpllFuncCntl_2 =
-		cpu_to_be32(pi->clock_registers.mpll_func_cntl_2);
-	table->MemoryACPILevel.MpllSs1 = cpu_to_be32(pi->clock_registers.mpll_ss1);
-	table->MemoryACPILevel.MpllSs2 = cpu_to_be32(pi->clock_registers.mpll_ss2);
-
-	table->MemoryACPILevel.EnabledForThrottle = 0;
-	table->MemoryACPILevel.EnabledForActivity = 0;
-	table->MemoryACPILevel.UpH = 0;
-	table->MemoryACPILevel.DownH = 100;
-	table->MemoryACPILevel.VoltageDownH = 0;
-	table->MemoryACPILevel.ActivityLevel =
-		cpu_to_be16((u16)pi->mclk_activity_target);
-
-	table->MemoryACPILevel.StutterEnable = false;
-	table->MemoryACPILevel.StrobeEnable = false;
-	table->MemoryACPILevel.EdcReadEnable = false;
-	table->MemoryACPILevel.EdcWriteEnable = false;
-	table->MemoryACPILevel.RttEnable = false;
-
-	return 0;
-}
-
-
-static int ci_enable_ulv(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ulv_parm *ulv = &pi->ulv;
-
-	if (ulv->supported) {
-		if (enable)
-			return (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_EnableULV) == PPSMC_Result_OK) ?
-				0 : -EINVAL;
-		else
-			return (amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_DisableULV) == PPSMC_Result_OK) ?
-				0 : -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ci_populate_ulv_level(struct amdgpu_device *adev,
-				 SMU7_Discrete_Ulv *state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u16 ulv_voltage = adev->pm.dpm.backbias_response_time;
-
-	state->CcPwrDynRm = 0;
-	state->CcPwrDynRm1 = 0;
-
-	if (ulv_voltage == 0) {
-		pi->ulv.supported = false;
-		return 0;
-	}
-
-	if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
-		if (ulv_voltage > adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v)
-			state->VddcOffset = 0;
-		else
-			state->VddcOffset =
-				adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v - ulv_voltage;
-	} else {
-		if (ulv_voltage > adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v)
-			state->VddcOffsetVid = 0;
-		else
-			state->VddcOffsetVid = (u8)
-				((adev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v - ulv_voltage) *
-				 VOLTAGE_VID_OFFSET_SCALE2 / VOLTAGE_VID_OFFSET_SCALE1);
-	}
-	state->VddcPhase = pi->vddc_phase_shed_control ? 0 : 1;
-
-	state->CcPwrDynRm = cpu_to_be32(state->CcPwrDynRm);
-	state->CcPwrDynRm1 = cpu_to_be32(state->CcPwrDynRm1);
-	state->VddcOffset = cpu_to_be16(state->VddcOffset);
-
-	return 0;
-}
-
-static int ci_calculate_sclk_params(struct amdgpu_device *adev,
-				    u32 engine_clock,
-				    SMU7_Discrete_GraphicsLevel *sclk)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct atom_clock_dividers dividers;
-	u32 spll_func_cntl_3 = pi->clock_registers.cg_spll_func_cntl_3;
-	u32 spll_func_cntl_4 = pi->clock_registers.cg_spll_func_cntl_4;
-	u32 cg_spll_spread_spectrum = pi->clock_registers.cg_spll_spread_spectrum;
-	u32 cg_spll_spread_spectrum_2 = pi->clock_registers.cg_spll_spread_spectrum_2;
-	u32 reference_clock = adev->clock.spll.reference_freq;
-	u32 reference_divider;
-	u32 fbdiv;
-	int ret;
-
-	ret = amdgpu_atombios_get_clock_dividers(adev,
-						 COMPUTE_GPUCLK_INPUT_FLAG_SCLK,
-						 engine_clock, false, &dividers);
-	if (ret)
-		return ret;
-
-	reference_divider = 1 + dividers.ref_div;
-	fbdiv = dividers.fb_div & 0x3FFFFFF;
-
-	spll_func_cntl_3 &= ~CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV_MASK;
-	spll_func_cntl_3 |= (fbdiv << CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV__SHIFT);
-	spll_func_cntl_3 |= CG_SPLL_FUNC_CNTL_3__SPLL_DITHEN_MASK;
-
-	if (pi->caps_sclk_ss_support) {
-		struct amdgpu_atom_ss ss;
-		u32 vco_freq = engine_clock * dividers.post_div;
-
-		if (amdgpu_atombios_get_asic_ss_info(adev, &ss,
-						     ASIC_INTERNAL_ENGINE_SS, vco_freq)) {
-			u32 clk_s = reference_clock * 5 / (reference_divider * ss.rate);
-			u32 clk_v = 4 * ss.percentage * fbdiv / (clk_s * 10000);
-
-			cg_spll_spread_spectrum &= ~(CG_SPLL_SPREAD_SPECTRUM__CLKS_MASK | CG_SPLL_SPREAD_SPECTRUM__SSEN_MASK);
-			cg_spll_spread_spectrum |= (clk_s << CG_SPLL_SPREAD_SPECTRUM__CLKS__SHIFT);
-			cg_spll_spread_spectrum |= (1 << CG_SPLL_SPREAD_SPECTRUM__SSEN__SHIFT);
-
-			cg_spll_spread_spectrum_2 &= ~CG_SPLL_SPREAD_SPECTRUM_2__CLKV_MASK;
-			cg_spll_spread_spectrum_2 |= (clk_v << CG_SPLL_SPREAD_SPECTRUM_2__CLKV__SHIFT);
-		}
-	}
-
-	sclk->SclkFrequency = engine_clock;
-	sclk->CgSpllFuncCntl3 = spll_func_cntl_3;
-	sclk->CgSpllFuncCntl4 = spll_func_cntl_4;
-	sclk->SpllSpreadSpectrum = cg_spll_spread_spectrum;
-	sclk->SpllSpreadSpectrum2  = cg_spll_spread_spectrum_2;
-	sclk->SclkDid = (u8)dividers.post_divider;
-
-	return 0;
-}
-
-static int ci_populate_single_graphic_level(struct amdgpu_device *adev,
-					    u32 engine_clock,
-					    u16 sclk_activity_level_t,
-					    SMU7_Discrete_GraphicsLevel *graphic_level)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-
-	ret = ci_calculate_sclk_params(adev, engine_clock, graphic_level);
-	if (ret)
-		return ret;
-
-	ret = ci_get_dependency_volt_by_clk(adev,
-					    &adev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
-					    engine_clock, &graphic_level->MinVddc);
-	if (ret)
-		return ret;
-
-	graphic_level->SclkFrequency = engine_clock;
-
-	graphic_level->Flags =  0;
-	graphic_level->MinVddcPhases = 1;
-
-	if (pi->vddc_phase_shed_control)
-		ci_populate_phase_value_based_on_sclk(adev,
-						      &adev->pm.dpm.dyn_state.phase_shedding_limits_table,
-						      engine_clock,
-						      &graphic_level->MinVddcPhases);
-
-	graphic_level->ActivityLevel = sclk_activity_level_t;
-
-	graphic_level->CcPwrDynRm = 0;
-	graphic_level->CcPwrDynRm1 = 0;
-	graphic_level->EnabledForThrottle = 1;
-	graphic_level->UpH = 0;
-	graphic_level->DownH = 0;
-	graphic_level->VoltageDownH = 0;
-	graphic_level->PowerThrottle = 0;
-
-	if (pi->caps_sclk_ds)
-		graphic_level->DeepSleepDivId = ci_get_sleep_divider_id_from_clock(engine_clock,
-										   CISLAND_MINIMUM_ENGINE_CLOCK);
-
-	graphic_level->DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
-
-	graphic_level->Flags = cpu_to_be32(graphic_level->Flags);
-	graphic_level->MinVddc = cpu_to_be32(graphic_level->MinVddc * VOLTAGE_SCALE);
-	graphic_level->MinVddcPhases = cpu_to_be32(graphic_level->MinVddcPhases);
-	graphic_level->SclkFrequency = cpu_to_be32(graphic_level->SclkFrequency);
-	graphic_level->ActivityLevel = cpu_to_be16(graphic_level->ActivityLevel);
-	graphic_level->CgSpllFuncCntl3 = cpu_to_be32(graphic_level->CgSpllFuncCntl3);
-	graphic_level->CgSpllFuncCntl4 = cpu_to_be32(graphic_level->CgSpllFuncCntl4);
-	graphic_level->SpllSpreadSpectrum = cpu_to_be32(graphic_level->SpllSpreadSpectrum);
-	graphic_level->SpllSpreadSpectrum2 = cpu_to_be32(graphic_level->SpllSpreadSpectrum2);
-	graphic_level->CcPwrDynRm = cpu_to_be32(graphic_level->CcPwrDynRm);
-	graphic_level->CcPwrDynRm1 = cpu_to_be32(graphic_level->CcPwrDynRm1);
-
-	return 0;
-}
-
-static int ci_populate_all_graphic_levels(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_dpm_table *dpm_table = &pi->dpm_table;
-	u32 level_array_address = pi->dpm_table_start +
-		offsetof(SMU7_Discrete_DpmTable, GraphicsLevel);
-	u32 level_array_size = sizeof(SMU7_Discrete_GraphicsLevel) *
-		SMU7_MAX_LEVELS_GRAPHICS;
-	SMU7_Discrete_GraphicsLevel *levels = pi->smc_state_table.GraphicsLevel;
-	u32 i, ret;
-
-	memset(levels, 0, level_array_size);
-
-	for (i = 0; i < dpm_table->sclk_table.count; i++) {
-		ret = ci_populate_single_graphic_level(adev,
-						       dpm_table->sclk_table.dpm_levels[i].value,
-						       (u16)pi->activity_target[i],
-						       &pi->smc_state_table.GraphicsLevel[i]);
-		if (ret)
-			return ret;
-		if (i > 1)
-			pi->smc_state_table.GraphicsLevel[i].DeepSleepDivId = 0;
-		if (i == (dpm_table->sclk_table.count - 1))
-			pi->smc_state_table.GraphicsLevel[i].DisplayWatermark =
-				PPSMC_DISPLAY_WATERMARK_HIGH;
-	}
-	pi->smc_state_table.GraphicsLevel[0].EnabledForActivity = 1;
-
-	pi->smc_state_table.GraphicsDpmLevelCount = (u8)dpm_table->sclk_table.count;
-	pi->dpm_level_enable_mask.sclk_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&dpm_table->sclk_table);
-
-	ret = amdgpu_ci_copy_bytes_to_smc(adev, level_array_address,
-				   (u8 *)levels, level_array_size,
-				   pi->sram_end);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static int ci_populate_ulv_state(struct amdgpu_device *adev,
-				 SMU7_Discrete_Ulv *ulv_level)
-{
-	return ci_populate_ulv_level(adev, ulv_level);
-}
-
-static int ci_populate_all_memory_levels(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_dpm_table *dpm_table = &pi->dpm_table;
-	u32 level_array_address = pi->dpm_table_start +
-		offsetof(SMU7_Discrete_DpmTable, MemoryLevel);
-	u32 level_array_size = sizeof(SMU7_Discrete_MemoryLevel) *
-		SMU7_MAX_LEVELS_MEMORY;
-	SMU7_Discrete_MemoryLevel *levels = pi->smc_state_table.MemoryLevel;
-	u32 i, ret;
-
-	memset(levels, 0, level_array_size);
-
-	for (i = 0; i < dpm_table->mclk_table.count; i++) {
-		if (dpm_table->mclk_table.dpm_levels[i].value == 0)
-			return -EINVAL;
-		ret = ci_populate_single_memory_level(adev,
-						      dpm_table->mclk_table.dpm_levels[i].value,
-						      &pi->smc_state_table.MemoryLevel[i]);
-		if (ret)
-			return ret;
-	}
-
-	if ((dpm_table->mclk_table.count >= 2) &&
-	    ((adev->pdev->device == 0x67B0) || (adev->pdev->device == 0x67B1))) {
-		pi->smc_state_table.MemoryLevel[1].MinVddc =
-			pi->smc_state_table.MemoryLevel[0].MinVddc;
-		pi->smc_state_table.MemoryLevel[1].MinVddcPhases =
-			pi->smc_state_table.MemoryLevel[0].MinVddcPhases;
-	}
-
-	pi->smc_state_table.MemoryLevel[0].ActivityLevel = cpu_to_be16(0x1F);
-
-	pi->smc_state_table.MemoryDpmLevelCount = (u8)dpm_table->mclk_table.count;
-	pi->dpm_level_enable_mask.mclk_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&dpm_table->mclk_table);
-
-	pi->smc_state_table.MemoryLevel[dpm_table->mclk_table.count - 1].DisplayWatermark =
-		PPSMC_DISPLAY_WATERMARK_HIGH;
-
-	ret = amdgpu_ci_copy_bytes_to_smc(adev, level_array_address,
-				   (u8 *)levels, level_array_size,
-				   pi->sram_end);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static void ci_reset_single_dpm_table(struct amdgpu_device *adev,
-				      struct ci_single_dpm_table* dpm_table,
-				      u32 count)
-{
-	u32 i;
-
-	dpm_table->count = count;
-	for (i = 0; i < MAX_REGULAR_DPM_NUMBER; i++)
-		dpm_table->dpm_levels[i].enabled = false;
-}
-
-static void ci_setup_pcie_table_entry(struct ci_single_dpm_table* dpm_table,
-				      u32 index, u32 pcie_gen, u32 pcie_lanes)
-{
-	dpm_table->dpm_levels[index].value = pcie_gen;
-	dpm_table->dpm_levels[index].param1 = pcie_lanes;
-	dpm_table->dpm_levels[index].enabled = true;
-}
-
-static int ci_setup_default_pcie_tables(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (!pi->use_pcie_performance_levels && !pi->use_pcie_powersaving_levels)
-		return -EINVAL;
-
-	if (pi->use_pcie_performance_levels && !pi->use_pcie_powersaving_levels) {
-		pi->pcie_gen_powersaving = pi->pcie_gen_performance;
-		pi->pcie_lane_powersaving = pi->pcie_lane_performance;
-	} else if (!pi->use_pcie_performance_levels && pi->use_pcie_powersaving_levels) {
-		pi->pcie_gen_performance = pi->pcie_gen_powersaving;
-		pi->pcie_lane_performance = pi->pcie_lane_powersaving;
-	}
-
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.pcie_speed_table,
-				  SMU7_MAX_LEVELS_LINK);
-
-	if (adev->asic_type == CHIP_BONAIRE)
-		ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 0,
-					  pi->pcie_gen_powersaving.min,
-					  pi->pcie_lane_powersaving.max);
-	else
-		ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 0,
-					  pi->pcie_gen_powersaving.min,
-					  pi->pcie_lane_powersaving.min);
-	ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 1,
-				  pi->pcie_gen_performance.min,
-				  pi->pcie_lane_performance.min);
-	ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 2,
-				  pi->pcie_gen_powersaving.min,
-				  pi->pcie_lane_powersaving.max);
-	ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 3,
-				  pi->pcie_gen_performance.min,
-				  pi->pcie_lane_performance.max);
-	ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 4,
-				  pi->pcie_gen_powersaving.max,
-				  pi->pcie_lane_powersaving.max);
-	ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 5,
-				  pi->pcie_gen_performance.max,
-				  pi->pcie_lane_performance.max);
-
-	pi->dpm_table.pcie_speed_table.count = 6;
-
-	return 0;
-}
-
-static int ci_setup_default_dpm_tables(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_clock_voltage_dependency_table *allowed_sclk_vddc_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
-	struct amdgpu_clock_voltage_dependency_table *allowed_mclk_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk;
-	struct amdgpu_cac_leakage_table *std_voltage_table =
-		&adev->pm.dpm.dyn_state.cac_leakage_table;
-	u32 i;
-
-	if (allowed_sclk_vddc_table == NULL)
-		return -EINVAL;
-	if (allowed_sclk_vddc_table->count < 1)
-		return -EINVAL;
-	if (allowed_mclk_table == NULL)
-		return -EINVAL;
-	if (allowed_mclk_table->count < 1)
-		return -EINVAL;
-
-	memset(&pi->dpm_table, 0, sizeof(struct ci_dpm_table));
-
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.sclk_table,
-				  SMU7_MAX_LEVELS_GRAPHICS);
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.mclk_table,
-				  SMU7_MAX_LEVELS_MEMORY);
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.vddc_table,
-				  SMU7_MAX_LEVELS_VDDC);
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.vddci_table,
-				  SMU7_MAX_LEVELS_VDDCI);
-	ci_reset_single_dpm_table(adev,
-				  &pi->dpm_table.mvdd_table,
-				  SMU7_MAX_LEVELS_MVDD);
-
-	pi->dpm_table.sclk_table.count = 0;
-	for (i = 0; i < allowed_sclk_vddc_table->count; i++) {
-		if ((i == 0) ||
-		    (pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count-1].value !=
-		     allowed_sclk_vddc_table->entries[i].clk)) {
-			pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].value =
-				allowed_sclk_vddc_table->entries[i].clk;
-			pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].enabled =
-				(i == 0) ? true : false;
-			pi->dpm_table.sclk_table.count++;
-		}
-	}
-
-	pi->dpm_table.mclk_table.count = 0;
-	for (i = 0; i < allowed_mclk_table->count; i++) {
-		if ((i == 0) ||
-		    (pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count-1].value !=
-		     allowed_mclk_table->entries[i].clk)) {
-			pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].value =
-				allowed_mclk_table->entries[i].clk;
-			pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].enabled =
-				(i == 0) ? true : false;
-			pi->dpm_table.mclk_table.count++;
-		}
-	}
-
-	for (i = 0; i < allowed_sclk_vddc_table->count; i++) {
-		pi->dpm_table.vddc_table.dpm_levels[i].value =
-			allowed_sclk_vddc_table->entries[i].v;
-		pi->dpm_table.vddc_table.dpm_levels[i].param1 =
-			std_voltage_table->entries[i].leakage;
-		pi->dpm_table.vddc_table.dpm_levels[i].enabled = true;
-	}
-	pi->dpm_table.vddc_table.count = allowed_sclk_vddc_table->count;
-
-	allowed_mclk_table = &adev->pm.dpm.dyn_state.vddci_dependency_on_mclk;
-	if (allowed_mclk_table) {
-		for (i = 0; i < allowed_mclk_table->count; i++) {
-			pi->dpm_table.vddci_table.dpm_levels[i].value =
-				allowed_mclk_table->entries[i].v;
-			pi->dpm_table.vddci_table.dpm_levels[i].enabled = true;
-		}
-		pi->dpm_table.vddci_table.count = allowed_mclk_table->count;
-	}
-
-	allowed_mclk_table = &adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk;
-	if (allowed_mclk_table) {
-		for (i = 0; i < allowed_mclk_table->count; i++) {
-			pi->dpm_table.mvdd_table.dpm_levels[i].value =
-				allowed_mclk_table->entries[i].v;
-			pi->dpm_table.mvdd_table.dpm_levels[i].enabled = true;
-		}
-		pi->dpm_table.mvdd_table.count = allowed_mclk_table->count;
-	}
-
-	ci_setup_default_pcie_tables(adev);
-
-	/* save a copy of the default DPM table */
-	memcpy(&(pi->golden_dpm_table), &(pi->dpm_table),
-			sizeof(struct ci_dpm_table));
-
-	return 0;
-}
-
-static int ci_find_boot_level(struct ci_single_dpm_table *table,
-			      u32 value, u32 *boot_level)
-{
-	u32 i;
-	int ret = -EINVAL;
-
-	for(i = 0; i < table->count; i++) {
-		if (value == table->dpm_levels[i].value) {
-			*boot_level = i;
-			ret = 0;
-		}
-	}
-
-	return ret;
-}
-
-static int ci_init_smc_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ulv_parm *ulv = &pi->ulv;
-	struct amdgpu_ps *amdgpu_boot_state = adev->pm.dpm.boot_ps;
-	SMU7_Discrete_DpmTable *table = &pi->smc_state_table;
-	int ret;
-
-	ret = ci_setup_default_dpm_tables(adev);
-	if (ret)
-		return ret;
-
-	if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_NONE)
-		ci_populate_smc_voltage_tables(adev, table);
-
-	ci_init_fps_limits(adev);
-
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_HARDWAREDC)
-		table->SystemFlags |= PPSMC_SYSTEMFLAG_GPIO_DC;
-
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_STEPVDDC)
-		table->SystemFlags |= PPSMC_SYSTEMFLAG_STEPVDDC;
-
-	if (adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5)
-		table->SystemFlags |= PPSMC_SYSTEMFLAG_GDDR5;
-
-	if (ulv->supported) {
-		ret = ci_populate_ulv_state(adev, &pi->smc_state_table.Ulv);
-		if (ret)
-			return ret;
-		WREG32_SMC(ixCG_ULV_PARAMETER, ulv->cg_ulv_parameter);
-	}
-
-	ret = ci_populate_all_graphic_levels(adev);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_all_memory_levels(adev);
-	if (ret)
-		return ret;
-
-	ci_populate_smc_link_level(adev, table);
-
-	ret = ci_populate_smc_acpi_level(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_vce_level(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_acp_level(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_samu_level(adev, table);
-	if (ret)
-		return ret;
-
-	ret = ci_do_program_memory_timing_parameters(adev);
-	if (ret)
-		return ret;
-
-	ret = ci_populate_smc_uvd_level(adev, table);
-	if (ret)
-		return ret;
-
-	table->UvdBootLevel  = 0;
-	table->VceBootLevel  = 0;
-	table->AcpBootLevel  = 0;
-	table->SamuBootLevel  = 0;
-	table->GraphicsBootLevel  = 0;
-	table->MemoryBootLevel  = 0;
-
-	ret = ci_find_boot_level(&pi->dpm_table.sclk_table,
-				 pi->vbios_boot_state.sclk_bootup_value,
-				 (u32 *)&pi->smc_state_table.GraphicsBootLevel);
-
-	ret = ci_find_boot_level(&pi->dpm_table.mclk_table,
-				 pi->vbios_boot_state.mclk_bootup_value,
-				 (u32 *)&pi->smc_state_table.MemoryBootLevel);
-
-	table->BootVddc = pi->vbios_boot_state.vddc_bootup_value;
-	table->BootVddci = pi->vbios_boot_state.vddci_bootup_value;
-	table->BootMVdd = pi->vbios_boot_state.mvdd_bootup_value;
-
-	ci_populate_smc_initial_state(adev, amdgpu_boot_state);
-
-	ret = ci_populate_bapm_parameters_in_dpm_table(adev);
-	if (ret)
-		return ret;
-
-	table->UVDInterval = 1;
-	table->VCEInterval = 1;
-	table->ACPInterval = 1;
-	table->SAMUInterval = 1;
-	table->GraphicsVoltageChangeEnable = 1;
-	table->GraphicsThermThrottleEnable = 1;
-	table->GraphicsInterval = 1;
-	table->VoltageInterval = 1;
-	table->ThermalInterval = 1;
-	table->TemperatureLimitHigh = (u16)((pi->thermal_temp_setting.temperature_high *
-					     CISLANDS_Q88_FORMAT_CONVERSION_UNIT) / 1000);
-	table->TemperatureLimitLow = (u16)((pi->thermal_temp_setting.temperature_low *
-					    CISLANDS_Q88_FORMAT_CONVERSION_UNIT) / 1000);
-	table->MemoryVoltageChangeEnable = 1;
-	table->MemoryInterval = 1;
-	table->VoltageResponseTime = 0;
-	table->VddcVddciDelta = 4000;
-	table->PhaseResponseTime = 0;
-	table->MemoryThermThrottleEnable = 1;
-	table->PCIeBootLinkLevel = pi->dpm_table.pcie_speed_table.count - 1;
-	table->PCIeGenInterval = 1;
-	if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2)
-		table->SVI2Enable  = 1;
-	else
-		table->SVI2Enable  = 0;
-
-	table->ThermGpio = 17;
-	table->SclkStepSize = 0x4000;
-
-	table->SystemFlags = cpu_to_be32(table->SystemFlags);
-	table->SmioMaskVddcVid = cpu_to_be32(table->SmioMaskVddcVid);
-	table->SmioMaskVddcPhase = cpu_to_be32(table->SmioMaskVddcPhase);
-	table->SmioMaskVddciVid = cpu_to_be32(table->SmioMaskVddciVid);
-	table->SmioMaskMvddVid = cpu_to_be32(table->SmioMaskMvddVid);
-	table->SclkStepSize = cpu_to_be32(table->SclkStepSize);
-	table->TemperatureLimitHigh = cpu_to_be16(table->TemperatureLimitHigh);
-	table->TemperatureLimitLow = cpu_to_be16(table->TemperatureLimitLow);
-	table->VddcVddciDelta = cpu_to_be16(table->VddcVddciDelta);
-	table->VoltageResponseTime = cpu_to_be16(table->VoltageResponseTime);
-	table->PhaseResponseTime = cpu_to_be16(table->PhaseResponseTime);
-	table->BootVddc = cpu_to_be16(table->BootVddc * VOLTAGE_SCALE);
-	table->BootVddci = cpu_to_be16(table->BootVddci * VOLTAGE_SCALE);
-	table->BootMVdd = cpu_to_be16(table->BootMVdd * VOLTAGE_SCALE);
-
-	ret = amdgpu_ci_copy_bytes_to_smc(adev,
-				   pi->dpm_table_start +
-				   offsetof(SMU7_Discrete_DpmTable, SystemFlags),
-				   (u8 *)&table->SystemFlags,
-				   sizeof(SMU7_Discrete_DpmTable) - 3 * sizeof(SMU7_PIDController),
-				   pi->sram_end);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static void ci_trim_single_dpm_states(struct amdgpu_device *adev,
-				      struct ci_single_dpm_table *dpm_table,
-				      u32 low_limit, u32 high_limit)
-{
-	u32 i;
-
-	for (i = 0; i < dpm_table->count; i++) {
-		if ((dpm_table->dpm_levels[i].value < low_limit) ||
-		    (dpm_table->dpm_levels[i].value > high_limit))
-			dpm_table->dpm_levels[i].enabled = false;
-		else
-			dpm_table->dpm_levels[i].enabled = true;
-	}
-}
-
-static void ci_trim_pcie_dpm_states(struct amdgpu_device *adev,
-				    u32 speed_low, u32 lanes_low,
-				    u32 speed_high, u32 lanes_high)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_single_dpm_table *pcie_table = &pi->dpm_table.pcie_speed_table;
-	u32 i, j;
-
-	for (i = 0; i < pcie_table->count; i++) {
-		if ((pcie_table->dpm_levels[i].value < speed_low) ||
-		    (pcie_table->dpm_levels[i].param1 < lanes_low) ||
-		    (pcie_table->dpm_levels[i].value > speed_high) ||
-		    (pcie_table->dpm_levels[i].param1 > lanes_high))
-			pcie_table->dpm_levels[i].enabled = false;
-		else
-			pcie_table->dpm_levels[i].enabled = true;
-	}
-
-	for (i = 0; i < pcie_table->count; i++) {
-		if (pcie_table->dpm_levels[i].enabled) {
-			for (j = i + 1; j < pcie_table->count; j++) {
-				if (pcie_table->dpm_levels[j].enabled) {
-					if ((pcie_table->dpm_levels[i].value == pcie_table->dpm_levels[j].value) &&
-					    (pcie_table->dpm_levels[i].param1 == pcie_table->dpm_levels[j].param1))
-						pcie_table->dpm_levels[j].enabled = false;
-				}
-			}
-		}
-	}
-}
-
-static int ci_trim_dpm_states(struct amdgpu_device *adev,
-			      struct amdgpu_ps *amdgpu_state)
-{
-	struct ci_ps *state = ci_get_ps(amdgpu_state);
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 high_limit_count;
-
-	if (state->performance_level_count < 1)
-		return -EINVAL;
-
-	if (state->performance_level_count == 1)
-		high_limit_count = 0;
-	else
-		high_limit_count = 1;
-
-	ci_trim_single_dpm_states(adev,
-				  &pi->dpm_table.sclk_table,
-				  state->performance_levels[0].sclk,
-				  state->performance_levels[high_limit_count].sclk);
-
-	ci_trim_single_dpm_states(adev,
-				  &pi->dpm_table.mclk_table,
-				  state->performance_levels[0].mclk,
-				  state->performance_levels[high_limit_count].mclk);
-
-	ci_trim_pcie_dpm_states(adev,
-				state->performance_levels[0].pcie_gen,
-				state->performance_levels[0].pcie_lane,
-				state->performance_levels[high_limit_count].pcie_gen,
-				state->performance_levels[high_limit_count].pcie_lane);
-
-	return 0;
-}
-
-static int ci_apply_disp_minimum_voltage_request(struct amdgpu_device *adev)
-{
-	struct amdgpu_clock_voltage_dependency_table *disp_voltage_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk;
-	struct amdgpu_clock_voltage_dependency_table *vddc_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
-	u32 requested_voltage = 0;
-	u32 i;
-
-	if (disp_voltage_table == NULL)
-		return -EINVAL;
-	if (!disp_voltage_table->count)
-		return -EINVAL;
-
-	for (i = 0; i < disp_voltage_table->count; i++) {
-		if (adev->clock.current_dispclk == disp_voltage_table->entries[i].clk)
-			requested_voltage = disp_voltage_table->entries[i].v;
-	}
-
-	for (i = 0; i < vddc_table->count; i++) {
-		if (requested_voltage <= vddc_table->entries[i].v) {
-			requested_voltage = vddc_table->entries[i].v;
-			return (amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-								  PPSMC_MSG_VddC_Request,
-								  requested_voltage * VOLTAGE_SCALE) == PPSMC_Result_OK) ?
-				0 : -EINVAL;
-		}
-	}
-
-	return -EINVAL;
-}
-
-static int ci_upload_dpm_level_enable_mask(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	PPSMC_Result result;
-
-	ci_apply_disp_minimum_voltage_request(adev);
-
-	if (!pi->sclk_dpm_key_disabled) {
-		if (pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
-			result = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-								   PPSMC_MSG_SCLKDPM_SetEnabledMask,
-								   pi->dpm_level_enable_mask.sclk_dpm_enable_mask);
-			if (result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	}
-
-	if (!pi->mclk_dpm_key_disabled) {
-		if (pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
-			result = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-								   PPSMC_MSG_MCLKDPM_SetEnabledMask,
-								   pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
-			if (result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	}
-
-#if 0
-	if (!pi->pcie_dpm_key_disabled) {
-		if (pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
-			result = amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-								   PPSMC_MSG_PCIeDPM_SetEnabledMask,
-								   pi->dpm_level_enable_mask.pcie_dpm_enable_mask);
-			if (result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-	}
-#endif
-
-	return 0;
-}
-
-static void ci_find_dpm_states_clocks_in_dpm_table(struct amdgpu_device *adev,
-						   struct amdgpu_ps *amdgpu_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *state = ci_get_ps(amdgpu_state);
-	struct ci_single_dpm_table *sclk_table = &pi->dpm_table.sclk_table;
-	u32 sclk = state->performance_levels[state->performance_level_count-1].sclk;
-	struct ci_single_dpm_table *mclk_table = &pi->dpm_table.mclk_table;
-	u32 mclk = state->performance_levels[state->performance_level_count-1].mclk;
-	u32 i;
-
-	pi->need_update_smu7_dpm_table = 0;
-
-	for (i = 0; i < sclk_table->count; i++) {
-		if (sclk == sclk_table->dpm_levels[i].value)
-			break;
-	}
-
-	if (i >= sclk_table->count) {
-		pi->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK;
-	} else {
-		/* XXX check display min clock requirements */
-		if (CISLAND_MINIMUM_ENGINE_CLOCK != CISLAND_MINIMUM_ENGINE_CLOCK)
-			pi->need_update_smu7_dpm_table |= DPMTABLE_UPDATE_SCLK;
-	}
-
-	for (i = 0; i < mclk_table->count; i++) {
-		if (mclk == mclk_table->dpm_levels[i].value)
-			break;
-	}
-
-	if (i >= mclk_table->count)
-		pi->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK;
-
-	if (adev->pm.dpm.current_active_crtc_count !=
-	    adev->pm.dpm.new_active_crtc_count)
-		pi->need_update_smu7_dpm_table |= DPMTABLE_UPDATE_MCLK;
-}
-
-static int ci_populate_and_upload_sclk_mclk_dpm_levels(struct amdgpu_device *adev,
-						       struct amdgpu_ps *amdgpu_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *state = ci_get_ps(amdgpu_state);
-	u32 sclk = state->performance_levels[state->performance_level_count-1].sclk;
-	u32 mclk = state->performance_levels[state->performance_level_count-1].mclk;
-	struct ci_dpm_table *dpm_table = &pi->dpm_table;
-	int ret;
-
-	if (!pi->need_update_smu7_dpm_table)
-		return 0;
-
-	if (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_SCLK)
-		dpm_table->sclk_table.dpm_levels[dpm_table->sclk_table.count-1].value = sclk;
-
-	if (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)
-		dpm_table->mclk_table.dpm_levels[dpm_table->mclk_table.count-1].value = mclk;
-
-	if (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK)) {
-		ret = ci_populate_all_graphic_levels(adev);
-		if (ret)
-			return ret;
-	}
-
-	if (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_MCLK | DPMTABLE_UPDATE_MCLK)) {
-		ret = ci_populate_all_memory_levels(adev);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-static int ci_enable_uvd_dpm(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct amdgpu_clock_and_voltage_limits *max_limits;
-	int i;
-
-	if (adev->pm.ac_power)
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-	else
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
-
-	if (enable) {
-		pi->dpm_level_enable_mask.uvd_dpm_enable_mask = 0;
-
-		for (i = adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
-			if (adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
-				pi->dpm_level_enable_mask.uvd_dpm_enable_mask |= 1 << i;
-
-				if (!pi->caps_uvd_dpm)
-					break;
-			}
-		}
-
-		amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-						  PPSMC_MSG_UVDDPM_SetEnabledMask,
-						  pi->dpm_level_enable_mask.uvd_dpm_enable_mask);
-
-		if (pi->last_mclk_dpm_enable_mask & 0x1) {
-			pi->uvd_enabled = true;
-			pi->dpm_level_enable_mask.mclk_dpm_enable_mask &= 0xFFFFFFFE;
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-							  PPSMC_MSG_MCLKDPM_SetEnabledMask,
-							  pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
-		}
-	} else {
-		if (pi->uvd_enabled) {
-			pi->uvd_enabled = false;
-			pi->dpm_level_enable_mask.mclk_dpm_enable_mask |= 1;
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-							  PPSMC_MSG_MCLKDPM_SetEnabledMask,
-							  pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
-		}
-	}
-
-	return (amdgpu_ci_send_msg_to_smc(adev, enable ?
-				   PPSMC_MSG_UVDDPM_Enable : PPSMC_MSG_UVDDPM_Disable) == PPSMC_Result_OK) ?
-		0 : -EINVAL;
-}
-
-static int ci_enable_vce_dpm(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct amdgpu_clock_and_voltage_limits *max_limits;
-	int i;
-
-	if (adev->pm.ac_power)
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-	else
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
-
-	if (enable) {
-		pi->dpm_level_enable_mask.vce_dpm_enable_mask = 0;
-		for (i = adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
-			if (adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
-				pi->dpm_level_enable_mask.vce_dpm_enable_mask |= 1 << i;
-
-				if (!pi->caps_vce_dpm)
-					break;
-			}
-		}
-
-		amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-						  PPSMC_MSG_VCEDPM_SetEnabledMask,
-						  pi->dpm_level_enable_mask.vce_dpm_enable_mask);
-	}
-
-	return (amdgpu_ci_send_msg_to_smc(adev, enable ?
-				   PPSMC_MSG_VCEDPM_Enable : PPSMC_MSG_VCEDPM_Disable) == PPSMC_Result_OK) ?
-		0 : -EINVAL;
-}
-
-#if 0
-static int ci_enable_samu_dpm(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct amdgpu_clock_and_voltage_limits *max_limits;
-	int i;
-
-	if (adev->pm.ac_power)
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-	else
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
-
-	if (enable) {
-		pi->dpm_level_enable_mask.samu_dpm_enable_mask = 0;
-		for (i = adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
-			if (adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
-				pi->dpm_level_enable_mask.samu_dpm_enable_mask |= 1 << i;
-
-				if (!pi->caps_samu_dpm)
-					break;
-			}
-		}
-
-		amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-						  PPSMC_MSG_SAMUDPM_SetEnabledMask,
-						  pi->dpm_level_enable_mask.samu_dpm_enable_mask);
-	}
-	return (amdgpu_ci_send_msg_to_smc(adev, enable ?
-				   PPSMC_MSG_SAMUDPM_Enable : PPSMC_MSG_SAMUDPM_Disable) == PPSMC_Result_OK) ?
-		0 : -EINVAL;
-}
-
-static int ci_enable_acp_dpm(struct amdgpu_device *adev, bool enable)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	const struct amdgpu_clock_and_voltage_limits *max_limits;
-	int i;
-
-	if (adev->pm.ac_power)
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-	else
-		max_limits = &adev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
-
-	if (enable) {
-		pi->dpm_level_enable_mask.acp_dpm_enable_mask = 0;
-		for (i = adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
-			if (adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
-				pi->dpm_level_enable_mask.acp_dpm_enable_mask |= 1 << i;
-
-				if (!pi->caps_acp_dpm)
-					break;
-			}
-		}
-
-		amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-						  PPSMC_MSG_ACPDPM_SetEnabledMask,
-						  pi->dpm_level_enable_mask.acp_dpm_enable_mask);
-	}
-
-	return (amdgpu_ci_send_msg_to_smc(adev, enable ?
-				   PPSMC_MSG_ACPDPM_Enable : PPSMC_MSG_ACPDPM_Disable) == PPSMC_Result_OK) ?
-		0 : -EINVAL;
-}
-#endif
-
-static int ci_update_uvd_dpm(struct amdgpu_device *adev, bool gate)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-	int ret = 0;
-
-	if (!gate) {
-		/* turn the clocks on when decoding */
-		if (pi->caps_uvd_dpm ||
-		    (adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count <= 0))
-			pi->smc_state_table.UvdBootLevel = 0;
-		else
-			pi->smc_state_table.UvdBootLevel =
-				adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count - 1;
-
-		tmp = RREG32_SMC(ixDPM_TABLE_475);
-		tmp &= ~DPM_TABLE_475__UvdBootLevel_MASK;
-		tmp |= (pi->smc_state_table.UvdBootLevel << DPM_TABLE_475__UvdBootLevel__SHIFT);
-		WREG32_SMC(ixDPM_TABLE_475, tmp);
-		ret = ci_enable_uvd_dpm(adev, true);
-	} else {
-		ret = ci_enable_uvd_dpm(adev, false);
-		if (ret)
-			return ret;
-	}
-
-	return ret;
-}
-
-static u8 ci_get_vce_boot_level(struct amdgpu_device *adev)
-{
-	u8 i;
-	u32 min_evclk = 30000; /* ??? */
-	struct amdgpu_vce_clock_voltage_dependency_table *table =
-		&adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
-
-	for (i = 0; i < table->count; i++) {
-		if (table->entries[i].evclk >= min_evclk)
-			return i;
-	}
-
-	return table->count - 1;
-}
-
-static int ci_update_vce_dpm(struct amdgpu_device *adev,
-			     struct amdgpu_ps *amdgpu_new_state,
-			     struct amdgpu_ps *amdgpu_current_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret = 0;
-	u32 tmp;
-
-	if (amdgpu_current_state->evclk != amdgpu_new_state->evclk) {
-		if (amdgpu_new_state->evclk) {
-			pi->smc_state_table.VceBootLevel = ci_get_vce_boot_level(adev);
-			tmp = RREG32_SMC(ixDPM_TABLE_475);
-			tmp &= ~DPM_TABLE_475__VceBootLevel_MASK;
-			tmp |= (pi->smc_state_table.VceBootLevel << DPM_TABLE_475__VceBootLevel__SHIFT);
-			WREG32_SMC(ixDPM_TABLE_475, tmp);
-
-			ret = ci_enable_vce_dpm(adev, true);
-		} else {
-			ret = ci_enable_vce_dpm(adev, false);
-			if (ret)
-				return ret;
-		}
-	}
-	return ret;
-}
-
-#if 0
-static int ci_update_samu_dpm(struct amdgpu_device *adev, bool gate)
-{
-	return ci_enable_samu_dpm(adev, gate);
-}
-
-static int ci_update_acp_dpm(struct amdgpu_device *adev, bool gate)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp;
-
-	if (!gate) {
-		pi->smc_state_table.AcpBootLevel = 0;
-
-		tmp = RREG32_SMC(ixDPM_TABLE_475);
-		tmp &= ~AcpBootLevel_MASK;
-		tmp |= AcpBootLevel(pi->smc_state_table.AcpBootLevel);
-		WREG32_SMC(ixDPM_TABLE_475, tmp);
-	}
-
-	return ci_enable_acp_dpm(adev, !gate);
-}
-#endif
-
-static int ci_generate_dpm_level_enable_mask(struct amdgpu_device *adev,
-					     struct amdgpu_ps *amdgpu_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-
-	ret = ci_trim_dpm_states(adev, amdgpu_state);
-	if (ret)
-		return ret;
-
-	pi->dpm_level_enable_mask.sclk_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&pi->dpm_table.sclk_table);
-	pi->dpm_level_enable_mask.mclk_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&pi->dpm_table.mclk_table);
-	pi->last_mclk_dpm_enable_mask =
-		pi->dpm_level_enable_mask.mclk_dpm_enable_mask;
-	if (pi->uvd_enabled) {
-		if (pi->dpm_level_enable_mask.mclk_dpm_enable_mask & 1)
-			pi->dpm_level_enable_mask.mclk_dpm_enable_mask &= 0xFFFFFFFE;
-	}
-	pi->dpm_level_enable_mask.pcie_dpm_enable_mask =
-		ci_get_dpm_level_enable_mask_value(&pi->dpm_table.pcie_speed_table);
-
-	return 0;
-}
-
-static u32 ci_get_lowest_enabled_level(struct amdgpu_device *adev,
-				       u32 level_mask)
-{
-	u32 level = 0;
-
-	while ((level_mask & (1 << level)) == 0)
-		level++;
-
-	return level;
-}
-
-
-static int ci_dpm_force_performance_level(void *handle,
-					  enum amd_dpm_forced_level level)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 tmp, levels, i;
-	int ret;
-
-	if (level == AMD_DPM_FORCED_LEVEL_HIGH) {
-		if ((!pi->pcie_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
-			levels = 0;
-			tmp = pi->dpm_level_enable_mask.pcie_dpm_enable_mask;
-			while (tmp >>= 1)
-				levels++;
-			if (levels) {
-				ret = ci_dpm_force_state_pcie(adev, level);
-				if (ret)
-					return ret;
-				for (i = 0; i < adev->usec_timeout; i++) {
-					tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_1) &
-					TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX_MASK) >>
-					TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX__SHIFT;
-					if (tmp == levels)
-						break;
-					udelay(1);
-				}
-			}
-		}
-		if ((!pi->sclk_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
-			levels = 0;
-			tmp = pi->dpm_level_enable_mask.sclk_dpm_enable_mask;
-			while (tmp >>= 1)
-				levels++;
-			if (levels) {
-				ret = ci_dpm_force_state_sclk(adev, levels);
-				if (ret)
-					return ret;
-				for (i = 0; i < adev->usec_timeout; i++) {
-					tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) &
-					TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >>
-					TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT;
-					if (tmp == levels)
-						break;
-					udelay(1);
-				}
-			}
-		}
-		if ((!pi->mclk_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
-			levels = 0;
-			tmp = pi->dpm_level_enable_mask.mclk_dpm_enable_mask;
-			while (tmp >>= 1)
-				levels++;
-			if (levels) {
-				ret = ci_dpm_force_state_mclk(adev, levels);
-				if (ret)
-					return ret;
-				for (i = 0; i < adev->usec_timeout; i++) {
-					tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) &
-					TARGET_AND_CURRENT_PROFILE_INDEX__CURR_MCLK_INDEX_MASK) >>
-					TARGET_AND_CURRENT_PROFILE_INDEX__CURR_MCLK_INDEX__SHIFT;
-					if (tmp == levels)
-						break;
-					udelay(1);
-				}
-			}
-		}
-	} else if (level == AMD_DPM_FORCED_LEVEL_LOW) {
-		if ((!pi->sclk_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
-			levels = ci_get_lowest_enabled_level(adev,
-							     pi->dpm_level_enable_mask.sclk_dpm_enable_mask);
-			ret = ci_dpm_force_state_sclk(adev, levels);
-			if (ret)
-				return ret;
-			for (i = 0; i < adev->usec_timeout; i++) {
-				tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) &
-				TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >>
-				TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT;
-				if (tmp == levels)
-					break;
-				udelay(1);
-			}
-		}
-		if ((!pi->mclk_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
-			levels = ci_get_lowest_enabled_level(adev,
-							     pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
-			ret = ci_dpm_force_state_mclk(adev, levels);
-			if (ret)
-				return ret;
-			for (i = 0; i < adev->usec_timeout; i++) {
-				tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) &
-				TARGET_AND_CURRENT_PROFILE_INDEX__CURR_MCLK_INDEX_MASK) >>
-				TARGET_AND_CURRENT_PROFILE_INDEX__CURR_MCLK_INDEX__SHIFT;
-				if (tmp == levels)
-					break;
-				udelay(1);
-			}
-		}
-		if ((!pi->pcie_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
-			levels = ci_get_lowest_enabled_level(adev,
-							     pi->dpm_level_enable_mask.pcie_dpm_enable_mask);
-			ret = ci_dpm_force_state_pcie(adev, levels);
-			if (ret)
-				return ret;
-			for (i = 0; i < adev->usec_timeout; i++) {
-				tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_1) &
-				TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX_MASK) >>
-				TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX__SHIFT;
-				if (tmp == levels)
-					break;
-				udelay(1);
-			}
-		}
-	} else if (level == AMD_DPM_FORCED_LEVEL_AUTO) {
-		if (!pi->pcie_dpm_key_disabled) {
-			PPSMC_Result smc_result;
-
-			smc_result = amdgpu_ci_send_msg_to_smc(adev,
-							       PPSMC_MSG_PCIeDPM_UnForceLevel);
-			if (smc_result != PPSMC_Result_OK)
-				return -EINVAL;
-		}
-		ret = ci_upload_dpm_level_enable_mask(adev);
-		if (ret)
-			return ret;
-	}
-
-	adev->pm.dpm.forced_level = level;
-
-	return 0;
-}
-
-static int ci_set_mc_special_registers(struct amdgpu_device *adev,
-				       struct ci_mc_reg_table *table)
-{
-	u8 i, j, k;
-	u32 temp_reg;
-
-	for (i = 0, j = table->last; i < table->last; i++) {
-		if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-			return -EINVAL;
-		switch(table->mc_reg_address[i].s1) {
-		case mmMC_SEQ_MISC1:
-			temp_reg = RREG32(mmMC_PMG_CMD_EMRS);
-			table->mc_reg_address[j].s1 = mmMC_PMG_CMD_EMRS;
-			table->mc_reg_address[j].s0 = mmMC_SEQ_PMG_CMD_EMRS_LP;
-			for (k = 0; k < table->num_entries; k++) {
-				table->mc_reg_table_entry[k].mc_data[j] =
-					((temp_reg & 0xffff0000)) | ((table->mc_reg_table_entry[k].mc_data[i] & 0xffff0000) >> 16);
-			}
-			j++;
-
-			if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-				return -EINVAL;
-			temp_reg = RREG32(mmMC_PMG_CMD_MRS);
-			table->mc_reg_address[j].s1 = mmMC_PMG_CMD_MRS;
-			table->mc_reg_address[j].s0 = mmMC_SEQ_PMG_CMD_MRS_LP;
-			for (k = 0; k < table->num_entries; k++) {
-				table->mc_reg_table_entry[k].mc_data[j] =
-					(temp_reg & 0xffff0000) | (table->mc_reg_table_entry[k].mc_data[i] & 0x0000ffff);
-				if (adev->gmc.vram_type != AMDGPU_VRAM_TYPE_GDDR5)
-					table->mc_reg_table_entry[k].mc_data[j] |= 0x100;
-			}
-			j++;
-
-			if (adev->gmc.vram_type != AMDGPU_VRAM_TYPE_GDDR5) {
-				if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-					return -EINVAL;
-				table->mc_reg_address[j].s1 = mmMC_PMG_AUTO_CMD;
-				table->mc_reg_address[j].s0 = mmMC_PMG_AUTO_CMD;
-				for (k = 0; k < table->num_entries; k++) {
-					table->mc_reg_table_entry[k].mc_data[j] =
-						(table->mc_reg_table_entry[k].mc_data[i] & 0xffff0000) >> 16;
-				}
-				j++;
-			}
-			break;
-		case mmMC_SEQ_RESERVE_M:
-			temp_reg = RREG32(mmMC_PMG_CMD_MRS1);
-			table->mc_reg_address[j].s1 = mmMC_PMG_CMD_MRS1;
-			table->mc_reg_address[j].s0 = mmMC_SEQ_PMG_CMD_MRS1_LP;
-			for (k = 0; k < table->num_entries; k++) {
-				table->mc_reg_table_entry[k].mc_data[j] =
-					(temp_reg & 0xffff0000) | (table->mc_reg_table_entry[k].mc_data[i] & 0x0000ffff);
-			}
-			j++;
-			break;
-		default:
-			break;
-		}
-
-	}
-
-	table->last = j;
-
-	return 0;
-}
-
-static bool ci_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
-{
-	bool result = true;
-
-	switch(in_reg) {
-	case mmMC_SEQ_RAS_TIMING:
-		*out_reg = mmMC_SEQ_RAS_TIMING_LP;
-		break;
-	case mmMC_SEQ_DLL_STBY:
-		*out_reg = mmMC_SEQ_DLL_STBY_LP;
-		break;
-	case mmMC_SEQ_G5PDX_CMD0:
-		*out_reg = mmMC_SEQ_G5PDX_CMD0_LP;
-		break;
-	case mmMC_SEQ_G5PDX_CMD1:
-		*out_reg = mmMC_SEQ_G5PDX_CMD1_LP;
-		break;
-	case mmMC_SEQ_G5PDX_CTRL:
-		*out_reg = mmMC_SEQ_G5PDX_CTRL_LP;
-		break;
-	case mmMC_SEQ_CAS_TIMING:
-		*out_reg = mmMC_SEQ_CAS_TIMING_LP;
-	    break;
-	case mmMC_SEQ_MISC_TIMING:
-		*out_reg = mmMC_SEQ_MISC_TIMING_LP;
-		break;
-	case mmMC_SEQ_MISC_TIMING2:
-		*out_reg = mmMC_SEQ_MISC_TIMING2_LP;
-		break;
-	case mmMC_SEQ_PMG_DVS_CMD:
-		*out_reg = mmMC_SEQ_PMG_DVS_CMD_LP;
-		break;
-	case mmMC_SEQ_PMG_DVS_CTL:
-		*out_reg = mmMC_SEQ_PMG_DVS_CTL_LP;
-		break;
-	case mmMC_SEQ_RD_CTL_D0:
-		*out_reg = mmMC_SEQ_RD_CTL_D0_LP;
-		break;
-	case mmMC_SEQ_RD_CTL_D1:
-		*out_reg = mmMC_SEQ_RD_CTL_D1_LP;
-		break;
-	case mmMC_SEQ_WR_CTL_D0:
-		*out_reg = mmMC_SEQ_WR_CTL_D0_LP;
-		break;
-	case mmMC_SEQ_WR_CTL_D1:
-		*out_reg = mmMC_SEQ_WR_CTL_D1_LP;
-		break;
-	case mmMC_PMG_CMD_EMRS:
-		*out_reg = mmMC_SEQ_PMG_CMD_EMRS_LP;
-		break;
-	case mmMC_PMG_CMD_MRS:
-		*out_reg = mmMC_SEQ_PMG_CMD_MRS_LP;
-		break;
-	case mmMC_PMG_CMD_MRS1:
-		*out_reg = mmMC_SEQ_PMG_CMD_MRS1_LP;
-		break;
-	case mmMC_SEQ_PMG_TIMING:
-		*out_reg = mmMC_SEQ_PMG_TIMING_LP;
-		break;
-	case mmMC_PMG_CMD_MRS2:
-		*out_reg = mmMC_SEQ_PMG_CMD_MRS2_LP;
-		break;
-	case mmMC_SEQ_WR_CTL_2:
-		*out_reg = mmMC_SEQ_WR_CTL_2_LP;
-		break;
-	default:
-		result = false;
-		break;
-	}
-
-	return result;
-}
-
-static void ci_set_valid_flag(struct ci_mc_reg_table *table)
-{
-	u8 i, j;
-
-	for (i = 0; i < table->last; i++) {
-		for (j = 1; j < table->num_entries; j++) {
-			if (table->mc_reg_table_entry[j-1].mc_data[i] !=
-			    table->mc_reg_table_entry[j].mc_data[i]) {
-				table->valid_flag |= 1 << i;
-				break;
-			}
-		}
-	}
-}
-
-static void ci_set_s0_mc_reg_index(struct ci_mc_reg_table *table)
-{
-	u32 i;
-	u16 address;
-
-	for (i = 0; i < table->last; i++) {
-		table->mc_reg_address[i].s0 =
-			ci_check_s0_mc_reg_index(table->mc_reg_address[i].s1, &address) ?
-			address : table->mc_reg_address[i].s1;
-	}
-}
-
-static int ci_copy_vbios_mc_reg_table(const struct atom_mc_reg_table *table,
-				      struct ci_mc_reg_table *ci_table)
-{
-	u8 i, j;
-
-	if (table->last > SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-		return -EINVAL;
-	if (table->num_entries > MAX_AC_TIMING_ENTRIES)
-		return -EINVAL;
-
-	for (i = 0; i < table->last; i++)
-		ci_table->mc_reg_address[i].s1 = table->mc_reg_address[i].s1;
-
-	ci_table->last = table->last;
-
-	for (i = 0; i < table->num_entries; i++) {
-		ci_table->mc_reg_table_entry[i].mclk_max =
-			table->mc_reg_table_entry[i].mclk_max;
-		for (j = 0; j < table->last; j++)
-			ci_table->mc_reg_table_entry[i].mc_data[j] =
-				table->mc_reg_table_entry[i].mc_data[j];
-	}
-	ci_table->num_entries = table->num_entries;
-
-	return 0;
-}
-
-static int ci_register_patching_mc_seq(struct amdgpu_device *adev,
-				       struct ci_mc_reg_table *table)
-{
-	u8 i, k;
-	u32 tmp;
-	bool patch;
-
-	tmp = RREG32(mmMC_SEQ_MISC0);
-	patch = ((tmp & 0x0000f00) == 0x300) ? true : false;
-
-	if (patch &&
-	    ((adev->pdev->device == 0x67B0) ||
-	     (adev->pdev->device == 0x67B1))) {
-		for (i = 0; i < table->last; i++) {
-			if (table->last >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-				return -EINVAL;
-			switch (table->mc_reg_address[i].s1) {
-			case mmMC_SEQ_MISC1:
-				for (k = 0; k < table->num_entries; k++) {
-					if ((table->mc_reg_table_entry[k].mclk_max == 125000) ||
-					    (table->mc_reg_table_entry[k].mclk_max == 137500))
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFFFFFF8) |
-							0x00000007;
-				}
-				break;
-			case mmMC_SEQ_WR_CTL_D0:
-				for (k = 0; k < table->num_entries; k++) {
-					if ((table->mc_reg_table_entry[k].mclk_max == 125000) ||
-					    (table->mc_reg_table_entry[k].mclk_max == 137500))
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFFF0F00) |
-							0x0000D0DD;
-				}
-				break;
-			case mmMC_SEQ_WR_CTL_D1:
-				for (k = 0; k < table->num_entries; k++) {
-					if ((table->mc_reg_table_entry[k].mclk_max == 125000) ||
-					    (table->mc_reg_table_entry[k].mclk_max == 137500))
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFFF0F00) |
-							0x0000D0DD;
-				}
-				break;
-			case mmMC_SEQ_WR_CTL_2:
-				for (k = 0; k < table->num_entries; k++) {
-					if ((table->mc_reg_table_entry[k].mclk_max == 125000) ||
-					    (table->mc_reg_table_entry[k].mclk_max == 137500))
-						table->mc_reg_table_entry[k].mc_data[i] = 0;
-				}
-				break;
-			case mmMC_SEQ_CAS_TIMING:
-				for (k = 0; k < table->num_entries; k++) {
-					if (table->mc_reg_table_entry[k].mclk_max == 125000)
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFE0FE0F) |
-							0x000C0140;
-					else if (table->mc_reg_table_entry[k].mclk_max == 137500)
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFE0FE0F) |
-							0x000C0150;
-				}
-				break;
-			case mmMC_SEQ_MISC_TIMING:
-				for (k = 0; k < table->num_entries; k++) {
-					if (table->mc_reg_table_entry[k].mclk_max == 125000)
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFFFFFE0) |
-							0x00000030;
-					else if (table->mc_reg_table_entry[k].mclk_max == 137500)
-						table->mc_reg_table_entry[k].mc_data[i] =
-							(table->mc_reg_table_entry[k].mc_data[i] & 0xFFFFFFE0) |
-							0x00000035;
-				}
-				break;
-			default:
-				break;
-			}
-		}
-
-		WREG32(mmMC_SEQ_IO_DEBUG_INDEX, 3);
-		tmp = RREG32(mmMC_SEQ_IO_DEBUG_DATA);
-		tmp = (tmp & 0xFFF8FFFF) | (1 << 16);
-		WREG32(mmMC_SEQ_IO_DEBUG_INDEX, 3);
-		WREG32(mmMC_SEQ_IO_DEBUG_DATA, tmp);
-	}
-
-	return 0;
-}
-
-static int ci_initialize_mc_reg_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct atom_mc_reg_table *table;
-	struct ci_mc_reg_table *ci_table = &pi->mc_reg_table;
-	u8 module_index = ci_get_memory_module_index(adev);
-	int ret;
-
-	table = kzalloc(sizeof(struct atom_mc_reg_table), GFP_KERNEL);
-	if (!table)
-		return -ENOMEM;
-
-	WREG32(mmMC_SEQ_RAS_TIMING_LP, RREG32(mmMC_SEQ_RAS_TIMING));
-	WREG32(mmMC_SEQ_CAS_TIMING_LP, RREG32(mmMC_SEQ_CAS_TIMING));
-	WREG32(mmMC_SEQ_DLL_STBY_LP, RREG32(mmMC_SEQ_DLL_STBY));
-	WREG32(mmMC_SEQ_G5PDX_CMD0_LP, RREG32(mmMC_SEQ_G5PDX_CMD0));
-	WREG32(mmMC_SEQ_G5PDX_CMD1_LP, RREG32(mmMC_SEQ_G5PDX_CMD1));
-	WREG32(mmMC_SEQ_G5PDX_CTRL_LP, RREG32(mmMC_SEQ_G5PDX_CTRL));
-	WREG32(mmMC_SEQ_PMG_DVS_CMD_LP, RREG32(mmMC_SEQ_PMG_DVS_CMD));
-	WREG32(mmMC_SEQ_PMG_DVS_CTL_LP, RREG32(mmMC_SEQ_PMG_DVS_CTL));
-	WREG32(mmMC_SEQ_MISC_TIMING_LP, RREG32(mmMC_SEQ_MISC_TIMING));
-	WREG32(mmMC_SEQ_MISC_TIMING2_LP, RREG32(mmMC_SEQ_MISC_TIMING2));
-	WREG32(mmMC_SEQ_PMG_CMD_EMRS_LP, RREG32(mmMC_PMG_CMD_EMRS));
-	WREG32(mmMC_SEQ_PMG_CMD_MRS_LP, RREG32(mmMC_PMG_CMD_MRS));
-	WREG32(mmMC_SEQ_PMG_CMD_MRS1_LP, RREG32(mmMC_PMG_CMD_MRS1));
-	WREG32(mmMC_SEQ_WR_CTL_D0_LP, RREG32(mmMC_SEQ_WR_CTL_D0));
-	WREG32(mmMC_SEQ_WR_CTL_D1_LP, RREG32(mmMC_SEQ_WR_CTL_D1));
-	WREG32(mmMC_SEQ_RD_CTL_D0_LP, RREG32(mmMC_SEQ_RD_CTL_D0));
-	WREG32(mmMC_SEQ_RD_CTL_D1_LP, RREG32(mmMC_SEQ_RD_CTL_D1));
-	WREG32(mmMC_SEQ_PMG_TIMING_LP, RREG32(mmMC_SEQ_PMG_TIMING));
-	WREG32(mmMC_SEQ_PMG_CMD_MRS2_LP, RREG32(mmMC_PMG_CMD_MRS2));
-	WREG32(mmMC_SEQ_WR_CTL_2_LP, RREG32(mmMC_SEQ_WR_CTL_2));
-
-	ret = amdgpu_atombios_init_mc_reg_table(adev, module_index, table);
-	if (ret)
-		goto init_mc_done;
-
-	ret = ci_copy_vbios_mc_reg_table(table, ci_table);
-	if (ret)
-		goto init_mc_done;
-
-	ci_set_s0_mc_reg_index(ci_table);
-
-	ret = ci_register_patching_mc_seq(adev, ci_table);
-	if (ret)
-		goto init_mc_done;
-
-	ret = ci_set_mc_special_registers(adev, ci_table);
-	if (ret)
-		goto init_mc_done;
-
-	ci_set_valid_flag(ci_table);
-
-init_mc_done:
-	kfree(table);
-
-	return ret;
-}
-
-static int ci_populate_mc_reg_addresses(struct amdgpu_device *adev,
-					SMU7_Discrete_MCRegisters *mc_reg_table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 i, j;
-
-	for (i = 0, j = 0; j < pi->mc_reg_table.last; j++) {
-		if (pi->mc_reg_table.valid_flag & (1 << j)) {
-			if (i >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
-				return -EINVAL;
-			mc_reg_table->address[i].s0 = cpu_to_be16(pi->mc_reg_table.mc_reg_address[j].s0);
-			mc_reg_table->address[i].s1 = cpu_to_be16(pi->mc_reg_table.mc_reg_address[j].s1);
-			i++;
-		}
-	}
-
-	mc_reg_table->last = (u8)i;
-
-	return 0;
-}
-
-static void ci_convert_mc_registers(const struct ci_mc_reg_entry *entry,
-				    SMU7_Discrete_MCRegisterSet *data,
-				    u32 num_entries, u32 valid_flag)
-{
-	u32 i, j;
-
-	for (i = 0, j = 0; j < num_entries; j++) {
-		if (valid_flag & (1 << j)) {
-			data->value[i] = cpu_to_be32(entry->mc_data[j]);
-			i++;
-		}
-	}
-}
-
-static void ci_convert_mc_reg_table_entry_to_smc(struct amdgpu_device *adev,
-						 const u32 memory_clock,
-						 SMU7_Discrete_MCRegisterSet *mc_reg_table_data)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 i = 0;
-
-	for(i = 0; i < pi->mc_reg_table.num_entries; i++) {
-		if (memory_clock <= pi->mc_reg_table.mc_reg_table_entry[i].mclk_max)
-			break;
-	}
-
-	if ((i == pi->mc_reg_table.num_entries) && (i > 0))
-		--i;
-
-	ci_convert_mc_registers(&pi->mc_reg_table.mc_reg_table_entry[i],
-				mc_reg_table_data, pi->mc_reg_table.last,
-				pi->mc_reg_table.valid_flag);
-}
-
-static void ci_convert_mc_reg_table_to_smc(struct amdgpu_device *adev,
-					   SMU7_Discrete_MCRegisters *mc_reg_table)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	u32 i;
-
-	for (i = 0; i < pi->dpm_table.mclk_table.count; i++)
-		ci_convert_mc_reg_table_entry_to_smc(adev,
-						     pi->dpm_table.mclk_table.dpm_levels[i].value,
-						     &mc_reg_table->data[i]);
-}
-
-static int ci_populate_initial_mc_reg_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	int ret;
-
-	memset(&pi->smc_mc_reg_table, 0, sizeof(SMU7_Discrete_MCRegisters));
-
-	ret = ci_populate_mc_reg_addresses(adev, &pi->smc_mc_reg_table);
-	if (ret)
-		return ret;
-	ci_convert_mc_reg_table_to_smc(adev, &pi->smc_mc_reg_table);
-
-	return amdgpu_ci_copy_bytes_to_smc(adev,
-				    pi->mc_reg_table_start,
-				    (u8 *)&pi->smc_mc_reg_table,
-				    sizeof(SMU7_Discrete_MCRegisters),
-				    pi->sram_end);
-}
-
-static int ci_update_and_upload_mc_reg_table(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (!(pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK))
-		return 0;
-
-	memset(&pi->smc_mc_reg_table, 0, sizeof(SMU7_Discrete_MCRegisters));
-
-	ci_convert_mc_reg_table_to_smc(adev, &pi->smc_mc_reg_table);
-
-	return amdgpu_ci_copy_bytes_to_smc(adev,
-				    pi->mc_reg_table_start +
-				    offsetof(SMU7_Discrete_MCRegisters, data[0]),
-				    (u8 *)&pi->smc_mc_reg_table.data[0],
-				    sizeof(SMU7_Discrete_MCRegisterSet) *
-				    pi->dpm_table.mclk_table.count,
-				    pi->sram_end);
-}
-
-static void ci_enable_voltage_control(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixGENERAL_PWRMGT);
-
-	tmp |= GENERAL_PWRMGT__VOLT_PWRMGT_EN_MASK;
-	WREG32_SMC(ixGENERAL_PWRMGT, tmp);
-}
-
-static enum amdgpu_pcie_gen ci_get_maximum_link_speed(struct amdgpu_device *adev,
-						      struct amdgpu_ps *amdgpu_state)
-{
-	struct ci_ps *state = ci_get_ps(amdgpu_state);
-	int i;
-	u16 pcie_speed, max_speed = 0;
-
-	for (i = 0; i < state->performance_level_count; i++) {
-		pcie_speed = state->performance_levels[i].pcie_gen;
-		if (max_speed < pcie_speed)
-			max_speed = pcie_speed;
-	}
-
-	return max_speed;
-}
-
-static u16 ci_get_current_pcie_speed(struct amdgpu_device *adev)
-{
-	u32 speed_cntl = 0;
-
-	speed_cntl = RREG32_PCIE(ixPCIE_LC_SPEED_CNTL) &
-		PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK;
-	speed_cntl >>= PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
-
-	return (u16)speed_cntl;
-}
-
-static int ci_get_current_pcie_lane_number(struct amdgpu_device *adev)
-{
-	u32 link_width = 0;
-
-	link_width = RREG32_PCIE(ixPCIE_LC_LINK_WIDTH_CNTL) &
-		PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK;
-	link_width >>= PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
-
-	switch (link_width) {
-	case 1:
-		return 1;
-	case 2:
-		return 2;
-	case 3:
-		return 4;
-	case 4:
-		return 8;
-	case 0:
-	case 6:
-	default:
-		return 16;
-	}
-}
-
-static void ci_request_link_speed_change_before_state_change(struct amdgpu_device *adev,
-							     struct amdgpu_ps *amdgpu_new_state,
-							     struct amdgpu_ps *amdgpu_current_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	enum amdgpu_pcie_gen target_link_speed =
-		ci_get_maximum_link_speed(adev, amdgpu_new_state);
-	enum amdgpu_pcie_gen current_link_speed;
-
-	if (pi->force_pcie_gen == AMDGPU_PCIE_GEN_INVALID)
-		current_link_speed = ci_get_maximum_link_speed(adev, amdgpu_current_state);
-	else
-		current_link_speed = pi->force_pcie_gen;
-
-	pi->force_pcie_gen = AMDGPU_PCIE_GEN_INVALID;
-	pi->pspp_notify_required = false;
-	if (target_link_speed > current_link_speed) {
-		switch (target_link_speed) {
-#ifdef CONFIG_ACPI
-		case AMDGPU_PCIE_GEN3:
-			if (amdgpu_acpi_pcie_performance_request(adev, PCIE_PERF_REQ_PECI_GEN3, false) == 0)
-				break;
-			pi->force_pcie_gen = AMDGPU_PCIE_GEN2;
-			if (current_link_speed == AMDGPU_PCIE_GEN2)
-				break;
-		case AMDGPU_PCIE_GEN2:
-			if (amdgpu_acpi_pcie_performance_request(adev, PCIE_PERF_REQ_PECI_GEN2, false) == 0)
-				break;
-#endif
-		default:
-			pi->force_pcie_gen = ci_get_current_pcie_speed(adev);
-			break;
-		}
-	} else {
-		if (target_link_speed < current_link_speed)
-			pi->pspp_notify_required = true;
-	}
-}
-
-static void ci_notify_link_speed_change_after_state_change(struct amdgpu_device *adev,
-							   struct amdgpu_ps *amdgpu_new_state,
-							   struct amdgpu_ps *amdgpu_current_state)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	enum amdgpu_pcie_gen target_link_speed =
-		ci_get_maximum_link_speed(adev, amdgpu_new_state);
-	u8 request;
-
-	if (pi->pspp_notify_required) {
-		if (target_link_speed == AMDGPU_PCIE_GEN3)
-			request = PCIE_PERF_REQ_PECI_GEN3;
-		else if (target_link_speed == AMDGPU_PCIE_GEN2)
-			request = PCIE_PERF_REQ_PECI_GEN2;
-		else
-			request = PCIE_PERF_REQ_PECI_GEN1;
-
-		if ((request == PCIE_PERF_REQ_PECI_GEN1) &&
-		    (ci_get_current_pcie_speed(adev) > 0))
-			return;
-
-#ifdef CONFIG_ACPI
-		amdgpu_acpi_pcie_performance_request(adev, request, false);
-#endif
-	}
-}
-
-static int ci_set_private_data_variables_based_on_pptable(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_clock_voltage_dependency_table *allowed_sclk_vddc_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
-	struct amdgpu_clock_voltage_dependency_table *allowed_mclk_vddc_table =
-		&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk;
-	struct amdgpu_clock_voltage_dependency_table *allowed_mclk_vddci_table =
-		&adev->pm.dpm.dyn_state.vddci_dependency_on_mclk;
-
-	if (allowed_sclk_vddc_table == NULL)
-		return -EINVAL;
-	if (allowed_sclk_vddc_table->count < 1)
-		return -EINVAL;
-	if (allowed_mclk_vddc_table == NULL)
-		return -EINVAL;
-	if (allowed_mclk_vddc_table->count < 1)
-		return -EINVAL;
-	if (allowed_mclk_vddci_table == NULL)
-		return -EINVAL;
-	if (allowed_mclk_vddci_table->count < 1)
-		return -EINVAL;
-
-	pi->min_vddc_in_pp_table = allowed_sclk_vddc_table->entries[0].v;
-	pi->max_vddc_in_pp_table =
-		allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].v;
-
-	pi->min_vddci_in_pp_table = allowed_mclk_vddci_table->entries[0].v;
-	pi->max_vddci_in_pp_table =
-		allowed_mclk_vddci_table->entries[allowed_mclk_vddci_table->count - 1].v;
-
-	adev->pm.dpm.dyn_state.max_clock_voltage_on_ac.sclk =
-		allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].clk;
-	adev->pm.dpm.dyn_state.max_clock_voltage_on_ac.mclk =
-		allowed_mclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].clk;
-	adev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddc =
-		allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].v;
-	adev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddci =
-		allowed_mclk_vddci_table->entries[allowed_mclk_vddci_table->count - 1].v;
-
-	return 0;
-}
-
-static void ci_patch_with_vddc_leakage(struct amdgpu_device *adev, u16 *vddc)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_leakage_voltage *leakage_table = &pi->vddc_leakage;
-	u32 leakage_index;
-
-	for (leakage_index = 0; leakage_index < leakage_table->count; leakage_index++) {
-		if (leakage_table->leakage_id[leakage_index] == *vddc) {
-			*vddc = leakage_table->actual_voltage[leakage_index];
-			break;
-		}
-	}
-}
-
-static void ci_patch_with_vddci_leakage(struct amdgpu_device *adev, u16 *vddci)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_leakage_voltage *leakage_table = &pi->vddci_leakage;
-	u32 leakage_index;
-
-	for (leakage_index = 0; leakage_index < leakage_table->count; leakage_index++) {
-		if (leakage_table->leakage_id[leakage_index] == *vddci) {
-			*vddci = leakage_table->actual_voltage[leakage_index];
-			break;
-		}
-	}
-}
-
-static void ci_patch_clock_voltage_dependency_table_with_vddc_leakage(struct amdgpu_device *adev,
-								      struct amdgpu_clock_voltage_dependency_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddc_leakage(adev, &table->entries[i].v);
-	}
-}
-
-static void ci_patch_clock_voltage_dependency_table_with_vddci_leakage(struct amdgpu_device *adev,
-								       struct amdgpu_clock_voltage_dependency_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddci_leakage(adev, &table->entries[i].v);
-	}
-}
-
-static void ci_patch_vce_clock_voltage_dependency_table_with_vddc_leakage(struct amdgpu_device *adev,
-									  struct amdgpu_vce_clock_voltage_dependency_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddc_leakage(adev, &table->entries[i].v);
-	}
-}
-
-static void ci_patch_uvd_clock_voltage_dependency_table_with_vddc_leakage(struct amdgpu_device *adev,
-									  struct amdgpu_uvd_clock_voltage_dependency_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddc_leakage(adev, &table->entries[i].v);
-	}
-}
-
-static void ci_patch_vddc_phase_shed_limit_table_with_vddc_leakage(struct amdgpu_device *adev,
-								   struct amdgpu_phase_shedding_limits_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddc_leakage(adev, &table->entries[i].voltage);
-	}
-}
-
-static void ci_patch_clock_voltage_limits_with_vddc_leakage(struct amdgpu_device *adev,
-							    struct amdgpu_clock_and_voltage_limits *table)
-{
-	if (table) {
-		ci_patch_with_vddc_leakage(adev, (u16 *)&table->vddc);
-		ci_patch_with_vddci_leakage(adev, (u16 *)&table->vddci);
-	}
-}
-
-static void ci_patch_cac_leakage_table_with_vddc_leakage(struct amdgpu_device *adev,
-							 struct amdgpu_cac_leakage_table *table)
-{
-	u32 i;
-
-	if (table) {
-		for (i = 0; i < table->count; i++)
-			ci_patch_with_vddc_leakage(adev, &table->entries[i].vddc);
-	}
-}
-
-static void ci_patch_dependency_tables_with_leakage(struct amdgpu_device *adev)
-{
-
-	ci_patch_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								  &adev->pm.dpm.dyn_state.vddc_dependency_on_sclk);
-	ci_patch_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								  &adev->pm.dpm.dyn_state.vddc_dependency_on_mclk);
-	ci_patch_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								  &adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk);
-	ci_patch_clock_voltage_dependency_table_with_vddci_leakage(adev,
-								   &adev->pm.dpm.dyn_state.vddci_dependency_on_mclk);
-	ci_patch_vce_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								      &adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table);
-	ci_patch_uvd_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								      &adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table);
-	ci_patch_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								  &adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table);
-	ci_patch_clock_voltage_dependency_table_with_vddc_leakage(adev,
-								  &adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table);
-	ci_patch_vddc_phase_shed_limit_table_with_vddc_leakage(adev,
-							       &adev->pm.dpm.dyn_state.phase_shedding_limits_table);
-	ci_patch_clock_voltage_limits_with_vddc_leakage(adev,
-							&adev->pm.dpm.dyn_state.max_clock_voltage_on_ac);
-	ci_patch_clock_voltage_limits_with_vddc_leakage(adev,
-							&adev->pm.dpm.dyn_state.max_clock_voltage_on_dc);
-	ci_patch_cac_leakage_table_with_vddc_leakage(adev,
-						     &adev->pm.dpm.dyn_state.cac_leakage_table);
-
-}
-
-static void ci_update_current_ps(struct amdgpu_device *adev,
-				 struct amdgpu_ps *rps)
-{
-	struct ci_ps *new_ps = ci_get_ps(rps);
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	pi->current_rps = *rps;
-	pi->current_ps = *new_ps;
-	pi->current_rps.ps_priv = &pi->current_ps;
-	adev->pm.dpm.current_ps = &pi->current_rps;
-}
-
-static void ci_update_requested_ps(struct amdgpu_device *adev,
-				   struct amdgpu_ps *rps)
-{
-	struct ci_ps *new_ps = ci_get_ps(rps);
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	pi->requested_rps = *rps;
-	pi->requested_ps = *new_ps;
-	pi->requested_rps.ps_priv = &pi->requested_ps;
-	adev->pm.dpm.requested_ps = &pi->requested_rps;
-}
-
-static int ci_dpm_pre_set_power_state(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps requested_ps = *adev->pm.dpm.requested_ps;
-	struct amdgpu_ps *new_ps = &requested_ps;
-
-	ci_update_requested_ps(adev, new_ps);
-
-	ci_apply_state_adjust_rules(adev, &pi->requested_rps);
-
-	return 0;
-}
-
-static void ci_dpm_post_set_power_state(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps *new_ps = &pi->requested_rps;
-
-	ci_update_current_ps(adev, new_ps);
-}
-
-
-static void ci_dpm_setup_asic(struct amdgpu_device *adev)
-{
-	ci_read_clock_registers(adev);
-	ci_enable_acpi_power_management(adev);
-	ci_init_sclk_t(adev);
-}
-
-static int ci_dpm_enable(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps *boot_ps = adev->pm.dpm.boot_ps;
-	int ret;
-
-	if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
-		ci_enable_voltage_control(adev);
-		ret = ci_construct_voltage_tables(adev);
-		if (ret) {
-			DRM_ERROR("ci_construct_voltage_tables failed\n");
-			return ret;
-		}
-	}
-	if (pi->caps_dynamic_ac_timing) {
-		ret = ci_initialize_mc_reg_table(adev);
-		if (ret)
-			pi->caps_dynamic_ac_timing = false;
-	}
-	if (pi->dynamic_ss)
-		ci_enable_spread_spectrum(adev, true);
-	if (pi->thermal_protection)
-		ci_enable_thermal_protection(adev, true);
-	ci_program_sstp(adev);
-	ci_enable_display_gap(adev);
-	ci_program_vc(adev);
-	ret = ci_upload_firmware(adev);
-	if (ret) {
-		DRM_ERROR("ci_upload_firmware failed\n");
-		return ret;
-	}
-	ret = ci_process_firmware_header(adev);
-	if (ret) {
-		DRM_ERROR("ci_process_firmware_header failed\n");
-		return ret;
-	}
-	ret = ci_initial_switch_from_arb_f0_to_f1(adev);
-	if (ret) {
-		DRM_ERROR("ci_initial_switch_from_arb_f0_to_f1 failed\n");
-		return ret;
-	}
-	ret = ci_init_smc_table(adev);
-	if (ret) {
-		DRM_ERROR("ci_init_smc_table failed\n");
-		return ret;
-	}
-	ret = ci_init_arb_table_index(adev);
-	if (ret) {
-		DRM_ERROR("ci_init_arb_table_index failed\n");
-		return ret;
-	}
-	if (pi->caps_dynamic_ac_timing) {
-		ret = ci_populate_initial_mc_reg_table(adev);
-		if (ret) {
-			DRM_ERROR("ci_populate_initial_mc_reg_table failed\n");
-			return ret;
-		}
-	}
-	ret = ci_populate_pm_base(adev);
-	if (ret) {
-		DRM_ERROR("ci_populate_pm_base failed\n");
-		return ret;
-	}
-	ci_dpm_start_smc(adev);
-	ci_enable_vr_hot_gpio_interrupt(adev);
-	ret = ci_notify_smc_display_change(adev, false);
-	if (ret) {
-		DRM_ERROR("ci_notify_smc_display_change failed\n");
-		return ret;
-	}
-	ci_enable_sclk_control(adev, true);
-	ret = ci_enable_ulv(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_ulv failed\n");
-		return ret;
-	}
-	ret = ci_enable_ds_master_switch(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_ds_master_switch failed\n");
-		return ret;
-	}
-	ret = ci_start_dpm(adev);
-	if (ret) {
-		DRM_ERROR("ci_start_dpm failed\n");
-		return ret;
-	}
-	ret = ci_enable_didt(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_didt failed\n");
-		return ret;
-	}
-	ret = ci_enable_smc_cac(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_smc_cac failed\n");
-		return ret;
-	}
-	ret = ci_enable_power_containment(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_power_containment failed\n");
-		return ret;
-	}
-
-	ret = ci_power_control_set_level(adev);
-	if (ret) {
-		DRM_ERROR("ci_power_control_set_level failed\n");
-		return ret;
-	}
-
-	ci_enable_auto_throttle_source(adev, AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, true);
-
-	ret = ci_enable_thermal_based_sclk_dpm(adev, true);
-	if (ret) {
-		DRM_ERROR("ci_enable_thermal_based_sclk_dpm failed\n");
-		return ret;
-	}
-
-	ci_thermal_start_thermal_controller(adev);
-
-	ci_update_current_ps(adev, boot_ps);
-
-	return 0;
-}
-
-static void ci_dpm_disable(struct amdgpu_device *adev)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps *boot_ps = adev->pm.dpm.boot_ps;
-
-	amdgpu_irq_put(adev, &adev->pm.dpm.thermal.irq,
-		       AMDGPU_THERMAL_IRQ_LOW_TO_HIGH);
-	amdgpu_irq_put(adev, &adev->pm.dpm.thermal.irq,
-		       AMDGPU_THERMAL_IRQ_HIGH_TO_LOW);
-
-	ci_dpm_powergate_uvd(adev, true);
-
-	if (!amdgpu_ci_is_smc_running(adev))
-		return;
-
-	ci_thermal_stop_thermal_controller(adev);
-
-	if (pi->thermal_protection)
-		ci_enable_thermal_protection(adev, false);
-	ci_enable_power_containment(adev, false);
-	ci_enable_smc_cac(adev, false);
-	ci_enable_didt(adev, false);
-	ci_enable_spread_spectrum(adev, false);
-	ci_enable_auto_throttle_source(adev, AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, false);
-	ci_stop_dpm(adev);
-	ci_enable_ds_master_switch(adev, false);
-	ci_enable_ulv(adev, false);
-	ci_clear_vc(adev);
-	ci_reset_to_default(adev);
-	ci_dpm_stop_smc(adev);
-	ci_force_switch_to_arb_f0(adev);
-	ci_enable_thermal_based_sclk_dpm(adev, false);
-
-	ci_update_current_ps(adev, boot_ps);
-}
-
-static int ci_dpm_set_power_state(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps *new_ps = &pi->requested_rps;
-	struct amdgpu_ps *old_ps = &pi->current_rps;
-	int ret;
-
-	ci_find_dpm_states_clocks_in_dpm_table(adev, new_ps);
-	if (pi->pcie_performance_request)
-		ci_request_link_speed_change_before_state_change(adev, new_ps, old_ps);
-	ret = ci_freeze_sclk_mclk_dpm(adev);
-	if (ret) {
-		DRM_ERROR("ci_freeze_sclk_mclk_dpm failed\n");
-		return ret;
-	}
-	ret = ci_populate_and_upload_sclk_mclk_dpm_levels(adev, new_ps);
-	if (ret) {
-		DRM_ERROR("ci_populate_and_upload_sclk_mclk_dpm_levels failed\n");
-		return ret;
-	}
-	ret = ci_generate_dpm_level_enable_mask(adev, new_ps);
-	if (ret) {
-		DRM_ERROR("ci_generate_dpm_level_enable_mask failed\n");
-		return ret;
-	}
-
-	ret = ci_update_vce_dpm(adev, new_ps, old_ps);
-	if (ret) {
-		DRM_ERROR("ci_update_vce_dpm failed\n");
-		return ret;
-	}
-
-	ret = ci_update_sclk_t(adev);
-	if (ret) {
-		DRM_ERROR("ci_update_sclk_t failed\n");
-		return ret;
-	}
-	if (pi->caps_dynamic_ac_timing) {
-		ret = ci_update_and_upload_mc_reg_table(adev);
-		if (ret) {
-			DRM_ERROR("ci_update_and_upload_mc_reg_table failed\n");
-			return ret;
-		}
-	}
-	ret = ci_program_memory_timing_parameters(adev);
-	if (ret) {
-		DRM_ERROR("ci_program_memory_timing_parameters failed\n");
-		return ret;
-	}
-	ret = ci_unfreeze_sclk_mclk_dpm(adev);
-	if (ret) {
-		DRM_ERROR("ci_unfreeze_sclk_mclk_dpm failed\n");
-		return ret;
-	}
-	ret = ci_upload_dpm_level_enable_mask(adev);
-	if (ret) {
-		DRM_ERROR("ci_upload_dpm_level_enable_mask failed\n");
-		return ret;
-	}
-	if (pi->pcie_performance_request)
-		ci_notify_link_speed_change_after_state_change(adev, new_ps, old_ps);
-
-	return 0;
-}
-
-#if 0
-static void ci_dpm_reset_asic(struct amdgpu_device *adev)
-{
-	ci_set_boot_state(adev);
-}
-#endif
-
-static void ci_dpm_display_configuration_changed(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	ci_program_display_gap(adev);
-}
-
-union power_info {
-	struct _ATOM_POWERPLAY_INFO info;
-	struct _ATOM_POWERPLAY_INFO_V2 info_2;
-	struct _ATOM_POWERPLAY_INFO_V3 info_3;
-	struct _ATOM_PPLIB_POWERPLAYTABLE pplib;
-	struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2;
-	struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3;
-};
-
-union pplib_clock_info {
-	struct _ATOM_PPLIB_R600_CLOCK_INFO r600;
-	struct _ATOM_PPLIB_RS780_CLOCK_INFO rs780;
-	struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO evergreen;
-	struct _ATOM_PPLIB_SUMO_CLOCK_INFO sumo;
-	struct _ATOM_PPLIB_SI_CLOCK_INFO si;
-	struct _ATOM_PPLIB_CI_CLOCK_INFO ci;
-};
-
-union pplib_power_state {
-	struct _ATOM_PPLIB_STATE v1;
-	struct _ATOM_PPLIB_STATE_V2 v2;
-};
-
-static void ci_parse_pplib_non_clock_info(struct amdgpu_device *adev,
-					  struct amdgpu_ps *rps,
-					  struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info,
-					  u8 table_rev)
-{
-	rps->caps = le32_to_cpu(non_clock_info->ulCapsAndSettings);
-	rps->class = le16_to_cpu(non_clock_info->usClassification);
-	rps->class2 = le16_to_cpu(non_clock_info->usClassification2);
-
-	if (ATOM_PPLIB_NONCLOCKINFO_VER1 < table_rev) {
-		rps->vclk = le32_to_cpu(non_clock_info->ulVCLK);
-		rps->dclk = le32_to_cpu(non_clock_info->ulDCLK);
-	} else {
-		rps->vclk = 0;
-		rps->dclk = 0;
-	}
-
-	if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT)
-		adev->pm.dpm.boot_ps = rps;
-	if (rps->class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE)
-		adev->pm.dpm.uvd_ps = rps;
-}
-
-static void ci_parse_pplib_clock_info(struct amdgpu_device *adev,
-				      struct amdgpu_ps *rps, int index,
-				      union pplib_clock_info *clock_info)
-{
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *ps = ci_get_ps(rps);
-	struct ci_pl *pl = &ps->performance_levels[index];
-
-	ps->performance_level_count = index + 1;
-
-	pl->sclk = le16_to_cpu(clock_info->ci.usEngineClockLow);
-	pl->sclk |= clock_info->ci.ucEngineClockHigh << 16;
-	pl->mclk = le16_to_cpu(clock_info->ci.usMemoryClockLow);
-	pl->mclk |= clock_info->ci.ucMemoryClockHigh << 16;
-
-	pl->pcie_gen = amdgpu_get_pcie_gen_support(adev,
-						   pi->sys_pcie_mask,
-						   pi->vbios_boot_state.pcie_gen_bootup_value,
-						   clock_info->ci.ucPCIEGen);
-	pl->pcie_lane = amdgpu_get_pcie_lane_support(adev,
-						     pi->vbios_boot_state.pcie_lane_bootup_value,
-						     le16_to_cpu(clock_info->ci.usPCIELane));
-
-	if (rps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) {
-		pi->acpi_pcie_gen = pl->pcie_gen;
-	}
-
-	if (rps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) {
-		pi->ulv.supported = true;
-		pi->ulv.pl = *pl;
-		pi->ulv.cg_ulv_parameter = CISLANDS_CGULVPARAMETER_DFLT;
-	}
-
-	/* patch up boot state */
-	if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT) {
-		pl->mclk = pi->vbios_boot_state.mclk_bootup_value;
-		pl->sclk = pi->vbios_boot_state.sclk_bootup_value;
-		pl->pcie_gen = pi->vbios_boot_state.pcie_gen_bootup_value;
-		pl->pcie_lane = pi->vbios_boot_state.pcie_lane_bootup_value;
-	}
-
-	switch (rps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) {
-	case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY:
-		pi->use_pcie_powersaving_levels = true;
-		if (pi->pcie_gen_powersaving.max < pl->pcie_gen)
-			pi->pcie_gen_powersaving.max = pl->pcie_gen;
-		if (pi->pcie_gen_powersaving.min > pl->pcie_gen)
-			pi->pcie_gen_powersaving.min = pl->pcie_gen;
-		if (pi->pcie_lane_powersaving.max < pl->pcie_lane)
-			pi->pcie_lane_powersaving.max = pl->pcie_lane;
-		if (pi->pcie_lane_powersaving.min > pl->pcie_lane)
-			pi->pcie_lane_powersaving.min = pl->pcie_lane;
-		break;
-	case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE:
-		pi->use_pcie_performance_levels = true;
-		if (pi->pcie_gen_performance.max < pl->pcie_gen)
-			pi->pcie_gen_performance.max = pl->pcie_gen;
-		if (pi->pcie_gen_performance.min > pl->pcie_gen)
-			pi->pcie_gen_performance.min = pl->pcie_gen;
-		if (pi->pcie_lane_performance.max < pl->pcie_lane)
-			pi->pcie_lane_performance.max = pl->pcie_lane;
-		if (pi->pcie_lane_performance.min > pl->pcie_lane)
-			pi->pcie_lane_performance.min = pl->pcie_lane;
-		break;
-	default:
-		break;
-	}
-}
-
-static int ci_parse_power_table(struct amdgpu_device *adev)
-{
-	struct amdgpu_mode_info *mode_info = &adev->mode_info;
-	struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
-	union pplib_power_state *power_state;
-	int i, j, k, non_clock_array_index, clock_array_index;
-	union pplib_clock_info *clock_info;
-	struct _StateArray *state_array;
-	struct _ClockInfoArray *clock_info_array;
-	struct _NonClockInfoArray *non_clock_info_array;
-	union power_info *power_info;
-	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-	u16 data_offset;
-	u8 frev, crev;
-	u8 *power_state_offset;
-	struct ci_ps *ps;
-
-	if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL,
-				   &frev, &crev, &data_offset))
-		return -EINVAL;
-	power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
-
-	amdgpu_add_thermal_controller(adev);
-
-	state_array = (struct _StateArray *)
-		(mode_info->atom_context->bios + data_offset +
-		 le16_to_cpu(power_info->pplib.usStateArrayOffset));
-	clock_info_array = (struct _ClockInfoArray *)
-		(mode_info->atom_context->bios + data_offset +
-		 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset));
-	non_clock_info_array = (struct _NonClockInfoArray *)
-		(mode_info->atom_context->bios + data_offset +
-		 le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset));
-
-	adev->pm.dpm.ps = kcalloc(state_array->ucNumEntries,
-				  sizeof(struct amdgpu_ps),
-				  GFP_KERNEL);
-	if (!adev->pm.dpm.ps)
-		return -ENOMEM;
-	power_state_offset = (u8 *)state_array->states;
-	for (i = 0; i < state_array->ucNumEntries; i++) {
-		u8 *idx;
-		power_state = (union pplib_power_state *)power_state_offset;
-		non_clock_array_index = power_state->v2.nonClockInfoIndex;
-		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
-			&non_clock_info_array->nonClockInfo[non_clock_array_index];
-		ps = kzalloc(sizeof(struct ci_ps), GFP_KERNEL);
-		if (ps == NULL) {
-			kfree(adev->pm.dpm.ps);
-			return -ENOMEM;
-		}
-		adev->pm.dpm.ps[i].ps_priv = ps;
-		ci_parse_pplib_non_clock_info(adev, &adev->pm.dpm.ps[i],
-					      non_clock_info,
-					      non_clock_info_array->ucEntrySize);
-		k = 0;
-		idx = (u8 *)&power_state->v2.clockInfoIndex[0];
-		for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
-			clock_array_index = idx[j];
-			if (clock_array_index >= clock_info_array->ucNumEntries)
-				continue;
-			if (k >= CISLANDS_MAX_HARDWARE_POWERLEVELS)
-				break;
-			clock_info = (union pplib_clock_info *)
-				((u8 *)&clock_info_array->clockInfo[0] +
-				 (clock_array_index * clock_info_array->ucEntrySize));
-			ci_parse_pplib_clock_info(adev,
-						  &adev->pm.dpm.ps[i], k,
-						  clock_info);
-			k++;
-		}
-		power_state_offset += 2 + power_state->v2.ucNumDPMLevels;
-	}
-	adev->pm.dpm.num_ps = state_array->ucNumEntries;
-
-	/* fill in the vce power states */
-	for (i = 0; i < adev->pm.dpm.num_of_vce_states; i++) {
-		u32 sclk, mclk;
-		clock_array_index = adev->pm.dpm.vce_states[i].clk_idx;
-		clock_info = (union pplib_clock_info *)
-			&clock_info_array->clockInfo[clock_array_index * clock_info_array->ucEntrySize];
-		sclk = le16_to_cpu(clock_info->ci.usEngineClockLow);
-		sclk |= clock_info->ci.ucEngineClockHigh << 16;
-		mclk = le16_to_cpu(clock_info->ci.usMemoryClockLow);
-		mclk |= clock_info->ci.ucMemoryClockHigh << 16;
-		adev->pm.dpm.vce_states[i].sclk = sclk;
-		adev->pm.dpm.vce_states[i].mclk = mclk;
-	}
-
-	return 0;
-}
-
-static int ci_get_vbios_boot_values(struct amdgpu_device *adev,
-				    struct ci_vbios_boot_state *boot_state)
-{
-	struct amdgpu_mode_info *mode_info = &adev->mode_info;
-	int index = GetIndexIntoMasterTable(DATA, FirmwareInfo);
-	ATOM_FIRMWARE_INFO_V2_2 *firmware_info;
-	u8 frev, crev;
-	u16 data_offset;
-
-	if (amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL,
-				   &frev, &crev, &data_offset)) {
-		firmware_info =
-			(ATOM_FIRMWARE_INFO_V2_2 *)(mode_info->atom_context->bios +
-						    data_offset);
-		boot_state->mvdd_bootup_value = le16_to_cpu(firmware_info->usBootUpMVDDCVoltage);
-		boot_state->vddc_bootup_value = le16_to_cpu(firmware_info->usBootUpVDDCVoltage);
-		boot_state->vddci_bootup_value = le16_to_cpu(firmware_info->usBootUpVDDCIVoltage);
-		boot_state->pcie_gen_bootup_value = ci_get_current_pcie_speed(adev);
-		boot_state->pcie_lane_bootup_value = ci_get_current_pcie_lane_number(adev);
-		boot_state->sclk_bootup_value = le32_to_cpu(firmware_info->ulDefaultEngineClock);
-		boot_state->mclk_bootup_value = le32_to_cpu(firmware_info->ulDefaultMemoryClock);
-
-		return 0;
-	}
-	return -EINVAL;
-}
-
-static void ci_dpm_fini(struct amdgpu_device *adev)
-{
-	int i;
-
-	for (i = 0; i < adev->pm.dpm.num_ps; i++) {
-		kfree(adev->pm.dpm.ps[i].ps_priv);
-	}
-	kfree(adev->pm.dpm.ps);
-	kfree(adev->pm.dpm.priv);
-	kfree(adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries);
-	amdgpu_free_extended_power_table(adev);
-}
-
-/**
- * ci_dpm_init_microcode - load ucode images from disk
- *
- * @adev: amdgpu_device pointer
- *
- * Use the firmware interface to load the ucode images into
- * the driver (not loaded into hw).
- * Returns 0 on success, error on failure.
- */
-static int ci_dpm_init_microcode(struct amdgpu_device *adev)
-{
-	const char *chip_name;
-	char fw_name[30];
-	int err;
-
-	DRM_DEBUG("\n");
-
-	switch (adev->asic_type) {
-	case CHIP_BONAIRE:
-		if ((adev->pdev->revision == 0x80) ||
-		    (adev->pdev->revision == 0x81) ||
-		    (adev->pdev->device == 0x665f))
-			chip_name = "bonaire_k";
-		else
-			chip_name = "bonaire";
-		break;
-	case CHIP_HAWAII:
-		if (adev->pdev->revision == 0x80)
-			chip_name = "hawaii_k";
-		else
-			chip_name = "hawaii";
-		break;
-	case CHIP_KAVERI:
-	case CHIP_KABINI:
-	case CHIP_MULLINS:
-	default: BUG();
-	}
-
-	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_smc.bin", chip_name);
-	err = request_firmware(&adev->pm.fw, fw_name, adev->dev);
-	if (err)
-		goto out;
-	err = amdgpu_ucode_validate(adev->pm.fw);
-
-out:
-	if (err) {
-		pr_err("cik_smc: Failed to load firmware \"%s\"\n", fw_name);
-		release_firmware(adev->pm.fw);
-		adev->pm.fw = NULL;
-	}
-	return err;
-}
-
-static int ci_dpm_init(struct amdgpu_device *adev)
-{
-	int index = GetIndexIntoMasterTable(DATA, ASIC_InternalSS_Info);
-	SMU7_Discrete_DpmTable *dpm_table;
-	struct amdgpu_gpio_rec gpio;
-	u16 data_offset, size;
-	u8 frev, crev;
-	struct ci_power_info *pi;
-	int ret;
-
-	pi = kzalloc(sizeof(struct ci_power_info), GFP_KERNEL);
-	if (pi == NULL)
-		return -ENOMEM;
-	adev->pm.dpm.priv = pi;
-
-	pi->sys_pcie_mask =
-		adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_MASK;
-
-	pi->force_pcie_gen = AMDGPU_PCIE_GEN_INVALID;
-
-	pi->pcie_gen_performance.max = AMDGPU_PCIE_GEN1;
-	pi->pcie_gen_performance.min = AMDGPU_PCIE_GEN3;
-	pi->pcie_gen_powersaving.max = AMDGPU_PCIE_GEN1;
-	pi->pcie_gen_powersaving.min = AMDGPU_PCIE_GEN3;
-
-	pi->pcie_lane_performance.max = 0;
-	pi->pcie_lane_performance.min = 16;
-	pi->pcie_lane_powersaving.max = 0;
-	pi->pcie_lane_powersaving.min = 16;
-
-	ret = ci_get_vbios_boot_values(adev, &pi->vbios_boot_state);
-	if (ret) {
-		ci_dpm_fini(adev);
-		return ret;
-	}
-
-	ret = amdgpu_get_platform_caps(adev);
-	if (ret) {
-		ci_dpm_fini(adev);
-		return ret;
-	}
-
-	ret = amdgpu_parse_extended_power_table(adev);
-	if (ret) {
-		ci_dpm_fini(adev);
-		return ret;
-	}
-
-	ret = ci_parse_power_table(adev);
-	if (ret) {
-		ci_dpm_fini(adev);
-		return ret;
-	}
-
-	pi->dll_default_on = false;
-	pi->sram_end = SMC_RAM_END;
-
-	pi->activity_target[0] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[1] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[2] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[3] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[4] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[5] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[6] = CISLAND_TARGETACTIVITY_DFLT;
-	pi->activity_target[7] = CISLAND_TARGETACTIVITY_DFLT;
-
-	pi->mclk_activity_target = CISLAND_MCLK_TARGETACTIVITY_DFLT;
-
-	pi->sclk_dpm_key_disabled = 0;
-	pi->mclk_dpm_key_disabled = 0;
-	pi->pcie_dpm_key_disabled = 0;
-	pi->thermal_sclk_dpm_enabled = 0;
-
-	if (adev->powerplay.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)
-		pi->caps_sclk_ds = true;
-	else
-		pi->caps_sclk_ds = false;
-
-	pi->mclk_strobe_mode_threshold = 40000;
-	pi->mclk_stutter_mode_threshold = 40000;
-	pi->mclk_edc_enable_threshold = 40000;
-	pi->mclk_edc_wr_enable_threshold = 40000;
-
-	ci_initialize_powertune_defaults(adev);
-
-	pi->caps_fps = false;
-
-	pi->caps_sclk_throttle_low_notification = false;
-
-	pi->caps_uvd_dpm = true;
-	pi->caps_vce_dpm = true;
-
-	ci_get_leakage_voltages(adev);
-	ci_patch_dependency_tables_with_leakage(adev);
-	ci_set_private_data_variables_based_on_pptable(adev);
-
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries =
-		kcalloc(4,
-			sizeof(struct amdgpu_clock_voltage_dependency_entry),
-			GFP_KERNEL);
-	if (!adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries) {
-		ci_dpm_fini(adev);
-		return -ENOMEM;
-	}
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.count = 4;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].clk = 0;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].v = 0;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[1].clk = 36000;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[1].v = 720;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[2].clk = 54000;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[2].v = 810;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[3].clk = 72000;
-	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[3].v = 900;
-
-	adev->pm.dpm.dyn_state.mclk_sclk_ratio = 4;
-	adev->pm.dpm.dyn_state.sclk_mclk_delta = 15000;
-	adev->pm.dpm.dyn_state.vddc_vddci_delta = 200;
-
-	adev->pm.dpm.dyn_state.valid_sclk_values.count = 0;
-	adev->pm.dpm.dyn_state.valid_sclk_values.values = NULL;
-	adev->pm.dpm.dyn_state.valid_mclk_values.count = 0;
-	adev->pm.dpm.dyn_state.valid_mclk_values.values = NULL;
-
-	if (adev->asic_type == CHIP_HAWAII) {
-		pi->thermal_temp_setting.temperature_low = 94500;
-		pi->thermal_temp_setting.temperature_high = 95000;
-		pi->thermal_temp_setting.temperature_shutdown = 104000;
-	} else {
-		pi->thermal_temp_setting.temperature_low = 99500;
-		pi->thermal_temp_setting.temperature_high = 100000;
-		pi->thermal_temp_setting.temperature_shutdown = 104000;
-	}
-
-	pi->uvd_enabled = false;
-
-	dpm_table = &pi->smc_state_table;
-
-	gpio = amdgpu_atombios_lookup_gpio(adev, VDDC_VRHOT_GPIO_PINID);
-	if (gpio.valid) {
-		dpm_table->VRHotGpio = gpio.shift;
-		adev->pm.dpm.platform_caps |= ATOM_PP_PLATFORM_CAP_REGULATOR_HOT;
-	} else {
-		dpm_table->VRHotGpio = CISLANDS_UNUSED_GPIO_PIN;
-		adev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_REGULATOR_HOT;
-	}
-
-	gpio = amdgpu_atombios_lookup_gpio(adev, PP_AC_DC_SWITCH_GPIO_PINID);
-	if (gpio.valid) {
-		dpm_table->AcDcGpio = gpio.shift;
-		adev->pm.dpm.platform_caps |= ATOM_PP_PLATFORM_CAP_HARDWAREDC;
-	} else {
-		dpm_table->AcDcGpio = CISLANDS_UNUSED_GPIO_PIN;
-		adev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_HARDWAREDC;
-	}
-
-	gpio = amdgpu_atombios_lookup_gpio(adev, VDDC_PCC_GPIO_PINID);
-	if (gpio.valid) {
-		u32 tmp = RREG32_SMC(ixCNB_PWRMGT_CNTL);
-
-		switch (gpio.shift) {
-		case 0:
-			tmp &= ~CNB_PWRMGT_CNTL__GNB_SLOW_MODE_MASK;
-			tmp |= 1 << CNB_PWRMGT_CNTL__GNB_SLOW_MODE__SHIFT;
-			break;
-		case 1:
-			tmp &= ~CNB_PWRMGT_CNTL__GNB_SLOW_MODE_MASK;
-			tmp |= 2 << CNB_PWRMGT_CNTL__GNB_SLOW_MODE__SHIFT;
-			break;
-		case 2:
-			tmp |= CNB_PWRMGT_CNTL__GNB_SLOW_MASK;
-			break;
-		case 3:
-			tmp |= CNB_PWRMGT_CNTL__FORCE_NB_PS1_MASK;
-			break;
-		case 4:
-			tmp |= CNB_PWRMGT_CNTL__DPM_ENABLED_MASK;
-			break;
-		default:
-			DRM_INFO("Invalid PCC GPIO: %u!\n", gpio.shift);
-			break;
-		}
-		WREG32_SMC(ixCNB_PWRMGT_CNTL, tmp);
-	}
-
-	pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_NONE;
-	pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_NONE;
-	pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_NONE;
-	if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_VDDC, VOLTAGE_OBJ_GPIO_LUT))
-		pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
-	else if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_VDDC, VOLTAGE_OBJ_SVID2))
-		pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
-
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL) {
-		if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_VDDCI, VOLTAGE_OBJ_GPIO_LUT))
-			pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
-		else if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_VDDCI, VOLTAGE_OBJ_SVID2))
-			pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
-		else
-			adev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL;
-	}
-
-	if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_MVDDCONTROL) {
-		if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_MVDDC, VOLTAGE_OBJ_GPIO_LUT))
-			pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
-		else if (amdgpu_atombios_is_voltage_gpio(adev, VOLTAGE_TYPE_MVDDC, VOLTAGE_OBJ_SVID2))
-			pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
-		else
-			adev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_MVDDCONTROL;
-	}
-
-	pi->vddc_phase_shed_control = true;
-
-#if defined(CONFIG_ACPI)
-	pi->pcie_performance_request =
-		amdgpu_acpi_is_pcie_performance_request_supported(adev);
-#else
-	pi->pcie_performance_request = false;
-#endif
-
-	if (amdgpu_atom_parse_data_header(adev->mode_info.atom_context, index, &size,
-				   &frev, &crev, &data_offset)) {
-		pi->caps_sclk_ss_support = true;
-		pi->caps_mclk_ss_support = true;
-		pi->dynamic_ss = true;
-	} else {
-		pi->caps_sclk_ss_support = false;
-		pi->caps_mclk_ss_support = false;
-		pi->dynamic_ss = true;
-	}
-
-	if (adev->pm.int_thermal_type != THERMAL_TYPE_NONE)
-		pi->thermal_protection = true;
-	else
-		pi->thermal_protection = false;
-
-	pi->caps_dynamic_ac_timing = true;
-
-	pi->uvd_power_gated = true;
-
-	/* make sure dc limits are valid */
-	if ((adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.sclk == 0) ||
-	    (adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.mclk == 0))
-		adev->pm.dpm.dyn_state.max_clock_voltage_on_dc =
-			adev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
-
-	pi->fan_ctrl_is_in_default_mode = true;
-
-	return 0;
-}
-
-static void
-ci_dpm_debugfs_print_current_performance_level(void *handle,
-					       struct seq_file *m)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct amdgpu_ps *rps = &pi->current_rps;
-	u32 sclk = ci_get_average_sclk_freq(adev);
-	u32 mclk = ci_get_average_mclk_freq(adev);
-	u32 activity_percent = 50;
-	int ret;
-
-	ret = ci_read_smc_soft_register(adev, offsetof(SMU7_SoftRegisters, AverageGraphicsA),
-					&activity_percent);
-
-	if (ret == 0) {
-		activity_percent += 0x80;
-		activity_percent >>= 8;
-		activity_percent = activity_percent > 100 ? 100 : activity_percent;
-	}
-
-	seq_printf(m, "uvd %sabled\n", pi->uvd_power_gated ? "dis" : "en");
-	seq_printf(m, "vce %sabled\n", rps->vce_active ? "en" : "dis");
-	seq_printf(m, "power level avg    sclk: %u mclk: %u\n",
-		   sclk, mclk);
-	seq_printf(m, "GPU load: %u %%\n", activity_percent);
-}
-
-static void ci_dpm_print_power_state(void *handle, void *current_ps)
-{
-	struct amdgpu_ps *rps = (struct amdgpu_ps *)current_ps;
-	struct ci_ps *ps = ci_get_ps(rps);
-	struct ci_pl *pl;
-	int i;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	amdgpu_dpm_print_class_info(rps->class, rps->class2);
-	amdgpu_dpm_print_cap_info(rps->caps);
-	printk("\tuvd    vclk: %d dclk: %d\n", rps->vclk, rps->dclk);
-	for (i = 0; i < ps->performance_level_count; i++) {
-		pl = &ps->performance_levels[i];
-		printk("\t\tpower level %d    sclk: %u mclk: %u pcie gen: %u pcie lanes: %u\n",
-		       i, pl->sclk, pl->mclk, pl->pcie_gen + 1, pl->pcie_lane);
-	}
-	amdgpu_dpm_print_ps_status(adev, rps);
-}
-
-static inline bool ci_are_power_levels_equal(const struct ci_pl *ci_cpl1,
-						const struct ci_pl *ci_cpl2)
-{
-	return ((ci_cpl1->mclk == ci_cpl2->mclk) &&
-		  (ci_cpl1->sclk == ci_cpl2->sclk) &&
-		  (ci_cpl1->pcie_gen == ci_cpl2->pcie_gen) &&
-		  (ci_cpl1->pcie_lane == ci_cpl2->pcie_lane));
-}
-
-static int ci_check_state_equal(void *handle,
-				void *current_ps,
-				void *request_ps,
-				bool *equal)
-{
-	struct ci_ps *ci_cps;
-	struct ci_ps *ci_rps;
-	int i;
-	struct amdgpu_ps *cps = (struct amdgpu_ps *)current_ps;
-	struct amdgpu_ps *rps = (struct amdgpu_ps *)request_ps;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (adev == NULL || cps == NULL || rps == NULL || equal == NULL)
-		return -EINVAL;
-
-	ci_cps = ci_get_ps((struct amdgpu_ps *)cps);
-	ci_rps = ci_get_ps((struct amdgpu_ps *)rps);
-
-	if (ci_cps == NULL) {
-		*equal = false;
-		return 0;
-	}
-
-	if (ci_cps->performance_level_count != ci_rps->performance_level_count) {
-
-		*equal = false;
-		return 0;
-	}
-
-	for (i = 0; i < ci_cps->performance_level_count; i++) {
-		if (!ci_are_power_levels_equal(&(ci_cps->performance_levels[i]),
-					&(ci_rps->performance_levels[i]))) {
-			*equal = false;
-			return 0;
-		}
-	}
-
-	/* If all performance levels are the same try to use the UVD clocks to break the tie.*/
-	*equal = ((cps->vclk == rps->vclk) && (cps->dclk == rps->dclk));
-	*equal &= ((cps->evclk == rps->evclk) && (cps->ecclk == rps->ecclk));
-
-	return 0;
-}
-
-static u32 ci_dpm_get_sclk(void *handle, bool low)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *requested_state = ci_get_ps(&pi->requested_rps);
-
-	if (low)
-		return requested_state->performance_levels[0].sclk;
-	else
-		return requested_state->performance_levels[requested_state->performance_level_count - 1].sclk;
-}
-
-static u32 ci_dpm_get_mclk(void *handle, bool low)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *requested_state = ci_get_ps(&pi->requested_rps);
-
-	if (low)
-		return requested_state->performance_levels[0].mclk;
-	else
-		return requested_state->performance_levels[requested_state->performance_level_count - 1].mclk;
-}
-
-/* get temperature in millidegrees */
-static int ci_dpm_get_temp(void *handle)
-{
-	u32 temp;
-	int actual_temp = 0;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	temp = (RREG32_SMC(ixCG_MULT_THERMAL_STATUS) & CG_MULT_THERMAL_STATUS__CTF_TEMP_MASK) >>
-		CG_MULT_THERMAL_STATUS__CTF_TEMP__SHIFT;
-
-	if (temp & 0x200)
-		actual_temp = 255;
-	else
-		actual_temp = temp & 0x1ff;
-
-	actual_temp = actual_temp * 1000;
-
-	return actual_temp;
-}
-
-static int ci_set_temperature_range(struct amdgpu_device *adev)
-{
-	int ret;
-
-	ret = ci_thermal_enable_alert(adev, false);
-	if (ret)
-		return ret;
-	ret = ci_thermal_set_temperature_range(adev, CISLANDS_TEMP_RANGE_MIN,
-					       CISLANDS_TEMP_RANGE_MAX);
-	if (ret)
-		return ret;
-	ret = ci_thermal_enable_alert(adev, true);
-	if (ret)
-		return ret;
-	return ret;
-}
-
-static int ci_dpm_early_init(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	adev->powerplay.pp_funcs = &ci_dpm_funcs;
-	adev->powerplay.pp_handle = adev;
-	ci_dpm_set_irq_funcs(adev);
-
-	return 0;
-}
-
-static int ci_dpm_late_init(void *handle)
-{
-	int ret;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (!adev->pm.dpm_enabled)
-		return 0;
-
-	/* init the sysfs and debugfs files late */
-	ret = amdgpu_pm_sysfs_init(adev);
-	if (ret)
-		return ret;
-
-	ret = ci_set_temperature_range(adev);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static int ci_dpm_sw_init(void *handle)
-{
-	int ret;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	ret = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 230,
-				&adev->pm.dpm.thermal.irq);
-	if (ret)
-		return ret;
-
-	ret = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 231,
-				&adev->pm.dpm.thermal.irq);
-	if (ret)
-		return ret;
-
-	/* default to balanced state */
-	adev->pm.dpm.state = POWER_STATE_TYPE_BALANCED;
-	adev->pm.dpm.user_state = POWER_STATE_TYPE_BALANCED;
-	adev->pm.dpm.forced_level = AMD_DPM_FORCED_LEVEL_AUTO;
-	adev->pm.default_sclk = adev->clock.default_sclk;
-	adev->pm.default_mclk = adev->clock.default_mclk;
-	adev->pm.current_sclk = adev->clock.default_sclk;
-	adev->pm.current_mclk = adev->clock.default_mclk;
-	adev->pm.int_thermal_type = THERMAL_TYPE_NONE;
-
-	ret = ci_dpm_init_microcode(adev);
-	if (ret)
-		return ret;
-
-	if (amdgpu_dpm == 0)
-		return 0;
-
-	INIT_WORK(&adev->pm.dpm.thermal.work, amdgpu_dpm_thermal_work_handler);
-	mutex_lock(&adev->pm.mutex);
-	ret = ci_dpm_init(adev);
-	if (ret)
-		goto dpm_failed;
-	adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps;
-	if (amdgpu_dpm == 1)
-		amdgpu_pm_print_power_states(adev);
-	mutex_unlock(&adev->pm.mutex);
-	DRM_INFO("amdgpu: dpm initialized\n");
-
-	return 0;
-
-dpm_failed:
-	ci_dpm_fini(adev);
-	mutex_unlock(&adev->pm.mutex);
-	DRM_ERROR("amdgpu: dpm initialization failed\n");
-	return ret;
-}
-
-static int ci_dpm_sw_fini(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	flush_work(&adev->pm.dpm.thermal.work);
-
-	mutex_lock(&adev->pm.mutex);
-	ci_dpm_fini(adev);
-	mutex_unlock(&adev->pm.mutex);
-
-	release_firmware(adev->pm.fw);
-	adev->pm.fw = NULL;
-
-	return 0;
-}
-
-static int ci_dpm_hw_init(void *handle)
-{
-	int ret;
-
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (!amdgpu_dpm) {
-		ret = ci_upload_firmware(adev);
-		if (ret) {
-			DRM_ERROR("ci_upload_firmware failed\n");
-			return ret;
-		}
-		ci_dpm_start_smc(adev);
-		return 0;
-	}
-
-	mutex_lock(&adev->pm.mutex);
-	ci_dpm_setup_asic(adev);
-	ret = ci_dpm_enable(adev);
-	if (ret)
-		adev->pm.dpm_enabled = false;
-	else
-		adev->pm.dpm_enabled = true;
-	mutex_unlock(&adev->pm.mutex);
-
-	return ret;
-}
-
-static int ci_dpm_hw_fini(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (adev->pm.dpm_enabled) {
-		mutex_lock(&adev->pm.mutex);
-		ci_dpm_disable(adev);
-		mutex_unlock(&adev->pm.mutex);
-	} else {
-		ci_dpm_stop_smc(adev);
-	}
-
-	return 0;
-}
-
-static int ci_dpm_suspend(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (adev->pm.dpm_enabled) {
-		mutex_lock(&adev->pm.mutex);
-		amdgpu_irq_put(adev, &adev->pm.dpm.thermal.irq,
-			       AMDGPU_THERMAL_IRQ_LOW_TO_HIGH);
-		amdgpu_irq_put(adev, &adev->pm.dpm.thermal.irq,
-			       AMDGPU_THERMAL_IRQ_HIGH_TO_LOW);
-		adev->pm.dpm.last_user_state = adev->pm.dpm.user_state;
-		adev->pm.dpm.last_state = adev->pm.dpm.state;
-		adev->pm.dpm.user_state = POWER_STATE_TYPE_INTERNAL_BOOT;
-		adev->pm.dpm.state = POWER_STATE_TYPE_INTERNAL_BOOT;
-		mutex_unlock(&adev->pm.mutex);
-		amdgpu_pm_compute_clocks(adev);
-
-	}
-
-	return 0;
-}
-
-static int ci_dpm_resume(void *handle)
-{
-	int ret;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (adev->pm.dpm_enabled) {
-		/* asic init will reset to the boot state */
-		mutex_lock(&adev->pm.mutex);
-		ci_dpm_setup_asic(adev);
-		ret = ci_dpm_enable(adev);
-		if (ret)
-			adev->pm.dpm_enabled = false;
-		else
-			adev->pm.dpm_enabled = true;
-		adev->pm.dpm.user_state = adev->pm.dpm.last_user_state;
-		adev->pm.dpm.state = adev->pm.dpm.last_state;
-		mutex_unlock(&adev->pm.mutex);
-		if (adev->pm.dpm_enabled)
-			amdgpu_pm_compute_clocks(adev);
-	}
-	return 0;
-}
-
-static bool ci_dpm_is_idle(void *handle)
-{
-	/* XXX */
-	return true;
-}
-
-static int ci_dpm_wait_for_idle(void *handle)
-{
-	/* XXX */
-	return 0;
-}
-
-static int ci_dpm_soft_reset(void *handle)
-{
-	return 0;
-}
-
-static int ci_dpm_set_interrupt_state(struct amdgpu_device *adev,
-				      struct amdgpu_irq_src *source,
-				      unsigned type,
-				      enum amdgpu_interrupt_state state)
-{
-	u32 cg_thermal_int;
-
-	switch (type) {
-	case AMDGPU_THERMAL_IRQ_LOW_TO_HIGH:
-		switch (state) {
-		case AMDGPU_IRQ_STATE_DISABLE:
-			cg_thermal_int = RREG32_SMC(ixCG_THERMAL_INT);
-			cg_thermal_int |= CG_THERMAL_INT_CTRL__THERM_INTH_MASK_MASK;
-			WREG32_SMC(ixCG_THERMAL_INT, cg_thermal_int);
-			break;
-		case AMDGPU_IRQ_STATE_ENABLE:
-			cg_thermal_int = RREG32_SMC(ixCG_THERMAL_INT);
-			cg_thermal_int &= ~CG_THERMAL_INT_CTRL__THERM_INTH_MASK_MASK;
-			WREG32_SMC(ixCG_THERMAL_INT, cg_thermal_int);
-			break;
-		default:
-			break;
-		}
-		break;
-
-	case AMDGPU_THERMAL_IRQ_HIGH_TO_LOW:
-		switch (state) {
-		case AMDGPU_IRQ_STATE_DISABLE:
-			cg_thermal_int = RREG32_SMC(ixCG_THERMAL_INT);
-			cg_thermal_int |= CG_THERMAL_INT_CTRL__THERM_INTL_MASK_MASK;
-			WREG32_SMC(ixCG_THERMAL_INT, cg_thermal_int);
-			break;
-		case AMDGPU_IRQ_STATE_ENABLE:
-			cg_thermal_int = RREG32_SMC(ixCG_THERMAL_INT);
-			cg_thermal_int &= ~CG_THERMAL_INT_CTRL__THERM_INTL_MASK_MASK;
-			WREG32_SMC(ixCG_THERMAL_INT, cg_thermal_int);
-			break;
-		default:
-			break;
-		}
-		break;
-
-	default:
-		break;
-	}
-	return 0;
-}
-
-static int ci_dpm_process_interrupt(struct amdgpu_device *adev,
-				    struct amdgpu_irq_src *source,
-				    struct amdgpu_iv_entry *entry)
-{
-	bool queue_thermal = false;
-
-	if (entry == NULL)
-		return -EINVAL;
-
-	switch (entry->src_id) {
-	case 230: /* thermal low to high */
-		DRM_DEBUG("IH: thermal low to high\n");
-		adev->pm.dpm.thermal.high_to_low = false;
-		queue_thermal = true;
-		break;
-	case 231: /* thermal high to low */
-		DRM_DEBUG("IH: thermal high to low\n");
-		adev->pm.dpm.thermal.high_to_low = true;
-		queue_thermal = true;
-		break;
-	default:
-		break;
-	}
-
-	if (queue_thermal)
-		schedule_work(&adev->pm.dpm.thermal.work);
-
-	return 0;
-}
-
-static int ci_dpm_set_clockgating_state(void *handle,
-					  enum amd_clockgating_state state)
-{
-	return 0;
-}
-
-static int ci_dpm_set_powergating_state(void *handle,
-					  enum amd_powergating_state state)
-{
-	return 0;
-}
-
-static int ci_dpm_print_clock_levels(void *handle,
-		enum pp_clock_type type, char *buf)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_single_dpm_table *sclk_table = &pi->dpm_table.sclk_table;
-	struct ci_single_dpm_table *mclk_table = &pi->dpm_table.mclk_table;
-	struct ci_single_dpm_table *pcie_table = &pi->dpm_table.pcie_speed_table;
-
-	int i, now, size = 0;
-	uint32_t clock, pcie_speed;
-
-	switch (type) {
-	case PP_SCLK:
-		amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_API_GetSclkFrequency);
-		clock = RREG32(mmSMC_MSG_ARG_0);
-
-		for (i = 0; i < sclk_table->count; i++) {
-			if (clock > sclk_table->dpm_levels[i].value)
-				continue;
-			break;
-		}
-		now = i;
-
-		for (i = 0; i < sclk_table->count; i++)
-			size += sprintf(buf + size, "%d: %uMhz %s\n",
-					i, sclk_table->dpm_levels[i].value / 100,
-					(i == now) ? "*" : "");
-		break;
-	case PP_MCLK:
-		amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_API_GetMclkFrequency);
-		clock = RREG32(mmSMC_MSG_ARG_0);
-
-		for (i = 0; i < mclk_table->count; i++) {
-			if (clock > mclk_table->dpm_levels[i].value)
-				continue;
-			break;
-		}
-		now = i;
-
-		for (i = 0; i < mclk_table->count; i++)
-			size += sprintf(buf + size, "%d: %uMhz %s\n",
-					i, mclk_table->dpm_levels[i].value / 100,
-					(i == now) ? "*" : "");
-		break;
-	case PP_PCIE:
-		pcie_speed = ci_get_current_pcie_speed(adev);
-		for (i = 0; i < pcie_table->count; i++) {
-			if (pcie_speed != pcie_table->dpm_levels[i].value)
-				continue;
-			break;
-		}
-		now = i;
-
-		for (i = 0; i < pcie_table->count; i++)
-			size += sprintf(buf + size, "%d: %s %s\n", i,
-					(pcie_table->dpm_levels[i].value == 0) ? "2.5GT/s, x1" :
-					(pcie_table->dpm_levels[i].value == 1) ? "5.0GT/s, x16" :
-					(pcie_table->dpm_levels[i].value == 2) ? "8.0GT/s, x16" : "",
-					(i == now) ? "*" : "");
-		break;
-	default:
-		break;
-	}
-
-	return size;
-}
-
-static int ci_dpm_force_clock_level(void *handle,
-		enum pp_clock_type type, uint32_t mask)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-
-	if (adev->pm.dpm.forced_level != AMD_DPM_FORCED_LEVEL_MANUAL)
-		return -EINVAL;
-
-	if (mask == 0)
-		return -EINVAL;
-
-	switch (type) {
-	case PP_SCLK:
-		if (!pi->sclk_dpm_key_disabled)
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-					PPSMC_MSG_SCLKDPM_SetEnabledMask,
-					pi->dpm_level_enable_mask.sclk_dpm_enable_mask & mask);
-		break;
-
-	case PP_MCLK:
-		if (!pi->mclk_dpm_key_disabled)
-			amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-					PPSMC_MSG_MCLKDPM_SetEnabledMask,
-					pi->dpm_level_enable_mask.mclk_dpm_enable_mask & mask);
-		break;
-
-	case PP_PCIE:
-	{
-		uint32_t tmp = mask & pi->dpm_level_enable_mask.pcie_dpm_enable_mask;
-
-		if (!pi->pcie_dpm_key_disabled) {
-			if (fls(tmp) != ffs(tmp))
-				amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_PCIeDPM_UnForceLevel);
-			else
-				amdgpu_ci_send_msg_to_smc_with_parameter(adev,
-					PPSMC_MSG_PCIeDPM_ForceLevel,
-					fls(tmp) - 1);
-		}
-		break;
-	}
-	default:
-		break;
-	}
-
-	return 0;
-}
-
-static int ci_dpm_get_sclk_od(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_single_dpm_table *sclk_table = &(pi->dpm_table.sclk_table);
-	struct ci_single_dpm_table *golden_sclk_table =
-			&(pi->golden_dpm_table.sclk_table);
-	int value;
-
-	value = (sclk_table->dpm_levels[sclk_table->count - 1].value -
-			golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value) *
-			100 /
-			golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value;
-
-	return value;
-}
-
-static int ci_dpm_set_sclk_od(void *handle, uint32_t value)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *ps = ci_get_ps(adev->pm.dpm.requested_ps);
-	struct ci_single_dpm_table *golden_sclk_table =
-			&(pi->golden_dpm_table.sclk_table);
-
-	if (value > 20)
-		value = 20;
-
-	ps->performance_levels[ps->performance_level_count - 1].sclk =
-			golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value *
-			value / 100 +
-			golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value;
-
-	return 0;
-}
-
-static int ci_dpm_get_mclk_od(void *handle)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_single_dpm_table *mclk_table = &(pi->dpm_table.mclk_table);
-	struct ci_single_dpm_table *golden_mclk_table =
-			&(pi->golden_dpm_table.mclk_table);
-	int value;
-
-	value = (mclk_table->dpm_levels[mclk_table->count - 1].value -
-			golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value) *
-			100 /
-			golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value;
-
-	return value;
-}
-
-static int ci_dpm_set_mclk_od(void *handle, uint32_t value)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct ci_power_info *pi = ci_get_pi(adev);
-	struct ci_ps *ps = ci_get_ps(adev->pm.dpm.requested_ps);
-	struct ci_single_dpm_table *golden_mclk_table =
-			&(pi->golden_dpm_table.mclk_table);
-
-	if (value > 20)
-		value = 20;
-
-	ps->performance_levels[ps->performance_level_count - 1].mclk =
-			golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value *
-			value / 100 +
-			golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value;
-
-	return 0;
-}
-
-static int ci_dpm_read_sensor(void *handle, int idx,
-			      void *value, int *size)
-{
-	u32 activity_percent = 50;
-	int ret;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	/* size must be at least 4 bytes for all sensors */
-	if (*size < 4)
-		return -EINVAL;
-
-	switch (idx) {
-	case AMDGPU_PP_SENSOR_GFX_SCLK:
-		*((uint32_t *)value) = ci_get_average_sclk_freq(adev);
-		*size = 4;
-		return 0;
-	case AMDGPU_PP_SENSOR_GFX_MCLK:
-		*((uint32_t *)value) = ci_get_average_mclk_freq(adev);
-		*size = 4;
-		return 0;
-	case AMDGPU_PP_SENSOR_GPU_TEMP:
-		*((uint32_t *)value) = ci_dpm_get_temp(adev);
-		*size = 4;
-		return 0;
-	case AMDGPU_PP_SENSOR_GPU_LOAD:
-		ret = ci_read_smc_soft_register(adev,
-						offsetof(SMU7_SoftRegisters,
-							 AverageGraphicsA),
-						&activity_percent);
-		if (ret == 0) {
-			activity_percent += 0x80;
-			activity_percent >>= 8;
-			activity_percent =
-				activity_percent > 100 ? 100 : activity_percent;
-		}
-		*((uint32_t *)value) = activity_percent;
-		*size = 4;
-		return 0;
-	default:
-		return -EINVAL;
-	}
-}
-
-static int ci_set_powergating_by_smu(void *handle,
-				uint32_t block_type, bool gate)
-{
-	switch (block_type) {
-	case AMD_IP_BLOCK_TYPE_UVD:
-		ci_dpm_powergate_uvd(handle, gate);
-		break;
-	default:
-		break;
-	}
-	return 0;
-}
-
-static const struct amd_ip_funcs ci_dpm_ip_funcs = {
-	.name = "ci_dpm",
-	.early_init = ci_dpm_early_init,
-	.late_init = ci_dpm_late_init,
-	.sw_init = ci_dpm_sw_init,
-	.sw_fini = ci_dpm_sw_fini,
-	.hw_init = ci_dpm_hw_init,
-	.hw_fini = ci_dpm_hw_fini,
-	.suspend = ci_dpm_suspend,
-	.resume = ci_dpm_resume,
-	.is_idle = ci_dpm_is_idle,
-	.wait_for_idle = ci_dpm_wait_for_idle,
-	.soft_reset = ci_dpm_soft_reset,
-	.set_clockgating_state = ci_dpm_set_clockgating_state,
-	.set_powergating_state = ci_dpm_set_powergating_state,
-};
-
-const struct amdgpu_ip_block_version ci_smu_ip_block =
-{
-	.type = AMD_IP_BLOCK_TYPE_SMC,
-	.major = 7,
-	.minor = 0,
-	.rev = 0,
-	.funcs = &ci_dpm_ip_funcs,
-};
-
-static const struct amd_pm_funcs ci_dpm_funcs = {
-	.pre_set_power_state = &ci_dpm_pre_set_power_state,
-	.set_power_state = &ci_dpm_set_power_state,
-	.post_set_power_state = &ci_dpm_post_set_power_state,
-	.display_configuration_changed = &ci_dpm_display_configuration_changed,
-	.get_sclk = &ci_dpm_get_sclk,
-	.get_mclk = &ci_dpm_get_mclk,
-	.print_power_state = &ci_dpm_print_power_state,
-	.debugfs_print_current_performance_level = &ci_dpm_debugfs_print_current_performance_level,
-	.force_performance_level = &ci_dpm_force_performance_level,
-	.vblank_too_short = &ci_dpm_vblank_too_short,
-	.set_powergating_by_smu = &ci_set_powergating_by_smu,
-	.set_fan_control_mode = &ci_dpm_set_fan_control_mode,
-	.get_fan_control_mode = &ci_dpm_get_fan_control_mode,
-	.set_fan_speed_percent = &ci_dpm_set_fan_speed_percent,
-	.get_fan_speed_percent = &ci_dpm_get_fan_speed_percent,
-	.print_clock_levels = ci_dpm_print_clock_levels,
-	.force_clock_level = ci_dpm_force_clock_level,
-	.get_sclk_od = ci_dpm_get_sclk_od,
-	.set_sclk_od = ci_dpm_set_sclk_od,
-	.get_mclk_od = ci_dpm_get_mclk_od,
-	.set_mclk_od = ci_dpm_set_mclk_od,
-	.check_state_equal = ci_check_state_equal,
-	.get_vce_clock_state = amdgpu_get_vce_clock_state,
-	.read_sensor = ci_dpm_read_sensor,
-};
-
-static const struct amdgpu_irq_src_funcs ci_dpm_irq_funcs = {
-	.set = ci_dpm_set_interrupt_state,
-	.process = ci_dpm_process_interrupt,
-};
-
-static void ci_dpm_set_irq_funcs(struct amdgpu_device *adev)
-{
-	adev->pm.dpm.thermal.irq.num_types = AMDGPU_THERMAL_IRQ_LAST;
-	adev->pm.dpm.thermal.irq.funcs = &ci_dpm_irq_funcs;
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.h b/drivers/gpu/drm/amd/amdgpu/ci_dpm.h
deleted file mode 100644
index 91be2996ae7c..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.h
+++ /dev/null
@@ -1,349 +0,0 @@
-/*
- * Copyright 2013 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 __CI_DPM_H__
-#define __CI_DPM_H__
-
-#include "amdgpu_atombios.h"
-#include "ppsmc.h"
-
-#define SMU__NUM_SCLK_DPM_STATE  8
-#define SMU__NUM_MCLK_DPM_LEVELS 6
-#define SMU__NUM_LCLK_DPM_LEVELS 8
-#define SMU__NUM_PCIE_DPM_LEVELS 8
-#include "smu7_discrete.h"
-
-#define CISLANDS_MAX_HARDWARE_POWERLEVELS 2
-
-#define CISLANDS_UNUSED_GPIO_PIN 0x7F
-
-struct ci_pl {
-	u32 mclk;
-	u32 sclk;
-	enum amdgpu_pcie_gen pcie_gen;
-	u16 pcie_lane;
-};
-
-struct ci_ps {
-	u16 performance_level_count;
-	bool dc_compatible;
-	u32 sclk_t;
-	struct ci_pl performance_levels[CISLANDS_MAX_HARDWARE_POWERLEVELS];
-};
-
-struct ci_dpm_level {
-	bool enabled;
-	u32 value;
-	u32 param1;
-};
-
-#define CISLAND_MAX_DEEPSLEEP_DIVIDER_ID 5
-#define MAX_REGULAR_DPM_NUMBER 8
-#define CISLAND_MINIMUM_ENGINE_CLOCK 800
-
-struct ci_single_dpm_table {
-	u32 count;
-	struct ci_dpm_level dpm_levels[MAX_REGULAR_DPM_NUMBER];
-};
-
-struct ci_dpm_table {
-	struct ci_single_dpm_table sclk_table;
-	struct ci_single_dpm_table mclk_table;
-	struct ci_single_dpm_table pcie_speed_table;
-	struct ci_single_dpm_table vddc_table;
-	struct ci_single_dpm_table vddci_table;
-	struct ci_single_dpm_table mvdd_table;
-};
-
-struct ci_mc_reg_entry {
-	u32 mclk_max;
-	u32 mc_data[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
-};
-
-struct ci_mc_reg_table {
-	u8 last;
-	u8 num_entries;
-	u16 valid_flag;
-	struct ci_mc_reg_entry mc_reg_table_entry[MAX_AC_TIMING_ENTRIES];
-	SMU7_Discrete_MCRegisterAddress mc_reg_address[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
-};
-
-struct ci_ulv_parm
-{
-	bool supported;
-	u32 cg_ulv_parameter;
-	u32 volt_change_delay;
-	struct ci_pl pl;
-};
-
-#define CISLANDS_MAX_LEAKAGE_COUNT  8
-
-struct ci_leakage_voltage {
-	u16 count;
-	u16 leakage_id[CISLANDS_MAX_LEAKAGE_COUNT];
-	u16 actual_voltage[CISLANDS_MAX_LEAKAGE_COUNT];
-};
-
-struct ci_dpm_level_enable_mask {
-	u32 uvd_dpm_enable_mask;
-	u32 vce_dpm_enable_mask;
-	u32 acp_dpm_enable_mask;
-	u32 samu_dpm_enable_mask;
-	u32 sclk_dpm_enable_mask;
-	u32 mclk_dpm_enable_mask;
-	u32 pcie_dpm_enable_mask;
-};
-
-struct ci_vbios_boot_state
-{
-	u16 mvdd_bootup_value;
-	u16 vddc_bootup_value;
-	u16 vddci_bootup_value;
-	u32 sclk_bootup_value;
-	u32 mclk_bootup_value;
-	u16 pcie_gen_bootup_value;
-	u16 pcie_lane_bootup_value;
-};
-
-struct ci_clock_registers {
-	u32 cg_spll_func_cntl;
-	u32 cg_spll_func_cntl_2;
-	u32 cg_spll_func_cntl_3;
-	u32 cg_spll_func_cntl_4;
-	u32 cg_spll_spread_spectrum;
-	u32 cg_spll_spread_spectrum_2;
-	u32 dll_cntl;
-	u32 mclk_pwrmgt_cntl;
-	u32 mpll_ad_func_cntl;
-	u32 mpll_dq_func_cntl;
-	u32 mpll_func_cntl;
-	u32 mpll_func_cntl_1;
-	u32 mpll_func_cntl_2;
-	u32 mpll_ss1;
-	u32 mpll_ss2;
-};
-
-struct ci_thermal_temperature_setting {
-	s32 temperature_low;
-	s32 temperature_high;
-	s32 temperature_shutdown;
-};
-
-struct ci_pcie_perf_range {
-	u16 max;
-	u16 min;
-};
-
-enum ci_pt_config_reg_type {
-	CISLANDS_CONFIGREG_MMR = 0,
-	CISLANDS_CONFIGREG_SMC_IND,
-	CISLANDS_CONFIGREG_DIDT_IND,
-	CISLANDS_CONFIGREG_CACHE,
-	CISLANDS_CONFIGREG_MAX
-};
-
-#define POWERCONTAINMENT_FEATURE_BAPM            0x00000001
-#define POWERCONTAINMENT_FEATURE_TDCLimit        0x00000002
-#define POWERCONTAINMENT_FEATURE_PkgPwrLimit     0x00000004
-
-struct ci_pt_config_reg {
-	u32 offset;
-	u32 mask;
-	u32 shift;
-	u32 value;
-	enum ci_pt_config_reg_type type;
-};
-
-struct ci_pt_defaults {
-	u8 svi_load_line_en;
-	u8 svi_load_line_vddc;
-	u8 tdc_vddc_throttle_release_limit_perc;
-	u8 tdc_mawt;
-	u8 tdc_waterfall_ctl;
-	u8 dte_ambient_temp_base;
-	u32 display_cac;
-	u32 bapm_temp_gradient;
-	u16 bapmti_r[SMU7_DTE_ITERATIONS * SMU7_DTE_SOURCES * SMU7_DTE_SINKS];
-	u16 bapmti_rc[SMU7_DTE_ITERATIONS * SMU7_DTE_SOURCES * SMU7_DTE_SINKS];
-};
-
-#define DPMTABLE_OD_UPDATE_SCLK     0x00000001
-#define DPMTABLE_OD_UPDATE_MCLK     0x00000002
-#define DPMTABLE_UPDATE_SCLK        0x00000004
-#define DPMTABLE_UPDATE_MCLK        0x00000008
-
-struct ci_power_info {
-	struct ci_dpm_table dpm_table;
-	struct ci_dpm_table golden_dpm_table;
-	u32 voltage_control;
-	u32 mvdd_control;
-	u32 vddci_control;
-	u32 active_auto_throttle_sources;
-	struct ci_clock_registers clock_registers;
-	u16 acpi_vddc;
-	u16 acpi_vddci;
-	enum amdgpu_pcie_gen force_pcie_gen;
-	enum amdgpu_pcie_gen acpi_pcie_gen;
-	struct ci_leakage_voltage vddc_leakage;
-	struct ci_leakage_voltage vddci_leakage;
-	u16 max_vddc_in_pp_table;
-	u16 min_vddc_in_pp_table;
-	u16 max_vddci_in_pp_table;
-	u16 min_vddci_in_pp_table;
-	u32 mclk_strobe_mode_threshold;
-	u32 mclk_stutter_mode_threshold;
-	u32 mclk_edc_enable_threshold;
-	u32 mclk_edc_wr_enable_threshold;
-	struct ci_vbios_boot_state vbios_boot_state;
-	/* smc offsets */
-	u32 sram_end;
-	u32 dpm_table_start;
-	u32 soft_regs_start;
-	u32 mc_reg_table_start;
-	u32 fan_table_start;
-	u32 arb_table_start;
-	/* smc tables */
-	SMU7_Discrete_DpmTable smc_state_table;
-	SMU7_Discrete_MCRegisters smc_mc_reg_table;
-	SMU7_Discrete_PmFuses smc_powertune_table;
-	/* other stuff */
-	struct ci_mc_reg_table mc_reg_table;
-	struct atom_voltage_table vddc_voltage_table;
-	struct atom_voltage_table vddci_voltage_table;
-	struct atom_voltage_table mvdd_voltage_table;
-	struct ci_ulv_parm ulv;
-	u32 power_containment_features;
-	const struct ci_pt_defaults *powertune_defaults;
-	u32 dte_tj_offset;
-	bool vddc_phase_shed_control;
-	struct ci_thermal_temperature_setting thermal_temp_setting;
-	struct ci_dpm_level_enable_mask dpm_level_enable_mask;
-	u32 need_update_smu7_dpm_table;
-	u32 sclk_dpm_key_disabled;
-	u32 mclk_dpm_key_disabled;
-	u32 pcie_dpm_key_disabled;
-	u32 thermal_sclk_dpm_enabled;
-	struct ci_pcie_perf_range pcie_gen_performance;
-	struct ci_pcie_perf_range pcie_lane_performance;
-	struct ci_pcie_perf_range pcie_gen_powersaving;
-	struct ci_pcie_perf_range pcie_lane_powersaving;
-	u32 activity_target[SMU7_MAX_LEVELS_GRAPHICS];
-	u32 mclk_activity_target;
-	u32 low_sclk_interrupt_t;
-	u32 last_mclk_dpm_enable_mask;
-	u32 sys_pcie_mask;
-	/* caps */
-	bool caps_power_containment;
-	bool caps_cac;
-	bool caps_sq_ramping;
-	bool caps_db_ramping;
-	bool caps_td_ramping;
-	bool caps_tcp_ramping;
-	bool caps_fps;
-	bool caps_sclk_ds;
-	bool caps_sclk_ss_support;
-	bool caps_mclk_ss_support;
-	bool caps_uvd_dpm;
-	bool caps_vce_dpm;
-	bool caps_samu_dpm;
-	bool caps_acp_dpm;
-	bool caps_automatic_dc_transition;
-	bool caps_sclk_throttle_low_notification;
-	bool caps_dynamic_ac_timing;
-	bool caps_od_fuzzy_fan_control_support;
-	/* flags */
-	bool thermal_protection;
-	bool pcie_performance_request;
-	bool dynamic_ss;
-	bool dll_default_on;
-	bool cac_enabled;
-	bool uvd_enabled;
-	bool battery_state;
-	bool pspp_notify_required;
-	bool enable_bapm_feature;
-	bool enable_tdc_limit_feature;
-	bool enable_pkg_pwr_tracking_feature;
-	bool use_pcie_performance_levels;
-	bool use_pcie_powersaving_levels;
-	bool uvd_power_gated;
-	/* driver states */
-	struct amdgpu_ps current_rps;
-	struct ci_ps current_ps;
-	struct amdgpu_ps requested_rps;
-	struct ci_ps requested_ps;
-	/* fan control */
-	bool fan_ctrl_is_in_default_mode;
-	bool fan_is_controlled_by_smc;
-	u32 t_min;
-	u32 fan_ctrl_default_mode;
-};
-
-#define CISLANDS_VOLTAGE_CONTROL_NONE                   0x0
-#define CISLANDS_VOLTAGE_CONTROL_BY_GPIO                0x1
-#define CISLANDS_VOLTAGE_CONTROL_BY_SVID2               0x2
-
-#define CISLANDS_Q88_FORMAT_CONVERSION_UNIT             256
-
-#define CISLANDS_VRC_DFLT0                              0x3FFFC000
-#define CISLANDS_VRC_DFLT1                              0x000400
-#define CISLANDS_VRC_DFLT2                              0xC00080
-#define CISLANDS_VRC_DFLT3                              0xC00200
-#define CISLANDS_VRC_DFLT4                              0xC01680
-#define CISLANDS_VRC_DFLT5                              0xC00033
-#define CISLANDS_VRC_DFLT6                              0xC00033
-#define CISLANDS_VRC_DFLT7                              0x3FFFC000
-
-#define CISLANDS_CGULVPARAMETER_DFLT                    0x00040035
-#define CISLAND_TARGETACTIVITY_DFLT                     30
-#define CISLAND_MCLK_TARGETACTIVITY_DFLT                10
-
-#define PCIE_PERF_REQ_REMOVE_REGISTRY   0
-#define PCIE_PERF_REQ_FORCE_LOWPOWER    1
-#define PCIE_PERF_REQ_PECI_GEN1         2
-#define PCIE_PERF_REQ_PECI_GEN2         3
-#define PCIE_PERF_REQ_PECI_GEN3         4
-
-#define CISLANDS_SSTU_DFLT                               0
-#define CISLANDS_SST_DFLT                                0x00C8
-
-/* XXX are these ok? */
-#define CISLANDS_TEMP_RANGE_MIN (90 * 1000)
-#define CISLANDS_TEMP_RANGE_MAX (120 * 1000)
-
-int amdgpu_ci_copy_bytes_to_smc(struct amdgpu_device *adev,
-			 u32 smc_start_address,
-			 const u8 *src, u32 byte_count, u32 limit);
-void amdgpu_ci_start_smc(struct amdgpu_device *adev);
-void amdgpu_ci_reset_smc(struct amdgpu_device *adev);
-int amdgpu_ci_program_jump_on_start(struct amdgpu_device *adev);
-void amdgpu_ci_stop_smc_clock(struct amdgpu_device *adev);
-void amdgpu_ci_start_smc_clock(struct amdgpu_device *adev);
-bool amdgpu_ci_is_smc_running(struct amdgpu_device *adev);
-PPSMC_Result amdgpu_ci_send_msg_to_smc(struct amdgpu_device *adev, PPSMC_Msg msg);
-PPSMC_Result amdgpu_ci_wait_for_smc_inactive(struct amdgpu_device *adev);
-int amdgpu_ci_load_smc_ucode(struct amdgpu_device *adev, u32 limit);
-int amdgpu_ci_read_smc_sram_dword(struct amdgpu_device *adev,
-			   u32 smc_address, u32 *value, u32 limit);
-int amdgpu_ci_write_smc_sram_dword(struct amdgpu_device *adev,
-			    u32 smc_address, u32 value, u32 limit);
-
-#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_smc.c b/drivers/gpu/drm/amd/amdgpu/ci_smc.c
deleted file mode 100644
index b8ba51e045b5..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/ci_smc.c
+++ /dev/null
@@ -1,279 +0,0 @@
-/*
- * Copyright 2011 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: Alex Deucher
- */
-
-#include <linux/firmware.h>
-#include <drm/drmP.h>
-#include "amdgpu.h"
-#include "cikd.h"
-#include "ppsmc.h"
-#include "amdgpu_ucode.h"
-#include "ci_dpm.h"
-
-#include "smu/smu_7_0_1_d.h"
-#include "smu/smu_7_0_1_sh_mask.h"
-
-static int ci_set_smc_sram_address(struct amdgpu_device *adev,
-				   u32 smc_address, u32 limit)
-{
-	if (smc_address & 3)
-		return -EINVAL;
-	if ((smc_address + 3) > limit)
-		return -EINVAL;
-
-	WREG32(mmSMC_IND_INDEX_0, smc_address);
-	WREG32_P(mmSMC_IND_ACCESS_CNTL, 0, ~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
-
-	return 0;
-}
-
-int amdgpu_ci_copy_bytes_to_smc(struct amdgpu_device *adev,
-			 u32 smc_start_address,
-			 const u8 *src, u32 byte_count, u32 limit)
-{
-	unsigned long flags;
-	u32 data, original_data;
-	u32 addr;
-	u32 extra_shift;
-	int ret = 0;
-
-	if (smc_start_address & 3)
-		return -EINVAL;
-	if ((smc_start_address + byte_count) > limit)
-		return -EINVAL;
-
-	addr = smc_start_address;
-
-	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	while (byte_count >= 4) {
-		/* SMC address space is BE */
-		data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3];
-
-		ret = ci_set_smc_sram_address(adev, addr, limit);
-		if (ret)
-			goto done;
-
-		WREG32(mmSMC_IND_DATA_0, data);
-
-		src += 4;
-		byte_count -= 4;
-		addr += 4;
-	}
-
-	/* RMW for the final bytes */
-	if (byte_count > 0) {
-		data = 0;
-
-		ret = ci_set_smc_sram_address(adev, addr, limit);
-		if (ret)
-			goto done;
-
-		original_data = RREG32(mmSMC_IND_DATA_0);
-
-		extra_shift = 8 * (4 - byte_count);
-
-		while (byte_count > 0) {
-			data = (data << 8) + *src++;
-			byte_count--;
-		}
-
-		data <<= extra_shift;
-
-		data |= (original_data & ~((~0UL) << extra_shift));
-
-		ret = ci_set_smc_sram_address(adev, addr, limit);
-		if (ret)
-			goto done;
-
-		WREG32(mmSMC_IND_DATA_0, data);
-	}
-
-done:
-	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
-
-	return ret;
-}
-
-void amdgpu_ci_start_smc(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixSMC_SYSCON_RESET_CNTL);
-
-	tmp &= ~SMC_SYSCON_RESET_CNTL__rst_reg_MASK;
-	WREG32_SMC(ixSMC_SYSCON_RESET_CNTL, tmp);
-}
-
-void amdgpu_ci_reset_smc(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixSMC_SYSCON_RESET_CNTL);
-
-	tmp |= SMC_SYSCON_RESET_CNTL__rst_reg_MASK;
-	WREG32_SMC(ixSMC_SYSCON_RESET_CNTL, tmp);
-}
-
-int amdgpu_ci_program_jump_on_start(struct amdgpu_device *adev)
-{
-	static u8 data[] = { 0xE0, 0x00, 0x80, 0x40 };
-
-	return amdgpu_ci_copy_bytes_to_smc(adev, 0x0, data, 4, sizeof(data)+1);
-}
-
-void amdgpu_ci_stop_smc_clock(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
-
-	tmp |= SMC_SYSCON_CLOCK_CNTL_0__ck_disable_MASK;
-
-	WREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0, tmp);
-}
-
-void amdgpu_ci_start_smc_clock(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
-
-	tmp &= ~SMC_SYSCON_CLOCK_CNTL_0__ck_disable_MASK;
-
-	WREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0, tmp);
-}
-
-bool amdgpu_ci_is_smc_running(struct amdgpu_device *adev)
-{
-	u32 clk = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
-	u32 pc_c = RREG32_SMC(ixSMC_PC_C);
-
-	if (!(clk & SMC_SYSCON_CLOCK_CNTL_0__ck_disable_MASK) && (0x20100 <= pc_c))
-		return true;
-
-	return false;
-}
-
-PPSMC_Result amdgpu_ci_send_msg_to_smc(struct amdgpu_device *adev, PPSMC_Msg msg)
-{
-	u32 tmp;
-	int i;
-
-	if (!amdgpu_ci_is_smc_running(adev))
-		return PPSMC_Result_Failed;
-
-	WREG32(mmSMC_MESSAGE_0, msg);
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(mmSMC_RESP_0);
-		if (tmp != 0)
-			break;
-		udelay(1);
-	}
-	tmp = RREG32(mmSMC_RESP_0);
-
-	return (PPSMC_Result)tmp;
-}
-
-PPSMC_Result amdgpu_ci_wait_for_smc_inactive(struct amdgpu_device *adev)
-{
-	u32 tmp;
-	int i;
-
-	if (!amdgpu_ci_is_smc_running(adev))
-		return PPSMC_Result_OK;
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
-		if ((tmp & SMC_SYSCON_CLOCK_CNTL_0__cken_MASK) == 0)
-			break;
-		udelay(1);
-	}
-
-	return PPSMC_Result_OK;
-}
-
-int amdgpu_ci_load_smc_ucode(struct amdgpu_device *adev, u32 limit)
-{
-	const struct smc_firmware_header_v1_0 *hdr;
-	unsigned long flags;
-	u32 ucode_start_address;
-	u32 ucode_size;
-	const u8 *src;
-	u32 data;
-
-	if (!adev->pm.fw)
-		return -EINVAL;
-
-	hdr = (const struct smc_firmware_header_v1_0 *)adev->pm.fw->data;
-	amdgpu_ucode_print_smc_hdr(&hdr->header);
-
-	adev->pm.fw_version = le32_to_cpu(hdr->header.ucode_version);
-	ucode_start_address = le32_to_cpu(hdr->ucode_start_addr);
-	ucode_size = le32_to_cpu(hdr->header.ucode_size_bytes);
-	src = (const u8 *)
-		(adev->pm.fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes));
-
-	if (ucode_size & 3)
-		return -EINVAL;
-
-	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	WREG32(mmSMC_IND_INDEX_0, ucode_start_address);
-	WREG32_P(mmSMC_IND_ACCESS_CNTL, SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK,
-		~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
-	while (ucode_size >= 4) {
-		/* SMC address space is BE */
-		data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3];
-
-		WREG32(mmSMC_IND_DATA_0, data);
-
-		src += 4;
-		ucode_size -= 4;
-	}
-	WREG32_P(mmSMC_IND_ACCESS_CNTL, 0, ~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
-	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
-
-	return 0;
-}
-
-int amdgpu_ci_read_smc_sram_dword(struct amdgpu_device *adev,
-			   u32 smc_address, u32 *value, u32 limit)
-{
-	unsigned long flags;
-	int ret;
-
-	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	ret = ci_set_smc_sram_address(adev, smc_address, limit);
-	if (ret == 0)
-		*value = RREG32(mmSMC_IND_DATA_0);
-	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
-
-	return ret;
-}
-
-int amdgpu_ci_write_smc_sram_dword(struct amdgpu_device *adev,
-			    u32 smc_address, u32 value, u32 limit)
-{
-	unsigned long flags;
-	int ret;
-
-	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	ret = ci_set_smc_sram_address(adev, smc_address, limit);
-	if (ret == 0)
-		WREG32(mmSMC_IND_DATA_0, value);
-	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
-
-	return ret;
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index 71c50d8900e3..07c1f239e9c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -1741,6 +1741,69 @@ static bool cik_need_full_reset(struct amdgpu_device *adev)
 	return true;
 }
 
+static void cik_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0,
+			       uint64_t *count1)
+{
+	uint32_t perfctr = 0;
+	uint64_t cnt0_of, cnt1_of;
+	int tmp;
+
+	/* This reports 0 on APUs, so return to avoid writing/reading registers
+	 * that may or may not be different from their GPU counterparts
+	 */
+	if (adev->flags & AMD_IS_APU)
+		return;
+
+	/* Set the 2 events that we wish to watch, defined above */
+	/* Reg 40 is # received msgs, Reg 104 is # of posted requests sent */
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT0_SEL, 40);
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT1_SEL, 104);
+
+	/* Write to enable desired perf counters */
+	WREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK, perfctr);
+	/* Zero out and enable the perf counters
+	 * Write 0x5:
+	 * Bit 0 = Start all counters(1)
+	 * Bit 2 = Global counter reset enable(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000005);
+
+	msleep(1000);
+
+	/* Load the shadow and disable the perf counters
+	 * Write 0x2:
+	 * Bit 0 = Stop counters(0)
+	 * Bit 1 = Load the shadow counters(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000002);
+
+	/* Read register values to get any >32bit overflow */
+	tmp = RREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK);
+	cnt0_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER0_UPPER);
+	cnt1_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER1_UPPER);
+
+	/* Get the values and add the overflow */
+	*count0 = RREG32_PCIE(ixPCIE_PERF_COUNT0_TXCLK) | (cnt0_of << 32);
+	*count1 = RREG32_PCIE(ixPCIE_PERF_COUNT1_TXCLK) | (cnt1_of << 32);
+}
+
+static bool cik_need_reset_on_init(struct amdgpu_device *adev)
+{
+	u32 clock_cntl, pc;
+
+	if (adev->flags & AMD_IS_APU)
+		return false;
+
+	/* check if the SMC is already running */
+	clock_cntl = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
+	pc = RREG32_SMC(ixSMC_PC_C);
+	if ((0 == REG_GET_FIELD(clock_cntl, SMC_SYSCON_CLOCK_CNTL_0, ck_disable)) &&
+	    (0x20100 <= pc))
+		return true;
+
+	return false;
+}
+
 static const struct amdgpu_asic_funcs cik_asic_funcs =
 {
 	.read_disabled_bios = &cik_read_disabled_bios,
@@ -1756,6 +1819,8 @@ static const struct amdgpu_asic_funcs cik_asic_funcs =
 	.invalidate_hdp = &cik_invalidate_hdp,
 	.need_full_reset = &cik_need_full_reset,
 	.init_doorbell_index = &legacy_doorbell_index_init,
+	.get_pcie_usage = &cik_get_pcie_usage,
+	.need_reset_on_init = &cik_need_reset_on_init,
 };
 
 static int cik_common_early_init(void *handle)
@@ -2005,10 +2070,7 @@ int cik_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &cik_ih_ip_block);
 		amdgpu_device_ip_block_add(adev, &gfx_v7_2_ip_block);
 		amdgpu_device_ip_block_add(adev, &cik_sdma_ip_block);
-		if (amdgpu_dpm == -1)
-			amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
-		else
-			amdgpu_device_ip_block_add(adev, &ci_smu_ip_block);
+		amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
 		if (adev->enable_virtual_display)
 			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
 #if defined(CONFIG_DRM_AMD_DC)
@@ -2026,10 +2088,7 @@ int cik_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &cik_ih_ip_block);
 		amdgpu_device_ip_block_add(adev, &gfx_v7_3_ip_block);
 		amdgpu_device_ip_block_add(adev, &cik_sdma_ip_block);
-		if (amdgpu_dpm == -1)
-			amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
-		else
-			amdgpu_device_ip_block_add(adev, &ci_smu_ip_block);
+		amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
 		if (adev->enable_virtual_display)
 			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
 #if defined(CONFIG_DRM_AMD_DC)
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_dpm.h b/drivers/gpu/drm/amd/amdgpu/cik_dpm.h
index 2a086610f74d..2fcc4b60153c 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_dpm.h
+++ b/drivers/gpu/drm/amd/amdgpu/cik_dpm.h
@@ -24,7 +24,6 @@
 #ifndef __CIK_DPM_H__
 #define __CIK_DPM_H__
 
-extern const struct amdgpu_ip_block_version ci_smu_ip_block;
 extern const struct amdgpu_ip_block_version kv_smu_ip_block;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c
index 8a8b4967a101..721c757156e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c
@@ -103,9 +103,9 @@ static void cik_ih_disable_interrupts(struct amdgpu_device *adev)
  */
 static int cik_ih_irq_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_ih_ring *ih = &adev->irq.ih;
 	int rb_bufsz;
 	u32 interrupt_cntl, ih_cntl, ih_rb_cntl;
-	u64 wptr_off;
 
 	/* disable irqs */
 	cik_ih_disable_interrupts(adev);
@@ -131,9 +131,8 @@ static int cik_ih_irq_init(struct amdgpu_device *adev)
 	ih_rb_cntl |= IH_RB_CNTL__WPTR_WRITEBACK_ENABLE_MASK;
 
 	/* set the writeback address whether it's enabled or not */
-	wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr));
+	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(ih->wptr_addr) & 0xFF);
 
 	WREG32(mmIH_RB_CNTL, ih_rb_cntl);
 
@@ -183,11 +182,12 @@ static void cik_ih_irq_disable(struct amdgpu_device *adev)
  * Used by cik_irq_process().
  * Returns the value of the wptr.
  */
-static u32 cik_ih_get_wptr(struct amdgpu_device *adev)
+static u32 cik_ih_get_wptr(struct amdgpu_device *adev,
+			   struct amdgpu_ih_ring *ih)
 {
 	u32 wptr, tmp;
 
-	wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
 
 	if (wptr & IH_RB_WPTR__RB_OVERFLOW_MASK) {
 		wptr &= ~IH_RB_WPTR__RB_OVERFLOW_MASK;
@@ -196,13 +196,13 @@ static u32 cik_ih_get_wptr(struct amdgpu_device *adev)
 		 * this should allow us to catchup.
 		 */
 		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, (wptr + 16) & adev->irq.ih.ptr_mask);
-		adev->irq.ih.rptr = (wptr + 16) & adev->irq.ih.ptr_mask;
+			 wptr, ih->rptr, (wptr + 16) & ih->ptr_mask);
+		ih->rptr = (wptr + 16) & ih->ptr_mask;
 		tmp = RREG32(mmIH_RB_CNTL);
 		tmp |= IH_RB_CNTL__WPTR_OVERFLOW_CLEAR_MASK;
 		WREG32(mmIH_RB_CNTL, tmp);
 	}
-	return (wptr & adev->irq.ih.ptr_mask);
+	return (wptr & ih->ptr_mask);
 }
 
 /*        CIK IV Ring
@@ -237,16 +237,17 @@ static u32 cik_ih_get_wptr(struct amdgpu_device *adev)
  * position and also advance the position.
  */
 static void cik_ih_decode_iv(struct amdgpu_device *adev,
+			     struct amdgpu_ih_ring *ih,
 			     struct amdgpu_iv_entry *entry)
 {
 	/* wptr/rptr are in bytes! */
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[4];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
 
 	entry->client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
 	entry->src_id = dw[0] & 0xff;
@@ -256,7 +257,7 @@ static void cik_ih_decode_iv(struct amdgpu_device *adev,
 	entry->pasid = (dw[2] >> 16) & 0xffff;
 
 	/* wptr/rptr are in bytes! */
-	adev->irq.ih.rptr += 16;
+	ih->rptr += 16;
 }
 
 /**
@@ -266,9 +267,10 @@ static void cik_ih_decode_iv(struct amdgpu_device *adev,
  *
  * Set the IH ring buffer rptr.
  */
-static void cik_ih_set_rptr(struct amdgpu_device *adev)
+static void cik_ih_set_rptr(struct amdgpu_device *adev,
+			    struct amdgpu_ih_ring *ih)
 {
-	WREG32(mmIH_RB_RPTR, adev->irq.ih.rptr);
+	WREG32(mmIH_RB_RPTR, ih->rptr);
 }
 
 static int cik_ih_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 45795191de1f..189599b694e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -220,7 +220,7 @@ static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void cik_sdma_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 extra_bits = vmid & 0xf;
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_ih.c b/drivers/gpu/drm/amd/amdgpu/cz_ih.c
index 9d3ea298e116..61024b9c7a4b 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_ih.c
@@ -103,9 +103,9 @@ static void cz_ih_disable_interrupts(struct amdgpu_device *adev)
  */
 static int cz_ih_irq_init(struct amdgpu_device *adev)
 {
-	int rb_bufsz;
+	struct amdgpu_ih_ring *ih = &adev->irq.ih;
 	u32 interrupt_cntl, ih_cntl, ih_rb_cntl;
-	u64 wptr_off;
+	int rb_bufsz;
 
 	/* disable irqs */
 	cz_ih_disable_interrupts(adev);
@@ -133,9 +133,8 @@ static int cz_ih_irq_init(struct amdgpu_device *adev)
 	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, WPTR_WRITEBACK_ENABLE, 1);
 
 	/* set the writeback address whether it's enabled or not */
-	wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr));
+	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(ih->wptr_addr) & 0xFF);
 
 	WREG32(mmIH_RB_CNTL, ih_rb_cntl);
 
@@ -185,11 +184,12 @@ static void cz_ih_irq_disable(struct amdgpu_device *adev)
  * Used by cz_irq_process(VI).
  * Returns the value of the wptr.
  */
-static u32 cz_ih_get_wptr(struct amdgpu_device *adev)
+static u32 cz_ih_get_wptr(struct amdgpu_device *adev,
+			  struct amdgpu_ih_ring *ih)
 {
 	u32 wptr, tmp;
 
-	wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
 
 	if (REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) {
 		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
@@ -198,13 +198,13 @@ static u32 cz_ih_get_wptr(struct amdgpu_device *adev)
 		 * this should allow us to catchup.
 		 */
 		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, (wptr + 16) & adev->irq.ih.ptr_mask);
-		adev->irq.ih.rptr = (wptr + 16) & adev->irq.ih.ptr_mask;
+			wptr, ih->rptr, (wptr + 16) & ih->ptr_mask);
+		ih->rptr = (wptr + 16) & ih->ptr_mask;
 		tmp = RREG32(mmIH_RB_CNTL);
 		tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
 		WREG32(mmIH_RB_CNTL, tmp);
 	}
-	return (wptr & adev->irq.ih.ptr_mask);
+	return (wptr & ih->ptr_mask);
 }
 
 /**
@@ -216,16 +216,17 @@ static u32 cz_ih_get_wptr(struct amdgpu_device *adev)
  * position and also advance the position.
  */
 static void cz_ih_decode_iv(struct amdgpu_device *adev,
-				 struct amdgpu_iv_entry *entry)
+			    struct amdgpu_ih_ring *ih,
+			    struct amdgpu_iv_entry *entry)
 {
 	/* wptr/rptr are in bytes! */
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[4];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
 
 	entry->client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
 	entry->src_id = dw[0] & 0xff;
@@ -235,7 +236,7 @@ static void cz_ih_decode_iv(struct amdgpu_device *adev,
 	entry->pasid = (dw[2] >> 16) & 0xffff;
 
 	/* wptr/rptr are in bytes! */
-	adev->irq.ih.rptr += 16;
+	ih->rptr += 16;
 }
 
 /**
@@ -245,9 +246,10 @@ static void cz_ih_decode_iv(struct amdgpu_device *adev,
  *
  * Set the IH ring buffer rptr.
  */
-static void cz_ih_set_rptr(struct amdgpu_device *adev)
+static void cz_ih_set_rptr(struct amdgpu_device *adev,
+			   struct amdgpu_ih_ring *ih)
 {
-	WREG32(mmIH_RB_RPTR, adev->irq.ih.rptr);
+	WREG32(mmIH_RB_RPTR, ih->rptr);
 }
 
 static int cz_ih_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
index fdace004544d..e4cc1d48eaab 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
@@ -167,19 +167,6 @@ static void dce_virtual_crtc_disable(struct drm_crtc *crtc)
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 
 	dce_virtual_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
-	if (crtc->primary->fb) {
-		int r;
-		struct amdgpu_bo *abo;
-
-		abo = gem_to_amdgpu_bo(crtc->primary->fb->obj[0]);
-		r = amdgpu_bo_reserve(abo, true);
-		if (unlikely(r))
-			DRM_ERROR("failed to reserve abo before unpin\n");
-		else {
-			amdgpu_bo_unpin(abo);
-			amdgpu_bo_unreserve(abo);
-		}
-	}
 
 	amdgpu_crtc->pll_id = ATOM_PPLL_INVALID;
 	amdgpu_crtc->encoder = NULL;
@@ -692,7 +679,9 @@ static int dce_virtual_pageflip(struct amdgpu_device *adev,
 	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
 
 	drm_crtc_vblank_put(&amdgpu_crtc->base);
-	schedule_work(&works->unpin_work);
+	amdgpu_bo_unref(&works->old_abo);
+	kfree(works->shared);
+	kfree(works);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 1dc3013ea1d5..305276c7e4bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -1842,13 +1842,13 @@ static void gfx_v6_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
 static void gfx_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 header, control = 0;
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (ctx_switch) {
+	if (flags & AMDGPU_HAVE_CTX_SWITCH) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 3a9fb6018c16..a59e0fdf5a97 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -2228,13 +2228,13 @@ static void gfx_v7_0_ring_emit_fence_compute(struct amdgpu_ring *ring,
 static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 header, control = 0;
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (ctx_switch) {
+	if (flags & AMDGPU_HAVE_CTX_SWITCH) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
@@ -2259,11 +2259,27 @@ static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 					  struct amdgpu_job *job,
 					  struct amdgpu_ib *ib,
-					  bool ctx_switch)
+					  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
+	/* Currently, there is a high possibility to get wave ID mismatch
+	 * between ME and GDS, leading to a hw deadlock, because ME generates
+	 * different wave IDs than the GDS expects. This situation happens
+	 * randomly when at least 5 compute pipes use GDS ordered append.
+	 * The wave IDs generated by ME are also wrong after suspend/resume.
+	 * Those are probably bugs somewhere else in the kernel driver.
+	 *
+	 * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and
+	 * GDS to 0 for this ring (me/pipe).
+	 */
+	if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) {
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		amdgpu_ring_write(ring, mmGDS_COMPUTE_MAX_WAVE_ID - PACKET3_SET_CONFIG_REG_START);
+		amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id);
+	}
+
 	amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
 	amdgpu_ring_write(ring,
 #ifdef __BIG_ENDIAN
@@ -5000,7 +5016,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 		7 + /* gfx_v7_0_ring_emit_pipeline_sync */
 		CIK_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v7_0_ring_emit_vm_flush */
 		7 + 7 + 7, /* gfx_v7_0_ring_emit_fence_compute x3 for user fence, vm fence */
-	.emit_ib_size =	4, /* gfx_v7_0_ring_emit_ib_compute */
+	.emit_ib_size =	7, /* gfx_v7_0_ring_emit_ib_compute */
 	.emit_ib = gfx_v7_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v7_0_ring_emit_fence_compute,
 	.emit_pipeline_sync = gfx_v7_0_ring_emit_pipeline_sync,
@@ -5057,6 +5073,7 @@ static void gfx_v7_0_set_gds_init(struct amdgpu_device *adev)
 	adev->gds.mem.total_size = RREG32(mmGDS_VMID0_SIZE);
 	adev->gds.gws.total_size = 64;
 	adev->gds.oa.total_size = 16;
+	adev->gds.gds_compute_max_wave_id = RREG32(mmGDS_COMPUTE_MAX_WAVE_ID);
 
 	if (adev->gds.mem.total_size == 64 * 1024) {
 		adev->gds.mem.gfx_partition_size = 4096;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 381f593b0cda..b8e50a34bdb3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4233,7 +4233,6 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev)
 	u32 tmp;
 	u32 rb_bufsz;
 	u64 rb_addr, rptr_addr, wptr_gpu_addr;
-	int r;
 
 	/* Set the write pointer delay */
 	WREG32(mmCP_RB_WPTR_DELAY, 0);
@@ -4278,9 +4277,8 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev)
 	amdgpu_ring_clear_ring(ring);
 	gfx_v8_0_cp_gfx_start(adev);
 	ring->sched.ready = true;
-	r = amdgpu_ring_test_helper(ring);
 
-	return r;
+	return 0;
 }
 
 static void gfx_v8_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
@@ -4369,10 +4367,9 @@ static int gfx_v8_0_kiq_kcq_enable(struct amdgpu_device *adev)
 		amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr));
 	}
 
-	r = amdgpu_ring_test_helper(kiq_ring);
-	if (r)
-		DRM_ERROR("KCQ enable failed\n");
-	return r;
+	amdgpu_ring_commit(kiq_ring);
+
+	return 0;
 }
 
 static int gfx_v8_0_deactivate_hqd(struct amdgpu_device *adev, u32 req)
@@ -4709,16 +4706,32 @@ static int gfx_v8_0_kcq_resume(struct amdgpu_device *adev)
 	if (r)
 		goto done;
 
-	/* Test KCQs - reversing the order of rings seems to fix ring test failure
-	 * after GPU reset
-	 */
-	for (i = adev->gfx.num_compute_rings - 1; i >= 0; i--) {
+done:
+	return r;
+}
+
+static int gfx_v8_0_cp_test_all_rings(struct amdgpu_device *adev)
+{
+	int r, i;
+	struct amdgpu_ring *ring;
+
+	/* collect all the ring_tests here, gfx, kiq, compute */
+	ring = &adev->gfx.gfx_ring[0];
+	r = amdgpu_ring_test_helper(ring);
+	if (r)
+		return r;
+
+	ring = &adev->gfx.kiq.ring;
+	r = amdgpu_ring_test_helper(ring);
+	if (r)
+		return r;
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
 		ring = &adev->gfx.compute_ring[i];
-		r = amdgpu_ring_test_helper(ring);
+		amdgpu_ring_test_helper(ring);
 	}
 
-done:
-	return r;
+	return 0;
 }
 
 static int gfx_v8_0_cp_resume(struct amdgpu_device *adev)
@@ -4739,6 +4752,11 @@ static int gfx_v8_0_cp_resume(struct amdgpu_device *adev)
 	r = gfx_v8_0_kcq_resume(adev);
 	if (r)
 		return r;
+
+	r = gfx_v8_0_cp_test_all_rings(adev);
+	if (r)
+		return r;
+
 	gfx_v8_0_enable_gui_idle_interrupt(adev, true);
 
 	return 0;
@@ -5086,6 +5104,8 @@ static int gfx_v8_0_post_soft_reset(void *handle)
 	    REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_GFX))
 		gfx_v8_0_cp_gfx_resume(adev);
 
+	gfx_v8_0_cp_test_all_rings(adev);
+
 	adev->gfx.rlc.funcs->start(adev);
 
 	return 0;
@@ -6027,7 +6047,7 @@ static void gfx_v8_0_ring_emit_vgt_flush(struct amdgpu_ring *ring)
 static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 header, control = 0;
@@ -6059,11 +6079,27 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 					  struct amdgpu_job *job,
 					  struct amdgpu_ib *ib,
-					  bool ctx_switch)
+					  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
+	/* Currently, there is a high possibility to get wave ID mismatch
+	 * between ME and GDS, leading to a hw deadlock, because ME generates
+	 * different wave IDs than the GDS expects. This situation happens
+	 * randomly when at least 5 compute pipes use GDS ordered append.
+	 * The wave IDs generated by ME are also wrong after suspend/resume.
+	 * Those are probably bugs somewhere else in the kernel driver.
+	 *
+	 * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and
+	 * GDS to 0 for this ring (me/pipe).
+	 */
+	if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) {
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		amdgpu_ring_write(ring, mmGDS_COMPUTE_MAX_WAVE_ID - PACKET3_SET_CONFIG_REG_START);
+		amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id);
+	}
+
 	amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
 	amdgpu_ring_write(ring,
 #ifdef __BIG_ENDIAN
@@ -6870,7 +6906,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
 		7 + /* gfx_v8_0_ring_emit_pipeline_sync */
 		VI_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v8_0_ring_emit_vm_flush */
 		7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_compute x3 for user fence, vm fence */
-	.emit_ib_size =	4, /* gfx_v8_0_ring_emit_ib_compute */
+	.emit_ib_size =	7, /* gfx_v8_0_ring_emit_ib_compute */
 	.emit_ib = gfx_v8_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v8_0_ring_emit_fence_compute,
 	.emit_pipeline_sync = gfx_v8_0_ring_emit_pipeline_sync,
@@ -6900,7 +6936,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = {
 		7 + /* gfx_v8_0_ring_emit_pipeline_sync */
 		17 + /* gfx_v8_0_ring_emit_vm_flush */
 		7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_kiq x3 for user fence, vm fence */
-	.emit_ib_size =	4, /* gfx_v8_0_ring_emit_ib_compute */
+	.emit_ib_size =	7, /* gfx_v8_0_ring_emit_ib_compute */
 	.emit_fence = gfx_v8_0_ring_emit_fence_kiq,
 	.test_ring = gfx_v8_0_ring_test_ring,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -6976,6 +7012,7 @@ static void gfx_v8_0_set_gds_init(struct amdgpu_device *adev)
 	adev->gds.mem.total_size = RREG32(mmGDS_VMID0_SIZE);
 	adev->gds.gws.total_size = 64;
 	adev->gds.oa.total_size = 16;
+	adev->gds.gds_compute_max_wave_id = RREG32(mmGDS_COMPUTE_MAX_WAVE_ID);
 
 	if (adev->gds.mem.total_size == 64 * 1024) {
 		adev->gds.mem.gfx_partition_size = 4096;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 7556716038d3..5533f6e4f4a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -113,7 +113,10 @@ static const struct soc15_reg_golden golden_settings_gc_9_0[] =
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x4a2c0e68),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_LO, 0xffffffff, 0xb5d3f197),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_CACHE_INVALIDATION, 0x3fff3af3, 0x19200000),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff)
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_MEC1_F32_INT_DIS, 0x00000000, 0x00000800),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_MEC2_F32_INT_DIS, 0x00000000, 0x00000800),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_DEBUG, 0x00000000, 0x00008000)
 };
 
 static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] =
@@ -135,10 +138,7 @@ static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] =
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_UTCL1_CNTL2, 0x00030000, 0x00020000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL_1, 0x0000000f, 0x01000107),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTD_CNTL, 0x00001800, 0x00000800),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x08000000, 0x08000080),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_MEC1_F32_INT_DIS, 0x00000000, 0x00000800),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_MEC2_F32_INT_DIS, 0x00000000, 0x00000800),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_DEBUG, 0x00000000, 0x00008000)
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x08000000, 0x08000080)
 };
 
 static const struct soc15_reg_golden golden_settings_gc_9_0_vg20[] =
@@ -3587,6 +3587,8 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 {
 	uint32_t data, def;
 
+	amdgpu_gfx_rlc_enter_safe_mode(adev);
+
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
 		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
@@ -3651,6 +3653,8 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 			WREG32_SOC15(GC, 0, mmCP_MEM_SLP_CNTL, data);
 		}
 	}
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev);
 }
 
 static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
@@ -3968,7 +3972,7 @@ static void gfx_v9_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 header, control = 0;
@@ -4001,11 +4005,27 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 					  struct amdgpu_job *job,
 					  struct amdgpu_ib *ib,
-					  bool ctx_switch)
+					  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
+	/* Currently, there is a high possibility to get wave ID mismatch
+	 * between ME and GDS, leading to a hw deadlock, because ME generates
+	 * different wave IDs than the GDS expects. This situation happens
+	 * randomly when at least 5 compute pipes use GDS ordered append.
+	 * The wave IDs generated by ME are also wrong after suspend/resume.
+	 * Those are probably bugs somewhere else in the kernel driver.
+	 *
+	 * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and
+	 * GDS to 0 for this ring (me/pipe).
+	 */
+	if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) {
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		amdgpu_ring_write(ring, mmGDS_COMPUTE_MAX_WAVE_ID);
+		amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id);
+	}
+
 	amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
 	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 	amdgpu_ring_write(ring,
@@ -4725,7 +4745,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 +
 		2 + /* gfx_v9_0_ring_emit_vm_flush */
 		8 + 8 + 8, /* gfx_v9_0_ring_emit_fence x3 for user fence, vm fence */
-	.emit_ib_size =	4, /* gfx_v9_0_ring_emit_ib_compute */
+	.emit_ib_size =	7, /* gfx_v9_0_ring_emit_ib_compute */
 	.emit_ib = gfx_v9_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v9_0_ring_emit_fence,
 	.emit_pipeline_sync = gfx_v9_0_ring_emit_pipeline_sync,
@@ -4760,7 +4780,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = {
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 +
 		2 + /* gfx_v9_0_ring_emit_vm_flush */
 		8 + 8 + 8, /* gfx_v9_0_ring_emit_fence_kiq x3 for user fence, vm fence */
-	.emit_ib_size =	4, /* gfx_v9_0_ring_emit_ib_compute */
+	.emit_ib_size =	7, /* gfx_v9_0_ring_emit_ib_compute */
 	.emit_fence = gfx_v9_0_ring_emit_fence_kiq,
 	.test_ring = gfx_v9_0_ring_test_ring,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -4842,6 +4862,26 @@ static void gfx_v9_0_set_gds_init(struct amdgpu_device *adev)
 		break;
 	}
 
+	switch (adev->asic_type) {
+	case CHIP_VEGA10:
+	case CHIP_VEGA20:
+		adev->gds.gds_compute_max_wave_id = 0x7ff;
+		break;
+	case CHIP_VEGA12:
+		adev->gds.gds_compute_max_wave_id = 0x27f;
+		break;
+	case CHIP_RAVEN:
+		if (adev->rev_id >= 0x8)
+			adev->gds.gds_compute_max_wave_id = 0x77; /* raven2 */
+		else
+			adev->gds.gds_compute_max_wave_id = 0x15f; /* raven1 */
+		break;
+	default:
+		/* this really depends on the chip */
+		adev->gds.gds_compute_max_wave_id = 0x7ff;
+		break;
+	}
+
 	adev->gds.gws.total_size = 64;
 	adev->gds.oa.total_size = 16;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 1ad7e6b8ed1d..34440672f938 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -1471,8 +1471,9 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,
 		gmc_v8_0_set_fault_enable_default(adev, false);
 
 	if (printk_ratelimit()) {
-		struct amdgpu_task_info task_info = { 0 };
+		struct amdgpu_task_info task_info;
 
+		memset(&task_info, 0, sizeof(struct amdgpu_task_info));
 		amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);
 
 		dev_err(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index bacdaef77b6c..600259b4e291 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -305,6 +305,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 				struct amdgpu_iv_entry *entry)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src];
+	bool retry_fault = !!(entry->src_data[1] & 0x80);
 	uint32_t status = 0;
 	u64 addr;
 
@@ -320,13 +321,16 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	}
 
 	if (printk_ratelimit()) {
-		struct amdgpu_task_info task_info = { 0 };
+		struct amdgpu_task_info task_info;
 
+		memset(&task_info, 0, sizeof(struct amdgpu_task_info));
 		amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);
 
 		dev_err(adev->dev,
-			"[%s] VMC page fault (src_id:%u ring:%u vmid:%u pasid:%u, for process %s pid %d thread %s pid %d)\n",
+			"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
+			"pasid:%u, for process %s pid %d thread %s pid %d)\n",
 			entry->vmid_src ? "mmhub" : "gfxhub",
+			retry_fault ? "retry" : "no-retry",
 			entry->src_id, entry->ring_id, entry->vmid,
 			entry->pasid, task_info.process_name, task_info.tgid,
 			task_info.task_name, task_info.pid);
@@ -961,7 +965,11 @@ static int gmc_v9_0_sw_init(void *handle)
 		 * vm size is 256TB (48bit), maximum size of Vega10,
 		 * block size 512 (9bit)
 		 */
-		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
+		/* sriov restrict max_pfn below AMDGPU_GMC_HOLE */
+		if (amdgpu_sriov_vf(adev))
+			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 47);
+		else
+			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c
index a3984d10b604..b1626e1d2f5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c
@@ -103,9 +103,9 @@ static void iceland_ih_disable_interrupts(struct amdgpu_device *adev)
  */
 static int iceland_ih_irq_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_ih_ring *ih = &adev->irq.ih;
 	int rb_bufsz;
 	u32 interrupt_cntl, ih_cntl, ih_rb_cntl;
-	u64 wptr_off;
 
 	/* disable irqs */
 	iceland_ih_disable_interrupts(adev);
@@ -133,9 +133,8 @@ static int iceland_ih_irq_init(struct amdgpu_device *adev)
 	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, WPTR_WRITEBACK_ENABLE, 1);
 
 	/* set the writeback address whether it's enabled or not */
-	wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr));
+	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(ih->wptr_addr) & 0xFF);
 
 	WREG32(mmIH_RB_CNTL, ih_rb_cntl);
 
@@ -185,11 +184,12 @@ static void iceland_ih_irq_disable(struct amdgpu_device *adev)
  * Used by cz_irq_process(VI).
  * Returns the value of the wptr.
  */
-static u32 iceland_ih_get_wptr(struct amdgpu_device *adev)
+static u32 iceland_ih_get_wptr(struct amdgpu_device *adev,
+			       struct amdgpu_ih_ring *ih)
 {
 	u32 wptr, tmp;
 
-	wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
 
 	if (REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) {
 		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
@@ -198,13 +198,13 @@ static u32 iceland_ih_get_wptr(struct amdgpu_device *adev)
 		 * this should allow us to catchup.
 		 */
 		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, (wptr + 16) & adev->irq.ih.ptr_mask);
-		adev->irq.ih.rptr = (wptr + 16) & adev->irq.ih.ptr_mask;
+			 wptr, ih->rptr, (wptr + 16) & ih->ptr_mask);
+		ih->rptr = (wptr + 16) & ih->ptr_mask;
 		tmp = RREG32(mmIH_RB_CNTL);
 		tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
 		WREG32(mmIH_RB_CNTL, tmp);
 	}
-	return (wptr & adev->irq.ih.ptr_mask);
+	return (wptr & ih->ptr_mask);
 }
 
 /**
@@ -216,16 +216,17 @@ static u32 iceland_ih_get_wptr(struct amdgpu_device *adev)
  * position and also advance the position.
  */
 static void iceland_ih_decode_iv(struct amdgpu_device *adev,
+				 struct amdgpu_ih_ring *ih,
 				 struct amdgpu_iv_entry *entry)
 {
 	/* wptr/rptr are in bytes! */
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[4];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
 
 	entry->client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
 	entry->src_id = dw[0] & 0xff;
@@ -235,7 +236,7 @@ static void iceland_ih_decode_iv(struct amdgpu_device *adev,
 	entry->pasid = (dw[2] >> 16) & 0xffff;
 
 	/* wptr/rptr are in bytes! */
-	adev->irq.ih.rptr += 16;
+	ih->rptr += 16;
 }
 
 /**
@@ -245,9 +246,10 @@ static void iceland_ih_decode_iv(struct amdgpu_device *adev,
  *
  * Set the IH ring buffer rptr.
  */
-static void iceland_ih_set_rptr(struct amdgpu_device *adev)
+static void iceland_ih_set_rptr(struct amdgpu_device *adev,
+				struct amdgpu_ih_ring *ih)
 {
-	WREG32(mmIH_RB_RPTR, adev->irq.ih.rptr);
+	WREG32(mmIH_RB_RPTR, ih->rptr);
 }
 
 static int iceland_ih_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index 8cbb4655896a..b11a1c17a7f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -174,7 +174,7 @@ static int xgpu_ai_send_access_requests(struct amdgpu_device *adev,
 			return r;
 		}
 		/* Retrieve checksum from mailbox2 */
-		if (req == IDH_REQ_GPU_INIT_ACCESS) {
+		if (req == IDH_REQ_GPU_INIT_ACCESS || req == IDH_REQ_GPU_RESET_ACCESS) {
 			adev->virt.fw_reserve.checksum_key =
 				RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0,
 					mmBIF_BX_PF0_MAILBOX_MSGBUF_RCV_DW2));
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
index accdedd63c98..cc967dbfd631 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
@@ -27,13 +27,9 @@
 #include "nbio/nbio_6_1_default.h"
 #include "nbio/nbio_6_1_offset.h"
 #include "nbio/nbio_6_1_sh_mask.h"
+#include "nbio/nbio_6_1_smn.h"
 #include "vega10_enum.h"
 
-#define smnCPM_CONTROL                                                                                  0x11180460
-#define smnPCIE_CNTL2                                                                                   0x11180070
-#define smnPCIE_CONFIG_CNTL                                                                             0x11180044
-#define smnPCIE_CI_CNTL                                                                                 0x11180080
-
 static u32 nbio_v6_1_get_rev_id(struct amdgpu_device *adev)
 {
         u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0);
@@ -72,7 +68,7 @@ static u32 nbio_v6_1_get_memsize(struct amdgpu_device *adev)
 }
 
 static void nbio_v6_1_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
-				  bool use_doorbell, int doorbell_index)
+			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
 	u32 reg = instance == 0 ? SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA0_DOORBELL_RANGE) :
 			SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA1_DOORBELL_RANGE);
@@ -81,7 +77,7 @@ static void nbio_v6_1_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 
 	if (use_doorbell) {
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, OFFSET, doorbell_index);
-		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 2);
+		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, doorbell_size);
 	} else
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 0);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c
index df34dc79d444..1cdb98ad2db3 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c
@@ -27,13 +27,11 @@
 #include "nbio/nbio_7_0_default.h"
 #include "nbio/nbio_7_0_offset.h"
 #include "nbio/nbio_7_0_sh_mask.h"
+#include "nbio/nbio_7_0_smn.h"
 #include "vega10_enum.h"
 
 #define smnNBIF_MGCG_CTRL_LCLK	0x1013a05c
 
-#define smnCPM_CONTROL                                                                                  0x11180460
-#define smnPCIE_CNTL2                                                                                   0x11180070
-
 static u32 nbio_v7_0_get_rev_id(struct amdgpu_device *adev)
 {
         u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0);
@@ -69,7 +67,7 @@ static u32 nbio_v7_0_get_memsize(struct amdgpu_device *adev)
 }
 
 static void nbio_v7_0_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
-					  bool use_doorbell, int doorbell_index)
+			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
 	u32 reg = instance == 0 ? SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA0_DOORBELL_RANGE) :
 			SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA1_DOORBELL_RANGE);
@@ -78,7 +76,7 @@ static void nbio_v7_0_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 
 	if (use_doorbell) {
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, OFFSET, doorbell_index);
-		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 2);
+		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, doorbell_size);
 	} else
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 0);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index 4cd31a276dcd..221f26e50322 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -26,13 +26,10 @@
 
 #include "nbio/nbio_7_4_offset.h"
 #include "nbio/nbio_7_4_sh_mask.h"
+#include "nbio/nbio_7_4_0_smn.h"
 
 #define smnNBIF_MGCG_CTRL_LCLK	0x1013a21c
 
-#define smnCPM_CONTROL                                                                                  0x11180460
-#define smnPCIE_CNTL2                                                                                   0x11180070
-#define smnPCIE_CI_CNTL                                                                                 0x11180080
-
 static u32 nbio_v7_4_get_rev_id(struct amdgpu_device *adev)
 {
     u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0);
@@ -68,7 +65,7 @@ static u32 nbio_v7_4_get_memsize(struct amdgpu_device *adev)
 }
 
 static void nbio_v7_4_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
-					  bool use_doorbell, int doorbell_index)
+			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
 	u32 reg = instance == 0 ? SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA0_DOORBELL_RANGE) :
 			SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA1_DOORBELL_RANGE);
@@ -77,7 +74,7 @@ static void nbio_v7_4_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 
 	if (use_doorbell) {
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, OFFSET, doorbell_index);
-		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 2);
+		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, doorbell_size);
 	} else
 		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 0);
 
@@ -93,7 +90,20 @@ static void nbio_v7_4_enable_doorbell_aperture(struct amdgpu_device *adev,
 static void nbio_v7_4_enable_doorbell_selfring_aperture(struct amdgpu_device *adev,
 							bool enable)
 {
+	u32 tmp = 0;
+
+	if (enable) {
+		tmp = REG_SET_FIELD(tmp, DOORBELL_SELFRING_GPA_APER_CNTL, DOORBELL_SELFRING_GPA_APER_EN, 1) |
+		      REG_SET_FIELD(tmp, DOORBELL_SELFRING_GPA_APER_CNTL, DOORBELL_SELFRING_GPA_APER_MODE, 1) |
+		      REG_SET_FIELD(tmp, DOORBELL_SELFRING_GPA_APER_CNTL, DOORBELL_SELFRING_GPA_APER_SIZE, 0);
+
+		WREG32_SOC15(NBIO, 0, mmDOORBELL_SELFRING_GPA_APER_BASE_LOW,
+			     lower_32_bits(adev->doorbell.base));
+		WREG32_SOC15(NBIO, 0, mmDOORBELL_SELFRING_GPA_APER_BASE_HIGH,
+			     upper_32_bits(adev->doorbell.base));
+	}
 
+	WREG32_SOC15(NBIO, 0, mmDOORBELL_SELFRING_GPA_APER_CNTL, tmp);
 }
 
 static void nbio_v7_4_ih_doorbell_range(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
index 0de00fbe9233..f3a7d207af07 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
@@ -191,7 +191,7 @@ enum psp_gfx_fw_type
     GFX_FW_TYPE_MMSCH       = 19,
     GFX_FW_TYPE_RLC_RESTORE_LIST_GPM_MEM        = 20,
     GFX_FW_TYPE_RLC_RESTORE_LIST_SRM_MEM        = 21,
-    GFX_FW_TYPE_RLC_RESTORE_LIST_CNTL           = 22,
+    GFX_FW_TYPE_RLC_RESTORE_LIST_SRM_CNTL       = 22,
     GFX_FW_TYPE_UVD1        = 23,
     GFX_FW_TYPE_MAX         = 24
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
index d78b4306a36f..77c2bc344dfc 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
@@ -38,75 +38,6 @@ MODULE_FIRMWARE("amdgpu/raven_asd.bin");
 MODULE_FIRMWARE("amdgpu/picasso_asd.bin");
 MODULE_FIRMWARE("amdgpu/raven2_asd.bin");
 
-static int
-psp_v10_0_get_fw_type(struct amdgpu_firmware_info *ucode, enum psp_gfx_fw_type *type)
-{
-	switch(ucode->ucode_id) {
-	case AMDGPU_UCODE_ID_SDMA0:
-		*type = GFX_FW_TYPE_SDMA0;
-		break;
-	case AMDGPU_UCODE_ID_SDMA1:
-		*type = GFX_FW_TYPE_SDMA1;
-		break;
-	case AMDGPU_UCODE_ID_CP_CE:
-		*type = GFX_FW_TYPE_CP_CE;
-		break;
-	case AMDGPU_UCODE_ID_CP_PFP:
-		*type = GFX_FW_TYPE_CP_PFP;
-		break;
-	case AMDGPU_UCODE_ID_CP_ME:
-		*type = GFX_FW_TYPE_CP_ME;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME1;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME2;
-		break;
-	case AMDGPU_UCODE_ID_RLC_G:
-		*type = GFX_FW_TYPE_RLC_G;
-		break;
-	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL:
-		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_CNTL;
-		break;
-	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM:
-		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_GPM_MEM;
-		break;
-	case AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM:
-		*type = GFX_FW_TYPE_RLC_RESTORE_LIST_SRM_MEM;
-		break;
-	case AMDGPU_UCODE_ID_SMC:
-		*type = GFX_FW_TYPE_SMU;
-		break;
-	case AMDGPU_UCODE_ID_UVD:
-		*type = GFX_FW_TYPE_UVD;
-		break;
-	case AMDGPU_UCODE_ID_VCE:
-		*type = GFX_FW_TYPE_VCE;
-		break;
-	case AMDGPU_UCODE_ID_VCN:
-		*type = GFX_FW_TYPE_VCN;
-		break;
-	case AMDGPU_UCODE_ID_DMCU_ERAM:
-		*type = GFX_FW_TYPE_DMCU_ERAM;
-		break;
-	case AMDGPU_UCODE_ID_DMCU_INTV:
-		*type = GFX_FW_TYPE_DMCU_ISR;
-		break;
-	case AMDGPU_UCODE_ID_MAXIMUM:
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static int psp_v10_0_init_microcode(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -158,26 +89,6 @@ out:
 	return err;
 }
 
-static int psp_v10_0_prep_cmd_buf(struct amdgpu_firmware_info *ucode,
-				  struct psp_gfx_cmd_resp *cmd)
-{
-	int ret;
-	uint64_t fw_mem_mc_addr = ucode->mc_addr;
-
-	memset(cmd, 0, sizeof(struct psp_gfx_cmd_resp));
-
-	cmd->cmd_id = GFX_CMD_ID_LOAD_IP_FW;
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_lo = lower_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_hi = upper_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_size = ucode->ucode_size;
-
-	ret = psp_v10_0_get_fw_type(ucode, &cmd->cmd.cmd_load_ip_fw.fw_type);
-	if (ret)
-		DRM_ERROR("Unknown firmware type\n");
-
-	return ret;
-}
-
 static int psp_v10_0_ring_init(struct psp_context *psp,
 			       enum psp_ring_type ring_type)
 {
@@ -454,7 +365,6 @@ static int psp_v10_0_mode1_reset(struct psp_context *psp)
 
 static const struct psp_funcs psp_v10_0_funcs = {
 	.init_microcode = psp_v10_0_init_microcode,
-	.prep_cmd_buf = psp_v10_0_prep_cmd_buf,
 	.ring_init = psp_v10_0_ring_init,
 	.ring_create = psp_v10_0_ring_create,
 	.ring_stop = psp_v10_0_ring_stop,
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 0c6e7f9b143f..860b70d80d3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -40,60 +40,6 @@ MODULE_FIRMWARE("amdgpu/vega20_ta.bin");
 /* address block */
 #define smnMP1_FIRMWARE_FLAGS		0x3010024
 
-static int
-psp_v11_0_get_fw_type(struct amdgpu_firmware_info *ucode, enum psp_gfx_fw_type *type)
-{
-	switch (ucode->ucode_id) {
-	case AMDGPU_UCODE_ID_SDMA0:
-		*type = GFX_FW_TYPE_SDMA0;
-		break;
-	case AMDGPU_UCODE_ID_SDMA1:
-		*type = GFX_FW_TYPE_SDMA1;
-		break;
-	case AMDGPU_UCODE_ID_CP_CE:
-		*type = GFX_FW_TYPE_CP_CE;
-		break;
-	case AMDGPU_UCODE_ID_CP_PFP:
-		*type = GFX_FW_TYPE_CP_PFP;
-		break;
-	case AMDGPU_UCODE_ID_CP_ME:
-		*type = GFX_FW_TYPE_CP_ME;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME1;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME2;
-		break;
-	case AMDGPU_UCODE_ID_RLC_G:
-		*type = GFX_FW_TYPE_RLC_G;
-		break;
-	case AMDGPU_UCODE_ID_SMC:
-		*type = GFX_FW_TYPE_SMU;
-		break;
-	case AMDGPU_UCODE_ID_UVD:
-		*type = GFX_FW_TYPE_UVD;
-		break;
-	case AMDGPU_UCODE_ID_VCE:
-		*type = GFX_FW_TYPE_VCE;
-		break;
-	case AMDGPU_UCODE_ID_UVD1:
-		*type = GFX_FW_TYPE_UVD1;
-		break;
-	case AMDGPU_UCODE_ID_MAXIMUM:
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static int psp_v11_0_init_microcode(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -152,18 +98,22 @@ static int psp_v11_0_init_microcode(struct psp_context *psp)
 
 	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)
-		goto out2;
-
-	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_xgmi_ucode_version = le32_to_cpu(ta_hdr->ta_xgmi_ucode_version);
-	adev->psp.ta_xgmi_ucode_size = le32_to_cpu(ta_hdr->ta_xgmi_size_bytes);
-	adev->psp.ta_xgmi_start_addr = (uint8_t *)ta_hdr +
-		le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes);
+	if (err) {
+		release_firmware(adev->psp.ta_fw);
+		adev->psp.ta_fw = NULL;
+		dev_info(adev->dev,
+			 "psp v11.0: 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_xgmi_ucode_version = le32_to_cpu(ta_hdr->ta_xgmi_ucode_version);
+		adev->psp.ta_xgmi_ucode_size = le32_to_cpu(ta_hdr->ta_xgmi_size_bytes);
+		adev->psp.ta_xgmi_start_addr = (uint8_t *)ta_hdr +
+			le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes);
+	}
 
 	return 0;
 
@@ -267,26 +217,6 @@ static int psp_v11_0_bootloader_load_sos(struct psp_context *psp)
 	return ret;
 }
 
-static int psp_v11_0_prep_cmd_buf(struct amdgpu_firmware_info *ucode,
-				 struct psp_gfx_cmd_resp *cmd)
-{
-	int ret;
-	uint64_t fw_mem_mc_addr = ucode->mc_addr;
-
-	memset(cmd, 0, sizeof(struct psp_gfx_cmd_resp));
-
-	cmd->cmd_id = GFX_CMD_ID_LOAD_IP_FW;
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_lo = lower_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_hi = upper_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_size = ucode->ucode_size;
-
-	ret = psp_v11_0_get_fw_type(ucode, &cmd->cmd.cmd_load_ip_fw.fw_type);
-	if (ret)
-		DRM_ERROR("Unknown firmware type\n");
-
-	return ret;
-}
-
 static int psp_v11_0_ring_init(struct psp_context *psp,
 			      enum psp_ring_type ring_type)
 {
@@ -753,7 +683,6 @@ static const struct psp_funcs psp_v11_0_funcs = {
 	.init_microcode = psp_v11_0_init_microcode,
 	.bootloader_load_sysdrv = psp_v11_0_bootloader_load_sysdrv,
 	.bootloader_load_sos = psp_v11_0_bootloader_load_sos,
-	.prep_cmd_buf = psp_v11_0_prep_cmd_buf,
 	.ring_init = psp_v11_0_ring_init,
 	.ring_create = psp_v11_0_ring_create,
 	.ring_stop = psp_v11_0_ring_stop,
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
index 79694ff16969..c63de945c021 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
@@ -47,57 +47,6 @@ MODULE_FIRMWARE("amdgpu/vega12_asd.bin");
 
 static uint32_t sos_old_versions[] = {1517616, 1510592, 1448594, 1446554};
 
-static int
-psp_v3_1_get_fw_type(struct amdgpu_firmware_info *ucode, enum psp_gfx_fw_type *type)
-{
-	switch(ucode->ucode_id) {
-	case AMDGPU_UCODE_ID_SDMA0:
-		*type = GFX_FW_TYPE_SDMA0;
-		break;
-	case AMDGPU_UCODE_ID_SDMA1:
-		*type = GFX_FW_TYPE_SDMA1;
-		break;
-	case AMDGPU_UCODE_ID_CP_CE:
-		*type = GFX_FW_TYPE_CP_CE;
-		break;
-	case AMDGPU_UCODE_ID_CP_PFP:
-		*type = GFX_FW_TYPE_CP_PFP;
-		break;
-	case AMDGPU_UCODE_ID_CP_ME:
-		*type = GFX_FW_TYPE_CP_ME;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC1_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME1;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2:
-		*type = GFX_FW_TYPE_CP_MEC;
-		break;
-	case AMDGPU_UCODE_ID_CP_MEC2_JT:
-		*type = GFX_FW_TYPE_CP_MEC_ME2;
-		break;
-	case AMDGPU_UCODE_ID_RLC_G:
-		*type = GFX_FW_TYPE_RLC_G;
-		break;
-	case AMDGPU_UCODE_ID_SMC:
-		*type = GFX_FW_TYPE_SMU;
-		break;
-	case AMDGPU_UCODE_ID_UVD:
-		*type = GFX_FW_TYPE_UVD;
-		break;
-	case AMDGPU_UCODE_ID_VCE:
-		*type = GFX_FW_TYPE_VCE;
-		break;
-	case AMDGPU_UCODE_ID_MAXIMUM:
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static int psp_v3_1_init_microcode(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -277,26 +226,6 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp)
 	return ret;
 }
 
-static int psp_v3_1_prep_cmd_buf(struct amdgpu_firmware_info *ucode,
-				 struct psp_gfx_cmd_resp *cmd)
-{
-	int ret;
-	uint64_t fw_mem_mc_addr = ucode->mc_addr;
-
-	memset(cmd, 0, sizeof(struct psp_gfx_cmd_resp));
-
-	cmd->cmd_id = GFX_CMD_ID_LOAD_IP_FW;
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_lo = lower_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_phy_addr_hi = upper_32_bits(fw_mem_mc_addr);
-	cmd->cmd.cmd_load_ip_fw.fw_size = ucode->ucode_size;
-
-	ret = psp_v3_1_get_fw_type(ucode, &cmd->cmd.cmd_load_ip_fw.fw_type);
-	if (ret)
-		DRM_ERROR("Unknown firmware type\n");
-
-	return ret;
-}
-
 static int psp_v3_1_ring_init(struct psp_context *psp,
 			      enum psp_ring_type ring_type)
 {
@@ -615,7 +544,6 @@ static const struct psp_funcs psp_v3_1_funcs = {
 	.init_microcode = psp_v3_1_init_microcode,
 	.bootloader_load_sysdrv = psp_v3_1_bootloader_load_sysdrv,
 	.bootloader_load_sos = psp_v3_1_bootloader_load_sos,
-	.prep_cmd_buf = psp_v3_1_prep_cmd_buf,
 	.ring_init = psp_v3_1_ring_init,
 	.ring_create = psp_v3_1_ring_create,
 	.ring_stop = psp_v3_1_ring_stop,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index 9f3cb2aec7c2..cca3552b36ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -247,7 +247,7 @@ static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void sdma_v2_4_ring_emit_ib(struct amdgpu_ring *ring,
 				   struct amdgpu_job *job,
 				   struct amdgpu_ib *ib,
-				   bool ctx_switch)
+				   uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 1bccc5fe2d9d..0ce8331baeb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -421,7 +421,7 @@ static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void sdma_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
 				   struct amdgpu_job *job,
 				   struct amdgpu_ib *ib,
-				   bool ctx_switch)
+				   uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
@@ -1145,8 +1145,7 @@ static int sdma_v3_0_sw_init(void *handle)
 		ring->ring_obj = NULL;
 		if (!amdgpu_sriov_vf(adev)) {
 			ring->use_doorbell = true;
-			ring->doorbell_index = (i == 0) ?
-				adev->doorbell_index.sdma_engine0 : adev->doorbell_index.sdma_engine1;
+			ring->doorbell_index = adev->doorbell_index.sdma_engine[i];
 		} else {
 			ring->use_pollmem = true;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index fd0bfe140ee0..127b85983e8f 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -78,7 +78,6 @@ static const struct soc15_reg_golden golden_settings_sdma_4[] = {
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xfc000000, 0x00000000),
-	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831f07),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CLK_CTRL, 0xffffffff, 0x3f000100),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GFX_IB_CNTL, 0x800f0100, 0x00000100),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000),
@@ -96,6 +95,7 @@ static const struct soc15_reg_golden golden_settings_sdma_4[] = {
 static const struct soc15_reg_golden golden_settings_sdma_vg10[] = {
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00104002),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104002),
+	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831d07),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0018773f, 0x00104002),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104002)
 };
@@ -103,6 +103,7 @@ static const struct soc15_reg_golden golden_settings_sdma_vg10[] = {
 static const struct soc15_reg_golden golden_settings_sdma_vg12[] = {
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00104001),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104001),
+	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831d07),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0018773f, 0x00104001),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104001)
 };
@@ -499,7 +500,7 @@ static void sdma_v4_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void sdma_v4_0_ring_emit_ib(struct amdgpu_ring *ring,
 				   struct amdgpu_job *job,
 				   struct amdgpu_ib *ib,
-				   bool ctx_switch)
+				   uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
@@ -833,8 +834,6 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 					OFFSET, ring->doorbell_index);
 	WREG32_SDMA(i, mmSDMA0_GFX_DOORBELL, doorbell);
 	WREG32_SDMA(i, mmSDMA0_GFX_DOORBELL_OFFSET, doorbell_offset);
-	adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell,
-					      ring->doorbell_index);
 
 	sdma_v4_0_ring_set_wptr(ring);
 
@@ -1521,9 +1520,7 @@ static int sdma_v4_0_sw_init(void *handle)
 				ring->use_doorbell?"true":"false");
 
 		/* doorbell size is 2 dwords, get DWORD offset */
-		ring->doorbell_index = (i == 0) ?
-			(adev->doorbell_index.sdma_engine0 << 1)
-			: (adev->doorbell_index.sdma_engine1 << 1);
+		ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
 
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024,
@@ -1542,9 +1539,7 @@ static int sdma_v4_0_sw_init(void *handle)
 			/* paging queue use same doorbell index/routing as gfx queue
 			 * with 0x400 (4096 dwords) offset on second doorbell page
 			 */
-			ring->doorbell_index = (i == 0) ?
-				(adev->doorbell_index.sdma_engine0 << 1)
-				: (adev->doorbell_index.sdma_engine1 << 1);
+			ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
 			ring->doorbell_index += 0x400;
 
 			sprintf(ring->name, "page%d", i);
diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c
index f8408f88cd37..79c1a9bbcc21 100644
--- a/drivers/gpu/drm/amd/amdgpu/si.c
+++ b/drivers/gpu/drm/amd/amdgpu/si.c
@@ -47,6 +47,7 @@
 #include "dce/dce_6_0_d.h"
 #include "uvd/uvd_4_0_d.h"
 #include "bif/bif_3_0_d.h"
+#include "bif/bif_3_0_sh_mask.h"
 
 static const u32 tahiti_golden_registers[] =
 {
@@ -1258,6 +1259,11 @@ static bool si_need_full_reset(struct amdgpu_device *adev)
 	return true;
 }
 
+static bool si_need_reset_on_init(struct amdgpu_device *adev)
+{
+	return false;
+}
+
 static int si_get_pcie_lanes(struct amdgpu_device *adev)
 {
 	u32 link_width_cntl;
@@ -1323,6 +1329,52 @@ static void si_set_pcie_lanes(struct amdgpu_device *adev, int lanes)
 	WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, link_width_cntl);
 }
 
+static void si_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0,
+			      uint64_t *count1)
+{
+	uint32_t perfctr = 0;
+	uint64_t cnt0_of, cnt1_of;
+	int tmp;
+
+	/* This reports 0 on APUs, so return to avoid writing/reading registers
+	 * that may or may not be different from their GPU counterparts
+	 */
+        if (adev->flags & AMD_IS_APU)
+                return;
+
+	/* Set the 2 events that we wish to watch, defined above */
+	/* Reg 40 is # received msgs, Reg 104 is # of posted requests sent */
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT0_SEL, 40);
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT1_SEL, 104);
+
+	/* Write to enable desired perf counters */
+	WREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK, perfctr);
+	/* Zero out and enable the perf counters
+	 * Write 0x5:
+	 * Bit 0 = Start all counters(1)
+	 * Bit 2 = Global counter reset enable(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000005);
+
+	msleep(1000);
+
+	/* Load the shadow and disable the perf counters
+	 * Write 0x2:
+	 * Bit 0 = Stop counters(0)
+	 * Bit 1 = Load the shadow counters(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000002);
+
+	/* Read register values to get any >32bit overflow */
+	tmp = RREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK);
+	cnt0_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER0_UPPER);
+	cnt1_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER1_UPPER);
+
+	/* Get the values and add the overflow */
+	*count0 = RREG32_PCIE(ixPCIE_PERF_COUNT0_TXCLK) | (cnt0_of << 32);
+	*count1 = RREG32_PCIE(ixPCIE_PERF_COUNT1_TXCLK) | (cnt1_of << 32);
+}
+
 static const struct amdgpu_asic_funcs si_asic_funcs =
 {
 	.read_disabled_bios = &si_read_disabled_bios,
@@ -1339,6 +1391,8 @@ static const struct amdgpu_asic_funcs si_asic_funcs =
 	.flush_hdp = &si_flush_hdp,
 	.invalidate_hdp = &si_invalidate_hdp,
 	.need_full_reset = &si_need_full_reset,
+	.get_pcie_usage = &si_get_pcie_usage,
+	.need_reset_on_init = &si_need_reset_on_init,
 };
 
 static uint32_t si_get_rev_id(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index b6e473134e19..f15f196684ba 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -63,7 +63,7 @@ static void si_dma_ring_set_wptr(struct amdgpu_ring *ring)
 static void si_dma_ring_emit_ib(struct amdgpu_ring *ring,
 				struct amdgpu_job *job,
 				struct amdgpu_ib *ib,
-				bool ctx_switch)
+				uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	/* The indirect buffer packet must end on an 8 DW boundary in the DMA ring.
diff --git a/drivers/gpu/drm/amd/amdgpu/si_ih.c b/drivers/gpu/drm/amd/amdgpu/si_ih.c
index 2938fb9f17cc..8c50c9cab455 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_ih.c
@@ -57,9 +57,9 @@ static void si_ih_disable_interrupts(struct amdgpu_device *adev)
 
 static int si_ih_irq_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_ih_ring *ih = &adev->irq.ih;
 	int rb_bufsz;
 	u32 interrupt_cntl, ih_cntl, ih_rb_cntl;
-	u64 wptr_off;
 
 	si_ih_disable_interrupts(adev);
 	WREG32(INTERRUPT_CNTL2, adev->irq.ih.gpu_addr >> 8);
@@ -76,9 +76,8 @@ static int si_ih_irq_init(struct amdgpu_device *adev)
 		     (rb_bufsz << 1) |
 		     IH_WPTR_WRITEBACK_ENABLE;
 
-	wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32(IH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32(IH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+	WREG32(IH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr));
+	WREG32(IH_RB_WPTR_ADDR_HI, upper_32_bits(ih->wptr_addr) & 0xFF);
 	WREG32(IH_RB_CNTL, ih_rb_cntl);
 	WREG32(IH_RB_RPTR, 0);
 	WREG32(IH_RB_WPTR, 0);
@@ -100,34 +99,36 @@ static void si_ih_irq_disable(struct amdgpu_device *adev)
 	mdelay(1);
 }
 
-static u32 si_ih_get_wptr(struct amdgpu_device *adev)
+static u32 si_ih_get_wptr(struct amdgpu_device *adev,
+			  struct amdgpu_ih_ring *ih)
 {
 	u32 wptr, tmp;
 
-	wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
 
 	if (wptr & IH_RB_WPTR__RB_OVERFLOW_MASK) {
 		wptr &= ~IH_RB_WPTR__RB_OVERFLOW_MASK;
 		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, (wptr + 16) & adev->irq.ih.ptr_mask);
-		adev->irq.ih.rptr = (wptr + 16) & adev->irq.ih.ptr_mask;
+			wptr, ih->rptr, (wptr + 16) & ih->ptr_mask);
+		ih->rptr = (wptr + 16) & ih->ptr_mask;
 		tmp = RREG32(IH_RB_CNTL);
 		tmp |= IH_RB_CNTL__WPTR_OVERFLOW_CLEAR_MASK;
 		WREG32(IH_RB_CNTL, tmp);
 	}
-	return (wptr & adev->irq.ih.ptr_mask);
+	return (wptr & ih->ptr_mask);
 }
 
 static void si_ih_decode_iv(struct amdgpu_device *adev,
-			     struct amdgpu_iv_entry *entry)
+			    struct amdgpu_ih_ring *ih,
+			    struct amdgpu_iv_entry *entry)
 {
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[4];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
 
 	entry->client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
 	entry->src_id = dw[0] & 0xff;
@@ -135,12 +136,13 @@ static void si_ih_decode_iv(struct amdgpu_device *adev,
 	entry->ring_id = dw[2] & 0xff;
 	entry->vmid = (dw[2] >> 8) & 0xff;
 
-	adev->irq.ih.rptr += 16;
+	ih->rptr += 16;
 }
 
-static void si_ih_set_rptr(struct amdgpu_device *adev)
+static void si_ih_set_rptr(struct amdgpu_device *adev,
+			   struct amdgpu_ih_ring *ih)
 {
-	WREG32(IH_RB_RPTR, adev->irq.ih.rptr);
+	WREG32(IH_RB_RPTR, ih->rptr);
 }
 
 static int si_ih_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 8849b74078d6..99ebcf29dcb0 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -43,6 +43,10 @@
 #include "hdp/hdp_4_0_sh_mask.h"
 #include "smuio/smuio_9_0_offset.h"
 #include "smuio/smuio_9_0_sh_mask.h"
+#include "nbio/nbio_7_0_default.h"
+#include "nbio/nbio_7_0_sh_mask.h"
+#include "nbio/nbio_7_0_smn.h"
+#include "mp/mp_9_0_offset.h"
 
 #include "soc15.h"
 #include "soc15_common.h"
@@ -385,14 +389,13 @@ void soc15_program_register_sequence(struct amdgpu_device *adev,
 
 }
 
-
-static int soc15_asic_reset(struct amdgpu_device *adev)
+static int soc15_asic_mode1_reset(struct amdgpu_device *adev)
 {
 	u32 i;
 
 	amdgpu_atombios_scratch_regs_engine_hung(adev, true);
 
-	dev_info(adev->dev, "GPU reset\n");
+	dev_info(adev->dev, "GPU mode1 reset\n");
 
 	/* disable BM */
 	pci_clear_master(adev->pdev);
@@ -417,6 +420,63 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
 	return 0;
 }
 
+static int soc15_asic_get_baco_capability(struct amdgpu_device *adev, bool *cap)
+{
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->get_asic_baco_capability) {
+		*cap = false;
+		return -ENOENT;
+	}
+
+	return pp_funcs->get_asic_baco_capability(pp_handle, cap);
+}
+
+static int soc15_asic_baco_reset(struct amdgpu_device *adev)
+{
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs ||!pp_funcs->get_asic_baco_state ||!pp_funcs->set_asic_baco_state)
+		return -ENOENT;
+
+	/* enter BACO state */
+	if (pp_funcs->set_asic_baco_state(pp_handle, 1))
+		return -EIO;
+
+	/* exit BACO state */
+	if (pp_funcs->set_asic_baco_state(pp_handle, 0))
+		return -EIO;
+
+	dev_info(adev->dev, "GPU BACO reset\n");
+
+	return 0;
+}
+
+static int soc15_asic_reset(struct amdgpu_device *adev)
+{
+	int ret;
+	bool baco_reset;
+
+	switch (adev->asic_type) {
+	case CHIP_VEGA10:
+	case CHIP_VEGA20:
+		soc15_asic_get_baco_capability(adev, &baco_reset);
+		break;
+	default:
+		baco_reset = false;
+		break;
+	}
+
+	if (baco_reset)
+		ret = soc15_asic_baco_reset(adev);
+	else
+		ret = soc15_asic_mode1_reset(adev);
+
+	return ret;
+}
+
 /*static int soc15_set_uvd_clock(struct amdgpu_device *adev, u32 clock,
 			u32 cntl_reg, u32 status_reg)
 {
@@ -535,10 +595,12 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &vega10_common_ip_block);
 		amdgpu_device_ip_block_add(adev, &gmc_v9_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block);
-		if (adev->asic_type == CHIP_VEGA20)
-			amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
-		else
-			amdgpu_device_ip_block_add(adev, &psp_v3_1_ip_block);
+		if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP)) {
+			if (adev->asic_type == CHIP_VEGA20)
+				amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
+			else
+				amdgpu_device_ip_block_add(adev, &psp_v3_1_ip_block);
+		}
 		amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &sdma_v4_0_ip_block);
 		if (!amdgpu_sriov_vf(adev))
@@ -560,7 +622,8 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &vega10_common_ip_block);
 		amdgpu_device_ip_block_add(adev, &gmc_v9_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block);
-		amdgpu_device_ip_block_add(adev, &psp_v10_0_ip_block);
+		if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP))
+			amdgpu_device_ip_block_add(adev, &psp_v10_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &sdma_v4_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
@@ -601,6 +664,68 @@ static bool soc15_need_full_reset(struct amdgpu_device *adev)
 	/* change this when we implement soft reset */
 	return true;
 }
+static void soc15_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0,
+				 uint64_t *count1)
+{
+	uint32_t perfctr = 0;
+	uint64_t cnt0_of, cnt1_of;
+	int tmp;
+
+	/* This reports 0 on APUs, so return to avoid writing/reading registers
+	 * that may or may not be different from their GPU counterparts
+	 */
+	 if (adev->flags & AMD_IS_APU)
+		 return;
+
+	/* Set the 2 events that we wish to watch, defined above */
+	/* Reg 40 is # received msgs, Reg 104 is # of posted requests sent */
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT0_SEL, 40);
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT1_SEL, 104);
+
+	/* Write to enable desired perf counters */
+	WREG32_PCIE(smnPCIE_PERF_CNTL_TXCLK, perfctr);
+	/* Zero out and enable the perf counters
+	 * Write 0x5:
+	 * Bit 0 = Start all counters(1)
+	 * Bit 2 = Global counter reset enable(1)
+	 */
+	WREG32_PCIE(smnPCIE_PERF_COUNT_CNTL, 0x00000005);
+
+	msleep(1000);
+
+	/* Load the shadow and disable the perf counters
+	 * Write 0x2:
+	 * Bit 0 = Stop counters(0)
+	 * Bit 1 = Load the shadow counters(1)
+	 */
+	WREG32_PCIE(smnPCIE_PERF_COUNT_CNTL, 0x00000002);
+
+	/* Read register values to get any >32bit overflow */
+	tmp = RREG32_PCIE(smnPCIE_PERF_CNTL_TXCLK);
+	cnt0_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER0_UPPER);
+	cnt1_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER1_UPPER);
+
+	/* Get the values and add the overflow */
+	*count0 = RREG32_PCIE(smnPCIE_PERF_COUNT0_TXCLK) | (cnt0_of << 32);
+	*count1 = RREG32_PCIE(smnPCIE_PERF_COUNT1_TXCLK) | (cnt1_of << 32);
+}
+
+static bool soc15_need_reset_on_init(struct amdgpu_device *adev)
+{
+	u32 sol_reg;
+
+	if (adev->flags & AMD_IS_APU)
+		return false;
+
+	/* Check sOS sign of life register to confirm sys driver and sOS
+	 * are already been loaded.
+	 */
+	sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81);
+	if (sol_reg)
+		return true;
+
+	return false;
+}
 
 static const struct amdgpu_asic_funcs soc15_asic_funcs =
 {
@@ -617,6 +742,8 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs =
 	.invalidate_hdp = &soc15_invalidate_hdp,
 	.need_full_reset = &soc15_need_full_reset,
 	.init_doorbell_index = &vega10_doorbell_index_init,
+	.get_pcie_usage = &soc15_get_pcie_usage,
+	.need_reset_on_init = &soc15_need_reset_on_init,
 };
 
 static const struct amdgpu_asic_funcs vega20_asic_funcs =
@@ -634,6 +761,8 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs =
 	.invalidate_hdp = &soc15_invalidate_hdp,
 	.need_full_reset = &soc15_need_full_reset,
 	.init_doorbell_index = &vega20_doorbell_index_init,
+	.get_pcie_usage = &soc15_get_pcie_usage,
+	.need_reset_on_init = &soc15_need_reset_on_init,
 };
 
 static int soc15_common_early_init(void *handle)
@@ -729,11 +858,13 @@ static int soc15_common_early_init(void *handle)
 	case CHIP_RAVEN:
 		adev->asic_funcs = &soc15_asic_funcs;
 		if (adev->rev_id >= 0x8)
-			adev->external_rev_id = adev->rev_id + 0x81;
+			adev->external_rev_id = adev->rev_id + 0x79;
 		else if (adev->pdev->device == 0x15d8)
 			adev->external_rev_id = adev->rev_id + 0x41;
+		else if (adev->rev_id == 1)
+			adev->external_rev_id = adev->rev_id + 0x20;
 		else
-			adev->external_rev_id = 0x1;
+			adev->external_rev_id = adev->rev_id + 0x01;
 
 		if (adev->rev_id >= 0x8) {
 			adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
@@ -840,6 +971,22 @@ static int soc15_common_sw_fini(void *handle)
 	return 0;
 }
 
+static void soc15_doorbell_range_init(struct amdgpu_device *adev)
+{
+	int i;
+	struct amdgpu_ring *ring;
+
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
+		adev->nbio_funcs->sdma_doorbell_range(adev, i,
+			ring->use_doorbell, ring->doorbell_index,
+			adev->doorbell_index.sdma_doorbell_range);
+	}
+
+	adev->nbio_funcs->ih_doorbell_range(adev, adev->irq.ih.use_doorbell,
+						adev->irq.ih.doorbell_index);
+}
+
 static int soc15_common_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -852,6 +999,12 @@ static int soc15_common_hw_init(void *handle)
 	adev->nbio_funcs->init_registers(adev);
 	/* enable the doorbell aperture */
 	soc15_enable_doorbell_aperture(adev, true);
+	/* HW doorbell routing policy: doorbell writing not
+	 * in SDMA/IH/MM/ACV range will be routed to CP. So
+	 * we need to init SDMA/IH/MM/ACV doorbell range prior
+	 * to CP ip block init and ring test.
+	 */
+	soc15_doorbell_range_init(adev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c
index 15da06ddeb75..a20b711a6756 100644
--- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c
@@ -99,9 +99,9 @@ static void tonga_ih_disable_interrupts(struct amdgpu_device *adev)
  */
 static int tonga_ih_irq_init(struct amdgpu_device *adev)
 {
-	int rb_bufsz;
 	u32 interrupt_cntl, ih_rb_cntl, ih_doorbell_rtpr;
-	u64 wptr_off;
+	struct amdgpu_ih_ring *ih = &adev->irq.ih;
+	int rb_bufsz;
 
 	/* disable irqs */
 	tonga_ih_disable_interrupts(adev);
@@ -118,10 +118,7 @@ static int tonga_ih_irq_init(struct amdgpu_device *adev)
 	WREG32(mmINTERRUPT_CNTL, interrupt_cntl);
 
 	/* Ring Buffer base. [39:8] of 40-bit address of the beginning of the ring buffer*/
-	if (adev->irq.ih.use_bus_addr)
-		WREG32(mmIH_RB_BASE, adev->irq.ih.rb_dma_addr >> 8);
-	else
-		WREG32(mmIH_RB_BASE, adev->irq.ih.gpu_addr >> 8);
+	WREG32(mmIH_RB_BASE, ih->gpu_addr >> 8);
 
 	rb_bufsz = order_base_2(adev->irq.ih.ring_size / 4);
 	ih_rb_cntl = REG_SET_FIELD(0, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
@@ -136,12 +133,8 @@ static int tonga_ih_irq_init(struct amdgpu_device *adev)
 	WREG32(mmIH_RB_CNTL, ih_rb_cntl);
 
 	/* set the writeback address whether it's enabled or not */
-	if (adev->irq.ih.use_bus_addr)
-		wptr_off = adev->irq.ih.rb_dma_addr + (adev->irq.ih.wptr_offs * 4);
-	else
-		wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+	WREG32(mmIH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr));
+	WREG32(mmIH_RB_WPTR_ADDR_HI, upper_32_bits(ih->wptr_addr) & 0xFF);
 
 	/* set rptr, wptr to 0 */
 	WREG32(mmIH_RB_RPTR, 0);
@@ -193,14 +186,12 @@ static void tonga_ih_irq_disable(struct amdgpu_device *adev)
  * Used by cz_irq_process(VI).
  * Returns the value of the wptr.
  */
-static u32 tonga_ih_get_wptr(struct amdgpu_device *adev)
+static u32 tonga_ih_get_wptr(struct amdgpu_device *adev,
+			     struct amdgpu_ih_ring *ih)
 {
 	u32 wptr, tmp;
 
-	if (adev->irq.ih.use_bus_addr)
-		wptr = le32_to_cpu(adev->irq.ih.ring[adev->irq.ih.wptr_offs]);
-	else
-		wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
 
 	if (REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) {
 		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
@@ -209,13 +200,13 @@ static u32 tonga_ih_get_wptr(struct amdgpu_device *adev)
 		 * this should allow us to catchup.
 		 */
 		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, (wptr + 16) & adev->irq.ih.ptr_mask);
-		adev->irq.ih.rptr = (wptr + 16) & adev->irq.ih.ptr_mask;
+			 wptr, ih->rptr, (wptr + 16) & ih->ptr_mask);
+		ih->rptr = (wptr + 16) & ih->ptr_mask;
 		tmp = RREG32(mmIH_RB_CNTL);
 		tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
 		WREG32(mmIH_RB_CNTL, tmp);
 	}
-	return (wptr & adev->irq.ih.ptr_mask);
+	return (wptr & ih->ptr_mask);
 }
 
 /**
@@ -227,16 +218,17 @@ static u32 tonga_ih_get_wptr(struct amdgpu_device *adev)
  * position and also advance the position.
  */
 static void tonga_ih_decode_iv(struct amdgpu_device *adev,
-				 struct amdgpu_iv_entry *entry)
+			       struct amdgpu_ih_ring *ih,
+			       struct amdgpu_iv_entry *entry)
 {
 	/* wptr/rptr are in bytes! */
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[4];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
 
 	entry->client_id = AMDGPU_IRQ_CLIENTID_LEGACY;
 	entry->src_id = dw[0] & 0xff;
@@ -246,7 +238,7 @@ static void tonga_ih_decode_iv(struct amdgpu_device *adev,
 	entry->pasid = (dw[2] >> 16) & 0xffff;
 
 	/* wptr/rptr are in bytes! */
-	adev->irq.ih.rptr += 16;
+	ih->rptr += 16;
 }
 
 /**
@@ -256,17 +248,15 @@ static void tonga_ih_decode_iv(struct amdgpu_device *adev,
  *
  * Set the IH ring buffer rptr.
  */
-static void tonga_ih_set_rptr(struct amdgpu_device *adev)
+static void tonga_ih_set_rptr(struct amdgpu_device *adev,
+			      struct amdgpu_ih_ring *ih)
 {
-	if (adev->irq.ih.use_doorbell) {
+	if (ih->use_doorbell) {
 		/* XXX check if swapping is necessary on BE */
-		if (adev->irq.ih.use_bus_addr)
-			adev->irq.ih.ring[adev->irq.ih.rptr_offs] = adev->irq.ih.rptr;
-		else
-			adev->wb.wb[adev->irq.ih.rptr_offs] = adev->irq.ih.rptr;
-		WDOORBELL32(adev->irq.ih.doorbell_index, adev->irq.ih.rptr);
+		*ih->rptr_cpu = ih->rptr;
+		WDOORBELL32(ih->doorbell_index, ih->rptr);
 	} else {
-		WREG32(mmIH_RB_RPTR, adev->irq.ih.rptr);
+		WREG32(mmIH_RB_RPTR, ih->rptr);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index d69c8f6daaf8..c4fb58667fd4 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -511,7 +511,7 @@ static int uvd_v4_2_ring_test_ring(struct amdgpu_ring *ring)
 static void uvd_v4_2_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	amdgpu_ring_write(ring, PACKET0(mmUVD_RBC_IB_BASE, 0));
 	amdgpu_ring_write(ring, ib->gpu_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index ee8cd06ddc38..52bd8a654734 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -526,7 +526,7 @@ static int uvd_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 static void uvd_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	amdgpu_ring_write(ring, PACKET0(mmUVD_LMI_RBC_IB_64BIT_BAR_LOW, 0));
 	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index d4f4a66f8324..c9edddf9f88a 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -977,7 +977,7 @@ static int uvd_v6_0_ring_test_ring(struct amdgpu_ring *ring)
 static void uvd_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
@@ -1003,7 +1003,7 @@ static void uvd_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
 static void uvd_v6_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index aef924026a28..dc461df48da0 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -1272,7 +1272,7 @@ static int uvd_v7_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p,
 static void uvd_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
@@ -1303,7 +1303,7 @@ static void uvd_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 static void uvd_v7_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index 2668effadd27..6ec65cf11112 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -834,7 +834,7 @@ out:
 static void vce_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
 				  struct amdgpu_job *job,
 				  struct amdgpu_ib *ib,
-				  bool ctx_switch)
+				  uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index 9fb34b7d8e03..aadc3e66ebd7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -947,7 +947,7 @@ static int vce_v4_0_set_powergating_state(void *handle,
 #endif
 
 static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
-					struct amdgpu_ib *ib, bool ctx_switch)
+					struct amdgpu_ib *ib, uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 89bb2fef90eb..3dbc51f9d3b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -1371,7 +1371,7 @@ static void vcn_v1_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
 static void vcn_v1_0_dec_ring_emit_ib(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
@@ -1531,7 +1531,7 @@ static void vcn_v1_0_enc_ring_insert_end(struct amdgpu_ring *ring)
 static void vcn_v1_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 
@@ -1736,7 +1736,7 @@ static void vcn_v1_0_jpeg_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u6
 static void vcn_v1_0_jpeg_ring_emit_ib(struct amdgpu_ring *ring,
 					struct amdgpu_job *job,
 					struct amdgpu_ib *ib,
-					bool ctx_switch)
+					uint32_t flags)
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index 2c250b01a903..6d1f804277f8 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -50,6 +50,22 @@ static void vega10_ih_enable_interrupts(struct amdgpu_device *adev)
 	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, ENABLE_INTR, 1);
 	WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl);
 	adev->irq.ih.enabled = true;
+
+	if (adev->irq.ih1.ring_size) {
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING1,
+					   RB_ENABLE, 1);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl);
+		adev->irq.ih1.enabled = true;
+	}
+
+	if (adev->irq.ih2.ring_size) {
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING2,
+					   RB_ENABLE, 1);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl);
+		adev->irq.ih2.enabled = true;
+	}
 }
 
 /**
@@ -71,6 +87,53 @@ static void vega10_ih_disable_interrupts(struct amdgpu_device *adev)
 	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR, 0);
 	adev->irq.ih.enabled = false;
 	adev->irq.ih.rptr = 0;
+
+	if (adev->irq.ih1.ring_size) {
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING1,
+					   RB_ENABLE, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl);
+		/* set rptr, wptr to 0 */
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING1, 0);
+		adev->irq.ih1.enabled = false;
+		adev->irq.ih1.rptr = 0;
+	}
+
+	if (adev->irq.ih2.ring_size) {
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING2,
+					   RB_ENABLE, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl);
+		/* set rptr, wptr to 0 */
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING2, 0);
+		adev->irq.ih2.enabled = false;
+		adev->irq.ih2.rptr = 0;
+	}
+}
+
+static uint32_t vega10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl)
+{
+	int rb_bufsz = order_base_2(ih->ring_size / 4);
+
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+				   MC_SPACE, ih->use_bus_addr ? 1 : 4);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+				   WPTR_OVERFLOW_CLEAR, 1);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+				   WPTR_OVERFLOW_ENABLE, 1);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RB_SIZE, rb_bufsz);
+	/* Ring Buffer write pointer writeback. If enabled, IH_RB_WPTR register
+	 * value is written to memory
+	 */
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+				   WPTR_WRITEBACK_ENABLE, 1);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_SNOOP, 1);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_RO, 0);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_VMID, 0);
+
+	return ih_rb_cntl;
 }
 
 /**
@@ -86,50 +149,32 @@ static void vega10_ih_disable_interrupts(struct amdgpu_device *adev)
  */
 static int vega10_ih_irq_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_ih_ring *ih;
 	int ret = 0;
-	int rb_bufsz;
 	u32 ih_rb_cntl, ih_doorbell_rtpr;
 	u32 tmp;
-	u64 wptr_off;
 
 	/* disable irqs */
 	vega10_ih_disable_interrupts(adev);
 
 	adev->nbio_funcs->ih_control(adev);
 
-	ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL);
+	ih = &adev->irq.ih;
 	/* Ring Buffer base. [39:8] of 40-bit address of the beginning of the ring buffer*/
-	if (adev->irq.ih.use_bus_addr) {
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE, adev->irq.ih.rb_dma_addr >> 8);
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI, ((u64)adev->irq.ih.rb_dma_addr >> 40) & 0xff);
-		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_SPACE, 1);
-	} else {
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE, adev->irq.ih.gpu_addr >> 8);
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI, (adev->irq.ih.gpu_addr >> 40) & 0xff);
-		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_SPACE, 4);
-	}
-	rb_bufsz = order_base_2(adev->irq.ih.ring_size / 4);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, WPTR_OVERFLOW_ENABLE, 1);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RB_SIZE, rb_bufsz);
-	/* Ring Buffer write pointer writeback. If enabled, IH_RB_WPTR register value is written to memory */
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, WPTR_WRITEBACK_ENABLE, 1);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_SNOOP, 1);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_RO, 0);
-	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_VMID, 0);
-
-	if (adev->irq.msi_enabled)
-		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RPTR_REARM, 1);
+	WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE, ih->gpu_addr >> 8);
+	WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI, (ih->gpu_addr >> 40) & 0xff);
 
+	ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL);
+	ih_rb_cntl = vega10_ih_rb_cntl(ih, ih_rb_cntl);
+	ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RPTR_REARM,
+				   !!adev->irq.msi_enabled);
 	WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl);
 
 	/* set the writeback address whether it's enabled or not */
-	if (adev->irq.ih.use_bus_addr)
-		wptr_off = adev->irq.ih.rb_dma_addr + (adev->irq.ih.wptr_offs * 4);
-	else
-		wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
-	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
-	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFFFF);
+	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_LO,
+		     lower_32_bits(ih->wptr_addr));
+	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI,
+		     upper_32_bits(ih->wptr_addr) & 0xFFFF);
 
 	/* set rptr, wptr to 0 */
 	WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0);
@@ -137,17 +182,48 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
 
 	ih_doorbell_rtpr = RREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR);
 	if (adev->irq.ih.use_doorbell) {
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, IH_DOORBELL_RPTR,
-						 OFFSET, adev->irq.ih.doorbell_index);
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, IH_DOORBELL_RPTR,
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR, OFFSET,
+						 adev->irq.ih.doorbell_index);
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR,
 						 ENABLE, 1);
 	} else {
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, IH_DOORBELL_RPTR,
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR,
 						 ENABLE, 0);
 	}
 	WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR, ih_doorbell_rtpr);
-	adev->nbio_funcs->ih_doorbell_range(adev, adev->irq.ih.use_doorbell,
-					    adev->irq.ih.doorbell_index);
+
+	ih = &adev->irq.ih1;
+	if (ih->ring_size) {
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_RING1, ih->gpu_addr >> 8);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI_RING1,
+			     (ih->gpu_addr >> 40) & 0xff);
+
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+		ih_rb_cntl = vega10_ih_rb_cntl(ih, ih_rb_cntl);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl);
+
+		/* set rptr, wptr to 0 */
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING1, 0);
+	}
+
+	ih = &adev->irq.ih2;
+	if (ih->ring_size) {
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_RING2, ih->gpu_addr >> 8);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI_RING2,
+			     (ih->gpu_addr >> 40) & 0xff);
+
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+		ih_rb_cntl = vega10_ih_rb_cntl(ih, ih_rb_cntl);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl);
+
+		/* set rptr, wptr to 0 */
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING2, 0);
+	}
 
 	tmp = RREG32_SOC15(OSSSYS, 0, mmIH_STORM_CLIENT_LIST_CNTL);
 	tmp = REG_SET_FIELD(tmp, IH_STORM_CLIENT_LIST_CNTL,
@@ -191,32 +267,58 @@ static void vega10_ih_irq_disable(struct amdgpu_device *adev)
  * ring buffer overflow and deal with it.
  * Returns the value of the wptr.
  */
-static u32 vega10_ih_get_wptr(struct amdgpu_device *adev)
+static u32 vega10_ih_get_wptr(struct amdgpu_device *adev,
+			      struct amdgpu_ih_ring *ih)
 {
-	u32 wptr, tmp;
+	u32 wptr, reg, tmp;
 
-	if (adev->irq.ih.use_bus_addr)
-		wptr = le32_to_cpu(adev->irq.ih.ring[adev->irq.ih.wptr_offs]);
+	wptr = le32_to_cpu(*ih->wptr_cpu);
+
+	if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW))
+		goto out;
+
+	/* Double check that the overflow wasn't already cleared. */
+
+	if (ih == &adev->irq.ih)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR);
+	else if (ih == &adev->irq.ih1)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR_RING1);
+	else if (ih == &adev->irq.ih2)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR_RING2);
 	else
-		wptr = le32_to_cpu(adev->wb.wb[adev->irq.ih.wptr_offs]);
-
-	if (REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) {
-		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
-
-		/* When a ring buffer overflow happen start parsing interrupt
-		 * from the last not overwritten vector (wptr + 32). Hopefully
-		 * this should allow us to catchup.
-		 */
-		tmp = (wptr + 32) & adev->irq.ih.ptr_mask;
-		dev_warn(adev->dev, "IH ring buffer overflow (0x%08X, 0x%08X, 0x%08X)\n",
-			wptr, adev->irq.ih.rptr, tmp);
-		adev->irq.ih.rptr = tmp;
-
-		tmp = RREG32_NO_KIQ(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL));
-		tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
-		WREG32_NO_KIQ(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL), tmp);
-	}
-	return (wptr & adev->irq.ih.ptr_mask);
+		BUG();
+
+	wptr = RREG32_NO_KIQ(reg);
+	if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW))
+		goto out;
+
+	wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
+
+	/* When a ring buffer overflow happen start parsing interrupt
+	 * from the last not overwritten vector (wptr + 32). Hopefully
+	 * this should allow us to catchup.
+	 */
+	tmp = (wptr + 32) & ih->ptr_mask;
+	dev_warn(adev->dev, "IH ring buffer overflow "
+		 "(0x%08X, 0x%08X, 0x%08X)\n",
+		 wptr, ih->rptr, tmp);
+	ih->rptr = tmp;
+
+	if (ih == &adev->irq.ih)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL);
+	else if (ih == &adev->irq.ih1)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+	else if (ih == &adev->irq.ih2)
+		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL_RING2);
+	else
+		BUG();
+
+	tmp = RREG32_NO_KIQ(reg);
+	tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1);
+	WREG32_NO_KIQ(reg, tmp);
+
+out:
+	return (wptr & ih->ptr_mask);
 }
 
 /**
@@ -228,20 +330,21 @@ static u32 vega10_ih_get_wptr(struct amdgpu_device *adev)
  * position and also advance the position.
  */
 static void vega10_ih_decode_iv(struct amdgpu_device *adev,
-				 struct amdgpu_iv_entry *entry)
+				struct amdgpu_ih_ring *ih,
+				struct amdgpu_iv_entry *entry)
 {
 	/* wptr/rptr are in bytes! */
-	u32 ring_index = adev->irq.ih.rptr >> 2;
+	u32 ring_index = ih->rptr >> 2;
 	uint32_t dw[8];
 
-	dw[0] = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]);
-	dw[1] = le32_to_cpu(adev->irq.ih.ring[ring_index + 1]);
-	dw[2] = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]);
-	dw[3] = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]);
-	dw[4] = le32_to_cpu(adev->irq.ih.ring[ring_index + 4]);
-	dw[5] = le32_to_cpu(adev->irq.ih.ring[ring_index + 5]);
-	dw[6] = le32_to_cpu(adev->irq.ih.ring[ring_index + 6]);
-	dw[7] = le32_to_cpu(adev->irq.ih.ring[ring_index + 7]);
+	dw[0] = le32_to_cpu(ih->ring[ring_index + 0]);
+	dw[1] = le32_to_cpu(ih->ring[ring_index + 1]);
+	dw[2] = le32_to_cpu(ih->ring[ring_index + 2]);
+	dw[3] = le32_to_cpu(ih->ring[ring_index + 3]);
+	dw[4] = le32_to_cpu(ih->ring[ring_index + 4]);
+	dw[5] = le32_to_cpu(ih->ring[ring_index + 5]);
+	dw[6] = le32_to_cpu(ih->ring[ring_index + 6]);
+	dw[7] = le32_to_cpu(ih->ring[ring_index + 7]);
 
 	entry->client_id = dw[0] & 0xff;
 	entry->src_id = (dw[0] >> 8) & 0xff;
@@ -257,9 +360,8 @@ static void vega10_ih_decode_iv(struct amdgpu_device *adev,
 	entry->src_data[2] = dw[6];
 	entry->src_data[3] = dw[7];
 
-
 	/* wptr/rptr are in bytes! */
-	adev->irq.ih.rptr += 32;
+	ih->rptr += 32;
 }
 
 /**
@@ -269,37 +371,95 @@ static void vega10_ih_decode_iv(struct amdgpu_device *adev,
  *
  * Set the IH ring buffer rptr.
  */
-static void vega10_ih_set_rptr(struct amdgpu_device *adev)
+static void vega10_ih_set_rptr(struct amdgpu_device *adev,
+			       struct amdgpu_ih_ring *ih)
 {
-	if (adev->irq.ih.use_doorbell) {
+	if (ih->use_doorbell) {
 		/* XXX check if swapping is necessary on BE */
-		if (adev->irq.ih.use_bus_addr)
-			adev->irq.ih.ring[adev->irq.ih.rptr_offs] = adev->irq.ih.rptr;
-		else
-			adev->wb.wb[adev->irq.ih.rptr_offs] = adev->irq.ih.rptr;
-		WDOORBELL32(adev->irq.ih.doorbell_index, adev->irq.ih.rptr);
-	} else {
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, adev->irq.ih.rptr);
+		*ih->rptr_cpu = ih->rptr;
+		WDOORBELL32(ih->doorbell_index, ih->rptr);
+	} else if (ih == &adev->irq.ih) {
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, ih->rptr);
+	} else if (ih == &adev->irq.ih1) {
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, ih->rptr);
+	} else if (ih == &adev->irq.ih2) {
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, ih->rptr);
 	}
 }
 
+/**
+ * vega10_ih_self_irq - dispatch work for ring 1 and 2
+ *
+ * @adev: amdgpu_device pointer
+ * @source: irq source
+ * @entry: IV with WPTR update
+ *
+ * Update the WPTR from the IV and schedule work to handle the entries.
+ */
+static int vega10_ih_self_irq(struct amdgpu_device *adev,
+			      struct amdgpu_irq_src *source,
+			      struct amdgpu_iv_entry *entry)
+{
+	uint32_t wptr = cpu_to_le32(entry->src_data[0]);
+
+	switch (entry->ring_id) {
+	case 1:
+		*adev->irq.ih1.wptr_cpu = wptr;
+		schedule_work(&adev->irq.ih1_work);
+		break;
+	case 2:
+		*adev->irq.ih2.wptr_cpu = wptr;
+		schedule_work(&adev->irq.ih2_work);
+		break;
+	default: break;
+	}
+	return 0;
+}
+
+static const struct amdgpu_irq_src_funcs vega10_ih_self_irq_funcs = {
+	.process = vega10_ih_self_irq,
+};
+
+static void vega10_ih_set_self_irq_funcs(struct amdgpu_device *adev)
+{
+	adev->irq.self_irq.num_types = 0;
+	adev->irq.self_irq.funcs = &vega10_ih_self_irq_funcs;
+}
+
 static int vega10_ih_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	vega10_ih_set_interrupt_funcs(adev);
+	vega10_ih_set_self_irq_funcs(adev);
 	return 0;
 }
 
 static int vega10_ih_sw_init(void *handle)
 {
-	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0,
+			      &adev->irq.self_irq);
+	if (r)
+		return r;
 
 	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, true);
 	if (r)
 		return r;
 
+	if (adev->asic_type == CHIP_VEGA10) {
+		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;
+	}
+
+	/* TODO add doorbell for IH1 & IH2 as well */
 	adev->irq.ih.use_doorbell = true;
 	adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1;
 
@@ -313,6 +473,8 @@ static int vega10_ih_sw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	amdgpu_irq_fini(adev);
+	amdgpu_ih_ring_fini(adev, &adev->irq.ih2);
+	amdgpu_ih_ring_fini(adev, &adev->irq.ih1);
 	amdgpu_ih_ring_fini(adev, &adev->irq.ih);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
index 422674bb3cdf..4b5d60ea3e78 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c
@@ -70,8 +70,8 @@ void vega10_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.userqueue_start = AMDGPU_DOORBELL64_USERQUEUE_START;
 	adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL64_USERQUEUE_END;
 	adev->doorbell_index.gfx_ring0 = AMDGPU_DOORBELL64_GFX_RING0;
-	adev->doorbell_index.sdma_engine0 = AMDGPU_DOORBELL64_sDMA_ENGINE0;
-	adev->doorbell_index.sdma_engine1 = AMDGPU_DOORBELL64_sDMA_ENGINE1;
+	adev->doorbell_index.sdma_engine[0] = AMDGPU_DOORBELL64_sDMA_ENGINE0;
+	adev->doorbell_index.sdma_engine[1] = AMDGPU_DOORBELL64_sDMA_ENGINE1;
 	adev->doorbell_index.ih = AMDGPU_DOORBELL64_IH;
 	adev->doorbell_index.uvd_vce.uvd_ring0_1 = AMDGPU_DOORBELL64_UVD_RING0_1;
 	adev->doorbell_index.uvd_vce.uvd_ring2_3 = AMDGPU_DOORBELL64_UVD_RING2_3;
@@ -83,5 +83,6 @@ void vega10_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.uvd_vce.vce_ring6_7 = AMDGPU_DOORBELL64_VCE_RING6_7;
 	/* In unit of dword doorbell */
 	adev->doorbell_index.max_assignment = AMDGPU_DOORBELL64_MAX_ASSIGNMENT << 1;
+	adev->doorbell_index.sdma_doorbell_range = 4;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
index edce413fda9a..53716c593b2b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
@@ -68,14 +68,14 @@ void vega20_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.userqueue_start = AMDGPU_VEGA20_DOORBELL_USERQUEUE_START;
 	adev->doorbell_index.userqueue_end = AMDGPU_VEGA20_DOORBELL_USERQUEUE_END;
 	adev->doorbell_index.gfx_ring0 = AMDGPU_VEGA20_DOORBELL_GFX_RING0;
-	adev->doorbell_index.sdma_engine0 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0;
-	adev->doorbell_index.sdma_engine1 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE1;
-	adev->doorbell_index.sdma_engine2 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE2;
-	adev->doorbell_index.sdma_engine3 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE3;
-	adev->doorbell_index.sdma_engine4 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE4;
-	adev->doorbell_index.sdma_engine5 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE5;
-	adev->doorbell_index.sdma_engine6 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE6;
-	adev->doorbell_index.sdma_engine7 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE7;
+	adev->doorbell_index.sdma_engine[0] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0;
+	adev->doorbell_index.sdma_engine[1] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE1;
+	adev->doorbell_index.sdma_engine[2] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE2;
+	adev->doorbell_index.sdma_engine[3] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE3;
+	adev->doorbell_index.sdma_engine[4] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE4;
+	adev->doorbell_index.sdma_engine[5] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE5;
+	adev->doorbell_index.sdma_engine[6] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE6;
+	adev->doorbell_index.sdma_engine[7] = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE7;
 	adev->doorbell_index.ih = AMDGPU_VEGA20_DOORBELL_IH;
 	adev->doorbell_index.uvd_vce.uvd_ring0_1 = AMDGPU_VEGA20_DOORBELL64_UVD_RING0_1;
 	adev->doorbell_index.uvd_vce.uvd_ring2_3 = AMDGPU_VEGA20_DOORBELL64_UVD_RING2_3;
@@ -86,5 +86,6 @@ void vega20_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.uvd_vce.vce_ring4_5 = AMDGPU_VEGA20_DOORBELL64_VCE_RING4_5;
 	adev->doorbell_index.uvd_vce.vce_ring6_7 = AMDGPU_VEGA20_DOORBELL64_VCE_RING6_7;
 	adev->doorbell_index.max_assignment = AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT << 1;
+	adev->doorbell_index.sdma_doorbell_range = 20;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 77e367459101..5e5b42a0744a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -941,6 +941,69 @@ static bool vi_need_full_reset(struct amdgpu_device *adev)
 	}
 }
 
+static void vi_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0,
+			      uint64_t *count1)
+{
+	uint32_t perfctr = 0;
+	uint64_t cnt0_of, cnt1_of;
+	int tmp;
+
+	/* This reports 0 on APUs, so return to avoid writing/reading registers
+	 * that may or may not be different from their GPU counterparts
+	 */
+	if (adev->flags & AMD_IS_APU)
+		return;
+
+	/* Set the 2 events that we wish to watch, defined above */
+	/* Reg 40 is # received msgs, Reg 104 is # of posted requests sent */
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT0_SEL, 40);
+	perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT1_SEL, 104);
+
+	/* Write to enable desired perf counters */
+	WREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK, perfctr);
+	/* Zero out and enable the perf counters
+	 * Write 0x5:
+	 * Bit 0 = Start all counters(1)
+	 * Bit 2 = Global counter reset enable(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000005);
+
+	msleep(1000);
+
+	/* Load the shadow and disable the perf counters
+	 * Write 0x2:
+	 * Bit 0 = Stop counters(0)
+	 * Bit 1 = Load the shadow counters(1)
+	 */
+	WREG32_PCIE(ixPCIE_PERF_COUNT_CNTL, 0x00000002);
+
+	/* Read register values to get any >32bit overflow */
+	tmp = RREG32_PCIE(ixPCIE_PERF_CNTL_TXCLK);
+	cnt0_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER0_UPPER);
+	cnt1_of = REG_GET_FIELD(tmp, PCIE_PERF_CNTL_TXCLK, COUNTER1_UPPER);
+
+	/* Get the values and add the overflow */
+	*count0 = RREG32_PCIE(ixPCIE_PERF_COUNT0_TXCLK) | (cnt0_of << 32);
+	*count1 = RREG32_PCIE(ixPCIE_PERF_COUNT1_TXCLK) | (cnt1_of << 32);
+}
+
+static bool vi_need_reset_on_init(struct amdgpu_device *adev)
+{
+	u32 clock_cntl, pc;
+
+	if (adev->flags & AMD_IS_APU)
+		return false;
+
+	/* check if the SMC is already running */
+	clock_cntl = RREG32_SMC(ixSMC_SYSCON_CLOCK_CNTL_0);
+	pc = RREG32_SMC(ixSMC_PC_C);
+	if ((0 == REG_GET_FIELD(clock_cntl, SMC_SYSCON_CLOCK_CNTL_0, ck_disable)) &&
+	    (0x20100 <= pc))
+		return true;
+
+	return false;
+}
+
 static const struct amdgpu_asic_funcs vi_asic_funcs =
 {
 	.read_disabled_bios = &vi_read_disabled_bios,
@@ -956,6 +1019,8 @@ static const struct amdgpu_asic_funcs vi_asic_funcs =
 	.invalidate_hdp = &vi_invalidate_hdp,
 	.need_full_reset = &vi_need_full_reset,
 	.init_doorbell_index = &legacy_doorbell_index_init,
+	.get_pcie_usage = &vi_get_pcie_usage,
+	.need_reset_on_init = &vi_need_reset_on_init,
 };
 
 #define CZ_REV_BRISTOL(rev)	 \
@@ -1726,8 +1791,8 @@ void legacy_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.mec_ring6 = AMDGPU_DOORBELL_MEC_RING6;
 	adev->doorbell_index.mec_ring7 = AMDGPU_DOORBELL_MEC_RING7;
 	adev->doorbell_index.gfx_ring0 = AMDGPU_DOORBELL_GFX_RING0;
-	adev->doorbell_index.sdma_engine0 = AMDGPU_DOORBELL_sDMA_ENGINE0;
-	adev->doorbell_index.sdma_engine1 = AMDGPU_DOORBELL_sDMA_ENGINE1;
+	adev->doorbell_index.sdma_engine[0] = AMDGPU_DOORBELL_sDMA_ENGINE0;
+	adev->doorbell_index.sdma_engine[1] = AMDGPU_DOORBELL_sDMA_ENGINE1;
 	adev->doorbell_index.ih = AMDGPU_DOORBELL_IH;
 	adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_MAX_ASSIGNMENT;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/Kconfig b/drivers/gpu/drm/amd/amdkfd/Kconfig
index fbf0ee5201c3..c3613604a4f8 100644
--- a/drivers/gpu/drm/amd/amdkfd/Kconfig
+++ b/drivers/gpu/drm/amd/amdkfd/Kconfig
@@ -4,8 +4,8 @@
 
 config HSA_AMD
 	bool "HSA kernel driver for AMD GPU devices"
-	depends on DRM_AMDGPU && X86_64
-	imply AMD_IOMMU_V2
+	depends on DRM_AMDGPU && (X86_64 || ARM64)
+	imply AMD_IOMMU_V2 if X86_64
 	select MMU_NOTIFIER
 	help
 	  Enable this if you want to use HSA features on AMD GPU devices.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index b7bc7d7d048f..2e7c44955f43 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -863,6 +863,7 @@ static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size,
 	return 0;
 }
 
+#ifdef CONFIG_X86_64
 static int kfd_fill_iolink_info_for_cpu(int numa_node_id, int *avail_size,
 				uint32_t *num_entries,
 				struct crat_subtype_iolink *sub_type_hdr)
@@ -905,6 +906,7 @@ static int kfd_fill_iolink_info_for_cpu(int numa_node_id, int *avail_size,
 
 	return 0;
 }
+#endif
 
 /* kfd_create_vcrat_image_cpu - Create Virtual CRAT for CPU
  *
@@ -920,7 +922,9 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 	struct crat_subtype_generic *sub_type_hdr;
 	int avail_size = *size;
 	int numa_node_id;
+#ifdef CONFIG_X86_64
 	uint32_t entries = 0;
+#endif
 	int ret = 0;
 
 	if (!pcrat_image || avail_size < VCRAT_SIZE_FOR_CPU)
@@ -982,6 +986,7 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 			sub_type_hdr->length);
 
 		/* Fill in Subtype: IO Link */
+#ifdef CONFIG_X86_64
 		ret = kfd_fill_iolink_info_for_cpu(numa_node_id, &avail_size,
 				&entries,
 				(struct crat_subtype_iolink *)sub_type_hdr);
@@ -992,6 +997,9 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 
 		sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
 				sub_type_hdr->length * entries);
+#else
+		pr_info("IO link not available for non x86 platforms\n");
+#endif
 
 		crat_table->num_domains++;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index 8018163414ff..932007eb9168 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -23,22 +23,7 @@
 #include <linux/sched.h>
 #include <linux/device.h>
 #include "kfd_priv.h"
-
-static const struct kgd2kfd_calls kgd2kfd = {
-	.exit		= kgd2kfd_exit,
-	.probe		= kgd2kfd_probe,
-	.device_init	= kgd2kfd_device_init,
-	.device_exit	= kgd2kfd_device_exit,
-	.interrupt	= kgd2kfd_interrupt,
-	.suspend	= kgd2kfd_suspend,
-	.resume		= kgd2kfd_resume,
-	.quiesce_mm	= kgd2kfd_quiesce_mm,
-	.resume_mm	= kgd2kfd_resume_mm,
-	.schedule_evict_and_restore_process =
-			  kgd2kfd_schedule_evict_and_restore_process,
-	.pre_reset	= kgd2kfd_pre_reset,
-	.post_reset	= kgd2kfd_post_reset,
-};
+#include "amdgpu_amdkfd.h"
 
 static int kfd_init(void)
 {
@@ -91,20 +76,10 @@ static void kfd_exit(void)
 	kfd_chardev_exit();
 }
 
-int kgd2kfd_init(unsigned int interface_version,
-		const struct kgd2kfd_calls **g2f)
+int kgd2kfd_init()
 {
-	int err;
-
-	err = kfd_init();
-	if (err)
-		return err;
-
-	*g2f = &kgd2kfd;
-
-	return 0;
+	return kfd_init();
 }
-EXPORT_SYMBOL(kgd2kfd_init);
 
 void kgd2kfd_exit(void)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 0689d4ccbbc0..12b66330fc6d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -266,14 +266,6 @@ struct kfd_dev {
 	bool pci_atomic_requested;
 };
 
-/* KGD2KFD callbacks */
-void kgd2kfd_exit(void);
-struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd,
-			struct pci_dev *pdev, const struct kfd2kgd_calls *f2g);
-bool kgd2kfd_device_init(struct kfd_dev *kfd,
-			const struct kgd2kfd_shared_resources *gpu_resources);
-void kgd2kfd_device_exit(struct kfd_dev *kfd);
-
 enum kfd_mempool {
 	KFD_MEMPOOL_SYSTEM_CACHEABLE = 1,
 	KFD_MEMPOOL_SYSTEM_WRITECOMBINE = 2,
@@ -541,11 +533,6 @@ struct qcm_process_device {
 /* Approx. time before evicting the process again */
 #define PROCESS_ACTIVE_TIME_MS 10
 
-int kgd2kfd_quiesce_mm(struct mm_struct *mm);
-int kgd2kfd_resume_mm(struct mm_struct *mm);
-int kgd2kfd_schedule_evict_and_restore_process(struct mm_struct *mm,
-					       struct dma_fence *fence);
-
 /* 8 byte handle containing GPU ID in the most significant 4 bytes and
  * idr_handle in the least significant 4 bytes
  */
@@ -800,20 +787,11 @@ int kfd_numa_node_to_apic_id(int numa_node_id);
 /* Interrupts */
 int kfd_interrupt_init(struct kfd_dev *dev);
 void kfd_interrupt_exit(struct kfd_dev *dev);
-void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
 bool enqueue_ih_ring_entry(struct kfd_dev *kfd,	const void *ih_ring_entry);
 bool interrupt_is_wanted(struct kfd_dev *dev,
 				const uint32_t *ih_ring_entry,
 				uint32_t *patched_ihre, bool *flag);
 
-/* Power Management */
-void kgd2kfd_suspend(struct kfd_dev *kfd);
-int kgd2kfd_resume(struct kfd_dev *kfd);
-
-/* GPU reset */
-int kgd2kfd_pre_reset(struct kfd_dev *kfd);
-int kgd2kfd_post_reset(struct kfd_dev *kfd);
-
 /* amdkfd Apertures */
 int kfd_init_apertures(struct kfd_process *process);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 5f5b2acedbac..09da91644f9f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1093,8 +1093,6 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
  *		the GPU device is not already present in the topology device
  *		list then return NULL. This means a new topology device has to
  *		be created for this GPU.
- * TODO: Rather than assiging @gpu to first topology device withtout
- *		gpu attached, it will better to have more stringent check.
  */
 static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 {
@@ -1102,12 +1100,20 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 	struct kfd_topology_device *out_dev = NULL;
 
 	down_write(&topology_lock);
-	list_for_each_entry(dev, &topology_device_list, list)
+	list_for_each_entry(dev, &topology_device_list, list) {
+		/* Discrete GPUs need their own topology device list
+		 * entries. Don't assign them to CPU/APU nodes.
+		 */
+		if (!gpu->device_info->needs_iommu_device &&
+		    dev->node_props.cpu_cores_count)
+			continue;
+
 		if (!dev->gpu && (dev->node_props.simd_count > 0)) {
 			dev->gpu = gpu;
 			out_dev = dev;
 			break;
 		}
+	}
 	up_write(&topology_lock);
 	return out_dev;
 }
@@ -1392,7 +1398,6 @@ int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev)
 
 static int kfd_cpumask_to_apic_id(const struct cpumask *cpumask)
 {
-	const struct cpuinfo_x86 *cpuinfo;
 	int first_cpu_of_numa_node;
 
 	if (!cpumask || cpumask == cpu_none_mask)
@@ -1400,9 +1405,11 @@ static int kfd_cpumask_to_apic_id(const struct cpumask *cpumask)
 	first_cpu_of_numa_node = cpumask_first(cpumask);
 	if (first_cpu_of_numa_node >= nr_cpu_ids)
 		return -1;
-	cpuinfo = &cpu_data(first_cpu_of_numa_node);
-
-	return cpuinfo->apicid;
+#ifdef CONFIG_X86_64
+	return cpu_data(first_cpu_of_numa_node).apicid;
+#else
+	return first_cpu_of_numa_node;
+#endif
 }
 
 /* kfd_numa_node_to_apic_id - Returns the APIC ID of the first logical processor
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 8a626d16e8e3..3a6f595f295e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -699,22 +699,36 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend)
 {
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_connector *connector;
+	struct drm_dp_mst_topology_mgr *mgr;
+	int ret;
+	bool need_hotplug = false;
 
 	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		   aconnector = to_amdgpu_dm_connector(connector);
-		   if (aconnector->dc_link->type == dc_connection_mst_branch &&
-				   !aconnector->mst_port) {
+	list_for_each_entry(connector, &dev->mode_config.connector_list,
+			    head) {
+		aconnector = to_amdgpu_dm_connector(connector);
+		if (aconnector->dc_link->type != dc_connection_mst_branch ||
+		    aconnector->mst_port)
+			continue;
 
-			   if (suspend)
-				   drm_dp_mst_topology_mgr_suspend(&aconnector->mst_mgr);
-			   else
-				   drm_dp_mst_topology_mgr_resume(&aconnector->mst_mgr);
-		   }
+		mgr = &aconnector->mst_mgr;
+
+		if (suspend) {
+			drm_dp_mst_topology_mgr_suspend(mgr);
+		} else {
+			ret = drm_dp_mst_topology_mgr_resume(mgr);
+			if (ret < 0) {
+				drm_dp_mst_topology_mgr_set_mst(mgr, false);
+				need_hotplug = true;
+			}
+		}
 	}
 
 	drm_modeset_unlock(&dev->mode_config.connection_mutex);
+
+	if (need_hotplug)
+		drm_kms_helper_hotplug_event(dev);
 }
 
 /**
@@ -898,7 +912,6 @@ static int dm_resume(void *handle)
 	struct drm_plane_state *new_plane_state;
 	struct dm_plane_state *dm_new_plane_state;
 	enum dc_connection_type new_connection_type = dc_connection_none;
-	int ret;
 	int i;
 
 	/* power on hardware */
@@ -971,13 +984,13 @@ static int dm_resume(void *handle)
 		}
 	}
 
-	ret = drm_atomic_helper_resume(ddev, dm->cached_state);
+	drm_atomic_helper_resume(ddev, dm->cached_state);
 
 	dm->cached_state = NULL;
 
 	amdgpu_dm_irq_resume_late(adev);
 
-	return ret;
+	return 0;
 }
 
 /**
@@ -1760,7 +1773,7 @@ static int amdgpu_dm_backlight_update_status(struct backlight_device *bd)
 		+ caps.min_input_signal * 0x101;
 
 	if (dc_link_set_backlight_level(dm->backlight_link,
-			brightness, 0, 0))
+			brightness, 0))
 		return 0;
 	else
 		return 1;
@@ -2284,6 +2297,71 @@ static int get_fb_info(const struct amdgpu_framebuffer *amdgpu_fb,
 	return r;
 }
 
+static inline uint64_t get_dcc_address(uint64_t address, uint64_t tiling_flags)
+{
+	uint32_t offset = AMDGPU_TILING_GET(tiling_flags, DCC_OFFSET_256B);
+
+	return offset ? (address + offset * 256) : 0;
+}
+
+static bool fill_plane_dcc_attributes(struct amdgpu_device *adev,
+				      const struct amdgpu_framebuffer *afb,
+				      struct dc_plane_state *plane_state,
+				      uint64_t info)
+{
+	struct dc *dc = adev->dm.dc;
+	struct dc_dcc_surface_param input;
+	struct dc_surface_dcc_cap output;
+	uint32_t offset = AMDGPU_TILING_GET(info, DCC_OFFSET_256B);
+	uint32_t i64b = AMDGPU_TILING_GET(info, DCC_INDEPENDENT_64B) != 0;
+	uint64_t dcc_address;
+
+	memset(&input, 0, sizeof(input));
+	memset(&output, 0, sizeof(output));
+
+	if (!offset)
+		return false;
+
+	if (!dc->cap_funcs.get_dcc_compression_cap)
+		return false;
+
+	input.format = plane_state->format;
+	input.surface_size.width =
+		plane_state->plane_size.grph.surface_size.width;
+	input.surface_size.height =
+		plane_state->plane_size.grph.surface_size.height;
+	input.swizzle_mode = plane_state->tiling_info.gfx9.swizzle;
+
+	if (plane_state->rotation == ROTATION_ANGLE_0 ||
+	    plane_state->rotation == ROTATION_ANGLE_180)
+		input.scan = SCAN_DIRECTION_HORIZONTAL;
+	else if (plane_state->rotation == ROTATION_ANGLE_90 ||
+		 plane_state->rotation == ROTATION_ANGLE_270)
+		input.scan = SCAN_DIRECTION_VERTICAL;
+
+	if (!dc->cap_funcs.get_dcc_compression_cap(dc, &input, &output))
+		return false;
+
+	if (!output.capable)
+		return false;
+
+	if (i64b == 0 && output.grph.rgb.independent_64b_blks != 0)
+		return false;
+
+	plane_state->dcc.enable = 1;
+	plane_state->dcc.grph.meta_pitch =
+		AMDGPU_TILING_GET(info, DCC_PITCH_MAX) + 1;
+	plane_state->dcc.grph.independent_64b_blks = i64b;
+
+	dcc_address = get_dcc_address(afb->address, info);
+	plane_state->address.grph.meta_addr.low_part =
+		lower_32_bits(dcc_address);
+	plane_state->address.grph.meta_addr.high_part =
+		upper_32_bits(dcc_address);
+
+	return true;
+}
+
 static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 					 struct dc_plane_state *plane_state,
 					 const struct amdgpu_framebuffer *amdgpu_fb)
@@ -2336,6 +2414,10 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 		return -EINVAL;
 	}
 
+	memset(&plane_state->address, 0, sizeof(plane_state->address));
+	memset(&plane_state->tiling_info, 0, sizeof(plane_state->tiling_info));
+	memset(&plane_state->dcc, 0, sizeof(plane_state->dcc));
+
 	if (plane_state->format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
 		plane_state->address.type = PLN_ADDR_TYPE_GRAPHICS;
 		plane_state->plane_size.grph.surface_size.x = 0;
@@ -2367,8 +2449,6 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 		plane_state->color_space = COLOR_SPACE_YCBCR709;
 	}
 
-	memset(&plane_state->tiling_info, 0, sizeof(plane_state->tiling_info));
-
 	/* Fill GFX8 params */
 	if (AMDGPU_TILING_GET(tiling_flags, ARRAY_MODE) == DC_ARRAY_2D_TILED_THIN1) {
 		unsigned int bankw, bankh, mtaspect, tile_split, num_banks;
@@ -2417,6 +2497,9 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 		plane_state->tiling_info.gfx9.swizzle =
 			AMDGPU_TILING_GET(tiling_flags, SWIZZLE_MODE);
 		plane_state->tiling_info.gfx9.shaderEnable = 1;
+
+		fill_plane_dcc_attributes(adev, amdgpu_fb, plane_state,
+					  tiling_flags);
 	}
 
 	plane_state->visible = true;
@@ -2580,7 +2663,7 @@ get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing)
 		 * according to HDMI spec, we use YCbCr709 and YCbCr601
 		 * respectively
 		 */
-		if (dc_crtc_timing->pix_clk_khz > 27030) {
+		if (dc_crtc_timing->pix_clk_100hz > 270300) {
 			if (dc_crtc_timing->flags.Y_ONLY)
 				color_space =
 					COLOR_SPACE_YCBCR709_LIMITED;
@@ -2623,7 +2706,7 @@ static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_
 	if (timing_out->display_color_depth <= COLOR_DEPTH_888)
 		return;
 	do {
-		normalized_clk = timing_out->pix_clk_khz;
+		normalized_clk = timing_out->pix_clk_100hz / 10;
 		/* YCbCr 4:2:0 requires additional adjustment of 1/2 */
 		if (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420)
 			normalized_clk /= 2;
@@ -2666,10 +2749,10 @@ fill_stream_properties_from_drm_display_mode(struct dc_stream_state *stream,
 	timing_out->v_border_bottom = 0;
 	/* TODO: un-hardcode */
 	if (drm_mode_is_420_only(info, mode_in)
-			&& stream->sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A)
+			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420;
 	else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCRCB444)
-			&& stream->sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A)
+			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR444;
 	else
 		timing_out->pixel_encoding = PIXEL_ENCODING_RGB;
@@ -2704,14 +2787,14 @@ fill_stream_properties_from_drm_display_mode(struct dc_stream_state *stream,
 		mode_in->crtc_vsync_start - mode_in->crtc_vdisplay;
 	timing_out->v_sync_width =
 		mode_in->crtc_vsync_end - mode_in->crtc_vsync_start;
-	timing_out->pix_clk_khz = mode_in->crtc_clock;
+	timing_out->pix_clk_100hz = mode_in->crtc_clock * 10;
 	timing_out->aspect_ratio = get_aspect_ratio(mode_in);
 
 	stream->output_color_space = get_output_color_space(timing_out);
 
 	stream->out_transfer_func->type = TF_TYPE_PREDEFINED;
 	stream->out_transfer_func->tf = TRANSFER_FUNCTION_SRGB;
-	if (stream->sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A)
+	if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
 		adjust_colour_depth_from_display_info(timing_out, info);
 }
 
@@ -2832,7 +2915,7 @@ static void set_master_stream(struct dc_stream_state *stream_set[],
 		if (stream_set[j] && stream_set[j]->triggered_crtc_reset.enabled) {
 			int refresh_rate = 0;
 
-			refresh_rate = (stream_set[j]->timing.pix_clk_khz*1000)/
+			refresh_rate = (stream_set[j]->timing.pix_clk_100hz*100)/
 				(stream_set[j]->timing.h_total*stream_set[j]->timing.v_total);
 			if (refresh_rate > highest_rfr) {
 				highest_rfr = refresh_rate;
@@ -2889,11 +2972,9 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 	drm_connector = &aconnector->base;
 
 	if (!aconnector->dc_sink) {
-		if (!aconnector->mst_port) {
-			sink = create_fake_sink(aconnector);
-			if (!sink)
-				return stream;
-		}
+		sink = create_fake_sink(aconnector);
+		if (!sink)
+			return stream;
 	} else {
 		sink = aconnector->dc_sink;
 	}
@@ -2905,6 +2986,8 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 		goto finish;
 	}
 
+	stream->dm_stream_context = aconnector;
+
 	list_for_each_entry(preferred_mode, &aconnector->base.modes, head) {
 		/* Search for preferred mode */
 		if (preferred_mode->type & DRM_MODE_TYPE_PREFERRED) {
@@ -2956,10 +3039,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 		drm_connector,
 		sink);
 
-	update_stream_signal(stream);
-
-	if (dm_state && dm_state->freesync_capable)
-		stream->ignore_msa_timing_param = true;
+	update_stream_signal(stream, sink);
 
 finish:
 	if (sink && sink->sink_signal == SIGNAL_TYPE_VIRTUAL && aconnector->base.force != DRM_FORCE_ON)
@@ -3532,6 +3612,7 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 	struct amdgpu_bo *rbo;
 	uint64_t chroma_addr = 0;
 	struct dm_plane_state *dm_plane_state_new, *dm_plane_state_old;
+	uint64_t tiling_flags, dcc_address;
 	unsigned int awidth;
 	uint32_t domain;
 	int r;
@@ -3572,6 +3653,9 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 		DRM_ERROR("%p bind failed\n", rbo);
 		return r;
 	}
+
+	amdgpu_bo_get_tiling_flags(rbo, &tiling_flags);
+
 	amdgpu_bo_unreserve(rbo);
 
 	afb->address = amdgpu_bo_gpu_offset(rbo);
@@ -3585,6 +3669,13 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 		if (plane_state->format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
 			plane_state->address.grph.addr.low_part = lower_32_bits(afb->address);
 			plane_state->address.grph.addr.high_part = upper_32_bits(afb->address);
+
+			dcc_address =
+				get_dcc_address(afb->address, tiling_flags);
+			plane_state->address.grph.meta_addr.low_part =
+				lower_32_bits(dcc_address);
+			plane_state->address.grph.meta_addr.high_part =
+				upper_32_bits(dcc_address);
 		} else {
 			awidth = ALIGN(new_state->fb->width, 64);
 			plane_state->address.type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE;
@@ -4070,7 +4161,8 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 	}
 
 	if (connector_type == DRM_MODE_CONNECTOR_HDMIA ||
-	    connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
+	    connector_type == DRM_MODE_CONNECTOR_DisplayPort ||
+	    connector_type == DRM_MODE_CONNECTOR_eDP) {
 		drm_connector_attach_vrr_capable_property(
 			&aconnector->base);
 	}
@@ -4456,20 +4548,6 @@ static void prepare_flip_isr(struct amdgpu_crtc *acrtc)
 						 acrtc->crtc_id);
 }
 
-struct dc_stream_status *dc_state_get_stream_status(
-	struct dc_state *state,
-	struct dc_stream_state *stream)
-{
-	uint8_t i;
-
-	for (i = 0; i < state->stream_count; i++) {
-		if (stream == state->streams[i])
-			return &state->stream_status[i];
-	}
-
-	return NULL;
-}
-
 static void update_freesync_state_on_stream(
 	struct amdgpu_display_manager *dm,
 	struct dm_crtc_state *new_crtc_state,
@@ -4523,12 +4601,12 @@ static void update_freesync_state_on_stream(
 		TRANSFER_FUNC_UNKNOWN,
 		&vrr_infopacket);
 
-	new_crtc_state->freesync_timing_changed =
+	new_crtc_state->freesync_timing_changed |=
 		(memcmp(&new_crtc_state->vrr_params.adjust,
 			&vrr_params.adjust,
 			sizeof(vrr_params.adjust)) != 0);
 
-	new_crtc_state->freesync_vrr_info_changed =
+	new_crtc_state->freesync_vrr_info_changed |=
 		(memcmp(&new_crtc_state->vrr_infopacket,
 			&vrr_infopacket,
 			sizeof(vrr_infopacket)) != 0);
@@ -4544,254 +4622,6 @@ static void update_freesync_state_on_stream(
 			      new_crtc_state->base.crtc->base.id,
 			      (int)new_crtc_state->base.vrr_enabled,
 			      (int)vrr_params.state);
-
-	if (new_crtc_state->freesync_timing_changed)
-		DRM_DEBUG_KMS("VRR timing update: crtc=%u min=%u max=%u\n",
-			      new_crtc_state->base.crtc->base.id,
-				  vrr_params.adjust.v_total_min,
-				  vrr_params.adjust.v_total_max);
-}
-
-/*
- * Executes flip
- *
- * Waits on all BO's fences and for proper vblank count
- */
-static void amdgpu_dm_do_flip(struct drm_crtc *crtc,
-			      struct drm_framebuffer *fb,
-			      uint32_t target,
-			      struct dc_state *state)
-{
-	unsigned long flags;
-	uint64_t timestamp_ns;
-	uint32_t target_vblank;
-	int r, vpos, hpos;
-	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_framebuffer *afb = to_amdgpu_framebuffer(fb);
-	struct amdgpu_bo *abo = gem_to_amdgpu_bo(fb->obj[0]);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-	bool async_flip = (crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0;
-	struct dc_flip_addrs addr = { {0} };
-	/* TODO eliminate or rename surface_update */
-	struct dc_surface_update surface_updates[1] = { {0} };
-	struct dc_stream_update stream_update = {0};
-	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
-	struct dc_stream_status *stream_status;
-	struct dc_plane_state *surface;
-
-
-	/* Prepare wait for target vblank early - before the fence-waits */
-	target_vblank = target - (uint32_t)drm_crtc_vblank_count(crtc) +
-			amdgpu_get_vblank_counter_kms(crtc->dev, acrtc->crtc_id);
-
-	/*
-	 * 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");
-		WARN_ON(1);
-	}
-
-	/* Wait for all fences on this FB */
-	WARN_ON(reservation_object_wait_timeout_rcu(abo->tbo.resv, true, false,
-								    MAX_SCHEDULE_TIMEOUT) < 0);
-
-	amdgpu_bo_unreserve(abo);
-
-	/*
-	 * Wait until we're out of the vertical blank period before the one
-	 * targeted by the flip
-	 */
-	while ((acrtc->enabled &&
-		(amdgpu_display_get_crtc_scanoutpos(adev->ddev, acrtc->crtc_id,
-						    0, &vpos, &hpos, NULL,
-						    NULL, &crtc->hwmode)
-		 & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) ==
-		(DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) &&
-		(int)(target_vblank -
-		  amdgpu_get_vblank_counter_kms(adev->ddev, acrtc->crtc_id)) > 0)) {
-		usleep_range(1000, 1100);
-	}
-
-	/* Flip */
-	spin_lock_irqsave(&crtc->dev->event_lock, flags);
-
-	WARN_ON(acrtc->pflip_status != AMDGPU_FLIP_NONE);
-	WARN_ON(!acrtc_state->stream);
-
-	addr.address.grph.addr.low_part = lower_32_bits(afb->address);
-	addr.address.grph.addr.high_part = upper_32_bits(afb->address);
-	addr.flip_immediate = async_flip;
-
-	timestamp_ns = ktime_get_ns();
-	addr.flip_timestamp_in_us = div_u64(timestamp_ns, 1000);
-
-
-	if (acrtc->base.state->event)
-		prepare_flip_isr(acrtc);
-
-	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
-
-	stream_status = dc_stream_get_status(acrtc_state->stream);
-	if (!stream_status) {
-		DRM_ERROR("No stream status for CRTC: id=%d\n",
-			acrtc->crtc_id);
-		return;
-	}
-
-	surface = stream_status->plane_states[0];
-	surface_updates->surface = surface;
-
-	if (!surface) {
-		DRM_ERROR("No surface for CRTC: id=%d\n",
-			acrtc->crtc_id);
-		return;
-	}
-	surface_updates->flip_addr = &addr;
-
-	if (acrtc_state->stream) {
-		update_freesync_state_on_stream(
-			&adev->dm,
-			acrtc_state,
-			acrtc_state->stream,
-			surface,
-			addr.flip_timestamp_in_us);
-
-		if (acrtc_state->freesync_timing_changed)
-			stream_update.adjust =
-				&acrtc_state->stream->adjust;
-
-		if (acrtc_state->freesync_vrr_info_changed)
-			stream_update.vrr_infopacket =
-				&acrtc_state->stream->vrr_infopacket;
-	}
-
-	/* Update surface timing information. */
-	surface->time.time_elapsed_in_us[surface->time.index] =
-		addr.flip_timestamp_in_us - surface->time.prev_update_time_in_us;
-	surface->time.prev_update_time_in_us = addr.flip_timestamp_in_us;
-	surface->time.index++;
-	if (surface->time.index >= DC_PLANE_UPDATE_TIMES_MAX)
-		surface->time.index = 0;
-
-	mutex_lock(&adev->dm.dc_lock);
-
-	dc_commit_updates_for_stream(adev->dm.dc,
-					     surface_updates,
-					     1,
-					     acrtc_state->stream,
-					     &stream_update,
-					     &surface_updates->surface,
-					     state);
-	mutex_unlock(&adev->dm.dc_lock);
-
-	DRM_DEBUG_DRIVER("%s Flipping to hi: 0x%x, low: 0x%x \n",
-			 __func__,
-			 addr.address.grph.addr.high_part,
-			 addr.address.grph.addr.low_part);
-}
-
-/*
- * TODO this whole function needs to go
- *
- * dc_surface_update is needlessly complex. See if we can just replace this
- * with a dc_plane_state and follow the atomic model a bit more closely here.
- */
-static bool commit_planes_to_stream(
-		struct amdgpu_display_manager *dm,
-		struct dc *dc,
-		struct dc_plane_state **plane_states,
-		uint8_t new_plane_count,
-		struct dm_crtc_state *dm_new_crtc_state,
-		struct dm_crtc_state *dm_old_crtc_state,
-		struct dc_state *state)
-{
-	/* no need to dynamically allocate this. it's pretty small */
-	struct dc_surface_update updates[MAX_SURFACES];
-	struct dc_flip_addrs *flip_addr;
-	struct dc_plane_info *plane_info;
-	struct dc_scaling_info *scaling_info;
-	int i;
-	struct dc_stream_state *dc_stream = dm_new_crtc_state->stream;
-	struct dc_stream_update *stream_update =
-			kzalloc(sizeof(struct dc_stream_update), GFP_KERNEL);
-	unsigned int abm_level;
-
-	if (!stream_update) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	flip_addr = kcalloc(MAX_SURFACES, sizeof(struct dc_flip_addrs),
-			    GFP_KERNEL);
-	plane_info = kcalloc(MAX_SURFACES, sizeof(struct dc_plane_info),
-			     GFP_KERNEL);
-	scaling_info = kcalloc(MAX_SURFACES, sizeof(struct dc_scaling_info),
-			       GFP_KERNEL);
-
-	if (!flip_addr || !plane_info || !scaling_info) {
-		kfree(flip_addr);
-		kfree(plane_info);
-		kfree(scaling_info);
-		kfree(stream_update);
-		return false;
-	}
-
-	memset(updates, 0, sizeof(updates));
-
-	stream_update->src = dc_stream->src;
-	stream_update->dst = dc_stream->dst;
-	stream_update->out_transfer_func = dc_stream->out_transfer_func;
-
-	if (dm_new_crtc_state->abm_level != dm_old_crtc_state->abm_level) {
-		abm_level = dm_new_crtc_state->abm_level;
-		stream_update->abm_level = &abm_level;
-	}
-
-	for (i = 0; i < new_plane_count; i++) {
-		updates[i].surface = plane_states[i];
-		updates[i].gamma =
-			(struct dc_gamma *)plane_states[i]->gamma_correction;
-		updates[i].in_transfer_func = plane_states[i]->in_transfer_func;
-		flip_addr[i].address = plane_states[i]->address;
-		flip_addr[i].flip_immediate = plane_states[i]->flip_immediate;
-		plane_info[i].color_space = plane_states[i]->color_space;
-		plane_info[i].format = plane_states[i]->format;
-		plane_info[i].plane_size = plane_states[i]->plane_size;
-		plane_info[i].rotation = plane_states[i]->rotation;
-		plane_info[i].horizontal_mirror = plane_states[i]->horizontal_mirror;
-		plane_info[i].stereo_format = plane_states[i]->stereo_format;
-		plane_info[i].tiling_info = plane_states[i]->tiling_info;
-		plane_info[i].visible = plane_states[i]->visible;
-		plane_info[i].per_pixel_alpha = plane_states[i]->per_pixel_alpha;
-		plane_info[i].dcc = plane_states[i]->dcc;
-		scaling_info[i].scaling_quality = plane_states[i]->scaling_quality;
-		scaling_info[i].src_rect = plane_states[i]->src_rect;
-		scaling_info[i].dst_rect = plane_states[i]->dst_rect;
-		scaling_info[i].clip_rect = plane_states[i]->clip_rect;
-
-		updates[i].flip_addr = &flip_addr[i];
-		updates[i].plane_info = &plane_info[i];
-		updates[i].scaling_info = &scaling_info[i];
-	}
-
-	mutex_lock(&dm->dc_lock);
-	dc_commit_updates_for_stream(
-			dc,
-			updates,
-			new_plane_count,
-			dc_stream, stream_update, plane_states, state);
-	mutex_unlock(&dm->dc_lock);
-
-	kfree(flip_addr);
-	kfree(plane_info);
-	kfree(scaling_info);
-	kfree(stream_update);
-	return true;
 }
 
 static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
@@ -4801,26 +4631,51 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 				    struct drm_crtc *pcrtc,
 				    bool *wait_for_vblank)
 {
-	uint32_t i;
+	uint32_t i, r;
+	uint64_t timestamp_ns;
 	struct drm_plane *plane;
 	struct drm_plane_state *old_plane_state, *new_plane_state;
-	struct dc_stream_state *dc_stream_attach;
-	struct dc_plane_state *plane_states_constructed[MAX_SURFACES];
 	struct amdgpu_crtc *acrtc_attach = to_amdgpu_crtc(pcrtc);
 	struct drm_crtc_state *new_pcrtc_state =
 			drm_atomic_get_new_crtc_state(state, pcrtc);
 	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(new_pcrtc_state);
 	struct dm_crtc_state *dm_old_crtc_state =
 			to_dm_crtc_state(drm_atomic_get_old_crtc_state(state, pcrtc));
-	int planes_count = 0;
+	int flip_count = 0, planes_count = 0, vpos, hpos;
 	unsigned long flags;
+	struct amdgpu_bo *abo;
+	uint64_t tiling_flags, dcc_address;
+	uint32_t target, target_vblank;
+
+	struct {
+		struct dc_surface_update surface_updates[MAX_SURFACES];
+		struct dc_flip_addrs flip_addrs[MAX_SURFACES];
+		struct dc_stream_update stream_update;
+	} *flip;
+
+	struct {
+		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_stream_update stream_update;
+	} *full;
+
+	flip = kzalloc(sizeof(*flip), GFP_KERNEL);
+	full = kzalloc(sizeof(*full), GFP_KERNEL);
+
+	if (!flip || !full) {
+		dm_error("Failed to allocate update bundles\n");
+		goto cleanup;
+	}
 
 	/* update planes when needed */
 	for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) {
 		struct drm_crtc *crtc = new_plane_state->crtc;
 		struct drm_crtc_state *new_crtc_state;
 		struct drm_framebuffer *fb = new_plane_state->fb;
+		struct amdgpu_framebuffer *afb = to_amdgpu_framebuffer(fb);
 		bool pflip_needed;
+		struct dc_plane_state *dc_plane;
 		struct dm_plane_state *dm_new_plane_state = to_dm_plane_state(new_plane_state);
 
 		if (plane->type == DRM_PLANE_TYPE_CURSOR) {
@@ -4835,73 +4690,193 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		if (!new_crtc_state->active)
 			continue;
 
-		pflip_needed = !state->allow_modeset;
+		pflip_needed = old_plane_state->fb &&
+			old_plane_state->fb != new_plane_state->fb;
 
-		spin_lock_irqsave(&crtc->dev->event_lock, flags);
-		if (acrtc_attach->pflip_status != AMDGPU_FLIP_NONE) {
-			DRM_ERROR("%s: acrtc %d, already busy\n",
-				  __func__,
-				  acrtc_attach->crtc_id);
-			/* In commit tail framework this cannot happen */
-			WARN_ON(1);
-		}
-		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
-
-		if (!pflip_needed || plane->type == DRM_PLANE_TYPE_OVERLAY) {
-			WARN_ON(!dm_new_plane_state->dc_state);
-
-			plane_states_constructed[planes_count] = dm_new_plane_state->dc_state;
+		dc_plane = dm_new_plane_state->dc_state;
 
-			dc_stream_attach = acrtc_state->stream;
-			planes_count++;
-
-		} else if (new_crtc_state->planes_changed) {
-			/* Assume even ONE crtc with immediate flip means
+		if (pflip_needed) {
+			/*
+			 * Assume even ONE crtc with immediate flip means
 			 * entire can't wait for VBLANK
 			 * TODO Check if it's correct
 			 */
-			*wait_for_vblank =
-					new_pcrtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC ?
-				false : true;
-
-			/* TODO: Needs rework for multiplane flip */
-			if (plane->type == DRM_PLANE_TYPE_PRIMARY)
-				drm_crtc_vblank_get(crtc);
-
-			amdgpu_dm_do_flip(
-				crtc,
-				fb,
-				(uint32_t)drm_crtc_vblank_count(crtc) + *wait_for_vblank,
-				dc_state);
+			if (new_pcrtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC)
+				*wait_for_vblank = false;
+
+			/*
+			 * 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
+			 */
+			abo = gem_to_amdgpu_bo(fb->obj[0]);
+			r = amdgpu_bo_reserve(abo, true);
+			if (unlikely(r != 0)) {
+				DRM_ERROR("failed to reserve buffer before flip\n");
+				WARN_ON(1);
+			}
+
+			/* Wait for all fences on this FB */
+			WARN_ON(reservation_object_wait_timeout_rcu(abo->tbo.resv, true, false,
+										    MAX_SCHEDULE_TIMEOUT) < 0);
+
+			amdgpu_bo_get_tiling_flags(abo, &tiling_flags);
+
+			amdgpu_bo_unreserve(abo);
+
+			flip->flip_addrs[flip_count].address.grph.addr.low_part = lower_32_bits(afb->address);
+			flip->flip_addrs[flip_count].address.grph.addr.high_part = upper_32_bits(afb->address);
+
+			dcc_address = get_dcc_address(afb->address, tiling_flags);
+			flip->flip_addrs[flip_count].address.grph.meta_addr.low_part = lower_32_bits(dcc_address);
+			flip->flip_addrs[flip_count].address.grph.meta_addr.high_part = upper_32_bits(dcc_address);
+
+			flip->flip_addrs[flip_count].flip_immediate =
+					(crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0;
+
+			timestamp_ns = ktime_get_ns();
+			flip->flip_addrs[flip_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000);
+			flip->surface_updates[flip_count].flip_addr = &flip->flip_addrs[flip_count];
+			flip->surface_updates[flip_count].surface = dc_plane;
+
+			if (!flip->surface_updates[flip_count].surface) {
+				DRM_ERROR("No surface for CRTC: id=%d\n",
+						acrtc_attach->crtc_id);
+				continue;
+			}
+
+			if (plane == pcrtc->primary)
+				update_freesync_state_on_stream(
+					dm,
+					acrtc_state,
+					acrtc_state->stream,
+					dc_plane,
+					flip->flip_addrs[flip_count].flip_timestamp_in_us);
+
+			DRM_DEBUG_DRIVER("%s Flipping to hi: 0x%x, low: 0x%x\n",
+					 __func__,
+					 flip->flip_addrs[flip_count].address.grph.addr.high_part,
+					 flip->flip_addrs[flip_count].address.grph.addr.low_part);
+
+			flip_count += 1;
 		}
 
+		full->surface_updates[planes_count].surface = dc_plane;
+		if (new_pcrtc_state->color_mgmt_changed) {
+			full->surface_updates[planes_count].gamma = dc_plane->gamma_correction;
+			full->surface_updates[planes_count].in_transfer_func = dc_plane->in_transfer_func;
+		}
+
+
+		full->scaling_infos[planes_count].scaling_quality = dc_plane->scaling_quality;
+		full->scaling_infos[planes_count].src_rect = dc_plane->src_rect;
+		full->scaling_infos[planes_count].dst_rect = dc_plane->dst_rect;
+		full->scaling_infos[planes_count].clip_rect = dc_plane->clip_rect;
+		full->surface_updates[planes_count].scaling_info = &full->scaling_infos[planes_count];
+
+
+		full->plane_infos[planes_count].color_space = dc_plane->color_space;
+		full->plane_infos[planes_count].format = dc_plane->format;
+		full->plane_infos[planes_count].plane_size = dc_plane->plane_size;
+		full->plane_infos[planes_count].rotation = dc_plane->rotation;
+		full->plane_infos[planes_count].horizontal_mirror = dc_plane->horizontal_mirror;
+		full->plane_infos[planes_count].stereo_format = dc_plane->stereo_format;
+		full->plane_infos[planes_count].tiling_info = dc_plane->tiling_info;
+		full->plane_infos[planes_count].visible = dc_plane->visible;
+		full->plane_infos[planes_count].per_pixel_alpha = dc_plane->per_pixel_alpha;
+		full->plane_infos[planes_count].dcc = dc_plane->dcc;
+		full->surface_updates[planes_count].plane_info = &full->plane_infos[planes_count];
+
+		planes_count += 1;
+
 	}
 
-	if (planes_count) {
-		unsigned long flags;
+	/*
+	 * TODO: For proper atomic behaviour, we should be calling into DC once with
+	 * all the changes.  However, DC refuses to do pageflips and non-pageflip
+	 * changes in the same call.  Change DC to respect atomic behaviour,
+	 * hopefully eliminating dc_*_update structs in their entirety.
+	 */
+	if (flip_count) {
+		target = (uint32_t)drm_crtc_vblank_count(pcrtc) + *wait_for_vblank;
+		/* Prepare wait for target vblank early - before the fence-waits */
+		target_vblank = target - (uint32_t)drm_crtc_vblank_count(pcrtc) +
+				amdgpu_get_vblank_counter_kms(pcrtc->dev, acrtc_attach->crtc_id);
 
-		if (new_pcrtc_state->event) {
+		/*
+		 * Wait until we're out of the vertical blank period before the one
+		 * targeted by the flip
+		 */
+		while ((acrtc_attach->enabled &&
+			(amdgpu_display_get_crtc_scanoutpos(dm->ddev, acrtc_attach->crtc_id,
+							    0, &vpos, &hpos, NULL,
+							    NULL, &pcrtc->hwmode)
+			 & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) ==
+			(DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) &&
+			(int)(target_vblank -
+			  amdgpu_get_vblank_counter_kms(dm->ddev, acrtc_attach->crtc_id)) > 0)) {
+			usleep_range(1000, 1100);
+		}
 
+		if (acrtc_attach->base.state->event) {
 			drm_crtc_vblank_get(pcrtc);
 
 			spin_lock_irqsave(&pcrtc->dev->event_lock, flags);
+
+			WARN_ON(acrtc_attach->pflip_status != AMDGPU_FLIP_NONE);
 			prepare_flip_isr(acrtc_attach);
+
 			spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags);
 		}
 
-		dc_stream_attach->abm_level = acrtc_state->abm_level;
+		if (acrtc_state->stream) {
 
-		if (false == commit_planes_to_stream(dm,
-							dm->dc,
-							plane_states_constructed,
-							planes_count,
-							acrtc_state,
-							dm_old_crtc_state,
-							dc_state))
-			dm_error("%s: Failed to attach plane!\n", __func__);
-	} else {
-		/*TODO BUG Here should go disable planes on CRTC. */
+			if (acrtc_state->freesync_timing_changed)
+				flip->stream_update.adjust =
+					&acrtc_state->stream->adjust;
+
+			if (acrtc_state->freesync_vrr_info_changed)
+				flip->stream_update.vrr_infopacket =
+					&acrtc_state->stream->vrr_infopacket;
+		}
+
+		mutex_lock(&dm->dc_lock);
+		dc_commit_updates_for_stream(dm->dc,
+						     flip->surface_updates,
+						     flip_count,
+						     acrtc_state->stream,
+						     &flip->stream_update,
+						     dc_state);
+		mutex_unlock(&dm->dc_lock);
+	}
+
+	if (planes_count) {
+		if (new_pcrtc_state->mode_changed) {
+			full->stream_update.src = acrtc_state->stream->src;
+			full->stream_update.dst = acrtc_state->stream->dst;
+		}
+
+		if (new_pcrtc_state->color_mgmt_changed)
+			full->stream_update.out_transfer_func = acrtc_state->stream->out_transfer_func;
+
+		acrtc_state->stream->abm_level = acrtc_state->abm_level;
+		if (acrtc_state->abm_level != dm_old_crtc_state->abm_level)
+			full->stream_update.abm_level = &acrtc_state->abm_level;
+
+		mutex_lock(&dm->dc_lock);
+		dc_commit_updates_for_stream(dm->dc,
+						     full->surface_updates,
+						     planes_count,
+						     acrtc_state->stream,
+						     &full->stream_update,
+						     dc_state);
+		mutex_unlock(&dm->dc_lock);
 	}
+
+cleanup:
+	kfree(flip);
+	kfree(full);
 }
 
 /*
@@ -4915,7 +4890,8 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 static void amdgpu_dm_crtc_copy_transient_flags(struct drm_crtc_state *crtc_state,
 						struct dc_stream_state *stream_state)
 {
-	stream_state->mode_changed = crtc_state->mode_changed;
+	stream_state->mode_changed =
+		crtc_state->mode_changed || crtc_state->active_changed;
 }
 
 static int amdgpu_dm_atomic_commit(struct drm_device *dev,
@@ -4936,10 +4912,25 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev,
 	 */
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+		struct dm_crtc_state *dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
 
-		if (drm_atomic_crtc_needs_modeset(new_crtc_state) && dm_old_crtc_state->stream)
+		if (drm_atomic_crtc_needs_modeset(new_crtc_state)
+		    && dm_old_crtc_state->stream) {
+			/*
+			 * If the stream is removed and CRC capture was
+			 * enabled on the CRTC the extra vblank reference
+			 * needs to be dropped since CRC capture will be
+			 * disabled.
+			 */
+			if (!dm_new_crtc_state->stream
+			    && dm_new_crtc_state->crc_enabled) {
+				drm_crtc_vblank_put(crtc);
+				dm_new_crtc_state->crc_enabled = false;
+			}
+
 			manage_dm_interrupts(adev, acrtc, false);
+		}
 	}
 	/*
 	 * Add check here for SoC's that support hardware cursor plane, to
@@ -5077,8 +5068,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 					dc_stream_get_status(dm_new_crtc_state->stream);
 
 			if (!status)
-				status = dc_state_get_stream_status(dc_state,
-								    dm_new_crtc_state->stream);
+				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);
@@ -5087,13 +5078,18 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		}
 	}
 
-	/* Handle scaling, underscan, and abm changes*/
+	/* Handle connector state changes */
 	for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) {
 		struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state);
 		struct dm_connector_state *dm_old_con_state = to_dm_connector_state(old_con_state);
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
+		struct dc_surface_update dummy_updates[MAX_SURFACES];
+		struct dc_stream_update stream_update;
 		struct dc_stream_status *status = NULL;
 
+		memset(&dummy_updates, 0, sizeof(dummy_updates));
+		memset(&stream_update, 0, sizeof(stream_update));
+
 		if (acrtc) {
 			new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base);
 			old_crtc_state = drm_atomic_get_old_crtc_state(state, &acrtc->base);
@@ -5103,37 +5099,48 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		if (!acrtc || drm_atomic_crtc_needs_modeset(new_crtc_state))
 			continue;
 
-
 		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
 
-		/* Skip anything that is not scaling or underscan changes */
 		if (!is_scaling_state_different(dm_new_con_state, dm_old_con_state) &&
 				(dm_new_crtc_state->abm_level == dm_old_crtc_state->abm_level))
 			continue;
 
-		update_stream_scaling_settings(&dm_new_con_state->base.crtc->mode,
-				dm_new_con_state, (struct dc_stream_state *)dm_new_crtc_state->stream);
+		if (is_scaling_state_different(dm_new_con_state, dm_old_con_state)) {
+			update_stream_scaling_settings(&dm_new_con_state->base.crtc->mode,
+					dm_new_con_state, (struct dc_stream_state *)dm_new_crtc_state->stream);
 
-		if (!dm_new_crtc_state->stream)
-			continue;
+			stream_update.src = dm_new_crtc_state->stream->src;
+			stream_update.dst = dm_new_crtc_state->stream->dst;
+		}
+
+		if (dm_new_crtc_state->abm_level != dm_old_crtc_state->abm_level) {
+			dm_new_crtc_state->stream->abm_level = dm_new_crtc_state->abm_level;
+
+			stream_update.abm_level = &dm_new_crtc_state->abm_level;
+		}
 
 		status = dc_stream_get_status(dm_new_crtc_state->stream);
 		WARN_ON(!status);
 		WARN_ON(!status->plane_count);
 
-		dm_new_crtc_state->stream->abm_level = dm_new_crtc_state->abm_level;
+		/*
+		 * TODO: DC refuses to perform stream updates without a dc_surface_update.
+		 * Here we create an empty update on each plane.
+		 * To fix this, DC should permit updating only stream properties.
+		 */
+		for (j = 0; j < status->plane_count; j++)
+			dummy_updates[j].surface = status->plane_states[0];
+
 
-		/*TODO How it works with MPO ?*/
-		if (!commit_planes_to_stream(
-				dm,
-				dm->dc,
-				status->plane_states,
-				status->plane_count,
-				dm_new_crtc_state,
-				to_dm_crtc_state(old_crtc_state),
-				dc_state))
-			dm_error("%s: Failed to update stream scaling!\n", __func__);
+		mutex_lock(&dm->dc_lock);
+		dc_commit_updates_for_stream(dm->dc,
+						     dummy_updates,
+						     status->plane_count,
+						     dm_new_crtc_state->stream,
+						     &stream_update,
+						     dc_state);
+		mutex_unlock(&dm->dc_lock);
 	}
 
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state,
@@ -5158,6 +5165,12 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 			continue;
 
 		manage_dm_interrupts(adev, acrtc, true);
+
+#ifdef CONFIG_DEBUG_FS
+		/* The stream has changed so CRC capture needs to re-enabled. */
+		if (dm_new_crtc_state->crc_enabled)
+			amdgpu_dm_crtc_set_crc_source(crtc, "auto");
+#endif
 	}
 
 	/* update planes when needed per crtc*/
@@ -5184,18 +5197,12 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 	}
 	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
 
+	/* Signal HW programming completion */
+	drm_atomic_helper_commit_hw_done(state);
 
 	if (wait_for_vblank)
 		drm_atomic_helper_wait_for_flip_done(dev, state);
 
-	/*
-	 * FIXME:
-	 * Delay hw_done() until flip_done() is signaled. This is to block
-	 * another commit from freeing the CRTC state while we're still
-	 * waiting on flip_done.
-	 */
-	drm_atomic_helper_commit_hw_done(state);
-
 	drm_atomic_helper_cleanup_planes(dev, state);
 
 	/*
@@ -5359,10 +5366,13 @@ static void get_freesync_config_for_crtc(
 	struct mod_freesync_config config = {0};
 	struct amdgpu_dm_connector *aconnector =
 			to_amdgpu_dm_connector(new_con_state->base.connector);
+	struct drm_display_mode *mode = &new_crtc_state->base.mode;
 
-	new_crtc_state->vrr_supported = new_con_state->freesync_capable;
+	new_crtc_state->vrr_supported = new_con_state->freesync_capable &&
+		aconnector->min_vfreq <= drm_mode_vrefresh(mode);
 
-	if (new_con_state->freesync_capable) {
+	if (new_crtc_state->vrr_supported) {
+		new_crtc_state->stream->ignore_msa_timing_param = true;
 		config.state = new_crtc_state->base.vrr_enabled ?
 				VRR_STATE_ACTIVE_VARIABLE :
 				VRR_STATE_INACTIVE;
@@ -5388,15 +5398,15 @@ static void reset_freesync_config_for_crtc(
 	       sizeof(new_crtc_state->vrr_infopacket));
 }
 
-static int dm_update_crtcs_state(struct amdgpu_display_manager *dm,
-				 struct drm_atomic_state *state,
-				 bool enable,
-				 bool *lock_and_validation_needed)
+static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
+				struct drm_atomic_state *state,
+				struct drm_crtc *crtc,
+				struct drm_crtc_state *old_crtc_state,
+				struct drm_crtc_state *new_crtc_state,
+				bool enable,
+				bool *lock_and_validation_needed)
 {
 	struct dm_atomic_state *dm_state = NULL;
-	struct drm_crtc *crtc;
-	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
-	int i;
 	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
 	struct dc_stream_state *new_stream;
 	int ret = 0;
@@ -5405,200 +5415,203 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm,
 	 * TODO Move this code into dm_crtc_atomic_check once we get rid of dc_validation_set
 	 * update changed items
 	 */
-	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
-		struct amdgpu_crtc *acrtc = NULL;
-		struct amdgpu_dm_connector *aconnector = NULL;
-		struct drm_connector_state *drm_new_conn_state = NULL, *drm_old_conn_state = NULL;
-		struct dm_connector_state *dm_new_conn_state = NULL, *dm_old_conn_state = NULL;
-		struct drm_plane_state *new_plane_state = NULL;
+	struct amdgpu_crtc *acrtc = NULL;
+	struct amdgpu_dm_connector *aconnector = NULL;
+	struct drm_connector_state *drm_new_conn_state = NULL, *drm_old_conn_state = NULL;
+	struct dm_connector_state *dm_new_conn_state = NULL, *dm_old_conn_state = NULL;
+	struct drm_plane_state *new_plane_state = NULL;
 
-		new_stream = NULL;
+	new_stream = NULL;
 
-		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
-		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
-		acrtc = to_amdgpu_crtc(crtc);
+	dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+	dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
+	acrtc = to_amdgpu_crtc(crtc);
 
-		new_plane_state = drm_atomic_get_new_plane_state(state, new_crtc_state->crtc->primary);
+	new_plane_state = drm_atomic_get_new_plane_state(state, new_crtc_state->crtc->primary);
 
-		if (new_crtc_state->enable && new_plane_state && !new_plane_state->fb) {
-			ret = -EINVAL;
-			goto fail;
-		}
+	if (new_crtc_state->enable && new_plane_state && !new_plane_state->fb) {
+		ret = -EINVAL;
+		goto fail;
+	}
 
-		aconnector = amdgpu_dm_find_first_crtc_matching_connector(state, crtc);
+	aconnector = amdgpu_dm_find_first_crtc_matching_connector(state, crtc);
 
-		/* TODO This hack should go away */
-		if (aconnector && enable) {
-			/* Make sure fake sink is created in plug-in scenario */
-			drm_new_conn_state = drm_atomic_get_new_connector_state(state,
- 								    &aconnector->base);
-			drm_old_conn_state = drm_atomic_get_old_connector_state(state,
-								    &aconnector->base);
+	/* TODO This hack should go away */
+	if (aconnector && enable) {
+		/* Make sure fake sink is created in plug-in scenario */
+		drm_new_conn_state = drm_atomic_get_new_connector_state(state,
+							    &aconnector->base);
+		drm_old_conn_state = drm_atomic_get_old_connector_state(state,
+							    &aconnector->base);
 
-			if (IS_ERR(drm_new_conn_state)) {
-				ret = PTR_ERR_OR_ZERO(drm_new_conn_state);
-				break;
-			}
+		if (IS_ERR(drm_new_conn_state)) {
+			ret = PTR_ERR_OR_ZERO(drm_new_conn_state);
+			goto fail;
+		}
 
-			dm_new_conn_state = to_dm_connector_state(drm_new_conn_state);
-			dm_old_conn_state = to_dm_connector_state(drm_old_conn_state);
+		dm_new_conn_state = to_dm_connector_state(drm_new_conn_state);
+		dm_old_conn_state = to_dm_connector_state(drm_old_conn_state);
 
-			new_stream = create_stream_for_sink(aconnector,
-							     &new_crtc_state->mode,
-							    dm_new_conn_state,
-							    dm_old_crtc_state->stream);
+		if (!drm_atomic_crtc_needs_modeset(new_crtc_state))
+			goto skip_modeset;
 
-			/*
-			 * we can have no stream on ACTION_SET if a display
-			 * was disconnected during S3, in this case it is not an
-			 * error, the OS will be updated after detection, and
-			 * will do the right thing on next atomic commit
-			 */
+		new_stream = create_stream_for_sink(aconnector,
+						     &new_crtc_state->mode,
+						    dm_new_conn_state,
+						    dm_old_crtc_state->stream);
 
-			if (!new_stream) {
-				DRM_DEBUG_DRIVER("%s: Failed to create new stream for crtc %d\n",
-						__func__, acrtc->base.base.id);
-				break;
-			}
-
-			dm_new_crtc_state->abm_level = dm_new_conn_state->abm_level;
+		/*
+		 * we can have no stream on ACTION_SET if a display
+		 * was disconnected during S3, in this case it is not an
+		 * error, the OS will be updated after detection, and
+		 * will do the right thing on next atomic commit
+		 */
 
-			if (dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) &&
-			    dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream)) {
-				new_crtc_state->mode_changed = false;
-				DRM_DEBUG_DRIVER("Mode change not required, setting mode_changed to %d",
-						 new_crtc_state->mode_changed);
-			}
+		if (!new_stream) {
+			DRM_DEBUG_DRIVER("%s: Failed to create new stream for crtc %d\n",
+					__func__, acrtc->base.base.id);
+			ret = -ENOMEM;
+			goto fail;
 		}
 
-		if (!drm_atomic_crtc_needs_modeset(new_crtc_state))
-			goto next_crtc;
+		dm_new_crtc_state->abm_level = dm_new_conn_state->abm_level;
 
-		DRM_DEBUG_DRIVER(
-			"amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, "
-			"planes_changed:%d, mode_changed:%d,active_changed:%d,"
-			"connectors_changed:%d\n",
-			acrtc->crtc_id,
-			new_crtc_state->enable,
-			new_crtc_state->active,
-			new_crtc_state->planes_changed,
-			new_crtc_state->mode_changed,
-			new_crtc_state->active_changed,
-			new_crtc_state->connectors_changed);
+		if (dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) &&
+		    dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream)) {
+			new_crtc_state->mode_changed = false;
+			DRM_DEBUG_DRIVER("Mode change not required, setting mode_changed to %d",
+					 new_crtc_state->mode_changed);
+		}
+	}
 
-		/* Remove stream for any changed/disabled CRTC */
-		if (!enable) {
+	/* mode_changed flag may get updated above, need to check again */
+	if (!drm_atomic_crtc_needs_modeset(new_crtc_state))
+		goto skip_modeset;
 
-			if (!dm_old_crtc_state->stream)
-				goto next_crtc;
+	DRM_DEBUG_DRIVER(
+		"amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, "
+		"planes_changed:%d, mode_changed:%d,active_changed:%d,"
+		"connectors_changed:%d\n",
+		acrtc->crtc_id,
+		new_crtc_state->enable,
+		new_crtc_state->active,
+		new_crtc_state->planes_changed,
+		new_crtc_state->mode_changed,
+		new_crtc_state->active_changed,
+		new_crtc_state->connectors_changed);
 
-			ret = dm_atomic_get_state(state, &dm_state);
-			if (ret)
-				goto fail;
+	/* Remove stream for any changed/disabled CRTC */
+	if (!enable) {
 
-			DRM_DEBUG_DRIVER("Disabling DRM crtc: %d\n",
-					crtc->base.id);
+		if (!dm_old_crtc_state->stream)
+			goto skip_modeset;
 
-			/* i.e. reset mode */
-			if (dc_remove_stream_from_ctx(
-					dm->dc,
-					dm_state->context,
-					dm_old_crtc_state->stream) != DC_OK) {
-				ret = -EINVAL;
-				goto fail;
-			}
+		ret = dm_atomic_get_state(state, &dm_state);
+		if (ret)
+			goto fail;
 
-			dc_stream_release(dm_old_crtc_state->stream);
-			dm_new_crtc_state->stream = NULL;
+		DRM_DEBUG_DRIVER("Disabling DRM crtc: %d\n",
+				crtc->base.id);
 
-			reset_freesync_config_for_crtc(dm_new_crtc_state);
+		/* i.e. reset mode */
+		if (dc_remove_stream_from_ctx(
+				dm->dc,
+				dm_state->context,
+				dm_old_crtc_state->stream) != DC_OK) {
+			ret = -EINVAL;
+			goto fail;
+		}
 
-			*lock_and_validation_needed = true;
+		dc_stream_release(dm_old_crtc_state->stream);
+		dm_new_crtc_state->stream = NULL;
 
-		} else {/* Add stream for any updated/enabled CRTC */
-			/*
-			 * Quick fix to prevent NULL pointer on new_stream when
-			 * added MST connectors not found in existing crtc_state in the chained mode
-			 * TODO: need to dig out the root cause of that
-			 */
-			if (!aconnector || (!aconnector->dc_sink && aconnector->mst_port))
-				goto next_crtc;
+		reset_freesync_config_for_crtc(dm_new_crtc_state);
+
+		*lock_and_validation_needed = true;
 
-			if (modereset_required(new_crtc_state))
-				goto next_crtc;
+	} else {/* Add stream for any updated/enabled CRTC */
+		/*
+		 * Quick fix to prevent NULL pointer on new_stream when
+		 * added MST connectors not found in existing crtc_state in the chained mode
+		 * TODO: need to dig out the root cause of that
+		 */
+		if (!aconnector || (!aconnector->dc_sink && aconnector->mst_port))
+			goto skip_modeset;
 
-			if (modeset_required(new_crtc_state, new_stream,
-					     dm_old_crtc_state->stream)) {
+		if (modereset_required(new_crtc_state))
+			goto skip_modeset;
 
-				WARN_ON(dm_new_crtc_state->stream);
+		if (modeset_required(new_crtc_state, new_stream,
+				     dm_old_crtc_state->stream)) {
 
-				ret = dm_atomic_get_state(state, &dm_state);
-				if (ret)
-					goto fail;
+			WARN_ON(dm_new_crtc_state->stream);
 
-				dm_new_crtc_state->stream = new_stream;
+			ret = dm_atomic_get_state(state, &dm_state);
+			if (ret)
+				goto fail;
 
-				dc_stream_retain(new_stream);
+			dm_new_crtc_state->stream = new_stream;
 
-				DRM_DEBUG_DRIVER("Enabling DRM crtc: %d\n",
-							crtc->base.id);
+			dc_stream_retain(new_stream);
 
-				if (dc_add_stream_to_ctx(
-						dm->dc,
-						dm_state->context,
-						dm_new_crtc_state->stream) != DC_OK) {
-					ret = -EINVAL;
-					goto fail;
-				}
+			DRM_DEBUG_DRIVER("Enabling DRM crtc: %d\n",
+						crtc->base.id);
 
-				*lock_and_validation_needed = true;
+			if (dc_add_stream_to_ctx(
+					dm->dc,
+					dm_state->context,
+					dm_new_crtc_state->stream) != DC_OK) {
+				ret = -EINVAL;
+				goto fail;
 			}
-		}
 
-next_crtc:
-		/* Release extra reference */
-		if (new_stream)
-			 dc_stream_release(new_stream);
+			*lock_and_validation_needed = true;
+		}
+	}
 
-		/*
-		 * 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))
-			continue;
-		/*
-		 * Given above conditions, the dc state cannot be NULL because:
-		 * 1. We're in the process of enabling CRTCs (just been added
-		 *    to the dc context, or already is on the context)
-		 * 2. Has a valid connector attached, and
-		 * 3. Is currently active and enabled.
-		 * => The dc stream state currently exists.
-		 */
-		BUG_ON(dm_new_crtc_state->stream == NULL);
+skip_modeset:
+	/* Release extra reference */
+	if (new_stream)
+		 dc_stream_release(new_stream);
 
-		/* Scaling or underscan settings */
-		if (is_scaling_state_different(dm_old_conn_state, dm_new_conn_state))
-			update_stream_scaling_settings(
-				&new_crtc_state->mode, dm_new_conn_state, dm_new_crtc_state->stream);
+	/*
+	 * 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))
+		return 0;
+	/*
+	 * Given above conditions, the dc state cannot be NULL because:
+	 * 1. We're in the process of enabling CRTCs (just been added
+	 *    to the dc context, or already is on the context)
+	 * 2. Has a valid connector attached, and
+	 * 3. Is currently active and enabled.
+	 * => The dc stream state currently exists.
+	 */
+	BUG_ON(dm_new_crtc_state->stream == NULL);
 
-		/*
-		 * Color management settings. We also update color properties
-		 * when a modeset is needed, to ensure it gets reprogrammed.
-		 */
-		if (dm_new_crtc_state->base.color_mgmt_changed ||
-		    drm_atomic_crtc_needs_modeset(new_crtc_state)) {
-			ret = amdgpu_dm_set_regamma_lut(dm_new_crtc_state);
-			if (ret)
-				goto fail;
-			amdgpu_dm_set_ctm(dm_new_crtc_state);
-		}
+	/* Scaling or underscan settings */
+	if (is_scaling_state_different(dm_old_conn_state, dm_new_conn_state))
+		update_stream_scaling_settings(
+			&new_crtc_state->mode, dm_new_conn_state, dm_new_crtc_state->stream);
 
-		/* Update Freesync settings. */
-		get_freesync_config_for_crtc(dm_new_crtc_state,
-					     dm_new_conn_state);
+	/*
+	 * Color management settings. We also update color properties
+	 * when a modeset is needed, to ensure it gets reprogrammed.
+	 */
+	if (dm_new_crtc_state->base.color_mgmt_changed ||
+	    drm_atomic_crtc_needs_modeset(new_crtc_state)) {
+		ret = amdgpu_dm_set_regamma_lut(dm_new_crtc_state);
+		if (ret)
+			goto fail;
+		amdgpu_dm_set_ctm(dm_new_crtc_state);
 	}
 
+	/* Update Freesync settings. */
+	get_freesync_config_for_crtc(dm_new_crtc_state,
+				     dm_new_conn_state);
+
 	return ret;
 
 fail:
@@ -5607,145 +5620,141 @@ fail:
 	return ret;
 }
 
-static int dm_update_planes_state(struct dc *dc,
-				  struct drm_atomic_state *state,
-				  bool enable,
-				  bool *lock_and_validation_needed)
+static int dm_update_plane_state(struct dc *dc,
+				 struct drm_atomic_state *state,
+				 struct drm_plane *plane,
+				 struct drm_plane_state *old_plane_state,
+				 struct drm_plane_state *new_plane_state,
+				 bool enable,
+				 bool *lock_and_validation_needed)
 {
 
 	struct dm_atomic_state *dm_state = NULL;
 	struct drm_crtc *new_plane_crtc, *old_plane_crtc;
 	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
-	struct drm_plane *plane;
-	struct drm_plane_state *old_plane_state, *new_plane_state;
 	struct dm_crtc_state *dm_new_crtc_state, *dm_old_crtc_state;
 	struct dm_plane_state *dm_new_plane_state, *dm_old_plane_state;
-	int i ;
 	/* TODO return page_flip_needed() function */
 	bool pflip_needed  = !state->allow_modeset;
 	int ret = 0;
 
 
-	/* Add new planes, in reverse order as DC expectation */
-	for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) {
-		new_plane_crtc = new_plane_state->crtc;
-		old_plane_crtc = old_plane_state->crtc;
-		dm_new_plane_state = to_dm_plane_state(new_plane_state);
-		dm_old_plane_state = to_dm_plane_state(old_plane_state);
+	new_plane_crtc = new_plane_state->crtc;
+	old_plane_crtc = old_plane_state->crtc;
+	dm_new_plane_state = to_dm_plane_state(new_plane_state);
+	dm_old_plane_state = to_dm_plane_state(old_plane_state);
 
-		/*TODO Implement atomic check for cursor plane */
-		if (plane->type == DRM_PLANE_TYPE_CURSOR)
-			continue;
+	/*TODO Implement atomic check for cursor plane */
+	if (plane->type == DRM_PLANE_TYPE_CURSOR)
+		return 0;
 
-		/* Remove any changed/removed planes */
-		if (!enable) {
-			if (pflip_needed &&
-			    plane->type != DRM_PLANE_TYPE_OVERLAY)
-				continue;
+	/* Remove any changed/removed planes */
+	if (!enable) {
+		if (pflip_needed &&
+		    plane->type != DRM_PLANE_TYPE_OVERLAY)
+			return 0;
 
-			if (!old_plane_crtc)
-				continue;
+		if (!old_plane_crtc)
+			return 0;
 
-			old_crtc_state = drm_atomic_get_old_crtc_state(
-					state, old_plane_crtc);
-			dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+		old_crtc_state = drm_atomic_get_old_crtc_state(
+				state, old_plane_crtc);
+		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
 
-			if (!dm_old_crtc_state->stream)
-				continue;
+		if (!dm_old_crtc_state->stream)
+			return 0;
 
-			DRM_DEBUG_ATOMIC("Disabling DRM plane: %d on DRM crtc %d\n",
-					plane->base.id, old_plane_crtc->base.id);
+		DRM_DEBUG_ATOMIC("Disabling DRM plane: %d on DRM crtc %d\n",
+				plane->base.id, old_plane_crtc->base.id);
 
-			ret = dm_atomic_get_state(state, &dm_state);
-			if (ret)
-				return ret;
+		ret = dm_atomic_get_state(state, &dm_state);
+		if (ret)
+			return ret;
 
-			if (!dc_remove_plane_from_context(
-					dc,
-					dm_old_crtc_state->stream,
-					dm_old_plane_state->dc_state,
-					dm_state->context)) {
+		if (!dc_remove_plane_from_context(
+				dc,
+				dm_old_crtc_state->stream,
+				dm_old_plane_state->dc_state,
+				dm_state->context)) {
 
-				ret = EINVAL;
-				return ret;
-			}
+			ret = EINVAL;
+			return ret;
+		}
 
 
-			dc_plane_state_release(dm_old_plane_state->dc_state);
-			dm_new_plane_state->dc_state = NULL;
+		dc_plane_state_release(dm_old_plane_state->dc_state);
+		dm_new_plane_state->dc_state = NULL;
 
-			*lock_and_validation_needed = true;
+		*lock_and_validation_needed = true;
 
-		} else { /* Add new planes */
-			struct dc_plane_state *dc_new_plane_state;
+	} else { /* Add new planes */
+		struct dc_plane_state *dc_new_plane_state;
 
-			if (drm_atomic_plane_disabling(plane->state, new_plane_state))
-				continue;
+		if (drm_atomic_plane_disabling(plane->state, new_plane_state))
+			return 0;
 
-			if (!new_plane_crtc)
-				continue;
+		if (!new_plane_crtc)
+			return 0;
 
-			new_crtc_state = drm_atomic_get_new_crtc_state(state, new_plane_crtc);
-			dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
+		new_crtc_state = drm_atomic_get_new_crtc_state(state, new_plane_crtc);
+		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 
-			if (!dm_new_crtc_state->stream)
-				continue;
+		if (!dm_new_crtc_state->stream)
+			return 0;
 
-			if (pflip_needed &&
-			    plane->type != DRM_PLANE_TYPE_OVERLAY)
-				continue;
+		if (pflip_needed && plane->type != DRM_PLANE_TYPE_OVERLAY)
+			return 0;
 
-			WARN_ON(dm_new_plane_state->dc_state);
+		WARN_ON(dm_new_plane_state->dc_state);
 
-			dc_new_plane_state = dc_create_plane_state(dc);
-			if (!dc_new_plane_state)
-				return -ENOMEM;
+		dc_new_plane_state = dc_create_plane_state(dc);
+		if (!dc_new_plane_state)
+			return -ENOMEM;
 
-			DRM_DEBUG_DRIVER("Enabling DRM plane: %d on DRM crtc %d\n",
-					plane->base.id, new_plane_crtc->base.id);
+		DRM_DEBUG_DRIVER("Enabling DRM plane: %d on DRM crtc %d\n",
+				plane->base.id, new_plane_crtc->base.id);
 
-			ret = fill_plane_attributes(
-				new_plane_crtc->dev->dev_private,
-				dc_new_plane_state,
-				new_plane_state,
-				new_crtc_state);
-			if (ret) {
-				dc_plane_state_release(dc_new_plane_state);
-				return ret;
-			}
+		ret = fill_plane_attributes(
+			new_plane_crtc->dev->dev_private,
+			dc_new_plane_state,
+			new_plane_state,
+			new_crtc_state);
+		if (ret) {
+			dc_plane_state_release(dc_new_plane_state);
+			return ret;
+		}
 
-			ret = dm_atomic_get_state(state, &dm_state);
-			if (ret) {
-				dc_plane_state_release(dc_new_plane_state);
-				return ret;
-			}
+		ret = dm_atomic_get_state(state, &dm_state);
+		if (ret) {
+			dc_plane_state_release(dc_new_plane_state);
+			return ret;
+		}
 
-			/*
-			 * Any atomic check errors that occur after this will
-			 * not need a release. The plane state will be attached
-			 * to the stream, and therefore part of the atomic
-			 * state. It'll be released when the atomic state is
-			 * cleaned.
-			 */
-			if (!dc_add_plane_to_context(
-					dc,
-					dm_new_crtc_state->stream,
-					dc_new_plane_state,
-					dm_state->context)) {
-
-				dc_plane_state_release(dc_new_plane_state);
-				return -EINVAL;
-			}
+		/*
+		 * Any atomic check errors that occur after this will
+		 * not need a release. The plane state will be attached
+		 * to the stream, and therefore part of the atomic
+		 * state. It'll be released when the atomic state is
+		 * cleaned.
+		 */
+		if (!dc_add_plane_to_context(
+				dc,
+				dm_new_crtc_state->stream,
+				dc_new_plane_state,
+				dm_state->context)) {
 
-			dm_new_plane_state->dc_state = dc_new_plane_state;
+			dc_plane_state_release(dc_new_plane_state);
+			return -EINVAL;
+		}
 
-			/* Tell DC to do a full surface update every time there
-			 * is a plane change. Inefficient, but works for now.
-			 */
-			dm_new_plane_state->dc_state->update_flags.bits.full_update = 1;
+		dm_new_plane_state->dc_state = dc_new_plane_state;
 
-			*lock_and_validation_needed = true;
-		}
+		/* Tell DC to do a full surface update every time there
+		 * is a plane change. Inefficient, but works for now.
+		 */
+		dm_new_plane_state->dc_state->update_flags.bits.full_update = 1;
+
+		*lock_and_validation_needed = true;
 	}
 
 
@@ -5769,11 +5778,13 @@ dm_determine_update_type_for_commit(struct dc *dc,
 	struct dm_crtc_state *new_dm_crtc_state, *old_dm_crtc_state;
 	struct dc_stream_status *status = NULL;
 
-	struct dc_surface_update *updates = kzalloc(MAX_SURFACES * sizeof(struct dc_surface_update), GFP_KERNEL);
-	struct dc_plane_state *surface = kzalloc(MAX_SURFACES * sizeof(struct dc_plane_state), GFP_KERNEL);
-	struct dc_stream_update stream_update;
+	struct dc_surface_update *updates;
+	struct dc_plane_state *surface;
 	enum surface_update_type update_type = UPDATE_TYPE_FAST;
 
+	updates = kcalloc(MAX_SURFACES, sizeof(*updates), GFP_KERNEL);
+	surface = kcalloc(MAX_SURFACES, sizeof(*surface), GFP_KERNEL);
+
 	if (!updates || !surface) {
 		DRM_ERROR("Plane or surface update failed to allocate");
 		/* Set type to FULL to avoid crashing in DC*/
@@ -5782,79 +5793,85 @@ dm_determine_update_type_for_commit(struct dc *dc,
 	}
 
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		struct dc_stream_update stream_update = { 0 };
+
 		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) {
-
-			for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, j) {
-				new_plane_crtc = new_plane_state->crtc;
-				old_plane_crtc = old_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 (!state->allow_modeset)
-					continue;
-
-				if (crtc == new_plane_crtc) {
-					updates[num_plane].surface = &surface[num_plane];
-
-					if (new_crtc_state->mode_changed) {
-						updates[num_plane].surface->src_rect =
-									new_dm_plane_state->dc_state->src_rect;
-						updates[num_plane].surface->dst_rect =
-									new_dm_plane_state->dc_state->dst_rect;
-						updates[num_plane].surface->rotation =
-									new_dm_plane_state->dc_state->rotation;
-						updates[num_plane].surface->in_transfer_func =
-									new_dm_plane_state->dc_state->in_transfer_func;
-						stream_update.dst = new_dm_crtc_state->stream->dst;
-						stream_update.src = new_dm_crtc_state->stream->src;
-					}
-
-					if (new_crtc_state->color_mgmt_changed) {
-						updates[num_plane].gamma =
-								new_dm_plane_state->dc_state->gamma_correction;
-						updates[num_plane].in_transfer_func =
-								new_dm_plane_state->dc_state->in_transfer_func;
-						stream_update.gamut_remap =
-								&new_dm_crtc_state->stream->gamut_remap_matrix;
-						stream_update.out_transfer_func =
-								new_dm_crtc_state->stream->out_transfer_func;
-					}
-
-					num_plane++;
-				}
+		if (!new_dm_crtc_state->stream) {
+			if (!new_dm_crtc_state->stream && old_dm_crtc_state->stream) {
+				update_type = UPDATE_TYPE_FULL;
+				goto cleanup;
 			}
 
-			if (num_plane > 0) {
-				ret = dm_atomic_get_state(state, &dm_state);
-				if (ret)
-					goto cleanup;
+			continue;
+		}
 
-				old_dm_state = dm_atomic_get_old_state(state);
-				if (!old_dm_state) {
-					ret = -EINVAL;
-					goto cleanup;
-				}
+		for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, j) {
+			new_plane_crtc = new_plane_state->crtc;
+			old_plane_crtc = old_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);
 
-				status = dc_state_get_stream_status(old_dm_state->context,
-								    new_dm_crtc_state->stream);
+			if (plane->type == DRM_PLANE_TYPE_CURSOR)
+				continue;
 
-				update_type = dc_check_update_surfaces_for_stream(dc, updates, num_plane,
-										  &stream_update, status);
+			if (!state->allow_modeset)
+				continue;
 
-				if (update_type > UPDATE_TYPE_MED) {
-					update_type = UPDATE_TYPE_FULL;
-					goto cleanup;
-				}
+			if (crtc != new_plane_crtc)
+				continue;
+
+			updates[num_plane].surface = &surface[num_plane];
+
+			if (new_crtc_state->mode_changed) {
+				updates[num_plane].surface->src_rect =
+						new_dm_plane_state->dc_state->src_rect;
+				updates[num_plane].surface->dst_rect =
+						new_dm_plane_state->dc_state->dst_rect;
+				updates[num_plane].surface->rotation =
+						new_dm_plane_state->dc_state->rotation;
+				updates[num_plane].surface->in_transfer_func =
+						new_dm_plane_state->dc_state->in_transfer_func;
+				stream_update.dst = new_dm_crtc_state->stream->dst;
+				stream_update.src = new_dm_crtc_state->stream->src;
+			}
+
+			if (new_crtc_state->color_mgmt_changed) {
+				updates[num_plane].gamma =
+						new_dm_plane_state->dc_state->gamma_correction;
+				updates[num_plane].in_transfer_func =
+						new_dm_plane_state->dc_state->in_transfer_func;
+				stream_update.gamut_remap =
+						&new_dm_crtc_state->stream->gamut_remap_matrix;
+				stream_update.out_transfer_func =
+						new_dm_crtc_state->stream->out_transfer_func;
 			}
 
-		} else if (!new_dm_crtc_state->stream && old_dm_crtc_state->stream) {
+			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);
+
+		update_type = dc_check_update_surfaces_for_stream(dc, updates, num_plane,
+								  &stream_update, status);
+
+		if (update_type > UPDATE_TYPE_MED) {
 			update_type = UPDATE_TYPE_FULL;
 			goto cleanup;
 		}
@@ -5903,6 +5920,8 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	struct drm_connector_state *old_con_state, *new_con_state;
 	struct drm_crtc *crtc;
 	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;
 
@@ -5921,7 +5940,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		if (!drm_atomic_crtc_needs_modeset(new_crtc_state) &&
 		    !new_crtc_state->color_mgmt_changed &&
-		    !new_crtc_state->vrr_enabled)
+		    old_crtc_state->vrr_enabled == new_crtc_state->vrr_enabled)
 			continue;
 
 		if (!new_crtc_state->enable)
@@ -5937,27 +5956,47 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	}
 
 	/* Remove exiting planes if they are modified */
-	ret = dm_update_planes_state(dc, state, false, &lock_and_validation_needed);
-	if (ret) {
-		goto fail;
+	for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) {
+		ret = dm_update_plane_state(dc, state, plane,
+					    old_plane_state,
+					    new_plane_state,
+					    false,
+					    &lock_and_validation_needed);
+		if (ret)
+			goto fail;
 	}
 
 	/* Disable all crtcs which require disable */
-	ret = dm_update_crtcs_state(&adev->dm, state, false, &lock_and_validation_needed);
-	if (ret) {
-		goto fail;
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		ret = dm_update_crtc_state(&adev->dm, state, crtc,
+					   old_crtc_state,
+					   new_crtc_state,
+					   false,
+					   &lock_and_validation_needed);
+		if (ret)
+			goto fail;
 	}
 
 	/* Enable all crtcs which require enable */
-	ret = dm_update_crtcs_state(&adev->dm, state, true, &lock_and_validation_needed);
-	if (ret) {
-		goto fail;
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		ret = dm_update_crtc_state(&adev->dm, state, crtc,
+					   old_crtc_state,
+					   new_crtc_state,
+					   true,
+					   &lock_and_validation_needed);
+		if (ret)
+			goto fail;
 	}
 
 	/* Add new/modified planes */
-	ret = dm_update_planes_state(dc, state, true, &lock_and_validation_needed);
-	if (ret) {
-		goto fail;
+	for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) {
+		ret = dm_update_plane_state(dc, state, plane,
+					    old_plane_state,
+					    new_plane_state,
+					    true,
+					    &lock_and_validation_needed);
+		if (ret)
+			goto fail;
 	}
 
 	/* Run this here since we want to validate the streams we created */
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 f088ac585978..a10e3a50d9ef 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
@@ -64,8 +64,10 @@ amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, const char *src_name,
 
 int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name)
 {
+	struct amdgpu_device *adev = crtc->dev->dev_private;
 	struct dm_crtc_state *crtc_state = to_dm_crtc_state(crtc->state);
 	struct dc_stream_state *stream_state = crtc_state->stream;
+	bool enable;
 
 	enum amdgpu_dm_pipe_crc_source source = dm_parse_crc_source(src_name);
 
@@ -80,29 +82,33 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name)
 		return -EINVAL;
 	}
 
-	/* When enabling CRC, we should also disable dithering. */
-	if (source == AMDGPU_DM_PIPE_CRC_SOURCE_AUTO) {
-		if (dc_stream_configure_crc(stream_state->ctx->dc,
-					    stream_state,
-					    true, true)) {
-			crtc_state->crc_enabled = true;
-			dc_stream_set_dither_option(stream_state,
-						    DITHER_OPTION_TRUN8);
-		}
-		else
-			return -EINVAL;
-	} else {
-		if (dc_stream_configure_crc(stream_state->ctx->dc,
-					    stream_state,
-					    false, false)) {
-			crtc_state->crc_enabled = false;
-			dc_stream_set_dither_option(stream_state,
-						    DITHER_OPTION_DEFAULT);
-		}
-		else
-			return -EINVAL;
+	enable = (source == AMDGPU_DM_PIPE_CRC_SOURCE_AUTO);
+
+	mutex_lock(&adev->dm.dc_lock);
+	if (!dc_stream_configure_crc(stream_state->ctx->dc, stream_state,
+				     enable, enable)) {
+		mutex_unlock(&adev->dm.dc_lock);
+		return -EINVAL;
 	}
 
+	/* When enabling CRC, we should also disable dithering. */
+	dc_stream_set_dither_option(stream_state,
+				    enable ? DITHER_OPTION_TRUN8
+					   : DITHER_OPTION_DEFAULT);
+
+	mutex_unlock(&adev->dm.dc_lock);
+
+	/*
+	 * Reading the CRC requires the vblank interrupt handler to be
+	 * enabled. Keep a reference until CRC capture stops.
+	 */
+	if (!crtc_state->crc_enabled && enable)
+		drm_crtc_vblank_get(crtc);
+	else if (crtc_state->crc_enabled && !enable)
+		drm_crtc_vblank_put(crtc);
+
+	crtc_state->crc_enabled = enable;
+
 	/* Reset crc_skipped on dm state */
 	crtc_state->crc_skip_count = 0;
 	return 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 9a7ac58eb18e..4a55cde027cf 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
@@ -671,6 +671,25 @@ static ssize_t dp_phy_test_pattern_debugfs_write(struct file *f, const char __us
 	return bytes_from_user;
 }
 
+/*
+ * Returns the min and max vrr vfreq through the connector's debugfs file.
+ * Example usage: cat /sys/kernel/debug/dri/0/DP-1/vrr_range
+ */
+static int vrr_range_show(struct seq_file *m, void *data)
+{
+	struct drm_connector *connector = m->private;
+	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
+
+	if (connector->status != connector_status_connected)
+		return -ENODEV;
+
+	seq_printf(m, "Min: %u\n", (unsigned int)aconnector->min_vfreq);
+	seq_printf(m, "Max: %u\n", (unsigned int)aconnector->max_vfreq);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(vrr_range);
+
 static const struct file_operations dp_link_settings_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_link_settings_read,
@@ -697,7 +716,8 @@ static const struct {
 } dp_debugfs_entries[] = {
 		{"link_settings", &dp_link_settings_debugfs_fops},
 		{"phy_settings", &dp_phy_settings_debugfs_fop},
-		{"test_pattern", &dp_phy_test_pattern_fops}
+		{"test_pattern", &dp_phy_test_pattern_fops},
+		{"vrr_range", &vrr_range_fops}
 };
 
 int connector_debugfs_init(struct amdgpu_dm_connector *connector)
@@ -783,6 +803,45 @@ static ssize_t dtn_log_write(
 	return size;
 }
 
+/*
+ * Backlight at this moment.  Read only.
+ * As written to display, taking ABM and backlight lut into account.
+ * Ranges from 0x0 to 0x10000 (= 100% PWM)
+ */
+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 dc *dc = adev->dm.dc;
+	unsigned int backlight = dc_get_current_backlight_pwm(dc);
+
+	seq_printf(m, "0x%x\n", backlight);
+	return 0;
+}
+
+/*
+ * Backlight value that is being approached.  Read only.
+ * As written to display, taking ABM and backlight lut into account.
+ * Ranges from 0x0 to 0x10000 (= 100% PWM)
+ */
+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 dc *dc = adev->dm.dc;
+	unsigned int backlight = dc_get_target_backlight_pwm(dc);
+
+	seq_printf(m, "0x%x\n", backlight);
+	return 0;
+}
+
+static const struct drm_info_list amdgpu_dm_debugfs_list[] = {
+	{"amdgpu_current_backlight_pwm", &current_backlight_read},
+	{"amdgpu_target_backlight_pwm", &target_backlight_read},
+};
+
 int dtn_debugfs_init(struct amdgpu_device *adev)
 {
 	static const struct file_operations dtn_log_fops = {
@@ -793,9 +852,15 @@ int dtn_debugfs_init(struct amdgpu_device *adev)
 	};
 
 	struct drm_minor *minor = adev->ddev->primary;
-	struct dentry *root = minor->debugfs_root;
+	struct dentry *ent, *root = minor->debugfs_root;
+	int ret;
+
+	ret = amdgpu_debugfs_add_files(adev, amdgpu_dm_debugfs_list,
+				ARRAY_SIZE(amdgpu_dm_debugfs_list));
+	if (ret)
+		return ret;
 
-	struct dentry *ent = debugfs_create_file(
+	ent = debugfs_create_file(
 		"amdgpu_dm_dtn_log",
 		0644,
 		root,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 39997d977efb..b39766bd2840 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -29,7 +29,7 @@
 #include <linux/i2c.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include <drm/drm_edid.h>
 
@@ -192,7 +192,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 	int bpp = 0;
 	int pbn = 0;
 
-	aconnector = stream->sink->priv;
+	aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
 
 	if (!aconnector || !aconnector->mst_port)
 		return false;
@@ -205,7 +205,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 	mst_port = aconnector->port;
 
 	if (enable) {
-		clock = stream->timing.pix_clk_khz;
+		clock = stream->timing.pix_clk_100hz / 10;
 
 		switch (stream->timing.display_color_depth) {
 
@@ -263,6 +263,13 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 	return true;
 }
 
+/*
+ * poll pending down reply before clear payload allocation table
+ */
+void dm_helpers_dp_mst_poll_pending_down_reply(
+	struct dc_context *ctx,
+	const struct dc_link *link)
+{}
 
 /*
  * Clear payload allocation table before enable MST DP link.
@@ -284,7 +291,7 @@ bool dm_helpers_dp_mst_poll_for_allocation_change_trigger(
 	struct drm_dp_mst_topology_mgr *mst_mgr;
 	int ret;
 
-	aconnector = stream->sink->priv;
+	aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
 
 	if (!aconnector || !aconnector->mst_port)
 		return false;
@@ -312,7 +319,7 @@ bool dm_helpers_dp_mst_send_payload_allocation(
 	struct drm_dp_mst_port *mst_port;
 	int ret;
 
-	aconnector = stream->sink->priv;
+	aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
 
 	if (!aconnector || !aconnector->mst_port)
 		return false;
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 24632727e127..f51d52eb52e6 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
@@ -35,6 +35,8 @@
 
 #include "dc_link_ddc.h"
 
+#include "i2caux_interface.h"
+
 /* #define TRACE_DPCD */
 
 #ifdef TRACE_DPCD
@@ -81,80 +83,24 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 				  struct drm_dp_aux_msg *msg)
 {
 	ssize_t result = 0;
-	enum i2caux_transaction_action action;
-	enum aux_transaction_type type;
+	struct aux_payload payload;
 
 	if (WARN_ON(msg->size > 16))
 		return -E2BIG;
 
-	switch (msg->request & ~DP_AUX_I2C_MOT) {
-	case DP_AUX_NATIVE_READ:
-		type = AUX_TRANSACTION_TYPE_DP;
-		action = I2CAUX_TRANSACTION_ACTION_DP_READ;
-
-		result = dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service,
-					      msg->address,
-					      &msg->reply,
-					      msg->buffer,
-					      msg->size,
-					      type,
-					      action);
-		break;
-	case DP_AUX_NATIVE_WRITE:
-		type = AUX_TRANSACTION_TYPE_DP;
-		action = I2CAUX_TRANSACTION_ACTION_DP_WRITE;
-
-		dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service,
-				     msg->address,
-				     &msg->reply,
-				     msg->buffer,
-				     msg->size,
-				     type,
-				     action);
-		result = msg->size;
-		break;
-	case DP_AUX_I2C_READ:
-		type = AUX_TRANSACTION_TYPE_I2C;
-		if (msg->request & DP_AUX_I2C_MOT)
-			action = I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT;
-		else
-			action = I2CAUX_TRANSACTION_ACTION_I2C_READ;
-
-		result = dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service,
-					      msg->address,
-					      &msg->reply,
-					      msg->buffer,
-					      msg->size,
-					      type,
-					      action);
-		break;
-	case DP_AUX_I2C_WRITE:
-		type = AUX_TRANSACTION_TYPE_I2C;
-		if (msg->request & DP_AUX_I2C_MOT)
-			action = I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT;
-		else
-			action = I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-
-		dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service,
-				     msg->address,
-				     &msg->reply,
-				     msg->buffer,
-				     msg->size,
-				     type,
-				     action);
-		result = msg->size;
-		break;
-	default:
-		return -EINVAL;
-	}
+	payload.address = msg->address;
+	payload.data = msg->buffer;
+	payload.length = msg->size;
+	payload.reply = &msg->reply;
+	payload.i2c_over_aux = (msg->request & DP_AUX_NATIVE_WRITE) == 0;
+	payload.write = (msg->request & DP_AUX_I2C_READ) == 0;
+	payload.mot = (msg->request & DP_AUX_I2C_MOT) != 0;
+	payload.defer_delay = 0;
 
-#ifdef TRACE_DPCD
-	log_dpcd(msg->request,
-		 msg->address,
-		 msg->buffer,
-		 msg->size,
-		 r == DDC_RESULT_SUCESSFULL);
-#endif
+	result = dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service, &payload);
+
+	if (payload.write)
+		result = msg->size;
 
 	if (result < 0) /* DC doesn't know about kernel error codes */
 		result = -EIO;
@@ -228,6 +174,11 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector)
 		aconnector->edid = edid;
 	}
 
+	if (aconnector->dc_sink && aconnector->dc_sink->sink_signal == SIGNAL_TYPE_VIRTUAL) {
+		dc_sink_release(aconnector->dc_sink);
+		aconnector->dc_sink = NULL;
+	}
+
 	if (!aconnector->dc_sink) {
 		struct dc_sink *dc_sink;
 		struct dc_sink_init_data init_params = {
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 9d2d6986b983..a114954d6a5b 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
@@ -25,7 +25,7 @@
 #include <linux/acpi.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include "dm_services.h"
 #include "amdgpu.h"
@@ -559,6 +559,58 @@ void pp_rv_set_pme_wa_enable(struct pp_smu *pp)
 	pp_funcs->notify_smu_enable_pwe(pp_handle);
 }
 
+void pp_rv_set_active_display_count(struct pp_smu *pp, int count)
+{
+	const struct dc_context *ctx = pp->dm;
+	struct amdgpu_device *adev = ctx->driver_context;
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->set_active_display_count)
+		return;
+
+	pp_funcs->set_active_display_count(pp_handle, count);
+}
+
+void pp_rv_set_min_deep_sleep_dcfclk(struct pp_smu *pp, int clock)
+{
+	const struct dc_context *ctx = pp->dm;
+	struct amdgpu_device *adev = ctx->driver_context;
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->set_min_deep_sleep_dcefclk)
+		return;
+
+	pp_funcs->set_min_deep_sleep_dcefclk(pp_handle, clock);
+}
+
+void pp_rv_set_hard_min_dcefclk_by_freq(struct pp_smu *pp, int clock)
+{
+	const struct dc_context *ctx = pp->dm;
+	struct amdgpu_device *adev = ctx->driver_context;
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->set_hard_min_dcefclk_by_freq)
+		return;
+
+	pp_funcs->set_hard_min_dcefclk_by_freq(pp_handle, clock);
+}
+
+void pp_rv_set_hard_min_fclk_by_freq(struct pp_smu *pp, int mhz)
+{
+	const struct dc_context *ctx = pp->dm;
+	struct amdgpu_device *adev = ctx->driver_context;
+	void *pp_handle = adev->powerplay.pp_handle;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->set_hard_min_fclk_by_freq)
+		return;
+
+	pp_funcs->set_hard_min_fclk_by_freq(pp_handle, mhz);
+}
+
 void dm_pp_get_funcs_rv(
 		struct dc_context *ctx,
 		struct pp_smu_funcs_rv *funcs)
@@ -567,4 +619,9 @@ void dm_pp_get_funcs_rv(
 	funcs->set_display_requirement = pp_rv_set_display_requirement;
 	funcs->set_wm_ranges = pp_rv_set_wm_ranges;
 	funcs->set_pme_wa_enable = pp_rv_set_pme_wa_enable;
+	funcs->set_display_count = pp_rv_set_active_display_count;
+	funcs->set_min_deep_sleep_dcfclk = pp_rv_set_min_deep_sleep_dcfclk;
+	funcs->set_hard_min_dcfclk_by_freq = pp_rv_set_hard_min_dcefclk_by_freq;
+	funcs->set_hard_min_fclk_by_freq = pp_rv_set_hard_min_fclk_by_freq;
 }
+
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
index 516795342dd2..d915e8c8769b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
@@ -27,7 +27,7 @@
 #include <linux/acpi.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/amdgpu_drm.h>
 #include "dm_services.h"
 #include "amdgpu.h"
diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile
index aed538a4d1ba..b8ddb4acccdb 100644
--- a/drivers/gpu/drm/amd/display/dc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/Makefile
@@ -23,7 +23,7 @@
 # Makefile for Display Core (dc) component.
 #
 
-DC_LIBS = basics bios calcs dce gpio i2caux irq virtual
+DC_LIBS = basics bios calcs dce gpio irq virtual
 
 ifdef CONFIG_DRM_AMD_DC_DCN1_0
 DC_LIBS += dcn10 dml
@@ -41,7 +41,8 @@ AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LI
 include $(AMD_DC)
 
 DISPLAY_CORE = dc.o dc_link.o dc_resource.o dc_hw_sequencer.o dc_sink.o \
-dc_surface.o dc_link_hwss.o dc_link_dp.o dc_link_ddc.o dc_debug.o dc_stream.o
+dc_surface.o dc_link_hwss.o dc_link_dp.o dc_link_ddc.o dc_debug.o dc_stream.o \
+dc_vm_helper.o
 
 AMD_DISPLAY_CORE = $(addprefix $(AMDDALPATH)/dc/core/,$(DISPLAY_CORE))
 
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
index c2ab026aee91..a4c97d32e751 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
@@ -835,18 +835,6 @@ static enum bp_result bios_parser_enable_crtc(
 	return bp->cmd_tbl.enable_crtc(bp, id, enable);
 }
 
-static enum bp_result bios_parser_crtc_source_select(
-	struct dc_bios *dcb,
-	struct bp_crtc_source_select *bp_params)
-{
-	struct bios_parser *bp = BP_FROM_DCB(dcb);
-
-	if (!bp->cmd_tbl.select_crtc_source)
-		return BP_RESULT_FAILURE;
-
-	return bp->cmd_tbl.select_crtc_source(bp, bp_params);
-}
-
 static enum bp_result bios_parser_enable_disp_power_gating(
 	struct dc_bios *dcb,
 	enum controller_id controller_id,
@@ -2842,8 +2830,6 @@ static const struct dc_vbios_funcs vbios_funcs = {
 
 	.program_crtc_timing = bios_parser_program_crtc_timing, /* still use.  should probably retire and program directly */
 
-	.crtc_source_select = bios_parser_crtc_source_select,  /* still use.  should probably retire and program directly */
-
 	.program_display_engine_pll = bios_parser_program_display_engine_pll,
 
 	.enable_disp_power_gating = bios_parser_enable_disp_power_gating,
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 c513ab6f3843..a1c56f29cfeb 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -1083,18 +1083,6 @@ static enum bp_result bios_parser_enable_crtc(
 	return bp->cmd_tbl.enable_crtc(bp, id, enable);
 }
 
-static enum bp_result bios_parser_crtc_source_select(
-	struct dc_bios *dcb,
-	struct bp_crtc_source_select *bp_params)
-{
-	struct bios_parser *bp = BP_FROM_DCB(dcb);
-
-	if (!bp->cmd_tbl.select_crtc_source)
-		return BP_RESULT_FAILURE;
-
-	return bp->cmd_tbl.select_crtc_source(bp, bp_params);
-}
-
 static enum bp_result bios_parser_enable_disp_power_gating(
 	struct dc_bios *dcb,
 	enum controller_id controller_id,
@@ -1899,8 +1887,6 @@ static const struct dc_vbios_funcs vbios_funcs = {
 
 	.is_accelerated_mode = bios_parser_is_accelerated_mode,
 
-	.is_active_display = bios_is_active_display,
-
 	.set_scratch_critical_state = bios_parser_set_scratch_critical_state,
 
 
@@ -1917,8 +1903,6 @@ static const struct dc_vbios_funcs vbios_funcs = {
 
 	.program_crtc_timing = bios_parser_program_crtc_timing,
 
-	.crtc_source_select = bios_parser_crtc_source_select,
-
 	.enable_disp_power_gating = bios_parser_enable_disp_power_gating,
 
 	.bios_parser_destroy = firmware_parser_destroy,
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
index fdda8aa8e303..fce46ab54c54 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
@@ -83,101 +83,7 @@ uint32_t bios_get_vga_enabled_displays(
 {
 	uint32_t active_disp = 1;
 
-	if (bios->regs->BIOS_SCRATCH_3) /*follow up with other asic, todo*/
-		active_disp = REG_READ(BIOS_SCRATCH_3) & 0XFFFF;
+	active_disp = REG_READ(BIOS_SCRATCH_3) & 0XFFFF;
 	return active_disp;
 }
 
-bool bios_is_active_display(
-		struct dc_bios *bios,
-		enum signal_type signal,
-		const struct connector_device_tag_info *device_tag)
-{
-	uint32_t active = 0;
-	uint32_t connected = 0;
-	uint32_t bios_scratch_0 = 0;
-	uint32_t bios_scratch_3 = 0;
-
-	switch (signal)	{
-	case SIGNAL_TYPE_DVI_SINGLE_LINK:
-	case SIGNAL_TYPE_DVI_DUAL_LINK:
-	case SIGNAL_TYPE_HDMI_TYPE_A:
-	case SIGNAL_TYPE_DISPLAY_PORT:
-	case SIGNAL_TYPE_DISPLAY_PORT_MST:
-		{
-			if (device_tag->dev_id.device_type == DEVICE_TYPE_DFP) {
-				switch (device_tag->dev_id.enum_id)	{
-				case 1:
-					{
-						active    = ATOM_S3_DFP1_ACTIVE;
-						connected = 0x0008;	//ATOM_DISPLAY_DFP1_CONNECT
-					}
-					break;
-
-				case 2:
-					{
-						active    = ATOM_S3_DFP2_ACTIVE;
-						connected = 0x0080; //ATOM_DISPLAY_DFP2_CONNECT
-					}
-					break;
-
-				case 3:
-					{
-						active    = ATOM_S3_DFP3_ACTIVE;
-						connected = 0x0200; //ATOM_DISPLAY_DFP3_CONNECT
-					}
-					break;
-
-				case 4:
-					{
-						active    = ATOM_S3_DFP4_ACTIVE;
-						connected = 0x0400;	//ATOM_DISPLAY_DFP4_CONNECT
-					}
-					break;
-
-				case 5:
-					{
-						active    = ATOM_S3_DFP5_ACTIVE;
-						connected = 0x0800; //ATOM_DISPLAY_DFP5_CONNECT
-					}
-					break;
-
-				case 6:
-					{
-						active    = ATOM_S3_DFP6_ACTIVE;
-						connected = 0x0040; //ATOM_DISPLAY_DFP6_CONNECT
-					}
-					break;
-
-				default:
-					break;
-				}
-				}
-			}
-			break;
-
-	case SIGNAL_TYPE_LVDS:
-	case SIGNAL_TYPE_EDP:
-		{
-			active    = ATOM_S3_LCD1_ACTIVE;
-			connected = 0x0002;	//ATOM_DISPLAY_LCD1_CONNECT
-		}
-		break;
-
-	default:
-		break;
-	}
-
-
-	if (bios->regs->BIOS_SCRATCH_0) /*follow up with other asic, todo*/
-		bios_scratch_0 = REG_READ(BIOS_SCRATCH_0);
-	if (bios->regs->BIOS_SCRATCH_3) /*follow up with other asic, todo*/
-		bios_scratch_3 = REG_READ(BIOS_SCRATCH_3);
-
-	bios_scratch_3 &= ATOM_S3_DEVICE_ACTIVE_MASK;
-	if ((active & bios_scratch_3) && (connected & bios_scratch_0))
-		return true;
-
-	return false;
-}
-
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
index f33cac2147e3..75a29e68fb27 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
@@ -35,10 +35,6 @@ bool bios_is_accelerated_mode(struct dc_bios *bios);
 void bios_set_scratch_acc_mode_change(struct dc_bios *bios);
 void bios_set_scratch_critical_state(struct dc_bios *bios, bool state);
 uint32_t bios_get_vga_enabled_displays(struct dc_bios *bios);
-bool bios_is_active_display(
-	struct dc_bios *bios,
-	enum signal_type signal,
-	const struct connector_device_tag_info *device_tag);
 
 #define GET_IMAGE(type, offset) ((type *) bios_get_image(&bp->base, offset, sizeof(type)))
 
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 2bd7cd97e00d..5815983caaf8 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c
@@ -55,7 +55,6 @@ static void init_adjust_display_pll(struct bios_parser *bp);
 static void init_dac_encoder_control(struct bios_parser *bp);
 static void init_dac_output_control(struct bios_parser *bp);
 static void init_set_crtc_timing(struct bios_parser *bp);
-static void init_select_crtc_source(struct bios_parser *bp);
 static void init_enable_crtc(struct bios_parser *bp);
 static void init_enable_crtc_mem_req(struct bios_parser *bp);
 static void init_external_encoder_control(struct bios_parser *bp);
@@ -73,7 +72,6 @@ void dal_bios_parser_init_cmd_tbl(struct bios_parser *bp)
 	init_dac_encoder_control(bp);
 	init_dac_output_control(bp);
 	init_set_crtc_timing(bp);
-	init_select_crtc_source(bp);
 	init_enable_crtc(bp);
 	init_enable_crtc_mem_req(bp);
 	init_program_clock(bp);
@@ -964,9 +962,9 @@ static enum bp_result set_pixel_clock_v3(
 	allocation.sPCLKInput.ucPostDiv =
 			(uint8_t)bp_params->pixel_clock_post_divider;
 
-	/* We need to convert from KHz units into 10KHz units */
+	/* We need to convert from 100Hz units into 10KHz units */
 	allocation.sPCLKInput.usPixelClock =
-			cpu_to_le16((uint16_t)(bp_params->target_pixel_clock / 10));
+			cpu_to_le16((uint16_t)(bp_params->target_pixel_clock_100hz / 100));
 
 	params = (PIXEL_CLOCK_PARAMETERS_V3 *)&allocation.sPCLKInput;
 	params->ucTransmitterId =
@@ -1042,9 +1040,9 @@ static enum bp_result set_pixel_clock_v5(
 				(uint8_t)bp->cmd_helper->encoder_mode_bp_to_atom(
 						bp_params->signal_type, false);
 
-		/* We need to convert from KHz units into 10KHz units */
+		/* We need to convert from 100Hz units into 10KHz units */
 		clk.sPCLKInput.usPixelClock =
-				cpu_to_le16((uint16_t)(bp_params->target_pixel_clock / 10));
+				cpu_to_le16((uint16_t)(bp_params->target_pixel_clock_100hz / 100));
 
 		if (bp_params->flags.FORCE_PROGRAMMING_OF_PLL)
 			clk.sPCLKInput.ucMiscInfo |=
@@ -1118,9 +1116,9 @@ static enum bp_result set_pixel_clock_v6(
 				(uint8_t) bp->cmd_helper->encoder_mode_bp_to_atom(
 						bp_params->signal_type, false);
 
-		/* We need to convert from KHz units into 10KHz units */
+		/* We need to convert from 100 Hz units into 10KHz units */
 		clk.sPCLKInput.ulCrtcPclkFreq.ulPixelClock =
-				cpu_to_le32(bp_params->target_pixel_clock / 10);
+				cpu_to_le32(bp_params->target_pixel_clock_100hz / 100);
 
 		if (bp_params->flags.FORCE_PROGRAMMING_OF_PLL) {
 			clk.sPCLKInput.ucMiscInfo |=
@@ -1182,8 +1180,7 @@ static enum bp_result set_pixel_clock_v7(
 		clk.ucTransmitterID = bp->cmd_helper->encoder_id_to_atom(dal_graphics_object_id_get_encoder_id(bp_params->encoder_object_id));
 		clk.ucEncoderMode = (uint8_t) bp->cmd_helper->encoder_mode_bp_to_atom(bp_params->signal_type, false);
 
-		/* We need to convert from KHz units into 10KHz units */
-		clk.ulPixelClock = cpu_to_le32(bp_params->target_pixel_clock * 10);
+		clk.ulPixelClock = cpu_to_le32(bp_params->target_pixel_clock_100hz);
 
 		clk.ucDeepColorRatio = (uint8_t) bp->cmd_helper->transmitter_color_depth_to_atom(bp_params->color_depth);
 
@@ -1899,120 +1896,6 @@ static enum bp_result set_crtc_using_dtd_timing_v3(
 /*******************************************************************************
  ********************************************************************************
  **
- **                  SELECT CRTC SOURCE
- **
- ********************************************************************************
- *******************************************************************************/
-
-static enum bp_result select_crtc_source_v2(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params);
-static enum bp_result select_crtc_source_v3(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params);
-
-static void init_select_crtc_source(struct bios_parser *bp)
-{
-	switch (BIOS_CMD_TABLE_PARA_REVISION(SelectCRTC_Source)) {
-	case 2:
-		bp->cmd_tbl.select_crtc_source = select_crtc_source_v2;
-		break;
-	case 3:
-		bp->cmd_tbl.select_crtc_source = select_crtc_source_v3;
-		break;
-	default:
-		dm_output_to_console("Don't select_crtc_source enable_crtc for v%d\n",
-			 BIOS_CMD_TABLE_PARA_REVISION(SelectCRTC_Source));
-		bp->cmd_tbl.select_crtc_source = NULL;
-		break;
-	}
-}
-
-static enum bp_result select_crtc_source_v2(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params)
-{
-	enum bp_result result = BP_RESULT_FAILURE;
-	SELECT_CRTC_SOURCE_PARAMETERS_V2 params;
-	uint8_t atom_controller_id;
-	uint32_t atom_engine_id;
-	enum signal_type s = bp_params->signal;
-
-	memset(&params, 0, sizeof(params));
-
-	/* set controller id */
-	if (bp->cmd_helper->controller_id_to_atom(
-			bp_params->controller_id, &atom_controller_id))
-		params.ucCRTC = atom_controller_id;
-	else
-		return BP_RESULT_FAILURE;
-
-	/* set encoder id */
-	if (bp->cmd_helper->engine_bp_to_atom(
-			bp_params->engine_id, &atom_engine_id))
-		params.ucEncoderID = (uint8_t)atom_engine_id;
-	else
-		return BP_RESULT_FAILURE;
-
-	if (SIGNAL_TYPE_EDP == s ||
-			(SIGNAL_TYPE_DISPLAY_PORT == s &&
-					SIGNAL_TYPE_LVDS == bp_params->sink_signal))
-		s = SIGNAL_TYPE_LVDS;
-
-	params.ucEncodeMode =
-			(uint8_t)bp->cmd_helper->encoder_mode_bp_to_atom(
-					s, bp_params->enable_dp_audio);
-
-	if (EXEC_BIOS_CMD_TABLE(SelectCRTC_Source, params))
-		result = BP_RESULT_OK;
-
-	return result;
-}
-
-static enum bp_result select_crtc_source_v3(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params)
-{
-	bool result = BP_RESULT_FAILURE;
-	SELECT_CRTC_SOURCE_PARAMETERS_V3 params;
-	uint8_t atom_controller_id;
-	uint32_t atom_engine_id;
-	enum signal_type s = bp_params->signal;
-
-	memset(&params, 0, sizeof(params));
-
-	if (bp->cmd_helper->controller_id_to_atom(bp_params->controller_id,
-			&atom_controller_id))
-		params.ucCRTC = atom_controller_id;
-	else
-		return result;
-
-	if (bp->cmd_helper->engine_bp_to_atom(bp_params->engine_id,
-			&atom_engine_id))
-		params.ucEncoderID = (uint8_t)atom_engine_id;
-	else
-		return result;
-
-	if (SIGNAL_TYPE_EDP == s ||
-			(SIGNAL_TYPE_DISPLAY_PORT == s &&
-					SIGNAL_TYPE_LVDS == bp_params->sink_signal))
-		s = SIGNAL_TYPE_LVDS;
-
-	params.ucEncodeMode =
-			bp->cmd_helper->encoder_mode_bp_to_atom(
-					s, bp_params->enable_dp_audio);
-	/* Needed for VBIOS Random Spatial Dithering feature */
-	params.ucDstBpc = (uint8_t)(bp_params->display_output_bit_depth);
-
-	if (EXEC_BIOS_CMD_TABLE(SelectCRTC_Source, params))
-		result = BP_RESULT_OK;
-
-	return result;
-}
-
-/*******************************************************************************
- ********************************************************************************
- **
  **                  ENABLE CRTC
  **
  ********************************************************************************
@@ -2164,7 +2047,7 @@ static enum bp_result program_clock_v5(
 	/* We need to convert from KHz units into 10KHz units */
 	params.sPCLKInput.ucPpll = (uint8_t) atom_pll_id;
 	params.sPCLKInput.usPixelClock =
-			cpu_to_le16((uint16_t) (bp_params->target_pixel_clock / 10));
+			cpu_to_le16((uint16_t) (bp_params->target_pixel_clock_100hz / 100));
 	params.sPCLKInput.ucCRTC = (uint8_t) ATOM_CRTC_INVALID;
 
 	if (bp_params->flags.SET_EXTERNAL_REF_DIV_SRC)
@@ -2196,7 +2079,7 @@ static enum bp_result program_clock_v6(
 	/* We need to convert from KHz units into 10KHz units */
 	params.sPCLKInput.ucPpll = (uint8_t)atom_pll_id;
 	params.sPCLKInput.ulDispEngClkFreq =
-			cpu_to_le32(bp_params->target_pixel_clock / 10);
+			cpu_to_le32(bp_params->target_pixel_clock_100hz / 100);
 
 	if (bp_params->flags.SET_EXTERNAL_REF_DIV_SRC)
 		params.sPCLKInput.ucMiscInfo |= PIXEL_CLOCK_MISC_REF_DIV_SRC;
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.h b/drivers/gpu/drm/amd/display/dc/bios/command_table.h
index 94f3d43a7471..ad533775e724 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.h
@@ -71,9 +71,6 @@ struct cmd_tbl {
 	enum bp_result (*set_crtc_timing)(
 		struct bios_parser *bp,
 		struct bp_hw_crtc_timing_parameters *bp_params);
-	enum bp_result (*select_crtc_source)(
-		struct bios_parser *bp,
-		struct bp_crtc_source_select *bp_params);
 	enum bp_result (*enable_crtc)(
 		struct bios_parser *bp,
 		enum controller_id controller_id,
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 2b5dc499a35e..bb2e8105e6ab 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
@@ -301,17 +301,17 @@ static enum bp_result set_pixel_clock_v7(
 			cmd_helper->encoder_mode_bp_to_atom(
 				bp_params->signal_type, false);
 
-		/* We need to convert from KHz units into 10KHz units */
-		clk.pixclk_100hz = cpu_to_le32(bp_params->target_pixel_clock *
-				10);
+		clk.pixclk_100hz = cpu_to_le32(bp_params->target_pixel_clock_100hz);
 
 		clk.deep_color_ratio =
 			(uint8_t) bp->cmd_helper->
 				transmitter_color_depth_to_atom(
 					bp_params->color_depth);
-		DC_LOG_BIOS("%s:program display clock = %d"\
-				"colorDepth = %d\n", __func__,\
-				bp_params->target_pixel_clock, bp_params->color_depth);
+
+		DC_LOG_BIOS("%s:program display clock = %d, tg = %d, pll = %d, "\
+				"colorDepth = %d\n", __func__,
+				bp_params->target_pixel_clock_100hz, (int)controller_id,
+				pll_id, bp_params->color_depth);
 
 		if (bp_params->flags.FORCE_PROGRAMMING_OF_PLL)
 			clk.miscinfo |= PIXEL_CLOCK_V7_MISC_FORCE_PROG_PPLL;
@@ -463,75 +463,6 @@ static enum bp_result set_crtc_using_dtd_timing_v3(
 /******************************************************************************
  ******************************************************************************
  **
- **                  SELECT CRTC SOURCE
- **
- ******************************************************************************
- *****************************************************************************/
-
-
-static enum bp_result select_crtc_source_v3(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params);
-
-static void init_select_crtc_source(struct bios_parser *bp)
-{
-	switch (BIOS_CMD_TABLE_PARA_REVISION(selectcrtc_source)) {
-	case 3:
-		bp->cmd_tbl.select_crtc_source = select_crtc_source_v3;
-		break;
-	default:
-		dm_output_to_console("Don't select_crtc_source enable_crtc for v%d\n",
-			 BIOS_CMD_TABLE_PARA_REVISION(selectcrtc_source));
-		bp->cmd_tbl.select_crtc_source = NULL;
-		break;
-	}
-}
-
-
-static enum bp_result select_crtc_source_v3(
-	struct bios_parser *bp,
-	struct bp_crtc_source_select *bp_params)
-{
-	bool result = BP_RESULT_FAILURE;
-	struct select_crtc_source_parameters_v2_3 params;
-	uint8_t atom_controller_id;
-	uint32_t atom_engine_id;
-	enum signal_type s = bp_params->signal;
-
-	memset(&params, 0, sizeof(params));
-
-	if (bp->cmd_helper->controller_id_to_atom(bp_params->controller_id,
-			&atom_controller_id))
-		params.crtc_id = atom_controller_id;
-	else
-		return result;
-
-	if (bp->cmd_helper->engine_bp_to_atom(bp_params->engine_id,
-			&atom_engine_id))
-		params.encoder_id = (uint8_t)atom_engine_id;
-	else
-		return result;
-
-	if (s == SIGNAL_TYPE_EDP ||
-		(s == SIGNAL_TYPE_DISPLAY_PORT && bp_params->sink_signal ==
-							SIGNAL_TYPE_LVDS))
-		s = SIGNAL_TYPE_LVDS;
-
-	params.encode_mode =
-			bp->cmd_helper->encoder_mode_bp_to_atom(
-					s, bp_params->enable_dp_audio);
-	/* Needed for VBIOS Random Spatial Dithering feature */
-	params.dst_bpc = (uint8_t)(bp_params->display_output_bit_depth);
-
-	if (EXEC_BIOS_CMD_TABLE(selectcrtc_source, params))
-		result = BP_RESULT_OK;
-
-	return result;
-}
-
-/******************************************************************************
- ******************************************************************************
- **
  **                  ENABLE CRTC
  **
  ******************************************************************************
@@ -808,7 +739,6 @@ void dal_firmware_parser_init_cmd_tbl(struct bios_parser *bp)
 
 	init_set_crtc_timing(bp);
 
-	init_select_crtc_source(bp);
 	init_enable_crtc(bp);
 
 	init_external_encoder_control(bp);
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.h b/drivers/gpu/drm/amd/display/dc/bios/command_table2.h
index ec1c0c9f3f1d..7a2af24dfe60 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.h
@@ -71,9 +71,6 @@ struct cmd_tbl {
 	enum bp_result (*set_crtc_timing)(
 		struct bios_parser *bp,
 		struct bp_hw_crtc_timing_parameters *bp_params);
-	enum bp_result (*select_crtc_source)(
-		struct bios_parser *bp,
-		struct bp_crtc_source_select *bp_params);
 	enum bp_result (*enable_crtc)(
 		struct bios_parser *bp,
 		enum controller_id controller_id,
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c
index 9ebe30ba4dab..f3aa7b53d2aa 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c
@@ -2792,7 +2792,7 @@ static void populate_initial_data(
 		data->lpt_en[num_displays + 4] = false;
 		data->h_total[num_displays + 4] = bw_int_to_fixed(pipe[i].stream->timing.h_total);
 		data->v_total[num_displays + 4] = bw_int_to_fixed(pipe[i].stream->timing.v_total);
-		data->pixel_rate[num_displays + 4] = bw_frc_to_fixed(pipe[i].stream->timing.pix_clk_khz, 1000);
+		data->pixel_rate[num_displays + 4] = bw_frc_to_fixed(pipe[i].stream->timing.pix_clk_100hz, 10000);
 		data->src_width[num_displays + 4] = bw_int_to_fixed(pipe[i].plane_res.scl_data.viewport.width);
 		data->pitch_in_pixels[num_displays + 4] = data->src_width[num_displays + 4];
 		data->src_height[num_displays + 4] = bw_int_to_fixed(pipe[i].plane_res.scl_data.viewport.height);
@@ -2881,7 +2881,7 @@ static void populate_initial_data(
 
 	/* Pipes without underlay after */
 	for (i = 0; i < pipe_count; i++) {
-		unsigned int pixel_clock_khz;
+		unsigned int pixel_clock_100hz;
 		if (!pipe[i].stream || pipe[i].bottom_pipe)
 			continue;
 
@@ -2890,10 +2890,10 @@ static void populate_initial_data(
 		data->lpt_en[num_displays + 4] = false;
 		data->h_total[num_displays + 4] = bw_int_to_fixed(pipe[i].stream->timing.h_total);
 		data->v_total[num_displays + 4] = bw_int_to_fixed(pipe[i].stream->timing.v_total);
-		pixel_clock_khz = pipe[i].stream->timing.pix_clk_khz;
+		pixel_clock_100hz = pipe[i].stream->timing.pix_clk_100hz;
 		if (pipe[i].stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
-			pixel_clock_khz *= 2;
-		data->pixel_rate[num_displays + 4] = bw_frc_to_fixed(pixel_clock_khz, 1000);
+			pixel_clock_100hz *= 2;
+		data->pixel_rate[num_displays + 4] = bw_frc_to_fixed(pixel_clock_100hz, 10000);
 		if (pipe[i].plane_state) {
 			data->src_width[num_displays + 4] = bw_int_to_fixed(pipe[i].plane_res.scl_data.viewport.width);
 			data->pitch_in_pixels[num_displays + 4] = data->src_width[num_displays + 4];
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c
index d0fc54f8fb1c..1ef0074302c5 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c
@@ -63,7 +63,7 @@ void scaler_settings_calculation(struct dcn_bw_internal_vars *v)
 		if (v->interlace_output[k] == 1.0) {
 			v->v_ratio[k] = 2.0 * v->v_ratio[k];
 		}
-		if ((v->underscan_output[k] == 1.0)) {
+		if (v->underscan_output[k] == 1.0) {
 			v->h_ratio[k] = v->h_ratio[k] * v->under_scan_factor;
 			v->v_ratio[k] = v->v_ratio[k] * v->under_scan_factor;
 		}
@@ -797,9 +797,40 @@ void mode_support_and_system_configuration(struct dcn_bw_internal_vars *v)
 				else {
 					v->maximum_vstartup = v->v_sync_plus_back_porch[k] - 1.0;
 				}
-				v->line_times_for_prefetch[k] = v->maximum_vstartup - v->urgent_latency / (v->htotal[k] / v->pixel_clock[k]) - (v->time_calc + v->time_setup) / (v->htotal[k] / v->pixel_clock[k]) - (v->dst_y_after_scaler + v->dst_x_after_scaler / v->htotal[k]);
-				v->line_times_for_prefetch[k] =dcn_bw_floor2(4.0 * (v->line_times_for_prefetch[k] + 0.125), 1.0) / 4;
-				v->prefetch_bw[k] = (v->meta_pte_bytes_per_frame[k] + 2.0 * v->meta_row_bytes[k] + 2.0 * v->dpte_bytes_per_row[k] + v->prefetch_lines_y[k] * v->swath_width_yper_state[i][j][k] *dcn_bw_ceil2(v->byte_per_pixel_in_dety[k], 1.0) + v->prefetch_lines_c[k] * v->swath_width_yper_state[i][j][k] / 2.0 *dcn_bw_ceil2(v->byte_per_pixel_in_detc[k], 2.0)) / (v->line_times_for_prefetch[k] * v->htotal[k] / v->pixel_clock[k]);
+
+				do {
+					v->line_times_for_prefetch[k] = v->maximum_vstartup - v->urgent_latency / (v->htotal[k] / v->pixel_clock[k]) - (v->time_calc + v->time_setup) / (v->htotal[k] / v->pixel_clock[k]) - (v->dst_y_after_scaler + v->dst_x_after_scaler / v->htotal[k]);
+					v->line_times_for_prefetch[k] =dcn_bw_floor2(4.0 * (v->line_times_for_prefetch[k] + 0.125), 1.0) / 4;
+					v->prefetch_bw[k] = (v->meta_pte_bytes_per_frame[k] + 2.0 * v->meta_row_bytes[k] + 2.0 * v->dpte_bytes_per_row[k] + v->prefetch_lines_y[k] * v->swath_width_yper_state[i][j][k] *dcn_bw_ceil2(v->byte_per_pixel_in_dety[k], 1.0) + v->prefetch_lines_c[k] * v->swath_width_yper_state[i][j][k] / 2.0 *dcn_bw_ceil2(v->byte_per_pixel_in_detc[k], 2.0)) / (v->line_times_for_prefetch[k] * v->htotal[k] / v->pixel_clock[k]);
+
+					if (v->pte_enable == dcn_bw_yes && v->dcc_enable[k] == dcn_bw_yes) {
+						v->time_for_meta_pte_without_immediate_flip = dcn_bw_max3(
+								v->meta_pte_bytes_frame[k] / v->prefetch_bandwidth[k],
+								v->extra_latency,
+								v->htotal[k] / v->pixel_clock[k] / 4.0);
+					} else {
+						v->time_for_meta_pte_without_immediate_flip = v->htotal[k] / v->pixel_clock[k] / 4.0;
+					}
+
+					if (v->pte_enable == dcn_bw_yes || v->dcc_enable[k] == dcn_bw_yes) {
+						v->time_for_meta_and_dpte_row_without_immediate_flip = dcn_bw_max3((
+								v->meta_row_bytes[k] + v->dpte_bytes_per_row[k]) / v->prefetch_bandwidth[k],
+								v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_without_immediate_flip,
+								v->extra_latency);
+					} else {
+						v->time_for_meta_and_dpte_row_without_immediate_flip = dcn_bw_max2(
+								v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_without_immediate_flip,
+								v->extra_latency - v->time_for_meta_pte_with_immediate_flip);
+					}
+
+					v->lines_for_meta_pte_without_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_pte_without_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
+					v->lines_for_meta_and_dpte_row_without_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_and_dpte_row_without_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
+					v->maximum_vstartup = v->maximum_vstartup - 1;
+
+					if (v->lines_for_meta_pte_without_immediate_flip[k] < 8.0 && v->lines_for_meta_and_dpte_row_without_immediate_flip[k] < 16.0)
+						break;
+
+				} while(1);
 			}
 			v->bw_available_for_immediate_flip = v->return_bw_per_state[i];
 			for (k = 0; k <= v->number_of_active_planes - 1; k++) {
@@ -814,24 +845,18 @@ void mode_support_and_system_configuration(struct dcn_bw_internal_vars *v)
 			for (k = 0; k <= v->number_of_active_planes - 1; k++) {
 				if (v->pte_enable == dcn_bw_yes && v->dcc_enable[k] == dcn_bw_yes) {
 					v->time_for_meta_pte_with_immediate_flip =dcn_bw_max5(v->meta_pte_bytes_per_frame[k] / v->prefetch_bw[k], v->meta_pte_bytes_per_frame[k] * v->total_immediate_flip_bytes[k] / (v->bw_available_for_immediate_flip * (v->meta_pte_bytes_per_frame[k] + v->meta_row_bytes[k] + v->dpte_bytes_per_row[k])), v->extra_latency, v->urgent_latency, v->htotal[k] / v->pixel_clock[k] / 4.0);
-					v->time_for_meta_pte_without_immediate_flip =dcn_bw_max3(v->meta_pte_bytes_per_frame[k] / v->prefetch_bw[k], v->extra_latency, v->htotal[k] / v->pixel_clock[k] / 4.0);
 				}
 				else {
 					v->time_for_meta_pte_with_immediate_flip = v->htotal[k] / v->pixel_clock[k] / 4.0;
-					v->time_for_meta_pte_without_immediate_flip = v->htotal[k] / v->pixel_clock[k] / 4.0;
 				}
 				if (v->pte_enable == dcn_bw_yes || v->dcc_enable[k] == dcn_bw_yes) {
 					v->time_for_meta_and_dpte_row_with_immediate_flip =dcn_bw_max5((v->meta_row_bytes[k] + v->dpte_bytes_per_row[k]) / v->prefetch_bw[k], (v->meta_row_bytes[k] + v->dpte_bytes_per_row[k]) * v->total_immediate_flip_bytes[k] / (v->bw_available_for_immediate_flip * (v->meta_pte_bytes_per_frame[k] + v->meta_row_bytes[k] + v->dpte_bytes_per_row[k])), v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_with_immediate_flip, v->extra_latency, 2.0 * v->urgent_latency);
-					v->time_for_meta_and_dpte_row_without_immediate_flip =dcn_bw_max3((v->meta_row_bytes[k] + v->dpte_bytes_per_row[k]) / v->prefetch_bw[k], v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_without_immediate_flip, v->extra_latency);
 				}
 				else {
 					v->time_for_meta_and_dpte_row_with_immediate_flip =dcn_bw_max2(v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_with_immediate_flip, v->extra_latency - v->time_for_meta_pte_with_immediate_flip);
-					v->time_for_meta_and_dpte_row_without_immediate_flip =dcn_bw_max2(v->htotal[k] / v->pixel_clock[k] - v->time_for_meta_pte_without_immediate_flip, v->extra_latency - v->time_for_meta_pte_without_immediate_flip);
 				}
 				v->lines_for_meta_pte_with_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_pte_with_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
-				v->lines_for_meta_pte_without_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_pte_without_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
 				v->lines_for_meta_and_dpte_row_with_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_and_dpte_row_with_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
-				v->lines_for_meta_and_dpte_row_without_immediate_flip[k] =dcn_bw_floor2(4.0 * (v->time_for_meta_and_dpte_row_without_immediate_flip / (v->htotal[k] / v->pixel_clock[k]) + 0.125), 1.0) / 4;
 				v->line_times_to_request_prefetch_pixel_data_with_immediate_flip = v->line_times_for_prefetch[k] - v->lines_for_meta_pte_with_immediate_flip[k] - v->lines_for_meta_and_dpte_row_with_immediate_flip[k];
 				v->line_times_to_request_prefetch_pixel_data_without_immediate_flip = v->line_times_for_prefetch[k] - v->lines_for_meta_pte_without_immediate_flip[k] - v->lines_for_meta_and_dpte_row_without_immediate_flip[k];
 				if (v->line_times_to_request_prefetch_pixel_data_with_immediate_flip > 0.0) {
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
index 43e4a2be0fa6..12d1842079ae 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
@@ -290,41 +290,34 @@ static void pipe_ctx_to_e2e_pipe_params (
 	switch (pipe->plane_state->tiling_info.gfx9.swizzle) {
 	/* for 4/8/16 high tiles */
 	case DC_SW_LINEAR:
-		input->src.is_display_sw = 1;
 		input->src.macro_tile_size = dm_4k_tile;
 		break;
 	case DC_SW_4KB_S:
 	case DC_SW_4KB_S_X:
-		input->src.is_display_sw = 0;
 		input->src.macro_tile_size = dm_4k_tile;
 		break;
 	case DC_SW_64KB_S:
 	case DC_SW_64KB_S_X:
 	case DC_SW_64KB_S_T:
-		input->src.is_display_sw = 0;
 		input->src.macro_tile_size = dm_64k_tile;
 		break;
 	case DC_SW_VAR_S:
 	case DC_SW_VAR_S_X:
-		input->src.is_display_sw = 0;
 		input->src.macro_tile_size = dm_256k_tile;
 		break;
 
 	/* For 64bpp 2 high tiles */
 	case DC_SW_4KB_D:
 	case DC_SW_4KB_D_X:
-		input->src.is_display_sw = 1;
 		input->src.macro_tile_size = dm_4k_tile;
 		break;
 	case DC_SW_64KB_D:
 	case DC_SW_64KB_D_X:
 	case DC_SW_64KB_D_T:
-		input->src.is_display_sw = 1;
 		input->src.macro_tile_size = dm_64k_tile;
 		break;
 	case DC_SW_VAR_D:
 	case DC_SW_VAR_D_X:
-		input->src.is_display_sw = 1;
 		input->src.macro_tile_size = dm_256k_tile;
 		break;
 
@@ -423,7 +416,7 @@ static void pipe_ctx_to_e2e_pipe_params (
 			- pipe->stream->timing.v_addressable
 			- pipe->stream->timing.v_border_bottom
 			- pipe->stream->timing.v_border_top;
-	input->dest.pixel_rate_mhz = pipe->stream->timing.pix_clk_khz/1000.0;
+	input->dest.pixel_rate_mhz = pipe->stream->timing.pix_clk_100hz/10000.0;
 	input->dest.vstartup_start = pipe->pipe_dlg_param.vstartup_start;
 	input->dest.vupdate_offset = pipe->pipe_dlg_param.vupdate_offset;
 	input->dest.vupdate_offset = pipe->pipe_dlg_param.vupdate_offset;
@@ -670,9 +663,9 @@ static void hack_disable_optional_pipe_split(struct dcn_bw_internal_vars *v)
 }
 
 static void hack_force_pipe_split(struct dcn_bw_internal_vars *v,
-		unsigned int pixel_rate_khz)
+		unsigned int pixel_rate_100hz)
 {
-	float pixel_rate_mhz = pixel_rate_khz / 1000;
+	float pixel_rate_mhz = pixel_rate_100hz / 10000;
 
 	/*
 	 * force enabling pipe split by lower dpp clock for DPM0 to just
@@ -695,7 +688,7 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v,
 
 	if (context->stream_count == 1 &&
 			dbg->force_single_disp_pipe_split)
-		hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_khz);
+		hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_100hz);
 }
 
 bool dcn_validate_bandwidth(
@@ -852,7 +845,7 @@ bool dcn_validate_bandwidth(
 		v->v_sync_plus_back_porch[input_idx] = pipe->stream->timing.v_total
 				- v->vactive[input_idx]
 				- pipe->stream->timing.v_front_porch;
-		v->pixel_clock[input_idx] = pipe->stream->timing.pix_clk_khz/1000.0;
+		v->pixel_clock[input_idx] = pipe->stream->timing.pix_clk_100hz/10000.0;
 		if (pipe->stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
 			v->pixel_clock[input_idx] *= 2;
 		if (!pipe->plane_state) {
@@ -961,7 +954,7 @@ bool dcn_validate_bandwidth(
 		v->dcc_rate[input_idx] = 1; /*TODO: Worst case? does this change?*/
 		v->output_format[input_idx] = pipe->stream->timing.pixel_encoding ==
 				PIXEL_ENCODING_YCBCR420 ? dcn_bw_420 : dcn_bw_444;
-		v->output[input_idx] = pipe->stream->sink->sink_signal ==
+		v->output[input_idx] = pipe->stream->signal ==
 				SIGNAL_TYPE_HDMI_TYPE_A ? dcn_bw_hdmi : dcn_bw_dp;
 		v->output_deep_color[input_idx] = dcn_bw_encoder_8bpc;
 		if (v->output[input_idx] == dcn_bw_hdmi) {
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 5fd52094d459..52f838442e21 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -384,7 +384,7 @@ void dc_stream_set_dither_option(struct dc_stream_state *stream,
 		enum dc_dither_option option)
 {
 	struct bit_depth_reduction_params params;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct pipe_ctx *pipes = NULL;
 	int i;
 
@@ -451,7 +451,7 @@ bool dc_stream_program_csc_matrix(struct dc *dc, struct dc_stream_state *stream)
 					pipes,
 					stream->output_color_space,
 					stream->csc_color_matrix.matrix,
-					pipes->plane_res.hubp->opp_id);
+					pipes->plane_res.hubp ? pipes->plane_res.hubp->opp_id : 0);
 			ret = true;
 		}
 	}
@@ -526,9 +526,8 @@ void dc_link_set_preferred_link_settings(struct dc *dc,
 
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
-		if (pipe->stream && pipe->stream->sink
-			&& pipe->stream->sink->link) {
-			if (pipe->stream->sink->link == link)
+		if (pipe->stream && pipe->stream->link) {
+			if (pipe->stream->link == link)
 				break;
 		}
 	}
@@ -586,9 +585,6 @@ static void destruct(struct dc *dc)
 	if (dc->ctx->gpio_service)
 		dal_gpio_service_destroy(&dc->ctx->gpio_service);
 
-	if (dc->ctx->i2caux)
-		dal_i2caux_destroy(&dc->ctx->i2caux);
-
 	if (dc->ctx->created_bios)
 		dal_bios_parser_destroy(&dc->ctx->dc_bios);
 
@@ -625,7 +621,6 @@ static bool construct(struct dc *dc,
 #endif
 
 	enum dce_version dc_version = DCE_VERSION_UNKNOWN;
-
 	dc_dceip = kzalloc(sizeof(*dc_dceip), GFP_KERNEL);
 	if (!dc_dceip) {
 		dm_error("%s: failed to create dceip\n", __func__);
@@ -670,6 +665,7 @@ static bool construct(struct dc *dc,
 	dc_ctx->dc = dc;
 	dc_ctx->asic_id = init_params->asic_id;
 	dc_ctx->dc_sink_id_count = 0;
+	dc_ctx->dc_stream_id_count = 0;
 	dc->ctx = dc_ctx;
 
 	dc->current_state = dc_create_state();
@@ -709,14 +705,6 @@ static bool construct(struct dc *dc,
 		dc_ctx->created_bios = true;
 		}
 
-	/* Create I2C AUX */
-	dc_ctx->i2caux = dal_i2caux_create(dc_ctx);
-
-	if (!dc_ctx->i2caux) {
-		ASSERT_CRITICAL(false);
-		goto fail;
-	}
-
 	dc_ctx->perf_trace = dc_perf_trace_create();
 	if (!dc_ctx->perf_trace) {
 		ASSERT_CRITICAL(false);
@@ -840,6 +828,11 @@ alloc_fail:
 	return NULL;
 }
 
+void dc_init_callbacks(struct dc *dc,
+		const struct dc_callback_init *init_params)
+{
+}
+
 void dc_destroy(struct dc **dc)
 {
 	destruct(*dc);
@@ -875,8 +868,9 @@ static void program_timing_sync(
 		struct dc *dc,
 		struct dc_state *ctx)
 {
-	int i, j;
+	int i, j, k;
 	int group_index = 0;
+	int num_group = 0;
 	int pipe_count = dc->res_pool->pipe_count;
 	struct pipe_ctx *unsynced_pipes[MAX_PIPES] = { NULL };
 
@@ -913,11 +907,11 @@ static void program_timing_sync(
 			}
 		}
 
-		/* set first unblanked pipe as master */
+		/* set first pipe with plane as master */
 		for (j = 0; j < group_size; j++) {
 			struct pipe_ctx *temp;
 
-			if (pipe_set[j]->stream_res.tg->funcs->is_blanked && !pipe_set[j]->stream_res.tg->funcs->is_blanked(pipe_set[j]->stream_res.tg)) {
+			if (pipe_set[j]->plane_state) {
 				if (j == 0)
 					break;
 
@@ -928,9 +922,21 @@ static void program_timing_sync(
 			}
 		}
 
-		/* remove any other unblanked pipes as they have already been synced */
+
+		for (k = 0; k < group_size; k++) {
+			struct dc_stream_status *status = dc_stream_get_status_from_state(ctx, pipe_set[k]->stream);
+
+			status->timing_sync_info.group_id = num_group;
+			status->timing_sync_info.group_size = group_size;
+			if (k == 0)
+				status->timing_sync_info.master = true;
+			else
+				status->timing_sync_info.master = false;
+
+		}
+		/* remove any other pipes with plane as they have already been synced */
 		for (j = j + 1; j < group_size; j++) {
-			if (pipe_set[j]->stream_res.tg->funcs->is_blanked && !pipe_set[j]->stream_res.tg->funcs->is_blanked(pipe_set[j]->stream_res.tg)) {
+			if (pipe_set[j]->plane_state) {
 				group_size--;
 				pipe_set[j] = pipe_set[group_size];
 				j--;
@@ -942,6 +948,7 @@ static void program_timing_sync(
 				dc, group_index, group_size, pipe_set);
 			group_index++;
 		}
+		num_group++;
 	}
 }
 
@@ -962,6 +969,52 @@ static bool context_changed(
 	return false;
 }
 
+bool dc_validate_seamless_boot_timing(struct dc *dc,
+				const struct dc_sink *sink,
+				struct dc_crtc_timing *crtc_timing)
+{
+	struct timing_generator *tg;
+	struct dc_link *link = sink->link;
+	unsigned int inst;
+
+	/* Check for enabled DIG to identify enabled display */
+	if (!link->link_enc->funcs->is_dig_enabled(link->link_enc))
+		return false;
+
+	/* Check for which front end is used by this encoder.
+	 * Note the inst is 1 indexed, where 0 is undefined.
+	 * Note that DIG_FE can source from different OTG but our
+	 * current implementation always map 1-to-1, so this code makes
+	 * the same assumption and doesn't check OTG source.
+	 */
+	inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1;
+
+	/* Instance should be within the range of the pool */
+	if (inst >= dc->res_pool->pipe_count)
+		return false;
+
+	tg = dc->res_pool->timing_generators[inst];
+
+	if (!tg->funcs->is_matching_timing)
+		return false;
+
+	if (!tg->funcs->is_matching_timing(tg, crtc_timing))
+		return false;
+
+	if (dc_is_dp_signal(link->connector_signal)) {
+		unsigned int pix_clk_100hz;
+
+		dc->res_pool->dp_clock_source->funcs->get_pixel_clk_frequency_100hz(
+			dc->res_pool->dp_clock_source,
+			inst, &pix_clk_100hz);
+
+		if (crtc_timing->pix_clk_100hz != pix_clk_100hz)
+			return false;
+	}
+
+	return true;
+}
+
 bool dc_enable_stereo(
 	struct dc *dc,
 	struct dc_state *context,
@@ -1040,7 +1093,11 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 
 	/* Program all planes within new context*/
 	for (i = 0; i < context->stream_count; i++) {
-		const struct dc_sink *sink = context->streams[i]->sink;
+		const struct dc_link *link = context->streams[i]->link;
+		struct dc_stream_status *status;
+
+		if (context->streams[i]->apply_seamless_boot_optimization)
+			context->streams[i]->apply_seamless_boot_optimization = false;
 
 		if (!context->streams[i]->mode_changed)
 			continue;
@@ -1065,12 +1122,15 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 			}
 		}
 
-		CONN_MSG_MODE(sink->link, "{%dx%d, %dx%d@%dKhz}",
+		status = dc_stream_get_status_from_state(context, context->streams[i]);
+		context->streams[i]->out.otg_offset = status->primary_otg_inst;
+
+		CONN_MSG_MODE(link, "{%dx%d, %dx%d@%dKhz}",
 				context->streams[i]->timing.h_addressable,
 				context->streams[i]->timing.v_addressable,
 				context->streams[i]->timing.h_total,
 				context->streams[i]->timing.v_total,
-				context->streams[i]->timing.pix_clk_khz);
+				context->streams[i]->timing.pix_clk_100hz / 10);
 	}
 
 	dc_enable_stereo(dc, context, dc_streams, context->stream_count);
@@ -1114,6 +1174,9 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc)
 	int i;
 	struct dc_state *context = dc->current_state;
 
+	if (dc->optimized_required == false)
+		return true;
+
 	post_surface_trace(dc);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++)
@@ -1215,6 +1278,12 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa
 		 */
 		update_flags->bits.bpp_change = 1;
 
+	if (u->plane_info->plane_size.grph.surface_pitch != u->surface->plane_size.grph.surface_pitch
+			|| u->plane_info->plane_size.video.luma_pitch != u->surface->plane_size.video.luma_pitch
+			|| u->plane_info->plane_size.video.chroma_pitch != u->surface->plane_size.video.chroma_pitch)
+		update_flags->bits.plane_size_change = 1;
+
+
 	if (memcmp(&u->plane_info->tiling_info, &u->surface->tiling_info,
 			sizeof(union dc_tiling_info)) != 0) {
 		update_flags->bits.swizzle_change = 1;
@@ -1236,7 +1305,7 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa
 			|| update_flags->bits.output_tf_change)
 		return UPDATE_TYPE_FULL;
 
-	return UPDATE_TYPE_MED;
+	return update_flags->raw ? UPDATE_TYPE_MED : UPDATE_TYPE_FAST;
 }
 
 static enum surface_update_type get_scaling_info_update_type(
@@ -1436,6 +1505,101 @@ static struct dc_stream_status *stream_get_status(
 
 static const enum surface_update_type update_surface_trace_level = UPDATE_TYPE_FULL;
 
+static void copy_surface_update_to_plane(
+		struct dc_plane_state *surface,
+		struct dc_surface_update *srf_update)
+{
+	if (srf_update->flip_addr) {
+		surface->address = srf_update->flip_addr->address;
+		surface->flip_immediate =
+			srf_update->flip_addr->flip_immediate;
+		surface->time.time_elapsed_in_us[surface->time.index] =
+			srf_update->flip_addr->flip_timestamp_in_us -
+				surface->time.prev_update_time_in_us;
+		surface->time.prev_update_time_in_us =
+			srf_update->flip_addr->flip_timestamp_in_us;
+		surface->time.index++;
+		if (surface->time.index >= DC_PLANE_UPDATE_TIMES_MAX)
+			surface->time.index = 0;
+	}
+
+	if (srf_update->scaling_info) {
+		surface->scaling_quality =
+				srf_update->scaling_info->scaling_quality;
+		surface->dst_rect =
+				srf_update->scaling_info->dst_rect;
+		surface->src_rect =
+				srf_update->scaling_info->src_rect;
+		surface->clip_rect =
+				srf_update->scaling_info->clip_rect;
+	}
+
+	if (srf_update->plane_info) {
+		surface->color_space =
+				srf_update->plane_info->color_space;
+		surface->format =
+				srf_update->plane_info->format;
+		surface->plane_size =
+				srf_update->plane_info->plane_size;
+		surface->rotation =
+				srf_update->plane_info->rotation;
+		surface->horizontal_mirror =
+				srf_update->plane_info->horizontal_mirror;
+		surface->stereo_format =
+				srf_update->plane_info->stereo_format;
+		surface->tiling_info =
+				srf_update->plane_info->tiling_info;
+		surface->visible =
+				srf_update->plane_info->visible;
+		surface->per_pixel_alpha =
+				srf_update->plane_info->per_pixel_alpha;
+		surface->global_alpha =
+				srf_update->plane_info->global_alpha;
+		surface->global_alpha_value =
+				srf_update->plane_info->global_alpha_value;
+		surface->dcc =
+				srf_update->plane_info->dcc;
+		surface->sdr_white_level =
+				srf_update->plane_info->sdr_white_level;
+	}
+
+	if (srf_update->gamma &&
+			(surface->gamma_correction !=
+					srf_update->gamma)) {
+		memcpy(&surface->gamma_correction->entries,
+			&srf_update->gamma->entries,
+			sizeof(struct dc_gamma_entries));
+		surface->gamma_correction->is_identity =
+			srf_update->gamma->is_identity;
+		surface->gamma_correction->num_entries =
+			srf_update->gamma->num_entries;
+		surface->gamma_correction->type =
+			srf_update->gamma->type;
+	}
+
+	if (srf_update->in_transfer_func &&
+			(surface->in_transfer_func !=
+				srf_update->in_transfer_func)) {
+		surface->in_transfer_func->sdr_ref_white_level =
+			srf_update->in_transfer_func->sdr_ref_white_level;
+		surface->in_transfer_func->tf =
+			srf_update->in_transfer_func->tf;
+		surface->in_transfer_func->type =
+			srf_update->in_transfer_func->type;
+		memcpy(&surface->in_transfer_func->tf_pts,
+			&srf_update->in_transfer_func->tf_pts,
+			sizeof(struct dc_transfer_func_distributed_points));
+	}
+
+	if (srf_update->input_csc_color_matrix)
+		surface->input_csc_color_matrix =
+			*srf_update->input_csc_color_matrix;
+
+	if (srf_update->coeff_reduction_factor)
+		surface->coeff_reduction_factor =
+			*srf_update->coeff_reduction_factor;
+}
+
 static void commit_planes_do_stream_update(struct dc *dc,
 		struct dc_stream_state *stream,
 		struct dc_stream_update *stream_update,
@@ -1459,11 +1623,13 @@ static void commit_planes_do_stream_update(struct dc *dc,
 					stream_update->adjust->v_total_min,
 					stream_update->adjust->v_total_max);
 
-			if (stream_update->periodic_fn_vsync_delta &&
-					pipe_ctx->stream_res.tg->funcs->program_vline_interrupt)
+			if (stream_update->periodic_vsync_config && pipe_ctx->stream_res.tg->funcs->program_vline_interrupt)
 				pipe_ctx->stream_res.tg->funcs->program_vline_interrupt(
-					pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing,
-					pipe_ctx->stream->periodic_fn_vsync_delta);
+					pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing, VLINE0, &stream->periodic_vsync_config);
+
+			if (stream_update->enhanced_sync_config && pipe_ctx->stream_res.tg->funcs->program_vline_interrupt)
+				pipe_ctx->stream_res.tg->funcs->program_vline_interrupt(
+					pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing, VLINE1, &stream->enhanced_sync_config);
 
 			if ((stream_update->hdr_static_metadata && !stream->use_dynamic_meta) ||
 					stream_update->vrr_infopacket ||
@@ -1605,7 +1771,6 @@ void dc_commit_updates_for_stream(struct dc *dc,
 		int surface_count,
 		struct dc_stream_state *stream,
 		struct dc_stream_update *stream_update,
-		struct dc_plane_state **plane_states,
 		struct dc_state *state)
 {
 	const struct dc_stream_status *stream_status;
@@ -1640,14 +1805,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 	for (i = 0; i < surface_count; i++) {
 		struct dc_plane_state *surface = srf_updates[i].surface;
 
-		/* TODO: On flip we don't build the state, so it still has the
-		 * old address. Which is why we are updating the address here
-		 */
-		if (srf_updates[i].flip_addr) {
-			surface->address = srf_updates[i].flip_addr->address;
-			surface->flip_immediate = srf_updates[i].flip_addr->flip_immediate;
-
-		}
+		copy_surface_update_to_plane(surface, &srf_updates[i]);
 
 		if (update_type >= UPDATE_TYPE_MED) {
 			for (j = 0; j < dc->res_pool->pipe_count; j++) {
@@ -1764,6 +1922,26 @@ void dc_resume(struct dc *dc)
 		core_link_resume(dc->links[i]);
 }
 
+unsigned int dc_get_current_backlight_pwm(struct dc *dc)
+{
+	struct abm *abm = dc->res_pool->abm;
+
+	if (abm)
+		return abm->funcs->get_current_backlight(abm);
+
+	return 0;
+}
+
+unsigned int dc_get_target_backlight_pwm(struct dc *dc)
+{
+	struct abm *abm = dc->res_pool->abm;
+
+	if (abm)
+		return abm->funcs->get_target_backlight(abm);
+
+	return 0;
+}
+
 bool dc_is_dmcu_initialized(struct dc *dc)
 {
 	struct dmcu *dmcu = dc->res_pool->dmcu;
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 52deacf39841..7f5a947ad31d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -43,10 +43,6 @@
 #include "dpcd_defs.h"
 #include "dmcu.h"
 
-#include "dce/dce_11_0_d.h"
-#include "dce/dce_11_0_enum.h"
-#include "dce/dce_11_0_sh_mask.h"
-
 #define DC_LOGGER_INIT(logger)
 
 
@@ -80,6 +76,12 @@ static void destruct(struct dc_link *link)
 {
 	int i;
 
+	if (link->hpd_gpio != NULL) {
+		dal_gpio_close(link->hpd_gpio);
+		dal_gpio_destroy_irq(&link->hpd_gpio);
+		link->hpd_gpio = NULL;
+	}
+
 	if (link->ddc)
 		dal_ddc_service_destroy(&link->ddc);
 
@@ -789,7 +791,7 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 			return false;
 		}
 
-		sink->dongle_max_pix_clk = sink_caps.max_hdmi_pixel_clock;
+		sink->link->dongle_max_pix_clk = sink_caps.max_hdmi_pixel_clock;
 		sink->converter_disable_audio = converter_disable_audio;
 
 		link->local_sink = sink;
@@ -935,18 +937,11 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 
 bool dc_link_get_hpd_state(struct dc_link *dc_link)
 {
-	struct gpio *hpd_pin;
 	uint32_t state;
 
-	hpd_pin = get_hpd_gpio(dc_link->ctx->dc_bios,
-					dc_link->link_id, dc_link->ctx->gpio_service);
-	if (hpd_pin == NULL)
-		ASSERT(false);
-
-	dal_gpio_open(hpd_pin, GPIO_MODE_INTERRUPT);
-	dal_gpio_get_value(hpd_pin, &state);
-	dal_gpio_close(hpd_pin);
-	dal_gpio_destroy_irq(&hpd_pin);
+	dal_gpio_lock_pin(dc_link->hpd_gpio);
+	dal_gpio_get_value(dc_link->hpd_gpio, &state);
+	dal_gpio_unlock_pin(dc_link->hpd_gpio);
 
 	return state;
 }
@@ -1102,7 +1097,6 @@ static bool construct(
 	const struct link_init_data *init_params)
 {
 	uint8_t i;
-	struct gpio *hpd_gpio = NULL;
 	struct ddc_service_init_data ddc_service_init_data = { { 0 } };
 	struct dc_context *dc_ctx = init_params->ctx;
 	struct encoder_init_data enc_init_data = { 0 };
@@ -1132,10 +1126,12 @@ static bool construct(
 	if (link->dc->res_pool->funcs->link_init)
 		link->dc->res_pool->funcs->link_init(link);
 
-	hpd_gpio = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service);
-
-	if (hpd_gpio != NULL)
-		link->irq_source_hpd = dal_irq_get_source(hpd_gpio);
+	link->hpd_gpio = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service);
+	if (link->hpd_gpio != NULL) {
+		dal_gpio_open(link->hpd_gpio, GPIO_MODE_INTERRUPT);
+		dal_gpio_unlock_pin(link->hpd_gpio);
+		link->irq_source_hpd = dal_irq_get_source(link->hpd_gpio);
+	}
 
 	switch (link->link_id.id) {
 	case CONNECTOR_ID_HDMI_TYPE_A:
@@ -1153,18 +1149,18 @@ static bool construct(
 	case CONNECTOR_ID_DISPLAY_PORT:
 		link->connector_signal =	SIGNAL_TYPE_DISPLAY_PORT;
 
-		if (hpd_gpio != NULL)
+		if (link->hpd_gpio != NULL)
 			link->irq_source_hpd_rx =
-					dal_irq_get_rx_source(hpd_gpio);
+					dal_irq_get_rx_source(link->hpd_gpio);
 
 		break;
 	case CONNECTOR_ID_EDP:
 		link->connector_signal = SIGNAL_TYPE_EDP;
 
-		if (hpd_gpio != NULL) {
+		if (link->hpd_gpio != NULL) {
 			link->irq_source_hpd = DC_IRQ_SOURCE_INVALID;
 			link->irq_source_hpd_rx =
-					dal_irq_get_rx_source(hpd_gpio);
+					dal_irq_get_rx_source(link->hpd_gpio);
 		}
 		break;
 	case CONNECTOR_ID_LVDS:
@@ -1175,10 +1171,7 @@ static bool construct(
 		goto create_fail;
 	}
 
-	if (hpd_gpio != NULL) {
-		dal_gpio_destroy_irq(&hpd_gpio);
-		hpd_gpio = NULL;
-	}
+
 
 	/* TODO: #DAL3 Implement id to str function.*/
 	LINK_INFO("Connector[%d] description:"
@@ -1281,8 +1274,9 @@ link_enc_create_fail:
 ddc_create_fail:
 create_fail:
 
-	if (hpd_gpio != NULL) {
-		dal_gpio_destroy_irq(&hpd_gpio);
+	if (link->hpd_gpio != NULL) {
+		dal_gpio_destroy_irq(&link->hpd_gpio);
+		link->hpd_gpio = NULL;
 	}
 
 	return false;
@@ -1372,7 +1366,7 @@ static void dpcd_configure_panel_mode(
 static void enable_stream_features(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	union down_spread_ctrl old_downspread;
 	union down_spread_ctrl new_downspread;
 
@@ -1397,7 +1391,7 @@ static enum dc_status enable_link_dp(
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	enum dc_status status;
 	bool skip_video_pattern;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct dc_link_settings link_settings = {0};
 	enum dp_panel_mode panel_mode;
 
@@ -1414,8 +1408,8 @@ static enum dc_status enable_link_dp(
 		pipe_ctx->clock_source->id,
 		&link_settings);
 
-	if (stream->sink->edid_caps.panel_patch.dppowerup_delay > 0) {
-		int delay_dp_power_up_in_ms = stream->sink->edid_caps.panel_patch.dppowerup_delay;
+	if (stream->sink_patches.dppowerup_delay > 0) {
+		int delay_dp_power_up_in_ms = stream->sink_patches.dppowerup_delay;
 
 		msleep(delay_dp_power_up_in_ms);
 	}
@@ -1448,7 +1442,7 @@ static enum dc_status enable_link_edp(
 {
 	enum dc_status status;
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	/*in case it is not on*/
 	link->dc->hwss.edp_power_control(link, true);
 	link->dc->hwss.edp_wait_for_hpd_ready(link, true);
@@ -1463,7 +1457,7 @@ static enum dc_status enable_link_dp_mst(
 		struct dc_state *state,
 		struct pipe_ctx *pipe_ctx)
 {
-	struct dc_link *link = pipe_ctx->stream->sink->link;
+	struct dc_link *link = pipe_ctx->stream->link;
 
 	/* sink signal type after MST branch is MST. Multiple MST sinks
 	 * share one link. Link DP PHY is enable or training only once.
@@ -1471,6 +1465,11 @@ static enum dc_status enable_link_dp_mst(
 	if (link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN)
 		return DC_OK;
 
+	/* to make sure the pending down rep can be processed
+	 * before clear payload table
+	 */
+	dm_helpers_dp_mst_poll_pending_down_reply(link->ctx, link);
+
 	/* clear payload table */
 	dm_helpers_dp_mst_clear_payload_allocation_table(link->ctx, link);
 
@@ -1597,7 +1596,7 @@ static bool i2c_write(struct pipe_ctx *pipe_ctx,
 	cmd.payloads = &payload;
 
 	if (dm_helpers_submit_i2c(pipe_ctx->stream->ctx,
-			pipe_ctx->stream->sink->link, &cmd))
+			pipe_ctx->stream->link, &cmd))
 		return true;
 
 	return false;
@@ -1651,7 +1650,7 @@ static void write_i2c_retimer_setting(
 				else {
 					i2c_success =
 						dal_ddc_service_query_ddc_data(
-						pipe_ctx->stream->sink->link->ddc,
+						pipe_ctx->stream->link->ddc,
 						slave_address, &offset, 1, &value, 1);
 					if (!i2c_success)
 						/* Write failure */
@@ -1704,7 +1703,7 @@ static void write_i2c_retimer_setting(
 					else {
 						i2c_success =
 								dal_ddc_service_query_ddc_data(
-								pipe_ctx->stream->sink->link->ddc,
+								pipe_ctx->stream->link->ddc,
 								slave_address, &offset, 1, &value, 1);
 						if (!i2c_success)
 							/* Write failure */
@@ -1929,7 +1928,7 @@ static void write_i2c_redriver_setting(
 static void enable_link_hdmi(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	enum dc_color_depth display_color_depth;
 	enum engine_id eng_id;
 	struct ext_hdmi_settings settings = {0};
@@ -1938,12 +1937,12 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx)
 			&& (stream->timing.v_addressable == 480);
 
 	if (stream->phy_pix_clk == 0)
-		stream->phy_pix_clk = stream->timing.pix_clk_khz;
+		stream->phy_pix_clk = stream->timing.pix_clk_100hz / 10;
 	if (stream->phy_pix_clk > 340000)
 		is_over_340mhz = true;
 
 	if (dc_is_hdmi_signal(pipe_ctx->stream->signal)) {
-		unsigned short masked_chip_caps = pipe_ctx->stream->sink->link->chip_caps &
+		unsigned short masked_chip_caps = pipe_ctx->stream->link->chip_caps &
 				EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK;
 		if (masked_chip_caps == EXT_DISPLAY_PATH_CAPS__HDMI20_TISN65DP159RSBT) {
 			/* DP159, Retimer settings */
@@ -1964,11 +1963,11 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx)
 
 	if (dc_is_hdmi_signal(pipe_ctx->stream->signal))
 		dal_ddc_service_write_scdc_data(
-			stream->sink->link->ddc,
+			stream->link->ddc,
 			stream->phy_pix_clk,
 			stream->timing.flags.LTE_340MCSC_SCRAMBLE);
 
-	memset(&stream->sink->link->cur_link_settings, 0,
+	memset(&stream->link->cur_link_settings, 0,
 			sizeof(struct dc_link_settings));
 
 	display_color_depth = stream->timing.display_color_depth;
@@ -1989,12 +1988,12 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx)
 static void enable_link_lvds(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 
 	if (stream->phy_pix_clk == 0)
-		stream->phy_pix_clk = stream->timing.pix_clk_khz;
+		stream->phy_pix_clk = stream->timing.pix_clk_100hz / 10;
 
-	memset(&stream->sink->link->cur_link_settings, 0,
+	memset(&stream->link->cur_link_settings, 0,
 			sizeof(struct dc_link_settings));
 
 	link->link_enc->funcs->enable_lvds_output(
@@ -2067,7 +2066,7 @@ static bool dp_active_dongle_validate_timing(
 		const struct dc_crtc_timing *timing,
 		const struct dpcd_caps *dpcd_caps)
 {
-	unsigned int required_pix_clk = timing->pix_clk_khz;
+	unsigned int required_pix_clk_100hz = timing->pix_clk_100hz;
 	const struct dc_dongle_caps *dongle_caps = &dpcd_caps->dongle_caps;
 
 	switch (dpcd_caps->dongle_type) {
@@ -2107,9 +2106,9 @@ static bool dp_active_dongle_validate_timing(
 
 	/* Check Color Depth and Pixel Clock */
 	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		required_pix_clk /= 2;
+		required_pix_clk_100hz /= 2;
 	else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-		required_pix_clk = required_pix_clk * 2 / 3;
+		required_pix_clk_100hz = required_pix_clk_100hz * 2 / 3;
 
 	switch (timing->display_color_depth) {
 	case COLOR_DEPTH_666:
@@ -2119,12 +2118,12 @@ static bool dp_active_dongle_validate_timing(
 	case COLOR_DEPTH_101010:
 		if (dongle_caps->dp_hdmi_max_bpc < 10)
 			return false;
-		required_pix_clk = required_pix_clk * 10 / 8;
+		required_pix_clk_100hz = required_pix_clk_100hz * 10 / 8;
 		break;
 	case COLOR_DEPTH_121212:
 		if (dongle_caps->dp_hdmi_max_bpc < 12)
 			return false;
-		required_pix_clk = required_pix_clk * 12 / 8;
+		required_pix_clk_100hz = required_pix_clk_100hz * 12 / 8;
 		break;
 
 	case COLOR_DEPTH_141414:
@@ -2134,7 +2133,7 @@ static bool dp_active_dongle_validate_timing(
 		return false;
 	}
 
-	if (required_pix_clk > dongle_caps->dp_hdmi_max_pixel_clk)
+	if (required_pix_clk_100hz > (dongle_caps->dp_hdmi_max_pixel_clk * 10))
 		return false;
 
 	return true;
@@ -2145,7 +2144,7 @@ enum dc_status dc_link_validate_mode_timing(
 		struct dc_link *link,
 		const struct dc_crtc_timing *timing)
 {
-	uint32_t max_pix_clk = stream->sink->dongle_max_pix_clk;
+	uint32_t max_pix_clk = stream->link->dongle_max_pix_clk * 10;
 	struct dpcd_caps *dpcd_caps = &link->dpcd_caps;
 
 	/* A hack to avoid failing any modes for EDID override feature on
@@ -2155,7 +2154,7 @@ enum dc_status dc_link_validate_mode_timing(
 		return DC_OK;
 
 	/* Passive Dongle */
-	if (0 != max_pix_clk && timing->pix_clk_khz > max_pix_clk)
+	if (0 != max_pix_clk && timing->pix_clk_100hz > max_pix_clk)
 		return DC_EXCEED_DONGLE_CAP;
 
 	/* Active Dongle*/
@@ -2190,8 +2189,7 @@ int dc_link_get_backlight_level(const struct dc_link *link)
 
 bool dc_link_set_backlight_level(const struct dc_link *link,
 		uint32_t backlight_pwm_u16_16,
-		uint32_t frame_ramp,
-		const struct dc_stream_state *stream)
+		uint32_t frame_ramp)
 {
 	struct dc  *core_dc = link->ctx->dc;
 	struct abm *abm = core_dc->res_pool->abm;
@@ -2206,10 +2204,6 @@ bool dc_link_set_backlight_level(const struct dc_link *link,
 		(abm->funcs->set_backlight_level_pwm == NULL))
 		return false;
 
-	if (stream)
-		((struct dc_stream_state *)stream)->bl_pwm_level =
-				backlight_pwm_u16_16;
-
 	use_smooth_brightness = dmcu->funcs->is_dmcu_initialized(dmcu);
 
 	DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n",
@@ -2219,7 +2213,7 @@ bool dc_link_set_backlight_level(const struct dc_link *link,
 		for (i = 0; i < MAX_PIPES; i++) {
 			if (core_dc->current_state->res_ctx.pipe_ctx[i].stream) {
 				if (core_dc->current_state->res_ctx.
-						pipe_ctx[i].stream->sink->link
+						pipe_ctx[i].stream->link
 						== link)
 					/* DMCU -1 for all controller id values,
 					 * therefore +1 here
@@ -2279,7 +2273,7 @@ void core_link_resume(struct dc_link *link)
 static struct fixed31_32 get_pbn_per_slot(struct dc_stream_state *stream)
 {
 	struct dc_link_settings *link_settings =
-			&stream->sink->link->cur_link_settings;
+			&stream->link->cur_link_settings;
 	uint32_t link_rate_in_mbps =
 			link_settings->link_rate * LINK_RATE_REF_FREQ_IN_MHZ;
 	struct fixed31_32 mbps = dc_fixpt_from_int(
@@ -2310,7 +2304,7 @@ static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx)
 	uint32_t denominator;
 
 	bpc = get_color_depth(pipe_ctx->stream_res.pix_clk_params.color_depth);
-	kbps = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk * bpc * 3;
+	kbps = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10 * bpc * 3;
 
 	/*
 	 * margin 5300ppm + 300ppm ~ 0.6% as per spec, factor is 1.006
@@ -2386,7 +2380,7 @@ static void update_mst_stream_alloc_table(
 static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct link_encoder *link_encoder = link->link_enc;
 	struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc;
 	struct dp_mst_stream_allocation_table proposed_table = {0};
@@ -2466,7 +2460,7 @@ static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx)
 static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct link_encoder *link_encoder = link->link_enc;
 	struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc;
 	struct dp_mst_stream_allocation_table proposed_table = {0};
@@ -2551,8 +2545,8 @@ void core_link_enable_stream(
 	DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger);
 
 	if (pipe_ctx->stream->signal != SIGNAL_TYPE_VIRTUAL) {
-		stream->sink->link->link_enc->funcs->setup(
-			stream->sink->link->link_enc,
+		stream->link->link_enc->funcs->setup(
+			stream->link->link_enc,
 			pipe_ctx->stream->signal);
 		pipe_ctx->stream_res.stream_enc->funcs->setup_stereo_sync(
 			pipe_ctx->stream_res.stream_enc,
@@ -2586,13 +2580,23 @@ void core_link_enable_stream(
 			&stream->timing);
 
 	if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) {
+		bool apply_edp_fast_boot_optimization =
+			pipe_ctx->stream->apply_edp_fast_boot_optimization;
+
+		pipe_ctx->stream->apply_edp_fast_boot_optimization = false;
+
 		resource_build_info_frame(pipe_ctx);
 		core_dc->hwss.update_info_frame(pipe_ctx);
 
+		/* Do not touch link on seamless boot optimization. */
+		if (pipe_ctx->stream->apply_seamless_boot_optimization) {
+			pipe_ctx->stream->dpms_off = false;
+			return;
+		}
+
 		/* eDP lit up by bios already, no need to enable again. */
 		if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP &&
-				pipe_ctx->stream->apply_edp_fast_boot_optimization) {
-			pipe_ctx->stream->apply_edp_fast_boot_optimization = false;
+					apply_edp_fast_boot_optimization) {
 			pipe_ctx->stream->dpms_off = false;
 			return;
 		}
@@ -2604,7 +2608,7 @@ void core_link_enable_stream(
 
 		if (status != DC_OK) {
 			DC_LOG_WARNING("enabling link %u failed: %d\n",
-			pipe_ctx->stream->sink->link->link_index,
+			pipe_ctx->stream->link->link_index,
 			status);
 
 			/* Abort stream enable *unless* the failure was due to
@@ -2619,6 +2623,8 @@ void core_link_enable_stream(
 			}
 		}
 
+		stream->link->link_status.link_active = true;
+
 		core_dc->hwss.enable_audio_stream(pipe_ctx);
 
 		/* turn off otg test pattern if enable */
@@ -2633,15 +2639,10 @@ void core_link_enable_stream(
 			allocate_mst_payload(pipe_ctx);
 
 		core_dc->hwss.unblank_stream(pipe_ctx,
-			&pipe_ctx->stream->sink->link->cur_link_settings);
+			&pipe_ctx->stream->link->cur_link_settings);
 
 		if (dc_is_dp_signal(pipe_ctx->stream->signal))
 			enable_stream_features(pipe_ctx);
-
-		dc_link_set_backlight_level(pipe_ctx->stream->sink->link,
-				pipe_ctx->stream->bl_pwm_level,
-				0,
-				pipe_ctx->stream);
 	}
 
 }
@@ -2657,7 +2658,9 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option)
 
 	core_dc->hwss.disable_stream(pipe_ctx, option);
 
-	disable_link(pipe_ctx->stream->sink->link, pipe_ctx->stream->signal);
+	disable_link(pipe_ctx->stream->link, pipe_ctx->stream->signal);
+
+	pipe_ctx->stream->link->link_status.link_active = false;
 }
 
 void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable)
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 506a97e16956..b7ee63cd8dc7 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
@@ -33,7 +33,7 @@
 #include "include/vector.h"
 #include "core_types.h"
 #include "dc_link_ddc.h"
-#include "aux_engine.h"
+#include "dce/dce_aux.h"
 
 #define AUX_POWER_UP_WA_DELAY 500
 #define I2C_OVER_AUX_DEFER_WA_DELAY 70
@@ -42,7 +42,6 @@
 #define CV_SMART_DONGLE_ADDRESS 0x20
 /* DVI-HDMI dongle slave address for retrieving dongle signature*/
 #define DVI_HDMI_DONGLE_ADDRESS 0x68
-static const int8_t dvi_hdmi_dongle_signature_str[] = "6140063500G";
 struct dvi_hdmi_dongle_signature_data {
 	int8_t vendor[3];/* "AMD" */
 	uint8_t version[2];
@@ -165,43 +164,6 @@ static void dal_ddc_i2c_payloads_destroy(struct i2c_payloads **p)
 
 }
 
-static struct aux_payloads *dal_ddc_aux_payloads_create(struct dc_context *ctx, uint32_t count)
-{
-	struct aux_payloads *payloads;
-
-	payloads = kzalloc(sizeof(struct aux_payloads), GFP_KERNEL);
-
-	if (!payloads)
-		return NULL;
-
-	if (dal_vector_construct(
-		&payloads->payloads, ctx, count, sizeof(struct aux_payload)))
-		return payloads;
-
-	kfree(payloads);
-	return NULL;
-}
-
-static struct aux_payload *dal_ddc_aux_payloads_get(struct aux_payloads *p)
-{
-	return (struct aux_payload *)p->payloads.container;
-}
-
-static uint32_t  dal_ddc_aux_payloads_get_count(struct aux_payloads *p)
-{
-	return p->payloads.count;
-}
-
-static void dal_ddc_aux_payloads_destroy(struct aux_payloads **p)
-{
-	if (!p || !*p)
-		return;
-
-	dal_vector_destruct(&(*p)->payloads);
-	kfree(*p);
-	*p = NULL;
-}
-
 #define DDC_MIN(a, b) (((a) < (b)) ? (a) : (b))
 
 void dal_ddc_i2c_payloads_add(
@@ -225,27 +187,6 @@ void dal_ddc_i2c_payloads_add(
 
 }
 
-void dal_ddc_aux_payloads_add(
-	struct aux_payloads *payloads,
-	uint32_t address,
-	uint32_t len,
-	uint8_t *data,
-	bool write)
-{
-	uint32_t payload_size = DEFAULT_AUX_MAX_DATA_SIZE;
-	uint32_t pos;
-
-	for (pos = 0; pos < len; pos += payload_size) {
-		struct aux_payload payload = {
-			.i2c_over_aux = true,
-			.write = write,
-			.address = address,
-			.length = DDC_MIN(payload_size, len - pos),
-			.data = data + pos };
-		dal_vector_append(&payloads->payloads, &payload);
-	}
-}
-
 static void construct(
 	struct ddc_service *ddc_service,
 	struct ddc_service_init_data *init_data)
@@ -574,32 +515,34 @@ bool dal_ddc_service_query_ddc_data(
 	/*TODO: len of payload data for i2c and aux is uint8!!!!,
 	 *  but we want to read 256 over i2c!!!!*/
 	if (dal_ddc_service_is_in_aux_transaction_mode(ddc)) {
-
-		struct aux_payloads *payloads =
-			dal_ddc_aux_payloads_create(ddc->ctx, payloads_num);
-
-		struct aux_command command = {
-			.payloads = dal_ddc_aux_payloads_get(payloads),
-			.number_of_payloads = 0,
+		struct aux_payload write_payload = {
+			.i2c_over_aux = true,
+			.write = true,
+			.mot = true,
+			.address = address,
+			.length = write_size,
+			.data = write_buf,
+			.reply = NULL,
 			.defer_delay = get_defer_delay(ddc),
-			.max_defer_write_retry = 0 };
+		};
 
-		dal_ddc_aux_payloads_add(
-			payloads, address, write_size, write_buf, true);
-
-		dal_ddc_aux_payloads_add(
-			payloads, address, read_size, read_buf, false);
-
-		command.number_of_payloads =
-			dal_ddc_aux_payloads_get_count(payloads);
+		struct aux_payload read_payload = {
+			.i2c_over_aux = true,
+			.write = false,
+			.mot = false,
+			.address = address,
+			.length = read_size,
+			.data = read_buf,
+			.reply = NULL,
+			.defer_delay = get_defer_delay(ddc),
+		};
 
-		ret = dal_i2caux_submit_aux_command(
-				ddc->ctx->i2caux,
-				ddc->ddc_pin,
-				&command);
+		ret = dc_link_aux_transfer_with_retries(ddc, &write_payload);
 
-		dal_ddc_aux_payloads_destroy(&payloads);
+		if (!ret)
+			return false;
 
+		ret = dc_link_aux_transfer_with_retries(ddc, &read_payload);
 	} else {
 		struct i2c_payloads *payloads =
 			dal_ddc_i2c_payloads_create(ddc->ctx, payloads_num);
@@ -631,56 +574,15 @@ bool dal_ddc_service_query_ddc_data(
 }
 
 int dc_link_aux_transfer(struct ddc_service *ddc,
-			     unsigned int address,
-			     uint8_t *reply,
-			     void *buffer,
-			     unsigned int size,
-			     enum aux_transaction_type type,
-			     enum i2caux_transaction_action action)
+		struct aux_payload *payload)
 {
-	struct ddc *ddc_pin = ddc->ddc_pin;
-	struct aux_engine *aux_engine;
-	enum aux_channel_operation_result operation_result;
-	struct aux_request_transaction_data aux_req;
-	struct aux_reply_transaction_data aux_rep;
-	uint8_t returned_bytes = 0;
-	int res = -1;
-	uint32_t status;
-
-	memset(&aux_req, 0, sizeof(aux_req));
-	memset(&aux_rep, 0, sizeof(aux_rep));
-
-	aux_engine = ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en];
-	aux_engine->funcs->acquire(aux_engine, ddc_pin);
-
-	aux_req.type = type;
-	aux_req.action = action;
-
-	aux_req.address = address;
-	aux_req.delay = 0;
-	aux_req.length = size;
-	aux_req.data = buffer;
-
-	aux_engine->funcs->submit_channel_request(aux_engine, &aux_req);
-	operation_result = aux_engine->funcs->get_channel_status(aux_engine, &returned_bytes);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		res = aux_engine->funcs->read_channel_reply(aux_engine, size,
-							buffer, reply,
-							&status);
-		break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		res = 0;
-		break;
-	case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN:
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
-		res = -1;
-		break;
-	}
-	aux_engine->funcs->release_engine(aux_engine);
-	return res;
+	return dce_aux_transfer(ddc, payload);
+}
+
+bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc,
+		struct aux_payload *payload)
+{
+	return dce_aux_transfer_with_retries(ddc, payload);
 }
 
 /*test only function*/
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 0caacb60b02f..09d301216076 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
@@ -49,6 +49,8 @@ static void wait_for_training_aux_rd_interval(
 {
 	union training_aux_rd_interval training_rd_interval;
 
+	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
@@ -117,6 +119,13 @@ static void dpcd_set_link_settings(
 	core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL,
 	&downspread.raw, sizeof(downspread));
 
+	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 &&
+		(link->dpcd_caps.link_rate_set >= 1 &&
+		link->dpcd_caps.link_rate_set <= 8)) {
+		core_link_write_dpcd(link, DP_LINK_RATE_SET,
+		&link->dpcd_caps.link_rate_set, 1);
+	}
+
 	DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x\n %x spread = %x\n",
 		__func__,
 		DP_LINK_BW_SET,
@@ -1542,7 +1551,7 @@ static uint32_t bandwidth_in_kbps_from_timing(
 
 	ASSERT(bits_per_channel != 0);
 
-	kbps = timing->pix_clk_khz;
+	kbps = timing->pix_clk_100hz / 10;
 	kbps *= bits_per_channel;
 
 	if (timing->flags.Y_ONLY != 1) {
@@ -1584,7 +1593,7 @@ bool dp_validate_mode_timing(
 	const struct dc_link_settings *link_setting;
 
 	/*always DP fail safe mode*/
-	if (timing->pix_clk_khz == (uint32_t) 25175 &&
+	if ((timing->pix_clk_100hz / 10) == (uint32_t) 25175 &&
 		timing->h_addressable == (uint32_t) 640 &&
 		timing->v_addressable == (uint32_t) 480)
 		return true;
@@ -1634,7 +1643,7 @@ void decide_link_settings(struct dc_stream_state *stream,
 
 	req_bw = bandwidth_in_kbps_from_timing(&stream->timing);
 
-	link = stream->sink->link;
+	link = stream->link;
 
 	/* if preferred is specified through AMDDP, use it, if it's enough
 	 * to drive the mode
@@ -1656,7 +1665,7 @@ void decide_link_settings(struct dc_stream_state *stream,
 	}
 
 	/* EDP use the link cap setting */
-	if (stream->sink->sink_signal == SIGNAL_TYPE_EDP) {
+	if (link->connector_signal == SIGNAL_TYPE_EDP) {
 		*link_setting = link->verified_link_cap;
 		return;
 	}
@@ -2002,11 +2011,7 @@ static void handle_automated_test(struct dc_link *link)
 		dp_test_send_phy_test_pattern(link);
 		test_response.bits.ACK = 1;
 	}
-	if (!test_request.raw)
-		/* no requests, revert all test signals
-		 * TODO: revert all test signals
-		 */
-		test_response.bits.ACK = 1;
+
 	/* send request acknowledgment */
 	if (test_response.bits.ACK)
 		core_link_write_dpcd(
@@ -2493,13 +2498,72 @@ bool detect_dp_sink_caps(struct dc_link *link)
 	/* TODO save sink caps in link->sink */
 }
 
+enum dc_link_rate linkRateInKHzToLinkRateMultiplier(uint32_t link_rate_in_khz)
+{
+	enum dc_link_rate link_rate;
+	// LinkRate is normally stored as a multiplier of 0.27 Gbps per lane. Do the translation.
+	switch (link_rate_in_khz) {
+	case 1620000:
+		link_rate = LINK_RATE_LOW;		// Rate_1 (RBR)		- 1.62 Gbps/Lane
+		break;
+	case 2160000:
+		link_rate = LINK_RATE_RATE_2;	// Rate_2			- 2.16 Gbps/Lane
+		break;
+	case 2430000:
+		link_rate = LINK_RATE_RATE_3;	// Rate_3			- 2.43 Gbps/Lane
+		break;
+	case 2700000:
+		link_rate = LINK_RATE_HIGH;		// Rate_4 (HBR)		- 2.70 Gbps/Lane
+		break;
+	case 3240000:
+		link_rate = LINK_RATE_RBR2;		// Rate_5 (RBR2)	- 3.24 Gbps/Lane
+		break;
+	case 4320000:
+		link_rate = LINK_RATE_RATE_6;	// Rate_6			- 4.32 Gbps/Lane
+		break;
+	case 5400000:
+		link_rate = LINK_RATE_HIGH2;	// Rate_7 (HBR2)	- 5.40 Gbps/Lane
+		break;
+	case 8100000:
+		link_rate = LINK_RATE_HIGH3;	// Rate_8 (HBR3)	- 8.10 Gbps/Lane
+		break;
+	default:
+		link_rate = LINK_RATE_UNKNOWN;
+		break;
+	}
+	return link_rate;
+}
+
 void detect_edp_sink_caps(struct dc_link *link)
 {
-	retrieve_link_cap(link);
+	uint8_t supported_link_rates[16] = {0};
+	uint32_t entry;
+	uint32_t link_rate_in_khz;
+	enum dc_link_rate link_rate = LINK_RATE_UNKNOWN;
 
-	if (link->reported_link_cap.link_rate == LINK_RATE_UNKNOWN)
-		link->reported_link_cap.link_rate = LINK_RATE_HIGH2;
+	retrieve_link_cap(link);
 
+	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14) {
+		// Read DPCD 00010h - 0001Fh 16 bytes at one shot
+		core_link_read_dpcd(link, DP_SUPPORTED_LINK_RATES,
+							supported_link_rates, sizeof(supported_link_rates));
+
+		link->dpcd_caps.link_rate_set = 0;
+		for (entry = 0; entry < 16; entry += 2) {
+			// DPCD register reports per-lane link rate = 16-bit link rate capability
+			// value X 200 kHz. Need multipler to find link rate in kHz.
+			link_rate_in_khz = (supported_link_rates[entry+1] * 0x100 +
+										supported_link_rates[entry]) * 200;
+
+			if (link_rate_in_khz != 0) {
+				link_rate = linkRateInKHzToLinkRateMultiplier(link_rate_in_khz);
+				if (link->reported_link_cap.link_rate < link_rate) {
+					link->reported_link_cap.link_rate = link_rate;
+					link->dpcd_caps.link_rate_set = entry;
+				}
+			}
+		}
+	}
 	link->verified_link_cap = link->reported_link_cap;
 }
 
@@ -2621,7 +2685,7 @@ bool dc_link_dp_set_test_pattern(
 	memset(&training_pattern, 0, sizeof(training_pattern));
 
 	for (i = 0; i < MAX_PIPES; i++) {
-		if (pipes[i].stream->sink->link == link) {
+		if (pipes[i].stream->link == link) {
 			pipe_ctx = &pipes[i];
 			break;
 		}
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 0065ec7d5330..f7f7515f65f4 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
@@ -70,13 +70,12 @@ void dp_enable_link_phy(
 	 */
 	for (i = 0; i < MAX_PIPES; i++) {
 		if (pipes[i].stream != NULL &&
-			pipes[i].stream->sink != NULL &&
-			pipes[i].stream->sink->link == link) {
+			pipes[i].stream->link == link) {
 			if (pipes[i].clock_source != NULL &&
 					pipes[i].clock_source->id != CLOCK_SOURCE_ID_DP_DTO) {
 				pipes[i].clock_source = dp_cs;
-				pipes[i].stream_res.pix_clk_params.requested_pix_clk =
-						pipes[i].stream->timing.pix_clk_khz;
+				pipes[i].stream_res.pix_clk_params.requested_pix_clk_100hz =
+						pipes[i].stream->timing.pix_clk_100hz;
 				pipes[i].clock_source->funcs->program_pix_clk(
 							pipes[i].clock_source,
 							&pipes[i].stream_res.pix_clk_params,
@@ -120,6 +119,10 @@ bool edp_receiver_ready_T9(struct dc_link *link)
 			break;
 		udelay(100); //MAx T9
 	} while (++tries < 50);
+
+	if (link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off > 0)
+		udelay(link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off * 1000);
+
 	return result;
 }
 bool edp_receiver_ready_T7(struct dc_link *link)
@@ -279,10 +282,8 @@ void dp_retrain_link_dp_test(struct dc_link *link,
 	for (i = 0; i < MAX_PIPES; i++) {
 		if (pipes[i].stream != NULL &&
 			!pipes[i].top_pipe &&
-			pipes[i].stream->sink != NULL &&
-			pipes[i].stream->sink->link != NULL &&
-			pipes[i].stream_res.stream_enc != NULL &&
-			pipes[i].stream->sink->link == link) {
+			pipes[i].stream->link != NULL &&
+			pipes[i].stream_res.stream_enc != NULL) {
 			udelay(100);
 
 			pipes[i].stream_res.stream_enc->funcs->dp_blank(
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 76137df74a53..349ab8017776 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -355,8 +355,8 @@ bool resource_are_streams_timing_synchronizable(
 				!= stream2->timing.v_addressable)
 		return false;
 
-	if (stream1->timing.pix_clk_khz
-				!= stream2->timing.pix_clk_khz)
+	if (stream1->timing.pix_clk_100hz
+				!= stream2->timing.pix_clk_100hz)
 		return false;
 
 	if (stream1->clamping.c_depth != stream2->clamping.c_depth)
@@ -1559,7 +1559,7 @@ static struct stream_encoder *find_first_free_match_stream_enc_for_link(
 {
 	int i;
 	int j = -1;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 
 	for (i = 0; i < pool->stream_enc_count; i++) {
 		if (!res_ctx->is_stream_enc_acquired[i] &&
@@ -1748,7 +1748,7 @@ static struct dc_stream_state *find_pll_sharable_stream(
 		if (resource_are_streams_timing_synchronizable(
 			stream_needs_pll, stream_has_pll)
 			&& !dc_is_dp_signal(stream_has_pll->signal)
-			&& stream_has_pll->sink->link->connector_signal
+			&& stream_has_pll->link->connector_signal
 			!= SIGNAL_TYPE_VIRTUAL)
 			return stream_has_pll;
 
@@ -1759,7 +1759,7 @@ static struct dc_stream_state *find_pll_sharable_stream(
 
 static int get_norm_pix_clk(const struct dc_crtc_timing *timing)
 {
-	uint32_t pix_clk = timing->pix_clk_khz;
+	uint32_t pix_clk = timing->pix_clk_100hz;
 	uint32_t normalized_pix_clk = pix_clk;
 
 	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
@@ -1791,15 +1791,60 @@ static void calculate_phy_pix_clks(struct dc_stream_state *stream)
 	/* update actual pixel clock on all streams */
 	if (dc_is_hdmi_signal(stream->signal))
 		stream->phy_pix_clk = get_norm_pix_clk(
-			&stream->timing);
+			&stream->timing) / 10;
 	else
 		stream->phy_pix_clk =
-			stream->timing.pix_clk_khz;
+			stream->timing.pix_clk_100hz / 10;
 
 	if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
 		stream->phy_pix_clk *= 2;
 }
 
+static int acquire_resource_from_hw_enabled_state(
+		struct resource_context *res_ctx,
+		const struct resource_pool *pool,
+		struct dc_stream_state *stream)
+{
+	struct dc_link *link = stream->link;
+	unsigned int inst;
+
+	/* Check for enabled DIG to identify enabled display */
+	if (!link->link_enc->funcs->is_dig_enabled(link->link_enc))
+		return -1;
+
+	/* Check for which front end is used by this encoder.
+	 * Note the inst is 1 indexed, where 0 is undefined.
+	 * Note that DIG_FE can source from different OTG but our
+	 * current implementation always map 1-to-1, so this code makes
+	 * the same assumption and doesn't check OTG source.
+	 */
+	inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1;
+
+	/* Instance should be within the range of the pool */
+	if (inst >= pool->pipe_count)
+		return -1;
+
+	if (!res_ctx->pipe_ctx[inst].stream) {
+		struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[inst];
+
+		pipe_ctx->stream_res.tg = pool->timing_generators[inst];
+		pipe_ctx->plane_res.mi = pool->mis[inst];
+		pipe_ctx->plane_res.hubp = pool->hubps[inst];
+		pipe_ctx->plane_res.ipp = pool->ipps[inst];
+		pipe_ctx->plane_res.xfm = pool->transforms[inst];
+		pipe_ctx->plane_res.dpp = pool->dpps[inst];
+		pipe_ctx->stream_res.opp = pool->opps[inst];
+		if (pool->dpps[inst])
+			pipe_ctx->plane_res.mpcc_inst = pool->dpps[inst]->inst;
+		pipe_ctx->pipe_idx = inst;
+
+		pipe_ctx->stream = stream;
+		return inst;
+	}
+
+	return -1;
+}
+
 enum dc_status resource_map_pool_resources(
 		const struct dc  *dc,
 		struct dc_state *context,
@@ -1824,8 +1869,15 @@ enum dc_status resource_map_pool_resources(
 
 	calculate_phy_pix_clks(stream);
 
-	/* acquire new resources */
-	pipe_idx = acquire_first_free_pipe(&context->res_ctx, pool, stream);
+	if (stream->apply_seamless_boot_optimization)
+		pipe_idx = acquire_resource_from_hw_enabled_state(
+				&context->res_ctx,
+				pool,
+				stream);
+
+	if (pipe_idx < 0)
+		/* acquire new resources */
+		pipe_idx = acquire_first_free_pipe(&context->res_ctx, pool, stream);
 
 #ifdef CONFIG_DRM_AMD_DC_DCN1_0
 	if (pipe_idx < 0)
@@ -1842,7 +1894,7 @@ enum dc_status resource_map_pool_resources(
 			&context->res_ctx, pool, stream);
 
 	if (!pipe_ctx->stream_res.stream_enc)
-		return DC_NO_STREAM_ENG_RESOURCE;
+		return DC_NO_STREAM_ENC_RESOURCE;
 
 	update_stream_engine_usage(
 		&context->res_ctx, pool,
@@ -1850,7 +1902,7 @@ enum dc_status resource_map_pool_resources(
 		true);
 
 	/* TODO: Add check if ASIC support and EDID audio */
-	if (!stream->sink->converter_disable_audio &&
+	if (!stream->converter_disable_audio &&
 	    dc_is_audio_capable_signal(pipe_ctx->stream->signal) &&
 	    stream->audio_info.mode_count) {
 		pipe_ctx->stream_res.audio = find_first_free_audio(
@@ -2112,7 +2164,7 @@ static void set_avi_info_frame(
 	itc = true;
 	itc_value = 1;
 
-	support = stream->sink->edid_caps.content_support;
+	support = stream->content_support;
 
 	if (itc) {
 		if (!support.bits.valid_content_type) {
@@ -2151,8 +2203,8 @@ static void set_avi_info_frame(
 
 	/* TODO : We should handle YCC quantization */
 	/* but we do not have matrix calculation */
-	if (stream->sink->edid_caps.qs_bit == 1 &&
-			stream->sink->edid_caps.qy_bit == 1) {
+	if (stream->qs_bit == 1 &&
+			stream->qy_bit == 1) {
 		if (color_space == COLOR_SPACE_SRGB ||
 			color_space == COLOR_SPACE_2020_RGB_FULLRANGE) {
 			hdmi_info.bits.Q0_Q1   = RGB_QUANTIZATION_FULL_RANGE;
@@ -2596,7 +2648,7 @@ void resource_build_bit_depth_reduction_params(struct dc_stream_state *stream,
 enum dc_status dc_validate_stream(struct dc *dc, struct dc_stream_state *stream)
 {
 	struct dc  *core_dc = dc;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct timing_generator *tg = core_dc->res_pool->timing_generators[0];
 	enum dc_status res = DC_OK;
 
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 66e5c4623a49..996298c35f42 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -35,20 +35,17 @@
 /*******************************************************************************
  * Private functions
  ******************************************************************************/
-void update_stream_signal(struct dc_stream_state *stream)
+void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink)
 {
-
-	struct dc_sink *dc_sink = stream->sink;
-
-	if (dc_sink->sink_signal == SIGNAL_TYPE_NONE)
-		stream->signal = stream->sink->link->connector_signal;
+	if (sink->sink_signal == SIGNAL_TYPE_NONE)
+		stream->signal = stream->link->connector_signal;
 	else
-		stream->signal = dc_sink->sink_signal;
+		stream->signal = sink->sink_signal;
 
 	if (dc_is_dvi_signal(stream->signal)) {
 		if (stream->ctx->dc->caps.dual_link_dvi &&
-		    stream->timing.pix_clk_khz > TMDS_MAX_PIXEL_CLOCK &&
-		    stream->sink->sink_signal != SIGNAL_TYPE_DVI_SINGLE_LINK)
+		    (stream->timing.pix_clk_100hz / 10) > TMDS_MAX_PIXEL_CLOCK &&
+		    sink->sink_signal != SIGNAL_TYPE_DVI_SINGLE_LINK)
 			stream->signal = SIGNAL_TYPE_DVI_DUAL_LINK;
 		else
 			stream->signal = SIGNAL_TYPE_DVI_SINGLE_LINK;
@@ -61,10 +58,15 @@ static void construct(struct dc_stream_state *stream,
 	uint32_t i = 0;
 
 	stream->sink = dc_sink_data;
-	stream->ctx = stream->sink->ctx;
-
 	dc_sink_retain(dc_sink_data);
 
+	stream->ctx = dc_sink_data->ctx;
+	stream->link = dc_sink_data->link;
+	stream->sink_patches = dc_sink_data->edid_caps.panel_patch;
+	stream->converter_disable_audio = dc_sink_data->converter_disable_audio;
+	stream->qs_bit = dc_sink_data->edid_caps.qs_bit;
+	stream->qy_bit = dc_sink_data->edid_caps.qy_bit;
+
 	/* Copy audio modes */
 	/* TODO - Remove this translation */
 	for (i = 0; i < (dc_sink_data->edid_caps.audio_mode_count); i++)
@@ -100,11 +102,14 @@ static void construct(struct dc_stream_state *stream,
 	/* EDID CAP translation for HDMI 2.0 */
 	stream->timing.flags.LTE_340MCSC_SCRAMBLE = dc_sink_data->edid_caps.lte_340mcsc_scramble;
 
-	update_stream_signal(stream);
+	update_stream_signal(stream, dc_sink_data);
 
 	stream->out_transfer_func = dc_create_transfer_func();
 	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++;
 }
 
 static void destruct(struct dc_stream_state *stream)
@@ -155,21 +160,43 @@ struct dc_stream_state *dc_create_stream_for_sink(
 	return stream;
 }
 
-struct dc_stream_status *dc_stream_get_status(
+/**
+ * dc_stream_get_status_from_state - Get stream status from given dc state
+ * @state: DC state to find the stream status in
+ * @stream: The stream to get the stream status for
+ *
+ * The given stream is expected to exist in the given dc state. Otherwise, NULL
+ * will be returned.
+ */
+struct dc_stream_status *dc_stream_get_status_from_state(
+	struct dc_state *state,
 	struct dc_stream_state *stream)
 {
 	uint8_t i;
-	struct dc  *dc = stream->ctx->dc;
 
-	for (i = 0; i < dc->current_state->stream_count; i++) {
-		if (stream == dc->current_state->streams[i])
-			return &dc->current_state->stream_status[i];
+	for (i = 0; i < state->stream_count; i++) {
+		if (stream == state->streams[i])
+			return &state->stream_status[i];
 	}
 
 	return NULL;
 }
 
 /**
+ * dc_stream_get_status() - Get current stream status of the given stream state
+ * @stream: The stream to get the stream status for.
+ *
+ * The given stream is expected to exist in dc->current_state. Otherwise, NULL
+ * will be returned.
+ */
+struct dc_stream_status *dc_stream_get_status(
+	struct dc_stream_state *stream)
+{
+	struct dc *dc = stream->ctx->dc;
+	return dc_stream_get_status_from_state(dc->current_state, stream);
+}
+
+/**
  * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address
  */
 bool dc_stream_set_cursor_attributes(
@@ -334,16 +361,12 @@ void dc_stream_log(const struct dc *dc, const struct dc_stream_state *stream)
 			stream->output_color_space);
 	DC_LOG_DC(
 			"\tpix_clk_khz: %d, h_total: %d, v_total: %d, pixelencoder:%d, displaycolorDepth:%d\n",
-			stream->timing.pix_clk_khz,
+			stream->timing.pix_clk_100hz / 10,
 			stream->timing.h_total,
 			stream->timing.v_total,
 			stream->timing.pixel_encoding,
 			stream->timing.display_color_depth);
 	DC_LOG_DC(
-			"\tsink name: %s, serial: %d\n",
-			stream->sink->edid_caps.display_name,
-			stream->sink->edid_caps.serial_number);
-	DC_LOG_DC(
 			"\tlink: %d\n",
-			stream->sink->link->link_index);
+			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 c60c9b4c3075..ee6bd50f60b8 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
@@ -40,11 +40,14 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state
 	plane_state->ctx = ctx;
 
 	plane_state->gamma_correction = dc_create_gamma();
-	plane_state->gamma_correction->is_identity = true;
+	if (plane_state->gamma_correction != NULL)
+		plane_state->gamma_correction->is_identity = true;
 
 	plane_state->in_transfer_func = dc_create_transfer_func();
-	plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
-	plane_state->in_transfer_func->ctx = ctx;
+	if (plane_state->in_transfer_func != NULL) {
+		plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
+		plane_state->in_transfer_func->ctx = ctx;
+	}
 }
 
 static void destruct(struct dc_plane_state *plane_state)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c
new file mode 100644
index 000000000000..6ce87b682a32
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c
@@ -0,0 +1,123 @@
+/*
+ * Copyright 2018 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
+ *
+ */
+
+#include "vm_helper.h"
+
+static void mark_vmid_used(struct vm_helper *vm_helper, unsigned int pos, uint8_t hubp_idx)
+{
+	struct vmid_usage vmids = vm_helper->hubp_vmid_usage[hubp_idx];
+
+	vmids.vmid_usage[0] = vmids.vmid_usage[1];
+	vmids.vmid_usage[1] = 1 << pos;
+}
+
+static void add_ptb_to_table(struct vm_helper *vm_helper, unsigned int vmid, uint64_t ptb)
+{
+	vm_helper->ptb_assigned_to_vmid[vmid] = ptb;
+	vm_helper->num_vmids_available--;
+}
+
+static void clear_entry_from_vmid_table(struct vm_helper *vm_helper, unsigned int vmid)
+{
+	vm_helper->ptb_assigned_to_vmid[vmid] = 0;
+	vm_helper->num_vmids_available++;
+}
+
+static void evict_vmids(struct vm_helper *vm_helper)
+{
+	int i;
+	uint16_t ord = 0;
+
+	for (i = 0; i < vm_helper->num_vmid; i++)
+		ord |= vm_helper->hubp_vmid_usage[i].vmid_usage[0] | vm_helper->hubp_vmid_usage[i].vmid_usage[1];
+
+	// At this point any positions with value 0 are unused vmids, evict them
+	for (i = 1; i < vm_helper->num_vmid; i++) {
+		if (ord & (1u << i))
+			clear_entry_from_vmid_table(vm_helper, i);
+	}
+}
+
+// Return value of -1 indicates vmid table unitialized or ptb dne in the table
+static int get_existing_vmid_for_ptb(struct vm_helper *vm_helper, uint64_t ptb)
+{
+	int i;
+
+	for (i = 0; i < vm_helper->num_vmid; i++) {
+		if (vm_helper->ptb_assigned_to_vmid[i] == ptb)
+			return i;
+	}
+
+	return -1;
+}
+
+// Expected to be called only when there's an available vmid
+static int get_next_available_vmid(struct vm_helper *vm_helper)
+{
+	int i;
+
+	for (i = 1; i < vm_helper->num_vmid; i++) {
+		if (vm_helper->ptb_assigned_to_vmid[i] == 0)
+			return i;
+	}
+
+	return -1;
+}
+
+uint8_t get_vmid_for_ptb(struct vm_helper *vm_helper, int64_t ptb, uint8_t hubp_idx)
+{
+	unsigned int vmid = 0;
+	int vmid_exists = -1;
+
+	// Physical address gets vmid 0
+	if (ptb == 0)
+		return 0;
+
+	vmid_exists = get_existing_vmid_for_ptb(vm_helper, ptb);
+
+	if (vmid_exists != -1) {
+		mark_vmid_used(vm_helper, vmid_exists, hubp_idx);
+		vmid = vmid_exists;
+	} else {
+		if (vm_helper->num_vmids_available == 0)
+			evict_vmids(vm_helper);
+
+		vmid = get_next_available_vmid(vm_helper);
+		mark_vmid_used(vm_helper, vmid, hubp_idx);
+		add_ptb_to_table(vm_helper, vmid, ptb);
+	}
+
+	return vmid;
+}
+
+void init_vm_helper(struct vm_helper *vm_helper, unsigned int num_vmid, unsigned int num_hubp)
+{
+	vm_helper->num_vmid = num_vmid;
+	vm_helper->num_hubp = num_hubp;
+	vm_helper->num_vmids_available = num_vmid - 1;
+
+	memset(vm_helper->hubp_vmid_usage, 0, sizeof(vm_helper->hubp_vmid_usage[0]) * MAX_HUBP);
+	memset(vm_helper->ptb_assigned_to_vmid, 0, sizeof(vm_helper->ptb_assigned_to_vmid[0]) * MAX_VMID);
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 4b5bbb13ce7f..1a7fd6aa77eb 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -39,7 +39,7 @@
 #include "inc/hw/dmcu.h"
 #include "dml/display_mode_lib.h"
 
-#define DC_VER "3.2.08"
+#define DC_VER "3.2.17"
 
 #define MAX_SURFACES 3
 #define MAX_STREAMS 6
@@ -255,6 +255,8 @@ struct dc_debug_options {
 	bool scl_reset_length10;
 	bool hdmi20_disable;
 	bool skip_detection_link_training;
+	unsigned int force_odm_combine; //bit vector based on otg inst
+	unsigned int force_fclk_khz;
 };
 
 struct dc_debug_data {
@@ -263,7 +265,6 @@ struct dc_debug_data {
 	uint32_t auxErrorCount;
 };
 
-
 struct dc_state;
 struct resource_pool;
 struct dce_hwseq;
@@ -339,8 +340,13 @@ struct dc_init_data {
 	uint32_t log_mask;
 };
 
-struct dc *dc_create(const struct dc_init_data *init_params);
+struct dc_callback_init {
+	uint8_t reserved;
+};
 
+struct dc *dc_create(const struct dc_init_data *init_params);
+void dc_init_callbacks(struct dc *dc,
+		const struct dc_callback_init *init_params);
 void dc_destroy(struct dc **dc);
 
 /*******************************************************************************
@@ -440,6 +446,7 @@ union surface_update_flags {
 		uint32_t coeff_reduction_change:1;
 		uint32_t output_tf_change:1;
 		uint32_t pixel_format_change:1;
+		uint32_t plane_size_change:1;
 
 		/* Full updates */
 		uint32_t new_plane:1;
@@ -587,6 +594,10 @@ struct dc_validation_set {
 	uint8_t plane_count;
 };
 
+bool dc_validate_seamless_boot_timing(struct dc *dc,
+				const struct dc_sink *sink,
+				struct dc_crtc_timing *crtc_timing);
+
 enum dc_status dc_validate_plane(struct dc *dc, const struct dc_plane_state *plane_state);
 
 void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx *info);
@@ -652,6 +663,7 @@ struct dpcd_caps {
 	int8_t branch_dev_name[6];
 	int8_t branch_hw_revision;
 	int8_t branch_fw_revision[2];
+	uint8_t link_rate_set;
 
 	bool allow_invalid_MSA_timing_param;
 	bool panel_mode_edp;
@@ -742,6 +754,9 @@ void dc_set_power_state(
 		struct dc *dc,
 		enum dc_acpi_cm_power_state power_state);
 void dc_resume(struct dc *dc);
+unsigned int dc_get_current_backlight_pwm(struct dc *dc);
+unsigned int dc_get_target_backlight_pwm(struct dc *dc);
+
 bool dc_is_dmcu_initialized(struct dc *dc);
 
 #endif /* DC_INTERFACE_H_ */
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 a8b3cedf9431..78c3b300ec45 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
@@ -86,10 +86,6 @@ struct dc_vbios_funcs {
 
 	bool (*is_accelerated_mode)(
 		struct dc_bios *bios);
-	bool (*is_active_display)(
-		struct dc_bios *bios,
-		enum signal_type signal,
-		const struct connector_device_tag_info *device_tag);
 	void (*set_scratch_critical_state)(
 		struct dc_bios *bios,
 		bool state);
@@ -125,10 +121,6 @@ struct dc_vbios_funcs {
 	enum bp_result (*program_crtc_timing)(
 		struct dc_bios *bios,
 		struct bp_hw_crtc_timing_parameters *bp_params);
-
-	enum bp_result (*crtc_source_select)(
-		struct dc_bios *bios,
-		struct bp_crtc_source_select *bp_params);
 	enum bp_result (*program_display_engine_pll)(
 		struct dc_bios *bios,
 		struct bp_pixel_clock_parameters *bp_params);
@@ -145,7 +137,6 @@ struct dc_vbios_funcs {
 };
 
 struct bios_registers {
-	uint32_t BIOS_SCRATCH_0;
 	uint32_t BIOS_SCRATCH_3;
 	uint32_t BIOS_SCRATCH_6;
 };
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 da93ab43f2d8..d4eab33c453b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -46,11 +46,14 @@ enum dc_lane_count {
  */
 enum dc_link_rate {
 	LINK_RATE_UNKNOWN = 0,
-	LINK_RATE_LOW = 0x06,
-	LINK_RATE_HIGH = 0x0A,
-	LINK_RATE_RBR2 = 0x0C,
-	LINK_RATE_HIGH2 = 0x14,
-	LINK_RATE_HIGH3 = 0x1E
+	LINK_RATE_LOW = 0x06,		// Rate_1 (RBR)	- 1.62 Gbps/Lane
+	LINK_RATE_RATE_2 = 0x08,	// Rate_2		- 2.16 Gbps/Lane
+	LINK_RATE_RATE_3 = 0x09,	// Rate_3		- 2.43 Gbps/Lane
+	LINK_RATE_HIGH = 0x0A,		// Rate_4 (HBR)	- 2.70 Gbps/Lane
+	LINK_RATE_RBR2 = 0x0C,		// Rate_5 (RBR2)- 3.24 Gbps/Lane
+	LINK_RATE_RATE_6 = 0x10,	// Rate_6		- 4.32 Gbps/Lane
+	LINK_RATE_HIGH2 = 0x14,		// Rate_7 (HBR2)- 5.40 Gbps/Lane
+	LINK_RATE_HIGH3 = 0x1E		// Rate_8 (HBR3)- 8.10 Gbps/Lane
 };
 
 enum dc_link_spread {
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 4842d2378bbf..597d38393379 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -29,31 +29,59 @@
 #include "dm_services.h"
 #include <stdarg.h>
 
+struct dc_reg_value_masks {
+	uint32_t value;
+	uint32_t mask;
+};
+
+struct dc_reg_sequence {
+	uint32_t addr;
+	struct dc_reg_value_masks value_masks;
+};
+
+static inline void set_reg_field_value_masks(
+	struct dc_reg_value_masks *field_value_mask,
+	uint32_t value,
+	uint32_t mask,
+	uint8_t shift)
+{
+	ASSERT(mask != 0);
+
+	field_value_mask->value = (field_value_mask->value & ~mask) | (mask & (value << shift));
+	field_value_mask->mask = field_value_mask->mask | mask;
+}
+
 uint32_t generic_reg_update_ex(const struct dc_context *ctx,
 		uint32_t addr, uint32_t reg_val, int n,
 		uint8_t shift1, uint32_t mask1, uint32_t field_value1,
 		...)
 {
+	struct dc_reg_value_masks field_value_mask = {0};
 	uint32_t shift, mask, field_value;
 	int i = 1;
 
 	va_list ap;
 	va_start(ap, field_value1);
 
-	reg_val = set_reg_field_value_ex(reg_val, field_value1, mask1, shift1);
+	/* gather all bits value/mask getting updated in this register */
+	set_reg_field_value_masks(&field_value_mask,
+			field_value1, mask1, shift1);
 
 	while (i < n) {
 		shift = va_arg(ap, uint32_t);
 		mask = va_arg(ap, uint32_t);
 		field_value = va_arg(ap, uint32_t);
 
-		reg_val = set_reg_field_value_ex(reg_val, field_value, mask, shift);
+		set_reg_field_value_masks(&field_value_mask,
+				field_value, mask, shift);
 		i++;
 	}
-
-	dm_write_reg(ctx, addr, reg_val);
 	va_end(ap);
 
+
+	/* mmio write directly */
+	reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value;
+	dm_write_reg(ctx, addr, reg_val);
 	return reg_val;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
index e72fce4eca65..da55d623647a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
@@ -97,6 +97,8 @@ struct dc_plane_address {
 			union large_integer chroma_dcc_const_color;
 		} video_progressive;
 	};
+
+	union large_integer page_table_base;
 };
 
 struct dc_size {
@@ -730,7 +732,7 @@ struct dc_crtc_timing {
 	uint32_t v_front_porch;
 	uint32_t v_sync_width;
 
-	uint32_t pix_clk_khz;
+	uint32_t pix_clk_100hz;
 
 	uint32_t vic;
 	uint32_t hdmi_vic;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index 29f19d57ff7a..8fc223defed4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -30,6 +30,7 @@
 #include "grph_object_defs.h"
 
 struct dc_link_status {
+	bool link_active;
 	struct dpcd_caps *dpcd_caps;
 };
 
@@ -110,6 +111,7 @@ struct dc_link {
 	union ddi_channel_mapping ddi_channel_mapping;
 	struct connector_device_tag_info device_tag;
 	struct dpcd_caps dpcd_caps;
+	uint32_t dongle_max_pix_clk;
 	unsigned short chip_caps;
 	unsigned int dpcd_sink_count;
 	enum edp_revision edp_revision;
@@ -124,6 +126,7 @@ struct dc_link {
 	struct dc_link_status link_status;
 
 	struct link_trace link_trace;
+	struct gpio *hpd_gpio;
 };
 
 const struct dc_link_status *dc_link_get_status(const struct dc_link *dc_link);
@@ -146,8 +149,7 @@ static inline struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_
  */
 bool dc_link_set_backlight_level(const struct dc_link *dc_link,
 		uint32_t backlight_pwm_u16_16,
-		uint32_t frame_ramp,
-		const struct dc_stream_state *stream);
+		uint32_t frame_ramp);
 
 int dc_link_get_backlight_level(const struct dc_link *dc_link);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index be34d638e15d..a798694992b9 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -32,17 +32,18 @@
 /*******************************************************************************
  * Stream Interfaces
  ******************************************************************************/
+struct timing_sync_info {
+	int group_id;
+	int group_size;
+	bool master;
+};
 
 struct dc_stream_status {
 	int primary_otg_inst;
 	int stream_enc_inst;
 	int plane_count;
+	struct timing_sync_info timing_sync_info;
 	struct dc_plane_state *plane_states[MAX_SURFACE_NUM];
-
-	/*
-	 * link this stream passes through
-	 */
-	struct dc_link *link;
 };
 
 // TODO: References to this needs to be removed..
@@ -50,8 +51,20 @@ struct freesync_context {
 	bool dummy;
 };
 
+union vline_config {
+	unsigned int line_number;
+	unsigned long long delta_in_ns;
+};
+
+
 struct dc_stream_state {
+	// sink is deprecated, new code should not reference
+	// this pointer
 	struct dc_sink *sink;
+
+	struct dc_link *link;
+	struct dc_panel_patch sink_patches;
+	union display_content_support content_support;
 	struct dc_crtc_timing timing;
 	struct dc_crtc_timing_adjust adjust;
 	struct dc_info_packet vrr_infopacket;
@@ -80,8 +93,9 @@ struct dc_stream_state {
 	enum view_3d_format view_format;
 
 	bool ignore_msa_timing_param;
-
-	unsigned long long periodic_fn_vsync_delta;
+	bool converter_disable_audio;
+	uint8_t qs_bit;
+	uint8_t qy_bit;
 
 	/* TODO: custom INFO packets */
 	/* TODO: ABM info (DMCU) */
@@ -91,7 +105,9 @@ struct dc_stream_state {
 
 	/* DMCU info */
 	unsigned int abm_level;
-	unsigned int bl_pwm_level;
+
+	union vline_config periodic_vsync_config;
+	union vline_config enhanced_sync_config;
 
 	/* from core_stream struct */
 	struct dc_context *ctx;
@@ -103,7 +119,8 @@ struct dc_stream_state {
 	int phy_pix_clk;
 	enum signal_type signal;
 	bool dpms_off;
-	bool apply_edp_fast_boot_optimization;
+
+	void *dm_stream_context;
 
 	struct dc_cursor_attributes cursor_attributes;
 	struct dc_cursor_position cursor_position;
@@ -117,6 +134,21 @@ struct dc_stream_state {
 	/* Computed state bits */
 	bool mode_changed : 1;
 
+	/* Output from DC when stream state is committed or altered
+	 * DC may only access these values during:
+	 * dc_commit_state, dc_commit_state_no_check, dc_commit_streams
+	 * values may not change outside of those calls
+	 */
+	struct {
+		// For interrupt management, some hardware instance
+		// offsets need to be exposed to DM
+		uint8_t otg_offset;
+	} out;
+
+	bool apply_edp_fast_boot_optimization;
+	bool apply_seamless_boot_optimization;
+
+	uint32_t stream_id;
 };
 
 struct dc_stream_update {
@@ -126,7 +158,9 @@ struct dc_stream_update {
 	struct dc_info_packet *hdr_static_metadata;
 	unsigned int *abm_level;
 
-	unsigned long long *periodic_fn_vsync_delta;
+	union vline_config *periodic_vsync_config;
+	union vline_config *enhanced_sync_config;
+
 	struct dc_crtc_timing_adjust *adjust;
 	struct dc_info_packet *vrr_infopacket;
 	struct dc_info_packet *vsc_infopacket;
@@ -163,7 +197,6 @@ void dc_commit_updates_for_stream(struct dc *dc,
 		int surface_count,
 		struct dc_stream_state *stream,
 		struct dc_stream_update *stream_update,
-		struct dc_plane_state **plane_states,
 		struct dc_state *state);
 /*
  * Log the current stream state.
@@ -256,11 +289,14 @@ enum surface_update_type dc_check_update_surfaces_for_stream(
  */
 struct dc_stream_state *dc_create_stream_for_sink(struct dc_sink *dc_sink);
 
-void update_stream_signal(struct dc_stream_state *stream);
+void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink);
 
 void dc_stream_retain(struct dc_stream_state *dc_stream);
 void dc_stream_release(struct dc_stream_state *dc_stream);
 
+struct dc_stream_status *dc_stream_get_status_from_state(
+	struct dc_state *state,
+	struct dc_stream_state *stream);
 struct dc_stream_status *dc_stream_get_status(
 	struct dc_stream_state *dc_stream);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 0b20ae23f169..da2009a108cf 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -97,8 +97,8 @@ struct dc_context {
 	struct dc_bios *dc_bios;
 	bool created_bios;
 	struct gpio_service *gpio_service;
-	struct i2caux *i2caux;
 	uint32_t dc_sink_id_count;
+	uint32_t dc_stream_id_count;
 	uint64_t fbc_gpu_addr;
 };
 
@@ -201,6 +201,7 @@ union display_content_support {
 struct dc_panel_patch {
 	unsigned int dppowerup_delay;
 	unsigned int extra_t12_ms;
+	unsigned int extra_delay_backlight_off;
 };
 
 struct dc_edid_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
index 2a342eae80fd..01e56f1a9f34 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
@@ -314,8 +314,8 @@ static bool dce_abm_immediate_disable(struct abm *abm)
 
 	/* setDMCUParam_ABMLevel */
 	REG_UPDATE_2(MASTER_COMM_CMD_REG,
-			MASTER_COMM_CMD_REG_BYTE0, MCP_ABM_LEVEL_SET,
-			MASTER_COMM_CMD_REG_BYTE2, MCP_DISABLE_ABM_IMMEDIATELY);
+			MASTER_COMM_CMD_REG_BYTE0, MCP_ABM_PIPE_SET,
+			MASTER_COMM_CMD_REG_BYTE1, MCP_DISABLE_ABM_IMMEDIATELY);
 
 	/* notifyDMCUMsg */
 	REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
index aaeb7faac0c4..4febf4ef7240 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
@@ -24,6 +24,7 @@
  */
 
 #include "dm_services.h"
+#include "core_types.h"
 #include "dce_aux.h"
 #include "dce/dce_11_0_sh_mask.h"
 
@@ -41,17 +42,17 @@
 	container_of((ptr), struct aux_engine_dce110, base)
 
 #define FROM_ENGINE(ptr) \
-	FROM_AUX_ENGINE(container_of((ptr), struct aux_engine, base))
+	FROM_AUX_ENGINE(container_of((ptr), struct dce_aux, base))
 
 #define FROM_AUX_ENGINE_ENGINE(ptr) \
-	container_of((ptr), struct aux_engine, base)
+	container_of((ptr), struct dce_aux, base)
 enum {
 	AUX_INVALID_REPLY_RETRY_COUNTER = 1,
 	AUX_TIMED_OUT_RETRY_COUNTER = 2,
 	AUX_DEFER_RETRY_COUNTER = 6
 };
 static void release_engine(
-	struct aux_engine *engine)
+	struct dce_aux *engine)
 {
 	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
 
@@ -66,7 +67,7 @@ static void release_engine(
 #define DMCU_CAN_ACCESS_AUX 2
 
 static bool is_engine_available(
-	struct aux_engine *engine)
+	struct dce_aux *engine)
 {
 	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
 
@@ -79,7 +80,7 @@ static bool is_engine_available(
 	return (field != DMCU_CAN_ACCESS_AUX);
 }
 static bool acquire_engine(
-	struct aux_engine *engine)
+	struct dce_aux *engine)
 {
 	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
 
@@ -155,7 +156,7 @@ static bool acquire_engine(
 	(0xFF & (address))
 
 static void submit_channel_request(
-	struct aux_engine *engine,
+	struct dce_aux *engine,
 	struct aux_request_transaction_data *request)
 {
 	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
@@ -247,7 +248,7 @@ static void submit_channel_request(
 	REG_UPDATE(AUX_SW_CONTROL, AUX_SW_GO, 1);
 }
 
-static int read_channel_reply(struct aux_engine *engine, uint32_t size,
+static int read_channel_reply(struct dce_aux *engine, uint32_t size,
 			      uint8_t *buffer, uint8_t *reply_result,
 			      uint32_t *sw_status)
 {
@@ -273,7 +274,8 @@ static int read_channel_reply(struct aux_engine *engine, uint32_t size,
 
 	REG_GET(AUX_SW_DATA, AUX_SW_DATA, &reply_result_32);
 	reply_result_32 = reply_result_32 >> 4;
-	*reply_result = (uint8_t)reply_result_32;
+	if (reply_result != NULL)
+		*reply_result = (uint8_t)reply_result_32;
 
 	if (reply_result_32 == 0) { /* ACK */
 		uint32_t i = 0;
@@ -299,61 +301,8 @@ static int read_channel_reply(struct aux_engine *engine, uint32_t size,
 	return 0;
 }
 
-static void process_channel_reply(
-	struct aux_engine *engine,
-	struct aux_reply_transaction_data *reply)
-{
-	int bytes_replied;
-	uint8_t reply_result;
-	uint32_t sw_status;
-
-	bytes_replied = read_channel_reply(engine, reply->length, reply->data,
-					   &reply_result, &sw_status);
-
-	/* in case HPD is LOW, exit AUX transaction */
-	if ((sw_status & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK)) {
-		reply->status = AUX_TRANSACTION_REPLY_HPD_DISCON;
-		return;
-	}
-
-	if (bytes_replied < 0) {
-		/* Need to handle an error case...
-		 * Hopefully, upper layer function won't call this function if
-		 * the number of bytes in the reply was 0, because there was
-		 * surely an error that was asserted that should have been
-		 * handled for hot plug case, this could happens
-		 */
-		if (!(sw_status & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK)) {
-			reply->status = AUX_TRANSACTION_REPLY_INVALID;
-			ASSERT_CRITICAL(false);
-			return;
-		}
-	} else {
-
-		switch (reply_result) {
-		case 0: /* ACK */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_ACK;
-		break;
-		case 1: /* NACK */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_NACK;
-		break;
-		case 2: /* DEFER */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_DEFER;
-		break;
-		case 4: /* AUX ACK / I2C NACK */
-			reply->status = AUX_TRANSACTION_REPLY_I2C_NACK;
-		break;
-		case 8: /* AUX ACK / I2C DEFER */
-			reply->status = AUX_TRANSACTION_REPLY_I2C_DEFER;
-		break;
-		default:
-			reply->status = AUX_TRANSACTION_REPLY_INVALID;
-		}
-	}
-}
-
 static enum aux_channel_operation_result get_channel_status(
-	struct aux_engine *engine,
+	struct dce_aux *engine,
 	uint8_t *returned_bytes)
 {
 	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
@@ -414,469 +363,22 @@ static enum aux_channel_operation_result get_channel_status(
 		return AUX_CHANNEL_OPERATION_FAILED_TIMEOUT;
 	}
 }
-static void process_read_reply(
-	struct aux_engine *engine,
-	struct read_command_context *ctx)
-{
-	engine->funcs->process_channel_reply(engine, &ctx->reply);
-
-	switch (ctx->reply.status) {
-	case AUX_TRANSACTION_REPLY_AUX_ACK:
-		ctx->defer_retry_aux = 0;
-		if (ctx->returned_byte > ctx->current_read_length) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else if (ctx->returned_byte < ctx->current_read_length) {
-			ctx->current_read_length -= ctx->returned_byte;
-
-			ctx->offset += ctx->returned_byte;
-
-			++ctx->invalid_reply_retry_aux_on_ack;
-
-			if (ctx->invalid_reply_retry_aux_on_ack >
-				AUX_INVALID_REPLY_RETRY_COUNTER) {
-				ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-				ctx->operation_succeeded = false;
-			}
-		} else {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-			ctx->transaction_complete = true;
-			ctx->operation_succeeded = true;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_NACK:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-		ctx->operation_succeeded = false;
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_DEFER:
-		++ctx->defer_retry_aux;
-
-		if (ctx->defer_retry_aux > AUX_DEFER_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_I2C_DEFER:
-		ctx->defer_retry_aux = 0;
-
-		++ctx->defer_retry_i2c;
-
-		if (ctx->defer_retry_i2c > AUX_DEFER_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-static void process_read_request(
-	struct aux_engine *engine,
-	struct read_command_context *ctx)
-{
-	enum aux_channel_operation_result operation_result;
 
-	engine->funcs->submit_channel_request(engine, &ctx->request);
-
-	operation_result = engine->funcs->get_channel_status(
-		engine, &ctx->returned_byte);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		if (ctx->returned_byte > ctx->current_read_length) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else {
-			ctx->timed_out_retry_aux = 0;
-			ctx->invalid_reply_retry_aux = 0;
-
-			ctx->reply.length = ctx->returned_byte;
-			ctx->reply.data = ctx->buffer;
-
-			process_read_reply(engine, ctx);
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-		++ctx->invalid_reply_retry_aux;
-
-		if (ctx->invalid_reply_retry_aux >
-			AUX_INVALID_REPLY_RETRY_COUNTER) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else
-			udelay(400);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
-		++ctx->timed_out_retry_aux;
-
-		if (ctx->timed_out_retry_aux > AUX_TIMED_OUT_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		} else {
-			/* DP 1.2a, table 2-58:
-			 * "S3: AUX Request CMD PENDING:
-			 * retry 3 times, with 400usec wait on each"
-			 * The HW timeout is set to 550usec,
-			 * so we should not wait here
-			 */
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-static bool read_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-	struct read_command_context ctx;
-
-	ctx.buffer = request->payload.data;
-	ctx.current_read_length = request->payload.length;
-	ctx.offset = 0;
-	ctx.timed_out_retry_aux = 0;
-	ctx.invalid_reply_retry_aux = 0;
-	ctx.defer_retry_aux = 0;
-	ctx.defer_retry_i2c = 0;
-	ctx.invalid_reply_retry_aux_on_ack = 0;
-	ctx.transaction_complete = false;
-	ctx.operation_succeeded = true;
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_DP;
-		ctx.request.action = I2CAUX_TRANSACTION_ACTION_DP_READ;
-		ctx.request.address = request->payload.address;
-	} else if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_I2C;
-		ctx.request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_READ;
-		ctx.request.address = request->payload.address >> 1;
-	} else {
-		/* in DAL2, there was no return in such case */
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	ctx.request.delay = 0;
-
-	do {
-		memset(ctx.buffer + ctx.offset, 0, ctx.current_read_length);
-
-		ctx.request.data = ctx.buffer + ctx.offset;
-		ctx.request.length = ctx.current_read_length;
-
-		process_read_request(engine, &ctx);
-
-		request->status = ctx.status;
-
-		if (ctx.operation_succeeded && !ctx.transaction_complete)
-			if (ctx.request.type == AUX_TRANSACTION_TYPE_I2C)
-				msleep(engine->delay);
-	} while (ctx.operation_succeeded && !ctx.transaction_complete);
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		DC_LOG_I2C_AUX("READ: addr:0x%x  value:0x%x Result:%d",
-				request->payload.address,
-				request->payload.data[0],
-				ctx.operation_succeeded);
-	}
-
-	return ctx.operation_succeeded;
-}
-
-static void process_write_reply(
-	struct aux_engine *engine,
-	struct write_command_context *ctx)
-{
-	engine->funcs->process_channel_reply(engine, &ctx->reply);
-
-	switch (ctx->reply.status) {
-	case AUX_TRANSACTION_REPLY_AUX_ACK:
-		ctx->operation_succeeded = true;
-
-		if (ctx->returned_byte) {
-			ctx->request.action = ctx->mot ?
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST;
-
-			ctx->current_write_length = 0;
-
-			++ctx->ack_m_retry;
-
-			if (ctx->ack_m_retry > AUX_DEFER_RETRY_COUNTER) {
-				ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-				ctx->operation_succeeded = false;
-			} else
-				udelay(300);
-		} else {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-			ctx->defer_retry_aux = 0;
-			ctx->ack_m_retry = 0;
-			ctx->transaction_complete = true;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_NACK:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-		ctx->operation_succeeded = false;
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_DEFER:
-		++ctx->defer_retry_aux;
-
-		if (ctx->defer_retry_aux > ctx->max_defer_retry) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_I2C_DEFER:
-		ctx->defer_retry_aux = 0;
-		ctx->current_write_length = 0;
-
-		ctx->request.action = ctx->mot ?
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST;
-
-		++ctx->defer_retry_i2c;
-
-		if (ctx->defer_retry_i2c > ctx->max_defer_retry) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-static void process_write_request(
-	struct aux_engine *engine,
-	struct write_command_context *ctx)
-{
-	enum aux_channel_operation_result operation_result;
-
-	engine->funcs->submit_channel_request(engine, &ctx->request);
-
-	operation_result = engine->funcs->get_channel_status(
-		engine, &ctx->returned_byte);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		ctx->timed_out_retry_aux = 0;
-		ctx->invalid_reply_retry_aux = 0;
-
-		ctx->reply.length = ctx->returned_byte;
-		ctx->reply.data = ctx->reply_data;
-
-		process_write_reply(engine, ctx);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-		++ctx->invalid_reply_retry_aux;
-
-		if (ctx->invalid_reply_retry_aux >
-			AUX_INVALID_REPLY_RETRY_COUNTER) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else
-			udelay(400);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
-		++ctx->timed_out_retry_aux;
-
-		if (ctx->timed_out_retry_aux > AUX_TIMED_OUT_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		} else {
-			/* DP 1.2a, table 2-58:
-			 * "S3: AUX Request CMD PENDING:
-			 * retry 3 times, with 400usec wait on each"
-			 * The HW timeout is set to 550usec,
-			 * so we should not wait here
-			 */
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-static bool write_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-	struct write_command_context ctx;
-
-	ctx.mot = middle_of_transaction;
-	ctx.buffer = request->payload.data;
-	ctx.current_write_length = request->payload.length;
-	ctx.timed_out_retry_aux = 0;
-	ctx.invalid_reply_retry_aux = 0;
-	ctx.defer_retry_aux = 0;
-	ctx.defer_retry_i2c = 0;
-	ctx.ack_m_retry = 0;
-	ctx.transaction_complete = false;
-	ctx.operation_succeeded = true;
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_DP;
-		ctx.request.action = I2CAUX_TRANSACTION_ACTION_DP_WRITE;
-		ctx.request.address = request->payload.address;
-	} else if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_I2C;
-		ctx.request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-		ctx.request.address = request->payload.address >> 1;
-	} else {
-		/* in DAL2, there was no return in such case */
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	ctx.request.delay = 0;
-
-	ctx.max_defer_retry =
-		(engine->max_defer_write_retry > AUX_DEFER_RETRY_COUNTER) ?
-			engine->max_defer_write_retry : AUX_DEFER_RETRY_COUNTER;
-
-	do {
-		ctx.request.data = ctx.buffer;
-		ctx.request.length = ctx.current_write_length;
-
-		process_write_request(engine, &ctx);
-
-		request->status = ctx.status;
-
-		if (ctx.operation_succeeded && !ctx.transaction_complete)
-			if (ctx.request.type == AUX_TRANSACTION_TYPE_I2C)
-				msleep(engine->delay);
-	} while (ctx.operation_succeeded && !ctx.transaction_complete);
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		DC_LOG_I2C_AUX("WRITE: addr:0x%x  value:0x%x Result:%d",
-				request->payload.address,
-				request->payload.data[0],
-				ctx.operation_succeeded);
-	}
-
-	return ctx.operation_succeeded;
-}
-static bool end_of_transaction_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request)
-{
-	struct i2caux_transaction_request dummy_request;
-	uint8_t dummy_data;
-
-	/* [tcheng] We only need to send the stop (read with MOT = 0)
-	 * for I2C-over-Aux, not native AUX
-	 */
-
-	if (request->payload.address_space !=
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C)
-		return false;
-
-	dummy_request.operation = request->operation;
-	dummy_request.payload.address_space = request->payload.address_space;
-	dummy_request.payload.address = request->payload.address;
-
-	/*
-	 * Add a dummy byte due to some receiver quirk
-	 * where one byte is sent along with MOT = 0.
-	 * Ideally this should be 0.
-	 */
-
-	dummy_request.payload.length = 0;
-	dummy_request.payload.data = &dummy_data;
-
-	if (request->operation == I2CAUX_TRANSACTION_READ)
-		return read_command(engine, &dummy_request, false);
-	else
-		return write_command(engine, &dummy_request, false);
-
-	/* according Syed, it does not need now DoDummyMOT */
-}
-static bool submit_request(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-
-	bool result;
-	bool mot_used = true;
-
-	switch (request->operation) {
-	case I2CAUX_TRANSACTION_READ:
-		result = read_command(engine, request, mot_used);
-	break;
-	case I2CAUX_TRANSACTION_WRITE:
-		result = write_command(engine, request, mot_used);
-	break;
-	default:
-		result = false;
-	}
-
-	/* [tcheng]
-	 * need to send stop for the last transaction to free up the AUX
-	 * if the above command fails, this would be the last transaction
-	 */
-
-	if (!middle_of_transaction || !result)
-		end_of_transaction_command(engine, request);
-
-	/* mask AUX interrupt */
-
-	return result;
-}
 enum i2caux_engine_type get_engine_type(
-		const struct aux_engine *engine)
+		const struct dce_aux *engine)
 {
 	return I2CAUX_ENGINE_TYPE_AUX;
 }
 
 static bool acquire(
-	struct aux_engine *engine,
+	struct dce_aux *engine,
 	struct ddc *ddc)
 {
 
 	enum gpio_result result;
 
-	if (engine->funcs->is_engine_available) {
-		/*check whether SW could use the engine*/
-		if (!engine->funcs->is_engine_available(engine))
-			return false;
-	}
+	if (!is_engine_available(engine))
+		return false;
 
 	result = dal_ddc_open(ddc, GPIO_MODE_HARDWARE,
 		GPIO_DDC_CONFIG_TYPE_MODE_AUX);
@@ -884,7 +386,7 @@ static bool acquire(
 	if (result != GPIO_RESULT_OK)
 		return false;
 
-	if (!engine->funcs->acquire_engine(engine)) {
+	if (!acquire_engine(engine)) {
 		dal_ddc_close(ddc);
 		return false;
 	}
@@ -894,21 +396,7 @@ static bool acquire(
 	return true;
 }
 
-static const struct aux_engine_funcs aux_engine_funcs = {
-	.acquire_engine = acquire_engine,
-	.submit_channel_request = submit_channel_request,
-	.process_channel_reply = process_channel_reply,
-	.read_channel_reply = read_channel_reply,
-	.get_channel_status = get_channel_status,
-	.is_engine_available = is_engine_available,
-	.release_engine = release_engine,
-	.destroy_engine = dce110_engine_destroy,
-	.submit_request = submit_request,
-	.get_engine_type = get_engine_type,
-	.acquire = acquire,
-};
-
-void dce110_engine_destroy(struct aux_engine **engine)
+void dce110_engine_destroy(struct dce_aux **engine)
 {
 
 	struct aux_engine_dce110 *engine110 = FROM_AUX_ENGINE(*engine);
@@ -917,7 +405,7 @@ void dce110_engine_destroy(struct aux_engine **engine)
 	*engine = NULL;
 
 }
-struct aux_engine *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_engine110,
+struct dce_aux *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_engine110,
 		struct dc_context *ctx,
 		uint32_t inst,
 		uint32_t timeout_period,
@@ -927,7 +415,6 @@ struct aux_engine *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_eng
 	aux_engine110->base.ctx = ctx;
 	aux_engine110->base.delay = 0;
 	aux_engine110->base.max_defer_write_retry = 0;
-	aux_engine110->base.funcs = &aux_engine_funcs;
 	aux_engine110->base.inst = inst;
 	aux_engine110->timeout_period = timeout_period;
 	aux_engine110->regs = regs;
@@ -935,3 +422,101 @@ struct aux_engine *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_eng
 	return &aux_engine110->base;
 }
 
+static enum i2caux_transaction_action i2caux_action_from_payload(struct aux_payload *payload)
+{
+	if (payload->i2c_over_aux) {
+		if (payload->write) {
+			if (payload->mot)
+				return I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT;
+			return I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
+		}
+		if (payload->mot)
+			return I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT;
+		return I2CAUX_TRANSACTION_ACTION_I2C_READ;
+	}
+	if (payload->write)
+		return I2CAUX_TRANSACTION_ACTION_DP_WRITE;
+	return I2CAUX_TRANSACTION_ACTION_DP_READ;
+}
+
+int dce_aux_transfer(struct ddc_service *ddc,
+		struct aux_payload *payload)
+{
+	struct ddc *ddc_pin = ddc->ddc_pin;
+	struct dce_aux *aux_engine;
+	enum aux_channel_operation_result operation_result;
+	struct aux_request_transaction_data aux_req;
+	struct aux_reply_transaction_data aux_rep;
+	uint8_t returned_bytes = 0;
+	int res = -1;
+	uint32_t status;
+
+	memset(&aux_req, 0, sizeof(aux_req));
+	memset(&aux_rep, 0, sizeof(aux_rep));
+
+	aux_engine = ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en];
+	acquire(aux_engine, ddc_pin);
+
+	if (payload->i2c_over_aux)
+		aux_req.type = AUX_TRANSACTION_TYPE_I2C;
+	else
+		aux_req.type = AUX_TRANSACTION_TYPE_DP;
+
+	aux_req.action = i2caux_action_from_payload(payload);
+
+	aux_req.address = payload->address;
+	aux_req.delay = payload->defer_delay * 10;
+	aux_req.length = payload->length;
+	aux_req.data = payload->data;
+
+	submit_channel_request(aux_engine, &aux_req);
+	operation_result = get_channel_status(aux_engine, &returned_bytes);
+
+	switch (operation_result) {
+	case AUX_CHANNEL_OPERATION_SUCCEEDED:
+		res = read_channel_reply(aux_engine, payload->length,
+							payload->data, payload->reply,
+							&status);
+		break;
+	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
+		res = 0;
+		break;
+	case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN:
+	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
+	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
+		res = -1;
+		break;
+	}
+	release_engine(aux_engine);
+	return res;
+}
+
+#define AUX_RETRY_MAX 7
+
+bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
+		struct aux_payload *payload)
+{
+	int i, ret = 0;
+	uint8_t reply;
+	bool payload_reply = true;
+
+	if (!payload->reply) {
+		payload_reply = false;
+		payload->reply = &reply;
+	}
+
+	for (i = 0; i < AUX_RETRY_MAX; i++) {
+		ret = dce_aux_transfer(ddc, payload);
+
+		if (ret >= 0) {
+			if (*payload->reply == 0) {
+				if (!payload_reply)
+					payload->reply = NULL;
+				return true;
+			}
+		}
+
+		udelay(1000);
+	}
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
index f7caab85dc80..d27f22c05e4b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
@@ -25,7 +25,9 @@
 
 #ifndef __DAL_AUX_ENGINE_DCE110_H__
 #define __DAL_AUX_ENGINE_DCE110_H__
-#include "aux_engine.h"
+
+#include "i2caux_interface.h"
+#include "inc/hw/aux_engine.h"
 
 #define AUX_COMMON_REG_LIST(id)\
 	SRI(AUX_CONTROL, DP_AUX, id), \
@@ -75,8 +77,20 @@ enum {	/* This is the timeout as defined in DP 1.2a,
 	 */
 	SW_AUX_TIMEOUT_PERIOD_MULTIPLIER = 4
 };
+
+struct dce_aux {
+	uint32_t inst;
+	struct ddc *ddc;
+	struct dc_context *ctx;
+	/* following values are expressed in milliseconds */
+	uint32_t delay;
+	uint32_t max_defer_write_retry;
+
+	bool acquire_reset;
+};
+
 struct aux_engine_dce110 {
-	struct aux_engine base;
+	struct dce_aux base;
 	const struct dce110_aux_registers *regs;
 	struct {
 		uint32_t aux_control;
@@ -96,16 +110,22 @@ struct aux_engine_dce110_init_data {
 	const struct dce110_aux_registers *regs;
 };
 
-struct aux_engine *dce110_aux_engine_construct(
+struct dce_aux *dce110_aux_engine_construct(
 		struct aux_engine_dce110 *aux_engine110,
 		struct dc_context *ctx,
 		uint32_t inst,
 		uint32_t timeout_period,
 		const struct dce110_aux_registers *regs);
 
-void dce110_engine_destroy(struct aux_engine **engine);
+void dce110_engine_destroy(struct dce_aux **engine);
 
 bool dce110_aux_engine_acquire(
-	struct aux_engine *aux_engine,
+	struct dce_aux *aux_engine,
 	struct ddc *ddc);
+
+int dce_aux_transfer(struct ddc_service *ddc,
+		struct aux_payload *cmd);
+
+bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
+		struct aux_payload *cmd);
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
index afd287f08bc9..bbe051736a18 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
@@ -194,8 +194,8 @@ static uint32_t get_max_pixel_clock_for_all_paths(struct dc_state *context)
 		if (pipe_ctx->top_pipe)
 			continue;
 
-		if (pipe_ctx->stream_res.pix_clk_params.requested_pix_clk > max_pix_clk)
-			max_pix_clk = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
+		if (pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10 > max_pix_clk)
+			max_pix_clk = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10;
 
 		/* raise clock state for HBR3/2 if required. Confirmed with HW DCE/DPCS
 		 * logic for HBR3 still needs Nominal (0.8V) on VDDC rail
@@ -257,7 +257,7 @@ static int dce_set_clock(
 				clk_mgr_dce->dentist_vco_freq_khz / 64);
 
 	/* Prepare to program display clock*/
-	pxl_clk_params.target_pixel_clock = requested_clk_khz;
+	pxl_clk_params.target_pixel_clock_100hz = requested_clk_khz * 10;
 	pxl_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
 
 	if (clk_mgr_dce->dfs_bypass_active)
@@ -450,6 +450,42 @@ void dce_clock_read_ss_info(struct dce_clk_mgr *clk_mgr_dce)
 	}
 }
 
+/**
+ * dce121_clock_patch_xgmi_ss_info() - Save XGMI spread spectrum info
+ * @clk_mgr: clock manager base structure
+ *
+ * Reads from VBIOS the XGMI spread spectrum info and saves it within
+ * the dce clock manager. This operation will overwrite the existing dprefclk
+ * SS values if the vBIOS query succeeds. Otherwise, it does nothing. It also
+ * sets the ->xgmi_enabled flag.
+ */
+void dce121_clock_patch_xgmi_ss_info(struct clk_mgr *clk_mgr)
+{
+	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+	enum bp_result result;
+	struct spread_spectrum_info info = { { 0 } };
+	struct dc_bios *bp = clk_mgr_dce->base.ctx->dc_bios;
+
+	clk_mgr_dce->xgmi_enabled = false;
+
+	result = bp->funcs->get_spread_spectrum_info(bp, AS_SIGNAL_TYPE_XGMI,
+						     0, &info);
+	if (result == BP_RESULT_OK && info.spread_spectrum_percentage != 0) {
+		clk_mgr_dce->xgmi_enabled = true;
+		clk_mgr_dce->ss_on_dprefclk = true;
+		clk_mgr_dce->dprefclk_ss_divider =
+				info.spread_percentage_divider;
+
+		if (info.type.CENTER_MODE == 0) {
+			/* Currently for DP Reference clock we
+			 * need only SS percentage for
+			 * downspread */
+			clk_mgr_dce->dprefclk_ss_percentage =
+					info.spread_spectrum_percentage;
+		}
+	}
+}
+
 void dce110_fill_display_configs(
 	const struct dc_state *context,
 	struct dm_pp_display_configuration *pp_display_cfg)
@@ -483,18 +519,18 @@ void dce110_fill_display_configs(
 		cfg->src_height = stream->src.height;
 		cfg->src_width = stream->src.width;
 		cfg->ddi_channel_mapping =
-			stream->sink->link->ddi_channel_mapping.raw;
+			stream->link->ddi_channel_mapping.raw;
 		cfg->transmitter =
-			stream->sink->link->link_enc->transmitter;
+			stream->link->link_enc->transmitter;
 		cfg->link_settings.lane_count =
-			stream->sink->link->cur_link_settings.lane_count;
+			stream->link->cur_link_settings.lane_count;
 		cfg->link_settings.link_rate =
-			stream->sink->link->cur_link_settings.link_rate;
+			stream->link->cur_link_settings.link_rate;
 		cfg->link_settings.link_spread =
-			stream->sink->link->cur_link_settings.link_spread;
+			stream->link->cur_link_settings.link_spread;
 		cfg->sym_clock = stream->phy_pix_clk;
 		/* Round v_refresh*/
-		cfg->v_refresh = stream->timing.pix_clk_khz * 1000;
+		cfg->v_refresh = stream->timing.pix_clk_100hz * 100;
 		cfg->v_refresh /= stream->timing.h_total;
 		cfg->v_refresh = (cfg->v_refresh + stream->timing.v_total / 2)
 							/ stream->timing.v_total;
@@ -518,7 +554,7 @@ static uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context)
 			 - stream->timing.v_addressable);
 
 		vertical_blank_time = vertical_blank_in_pixels
-			* 1000 / stream->timing.pix_clk_khz;
+			* 10000 / stream->timing.pix_clk_100hz;
 
 		if (min_vertical_blank_time > vertical_blank_time)
 			min_vertical_blank_time = vertical_blank_time;
@@ -591,7 +627,15 @@ static void dce11_pplib_apply_display_requirements(
 			dc,
 			context->bw.dce.sclk_khz);
 
-	pp_display_cfg->min_dcfclock_khz = pp_display_cfg->min_engine_clock_khz;
+	/*
+	 * As workaround for >4x4K lightup set dcfclock to min_engine_clock value.
+	 * This is not required for less than 5 displays,
+	 * thus don't request decfclk in dc to avoid impact
+	 * on power saving.
+	 *
+	 */
+	pp_display_cfg->min_dcfclock_khz = (context->stream_count > 4)?
+			pp_display_cfg->min_engine_clock_khz : 0;
 
 	pp_display_cfg->min_engine_clock_deep_sleep_khz
 			= context->bw.dce.sclk_deep_sleep_khz;
@@ -612,7 +656,7 @@ static void dce11_pplib_apply_display_requirements(
 
 		pp_display_cfg->crtc_index =
 			pp_display_cfg->disp_configs[0].pipe_idx;
-		pp_display_cfg->line_time_in_us = timing->h_total * 1000 / timing->pix_clk_khz;
+		pp_display_cfg->line_time_in_us = timing->h_total * 10000 / timing->pix_clk_100hz;
 	}
 
 	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)
@@ -625,11 +669,11 @@ static void dce_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_power_level_change_request level_change_req;
-	int unpatched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
-		context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
+		patched_disp_clk = patched_disp_clk * 115 / 100;
 
 	level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context);
 	/* get max clock state from PPLIB */
@@ -639,13 +683,11 @@ static void dce_update_clocks(struct clk_mgr *clk_mgr,
 			clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
 	}
 
-	if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
-		context->bw.dce.dispclk_khz = dce_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
-		clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz; 
+	if (should_set_clock(safe_to_lower, patched_disp_clk, clk_mgr->clks.dispclk_khz)) {
+		patched_disp_clk = dce_set_clock(clk_mgr, patched_disp_clk);
+		clk_mgr->clks.dispclk_khz = patched_disp_clk;
 	}
 	dce_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
-
-	context->bw.dce.dispclk_khz = unpatched_disp_clk;
 }
 
 static void dce11_update_clocks(struct clk_mgr *clk_mgr,
@@ -676,11 +718,11 @@ static void dce112_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_power_level_change_request level_change_req;
-	int unpatched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
-		context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
+		patched_disp_clk = patched_disp_clk * 115 / 100;
 
 	level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context);
 	/* get max clock state from PPLIB */
@@ -690,13 +732,11 @@ static void dce112_update_clocks(struct clk_mgr *clk_mgr,
 			clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
 	}
 
-	if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
-		context->bw.dce.dispclk_khz = dce112_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
-		clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+	if (should_set_clock(safe_to_lower, patched_disp_clk, clk_mgr->clks.dispclk_khz)) {
+		patched_disp_clk = dce112_set_clock(clk_mgr, patched_disp_clk);
+		clk_mgr->clks.dispclk_khz = patched_disp_clk;
 	}
 	dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
-
-	context->bw.dce.dispclk_khz = unpatched_disp_clk;
 }
 
 static void dce12_update_clocks(struct clk_mgr *clk_mgr,
@@ -706,17 +746,23 @@ static void dce12_update_clocks(struct clk_mgr *clk_mgr,
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
 	int max_pix_clk = get_max_pixel_clock_for_all_paths(context);
-	int unpatched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
-		context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
+		patched_disp_clk = patched_disp_clk * 115 / 100;
 
-	if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
+	if (should_set_clock(safe_to_lower, patched_disp_clk, clk_mgr->clks.dispclk_khz)) {
 		clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DISPLAY_CLK;
-		clock_voltage_req.clocks_in_khz = context->bw.dce.dispclk_khz;
-		context->bw.dce.dispclk_khz = dce112_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
-		clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+		/*
+		 * When xGMI is enabled, the display clk needs to be adjusted
+		 * with the WAFL link's SS percentage.
+		 */
+		if (clk_mgr_dce->xgmi_enabled)
+			patched_disp_clk = clk_mgr_adjust_dp_ref_freq_for_ss(
+					clk_mgr_dce, patched_disp_clk);
+		clock_voltage_req.clocks_in_khz = patched_disp_clk;
+		clk_mgr->clks.dispclk_khz = dce112_set_clock(clk_mgr, patched_disp_clk);
 
 		dm_pp_apply_clock_for_voltage_request(clk_mgr->ctx, &clock_voltage_req);
 	}
@@ -729,8 +775,6 @@ static void dce12_update_clocks(struct clk_mgr *clk_mgr,
 		dm_pp_apply_clock_for_voltage_request(clk_mgr->ctx, &clock_voltage_req);
 	}
 	dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
-
-	context->bw.dce.dispclk_khz = unpatched_disp_clk;
 }
 
 static const struct clk_mgr_funcs dce120_funcs = {
@@ -882,6 +926,27 @@ struct clk_mgr *dce120_clk_mgr_create(struct dc_context *ctx)
 	return &clk_mgr_dce->base;
 }
 
+struct clk_mgr *dce121_clk_mgr_create(struct dc_context *ctx)
+{
+	struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce),
+						  GFP_KERNEL);
+
+	if (clk_mgr_dce == NULL) {
+		BREAK_TO_DEBUGGER();
+		return NULL;
+	}
+
+	memcpy(clk_mgr_dce->max_clks_by_state, dce120_max_clks_by_state,
+	       sizeof(dce120_max_clks_by_state));
+
+	dce_clk_mgr_construct(clk_mgr_dce, ctx, NULL, NULL, NULL);
+
+	clk_mgr_dce->dprefclk_khz = 625000;
+	clk_mgr_dce->base.funcs = &dce120_funcs;
+
+	return &clk_mgr_dce->base;
+}
+
 void dce_clk_mgr_destroy(struct clk_mgr **clk_mgr)
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(*clk_mgr);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h
index 3bceb31d910d..c8f8c442142a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h
@@ -94,11 +94,37 @@ struct dce_clk_mgr {
 	 * This is basically "Crystal Frequency In KHz" (XTALIN) frequency */
 	int dfs_bypass_disp_clk;
 
-	/* Flag for Enabled SS on DPREFCLK */
+	/**
+	 * @ss_on_dprefclk:
+	 *
+	 * True if spread spectrum is enabled on the DP ref clock.
+	 */
 	bool ss_on_dprefclk;
-	/* DPREFCLK SS percentage (if down-spread enabled) */
+
+	/**
+	 * @xgmi_enabled:
+	 *
+	 * True if xGMI is enabled. On VG20, both audio and display clocks need
+	 * to be adjusted with the WAFL link's SS info if xGMI is enabled.
+	 */
+	bool xgmi_enabled;
+
+	/**
+	 * @dprefclk_ss_percentage:
+	 *
+	 * DPREFCLK SS percentage (if down-spread enabled).
+	 *
+	 * Note that if XGMI is enabled, the SS info (percentage and divider)
+	 * from the WAFL link is used instead. This is decided during
+	 * dce_clk_mgr initialization.
+	 */
 	int dprefclk_ss_percentage;
-	/* DPREFCLK SS percentage Divider (100 or 1000) */
+
+	/**
+	 * @dprefclk_ss_divider:
+	 *
+	 * DPREFCLK SS percentage Divider (100 or 1000).
+	 */
 	int dprefclk_ss_divider;
 	int dprefclk_khz;
 
@@ -163,6 +189,9 @@ struct clk_mgr *dce112_clk_mgr_create(
 
 struct clk_mgr *dce120_clk_mgr_create(struct dc_context *ctx);
 
+struct clk_mgr *dce121_clk_mgr_create(struct dc_context *ctx);
+void dce121_clock_patch_xgmi_ss_info(struct clk_mgr *clk_mgr);
+
 void dce_clk_mgr_destroy(struct clk_mgr **clk_mgr);
 
 int dentist_get_divider_from_did(int did);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
index 723ce80ed89c..71d5777de961 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
@@ -108,28 +108,28 @@ static const struct spread_spectrum_data *get_ss_data_entry(
 }
 
 /**
-* Function: calculate_fb_and_fractional_fb_divider
-*
-* * DESCRIPTION: Calculates feedback and fractional feedback dividers values
-*
-*PARAMETERS:
-* targetPixelClock             Desired frequency in 10 KHz
-* ref_divider                  Reference divider (already known)
-* postDivider                  Post Divider (already known)
-* feedback_divider_param       Pointer where to store
-*					calculated feedback divider value
-* fract_feedback_divider_param Pointer where to store
-*					calculated fract feedback divider value
-*
-*RETURNS:
-* It fills the locations pointed by feedback_divider_param
-*					and fract_feedback_divider_param
-* It returns	- true if feedback divider not 0
-*		- false should never happen)
-*/
+ * Function: calculate_fb_and_fractional_fb_divider
+ *
+ * * DESCRIPTION: Calculates feedback and fractional feedback dividers values
+ *
+ *PARAMETERS:
+ * targetPixelClock             Desired frequency in 100 Hz
+ * ref_divider                  Reference divider (already known)
+ * postDivider                  Post Divider (already known)
+ * feedback_divider_param       Pointer where to store
+ *					calculated feedback divider value
+ * fract_feedback_divider_param Pointer where to store
+ *					calculated fract feedback divider value
+ *
+ *RETURNS:
+ * It fills the locations pointed by feedback_divider_param
+ *					and fract_feedback_divider_param
+ * It returns	- true if feedback divider not 0
+ *		- false should never happen)
+ */
 static bool calculate_fb_and_fractional_fb_divider(
 		struct calc_pll_clock_source *calc_pll_cs,
-		uint32_t target_pix_clk_khz,
+		uint32_t target_pix_clk_100hz,
 		uint32_t ref_divider,
 		uint32_t post_divider,
 		uint32_t *feedback_divider_param,
@@ -138,11 +138,11 @@ static bool calculate_fb_and_fractional_fb_divider(
 	uint64_t feedback_divider;
 
 	feedback_divider =
-		(uint64_t)target_pix_clk_khz * ref_divider * post_divider;
+		(uint64_t)target_pix_clk_100hz * ref_divider * post_divider;
 	feedback_divider *= 10;
 	/* additional factor, since we divide by 10 afterwards */
 	feedback_divider *= (uint64_t)(calc_pll_cs->fract_fb_divider_factor);
-	feedback_divider = div_u64(feedback_divider, calc_pll_cs->ref_freq_khz);
+	feedback_divider = div_u64(feedback_divider, calc_pll_cs->ref_freq_khz * 10ull);
 
 /*Round to the number of precision
  * The following code replace the old code (ullfeedbackDivider + 5)/10
@@ -195,36 +195,36 @@ static bool calc_fb_divider_checking_tolerance(
 {
 	uint32_t feedback_divider;
 	uint32_t fract_feedback_divider;
-	uint32_t actual_calculated_clock_khz;
+	uint32_t actual_calculated_clock_100hz;
 	uint32_t abs_err;
-	uint64_t actual_calc_clk_khz;
+	uint64_t actual_calc_clk_100hz;
 
 	calculate_fb_and_fractional_fb_divider(
 			calc_pll_cs,
-			pll_settings->adjusted_pix_clk,
+			pll_settings->adjusted_pix_clk_100hz,
 			ref_divider,
 			post_divider,
 			&feedback_divider,
 			&fract_feedback_divider);
 
 	/*Actual calculated value*/
-	actual_calc_clk_khz = (uint64_t)feedback_divider *
+	actual_calc_clk_100hz = (uint64_t)feedback_divider *
 					calc_pll_cs->fract_fb_divider_factor +
 							fract_feedback_divider;
-	actual_calc_clk_khz *= calc_pll_cs->ref_freq_khz;
-	actual_calc_clk_khz =
-		div_u64(actual_calc_clk_khz,
+	actual_calc_clk_100hz *= calc_pll_cs->ref_freq_khz * 10;
+	actual_calc_clk_100hz =
+		div_u64(actual_calc_clk_100hz,
 			ref_divider * post_divider *
 				calc_pll_cs->fract_fb_divider_factor);
 
-	actual_calculated_clock_khz = (uint32_t)(actual_calc_clk_khz);
+	actual_calculated_clock_100hz = (uint32_t)(actual_calc_clk_100hz);
 
-	abs_err = (actual_calculated_clock_khz >
-					pll_settings->adjusted_pix_clk)
-			? actual_calculated_clock_khz -
-					pll_settings->adjusted_pix_clk
-			: pll_settings->adjusted_pix_clk -
-						actual_calculated_clock_khz;
+	abs_err = (actual_calculated_clock_100hz >
+					pll_settings->adjusted_pix_clk_100hz)
+			? actual_calculated_clock_100hz -
+					pll_settings->adjusted_pix_clk_100hz
+			: pll_settings->adjusted_pix_clk_100hz -
+						actual_calculated_clock_100hz;
 
 	if (abs_err <= tolerance) {
 		/*found good values*/
@@ -233,10 +233,10 @@ static bool calc_fb_divider_checking_tolerance(
 		pll_settings->feedback_divider = feedback_divider;
 		pll_settings->fract_feedback_divider = fract_feedback_divider;
 		pll_settings->pix_clk_post_divider = post_divider;
-		pll_settings->calculated_pix_clk =
-			actual_calculated_clock_khz;
+		pll_settings->calculated_pix_clk_100hz =
+			actual_calculated_clock_100hz;
 		pll_settings->vco_freq =
-			actual_calculated_clock_khz * post_divider;
+			actual_calculated_clock_100hz * post_divider / 10;
 		return true;
 	}
 	return false;
@@ -257,8 +257,8 @@ static bool calc_pll_dividers_in_range(
 
 /* This is err_tolerance / 10000 = 0.0025 - acceptable error of 0.25%
  * This is errorTolerance / 10000 = 0.0001 - acceptable error of 0.01%*/
-	tolerance = (pll_settings->adjusted_pix_clk * err_tolerance) /
-									10000;
+	tolerance = (pll_settings->adjusted_pix_clk_100hz * err_tolerance) /
+									100000;
 	if (tolerance < CALC_PLL_CLK_SRC_ERR_TOLERANCE)
 		tolerance = CALC_PLL_CLK_SRC_ERR_TOLERANCE;
 
@@ -294,7 +294,7 @@ static uint32_t calculate_pixel_clock_pll_dividers(
 	uint32_t min_ref_divider;
 	uint32_t max_ref_divider;
 
-	if (pll_settings->adjusted_pix_clk == 0) {
+	if (pll_settings->adjusted_pix_clk_100hz == 0) {
 		DC_LOG_ERROR(
 			"%s Bad requested pixel clock", __func__);
 		return MAX_PLL_CALC_ERROR;
@@ -306,21 +306,21 @@ static uint32_t calculate_pixel_clock_pll_dividers(
 		max_post_divider = pll_settings->pix_clk_post_divider;
 	} else {
 		min_post_divider = calc_pll_cs->min_pix_clock_pll_post_divider;
-		if (min_post_divider * pll_settings->adjusted_pix_clk <
-						calc_pll_cs->min_vco_khz) {
-			min_post_divider = calc_pll_cs->min_vco_khz /
-					pll_settings->adjusted_pix_clk;
+		if (min_post_divider * pll_settings->adjusted_pix_clk_100hz <
+						calc_pll_cs->min_vco_khz * 10) {
+			min_post_divider = calc_pll_cs->min_vco_khz * 10 /
+					pll_settings->adjusted_pix_clk_100hz;
 			if ((min_post_divider *
-					pll_settings->adjusted_pix_clk) <
-						calc_pll_cs->min_vco_khz)
+					pll_settings->adjusted_pix_clk_100hz) <
+						calc_pll_cs->min_vco_khz * 10)
 				min_post_divider++;
 		}
 
 		max_post_divider = calc_pll_cs->max_pix_clock_pll_post_divider;
-		if (max_post_divider * pll_settings->adjusted_pix_clk
-				> calc_pll_cs->max_vco_khz)
-			max_post_divider = calc_pll_cs->max_vco_khz /
-					pll_settings->adjusted_pix_clk;
+		if (max_post_divider * pll_settings->adjusted_pix_clk_100hz
+				> calc_pll_cs->max_vco_khz * 10)
+			max_post_divider = calc_pll_cs->max_vco_khz * 10 /
+					pll_settings->adjusted_pix_clk_100hz;
 	}
 
 /* 2) Find Reference divider ranges
@@ -392,47 +392,47 @@ static bool pll_adjust_pix_clk(
 		struct pixel_clk_params *pix_clk_params,
 		struct pll_settings *pll_settings)
 {
-	uint32_t actual_pix_clk_khz = 0;
-	uint32_t requested_clk_khz = 0;
+	uint32_t actual_pix_clk_100hz = 0;
+	uint32_t requested_clk_100hz = 0;
 	struct bp_adjust_pixel_clock_parameters bp_adjust_pixel_clock_params = {
 							0 };
 	enum bp_result bp_result;
 	switch (pix_clk_params->signal_type) {
 	case SIGNAL_TYPE_HDMI_TYPE_A: {
-		requested_clk_khz = pix_clk_params->requested_pix_clk;
+		requested_clk_100hz = pix_clk_params->requested_pix_clk_100hz;
 		if (pix_clk_params->pixel_encoding != PIXEL_ENCODING_YCBCR422) {
 			switch (pix_clk_params->color_depth) {
 			case COLOR_DEPTH_101010:
-				requested_clk_khz = (requested_clk_khz * 5) >> 2;
+				requested_clk_100hz = (requested_clk_100hz * 5) >> 2;
 				break; /* x1.25*/
 			case COLOR_DEPTH_121212:
-				requested_clk_khz = (requested_clk_khz * 6) >> 2;
+				requested_clk_100hz = (requested_clk_100hz * 6) >> 2;
 				break; /* x1.5*/
 			case COLOR_DEPTH_161616:
-				requested_clk_khz = requested_clk_khz * 2;
+				requested_clk_100hz = requested_clk_100hz * 2;
 				break; /* x2.0*/
 			default:
 				break;
 			}
 		}
-		actual_pix_clk_khz = requested_clk_khz;
+		actual_pix_clk_100hz = requested_clk_100hz;
 	}
 		break;
 
 	case SIGNAL_TYPE_DISPLAY_PORT:
 	case SIGNAL_TYPE_DISPLAY_PORT_MST:
 	case SIGNAL_TYPE_EDP:
-		requested_clk_khz = pix_clk_params->requested_sym_clk;
-		actual_pix_clk_khz = pix_clk_params->requested_pix_clk;
+		requested_clk_100hz = pix_clk_params->requested_sym_clk * 10;
+		actual_pix_clk_100hz = pix_clk_params->requested_pix_clk_100hz;
 		break;
 
 	default:
-		requested_clk_khz = pix_clk_params->requested_pix_clk;
-		actual_pix_clk_khz = pix_clk_params->requested_pix_clk;
+		requested_clk_100hz = pix_clk_params->requested_pix_clk_100hz;
+		actual_pix_clk_100hz = pix_clk_params->requested_pix_clk_100hz;
 		break;
 	}
 
-	bp_adjust_pixel_clock_params.pixel_clock = requested_clk_khz;
+	bp_adjust_pixel_clock_params.pixel_clock = requested_clk_100hz / 10;
 	bp_adjust_pixel_clock_params.
 		encoder_object_id = pix_clk_params->encoder_object_id;
 	bp_adjust_pixel_clock_params.signal_type = pix_clk_params->signal_type;
@@ -441,9 +441,9 @@ static bool pll_adjust_pix_clk(
 	bp_result = clk_src->bios->funcs->adjust_pixel_clock(
 			clk_src->bios, &bp_adjust_pixel_clock_params);
 	if (bp_result == BP_RESULT_OK) {
-		pll_settings->actual_pix_clk = actual_pix_clk_khz;
-		pll_settings->adjusted_pix_clk =
-			bp_adjust_pixel_clock_params.adjusted_pixel_clock;
+		pll_settings->actual_pix_clk_100hz = actual_pix_clk_100hz;
+		pll_settings->adjusted_pix_clk_100hz =
+			bp_adjust_pixel_clock_params.adjusted_pixel_clock * 10;
 		pll_settings->reference_divider =
 			bp_adjust_pixel_clock_params.reference_divider;
 		pll_settings->pix_clk_post_divider =
@@ -490,7 +490,7 @@ static uint32_t dce110_get_pix_clk_dividers_helper (
 		const struct spread_spectrum_data *ss_data = get_ss_data_entry(
 					clk_src,
 					pix_clk_params->signal_type,
-					pll_settings->adjusted_pix_clk);
+					pll_settings->adjusted_pix_clk_100hz / 10);
 
 		if (NULL != ss_data)
 			pll_settings->ss_percentage = ss_data->percentage;
@@ -502,13 +502,13 @@ static uint32_t dce110_get_pix_clk_dividers_helper (
 		 * to continue. */
 		DC_LOG_ERROR(
 			"%s: Failed to adjust pixel clock!!", __func__);
-		pll_settings->actual_pix_clk =
-				pix_clk_params->requested_pix_clk;
-		pll_settings->adjusted_pix_clk =
-				pix_clk_params->requested_pix_clk;
+		pll_settings->actual_pix_clk_100hz =
+				pix_clk_params->requested_pix_clk_100hz;
+		pll_settings->adjusted_pix_clk_100hz =
+				pix_clk_params->requested_pix_clk_100hz;
 
 		if (dc_is_dp_signal(pix_clk_params->signal_type))
-			pll_settings->adjusted_pix_clk = 100000;
+			pll_settings->adjusted_pix_clk_100hz = 1000000;
 	}
 
 	/* Calculate Dividers */
@@ -533,28 +533,28 @@ static void dce112_get_pix_clk_dividers_helper (
 		struct pll_settings *pll_settings,
 		struct pixel_clk_params *pix_clk_params)
 {
-	uint32_t actualPixelClockInKHz;
+	uint32_t actual_pixel_clock_100hz;
 
-	actualPixelClockInKHz = pix_clk_params->requested_pix_clk;
+	actual_pixel_clock_100hz = pix_clk_params->requested_pix_clk_100hz;
 	/* Calculate Dividers */
 	if (pix_clk_params->signal_type == SIGNAL_TYPE_HDMI_TYPE_A) {
 		switch (pix_clk_params->color_depth) {
 		case COLOR_DEPTH_101010:
-			actualPixelClockInKHz = (actualPixelClockInKHz * 5) >> 2;
+			actual_pixel_clock_100hz = (actual_pixel_clock_100hz * 5) >> 2;
 			break;
 		case COLOR_DEPTH_121212:
-			actualPixelClockInKHz = (actualPixelClockInKHz * 6) >> 2;
+			actual_pixel_clock_100hz = (actual_pixel_clock_100hz * 6) >> 2;
 			break;
 		case COLOR_DEPTH_161616:
-			actualPixelClockInKHz = actualPixelClockInKHz * 2;
+			actual_pixel_clock_100hz = actual_pixel_clock_100hz * 2;
 			break;
 		default:
 			break;
 		}
 	}
-	pll_settings->actual_pix_clk = actualPixelClockInKHz;
-	pll_settings->adjusted_pix_clk = actualPixelClockInKHz;
-	pll_settings->calculated_pix_clk = pix_clk_params->requested_pix_clk;
+	pll_settings->actual_pix_clk_100hz = actual_pixel_clock_100hz;
+	pll_settings->adjusted_pix_clk_100hz = actual_pixel_clock_100hz;
+	pll_settings->calculated_pix_clk_100hz = pix_clk_params->requested_pix_clk_100hz;
 }
 
 static uint32_t dce110_get_pix_clk_dividers(
@@ -567,7 +567,7 @@ static uint32_t dce110_get_pix_clk_dividers(
 	DC_LOGGER_INIT();
 
 	if (pix_clk_params == NULL || pll_settings == NULL
-			|| pix_clk_params->requested_pix_clk == 0) {
+			|| pix_clk_params->requested_pix_clk_100hz == 0) {
 		DC_LOG_ERROR(
 			"%s: Invalid parameters!!\n", __func__);
 		return pll_calc_error;
@@ -577,10 +577,10 @@ static uint32_t dce110_get_pix_clk_dividers(
 
 	if (cs->id == CLOCK_SOURCE_ID_DP_DTO ||
 			cs->id == CLOCK_SOURCE_ID_EXTERNAL) {
-		pll_settings->adjusted_pix_clk = clk_src->ext_clk_khz;
-		pll_settings->calculated_pix_clk = clk_src->ext_clk_khz;
-		pll_settings->actual_pix_clk =
-					pix_clk_params->requested_pix_clk;
+		pll_settings->adjusted_pix_clk_100hz = clk_src->ext_clk_khz * 10;
+		pll_settings->calculated_pix_clk_100hz = clk_src->ext_clk_khz * 10;
+		pll_settings->actual_pix_clk_100hz =
+					pix_clk_params->requested_pix_clk_100hz;
 		return 0;
 	}
 
@@ -599,7 +599,7 @@ static uint32_t dce112_get_pix_clk_dividers(
 	DC_LOGGER_INIT();
 
 	if (pix_clk_params == NULL || pll_settings == NULL
-			|| pix_clk_params->requested_pix_clk == 0) {
+			|| pix_clk_params->requested_pix_clk_100hz == 0) {
 		DC_LOG_ERROR(
 			"%s: Invalid parameters!!\n", __func__);
 		return -1;
@@ -609,10 +609,10 @@ static uint32_t dce112_get_pix_clk_dividers(
 
 	if (cs->id == CLOCK_SOURCE_ID_DP_DTO ||
 			cs->id == CLOCK_SOURCE_ID_EXTERNAL) {
-		pll_settings->adjusted_pix_clk = clk_src->ext_clk_khz;
-		pll_settings->calculated_pix_clk = clk_src->ext_clk_khz;
-		pll_settings->actual_pix_clk =
-					pix_clk_params->requested_pix_clk;
+		pll_settings->adjusted_pix_clk_100hz = clk_src->ext_clk_khz * 10;
+		pll_settings->calculated_pix_clk_100hz = clk_src->ext_clk_khz * 10;
+		pll_settings->actual_pix_clk_100hz =
+					pix_clk_params->requested_pix_clk_100hz;
 		return -1;
 	}
 
@@ -714,7 +714,7 @@ static bool enable_spread_spectrum(
 	ss_data = get_ss_data_entry(
 			clk_src,
 			signal,
-			pll_settings->calculated_pix_clk);
+			pll_settings->calculated_pix_clk_100hz / 10);
 
 /* Pixel clock PLL has been programmed to generate desired pixel clock,
  * now enable SS on pixel clock */
@@ -853,7 +853,7 @@ static bool dce110_program_pix_clk(
 	/*ATOMBIOS expects pixel rate adjusted by deep color ratio)*/
 	bp_pc_params.controller_id = pix_clk_params->controller_id;
 	bp_pc_params.pll_id = clock_source->id;
-	bp_pc_params.target_pixel_clock = pll_settings->actual_pix_clk;
+	bp_pc_params.target_pixel_clock_100hz = pll_settings->actual_pix_clk_100hz;
 	bp_pc_params.encoder_object_id = pix_clk_params->encoder_object_id;
 	bp_pc_params.signal_type = pix_clk_params->signal_type;
 
@@ -903,12 +903,12 @@ static bool dce112_program_pix_clk(
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
 	if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) {
 		unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0;
-		unsigned dp_dto_ref_kHz = 700000;
-		unsigned clock_kHz = pll_settings->actual_pix_clk;
+		unsigned dp_dto_ref_100hz = 7000000;
+		unsigned clock_100hz = pll_settings->actual_pix_clk_100hz;
 
 		/* Set DTO values: phase = target clock, modulo = reference clock */
-		REG_WRITE(PHASE[inst], clock_kHz);
-		REG_WRITE(MODULO[inst], dp_dto_ref_kHz);
+		REG_WRITE(PHASE[inst], clock_100hz);
+		REG_WRITE(MODULO[inst], dp_dto_ref_100hz);
 
 		/* Enable DTO */
 		REG_UPDATE(PIXEL_RATE_CNTL[inst], DP_DTO0_ENABLE, 1);
@@ -927,7 +927,7 @@ static bool dce112_program_pix_clk(
 	/*ATOMBIOS expects pixel rate adjusted by deep color ratio)*/
 	bp_pc_params.controller_id = pix_clk_params->controller_id;
 	bp_pc_params.pll_id = clock_source->id;
-	bp_pc_params.target_pixel_clock = pll_settings->actual_pix_clk;
+	bp_pc_params.target_pixel_clock_100hz = pll_settings->actual_pix_clk_100hz;
 	bp_pc_params.encoder_object_id = pix_clk_params->encoder_object_id;
 	bp_pc_params.signal_type = pix_clk_params->signal_type;
 
@@ -977,6 +977,28 @@ static bool dce110_clock_source_power_down(
 	return bp_result == BP_RESULT_OK;
 }
 
+static bool get_pixel_clk_frequency_100hz(
+		struct clock_source *clock_source,
+		unsigned int inst,
+		unsigned int *pixel_clk_khz)
+{
+	struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source);
+	unsigned int clock_hz = 0;
+
+	if (clock_source->id == CLOCK_SOURCE_ID_DP_DTO) {
+		clock_hz = REG_READ(PHASE[inst]);
+
+		/* NOTE: There is agreement with VBIOS here that MODULO is
+		 * programmed equal to DPREFCLK, in which case PHASE will be
+		 * equivalent to pixel clock.
+		 */
+		*pixel_clk_khz = clock_hz / 100;
+		return true;
+	}
+
+	return false;
+}
+
 /*****************************************/
 /* Constructor                           */
 /*****************************************/
@@ -984,12 +1006,14 @@ static bool dce110_clock_source_power_down(
 static const struct clock_source_funcs dce112_clk_src_funcs = {
 	.cs_power_down = dce110_clock_source_power_down,
 	.program_pix_clk = dce112_program_pix_clk,
-	.get_pix_clk_dividers = dce112_get_pix_clk_dividers
+	.get_pix_clk_dividers = dce112_get_pix_clk_dividers,
+	.get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz
 };
 static const struct clock_source_funcs dce110_clk_src_funcs = {
 	.cs_power_down = dce110_clock_source_power_down,
 	.program_pix_clk = dce110_program_pix_clk,
-	.get_pix_clk_dividers = dce110_get_pix_clk_dividers
+	.get_pix_clk_dividers = dce110_get_pix_clk_dividers,
+	.get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
index dea40b322191..c2926cf19dee 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
@@ -51,7 +51,6 @@
 #define PSR_SET_WAITLOOP 0x31
 #define MCP_INIT_DMCU 0x88
 #define MCP_INIT_IRAM 0x89
-#define MCP_DMCU_VERSION 0x90
 #define MASTER_COMM_CNTL_REG__MASTER_COMM_INTERRUPT_MASK   0x00000001L
 
 static bool dce_dmcu_init(struct dmcu *dmcu)
@@ -317,38 +316,11 @@ static void dce_get_psr_wait_loop(
 }
 
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
-static void dcn10_get_dmcu_state(struct dmcu *dmcu)
-{
-	struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu);
-	uint32_t dmcu_state_offset = 0xf6;
-
-	/* Enable write access to IRAM */
-	REG_UPDATE_2(DMCU_RAM_ACCESS_CTRL,
-			IRAM_HOST_ACCESS_EN, 1,
-			IRAM_RD_ADDR_AUTO_INC, 1);
-
-	REG_WAIT(DMU_MEM_PWR_CNTL, DMCU_IRAM_MEM_PWR_STATE, 0, 2, 10);
-
-	/* Write address to IRAM_RD_ADDR in DMCU_IRAM_RD_CTRL */
-	REG_WRITE(DMCU_IRAM_RD_CTRL, dmcu_state_offset);
-
-	/* Read data from IRAM_RD_DATA in DMCU_IRAM_RD_DATA*/
-	dmcu->dmcu_state = REG_READ(DMCU_IRAM_RD_DATA);
-
-	/* Disable write access to IRAM to allow dynamic sleep state */
-	REG_UPDATE_2(DMCU_RAM_ACCESS_CTRL,
-			IRAM_HOST_ACCESS_EN, 0,
-			IRAM_RD_ADDR_AUTO_INC, 0);
-}
-
 static void dcn10_get_dmcu_version(struct dmcu *dmcu)
 {
 	struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu);
 	uint32_t dmcu_version_offset = 0xf1;
 
-	/* Clear scratch */
-	REG_WRITE(DC_DMCU_SCRATCH, 0);
-
 	/* Enable write access to IRAM */
 	REG_UPDATE_2(DMCU_RAM_ACCESS_CTRL,
 			IRAM_HOST_ACCESS_EN, 1,
@@ -359,85 +331,74 @@ static void dcn10_get_dmcu_version(struct dmcu *dmcu)
 	/* Write address to IRAM_RD_ADDR and read from DATA register */
 	REG_WRITE(DMCU_IRAM_RD_CTRL, dmcu_version_offset);
 	dmcu->dmcu_version.interface_version = REG_READ(DMCU_IRAM_RD_DATA);
-	dmcu->dmcu_version.year = ((REG_READ(DMCU_IRAM_RD_DATA) << 8) |
+	dmcu->dmcu_version.abm_version = REG_READ(DMCU_IRAM_RD_DATA);
+	dmcu->dmcu_version.psr_version = REG_READ(DMCU_IRAM_RD_DATA);
+	dmcu->dmcu_version.build_version = ((REG_READ(DMCU_IRAM_RD_DATA) << 8) |
 						REG_READ(DMCU_IRAM_RD_DATA));
-	dmcu->dmcu_version.month = REG_READ(DMCU_IRAM_RD_DATA);
-	dmcu->dmcu_version.date = REG_READ(DMCU_IRAM_RD_DATA);
 
 	/* Disable write access to IRAM to allow dynamic sleep state */
 	REG_UPDATE_2(DMCU_RAM_ACCESS_CTRL,
 			IRAM_HOST_ACCESS_EN, 0,
 			IRAM_RD_ADDR_AUTO_INC, 0);
-
-	/* Send MCP command message to DMCU to get version reply from FW.
-	 * We expect this version should match the one in IRAM, otherwise
-	 * something is wrong with DMCU and we should fail and disable UC.
-	 */
-	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
-
-	/* Set command to get DMCU version from microcontroller */
-	REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0,
-			MCP_DMCU_VERSION);
-
-	/* Notify microcontroller of new command */
-	REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
-
-	/* Ensure command has been executed before continuing */
-	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
-
-	/* Somehow version does not match, so fail and return version 0 */
-	if (dmcu->dmcu_version.interface_version != REG_READ(DC_DMCU_SCRATCH))
-		dmcu->dmcu_version.interface_version = 0;
 }
 
 static bool dcn10_dmcu_init(struct dmcu *dmcu)
 {
 	struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu);
+	bool status = false;
 
-	/* DMCU FW should populate the scratch register if running */
-	if (REG_READ(DC_DMCU_SCRATCH) == 0)
-		return false;
-
-	/* Check state is uninitialized */
-	dcn10_get_dmcu_state(dmcu);
-
-	/* If microcontroller is already initialized, do nothing */
-	if (dmcu->dmcu_state == DMCU_RUNNING)
-		return true;
-
-	/* Retrieve and cache the DMCU firmware version. */
-	dcn10_get_dmcu_version(dmcu);
-
-	/* Check interface version to confirm firmware is loaded and running */
-	if (dmcu->dmcu_version.interface_version == 0)
-		return false;
+	/*  Definition of DC_DMCU_SCRATCH
+	 *  0 : firmare not loaded
+	 *  1 : PSP load DMCU FW but not initialized
+	 *  2 : Firmware already initialized
+	 */
+	dmcu->dmcu_state = REG_READ(DC_DMCU_SCRATCH);
 
-	/* Wait until microcontroller is ready to process interrupt */
-	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
+	switch (dmcu->dmcu_state) {
+	case DMCU_UNLOADED:
+		status = false;
+		break;
+	case DMCU_LOADED_UNINITIALIZED:
+		/* Wait until microcontroller is ready to process interrupt */
+		REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
 
-	/* Set initialized ramping boundary value */
-	REG_WRITE(MASTER_COMM_DATA_REG1, 0xFFFF);
+		/* Set initialized ramping boundary value */
+		REG_WRITE(MASTER_COMM_DATA_REG1, 0xFFFF);
 
-	/* Set command to initialize microcontroller */
-	REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0,
+		/* Set command to initialize microcontroller */
+		REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0,
 			MCP_INIT_DMCU);
 
-	/* Notify microcontroller of new command */
-	REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
+		/* Notify microcontroller of new command */
+		REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
 
-	/* Ensure command has been executed before continuing */
-	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
+		/* Ensure command has been executed before continuing */
+		REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
 
-	// Check state is initialized
-	dcn10_get_dmcu_state(dmcu);
+		// Check state is initialized
+		dmcu->dmcu_state = REG_READ(DC_DMCU_SCRATCH);
 
-	// If microcontroller is not in running state, fail
-	if (dmcu->dmcu_state != DMCU_RUNNING)
-		return false;
+		// If microcontroller is not in running state, fail
+		if (dmcu->dmcu_state == DMCU_RUNNING) {
+			/* Retrieve and cache the DMCU firmware version. */
+			dcn10_get_dmcu_version(dmcu);
+			status = true;
+		} else
+			status = false;
 
-	return true;
+		break;
+	case DMCU_RUNNING:
+		status = true;
+		break;
+	default:
+		status = false;
+		break;
+	}
+
+	return status;
 }
 
+
 static bool dcn10_dmcu_load_iram(struct dmcu *dmcu,
 		unsigned int start_offset,
 		const char *src,
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 c83a7f05f14c..956bdf14503f 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
@@ -133,6 +133,10 @@
 	SR(DCHUB_AGP_TOP), \
 	BL_REG_LIST()
 
+#define HWSEQ_VG20_REG_LIST() \
+	HWSEQ_DCE120_REG_LIST(),\
+	MMHUB_SR(MC_VM_XGMI_LFB_CNTL)
+
 #define HWSEQ_DCE112_REG_LIST() \
 	HWSEQ_DCE10_REG_LIST(), \
 	HWSEQ_PIXEL_RATE_REG_LIST(CRTC), \
@@ -298,6 +302,7 @@ struct dce_hwseq_registers {
 	uint32_t MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;
 	uint32_t MC_VM_SYSTEM_APERTURE_LOW_ADDR;
 	uint32_t MC_VM_SYSTEM_APERTURE_HIGH_ADDR;
+	uint32_t MC_VM_XGMI_LFB_CNTL;
 	uint32_t AZALIA_AUDIO_DTO;
 	uint32_t AZALIA_CONTROLLER_CLOCK_GATING;
 };
@@ -382,6 +387,11 @@ struct dce_hwseq_registers {
 	HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \
 	HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh)
 
+#define HWSEQ_VG20_MASK_SH_LIST(mask_sh)\
+	HWSEQ_DCE12_MASK_SH_LIST(mask_sh),\
+	HWS_SF(, MC_VM_XGMI_LFB_CNTL, PF_LFB_REGION, mask_sh),\
+	HWS_SF(, MC_VM_XGMI_LFB_CNTL, PF_MAX_REGION, mask_sh)
+
 #define HWSEQ_DCN_MASK_SH_LIST(mask_sh)\
 	HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, OTG0_),\
 	HWS_SF1(OTG0_, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh), \
@@ -470,6 +480,8 @@ struct dce_hwseq_registers {
 	type PHYSICAL_PAGE_NUMBER_MSB;\
 	type PHYSICAL_PAGE_NUMBER_LSB;\
 	type LOGICAL_ADDR; \
+	type PF_LFB_REGION;\
+	type PF_MAX_REGION;\
 	type ENABLE_L1_TLB;\
 	type SYSTEM_ACCESS_MODE;\
 	type LVTMA_BLON;\
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 3e18ea84b1f9..314c04a915d2 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
@@ -599,12 +599,12 @@ bool dce110_link_encoder_validate_dvi_output(
 	if ((connector_signal == SIGNAL_TYPE_DVI_SINGLE_LINK ||
 		connector_signal == SIGNAL_TYPE_HDMI_TYPE_A) &&
 		signal != SIGNAL_TYPE_HDMI_TYPE_A &&
-		crtc_timing->pix_clk_khz > TMDS_MAX_PIXEL_CLOCK)
+		crtc_timing->pix_clk_100hz > (TMDS_MAX_PIXEL_CLOCK * 10))
 		return false;
-	if (crtc_timing->pix_clk_khz < TMDS_MIN_PIXEL_CLOCK)
+	if (crtc_timing->pix_clk_100hz < (TMDS_MIN_PIXEL_CLOCK * 10))
 		return false;
 
-	if (crtc_timing->pix_clk_khz > max_pixel_clock)
+	if (crtc_timing->pix_clk_100hz > (max_pixel_clock * 10))
 		return false;
 
 	/* DVI supports 6/8bpp single-link and 10/16bpp dual-link */
@@ -788,7 +788,7 @@ bool dce110_link_encoder_validate_output_with_stream(
 	case SIGNAL_TYPE_DVI_DUAL_LINK:
 		is_valid = dce110_link_encoder_validate_dvi_output(
 			enc110,
-			stream->sink->link->connector_signal,
+			stream->link->connector_signal,
 			stream->signal,
 			&stream->timing);
 	break;
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 cce0d18f91da..1fa2d4fd7a35 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
@@ -288,9 +288,18 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 #endif
 
 	struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc);
-
+	struct dc_crtc_timing hw_crtc_timing = *crtc_timing;
+	if (hw_crtc_timing.flags.INTERLACE) {
+		/*the input timing is in VESA spec format with Interlace flag =1*/
+		hw_crtc_timing.v_total /= 2;
+		hw_crtc_timing.v_border_top /= 2;
+		hw_crtc_timing.v_addressable /= 2;
+		hw_crtc_timing.v_border_bottom /= 2;
+		hw_crtc_timing.v_front_porch /= 2;
+		hw_crtc_timing.v_sync_width /= 2;
+	}
 	/* set pixel encoding */
-	switch (crtc_timing->pixel_encoding) {
+	switch (hw_crtc_timing.pixel_encoding) {
 	case PIXEL_ENCODING_YCBCR422:
 		REG_UPDATE(DP_PIXEL_FORMAT, DP_PIXEL_ENCODING,
 				DP_PIXEL_ENCODING_TYPE_YCBCR422);
@@ -299,8 +308,8 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 		REG_UPDATE(DP_PIXEL_FORMAT, DP_PIXEL_ENCODING,
 				DP_PIXEL_ENCODING_TYPE_YCBCR444);
 
-		if (crtc_timing->flags.Y_ONLY)
-			if (crtc_timing->display_color_depth != COLOR_DEPTH_666)
+		if (hw_crtc_timing.flags.Y_ONLY)
+			if (hw_crtc_timing.display_color_depth != COLOR_DEPTH_666)
 				/* HW testing only, no use case yet.
 				 * Color depth of Y-only could be
 				 * 8, 10, 12, 16 bits */
@@ -335,7 +344,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 
 	/* set color depth */
 
-	switch (crtc_timing->display_color_depth) {
+	switch (hw_crtc_timing.display_color_depth) {
 	case COLOR_DEPTH_666:
 		REG_UPDATE(DP_PIXEL_FORMAT, DP_COMPONENT_DEPTH,
 				0);
@@ -363,7 +372,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 
 
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
-	switch (crtc_timing->display_color_depth) {
+	switch (hw_crtc_timing.display_color_depth) {
 	case COLOR_DEPTH_666:
 		colorimetry_bpc = 0;
 		break;
@@ -401,9 +410,9 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 			misc0 = misc0 | 0x8; /* bit3=1, bit4=0 */
 			misc1 = misc1 & ~0x80; /* bit7 = 0*/
 			dynamic_range_ycbcr = 0; /*bt601*/
-			if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+			if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR422)
 				misc0 = misc0 | 0x2; /* bit2=0, bit1=1 */
-			else if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR444)
+			else if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR444)
 				misc0 = misc0 | 0x4; /* bit2=1, bit1=0 */
 			break;
 		case COLOR_SPACE_YCBCR709:
@@ -411,9 +420,9 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 			misc0 = misc0 | 0x18; /* bit3=1, bit4=1 */
 			misc1 = misc1 & ~0x80; /* bit7 = 0*/
 			dynamic_range_ycbcr = 1; /*bt709*/
-			if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+			if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR422)
 				misc0 = misc0 | 0x2; /* bit2=0, bit1=1 */
-			else if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR444)
+			else if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR444)
 				misc0 = misc0 | 0x4; /* bit2=1, bit1=0 */
 			break;
 		case COLOR_SPACE_2020_RGB_LIMITEDRANGE:
@@ -453,27 +462,27 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 	 */
 		if (REG(DP_MSA_TIMING_PARAM1))
 			REG_SET_2(DP_MSA_TIMING_PARAM1, 0,
-					DP_MSA_HTOTAL, crtc_timing->h_total,
-					DP_MSA_VTOTAL, crtc_timing->v_total);
+					DP_MSA_HTOTAL, hw_crtc_timing.h_total,
+					DP_MSA_VTOTAL, hw_crtc_timing.v_total);
 #endif
 
 		/* calcuate from vesa timing parameters
 		 * h_active_start related to leading edge of sync
 		 */
 
-		h_blank = crtc_timing->h_total - crtc_timing->h_border_left -
-				crtc_timing->h_addressable - crtc_timing->h_border_right;
+		h_blank = hw_crtc_timing.h_total - hw_crtc_timing.h_border_left -
+				hw_crtc_timing.h_addressable - hw_crtc_timing.h_border_right;
 
-		h_back_porch = h_blank - crtc_timing->h_front_porch -
-				crtc_timing->h_sync_width;
+		h_back_porch = h_blank - hw_crtc_timing.h_front_porch -
+				hw_crtc_timing.h_sync_width;
 
 		/* start at begining of left border */
-		h_active_start = crtc_timing->h_sync_width + h_back_porch;
+		h_active_start = hw_crtc_timing.h_sync_width + h_back_porch;
 
 
-		v_active_start = crtc_timing->v_total - crtc_timing->v_border_top -
-				crtc_timing->v_addressable - crtc_timing->v_border_bottom -
-				crtc_timing->v_front_porch;
+		v_active_start = hw_crtc_timing.v_total - hw_crtc_timing.v_border_top -
+				hw_crtc_timing.v_addressable - hw_crtc_timing.v_border_bottom -
+				hw_crtc_timing.v_front_porch;
 
 
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
@@ -486,21 +495,21 @@ static void dce110_stream_encoder_dp_set_stream_attribute(
 		if (REG(DP_MSA_TIMING_PARAM3))
 			REG_SET_4(DP_MSA_TIMING_PARAM3, 0,
 					DP_MSA_HSYNCWIDTH,
-					crtc_timing->h_sync_width,
+					hw_crtc_timing.h_sync_width,
 					DP_MSA_HSYNCPOLARITY,
-					!crtc_timing->flags.HSYNC_POSITIVE_POLARITY,
+					!hw_crtc_timing.flags.HSYNC_POSITIVE_POLARITY,
 					DP_MSA_VSYNCWIDTH,
-					crtc_timing->v_sync_width,
+					hw_crtc_timing.v_sync_width,
 					DP_MSA_VSYNCPOLARITY,
-					!crtc_timing->flags.VSYNC_POSITIVE_POLARITY);
+					!hw_crtc_timing.flags.VSYNC_POSITIVE_POLARITY);
 
 		/* HWDITH include border or overscan */
 		if (REG(DP_MSA_TIMING_PARAM4))
 			REG_SET_2(DP_MSA_TIMING_PARAM4, 0,
-				DP_MSA_HWIDTH, crtc_timing->h_border_left +
-				crtc_timing->h_addressable + crtc_timing->h_border_right,
-				DP_MSA_VHEIGHT, crtc_timing->v_border_top +
-				crtc_timing->v_addressable + crtc_timing->v_border_bottom);
+				DP_MSA_HWIDTH, hw_crtc_timing.h_border_left +
+				hw_crtc_timing.h_addressable + hw_crtc_timing.h_border_right,
+				DP_MSA_VHEIGHT, hw_crtc_timing.v_border_top +
+				hw_crtc_timing.v_addressable + hw_crtc_timing.v_border_bottom);
 #endif
 	}
 #endif
@@ -662,7 +671,7 @@ static void dce110_stream_encoder_dvi_set_stream_attribute(
 	cntl.signal = is_dual_link ?
 			SIGNAL_TYPE_DVI_DUAL_LINK : SIGNAL_TYPE_DVI_SINGLE_LINK;
 	cntl.enable_dp_audio = false;
-	cntl.pixel_clock = crtc_timing->pix_clk_khz;
+	cntl.pixel_clock = crtc_timing->pix_clk_100hz / 10;
 	cntl.lanes_number = (is_dual_link) ? LANE_COUNT_EIGHT : LANE_COUNT_FOUR;
 
 	if (enc110->base.bp->funcs->encoder_control(
@@ -686,7 +695,7 @@ static void dce110_stream_encoder_lvds_set_stream_attribute(
 	cntl.engine_id = enc110->base.id;
 	cntl.signal = SIGNAL_TYPE_LVDS;
 	cntl.enable_dp_audio = false;
-	cntl.pixel_clock = crtc_timing->pix_clk_khz;
+	cntl.pixel_clock = crtc_timing->pix_clk_100hz / 10;
 	cntl.lanes_number = LANE_COUNT_FOUR;
 
 	if (enc110->base.bp->funcs->encoder_control(
@@ -1575,6 +1584,14 @@ static void setup_stereo_sync(
 	REG_UPDATE(DIG_FE_CNTL, DIG_STEREOSYNC_GATE_EN, !enable);
 }
 
+static void dig_connect_to_otg(
+	struct stream_encoder *enc,
+	int tg_inst)
+{
+	struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc);
+
+	REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst);
+}
 
 static const struct stream_encoder_funcs dce110_str_enc_funcs = {
 	.dp_set_stream_attribute =
@@ -1609,7 +1626,7 @@ static const struct stream_encoder_funcs dce110_str_enc_funcs = {
 	.hdmi_audio_disable = dce110_se_hdmi_audio_disable,
 	.setup_stereo_sync  = setup_stereo_sync,
 	.set_avmute = dce110_stream_encoder_set_avmute,
-
+	.dig_connect_to_otg  = dig_connect_to_otg,
 };
 
 void dce110_stream_encoder_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.h
index 6c28229c76eb..f9cdf2b5242c 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.h
@@ -199,7 +199,8 @@
 	SE_SF(DP_SEC_CNTL, DP_SEC_ATP_ENABLE, mask_sh),\
 	SE_SF(DP_SEC_CNTL, DP_SEC_AIP_ENABLE, mask_sh),\
 	SE_SF(DP_SEC_CNTL, DP_SEC_ACM_ENABLE, mask_sh),\
-	SE_SF(AFMT_AUDIO_PACKET_CONTROL, AFMT_AUDIO_SAMPLE_SEND, mask_sh)
+	SE_SF(AFMT_AUDIO_PACKET_CONTROL, AFMT_AUDIO_SAMPLE_SEND, mask_sh),\
+	SE_SF(DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh)
 
 #define SE_COMMON_MASK_SH_LIST_DCE_COMMON(mask_sh)\
 	SE_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh)
@@ -284,7 +285,8 @@
 	SE_SF(DIG0_DIG_FE_CNTL, TMDS_PIXEL_ENCODING, mask_sh),\
 	SE_SF(DIG0_DIG_FE_CNTL, TMDS_COLOR_FORMAT, mask_sh),\
 	SE_SF(DIG0_DIG_FE_CNTL, DIG_STEREOSYNC_SELECT, mask_sh),\
-	SE_SF(DIG0_DIG_FE_CNTL, DIG_STEREOSYNC_GATE_EN, mask_sh)
+	SE_SF(DIG0_DIG_FE_CNTL, DIG_STEREOSYNC_GATE_EN, mask_sh),\
+	SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh)
 
 #define SE_COMMON_MASK_SH_LIST_SOC(mask_sh)\
 	SE_COMMON_MASK_SH_LIST_SOC_BASE(mask_sh)
@@ -494,6 +496,7 @@ struct dce_stream_encoder_shift {
 	uint8_t HDMI_DB_DISABLE;
 	uint8_t DP_VID_N_MUL;
 	uint8_t DP_VID_M_DOUBLE_VALUE_EN;
+	uint8_t DIG_SOURCE_SELECT;
 };
 
 struct dce_stream_encoder_mask {
@@ -624,6 +627,7 @@ struct dce_stream_encoder_mask {
 	uint32_t HDMI_DB_DISABLE;
 	uint32_t DP_VID_N_MUL;
 	uint32_t DP_VID_M_DOUBLE_VALUE_EN;
+	uint32_t DIG_SOURCE_SELECT;
 };
 
 struct dce110_stream_enc_registers {
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 6ae51a5dfc04..23044e6723e8 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -76,6 +76,7 @@
 
 #ifndef mmBIOS_SCRATCH_2
 	#define mmBIOS_SCRATCH_2 0x05CB
+	#define mmBIOS_SCRATCH_3 0x05CC
 	#define mmBIOS_SCRATCH_6 0x05CF
 #endif
 
@@ -365,6 +366,7 @@ static const struct dce_abm_mask abm_mask = {
 #define DCFE_MEM_PWR_CTRL_REG_BASE 0x1b03
 
 static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3,
 	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6
 };
 
@@ -587,7 +589,7 @@ struct output_pixel_processor *dce100_opp_create(
 	return &opp->base;
 }
 
-struct aux_engine *dce100_aux_engine_create(
+struct dce_aux *dce100_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c
index 52d50e24a995..7b23239d33fe 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c
@@ -62,8 +62,6 @@ static const struct dce110_compressor_reg_offsets reg_offsets[] = {
 }
 };
 
-static const uint32_t dce11_one_lpt_channel_max_resolution = 2560 * 1600;
-
 static uint32_t align_to_chunks_number_per_line(uint32_t pixels)
 {
 	return 256 * ((pixels + 255) / 256);
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 4bf24758217f..e1b285ea01ac 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
@@ -614,55 +614,6 @@ dce110_set_output_transfer_func(struct pipe_ctx *pipe_ctx,
 	return true;
 }
 
-static enum dc_status bios_parser_crtc_source_select(
-		struct pipe_ctx *pipe_ctx)
-{
-	struct dc_bios *dcb;
-	/* call VBIOS table to set CRTC source for the HW
-	 * encoder block
-	 * note: video bios clears all FMT setting here. */
-	struct bp_crtc_source_select crtc_source_select = {0};
-	const struct dc_sink *sink = pipe_ctx->stream->sink;
-
-	crtc_source_select.engine_id = pipe_ctx->stream_res.stream_enc->id;
-	crtc_source_select.controller_id = pipe_ctx->stream_res.tg->inst + 1;
-	/*TODO: Need to un-hardcode color depth, dp_audio and account for
-	 * the case where signal and sink signal is different (translator
-	 * encoder)*/
-	crtc_source_select.signal = pipe_ctx->stream->signal;
-	crtc_source_select.enable_dp_audio = false;
-	crtc_source_select.sink_signal = pipe_ctx->stream->signal;
-
-	switch (pipe_ctx->stream->timing.display_color_depth) {
-	case COLOR_DEPTH_666:
-		crtc_source_select.display_output_bit_depth = PANEL_6BIT_COLOR;
-		break;
-	case COLOR_DEPTH_888:
-		crtc_source_select.display_output_bit_depth = PANEL_8BIT_COLOR;
-		break;
-	case COLOR_DEPTH_101010:
-		crtc_source_select.display_output_bit_depth = PANEL_10BIT_COLOR;
-		break;
-	case COLOR_DEPTH_121212:
-		crtc_source_select.display_output_bit_depth = PANEL_12BIT_COLOR;
-		break;
-	default:
-		BREAK_TO_DEBUGGER();
-		crtc_source_select.display_output_bit_depth = PANEL_8BIT_COLOR;
-		break;
-	}
-
-	dcb = sink->ctx->dc_bios;
-
-	if (BP_RESULT_OK != dcb->funcs->crtc_source_select(
-		dcb,
-		&crtc_source_select)) {
-		return DC_ERROR_UNEXPECTED;
-	}
-
-	return DC_OK;
-}
-
 void dce110_update_info_frame(struct pipe_ctx *pipe_ctx)
 {
 	bool is_hdmi;
@@ -692,10 +643,10 @@ void dce110_update_info_frame(struct pipe_ctx *pipe_ctx)
 void dce110_enable_stream(struct pipe_ctx *pipe_ctx)
 {
 	enum dc_lane_count lane_count =
-		pipe_ctx->stream->sink->link->cur_link_settings.lane_count;
+		pipe_ctx->stream->link->cur_link_settings.lane_count;
 
 	struct dc_crtc_timing *timing = &pipe_ctx->stream->timing;
-	struct dc_link *link = pipe_ctx->stream->sink->link;
+	struct dc_link *link = pipe_ctx->stream->link;
 
 
 	uint32_t active_total_with_borders;
@@ -1000,7 +951,7 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
 
 		pipe_ctx->stream_res.audio->funcs->az_enable(pipe_ctx->stream_res.audio);
 
-		if (num_audio == 1 && pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+		if (num_audio >= 1 && pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
 			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
 			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
 		/* un-mute audio */
@@ -1017,6 +968,8 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 	pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control(
 			pipe_ctx->stream_res.stream_enc, true);
 	if (pipe_ctx->stream_res.audio) {
+		struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
+
 		if (option != KEEP_ACQUIRED_RESOURCE ||
 				!dc->debug.az_endpoint_mute_only) {
 			/*only disalbe az_endpoint if power down or free*/
@@ -1036,6 +989,9 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false);
 			pipe_ctx->stream_res.audio = NULL;
 		}
+		if (pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
+			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
 
 		/* TODO: notify audio driver for if audio modes list changed
 		 * add audio mode list change flag */
@@ -1048,7 +1004,7 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 	struct dc *dc = pipe_ctx->stream->ctx->dc;
 
 	if (dc_is_hdmi_signal(pipe_ctx->stream->signal))
@@ -1073,11 +1029,10 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx,
 {
 	struct encoder_unblank_param params = { { 0 } };
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 
 	/* only 3 items below are used by unblank */
-	params.pixel_clk_khz =
-		pipe_ctx->stream->timing.pix_clk_khz;
+	params.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10;
 	params.link_settings.link_rate = link_settings->link_rate;
 
 	if (dc_is_dp_signal(pipe_ctx->stream->signal))
@@ -1087,10 +1042,11 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx,
 		link->dc->hwss.edp_backlight_control(link, true);
 	}
 }
+
 void dce110_blank_stream(struct pipe_ctx *pipe_ctx)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->sink->link;
+	struct dc_link *link = stream->link;
 
 	if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
 		link->dc->hwss.edp_backlight_control(link, false);
@@ -1163,27 +1119,27 @@ static void build_audio_output(
 			stream->timing.flags.INTERLACE;
 
 	audio_output->crtc_info.refresh_rate =
-		(stream->timing.pix_clk_khz*1000)/
+		(stream->timing.pix_clk_100hz*10000)/
 		(stream->timing.h_total*stream->timing.v_total);
 
 	audio_output->crtc_info.color_depth =
 		stream->timing.display_color_depth;
 
 	audio_output->crtc_info.requested_pixel_clock =
-			pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
+			pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10;
 
 	audio_output->crtc_info.calculated_pixel_clock =
-			pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
+			pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10;
 
 /*for HDMI, audio ACR is with deep color ratio factor*/
 	if (dc_is_hdmi_signal(pipe_ctx->stream->signal) &&
 		audio_output->crtc_info.requested_pixel_clock ==
-				stream->timing.pix_clk_khz) {
+				(stream->timing.pix_clk_100hz / 10)) {
 		if (pipe_ctx->stream_res.pix_clk_params.pixel_encoding == PIXEL_ENCODING_YCBCR420) {
 			audio_output->crtc_info.requested_pixel_clock =
 					audio_output->crtc_info.requested_pixel_clock/2;
 			audio_output->crtc_info.calculated_pixel_clock =
-					pipe_ctx->stream_res.pix_clk_params.requested_pix_clk/2;
+					pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz/20;
 
 		}
 	}
@@ -1294,8 +1250,6 @@ static enum dc_status dce110_enable_stream_timing(
 	struct pipe_ctx *pipe_ctx_old = &dc->current_state->res_ctx.
 			pipe_ctx[pipe_ctx->pipe_idx];
 	struct tg_color black_color = {0};
-	struct drr_params params = {0};
-	unsigned int event_triggers = 0;
 
 	if (!pipe_ctx_old->stream) {
 
@@ -1324,20 +1278,6 @@ static enum dc_status dce110_enable_stream_timing(
 				pipe_ctx->stream_res.tg,
 				&stream->timing,
 				true);
-
-		params.vertical_total_min = stream->adjust.v_total_min;
-		params.vertical_total_max = stream->adjust.v_total_max;
-		if (pipe_ctx->stream_res.tg->funcs->set_drr)
-			pipe_ctx->stream_res.tg->funcs->set_drr(
-				pipe_ctx->stream_res.tg, &params);
-
-		// DRR should set trigger event to monitor surface update event
-		if (stream->adjust.v_total_min != 0 &&
-				stream->adjust.v_total_max != 0)
-			event_triggers = 0x80;
-		if (pipe_ctx->stream_res.tg->funcs->set_static_screen_control)
-			pipe_ctx->stream_res.tg->funcs->set_static_screen_control(
-				pipe_ctx->stream_res.tg, event_triggers);
 	}
 
 	if (!pipe_ctx_old->stream) {
@@ -1357,6 +1297,8 @@ static enum dc_status apply_single_controller_ctx_to_hw(
 		struct dc *dc)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
+	struct drr_params params = {0};
+	unsigned int event_triggers = 0;
 
 	if (pipe_ctx->stream_res.audio != NULL) {
 		struct audio_output audio_output;
@@ -1383,14 +1325,32 @@ static enum dc_status apply_single_controller_ctx_to_hw(
 	}
 
 	/*  */
-	dc->hwss.enable_stream_timing(pipe_ctx, context, dc);
+	/* Do not touch stream timing on seamless boot optimization. */
+	if (!pipe_ctx->stream->apply_seamless_boot_optimization)
+		dc->hwss.enable_stream_timing(pipe_ctx, context, dc);
+
+	if (pipe_ctx->stream_res.tg->funcs->program_vupdate_interrupt)
+		pipe_ctx->stream_res.tg->funcs->program_vupdate_interrupt(
+						pipe_ctx->stream_res.tg,
+							&stream->timing);
+
+	params.vertical_total_min = stream->adjust.v_total_min;
+	params.vertical_total_max = stream->adjust.v_total_max;
+	if (pipe_ctx->stream_res.tg->funcs->set_drr)
+		pipe_ctx->stream_res.tg->funcs->set_drr(
+			pipe_ctx->stream_res.tg, &params);
+
+	// DRR should set trigger event to monitor surface update event
+	if (stream->adjust.v_total_min != 0 && stream->adjust.v_total_max != 0)
+		event_triggers = 0x80;
+	if (pipe_ctx->stream_res.tg->funcs->set_static_screen_control)
+		pipe_ctx->stream_res.tg->funcs->set_static_screen_control(
+				pipe_ctx->stream_res.tg, event_triggers);
 
-	/* TODO: move to stream encoder */
 	if (pipe_ctx->stream->signal != SIGNAL_TYPE_VIRTUAL)
-		if (DC_OK != bios_parser_crtc_source_select(pipe_ctx)) {
-			BREAK_TO_DEBUGGER();
-			return DC_ERROR_UNEXPECTED;
-		}
+		pipe_ctx->stream_res.stream_enc->funcs->dig_connect_to_otg(
+			pipe_ctx->stream_res.stream_enc,
+			pipe_ctx->stream_res.tg->inst);
 
 	pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion(
 			pipe_ctx->stream_res.opp,
@@ -1408,7 +1368,7 @@ static enum dc_status apply_single_controller_ctx_to_hw(
 
 	pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0;
 
-	pipe_ctx->stream->sink->link->psr_enabled = false;
+	pipe_ctx->stream->link->psr_enabled = false;
 
 	return DC_OK;
 }
@@ -1518,7 +1478,7 @@ static struct dc_link *get_link_for_edp(struct dc *dc)
 	return NULL;
 }
 
-static struct dc_link *get_link_for_edp_not_in_use(
+static struct dc_link *get_link_for_edp_to_turn_off(
 		struct dc *dc,
 		struct dc_state *context)
 {
@@ -1527,8 +1487,12 @@ static struct dc_link *get_link_for_edp_not_in_use(
 
 	/* check if eDP panel is suppose to be set mode, if yes, no need to disable */
 	for (i = 0; i < context->stream_count; i++) {
-		if (context->streams[i]->signal == SIGNAL_TYPE_EDP)
-			return NULL;
+		if (context->streams[i]->signal == SIGNAL_TYPE_EDP) {
+			if (context->streams[i]->dpms_off == true)
+				return context->streams[i]->sink->link;
+			else
+				return NULL;
+		}
 	}
 
 	/* check if there is an eDP panel not in use */
@@ -1555,7 +1519,6 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 	int i;
 	struct dc_link *edp_link_to_turnoff = NULL;
 	struct dc_link *edp_link = get_link_for_edp(dc);
-	struct dc_bios *bios = dc->ctx->dc_bios;
 	bool can_edp_fast_boot_optimize = false;
 	bool apply_edp_fast_boot_optimization = false;
 
@@ -1571,7 +1534,7 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 	}
 
 	if (can_edp_fast_boot_optimize)
-		edp_link_to_turnoff = get_link_for_edp_not_in_use(dc, context);
+		edp_link_to_turnoff = get_link_for_edp_to_turn_off(dc, context);
 
 	/* if OS doesn't light up eDP and eDP link is available, we want to disable
 	 * If resume from S4/S5, should optimization.
@@ -1582,20 +1545,6 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 			if (context->streams[i]->signal == SIGNAL_TYPE_EDP) {
 				context->streams[i]->apply_edp_fast_boot_optimization = true;
 				apply_edp_fast_boot_optimization = true;
-
-				/* When after S4 and S5, vbios may post edp and previous dpms_off
-				 * doesn't make sense.
-				 * Update dpms_off state to align hw and sw state via check
-				 * vBios scratch register.
-				 */
-				if (bios->funcs->is_active_display)	{
-					const struct connector_device_tag_info *device_tag = &(edp_link->device_tag);
-
-					if (bios->funcs->is_active_display(bios,
-							context->streams[i]->signal,
-							device_tag))
-						context->streams[i]->dpms_off = false;
-				}
 			}
 		}
 	}
@@ -1624,8 +1573,8 @@ static uint32_t compute_pstate_blackout_duration(
 	pstate_blackout_duration_ns = 1000 * blackout_duration.value >> 24;
 
 	total_dest_line_time_ns = 1000000UL *
-		stream->timing.h_total /
-		stream->timing.pix_clk_khz +
+		(stream->timing.h_total * 10) /
+		stream->timing.pix_clk_100hz +
 		pstate_blackout_duration_ns;
 
 	return total_dest_line_time_ns;
@@ -1813,18 +1762,15 @@ static bool should_enable_fbc(struct dc *dc,
 	if (i == dc->res_pool->pipe_count)
 		return false;
 
-	if (!pipe_ctx->stream->sink)
-		return false;
-
-	if (!pipe_ctx->stream->sink->link)
+	if (!pipe_ctx->stream->link)
 		return false;
 
 	/* Only supports eDP */
-	if (pipe_ctx->stream->sink->link->connector_signal != SIGNAL_TYPE_EDP)
+	if (pipe_ctx->stream->link->connector_signal != SIGNAL_TYPE_EDP)
 		return false;
 
 	/* PSR should not be enabled */
-	if (pipe_ctx->stream->sink->link->psr_enabled)
+	if (pipe_ctx->stream->link->psr_enabled)
 		return false;
 
 	/* Nothing to compress */
@@ -2329,6 +2275,11 @@ static void dce110_enable_per_frame_crtc_position_reset(
 
 }
 
+static void init_pipes(struct dc *dc, struct dc_state *context)
+{
+	// Do nothing
+}
+
 static void init_hw(struct dc *dc)
 {
 	int i;
@@ -2573,7 +2524,7 @@ static void dce110_apply_ctx_for_surface(
 				pipe_ctx->plane_res.mi,
 				pipe_ctx->stream->timing.h_total,
 				pipe_ctx->stream->timing.v_total,
-				pipe_ctx->stream->timing.pix_clk_khz,
+				pipe_ctx->stream->timing.pix_clk_100hz / 10,
 				context->stream_count);
 
 		dce110_program_front_end_for_pipe(dc, pipe_ctx);
@@ -2595,7 +2546,7 @@ static void dce110_apply_ctx_for_surface(
 	}
 
 	if (dc->fbc_compressor)
-		enable_fbc(dc, dc->current_state);
+		enable_fbc(dc, context);
 }
 
 static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx)
@@ -2622,13 +2573,35 @@ static void dce110_wait_for_mpcc_disconnect(
 	/* do nothing*/
 }
 
+static void program_output_csc(struct dc *dc,
+		struct pipe_ctx *pipe_ctx,
+		enum dc_color_space colorspace,
+		uint16_t *matrix,
+		int opp_id)
+{
+	int i;
+	struct out_csc_color_matrix tbl_entry;
+
+	if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) {
+		enum dc_color_space 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];
+
+		tbl_entry.color_space = color_space;
+
+		pipe_ctx->plane_res.xfm->funcs->opp_set_csc_adjustment(
+				pipe_ctx->plane_res.xfm, &tbl_entry);
+	}
+}
+
 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;
 	struct mem_input *mi = pipe_ctx->plane_res.mi;
 	struct dc_cursor_mi_param param = {
-		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_khz,
+		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10,
 		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clock_inKhz,
 		.viewport = pipe_ctx->plane_res.scl_data.viewport,
 		.h_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.horz,
@@ -2672,7 +2645,9 @@ void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx)
 
 static const struct hw_sequencer_funcs dce110_funcs = {
 	.program_gamut_remap = program_gamut_remap,
+	.program_output_csc = program_output_csc,
 	.init_hw = init_hw,
+	.init_pipes = init_pipes,
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = dce110_apply_ctx_for_surface,
 	.update_plane_addr = update_plane_addr,
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
index e33d11785b1f..7549adaa1542 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
@@ -84,6 +84,7 @@
 
 #ifndef mmBIOS_SCRATCH_2
 	#define mmBIOS_SCRATCH_2 0x05CB
+	#define mmBIOS_SCRATCH_3 0x05CC
 	#define mmBIOS_SCRATCH_6 0x05CF
 #endif
 
@@ -369,6 +370,7 @@ static const struct dce110_clk_src_mask cs_mask = {
 };
 
 static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3,
 	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6
 };
 
@@ -606,7 +608,7 @@ static struct output_pixel_processor *dce110_opp_create(
 	return &opp->base;
 }
 
-struct aux_engine *dce110_aux_engine_create(
+struct dce_aux *dce110_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -779,8 +781,8 @@ static void get_pixel_clock_parameters(
 	 * the pixel clock normalization for hdmi up to here instead of doing it
 	 * in pll_adjust_pix_clk
 	 */
-	pixel_clk_params->requested_pix_clk = stream->timing.pix_clk_khz;
-	pixel_clk_params->encoder_object_id = stream->sink->link->link_enc->id;
+	pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz;
+	pixel_clk_params->encoder_object_id = stream->link->link_enc->id;
 	pixel_clk_params->signal_type = pipe_ctx->stream->signal;
 	pixel_clk_params->controller_id = pipe_ctx->stream_res.tg->inst + 1;
 	/* TODO: un-hardcode*/
@@ -797,10 +799,10 @@ static void get_pixel_clock_parameters(
 		pixel_clk_params->color_depth = COLOR_DEPTH_888;
 	}
 	if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) {
-		pixel_clk_params->requested_pix_clk  = pixel_clk_params->requested_pix_clk / 2;
+		pixel_clk_params->requested_pix_clk_100hz  = pixel_clk_params->requested_pix_clk_100hz / 2;
 	}
 	if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
-		pixel_clk_params->requested_pix_clk *= 2;
+		pixel_clk_params->requested_pix_clk_100hz *= 2;
 
 }
 
@@ -874,7 +876,7 @@ static bool dce110_validate_bandwidth(
 			__func__,
 			context->streams[0]->timing.h_addressable,
 			context->streams[0]->timing.v_addressable,
-			context->streams[0]->timing.pix_clk_khz);
+			context->streams[0]->timing.pix_clk_100hz / 10);
 
 	if (memcmp(&dc->current_state->bw.dce,
 			&context->bw.dce, sizeof(context->bw.dce))) {
@@ -1055,7 +1057,7 @@ static struct pipe_ctx *dce110_acquire_underlay(
 		pipe_ctx->plane_res.mi->funcs->allocate_mem_input(pipe_ctx->plane_res.mi,
 				stream->timing.h_total,
 				stream->timing.v_total,
-				stream->timing.pix_clk_khz,
+				stream->timing.pix_clk_100hz / 10,
 				context->stream_count);
 
 		color_space_to_black_color(dc,
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 969d4e72dc94..ea3065d63372 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -76,6 +76,7 @@
 
 #ifndef mmBIOS_SCRATCH_2
 	#define mmBIOS_SCRATCH_2 0x05CB
+	#define mmBIOS_SCRATCH_3 0x05CC
 	#define mmBIOS_SCRATCH_6 0x05CF
 #endif
 
@@ -376,6 +377,7 @@ static const struct dce110_clk_src_mask cs_mask = {
 };
 
 static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3,
 	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6
 };
 
@@ -607,7 +609,7 @@ struct output_pixel_processor *dce112_opp_create(
 	return &opp->base;
 }
 
-struct aux_engine *dce112_aux_engine_create(
+struct dce_aux *dce112_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -763,7 +765,7 @@ static struct clock_source *find_matching_pll(
 		const struct resource_pool *pool,
 		const struct dc_stream_state *const stream)
 {
-	switch (stream->sink->link->link_enc->transmitter) {
+	switch (stream->link->link_enc->transmitter) {
 	case TRANSMITTER_UNIPHY_A:
 		return pool->clock_sources[DCE112_CLK_SRC_PLL0];
 	case TRANSMITTER_UNIPHY_B:
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
index eb0f5f9a973b..1ca30928025e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
@@ -244,6 +244,21 @@ static void dce120_update_dchub(
 	dh_data->dchub_info_valid = false;
 }
 
+/**
+ * dce121_xgmi_enabled() - Check if xGMI is enabled
+ * @hws: DCE hardware sequencer object
+ *
+ * Return true if xGMI is enabled. False otherwise.
+ */
+bool dce121_xgmi_enabled(struct dce_hwseq *hws)
+{
+	uint32_t pf_max_region;
+
+	REG_GET(MC_VM_XGMI_LFB_CNTL, PF_MAX_REGION, &pf_max_region);
+	/* PF_MAX_REGION == 0 means xgmi is disabled */
+	return !!pf_max_region;
+}
+
 void dce120_hw_sequencer_construct(struct dc *dc)
 {
 	/* All registers used by dce11.2 match those in dce11 in offset and
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h
index 77a6b86d7606..c51afbd0b012 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h
@@ -30,6 +30,7 @@
 
 struct dc;
 
+bool dce121_xgmi_enabled(struct dce_hwseq *hws);
 void dce120_hw_sequencer_construct(struct dc *dc);
 
 #endif /* __DC_HWSS_DCE112_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
index f12696674eb0..312a0aebf91f 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
@@ -62,6 +62,8 @@
 #include "soc15_hw_ip.h"
 #include "vega10_ip_offset.h"
 #include "nbio/nbio_6_1_offset.h"
+#include "mmhub/mmhub_9_4_0_offset.h"
+#include "mmhub/mmhub_9_4_0_sh_mask.h"
 #include "reg_helper.h"
 
 #include "dce100/dce100_resource.h"
@@ -139,6 +141,17 @@ static const struct dce110_timing_generator_offsets dce120_tg_offsets[] = {
 	.reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
 					mm ## block ## id ## _ ## reg_name
 
+/* MMHUB */
+#define MMHUB_BASE_INNER(seg) \
+	MMHUB_BASE__INST0_SEG ## seg
+
+#define MMHUB_BASE(seg) \
+	MMHUB_BASE_INNER(seg)
+
+#define MMHUB_SR(reg_name)\
+		.reg_name = MMHUB_BASE(mm ## reg_name ## _BASE_IDX) +  \
+					mm ## reg_name
+
 /* macros to expend register list macro defined in HW object header file
  * end *********************/
 
@@ -378,7 +391,7 @@ struct output_pixel_processor *dce120_opp_create(
 			     ctx, inst, &opp_regs[inst], &opp_shift, &opp_mask);
 	return &opp->base;
 }
-struct aux_engine *dce120_aux_engine_create(
+struct dce_aux *dce120_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -429,6 +442,7 @@ struct dce_i2c_hw *dce120_i2c_hw_create(
 	return dce_i2c_hw;
 }
 static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3 + NBIO_BASE(mmBIOS_SCRATCH_3_BASE_IDX),
 	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6 + NBIO_BASE(mmBIOS_SCRATCH_6_BASE_IDX)
 };
 
@@ -681,6 +695,19 @@ static const struct dce_hwseq_mask hwseq_mask = {
 		HWSEQ_DCE12_MASK_SH_LIST(_MASK)
 };
 
+/* HWSEQ regs for VG20 */
+static const struct dce_hwseq_registers dce121_hwseq_reg = {
+		HWSEQ_VG20_REG_LIST()
+};
+
+static const struct dce_hwseq_shift dce121_hwseq_shift = {
+		HWSEQ_VG20_MASK_SH_LIST(__SHIFT)
+};
+
+static const struct dce_hwseq_mask dce121_hwseq_mask = {
+		HWSEQ_VG20_MASK_SH_LIST(_MASK)
+};
+
 static struct dce_hwseq *dce120_hwseq_create(
 	struct dc_context *ctx)
 {
@@ -695,6 +722,20 @@ static struct dce_hwseq *dce120_hwseq_create(
 	return hws;
 }
 
+static struct dce_hwseq *dce121_hwseq_create(
+	struct dc_context *ctx)
+{
+	struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL);
+
+	if (hws) {
+		hws->ctx = ctx;
+		hws->regs = &dce121_hwseq_reg;
+		hws->shifts = &dce121_hwseq_shift;
+		hws->masks = &dce121_hwseq_mask;
+	}
+	return hws;
+}
+
 static const struct resource_create_funcs res_create_funcs = {
 	.read_dce_straps = read_dce_straps,
 	.create_audio = create_audio,
@@ -702,6 +743,14 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dce120_hwseq_create,
 };
 
+static const struct resource_create_funcs dce121_res_create_funcs = {
+	.read_dce_straps = read_dce_straps,
+	.create_audio = create_audio,
+	.create_stream_encoder = dce120_stream_encoder_create,
+	.create_hwseq = dce121_hwseq_create,
+};
+
+
 #define mi_inst_regs(id) { MI_DCE12_REG_LIST(id) }
 static const struct dce_mem_input_registers mi_regs[] = {
 		mi_inst_regs(0),
@@ -911,7 +960,8 @@ static bool construct(
 	int j;
 	struct dc_context *ctx = dc->ctx;
 	struct irq_service_init_data irq_init_data;
-	bool harvest_enabled = ASICREV_IS_VEGA20_P(ctx->asic_id.hw_internal_rev);
+	static const struct resource_create_funcs *res_funcs;
+	bool is_vg20 = ASICREV_IS_VEGA20_P(ctx->asic_id.hw_internal_rev);
 	uint32_t pipe_fuses;
 
 	ctx->dc_bios->regs = &bios_regs;
@@ -975,7 +1025,11 @@ static bool construct(
 		}
 	}
 
-	pool->base.clk_mgr = dce120_clk_mgr_create(ctx);
+	if (is_vg20)
+		pool->base.clk_mgr = dce121_clk_mgr_create(ctx);
+	else
+		pool->base.clk_mgr = dce120_clk_mgr_create(ctx);
+
 	if (pool->base.clk_mgr == NULL) {
 		dm_error("DC: failed to create display clock!\n");
 		BREAK_TO_DEBUGGER();
@@ -1008,14 +1062,14 @@ static bool construct(
 	if (!pool->base.irqs)
 		goto irqs_create_fail;
 
-	/* retrieve valid pipe fuses */
-	if (harvest_enabled)
+	/* VG20: Pipe harvesting enabled, retrieve valid pipe fuses */
+	if (is_vg20)
 		pipe_fuses = read_pipe_fuses(ctx);
 
 	/* index to valid pipe resource */
 	j = 0;
 	for (i = 0; i < pool->base.pipe_count; i++) {
-		if (harvest_enabled) {
+		if (is_vg20) {
 			if ((pipe_fuses & (1 << i)) != 0) {
 				dm_error("DC: skip invalid pipe %d!\n", i);
 				continue;
@@ -1093,10 +1147,24 @@ static bool construct(
 	pool->base.pipe_count = j;
 	pool->base.timing_generator_count = j;
 
-	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			 &res_create_funcs))
+	if (is_vg20)
+		res_funcs = &dce121_res_create_funcs;
+	else
+		res_funcs = &res_create_funcs;
+
+	if (!resource_construct(num_virtual_links, dc, &pool->base, res_funcs))
 		goto res_create_fail;
 
+	/*
+	 * This is a bit of a hack. The xGMI enabled info is used to determine
+	 * if audio and display clocks need to be adjusted with the WAFL link's
+	 * SS info. This is a responsiblity of the clk_mgr. But since MMHUB is
+	 * under hwseq, and the relevant register is in MMHUB, we have to do it
+	 * here.
+	 */
+	if (is_vg20 && dce121_xgmi_enabled(dc->hwseq))
+		dce121_clock_patch_xgmi_ss_info(pool->base.clk_mgr);
+
 	/* Create hardware sequencer */
 	if (!dce120_hw_sequencer_create(dc))
 		goto controller_create_fail;
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
index cdd1d6b7b9f2..2eca81b5cf2f 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
@@ -77,6 +77,7 @@
 
 #ifndef mmBIOS_SCRATCH_2
 	#define mmBIOS_SCRATCH_2 0x05CB
+	#define mmBIOS_SCRATCH_3 0x05CC
 	#define mmBIOS_SCRATCH_6 0x05CF
 #endif
 
@@ -358,6 +359,7 @@ static const struct dce110_clk_src_mask cs_mask = {
 };
 
 static const struct bios_registers bios_regs = {
+	.BIOS_SCRATCH_3 = mmBIOS_SCRATCH_3,
 	.BIOS_SCRATCH_6 = mmBIOS_SCRATCH_6
 };
 
@@ -467,7 +469,7 @@ static struct output_pixel_processor *dce80_opp_create(
 	return &opp->base;
 }
 
-struct aux_engine *dce80_aux_engine_create(
+struct dce_aux *dce80_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
index 3ba4712a35ab..8b5ce557ee71 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
@@ -84,17 +84,17 @@ static const struct dce110_timing_generator_offsets reg_offsets[] = {
 #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_khz)
+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_khz == 0)
+	if (pix_clk_100hz == 0)
 		return;
 
-	pix_dur = 1000000000 / pix_clk_khz;
+	pix_dur = div_u64(10000000000ull, pix_clk_100hz);
 
 	set_reg_field_value(
 		value,
@@ -110,7 +110,7 @@ static void program_timing(struct timing_generator *tg,
 	bool use_vbios)
 {
 	if (!use_vbios)
-		program_pix_dur(tg, timing->pix_clk_khz);
+		program_pix_dur(tg, timing->pix_clk_100hz);
 
 	dce110_tg_program_timing(tg, timing, use_vbios);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
index 54abedbf1b43..afe8c42211cd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
@@ -161,69 +161,17 @@ static int get_active_display_cnt(
 	return display_count;
 }
 
-static void notify_deep_sleep_dcfclk_to_smu(
-		struct pp_smu_funcs_rv *pp_smu, int min_dcef_deep_sleep_clk_khz)
-{
-	int min_dcef_deep_sleep_clk_mhz; //minimum required DCEF Deep Sleep clock in mhz
-	/*
-	 * if function pointer not set up, this message is
-	 * sent as part of pplib_apply_display_requirements.
-	 * So just return.
-	 */
-	if (!pp_smu || !pp_smu->set_min_deep_sleep_dcfclk)
-		return;
-
-	min_dcef_deep_sleep_clk_mhz = (min_dcef_deep_sleep_clk_khz + 999) / 1000; //Round up
-	pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, min_dcef_deep_sleep_clk_mhz);
-}
-
-static void notify_hard_min_dcfclk_to_smu(
-		struct pp_smu_funcs_rv *pp_smu, int min_dcf_clk_khz)
-{
-	int min_dcf_clk_mhz; //minimum required DCF clock in mhz
-
-	/*
-	 * if function pointer not set up, this message is
-	 * sent as part of pplib_apply_display_requirements.
-	 * So just return.
-	 */
-	if (!pp_smu || !pp_smu->set_hard_min_dcfclk_by_freq)
-		return;
-
-	min_dcf_clk_mhz = min_dcf_clk_khz / 1000;
-
-	pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, min_dcf_clk_mhz);
-}
-
-static void notify_hard_min_fclk_to_smu(
-		struct pp_smu_funcs_rv *pp_smu, int min_f_clk_khz)
-{
-	int min_f_clk_mhz; //minimum required F clock in mhz
-
-	/*
-	 * if function pointer not set up, this message is
-	 * sent as part of pplib_apply_display_requirements.
-	 * So just return.
-	 */
-	if (!pp_smu || !pp_smu->set_hard_min_fclk_by_freq)
-		return;
-
-	min_f_clk_mhz = min_f_clk_khz / 1000;
-
-	pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, min_f_clk_mhz);
-}
-
 static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 			struct dc_state *context,
 			bool safe_to_lower)
 {
 	struct dc *dc = clk_mgr->ctx->dc;
+	struct dc_debug_options *debug = &dc->debug;
 	struct dc_clocks *new_clocks = &context->bw.dcn.clk;
 	struct pp_smu_display_requirement_rv *smu_req_cur =
 			&dc->res_pool->pp_smu_req;
 	struct pp_smu_display_requirement_rv smu_req = *smu_req_cur;
 	struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
-	uint32_t requested_dcf_clock_in_khz = 0;
 	bool send_request_to_increase = false;
 	bool send_request_to_lower = false;
 	int display_count;
@@ -243,9 +191,8 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 		 */
 		if (pp_smu->set_display_count)
 			pp_smu->set_display_count(&pp_smu->pp_smu, display_count);
-		else
-			smu_req.display_count = display_count;
 
+		smu_req.display_count = display_count;
 	}
 
 	if (new_clocks->dispclk_khz > clk_mgr->clks.dispclk_khz
@@ -261,12 +208,13 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 	}
 
 	// F Clock
+	if (debug->force_fclk_khz != 0)
+		new_clocks->fclk_khz = debug->force_fclk_khz;
+
 	if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr->clks.fclk_khz)) {
 		clk_mgr->clks.fclk_khz = new_clocks->fclk_khz;
 		smu_req.hard_min_fclk_mhz = new_clocks->fclk_khz / 1000;
 
-		notify_hard_min_fclk_to_smu(pp_smu, new_clocks->fclk_khz);
-
 		send_request_to_lower = true;
 	}
 
@@ -281,7 +229,7 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 	if (should_set_clock(safe_to_lower,
 			new_clocks->dcfclk_deep_sleep_khz, clk_mgr->clks.dcfclk_deep_sleep_khz)) {
 		clk_mgr->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
-		smu_req.min_deep_sleep_dcefclk_mhz = new_clocks->dcfclk_deep_sleep_khz / 1000;
+		smu_req.min_deep_sleep_dcefclk_mhz = (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000;
 
 		send_request_to_lower = true;
 	}
@@ -291,15 +239,18 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 	 */
 	if (send_request_to_increase) {
 		/*use dcfclk to request voltage*/
-		requested_dcf_clock_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
-
-		notify_hard_min_dcfclk_to_smu(pp_smu, requested_dcf_clock_in_khz);
-
-		if (pp_smu->set_display_requirement)
-			pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
-
-		notify_deep_sleep_dcfclk_to_smu(pp_smu, clk_mgr->clks.dcfclk_deep_sleep_khz);
-		dcn1_pplib_apply_display_requirements(dc, context);
+		if (pp_smu->set_hard_min_fclk_by_freq &&
+				pp_smu->set_hard_min_dcfclk_by_freq &&
+				pp_smu->set_min_deep_sleep_dcfclk) {
+
+			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz);
+			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz);
+			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz);
+		} else {
+			if (pp_smu->set_display_requirement)
+				pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
+			dcn1_pplib_apply_display_requirements(dc, context);
+		}
 	}
 
 	/* dcn1 dppclk is tied to dispclk */
@@ -314,18 +265,20 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 
 	if (!send_request_to_increase && send_request_to_lower) {
 		/*use dcfclk to request voltage*/
-		requested_dcf_clock_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
-
-		notify_hard_min_dcfclk_to_smu(pp_smu, requested_dcf_clock_in_khz);
-
-		if (pp_smu->set_display_requirement)
-			pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
-
-		notify_deep_sleep_dcfclk_to_smu(pp_smu, clk_mgr->clks.dcfclk_deep_sleep_khz);
-		dcn1_pplib_apply_display_requirements(dc, context);
+		if (pp_smu->set_hard_min_fclk_by_freq &&
+				pp_smu->set_hard_min_dcfclk_by_freq &&
+				pp_smu->set_min_deep_sleep_dcfclk) {
+
+			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz);
+			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz);
+			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz);
+		} else {
+			if (pp_smu->set_display_requirement)
+				pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
+			dcn1_pplib_apply_display_requirements(dc, context);
+		}
 	}
 
-
 	*smu_req_cur = smu_req;
 }
 static const struct clk_mgr_funcs dcn1_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
index dcb3c5530236..cd1ebe57ed59 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
@@ -463,7 +463,7 @@ void dpp1_set_cursor_position(
 	if (src_y_offset >= (int)param->viewport.height)
 		cur_en = 0;  /* not visible beyond bottom edge*/
 
-	if (src_y_offset < 0)
+	if (src_y_offset + (int)height <= 0)
 		cur_en = 0;  /* not visible beyond top edge*/
 
 	REG_UPDATE(CURSOR0_CONTROL,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
index 116977eb24e2..41f0f4c912e7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
@@ -51,10 +51,6 @@
 
 #define NUM_ELEMENTS(a) (sizeof(a) / sizeof((a)[0]))
 
-struct dcn10_input_csc_matrix {
-	enum dc_color_space color_space;
-	uint16_t regval[12];
-};
 
 enum dcn10_coef_filter_type_sel {
 	SCL_COEF_LUMA_VERT_FILTER = 0,
@@ -99,7 +95,7 @@ enum gamut_remap_select {
 	GAMUT_REMAP_COMB_COEFF
 };
 
-static const struct dcn10_input_csc_matrix dcn10_input_csc_matrix[] = {
+static const struct dpp_input_csc_matrix dpp_input_csc_matrix[] = {
 	{COLOR_SPACE_SRGB,
 		{0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} },
 	{COLOR_SPACE_SRGB_LIMITED,
@@ -454,7 +450,7 @@ void dpp1_program_input_csc(
 {
 	struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base);
 	int i;
-	int arr_size = sizeof(dcn10_input_csc_matrix)/sizeof(struct dcn10_input_csc_matrix);
+	int arr_size = sizeof(dpp_input_csc_matrix)/sizeof(struct dpp_input_csc_matrix);
 	const uint16_t *regval = NULL;
 	uint32_t cur_select = 0;
 	enum dcn10_input_csc_select select;
@@ -467,8 +463,8 @@ void dpp1_program_input_csc(
 
 	if (tbl_entry == NULL) {
 		for (i = 0; i < arr_size; i++)
-			if (dcn10_input_csc_matrix[i].color_space == color_space) {
-				regval = dcn10_input_csc_matrix[i].regval;
+			if (dpp_input_csc_matrix[i].color_space == color_space) {
+				regval = dpp_input_csc_matrix[i].regval;
 				break;
 			}
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
index 4a863a5dab41..c7642e748297 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
@@ -597,11 +597,13 @@ static void dpp1_dscl_set_manual_ratio_init(
 		SCL_V_INIT_FRAC, init_frac,
 		SCL_V_INIT_INT, init_int);
 
-	init_frac = dc_fixpt_u0d19(data->inits.v_bot) << 5;
-	init_int = dc_fixpt_floor(data->inits.v_bot);
-	REG_SET_2(SCL_VERT_FILTER_INIT_BOT, 0,
-		SCL_V_INIT_FRAC_BOT, init_frac,
-		SCL_V_INIT_INT_BOT, init_int);
+	if (REG(SCL_VERT_FILTER_INIT_BOT)) {
+		init_frac = dc_fixpt_u0d19(data->inits.v_bot) << 5;
+		init_int = dc_fixpt_floor(data->inits.v_bot);
+		REG_SET_2(SCL_VERT_FILTER_INIT_BOT, 0,
+			SCL_V_INIT_FRAC_BOT, init_frac,
+			SCL_V_INIT_INT_BOT, init_int);
+	}
 
 	init_frac = dc_fixpt_u0d19(data->inits.v_c) << 5;
 	init_int = dc_fixpt_floor(data->inits.v_c);
@@ -609,11 +611,13 @@ static void dpp1_dscl_set_manual_ratio_init(
 		SCL_V_INIT_FRAC_C, init_frac,
 		SCL_V_INIT_INT_C, init_int);
 
-	init_frac = dc_fixpt_u0d19(data->inits.v_c_bot) << 5;
-	init_int = dc_fixpt_floor(data->inits.v_c_bot);
-	REG_SET_2(SCL_VERT_FILTER_INIT_BOT_C, 0,
-		SCL_V_INIT_FRAC_BOT_C, init_frac,
-		SCL_V_INIT_INT_BOT_C, init_int);
+	if (REG(SCL_VERT_FILTER_INIT_BOT_C)) {
+		init_frac = dc_fixpt_u0d19(data->inits.v_c_bot) << 5;
+		init_int = dc_fixpt_floor(data->inits.v_c_bot);
+		REG_SET_2(SCL_VERT_FILTER_INIT_BOT_C, 0,
+			SCL_V_INIT_FRAC_BOT_C, init_frac,
+			SCL_V_INIT_INT_BOT_C, init_int);
+	}
 }
 
 
@@ -688,15 +692,17 @@ void dpp1_dscl_set_scaler_manual_scale(
 		return;
 
 	/* Black offsets */
-	if (ycbcr)
-		REG_SET_2(SCL_BLACK_OFFSET, 0,
-				SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y,
-				SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_CBCR);
-	else
+	if (REG(SCL_BLACK_OFFSET)) {
+		if (ycbcr)
+			REG_SET_2(SCL_BLACK_OFFSET, 0,
+					SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y,
+					SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_CBCR);
+		else
 
-		REG_SET_2(SCL_BLACK_OFFSET, 0,
-				SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y,
-				SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_RGB_Y);
+			REG_SET_2(SCL_BLACK_OFFSET, 0,
+					SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y,
+					SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_RGB_Y);
+	}
 
 	/* Manually calculate scale ratio and init values */
 	dpp1_dscl_set_manual_ratio_init(dpp, scl_data);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
index c7d1e678ebf5..e161ad836812 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
@@ -29,19 +29,20 @@
 #include "reg_helper.h"
 
 #define CTX \
-	hubbub->ctx
+	hubbub1->base.ctx
 #define DC_LOGGER \
-	hubbub->ctx->logger
+	hubbub1->base.ctx->logger
 #define REG(reg)\
-	hubbub->regs->reg
+	hubbub1->regs->reg
 
 #undef FN
 #define FN(reg_name, field_name) \
-	hubbub->shifts->field_name, hubbub->masks->field_name
+	hubbub1->shifts->field_name, hubbub1->masks->field_name
 
 void hubbub1_wm_read_state(struct hubbub *hubbub,
 		struct dcn_hubbub_wm *wm)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
 	struct dcn_hubbub_wm_set *s;
 
 	memset(wm, 0, sizeof(struct dcn_hubbub_wm));
@@ -87,14 +88,23 @@ void hubbub1_wm_read_state(struct hubbub *hubbub,
 	s->dram_clk_chanage = REG_READ(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D);
 }
 
-void hubbub1_disable_allow_self_refresh(struct hubbub *hubbub)
+void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow)
 {
-	REG_UPDATE(DCHUBBUB_ARB_DRAM_STATE_CNTL,
-			DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, 0);
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
+	/*
+	 * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 1 means do not allow stutter
+	 * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 means allow stutter
+	 */
+
+	REG_UPDATE_2(DCHUBBUB_ARB_DRAM_STATE_CNTL,
+			DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_VALUE, 0,
+			DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, !allow);
 }
 
 bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubbub)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
 	uint32_t enable = 0;
 
 	REG_GET(DCHUBBUB_ARB_DRAM_STATE_CNTL,
@@ -107,6 +117,8 @@ bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubbub)
 bool hubbub1_verify_allow_pstate_change_high(
 	struct hubbub *hubbub)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
 	/* pstate latency is ~20us so if we wait over 40us and pstate allow
 	 * still not asserted, we are probably stuck and going to hang
 	 *
@@ -193,7 +205,7 @@ bool hubbub1_verify_allow_pstate_change_high(
 	 * 31:    SOC pstate change request
 	 */
 
-	REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, hubbub->debug_test_index_pstate);
+	REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, hubbub1->debug_test_index_pstate);
 
 	for (i = 0; i < pstate_wait_timeout_us; i++) {
 		debug_data = REG_READ(DCHUBBUB_TEST_DEBUG_DATA);
@@ -244,6 +256,8 @@ static uint32_t convert_and_clamp(
 
 void hubbub1_wm_change_req_wa(struct hubbub *hubbub)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
 	REG_UPDATE_SEQ(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL,
 			DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 0, 1);
 }
@@ -254,7 +268,7 @@ void hubbub1_program_watermarks(
 		unsigned int refclk_mhz,
 		bool safe_to_lower)
 {
-	uint32_t force_en = hubbub->ctx->dc->debug.disable_stutter ? 1 : 0;
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
 	/*
 	 * Need to clamp to max of the register values (i.e. no wrap)
 	 * for dcn1, all wm registers are 21-bit wide
@@ -264,8 +278,8 @@ void hubbub1_program_watermarks(
 
 	/* Repeat for water mark set A, B, C and D. */
 	/* clock state A */
-	if (safe_to_lower || watermarks->a.urgent_ns > hubbub->watermarks.a.urgent_ns) {
-		hubbub->watermarks.a.urgent_ns = watermarks->a.urgent_ns;
+	if (safe_to_lower || watermarks->a.urgent_ns > hubbub1->watermarks.a.urgent_ns) {
+		hubbub1->watermarks.a.urgent_ns = watermarks->a.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->a.urgent_ns,
 				refclk_mhz, 0x1fffff);
 		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value);
@@ -275,20 +289,22 @@ void hubbub1_program_watermarks(
 			watermarks->a.urgent_ns, prog_wm_value);
 	}
 
-	if (safe_to_lower || watermarks->a.pte_meta_urgent_ns > hubbub->watermarks.a.pte_meta_urgent_ns) {
-		hubbub->watermarks.a.pte_meta_urgent_ns = watermarks->a.pte_meta_urgent_ns;
-		prog_wm_value = convert_and_clamp(watermarks->a.pte_meta_urgent_ns,
-				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A, prog_wm_value);
-		DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_A calculated =%d\n"
-			"HW register value = 0x%x\n",
-			watermarks->a.pte_meta_urgent_ns, prog_wm_value);
+	if (REG(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A)) {
+		if (safe_to_lower || watermarks->a.pte_meta_urgent_ns > hubbub1->watermarks.a.pte_meta_urgent_ns) {
+			hubbub1->watermarks.a.pte_meta_urgent_ns = watermarks->a.pte_meta_urgent_ns;
+			prog_wm_value = convert_and_clamp(watermarks->a.pte_meta_urgent_ns,
+					refclk_mhz, 0x1fffff);
+			REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A, prog_wm_value);
+			DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_A calculated =%d\n"
+				"HW register value = 0x%x\n",
+				watermarks->a.pte_meta_urgent_ns, prog_wm_value);
+		}
 	}
 
 	if (REG(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A)) {
 		if (safe_to_lower || watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns
-				> hubbub->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) {
-			hubbub->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
+				> hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) {
+			hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
 					watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns,
@@ -300,8 +316,8 @@ void hubbub1_program_watermarks(
 		}
 
 		if (safe_to_lower || watermarks->a.cstate_pstate.cstate_exit_ns
-				> hubbub->watermarks.a.cstate_pstate.cstate_exit_ns) {
-			hubbub->watermarks.a.cstate_pstate.cstate_exit_ns =
+				> hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) {
+			hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns =
 					watermarks->a.cstate_pstate.cstate_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->a.cstate_pstate.cstate_exit_ns,
@@ -314,8 +330,8 @@ void hubbub1_program_watermarks(
 	}
 
 	if (safe_to_lower || watermarks->a.cstate_pstate.pstate_change_ns
-			> hubbub->watermarks.a.cstate_pstate.pstate_change_ns) {
-		hubbub->watermarks.a.cstate_pstate.pstate_change_ns =
+			> hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) {
+		hubbub1->watermarks.a.cstate_pstate.pstate_change_ns =
 				watermarks->a.cstate_pstate.pstate_change_ns;
 		prog_wm_value = convert_and_clamp(
 				watermarks->a.cstate_pstate.pstate_change_ns,
@@ -327,8 +343,8 @@ void hubbub1_program_watermarks(
 	}
 
 	/* clock state B */
-	if (safe_to_lower || watermarks->b.urgent_ns > hubbub->watermarks.b.urgent_ns) {
-		hubbub->watermarks.b.urgent_ns = watermarks->b.urgent_ns;
+	if (safe_to_lower || watermarks->b.urgent_ns > hubbub1->watermarks.b.urgent_ns) {
+		hubbub1->watermarks.b.urgent_ns = watermarks->b.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->b.urgent_ns,
 				refclk_mhz, 0x1fffff);
 		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value);
@@ -338,20 +354,22 @@ void hubbub1_program_watermarks(
 			watermarks->b.urgent_ns, prog_wm_value);
 	}
 
-	if (safe_to_lower || watermarks->b.pte_meta_urgent_ns > hubbub->watermarks.b.pte_meta_urgent_ns) {
-		hubbub->watermarks.b.pte_meta_urgent_ns = watermarks->b.pte_meta_urgent_ns;
-		prog_wm_value = convert_and_clamp(watermarks->b.pte_meta_urgent_ns,
-				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B, prog_wm_value);
-		DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_B calculated =%d\n"
-			"HW register value = 0x%x\n",
-			watermarks->b.pte_meta_urgent_ns, prog_wm_value);
+	if (REG(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B)) {
+		if (safe_to_lower || watermarks->b.pte_meta_urgent_ns > hubbub1->watermarks.b.pte_meta_urgent_ns) {
+			hubbub1->watermarks.b.pte_meta_urgent_ns = watermarks->b.pte_meta_urgent_ns;
+			prog_wm_value = convert_and_clamp(watermarks->b.pte_meta_urgent_ns,
+					refclk_mhz, 0x1fffff);
+			REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B, prog_wm_value);
+			DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_B calculated =%d\n"
+				"HW register value = 0x%x\n",
+				watermarks->b.pte_meta_urgent_ns, prog_wm_value);
+		}
 	}
 
 	if (REG(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B)) {
 		if (safe_to_lower || watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns
-				> hubbub->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) {
-			hubbub->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns =
+				> hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) {
+			hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns =
 					watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns,
@@ -363,8 +381,8 @@ void hubbub1_program_watermarks(
 		}
 
 		if (safe_to_lower || watermarks->b.cstate_pstate.cstate_exit_ns
-				> hubbub->watermarks.b.cstate_pstate.cstate_exit_ns) {
-			hubbub->watermarks.b.cstate_pstate.cstate_exit_ns =
+				> hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) {
+			hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns =
 					watermarks->b.cstate_pstate.cstate_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->b.cstate_pstate.cstate_exit_ns,
@@ -377,8 +395,8 @@ void hubbub1_program_watermarks(
 	}
 
 	if (safe_to_lower || watermarks->b.cstate_pstate.pstate_change_ns
-			> hubbub->watermarks.b.cstate_pstate.pstate_change_ns) {
-		hubbub->watermarks.b.cstate_pstate.pstate_change_ns =
+			> hubbub1->watermarks.b.cstate_pstate.pstate_change_ns) {
+		hubbub1->watermarks.b.cstate_pstate.pstate_change_ns =
 				watermarks->b.cstate_pstate.pstate_change_ns;
 		prog_wm_value = convert_and_clamp(
 				watermarks->b.cstate_pstate.pstate_change_ns,
@@ -390,8 +408,8 @@ void hubbub1_program_watermarks(
 	}
 
 	/* clock state C */
-	if (safe_to_lower || watermarks->c.urgent_ns > hubbub->watermarks.c.urgent_ns) {
-		hubbub->watermarks.c.urgent_ns = watermarks->c.urgent_ns;
+	if (safe_to_lower || watermarks->c.urgent_ns > hubbub1->watermarks.c.urgent_ns) {
+		hubbub1->watermarks.c.urgent_ns = watermarks->c.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->c.urgent_ns,
 				refclk_mhz, 0x1fffff);
 		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value);
@@ -401,20 +419,22 @@ void hubbub1_program_watermarks(
 			watermarks->c.urgent_ns, prog_wm_value);
 	}
 
-	if (safe_to_lower || watermarks->c.pte_meta_urgent_ns > hubbub->watermarks.c.pte_meta_urgent_ns) {
-		hubbub->watermarks.c.pte_meta_urgent_ns = watermarks->c.pte_meta_urgent_ns;
-		prog_wm_value = convert_and_clamp(watermarks->c.pte_meta_urgent_ns,
-				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C, prog_wm_value);
-		DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_C calculated =%d\n"
-			"HW register value = 0x%x\n",
-			watermarks->c.pte_meta_urgent_ns, prog_wm_value);
+	if (REG(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C)) {
+		if (safe_to_lower || watermarks->c.pte_meta_urgent_ns > hubbub1->watermarks.c.pte_meta_urgent_ns) {
+			hubbub1->watermarks.c.pte_meta_urgent_ns = watermarks->c.pte_meta_urgent_ns;
+			prog_wm_value = convert_and_clamp(watermarks->c.pte_meta_urgent_ns,
+					refclk_mhz, 0x1fffff);
+			REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C, prog_wm_value);
+			DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_C calculated =%d\n"
+				"HW register value = 0x%x\n",
+				watermarks->c.pte_meta_urgent_ns, prog_wm_value);
+		}
 	}
 
 	if (REG(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C)) {
 		if (safe_to_lower || watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns
-				> hubbub->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) {
-			hubbub->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns =
+				> hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) {
+			hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns =
 					watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns,
@@ -426,8 +446,8 @@ void hubbub1_program_watermarks(
 		}
 
 		if (safe_to_lower || watermarks->c.cstate_pstate.cstate_exit_ns
-				> hubbub->watermarks.c.cstate_pstate.cstate_exit_ns) {
-			hubbub->watermarks.c.cstate_pstate.cstate_exit_ns =
+				> hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) {
+			hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns =
 					watermarks->c.cstate_pstate.cstate_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->c.cstate_pstate.cstate_exit_ns,
@@ -440,8 +460,8 @@ void hubbub1_program_watermarks(
 	}
 
 	if (safe_to_lower || watermarks->c.cstate_pstate.pstate_change_ns
-			> hubbub->watermarks.c.cstate_pstate.pstate_change_ns) {
-		hubbub->watermarks.c.cstate_pstate.pstate_change_ns =
+			> hubbub1->watermarks.c.cstate_pstate.pstate_change_ns) {
+		hubbub1->watermarks.c.cstate_pstate.pstate_change_ns =
 				watermarks->c.cstate_pstate.pstate_change_ns;
 		prog_wm_value = convert_and_clamp(
 				watermarks->c.cstate_pstate.pstate_change_ns,
@@ -453,8 +473,8 @@ void hubbub1_program_watermarks(
 	}
 
 	/* clock state D */
-	if (safe_to_lower || watermarks->d.urgent_ns > hubbub->watermarks.d.urgent_ns) {
-		hubbub->watermarks.d.urgent_ns = watermarks->d.urgent_ns;
+	if (safe_to_lower || watermarks->d.urgent_ns > hubbub1->watermarks.d.urgent_ns) {
+		hubbub1->watermarks.d.urgent_ns = watermarks->d.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->d.urgent_ns,
 				refclk_mhz, 0x1fffff);
 		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value);
@@ -464,20 +484,22 @@ void hubbub1_program_watermarks(
 			watermarks->d.urgent_ns, prog_wm_value);
 	}
 
-	if (safe_to_lower || watermarks->d.pte_meta_urgent_ns > hubbub->watermarks.d.pte_meta_urgent_ns) {
-		hubbub->watermarks.d.pte_meta_urgent_ns = watermarks->d.pte_meta_urgent_ns;
-		prog_wm_value = convert_and_clamp(watermarks->d.pte_meta_urgent_ns,
-				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D, prog_wm_value);
-		DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_D calculated =%d\n"
-			"HW register value = 0x%x\n",
-			watermarks->d.pte_meta_urgent_ns, prog_wm_value);
+	if (REG(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D)) {
+		if (safe_to_lower || watermarks->d.pte_meta_urgent_ns > hubbub1->watermarks.d.pte_meta_urgent_ns) {
+			hubbub1->watermarks.d.pte_meta_urgent_ns = watermarks->d.pte_meta_urgent_ns;
+			prog_wm_value = convert_and_clamp(watermarks->d.pte_meta_urgent_ns,
+					refclk_mhz, 0x1fffff);
+			REG_WRITE(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D, prog_wm_value);
+			DC_LOG_BANDWIDTH_CALCS("PTE_META_URGENCY_WATERMARK_D calculated =%d\n"
+				"HW register value = 0x%x\n",
+				watermarks->d.pte_meta_urgent_ns, prog_wm_value);
+		}
 	}
 
 	if (REG(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D)) {
 		if (safe_to_lower || watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns
-				> hubbub->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) {
-			hubbub->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns =
+				> hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) {
+			hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns =
 					watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns,
@@ -489,8 +511,8 @@ void hubbub1_program_watermarks(
 		}
 
 		if (safe_to_lower || watermarks->d.cstate_pstate.cstate_exit_ns
-				> hubbub->watermarks.d.cstate_pstate.cstate_exit_ns) {
-			hubbub->watermarks.d.cstate_pstate.cstate_exit_ns =
+				> hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) {
+			hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns =
 					watermarks->d.cstate_pstate.cstate_exit_ns;
 			prog_wm_value = convert_and_clamp(
 					watermarks->d.cstate_pstate.cstate_exit_ns,
@@ -503,8 +525,8 @@ void hubbub1_program_watermarks(
 	}
 
 	if (safe_to_lower || watermarks->d.cstate_pstate.pstate_change_ns
-			> hubbub->watermarks.d.cstate_pstate.pstate_change_ns) {
-		hubbub->watermarks.d.cstate_pstate.pstate_change_ns =
+			> hubbub1->watermarks.d.cstate_pstate.pstate_change_ns) {
+		hubbub1->watermarks.d.cstate_pstate.pstate_change_ns =
 				watermarks->d.cstate_pstate.pstate_change_ns;
 		prog_wm_value = convert_and_clamp(
 				watermarks->d.cstate_pstate.pstate_change_ns,
@@ -520,9 +542,7 @@ void hubbub1_program_watermarks(
 	REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND,
 			DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 68);
 
-	REG_UPDATE_2(DCHUBBUB_ARB_DRAM_STATE_CNTL,
-			DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_VALUE, 0,
-			DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, force_en);
+	hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter);
 
 #if 0
 	REG_UPDATE_2(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL,
@@ -535,6 +555,8 @@ void hubbub1_update_dchub(
 	struct hubbub *hubbub,
 	struct dchub_init_data *dh_data)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
 	if (REG(DCHUBBUB_SDPIF_FB_TOP) == 0) {
 		ASSERT(false);
 		/*should not come here*/
@@ -594,6 +616,8 @@ void hubbub1_update_dchub(
 
 void hubbub1_toggle_watermark_change_req(struct hubbub *hubbub)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
 	uint32_t watermark_change_req;
 
 	REG_GET(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL,
@@ -610,6 +634,8 @@ void hubbub1_toggle_watermark_change_req(struct hubbub *hubbub)
 
 void hubbub1_soft_reset(struct hubbub *hubbub, bool reset)
 {
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
 	uint32_t reset_en = reset ? 1 : 0;
 
 	REG_UPDATE(DCHUBBUB_SOFT_RESET,
@@ -752,7 +778,9 @@ static bool hubbub1_get_dcc_compression_cap(struct hubbub *hubbub,
 		const struct dc_dcc_surface_param *input,
 		struct dc_surface_dcc_cap *output)
 {
-	struct dc *dc = hubbub->ctx->dc;
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+	struct dc *dc = hubbub1->base.ctx->dc;
+
 	/* implement section 1.6.2.1 of DCN1_Programming_Guide.docx */
 	enum dcc_control dcc_control;
 	unsigned int bpe;
@@ -764,10 +792,10 @@ static bool hubbub1_get_dcc_compression_cap(struct hubbub *hubbub,
 	if (dc->debug.disable_dcc == DCC_DISABLE)
 		return false;
 
-	if (!hubbub->funcs->dcc_support_pixel_format(input->format, &bpe))
+	if (!hubbub1->base.funcs->dcc_support_pixel_format(input->format, &bpe))
 		return false;
 
-	if (!hubbub->funcs->dcc_support_swizzle(input->swizzle_mode, bpe,
+	if (!hubbub1->base.funcs->dcc_support_swizzle(input->swizzle_mode, bpe,
 			&segment_order_horz, &segment_order_vert))
 		return false;
 
@@ -837,6 +865,7 @@ static const struct hubbub_funcs hubbub1_funcs = {
 	.dcc_support_swizzle = hubbub1_dcc_support_swizzle,
 	.dcc_support_pixel_format = hubbub1_dcc_support_pixel_format,
 	.get_dcc_compression_cap = hubbub1_get_dcc_compression_cap,
+	.wm_read_state = hubbub1_wm_read_state,
 };
 
 void hubbub1_construct(struct hubbub *hubbub,
@@ -845,18 +874,20 @@ void hubbub1_construct(struct hubbub *hubbub,
 	const struct dcn_hubbub_shift *hubbub_shift,
 	const struct dcn_hubbub_mask *hubbub_mask)
 {
-	hubbub->ctx = ctx;
+	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
+
+	hubbub1->base.ctx = ctx;
 
-	hubbub->funcs = &hubbub1_funcs;
+	hubbub1->base.funcs = &hubbub1_funcs;
 
-	hubbub->regs = hubbub_regs;
-	hubbub->shifts = hubbub_shift;
-	hubbub->masks = hubbub_mask;
+	hubbub1->regs = hubbub_regs;
+	hubbub1->shifts = hubbub_shift;
+	hubbub1->masks = hubbub_mask;
 
-	hubbub->debug_test_index_pstate = 0x7;
+	hubbub1->debug_test_index_pstate = 0x7;
 #if defined(CONFIG_DRM_AMD_DC_DCN1_01)
 	if (ctx->dce_version == DCN_VERSION_1_01)
-		hubbub->debug_test_index_pstate = 0xB;
+		hubbub1->debug_test_index_pstate = 0xB;
 #endif
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
index d0f03d152913..9cd4a5194154 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
@@ -29,6 +29,9 @@
 #include "core_types.h"
 #include "dchubbub.h"
 
+#define TO_DCN10_HUBBUB(hubbub)\
+	container_of(hubbub, struct dcn10_hubbub, base)
+
 #define HUBHUB_REG_LIST_DCN()\
 	SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A),\
 	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A),\
@@ -107,6 +110,12 @@ struct dcn_hubbub_registers {
 	uint32_t DCHUBBUB_SDPIF_AGP_TOP;
 	uint32_t DCHUBBUB_CRC_CTRL;
 	uint32_t DCHUBBUB_SOFT_RESET;
+	uint32_t DCN_VM_FB_LOCATION_BASE;
+	uint32_t DCN_VM_FB_LOCATION_TOP;
+	uint32_t DCN_VM_FB_OFFSET;
+	uint32_t DCN_VM_AGP_BOT;
+	uint32_t DCN_VM_AGP_TOP;
+	uint32_t DCN_VM_AGP_BASE;
 };
 
 /* set field name */
@@ -152,7 +161,13 @@ struct dcn_hubbub_registers {
 		type SDPIF_FB_OFFSET;\
 		type SDPIF_AGP_BASE;\
 		type SDPIF_AGP_BOT;\
-		type SDPIF_AGP_TOP
+		type SDPIF_AGP_TOP;\
+		type FB_BASE;\
+		type FB_TOP;\
+		type FB_OFFSET;\
+		type AGP_BOT;\
+		type AGP_TOP;\
+		type AGP_BASE
 
 
 struct dcn_hubbub_shift {
@@ -165,22 +180,8 @@ struct dcn_hubbub_mask {
 
 struct dc;
 
-struct dcn_hubbub_wm_set {
-	uint32_t wm_set;
-	uint32_t data_urgent;
-	uint32_t pte_meta_urgent;
-	uint32_t sr_enter;
-	uint32_t sr_exit;
-	uint32_t dram_clk_chanage;
-};
-
-struct dcn_hubbub_wm {
-	struct dcn_hubbub_wm_set sets[4];
-};
-
-struct hubbub {
-	const struct hubbub_funcs *funcs;
-	struct dc_context *ctx;
+struct dcn10_hubbub {
+	struct hubbub base;
 	const struct dcn_hubbub_registers *regs;
 	const struct dcn_hubbub_shift *shifts;
 	const struct dcn_hubbub_mask *masks;
@@ -203,7 +204,7 @@ void hubbub1_program_watermarks(
 		unsigned int refclk_mhz,
 		bool safe_to_lower);
 
-void hubbub1_disable_allow_self_refresh(struct hubbub *hubbub);
+void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow);
 
 bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubub);
 
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 345af015d061..683829466a44 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
@@ -115,7 +115,7 @@ static void hubp1_set_hubp_blank_en(struct hubp *hubp, bool blank)
 	REG_UPDATE(DCHUBP_CNTL, HUBP_BLANK_EN, blank_en);
 }
 
-static void hubp1_vready_workaround(struct hubp *hubp,
+void hubp1_vready_workaround(struct hubp *hubp,
 		struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest)
 {
 	uint32_t value = 0;
@@ -317,7 +317,8 @@ void hubp1_program_pixel_format(
 bool hubp1_program_surface_flip_and_addr(
 	struct hubp *hubp,
 	const struct dc_plane_address *address,
-	bool flip_immediate)
+	bool flip_immediate,
+	uint8_t vmid)
 {
 	struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp);
 
@@ -1140,7 +1141,7 @@ void hubp1_cursor_set_position(
 	if (src_y_offset >= (int)param->viewport.height)
 		cur_en = 0;  /* not visible beyond bottom edge*/
 
-	if (src_y_offset < 0) //+ (int)hubp->curs_attr.height
+	if (src_y_offset + (int)hubp->curs_attr.height <= 0)
 		cur_en = 0;  /* not visible beyond top edge*/
 
 	if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0)
@@ -1149,9 +1150,28 @@ void hubp1_cursor_set_position(
 	REG_UPDATE(CURSOR_CONTROL,
 			CURSOR_ENABLE, cur_en);
 
-	REG_SET_2(CURSOR_POSITION, 0,
-			CURSOR_X_POSITION, pos->x,
+	//account for cases where we see negative offset relative to overlay plane
+	if (src_x_offset < 0 && src_y_offset < 0) {
+		REG_SET_2(CURSOR_POSITION, 0,
+			CURSOR_X_POSITION, 0,
+			CURSOR_Y_POSITION, 0);
+		x_hotspot -= src_x_offset;
+		y_hotspot -= src_y_offset;
+	} else if (src_x_offset < 0) {
+		REG_SET_2(CURSOR_POSITION, 0,
+			CURSOR_X_POSITION, 0,
 			CURSOR_Y_POSITION, pos->y);
+		x_hotspot -= src_x_offset;
+	} else if (src_y_offset < 0) {
+		REG_SET_2(CURSOR_POSITION, 0,
+			CURSOR_X_POSITION, pos->x,
+			CURSOR_Y_POSITION, 0);
+		y_hotspot -= src_y_offset;
+	} else {
+		REG_SET_2(CURSOR_POSITION, 0,
+				CURSOR_X_POSITION, pos->x,
+				CURSOR_Y_POSITION, pos->y);
+	}
 
 	REG_SET_2(CURSOR_HOT_SPOT, 0,
 			CURSOR_HOT_SPOT_X, x_hotspot,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
index 62d4232e7796..a6d6dfe00617 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
@@ -707,11 +707,6 @@ void hubp1_dcc_control(struct hubp *hubp,
 		bool enable,
 		bool independent_64b_blks);
 
-bool hubp1_program_surface_flip_and_addr(
-	struct hubp *hubp,
-	const struct dc_plane_address *address,
-	bool flip_immediate);
-
 bool hubp1_is_flip_pending(struct hubp *hubp);
 
 void hubp1_cursor_set_attributes(
@@ -745,5 +740,7 @@ void hubp1_clear_underflow(struct hubp *hubp);
 
 enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch);
 
+void hubp1_vready_workaround(struct hubp *hubp,
+		struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest);
 
 #endif
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 91e015e14355..117d9d8227f7 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
@@ -40,7 +40,6 @@
 #include "ipp.h"
 #include "mpc.h"
 #include "reg_helper.h"
-#include "custom_float.h"
 #include "dcn10_hubp.h"
 #include "dcn10_hubbub.h"
 #include "dcn10_cm_common.h"
@@ -92,10 +91,11 @@ static void log_mpc_crc(struct dc *dc,
 void dcn10_log_hubbub_state(struct dc *dc, struct dc_log_buffer_ctx *log_ctx)
 {
 	struct dc_context *dc_ctx = dc->ctx;
-	struct dcn_hubbub_wm wm = {0};
+	struct dcn_hubbub_wm wm;
 	int i;
 
-	hubbub1_wm_read_state(dc->res_pool->hubbub, &wm);
+	memset(&wm, 0, sizeof(struct dcn_hubbub_wm));
+	dc->res_pool->hubbub->funcs->wm_read_state(dc->res_pool->hubbub, &wm);
 
 	DTN_INFO("HUBBUB WM:      data_urgent  pte_meta_urgent"
 			"         sr_enter          sr_exit  dram_clk_change\n");
@@ -636,8 +636,6 @@ static enum dc_status dcn10_enable_stream_timing(
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	enum dc_color_space color_space;
 	struct tg_color black_color = {0};
-	struct drr_params params = {0};
-	unsigned int event_triggers = 0;
 
 	/* by upper caller loop, pipe0 is parent pipe and be called first.
 	 * back end is set up by for pipe0. Other children pipe share back end
@@ -705,19 +703,6 @@ static enum dc_status dcn10_enable_stream_timing(
 		return DC_ERROR_UNEXPECTED;
 	}
 
-	params.vertical_total_min = stream->adjust.v_total_min;
-	params.vertical_total_max = stream->adjust.v_total_max;
-	if (pipe_ctx->stream_res.tg->funcs->set_drr)
-		pipe_ctx->stream_res.tg->funcs->set_drr(
-			pipe_ctx->stream_res.tg, &params);
-
-	// DRR should set trigger event to monitor surface update event
-	if (stream->adjust.v_total_min != 0 && stream->adjust.v_total_max != 0)
-		event_triggers = 0x80;
-	if (pipe_ctx->stream_res.tg->funcs->set_static_screen_control)
-		pipe_ctx->stream_res.tg->funcs->set_static_screen_control(
-				pipe_ctx->stream_res.tg, event_triggers);
-
 	/* TODO program crtc source select for non-virtual signal*/
 	/* TODO program FMT */
 	/* TODO setup link_enc */
@@ -971,92 +956,34 @@ static void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx)
 					pipe_ctx->pipe_idx);
 }
 
-static void dcn10_init_hw(struct dc *dc)
+static void dcn10_init_pipes(struct dc *dc, struct dc_state *context)
 {
 	int i;
-	struct abm *abm = dc->res_pool->abm;
-	struct dmcu *dmcu = dc->res_pool->dmcu;
-	struct dce_hwseq *hws = dc->hwseq;
-	struct dc_bios *dcb = dc->ctx->dc_bios;
-	struct dc_state  *context = dc->current_state;
-
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		REG_WRITE(REFCLK_CNTL, 0);
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
-		REG_WRITE(DIO_MEM_PWR_CTRL, 0);
-
-		if (!dc->debug.disable_clock_gate) {
-			/* enable all DCN clock gating */
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
-
-			REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
-		}
-
-		enable_power_gating_plane(dc->hwseq, true);
-	} else {
-
-		if (!dcb->funcs->is_accelerated_mode(dcb)) {
-			bool allow_self_fresh_force_enable =
-					hububu1_is_allow_self_refresh_enabled(dc->res_pool->hubbub);
-
-			bios_golden_init(dc);
-
-			/* WA for making DF sleep when idle after resume from S0i3.
-			 * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by
-			 * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0
-			 * before calling command table and it changed to 1 after,
-			 * it should be set back to 0.
-			 */
-			if (allow_self_fresh_force_enable == false &&
-					hububu1_is_allow_self_refresh_enabled(dc->res_pool->hubbub))
-				hubbub1_disable_allow_self_refresh(dc->res_pool->hubbub);
-
-			disable_vga(dc->hwseq);
-		}
-
-		for (i = 0; i < dc->link_count; i++) {
-			/* Power up AND update implementation according to the
-			 * required signal (which may be different from the
-			 * default signal on connector).
-			 */
-			struct dc_link *link = dc->links[i];
-
-			if (link->link_enc->connector.id == CONNECTOR_ID_EDP)
-				dc->hwss.edp_power_control(link, true);
-
-			link->link_enc->funcs->hw_init(link->link_enc);
-		}
-	}
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct timing_generator *tg = dc->res_pool->timing_generators[i];
 
 		if (tg->funcs->is_tg_enabled(tg))
 			tg->funcs->lock(tg);
-	}
-
-	/* Blank controller using driver code instead of
-	 * command table.
-	 */
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct timing_generator *tg = dc->res_pool->timing_generators[i];
 
+		/* Blank controller using driver code instead of
+		 * command table.
+		 */
 		if (tg->funcs->is_tg_enabled(tg)) {
 			tg->funcs->set_blank(tg, true);
 			hwss_wait_for_blank_complete(tg);
 		}
 	}
 
-	/* Reset all MPCC muxes */
 	dc->res_pool->mpc->funcs->mpc_init(dc->res_pool->mpc);
 
-	for (i = 0; i < dc->res_pool->timing_generator_count; i++) {
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct timing_generator *tg = dc->res_pool->timing_generators[i];
-		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 		struct hubp *hubp = dc->res_pool->hubps[i];
 		struct dpp *dpp = dc->res_pool->dpps[i];
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+
+		dpp->funcs->dpp_reset(dpp);
 
 		pipe_ctx->stream_res.tg = tg;
 		pipe_ctx->pipe_idx = i;
@@ -1074,18 +1001,9 @@ static void dcn10_init_hw(struct dc *dc)
 		pipe_ctx->stream_res.opp = dc->res_pool->opps[i];
 
 		hwss1_plane_atomic_disconnect(dc, pipe_ctx);
-	}
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct timing_generator *tg = dc->res_pool->timing_generators[i];
 
 		if (tg->funcs->is_tg_enabled(tg))
 			tg->funcs->unlock(tg);
-	}
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct timing_generator *tg = dc->res_pool->timing_generators[i];
-		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
 		dcn10_disable_plane(dc, pipe_ctx);
 
@@ -1094,10 +1012,73 @@ static void dcn10_init_hw(struct dc *dc)
 
 		tg->funcs->tg_init(tg);
 	}
+}
 
-	/* end of FPGA. Below if real ASIC */
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
+static void dcn10_init_hw(struct dc *dc)
+{
+	int i;
+	struct abm *abm = dc->res_pool->abm;
+	struct dmcu *dmcu = dc->res_pool->dmcu;
+	struct dce_hwseq *hws = dc->hwseq;
+	struct dc_bios *dcb = dc->ctx->dc_bios;
+
+	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+		REG_WRITE(REFCLK_CNTL, 0);
+		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
+		REG_WRITE(DIO_MEM_PWR_CTRL, 0);
+
+		if (!dc->debug.disable_clock_gate) {
+			/* enable all DCN clock gating */
+			REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
+
+			REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
+
+			REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
+		}
+
+		enable_power_gating_plane(dc->hwseq, true);
+
+		/* end of FPGA. Below if real ASIC */
 		return;
+	}
+
+	if (!dcb->funcs->is_accelerated_mode(dcb)) {
+		bool allow_self_fresh_force_enable =
+			hububu1_is_allow_self_refresh_enabled(
+						dc->res_pool->hubbub);
+
+		bios_golden_init(dc);
+
+		/* WA for making DF sleep when idle after resume from S0i3.
+		 * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by
+		 * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0
+		 * before calling command table and it changed to 1 after,
+		 * it should be set back to 0.
+		 */
+		if (allow_self_fresh_force_enable == false &&
+				hububu1_is_allow_self_refresh_enabled(dc->res_pool->hubbub))
+			hubbub1_allow_self_refresh_control(dc->res_pool->hubbub, true);
+
+		disable_vga(dc->hwseq);
+	}
+
+	for (i = 0; i < dc->link_count; i++) {
+		/* Power up AND update implementation according to the
+		 * required signal (which may be different from the
+		 * default signal on connector).
+		 */
+		struct dc_link *link = dc->links[i];
+
+		if (link->link_enc->connector.id == CONNECTOR_ID_EDP)
+			dc->hwss.edp_power_control(link, true);
+
+		link->link_enc->funcs->hw_init(link->link_enc);
+
+		/* Check for enabled DIG to identify enabled display */
+		if (link->link_enc->funcs->is_dig_enabled &&
+			link->link_enc->funcs->is_dig_enabled(link->link_enc))
+			link->link_status.link_active = true;
+	}
 
 	for (i = 0; i < dc->res_pool->audio_count; i++) {
 		struct audio *audio = dc->res_pool->audios[i];
@@ -1128,6 +1109,9 @@ static void dcn10_init_hw(struct dc *dc)
 	enable_power_gating_plane(dc->hwseq, true);
 
 	memset(&dc->res_pool->clk_mgr->clks, 0, sizeof(dc->res_pool->clk_mgr->clks));
+
+	if (dc->hwss.init_pipes)
+		dc->hwss.init_pipes(dc, dc->current_state);
 }
 
 static void reset_hw_ctx_wrap(
@@ -1202,7 +1186,8 @@ static void dcn10_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_c
 	pipe_ctx->plane_res.hubp->funcs->hubp_program_surface_flip_and_addr(
 			pipe_ctx->plane_res.hubp,
 			&plane_state->address,
-			plane_state->flip_immediate);
+			plane_state->flip_immediate,
+			0);
 
 	plane_state->status.requested_address = plane_state->address;
 
@@ -2048,7 +2033,7 @@ void update_dchubp_dpp(
 			dc->res_pool->dccg->funcs->update_dpp_dto(
 					dc->res_pool->dccg,
 					dpp->inst,
-					pipe_ctx->plane_res.bw.calc.dppclk_khz);
+					pipe_ctx->plane_res.bw.dppclk_khz);
 		else
 			dc->res_pool->clk_mgr->clks.dppclk_khz = should_divided_by_2 ?
 						dc->res_pool->clk_mgr->clks.dispclk_khz / 2 :
@@ -2125,7 +2110,8 @@ void update_dchubp_dpp(
 		plane_state->update_flags.bits.swizzle_change ||
 		plane_state->update_flags.bits.dcc_change ||
 		plane_state->update_flags.bits.bpp_change ||
-		plane_state->update_flags.bits.scaling_change) {
+		plane_state->update_flags.bits.scaling_change ||
+		plane_state->update_flags.bits.plane_size_change) {
 		hubp->funcs->hubp_program_surface_config(
 			hubp,
 			plane_state->format,
@@ -2252,13 +2238,11 @@ static void program_all_pipe_in_tree(
 
 	}
 
-	if (pipe_ctx->plane_state != NULL) {
+	if (pipe_ctx->plane_state != NULL)
 		dcn10_program_pipe(dc, pipe_ctx, context);
-	}
 
-	if (pipe_ctx->bottom_pipe != NULL && pipe_ctx->bottom_pipe != pipe_ctx) {
+	if (pipe_ctx->bottom_pipe != NULL && pipe_ctx->bottom_pipe != pipe_ctx)
 		program_all_pipe_in_tree(dc, pipe_ctx->bottom_pipe, context);
-	}
 }
 
 struct pipe_ctx *find_top_pipe_for_stream(
@@ -2334,9 +2318,10 @@ static void dcn10_apply_ctx_for_surface(
 			}
 		}
 
-		if (!pipe_ctx->plane_state &&
-			old_pipe_ctx->plane_state &&
-			old_pipe_ctx->stream_res.tg == tg) {
+		if ((!pipe_ctx->plane_state ||
+		     pipe_ctx->stream_res.tg != old_pipe_ctx->stream_res.tg) &&
+		    old_pipe_ctx->plane_state &&
+		    old_pipe_ctx->stream_res.tg == tg) {
 
 			dc->hwss.plane_atomic_disconnect(dc, old_pipe_ctx);
 			removed_pipe[i] = true;
@@ -2355,29 +2340,22 @@ static void dcn10_apply_ctx_for_surface(
 			top_pipe_to_program->plane_state->update_flags.bits.full_update)
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
+			tg = pipe_ctx->stream_res.tg;
 			/* Skip inactive pipes and ones already updated */
 			if (!pipe_ctx->stream || pipe_ctx->stream == stream
-					|| !pipe_ctx->plane_state)
+					|| !pipe_ctx->plane_state
+					|| !tg->funcs->is_tg_enabled(tg))
 				continue;
 
-			pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg);
+			tg->funcs->lock(tg);
 
 			pipe_ctx->plane_res.hubp->funcs->hubp_setup_interdependent(
 				pipe_ctx->plane_res.hubp,
 				&pipe_ctx->dlg_regs,
 				&pipe_ctx->ttu_regs);
-		}
-
-	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 || pipe_ctx->stream == stream
-				|| !pipe_ctx->plane_state)
-			continue;
 
-		dcn10_pipe_control_lock(dc, pipe_ctx, false);
-	}
+			tg->funcs->unlock(tg);
+		}
 
 	if (num_planes == 0)
 		false_optc_underflow_wa(dc, stream, tg);
@@ -2390,6 +2368,22 @@ static void dcn10_apply_ctx_for_surface(
 		hubbub1_wm_change_req_wa(dc->res_pool->hubbub);
 }
 
+static void dcn10_stereo_hw_frame_pack_wa(struct dc *dc, struct dc_state *context)
+{
+	uint8_t i;
+
+	for (i = 0; i < context->stream_count; i++) {
+		if (context->streams[i]->timing.timing_3d_format
+				== TIMING_3D_FORMAT_HW_FRAME_PACKING) {
+			/*
+			 * Disable stutter
+			 */
+			hubbub1_allow_self_refresh_control(dc->res_pool->hubbub, false);
+			break;
+		}
+	}
+}
+
 static void dcn10_prepare_bandwidth(
 		struct dc *dc,
 		struct dc_state *context)
@@ -2411,6 +2405,7 @@ static void dcn10_prepare_bandwidth(
 			&context->bw.dcn.watermarks,
 			dc->res_pool->ref_clock_inKhz / 1000,
 			true);
+	dcn10_stereo_hw_frame_pack_wa(dc, context);
 
 	if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE)
 		dcn_bw_notify_pplib_of_wm_ranges(dc);
@@ -2440,6 +2435,7 @@ static void dcn10_optimize_bandwidth(
 			&context->bw.dcn.watermarks,
 			dc->res_pool->ref_clock_inKhz / 1000,
 			true);
+	dcn10_stereo_hw_frame_pack_wa(dc, context);
 
 	if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE)
 		dcn_bw_notify_pplib_of_wm_ranges(dc);
@@ -2525,7 +2521,7 @@ static void dcn10_config_stereo_parameters(
 			timing_3d_format == TIMING_3D_FORMAT_DP_HDMI_INBAND_FA ||
 			timing_3d_format == TIMING_3D_FORMAT_SIDEBAND_FA) {
 			enum display_dongle_type dongle = \
-					stream->sink->link->ddc->dongle_type;
+					stream->link->ddc->dongle_type;
 			if (dongle == DISPLAY_DONGLE_DP_VGA_CONVERTER ||
 				dongle == DISPLAY_DONGLE_DP_DVI_CONVERTER ||
 				dongle == DISPLAY_DONGLE_DP_HDMI_CONVERTER)
@@ -2656,7 +2652,7 @@ static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	struct hubp *hubp = pipe_ctx->plane_res.hubp;
 	struct dpp *dpp = pipe_ctx->plane_res.dpp;
 	struct dc_cursor_mi_param param = {
-		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_khz,
+		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10,
 		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clock_inKhz,
 		.viewport = pipe_ctx->plane_res.scl_data.viewport,
 		.h_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.horz,
@@ -2716,6 +2712,7 @@ static void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx)
 static const struct hw_sequencer_funcs dcn10_funcs = {
 	.program_gamut_remap = program_gamut_remap,
 	.init_hw = dcn10_init_hw,
+	.init_pipes = dcn10_init_pipes,
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = dcn10_apply_ctx_for_surface,
 	.update_plane_addr = dcn10_update_plane_addr,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
index cd469014baa3..98f41d250978 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
@@ -40,7 +40,6 @@
 #include "ipp.h"
 #include "mpc.h"
 #include "reg_helper.h"
-#include "custom_float.h"
 #include "dcn10_hubp.h"
 #include "dcn10_hubbub.h"
 #include "dcn10_cm_common.h"
@@ -72,7 +71,7 @@ static unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt,
 static unsigned int dcn10_get_hubbub_state(struct dc *dc, char *pBuf, unsigned int bufSize)
 {
 	struct dc_context *dc_ctx = dc->ctx;
-	struct dcn_hubbub_wm wm = {0};
+	struct dcn_hubbub_wm wm;
 	int i;
 
 	unsigned int chars_printed = 0;
@@ -81,7 +80,8 @@ static unsigned int dcn10_get_hubbub_state(struct dc *dc, char *pBuf, unsigned i
 	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clock_inKhz / 1000;
 	static const unsigned int frac = 1000;
 
-	hubbub1_wm_read_state(dc->res_pool->hubbub, &wm);
+	memset(&wm, 0, sizeof(struct dcn_hubbub_wm));
+	dc->res_pool->hubbub->funcs->wm_read_state(dc->res_pool->hubbub, &wm);
 
 	chars_printed = snprintf_count(pBuf, remaining_buffer, "wm_set_index,data_urgent,pte_meta_urgent,sr_enter,sr_exit,dram_clk_chanage\n");
 	remaining_buffer -= chars_printed;
@@ -419,20 +419,22 @@ static unsigned int dcn10_get_otg_states(struct dc *dc, char *pBuf, unsigned int
 	unsigned int remaining_buffer = bufSize;
 
 	chars_printed = snprintf_count(pBuf, remaining_buffer, "instance,v_bs,v_be,v_ss,v_se,vpol,vmax,vmin,vmax_sel,vmin_sel,"
-			"h_bs,h_be,h_ss,h_se,hpol,htot,vtot,underflow\n");
+			"h_bs,h_be,h_ss,h_se,hpol,htot,vtot,underflow,pixelclk[khz]\n");
 	remaining_buffer -= chars_printed;
 	pBuf += chars_printed;
 
 	for (i = 0; i < pool->timing_generator_count; i++) {
 		struct timing_generator *tg = pool->timing_generators[i];
 		struct dcn_otg_state s = {0};
+		int pix_clk = 0;
 
 		optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s);
+		pix_clk = dc->current_state->res_ctx.pipe_ctx[i].stream_res.pix_clk_params.requested_pix_clk_100hz / 10;
 
 		//only print if OTG master is enabled
 		if (s.otg_enabled & 1) {
 			chars_printed = snprintf_count(pBuf, remaining_buffer, "%x,%d,%d,%d,%d,%d,%d,%d,%d,%d,"
-				"%d,%d,%d,%d,%d,%d,%d,%d"
+				"%d,%d,%d,%d,%d,%d,%d,%d,%d"
 				"\n",
 				tg->inst,
 				s.v_blank_start,
@@ -451,7 +453,8 @@ static unsigned int dcn10_get_otg_states(struct dc *dc, char *pBuf, unsigned int
 				s.h_sync_a_pol,
 				s.h_total,
 				s.v_total,
-				s.underflow_occurred_status);
+				s.underflow_occurred_status,
+				pix_clk);
 
 			remaining_buffer -= chars_printed;
 			pBuf += chars_printed;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
index 477ab9222216..a9db372688ff 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
@@ -85,6 +85,7 @@ static const struct link_encoder_funcs dcn10_lnk_enc_funcs = {
 	.enable_hpd = dcn10_link_encoder_enable_hpd,
 	.disable_hpd = dcn10_link_encoder_disable_hpd,
 	.is_dig_enabled = dcn10_is_dig_enabled,
+	.get_dig_frontend = dcn10_get_dig_frontend,
 	.destroy = dcn10_link_encoder_destroy
 };
 
@@ -440,7 +441,7 @@ static uint8_t get_frontend_source(
 	}
 }
 
-void configure_encoder(
+void enc1_configure_encoder(
 	struct dcn10_link_encoder *enc10,
 	const struct dc_link_settings *link_settings)
 {
@@ -495,6 +496,15 @@ bool dcn10_is_dig_enabled(struct link_encoder *enc)
 	return value;
 }
 
+unsigned int dcn10_get_dig_frontend(struct link_encoder *enc)
+{
+	struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc);
+	uint32_t value;
+
+	REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &value);
+	return value;
+}
+
 static void link_encoder_disable(struct dcn10_link_encoder *enc10)
 {
 	/* reset training pattern */
@@ -543,12 +553,12 @@ bool dcn10_link_encoder_validate_dvi_output(
 	if ((connector_signal == SIGNAL_TYPE_DVI_SINGLE_LINK ||
 		connector_signal == SIGNAL_TYPE_HDMI_TYPE_A) &&
 		signal != SIGNAL_TYPE_HDMI_TYPE_A &&
-		crtc_timing->pix_clk_khz > TMDS_MAX_PIXEL_CLOCK)
+		crtc_timing->pix_clk_100hz > (TMDS_MAX_PIXEL_CLOCK * 10))
 		return false;
-	if (crtc_timing->pix_clk_khz < TMDS_MIN_PIXEL_CLOCK)
+	if (crtc_timing->pix_clk_100hz < (TMDS_MIN_PIXEL_CLOCK * 10))
 		return false;
 
-	if (crtc_timing->pix_clk_khz > max_pixel_clock)
+	if (crtc_timing->pix_clk_100hz > (max_pixel_clock * 10))
 		return false;
 
 	/* DVI supports 6/8bpp single-link and 10/16bpp dual-link */
@@ -571,7 +581,7 @@ bool dcn10_link_encoder_validate_dvi_output(
 static bool dcn10_link_encoder_validate_hdmi_output(
 	const struct dcn10_link_encoder *enc10,
 	const struct dc_crtc_timing *crtc_timing,
-	int adjusted_pix_clk_khz)
+	int adjusted_pix_clk_100hz)
 {
 	enum dc_color_depth max_deep_color =
 			enc10->base.features.max_hdmi_deep_color;
@@ -581,11 +591,11 @@ static bool dcn10_link_encoder_validate_hdmi_output(
 
 	if (crtc_timing->display_color_depth < COLOR_DEPTH_888)
 		return false;
-	if (adjusted_pix_clk_khz < TMDS_MIN_PIXEL_CLOCK)
+	if (adjusted_pix_clk_100hz < (TMDS_MIN_PIXEL_CLOCK * 10))
 		return false;
 
-	if ((adjusted_pix_clk_khz == 0) ||
-		(adjusted_pix_clk_khz > enc10->base.features.max_hdmi_pixel_clock))
+	if ((adjusted_pix_clk_100hz == 0) ||
+		(adjusted_pix_clk_100hz > (enc10->base.features.max_hdmi_pixel_clock * 10)))
 		return false;
 
 	/* DCE11 HW does not support 420 */
@@ -594,7 +604,7 @@ static bool dcn10_link_encoder_validate_hdmi_output(
 		return false;
 
 	if (!enc10->base.features.flags.bits.HDMI_6GB_EN &&
-		adjusted_pix_clk_khz >= 300000)
+		adjusted_pix_clk_100hz >= 3000000)
 		return false;
 	if (enc10->base.ctx->dc->debug.hdmi20_disable &&
 		crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
@@ -738,7 +748,7 @@ bool dcn10_link_encoder_validate_output_with_stream(
 	case SIGNAL_TYPE_DVI_DUAL_LINK:
 		is_valid = dcn10_link_encoder_validate_dvi_output(
 			enc10,
-			stream->sink->link->connector_signal,
+			stream->link->connector_signal,
 			stream->signal,
 			&stream->timing);
 	break;
@@ -746,7 +756,7 @@ bool dcn10_link_encoder_validate_output_with_stream(
 		is_valid = dcn10_link_encoder_validate_hdmi_output(
 				enc10,
 				&stream->timing,
-				stream->phy_pix_clk);
+				stream->phy_pix_clk * 10);
 	break;
 	case SIGNAL_TYPE_DISPLAY_PORT:
 	case SIGNAL_TYPE_DISPLAY_PORT_MST:
@@ -910,7 +920,7 @@ void dcn10_link_encoder_enable_dp_output(
 	 * but it's not passed to asic_control.
 	 * We need to set number of lanes manually.
 	 */
-	configure_encoder(enc10, link_settings);
+	enc1_configure_encoder(enc10, link_settings);
 
 	cntl.action = TRANSMITTER_CONTROL_ENABLE;
 	cntl.engine_id = enc->preferred_engine;
@@ -949,7 +959,7 @@ void dcn10_link_encoder_enable_dp_mst_output(
 	 * but it's not passed to asic_control.
 	 * We need to set number of lanes manually.
 	 */
-	configure_encoder(enc10, link_settings);
+	enc1_configure_encoder(enc10, link_settings);
 
 	cntl.action = TRANSMITTER_CONTROL_ENABLE;
 	cntl.engine_id = ENGINE_ID_UNKNOWN;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h
index 49ead12b2532..b74b80a247ec 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h
@@ -271,7 +271,7 @@ void dcn10_link_encoder_setup(
 	struct link_encoder *enc,
 	enum signal_type signal);
 
-void configure_encoder(
+void enc1_configure_encoder(
 	struct dcn10_link_encoder *enc10,
 	const struct dc_link_settings *link_settings);
 
@@ -336,6 +336,8 @@ void dcn10_psr_program_secondary_packet(struct link_encoder *enc,
 
 bool dcn10_is_dig_enabled(struct link_encoder *enc);
 
+unsigned int dcn10_get_dig_frontend(struct link_encoder *enc);
+
 void dcn10_aux_initialize(struct dcn10_link_encoder *enc10);
 
 #endif /* __DC_LINK_ENCODER__DCN10_H__ */
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 7c138615f17d..2f78a84f0dcb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -97,6 +97,7 @@ static uint32_t get_start_vline(struct timing_generator *optc, const struct dc_c
 	struct dc_crtc_timing patched_crtc_timing;
 	int vesa_sync_start;
 	int asic_blank_end;
+	int interlace_factor;
 	int vertical_line_start;
 
 	patched_crtc_timing = *dc_crtc_timing;
@@ -110,13 +111,16 @@ static uint32_t get_start_vline(struct timing_generator *optc, const struct dc_c
 			vesa_sync_start -
 			patched_crtc_timing.h_border_left;
 
+	interlace_factor = patched_crtc_timing.flags.INTERLACE ? 2 : 1;
+
 	vesa_sync_start = patched_crtc_timing.v_addressable +
 			patched_crtc_timing.v_border_bottom +
 			patched_crtc_timing.v_front_porch;
 
 	asic_blank_end = (patched_crtc_timing.v_total -
 			vesa_sync_start -
-			patched_crtc_timing.v_border_top);
+			patched_crtc_timing.v_border_top)
+			* interlace_factor;
 
 	vertical_line_start = asic_blank_end - optc->dlg_otg_param.vstartup_start + 1;
 	if (vertical_line_start < 0) {
@@ -127,40 +131,95 @@ static uint32_t get_start_vline(struct timing_generator *optc, const struct dc_c
 	return vertical_line_start;
 }
 
-void optc1_program_vline_interrupt(
+static void calc_vline_position(
 		struct timing_generator *optc,
 		const struct dc_crtc_timing *dc_crtc_timing,
-		unsigned long long vsync_delta)
+		unsigned long long vsync_delta,
+		uint32_t *start_line,
+		uint32_t *end_line)
 {
-
-	struct optc *optc1 = DCN10TG_FROM_TG(optc);
-
 	unsigned long long req_delta_tens_of_usec = div64_u64((vsync_delta + 9999), 10000);
-	unsigned long long pix_clk_hundreds_khz = div64_u64((dc_crtc_timing->pix_clk_khz + 99), 100);
+	unsigned long long pix_clk_hundreds_khz = div64_u64((dc_crtc_timing->pix_clk_100hz + 999), 1000);
 	uint32_t req_delta_lines = (uint32_t) div64_u64(
 			(req_delta_tens_of_usec * pix_clk_hundreds_khz + dc_crtc_timing->h_total - 1),
 								dc_crtc_timing->h_total);
 
 	uint32_t vsync_line = get_start_vline(optc, dc_crtc_timing);
-	uint32_t start_line = 0;
-	uint32_t endLine = 0;
 
 	if (req_delta_lines != 0)
-		req_delta_lines--;
+			req_delta_lines--;
 
-	if (req_delta_lines > vsync_line)
-		start_line = dc_crtc_timing->v_total - (req_delta_lines - vsync_line) + 2;
-	else
-		start_line = vsync_line - req_delta_lines;
+		if (req_delta_lines > vsync_line)
+			*start_line = dc_crtc_timing->v_total - (req_delta_lines - vsync_line) - 1;
+		else
+			*start_line = vsync_line - req_delta_lines;
+
+		*end_line = *start_line + 2;
+
+		if (*end_line >= dc_crtc_timing->v_total)
+			*end_line = 2;
+}
+
+void optc1_program_vline_interrupt(
+		struct timing_generator *optc,
+		const struct dc_crtc_timing *dc_crtc_timing,
+		enum vline_select vline,
+		const union vline_config *vline_config)
+{
+	struct optc *optc1 = DCN10TG_FROM_TG(optc);
+	uint32_t start_line = 0;
+	uint32_t end_line = 0;
+
+	switch (vline) {
+	case VLINE0:
+		calc_vline_position(optc, dc_crtc_timing, vline_config->delta_in_ns, &start_line, &end_line);
+		REG_SET_2(OTG_VERTICAL_INTERRUPT0_POSITION, 0,
+				OTG_VERTICAL_INTERRUPT0_LINE_START, start_line,
+				OTG_VERTICAL_INTERRUPT0_LINE_END, end_line);
+		break;
+	case VLINE1:
+		REG_SET(OTG_VERTICAL_INTERRUPT1_POSITION, 0,
+					OTG_VERTICAL_INTERRUPT1_LINE_START, vline_config->line_number);
+		break;
+	default:
+		break;
+	}
+}
+
+void optc1_program_vupdate_interrupt(
+		struct timing_generator *optc,
+		const struct dc_crtc_timing *dc_crtc_timing)
+{
+	struct optc *optc1 = DCN10TG_FROM_TG(optc);
+	int32_t vertical_line_start;
+	uint32_t asic_blank_end;
+	uint32_t vesa_sync_start;
+	struct dc_crtc_timing patched_crtc_timing;
+
+	patched_crtc_timing = *dc_crtc_timing;
+	optc1_apply_front_porch_workaround(optc, &patched_crtc_timing);
+
+	/* asic_h_blank_end = HsyncWidth + HbackPorch =
+	 * vesa. usHorizontalTotal - vesa. usHorizontalSyncStart -
+	 * vesa.h_left_border
+	 */
+	vesa_sync_start = patched_crtc_timing.h_addressable +
+			patched_crtc_timing.h_border_right +
+			patched_crtc_timing.h_front_porch;
 
-	endLine = start_line + 2;
+	asic_blank_end = patched_crtc_timing.h_total -
+			vesa_sync_start -
+			patched_crtc_timing.h_border_left;
 
-	if (endLine >= dc_crtc_timing->v_total)
-		endLine = 2;
+	/* Use OTG_VERTICAL_INTERRUPT2 replace VUPDATE interrupt,
+	 * program the reg for interrupt postition.
+	 */
+	vertical_line_start = asic_blank_end - optc->dlg_otg_param.vstartup_start + 1;
+	if (vertical_line_start < 0)
+		vertical_line_start = 0;
 
-	REG_SET_2(OTG_VERTICAL_INTERRUPT0_POSITION, 0,
-			OTG_VERTICAL_INTERRUPT0_LINE_START, start_line,
-			OTG_VERTICAL_INTERRUPT0_LINE_END, endLine);
+	REG_SET(OTG_VERTICAL_INTERRUPT2_POSITION, 0,
+			OTG_VERTICAL_INTERRUPT2_LINE_START, vertical_line_start);
 }
 
 /**
@@ -265,22 +324,14 @@ void optc1_program_timing(
 			patched_crtc_timing.v_addressable +
 			patched_crtc_timing.v_border_bottom);
 
-	REG_UPDATE_2(OTG_V_BLANK_START_END,
-			OTG_V_BLANK_START, asic_blank_start,
-			OTG_V_BLANK_END, asic_blank_end);
-
-	/* Use OTG_VERTICAL_INTERRUPT2 replace VUPDATE interrupt,
-	 * program the reg for interrupt postition.
-	 */
 	vertical_line_start = asic_blank_end - optc->dlg_otg_param.vstartup_start + 1;
 	v_fp2 = 0;
 	if (vertical_line_start < 0)
 		v_fp2 = -vertical_line_start;
-	if (vertical_line_start < 0)
-		vertical_line_start = 0;
 
-	REG_SET(OTG_VERTICAL_INTERRUPT2_POSITION, 0,
-			OTG_VERTICAL_INTERRUPT2_LINE_START, vertical_line_start);
+	REG_UPDATE_2(OTG_V_BLANK_START_END,
+			OTG_V_BLANK_START, asic_blank_start,
+			OTG_V_BLANK_END, asic_blank_end);
 
 	/* v_sync polarity */
 	v_sync_polarity = patched_crtc_timing.flags.VSYNC_POSITIVE_POLARITY ?
@@ -299,16 +350,17 @@ void optc1_program_timing(
 	}
 
 	/* Interlace */
-	if (patched_crtc_timing.flags.INTERLACE == 1) {
-		REG_UPDATE(OTG_INTERLACE_CONTROL,
-				OTG_INTERLACE_ENABLE, 1);
-		v_init = v_init / 2;
-		if ((optc->dlg_otg_param.vstartup_start/2)*2 > asic_blank_end)
-			v_fp2 = v_fp2 / 2;
-	} else
-		REG_UPDATE(OTG_INTERLACE_CONTROL,
-				OTG_INTERLACE_ENABLE, 0);
-
+	if (REG(OTG_INTERLACE_CONTROL)) {
+		if (patched_crtc_timing.flags.INTERLACE == 1) {
+			REG_UPDATE(OTG_INTERLACE_CONTROL,
+					OTG_INTERLACE_ENABLE, 1);
+			v_init = v_init / 2;
+			if ((optc->dlg_otg_param.vstartup_start/2)*2 > asic_blank_end)
+				v_fp2 = v_fp2 / 2;
+		} else
+			REG_UPDATE(OTG_INTERLACE_CONTROL,
+					OTG_INTERLACE_ENABLE, 0);
+	}
 
 	/* VTG enable set to 0 first VInit */
 	REG_UPDATE(CONTROL,
@@ -338,7 +390,7 @@ void optc1_program_timing(
 
 	h_div_2 = optc1_is_two_pixels_per_containter(&patched_crtc_timing);
 	REG_UPDATE(OTG_H_TIMING_CNTL,
-			OTG_H_TIMING_DIV_BY2, h_div_2);
+			OTG_H_TIMING_DIV_BY2, h_div_2 || optc1->comb_opp_id != 0xf);
 
 }
 
@@ -1184,6 +1236,64 @@ bool optc1_is_stereo_left_eye(struct timing_generator *optc)
 	return ret;
 }
 
+bool optc1_is_matching_timing(struct timing_generator *tg,
+		const struct dc_crtc_timing *otg_timing)
+{
+	struct dc_crtc_timing hw_crtc_timing = {0};
+	struct dcn_otg_state s = {0};
+
+	if (tg == NULL || otg_timing == NULL)
+		return false;
+
+	optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s);
+
+	hw_crtc_timing.h_total = s.h_total + 1;
+	hw_crtc_timing.h_addressable = s.h_total - ((s.h_total - s.h_blank_start) + s.h_blank_end);
+	hw_crtc_timing.h_front_porch = s.h_total + 1 - s.h_blank_start;
+	hw_crtc_timing.h_sync_width = s.h_sync_a_end - s.h_sync_a_start;
+
+	hw_crtc_timing.v_total = s.v_total + 1;
+	hw_crtc_timing.v_addressable = s.v_total - ((s.v_total - s.v_blank_start) + s.v_blank_end);
+	hw_crtc_timing.v_front_porch = s.v_total + 1 - s.v_blank_start;
+	hw_crtc_timing.v_sync_width = s.v_sync_a_end - s.v_sync_a_start;
+
+	if (otg_timing->h_total != hw_crtc_timing.h_total)
+		return false;
+
+	if (otg_timing->h_border_left != hw_crtc_timing.h_border_left)
+		return false;
+
+	if (otg_timing->h_addressable != hw_crtc_timing.h_addressable)
+		return false;
+
+	if (otg_timing->h_border_right != hw_crtc_timing.h_border_right)
+		return false;
+
+	if (otg_timing->h_front_porch != hw_crtc_timing.h_front_porch)
+		return false;
+
+	if (otg_timing->h_sync_width != hw_crtc_timing.h_sync_width)
+		return false;
+
+	if (otg_timing->v_total != hw_crtc_timing.v_total)
+		return false;
+
+	if (otg_timing->v_border_top != hw_crtc_timing.v_border_top)
+		return false;
+
+	if (otg_timing->v_addressable != hw_crtc_timing.v_addressable)
+		return false;
+
+	if (otg_timing->v_border_bottom != hw_crtc_timing.v_border_bottom)
+		return false;
+
+	if (otg_timing->v_sync_width != hw_crtc_timing.v_sync_width)
+		return false;
+
+	return true;
+}
+
+
 void optc1_read_otg_state(struct optc *optc1,
 		struct dcn_otg_state *s)
 {
@@ -1371,6 +1481,7 @@ static const struct timing_generator_funcs dcn10_tg_funcs = {
 		.validate_timing = optc1_validate_timing,
 		.program_timing = optc1_program_timing,
 		.program_vline_interrupt = optc1_program_vline_interrupt,
+		.program_vupdate_interrupt = optc1_program_vupdate_interrupt,
 		.program_global_sync = optc1_program_global_sync,
 		.enable_crtc = optc1_enable_crtc,
 		.disable_crtc = optc1_disable_crtc,
@@ -1380,6 +1491,7 @@ static const struct timing_generator_funcs dcn10_tg_funcs = {
 		.get_frame_count = optc1_get_vblank_counter,
 		.get_scanoutpos = optc1_get_crtc_scanoutpos,
 		.get_otg_active_size = optc1_get_otg_active_size,
+		.is_matching_timing = optc1_is_matching_timing,
 		.set_early_control = optc1_set_early_control,
 		/* used by enable_timing_synchronization. Not need for FPGA */
 		.wait_for_state = optc1_wait_for_state,
@@ -1419,10 +1531,13 @@ void dcn10_timing_generator_init(struct optc *optc1)
 	optc1->min_v_blank_interlace = 5;
 	optc1->min_h_sync_width = 8;
 	optc1->min_v_sync_width = 1;
+	optc1->comb_opp_id = 0xf;
 }
 
 bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing)
 {
-	return timing->pixel_encoding == PIXEL_ENCODING_YCBCR420;
+	bool two_pix = timing->pixel_encoding == PIXEL_ENCODING_YCBCR420;
+
+	return two_pix;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
index 8bacf0b6e27e..24452f11c598 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
@@ -67,6 +67,8 @@
 	SRI(OTG_CLOCK_CONTROL, OTG, inst),\
 	SRI(OTG_VERTICAL_INTERRUPT0_CONTROL, OTG, inst),\
 	SRI(OTG_VERTICAL_INTERRUPT0_POSITION, OTG, inst),\
+	SRI(OTG_VERTICAL_INTERRUPT1_CONTROL, OTG, inst),\
+	SRI(OTG_VERTICAL_INTERRUPT1_POSITION, OTG, inst),\
 	SRI(OTG_VERTICAL_INTERRUPT2_CONTROL, OTG, inst),\
 	SRI(OTG_VERTICAL_INTERRUPT2_POSITION, OTG, inst),\
 	SRI(OPTC_INPUT_CLOCK_CONTROL, ODM, inst),\
@@ -135,6 +137,8 @@ struct dcn_optc_registers {
 	uint32_t OTG_CLOCK_CONTROL;
 	uint32_t OTG_VERTICAL_INTERRUPT0_CONTROL;
 	uint32_t OTG_VERTICAL_INTERRUPT0_POSITION;
+	uint32_t OTG_VERTICAL_INTERRUPT1_CONTROL;
+	uint32_t OTG_VERTICAL_INTERRUPT1_POSITION;
 	uint32_t OTG_VERTICAL_INTERRUPT2_CONTROL;
 	uint32_t OTG_VERTICAL_INTERRUPT2_POSITION;
 	uint32_t OPTC_INPUT_CLOCK_CONTROL;
@@ -227,6 +231,8 @@ struct dcn_optc_registers {
 	SF(OTG0_OTG_VERTICAL_INTERRUPT0_CONTROL, OTG_VERTICAL_INTERRUPT0_INT_ENABLE, mask_sh),\
 	SF(OTG0_OTG_VERTICAL_INTERRUPT0_POSITION, OTG_VERTICAL_INTERRUPT0_LINE_START, mask_sh),\
 	SF(OTG0_OTG_VERTICAL_INTERRUPT0_POSITION, OTG_VERTICAL_INTERRUPT0_LINE_END, mask_sh),\
+	SF(OTG0_OTG_VERTICAL_INTERRUPT1_CONTROL, OTG_VERTICAL_INTERRUPT1_INT_ENABLE, mask_sh),\
+	SF(OTG0_OTG_VERTICAL_INTERRUPT1_POSITION, OTG_VERTICAL_INTERRUPT1_LINE_START, mask_sh),\
 	SF(OTG0_OTG_VERTICAL_INTERRUPT2_CONTROL, OTG_VERTICAL_INTERRUPT2_INT_ENABLE, mask_sh),\
 	SF(OTG0_OTG_VERTICAL_INTERRUPT2_POSITION, OTG_VERTICAL_INTERRUPT2_LINE_START, mask_sh),\
 	SF(ODM0_OPTC_INPUT_CLOCK_CONTROL, OPTC_INPUT_CLK_EN, mask_sh),\
@@ -361,6 +367,8 @@ struct dcn_optc_registers {
 	type OTG_VERTICAL_INTERRUPT0_INT_ENABLE;\
 	type OTG_VERTICAL_INTERRUPT0_LINE_START;\
 	type OTG_VERTICAL_INTERRUPT0_LINE_END;\
+	type OTG_VERTICAL_INTERRUPT1_INT_ENABLE;\
+	type OTG_VERTICAL_INTERRUPT1_LINE_START;\
 	type OTG_VERTICAL_INTERRUPT2_INT_ENABLE;\
 	type OTG_VERTICAL_INTERRUPT2_LINE_START;\
 	type OPTC_INPUT_CLK_EN;\
@@ -427,7 +435,7 @@ struct optc {
 	const struct dcn_optc_shift *tg_shift;
 	const struct dcn_optc_mask *tg_mask;
 
-	enum controller_id controller_id;
+	int comb_opp_id;
 
 	uint32_t max_h_total;
 	uint32_t max_v_total;
@@ -475,9 +483,11 @@ void optc1_program_timing(
 	const struct dc_crtc_timing *dc_crtc_timing,
 	bool use_vbios);
 
-void optc1_program_vline_interrupt(struct timing_generator *optc,
+void optc1_program_vline_interrupt(
+		struct timing_generator *optc,
 		const struct dc_crtc_timing *dc_crtc_timing,
-		unsigned long long vsync_delta);
+		enum vline_select vline,
+		const union vline_config *vline_config);
 
 void optc1_program_global_sync(
 		struct timing_generator *optc);
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 5d4772dec0ba..09d74070a49b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -70,7 +70,7 @@
 const struct _vcs_dpi_ip_params_st dcn1_0_ip = {
 	.rob_buffer_size_kbytes = 64,
 	.det_buffer_size_kbytes = 164,
-	.dpte_buffer_size_in_pte_reqs = 42,
+	.dpte_buffer_size_in_pte_reqs_luma = 42,
 	.dpp_output_buffer_pixels = 2560,
 	.opp_output_buffer_lines = 1,
 	.pixel_chunk_size_kbytes = 8,
@@ -436,7 +436,6 @@ static const struct dcn_optc_mask tg_mask = {
 };
 
 static const struct bios_registers bios_regs = {
-		NBIO_SR(BIOS_SCRATCH_0),
 		NBIO_SR(BIOS_SCRATCH_3),
 		NBIO_SR(BIOS_SCRATCH_6)
 };
@@ -609,7 +608,7 @@ static struct output_pixel_processor *dcn10_opp_create(
 	return &opp->base;
 }
 
-struct aux_engine *dcn10_aux_engine_create(
+struct dce_aux *dcn10_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
 {
@@ -678,18 +677,18 @@ static struct mpc *dcn10_mpc_create(struct dc_context *ctx)
 
 static struct hubbub *dcn10_hubbub_create(struct dc_context *ctx)
 {
-	struct hubbub *hubbub = kzalloc(sizeof(struct hubbub),
+	struct dcn10_hubbub *dcn10_hubbub = kzalloc(sizeof(struct dcn10_hubbub),
 					  GFP_KERNEL);
 
-	if (!hubbub)
+	if (!dcn10_hubbub)
 		return NULL;
 
-	hubbub1_construct(hubbub, ctx,
+	hubbub1_construct(&dcn10_hubbub->base, ctx,
 			&hubbub_reg,
 			&hubbub_shift,
 			&hubbub_mask);
 
-	return hubbub;
+	return &dcn10_hubbub->base;
 }
 
 static struct timing_generator *dcn10_timing_generator_create(
@@ -911,7 +910,7 @@ static void destruct(struct dcn10_resource_pool *pool)
 
 	for (i = 0; i < pool->base.res_cap->num_ddc; i++) {
 		if (pool->base.engines[i] != NULL)
-			pool->base.engines[i]->funcs->destroy_engine(&pool->base.engines[i]);
+			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;
@@ -974,8 +973,8 @@ static void get_pixel_clock_parameters(
 	struct pixel_clk_params *pixel_clk_params)
 {
 	const struct dc_stream_state *stream = pipe_ctx->stream;
-	pixel_clk_params->requested_pix_clk = stream->timing.pix_clk_khz;
-	pixel_clk_params->encoder_object_id = stream->sink->link->link_enc->id;
+	pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz;
+	pixel_clk_params->encoder_object_id = stream->link->link_enc->id;
 	pixel_clk_params->signal_type = pipe_ctx->stream->signal;
 	pixel_clk_params->controller_id = pipe_ctx->stream_res.tg->inst + 1;
 	/* TODO: un-hardcode*/
@@ -991,9 +990,9 @@ static void get_pixel_clock_parameters(
 		pixel_clk_params->color_depth = COLOR_DEPTH_888;
 
 	if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		pixel_clk_params->requested_pix_clk  /= 2;
+		pixel_clk_params->requested_pix_clk_100hz  /= 2;
 	if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
-		pixel_clk_params->requested_pix_clk *= 2;
+		pixel_clk_params->requested_pix_clk_100hz *= 2;
 
 }
 
@@ -1131,6 +1130,56 @@ static enum dc_status dcn10_validate_plane(const struct dc_plane_state *plane_st
 	return DC_OK;
 }
 
+static enum dc_status dcn10_validate_global(struct dc *dc, struct dc_state *context)
+{
+	int i, j;
+	bool video_down_scaled = false;
+	bool video_large = false;
+	bool desktop_large = false;
+	bool dcc_disabled = false;
+
+	for (i = 0; i < context->stream_count; i++) {
+		if (context->stream_status[i].plane_count == 0)
+			continue;
+
+		if (context->stream_status[i].plane_count > 2)
+			return false;
+
+		for (j = 0; j < context->stream_status[i].plane_count; j++) {
+			struct dc_plane_state *plane =
+				context->stream_status[i].plane_states[j];
+
+
+			if (plane->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
+
+				if (plane->src_rect.width > plane->dst_rect.width ||
+						plane->src_rect.height > plane->dst_rect.height)
+					video_down_scaled = true;
+
+				if (plane->src_rect.width >= 3840)
+					video_large = true;
+
+			} else {
+				if (plane->src_rect.width >= 3840)
+					desktop_large = true;
+				if (!plane->dcc.enable)
+					dcc_disabled = true;
+			}
+		}
+	}
+
+	/*
+	 * Workaround: On DCN10 there is UMC issue that causes underflow when
+	 * playing 4k video on 4k desktop with video downscaled and single channel
+	 * memory
+	 */
+	if (video_large && desktop_large && video_down_scaled && dcc_disabled &&
+			dc->dcn_soc->number_of_channels == 1)
+		return DC_FAIL_SURFACE_VALIDATE;
+
+	return DC_OK;
+}
+
 static enum dc_status dcn10_get_default_swizzle_mode(struct dc_plane_state *plane_state)
 {
 	enum dc_status result = DC_OK;
@@ -1159,6 +1208,7 @@ static const struct resource_funcs dcn10_res_pool_funcs = {
 	.validate_bandwidth = dcn_validate_bandwidth,
 	.acquire_idle_pipe_for_layer = dcn10_acquire_idle_pipe_for_layer,
 	.validate_plane = dcn10_validate_plane,
+	.validate_global = dcn10_validate_global,
 	.add_stream_to_ctx = dcn10_add_stream_to_ctx,
 	.get_default_swizzle_mode = dcn10_get_default_swizzle_mode
 };
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 b8b5525a389a..b08254121251 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
@@ -261,17 +261,29 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 	uint8_t dp_component_depth = 0;
 
 	struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc);
+	struct dc_crtc_timing hw_crtc_timing = *crtc_timing;
+
+	if (hw_crtc_timing.flags.INTERLACE) {
+		/*the input timing is in VESA spec format with Interlace flag =1*/
+		hw_crtc_timing.v_total /= 2;
+		hw_crtc_timing.v_border_top /= 2;
+		hw_crtc_timing.v_addressable /= 2;
+		hw_crtc_timing.v_border_bottom /= 2;
+		hw_crtc_timing.v_front_porch /= 2;
+		hw_crtc_timing.v_sync_width /= 2;
+	}
+
 
 	/* set pixel encoding */
-	switch (crtc_timing->pixel_encoding) {
+	switch (hw_crtc_timing.pixel_encoding) {
 	case PIXEL_ENCODING_YCBCR422:
 		dp_pixel_encoding = DP_PIXEL_ENCODING_TYPE_YCBCR422;
 		break;
 	case PIXEL_ENCODING_YCBCR444:
 		dp_pixel_encoding = DP_PIXEL_ENCODING_TYPE_YCBCR444;
 
-		if (crtc_timing->flags.Y_ONLY)
-			if (crtc_timing->display_color_depth != COLOR_DEPTH_666)
+		if (hw_crtc_timing.flags.Y_ONLY)
+			if (hw_crtc_timing.display_color_depth != COLOR_DEPTH_666)
 				/* HW testing only, no use case yet.
 				 * Color depth of Y-only could be
 				 * 8, 10, 12, 16 bits
@@ -299,7 +311,7 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 	 * Pixel Encoding/Colorimetry Format and that a Sink device shall ignore MISC1, bit 7,
 	 * and MISC0, bits 7:1 (MISC1, bit 7, and MISC0, bits 7:1, become "don't care").
 	 */
-	if ((crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) ||
+	if ((hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) ||
 			(output_color_space == COLOR_SPACE_2020_YCBCR) ||
 			(output_color_space == COLOR_SPACE_2020_RGB_FULLRANGE) ||
 			(output_color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE))
@@ -308,7 +320,7 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 		misc1 = misc1 & ~0x40;
 
 	/* set color depth */
-	switch (crtc_timing->display_color_depth) {
+	switch (hw_crtc_timing.display_color_depth) {
 	case COLOR_DEPTH_666:
 		dp_component_depth = DP_COMPONENT_PIXEL_DEPTH_6BPC;
 		break;
@@ -336,7 +348,7 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 
 	/* set dynamic range and YCbCr range */
 
-	switch (crtc_timing->display_color_depth) {
+	switch (hw_crtc_timing.display_color_depth) {
 	case COLOR_DEPTH_666:
 		colorimetry_bpc = 0;
 		break;
@@ -372,9 +384,9 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 		misc0 = misc0 | 0x8; /* bit3=1, bit4=0 */
 		misc1 = misc1 & ~0x80; /* bit7 = 0*/
 		dynamic_range_ycbcr = 0; /*bt601*/
-		if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+		if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR422)
 			misc0 = misc0 | 0x2; /* bit2=0, bit1=1 */
-		else if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR444)
+		else if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR444)
 			misc0 = misc0 | 0x4; /* bit2=1, bit1=0 */
 		break;
 	case COLOR_SPACE_YCBCR709:
@@ -382,9 +394,9 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 		misc0 = misc0 | 0x18; /* bit3=1, bit4=1 */
 		misc1 = misc1 & ~0x80; /* bit7 = 0*/
 		dynamic_range_ycbcr = 1; /*bt709*/
-		if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+		if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR422)
 			misc0 = misc0 | 0x2; /* bit2=0, bit1=1 */
-		else if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR444)
+		else if (hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR444)
 			misc0 = misc0 | 0x4; /* bit2=1, bit1=0 */
 		break;
 	case COLOR_SPACE_2020_RGB_LIMITEDRANGE:
@@ -414,26 +426,26 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 	 * dc_crtc_timing is vesa dmt struct. data from edid
 	 */
 	REG_SET_2(DP_MSA_TIMING_PARAM1, 0,
-			DP_MSA_HTOTAL, crtc_timing->h_total,
-			DP_MSA_VTOTAL, crtc_timing->v_total);
+			DP_MSA_HTOTAL, hw_crtc_timing.h_total,
+			DP_MSA_VTOTAL, hw_crtc_timing.v_total);
 
 	/* calculate from vesa timing parameters
 	 * h_active_start related to leading edge of sync
 	 */
 
-	h_blank = crtc_timing->h_total - crtc_timing->h_border_left -
-			crtc_timing->h_addressable - crtc_timing->h_border_right;
+	h_blank = hw_crtc_timing.h_total - hw_crtc_timing.h_border_left -
+			hw_crtc_timing.h_addressable - hw_crtc_timing.h_border_right;
 
-	h_back_porch = h_blank - crtc_timing->h_front_porch -
-			crtc_timing->h_sync_width;
+	h_back_porch = h_blank - hw_crtc_timing.h_front_porch -
+			hw_crtc_timing.h_sync_width;
 
 	/* start at beginning of left border */
-	h_active_start = crtc_timing->h_sync_width + h_back_porch;
+	h_active_start = hw_crtc_timing.h_sync_width + h_back_porch;
 
 
-	v_active_start = crtc_timing->v_total - crtc_timing->v_border_top -
-			crtc_timing->v_addressable - crtc_timing->v_border_bottom -
-			crtc_timing->v_front_porch;
+	v_active_start = hw_crtc_timing.v_total - hw_crtc_timing.v_border_top -
+			hw_crtc_timing.v_addressable - hw_crtc_timing.v_border_bottom -
+			hw_crtc_timing.v_front_porch;
 
 
 	/* start at beginning of left border */
@@ -443,20 +455,20 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 
 	REG_SET_4(DP_MSA_TIMING_PARAM3, 0,
 			DP_MSA_HSYNCWIDTH,
-			crtc_timing->h_sync_width,
+			hw_crtc_timing.h_sync_width,
 			DP_MSA_HSYNCPOLARITY,
-			!crtc_timing->flags.HSYNC_POSITIVE_POLARITY,
+			!hw_crtc_timing.flags.HSYNC_POSITIVE_POLARITY,
 			DP_MSA_VSYNCWIDTH,
-			crtc_timing->v_sync_width,
+			hw_crtc_timing.v_sync_width,
 			DP_MSA_VSYNCPOLARITY,
-			!crtc_timing->flags.VSYNC_POSITIVE_POLARITY);
+			!hw_crtc_timing.flags.VSYNC_POSITIVE_POLARITY);
 
 	/* HWDITH include border or overscan */
 	REG_SET_2(DP_MSA_TIMING_PARAM4, 0,
-		DP_MSA_HWIDTH, crtc_timing->h_border_left +
-		crtc_timing->h_addressable + crtc_timing->h_border_right,
-		DP_MSA_VHEIGHT, crtc_timing->v_border_top +
-		crtc_timing->v_addressable + crtc_timing->v_border_bottom);
+		DP_MSA_HWIDTH, hw_crtc_timing.h_border_left +
+		hw_crtc_timing.h_addressable + hw_crtc_timing.h_border_right,
+		DP_MSA_VHEIGHT, hw_crtc_timing.v_border_top +
+		hw_crtc_timing.v_addressable + hw_crtc_timing.v_border_bottom);
 }
 
 static void enc1_stream_encoder_set_stream_attribute_helper(
@@ -594,7 +606,7 @@ void enc1_stream_encoder_dvi_set_stream_attribute(
 	cntl.signal = is_dual_link ?
 			SIGNAL_TYPE_DVI_DUAL_LINK : SIGNAL_TYPE_DVI_SINGLE_LINK;
 	cntl.enable_dp_audio = false;
-	cntl.pixel_clock = crtc_timing->pix_clk_khz;
+	cntl.pixel_clock = crtc_timing->pix_clk_100hz / 10;
 	cntl.lanes_number = (is_dual_link) ? LANE_COUNT_EIGHT : LANE_COUNT_FOUR;
 
 	if (enc1->base.bp->funcs->encoder_control(
@@ -1413,6 +1425,14 @@ void enc1_setup_stereo_sync(
 	REG_UPDATE(DIG_FE_CNTL, DIG_STEREOSYNC_GATE_EN, !enable);
 }
 
+void enc1_dig_connect_to_otg(
+	struct stream_encoder *enc,
+	int tg_inst)
+{
+	struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc);
+
+	REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst);
+}
 
 static const struct stream_encoder_funcs dcn10_str_enc_funcs = {
 	.dp_set_stream_attribute =
@@ -1445,6 +1465,7 @@ static const struct stream_encoder_funcs dcn10_str_enc_funcs = {
 	.hdmi_audio_disable = enc1_se_hdmi_audio_disable,
 	.setup_stereo_sync  = enc1_setup_stereo_sync,
 	.set_avmute = enc1_stream_encoder_set_avmute,
+	.dig_connect_to_otg  = enc1_dig_connect_to_otg,
 };
 
 void dcn10_stream_encoder_construct(
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 67f3e4dd95c1..b7c800e10a32 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
@@ -274,7 +274,8 @@ struct dcn10_stream_enc_registers {
 	SE_SF(DP0_DP_MSA_TIMING_PARAM4, DP_MSA_HWIDTH, mask_sh),\
 	SE_SF(DP0_DP_MSA_TIMING_PARAM4, DP_MSA_VHEIGHT, mask_sh),\
 	SE_SF(DIG0_HDMI_DB_CONTROL, HDMI_DB_DISABLE, mask_sh),\
-	SE_SF(DP0_DP_VID_TIMING, DP_VID_N_MUL, mask_sh)
+	SE_SF(DP0_DP_VID_TIMING, DP_VID_N_MUL, mask_sh),\
+	SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh)
 
 #define SE_COMMON_MASK_SH_LIST_SOC(mask_sh)\
 	SE_COMMON_MASK_SH_LIST_SOC_BASE(mask_sh)
@@ -426,7 +427,8 @@ struct dcn10_stream_enc_registers {
 	type DP_MSA_VHEIGHT;\
 	type HDMI_DB_DISABLE;\
 	type DP_VID_N_MUL;\
-	type DP_VID_M_DOUBLE_VALUE_EN
+	type DP_VID_M_DOUBLE_VALUE_EN;\
+	type DIG_SOURCE_SELECT
 
 struct dcn10_stream_encoder_shift {
 	SE_REG_FIELD_LIST_DCN1_0(uint8_t);
@@ -523,4 +525,8 @@ void enc1_se_hdmi_audio_setup(
 void enc1_se_hdmi_audio_disable(
 	struct stream_encoder *enc);
 
+void enc1_dig_connect_to_otg(
+	struct stream_encoder *enc,
+	int tg_inst);
+
 #endif /* __DC_STREAM_ENCODER_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
index 5d4527d03045..e81b24374bcb 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
@@ -58,6 +58,13 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 		bool enable);
 
 /*
+ * poll pending down reply before clear payload allocation table
+ */
+void dm_helpers_dp_mst_poll_pending_down_reply(
+	struct dc_context *ctx,
+	const struct dc_link *link);
+
+/*
  * Clear payload allocation table before enable MST DP link.
  */
 void dm_helpers_dp_mst_clear_payload_allocation_table(
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 0029a39efb1c..14bed5b1fa97 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
@@ -38,7 +38,8 @@ enum pp_smu_ver {
 	 * of interface sharing between families of ASIcs.
 	 */
 	PP_SMU_UNSUPPORTED,
-	PP_SMU_VER_RV
+	PP_SMU_VER_RV,
+	PP_SMU_VER_MAX
 };
 
 struct pp_smu {
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
index 1af8c777b3ac..77200711abbe 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
@@ -82,9 +82,17 @@ enum dm_pp_clock_type {
 #define DC_DECODE_PP_CLOCK_TYPE(clk_type) \
 	(clk_type) == DM_PP_CLOCK_TYPE_DISPLAY_CLK ? "Display" : \
 	(clk_type) == DM_PP_CLOCK_TYPE_ENGINE_CLK ? "Engine" : \
-	(clk_type) == DM_PP_CLOCK_TYPE_MEMORY_CLK ? "Memory" : "Invalid"
-
-#define DM_PP_MAX_CLOCK_LEVELS 8
+	(clk_type) == DM_PP_CLOCK_TYPE_MEMORY_CLK ? "Memory" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_DCFCLK ? "DCF" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_DCEFCLK ? "DCEF" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_SOCCLK ? "SoC" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_PIXELCLK ? "Pixel" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_DISPLAYPHYCLK ? "Display PHY" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_DPPCLK ? "DPP" : \
+	(clk_type) == DM_PP_CLOCK_TYPE_FCLK ? "F" : \
+	"Invalid"
+
+#define DM_PP_MAX_CLOCK_LEVELS 16
 
 struct dm_pp_clock_levels {
 	uint32_t num_levels;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
index bea4e61b94c7..c59e582c1f40 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
@@ -121,4 +121,30 @@ enum self_refresh_affinity {
 	dm_neither_self_refresh_nor_mclk_switch
 };
 
+enum dm_validation_status {
+	DML_VALIDATION_OK,
+	DML_FAIL_SCALE_RATIO_TAP,
+	DML_FAIL_SOURCE_PIXEL_FORMAT,
+	DML_FAIL_VIEWPORT_SIZE,
+	DML_FAIL_TOTAL_V_ACTIVE_BW,
+	DML_FAIL_DIO_SUPPORT,
+	DML_FAIL_NOT_ENOUGH_DSC,
+	DML_FAIL_DSC_CLK_REQUIRED,
+	DML_FAIL_URGENT_LATENCY,
+	DML_FAIL_REORDERING_BUFFER,
+	DML_FAIL_DISPCLK_DPPCLK,
+	DML_FAIL_TOTAL_AVAILABLE_PIPES,
+	DML_FAIL_NUM_OTG,
+	DML_FAIL_WRITEBACK_MODE,
+	DML_FAIL_WRITEBACK_LATENCY,
+	DML_FAIL_WRITEBACK_SCALE_RATIO_TAP,
+	DML_FAIL_CURSOR_SUPPORT,
+	DML_FAIL_PITCH_SUPPORT,
+	DML_FAIL_PTE_BUFFER_SIZE,
+	DML_FAIL_HOST_VM_IMMEDIATE_FLIP,
+	DML_FAIL_DSC_INPUT_BPC,
+	DML_FAIL_PREFETCH_SUPPORT,
+	DML_FAIL_V_RATIO_PREFETCH,
+};
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
index dddeb0d4db8f..d303b789adfe 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
@@ -62,3 +62,31 @@ void dml_init_instance(struct display_mode_lib *lib, enum dml_project project)
 	}
 }
 
+const char *dml_get_status_message(enum dm_validation_status status)
+{
+	switch (status) {
+	case DML_VALIDATION_OK:                   return "Validation OK";
+	case DML_FAIL_SCALE_RATIO_TAP:            return "Scale ratio/tap";
+	case DML_FAIL_SOURCE_PIXEL_FORMAT:        return "Source pixel format";
+	case DML_FAIL_VIEWPORT_SIZE:              return "Viewport size";
+	case DML_FAIL_TOTAL_V_ACTIVE_BW:          return "Total vertical active bandwidth";
+	case DML_FAIL_DIO_SUPPORT:                return "DIO support";
+	case DML_FAIL_NOT_ENOUGH_DSC:             return "Not enough DSC Units";
+	case DML_FAIL_DSC_CLK_REQUIRED:           return "DSC clock required";
+	case DML_FAIL_URGENT_LATENCY:             return "Urgent latency";
+	case DML_FAIL_REORDERING_BUFFER:          return "Re-ordering buffer";
+	case DML_FAIL_DISPCLK_DPPCLK:             return "Dispclk and Dppclk";
+	case DML_FAIL_TOTAL_AVAILABLE_PIPES:      return "Total available pipes";
+	case DML_FAIL_NUM_OTG:                    return "Number of OTG";
+	case DML_FAIL_WRITEBACK_MODE:             return "Writeback mode";
+	case DML_FAIL_WRITEBACK_LATENCY:          return "Writeback latency";
+	case DML_FAIL_WRITEBACK_SCALE_RATIO_TAP:  return "Writeback scale ratio/tap";
+	case DML_FAIL_CURSOR_SUPPORT:             return "Cursor support";
+	case DML_FAIL_PITCH_SUPPORT:              return "Pitch support";
+	case DML_FAIL_PTE_BUFFER_SIZE:            return "PTE buffer size";
+	case DML_FAIL_DSC_INPUT_BPC:              return "DSC input bpc";
+	case DML_FAIL_PREFETCH_SUPPORT:           return "Prefetch support";
+	case DML_FAIL_V_RATIO_PREFETCH:           return "Vertical ratio prefetch";
+	default:                                  return "Unknown Status";
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
index 635206248889..a730e0209c05 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
@@ -43,4 +43,6 @@ struct display_mode_lib {
 
 void dml_init_instance(struct display_mode_lib *lib, enum dml_project project);
 
+const char *dml_get_status_message(enum dm_validation_status status);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
index 5dd04520ceca..391183e3428f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
@@ -30,22 +30,15 @@ typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st;
 typedef struct _vcs_dpi_ip_params_st ip_params_st;
 typedef struct _vcs_dpi_display_pipe_source_params_st display_pipe_source_params_st;
 typedef struct _vcs_dpi_display_output_params_st display_output_params_st;
-typedef struct _vcs_dpi_display_bandwidth_st display_bandwidth_st;
 typedef struct _vcs_dpi_scaler_ratio_depth_st scaler_ratio_depth_st;
 typedef struct _vcs_dpi_scaler_taps_st scaler_taps_st;
 typedef struct _vcs_dpi_display_pipe_dest_params_st display_pipe_dest_params_st;
 typedef struct _vcs_dpi_display_pipe_params_st display_pipe_params_st;
 typedef struct _vcs_dpi_display_clocks_and_cfg_st display_clocks_and_cfg_st;
 typedef struct _vcs_dpi_display_e2e_pipe_params_st display_e2e_pipe_params_st;
-typedef struct _vcs_dpi_dchub_buffer_sizing_st dchub_buffer_sizing_st;
-typedef struct _vcs_dpi_watermarks_perf_st watermarks_perf_st;
-typedef struct _vcs_dpi_cstate_pstate_watermarks_st cstate_pstate_watermarks_st;
-typedef struct _vcs_dpi_wm_calc_pipe_params_st wm_calc_pipe_params_st;
-typedef struct _vcs_dpi_vratio_pre_st vratio_pre_st;
 typedef struct _vcs_dpi_display_data_rq_misc_params_st display_data_rq_misc_params_st;
 typedef struct _vcs_dpi_display_data_rq_sizing_params_st display_data_rq_sizing_params_st;
 typedef struct _vcs_dpi_display_data_rq_dlg_params_st display_data_rq_dlg_params_st;
-typedef struct _vcs_dpi_display_cur_rq_dlg_params_st display_cur_rq_dlg_params_st;
 typedef struct _vcs_dpi_display_rq_dlg_params_st display_rq_dlg_params_st;
 typedef struct _vcs_dpi_display_rq_sizing_params_st display_rq_sizing_params_st;
 typedef struct _vcs_dpi_display_rq_misc_params_st display_rq_misc_params_st;
@@ -55,8 +48,6 @@ typedef struct _vcs_dpi_display_ttu_regs_st display_ttu_regs_st;
 typedef struct _vcs_dpi_display_data_rq_regs_st display_data_rq_regs_st;
 typedef struct _vcs_dpi_display_rq_regs_st display_rq_regs_st;
 typedef struct _vcs_dpi_display_dlg_sys_params_st display_dlg_sys_params_st;
-typedef struct _vcs_dpi_display_dlg_prefetch_param_st display_dlg_prefetch_param_st;
-typedef struct _vcs_dpi_display_pipe_clock_st display_pipe_clock_st;
 typedef struct _vcs_dpi_display_arb_params_st display_arb_params_st;
 
 struct _vcs_dpi_voltage_scaling_st {
@@ -111,8 +102,6 @@ struct _vcs_dpi_soc_bounding_box_st {
 	double xfc_bus_transport_time_us;
 	double xfc_xbuf_latency_tolerance_us;
 	int use_urgent_burst_bw;
-	double max_hscl_ratio;
-	double max_vscl_ratio;
 	unsigned int num_states;
 	struct _vcs_dpi_voltage_scaling_st clock_limits[8];
 };
@@ -129,7 +118,8 @@ struct _vcs_dpi_ip_params_st {
 	unsigned int odm_capable;
 	unsigned int rob_buffer_size_kbytes;
 	unsigned int det_buffer_size_kbytes;
-	unsigned int dpte_buffer_size_in_pte_reqs;
+	unsigned int dpte_buffer_size_in_pte_reqs_luma;
+	unsigned int dpte_buffer_size_in_pte_reqs_chroma;
 	unsigned int pde_proc_buffer_size_64k_reqs;
 	unsigned int dpp_output_buffer_pixels;
 	unsigned int opp_output_buffer_lines;
@@ -192,7 +182,6 @@ struct _vcs_dpi_display_xfc_params_st {
 struct _vcs_dpi_display_pipe_source_params_st {
 	int source_format;
 	unsigned char dcc;
-	unsigned int dcc_override;
 	unsigned int dcc_rate;
 	unsigned char dcc_use_global;
 	unsigned char vm;
@@ -205,7 +194,6 @@ struct _vcs_dpi_display_pipe_source_params_st {
 	int source_scan;
 	int sw_mode;
 	int macro_tile_size;
-	unsigned char is_display_sw;
 	unsigned int viewport_width;
 	unsigned int viewport_height;
 	unsigned int viewport_y_y;
@@ -252,16 +240,10 @@ struct _vcs_dpi_display_output_params_st {
 	int output_bpc;
 	int output_type;
 	int output_format;
-	int output_standard;
 	int dsc_slices;
 	struct writeback_st wb;
 };
 
-struct _vcs_dpi_display_bandwidth_st {
-	double total_bw_consumed_gbps;
-	double guaranteed_urgent_return_bw_gbps;
-};
-
 struct _vcs_dpi_scaler_ratio_depth_st {
 	double hscl_ratio;
 	double vscl_ratio;
@@ -300,11 +282,9 @@ struct _vcs_dpi_display_pipe_dest_params_st {
 	unsigned int vupdate_width;
 	unsigned int vready_offset;
 	unsigned char interlaced;
-	unsigned char underscan;
 	double pixel_rate_mhz;
 	unsigned char synchronized_vblank_all_planes;
 	unsigned char otg_inst;
-	unsigned char odm_split_cnt;
 	unsigned char odm_combine;
 	unsigned char use_maximum_vstartup;
 };
@@ -331,65 +311,6 @@ struct _vcs_dpi_display_e2e_pipe_params_st {
 	display_clocks_and_cfg_st clks_cfg;
 };
 
-struct _vcs_dpi_dchub_buffer_sizing_st {
-	unsigned int swath_width_y;
-	unsigned int swath_height_y;
-	unsigned int swath_height_c;
-	unsigned int detail_buffer_size_y;
-};
-
-struct _vcs_dpi_watermarks_perf_st {
-	double stutter_eff_in_active_region_percent;
-	double urgent_latency_supported_us;
-	double non_urgent_latency_supported_us;
-	double dram_clock_change_margin_us;
-	double dram_access_eff_percent;
-};
-
-struct _vcs_dpi_cstate_pstate_watermarks_st {
-	double cstate_exit_us;
-	double cstate_enter_plus_exit_us;
-	double pstate_change_us;
-};
-
-struct _vcs_dpi_wm_calc_pipe_params_st {
-	unsigned int num_dpp;
-	int voltage;
-	int output_type;
-	double dcfclk_mhz;
-	double socclk_mhz;
-	double dppclk_mhz;
-	double pixclk_mhz;
-	unsigned char interlace_en;
-	unsigned char pte_enable;
-	unsigned char dcc_enable;
-	double dcc_rate;
-	double bytes_per_pixel_c;
-	double bytes_per_pixel_y;
-	unsigned int swath_width_y;
-	unsigned int swath_height_y;
-	unsigned int swath_height_c;
-	unsigned int det_buffer_size_y;
-	double h_ratio;
-	double v_ratio;
-	unsigned int h_taps;
-	unsigned int h_total;
-	unsigned int v_total;
-	unsigned int v_active;
-	unsigned int e2e_index;
-	double display_pipe_line_delivery_time;
-	double read_bw;
-	unsigned int lines_in_det_y;
-	unsigned int lines_in_det_y_rounded_down_to_swath;
-	double full_det_buffering_time;
-	double dcfclk_deepsleep_mhz_per_plane;
-};
-
-struct _vcs_dpi_vratio_pre_st {
-	double vratio_pre_l;
-	double vratio_pre_c;
-};
-
 struct _vcs_dpi_display_data_rq_misc_params_st {
 	unsigned int full_swath_bytes;
 	unsigned int stored_swath_bytes;
@@ -423,16 +344,9 @@ struct _vcs_dpi_display_data_rq_dlg_params_st {
 	unsigned int meta_bytes_per_row_ub;
 };
 
-struct _vcs_dpi_display_cur_rq_dlg_params_st {
-	unsigned char enable;
-	unsigned int swath_height;
-	unsigned int req_per_line;
-};
-
 struct _vcs_dpi_display_rq_dlg_params_st {
 	display_data_rq_dlg_params_st rq_l;
 	display_data_rq_dlg_params_st rq_c;
-	display_cur_rq_dlg_params_st rq_cur0;
 };
 
 struct _vcs_dpi_display_rq_sizing_params_st {
@@ -498,6 +412,10 @@ struct _vcs_dpi_display_dlg_regs_st {
 	unsigned int xfc_reg_remote_surface_flip_latency;
 	unsigned int xfc_reg_prefetch_margin;
 	unsigned int dst_y_delta_drq_limit;
+	unsigned int refcyc_per_vm_group_vblank;
+	unsigned int refcyc_per_vm_group_flip;
+	unsigned int refcyc_per_vm_req_vblank;
+	unsigned int refcyc_per_vm_req_flip;
 };
 
 struct _vcs_dpi_display_ttu_regs_st {
@@ -556,19 +474,6 @@ struct _vcs_dpi_display_dlg_sys_params_st {
 	unsigned int total_flip_bytes;
 };
 
-struct _vcs_dpi_display_dlg_prefetch_param_st {
-	double prefetch_bw;
-	unsigned int flip_bytes;
-};
-
-struct _vcs_dpi_display_pipe_clock_st {
-	double dcfclk_mhz;
-	double dispclk_mhz;
-	double socclk_mhz;
-	double dscclk_mhz[6];
-	double dppclk_mhz[6];
-};
-
 struct _vcs_dpi_display_arb_params_st {
 	int max_req_outstanding;
 	int min_req_outstanding;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c
index c2037daa8e66..ad8571f5a142 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c
@@ -459,7 +459,7 @@ static void dml1_rq_dlg_get_row_heights(
 	/* dpte   */
 	/* ------ */
 	log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes);
-	dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs;
+	dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma;
 
 	log2_vmpg_height = 0;
 	log2_vmpg_width = 0;
@@ -776,7 +776,7 @@ static void get_surf_rq_param(
 	/* dpte   */
 	/* ------ */
 	log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes);
-	dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs;
+	dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma;
 
 	log2_vmpg_height = 0;
 	log2_vmpg_width = 0;
@@ -881,7 +881,7 @@ static void get_surf_rq_param(
 	/* the dpte_group_bytes is reduced for the specific case of vertical
 	 * access of a tile surface that has dpte request of 8x1 ptes.
 	 */
-	if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) /*reduced, in this case, will have page fault within a group */
+	if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) /*reduced, in this case, will have page fault within a group */
 		rq_sizing_param->dpte_group_bytes = 512;
 	else
 		/*full size */
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c
index 1d1efd72b291..cf76ea2d9f5a 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c
@@ -101,6 +101,18 @@ enum gpio_mode dal_gpio_get_mode(
 	return gpio->mode;
 }
 
+enum gpio_result dal_gpio_lock_pin(
+	struct gpio *gpio)
+{
+	return dal_gpio_service_lock(gpio->service, gpio->id, gpio->en);
+}
+
+enum gpio_result dal_gpio_unlock_pin(
+	struct gpio *gpio)
+{
+	return dal_gpio_service_unlock(gpio->service, gpio->id, gpio->en);
+}
+
 enum gpio_result dal_gpio_change_mode(
 	struct gpio *gpio,
 	enum gpio_mode mode)
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
index dada04296025..3c63a3c04dbb 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
@@ -192,6 +192,34 @@ static void set_pin_free(
 	service->busyness[id][en] = false;
 }
 
+enum gpio_result dal_gpio_service_lock(
+	struct gpio_service *service,
+	enum gpio_id id,
+	uint32_t en)
+{
+	if (!service->busyness[id]) {
+		ASSERT_CRITICAL(false);
+		return GPIO_RESULT_OPEN_FAILED;
+	}
+
+	set_pin_busy(service, id, en);
+	return GPIO_RESULT_OK;
+}
+
+enum gpio_result dal_gpio_service_unlock(
+	struct gpio_service *service,
+	enum gpio_id id,
+	uint32_t en)
+{
+	if (!service->busyness[id]) {
+		ASSERT_CRITICAL(false);
+		return GPIO_RESULT_OPEN_FAILED;
+	}
+
+	set_pin_free(service, id, en);
+	return GPIO_RESULT_OK;
+}
+
 enum gpio_result dal_gpio_service_open(
 	struct gpio_service *service,
 	enum gpio_id id,
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h
index 1d501a43d13b..0c678af75331 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h
+++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h
@@ -52,4 +52,14 @@ void dal_gpio_service_close(
 	struct gpio_service *service,
 	struct hw_gpio_pin **ptr);
 
+enum gpio_result dal_gpio_service_lock(
+	struct gpio_service *service,
+	enum gpio_id id,
+	uint32_t en);
+
+enum gpio_result dal_gpio_service_unlock(
+	struct gpio_service *service,
+	enum gpio_id id,
+	uint32_t en);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/Makefile b/drivers/gpu/drm/amd/display/dc/i2caux/Makefile
deleted file mode 100644
index 352885cb4d07..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/Makefile
+++ /dev/null
@@ -1,99 +0,0 @@
-#
-# 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.
-#
-#
-# Makefile for the 'i2c' sub-component of DAL.
-# It provides the control and status of HW i2c engine of the adapter.
-
-I2CAUX = aux_engine.o engine_base.o i2caux.o i2c_engine.o \
-	 i2c_generic_hw_engine.o i2c_hw_engine.o i2c_sw_engine.o
-
-AMD_DAL_I2CAUX = $(addprefix $(AMDDALPATH)/dc/i2caux/,$(I2CAUX))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX)
-
-###############################################################################
-# DCE 8x family
-###############################################################################
-I2CAUX_DCE80 = i2caux_dce80.o i2c_hw_engine_dce80.o \
-	i2c_sw_engine_dce80.o
-
-AMD_DAL_I2CAUX_DCE80 = $(addprefix $(AMDDALPATH)/dc/i2caux/dce80/,$(I2CAUX_DCE80))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCE80)
-
-###############################################################################
-# DCE 100 family
-###############################################################################
-I2CAUX_DCE100 = i2caux_dce100.o
-
-AMD_DAL_I2CAUX_DCE100 = $(addprefix $(AMDDALPATH)/dc/i2caux/dce100/,$(I2CAUX_DCE100))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCE100)
-
-###############################################################################
-# DCE 110 family
-###############################################################################
-I2CAUX_DCE110 = i2caux_dce110.o i2c_sw_engine_dce110.o i2c_hw_engine_dce110.o \
-	aux_engine_dce110.o
-
-AMD_DAL_I2CAUX_DCE110 = $(addprefix $(AMDDALPATH)/dc/i2caux/dce110/,$(I2CAUX_DCE110))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCE110)
-
-###############################################################################
-# DCE 112 family
-###############################################################################
-I2CAUX_DCE112 = i2caux_dce112.o
-
-AMD_DAL_I2CAUX_DCE112 = $(addprefix $(AMDDALPATH)/dc/i2caux/dce112/,$(I2CAUX_DCE112))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCE112)
-
-###############################################################################
-# DCN 1.0 family
-###############################################################################
-ifdef CONFIG_DRM_AMD_DC_DCN1_0
-I2CAUX_DCN1 = i2caux_dcn10.o
-
-AMD_DAL_I2CAUX_DCN1 = $(addprefix $(AMDDALPATH)/dc/i2caux/dcn10/,$(I2CAUX_DCN1))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCN1)
-endif
-
-###############################################################################
-# DCE 120 family
-###############################################################################
-I2CAUX_DCE120 = i2caux_dce120.o
-
-AMD_DAL_I2CAUX_DCE120 = $(addprefix $(AMDDALPATH)/dc/i2caux/dce120/,$(I2CAUX_DCE120))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DCE120)
-
-###############################################################################
-# Diagnostics on FPGA
-###############################################################################
-I2CAUX_DIAG = i2caux_diag.o
-
-AMD_DAL_I2CAUX_DIAG = $(addprefix $(AMDDALPATH)/dc/i2caux/diagnostics/,$(I2CAUX_DIAG))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_I2CAUX_DIAG)
-
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.c b/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.c
deleted file mode 100644
index 8cbf38b2470d..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.c
+++ /dev/null
@@ -1,606 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-#include "dm_event_log.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "aux_engine.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-#include "include/link_service_types.h"
-
-/*
- * This unit
- */
-
-enum {
-	AUX_INVALID_REPLY_RETRY_COUNTER = 1,
-	AUX_TIMED_OUT_RETRY_COUNTER = 2,
-	AUX_DEFER_RETRY_COUNTER = 6
-};
-
-#define FROM_ENGINE(ptr) \
-	container_of((ptr), struct aux_engine, base)
-#define DC_LOGGER \
-	engine->base.ctx->logger
-
-enum i2caux_engine_type dal_aux_engine_get_engine_type(
-	const struct engine *engine)
-{
-	return I2CAUX_ENGINE_TYPE_AUX;
-}
-
-bool dal_aux_engine_acquire(
-	struct engine *engine,
-	struct ddc *ddc)
-{
-	struct aux_engine *aux_engine = FROM_ENGINE(engine);
-
-	enum gpio_result result;
-	if (aux_engine->funcs->is_engine_available) {
-		/*check whether SW could use the engine*/
-		if (!aux_engine->funcs->is_engine_available(aux_engine)) {
-			return false;
-		}
-	}
-
-	result = dal_ddc_open(ddc, GPIO_MODE_HARDWARE,
-		GPIO_DDC_CONFIG_TYPE_MODE_AUX);
-
-	if (result != GPIO_RESULT_OK)
-		return false;
-
-	if (!aux_engine->funcs->acquire_engine(aux_engine)) {
-		dal_ddc_close(ddc);
-		return false;
-	}
-
-	engine->ddc = ddc;
-
-	return true;
-}
-
-struct read_command_context {
-	uint8_t *buffer;
-	uint32_t current_read_length;
-	uint32_t offset;
-	enum i2caux_transaction_status status;
-
-	struct aux_request_transaction_data request;
-	struct aux_reply_transaction_data reply;
-
-	uint8_t returned_byte;
-
-	uint32_t timed_out_retry_aux;
-	uint32_t invalid_reply_retry_aux;
-	uint32_t defer_retry_aux;
-	uint32_t defer_retry_i2c;
-	uint32_t invalid_reply_retry_aux_on_ack;
-
-	bool transaction_complete;
-	bool operation_succeeded;
-};
-
-static void process_read_reply(
-	struct aux_engine *engine,
-	struct read_command_context *ctx)
-{
-	engine->funcs->process_channel_reply(engine, &ctx->reply);
-
-	switch (ctx->reply.status) {
-	case AUX_TRANSACTION_REPLY_AUX_ACK:
-		ctx->defer_retry_aux = 0;
-		if (ctx->returned_byte > ctx->current_read_length) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else if (ctx->returned_byte < ctx->current_read_length) {
-			ctx->current_read_length -= ctx->returned_byte;
-
-			ctx->offset += ctx->returned_byte;
-
-			++ctx->invalid_reply_retry_aux_on_ack;
-
-			if (ctx->invalid_reply_retry_aux_on_ack >
-				AUX_INVALID_REPLY_RETRY_COUNTER) {
-				ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-				ctx->operation_succeeded = false;
-			}
-		} else {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-			ctx->transaction_complete = true;
-			ctx->operation_succeeded = true;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_NACK:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-		ctx->operation_succeeded = false;
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_DEFER:
-		++ctx->defer_retry_aux;
-
-		if (ctx->defer_retry_aux > AUX_DEFER_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_I2C_DEFER:
-		ctx->defer_retry_aux = 0;
-
-		++ctx->defer_retry_i2c;
-
-		if (ctx->defer_retry_i2c > AUX_DEFER_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-
-static void process_read_request(
-	struct aux_engine *engine,
-	struct read_command_context *ctx)
-{
-	enum aux_channel_operation_result operation_result;
-
-	engine->funcs->submit_channel_request(engine, &ctx->request);
-
-	operation_result = engine->funcs->get_channel_status(
-		engine, &ctx->returned_byte);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		if (ctx->returned_byte > ctx->current_read_length) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else {
-			ctx->timed_out_retry_aux = 0;
-			ctx->invalid_reply_retry_aux = 0;
-
-			ctx->reply.length = ctx->returned_byte;
-			ctx->reply.data = ctx->buffer;
-
-			process_read_reply(engine, ctx);
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-		++ctx->invalid_reply_retry_aux;
-
-		if (ctx->invalid_reply_retry_aux >
-			AUX_INVALID_REPLY_RETRY_COUNTER) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else
-			udelay(400);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
-		++ctx->timed_out_retry_aux;
-
-		if (ctx->timed_out_retry_aux > AUX_TIMED_OUT_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		} else {
-			/* DP 1.2a, table 2-58:
-			 * "S3: AUX Request CMD PENDING:
-			 * retry 3 times, with 400usec wait on each"
-			 * The HW timeout is set to 550usec,
-			 * so we should not wait here */
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-
-static bool read_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-	struct read_command_context ctx;
-
-	ctx.buffer = request->payload.data;
-	ctx.current_read_length = request->payload.length;
-	ctx.offset = 0;
-	ctx.timed_out_retry_aux = 0;
-	ctx.invalid_reply_retry_aux = 0;
-	ctx.defer_retry_aux = 0;
-	ctx.defer_retry_i2c = 0;
-	ctx.invalid_reply_retry_aux_on_ack = 0;
-	ctx.transaction_complete = false;
-	ctx.operation_succeeded = true;
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_DP;
-		ctx.request.action = I2CAUX_TRANSACTION_ACTION_DP_READ;
-		ctx.request.address = request->payload.address;
-	} else if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_I2C;
-		ctx.request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_READ;
-		ctx.request.address = request->payload.address >> 1;
-	} else {
-		/* in DAL2, there was no return in such case */
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	ctx.request.delay = 0;
-
-	do {
-		memset(ctx.buffer + ctx.offset, 0, ctx.current_read_length);
-
-		ctx.request.data = ctx.buffer + ctx.offset;
-		ctx.request.length = ctx.current_read_length;
-
-		process_read_request(engine, &ctx);
-
-		request->status = ctx.status;
-
-		if (ctx.operation_succeeded && !ctx.transaction_complete)
-			if (ctx.request.type == AUX_TRANSACTION_TYPE_I2C)
-				msleep(engine->delay);
-	} while (ctx.operation_succeeded && !ctx.transaction_complete);
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		DC_LOG_I2C_AUX("READ: addr:0x%x  value:0x%x Result:%d",
-				request->payload.address,
-				request->payload.data[0],
-				ctx.operation_succeeded);
-	}
-
-	return ctx.operation_succeeded;
-}
-
-struct write_command_context {
-	bool mot;
-
-	uint8_t *buffer;
-	uint32_t current_write_length;
-	enum i2caux_transaction_status status;
-
-	struct aux_request_transaction_data request;
-	struct aux_reply_transaction_data reply;
-
-	uint8_t returned_byte;
-
-	uint32_t timed_out_retry_aux;
-	uint32_t invalid_reply_retry_aux;
-	uint32_t defer_retry_aux;
-	uint32_t defer_retry_i2c;
-	uint32_t max_defer_retry;
-	uint32_t ack_m_retry;
-
-	uint8_t reply_data[DEFAULT_AUX_MAX_DATA_SIZE];
-
-	bool transaction_complete;
-	bool operation_succeeded;
-};
-
-static void process_write_reply(
-	struct aux_engine *engine,
-	struct write_command_context *ctx)
-{
-	engine->funcs->process_channel_reply(engine, &ctx->reply);
-
-	switch (ctx->reply.status) {
-	case AUX_TRANSACTION_REPLY_AUX_ACK:
-		ctx->operation_succeeded = true;
-
-		if (ctx->returned_byte) {
-			ctx->request.action = ctx->mot ?
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST;
-
-			ctx->current_write_length = 0;
-
-			++ctx->ack_m_retry;
-
-			if (ctx->ack_m_retry > AUX_DEFER_RETRY_COUNTER) {
-				ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-				ctx->operation_succeeded = false;
-			} else
-				udelay(300);
-		} else {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-			ctx->defer_retry_aux = 0;
-			ctx->ack_m_retry = 0;
-			ctx->transaction_complete = true;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_NACK:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-		ctx->operation_succeeded = false;
-	break;
-	case AUX_TRANSACTION_REPLY_AUX_DEFER:
-		++ctx->defer_retry_aux;
-
-		if (ctx->defer_retry_aux > ctx->max_defer_retry) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_I2C_DEFER:
-		ctx->defer_retry_aux = 0;
-		ctx->current_write_length = 0;
-
-		ctx->request.action = ctx->mot ?
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_STATUS_REQUEST;
-
-		++ctx->defer_retry_i2c;
-
-		if (ctx->defer_retry_i2c > ctx->max_defer_retry) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		}
-	break;
-	case AUX_TRANSACTION_REPLY_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-
-static void process_write_request(
-	struct aux_engine *engine,
-	struct write_command_context *ctx)
-{
-	enum aux_channel_operation_result operation_result;
-
-	engine->funcs->submit_channel_request(engine, &ctx->request);
-
-	operation_result = engine->funcs->get_channel_status(
-		engine, &ctx->returned_byte);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		ctx->timed_out_retry_aux = 0;
-		ctx->invalid_reply_retry_aux = 0;
-
-		ctx->reply.length = ctx->returned_byte;
-		ctx->reply.data = ctx->reply_data;
-
-		process_write_reply(engine, ctx);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-		++ctx->invalid_reply_retry_aux;
-
-		if (ctx->invalid_reply_retry_aux >
-			AUX_INVALID_REPLY_RETRY_COUNTER) {
-			ctx->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR;
-			ctx->operation_succeeded = false;
-		} else
-			udelay(400);
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
-		++ctx->timed_out_retry_aux;
-
-		if (ctx->timed_out_retry_aux > AUX_TIMED_OUT_RETRY_COUNTER) {
-			ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			ctx->operation_succeeded = false;
-		} else {
-			/* DP 1.2a, table 2-58:
-			 * "S3: AUX Request CMD PENDING:
-			 * retry 3 times, with 400usec wait on each"
-			 * The HW timeout is set to 550usec,
-			 * so we should not wait here */
-		}
-	break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON;
-		ctx->operation_succeeded = false;
-	break;
-	default:
-		ctx->status = I2CAUX_TRANSACTION_STATUS_UNKNOWN;
-		ctx->operation_succeeded = false;
-	}
-}
-
-static bool write_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-	struct write_command_context ctx;
-
-	ctx.mot = middle_of_transaction;
-	ctx.buffer = request->payload.data;
-	ctx.current_write_length = request->payload.length;
-	ctx.timed_out_retry_aux = 0;
-	ctx.invalid_reply_retry_aux = 0;
-	ctx.defer_retry_aux = 0;
-	ctx.defer_retry_i2c = 0;
-	ctx.ack_m_retry = 0;
-	ctx.transaction_complete = false;
-	ctx.operation_succeeded = true;
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_DP;
-		ctx.request.action = I2CAUX_TRANSACTION_ACTION_DP_WRITE;
-		ctx.request.address = request->payload.address;
-	} else if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C) {
-		ctx.request.type = AUX_TRANSACTION_TYPE_I2C;
-		ctx.request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-		ctx.request.address = request->payload.address >> 1;
-	} else {
-		/* in DAL2, there was no return in such case */
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	ctx.request.delay = 0;
-
-	ctx.max_defer_retry =
-		(engine->max_defer_write_retry > AUX_DEFER_RETRY_COUNTER) ?
-			engine->max_defer_write_retry : AUX_DEFER_RETRY_COUNTER;
-
-	do {
-		ctx.request.data = ctx.buffer;
-		ctx.request.length = ctx.current_write_length;
-
-		process_write_request(engine, &ctx);
-
-		request->status = ctx.status;
-
-		if (ctx.operation_succeeded && !ctx.transaction_complete)
-			if (ctx.request.type == AUX_TRANSACTION_TYPE_I2C)
-				msleep(engine->delay);
-	} while (ctx.operation_succeeded && !ctx.transaction_complete);
-
-	if (request->payload.address_space ==
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD) {
-		DC_LOG_I2C_AUX("WRITE: addr:0x%x  value:0x%x Result:%d",
-				request->payload.address,
-				request->payload.data[0],
-				ctx.operation_succeeded);
-	}
-
-	return ctx.operation_succeeded;
-}
-
-static bool end_of_transaction_command(
-	struct aux_engine *engine,
-	struct i2caux_transaction_request *request)
-{
-	struct i2caux_transaction_request dummy_request;
-	uint8_t dummy_data;
-
-	/* [tcheng] We only need to send the stop (read with MOT = 0)
-	 * for I2C-over-Aux, not native AUX */
-
-	if (request->payload.address_space !=
-		I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C)
-		return false;
-
-	dummy_request.operation = request->operation;
-	dummy_request.payload.address_space = request->payload.address_space;
-	dummy_request.payload.address = request->payload.address;
-
-	/*
-	 * Add a dummy byte due to some receiver quirk
-	 * where one byte is sent along with MOT = 0.
-	 * Ideally this should be 0.
-	 */
-
-	dummy_request.payload.length = 0;
-	dummy_request.payload.data = &dummy_data;
-
-	if (request->operation == I2CAUX_TRANSACTION_READ)
-		return read_command(engine, &dummy_request, false);
-	else
-		return write_command(engine, &dummy_request, false);
-
-	/* according Syed, it does not need now DoDummyMOT */
-}
-
-bool dal_aux_engine_submit_request(
-	struct engine *engine,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction)
-{
-	struct aux_engine *aux_engine = FROM_ENGINE(engine);
-
-	bool result;
-	bool mot_used = true;
-
-	switch (request->operation) {
-	case I2CAUX_TRANSACTION_READ:
-		result = read_command(aux_engine, request, mot_used);
-	break;
-	case I2CAUX_TRANSACTION_WRITE:
-		result = write_command(aux_engine, request, mot_used);
-	break;
-	default:
-		result = false;
-	}
-
-	/* [tcheng]
-	 * need to send stop for the last transaction to free up the AUX
-	 * if the above command fails, this would be the last transaction */
-
-	if (!middle_of_transaction || !result)
-		end_of_transaction_command(aux_engine, request);
-
-	/* mask AUX interrupt */
-
-	return result;
-}
-
-void dal_aux_engine_construct(
-	struct aux_engine *engine,
-	struct dc_context *ctx)
-{
-	dal_i2caux_construct_engine(&engine->base, ctx);
-	engine->delay = 0;
-	engine->max_defer_write_retry = 0;
-}
-
-void dal_aux_engine_destruct(
-	struct aux_engine *engine)
-{
-	dal_i2caux_destruct_engine(&engine->base);
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.h
deleted file mode 100644
index c33a2898d967..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/aux_engine.h
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_AUX_ENGINE_H__
-#define __DAL_AUX_ENGINE_H__
-
-#include "dc_ddc_types.h"
-
-struct aux_engine;
-
-struct aux_engine_funcs {
-	void (*destroy)(
-		struct aux_engine **ptr);
-	bool (*acquire_engine)(
-		struct aux_engine *engine);
-	void (*configure)(
-		struct aux_engine *engine,
-		union aux_config cfg);
-	void (*submit_channel_request)(
-		struct aux_engine *engine,
-		struct aux_request_transaction_data *request);
-	void (*process_channel_reply)(
-		struct aux_engine *engine,
-		struct aux_reply_transaction_data *reply);
-	int (*read_channel_reply)(
-		struct aux_engine *engine,
-		uint32_t size,
-		uint8_t *buffer,
-		uint8_t *reply_result,
-		uint32_t *sw_status);
-	enum aux_channel_operation_result (*get_channel_status)(
-		struct aux_engine *engine,
-		uint8_t *returned_bytes);
-	bool (*is_engine_available) (
-		struct aux_engine *engine);
-};
-
-struct aux_engine {
-	struct engine base;
-	const struct aux_engine_funcs *funcs;
-	/* following values are expressed in milliseconds */
-	uint32_t delay;
-	uint32_t max_defer_write_retry;
-
-	bool acquire_reset;
-};
-
-void dal_aux_engine_construct(
-	struct aux_engine *engine,
-	struct dc_context *ctx);
-
-void dal_aux_engine_destruct(
-	struct aux_engine *engine);
-bool dal_aux_engine_submit_request(
-	struct engine *ptr,
-	struct i2caux_transaction_request *request,
-	bool middle_of_transaction);
-bool dal_aux_engine_acquire(
-	struct engine *ptr,
-	struct ddc *ddc);
-enum i2caux_engine_type dal_aux_engine_get_engine_type(
-	const struct engine *engine);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce100/i2caux_dce100.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce100/i2caux_dce100.c
deleted file mode 100644
index 8b704ab0471c..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce100/i2caux_dce100.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-#include "../dce110/aux_engine_dce110.h"
-#include "../dce110/i2c_hw_engine_dce110.h"
-#include "../dce110/i2caux_dce110.h"
-
-#include "dce/dce_10_0_d.h"
-#include "dce/dce_10_0_sh_mask.h"
-
-/* 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 aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = 0 \
-}
-
-#define hw_engine_regs(id)\
-{\
-		I2C_HW_ENGINE_COMMON_REG_LIST(id) \
-}
-
-static const struct dce110_aux_registers dce100_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5),
-};
-
-static const struct dce110_i2c_hw_engine_registers dce100_hw_engine_regs[] = {
-		hw_engine_regs(1),
-		hw_engine_regs(2),
-		hw_engine_regs(3),
-		hw_engine_regs(4),
-		hw_engine_regs(5),
-		hw_engine_regs(6)
-};
-
-static const struct dce110_i2c_hw_engine_shift i2c_shift = {
-		I2C_COMMON_MASK_SH_LIST_DCE100(__SHIFT)
-};
-
-static const struct dce110_i2c_hw_engine_mask i2c_mask = {
-		I2C_COMMON_MASK_SH_LIST_DCE100(_MASK)
-};
-
-struct i2caux *dal_i2caux_dce100_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce110 *i2caux_dce110 =
-		kzalloc(sizeof(struct i2caux_dce110), GFP_KERNEL);
-
-	if (!i2caux_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	dal_i2caux_dce110_construct(i2caux_dce110,
-				    ctx,
-				    ARRAY_SIZE(dce100_aux_regs),
-				    dce100_aux_regs,
-				    dce100_hw_engine_regs,
-				    &i2c_shift,
-				    &i2c_mask);
-	return &i2caux_dce110->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.c
deleted file mode 100644
index 59c3ed43d609..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.c
+++ /dev/null
@@ -1,505 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-#include "dm_event_log.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../engine.h"
-#include "../aux_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "aux_engine_dce110.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-#include "dce/dce_11_0_sh_mask.h"
-
-#define CTX \
-	aux110->base.base.ctx
-#define REG(reg_name)\
-	(aux110->regs->reg_name)
-#include "reg_helper.h"
-
-/*
- * This unit
- */
-
-/*
- * @brief
- * Cast 'struct aux_engine *'
- * to 'struct aux_engine_dce110 *'
- */
-#define FROM_AUX_ENGINE(ptr) \
-	container_of((ptr), struct aux_engine_dce110, base)
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct aux_engine_dce110 *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_AUX_ENGINE(container_of((ptr), struct aux_engine, base))
-
-static void release_engine(
-	struct engine *engine)
-{
-	struct aux_engine_dce110 *aux110 = FROM_ENGINE(engine);
-
-	REG_UPDATE(AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, 1);
-}
-
-static void destruct(
-	struct aux_engine_dce110 *engine);
-
-static void destroy(
-	struct aux_engine **aux_engine)
-{
-	struct aux_engine_dce110 *engine = FROM_AUX_ENGINE(*aux_engine);
-
-	destruct(engine);
-
-	kfree(engine);
-
-	*aux_engine = NULL;
-}
-
-#define SW_CAN_ACCESS_AUX 1
-#define DMCU_CAN_ACCESS_AUX 2
-
-static bool is_engine_available(
-	struct aux_engine *engine)
-{
-	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
-
-	uint32_t value = REG_READ(AUX_ARB_CONTROL);
-	uint32_t field = get_reg_field_value(
-			value,
-			AUX_ARB_CONTROL,
-			AUX_REG_RW_CNTL_STATUS);
-
-	return (field != DMCU_CAN_ACCESS_AUX);
-}
-static bool acquire_engine(
-	struct aux_engine *engine)
-{
-	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
-
-	uint32_t value = REG_READ(AUX_ARB_CONTROL);
-	uint32_t field = get_reg_field_value(
-			value,
-			AUX_ARB_CONTROL,
-			AUX_REG_RW_CNTL_STATUS);
-	if (field == DMCU_CAN_ACCESS_AUX)
-	 return false;
-	/* enable AUX before request SW to access AUX */
-	value = REG_READ(AUX_CONTROL);
-	field = get_reg_field_value(value,
-				AUX_CONTROL,
-				AUX_EN);
-
-	if (field == 0) {
-		set_reg_field_value(
-				value,
-				1,
-				AUX_CONTROL,
-				AUX_EN);
-
-		if (REG(AUX_RESET_MASK)) {
-			/*DP_AUX block as part of the enable sequence*/
-			set_reg_field_value(
-				value,
-				1,
-				AUX_CONTROL,
-				AUX_RESET);
-		}
-
-		REG_WRITE(AUX_CONTROL, value);
-
-		if (REG(AUX_RESET_MASK)) {
-			/*poll HW to make sure reset it done*/
-
-			REG_WAIT(AUX_CONTROL, AUX_RESET_DONE, 1,
-					1, 11);
-
-			set_reg_field_value(
-				value,
-				0,
-				AUX_CONTROL,
-				AUX_RESET);
-
-			REG_WRITE(AUX_CONTROL, value);
-
-			REG_WAIT(AUX_CONTROL, AUX_RESET_DONE, 0,
-					1, 11);
-		}
-	} /*if (field)*/
-
-	/* request SW to access AUX */
-	REG_UPDATE(AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, 1);
-
-	value = REG_READ(AUX_ARB_CONTROL);
-	field = get_reg_field_value(
-			value,
-			AUX_ARB_CONTROL,
-			AUX_REG_RW_CNTL_STATUS);
-
-	return (field == SW_CAN_ACCESS_AUX);
-}
-
-#define COMPOSE_AUX_SW_DATA_16_20(command, address) \
-	((command) | ((0xF0000 & (address)) >> 16))
-
-#define COMPOSE_AUX_SW_DATA_8_15(address) \
-	((0xFF00 & (address)) >> 8)
-
-#define COMPOSE_AUX_SW_DATA_0_7(address) \
-	(0xFF & (address))
-
-static void submit_channel_request(
-	struct aux_engine *engine,
-	struct aux_request_transaction_data *request)
-{
-	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
-	uint32_t value;
-	uint32_t length;
-
-	bool is_write =
-		((request->type == AUX_TRANSACTION_TYPE_DP) &&
-		 (request->action == I2CAUX_TRANSACTION_ACTION_DP_WRITE)) ||
-		((request->type == AUX_TRANSACTION_TYPE_I2C) &&
-		((request->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE) ||
-		 (request->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT)));
-	if (REG(AUXN_IMPCAL)) {
-		/* clear_aux_error */
-		REG_UPDATE_SEQ(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK,
-				1,
-				0);
-
-		REG_UPDATE_SEQ(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK,
-				1,
-				0);
-
-		/* force_default_calibrate */
-		REG_UPDATE_1BY1_2(AUXN_IMPCAL,
-				AUXN_IMPCAL_ENABLE, 1,
-				AUXN_IMPCAL_OVERRIDE_ENABLE, 0);
-
-		/* bug? why AUXN update EN and OVERRIDE_EN 1 by 1 while AUX P toggles OVERRIDE? */
-
-		REG_UPDATE_SEQ(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE,
-				1,
-				0);
-	}
-	/* set the delay and the number of bytes to write */
-
-	/* The length include
-	 * the 4 bit header and the 20 bit address
-	 * (that is 3 byte).
-	 * If the requested length is non zero this means
-	 * an addition byte specifying the length is required. */
-
-	length = request->length ? 4 : 3;
-	if (is_write)
-		length += request->length;
-
-	REG_UPDATE_2(AUX_SW_CONTROL,
-			AUX_SW_START_DELAY, request->delay,
-			AUX_SW_WR_BYTES, length);
-
-	/* program action and address and payload data (if 'is_write') */
-	value = REG_UPDATE_4(AUX_SW_DATA,
-			AUX_SW_INDEX, 0,
-			AUX_SW_DATA_RW, 0,
-			AUX_SW_AUTOINCREMENT_DISABLE, 1,
-			AUX_SW_DATA, COMPOSE_AUX_SW_DATA_16_20(request->action, request->address));
-
-	value = REG_SET_2(AUX_SW_DATA, value,
-			AUX_SW_AUTOINCREMENT_DISABLE, 0,
-			AUX_SW_DATA, COMPOSE_AUX_SW_DATA_8_15(request->address));
-
-	value = REG_SET(AUX_SW_DATA, value,
-			AUX_SW_DATA, COMPOSE_AUX_SW_DATA_0_7(request->address));
-
-	if (request->length) {
-		value = REG_SET(AUX_SW_DATA, value,
-				AUX_SW_DATA, request->length - 1);
-	}
-
-	if (is_write) {
-		/* Load the HW buffer with the Data to be sent.
-		 * This is relevant for write operation.
-		 * For read, the data recived data will be
-		 * processed in process_channel_reply(). */
-		uint32_t i = 0;
-
-		while (i < request->length) {
-			value = REG_SET(AUX_SW_DATA, value,
-					AUX_SW_DATA, request->data[i]);
-
-			++i;
-		}
-	}
-
-	REG_UPDATE(AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, 1);
-	REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 0,
-				10, aux110->timeout_period/10);
-	REG_UPDATE(AUX_SW_CONTROL, AUX_SW_GO, 1);
-	EVENT_LOG_AUX_REQ(engine->base.ddc->pin_data->en, EVENT_LOG_AUX_ORIGIN_NATIVE,
-					request->action, request->address, request->length, request->data);
-}
-
-static int read_channel_reply(struct aux_engine *engine, uint32_t size,
-			      uint8_t *buffer, uint8_t *reply_result,
-			      uint32_t *sw_status)
-{
-	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
-	uint32_t bytes_replied;
-	uint32_t reply_result_32;
-
-	*sw_status = REG_GET(AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT,
-			     &bytes_replied);
-
-	/* In case HPD is LOW, exit AUX transaction */
-	if ((*sw_status & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK))
-		return -1;
-
-	/* Need at least the status byte */
-	if (!bytes_replied)
-		return -1;
-
-	REG_UPDATE_1BY1_3(AUX_SW_DATA,
-			  AUX_SW_INDEX, 0,
-			  AUX_SW_AUTOINCREMENT_DISABLE, 1,
-			  AUX_SW_DATA_RW, 1);
-
-	REG_GET(AUX_SW_DATA, AUX_SW_DATA, &reply_result_32);
-	reply_result_32 = reply_result_32 >> 4;
-	*reply_result = (uint8_t)reply_result_32;
-
-	if (reply_result_32 == 0) { /* ACK */
-		uint32_t i = 0;
-
-		/* First byte was already used to get the command status */
-		--bytes_replied;
-
-		/* Do not overflow buffer */
-		if (bytes_replied > size)
-			return -1;
-
-		while (i < bytes_replied) {
-			uint32_t aux_sw_data_val;
-
-			REG_GET(AUX_SW_DATA, AUX_SW_DATA, &aux_sw_data_val);
-			buffer[i] = aux_sw_data_val;
-			++i;
-		}
-
-		return i;
-	}
-
-	return 0;
-}
-
-static void process_channel_reply(
-	struct aux_engine *engine,
-	struct aux_reply_transaction_data *reply)
-{
-	int bytes_replied;
-	uint8_t reply_result;
-	uint32_t sw_status;
-
-	bytes_replied = read_channel_reply(engine, reply->length, reply->data,
-						&reply_result, &sw_status);
-	EVENT_LOG_AUX_REP(engine->base.ddc->pin_data->en,
-					EVENT_LOG_AUX_ORIGIN_NATIVE, reply_result,
-					bytes_replied, reply->data);
-
-	/* in case HPD is LOW, exit AUX transaction */
-	if ((sw_status & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK)) {
-		reply->status = AUX_TRANSACTION_REPLY_HPD_DISCON;
-		return;
-	}
-
-	if (bytes_replied < 0) {
-		/* Need to handle an error case...
-		 * Hopefully, upper layer function won't call this function if
-		 * the number of bytes in the reply was 0, because there was
-		 * surely an error that was asserted that should have been
-		 * handled for hot plug case, this could happens
-		 */
-		if (!(sw_status & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK)) {
-			reply->status = AUX_TRANSACTION_REPLY_INVALID;
-			ASSERT_CRITICAL(false);
-			return;
-		}
-	} else {
-
-		switch (reply_result) {
-		case 0: /* ACK */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_ACK;
-		break;
-		case 1: /* NACK */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_NACK;
-		break;
-		case 2: /* DEFER */
-			reply->status = AUX_TRANSACTION_REPLY_AUX_DEFER;
-		break;
-		case 4: /* AUX ACK / I2C NACK */
-			reply->status = AUX_TRANSACTION_REPLY_I2C_NACK;
-		break;
-		case 8: /* AUX ACK / I2C DEFER */
-			reply->status = AUX_TRANSACTION_REPLY_I2C_DEFER;
-		break;
-		default:
-			reply->status = AUX_TRANSACTION_REPLY_INVALID;
-		}
-	}
-}
-
-static enum aux_channel_operation_result get_channel_status(
-	struct aux_engine *engine,
-	uint8_t *returned_bytes)
-{
-	struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(engine);
-
-	uint32_t value;
-
-	if (returned_bytes == NULL) {
-		/*caller pass NULL pointer*/
-		ASSERT_CRITICAL(false);
-		return AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN;
-	}
-	*returned_bytes = 0;
-
-	/* poll to make sure that SW_DONE is asserted */
-	value = REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 1,
-				10, aux110->timeout_period/10);
-
-	/* in case HPD is LOW, exit AUX transaction */
-	if ((value & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK))
-		return AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON;
-
-	/* Note that the following bits are set in 'status.bits'
-	 * during CTS 4.2.1.2 (FW 3.3.1):
-	 * AUX_SW_RX_MIN_COUNT_VIOL, AUX_SW_RX_INVALID_STOP,
-	 * AUX_SW_RX_RECV_NO_DET, AUX_SW_RX_RECV_INVALID_H.
-	 *
-	 * AUX_SW_RX_MIN_COUNT_VIOL is an internal,
-	 * HW debugging bit and should be ignored. */
-	if (value & AUX_SW_STATUS__AUX_SW_DONE_MASK) {
-		if ((value & AUX_SW_STATUS__AUX_SW_RX_TIMEOUT_STATE_MASK) ||
-			(value & AUX_SW_STATUS__AUX_SW_RX_TIMEOUT_MASK))
-			return AUX_CHANNEL_OPERATION_FAILED_TIMEOUT;
-
-		else if ((value & AUX_SW_STATUS__AUX_SW_RX_INVALID_STOP_MASK) ||
-			(value & AUX_SW_STATUS__AUX_SW_RX_RECV_NO_DET_MASK) ||
-			(value &
-				AUX_SW_STATUS__AUX_SW_RX_RECV_INVALID_H_MASK) ||
-			(value & AUX_SW_STATUS__AUX_SW_RX_RECV_INVALID_L_MASK))
-			return AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY;
-
-		*returned_bytes = get_reg_field_value(value,
-				AUX_SW_STATUS,
-				AUX_SW_REPLY_BYTE_COUNT);
-
-		if (*returned_bytes == 0)
-			return
-			AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY;
-		else {
-			*returned_bytes -= 1;
-			return AUX_CHANNEL_OPERATION_SUCCEEDED;
-		}
-	} else {
-		/*time_elapsed >= aux_engine->timeout_period
-		 *  AUX_SW_STATUS__AUX_SW_HPD_DISCON = at this point
-		 */
-		ASSERT_CRITICAL(false);
-		return AUX_CHANNEL_OPERATION_FAILED_TIMEOUT;
-	}
-}
-
-static const struct aux_engine_funcs aux_engine_funcs = {
-	.destroy = destroy,
-	.acquire_engine = acquire_engine,
-	.submit_channel_request = submit_channel_request,
-	.process_channel_reply = process_channel_reply,
-	.read_channel_reply = read_channel_reply,
-	.get_channel_status = get_channel_status,
-	.is_engine_available = is_engine_available,
-};
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.submit_request = dal_aux_engine_submit_request,
-	.get_engine_type = dal_aux_engine_get_engine_type,
-	.acquire = dal_aux_engine_acquire,
-};
-
-static void construct(
-	struct aux_engine_dce110 *engine,
-	const struct aux_engine_dce110_init_data *aux_init_data)
-{
-	dal_aux_engine_construct(&engine->base, aux_init_data->ctx);
-	engine->base.base.funcs = &engine_funcs;
-	engine->base.funcs = &aux_engine_funcs;
-
-	engine->timeout_period = aux_init_data->timeout_period;
-	engine->regs = aux_init_data->regs;
-}
-
-static void destruct(
-	struct aux_engine_dce110 *engine)
-{
-	dal_aux_engine_destruct(&engine->base);
-}
-
-struct aux_engine *dal_aux_engine_dce110_create(
-	const struct aux_engine_dce110_init_data *aux_init_data)
-{
-	struct aux_engine_dce110 *engine;
-
-	if (!aux_init_data) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	engine = kzalloc(sizeof(*engine), GFP_KERNEL);
-
-	if (!engine) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	construct(engine, aux_init_data);
-	return &engine->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.h b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.h
deleted file mode 100644
index 85ee82162590..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/aux_engine_dce110.h
+++ /dev/null
@@ -1,78 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_AUX_ENGINE_DCE110_H__
-#define __DAL_AUX_ENGINE_DCE110_H__
-
-#include "../aux_engine.h"
-
-#define AUX_COMMON_REG_LIST(id)\
-	SRI(AUX_CONTROL, DP_AUX, id), \
-	SRI(AUX_ARB_CONTROL, DP_AUX, id), \
-	SRI(AUX_SW_DATA, DP_AUX, id), \
-	SRI(AUX_SW_CONTROL, DP_AUX, id), \
-	SRI(AUX_INTERRUPT_CONTROL, DP_AUX, id), \
-	SRI(AUX_SW_STATUS, DP_AUX, id), \
-	SR(AUXN_IMPCAL), \
-	SR(AUXP_IMPCAL)
-
-struct dce110_aux_registers {
-	uint32_t AUX_CONTROL;
-	uint32_t AUX_ARB_CONTROL;
-	uint32_t AUX_SW_DATA;
-	uint32_t AUX_SW_CONTROL;
-	uint32_t AUX_INTERRUPT_CONTROL;
-	uint32_t AUX_SW_STATUS;
-	uint32_t AUXN_IMPCAL;
-	uint32_t AUXP_IMPCAL;
-
-	uint32_t AUX_RESET_MASK;
-};
-
-struct aux_engine_dce110 {
-	struct aux_engine base;
-	const struct dce110_aux_registers *regs;
-	struct {
-		uint32_t aux_control;
-		uint32_t aux_arb_control;
-		uint32_t aux_sw_data;
-		uint32_t aux_sw_control;
-		uint32_t aux_interrupt_control;
-		uint32_t aux_sw_status;
-	} addr;
-	uint32_t timeout_period;
-};
-
-struct aux_engine_dce110_init_data {
-	uint32_t engine_id;
-	uint32_t timeout_period;
-	struct dc_context *ctx;
-	const struct dce110_aux_registers *regs;
-};
-
-struct aux_engine *dal_aux_engine_dce110_create(
-	const struct aux_engine_dce110_init_data *aux_init_data);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.c
deleted file mode 100644
index 9cbe1a7a6bcb..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.c
+++ /dev/null
@@ -1,574 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-#include "include/logger_interface.h"
-/*
- * Pre-requisites: headers required by header of this unit
- */
-
-#include "include/i2caux_interface.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_hw_engine.h"
-#include "../i2c_generic_hw_engine.h"
-/*
- * Header of this unit
- */
-
-#include "i2c_hw_engine_dce110.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-#include "reg_helper.h"
-
-/*
- * This unit
- */
-#define DC_LOGGER \
-		hw_engine->base.base.base.ctx->logger
-
-enum dc_i2c_status {
-	DC_I2C_STATUS__DC_I2C_STATUS_IDLE,
-	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_SW,
-	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_HW
-};
-
-enum dc_i2c_arbitration {
-	DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_NORMAL,
-	DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_HIGH
-};
-
-
-
-/*
- * @brief
- * Cast pointer to 'struct i2c_hw_engine *'
- * to pointer 'struct i2c_hw_engine_dce110 *'
- */
-#define FROM_I2C_HW_ENGINE(ptr) \
-	container_of((ptr), struct i2c_hw_engine_dce110, base)
-/*
- * @brief
- * Cast pointer to 'struct i2c_engine *'
- * to pointer to 'struct i2c_hw_engine_dce110 *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	FROM_I2C_HW_ENGINE(container_of((ptr), struct i2c_hw_engine, base))
-
-/*
- * @brief
- * Cast pointer to 'struct engine *'
- * to 'pointer to struct i2c_hw_engine_dce110 *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-#define CTX \
-		hw_engine->base.base.base.ctx
-
-#define REG(reg_name)\
-	(hw_engine->regs->reg_name)
-
-#undef FN
-#define FN(reg_name, field_name) \
-	hw_engine->i2c_shift->field_name, hw_engine->i2c_mask->field_name
-
-#include "reg_helper.h"
-
-static void disable_i2c_hw_engine(
-	struct i2c_hw_engine_dce110 *hw_engine)
-{
-	REG_UPDATE_N(SETUP, 1, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE), 0);
-}
-
-static void release_engine(
-	struct engine *engine)
-{
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_ENGINE(engine);
-
-	struct i2c_engine *base = NULL;
-	bool safe_to_reset;
-
-	base = &hw_engine->base.base;
-
-	/* Restore original HW engine speed */
-
-	base->funcs->set_speed(base, hw_engine->base.original_speed);
-
-	/* Release I2C */
-	REG_UPDATE(DC_I2C_ARBITRATION, DC_I2C_SW_DONE_USING_I2C_REG, 1);
-
-	/* Reset HW engine */
-	{
-		uint32_t i2c_sw_status = 0;
-		REG_GET(DC_I2C_SW_STATUS, DC_I2C_SW_STATUS, &i2c_sw_status);
-		/* if used by SW, safe to reset */
-		safe_to_reset = (i2c_sw_status == 1);
-	}
-
-	if (safe_to_reset)
-		REG_UPDATE_2(
-			DC_I2C_CONTROL,
-			DC_I2C_SOFT_RESET, 1,
-			DC_I2C_SW_STATUS_RESET, 1);
-	else
-		REG_UPDATE(DC_I2C_CONTROL, DC_I2C_SW_STATUS_RESET, 1);
-
-	/* HW I2c engine - clock gating feature */
-	if (!hw_engine->engine_keep_power_up_count)
-		disable_i2c_hw_engine(hw_engine);
-}
-
-static bool setup_engine(
-	struct i2c_engine *i2c_engine)
-{
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_I2C_ENGINE(i2c_engine);
-	uint32_t i2c_setup_limit = I2C_SETUP_TIME_LIMIT_DCE;
-	uint32_t  reset_length = 0;
-
-	if (hw_engine->base.base.setup_limit != 0)
-		i2c_setup_limit = hw_engine->base.base.setup_limit;
-
-	/* Program pin select */
-	REG_UPDATE_6(
-			DC_I2C_CONTROL,
-			DC_I2C_GO, 0,
-			DC_I2C_SOFT_RESET, 0,
-			DC_I2C_SEND_RESET, 0,
-			DC_I2C_SW_STATUS_RESET, 1,
-			DC_I2C_TRANSACTION_COUNT, 0,
-			DC_I2C_DDC_SELECT, hw_engine->engine_id);
-
-	/* Program time limit */
-	if (hw_engine->base.base.send_reset_length == 0) {
-		/*pre-dcn*/
-		REG_UPDATE_N(
-				SETUP, 2,
-				FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_TIME_LIMIT), i2c_setup_limit,
-				FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE), 1);
-	} else {
-		reset_length = hw_engine->base.base.send_reset_length;
-	}
-	/* Program HW priority
-	 * set to High - interrupt software I2C at any time
-	 * Enable restart of SW I2C that was interrupted by HW
-	 * disable queuing of software while I2C is in use by HW */
-	REG_UPDATE_2(
-			DC_I2C_ARBITRATION,
-			DC_I2C_NO_QUEUED_SW_GO, 0,
-			DC_I2C_SW_PRIORITY, DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_NORMAL);
-
-	return true;
-}
-
-static uint32_t get_speed(
-	const struct i2c_engine *i2c_engine)
-{
-	const struct i2c_hw_engine_dce110 *hw_engine = FROM_I2C_ENGINE(i2c_engine);
-	uint32_t pre_scale = 0;
-
-	REG_GET(SPEED, DC_I2C_DDC1_PRESCALE, &pre_scale);
-
-	/* [anaumov] it seems following is unnecessary */
-	/*ASSERT(value.bits.DC_I2C_DDC1_PRESCALE);*/
-	return pre_scale ?
-		hw_engine->reference_frequency / pre_scale :
-		hw_engine->base.default_speed;
-}
-
-static void set_speed(
-	struct i2c_engine *i2c_engine,
-	uint32_t speed)
-{
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_I2C_ENGINE(i2c_engine);
-
-	if (speed) {
-		if (hw_engine->i2c_mask->DC_I2C_DDC1_START_STOP_TIMING_CNTL)
-			REG_UPDATE_N(
-				SPEED, 3,
-				FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), hw_engine->reference_frequency / speed,
-				FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2,
-				FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1);
-		else
-			REG_UPDATE_N(
-				SPEED, 2,
-				FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), hw_engine->reference_frequency / speed,
-				FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2);
-	}
-}
-
-static inline void reset_hw_engine(struct engine *engine)
-{
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_ENGINE(engine);
-
-	REG_UPDATE_2(
-			DC_I2C_CONTROL,
-			DC_I2C_SW_STATUS_RESET, 1,
-			DC_I2C_SW_STATUS_RESET, 1);
-}
-
-static bool is_hw_busy(struct engine *engine)
-{
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_ENGINE(engine);
-	uint32_t i2c_sw_status = 0;
-
-	REG_GET(DC_I2C_SW_STATUS, DC_I2C_SW_STATUS, &i2c_sw_status);
-	if (i2c_sw_status == DC_I2C_STATUS__DC_I2C_STATUS_IDLE)
-		return false;
-
-	reset_hw_engine(engine);
-
-	REG_GET(DC_I2C_SW_STATUS, DC_I2C_SW_STATUS, &i2c_sw_status);
-	return i2c_sw_status != DC_I2C_STATUS__DC_I2C_STATUS_IDLE;
-}
-
-
-#define STOP_TRANS_PREDICAT \
-		((hw_engine->transaction_count == 3) ||	\
-				(request->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE) ||	\
-				(request->action & I2CAUX_TRANSACTION_ACTION_I2C_READ))
-
-#define SET_I2C_TRANSACTION(id)	\
-		do {	\
-			REG_UPDATE_N(DC_I2C_TRANSACTION##id, 5,	\
-				FN(DC_I2C_TRANSACTION0, DC_I2C_STOP_ON_NACK0), 1,	\
-				FN(DC_I2C_TRANSACTION0, DC_I2C_START0), 1,	\
-				FN(DC_I2C_TRANSACTION0, DC_I2C_STOP0), STOP_TRANS_PREDICAT ? 1:0,	\
-				FN(DC_I2C_TRANSACTION0, DC_I2C_RW0), (0 != (request->action & I2CAUX_TRANSACTION_ACTION_I2C_READ)),	\
-				FN(DC_I2C_TRANSACTION0, DC_I2C_COUNT0), length);	\
-				if (STOP_TRANS_PREDICAT)	\
-					last_transaction = true;	\
-		} while (false)
-
-
-static bool process_transaction(
-	struct i2c_hw_engine_dce110 *hw_engine,
-	struct i2c_request_transaction_data *request)
-{
-	uint32_t length = request->length;
-	uint8_t *buffer = request->data;
-	uint32_t value = 0;
-
-	bool last_transaction = false;
-
-	struct dc_context *ctx = NULL;
-
-	ctx = hw_engine->base.base.base.ctx;
-
-
-
-	switch (hw_engine->transaction_count) {
-	case 0:
-		SET_I2C_TRANSACTION(0);
-		break;
-	case 1:
-		SET_I2C_TRANSACTION(1);
-		break;
-	case 2:
-		SET_I2C_TRANSACTION(2);
-		break;
-	case 3:
-		SET_I2C_TRANSACTION(3);
-		break;
-	default:
-		/* TODO Warning ? */
-		break;
-	}
-
-
-	/* Write the I2C address and I2C data
-	 * into the hardware circular buffer, one byte per entry.
-	 * As an example, the 7-bit I2C slave address for CRT monitor
-	 * for reading DDC/EDID information is 0b1010001.
-	 * For an I2C send operation, the LSB must be programmed to 0;
-	 * for I2C receive operation, the LSB must be programmed to 1. */
-	if (hw_engine->transaction_count == 0) {
-		value = REG_SET_4(DC_I2C_DATA, 0,
-				  DC_I2C_DATA_RW, false,
-				  DC_I2C_DATA, request->address,
-				  DC_I2C_INDEX, 0,
-				  DC_I2C_INDEX_WRITE, 1);
-		hw_engine->buffer_used_write = 0;
-	} else
-		value = REG_SET_2(DC_I2C_DATA, 0,
-				  DC_I2C_DATA_RW, false,
-				  DC_I2C_DATA, request->address);
-
-	hw_engine->buffer_used_write++;
-
-	if (!(request->action & I2CAUX_TRANSACTION_ACTION_I2C_READ)) {
-		while (length) {
-			REG_SET_2(DC_I2C_DATA, value,
-					DC_I2C_INDEX_WRITE, 0,
-					DC_I2C_DATA, *buffer++);
-			hw_engine->buffer_used_write++;
-			--length;
-		}
-	}
-
-	++hw_engine->transaction_count;
-	hw_engine->buffer_used_bytes += length + 1;
-
-	return last_transaction;
-}
-
-static void execute_transaction(
-	struct i2c_hw_engine_dce110 *hw_engine)
-{
-	REG_UPDATE_N(SETUP, 5,
-		FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_DATA_DRIVE_EN), 0,
-		FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_CLK_DRIVE_EN), 0,
-		FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_DATA_DRIVE_SEL), 0,
-		FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_INTRA_TRANSACTION_DELAY), 0,
-		FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_INTRA_BYTE_DELAY), 0);
-
-
-	REG_UPDATE_5(DC_I2C_CONTROL,
-		DC_I2C_SOFT_RESET, 0,
-		DC_I2C_SW_STATUS_RESET, 0,
-		DC_I2C_SEND_RESET, 0,
-		DC_I2C_GO, 0,
-		DC_I2C_TRANSACTION_COUNT, hw_engine->transaction_count - 1);
-
-	/* start I2C transfer */
-	REG_UPDATE(DC_I2C_CONTROL, DC_I2C_GO, 1);
-
-	/* all transactions were executed and HW buffer became empty
-	 * (even though it actually happens when status becomes DONE) */
-	hw_engine->transaction_count = 0;
-	hw_engine->buffer_used_bytes = 0;
-}
-
-static void submit_channel_request(
-	struct i2c_engine *engine,
-	struct i2c_request_transaction_data *request)
-{
-	request->status = I2C_CHANNEL_OPERATION_SUCCEEDED;
-
-	if (!process_transaction(FROM_I2C_ENGINE(engine), request))
-		return;
-
-	if (is_hw_busy(&engine->base)) {
-		request->status = I2C_CHANNEL_OPERATION_ENGINE_BUSY;
-		return;
-	}
-
-	execute_transaction(FROM_I2C_ENGINE(engine));
-}
-
-static void process_channel_reply(
-	struct i2c_engine *engine,
-	struct i2c_reply_transaction_data *reply)
-{
-	uint32_t length = reply->length;
-	uint8_t *buffer = reply->data;
-
-	struct i2c_hw_engine_dce110 *hw_engine =
-		FROM_I2C_ENGINE(engine);
-
-
-	REG_SET_3(DC_I2C_DATA, 0,
-			DC_I2C_INDEX, hw_engine->buffer_used_write,
-			DC_I2C_DATA_RW, 1,
-			DC_I2C_INDEX_WRITE, 1);
-
-	while (length) {
-		/* after reading the status,
-		 * if the I2C operation executed successfully
-		 * (i.e. DC_I2C_STATUS_DONE = 1) then the I2C controller
-		 * should read data bytes from I2C circular data buffer */
-
-		uint32_t i2c_data;
-
-		REG_GET(DC_I2C_DATA, DC_I2C_DATA, &i2c_data);
-		*buffer++ = i2c_data;
-
-		--length;
-	}
-}
-
-static enum i2c_channel_operation_result get_channel_status(
-	struct i2c_engine *i2c_engine,
-	uint8_t *returned_bytes)
-{
-	uint32_t i2c_sw_status = 0;
-	struct i2c_hw_engine_dce110 *hw_engine = FROM_I2C_ENGINE(i2c_engine);
-	uint32_t value =
-			REG_GET(DC_I2C_SW_STATUS, DC_I2C_SW_STATUS, &i2c_sw_status);
-
-	if (i2c_sw_status == DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_SW)
-		return I2C_CHANNEL_OPERATION_ENGINE_BUSY;
-	else if (value & hw_engine->i2c_mask->DC_I2C_SW_STOPPED_ON_NACK)
-		return I2C_CHANNEL_OPERATION_NO_RESPONSE;
-	else if (value & hw_engine->i2c_mask->DC_I2C_SW_TIMEOUT)
-		return I2C_CHANNEL_OPERATION_TIMEOUT;
-	else if (value & hw_engine->i2c_mask->DC_I2C_SW_ABORTED)
-		return I2C_CHANNEL_OPERATION_FAILED;
-	else if (value & hw_engine->i2c_mask->DC_I2C_SW_DONE)
-		return I2C_CHANNEL_OPERATION_SUCCEEDED;
-
-	/*
-	 * this is the case when HW used for communication, I2C_SW_STATUS
-	 * could be zero
-	 */
-	return I2C_CHANNEL_OPERATION_SUCCEEDED;
-}
-
-static uint32_t get_hw_buffer_available_size(
-	const struct i2c_hw_engine *engine)
-{
-	return I2C_HW_BUFFER_SIZE -
-		FROM_I2C_HW_ENGINE(engine)->buffer_used_bytes;
-}
-
-static uint32_t get_transaction_timeout(
-	const struct i2c_hw_engine *engine,
-	uint32_t length)
-{
-	uint32_t speed = engine->base.funcs->get_speed(&engine->base);
-
-	uint32_t period_timeout;
-	uint32_t num_of_clock_stretches;
-
-	if (!speed)
-		return 0;
-
-	period_timeout = (1000 * TRANSACTION_TIMEOUT_IN_I2C_CLOCKS) / speed;
-
-	num_of_clock_stretches = 1 + (length << 3) + 1;
-	num_of_clock_stretches +=
-		(FROM_I2C_HW_ENGINE(engine)->buffer_used_bytes << 3) +
-		(FROM_I2C_HW_ENGINE(engine)->transaction_count << 1);
-
-	return period_timeout * num_of_clock_stretches;
-}
-
-static void destroy(
-	struct i2c_engine **i2c_engine)
-{
-	struct i2c_hw_engine_dce110 *engine_dce110 =
-			FROM_I2C_ENGINE(*i2c_engine);
-
-	dal_i2c_hw_engine_destruct(&engine_dce110->base);
-
-	kfree(engine_dce110);
-
-	*i2c_engine = NULL;
-}
-
-static const struct i2c_engine_funcs i2c_engine_funcs = {
-	.destroy = destroy,
-	.get_speed = get_speed,
-	.set_speed = set_speed,
-	.setup_engine = setup_engine,
-	.submit_channel_request = submit_channel_request,
-	.process_channel_reply = process_channel_reply,
-	.get_channel_status = get_channel_status,
-	.acquire_engine = dal_i2c_hw_engine_acquire_engine,
-};
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.get_engine_type = dal_i2c_hw_engine_get_engine_type,
-	.acquire = dal_i2c_engine_acquire,
-	.submit_request = dal_i2c_hw_engine_submit_request,
-};
-
-static const struct i2c_hw_engine_funcs i2c_hw_engine_funcs = {
-	.get_hw_buffer_available_size = get_hw_buffer_available_size,
-	.get_transaction_timeout = get_transaction_timeout,
-	.wait_on_operation_result = dal_i2c_hw_engine_wait_on_operation_result,
-};
-
-static void construct(
-	struct i2c_hw_engine_dce110 *hw_engine,
-	const struct i2c_hw_engine_dce110_create_arg *arg)
-{
-	uint32_t xtal_ref_div = 0;
-
-	dal_i2c_hw_engine_construct(&hw_engine->base, arg->ctx);
-
-	hw_engine->base.base.base.funcs = &engine_funcs;
-	hw_engine->base.base.funcs = &i2c_engine_funcs;
-	hw_engine->base.funcs = &i2c_hw_engine_funcs;
-	hw_engine->base.default_speed = arg->default_speed;
-
-	hw_engine->regs = arg->regs;
-	hw_engine->i2c_shift = arg->i2c_shift;
-	hw_engine->i2c_mask = arg->i2c_mask;
-
-	hw_engine->engine_id = arg->engine_id;
-
-	hw_engine->buffer_used_bytes = 0;
-	hw_engine->transaction_count = 0;
-	hw_engine->engine_keep_power_up_count = 1;
-
-
-	REG_GET(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, &xtal_ref_div);
-
-	if (xtal_ref_div == 0) {
-		DC_LOG_WARNING("Invalid base timer divider [%s]\n",
-				__func__);
-		xtal_ref_div = 2;
-	}
-
-	/*Calculating Reference Clock by divding original frequency by
-	 * XTAL_REF_DIV.
-	 * At upper level, uint32_t reference_frequency =
-	 *  dal_i2caux_get_reference_clock(as) >> 1
-	 *  which already divided by 2. So we need x2 to get original
-	 *  reference clock from ppll_info
-	 */
-	hw_engine->reference_frequency =
-		(arg->reference_frequency * 2) / xtal_ref_div;
-}
-
-struct i2c_engine *dal_i2c_hw_engine_dce110_create(
-	const struct i2c_hw_engine_dce110_create_arg *arg)
-{
-	struct i2c_hw_engine_dce110 *engine_dce10;
-
-	if (!arg) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-	if (!arg->reference_frequency) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	engine_dce10 = kzalloc(sizeof(struct i2c_hw_engine_dce110),
-			       GFP_KERNEL);
-
-	if (!engine_dce10) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	construct(engine_dce10, arg);
-	return &engine_dce10->base.base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.h b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.h
deleted file mode 100644
index fea2946906ed..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_hw_engine_dce110.h
+++ /dev/null
@@ -1,218 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_HW_ENGINE_DCE110_H__
-#define __DAL_I2C_HW_ENGINE_DCE110_H__
-
-#define I2C_HW_ENGINE_COMMON_REG_LIST(id)\
-	SRI(SETUP, DC_I2C_DDC, id),\
-	SRI(SPEED, DC_I2C_DDC, id),\
-	SR(DC_I2C_ARBITRATION),\
-	SR(DC_I2C_CONTROL),\
-	SR(DC_I2C_SW_STATUS),\
-	SR(DC_I2C_TRANSACTION0),\
-	SR(DC_I2C_TRANSACTION1),\
-	SR(DC_I2C_TRANSACTION2),\
-	SR(DC_I2C_TRANSACTION3),\
-	SR(DC_I2C_DATA),\
-	SR(MICROSECOND_TIME_BASE_DIV)
-
-#define I2C_SF(reg_name, field_name, post_fix)\
-	.field_name = reg_name ## __ ## field_name ## post_fix
-
-#define I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh)\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_TIME_LIMIT, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_DATA_DRIVE_EN, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_CLK_DRIVE_EN, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_DATA_DRIVE_SEL, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_INTRA_TRANSACTION_DELAY, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_INTRA_BYTE_DELAY, mask_sh),\
-	I2C_SF(DC_I2C_ARBITRATION, DC_I2C_SW_DONE_USING_I2C_REG, mask_sh),\
-	I2C_SF(DC_I2C_ARBITRATION, DC_I2C_NO_QUEUED_SW_GO, mask_sh),\
-	I2C_SF(DC_I2C_ARBITRATION, DC_I2C_SW_PRIORITY, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_SOFT_RESET, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_SW_STATUS_RESET, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_GO, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_SEND_RESET, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_TRANSACTION_COUNT, mask_sh),\
-	I2C_SF(DC_I2C_CONTROL, DC_I2C_DDC_SELECT, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE, mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD, mask_sh),\
-	I2C_SF(DC_I2C_SW_STATUS, DC_I2C_SW_STOPPED_ON_NACK, mask_sh),\
-	I2C_SF(DC_I2C_SW_STATUS, DC_I2C_SW_TIMEOUT, mask_sh),\
-	I2C_SF(DC_I2C_SW_STATUS, DC_I2C_SW_ABORTED, mask_sh),\
-	I2C_SF(DC_I2C_SW_STATUS, DC_I2C_SW_DONE, mask_sh),\
-	I2C_SF(DC_I2C_SW_STATUS, DC_I2C_SW_STATUS, mask_sh),\
-	I2C_SF(DC_I2C_TRANSACTION0, DC_I2C_STOP_ON_NACK0, mask_sh),\
-	I2C_SF(DC_I2C_TRANSACTION0, DC_I2C_START0, mask_sh),\
-	I2C_SF(DC_I2C_TRANSACTION0, DC_I2C_RW0, mask_sh),\
-	I2C_SF(DC_I2C_TRANSACTION0, DC_I2C_STOP0, mask_sh),\
-	I2C_SF(DC_I2C_TRANSACTION0, DC_I2C_COUNT0, mask_sh),\
-	I2C_SF(DC_I2C_DATA, DC_I2C_DATA_RW, mask_sh),\
-	I2C_SF(DC_I2C_DATA, DC_I2C_DATA, mask_sh),\
-	I2C_SF(DC_I2C_DATA, DC_I2C_INDEX, mask_sh),\
-	I2C_SF(DC_I2C_DATA, DC_I2C_INDEX_WRITE, mask_sh),\
-	I2C_SF(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, mask_sh)
-
-#define I2C_COMMON_MASK_SH_LIST_DCE100(mask_sh)\
-	I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh)
-
-#define I2C_COMMON_MASK_SH_LIST_DCE110(mask_sh)\
-	I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh),\
-	I2C_SF(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL, mask_sh)
-
-struct dce110_i2c_hw_engine_shift {
-	uint8_t DC_I2C_DDC1_ENABLE;
-	uint8_t DC_I2C_DDC1_TIME_LIMIT;
-	uint8_t DC_I2C_DDC1_DATA_DRIVE_EN;
-	uint8_t DC_I2C_DDC1_CLK_DRIVE_EN;
-	uint8_t DC_I2C_DDC1_DATA_DRIVE_SEL;
-	uint8_t DC_I2C_DDC1_INTRA_TRANSACTION_DELAY;
-	uint8_t DC_I2C_DDC1_INTRA_BYTE_DELAY;
-	uint8_t DC_I2C_SW_DONE_USING_I2C_REG;
-	uint8_t DC_I2C_NO_QUEUED_SW_GO;
-	uint8_t DC_I2C_SW_PRIORITY;
-	uint8_t DC_I2C_SOFT_RESET;
-	uint8_t DC_I2C_SW_STATUS_RESET;
-	uint8_t DC_I2C_GO;
-	uint8_t DC_I2C_SEND_RESET;
-	uint8_t DC_I2C_TRANSACTION_COUNT;
-	uint8_t DC_I2C_DDC_SELECT;
-	uint8_t DC_I2C_DDC1_PRESCALE;
-	uint8_t DC_I2C_DDC1_THRESHOLD;
-	uint8_t DC_I2C_DDC1_START_STOP_TIMING_CNTL;
-	uint8_t DC_I2C_SW_STOPPED_ON_NACK;
-	uint8_t DC_I2C_SW_TIMEOUT;
-	uint8_t DC_I2C_SW_ABORTED;
-	uint8_t DC_I2C_SW_DONE;
-	uint8_t DC_I2C_SW_STATUS;
-	uint8_t DC_I2C_STOP_ON_NACK0;
-	uint8_t DC_I2C_START0;
-	uint8_t DC_I2C_RW0;
-	uint8_t DC_I2C_STOP0;
-	uint8_t DC_I2C_COUNT0;
-	uint8_t DC_I2C_DATA_RW;
-	uint8_t DC_I2C_DATA;
-	uint8_t DC_I2C_INDEX;
-	uint8_t DC_I2C_INDEX_WRITE;
-	uint8_t XTAL_REF_DIV;
-};
-
-struct dce110_i2c_hw_engine_mask {
-	uint32_t DC_I2C_DDC1_ENABLE;
-	uint32_t DC_I2C_DDC1_TIME_LIMIT;
-	uint32_t DC_I2C_DDC1_DATA_DRIVE_EN;
-	uint32_t DC_I2C_DDC1_CLK_DRIVE_EN;
-	uint32_t DC_I2C_DDC1_DATA_DRIVE_SEL;
-	uint32_t DC_I2C_DDC1_INTRA_TRANSACTION_DELAY;
-	uint32_t DC_I2C_DDC1_INTRA_BYTE_DELAY;
-	uint32_t DC_I2C_SW_DONE_USING_I2C_REG;
-	uint32_t DC_I2C_NO_QUEUED_SW_GO;
-	uint32_t DC_I2C_SW_PRIORITY;
-	uint32_t DC_I2C_SOFT_RESET;
-	uint32_t DC_I2C_SW_STATUS_RESET;
-	uint32_t DC_I2C_GO;
-	uint32_t DC_I2C_SEND_RESET;
-	uint32_t DC_I2C_TRANSACTION_COUNT;
-	uint32_t DC_I2C_DDC_SELECT;
-	uint32_t DC_I2C_DDC1_PRESCALE;
-	uint32_t DC_I2C_DDC1_THRESHOLD;
-	uint32_t DC_I2C_DDC1_START_STOP_TIMING_CNTL;
-	uint32_t DC_I2C_SW_STOPPED_ON_NACK;
-	uint32_t DC_I2C_SW_TIMEOUT;
-	uint32_t DC_I2C_SW_ABORTED;
-	uint32_t DC_I2C_SW_DONE;
-	uint32_t DC_I2C_SW_STATUS;
-	uint32_t DC_I2C_STOP_ON_NACK0;
-	uint32_t DC_I2C_START0;
-	uint32_t DC_I2C_RW0;
-	uint32_t DC_I2C_STOP0;
-	uint32_t DC_I2C_COUNT0;
-	uint32_t DC_I2C_DATA_RW;
-	uint32_t DC_I2C_DATA;
-	uint32_t DC_I2C_INDEX;
-	uint32_t DC_I2C_INDEX_WRITE;
-	uint32_t XTAL_REF_DIV;
-};
-
-struct dce110_i2c_hw_engine_registers {
-	uint32_t SETUP;
-	uint32_t SPEED;
-	uint32_t DC_I2C_ARBITRATION;
-	uint32_t DC_I2C_CONTROL;
-	uint32_t DC_I2C_SW_STATUS;
-	uint32_t DC_I2C_TRANSACTION0;
-	uint32_t DC_I2C_TRANSACTION1;
-	uint32_t DC_I2C_TRANSACTION2;
-	uint32_t DC_I2C_TRANSACTION3;
-	uint32_t DC_I2C_DATA;
-	uint32_t MICROSECOND_TIME_BASE_DIV;
-};
-
-struct i2c_hw_engine_dce110 {
-	struct i2c_hw_engine base;
-	const struct dce110_i2c_hw_engine_registers *regs;
-	const struct dce110_i2c_hw_engine_shift *i2c_shift;
-	const struct dce110_i2c_hw_engine_mask *i2c_mask;
-	struct {
-		uint32_t DC_I2C_DDCX_SETUP;
-		uint32_t DC_I2C_DDCX_SPEED;
-	} addr;
-	uint32_t engine_id;
-	/* expressed in kilohertz */
-	uint32_t reference_frequency;
-	/* number of bytes currently used in HW buffer */
-	uint32_t buffer_used_bytes;
-	/* number of bytes used for write transaction in HW buffer
-	 * - this will be used as the index to read from*/
-	uint32_t buffer_used_write;
-	/* number of pending transactions (before GO) */
-	uint32_t transaction_count;
-	uint32_t engine_keep_power_up_count;
-	uint32_t i2_setup_time_limit;
-};
-
-struct i2c_hw_engine_dce110_create_arg {
-	uint32_t engine_id;
-	uint32_t reference_frequency;
-	uint32_t default_speed;
-	struct dc_context *ctx;
-	const struct dce110_i2c_hw_engine_registers *regs;
-	const struct dce110_i2c_hw_engine_shift *i2c_shift;
-	const struct dce110_i2c_hw_engine_mask *i2c_mask;
-};
-
-struct i2c_engine *dal_i2c_hw_engine_dce110_create(
-	const struct i2c_hw_engine_dce110_create_arg *arg);
-
-enum {
-	I2C_SETUP_TIME_LIMIT_DCE = 255,
-	I2C_SETUP_TIME_LIMIT_DCN = 3,
-	I2C_HW_BUFFER_SIZE = 538,
-	I2C_SEND_RESET_LENGTH_9 = 9,
-	I2C_SEND_RESET_LENGTH_10 = 10,
-};
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_sw_engine_dce110.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_sw_engine_dce110.c
deleted file mode 100644
index 3aa7f791e523..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_sw_engine_dce110.c
+++ /dev/null
@@ -1,160 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_sw_engine_dce110.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-/*
- * @brief
- * Cast 'struct i2c_sw_engine *'
- * to 'struct i2c_sw_engine_dce110 *'
- */
-#define FROM_I2C_SW_ENGINE(ptr) \
-	container_of((ptr), struct i2c_sw_engine_dce110, base)
-/*
- * @brief
- * Cast 'struct i2c_engine *'
- * to 'struct i2c_sw_engine_dce80 *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	FROM_I2C_SW_ENGINE(container_of((ptr), struct i2c_sw_engine, base))
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct i2c_sw_engine_dce80 *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-static void release_engine(
-	struct engine *engine)
-{
-}
-
-static void destruct(
-	struct i2c_sw_engine_dce110 *engine)
-{
-	dal_i2c_sw_engine_destruct(&engine->base);
-}
-
-static void destroy(
-	struct i2c_engine **engine)
-{
-	struct i2c_sw_engine_dce110 *sw_engine = FROM_I2C_ENGINE(*engine);
-
-	destruct(sw_engine);
-
-	kfree(sw_engine);
-
-	*engine = NULL;
-}
-
-static bool acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc_handle)
-{
-	return dal_i2caux_i2c_sw_engine_acquire_engine(engine, ddc_handle);
-}
-
-static const struct i2c_engine_funcs i2c_engine_funcs = {
-	.acquire_engine = acquire_engine,
-	.destroy = destroy,
-	.get_speed = dal_i2c_sw_engine_get_speed,
-	.set_speed = dal_i2c_sw_engine_set_speed,
-	.setup_engine = dal_i2c_engine_setup_i2c_engine,
-	.submit_channel_request = dal_i2c_sw_engine_submit_channel_request,
-	.process_channel_reply = dal_i2c_engine_process_channel_reply,
-	.get_channel_status = dal_i2c_sw_engine_get_channel_status,
-};
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.get_engine_type = dal_i2c_sw_engine_get_engine_type,
-	.acquire = dal_i2c_engine_acquire,
-	.submit_request = dal_i2c_sw_engine_submit_request,
-};
-
-static void construct(
-	struct i2c_sw_engine_dce110 *engine_dce110,
-	const struct i2c_sw_engine_dce110_create_arg *arg_dce110)
-{
-	struct i2c_sw_engine_create_arg arg_base;
-
-	arg_base.ctx = arg_dce110->ctx;
-	arg_base.default_speed = arg_dce110->default_speed;
-
-	dal_i2c_sw_engine_construct(&engine_dce110->base, &arg_base);
-
-	/*struct engine   struct engine_funcs*/
-	engine_dce110->base.base.base.funcs = &engine_funcs;
-	/*struct i2c_engine  struct i2c_engine_funcs*/
-	engine_dce110->base.base.funcs = &i2c_engine_funcs;
-	engine_dce110->base.default_speed = arg_dce110->default_speed;
-	engine_dce110->engine_id = arg_dce110->engine_id;
-}
-
-struct i2c_engine *dal_i2c_sw_engine_dce110_create(
-	const struct i2c_sw_engine_dce110_create_arg *arg)
-{
-	struct i2c_sw_engine_dce110 *engine_dce110;
-
-	if (!arg) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	engine_dce110 = kzalloc(sizeof(struct i2c_sw_engine_dce110),
-				GFP_KERNEL);
-
-	if (!engine_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	construct(engine_dce110, arg);
-	return &engine_dce110->base.base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.c
deleted file mode 100644
index 1d748ac1d6d6..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.c
+++ /dev/null
@@ -1,329 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-/*
- * Header of this unit
- */
-#include "i2caux_dce110.h"
-
-#include "i2c_sw_engine_dce110.h"
-#include "i2c_hw_engine_dce110.h"
-#include "aux_engine_dce110.h"
-#include "../../dc.h"
-#include "dc_types.h"
-
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-/*cast pointer to struct i2caux TO pointer to struct i2caux_dce110*/
-#define FROM_I2C_AUX(ptr) \
-	container_of((ptr), struct i2caux_dce110, base)
-
-static void destruct(
-	struct i2caux_dce110 *i2caux_dce110)
-{
-	dal_i2caux_destruct(&i2caux_dce110->base);
-}
-
-static void destroy(
-	struct i2caux **i2c_engine)
-{
-	struct i2caux_dce110 *i2caux_dce110 = FROM_I2C_AUX(*i2c_engine);
-
-	destruct(i2caux_dce110);
-
-	kfree(i2caux_dce110);
-
-	*i2c_engine = NULL;
-}
-
-static struct i2c_engine *acquire_i2c_hw_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc)
-{
-	struct i2caux_dce110 *i2caux_dce110 = FROM_I2C_AUX(i2caux);
-
-	struct i2c_engine *engine = NULL;
-	/* generic hw engine is not used for EDID read
-	 * It may be needed for external i2c device, like thermal chip,
-	 * TODO will be implemented when needed.
-	 * check dce80 bool non_generic for generic hw engine;
-	 */
-
-	if (!ddc)
-		return NULL;
-
-	if (ddc->hw_info.hw_supported) {
-		enum gpio_ddc_line line = dal_ddc_get_line(ddc);
-
-		if (line < GPIO_DDC_LINE_COUNT)
-			engine = i2caux->i2c_hw_engines[line];
-	}
-
-	if (!engine)
-		return NULL;
-
-	if (!i2caux_dce110->i2c_hw_buffer_in_use &&
-		engine->base.funcs->acquire(&engine->base, ddc)) {
-		i2caux_dce110->i2c_hw_buffer_in_use = true;
-		return engine;
-	}
-
-	return NULL;
-}
-
-static void release_engine(
-	struct i2caux *i2caux,
-	struct engine *engine)
-{
-	struct i2caux_dce110 *i2caux_dce110 = FROM_I2C_AUX(i2caux);
-
-	if (engine->funcs->get_engine_type(engine) ==
-		I2CAUX_ENGINE_TYPE_I2C_DDC_HW)
-		i2caux_dce110->i2c_hw_buffer_in_use = false;
-
-	dal_i2caux_release_engine(i2caux, engine);
-}
-
-static const enum gpio_ddc_line hw_ddc_lines[] = {
-	GPIO_DDC_LINE_DDC1,
-	GPIO_DDC_LINE_DDC2,
-	GPIO_DDC_LINE_DDC3,
-	GPIO_DDC_LINE_DDC4,
-	GPIO_DDC_LINE_DDC5,
-	GPIO_DDC_LINE_DDC6,
-};
-
-static const enum gpio_ddc_line hw_aux_lines[] = {
-	GPIO_DDC_LINE_DDC1,
-	GPIO_DDC_LINE_DDC2,
-	GPIO_DDC_LINE_DDC3,
-	GPIO_DDC_LINE_DDC4,
-	GPIO_DDC_LINE_DDC5,
-	GPIO_DDC_LINE_DDC6,
-};
-
-/* function table */
-static const struct i2caux_funcs i2caux_funcs = {
-	.destroy = destroy,
-	.acquire_i2c_hw_engine = acquire_i2c_hw_engine,
-	.release_engine = release_engine,
-	.acquire_i2c_sw_engine = dal_i2caux_acquire_i2c_sw_engine,
-	.acquire_aux_engine = dal_i2caux_acquire_aux_engine,
-};
-
-#include "dce/dce_11_0_d.h"
-#include "dce/dce_11_0_sh_mask.h"
-
-/* 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 aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = AUX_CONTROL__AUX_RESET_MASK \
-}
-
-#define hw_engine_regs(id)\
-{\
-		I2C_HW_ENGINE_COMMON_REG_LIST(id) \
-}
-
-static const struct dce110_aux_registers dce110_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5)
-};
-
-static const struct dce110_i2c_hw_engine_registers i2c_hw_engine_regs[] = {
-		hw_engine_regs(1),
-		hw_engine_regs(2),
-		hw_engine_regs(3),
-		hw_engine_regs(4),
-		hw_engine_regs(5),
-		hw_engine_regs(6)
-};
-
-static const struct dce110_i2c_hw_engine_shift i2c_shift = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(__SHIFT)
-};
-
-static const struct dce110_i2c_hw_engine_mask i2c_mask = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(_MASK)
-};
-
-void dal_i2caux_dce110_construct(
-	struct i2caux_dce110 *i2caux_dce110,
-	struct dc_context *ctx,
-	unsigned int num_i2caux_inst,
-	const struct dce110_aux_registers aux_regs[],
-	const struct dce110_i2c_hw_engine_registers i2c_hw_engine_regs[],
-	const struct dce110_i2c_hw_engine_shift *i2c_shift,
-	const struct dce110_i2c_hw_engine_mask *i2c_mask)
-{
-	uint32_t i = 0;
-	uint32_t reference_frequency = 0;
-	bool use_i2c_sw_engine = false;
-	struct i2caux *base = NULL;
-	/*TODO: For CZ bring up, if dal_i2caux_get_reference_clock
-	 * does not return 48KHz, we need hard coded for 48Khz.
-	 * Some BIOS setting incorrect cause this
-	 * For production, we always get value from BIOS*/
-	reference_frequency =
-		dal_i2caux_get_reference_clock(ctx->dc_bios) >> 1;
-
-	base = &i2caux_dce110->base;
-
-	dal_i2caux_construct(base, ctx);
-
-	i2caux_dce110->base.funcs = &i2caux_funcs;
-	i2caux_dce110->i2c_hw_buffer_in_use = false;
-	/* Create I2C engines (DDC lines per connector)
-	 * different I2C/AUX usage cases, DDC, Generic GPIO, AUX.
-	 */
-	do {
-		enum gpio_ddc_line line_id = hw_ddc_lines[i];
-
-		struct i2c_hw_engine_dce110_create_arg hw_arg_dce110;
-
-		if (use_i2c_sw_engine) {
-			struct i2c_sw_engine_dce110_create_arg sw_arg;
-
-			sw_arg.engine_id = i;
-			sw_arg.default_speed = base->default_i2c_sw_speed;
-			sw_arg.ctx = ctx;
-			base->i2c_sw_engines[line_id] =
-				dal_i2c_sw_engine_dce110_create(&sw_arg);
-		}
-
-		hw_arg_dce110.engine_id = i;
-		hw_arg_dce110.reference_frequency = reference_frequency;
-		hw_arg_dce110.default_speed = base->default_i2c_hw_speed;
-		hw_arg_dce110.ctx = ctx;
-		hw_arg_dce110.regs = &i2c_hw_engine_regs[i];
-		hw_arg_dce110.i2c_shift = i2c_shift;
-		hw_arg_dce110.i2c_mask = i2c_mask;
-
-		base->i2c_hw_engines[line_id] =
-			dal_i2c_hw_engine_dce110_create(&hw_arg_dce110);
-		if (base->i2c_hw_engines[line_id] != NULL) {
-			switch (ctx->dce_version) {
-			case DCN_VERSION_1_0:
-				base->i2c_hw_engines[line_id]->setup_limit =
-					I2C_SETUP_TIME_LIMIT_DCN;
-				base->i2c_hw_engines[line_id]->send_reset_length  = 0;
-			break;
-			default:
-				base->i2c_hw_engines[line_id]->setup_limit =
-					I2C_SETUP_TIME_LIMIT_DCE;
-				base->i2c_hw_engines[line_id]->send_reset_length  = 0;
-				break;
-			}
-		}
-		++i;
-	} while (i < num_i2caux_inst);
-
-	/* Create AUX engines for all lines which has assisted HW AUX
-	 * 'i' (loop counter) used as DDC/AUX engine_id */
-
-	i = 0;
-
-	do {
-		enum gpio_ddc_line line_id = hw_aux_lines[i];
-
-		struct aux_engine_dce110_init_data aux_init_data;
-
-		aux_init_data.engine_id = i;
-		aux_init_data.timeout_period = base->aux_timeout_period;
-		aux_init_data.ctx = ctx;
-		aux_init_data.regs = &aux_regs[i];
-
-		base->aux_engines[line_id] =
-			dal_aux_engine_dce110_create(&aux_init_data);
-
-		++i;
-	} while (i < num_i2caux_inst);
-
-	/*TODO Generic I2C SW and HW*/
-}
-
-/*
- * dal_i2caux_dce110_create
- *
- * @brief
- * public interface to allocate memory for DCE11 I2CAUX
- *
- * @param
- * struct adapter_service *as - [in]
- * struct dc_context *ctx - [in]
- *
- * @return
- * pointer to the base struct of DCE11 I2CAUX
- */
-struct i2caux *dal_i2caux_dce110_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce110 *i2caux_dce110 =
-		kzalloc(sizeof(struct i2caux_dce110), GFP_KERNEL);
-
-	if (!i2caux_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	dal_i2caux_dce110_construct(i2caux_dce110,
-				    ctx,
-				    ARRAY_SIZE(dce110_aux_regs),
-				    dce110_aux_regs,
-				    i2c_hw_engine_regs,
-				    &i2c_shift,
-				    &i2c_mask);
-	return &i2caux_dce110->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.h b/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.h
deleted file mode 100644
index d3d8cc58666a..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2caux_dce110.h
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_AUX_DCE110_H__
-#define __DAL_I2C_AUX_DCE110_H__
-
-#include "../i2caux.h"
-
-struct i2caux_dce110 {
-	struct i2caux base;
-	/* indicate the I2C HW circular buffer is in use */
-	bool i2c_hw_buffer_in_use;
-};
-
-struct dce110_aux_registers;
-struct dce110_i2c_hw_engine_registers;
-struct dce110_i2c_hw_engine_shift;
-struct dce110_i2c_hw_engine_mask;
-
-struct i2caux *dal_i2caux_dce110_create(
-	struct dc_context *ctx);
-
-void dal_i2caux_dce110_construct(
-	struct i2caux_dce110 *i2caux_dce110,
-	struct dc_context *ctx,
-	unsigned int num_i2caux_inst,
-	const struct dce110_aux_registers *aux_regs,
-	const struct dce110_i2c_hw_engine_registers *i2c_hw_engine_regs,
-	const struct dce110_i2c_hw_engine_shift *i2c_shift,
-	const struct dce110_i2c_hw_engine_mask *i2c_mask);
-
-#endif /* __DAL_I2C_AUX_DCE110_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.c
deleted file mode 100644
index a9db04738724..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.c
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-#include "../dce110/i2caux_dce110.h"
-#include "i2caux_dce112.h"
-
-#include "../dce110/aux_engine_dce110.h"
-
-#include "../dce110/i2c_hw_engine_dce110.h"
-
-#include "dce/dce_11_2_d.h"
-#include "dce/dce_11_2_sh_mask.h"
-
-/* 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 aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = AUX_CONTROL__AUX_RESET_MASK \
-}
-
-#define hw_engine_regs(id)\
-{\
-		I2C_HW_ENGINE_COMMON_REG_LIST(id) \
-}
-
-static const struct dce110_aux_registers dce112_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5),
-};
-
-static const struct dce110_i2c_hw_engine_registers dce112_hw_engine_regs[] = {
-		hw_engine_regs(1),
-		hw_engine_regs(2),
-		hw_engine_regs(3),
-		hw_engine_regs(4),
-		hw_engine_regs(5),
-		hw_engine_regs(6)
-};
-
-static const struct dce110_i2c_hw_engine_shift i2c_shift = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(__SHIFT)
-};
-
-static const struct dce110_i2c_hw_engine_mask i2c_mask = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(_MASK)
-};
-
-static void construct(
-	struct i2caux_dce110 *i2caux_dce110,
-	struct dc_context *ctx)
-{
-	dal_i2caux_dce110_construct(i2caux_dce110,
-				    ctx,
-				    ARRAY_SIZE(dce112_aux_regs),
-				    dce112_aux_regs,
-				    dce112_hw_engine_regs,
-				    &i2c_shift,
-				    &i2c_mask);
-}
-
-/*
- * dal_i2caux_dce110_create
- *
- * @brief
- * public interface to allocate memory for DCE11 I2CAUX
- *
- * @param
- * struct adapter_service *as - [in]
- * struct dc_context *ctx - [in]
- *
- * @return
- * pointer to the base struct of DCE11 I2CAUX
- */
-struct i2caux *dal_i2caux_dce112_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce110 *i2caux_dce110 =
-		kzalloc(sizeof(struct i2caux_dce110), GFP_KERNEL);
-
-	if (!i2caux_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	construct(i2caux_dce110, ctx);
-	return &i2caux_dce110->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.h b/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.h
deleted file mode 100644
index 8d35453c25b6..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce112/i2caux_dce112.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_AUX_DCE112_H__
-#define __DAL_I2C_AUX_DCE112_H__
-
-struct i2caux *dal_i2caux_dce112_create(
-	struct dc_context *ctx);
-
-#endif /* __DAL_I2C_AUX_DCE112_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce120/i2caux_dce120.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce120/i2caux_dce120.c
deleted file mode 100644
index 6a4f344c1db4..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce120/i2caux_dce120.c
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
- * Copyright 2012-16 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
- *
- */
-
-#include "dm_services.h"
-
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-#include "../dce110/i2c_hw_engine_dce110.h"
-#include "../dce110/aux_engine_dce110.h"
-#include "../dce110/i2caux_dce110.h"
-
-#include "dce/dce_12_0_offset.h"
-#include "dce/dce_12_0_sh_mask.h"
-#include "soc15_hw_ip.h"
-#include "vega10_ip_offset.h"
-
-/* begin *********************
- * macros to expend register list macro defined in HW object header file */
-
-#define BASE_INNER(seg) \
-	DCE_BASE__INST0_SEG ## seg
-
-/* compile time expand base address. */
-#define BASE(seg) \
-	BASE_INNER(seg)
-
-#define SR(reg_name)\
-		.reg_name = BASE(mm ## reg_name ## _BASE_IDX) +  \
-					mm ## reg_name
-
-#define SRI(reg_name, block, id)\
-	.reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
-					mm ## block ## id ## _ ## reg_name
-/* macros to expend register list macro defined in HW object header file
- * end *********************/
-
-#define aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = DP_AUX0_AUX_CONTROL__AUX_RESET_MASK \
-}
-
-static const struct dce110_aux_registers dce120_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5),
-};
-
-#define hw_engine_regs(id)\
-{\
-		I2C_HW_ENGINE_COMMON_REG_LIST(id) \
-}
-
-static const struct dce110_i2c_hw_engine_registers dce120_hw_engine_regs[] = {
-		hw_engine_regs(1),
-		hw_engine_regs(2),
-		hw_engine_regs(3),
-		hw_engine_regs(4),
-		hw_engine_regs(5),
-		hw_engine_regs(6)
-};
-
-static const struct dce110_i2c_hw_engine_shift i2c_shift = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(__SHIFT)
-};
-
-static const struct dce110_i2c_hw_engine_mask i2c_mask = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(_MASK)
-};
-
-struct i2caux *dal_i2caux_dce120_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce110 *i2caux_dce110 =
-		kzalloc(sizeof(struct i2caux_dce110), GFP_KERNEL);
-
-	if (!i2caux_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	dal_i2caux_dce110_construct(i2caux_dce110,
-				    ctx,
-				    ARRAY_SIZE(dce120_aux_regs),
-				    dce120_aux_regs,
-				    dce120_hw_engine_regs,
-				    &i2c_shift,
-				    &i2c_mask);
-	return &i2caux_dce110->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.c
deleted file mode 100644
index fd0832dd2c75..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.c
+++ /dev/null
@@ -1,875 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_hw_engine.h"
-#include "../i2c_generic_hw_engine.h"
-/*
- * Header of this unit
- */
-
-#include "i2c_hw_engine_dce80.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-#include "dce/dce_8_0_d.h"
-#include "dce/dce_8_0_sh_mask.h"
-/*
- * This unit
- */
-
-enum dc_i2c_status {
-	DC_I2C_STATUS__DC_I2C_STATUS_IDLE,
-	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_SW,
-	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_HW
-};
-
-enum dc_i2c_arbitration {
-	DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_NORMAL,
-	DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_HIGH
-};
-
-enum {
-	/* No timeout in HW
-	 * (timeout implemented in SW by querying status) */
-	I2C_SETUP_TIME_LIMIT = 255,
-	I2C_HW_BUFFER_SIZE = 144
-};
-
-/*
- * @brief
- * Cast 'struct i2c_hw_engine *'
- * to 'struct i2c_hw_engine_dce80 *'
- */
-#define FROM_I2C_HW_ENGINE(ptr) \
-	container_of((ptr), struct i2c_hw_engine_dce80, base)
-
-/*
- * @brief
- * Cast pointer to 'struct i2c_engine *'
- * to pointer to 'struct i2c_hw_engine_dce80 *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	FROM_I2C_HW_ENGINE(container_of((ptr), struct i2c_hw_engine, base))
-
-/*
- * @brief
- * Cast pointer to 'struct engine *'
- * to 'pointer to struct i2c_hw_engine_dce80 *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-static void disable_i2c_hw_engine(
-	struct i2c_hw_engine_dce80 *engine)
-{
-	const uint32_t addr = engine->addr.DC_I2C_DDCX_SETUP;
-	uint32_t value = 0;
-
-	struct dc_context *ctx = NULL;
-
-	ctx = engine->base.base.base.ctx;
-
-	value = dm_read_reg(ctx, addr);
-
-	set_reg_field_value(
-		value,
-		0,
-		DC_I2C_DDC1_SETUP,
-		DC_I2C_DDC1_ENABLE);
-
-	dm_write_reg(ctx, addr, value);
-}
-
-static void release_engine(
-	struct engine *engine)
-{
-	struct i2c_hw_engine_dce80 *hw_engine = FROM_ENGINE(engine);
-
-	struct i2c_engine *base = NULL;
-	bool safe_to_reset;
-	uint32_t value = 0;
-
-	base = &hw_engine->base.base;
-
-	/* Restore original HW engine speed */
-
-	base->funcs->set_speed(base, hw_engine->base.original_speed);
-
-	/* Release I2C */
-	{
-		value = dm_read_reg(engine->ctx, mmDC_I2C_ARBITRATION);
-
-		set_reg_field_value(
-				value,
-				1,
-				DC_I2C_ARBITRATION,
-				DC_I2C_SW_DONE_USING_I2C_REG);
-
-		dm_write_reg(engine->ctx, mmDC_I2C_ARBITRATION, value);
-	}
-
-	/* Reset HW engine */
-	{
-		uint32_t i2c_sw_status = 0;
-
-		value = dm_read_reg(engine->ctx, mmDC_I2C_SW_STATUS);
-
-		i2c_sw_status = get_reg_field_value(
-				value,
-				DC_I2C_SW_STATUS,
-				DC_I2C_SW_STATUS);
-		/* if used by SW, safe to reset */
-		safe_to_reset = (i2c_sw_status == 1);
-	}
-	{
-		value = dm_read_reg(engine->ctx, mmDC_I2C_CONTROL);
-
-		if (safe_to_reset)
-			set_reg_field_value(
-				value,
-				1,
-				DC_I2C_CONTROL,
-				DC_I2C_SOFT_RESET);
-
-		set_reg_field_value(
-			value,
-			1,
-			DC_I2C_CONTROL,
-			DC_I2C_SW_STATUS_RESET);
-
-		dm_write_reg(engine->ctx, mmDC_I2C_CONTROL, value);
-	}
-
-	/* HW I2c engine - clock gating feature */
-	if (!hw_engine->engine_keep_power_up_count)
-		disable_i2c_hw_engine(hw_engine);
-}
-
-static void destruct(
-	struct i2c_hw_engine_dce80 *engine)
-{
-	dal_i2c_hw_engine_destruct(&engine->base);
-}
-
-static void destroy(
-	struct i2c_engine **i2c_engine)
-{
-	struct i2c_hw_engine_dce80 *engine = FROM_I2C_ENGINE(*i2c_engine);
-
-	destruct(engine);
-
-	kfree(engine);
-
-	*i2c_engine = NULL;
-}
-
-static bool setup_engine(
-	struct i2c_engine *i2c_engine)
-{
-	uint32_t value = 0;
-	struct i2c_hw_engine_dce80 *engine = FROM_I2C_ENGINE(i2c_engine);
-
-	/* Program pin select */
-	{
-		const uint32_t addr = mmDC_I2C_CONTROL;
-
-		value = dm_read_reg(i2c_engine->base.ctx, addr);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_GO);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SOFT_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SEND_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SW_STATUS_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_TRANSACTION_COUNT);
-
-		set_reg_field_value(
-			value,
-			engine->engine_id,
-			DC_I2C_CONTROL,
-			DC_I2C_DDC_SELECT);
-
-		dm_write_reg(i2c_engine->base.ctx, addr, value);
-	}
-
-	/* Program time limit */
-	{
-		const uint32_t addr = engine->addr.DC_I2C_DDCX_SETUP;
-
-		value = dm_read_reg(i2c_engine->base.ctx, addr);
-
-		set_reg_field_value(
-			value,
-			I2C_SETUP_TIME_LIMIT,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_TIME_LIMIT);
-
-		set_reg_field_value(
-			value,
-			1,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_ENABLE);
-
-		dm_write_reg(i2c_engine->base.ctx, addr, value);
-	}
-
-	/* Program HW priority
-	 * set to High - interrupt software I2C at any time
-	 * Enable restart of SW I2C that was interrupted by HW
-	 * disable queuing of software while I2C is in use by HW */
-	{
-		value = dm_read_reg(i2c_engine->base.ctx,
-				mmDC_I2C_ARBITRATION);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_ARBITRATION,
-			DC_I2C_NO_QUEUED_SW_GO);
-
-		set_reg_field_value(
-			value,
-			DC_I2C_ARBITRATION__DC_I2C_SW_PRIORITY_NORMAL,
-			DC_I2C_ARBITRATION,
-			DC_I2C_SW_PRIORITY);
-
-		dm_write_reg(i2c_engine->base.ctx,
-				mmDC_I2C_ARBITRATION, value);
-	}
-
-	return true;
-}
-
-static uint32_t get_speed(
-	const struct i2c_engine *i2c_engine)
-{
-	const struct i2c_hw_engine_dce80 *engine = FROM_I2C_ENGINE(i2c_engine);
-
-	const uint32_t addr = engine->addr.DC_I2C_DDCX_SPEED;
-
-	uint32_t pre_scale = 0;
-
-	uint32_t value = dm_read_reg(i2c_engine->base.ctx, addr);
-
-	pre_scale = get_reg_field_value(
-			value,
-			DC_I2C_DDC1_SPEED,
-			DC_I2C_DDC1_PRESCALE);
-
-	/* [anaumov] it seems following is unnecessary */
-	/*ASSERT(value.bits.DC_I2C_DDC1_PRESCALE);*/
-
-	return pre_scale ?
-		engine->reference_frequency / pre_scale :
-		engine->base.default_speed;
-}
-
-static void set_speed(
-	struct i2c_engine *i2c_engine,
-	uint32_t speed)
-{
-	struct i2c_hw_engine_dce80 *engine = FROM_I2C_ENGINE(i2c_engine);
-
-	if (speed) {
-		const uint32_t addr = engine->addr.DC_I2C_DDCX_SPEED;
-
-		uint32_t value = dm_read_reg(i2c_engine->base.ctx, addr);
-
-		set_reg_field_value(
-			value,
-			engine->reference_frequency / speed,
-			DC_I2C_DDC1_SPEED,
-			DC_I2C_DDC1_PRESCALE);
-
-		set_reg_field_value(
-			value,
-			2,
-			DC_I2C_DDC1_SPEED,
-			DC_I2C_DDC1_THRESHOLD);
-
-		dm_write_reg(i2c_engine->base.ctx, addr, value);
-	}
-}
-
-static inline void reset_hw_engine(struct engine *engine)
-{
-	uint32_t value = dm_read_reg(engine->ctx, mmDC_I2C_CONTROL);
-
-	set_reg_field_value(
-		value,
-		1,
-		DC_I2C_CONTROL,
-		DC_I2C_SOFT_RESET);
-
-	set_reg_field_value(
-		value,
-		1,
-		DC_I2C_CONTROL,
-		DC_I2C_SW_STATUS_RESET);
-
-	dm_write_reg(engine->ctx, mmDC_I2C_CONTROL, value);
-}
-
-static bool is_hw_busy(struct engine *engine)
-{
-	uint32_t i2c_sw_status = 0;
-
-	uint32_t value = dm_read_reg(engine->ctx, mmDC_I2C_SW_STATUS);
-
-	i2c_sw_status = get_reg_field_value(
-			value,
-			DC_I2C_SW_STATUS,
-			DC_I2C_SW_STATUS);
-
-	if (i2c_sw_status == DC_I2C_STATUS__DC_I2C_STATUS_IDLE)
-		return false;
-
-	reset_hw_engine(engine);
-
-	value = dm_read_reg(engine->ctx, mmDC_I2C_SW_STATUS);
-
-	i2c_sw_status = get_reg_field_value(
-			value,
-			DC_I2C_SW_STATUS,
-			DC_I2C_SW_STATUS);
-
-	return i2c_sw_status != DC_I2C_STATUS__DC_I2C_STATUS_IDLE;
-}
-
-/*
- * @brief
- * DC_GPIO_DDC MM register offsets
- */
-static const uint32_t transaction_addr[] = {
-	mmDC_I2C_TRANSACTION0,
-	mmDC_I2C_TRANSACTION1,
-	mmDC_I2C_TRANSACTION2,
-	mmDC_I2C_TRANSACTION3
-};
-
-static bool process_transaction(
-	struct i2c_hw_engine_dce80 *engine,
-	struct i2c_request_transaction_data *request)
-{
-	uint32_t length = request->length;
-	uint8_t *buffer = request->data;
-
-	bool last_transaction = false;
-	uint32_t value = 0;
-
-	struct dc_context *ctx = NULL;
-
-	ctx = engine->base.base.base.ctx;
-
-	{
-		const uint32_t addr =
-			transaction_addr[engine->transaction_count];
-
-		value = dm_read_reg(ctx, addr);
-
-		set_reg_field_value(
-			value,
-			1,
-			DC_I2C_TRANSACTION0,
-			DC_I2C_STOP_ON_NACK0);
-
-		set_reg_field_value(
-			value,
-			1,
-			DC_I2C_TRANSACTION0,
-			DC_I2C_START0);
-
-		if ((engine->transaction_count == 3) ||
-		(request->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE) ||
-		(request->action & I2CAUX_TRANSACTION_ACTION_I2C_READ)) {
-
-			set_reg_field_value(
-				value,
-				1,
-				DC_I2C_TRANSACTION0,
-				DC_I2C_STOP0);
-
-			last_transaction = true;
-		} else
-			set_reg_field_value(
-				value,
-				0,
-				DC_I2C_TRANSACTION0,
-				DC_I2C_STOP0);
-
-		set_reg_field_value(
-			value,
-			(0 != (request->action &
-					I2CAUX_TRANSACTION_ACTION_I2C_READ)),
-			DC_I2C_TRANSACTION0,
-			DC_I2C_RW0);
-
-		set_reg_field_value(
-			value,
-			length,
-			DC_I2C_TRANSACTION0,
-			DC_I2C_COUNT0);
-
-		dm_write_reg(ctx, addr, value);
-	}
-
-	/* Write the I2C address and I2C data
-	 * into the hardware circular buffer, one byte per entry.
-	 * As an example, the 7-bit I2C slave address for CRT monitor
-	 * for reading DDC/EDID information is 0b1010001.
-	 * For an I2C send operation, the LSB must be programmed to 0;
-	 * for I2C receive operation, the LSB must be programmed to 1. */
-
-	{
-		value = 0;
-
-		set_reg_field_value(
-			value,
-			false,
-			DC_I2C_DATA,
-			DC_I2C_DATA_RW);
-
-		set_reg_field_value(
-			value,
-			request->address,
-			DC_I2C_DATA,
-			DC_I2C_DATA);
-
-		if (engine->transaction_count == 0) {
-			set_reg_field_value(
-				value,
-				0,
-				DC_I2C_DATA,
-				DC_I2C_INDEX);
-
-			/*enable index write*/
-			set_reg_field_value(
-				value,
-				1,
-				DC_I2C_DATA,
-				DC_I2C_INDEX_WRITE);
-		}
-
-		dm_write_reg(ctx, mmDC_I2C_DATA, value);
-
-		if (!(request->action & I2CAUX_TRANSACTION_ACTION_I2C_READ)) {
-
-			set_reg_field_value(
-				value,
-				0,
-				DC_I2C_DATA,
-				DC_I2C_INDEX_WRITE);
-
-			while (length) {
-
-				set_reg_field_value(
-					value,
-					*buffer++,
-					DC_I2C_DATA,
-					DC_I2C_DATA);
-
-				dm_write_reg(ctx, mmDC_I2C_DATA, value);
-				--length;
-			}
-		}
-	}
-
-	++engine->transaction_count;
-	engine->buffer_used_bytes += length + 1;
-
-	return last_transaction;
-}
-
-static void execute_transaction(
-	struct i2c_hw_engine_dce80 *engine)
-{
-	uint32_t value = 0;
-	struct dc_context *ctx = NULL;
-
-	ctx = engine->base.base.base.ctx;
-
-	{
-		const uint32_t addr = engine->addr.DC_I2C_DDCX_SETUP;
-
-		value = dm_read_reg(ctx, addr);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_DATA_DRIVE_EN);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_CLK_DRIVE_EN);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_DATA_DRIVE_SEL);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_INTRA_TRANSACTION_DELAY);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_DDC1_SETUP,
-			DC_I2C_DDC1_INTRA_BYTE_DELAY);
-
-		dm_write_reg(ctx, addr, value);
-	}
-
-	{
-		const uint32_t addr = mmDC_I2C_CONTROL;
-
-		value = dm_read_reg(ctx, addr);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SOFT_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SW_STATUS_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_SEND_RESET);
-
-		set_reg_field_value(
-			value,
-			0,
-			DC_I2C_CONTROL,
-			DC_I2C_GO);
-
-		set_reg_field_value(
-			value,
-			engine->transaction_count - 1,
-			DC_I2C_CONTROL,
-			DC_I2C_TRANSACTION_COUNT);
-
-		dm_write_reg(ctx, addr, value);
-	}
-
-	/* start I2C transfer */
-	{
-		const uint32_t addr = mmDC_I2C_CONTROL;
-
-		value	= dm_read_reg(ctx, addr);
-
-		set_reg_field_value(
-			value,
-			1,
-			DC_I2C_CONTROL,
-			DC_I2C_GO);
-
-		dm_write_reg(ctx, addr, value);
-	}
-
-	/* all transactions were executed and HW buffer became empty
-	 * (even though it actually happens when status becomes DONE) */
-	engine->transaction_count = 0;
-	engine->buffer_used_bytes = 0;
-}
-
-static void submit_channel_request(
-	struct i2c_engine *engine,
-	struct i2c_request_transaction_data *request)
-{
-	request->status = I2C_CHANNEL_OPERATION_SUCCEEDED;
-
-	if (!process_transaction(FROM_I2C_ENGINE(engine), request))
-		return;
-
-	if (is_hw_busy(&engine->base)) {
-		request->status = I2C_CHANNEL_OPERATION_ENGINE_BUSY;
-		return;
-	}
-
-	execute_transaction(FROM_I2C_ENGINE(engine));
-}
-
-static void process_channel_reply(
-	struct i2c_engine *engine,
-	struct i2c_reply_transaction_data *reply)
-{
-	uint32_t length = reply->length;
-	uint8_t *buffer = reply->data;
-
-	uint32_t value = 0;
-
-	/*set index*/
-	set_reg_field_value(
-		value,
-		length - 1,
-		DC_I2C_DATA,
-		DC_I2C_INDEX);
-
-	set_reg_field_value(
-		value,
-		1,
-		DC_I2C_DATA,
-		DC_I2C_DATA_RW);
-
-	set_reg_field_value(
-		value,
-		1,
-		DC_I2C_DATA,
-		DC_I2C_INDEX_WRITE);
-
-	dm_write_reg(engine->base.ctx, mmDC_I2C_DATA, value);
-
-	while (length) {
-		/* after reading the status,
-		 * if the I2C operation executed successfully
-		 * (i.e. DC_I2C_STATUS_DONE = 1) then the I2C controller
-		 * should read data bytes from I2C circular data buffer */
-
-		value = dm_read_reg(engine->base.ctx, mmDC_I2C_DATA);
-
-		*buffer++ = get_reg_field_value(
-				value,
-				DC_I2C_DATA,
-				DC_I2C_DATA);
-
-		--length;
-	}
-}
-
-static enum i2c_channel_operation_result get_channel_status(
-	struct i2c_engine *engine,
-	uint8_t *returned_bytes)
-{
-	uint32_t i2c_sw_status = 0;
-	uint32_t value = dm_read_reg(engine->base.ctx, mmDC_I2C_SW_STATUS);
-
-	i2c_sw_status = get_reg_field_value(
-			value,
-			DC_I2C_SW_STATUS,
-			DC_I2C_SW_STATUS);
-
-	if (i2c_sw_status == DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_SW)
-		return I2C_CHANNEL_OPERATION_ENGINE_BUSY;
-	else if (value & DC_I2C_SW_STATUS__DC_I2C_SW_STOPPED_ON_NACK_MASK)
-		return I2C_CHANNEL_OPERATION_NO_RESPONSE;
-	else if (value & DC_I2C_SW_STATUS__DC_I2C_SW_TIMEOUT_MASK)
-		return I2C_CHANNEL_OPERATION_TIMEOUT;
-	else if (value & DC_I2C_SW_STATUS__DC_I2C_SW_ABORTED_MASK)
-		return I2C_CHANNEL_OPERATION_FAILED;
-	else if (value & DC_I2C_SW_STATUS__DC_I2C_SW_DONE_MASK)
-		return I2C_CHANNEL_OPERATION_SUCCEEDED;
-
-	/*
-	 * this is the case when HW used for communication, I2C_SW_STATUS
-	 * could be zero
-	 */
-	return I2C_CHANNEL_OPERATION_SUCCEEDED;
-}
-
-static uint32_t get_hw_buffer_available_size(
-	const struct i2c_hw_engine *engine)
-{
-	return I2C_HW_BUFFER_SIZE -
-		FROM_I2C_HW_ENGINE(engine)->buffer_used_bytes;
-}
-
-static uint32_t get_transaction_timeout(
-	const struct i2c_hw_engine *engine,
-	uint32_t length)
-{
-	uint32_t speed = engine->base.funcs->get_speed(&engine->base);
-
-	uint32_t period_timeout;
-	uint32_t num_of_clock_stretches;
-
-	if (!speed)
-		return 0;
-
-	period_timeout = (1000 * TRANSACTION_TIMEOUT_IN_I2C_CLOCKS) / speed;
-
-	num_of_clock_stretches = 1 + (length << 3) + 1;
-	num_of_clock_stretches +=
-		(FROM_I2C_HW_ENGINE(engine)->buffer_used_bytes << 3) +
-		(FROM_I2C_HW_ENGINE(engine)->transaction_count << 1);
-
-	return period_timeout * num_of_clock_stretches;
-}
-
-/*
- * @brief
- * DC_I2C_DDC1_SETUP MM register offsets
- *
- * @note
- * The indices of this offset array are DDC engine IDs
- */
-static const int32_t ddc_setup_offset[] = {
-
-	mmDC_I2C_DDC1_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 1 */
-	mmDC_I2C_DDC2_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 2 */
-	mmDC_I2C_DDC3_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 3 */
-	mmDC_I2C_DDC4_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 4 */
-	mmDC_I2C_DDC5_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 5 */
-	mmDC_I2C_DDC6_SETUP - mmDC_I2C_DDC1_SETUP, /* DDC Engine 6 */
-	mmDC_I2C_DDCVGA_SETUP - mmDC_I2C_DDC1_SETUP /* DDC Engine 7 */
-};
-
-/*
- * @brief
- * DC_I2C_DDC1_SPEED MM register offsets
- *
- * @note
- * The indices of this offset array are DDC engine IDs
- */
-static const int32_t ddc_speed_offset[] = {
-	mmDC_I2C_DDC1_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 1 */
-	mmDC_I2C_DDC2_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 2 */
-	mmDC_I2C_DDC3_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 3 */
-	mmDC_I2C_DDC4_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 4 */
-	mmDC_I2C_DDC5_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 5 */
-	mmDC_I2C_DDC6_SPEED - mmDC_I2C_DDC1_SPEED, /* DDC Engine 6 */
-	mmDC_I2C_DDCVGA_SPEED - mmDC_I2C_DDC1_SPEED /* DDC Engine 7 */
-};
-
-static const struct i2c_engine_funcs i2c_engine_funcs = {
-	.destroy = destroy,
-	.get_speed = get_speed,
-	.set_speed = set_speed,
-	.setup_engine = setup_engine,
-	.submit_channel_request = submit_channel_request,
-	.process_channel_reply = process_channel_reply,
-	.get_channel_status = get_channel_status,
-	.acquire_engine = dal_i2c_hw_engine_acquire_engine,
-};
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.get_engine_type = dal_i2c_hw_engine_get_engine_type,
-	.acquire = dal_i2c_engine_acquire,
-	.submit_request = dal_i2c_hw_engine_submit_request,
-};
-
-static const struct i2c_hw_engine_funcs i2c_hw_engine_funcs = {
-	.get_hw_buffer_available_size =
-		get_hw_buffer_available_size,
-	.get_transaction_timeout =
-		get_transaction_timeout,
-	.wait_on_operation_result =
-		dal_i2c_hw_engine_wait_on_operation_result,
-};
-
-static void construct(
-	struct i2c_hw_engine_dce80 *engine,
-	const struct i2c_hw_engine_dce80_create_arg *arg)
-{
-	dal_i2c_hw_engine_construct(&engine->base, arg->ctx);
-
-	engine->base.base.base.funcs = &engine_funcs;
-	engine->base.base.funcs = &i2c_engine_funcs;
-	engine->base.funcs = &i2c_hw_engine_funcs;
-	engine->base.default_speed = arg->default_speed;
-	engine->addr.DC_I2C_DDCX_SETUP =
-		mmDC_I2C_DDC1_SETUP + ddc_setup_offset[arg->engine_id];
-	engine->addr.DC_I2C_DDCX_SPEED =
-		mmDC_I2C_DDC1_SPEED + ddc_speed_offset[arg->engine_id];
-
-	engine->engine_id = arg->engine_id;
-	engine->reference_frequency = arg->reference_frequency;
-	engine->buffer_used_bytes = 0;
-	engine->transaction_count = 0;
-	engine->engine_keep_power_up_count = 1;
-}
-
-struct i2c_engine *dal_i2c_hw_engine_dce80_create(
-	const struct i2c_hw_engine_dce80_create_arg *arg)
-{
-	struct i2c_hw_engine_dce80 *engine;
-
-	if (!arg) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	if ((arg->engine_id >= sizeof(ddc_setup_offset) / sizeof(int32_t)) ||
-	    (arg->engine_id >= sizeof(ddc_speed_offset) / sizeof(int32_t)) ||
-	    !arg->reference_frequency) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	engine = kzalloc(sizeof(struct i2c_hw_engine_dce80), GFP_KERNEL);
-
-	if (!engine) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	construct(engine, arg);
-	return &engine->base.base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.h b/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.h
deleted file mode 100644
index 5c6116fb5479..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_hw_engine_dce80.h
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_HW_ENGINE_DCE80_H__
-#define __DAL_I2C_HW_ENGINE_DCE80_H__
-
-struct i2c_hw_engine_dce80 {
-	struct i2c_hw_engine base;
-	struct {
-		uint32_t DC_I2C_DDCX_SETUP;
-		uint32_t DC_I2C_DDCX_SPEED;
-	} addr;
-	uint32_t engine_id;
-	/* expressed in kilohertz */
-	uint32_t reference_frequency;
-	/* number of bytes currently used in HW buffer */
-	uint32_t buffer_used_bytes;
-	/* number of pending transactions (before GO) */
-	uint32_t transaction_count;
-	uint32_t engine_keep_power_up_count;
-};
-
-struct i2c_hw_engine_dce80_create_arg {
-	uint32_t engine_id;
-	uint32_t reference_frequency;
-	uint32_t default_speed;
-	struct dc_context *ctx;
-};
-
-struct i2c_engine *dal_i2c_hw_engine_dce80_create(
-	const struct i2c_hw_engine_dce80_create_arg *arg);
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_sw_engine_dce80.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_sw_engine_dce80.c
deleted file mode 100644
index 4853ee26096a..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_sw_engine_dce80.c
+++ /dev/null
@@ -1,173 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_sw_engine_dce80.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-#include "dce/dce_8_0_d.h"
-#include "dce/dce_8_0_sh_mask.h"
-
-/*
- * This unit
- */
-
-static const uint32_t ddc_hw_status_addr[] = {
-	mmDC_I2C_DDC1_HW_STATUS,
-	mmDC_I2C_DDC2_HW_STATUS,
-	mmDC_I2C_DDC3_HW_STATUS,
-	mmDC_I2C_DDC4_HW_STATUS,
-	mmDC_I2C_DDC5_HW_STATUS,
-	mmDC_I2C_DDC6_HW_STATUS,
-	mmDC_I2C_DDCVGA_HW_STATUS
-};
-
-/*
- * @brief
- * Cast 'struct i2c_sw_engine *'
- * to 'struct i2c_sw_engine_dce80 *'
- */
-#define FROM_I2C_SW_ENGINE(ptr) \
-	container_of((ptr), struct i2c_sw_engine_dce80, base)
-
-/*
- * @brief
- * Cast 'struct i2c_engine *'
- * to 'struct i2c_sw_engine_dce80 *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	FROM_I2C_SW_ENGINE(container_of((ptr), struct i2c_sw_engine, base))
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct i2c_sw_engine_dce80 *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-static void release_engine(
-	struct engine *engine)
-{
-
-}
-
-static void destruct(
-	struct i2c_sw_engine_dce80 *engine)
-{
-	dal_i2c_sw_engine_destruct(&engine->base);
-}
-
-static void destroy(
-	struct i2c_engine **engine)
-{
-	struct i2c_sw_engine_dce80 *sw_engine = FROM_I2C_ENGINE(*engine);
-
-	destruct(sw_engine);
-
-	kfree(sw_engine);
-
-	*engine = NULL;
-}
-
-static bool acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc_handle)
-{
-	return dal_i2caux_i2c_sw_engine_acquire_engine(engine, ddc_handle);
-}
-
-static const struct i2c_engine_funcs i2c_engine_funcs = {
-	.acquire_engine = acquire_engine,
-	.destroy = destroy,
-	.get_speed = dal_i2c_sw_engine_get_speed,
-	.set_speed = dal_i2c_sw_engine_set_speed,
-	.setup_engine = dal_i2c_engine_setup_i2c_engine,
-	.submit_channel_request = dal_i2c_sw_engine_submit_channel_request,
-	.process_channel_reply = dal_i2c_engine_process_channel_reply,
-	.get_channel_status = dal_i2c_sw_engine_get_channel_status,
-};
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.get_engine_type = dal_i2c_sw_engine_get_engine_type,
-	.acquire = dal_i2c_engine_acquire,
-	.submit_request = dal_i2c_sw_engine_submit_request,
-};
-
-static void construct(
-	struct i2c_sw_engine_dce80 *engine,
-	const struct i2c_sw_engine_dce80_create_arg *arg)
-{
-	struct i2c_sw_engine_create_arg arg_base;
-
-	arg_base.ctx = arg->ctx;
-	arg_base.default_speed = arg->default_speed;
-
-	dal_i2c_sw_engine_construct(&engine->base, &arg_base);
-
-	engine->base.base.base.funcs = &engine_funcs;
-	engine->base.base.funcs = &i2c_engine_funcs;
-	engine->base.default_speed = arg->default_speed;
-	engine->engine_id = arg->engine_id;
-}
-
-struct i2c_engine *dal_i2c_sw_engine_dce80_create(
-	const struct i2c_sw_engine_dce80_create_arg *arg)
-{
-	struct i2c_sw_engine_dce80 *engine;
-
-	if (!arg) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	engine = kzalloc(sizeof(struct i2c_sw_engine_dce80), GFP_KERNEL);
-
-	if (!engine) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	construct(engine, arg);
-	return &engine->base.base;
-}
-
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2caux_dce80.c b/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2caux_dce80.c
deleted file mode 100644
index ed48596dd2a5..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2caux_dce80.c
+++ /dev/null
@@ -1,284 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2caux_dce80.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "i2c_sw_engine_dce80.h"
-#include "../i2c_hw_engine.h"
-#include "i2c_hw_engine_dce80.h"
-#include "../i2c_generic_hw_engine.h"
-#include "../aux_engine.h"
-
-
-#include "../dce110/aux_engine_dce110.h"
-#include "../dce110/i2caux_dce110.h"
-
-#include "dce/dce_8_0_d.h"
-#include "dce/dce_8_0_sh_mask.h"
-
-
-/* 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 aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = 0 \
-}
-
-static const struct dce110_aux_registers dce80_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5)
-};
-
-/*
- * This unit
- */
-
-#define FROM_I2C_AUX(ptr) \
-	container_of((ptr), struct i2caux_dce80, base)
-
-static void destruct(
-	struct i2caux_dce80 *i2caux_dce80)
-{
-	dal_i2caux_destruct(&i2caux_dce80->base);
-}
-
-static void destroy(
-	struct i2caux **i2c_engine)
-{
-	struct i2caux_dce80 *i2caux_dce80 = FROM_I2C_AUX(*i2c_engine);
-
-	destruct(i2caux_dce80);
-
-	kfree(i2caux_dce80);
-
-	*i2c_engine = NULL;
-}
-
-static struct i2c_engine *acquire_i2c_hw_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc)
-{
-	struct i2caux_dce80 *i2caux_dce80 = FROM_I2C_AUX(i2caux);
-
-	struct i2c_engine *engine = NULL;
-	bool non_generic;
-
-	if (!ddc)
-		return NULL;
-
-	if (ddc->hw_info.hw_supported) {
-		enum gpio_ddc_line line = dal_ddc_get_line(ddc);
-
-		if (line < GPIO_DDC_LINE_COUNT) {
-			non_generic = true;
-			engine = i2caux->i2c_hw_engines[line];
-		}
-	}
-
-	if (!engine) {
-		non_generic = false;
-		engine = i2caux->i2c_generic_hw_engine;
-	}
-
-	if (!engine)
-		return NULL;
-
-	if (non_generic) {
-		if (!i2caux_dce80->i2c_hw_buffer_in_use &&
-			engine->base.funcs->acquire(&engine->base, ddc)) {
-			i2caux_dce80->i2c_hw_buffer_in_use = true;
-			return engine;
-		}
-	} else {
-		if (engine->base.funcs->acquire(&engine->base, ddc))
-			return engine;
-	}
-
-	return NULL;
-}
-
-static void release_engine(
-	struct i2caux *i2caux,
-	struct engine *engine)
-{
-	if (engine->funcs->get_engine_type(engine) ==
-		I2CAUX_ENGINE_TYPE_I2C_DDC_HW)
-		FROM_I2C_AUX(i2caux)->i2c_hw_buffer_in_use = false;
-
-	dal_i2caux_release_engine(i2caux, engine);
-}
-
-static const enum gpio_ddc_line hw_ddc_lines[] = {
-	GPIO_DDC_LINE_DDC1,
-	GPIO_DDC_LINE_DDC2,
-	GPIO_DDC_LINE_DDC3,
-	GPIO_DDC_LINE_DDC4,
-	GPIO_DDC_LINE_DDC5,
-	GPIO_DDC_LINE_DDC6,
-	GPIO_DDC_LINE_DDC_VGA
-};
-
-static const enum gpio_ddc_line hw_aux_lines[] = {
-	GPIO_DDC_LINE_DDC1,
-	GPIO_DDC_LINE_DDC2,
-	GPIO_DDC_LINE_DDC3,
-	GPIO_DDC_LINE_DDC4,
-	GPIO_DDC_LINE_DDC5,
-	GPIO_DDC_LINE_DDC6
-};
-
-static const struct i2caux_funcs i2caux_funcs = {
-	.destroy = destroy,
-	.acquire_i2c_hw_engine = acquire_i2c_hw_engine,
-	.release_engine = release_engine,
-	.acquire_i2c_sw_engine = dal_i2caux_acquire_i2c_sw_engine,
-	.acquire_aux_engine = dal_i2caux_acquire_aux_engine,
-};
-
-static void construct(
-	struct i2caux_dce80 *i2caux_dce80,
-	struct dc_context *ctx)
-{
-	/* Entire family have I2C engine reference clock frequency
-	 * changed from XTALIN (27) to XTALIN/2 (13.5) */
-
-	struct i2caux *base = &i2caux_dce80->base;
-
-	uint32_t reference_frequency =
-		dal_i2caux_get_reference_clock(ctx->dc_bios) >> 1;
-
-	/*bool use_i2c_sw_engine = dal_adapter_service_is_feature_supported(as,
-		FEATURE_RESTORE_USAGE_I2C_SW_ENGINE);*/
-
-	/* Use SWI2C for dce8 currently, sicne we have bug with hwi2c */
-	bool use_i2c_sw_engine = true;
-
-	uint32_t i;
-
-	dal_i2caux_construct(base, ctx);
-
-	i2caux_dce80->base.funcs = &i2caux_funcs;
-	i2caux_dce80->i2c_hw_buffer_in_use = false;
-
-	/* Create I2C HW engines (HW + SW pairs)
-	 * for all lines which has assisted HW DDC
-	 * 'i' (loop counter) used as DDC/AUX engine_id */
-
-	i = 0;
-
-	do {
-		enum gpio_ddc_line line_id = hw_ddc_lines[i];
-
-		struct i2c_hw_engine_dce80_create_arg hw_arg;
-
-		if (use_i2c_sw_engine) {
-			struct i2c_sw_engine_dce80_create_arg sw_arg;
-
-			sw_arg.engine_id = i;
-			sw_arg.default_speed = base->default_i2c_sw_speed;
-			sw_arg.ctx = ctx;
-			base->i2c_sw_engines[line_id] =
-				dal_i2c_sw_engine_dce80_create(&sw_arg);
-		}
-
-		hw_arg.engine_id = i;
-		hw_arg.reference_frequency = reference_frequency;
-		hw_arg.default_speed = base->default_i2c_hw_speed;
-		hw_arg.ctx = ctx;
-
-		base->i2c_hw_engines[line_id] =
-			dal_i2c_hw_engine_dce80_create(&hw_arg);
-
-		++i;
-	} while (i < ARRAY_SIZE(hw_ddc_lines));
-
-	/* Create AUX engines for all lines which has assisted HW AUX
-	 * 'i' (loop counter) used as DDC/AUX engine_id */
-
-	i = 0;
-
-	do {
-		enum gpio_ddc_line line_id = hw_aux_lines[i];
-
-		struct aux_engine_dce110_init_data arg;
-
-		arg.engine_id = i;
-		arg.timeout_period = base->aux_timeout_period;
-		arg.ctx = ctx;
-		arg.regs = &dce80_aux_regs[i];
-
-		base->aux_engines[line_id] =
-			dal_aux_engine_dce110_create(&arg);
-
-		++i;
-	} while (i < ARRAY_SIZE(hw_aux_lines));
-
-	/* TODO Generic I2C SW and HW */
-}
-
-struct i2caux *dal_i2caux_dce80_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce80 *i2caux_dce80 =
-		kzalloc(sizeof(struct i2caux_dce80), GFP_KERNEL);
-
-	if (!i2caux_dce80) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	construct(i2caux_dce80, ctx);
-	return &i2caux_dce80->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.c b/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.c
deleted file mode 100644
index a59c1f50c1e8..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.c
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-#include "../dce110/aux_engine_dce110.h"
-#include "../dce110/i2c_hw_engine_dce110.h"
-#include "../dce110/i2caux_dce110.h"
-
-#include "dcn/dcn_1_0_offset.h"
-#include "dcn/dcn_1_0_sh_mask.h"
-#include "soc15_hw_ip.h"
-#include "vega10_ip_offset.h"
-
-/* begin *********************
- * macros to expend register list macro defined in HW object header file */
-
-#define BASE_INNER(seg) \
-	DCE_BASE__INST0_SEG ## seg
-
-/* compile time expand base address. */
-#define BASE(seg) \
-	BASE_INNER(seg)
-
-#define SR(reg_name)\
-		.reg_name = BASE(mm ## reg_name ## _BASE_IDX) +  \
-					mm ## reg_name
-
-#define SRI(reg_name, block, id)\
-	.reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
-					mm ## block ## id ## _ ## reg_name
-/* macros to expend register list macro defined in HW object header file
- * end *********************/
-
-#define aux_regs(id)\
-[id] = {\
-	AUX_COMMON_REG_LIST(id), \
-	.AUX_RESET_MASK = DP_AUX0_AUX_CONTROL__AUX_RESET_MASK \
-}
-
-#define hw_engine_regs(id)\
-{\
-		I2C_HW_ENGINE_COMMON_REG_LIST(id) \
-}
-
-static const struct dce110_aux_registers dcn10_aux_regs[] = {
-		aux_regs(0),
-		aux_regs(1),
-		aux_regs(2),
-		aux_regs(3),
-		aux_regs(4),
-		aux_regs(5),
-};
-
-static const struct dce110_i2c_hw_engine_registers dcn10_hw_engine_regs[] = {
-		hw_engine_regs(1),
-		hw_engine_regs(2),
-		hw_engine_regs(3),
-		hw_engine_regs(4),
-		hw_engine_regs(5),
-		hw_engine_regs(6)
-};
-
-static const struct dce110_i2c_hw_engine_shift i2c_shift = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(__SHIFT)
-};
-
-static const struct dce110_i2c_hw_engine_mask i2c_mask = {
-		I2C_COMMON_MASK_SH_LIST_DCE110(_MASK)
-};
-
-struct i2caux *dal_i2caux_dcn10_create(
-	struct dc_context *ctx)
-{
-	struct i2caux_dce110 *i2caux_dce110 =
-		kzalloc(sizeof(struct i2caux_dce110), GFP_KERNEL);
-
-	if (!i2caux_dce110) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	dal_i2caux_dce110_construct(i2caux_dce110,
-				    ctx,
-				    ARRAY_SIZE(dcn10_aux_regs),
-				    dcn10_aux_regs,
-				    dcn10_hw_engine_regs,
-				    &i2c_shift,
-				    &i2c_mask);
-	return &i2caux_dce110->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.h b/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.h
deleted file mode 100644
index aeb4a86463d4..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dcn10/i2caux_dcn10.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_AUX_DCN10_H__
-#define __DAL_I2C_AUX_DCN10_H__
-
-struct i2caux *dal_i2caux_dcn10_create(
-	struct dc_context *ctx);
-
-#endif /* __DAL_I2C_AUX_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.c b/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.c
deleted file mode 100644
index e6408f644086..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.c
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
- * Copyright 2012-16 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "../i2caux.h"
-#include "../engine.h"
-#include "../i2c_engine.h"
-#include "../i2c_sw_engine.h"
-#include "../i2c_hw_engine.h"
-
-/*
- * Header of this unit
- */
-#include "i2caux_diag.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-static void destruct(
-	struct i2caux *i2caux)
-{
-	dal_i2caux_destruct(i2caux);
-}
-
-static void destroy(
-	struct i2caux **i2c_engine)
-{
-	destruct(*i2c_engine);
-
-	kfree(*i2c_engine);
-
-	*i2c_engine = NULL;
-}
-
-/* function table */
-static const struct i2caux_funcs i2caux_funcs = {
-	.destroy = destroy,
-	.acquire_i2c_hw_engine = NULL,
-	.release_engine = NULL,
-	.acquire_i2c_sw_engine = NULL,
-	.acquire_aux_engine = NULL,
-};
-
-static void construct(
-	struct i2caux *i2caux,
-	struct dc_context *ctx)
-{
-	dal_i2caux_construct(i2caux, ctx);
-	i2caux->funcs = &i2caux_funcs;
-}
-
-struct i2caux *dal_i2caux_diag_fpga_create(
-	struct dc_context *ctx)
-{
-	struct i2caux *i2caux =	kzalloc(sizeof(struct i2caux),
-					       GFP_KERNEL);
-
-	if (!i2caux) {
-		ASSERT_CRITICAL(false);
-		return NULL;
-	}
-
-	construct(i2caux, ctx);
-	return i2caux;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.h b/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.h
deleted file mode 100644
index a83eeb748283..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/diagnostics/i2caux_diag.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * Copyright 2012-16 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 __DAL_I2C_AUX_DIAG_FPGA_H__
-#define __DAL_I2C_AUX_DIAG_FPGA_H__
-
-struct i2caux *dal_i2caux_diag_fpga_create(
-	struct dc_context *ctx);
-
-#endif /* __DAL_I2C_AUX_DIAG_FPGA_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/engine.h
deleted file mode 100644
index b16fb1ff687d..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/engine.h
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_ENGINE_H__
-#define __DAL_ENGINE_H__
-
-#include "dc_ddc_types.h"
-
-enum i2caux_transaction_operation {
-	I2CAUX_TRANSACTION_READ,
-	I2CAUX_TRANSACTION_WRITE
-};
-
-enum i2caux_transaction_address_space {
-	I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C = 1,
-	I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD
-};
-
-struct i2caux_transaction_payload {
-	enum i2caux_transaction_address_space address_space;
-	uint32_t address;
-	uint32_t length;
-	uint8_t *data;
-};
-
-enum i2caux_transaction_status {
-	I2CAUX_TRANSACTION_STATUS_UNKNOWN = (-1L),
-	I2CAUX_TRANSACTION_STATUS_SUCCEEDED,
-	I2CAUX_TRANSACTION_STATUS_FAILED_CHANNEL_BUSY,
-	I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT,
-	I2CAUX_TRANSACTION_STATUS_FAILED_PROTOCOL_ERROR,
-	I2CAUX_TRANSACTION_STATUS_FAILED_NACK,
-	I2CAUX_TRANSACTION_STATUS_FAILED_INCOMPLETE,
-	I2CAUX_TRANSACTION_STATUS_FAILED_OPERATION,
-	I2CAUX_TRANSACTION_STATUS_FAILED_INVALID_OPERATION,
-	I2CAUX_TRANSACTION_STATUS_FAILED_BUFFER_OVERFLOW,
-	I2CAUX_TRANSACTION_STATUS_FAILED_HPD_DISCON
-};
-
-struct i2caux_transaction_request {
-	enum i2caux_transaction_operation operation;
-	struct i2caux_transaction_payload payload;
-	enum i2caux_transaction_status status;
-};
-
-enum i2caux_engine_type {
-	I2CAUX_ENGINE_TYPE_UNKNOWN = (-1L),
-	I2CAUX_ENGINE_TYPE_AUX,
-	I2CAUX_ENGINE_TYPE_I2C_DDC_HW,
-	I2CAUX_ENGINE_TYPE_I2C_GENERIC_HW,
-	I2CAUX_ENGINE_TYPE_I2C_SW
-};
-
-enum i2c_default_speed {
-	I2CAUX_DEFAULT_I2C_HW_SPEED = 50,
-	I2CAUX_DEFAULT_I2C_SW_SPEED = 50
-};
-
-struct engine;
-
-struct engine_funcs {
-	enum i2caux_engine_type (*get_engine_type)(
-		const struct engine *engine);
-	bool (*acquire)(
-		struct engine *engine,
-		struct ddc *ddc);
-	bool (*submit_request)(
-		struct engine *engine,
-		struct i2caux_transaction_request *request,
-		bool middle_of_transaction);
-	void (*release_engine)(
-		struct engine *engine);
-};
-
-struct engine {
-	const struct engine_funcs *funcs;
-	uint32_t inst;
-	struct ddc *ddc;
-	struct dc_context *ctx;
-};
-
-void dal_i2caux_construct_engine(
-	struct engine *engine,
-	struct dc_context *ctx);
-
-void dal_i2caux_destruct_engine(
-	struct engine *engine);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.c
deleted file mode 100644
index 70e20bd47ce4..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.c
+++ /dev/null
@@ -1,118 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_engine.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-#define FROM_ENGINE(ptr) \
-	container_of((ptr), struct i2c_engine, base)
-
-bool dal_i2c_engine_acquire(
-	struct engine *engine,
-	struct ddc *ddc_handle)
-{
-	struct i2c_engine *i2c_engine = FROM_ENGINE(engine);
-
-	uint32_t counter = 0;
-	bool result;
-
-	do {
-		result = i2c_engine->funcs->acquire_engine(
-			i2c_engine, ddc_handle);
-
-		if (result)
-			break;
-
-		/* i2c_engine is busy by VBios, lets wait and retry */
-
-		udelay(10);
-
-		++counter;
-	} while (counter < 2);
-
-	if (result) {
-		if (!i2c_engine->funcs->setup_engine(i2c_engine)) {
-			engine->funcs->release_engine(engine);
-			result = false;
-		}
-	}
-
-	return result;
-}
-
-bool dal_i2c_engine_setup_i2c_engine(
-	struct i2c_engine *engine)
-{
-	/* Derivative classes do not have to override this */
-
-	return true;
-}
-
-void dal_i2c_engine_submit_channel_request(
-	struct i2c_engine *engine,
-	struct i2c_request_transaction_data *request)
-{
-
-}
-
-void dal_i2c_engine_process_channel_reply(
-	struct i2c_engine *engine,
-	struct i2c_reply_transaction_data *reply)
-{
-
-}
-
-void dal_i2c_engine_construct(
-	struct i2c_engine *engine,
-	struct dc_context *ctx)
-{
-	dal_i2caux_construct_engine(&engine->base, ctx);
-	engine->timeout_delay = 0;
-}
-
-void dal_i2c_engine_destruct(
-	struct i2c_engine *engine)
-{
-	dal_i2caux_destruct_engine(&engine->base);
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.h
deleted file mode 100644
index ded6ea34b714..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_engine.h
+++ /dev/null
@@ -1,115 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_ENGINE_H__
-#define __DAL_I2C_ENGINE_H__
-
-enum i2c_channel_operation_result {
-	I2C_CHANNEL_OPERATION_SUCCEEDED,
-	I2C_CHANNEL_OPERATION_FAILED,
-	I2C_CHANNEL_OPERATION_NOT_GRANTED,
-	I2C_CHANNEL_OPERATION_IS_BUSY,
-	I2C_CHANNEL_OPERATION_NO_HANDLE_PROVIDED,
-	I2C_CHANNEL_OPERATION_CHANNEL_IN_USE,
-	I2C_CHANNEL_OPERATION_CHANNEL_CLIENT_MAX_ALLOWED,
-	I2C_CHANNEL_OPERATION_ENGINE_BUSY,
-	I2C_CHANNEL_OPERATION_TIMEOUT,
-	I2C_CHANNEL_OPERATION_NO_RESPONSE,
-	I2C_CHANNEL_OPERATION_HW_REQUEST_I2C_BUS,
-	I2C_CHANNEL_OPERATION_WRONG_PARAMETER,
-	I2C_CHANNEL_OPERATION_OUT_NB_OF_RETRIES,
-	I2C_CHANNEL_OPERATION_NOT_STARTED
-};
-
-struct i2c_request_transaction_data {
-	enum i2caux_transaction_action action;
-	enum i2c_channel_operation_result status;
-	uint8_t address;
-	uint32_t length;
-	uint8_t *data;
-};
-
-struct i2c_reply_transaction_data {
-	uint32_t length;
-	uint8_t *data;
-};
-
-struct i2c_engine;
-
-struct i2c_engine_funcs {
-	void (*destroy)(
-		struct i2c_engine **ptr);
-	uint32_t (*get_speed)(
-		const struct i2c_engine *engine);
-	void (*set_speed)(
-		struct i2c_engine *engine,
-		uint32_t speed);
-	bool (*acquire_engine)(
-		struct i2c_engine *engine,
-		struct ddc *ddc);
-	bool (*setup_engine)(
-		struct i2c_engine *engine);
-	void (*submit_channel_request)(
-		struct i2c_engine *engine,
-		struct i2c_request_transaction_data *request);
-	void (*process_channel_reply)(
-		struct i2c_engine *engine,
-		struct i2c_reply_transaction_data *reply);
-	enum i2c_channel_operation_result (*get_channel_status)(
-		struct i2c_engine *engine,
-		uint8_t *returned_bytes);
-};
-
-struct i2c_engine {
-	struct engine base;
-	const struct i2c_engine_funcs *funcs;
-	uint32_t timeout_delay;
-	uint32_t setup_limit;
-	uint32_t send_reset_length;
-};
-
-void dal_i2c_engine_construct(
-	struct i2c_engine *engine,
-	struct dc_context *ctx);
-
-void dal_i2c_engine_destruct(
-	struct i2c_engine *engine);
-
-bool dal_i2c_engine_setup_i2c_engine(
-	struct i2c_engine *engine);
-
-void dal_i2c_engine_submit_channel_request(
-	struct i2c_engine *engine,
-	struct i2c_request_transaction_data *request);
-
-void dal_i2c_engine_process_channel_reply(
-	struct i2c_engine *engine,
-	struct i2c_reply_transaction_data *reply);
-
-bool dal_i2c_engine_acquire(
-	struct engine *ptr,
-	struct ddc *ddc_handle);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.c
deleted file mode 100644
index 5a4295e0fae5..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.c
+++ /dev/null
@@ -1,284 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "engine.h"
-#include "i2c_engine.h"
-#include "i2c_hw_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_generic_hw_engine.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-/*
- * @brief
- * Cast 'struct i2c_hw_engine *'
- * to 'struct i2c_generic_hw_engine *'
- */
-#define FROM_I2C_HW_ENGINE(ptr) \
-	container_of((ptr), struct i2c_generic_hw_engine, base)
-
-/*
- * @brief
- * Cast 'struct i2c_engine *'
- * to 'struct i2c_generic_hw_engine *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	FROM_I2C_HW_ENGINE(container_of((ptr), struct i2c_hw_engine, base))
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct i2c_generic_hw_engine *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-enum i2caux_engine_type dal_i2c_generic_hw_engine_get_engine_type(
-	const struct engine *engine)
-{
-	return I2CAUX_ENGINE_TYPE_I2C_GENERIC_HW;
-}
-
-/*
- * @brief
- * Single transaction handling.
- * Since transaction may be bigger than HW buffer size,
- * it divides transaction to sub-transactions
- * and uses batch transaction feature of the engine.
- */
-bool dal_i2c_generic_hw_engine_submit_request(
-	struct engine *engine,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction)
-{
-	struct i2c_generic_hw_engine *hw_engine = FROM_ENGINE(engine);
-
-	struct i2c_hw_engine *base = &hw_engine->base;
-
-	uint32_t max_payload_size =
-		base->funcs->get_hw_buffer_available_size(base);
-
-	bool initial_stop_bit = !middle_of_transaction;
-
-	struct i2c_generic_transaction_attributes attributes;
-
-	enum i2c_channel_operation_result operation_result =
-		I2C_CHANNEL_OPERATION_FAILED;
-
-	bool result = false;
-
-	/* setup transaction initial properties */
-
-	uint8_t address = i2caux_request->payload.address;
-	uint8_t *current_payload = i2caux_request->payload.data;
-	uint32_t remaining_payload_size = i2caux_request->payload.length;
-
-	bool first_iteration = true;
-
-	if (i2caux_request->operation == I2CAUX_TRANSACTION_READ)
-		attributes.action = I2CAUX_TRANSACTION_ACTION_I2C_READ;
-	else if (i2caux_request->operation == I2CAUX_TRANSACTION_WRITE)
-		attributes.action = I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-	else {
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_INVALID_OPERATION;
-		return false;
-	}
-
-	/* Do batch transaction.
-	 * Divide read/write data into payloads which fit HW buffer size.
-	 * 1. Single transaction:
-	 *    start_bit = 1, stop_bit depends on session state, ack_on_read = 0;
-	 * 2. Start of batch transaction:
-	 *    start_bit = 1, stop_bit = 0, ack_on_read = 1;
-	 * 3. Middle of batch transaction:
-	 *    start_bit = 0, stop_bit = 0, ack_on_read = 1;
-	 * 4. End of batch transaction:
-	 *    start_bit = 0, stop_bit depends on session state, ack_on_read = 0.
-	 * Session stop bit is set if 'middle_of_transaction' = 0. */
-
-	while (remaining_payload_size) {
-		uint32_t current_transaction_size;
-		uint32_t current_payload_size;
-
-		bool last_iteration;
-		bool stop_bit;
-
-		/* Calculate current transaction size and payload size.
-		 * Transaction size = total number of bytes in transaction,
-		 * including slave's address;
-		 * Payload size = number of data bytes in transaction. */
-
-		if (first_iteration) {
-			/* In the first sub-transaction we send slave's address
-			 * thus we need to reserve one byte for it */
-			current_transaction_size =
-			(remaining_payload_size > max_payload_size - 1) ?
-				max_payload_size :
-				remaining_payload_size + 1;
-
-			current_payload_size = current_transaction_size - 1;
-		} else {
-			/* Second and further sub-transactions will have
-			 * entire buffer reserved for data */
-			current_transaction_size =
-				(remaining_payload_size > max_payload_size) ?
-				max_payload_size :
-				remaining_payload_size;
-
-			current_payload_size = current_transaction_size;
-		}
-
-		last_iteration =
-			(remaining_payload_size == current_payload_size);
-
-		stop_bit = last_iteration ? initial_stop_bit : false;
-
-		/* write slave device address */
-
-		if (first_iteration)
-			hw_engine->funcs->write_address(hw_engine, address);
-
-		/* write current portion of data, if requested */
-
-		if (i2caux_request->operation == I2CAUX_TRANSACTION_WRITE)
-			hw_engine->funcs->write_data(
-				hw_engine,
-				current_payload,
-				current_payload_size);
-
-		/* execute transaction */
-
-		attributes.start_bit = first_iteration;
-		attributes.stop_bit = stop_bit;
-		attributes.last_read = last_iteration;
-		attributes.transaction_size = current_transaction_size;
-
-		hw_engine->funcs->execute_transaction(hw_engine, &attributes);
-
-		/* wait until transaction is processed; if it fails - quit */
-
-		operation_result = base->funcs->wait_on_operation_result(
-			base,
-			base->funcs->get_transaction_timeout(
-				base, current_transaction_size),
-			I2C_CHANNEL_OPERATION_ENGINE_BUSY);
-
-		if (operation_result != I2C_CHANNEL_OPERATION_SUCCEEDED)
-			break;
-
-		/* read current portion of data, if requested */
-
-		/* the read offset should be 1 for first sub-transaction,
-		 * and 0 for any next one */
-
-		if (i2caux_request->operation == I2CAUX_TRANSACTION_READ)
-			hw_engine->funcs->read_data(hw_engine, current_payload,
-				current_payload_size, first_iteration ? 1 : 0);
-
-		/* update loop variables */
-
-		first_iteration = false;
-		current_payload += current_payload_size;
-		remaining_payload_size -= current_payload_size;
-	}
-
-	/* update transaction status */
-
-	switch (operation_result) {
-	case I2C_CHANNEL_OPERATION_SUCCEEDED:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-		result = true;
-	break;
-	case I2C_CHANNEL_OPERATION_NO_RESPONSE:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-	break;
-	case I2C_CHANNEL_OPERATION_TIMEOUT:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-	break;
-	case I2C_CHANNEL_OPERATION_FAILED:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_INCOMPLETE;
-	break;
-	default:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_OPERATION;
-	}
-
-	return result;
-}
-
-/*
- * @brief
- * Returns number of microseconds to wait until timeout to be considered
- */
-uint32_t dal_i2c_generic_hw_engine_get_transaction_timeout(
-	const struct i2c_hw_engine *engine,
-	uint32_t length)
-{
-	const struct i2c_engine *base = &engine->base;
-
-	uint32_t speed = base->funcs->get_speed(base);
-
-	if (!speed)
-		return 0;
-
-	/* total timeout = period_timeout * (start + data bits count + stop) */
-
-	return ((1000 * TRANSACTION_TIMEOUT_IN_I2C_CLOCKS) / speed) *
-		(1 + (length << 3) + 1);
-}
-
-void dal_i2c_generic_hw_engine_construct(
-	struct i2c_generic_hw_engine *engine,
-	struct dc_context *ctx)
-{
-	dal_i2c_hw_engine_construct(&engine->base, ctx);
-}
-
-void dal_i2c_generic_hw_engine_destruct(
-	struct i2c_generic_hw_engine *engine)
-{
-	dal_i2c_hw_engine_destruct(&engine->base);
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.h
deleted file mode 100644
index 1da0397b04a2..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_generic_hw_engine.h
+++ /dev/null
@@ -1,77 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_GENERIC_HW_ENGINE_H__
-#define __DAL_I2C_GENERIC_HW_ENGINE_H__
-
-struct i2c_generic_transaction_attributes {
-	enum i2caux_transaction_action action;
-	uint32_t transaction_size;
-	bool start_bit;
-	bool stop_bit;
-	bool last_read;
-};
-
-struct i2c_generic_hw_engine;
-
-struct i2c_generic_hw_engine_funcs {
-	void (*write_address)(
-		struct i2c_generic_hw_engine *engine,
-		uint8_t address);
-	void (*write_data)(
-		struct i2c_generic_hw_engine *engine,
-		const uint8_t *buffer,
-		uint32_t length);
-	void (*read_data)(
-		struct i2c_generic_hw_engine *engine,
-		uint8_t *buffer,
-		uint32_t length,
-		uint32_t offset);
-	void (*execute_transaction)(
-		struct i2c_generic_hw_engine *engine,
-		struct i2c_generic_transaction_attributes *attributes);
-};
-
-struct i2c_generic_hw_engine {
-	struct i2c_hw_engine base;
-	const struct i2c_generic_hw_engine_funcs *funcs;
-};
-
-void dal_i2c_generic_hw_engine_construct(
-	struct i2c_generic_hw_engine *engine,
-	struct dc_context *ctx);
-
-void dal_i2c_generic_hw_engine_destruct(
-	struct i2c_generic_hw_engine *engine);
-enum i2caux_engine_type dal_i2c_generic_hw_engine_get_engine_type(
-	const struct engine *engine);
-bool dal_i2c_generic_hw_engine_submit_request(
-	struct engine *ptr,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction);
-uint32_t dal_i2c_generic_hw_engine_get_transaction_timeout(
-	const struct i2c_hw_engine *engine,
-	uint32_t length);
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.c
deleted file mode 100644
index 141898533e8e..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.c
+++ /dev/null
@@ -1,251 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-#include "dm_event_log.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "engine.h"
-#include "i2c_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_hw_engine.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-/*
- * @brief
- * Cast 'struct i2c_engine *'
- * to 'struct i2c_hw_engine *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	container_of((ptr), struct i2c_hw_engine, base)
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct i2c_hw_engine *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-enum i2caux_engine_type dal_i2c_hw_engine_get_engine_type(
-	const struct engine *engine)
-{
-	return I2CAUX_ENGINE_TYPE_I2C_DDC_HW;
-}
-
-bool dal_i2c_hw_engine_submit_request(
-	struct engine *engine,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction)
-{
-	struct i2c_hw_engine *hw_engine = FROM_ENGINE(engine);
-
-	struct i2c_request_transaction_data request;
-
-	uint32_t transaction_timeout;
-
-	enum i2c_channel_operation_result operation_result;
-
-	bool result = false;
-
-	/* We need following:
-	 * transaction length will not exceed
-	 * the number of free bytes in HW buffer (minus one for address)*/
-
-	if (i2caux_request->payload.length >=
-		hw_engine->funcs->get_hw_buffer_available_size(hw_engine)) {
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_BUFFER_OVERFLOW;
-		return false;
-	}
-
-	if (i2caux_request->operation == I2CAUX_TRANSACTION_READ)
-		request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_READ;
-	else if (i2caux_request->operation == I2CAUX_TRANSACTION_WRITE)
-		request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-	else {
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_INVALID_OPERATION;
-		/* [anaumov] in DAL2, there was no "return false" */
-		return false;
-	}
-
-	request.address = (uint8_t)i2caux_request->payload.address;
-	request.length = i2caux_request->payload.length;
-	request.data = i2caux_request->payload.data;
-
-	/* obtain timeout value before submitting request */
-
-	transaction_timeout = hw_engine->funcs->get_transaction_timeout(
-		hw_engine, i2caux_request->payload.length + 1);
-
-	hw_engine->base.funcs->submit_channel_request(
-		&hw_engine->base, &request);
-	/* EVENT_LOG_AUX_REQ(engine->ddc->pin_data->en, EVENT_LOG_AUX_ORIGIN_I2C, */
-	/* request.action, request.address, request.length, request.data); */
-
-	if ((request.status == I2C_CHANNEL_OPERATION_FAILED) ||
-		(request.status == I2C_CHANNEL_OPERATION_ENGINE_BUSY)) {
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_CHANNEL_BUSY;
-		return false;
-	}
-
-	/* wait until transaction proceed */
-
-	operation_result = hw_engine->funcs->wait_on_operation_result(
-		hw_engine,
-		transaction_timeout,
-		I2C_CHANNEL_OPERATION_ENGINE_BUSY);
-
-	/* update transaction status */
-
-	switch (operation_result) {
-	case I2C_CHANNEL_OPERATION_SUCCEEDED:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-		result = true;
-	break;
-	case I2C_CHANNEL_OPERATION_NO_RESPONSE:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-	break;
-	case I2C_CHANNEL_OPERATION_TIMEOUT:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-	break;
-	case I2C_CHANNEL_OPERATION_FAILED:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_INCOMPLETE;
-	break;
-	default:
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_OPERATION;
-	}
-
-	if (result && (i2caux_request->operation == I2CAUX_TRANSACTION_READ)) {
-		struct i2c_reply_transaction_data reply;
-
-		reply.data = i2caux_request->payload.data;
-		reply.length = i2caux_request->payload.length;
-
-		hw_engine->base.funcs->
-			process_channel_reply(&hw_engine->base, &reply);
-		/* EVENT_LOG_AUX_REP(engine->ddc->pin_data->en, EVENT_LOG_AUX_ORIGIN_I2C, */
-		/* AUX_TRANSACTION_REPLY_I2C_ACK, reply.length, reply.data); */
-	}
-
-
-
-	return result;
-}
-
-bool dal_i2c_hw_engine_acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc)
-{
-	enum gpio_result result;
-	uint32_t current_speed;
-
-	result = dal_ddc_open(ddc, GPIO_MODE_HARDWARE,
-		GPIO_DDC_CONFIG_TYPE_MODE_I2C);
-
-	if (result != GPIO_RESULT_OK)
-		return false;
-
-	engine->base.ddc = ddc;
-
-	current_speed = engine->funcs->get_speed(engine);
-
-	if (current_speed)
-		FROM_I2C_ENGINE(engine)->original_speed = current_speed;
-
-	return true;
-}
-/*
- * @brief
- * Queries in a loop for current engine status
- * until retrieved status matches 'expected_result', or timeout occurs.
- * Timeout given in microseconds
- * and the status query frequency is also one per microsecond.
- */
-enum i2c_channel_operation_result dal_i2c_hw_engine_wait_on_operation_result(
-	struct i2c_hw_engine *engine,
-	uint32_t timeout,
-	enum i2c_channel_operation_result expected_result)
-{
-	enum i2c_channel_operation_result result;
-	uint32_t i = 0;
-
-	if (!timeout)
-		return I2C_CHANNEL_OPERATION_SUCCEEDED;
-
-	do {
-		result = engine->base.funcs->get_channel_status(
-			&engine->base, NULL);
-
-		if (result != expected_result)
-			break;
-
-		udelay(1);
-
-		++i;
-	} while (i < timeout);
-
-	return result;
-}
-
-void dal_i2c_hw_engine_construct(
-	struct i2c_hw_engine *engine,
-	struct dc_context *ctx)
-{
-	dal_i2c_engine_construct(&engine->base, ctx);
-	engine->original_speed = I2CAUX_DEFAULT_I2C_HW_SPEED;
-	engine->default_speed = I2CAUX_DEFAULT_I2C_HW_SPEED;
-}
-
-void dal_i2c_hw_engine_destruct(
-	struct i2c_hw_engine *engine)
-{
-	dal_i2c_engine_destruct(&engine->base);
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.h
deleted file mode 100644
index 8936a994804a..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_hw_engine.h
+++ /dev/null
@@ -1,80 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_HW_ENGINE_H__
-#define __DAL_I2C_HW_ENGINE_H__
-
-enum {
-	TRANSACTION_TIMEOUT_IN_I2C_CLOCKS = 32
-};
-
-struct i2c_hw_engine;
-
-struct i2c_hw_engine_funcs {
-	uint32_t (*get_hw_buffer_available_size)(
-		const struct i2c_hw_engine *engine);
-	enum i2c_channel_operation_result (*wait_on_operation_result)(
-		struct i2c_hw_engine *engine,
-		uint32_t timeout,
-		enum i2c_channel_operation_result expected_result);
-	uint32_t (*get_transaction_timeout)(
-		const struct i2c_hw_engine *engine,
-		uint32_t length);
-};
-
-struct i2c_hw_engine {
-	struct i2c_engine base;
-	const struct i2c_hw_engine_funcs *funcs;
-
-	/* Values below are in kilohertz */
-	uint32_t original_speed;
-	uint32_t default_speed;
-};
-
-void dal_i2c_hw_engine_construct(
-	struct i2c_hw_engine *engine,
-	struct dc_context *ctx);
-
-void dal_i2c_hw_engine_destruct(
-	struct i2c_hw_engine *engine);
-
-enum i2c_channel_operation_result dal_i2c_hw_engine_wait_on_operation_result(
-	struct i2c_hw_engine *engine,
-	uint32_t timeout,
-	enum i2c_channel_operation_result expected_result);
-
-bool dal_i2c_hw_engine_acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc);
-
-bool dal_i2c_hw_engine_submit_request(
-	struct engine *ptr,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction);
-
-enum i2caux_engine_type dal_i2c_hw_engine_get_engine_type(
-	const struct engine *engine);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.c
deleted file mode 100644
index 8e19bb629394..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.c
+++ /dev/null
@@ -1,601 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "engine.h"
-#include "i2c_engine.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2c_sw_engine.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-/*
- * This unit
- */
-
-#define SCL false
-#define SDA true
-
-static inline bool read_bit_from_ddc(
-	struct ddc *ddc,
-	bool data_nor_clock)
-{
-	uint32_t value = 0;
-
-	if (data_nor_clock)
-		dal_gpio_get_value(ddc->pin_data, &value);
-	else
-		dal_gpio_get_value(ddc->pin_clock, &value);
-
-	return (value != 0);
-}
-
-static inline void write_bit_to_ddc(
-	struct ddc *ddc,
-	bool data_nor_clock,
-	bool bit)
-{
-	uint32_t value = bit ? 1 : 0;
-
-	if (data_nor_clock)
-		dal_gpio_set_value(ddc->pin_data, value);
-	else
-		dal_gpio_set_value(ddc->pin_clock, value);
-}
-
-static bool wait_for_scl_high(
-	struct dc_context *ctx,
-	struct ddc *ddc,
-	uint16_t clock_delay_div_4)
-{
-	uint32_t scl_retry = 0;
-	uint32_t scl_retry_max = I2C_SW_TIMEOUT_DELAY / clock_delay_div_4;
-
-	udelay(clock_delay_div_4);
-
-	/* 3 milliseconds delay
-	 * to wake up some displays from "low power" state.
-	 */
-
-	do {
-		if (read_bit_from_ddc(ddc, SCL))
-			return true;
-
-		udelay(clock_delay_div_4);
-
-		++scl_retry;
-	} while (scl_retry <= scl_retry_max);
-
-	return false;
-}
-
-static bool start_sync(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4)
-{
-	uint32_t retry = 0;
-
-	/* The I2C communications start signal is:
-	 * the SDA going low from high, while the SCL is high. */
-
-	write_bit_to_ddc(ddc_handle, SCL, true);
-
-	udelay(clock_delay_div_4);
-
-	do {
-		write_bit_to_ddc(ddc_handle, SDA, true);
-
-		if (!read_bit_from_ddc(ddc_handle, SDA)) {
-			++retry;
-			continue;
-		}
-
-		udelay(clock_delay_div_4);
-
-		write_bit_to_ddc(ddc_handle, SCL, true);
-
-		if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-			break;
-
-		write_bit_to_ddc(ddc_handle, SDA, false);
-
-		udelay(clock_delay_div_4);
-
-		write_bit_to_ddc(ddc_handle, SCL, false);
-
-		udelay(clock_delay_div_4);
-
-		return true;
-	} while (retry <= I2C_SW_RETRIES);
-
-	return false;
-}
-
-static bool stop_sync(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4)
-{
-	uint32_t retry = 0;
-
-	/* The I2C communications stop signal is:
-	 * the SDA going high from low, while the SCL is high. */
-
-	write_bit_to_ddc(ddc_handle, SCL, false);
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SDA, false);
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SCL, true);
-
-	if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-		return false;
-
-	write_bit_to_ddc(ddc_handle, SDA, true);
-
-	do {
-		udelay(clock_delay_div_4);
-
-		if (read_bit_from_ddc(ddc_handle, SDA))
-			return true;
-
-		++retry;
-	} while (retry <= 2);
-
-	return false;
-}
-
-static bool write_byte(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4,
-	uint8_t byte)
-{
-	int32_t shift = 7;
-	bool ack;
-
-	/* bits are transmitted serially, starting from MSB */
-
-	do {
-		udelay(clock_delay_div_4);
-
-		write_bit_to_ddc(ddc_handle, SDA, (byte >> shift) & 1);
-
-		udelay(clock_delay_div_4);
-
-		write_bit_to_ddc(ddc_handle, SCL, true);
-
-		if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-			return false;
-
-		write_bit_to_ddc(ddc_handle, SCL, false);
-
-		--shift;
-	} while (shift >= 0);
-
-	/* The display sends ACK by preventing the SDA from going high
-	 * after the SCL pulse we use to send our last data bit.
-	 * If the SDA goes high after that bit, it's a NACK */
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SDA, true);
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SCL, true);
-
-	if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-		return false;
-
-	/* read ACK bit */
-
-	ack = !read_bit_from_ddc(ddc_handle, SDA);
-
-	udelay(clock_delay_div_4 << 1);
-
-	write_bit_to_ddc(ddc_handle, SCL, false);
-
-	udelay(clock_delay_div_4 << 1);
-
-	return ack;
-}
-
-static bool read_byte(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4,
-	uint8_t *byte,
-	bool more)
-{
-	int32_t shift = 7;
-
-	uint8_t data = 0;
-
-	/* The data bits are read from MSB to LSB;
-	 * bit is read while SCL is high */
-
-	do {
-		write_bit_to_ddc(ddc_handle, SCL, true);
-
-		if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-			return false;
-
-		if (read_bit_from_ddc(ddc_handle, SDA))
-			data |= (1 << shift);
-
-		write_bit_to_ddc(ddc_handle, SCL, false);
-
-		udelay(clock_delay_div_4 << 1);
-
-		--shift;
-	} while (shift >= 0);
-
-	/* read only whole byte */
-
-	*byte = data;
-
-	udelay(clock_delay_div_4);
-
-	/* send the acknowledge bit:
-	 * SDA low means ACK, SDA high means NACK */
-
-	write_bit_to_ddc(ddc_handle, SDA, !more);
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SCL, true);
-
-	if (!wait_for_scl_high(ctx, ddc_handle, clock_delay_div_4))
-		return false;
-
-	write_bit_to_ddc(ddc_handle, SCL, false);
-
-	udelay(clock_delay_div_4);
-
-	write_bit_to_ddc(ddc_handle, SDA, true);
-
-	udelay(clock_delay_div_4);
-
-	return true;
-}
-
-static bool i2c_write(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4,
-	uint8_t address,
-	uint32_t length,
-	const uint8_t *data)
-{
-	uint32_t i = 0;
-
-	if (!write_byte(ctx, ddc_handle, clock_delay_div_4, address))
-		return false;
-
-	while (i < length) {
-		if (!write_byte(ctx, ddc_handle, clock_delay_div_4, data[i]))
-			return false;
-		++i;
-	}
-
-	return true;
-}
-
-static bool i2c_read(
-	struct dc_context *ctx,
-	struct ddc *ddc_handle,
-	uint16_t clock_delay_div_4,
-	uint8_t address,
-	uint32_t length,
-	uint8_t *data)
-{
-	uint32_t i = 0;
-
-	if (!write_byte(ctx, ddc_handle, clock_delay_div_4, address))
-		return false;
-
-	while (i < length) {
-		if (!read_byte(ctx, ddc_handle, clock_delay_div_4, data + i,
-			i < length - 1))
-			return false;
-		++i;
-	}
-
-	return true;
-}
-
-/*
- * @brief
- * Cast 'struct i2c_engine *'
- * to 'struct i2c_sw_engine *'
- */
-#define FROM_I2C_ENGINE(ptr) \
-	container_of((ptr), struct i2c_sw_engine, base)
-
-/*
- * @brief
- * Cast 'struct engine *'
- * to 'struct i2c_sw_engine *'
- */
-#define FROM_ENGINE(ptr) \
-	FROM_I2C_ENGINE(container_of((ptr), struct i2c_engine, base))
-
-enum i2caux_engine_type dal_i2c_sw_engine_get_engine_type(
-	const struct engine *engine)
-{
-	return I2CAUX_ENGINE_TYPE_I2C_SW;
-}
-
-bool dal_i2c_sw_engine_submit_request(
-	struct engine *engine,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction)
-{
-	struct i2c_sw_engine *sw_engine = FROM_ENGINE(engine);
-
-	struct i2c_engine *base = &sw_engine->base;
-
-	struct i2c_request_transaction_data request;
-	bool operation_succeeded = false;
-
-	if (i2caux_request->operation == I2CAUX_TRANSACTION_READ)
-		request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_READ;
-	else if (i2caux_request->operation == I2CAUX_TRANSACTION_WRITE)
-		request.action = middle_of_transaction ?
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT :
-			I2CAUX_TRANSACTION_ACTION_I2C_WRITE;
-	else {
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_INVALID_OPERATION;
-		/* in DAL2, there was no "return false" */
-		return false;
-	}
-
-	request.address = (uint8_t)i2caux_request->payload.address;
-	request.length = i2caux_request->payload.length;
-	request.data = i2caux_request->payload.data;
-
-	base->funcs->submit_channel_request(base, &request);
-
-	if ((request.status == I2C_CHANNEL_OPERATION_ENGINE_BUSY) ||
-		(request.status == I2C_CHANNEL_OPERATION_FAILED))
-		i2caux_request->status =
-			I2CAUX_TRANSACTION_STATUS_FAILED_CHANNEL_BUSY;
-	else {
-		enum i2c_channel_operation_result operation_result;
-
-		do {
-			operation_result =
-				base->funcs->get_channel_status(base, NULL);
-
-			switch (operation_result) {
-			case I2C_CHANNEL_OPERATION_SUCCEEDED:
-				i2caux_request->status =
-					I2CAUX_TRANSACTION_STATUS_SUCCEEDED;
-				operation_succeeded = true;
-			break;
-			case I2C_CHANNEL_OPERATION_NO_RESPONSE:
-				i2caux_request->status =
-					I2CAUX_TRANSACTION_STATUS_FAILED_NACK;
-			break;
-			case I2C_CHANNEL_OPERATION_TIMEOUT:
-				i2caux_request->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_TIMEOUT;
-			break;
-			case I2C_CHANNEL_OPERATION_FAILED:
-				i2caux_request->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_INCOMPLETE;
-			break;
-			default:
-				i2caux_request->status =
-				I2CAUX_TRANSACTION_STATUS_FAILED_OPERATION;
-			break;
-			}
-		} while (operation_result == I2C_CHANNEL_OPERATION_ENGINE_BUSY);
-	}
-
-	return operation_succeeded;
-}
-
-uint32_t dal_i2c_sw_engine_get_speed(
-	const struct i2c_engine *engine)
-{
-	return FROM_I2C_ENGINE(engine)->speed;
-}
-
-void dal_i2c_sw_engine_set_speed(
-	struct i2c_engine *engine,
-	uint32_t speed)
-{
-	struct i2c_sw_engine *sw_engine = FROM_I2C_ENGINE(engine);
-
-	ASSERT(speed);
-
-	sw_engine->speed = speed ? speed : I2CAUX_DEFAULT_I2C_SW_SPEED;
-
-	sw_engine->clock_delay = 1000 / sw_engine->speed;
-
-	if (sw_engine->clock_delay < 12)
-		sw_engine->clock_delay = 12;
-}
-
-bool dal_i2caux_i2c_sw_engine_acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc)
-{
-	enum gpio_result result;
-
-	result = dal_ddc_open(ddc, GPIO_MODE_FAST_OUTPUT,
-		GPIO_DDC_CONFIG_TYPE_MODE_I2C);
-
-	if (result != GPIO_RESULT_OK)
-		return false;
-
-	engine->base.ddc = ddc;
-
-	return true;
-}
-
-void dal_i2c_sw_engine_submit_channel_request(
-	struct i2c_engine *engine,
-	struct i2c_request_transaction_data *req)
-{
-	struct i2c_sw_engine *sw_engine = FROM_I2C_ENGINE(engine);
-
-	struct ddc *ddc = engine->base.ddc;
-	uint16_t clock_delay_div_4 = sw_engine->clock_delay >> 2;
-
-	/* send sync (start / repeated start) */
-
-	bool result = start_sync(engine->base.ctx, ddc, clock_delay_div_4);
-
-	/* process payload */
-
-	if (result) {
-		switch (req->action) {
-		case I2CAUX_TRANSACTION_ACTION_I2C_WRITE:
-		case I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT:
-			result = i2c_write(engine->base.ctx, ddc, clock_delay_div_4,
-				req->address, req->length, req->data);
-		break;
-		case I2CAUX_TRANSACTION_ACTION_I2C_READ:
-		case I2CAUX_TRANSACTION_ACTION_I2C_READ_MOT:
-			result = i2c_read(engine->base.ctx, ddc, clock_delay_div_4,
-				req->address, req->length, req->data);
-		break;
-		default:
-			result = false;
-		break;
-		}
-	}
-
-	/* send stop if not 'mot' or operation failed */
-
-	if (!result ||
-		(req->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE) ||
-		(req->action == I2CAUX_TRANSACTION_ACTION_I2C_READ))
-		if (!stop_sync(engine->base.ctx, ddc, clock_delay_div_4))
-			result = false;
-
-	req->status = result ?
-		I2C_CHANNEL_OPERATION_SUCCEEDED :
-		I2C_CHANNEL_OPERATION_FAILED;
-}
-
-enum i2c_channel_operation_result dal_i2c_sw_engine_get_channel_status(
-	struct i2c_engine *engine,
-	uint8_t *returned_bytes)
-{
-	/* No arbitration with VBIOS is performed since DCE 6.0 */
-	return I2C_CHANNEL_OPERATION_SUCCEEDED;
-}
-
-void dal_i2c_sw_engine_destruct(
-	struct i2c_sw_engine *engine)
-{
-	dal_i2c_engine_destruct(&engine->base);
-}
-
-static void destroy(
-	struct i2c_engine **ptr)
-{
-	dal_i2c_sw_engine_destruct(FROM_I2C_ENGINE(*ptr));
-
-	kfree(*ptr);
-	*ptr = NULL;
-}
-
-static const struct i2c_engine_funcs i2c_engine_funcs = {
-	.acquire_engine = dal_i2caux_i2c_sw_engine_acquire_engine,
-	.destroy = destroy,
-	.get_speed = dal_i2c_sw_engine_get_speed,
-	.set_speed = dal_i2c_sw_engine_set_speed,
-	.setup_engine = dal_i2c_engine_setup_i2c_engine,
-	.submit_channel_request = dal_i2c_sw_engine_submit_channel_request,
-	.process_channel_reply = dal_i2c_engine_process_channel_reply,
-	.get_channel_status = dal_i2c_sw_engine_get_channel_status,
-};
-
-static void release_engine(
-	struct engine *engine)
-{
-
-}
-
-static const struct engine_funcs engine_funcs = {
-	.release_engine = release_engine,
-	.get_engine_type = dal_i2c_sw_engine_get_engine_type,
-	.acquire = dal_i2c_engine_acquire,
-	.submit_request = dal_i2c_sw_engine_submit_request,
-};
-
-void dal_i2c_sw_engine_construct(
-	struct i2c_sw_engine *engine,
-	const struct i2c_sw_engine_create_arg *arg)
-{
-	dal_i2c_engine_construct(&engine->base, arg->ctx);
-	dal_i2c_sw_engine_set_speed(&engine->base, arg->default_speed);
-	engine->base.funcs = &i2c_engine_funcs;
-	engine->base.base.funcs = &engine_funcs;
-}
-
-struct i2c_engine *dal_i2c_sw_engine_create(
-	const struct i2c_sw_engine_create_arg *arg)
-{
-	struct i2c_sw_engine *engine;
-
-	if (!arg) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	engine = kzalloc(sizeof(struct i2c_sw_engine), GFP_KERNEL);
-
-	if (!engine) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-
-	dal_i2c_sw_engine_construct(engine, arg);
-	return &engine->base;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.h b/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.h
deleted file mode 100644
index 546f15b0d3f1..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2c_sw_engine.h
+++ /dev/null
@@ -1,81 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_SW_ENGINE_H__
-#define __DAL_I2C_SW_ENGINE_H__
-
-enum {
-	I2C_SW_RETRIES = 10,
-	I2C_SW_SCL_READ_RETRIES = 128,
-	/* following value is in microseconds */
-	I2C_SW_TIMEOUT_DELAY = 3000
-};
-
-struct i2c_sw_engine;
-
-struct i2c_sw_engine {
-	struct i2c_engine base;
-	uint32_t clock_delay;
-	/* Values below are in KHz */
-	uint32_t speed;
-	uint32_t default_speed;
-};
-
-struct i2c_sw_engine_create_arg {
-	uint32_t default_speed;
-	struct dc_context *ctx;
-};
-
-void dal_i2c_sw_engine_construct(
-	struct i2c_sw_engine *engine,
-	const struct i2c_sw_engine_create_arg *arg);
-
-bool dal_i2caux_i2c_sw_engine_acquire_engine(
-	struct i2c_engine *engine,
-	struct ddc *ddc_handle);
-
-void dal_i2c_sw_engine_destruct(
-	struct i2c_sw_engine *engine);
-
-struct i2c_engine *dal_i2c_sw_engine_create(
-	const struct i2c_sw_engine_create_arg *arg);
-enum i2caux_engine_type dal_i2c_sw_engine_get_engine_type(
-	const struct engine *engine);
-bool dal_i2c_sw_engine_submit_request(
-	struct engine *ptr,
-	struct i2caux_transaction_request *i2caux_request,
-	bool middle_of_transaction);
-uint32_t dal_i2c_sw_engine_get_speed(
-	const struct i2c_engine *engine);
-void dal_i2c_sw_engine_set_speed(
-	struct i2c_engine *ptr,
-	uint32_t speed);
-void dal_i2c_sw_engine_submit_channel_request(
-	struct i2c_engine *ptr,
-	struct i2c_request_transaction_data *req);
-enum i2c_channel_operation_result dal_i2c_sw_engine_get_channel_status(
-	struct i2c_engine *engine,
-	uint8_t *returned_bytes);
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c
deleted file mode 100644
index 1ad6e49102ff..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c
+++ /dev/null
@@ -1,491 +0,0 @@
-/*
- * Copyright 2012-15 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
- *
- */
-
-#include "dm_services.h"
-
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-#include "dc_bios_types.h"
-
-/*
- * Header of this unit
- */
-
-#include "i2caux.h"
-
-/*
- * Post-requisites: headers required by this unit
- */
-
-#include "engine.h"
-#include "i2c_engine.h"
-#include "aux_engine.h"
-
-/*
- * This unit
- */
-
-#include "dce80/i2caux_dce80.h"
-
-#include "dce100/i2caux_dce100.h"
-
-#include "dce110/i2caux_dce110.h"
-
-#include "dce112/i2caux_dce112.h"
-
-#include "dce120/i2caux_dce120.h"
-
-#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
-#include "dcn10/i2caux_dcn10.h"
-#endif
-
-#include "diagnostics/i2caux_diag.h"
-
-/*
- * @brief
- * Plain API, available publicly
- */
-
-struct i2caux *dal_i2caux_create(
-	struct dc_context *ctx)
-{
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		return dal_i2caux_diag_fpga_create(ctx);
-	}
-
-	switch (ctx->dce_version) {
-	case DCE_VERSION_8_0:
-	case DCE_VERSION_8_1:
-	case DCE_VERSION_8_3:
-		return dal_i2caux_dce80_create(ctx);
-	case DCE_VERSION_11_2:
-	case DCE_VERSION_11_22:
-		return dal_i2caux_dce112_create(ctx);
-	case DCE_VERSION_11_0:
-		return dal_i2caux_dce110_create(ctx);
-	case DCE_VERSION_10_0:
-		return dal_i2caux_dce100_create(ctx);
-	case DCE_VERSION_12_0:
-	case DCE_VERSION_12_1:
-		return dal_i2caux_dce120_create(ctx);
-#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
-	case DCN_VERSION_1_0:
-		return dal_i2caux_dcn10_create(ctx);
-#endif
-
-#if defined(CONFIG_DRM_AMD_DC_DCN1_01)
-	case DCN_VERSION_1_01:
-		return dal_i2caux_dcn10_create(ctx);
-#endif
-	default:
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
-}
-
-bool dal_i2caux_submit_i2c_command(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	struct i2c_command *cmd)
-{
-	struct i2c_engine *engine;
-	uint8_t index_of_payload = 0;
-	bool result;
-
-	if (!ddc) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	if (!cmd) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	/*
-	 * default will be SW, however there is a feature flag in adapter
-	 * service that determines whether SW i2c_engine will be available or
-	 * not, if sw i2c is not available we will fallback to hw. This feature
-	 * flag is set to not creating sw i2c engine for every dce except dce80
-	 * currently
-	 */
-	switch (cmd->engine) {
-	case I2C_COMMAND_ENGINE_DEFAULT:
-	case I2C_COMMAND_ENGINE_SW:
-		/* try to acquire SW engine first,
-		 * acquire HW engine if SW engine not available */
-		engine = i2caux->funcs->acquire_i2c_sw_engine(i2caux, ddc);
-
-		if (!engine)
-			engine = i2caux->funcs->acquire_i2c_hw_engine(
-				i2caux, ddc);
-	break;
-	case I2C_COMMAND_ENGINE_HW:
-	default:
-		/* try to acquire HW engine first,
-		 * acquire SW engine if HW engine not available */
-		engine = i2caux->funcs->acquire_i2c_hw_engine(i2caux, ddc);
-
-		if (!engine)
-			engine = i2caux->funcs->acquire_i2c_sw_engine(
-				i2caux, ddc);
-	}
-
-	if (!engine)
-		return false;
-
-	engine->funcs->set_speed(engine, cmd->speed);
-
-	result = true;
-
-	while (index_of_payload < cmd->number_of_payloads) {
-		bool mot = (index_of_payload != cmd->number_of_payloads - 1);
-
-		struct i2c_payload *payload = cmd->payloads + index_of_payload;
-
-		struct i2caux_transaction_request request = { 0 };
-
-		request.operation = payload->write ?
-			I2CAUX_TRANSACTION_WRITE :
-			I2CAUX_TRANSACTION_READ;
-
-		request.payload.address_space =
-			I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C;
-		request.payload.address = (payload->address << 1) |
-			!payload->write;
-		request.payload.length = payload->length;
-		request.payload.data = payload->data;
-
-		if (!engine->base.funcs->submit_request(
-			&engine->base, &request, mot)) {
-			result = false;
-			break;
-		}
-
-		++index_of_payload;
-	}
-
-	i2caux->funcs->release_engine(i2caux, &engine->base);
-
-	return result;
-}
-
-bool dal_i2caux_submit_aux_command(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	struct aux_command *cmd)
-{
-	struct aux_engine *engine;
-	uint8_t index_of_payload = 0;
-	bool result;
-	bool mot;
-
-	if (!ddc) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	if (!cmd) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	engine = i2caux->funcs->acquire_aux_engine(i2caux, ddc);
-
-	if (!engine)
-		return false;
-
-	engine->delay = cmd->defer_delay;
-	engine->max_defer_write_retry = cmd->max_defer_write_retry;
-
-	result = true;
-
-	while (index_of_payload < cmd->number_of_payloads) {
-		struct aux_payload *payload = cmd->payloads + index_of_payload;
-		struct i2caux_transaction_request request = { 0 };
-
-		if (cmd->mot == I2C_MOT_UNDEF)
-			mot = (index_of_payload != cmd->number_of_payloads - 1);
-		else
-			mot = (cmd->mot == I2C_MOT_TRUE);
-
-		request.operation = payload->write ?
-			I2CAUX_TRANSACTION_WRITE :
-			I2CAUX_TRANSACTION_READ;
-
-		if (payload->i2c_over_aux) {
-			request.payload.address_space =
-				I2CAUX_TRANSACTION_ADDRESS_SPACE_I2C;
-
-			request.payload.address = (payload->address << 1) |
-				!payload->write;
-		} else {
-			request.payload.address_space =
-				I2CAUX_TRANSACTION_ADDRESS_SPACE_DPCD;
-
-			request.payload.address = payload->address;
-		}
-
-		request.payload.length = payload->length;
-		request.payload.data = payload->data;
-
-		if (!engine->base.funcs->submit_request(
-			&engine->base, &request, mot)) {
-			result = false;
-			break;
-		}
-
-		++index_of_payload;
-	}
-
-	i2caux->funcs->release_engine(i2caux, &engine->base);
-
-	return result;
-}
-
-static bool get_hw_supported_ddc_line(
-	struct ddc *ddc,
-	enum gpio_ddc_line *line)
-{
-	enum gpio_ddc_line line_found;
-
-	*line = GPIO_DDC_LINE_UNKNOWN;
-
-	if (!ddc) {
-		BREAK_TO_DEBUGGER();
-		return false;
-	}
-
-	if (!ddc->hw_info.hw_supported)
-		return false;
-
-	line_found = dal_ddc_get_line(ddc);
-
-	if (line_found >= GPIO_DDC_LINE_COUNT)
-		return false;
-
-	*line = line_found;
-
-	return true;
-}
-
-void dal_i2caux_configure_aux(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	union aux_config cfg)
-{
-	struct aux_engine *engine =
-		i2caux->funcs->acquire_aux_engine(i2caux, ddc);
-
-	if (!engine)
-		return;
-
-	engine->funcs->configure(engine, cfg);
-
-	i2caux->funcs->release_engine(i2caux, &engine->base);
-}
-
-void dal_i2caux_destroy(
-	struct i2caux **i2caux)
-{
-	if (!i2caux || !*i2caux) {
-		BREAK_TO_DEBUGGER();
-		return;
-	}
-
-	(*i2caux)->funcs->destroy(i2caux);
-
-	*i2caux = NULL;
-}
-
-/*
- * @brief
- * An utility function used by 'struct i2caux' and its descendants
- */
-
-uint32_t dal_i2caux_get_reference_clock(
-		struct dc_bios *bios)
-{
-	struct dc_firmware_info info = { { 0 } };
-
-	if (bios->funcs->get_firmware_info(bios, &info) != BP_RESULT_OK)
-		return 0;
-
-	return info.pll_info.crystal_frequency;
-}
-
-/*
- * @brief
- * i2caux
- */
-
-enum {
-	/* following are expressed in KHz */
-	DEFAULT_I2C_SW_SPEED = 50,
-	DEFAULT_I2C_HW_SPEED = 50,
-
-	DEFAULT_I2C_SW_SPEED_100KHZ = 100,
-	DEFAULT_I2C_HW_SPEED_100KHZ = 100,
-
-	/* This is the timeout as defined in DP 1.2a,
-	 * 2.3.4 "Detailed uPacket TX AUX CH State Description". */
-	AUX_TIMEOUT_PERIOD = 400,
-
-	/* Ideally, the SW timeout should be just above 550usec
-	 * which is programmed in HW.
-	 * But the SW timeout of 600usec is not reliable,
-	 * because on some systems, delay_in_microseconds()
-	 * returns faster than it should.
-	 * EPR #379763: by trial-and-error on different systems,
-	 * 700usec is the minimum reliable SW timeout for polling
-	 * the AUX_SW_STATUS.AUX_SW_DONE bit.
-	 * This timeout expires *only* when there is
-	 * AUX Error or AUX Timeout conditions - not during normal operation.
-	 * During normal operation, AUX_SW_STATUS.AUX_SW_DONE bit is set
-	 * at most within ~240usec. That means,
-	 * increasing this timeout will not affect normal operation,
-	 * and we'll timeout after
-	 * SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD = 1600usec.
-	 * This timeout is especially important for
-	 * resume from S3 and CTS. */
-	SW_AUX_TIMEOUT_PERIOD_MULTIPLIER = 4
-};
-
-struct i2c_engine *dal_i2caux_acquire_i2c_sw_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc)
-{
-	enum gpio_ddc_line line;
-	struct i2c_engine *engine = NULL;
-
-	if (get_hw_supported_ddc_line(ddc, &line))
-		engine = i2caux->i2c_sw_engines[line];
-
-	if (!engine)
-		engine = i2caux->i2c_generic_sw_engine;
-
-	if (!engine)
-		return NULL;
-
-	if (!engine->base.funcs->acquire(&engine->base, ddc))
-		return NULL;
-
-	return engine;
-}
-
-struct aux_engine *dal_i2caux_acquire_aux_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc)
-{
-	enum gpio_ddc_line line;
-	struct aux_engine *engine;
-
-	if (!get_hw_supported_ddc_line(ddc, &line))
-		return NULL;
-
-	engine = i2caux->aux_engines[line];
-
-	if (!engine)
-		return NULL;
-
-	if (!engine->base.funcs->acquire(&engine->base, ddc))
-		return NULL;
-
-	return engine;
-}
-
-void dal_i2caux_release_engine(
-	struct i2caux *i2caux,
-	struct engine *engine)
-{
-	engine->funcs->release_engine(engine);
-
-	dal_ddc_close(engine->ddc);
-
-	engine->ddc = NULL;
-}
-
-void dal_i2caux_construct(
-	struct i2caux *i2caux,
-	struct dc_context *ctx)
-{
-	uint32_t i = 0;
-
-	i2caux->ctx = ctx;
-	do {
-		i2caux->i2c_sw_engines[i] = NULL;
-		i2caux->i2c_hw_engines[i] = NULL;
-		i2caux->aux_engines[i] = NULL;
-
-		++i;
-	} while (i < GPIO_DDC_LINE_COUNT);
-
-	i2caux->i2c_generic_sw_engine = NULL;
-	i2caux->i2c_generic_hw_engine = NULL;
-
-	i2caux->aux_timeout_period =
-		SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD;
-
-	if (ctx->dce_version >= DCE_VERSION_11_2) {
-		i2caux->default_i2c_hw_speed = DEFAULT_I2C_HW_SPEED_100KHZ;
-		i2caux->default_i2c_sw_speed = DEFAULT_I2C_SW_SPEED_100KHZ;
-	} else {
-		i2caux->default_i2c_hw_speed = DEFAULT_I2C_HW_SPEED;
-		i2caux->default_i2c_sw_speed = DEFAULT_I2C_SW_SPEED;
-	}
-}
-
-void dal_i2caux_destruct(
-	struct i2caux *i2caux)
-{
-	uint32_t i = 0;
-
-	if (i2caux->i2c_generic_hw_engine)
-		i2caux->i2c_generic_hw_engine->funcs->destroy(
-			&i2caux->i2c_generic_hw_engine);
-
-	if (i2caux->i2c_generic_sw_engine)
-		i2caux->i2c_generic_sw_engine->funcs->destroy(
-			&i2caux->i2c_generic_sw_engine);
-
-	do {
-		if (i2caux->aux_engines[i])
-			i2caux->aux_engines[i]->funcs->destroy(
-				&i2caux->aux_engines[i]);
-
-		if (i2caux->i2c_hw_engines[i])
-			i2caux->i2c_hw_engines[i]->funcs->destroy(
-				&i2caux->i2c_hw_engines[i]);
-
-		if (i2caux->i2c_sw_engines[i])
-			i2caux->i2c_sw_engines[i]->funcs->destroy(
-				&i2caux->i2c_sw_engines[i]);
-
-		++i;
-	} while (i < GPIO_DDC_LINE_COUNT);
-}
-
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.h b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.h
deleted file mode 100644
index 64f51bb06915..000000000000
--- a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.h
+++ /dev/null
@@ -1,122 +0,0 @@
-/*
- * Copyright 2012-15 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 __DAL_I2C_AUX_H__
-#define __DAL_I2C_AUX_H__
-
-uint32_t dal_i2caux_get_reference_clock(
-	struct dc_bios *bios);
-
-struct i2caux;
-
-struct engine;
-
-struct i2caux_funcs {
-	void (*destroy)(struct i2caux **ptr);
-	struct i2c_engine * (*acquire_i2c_sw_engine)(
-		struct i2caux *i2caux,
-		struct ddc *ddc);
-	struct i2c_engine * (*acquire_i2c_hw_engine)(
-		struct i2caux *i2caux,
-		struct ddc *ddc);
-	struct aux_engine * (*acquire_aux_engine)(
-		struct i2caux *i2caux,
-		struct ddc *ddc);
-	void (*release_engine)(
-		struct i2caux *i2caux,
-		struct engine *engine);
-};
-
-struct i2c_engine;
-struct aux_engine;
-
-struct i2caux {
-	struct dc_context *ctx;
-	const struct i2caux_funcs *funcs;
-	/* On ASIC we have certain amount of lines with HW DDC engine
-	 * (4, 6, or maybe more in the future).
-	 * For every such line, we create separate HW DDC engine
-	 * (since we have these engines in HW) and separate SW DDC engine
-	 * (to allow concurrent use of few lines).
-	 * In similar way we have AUX engines. */
-
-	/* I2C SW engines, per DDC line.
-	 * Only lines with HW DDC support will be initialized */
-	struct i2c_engine *i2c_sw_engines[GPIO_DDC_LINE_COUNT];
-
-	/* I2C HW engines, per DDC line.
-	 * Only lines with HW DDC support will be initialized */
-	struct i2c_engine *i2c_hw_engines[GPIO_DDC_LINE_COUNT];
-
-	/* AUX engines, per DDC line.
-	 * Only lines with HW AUX support will be initialized */
-	struct aux_engine *aux_engines[GPIO_DDC_LINE_COUNT];
-
-	/* For all other lines, we can use
-	 * single instance of generic I2C HW engine
-	 * (since in HW, there is single instance of it)
-	 * or single instance of generic I2C SW engine.
-	 * AUX is not supported for other lines. */
-
-	/* General-purpose I2C SW engine.
-	 * Can be assigned dynamically to any line per transaction */
-	struct i2c_engine *i2c_generic_sw_engine;
-
-	/* General-purpose I2C generic HW engine.
-	 * Can be assigned dynamically to almost any line per transaction */
-	struct i2c_engine *i2c_generic_hw_engine;
-
-	/* [anaumov] in DAL2, there is a Mutex */
-
-	uint32_t aux_timeout_period;
-
-	/* expressed in KHz */
-	uint32_t default_i2c_sw_speed;
-	uint32_t default_i2c_hw_speed;
-};
-
-void dal_i2caux_construct(
-	struct i2caux *i2caux,
-	struct dc_context *ctx);
-
-void dal_i2caux_release_engine(
-	struct i2caux *i2caux,
-	struct engine *engine);
-
-void dal_i2caux_destruct(
-	struct i2caux *i2caux);
-
-void dal_i2caux_destroy(
-	struct i2caux **ptr);
-
-struct i2c_engine *dal_i2caux_acquire_i2c_sw_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc);
-
-struct aux_engine *dal_i2caux_acquire_aux_engine(
-	struct i2caux *i2caux,
-	struct ddc *ddc);
-
-#endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/clock_source.h b/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
index 47ef90495376..fe6301cb8681 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
@@ -78,7 +78,7 @@ struct csdp_ref_clk_ds_params {
 };
 
 struct pixel_clk_params {
-	uint32_t requested_pix_clk; /* in KHz */
+	uint32_t requested_pix_clk_100hz;
 /*> Requested Pixel Clock
  * (based on Video Timing standard used for requested mode)*/
 	uint32_t requested_sym_clk; /* in KHz */
@@ -104,9 +104,9 @@ struct pixel_clk_params {
  *  with actually calculated Clock and reference Crystal frequency
  */
 struct pll_settings {
-	uint32_t actual_pix_clk;
-	uint32_t adjusted_pix_clk;
-	uint32_t calculated_pix_clk;
+	uint32_t actual_pix_clk_100hz;
+	uint32_t adjusted_pix_clk_100hz;
+	uint32_t calculated_pix_clk_100hz;
 	uint32_t vco_freq;
 	uint32_t reference_freq;
 	uint32_t reference_divider;
@@ -166,6 +166,10 @@ struct clock_source_funcs {
 			struct clock_source *,
 			struct pixel_clk_params *,
 			struct pll_settings *);
+	bool (*get_pixel_clk_frequency_100hz)(
+			struct clock_source *clock_source,
+			unsigned int inst,
+			unsigned int *pixel_clk_khz);
 };
 
 struct clock_source {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
index 94fc31080fda..2e61a22ef4b2 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
@@ -30,7 +30,7 @@ enum dc_status {
 	DC_OK = 1,
 
 	DC_NO_CONTROLLER_RESOURCE = 2,
-	DC_NO_STREAM_ENG_RESOURCE = 3,
+	DC_NO_STREAM_ENC_RESOURCE = 3,
 	DC_NO_CLOCK_SOURCE_RESOURCE = 4,
 	DC_FAIL_CONTROLLER_VALIDATE = 5,
 	DC_FAIL_ENC_VALIDATE = 6,
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 b168a5e9dd9d..986ed1728644 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -146,7 +146,7 @@ struct resource_pool {
 	struct mpc *mpc;
 	struct pp_smu_funcs_rv *pp_smu;
 	struct pp_smu_display_requirement_rv pp_smu_req;
-	struct aux_engine *engines[MAX_PIPES];
+	struct dce_aux *engines[MAX_PIPES];
 	struct dce_i2c_hw *hw_i2cs[MAX_PIPES];
 	struct dce_i2c_sw *sw_i2cs[MAX_PIPES];
 	bool i2c_hw_buffer_in_use;
@@ -180,13 +180,8 @@ struct resource_pool {
 	const struct resource_caps *res_cap;
 };
 
-struct dcn_fe_clocks {
-	int dppclk_khz;
-};
-
 struct dcn_fe_bandwidth {
-	struct dcn_fe_clocks calc;
-	struct dcn_fe_clocks cur;
+	int dppclk_khz;
 };
 
 struct stream_resource {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
index 538b83303b86..16fd4dc6c4dd 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
@@ -64,13 +64,6 @@ void dal_ddc_i2c_payloads_add(
 		uint8_t *data,
 		bool write);
 
-void dal_ddc_aux_payloads_add(
-		struct aux_payloads *payloads,
-		uint32_t address,
-		uint32_t len,
-		uint8_t *data,
-		bool write);
-
 struct ddc_service_init_data {
 	struct graphics_object_id id;
 	struct dc_context *ctx;
@@ -103,12 +96,10 @@ bool dal_ddc_service_query_ddc_data(
 		uint32_t read_size);
 
 int dc_link_aux_transfer(struct ddc_service *ddc,
-			     unsigned int address,
-			     uint8_t *reply,
-			     void *buffer,
-			     unsigned int size,
-			     enum aux_transaction_type type,
-			     enum i2caux_transaction_action action);
+		struct aux_payload *payload);
+
+bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc,
+		struct aux_payload *payload);
 
 void dal_ddc_service_write_scdc_data(
 		struct ddc_service *ddc_service,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index 02f757dd70d4..9d2d8e51306c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -39,6 +39,18 @@ enum segment_order {
 	segment_order__non_contiguous,
 };
 
+struct dcn_hubbub_wm_set {
+	uint32_t wm_set;
+	uint32_t data_urgent;
+	uint32_t pte_meta_urgent;
+	uint32_t sr_enter;
+	uint32_t sr_exit;
+	uint32_t dram_clk_chanage;
+};
+
+struct dcn_hubbub_wm {
+	struct dcn_hubbub_wm_set sets[4];
+};
 
 struct hubbub_funcs {
 	void (*update_dchub)(
@@ -58,7 +70,14 @@ struct hubbub_funcs {
 	bool (*dcc_support_pixel_format)(
 			enum surface_pixel_format format,
 			unsigned int *bytes_per_element);
+
+	void (*wm_read_state)(struct hubbub *hubbub,
+			struct dcn_hubbub_wm *wm);
 };
 
+struct hubbub {
+	const struct hubbub_funcs *funcs;
+	struct dc_context *ctx;
+};
 
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
index cb85eaa9857f..cbaa43853611 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
@@ -27,16 +27,22 @@
 
 #include "dm_services_types.h"
 
+/* If HW itself ever powered down it will be 0.
+ * fwDmcuInit will write to 1.
+ * Driver will only call MCP init if current state is 1,
+ * and the MCP command will transition this to 2.
+ */
 enum dmcu_state {
-	DMCU_NOT_INITIALIZED = 0,
-	DMCU_RUNNING = 1
+	DMCU_UNLOADED = 0,
+	DMCU_LOADED_UNINITIALIZED = 1,
+	DMCU_RUNNING = 2,
 };
 
 struct dmcu_version {
-	unsigned int date;
-	unsigned int month;
-	unsigned int year;
 	unsigned int interface_version;
+	unsigned int abm_version;
+	unsigned int psr_version;
+	unsigned int build_version;
 };
 
 struct dmcu {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
index e894e649ce5a..fb7967b39edb 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
@@ -39,6 +39,11 @@ struct dpp {
 
 };
 
+struct dpp_input_csc_matrix {
+	enum dc_color_space color_space;
+	uint16_t regval[12];
+};
+
 struct dpp_grph_csc_adjustment {
 	struct fixed31_32 temperature_matrix[CSC_TEMPERATURE_MATRIX_SIZE];
 	enum graphics_gamut_adjust_type gamut_adjust_type;
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
index 04c6989aac58..1cd07e94ee63 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
@@ -78,7 +78,8 @@ struct hubp_funcs {
 	bool (*hubp_program_surface_flip_and_addr)(
 		struct hubp *hubp,
 		const struct dc_plane_address *address,
-		bool flip_immediate);
+		bool flip_immediate,
+		uint8_t vmid);
 
 	void (*hubp_program_pte_vm)(
 		struct hubp *hubp,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
index c20fdcaac53b..c9d3e37e9531 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
@@ -153,6 +153,7 @@ struct link_encoder_funcs {
 	void (*enable_hpd)(struct link_encoder *enc);
 	void (*disable_hpd)(struct link_encoder *enc);
 	bool (*is_dig_enabled)(struct link_encoder *enc);
+	unsigned int (*get_dig_frontend)(struct link_encoder *enc);
 	void (*destroy)(struct link_encoder **enc);
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
index 06df02ddff6a..da89c2edb07c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
@@ -31,7 +31,7 @@
 #include "dml/display_mode_structs.h"
 
 struct dchub_init_data;
-struct cstate_pstate_watermarks_st1 {
+struct cstate_pstate_watermarks_st {
 	uint32_t cstate_exit_ns;
 	uint32_t cstate_enter_plus_exit_ns;
 	uint32_t pstate_change_ns;
@@ -40,7 +40,7 @@ struct cstate_pstate_watermarks_st1 {
 struct dcn_watermarks {
 	uint32_t pte_meta_urgent_ns;
 	uint32_t urgent_ns;
-	struct cstate_pstate_watermarks_st1 cstate_pstate;
+	struct cstate_pstate_watermarks_st cstate_pstate;
 };
 
 struct dcn_watermark_set {
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 53a9b64df11a..4051493557bc 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
@@ -161,6 +161,10 @@ struct stream_encoder_funcs {
 	void (*set_avmute)(
 		struct stream_encoder *enc, bool enable);
 
+	void (*dig_connect_to_otg)(
+		struct stream_encoder *enc,
+		int tg_inst);
+
 };
 
 #endif /* STREAM_ENCODER_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
index af700c7dac50..03ae941895f3 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
@@ -134,15 +134,29 @@ struct dc_crtc_timing;
 
 struct drr_params;
 
+union vline_config;
+
+
+enum vline_select {
+	VLINE0,
+	VLINE1,
+	VLINE2
+};
+
 struct timing_generator_funcs {
 	bool (*validate_timing)(struct timing_generator *tg,
 							const struct dc_crtc_timing *timing);
 	void (*program_timing)(struct timing_generator *tg,
 							const struct dc_crtc_timing *timing,
 							bool use_vbios);
-	void (*program_vline_interrupt)(struct timing_generator *optc,
+	void (*program_vline_interrupt)(
+			struct timing_generator *optc,
 			const struct dc_crtc_timing *dc_crtc_timing,
-			unsigned long long vsync_delta);
+			enum vline_select vline,
+			const union vline_config *vline_config);
+
+	void (*program_vupdate_interrupt)(struct timing_generator *optc,
+			const struct dc_crtc_timing *dc_crtc_timing);
 	bool (*enable_crtc)(struct timing_generator *tg);
 	bool (*disable_crtc)(struct timing_generator *tg);
 	bool (*is_counter_moving)(struct timing_generator *tg);
@@ -159,6 +173,8 @@ struct timing_generator_funcs {
 	bool (*get_otg_active_size)(struct timing_generator *optc,
 			uint32_t *otg_active_width,
 			uint32_t *otg_active_height);
+	bool (*is_matching_timing)(struct timing_generator *tg,
+			const struct dc_crtc_timing *otg_timing);
 	void (*set_early_control)(struct timing_generator *tg,
 							   uint32_t early_cntl);
 	void (*wait_for_state)(struct timing_generator *tg,
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/engine_base.c b/drivers/gpu/drm/amd/display/dc/inc/hw/vmid.h
index 5d155d36d353..037beb0a2a27 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/engine_base.c
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/vmid.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-15 Advanced Micro Devices, Inc.
+ * Copyright 2018 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"),
@@ -23,30 +23,27 @@
  *
  */
 
-#include "dm_services.h"
+#ifndef DAL_DC_INC_HW_VMID_H_
+#define DAL_DC_INC_HW_VMID_H_
 
-/*
- * Pre-requisites: headers required by header of this unit
- */
-#include "include/i2caux_interface.h"
-
-/*
- * Header of this unit
- */
-
-#include "engine.h"
+#include "core_types.h"
+#include "dchubbub.h"
 
-void dal_i2caux_construct_engine(
-	struct engine *engine,
-	struct dc_context *ctx)
-{
-	engine->ddc = NULL;
-	engine->ctx = ctx;
-}
+struct dcn_vmid_registers {
+	uint32_t CNTL;
+	uint32_t PAGE_TABLE_BASE_ADDR_HI32;
+	uint32_t PAGE_TABLE_BASE_ADDR_LO32;
+	uint32_t PAGE_TABLE_START_ADDR_HI32;
+	uint32_t PAGE_TABLE_START_ADDR_LO32;
+	uint32_t PAGE_TABLE_END_ADDR_HI32;
+	uint32_t PAGE_TABLE_END_ADDR_LO32;
+};
 
-void dal_i2caux_destruct_engine(
-	struct engine *engine)
-{
-	/* nothing to do */
-}
+struct dcn_vmid_page_table_config {
+	uint64_t	page_table_start_addr;
+	uint64_t	page_table_end_addr;
+	enum dcn_hubbub_page_table_depth	depth;
+	enum dcn_hubbub_page_table_block_size	block_size;
+};
 
+#endif /* DAL_DC_INC_HW_VMID_H_ */
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 d6a85f48b6d1..341b4810288c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -70,6 +70,8 @@ struct hw_sequencer_funcs {
 
 	void (*init_hw)(struct dc *dc);
 
+	void (*init_pipes)(struct dc *dc, struct dc_state *context);
+
 	enum dc_status (*apply_ctx_to_hw)(
 			struct dc *dc, struct dc_state *context);
 
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_sw_engine_dce110.h b/drivers/gpu/drm/amd/display/dc/inc/vm_helper.h
index c48c61f540a8..193407f76a80 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce110/i2c_sw_engine_dce110.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/vm_helper.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-15 Advanced Micro Devices, Inc.
+ * Copyright 2018 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"),
@@ -23,21 +23,34 @@
  *
  */
 
-#ifndef __DAL_I2C_SW_ENGINE_DCE110_H__
-#define __DAL_I2C_SW_ENGINE_DCE110_H__
+#ifndef DC_INC_VM_HELPER_H_
+#define DC_INC_VM_HELPER_H_
 
-struct i2c_sw_engine_dce110 {
-	struct i2c_sw_engine base;
-	uint32_t engine_id;
+#include "dc_types.h"
+
+#define MAX_VMID 16
+#define MAX_HUBP 6
+
+struct vmid_usage {
+	uint16_t vmid_usage[2];
 };
 
-struct i2c_sw_engine_dce110_create_arg {
-	uint32_t engine_id;
-	uint32_t default_speed;
-	struct dc_context *ctx;
+struct vm_helper {
+	unsigned int num_vmid;
+	unsigned int num_hubp;
+	unsigned int num_vmids_available;
+	uint64_t ptb_assigned_to_vmid[MAX_VMID];
+	struct vmid_usage hubp_vmid_usage[MAX_HUBP];
 };
 
-struct i2c_engine *dal_i2c_sw_engine_dce110_create(
-	const struct i2c_sw_engine_dce110_create_arg *arg);
+uint8_t get_vmid_for_ptb(
+		struct vm_helper *vm_helper,
+		int64_t ptb,
+		uint8_t pipe_idx);
+
+void init_vm_helper(
+	struct vm_helper *vm_helper,
+	unsigned int num_vmid,
+	unsigned int num_hubp);
 
-#endif
+#endif /* DC_INC_VM_HELPER_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/irq_types.h b/drivers/gpu/drm/amd/display/dc/irq_types.h
index 0b5f3a278c22..d0ccd81ad5b4 100644
--- a/drivers/gpu/drm/amd/display/dc/irq_types.h
+++ b/drivers/gpu/drm/amd/display/dc/irq_types.h
@@ -144,6 +144,14 @@ enum dc_irq_source {
 	DC_IRQ_SOURCE_DC5_VLINE0,
 	DC_IRQ_SOURCE_DC6_VLINE0,
 
+	DC_IRQ_SOURCE_DC1_VLINE1,
+	DC_IRQ_SOURCE_DC2_VLINE1,
+	DC_IRQ_SOURCE_DC3_VLINE1,
+	DC_IRQ_SOURCE_DC4_VLINE1,
+	DC_IRQ_SOURCE_DC5_VLINE1,
+	DC_IRQ_SOURCE_DC6_VLINE1,
+
+
 	DAL_IRQ_SOURCES_NUMBER
 };
 
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 7fd78a696800..01bf01a34a08 100644
--- a/drivers/gpu/drm/amd/display/include/bios_parser_types.h
+++ b/drivers/gpu/drm/amd/display/include/bios_parser_types.h
@@ -211,8 +211,8 @@ struct bp_pixel_clock_parameters {
 	/* signal_type -> Encoder Mode - needed by VBIOS Exec table */
 	enum signal_type signal_type;
 	/* Adjusted Pixel Clock (after VBIOS exec table)
-	 * that becomes Target Pixel Clock (KHz) */
-	uint32_t target_pixel_clock;
+	 * that becomes Target Pixel Clock (100 Hz units) */
+	uint32_t target_pixel_clock_100hz;
 	/* Calculated Reference divider of Display PLL */
 	uint32_t reference_divider;
 	/* Calculated Feedback divider of Display PLL */
diff --git a/drivers/gpu/drm/amd/display/include/gpio_interface.h b/drivers/gpu/drm/amd/display/include/gpio_interface.h
index e4fd31024b92..7de64195dc33 100644
--- a/drivers/gpu/drm/amd/display/include/gpio_interface.h
+++ b/drivers/gpu/drm/amd/display/include/gpio_interface.h
@@ -59,6 +59,14 @@ enum gpio_result dal_gpio_change_mode(
 	struct gpio *gpio,
 	enum gpio_mode mode);
 
+/* Lock Pin */
+enum gpio_result dal_gpio_lock_pin(
+	struct gpio *gpio);
+
+/* Unlock Pin */
+enum gpio_result dal_gpio_unlock_pin(
+	struct gpio *gpio);
+
 /* Get the GPIO id */
 enum gpio_id dal_gpio_get_id(
 	const struct gpio *gpio);
diff --git a/drivers/gpu/drm/amd/display/include/i2caux_interface.h b/drivers/gpu/drm/amd/display/include/i2caux_interface.h
index 13a3c82d118f..bb012cb1a9f5 100644
--- a/drivers/gpu/drm/amd/display/include/i2caux_interface.h
+++ b/drivers/gpu/drm/amd/display/include/i2caux_interface.h
@@ -40,9 +40,19 @@ struct aux_payload {
 	/* set following flag to write data,
 	 * reset it to read data */
 	bool write;
+	bool mot;
 	uint32_t address;
 	uint8_t length;
 	uint8_t *data;
+	/*
+	 * used to return the reply type of the transaction
+	 * ignored if NULL
+	 */
+	uint8_t *reply;
+	/* expressed in milliseconds
+	 * zero means "use default value"
+	 */
+	uint32_t defer_delay;
 };
 
 struct aux_command {
@@ -66,27 +76,4 @@ union aux_config {
 	uint32_t raw;
 };
 
-struct i2caux;
-
-struct i2caux *dal_i2caux_create(
-	struct dc_context *ctx);
-
-bool dal_i2caux_submit_i2c_command(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	struct i2c_command *cmd);
-
-bool dal_i2caux_submit_aux_command(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	struct aux_command *cmd);
-
-void dal_i2caux_configure_aux(
-	struct i2caux *i2caux,
-	struct ddc *ddc,
-	union aux_config cfg);
-
-void dal_i2caux_destroy(
-	struct i2caux **ptr);
-
 #endif
diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
index 479b77c2e89e..0fbc8fbc3541 100644
--- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
+++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
@@ -823,7 +823,7 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma,
 	bool is_clipped = false;
 	struct fixed31_32 sdr_white_level;
 
-	if (fs_params == NULL || fs_params->max_content == 0 ||
+	if (fs_params->max_content == 0 ||
 			fs_params->max_display == 0)
 		return false;
 
@@ -1508,7 +1508,7 @@ static bool map_regamma_hw_to_x_user(
 	struct hw_x_point *coords = coords_x;
 	const struct pwl_float_data_ex *regamma = rgb_regamma;
 
-	if (mapUserRamp) {
+	if (ramp && mapUserRamp) {
 		copy_rgb_regamma_to_coordinates_x(coords,
 				hw_points_num,
 				rgb_regamma);
@@ -1545,7 +1545,7 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
 
 	struct pwl_float_data *rgb_user = NULL;
 	struct pwl_float_data_ex *rgb_regamma = NULL;
-	struct gamma_pixel *axix_x = NULL;
+	struct gamma_pixel *axis_x = NULL;
 	struct pixel_gamma_point *coeff = NULL;
 	enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB;
 	bool ret = false;
@@ -1555,47 +1555,54 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
 
 	/* we can use hardcoded curve for plain SRGB TF */
 	if (output_tf->type == TF_TYPE_PREDEFINED && canRomBeUsed == true &&
-			output_tf->tf == TRANSFER_FUNCTION_SRGB &&
-			(ramp->is_identity || (!mapUserRamp && ramp->type == GAMMA_RGB_256)))
-		return true;
+			output_tf->tf == TRANSFER_FUNCTION_SRGB) {
+		if (ramp == NULL)
+			return true;
+		if (ramp->is_identity || (!mapUserRamp && ramp->type == GAMMA_RGB_256))
+			return true;
+	}
 
 	output_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
 
-	rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS,
+	if (ramp && (mapUserRamp || ramp->type != GAMMA_RGB_256)) {
+		rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS,
 			    sizeof(*rgb_user),
 			    GFP_KERNEL);
-	if (!rgb_user)
-		goto rgb_user_alloc_fail;
+		if (!rgb_user)
+			goto rgb_user_alloc_fail;
+
+		axis_x = kvcalloc(ramp->num_entries + 3, sizeof(*axis_x),
+				GFP_KERNEL);
+		if (!axis_x)
+			goto axis_x_alloc_fail;
+
+		dividers.divider1 = dc_fixpt_from_fraction(3, 2);
+		dividers.divider2 = dc_fixpt_from_int(2);
+		dividers.divider3 = dc_fixpt_from_fraction(5, 2);
+
+		build_evenly_distributed_points(
+				axis_x,
+				ramp->num_entries,
+				dividers);
+
+		if (ramp->type == GAMMA_RGB_256 && mapUserRamp)
+			scale_gamma(rgb_user, ramp, dividers);
+		else if (ramp->type == GAMMA_RGB_FLOAT_1024)
+			scale_gamma_dx(rgb_user, ramp, dividers);
+	}
+
 	rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS,
 			       sizeof(*rgb_regamma),
 			       GFP_KERNEL);
 	if (!rgb_regamma)
 		goto rgb_regamma_alloc_fail;
-	axix_x = kvcalloc(ramp->num_entries + 3, sizeof(*axix_x),
-			  GFP_KERNEL);
-	if (!axix_x)
-		goto axix_x_alloc_fail;
+
 	coeff = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*coeff),
 			 GFP_KERNEL);
 	if (!coeff)
 		goto coeff_alloc_fail;
 
-	dividers.divider1 = dc_fixpt_from_fraction(3, 2);
-	dividers.divider2 = dc_fixpt_from_int(2);
-	dividers.divider3 = dc_fixpt_from_fraction(5, 2);
-
 	tf = output_tf->tf;
-
-	build_evenly_distributed_points(
-			axix_x,
-			ramp->num_entries,
-			dividers);
-
-	if (ramp->type == GAMMA_RGB_256 && mapUserRamp)
-		scale_gamma(rgb_user, ramp, dividers);
-	else if (ramp->type == GAMMA_RGB_FLOAT_1024)
-		scale_gamma_dx(rgb_user, ramp, dividers);
-
 	if (tf == TRANSFER_FUNCTION_PQ) {
 		tf_pts->end_exponent = 7;
 		tf_pts->x_point_at_y1_red = 125;
@@ -1623,22 +1630,22 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
 				coordinates_x, tf == TRANSFER_FUNCTION_SRGB ? true:false);
 	}
 	map_regamma_hw_to_x_user(ramp, coeff, rgb_user,
-			coordinates_x, axix_x, rgb_regamma,
+			coordinates_x, axis_x, rgb_regamma,
 			MAX_HW_POINTS, tf_pts,
-			(mapUserRamp || ramp->type != GAMMA_RGB_256) &&
-			ramp->type != GAMMA_CS_TFM_1D);
+			(mapUserRamp || (ramp && ramp->type != GAMMA_RGB_256)) &&
+			(ramp && ramp->type != GAMMA_CS_TFM_1D));
 
-	if (ramp->type == GAMMA_CS_TFM_1D)
+	if (ramp && ramp->type == GAMMA_CS_TFM_1D)
 		apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts);
 
 	ret = true;
 
 	kvfree(coeff);
 coeff_alloc_fail:
-	kvfree(axix_x);
-axix_x_alloc_fail:
 	kvfree(rgb_regamma);
 rgb_regamma_alloc_fail:
+	kvfree(axis_x);
+axis_x_alloc_fail:
 	kvfree(rgb_user);
 rgb_user_alloc_fail:
 	return ret;
@@ -1758,69 +1765,85 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf,
 {
 	struct dc_transfer_func_distributed_points *tf_pts = &input_tf->tf_pts;
 	struct dividers dividers;
-
 	struct pwl_float_data *rgb_user = NULL;
 	struct pwl_float_data_ex *curve = NULL;
 	struct gamma_pixel *axis_x = NULL;
 	struct pixel_gamma_point *coeff = NULL;
 	enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB;
+	uint32_t i;
 	bool ret = false;
 
 	if (input_tf->type == TF_TYPE_BYPASS)
 		return false;
 
-	/* we can use hardcoded curve for plain SRGB TF */
+	/* we can use hardcoded curve for plain SRGB TF
+	 * If linear, it's bypass if on user ramp
+	 */
 	if (input_tf->type == TF_TYPE_PREDEFINED &&
-			input_tf->tf == TRANSFER_FUNCTION_SRGB &&
-			(!mapUserRamp &&
-			(ramp->type == GAMMA_RGB_256 || ramp->num_entries == 0)))
+			(input_tf->tf == TRANSFER_FUNCTION_SRGB ||
+					input_tf->tf == TRANSFER_FUNCTION_LINEAR) &&
+					!mapUserRamp)
 		return true;
 
 	input_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
 
-	rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS,
-			    sizeof(*rgb_user),
-			    GFP_KERNEL);
-	if (!rgb_user)
-		goto rgb_user_alloc_fail;
+	if (mapUserRamp && ramp && ramp->type == GAMMA_RGB_256) {
+		rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS,
+				sizeof(*rgb_user),
+				GFP_KERNEL);
+		if (!rgb_user)
+			goto rgb_user_alloc_fail;
+
+		axis_x = kvcalloc(ramp->num_entries + _EXTRA_POINTS, sizeof(*axis_x),
+				GFP_KERNEL);
+		if (!axis_x)
+			goto axis_x_alloc_fail;
+
+		dividers.divider1 = dc_fixpt_from_fraction(3, 2);
+		dividers.divider2 = dc_fixpt_from_int(2);
+		dividers.divider3 = dc_fixpt_from_fraction(5, 2);
+
+		build_evenly_distributed_points(
+				axis_x,
+				ramp->num_entries,
+				dividers);
+
+		scale_gamma(rgb_user, ramp, dividers);
+	}
+
 	curve = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*curve),
-			 GFP_KERNEL);
+			GFP_KERNEL);
 	if (!curve)
 		goto curve_alloc_fail;
-	axis_x = kvcalloc(ramp->num_entries + _EXTRA_POINTS, sizeof(*axis_x),
-			  GFP_KERNEL);
-	if (!axis_x)
-		goto axis_x_alloc_fail;
+
 	coeff = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*coeff),
-			 GFP_KERNEL);
+			GFP_KERNEL);
 	if (!coeff)
 		goto coeff_alloc_fail;
 
-	dividers.divider1 = dc_fixpt_from_fraction(3, 2);
-	dividers.divider2 = dc_fixpt_from_int(2);
-	dividers.divider3 = dc_fixpt_from_fraction(5, 2);
-
 	tf = input_tf->tf;
 
-	build_evenly_distributed_points(
-			axis_x,
-			ramp->num_entries,
-			dividers);
-
-	if (ramp->type == GAMMA_RGB_256 && mapUserRamp)
-		scale_gamma(rgb_user, ramp, dividers);
-	else if (ramp->type == GAMMA_RGB_FLOAT_1024)
-		scale_gamma_dx(rgb_user, ramp, dividers);
-
 	if (tf == TRANSFER_FUNCTION_PQ)
 		build_de_pq(curve,
 				MAX_HW_POINTS,
 				coordinates_x);
-	else
+	else if (tf == TRANSFER_FUNCTION_SRGB ||
+			tf == TRANSFER_FUNCTION_BT709)
 		build_degamma(curve,
 				MAX_HW_POINTS,
 				coordinates_x,
-				tf == TRANSFER_FUNCTION_SRGB ? true:false);
+				tf == TRANSFER_FUNCTION_SRGB ? true : false);
+	else if (tf == TRANSFER_FUNCTION_LINEAR) {
+		// just copy coordinates_x into curve
+		i = 0;
+		while (i != MAX_HW_POINTS + 1) {
+			curve[i].r = coordinates_x[i].x;
+			curve[i].g = curve[i].r;
+			curve[i].b = curve[i].r;
+			i++;
+		}
+	} else
+		goto invalid_tf_fail;
 
 	tf_pts->end_exponent = 0;
 	tf_pts->x_point_at_y1_red = 1;
@@ -1830,23 +1853,21 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf,
 	map_regamma_hw_to_x_user(ramp, coeff, rgb_user,
 			coordinates_x, axis_x, curve,
 			MAX_HW_POINTS, tf_pts,
-			mapUserRamp && ramp->type != GAMMA_CUSTOM);
-	if (ramp->type == GAMMA_CUSTOM)
-		apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts);
+			mapUserRamp && ramp && ramp->type == GAMMA_RGB_256);
 
 	ret = true;
 
+invalid_tf_fail:
 	kvfree(coeff);
 coeff_alloc_fail:
-	kvfree(axis_x);
-axis_x_alloc_fail:
 	kvfree(curve);
 curve_alloc_fail:
+	kvfree(axis_x);
+axis_x_alloc_fail:
 	kvfree(rgb_user);
 rgb_user_alloc_fail:
 
 	return ret;
-
 }
 
 
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 1544ed3f1747..94a84bc57c7a 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -108,8 +108,8 @@ static unsigned int calc_duration_in_us_from_v_total(
 {
 	unsigned int duration_in_us =
 			(unsigned int)(div64_u64(((unsigned long long)(v_total)
-				* 1000) * stream->timing.h_total,
-					stream->timing.pix_clk_khz));
+				* 10000) * stream->timing.h_total,
+					stream->timing.pix_clk_100hz));
 
 	return duration_in_us;
 }
@@ -126,7 +126,7 @@ static unsigned int calc_v_total_from_refresh(
 					refresh_in_uhz)));
 
 	v_total = div64_u64(div64_u64(((unsigned long long)(
-			frame_duration_in_ns) * stream->timing.pix_clk_khz),
+			frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)),
 			stream->timing.h_total), 1000000);
 
 	/* v_total cannot be less than nominal */
@@ -152,7 +152,7 @@ static unsigned int calc_v_total_from_duration(
 		duration_in_us = vrr->max_duration_in_us;
 
 	v_total = div64_u64(div64_u64(((unsigned long long)(
-				duration_in_us) * stream->timing.pix_clk_khz),
+				duration_in_us) * (stream->timing.pix_clk_100hz / 10)),
 				stream->timing.h_total), 1000);
 
 	/* v_total cannot be less than nominal */
@@ -227,7 +227,7 @@ static void update_v_total_for_static_ramp(
 	}
 
 	v_total = div64_u64(div64_u64(((unsigned long long)(
-			current_duration_in_us) * stream->timing.pix_clk_khz),
+			current_duration_in_us) * (stream->timing.pix_clk_100hz / 10)),
 				stream->timing.h_total), 1000);
 
 	in_out_vrr->adjust.v_total_min = v_total;
@@ -461,6 +461,26 @@ bool mod_freesync_get_v_position(struct mod_freesync *mod_freesync,
 	return false;
 }
 
+static void build_vrr_infopacket_header_vtem(enum signal_type signal,
+		struct dc_info_packet *infopacket)
+{
+	// HEADER
+
+	// HB0, HB1, HB2 indicates PacketType VTEMPacket
+	infopacket->hb0 = 0x7F;
+	infopacket->hb1 = 0xC0;
+	infopacket->hb2 = 0x00;
+	/* HB3 Bit Fields
+	 * Reserved :1 = 0
+	 * Sync     :1 = 0
+	 * VFR      :1 = 1
+	 * Ds_Type  :2 = 0
+	 * End      :1 = 0
+	 * New      :1 = 0
+	 */
+	infopacket->hb3 = 0x20;
+}
+
 static void build_vrr_infopacket_header_v1(enum signal_type signal,
 		struct dc_info_packet *infopacket,
 		unsigned int *payload_size)
@@ -559,6 +579,54 @@ static void build_vrr_infopacket_header_v2(enum signal_type signal,
 	}
 }
 
+static void build_vrr_vtem_infopacket_data(const struct dc_stream_state *stream,
+		const struct mod_vrr_params *vrr,
+		struct dc_info_packet *infopacket)
+{
+	/* dc_info_packet to VtemPacket Translation of Bit-fields,
+	 * SB[6]
+	 * unsigned char VRR_EN        :1
+	 * unsigned char M_CONST       :1
+	 * unsigned char Reserved2     :2
+	 * unsigned char FVA_Factor_M1 :4
+	 * SB[7]
+	 * unsigned char Base_Vfront   :8
+	 * SB[8]
+	 * unsigned char Base_Refresh_Rate_98 :2
+	 * unsigned char RB                   :1
+	 * unsigned char Reserved3            :5
+	 * SB[9]
+	 * unsigned char Base_RefreshRate_07  :8
+	 */
+	unsigned int fieldRateInHz;
+
+	if (vrr->state == VRR_STATE_ACTIVE_VARIABLE ||
+				vrr->state == VRR_STATE_ACTIVE_FIXED){
+		infopacket->sb[6] |= 0x80; //VRR_EN Bit = 1
+	} else {
+		infopacket->sb[6] &= 0x7F; //VRR_EN Bit = 0
+	}
+
+	if (!stream->timing.vic) {
+		infopacket->sb[7] = stream->timing.v_front_porch;
+
+		/* TODO: In dal2, we check mode flags for a reduced blanking timing.
+		 * Need a way to relay that information to this function.
+		 * if("ReducedBlanking")
+		 * {
+		 *   infopacket->sb[8] |= 0x20; //Set 3rd bit to 1
+		 * }
+		 */
+		fieldRateInHz = (stream->timing.pix_clk_100hz * 100)/
+				(stream->timing.h_total * stream->timing.v_total);
+
+		infopacket->sb[8] |= ((fieldRateInHz & 0x300) >> 2);
+		infopacket->sb[9] |= fieldRateInHz & 0xFF;
+
+	}
+	infopacket->valid = true;
+}
+
 static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr,
 		struct dc_info_packet *infopacket)
 {
@@ -672,6 +740,19 @@ static void build_vrr_infopacket_v2(enum signal_type signal,
 	infopacket->valid = true;
 }
 
+static void build_vrr_infopacket_vtem(const struct dc_stream_state *stream,
+		const struct mod_vrr_params *vrr,
+		struct dc_info_packet *infopacket)
+{
+	//VTEM info packet for HdmiVrr
+
+	//VTEM Packet is structured differently
+	build_vrr_infopacket_header_vtem(stream->signal, infopacket);
+	build_vrr_vtem_infopacket_data(stream, vrr, infopacket);
+
+	infopacket->valid = true;
+}
+
 void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync,
 		const struct dc_stream_state *stream,
 		const struct mod_vrr_params *vrr,
@@ -679,18 +760,21 @@ void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync,
 		const enum color_transfer_func *app_tf,
 		struct dc_info_packet *infopacket)
 {
-	/* SPD info packet for FreeSync */
-
-	/* Check if Freesync is supported. Return if false. If true,
+	/* SPD info packet for FreeSync
+	 * VTEM info packet for HdmiVRR
+	 * Check if Freesync is supported. Return if false. If true,
 	 * set the corresponding bit in the info packet
 	 */
-	if (!vrr->supported || !vrr->send_vsif)
+	if (!vrr->supported || (!vrr->send_info_frame && packet_type != PACKET_TYPE_VTEM))
 		return;
 
 	switch (packet_type) {
 	case PACKET_TYPE_FS2:
 		build_vrr_infopacket_v2(stream->signal, vrr, app_tf, infopacket);
 		break;
+	case PACKET_TYPE_VTEM:
+		build_vrr_infopacket_vtem(stream, vrr, infopacket);
+		break;
 	case PACKET_TYPE_VRR:
 	case PACKET_TYPE_FS1:
 	default:
@@ -739,7 +823,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync,
 		return;
 
 	in_out_vrr->state = in_config->state;
-	in_out_vrr->send_vsif = in_config->vsif_supported;
+	in_out_vrr->send_info_frame = in_config->vsif_supported;
 
 	if (in_config->state == VRR_STATE_UNSUPPORTED) {
 		in_out_vrr->state = VRR_STATE_UNSUPPORTED;
@@ -972,7 +1056,7 @@ unsigned long long mod_freesync_calc_nominal_field_rate(
 	unsigned long long nominal_field_rate_in_uhz = 0;
 
 	/* Calculate nominal field rate for stream */
-	nominal_field_rate_in_uhz = stream->timing.pix_clk_khz;
+	nominal_field_rate_in_uhz = stream->timing.pix_clk_100hz / 10;
 	nominal_field_rate_in_uhz *= 1000ULL * 1000ULL * 1000ULL;
 	nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz,
 						stream->timing.h_total);
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h
index 949a8b62aa98..4222e403b151 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h
@@ -104,7 +104,7 @@ struct mod_vrr_params_fixed_refresh {
 
 struct mod_vrr_params {
 	bool supported;
-	bool send_vsif;
+	bool send_info_frame;
 	enum mod_vrr_state state;
 
 	uint32_t min_refresh_in_uhz;
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h
index 1bd02c0ac30c..b711e7e6c204 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h
@@ -41,7 +41,8 @@ enum color_transfer_func {
 enum vrr_packet_type {
 	PACKET_TYPE_VRR,
 	PACKET_TYPE_FS1,
-	PACKET_TYPE_FS2
+	PACKET_TYPE_FS2,
+	PACKET_TYPE_VTEM
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 00f63b7dd32f..3ba87b076287 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -41,6 +41,17 @@ static const unsigned char min_reduction_table[13] = {
 static const unsigned char max_reduction_table[13] = {
 0xf5, 0xe5, 0xd9, 0xcd, 0xb1, 0xa5, 0xa5, 0x80, 0x65, 0x4d, 0x4d, 0x4d, 0x32};
 
+/* ABM 2.2 Min Reduction effectively disabled (100% for all configs)*/
+static const unsigned char min_reduction_table_v_2_2[13] = {
+0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+
+/* Possible ABM 2.2 Max Reduction configs from least aggressive to most aggressive
+ *  0    1     2     3     4     5     6     7     8     9     10    11   12
+ * 96.1 89.8 74.9  69.4  64.7  52.2  48.6  39.6  30.2  25.1  19.6  12.5  12.5 %
+ */
+static const unsigned char max_reduction_table_v_2_2[13] = {
+0xf5, 0xe5, 0xbf, 0xb1, 0xa5, 0x85, 0x7c, 0x65, 0x4d, 0x40, 0x32, 0x20, 0x20};
+
 /* Predefined ABM configuration sets. We may have different configuration sets
  * in order to satisfy different power/quality requirements.
  */
@@ -56,7 +67,15 @@ static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_le
 #define NUM_AGGR_LEVEL    4
 #define NUM_POWER_FN_SEGS 8
 #define NUM_BL_CURVE_SEGS 16
+#define IRAM_SIZE 256
 
+#define IRAM_RESERVE_AREA_START_V2 0xF0  // reserve 0xF0~0xF6 are write by DMCU only
+#define IRAM_RESERVE_AREA_END_V2 0xF6  // reserve 0xF0~0xF6 are write by DMCU only
+
+#define IRAM_RESERVE_AREA_START_V2_2 0xF0  // reserve 0xF0~0xFF are write by DMCU only
+#define IRAM_RESERVE_AREA_END_V2_2 0xFF  // reserve 0xF0~0xFF are write by DMCU only
+
+#pragma pack(push, 1)
 /* NOTE: iRAM is 256B in size */
 struct iram_table_v_2 {
 	/* flags                      */
@@ -85,11 +104,10 @@ struct iram_table_v_2 {
 
 	/* For reading PSR State directly from IRAM */
 	uint8_t psr_state;						/* 0xf0       */
-	uint8_t dmcu_interface_version;					/* 0xf1       */
-	uint8_t dmcu_date_version_year_b0;				/* 0xf2       */
-	uint8_t dmcu_date_version_year_b1;				/* 0xf3       */
-	uint8_t dmcu_date_version_month;				/* 0xf4       */
-	uint8_t dmcu_date_version_day;					/* 0xf5       */
+	uint8_t dmcu_mcp_interface_version;							/* 0xf1       */
+	uint8_t dmcu_abm_feature_version;							/* 0xf2       */
+	uint8_t dmcu_psr_feature_version;							/* 0xf3       */
+	uint16_t dmcu_version;										/* 0xf4       */
 	uint8_t dmcu_state;						/* 0xf6       */
 
 	uint16_t blRampReduction;					/* 0xf7       */
@@ -101,6 +119,52 @@ struct iram_table_v_2 {
 	uint8_t dummy9;							/* 0xff       */
 };
 
+struct iram_table_v_2_2 {
+	/* flags                      */
+	uint16_t flags;							/* 0x00 U16  */
+
+	/* parameters for ABM2.2 algorithm */
+	uint8_t min_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL];		/* 0x02 U0.8 */
+	uint8_t max_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL];		/* 0x16 U0.8 */
+	uint8_t bright_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL];	/* 0x2a U2.6 */
+	uint8_t dark_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL];		/* 0x3e U2.6 */
+	uint8_t hybridFactor[NUM_AGGR_LEVEL];						/* 0x52 U0.8 */
+	uint8_t contrastFactor[NUM_AGGR_LEVEL];						/* 0x56 U0.8 */
+	uint8_t deviation_gain[NUM_AGGR_LEVEL];						/* 0x5a U0.8 */
+	uint8_t iir_curve[NUM_AMBI_LEVEL];							/* 0x5e U0.8 */
+	uint8_t pad[29];											/* 0x63 U0.8 */
+
+	/* parameters for crgb conversion */
+	uint16_t crgb_thresh[NUM_POWER_FN_SEGS];					/* 0x80 U3.13 */
+	uint16_t crgb_offset[NUM_POWER_FN_SEGS];					/* 0x90 U1.15 */
+	uint16_t crgb_slope[NUM_POWER_FN_SEGS];						/* 0xa0 U4.12 */
+
+	/* parameters for custom curve */
+	/* thresholds for brightness --> backlight */
+	uint16_t backlight_thresholds[NUM_BL_CURVE_SEGS];			/* 0xb0 U16.0 */
+	/* offsets for brightness --> backlight */
+	uint16_t backlight_offsets[NUM_BL_CURVE_SEGS];				/* 0xd0 U16.0 */
+
+	/* For reading PSR State directly from IRAM */
+	uint8_t psr_state;											/* 0xf0       */
+	uint8_t dmcu_mcp_interface_version;							/* 0xf1       */
+	uint8_t dmcu_abm_feature_version;							/* 0xf2       */
+	uint8_t dmcu_psr_feature_version;							/* 0xf3       */
+	uint16_t dmcu_version;										/* 0xf4       */
+	uint8_t dmcu_state;											/* 0xf6       */
+
+	uint8_t dummy1;												/* 0xf7       */
+	uint8_t dummy2;												/* 0xf8       */
+	uint8_t dummy3;												/* 0xf9       */
+	uint8_t dummy4;												/* 0xfa       */
+	uint8_t dummy5;												/* 0xfb       */
+	uint8_t dummy6;												/* 0xfc       */
+	uint8_t dummy7;												/* 0xfd       */
+	uint8_t dummy8;												/* 0xfe       */
+	uint8_t dummy9;												/* 0xff       */
+};
+#pragma pack(pop)
+
 static uint16_t backlight_8_to_16(unsigned int backlight_8bit)
 {
 	return (uint16_t)(backlight_8bit * 0x101);
@@ -143,11 +207,363 @@ static void fill_backlight_transform_table(struct dmcu_iram_parameters params,
 	}
 }
 
+static void fill_backlight_transform_table_v_2_2(struct dmcu_iram_parameters params,
+		struct iram_table_v_2_2 *table)
+{
+	unsigned int i;
+	unsigned int num_entries = NUM_BL_CURVE_SEGS;
+	unsigned int query_input_8bit;
+	unsigned int query_output_8bit;
+	unsigned int lut_index;
+
+	table->backlight_thresholds[0] = 0;
+	table->backlight_offsets[0] = params.backlight_lut_array[0];
+	table->backlight_thresholds[num_entries-1] = 0xFFFF;
+	table->backlight_offsets[num_entries-1] =
+		params.backlight_lut_array[params.backlight_lut_array_size - 1];
+
+	/* Setup all brightness levels between 0% and 100% exclusive
+	 * Fills brightness-to-backlight transform table. Backlight custom curve
+	 * describes transform from brightness to backlight. It will be defined
+	 * as set of thresholds and set of offsets, together, implying
+	 * extrapolation of custom curve into 16 uniformly spanned linear
+	 * segments.  Each threshold/offset represented by 16 bit entry in
+	 * format U4.10.
+	 */
+	for (i = 1; i+1 < num_entries; i++) {
+		query_input_8bit = DIV_ROUNDUP((i * 256), num_entries);
+
+		lut_index = (params.backlight_lut_array_size - 1) * i / (num_entries - 1);
+		ASSERT(lut_index < params.backlight_lut_array_size);
+		query_output_8bit = params.backlight_lut_array[lut_index] >> 8;
+
+		table->backlight_thresholds[i] =
+				backlight_8_to_16(query_input_8bit);
+		table->backlight_offsets[i] =
+				backlight_8_to_16(query_output_8bit);
+	}
+}
+
+void fill_iram_v_2(struct iram_table_v_2 *ram_table, struct dmcu_iram_parameters params)
+{
+	unsigned int set = params.set;
+
+	ram_table->flags = 0x0;
+	ram_table->deviation_gain = 0xb3;
+
+	ram_table->blRampReduction =
+		cpu_to_be16(params.backlight_ramping_reduction);
+	ram_table->blRampStart =
+		cpu_to_be16(params.backlight_ramping_start);
+
+	ram_table->min_reduction[0][0] = min_reduction_table[abm_config[set][0]];
+	ram_table->min_reduction[1][0] = min_reduction_table[abm_config[set][0]];
+	ram_table->min_reduction[2][0] = min_reduction_table[abm_config[set][0]];
+	ram_table->min_reduction[3][0] = min_reduction_table[abm_config[set][0]];
+	ram_table->min_reduction[4][0] = min_reduction_table[abm_config[set][0]];
+	ram_table->max_reduction[0][0] = max_reduction_table[abm_config[set][0]];
+	ram_table->max_reduction[1][0] = max_reduction_table[abm_config[set][0]];
+	ram_table->max_reduction[2][0] = max_reduction_table[abm_config[set][0]];
+	ram_table->max_reduction[3][0] = max_reduction_table[abm_config[set][0]];
+	ram_table->max_reduction[4][0] = max_reduction_table[abm_config[set][0]];
+
+	ram_table->min_reduction[0][1] = min_reduction_table[abm_config[set][1]];
+	ram_table->min_reduction[1][1] = min_reduction_table[abm_config[set][1]];
+	ram_table->min_reduction[2][1] = min_reduction_table[abm_config[set][1]];
+	ram_table->min_reduction[3][1] = min_reduction_table[abm_config[set][1]];
+	ram_table->min_reduction[4][1] = min_reduction_table[abm_config[set][1]];
+	ram_table->max_reduction[0][1] = max_reduction_table[abm_config[set][1]];
+	ram_table->max_reduction[1][1] = max_reduction_table[abm_config[set][1]];
+	ram_table->max_reduction[2][1] = max_reduction_table[abm_config[set][1]];
+	ram_table->max_reduction[3][1] = max_reduction_table[abm_config[set][1]];
+	ram_table->max_reduction[4][1] = max_reduction_table[abm_config[set][1]];
+
+	ram_table->min_reduction[0][2] = min_reduction_table[abm_config[set][2]];
+	ram_table->min_reduction[1][2] = min_reduction_table[abm_config[set][2]];
+	ram_table->min_reduction[2][2] = min_reduction_table[abm_config[set][2]];
+	ram_table->min_reduction[3][2] = min_reduction_table[abm_config[set][2]];
+	ram_table->min_reduction[4][2] = min_reduction_table[abm_config[set][2]];
+	ram_table->max_reduction[0][2] = max_reduction_table[abm_config[set][2]];
+	ram_table->max_reduction[1][2] = max_reduction_table[abm_config[set][2]];
+	ram_table->max_reduction[2][2] = max_reduction_table[abm_config[set][2]];
+	ram_table->max_reduction[3][2] = max_reduction_table[abm_config[set][2]];
+	ram_table->max_reduction[4][2] = max_reduction_table[abm_config[set][2]];
+
+	ram_table->min_reduction[0][3] = min_reduction_table[abm_config[set][3]];
+	ram_table->min_reduction[1][3] = min_reduction_table[abm_config[set][3]];
+	ram_table->min_reduction[2][3] = min_reduction_table[abm_config[set][3]];
+	ram_table->min_reduction[3][3] = min_reduction_table[abm_config[set][3]];
+	ram_table->min_reduction[4][3] = min_reduction_table[abm_config[set][3]];
+	ram_table->max_reduction[0][3] = max_reduction_table[abm_config[set][3]];
+	ram_table->max_reduction[1][3] = max_reduction_table[abm_config[set][3]];
+	ram_table->max_reduction[2][3] = max_reduction_table[abm_config[set][3]];
+	ram_table->max_reduction[3][3] = max_reduction_table[abm_config[set][3]];
+	ram_table->max_reduction[4][3] = max_reduction_table[abm_config[set][3]];
+
+	ram_table->bright_pos_gain[0][0] = 0x20;
+	ram_table->bright_pos_gain[0][1] = 0x20;
+	ram_table->bright_pos_gain[0][2] = 0x20;
+	ram_table->bright_pos_gain[0][3] = 0x20;
+	ram_table->bright_pos_gain[1][0] = 0x20;
+	ram_table->bright_pos_gain[1][1] = 0x20;
+	ram_table->bright_pos_gain[1][2] = 0x20;
+	ram_table->bright_pos_gain[1][3] = 0x20;
+	ram_table->bright_pos_gain[2][0] = 0x20;
+	ram_table->bright_pos_gain[2][1] = 0x20;
+	ram_table->bright_pos_gain[2][2] = 0x20;
+	ram_table->bright_pos_gain[2][3] = 0x20;
+	ram_table->bright_pos_gain[3][0] = 0x20;
+	ram_table->bright_pos_gain[3][1] = 0x20;
+	ram_table->bright_pos_gain[3][2] = 0x20;
+	ram_table->bright_pos_gain[3][3] = 0x20;
+	ram_table->bright_pos_gain[4][0] = 0x20;
+	ram_table->bright_pos_gain[4][1] = 0x20;
+	ram_table->bright_pos_gain[4][2] = 0x20;
+	ram_table->bright_pos_gain[4][3] = 0x20;
+	ram_table->bright_neg_gain[0][1] = 0x00;
+	ram_table->bright_neg_gain[0][2] = 0x00;
+	ram_table->bright_neg_gain[0][3] = 0x00;
+	ram_table->bright_neg_gain[1][0] = 0x00;
+	ram_table->bright_neg_gain[1][1] = 0x00;
+	ram_table->bright_neg_gain[1][2] = 0x00;
+	ram_table->bright_neg_gain[1][3] = 0x00;
+	ram_table->bright_neg_gain[2][0] = 0x00;
+	ram_table->bright_neg_gain[2][1] = 0x00;
+	ram_table->bright_neg_gain[2][2] = 0x00;
+	ram_table->bright_neg_gain[2][3] = 0x00;
+	ram_table->bright_neg_gain[3][0] = 0x00;
+	ram_table->bright_neg_gain[3][1] = 0x00;
+	ram_table->bright_neg_gain[3][2] = 0x00;
+	ram_table->bright_neg_gain[3][3] = 0x00;
+	ram_table->bright_neg_gain[4][0] = 0x00;
+	ram_table->bright_neg_gain[4][1] = 0x00;
+	ram_table->bright_neg_gain[4][2] = 0x00;
+	ram_table->bright_neg_gain[4][3] = 0x00;
+	ram_table->dark_pos_gain[0][0] = 0x00;
+	ram_table->dark_pos_gain[0][1] = 0x00;
+	ram_table->dark_pos_gain[0][2] = 0x00;
+	ram_table->dark_pos_gain[0][3] = 0x00;
+	ram_table->dark_pos_gain[1][0] = 0x00;
+	ram_table->dark_pos_gain[1][1] = 0x00;
+	ram_table->dark_pos_gain[1][2] = 0x00;
+	ram_table->dark_pos_gain[1][3] = 0x00;
+	ram_table->dark_pos_gain[2][0] = 0x00;
+	ram_table->dark_pos_gain[2][1] = 0x00;
+	ram_table->dark_pos_gain[2][2] = 0x00;
+	ram_table->dark_pos_gain[2][3] = 0x00;
+	ram_table->dark_pos_gain[3][0] = 0x00;
+	ram_table->dark_pos_gain[3][1] = 0x00;
+	ram_table->dark_pos_gain[3][2] = 0x00;
+	ram_table->dark_pos_gain[3][3] = 0x00;
+	ram_table->dark_pos_gain[4][0] = 0x00;
+	ram_table->dark_pos_gain[4][1] = 0x00;
+	ram_table->dark_pos_gain[4][2] = 0x00;
+	ram_table->dark_pos_gain[4][3] = 0x00;
+	ram_table->dark_neg_gain[0][0] = 0x00;
+	ram_table->dark_neg_gain[0][1] = 0x00;
+	ram_table->dark_neg_gain[0][2] = 0x00;
+	ram_table->dark_neg_gain[0][3] = 0x00;
+	ram_table->dark_neg_gain[1][0] = 0x00;
+	ram_table->dark_neg_gain[1][1] = 0x00;
+	ram_table->dark_neg_gain[1][2] = 0x00;
+	ram_table->dark_neg_gain[1][3] = 0x00;
+	ram_table->dark_neg_gain[2][0] = 0x00;
+	ram_table->dark_neg_gain[2][1] = 0x00;
+	ram_table->dark_neg_gain[2][2] = 0x00;
+	ram_table->dark_neg_gain[2][3] = 0x00;
+	ram_table->dark_neg_gain[3][0] = 0x00;
+	ram_table->dark_neg_gain[3][1] = 0x00;
+	ram_table->dark_neg_gain[3][2] = 0x00;
+	ram_table->dark_neg_gain[3][3] = 0x00;
+	ram_table->dark_neg_gain[4][0] = 0x00;
+	ram_table->dark_neg_gain[4][1] = 0x00;
+	ram_table->dark_neg_gain[4][2] = 0x00;
+	ram_table->dark_neg_gain[4][3] = 0x00;
+
+	ram_table->iir_curve[0] = 0x65;
+	ram_table->iir_curve[1] = 0x65;
+	ram_table->iir_curve[2] = 0x65;
+	ram_table->iir_curve[3] = 0x65;
+	ram_table->iir_curve[4] = 0x65;
+
+	//Gamma 2.4
+	ram_table->crgb_thresh[0] = cpu_to_be16(0x13b6);
+	ram_table->crgb_thresh[1] = cpu_to_be16(0x1648);
+	ram_table->crgb_thresh[2] = cpu_to_be16(0x18e3);
+	ram_table->crgb_thresh[3] = cpu_to_be16(0x1b41);
+	ram_table->crgb_thresh[4] = cpu_to_be16(0x1d46);
+	ram_table->crgb_thresh[5] = cpu_to_be16(0x1f21);
+	ram_table->crgb_thresh[6] = cpu_to_be16(0x2167);
+	ram_table->crgb_thresh[7] = cpu_to_be16(0x2384);
+	ram_table->crgb_offset[0] = cpu_to_be16(0x2999);
+	ram_table->crgb_offset[1] = cpu_to_be16(0x3999);
+	ram_table->crgb_offset[2] = cpu_to_be16(0x4666);
+	ram_table->crgb_offset[3] = cpu_to_be16(0x5999);
+	ram_table->crgb_offset[4] = cpu_to_be16(0x6333);
+	ram_table->crgb_offset[5] = cpu_to_be16(0x7800);
+	ram_table->crgb_offset[6] = cpu_to_be16(0x8c00);
+	ram_table->crgb_offset[7] = cpu_to_be16(0xa000);
+	ram_table->crgb_slope[0]  = cpu_to_be16(0x3147);
+	ram_table->crgb_slope[1]  = cpu_to_be16(0x2978);
+	ram_table->crgb_slope[2]  = cpu_to_be16(0x23a2);
+	ram_table->crgb_slope[3]  = cpu_to_be16(0x1f55);
+	ram_table->crgb_slope[4]  = cpu_to_be16(0x1c63);
+	ram_table->crgb_slope[5]  = cpu_to_be16(0x1a0f);
+	ram_table->crgb_slope[6]  = cpu_to_be16(0x178d);
+	ram_table->crgb_slope[7]  = cpu_to_be16(0x15ab);
+
+	fill_backlight_transform_table(
+			params, ram_table);
+}
+
+void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parameters params)
+{
+	unsigned int set = params.set;
+
+	ram_table->flags = 0x0;
+
+	ram_table->deviation_gain[0] = 0xb3;
+	ram_table->deviation_gain[1] = 0xb3;
+	ram_table->deviation_gain[2] = 0xb3;
+	ram_table->deviation_gain[3] = 0xb3;
+
+	ram_table->min_reduction[0][0] = min_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->min_reduction[1][0] = min_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->min_reduction[2][0] = min_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->min_reduction[3][0] = min_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->min_reduction[4][0] = min_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->max_reduction[0][0] = max_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->max_reduction[1][0] = max_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->max_reduction[2][0] = max_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->max_reduction[3][0] = max_reduction_table_v_2_2[abm_config[set][0]];
+	ram_table->max_reduction[4][0] = max_reduction_table_v_2_2[abm_config[set][0]];
+
+	ram_table->min_reduction[0][1] = min_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->min_reduction[1][1] = min_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->min_reduction[2][1] = min_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->min_reduction[3][1] = min_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->min_reduction[4][1] = min_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->max_reduction[0][1] = max_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->max_reduction[1][1] = max_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->max_reduction[2][1] = max_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->max_reduction[3][1] = max_reduction_table_v_2_2[abm_config[set][1]];
+	ram_table->max_reduction[4][1] = max_reduction_table_v_2_2[abm_config[set][1]];
+
+	ram_table->min_reduction[0][2] = min_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->min_reduction[1][2] = min_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->min_reduction[2][2] = min_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->min_reduction[3][2] = min_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->min_reduction[4][2] = min_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->max_reduction[0][2] = max_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->max_reduction[1][2] = max_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->max_reduction[2][2] = max_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->max_reduction[3][2] = max_reduction_table_v_2_2[abm_config[set][2]];
+	ram_table->max_reduction[4][2] = max_reduction_table_v_2_2[abm_config[set][2]];
+
+	ram_table->min_reduction[0][3] = min_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->min_reduction[1][3] = min_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->min_reduction[2][3] = min_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->min_reduction[3][3] = min_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->min_reduction[4][3] = min_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->max_reduction[0][3] = max_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->max_reduction[1][3] = max_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->max_reduction[2][3] = max_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->max_reduction[3][3] = max_reduction_table_v_2_2[abm_config[set][3]];
+	ram_table->max_reduction[4][3] = max_reduction_table_v_2_2[abm_config[set][3]];
+
+	ram_table->bright_pos_gain[0][0] = 0x20;
+	ram_table->bright_pos_gain[0][1] = 0x20;
+	ram_table->bright_pos_gain[0][2] = 0x20;
+	ram_table->bright_pos_gain[0][3] = 0x20;
+	ram_table->bright_pos_gain[1][0] = 0x20;
+	ram_table->bright_pos_gain[1][1] = 0x20;
+	ram_table->bright_pos_gain[1][2] = 0x20;
+	ram_table->bright_pos_gain[1][3] = 0x20;
+	ram_table->bright_pos_gain[2][0] = 0x20;
+	ram_table->bright_pos_gain[2][1] = 0x20;
+	ram_table->bright_pos_gain[2][2] = 0x20;
+	ram_table->bright_pos_gain[2][3] = 0x20;
+	ram_table->bright_pos_gain[3][0] = 0x20;
+	ram_table->bright_pos_gain[3][1] = 0x20;
+	ram_table->bright_pos_gain[3][2] = 0x20;
+	ram_table->bright_pos_gain[3][3] = 0x20;
+	ram_table->bright_pos_gain[4][0] = 0x20;
+	ram_table->bright_pos_gain[4][1] = 0x20;
+	ram_table->bright_pos_gain[4][2] = 0x20;
+	ram_table->bright_pos_gain[4][3] = 0x20;
+
+	ram_table->dark_pos_gain[0][0] = 0x00;
+	ram_table->dark_pos_gain[0][1] = 0x00;
+	ram_table->dark_pos_gain[0][2] = 0x00;
+	ram_table->dark_pos_gain[0][3] = 0x00;
+	ram_table->dark_pos_gain[1][0] = 0x00;
+	ram_table->dark_pos_gain[1][1] = 0x00;
+	ram_table->dark_pos_gain[1][2] = 0x00;
+	ram_table->dark_pos_gain[1][3] = 0x00;
+	ram_table->dark_pos_gain[2][0] = 0x00;
+	ram_table->dark_pos_gain[2][1] = 0x00;
+	ram_table->dark_pos_gain[2][2] = 0x00;
+	ram_table->dark_pos_gain[2][3] = 0x00;
+	ram_table->dark_pos_gain[3][0] = 0x00;
+	ram_table->dark_pos_gain[3][1] = 0x00;
+	ram_table->dark_pos_gain[3][2] = 0x00;
+	ram_table->dark_pos_gain[3][3] = 0x00;
+	ram_table->dark_pos_gain[4][0] = 0x00;
+	ram_table->dark_pos_gain[4][1] = 0x00;
+	ram_table->dark_pos_gain[4][2] = 0x00;
+	ram_table->dark_pos_gain[4][3] = 0x00;
+
+	ram_table->hybridFactor[0] = 0xff;
+	ram_table->hybridFactor[1] = 0xff;
+	ram_table->hybridFactor[2] = 0xff;
+	ram_table->hybridFactor[3] = 0xc0;
+
+	ram_table->contrastFactor[0] = 0x99;
+	ram_table->contrastFactor[1] = 0x99;
+	ram_table->contrastFactor[2] = 0x99;
+	ram_table->contrastFactor[3] = 0x80;
+
+	ram_table->iir_curve[0] = 0x65;
+	ram_table->iir_curve[1] = 0x65;
+	ram_table->iir_curve[2] = 0x65;
+	ram_table->iir_curve[3] = 0x65;
+	ram_table->iir_curve[4] = 0x65;
+
+	//Gamma 2.2
+	ram_table->crgb_thresh[0] = cpu_to_be16(0x127c);
+	ram_table->crgb_thresh[1] = cpu_to_be16(0x151b);
+	ram_table->crgb_thresh[2] = cpu_to_be16(0x17d5);
+	ram_table->crgb_thresh[3] = cpu_to_be16(0x1a56);
+	ram_table->crgb_thresh[4] = cpu_to_be16(0x1c83);
+	ram_table->crgb_thresh[5] = cpu_to_be16(0x1e72);
+	ram_table->crgb_thresh[6] = cpu_to_be16(0x20f0);
+	ram_table->crgb_thresh[7] = cpu_to_be16(0x232b);
+	ram_table->crgb_offset[0] = cpu_to_be16(0x2999);
+	ram_table->crgb_offset[1] = cpu_to_be16(0x3999);
+	ram_table->crgb_offset[2] = cpu_to_be16(0x4666);
+	ram_table->crgb_offset[3] = cpu_to_be16(0x5999);
+	ram_table->crgb_offset[4] = cpu_to_be16(0x6333);
+	ram_table->crgb_offset[5] = cpu_to_be16(0x7800);
+	ram_table->crgb_offset[6] = cpu_to_be16(0x8c00);
+	ram_table->crgb_offset[7] = cpu_to_be16(0xa000);
+	ram_table->crgb_slope[0]  = cpu_to_be16(0x3609);
+	ram_table->crgb_slope[1]  = cpu_to_be16(0x2dfa);
+	ram_table->crgb_slope[2]  = cpu_to_be16(0x27ea);
+	ram_table->crgb_slope[3]  = cpu_to_be16(0x235d);
+	ram_table->crgb_slope[4]  = cpu_to_be16(0x2042);
+	ram_table->crgb_slope[5]  = cpu_to_be16(0x1dc3);
+	ram_table->crgb_slope[6]  = cpu_to_be16(0x1b1a);
+	ram_table->crgb_slope[7]  = cpu_to_be16(0x1910);
+
+	fill_backlight_transform_table_v_2_2(
+			params, ram_table);
+}
+
 bool dmcu_load_iram(struct dmcu *dmcu,
 	struct dmcu_iram_parameters params)
 {
-	struct iram_table_v_2 ram_table;
-	unsigned int set = params.set;
+	unsigned char ram_table[IRAM_SIZE];
+	bool result = false;
 
 	if (dmcu == NULL)
 		return false;
@@ -157,170 +573,23 @@ bool dmcu_load_iram(struct dmcu *dmcu,
 
 	memset(&ram_table, 0, sizeof(ram_table));
 
-	ram_table.flags = 0x0;
-	ram_table.deviation_gain = 0xb3;
+	if (dmcu->dmcu_version.abm_version == 0x22) {
+		fill_iram_v_2_2((struct iram_table_v_2_2 *)ram_table, params);
 
-	ram_table.blRampReduction =
-		cpu_to_be16(params.backlight_ramping_reduction);
-	ram_table.blRampStart =
-		cpu_to_be16(params.backlight_ramping_start);
+		result = dmcu->funcs->load_iram(
+				dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2);
+	} else {
+		fill_iram_v_2((struct iram_table_v_2 *)ram_table, params);
 
-	ram_table.min_reduction[0][0] = min_reduction_table[abm_config[set][0]];
-	ram_table.min_reduction[1][0] = min_reduction_table[abm_config[set][0]];
-	ram_table.min_reduction[2][0] = min_reduction_table[abm_config[set][0]];
-	ram_table.min_reduction[3][0] = min_reduction_table[abm_config[set][0]];
-	ram_table.min_reduction[4][0] = min_reduction_table[abm_config[set][0]];
-	ram_table.max_reduction[0][0] = max_reduction_table[abm_config[set][0]];
-	ram_table.max_reduction[1][0] = max_reduction_table[abm_config[set][0]];
-	ram_table.max_reduction[2][0] = max_reduction_table[abm_config[set][0]];
-	ram_table.max_reduction[3][0] = max_reduction_table[abm_config[set][0]];
-	ram_table.max_reduction[4][0] = max_reduction_table[abm_config[set][0]];
-
-	ram_table.min_reduction[0][1] = min_reduction_table[abm_config[set][1]];
-	ram_table.min_reduction[1][1] = min_reduction_table[abm_config[set][1]];
-	ram_table.min_reduction[2][1] = min_reduction_table[abm_config[set][1]];
-	ram_table.min_reduction[3][1] = min_reduction_table[abm_config[set][1]];
-	ram_table.min_reduction[4][1] = min_reduction_table[abm_config[set][1]];
-	ram_table.max_reduction[0][1] = max_reduction_table[abm_config[set][1]];
-	ram_table.max_reduction[1][1] = max_reduction_table[abm_config[set][1]];
-	ram_table.max_reduction[2][1] = max_reduction_table[abm_config[set][1]];
-	ram_table.max_reduction[3][1] = max_reduction_table[abm_config[set][1]];
-	ram_table.max_reduction[4][1] = max_reduction_table[abm_config[set][1]];
-
-	ram_table.min_reduction[0][2] = min_reduction_table[abm_config[set][2]];
-	ram_table.min_reduction[1][2] = min_reduction_table[abm_config[set][2]];
-	ram_table.min_reduction[2][2] = min_reduction_table[abm_config[set][2]];
-	ram_table.min_reduction[3][2] = min_reduction_table[abm_config[set][2]];
-	ram_table.min_reduction[4][2] = min_reduction_table[abm_config[set][2]];
-	ram_table.max_reduction[0][2] = max_reduction_table[abm_config[set][2]];
-	ram_table.max_reduction[1][2] = max_reduction_table[abm_config[set][2]];
-	ram_table.max_reduction[2][2] = max_reduction_table[abm_config[set][2]];
-	ram_table.max_reduction[3][2] = max_reduction_table[abm_config[set][2]];
-	ram_table.max_reduction[4][2] = max_reduction_table[abm_config[set][2]];
-
-	ram_table.min_reduction[0][3] = min_reduction_table[abm_config[set][3]];
-	ram_table.min_reduction[1][3] = min_reduction_table[abm_config[set][3]];
-	ram_table.min_reduction[2][3] = min_reduction_table[abm_config[set][3]];
-	ram_table.min_reduction[3][3] = min_reduction_table[abm_config[set][3]];
-	ram_table.min_reduction[4][3] = min_reduction_table[abm_config[set][3]];
-	ram_table.max_reduction[0][3] = max_reduction_table[abm_config[set][3]];
-	ram_table.max_reduction[1][3] = max_reduction_table[abm_config[set][3]];
-	ram_table.max_reduction[2][3] = max_reduction_table[abm_config[set][3]];
-	ram_table.max_reduction[3][3] = max_reduction_table[abm_config[set][3]];
-	ram_table.max_reduction[4][3] = max_reduction_table[abm_config[set][3]];
-
-	ram_table.bright_pos_gain[0][0] = 0x20;
-	ram_table.bright_pos_gain[0][1] = 0x20;
-	ram_table.bright_pos_gain[0][2] = 0x20;
-	ram_table.bright_pos_gain[0][3] = 0x20;
-	ram_table.bright_pos_gain[1][0] = 0x20;
-	ram_table.bright_pos_gain[1][1] = 0x20;
-	ram_table.bright_pos_gain[1][2] = 0x20;
-	ram_table.bright_pos_gain[1][3] = 0x20;
-	ram_table.bright_pos_gain[2][0] = 0x20;
-	ram_table.bright_pos_gain[2][1] = 0x20;
-	ram_table.bright_pos_gain[2][2] = 0x20;
-	ram_table.bright_pos_gain[2][3] = 0x20;
-	ram_table.bright_pos_gain[3][0] = 0x20;
-	ram_table.bright_pos_gain[3][1] = 0x20;
-	ram_table.bright_pos_gain[3][2] = 0x20;
-	ram_table.bright_pos_gain[3][3] = 0x20;
-	ram_table.bright_pos_gain[4][0] = 0x20;
-	ram_table.bright_pos_gain[4][1] = 0x20;
-	ram_table.bright_pos_gain[4][2] = 0x20;
-	ram_table.bright_pos_gain[4][3] = 0x20;
-	ram_table.bright_neg_gain[0][1] = 0x00;
-	ram_table.bright_neg_gain[0][2] = 0x00;
-	ram_table.bright_neg_gain[0][3] = 0x00;
-	ram_table.bright_neg_gain[1][0] = 0x00;
-	ram_table.bright_neg_gain[1][1] = 0x00;
-	ram_table.bright_neg_gain[1][2] = 0x00;
-	ram_table.bright_neg_gain[1][3] = 0x00;
-	ram_table.bright_neg_gain[2][0] = 0x00;
-	ram_table.bright_neg_gain[2][1] = 0x00;
-	ram_table.bright_neg_gain[2][2] = 0x00;
-	ram_table.bright_neg_gain[2][3] = 0x00;
-	ram_table.bright_neg_gain[3][0] = 0x00;
-	ram_table.bright_neg_gain[3][1] = 0x00;
-	ram_table.bright_neg_gain[3][2] = 0x00;
-	ram_table.bright_neg_gain[3][3] = 0x00;
-	ram_table.bright_neg_gain[4][0] = 0x00;
-	ram_table.bright_neg_gain[4][1] = 0x00;
-	ram_table.bright_neg_gain[4][2] = 0x00;
-	ram_table.bright_neg_gain[4][3] = 0x00;
-	ram_table.dark_pos_gain[0][0] = 0x00;
-	ram_table.dark_pos_gain[0][1] = 0x00;
-	ram_table.dark_pos_gain[0][2] = 0x00;
-	ram_table.dark_pos_gain[0][3] = 0x00;
-	ram_table.dark_pos_gain[1][0] = 0x00;
-	ram_table.dark_pos_gain[1][1] = 0x00;
-	ram_table.dark_pos_gain[1][2] = 0x00;
-	ram_table.dark_pos_gain[1][3] = 0x00;
-	ram_table.dark_pos_gain[2][0] = 0x00;
-	ram_table.dark_pos_gain[2][1] = 0x00;
-	ram_table.dark_pos_gain[2][2] = 0x00;
-	ram_table.dark_pos_gain[2][3] = 0x00;
-	ram_table.dark_pos_gain[3][0] = 0x00;
-	ram_table.dark_pos_gain[3][1] = 0x00;
-	ram_table.dark_pos_gain[3][2] = 0x00;
-	ram_table.dark_pos_gain[3][3] = 0x00;
-	ram_table.dark_pos_gain[4][0] = 0x00;
-	ram_table.dark_pos_gain[4][1] = 0x00;
-	ram_table.dark_pos_gain[4][2] = 0x00;
-	ram_table.dark_pos_gain[4][3] = 0x00;
-	ram_table.dark_neg_gain[0][0] = 0x00;
-	ram_table.dark_neg_gain[0][1] = 0x00;
-	ram_table.dark_neg_gain[0][2] = 0x00;
-	ram_table.dark_neg_gain[0][3] = 0x00;
-	ram_table.dark_neg_gain[1][0] = 0x00;
-	ram_table.dark_neg_gain[1][1] = 0x00;
-	ram_table.dark_neg_gain[1][2] = 0x00;
-	ram_table.dark_neg_gain[1][3] = 0x00;
-	ram_table.dark_neg_gain[2][0] = 0x00;
-	ram_table.dark_neg_gain[2][1] = 0x00;
-	ram_table.dark_neg_gain[2][2] = 0x00;
-	ram_table.dark_neg_gain[2][3] = 0x00;
-	ram_table.dark_neg_gain[3][0] = 0x00;
-	ram_table.dark_neg_gain[3][1] = 0x00;
-	ram_table.dark_neg_gain[3][2] = 0x00;
-	ram_table.dark_neg_gain[3][3] = 0x00;
-	ram_table.dark_neg_gain[4][0] = 0x00;
-	ram_table.dark_neg_gain[4][1] = 0x00;
-	ram_table.dark_neg_gain[4][2] = 0x00;
-	ram_table.dark_neg_gain[4][3] = 0x00;
-	ram_table.iir_curve[0] = 0x65;
-	ram_table.iir_curve[1] = 0x65;
-	ram_table.iir_curve[2] = 0x65;
-	ram_table.iir_curve[3] = 0x65;
-	ram_table.iir_curve[4] = 0x65;
-	ram_table.crgb_thresh[0] = cpu_to_be16(0x13b6);
-	ram_table.crgb_thresh[1] = cpu_to_be16(0x1648);
-	ram_table.crgb_thresh[2] = cpu_to_be16(0x18e3);
-	ram_table.crgb_thresh[3] = cpu_to_be16(0x1b41);
-	ram_table.crgb_thresh[4] = cpu_to_be16(0x1d46);
-	ram_table.crgb_thresh[5] = cpu_to_be16(0x1f21);
-	ram_table.crgb_thresh[6] = cpu_to_be16(0x2167);
-	ram_table.crgb_thresh[7] = cpu_to_be16(0x2384);
-	ram_table.crgb_offset[0] = cpu_to_be16(0x2999);
-	ram_table.crgb_offset[1] = cpu_to_be16(0x3999);
-	ram_table.crgb_offset[2] = cpu_to_be16(0x4666);
-	ram_table.crgb_offset[3] = cpu_to_be16(0x5999);
-	ram_table.crgb_offset[4] = cpu_to_be16(0x6333);
-	ram_table.crgb_offset[5] = cpu_to_be16(0x7800);
-	ram_table.crgb_offset[6] = cpu_to_be16(0x8c00);
-	ram_table.crgb_offset[7] = cpu_to_be16(0xa000);
-	ram_table.crgb_slope[0]  = cpu_to_be16(0x3147);
-	ram_table.crgb_slope[1]  = cpu_to_be16(0x2978);
-	ram_table.crgb_slope[2]  = cpu_to_be16(0x23a2);
-	ram_table.crgb_slope[3]  = cpu_to_be16(0x1f55);
-	ram_table.crgb_slope[4]  = cpu_to_be16(0x1c63);
-	ram_table.crgb_slope[5]  = cpu_to_be16(0x1a0f);
-	ram_table.crgb_slope[6]  = cpu_to_be16(0x178d);
-	ram_table.crgb_slope[7]  = cpu_to_be16(0x15ab);
+		result = dmcu->funcs->load_iram(
+				dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2);
 
-	fill_backlight_transform_table(
-			params, &ram_table);
+		if (result)
+			result = dmcu->funcs->load_iram(
+					dmcu, IRAM_RESERVE_AREA_END_V2 + 1,
+					(char *)(&ram_table) + IRAM_RESERVE_AREA_END_V2 + 1,
+					sizeof(ram_table) - IRAM_RESERVE_AREA_END_V2 - 1);
+	}
 
-	return dmcu->funcs->load_iram(
-			dmcu, 0, (char *)(&ram_table), sizeof(ram_table));
+	return result;
 }
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_offset.h
index 13d4de645190..d8e0dd192fdd 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_offset.h
@@ -2247,6 +2247,8 @@
 
 // addressBlock: nbio_nbif_rcc_strap_BIFDEC1[13440..14975]
 // base address: 0x3480
+#define mmRCC_BIF_STRAP0                                                                               0x0000
+#define mmRCC_BIF_STRAP0_BASE_IDX                                                                      2
 #define mmRCC_DEV0_EPF0_STRAP0                                                                         0x000f
 #define mmRCC_DEV0_EPF0_STRAP0_BASE_IDX                                                                2
 
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_sh_mask.h
index a02b67943372..29af5167cd00 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_sh_mask.h
@@ -16838,6 +16838,10 @@
 
 
 // addressBlock: nbio_nbif_rcc_strap_BIFDEC1[13440..14975]
+//RCC_BIF_STRAP0
+#define RCC_BIF_STRAP0__STRAP_PX_CAPABLE__SHIFT                                                               0x7
+#define RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK                                                                 0x00000080L
+
 //RCC_DEV0_EPF0_STRAP0
 #define RCC_DEV0_EPF0_STRAP0__STRAP_DEVICE_ID_DEV0_F0__SHIFT                                                  0x0
 #define RCC_DEV0_EPF0_STRAP0__STRAP_MAJOR_REV_ID_DEV0_F0__SHIFT                                               0x10
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_smn.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_smn.h
new file mode 100644
index 000000000000..8c75669eb500
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_6_1_smn.h
@@ -0,0 +1,58 @@
+/*
+ * Copyright (C) 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) 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 _nbio_6_1_SMN_HEADER
+#define _nbio_6_1_SMN_HEADER
+
+
+#define smnCPM_CONTROL					0x11180460
+#define smnPCIE_CNTL2					0x11180070
+#define smnPCIE_CONFIG_CNTL				0x11180044
+#define smnPCIE_CI_CNTL					0x11180080
+
+
+#define smnPCIE_PERF_COUNT_CNTL				0x11180200
+#define smnPCIE_PERF_CNTL_TXCLK				0x11180204
+#define smnPCIE_PERF_COUNT0_TXCLK			0x11180208
+#define smnPCIE_PERF_COUNT1_TXCLK			0x1118020c
+#define smnPCIE_PERF_CNTL_MST_R_CLK			0x11180210
+#define smnPCIE_PERF_COUNT0_MST_R_CLK			0x11180214
+#define smnPCIE_PERF_COUNT1_MST_R_CLK			0x11180218
+#define smnPCIE_PERF_CNTL_MST_C_CLK			0x1118021c
+#define smnPCIE_PERF_COUNT0_MST_C_CLK			0x11180220
+#define smnPCIE_PERF_COUNT1_MST_C_CLK			0x11180224
+#define smnPCIE_PERF_CNTL_SLV_R_CLK			0x11180228
+#define smnPCIE_PERF_COUNT0_SLV_R_CLK			0x1118022c
+#define smnPCIE_PERF_COUNT1_SLV_R_CLK			0x11180230
+#define smnPCIE_PERF_CNTL_SLV_S_C_CLK			0x11180234
+#define smnPCIE_PERF_COUNT0_SLV_S_C_CLK			0x11180238
+#define smnPCIE_PERF_COUNT1_SLV_S_C_CLK			0x1118023c
+#define smnPCIE_PERF_CNTL_SLV_NS_C_CLK			0x11180240
+#define smnPCIE_PERF_COUNT0_SLV_NS_C_CLK		0x11180244
+#define smnPCIE_PERF_COUNT1_SLV_NS_C_CLK		0x11180248
+#define smnPCIE_PERF_CNTL_EVENT0_PORT_SEL		0x1118024c
+#define smnPCIE_PERF_CNTL_EVENT1_PORT_SEL		0x11180250
+#define smnPCIE_PERF_CNTL_TXCLK2			0x11180254
+#define smnPCIE_PERF_COUNT0_TXCLK2			0x11180258
+#define smnPCIE_PERF_COUNT1_TXCLK2			0x1118025c
+
+#endif	// _nbio_6_1_SMN_HEADER
+
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_smn.h
new file mode 100644
index 000000000000..5563f0715896
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_smn.h
@@ -0,0 +1,54 @@
+/*
+ * Copyright (C) 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) 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 _nbio_7_0_SMN_HEADER
+#define _nbio_7_0_SMN_HEADER
+
+
+#define smnCPM_CONTROL					0x11180460
+#define smnPCIE_CNTL2					0x11180070
+
+#define smnPCIE_PERF_COUNT_CNTL				0x11180200
+#define smnPCIE_PERF_CNTL_TXCLK				0x11180204
+#define smnPCIE_PERF_COUNT0_TXCLK			0x11180208
+#define smnPCIE_PERF_COUNT1_TXCLK			0x1118020c
+#define smnPCIE_PERF_CNTL_MST_R_CLK			0x11180210
+#define smnPCIE_PERF_COUNT0_MST_R_CLK			0x11180214
+#define smnPCIE_PERF_COUNT1_MST_R_CLK			0x11180218
+#define smnPCIE_PERF_CNTL_MST_C_CLK			0x1118021c
+#define smnPCIE_PERF_COUNT0_MST_C_CLK			0x11180220
+#define smnPCIE_PERF_COUNT1_MST_C_CLK			0x11180224
+#define smnPCIE_PERF_CNTL_SLV_R_CLK			0x11180228
+#define smnPCIE_PERF_COUNT0_SLV_R_CLK			0x1118022c
+#define smnPCIE_PERF_COUNT1_SLV_R_CLK			0x11180230
+#define smnPCIE_PERF_CNTL_SLV_S_C_CLK			0x11180234
+#define smnPCIE_PERF_COUNT0_SLV_S_C_CLK			0x11180238
+#define smnPCIE_PERF_COUNT1_SLV_S_C_CLK			0x1118023c
+#define smnPCIE_PERF_CNTL_SLV_NS_C_CLK			0x11180240
+#define smnPCIE_PERF_COUNT0_SLV_NS_C_CLK		0x11180244
+#define smnPCIE_PERF_COUNT1_SLV_NS_C_CLK		0x11180248
+#define smnPCIE_PERF_CNTL_EVENT0_PORT_SEL		0x1118024c
+#define smnPCIE_PERF_CNTL_EVENT1_PORT_SEL		0x11180250
+#define smnPCIE_PERF_CNTL_TXCLK2			0x11180254
+#define smnPCIE_PERF_COUNT0_TXCLK2			0x11180258
+#define smnPCIE_PERF_COUNT1_TXCLK2			0x1118025c
+
+#endif	// _nbio_7_0_SMN_HEADER
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_0_smn.h
new file mode 100644
index 000000000000..c1457d880c4d
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_0_smn.h
@@ -0,0 +1,53 @@
+/*
+ * Copyright (C) 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) 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 _nbio_7_4_0_SMN_HEADER
+#define _nbio_7_4_0_SMN_HEADER
+
+
+#define smnNBIF_MGCG_CTRL_LCLK				0x1013a21c
+#define smnCPM_CONTROL					0x11180460
+#define smnPCIE_CNTL2					0x11180070
+#define smnPCIE_CI_CNTL					0x11180080
+
+#define smnPCIE_PERF_COUNT_CNTL				0x11180200
+#define smnPCIE_PERF_CNTL_TXCLK1			0x11180204
+#define smnPCIE_PERF_COUNT0_TXCLK1			0x11180208
+#define smnPCIE_PERF_COUNT1_TXCLK1			0x1118020c
+#define smnPCIE_PERF_CNTL_TXCLK2			0x11180210
+#define smnPCIE_PERF_COUNT0_TXCLK2			0x11180214
+#define smnPCIE_PERF_COUNT1_TXCLK2			0x11180218
+#define smnPCIE_PERF_CNTL_TXCLK3			0x1118021c
+#define smnPCIE_PERF_COUNT0_TXCLK3			0x11180220
+#define smnPCIE_PERF_COUNT1_TXCLK3			0x11180224
+#define smnPCIE_PERF_CNTL_TXCLK4			0x11180228
+#define smnPCIE_PERF_COUNT0_TXCLK4			0x1118022c
+#define smnPCIE_PERF_COUNT1_TXCLK4			0x11180230
+#define smnPCIE_PERF_CNTL_SCLK1				0x11180234
+#define smnPCIE_PERF_COUNT0_SCLK1			0x11180238
+#define smnPCIE_PERF_COUNT1_SCLK1			0x1118023c
+#define smnPCIE_PERF_CNTL_SCLK2				0x11180240
+#define smnPCIE_PERF_COUNT0_SCLK2			0x11180244
+#define smnPCIE_PERF_COUNT1_SCLK2			0x11180248
+#define smnPCIE_PERF_CNTL_EVENT_LC_PORT_SEL		0x1118024c
+#define smnPCIE_PERF_CNTL_EVENT_CI_PORT_SEL		0x11180250
+
+#endif	// _nbio_7_4_0_SMN_HEADER
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h
index e932213f87f0..994e796a28d7 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h
@@ -2567,6 +2567,8 @@
 
 // addressBlock: nbio_nbif0_rcc_strap_BIFDEC1
 // base address: 0x0
+#define mmRCC_BIF_STRAP0                                                                               0x0000
+#define mmRCC_BIF_STRAP0_BASE_IDX                                                                      2
 #define mmRCC_DEV0_EPF0_STRAP0                                                                         0x0011
 #define mmRCC_DEV0_EPF0_STRAP0_BASE_IDX                                                                2
 
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_sh_mask.h
index d3704b438f2d..d467b939c971 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_sh_mask.h
@@ -19690,6 +19690,9 @@
 
 
 // addressBlock: nbio_nbif0_rcc_strap_BIFDEC1
+//RCC_BIF_STRAP0
+#define RCC_BIF_STRAP0__STRAP_PX_CAPABLE__SHIFT                                                               0x7
+#define RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK                                                                 0x00000080L
 //RCC_DEV0_EPF0_STRAP0
 #define RCC_DEV0_EPF0_STRAP0__STRAP_DEVICE_ID_DEV0_F0__SHIFT                                                  0x0
 #define RCC_DEV0_EPF0_STRAP0__STRAP_MAJOR_REV_ID_DEV0_F0__SHIFT                                               0x10
diff --git a/drivers/gpu/drm/amd/include/asic_reg/thm/thm_11_0_2_offset.h b/drivers/gpu/drm/amd/include/asic_reg/thm/thm_11_0_2_offset.h
index a9eb57a53e59..a485526f3a51 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/thm/thm_11_0_2_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/thm/thm_11_0_2_offset.h
@@ -46,4 +46,7 @@
 #define mmTHM_TCON_THERM_TRIP                                                                          0x0002
 #define mmTHM_TCON_THERM_TRIP_BASE_IDX                                                                 0
 
+#define mmTHM_BACO_CNTL                                                                                0x0081
+#define mmTHM_BACO_CNTL_BASE_IDX                                                                       0
+
 #endif
diff --git a/drivers/gpu/drm/amd/include/atombios.h b/drivers/gpu/drm/amd/include/atombios.h
index 7931502fa54f..8ba21747b40a 100644
--- a/drivers/gpu/drm/amd/include/atombios.h
+++ b/drivers/gpu/drm/amd/include/atombios.h
@@ -4106,7 +4106,7 @@ typedef struct  _ATOM_LCD_MODE_CONTROL_CAP
 typedef struct _ATOM_FAKE_EDID_PATCH_RECORD
 {
   UCHAR ucRecordType;
-  UCHAR ucFakeEDIDLength;       // = 128 means EDID lenght is 128 bytes, otherwise the EDID length = ucFakeEDIDLength*128
+  UCHAR ucFakeEDIDLength;       // = 128 means EDID length is 128 bytes, otherwise the EDID length = ucFakeEDIDLength*128
   UCHAR ucFakeEDIDString[1];    // This actually has ucFakeEdidLength elements.
 } ATOM_FAKE_EDID_PATCH_RECORD;
 
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 8154d67388cc..83d960110d23 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -34,7 +34,6 @@
 
 struct pci_dev;
 
-#define KFD_INTERFACE_VERSION 2
 #define KGD_MAX_QUEUES 128
 
 struct kfd_dev;
@@ -330,56 +329,4 @@ struct kfd2kgd_calls {
 
 };
 
-/**
- * struct kgd2kfd_calls
- *
- * @exit: Notifies amdkfd that kgd module is unloaded
- *
- * @probe: Notifies amdkfd about a probe done on a device in the kgd driver.
- *
- * @device_init: Initialize the newly probed device (if it is a device that
- * amdkfd supports)
- *
- * @device_exit: Notifies amdkfd about a removal of a kgd device
- *
- * @suspend: Notifies amdkfd about a suspend action done to a kgd device
- *
- * @resume: Notifies amdkfd about a resume action done to a kgd device
- *
- * @quiesce_mm: Quiesce all user queue access to specified MM address space
- *
- * @resume_mm: Resume user queue access to specified MM address space
- *
- * @schedule_evict_and_restore_process: Schedules work queue that will prepare
- * for safe eviction of KFD BOs that belong to the specified process.
- *
- * @pre_reset: Notifies amdkfd that amdgpu about to reset the gpu
- *
- * @post_reset: Notify amdkfd that amgpu successfully reseted the gpu
- *
- * This structure contains function callback pointers so the kgd driver
- * will notify to the amdkfd about certain status changes.
- *
- */
-struct kgd2kfd_calls {
-	void (*exit)(void);
-	struct kfd_dev* (*probe)(struct kgd_dev *kgd, struct pci_dev *pdev,
-		const struct kfd2kgd_calls *f2g);
-	bool (*device_init)(struct kfd_dev *kfd,
-			const struct kgd2kfd_shared_resources *gpu_resources);
-	void (*device_exit)(struct kfd_dev *kfd);
-	void (*interrupt)(struct kfd_dev *kfd, const void *ih_ring_entry);
-	void (*suspend)(struct kfd_dev *kfd);
-	int (*resume)(struct kfd_dev *kfd);
-	int (*quiesce_mm)(struct mm_struct *mm);
-	int (*resume_mm)(struct mm_struct *mm);
-	int (*schedule_evict_and_restore_process)(struct mm_struct *mm,
-			struct dma_fence *fence);
-	int  (*pre_reset)(struct kfd_dev *kfd);
-	int  (*post_reset)(struct kfd_dev *kfd);
-};
-
-int kgd2kfd_init(unsigned interface_version,
-		const struct kgd2kfd_calls **g2f);
-
 #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 1479ea1dc3e7..2b579ba9b685 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -92,6 +92,9 @@ enum pp_clock_type {
 	PP_SCLK,
 	PP_MCLK,
 	PP_PCIE,
+	PP_SOCCLK,
+	PP_FCLK,
+	PP_DCEFCLK,
 	OD_SCLK,
 	OD_MCLK,
 	OD_VDDC_CURVE,
@@ -127,12 +130,13 @@ enum amd_pp_task {
 };
 
 enum PP_SMC_POWER_PROFILE {
-	PP_SMC_POWER_PROFILE_FULLSCREEN3D = 0x0,
-	PP_SMC_POWER_PROFILE_POWERSAVING  = 0x1,
-	PP_SMC_POWER_PROFILE_VIDEO        = 0x2,
-	PP_SMC_POWER_PROFILE_VR           = 0x3,
-	PP_SMC_POWER_PROFILE_COMPUTE      = 0x4,
-	PP_SMC_POWER_PROFILE_CUSTOM       = 0x5,
+	PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT = 0x0,
+	PP_SMC_POWER_PROFILE_FULLSCREEN3D = 0x1,
+	PP_SMC_POWER_PROFILE_POWERSAVING  = 0x2,
+	PP_SMC_POWER_PROFILE_VIDEO        = 0x3,
+	PP_SMC_POWER_PROFILE_VR           = 0x4,
+	PP_SMC_POWER_PROFILE_COMPUTE      = 0x5,
+	PP_SMC_POWER_PROFILE_CUSTOM       = 0x6,
 };
 
 enum {
@@ -280,6 +284,11 @@ struct amd_pm_funcs {
 	int (*set_hard_min_dcefclk_by_freq)(void *handle, uint32_t clock);
 	int (*set_hard_min_fclk_by_freq)(void *handle, uint32_t clock);
 	int (*set_min_deep_sleep_dcefclk)(void *handle, uint32_t clock);
+	int (*get_asic_baco_capability)(void *handle, bool *cap);
+	int (*get_asic_baco_state)(void *handle, int *state);
+	int (*set_asic_baco_state)(void *handle, int state);
+	int (*get_ppfeature_status)(void *handle, char *buf);
+	int (*set_ppfeature_status)(void *handle, uint64_t ppfeature_masks);
 };
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index 9bc27f468d5b..3f73f7cd18b9 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -1404,6 +1404,97 @@ static int pp_set_active_display_count(void *handle, uint32_t count)
 	return ret;
 }
 
+static int pp_get_asic_baco_capability(void *handle, bool *cap)
+{
+	struct pp_hwmgr *hwmgr = handle;
+
+	if (!hwmgr)
+		return -EINVAL;
+
+	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_capability)
+		return 0;
+
+	mutex_lock(&hwmgr->smu_lock);
+	hwmgr->hwmgr_func->get_asic_baco_capability(hwmgr, cap);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return 0;
+}
+
+static int pp_get_asic_baco_state(void *handle, int *state)
+{
+	struct pp_hwmgr *hwmgr = handle;
+
+	if (!hwmgr)
+		return -EINVAL;
+
+	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_state)
+		return 0;
+
+	mutex_lock(&hwmgr->smu_lock);
+	hwmgr->hwmgr_func->get_asic_baco_state(hwmgr, (enum BACO_STATE *)state);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return 0;
+}
+
+static int pp_set_asic_baco_state(void *handle, int state)
+{
+	struct pp_hwmgr *hwmgr = handle;
+
+	if (!hwmgr)
+		return -EINVAL;
+
+	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_asic_baco_state)
+		return 0;
+
+	mutex_lock(&hwmgr->smu_lock);
+	hwmgr->hwmgr_func->set_asic_baco_state(hwmgr, (enum BACO_STATE)state);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return 0;
+}
+
+static int pp_get_ppfeature_status(void *handle, char *buf)
+{
+	struct pp_hwmgr *hwmgr = handle;
+	int ret = 0;
+
+	if (!hwmgr || !hwmgr->pm_en || !buf)
+		return -EINVAL;
+
+	if (hwmgr->hwmgr_func->get_ppfeature_status == NULL) {
+		pr_info_ratelimited("%s was not implemented.\n", __func__);
+		return -EINVAL;
+	}
+
+	mutex_lock(&hwmgr->smu_lock);
+	ret = hwmgr->hwmgr_func->get_ppfeature_status(hwmgr, buf);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return ret;
+}
+
+static int pp_set_ppfeature_status(void *handle, uint64_t ppfeature_masks)
+{
+	struct pp_hwmgr *hwmgr = handle;
+	int ret = 0;
+
+	if (!hwmgr || !hwmgr->pm_en)
+		return -EINVAL;
+
+	if (hwmgr->hwmgr_func->set_ppfeature_status == NULL) {
+		pr_info_ratelimited("%s was not implemented.\n", __func__);
+		return -EINVAL;
+	}
+
+	mutex_lock(&hwmgr->smu_lock);
+	ret = hwmgr->hwmgr_func->set_ppfeature_status(hwmgr, ppfeature_masks);
+	mutex_unlock(&hwmgr->smu_lock);
+
+	return ret;
+}
+
 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,
@@ -1454,4 +1545,9 @@ static const struct amd_pm_funcs pp_dpm_funcs = {
 	.set_min_deep_sleep_dcefclk = pp_set_min_deep_sleep_dcefclk,
 	.set_hard_min_dcefclk_by_freq = pp_set_hard_min_dcefclk_by_freq,
 	.set_hard_min_fclk_by_freq = pp_set_hard_min_fclk_by_freq,
+	.get_asic_baco_capability = pp_get_asic_baco_capability,
+	.get_asic_baco_state = pp_get_asic_baco_state,
+	.set_asic_baco_state = pp_set_asic_baco_state,
+	.get_ppfeature_status = pp_get_ppfeature_status,
+	.set_ppfeature_status = pp_set_ppfeature_status,
 };
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile b/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
index ade8973b6f4d..0b3c6d1d52e4 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
@@ -35,7 +35,7 @@ HARDWARE_MGR = hwmgr.o processpptables.o \
 		vega12_thermal.o \
 		pp_overdriver.o smu_helper.o \
 		vega20_processpptables.o vega20_hwmgr.o vega20_powertune.o \
-		vega20_thermal.o
+		vega20_thermal.o common_baco.o vega10_baco.o  vega20_baco.o
 
 AMD_PP_HWMGR = $(addprefix $(AMD_PP_PATH)/hwmgr/,$(HARDWARE_MGR))
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.c
new file mode 100644
index 000000000000..9c57c1f67749
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.c
@@ -0,0 +1,101 @@
+/*
+ * Copyright 2018 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 "common_baco.h"
+
+
+static bool baco_wait_register(struct pp_hwmgr *hwmgr, u32 reg, u32 mask, u32 value)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	u32 timeout = 5000, data;
+
+	do {
+		msleep(1);
+		data = RREG32(reg);
+		timeout--;
+	} while (value != (data & mask) && (timeout != 0));
+
+	if (timeout == 0)
+		return false;
+
+	return true;
+}
+
+static bool baco_cmd_handler(struct pp_hwmgr *hwmgr, u32 command, u32 reg, u32 mask,
+			        u32 shift, u32 value, u32 timeout)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	u32 data;
+	bool ret = true;
+
+	switch (command) {
+	case CMD_WRITE:
+		WREG32(reg, value << shift);
+		break;
+	case CMD_READMODIFYWRITE:
+		data = RREG32(reg);
+		data = (data & (~mask)) | (value << shift);
+		WREG32(reg, data);
+		break;
+	case CMD_WAITFOR:
+		ret = baco_wait_register(hwmgr, reg, mask, value);
+		break;
+	case CMD_DELAY_MS:
+		if (timeout)
+			/* Delay in milli Seconds */
+			msleep(timeout);
+		break;
+	case CMD_DELAY_US:
+		if (timeout)
+			/* Delay in micro Seconds */
+			udelay(timeout);
+		break;
+
+	default:
+		dev_warn(adev->dev, "Invalid BACO command.\n");
+		ret = false;
+	}
+
+	return ret;
+}
+
+bool soc15_baco_program_registers(struct pp_hwmgr *hwmgr,
+				 const struct soc15_baco_cmd_entry *entry,
+				 const u32 array_size)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	u32 i, reg = 0;
+
+	for (i = 0; i < array_size; i++) {
+		if ((entry[i].cmd == CMD_WRITE) ||
+		    (entry[i].cmd == CMD_READMODIFYWRITE) ||
+		    (entry[i].cmd == CMD_WAITFOR))
+			reg = adev->reg_offset[entry[i].hwip][entry[i].inst][entry[i].seg]
+				+ entry[i].reg_offset;
+		if (!baco_cmd_handler(hwmgr, entry[i].cmd, reg, entry[i].mask,
+				     entry[i].shift, entry[i].val, entry[i].timeout))
+			return false;
+	}
+
+	return true;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_sw_engine_dce80.h b/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.h
index 26355c088746..95296c916f4e 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2c_sw_engine_dce80.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/common_baco.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-15 Advanced Micro Devices, Inc.
+ * Copyright 2018 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"),
@@ -19,25 +19,32 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  *
- * Authors: AMD
- *
  */
+#ifndef __COMMON_BOCO_H__
+#define __COMMON_BOCO_H__
+#include "hwmgr.h"
 
-#ifndef __DAL_I2C_SW_ENGINE_DCE80_H__
-#define __DAL_I2C_SW_ENGINE_DCE80_H__
 
-struct i2c_sw_engine_dce80 {
-	struct i2c_sw_engine base;
-	uint32_t engine_id;
+enum baco_cmd_type {
+	CMD_WRITE = 0,
+	CMD_READMODIFYWRITE,
+	CMD_WAITFOR,
+	CMD_DELAY_MS,
+	CMD_DELAY_US,
 };
 
-struct i2c_sw_engine_dce80_create_arg {
-	uint32_t engine_id;
-	uint32_t default_speed;
-	struct dc_context *ctx;
+struct soc15_baco_cmd_entry {
+	enum baco_cmd_type cmd;
+	uint32_t 	hwip;
+	uint32_t 	inst;
+	uint32_t 	seg;
+	uint32_t 	reg_offset;
+	uint32_t     	mask;
+	uint32_t     	shift;
+	uint32_t     	timeout;
+	uint32_t     	val;
 };
-
-struct i2c_engine *dal_i2c_sw_engine_dce80_create(
-	const struct i2c_sw_engine_dce80_create_arg *arg);
-
+extern bool soc15_baco_program_registers(struct pp_hwmgr *hwmgr,
+					const struct soc15_baco_cmd_entry *entry,
+					const u32 array_size);
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
index 1f92a9f4c9e3..c1c51c115e57 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
@@ -154,15 +154,6 @@ int phm_powerdown_uvd(struct pp_hwmgr *hwmgr)
 	return 0;
 }
 
-int phm_enable_clock_power_gatings(struct pp_hwmgr *hwmgr)
-{
-	PHM_FUNC_CHECK(hwmgr);
-
-	if (NULL != hwmgr->hwmgr_func->enable_clock_power_gating)
-		return hwmgr->hwmgr_func->enable_clock_power_gating(hwmgr);
-
-	return 0;
-}
 
 int phm_disable_clock_power_gatings(struct pp_hwmgr *hwmgr)
 {
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
index 0173d0480024..6cd6497c6fc2 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
@@ -64,17 +64,19 @@ static int ci_set_asic_special_caps(struct pp_hwmgr *hwmgr);
 
 static void hwmgr_init_workload_prority(struct pp_hwmgr *hwmgr)
 {
-	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D] = 2;
-	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_POWERSAVING] = 0;
-	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VIDEO] = 1;
-	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VR] = 3;
-	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_COMPUTE] = 4;
-
-	hwmgr->workload_setting[0] = PP_SMC_POWER_PROFILE_POWERSAVING;
-	hwmgr->workload_setting[1] = PP_SMC_POWER_PROFILE_VIDEO;
-	hwmgr->workload_setting[2] = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
-	hwmgr->workload_setting[3] = PP_SMC_POWER_PROFILE_VR;
-	hwmgr->workload_setting[4] = PP_SMC_POWER_PROFILE_COMPUTE;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT] = 0;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D] = 1;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_POWERSAVING] = 2;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VIDEO] = 3;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VR] = 4;
+	hwmgr->workload_prority[PP_SMC_POWER_PROFILE_COMPUTE] = 5;
+
+	hwmgr->workload_setting[0] = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	hwmgr->workload_setting[1] = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
+	hwmgr->workload_setting[2] = PP_SMC_POWER_PROFILE_POWERSAVING;
+	hwmgr->workload_setting[3] = PP_SMC_POWER_PROFILE_VIDEO;
+	hwmgr->workload_setting[4] = PP_SMC_POWER_PROFILE_VR;
+	hwmgr->workload_setting[5] = PP_SMC_POWER_PROFILE_COMPUTE;
 }
 
 int hwmgr_early_init(struct pp_hwmgr *hwmgr)
@@ -271,7 +273,7 @@ int hwmgr_hw_fini(struct pp_hwmgr *hwmgr)
 
 	phm_stop_thermal_controller(hwmgr);
 	psm_set_boot_states(hwmgr);
-	psm_adjust_power_state_dynamic(hwmgr, false, NULL);
+	psm_adjust_power_state_dynamic(hwmgr, true, NULL);
 	phm_disable_dynamic_state_management(hwmgr);
 	phm_disable_clock_power_gatings(hwmgr);
 
@@ -293,7 +295,7 @@ int hwmgr_suspend(struct pp_hwmgr *hwmgr)
 	ret = psm_set_boot_states(hwmgr);
 	if (ret)
 		return ret;
-	ret = psm_adjust_power_state_dynamic(hwmgr, false, NULL);
+	ret = psm_adjust_power_state_dynamic(hwmgr, true, NULL);
 	if (ret)
 		return ret;
 	ret = phm_power_down_asic(hwmgr);
@@ -323,7 +325,7 @@ int hwmgr_resume(struct pp_hwmgr *hwmgr)
 	if (ret)
 		return ret;
 
-	ret = psm_adjust_power_state_dynamic(hwmgr, false, NULL);
+	ret = psm_adjust_power_state_dynamic(hwmgr, true, NULL);
 
 	return ret;
 }
@@ -377,12 +379,12 @@ int hwmgr_handle_task(struct pp_hwmgr *hwmgr, enum amd_pp_task task_id,
 		ret = psm_set_user_performance_state(hwmgr, requested_ui_label, &requested_ps);
 		if (ret)
 			return ret;
-		ret = psm_adjust_power_state_dynamic(hwmgr, false, requested_ps);
+		ret = psm_adjust_power_state_dynamic(hwmgr, true, requested_ps);
 		break;
 	}
 	case AMD_PP_TASK_COMPLETE_INIT:
 	case AMD_PP_TASK_READJUST_POWER_STATE:
-		ret = psm_adjust_power_state_dynamic(hwmgr, false, NULL);
+		ret = psm_adjust_power_state_dynamic(hwmgr, true, NULL);
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c
index 56437866d120..ce177d7f04cb 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c
@@ -256,16 +256,14 @@ static void power_state_management(struct pp_hwmgr *hwmgr,
 	}
 }
 
-int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip,
+int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip_display_settings,
 						struct pp_power_state *new_ps)
 {
 	uint32_t index;
 	long workload;
 
-	if (skip)
-		return 0;
-
-	phm_display_configuration_changed(hwmgr);
+	if (!skip_display_settings)
+		phm_display_configuration_changed(hwmgr);
 
 	if (hwmgr->ps)
 		power_state_management(hwmgr, new_ps);
@@ -276,9 +274,11 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip,
 		 */
 		phm_apply_clock_adjust_rules(hwmgr);
 
-	phm_notify_smc_display_config_after_ps_adjustment(hwmgr);
+	if (!skip_display_settings)
+		phm_notify_smc_display_config_after_ps_adjustment(hwmgr);
 
-	if (!phm_force_dpm_levels(hwmgr, hwmgr->request_dpm_level))
+	if ((hwmgr->request_dpm_level != hwmgr->dpm_level) &&
+	    !phm_force_dpm_levels(hwmgr, hwmgr->request_dpm_level))
 		hwmgr->dpm_level = hwmgr->request_dpm_level;
 
 	if (hwmgr->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) {
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h
index fa1b6825036a..b62d55f1f289 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.h
@@ -34,7 +34,7 @@ int psm_set_user_performance_state(struct pp_hwmgr *hwmgr,
 					enum PP_StateUILabel label_id,
 					struct pp_power_state **state);
 int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr,
-				bool skip,
+				bool skip_display_settings,
 				struct pp_power_state *new_ps);
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
index f95c5f50eb0f..5273de3c5b98 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
@@ -1033,6 +1033,7 @@ static int smu10_get_clock_by_type_with_latency(struct pp_hwmgr *hwmgr,
 		break;
 	case amd_pp_dpp_clock:
 		pclk_vol_table = pinfo->vdd_dep_on_dppclk;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
index d91390459326..c8f5c00dd1e7 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
@@ -77,8 +77,9 @@
 #define PCIE_BUS_CLK                10000
 #define TCLK                        (PCIE_BUS_CLK / 10)
 
-static const struct profile_mode_setting smu7_profiling[6] =
-					{{1, 0, 100, 30, 1, 0, 100, 10},
+static const struct profile_mode_setting smu7_profiling[7] =
+					{{0, 0, 0, 0, 0, 0, 0, 0},
+					 {1, 0, 100, 30, 1, 0, 100, 10},
 					 {1, 10, 0, 30, 0, 0, 0, 0},
 					 {0, 0, 0, 0, 1, 10, 16, 31},
 					 {1, 0, 11, 50, 1, 0, 100, 10},
@@ -4889,7 +4890,8 @@ static int smu7_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf)
 	uint32_t i, size = 0;
 	uint32_t len;
 
-	static const char *profile_name[6] = {"3D_FULL_SCREEN",
+	static const char *profile_name[7] = {"BOOTUP_DEFAULT",
+					"3D_FULL_SCREEN",
 					"POWER_SAVING",
 					"VIDEO",
 					"VR",
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
new file mode 100644
index 000000000000..f94dab27f486
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
@@ -0,0 +1,136 @@
+#include "amdgpu.h"
+#include "soc15.h"
+#include "soc15_hw_ip.h"
+#include "vega10_ip_offset.h"
+#include "soc15_common.h"
+#include "vega10_inc.h"
+#include "vega10_ppsmc.h"
+#include "vega10_baco.h"
+
+
+
+static const struct soc15_baco_cmd_entry  pre_baco_tbl[] =
+{
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIF_DOORBELL_CNTL), BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN_MASK, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN__SHIFT, 0, 1},
+	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIF_FB_EN), 0, 0, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DSTATE_BYPASS_MASK, BACO_CNTL__BACO_DSTATE_BYPASS__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_RST_INTR_MASK_MASK, BACO_CNTL__BACO_RST_INTR_MASK__SHIFT, 0, 1}
+};
+
+static const struct soc15_baco_cmd_entry enter_baco_tbl[] =
+{
+	{CMD_WAITFOR, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__SOC_DOMAIN_IDLE_MASK, THM_BACO_CNTL__SOC_DOMAIN_IDLE__SHIFT, 0xffffffff, 0x80000000},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_VDCI_RESET_MASK, THM_BACO_CNTL__BACO_SOC_VDCI_RESET__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT,0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT,0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK,     THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 1},
+	{CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 5, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_RESET_EN_MASK, THM_BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_PWROKRAW_CNTL_MASK, THM_BACO_CNTL__BACO_PWROKRAW_CNTL__SHIFT, 0, 0},
+	{CMD_WAITFOR, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_MODE_MASK, BACO_CNTL__BACO_MODE__SHIFT, 0xffffffff, 0x100}
+};
+
+static const struct soc15_baco_cmd_entry exit_baco_tbl[] =
+{
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0},
+	{CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 10,0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0,0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT,0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_PWROKRAW_CNTL_MASK, THM_BACO_CNTL__BACO_PWROKRAW_CNTL__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_VDCI_RESET_MASK, THM_BACO_CNTL__BACO_SOC_VDCI_RESET__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_EXIT_MASK, THM_BACO_CNTL__BACO_EXIT__SHIFT, 0, 1},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_RESET_EN_MASK, THM_BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0},
+	{CMD_WAITFOR, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_EXIT_MASK, 0, 0xffffffff, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SB_AXI_FENCE_MASK, THM_BACO_CNTL__BACO_SB_AXI_FENCE__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT,  0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK ,BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 0},
+	{CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_EN_MASK , BACO_CNTL__BACO_EN__SHIFT, 0,0},
+	{CMD_WAITFOR, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0}
+ };
+
+static const struct soc15_baco_cmd_entry clean_baco_tbl[] =
+{
+	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_6), 0, 0, 0, 0},
+	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_7), 0, 0, 0, 0},
+};
+
+int vega10_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg, data;
+
+	*cap = false;
+	if (!phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_BACO))
+		return 0;
+
+	WREG32(0x12074, 0xFFF0003B);
+	data = RREG32(0x12075);
+
+	if (data == 0x1) {
+		reg = RREG32_SOC15(NBIF, 0, mmRCC_BIF_STRAP0);
+
+		if (reg & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK)
+			*cap = true;
+	}
+
+	return 0;
+}
+
+int vega10_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg;
+
+	reg = RREG32_SOC15(NBIF, 0, mmBACO_CNTL);
+
+	if (reg & BACO_CNTL__BACO_MODE_MASK)
+		/* gfx has already entered BACO state */
+		*state = BACO_STATE_IN;
+	else
+		*state = BACO_STATE_OUT;
+	return 0;
+}
+
+int vega10_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
+{
+	enum BACO_STATE cur_state;
+
+	vega10_baco_get_state(hwmgr, &cur_state);
+
+	if (cur_state == state)
+		/* aisc already in the target state */
+		return 0;
+
+	if (state == BACO_STATE_IN) {
+		if (soc15_baco_program_registers(hwmgr, pre_baco_tbl,
+					     ARRAY_SIZE(pre_baco_tbl))) {
+			if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnterBaco))
+				return -1;
+
+			if (soc15_baco_program_registers(hwmgr, enter_baco_tbl,
+						   ARRAY_SIZE(enter_baco_tbl)))
+				return 0;
+		}
+	} else if (state == BACO_STATE_OUT) {
+		/* HW requires at least 20ms between regulator off and on */
+		msleep(20);
+		/* Execute Hardware BACO exit sequence */
+		if (soc15_baco_program_registers(hwmgr, exit_baco_tbl,
+					     ARRAY_SIZE(exit_baco_tbl))) {
+			if (soc15_baco_program_registers(hwmgr, clean_baco_tbl,
+						     ARRAY_SIZE(clean_baco_tbl)))
+				return 0;
+		}
+	}
+
+	return -1;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2caux_dce80.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
index 21908629e973..a93b1e6d1c66 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce80/i2caux_dce80.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-15 Advanced Micro Devices, Inc.
+ * Copyright 2018 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"),
@@ -19,20 +19,14 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  *
- * Authors: AMD
- *
  */
+#ifndef __VEGA10_BOCO_H__
+#define __VEGA10_BOCO_H__
+#include "hwmgr.h"
+#include "common_baco.h"
 
-#ifndef __DAL_I2C_AUX_DCE80_H__
-#define __DAL_I2C_AUX_DCE80_H__
-
-struct i2caux_dce80 {
-	struct i2caux base;
-	/* indicate the I2C HW circular buffer is in use */
-	bool i2c_hw_buffer_in_use;
-};
-
-struct i2caux *dal_i2caux_dce80_create(
-	struct dc_context *ctx);
+extern int vega10_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap);
+extern int vega10_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
+extern int vega10_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
index 79c86247d0ac..5479125ff4f6 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
@@ -48,6 +48,7 @@
 #include "ppinterrupt.h"
 #include "pp_overdriver.h"
 #include "pp_thermal.h"
+#include "vega10_baco.h"
 
 #include "smuio/smuio_9_0_offset.h"
 #include "smuio/smuio_9_0_sh_mask.h"
@@ -71,6 +72,21 @@ static const uint32_t channel_number[] = {1, 2, 0, 4, 0, 8, 0, 16, 2};
 #define DF_CS_AON0_DramBaseAddress0__IntLvAddrSel_MASK                                                        0x00000700L
 #define DF_CS_AON0_DramBaseAddress0__DramBaseAddr_MASK                                                        0xFFFFF000L
 
+typedef enum {
+	CLK_SMNCLK = 0,
+	CLK_SOCCLK,
+	CLK_MP0CLK,
+	CLK_MP1CLK,
+	CLK_LCLK,
+	CLK_DCEFCLK,
+	CLK_VCLK,
+	CLK_DCLK,
+	CLK_ECLK,
+	CLK_UCLK,
+	CLK_GFXCLK,
+	CLK_COUNT,
+} CLOCK_ID_e;
+
 static const ULONG PhwVega10_Magic = (ULONG)(PHM_VIslands_Magic);
 
 struct vega10_power_state *cast_phw_vega10_power_state(
@@ -804,9 +820,9 @@ static int vega10_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 
 	hwmgr->backend = data;
 
-	hwmgr->workload_mask = 1 << hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VIDEO];
-	hwmgr->power_profile_mode = PP_SMC_POWER_PROFILE_VIDEO;
-	hwmgr->default_power_profile_mode = PP_SMC_POWER_PROFILE_VIDEO;
+	hwmgr->workload_mask = 1 << hwmgr->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT];
+	hwmgr->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	hwmgr->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 
 	vega10_set_default_registry_data(hwmgr);
 	data->disable_dpm_mask = 0xff;
@@ -3485,6 +3501,17 @@ static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr)
 		}
 	}
 
+	if (!data->registry_data.socclk_dpm_key_disabled) {
+		if (data->smc_state_table.soc_boot_level !=
+				data->dpm_table.soc_table.dpm_state.soft_min_level) {
+			smum_send_msg_to_smc_with_parameter(hwmgr,
+				PPSMC_MSG_SetSoftMinSocclkByIndex,
+				data->smc_state_table.soc_boot_level);
+			data->dpm_table.soc_table.dpm_state.soft_min_level =
+					data->smc_state_table.soc_boot_level;
+		}
+	}
+
 	return 0;
 }
 
@@ -3516,6 +3543,17 @@ static int vega10_upload_dpm_max_level(struct pp_hwmgr *hwmgr)
 		}
 	}
 
+	if (!data->registry_data.socclk_dpm_key_disabled) {
+		if (data->smc_state_table.soc_max_level !=
+			data->dpm_table.soc_table.dpm_state.soft_max_level) {
+			smum_send_msg_to_smc_with_parameter(hwmgr,
+				PPSMC_MSG_SetSoftMaxSocclkByIndex,
+				data->smc_state_table.soc_max_level);
+			data->dpm_table.soc_table.dpm_state.soft_max_level =
+					data->smc_state_table.soc_max_level;
+		}
+	}
+
 	return 0;
 }
 
@@ -3541,6 +3579,10 @@ static int vega10_generate_dpm_level_enable_mask(
 			vega10_find_lowest_dpm_level(&(data->dpm_table.mem_table));
 	data->smc_state_table.mem_max_level =
 			vega10_find_highest_dpm_level(&(data->dpm_table.mem_table));
+	data->smc_state_table.soc_boot_level =
+			vega10_find_lowest_dpm_level(&(data->dpm_table.soc_table));
+	data->smc_state_table.soc_max_level =
+			vega10_find_highest_dpm_level(&(data->dpm_table.soc_table));
 
 	PP_ASSERT_WITH_CODE(!vega10_upload_dpm_bootup_level(hwmgr),
 			"Attempt to upload DPM Bootup Levels Failed!",
@@ -3555,6 +3597,9 @@ static int vega10_generate_dpm_level_enable_mask(
 	for(i = data->smc_state_table.mem_boot_level; i < data->smc_state_table.mem_max_level; i++)
 		data->dpm_table.mem_table.dpm_levels[i].enabled = true;
 
+	for (i = data->smc_state_table.soc_boot_level; i < data->smc_state_table.soc_max_level; i++)
+		data->dpm_table.soc_table.dpm_levels[i].enabled = true;
+
 	return 0;
 }
 
@@ -4028,6 +4073,24 @@ static int vega10_force_clock_level(struct pp_hwmgr *hwmgr,
 
 		break;
 
+	case PP_SOCCLK:
+		data->smc_state_table.soc_boot_level = mask ? (ffs(mask) - 1) : 0;
+		data->smc_state_table.soc_max_level = mask ? (fls(mask) - 1) : 0;
+
+		PP_ASSERT_WITH_CODE(!vega10_upload_dpm_bootup_level(hwmgr),
+			"Failed to upload boot level to lowest!",
+			return -EINVAL);
+
+		PP_ASSERT_WITH_CODE(!vega10_upload_dpm_max_level(hwmgr),
+			"Failed to upload dpm max level to highest!",
+			return -EINVAL);
+
+		break;
+
+	case PP_DCEFCLK:
+		pr_info("Setting DCEFCLK min/max dpm level is not supported!\n");
+		break;
+
 	case PP_PCIE:
 	default:
 		break;
@@ -4267,12 +4330,113 @@ static int vega10_set_watermarks_for_clocks_ranges(struct pp_hwmgr *hwmgr,
 	return result;
 }
 
+static int vega10_get_ppfeature_status(struct pp_hwmgr *hwmgr, char *buf)
+{
+	static const char *ppfeature_name[] = {
+				"DPM_PREFETCHER",
+				"GFXCLK_DPM",
+				"UCLK_DPM",
+				"SOCCLK_DPM",
+				"UVD_DPM",
+				"VCE_DPM",
+				"ULV",
+				"MP0CLK_DPM",
+				"LINK_DPM",
+				"DCEFCLK_DPM",
+				"AVFS",
+				"GFXCLK_DS",
+				"SOCCLK_DS",
+				"LCLK_DS",
+				"PPT",
+				"TDC",
+				"THERMAL",
+				"GFX_PER_CU_CG",
+				"RM",
+				"DCEFCLK_DS",
+				"ACDC",
+				"VR0HOT",
+				"VR1HOT",
+				"FW_CTF",
+				"LED_DISPLAY",
+				"FAN_CONTROL",
+				"FAST_PPT",
+				"DIDT",
+				"ACG",
+				"PCC_LIMIT"};
+	static const char *output_title[] = {
+				"FEATURES",
+				"BITMASK",
+				"ENABLEMENT"};
+	uint64_t features_enabled;
+	int i;
+	int ret = 0;
+	int size = 0;
+
+	ret = vega10_get_enabled_smc_features(hwmgr, &features_enabled);
+	PP_ASSERT_WITH_CODE(!ret,
+			"[EnableAllSmuFeatures] Failed to get enabled smc features!",
+			return ret);
+
+	size += sprintf(buf + size, "Current ppfeatures: 0x%016llx\n", features_enabled);
+	size += sprintf(buf + size, "%-19s %-22s %s\n",
+				output_title[0],
+				output_title[1],
+				output_title[2]);
+	for (i = 0; i < GNLD_FEATURES_MAX; i++) {
+		size += sprintf(buf + size, "%-19s 0x%016llx %6s\n",
+					ppfeature_name[i],
+					1ULL << i,
+					(features_enabled & (1ULL << i)) ? "Y" : "N");
+	}
+
+	return size;
+}
+
+static int vega10_set_ppfeature_status(struct pp_hwmgr *hwmgr, uint64_t new_ppfeature_masks)
+{
+	uint64_t features_enabled;
+	uint64_t features_to_enable;
+	uint64_t features_to_disable;
+	int ret = 0;
+
+	if (new_ppfeature_masks >= (1ULL << GNLD_FEATURES_MAX))
+		return -EINVAL;
+
+	ret = vega10_get_enabled_smc_features(hwmgr, &features_enabled);
+	if (ret)
+		return ret;
+
+	features_to_disable =
+		(features_enabled ^ new_ppfeature_masks) & features_enabled;
+	features_to_enable =
+		(features_enabled ^ new_ppfeature_masks) ^ features_to_disable;
+
+	pr_debug("features_to_disable 0x%llx\n", features_to_disable);
+	pr_debug("features_to_enable 0x%llx\n", features_to_enable);
+
+	if (features_to_disable) {
+		ret = vega10_enable_smc_features(hwmgr, false, features_to_disable);
+		if (ret)
+			return ret;
+	}
+
+	if (features_to_enable) {
+		ret = vega10_enable_smc_features(hwmgr, true, features_to_enable);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, char *buf)
 {
 	struct vega10_hwmgr *data = hwmgr->backend;
 	struct vega10_single_dpm_table *sclk_table = &(data->dpm_table.gfx_table);
 	struct vega10_single_dpm_table *mclk_table = &(data->dpm_table.mem_table);
+	struct vega10_single_dpm_table *soc_table = &(data->dpm_table.soc_table);
+	struct vega10_single_dpm_table *dcef_table = &(data->dpm_table.dcef_table);
 	struct vega10_pcie_table *pcie_table = &(data->dpm_table.pcie_table);
 	struct vega10_odn_clock_voltage_dependency_table *podn_vdd_dep = NULL;
 
@@ -4303,6 +4467,32 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr,
 					i, mclk_table->dpm_levels[i].value / 100,
 					(i == now) ? "*" : "");
 		break;
+	case PP_SOCCLK:
+		if (data->registry_data.socclk_dpm_key_disabled)
+			break;
+
+		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentSocclkIndex);
+		now = smum_get_argument(hwmgr);
+
+		for (i = 0; i < soc_table->count; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+					i, soc_table->dpm_levels[i].value / 100,
+					(i == now) ? "*" : "");
+		break;
+	case PP_DCEFCLK:
+		if (data->registry_data.dcefclk_dpm_key_disabled)
+			break;
+
+		smum_send_msg_to_smc_with_parameter(hwmgr,
+				PPSMC_MSG_GetClockFreqMHz, CLK_DCEFCLK);
+		now = smum_get_argument(hwmgr);
+
+		for (i = 0; i < dcef_table->count; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+					i, dcef_table->dpm_levels[i].value / 100,
+					(dcef_table->dpm_levels[i].value / 100 == now) ?
+					"*" : "");
+		break;
 	case PP_PCIE:
 		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentLinkIndex);
 		now = smum_get_argument(hwmgr);
@@ -4668,13 +4858,15 @@ static int vega10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf)
 {
 	struct vega10_hwmgr *data = hwmgr->backend;
 	uint32_t i, size = 0;
-	static const uint8_t profile_mode_setting[5][4] = {{70, 60, 1, 3,},
+	static const uint8_t profile_mode_setting[6][4] = {{70, 60, 0, 0,},
+						{70, 60, 1, 3,},
 						{90, 60, 0, 0,},
 						{70, 60, 0, 0,},
 						{70, 90, 0, 0,},
 						{30, 60, 0, 6,},
 						};
-	static const char *profile_name[6] = {"3D_FULL_SCREEN",
+	static const char *profile_name[7] = {"BOOTUP_DEFAULT",
+					"3D_FULL_SCREEN",
 					"POWER_SAVING",
 					"VIDEO",
 					"VR",
@@ -4978,6 +5170,12 @@ static const struct pp_hwmgr_func vega10_hwmgr_funcs = {
 	.set_power_limit = vega10_set_power_limit,
 	.odn_edit_dpm_table = vega10_odn_edit_dpm_table,
 	.get_performance_level = vega10_get_performance_level,
+	.get_asic_baco_capability = vega10_baco_get_capability,
+	.get_asic_baco_state = vega10_baco_get_state,
+	.set_asic_baco_state = vega10_baco_set_state,
+	.enable_mgpu_fan_boost = vega10_enable_mgpu_fan_boost,
+	.get_ppfeature_status = vega10_get_ppfeature_status,
+	.set_ppfeature_status = vega10_set_ppfeature_status,
 };
 
 int vega10_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h
index 89870556de1b..f752b4ad0c8a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h
@@ -199,6 +199,7 @@ struct vega10_smc_state_table {
 	uint32_t        vce_boot_level;
 	uint32_t        gfx_max_level;
 	uint32_t        mem_max_level;
+	uint32_t        soc_max_level;
 	uint8_t         vr_hot_gpio;
 	uint8_t         ac_dc_gpio;
 	uint8_t         therm_out_gpio;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h
index b3e63003a789..c934e9612c1b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_pptable.h
@@ -282,6 +282,30 @@ typedef struct _ATOM_Vega10_Fan_Table_V2 {
 	UCHAR   ucFanMaxRPM;
 } ATOM_Vega10_Fan_Table_V2;
 
+typedef struct _ATOM_Vega10_Fan_Table_V3 {
+	UCHAR   ucRevId;
+	USHORT  usFanOutputSensitivity;
+	USHORT  usFanAcousticLimitRpm;
+	USHORT  usThrottlingRPM;
+	USHORT  usTargetTemperature;
+	USHORT  usMinimumPWMLimit;
+	USHORT  usTargetGfxClk;
+	USHORT  usFanGainEdge;
+	USHORT  usFanGainHotspot;
+	USHORT  usFanGainLiquid;
+	USHORT  usFanGainVrVddc;
+	USHORT  usFanGainVrMvdd;
+	USHORT  usFanGainPlx;
+	USHORT  usFanGainHbm;
+	UCHAR   ucEnableZeroRPM;
+	USHORT  usFanStopTemperature;
+	USHORT  usFanStartTemperature;
+	UCHAR   ucFanParameters;
+	UCHAR   ucFanMinRPM;
+	UCHAR   ucFanMaxRPM;
+	USHORT  usMGpuThrottlingRPM;
+} ATOM_Vega10_Fan_Table_V3;
+
 typedef struct _ATOM_Vega10_Thermal_Controller {
 	UCHAR ucRevId;
 	UCHAR ucType;           /* one of ATOM_VEGA10_PP_THERMALCONTROLLER_*/
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c
index b8747a5c9204..b6767d74dc85 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c
@@ -32,6 +32,7 @@
 #include "vega10_pptable.h"
 
 #define NUM_DSPCLK_LEVELS 8
+#define VEGA10_ENGINECLOCK_HARDMAX 198000
 
 static void set_hw_cap(struct pp_hwmgr *hwmgr, bool enable,
 		enum phm_platform_caps cap)
@@ -122,6 +123,7 @@ static int init_thermal_controller(
 	const Vega10_PPTable_Generic_SubTable_Header *header;
 	const ATOM_Vega10_Fan_Table *fan_table_v1;
 	const ATOM_Vega10_Fan_Table_V2 *fan_table_v2;
+	const ATOM_Vega10_Fan_Table_V3 *fan_table_v3;
 
 	thermal_controller = (ATOM_Vega10_Thermal_Controller *)
 			(((unsigned long)powerplay_table) +
@@ -206,7 +208,7 @@ static int init_thermal_controller(
 				le16_to_cpu(fan_table_v1->usFanStopTemperature);
 		hwmgr->thermal_controller.advanceFanControlParameters.usZeroRPMStartTemperature =
 				le16_to_cpu(fan_table_v1->usFanStartTemperature);
-	} else if (header->ucRevId > 10) {
+	} else if (header->ucRevId == 0xb) {
 		fan_table_v2 = (ATOM_Vega10_Fan_Table_V2 *)header;
 
 		hwmgr->thermal_controller.fanInfo.ucTachometerPulsesPerRevolution =
@@ -250,7 +252,54 @@ static int init_thermal_controller(
 				le16_to_cpu(fan_table_v2->usFanStopTemperature);
 		hwmgr->thermal_controller.advanceFanControlParameters.usZeroRPMStartTemperature =
 				le16_to_cpu(fan_table_v2->usFanStartTemperature);
+	} else if (header->ucRevId > 0xb) {
+		fan_table_v3 = (ATOM_Vega10_Fan_Table_V3 *)header;
+
+		hwmgr->thermal_controller.fanInfo.ucTachometerPulsesPerRevolution =
+				fan_table_v3->ucFanParameters & ATOM_VEGA10_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK;
+		hwmgr->thermal_controller.fanInfo.ulMinRPM = fan_table_v3->ucFanMinRPM * 100UL;
+		hwmgr->thermal_controller.fanInfo.ulMaxRPM = fan_table_v3->ucFanMaxRPM * 100UL;
+		phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+				PHM_PlatformCaps_MicrocodeFanControl);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanOutputSensitivity =
+				le16_to_cpu(fan_table_v3->usFanOutputSensitivity);
+		hwmgr->thermal_controller.advanceFanControlParameters.usMaxFanRPM =
+				fan_table_v3->ucFanMaxRPM * 100UL;
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanRPMMaxLimit =
+				le16_to_cpu(fan_table_v3->usThrottlingRPM);
+		hwmgr->thermal_controller.advanceFanControlParameters.ulMinFanSCLKAcousticLimit =
+				le16_to_cpu(fan_table_v3->usFanAcousticLimitRpm);
+		hwmgr->thermal_controller.advanceFanControlParameters.usTMax =
+				le16_to_cpu(fan_table_v3->usTargetTemperature);
+		hwmgr->thermal_controller.advanceFanControlParameters.usPWMMin =
+				le16_to_cpu(fan_table_v3->usMinimumPWMLimit);
+		hwmgr->thermal_controller.advanceFanControlParameters.ulTargetGfxClk =
+				le16_to_cpu(fan_table_v3->usTargetGfxClk);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainEdge =
+				le16_to_cpu(fan_table_v3->usFanGainEdge);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainHotspot =
+				le16_to_cpu(fan_table_v3->usFanGainHotspot);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainLiquid =
+				le16_to_cpu(fan_table_v3->usFanGainLiquid);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainVrVddc =
+				le16_to_cpu(fan_table_v3->usFanGainVrVddc);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainVrMvdd =
+				le16_to_cpu(fan_table_v3->usFanGainVrMvdd);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainPlx =
+				le16_to_cpu(fan_table_v3->usFanGainPlx);
+		hwmgr->thermal_controller.advanceFanControlParameters.usFanGainHbm =
+				le16_to_cpu(fan_table_v3->usFanGainHbm);
+
+		hwmgr->thermal_controller.advanceFanControlParameters.ucEnableZeroRPM =
+				fan_table_v3->ucEnableZeroRPM;
+		hwmgr->thermal_controller.advanceFanControlParameters.usZeroRPMStopTemperature =
+				le16_to_cpu(fan_table_v3->usFanStopTemperature);
+		hwmgr->thermal_controller.advanceFanControlParameters.usZeroRPMStartTemperature =
+				le16_to_cpu(fan_table_v3->usFanStartTemperature);
+		hwmgr->thermal_controller.advanceFanControlParameters.usMGpuThrottlingRPMLimit =
+				le16_to_cpu(fan_table_v3->usMGpuThrottlingRPM);
 	}
+
 	return 0;
 }
 
@@ -258,7 +307,26 @@ static int init_over_drive_limits(
 		struct pp_hwmgr *hwmgr,
 		const ATOM_Vega10_POWERPLAYTABLE *powerplay_table)
 {
-	hwmgr->platform_descriptor.overdriveLimit.engineClock =
+	const ATOM_Vega10_GFXCLK_Dependency_Table *gfxclk_dep_table =
+			(const ATOM_Vega10_GFXCLK_Dependency_Table *)
+			(((unsigned long) powerplay_table) +
+			le16_to_cpu(powerplay_table->usGfxclkDependencyTableOffset));
+	bool is_acg_enabled = false;
+	ATOM_Vega10_GFXCLK_Dependency_Record_V2 *patom_record_v2;
+
+	if (gfxclk_dep_table->ucRevId == 1) {
+		patom_record_v2 =
+			(ATOM_Vega10_GFXCLK_Dependency_Record_V2 *)gfxclk_dep_table->entries;
+		is_acg_enabled =
+			(bool)patom_record_v2[gfxclk_dep_table->ucNumEntries-1].ucACGEnable;
+	}
+
+	if (powerplay_table->ulMaxODEngineClock > VEGA10_ENGINECLOCK_HARDMAX &&
+		!is_acg_enabled)
+		hwmgr->platform_descriptor.overdriveLimit.engineClock =
+			VEGA10_ENGINECLOCK_HARDMAX;
+	else
+		hwmgr->platform_descriptor.overdriveLimit.engineClock =
 			le32_to_cpu(powerplay_table->ulMaxODEngineClock);
 	hwmgr->platform_descriptor.overdriveLimit.memoryClock =
 			le32_to_cpu(powerplay_table->ulMaxODMemoryClock);
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
index 3f807d6c95ce..ba8763daa380 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
@@ -556,6 +556,43 @@ int vega10_thermal_setup_fan_table(struct pp_hwmgr *hwmgr)
 	return ret;
 }
 
+int vega10_enable_mgpu_fan_boost(struct pp_hwmgr *hwmgr)
+{
+	struct vega10_hwmgr *data = hwmgr->backend;
+	PPTable_t *table = &(data->smc_state_table.pp_table);
+	int ret;
+
+	if (!data->smu_features[GNLD_FAN_CONTROL].supported)
+		return 0;
+
+	if (!hwmgr->thermal_controller.advanceFanControlParameters.
+			usMGpuThrottlingRPMLimit)
+		return 0;
+
+	table->FanThrottlingRpm = hwmgr->thermal_controller.
+			advanceFanControlParameters.usMGpuThrottlingRPMLimit;
+
+	ret = smum_smc_table_manager(hwmgr,
+				(uint8_t *)(&(data->smc_state_table.pp_table)),
+				PPTABLE, false);
+	if (ret) {
+		pr_info("Failed to update fan control table in pptable!");
+		return ret;
+	}
+
+	ret = vega10_disable_fan_control_feature(hwmgr);
+	if (ret) {
+		pr_info("Attempt to disable SMC fan control feature failed!");
+		return ret;
+	}
+
+	ret = vega10_enable_fan_control_feature(hwmgr);
+	if (ret)
+		pr_info("Attempt to enable SMC fan control feature failed!");
+
+	return ret;
+}
+
 /**
 * Start the fan control on the SMC.
 * @param    hwmgr  the address of the powerplay hardware manager.
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h
index 21e7c4dfa2ca..4a0ede7c1f07 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.h
@@ -73,6 +73,7 @@ extern int vega10_thermal_disable_alert(struct pp_hwmgr *hwmgr);
 extern int vega10_fan_ctrl_start_smc_fan_control(struct pp_hwmgr *hwmgr);
 extern int vega10_start_thermal_controller(struct pp_hwmgr *hwmgr,
 				struct PP_TemperatureRange *range);
+extern int vega10_enable_mgpu_fan_boost(struct pp_hwmgr *hwmgr);
 
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
index 54364444ecd1..6c8e78611c03 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
@@ -753,6 +753,22 @@ static int vega12_init_smc_table(struct pp_hwmgr *hwmgr)
 	return 0;
 }
 
+static int vega12_run_acg_btc(struct pp_hwmgr *hwmgr)
+{
+	uint32_t result;
+
+	PP_ASSERT_WITH_CODE(
+		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc) == 0,
+		"[Run_ACG_BTC] Attempt to run ACG BTC failed!",
+		return -EINVAL);
+
+	result = smum_get_argument(hwmgr);
+	PP_ASSERT_WITH_CODE(result == 1,
+			"Failed to run ACG BTC!", return -EINVAL);
+
+	return 0;
+}
+
 static int vega12_set_allowed_featuresmask(struct pp_hwmgr *hwmgr)
 {
 	struct vega12_hwmgr *data =
@@ -931,6 +947,11 @@ static int vega12_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
 			"Failed to initialize SMC table!",
 			result = tmp_result);
 
+	tmp_result = vega12_run_acg_btc(hwmgr);
+	PP_ASSERT_WITH_CODE(!tmp_result,
+			"Failed to run ACG BTC!",
+			result = tmp_result);
+
 	result = vega12_enable_all_smu_features(hwmgr);
 	PP_ASSERT_WITH_CODE(!result,
 			"Failed to enable all smu features!",
@@ -1072,6 +1093,16 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr)
 					return ret);
 	}
 
+	if (data->smu_features[GNLD_DPM_DCEFCLK].enabled) {
+		min_freq = data->dpm_table.dcef_table.dpm_state.hard_min_level;
+
+		PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(
+					hwmgr, PPSMC_MSG_SetHardMinByFreq,
+					(PPCLK_DCEFCLK << 16) | (min_freq & 0xffff))),
+					"Failed to set hard min dcefclk!",
+					return ret);
+	}
+
 	return ret;
 
 }
@@ -1797,7 +1828,7 @@ static int vega12_force_clock_level(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, uint32_t mask)
 {
 	struct vega12_hwmgr *data = (struct vega12_hwmgr *)(hwmgr->backend);
-	uint32_t soft_min_level, soft_max_level;
+	uint32_t soft_min_level, soft_max_level, hard_min_level;
 	int ret = 0;
 
 	switch (type) {
@@ -1842,6 +1873,56 @@ static int vega12_force_clock_level(struct pp_hwmgr *hwmgr,
 
 		break;
 
+	case PP_SOCCLK:
+		soft_min_level = mask ? (ffs(mask) - 1) : 0;
+		soft_max_level = mask ? (fls(mask) - 1) : 0;
+
+		if (soft_max_level >= data->dpm_table.soc_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level,
+					data->dpm_table.soc_table.count - 1);
+			return -EINVAL;
+		}
+
+		data->dpm_table.soc_table.dpm_state.soft_min_level =
+			data->dpm_table.soc_table.dpm_levels[soft_min_level].value;
+		data->dpm_table.soc_table.dpm_state.soft_max_level =
+			data->dpm_table.soc_table.dpm_levels[soft_max_level].value;
+
+		ret = vega12_upload_dpm_min_level(hwmgr);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload boot level to lowest!",
+			return ret);
+
+		ret = vega12_upload_dpm_max_level(hwmgr);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload dpm max level to highest!",
+			return ret);
+
+		break;
+
+	case PP_DCEFCLK:
+		hard_min_level = mask ? (ffs(mask) - 1) : 0;
+
+		if (hard_min_level >= data->dpm_table.dcef_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					hard_min_level,
+					data->dpm_table.dcef_table.count - 1);
+			return -EINVAL;
+		}
+
+		data->dpm_table.dcef_table.dpm_state.hard_min_level =
+			data->dpm_table.dcef_table.dpm_levels[hard_min_level].value;
+
+		ret = vega12_upload_dpm_min_level(hwmgr);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload boot level to lowest!",
+			return ret);
+
+		//TODO: Setting DCEFCLK max dpm level is not supported
+
+		break;
+
 	case PP_PCIE:
 		break;
 
@@ -1852,6 +1933,104 @@ static int vega12_force_clock_level(struct pp_hwmgr *hwmgr,
 	return 0;
 }
 
+static int vega12_get_ppfeature_status(struct pp_hwmgr *hwmgr, char *buf)
+{
+	static const char *ppfeature_name[] = {
+			"DPM_PREFETCHER",
+			"GFXCLK_DPM",
+			"UCLK_DPM",
+			"SOCCLK_DPM",
+			"UVD_DPM",
+			"VCE_DPM",
+			"ULV",
+			"MP0CLK_DPM",
+			"LINK_DPM",
+			"DCEFCLK_DPM",
+			"GFXCLK_DS",
+			"SOCCLK_DS",
+			"LCLK_DS",
+			"PPT",
+			"TDC",
+			"THERMAL",
+			"GFX_PER_CU_CG",
+			"RM",
+			"DCEFCLK_DS",
+			"ACDC",
+			"VR0HOT",
+			"VR1HOT",
+			"FW_CTF",
+			"LED_DISPLAY",
+			"FAN_CONTROL",
+			"DIDT",
+			"GFXOFF",
+			"CG",
+			"ACG"};
+	static const char *output_title[] = {
+			"FEATURES",
+			"BITMASK",
+			"ENABLEMENT"};
+	uint64_t features_enabled;
+	int i;
+	int ret = 0;
+	int size = 0;
+
+	ret = vega12_get_enabled_smc_features(hwmgr, &features_enabled);
+	PP_ASSERT_WITH_CODE(!ret,
+		"[EnableAllSmuFeatures] Failed to get enabled smc features!",
+		return ret);
+
+	size += sprintf(buf + size, "Current ppfeatures: 0x%016llx\n", features_enabled);
+	size += sprintf(buf + size, "%-19s %-22s %s\n",
+				output_title[0],
+				output_title[1],
+				output_title[2]);
+	for (i = 0; i < GNLD_FEATURES_MAX; i++) {
+		size += sprintf(buf + size, "%-19s 0x%016llx %6s\n",
+				ppfeature_name[i],
+				1ULL << i,
+				(features_enabled & (1ULL << i)) ? "Y" : "N");
+	}
+
+	return size;
+}
+
+static int vega12_set_ppfeature_status(struct pp_hwmgr *hwmgr, uint64_t new_ppfeature_masks)
+{
+	uint64_t features_enabled;
+	uint64_t features_to_enable;
+	uint64_t features_to_disable;
+	int ret = 0;
+
+	if (new_ppfeature_masks >= (1ULL << GNLD_FEATURES_MAX))
+		return -EINVAL;
+
+	ret = vega12_get_enabled_smc_features(hwmgr, &features_enabled);
+	if (ret)
+		return ret;
+
+	features_to_disable =
+		(features_enabled ^ new_ppfeature_masks) & features_enabled;
+	features_to_enable =
+		(features_enabled ^ new_ppfeature_masks) ^ features_to_disable;
+
+	pr_debug("features_to_disable 0x%llx\n", features_to_disable);
+	pr_debug("features_to_enable 0x%llx\n", features_to_enable);
+
+	if (features_to_disable) {
+		ret = vega12_enable_smc_features(hwmgr, false, features_to_disable);
+		if (ret)
+			return ret;
+	}
+
+	if (features_to_enable) {
+		ret = vega12_enable_smc_features(hwmgr, true, features_to_enable);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int vega12_print_clock_levels(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, char *buf)
 {
@@ -1891,6 +2070,42 @@ static int vega12_print_clock_levels(struct pp_hwmgr *hwmgr,
 				(clocks.data[i].clocks_in_khz / 1000 == now / 100) ? "*" : "");
 		break;
 
+	case PP_SOCCLK:
+		PP_ASSERT_WITH_CODE(
+				smum_send_msg_to_smc_with_parameter(hwmgr,
+					PPSMC_MSG_GetDpmClockFreq, (PPCLK_SOCCLK << 16)) == 0,
+				"Attempt to get Current SOCCLK Frequency Failed!",
+				return -EINVAL);
+		now = smum_get_argument(hwmgr);
+
+		PP_ASSERT_WITH_CODE(
+				vega12_get_socclocks(hwmgr, &clocks) == 0,
+				"Attempt to get soc clk levels Failed!",
+				return -1);
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz / 1000 == now) ? "*" : "");
+		break;
+
+	case PP_DCEFCLK:
+		PP_ASSERT_WITH_CODE(
+				smum_send_msg_to_smc_with_parameter(hwmgr,
+					PPSMC_MSG_GetDpmClockFreq, (PPCLK_DCEFCLK << 16)) == 0,
+				"Attempt to get Current DCEFCLK Frequency Failed!",
+				return -EINVAL);
+		now = smum_get_argument(hwmgr);
+
+		PP_ASSERT_WITH_CODE(
+				vega12_get_dcefclocks(hwmgr, &clocks) == 0,
+				"Attempt to get dcef clk levels Failed!",
+				return -1);
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz / 1000 == now) ? "*" : "");
+		break;
+
 	case PP_PCIE:
 		break;
 
@@ -2411,6 +2626,8 @@ static const struct pp_hwmgr_func vega12_hwmgr_funcs = {
 	.start_thermal_controller = vega12_start_thermal_controller,
 	.powergate_gfx = vega12_gfx_off_control,
 	.get_performance_level = vega12_get_performance_level,
+	.get_ppfeature_status = vega12_get_ppfeature_status,
+	.set_ppfeature_status = vega12_set_ppfeature_status,
 };
 
 int vega12_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
new file mode 100644
index 000000000000..0d883b358df2
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
@@ -0,0 +1,81 @@
+#include "amdgpu.h"
+#include "soc15.h"
+#include "soc15_hw_ip.h"
+#include "soc15_common.h"
+#include "vega20_inc.h"
+#include "vega20_ppsmc.h"
+#include "vega20_baco.h"
+
+
+
+static const struct soc15_baco_cmd_entry clean_baco_tbl[] =
+{
+	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_6), 0, 0, 0, 0},
+	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_7), 0, 0, 0, 0},
+};
+
+int vega20_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg;
+
+	*cap = false;
+	if (!phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_BACO))
+		return 0;
+
+	if (((RREG32(0x17569) & 0x20000000) >> 29) == 0x1) {
+		reg = RREG32_SOC15(NBIF, 0, mmRCC_BIF_STRAP0);
+
+		if (reg & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK)
+			*cap = true;
+	}
+
+	return 0;
+}
+
+int vega20_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg;
+
+	reg = RREG32_SOC15(NBIF, 0, mmBACO_CNTL);
+
+	if (reg & BACO_CNTL__BACO_MODE_MASK)
+		/* gfx has already entered BACO state */
+		*state = BACO_STATE_IN;
+	else
+		*state = BACO_STATE_OUT;
+	return 0;
+}
+
+int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	enum BACO_STATE cur_state;
+	uint32_t data;
+
+	vega20_baco_get_state(hwmgr, &cur_state);
+
+	if (cur_state == state)
+		/* aisc already in the target state */
+		return 0;
+
+	if (state == BACO_STATE_IN) {
+		data = RREG32_SOC15(THM, 0, mmTHM_BACO_CNTL);
+		data |= 0x80000000;
+		WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data);
+
+
+		if(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0))
+			return -1;
+
+	} else if (state == BACO_STATE_OUT) {
+		if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ExitBaco))
+			return -1;
+		if (!soc15_baco_program_registers(hwmgr, clean_baco_tbl,
+						     ARRAY_SIZE(clean_baco_tbl)))
+			return -1;
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce100/i2caux_dce100.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
index 2b508d3e0ef4..c51988a9ed77 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce100/i2caux_dce100.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-15 Advanced Micro Devices, Inc.
+ * Copyright 2018 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"),
@@ -19,14 +19,14 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  *
- * Authors: AMD
- *
  */
+#ifndef __VEGA20_BOCO_H__
+#define __VEGA20_BOCO_H__
+#include "hwmgr.h"
+#include "common_baco.h"
 
-#ifndef __DAL_I2C_AUX_DCE100_H__
-#define __DAL_I2C_AUX_DCE100_H__
-
-struct i2caux *dal_i2caux_dce100_create(
-	struct dc_context *ctx);
+extern int vega20_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap);
+extern int vega20_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
+extern int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
 
-#endif /* __DAL_I2C_AUX_DCE100_H__ */
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
index 26154f9b2178..0769b1ec562b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
@@ -47,6 +47,7 @@
 #include "pp_overdriver.h"
 #include "pp_thermal.h"
 #include "soc15_common.h"
+#include "vega20_baco.h"
 #include "smuio/smuio_9_0_offset.h"
 #include "smuio/smuio_9_0_sh_mask.h"
 #include "nbio/nbio_7_4_sh_mask.h"
@@ -390,9 +391,9 @@ static int vega20_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 
 	hwmgr->backend = data;
 
-	hwmgr->workload_mask = 1 << hwmgr->workload_prority[PP_SMC_POWER_PROFILE_VIDEO];
-	hwmgr->power_profile_mode = PP_SMC_POWER_PROFILE_VIDEO;
-	hwmgr->default_power_profile_mode = PP_SMC_POWER_PROFILE_VIDEO;
+	hwmgr->workload_mask = 1 << hwmgr->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT];
+	hwmgr->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	hwmgr->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 
 	vega20_set_default_registry_data(hwmgr);
 
@@ -770,6 +771,54 @@ static int vega20_init_smc_table(struct pp_hwmgr *hwmgr)
 	return 0;
 }
 
+/*
+ * Override PCIe link speed and link width for DPM Level 1. PPTable entries
+ * reflect the ASIC capabilities and not the system capabilities. For e.g.
+ * Vega20 board in a PCI Gen3 system. In this case, when SMU's tries to switch
+ * to DPM1, it fails as system doesn't support Gen4.
+ */
+static int vega20_override_pcie_parameters(struct pp_hwmgr *hwmgr)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t pcie_gen = 0, pcie_width = 0, smu_pcie_arg;
+	int ret;
+
+	if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
+		pcie_gen = 3;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
+		pcie_gen = 2;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
+		pcie_gen = 1;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
+		pcie_gen = 0;
+
+	if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
+		pcie_width = 6;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
+		pcie_width = 5;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
+		pcie_width = 4;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
+		pcie_width = 3;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
+		pcie_width = 2;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
+		pcie_width = 1;
+
+	/* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
+	 * Bit 15:8:  PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
+	 * Bit 7:0:   PCIE lane width, 1 to 7 corresponds is x1 to x32
+	 */
+	smu_pcie_arg = (1 << 16) | (pcie_gen << 8) | pcie_width;
+	ret = smum_send_msg_to_smc_with_parameter(hwmgr,
+			PPSMC_MSG_OverridePcieParameters, smu_pcie_arg);
+	PP_ASSERT_WITH_CODE(!ret,
+		"[OverridePcieParameters] Attempt to override pcie params failed!",
+		return ret);
+
+	return 0;
+}
+
 static int vega20_set_allowed_featuresmask(struct pp_hwmgr *hwmgr)
 {
 	struct vega20_hwmgr *data =
@@ -803,6 +852,11 @@ static int vega20_set_allowed_featuresmask(struct pp_hwmgr *hwmgr)
 	return 0;
 }
 
+static int vega20_run_btc(struct pp_hwmgr *hwmgr)
+{
+	return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunBtc);
+}
+
 static int vega20_run_btc_afll(struct pp_hwmgr *hwmgr)
 {
 	return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAfllBtc);
@@ -980,6 +1034,9 @@ static int vega20_od8_set_feature_capabilities(
 	    pp_table->FanZeroRpmEnable)
 		od_settings->overdrive8_capabilities |= OD8_FAN_ZERO_RPM_CONTROL;
 
+	if (!od_settings->overdrive8_capabilities)
+		hwmgr->od_enabled = false;
+
 	return 0;
 }
 
@@ -1561,6 +1618,11 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
 			"[EnableDPMTasks] Failed to initialize SMC table!",
 			return result);
 
+	result = vega20_run_btc(hwmgr);
+	PP_ASSERT_WITH_CODE(!result,
+			"[EnableDPMTasks] Failed to run btc!",
+			return result);
+
 	result = vega20_run_btc_afll(hwmgr);
 	PP_ASSERT_WITH_CODE(!result,
 			"[EnableDPMTasks] Failed to run btc afll!",
@@ -1571,6 +1633,11 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
 			"[EnableDPMTasks] Failed to enable all smu features!",
 			return result);
 
+	result = vega20_override_pcie_parameters(hwmgr);
+	PP_ASSERT_WITH_CODE(!result,
+			"[EnableDPMTasks] Failed to override pcie parameters!",
+			return result);
+
 	result = vega20_notify_smc_display_change(hwmgr);
 	PP_ASSERT_WITH_CODE(!result,
 			"[EnableDPMTasks] Failed to notify smc display change!",
@@ -1689,13 +1756,6 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_
 					(PPCLK_UCLK << 16) | (min_freq & 0xffff))),
 					"Failed to set soft min memclk !",
 					return ret);
-
-		min_freq = data->dpm_table.mem_table.dpm_state.hard_min_level;
-		PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(
-					hwmgr, PPSMC_MSG_SetHardMinByFreq,
-					(PPCLK_UCLK << 16) | (min_freq & 0xffff))),
-					"Failed to set hard min memclk !",
-					return ret);
 	}
 
 	if (data->smu_features[GNLD_DPM_UVD].enabled &&
@@ -1739,6 +1799,28 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_
 					return ret);
 	}
 
+	if (data->smu_features[GNLD_DPM_FCLK].enabled &&
+	   (feature_mask & FEATURE_DPM_FCLK_MASK)) {
+		min_freq = data->dpm_table.fclk_table.dpm_state.soft_min_level;
+
+		PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(
+					hwmgr, PPSMC_MSG_SetSoftMinByFreq,
+					(PPCLK_FCLK << 16) | (min_freq & 0xffff))),
+					"Failed to set soft min fclk!",
+					return ret);
+	}
+
+	if (data->smu_features[GNLD_DPM_DCEFCLK].enabled &&
+	   (feature_mask & FEATURE_DPM_DCEFCLK_MASK)) {
+		min_freq = data->dpm_table.dcef_table.dpm_state.hard_min_level;
+
+		PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(
+					hwmgr, PPSMC_MSG_SetHardMinByFreq,
+					(PPCLK_DCEFCLK << 16) | (min_freq & 0xffff))),
+					"Failed to set hard min dcefclk!",
+					return ret);
+	}
+
 	return ret;
 }
 
@@ -1811,6 +1893,17 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_
 					return ret);
 	}
 
+	if (data->smu_features[GNLD_DPM_FCLK].enabled &&
+	   (feature_mask & FEATURE_DPM_FCLK_MASK)) {
+		max_freq = data->dpm_table.fclk_table.dpm_state.soft_max_level;
+
+		PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(
+					hwmgr, PPSMC_MSG_SetSoftMaxByFreq,
+					(PPCLK_FCLK << 16) | (max_freq & 0xffff))),
+					"Failed to set soft max fclk!",
+					return ret);
+	}
+
 	return ret;
 }
 
@@ -1918,16 +2011,36 @@ 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)
+{
+	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 (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
+		memcpy(metrics_table, &data->metrics_table, sizeof(SmuMetrics_t));
+
+	return ret;
+}
+
 static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr,
 		uint32_t *query)
 {
 	int ret = 0;
 	SmuMetrics_t metrics_table;
 
-	ret = smum_smc_table_manager(hwmgr, (uint8_t *)&metrics_table, TABLE_SMU_METRICS, true);
-	PP_ASSERT_WITH_CODE(!ret,
-			"Failed to export SMU METRICS table!",
-			return ret);
+	ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+	if (ret)
+		return ret;
 
 	*query = metrics_table.CurrSocketPower << 8;
 
@@ -1958,10 +2071,9 @@ static int vega20_get_current_activity_percent(struct pp_hwmgr *hwmgr,
 	int ret = 0;
 	SmuMetrics_t metrics_table;
 
-	ret = smum_smc_table_manager(hwmgr, (uint8_t *)&metrics_table, TABLE_SMU_METRICS, true);
-	PP_ASSERT_WITH_CODE(!ret,
-			"Failed to export SMU METRICS table!",
-			return ret);
+	ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+	if (ret)
+		return ret;
 
 	*activity_percent = metrics_table.AverageGfxActivity;
 
@@ -1973,16 +2085,18 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx,
 {
 	struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend);
 	struct amdgpu_device *adev = hwmgr->adev;
+	SmuMetrics_t metrics_table;
 	uint32_t val_vid;
 	int ret = 0;
 
 	switch (idx) {
 	case AMDGPU_PP_SENSOR_GFX_SCLK:
-		ret = vega20_get_current_clk_freq(hwmgr,
-				PPCLK_GFXCLK,
-				(uint32_t *)value);
-		if (!ret)
-			*size = 4;
+		ret = vega20_get_metrics_table(hwmgr, &metrics_table);
+		if (ret)
+			return ret;
+
+		*((uint32_t *)value) = metrics_table.AverageGfxclkFrequency * 100;
+		*size = 4;
 		break;
 	case AMDGPU_PP_SENSOR_GFX_MCLK:
 		ret = vega20_get_current_clk_freq(hwmgr,
@@ -2140,6 +2254,12 @@ static int vega20_force_dpm_highest(struct pp_hwmgr *hwmgr)
 		data->dpm_table.mem_table.dpm_state.soft_max_level =
 		data->dpm_table.mem_table.dpm_levels[soft_level].value;
 
+	soft_level = vega20_find_highest_dpm_level(&(data->dpm_table.soc_table));
+
+	data->dpm_table.soc_table.dpm_state.soft_min_level =
+		data->dpm_table.soc_table.dpm_state.soft_max_level =
+		data->dpm_table.soc_table.dpm_levels[soft_level].value;
+
 	ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF);
 	PP_ASSERT_WITH_CODE(!ret,
 			"Failed to upload boot level to highest!",
@@ -2172,6 +2292,12 @@ static int vega20_force_dpm_lowest(struct pp_hwmgr *hwmgr)
 		data->dpm_table.mem_table.dpm_state.soft_max_level =
 		data->dpm_table.mem_table.dpm_levels[soft_level].value;
 
+	soft_level = vega20_find_lowest_dpm_level(&(data->dpm_table.soc_table));
+
+	data->dpm_table.soc_table.dpm_state.soft_min_level =
+		data->dpm_table.soc_table.dpm_state.soft_max_level =
+		data->dpm_table.soc_table.dpm_levels[soft_level].value;
+
 	ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF);
 	PP_ASSERT_WITH_CODE(!ret,
 			"Failed to upload boot level to highest!",
@@ -2188,8 +2314,32 @@ static int vega20_force_dpm_lowest(struct pp_hwmgr *hwmgr)
 
 static int vega20_unforce_dpm_levels(struct pp_hwmgr *hwmgr)
 {
+	struct vega20_hwmgr *data =
+			(struct vega20_hwmgr *)(hwmgr->backend);
+	uint32_t soft_min_level, soft_max_level;
 	int ret = 0;
 
+	soft_min_level = vega20_find_lowest_dpm_level(&(data->dpm_table.gfx_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(data->dpm_table.gfx_table));
+	data->dpm_table.gfx_table.dpm_state.soft_min_level =
+		data->dpm_table.gfx_table.dpm_levels[soft_min_level].value;
+	data->dpm_table.gfx_table.dpm_state.soft_max_level =
+		data->dpm_table.gfx_table.dpm_levels[soft_max_level].value;
+
+	soft_min_level = vega20_find_lowest_dpm_level(&(data->dpm_table.mem_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(data->dpm_table.mem_table));
+	data->dpm_table.mem_table.dpm_state.soft_min_level =
+		data->dpm_table.mem_table.dpm_levels[soft_min_level].value;
+	data->dpm_table.mem_table.dpm_state.soft_max_level =
+		data->dpm_table.mem_table.dpm_levels[soft_max_level].value;
+
+	soft_min_level = vega20_find_lowest_dpm_level(&(data->dpm_table.soc_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(data->dpm_table.soc_table));
+	data->dpm_table.soc_table.dpm_state.soft_min_level =
+		data->dpm_table.soc_table.dpm_levels[soft_min_level].value;
+	data->dpm_table.soc_table.dpm_state.soft_max_level =
+		data->dpm_table.soc_table.dpm_levels[soft_max_level].value;
+
 	ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF);
 	PP_ASSERT_WITH_CODE(!ret,
 			"Failed to upload DPM Bootup Levels!",
@@ -2240,7 +2390,7 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, uint32_t mask)
 {
 	struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend);
-	uint32_t soft_min_level, soft_max_level;
+	uint32_t soft_min_level, soft_max_level, hard_min_level;
 	int ret = 0;
 
 	switch (type) {
@@ -2248,6 +2398,13 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr,
 		soft_min_level = mask ? (ffs(mask) - 1) : 0;
 		soft_max_level = mask ? (fls(mask) - 1) : 0;
 
+		if (soft_max_level >= data->dpm_table.gfx_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level,
+					data->dpm_table.gfx_table.count - 1);
+			return -EINVAL;
+		}
+
 		data->dpm_table.gfx_table.dpm_state.soft_min_level =
 			data->dpm_table.gfx_table.dpm_levels[soft_min_level].value;
 		data->dpm_table.gfx_table.dpm_state.soft_max_level =
@@ -2268,6 +2425,13 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr,
 		soft_min_level = mask ? (ffs(mask) - 1) : 0;
 		soft_max_level = mask ? (fls(mask) - 1) : 0;
 
+		if (soft_max_level >= data->dpm_table.mem_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level,
+					data->dpm_table.mem_table.count - 1);
+			return -EINVAL;
+		}
+
 		data->dpm_table.mem_table.dpm_state.soft_min_level =
 			data->dpm_table.mem_table.dpm_levels[soft_min_level].value;
 		data->dpm_table.mem_table.dpm_state.soft_max_level =
@@ -2285,6 +2449,84 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr,
 
 		break;
 
+	case PP_SOCCLK:
+		soft_min_level = mask ? (ffs(mask) - 1) : 0;
+		soft_max_level = mask ? (fls(mask) - 1) : 0;
+
+		if (soft_max_level >= data->dpm_table.soc_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level,
+					data->dpm_table.soc_table.count - 1);
+			return -EINVAL;
+		}
+
+		data->dpm_table.soc_table.dpm_state.soft_min_level =
+			data->dpm_table.soc_table.dpm_levels[soft_min_level].value;
+		data->dpm_table.soc_table.dpm_state.soft_max_level =
+			data->dpm_table.soc_table.dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_SOCCLK_MASK);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload boot level to lowest!",
+			return ret);
+
+		ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_SOCCLK_MASK);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload dpm max level to highest!",
+			return ret);
+
+		break;
+
+	case PP_FCLK:
+		soft_min_level = mask ? (ffs(mask) - 1) : 0;
+		soft_max_level = mask ? (fls(mask) - 1) : 0;
+
+		if (soft_max_level >= data->dpm_table.fclk_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level,
+					data->dpm_table.fclk_table.count - 1);
+			return -EINVAL;
+		}
+
+		data->dpm_table.fclk_table.dpm_state.soft_min_level =
+			data->dpm_table.fclk_table.dpm_levels[soft_min_level].value;
+		data->dpm_table.fclk_table.dpm_state.soft_max_level =
+			data->dpm_table.fclk_table.dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_FCLK_MASK);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload boot level to lowest!",
+			return ret);
+
+		ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_FCLK_MASK);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload dpm max level to highest!",
+			return ret);
+
+		break;
+
+	case PP_DCEFCLK:
+		hard_min_level = mask ? (ffs(mask) - 1) : 0;
+
+		if (hard_min_level >= data->dpm_table.dcef_table.count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					hard_min_level,
+					data->dpm_table.dcef_table.count - 1);
+			return -EINVAL;
+		}
+
+		data->dpm_table.dcef_table.dpm_state.hard_min_level =
+			data->dpm_table.dcef_table.dpm_levels[hard_min_level].value;
+
+		ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_DCEFCLK_MASK);
+		PP_ASSERT_WITH_CODE(!ret,
+			"Failed to upload boot level to lowest!",
+			return ret);
+
+		//TODO: Setting DCEFCLK max dpm level is not supported
+
+		break;
+
 	case PP_PCIE:
 		soft_min_level = mask ? (ffs(mask) - 1) : 0;
 		soft_max_level = mask ? (fls(mask) - 1) : 0;
@@ -2335,6 +2577,7 @@ static int vega20_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 			return ret;
 		vega20_force_clock_level(hwmgr, PP_SCLK, 1 << sclk_mask);
 		vega20_force_clock_level(hwmgr, PP_MCLK, 1 << mclk_mask);
+		vega20_force_clock_level(hwmgr, PP_SOCCLK, 1 << soc_mask);
 		break;
 
 	case AMD_DPM_FORCED_LEVEL_MANUAL:
@@ -2765,6 +3008,108 @@ static int vega20_odn_edit_dpm_table(struct pp_hwmgr *hwmgr,
 	return 0;
 }
 
+static int vega20_get_ppfeature_status(struct pp_hwmgr *hwmgr, char *buf)
+{
+	static const char *ppfeature_name[] = {
+				"DPM_PREFETCHER",
+				"GFXCLK_DPM",
+				"UCLK_DPM",
+				"SOCCLK_DPM",
+				"UVD_DPM",
+				"VCE_DPM",
+				"ULV",
+				"MP0CLK_DPM",
+				"LINK_DPM",
+				"DCEFCLK_DPM",
+				"GFXCLK_DS",
+				"SOCCLK_DS",
+				"LCLK_DS",
+				"PPT",
+				"TDC",
+				"THERMAL",
+				"GFX_PER_CU_CG",
+				"RM",
+				"DCEFCLK_DS",
+				"ACDC",
+				"VR0HOT",
+				"VR1HOT",
+				"FW_CTF",
+				"LED_DISPLAY",
+				"FAN_CONTROL",
+				"GFX_EDC",
+				"GFXOFF",
+				"CG",
+				"FCLK_DPM",
+				"FCLK_DS",
+				"MP1CLK_DS",
+				"MP0CLK_DS",
+				"XGMI"};
+	static const char *output_title[] = {
+				"FEATURES",
+				"BITMASK",
+				"ENABLEMENT"};
+	uint64_t features_enabled;
+	int i;
+	int ret = 0;
+	int size = 0;
+
+	ret = vega20_get_enabled_smc_features(hwmgr, &features_enabled);
+	PP_ASSERT_WITH_CODE(!ret,
+			"[EnableAllSmuFeatures] Failed to get enabled smc features!",
+			return ret);
+
+	size += sprintf(buf + size, "Current ppfeatures: 0x%016llx\n", features_enabled);
+	size += sprintf(buf + size, "%-19s %-22s %s\n",
+				output_title[0],
+				output_title[1],
+				output_title[2]);
+	for (i = 0; i < GNLD_FEATURES_MAX; i++) {
+		size += sprintf(buf + size, "%-19s 0x%016llx %6s\n",
+					ppfeature_name[i],
+					1ULL << i,
+					(features_enabled & (1ULL << i)) ? "Y" : "N");
+	}
+
+	return size;
+}
+
+static int vega20_set_ppfeature_status(struct pp_hwmgr *hwmgr, uint64_t new_ppfeature_masks)
+{
+	uint64_t features_enabled;
+	uint64_t features_to_enable;
+	uint64_t features_to_disable;
+	int ret = 0;
+
+	if (new_ppfeature_masks >= (1ULL << GNLD_FEATURES_MAX))
+		return -EINVAL;
+
+	ret = vega20_get_enabled_smc_features(hwmgr, &features_enabled);
+	if (ret)
+		return ret;
+
+	features_to_disable =
+		(features_enabled ^ new_ppfeature_masks) & features_enabled;
+	features_to_enable =
+		(features_enabled ^ new_ppfeature_masks) ^ features_to_disable;
+
+	pr_debug("features_to_disable 0x%llx\n", features_to_disable);
+	pr_debug("features_to_enable 0x%llx\n", features_to_enable);
+
+	if (features_to_disable) {
+		ret = vega20_enable_smc_features(hwmgr, false, features_to_disable);
+		if (ret)
+			return ret;
+	}
+
+	if (features_to_enable) {
+		ret = vega20_enable_smc_features(hwmgr, true, features_to_enable);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 		enum pp_clock_type type, char *buf)
 {
@@ -2779,6 +3124,8 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 	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);
 	int i, now, size = 0;
 	int ret = 0;
 	uint32_t gen_speed, lane_width;
@@ -2818,6 +3165,52 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
 				(clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
 		break;
 
+	case PP_SOCCLK:
+		ret = vega20_get_current_clk_freq(hwmgr, PPCLK_SOCCLK, &now);
+		PP_ASSERT_WITH_CODE(!ret,
+				"Attempt to get current socclk freq Failed!",
+				return ret);
+
+		ret = vega20_get_socclocks(hwmgr, &clocks);
+		PP_ASSERT_WITH_CODE(!ret,
+				"Attempt to get soc clk levels Failed!",
+				return ret);
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
+		break;
+
+	case PP_FCLK:
+		ret = vega20_get_current_clk_freq(hwmgr, PPCLK_FCLK, &now);
+		PP_ASSERT_WITH_CODE(!ret,
+				"Attempt to get current fclk freq Failed!",
+				return ret);
+
+		for (i = 0; i < fclk_dpm_table->count; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, fclk_dpm_table->dpm_levels[i].value,
+				fclk_dpm_table->dpm_levels[i].value == (now / 100) ? "*" : "");
+		break;
+
+	case PP_DCEFCLK:
+		ret = vega20_get_current_clk_freq(hwmgr, PPCLK_DCEFCLK, &now);
+		PP_ASSERT_WITH_CODE(!ret,
+				"Attempt to get current dcefclk freq Failed!",
+				return ret);
+
+		ret = vega20_get_dcefclocks(hwmgr, &clocks);
+		PP_ASSERT_WITH_CODE(!ret,
+				"Attempt to get dcefclk levels Failed!",
+				return ret);
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
+		break;
+
 	case PP_PCIE:
 		gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
 			     PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
@@ -3261,6 +3654,9 @@ static int conv_power_profile_to_pplib_workload(int power_profile)
 	int pplib_workload = 0;
 
 	switch (power_profile) {
+	case PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT:
+		pplib_workload = WORKLOAD_DEFAULT_BIT;
+		break;
 	case PP_SMC_POWER_PROFILE_FULLSCREEN3D:
 		pplib_workload = WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT;
 		break;
@@ -3290,6 +3686,7 @@ static int vega20_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf)
 	uint32_t i, size = 0;
 	uint16_t workload_type = 0;
 	static const char *profile_name[] = {
+					"BOOTUP_DEFAULT",
 					"3D_FULL_SCREEN",
 					"POWER_SAVING",
 					"VIDEO",
@@ -3557,6 +3954,8 @@ static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
 	.force_clock_level = vega20_force_clock_level,
 	.print_clock_levels = vega20_print_clock_levels,
 	.read_sensor = vega20_read_sensor,
+	.get_ppfeature_status = vega20_get_ppfeature_status,
+	.set_ppfeature_status = vega20_set_ppfeature_status,
 	/* powergate related */
 	.powergate_uvd = vega20_power_gate_uvd,
 	.powergate_vce = vega20_power_gate_vce,
@@ -3577,6 +3976,10 @@ static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
 	/* smu memory related */
 	.notify_cac_buffer_info = vega20_notify_cac_buffer_info,
 	.enable_mgpu_fan_boost = vega20_enable_mgpu_fan_boost,
+	/* BACO related */
+	.get_asic_baco_capability = vega20_baco_get_capability,
+	.get_asic_baco_state = vega20_baco_get_state,
+	.set_asic_baco_state = vega20_baco_set_state,
 };
 
 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/powerplay/hwmgr/vega20_hwmgr.h
index 25faaa5c5b10..37f5f5e657da 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
@@ -520,6 +520,9 @@ struct vega20_hwmgr {
 	/* ---- Gfxoff ---- */
 	bool                           gfxoff_allowed;
 	uint32_t                       counter_gfxoff;
+
+	unsigned long                  metrics_time;
+	SmuMetrics_t                   metrics_table;
 };
 
 #define VEGA20_DPM2_NEAR_TDP_DEC                      10
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h
index 6738bad53602..613cb1989b3d 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_inc.h
@@ -31,5 +31,6 @@
 #include "asic_reg/mp/mp_9_0_sh_mask.h"
 
 #include "asic_reg/nbio/nbio_7_4_offset.h"
+#include "asic_reg/nbio/nbio_7_4_sh_mask.h"
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h
index f4dab979a3a1..6e0be6027705 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h
@@ -397,7 +397,6 @@ struct phm_odn_clock_levels {
 };
 
 extern int phm_disable_clock_power_gatings(struct pp_hwmgr *hwmgr);
-extern int phm_enable_clock_power_gatings(struct pp_hwmgr *hwmgr);
 extern int phm_powerdown_uvd(struct pp_hwmgr *hwmgr);
 extern int phm_setup_asic(struct pp_hwmgr *hwmgr);
 extern int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr);
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
index 0d298a0409f5..bac3d85e3b82 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
@@ -47,6 +47,11 @@ enum DISPLAY_GAP {
 };
 typedef enum DISPLAY_GAP DISPLAY_GAP;
 
+enum BACO_STATE {
+	BACO_STATE_OUT = 0,
+	BACO_STATE_IN,
+};
+
 struct vi_dpm_level {
 	bool enabled;
 	uint32_t value;
@@ -251,7 +256,6 @@ struct pp_hwmgr_func {
 	uint32_t (*get_sclk)(struct pp_hwmgr *hwmgr, bool low);
 	int (*power_state_set)(struct pp_hwmgr *hwmgr,
 						const void *state);
-	int (*enable_clock_power_gating)(struct pp_hwmgr *hwmgr);
 	int (*notify_smc_display_config_after_ps_adjustment)(struct pp_hwmgr *hwmgr);
 	int (*pre_display_config_changed)(struct pp_hwmgr *hwmgr);
 	int (*display_config_changed)(struct pp_hwmgr *hwmgr);
@@ -334,6 +338,11 @@ 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 (*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);
+	int (*get_ppfeature_status)(struct pp_hwmgr *hwmgr, char *buf);
+	int (*set_ppfeature_status)(struct pp_hwmgr *hwmgr, uint64_t ppfeature_masks);
 };
 
 struct pp_table_func {
@@ -678,6 +687,7 @@ struct pp_advance_fan_control_parameters {
 	uint32_t  ulTargetGfxClk;
 	uint16_t  usZeroRPMStartTemperature;
 	uint16_t  usZeroRPMStopTemperature;
+	uint16_t  usMGpuThrottlingRPMLimit;
 };
 
 struct pp_thermal_controller_info {
@@ -705,7 +715,7 @@ enum PP_TABLE_VERSION {
 /**
  * The main hardware manager structure.
  */
-#define Workload_Policy_Max 5
+#define Workload_Policy_Max 6
 
 struct pp_hwmgr {
 	void *adev;
diff --git a/drivers/gpu/drm/arc/arcpgu_crtc.c b/drivers/gpu/drm/arc/arcpgu_crtc.c
index 155ab177ce0b..73e508e00e30 100644
--- a/drivers/gpu/drm/arc/arcpgu_crtc.c
+++ b/drivers/gpu/drm/arc/arcpgu_crtc.c
@@ -15,12 +15,12 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_vblank.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/platform_data/simplefb.h>
 
diff --git a/drivers/gpu/drm/arc/arcpgu_drv.c b/drivers/gpu/drm/arc/arcpgu_drv.c
index 39a79f5718c4..c9f78397d345 100644
--- a/drivers/gpu/drm/arc/arcpgu_drv.c
+++ b/drivers/gpu/drm/arc/arcpgu_drv.c
@@ -15,15 +15,15 @@
  */
 
 #include <linux/clk.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_device.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_debugfs.h>
+#include <drm/drm_device.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/dma-mapping.h>
 #include <linux/module.h>
 #include <linux/of_reserved_mem.h>
diff --git a/drivers/gpu/drm/arc/arcpgu_sim.c b/drivers/gpu/drm/arc/arcpgu_sim.c
index 6530d88f7293..5ea053cf805c 100644
--- a/drivers/gpu/drm/arc/arcpgu_sim.c
+++ b/drivers/gpu/drm/arc/arcpgu_sim.c
@@ -14,8 +14,9 @@
  *
  */
 
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_probe_helper.h>
 
 #include "arcpgu.h"
 
diff --git a/drivers/gpu/drm/arm/Kconfig b/drivers/gpu/drm/arm/Kconfig
index 9a18e1bd57b4..a204103b3efb 100644
--- a/drivers/gpu/drm/arm/Kconfig
+++ b/drivers/gpu/drm/arm/Kconfig
@@ -1,13 +1,10 @@
-config DRM_ARM
-	bool
-	help
-	  Choose this option to select drivers for ARM's devices
+# SPDX-License-Identifier: GPL-2.0
+menu "ARM devices"
 
 config DRM_HDLCD
 	tristate "ARM HDLCD"
 	depends on DRM && OF && (ARM || ARM64)
 	depends on COMMON_CLK
-	select DRM_ARM
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	help
@@ -29,7 +26,6 @@ config DRM_MALI_DISPLAY
 	tristate "ARM Mali Display Processor"
 	depends on DRM && OF && (ARM || ARM64)
 	depends on COMMON_CLK
-	select DRM_ARM
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
@@ -40,3 +36,7 @@ config DRM_MALI_DISPLAY
 	  of the hardware.
 
 	  If compiled as a module it will be called mali-dp.
+
+source "drivers/gpu/drm/arm/display/Kconfig"
+
+endmenu
diff --git a/drivers/gpu/drm/arm/Makefile b/drivers/gpu/drm/arm/Makefile
index 3bf31d1a4722..120bef801fcf 100644
--- a/drivers/gpu/drm/arm/Makefile
+++ b/drivers/gpu/drm/arm/Makefile
@@ -3,3 +3,4 @@ obj-$(CONFIG_DRM_HDLCD)	+= hdlcd.o
 mali-dp-y := malidp_drv.o malidp_hw.o malidp_planes.o malidp_crtc.o
 mali-dp-y += malidp_mw.o
 obj-$(CONFIG_DRM_MALI_DISPLAY)	+= mali-dp.o
+obj-$(CONFIG_DRM_KOMEDA) += display/
diff --git a/drivers/gpu/drm/arm/display/Kbuild b/drivers/gpu/drm/arm/display/Kbuild
new file mode 100644
index 000000000000..382f1ca831e4
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/Kbuild
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_DRM_KOMEDA) += komeda/
diff --git a/drivers/gpu/drm/arm/display/Kconfig b/drivers/gpu/drm/arm/display/Kconfig
new file mode 100644
index 000000000000..cec0639e3aa1
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/Kconfig
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0
+config DRM_KOMEDA
+	tristate "ARM Komeda display driver"
+	depends on DRM && OF
+	depends on COMMON_CLK
+	select DRM_KMS_HELPER
+	select DRM_KMS_CMA_HELPER
+	select DRM_GEM_CMA_HELPER
+	select VIDEOMODE_HELPERS
+	help
+	  Choose this option if you want to compile the ARM Komeda display
+	  Processor driver. It supports the D71 variants of the hardware.
+
+	  If compiled as a module it will be called komeda.
diff --git a/drivers/gpu/drm/arm/display/include/malidp_io.h b/drivers/gpu/drm/arm/display/include/malidp_io.h
new file mode 100644
index 000000000000..4fb3caf864ce
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/include/malidp_io.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _MALIDP_IO_H_
+#define _MALIDP_IO_H_
+
+#include <linux/io.h>
+
+static inline u32
+malidp_read32(u32 __iomem *base, u32 offset)
+{
+	return readl((base + (offset >> 2)));
+}
+
+static inline void
+malidp_write32(u32 __iomem *base, u32 offset, u32 v)
+{
+	writel(v, (base + (offset >> 2)));
+}
+
+static inline void
+malidp_write32_mask(u32 __iomem *base, u32 offset, u32 m, u32 v)
+{
+	u32 tmp = malidp_read32(base, offset);
+
+	tmp &= (~m);
+	malidp_write32(base, offset, v | tmp);
+}
+
+static inline void
+malidp_write_group(u32 __iomem *base, u32 offset, int num, const u32 *values)
+{
+	int i;
+
+	for (i = 0; i < num; i++)
+		malidp_write32(base, offset + i * 4, values[i]);
+}
+
+#endif /*_MALIDP_IO_H_*/
diff --git a/drivers/gpu/drm/arm/display/include/malidp_product.h b/drivers/gpu/drm/arm/display/include/malidp_product.h
new file mode 100644
index 000000000000..b35fc5db866b
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/include/malidp_product.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _MALIDP_PRODUCT_H_
+#define _MALIDP_PRODUCT_H_
+
+/* Product identification */
+#define MALIDP_CORE_ID(__product, __major, __minor, __status) \
+	((((__product) & 0xFFFF) << 16) | (((__major) & 0xF) << 12) | \
+	(((__minor) & 0xF) << 8) | ((__status) & 0xFF))
+
+#define MALIDP_CORE_ID_PRODUCT_ID(__core_id) ((__u32)(__core_id) >> 16)
+#define MALIDP_CORE_ID_MAJOR(__core_id)      (((__u32)(__core_id) >> 12) & 0xF)
+#define MALIDP_CORE_ID_MINOR(__core_id)      (((__u32)(__core_id) >> 8) & 0xF)
+#define MALIDP_CORE_ID_STATUS(__core_id)     (((__u32)(__core_id)) & 0xFF)
+
+/* Mali-display product IDs */
+#define MALIDP_D71_PRODUCT_ID   0x0071
+
+#endif /* _MALIDP_PRODUCT_H_ */
diff --git a/drivers/gpu/drm/arm/display/include/malidp_utils.h b/drivers/gpu/drm/arm/display/include/malidp_utils.h
new file mode 100644
index 000000000000..63cc47cefcf8
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/include/malidp_utils.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _MALIDP_UTILS_
+#define _MALIDP_UTILS_
+
+#define has_bit(nr, mask)	(BIT(nr) & (mask))
+#define has_bits(bits, mask)	(((bits) & (mask)) == (bits))
+
+#define dp_for_each_set_bit(bit, mask) \
+	for_each_set_bit((bit), ((unsigned long *)&(mask)), sizeof(mask) * 8)
+
+#endif /* _MALIDP_UTILS_ */
diff --git a/drivers/gpu/drm/arm/display/komeda/Makefile b/drivers/gpu/drm/arm/display/komeda/Makefile
new file mode 100644
index 000000000000..1b875e5dc0f6
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/Makefile
@@ -0,0 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0
+
+ccflags-y := \
+	-I$(src)/../include \
+	-I$(src)
+
+komeda-y := \
+	komeda_drv.o \
+	komeda_dev.o \
+	komeda_format_caps.o \
+	komeda_pipeline.o \
+	komeda_framebuffer.o \
+	komeda_kms.o \
+	komeda_crtc.o \
+	komeda_plane.o \
+	komeda_private_obj.o
+
+komeda-y += \
+	d71/d71_dev.o
+
+obj-$(CONFIG_DRM_KOMEDA) += komeda.o
diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c
new file mode 100644
index 000000000000..edbf9daa1545
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include "malidp_io.h"
+#include "komeda_dev.h"
+
+static int d71_enum_resources(struct komeda_dev *mdev)
+{
+	/* TODO add enum resources */
+	return -1;
+}
+
+#define __HW_ID(__group, __format) \
+	((((__group) & 0x7) << 3) | ((__format) & 0x7))
+
+#define RICH		KOMEDA_FMT_RICH_LAYER
+#define SIMPLE		KOMEDA_FMT_SIMPLE_LAYER
+#define RICH_SIMPLE	(KOMEDA_FMT_RICH_LAYER | KOMEDA_FMT_SIMPLE_LAYER)
+#define RICH_WB		(KOMEDA_FMT_RICH_LAYER | KOMEDA_FMT_WB_LAYER)
+#define RICH_SIMPLE_WB	(RICH_SIMPLE | KOMEDA_FMT_WB_LAYER)
+
+#define Rot_0		DRM_MODE_ROTATE_0
+#define Flip_H_V	(DRM_MODE_REFLECT_X | DRM_MODE_REFLECT_Y | Rot_0)
+#define Rot_ALL_H_V	(DRM_MODE_ROTATE_MASK | Flip_H_V)
+
+#define LYT_NM		BIT(AFBC_FORMAT_MOD_BLOCK_SIZE_16x16)
+#define LYT_WB		BIT(AFBC_FORMAT_MOD_BLOCK_SIZE_32x8)
+#define LYT_NM_WB	(LYT_NM | LYT_WB)
+
+#define AFB_TH		AFBC(_TILED | _SPARSE)
+#define AFB_TH_SC_YTR	AFBC(_TILED | _SC | _SPARSE | _YTR)
+#define AFB_TH_SC_YTR_BS AFBC(_TILED | _SC | _SPARSE | _YTR | _SPLIT)
+
+static struct komeda_format_caps d71_format_caps_table[] = {
+	/*   HW_ID    |        fourcc        | tile_sz |   layer_types |   rots    | afbc_layouts | afbc_features */
+	/* ABGR_2101010*/
+	{__HW_ID(0, 0),	DRM_FORMAT_ARGB2101010,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(0, 1),	DRM_FORMAT_ABGR2101010,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(0, 1),	DRM_FORMAT_ABGR2101010,	1,	RICH_SIMPLE,	Rot_ALL_H_V,	LYT_NM_WB, AFB_TH_SC_YTR_BS}, /* afbc */
+	{__HW_ID(0, 2),	DRM_FORMAT_RGBA1010102,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(0, 3),	DRM_FORMAT_BGRA1010102,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	/* ABGR_8888*/
+	{__HW_ID(1, 0),	DRM_FORMAT_ARGB8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(1, 1),	DRM_FORMAT_ABGR8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(1, 1),	DRM_FORMAT_ABGR8888,	1,	RICH_SIMPLE,	Rot_ALL_H_V,	LYT_NM_WB, AFB_TH_SC_YTR_BS}, /* afbc */
+	{__HW_ID(1, 2),	DRM_FORMAT_RGBA8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(1, 3),	DRM_FORMAT_BGRA8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	/* XBGB_8888 */
+	{__HW_ID(2, 0),	DRM_FORMAT_XRGB8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(2, 1),	DRM_FORMAT_XBGR8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(2, 2),	DRM_FORMAT_RGBX8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	{__HW_ID(2, 3),	DRM_FORMAT_BGRX8888,	1,	RICH_SIMPLE_WB,	Flip_H_V,		0, 0},
+	/* BGR_888 */ /* none-afbc RGB888 doesn't support rotation and flip */
+	{__HW_ID(3, 0),	DRM_FORMAT_RGB888,	1,	RICH_SIMPLE_WB,	Rot_0,			0, 0},
+	{__HW_ID(3, 1),	DRM_FORMAT_BGR888,	1,	RICH_SIMPLE_WB,	Rot_0,			0, 0},
+	{__HW_ID(3, 1),	DRM_FORMAT_BGR888,	1,	RICH_SIMPLE,	Rot_ALL_H_V,	LYT_NM_WB, AFB_TH_SC_YTR_BS}, /* afbc */
+	/* BGR 16bpp */
+	{__HW_ID(4, 0),	DRM_FORMAT_RGBA5551,	1,	RICH_SIMPLE,	Flip_H_V,		0, 0},
+	{__HW_ID(4, 1),	DRM_FORMAT_ABGR1555,	1,	RICH_SIMPLE,	Flip_H_V,		0, 0},
+	{__HW_ID(4, 1),	DRM_FORMAT_ABGR1555,	1,	RICH_SIMPLE,	Rot_ALL_H_V,	LYT_NM_WB, AFB_TH_SC_YTR}, /* afbc */
+	{__HW_ID(4, 2),	DRM_FORMAT_RGB565,	1,	RICH_SIMPLE,	Flip_H_V,		0, 0},
+	{__HW_ID(4, 3),	DRM_FORMAT_BGR565,	1,	RICH_SIMPLE,	Flip_H_V,		0, 0},
+	{__HW_ID(4, 3),	DRM_FORMAT_BGR565,	1,	RICH_SIMPLE,	Rot_ALL_H_V,	LYT_NM_WB, AFB_TH_SC_YTR}, /* afbc */
+	{__HW_ID(4, 4), DRM_FORMAT_R8,		1,	SIMPLE,		Rot_0,			0, 0},
+	/* YUV 444/422/420 8bit  */
+	{__HW_ID(5, 0),	0 /*XYUV8888*/,		1,	0,		0,			0, 0},
+	/* XYUV unsupported*/
+	{__HW_ID(5, 1),	DRM_FORMAT_YUYV,	1,	RICH,		Rot_ALL_H_V,	LYT_NM, AFB_TH}, /* afbc */
+	{__HW_ID(5, 2),	DRM_FORMAT_YUYV,	1,	RICH,		Flip_H_V,		0, 0},
+	{__HW_ID(5, 3),	DRM_FORMAT_UYVY,	1,	RICH,		Flip_H_V,		0, 0},
+	{__HW_ID(5, 4),	0, /*X0L0 */		2,		0,			0, 0}, /* Y0L0 unsupported */
+	{__HW_ID(5, 6),	DRM_FORMAT_NV12,	1,	RICH,		Flip_H_V,		0, 0},
+	{__HW_ID(5, 6),	0/*DRM_FORMAT_YUV420_8BIT*/,	1,	RICH,	Rot_ALL_H_V,	LYT_NM, AFB_TH}, /* afbc */
+	{__HW_ID(5, 7),	DRM_FORMAT_YUV420,	1,	RICH,		Flip_H_V,		0, 0},
+	/* YUV 10bit*/
+	{__HW_ID(6, 0),	0,/*XVYU2101010*/	1,	0,		0,			0, 0},/* VYV30 unsupported */
+	{__HW_ID(6, 6),	0/*DRM_FORMAT_X0L2*/,	2,	RICH,		Flip_H_V,		0, 0},
+	{__HW_ID(6, 7),	0/*DRM_FORMAT_P010*/,	1,	RICH,		Flip_H_V,		0, 0},
+	{__HW_ID(6, 7),	0/*DRM_FORMAT_YUV420_10BIT*/, 1,	RICH,	Rot_ALL_H_V,	LYT_NM, AFB_TH},
+};
+
+static void d71_init_fmt_tbl(struct komeda_dev *mdev)
+{
+	struct komeda_format_caps_table *table = &mdev->fmt_tbl;
+
+	table->format_caps = d71_format_caps_table;
+	table->n_formats = ARRAY_SIZE(d71_format_caps_table);
+}
+
+static struct komeda_dev_funcs d71_chip_funcs = {
+	.init_format_table = d71_init_fmt_tbl,
+	.enum_resources	= d71_enum_resources,
+	.cleanup	= NULL,
+};
+
+#define GLB_ARCH_ID		0x000
+#define GLB_CORE_ID		0x004
+#define GLB_CORE_INFO		0x008
+
+struct komeda_dev_funcs *
+d71_identify(u32 __iomem *reg_base, struct komeda_chip_info *chip)
+{
+	chip->arch_id	= malidp_read32(reg_base, GLB_ARCH_ID);
+	chip->core_id	= malidp_read32(reg_base, GLB_CORE_ID);
+	chip->core_info	= malidp_read32(reg_base, GLB_CORE_INFO);
+
+	return &d71_chip_funcs;
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c
new file mode 100644
index 000000000000..3ca5718aa0c2
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c
@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <linux/clk.h>
+#include <linux/pm_runtime.h>
+#include <linux/spinlock.h>
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_print.h>
+#include <drm/drm_vblank.h>
+
+#include "komeda_dev.h"
+#include "komeda_kms.h"
+
+struct drm_crtc_helper_funcs komeda_crtc_helper_funcs = {
+};
+
+static const struct drm_crtc_funcs komeda_crtc_funcs = {
+};
+
+int komeda_kms_setup_crtcs(struct komeda_kms_dev *kms,
+			   struct komeda_dev *mdev)
+{
+	struct komeda_crtc *crtc;
+	struct komeda_pipeline *master;
+	char str[16];
+	int i;
+
+	kms->n_crtcs = 0;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		crtc = &kms->crtcs[kms->n_crtcs];
+		master = mdev->pipelines[i];
+
+		crtc->master = master;
+		crtc->slave  = NULL;
+
+		if (crtc->slave)
+			sprintf(str, "pipe-%d", crtc->slave->id);
+		else
+			sprintf(str, "None");
+
+		DRM_INFO("crtc%d: master(pipe-%d) slave(%s) output: %s.\n",
+			 kms->n_crtcs, master->id, str,
+			 master->of_output_dev ?
+			 master->of_output_dev->full_name : "None");
+
+		kms->n_crtcs++;
+	}
+
+	return 0;
+}
+
+static struct drm_plane *
+get_crtc_primary(struct komeda_kms_dev *kms, struct komeda_crtc *crtc)
+{
+	struct komeda_plane *kplane;
+	struct drm_plane *plane;
+
+	drm_for_each_plane(plane, &kms->base) {
+		if (plane->type != DRM_PLANE_TYPE_PRIMARY)
+			continue;
+
+		kplane = to_kplane(plane);
+		/* only master can be primary */
+		if (kplane->layer->base.pipeline == crtc->master)
+			return plane;
+	}
+
+	return NULL;
+}
+
+static int komeda_crtc_add(struct komeda_kms_dev *kms,
+			   struct komeda_crtc *kcrtc)
+{
+	struct drm_crtc *crtc = &kcrtc->base;
+	int err;
+
+	err = drm_crtc_init_with_planes(&kms->base, crtc,
+					get_crtc_primary(kms, kcrtc), NULL,
+					&komeda_crtc_funcs, NULL);
+	if (err)
+		return err;
+
+	drm_crtc_helper_add(crtc, &komeda_crtc_helper_funcs);
+	drm_crtc_vblank_reset(crtc);
+
+	crtc->port = kcrtc->master->of_output_port;
+
+	return 0;
+}
+
+int komeda_kms_add_crtcs(struct komeda_kms_dev *kms, struct komeda_dev *mdev)
+{
+	int i, err;
+
+	for (i = 0; i < kms->n_crtcs; i++) {
+		err = komeda_crtc_add(kms, &kms->crtcs[i]);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
new file mode 100644
index 000000000000..70e9bb7fa30c
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
@@ -0,0 +1,190 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <linux/io.h>
+#include <linux/of_device.h>
+#include <linux/of_graph.h>
+#include <linux/platform_device.h>
+
+#include <drm/drm_print.h>
+
+#include "komeda_dev.h"
+
+static int komeda_parse_pipe_dt(struct komeda_dev *mdev, struct device_node *np)
+{
+	struct komeda_pipeline *pipe;
+	struct clk *clk;
+	u32 pipe_id;
+	int ret = 0;
+
+	ret = of_property_read_u32(np, "reg", &pipe_id);
+	if (ret != 0 || pipe_id >= mdev->n_pipelines)
+		return -EINVAL;
+
+	pipe = mdev->pipelines[pipe_id];
+
+	clk = of_clk_get_by_name(np, "aclk");
+	if (IS_ERR(clk)) {
+		DRM_ERROR("get aclk for pipeline %d failed!\n", pipe_id);
+		return PTR_ERR(clk);
+	}
+	pipe->aclk = clk;
+
+	clk = of_clk_get_by_name(np, "pxclk");
+	if (IS_ERR(clk)) {
+		DRM_ERROR("get pxclk for pipeline %d failed!\n", pipe_id);
+		return PTR_ERR(clk);
+	}
+	pipe->pxlclk = clk;
+
+	/* enum ports */
+	pipe->of_output_dev =
+		of_graph_get_remote_node(np, KOMEDA_OF_PORT_OUTPUT, 0);
+	pipe->of_output_port =
+		of_graph_get_port_by_id(np, KOMEDA_OF_PORT_OUTPUT);
+
+	pipe->of_node = np;
+
+	return 0;
+}
+
+static int komeda_parse_dt(struct device *dev, struct komeda_dev *mdev)
+{
+	struct device_node *child, *np = dev->of_node;
+	struct clk *clk;
+	int ret;
+
+	clk = devm_clk_get(dev, "mclk");
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+
+	mdev->mclk = clk;
+
+	for_each_available_child_of_node(np, child) {
+		if (of_node_cmp(child->name, "pipeline") == 0) {
+			ret = komeda_parse_pipe_dt(mdev, child);
+			if (ret) {
+				DRM_ERROR("parse pipeline dt error!\n");
+				of_node_put(child);
+				break;
+			}
+		}
+	}
+
+	return ret;
+}
+
+struct komeda_dev *komeda_dev_create(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	const struct komeda_product_data *product;
+	struct komeda_dev *mdev;
+	struct resource *io_res;
+	int err = 0;
+
+	product = of_device_get_match_data(dev);
+	if (!product)
+		return ERR_PTR(-ENODEV);
+
+	io_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!io_res) {
+		DRM_ERROR("No registers defined.\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	mdev = devm_kzalloc(dev, sizeof(*mdev), GFP_KERNEL);
+	if (!mdev)
+		return ERR_PTR(-ENOMEM);
+
+	mdev->dev = dev;
+	mdev->reg_base = devm_ioremap_resource(dev, io_res);
+	if (IS_ERR(mdev->reg_base)) {
+		DRM_ERROR("Map register space failed.\n");
+		err = PTR_ERR(mdev->reg_base);
+		mdev->reg_base = NULL;
+		goto err_cleanup;
+	}
+
+	mdev->pclk = devm_clk_get(dev, "pclk");
+	if (IS_ERR(mdev->pclk)) {
+		DRM_ERROR("Get APB clk failed.\n");
+		err = PTR_ERR(mdev->pclk);
+		mdev->pclk = NULL;
+		goto err_cleanup;
+	}
+
+	/* Enable APB clock to access the registers */
+	clk_prepare_enable(mdev->pclk);
+
+	mdev->funcs = product->identify(mdev->reg_base, &mdev->chip);
+	if (!komeda_product_match(mdev, product->product_id)) {
+		DRM_ERROR("DT configured %x mismatch with real HW %x.\n",
+			  product->product_id,
+			  MALIDP_CORE_ID_PRODUCT_ID(mdev->chip.core_id));
+		err = -ENODEV;
+		goto err_cleanup;
+	}
+
+	DRM_INFO("Found ARM Mali-D%x version r%dp%d\n",
+		 MALIDP_CORE_ID_PRODUCT_ID(mdev->chip.core_id),
+		 MALIDP_CORE_ID_MAJOR(mdev->chip.core_id),
+		 MALIDP_CORE_ID_MINOR(mdev->chip.core_id));
+
+	mdev->funcs->init_format_table(mdev);
+
+	err = mdev->funcs->enum_resources(mdev);
+	if (err) {
+		DRM_ERROR("enumerate display resource failed.\n");
+		goto err_cleanup;
+	}
+
+	err = komeda_parse_dt(dev, mdev);
+	if (err) {
+		DRM_ERROR("parse device tree failed.\n");
+		goto err_cleanup;
+	}
+
+	return mdev;
+
+err_cleanup:
+	komeda_dev_destroy(mdev);
+	return ERR_PTR(err);
+}
+
+void komeda_dev_destroy(struct komeda_dev *mdev)
+{
+	struct device *dev = mdev->dev;
+	struct komeda_dev_funcs *funcs = mdev->funcs;
+	int i;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		komeda_pipeline_destroy(mdev, mdev->pipelines[i]);
+		mdev->pipelines[i] = NULL;
+	}
+
+	mdev->n_pipelines = 0;
+
+	if (funcs && funcs->cleanup)
+		funcs->cleanup(mdev);
+
+	if (mdev->reg_base) {
+		devm_iounmap(dev, mdev->reg_base);
+		mdev->reg_base = NULL;
+	}
+
+	if (mdev->mclk) {
+		devm_clk_put(dev, mdev->mclk);
+		mdev->mclk = NULL;
+	}
+
+	if (mdev->pclk) {
+		clk_disable_unprepare(mdev->pclk);
+		devm_clk_put(dev, mdev->pclk);
+		mdev->pclk = NULL;
+	}
+
+	devm_kfree(dev, mdev);
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h
new file mode 100644
index 000000000000..0f77dead6a23
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h
@@ -0,0 +1,110 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _KOMEDA_DEV_H_
+#define _KOMEDA_DEV_H_
+
+#include <linux/device.h>
+#include <linux/clk.h>
+#include "komeda_pipeline.h"
+#include "malidp_product.h"
+#include "komeda_format_caps.h"
+
+/* malidp device id */
+enum {
+	MALI_D71 = 0,
+};
+
+/* pipeline DT ports */
+enum {
+	KOMEDA_OF_PORT_OUTPUT		= 0,
+	KOMEDA_OF_PORT_COPROC		= 1,
+};
+
+struct komeda_chip_info {
+	u32 arch_id;
+	u32 core_id;
+	u32 core_info;
+	u32 bus_width;
+};
+
+struct komeda_product_data {
+	u32 product_id;
+	struct komeda_dev_funcs *(*identify)(u32 __iomem *reg,
+					     struct komeda_chip_info *info);
+};
+
+struct komeda_dev;
+
+/**
+ * struct komeda_dev_funcs
+ *
+ * Supplied by chip level and returned by the chip entry function xxx_identify,
+ */
+struct komeda_dev_funcs {
+	/**
+	 * @init_format_table:
+	 *
+	 * initialize &komeda_dev->format_table, this function should be called
+	 * before the &enum_resource
+	 */
+	void (*init_format_table)(struct komeda_dev *mdev);
+	/**
+	 * @enum_resources:
+	 *
+	 * for CHIP to report or add pipeline and component resources to CORE
+	 */
+	int (*enum_resources)(struct komeda_dev *mdev);
+	/** @cleanup: call to chip to cleanup komeda_dev->chip data */
+	void (*cleanup)(struct komeda_dev *mdev);
+};
+
+/**
+ * struct komeda_dev
+ *
+ * Pipeline and component are used to describe how to handle the pixel data.
+ * komeda_device is for describing the whole view of the device, and the
+ * control-abilites of device.
+ */
+struct komeda_dev {
+	struct device *dev;
+	u32 __iomem   *reg_base;
+
+	struct komeda_chip_info chip;
+	/** @fmt_tbl: initialized by &komeda_dev_funcs->init_format_table */
+	struct komeda_format_caps_table fmt_tbl;
+	/** @pclk: APB clock for register access */
+	struct clk *pclk;
+	/** @mck: HW main engine clk */
+	struct clk *mclk;
+
+	int n_pipelines;
+	struct komeda_pipeline *pipelines[KOMEDA_MAX_PIPELINES];
+
+	/** @funcs: chip funcs to access to HW */
+	struct komeda_dev_funcs *funcs;
+	/**
+	 * @chip_data:
+	 *
+	 * chip data will be added by &komeda_dev_funcs.enum_resources() and
+	 * destroyed by &komeda_dev_funcs.cleanup()
+	 */
+	void *chip_data;
+};
+
+static inline bool
+komeda_product_match(struct komeda_dev *mdev, u32 target)
+{
+	return MALIDP_CORE_ID_PRODUCT_ID(mdev->chip.core_id) == target;
+}
+
+struct komeda_dev_funcs *
+d71_identify(u32 __iomem *reg, struct komeda_chip_info *chip);
+
+struct komeda_dev *komeda_dev_create(struct device *dev);
+void komeda_dev_destroy(struct komeda_dev *mdev);
+
+#endif /*_KOMEDA_DEV_H_*/
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c
new file mode 100644
index 000000000000..2bdd189b041d
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/component.h>
+#include <drm/drm_of.h>
+#include "komeda_dev.h"
+#include "komeda_kms.h"
+
+struct komeda_drv {
+	struct komeda_dev *mdev;
+	struct komeda_kms_dev *kms;
+};
+
+static void komeda_unbind(struct device *dev)
+{
+	struct komeda_drv *mdrv = dev_get_drvdata(dev);
+
+	if (!mdrv)
+		return;
+
+	komeda_kms_detach(mdrv->kms);
+	komeda_dev_destroy(mdrv->mdev);
+
+	dev_set_drvdata(dev, NULL);
+	devm_kfree(dev, mdrv);
+}
+
+static int komeda_bind(struct device *dev)
+{
+	struct komeda_drv *mdrv;
+	int err;
+
+	mdrv = devm_kzalloc(dev, sizeof(*mdrv), GFP_KERNEL);
+	if (!mdrv)
+		return -ENOMEM;
+
+	mdrv->mdev = komeda_dev_create(dev);
+	if (IS_ERR(mdrv->mdev)) {
+		err = PTR_ERR(mdrv->mdev);
+		goto free_mdrv;
+	}
+
+	mdrv->kms = komeda_kms_attach(mdrv->mdev);
+	if (IS_ERR(mdrv->kms)) {
+		err = PTR_ERR(mdrv->kms);
+		goto destroy_mdev;
+	}
+
+	dev_set_drvdata(dev, mdrv);
+
+	return 0;
+
+destroy_mdev:
+	komeda_dev_destroy(mdrv->mdev);
+
+free_mdrv:
+	devm_kfree(dev, mdrv);
+	return err;
+}
+
+static const struct component_master_ops komeda_master_ops = {
+	.bind	= komeda_bind,
+	.unbind	= komeda_unbind,
+};
+
+static int compare_of(struct device *dev, void *data)
+{
+	return dev->of_node == data;
+}
+
+static void komeda_add_slave(struct device *master,
+			     struct component_match **match,
+			     struct device_node *np, int port)
+{
+	struct device_node *remote;
+
+	remote = of_graph_get_remote_node(np, port, 0);
+	if (remote) {
+		drm_of_component_match_add(master, match, compare_of, remote);
+		of_node_put(remote);
+	}
+}
+
+static int komeda_platform_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct component_match *match = NULL;
+	struct device_node *child;
+
+	if (!dev->of_node)
+		return -ENODEV;
+
+	for_each_available_child_of_node(dev->of_node, child) {
+		if (of_node_cmp(child->name, "pipeline") != 0)
+			continue;
+
+		/* add connector */
+		komeda_add_slave(dev, &match, child, KOMEDA_OF_PORT_OUTPUT);
+	}
+
+	return component_master_add_with_match(dev, &komeda_master_ops, match);
+}
+
+static int komeda_platform_remove(struct platform_device *pdev)
+{
+	component_master_del(&pdev->dev, &komeda_master_ops);
+	return 0;
+}
+
+static const struct komeda_product_data komeda_products[] = {
+	[MALI_D71] = {
+		.product_id = MALIDP_D71_PRODUCT_ID,
+		.identify = d71_identify,
+	},
+};
+
+const struct of_device_id komeda_of_match[] = {
+	{ .compatible = "arm,mali-d71", .data = &komeda_products[MALI_D71], },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, komeda_of_match);
+
+static struct platform_driver komeda_platform_driver = {
+	.probe	= komeda_platform_probe,
+	.remove	= komeda_platform_remove,
+	.driver	= {
+		.name = "komeda",
+		.of_match_table	= komeda_of_match,
+		.pm = NULL,
+	},
+};
+
+module_platform_driver(komeda_platform_driver);
+
+MODULE_AUTHOR("James.Qian.Wang <james.qian.wang@arm.com>");
+MODULE_DESCRIPTION("Komeda KMS driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.c b/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.c
new file mode 100644
index 000000000000..1e17bd6107a4
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.c
@@ -0,0 +1,75 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+
+#include <linux/slab.h>
+#include "komeda_format_caps.h"
+#include "malidp_utils.h"
+
+const struct komeda_format_caps *
+komeda_get_format_caps(struct komeda_format_caps_table *table,
+		       u32 fourcc, u64 modifier)
+{
+	const struct komeda_format_caps *caps;
+	u64 afbc_features = modifier & ~(AFBC_FORMAT_MOD_BLOCK_SIZE_MASK);
+	u32 afbc_layout = modifier & AFBC_FORMAT_MOD_BLOCK_SIZE_MASK;
+	int id;
+
+	for (id = 0; id < table->n_formats; id++) {
+		caps = &table->format_caps[id];
+
+		if (fourcc != caps->fourcc)
+			continue;
+
+		if ((modifier == 0ULL) && (caps->supported_afbc_layouts == 0))
+			return caps;
+
+		if (has_bits(afbc_features, caps->supported_afbc_features) &&
+		    has_bit(afbc_layout, caps->supported_afbc_layouts))
+			return caps;
+	}
+
+	return NULL;
+}
+
+u32 *komeda_get_layer_fourcc_list(struct komeda_format_caps_table *table,
+				  u32 layer_type, u32 *n_fmts)
+{
+	const struct komeda_format_caps *cap;
+	u32 *fmts;
+	int i, j, n = 0;
+
+	fmts = kcalloc(table->n_formats, sizeof(u32), GFP_KERNEL);
+	if (!fmts)
+		return NULL;
+
+	for (i = 0; i < table->n_formats; i++) {
+		cap = &table->format_caps[i];
+		if (!(layer_type & cap->supported_layer_types) ||
+		    (cap->fourcc == 0))
+			continue;
+
+		/* one fourcc may has two caps items in table (afbc/none-afbc),
+		 * so check the existing list to avoid adding a duplicated one.
+		 */
+		for (j = n - 1; j >= 0; j--)
+			if (fmts[j] == cap->fourcc)
+				break;
+
+		if (j < 0)
+			fmts[n++] = cap->fourcc;
+	}
+
+	if (n_fmts)
+		*n_fmts = n;
+
+	return fmts;
+}
+
+void komeda_put_fourcc_list(u32 *fourcc_list)
+{
+	kfree(fourcc_list);
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.h b/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.h
new file mode 100644
index 000000000000..60f39e77b098
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_format_caps.h
@@ -0,0 +1,89 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+
+#ifndef _KOMEDA_FORMAT_CAPS_H_
+#define _KOMEDA_FORMAT_CAPS_H_
+
+#include <linux/types.h>
+#include <uapi/drm/drm_fourcc.h>
+#include <drm/drm_fourcc.h>
+
+#define AFBC(x)		DRM_FORMAT_MOD_ARM_AFBC(x)
+
+/* afbc layerout */
+#define AFBC_16x16(x)	AFBC(AFBC_FORMAT_MOD_BLOCK_SIZE_16x16 | (x))
+#define AFBC_32x8(x)	AFBC(AFBC_FORMAT_MOD_BLOCK_SIZE_32x8 | (x))
+
+/* afbc features */
+#define _YTR		AFBC_FORMAT_MOD_YTR
+#define _SPLIT		AFBC_FORMAT_MOD_SPLIT
+#define _SPARSE		AFBC_FORMAT_MOD_SPARSE
+#define _CBR		AFBC_FORMAT_MOD_CBR
+#define _TILED		AFBC_FORMAT_MOD_TILED
+#define _SC		AFBC_FORMAT_MOD_SC
+
+/* layer_type */
+#define KOMEDA_FMT_RICH_LAYER		BIT(0)
+#define KOMEDA_FMT_SIMPLE_LAYER		BIT(1)
+#define KOMEDA_FMT_WB_LAYER		BIT(2)
+
+#define AFBC_TH_LAYOUT_ALIGNMENT	8
+#define AFBC_HEADER_SIZE		16
+#define AFBC_SUPERBLK_ALIGNMENT		128
+#define AFBC_SUPERBLK_PIXELS		256
+#define AFBC_BODY_START_ALIGNMENT	1024
+#define AFBC_TH_BODY_START_ALIGNMENT	4096
+
+/**
+ * struct komeda_format_caps
+ *
+ * komeda_format_caps is for describing ARM display specific features and
+ * limitations for a specific format, and format_caps will be linked into
+ * &komeda_framebuffer like a extension of &drm_format_info.
+ *
+ * NOTE: one fourcc may has two different format_caps items for fourcc and
+ * fourcc+modifier
+ *
+ * @hw_id: hw format id, hw specific value.
+ * @fourcc: drm fourcc format.
+ * @tile_size: format tiled size, used by ARM format X0L0/X0L2
+ * @supported_layer_types: indicate which layer supports this format
+ * @supported_rots: allowed rotations for this format
+ * @supported_afbc_layouts: supported afbc layerout
+ * @supported_afbc_features: supported afbc features
+ */
+struct komeda_format_caps {
+	u32 hw_id;
+	u32 fourcc;
+	u32 tile_size;
+	u32 supported_layer_types;
+	u32 supported_rots;
+	u32 supported_afbc_layouts;
+	u64 supported_afbc_features;
+};
+
+/**
+ * struct komeda_format_caps_table - format_caps mananger
+ *
+ * @n_formats: the size of format_caps list.
+ * @format_caps: format_caps list.
+ */
+struct komeda_format_caps_table {
+	u32 n_formats;
+	const struct komeda_format_caps *format_caps;
+};
+
+const struct komeda_format_caps *
+komeda_get_format_caps(struct komeda_format_caps_table *table,
+		       u32 fourcc, u64 modifier);
+
+u32 *komeda_get_layer_fourcc_list(struct komeda_format_caps_table *table,
+				  u32 layer_type, u32 *n_fmts);
+
+void komeda_put_fourcc_list(u32 *fourcc_list);
+
+#endif
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c
new file mode 100644
index 000000000000..9cc9935024f7
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c
@@ -0,0 +1,167 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <drm/drm_device.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+
+#include "komeda_framebuffer.h"
+#include "komeda_dev.h"
+
+static void komeda_fb_destroy(struct drm_framebuffer *fb)
+{
+	struct komeda_fb *kfb = to_kfb(fb);
+	u32 i;
+
+	for (i = 0; i < fb->format->num_planes; i++)
+		drm_gem_object_put_unlocked(fb->obj[i]);
+
+	drm_framebuffer_cleanup(fb);
+	kfree(kfb);
+}
+
+static int komeda_fb_create_handle(struct drm_framebuffer *fb,
+				   struct drm_file *file, u32 *handle)
+{
+	return drm_gem_handle_create(file, fb->obj[0], handle);
+}
+
+static const struct drm_framebuffer_funcs komeda_fb_funcs = {
+	.destroy	= komeda_fb_destroy,
+	.create_handle	= komeda_fb_create_handle,
+};
+
+static int
+komeda_fb_none_afbc_size_check(struct komeda_dev *mdev, struct komeda_fb *kfb,
+			       struct drm_file *file,
+			       const struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	struct drm_framebuffer *fb = &kfb->base;
+	struct drm_gem_object *obj;
+	u32 min_size = 0;
+	u32 i;
+
+	for (i = 0; i < fb->format->num_planes; i++) {
+		obj = drm_gem_object_lookup(file, mode_cmd->handles[i]);
+		if (!obj) {
+			DRM_DEBUG_KMS("Failed to lookup GEM object\n");
+			fb->obj[i] = NULL;
+
+			return -ENOENT;
+		}
+
+		kfb->aligned_w = fb->width / (i ? fb->format->hsub : 1);
+		kfb->aligned_h = fb->height / (i ? fb->format->vsub : 1);
+
+		if (fb->pitches[i] % mdev->chip.bus_width) {
+			DRM_DEBUG_KMS("Pitch[%d]: 0x%x doesn't align to 0x%x\n",
+				      i, fb->pitches[i], mdev->chip.bus_width);
+			drm_gem_object_put_unlocked(obj);
+			fb->obj[i] = NULL;
+
+			return -EINVAL;
+		}
+
+		min_size = ((kfb->aligned_h / kfb->format_caps->tile_size - 1)
+			    * fb->pitches[i])
+			    + (kfb->aligned_w * fb->format->cpp[i]
+			       * kfb->format_caps->tile_size)
+			    + fb->offsets[i];
+
+		if (obj->size < min_size) {
+			DRM_DEBUG_KMS("Fail to check none afbc fb size.\n");
+			drm_gem_object_put_unlocked(obj);
+			fb->obj[i] = NULL;
+
+			return -EINVAL;
+		}
+
+		fb->obj[i] = obj;
+	}
+
+	if (fb->format->num_planes == 3) {
+		if (fb->pitches[1] != fb->pitches[2]) {
+			DRM_DEBUG_KMS("The pitch[1] and [2] are not same\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+struct drm_framebuffer *
+komeda_fb_create(struct drm_device *dev, struct drm_file *file,
+		 const struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	struct komeda_dev *mdev = dev->dev_private;
+	struct komeda_fb *kfb;
+	int ret = 0, i;
+
+	kfb = kzalloc(sizeof(*kfb), GFP_KERNEL);
+	if (!kfb)
+		return ERR_PTR(-ENOMEM);
+
+	kfb->format_caps = komeda_get_format_caps(&mdev->fmt_tbl,
+						  mode_cmd->pixel_format,
+						  mode_cmd->modifier[0]);
+	if (!kfb->format_caps) {
+		DRM_DEBUG_KMS("FMT %x is not supported.\n",
+			      mode_cmd->pixel_format);
+		kfree(kfb);
+		return ERR_PTR(-EINVAL);
+	}
+
+	drm_helper_mode_fill_fb_struct(dev, &kfb->base, mode_cmd);
+
+	ret = komeda_fb_none_afbc_size_check(mdev, kfb, file, mode_cmd);
+	if (ret < 0)
+		goto err_cleanup;
+
+	ret = drm_framebuffer_init(dev, &kfb->base, &komeda_fb_funcs);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("failed to initialize fb\n");
+
+		goto err_cleanup;
+	}
+
+	return &kfb->base;
+
+err_cleanup:
+	for (i = 0; i < kfb->base.format->num_planes; i++)
+		drm_gem_object_put_unlocked(kfb->base.obj[i]);
+
+	kfree(kfb);
+	return ERR_PTR(ret);
+}
+
+dma_addr_t
+komeda_fb_get_pixel_addr(struct komeda_fb *kfb, int x, int y, int plane)
+{
+	struct drm_framebuffer *fb = &kfb->base;
+	const struct drm_gem_cma_object *obj;
+	u32 plane_x, plane_y, cpp, pitch, offset;
+
+	if (plane >= fb->format->num_planes) {
+		DRM_DEBUG_KMS("Out of max plane num.\n");
+		return -EINVAL;
+	}
+
+	obj = drm_fb_cma_get_gem_obj(fb, plane);
+
+	offset = fb->offsets[plane];
+	if (!fb->modifier) {
+		plane_x = x / (plane ? fb->format->hsub : 1);
+		plane_y = y / (plane ? fb->format->vsub : 1);
+		cpp = fb->format->cpp[plane];
+		pitch = fb->pitches[plane];
+		offset += plane_x * cpp *  kfb->format_caps->tile_size +
+				(plane_y * pitch) / kfb->format_caps->tile_size;
+	}
+
+	return obj->paddr + offset;
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h
new file mode 100644
index 000000000000..0de2e4a2afd2
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _KOMEDA_FRAMEBUFFER_H_
+#define _KOMEDA_FRAMEBUFFER_H_
+
+#include <drm/drm_framebuffer.h>
+#include "komeda_format_caps.h"
+
+/** struct komeda_fb - entend drm_framebuffer with komeda attribute */
+struct komeda_fb {
+	/** @base: &drm_framebuffer */
+	struct drm_framebuffer base;
+	/* @format_caps: &komeda_format_caps */
+	const struct komeda_format_caps *format_caps;
+	/** @aligned_w: aligned frame buffer width */
+	u32 aligned_w;
+	/** @aligned_h: aligned frame buffer height */
+	u32 aligned_h;
+};
+
+#define to_kfb(dfb)	container_of(dfb, struct komeda_fb, base)
+
+struct drm_framebuffer *
+komeda_fb_create(struct drm_device *dev, struct drm_file *file,
+		 const struct drm_mode_fb_cmd2 *mode_cmd);
+dma_addr_t
+komeda_fb_get_pixel_addr(struct komeda_fb *kfb, int x, int y, int plane);
+bool komeda_fb_is_layer_supported(struct komeda_fb *kfb, u32 layer_type);
+
+#endif
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c
new file mode 100644
index 000000000000..47a58ab20434
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c
@@ -0,0 +1,171 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <linux/component.h>
+#include <linux/interrupt.h>
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.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_vblank.h>
+
+#include "komeda_dev.h"
+#include "komeda_framebuffer.h"
+#include "komeda_kms.h"
+
+DEFINE_DRM_GEM_CMA_FOPS(komeda_cma_fops);
+
+static int komeda_gem_cma_dumb_create(struct drm_file *file,
+				      struct drm_device *dev,
+				      struct drm_mode_create_dumb *args)
+{
+	u32 alignment = 16; /* TODO get alignment from dev */
+
+	args->pitch = ALIGN(DIV_ROUND_UP(args->width * args->bpp, 8),
+			    alignment);
+
+	return drm_gem_cma_dumb_create_internal(file, dev, args);
+}
+
+static struct drm_driver komeda_kms_driver = {
+	.driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC |
+			   DRIVER_PRIME,
+	.lastclose			= drm_fb_helper_lastclose,
+	.gem_free_object_unlocked	= drm_gem_cma_free_object,
+	.gem_vm_ops			= &drm_gem_cma_vm_ops,
+	.dumb_create			= komeda_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_export		= drm_gem_prime_export,
+	.gem_prime_import		= drm_gem_prime_import,
+	.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 = &komeda_cma_fops,
+	.name = "komeda",
+	.desc = "Arm Komeda Display Processor driver",
+	.date = "20181101",
+	.major = 0,
+	.minor = 1,
+};
+
+static void komeda_kms_commit_tail(struct drm_atomic_state *old_state)
+{
+	struct drm_device *dev = old_state->dev;
+
+	drm_atomic_helper_commit_modeset_disables(dev, old_state);
+
+	drm_atomic_helper_commit_planes(dev, old_state, 0);
+
+	drm_atomic_helper_commit_modeset_enables(dev, old_state);
+
+	drm_atomic_helper_wait_for_flip_done(dev, old_state);
+
+	drm_atomic_helper_commit_hw_done(old_state);
+
+	drm_atomic_helper_cleanup_planes(dev, old_state);
+}
+
+static const struct drm_mode_config_helper_funcs komeda_mode_config_helpers = {
+	.atomic_commit_tail = komeda_kms_commit_tail,
+};
+
+static const struct drm_mode_config_funcs komeda_mode_config_funcs = {
+	.fb_create		= komeda_fb_create,
+	.atomic_check		= drm_atomic_helper_check,
+	.atomic_commit		= drm_atomic_helper_commit,
+};
+
+static void komeda_kms_mode_config_init(struct komeda_kms_dev *kms,
+					struct komeda_dev *mdev)
+{
+	struct drm_mode_config *config = &kms->base.mode_config;
+
+	drm_mode_config_init(&kms->base);
+
+	komeda_kms_setup_crtcs(kms, mdev);
+
+	/* Get value from dev */
+	config->min_width	= 0;
+	config->min_height	= 0;
+	config->max_width	= 4096;
+	config->max_height	= 4096;
+	config->allow_fb_modifiers = false;
+
+	config->funcs = &komeda_mode_config_funcs;
+	config->helper_private = &komeda_mode_config_helpers;
+}
+
+struct komeda_kms_dev *komeda_kms_attach(struct komeda_dev *mdev)
+{
+	struct komeda_kms_dev *kms = kzalloc(sizeof(*kms), GFP_KERNEL);
+	struct drm_device *drm;
+	int err;
+
+	if (!kms)
+		return ERR_PTR(-ENOMEM);
+
+	drm = &kms->base;
+	err = drm_dev_init(drm, &komeda_kms_driver, mdev->dev);
+	if (err)
+		goto free_kms;
+
+	drm->dev_private = mdev;
+
+	komeda_kms_mode_config_init(kms, mdev);
+
+	err = komeda_kms_add_private_objs(kms, mdev);
+	if (err)
+		goto cleanup_mode_config;
+
+	err = komeda_kms_add_planes(kms, mdev);
+	if (err)
+		goto cleanup_mode_config;
+
+	err = drm_vblank_init(drm, kms->n_crtcs);
+	if (err)
+		goto cleanup_mode_config;
+
+	err = komeda_kms_add_crtcs(kms, mdev);
+	if (err)
+		goto cleanup_mode_config;
+
+	err = component_bind_all(mdev->dev, kms);
+	if (err)
+		goto cleanup_mode_config;
+
+	drm_mode_config_reset(drm);
+
+	err = drm_dev_register(drm, 0);
+	if (err)
+		goto cleanup_mode_config;
+
+	return kms;
+
+cleanup_mode_config:
+	drm_mode_config_cleanup(drm);
+free_kms:
+	kfree(kms);
+	return ERR_PTR(err);
+}
+
+void komeda_kms_detach(struct komeda_kms_dev *kms)
+{
+	struct drm_device *drm = &kms->base;
+	struct komeda_dev *mdev = drm->dev_private;
+
+	drm_dev_unregister(drm);
+	component_unbind_all(mdev->dev, drm);
+	komeda_kms_cleanup_private_objs(mdev);
+	drm_mode_config_cleanup(drm);
+	drm->dev_private = NULL;
+	drm_dev_put(drm);
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.h b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h
new file mode 100644
index 000000000000..874e9c9f0749
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h
@@ -0,0 +1,114 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _KOMEDA_KMS_H_
+#define _KOMEDA_KMS_H_
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_writeback.h>
+
+/** struct komeda_plane - komeda instance of drm_plane */
+struct komeda_plane {
+	/** @base: &drm_plane */
+	struct drm_plane base;
+	/**
+	 * @layer:
+	 *
+	 * represents available layer input pipelines for this plane.
+	 *
+	 * NOTE:
+	 * the layer is not for a specific Layer, but indicate a group of
+	 * Layers with same capabilities.
+	 */
+	struct komeda_layer *layer;
+};
+
+/**
+ * struct komeda_plane_state
+ *
+ * The plane_state can be split into two data flow (left/right) and handled
+ * by two layers &komeda_plane.layer and &komeda_plane.layer.right
+ */
+struct komeda_plane_state {
+	/** @base: &drm_plane_state */
+	struct drm_plane_state base;
+
+	/* private properties */
+};
+
+/**
+ * struct komeda_wb_connector
+ */
+struct komeda_wb_connector {
+	/** @base: &drm_writeback_connector */
+	struct drm_writeback_connector base;
+
+	/** @wb_layer: represents associated writeback pipeline of komeda */
+	struct komeda_layer *wb_layer;
+};
+
+/**
+ * struct komeda_crtc
+ */
+struct komeda_crtc {
+	/** @base: &drm_crtc */
+	struct drm_crtc base;
+	/** @master: only master has display output */
+	struct komeda_pipeline *master;
+	/**
+	 * @slave: optional
+	 *
+	 * Doesn't have its own display output, the handled data flow will
+	 * merge into the master.
+	 */
+	struct komeda_pipeline *slave;
+};
+
+/** struct komeda_crtc_state */
+struct komeda_crtc_state {
+	/** @base: &drm_crtc_state */
+	struct drm_crtc_state base;
+
+	/* private properties */
+
+	/* computed state which are used by validate/check */
+	u32 affected_pipes;
+	u32 active_pipes;
+};
+
+/** struct komeda_kms_dev - for gather KMS related things */
+struct komeda_kms_dev {
+	/** @base: &drm_device */
+	struct drm_device base;
+
+	/** @n_crtcs: valid numbers of crtcs in &komeda_kms_dev.crtcs */
+	int n_crtcs;
+	/** @crtcs: crtcs list */
+	struct komeda_crtc crtcs[KOMEDA_MAX_PIPELINES];
+};
+
+#define to_kplane(p)	container_of(p, struct komeda_plane, base)
+#define to_kplane_st(p)	container_of(p, struct komeda_plane_state, base)
+#define to_kconn(p)	container_of(p, struct komeda_wb_connector, base)
+#define to_kcrtc(p)	container_of(p, struct komeda_crtc, base)
+#define to_kcrtc_st(p)	container_of(p, struct komeda_crtc_state, base)
+#define to_kdev(p)	container_of(p, struct komeda_kms_dev, base)
+
+int komeda_kms_setup_crtcs(struct komeda_kms_dev *kms, struct komeda_dev *mdev);
+
+int komeda_kms_add_crtcs(struct komeda_kms_dev *kms, struct komeda_dev *mdev);
+int komeda_kms_add_planes(struct komeda_kms_dev *kms, struct komeda_dev *mdev);
+int komeda_kms_add_private_objs(struct komeda_kms_dev *kms,
+				struct komeda_dev *mdev);
+void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev);
+
+struct komeda_kms_dev *komeda_kms_attach(struct komeda_dev *mdev);
+void komeda_kms_detach(struct komeda_kms_dev *kms);
+
+#endif /*_KOMEDA_KMS_H_*/
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c
new file mode 100644
index 000000000000..f1908e9ef128
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <drm/drm_print.h>
+
+#include "komeda_dev.h"
+#include "komeda_pipeline.h"
+
+/** komeda_pipeline_add - Add a pipeline to &komeda_dev */
+struct komeda_pipeline *
+komeda_pipeline_add(struct komeda_dev *mdev, size_t size,
+		    struct komeda_pipeline_funcs *funcs)
+{
+	struct komeda_pipeline *pipe;
+
+	if (mdev->n_pipelines + 1 > KOMEDA_MAX_PIPELINES) {
+		DRM_ERROR("Exceed max support %d pipelines.\n",
+			  KOMEDA_MAX_PIPELINES);
+		return NULL;
+	}
+
+	if (size < sizeof(*pipe)) {
+		DRM_ERROR("Request pipeline size too small.\n");
+		return NULL;
+	}
+
+	pipe = devm_kzalloc(mdev->dev, size, GFP_KERNEL);
+	if (!pipe)
+		return NULL;
+
+	pipe->mdev = mdev;
+	pipe->id   = mdev->n_pipelines;
+	pipe->funcs = funcs;
+
+	mdev->pipelines[mdev->n_pipelines] = pipe;
+	mdev->n_pipelines++;
+
+	return pipe;
+}
+
+void komeda_pipeline_destroy(struct komeda_dev *mdev,
+			     struct komeda_pipeline *pipe)
+{
+	struct komeda_component *c;
+	int i;
+
+	dp_for_each_set_bit(i, pipe->avail_comps) {
+		c = komeda_pipeline_get_component(pipe, i);
+		komeda_component_destroy(mdev, c);
+	}
+
+	clk_put(pipe->pxlclk);
+	clk_put(pipe->aclk);
+
+	of_node_put(pipe->of_output_dev);
+	of_node_put(pipe->of_output_port);
+	of_node_put(pipe->of_node);
+
+	devm_kfree(mdev->dev, pipe);
+}
+
+struct komeda_component **
+komeda_pipeline_get_component_pos(struct komeda_pipeline *pipe, int id)
+{
+	struct komeda_dev *mdev = pipe->mdev;
+	struct komeda_pipeline *temp = NULL;
+	struct komeda_component **pos = NULL;
+
+	switch (id) {
+	case KOMEDA_COMPONENT_LAYER0:
+	case KOMEDA_COMPONENT_LAYER1:
+	case KOMEDA_COMPONENT_LAYER2:
+	case KOMEDA_COMPONENT_LAYER3:
+		pos = to_cpos(pipe->layers[id - KOMEDA_COMPONENT_LAYER0]);
+		break;
+	case KOMEDA_COMPONENT_WB_LAYER:
+		pos = to_cpos(pipe->wb_layer);
+		break;
+	case KOMEDA_COMPONENT_COMPIZ0:
+	case KOMEDA_COMPONENT_COMPIZ1:
+		temp = mdev->pipelines[id - KOMEDA_COMPONENT_COMPIZ0];
+		if (!temp) {
+			DRM_ERROR("compiz-%d doesn't exist.\n", id);
+			return NULL;
+		}
+		pos = to_cpos(temp->compiz);
+		break;
+	case KOMEDA_COMPONENT_SCALER0:
+	case KOMEDA_COMPONENT_SCALER1:
+		pos = to_cpos(pipe->scalers[id - KOMEDA_COMPONENT_SCALER0]);
+		break;
+	case KOMEDA_COMPONENT_IPS0:
+	case KOMEDA_COMPONENT_IPS1:
+		temp = mdev->pipelines[id - KOMEDA_COMPONENT_IPS0];
+		if (!temp) {
+			DRM_ERROR("ips-%d doesn't exist.\n", id);
+			return NULL;
+		}
+		pos = to_cpos(temp->improc);
+		break;
+	case KOMEDA_COMPONENT_TIMING_CTRLR:
+		pos = to_cpos(pipe->ctrlr);
+		break;
+	default:
+		pos = NULL;
+		DRM_ERROR("Unknown pipeline resource ID: %d.\n", id);
+		break;
+	}
+
+	return pos;
+}
+
+struct komeda_component *
+komeda_pipeline_get_component(struct komeda_pipeline *pipe, int id)
+{
+	struct komeda_component **pos = NULL;
+	struct komeda_component *c = NULL;
+
+	pos = komeda_pipeline_get_component_pos(pipe, id);
+	if (pos)
+		c = *pos;
+
+	return c;
+}
+
+/** komeda_component_add - Add a component to &komeda_pipeline */
+struct komeda_component *
+komeda_component_add(struct komeda_pipeline *pipe,
+		     size_t comp_sz, u32 id, u32 hw_id,
+		     struct komeda_component_funcs *funcs,
+		     u8 max_active_inputs, u32 supported_inputs,
+		     u8 max_active_outputs, u32 __iomem *reg,
+		     const char *name_fmt, ...)
+{
+	struct komeda_component **pos;
+	struct komeda_component *c;
+	int idx, *num = NULL;
+
+	if (max_active_inputs > KOMEDA_COMPONENT_N_INPUTS) {
+		WARN(1, "please large KOMEDA_COMPONENT_N_INPUTS to %d.\n",
+		     max_active_inputs);
+		return NULL;
+	}
+
+	pos = komeda_pipeline_get_component_pos(pipe, id);
+	if (!pos || (*pos))
+		return NULL;
+
+	if (has_bit(id, KOMEDA_PIPELINE_LAYERS)) {
+		idx = id - KOMEDA_COMPONENT_LAYER0;
+		num = &pipe->n_layers;
+		if (idx != pipe->n_layers) {
+			DRM_ERROR("please add Layer by id sequence.\n");
+			return NULL;
+		}
+	} else if (has_bit(id,  KOMEDA_PIPELINE_SCALERS)) {
+		idx = id - KOMEDA_COMPONENT_SCALER0;
+		num = &pipe->n_scalers;
+		if (idx != pipe->n_scalers) {
+			DRM_ERROR("please add Scaler by id sequence.\n");
+			return NULL;
+		}
+	}
+
+	c = devm_kzalloc(pipe->mdev->dev, comp_sz, GFP_KERNEL);
+	if (!c)
+		return NULL;
+
+	c->id = id;
+	c->hw_id = hw_id;
+	c->reg = reg;
+	c->pipeline = pipe;
+	c->max_active_inputs = max_active_inputs;
+	c->max_active_outputs = max_active_outputs;
+	c->supported_inputs = supported_inputs;
+	c->funcs = funcs;
+
+	if (name_fmt) {
+		va_list args;
+
+		va_start(args, name_fmt);
+		vsnprintf(c->name, sizeof(c->name), name_fmt, args);
+		va_end(args);
+	}
+
+	if (num)
+		*num = *num + 1;
+
+	pipe->avail_comps |= BIT(c->id);
+	*pos = c;
+
+	return c;
+}
+
+void komeda_component_destroy(struct komeda_dev *mdev,
+			      struct komeda_component *c)
+{
+	devm_kfree(mdev->dev, c);
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h
new file mode 100644
index 000000000000..8c950bc8ae96
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h
@@ -0,0 +1,359 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _KOMEDA_PIPELINE_H_
+#define _KOMEDA_PIPELINE_H_
+
+#include <linux/types.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include "malidp_utils.h"
+
+#define KOMEDA_MAX_PIPELINES		2
+#define KOMEDA_PIPELINE_MAX_LAYERS	4
+#define KOMEDA_PIPELINE_MAX_SCALERS	2
+#define KOMEDA_COMPONENT_N_INPUTS	5
+
+/* pipeline component IDs */
+enum {
+	KOMEDA_COMPONENT_LAYER0		= 0,
+	KOMEDA_COMPONENT_LAYER1		= 1,
+	KOMEDA_COMPONENT_LAYER2		= 2,
+	KOMEDA_COMPONENT_LAYER3		= 3,
+	KOMEDA_COMPONENT_WB_LAYER	= 7, /* write back layer */
+	KOMEDA_COMPONENT_SCALER0	= 8,
+	KOMEDA_COMPONENT_SCALER1	= 9,
+	KOMEDA_COMPONENT_SPLITTER	= 12,
+	KOMEDA_COMPONENT_MERGER		= 14,
+	KOMEDA_COMPONENT_COMPIZ0	= 16, /* compositor */
+	KOMEDA_COMPONENT_COMPIZ1	= 17,
+	KOMEDA_COMPONENT_IPS0		= 20, /* post image processor */
+	KOMEDA_COMPONENT_IPS1		= 21,
+	KOMEDA_COMPONENT_TIMING_CTRLR	= 22, /* timing controller */
+};
+
+#define KOMEDA_PIPELINE_LAYERS		(BIT(KOMEDA_COMPONENT_LAYER0) |\
+					 BIT(KOMEDA_COMPONENT_LAYER1) |\
+					 BIT(KOMEDA_COMPONENT_LAYER2) |\
+					 BIT(KOMEDA_COMPONENT_LAYER3))
+
+#define KOMEDA_PIPELINE_SCALERS		(BIT(KOMEDA_COMPONENT_SCALER0) |\
+					 BIT(KOMEDA_COMPONENT_SCALER1))
+
+#define KOMEDA_PIPELINE_COMPIZS		(BIT(KOMEDA_COMPONENT_COMPIZ0) |\
+					 BIT(KOMEDA_COMPONENT_COMPIZ1))
+
+#define KOMEDA_PIPELINE_IMPROCS		(BIT(KOMEDA_COMPONENT_IPS0) |\
+					 BIT(KOMEDA_COMPONENT_IPS1))
+struct komeda_component;
+struct komeda_component_state;
+
+/** komeda_component_funcs - component control functions */
+struct komeda_component_funcs {
+	/** @validate: optional,
+	 * component may has special requirements or limitations, this function
+	 * supply HW the ability to do the further HW specific check.
+	 */
+	int (*validate)(struct komeda_component *c,
+			struct komeda_component_state *state);
+	/** @update: update is a active update */
+	void (*update)(struct komeda_component *c,
+		       struct komeda_component_state *state);
+	/** @disable: disable component */
+	void (*disable)(struct komeda_component *c);
+	/** @dump_register: Optional, dump registers to seq_file */
+	void (*dump_register)(struct komeda_component *c, struct seq_file *seq);
+};
+
+/**
+ * struct komeda_component
+ *
+ * struct komeda_component describe the data flow capabilities for how to link a
+ * component into the display pipeline.
+ * all specified components are subclass of this structure.
+ */
+struct komeda_component {
+	/** @obj: treat component as private obj */
+	struct drm_private_obj obj;
+	/** @pipeline: the komeda pipeline this component belongs to */
+	struct komeda_pipeline *pipeline;
+	/** @name: component name */
+	char name[32];
+	/**
+	 * @reg:
+	 * component register base,
+	 * which is initialized by chip and used by chip only
+	 */
+	u32 __iomem *reg;
+	/** @id: component id */
+	u32 id;
+	/** @hw_ic: component hw id,
+	 *  which is initialized by chip and used by chip only
+	 */
+	u32 hw_id;
+
+	/**
+	 * @max_active_inputs:
+	 * @max_active_outpus:
+	 *
+	 * maximum number of inputs/outputs that can be active in the same time
+	 * Note:
+	 * the number isn't the bit number of @supported_inputs or
+	 * @supported_outputs, but may be less than it, since component may not
+	 * support enabling all @supported_inputs/outputs at the same time.
+	 */
+	u8 max_active_inputs;
+	u8 max_active_outputs;
+	/**
+	 * @supported_inputs:
+	 * @supported_outputs:
+	 *
+	 * bitmask of BIT(component->id) for the supported inputs/outputs
+	 * describes the possibilities of how a component is linked into a
+	 * pipeline.
+	 */
+	u32 supported_inputs;
+	u32 supported_outputs;
+
+	/**
+	 * @funcs: chip functions to access HW
+	 */
+	struct komeda_component_funcs *funcs;
+};
+
+/**
+ * struct komeda_component_output
+ *
+ * a component has multiple outputs, if want to know where the data
+ * comes from, only know the component is not enough, we still need to know
+ * its output port
+ */
+struct komeda_component_output {
+	/** @component: indicate which component the data comes from */
+	struct komeda_component *component;
+	/** @output_port:
+	 * the output port of the &komeda_component_output.component
+	 */
+	u8 output_port;
+};
+
+/**
+ * struct komeda_component_state
+ *
+ * component_state is the data flow configuration of the component, and it's
+ * the superclass of all specific component_state like @komeda_layer_state,
+ * @komeda_scaler_state
+ */
+struct komeda_component_state {
+	/** @obj: tracking component_state by drm_atomic_state */
+	struct drm_private_state obj;
+	struct komeda_component *component;
+	/**
+	 * @binding_user:
+	 * currently bound user, the user can be crtc/plane/wb_conn, which is
+	 * valid decided by @component and @inputs
+	 *
+	 * -  Layer: its user always is plane.
+	 * -  compiz/improc/timing_ctrlr: the user is crtc.
+	 * -  wb_layer: wb_conn;
+	 * -  scaler: plane when input is layer, wb_conn if input is compiz.
+	 */
+	union {
+		struct drm_crtc *crtc;
+		struct drm_plane *plane;
+		struct drm_connector *wb_conn;
+		void *binding_user;
+	};
+	/**
+	 * @active_inputs:
+	 *
+	 * active_inputs is bitmask of @inputs index
+	 *
+	 * -  active_inputs = changed_active_inputs + unchanged_active_inputs
+	 * -  affected_inputs = old->active_inputs + new->active_inputs;
+	 * -  disabling_inputs = affected_inputs ^ active_inputs;
+	 * -  changed_inputs = disabling_inputs + changed_active_inputs;
+	 *
+	 * NOTE:
+	 * changed_inputs doesn't include all active_input but only
+	 * @changed_active_inputs, and this bitmask can be used in chip
+	 * level for dirty update.
+	 */
+	u16 active_inputs;
+	u16 changed_active_inputs;
+	u16 affected_inputs;
+	/**
+	 * @inputs:
+	 *
+	 * the specific inputs[i] only valid on BIT(i) has been set in
+	 * @active_inputs, if not the inputs[i] is undefined.
+	 */
+	struct komeda_component_output inputs[KOMEDA_COMPONENT_N_INPUTS];
+};
+
+static inline u16 component_disabling_inputs(struct komeda_component_state *st)
+{
+	return st->affected_inputs ^ st->active_inputs;
+}
+
+static inline u16 component_changed_inputs(struct komeda_component_state *st)
+{
+	return component_disabling_inputs(st) | st->changed_active_inputs;
+}
+
+#define to_comp(__c)	(((__c) == NULL) ? NULL : &((__c)->base))
+#define to_cpos(__c)	((struct komeda_component **)&(__c))
+
+/* these structures are going to be filled in in uture patches */
+struct komeda_layer {
+	struct komeda_component base;
+	/* layer specific features and caps */
+	int layer_type; /* RICH, SIMPLE or WB */
+};
+
+struct komeda_layer_state {
+	struct komeda_component_state base;
+	/* layer specific configuration state */
+};
+
+struct komeda_compiz {
+	struct komeda_component base;
+	/* compiz specific features and caps */
+};
+
+struct komeda_compiz_state {
+	struct komeda_component_state base;
+	/* compiz specific configuration state */
+};
+
+struct komeda_scaler {
+	struct komeda_component base;
+	/* scaler features and caps */
+};
+
+struct komeda_scaler_state {
+	struct komeda_component_state base;
+};
+
+struct komeda_improc {
+	struct komeda_component base;
+};
+
+struct komeda_improc_state {
+	struct komeda_component_state base;
+};
+
+/* display timing controller */
+struct komeda_timing_ctrlr {
+	struct komeda_component base;
+};
+
+struct komeda_timing_ctrlr_state {
+	struct komeda_component_state base;
+};
+
+/** struct komeda_pipeline_funcs */
+struct komeda_pipeline_funcs {
+	/* dump_register: Optional, dump registers to seq_file */
+	void (*dump_register)(struct komeda_pipeline *pipe,
+			      struct seq_file *sf);
+};
+
+/**
+ * struct komeda_pipeline
+ *
+ * Represent a complete display pipeline and hold all functional components.
+ */
+struct komeda_pipeline {
+	/** @obj: link pipeline as private obj of drm_atomic_state */
+	struct drm_private_obj obj;
+	/** @mdev: the parent komeda_dev */
+	struct komeda_dev *mdev;
+	/** @pxlclk: pixel clock */
+	struct clk *pxlclk;
+	/** @aclk: AXI clock */
+	struct clk *aclk;
+	/** @id: pipeline id */
+	int id;
+	/** @avail_comps: available components mask of pipeline */
+	u32 avail_comps;
+	int n_layers;
+	struct komeda_layer *layers[KOMEDA_PIPELINE_MAX_LAYERS];
+	int n_scalers;
+	struct komeda_scaler *scalers[KOMEDA_PIPELINE_MAX_SCALERS];
+	struct komeda_compiz *compiz;
+	struct komeda_layer  *wb_layer;
+	struct komeda_improc *improc;
+	struct komeda_timing_ctrlr *ctrlr;
+	struct komeda_pipeline_funcs *funcs; /* private pipeline functions */
+
+	/** @of_node: pipeline dt node */
+	struct device_node *of_node;
+	/** @of_output_port: pipeline output port */
+	struct device_node *of_output_port;
+	/** @of_output_dev: output connector device node */
+	struct device_node *of_output_dev;
+};
+
+/**
+ * struct komeda_pipeline_state
+ *
+ * NOTE:
+ * Unlike the pipeline, pipeline_state doesn’t gather any component_state
+ * into it. It because all component will be managed by drm_atomic_state.
+ */
+struct komeda_pipeline_state {
+	/** @obj: tracking pipeline_state by drm_atomic_state */
+	struct drm_private_state obj;
+	struct komeda_pipeline *pipe;
+	/** @crtc: currently bound crtc */
+	struct drm_crtc *crtc;
+	/**
+	 * @active_comps:
+	 *
+	 * bitmask - BIT(component->id) of active components
+	 */
+	u32 active_comps;
+};
+
+#define to_layer(c)	container_of(c, struct komeda_layer, base)
+#define to_compiz(c)	container_of(c, struct komeda_compiz, base)
+#define to_scaler(c)	container_of(c, struct komeda_scaler, base)
+#define to_improc(c)	container_of(c, struct komeda_improc, base)
+#define to_ctrlr(c)	container_of(c, struct komeda_timing_ctrlr, base)
+
+#define to_layer_st(c)	container_of(c, struct komeda_layer_state, base)
+#define to_compiz_st(c)	container_of(c, struct komeda_compiz_state, base)
+#define to_scaler_st(c) container_of(c, struct komeda_scaler_state, base)
+#define to_improc_st(c)	container_of(c, struct komeda_improc_state, base)
+#define to_ctrlr_st(c)	container_of(c, struct komeda_timing_ctrlr_state, base)
+
+#define priv_to_comp_st(o) container_of(o, struct komeda_component_state, obj)
+#define priv_to_pipe_st(o)  container_of(o, struct komeda_pipeline_state, obj)
+
+/* pipeline APIs */
+struct komeda_pipeline *
+komeda_pipeline_add(struct komeda_dev *mdev, size_t size,
+		    struct komeda_pipeline_funcs *funcs);
+void komeda_pipeline_destroy(struct komeda_dev *mdev,
+			     struct komeda_pipeline *pipe);
+
+struct komeda_component *
+komeda_pipeline_get_component(struct komeda_pipeline *pipe, int id);
+
+/* component APIs */
+struct komeda_component *
+komeda_component_add(struct komeda_pipeline *pipe,
+		     size_t comp_sz, u32 id, u32 hw_id,
+		     struct komeda_component_funcs *funcs,
+		     u8 max_active_inputs, u32 supported_inputs,
+		     u8 max_active_outputs, u32 __iomem *reg,
+		     const char *name_fmt, ...);
+
+void komeda_component_destroy(struct komeda_dev *mdev,
+			      struct komeda_component *c);
+
+#endif /* _KOMEDA_PIPELINE_H_*/
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_plane.c b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c
new file mode 100644
index 000000000000..0a4953a9a909
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c
@@ -0,0 +1,109 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_plane_helper.h>
+#include "komeda_dev.h"
+#include "komeda_kms.h"
+
+static const struct drm_plane_helper_funcs komeda_plane_helper_funcs = {
+};
+
+static void komeda_plane_destroy(struct drm_plane *plane)
+{
+	drm_plane_cleanup(plane);
+
+	kfree(to_kplane(plane));
+}
+
+static const struct drm_plane_funcs komeda_plane_funcs = {
+};
+
+/* for komeda, which is pipeline can be share between crtcs */
+static u32 get_possible_crtcs(struct komeda_kms_dev *kms,
+			      struct komeda_pipeline *pipe)
+{
+	struct komeda_crtc *crtc;
+	u32 possible_crtcs = 0;
+	int i;
+
+	for (i = 0; i < kms->n_crtcs; i++) {
+		crtc = &kms->crtcs[i];
+
+		if ((pipe == crtc->master) || (pipe == crtc->slave))
+			possible_crtcs |= BIT(i);
+	}
+
+	return possible_crtcs;
+}
+
+/* use Layer0 as primary */
+static u32 get_plane_type(struct komeda_kms_dev *kms,
+			  struct komeda_component *c)
+{
+	bool is_primary = (c->id == KOMEDA_COMPONENT_LAYER0);
+
+	return is_primary ? DRM_PLANE_TYPE_PRIMARY : DRM_PLANE_TYPE_OVERLAY;
+}
+
+static int komeda_plane_add(struct komeda_kms_dev *kms,
+			    struct komeda_layer *layer)
+{
+	struct komeda_dev *mdev = kms->base.dev_private;
+	struct komeda_component *c = &layer->base;
+	struct komeda_plane *kplane;
+	struct drm_plane *plane;
+	u32 *formats, n_formats = 0;
+	int err;
+
+	kplane = kzalloc(sizeof(*kplane), GFP_KERNEL);
+	if (!kplane)
+		return -ENOMEM;
+
+	plane = &kplane->base;
+	kplane->layer = layer;
+
+	formats = komeda_get_layer_fourcc_list(&mdev->fmt_tbl,
+					       layer->layer_type, &n_formats);
+
+	err = drm_universal_plane_init(&kms->base, plane,
+			get_possible_crtcs(kms, c->pipeline),
+			&komeda_plane_funcs,
+			formats, n_formats, NULL,
+			get_plane_type(kms, c),
+			"%s", c->name);
+
+	komeda_put_fourcc_list(formats);
+
+	if (err)
+		goto cleanup;
+
+	drm_plane_helper_add(plane, &komeda_plane_helper_funcs);
+
+	return 0;
+cleanup:
+	komeda_plane_destroy(plane);
+	return err;
+}
+
+int komeda_kms_add_planes(struct komeda_kms_dev *kms, struct komeda_dev *mdev)
+{
+	struct komeda_pipeline *pipe;
+	int i, j, err;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		pipe = mdev->pipelines[i];
+
+		for (j = 0; j < pipe->n_layers; j++) {
+			err = komeda_plane_add(kms, pipe->layers[j]);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c
new file mode 100644
index 000000000000..f1c9e3fefa86
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c
@@ -0,0 +1,88 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#include "komeda_dev.h"
+#include "komeda_kms.h"
+
+static struct drm_private_state *
+komeda_pipeline_atomic_duplicate_state(struct drm_private_obj *obj)
+{
+	struct komeda_pipeline_state *st;
+
+	st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return NULL;
+
+	st->active_comps = 0;
+
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &st->obj);
+
+	return &st->obj;
+}
+
+static void
+komeda_pipeline_atomic_destroy_state(struct drm_private_obj *obj,
+				     struct drm_private_state *state)
+{
+	kfree(priv_to_pipe_st(state));
+}
+
+static const struct drm_private_state_funcs komeda_pipeline_obj_funcs = {
+	.atomic_duplicate_state	= komeda_pipeline_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_pipeline_atomic_destroy_state,
+};
+
+static int komeda_pipeline_obj_add(struct komeda_kms_dev *kms,
+				   struct komeda_pipeline *pipe)
+{
+	struct komeda_pipeline_state *st;
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return -ENOMEM;
+
+	st->pipe = pipe;
+	drm_atomic_private_obj_init(&kms->base, &pipe->obj, &st->obj,
+				    &komeda_pipeline_obj_funcs);
+
+	return 0;
+}
+
+int komeda_kms_add_private_objs(struct komeda_kms_dev *kms,
+				struct komeda_dev *mdev)
+{
+	struct komeda_pipeline *pipe;
+	int i, err;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		pipe = mdev->pipelines[i];
+
+		err = komeda_pipeline_obj_add(kms, pipe);
+		if (err)
+			return err;
+
+		/* Add component */
+	}
+
+	return 0;
+}
+
+void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev)
+{
+	struct komeda_pipeline *pipe;
+	struct komeda_component *c;
+	int i, id;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		pipe = mdev->pipelines[i];
+		dp_for_each_set_bit(id, pipe->avail_comps) {
+			c = komeda_pipeline_get_component(pipe, id);
+
+			drm_atomic_private_obj_fini(&c->obj);
+		}
+		drm_atomic_private_obj_fini(&pipe->obj);
+	}
+}
diff --git a/drivers/gpu/drm/arm/hdlcd_crtc.c b/drivers/gpu/drm/arm/hdlcd_crtc.c
index e4d67b70244d..0b2b62f8fa3c 100644
--- a/drivers/gpu/drm/arm/hdlcd_crtc.c
+++ b/drivers/gpu/drm/arm/hdlcd_crtc.c
@@ -13,12 +13,12 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_fb_helper.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 <linux/clk.h>
 #include <linux/of_graph.h>
 #include <linux/platform_data/simplefb.h>
diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c
index dfad8d06d108..8fc0b884c428 100644
--- a/drivers/gpu/drm/arm/hdlcd_drv.c
+++ b/drivers/gpu/drm/arm/hdlcd_drv.c
@@ -22,13 +22,13 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include "hdlcd_drv.h"
 #include "hdlcd_regs.h"
@@ -229,7 +229,7 @@ static int hdlcd_debugfs_init(struct drm_minor *minor)
 DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver hdlcd_driver = {
-	.driver_features = DRIVER_HAVE_IRQ | DRIVER_GEM |
+	.driver_features = DRIVER_GEM |
 			   DRIVER_MODESET | DRIVER_PRIME |
 			   DRIVER_ATOMIC,
 	.irq_handler = hdlcd_irq,
diff --git a/drivers/gpu/drm/arm/malidp_crtc.c b/drivers/gpu/drm/arm/malidp_crtc.c
index e1b72782848c..56aad288666e 100644
--- a/drivers/gpu/drm/arm/malidp_crtc.c
+++ b/drivers/gpu/drm/arm/malidp_crtc.c
@@ -14,7 +14,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/pm_runtime.h>
 #include <video/videomode.h>
diff --git a/drivers/gpu/drm/arm/malidp_drv.c b/drivers/gpu/drm/arm/malidp_drv.c
index 505f316a192e..ab50ad06e271 100644
--- a/drivers/gpu/drm/arm/malidp_drv.c
+++ b/drivers/gpu/drm/arm/malidp_drv.c
@@ -23,7 +23,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
diff --git a/drivers/gpu/drm/arm/malidp_mw.c b/drivers/gpu/drm/arm/malidp_mw.c
index 91472e5e0c8b..041a64dc7167 100644
--- a/drivers/gpu/drm/arm/malidp_mw.c
+++ b/drivers/gpu/drm/arm/malidp_mw.c
@@ -8,7 +8,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drmP.h>
diff --git a/drivers/gpu/drm/armada/armada_510.c b/drivers/gpu/drm/armada/armada_510.c
index 2f7c048c5361..0e91d27921bd 100644
--- a/drivers/gpu/drm/armada/armada_510.c
+++ b/drivers/gpu/drm/armada/armada_510.c
@@ -9,7 +9,7 @@
  */
 #include <linux/clk.h>
 #include <linux/io.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "armada_crtc.h"
 #include "armada_drm.h"
 #include "armada_hw.h"
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 20dfb29561c2..ba4a3fab7745 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -12,7 +12,7 @@
 #include <linux/platform_device.h>
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_plane_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include "armada_crtc.h"
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index 7ebd337b60af..08761ff01739 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -8,6 +8,8 @@
 #ifndef ARMADA_CRTC_H
 #define ARMADA_CRTC_H
 
+#include <drm/drm_crtc.h>
+
 struct armada_gem_object;
 
 struct armada_regs {
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index fa31589b4fc0..e660c5ca52ae 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -10,7 +10,7 @@
 #include <linux/module.h>
 #include <linux/of_graph.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_of.h>
 #include "armada_crtc.h"
diff --git a/drivers/gpu/drm/armada/armada_fb.c b/drivers/gpu/drm/armada/armada_fb.c
index 6bd638a54579..058ac7d9920f 100644
--- a/drivers/gpu/drm/armada/armada_fb.c
+++ b/drivers/gpu/drm/armada/armada_fb.c
@@ -5,7 +5,7 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include "armada_drm.h"
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index bf589c53b908..3871b39d4dea 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -30,6 +30,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "ast_drv.h"
 
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 373700c05a00..2854399856ba 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -639,13 +639,9 @@ int ast_dumb_create(struct drm_file *file,
 
 static void ast_bo_unref(struct ast_bo **bo)
 {
-	struct ttm_buffer_object *tbo;
-
 	if ((*bo) == NULL)
 		return;
-
-	tbo = &((*bo)->bo);
-	ttm_bo_unref(&tbo);
+	ttm_bo_put(&((*bo)->bo));
 	*bo = NULL;
 }
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index 8bb355d5d43d..97fed0627d1c 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -32,6 +32,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "ast_drv.h"
 
 #include "ast_tables.h"
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
index 96f4082671fe..8070a558d7b1 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
@@ -24,7 +24,7 @@
 #include <linux/pinctrl/consumer.h>
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include <video/videomode.h>
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
index 034a91112098..0be13eceedba 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
@@ -720,7 +720,7 @@ static void atmel_hlcdc_dc_irq_uninstall(struct drm_device *dev)
 DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver atmel_hlcdc_dc_driver = {
-	.driver_features = DRIVER_HAVE_IRQ | DRIVER_GEM |
+	.driver_features = DRIVER_GEM |
 			   DRIVER_MODESET | DRIVER_PRIME |
 			   DRIVER_ATOMIC,
 	.irq_handler = atmel_hlcdc_dc_irq_handler,
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.h b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.h
index 4cc1e03f0aee..70bd540d644e 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.h
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.h
@@ -31,7 +31,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
index 9330a076e15a..e836e2de35ce 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
@@ -549,7 +549,8 @@ atmel_hlcdc_plane_prepare_disc_area(struct drm_crtc_state *c_state)
 
 		ovl_state = drm_plane_state_to_atmel_hlcdc_plane_state(ovl_s);
 
-		if (!ovl_s->fb ||
+		if (!ovl_s->visible ||
+		    !ovl_s->fb ||
 		    ovl_s->fb->format->has_alpha ||
 		    ovl_s->alpha != DRM_BLEND_ALPHA_OPAQUE)
 			continue;
@@ -601,15 +602,10 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 	struct drm_framebuffer *fb = state->base.fb;
 	const struct drm_display_mode *mode;
 	struct drm_crtc_state *crtc_state;
-	unsigned int patched_crtc_w;
-	unsigned int patched_crtc_h;
-	unsigned int patched_src_w;
-	unsigned int patched_src_h;
 	unsigned int tmp;
-	int x_offset = 0;
-	int y_offset = 0;
 	int hsub = 1;
 	int vsub = 1;
+	int ret;
 	int i;
 
 	if (!state->base.crtc || !fb)
@@ -618,14 +614,21 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 	crtc_state = drm_atomic_get_existing_crtc_state(s->state, s->crtc);
 	mode = &crtc_state->adjusted_mode;
 
-	state->src_x = s->src_x;
-	state->src_y = s->src_y;
-	state->src_h = s->src_h;
-	state->src_w = s->src_w;
-	state->crtc_x = s->crtc_x;
-	state->crtc_y = s->crtc_y;
-	state->crtc_h = s->crtc_h;
-	state->crtc_w = s->crtc_w;
+	ret = drm_atomic_helper_check_plane_state(s, crtc_state,
+						  (1 << 16) / 2048,
+						  INT_MAX, true, true);
+	if (ret || !s->visible)
+		return ret;
+
+	state->src_x = s->src.x1;
+	state->src_y = s->src.y1;
+	state->src_w = drm_rect_width(&s->src);
+	state->src_h = drm_rect_height(&s->src);
+	state->crtc_x = s->dst.x1;
+	state->crtc_y = s->dst.y1;
+	state->crtc_w = drm_rect_width(&s->dst);
+	state->crtc_h = drm_rect_height(&s->dst);
+
 	if ((state->src_x | state->src_y | state->src_w | state->src_h) &
 	    SUBPIXEL_MASK)
 		return -EINVAL;
@@ -639,45 +642,6 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 	if (state->nplanes > ATMEL_HLCDC_LAYER_MAX_PLANES)
 		return -EINVAL;
 
-	/*
-	 * Swap width and size in case of 90 or 270 degrees rotation
-	 */
-	if (drm_rotation_90_or_270(state->base.rotation)) {
-		tmp = state->crtc_w;
-		state->crtc_w = state->crtc_h;
-		state->crtc_h = tmp;
-		tmp = state->src_w;
-		state->src_w = state->src_h;
-		state->src_h = tmp;
-	}
-
-	if (state->crtc_x + state->crtc_w > mode->hdisplay)
-		patched_crtc_w = mode->hdisplay - state->crtc_x;
-	else
-		patched_crtc_w = state->crtc_w;
-
-	if (state->crtc_x < 0) {
-		patched_crtc_w += state->crtc_x;
-		x_offset = -state->crtc_x;
-		state->crtc_x = 0;
-	}
-
-	if (state->crtc_y + state->crtc_h > mode->vdisplay)
-		patched_crtc_h = mode->vdisplay - state->crtc_y;
-	else
-		patched_crtc_h = state->crtc_h;
-
-	if (state->crtc_y < 0) {
-		patched_crtc_h += state->crtc_y;
-		y_offset = -state->crtc_y;
-		state->crtc_y = 0;
-	}
-
-	patched_src_w = DIV_ROUND_CLOSEST(patched_crtc_w * state->src_w,
-					  state->crtc_w);
-	patched_src_h = DIV_ROUND_CLOSEST(patched_crtc_h * state->src_h,
-					  state->crtc_h);
-
 	hsub = drm_format_horz_chroma_subsampling(fb->format->format);
 	vsub = drm_format_vert_chroma_subsampling(fb->format->format);
 
@@ -692,41 +656,38 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 
 		switch (state->base.rotation & DRM_MODE_ROTATE_MASK) {
 		case DRM_MODE_ROTATE_90:
-			offset = ((y_offset + state->src_y + patched_src_w - 1) /
-				  ydiv) * fb->pitches[i];
-			offset += ((x_offset + state->src_x) / xdiv) *
-				  state->bpp[i];
-			state->xstride[i] = ((patched_src_w - 1) / ydiv) *
-					  fb->pitches[i];
-			state->pstride[i] = -fb->pitches[i] - state->bpp[i];
+			offset = (state->src_y / ydiv) *
+				 fb->pitches[i];
+			offset += ((state->src_x + state->src_w - 1) /
+				   xdiv) * state->bpp[i];
+			state->xstride[i] = -(((state->src_h - 1) / ydiv) *
+					    fb->pitches[i]) -
+					  (2 * state->bpp[i]);
+			state->pstride[i] = fb->pitches[i] - state->bpp[i];
 			break;
 		case DRM_MODE_ROTATE_180:
-			offset = ((y_offset + state->src_y + patched_src_h - 1) /
+			offset = ((state->src_y + state->src_h - 1) /
 				  ydiv) * fb->pitches[i];
-			offset += ((x_offset + state->src_x + patched_src_w - 1) /
+			offset += ((state->src_x + state->src_w - 1) /
 				   xdiv) * state->bpp[i];
-			state->xstride[i] = ((((patched_src_w - 1) / xdiv) - 1) *
+			state->xstride[i] = ((((state->src_w - 1) / xdiv) - 1) *
 					   state->bpp[i]) - fb->pitches[i];
 			state->pstride[i] = -2 * state->bpp[i];
 			break;
 		case DRM_MODE_ROTATE_270:
-			offset = ((y_offset + state->src_y) / ydiv) *
-				 fb->pitches[i];
-			offset += ((x_offset + state->src_x + patched_src_h - 1) /
-				   xdiv) * state->bpp[i];
-			state->xstride[i] = -(((patched_src_w - 1) / ydiv) *
-					    fb->pitches[i]) -
-					  (2 * state->bpp[i]);
-			state->pstride[i] = fb->pitches[i] - state->bpp[i];
+			offset = ((state->src_y + state->src_h - 1) /
+				  ydiv) * fb->pitches[i];
+			offset += (state->src_x / xdiv) * state->bpp[i];
+			state->xstride[i] = ((state->src_h - 1) / ydiv) *
+					  fb->pitches[i];
+			state->pstride[i] = -fb->pitches[i] - state->bpp[i];
 			break;
 		case DRM_MODE_ROTATE_0:
 		default:
-			offset = ((y_offset + state->src_y) / ydiv) *
-				 fb->pitches[i];
-			offset += ((x_offset + state->src_x) / xdiv) *
-				  state->bpp[i];
+			offset = (state->src_y / ydiv) * fb->pitches[i];
+			offset += (state->src_x / xdiv) * state->bpp[i];
 			state->xstride[i] = fb->pitches[i] -
-					  ((patched_src_w / xdiv) *
+					  ((state->src_w / xdiv) *
 					   state->bpp[i]);
 			state->pstride[i] = 0;
 			break;
@@ -735,35 +696,45 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 		state->offsets[i] = offset + fb->offsets[i];
 	}
 
-	state->src_w = patched_src_w;
-	state->src_h = patched_src_h;
-	state->crtc_w = patched_crtc_w;
-	state->crtc_h = patched_crtc_h;
+	/*
+	 * Swap width and size in case of 90 or 270 degrees rotation
+	 */
+	if (drm_rotation_90_or_270(state->base.rotation)) {
+		tmp = state->src_w;
+		state->src_w = state->src_h;
+		state->src_h = tmp;
+	}
 
 	if (!desc->layout.size &&
 	    (mode->hdisplay != state->crtc_w ||
 	     mode->vdisplay != state->crtc_h))
 		return -EINVAL;
 
-	if (desc->max_height && state->crtc_h > desc->max_height)
-		return -EINVAL;
-
-	if (desc->max_width && state->crtc_w > desc->max_width)
-		return -EINVAL;
-
 	if ((state->crtc_h != state->src_h || state->crtc_w != state->src_w) &&
 	    (!desc->layout.memsize ||
 	     state->base.fb->format->has_alpha))
 		return -EINVAL;
 
-	if (state->crtc_x < 0 || state->crtc_y < 0)
-		return -EINVAL;
+	return 0;
+}
 
-	if (state->crtc_w + state->crtc_x > mode->hdisplay ||
-	    state->crtc_h + state->crtc_y > mode->vdisplay)
-		return -EINVAL;
+static void atmel_hlcdc_plane_atomic_disable(struct drm_plane *p,
+					     struct drm_plane_state *old_state)
+{
+	struct atmel_hlcdc_plane *plane = drm_plane_to_atmel_hlcdc_plane(p);
 
-	return 0;
+	/* Disable interrupts */
+	atmel_hlcdc_layer_write_reg(&plane->layer, ATMEL_HLCDC_LAYER_IDR,
+				    0xffffffff);
+
+	/* Disable the layer */
+	atmel_hlcdc_layer_write_reg(&plane->layer, ATMEL_HLCDC_LAYER_CHDR,
+				    ATMEL_HLCDC_LAYER_RST |
+				    ATMEL_HLCDC_LAYER_A2Q |
+				    ATMEL_HLCDC_LAYER_UPDATE);
+
+	/* Clear all pending interrupts */
+	atmel_hlcdc_layer_read_reg(&plane->layer, ATMEL_HLCDC_LAYER_ISR);
 }
 
 static void atmel_hlcdc_plane_atomic_update(struct drm_plane *p,
@@ -777,6 +748,11 @@ static void atmel_hlcdc_plane_atomic_update(struct drm_plane *p,
 	if (!p->state->crtc || !p->state->fb)
 		return;
 
+	if (!state->base.visible) {
+		atmel_hlcdc_plane_atomic_disable(p, old_s);
+		return;
+	}
+
 	atmel_hlcdc_plane_update_pos_and_size(plane, state);
 	atmel_hlcdc_plane_update_general_settings(plane, state);
 	atmel_hlcdc_plane_update_format(plane, state);
@@ -798,25 +774,6 @@ static void atmel_hlcdc_plane_atomic_update(struct drm_plane *p,
 			 ATMEL_HLCDC_LAYER_A2Q : ATMEL_HLCDC_LAYER_EN));
 }
 
-static void atmel_hlcdc_plane_atomic_disable(struct drm_plane *p,
-					     struct drm_plane_state *old_state)
-{
-	struct atmel_hlcdc_plane *plane = drm_plane_to_atmel_hlcdc_plane(p);
-
-	/* Disable interrupts */
-	atmel_hlcdc_layer_write_reg(&plane->layer, ATMEL_HLCDC_LAYER_IDR,
-				    0xffffffff);
-
-	/* Disable the layer */
-	atmel_hlcdc_layer_write_reg(&plane->layer, ATMEL_HLCDC_LAYER_CHDR,
-				    ATMEL_HLCDC_LAYER_RST |
-				    ATMEL_HLCDC_LAYER_A2Q |
-				    ATMEL_HLCDC_LAYER_UPDATE);
-
-	/* Clear all pending interrupts */
-	atmel_hlcdc_layer_read_reg(&plane->layer, ATMEL_HLCDC_LAYER_ISR);
-}
-
 static int atmel_hlcdc_plane_init_properties(struct atmel_hlcdc_plane *plane)
 {
 	const struct atmel_hlcdc_layer_desc *desc = plane->layer.desc;
diff --git a/drivers/gpu/drm/bochs/bochs_drv.c b/drivers/gpu/drm/bochs/bochs_drv.c
index cea42ac64d7e..cb55bdc36f3f 100644
--- a/drivers/gpu/drm/bochs/bochs_drv.c
+++ b/drivers/gpu/drm/bochs/bochs_drv.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "bochs.h"
 
diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c
index e9d5dbc34676..9cd82e3631fb 100644
--- a/drivers/gpu/drm/bochs/bochs_kms.c
+++ b/drivers/gpu/drm/bochs/bochs_kms.c
@@ -10,6 +10,7 @@
 #include <drm/drm_plane_helper.h>
 #include <drm/drm_atomic_uapi.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 static int defx = 1024;
 static int defy = 768;
diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c
index 641a33f134ee..49463348a07a 100644
--- a/drivers/gpu/drm/bochs/bochs_mm.c
+++ b/drivers/gpu/drm/bochs/bochs_mm.c
@@ -442,5 +442,6 @@ int bochs_gem_prime_mmap(struct drm_gem_object *obj,
 {
 	struct bochs_bo *bo = gem_to_bochs_bo(obj);
 
-	return ttm_fbdev_mmap(vma, &bo->bo);
+	bo->gem.vma_node.vm_node.start = bo->bo.vma_node.vm_node.start;
+	return drm_gem_prime_mmap(obj, vma);
 }
diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 2fee47b0d50b..8840f396a7b6 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -30,6 +30,7 @@ config DRM_CDNS_DSI
 	select DRM_KMS_HELPER
 	select DRM_MIPI_DSI
 	select DRM_PANEL_BRIDGE
+	select GENERIC_PHY_MIPI_DPHY
 	depends on OF
 	help
 	  Support Cadence DPI to DSI bridge. This is an internal
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511.h b/drivers/gpu/drm/bridge/adv7511/adv7511.h
index 2b6e0832d1cf..996a7e7dbfd6 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511.h
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511.h
@@ -14,8 +14,10 @@
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
 
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_connector.h>
 #include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
 
 #define ADV7511_REG_CHIP_REVISION		0x00
 #define ADV7511_REG_N0				0x01
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index d0e98caa2e2a..ec2ca71e1323 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -17,6 +17,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include <media/cec.h>
 
diff --git a/drivers/gpu/drm/bridge/analogix-anx78xx.c b/drivers/gpu/drm/bridge/analogix-anx78xx.c
index 4cf7bc17ae14..c09aaf93ae1b 100644
--- a/drivers/gpu/drm/bridge/analogix-anx78xx.c
+++ b/drivers/gpu/drm/bridge/analogix-anx78xx.c
@@ -31,9 +31,9 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include "analogix-anx78xx.h"
 
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
index 4d5b47585834..225f5e5dd69b 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
@@ -26,8 +26,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include <drm/bridge/analogix_dp.h>
 
diff --git a/drivers/gpu/drm/bridge/cdns-dsi.c b/drivers/gpu/drm/bridge/cdns-dsi.c
index ce9496d13986..6166dca6be81 100644
--- a/drivers/gpu/drm/bridge/cdns-dsi.c
+++ b/drivers/gpu/drm/bridge/cdns-dsi.c
@@ -7,12 +7,14 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <video/mipi_display.h>
 
 #include <linux/clk.h>
+#include <linux/interrupt.h>
 #include <linux/iopoll.h>
 #include <linux/module.h>
 #include <linux/of_address.h>
@@ -21,6 +23,9 @@
 #include <linux/pm_runtime.h>
 #include <linux/reset.h>
 
+#include <linux/phy/phy.h>
+#include <linux/phy/phy-mipi-dphy.h>
+
 #define IP_CONF				0x0
 #define SP_HS_FIFO_DEPTH(x)		(((x) & GENMASK(30, 26)) >> 26)
 #define SP_LP_FIFO_DEPTH(x)		(((x) & GENMASK(25, 21)) >> 21)
@@ -419,44 +424,11 @@
 #define DSI_NULL_FRAME_OVERHEAD		6
 #define DSI_EOT_PKT_SIZE		4
 
-#define REG_WAKEUP_TIME_NS		800
-#define DPHY_PLL_RATE_HZ		108000000
-
-/* DPHY registers */
-#define DPHY_PMA_CMN(reg)		(reg)
-#define DPHY_PMA_LCLK(reg)		(0x100 + (reg))
-#define DPHY_PMA_LDATA(lane, reg)	(0x200 + ((lane) * 0x100) + (reg))
-#define DPHY_PMA_RCLK(reg)		(0x600 + (reg))
-#define DPHY_PMA_RDATA(lane, reg)	(0x700 + ((lane) * 0x100) + (reg))
-#define DPHY_PCS(reg)			(0xb00 + (reg))
-
-#define DPHY_CMN_SSM			DPHY_PMA_CMN(0x20)
-#define DPHY_CMN_SSM_EN			BIT(0)
-#define DPHY_CMN_TX_MODE_EN		BIT(9)
-
-#define DPHY_CMN_PWM			DPHY_PMA_CMN(0x40)
-#define DPHY_CMN_PWM_DIV(x)		((x) << 20)
-#define DPHY_CMN_PWM_LOW(x)		((x) << 10)
-#define DPHY_CMN_PWM_HIGH(x)		(x)
-
-#define DPHY_CMN_FBDIV			DPHY_PMA_CMN(0x4c)
-#define DPHY_CMN_FBDIV_VAL(low, high)	(((high) << 11) | ((low) << 22))
-#define DPHY_CMN_FBDIV_FROM_REG		(BIT(10) | BIT(21))
-
-#define DPHY_CMN_OPIPDIV		DPHY_PMA_CMN(0x50)
-#define DPHY_CMN_IPDIV_FROM_REG		BIT(0)
-#define DPHY_CMN_IPDIV(x)		((x) << 1)
-#define DPHY_CMN_OPDIV_FROM_REG		BIT(6)
-#define DPHY_CMN_OPDIV(x)		((x) << 7)
-
-#define DPHY_PSM_CFG			DPHY_PCS(0x4)
-#define DPHY_PSM_CFG_FROM_REG		BIT(0)
-#define DPHY_PSM_CLK_DIV(x)		((x) << 1)
-
 struct cdns_dsi_output {
 	struct mipi_dsi_device *dev;
 	struct drm_panel *panel;
 	struct drm_bridge *bridge;
+	union phy_configure_opts phy_opts;
 };
 
 enum cdns_dsi_input_id {
@@ -465,14 +437,6 @@ enum cdns_dsi_input_id {
 	CDNS_DSC_INPUT,
 };
 
-struct cdns_dphy_cfg {
-	u8 pll_ipdiv;
-	u8 pll_opdiv;
-	u16 pll_fbdiv;
-	unsigned long lane_bps;
-	unsigned int nlanes;
-};
-
 struct cdns_dsi_cfg {
 	unsigned int hfp;
 	unsigned int hsa;
@@ -481,34 +445,6 @@ struct cdns_dsi_cfg {
 	unsigned int htotal;
 };
 
-struct cdns_dphy;
-
-enum cdns_dphy_clk_lane_cfg {
-	DPHY_CLK_CFG_LEFT_DRIVES_ALL = 0,
-	DPHY_CLK_CFG_LEFT_DRIVES_RIGHT = 1,
-	DPHY_CLK_CFG_LEFT_DRIVES_LEFT = 2,
-	DPHY_CLK_CFG_RIGHT_DRIVES_ALL = 3,
-};
-
-struct cdns_dphy_ops {
-	int (*probe)(struct cdns_dphy *dphy);
-	void (*remove)(struct cdns_dphy *dphy);
-	void (*set_psm_div)(struct cdns_dphy *dphy, u8 div);
-	void (*set_clk_lane_cfg)(struct cdns_dphy *dphy,
-				 enum cdns_dphy_clk_lane_cfg cfg);
-	void (*set_pll_cfg)(struct cdns_dphy *dphy,
-			    const struct cdns_dphy_cfg *cfg);
-	unsigned long (*get_wakeup_time_ns)(struct cdns_dphy *dphy);
-};
-
-struct cdns_dphy {
-	struct cdns_dphy_cfg cfg;
-	void __iomem *regs;
-	struct clk *psm_clk;
-	struct clk *pll_ref_clk;
-	const struct cdns_dphy_ops *ops;
-};
-
 struct cdns_dsi_input {
 	enum cdns_dsi_input_id id;
 	struct drm_bridge bridge;
@@ -526,7 +462,7 @@ struct cdns_dsi {
 	struct reset_control *dsi_p_rst;
 	struct clk *dsi_sys_clk;
 	bool link_initialized;
-	struct cdns_dphy *dphy;
+	struct phy *dphy;
 };
 
 static inline struct cdns_dsi *input_to_dsi(struct cdns_dsi_input *input)
@@ -545,173 +481,13 @@ bridge_to_cdns_dsi_input(struct drm_bridge *bridge)
 	return container_of(bridge, struct cdns_dsi_input, bridge);
 }
 
-static int cdns_dsi_get_dphy_pll_cfg(struct cdns_dphy *dphy,
-				     struct cdns_dphy_cfg *cfg,
-				     unsigned int dpi_htotal,
-				     unsigned int dpi_bpp,
-				     unsigned int dpi_hz,
-				     unsigned int dsi_htotal,
-				     unsigned int dsi_nlanes,
-				     unsigned int *dsi_hfp_ext)
-{
-	u64 dlane_bps, dlane_bps_max, fbdiv, fbdiv_max, adj_dsi_htotal;
-	unsigned long pll_ref_hz = clk_get_rate(dphy->pll_ref_clk);
-
-	memset(cfg, 0, sizeof(*cfg));
-
-	cfg->nlanes = dsi_nlanes;
-
-	if (pll_ref_hz < 9600000 || pll_ref_hz >= 150000000)
-		return -EINVAL;
-	else if (pll_ref_hz < 19200000)
-		cfg->pll_ipdiv = 1;
-	else if (pll_ref_hz < 38400000)
-		cfg->pll_ipdiv = 2;
-	else if (pll_ref_hz < 76800000)
-		cfg->pll_ipdiv = 4;
-	else
-		cfg->pll_ipdiv = 8;
-
-	/*
-	 * Make sure DSI htotal is aligned on a lane boundary when calculating
-	 * the expected data rate. This is done by extending HFP in case of
-	 * misalignment.
-	 */
-	adj_dsi_htotal = dsi_htotal;
-	if (dsi_htotal % dsi_nlanes)
-		adj_dsi_htotal += dsi_nlanes - (dsi_htotal % dsi_nlanes);
-
-	dlane_bps = (u64)dpi_hz * adj_dsi_htotal;
-
-	/* data rate in bytes/sec is not an integer, refuse the mode. */
-	if (do_div(dlane_bps, dsi_nlanes * dpi_htotal))
-		return -EINVAL;
-
-	/* data rate was in bytes/sec, convert to bits/sec. */
-	dlane_bps *= 8;
-
-	if (dlane_bps > 2500000000UL || dlane_bps < 160000000UL)
-		return -EINVAL;
-	else if (dlane_bps >= 1250000000)
-		cfg->pll_opdiv = 1;
-	else if (dlane_bps >= 630000000)
-		cfg->pll_opdiv = 2;
-	else if (dlane_bps >= 320000000)
-		cfg->pll_opdiv = 4;
-	else if (dlane_bps >= 160000000)
-		cfg->pll_opdiv = 8;
-
-	/*
-	 * Allow a deviation of 0.2% on the per-lane data rate to try to
-	 * recover a potential mismatch between DPI and PPI clks.
-	 */
-	dlane_bps_max = dlane_bps + DIV_ROUND_DOWN_ULL(dlane_bps, 500);
-	fbdiv_max = DIV_ROUND_DOWN_ULL(dlane_bps_max * 2 *
-				       cfg->pll_opdiv * cfg->pll_ipdiv,
-				       pll_ref_hz);
-	fbdiv = DIV_ROUND_UP_ULL(dlane_bps * 2 * cfg->pll_opdiv *
-				 cfg->pll_ipdiv,
-				 pll_ref_hz);
-
-	/*
-	 * Iterate over all acceptable fbdiv and try to find an adjusted DSI
-	 * htotal length providing an exact match.
-	 *
-	 * Note that we could do something even trickier by relying on the fact
-	 * that a new line is not necessarily aligned on a lane boundary, so,
-	 * by making adj_dsi_htotal non aligned on a dsi_lanes we can improve a
-	 * bit the precision. With this, the step would be
-	 *
-	 *	pll_ref_hz / (2 * opdiv * ipdiv * nlanes)
-	 *
-	 * instead of
-	 *
-	 *	pll_ref_hz / (2 * opdiv * ipdiv)
-	 *
-	 * The drawback of this approach is that we would need to make sure the
-	 * number or lines is a multiple of the realignment periodicity which is
-	 * a function of the number of lanes and the original misalignment. For
-	 * example, for NLANES = 4 and HTOTAL % NLANES = 3, it takes 4 lines
-	 * to realign on a lane:
-	 * LINE 0: expected number of bytes, starts emitting first byte of
-	 *	   LINE 1 on LANE 3
-	 * LINE 1: expected number of bytes, starts emitting first 2 bytes of
-	 *	   LINE 2 on LANES 2 and 3
-	 * LINE 2: expected number of bytes, starts emitting first 3 bytes of
-	 *	   of LINE 3 on LANES 1, 2 and 3
-	 * LINE 3: one byte less, now things are realigned on LANE 0 for LINE 4
-	 *
-	 * I figured this extra complexity was not worth the benefit, but if
-	 * someone really has unfixable mismatch, that would be something to
-	 * investigate.
-	 */
-	for (; fbdiv <= fbdiv_max; fbdiv++) {
-		u32 rem;
-
-		adj_dsi_htotal = (u64)fbdiv * pll_ref_hz * dsi_nlanes *
-				 dpi_htotal;
-
-		/*
-		 * Do the division in 2 steps to avoid an overflow on the
-		 * divider.
-		 */
-		rem = do_div(adj_dsi_htotal, dpi_hz);
-		if (rem)
-			continue;
-
-		rem = do_div(adj_dsi_htotal,
-			     cfg->pll_opdiv * cfg->pll_ipdiv * 2 * 8);
-		if (rem)
-			continue;
-
-		cfg->pll_fbdiv = fbdiv;
-		*dsi_hfp_ext = adj_dsi_htotal - dsi_htotal;
-		break;
-	}
-
-	/* No match, let's just reject the display mode. */
-	if (!cfg->pll_fbdiv)
-		return -EINVAL;
-
-	dlane_bps = DIV_ROUND_DOWN_ULL((u64)dpi_hz * adj_dsi_htotal * 8,
-				       dsi_nlanes * dpi_htotal);
-	cfg->lane_bps = dlane_bps;
-
-	return 0;
-}
-
-static int cdns_dphy_setup_psm(struct cdns_dphy *dphy)
-{
-	unsigned long psm_clk_hz = clk_get_rate(dphy->psm_clk);
-	unsigned long psm_div;
-
-	if (!psm_clk_hz || psm_clk_hz > 100000000)
-		return -EINVAL;
-
-	psm_div = DIV_ROUND_CLOSEST(psm_clk_hz, 1000000);
-	if (dphy->ops->set_psm_div)
-		dphy->ops->set_psm_div(dphy, psm_div);
-
-	return 0;
-}
-
-static void cdns_dphy_set_clk_lane_cfg(struct cdns_dphy *dphy,
-				       enum cdns_dphy_clk_lane_cfg cfg)
-{
-	if (dphy->ops->set_clk_lane_cfg)
-		dphy->ops->set_clk_lane_cfg(dphy, cfg);
-}
-
-static void cdns_dphy_set_pll_cfg(struct cdns_dphy *dphy,
-				  const struct cdns_dphy_cfg *cfg)
+static unsigned int mode_to_dpi_hfp(const struct drm_display_mode *mode,
+				    bool mode_valid_check)
 {
-	if (dphy->ops->set_pll_cfg)
-		dphy->ops->set_pll_cfg(dphy, cfg);
-}
+	if (mode_valid_check)
+		return mode->hsync_start - mode->hdisplay;
 
-static unsigned long cdns_dphy_get_wakeup_time_ns(struct cdns_dphy *dphy)
-{
-	return dphy->ops->get_wakeup_time_ns(dphy);
+	return mode->crtc_hsync_start - mode->crtc_hdisplay;
 }
 
 static unsigned int dpi_to_dsi_timing(unsigned int dpi_timing,
@@ -731,14 +507,12 @@ static unsigned int dpi_to_dsi_timing(unsigned int dpi_timing,
 static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 			     const struct drm_display_mode *mode,
 			     struct cdns_dsi_cfg *dsi_cfg,
-			     struct cdns_dphy_cfg *dphy_cfg,
 			     bool mode_valid_check)
 {
-	unsigned long dsi_htotal = 0, dsi_hss_hsa_hse_hbp = 0;
 	struct cdns_dsi_output *output = &dsi->output;
-	unsigned int dsi_hfp_ext = 0, dpi_hfp, tmp;
+	unsigned int tmp;
 	bool sync_pulse = false;
-	int bpp, nlanes, ret;
+	int bpp, nlanes;
 
 	memset(dsi_cfg, 0, sizeof(*dsi_cfg));
 
@@ -757,8 +531,6 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 		       mode->crtc_hsync_end : mode->crtc_hsync_start);
 
 	dsi_cfg->hbp = dpi_to_dsi_timing(tmp, bpp, DSI_HBP_FRAME_OVERHEAD);
-	dsi_htotal += dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
-	dsi_hss_hsa_hse_hbp += dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
 
 	if (sync_pulse) {
 		if (mode_valid_check)
@@ -768,49 +540,104 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 
 		dsi_cfg->hsa = dpi_to_dsi_timing(tmp, bpp,
 						 DSI_HSA_FRAME_OVERHEAD);
-		dsi_htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
-		dsi_hss_hsa_hse_hbp += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
 	}
 
 	dsi_cfg->hact = dpi_to_dsi_timing(mode_valid_check ?
 					  mode->hdisplay : mode->crtc_hdisplay,
 					  bpp, 0);
-	dsi_htotal += dsi_cfg->hact;
+	dsi_cfg->hfp = dpi_to_dsi_timing(mode_to_dpi_hfp(mode, mode_valid_check),
+					 bpp, DSI_HFP_FRAME_OVERHEAD);
 
-	if (mode_valid_check)
-		dpi_hfp = mode->hsync_start - mode->hdisplay;
-	else
-		dpi_hfp = mode->crtc_hsync_start - mode->crtc_hdisplay;
+	return 0;
+}
+
+static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
+			      struct cdns_dsi_cfg *dsi_cfg,
+			      struct phy_configure_opts_mipi_dphy *phy_cfg,
+			      const struct drm_display_mode *mode,
+			      bool mode_valid_check)
+{
+	struct cdns_dsi_output *output = &dsi->output;
+	unsigned long long dlane_bps;
+	unsigned long adj_dsi_htotal;
+	unsigned long dsi_htotal;
+	unsigned long dpi_htotal;
+	unsigned long dpi_hz;
+	unsigned int dsi_hfp_ext;
+	unsigned int lanes = output->dev->lanes;
+
+	dsi_htotal = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
+	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
+		dsi_htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
 
-	dsi_cfg->hfp = dpi_to_dsi_timing(dpi_hfp, bpp, DSI_HFP_FRAME_OVERHEAD);
+	dsi_htotal += dsi_cfg->hact;
 	dsi_htotal += dsi_cfg->hfp + DSI_HFP_FRAME_OVERHEAD;
 
-	if (mode_valid_check)
-		ret = cdns_dsi_get_dphy_pll_cfg(dsi->dphy, dphy_cfg,
-						mode->htotal, bpp,
-						mode->clock * 1000,
-						dsi_htotal, nlanes,
-						&dsi_hfp_ext);
-	else
-		ret = cdns_dsi_get_dphy_pll_cfg(dsi->dphy, dphy_cfg,
-						mode->crtc_htotal, bpp,
-						mode->crtc_clock * 1000,
-						dsi_htotal, nlanes,
-						&dsi_hfp_ext);
+	/*
+	 * Make sure DSI htotal is aligned on a lane boundary when calculating
+	 * the expected data rate. This is done by extending HFP in case of
+	 * misalignment.
+	 */
+	adj_dsi_htotal = dsi_htotal;
+	if (dsi_htotal % lanes)
+		adj_dsi_htotal += lanes - (dsi_htotal % lanes);
+
+	dpi_hz = (mode_valid_check ? mode->clock : mode->crtc_clock) * 1000;
+	dlane_bps = (unsigned long long)dpi_hz * adj_dsi_htotal;
+
+	/* data rate in bytes/sec is not an integer, refuse the mode. */
+	dpi_htotal = mode_valid_check ? mode->htotal : mode->crtc_htotal;
+	if (do_div(dlane_bps, lanes * dpi_htotal))
+		return -EINVAL;
+
+	/* data rate was in bytes/sec, convert to bits/sec. */
+	phy_cfg->hs_clk_rate = dlane_bps * 8;
 
+	dsi_hfp_ext = adj_dsi_htotal - dsi_htotal;
+	dsi_cfg->hfp += dsi_hfp_ext;
+	dsi_cfg->htotal = dsi_htotal + dsi_hfp_ext;
+
+	return 0;
+}
+
+static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
+			       const struct drm_display_mode *mode,
+			       struct cdns_dsi_cfg *dsi_cfg,
+			       bool mode_valid_check)
+{
+	struct cdns_dsi_output *output = &dsi->output;
+	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
+	unsigned long dsi_hss_hsa_hse_hbp;
+	unsigned int nlanes = output->dev->lanes;
+	int ret;
+
+	ret = cdns_dsi_mode2cfg(dsi, mode, dsi_cfg, mode_valid_check);
 	if (ret)
 		return ret;
 
-	dsi_cfg->hfp += dsi_hfp_ext;
-	dsi_htotal += dsi_hfp_ext;
-	dsi_cfg->htotal = dsi_htotal;
+	phy_mipi_dphy_get_default_config(mode->crtc_clock * 1000,
+					 mipi_dsi_pixel_format_to_bpp(output->dev->format),
+					 nlanes, phy_cfg);
+
+	ret = cdns_dsi_adjust_phy_config(dsi, dsi_cfg, phy_cfg, mode, mode_valid_check);
+	if (ret)
+		return ret;
+
+	ret = phy_validate(dsi->dphy, PHY_MODE_MIPI_DPHY, 0, &output->phy_opts);
+	if (ret)
+		return ret;
+
+	dsi_hss_hsa_hse_hbp = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
+	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
+		dsi_hss_hsa_hse_hbp += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
 
 	/*
 	 * Make sure DPI(HFP) > DSI(HSS+HSA+HSE+HBP) to guarantee that the FIFO
 	 * is empty before we start a receiving a new line on the DPI
 	 * interface.
 	 */
-	if ((u64)dphy_cfg->lane_bps * dpi_hfp * nlanes <
+	if ((u64)phy_cfg->hs_clk_rate *
+	    mode_to_dpi_hfp(mode, mode_valid_check) * nlanes <
 	    (u64)dsi_hss_hsa_hse_hbp *
 	    (mode_valid_check ? mode->clock : mode->crtc_clock) * 1000)
 		return -EINVAL;
@@ -840,9 +667,8 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_output *output = &dsi->output;
-	struct cdns_dphy_cfg dphy_cfg;
 	struct cdns_dsi_cfg dsi_cfg;
-	int bpp, nlanes, ret;
+	int bpp, ret;
 
 	/*
 	 * VFP_DSI should be less than VFP_DPI and VFP_DSI should be at
@@ -860,11 +686,9 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	if ((mode->hdisplay * bpp) % 32)
 		return MODE_H_ILLEGAL;
 
-	nlanes = output->dev->lanes;
-
-	ret = cdns_dsi_mode2cfg(dsi, mode, &dsi_cfg, &dphy_cfg, true);
+	ret = cdns_dsi_check_conf(dsi, mode, &dsi_cfg, true);
 	if (ret)
-		return MODE_CLOCK_RANGE;
+		return MODE_BAD;
 
 	return MODE_OK;
 }
@@ -885,9 +709,9 @@ static void cdns_dsi_bridge_disable(struct drm_bridge *bridge)
 	pm_runtime_put(dsi->base.dev);
 }
 
-static void cdns_dsi_hs_init(struct cdns_dsi *dsi,
-			     const struct cdns_dphy_cfg *dphy_cfg)
+static void cdns_dsi_hs_init(struct cdns_dsi *dsi)
 {
+	struct cdns_dsi_output *output = &dsi->output;
 	u32 status;
 
 	/*
@@ -898,30 +722,10 @@ static void cdns_dsi_hs_init(struct cdns_dsi *dsi,
 	       DPHY_CMN_PDN | DPHY_PLL_PDN,
 	       dsi->regs + MCTL_DPHY_CFG0);
 
-	/*
-	 * Configure the internal PSM clk divider so that the DPHY has a
-	 * 1MHz clk (or something close).
-	 */
-	WARN_ON_ONCE(cdns_dphy_setup_psm(dsi->dphy));
-
-	/*
-	 * Configure attach clk lanes to data lanes: the DPHY has 2 clk lanes
-	 * and 8 data lanes, each clk lane can be attache different set of
-	 * data lanes. The 2 groups are named 'left' and 'right', so here we
-	 * just say that we want the 'left' clk lane to drive the 'left' data
-	 * lanes.
-	 */
-	cdns_dphy_set_clk_lane_cfg(dsi->dphy, DPHY_CLK_CFG_LEFT_DRIVES_LEFT);
-
-	/*
-	 * Configure the DPHY PLL that will be used to generate the TX byte
-	 * clk.
-	 */
-	cdns_dphy_set_pll_cfg(dsi->dphy, dphy_cfg);
-
-	/* Start TX state machine. */
-	writel(DPHY_CMN_SSM_EN | DPHY_CMN_TX_MODE_EN,
-	       dsi->dphy->regs + DPHY_CMN_SSM);
+	phy_init(dsi->dphy);
+	phy_set_mode(dsi->dphy, PHY_MODE_MIPI_DPHY);
+	phy_configure(dsi->dphy, &output->phy_opts);
+	phy_power_on(dsi->dphy);
 
 	/* Activate the PLL and wait until it's locked. */
 	writel(PLL_LOCKED, dsi->regs + MCTL_MAIN_STS_CLR);
@@ -931,7 +735,7 @@ static void cdns_dsi_hs_init(struct cdns_dsi *dsi,
 					status & PLL_LOCKED, 100, 100));
 	/* De-assert data and clock reset lines. */
 	writel(DPHY_CMN_PSO | DPHY_ALL_D_PDN | DPHY_C_PDN | DPHY_CMN_PDN |
-	       DPHY_D_RSTB(dphy_cfg->nlanes) | DPHY_C_RSTB,
+	       DPHY_D_RSTB(output->dev->lanes) | DPHY_C_RSTB,
 	       dsi->regs + MCTL_DPHY_CFG0);
 }
 
@@ -977,7 +781,7 @@ static void cdns_dsi_bridge_enable(struct drm_bridge *bridge)
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_output *output = &dsi->output;
 	struct drm_display_mode *mode;
-	struct cdns_dphy_cfg dphy_cfg;
+	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
 	unsigned long tx_byte_period;
 	struct cdns_dsi_cfg dsi_cfg;
 	u32 tmp, reg_wakeup, div;
@@ -990,9 +794,9 @@ static void cdns_dsi_bridge_enable(struct drm_bridge *bridge)
 	bpp = mipi_dsi_pixel_format_to_bpp(output->dev->format);
 	nlanes = output->dev->lanes;
 
-	WARN_ON_ONCE(cdns_dsi_mode2cfg(dsi, mode, &dsi_cfg, &dphy_cfg, false));
+	WARN_ON_ONCE(cdns_dsi_check_conf(dsi, mode, &dsi_cfg, false));
 
-	cdns_dsi_hs_init(dsi, &dphy_cfg);
+	cdns_dsi_hs_init(dsi);
 	cdns_dsi_init_link(dsi);
 
 	writel(HBP_LEN(dsi_cfg.hbp) | HSA_LEN(dsi_cfg.hsa),
@@ -1028,9 +832,8 @@ static void cdns_dsi_bridge_enable(struct drm_bridge *bridge)
 		tmp -= DIV_ROUND_UP(DSI_EOT_PKT_SIZE, nlanes);
 
 	tx_byte_period = DIV_ROUND_DOWN_ULL((u64)NSEC_PER_SEC * 8,
-					    dphy_cfg.lane_bps);
-	reg_wakeup = cdns_dphy_get_wakeup_time_ns(dsi->dphy) /
-		     tx_byte_period;
+					    phy_cfg->hs_clk_rate);
+	reg_wakeup = (phy_cfg->hs_prepare + phy_cfg->hs_zero) / tx_byte_period;
 	writel(REG_WAKEUP_TIME(reg_wakeup) | REG_LINE_DURATION(tmp),
 	       dsi->regs + VID_DPHY_TIME);
 
@@ -1344,8 +1147,6 @@ static int __maybe_unused cdns_dsi_resume(struct device *dev)
 	reset_control_deassert(dsi->dsi_p_rst);
 	clk_prepare_enable(dsi->dsi_p_clk);
 	clk_prepare_enable(dsi->dsi_sys_clk);
-	clk_prepare_enable(dsi->dphy->psm_clk);
-	clk_prepare_enable(dsi->dphy->pll_ref_clk);
 
 	return 0;
 }
@@ -1354,8 +1155,6 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev)
 {
 	struct cdns_dsi *dsi = dev_get_drvdata(dev);
 
-	clk_disable_unprepare(dsi->dphy->pll_ref_clk);
-	clk_disable_unprepare(dsi->dphy->psm_clk);
 	clk_disable_unprepare(dsi->dsi_sys_clk);
 	clk_disable_unprepare(dsi->dsi_p_clk);
 	reset_control_assert(dsi->dsi_p_rst);
@@ -1366,121 +1165,6 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev)
 static UNIVERSAL_DEV_PM_OPS(cdns_dsi_pm_ops, cdns_dsi_suspend, cdns_dsi_resume,
 			    NULL);
 
-static unsigned long cdns_dphy_ref_get_wakeup_time_ns(struct cdns_dphy *dphy)
-{
-	/* Default wakeup time is 800 ns (in a simulated environment). */
-	return 800;
-}
-
-static void cdns_dphy_ref_set_pll_cfg(struct cdns_dphy *dphy,
-				      const struct cdns_dphy_cfg *cfg)
-{
-	u32 fbdiv_low, fbdiv_high;
-
-	fbdiv_low = (cfg->pll_fbdiv / 4) - 2;
-	fbdiv_high = cfg->pll_fbdiv - fbdiv_low - 2;
-
-	writel(DPHY_CMN_IPDIV_FROM_REG | DPHY_CMN_OPDIV_FROM_REG |
-	       DPHY_CMN_IPDIV(cfg->pll_ipdiv) |
-	       DPHY_CMN_OPDIV(cfg->pll_opdiv),
-	       dphy->regs + DPHY_CMN_OPIPDIV);
-	writel(DPHY_CMN_FBDIV_FROM_REG |
-	       DPHY_CMN_FBDIV_VAL(fbdiv_low, fbdiv_high),
-	       dphy->regs + DPHY_CMN_FBDIV);
-	writel(DPHY_CMN_PWM_HIGH(6) | DPHY_CMN_PWM_LOW(0x101) |
-	       DPHY_CMN_PWM_DIV(0x8),
-	       dphy->regs + DPHY_CMN_PWM);
-}
-
-static void cdns_dphy_ref_set_psm_div(struct cdns_dphy *dphy, u8 div)
-{
-	writel(DPHY_PSM_CFG_FROM_REG | DPHY_PSM_CLK_DIV(div),
-	       dphy->regs + DPHY_PSM_CFG);
-}
-
-/*
- * This is the reference implementation of DPHY hooks. Specific integration of
- * this IP may have to re-implement some of them depending on how they decided
- * to wire things in the SoC.
- */
-static const struct cdns_dphy_ops ref_dphy_ops = {
-	.get_wakeup_time_ns = cdns_dphy_ref_get_wakeup_time_ns,
-	.set_pll_cfg = cdns_dphy_ref_set_pll_cfg,
-	.set_psm_div = cdns_dphy_ref_set_psm_div,
-};
-
-static const struct of_device_id cdns_dphy_of_match[] = {
-	{ .compatible = "cdns,dphy", .data = &ref_dphy_ops },
-	{ /* sentinel */ },
-};
-
-static struct cdns_dphy *cdns_dphy_probe(struct platform_device *pdev)
-{
-	const struct of_device_id *match;
-	struct cdns_dphy *dphy;
-	struct of_phandle_args args;
-	struct resource res;
-	int ret;
-
-	ret = of_parse_phandle_with_args(pdev->dev.of_node, "phys",
-					 "#phy-cells", 0, &args);
-	if (ret)
-		return ERR_PTR(-ENOENT);
-
-	match = of_match_node(cdns_dphy_of_match, args.np);
-	if (!match || !match->data)
-		return ERR_PTR(-EINVAL);
-
-	dphy = devm_kzalloc(&pdev->dev, sizeof(*dphy), GFP_KERNEL);
-	if (!dphy)
-		return ERR_PTR(-ENOMEM);
-
-	dphy->ops = match->data;
-
-	ret = of_address_to_resource(args.np, 0, &res);
-	if (ret)
-		return ERR_PTR(ret);
-
-	dphy->regs = devm_ioremap_resource(&pdev->dev, &res);
-	if (IS_ERR(dphy->regs))
-		return ERR_CAST(dphy->regs);
-
-	dphy->psm_clk = of_clk_get_by_name(args.np, "psm");
-	if (IS_ERR(dphy->psm_clk))
-		return ERR_CAST(dphy->psm_clk);
-
-	dphy->pll_ref_clk = of_clk_get_by_name(args.np, "pll_ref");
-	if (IS_ERR(dphy->pll_ref_clk)) {
-		ret = PTR_ERR(dphy->pll_ref_clk);
-		goto err_put_psm_clk;
-	}
-
-	if (dphy->ops->probe) {
-		ret = dphy->ops->probe(dphy);
-		if (ret)
-			goto err_put_pll_ref_clk;
-	}
-
-	return dphy;
-
-err_put_pll_ref_clk:
-	clk_put(dphy->pll_ref_clk);
-
-err_put_psm_clk:
-	clk_put(dphy->psm_clk);
-
-	return ERR_PTR(ret);
-}
-
-static void cdns_dphy_remove(struct cdns_dphy *dphy)
-{
-	if (dphy->ops->remove)
-		dphy->ops->remove(dphy);
-
-	clk_put(dphy->pll_ref_clk);
-	clk_put(dphy->psm_clk);
-}
-
 static int cdns_dsi_drm_probe(struct platform_device *pdev)
 {
 	struct cdns_dsi *dsi;
@@ -1519,13 +1203,13 @@ static int cdns_dsi_drm_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return irq;
 
-	dsi->dphy = cdns_dphy_probe(pdev);
+	dsi->dphy = devm_phy_get(&pdev->dev, "dphy");
 	if (IS_ERR(dsi->dphy))
 		return PTR_ERR(dsi->dphy);
 
 	ret = clk_prepare_enable(dsi->dsi_p_clk);
 	if (ret)
-		goto err_remove_dphy;
+		return ret;
 
 	val = readl(dsi->regs + ID_REG);
 	if (REV_VENDOR_ID(val) != 0xcad) {
@@ -1583,9 +1267,6 @@ err_disable_runtime_pm:
 err_disable_pclk:
 	clk_disable_unprepare(dsi->dsi_p_clk);
 
-err_remove_dphy:
-	cdns_dphy_remove(dsi->dphy);
-
 	return ret;
 }
 
@@ -1595,7 +1276,6 @@ static int cdns_dsi_drm_remove(struct platform_device *pdev)
 
 	mipi_dsi_host_unregister(&dsi->base);
 	pm_runtime_disable(&pdev->dev);
-	cdns_dphy_remove(dsi->dphy);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/bridge/dumb-vga-dac.c b/drivers/gpu/drm/bridge/dumb-vga-dac.c
index 9b706789a341..0805801f4e94 100644
--- a/drivers/gpu/drm/bridge/dumb-vga-dac.c
+++ b/drivers/gpu/drm/bridge/dumb-vga-dac.c
@@ -18,7 +18,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 struct dumb_vga {
 	struct drm_bridge	bridge;
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 2136c97aeb8e..a01028ec4de6 100644
--- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
+++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
@@ -36,8 +36,8 @@
 #include <linux/of.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #define EDID_EXT_BLOCK_CNT 0x7E
diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c
index a3e817abace1..fb335afea4cf 100644
--- a/drivers/gpu/drm/bridge/nxp-ptn3460.c
+++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c
@@ -22,10 +22,10 @@
 #include <linux/of_gpio.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #define PTN3460_EDID_ADDR			0x0
diff --git a/drivers/gpu/drm/bridge/panel.c b/drivers/gpu/drm/bridge/panel.c
index 37baa79e95c3..38eeaf8ba959 100644
--- a/drivers/gpu/drm/bridge/panel.c
+++ b/drivers/gpu/drm/bridge/panel.c
@@ -12,9 +12,9 @@
 #include <drm/drm_panel.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_connector.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_panel.h>
 
 struct panel_bridge {
diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c
index 7334d1b62b71..fda1395b7481 100644
--- a/drivers/gpu/drm/bridge/parade-ps8622.c
+++ b/drivers/gpu/drm/bridge/parade-ps8622.c
@@ -26,9 +26,9 @@
 #include <linux/regulator/consumer.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 /* Brightness scale on the Parade chip */
diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c
index a5d58f7035c1..08e12fef1349 100644
--- a/drivers/gpu/drm/bridge/sii902x.c
+++ b/drivers/gpu/drm/bridge/sii902x.c
@@ -30,8 +30,8 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #define SII902X_TPI_VIDEO_DATA			0x0
 
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
index 129f464cbeb1..a63e5f0dae56 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
@@ -25,9 +25,10 @@
 #include <drm/drm_of.h>
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder_slave.h>
+#include <drm/drm_scdc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/bridge/dw_hdmi.h>
 
 #include <uapi/linux/media-bus-format.h>
@@ -43,6 +44,11 @@
 
 #define HDMI_EDID_LEN		512
 
+/* DW-HDMI Controller >= 0x200a are at least compliant with SCDC version 1 */
+#define SCDC_MIN_SOURCE_VERSION	0x1
+
+#define HDMI14_MAX_TMDSCLK	340000000
+
 enum hdmi_datamap {
 	RGB444_8B = 0x01,
 	RGB444_10B = 0x03,
@@ -93,6 +99,7 @@ struct hdmi_vmode {
 	unsigned int mpixelclock;
 	unsigned int mpixelrepetitioninput;
 	unsigned int mpixelrepetitionoutput;
+	unsigned int mtmdsclock;
 };
 
 struct hdmi_data_info {
@@ -537,7 +544,7 @@ static void hdmi_init_clk_regenerator(struct dw_hdmi *hdmi)
 static void hdmi_clk_regenerator_update_pixel_clock(struct dw_hdmi *hdmi)
 {
 	mutex_lock(&hdmi->audio_mutex);
-	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mpixelclock,
+	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mtmdsclock,
 				 hdmi->sample_rate);
 	mutex_unlock(&hdmi->audio_mutex);
 }
@@ -546,7 +553,7 @@ void dw_hdmi_set_sample_rate(struct dw_hdmi *hdmi, unsigned int rate)
 {
 	mutex_lock(&hdmi->audio_mutex);
 	hdmi->sample_rate = rate;
-	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mpixelclock,
+	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mtmdsclock,
 				 hdmi->sample_rate);
 	mutex_unlock(&hdmi->audio_mutex);
 }
@@ -647,6 +654,20 @@ static bool hdmi_bus_fmt_is_yuv422(unsigned int bus_format)
 	}
 }
 
+static bool hdmi_bus_fmt_is_yuv420(unsigned int bus_format)
+{
+	switch (bus_format) {
+	case MEDIA_BUS_FMT_UYYVYY8_0_5X24:
+	case MEDIA_BUS_FMT_UYYVYY10_0_5X30:
+	case MEDIA_BUS_FMT_UYYVYY12_0_5X36:
+	case MEDIA_BUS_FMT_UYYVYY16_0_5X48:
+		return true;
+
+	default:
+		return false;
+	}
+}
+
 static int hdmi_bus_fmt_color_depth(unsigned int bus_format)
 {
 	switch (bus_format) {
@@ -876,7 +897,8 @@ static void hdmi_video_packetize(struct dw_hdmi *hdmi)
 	u8 val, vp_conf;
 
 	if (hdmi_bus_fmt_is_rgb(hdmi->hdmi_data.enc_out_bus_format) ||
-	    hdmi_bus_fmt_is_yuv444(hdmi->hdmi_data.enc_out_bus_format)) {
+	    hdmi_bus_fmt_is_yuv444(hdmi->hdmi_data.enc_out_bus_format) ||
+	    hdmi_bus_fmt_is_yuv420(hdmi->hdmi_data.enc_out_bus_format)) {
 		switch (hdmi_bus_fmt_color_depth(
 					hdmi->hdmi_data.enc_out_bus_format)) {
 		case 8:
@@ -1015,6 +1037,33 @@ void dw_hdmi_phy_i2c_write(struct dw_hdmi *hdmi, unsigned short data,
 }
 EXPORT_SYMBOL_GPL(dw_hdmi_phy_i2c_write);
 
+/*
+ * HDMI2.0 Specifies the following procedure for High TMDS Bit Rates:
+ * - The Source shall suspend transmission of the TMDS clock and data
+ * - The Source shall write to the TMDS_Bit_Clock_Ratio bit to change it
+ * from a 0 to a 1 or from a 1 to a 0
+ * - The Source shall allow a minimum of 1 ms and a maximum of 100 ms from
+ * the time the TMDS_Bit_Clock_Ratio bit is written until resuming
+ * transmission of TMDS clock and data
+ *
+ * To respect the 100ms maximum delay, the dw_hdmi_set_high_tmds_clock_ratio()
+ * helper should called right before enabling the TMDS Clock and Data in
+ * the PHY configuration callback.
+ */
+void dw_hdmi_set_high_tmds_clock_ratio(struct dw_hdmi *hdmi)
+{
+	unsigned long mtmdsclock = hdmi->hdmi_data.video_mode.mtmdsclock;
+
+	/* Control for TMDS Bit Period/TMDS Clock-Period Ratio */
+	if (hdmi->connector.display_info.hdmi.scdc.supported) {
+		if (mtmdsclock > HDMI14_MAX_TMDSCLK)
+			drm_scdc_set_high_tmds_clock_ratio(hdmi->ddc, 1);
+		else
+			drm_scdc_set_high_tmds_clock_ratio(hdmi->ddc, 0);
+	}
+}
+EXPORT_SYMBOL_GPL(dw_hdmi_set_high_tmds_clock_ratio);
+
 static void dw_hdmi_phy_enable_powerdown(struct dw_hdmi *hdmi, bool enable)
 {
 	hdmi_mask_writeb(hdmi, !enable, HDMI_PHY_CONF0,
@@ -1165,6 +1214,8 @@ static int hdmi_phy_configure_dwc_hdmi_3d_tx(struct dw_hdmi *hdmi,
 	const struct dw_hdmi_curr_ctrl *curr_ctrl = pdata->cur_ctr;
 	const struct dw_hdmi_phy_config *phy_config = pdata->phy_config;
 
+	/* TOFIX Will need 420 specific PHY configuration tables */
+
 	/* PLL/MPLL Cfg - always match on final entry */
 	for (; mpll_config->mpixelclock != ~0UL; mpll_config++)
 		if (mpixelclock <= mpll_config->mpixelclock)
@@ -1212,10 +1263,13 @@ static int hdmi_phy_configure(struct dw_hdmi *hdmi)
 	const struct dw_hdmi_phy_data *phy = hdmi->phy.data;
 	const struct dw_hdmi_plat_data *pdata = hdmi->plat_data;
 	unsigned long mpixelclock = hdmi->hdmi_data.video_mode.mpixelclock;
+	unsigned long mtmdsclock = hdmi->hdmi_data.video_mode.mtmdsclock;
 	int ret;
 
 	dw_hdmi_phy_power_off(hdmi);
 
+	dw_hdmi_set_high_tmds_clock_ratio(hdmi);
+
 	/* Leave low power consumption mode by asserting SVSRET. */
 	if (phy->has_svsret)
 		dw_hdmi_phy_enable_svsret(hdmi, 1);
@@ -1237,6 +1291,10 @@ static int hdmi_phy_configure(struct dw_hdmi *hdmi)
 		return ret;
 	}
 
+	/* Wait for resuming transmission of TMDS clock and data */
+	if (mtmdsclock > HDMI14_MAX_TMDSCLK)
+		msleep(100);
+
 	return dw_hdmi_phy_power_on(hdmi);
 }
 
@@ -1351,6 +1409,8 @@ static void hdmi_config_AVI(struct dw_hdmi *hdmi, struct drm_display_mode *mode)
 		frame.colorspace = HDMI_COLORSPACE_YUV444;
 	else if (hdmi_bus_fmt_is_yuv422(hdmi->hdmi_data.enc_out_bus_format))
 		frame.colorspace = HDMI_COLORSPACE_YUV422;
+	else if (hdmi_bus_fmt_is_yuv420(hdmi->hdmi_data.enc_out_bus_format))
+		frame.colorspace = HDMI_COLORSPACE_YUV420;
 	else
 		frame.colorspace = HDMI_COLORSPACE_RGB;
 
@@ -1504,17 +1564,23 @@ static void hdmi_config_vendor_specific_infoframe(struct dw_hdmi *hdmi,
 static void hdmi_av_composer(struct dw_hdmi *hdmi,
 			     const struct drm_display_mode *mode)
 {
-	u8 inv_val;
+	u8 inv_val, bytes;
+	struct drm_hdmi_info *hdmi_info = &hdmi->connector.display_info.hdmi;
 	struct hdmi_vmode *vmode = &hdmi->hdmi_data.video_mode;
 	int hblank, vblank, h_de_hs, v_de_vs, hsync_len, vsync_len;
-	unsigned int vdisplay;
+	unsigned int vdisplay, hdisplay;
 
-	vmode->mpixelclock = mode->clock * 1000;
+	vmode->mtmdsclock = vmode->mpixelclock = mode->clock * 1000;
 
 	dev_dbg(hdmi->dev, "final pixclk = %d\n", vmode->mpixelclock);
 
+	if (hdmi_bus_fmt_is_yuv420(hdmi->hdmi_data.enc_out_bus_format))
+		vmode->mtmdsclock /= 2;
+
 	/* Set up HDMI_FC_INVIDCONF */
-	inv_val = (hdmi->hdmi_data.hdcp_enable ?
+	inv_val = (hdmi->hdmi_data.hdcp_enable ||
+		   vmode->mtmdsclock > HDMI14_MAX_TMDSCLK ||
+		   hdmi_info->scdc.scrambling.low_rates ?
 		HDMI_FC_INVIDCONF_HDCP_KEEPOUT_ACTIVE :
 		HDMI_FC_INVIDCONF_HDCP_KEEPOUT_INACTIVE);
 
@@ -1547,6 +1613,22 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 
 	hdmi_writeb(hdmi, inv_val, HDMI_FC_INVIDCONF);
 
+	hdisplay = mode->hdisplay;
+	hblank = mode->htotal - mode->hdisplay;
+	h_de_hs = mode->hsync_start - mode->hdisplay;
+	hsync_len = mode->hsync_end - mode->hsync_start;
+
+	/*
+	 * When we're setting a YCbCr420 mode, we need
+	 * to adjust the horizontal timing to suit.
+	 */
+	if (hdmi_bus_fmt_is_yuv420(hdmi->hdmi_data.enc_out_bus_format)) {
+		hdisplay /= 2;
+		hblank /= 2;
+		h_de_hs /= 2;
+		hsync_len /= 2;
+	}
+
 	vdisplay = mode->vdisplay;
 	vblank = mode->vtotal - mode->vdisplay;
 	v_de_vs = mode->vsync_start - mode->vdisplay;
@@ -1563,16 +1645,54 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 		vsync_len /= 2;
 	}
 
+	/* Scrambling Control */
+	if (hdmi_info->scdc.supported) {
+		if (vmode->mtmdsclock > HDMI14_MAX_TMDSCLK ||
+		    hdmi_info->scdc.scrambling.low_rates) {
+			/*
+			 * HDMI2.0 Specifies the following procedure:
+			 * After the Source Device has determined that
+			 * SCDC_Present is set (=1), the Source Device should
+			 * write the accurate Version of the Source Device
+			 * to the Source Version field in the SCDCS.
+			 * Source Devices compliant shall set the
+			 * Source Version = 1.
+			 */
+			drm_scdc_readb(&hdmi->i2c->adap, SCDC_SINK_VERSION,
+				       &bytes);
+			drm_scdc_writeb(&hdmi->i2c->adap, SCDC_SOURCE_VERSION,
+				min_t(u8, bytes, SCDC_MIN_SOURCE_VERSION));
+
+			/* Enabled Scrambling in the Sink */
+			drm_scdc_set_scrambling(&hdmi->i2c->adap, 1);
+
+			/*
+			 * To activate the scrambler feature, you must ensure
+			 * that the quasi-static configuration bit
+			 * fc_invidconf.HDCP_keepout is set at configuration
+			 * time, before the required mc_swrstzreq.tmdsswrst_req
+			 * reset request is issued.
+			 */
+			hdmi_writeb(hdmi, (u8)~HDMI_MC_SWRSTZ_TMDSSWRST_REQ,
+				    HDMI_MC_SWRSTZ);
+			hdmi_writeb(hdmi, 1, HDMI_FC_SCRAMBLER_CTRL);
+		} else {
+			hdmi_writeb(hdmi, 0, HDMI_FC_SCRAMBLER_CTRL);
+			hdmi_writeb(hdmi, (u8)~HDMI_MC_SWRSTZ_TMDSSWRST_REQ,
+				    HDMI_MC_SWRSTZ);
+			drm_scdc_set_scrambling(&hdmi->i2c->adap, 0);
+		}
+	}
+
 	/* Set up horizontal active pixel width */
-	hdmi_writeb(hdmi, mode->hdisplay >> 8, HDMI_FC_INHACTV1);
-	hdmi_writeb(hdmi, mode->hdisplay, HDMI_FC_INHACTV0);
+	hdmi_writeb(hdmi, hdisplay >> 8, HDMI_FC_INHACTV1);
+	hdmi_writeb(hdmi, hdisplay, HDMI_FC_INHACTV0);
 
 	/* Set up vertical active lines */
 	hdmi_writeb(hdmi, vdisplay >> 8, HDMI_FC_INVACTV1);
 	hdmi_writeb(hdmi, vdisplay, HDMI_FC_INVACTV0);
 
 	/* Set up horizontal blanking pixel region width */
-	hblank = mode->htotal - mode->hdisplay;
 	hdmi_writeb(hdmi, hblank >> 8, HDMI_FC_INHBLANK1);
 	hdmi_writeb(hdmi, hblank, HDMI_FC_INHBLANK0);
 
@@ -1580,7 +1700,6 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	hdmi_writeb(hdmi, vblank, HDMI_FC_INVBLANK);
 
 	/* Set up HSYNC active edge delay width (in pixel clks) */
-	h_de_hs = mode->hsync_start - mode->hdisplay;
 	hdmi_writeb(hdmi, h_de_hs >> 8, HDMI_FC_HSYNCINDELAY1);
 	hdmi_writeb(hdmi, h_de_hs, HDMI_FC_HSYNCINDELAY0);
 
@@ -1588,7 +1707,6 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	hdmi_writeb(hdmi, v_de_vs, HDMI_FC_VSYNCINDELAY);
 
 	/* Set up HSYNC active pulse width (in pixel clks) */
-	hsync_len = mode->hsync_end - mode->hsync_start;
 	hdmi_writeb(hdmi, hsync_len >> 8, HDMI_FC_HSYNCINWIDTH1);
 	hdmi_writeb(hdmi, hsync_len, HDMI_FC_HSYNCINWIDTH0);
 
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
index 9d90eb9c46e5..3f3c616eba97 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
@@ -255,6 +255,7 @@
 #define HDMI_FC_MASK2                           0x10DA
 #define HDMI_FC_POL2                            0x10DB
 #define HDMI_FC_PRCONF                          0x10E0
+#define HDMI_FC_SCRAMBLER_CTRL                  0x10E1
 
 #define HDMI_FC_GMD_STAT                        0x1100
 #define HDMI_FC_GMD_EN                          0x1101
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
index 23a5977a3b0a..e915ae8c9a92 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
@@ -19,9 +19,9 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/bridge/dw_mipi_dsi.h>
 #include <video/mipi_display.h>
 
diff --git a/drivers/gpu/drm/bridge/tc358764.c b/drivers/gpu/drm/bridge/tc358764.c
index afd491018bfc..a20e454ddd64 100644
--- a/drivers/gpu/drm/bridge/tc358764.c
+++ b/drivers/gpu/drm/bridge/tc358764.c
@@ -9,11 +9,11 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.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_probe_helper.h>
 #include <drm/drmP.h>
 #include <linux/gpio/consumer.h>
 #include <linux/of_graph.h>
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 4df07f4cbff5..888980d4bc74 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -34,11 +34,11 @@
 #include <linux/slab.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 /* Registers */
 
@@ -98,6 +98,8 @@
 #define DP0_STARTVAL		0x064c
 #define DP0_ACTIVEVAL		0x0650
 #define DP0_SYNCVAL		0x0654
+#define SYNCVAL_HS_POL_ACTIVE_LOW	(1 << 15)
+#define SYNCVAL_VS_POL_ACTIVE_LOW	(1 << 31)
 #define DP0_MISC		0x0658
 #define TU_SIZE_RECOMMENDED		(63) /* LSCLK cycles per TU */
 #define BPC_6				(0 << 5)
@@ -142,6 +144,8 @@
 #define DP0_LTLOOPCTRL		0x06d8
 #define DP0_SNKLTCTRL		0x06e4
 
+#define DP1_SRCCTRL		0x07a0
+
 /* PHY */
 #define DP_PHY_CTRL		0x0800
 #define DP_PHY_RST			BIT(28)  /* DP PHY Global Soft Reset */
@@ -150,6 +154,7 @@
 #define PHY_M1_RST			BIT(12)  /* Reset PHY1 Main Channel */
 #define PHY_RDY				BIT(16)  /* PHY Main Channels Ready */
 #define PHY_M0_RST			BIT(8)   /* Reset PHY0 Main Channel */
+#define PHY_2LANE			BIT(2)   /* PHY Enable 2 lanes */
 #define PHY_A0_EN			BIT(1)   /* PHY Aux Channel0 Enable */
 #define PHY_M0_EN			BIT(0)   /* PHY Main Channel0 Enable */
 
@@ -540,6 +545,7 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	unsigned long rate;
 	u32 value;
 	int ret;
+	u32 dp_phy_ctrl;
 
 	rate = clk_get_rate(tc->refclk);
 	switch (rate) {
@@ -564,7 +570,10 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	value |= SYSCLK_SEL_LSCLK | LSCLK_DIV_2;
 	tc_write(SYS_PLLPARAM, value);
 
-	tc_write(DP_PHY_CTRL, BGREN | PWR_SW_EN | BIT(2) | PHY_A0_EN);
+	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_A0_EN;
+	if (tc->link.base.num_lanes == 2)
+		dp_phy_ctrl |= PHY_2LANE;
+	tc_write(DP_PHY_CTRL, dp_phy_ctrl);
 
 	/*
 	 * Initially PLLs are in bypass. Force PLL parameter update,
@@ -720,7 +729,9 @@ static int tc_set_video_mode(struct tc_data *tc,
 
 	tc_write(DP0_ACTIVEVAL, (mode->vdisplay << 16) | (mode->hdisplay));
 
-	tc_write(DP0_SYNCVAL, (vsync_len << 16) | (hsync_len << 0));
+	tc_write(DP0_SYNCVAL, (vsync_len << 16) | (hsync_len << 0) |
+		 ((mode->flags & DRM_MODE_FLAG_NHSYNC) ? SYNCVAL_HS_POL_ACTIVE_LOW : 0) |
+		 ((mode->flags & DRM_MODE_FLAG_NVSYNC) ? SYNCVAL_VS_POL_ACTIVE_LOW : 0));
 
 	tc_write(DPIPXLFMT, VS_POL_ACTIVE_LOW | HS_POL_ACTIVE_LOW |
 		 DE_POL_ACTIVE_HIGH | SUB_CFG_TYPE_CONFIG1 | DPI_BPP_RGB888);
@@ -830,12 +841,11 @@ static int tc_main_link_setup(struct tc_data *tc)
 	if (!tc->mode)
 		return -EINVAL;
 
-	/* from excel file - DP0_SrcCtrl */
-	tc_write(DP0_SRCCTRL, DP0_SRCCTRL_SCRMBLDIS | DP0_SRCCTRL_EN810B |
-		 DP0_SRCCTRL_LANESKEW | DP0_SRCCTRL_LANES_2 |
-		 DP0_SRCCTRL_BW27 | DP0_SRCCTRL_AUTOCORRECT);
-	/* from excel file - DP1_SrcCtrl */
-	tc_write(0x07a0, 0x00003083);
+	tc_write(DP0_SRCCTRL, tc_srcctrl(tc));
+	/* SSCG and BW27 on DP1 must be set to the same as on DP0 */
+	tc_write(DP1_SRCCTRL,
+		 (tc->link.spread ? DP0_SRCCTRL_SSCG : 0) |
+		 ((tc->link.base.rate != 162000) ? DP0_SRCCTRL_BW27 : 0));
 
 	rate = clk_get_rate(tc->refclk);
 	switch (rate) {
@@ -856,8 +866,11 @@ static int tc_main_link_setup(struct tc_data *tc)
 	}
 	value |= SYSCLK_SEL_LSCLK | LSCLK_DIV_2;
 	tc_write(SYS_PLLPARAM, value);
+
 	/* Setup Main Link */
-	dp_phy_ctrl = BGREN | PWR_SW_EN | BIT(2) | PHY_A0_EN |  PHY_M0_EN;
+	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_A0_EN | PHY_M0_EN;
+	if (tc->link.base.num_lanes == 2)
+		dp_phy_ctrl |= PHY_2LANE;
 	tc_write(DP_PHY_CTRL, dp_phy_ctrl);
 	msleep(100);
 
@@ -1106,10 +1119,20 @@ static bool tc_bridge_mode_fixup(struct drm_bridge *bridge,
 static enum drm_mode_status tc_connector_mode_valid(struct drm_connector *connector,
 				   struct drm_display_mode *mode)
 {
+	struct tc_data *tc = connector_to_tc(connector);
+	u32 req, avail;
+	u32 bits_per_pixel = 24;
+
 	/* DPI interface clock limitation: upto 154 MHz */
 	if (mode->clock > 154000)
 		return MODE_CLOCK_HIGH;
 
+	req = mode->clock * bits_per_pixel / 8;
+	avail = tc->link.base.num_lanes * tc->link.base.rate;
+
+	if (req > avail)
+		return MODE_BAD;
+
 	return MODE_OK;
 }
 
@@ -1187,7 +1210,8 @@ static int tc_bridge_attach(struct drm_bridge *bridge)
 	/* Create eDP connector */
 	drm_connector_helper_add(&tc->connector, &tc_connector_helper_funcs);
 	ret = drm_connector_init(drm, &tc->connector, &tc_connector_funcs,
-				 DRM_MODE_CONNECTOR_eDP);
+				 tc->panel ? DRM_MODE_CONNECTOR_eDP :
+				 DRM_MODE_CONNECTOR_DisplayPort);
 	if (ret)
 		return ret;
 
@@ -1196,6 +1220,10 @@ static int tc_bridge_attach(struct drm_bridge *bridge)
 
 	drm_display_info_set_bus_formats(&tc->connector.display_info,
 					 &bus_format, 1);
+	tc->connector.display_info.bus_flags =
+		DRM_BUS_FLAG_DE_HIGH |
+		DRM_BUS_FLAG_PIXDATA_NEGEDGE |
+		DRM_BUS_FLAG_SYNC_NEGEDGE;
 	drm_connector_attach_encoder(&tc->connector, tc->bridge.encoder);
 
 	return 0;
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 10243965ee7c..f72ee137e5f1 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -6,11 +6,11 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.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>
 #include <linux/clk.h>
 #include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c
index c3e32138c6bb..7bfb4f338813 100644
--- a/drivers/gpu/drm/bridge/ti-tfp410.c
+++ b/drivers/gpu/drm/bridge/ti-tfp410.c
@@ -20,7 +20,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #define HOTPLUG_DEBOUNCE_MS		1100
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.c b/drivers/gpu/drm/cirrus/cirrus_drv.c
index db40b77c7f7c..8ec880f3a322 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.c
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.c
@@ -12,6 +12,7 @@
 #include <linux/console.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "cirrus_drv.h"
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
index ed7dcf212a34..7f9bc32af685 100644
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ b/drivers/gpu/drm/cirrus/cirrus_mode.c
@@ -17,6 +17,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <video/cirrus.h>
 
@@ -359,10 +360,70 @@ static const struct drm_crtc_helper_funcs cirrus_helper_funcs = {
 };
 
 /* CRTC setup */
+static const uint32_t cirrus_formats_16[] = {
+	DRM_FORMAT_RGB565,
+};
+
+static const uint32_t cirrus_formats_24[] = {
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_RGB565,
+};
+
+static const uint32_t cirrus_formats_32[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_RGB565,
+};
+
+static struct drm_plane *cirrus_primary_plane(struct drm_device *dev)
+{
+	const uint32_t *formats;
+	uint32_t nformats;
+	struct drm_plane *primary;
+	int ret;
+
+	switch (cirrus_bpp) {
+	case 16:
+		formats = cirrus_formats_16;
+		nformats = ARRAY_SIZE(cirrus_formats_16);
+		break;
+	case 24:
+		formats = cirrus_formats_24;
+		nformats = ARRAY_SIZE(cirrus_formats_24);
+		break;
+	case 32:
+		formats = cirrus_formats_32;
+		nformats = ARRAY_SIZE(cirrus_formats_32);
+		break;
+	default:
+		return NULL;
+	}
+
+	primary = kzalloc(sizeof(*primary), GFP_KERNEL);
+	if (primary == NULL) {
+		DRM_DEBUG_KMS("Failed to allocate primary plane\n");
+		return NULL;
+	}
+
+	ret = drm_universal_plane_init(dev, primary, 0,
+				       &drm_primary_helper_funcs,
+				       formats, nformats,
+				       NULL,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret) {
+		kfree(primary);
+		primary = NULL;
+	}
+
+	return primary;
+}
+
 static void cirrus_crtc_init(struct drm_device *dev)
 {
 	struct cirrus_device *cdev = dev->dev_private;
 	struct cirrus_crtc *cirrus_crtc;
+	struct drm_plane *primary;
 
 	cirrus_crtc = kzalloc(sizeof(struct cirrus_crtc) +
 			      (CIRRUSFB_CONN_LIMIT * sizeof(struct drm_connector *)),
@@ -371,7 +432,15 @@ static void cirrus_crtc_init(struct drm_device *dev)
 	if (cirrus_crtc == NULL)
 		return;
 
-	drm_crtc_init(dev, &cirrus_crtc->base, &cirrus_crtc_funcs);
+	primary = cirrus_primary_plane(dev);
+	if (primary == NULL) {
+		kfree(cirrus_crtc);
+		return;
+	}
+
+	drm_crtc_init_with_planes(dev, &cirrus_crtc->base,
+				  primary, NULL,
+				  &cirrus_crtc_funcs, NULL);
 
 	drm_mode_crtc_set_gamma_size(&cirrus_crtc->base, CIRRUS_LUT_SIZE);
 	cdev->mode_info.crtc = cirrus_crtc;
diff --git a/drivers/gpu/drm/drm_agpsupport.c b/drivers/gpu/drm/drm_agpsupport.c
index 737f02885c28..40fba1c04dfc 100644
--- a/drivers/gpu/drm/drm_agpsupport.c
+++ b/drivers/gpu/drm/drm_agpsupport.c
@@ -348,7 +348,7 @@ int drm_agp_bind_ioctl(struct drm_device *dev, void *data,
  * \return zero on success or a negative number on failure.
  *
  * Verifies the AGP device is present and has been acquired and looks up the
- * AGP memory entry. If the memory it's currently bound, unbind it via
+ * AGP memory entry. If the memory is currently bound, unbind it via
  * unbind_agp(). Frees it via free_agp() as well as the entry itself
  * and unlinks from the doubly linked list it's inserted in.
  */
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 54e2ae614dcc..c53ecbd9abdd 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -29,7 +29,6 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_uapi.h>
 #include <drm/drm_plane_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_writeback.h>
 #include <drm/drm_damage_helper.h>
@@ -331,10 +330,17 @@ update_connector_routing(struct drm_atomic_state *state,
 	 * Since the connector can be unregistered at any point during an
 	 * atomic check or commit, this is racy. But that's OK: all we care
 	 * about is ensuring that userspace can't do anything but shut off the
-	 * display on a connector that was destroyed after its been notified,
+	 * display on a connector that was destroyed after it's been notified,
 	 * not before.
+	 *
+	 * Additionally, we also want to ignore connector registration when
+	 * we're trying to restore an atomic state during system resume since
+	 * there's a chance the connector may have been destroyed during the
+	 * process, but it's better to ignore that then cause
+	 * drm_atomic_helper_resume() to fail.
 	 */
-	if (drm_connector_is_unregistered(connector) && crtc_state->active) {
+	if (!state->duplicated && drm_connector_is_unregistered(connector) &&
+	    crtc_state->active) {
 		DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] is not registered\n",
 				 connector->base.id, connector->name);
 		return -EINVAL;
@@ -686,7 +692,7 @@ drm_atomic_helper_check_modeset(struct drm_device *dev,
 
 	/*
 	 * After all the routing has been prepared we need to add in any
-	 * connector which is itself unchanged, but who's crtc changes it's
+	 * connector which is itself unchanged, but whose crtc changes its
 	 * configuration. This must be done before calling mode_fixup in case a
 	 * crtc only changed its mode but has the same set of connectors.
 	 */
@@ -1671,7 +1677,7 @@ EXPORT_SYMBOL(drm_atomic_helper_async_commit);
  * drm_atomic_helper_setup_commit() and related functions.
  *
  * Committing the actual hardware state is done through the
- * &drm_mode_config_helper_funcs.atomic_commit_tail callback, or it's default
+ * &drm_mode_config_helper_funcs.atomic_commit_tail callback, or its default
  * implementation drm_atomic_helper_commit_tail().
  *
  * RETURNS:
@@ -1894,7 +1900,7 @@ crtc_or_fake_commit(struct drm_atomic_state *state, struct drm_crtc *crtc)
  * functions. drm_atomic_helper_wait_for_dependencies() must be called before
  * actually committing the hardware state, and for nonblocking commits this call
  * must be placed in the async worker. See also drm_atomic_helper_swap_state()
- * and it's stall parameter, for when a driver's commit hooks look at the
+ * and its stall parameter, for when a driver's commit hooks look at the
  * &drm_crtc.state, &drm_plane.state or &drm_connector.state pointer directly.
  *
  * Completion of the hardware commit step must be signalled using
@@ -3181,6 +3187,7 @@ drm_atomic_helper_duplicate_state(struct drm_device *dev,
 		return ERR_PTR(-ENOMEM);
 
 	state->acquire_ctx = ctx;
+	state->duplicated = true;
 
 	drm_for_each_crtc(crtc, dev) {
 		struct drm_crtc_state *crtc_state;
diff --git a/drivers/gpu/drm/drm_atomic_uapi.c b/drivers/gpu/drm/drm_atomic_uapi.c
index c40889888a16..0aabd401d3ca 100644
--- a/drivers/gpu/drm/drm_atomic_uapi.c
+++ b/drivers/gpu/drm/drm_atomic_uapi.c
@@ -44,8 +44,8 @@
  * DOC: overview
  *
  * This file contains the marshalling and demarshalling glue for the atomic UAPI
- * in all it's form: The monster ATOMIC IOCTL itself, code for GET_PROPERTY and
- * SET_PROPERTY IOCTls. Plus interface functions for compatibility helpers and
+ * in all its forms: The monster ATOMIC IOCTL itself, code for GET_PROPERTY and
+ * SET_PROPERTY IOCTLs. Plus interface functions for compatibility helpers and
  * drivers which have special needs to construct their own atomic updates, e.g.
  * for load detect or similiar.
  */
@@ -1296,12 +1296,11 @@ int drm_mode_atomic_ioctl(struct drm_device *dev,
 			(arg->flags & DRM_MODE_PAGE_FLIP_EVENT))
 		return -EINVAL;
 
-	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
-
 	state = drm_atomic_state_alloc(dev);
 	if (!state)
 		return -ENOMEM;
 
+	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
 	state->acquire_ctx = &ctx;
 	state->allow_modeset = !!(arg->flags & DRM_MODE_ATOMIC_ALLOW_MODESET);
 
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index a39ab2193bfe..e407adb033e7 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -494,7 +494,7 @@ int drm_legacy_getmap_ioctl(struct drm_device *dev, void *data,
  * isn't in use.
  *
  * Searches the map on drm_device::maplist, removes it from the list, see if
- * its being used, and free any associate resource (such as MTRR's) if it's not
+ * it's being used, and free any associated resource (such as MTRR's) if it's not
  * being on use.
  *
  * \sa drm_legacy_addmap
@@ -621,7 +621,7 @@ int drm_legacy_rmmap_ioctl(struct drm_device *dev, void *data,
 		}
 	}
 
-	/* List has wrapped around to the head pointer, or its empty we didn't
+	/* List has wrapped around to the head pointer, or it's empty we didn't
 	 * find anything.
 	 */
 	if (list_empty(&dev->maplist) || !map) {
diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index 847539645558..dd40eff0911c 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -1066,7 +1066,7 @@ EXPORT_SYMBOL(drm_mode_create_dvi_i_properties);
  *
  * content type (HDMI specific):
  *	Indicates content type setting to be used in HDMI infoframes to indicate
- *	content type for the external device, so that it adjusts it's display
+ *	content type for the external device, so that it adjusts its display
  *	settings accordingly.
  *
  *	The value of this property can be one of the following:
@@ -1367,7 +1367,7 @@ EXPORT_SYMBOL(drm_mode_create_scaling_mode_property);
  *
  *	Absence of the property should indicate absence of support.
  *
- * "vrr_enabled":
+ * "VRR_ENABLED":
  *	Default &drm_crtc boolean property that notifies the driver that the
  *	content on the CRTC is suitable for variable refresh rate presentation.
  *	The driver will take this property as a hint to enable variable
diff --git a/drivers/gpu/drm/drm_damage_helper.c b/drivers/gpu/drm/drm_damage_helper.c
index e16aa5ae00b4..ee67c96841fa 100644
--- a/drivers/gpu/drm/drm_damage_helper.c
+++ b/drivers/gpu/drm/drm_damage_helper.c
@@ -32,6 +32,7 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_damage_helper.h>
+#include <drm/drm_device.h>
 
 /**
  * DOC: overview
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 54120b6319e7..54a6414c5d96 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -1360,7 +1360,20 @@ int drm_dp_read_desc(struct drm_dp_aux *aux, struct drm_dp_desc *desc,
 EXPORT_SYMBOL(drm_dp_read_desc);
 
 /**
- * DRM DP Helpers for DSC
+ * drm_dp_dsc_sink_max_slice_count() - Get the max slice count
+ * supported by the DSC sink.
+ * @dsc_dpcd: DSC capabilities from DPCD
+ * @is_edp: true if its eDP, false for DP
+ *
+ * Read the slice capabilities DPCD register from DSC sink to get
+ * the maximum slice count supported. This is used to populate
+ * the DSC parameters in the &struct drm_dsc_config by the driver.
+ * Driver creates an infoframe using these parameters to populate
+ * &struct drm_dsc_pps_infoframe. These are sent to the sink using DSC
+ * infoframe using the helper function drm_dsc_pps_infoframe_pack()
+ *
+ * Returns:
+ * Maximum slice count supported by DSC sink or 0 its invalid
  */
 u8 drm_dp_dsc_sink_max_slice_count(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE],
 				   bool is_edp)
@@ -1405,6 +1418,21 @@ u8 drm_dp_dsc_sink_max_slice_count(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE],
 }
 EXPORT_SYMBOL(drm_dp_dsc_sink_max_slice_count);
 
+/**
+ * drm_dp_dsc_sink_line_buf_depth() - Get the line buffer depth in bits
+ * @dsc_dpcd: DSC capabilities from DPCD
+ *
+ * Read the DSC DPCD register to parse the line buffer depth in bits which is
+ * number of bits of precision within the decoder line buffer supported by
+ * the DSC sink. This is used to populate the DSC parameters in the
+ * &struct drm_dsc_config by the driver.
+ * Driver creates an infoframe using these parameters to populate
+ * &struct drm_dsc_pps_infoframe. These are sent to the sink using DSC
+ * infoframe using the helper function drm_dsc_pps_infoframe_pack()
+ *
+ * Returns:
+ * Line buffer depth supported by DSC panel or 0 its invalid
+ */
 u8 drm_dp_dsc_sink_line_buf_depth(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE])
 {
 	u8 line_buf_depth = dsc_dpcd[DP_DSC_LINE_BUF_BIT_DEPTH - DP_DSC_SUPPORT];
@@ -1434,6 +1462,23 @@ u8 drm_dp_dsc_sink_line_buf_depth(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE])
 }
 EXPORT_SYMBOL(drm_dp_dsc_sink_line_buf_depth);
 
+/**
+ * drm_dp_dsc_sink_supported_input_bpcs() - Get all the input bits per component
+ * values supported by the DSC sink.
+ * @dsc_dpcd: DSC capabilities from DPCD
+ * @dsc_bpc: An array to be filled by this helper with supported
+ *           input bpcs.
+ *
+ * Read the DSC DPCD from the sink device to parse the supported bits per
+ * component values. This is used to populate the DSC parameters
+ * in the &struct drm_dsc_config by the driver.
+ * Driver creates an infoframe using these parameters to populate
+ * &struct drm_dsc_pps_infoframe. These are sent to the sink using DSC
+ * infoframe using the helper function drm_dsc_pps_infoframe_pack()
+ *
+ * Returns:
+ * Number of input BPC values parsed from the DPCD
+ */
 int drm_dp_dsc_sink_supported_input_bpcs(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE],
 					 u8 dsc_bpc[3])
 {
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index 196ebba8af5f..dc7ac0c60547 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -33,7 +33,7 @@
 #include <drm/drm_fixed.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 /**
  * DOC: dp mst helper
@@ -67,6 +67,64 @@ static bool drm_dp_validate_guid(struct drm_dp_mst_topology_mgr *mgr,
 static int drm_dp_mst_register_i2c_bus(struct drm_dp_aux *aux);
 static void drm_dp_mst_unregister_i2c_bus(struct drm_dp_aux *aux);
 static void drm_dp_mst_kick_tx(struct drm_dp_mst_topology_mgr *mgr);
+
+#define DP_STR(x) [DP_ ## x] = #x
+
+static const char *drm_dp_mst_req_type_str(u8 req_type)
+{
+	static const char * const req_type_str[] = {
+		DP_STR(GET_MSG_TRANSACTION_VERSION),
+		DP_STR(LINK_ADDRESS),
+		DP_STR(CONNECTION_STATUS_NOTIFY),
+		DP_STR(ENUM_PATH_RESOURCES),
+		DP_STR(ALLOCATE_PAYLOAD),
+		DP_STR(QUERY_PAYLOAD),
+		DP_STR(RESOURCE_STATUS_NOTIFY),
+		DP_STR(CLEAR_PAYLOAD_ID_TABLE),
+		DP_STR(REMOTE_DPCD_READ),
+		DP_STR(REMOTE_DPCD_WRITE),
+		DP_STR(REMOTE_I2C_READ),
+		DP_STR(REMOTE_I2C_WRITE),
+		DP_STR(POWER_UP_PHY),
+		DP_STR(POWER_DOWN_PHY),
+		DP_STR(SINK_EVENT_NOTIFY),
+		DP_STR(QUERY_STREAM_ENC_STATUS),
+	};
+
+	if (req_type >= ARRAY_SIZE(req_type_str) ||
+	    !req_type_str[req_type])
+		return "unknown";
+
+	return req_type_str[req_type];
+}
+
+#undef DP_STR
+#define DP_STR(x) [DP_NAK_ ## x] = #x
+
+static const char *drm_dp_mst_nak_reason_str(u8 nak_reason)
+{
+	static const char * const nak_reason_str[] = {
+		DP_STR(WRITE_FAILURE),
+		DP_STR(INVALID_READ),
+		DP_STR(CRC_FAILURE),
+		DP_STR(BAD_PARAM),
+		DP_STR(DEFER),
+		DP_STR(LINK_FAILURE),
+		DP_STR(NO_RESOURCES),
+		DP_STR(DPCD_FAIL),
+		DP_STR(I2C_NAK),
+		DP_STR(ALLOCATE_FAIL),
+	};
+
+	if (nak_reason >= ARRAY_SIZE(nak_reason_str) ||
+	    !nak_reason_str[nak_reason])
+		return "unknown";
+
+	return nak_reason_str[nak_reason];
+}
+
+#undef DP_STR
+
 /* sideband msg handling */
 static u8 drm_dp_msg_header_crc4(const uint8_t *data, size_t num_nibbles)
 {
@@ -568,7 +626,7 @@ static bool drm_dp_sideband_parse_reply(struct drm_dp_sideband_msg_rx *raw,
 	msg->reply_type = (raw->msg[0] & 0x80) >> 7;
 	msg->req_type = (raw->msg[0] & 0x7f);
 
-	if (msg->reply_type) {
+	if (msg->reply_type == DP_SIDEBAND_REPLY_NAK) {
 		memcpy(msg->u.nak.guid, &raw->msg[1], 16);
 		msg->u.nak.reason = raw->msg[17];
 		msg->u.nak.nak_data = raw->msg[18];
@@ -594,7 +652,8 @@ static bool drm_dp_sideband_parse_reply(struct drm_dp_sideband_msg_rx *raw,
 	case DP_POWER_UP_PHY:
 		return drm_dp_sideband_parse_power_updown_phy_ack(raw, msg);
 	default:
-		DRM_ERROR("Got unknown reply 0x%02x\n", msg->req_type);
+		DRM_ERROR("Got unknown reply 0x%02x (%s)\n", msg->req_type,
+			  drm_dp_mst_req_type_str(msg->req_type));
 		return false;
 	}
 }
@@ -661,7 +720,8 @@ static bool drm_dp_sideband_parse_req(struct drm_dp_sideband_msg_rx *raw,
 	case DP_RESOURCE_STATUS_NOTIFY:
 		return drm_dp_sideband_parse_resource_status_notify(raw, msg);
 	default:
-		DRM_ERROR("Got unknown request 0x%02x\n", msg->req_type);
+		DRM_ERROR("Got unknown request 0x%02x (%s)\n", msg->req_type,
+			  drm_dp_mst_req_type_str(msg->req_type));
 		return false;
 	}
 }
@@ -881,7 +941,7 @@ static void drm_dp_free_mst_branch_device(struct kref *kref)
  * in-memory topology state from being changed in the middle of critical
  * operations like changing the internal state of payload allocations. This
  * means each branch and port will be considered to be connected to the rest
- * of the topology until it's topology refcount reaches zero. Additionally,
+ * of the topology until its topology refcount reaches zero. Additionally,
  * for ports this means that their associated &struct drm_connector will stay
  * registered with userspace until the port's refcount reaches 0.
  *
@@ -919,8 +979,8 @@ static void drm_dp_free_mst_branch_device(struct kref *kref)
  *    same way as the C pointers used to reference a structure.
  *
  * As you can see in the above figure, every branch increments the topology
- * refcount of it's children, and increments the malloc refcount of it's
- * parent. Additionally, every payload increments the malloc refcount of it's
+ * refcount of its children, and increments the malloc refcount of its
+ * parent. Additionally, every payload increments the malloc refcount of its
  * assigned port by 1.
  *
  * So, what would happen if MSTB #3 from the above figure was unplugged from
@@ -937,9 +997,9 @@ static void drm_dp_free_mst_branch_device(struct kref *kref)
  * of its parent, and finally its own malloc refcount. For MSTB #4 and port
  * #4, this means they both have been disconnected from the topology and freed
  * from memory. But, because payload #2 is still holding a reference to port
- * #3, port #3 is removed from the topology but it's &struct drm_dp_mst_port
+ * #3, port #3 is removed from the topology but its &struct drm_dp_mst_port
  * is still accessible from memory. This also means port #3 has not yet
- * decremented the malloc refcount of MSTB #3, so it's &struct
+ * decremented the malloc refcount of MSTB #3, so its &struct
  * drm_dp_mst_branch will also stay allocated in memory until port #3's
  * malloc refcount reaches 0.
  *
@@ -1079,7 +1139,7 @@ static void drm_dp_destroy_mst_branch_device(struct kref *kref)
 
 /**
  * drm_dp_mst_topology_try_get_mstb() - Increment the topology refcount of a
- * branch device unless its zero
+ * branch device unless it's zero
  * @mstb: &struct drm_dp_mst_branch to increment the topology refcount of
  *
  * Attempts to grab a topology reference to @mstb, if it hasn't yet been
@@ -1205,7 +1265,7 @@ static void drm_dp_destroy_port(struct kref *kref)
 
 /**
  * drm_dp_mst_topology_try_get_port() - Increment the topology refcount of a
- * port unless its zero
+ * port unless it's zero
  * @port: &struct drm_dp_mst_port to increment the topology refcount of
  *
  * Attempts to grab a topology reference to @port, if it hasn't yet been
@@ -1411,7 +1471,7 @@ static bool drm_dp_port_setup_pdt(struct drm_dp_mst_port *port)
 			port->mstb->port_parent = port;
 			/*
 			 * Make sure this port's memory allocation stays
-			 * around until it's child MSTB releases it
+			 * around until its child MSTB releases it
 			 */
 			drm_dp_mst_get_port_malloc(port);
 
@@ -1969,9 +2029,9 @@ static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr,
 	if (ret > 0) {
 		int i;
 
-		if (txmsg->reply.reply_type == 1)
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) {
 			DRM_DEBUG_KMS("link address nak received\n");
-		else {
+		} else {
 			DRM_DEBUG_KMS("link address reply: %d\n", txmsg->reply.u.link_addr.nports);
 			for (i = 0; i < txmsg->reply.u.link_addr.nports; i++) {
 				DRM_DEBUG_KMS("port %d: input %d, pdt: %d, pn: %d, dpcd_rev: %02x, mcs: %d, ddps: %d, ldps %d, sdp %d/%d\n", i,
@@ -2020,9 +2080,9 @@ static int drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr,
 
 	ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
 	if (ret > 0) {
-		if (txmsg->reply.reply_type == 1)
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) {
 			DRM_DEBUG_KMS("enum path resources nak received\n");
-		else {
+		} else {
 			if (port->port_num != txmsg->reply.u.path_resources.port_number)
 				DRM_ERROR("got incorrect port in response\n");
 			DRM_DEBUG_KMS("enum path resources %d: %d %d\n", txmsg->reply.u.path_resources.port_number, txmsg->reply.u.path_resources.full_payload_bw_number,
@@ -2132,7 +2192,7 @@ static int drm_dp_payload_send_msg(struct drm_dp_mst_topology_mgr *mgr,
 	 */
 	ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
 	if (ret > 0) {
-		if (txmsg->reply.reply_type == 1)
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
 			ret = -EINVAL;
 		else
 			ret = 0;
@@ -2165,7 +2225,7 @@ int drm_dp_send_power_updown_phy(struct drm_dp_mst_topology_mgr *mgr,
 
 	ret = drm_dp_mst_wait_tx_reply(port->parent, txmsg);
 	if (ret > 0) {
-		if (txmsg->reply.reply_type == 1)
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
 			ret = -EINVAL;
 		else
 			ret = 0;
@@ -2211,7 +2271,7 @@ static int drm_dp_destroy_payload_step1(struct drm_dp_mst_topology_mgr *mgr,
 					struct drm_dp_payload *payload)
 {
 	DRM_DEBUG_KMS("\n");
-	/* its okay for these to fail */
+	/* it's okay for these to fail */
 	if (port) {
 		drm_dp_payload_send_msg(mgr, port, id, 0);
 	}
@@ -2423,9 +2483,9 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr,
 
 	ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
 	if (ret > 0) {
-		if (txmsg->reply.reply_type == 1) {
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
 			ret = -EINVAL;
-		} else
+		else
 			ret = 0;
 	}
 	kfree(txmsg);
@@ -2438,7 +2498,7 @@ static int drm_dp_encode_up_ack_reply(struct drm_dp_sideband_msg_tx *msg, u8 req
 {
 	struct drm_dp_sideband_msg_reply_body reply;
 
-	reply.reply_type = 0;
+	reply.reply_type = DP_SIDEBAND_REPLY_ACK;
 	reply.req_type = req_type;
 	drm_dp_encode_sideband_reply(&reply, msg);
 	return 0;
@@ -2745,9 +2805,14 @@ static int drm_dp_mst_handle_down_rep(struct drm_dp_mst_topology_mgr *mgr)
 		}
 
 		drm_dp_sideband_parse_reply(&mgr->down_rep_recv, &txmsg->reply);
-		if (txmsg->reply.reply_type == 1) {
-			DRM_DEBUG_KMS("Got NAK reply: req 0x%02x, reason 0x%02x, nak data 0x%02x\n", txmsg->reply.req_type, txmsg->reply.u.nak.reason, txmsg->reply.u.nak.nak_data);
-		}
+
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK)
+			DRM_DEBUG_KMS("Got NAK reply: req 0x%02x (%s), reason 0x%02x (%s), nak data 0x%02x\n",
+				      txmsg->reply.req_type,
+				      drm_dp_mst_req_type_str(txmsg->reply.req_type),
+				      txmsg->reply.u.nak.reason,
+				      drm_dp_mst_nak_reason_str(txmsg->reply.u.nak.reason),
+				      txmsg->reply.u.nak.nak_data);
 
 		memset(&mgr->down_rep_recv, 0, sizeof(struct drm_dp_sideband_msg_rx));
 		drm_dp_mst_topology_put_mstb(mstb);
@@ -2882,7 +2947,7 @@ enum drm_connector_status drm_dp_mst_detect_port(struct drm_connector *connector
 {
 	enum drm_connector_status status = connector_status_disconnected;
 
-	/* we need to search for the port in the mgr in case its gone */
+	/* we need to search for the port in the mgr in case it's gone */
 	port = drm_dp_mst_topology_get_port_validated(mgr, port);
 	if (!port)
 		return connector_status_disconnected;
@@ -2948,7 +3013,7 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_
 {
 	struct edid *edid = NULL;
 
-	/* we need to search for the port in the mgr in case its gone */
+	/* we need to search for the port in the mgr in case it's gone */
 	port = drm_dp_mst_topology_get_port_validated(mgr, port);
 	if (!port)
 		return NULL;
@@ -3052,10 +3117,6 @@ int drm_dp_atomic_find_vcpi_slots(struct drm_atomic_state *state,
 	if (IS_ERR(topology_state))
 		return PTR_ERR(topology_state);
 
-	port = drm_dp_mst_topology_get_port_validated(mgr, port);
-	if (port == NULL)
-		return -EINVAL;
-
 	/* Find the current allocation for this port, if any */
 	list_for_each_entry(pos, &topology_state->vcpis, next) {
 		if (pos->port == port) {
@@ -3088,10 +3149,8 @@ int drm_dp_atomic_find_vcpi_slots(struct drm_atomic_state *state,
 	/* Add the new allocation to the state */
 	if (!vcpi) {
 		vcpi = kzalloc(sizeof(*vcpi), GFP_KERNEL);
-		if (!vcpi) {
-			ret = -ENOMEM;
-			goto out;
-		}
+		if (!vcpi)
+			return -ENOMEM;
 
 		drm_dp_mst_get_port_malloc(port);
 		vcpi->port = port;
@@ -3100,8 +3159,6 @@ int drm_dp_atomic_find_vcpi_slots(struct drm_atomic_state *state,
 	vcpi->vcpi = req_slots;
 
 	ret = req_slots;
-out:
-	drm_dp_mst_topology_put_port(port);
 	return ret;
 }
 EXPORT_SYMBOL(drm_dp_atomic_find_vcpi_slots);
@@ -3115,7 +3172,7 @@ EXPORT_SYMBOL(drm_dp_atomic_find_vcpi_slots);
  * Releases any VCPI slots that have been allocated to a port in the atomic
  * state. Any atomic drivers which support MST must call this function in
  * their &drm_connector_helper_funcs.atomic_check() callback when the
- * connector will no longer have VCPI allocated (e.g. because it's CRTC was
+ * connector will no longer have VCPI allocated (e.g. because its CRTC was
  * removed) when it had VCPI allocated in the previous atomic state.
  *
  * It is OK to call this even if @port has been removed from the system.
@@ -3203,7 +3260,7 @@ bool drm_dp_mst_allocate_vcpi(struct drm_dp_mst_topology_mgr *mgr,
 	DRM_DEBUG_KMS("initing vcpi for pbn=%d slots=%d\n",
 		      pbn, port->vcpi.num_slots);
 
-	/* Keep port allocated until it's payload has been removed */
+	/* Keep port allocated until its payload has been removed */
 	drm_dp_mst_get_port_malloc(port);
 	drm_dp_mst_topology_put_port(port);
 	return true;
@@ -3235,7 +3292,7 @@ EXPORT_SYMBOL(drm_dp_mst_get_vcpi_slots);
 void drm_dp_mst_reset_vcpi_slots(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port)
 {
 	/*
-	 * A port with VCPI will remain allocated until it's VCPI is
+	 * A port with VCPI will remain allocated until its VCPI is
 	 * released, no verified ref needed
 	 */
 
@@ -3246,15 +3303,16 @@ EXPORT_SYMBOL(drm_dp_mst_reset_vcpi_slots);
 /**
  * drm_dp_mst_deallocate_vcpi() - deallocate a VCPI
  * @mgr: manager for this port
- * @port: unverified port to deallocate vcpi for
+ * @port: port to deallocate vcpi for
+ *
+ * This can be called unconditionally, regardless of whether
+ * drm_dp_mst_allocate_vcpi() succeeded or not.
  */
 void drm_dp_mst_deallocate_vcpi(struct drm_dp_mst_topology_mgr *mgr,
 				struct drm_dp_mst_port *port)
 {
-	/*
-	 * A port with VCPI will remain allocated until it's VCPI is
-	 * released, no verified ref needed
-	 */
+	if (!port->vcpi.vcpi)
+		return;
 
 	drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi);
 	port->vcpi.num_slots = 0;
@@ -3893,7 +3951,7 @@ static int drm_dp_mst_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs
 	ret = drm_dp_mst_wait_tx_reply(mstb, txmsg);
 	if (ret > 0) {
 
-		if (txmsg->reply.reply_type == 1) { /* got a NAK back */
+		if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) {
 			ret = -EREMOTEIO;
 			goto out;
 		}
diff --git a/drivers/gpu/drm/drm_dsc.c b/drivers/gpu/drm/drm_dsc.c
index bc2b23adb072..bce99f95c1a3 100644
--- a/drivers/gpu/drm/drm_dsc.c
+++ b/drivers/gpu/drm/drm_dsc.c
@@ -17,6 +17,12 @@
 /**
  * DOC: dsc helpers
  *
+ * VESA specification for DP 1.4 adds a new feature called Display Stream
+ * Compression (DSC) used to compress the pixel bits before sending it on
+ * DP/eDP/MIPI DSI interface. DSC is required to be enabled so that the existing
+ * display interfaces can support high resolutions at higher frames rates uisng
+ * the maximum available link capacity of these interfaces.
+ *
  * These functions contain some common logic and helpers to deal with VESA
  * Display Stream Compression standard required for DSC on Display Port/eDP or
  * MIPI display interfaces.
@@ -26,6 +32,13 @@
  * drm_dsc_dp_pps_header_init() - Initializes the PPS Header
  * for DisplayPort as per the DP 1.4 spec.
  * @pps_sdp: Secondary data packet for DSC Picture Parameter Set
+ *           as defined in &struct drm_dsc_pps_infoframe
+ *
+ * DP 1.4 spec defines the secondary data packet for sending the
+ * picture parameter infoframes from the source to the sink.
+ * This function populates the pps header defined in
+ * &struct drm_dsc_pps_infoframe as per the header bytes defined
+ * in &struct dp_sdp_header.
  */
 void drm_dsc_dp_pps_header_init(struct drm_dsc_pps_infoframe *pps_sdp)
 {
@@ -38,15 +51,20 @@ EXPORT_SYMBOL(drm_dsc_dp_pps_header_init);
 
 /**
  * drm_dsc_pps_infoframe_pack() - Populates the DSC PPS infoframe
- * using the DSC configuration parameters in the order expected
- * by the DSC Display Sink device. For the DSC, the sink device
- * expects the PPS payload in the big endian format for the fields
- * that span more than 1 byte.
  *
  * @pps_sdp:
- * Secondary data packet for DSC Picture Parameter Set
+ * Secondary data packet for DSC Picture Parameter Set. This is defined
+ * by &struct drm_dsc_pps_infoframe
  * @dsc_cfg:
- * DSC Configuration data filled by driver
+ * DSC Configuration data filled by driver as defined by
+ * &struct drm_dsc_config
+ *
+ * DSC source device sends a secondary data packet filled with all the
+ * picture parameter set (PPS) information required by the sink to decode
+ * the compressed frame. Driver populates the dsC PPS infoframe using the DSC
+ * configuration parameters in the order expected by the DSC Display Sink
+ * device. For the DSC, the sink device expects the PPS payload in the big
+ * endian format for the fields that span more than 1 byte.
  */
 void drm_dsc_pps_infoframe_pack(struct drm_dsc_pps_infoframe *pps_sdp,
 				const struct drm_dsc_config *dsc_cfg)
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 31fcf94bf825..0e9349ff2d16 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -1621,6 +1621,64 @@ static bool drm_fb_pixel_format_equal(const struct fb_var_screeninfo *var_1,
 	       var_1->transp.msb_right == var_2->transp.msb_right;
 }
 
+static void drm_fb_helper_fill_pixel_fmt(struct fb_var_screeninfo *var,
+					 u8 depth)
+{
+	switch (depth) {
+	case 8:
+		var->red.offset = 0;
+		var->green.offset = 0;
+		var->blue.offset = 0;
+		var->red.length = 8; /* 8bit DAC */
+		var->green.length = 8;
+		var->blue.length = 8;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+	case 15:
+		var->red.offset = 10;
+		var->green.offset = 5;
+		var->blue.offset = 0;
+		var->red.length = 5;
+		var->green.length = 5;
+		var->blue.length = 5;
+		var->transp.offset = 15;
+		var->transp.length = 1;
+		break;
+	case 16:
+		var->red.offset = 11;
+		var->green.offset = 5;
+		var->blue.offset = 0;
+		var->red.length = 5;
+		var->green.length = 6;
+		var->blue.length = 5;
+		var->transp.offset = 0;
+		break;
+	case 24:
+		var->red.offset = 16;
+		var->green.offset = 8;
+		var->blue.offset = 0;
+		var->red.length = 8;
+		var->green.length = 8;
+		var->blue.length = 8;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+	case 32:
+		var->red.offset = 16;
+		var->green.offset = 8;
+		var->blue.offset = 0;
+		var->red.length = 8;
+		var->green.length = 8;
+		var->blue.length = 8;
+		var->transp.offset = 24;
+		var->transp.length = 8;
+		break;
+	default:
+		break;
+	}
+}
+
 /**
  * drm_fb_helper_check_var - implementation for &fb_ops.fb_check_var
  * @var: screeninfo to check
@@ -1632,9 +1690,14 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 	struct drm_fb_helper *fb_helper = info->par;
 	struct drm_framebuffer *fb = fb_helper->fb;
 
-	if (var->pixclock != 0 || in_dbg_master())
+	if (in_dbg_master())
 		return -EINVAL;
 
+	if (var->pixclock != 0) {
+		DRM_DEBUG("fbdev emulation doesn't support changing the pixel clock, value of pixclock is ignored\n");
+		var->pixclock = 0;
+	}
+
 	if ((drm_format_info_block_width(fb->format, 0) > 1) ||
 	    (drm_format_info_block_height(fb->format, 0) > 1))
 		return -EINVAL;
@@ -1655,6 +1718,20 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 	}
 
 	/*
+	 * Workaround for SDL 1.2, which is known to be setting all pixel format
+	 * fields values to zero in some cases. We treat this situation as a
+	 * kind of "use some reasonable autodetected values".
+	 */
+	if (!var->red.offset     && !var->green.offset    &&
+	    !var->blue.offset    && !var->transp.offset   &&
+	    !var->red.length     && !var->green.length    &&
+	    !var->blue.length    && !var->transp.length   &&
+	    !var->red.msb_right  && !var->green.msb_right &&
+	    !var->blue.msb_right && !var->transp.msb_right) {
+		drm_fb_helper_fill_pixel_fmt(var, fb->format->depth);
+	}
+
+	/*
 	 * drm fbdev emulation doesn't support changing the pixel format at all,
 	 * so reject all pixel format changing requests.
 	 */
@@ -2020,59 +2097,7 @@ void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helpe
 	info->var.yoffset = 0;
 	info->var.activate = FB_ACTIVATE_NOW;
 
-	switch (fb->format->depth) {
-	case 8:
-		info->var.red.offset = 0;
-		info->var.green.offset = 0;
-		info->var.blue.offset = 0;
-		info->var.red.length = 8; /* 8bit DAC */
-		info->var.green.length = 8;
-		info->var.blue.length = 8;
-		info->var.transp.offset = 0;
-		info->var.transp.length = 0;
-		break;
-	case 15:
-		info->var.red.offset = 10;
-		info->var.green.offset = 5;
-		info->var.blue.offset = 0;
-		info->var.red.length = 5;
-		info->var.green.length = 5;
-		info->var.blue.length = 5;
-		info->var.transp.offset = 15;
-		info->var.transp.length = 1;
-		break;
-	case 16:
-		info->var.red.offset = 11;
-		info->var.green.offset = 5;
-		info->var.blue.offset = 0;
-		info->var.red.length = 5;
-		info->var.green.length = 6;
-		info->var.blue.length = 5;
-		info->var.transp.offset = 0;
-		break;
-	case 24:
-		info->var.red.offset = 16;
-		info->var.green.offset = 8;
-		info->var.blue.offset = 0;
-		info->var.red.length = 8;
-		info->var.green.length = 8;
-		info->var.blue.length = 8;
-		info->var.transp.offset = 0;
-		info->var.transp.length = 0;
-		break;
-	case 32:
-		info->var.red.offset = 16;
-		info->var.green.offset = 8;
-		info->var.blue.offset = 0;
-		info->var.red.length = 8;
-		info->var.green.length = 8;
-		info->var.blue.length = 8;
-		info->var.transp.offset = 24;
-		info->var.transp.length = 8;
-		break;
-	default:
-		break;
-	}
+	drm_fb_helper_fill_pixel_fmt(&info->var, fb->format->depth);
 
 	info->var.xres = fb_width;
 	info->var.yres = fb_height;
@@ -2483,7 +2508,7 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper,
 /*
  * This function checks if rotation is necessary because of panel orientation
  * and if it is, if it is supported.
- * If rotation is necessary and supported, its gets set in fb_crtc.rotation.
+ * If rotation is necessary and supported, it gets set in fb_crtc.rotation.
  * If rotation is necessary but not supported, a DRM_MODE_ROTATE_* flag gets
  * or-ed into fb_helper->sw_rotations. In drm_setup_crtcs_fb() we check if only
  * one bit is set and then we set fb_info.fbcon_rotate_hint to make fbcon do
@@ -3185,9 +3210,7 @@ static void drm_fbdev_client_unregister(struct drm_client_dev *client)
 
 static int drm_fbdev_client_restore(struct drm_client_dev *client)
 {
-	struct drm_fb_helper *fb_helper = drm_fb_helper_from_client(client);
-
-	drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper);
+	drm_fb_helper_lastclose(client->dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index 46f48f245eb5..b1838a41ad43 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -701,7 +701,7 @@ int drm_event_reserve_init(struct drm_device *dev,
 EXPORT_SYMBOL(drm_event_reserve_init);
 
 /**
- * drm_event_cancel_free - free a DRM event and release it's space
+ * drm_event_cancel_free - free a DRM event and release its space
  * @dev: DRM device
  * @p: tracking structure for the pending event
  *
diff --git a/drivers/gpu/drm/drm_fourcc.c b/drivers/gpu/drm/drm_fourcc.c
index d90ee03a84c6..ba7e19d4336c 100644
--- a/drivers/gpu/drm/drm_fourcc.c
+++ b/drivers/gpu/drm/drm_fourcc.c
@@ -238,6 +238,15 @@ const struct drm_format_info *__drm_format_info(u32 format)
 		{ .format = DRM_FORMAT_X0L2,		.depth = 0,  .num_planes = 1,
 		  .char_per_block = { 8, 0, 0 }, .block_w = { 2, 0, 0 }, .block_h = { 2, 0, 0 },
 		  .hsub = 2, .vsub = 2, .is_yuv = true },
+		{ .format = DRM_FORMAT_P010,            .depth = 0,  .num_planes = 2,
+		  .char_per_block = { 2, 4, 0 }, .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 },
+		  .hsub = 2, .vsub = 2, .is_yuv = true},
+		{ .format = DRM_FORMAT_P012,		.depth = 0,  .num_planes = 2,
+		  .char_per_block = { 2, 4, 0 }, .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 },
+		   .hsub = 2, .vsub = 2, .is_yuv = true},
+		{ .format = DRM_FORMAT_P016,		.depth = 0,  .num_planes = 2,
+		  .char_per_block = { 2, 4, 0 }, .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 },
+		  .hsub = 2, .vsub = 2, .is_yuv = true},
 	};
 
 	unsigned int i;
diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
index 7abcb265a108..d8d75e25f6fb 100644
--- a/drivers/gpu/drm/drm_framebuffer.c
+++ b/drivers/gpu/drm/drm_framebuffer.c
@@ -773,7 +773,7 @@ EXPORT_SYMBOL(drm_framebuffer_lookup);
  * @fb: fb to unregister
  *
  * Drivers need to call this when cleaning up driver-private framebuffers, e.g.
- * those used for fbdev. Note that the caller must hold a reference of it's own,
+ * those used for fbdev. Note that the caller must hold a reference of its own,
  * i.e. the object may not be destroyed through this call (since it'll lead to a
  * locking inversion).
  *
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 2896ff60552f..d0b9f6a9953f 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -660,7 +660,7 @@ void drm_gem_put_pages(struct drm_gem_object *obj, struct page **pages,
 EXPORT_SYMBOL(drm_gem_put_pages);
 
 /**
- * drm_gem_object_lookup - look up a GEM object from it's handle
+ * drm_gem_object_lookup - look up a GEM object from its handle
  * @filp: DRM file private date
  * @handle: userspace handle
  *
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 45a07652fa00..9bd8908d5fd8 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -103,9 +103,6 @@ int drm_irq_install(struct drm_device *dev, int irq)
 	int ret;
 	unsigned long sh_flags = 0;
 
-	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		return -EOPNOTSUPP;
-
 	if (irq == 0)
 		return -EINVAL;
 
@@ -123,8 +120,8 @@ int drm_irq_install(struct drm_device *dev, int irq)
 	if (dev->driver->irq_preinstall)
 		dev->driver->irq_preinstall(dev);
 
-	/* Install handler */
-	if (drm_core_check_feature(dev, DRIVER_IRQ_SHARED))
+	/* PCI devices require shared interrupts. */
+	if (dev->pdev)
 		sh_flags = IRQF_SHARED;
 
 	ret = request_irq(irq, dev->driver->irq_handler,
@@ -174,9 +171,6 @@ int drm_irq_uninstall(struct drm_device *dev)
 	bool irq_enabled;
 	int i;
 
-	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		return -EOPNOTSUPP;
-
 	irq_enabled = dev->irq_enabled;
 	dev->irq_enabled = false;
 
diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c
index b735704653cb..603b0bd9c5ce 100644
--- a/drivers/gpu/drm/drm_lease.c
+++ b/drivers/gpu/drm/drm_lease.c
@@ -528,7 +528,8 @@ int drm_mode_create_lease_ioctl(struct drm_device *dev,
 
 	object_count = cl->object_count;
 
-	object_ids = memdup_user(u64_to_user_ptr(cl->object_ids), object_count * sizeof(__u32));
+	object_ids = memdup_user(u64_to_user_ptr(cl->object_ids),
+			array_size(object_count, sizeof(__u32)));
 	if (IS_ERR(object_ids))
 		return PTR_ERR(object_ids);
 
diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c
index 3cc5fbd78ee2..2b4f373736c7 100644
--- a/drivers/gpu/drm/drm_mm.c
+++ b/drivers/gpu/drm/drm_mm.c
@@ -816,7 +816,7 @@ EXPORT_SYMBOL(drm_mm_scan_add_block);
  * When the scan list is empty, the selected memory nodes can be freed. An
  * immediately following drm_mm_insert_node_in_range_generic() or one of the
  * simpler versions of that function with !DRM_MM_SEARCH_BEST will then return
- * the just freed block (because its at the top of the free_stack list).
+ * the just freed block (because it's at the top of the free_stack list).
  *
  * Returns:
  * True if this block should be evicted, false otherwise. Will always
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index adce9a26bac9..869ac6f4671e 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -751,7 +751,7 @@ int drm_mode_hsync(const struct drm_display_mode *mode)
 	if (mode->hsync)
 		return mode->hsync;
 
-	if (mode->htotal < 0)
+	if (mode->htotal <= 0)
 		return 0;
 
 	calc_val = (mode->clock * 1000) / mode->htotal; /* hsync in Hz */
@@ -1272,7 +1272,7 @@ const char *drm_get_mode_status_name(enum drm_mode_status status)
  * @verbose: be verbose about it
  *
  * This helper function can be used to prune a display mode list after
- * validation has been completed. All modes who's status is not MODE_OK will be
+ * validation has been completed. All modes whose status is not MODE_OK will be
  * removed from the list, and if @verbose the status code and mode name is also
  * printed to dmesg.
  */
diff --git a/drivers/gpu/drm/drm_modeset_helper.c b/drivers/gpu/drm/drm_modeset_helper.c
index 9150fa385bba..da483125e063 100644
--- a/drivers/gpu/drm/drm_modeset_helper.c
+++ b/drivers/gpu/drm/drm_modeset_helper.c
@@ -21,10 +21,12 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
 
 /**
  * DOC: aux kms helpers
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index a9d9df6c85ad..693748ad8b88 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -61,8 +61,9 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali
 		return NULL;
 
 	dmah->size = size;
-	dmah->vaddr = dma_zalloc_coherent(&dev->pdev->dev, size, &dmah->busaddr,
-						GFP_KERNEL | __GFP_COMP);
+	dmah->vaddr = dma_alloc_coherent(&dev->pdev->dev, size,
+					 &dmah->busaddr,
+					 GFP_KERNEL | __GFP_COMP);
 
 	if (dmah->vaddr == NULL) {
 		kfree(dmah);
diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c
index a1bb157bfdfa..6fd08e04b323 100644
--- a/drivers/gpu/drm/drm_probe_helper.c
+++ b/drivers/gpu/drm/drm_probe_helper.c
@@ -36,10 +36,10 @@
 #include <drm/drm_client.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_fourcc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_probe_helper.h>
 
 #include "drm_crtc_helper_internal.h"
 
diff --git a/drivers/gpu/drm/drm_property.c b/drivers/gpu/drm/drm_property.c
index 79c77c3cad86..f8ec8f9c3e7a 100644
--- a/drivers/gpu/drm/drm_property.c
+++ b/drivers/gpu/drm/drm_property.c
@@ -866,7 +866,7 @@ err:
  * value doesn't become invalid part way through the property update due to
  * race).  The value returned by reference via 'obj' should be passed back
  * to drm_property_change_valid_put() after the property is set (and the
- * object to which the property is attached has a chance to take it's own
+ * object to which the property is attached has a chance to take its own
  * reference).
  */
 bool drm_property_change_valid_get(struct drm_property *property,
diff --git a/drivers/gpu/drm/drm_rect.c b/drivers/gpu/drm/drm_rect.c
index 8c057829b804..66c41b12719c 100644
--- a/drivers/gpu/drm/drm_rect.c
+++ b/drivers/gpu/drm/drm_rect.c
@@ -208,114 +208,6 @@ int drm_rect_calc_vscale(const struct drm_rect *src,
 EXPORT_SYMBOL(drm_rect_calc_vscale);
 
 /**
- * drm_calc_hscale_relaxed - calculate the horizontal scaling factor
- * @src: source window rectangle
- * @dst: destination window rectangle
- * @min_hscale: minimum allowed horizontal scaling factor
- * @max_hscale: maximum allowed horizontal scaling factor
- *
- * Calculate the horizontal scaling factor as
- * (@src width) / (@dst width).
- *
- * If the calculated scaling factor is below @min_vscale,
- * decrease the height of rectangle @dst to compensate.
- *
- * If the calculated scaling factor is above @max_vscale,
- * decrease the height of rectangle @src to compensate.
- *
- * If the scale is below 1 << 16, round down. If the scale is above
- * 1 << 16, round up. This will calculate the scale with the most
- * pessimistic limit calculation.
- *
- * RETURNS:
- * The horizontal scaling factor.
- */
-int drm_rect_calc_hscale_relaxed(struct drm_rect *src,
-				 struct drm_rect *dst,
-				 int min_hscale, int max_hscale)
-{
-	int src_w = drm_rect_width(src);
-	int dst_w = drm_rect_width(dst);
-	int hscale = drm_calc_scale(src_w, dst_w);
-
-	if (hscale < 0 || dst_w == 0)
-		return hscale;
-
-	if (hscale < min_hscale) {
-		int max_dst_w = src_w / min_hscale;
-
-		drm_rect_adjust_size(dst, max_dst_w - dst_w, 0);
-
-		return min_hscale;
-	}
-
-	if (hscale > max_hscale) {
-		int max_src_w = dst_w * max_hscale;
-
-		drm_rect_adjust_size(src, max_src_w - src_w, 0);
-
-		return max_hscale;
-	}
-
-	return hscale;
-}
-EXPORT_SYMBOL(drm_rect_calc_hscale_relaxed);
-
-/**
- * drm_rect_calc_vscale_relaxed - calculate the vertical scaling factor
- * @src: source window rectangle
- * @dst: destination window rectangle
- * @min_vscale: minimum allowed vertical scaling factor
- * @max_vscale: maximum allowed vertical scaling factor
- *
- * Calculate the vertical scaling factor as
- * (@src height) / (@dst height).
- *
- * If the calculated scaling factor is below @min_vscale,
- * decrease the height of rectangle @dst to compensate.
- *
- * If the calculated scaling factor is above @max_vscale,
- * decrease the height of rectangle @src to compensate.
- *
- * If the scale is below 1 << 16, round down. If the scale is above
- * 1 << 16, round up. This will calculate the scale with the most
- * pessimistic limit calculation.
- *
- * RETURNS:
- * The vertical scaling factor.
- */
-int drm_rect_calc_vscale_relaxed(struct drm_rect *src,
-				 struct drm_rect *dst,
-				 int min_vscale, int max_vscale)
-{
-	int src_h = drm_rect_height(src);
-	int dst_h = drm_rect_height(dst);
-	int vscale = drm_calc_scale(src_h, dst_h);
-
-	if (vscale < 0 || dst_h == 0)
-		return vscale;
-
-	if (vscale < min_vscale) {
-		int max_dst_h = src_h / min_vscale;
-
-		drm_rect_adjust_size(dst, 0, max_dst_h - dst_h);
-
-		return min_vscale;
-	}
-
-	if (vscale > max_vscale) {
-		int max_src_h = dst_h * max_vscale;
-
-		drm_rect_adjust_size(src, 0, max_src_h - src_h);
-
-		return max_vscale;
-	}
-
-	return vscale;
-}
-EXPORT_SYMBOL(drm_rect_calc_vscale_relaxed);
-
-/**
  * drm_rect_debug_print - print the rectangle information
  * @prefix: prefix string
  * @r: rectangle to print
diff --git a/drivers/gpu/drm/drm_simple_kms_helper.c b/drivers/gpu/drm/drm_simple_kms_helper.c
index 917812448d1b..a32f14cd7398 100644
--- a/drivers/gpu/drm/drm_simple_kms_helper.c
+++ b/drivers/gpu/drm/drm_simple_kms_helper.c
@@ -10,8 +10,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_simple_kms_helper.h>
 #include <linux/slab.h>
 
diff --git a/drivers/gpu/drm/drm_vblank.c b/drivers/gpu/drm/drm_vblank.c
index cde71ee95a8f..a1b65d26d761 100644
--- a/drivers/gpu/drm/drm_vblank.c
+++ b/drivers/gpu/drm/drm_vblank.c
@@ -48,7 +48,7 @@
  * Drivers must initialize the vertical blanking handling core with a call to
  * drm_vblank_init(). Minimally, a driver needs to implement
  * &drm_crtc_funcs.enable_vblank and &drm_crtc_funcs.disable_vblank plus call
- * drm_crtc_handle_vblank() in it's vblank interrupt handler for working vblank
+ * drm_crtc_handle_vblank() in its vblank interrupt handler for working vblank
  * support.
  *
  * Vertical blanking interrupts can be enabled by the DRM core or by drivers
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
index 4bf698de5996..a6a7ded37ef1 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
@@ -21,7 +21,6 @@
 #include <linux/mm_types.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem.h>
 #include <drm/etnaviv_drm.h>
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
index 49a6763693f1..67ae26602024 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
@@ -109,16 +109,19 @@ static void etnaviv_sched_timedout_job(struct drm_sched_job *sched_job)
 	}
 
 	/* block scheduler */
-	kthread_park(gpu->sched.thread);
-	drm_sched_hw_job_reset(&gpu->sched, sched_job);
+	drm_sched_stop(&gpu->sched);
+
+	if(sched_job)
+		drm_sched_increase_karma(sched_job);
 
 	/* get the GPU back into the init state */
 	etnaviv_core_dump(gpu);
 	etnaviv_gpu_recover_hang(gpu);
 
+	drm_sched_resubmit_jobs(&gpu->sched);
+
 	/* restart scheduler after GPU is usable again */
-	drm_sched_job_recovery(&gpu->sched);
-	kthread_unpark(gpu->sched.thread);
+	drm_sched_start(&gpu->sched, true);
 }
 
 static void etnaviv_sched_free_job(struct drm_sched_job *sched_job)
diff --git a/drivers/gpu/drm/exynos/exynos_dp.c b/drivers/gpu/drm/exynos/exynos_dp.c
index c8449ae4f4fe..471242a5e580 100644
--- a/drivers/gpu/drm/exynos/exynos_dp.c
+++ b/drivers/gpu/drm/exynos/exynos_dp.c
@@ -22,10 +22,11 @@
 #include <video/videomode.h>
 
 #include <drm/drmP.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include <drm/bridge/analogix_dp.h>
 #include <drm/exynos_drm.h>
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 2696289ecc78..96ee83a798c4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -13,10 +13,10 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_encoder.h>
+#include <drm/drm_probe_helper.h>
 
 #include "exynos_drm_crtc.h"
 #include "exynos_drm_drv.h"
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 2f0babb67c51..ae425c9a3f7b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -11,9 +11,9 @@
 */
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_panel.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/of_graph.h>
 #include <linux/regulator/consumer.h>
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 2c75e789b2a7..e1ef9dc9ebf3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -15,8 +15,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/component.h>
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index d81e62ae286a..a4253dd55f86 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -13,11 +13,11 @@
 #include <asm/unaligned.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
-#include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/clk.h>
 #include <linux/gpio/consumer.h>
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 31eb538a44ae..1f11ab0f8e9d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -13,12 +13,12 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <uapi/drm/exynos_drm.h>
 
 #include "exynos_drm_drv.h"
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index ce9604ca8041..c30dd88cdb25 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -15,7 +15,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/exynos_drm.h>
 
 #include <linux/console.h>
@@ -88,7 +88,6 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	}
 
 	fbi->par = helper;
-	fbi->flags = FBINFO_FLAG_DEFAULT;
 	fbi->fbops = &exynos_drm_fb_ops;
 
 	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
index 8d67b2a54be3..05abfed6f7f8 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
@@ -356,6 +356,11 @@ static int rotator_runtime_resume(struct device *dev)
 }
 #endif
 
+static const struct drm_exynos_ipp_limit rotator_s5pv210_rbg888_limits[] = {
+	{ IPP_SIZE_LIMIT(BUFFER, .h = { 8, SZ_16K }, .v = { 8, SZ_16K }) },
+	{ IPP_SIZE_LIMIT(AREA, .h.align = 2, .v.align = 2) },
+};
+
 static const struct drm_exynos_ipp_limit rotator_4210_rbg888_limits[] = {
 	{ IPP_SIZE_LIMIT(BUFFER, .h = { 8, SZ_16K }, .v = { 8, SZ_16K }) },
 	{ IPP_SIZE_LIMIT(AREA, .h.align = 4, .v.align = 4) },
@@ -371,6 +376,11 @@ static const struct drm_exynos_ipp_limit rotator_5250_rbg888_limits[] = {
 	{ IPP_SIZE_LIMIT(AREA, .h.align = 2, .v.align = 2) },
 };
 
+static const struct drm_exynos_ipp_limit rotator_s5pv210_yuv_limits[] = {
+	{ IPP_SIZE_LIMIT(BUFFER, .h = { 32, SZ_64K }, .v = { 32, SZ_64K }) },
+	{ IPP_SIZE_LIMIT(AREA, .h.align = 8, .v.align = 8) },
+};
+
 static const struct drm_exynos_ipp_limit rotator_4210_yuv_limits[] = {
 	{ IPP_SIZE_LIMIT(BUFFER, .h = { 32, SZ_64K }, .v = { 32, SZ_64K }) },
 	{ IPP_SIZE_LIMIT(AREA, .h.align = 8, .v.align = 8) },
@@ -381,6 +391,11 @@ static const struct drm_exynos_ipp_limit rotator_4412_yuv_limits[] = {
 	{ IPP_SIZE_LIMIT(AREA, .h.align = 8, .v.align = 8) },
 };
 
+static const struct exynos_drm_ipp_formats rotator_s5pv210_formats[] = {
+	{ IPP_SRCDST_FORMAT(XRGB8888, rotator_s5pv210_rbg888_limits) },
+	{ IPP_SRCDST_FORMAT(NV12, rotator_s5pv210_yuv_limits) },
+};
+
 static const struct exynos_drm_ipp_formats rotator_4210_formats[] = {
 	{ IPP_SRCDST_FORMAT(XRGB8888, rotator_4210_rbg888_limits) },
 	{ IPP_SRCDST_FORMAT(NV12, rotator_4210_yuv_limits) },
@@ -396,6 +411,11 @@ static const struct exynos_drm_ipp_formats rotator_5250_formats[] = {
 	{ IPP_SRCDST_FORMAT(NV12, rotator_4412_yuv_limits) },
 };
 
+static const struct rot_variant rotator_s5pv210_data = {
+	.formats = rotator_s5pv210_formats,
+	.num_formats = ARRAY_SIZE(rotator_s5pv210_formats),
+};
+
 static const struct rot_variant rotator_4210_data = {
 	.formats = rotator_4210_formats,
 	.num_formats = ARRAY_SIZE(rotator_4210_formats),
@@ -413,6 +433,9 @@ static const struct rot_variant rotator_5250_data = {
 
 static const struct of_device_id exynos_rotator_match[] = {
 	{
+		.compatible = "samsung,s5pv210-rotator",
+		.data = &rotator_s5pv210_data,
+	}, {
 		.compatible = "samsung,exynos4210-rotator",
 		.data = &rotator_4210_data,
 	}, {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
index 71270efa64f3..ed1dd1aec902 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
@@ -1,7 +1,7 @@
 /*
  * Copyright (C) 2017 Samsung Electronics Co.Ltd
  * Author:
- *	Andrzej Pietrasiewicz <andrzej.p@samsung.com>
+ *	Andrzej Pietrasiewicz <andrzejtp2010@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 19697c1362d8..29f4c1932aed 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -19,9 +19,9 @@
 
 #include <drm/exynos_drm.h>
 
-#include <drm/drm_edid.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include "exynos_drm_drv.h"
 #include "exynos_drm_crtc.h"
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index b857df67aff0..8e2c02fc66e8 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -15,9 +15,9 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_edid.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include "regs-hdmi.h"
 
diff --git a/drivers/gpu/drm/exynos/regs-scaler.h b/drivers/gpu/drm/exynos/regs-scaler.h
index fc7ccad75e74..512a2baced11 100644
--- a/drivers/gpu/drm/exynos/regs-scaler.h
+++ b/drivers/gpu/drm/exynos/regs-scaler.h
@@ -2,7 +2,7 @@
  *
  * Copyright (c) 2017 Samsung Electronics Co., Ltd.
  *		http://www.samsung.com/
- * Author: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
+ * Author: Andrzej Pietrasiewicz <andrzejtp2010@gmail.com>
  *
  * Register definition file for Samsung scaler driver
  *
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
index 18afc94e4dff..bf256971063d 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
@@ -16,7 +16,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <video/videomode.h>
 
 #include "fsl_dcu_drm_crtc.h"
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
index ceddc3e29258..dfc73aade325 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
@@ -24,11 +24,11 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_modeset_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "fsl_dcu_drm_crtc.h"
 #include "fsl_dcu_drm_drv.h"
@@ -137,7 +137,7 @@ static irqreturn_t fsl_dcu_drm_irq(int irq, void *arg)
 DEFINE_DRM_GEM_CMA_FOPS(fsl_dcu_drm_fops);
 
 static struct drm_driver fsl_dcu_drm_driver = {
-	.driver_features	= DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET
+	.driver_features	= DRIVER_GEM | DRIVER_MODESET
 				| DRIVER_PRIME | DRIVER_ATOMIC,
 	.load			= fsl_dcu_load,
 	.unload			= fsl_dcu_unload,
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_kms.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_kms.c
index ddc68e476a4d..e447f7d0c304 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_kms.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_kms.c
@@ -11,9 +11,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "fsl_dcu_drm_crtc.h"
 #include "fsl_dcu_drm_drv.h"
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c
index 9554b245746e..2a9e8a82c06a 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c
@@ -14,10 +14,10 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "fsl_dcu_drm_drv.h"
 #include "fsl_dcu_drm_plane.h"
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 2298ed2a9e1c..0a3a62b08240 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c
@@ -14,9 +14,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "fsl_dcu_drm_drv.h"
 #include "fsl_tcon.h"
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index adefae58b5fc..c934b3df1f81 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -405,7 +405,6 @@ static int psbfb_create(struct psb_fbdev *fbdev,
 	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
 	strcpy(info->fix.id, "psbdrmfb");
 
-	info->flags = FBINFO_DEFAULT;
 	if (dev_priv->ops->accel_2d && pitch_lines > 8)	/* 2D engine */
 		info->fbops = &psbfb_ops;
 	else if (gtt_roll) {	/* GTT rolling seems best */
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index ac32ab5aa002..eefaf4daff2b 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -468,8 +468,7 @@ static const struct file_operations psb_gem_fops = {
 };
 
 static struct drm_driver driver = {
-	.driver_features = DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | \
-			   DRIVER_MODESET | DRIVER_GEM,
+	.driver_features = DRIVER_MODESET | DRIVER_GEM,
 	.load = psb_driver_load,
 	.unload = psb_driver_unload,
 	.lastclose = drm_fb_helper_lastclose,
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index e05e5399af2d..8280a923b916 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -24,6 +24,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/gpio.h>
 #include "gma_display.h"
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
index a956545774a3..9316b724e7a2 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_de.c
@@ -18,8 +18,8 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "hibmc_drm_drv.h"
 #include "hibmc_drm_regs.h"
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index 68c0c297b3a5..8ed94fcd42a7 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -20,7 +20,7 @@
 #include <linux/module.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "hibmc_drm_drv.h"
 #include "hibmc_drm_regs.h"
@@ -56,8 +56,7 @@ static irqreturn_t hibmc_drm_interrupt(int irq, void *arg)
 }
 
 static struct drm_driver hibmc_driver = {
-	.driver_features	= DRIVER_GEM | DRIVER_MODESET |
-				  DRIVER_ATOMIC | DRIVER_HAVE_IRQ,
+	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC,
 	.fops			= &hibmc_fops,
 	.name			= "hibmc",
 	.date			= "20160828",
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
index edcca1761500..de9d7cc97e44 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
@@ -17,8 +17,8 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "hibmc_drm_drv.h"
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index 744956cea749..d2cf7317930a 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -17,7 +17,7 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "hibmc_drm_drv.h"
 #include "hibmc_drm_regs.h"
diff --git a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
index b4c7af3ab6ae..3d6c45097f51 100644
--- a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
+++ b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
@@ -17,12 +17,17 @@
 
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
 
-#include <drm/drm_of.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_mipi_dsi.h>
-#include <drm/drm_encoder_slave.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_encoder_slave.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
 
 #include "dw_dsi_reg.h"
 
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
index bb774202a5a1..73611a92d96c 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
@@ -23,13 +23,13 @@
 #include <linux/reset.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_crtc.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "kirin_drm_drv.h"
 #include "kirin_ade_reg.h"
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
index 15e32e5d9101..7cb7c042b93f 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_drv.c
@@ -20,13 +20,13 @@
 #include <linux/of_graph.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include "kirin_drm_drv.h"
 
diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/i2c/ch7006_priv.h
index dc6414af5d79..b6e091935977 100644
--- a/drivers/gpu/drm/i2c/ch7006_priv.h
+++ b/drivers/gpu/drm/i2c/ch7006_priv.h
@@ -30,6 +30,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder_slave.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/i2c/ch7006.h>
 
 typedef int64_t fixed;
diff --git a/drivers/gpu/drm/i2c/sil164_drv.c b/drivers/gpu/drm/i2c/sil164_drv.c
index c52d7a3af786..878ba8d06ce2 100644
--- a/drivers/gpu/drm/i2c/sil164_drv.c
+++ b/drivers/gpu/drm/i2c/sil164_drv.c
@@ -27,8 +27,8 @@
 #include <linux/module.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder_slave.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/i2c/sil164.h>
 
 struct sil164_priv {
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index ecdb8070ed35..7f34601bb515 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -26,9 +26,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/i2c/tda998x.h>
 
 #include <media/cec-notifier.h>
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 9c106e47e640..bc64b810e0d5 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -2799,6 +2799,7 @@ static int init_broadwell_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(_MMIO(0xe2a0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0xe2b0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0xe2c0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(_MMIO(0x21f0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/gvt/hypercall.h b/drivers/gpu/drm/i915/gvt/hypercall.h
index 5e01cc8d9b16..4862fb12778e 100644
--- a/drivers/gpu/drm/i915/gvt/hypercall.h
+++ b/drivers/gpu/drm/i915/gvt/hypercall.h
@@ -47,7 +47,7 @@ struct intel_gvt_mpt {
 	int (*host_init)(struct device *dev, void *gvt, const void *ops);
 	void (*host_exit)(struct device *dev);
 	int (*attach_vgpu)(void *vgpu, unsigned long *handle);
-	void (*detach_vgpu)(unsigned long handle);
+	void (*detach_vgpu)(void *vgpu);
 	int (*inject_msi)(unsigned long handle, u32 addr, u16 data);
 	unsigned long (*from_virt_to_mfn)(void *p);
 	int (*enable_page_track)(unsigned long handle, u64 gfn);
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index 63eef86a2a85..d5fcc447d22f 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -1147,7 +1147,7 @@ static int intel_vgpu_mmap(struct mdev_device *mdev, struct vm_area_struct *vma)
 {
 	unsigned int index;
 	u64 virtaddr;
-	unsigned long req_size, pgoff = 0;
+	unsigned long req_size, pgoff, req_start;
 	pgprot_t pg_prot;
 	struct intel_vgpu *vgpu = mdev_get_drvdata(mdev);
 
@@ -1165,7 +1165,17 @@ static int intel_vgpu_mmap(struct mdev_device *mdev, struct vm_area_struct *vma)
 	pg_prot = vma->vm_page_prot;
 	virtaddr = vma->vm_start;
 	req_size = vma->vm_end - vma->vm_start;
-	pgoff = vgpu_aperture_pa_base(vgpu) >> PAGE_SHIFT;
+	pgoff = vma->vm_pgoff &
+		((1U << (VFIO_PCI_OFFSET_SHIFT - PAGE_SHIFT)) - 1);
+	req_start = pgoff << PAGE_SHIFT;
+
+	if (!intel_vgpu_in_aperture(vgpu, req_start))
+		return -EINVAL;
+	if (req_start + req_size >
+	    vgpu_aperture_offset(vgpu) + vgpu_aperture_sz(vgpu))
+		return -EINVAL;
+
+	pgoff = (gvt_aperture_pa_base(vgpu->gvt) >> PAGE_SHIFT) + pgoff;
 
 	return remap_pfn_range(vma, virtaddr, pgoff, req_size, pg_prot);
 }
@@ -1813,9 +1823,21 @@ static int kvmgt_attach_vgpu(void *vgpu, unsigned long *handle)
 	return 0;
 }
 
-static void kvmgt_detach_vgpu(unsigned long handle)
+static void kvmgt_detach_vgpu(void *p_vgpu)
 {
-	/* nothing to do here */
+	int i;
+	struct intel_vgpu *vgpu = (struct intel_vgpu *)p_vgpu;
+
+	if (!vgpu->vdev.region)
+		return;
+
+	for (i = 0; i < vgpu->vdev.num_regions; i++)
+		if (vgpu->vdev.region[i].ops->release)
+			vgpu->vdev.region[i].ops->release(vgpu,
+					&vgpu->vdev.region[i]);
+	vgpu->vdev.num_regions = 0;
+	kfree(vgpu->vdev.region);
+	vgpu->vdev.region = NULL;
 }
 
 static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data)
diff --git a/drivers/gpu/drm/i915/gvt/mpt.h b/drivers/gpu/drm/i915/gvt/mpt.h
index 5d8b8f228d8f..0f9440128123 100644
--- a/drivers/gpu/drm/i915/gvt/mpt.h
+++ b/drivers/gpu/drm/i915/gvt/mpt.h
@@ -99,7 +99,7 @@ static inline void intel_gvt_hypervisor_detach_vgpu(struct intel_vgpu *vgpu)
 	if (!intel_gvt_host.mpt->detach_vgpu)
 		return;
 
-	intel_gvt_host.mpt->detach_vgpu(vgpu->handle);
+	intel_gvt_host.mpt->detach_vgpu(vgpu);
 }
 
 #define MSI_CAP_CONTROL(offset) (offset + 2)
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index b7957eefb976..1bb8f936fdaa 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -333,6 +333,9 @@ static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 
 	i915_gem_object_unpin_map(wa_ctx->indirect_ctx.obj);
 	i915_gem_object_put(wa_ctx->indirect_ctx.obj);
+
+	wa_ctx->indirect_ctx.obj = NULL;
+	wa_ctx->indirect_ctx.shadow_va = NULL;
 }
 
 static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
@@ -357,6 +360,33 @@ static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 	return 0;
 }
 
+static int
+intel_gvt_workload_req_alloc(struct intel_vgpu_workload *workload)
+{
+	struct intel_vgpu *vgpu = workload->vgpu;
+	struct intel_vgpu_submission *s = &vgpu->submission;
+	struct i915_gem_context *shadow_ctx = s->shadow_ctx;
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_engine_cs *engine = dev_priv->engine[workload->ring_id];
+	struct i915_request *rq;
+	int ret = 0;
+
+	lockdep_assert_held(&dev_priv->drm.struct_mutex);
+
+	if (workload->req)
+		goto out;
+
+	rq = i915_request_alloc(engine, shadow_ctx);
+	if (IS_ERR(rq)) {
+		gvt_vgpu_err("fail to allocate gem request\n");
+		ret = PTR_ERR(rq);
+		goto out;
+	}
+	workload->req = i915_request_get(rq);
+out:
+	return ret;
+}
+
 /**
  * intel_gvt_scan_and_shadow_workload - audit the workload by scanning and
  * shadow it as well, include ringbuffer,wa_ctx and ctx.
@@ -373,12 +403,11 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload)
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine = dev_priv->engine[workload->ring_id];
 	struct intel_context *ce;
-	struct i915_request *rq;
 	int ret;
 
 	lockdep_assert_held(&dev_priv->drm.struct_mutex);
 
-	if (workload->req)
+	if (workload->shadow)
 		return 0;
 
 	ret = set_context_ppgtt_from_shadow(workload, shadow_ctx);
@@ -418,22 +447,8 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload)
 			goto err_shadow;
 	}
 
-	rq = i915_request_alloc(engine, shadow_ctx);
-	if (IS_ERR(rq)) {
-		gvt_vgpu_err("fail to allocate gem request\n");
-		ret = PTR_ERR(rq);
-		goto err_shadow;
-	}
-	workload->req = i915_request_get(rq);
-
-	ret = populate_shadow_context(workload);
-	if (ret)
-		goto err_req;
-
+	workload->shadow = true;
 	return 0;
-err_req:
-	rq = fetch_and_zero(&workload->req);
-	i915_request_put(rq);
 err_shadow:
 	release_shadow_wa_ctx(&workload->wa_ctx);
 err_unpin:
@@ -672,23 +687,31 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
 	mutex_lock(&vgpu->vgpu_lock);
 	mutex_lock(&dev_priv->drm.struct_mutex);
 
+	ret = intel_gvt_workload_req_alloc(workload);
+	if (ret)
+		goto err_req;
+
 	ret = intel_gvt_scan_and_shadow_workload(workload);
 	if (ret)
 		goto out;
 
-	ret = prepare_workload(workload);
+	ret = populate_shadow_context(workload);
+	if (ret) {
+		release_shadow_wa_ctx(&workload->wa_ctx);
+		goto out;
+	}
 
+	ret = prepare_workload(workload);
 out:
-	if (ret)
-		workload->status = ret;
-
 	if (!IS_ERR_OR_NULL(workload->req)) {
 		gvt_dbg_sched("ring id %d submit workload to i915 %p\n",
 				ring_id, workload->req);
 		i915_request_add(workload->req);
 		workload->dispatched = true;
 	}
-
+err_req:
+	if (ret)
+		workload->status = ret;
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 	mutex_unlock(&vgpu->vgpu_lock);
 	return ret;
@@ -892,11 +915,6 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 
 	list_del_init(&workload->list);
 
-	if (!workload->status) {
-		release_shadow_batch_buffer(workload);
-		release_shadow_wa_ctx(&workload->wa_ctx);
-	}
-
 	if (workload->status || (vgpu->resetting_eng & ENGINE_MASK(ring_id))) {
 		/* if workload->status is not successful means HW GPU
 		 * has occurred GPU hang or something wrong with i915/GVT,
@@ -1262,6 +1280,9 @@ void intel_vgpu_destroy_workload(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu_submission *s = &workload->vgpu->submission;
 
+	release_shadow_batch_buffer(workload);
+	release_shadow_wa_ctx(&workload->wa_ctx);
+
 	if (workload->shadow_mm)
 		intel_vgpu_mm_put(workload->shadow_mm);
 
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h
index 1e9eec6a32fe..0635b2c4bed7 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.h
+++ b/drivers/gpu/drm/i915/gvt/scheduler.h
@@ -83,6 +83,7 @@ struct intel_vgpu_workload {
 	struct i915_request *req;
 	/* if this workload has been dispatched to i915? */
 	bool dispatched;
+	bool shadow;      /* if workload has done shadow of guest request */
 	int status;
 
 	struct intel_vgpu_mm *shadow_mm;
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index ca8fa4461fc9..2aeea977283f 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -26,9 +26,10 @@
  *
  */
 
-#include <linux/debugfs.h>
 #include <linux/sort.h>
 #include <linux/sched/mm.h>
+#include <drm/drm_debugfs.h>
+#include <drm/drm_fourcc.h>
 #include "intel_drv.h"
 #include "intel_guc_submission.h"
 
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 349ba9e23a5b..6630212f2faf 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -41,8 +41,10 @@
 #include <linux/vt.h>
 #include <acpi/video.h>
 
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_ioctl.h>
+#include <drm/drm_irq.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/i915_drm.h>
 
 #include "i915_drv.h"
@@ -3018,7 +3020,7 @@ static struct drm_driver driver = {
 	 * deal with them for Intel hardware.
 	 */
 	.driver_features =
-	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM | DRIVER_PRIME |
+	    DRIVER_GEM | DRIVER_PRIME |
 	    DRIVER_RENDER | DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_SYNCOBJ,
 	.release = i915_driver_release,
 	.open = i915_driver_open,
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 5c1b9d44b7d3..b421bc7a2e26 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -26,6 +26,7 @@
  */
 
 #include <drm/drm_vma_manager.h>
+#include <drm/drm_pci.h>
 #include <drm/i915_drm.h>
 #include <linux/dma-fence-array.h>
 #include <linux/kthread.h>
@@ -36,6 +37,7 @@
 #include <linux/swap.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
+#include <linux/mman.h>
 
 #include "i915_drv.h"
 #include "i915_gem_clflush.h"
diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c
index 73c3e8f519e8..c1007245f46d 100644
--- a/drivers/gpu/drm/i915/i915_ioc32.c
+++ b/drivers/gpu/drm/i915/i915_ioc32.c
@@ -29,6 +29,7 @@
 #include <linux/compat.h>
 
 #include <drm/i915_drm.h>
+#include <drm/drm_ioctl.h>
 #include "i915_drv.h"
 
 struct drm_i915_getparam32 {
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 292e31cb8fb1..92bb32ed27fb 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -31,6 +31,8 @@
 #include <linux/sysrq.h>
 #include <linux/slab.h>
 #include <linux/circ_buf.h>
+#include <drm/drm_irq.h>
+#include <drm/drm_drv.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 #include "i915_trace.h"
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index e1529a5a0af4..c4d6b8da9b03 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -26,6 +26,8 @@
 #include <linux/vgaarb.h>
 #include <linux/vga_switcheroo.h>
 
+#include <drm/drm_drv.h>
+
 #include "i915_active.h"
 #include "i915_drv.h"
 #include "i915_selftest.h"
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index b3b4d78d1dc0..0d9cedb892b0 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -6,6 +6,8 @@
 #include <linux/types.h>
 #include <linux/tracepoint.h>
 
+#include <drm/drm_drv.h>
+
 #include "i915_drv.h"
 #include "intel_drv.h"
 #include "intel_ringbuffer.h"
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 16263add3cdd..7cf9290ea34a 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -31,7 +31,9 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
+
 #include "intel_drv.h"
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c
index 021b94b78862..7961cf0e6951 100644
--- a/drivers/gpu/drm/i915/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.c
@@ -32,7 +32,9 @@
  */
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
+
 #include "intel_drv.h"
 
 struct intel_plane *intel_plane_alloc(void)
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index c2e799a5e63e..3716b2ee362f 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -29,8 +29,8 @@
 #include <linux/slab.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 #include "intel_drv.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e53d33edc85d..415d8968f2c5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -24,23 +24,32 @@
  *	Eric Anholt <eric@anholt.net>
  */
 
-#include <linux/module.h>
-#include <linux/input.h>
 #include <linux/i2c.h>
+#include <linux/input.h>
+#include <linux/intel-iommu.h>
 #include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/reservation.h>
 #include <linux/slab.h>
 #include <linux/vgaarb.h>
-#include <drm/drm_edid.h>
-#include <drm/i915_drm.h>
+
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_uapi.h>
 #include <drm/drm_dp_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
-#include <drm/drm_atomic_uapi.h>
-#include <linux/intel-iommu.h>
-#include <linux/reservation.h>
+#include <drm/i915_drm.h>
+
+#include "i915_drv.h"
+#include "i915_gem_clflush.h"
+#include "i915_trace.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+#include "intel_frontbuffer.h"
 
 #include "intel_drv.h"
 #include "intel_dsi.h"
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 8f38d932f2d1..cf709835fb9a 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -34,10 +34,10 @@
 #include <asm/byteorder.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_hdcp.h>
+#include <drm/drm_probe_helper.h>
 #include "intel_drv.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index f90041cb336d..fb67cd931117 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -26,8 +26,8 @@
 #include "i915_drv.h"
 #include "intel_drv.h"
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
 				       struct intel_crtc_state *pipe_config,
@@ -80,17 +80,12 @@ static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	mst_pbn = drm_dp_calc_pbn_mode(adjusted_mode->crtc_clock, bpp);
 	pipe_config->pbn = mst_pbn;
 
-	/* Zombie connectors can't have VCPI slots */
-	if (!drm_connector_is_unregistered(connector)) {
-		slots = drm_dp_atomic_find_vcpi_slots(state,
-						      &intel_dp->mst_mgr,
-						      port,
-						      mst_pbn);
-		if (slots < 0) {
-			DRM_DEBUG_KMS("failed finding vcpi slots:%d\n",
-				      slots);
-			return slots;
-		}
+	slots = drm_dp_atomic_find_vcpi_slots(state, &intel_dp->mst_mgr, port,
+					      mst_pbn);
+	if (slots < 0) {
+		DRM_DEBUG_KMS("failed finding vcpi slots:%d\n",
+			      slots);
+		return slots;
 	}
 
 	intel_link_compute_m_n(bpp, lane_count,
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index b4e29b823b04..eec4ed93c335 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -33,12 +33,13 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_dp_dual_mode_helper.h>
 #include <drm/drm_dp_mst_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
+#include <drm/drm_vblank.h>
 #include <drm/drm_atomic.h>
 #include <media/cec-notifier.h>
 
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index ec72be4b7a7b..656e684e7c9a 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -38,6 +38,8 @@
  * forcibly disable it to allow proper screen updates.
  */
 
+#include <drm/drm_fourcc.h>
+
 #include "intel_drv.h"
 #include "i915_drv.h"
 
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index fc02e6f81309..e8f694b57b8a 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -39,6 +39,8 @@
 
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_fourcc.h>
+
 #include "intel_drv.h"
 #include "intel_frontbuffer.h"
 #include <drm/i915_drm.h>
diff --git a/drivers/gpu/drm/i915/intel_lpe_audio.c b/drivers/gpu/drm/i915/intel_lpe_audio.c
index 5d5336fbe7b0..f8239bca3820 100644
--- a/drivers/gpu/drm/i915/intel_lpe_audio.c
+++ b/drivers/gpu/drm/i915/intel_lpe_audio.c
@@ -65,6 +65,7 @@
 #include <linux/irq.h>
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/platform_device.h>
 
 #include "i915_drv.h"
 #include <linux/delay.h>
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index fc2e283d326b..c0df1dbb0069 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -26,6 +26,8 @@
  * Derived from Xorg ddx, xf86-video-intel, src/i830_video.c
  */
 #include <drm/i915_drm.h>
+#include <drm/drm_fourcc.h>
+
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_drv.h"
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index c7ec9b169046..4c0e43caa5cd 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -26,13 +26,16 @@
  */
 
 #include <linux/cpufreq.h>
+#include <linux/module.h>
 #include <linux/pm_runtime.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
+
 #include "i915_drv.h"
 #include "intel_drv.h"
 #include "../../../platform/x86/intel_ips.h"
-#include <linux/module.h>
-#include <drm/drm_atomic_helper.h>
 
 /**
  * DOC: RC6
diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c
index 77a26fd3a44a..06393cd1067d 100644
--- a/drivers/gpu/drm/imx/dw_hdmi-imx.c
+++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c
@@ -13,7 +13,7 @@
 #include <linux/regmap.h>
 #include <drm/drm_of.h>
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder_slave.h>
 
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index 820c7e3878f0..44da0f5d0ed9 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -12,13 +12,13 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_plane_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <video/imx-ipu-v3.h>
 
 #include "imx-drm.h"
diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c
index 2c5bbe317353..383733302280 100644
--- a/drivers/gpu/drm/imx/imx-ldb.c
+++ b/drivers/gpu/drm/imx/imx-ldb.c
@@ -12,9 +12,9 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/mfd/syscon.h>
 #include <linux/mfd/syscon/imx6q-iomuxc-gpr.h>
 #include <linux/of_device.h>
@@ -643,8 +643,10 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 		int bus_format;
 
 		ret = of_property_read_u32(child, "reg", &i);
-		if (ret || i < 0 || i > 1)
-			return -EINVAL;
+		if (ret || i < 0 || i > 1) {
+			ret = -EINVAL;
+			goto free_child;
+		}
 
 		if (!of_device_is_available(child))
 			continue;
@@ -657,7 +659,6 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 		channel = &imx_ldb->channel[i];
 		channel->ldb = imx_ldb;
 		channel->chno = i;
-		channel->child = child;
 
 		/*
 		 * The output port is port@4 with an external 4-port mux or
@@ -667,13 +668,13 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 						  imx_ldb->lvds_mux ? 4 : 2, 0,
 						  &channel->panel, &channel->bridge);
 		if (ret && ret != -ENODEV)
-			return ret;
+			goto free_child;
 
 		/* panel ddc only if there is no bridge */
 		if (!channel->bridge) {
 			ret = imx_ldb_panel_ddc(dev, channel, child);
 			if (ret)
-				return ret;
+				goto free_child;
 		}
 
 		bus_format = of_get_bus_format(dev, child);
@@ -689,18 +690,26 @@ static int imx_ldb_bind(struct device *dev, struct device *master, void *data)
 		if (bus_format < 0) {
 			dev_err(dev, "could not determine data mapping: %d\n",
 				bus_format);
-			return bus_format;
+			ret = bus_format;
+			goto free_child;
 		}
 		channel->bus_format = bus_format;
+		channel->child = child;
 
 		ret = imx_ldb_register(drm, channel);
-		if (ret)
-			return ret;
+		if (ret) {
+			channel->child = NULL;
+			goto free_child;
+		}
 	}
 
 	dev_set_drvdata(dev, imx_ldb);
 
 	return 0;
+
+free_child:
+	of_node_put(child);
+	return ret;
 }
 
 static void imx_ldb_unbind(struct device *dev, struct device *master,
diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c
index 293dd5752583..e725af8a0025 100644
--- a/drivers/gpu/drm/imx/imx-tve.c
+++ b/drivers/gpu/drm/imx/imx-tve.c
@@ -17,7 +17,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <video/imx-ipu-v3.h>
 
 #include "imx-drm.h"
diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c
index 058b53c0aa7e..3c62167a9251 100644
--- a/drivers/gpu/drm/imx/ipuv3-crtc.c
+++ b/drivers/gpu/drm/imx/ipuv3-crtc.c
@@ -4,19 +4,19 @@
  *
  * Copyright (C) 2011 Sascha Hauer, Pengutronix
  */
+#include <linux/clk.h>
 #include <linux/component.h>
-#include <linux/module.h>
-#include <linux/export.h>
 #include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <linux/module.h>
 #include <linux/platform_device.h>
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
-#include <linux/clk.h>
-#include <linux/errno.h>
-#include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <video/imx-ipu-v3.h>
 #include "imx-drm.h"
diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index c390924de93d..21e964f6ab5c 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -370,9 +370,9 @@ static int ipu_plane_atomic_check(struct drm_plane *plane,
 	if (ret)
 		return ret;
 
-	/* CRTC should be enabled */
+	/* nothing to check when disabling or disabled */
 	if (!crtc_state->enable)
-		return -EINVAL;
+		return 0;
 
 	switch (plane->type) {
 	case DRM_PLANE_TYPE_PRIMARY:
diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c
index f3ce51121dd6..1a76de1e8e7b 100644
--- a/drivers/gpu/drm/imx/parallel-display.c
+++ b/drivers/gpu/drm/imx/parallel-display.c
@@ -10,9 +10,9 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/videodev2.h>
 #include <video/of_display_timing.h>
 
diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c
index 62a9d47df948..22e68a100e7b 100644
--- a/drivers/gpu/drm/mediatek/mtk_dpi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dpi.c
@@ -13,7 +13,7 @@
  */
 #include <drm/drmP.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_of.h>
 #include <linux/kernel.h>
 #include <linux/component.h>
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c
index 92ecb9bf982c..acad088173da 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c
@@ -14,8 +14,8 @@
 #include <asm/barrier.h>
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/pm_runtime.h>
 #include <soc/mediatek/smi.h>
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
index 6422e99952fe..cf59ea9bccfd 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
@@ -15,10 +15,10 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/component.h>
 #include <linux/iommu.h>
 #include <linux/of_address.h>
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_fb.c b/drivers/gpu/drm/mediatek/mtk_drm_fb.c
index be5f6f1daf55..e20fcaef2851 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_fb.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_fb.c
@@ -12,7 +12,7 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_gem_framebuffer_helper.h>
diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c
index 27b507eb4a99..b00eb2d2e086 100644
--- a/drivers/gpu/drm/mediatek/mtk_dsi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dsi.c
@@ -13,10 +13,10 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/iopoll.h>
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c
index 12ad00d01063..915cc84621ae 100644
--- a/drivers/gpu/drm/mediatek/mtk_hdmi.c
+++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c
@@ -14,7 +14,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_edid.h>
 #include <linux/arm-smccc.h>
 #include <linux/clk.h>
diff --git a/drivers/gpu/drm/meson/meson_crtc.c b/drivers/gpu/drm/meson/meson_crtc.c
index 75d97f1b2e8f..43e29984f8b1 100644
--- a/drivers/gpu/drm/meson/meson_crtc.c
+++ b/drivers/gpu/drm/meson/meson_crtc.c
@@ -30,7 +30,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_flip_work.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "meson_crtc.h"
 #include "meson_plane.h"
@@ -46,7 +46,6 @@ struct meson_crtc {
 	struct drm_crtc base;
 	struct drm_pending_vblank_event *event;
 	struct meson_drm *priv;
-	bool enabled;
 };
 #define to_meson_crtc(x) container_of(x, struct meson_crtc, base)
 
@@ -82,7 +81,8 @@ static const struct drm_crtc_funcs meson_crtc_funcs = {
 
 };
 
-static void meson_crtc_enable(struct drm_crtc *crtc)
+static void meson_crtc_atomic_enable(struct drm_crtc *crtc,
+				     struct drm_crtc_state *old_state)
 {
 	struct meson_crtc *meson_crtc = to_meson_crtc(crtc);
 	struct drm_crtc_state *crtc_state = crtc->state;
@@ -108,20 +108,6 @@ static void meson_crtc_enable(struct drm_crtc *crtc)
 
 	drm_crtc_vblank_on(crtc);
 
-	meson_crtc->enabled = true;
-}
-
-static void meson_crtc_atomic_enable(struct drm_crtc *crtc,
-				     struct drm_crtc_state *old_state)
-{
-	struct meson_crtc *meson_crtc = to_meson_crtc(crtc);
-	struct meson_drm *priv = meson_crtc->priv;
-
-	DRM_DEBUG_DRIVER("\n");
-
-	if (!meson_crtc->enabled)
-		meson_crtc_enable(crtc);
-
 	priv->viu.osd1_enabled = true;
 }
 
@@ -153,8 +139,6 @@ static void meson_crtc_atomic_disable(struct drm_crtc *crtc,
 
 		crtc->state->event = NULL;
 	}
-
-	meson_crtc->enabled = false;
 }
 
 static void meson_crtc_atomic_begin(struct drm_crtc *crtc,
@@ -163,9 +147,6 @@ static void meson_crtc_atomic_begin(struct drm_crtc *crtc,
 	struct meson_crtc *meson_crtc = to_meson_crtc(crtc);
 	unsigned long flags;
 
-	if (crtc->state->enable && !meson_crtc->enabled)
-		meson_crtc_enable(crtc);
-
 	if (crtc->state->event) {
 		WARN_ON(drm_crtc_vblank_get(crtc) != 0);
 
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index 1e360828ac38..2281ed3eb774 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -30,14 +30,14 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_flip_work.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_plane_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
-#include <drm/drm_fb_helper.h>
 
 #include "meson_drv.h"
 #include "meson_plane.h"
@@ -75,6 +75,10 @@ static const struct drm_mode_config_funcs meson_mode_config_funcs = {
 	.fb_create           = drm_gem_fb_create,
 };
 
+static const struct drm_mode_config_helper_funcs meson_mode_config_helpers = {
+	.atomic_commit_tail = drm_atomic_helper_commit_tail_rpm,
+};
+
 static irqreturn_t meson_irq(int irq, void *arg)
 {
 	struct drm_device *dev = arg;
@@ -90,7 +94,7 @@ static irqreturn_t meson_irq(int irq, void *arg)
 DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver meson_driver = {
-	.driver_features	= DRIVER_HAVE_IRQ | DRIVER_GEM |
+	.driver_features	= DRIVER_GEM |
 				  DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 
@@ -286,6 +290,7 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 	drm->mode_config.max_width = 3840;
 	drm->mode_config.max_height = 2160;
 	drm->mode_config.funcs = &meson_mode_config_funcs;
+	drm->mode_config.helper_private	= &meson_mode_config_helpers;
 
 	/* Hardware Initialization */
 
@@ -408,8 +413,10 @@ static int meson_probe_remote(struct platform_device *pdev,
 		remote_node = of_graph_get_remote_port_parent(ep);
 		if (!remote_node ||
 		    remote_node == parent || /* Ignore parent endpoint */
-		    !of_device_is_available(remote_node))
+		    !of_device_is_available(remote_node)) {
+			of_node_put(remote_node);
 			continue;
+		}
 
 		count += meson_probe_remote(pdev, match, remote, remote_node);
 
@@ -428,10 +435,13 @@ static int meson_drv_probe(struct platform_device *pdev)
 
 	for_each_endpoint_of_node(np, ep) {
 		remote = of_graph_get_remote_port_parent(ep);
-		if (!remote || !of_device_is_available(remote))
+		if (!remote || !of_device_is_available(remote)) {
+			of_node_put(remote);
 			continue;
+		}
 
 		count += meson_probe_remote(pdev, &match, np, remote);
+		of_node_put(remote);
 	}
 
 	if (count && !match)
diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c
index bc25001b8207..e28814f4ea6c 100644
--- a/drivers/gpu/drm/meson/meson_dw_hdmi.c
+++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c
@@ -26,9 +26,9 @@
 #include <linux/regulator/consumer.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_edid.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/bridge/dw_hdmi.h>
 
 #include <uapi/linux/media-bus-format.h>
@@ -365,7 +365,8 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 	unsigned int wr_clk =
 		readl_relaxed(priv->io_base + _REG(VPU_HDMI_SETTING));
 
-	DRM_DEBUG_DRIVER("\"%s\"\n", mode->name);
+	DRM_DEBUG_DRIVER("\"%s\" div%d\n", mode->name,
+			 mode->clock > 340000 ? 40 : 10);
 
 	/* Enable clocks */
 	regmap_update_bits(priv->hhi, HHI_HDMI_CLK_CNTL, 0xffff, 0x100);
@@ -385,9 +386,17 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 	/* Enable normal output to PHY */
 	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_BIST_CNTL, BIT(12));
 
-	/* TMDS pattern setup (TOFIX pattern for 4k2k scrambling) */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01, 0x001f001f);
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23, 0x001f001f);
+	/* TMDS pattern setup (TOFIX Handle the YUV420 case) */
+	if (mode->clock > 340000) {
+		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01, 0);
+		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
+				  0x03ff03ff);
+	} else {
+		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01,
+				  0x001f001f);
+		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
+				  0x001f001f);
+	}
 
 	/* Load TMDS pattern */
 	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_CNTL, 0x1);
@@ -413,6 +422,8 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 	/* Disable clock, fifo, fifo_wr */
 	regmap_update_bits(priv->hhi, HHI_HDMI_PHY_CNTL1, 0xf, 0);
 
+	dw_hdmi_set_high_tmds_clock_ratio(hdmi);
+
 	msleep(100);
 
 	/* Reset PHY 3 times in a row */
@@ -557,6 +568,10 @@ dw_hdmi_mode_valid(struct drm_connector *connector,
 
 	DRM_DEBUG_DRIVER("Modeline " DRM_MODE_FMT "\n", DRM_MODE_ARG(mode));
 
+	/* If sink max TMDS clock, we reject the mode */
+	if (mode->clock > connector->display_info.max_tmds_clock)
+		return MODE_BAD;
+
 	/* Check against non-VIC supported modes */
 	if (!vic) {
 		status = meson_venc_hdmi_supported_mode(mode);
diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index 0ba04f6813e6..66d73a932d19 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -848,6 +848,8 @@ struct meson_hdmi_venc_vic_mode {
 	{ 93, &meson_hdmi_encp_mode_2160p24 },
 	{ 94, &meson_hdmi_encp_mode_2160p25 },
 	{ 95, &meson_hdmi_encp_mode_2160p30 },
+	{ 96, &meson_hdmi_encp_mode_2160p25 },
+	{ 97, &meson_hdmi_encp_mode_2160p30 },
 	{ 0, NULL}, /* sentinel */
 };
 
diff --git a/drivers/gpu/drm/meson/meson_venc_cvbs.c b/drivers/gpu/drm/meson/meson_venc_cvbs.c
index f7945bae3b4a..d622d817b6df 100644
--- a/drivers/gpu/drm/meson/meson_venc_cvbs.c
+++ b/drivers/gpu/drm/meson/meson_venc_cvbs.c
@@ -26,9 +26,9 @@
 #include <linux/of_graph.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_edid.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include "meson_venc_cvbs.h"
 #include "meson_venc.h"
diff --git a/drivers/gpu/drm/mga/mga_drv.c b/drivers/gpu/drm/mga/mga_drv.c
index 1aad27813c23..6e1d1054ad06 100644
--- a/drivers/gpu/drm/mga/mga_drv.c
+++ b/drivers/gpu/drm/mga/mga_drv.c
@@ -57,7 +57,7 @@ static const struct file_operations mga_driver_fops = {
 static struct drm_driver driver = {
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_PCI_DMA | DRIVER_LEGACY |
-	    DRIVER_HAVE_DMA | DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED,
+	    DRIVER_HAVE_DMA | DRIVER_HAVE_IRQ,
 	.dev_priv_size = sizeof(drm_mga_buf_priv_t),
 	.load = mga_driver_load,
 	.unload = mga_driver_unload,
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index 79d54103d470..163255099779 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -33,7 +33,7 @@ int mgag200_framebuffer_init(struct drm_device *dev,
 			     struct drm_gem_object *obj)
 {
 	int ret;
-	
+
 	drm_helper_mode_fill_fb_struct(dev, &gfb->base, mode_cmd);
 	gfb->obj = obj;
 	ret = drm_framebuffer_init(dev, &gfb->base, &mga_fb_funcs);
@@ -318,13 +318,9 @@ int mgag200_dumb_create(struct drm_file *file,
 
 static void mgag200_bo_unref(struct mgag200_bo **bo)
 {
-	struct ttm_buffer_object *tbo;
-
 	if ((*bo) == NULL)
 		return;
-
-	tbo = &((*bo)->bo);
-	ttm_bo_unref(&tbo);
+	ttm_bo_put(&((*bo)->bo));
 	*bo = NULL;
 }
 
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index acf7bfe68454..7481a3d556ad 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -16,6 +16,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mgag200_drv.h"
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 5beb83d1cf87..ce1b3cc4bf6d 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -944,7 +944,7 @@ static u32 a6xx_gmu_get_arc_level(struct device *dev, unsigned long freq)
 	np = dev_pm_opp_get_of_node(opp);
 
 	if (np) {
-		of_property_read_u32(np, "qcom,level", &val);
+		of_property_read_u32(np, "opp-level", &val);
 		of_node_put(np);
 	}
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 2e4372ef17a3..2cfee1a4fe0b 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -765,7 +765,6 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	adreno_gpu->rev = config->rev;
 
 	adreno_gpu_config.ioname = "kgsl_3d0_reg_memory";
-	adreno_gpu_config.irqname = "kgsl_3d0_irq";
 
 	adreno_gpu_config.va_start = SZ_16M;
 	adreno_gpu_config.va_end = 0xffffffff;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index 9be7c355debd..b776fca571f3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -20,10 +20,10 @@
 #include <linux/sort.h>
 #include <linux/debugfs.h>
 #include <linux/ktime.h>
-#include <drm/drm_mode.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_flip_work.h>
+#include <drm/drm_mode.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
 
 #include "dpu_kms.h"
@@ -465,8 +465,6 @@ static void _dpu_crtc_setup_mixer_for_encoder(
 			return;
 		}
 
-		mixer->encoder = enc;
-
 		cstate->num_mixers++;
 		DPU_DEBUG("setup mixer %d: lm %d\n",
 				i, mixer->hw_lm->idx - LM_0);
@@ -718,11 +716,8 @@ void dpu_crtc_commit_kickoff(struct drm_crtc *crtc, bool async)
 	 * may delay and flush at an irq event (e.g. ppdone)
 	 */
 	drm_for_each_encoder_mask(encoder, crtc->dev,
-				  crtc->state->encoder_mask) {
-		struct dpu_encoder_kickoff_params params = { 0 };
-		dpu_encoder_prepare_for_kickoff(encoder, &params, async);
-	}
-
+				  crtc->state->encoder_mask)
+		dpu_encoder_prepare_for_kickoff(encoder, async);
 
 	if (!async) {
 		/* wait for frame_event_done completion */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h
index dbfb38a1986c..e59d62be4980 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h
@@ -84,14 +84,12 @@ struct dpu_crtc_smmu_state_data {
  * struct dpu_crtc_mixer: stores the map for each virtual pipeline in the CRTC
  * @hw_lm:	LM HW Driver context
  * @lm_ctl:	CTL Path HW driver context
- * @encoder:	Encoder attached to this lm & ctl
  * @mixer_op_mode:	mixer blending operation mode
  * @flush_mask:	mixer flush mask for ctl, mixer and pipe
  */
 struct dpu_crtc_mixer {
 	struct dpu_hw_mixer *hw_lm;
 	struct dpu_hw_ctl *lm_ctl;
-	struct drm_encoder *encoder;
 	u32 mixer_op_mode;
 	u32 flush_mask;
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 36158b7d99cd..5aa3307f3f0c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -24,7 +24,7 @@
 #include "msm_drv.h"
 #include "dpu_kms.h"
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "dpu_hwio.h"
 #include "dpu_hw_catalog.h"
 #include "dpu_hw_intf.h"
@@ -205,7 +205,7 @@ struct dpu_encoder_virt {
 	bool idle_pc_supported;
 	struct mutex rc_lock;
 	enum dpu_enc_rc_states rc_state;
-	struct kthread_delayed_work delayed_off_work;
+	struct delayed_work delayed_off_work;
 	struct kthread_work vsync_event_work;
 	struct msm_display_topology topology;
 	bool mode_set_complete;
@@ -742,7 +742,6 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc,
 {
 	struct dpu_encoder_virt *dpu_enc;
 	struct msm_drm_private *priv;
-	struct msm_drm_thread *disp_thread;
 	bool is_vid_mode = false;
 
 	if (!drm_enc || !drm_enc->dev || !drm_enc->dev->dev_private ||
@@ -755,12 +754,6 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc,
 	is_vid_mode = dpu_enc->disp_info.capabilities &
 						MSM_DISPLAY_CAP_VID_MODE;
 
-	if (drm_enc->crtc->index >= ARRAY_SIZE(priv->disp_thread)) {
-		DPU_ERROR("invalid crtc index\n");
-		return -EINVAL;
-	}
-	disp_thread = &priv->disp_thread[drm_enc->crtc->index];
-
 	/*
 	 * when idle_pc is not supported, process only KICKOFF, STOP and MODESET
 	 * events and return early for other events (ie wb display).
@@ -777,8 +770,7 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc,
 	switch (sw_event) {
 	case DPU_ENC_RC_EVENT_KICKOFF:
 		/* cancel delayed off work, if any */
-		if (kthread_cancel_delayed_work_sync(
-				&dpu_enc->delayed_off_work))
+		if (cancel_delayed_work_sync(&dpu_enc->delayed_off_work))
 			DPU_DEBUG_ENC(dpu_enc, "sw_event:%d, work cancelled\n",
 					sw_event);
 
@@ -837,10 +829,8 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc,
 			return 0;
 		}
 
-		kthread_queue_delayed_work(
-			&disp_thread->worker,
-			&dpu_enc->delayed_off_work,
-			msecs_to_jiffies(dpu_enc->idle_timeout));
+		queue_delayed_work(priv->wq, &dpu_enc->delayed_off_work,
+				   msecs_to_jiffies(dpu_enc->idle_timeout));
 
 		trace_dpu_enc_rc(DRMID(drm_enc), sw_event,
 				 dpu_enc->idle_pc_supported, dpu_enc->rc_state,
@@ -849,8 +839,7 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc,
 
 	case DPU_ENC_RC_EVENT_PRE_STOP:
 		/* cancel delayed off work, if any */
-		if (kthread_cancel_delayed_work_sync(
-				&dpu_enc->delayed_off_work))
+		if (cancel_delayed_work_sync(&dpu_enc->delayed_off_work))
 			DPU_DEBUG_ENC(dpu_enc, "sw_event:%d, work cancelled\n",
 					sw_event);
 
@@ -1368,7 +1357,7 @@ static void dpu_encoder_frame_done_callback(
 	}
 }
 
-static void dpu_encoder_off_work(struct kthread_work *work)
+static void dpu_encoder_off_work(struct work_struct *work)
 {
 	struct dpu_encoder_virt *dpu_enc = container_of(work,
 			struct dpu_encoder_virt, delayed_off_work.work);
@@ -1756,15 +1745,14 @@ static void dpu_encoder_vsync_event_work_handler(struct kthread_work *work)
 			nsecs_to_jiffies(ktime_to_ns(wakeup_time)));
 }
 
-void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc,
-		struct dpu_encoder_kickoff_params *params, bool async)
+void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc, bool async)
 {
 	struct dpu_encoder_virt *dpu_enc;
 	struct dpu_encoder_phys *phys;
 	bool needs_hw_reset = false;
 	unsigned int i;
 
-	if (!drm_enc || !params) {
+	if (!drm_enc) {
 		DPU_ERROR("invalid args\n");
 		return;
 	}
@@ -1778,7 +1766,7 @@ void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc,
 		phys = dpu_enc->phys_encs[i];
 		if (phys) {
 			if (phys->ops.prepare_for_kickoff)
-				phys->ops.prepare_for_kickoff(phys, params);
+				phys->ops.prepare_for_kickoff(phys);
 			if (phys->enable_state == DPU_ENC_ERR_NEEDS_HW_RESET)
 				needs_hw_reset = true;
 		}
@@ -2193,7 +2181,7 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
 
 
 	mutex_init(&dpu_enc->rc_lock);
-	kthread_init_delayed_work(&dpu_enc->delayed_off_work,
+	INIT_DELAYED_WORK(&dpu_enc->delayed_off_work,
 			dpu_encoder_off_work);
 	dpu_enc->idle_timeout = IDLE_TIMEOUT;
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
index 3f5dafe00580..d77f74fb26d4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
@@ -38,15 +38,6 @@ struct dpu_encoder_hw_resources {
 };
 
 /**
- * dpu_encoder_kickoff_params - info encoder requires at kickoff
- * @affected_displays:  bitmask, bit set means the ROI of the commit lies within
- *                      the bounds of the physical display at the bit index
- */
-struct dpu_encoder_kickoff_params {
-	unsigned long affected_displays;
-};
-
-/**
  * dpu_encoder_get_hw_resources - Populate table of required hardware resources
  * @encoder:	encoder pointer
  * @hw_res:	resource table to populate with encoder required resources
@@ -88,11 +79,9 @@ void dpu_encoder_register_frame_event_callback(struct drm_encoder *encoder,
  *	Immediately: if no previous commit is outstanding.
  *	Delayed: Block until next trigger can be issued.
  * @encoder:	encoder pointer
- * @params:	kickoff time parameters
  * @async:	true if this is an asynchronous commit
  */
-void dpu_encoder_prepare_for_kickoff(struct drm_encoder *encoder,
-		struct dpu_encoder_kickoff_params *params, bool async);
+void dpu_encoder_prepare_for_kickoff(struct drm_encoder *encoder,  bool async);
 
 /**
  * dpu_encoder_trigger_kickoff_pending - Clear the flush bits from previous
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index 44e6f8b68e70..db94f3d3bea3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -144,8 +144,7 @@ struct dpu_encoder_phys_ops {
 	int (*wait_for_commit_done)(struct dpu_encoder_phys *phys_enc);
 	int (*wait_for_tx_complete)(struct dpu_encoder_phys *phys_enc);
 	int (*wait_for_vblank)(struct dpu_encoder_phys *phys_enc);
-	void (*prepare_for_kickoff)(struct dpu_encoder_phys *phys_enc,
-			struct dpu_encoder_kickoff_params *params);
+	void (*prepare_for_kickoff)(struct dpu_encoder_phys *phys_enc);
 	void (*handle_post_kickoff)(struct dpu_encoder_phys *phys_enc);
 	void (*trigger_start)(struct dpu_encoder_phys *phys_enc);
 	bool (*needs_single_flush)(struct dpu_encoder_phys *phys_enc);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 99ab5ca9bed3..a399e1edd313 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -594,8 +594,7 @@ static void dpu_encoder_phys_cmd_get_hw_resources(
 }
 
 static void dpu_encoder_phys_cmd_prepare_for_kickoff(
-		struct dpu_encoder_phys *phys_enc,
-		struct dpu_encoder_kickoff_params *params)
+		struct dpu_encoder_phys *phys_enc)
 {
 	struct dpu_encoder_phys_cmd *cmd_enc =
 			to_dpu_encoder_phys_cmd(phys_enc);
@@ -693,7 +692,7 @@ static int dpu_encoder_phys_cmd_wait_for_commit_done(
 
 	/* required for both controllers */
 	if (!rc && cmd_enc->serialize_wait4pp)
-		dpu_encoder_phys_cmd_prepare_for_kickoff(phys_enc, NULL);
+		dpu_encoder_phys_cmd_prepare_for_kickoff(phys_enc);
 
 	return rc;
 }
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 acdab5b0db18..3c4eb470a82c 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
@@ -587,14 +587,13 @@ static int dpu_encoder_phys_vid_wait_for_vblank(
 }
 
 static void dpu_encoder_phys_vid_prepare_for_kickoff(
-		struct dpu_encoder_phys *phys_enc,
-		struct dpu_encoder_kickoff_params *params)
+		struct dpu_encoder_phys *phys_enc)
 {
 	struct dpu_encoder_phys_vid *vid_enc;
 	struct dpu_hw_ctl *ctl;
 	int rc;
 
-	if (!phys_enc || !params) {
+	if (!phys_enc) {
 		DPU_ERROR("invalid encoder/parameters\n");
 		return;
 	}
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
index 0874f0a53bf9..f59fe1a9f4b9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
@@ -263,13 +263,13 @@ static const struct dpu_format dpu_format_map[] = {
 
 	INTERLEAVED_RGB_FMT(RGB565,
 		0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT,
-		C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3,
+		C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3,
 		false, 2, 0,
 		DPU_FETCH_LINEAR, 1),
 
 	INTERLEAVED_RGB_FMT(BGR565,
 		0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT,
-		C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3,
+		C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3,
 		false, 2, 0,
 		DPU_FETCH_LINEAR, 1),
 
@@ -1137,36 +1137,3 @@ const struct msm_format *dpu_get_msm_format(
 		return &fmt->base;
 	return NULL;
 }
-
-uint32_t dpu_populate_formats(
-		const struct dpu_format_extended *format_list,
-		uint32_t *pixel_formats,
-		uint64_t *pixel_modifiers,
-		uint32_t pixel_formats_max)
-{
-	uint32_t i, fourcc_format;
-
-	if (!format_list || !pixel_formats)
-		return 0;
-
-	for (i = 0, fourcc_format = 0;
-			format_list->fourcc_format && i < pixel_formats_max;
-			++format_list) {
-		/* verify if listed format is in dpu_format_map? */
-
-		/* optionally return modified formats */
-		if (pixel_modifiers) {
-			/* assume same modifier for all fb planes */
-			pixel_formats[i] = format_list->fourcc_format;
-			pixel_modifiers[i++] = format_list->modifier;
-		} else {
-			/* assume base formats grouped together */
-			if (fourcc_format != format_list->fourcc_format) {
-				fourcc_format = format_list->fourcc_format;
-				pixel_formats[i++] = fourcc_format;
-			}
-		}
-	}
-
-	return i;
-}
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
index a54451d8d011..c02c81e7a667 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
@@ -41,20 +41,6 @@ const struct msm_format *dpu_get_msm_format(
 		const uint64_t modifiers);
 
 /**
- * dpu_populate_formats - populate the given array with fourcc codes supported
- * @format_list:       pointer to list of possible formats
- * @pixel_formats:     array to populate with fourcc codes
- * @pixel_modifiers:   array to populate with drm modifiers, can be NULL
- * @pixel_formats_max: length of pixel formats array
- * Return: number of elements populated
- */
-uint32_t dpu_populate_formats(
-		const struct dpu_format_extended *format_list,
-		uint32_t *pixel_formats,
-		uint64_t *pixel_modifiers,
-		uint32_t pixel_formats_max);
-
-/**
  * dpu_format_check_modified_format - validate format and buffers for
  *                   dpu non-standard, i.e. modified format
  * @kms:             kms driver
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 512ac0834d2b..df6852cc98b9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -151,7 +151,9 @@ static const struct dpu_sspp_blks_common sdm845_sspp_common = {
 		.id = DPU_SSPP_CSC_10BIT, \
 		.base = 0x1a00, .len = 0x100,}, \
 	.format_list = plane_formats_yuv, \
+	.num_formats = ARRAY_SIZE(plane_formats_yuv), \
 	.virt_format_list = plane_formats, \
+	.virt_num_formats = ARRAY_SIZE(plane_formats), \
 	}
 
 #define _DMA_SBLK(num, sdma_pri) \
@@ -163,7 +165,9 @@ static const struct dpu_sspp_blks_common sdm845_sspp_common = {
 	.src_blk = {.name = STRCAT("sspp_src_", num), \
 		.id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \
 	.format_list = plane_formats, \
+	.num_formats = ARRAY_SIZE(plane_formats), \
 	.virt_format_list = plane_formats, \
+	.virt_num_formats = ARRAY_SIZE(plane_formats), \
 	}
 
 static const struct dpu_sspp_sub_blks sdm845_vig_sblk_0 = _VIG_SBLK("0", 5);
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 144358a3d0fb..a55653b2e466 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -252,17 +252,6 @@ struct dpu_pp_blk {
 };
 
 /**
- * struct dpu_format_extended - define dpu specific pixel format+modifier
- * @fourcc_format: Base FOURCC pixel format code
- * @modifier: 64-bit drm format modifier, same modifier must be applied to all
- *            framebuffer planes
- */
-struct dpu_format_extended {
-	uint32_t fourcc_format;
-	uint64_t modifier;
-};
-
-/**
  * enum dpu_qos_lut_usage - define QoS LUT use cases
  */
 enum dpu_qos_lut_usage {
@@ -348,7 +337,9 @@ struct dpu_sspp_blks_common {
  * @pcc_blk:
  * @igc_blk:
  * @format_list: Pointer to list of supported formats
+ * @num_formats: Number of supported formats
  * @virt_format_list: Pointer to list of supported formats for virtual planes
+ * @virt_num_formats: Number of supported formats for virtual planes
  */
 struct dpu_sspp_sub_blks {
 	const struct dpu_sspp_blks_common *common;
@@ -366,8 +357,10 @@ struct dpu_sspp_sub_blks {
 	struct dpu_pp_blk pcc_blk;
 	struct dpu_pp_blk igc_blk;
 
-	const struct dpu_format_extended *format_list;
-	const struct dpu_format_extended *virt_format_list;
+	const u32 *format_list;
+	u32 num_formats;
+	const u32 *virt_format_list;
+	u32 virt_num_formats;
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog_format.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog_format.h
index 3c9f028628ef..d09730985951 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog_format.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog_format.h
@@ -12,157 +12,81 @@
 
 #include "dpu_hw_mdss.h"
 
-static const struct dpu_format_extended plane_formats[] = {
-	{DRM_FORMAT_ARGB8888, 0},
-	{DRM_FORMAT_ABGR8888, 0},
-	{DRM_FORMAT_RGBA8888, 0},
-	{DRM_FORMAT_ABGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_BGRA8888, 0},
-	{DRM_FORMAT_XRGB8888, 0},
-	{DRM_FORMAT_RGBX8888, 0},
-	{DRM_FORMAT_BGRX8888, 0},
-	{DRM_FORMAT_XBGR8888, 0},
-	{DRM_FORMAT_XBGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_RGB888, 0},
-	{DRM_FORMAT_BGR888, 0},
-	{DRM_FORMAT_RGB565, 0},
-	{DRM_FORMAT_BGR565, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_BGR565, 0},
-	{DRM_FORMAT_ARGB1555, 0},
-	{DRM_FORMAT_ABGR1555, 0},
-	{DRM_FORMAT_RGBA5551, 0},
-	{DRM_FORMAT_BGRA5551, 0},
-	{DRM_FORMAT_XRGB1555, 0},
-	{DRM_FORMAT_XBGR1555, 0},
-	{DRM_FORMAT_RGBX5551, 0},
-	{DRM_FORMAT_BGRX5551, 0},
-	{DRM_FORMAT_ARGB4444, 0},
-	{DRM_FORMAT_ABGR4444, 0},
-	{DRM_FORMAT_RGBA4444, 0},
-	{DRM_FORMAT_BGRA4444, 0},
-	{DRM_FORMAT_XRGB4444, 0},
-	{DRM_FORMAT_XBGR4444, 0},
-	{DRM_FORMAT_RGBX4444, 0},
-	{DRM_FORMAT_BGRX4444, 0},
-	{0, 0},
+static const uint32_t qcom_compressed_supported_formats[] = {
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_BGR565,
 };
 
-static const struct dpu_format_extended plane_formats_yuv[] = {
-	{DRM_FORMAT_ARGB8888, 0},
-	{DRM_FORMAT_ABGR8888, 0},
-	{DRM_FORMAT_RGBA8888, 0},
-	{DRM_FORMAT_BGRX8888, 0},
-	{DRM_FORMAT_ABGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_BGRA8888, 0},
-	{DRM_FORMAT_XRGB8888, 0},
-	{DRM_FORMAT_XBGR8888, 0},
-	{DRM_FORMAT_RGBX8888, 0},
-	{DRM_FORMAT_XBGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_RGB888, 0},
-	{DRM_FORMAT_BGR888, 0},
-	{DRM_FORMAT_RGB565, 0},
-	{DRM_FORMAT_BGR565, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_BGR565, 0},
-	{DRM_FORMAT_ARGB1555, 0},
-	{DRM_FORMAT_ABGR1555, 0},
-	{DRM_FORMAT_RGBA5551, 0},
-	{DRM_FORMAT_BGRA5551, 0},
-	{DRM_FORMAT_XRGB1555, 0},
-	{DRM_FORMAT_XBGR1555, 0},
-	{DRM_FORMAT_RGBX5551, 0},
-	{DRM_FORMAT_BGRX5551, 0},
-	{DRM_FORMAT_ARGB4444, 0},
-	{DRM_FORMAT_ABGR4444, 0},
-	{DRM_FORMAT_RGBA4444, 0},
-	{DRM_FORMAT_BGRA4444, 0},
-	{DRM_FORMAT_XRGB4444, 0},
-	{DRM_FORMAT_XBGR4444, 0},
-	{DRM_FORMAT_RGBX4444, 0},
-	{DRM_FORMAT_BGRX4444, 0},
-
-	{DRM_FORMAT_NV12, 0},
-	{DRM_FORMAT_NV12, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_NV21, 0},
-	{DRM_FORMAT_NV16, 0},
-	{DRM_FORMAT_NV61, 0},
-	{DRM_FORMAT_VYUY, 0},
-	{DRM_FORMAT_UYVY, 0},
-	{DRM_FORMAT_YUYV, 0},
-	{DRM_FORMAT_YVYU, 0},
-	{DRM_FORMAT_YUV420, 0},
-	{DRM_FORMAT_YVU420, 0},
-	{0, 0},
-};
-
-static const struct dpu_format_extended cursor_formats[] = {
-	{DRM_FORMAT_ARGB8888, 0},
-	{DRM_FORMAT_ABGR8888, 0},
-	{DRM_FORMAT_RGBA8888, 0},
-	{DRM_FORMAT_BGRA8888, 0},
-	{DRM_FORMAT_XRGB8888, 0},
-	{DRM_FORMAT_ARGB1555, 0},
-	{DRM_FORMAT_ABGR1555, 0},
-	{DRM_FORMAT_RGBA5551, 0},
-	{DRM_FORMAT_BGRA5551, 0},
-	{DRM_FORMAT_ARGB4444, 0},
-	{DRM_FORMAT_ABGR4444, 0},
-	{DRM_FORMAT_RGBA4444, 0},
-	{DRM_FORMAT_BGRA4444, 0},
-	{0, 0},
+static const uint32_t plane_formats[] = {
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGBX8888,
+	DRM_FORMAT_BGRX8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGR888,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR565,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_RGBA5551,
+	DRM_FORMAT_BGRA5551,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_RGBX5551,
+	DRM_FORMAT_BGRX5551,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_ABGR4444,
+	DRM_FORMAT_RGBA4444,
+	DRM_FORMAT_BGRA4444,
+	DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_XBGR4444,
+	DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_BGRX4444,
 };
 
-static const struct dpu_format_extended wb2_formats[] = {
-	{DRM_FORMAT_RGB565, 0},
-	{DRM_FORMAT_BGR565, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_RGB888, 0},
-	{DRM_FORMAT_ARGB8888, 0},
-	{DRM_FORMAT_RGBA8888, 0},
-	{DRM_FORMAT_ABGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_XRGB8888, 0},
-	{DRM_FORMAT_RGBX8888, 0},
-	{DRM_FORMAT_XBGR8888, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_ARGB1555, 0},
-	{DRM_FORMAT_RGBA5551, 0},
-	{DRM_FORMAT_XRGB1555, 0},
-	{DRM_FORMAT_RGBX5551, 0},
-	{DRM_FORMAT_ARGB4444, 0},
-	{DRM_FORMAT_RGBA4444, 0},
-	{DRM_FORMAT_RGBX4444, 0},
-	{DRM_FORMAT_XRGB4444, 0},
-
-	{DRM_FORMAT_BGR565, 0},
-	{DRM_FORMAT_BGR888, 0},
-	{DRM_FORMAT_ABGR8888, 0},
-	{DRM_FORMAT_BGRA8888, 0},
-	{DRM_FORMAT_BGRX8888, 0},
-	{DRM_FORMAT_XBGR8888, 0},
-	{DRM_FORMAT_ABGR1555, 0},
-	{DRM_FORMAT_BGRA5551, 0},
-	{DRM_FORMAT_XBGR1555, 0},
-	{DRM_FORMAT_BGRX5551, 0},
-	{DRM_FORMAT_ABGR4444, 0},
-	{DRM_FORMAT_BGRA4444, 0},
-	{DRM_FORMAT_BGRX4444, 0},
-	{DRM_FORMAT_XBGR4444, 0},
-
-	{DRM_FORMAT_YUV420, 0},
-	{DRM_FORMAT_NV12, 0},
-	{DRM_FORMAT_NV12, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_NV16, 0},
-	{DRM_FORMAT_YUYV, 0},
-
-	{0, 0},
-};
+static const uint32_t plane_formats_yuv[] = {
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_BGRX8888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGBX8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGR888,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR565,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_RGBA5551,
+	DRM_FORMAT_BGRA5551,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_RGBX5551,
+	DRM_FORMAT_BGRX5551,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_ABGR4444,
+	DRM_FORMAT_RGBA4444,
+	DRM_FORMAT_BGRA4444,
+	DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_XBGR4444,
+	DRM_FORMAT_RGBX4444,
+	DRM_FORMAT_BGRX4444,
 
-static const struct dpu_format_extended rgb_10bit_formats[] = {
-	{DRM_FORMAT_BGRA1010102, 0},
-	{DRM_FORMAT_BGRX1010102, 0},
-	{DRM_FORMAT_RGBA1010102, 0},
-	{DRM_FORMAT_RGBX1010102, 0},
-	{DRM_FORMAT_ABGR2101010, 0},
-	{DRM_FORMAT_ABGR2101010, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_XBGR2101010, 0},
-	{DRM_FORMAT_XBGR2101010, DRM_FORMAT_MOD_QCOM_COMPRESSED},
-	{DRM_FORMAT_ARGB2101010, 0},
-	{DRM_FORMAT_XRGB2101010, 0},
+	DRM_FORMAT_NV12,
+	DRM_FORMAT_NV21,
+	DRM_FORMAT_NV16,
+	DRM_FORMAT_NV61,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_YUV420,
+	DRM_FORMAT_YVU420,
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
index c0b7f0049365..8a28a03ac6a9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
@@ -170,10 +170,6 @@
 /**
  * AD4 interrupt status bit definitions
  */
-#define DPU_INTR_BRIGHTPR_UPDATED BIT(4)
-#define DPU_INTR_DARKENH_UPDATED BIT(3)
-#define DPU_INTR_STREN_OUTROI_UPDATED BIT(2)
-#define DPU_INTR_STREN_INROI_UPDATED BIT(1)
 #define DPU_INTR_BACKLIGHT_UPDATED BIT(0)
 /**
  * struct dpu_intr_reg - array of DPU register sets
@@ -782,18 +778,6 @@ static int dpu_hw_intr_irqidx_lookup(enum dpu_intr_type intr_type,
 	return -EINVAL;
 }
 
-static void dpu_hw_intr_set_mask(struct dpu_hw_intr *intr, uint32_t reg_off,
-		uint32_t mask)
-{
-	if (!intr)
-		return;
-
-	DPU_REG_WRITE(&intr->hw, reg_off, mask);
-
-	/* ensure register writes go through */
-	wmb();
-}
-
 static void dpu_hw_intr_dispatch_irq(struct dpu_hw_intr *intr,
 		void (*cbfunc)(void *, int),
 		void *arg)
@@ -1004,18 +988,6 @@ static int dpu_hw_intr_disable_irqs(struct dpu_hw_intr *intr)
 	return 0;
 }
 
-static int dpu_hw_intr_get_valid_interrupts(struct dpu_hw_intr *intr,
-		uint32_t *mask)
-{
-	if (!intr || !mask)
-		return -EINVAL;
-
-	*mask = IRQ_SOURCE_MDP | IRQ_SOURCE_DSI0 | IRQ_SOURCE_DSI1
-		| IRQ_SOURCE_HDMI | IRQ_SOURCE_EDP;
-
-	return 0;
-}
-
 static void dpu_hw_intr_get_interrupt_statuses(struct dpu_hw_intr *intr)
 {
 	int i;
@@ -1065,19 +1037,6 @@ static void dpu_hw_intr_clear_intr_status_nolock(struct dpu_hw_intr *intr,
 	wmb();
 }
 
-static void dpu_hw_intr_clear_interrupt_status(struct dpu_hw_intr *intr,
-		int irq_idx)
-{
-	unsigned long irq_flags;
-
-	if (!intr)
-		return;
-
-	spin_lock_irqsave(&intr->irq_lock, irq_flags);
-	dpu_hw_intr_clear_intr_status_nolock(intr, irq_idx);
-	spin_unlock_irqrestore(&intr->irq_lock, irq_flags);
-}
-
 static u32 dpu_hw_intr_get_interrupt_status(struct dpu_hw_intr *intr,
 		int irq_idx, bool clear)
 {
@@ -1113,16 +1072,13 @@ static u32 dpu_hw_intr_get_interrupt_status(struct dpu_hw_intr *intr,
 
 static void __setup_intr_ops(struct dpu_hw_intr_ops *ops)
 {
-	ops->set_mask = dpu_hw_intr_set_mask;
 	ops->irq_idx_lookup = dpu_hw_intr_irqidx_lookup;
 	ops->enable_irq = dpu_hw_intr_enable_irq;
 	ops->disable_irq = dpu_hw_intr_disable_irq;
 	ops->dispatch_irqs = dpu_hw_intr_dispatch_irq;
 	ops->clear_all_irqs = dpu_hw_intr_clear_irqs;
 	ops->disable_all_irqs = dpu_hw_intr_disable_irqs;
-	ops->get_valid_interrupts = dpu_hw_intr_get_valid_interrupts;
 	ops->get_interrupt_statuses = dpu_hw_intr_get_interrupt_statuses;
-	ops->clear_interrupt_status = dpu_hw_intr_clear_interrupt_status;
 	ops->clear_intr_status_nolock = dpu_hw_intr_clear_intr_status_nolock;
 	ops->get_interrupt_status = dpu_hw_intr_get_interrupt_status;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
index 61e4cba36562..4d7a1c727ce2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
@@ -20,13 +20,6 @@
 #include "dpu_hw_util.h"
 #include "dpu_hw_mdss.h"
 
-#define IRQ_SOURCE_MDP		BIT(0)
-#define IRQ_SOURCE_DSI0		BIT(4)
-#define IRQ_SOURCE_DSI1		BIT(5)
-#define IRQ_SOURCE_HDMI		BIT(8)
-#define IRQ_SOURCE_EDP		BIT(12)
-#define IRQ_SOURCE_MHL		BIT(16)
-
 /**
  * dpu_intr_type - HW Interrupt Type
  * @DPU_IRQ_TYPE_WB_ROT_COMP:		WB rotator done
@@ -96,18 +89,6 @@ struct dpu_hw_intr;
  */
 struct dpu_hw_intr_ops {
 	/**
-	 * set_mask - Programs the given interrupt register with the
-	 *            given interrupt mask. Register value will get overwritten.
-	 * @intr:	HW interrupt handle
-	 * @reg_off:	MDSS HW register offset
-	 * @irqmask:	IRQ mask value
-	 */
-	void (*set_mask)(
-			struct dpu_hw_intr *intr,
-			uint32_t reg,
-			uint32_t irqmask);
-
-	/**
 	 * irq_idx_lookup - Lookup IRQ index on the HW interrupt type
 	 *                 Used for all irq related ops
 	 * @intr_type:		Interrupt type defined in dpu_intr_type
@@ -177,16 +158,6 @@ struct dpu_hw_intr_ops {
 			struct dpu_hw_intr *intr);
 
 	/**
-	 * clear_interrupt_status - Clears HW interrupt status based on given
-	 *                          lookup IRQ index.
-	 * @intr:	HW interrupt handle
-	 * @irq_idx:	Lookup irq index return from irq_idx_lookup
-	 */
-	void (*clear_interrupt_status)(
-			struct dpu_hw_intr *intr,
-			int irq_idx);
-
-	/**
 	 * clear_intr_status_nolock() - clears the HW interrupts without lock
 	 * @intr:	HW interrupt handle
 	 * @irq_idx:	Lookup irq index return from irq_idx_lookup
@@ -206,21 +177,6 @@ struct dpu_hw_intr_ops {
 			struct dpu_hw_intr *intr,
 			int irq_idx,
 			bool clear);
-
-	/**
-	 * get_valid_interrupts - Gets a mask of all valid interrupt sources
-	 *                        within DPU. These are actually status bits
-	 *                        within interrupt registers that specify the
-	 *                        source of the interrupt in IRQs. For example,
-	 *                        valid interrupt sources can be MDP, DSI,
-	 *                        HDMI etc.
-	 * @intr:	HW interrupt handle
-	 * @mask:	Returning the interrupt source MASK
-	 * @return:	0 for success, otherwise failure
-	 */
-	int (*get_valid_interrupts)(
-			struct dpu_hw_intr *intr,
-			uint32_t *mask);
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index 68c54d2c9677..1ab8d4a889f7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -258,12 +258,6 @@ enum dpu_vbif {
 	VBIF_NRT = VBIF_1
 };
 
-enum dpu_iommu_domain {
-	DPU_IOMMU_DOMAIN_UNSECURE,
-	DPU_IOMMU_DOMAIN_SECURE,
-	DPU_IOMMU_DOMAIN_MAX
-};
-
 /**
  * DPU HW,Component order color map
  */
@@ -358,7 +352,6 @@ enum dpu_3d_blend_mode {
  * @alpha_enable: whether the format has an alpha channel
  * @num_planes: number of planes (including meta data planes)
  * @fetch_mode: linear, tiled, or ubwc hw fetch behavior
- * @is_yuv: is format a yuv variant
  * @flag: usage bit flags
  * @tile_width: format tile width
  * @tile_height: format tile height
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
index 321fc64ddd0e..efe70c508ee0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
@@ -18,7 +18,6 @@
 #include "dpu_hw_mdss.h"
 
 #define REG_MASK(n)                     ((BIT(n)) - 1)
-struct dpu_format_extended;
 
 /*
  * This is the common struct maintained by each sub block
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 4d67b3c96702..885bf88afa3e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -405,35 +405,38 @@ static void dpu_kms_wait_for_commit_done(struct msm_kms *kms,
 	}
 }
 
-static void _dpu_kms_initialize_dsi(struct drm_device *dev,
+static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 				    struct msm_drm_private *priv,
 				    struct dpu_kms *dpu_kms)
 {
 	struct drm_encoder *encoder = NULL;
-	int i, rc;
+	int i, rc = 0;
+
+	if (!(priv->dsi[0] || priv->dsi[1]))
+		return rc;
 
 	/*TODO: Support two independent DSI connectors */
 	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI);
-	if (IS_ERR_OR_NULL(encoder)) {
+	if (IS_ERR(encoder)) {
 		DPU_ERROR("encoder init failed for dsi display\n");
-		return;
+		return PTR_ERR(encoder);
 	}
 
 	priv->encoders[priv->num_encoders++] = encoder;
 
 	for (i = 0; i < ARRAY_SIZE(priv->dsi); i++) {
-		if (!priv->dsi[i]) {
-			DPU_DEBUG("invalid msm_dsi for ctrl %d\n", i);
-			return;
-		}
+		if (!priv->dsi[i])
+			continue;
 
 		rc = msm_dsi_modeset_init(priv->dsi[i], dev, encoder);
 		if (rc) {
 			DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n",
 				i, rc);
-			continue;
+			break;
 		}
 	}
+
+	return rc;
 }
 
 /**
@@ -444,16 +447,16 @@ static void _dpu_kms_initialize_dsi(struct drm_device *dev,
  * @dpu_kms:    Pointer to dpu kms structure
  * Returns:     Zero on success
  */
-static void _dpu_kms_setup_displays(struct drm_device *dev,
+static int _dpu_kms_setup_displays(struct drm_device *dev,
 				    struct msm_drm_private *priv,
 				    struct dpu_kms *dpu_kms)
 {
-	_dpu_kms_initialize_dsi(dev, priv, dpu_kms);
-
 	/**
 	 * Extend this function to initialize other
 	 * types of displays
 	 */
+
+	return _dpu_kms_initialize_dsi(dev, priv, dpu_kms);
 }
 
 static void _dpu_kms_drm_obj_destroy(struct dpu_kms *dpu_kms)
@@ -516,7 +519,9 @@ static int _dpu_kms_drm_obj_init(struct dpu_kms *dpu_kms)
 	 * Create encoder and query display drivers to create
 	 * bridges and connectors
 	 */
-	_dpu_kms_setup_displays(dev, priv, dpu_kms);
+	ret = _dpu_kms_setup_displays(dev, priv, dpu_kms);
+	if (ret)
+		goto fail;
 
 	max_crtc_count = min(catalog->mixer_count, priv->num_encoders);
 
@@ -627,6 +632,10 @@ static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms)
 		devm_iounmap(&dpu_kms->pdev->dev, dpu_kms->vbif[VBIF_RT]);
 	dpu_kms->vbif[VBIF_RT] = NULL;
 
+	if (dpu_kms->hw_mdp)
+		dpu_hw_mdp_destroy(dpu_kms->hw_mdp);
+	dpu_kms->hw_mdp = NULL;
+
 	if (dpu_kms->mmio)
 		devm_iounmap(&dpu_kms->pdev->dev, dpu_kms->mmio);
 	dpu_kms->mmio = NULL;
@@ -877,8 +886,7 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 		goto power_error;
 	}
 
-	rc = dpu_rm_init(&dpu_kms->rm, dpu_kms->catalog, dpu_kms->mmio,
-			dpu_kms->dev);
+	rc = dpu_rm_init(&dpu_kms->rm, dpu_kms->catalog, dpu_kms->mmio);
 	if (rc) {
 		DPU_ERROR("rm init failed: %d\n", rc);
 		goto power_error;
@@ -886,11 +894,10 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 
 	dpu_kms->rm_init = true;
 
-	dpu_kms->hw_mdp = dpu_rm_get_mdp(&dpu_kms->rm);
-	if (IS_ERR_OR_NULL(dpu_kms->hw_mdp)) {
+	dpu_kms->hw_mdp = dpu_hw_mdptop_init(MDP_TOP, dpu_kms->mmio,
+					     dpu_kms->catalog);
+	if (IS_ERR(dpu_kms->hw_mdp)) {
 		rc = PTR_ERR(dpu_kms->hw_mdp);
-		if (!dpu_kms->hw_mdp)
-			rc = -EINVAL;
 		DPU_ERROR("failed to get hw_mdp: %d\n", rc);
 		dpu_kms->hw_mdp = NULL;
 		goto power_error;
@@ -926,16 +933,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 		goto hw_intr_init_err;
 	}
 
-	/*
-	 * _dpu_kms_drm_obj_init should create the DRM related objects
-	 * i.e. CRTCs, planes, encoders, connectors and so forth
-	 */
-	rc = _dpu_kms_drm_obj_init(dpu_kms);
-	if (rc) {
-		DPU_ERROR("modeset init failed: %d\n", rc);
-		goto drm_obj_init_err;
-	}
-
 	dev->mode_config.min_width = 0;
 	dev->mode_config.min_height = 0;
 
@@ -952,6 +949,16 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 	 */
 	dev->mode_config.allow_fb_modifiers = true;
 
+	/*
+	 * _dpu_kms_drm_obj_init should create the DRM related objects
+	 * i.e. CRTCs, planes, encoders, connectors and so forth
+	 */
+	rc = _dpu_kms_drm_obj_init(dpu_kms);
+	if (rc) {
+		DPU_ERROR("modeset init failed: %d\n", rc);
+		goto drm_obj_init_err;
+	}
+
 	dpu_vbif_init_memtypes(dpu_kms);
 
 	pm_runtime_put_sync(&dpu_kms->pdev->dev);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
index cb307a2abf06..7316b4ab1b85 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
@@ -23,11 +23,14 @@ struct dpu_mdss {
 	struct dpu_irq_controller irq_controller;
 };
 
-static irqreturn_t dpu_mdss_irq(int irq, void *arg)
+static void dpu_mdss_irq(struct irq_desc *desc)
 {
-	struct dpu_mdss *dpu_mdss = arg;
+	struct dpu_mdss *dpu_mdss = irq_desc_get_handler_data(desc);
+	struct irq_chip *chip = irq_desc_get_chip(desc);
 	u32 interrupts;
 
+	chained_irq_enter(chip, desc);
+
 	interrupts = readl_relaxed(dpu_mdss->mmio + HW_INTR_STATUS);
 
 	while (interrupts) {
@@ -39,20 +42,20 @@ static irqreturn_t dpu_mdss_irq(int irq, void *arg)
 					   hwirq);
 		if (mapping == 0) {
 			DRM_ERROR("couldn't find irq mapping for %lu\n", hwirq);
-			return IRQ_NONE;
+			break;
 		}
 
 		rc = generic_handle_irq(mapping);
 		if (rc < 0) {
 			DRM_ERROR("handle irq fail: irq=%lu mapping=%u rc=%d\n",
 				  hwirq, mapping, rc);
-			return IRQ_NONE;
+			break;
 		}
 
 		interrupts &= ~(1 << hwirq);
 	}
 
-	return IRQ_HANDLED;
+	chained_irq_exit(chip, desc);
 }
 
 static void dpu_mdss_irq_mask(struct irq_data *irqd)
@@ -83,16 +86,16 @@ static struct irq_chip dpu_mdss_irq_chip = {
 	.irq_unmask = dpu_mdss_irq_unmask,
 };
 
+static struct lock_class_key dpu_mdss_lock_key, dpu_mdss_request_key;
+
 static int dpu_mdss_irqdomain_map(struct irq_domain *domain,
 		unsigned int irq, irq_hw_number_t hwirq)
 {
 	struct dpu_mdss *dpu_mdss = domain->host_data;
-	int ret;
 
+	irq_set_lockdep_class(irq, &dpu_mdss_lock_key, &dpu_mdss_request_key);
 	irq_set_chip_and_handler(irq, &dpu_mdss_irq_chip, handle_level_irq);
-	ret = irq_set_chip_data(irq, dpu_mdss);
-
-	return ret;
+	return irq_set_chip_data(irq, dpu_mdss);
 }
 
 static const struct irq_domain_ops dpu_mdss_irqdomain_ops = {
@@ -159,11 +162,13 @@ static void dpu_mdss_destroy(struct drm_device *dev)
 	struct msm_drm_private *priv = dev->dev_private;
 	struct dpu_mdss *dpu_mdss = to_dpu_mdss(priv->mdss);
 	struct dss_module_power *mp = &dpu_mdss->mp;
+	int irq;
 
 	pm_runtime_suspend(dev->dev);
 	pm_runtime_disable(dev->dev);
 	_dpu_mdss_irq_domain_fini(dpu_mdss);
-	free_irq(platform_get_irq(pdev, 0), dpu_mdss);
+	irq = platform_get_irq(pdev, 0);
+	irq_set_chained_handler_and_data(irq, NULL, NULL);
 	msm_dss_put_clk(mp->clk_config, mp->num_clk);
 	devm_kfree(&pdev->dev, mp->clk_config);
 
@@ -187,6 +192,7 @@ int dpu_mdss_init(struct drm_device *dev)
 	struct dpu_mdss *dpu_mdss;
 	struct dss_module_power *mp;
 	int ret = 0;
+	int irq;
 
 	dpu_mdss = devm_kzalloc(dev->dev, sizeof(*dpu_mdss), GFP_KERNEL);
 	if (!dpu_mdss)
@@ -219,12 +225,12 @@ int dpu_mdss_init(struct drm_device *dev)
 	if (ret)
 		goto irq_domain_error;
 
-	ret = request_irq(platform_get_irq(pdev, 0),
-			dpu_mdss_irq, 0, "dpu_mdss_isr", dpu_mdss);
-	if (ret) {
-		DPU_ERROR("failed to init irq: %d\n", ret);
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
 		goto irq_error;
-	}
+
+	irq_set_chained_handler_and_data(irq, dpu_mdss_irq,
+					 dpu_mdss);
 
 	pm_runtime_enable(dev->dev);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index fd75870eb17f..b01183b309b9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -95,8 +95,6 @@ struct dpu_plane {
 
 	enum dpu_sspp pipe;
 	uint32_t features;      /* capabilities from catalog */
-	uint32_t nformats;
-	uint32_t formats[64];
 
 	struct dpu_hw_pipe *pipe_hw;
 	struct dpu_hw_pipe_cfg pipe_cfg;
@@ -121,6 +119,12 @@ struct dpu_plane {
 	bool debugfs_default_scale;
 };
 
+static const uint64_t supported_format_modifiers[] = {
+	DRM_FORMAT_MOD_QCOM_COMPRESSED,
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID
+};
+
 #define to_dpu_plane(x) container_of(x, struct dpu_plane, base)
 
 static struct dpu_kms *_dpu_plane_get_kms(struct drm_plane *plane)
@@ -365,19 +369,6 @@ static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 			&pdpu->pipe_qos_cfg);
 }
 
-static void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable)
-{
-	struct dpu_plane *pdpu = to_dpu_plane(plane);
-	struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane);
-
-	if (!pdpu->is_rt_pipe)
-		return;
-
-	pm_runtime_get_sync(&dpu_kms->pdev->dev);
-	_dpu_plane_set_qos_ctrl(plane, enable, DPU_PLANE_QOS_PANIC_CTRL);
-	pm_runtime_put_sync(&dpu_kms->pdev->dev);
-}
-
 /**
  * _dpu_plane_set_ot_limit - set OT limit for the given plane
  * @plane:		Pointer to drm plane
@@ -1248,6 +1239,19 @@ static void dpu_plane_reset(struct drm_plane *plane)
 }
 
 #ifdef CONFIG_DEBUG_FS
+static void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable)
+{
+	struct dpu_plane *pdpu = to_dpu_plane(plane);
+	struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane);
+
+	if (!pdpu->is_rt_pipe)
+		return;
+
+	pm_runtime_get_sync(&dpu_kms->pdev->dev);
+	_dpu_plane_set_qos_ctrl(plane, enable, DPU_PLANE_QOS_PANIC_CTRL);
+	pm_runtime_put_sync(&dpu_kms->pdev->dev);
+}
+
 static ssize_t _dpu_plane_danger_read(struct file *file,
 			char __user *buff, size_t count, loff_t *ppos)
 {
@@ -1410,6 +1414,23 @@ static void dpu_plane_early_unregister(struct drm_plane *plane)
 	debugfs_remove_recursive(pdpu->debugfs_root);
 }
 
+static bool dpu_plane_format_mod_supported(struct drm_plane *plane,
+		uint32_t format, uint64_t modifier)
+{
+	if (modifier == DRM_FORMAT_MOD_LINEAR)
+		return true;
+
+	if (modifier == DRM_FORMAT_MOD_QCOM_COMPRESSED) {
+		int i;
+		for (i = 0; i < ARRAY_SIZE(qcom_compressed_supported_formats); i++) {
+			if (format == qcom_compressed_supported_formats[i])
+				return true;
+		}
+	}
+
+	return false;
+}
+
 static const struct drm_plane_funcs dpu_plane_funcs = {
 		.update_plane = drm_atomic_helper_update_plane,
 		.disable_plane = drm_atomic_helper_disable_plane,
@@ -1419,6 +1440,7 @@ static const struct drm_plane_funcs dpu_plane_funcs = {
 		.atomic_destroy_state = dpu_plane_destroy_state,
 		.late_register = dpu_plane_late_register,
 		.early_unregister = dpu_plane_early_unregister,
+		.format_mod_supported = dpu_plane_format_mod_supported,
 };
 
 static const struct drm_plane_helper_funcs dpu_plane_helper_funcs = {
@@ -1444,11 +1466,12 @@ struct drm_plane *dpu_plane_init(struct drm_device *dev,
 		unsigned long possible_crtcs, u32 master_plane_id)
 {
 	struct drm_plane *plane = NULL, *master_plane = NULL;
-	const struct dpu_format_extended *format_list;
+	const uint32_t *format_list;
 	struct dpu_plane *pdpu;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct dpu_kms *kms = to_dpu_kms(priv->kms);
 	int zpos_max = DPU_ZPOS_MAX;
+	uint32_t num_formats;
 	int ret = -EINVAL;
 
 	/* create and zero local structure */
@@ -1491,24 +1514,18 @@ struct drm_plane *dpu_plane_init(struct drm_device *dev,
 		goto clean_sspp;
 	}
 
-	if (!master_plane_id)
-		format_list = pdpu->pipe_sblk->format_list;
-	else
+	if (pdpu->is_virtual) {
 		format_list = pdpu->pipe_sblk->virt_format_list;
-
-	pdpu->nformats = dpu_populate_formats(format_list,
-				pdpu->formats,
-				0,
-				ARRAY_SIZE(pdpu->formats));
-
-	if (!pdpu->nformats) {
-		DPU_ERROR("[%u]no valid formats for plane\n", pipe);
-		goto clean_sspp;
+		num_formats = pdpu->pipe_sblk->virt_num_formats;
+	}
+	else {
+		format_list = pdpu->pipe_sblk->format_list;
+		num_formats = pdpu->pipe_sblk->num_formats;
 	}
 
 	ret = drm_universal_plane_init(dev, plane, 0xff, &dpu_plane_funcs,
-				pdpu->formats, pdpu->nformats,
-				NULL, type, NULL);
+				format_list, num_formats,
+				supported_format_modifiers, type, NULL);
 	if (ret)
 		goto clean_sspp;
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
index 7fed0b627708..0e6063acd041 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
@@ -28,23 +28,18 @@
 /**
  * struct dpu_plane_state: Define dpu extension of drm plane state object
  * @base:	base drm plane state object
- * @property_state: Local storage for msm_prop properties
- * @property_values:	cached plane property values
  * @aspace:	pointer to address space for input/output buffers
- * @input_fence:	dereferenced input fence pointer
  * @stage:	assigned by crtc blender
  * @multirect_index: index of the rectangle of SSPP
  * @multirect_mode: parallel or time multiplex multirect mode
  * @pending:	whether the current update is still pending
  * @scaler3_cfg: configuration data for scaler3
  * @pixel_ext: configuration data for pixel extensions
- * @scaler_check_state: indicates status of user provided pixel extension data
  * @cdp_cfg:	CDP configuration
  */
 struct dpu_plane_state {
 	struct drm_plane_state base;
 	struct msm_gem_address_space *aspace;
-	void *input_fence;
 	enum dpu_stage stage;
 	uint32_t multirect_index;
 	uint32_t multirect_mode;
@@ -107,12 +102,6 @@ void dpu_plane_restore(struct drm_plane *plane);
 void dpu_plane_flush(struct drm_plane *plane);
 
 /**
- * dpu_plane_kickoff - final plane operations before commit kickoff
- * @plane: Pointer to drm plane structure
- */
-void dpu_plane_kickoff(struct drm_plane *plane);
-
-/**
  * dpu_plane_set_error: enable/disable error condition
  * @plane: pointer to drm_plane structure
  */
@@ -147,14 +136,6 @@ int dpu_plane_validate_multirect_v2(struct dpu_multirect_plane_states *plane);
 void dpu_plane_clear_multirect(const struct drm_plane_state *drm_state);
 
 /**
- * dpu_plane_wait_input_fence - wait for input fence object
- * @plane:   Pointer to DRM plane object
- * @wait_ms: Wait timeout value
- * Returns: Zero on success
- */
-int dpu_plane_wait_input_fence(struct drm_plane *plane, uint32_t wait_ms);
-
-/**
  * dpu_plane_color_fill - enables color fill on plane
  * @plane:  Pointer to DRM plane object
  * @color:  RGB fill color value, [23..16] Blue, [15..8] Green, [7..0] Red
@@ -164,12 +145,4 @@ int dpu_plane_wait_input_fence(struct drm_plane *plane, uint32_t wait_ms);
 int dpu_plane_color_fill(struct drm_plane *plane,
 		uint32_t color, uint32_t alpha);
 
-/**
- * dpu_plane_set_revalidate - sets revalidate flag which forces a full
- *	validation of the plane properties in the next atomic check
- * @plane: Pointer to DRM plane object
- * @enable: Boolean to set/unset the flag
- */
-void dpu_plane_set_revalidate(struct drm_plane *plane, bool enable);
-
 #endif /* _DPU_PLANE_H_ */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index bdb117709674..037d9f4187f9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -21,8 +21,8 @@
 #include "dpu_encoder.h"
 #include "dpu_trace.h"
 
-#define RESERVED_BY_OTHER(h, r) \
-	((h)->rsvp && ((h)->rsvp->enc_id != (r)->enc_id))
+#define RESERVED_BY_OTHER(h, r)  \
+		((h)->enc_id && (h)->enc_id != r)
 
 /**
  * struct dpu_rm_requirements - Reservation requirements parameter bundle
@@ -34,90 +34,21 @@ struct dpu_rm_requirements {
 	struct dpu_encoder_hw_resources hw_res;
 };
 
-/**
- * struct dpu_rm_rsvp - Use Case Reservation tagging structure
- *	Used to tag HW blocks as reserved by a CRTC->Encoder->Connector chain
- *	By using as a tag, rather than lists of pointers to HW blocks used
- *	we can avoid some list management since we don't know how many blocks
- *	of each type a given use case may require.
- * @list:	List head for list of all reservations
- * @seq:	Global RSVP sequence number for debugging, especially for
- *		differentiating differenct allocations for same encoder.
- * @enc_id:	Reservations are tracked by Encoder DRM object ID.
- *		CRTCs may be connected to multiple Encoders.
- *		An encoder or connector id identifies the display path.
- */
-struct dpu_rm_rsvp {
-	struct list_head list;
-	uint32_t seq;
-	uint32_t enc_id;
-};
 
 /**
  * struct dpu_rm_hw_blk - hardware block tracking list member
  * @list:	List head for list of all hardware blocks tracking items
- * @rsvp:	Pointer to use case reservation if reserved by a client
- * @rsvp_nxt:	Temporary pointer used during reservation to the incoming
- *		request. Will be swapped into rsvp if proposal is accepted
- * @type:	Type of hardware block this structure tracks
  * @id:		Hardware ID number, within it's own space, ie. LM_X
- * @catalog:	Pointer to the hardware catalog entry for this block
+ * @enc_id:	Encoder id to which this blk is binded
  * @hw:		Pointer to the hardware register access object for this block
  */
 struct dpu_rm_hw_blk {
 	struct list_head list;
-	struct dpu_rm_rsvp *rsvp;
-	struct dpu_rm_rsvp *rsvp_nxt;
-	enum dpu_hw_blk_type type;
 	uint32_t id;
+	uint32_t enc_id;
 	struct dpu_hw_blk *hw;
 };
 
-/**
- * dpu_rm_dbg_rsvp_stage - enum of steps in making reservation for event logging
- */
-enum dpu_rm_dbg_rsvp_stage {
-	DPU_RM_STAGE_BEGIN,
-	DPU_RM_STAGE_AFTER_CLEAR,
-	DPU_RM_STAGE_AFTER_RSVPNEXT,
-	DPU_RM_STAGE_FINAL
-};
-
-static void _dpu_rm_print_rsvps(
-		struct dpu_rm *rm,
-		enum dpu_rm_dbg_rsvp_stage stage)
-{
-	struct dpu_rm_rsvp *rsvp;
-	struct dpu_rm_hw_blk *blk;
-	enum dpu_hw_blk_type type;
-
-	DPU_DEBUG("%d\n", stage);
-
-	list_for_each_entry(rsvp, &rm->rsvps, list) {
-		DRM_DEBUG_KMS("%d rsvp[s%ue%u]\n", stage, rsvp->seq,
-			      rsvp->enc_id);
-	}
-
-	for (type = 0; type < DPU_HW_BLK_MAX; type++) {
-		list_for_each_entry(blk, &rm->hw_blks[type], list) {
-			if (!blk->rsvp && !blk->rsvp_nxt)
-				continue;
-
-			DRM_DEBUG_KMS("%d rsvp[s%ue%u->s%ue%u] %d %d\n", stage,
-				(blk->rsvp) ? blk->rsvp->seq : 0,
-				(blk->rsvp) ? blk->rsvp->enc_id : 0,
-				(blk->rsvp_nxt) ? blk->rsvp_nxt->seq : 0,
-				(blk->rsvp_nxt) ? blk->rsvp_nxt->enc_id : 0,
-				blk->type, blk->id);
-		}
-	}
-}
-
-struct dpu_hw_mdp *dpu_rm_get_mdp(struct dpu_rm *rm)
-{
-	return rm->hw_mdp;
-}
-
 void dpu_rm_init_hw_iter(
 		struct dpu_rm_hw_iter *iter,
 		uint32_t enc_id,
@@ -148,15 +79,7 @@ static bool _dpu_rm_get_hw_locked(struct dpu_rm *rm, struct dpu_rm_hw_iter *i)
 	i->blk = list_prepare_entry(i->blk, blk_list, list);
 
 	list_for_each_entry_continue(i->blk, blk_list, list) {
-		struct dpu_rm_rsvp *rsvp = i->blk->rsvp;
-
-		if (i->blk->type != i->type) {
-			DPU_ERROR("found incorrect block type %d on %d list\n",
-					i->blk->type, i->type);
-			return false;
-		}
-
-		if ((i->enc_id == 0) || (rsvp && rsvp->enc_id == i->enc_id)) {
+		if (i->enc_id == i->blk->enc_id) {
 			i->hw = i->blk->hw;
 			DPU_DEBUG("found type %d id %d for enc %d\n",
 					i->type, i->blk->id, i->enc_id);
@@ -208,34 +131,18 @@ static void _dpu_rm_hw_destroy(enum dpu_hw_blk_type type, void *hw)
 
 int dpu_rm_destroy(struct dpu_rm *rm)
 {
-
-	struct dpu_rm_rsvp *rsvp_cur, *rsvp_nxt;
 	struct dpu_rm_hw_blk *hw_cur, *hw_nxt;
 	enum dpu_hw_blk_type type;
 
-	if (!rm) {
-		DPU_ERROR("invalid rm\n");
-		return -EINVAL;
-	}
-
-	list_for_each_entry_safe(rsvp_cur, rsvp_nxt, &rm->rsvps, list) {
-		list_del(&rsvp_cur->list);
-		kfree(rsvp_cur);
-	}
-
-
 	for (type = 0; type < DPU_HW_BLK_MAX; type++) {
 		list_for_each_entry_safe(hw_cur, hw_nxt, &rm->hw_blks[type],
 				list) {
 			list_del(&hw_cur->list);
-			_dpu_rm_hw_destroy(hw_cur->type, hw_cur->hw);
+			_dpu_rm_hw_destroy(type, hw_cur->hw);
 			kfree(hw_cur);
 		}
 	}
 
-	dpu_hw_mdp_destroy(rm->hw_mdp);
-	rm->hw_mdp = NULL;
-
 	mutex_destroy(&rm->rm_lock);
 
 	return 0;
@@ -250,11 +157,8 @@ static int _dpu_rm_hw_blk_create(
 		void *hw_catalog_info)
 {
 	struct dpu_rm_hw_blk *blk;
-	struct dpu_hw_mdp *hw_mdp;
 	void *hw;
 
-	hw_mdp = rm->hw_mdp;
-
 	switch (type) {
 	case DPU_HW_BLK_LM:
 		hw = dpu_hw_lm_init(id, mmio, cat);
@@ -290,9 +194,9 @@ static int _dpu_rm_hw_blk_create(
 		return -ENOMEM;
 	}
 
-	blk->type = type;
 	blk->id = id;
 	blk->hw = hw;
+	blk->enc_id = 0;
 	list_add_tail(&blk->list, &rm->hw_blks[type]);
 
 	return 0;
@@ -300,13 +204,12 @@ static int _dpu_rm_hw_blk_create(
 
 int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_mdss_cfg *cat,
-		void __iomem *mmio,
-		struct drm_device *dev)
+		void __iomem *mmio)
 {
 	int rc, i;
 	enum dpu_hw_blk_type type;
 
-	if (!rm || !cat || !mmio || !dev) {
+	if (!rm || !cat || !mmio) {
 		DPU_ERROR("invalid kms\n");
 		return -EINVAL;
 	}
@@ -316,21 +219,9 @@ int dpu_rm_init(struct dpu_rm *rm,
 
 	mutex_init(&rm->rm_lock);
 
-	INIT_LIST_HEAD(&rm->rsvps);
 	for (type = 0; type < DPU_HW_BLK_MAX; type++)
 		INIT_LIST_HEAD(&rm->hw_blks[type]);
 
-	rm->dev = dev;
-
-	/* Some of the sub-blocks require an mdptop to be created */
-	rm->hw_mdp = dpu_hw_mdptop_init(MDP_TOP, mmio, cat);
-	if (IS_ERR_OR_NULL(rm->hw_mdp)) {
-		rc = PTR_ERR(rm->hw_mdp);
-		rm->hw_mdp = NULL;
-		DPU_ERROR("failed: mdp hw not available\n");
-		goto fail;
-	}
-
 	/* Interrogate HW catalog and create tracking items for hw blocks */
 	for (i = 0; i < cat->mixer_count; i++) {
 		struct dpu_lm_cfg *lm = &cat->mixer[i];
@@ -410,7 +301,7 @@ static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top)
  *	proposed use case requirements, incl. hardwired dependent blocks like
  *	pingpong
  * @rm: dpu resource manager handle
- * @rsvp: reservation currently being created
+ * @enc_id: encoder id requesting for allocation
  * @reqs: proposed use case requirements
  * @lm: proposed layer mixer, function checks if lm, and all other hardwired
  *      blocks connected to the lm (pp) is available and appropriate
@@ -422,7 +313,7 @@ static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top)
  */
 static bool _dpu_rm_check_lm_and_get_connected_blks(
 		struct dpu_rm *rm,
-		struct dpu_rm_rsvp *rsvp,
+		uint32_t enc_id,
 		struct dpu_rm_requirements *reqs,
 		struct dpu_rm_hw_blk *lm,
 		struct dpu_rm_hw_blk **pp,
@@ -449,7 +340,7 @@ static bool _dpu_rm_check_lm_and_get_connected_blks(
 	}
 
 	/* Already reserved? */
-	if (RESERVED_BY_OTHER(lm, rsvp)) {
+	if (RESERVED_BY_OTHER(lm, enc_id)) {
 		DPU_DEBUG("lm %d already reserved\n", lm_cfg->id);
 		return false;
 	}
@@ -467,7 +358,7 @@ static bool _dpu_rm_check_lm_and_get_connected_blks(
 		return false;
 	}
 
-	if (RESERVED_BY_OTHER(*pp, rsvp)) {
+	if (RESERVED_BY_OTHER(*pp, enc_id)) {
 		DPU_DEBUG("lm %d pp %d already reserved\n", lm->id,
 				(*pp)->id);
 		return false;
@@ -476,10 +367,8 @@ static bool _dpu_rm_check_lm_and_get_connected_blks(
 	return true;
 }
 
-static int _dpu_rm_reserve_lms(
-		struct dpu_rm *rm,
-		struct dpu_rm_rsvp *rsvp,
-		struct dpu_rm_requirements *reqs)
+static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id,
+			       struct dpu_rm_requirements *reqs)
 
 {
 	struct dpu_rm_hw_blk *lm[MAX_BLOCKS];
@@ -504,7 +393,7 @@ static int _dpu_rm_reserve_lms(
 		lm[lm_count] = iter_i.blk;
 
 		if (!_dpu_rm_check_lm_and_get_connected_blks(
-				rm, rsvp, reqs, lm[lm_count],
+				rm, enc_id, reqs, lm[lm_count],
 				&pp[lm_count], NULL))
 			continue;
 
@@ -519,7 +408,7 @@ static int _dpu_rm_reserve_lms(
 				continue;
 
 			if (!_dpu_rm_check_lm_and_get_connected_blks(
-					rm, rsvp, reqs, iter_j.blk,
+					rm, enc_id, reqs, iter_j.blk,
 					&pp[lm_count], iter_i.blk))
 				continue;
 
@@ -537,11 +426,10 @@ static int _dpu_rm_reserve_lms(
 		if (!lm[i])
 			break;
 
-		lm[i]->rsvp_nxt = rsvp;
-		pp[i]->rsvp_nxt = rsvp;
+		lm[i]->enc_id = enc_id;
+		pp[i]->enc_id = enc_id;
 
-		trace_dpu_rm_reserve_lms(lm[i]->id, lm[i]->type, rsvp->enc_id,
-					 pp[i]->id);
+		trace_dpu_rm_reserve_lms(lm[i]->id, enc_id, pp[i]->id);
 	}
 
 	return rc;
@@ -549,7 +437,7 @@ static int _dpu_rm_reserve_lms(
 
 static int _dpu_rm_reserve_ctls(
 		struct dpu_rm *rm,
-		struct dpu_rm_rsvp *rsvp,
+		uint32_t enc_id,
 		const struct msm_display_topology *top)
 {
 	struct dpu_rm_hw_blk *ctls[MAX_BLOCKS];
@@ -570,7 +458,7 @@ static int _dpu_rm_reserve_ctls(
 		unsigned long features = ctl->caps->features;
 		bool has_split_display;
 
-		if (RESERVED_BY_OTHER(iter.blk, rsvp))
+		if (RESERVED_BY_OTHER(iter.blk, enc_id))
 			continue;
 
 		has_split_display = BIT(DPU_CTL_SPLIT_DISPLAY) & features;
@@ -591,9 +479,8 @@ static int _dpu_rm_reserve_ctls(
 		return -ENAVAIL;
 
 	for (i = 0; i < ARRAY_SIZE(ctls) && i < num_ctls; i++) {
-		ctls[i]->rsvp_nxt = rsvp;
-		trace_dpu_rm_reserve_ctls(ctls[i]->id, ctls[i]->type,
-					  rsvp->enc_id);
+		ctls[i]->enc_id = enc_id;
+		trace_dpu_rm_reserve_ctls(ctls[i]->id, enc_id);
 	}
 
 	return 0;
@@ -601,7 +488,7 @@ static int _dpu_rm_reserve_ctls(
 
 static int _dpu_rm_reserve_intf(
 		struct dpu_rm *rm,
-		struct dpu_rm_rsvp *rsvp,
+		uint32_t enc_id,
 		uint32_t id,
 		enum dpu_hw_blk_type type)
 {
@@ -614,14 +501,13 @@ static int _dpu_rm_reserve_intf(
 		if (iter.blk->id != id)
 			continue;
 
-		if (RESERVED_BY_OTHER(iter.blk, rsvp)) {
+		if (RESERVED_BY_OTHER(iter.blk, enc_id)) {
 			DPU_ERROR("type %d id %d already reserved\n", type, id);
 			return -ENAVAIL;
 		}
 
-		iter.blk->rsvp_nxt = rsvp;
-		trace_dpu_rm_reserve_intf(iter.blk->id, iter.blk->type,
-					  rsvp->enc_id);
+		iter.blk->enc_id = enc_id;
+		trace_dpu_rm_reserve_intf(iter.blk->id, enc_id);
 		break;
 	}
 
@@ -636,7 +522,7 @@ static int _dpu_rm_reserve_intf(
 
 static int _dpu_rm_reserve_intf_related_hw(
 		struct dpu_rm *rm,
-		struct dpu_rm_rsvp *rsvp,
+		uint32_t enc_id,
 		struct dpu_encoder_hw_resources *hw_res)
 {
 	int i, ret = 0;
@@ -646,7 +532,7 @@ static int _dpu_rm_reserve_intf_related_hw(
 		if (hw_res->intfs[i] == INTF_MODE_NONE)
 			continue;
 		id = i + INTF_0;
-		ret = _dpu_rm_reserve_intf(rm, rsvp, id,
+		ret = _dpu_rm_reserve_intf(rm, enc_id, id,
 				DPU_HW_BLK_INTF);
 		if (ret)
 			return ret;
@@ -655,33 +541,27 @@ static int _dpu_rm_reserve_intf_related_hw(
 	return ret;
 }
 
-static int _dpu_rm_make_next_rsvp(
+static int _dpu_rm_make_reservation(
 		struct dpu_rm *rm,
 		struct drm_encoder *enc,
 		struct drm_crtc_state *crtc_state,
-		struct dpu_rm_rsvp *rsvp,
 		struct dpu_rm_requirements *reqs)
 {
 	int ret;
 
-	/* Create reservation info, tag reserved blocks with it as we go */
-	rsvp->seq = ++rm->rsvp_next_seq;
-	rsvp->enc_id = enc->base.id;
-	list_add_tail(&rsvp->list, &rm->rsvps);
-
-	ret = _dpu_rm_reserve_lms(rm, rsvp, reqs);
+	ret = _dpu_rm_reserve_lms(rm, enc->base.id, reqs);
 	if (ret) {
 		DPU_ERROR("unable to find appropriate mixers\n");
 		return ret;
 	}
 
-	ret = _dpu_rm_reserve_ctls(rm, rsvp, &reqs->topology);
+	ret = _dpu_rm_reserve_ctls(rm, enc->base.id, &reqs->topology);
 	if (ret) {
 		DPU_ERROR("unable to find appropriate CTL\n");
 		return ret;
 	}
 
-	ret = _dpu_rm_reserve_intf_related_hw(rm, rsvp, &reqs->hw_res);
+	ret = _dpu_rm_reserve_intf_related_hw(rm, enc->base.id, &reqs->hw_res);
 	if (ret)
 		return ret;
 
@@ -706,108 +586,31 @@ static int _dpu_rm_populate_requirements(
 	return 0;
 }
 
-static struct dpu_rm_rsvp *_dpu_rm_get_rsvp(
-		struct dpu_rm *rm,
-		struct drm_encoder *enc)
+static void _dpu_rm_release_reservation(struct dpu_rm *rm, uint32_t enc_id)
 {
-	struct dpu_rm_rsvp *i;
-
-	if (!rm || !enc) {
-		DPU_ERROR("invalid params\n");
-		return NULL;
-	}
-
-	if (list_empty(&rm->rsvps))
-		return NULL;
-
-	list_for_each_entry(i, &rm->rsvps, list)
-		if (i->enc_id == enc->base.id)
-			return i;
-
-	return NULL;
-}
-
-/**
- * _dpu_rm_release_rsvp - release resources and release a reservation
- * @rm:	KMS handle
- * @rsvp:	RSVP pointer to release and release resources for
- */
-static void _dpu_rm_release_rsvp(struct dpu_rm *rm, struct dpu_rm_rsvp *rsvp)
-{
-	struct dpu_rm_rsvp *rsvp_c, *rsvp_n;
 	struct dpu_rm_hw_blk *blk;
 	enum dpu_hw_blk_type type;
 
-	if (!rsvp)
-		return;
-
-	DPU_DEBUG("rel rsvp %d enc %d\n", rsvp->seq, rsvp->enc_id);
-
-	list_for_each_entry_safe(rsvp_c, rsvp_n, &rm->rsvps, list) {
-		if (rsvp == rsvp_c) {
-			list_del(&rsvp_c->list);
-			break;
-		}
-	}
-
 	for (type = 0; type < DPU_HW_BLK_MAX; type++) {
 		list_for_each_entry(blk, &rm->hw_blks[type], list) {
-			if (blk->rsvp == rsvp) {
-				blk->rsvp = NULL;
-				DPU_DEBUG("rel rsvp %d enc %d %d %d\n",
-						rsvp->seq, rsvp->enc_id,
-						blk->type, blk->id);
-			}
-			if (blk->rsvp_nxt == rsvp) {
-				blk->rsvp_nxt = NULL;
-				DPU_DEBUG("rel rsvp_nxt %d enc %d %d %d\n",
-						rsvp->seq, rsvp->enc_id,
-						blk->type, blk->id);
+			if (blk->enc_id == enc_id) {
+				blk->enc_id = 0;
+				DPU_DEBUG("rel enc %d %d %d\n", enc_id,
+					  type, blk->id);
 			}
 		}
 	}
-
-	kfree(rsvp);
 }
 
 void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc)
 {
-	struct dpu_rm_rsvp *rsvp;
-
-	if (!rm || !enc) {
-		DPU_ERROR("invalid params\n");
-		return;
-	}
-
 	mutex_lock(&rm->rm_lock);
 
-	rsvp = _dpu_rm_get_rsvp(rm, enc);
-	if (!rsvp) {
-		DPU_ERROR("failed to find rsvp for enc %d\n", enc->base.id);
-		goto end;
-	}
+	_dpu_rm_release_reservation(rm, enc->base.id);
 
-	_dpu_rm_release_rsvp(rm, rsvp);
-end:
 	mutex_unlock(&rm->rm_lock);
 }
 
-static void _dpu_rm_commit_rsvp(struct dpu_rm *rm, struct dpu_rm_rsvp *rsvp)
-{
-	struct dpu_rm_hw_blk *blk;
-	enum dpu_hw_blk_type type;
-
-	/* Swap next rsvp to be the active */
-	for (type = 0; type < DPU_HW_BLK_MAX; type++) {
-		list_for_each_entry(blk, &rm->hw_blks[type], list) {
-			if (blk->rsvp_nxt) {
-				blk->rsvp = blk->rsvp_nxt;
-				blk->rsvp_nxt = NULL;
-			}
-		}
-	}
-}
-
 int dpu_rm_reserve(
 		struct dpu_rm *rm,
 		struct drm_encoder *enc,
@@ -815,7 +618,6 @@ int dpu_rm_reserve(
 		struct msm_display_topology topology,
 		bool test_only)
 {
-	struct dpu_rm_rsvp *rsvp_cur, *rsvp_nxt;
 	struct dpu_rm_requirements reqs;
 	int ret;
 
@@ -828,8 +630,6 @@ int dpu_rm_reserve(
 
 	mutex_lock(&rm->rm_lock);
 
-	_dpu_rm_print_rsvps(rm, DPU_RM_STAGE_BEGIN);
-
 	ret = _dpu_rm_populate_requirements(rm, enc, crtc_state, &reqs,
 					    topology);
 	if (ret) {
@@ -837,50 +637,17 @@ int dpu_rm_reserve(
 		goto end;
 	}
 
-	/*
-	 * We only support one active reservation per-hw-block. But to implement
-	 * transactional semantics for test-only, and for allowing failure while
-	 * modifying your existing reservation, over the course of this
-	 * function we can have two reservations:
-	 * Current: Existing reservation
-	 * Next: Proposed reservation. The proposed reservation may fail, or may
-	 *       be discarded if in test-only mode.
-	 * If reservation is successful, and we're not in test-only, then we
-	 * replace the current with the next.
-	 */
-	rsvp_nxt = kzalloc(sizeof(*rsvp_nxt), GFP_KERNEL);
-	if (!rsvp_nxt) {
-		ret = -ENOMEM;
-		goto end;
-	}
-
-	rsvp_cur = _dpu_rm_get_rsvp(rm, enc);
-
-	/* Check the proposed reservation, store it in hw's "next" field */
-	ret = _dpu_rm_make_next_rsvp(rm, enc, crtc_state, rsvp_nxt, &reqs);
-
-	_dpu_rm_print_rsvps(rm, DPU_RM_STAGE_AFTER_RSVPNEXT);
-
+	ret = _dpu_rm_make_reservation(rm, enc, crtc_state, &reqs);
 	if (ret) {
 		DPU_ERROR("failed to reserve hw resources: %d\n", ret);
-		_dpu_rm_release_rsvp(rm, rsvp_nxt);
+		_dpu_rm_release_reservation(rm, enc->base.id);
 	} else if (test_only) {
-		/*
-		 * Normally, if test_only, test the reservation and then undo
-		 * However, if the user requests LOCK, then keep the reservation
-		 * made during the atomic_check phase.
-		 */
-		DPU_DEBUG("test_only: discard test rsvp[s%de%d]\n",
-				rsvp_nxt->seq, rsvp_nxt->enc_id);
-		_dpu_rm_release_rsvp(rm, rsvp_nxt);
-	} else {
-		_dpu_rm_release_rsvp(rm, rsvp_cur);
-
-		_dpu_rm_commit_rsvp(rm, rsvp_nxt);
+		 /* test_only: test the reservation and then undo */
+		DPU_DEBUG("test_only: discard test [enc: %d]\n",
+				enc->base.id);
+		_dpu_rm_release_reservation(rm, enc->base.id);
 	}
 
-	_dpu_rm_print_rsvps(rm, DPU_RM_STAGE_FINAL);
-
 end:
 	mutex_unlock(&rm->rm_lock);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
index b8273bd23801..381611fc5877 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
@@ -22,22 +22,14 @@
 
 /**
  * struct dpu_rm - DPU dynamic hardware resource manager
- * @dev: device handle for event logging purposes
- * @rsvps: list of hardware reservations by each crtc->encoder->connector
  * @hw_blks: array of lists of hardware resources present in the system, one
  *	list per type of hardware block
- * @hw_mdp: hardware object for mdp_top
  * @lm_max_width: cached layer mixer maximum width
- * @rsvp_next_seq: sequence number for next reservation for debugging purposes
  * @rm_lock: resource manager mutex
  */
 struct dpu_rm {
-	struct drm_device *dev;
-	struct list_head rsvps;
 	struct list_head hw_blks[DPU_HW_BLK_MAX];
-	struct dpu_hw_mdp *hw_mdp;
 	uint32_t lm_max_width;
-	uint32_t rsvp_next_seq;
 	struct mutex rm_lock;
 };
 
@@ -67,13 +59,11 @@ struct dpu_rm_hw_iter {
  * @rm: DPU Resource Manager handle
  * @cat: Pointer to hardware catalog
  * @mmio: mapped register io address of MDP
- * @dev: device handle for event logging purposes
  * @Return: 0 on Success otherwise -ERROR
  */
 int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_mdss_cfg *cat,
-		void __iomem *mmio,
-		struct drm_device *dev);
+		void __iomem *mmio);
 
 /**
  * dpu_rm_destroy - Free all memory allocated by dpu_rm_init
@@ -112,14 +102,6 @@ int dpu_rm_reserve(struct dpu_rm *rm,
 void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc);
 
 /**
- * dpu_rm_get_mdp - Retrieve HW block for MDP TOP.
- *	This is never reserved, and is usable by any display.
- * @rm: DPU Resource Manager handle
- * @Return: Pointer to hw block or NULL
- */
-struct dpu_hw_mdp *dpu_rm_get_mdp(struct dpu_rm *rm);
-
-/**
  * dpu_rm_init_hw_iter - setup given iterator for new iteration over hw list
  *	using dpu_rm_get_hw
  * @iter: iter object to initialize
@@ -144,12 +126,4 @@ void dpu_rm_init_hw_iter(
  * @Return: true on match found, false on no match found
  */
 bool dpu_rm_get_hw(struct dpu_rm *rm, struct dpu_rm_hw_iter *iter);
-
-/**
- * dpu_rm_check_property_topctl - validate property bitmask before it is set
- * @val: user's proposed topology control bitmask
- * @Return: 0 on success or error
- */
-int dpu_rm_check_property_topctl(uint64_t val);
-
 #endif /* __DPU_RM_H__ */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
index c78b521ceda1..8bb46090bd16 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
@@ -831,48 +831,42 @@ TRACE_EVENT(dpu_plane_disable,
 );
 
 DECLARE_EVENT_CLASS(dpu_rm_iter_template,
-	TP_PROTO(uint32_t id, enum dpu_hw_blk_type type, uint32_t enc_id),
-	TP_ARGS(id, type, enc_id),
+	TP_PROTO(uint32_t id, uint32_t enc_id),
+	TP_ARGS(id, enc_id),
 	TP_STRUCT__entry(
 		__field(	uint32_t,		id	)
-		__field(	enum dpu_hw_blk_type,	type	)
 		__field(	uint32_t,		enc_id	)
 	),
 	TP_fast_assign(
 		__entry->id = id;
-		__entry->type = type;
 		__entry->enc_id = enc_id;
 	),
-	TP_printk("id:%d type:%d enc_id:%u", __entry->id, __entry->type,
-		  __entry->enc_id)
+	TP_printk("id:%d enc_id:%u", __entry->id, __entry->enc_id)
 );
 DEFINE_EVENT(dpu_rm_iter_template, dpu_rm_reserve_intf,
-	TP_PROTO(uint32_t id, enum dpu_hw_blk_type type, uint32_t enc_id),
-	TP_ARGS(id, type, enc_id)
+	TP_PROTO(uint32_t id, uint32_t enc_id),
+	TP_ARGS(id, enc_id)
 );
 DEFINE_EVENT(dpu_rm_iter_template, dpu_rm_reserve_ctls,
-	TP_PROTO(uint32_t id, enum dpu_hw_blk_type type, uint32_t enc_id),
-	TP_ARGS(id, type, enc_id)
+	TP_PROTO(uint32_t id, uint32_t enc_id),
+	TP_ARGS(id, enc_id)
 );
 
 TRACE_EVENT(dpu_rm_reserve_lms,
-	TP_PROTO(uint32_t id, enum dpu_hw_blk_type type, uint32_t enc_id,
-		 uint32_t pp_id),
-	TP_ARGS(id, type, enc_id, pp_id),
+	TP_PROTO(uint32_t id, uint32_t enc_id, uint32_t pp_id),
+	TP_ARGS(id, enc_id, pp_id),
 	TP_STRUCT__entry(
 		__field(	uint32_t,		id	)
-		__field(	enum dpu_hw_blk_type,	type	)
 		__field(	uint32_t,		enc_id	)
 		__field(	uint32_t,		pp_id	)
 	),
 	TP_fast_assign(
 		__entry->id = id;
-		__entry->type = type;
 		__entry->enc_id = enc_id;
 		__entry->pp_id = pp_id;
 	),
-	TP_printk("id:%d type:%d enc_id:%u pp_id:%u", __entry->id,
-		  __entry->type, __entry->enc_id, __entry->pp_id)
+	TP_printk("id:%d enc_id:%u pp_id:%u", __entry->id,
+		  __entry->enc_id, __entry->pp_id)
 );
 
 TRACE_EVENT(dpu_vbif_wait_xin_halt_fail,
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c
index cc32ea5f4289..0cfd4c06b610 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c
@@ -16,9 +16,9 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_flip_work.h>
 #include <drm/drm_mode.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp4_kms.h"
 
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c
index 48ce218b8463..caa39b4621e3 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c
@@ -18,7 +18,7 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp4_kms.h"
 
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 ff8f2da160c0..259d51971401 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c
@@ -16,7 +16,7 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "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 fff77a4b12c2..df6f9803a1d7 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
@@ -17,7 +17,7 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp4_kms.h"
 
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 976585d8bfd6..9bf9d6065c55 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
@@ -12,7 +12,7 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp5_kms.h"
 
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
index 2f95e6525589..b0cf63c4e3d7 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
@@ -19,8 +19,8 @@
 #include <linux/sort.h>
 #include <drm/drm_mode.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_flip_work.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp5_kms.h"
 
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
index d2a56e55e53d..820a62c40063 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c
@@ -17,7 +17,7 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "mdp5_kms.h"
 
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index d2cdc7b553fe..0bdd93648761 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -207,62 +207,44 @@ u32 msm_readl(const void __iomem *addr)
 	return val;
 }
 
-struct vblank_event {
-	struct list_head node;
+struct msm_vblank_work {
+	struct work_struct work;
 	int crtc_id;
 	bool enable;
+	struct msm_drm_private *priv;
 };
 
-static void vblank_ctrl_worker(struct kthread_work *work)
+static void vblank_ctrl_worker(struct work_struct *work)
 {
-	struct msm_vblank_ctrl *vbl_ctrl = container_of(work,
-						struct msm_vblank_ctrl, work);
-	struct msm_drm_private *priv = container_of(vbl_ctrl,
-					struct msm_drm_private, vblank_ctrl);
+	struct msm_vblank_work *vbl_work = container_of(work,
+						struct msm_vblank_work, work);
+	struct msm_drm_private *priv = vbl_work->priv;
 	struct msm_kms *kms = priv->kms;
-	struct vblank_event *vbl_ev, *tmp;
-	unsigned long flags;
-
-	spin_lock_irqsave(&vbl_ctrl->lock, flags);
-	list_for_each_entry_safe(vbl_ev, tmp, &vbl_ctrl->event_list, node) {
-		list_del(&vbl_ev->node);
-		spin_unlock_irqrestore(&vbl_ctrl->lock, flags);
-
-		if (vbl_ev->enable)
-			kms->funcs->enable_vblank(kms,
-						priv->crtcs[vbl_ev->crtc_id]);
-		else
-			kms->funcs->disable_vblank(kms,
-						priv->crtcs[vbl_ev->crtc_id]);
-
-		kfree(vbl_ev);
 
-		spin_lock_irqsave(&vbl_ctrl->lock, flags);
-	}
+	if (vbl_work->enable)
+		kms->funcs->enable_vblank(kms, priv->crtcs[vbl_work->crtc_id]);
+	else
+		kms->funcs->disable_vblank(kms,	priv->crtcs[vbl_work->crtc_id]);
 
-	spin_unlock_irqrestore(&vbl_ctrl->lock, flags);
+	kfree(vbl_work);
 }
 
 static int vblank_ctrl_queue_work(struct msm_drm_private *priv,
 					int crtc_id, bool enable)
 {
-	struct msm_vblank_ctrl *vbl_ctrl = &priv->vblank_ctrl;
-	struct vblank_event *vbl_ev;
-	unsigned long flags;
+	struct msm_vblank_work *vbl_work;
 
-	vbl_ev = kzalloc(sizeof(*vbl_ev), GFP_ATOMIC);
-	if (!vbl_ev)
+	vbl_work = kzalloc(sizeof(*vbl_work), GFP_ATOMIC);
+	if (!vbl_work)
 		return -ENOMEM;
 
-	vbl_ev->crtc_id = crtc_id;
-	vbl_ev->enable = enable;
+	INIT_WORK(&vbl_work->work, vblank_ctrl_worker);
 
-	spin_lock_irqsave(&vbl_ctrl->lock, flags);
-	list_add_tail(&vbl_ev->node, &vbl_ctrl->event_list);
-	spin_unlock_irqrestore(&vbl_ctrl->lock, flags);
+	vbl_work->crtc_id = crtc_id;
+	vbl_work->enable = enable;
+	vbl_work->priv = priv;
 
-	kthread_queue_work(&priv->disp_thread[crtc_id].worker,
-			&vbl_ctrl->work);
+	queue_work(priv->wq, &vbl_work->work);
 
 	return 0;
 }
@@ -274,31 +256,20 @@ static int msm_drm_uninit(struct device *dev)
 	struct msm_drm_private *priv = ddev->dev_private;
 	struct msm_kms *kms = priv->kms;
 	struct msm_mdss *mdss = priv->mdss;
-	struct msm_vblank_ctrl *vbl_ctrl = &priv->vblank_ctrl;
-	struct vblank_event *vbl_ev, *tmp;
 	int i;
 
 	/* We must cancel and cleanup any pending vblank enable/disable
 	 * work before drm_irq_uninstall() to avoid work re-enabling an
 	 * irq after uninstall has disabled it.
 	 */
-	kthread_flush_work(&vbl_ctrl->work);
-	list_for_each_entry_safe(vbl_ev, tmp, &vbl_ctrl->event_list, node) {
-		list_del(&vbl_ev->node);
-		kfree(vbl_ev);
-	}
 
-	/* clean up display commit/event worker threads */
-	for (i = 0; i < priv->num_crtcs; i++) {
-		if (priv->disp_thread[i].thread) {
-			kthread_flush_worker(&priv->disp_thread[i].worker);
-			kthread_stop(priv->disp_thread[i].thread);
-			priv->disp_thread[i].thread = NULL;
-		}
+	flush_workqueue(priv->wq);
+	destroy_workqueue(priv->wq);
 
+	/* clean up event worker threads */
+	for (i = 0; i < priv->num_crtcs; i++) {
 		if (priv->event_thread[i].thread) {
-			kthread_flush_worker(&priv->event_thread[i].worker);
-			kthread_stop(priv->event_thread[i].thread);
+			kthread_destroy_worker(&priv->event_thread[i].worker);
 			priv->event_thread[i].thread = NULL;
 		}
 	}
@@ -323,9 +294,6 @@ static int msm_drm_uninit(struct device *dev)
 	drm_irq_uninstall(ddev);
 	pm_runtime_put_sync(dev);
 
-	flush_workqueue(priv->wq);
-	destroy_workqueue(priv->wq);
-
 	if (kms && kms->funcs)
 		kms->funcs->destroy(kms);
 
@@ -490,9 +458,6 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv)
 	priv->wq = alloc_ordered_workqueue("msm", 0);
 
 	INIT_LIST_HEAD(&priv->inactive_list);
-	INIT_LIST_HEAD(&priv->vblank_ctrl.event_list);
-	kthread_init_work(&priv->vblank_ctrl.work, vblank_ctrl_worker);
-	spin_lock_init(&priv->vblank_ctrl.lock);
 
 	drm_mode_config_init(ddev);
 
@@ -554,27 +519,6 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv)
 	 */
 	param.sched_priority = 16;
 	for (i = 0; i < priv->num_crtcs; i++) {
-
-		/* initialize display thread */
-		priv->disp_thread[i].crtc_id = priv->crtcs[i]->base.id;
-		kthread_init_worker(&priv->disp_thread[i].worker);
-		priv->disp_thread[i].dev = ddev;
-		priv->disp_thread[i].thread =
-			kthread_run(kthread_worker_fn,
-				&priv->disp_thread[i].worker,
-				"crtc_commit:%d", priv->disp_thread[i].crtc_id);
-		if (IS_ERR(priv->disp_thread[i].thread)) {
-			DRM_DEV_ERROR(dev, "failed to create crtc_commit kthread\n");
-			priv->disp_thread[i].thread = NULL;
-			goto err_msm_uninit;
-		}
-
-		ret = sched_setscheduler(priv->disp_thread[i].thread,
-					 SCHED_FIFO, &param);
-		if (ret)
-			dev_warn(dev, "disp_thread set priority failed: %d\n",
-				 ret);
-
 		/* initialize event thread */
 		priv->event_thread[i].crtc_id = priv->crtcs[i]->base.id;
 		kthread_init_worker(&priv->event_thread[i].worker);
@@ -589,13 +533,6 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv)
 			goto err_msm_uninit;
 		}
 
-		/**
-		 * event thread should also run at same priority as disp_thread
-		 * because it is handling frame_done events. A lower priority
-		 * event thread and higher priority disp_thread can causes
-		 * frame_pending counters beyond 2. This can lead to commit
-		 * failure at crtc commit level.
-		 */
 		ret = sched_setscheduler(priv->event_thread[i].thread,
 					 SCHED_FIFO, &param);
 		if (ret)
@@ -914,8 +851,12 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data,
 			ret = -EINVAL;
 			break;
 		}
-		ret = copy_from_user(msm_obj->name,
-			u64_to_user_ptr(args->value), args->len);
+		if (copy_from_user(msm_obj->name, u64_to_user_ptr(args->value),
+				   args->len)) {
+			msm_obj->name[0] = '\0';
+			ret = -EFAULT;
+			break;
+		}
 		msm_obj->name[args->len] = '\0';
 		for (i = 0; i < args->len; i++) {
 			if (!isprint(msm_obj->name[i])) {
@@ -931,8 +872,9 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data,
 		}
 		args->len = strlen(msm_obj->name);
 		if (args->value) {
-			ret = copy_to_user(u64_to_user_ptr(args->value),
-					msm_obj->name, args->len);
+			if (copy_to_user(u64_to_user_ptr(args->value),
+					 msm_obj->name, args->len))
+				ret = -EFAULT;
 		}
 		break;
 	}
@@ -1063,8 +1005,7 @@ static const struct file_operations fops = {
 };
 
 static struct drm_driver msm_driver = {
-	.driver_features    = DRIVER_HAVE_IRQ |
-				DRIVER_GEM |
+	.driver_features    = DRIVER_GEM |
 				DRIVER_PRIME |
 				DRIVER_RENDER |
 				DRIVER_ATOMIC |
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 9cd6a96c6bf2..c56dade2c1dc 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -39,8 +39,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/msm_drm.h>
 #include <drm/drm_gem.h>
@@ -77,12 +77,6 @@ enum msm_mdp_plane_property {
 	PLANE_PROP_MAX_NUM
 };
 
-struct msm_vblank_ctrl {
-	struct kthread_work work;
-	struct list_head event_list;
-	spinlock_t lock;
-};
-
 #define MSM_GPU_MAX_RINGS 4
 #define MAX_H_TILES_PER_DISPLAY 2
 
@@ -126,7 +120,7 @@ struct msm_display_topology {
 
 /**
  * struct msm_display_info - defines display properties
- * @intf_type:          DRM_MODE_CONNECTOR_ display type
+ * @intf_type:          DRM_MODE_ENCODER_ type
  * @capabilities:       Bitmask of display flags
  * @num_of_h_tiles:     Number of horizontal tiles in case of split interface
  * @h_tile_instance:    Controller instance used per tile. Number of elements is
@@ -199,7 +193,6 @@ struct msm_drm_private {
 	unsigned int num_crtcs;
 	struct drm_crtc *crtcs[MAX_CRTCS];
 
-	struct msm_drm_thread disp_thread[MAX_CRTCS];
 	struct msm_drm_thread event_thread[MAX_CRTCS];
 
 	unsigned int num_encoders;
@@ -228,7 +221,6 @@ struct msm_drm_private {
 	struct notifier_block vmap_notifier;
 	struct shrinker shrinker;
 
-	struct msm_vblank_ctrl vblank_ctrl;
 	struct drm_atomic_state *pm_state;
 };
 
@@ -250,7 +242,8 @@ void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
 void msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma);
 int msm_gem_map_vma(struct msm_gem_address_space *aspace,
-		struct msm_gem_vma *vma, struct sg_table *sgt, int npages);
+		struct msm_gem_vma *vma, int prot,
+		struct sg_table *sgt, int npages);
 void msm_gem_close_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma);
 
@@ -333,6 +326,7 @@ void msm_gem_kernel_put(struct drm_gem_object *bo,
 struct drm_gem_object *msm_gem_import(struct drm_device *dev,
 		struct dma_buf *dmabuf, struct sg_table *sgt);
 
+__printf(2, 3)
 void msm_gem_object_set_name(struct drm_gem_object *bo, const char *fmt, ...);
 
 int msm_framebuffer_prepare(struct drm_framebuffer *fb,
@@ -396,12 +390,14 @@ void msm_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m);
 int msm_debugfs_late_init(struct drm_device *dev);
 int msm_rd_debugfs_init(struct drm_minor *minor);
 void msm_rd_debugfs_cleanup(struct msm_drm_private *priv);
+__printf(3, 4)
 void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit,
 		const char *fmt, ...);
 int msm_perf_debugfs_init(struct drm_minor *minor);
 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_debugfs_cleanup(struct msm_drm_private *priv) {}
diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c
index 67dfd8d3dc12..136058978e0f 100644
--- a/drivers/gpu/drm/msm/msm_fb.c
+++ b/drivers/gpu/drm/msm/msm_fb.c
@@ -16,8 +16,8 @@
  */
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "msm_drv.h"
 #include "msm_kms.h"
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 20c979a7fc9c..18ca651ab942 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -391,6 +391,10 @@ static int msm_gem_pin_iova(struct drm_gem_object *obj,
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct msm_gem_vma *vma;
 	struct page **pages;
+	int prot = IOMMU_READ;
+
+	if (!(msm_obj->flags & MSM_BO_GPU_READONLY))
+		prot |= IOMMU_WRITE;
 
 	WARN_ON(!mutex_is_locked(&msm_obj->lock));
 
@@ -405,8 +409,8 @@ static int msm_gem_pin_iova(struct drm_gem_object *obj,
 	if (IS_ERR(pages))
 		return PTR_ERR(pages);
 
-	return msm_gem_map_vma(aspace, vma, msm_obj->sgt,
-			obj->size >> PAGE_SHIFT);
+	return msm_gem_map_vma(aspace, vma, prot,
+			msm_obj->sgt, obj->size >> PAGE_SHIFT);
 }
 
 /* get iova and pin it. Should have a matching put */
diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 557360788084..49c04829cf34 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -68,7 +68,8 @@ void msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
 
 int
 msm_gem_map_vma(struct msm_gem_address_space *aspace,
-		struct msm_gem_vma *vma, struct sg_table *sgt, int npages)
+		struct msm_gem_vma *vma, int prot,
+		struct sg_table *sgt, int npages)
 {
 	unsigned size = npages << PAGE_SHIFT;
 	int ret = 0;
@@ -86,7 +87,7 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace,
 
 	if (aspace->mmu)
 		ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt,
-				size, IOMMU_READ | IOMMU_WRITE);
+				size, prot);
 
 	if (ret)
 		vma->mapped = false;
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 5f3eff304355..10babd18e286 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -900,7 +900,7 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	}
 
 	/* Get Interrupt: */
-	gpu->irq = platform_get_irq_byname(pdev, config->irqname);
+	gpu->irq = platform_get_irq(pdev, 0);
 	if (gpu->irq < 0) {
 		ret = gpu->irq;
 		DRM_DEV_ERROR(drm->dev, "failed to get irq: %d\n", ret);
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index efb49bb64191..ca17086f72c9 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -31,7 +31,6 @@ struct msm_gpu_state;
 
 struct msm_gpu_config {
 	const char *ioname;
-	const char *irqname;
 	uint64_t va_start;
 	uint64_t va_end;
 	unsigned int nr_rings;
@@ -63,7 +62,7 @@ struct msm_gpu_funcs {
 	struct msm_ringbuffer *(*active_ring)(struct msm_gpu *gpu);
 	void (*recover)(struct msm_gpu *gpu);
 	void (*destroy)(struct msm_gpu *gpu);
-#ifdef CONFIG_DEBUG_FS
+#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP)
 	/* show GPU status in debugfs: */
 	void (*show)(struct msm_gpu *gpu, struct msm_gpu_state *state,
 			struct drm_printer *p);
diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c
index 90e9d0a48dc0..d21172933d92 100644
--- a/drivers/gpu/drm/msm/msm_rd.c
+++ b/drivers/gpu/drm/msm/msm_rd.c
@@ -115,7 +115,9 @@ static void rd_write(struct msm_rd_state *rd, const void *buf, int sz)
 		char *fptr = &fifo->buf[fifo->head];
 		int n;
 
-		wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0);
+		wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0 || !rd->open);
+		if (!rd->open)
+			return;
 
 		/* Note that smp_load_acquire() is not strictly required
 		 * as CIRC_SPACE_TO_END() does not access the tail more
@@ -213,7 +215,10 @@ out:
 static int rd_release(struct inode *inode, struct file *file)
 {
 	struct msm_rd_state *rd = inode->i_private;
+
 	rd->open = false;
+	wake_up_all(&rd->fifo_event);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c b/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
index 24b1f0c1432e..0ee1ca8a316a 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
@@ -19,12 +19,12 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_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 <linux/clk.h>
 #include <linux/iopoll.h>
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.c b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
index 13e778825098..967379f3f571 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_drv.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
@@ -31,13 +31,13 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.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 "mxsfb_drv.h"
@@ -322,8 +322,7 @@ DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver mxsfb_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET |
-				  DRIVER_PRIME | DRIVER_ATOMIC |
-				  DRIVER_HAVE_IRQ,
+				  DRIVER_PRIME | DRIVER_ATOMIC,
 	.irq_handler		= mxsfb_irq_handler,
 	.irq_preinstall		= mxsfb_irq_preinstall,
 	.irq_uninstall		= mxsfb_irq_preinstall,
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_out.c b/drivers/gpu/drm/mxsfb/mxsfb_out.c
index e5edf016a439..27add9976931 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_out.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_out.c
@@ -16,12 +16,12 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.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 <drm/drmP.h>
 
diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild
index b17843dd050d..581404e6544d 100644
--- a/drivers/gpu/drm/nouveau/Kbuild
+++ b/drivers/gpu/drm/nouveau/Kbuild
@@ -30,6 +30,8 @@ nouveau-y += nouveau_vga.o
 # DRM - memory management
 nouveau-y += nouveau_bo.o
 nouveau-y += nouveau_gem.o
+nouveau-$(CONFIG_DRM_NOUVEAU_SVM) += nouveau_svm.o
+nouveau-$(CONFIG_DRM_NOUVEAU_SVM) += nouveau_dmem.o
 nouveau-y += nouveau_mem.o
 nouveau-y += nouveau_prime.o
 nouveau-y += nouveau_sgdma.o
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index 432c440223bb..00cd9ab8948d 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -71,3 +71,15 @@ config DRM_NOUVEAU_BACKLIGHT
 	help
 	  Say Y here if you want to control the backlight of your display
 	  (e.g. a laptop panel).
+
+config DRM_NOUVEAU_SVM
+	bool "(EXPERIMENTAL) Enable SVM (Shared Virtual Memory) support"
+	depends on ARCH_HAS_HMM
+	depends on DRM_NOUVEAU
+	depends on STAGING
+	select HMM_MIRROR
+	select DEVICE_PRIVATE
+	default n
+	help
+	  Say Y here if you want to enable experimental support for
+	  Shared Virtual Memory (SVM).
diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
index 2c569e264df3..f22f01020625 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
@@ -40,6 +40,7 @@
 #include "nvreg.h"
 #include "nouveau_fbcon.h"
 #include "disp.h"
+#include "nouveau_dma.h"
 
 #include <subdev/bios/pll.h>
 #include <subdev/clk.h>
@@ -1077,12 +1078,223 @@ nouveau_crtc_set_config(struct drm_mode_set *set,
 	return ret;
 }
 
+struct nv04_page_flip_state {
+	struct list_head head;
+	struct drm_pending_vblank_event *event;
+	struct drm_crtc *crtc;
+	int bpp, pitch;
+	u64 offset;
+};
+
+static int
+nv04_finish_page_flip(struct nouveau_channel *chan,
+		      struct nv04_page_flip_state *ps)
+{
+	struct nouveau_fence_chan *fctx = chan->fence;
+	struct nouveau_drm *drm = chan->drm;
+	struct drm_device *dev = drm->dev;
+	struct nv04_page_flip_state *s;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+
+	if (list_empty(&fctx->flip)) {
+		NV_ERROR(drm, "unexpected pageflip\n");
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+		return -EINVAL;
+	}
+
+	s = list_first_entry(&fctx->flip, struct nv04_page_flip_state, head);
+	if (s->event) {
+		drm_crtc_arm_vblank_event(s->crtc, s->event);
+	} else {
+		/* Give up ownership of vblank for page-flipped crtc */
+		drm_crtc_vblank_put(s->crtc);
+	}
+
+	list_del(&s->head);
+	if (ps)
+		*ps = *s;
+	kfree(s);
+
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+	return 0;
+}
+
+int
+nv04_flip_complete(struct nvif_notify *notify)
+{
+	struct nouveau_cli *cli = (void *)notify->object->client;
+	struct nouveau_drm *drm = cli->drm;
+	struct nouveau_channel *chan = drm->channel;
+	struct nv04_page_flip_state state;
+
+	if (!nv04_finish_page_flip(chan, &state)) {
+		nv_set_crtc_base(drm->dev, drm_crtc_index(state.crtc),
+				 state.offset + state.crtc->y *
+				 state.pitch + state.crtc->x *
+				 state.bpp / 8);
+	}
+
+	return NVIF_NOTIFY_KEEP;
+}
+
+static int
+nv04_page_flip_emit(struct nouveau_channel *chan,
+		    struct nouveau_bo *old_bo,
+		    struct nouveau_bo *new_bo,
+		    struct nv04_page_flip_state *s,
+		    struct nouveau_fence **pfence)
+{
+	struct nouveau_fence_chan *fctx = chan->fence;
+	struct nouveau_drm *drm = chan->drm;
+	struct drm_device *dev = drm->dev;
+	unsigned long flags;
+	int ret;
+
+	/* Queue it to the pending list */
+	spin_lock_irqsave(&dev->event_lock, flags);
+	list_add_tail(&s->head, &fctx->flip);
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
+	/* Synchronize with the old framebuffer */
+	ret = nouveau_fence_sync(old_bo, chan, false, false);
+	if (ret)
+		goto fail;
+
+	/* Emit the pageflip */
+	ret = RING_SPACE(chan, 2);
+	if (ret)
+		goto fail;
+
+	BEGIN_NV04(chan, NvSubSw, NV_SW_PAGE_FLIP, 1);
+	OUT_RING  (chan, 0x00000000);
+	FIRE_RING (chan);
+
+	ret = nouveau_fence_new(chan, false, pfence);
+	if (ret)
+		goto fail;
+
+	return 0;
+fail:
+	spin_lock_irqsave(&dev->event_lock, flags);
+	list_del(&s->head);
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+	return ret;
+}
+
+static int
+nv04_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+		    struct drm_pending_vblank_event *event, u32 flags,
+		    struct drm_modeset_acquire_ctx *ctx)
+{
+	const int swap_interval = (flags & DRM_MODE_PAGE_FLIP_ASYNC) ? 0 : 1;
+	struct drm_device *dev = crtc->dev;
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_bo *old_bo = nouveau_framebuffer(crtc->primary->fb)->nvbo;
+	struct nouveau_bo *new_bo = nouveau_framebuffer(fb)->nvbo;
+	struct nv04_page_flip_state *s;
+	struct nouveau_channel *chan;
+	struct nouveau_cli *cli;
+	struct nouveau_fence *fence;
+	struct nv04_display *dispnv04 = nv04_display(dev);
+	int head = nouveau_crtc(crtc)->index;
+	int ret;
+
+	chan = drm->channel;
+	if (!chan)
+		return -ENODEV;
+	cli = (void *)chan->user.client;
+
+	s = kzalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return -ENOMEM;
+
+	if (new_bo != old_bo) {
+		ret = nouveau_bo_pin(new_bo, TTM_PL_FLAG_VRAM, true);
+		if (ret)
+			goto fail_free;
+	}
+
+	mutex_lock(&cli->mutex);
+	ret = ttm_bo_reserve(&new_bo->bo, true, false, NULL);
+	if (ret)
+		goto fail_unpin;
+
+	/* synchronise rendering channel with the kernel's channel */
+	ret = nouveau_fence_sync(new_bo, chan, false, true);
+	if (ret) {
+		ttm_bo_unreserve(&new_bo->bo);
+		goto fail_unpin;
+	}
+
+	if (new_bo != old_bo) {
+		ttm_bo_unreserve(&new_bo->bo);
+
+		ret = ttm_bo_reserve(&old_bo->bo, true, false, NULL);
+		if (ret)
+			goto fail_unpin;
+	}
+
+	/* Initialize a page flip struct */
+	*s = (struct nv04_page_flip_state)
+		{ { }, event, crtc, fb->format->cpp[0] * 8, fb->pitches[0],
+		  new_bo->bo.offset };
+
+	/* Keep vblanks on during flip, for the target crtc of this flip */
+	drm_crtc_vblank_get(crtc);
+
+	/* Emit a page flip */
+	if (swap_interval) {
+		ret = RING_SPACE(chan, 8);
+		if (ret)
+			goto fail_unreserve;
+
+		BEGIN_NV04(chan, NvSubImageBlit, 0x012c, 1);
+		OUT_RING  (chan, 0);
+		BEGIN_NV04(chan, NvSubImageBlit, 0x0134, 1);
+		OUT_RING  (chan, head);
+		BEGIN_NV04(chan, NvSubImageBlit, 0x0100, 1);
+		OUT_RING  (chan, 0);
+		BEGIN_NV04(chan, NvSubImageBlit, 0x0130, 1);
+		OUT_RING  (chan, 0);
+	}
+
+	nouveau_bo_ref(new_bo, &dispnv04->image[head]);
+
+	ret = nv04_page_flip_emit(chan, old_bo, new_bo, s, &fence);
+	if (ret)
+		goto fail_unreserve;
+	mutex_unlock(&cli->mutex);
+
+	/* Update the crtc struct and cleanup */
+	crtc->primary->fb = fb;
+
+	nouveau_bo_fence(old_bo, fence, false);
+	ttm_bo_unreserve(&old_bo->bo);
+	if (old_bo != new_bo)
+		nouveau_bo_unpin(old_bo);
+	nouveau_fence_unref(&fence);
+	return 0;
+
+fail_unreserve:
+	drm_crtc_vblank_put(crtc);
+	ttm_bo_unreserve(&old_bo->bo);
+fail_unpin:
+	mutex_unlock(&cli->mutex);
+	if (old_bo != new_bo)
+		nouveau_bo_unpin(new_bo);
+fail_free:
+	kfree(s);
+	return ret;
+}
+
 static const struct drm_crtc_funcs nv04_crtc_funcs = {
 	.cursor_set = nv04_crtc_cursor_set,
 	.cursor_move = nv04_crtc_cursor_move,
 	.gamma_set = nv_crtc_gamma_set,
 	.set_config = nouveau_crtc_set_config,
-	.page_flip = nouveau_crtc_page_flip,
+	.page_flip = nv04_crtc_page_flip,
 	.destroy = nv_crtc_destroy,
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 1727d399833c..5713bacaee80 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -30,6 +30,160 @@
 #include "hw.h"
 #include "nouveau_encoder.h"
 #include "nouveau_connector.h"
+#include "nouveau_bo.h"
+
+#include <nvif/if0004.h>
+
+static void
+nv04_display_fini(struct drm_device *dev, bool suspend)
+{
+	struct nv04_display *disp = nv04_display(dev);
+	struct drm_crtc *crtc;
+
+	/* Disable flip completion events. */
+	nvif_notify_put(&disp->flip);
+
+	/* Disable vblank interrupts. */
+	NVWriteCRTC(dev, 0, NV_PCRTC_INTR_EN_0, 0);
+	if (nv_two_heads(dev))
+		NVWriteCRTC(dev, 1, NV_PCRTC_INTR_EN_0, 0);
+
+	if (!suspend)
+		return;
+
+	/* Un-pin FB and cursors so they'll be evicted to system memory. */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_framebuffer *nouveau_fb;
+
+		nouveau_fb = nouveau_framebuffer(crtc->primary->fb);
+		if (!nouveau_fb || !nouveau_fb->nvbo)
+			continue;
+
+		nouveau_bo_unpin(nouveau_fb->nvbo);
+	}
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+		if (nv_crtc->cursor.nvbo) {
+			if (nv_crtc->cursor.set_offset)
+				nouveau_bo_unmap(nv_crtc->cursor.nvbo);
+			nouveau_bo_unpin(nv_crtc->cursor.nvbo);
+		}
+	}
+}
+
+static int
+nv04_display_init(struct drm_device *dev, bool resume, bool runtime)
+{
+	struct nv04_display *disp = nv04_display(dev);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_encoder *encoder;
+	struct drm_crtc *crtc;
+	int ret;
+
+	/* meh.. modeset apparently doesn't setup all the regs and depends
+	 * on pre-existing state, for now load the state of the card *before*
+	 * nouveau was loaded, and then do a modeset.
+	 *
+	 * best thing to do probably is to make save/restore routines not
+	 * save/restore "pre-load" state, but more general so we can save
+	 * on suspend too.
+	 */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+		nv_crtc->save(&nv_crtc->base);
+	}
+
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head)
+		encoder->enc_save(&encoder->base.base);
+
+	/* Enable flip completion events. */
+	nvif_notify_get(&disp->flip);
+
+	if (!resume)
+		return 0;
+
+	/* Re-pin FB/cursors. */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_framebuffer *nouveau_fb;
+
+		nouveau_fb = nouveau_framebuffer(crtc->primary->fb);
+		if (!nouveau_fb || !nouveau_fb->nvbo)
+			continue;
+
+		ret = nouveau_bo_pin(nouveau_fb->nvbo, TTM_PL_FLAG_VRAM, true);
+		if (ret)
+			NV_ERROR(drm, "Could not pin framebuffer\n");
+	}
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+		if (!nv_crtc->cursor.nvbo)
+			continue;
+
+		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo, TTM_PL_FLAG_VRAM, true);
+		if (!ret && nv_crtc->cursor.set_offset)
+			ret = nouveau_bo_map(nv_crtc->cursor.nvbo);
+		if (ret)
+			NV_ERROR(drm, "Could not pin/map cursor.\n");
+	}
+
+	/* Force CLUT to get re-loaded during modeset. */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+
+		nv_crtc->lut.depth = 0;
+	}
+
+	/* This should ensure we don't hit a locking problem when someone
+	 * wakes us up via a connector.  We should never go into suspend
+	 * while the display is on anyways.
+	 */
+	if (runtime)
+		return 0;
+
+	/* Restore mode. */
+	drm_helper_resume_force_mode(dev);
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+
+		if (!nv_crtc->cursor.nvbo)
+			continue;
+
+		if (nv_crtc->cursor.set_offset)
+			nv_crtc->cursor.set_offset(nv_crtc, nv_crtc->cursor.nvbo->bo.offset);
+		nv_crtc->cursor.set_pos(nv_crtc, nv_crtc->cursor_saved_x,
+						 nv_crtc->cursor_saved_y);
+	}
+
+	return 0;
+}
+
+static void
+nv04_display_destroy(struct drm_device *dev)
+{
+	struct nv04_display *disp = nv04_display(dev);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_encoder *encoder;
+	struct nouveau_crtc *nv_crtc;
+
+	/* Restore state */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head)
+		encoder->enc_restore(&encoder->base.base);
+
+	list_for_each_entry(nv_crtc, &dev->mode_config.crtc_list, base.head)
+		nv_crtc->restore(&nv_crtc->base);
+
+	nouveau_hw_save_vga_fonts(dev, 0);
+
+	nvif_notify_fini(&disp->flip);
+
+	nouveau_display(dev)->priv = NULL;
+	kfree(disp);
+
+	nvif_object_unmap(&drm->client.device.object);
+}
 
 int
 nv04_display_create(struct drm_device *dev)
@@ -58,6 +212,13 @@ nv04_display_create(struct drm_device *dev)
 	/* Pre-nv50 doesn't support atomic, so don't expose the ioctls */
 	dev->driver->driver_features &= ~DRIVER_ATOMIC;
 
+	/* Request page flip completion event. */
+	if (drm->nvsw.client) {
+		nvif_notify_init(&drm->nvsw, nv04_flip_complete,
+				 false, NV04_NVSW_NTFY_UEVENT,
+				 NULL, 0, 0, &disp->flip);
+	}
+
 	nouveau_hw_save_vga_fonts(dev, 1);
 
 	nv04_crtc_create(dev, 0);
@@ -121,58 +282,3 @@ nv04_display_create(struct drm_device *dev)
 
 	return 0;
 }
-
-void
-nv04_display_destroy(struct drm_device *dev)
-{
-	struct nv04_display *disp = nv04_display(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_encoder *encoder;
-	struct nouveau_crtc *nv_crtc;
-
-	/* Restore state */
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head)
-		encoder->enc_restore(&encoder->base.base);
-
-	list_for_each_entry(nv_crtc, &dev->mode_config.crtc_list, base.head)
-		nv_crtc->restore(&nv_crtc->base);
-
-	nouveau_hw_save_vga_fonts(dev, 0);
-
-	nouveau_display(dev)->priv = NULL;
-	kfree(disp);
-
-	nvif_object_unmap(&drm->client.device.object);
-}
-
-int
-nv04_display_init(struct drm_device *dev)
-{
-	struct nouveau_encoder *encoder;
-	struct nouveau_crtc *crtc;
-
-	/* meh.. modeset apparently doesn't setup all the regs and depends
-	 * on pre-existing state, for now load the state of the card *before*
-	 * nouveau was loaded, and then do a modeset.
-	 *
-	 * best thing to do probably is to make save/restore routines not
-	 * save/restore "pre-load" state, but more general so we can save
-	 * on suspend too.
-	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head)
-		crtc->save(&crtc->base);
-
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.base.head)
-		encoder->enc_save(&encoder->base.base);
-
-	return 0;
-}
-
-void
-nv04_display_fini(struct drm_device *dev)
-{
-	/* disable vblank interrupts */
-	NVWriteCRTC(dev, 0, NV_PCRTC_INTR_EN_0, 0);
-	if (nv_two_heads(dev))
-		NVWriteCRTC(dev, 1, NV_PCRTC_INTR_EN_0, 0);
-}
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.h b/drivers/gpu/drm/nouveau/dispnv04/disp.h
index f74f1f2b186e..c6ed20a09f4a 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.h
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.h
@@ -82,6 +82,7 @@ struct nv04_display {
 	uint32_t saved_vga_font[4][16384];
 	uint32_t dac_users[4];
 	struct nouveau_bo *image[2];
+	struct nvif_notify flip;
 };
 
 static inline struct nv04_display *
@@ -92,9 +93,6 @@ nv04_display(struct drm_device *dev)
 
 /* nv04_display.c */
 int nv04_display_create(struct drm_device *);
-void nv04_display_destroy(struct drm_device *);
-int nv04_display_init(struct drm_device *);
-void nv04_display_fini(struct drm_device *);
 
 /* nv04_crtc.c */
 int nv04_crtc_create(struct drm_device *, int index);
@@ -176,4 +174,5 @@ nouveau_bios_run_init_table(struct drm_device *dev, u16 table,
 	);
 }
 
+int nv04_flip_complete(struct nvif_notify *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
index 8fd8124d72ba..26fd71c06626 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
@@ -26,6 +26,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "nouveau_drv.h"
 #include "nouveau_reg.h"
 #include "nouveau_encoder.h"
diff --git a/drivers/gpu/drm/nouveau/dispnv50/atom.h b/drivers/gpu/drm/nouveau/dispnv50/atom.h
index a194990d2b0d..b5fae5ab3fa8 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/atom.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/atom.h
@@ -116,6 +116,12 @@ struct nv50_head_atom {
 		u8 depth:4;
 	} or;
 
+	/* Currently only used for MST */
+	struct {
+		int pbn;
+		u8 tu:6;
+	} dp;
+
 	union nv50_head_atom_mask {
 		struct {
 			bool olut:1;
diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.c b/drivers/gpu/drm/nouveau/dispnv50/core.c
index c25e0ebe3c92..27ea3f34706d 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/core.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/core.c
@@ -42,7 +42,7 @@ nv50_core_new(struct nouveau_drm *drm, struct nv50_core **pcore)
 		int version;
 		int (*new)(struct nouveau_drm *, s32, struct nv50_core **);
 	} cores[] = {
-		{ TU104_DISP_CORE_CHANNEL_DMA, 0, corec57d_new },
+		{ TU102_DISP_CORE_CHANNEL_DMA, 0, corec57d_new },
 		{ GV100_DISP_CORE_CHANNEL_DMA, 0, corec37d_new },
 		{ GP102_DISP_CORE_CHANNEL_DMA, 0, core917d_new },
 		{ GP100_DISP_CORE_CHANNEL_DMA, 0, core917d_new },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs.c b/drivers/gpu/drm/nouveau/dispnv50/curs.c
index cb6e4d2b1b45..121c24a18f11 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/curs.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/curs.c
@@ -31,7 +31,7 @@ nv50_curs_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw)
 		int version;
 		int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **);
 	} curses[] = {
-		{ TU104_DISP_CURSOR, 0, cursc37a_new },
+		{ TU102_DISP_CURSOR, 0, cursc37a_new },
 		{ GV100_DISP_CURSOR, 0, cursc37a_new },
 		{ GK104_DISP_CURSOR, 0, curs907a_new },
 		{ GF110_DISP_CURSOR, 0, curs907a_new },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 67107f0b1299..4b1650f51955 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -32,10 +32,10 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_scdc_helper.h>
 #include <drm/drm_edid.h>
 
@@ -659,8 +659,6 @@ struct nv50_mstc {
 
 	struct drm_display_mode *native;
 	struct edid *edid;
-
-	int pbn;
 };
 
 struct nv50_msto {
@@ -765,18 +763,26 @@ nv50_msto_atomic_check(struct drm_encoder *encoder,
 	struct drm_connector *connector = conn_state->connector;
 	struct nv50_mstc *mstc = nv50_mstc(connector);
 	struct nv50_mstm *mstm = mstc->mstm;
-	int bpp = connector->display_info.bpc * 3;
+	struct nv50_head_atom *asyh = nv50_head_atom(crtc_state);
 	int slots;
 
-	mstc->pbn = drm_dp_calc_pbn_mode(crtc_state->adjusted_mode.clock,
-					 bpp);
+	/* When restoring duplicated states, we need to make sure that the
+	 * bw remains the same and avoid recalculating it, as the connector's
+	 * bpc may have changed after the state was duplicated
+	 */
+	if (!state->duplicated)
+		asyh->dp.pbn =
+			drm_dp_calc_pbn_mode(crtc_state->adjusted_mode.clock,
+					     connector->display_info.bpc * 3);
 
-	if (drm_atomic_crtc_needs_modeset(crtc_state) &&
-	    !drm_connector_is_unregistered(connector)) {
+	if (drm_atomic_crtc_needs_modeset(crtc_state)) {
 		slots = drm_dp_atomic_find_vcpi_slots(state, &mstm->mgr,
-						      mstc->port, mstc->pbn);
+						      mstc->port,
+						      asyh->dp.pbn);
 		if (slots < 0)
 			return slots;
+
+		asyh->dp.tu = slots;
 	}
 
 	return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state,
@@ -787,13 +793,13 @@ static void
 nv50_msto_enable(struct drm_encoder *encoder)
 {
 	struct nv50_head *head = nv50_head(encoder->crtc);
+	struct nv50_head_atom *armh = nv50_head_atom(head->base.base.state);
 	struct nv50_msto *msto = nv50_msto(encoder);
 	struct nv50_mstc *mstc = NULL;
 	struct nv50_mstm *mstm = NULL;
 	struct drm_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 	u8 proto, depth;
-	int slots;
 	bool r;
 
 	drm_connector_list_iter_begin(encoder->dev, &conn_iter);
@@ -809,9 +815,10 @@ nv50_msto_enable(struct drm_encoder *encoder)
 	if (WARN_ON(!mstc))
 		return;
 
-	slots = drm_dp_find_vcpi_slots(&mstm->mgr, mstc->pbn);
-	r = drm_dp_mst_allocate_vcpi(&mstm->mgr, mstc->port, mstc->pbn, slots);
-	WARN_ON(!r);
+	r = drm_dp_mst_allocate_vcpi(&mstm->mgr, mstc->port, armh->dp.pbn,
+				     armh->dp.tu);
+	if (!r)
+		DRM_DEBUG_KMS("Failed to allocate VCPI\n");
 
 	if (!mstm->links++)
 		nv50_outp_acquire(mstm->outp);
@@ -828,8 +835,7 @@ nv50_msto_enable(struct drm_encoder *encoder)
 	default: depth = 0x6; break;
 	}
 
-	mstm->outp->update(mstm->outp, head->base.index,
-			   nv50_head_atom(head->base.base.state), proto, depth);
+	mstm->outp->update(mstm->outp, head->base.index, armh, proto, depth);
 
 	msto->head = head;
 	msto->mstc = mstc;
@@ -2215,8 +2221,8 @@ nv50_disp_func = {
  * Init
  *****************************************************************************/
 
-void
-nv50_display_fini(struct drm_device *dev)
+static void
+nv50_display_fini(struct drm_device *dev, bool suspend)
 {
 	struct nouveau_encoder *nv_encoder;
 	struct drm_encoder *encoder;
@@ -2237,8 +2243,8 @@ nv50_display_fini(struct drm_device *dev)
 	}
 }
 
-int
-nv50_display_init(struct drm_device *dev)
+static int
+nv50_display_init(struct drm_device *dev, bool resume, bool runtime)
 {
 	struct nv50_core *core = nv50_disp(dev)->core;
 	struct drm_encoder *encoder;
@@ -2264,7 +2270,7 @@ nv50_display_init(struct drm_device *dev)
 	return 0;
 }
 
-void
+static void
 nv50_display_destroy(struct drm_device *dev)
 {
 	struct nv50_disp *disp = nv50_disp(dev);
diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c
index ac97ebce5b35..2e7a0c347ddb 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/head.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/head.c
@@ -413,6 +413,7 @@ nv50_head_atomic_duplicate_state(struct drm_crtc *crtc)
 	asyh->ovly = armh->ovly;
 	asyh->dither = armh->dither;
 	asyh->procamp = armh->procamp;
+	asyh->dp = armh->dp;
 	asyh->clr.mask = 0;
 	asyh->set.mask = 0;
 	return &asyh->state;
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimm.c b/drivers/gpu/drm/nouveau/dispnv50/wimm.c
index bc9eeaf212ae..a1ac153d5e98 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wimm.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wimm.c
@@ -31,7 +31,7 @@ nv50_wimm_init(struct nouveau_drm *drm, struct nv50_wndw *wndw)
 		int version;
 		int (*init)(struct nouveau_drm *, s32, struct nv50_wndw *);
 	} wimms[] = {
-		{ TU104_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
+		{ TU102_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
 		{ GV100_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
 		{}
 	};
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
index ba9eea2ff16b..b95181027b31 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
@@ -626,7 +626,7 @@ nv50_wndw_new(struct nouveau_drm *drm, enum drm_plane_type type, int index,
 		int (*new)(struct nouveau_drm *, enum drm_plane_type,
 			   int, s32, struct nv50_wndw **);
 	} wndws[] = {
-		{ TU104_DISP_WINDOW_CHANNEL_DMA, 0, wndwc57e_new },
+		{ TU102_DISP_WINDOW_CHANNEL_DMA, 0, wndwc57e_new },
 		{ GV100_DISP_WINDOW_CHANNEL_DMA, 0, wndwc37e_new },
 		{}
 	};
diff --git a/drivers/gpu/drm/nouveau/include/nvif/class.h b/drivers/gpu/drm/nouveau/include/nvif/class.h
index 1d82cbf70cf4..7d556a1c92fa 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/class.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/class.h
@@ -54,6 +54,9 @@
 
 #define VOLTA_USERMODE_A                                             0x0000c361
 
+#define MAXWELL_FAULT_BUFFER_A                        /* clb069.h */ 0x0000b069
+#define VOLTA_FAULT_BUFFER_A                          /* clb069.h */ 0x0000c369
+
 #define NV03_CHANNEL_DMA                              /* cl506b.h */ 0x0000006b
 #define NV10_CHANNEL_DMA                              /* cl506b.h */ 0x0000006e
 #define NV17_CHANNEL_DMA                              /* cl506b.h */ 0x0000176e
@@ -84,7 +87,7 @@
 #define GP100_DISP                                    /* cl5070.h */ 0x00009770
 #define GP102_DISP                                    /* cl5070.h */ 0x00009870
 #define GV100_DISP                                    /* cl5070.h */ 0x0000c370
-#define TU104_DISP                                    /* cl5070.h */ 0x0000c570
+#define TU102_DISP                                    /* cl5070.h */ 0x0000c570
 
 #define NV31_MPEG                                                    0x00003174
 #define G82_MPEG                                                     0x00008274
@@ -97,7 +100,7 @@
 #define GF110_DISP_CURSOR                             /* cl507a.h */ 0x0000907a
 #define GK104_DISP_CURSOR                             /* cl507a.h */ 0x0000917a
 #define GV100_DISP_CURSOR                             /* cl507a.h */ 0x0000c37a
-#define TU104_DISP_CURSOR                             /* cl507a.h */ 0x0000c57a
+#define TU102_DISP_CURSOR                             /* cl507a.h */ 0x0000c57a
 
 #define NV50_DISP_OVERLAY                             /* cl507b.h */ 0x0000507b
 #define G82_DISP_OVERLAY                              /* cl507b.h */ 0x0000827b
@@ -106,7 +109,7 @@
 #define GK104_DISP_OVERLAY                            /* cl507b.h */ 0x0000917b
 
 #define GV100_DISP_WINDOW_IMM_CHANNEL_DMA             /* clc37b.h */ 0x0000c37b
-#define TU104_DISP_WINDOW_IMM_CHANNEL_DMA             /* clc37b.h */ 0x0000c57b
+#define TU102_DISP_WINDOW_IMM_CHANNEL_DMA             /* clc37b.h */ 0x0000c57b
 
 #define NV50_DISP_BASE_CHANNEL_DMA                    /* cl507c.h */ 0x0000507c
 #define G82_DISP_BASE_CHANNEL_DMA                     /* cl507c.h */ 0x0000827c
@@ -129,7 +132,7 @@
 #define GP100_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000977d
 #define GP102_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000987d
 #define GV100_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000c37d
-#define TU104_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000c57d
+#define TU102_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000c57d
 
 #define NV50_DISP_OVERLAY_CHANNEL_DMA                 /* cl507e.h */ 0x0000507e
 #define G82_DISP_OVERLAY_CHANNEL_DMA                  /* cl507e.h */ 0x0000827e
@@ -139,7 +142,7 @@
 #define GK104_DISP_OVERLAY_CONTROL_DMA                /* cl507e.h */ 0x0000917e
 
 #define GV100_DISP_WINDOW_CHANNEL_DMA                 /* clc37e.h */ 0x0000c37e
-#define TU104_DISP_WINDOW_CHANNEL_DMA                 /* clc37e.h */ 0x0000c57e
+#define TU102_DISP_WINDOW_CHANNEL_DMA                 /* clc37e.h */ 0x0000c57e
 
 #define NV50_TESLA                                                   0x00005097
 #define G82_TESLA                                                    0x00008297
diff --git a/drivers/gpu/drm/nouveau/include/nvif/clb069.h b/drivers/gpu/drm/nouveau/include/nvif/clb069.h
new file mode 100644
index 000000000000..eef5d0227bab
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvif/clb069.h
@@ -0,0 +1,12 @@
+#ifndef __NVIF_CLB069_H__
+#define __NVIF_CLB069_H__
+struct nvif_clb069_v0 {
+	__u8  version;
+	__u8  pad01[3];
+	__u32 entries;
+	__u32 get;
+	__u32 put;
+};
+
+#define NVB069_V0_NTFY_FAULT                                                0x00
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/if000c.h b/drivers/gpu/drm/nouveau/include/nvif/if000c.h
index 2928ecd989ad..d6dd40f21eed 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/if000c.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/if000c.h
@@ -3,7 +3,8 @@
 struct nvif_vmm_v0 {
 	__u8  version;
 	__u8  page_nr;
-	__u8  pad02[6];
+	__u8  managed;
+	__u8  pad03[5];
 	__u64 addr;
 	__u64 size;
 	__u8  data[];
@@ -14,6 +15,9 @@ struct nvif_vmm_v0 {
 #define NVIF_VMM_V0_PUT                                                    0x02
 #define NVIF_VMM_V0_MAP                                                    0x03
 #define NVIF_VMM_V0_UNMAP                                                  0x04
+#define NVIF_VMM_V0_PFNMAP                                                 0x05
+#define NVIF_VMM_V0_PFNCLR                                                 0x06
+#define NVIF_VMM_V0_MTHD(i)                                         ((i) + 0x80)
 
 struct nvif_vmm_page_v0 {
 	__u8  version;
@@ -61,4 +65,28 @@ struct nvif_vmm_unmap_v0 {
 	__u8  pad01[7];
 	__u64 addr;
 };
+
+struct nvif_vmm_pfnmap_v0 {
+	__u8  version;
+	__u8  page;
+	__u8  pad02[6];
+	__u64 addr;
+	__u64 size;
+#define NVIF_VMM_PFNMAP_V0_ADDR                           0xfffffffffffff000ULL
+#define NVIF_VMM_PFNMAP_V0_ADDR_SHIFT                                        12
+#define NVIF_VMM_PFNMAP_V0_APER                           0x00000000000000f0ULL
+#define NVIF_VMM_PFNMAP_V0_HOST                           0x0000000000000000ULL
+#define NVIF_VMM_PFNMAP_V0_VRAM                           0x0000000000000010ULL
+#define NVIF_VMM_PFNMAP_V0_W                              0x0000000000000002ULL
+#define NVIF_VMM_PFNMAP_V0_V                              0x0000000000000001ULL
+#define NVIF_VMM_PFNMAP_V0_NONE                           0x0000000000000000ULL
+	__u64 phys[];
+};
+
+struct nvif_vmm_pfnclr_v0 {
+	__u8  version;
+	__u8  pad01[7];
+	__u64 addr;
+	__u64 size;
+};
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/ifc00d.h b/drivers/gpu/drm/nouveau/include/nvif/ifc00d.h
index 1d9c637859f3..4cabd613a280 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/ifc00d.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/ifc00d.h
@@ -6,6 +6,12 @@ struct gp100_vmm_vn {
 	/* nvif_vmm_vX ... */
 };
 
+struct gp100_vmm_v0 {
+	/* nvif_vmm_vX ... */
+	__u8  version;
+	__u8  fault_replay;
+};
+
 struct gp100_vmm_map_vn {
 	/* nvif_vmm_map_vX ... */
 };
@@ -18,4 +24,19 @@ struct gp100_vmm_map_v0 {
 	__u8  priv;
 	__u8  kind;
 };
+
+#define GP100_VMM_VN_FAULT_REPLAY                         NVIF_VMM_V0_MTHD(0x00)
+#define GP100_VMM_VN_FAULT_CANCEL                         NVIF_VMM_V0_MTHD(0x01)
+
+struct gp100_vmm_fault_replay_vn {
+};
+
+struct gp100_vmm_fault_cancel_v0 {
+	__u8  version;
+	__u8  hub;
+	__u8  gpc;
+	__u8  client;
+	__u8  pad04[4];
+	__u64 inst;
+};
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/vmm.h b/drivers/gpu/drm/nouveau/include/nvif/vmm.h
index c5db8a2e82df..79bf85d2f43a 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/vmm.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/vmm.h
@@ -30,8 +30,8 @@ struct nvif_vmm {
 	int page_nr;
 };
 
-int nvif_vmm_init(struct nvif_mmu *, s32 oclass, u64 addr, u64 size,
-		  void *argv, u32 argc, struct nvif_vmm *);
+int nvif_vmm_init(struct nvif_mmu *, s32 oclass, bool managed, u64 addr,
+		  u64 size, void *argv, u32 argc, struct nvif_vmm *);
 void nvif_vmm_fini(struct nvif_vmm *);
 int nvif_vmm_get(struct nvif_vmm *, enum nvif_vmm_get, bool sparse,
 		 u8 page, u8 align, u64 size, struct nvif_vma *);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
index 72e4dc1f0236..642492344196 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
@@ -28,6 +28,7 @@ enum nvkm_devidx {
 	NVKM_SUBDEV_ICCSENSE,
 	NVKM_SUBDEV_THERM,
 	NVKM_SUBDEV_CLK,
+	NVKM_SUBDEV_GSP,
 	NVKM_SUBDEV_SECBOOT,
 
 	NVKM_ENGINE_BSP,
@@ -137,6 +138,7 @@ struct nvkm_device {
 	struct nvkm_fb *fb;
 	struct nvkm_fuse *fuse;
 	struct nvkm_gpio *gpio;
+	struct nvkm_gsp *gsp;
 	struct nvkm_i2c *i2c;
 	struct nvkm_subdev *ibus;
 	struct nvkm_iccsense *iccsense;
@@ -209,6 +211,7 @@ struct nvkm_device_chip {
 	int (*fb      )(struct nvkm_device *, int idx, struct nvkm_fb **);
 	int (*fuse    )(struct nvkm_device *, int idx, struct nvkm_fuse **);
 	int (*gpio    )(struct nvkm_device *, int idx, struct nvkm_gpio **);
+	int (*gsp     )(struct nvkm_device *, int idx, struct nvkm_gsp **);
 	int (*i2c     )(struct nvkm_device *, int idx, struct nvkm_i2c **);
 	int (*ibus    )(struct nvkm_device *, int idx, struct nvkm_subdev **);
 	int (*iccsense)(struct nvkm_device *, int idx, struct nvkm_iccsense **);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
index 86abe76023c2..5f3650692e4d 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
@@ -11,5 +11,5 @@ int gm200_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 int gp100_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 int gp102_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 int gv100_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
-int tu104_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
+int tu102_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
index 5ca86e178bb9..3026b22d44fb 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
@@ -36,5 +36,5 @@ int gm200_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gp100_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gp102_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gv100_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
-int tu104_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
+int tu102_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
index 3b2b685778eb..b7fc04dd1628 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
@@ -74,5 +74,5 @@ int gm20b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gp100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gp10b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gv100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
-int tu104_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
+int tu102_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
index ba1518ff8b66..1e924c7f7ba7 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
@@ -10,6 +10,9 @@ struct nvkm_gr {
 
 u64 nvkm_gr_units(struct nvkm_gr *);
 int nvkm_gr_tlb_flush(struct nvkm_gr *);
+int nvkm_gr_ctxsw_pause(struct nvkm_device *);
+int nvkm_gr_ctxsw_resume(struct nvkm_device *);
+u32 nvkm_gr_ctxsw_inst(struct nvkm_device *);
 
 int nv04_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 int nv10_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
index fe716859d4a9..b72a4844c5f7 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
@@ -6,6 +6,8 @@
 
 struct nvkm_nvdec {
 	struct nvkm_engine engine;
+	u32 addr;
+
 	struct nvkm_falcon *falcon;
 };
 
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/sec2.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/sec2.h
index f7d89822b905..c93ad332461a 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/sec2.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/sec2.h
@@ -5,10 +5,13 @@
 
 struct nvkm_sec2 {
 	struct nvkm_engine engine;
+	u32 addr;
+
 	struct nvkm_falcon *falcon;
 	struct nvkm_msgqueue *queue;
 	struct work_struct work;
 };
 
 int gp102_sec2_new(struct nvkm_device *, int, struct nvkm_sec2 **);
+int tu102_sec2_new(struct nvkm_device *, int, struct nvkm_sec2 **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h
index fd9d713b611c..da14486317ca 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h
@@ -29,5 +29,5 @@ int gf100_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
 int gk20a_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
 int gm107_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
 int gm20b_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
-int tu104_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
+int tu102_bar_new(struct nvkm_device *, int, struct nvkm_bar **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
index 1b71812a790b..8ba982c2fdfb 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
@@ -31,5 +31,5 @@ int gf100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int gm107_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int gm200_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int gv100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
-int tu104_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
+int tu102_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h
index 127f48066026..97322f95b3ee 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h
@@ -13,6 +13,8 @@ struct nvkm_fault {
 	struct nvkm_event event;
 
 	struct nvkm_notify nrpfb;
+
+	struct nvkm_device_oclass user;
 };
 
 struct nvkm_fault_data {
@@ -30,5 +32,5 @@ struct nvkm_fault_data {
 
 int gp100_fault_new(struct nvkm_device *, int, struct nvkm_fault **);
 int gv100_fault_new(struct nvkm_device *, int, struct nvkm_fault **);
-int tu104_fault_new(struct nvkm_device *, int, struct nvkm_fault **);
+int tu102_fault_new(struct nvkm_device *, int, struct nvkm_fault **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
new file mode 100644
index 000000000000..4c672a5c4cd5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
@@ -0,0 +1,14 @@
+#ifndef __NVKM_GSP_H__
+#define __NVKM_GSP_H__
+#define nvkm_gsp(p) container_of((p), struct nvkm_gsp, subdev)
+#include <core/subdev.h>
+
+struct nvkm_gsp {
+	struct nvkm_subdev subdev;
+	u32 addr;
+
+	struct nvkm_falcon *falcon;
+};
+
+int gv100_gsp_new(struct nvkm_device *, int, struct nvkm_gsp **);
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h
index b66dedd8abb6..e38f4958dea2 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h
@@ -31,5 +31,5 @@ int gk104_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
 int gk20a_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
 int gp100_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
 int gp10b_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
-int tu104_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
+int tu102_mc_new(struct nvkm_device *, int, struct nvkm_mc **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
index 0a0e064f22e5..28ade86f74c5 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
@@ -17,6 +17,7 @@ struct nvkm_vma {
 	bool part:1; /* Region was split from an allocated region by map(). */
 	bool user:1; /* Region user-allocated. */
 	bool busy:1; /* Region busy (for temporarily preventing user access). */
+	bool mapped:1; /* Region contains valid pages. */
 	struct nvkm_memory *memory; /* Memory currently mapped into VMA. */
 	struct nvkm_tags *tags; /* Compression tag reference. */
 };
@@ -44,6 +45,8 @@ struct nvkm_vmm {
 
 	dma_addr_t null;
 	void *nullp;
+
+	bool replay;
 };
 
 int nvkm_vmm_new(struct nvkm_device *, u64 addr, u64 size, void *argv, u32 argc,
@@ -63,6 +66,7 @@ struct nvkm_vmm_map {
 	struct nvkm_mm_node *mem;
 	struct scatterlist *sgl;
 	dma_addr_t *dma;
+	u64 *pfn;
 	u64 off;
 
 	const struct nvkm_vmm_page *page;
@@ -130,5 +134,5 @@ int gm20b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
 int gp100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
 int gp10b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
 int gv100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
-int tu104_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
+int tu102_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/top.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/top.h
index f7d3eb647e2e..2904e67d79d2 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/top.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/top.h
@@ -9,6 +9,7 @@ struct nvkm_top {
 	struct list_head device;
 };
 
+u32 nvkm_top_addr(struct nvkm_device *, enum nvkm_devidx);
 u32 nvkm_top_reset(struct nvkm_device *, enum nvkm_devidx);
 u32 nvkm_top_intr(struct nvkm_device *, u32 intr, u64 *subdevs);
 u32 nvkm_top_intr_mask(struct nvkm_device *, enum nvkm_devidx);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
index 8a0f85f5fc1a..6a765682fbfa 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
@@ -38,6 +38,7 @@ int nvkm_volt_set_id(struct nvkm_volt *, u8 id, u8 min_id, u8 temp,
 
 int nv40_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gf100_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
+int gf117_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gk104_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gk20a_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gm20b_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index b06cdac8f3a2..c3fd5dd39ed9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -214,6 +214,7 @@ nouveau_abi16_ioctl_getparam(ABI16_IOCTL_ARGS)
 			WARN_ON(1);
 			break;
 		}
+		break;
 	case NOUVEAU_GETPARAM_FB_SIZE:
 		getparam->value = drm->gem.vram_available;
 		break;
@@ -338,7 +339,8 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 		goto done;
 
 	if (device->info.family >= NV_DEVICE_INFO_V0_TESLA) {
-		ret = nouveau_vma_new(chan->ntfy, &cli->vmm, &chan->ntfy_vma);
+		ret = nouveau_vma_new(chan->ntfy, chan->chan->vmm,
+				      &chan->ntfy_vma);
 		if (ret)
 			goto done;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c
index 5f5be6368aed..c7a94c94dbf3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_backlight.c
+++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c
@@ -253,6 +253,9 @@ nouveau_backlight_init(struct drm_connector *connector)
 	case NV_DEVICE_INFO_V0_FERMI:
 	case NV_DEVICE_INFO_V0_KEPLER:
 	case NV_DEVICE_INFO_V0_MAXWELL:
+	case NV_DEVICE_INFO_V0_PASCAL:
+	case NV_DEVICE_INFO_V0_VOLTA:
+	case NV_DEVICE_INFO_V0_TURING:
 		ret = nv50_backlight_init(nv_encoder, &props, &ops);
 		break;
 	default:
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 73eff52036d2..34a998012bf6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -194,7 +194,7 @@ nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align,
 	struct nouveau_drm *drm = cli->drm;
 	struct nouveau_bo *nvbo;
 	struct nvif_mmu *mmu = &cli->mmu;
-	struct nvif_vmm *vmm = &cli->vmm.vmm;
+	struct nvif_vmm *vmm = cli->svm.cli ? &cli->svm.vmm : &cli->vmm.vmm;
 	size_t acc_size;
 	int type = ttm_bo_type_device;
 	int ret, i, pi = -1;
@@ -1434,7 +1434,7 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *reg)
 		if (drm->client.mem->oclass < NVIF_CLASS_MEM_NV50 || !mem->kind)
 			/* untiled */
 			break;
-		/* fallthrough, tiled memory */
+		/* fall through - tiled memory */
 	case TTM_PL_VRAM:
 		reg->bus.offset = reg->start << PAGE_SHIFT;
 		reg->bus.base = device->func->resource_addr(device, 1);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 73c48440d4d7..846f4bdec0de 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -61,12 +61,14 @@ nouveau_bo_ref(struct nouveau_bo *ref, struct nouveau_bo **pnvbo)
 		return -EINVAL;
 	prev = *pnvbo;
 
-	*pnvbo = ref ? nouveau_bo(ttm_bo_reference(&ref->bo)) : NULL;
-	if (prev) {
-		struct ttm_buffer_object *bo = &prev->bo;
-
-		ttm_bo_unref(&bo);
+	if (ref) {
+		ttm_bo_get(&ref->bo);
+		*pnvbo = nouveau_bo(&ref->bo);
+	} else {
+		*pnvbo = NULL;
 	}
+	if (prev)
+		ttm_bo_put(&prev->bo);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index 668afbc29c3e..282fd90b65e1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -42,6 +42,7 @@
 #include "nouveau_fence.h"
 #include "nouveau_abi16.h"
 #include "nouveau_vmm.h"
+#include "nouveau_svm.h"
 
 MODULE_PARM_DESC(vram_pushbuf, "Create DMA push buffers in VRAM");
 int nouveau_vram_pushbuf;
@@ -95,6 +96,10 @@ nouveau_channel_del(struct nouveau_channel **pchan)
 
 		if (chan->fence)
 			nouveau_fence(chan->drm)->context_del(chan);
+
+		if (cli)
+			nouveau_svmm_part(chan->vmm->svmm, chan->inst);
+
 		nvif_object_fini(&chan->nvsw);
 		nvif_object_fini(&chan->gart);
 		nvif_object_fini(&chan->vram);
@@ -130,6 +135,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 
 	chan->device = device;
 	chan->drm = drm;
+	chan->vmm = cli->svm.cli ? &cli->svm : &cli->vmm;
 	atomic_set(&chan->killed, 0);
 
 	/* allocate memory for dma push buffer */
@@ -157,7 +163,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 	chan->push.addr = chan->push.buffer->bo.offset;
 
 	if (device->info.family >= NV_DEVICE_INFO_V0_TESLA) {
-		ret = nouveau_vma_new(chan->push.buffer, &cli->vmm,
+		ret = nouveau_vma_new(chan->push.buffer, chan->vmm,
 				      &chan->push.vma);
 		if (ret) {
 			nouveau_channel_del(pchan);
@@ -172,7 +178,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 		args.target = NV_DMA_V0_TARGET_VM;
 		args.access = NV_DMA_V0_ACCESS_VM;
 		args.start = 0;
-		args.limit = cli->vmm.vmm.limit - 1;
+		args.limit = chan->vmm->vmm.limit - 1;
 	} else
 	if (chan->push.buffer->bo.mem.mem_type == TTM_PL_VRAM) {
 		if (device->info.family == NV_DEVICE_INFO_V0_TNT) {
@@ -202,7 +208,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 			args.target = NV_DMA_V0_TARGET_VM;
 			args.access = NV_DMA_V0_ACCESS_RDWR;
 			args.start = 0;
-			args.limit = cli->vmm.vmm.limit - 1;
+			args.limit = chan->vmm->vmm.limit - 1;
 		}
 	}
 
@@ -220,7 +226,6 @@ static int
 nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 		    u64 runlist, bool priv, struct nouveau_channel **pchan)
 {
-	struct nouveau_cli *cli = (void *)device->object.client;
 	static const u16 oclasses[] = { TURING_CHANNEL_GPFIFO_A,
 					VOLTA_CHANNEL_GPFIFO_A,
 					PASCAL_CHANNEL_GPFIFO_A,
@@ -255,7 +260,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 			args.volta.ilength = 0x02000;
 			args.volta.ioffset = 0x10000 + chan->push.addr;
 			args.volta.runlist = runlist;
-			args.volta.vmm = nvif_handle(&cli->vmm.vmm.object);
+			args.volta.vmm = nvif_handle(&chan->vmm->vmm.object);
 			args.volta.priv = priv;
 			size = sizeof(args.volta);
 		} else
@@ -264,7 +269,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 			args.kepler.ilength = 0x02000;
 			args.kepler.ioffset = 0x10000 + chan->push.addr;
 			args.kepler.runlist = runlist;
-			args.kepler.vmm = nvif_handle(&cli->vmm.vmm.object);
+			args.kepler.vmm = nvif_handle(&chan->vmm->vmm.object);
 			args.kepler.priv = priv;
 			size = sizeof(args.kepler);
 		} else
@@ -272,14 +277,14 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 			args.fermi.version = 0;
 			args.fermi.ilength = 0x02000;
 			args.fermi.ioffset = 0x10000 + chan->push.addr;
-			args.fermi.vmm = nvif_handle(&cli->vmm.vmm.object);
+			args.fermi.vmm = nvif_handle(&chan->vmm->vmm.object);
 			size = sizeof(args.fermi);
 		} else {
 			args.nv50.version = 0;
 			args.nv50.ilength = 0x02000;
 			args.nv50.ioffset = 0x10000 + chan->push.addr;
 			args.nv50.pushbuf = nvif_handle(&chan->push.ctxdma);
-			args.nv50.vmm = nvif_handle(&cli->vmm.vmm.object);
+			args.nv50.vmm = nvif_handle(&chan->vmm->vmm.object);
 			size = sizeof(args.nv50);
 		}
 
@@ -350,7 +355,6 @@ static int
 nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 {
 	struct nvif_device *device = chan->device;
-	struct nouveau_cli *cli = (void *)chan->user.client;
 	struct nouveau_drm *drm = chan->drm;
 	struct nv_dma_v0 args = {};
 	int ret, i;
@@ -376,7 +380,7 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 			args.target = NV_DMA_V0_TARGET_VM;
 			args.access = NV_DMA_V0_ACCESS_VM;
 			args.start = 0;
-			args.limit = cli->vmm.vmm.limit - 1;
+			args.limit = chan->vmm->vmm.limit - 1;
 		} else {
 			args.target = NV_DMA_V0_TARGET_VRAM;
 			args.access = NV_DMA_V0_ACCESS_RDWR;
@@ -393,7 +397,7 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 			args.target = NV_DMA_V0_TARGET_VM;
 			args.access = NV_DMA_V0_ACCESS_VM;
 			args.start = 0;
-			args.limit = cli->vmm.vmm.limit - 1;
+			args.limit = chan->vmm->vmm.limit - 1;
 		} else
 		if (chan->drm->agp.bridge) {
 			args.target = NV_DMA_V0_TARGET_AGP;
@@ -405,7 +409,7 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 			args.target = NV_DMA_V0_TARGET_VM;
 			args.access = NV_DMA_V0_ACCESS_RDWR;
 			args.start = 0;
-			args.limit = cli->vmm.vmm.limit - 1;
+			args.limit = chan->vmm->vmm.limit - 1;
 		}
 
 		ret = nvif_object_init(&chan->user, gart, NV_DMA_IN_MEMORY,
@@ -495,6 +499,10 @@ nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device,
 		nouveau_channel_del(pchan);
 	}
 
+	ret = nouveau_svmm_join((*pchan)->vmm->svmm, (*pchan)->inst);
+	if (ret)
+		nouveau_channel_del(pchan);
+
 done:
 	cli->base.super = super;
 	return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h
index 28418f4e5748..93814d1d31e4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.h
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.h
@@ -8,6 +8,7 @@ struct nvif_device;
 struct nouveau_channel {
 	struct nvif_device *device;
 	struct nouveau_drm *drm;
+	struct nouveau_vmm *vmm;
 
 	int chid;
 	u64 inst;
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 3f463c91314a..4116ee62adaf 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -33,6 +33,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_atomic.h>
 
 #include "nouveau_reg.h"
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 5d273a655479..55c0fa451163 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -30,19 +30,15 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
-
-#include <nvif/class.h>
+#include <drm/drm_probe_helper.h>
 
 #include "nouveau_fbcon.h"
-#include "dispnv04/hw.h"
 #include "nouveau_crtc.h"
-#include "nouveau_dma.h"
 #include "nouveau_gem.h"
 #include "nouveau_connector.h"
 #include "nv50_display.h"
 
-#include "nouveau_fence.h"
-
+#include <nvif/class.h>
 #include <nvif/cl0046.h>
 #include <nvif/event.h>
 
@@ -411,15 +407,14 @@ nouveau_display_acpi_ntfy(struct notifier_block *nb, unsigned long val,
 #endif
 
 int
-nouveau_display_init(struct drm_device *dev)
+nouveau_display_init(struct drm_device *dev, bool resume, bool runtime)
 {
 	struct nouveau_display *disp = nouveau_display(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 	int ret;
 
-	ret = disp->init(dev);
+	ret = disp->init(dev, resume, runtime);
 	if (ret)
 		return ret;
 
@@ -436,8 +431,6 @@ nouveau_display_init(struct drm_device *dev)
 	}
 	drm_connector_list_iter_end(&conn_iter);
 
-	/* enable flip completion events */
-	nvif_notify_get(&drm->flip);
 	return ret;
 }
 
@@ -456,9 +449,6 @@ nouveau_display_fini(struct drm_device *dev, bool suspend, bool runtime)
 			drm_helper_force_disable_all(dev);
 	}
 
-	/* disable flip completion events */
-	nvif_notify_put(&drm->flip);
-
 	/* disable hotplug interrupts */
 	drm_connector_list_iter_begin(dev, &conn_iter);
 	nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
@@ -471,7 +461,7 @@ 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);
+	disp->fini(dev, suspend);
 }
 
 static void
@@ -624,7 +614,6 @@ int
 nouveau_display_suspend(struct drm_device *dev, bool runtime)
 {
 	struct nouveau_display *disp = nouveau_display(dev);
-	struct drm_crtc *crtc;
 
 	if (drm_drv_uses_atomic_modeset(dev)) {
 		if (!runtime) {
@@ -635,32 +624,9 @@ nouveau_display_suspend(struct drm_device *dev, bool runtime)
 				return ret;
 			}
 		}
-
-		nouveau_display_fini(dev, true, runtime);
-		return 0;
 	}
 
 	nouveau_display_fini(dev, true, runtime);
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_framebuffer *nouveau_fb;
-
-		nouveau_fb = nouveau_framebuffer(crtc->primary->fb);
-		if (!nouveau_fb || !nouveau_fb->nvbo)
-			continue;
-
-		nouveau_bo_unpin(nouveau_fb->nvbo);
-	}
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-		if (nv_crtc->cursor.nvbo) {
-			if (nv_crtc->cursor.set_offset)
-				nouveau_bo_unmap(nv_crtc->cursor.nvbo);
-			nouveau_bo_unpin(nv_crtc->cursor.nvbo);
-		}
-	}
-
 	return 0;
 }
 
@@ -668,275 +634,16 @@ void
 nouveau_display_resume(struct drm_device *dev, bool runtime)
 {
 	struct nouveau_display *disp = nouveau_display(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct drm_crtc *crtc;
-	int ret;
+
+	nouveau_display_init(dev, true, runtime);
 
 	if (drm_drv_uses_atomic_modeset(dev)) {
-		nouveau_display_init(dev);
 		if (disp->suspend) {
 			drm_atomic_helper_resume(dev, disp->suspend);
 			disp->suspend = NULL;
 		}
 		return;
 	}
-
-	/* re-pin fb/cursors */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_framebuffer *nouveau_fb;
-
-		nouveau_fb = nouveau_framebuffer(crtc->primary->fb);
-		if (!nouveau_fb || !nouveau_fb->nvbo)
-			continue;
-
-		ret = nouveau_bo_pin(nouveau_fb->nvbo, TTM_PL_FLAG_VRAM, true);
-		if (ret)
-			NV_ERROR(drm, "Could not pin framebuffer\n");
-	}
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-		if (!nv_crtc->cursor.nvbo)
-			continue;
-
-		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo, TTM_PL_FLAG_VRAM, true);
-		if (!ret && nv_crtc->cursor.set_offset)
-			ret = nouveau_bo_map(nv_crtc->cursor.nvbo);
-		if (ret)
-			NV_ERROR(drm, "Could not pin/map cursor.\n");
-	}
-
-	nouveau_display_init(dev);
-
-	/* Force CLUT to get re-loaded during modeset */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-
-		nv_crtc->lut.depth = 0;
-	}
-
-	/* This should ensure we don't hit a locking problem when someone
-	 * wakes us up via a connector.  We should never go into suspend
-	 * while the display is on anyways.
-	 */
-	if (runtime)
-		return;
-
-	drm_helper_resume_force_mode(dev);
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-
-		if (!nv_crtc->cursor.nvbo)
-			continue;
-
-		if (nv_crtc->cursor.set_offset)
-			nv_crtc->cursor.set_offset(nv_crtc, nv_crtc->cursor.nvbo->bo.offset);
-		nv_crtc->cursor.set_pos(nv_crtc, nv_crtc->cursor_saved_x,
-						 nv_crtc->cursor_saved_y);
-	}
-}
-
-static int
-nouveau_page_flip_emit(struct nouveau_channel *chan,
-		       struct nouveau_bo *old_bo,
-		       struct nouveau_bo *new_bo,
-		       struct nouveau_page_flip_state *s,
-		       struct nouveau_fence **pfence)
-{
-	struct nouveau_fence_chan *fctx = chan->fence;
-	struct nouveau_drm *drm = chan->drm;
-	struct drm_device *dev = drm->dev;
-	unsigned long flags;
-	int ret;
-
-	/* Queue it to the pending list */
-	spin_lock_irqsave(&dev->event_lock, flags);
-	list_add_tail(&s->head, &fctx->flip);
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-
-	/* Synchronize with the old framebuffer */
-	ret = nouveau_fence_sync(old_bo, chan, false, false);
-	if (ret)
-		goto fail;
-
-	/* Emit the pageflip */
-	ret = RING_SPACE(chan, 2);
-	if (ret)
-		goto fail;
-
-	BEGIN_NV04(chan, NvSubSw, NV_SW_PAGE_FLIP, 1);
-	OUT_RING  (chan, 0x00000000);
-	FIRE_RING (chan);
-
-	ret = nouveau_fence_new(chan, false, pfence);
-	if (ret)
-		goto fail;
-
-	return 0;
-fail:
-	spin_lock_irqsave(&dev->event_lock, flags);
-	list_del(&s->head);
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-	return ret;
-}
-
-int
-nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
-		       struct drm_pending_vblank_event *event, u32 flags,
-		       struct drm_modeset_acquire_ctx *ctx)
-{
-	const int swap_interval = (flags & DRM_MODE_PAGE_FLIP_ASYNC) ? 0 : 1;
-	struct drm_device *dev = crtc->dev;
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_bo *old_bo = nouveau_framebuffer(crtc->primary->fb)->nvbo;
-	struct nouveau_bo *new_bo = nouveau_framebuffer(fb)->nvbo;
-	struct nouveau_page_flip_state *s;
-	struct nouveau_channel *chan;
-	struct nouveau_cli *cli;
-	struct nouveau_fence *fence;
-	struct nv04_display *dispnv04 = nv04_display(dev);
-	int head = nouveau_crtc(crtc)->index;
-	int ret;
-
-	chan = drm->channel;
-	if (!chan)
-		return -ENODEV;
-	cli = (void *)chan->user.client;
-
-	s = kzalloc(sizeof(*s), GFP_KERNEL);
-	if (!s)
-		return -ENOMEM;
-
-	if (new_bo != old_bo) {
-		ret = nouveau_bo_pin(new_bo, TTM_PL_FLAG_VRAM, true);
-		if (ret)
-			goto fail_free;
-	}
-
-	mutex_lock(&cli->mutex);
-	ret = ttm_bo_reserve(&new_bo->bo, true, false, NULL);
-	if (ret)
-		goto fail_unpin;
-
-	/* synchronise rendering channel with the kernel's channel */
-	ret = nouveau_fence_sync(new_bo, chan, false, true);
-	if (ret) {
-		ttm_bo_unreserve(&new_bo->bo);
-		goto fail_unpin;
-	}
-
-	if (new_bo != old_bo) {
-		ttm_bo_unreserve(&new_bo->bo);
-
-		ret = ttm_bo_reserve(&old_bo->bo, true, false, NULL);
-		if (ret)
-			goto fail_unpin;
-	}
-
-	/* Initialize a page flip struct */
-	*s = (struct nouveau_page_flip_state)
-		{ { }, event, crtc, fb->format->cpp[0] * 8, fb->pitches[0],
-		  new_bo->bo.offset };
-
-	/* Keep vblanks on during flip, for the target crtc of this flip */
-	drm_crtc_vblank_get(crtc);
-
-	/* Emit a page flip */
-	if (swap_interval) {
-		ret = RING_SPACE(chan, 8);
-		if (ret)
-			goto fail_unreserve;
-
-		BEGIN_NV04(chan, NvSubImageBlit, 0x012c, 1);
-		OUT_RING  (chan, 0);
-		BEGIN_NV04(chan, NvSubImageBlit, 0x0134, 1);
-		OUT_RING  (chan, head);
-		BEGIN_NV04(chan, NvSubImageBlit, 0x0100, 1);
-		OUT_RING  (chan, 0);
-		BEGIN_NV04(chan, NvSubImageBlit, 0x0130, 1);
-		OUT_RING  (chan, 0);
-	}
-
-	nouveau_bo_ref(new_bo, &dispnv04->image[head]);
-
-	ret = nouveau_page_flip_emit(chan, old_bo, new_bo, s, &fence);
-	if (ret)
-		goto fail_unreserve;
-	mutex_unlock(&cli->mutex);
-
-	/* Update the crtc struct and cleanup */
-	crtc->primary->fb = fb;
-
-	nouveau_bo_fence(old_bo, fence, false);
-	ttm_bo_unreserve(&old_bo->bo);
-	if (old_bo != new_bo)
-		nouveau_bo_unpin(old_bo);
-	nouveau_fence_unref(&fence);
-	return 0;
-
-fail_unreserve:
-	drm_crtc_vblank_put(crtc);
-	ttm_bo_unreserve(&old_bo->bo);
-fail_unpin:
-	mutex_unlock(&cli->mutex);
-	if (old_bo != new_bo)
-		nouveau_bo_unpin(new_bo);
-fail_free:
-	kfree(s);
-	return ret;
-}
-
-int
-nouveau_finish_page_flip(struct nouveau_channel *chan,
-			 struct nouveau_page_flip_state *ps)
-{
-	struct nouveau_fence_chan *fctx = chan->fence;
-	struct nouveau_drm *drm = chan->drm;
-	struct drm_device *dev = drm->dev;
-	struct nouveau_page_flip_state *s;
-	unsigned long flags;
-
-	spin_lock_irqsave(&dev->event_lock, flags);
-
-	if (list_empty(&fctx->flip)) {
-		NV_ERROR(drm, "unexpected pageflip\n");
-		spin_unlock_irqrestore(&dev->event_lock, flags);
-		return -EINVAL;
-	}
-
-	s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head);
-	if (s->event) {
-		drm_crtc_arm_vblank_event(s->crtc, s->event);
-	} else {
-		/* Give up ownership of vblank for page-flipped crtc */
-		drm_crtc_vblank_put(s->crtc);
-	}
-
-	list_del(&s->head);
-	if (ps)
-		*ps = *s;
-	kfree(s);
-
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-	return 0;
-}
-
-int
-nouveau_flip_complete(struct nvif_notify *notify)
-{
-	struct nouveau_drm *drm = container_of(notify, typeof(*drm), flip);
-	struct nouveau_channel *chan = drm->channel;
-	struct nouveau_page_flip_state state;
-
-	if (!nouveau_finish_page_flip(chan, &state)) {
-		nv_set_crtc_base(drm->dev, drm_crtc_index(state.crtc),
-				 state.offset + state.crtc->y *
-				 state.pitch + state.crtc->x *
-				 state.bpp / 8);
-	}
-
-	return NVIF_NOTIFY_KEEP;
 }
 
 int
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index eb77e41c2d4e..311e175f0513 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -25,19 +25,11 @@ int nouveau_framebuffer_new(struct drm_device *,
 			    const struct drm_mode_fb_cmd2 *,
 			    struct nouveau_bo *, struct nouveau_framebuffer **);
 
-struct nouveau_page_flip_state {
-	struct list_head head;
-	struct drm_pending_vblank_event *event;
-	struct drm_crtc *crtc;
-	int bpp, pitch;
-	u64 offset;
-};
-
 struct nouveau_display {
 	void *priv;
 	void (*dtor)(struct drm_device *);
-	int  (*init)(struct drm_device *);
-	void (*fini)(struct drm_device *);
+	int  (*init)(struct drm_device *, bool resume, bool runtime);
+	void (*fini)(struct drm_device *, bool suspend);
 
 	struct nvif_disp disp;
 
@@ -61,7 +53,7 @@ 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);
+int  nouveau_display_init(struct drm_device *dev, bool resume, bool runtime);
 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);
@@ -71,13 +63,6 @@ bool  nouveau_display_scanoutpos(struct drm_device *, unsigned int,
 				 bool, int *, int *, ktime_t *,
 				 ktime_t *, const struct drm_display_mode *);
 
-int  nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
-			    struct drm_pending_vblank_event *event,
-			    uint32_t page_flip_flags,
-			    struct drm_modeset_acquire_ctx *ctx);
-int  nouveau_finish_page_flip(struct nouveau_channel *,
-			      struct nouveau_page_flip_state *);
-
 int  nouveau_display_dumb_create(struct drm_file *, struct drm_device *,
 				 struct drm_mode_create_dumb *args);
 int  nouveau_display_dumb_map_offset(struct drm_file *, struct drm_device *,
diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
new file mode 100644
index 000000000000..8be7a83ced9b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
@@ -0,0 +1,887 @@
+/*
+ * Copyright 2018 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "nouveau_dmem.h"
+#include "nouveau_drv.h"
+#include "nouveau_chan.h"
+#include "nouveau_dma.h"
+#include "nouveau_mem.h"
+#include "nouveau_bo.h"
+
+#include <nvif/class.h>
+#include <nvif/object.h>
+#include <nvif/if500b.h>
+#include <nvif/if900b.h>
+
+#include <linux/sched/mm.h>
+#include <linux/hmm.h>
+
+/*
+ * FIXME: this is ugly right now we are using TTM to allocate vram and we pin
+ * it in vram while in use. We likely want to overhaul memory management for
+ * nouveau to be more page like (not necessarily with system page size but a
+ * bigger page size) at lowest level and have some shim layer on top that would
+ * provide the same functionality as TTM.
+ */
+#define DMEM_CHUNK_SIZE (2UL << 20)
+#define DMEM_CHUNK_NPAGES (DMEM_CHUNK_SIZE >> PAGE_SHIFT)
+
+struct nouveau_migrate;
+
+enum nouveau_aper {
+	NOUVEAU_APER_VIRT,
+	NOUVEAU_APER_VRAM,
+	NOUVEAU_APER_HOST,
+};
+
+typedef int (*nouveau_migrate_copy_t)(struct nouveau_drm *drm, u64 npages,
+				      enum nouveau_aper, u64 dst_addr,
+				      enum nouveau_aper, u64 src_addr);
+
+struct nouveau_dmem_chunk {
+	struct list_head list;
+	struct nouveau_bo *bo;
+	struct nouveau_drm *drm;
+	unsigned long pfn_first;
+	unsigned long callocated;
+	unsigned long bitmap[BITS_TO_LONGS(DMEM_CHUNK_NPAGES)];
+	spinlock_t lock;
+};
+
+struct nouveau_dmem_migrate {
+	nouveau_migrate_copy_t copy_func;
+	struct nouveau_channel *chan;
+};
+
+struct nouveau_dmem {
+	struct hmm_devmem *devmem;
+	struct nouveau_dmem_migrate migrate;
+	struct list_head chunk_free;
+	struct list_head chunk_full;
+	struct list_head chunk_empty;
+	struct mutex mutex;
+};
+
+struct nouveau_dmem_fault {
+	struct nouveau_drm *drm;
+	struct nouveau_fence *fence;
+	dma_addr_t *dma;
+	unsigned long npages;
+};
+
+struct nouveau_migrate {
+	struct vm_area_struct *vma;
+	struct nouveau_drm *drm;
+	struct nouveau_fence *fence;
+	unsigned long npages;
+	dma_addr_t *dma;
+	unsigned long dma_nr;
+};
+
+static void
+nouveau_dmem_free(struct hmm_devmem *devmem, struct page *page)
+{
+	struct nouveau_dmem_chunk *chunk;
+	struct nouveau_drm *drm;
+	unsigned long idx;
+
+	chunk = (void *)hmm_devmem_page_get_drvdata(page);
+	idx = page_to_pfn(page) - chunk->pfn_first;
+	drm = chunk->drm;
+
+	/*
+	 * FIXME:
+	 *
+	 * This is really a bad example, we need to overhaul nouveau memory
+	 * management to be more page focus and allow lighter locking scheme
+	 * to be use in the process.
+	 */
+	spin_lock(&chunk->lock);
+	clear_bit(idx, chunk->bitmap);
+	WARN_ON(!chunk->callocated);
+	chunk->callocated--;
+	/*
+	 * FIXME when chunk->callocated reach 0 we should add the chunk to
+	 * a reclaim list so that it can be freed in case of memory pressure.
+	 */
+	spin_unlock(&chunk->lock);
+}
+
+static void
+nouveau_dmem_fault_alloc_and_copy(struct vm_area_struct *vma,
+				  const unsigned long *src_pfns,
+				  unsigned long *dst_pfns,
+				  unsigned long start,
+				  unsigned long end,
+				  void *private)
+{
+	struct nouveau_dmem_fault *fault = private;
+	struct nouveau_drm *drm = fault->drm;
+	struct device *dev = drm->dev->dev;
+	unsigned long addr, i, npages = 0;
+	nouveau_migrate_copy_t copy;
+	int ret;
+
+
+	/* First allocate new memory */
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, i++) {
+		struct page *dpage, *spage;
+
+		dst_pfns[i] = 0;
+		spage = migrate_pfn_to_page(src_pfns[i]);
+		if (!spage || !(src_pfns[i] & MIGRATE_PFN_MIGRATE))
+			continue;
+
+		dpage = hmm_vma_alloc_locked_page(vma, addr);
+		if (!dpage) {
+			dst_pfns[i] = MIGRATE_PFN_ERROR;
+			continue;
+		}
+
+		dst_pfns[i] = migrate_pfn(page_to_pfn(dpage)) |
+			      MIGRATE_PFN_LOCKED;
+		npages++;
+	}
+
+	/* Allocate storage for DMA addresses, so we can unmap later. */
+	fault->dma = kmalloc(sizeof(*fault->dma) * npages, GFP_KERNEL);
+	if (!fault->dma)
+		goto error;
+
+	/* Copy things over */
+	copy = drm->dmem->migrate.copy_func;
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, i++) {
+		struct nouveau_dmem_chunk *chunk;
+		struct page *spage, *dpage;
+		u64 src_addr, dst_addr;
+
+		dpage = migrate_pfn_to_page(dst_pfns[i]);
+		if (!dpage || dst_pfns[i] == MIGRATE_PFN_ERROR)
+			continue;
+
+		spage = migrate_pfn_to_page(src_pfns[i]);
+		if (!spage || !(src_pfns[i] & MIGRATE_PFN_MIGRATE)) {
+			dst_pfns[i] = MIGRATE_PFN_ERROR;
+			__free_page(dpage);
+			continue;
+		}
+
+		fault->dma[fault->npages] =
+			dma_map_page_attrs(dev, dpage, 0, PAGE_SIZE,
+					   PCI_DMA_BIDIRECTIONAL,
+					   DMA_ATTR_SKIP_CPU_SYNC);
+		if (dma_mapping_error(dev, fault->dma[fault->npages])) {
+			dst_pfns[i] = MIGRATE_PFN_ERROR;
+			__free_page(dpage);
+			continue;
+		}
+
+		dst_addr = fault->dma[fault->npages++];
+
+		chunk = (void *)hmm_devmem_page_get_drvdata(spage);
+		src_addr = page_to_pfn(spage) - chunk->pfn_first;
+		src_addr = (src_addr << PAGE_SHIFT) + chunk->bo->bo.offset;
+
+		ret = copy(drm, 1, NOUVEAU_APER_HOST, dst_addr,
+				   NOUVEAU_APER_VRAM, src_addr);
+		if (ret) {
+			dst_pfns[i] = MIGRATE_PFN_ERROR;
+			__free_page(dpage);
+			continue;
+		}
+	}
+
+	nouveau_fence_new(drm->dmem->migrate.chan, false, &fault->fence);
+
+	return;
+
+error:
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, ++i) {
+		struct page *page;
+
+		if (!dst_pfns[i] || dst_pfns[i] == MIGRATE_PFN_ERROR)
+			continue;
+
+		page = migrate_pfn_to_page(dst_pfns[i]);
+		dst_pfns[i] = MIGRATE_PFN_ERROR;
+		if (page == NULL)
+			continue;
+
+		__free_page(page);
+	}
+}
+
+void nouveau_dmem_fault_finalize_and_map(struct vm_area_struct *vma,
+					 const unsigned long *src_pfns,
+					 const unsigned long *dst_pfns,
+					 unsigned long start,
+					 unsigned long end,
+					 void *private)
+{
+	struct nouveau_dmem_fault *fault = private;
+	struct nouveau_drm *drm = fault->drm;
+
+	if (fault->fence) {
+		nouveau_fence_wait(fault->fence, true, false);
+		nouveau_fence_unref(&fault->fence);
+	} else {
+		/*
+		 * FIXME wait for channel to be IDLE before calling finalizing
+		 * the hmem object below (nouveau_migrate_hmem_fini()).
+		 */
+	}
+
+	while (fault->npages--) {
+		dma_unmap_page(drm->dev->dev, fault->dma[fault->npages],
+			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+	}
+	kfree(fault->dma);
+}
+
+static const struct migrate_vma_ops nouveau_dmem_fault_migrate_ops = {
+	.alloc_and_copy		= nouveau_dmem_fault_alloc_and_copy,
+	.finalize_and_map	= nouveau_dmem_fault_finalize_and_map,
+};
+
+static int
+nouveau_dmem_fault(struct hmm_devmem *devmem,
+		   struct vm_area_struct *vma,
+		   unsigned long addr,
+		   const struct page *page,
+		   unsigned int flags,
+		   pmd_t *pmdp)
+{
+	struct drm_device *drm_dev = dev_get_drvdata(devmem->device);
+	unsigned long src[1] = {0}, dst[1] = {0};
+	struct nouveau_dmem_fault fault = {0};
+	int ret;
+
+
+
+	/*
+	 * FIXME what we really want is to find some heuristic to migrate more
+	 * than just one page on CPU fault. When such fault happens it is very
+	 * likely that more surrounding page will CPU fault too.
+	 */
+	fault.drm = nouveau_drm(drm_dev);
+	ret = migrate_vma(&nouveau_dmem_fault_migrate_ops, vma, addr,
+			  addr + PAGE_SIZE, src, dst, &fault);
+	if (ret)
+		return VM_FAULT_SIGBUS;
+
+	if (dst[0] == MIGRATE_PFN_ERROR)
+		return VM_FAULT_SIGBUS;
+
+	return 0;
+}
+
+static const struct hmm_devmem_ops
+nouveau_dmem_devmem_ops = {
+	.free = nouveau_dmem_free,
+	.fault = nouveau_dmem_fault,
+};
+
+static int
+nouveau_dmem_chunk_alloc(struct nouveau_drm *drm)
+{
+	struct nouveau_dmem_chunk *chunk;
+	int ret;
+
+	if (drm->dmem == NULL)
+		return -EINVAL;
+
+	mutex_lock(&drm->dmem->mutex);
+	chunk = list_first_entry_or_null(&drm->dmem->chunk_empty,
+					 struct nouveau_dmem_chunk,
+					 list);
+	if (chunk == NULL) {
+		mutex_unlock(&drm->dmem->mutex);
+		return -ENOMEM;
+	}
+
+	list_del(&chunk->list);
+	mutex_unlock(&drm->dmem->mutex);
+
+	ret = nouveau_bo_new(&drm->client, DMEM_CHUNK_SIZE, 0,
+			     TTM_PL_FLAG_VRAM, 0, 0, NULL, NULL,
+			     &chunk->bo);
+	if (ret)
+		goto out;
+
+	ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+	if (ret) {
+		nouveau_bo_ref(NULL, &chunk->bo);
+		goto out;
+	}
+
+	bitmap_zero(chunk->bitmap, DMEM_CHUNK_NPAGES);
+	spin_lock_init(&chunk->lock);
+
+out:
+	mutex_lock(&drm->dmem->mutex);
+	if (chunk->bo)
+		list_add(&chunk->list, &drm->dmem->chunk_empty);
+	else
+		list_add_tail(&chunk->list, &drm->dmem->chunk_empty);
+	mutex_unlock(&drm->dmem->mutex);
+
+	return ret;
+}
+
+static struct nouveau_dmem_chunk *
+nouveau_dmem_chunk_first_free_locked(struct nouveau_drm *drm)
+{
+	struct nouveau_dmem_chunk *chunk;
+
+	chunk = list_first_entry_or_null(&drm->dmem->chunk_free,
+					 struct nouveau_dmem_chunk,
+					 list);
+	if (chunk)
+		return chunk;
+
+	chunk = list_first_entry_or_null(&drm->dmem->chunk_empty,
+					 struct nouveau_dmem_chunk,
+					 list);
+	if (chunk->bo)
+		return chunk;
+
+	return NULL;
+}
+
+static int
+nouveau_dmem_pages_alloc(struct nouveau_drm *drm,
+			 unsigned long npages,
+			 unsigned long *pages)
+{
+	struct nouveau_dmem_chunk *chunk;
+	unsigned long c;
+	int ret;
+
+	memset(pages, 0xff, npages * sizeof(*pages));
+
+	mutex_lock(&drm->dmem->mutex);
+	for (c = 0; c < npages;) {
+		unsigned long i;
+
+		chunk = nouveau_dmem_chunk_first_free_locked(drm);
+		if (chunk == NULL) {
+			mutex_unlock(&drm->dmem->mutex);
+			ret = nouveau_dmem_chunk_alloc(drm);
+			if (ret) {
+				if (c)
+					break;
+				return ret;
+			}
+			continue;
+		}
+
+		spin_lock(&chunk->lock);
+		i = find_first_zero_bit(chunk->bitmap, DMEM_CHUNK_NPAGES);
+		while (i < DMEM_CHUNK_NPAGES && c < npages) {
+			pages[c] = chunk->pfn_first + i;
+			set_bit(i, chunk->bitmap);
+			chunk->callocated++;
+			c++;
+
+			i = find_next_zero_bit(chunk->bitmap,
+					DMEM_CHUNK_NPAGES, i);
+		}
+		spin_unlock(&chunk->lock);
+	}
+	mutex_unlock(&drm->dmem->mutex);
+
+	return 0;
+}
+
+static struct page *
+nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm)
+{
+	unsigned long pfns[1];
+	struct page *page;
+	int ret;
+
+	/* FIXME stop all the miss-match API ... */
+	ret = nouveau_dmem_pages_alloc(drm, 1, pfns);
+	if (ret)
+		return NULL;
+
+	page = pfn_to_page(pfns[0]);
+	get_page(page);
+	lock_page(page);
+	return page;
+}
+
+static void
+nouveau_dmem_page_free_locked(struct nouveau_drm *drm, struct page *page)
+{
+	unlock_page(page);
+	put_page(page);
+}
+
+void
+nouveau_dmem_resume(struct nouveau_drm *drm)
+{
+	struct nouveau_dmem_chunk *chunk;
+	int ret;
+
+	if (drm->dmem == NULL)
+		return;
+
+	mutex_lock(&drm->dmem->mutex);
+	list_for_each_entry (chunk, &drm->dmem->chunk_free, list) {
+		ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+		/* FIXME handle pin failure */
+		WARN_ON(ret);
+	}
+	list_for_each_entry (chunk, &drm->dmem->chunk_full, list) {
+		ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+		/* FIXME handle pin failure */
+		WARN_ON(ret);
+	}
+	list_for_each_entry (chunk, &drm->dmem->chunk_empty, list) {
+		ret = nouveau_bo_pin(chunk->bo, TTM_PL_FLAG_VRAM, false);
+		/* FIXME handle pin failure */
+		WARN_ON(ret);
+	}
+	mutex_unlock(&drm->dmem->mutex);
+}
+
+void
+nouveau_dmem_suspend(struct nouveau_drm *drm)
+{
+	struct nouveau_dmem_chunk *chunk;
+
+	if (drm->dmem == NULL)
+		return;
+
+	mutex_lock(&drm->dmem->mutex);
+	list_for_each_entry (chunk, &drm->dmem->chunk_free, list) {
+		nouveau_bo_unpin(chunk->bo);
+	}
+	list_for_each_entry (chunk, &drm->dmem->chunk_full, list) {
+		nouveau_bo_unpin(chunk->bo);
+	}
+	list_for_each_entry (chunk, &drm->dmem->chunk_empty, list) {
+		nouveau_bo_unpin(chunk->bo);
+	}
+	mutex_unlock(&drm->dmem->mutex);
+}
+
+void
+nouveau_dmem_fini(struct nouveau_drm *drm)
+{
+	struct nouveau_dmem_chunk *chunk, *tmp;
+
+	if (drm->dmem == NULL)
+		return;
+
+	mutex_lock(&drm->dmem->mutex);
+
+	WARN_ON(!list_empty(&drm->dmem->chunk_free));
+	WARN_ON(!list_empty(&drm->dmem->chunk_full));
+
+	list_for_each_entry_safe (chunk, tmp, &drm->dmem->chunk_empty, list) {
+		if (chunk->bo) {
+			nouveau_bo_unpin(chunk->bo);
+			nouveau_bo_ref(NULL, &chunk->bo);
+		}
+		list_del(&chunk->list);
+		kfree(chunk);
+	}
+
+	mutex_unlock(&drm->dmem->mutex);
+}
+
+static int
+nvc0b5_migrate_copy(struct nouveau_drm *drm, u64 npages,
+		    enum nouveau_aper dst_aper, u64 dst_addr,
+		    enum nouveau_aper src_aper, u64 src_addr)
+{
+	struct nouveau_channel *chan = drm->dmem->migrate.chan;
+	u32 launch_dma = (1 << 9) /* MULTI_LINE_ENABLE. */ |
+			 (1 << 8) /* DST_MEMORY_LAYOUT_PITCH. */ |
+			 (1 << 7) /* SRC_MEMORY_LAYOUT_PITCH. */ |
+			 (1 << 2) /* FLUSH_ENABLE_TRUE. */ |
+			 (2 << 0) /* DATA_TRANSFER_TYPE_NON_PIPELINED. */;
+	int ret;
+
+	ret = RING_SPACE(chan, 13);
+	if (ret)
+		return ret;
+
+	if (src_aper != NOUVEAU_APER_VIRT) {
+		switch (src_aper) {
+		case NOUVEAU_APER_VRAM:
+			BEGIN_IMC0(chan, NvSubCopy, 0x0260, 0);
+			break;
+		case NOUVEAU_APER_HOST:
+			BEGIN_IMC0(chan, NvSubCopy, 0x0260, 1);
+			break;
+		default:
+			return -EINVAL;
+		}
+		launch_dma |= 0x00001000; /* SRC_TYPE_PHYSICAL. */
+	}
+
+	if (dst_aper != NOUVEAU_APER_VIRT) {
+		switch (dst_aper) {
+		case NOUVEAU_APER_VRAM:
+			BEGIN_IMC0(chan, NvSubCopy, 0x0264, 0);
+			break;
+		case NOUVEAU_APER_HOST:
+			BEGIN_IMC0(chan, NvSubCopy, 0x0264, 1);
+			break;
+		default:
+			return -EINVAL;
+		}
+		launch_dma |= 0x00002000; /* DST_TYPE_PHYSICAL. */
+	}
+
+	BEGIN_NVC0(chan, NvSubCopy, 0x0400, 8);
+	OUT_RING  (chan, upper_32_bits(src_addr));
+	OUT_RING  (chan, lower_32_bits(src_addr));
+	OUT_RING  (chan, upper_32_bits(dst_addr));
+	OUT_RING  (chan, lower_32_bits(dst_addr));
+	OUT_RING  (chan, PAGE_SIZE);
+	OUT_RING  (chan, PAGE_SIZE);
+	OUT_RING  (chan, PAGE_SIZE);
+	OUT_RING  (chan, npages);
+	BEGIN_NVC0(chan, NvSubCopy, 0x0300, 1);
+	OUT_RING  (chan, launch_dma);
+	return 0;
+}
+
+static int
+nouveau_dmem_migrate_init(struct nouveau_drm *drm)
+{
+	switch (drm->ttm.copy.oclass) {
+	case PASCAL_DMA_COPY_A:
+	case PASCAL_DMA_COPY_B:
+	case  VOLTA_DMA_COPY_A:
+	case TURING_DMA_COPY_A:
+		drm->dmem->migrate.copy_func = nvc0b5_migrate_copy;
+		drm->dmem->migrate.chan = drm->ttm.chan;
+		return 0;
+	default:
+		break;
+	}
+	return -ENODEV;
+}
+
+void
+nouveau_dmem_init(struct nouveau_drm *drm)
+{
+	struct device *device = drm->dev->dev;
+	unsigned long i, size;
+	int ret;
+
+	/* This only make sense on PASCAL or newer */
+	if (drm->client.device.info.family < NV_DEVICE_INFO_V0_PASCAL)
+		return;
+
+	if (!(drm->dmem = kzalloc(sizeof(*drm->dmem), GFP_KERNEL)))
+		return;
+
+	mutex_init(&drm->dmem->mutex);
+	INIT_LIST_HEAD(&drm->dmem->chunk_free);
+	INIT_LIST_HEAD(&drm->dmem->chunk_full);
+	INIT_LIST_HEAD(&drm->dmem->chunk_empty);
+
+	size = ALIGN(drm->client.device.info.ram_user, DMEM_CHUNK_SIZE);
+
+	/* Initialize migration dma helpers before registering memory */
+	ret = nouveau_dmem_migrate_init(drm);
+	if (ret) {
+		kfree(drm->dmem);
+		drm->dmem = NULL;
+		return;
+	}
+
+	/*
+	 * FIXME we need some kind of policy to decide how much VRAM we
+	 * want to register with HMM. For now just register everything
+	 * and latter if we want to do thing like over commit then we
+	 * could revisit this.
+	 */
+	drm->dmem->devmem = hmm_devmem_add(&nouveau_dmem_devmem_ops,
+					   device, size);
+	if (drm->dmem->devmem == NULL) {
+		kfree(drm->dmem);
+		drm->dmem = NULL;
+		return;
+	}
+
+	for (i = 0; i < (size / DMEM_CHUNK_SIZE); ++i) {
+		struct nouveau_dmem_chunk *chunk;
+		struct page *page;
+		unsigned long j;
+
+		chunk = kzalloc(sizeof(*chunk), GFP_KERNEL);
+		if (chunk == NULL) {
+			nouveau_dmem_fini(drm);
+			return;
+		}
+
+		chunk->drm = drm;
+		chunk->pfn_first = drm->dmem->devmem->pfn_first;
+		chunk->pfn_first += (i * DMEM_CHUNK_NPAGES);
+		list_add_tail(&chunk->list, &drm->dmem->chunk_empty);
+
+		page = pfn_to_page(chunk->pfn_first);
+		for (j = 0; j < DMEM_CHUNK_NPAGES; ++j, ++page) {
+			hmm_devmem_page_set_drvdata(page, (long)chunk);
+		}
+	}
+
+	NV_INFO(drm, "DMEM: registered %ldMB of device memory\n", size >> 20);
+}
+
+static void
+nouveau_dmem_migrate_alloc_and_copy(struct vm_area_struct *vma,
+				    const unsigned long *src_pfns,
+				    unsigned long *dst_pfns,
+				    unsigned long start,
+				    unsigned long end,
+				    void *private)
+{
+	struct nouveau_migrate *migrate = private;
+	struct nouveau_drm *drm = migrate->drm;
+	struct device *dev = drm->dev->dev;
+	unsigned long addr, i, npages = 0;
+	nouveau_migrate_copy_t copy;
+	int ret;
+
+	/* First allocate new memory */
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, i++) {
+		struct page *dpage, *spage;
+
+		dst_pfns[i] = 0;
+		spage = migrate_pfn_to_page(src_pfns[i]);
+		if (!spage || !(src_pfns[i] & MIGRATE_PFN_MIGRATE))
+			continue;
+
+		dpage = nouveau_dmem_page_alloc_locked(drm);
+		if (!dpage)
+			continue;
+
+		dst_pfns[i] = migrate_pfn(page_to_pfn(dpage)) |
+			      MIGRATE_PFN_LOCKED |
+			      MIGRATE_PFN_DEVICE;
+		npages++;
+	}
+
+	if (!npages)
+		return;
+
+	/* Allocate storage for DMA addresses, so we can unmap later. */
+	migrate->dma = kmalloc(sizeof(*migrate->dma) * npages, GFP_KERNEL);
+	if (!migrate->dma)
+		goto error;
+
+	/* Copy things over */
+	copy = drm->dmem->migrate.copy_func;
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, i++) {
+		struct nouveau_dmem_chunk *chunk;
+		struct page *spage, *dpage;
+		u64 src_addr, dst_addr;
+
+		dpage = migrate_pfn_to_page(dst_pfns[i]);
+		if (!dpage || dst_pfns[i] == MIGRATE_PFN_ERROR)
+			continue;
+
+		chunk = (void *)hmm_devmem_page_get_drvdata(dpage);
+		dst_addr = page_to_pfn(dpage) - chunk->pfn_first;
+		dst_addr = (dst_addr << PAGE_SHIFT) + chunk->bo->bo.offset;
+
+		spage = migrate_pfn_to_page(src_pfns[i]);
+		if (!spage || !(src_pfns[i] & MIGRATE_PFN_MIGRATE)) {
+			nouveau_dmem_page_free_locked(drm, dpage);
+			dst_pfns[i] = 0;
+			continue;
+		}
+
+		migrate->dma[migrate->dma_nr] =
+			dma_map_page_attrs(dev, spage, 0, PAGE_SIZE,
+					   PCI_DMA_BIDIRECTIONAL,
+					   DMA_ATTR_SKIP_CPU_SYNC);
+		if (dma_mapping_error(dev, migrate->dma[migrate->dma_nr])) {
+			nouveau_dmem_page_free_locked(drm, dpage);
+			dst_pfns[i] = 0;
+			continue;
+		}
+
+		src_addr = migrate->dma[migrate->dma_nr++];
+
+		ret = copy(drm, 1, NOUVEAU_APER_VRAM, dst_addr,
+				   NOUVEAU_APER_HOST, src_addr);
+		if (ret) {
+			nouveau_dmem_page_free_locked(drm, dpage);
+			dst_pfns[i] = 0;
+			continue;
+		}
+	}
+
+	nouveau_fence_new(drm->dmem->migrate.chan, false, &migrate->fence);
+
+	return;
+
+error:
+	for (addr = start, i = 0; addr < end; addr += PAGE_SIZE, ++i) {
+		struct page *page;
+
+		if (!dst_pfns[i] || dst_pfns[i] == MIGRATE_PFN_ERROR)
+			continue;
+
+		page = migrate_pfn_to_page(dst_pfns[i]);
+		dst_pfns[i] = MIGRATE_PFN_ERROR;
+		if (page == NULL)
+			continue;
+
+		__free_page(page);
+	}
+}
+
+void nouveau_dmem_migrate_finalize_and_map(struct vm_area_struct *vma,
+					   const unsigned long *src_pfns,
+					   const unsigned long *dst_pfns,
+					   unsigned long start,
+					   unsigned long end,
+					   void *private)
+{
+	struct nouveau_migrate *migrate = private;
+	struct nouveau_drm *drm = migrate->drm;
+
+	if (migrate->fence) {
+		nouveau_fence_wait(migrate->fence, true, false);
+		nouveau_fence_unref(&migrate->fence);
+	} else {
+		/*
+		 * FIXME wait for channel to be IDLE before finalizing
+		 * the hmem object below (nouveau_migrate_hmem_fini()) ?
+		 */
+	}
+
+	while (migrate->dma_nr--) {
+		dma_unmap_page(drm->dev->dev, migrate->dma[migrate->dma_nr],
+			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+	}
+	kfree(migrate->dma);
+
+	/*
+	 * FIXME optimization: update GPU page table to point to newly
+	 * migrated memory.
+	 */
+}
+
+static const struct migrate_vma_ops nouveau_dmem_migrate_ops = {
+	.alloc_and_copy		= nouveau_dmem_migrate_alloc_and_copy,
+	.finalize_and_map	= nouveau_dmem_migrate_finalize_and_map,
+};
+
+int
+nouveau_dmem_migrate_vma(struct nouveau_drm *drm,
+			 struct vm_area_struct *vma,
+			 unsigned long start,
+			 unsigned long end)
+{
+	unsigned long *src_pfns, *dst_pfns, npages;
+	struct nouveau_migrate migrate = {0};
+	unsigned long i, c, max;
+	int ret = 0;
+
+	npages = (end - start) >> PAGE_SHIFT;
+	max = min(SG_MAX_SINGLE_ALLOC, npages);
+	src_pfns = kzalloc(sizeof(long) * max, GFP_KERNEL);
+	if (src_pfns == NULL)
+		return -ENOMEM;
+	dst_pfns = kzalloc(sizeof(long) * max, GFP_KERNEL);
+	if (dst_pfns == NULL) {
+		kfree(src_pfns);
+		return -ENOMEM;
+	}
+
+	migrate.drm = drm;
+	migrate.vma = vma;
+	migrate.npages = npages;
+	for (i = 0; i < npages; i += c) {
+		unsigned long next;
+
+		c = min(SG_MAX_SINGLE_ALLOC, npages);
+		next = start + (c << PAGE_SHIFT);
+		ret = migrate_vma(&nouveau_dmem_migrate_ops, vma, start,
+				  next, src_pfns, dst_pfns, &migrate);
+		if (ret)
+			goto out;
+		start = next;
+	}
+
+out:
+	kfree(dst_pfns);
+	kfree(src_pfns);
+	return ret;
+}
+
+static inline bool
+nouveau_dmem_page(struct nouveau_drm *drm, struct page *page)
+{
+	if (!is_device_private_page(page))
+		return false;
+
+	if (drm->dmem->devmem != page->pgmap->data)
+		return false;
+
+	return true;
+}
+
+void
+nouveau_dmem_convert_pfn(struct nouveau_drm *drm,
+			 struct hmm_range *range)
+{
+	unsigned long i, npages;
+
+	npages = (range->end - range->start) >> PAGE_SHIFT;
+	for (i = 0; i < npages; ++i) {
+		struct nouveau_dmem_chunk *chunk;
+		struct page *page;
+		uint64_t addr;
+
+		page = hmm_pfn_to_page(range, range->pfns[i]);
+		if (page == NULL)
+			continue;
+
+		if (!(range->pfns[i] & range->flags[HMM_PFN_DEVICE_PRIVATE])) {
+			continue;
+		}
+
+		if (!nouveau_dmem_page(drm, page)) {
+			WARN(1, "Some unknown device memory !\n");
+			range->pfns[i] = 0;
+			continue;
+		}
+
+		chunk = (void *)hmm_devmem_page_get_drvdata(page);
+		addr = page_to_pfn(page) - chunk->pfn_first;
+		addr = (addr + chunk->bo->bo.mem.start) << PAGE_SHIFT;
+
+		range->pfns[i] &= ((1UL << range->pfn_shift) - 1);
+		range->pfns[i] |= (addr >> PAGE_SHIFT) << range->pfn_shift;
+	}
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.h b/drivers/gpu/drm/nouveau/nouveau_dmem.h
new file mode 100644
index 000000000000..9d97d756fb7d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.h
@@ -0,0 +1,60 @@
+/*
+ * Copyright 2018 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#ifndef __NOUVEAU_DMEM_H__
+#define __NOUVEAU_DMEM_H__
+#include <nvif/os.h>
+struct drm_device;
+struct drm_file;
+struct nouveau_drm;
+struct hmm_range;
+
+#if IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM)
+void nouveau_dmem_init(struct nouveau_drm *);
+void nouveau_dmem_fini(struct nouveau_drm *);
+void nouveau_dmem_suspend(struct nouveau_drm *);
+void nouveau_dmem_resume(struct nouveau_drm *);
+
+int nouveau_dmem_migrate_vma(struct nouveau_drm *drm,
+			     struct vm_area_struct *vma,
+			     unsigned long start,
+			     unsigned long end);
+
+void nouveau_dmem_convert_pfn(struct nouveau_drm *drm,
+			      struct hmm_range *range);
+#else /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */
+static inline void nouveau_dmem_init(struct nouveau_drm *drm) {}
+static inline void nouveau_dmem_fini(struct nouveau_drm *drm) {}
+static inline void nouveau_dmem_suspend(struct nouveau_drm *drm) {}
+static inline void nouveau_dmem_resume(struct nouveau_drm *drm) {}
+
+static inline int nouveau_dmem_migrate_vma(struct nouveau_drm *drm,
+					   struct vm_area_struct *vma,
+					   unsigned long start,
+					   unsigned long end)
+{
+	return 0;
+}
+
+static inline void nouveau_dmem_convert_pfn(struct nouveau_drm *drm,
+					    struct hmm_range *range) {}
+#endif /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index f900e94592f8..5020265bfbd9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -44,7 +44,6 @@
 #include <nvif/class.h>
 #include <nvif/cl0002.h>
 #include <nvif/cla06f.h>
-#include <nvif/if0004.h>
 
 #include "nouveau_drv.h"
 #include "nouveau_dma.h"
@@ -63,6 +62,8 @@
 #include "nouveau_usif.h"
 #include "nouveau_connector.h"
 #include "nouveau_platform.h"
+#include "nouveau_svm.h"
+#include "nouveau_dmem.h"
 
 MODULE_PARM_DESC(config, "option string to pass to driver core");
 static char *nouveau_config;
@@ -173,6 +174,7 @@ nouveau_cli_fini(struct nouveau_cli *cli)
 	WARN_ON(!list_empty(&cli->worker));
 
 	usif_client_fini(cli);
+	nouveau_vmm_fini(&cli->svm);
 	nouveau_vmm_fini(&cli->vmm);
 	nvif_mmu_fini(&cli->mmu);
 	nvif_device_fini(&cli->device);
@@ -283,19 +285,134 @@ done:
 }
 
 static void
-nouveau_accel_fini(struct nouveau_drm *drm)
+nouveau_accel_ce_fini(struct nouveau_drm *drm)
+{
+	nouveau_channel_idle(drm->cechan);
+	nvif_object_fini(&drm->ttm.copy);
+	nouveau_channel_del(&drm->cechan);
+}
+
+static void
+nouveau_accel_ce_init(struct nouveau_drm *drm)
+{
+	struct nvif_device *device = &drm->client.device;
+	int ret = 0;
+
+	/* Allocate channel that has access to a (preferably async) copy
+	 * engine, to use for TTM buffer moves.
+	 */
+	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
+		ret = nouveau_channel_new(drm, device,
+					  nvif_fifo_runlist_ce(device), 0,
+					  true, &drm->cechan);
+	} else
+	if (device->info.chipset >= 0xa3 &&
+	    device->info.chipset != 0xaa &&
+	    device->info.chipset != 0xac) {
+		/* Prior to Kepler, there's only a single runlist, so all
+		 * engines can be accessed from any channel.
+		 *
+		 * We still want to use a separate channel though.
+		 */
+		ret = nouveau_channel_new(drm, device, NvDmaFB, NvDmaTT, false,
+					  &drm->cechan);
+	}
+
+	if (ret)
+		NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
+}
+
+static void
+nouveau_accel_gr_fini(struct nouveau_drm *drm)
 {
 	nouveau_channel_idle(drm->channel);
 	nvif_object_fini(&drm->ntfy);
 	nvkm_gpuobj_del(&drm->notify);
-	nvif_notify_fini(&drm->flip);
 	nvif_object_fini(&drm->nvsw);
 	nouveau_channel_del(&drm->channel);
+}
 
-	nouveau_channel_idle(drm->cechan);
-	nvif_object_fini(&drm->ttm.copy);
-	nouveau_channel_del(&drm->cechan);
+static void
+nouveau_accel_gr_init(struct nouveau_drm *drm)
+{
+	struct nvif_device *device = &drm->client.device;
+	u32 arg0, arg1;
+	int ret;
+
+	/* Allocate channel that has access to the graphics engine. */
+	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
+		arg0 = nvif_fifo_runlist(device, NV_DEVICE_INFO_ENGINE_GR);
+		arg1 = 1;
+	} else {
+		arg0 = NvDmaFB;
+		arg1 = NvDmaTT;
+	}
 
+	ret = nouveau_channel_new(drm, device, arg0, arg1, false,
+				  &drm->channel);
+	if (ret) {
+		NV_ERROR(drm, "failed to create kernel channel, %d\n", ret);
+		nouveau_accel_gr_fini(drm);
+		return;
+	}
+
+	/* A SW class is used on pre-NV50 HW to assist with handling the
+	 * synchronisation of page flips, as well as to implement fences
+	 * on TNT/TNT2 HW that lacks any kind of support in host.
+	 */
+	if (device->info.family < NV_DEVICE_INFO_V0_TESLA) {
+		ret = nvif_object_init(&drm->channel->user, NVDRM_NVSW,
+				       nouveau_abi16_swclass(drm), NULL, 0,
+				       &drm->nvsw);
+		if (ret == 0) {
+			ret = RING_SPACE(drm->channel, 2);
+			if (ret == 0) {
+				BEGIN_NV04(drm->channel, NvSubSw, 0, 1);
+				OUT_RING  (drm->channel, drm->nvsw.handle);
+			}
+		}
+
+		if (ret) {
+			NV_ERROR(drm, "failed to allocate sw class, %d\n", ret);
+			nouveau_accel_gr_fini(drm);
+			return;
+		}
+	}
+
+	/* NvMemoryToMemoryFormat requires a notifier ctxdma for some reason,
+	 * even if notification is never requested, so, allocate a ctxdma on
+	 * any GPU where it's possible we'll end up using M2MF for BO moves.
+	 */
+	if (device->info.family < NV_DEVICE_INFO_V0_FERMI) {
+		ret = nvkm_gpuobj_new(nvxx_device(device), 32, 0, false, NULL,
+				      &drm->notify);
+		if (ret) {
+			NV_ERROR(drm, "failed to allocate notifier, %d\n", ret);
+			nouveau_accel_gr_fini(drm);
+			return;
+		}
+
+		ret = nvif_object_init(&drm->channel->user, NvNotify0,
+				       NV_DMA_IN_MEMORY,
+				       &(struct nv_dma_v0) {
+						.target = NV_DMA_V0_TARGET_VRAM,
+						.access = NV_DMA_V0_ACCESS_RDWR,
+						.start = drm->notify->addr,
+						.limit = drm->notify->addr + 31
+				       }, sizeof(struct nv_dma_v0),
+				       &drm->ntfy);
+		if (ret) {
+			nouveau_accel_gr_fini(drm);
+			return;
+		}
+	}
+}
+
+static void
+nouveau_accel_fini(struct nouveau_drm *drm)
+{
+	nouveau_accel_ce_fini(drm);
+	nouveau_accel_gr_fini(drm);
 	if (drm->fence)
 		nouveau_fence(drm)->dtor(drm);
 }
@@ -305,23 +422,16 @@ nouveau_accel_init(struct nouveau_drm *drm)
 {
 	struct nvif_device *device = &drm->client.device;
 	struct nvif_sclass *sclass;
-	u32 arg0, arg1;
 	int ret, i, n;
 
 	if (nouveau_noaccel)
 		return;
 
+	/* Initialise global support for channels, and synchronisation. */
 	ret = nouveau_channels_init(drm);
 	if (ret)
 		return;
 
-	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_VOLTA) {
-		ret = nvif_user_init(device);
-		if (ret)
-			return;
-	}
-
-	/* initialise synchronisation routines */
 	/*XXX: this is crap, but the fence/channel stuff is a little
 	 *     backwards in some places.  this will be fixed.
 	 */
@@ -368,95 +478,18 @@ nouveau_accel_init(struct nouveau_drm *drm)
 		return;
 	}
 
-	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
-		ret = nouveau_channel_new(drm, &drm->client.device,
-					  nvif_fifo_runlist_ce(device), 0,
-					  true, &drm->cechan);
-		if (ret)
-			NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
-
-		arg0 = nvif_fifo_runlist(device, NV_DEVICE_INFO_ENGINE_GR);
-		arg1 = 1;
-	} else
-	if (device->info.chipset >= 0xa3 &&
-	    device->info.chipset != 0xaa &&
-	    device->info.chipset != 0xac) {
-		ret = nouveau_channel_new(drm, &drm->client.device,
-					  NvDmaFB, NvDmaTT, false,
-					  &drm->cechan);
+	/* Volta requires access to a doorbell register for kickoff. */
+	if (drm->client.device.info.family >= NV_DEVICE_INFO_V0_VOLTA) {
+		ret = nvif_user_init(device);
 		if (ret)
-			NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
-
-		arg0 = NvDmaFB;
-		arg1 = NvDmaTT;
-	} else {
-		arg0 = NvDmaFB;
-		arg1 = NvDmaTT;
-	}
-
-	ret = nouveau_channel_new(drm, &drm->client.device,
-				  arg0, arg1, false, &drm->channel);
-	if (ret) {
-		NV_ERROR(drm, "failed to create kernel channel, %d\n", ret);
-		nouveau_accel_fini(drm);
-		return;
-	}
-
-	if (device->info.family < NV_DEVICE_INFO_V0_TESLA) {
-		ret = nvif_object_init(&drm->channel->user, NVDRM_NVSW,
-				       nouveau_abi16_swclass(drm), NULL, 0,
-				       &drm->nvsw);
-		if (ret == 0) {
-			ret = RING_SPACE(drm->channel, 2);
-			if (ret == 0) {
-				BEGIN_NV04(drm->channel, NvSubSw, 0, 1);
-				OUT_RING  (drm->channel, drm->nvsw.handle);
-			}
-
-			ret = nvif_notify_init(&drm->nvsw,
-					       nouveau_flip_complete,
-					       false, NV04_NVSW_NTFY_UEVENT,
-					       NULL, 0, 0, &drm->flip);
-			if (ret == 0)
-				ret = nvif_notify_get(&drm->flip);
-			if (ret) {
-				nouveau_accel_fini(drm);
-				return;
-			}
-		}
-
-		if (ret) {
-			NV_ERROR(drm, "failed to allocate sw class, %d\n", ret);
-			nouveau_accel_fini(drm);
-			return;
-		}
-	}
-
-	if (device->info.family < NV_DEVICE_INFO_V0_FERMI) {
-		ret = nvkm_gpuobj_new(nvxx_device(&drm->client.device), 32, 0,
-				      false, NULL, &drm->notify);
-		if (ret) {
-			NV_ERROR(drm, "failed to allocate notifier, %d\n", ret);
-			nouveau_accel_fini(drm);
 			return;
-		}
-
-		ret = nvif_object_init(&drm->channel->user, NvNotify0,
-				       NV_DMA_IN_MEMORY,
-				       &(struct nv_dma_v0) {
-						.target = NV_DMA_V0_TARGET_VRAM,
-						.access = NV_DMA_V0_ACCESS_RDWR,
-						.start = drm->notify->addr,
-						.limit = drm->notify->addr + 31
-				       }, sizeof(struct nv_dma_v0),
-				       &drm->ntfy);
-		if (ret) {
-			nouveau_accel_fini(drm);
-			return;
-		}
 	}
 
+	/* Allocate channels we need to support various functions. */
+	nouveau_accel_gr_init(drm);
+	nouveau_accel_ce_init(drm);
 
+	/* Initialise accelerated TTM buffer moves. */
 	nouveau_bo_move_init(drm);
 }
 
@@ -504,19 +537,22 @@ nouveau_drm_device_init(struct drm_device *dev)
 	if (ret)
 		goto fail_bios;
 
+	nouveau_accel_init(drm);
+
 	ret = nouveau_display_create(dev);
 	if (ret)
 		goto fail_dispctor;
 
 	if (dev->mode_config.num_crtc) {
-		ret = nouveau_display_init(dev);
+		ret = nouveau_display_init(dev, false, false);
 		if (ret)
 			goto fail_dispinit;
 	}
 
 	nouveau_debugfs_init(drm);
 	nouveau_hwmon_init(dev);
-	nouveau_accel_init(drm);
+	nouveau_svm_init(drm);
+	nouveau_dmem_init(drm);
 	nouveau_fbcon_init(dev);
 	nouveau_led_init(dev);
 
@@ -534,6 +570,7 @@ nouveau_drm_device_init(struct drm_device *dev)
 fail_dispinit:
 	nouveau_display_destroy(dev);
 fail_dispctor:
+	nouveau_accel_fini(drm);
 	nouveau_bios_takedown(dev);
 fail_bios:
 	nouveau_ttm_fini(drm);
@@ -559,7 +596,8 @@ nouveau_drm_device_fini(struct drm_device *dev)
 
 	nouveau_led_fini(dev);
 	nouveau_fbcon_fini(dev);
-	nouveau_accel_fini(drm);
+	nouveau_dmem_fini(drm);
+	nouveau_svm_fini(drm);
 	nouveau_hwmon_fini(dev);
 	nouveau_debugfs_fini(drm);
 
@@ -567,6 +605,7 @@ nouveau_drm_device_fini(struct drm_device *dev)
 		nouveau_display_fini(dev, false, false);
 	nouveau_display_destroy(dev);
 
+	nouveau_accel_fini(drm);
 	nouveau_bios_takedown(dev);
 
 	nouveau_ttm_fini(drm);
@@ -704,6 +743,8 @@ nouveau_do_suspend(struct drm_device *dev, bool runtime)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	int ret;
 
+	nouveau_svm_suspend(drm);
+	nouveau_dmem_suspend(drm);
 	nouveau_led_suspend(dev);
 
 	if (dev->mode_config.num_crtc) {
@@ -780,7 +821,8 @@ nouveau_do_resume(struct drm_device *dev, bool runtime)
 	}
 
 	nouveau_led_resume(dev);
-
+	nouveau_dmem_resume(drm);
+	nouveau_svm_resume(drm);
 	return 0;
 }
 
@@ -1000,6 +1042,8 @@ nouveau_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(NOUVEAU_GROBJ_ALLOC, nouveau_abi16_ioctl_grobj_alloc, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(NOUVEAU_NOTIFIEROBJ_ALLOC, nouveau_abi16_ioctl_notifierobj_alloc, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(NOUVEAU_GPUOBJ_FREE, nouveau_abi16_ioctl_gpuobj_free, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_SVM_INIT, nouveau_svmm_init, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_SVM_BIND, nouveau_svmm_bind, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_NEW, nouveau_gem_ioctl_new, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_PUSHBUF, nouveau_gem_ioctl_pushbuf, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_CPU_PREP, nouveau_gem_ioctl_cpu_prep, DRM_AUTH|DRM_RENDER_ALLOW),
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index d20b9ba4b1c1..da847244479d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -96,6 +96,7 @@ struct nouveau_cli {
 	struct nvif_device device;
 	struct nvif_mmu mmu;
 	struct nouveau_vmm vmm;
+	struct nouveau_vmm svm;
 	const struct nvif_mclass *mem;
 
 	struct list_head head;
@@ -181,7 +182,6 @@ struct nouveau_drm {
 	struct nouveau_fbdev *fbcon;
 	struct nvif_object nvsw;
 	struct nvif_object ntfy;
-	struct nvif_notify flip;
 
 	/* nv10-nv40 tiling regions */
 	struct {
@@ -210,6 +210,10 @@ struct nouveau_drm {
 	bool have_disp_power_ref;
 
 	struct dev_pm_domain vga_pm_domain;
+
+	struct nouveau_svm *svm;
+
+	struct nouveau_dmem *dmem;
 };
 
 static inline struct nouveau_drm *
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index 032317c81bf0..0d3cd4e05728 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -353,7 +353,7 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 
 	chan = nouveau_nofbaccel ? NULL : drm->channel;
 	if (chan && device->info.family >= NV_DEVICE_INFO_V0_TESLA) {
-		ret = nouveau_vma_new(nvbo, &drm->client.vmm, &fb->vma);
+		ret = nouveau_vma_new(nvbo, chan->vmm, &fb->vma);
 		if (ret) {
 			NV_ERROR(drm, "failed to map fb into chan: %d\n", ret);
 			chan = NULL;
@@ -374,9 +374,9 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 
 	strcpy(info->fix.id, "nouveaufb");
 	if (!chan)
-		info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_DISABLED;
+		info->flags = FBINFO_HWACCEL_DISABLED;
 	else
-		info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_COPYAREA |
+		info->flags = FBINFO_HWACCEL_COPYAREA |
 			      FBINFO_HWACCEL_FILLRECT |
 			      FBINFO_HWACCEL_IMAGEBLIT;
 	info->fbops = &nouveau_fbcon_sw_ops;
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h
index b999e6058046..ad27caeca0fd 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.h
@@ -82,8 +82,6 @@ int nv50_fence_create(struct nouveau_drm *);
 int nv84_fence_create(struct nouveau_drm *);
 int nvc0_fence_create(struct nouveau_drm *);
 
-int nouveau_flip_complete(struct nvif_notify *);
-
 struct nv84_fence_chan {
 	struct nouveau_fence_chan base;
 	struct nouveau_vma *vma;
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index b56524d343c3..b4bda716564d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -41,7 +41,6 @@ nouveau_gem_object_del(struct drm_gem_object *gem)
 {
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
-	struct ttm_buffer_object *bo = &nvbo->bo;
 	struct device *dev = drm->dev->dev;
 	int ret;
 
@@ -56,7 +55,7 @@ nouveau_gem_object_del(struct drm_gem_object *gem)
 
 	/* reset filp so nouveau_bo_del_ttm() can test for it */
 	gem->filp = NULL;
-	ttm_bo_unref(&bo);
+	ttm_bo_put(&nvbo->bo);
 
 	pm_runtime_mark_last_busy(dev);
 	pm_runtime_put_autosuspend(dev);
@@ -69,10 +68,11 @@ nouveau_gem_object_open(struct drm_gem_object *gem, struct drm_file *file_priv)
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	struct device *dev = drm->dev->dev;
+	struct nouveau_vmm *vmm = cli->svm.cli ? &cli->svm : &cli->vmm;
 	struct nouveau_vma *vma;
 	int ret;
 
-	if (cli->vmm.vmm.object.oclass < NVIF_CLASS_VMM_NV50)
+	if (vmm->vmm.object.oclass < NVIF_CLASS_VMM_NV50)
 		return 0;
 
 	ret = ttm_bo_reserve(&nvbo->bo, false, false, NULL);
@@ -83,7 +83,7 @@ nouveau_gem_object_open(struct drm_gem_object *gem, struct drm_file *file_priv)
 	if (ret < 0 && ret != -EACCES)
 		goto out;
 
-	ret = nouveau_vma_new(nvbo, &cli->vmm, &vma);
+	ret = nouveau_vma_new(nvbo, vmm, &vma);
 	pm_runtime_mark_last_busy(dev);
 	pm_runtime_put_autosuspend(dev);
 out:
@@ -143,17 +143,18 @@ nouveau_gem_object_close(struct drm_gem_object *gem, struct drm_file *file_priv)
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
 	struct device *dev = drm->dev->dev;
+	struct nouveau_vmm *vmm = cli->svm.cli ? &cli->svm : & cli->vmm;
 	struct nouveau_vma *vma;
 	int ret;
 
-	if (cli->vmm.vmm.object.oclass < NVIF_CLASS_VMM_NV50)
+	if (vmm->vmm.object.oclass < NVIF_CLASS_VMM_NV50)
 		return;
 
 	ret = ttm_bo_reserve(&nvbo->bo, false, false, NULL);
 	if (ret)
 		return;
 
-	vma = nouveau_vma_find(nvbo, &cli->vmm);
+	vma = nouveau_vma_find(nvbo, vmm);
 	if (vma) {
 		if (--vma->refs == 0) {
 			ret = pm_runtime_get_sync(dev);
@@ -220,6 +221,7 @@ nouveau_gem_info(struct drm_file *file_priv, struct drm_gem_object *gem,
 {
 	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
+	struct nouveau_vmm *vmm = cli->svm.cli ? &cli->svm : &cli->vmm;
 	struct nouveau_vma *vma;
 
 	if (is_power_of_2(nvbo->valid_domains))
@@ -229,8 +231,8 @@ nouveau_gem_info(struct drm_file *file_priv, struct drm_gem_object *gem,
 	else
 		rep->domain = NOUVEAU_GEM_DOMAIN_VRAM;
 	rep->offset = nvbo->bo.offset;
-	if (cli->vmm.vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
-		vma = nouveau_vma_find(nvbo, &cli->vmm);
+	if (vmm->vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
+		vma = nouveau_vma_find(nvbo, vmm);
 		if (!vma)
 			return -EINVAL;
 
@@ -322,7 +324,8 @@ struct validate_op {
 };
 
 static void
-validate_fini_no_ticket(struct validate_op *op, struct nouveau_fence *fence,
+validate_fini_no_ticket(struct validate_op *op, struct nouveau_channel *chan,
+			struct nouveau_fence *fence,
 			struct drm_nouveau_gem_pushbuf_bo *pbbo)
 {
 	struct nouveau_bo *nvbo;
@@ -333,13 +336,11 @@ validate_fini_no_ticket(struct validate_op *op, struct nouveau_fence *fence,
 		b = &pbbo[nvbo->pbbo_index];
 
 		if (likely(fence)) {
-			struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev);
-			struct nouveau_vma *vma;
-
 			nouveau_bo_fence(nvbo, fence, !!b->write_domains);
 
-			if (drm->client.vmm.vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
-				vma = (void *)(unsigned long)b->user_priv;
+			if (chan->vmm->vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
+				struct nouveau_vma *vma =
+					(void *)(unsigned long)b->user_priv;
 				nouveau_fence_unref(&vma->fence);
 				dma_fence_get(&fence->base);
 				vma->fence = fence;
@@ -359,10 +360,11 @@ validate_fini_no_ticket(struct validate_op *op, struct nouveau_fence *fence,
 }
 
 static void
-validate_fini(struct validate_op *op, struct nouveau_fence *fence,
+validate_fini(struct validate_op *op, struct nouveau_channel *chan,
+	      struct nouveau_fence *fence,
 	      struct drm_nouveau_gem_pushbuf_bo *pbbo)
 {
-	validate_fini_no_ticket(op, fence, pbbo);
+	validate_fini_no_ticket(op, chan, fence, pbbo);
 	ww_acquire_fini(&op->ticket);
 }
 
@@ -417,7 +419,7 @@ retry:
 			list_splice_tail_init(&vram_list, &op->list);
 			list_splice_tail_init(&gart_list, &op->list);
 			list_splice_tail_init(&both_list, &op->list);
-			validate_fini_no_ticket(op, NULL, NULL);
+			validate_fini_no_ticket(op, chan, NULL, NULL);
 			if (unlikely(ret == -EDEADLK)) {
 				ret = ttm_bo_reserve_slowpath(&nvbo->bo, true,
 							      &op->ticket);
@@ -431,8 +433,8 @@ retry:
 			}
 		}
 
-		if (cli->vmm.vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
-			struct nouveau_vmm *vmm = &cli->vmm;
+		if (chan->vmm->vmm.object.oclass >= NVIF_CLASS_VMM_NV50) {
+			struct nouveau_vmm *vmm = chan->vmm;
 			struct nouveau_vma *vma = nouveau_vma_find(nvbo, vmm);
 			if (!vma) {
 				NV_PRINTK(err, cli, "vma not found!\n");
@@ -472,7 +474,7 @@ retry:
 	list_splice_tail(&gart_list, &op->list);
 	list_splice_tail(&both_list, &op->list);
 	if (ret)
-		validate_fini(op, NULL, NULL);
+		validate_fini(op, chan, NULL, NULL);
 	return ret;
 
 }
@@ -564,7 +566,7 @@ nouveau_gem_pushbuf_validate(struct nouveau_channel *chan,
 	if (unlikely(ret < 0)) {
 		if (ret != -ERESTARTSYS)
 			NV_PRINTK(err, cli, "validating bo list\n");
-		validate_fini(op, NULL, NULL);
+		validate_fini(op, chan, NULL, NULL);
 		return ret;
 	}
 	*apply_relocs = ret;
@@ -843,7 +845,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 	}
 
 out:
-	validate_fini(&op, fence, bo);
+	validate_fini(&op, chan, fence, bo);
 	nouveau_fence_unref(&fence);
 
 out_prevalid:
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c
new file mode 100644
index 000000000000..93ed43c413f0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -0,0 +1,835 @@
+/*
+ * Copyright 2018 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "nouveau_svm.h"
+#include "nouveau_drv.h"
+#include "nouveau_chan.h"
+#include "nouveau_dmem.h"
+
+#include <nvif/notify.h>
+#include <nvif/object.h>
+#include <nvif/vmm.h>
+
+#include <nvif/class.h>
+#include <nvif/clb069.h>
+#include <nvif/ifc00d.h>
+
+#include <linux/sched/mm.h>
+#include <linux/sort.h>
+#include <linux/hmm.h>
+
+struct nouveau_svm {
+	struct nouveau_drm *drm;
+	struct mutex mutex;
+	struct list_head inst;
+
+	struct nouveau_svm_fault_buffer {
+		int id;
+		struct nvif_object object;
+		u32 entries;
+		u32 getaddr;
+		u32 putaddr;
+		u32 get;
+		u32 put;
+		struct nvif_notify notify;
+
+		struct nouveau_svm_fault {
+			u64 inst;
+			u64 addr;
+			u64 time;
+			u32 engine;
+			u8  gpc;
+			u8  hub;
+			u8  access;
+			u8  client;
+			u8  fault;
+			struct nouveau_svmm *svmm;
+		} **fault;
+		int fault_nr;
+	} buffer[1];
+};
+
+#define SVM_DBG(s,f,a...) NV_DEBUG((s)->drm, "svm: "f"\n", ##a)
+#define SVM_ERR(s,f,a...) NV_WARN((s)->drm, "svm: "f"\n", ##a)
+
+struct nouveau_ivmm {
+	struct nouveau_svmm *svmm;
+	u64 inst;
+	struct list_head head;
+};
+
+static struct nouveau_ivmm *
+nouveau_ivmm_find(struct nouveau_svm *svm, u64 inst)
+{
+	struct nouveau_ivmm *ivmm;
+	list_for_each_entry(ivmm, &svm->inst, head) {
+		if (ivmm->inst == inst)
+			return ivmm;
+	}
+	return NULL;
+}
+
+struct nouveau_svmm {
+	struct nouveau_vmm *vmm;
+	struct {
+		unsigned long start;
+		unsigned long limit;
+	} unmanaged;
+
+	struct mutex mutex;
+
+	struct mm_struct *mm;
+	struct hmm_mirror mirror;
+};
+
+#define SVMM_DBG(s,f,a...)                                                     \
+	NV_DEBUG((s)->vmm->cli->drm, "svm-%p: "f"\n", (s), ##a)
+#define SVMM_ERR(s,f,a...)                                                     \
+	NV_WARN((s)->vmm->cli->drm, "svm-%p: "f"\n", (s), ##a)
+
+int
+nouveau_svmm_bind(struct drm_device *dev, void *data,
+		  struct drm_file *file_priv)
+{
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
+	struct drm_nouveau_svm_bind *args = data;
+	unsigned target, cmd, priority;
+	unsigned long addr, end, size;
+	struct mm_struct *mm;
+
+	args->va_start &= PAGE_MASK;
+	args->va_end &= PAGE_MASK;
+
+	/* Sanity check arguments */
+	if (args->reserved0 || args->reserved1)
+		return -EINVAL;
+	if (args->header & (~NOUVEAU_SVM_BIND_VALID_MASK))
+		return -EINVAL;
+	if (args->va_start >= args->va_end)
+		return -EINVAL;
+	if (!args->npages)
+		return -EINVAL;
+
+	cmd = args->header >> NOUVEAU_SVM_BIND_COMMAND_SHIFT;
+	cmd &= NOUVEAU_SVM_BIND_COMMAND_MASK;
+	switch (cmd) {
+	case NOUVEAU_SVM_BIND_COMMAND__MIGRATE:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	priority = args->header >> NOUVEAU_SVM_BIND_PRIORITY_SHIFT;
+	priority &= NOUVEAU_SVM_BIND_PRIORITY_MASK;
+
+	/* FIXME support CPU target ie all target value < GPU_VRAM */
+	target = args->header >> NOUVEAU_SVM_BIND_TARGET_SHIFT;
+	target &= NOUVEAU_SVM_BIND_TARGET_MASK;
+	switch (target) {
+	case NOUVEAU_SVM_BIND_TARGET__GPU_VRAM:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/*
+	 * FIXME: For now refuse non 0 stride, we need to change the migrate
+	 * kernel function to handle stride to avoid to create a mess within
+	 * each device driver.
+	 */
+	if (args->stride)
+		return -EINVAL;
+
+	size = ((unsigned long)args->npages) << PAGE_SHIFT;
+	if ((args->va_start + size) <= args->va_start)
+		return -EINVAL;
+	if ((args->va_start + size) > args->va_end)
+		return -EINVAL;
+
+	/*
+	 * Ok we are ask to do something sane, for now we only support migrate
+	 * commands but we will add things like memory policy (what to do on
+	 * page fault) and maybe some other commands.
+	 */
+
+	mm = get_task_mm(current);
+	down_read(&mm->mmap_sem);
+
+	for (addr = args->va_start, end = args->va_start + size; addr < end;) {
+		struct vm_area_struct *vma;
+		unsigned long next;
+
+		vma = find_vma_intersection(mm, addr, end);
+		if (!vma)
+			break;
+
+		next = min(vma->vm_end, end);
+		/* This is a best effort so we ignore errors */
+		nouveau_dmem_migrate_vma(cli->drm, vma, addr, next);
+		addr = next;
+	}
+
+	/*
+	 * FIXME Return the number of page we have migrated, again we need to
+	 * update the migrate API to return that information so that we can
+	 * report it to user space.
+	 */
+	args->result = 0;
+
+	up_read(&mm->mmap_sem);
+	mmput(mm);
+
+	return 0;
+}
+
+/* Unlink channel instance from SVMM. */
+void
+nouveau_svmm_part(struct nouveau_svmm *svmm, u64 inst)
+{
+	struct nouveau_ivmm *ivmm;
+	if (svmm) {
+		mutex_lock(&svmm->vmm->cli->drm->svm->mutex);
+		ivmm = nouveau_ivmm_find(svmm->vmm->cli->drm->svm, inst);
+		if (ivmm) {
+			list_del(&ivmm->head);
+			kfree(ivmm);
+		}
+		mutex_unlock(&svmm->vmm->cli->drm->svm->mutex);
+	}
+}
+
+/* Link channel instance to SVMM. */
+int
+nouveau_svmm_join(struct nouveau_svmm *svmm, u64 inst)
+{
+	struct nouveau_ivmm *ivmm;
+	if (svmm) {
+		if (!(ivmm = kmalloc(sizeof(*ivmm), GFP_KERNEL)))
+			return -ENOMEM;
+		ivmm->svmm = svmm;
+		ivmm->inst = inst;
+
+		mutex_lock(&svmm->vmm->cli->drm->svm->mutex);
+		list_add(&ivmm->head, &svmm->vmm->cli->drm->svm->inst);
+		mutex_unlock(&svmm->vmm->cli->drm->svm->mutex);
+	}
+	return 0;
+}
+
+/* Invalidate SVMM address-range on GPU. */
+static void
+nouveau_svmm_invalidate(struct nouveau_svmm *svmm, u64 start, u64 limit)
+{
+	if (limit > start) {
+		bool super = svmm->vmm->vmm.object.client->super;
+		svmm->vmm->vmm.object.client->super = true;
+		nvif_object_mthd(&svmm->vmm->vmm.object, NVIF_VMM_V0_PFNCLR,
+				 &(struct nvif_vmm_pfnclr_v0) {
+					.addr = start,
+					.size = limit - start,
+				 }, sizeof(struct nvif_vmm_pfnclr_v0));
+		svmm->vmm->vmm.object.client->super = super;
+	}
+}
+
+static int
+nouveau_svmm_sync_cpu_device_pagetables(struct hmm_mirror *mirror,
+					const struct hmm_update *update)
+{
+	struct nouveau_svmm *svmm = container_of(mirror, typeof(*svmm), mirror);
+	unsigned long start = update->start;
+	unsigned long limit = update->end;
+
+	if (!update->blockable)
+		return -EAGAIN;
+
+	SVMM_DBG(svmm, "invalidate %016lx-%016lx", start, limit);
+
+	mutex_lock(&svmm->mutex);
+	if (limit > svmm->unmanaged.start && start < svmm->unmanaged.limit) {
+		if (start < svmm->unmanaged.start) {
+			nouveau_svmm_invalidate(svmm, start,
+						svmm->unmanaged.limit);
+		}
+		start = svmm->unmanaged.limit;
+	}
+
+	nouveau_svmm_invalidate(svmm, start, limit);
+	mutex_unlock(&svmm->mutex);
+	return 0;
+}
+
+static void
+nouveau_svmm_release(struct hmm_mirror *mirror)
+{
+}
+
+static const struct hmm_mirror_ops
+nouveau_svmm = {
+	.sync_cpu_device_pagetables = nouveau_svmm_sync_cpu_device_pagetables,
+	.release = nouveau_svmm_release,
+};
+
+void
+nouveau_svmm_fini(struct nouveau_svmm **psvmm)
+{
+	struct nouveau_svmm *svmm = *psvmm;
+	if (svmm) {
+		hmm_mirror_unregister(&svmm->mirror);
+		kfree(*psvmm);
+		*psvmm = NULL;
+	}
+}
+
+int
+nouveau_svmm_init(struct drm_device *dev, void *data,
+		  struct drm_file *file_priv)
+{
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
+	struct nouveau_svmm *svmm;
+	struct drm_nouveau_svm_init *args = data;
+	int ret;
+
+	/* Allocate tracking for SVM-enabled VMM. */
+	if (!(svmm = kzalloc(sizeof(*svmm), GFP_KERNEL)))
+		return -ENOMEM;
+	svmm->vmm = &cli->svm;
+	svmm->unmanaged.start = args->unmanaged_addr;
+	svmm->unmanaged.limit = args->unmanaged_addr + args->unmanaged_size;
+	mutex_init(&svmm->mutex);
+
+	/* Check that SVM isn't already enabled for the client. */
+	mutex_lock(&cli->mutex);
+	if (cli->svm.cli) {
+		ret = -EBUSY;
+		goto done;
+	}
+
+	/* Allocate a new GPU VMM that can support SVM (managed by the
+	 * client, with replayable faults enabled).
+	 *
+	 * All future channel/memory allocations will make use of this
+	 * VMM instead of the standard one.
+	 */
+	ret = nvif_vmm_init(&cli->mmu, cli->vmm.vmm.object.oclass, true,
+			    args->unmanaged_addr, args->unmanaged_size,
+			    &(struct gp100_vmm_v0) {
+				.fault_replay = true,
+			    }, sizeof(struct gp100_vmm_v0), &cli->svm.vmm);
+	if (ret)
+		goto done;
+
+	/* Enable HMM mirroring of CPU address-space to VMM. */
+	svmm->mm = get_task_mm(current);
+	down_write(&svmm->mm->mmap_sem);
+	svmm->mirror.ops = &nouveau_svmm;
+	ret = hmm_mirror_register(&svmm->mirror, svmm->mm);
+	if (ret == 0) {
+		cli->svm.svmm = svmm;
+		cli->svm.cli = cli;
+	}
+	up_write(&svmm->mm->mmap_sem);
+	mmput(svmm->mm);
+
+done:
+	if (ret)
+		nouveau_svmm_fini(&svmm);
+	mutex_unlock(&cli->mutex);
+	return ret;
+}
+
+static const u64
+nouveau_svm_pfn_flags[HMM_PFN_FLAG_MAX] = {
+	[HMM_PFN_VALID         ] = NVIF_VMM_PFNMAP_V0_V,
+	[HMM_PFN_WRITE         ] = NVIF_VMM_PFNMAP_V0_W,
+	[HMM_PFN_DEVICE_PRIVATE] = NVIF_VMM_PFNMAP_V0_VRAM,
+};
+
+static const u64
+nouveau_svm_pfn_values[HMM_PFN_VALUE_MAX] = {
+	[HMM_PFN_ERROR  ] = ~NVIF_VMM_PFNMAP_V0_V,
+	[HMM_PFN_NONE   ] =  NVIF_VMM_PFNMAP_V0_NONE,
+	[HMM_PFN_SPECIAL] = ~NVIF_VMM_PFNMAP_V0_V,
+};
+
+/* Issue fault replay for GPU to retry accesses that faulted previously. */
+static void
+nouveau_svm_fault_replay(struct nouveau_svm *svm)
+{
+	SVM_DBG(svm, "replay");
+	WARN_ON(nvif_object_mthd(&svm->drm->client.vmm.vmm.object,
+				 GP100_VMM_VN_FAULT_REPLAY,
+				 &(struct gp100_vmm_fault_replay_vn) {},
+				 sizeof(struct gp100_vmm_fault_replay_vn)));
+}
+
+/* Cancel a replayable fault that could not be handled.
+ *
+ * Cancelling the fault will trigger recovery to reset the engine
+ * and kill the offending channel (ie. GPU SIGSEGV).
+ */
+static void
+nouveau_svm_fault_cancel(struct nouveau_svm *svm,
+			 u64 inst, u8 hub, u8 gpc, u8 client)
+{
+	SVM_DBG(svm, "cancel %016llx %d %02x %02x", inst, hub, gpc, client);
+	WARN_ON(nvif_object_mthd(&svm->drm->client.vmm.vmm.object,
+				 GP100_VMM_VN_FAULT_CANCEL,
+				 &(struct gp100_vmm_fault_cancel_v0) {
+					.hub = hub,
+					.gpc = gpc,
+					.client = client,
+					.inst = inst,
+				 }, sizeof(struct gp100_vmm_fault_cancel_v0)));
+}
+
+static void
+nouveau_svm_fault_cancel_fault(struct nouveau_svm *svm,
+			       struct nouveau_svm_fault *fault)
+{
+	nouveau_svm_fault_cancel(svm, fault->inst,
+				      fault->hub,
+				      fault->gpc,
+				      fault->client);
+}
+
+static int
+nouveau_svm_fault_cmp(const void *a, const void *b)
+{
+	const struct nouveau_svm_fault *fa = *(struct nouveau_svm_fault **)a;
+	const struct nouveau_svm_fault *fb = *(struct nouveau_svm_fault **)b;
+	int ret;
+	if ((ret = (s64)fa->inst - fb->inst))
+		return ret;
+	if ((ret = (s64)fa->addr - fb->addr))
+		return ret;
+	/*XXX: atomic? */
+	return (fa->access == 0 || fa->access == 3) -
+	       (fb->access == 0 || fb->access == 3);
+}
+
+static void
+nouveau_svm_fault_cache(struct nouveau_svm *svm,
+			struct nouveau_svm_fault_buffer *buffer, u32 offset)
+{
+	struct nvif_object *memory = &buffer->object;
+	const u32 instlo = nvif_rd32(memory, offset + 0x00);
+	const u32 insthi = nvif_rd32(memory, offset + 0x04);
+	const u32 addrlo = nvif_rd32(memory, offset + 0x08);
+	const u32 addrhi = nvif_rd32(memory, offset + 0x0c);
+	const u32 timelo = nvif_rd32(memory, offset + 0x10);
+	const u32 timehi = nvif_rd32(memory, offset + 0x14);
+	const u32 engine = nvif_rd32(memory, offset + 0x18);
+	const u32   info = nvif_rd32(memory, offset + 0x1c);
+	const u64   inst = (u64)insthi << 32 | instlo;
+	const u8     gpc = (info & 0x1f000000) >> 24;
+	const u8     hub = (info & 0x00100000) >> 20;
+	const u8  client = (info & 0x00007f00) >> 8;
+	struct nouveau_svm_fault *fault;
+
+	//XXX: i think we're supposed to spin waiting */
+	if (WARN_ON(!(info & 0x80000000)))
+		return;
+
+	nvif_mask(memory, offset + 0x1c, 0x80000000, 0x00000000);
+
+	if (!buffer->fault[buffer->fault_nr]) {
+		fault = kmalloc(sizeof(*fault), GFP_KERNEL);
+		if (WARN_ON(!fault)) {
+			nouveau_svm_fault_cancel(svm, inst, hub, gpc, client);
+			return;
+		}
+		buffer->fault[buffer->fault_nr] = fault;
+	}
+
+	fault = buffer->fault[buffer->fault_nr++];
+	fault->inst   = inst;
+	fault->addr   = (u64)addrhi << 32 | addrlo;
+	fault->time   = (u64)timehi << 32 | timelo;
+	fault->engine = engine;
+	fault->gpc    = gpc;
+	fault->hub    = hub;
+	fault->access = (info & 0x000f0000) >> 16;
+	fault->client = client;
+	fault->fault  = (info & 0x0000001f);
+
+	SVM_DBG(svm, "fault %016llx %016llx %02x",
+		fault->inst, fault->addr, fault->access);
+}
+
+static int
+nouveau_svm_fault(struct nvif_notify *notify)
+{
+	struct nouveau_svm_fault_buffer *buffer =
+		container_of(notify, typeof(*buffer), notify);
+	struct nouveau_svm *svm =
+		container_of(buffer, typeof(*svm), buffer[buffer->id]);
+	struct nvif_object *device = &svm->drm->client.device.object;
+	struct nouveau_svmm *svmm;
+	struct {
+		struct {
+			struct nvif_ioctl_v0 i;
+			struct nvif_ioctl_mthd_v0 m;
+			struct nvif_vmm_pfnmap_v0 p;
+		} i;
+		u64 phys[16];
+	} args;
+	struct hmm_range range;
+	struct vm_area_struct *vma;
+	u64 inst, start, limit;
+	int fi, fn, pi, fill;
+	int replay = 0, ret;
+
+	/* Parse available fault buffer entries into a cache, and update
+	 * the GET pointer so HW can reuse the entries.
+	 */
+	SVM_DBG(svm, "fault handler");
+	if (buffer->get == buffer->put) {
+		buffer->put = nvif_rd32(device, buffer->putaddr);
+		buffer->get = nvif_rd32(device, buffer->getaddr);
+		if (buffer->get == buffer->put)
+			return NVIF_NOTIFY_KEEP;
+	}
+	buffer->fault_nr = 0;
+
+	SVM_DBG(svm, "get %08x put %08x", buffer->get, buffer->put);
+	while (buffer->get != buffer->put) {
+		nouveau_svm_fault_cache(svm, buffer, buffer->get * 0x20);
+		if (++buffer->get == buffer->entries)
+			buffer->get = 0;
+	}
+	nvif_wr32(device, buffer->getaddr, buffer->get);
+	SVM_DBG(svm, "%d fault(s) pending", buffer->fault_nr);
+
+	/* Sort parsed faults by instance pointer to prevent unnecessary
+	 * instance to SVMM translations, followed by address and access
+	 * type to reduce the amount of work when handling the faults.
+	 */
+	sort(buffer->fault, buffer->fault_nr, sizeof(*buffer->fault),
+	     nouveau_svm_fault_cmp, NULL);
+
+	/* Lookup SVMM structure for each unique instance pointer. */
+	mutex_lock(&svm->mutex);
+	for (fi = 0, svmm = NULL; fi < buffer->fault_nr; fi++) {
+		if (!svmm || buffer->fault[fi]->inst != inst) {
+			struct nouveau_ivmm *ivmm =
+				nouveau_ivmm_find(svm, buffer->fault[fi]->inst);
+			svmm = ivmm ? ivmm->svmm : NULL;
+			inst = buffer->fault[fi]->inst;
+			SVM_DBG(svm, "inst %016llx -> svm-%p", inst, svmm);
+		}
+		buffer->fault[fi]->svmm = svmm;
+	}
+	mutex_unlock(&svm->mutex);
+
+	/* Process list of faults. */
+	args.i.i.version = 0;
+	args.i.i.type = NVIF_IOCTL_V0_MTHD;
+	args.i.m.version = 0;
+	args.i.m.method = NVIF_VMM_V0_PFNMAP;
+	args.i.p.version = 0;
+
+	for (fi = 0; fn = fi + 1, fi < buffer->fault_nr; fi = fn) {
+		/* Cancel any faults from non-SVM channels. */
+		if (!(svmm = buffer->fault[fi]->svmm)) {
+			nouveau_svm_fault_cancel_fault(svm, buffer->fault[fi]);
+			continue;
+		}
+		SVMM_DBG(svmm, "addr %016llx", buffer->fault[fi]->addr);
+
+		/* We try and group handling of faults within a small
+		 * window into a single update.
+		 */
+		start = buffer->fault[fi]->addr;
+		limit = start + (ARRAY_SIZE(args.phys) << PAGE_SHIFT);
+		if (start < svmm->unmanaged.limit)
+			limit = min_t(u64, limit, svmm->unmanaged.start);
+		else
+		if (limit > svmm->unmanaged.start)
+			start = max_t(u64, start, svmm->unmanaged.limit);
+		SVMM_DBG(svmm, "wndw %016llx-%016llx", start, limit);
+
+		/* Intersect fault window with the CPU VMA, cancelling
+		 * the fault if the address is invalid.
+		 */
+		down_read(&svmm->mm->mmap_sem);
+		vma = find_vma_intersection(svmm->mm, start, limit);
+		if (!vma) {
+			SVMM_ERR(svmm, "wndw %016llx-%016llx", start, limit);
+			up_read(&svmm->mm->mmap_sem);
+			nouveau_svm_fault_cancel_fault(svm, buffer->fault[fi]);
+			continue;
+		}
+		start = max_t(u64, start, vma->vm_start);
+		limit = min_t(u64, limit, vma->vm_end);
+		SVMM_DBG(svmm, "wndw %016llx-%016llx", start, limit);
+
+		if (buffer->fault[fi]->addr != start) {
+			SVMM_ERR(svmm, "addr %016llx", buffer->fault[fi]->addr);
+			up_read(&svmm->mm->mmap_sem);
+			nouveau_svm_fault_cancel_fault(svm, buffer->fault[fi]);
+			continue;
+		}
+
+		/* Prepare the GPU-side update of all pages within the
+		 * fault window, determining required pages and access
+		 * permissions based on pending faults.
+		 */
+		args.i.p.page = PAGE_SHIFT;
+		args.i.p.addr = start;
+		for (fn = fi, pi = 0;;) {
+			/* Determine required permissions based on GPU fault
+			 * access flags.
+			 *XXX: atomic?
+			 */
+			if (buffer->fault[fn]->access != 0 /* READ. */ &&
+			    buffer->fault[fn]->access != 3 /* PREFETCH. */) {
+				args.phys[pi++] = NVIF_VMM_PFNMAP_V0_V |
+						  NVIF_VMM_PFNMAP_V0_W;
+			} else {
+				args.phys[pi++] = NVIF_VMM_PFNMAP_V0_V;
+			}
+			args.i.p.size = pi << PAGE_SHIFT;
+
+			/* It's okay to skip over duplicate addresses from the
+			 * same SVMM as faults are ordered by access type such
+			 * that only the first one needs to be handled.
+			 *
+			 * ie. WRITE faults appear first, thus any handling of
+			 * pending READ faults will already be satisfied.
+			 */
+			while (++fn < buffer->fault_nr &&
+			       buffer->fault[fn]->svmm == svmm &&
+			       buffer->fault[fn    ]->addr ==
+			       buffer->fault[fn - 1]->addr);
+
+			/* If the next fault is outside the window, or all GPU
+			 * faults have been dealt with, we're done here.
+			 */
+			if (fn >= buffer->fault_nr ||
+			    buffer->fault[fn]->svmm != svmm ||
+			    buffer->fault[fn]->addr >= limit)
+				break;
+
+			/* Fill in the gap between this fault and the next. */
+			fill = (buffer->fault[fn    ]->addr -
+				buffer->fault[fn - 1]->addr) >> PAGE_SHIFT;
+			while (--fill)
+				args.phys[pi++] = NVIF_VMM_PFNMAP_V0_NONE;
+		}
+
+		SVMM_DBG(svmm, "wndw %016llx-%016llx covering %d fault(s)",
+			 args.i.p.addr,
+			 args.i.p.addr + args.i.p.size, fn - fi);
+
+		/* Have HMM fault pages within the fault window to the GPU. */
+		range.vma = vma;
+		range.start = args.i.p.addr;
+		range.end = args.i.p.addr + args.i.p.size;
+		range.pfns = args.phys;
+		range.flags = nouveau_svm_pfn_flags;
+		range.values = nouveau_svm_pfn_values;
+		range.pfn_shift = NVIF_VMM_PFNMAP_V0_ADDR_SHIFT;
+again:
+		ret = hmm_vma_fault(&range, true);
+		if (ret == 0) {
+			mutex_lock(&svmm->mutex);
+			if (!hmm_vma_range_done(&range)) {
+				mutex_unlock(&svmm->mutex);
+				goto again;
+			}
+
+			nouveau_dmem_convert_pfn(svm->drm, &range);
+
+			svmm->vmm->vmm.object.client->super = true;
+			ret = nvif_object_ioctl(&svmm->vmm->vmm.object,
+						&args, sizeof(args.i) +
+						pi * sizeof(args.phys[0]),
+						NULL);
+			svmm->vmm->vmm.object.client->super = false;
+			mutex_unlock(&svmm->mutex);
+		}
+		up_read(&svmm->mm->mmap_sem);
+
+		/* Cancel any faults in the window whose pages didn't manage
+		 * to keep their valid bit, or stay writeable when required.
+		 *
+		 * If handling failed completely, cancel all faults.
+		 */
+		while (fi < fn) {
+			struct nouveau_svm_fault *fault = buffer->fault[fi++];
+			pi = (fault->addr - range.start) >> PAGE_SHIFT;
+			if (ret ||
+			     !(range.pfns[pi] & NVIF_VMM_PFNMAP_V0_V) ||
+			    (!(range.pfns[pi] & NVIF_VMM_PFNMAP_V0_W) &&
+			     fault->access != 0 && fault->access != 3)) {
+				nouveau_svm_fault_cancel_fault(svm, fault);
+				continue;
+			}
+			replay++;
+		}
+	}
+
+	/* Issue fault replay to the GPU. */
+	if (replay)
+		nouveau_svm_fault_replay(svm);
+	return NVIF_NOTIFY_KEEP;
+}
+
+static void
+nouveau_svm_fault_buffer_fini(struct nouveau_svm *svm, int id)
+{
+	struct nouveau_svm_fault_buffer *buffer = &svm->buffer[id];
+	nvif_notify_put(&buffer->notify);
+}
+
+static int
+nouveau_svm_fault_buffer_init(struct nouveau_svm *svm, int id)
+{
+	struct nouveau_svm_fault_buffer *buffer = &svm->buffer[id];
+	struct nvif_object *device = &svm->drm->client.device.object;
+	buffer->get = nvif_rd32(device, buffer->getaddr);
+	buffer->put = nvif_rd32(device, buffer->putaddr);
+	SVM_DBG(svm, "get %08x put %08x (init)", buffer->get, buffer->put);
+	return nvif_notify_get(&buffer->notify);
+}
+
+static void
+nouveau_svm_fault_buffer_dtor(struct nouveau_svm *svm, int id)
+{
+	struct nouveau_svm_fault_buffer *buffer = &svm->buffer[id];
+	int i;
+
+	if (buffer->fault) {
+		for (i = 0; buffer->fault[i] && i < buffer->entries; i++)
+			kfree(buffer->fault[i]);
+		kvfree(buffer->fault);
+	}
+
+	nouveau_svm_fault_buffer_fini(svm, id);
+
+	nvif_notify_fini(&buffer->notify);
+	nvif_object_fini(&buffer->object);
+}
+
+static int
+nouveau_svm_fault_buffer_ctor(struct nouveau_svm *svm, s32 oclass, int id)
+{
+	struct nouveau_svm_fault_buffer *buffer = &svm->buffer[id];
+	struct nouveau_drm *drm = svm->drm;
+	struct nvif_object *device = &drm->client.device.object;
+	struct nvif_clb069_v0 args = {};
+	int ret;
+
+	buffer->id = id;
+
+	ret = nvif_object_init(device, 0, oclass, &args, sizeof(args),
+			       &buffer->object);
+	if (ret < 0) {
+		SVM_ERR(svm, "Fault buffer allocation failed: %d", ret);
+		return ret;
+	}
+
+	nvif_object_map(&buffer->object, NULL, 0);
+	buffer->entries = args.entries;
+	buffer->getaddr = args.get;
+	buffer->putaddr = args.put;
+
+	ret = nvif_notify_init(&buffer->object, nouveau_svm_fault, true,
+			       NVB069_V0_NTFY_FAULT, NULL, 0, 0,
+			       &buffer->notify);
+	if (ret)
+		return ret;
+
+	buffer->fault = kvzalloc(sizeof(*buffer->fault) * buffer->entries, GFP_KERNEL);
+	if (!buffer->fault)
+		return -ENOMEM;
+
+	return nouveau_svm_fault_buffer_init(svm, id);
+}
+
+void
+nouveau_svm_resume(struct nouveau_drm *drm)
+{
+	struct nouveau_svm *svm = drm->svm;
+	if (svm)
+		nouveau_svm_fault_buffer_init(svm, 0);
+}
+
+void
+nouveau_svm_suspend(struct nouveau_drm *drm)
+{
+	struct nouveau_svm *svm = drm->svm;
+	if (svm)
+		nouveau_svm_fault_buffer_fini(svm, 0);
+}
+
+void
+nouveau_svm_fini(struct nouveau_drm *drm)
+{
+	struct nouveau_svm *svm = drm->svm;
+	if (svm) {
+		nouveau_svm_fault_buffer_dtor(svm, 0);
+		kfree(drm->svm);
+		drm->svm = NULL;
+	}
+}
+
+void
+nouveau_svm_init(struct nouveau_drm *drm)
+{
+	static const struct nvif_mclass buffers[] = {
+		{   VOLTA_FAULT_BUFFER_A, 0 },
+		{ MAXWELL_FAULT_BUFFER_A, 0 },
+		{}
+	};
+	struct nouveau_svm *svm;
+	int ret;
+
+	/* Disable on Volta and newer until channel recovery is fixed,
+	 * otherwise clients will have a trivial way to trash the GPU
+	 * for everyone.
+	 */
+	if (drm->client.device.info.family > NV_DEVICE_INFO_V0_PASCAL)
+		return;
+
+	if (!(drm->svm = svm = kzalloc(sizeof(*drm->svm), GFP_KERNEL)))
+		return;
+
+	drm->svm->drm = drm;
+	mutex_init(&drm->svm->mutex);
+	INIT_LIST_HEAD(&drm->svm->inst);
+
+	ret = nvif_mclass(&drm->client.device.object, buffers);
+	if (ret < 0) {
+		SVM_DBG(svm, "No supported fault buffer class");
+		nouveau_svm_fini(drm);
+		return;
+	}
+
+	ret = nouveau_svm_fault_buffer_ctor(svm, buffers[ret].oclass, 0);
+	if (ret) {
+		nouveau_svm_fini(drm);
+		return;
+	}
+
+	SVM_DBG(svm, "Initialised");
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.h b/drivers/gpu/drm/nouveau/nouveau_svm.h
new file mode 100644
index 000000000000..e839d8189461
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.h
@@ -0,0 +1,48 @@
+#ifndef __NOUVEAU_SVM_H__
+#define __NOUVEAU_SVM_H__
+#include <nvif/os.h>
+struct drm_device;
+struct drm_file;
+struct nouveau_drm;
+
+struct nouveau_svmm;
+
+#if IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM)
+void nouveau_svm_init(struct nouveau_drm *);
+void nouveau_svm_fini(struct nouveau_drm *);
+void nouveau_svm_suspend(struct nouveau_drm *);
+void nouveau_svm_resume(struct nouveau_drm *);
+
+int nouveau_svmm_init(struct drm_device *, void *, struct drm_file *);
+void nouveau_svmm_fini(struct nouveau_svmm **);
+int nouveau_svmm_join(struct nouveau_svmm *, u64 inst);
+void nouveau_svmm_part(struct nouveau_svmm *, u64 inst);
+int nouveau_svmm_bind(struct drm_device *, void *, struct drm_file *);
+#else /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */
+static inline void nouveau_svm_init(struct nouveau_drm *drm) {}
+static inline void nouveau_svm_fini(struct nouveau_drm *drm) {}
+static inline void nouveau_svm_suspend(struct nouveau_drm *drm) {}
+static inline void nouveau_svm_resume(struct nouveau_drm *drm) {}
+
+static inline int nouveau_svmm_init(struct drm_device *device, void *p,
+				    struct drm_file *file)
+{
+	return -ENOSYS;
+}
+
+static inline void nouveau_svmm_fini(struct nouveau_svmm **svmmp) {}
+
+static inline int nouveau_svmm_join(struct nouveau_svmm *svmm, u64 inst)
+{
+	return 0;
+}
+
+static inline void nouveau_svmm_part(struct nouveau_svmm *svmm, u64 inst) {}
+
+static inline int nouveau_svmm_bind(struct drm_device *device, void *p,
+				    struct drm_file *file)
+{
+	return -ENOSYS;
+}
+#endif /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.c b/drivers/gpu/drm/nouveau/nouveau_vmm.c
index 2032c3e4f6e5..77061182a1cf 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vmm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vmm.c
@@ -22,6 +22,7 @@
 #include "nouveau_vmm.h"
 #include "nouveau_drv.h"
 #include "nouveau_bo.h"
+#include "nouveau_svm.h"
 #include "nouveau_mem.h"
 
 void
@@ -119,6 +120,7 @@ done:
 void
 nouveau_vmm_fini(struct nouveau_vmm *vmm)
 {
+	nouveau_svmm_fini(&vmm->svmm);
 	nvif_vmm_fini(&vmm->vmm);
 	vmm->cli = NULL;
 }
@@ -126,7 +128,7 @@ nouveau_vmm_fini(struct nouveau_vmm *vmm)
 int
 nouveau_vmm_init(struct nouveau_cli *cli, s32 oclass, struct nouveau_vmm *vmm)
 {
-	int ret = nvif_vmm_init(&cli->mmu, oclass, PAGE_SIZE, 0, NULL, 0,
+	int ret = nvif_vmm_init(&cli->mmu, oclass, false, PAGE_SIZE, 0, NULL, 0,
 				&vmm->vmm);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.h b/drivers/gpu/drm/nouveau/nouveau_vmm.h
index ede872f6f668..2b98d975f37e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vmm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_vmm.h
@@ -25,6 +25,7 @@ void nouveau_vma_unmap(struct nouveau_vma *);
 struct nouveau_vmm {
 	struct nouveau_cli *cli;
 	struct nvif_vmm vmm;
+	struct nouveau_svmm *svmm;
 };
 
 int nouveau_vmm_init(struct nouveau_cli *, s32 oclass, struct nouveau_vmm *);
diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c
index e721bb2163a0..f07da00f285f 100644
--- a/drivers/gpu/drm/nouveau/nv84_fence.c
+++ b/drivers/gpu/drm/nouveau/nv84_fence.c
@@ -109,7 +109,6 @@ nv84_fence_context_del(struct nouveau_channel *chan)
 int
 nv84_fence_context_new(struct nouveau_channel *chan)
 {
-	struct nouveau_cli *cli = (void *)chan->user.client;
 	struct nv84_fence_priv *priv = chan->drm->fence;
 	struct nv84_fence_chan *fctx;
 	int ret;
@@ -127,7 +126,7 @@ nv84_fence_context_new(struct nouveau_channel *chan)
 	fctx->base.sequence = nv84_fence_read(chan);
 
 	mutex_lock(&priv->mutex);
-	ret = nouveau_vma_new(priv->bo, &cli->vmm, &fctx->vma);
+	ret = nouveau_vma_new(priv->bo, chan->vmm, &fctx->vma);
 	mutex_unlock(&priv->mutex);
 
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nvif/disp.c b/drivers/gpu/drm/nouveau/nvif/disp.c
index ef97dd223a32..61638b3b9d3d 100644
--- a/drivers/gpu/drm/nouveau/nvif/disp.c
+++ b/drivers/gpu/drm/nouveau/nvif/disp.c
@@ -34,7 +34,7 @@ int
 nvif_disp_ctor(struct nvif_device *device, s32 oclass, struct nvif_disp *disp)
 {
 	static const struct nvif_mclass disps[] = {
-		{ TU104_DISP, -1 },
+		{ TU102_DISP, -1 },
 		{ GV100_DISP, -1 },
 		{ GP102_DISP, -1 },
 		{ GP100_DISP, -1 },
diff --git a/drivers/gpu/drm/nouveau/nvif/vmm.c b/drivers/gpu/drm/nouveau/nvif/vmm.c
index 6b9c5776547f..11487c00b909 100644
--- a/drivers/gpu/drm/nouveau/nvif/vmm.c
+++ b/drivers/gpu/drm/nouveau/nvif/vmm.c
@@ -112,8 +112,8 @@ nvif_vmm_fini(struct nvif_vmm *vmm)
 }
 
 int
-nvif_vmm_init(struct nvif_mmu *mmu, s32 oclass, u64 addr, u64 size,
-	      void *argv, u32 argc, struct nvif_vmm *vmm)
+nvif_vmm_init(struct nvif_mmu *mmu, s32 oclass, bool managed, u64 addr,
+	      u64 size, void *argv, u32 argc, struct nvif_vmm *vmm)
 {
 	struct nvif_vmm_v0 *args;
 	u32 argn = sizeof(*args) + argc;
@@ -125,6 +125,7 @@ nvif_vmm_init(struct nvif_mmu *mmu, s32 oclass, u64 addr, u64 size,
 	if (!(args = kmalloc(argn, GFP_KERNEL)))
 		return -ENOMEM;
 	args->version = 0;
+	args->managed = managed;
 	args->addr = addr;
 	args->size = size;
 	memcpy(args->data, argv, argc);
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
index c61b467cf45e..245990de1e90 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
+++ b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
@@ -39,6 +39,7 @@ nvkm_subdev_name[NVKM_SUBDEV_NR] = {
 	[NVKM_SUBDEV_FB      ] = "fb",
 	[NVKM_SUBDEV_FUSE    ] = "fuse",
 	[NVKM_SUBDEV_GPIO    ] = "gpio",
+	[NVKM_SUBDEV_GSP     ] = "gsp",
 	[NVKM_SUBDEV_I2C     ] = "i2c",
 	[NVKM_SUBDEV_IBUS    ] = "priv",
 	[NVKM_SUBDEV_ICCSENSE] = "iccsense",
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
index 177a23301d6a..9211663239af 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
@@ -6,4 +6,4 @@ nvkm-y += nvkm/engine/ce/gm200.o
 nvkm-y += nvkm/engine/ce/gp100.o
 nvkm-y += nvkm/engine/ce/gp102.o
 nvkm-y += nvkm/engine/ce/gv100.o
-nvkm-y += nvkm/engine/ce/tu104.o
+nvkm-y += nvkm/engine/ce/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c
index 3c25043bbb33..b4308e2d8c75 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c
@@ -24,7 +24,7 @@
 #include <nvif/class.h>
 
 static const struct nvkm_engine_func
-tu104_ce = {
+tu102_ce = {
 	.intr = gp100_ce_intr,
 	.sclass = {
 		{ -1, -1, TURING_DMA_COPY_A },
@@ -33,8 +33,8 @@ tu104_ce = {
 };
 
 int
-tu104_ce_new(struct nvkm_device *device, int index,
+tu102_ce_new(struct nvkm_device *device, int index,
 	     struct nvkm_engine **pengine)
 {
-	return nvkm_engine_new_(&tu104_ce, device, index, true, pengine);
+	return nvkm_engine_new_(&tu102_ce, device, index, true, pengine);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index bfbc9341e0c2..7971096b6767 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1613,7 +1613,7 @@ nvd7_chipset = {
 	.pci = gf106_pci_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = gf100_volt_new,
+	.volt = gf117_volt_new,
 	.ce[0] = gf100_ce_new,
 	.disp = gf119_disp_new,
 	.dma = gf119_dma_new,
@@ -2405,6 +2405,7 @@ nv140_chipset = {
 	.fb = gv100_fb_new,
 	.fuse = gm107_fuse_new,
 	.gpio = gk104_gpio_new,
+	.gsp = gv100_gsp_new,
 	.i2c = gm200_i2c_new,
 	.ibus = gm200_ibus_new,
 	.imem = nv50_instmem_new,
@@ -2435,67 +2436,108 @@ nv140_chipset = {
 };
 
 static const struct nvkm_device_chip
+nv162_chipset = {
+	.name = "TU102",
+	.bar = tu102_bar_new,
+	.bios = nvkm_bios_new,
+	.bus = gf100_bus_new,
+	.devinit = tu102_devinit_new,
+	.fault = tu102_fault_new,
+	.fb = gv100_fb_new,
+	.fuse = gm107_fuse_new,
+	.gpio = gk104_gpio_new,
+	.gsp = gv100_gsp_new,
+	.i2c = gm200_i2c_new,
+	.ibus = gm200_ibus_new,
+	.imem = nv50_instmem_new,
+	.ltc = gp102_ltc_new,
+	.mc = tu102_mc_new,
+	.mmu = tu102_mmu_new,
+	.pci = gp100_pci_new,
+	.pmu = gp102_pmu_new,
+	.therm = gp100_therm_new,
+	.timer = gk20a_timer_new,
+	.top = gk104_top_new,
+	.ce[0] = tu102_ce_new,
+	.ce[1] = tu102_ce_new,
+	.ce[2] = tu102_ce_new,
+	.ce[3] = tu102_ce_new,
+	.ce[4] = tu102_ce_new,
+	.disp = tu102_disp_new,
+	.dma = gv100_dma_new,
+	.fifo = tu102_fifo_new,
+	.nvdec[0] = gp102_nvdec_new,
+	.sec2 = tu102_sec2_new,
+};
+
+static const struct nvkm_device_chip
 nv164_chipset = {
 	.name = "TU104",
-	.bar = tu104_bar_new,
+	.bar = tu102_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
-	.devinit = tu104_devinit_new,
-	.fault = tu104_fault_new,
+	.devinit = tu102_devinit_new,
+	.fault = tu102_fault_new,
 	.fb = gv100_fb_new,
 	.fuse = gm107_fuse_new,
 	.gpio = gk104_gpio_new,
+	.gsp = gv100_gsp_new,
 	.i2c = gm200_i2c_new,
 	.ibus = gm200_ibus_new,
 	.imem = nv50_instmem_new,
 	.ltc = gp102_ltc_new,
-	.mc = tu104_mc_new,
-	.mmu = tu104_mmu_new,
+	.mc = tu102_mc_new,
+	.mmu = tu102_mmu_new,
 	.pci = gp100_pci_new,
 	.pmu = gp102_pmu_new,
 	.therm = gp100_therm_new,
 	.timer = gk20a_timer_new,
 	.top = gk104_top_new,
-	.ce[0] = tu104_ce_new,
-	.ce[1] = tu104_ce_new,
-	.ce[2] = tu104_ce_new,
-	.ce[3] = tu104_ce_new,
-	.ce[4] = tu104_ce_new,
-	.disp = tu104_disp_new,
+	.ce[0] = tu102_ce_new,
+	.ce[1] = tu102_ce_new,
+	.ce[2] = tu102_ce_new,
+	.ce[3] = tu102_ce_new,
+	.ce[4] = tu102_ce_new,
+	.disp = tu102_disp_new,
 	.dma = gv100_dma_new,
-	.fifo = tu104_fifo_new,
+	.fifo = tu102_fifo_new,
+	.nvdec[0] = gp102_nvdec_new,
+	.sec2 = tu102_sec2_new,
 };
 
 static const struct nvkm_device_chip
 nv166_chipset = {
 	.name = "TU106",
-	.bar = tu104_bar_new,
+	.bar = tu102_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
-	.devinit = tu104_devinit_new,
-	.fault = tu104_fault_new,
+	.devinit = tu102_devinit_new,
+	.fault = tu102_fault_new,
 	.fb = gv100_fb_new,
 	.fuse = gm107_fuse_new,
 	.gpio = gk104_gpio_new,
+	.gsp = gv100_gsp_new,
 	.i2c = gm200_i2c_new,
 	.ibus = gm200_ibus_new,
 	.imem = nv50_instmem_new,
 	.ltc = gp102_ltc_new,
-	.mc = tu104_mc_new,
-	.mmu = tu104_mmu_new,
+	.mc = tu102_mc_new,
+	.mmu = tu102_mmu_new,
 	.pci = gp100_pci_new,
 	.pmu = gp102_pmu_new,
 	.therm = gp100_therm_new,
 	.timer = gk20a_timer_new,
 	.top = gk104_top_new,
-	.ce[0] = tu104_ce_new,
-	.ce[1] = tu104_ce_new,
-	.ce[2] = tu104_ce_new,
-	.ce[3] = tu104_ce_new,
-	.ce[4] = tu104_ce_new,
-	.disp = tu104_disp_new,
+	.ce[0] = tu102_ce_new,
+	.ce[1] = tu102_ce_new,
+	.ce[2] = tu102_ce_new,
+	.ce[3] = tu102_ce_new,
+	.ce[4] = tu102_ce_new,
+	.disp = tu102_disp_new,
 	.dma = gv100_dma_new,
-	.fifo = tu104_fifo_new,
+	.fifo = tu102_fifo_new,
+	.nvdec[0] = gp102_nvdec_new,
+	.sec2 = tu102_sec2_new,
 };
 
 static int
@@ -2535,6 +2577,7 @@ nvkm_device_subdev(struct nvkm_device *device, int index)
 	_(FB      , device->fb      , &device->fb->subdev);
 	_(FUSE    , device->fuse    , &device->fuse->subdev);
 	_(GPIO    , device->gpio    , &device->gpio->subdev);
+	_(GSP     , device->gsp     , &device->gsp->subdev);
 	_(I2C     , device->i2c     , &device->i2c->subdev);
 	_(IBUS    , device->ibus    ,  device->ibus);
 	_(ICCSENSE, device->iccsense, &device->iccsense->subdev);
@@ -2950,6 +2993,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 		case 0x138: device->chip = &nv138_chipset; break;
 		case 0x13b: device->chip = &nv13b_chipset; break;
 		case 0x140: device->chip = &nv140_chipset; break;
+		case 0x162: device->chip = &nv162_chipset; break;
 		case 0x164: device->chip = &nv164_chipset; break;
 		case 0x166: device->chip = &nv166_chipset; break;
 		default:
@@ -3017,6 +3061,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 		_(NVKM_SUBDEV_FB      ,       fb);
 		_(NVKM_SUBDEV_FUSE    ,     fuse);
 		_(NVKM_SUBDEV_GPIO    ,     gpio);
+		_(NVKM_SUBDEV_GSP     ,      gsp);
 		_(NVKM_SUBDEV_I2C     ,      i2c);
 		_(NVKM_SUBDEV_IBUS    ,     ibus);
 		_(NVKM_SUBDEV_ICCSENSE, iccsense);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
index 253ab914a8ef..2a53e37dfa7a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
@@ -12,6 +12,7 @@
 #include <subdev/fb.h>
 #include <subdev/fuse.h>
 #include <subdev/gpio.h>
+#include <subdev/gsp.h>
 #include <subdev/i2c.h>
 #include <subdev/ibus.h>
 #include <subdev/iccsense.h>
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
index 092ddc4ffefa..03c6d9aef075 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
@@ -365,16 +365,15 @@ nvkm_udevice_child_get(struct nvkm_object *object, int index,
 	}
 
 	if (!sclass) {
-		switch (index) {
-		case 0: sclass = &nvkm_control_oclass; break;
-		case 1:
-			if (!device->mmu)
-				return -EINVAL;
+		if (index-- == 0)
+			sclass = &nvkm_control_oclass;
+		else if (device->mmu && index-- == 0)
 			sclass = &device->mmu->user;
-			break;
-		default:
+		else if (device->fault && index-- == 0)
+			sclass = &device->fault->user;
+		else
 			return -EINVAL;
-		}
+
 		oclass->base = sclass->base;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
index c6a257ba4347..2c28a5e747cc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
@@ -15,7 +15,7 @@ nvkm-y += nvkm/engine/disp/gm200.o
 nvkm-y += nvkm/engine/disp/gp100.o
 nvkm-y += nvkm/engine/disp/gp102.o
 nvkm-y += nvkm/engine/disp/gv100.o
-nvkm-y += nvkm/engine/disp/tu104.o
+nvkm-y += nvkm/engine/disp/tu102.o
 nvkm-y += nvkm/engine/disp/vga.o
 
 nvkm-y += nvkm/engine/disp/head.o
@@ -39,7 +39,7 @@ nvkm-y += nvkm/engine/disp/sorgk104.o
 nvkm-y += nvkm/engine/disp/sorgm107.o
 nvkm-y += nvkm/engine/disp/sorgm200.o
 nvkm-y += nvkm/engine/disp/sorgv100.o
-nvkm-y += nvkm/engine/disp/sortu104.o
+nvkm-y += nvkm/engine/disp/sortu102.o
 
 nvkm-y += nvkm/engine/disp/outp.o
 nvkm-y += nvkm/engine/disp/dp.o
@@ -71,7 +71,7 @@ nvkm-y += nvkm/engine/disp/rootgm200.o
 nvkm-y += nvkm/engine/disp/rootgp100.o
 nvkm-y += nvkm/engine/disp/rootgp102.o
 nvkm-y += nvkm/engine/disp/rootgv100.o
-nvkm-y += nvkm/engine/disp/roottu104.o
+nvkm-y += nvkm/engine/disp/roottu102.o
 
 nvkm-y += nvkm/engine/disp/channv50.o
 nvkm-y += nvkm/engine/disp/changf119.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c
index 794e90982641..e675d9b9d5d7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf119.c
@@ -91,15 +91,21 @@ gf119_disp_intr_error(struct nv50_disp *disp, int chid)
 {
 	struct nvkm_subdev *subdev = &disp->base.engine.subdev;
 	struct nvkm_device *device = subdev->device;
-	u32 mthd = nvkm_rd32(device, 0x6101f0 + (chid * 12));
+	u32 stat = nvkm_rd32(device, 0x6101f0 + (chid * 12));
+	u32 type = (stat & 0x00007000) >> 12;
+	u32 mthd = (stat & 0x00000ffc);
 	u32 data = nvkm_rd32(device, 0x6101f4 + (chid * 12));
-	u32 unkn = nvkm_rd32(device, 0x6101f8 + (chid * 12));
+	u32 code = nvkm_rd32(device, 0x6101f8 + (chid * 12));
+	const struct nvkm_enum *reason =
+		nvkm_enum_find(nv50_disp_intr_error_type, type);
 
-	nvkm_error(subdev, "chid %d mthd %04x data %08x %08x %08x\n",
-		   chid, (mthd & 0x0000ffc), data, mthd, unkn);
+	nvkm_error(subdev, "chid %d stat %08x reason %d [%s] mthd %04x "
+			   "data %08x code %08x\n",
+		   chid, stat, type, reason ? reason->name : "",
+		   mthd, data, code);
 
 	if (chid < ARRAY_SIZE(disp->chan)) {
-		switch (mthd & 0xffc) {
+		switch (mthd) {
 		case 0x0080:
 			nv50_disp_chan_mthd(disp->chan[chid], NV_DBG_ERROR);
 			break;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
index 47be0ba4aebe..892be6c9b76c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
@@ -103,10 +103,13 @@ gv100_disp_exception(struct nv50_disp *disp, int chid)
 	u32 mthd = (stat & 0x00000fff) << 2;
 	u32 data = nvkm_rd32(device, 0x611024 + (chid * 12));
 	u32 code = nvkm_rd32(device, 0x611028 + (chid * 12));
+	const struct nvkm_enum *reason =
+		nvkm_enum_find(nv50_disp_intr_error_type, type);
 
-	nvkm_error(subdev, "chid %d %08x [type %d mthd %04x] "
+	nvkm_error(subdev, "chid %d stat %08x reason %d [%s] mthd %04x "
 			   "data %08x code %08x\n",
-		   chid, stat, type, mthd, data, code);
+		   chid, stat, type, reason ? reason->name : "",
+		   mthd, data, code);
 
 	if (chid < ARRAY_SIZE(disp->chan) && disp->chan[chid]) {
 		switch (mthd) {
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
index 790e42f460fd..1681ddccd298 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h
@@ -201,5 +201,5 @@ int gm200_sor_new(struct nvkm_disp *, int);
 int gv100_sor_cnt(struct nvkm_disp *, unsigned long *);
 int gv100_sor_new(struct nvkm_disp *, int);
 
-int tu104_sor_new(struct nvkm_disp *, int);
+int tu102_sor_new(struct nvkm_disp *, int);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
index def005dd5fda..e21556bf2cb1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
@@ -28,7 +28,6 @@
 #include "rootnv50.h"
 
 #include <core/client.h>
-#include <core/enum.h>
 #include <core/ramht.h>
 #include <subdev/bios.h>
 #include <subdev/bios/disp.h>
@@ -593,12 +592,15 @@ nv50_disp_super(struct work_struct *work)
 	nvkm_wr32(device, 0x610030, 0x80000000);
 }
 
-static const struct nvkm_enum
+const struct nvkm_enum
 nv50_disp_intr_error_type[] = {
-	{ 3, "ILLEGAL_MTHD" },
-	{ 4, "INVALID_VALUE" },
+	{ 0, "NONE" },
+	{ 1, "PUSHBUFFER_ERR" },
+	{ 2, "TRAP" },
+	{ 3, "RESERVED_METHOD" },
+	{ 4, "INVALID_ARG" },
 	{ 5, "INVALID_STATE" },
-	{ 7, "INVALID_HANDLE" },
+	{ 7, "UNRESOLVABLE_HANDLE" },
 	{}
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
index c36a8a7cafa1..e5d00f478bb1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h
@@ -5,6 +5,8 @@
 #include "priv.h"
 struct nvkm_head;
 
+#include <core/enum.h>
+
 struct nv50_disp {
 	const struct nv50_disp_func *func;
 	struct nvkm_disp base;
@@ -71,6 +73,7 @@ int nv50_disp_init(struct nv50_disp *);
 void nv50_disp_fini(struct nv50_disp *);
 void nv50_disp_intr(struct nv50_disp *);
 void nv50_disp_super(struct work_struct *);
+extern const struct nvkm_enum nv50_disp_intr_error_type[];
 
 int gf119_disp_init(struct nv50_disp *);
 void gf119_disp_fini(struct nv50_disp *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
index 97de928cbde1..aee9822a7a87 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
@@ -37,5 +37,5 @@ extern const struct nvkm_disp_oclass gm200_disp_root_oclass;
 extern const struct nvkm_disp_oclass gp100_disp_root_oclass;
 extern const struct nvkm_disp_oclass gp102_disp_root_oclass;
 extern const struct nvkm_disp_oclass gv100_disp_root_oclass;
-extern const struct nvkm_disp_oclass tu104_disp_root_oclass;
+extern const struct nvkm_disp_oclass tu102_disp_root_oclass;
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu102.c
index ad438c62f66c..579a5d02308a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu102.c
@@ -25,28 +25,28 @@
 #include <nvif/class.h>
 
 static const struct nv50_disp_root_func
-tu104_disp_root = {
+tu102_disp_root = {
 	.user = {
-		{{0,0,TU104_DISP_CURSOR                }, gv100_disp_curs_new },
-		{{0,0,TU104_DISP_WINDOW_IMM_CHANNEL_DMA}, gv100_disp_wimm_new },
-		{{0,0,TU104_DISP_CORE_CHANNEL_DMA      }, gv100_disp_core_new },
-		{{0,0,TU104_DISP_WINDOW_CHANNEL_DMA    }, gv100_disp_wndw_new },
+		{{0,0,TU102_DISP_CURSOR                }, gv100_disp_curs_new },
+		{{0,0,TU102_DISP_WINDOW_IMM_CHANNEL_DMA}, gv100_disp_wimm_new },
+		{{0,0,TU102_DISP_CORE_CHANNEL_DMA      }, gv100_disp_core_new },
+		{{0,0,TU102_DISP_WINDOW_CHANNEL_DMA    }, gv100_disp_wndw_new },
 		{}
 	},
 };
 
 static int
-tu104_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass,
+tu102_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass,
 		    void *data, u32 size, struct nvkm_object **pobject)
 {
-	return nv50_disp_root_new_(&tu104_disp_root, disp, oclass,
+	return nv50_disp_root_new_(&tu102_disp_root, disp, oclass,
 				   data, size, pobject);
 }
 
 const struct nvkm_disp_oclass
-tu104_disp_root_oclass = {
-	.base.oclass = TU104_DISP,
+tu102_disp_root_oclass = {
+	.base.oclass = TU102_DISP,
 	.base.minver = -1,
 	.base.maxver = -1,
-	.ctor = tu104_disp_root_new,
+	.ctor = tu102_disp_root_new,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu102.c
index df026a525ef1..d57b73ada89e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu102.c
@@ -24,7 +24,7 @@
 #include <subdev/timer.h>
 
 static void
-tu104_sor_dp_vcpi(struct nvkm_ior *sor, int head,
+tu102_sor_dp_vcpi(struct nvkm_ior *sor, int head,
 		  u8 slot, u8 slot_nr, u16 pbn, u16 aligned)
 {
 	struct nvkm_device *device = sor->disp->engine.subdev.device;
@@ -35,7 +35,7 @@ tu104_sor_dp_vcpi(struct nvkm_ior *sor, int head,
 }
 
 static int
-tu104_sor_dp_links(struct nvkm_ior *sor, struct nvkm_i2c_aux *aux)
+tu102_sor_dp_links(struct nvkm_ior *sor, struct nvkm_i2c_aux *aux)
 {
 	struct nvkm_device *device = sor->disp->engine.subdev.device;
 	const u32 soff = nv50_ior_base(sor);
@@ -62,7 +62,7 @@ tu104_sor_dp_links(struct nvkm_ior *sor, struct nvkm_i2c_aux *aux)
 }
 
 static const struct nvkm_ior_func
-tu104_sor = {
+tu102_sor = {
 	.route = {
 		.get = gm200_sor_route_get,
 		.set = gm200_sor_route_set,
@@ -75,11 +75,11 @@ tu104_sor = {
 	},
 	.dp = {
 		.lanes = { 0, 1, 2, 3 },
-		.links = tu104_sor_dp_links,
+		.links = tu102_sor_dp_links,
 		.power = g94_sor_dp_power,
 		.pattern = gm107_sor_dp_pattern,
 		.drive = gm200_sor_dp_drive,
-		.vcpi = tu104_sor_dp_vcpi,
+		.vcpi = tu102_sor_dp_vcpi,
 		.audio = gv100_sor_dp_audio,
 		.audio_sym = gv100_sor_dp_audio_sym,
 		.watermark = gv100_sor_dp_watermark,
@@ -91,7 +91,7 @@ tu104_sor = {
 };
 
 int
-tu104_sor_new(struct nvkm_disp *disp, int id)
+tu102_sor_new(struct nvkm_disp *disp, int id)
 {
-	return nvkm_ior_new_(&tu104_sor, disp, SOR, id);
+	return nvkm_ior_new_(&tu102_sor, disp, SOR, id);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu102.c
index 13fa21459d38..883ae4151ff8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu102.c
@@ -29,7 +29,7 @@
 #include <subdev/timer.h>
 
 static int
-tu104_disp_init(struct nv50_disp *disp)
+tu102_disp_init(struct nv50_disp *disp)
 {
 	struct nvkm_device *device = disp->base.engine.subdev.device;
 	struct nvkm_head *head;
@@ -132,21 +132,21 @@ tu104_disp_init(struct nv50_disp *disp)
 }
 
 static const struct nv50_disp_func
-tu104_disp = {
-	.init = tu104_disp_init,
+tu102_disp = {
+	.init = tu102_disp_init,
 	.fini = gv100_disp_fini,
 	.intr = gv100_disp_intr,
 	.uevent = &gv100_disp_chan_uevent,
 	.super = gv100_disp_super,
-	.root = &tu104_disp_root_oclass,
+	.root = &tu102_disp_root_oclass,
 	.wndw = { .cnt = gv100_disp_wndw_cnt },
 	.head = { .cnt = gv100_head_cnt, .new = gv100_head_new },
-	.sor = { .cnt = gv100_sor_cnt, .new = tu104_sor_new },
+	.sor = { .cnt = gv100_sor_cnt, .new = tu102_sor_new },
 	.ramht_size = 0x2000,
 };
 
 int
-tu104_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp)
+tu102_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp)
 {
-	return nv50_disp_new_(&tu104_disp, device, index, pdisp);
+	return nv50_disp_new_(&tu102_disp, device, index, pdisp);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/falcon.c b/drivers/gpu/drm/nouveau/nvkm/engine/falcon.c
index 816ccaedfc73..8675613e142b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/falcon.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/falcon.c
@@ -22,6 +22,7 @@
 #include <engine/falcon.h>
 
 #include <core/gpuobj.h>
+#include <subdev/mc.h>
 #include <subdev/timer.h>
 #include <engine/fifo.h>
 
@@ -107,8 +108,10 @@ nvkm_falcon_fini(struct nvkm_engine *engine, bool suspend)
 		}
 	}
 
-	nvkm_mask(device, base + 0x048, 0x00000003, 0x00000000);
-	nvkm_wr32(device, base + 0x014, 0xffffffff);
+	if (nvkm_mc_enabled(device, engine->subdev.index)) {
+		nvkm_mask(device, base + 0x048, 0x00000003, 0x00000000);
+		nvkm_wr32(device, base + 0x014, 0xffffffff);
+	}
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
index 87d8e054e40a..05aada541ea5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
@@ -16,7 +16,7 @@ nvkm-y += nvkm/engine/fifo/gm20b.o
 nvkm-y += nvkm/engine/fifo/gp100.o
 nvkm-y += nvkm/engine/fifo/gp10b.o
 nvkm-y += nvkm/engine/fifo/gv100.o
-nvkm-y += nvkm/engine/fifo/tu104.o
+nvkm-y += nvkm/engine/fifo/tu102.o
 
 nvkm-y += nvkm/engine/fifo/chan.o
 nvkm-y += nvkm/engine/fifo/channv50.o
@@ -34,7 +34,7 @@ nvkm-y += nvkm/engine/fifo/gpfifog84.o
 nvkm-y += nvkm/engine/fifo/gpfifogf100.o
 nvkm-y += nvkm/engine/fifo/gpfifogk104.o
 nvkm-y += nvkm/engine/fifo/gpfifogv100.o
-nvkm-y += nvkm/engine/fifo/gpfifotu104.o
+nvkm-y += nvkm/engine/fifo/gpfifotu102.o
 
 nvkm-y += nvkm/engine/fifo/usergv100.o
-nvkm-y += nvkm/engine/fifo/usertu104.o
+nvkm-y += nvkm/engine/fifo/usertu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
index a14545d871d8..f8557cdfbd81 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
@@ -47,6 +47,6 @@ int gv100_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *,
 int gv100_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *,
 				  struct nvkm_engine *, bool);
 
-int tu104_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *,
+int tu102_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *,
 			  void *data, u32 size, struct nvkm_object **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu102.c
index ff70484dd01a..abef7fb6e2d3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu102.c
@@ -29,14 +29,14 @@
 #include <nvif/unpack.h>
 
 static u32
-tu104_fifo_gpfifo_submit_token(struct nvkm_fifo_chan *base)
+tu102_fifo_gpfifo_submit_token(struct nvkm_fifo_chan *base)
 {
 	struct gk104_fifo_chan *chan = gk104_fifo_chan(base);
 	return (chan->runl << 16) | chan->base.chid;
 }
 
 static const struct nvkm_fifo_chan_func
-tu104_fifo_gpfifo = {
+tu102_fifo_gpfifo = {
 	.dtor = gk104_fifo_gpfifo_dtor,
 	.init = gk104_fifo_gpfifo_init,
 	.fini = gk104_fifo_gpfifo_fini,
@@ -45,11 +45,11 @@ tu104_fifo_gpfifo = {
 	.engine_dtor = gk104_fifo_gpfifo_engine_dtor,
 	.engine_init = gv100_fifo_gpfifo_engine_init,
 	.engine_fini = gv100_fifo_gpfifo_engine_fini,
-	.submit_token = tu104_fifo_gpfifo_submit_token,
+	.submit_token = tu102_fifo_gpfifo_submit_token,
 };
 
 int
-tu104_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass,
+tu102_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass,
 		      void *data, u32 size, struct nvkm_object **pobject)
 {
 	struct nvkm_object *parent = oclass->parent;
@@ -67,7 +67,7 @@ tu104_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass,
 			   args->v0.ilength, args->v0.runlist, args->v0.priv);
 		if (args->v0.priv && !oclass->client->super)
 			return -EINVAL;
-		return gv100_fifo_gpfifo_new_(&tu104_fifo_gpfifo, fifo,
+		return gv100_fifo_gpfifo_new_(&tu102_fifo_gpfifo, fifo,
 					      &args->v0.runlist,
 					      &args->v0.chid,
 					       args->v0.vmm,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c
index 98c80705bc61..005f3e1729b9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c
@@ -29,7 +29,7 @@
 #include <nvif/class.h>
 
 static void
-tu104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl,
+tu102_fifo_runlist_commit(struct gk104_fifo *fifo, int runl,
 			  struct nvkm_memory *mem, int nr)
 {
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
@@ -44,15 +44,15 @@ tu104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl,
 }
 
 const struct gk104_fifo_runlist_func
-tu104_fifo_runlist = {
+tu102_fifo_runlist = {
 	.size = 16,
 	.cgrp = gv100_fifo_runlist_cgrp,
 	.chan = gv100_fifo_runlist_chan,
-	.commit = tu104_fifo_runlist_commit,
+	.commit = tu102_fifo_runlist_commit,
 };
 
 static const struct nvkm_enum
-tu104_fifo_fault_engine[] = {
+tu102_fifo_fault_engine[] = {
 	{ 0x01, "DISPLAY" },
 	{ 0x03, "PTP" },
 	{ 0x06, "PWR_PMU" },
@@ -80,7 +80,7 @@ tu104_fifo_fault_engine[] = {
 };
 
 static void
-tu104_fifo_pbdma_init(struct gk104_fifo *fifo)
+tu102_fifo_pbdma_init(struct gk104_fifo *fifo)
 {
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	const u32 mask = (1 << fifo->pbdma_nr) - 1;
@@ -89,28 +89,28 @@ tu104_fifo_pbdma_init(struct gk104_fifo *fifo)
 }
 
 static const struct gk104_fifo_pbdma_func
-tu104_fifo_pbdma = {
+tu102_fifo_pbdma = {
 	.nr = gm200_fifo_pbdma_nr,
-	.init = tu104_fifo_pbdma_init,
+	.init = tu102_fifo_pbdma_init,
 	.init_timeout = gk208_fifo_pbdma_init_timeout,
 };
 
 static const struct gk104_fifo_func
-tu104_fifo = {
-	.pbdma = &tu104_fifo_pbdma,
+tu102_fifo = {
+	.pbdma = &tu102_fifo_pbdma,
 	.fault.access = gv100_fifo_fault_access,
-	.fault.engine = tu104_fifo_fault_engine,
+	.fault.engine = tu102_fifo_fault_engine,
 	.fault.reason = gv100_fifo_fault_reason,
 	.fault.hubclient = gv100_fifo_fault_hubclient,
 	.fault.gpcclient = gv100_fifo_fault_gpcclient,
-	.runlist = &tu104_fifo_runlist,
-	.user = {{-1,-1,VOLTA_USERMODE_A       }, tu104_fifo_user_new   },
-	.chan = {{ 0, 0,TURING_CHANNEL_GPFIFO_A}, tu104_fifo_gpfifo_new },
+	.runlist = &tu102_fifo_runlist,
+	.user = {{-1,-1,VOLTA_USERMODE_A       }, tu102_fifo_user_new   },
+	.chan = {{ 0, 0,TURING_CHANNEL_GPFIFO_A}, tu102_fifo_gpfifo_new },
 	.cgrp_force = true,
 };
 
 int
-tu104_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
+tu102_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
 {
-	return gk104_fifo_new_(&tu104_fifo, device, index, 4096, pfifo);
+	return gk104_fifo_new_(&tu102_fifo, device, index, 4096, pfifo);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h
index 14b0c6bde8eb..54a3a3092cc0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h
@@ -3,6 +3,6 @@
 #include "priv.h"
 int gv100_fifo_user_new(const struct nvkm_oclass *, void *, u32,
 			struct nvkm_object **);
-int tu104_fifo_user_new(const struct nvkm_oclass *, void *, u32,
+int tu102_fifo_user_new(const struct nvkm_oclass *, void *, u32,
 			struct nvkm_object **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu102.c
index 8f98548a21f6..217268f8ccad 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu102.c
@@ -22,7 +22,7 @@
 #include "user.h"
 
 static int
-tu104_fifo_user_map(struct nvkm_object *object, void *argv, u32 argc,
+tu102_fifo_user_map(struct nvkm_object *object, void *argv, u32 argc,
 		    enum nvkm_object_map *type, u64 *addr, u64 *size)
 {
 	struct nvkm_device *device = object->engine->subdev.device;
@@ -33,13 +33,13 @@ tu104_fifo_user_map(struct nvkm_object *object, void *argv, u32 argc,
 }
 
 static const struct nvkm_object_func
-tu104_fifo_user = {
-	.map = tu104_fifo_user_map,
+tu102_fifo_user = {
+	.map = tu102_fifo_user_map,
 };
 
 int
-tu104_fifo_user_new(const struct nvkm_oclass *oclass, void *argv, u32 argc,
+tu102_fifo_user_new(const struct nvkm_oclass *oclass, void *argv, u32 argc,
 		    struct nvkm_object **pobject)
 {
-	return nvkm_object_new_(&tu104_fifo_user, oclass, argv, argc, pobject);
+	return nvkm_object_new_(&tu102_fifo_user, oclass, argv, argc, pobject);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/base.c
index cd8cf6f7024c..d41fb94524e9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/base.c
@@ -25,6 +25,33 @@
 
 #include <engine/fifo.h>
 
+u32
+nvkm_gr_ctxsw_inst(struct nvkm_device *device)
+{
+	struct nvkm_gr *gr = device->gr;
+	if (gr && gr->func->ctxsw.inst)
+		return gr->func->ctxsw.inst(gr);
+	return 0;
+}
+
+int
+nvkm_gr_ctxsw_resume(struct nvkm_device *device)
+{
+	struct nvkm_gr *gr = device->gr;
+	if (gr && gr->func->ctxsw.resume)
+		return gr->func->ctxsw.resume(gr);
+	return 0;
+}
+
+int
+nvkm_gr_ctxsw_pause(struct nvkm_device *device)
+{
+	struct nvkm_gr *gr = device->gr;
+	if (gr && gr->func->ctxsw.pause)
+		return gr->func->ctxsw.pause(gr);
+	return 0;
+}
+
 static bool
 nvkm_gr_chsw_load(struct nvkm_engine *engine)
 {
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
index e813a3f8ea93..85f2d1e950e8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
@@ -1523,13 +1523,9 @@ gf100_grctx_generate(struct gf100_gr *gr)
 	/* Make channel current. */
 	addr = nvkm_memory_addr(inst) >> 12;
 	if (gr->firmware) {
-		nvkm_wr32(device, 0x409840, 0x00000030);
-		nvkm_wr32(device, 0x409500, 0x80000000 | addr);
-		nvkm_wr32(device, 0x409504, 0x00000003);
-		nvkm_msec(device, 2000,
-			if (nvkm_rd32(device, 0x409800) & 0x00000010)
-				break;
-		);
+		ret = gf100_gr_fecs_bind_pointer(gr, 0x80000000 | addr);
+		if (ret)
+			goto done;
 
 		nvkm_kmap(data);
 		nvkm_wo32(data, 0x1c, 1);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
index 70d3d41e616c..81a13cf9a292 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
@@ -715,6 +715,211 @@ gf100_gr_pack_mmio[] = {
  * PGRAPH engine/subdev functions
  ******************************************************************************/
 
+static u32
+gf100_gr_ctxsw_inst(struct nvkm_gr *gr)
+{
+	return nvkm_rd32(gr->engine.subdev.device, 0x409b00);
+}
+
+static int
+gf100_gr_fecs_ctrl_ctxsw(struct gf100_gr *gr, u32 mthd)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409804, 0xffffffff);
+	nvkm_wr32(device, 0x409840, 0xffffffff);
+	nvkm_wr32(device, 0x409500, 0xffffffff);
+	nvkm_wr32(device, 0x409504, mthd);
+	nvkm_msec(device, 2000,
+		u32 stat = nvkm_rd32(device, 0x409804);
+		if (stat == 0x00000002)
+			return -EIO;
+		if (stat == 0x00000001)
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+int
+gf100_gr_fecs_start_ctxsw(struct nvkm_gr *base)
+{
+	struct gf100_gr *gr = gf100_gr(base);
+	int ret = 0;
+
+	mutex_lock(&gr->fecs.mutex);
+	if (!--gr->fecs.disable) {
+		if (WARN_ON(ret = gf100_gr_fecs_ctrl_ctxsw(gr, 0x39)))
+			gr->fecs.disable++;
+	}
+	mutex_unlock(&gr->fecs.mutex);
+	return ret;
+}
+
+int
+gf100_gr_fecs_stop_ctxsw(struct nvkm_gr *base)
+{
+	struct gf100_gr *gr = gf100_gr(base);
+	int ret = 0;
+
+	mutex_lock(&gr->fecs.mutex);
+	if (!gr->fecs.disable++) {
+		if (WARN_ON(ret = gf100_gr_fecs_ctrl_ctxsw(gr, 0x38)))
+			gr->fecs.disable--;
+	}
+	mutex_unlock(&gr->fecs.mutex);
+	return ret;
+}
+
+int
+gf100_gr_fecs_bind_pointer(struct gf100_gr *gr, u32 inst)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409840, 0x00000030);
+	nvkm_wr32(device, 0x409500, inst);
+	nvkm_wr32(device, 0x409504, 0x00000003);
+	nvkm_msec(device, 2000,
+		u32 stat = nvkm_rd32(device, 0x409800);
+		if (stat & 0x00000020)
+			return -EIO;
+		if (stat & 0x00000010)
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_set_reglist_virtual_address(struct gf100_gr *gr, u64 addr)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409810, addr >> 8);
+	nvkm_wr32(device, 0x409800, 0x00000000);
+	nvkm_wr32(device, 0x409500, 0x00000001);
+	nvkm_wr32(device, 0x409504, 0x00000032);
+	nvkm_msec(device, 2000,
+		if (nvkm_rd32(device, 0x409800) == 0x00000001)
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_set_reglist_bind_instance(struct gf100_gr *gr, u32 inst)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409810, inst);
+	nvkm_wr32(device, 0x409800, 0x00000000);
+	nvkm_wr32(device, 0x409500, 0x00000001);
+	nvkm_wr32(device, 0x409504, 0x00000031);
+	nvkm_msec(device, 2000,
+		if (nvkm_rd32(device, 0x409800) == 0x00000001)
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_discover_reglist_image_size(struct gf100_gr *gr, u32 *psize)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409800, 0x00000000);
+	nvkm_wr32(device, 0x409500, 0x00000001);
+	nvkm_wr32(device, 0x409504, 0x00000030);
+	nvkm_msec(device, 2000,
+		if ((*psize = nvkm_rd32(device, 0x409800)))
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_elpg_bind(struct gf100_gr *gr)
+{
+	u32 size;
+	int ret;
+
+	ret = gf100_gr_fecs_discover_reglist_image_size(gr, &size);
+	if (ret)
+		return ret;
+
+	/*XXX: We need to allocate + map the above into PMU's inst block,
+	 *     which which means we probably need a proper PMU before we
+	 *     even bother.
+	 */
+
+	ret = gf100_gr_fecs_set_reglist_bind_instance(gr, 0);
+	if (ret)
+		return ret;
+
+	return gf100_gr_fecs_set_reglist_virtual_address(gr, 0);
+}
+
+static int
+gf100_gr_fecs_discover_pm_image_size(struct gf100_gr *gr, u32 *psize)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409840, 0xffffffff);
+	nvkm_wr32(device, 0x409500, 0x00000000);
+	nvkm_wr32(device, 0x409504, 0x00000025);
+	nvkm_msec(device, 2000,
+		if ((*psize = nvkm_rd32(device, 0x409800)))
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_discover_zcull_image_size(struct gf100_gr *gr, u32 *psize)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409840, 0xffffffff);
+	nvkm_wr32(device, 0x409500, 0x00000000);
+	nvkm_wr32(device, 0x409504, 0x00000016);
+	nvkm_msec(device, 2000,
+		if ((*psize = nvkm_rd32(device, 0x409800)))
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gf100_gr_fecs_discover_image_size(struct gf100_gr *gr, u32 *psize)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409840, 0xffffffff);
+	nvkm_wr32(device, 0x409500, 0x00000000);
+	nvkm_wr32(device, 0x409504, 0x00000010);
+	nvkm_msec(device, 2000,
+		if ((*psize = nvkm_rd32(device, 0x409800)))
+			return 0;
+	);
+
+	return -ETIMEDOUT;
+}
+
+static void
+gf100_gr_fecs_set_watchdog_timeout(struct gf100_gr *gr, u32 timeout)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+
+	nvkm_wr32(device, 0x409840, 0xffffffff);
+	nvkm_wr32(device, 0x409500, timeout);
+	nvkm_wr32(device, 0x409504, 0x00000021);
+}
+
 static bool
 gf100_gr_chsw_load(struct nvkm_gr *base)
 {
@@ -1487,6 +1692,7 @@ gf100_gr_init_ctxctl_ext(struct gf100_gr *gr)
 	struct nvkm_device *device = subdev->device;
 	struct nvkm_secboot *sb = device->secboot;
 	u32 secboot_mask = 0;
+	int ret;
 
 	/* load fuc microcode */
 	nvkm_mc_unk260(device, 0);
@@ -1495,12 +1701,12 @@ gf100_gr_init_ctxctl_ext(struct gf100_gr *gr)
 	if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_FECS))
 		secboot_mask |= BIT(NVKM_SECBOOT_FALCON_FECS);
 	else
-		gf100_gr_init_fw(gr->fecs, &gr->fuc409c, &gr->fuc409d);
+		gf100_gr_init_fw(gr->fecs.falcon, &gr->fuc409c, &gr->fuc409d);
 
 	if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_GPCCS))
 		secboot_mask |= BIT(NVKM_SECBOOT_FALCON_GPCCS);
 	else
-		gf100_gr_init_fw(gr->gpccs, &gr->fuc41ac, &gr->fuc41ad);
+		gf100_gr_init_fw(gr->gpccs.falcon, &gr->fuc41ac, &gr->fuc41ad);
 
 	if (secboot_mask != 0) {
 		int ret = nvkm_secboot_reset(sb, secboot_mask);
@@ -1515,8 +1721,8 @@ gf100_gr_init_ctxctl_ext(struct gf100_gr *gr)
 	nvkm_wr32(device, 0x41a10c, 0x00000000);
 	nvkm_wr32(device, 0x40910c, 0x00000000);
 
-	nvkm_falcon_start(gr->gpccs);
-	nvkm_falcon_start(gr->fecs);
+	nvkm_falcon_start(gr->gpccs.falcon);
+	nvkm_falcon_start(gr->fecs.falcon);
 
 	if (nvkm_msec(device, 2000,
 		if (nvkm_rd32(device, 0x409800) & 0x00000001)
@@ -1524,72 +1730,36 @@ gf100_gr_init_ctxctl_ext(struct gf100_gr *gr)
 	) < 0)
 		return -EBUSY;
 
-	nvkm_wr32(device, 0x409840, 0xffffffff);
-	nvkm_wr32(device, 0x409500, 0x7fffffff);
-	nvkm_wr32(device, 0x409504, 0x00000021);
+	gf100_gr_fecs_set_watchdog_timeout(gr, 0x7fffffff);
 
-	nvkm_wr32(device, 0x409840, 0xffffffff);
-	nvkm_wr32(device, 0x409500, 0x00000000);
-	nvkm_wr32(device, 0x409504, 0x00000010);
-	if (nvkm_msec(device, 2000,
-		if ((gr->size = nvkm_rd32(device, 0x409800)))
-			break;
-	) < 0)
-		return -EBUSY;
-
-	nvkm_wr32(device, 0x409840, 0xffffffff);
-	nvkm_wr32(device, 0x409500, 0x00000000);
-	nvkm_wr32(device, 0x409504, 0x00000016);
-	if (nvkm_msec(device, 2000,
-		if (nvkm_rd32(device, 0x409800))
-			break;
-	) < 0)
-		return -EBUSY;
+	/* Determine how much memory is required to store main context image. */
+	ret = gf100_gr_fecs_discover_image_size(gr, &gr->size);
+	if (ret)
+		return ret;
 
-	nvkm_wr32(device, 0x409840, 0xffffffff);
-	nvkm_wr32(device, 0x409500, 0x00000000);
-	nvkm_wr32(device, 0x409504, 0x00000025);
-	if (nvkm_msec(device, 2000,
-		if (nvkm_rd32(device, 0x409800))
-			break;
-	) < 0)
-		return -EBUSY;
+	/* Determine how much memory is required to store ZCULL image. */
+	ret = gf100_gr_fecs_discover_zcull_image_size(gr, &gr->size_zcull);
+	if (ret)
+		return ret;
 
-	if (device->chipset >= 0xe0) {
-		nvkm_wr32(device, 0x409800, 0x00000000);
-		nvkm_wr32(device, 0x409500, 0x00000001);
-		nvkm_wr32(device, 0x409504, 0x00000030);
-		if (nvkm_msec(device, 2000,
-			if (nvkm_rd32(device, 0x409800))
-				break;
-		) < 0)
-			return -EBUSY;
-
-		nvkm_wr32(device, 0x409810, 0xb00095c8);
-		nvkm_wr32(device, 0x409800, 0x00000000);
-		nvkm_wr32(device, 0x409500, 0x00000001);
-		nvkm_wr32(device, 0x409504, 0x00000031);
-		if (nvkm_msec(device, 2000,
-			if (nvkm_rd32(device, 0x409800))
-				break;
-		) < 0)
-			return -EBUSY;
-
-		nvkm_wr32(device, 0x409810, 0x00080420);
-		nvkm_wr32(device, 0x409800, 0x00000000);
-		nvkm_wr32(device, 0x409500, 0x00000001);
-		nvkm_wr32(device, 0x409504, 0x00000032);
-		if (nvkm_msec(device, 2000,
-			if (nvkm_rd32(device, 0x409800))
-				break;
-		) < 0)
-			return -EBUSY;
+	/* Determine how much memory is required to store PerfMon image. */
+	ret = gf100_gr_fecs_discover_pm_image_size(gr, &gr->size_pm);
+	if (ret)
+		return ret;
 
-		nvkm_wr32(device, 0x409614, 0x00000070);
-		nvkm_wr32(device, 0x409614, 0x00000770);
-		nvkm_wr32(device, 0x40802c, 0x00000001);
+	/*XXX: We (likely) require PMU support to even bother with this.
+	 *
+	 *     Also, it seems like not all GPUs support ELPG.  Traces I
+	 *     have here show RM enabling it on Kepler/Turing, but none
+	 *     of the GPUs between those.  NVGPU decides this by PCIID.
+	 */
+	if (0) {
+		ret = gf100_gr_fecs_elpg_bind(gr);
+		if (ret)
+			return ret;
 	}
 
+	/* Generate golden context image. */
 	if (gr->data == NULL) {
 		int ret = gf100_grctx_generate(gr);
 		if (ret) {
@@ -1614,15 +1784,19 @@ gf100_gr_init_ctxctl_int(struct gf100_gr *gr)
 
 	/* load HUB microcode */
 	nvkm_mc_unk260(device, 0);
-	nvkm_falcon_load_dmem(gr->fecs, gr->func->fecs.ucode->data.data, 0x0,
+	nvkm_falcon_load_dmem(gr->fecs.falcon,
+			      gr->func->fecs.ucode->data.data, 0x0,
 			      gr->func->fecs.ucode->data.size, 0);
-	nvkm_falcon_load_imem(gr->fecs, gr->func->fecs.ucode->code.data, 0x0,
+	nvkm_falcon_load_imem(gr->fecs.falcon,
+			      gr->func->fecs.ucode->code.data, 0x0,
 			      gr->func->fecs.ucode->code.size, 0, 0, false);
 
 	/* load GPC microcode */
-	nvkm_falcon_load_dmem(gr->gpccs, gr->func->gpccs.ucode->data.data, 0x0,
+	nvkm_falcon_load_dmem(gr->gpccs.falcon,
+			      gr->func->gpccs.ucode->data.data, 0x0,
 			      gr->func->gpccs.ucode->data.size, 0);
-	nvkm_falcon_load_imem(gr->gpccs, gr->func->gpccs.ucode->code.data, 0x0,
+	nvkm_falcon_load_imem(gr->gpccs.falcon,
+			      gr->func->gpccs.ucode->code.data, 0x0,
 			      gr->func->gpccs.ucode->code.size, 0, 0, false);
 	nvkm_mc_unk260(device, 1);
 
@@ -1769,11 +1943,13 @@ gf100_gr_oneinit(struct nvkm_gr *base)
 	int i, j;
 	int ret;
 
-	ret = nvkm_falcon_v1_new(subdev, "FECS", 0x409000, &gr->fecs);
+	ret = nvkm_falcon_v1_new(subdev, "FECS", 0x409000, &gr->fecs.falcon);
 	if (ret)
 		return ret;
 
-	ret = nvkm_falcon_v1_new(subdev, "GPCCS", 0x41a000, &gr->gpccs);
+	mutex_init(&gr->fecs.mutex);
+
+	ret = nvkm_falcon_v1_new(subdev, "GPCCS", 0x41a000, &gr->gpccs.falcon);
 	if (ret)
 		return ret;
 
@@ -1816,11 +1992,11 @@ gf100_gr_init_(struct nvkm_gr *base)
 
 	nvkm_pmu_pgob(gr->base.engine.subdev.device->pmu, false);
 
-	ret = nvkm_falcon_get(gr->fecs, subdev);
+	ret = nvkm_falcon_get(gr->fecs.falcon, subdev);
 	if (ret)
 		return ret;
 
-	ret = nvkm_falcon_get(gr->gpccs, subdev);
+	ret = nvkm_falcon_get(gr->gpccs.falcon, subdev);
 	if (ret)
 		return ret;
 
@@ -1832,8 +2008,8 @@ gf100_gr_fini_(struct nvkm_gr *base, bool suspend)
 {
 	struct gf100_gr *gr = gf100_gr(base);
 	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
-	nvkm_falcon_put(gr->gpccs, subdev);
-	nvkm_falcon_put(gr->fecs, subdev);
+	nvkm_falcon_put(gr->gpccs.falcon, subdev);
+	nvkm_falcon_put(gr->fecs.falcon, subdev);
 	return 0;
 }
 
@@ -1859,8 +2035,8 @@ gf100_gr_dtor(struct nvkm_gr *base)
 		gr->func->dtor(gr);
 	kfree(gr->data);
 
-	nvkm_falcon_del(&gr->gpccs);
-	nvkm_falcon_del(&gr->fecs);
+	nvkm_falcon_del(&gr->gpccs.falcon);
+	nvkm_falcon_del(&gr->fecs.falcon);
 
 	gf100_gr_dtor_fw(&gr->fuc409c);
 	gf100_gr_dtor_fw(&gr->fuc409d);
@@ -1886,6 +2062,9 @@ gf100_gr_ = {
 	.chan_new = gf100_gr_chan_new,
 	.object_get = gf100_gr_object_get,
 	.chsw_load = gf100_gr_chsw_load,
+	.ctxsw.pause = gf100_gr_fecs_stop_ctxsw,
+	.ctxsw.resume = gf100_gr_fecs_start_ctxsw,
+	.ctxsw.inst = gf100_gr_ctxsw_inst,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
index dc46cf0131db..fafdd0bbea9b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
@@ -82,8 +82,16 @@ struct gf100_gr {
 	const struct gf100_gr_func *func;
 	struct nvkm_gr base;
 
-	struct nvkm_falcon *fecs;
-	struct nvkm_falcon *gpccs;
+	struct {
+		struct nvkm_falcon *falcon;
+		struct mutex mutex;
+		u32 disable;
+	} fecs;
+
+	struct {
+		struct nvkm_falcon *falcon;
+	} gpccs;
+
 	struct gf100_gr_fuc fuc409c;
 	struct gf100_gr_fuc fuc409d;
 	struct gf100_gr_fuc fuc41ac;
@@ -128,6 +136,8 @@ struct gf100_gr {
 	struct gf100_gr_mmio mmio_list[4096/8];
 	u32  size;
 	u32 *data;
+	u32 size_zcull;
+	u32 size_pm;
 };
 
 int gf100_gr_ctor(const struct gf100_gr_func *, struct nvkm_device *,
@@ -136,6 +146,8 @@ int gf100_gr_new_(const struct gf100_gr_func *, struct nvkm_device *,
 		  int, struct nvkm_gr **);
 void *gf100_gr_dtor(struct nvkm_gr *);
 
+int gf100_gr_fecs_bind_pointer(struct gf100_gr *, u32 inst);
+
 struct gf100_gr_func_zbc {
 	void (*clear_color)(struct gf100_gr *, int zbc);
 	void (*clear_depth)(struct gf100_gr *, int zbc);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/priv.h
index 66359c23cbce..d4d5601c51e7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/priv.h
@@ -27,6 +27,11 @@ struct nvkm_gr_func {
 	 */
 	u64 (*units)(struct nvkm_gr *);
 	bool (*chsw_load)(struct nvkm_gr *);
+	struct {
+		int (*pause)(struct nvkm_gr *);
+		int (*resume)(struct nvkm_gr *);
+		u32 (*inst)(struct nvkm_gr *);
+	} ctxsw;
 	struct nvkm_sclass sclass[];
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/base.c
index 4807021fd990..4a63581bdd5e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/base.c
@@ -21,13 +21,21 @@
  */
 #include "priv.h"
 
+#include <subdev/top.h>
 #include <engine/falcon.h>
 
 static int
 nvkm_nvdec_oneinit(struct nvkm_engine *engine)
 {
 	struct nvkm_nvdec *nvdec = nvkm_nvdec(engine);
-	return nvkm_falcon_v1_new(&nvdec->engine.subdev, "NVDEC", 0x84000,
+	struct nvkm_subdev *subdev = &nvdec->engine.subdev;
+
+	nvdec->addr = nvkm_top_addr(subdev->device, subdev->index);
+	if (!nvdec->addr)
+		return -EINVAL;
+
+	/*XXX: fix naming of this when adding support for multiple-NVDEC */
+	return nvkm_falcon_v1_new(subdev, "NVDEC", nvdec->addr,
 				  &nvdec->falcon);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/Kbuild
index 4b17254cfbd0..d9cdea7d9353 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/Kbuild
@@ -1,2 +1,3 @@
 nvkm-y += nvkm/engine/sec2/base.o
 nvkm-y += nvkm/engine/sec2/gp102.o
+nvkm-y += nvkm/engine/sec2/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/base.c
index f865d2a3e184..1b49e5b6717f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/base.c
@@ -22,6 +22,7 @@
 #include "priv.h"
 
 #include <core/msgqueue.h>
+#include <subdev/top.h>
 #include <engine/falcon.h>
 
 static void *
@@ -39,18 +40,18 @@ nvkm_sec2_intr(struct nvkm_engine *engine)
 	struct nvkm_sec2 *sec2 = nvkm_sec2(engine);
 	struct nvkm_subdev *subdev = &engine->subdev;
 	struct nvkm_device *device = subdev->device;
-	u32 disp = nvkm_rd32(device, 0x8701c);
-	u32 intr = nvkm_rd32(device, 0x87008) & disp & ~(disp >> 16);
+	u32 disp = nvkm_rd32(device, sec2->addr + 0x01c);
+	u32 intr = nvkm_rd32(device, sec2->addr + 0x008) & disp & ~(disp >> 16);
 
 	if (intr & 0x00000040) {
 		schedule_work(&sec2->work);
-		nvkm_wr32(device, 0x87004, 0x00000040);
+		nvkm_wr32(device, sec2->addr + 0x004, 0x00000040);
 		intr &= ~0x00000040;
 	}
 
 	if (intr) {
 		nvkm_error(subdev, "unhandled intr %08x\n", intr);
-		nvkm_wr32(device, 0x87004, intr);
+		nvkm_wr32(device, sec2->addr + 0x004, intr);
 
 	}
 }
@@ -74,8 +75,15 @@ static int
 nvkm_sec2_oneinit(struct nvkm_engine *engine)
 {
 	struct nvkm_sec2 *sec2 = nvkm_sec2(engine);
-	return nvkm_falcon_v1_new(&sec2->engine.subdev, "SEC2", 0x87000,
-				  &sec2->falcon);
+	struct nvkm_subdev *subdev = &sec2->engine.subdev;
+
+	if (!sec2->addr) {
+		sec2->addr = nvkm_top_addr(subdev->device, subdev->index);
+		if (WARN_ON(!sec2->addr))
+			return -EINVAL;
+	}
+
+	return nvkm_falcon_v1_new(subdev, "SEC2", sec2->addr, &sec2->falcon);
 }
 
 static int
@@ -95,13 +103,14 @@ nvkm_sec2 = {
 };
 
 int
-nvkm_sec2_new_(struct nvkm_device *device, int index,
+nvkm_sec2_new_(struct nvkm_device *device, int index, u32 addr,
 	       struct nvkm_sec2 **psec2)
 {
 	struct nvkm_sec2 *sec2;
 
 	if (!(sec2 = *psec2 = kzalloc(sizeof(*sec2), GFP_KERNEL)))
 		return -ENOMEM;
+	sec2->addr = addr;
 	INIT_WORK(&sec2->work, nvkm_sec2_recv);
 
 	return nvkm_engine_ctor(&nvkm_sec2, device, index, true, &sec2->engine);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp102.c b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp102.c
index 9be1524c08f5..858cf27fa010 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp102.c
@@ -26,5 +26,5 @@ int
 gp102_sec2_new(struct nvkm_device *device, int index,
 	       struct nvkm_sec2 **psec2)
 {
-	return nvkm_sec2_new_(device, index, psec2);
+	return nvkm_sec2_new_(device, index, 0, psec2);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/priv.h
index 2f97c806a79d..ab0165e2d1a3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/priv.h
@@ -5,6 +5,5 @@
 
 #define nvkm_sec2(p) container_of((p), struct nvkm_sec2, engine)
 
-int nvkm_sec2_new_(struct nvkm_device *, int, struct nvkm_sec2 **);
-
+int nvkm_sec2_new_(struct nvkm_device *, int, u32 addr, struct nvkm_sec2 **);
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/dce120/i2caux_dce120.h b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/tu102.c
index b6ac47617c70..d655576164b1 100644
--- a/drivers/gpu/drm/amd/display/dc/i2caux/dce120/i2caux_dce120.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/tu102.c
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-16 Advanced Micro Devices, Inc.
+ * Copyright 2019 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -18,15 +18,16 @@
  * 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_I2C_AUX_DCE120_H__
-#define __DAL_I2C_AUX_DCE120_H__
-
-struct i2caux *dal_i2caux_dce120_create(
-	struct dc_context *ctx);
+#include "priv.h"
 
-#endif /* __DAL_I2C_AUX_DCE120_H__ */
+int
+tu102_sec2_new(struct nvkm_device *device, int index,
+	       struct nvkm_sec2 **psec2)
+{
+	/* TOP info wasn't updated on Turing to reflect the PRI
+	 * address change for some reason.  We override it here.
+	 */
+	return nvkm_sec2_new_(device, index, 0x840000, psec2);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/falcon/base.c b/drivers/gpu/drm/nouveau/nvkm/falcon/base.c
index 427340153640..366c87de6e72 100644
--- a/drivers/gpu/drm/nouveau/nvkm/falcon/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/falcon/base.c
@@ -204,6 +204,9 @@ nvkm_falcon_ctor(const struct nvkm_falcon_func *func,
 		debug_reg = 0x408;
 		falcon->has_emem = true;
 		break;
+	case NVKM_SUBDEV_GSP:
+		debug_reg = 0x0; /*XXX*/
+		break;
 	default:
 		nvkm_warn(subdev, "unsupported falcon %s!\n",
 			  nvkm_subdev_name[subdev->index]);
diff --git a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
index 771e16a16267..a8bee1e046aa 100644
--- a/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
+++ b/drivers/gpu/drm/nouveau/nvkm/falcon/msgqueue.c
@@ -269,7 +269,7 @@ cmd_write(struct nvkm_msgqueue *priv, struct nvkm_msgqueue_hdr *cmd,
 		commit = false;
 	}
 
-	   cmd_queue_close(priv, queue, commit);
+	cmd_queue_close(priv, queue, commit);
 
 	return ret;
 }
@@ -347,7 +347,7 @@ nvkm_msgqueue_post(struct nvkm_msgqueue *priv, enum msgqueue_msg_priority prio,
 	ret = cmd_write(priv, cmd, queue);
 	if (ret) {
 		seq->state = SEQ_STATE_PENDING;
-		      msgqueue_seq_release(priv, seq);
+		msgqueue_seq_release(priv, seq);
 	}
 
 	return ret;
@@ -373,7 +373,7 @@ msgqueue_msg_handle(struct nvkm_msgqueue *priv, struct nvkm_msgqueue_hdr *hdr)
 	if (seq->completion)
 		complete(seq->completion);
 
-	   msgqueue_seq_release(priv, seq);
+	msgqueue_seq_release(priv, seq);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
index cfdffef1afb9..a339fe03d423 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
@@ -7,6 +7,7 @@ include $(src)/nvkm/subdev/fault/Kbuild
 include $(src)/nvkm/subdev/fb/Kbuild
 include $(src)/nvkm/subdev/fuse/Kbuild
 include $(src)/nvkm/subdev/gpio/Kbuild
+include $(src)/nvkm/subdev/gsp/Kbuild
 include $(src)/nvkm/subdev/i2c/Kbuild
 include $(src)/nvkm/subdev/ibus/Kbuild
 include $(src)/nvkm/subdev/iccsense/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
index ab0282dc0736..dc300600c019 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
@@ -5,4 +5,4 @@ nvkm-y += nvkm/subdev/bar/gf100.o
 nvkm-y += nvkm/subdev/bar/gk20a.o
 nvkm-y += nvkm/subdev/bar/gm107.o
 nvkm-y += nvkm/subdev/bar/gm20b.o
-nvkm-y += nvkm/subdev/bar/tu104.o
+nvkm-y += nvkm/subdev/bar/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu102.c
index ecaead156e9b..798f65ec3a86 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu102.c
@@ -25,7 +25,7 @@
 #include <subdev/timer.h>
 
 static void
-tu104_bar_bar2_wait(struct nvkm_bar *bar)
+tu102_bar_bar2_wait(struct nvkm_bar *bar)
 {
 	struct nvkm_device *device = bar->subdev.device;
 	nvkm_msec(device, 2000,
@@ -35,13 +35,13 @@ tu104_bar_bar2_wait(struct nvkm_bar *bar)
 }
 
 static void
-tu104_bar_bar2_fini(struct nvkm_bar *bar)
+tu102_bar_bar2_fini(struct nvkm_bar *bar)
 {
 	nvkm_mask(bar->subdev.device, 0xb80f48, 0x80000000, 0x00000000);
 }
 
 static void
-tu104_bar_bar2_init(struct nvkm_bar *base)
+tu102_bar_bar2_init(struct nvkm_bar *base)
 {
 	struct nvkm_device *device = base->subdev.device;
 	struct gf100_bar *bar = gf100_bar(base);
@@ -52,7 +52,7 @@ tu104_bar_bar2_init(struct nvkm_bar *base)
 }
 
 static void
-tu104_bar_bar1_wait(struct nvkm_bar *bar)
+tu102_bar_bar1_wait(struct nvkm_bar *bar)
 {
 	struct nvkm_device *device = bar->subdev.device;
 	nvkm_msec(device, 2000,
@@ -62,13 +62,13 @@ tu104_bar_bar1_wait(struct nvkm_bar *bar)
 }
 
 static void
-tu104_bar_bar1_fini(struct nvkm_bar *bar)
+tu102_bar_bar1_fini(struct nvkm_bar *bar)
 {
 	nvkm_mask(bar->subdev.device, 0xb80f40, 0x80000000, 0x00000000);
 }
 
 static void
-tu104_bar_bar1_init(struct nvkm_bar *base)
+tu102_bar_bar1_init(struct nvkm_bar *base)
 {
 	struct nvkm_device *device = base->subdev.device;
 	struct gf100_bar *bar = gf100_bar(base);
@@ -77,22 +77,22 @@ tu104_bar_bar1_init(struct nvkm_bar *base)
 }
 
 static const struct nvkm_bar_func
-tu104_bar = {
+tu102_bar = {
 	.dtor = gf100_bar_dtor,
 	.oneinit = gf100_bar_oneinit,
-	.bar1.init = tu104_bar_bar1_init,
-	.bar1.fini = tu104_bar_bar1_fini,
-	.bar1.wait = tu104_bar_bar1_wait,
+	.bar1.init = tu102_bar_bar1_init,
+	.bar1.fini = tu102_bar_bar1_fini,
+	.bar1.wait = tu102_bar_bar1_wait,
 	.bar1.vmm = gf100_bar_bar1_vmm,
-	.bar2.init = tu104_bar_bar2_init,
-	.bar2.fini = tu104_bar_bar2_fini,
-	.bar2.wait = tu104_bar_bar2_wait,
+	.bar2.init = tu102_bar_bar2_init,
+	.bar2.fini = tu102_bar_bar2_fini,
+	.bar2.wait = tu102_bar_bar2_wait,
 	.bar2.vmm = gf100_bar_bar2_vmm,
 	.flush = g84_bar_flush,
 };
 
 int
-tu104_bar_new(struct nvkm_device *device, int index, struct nvkm_bar **pbar)
+tu102_bar_new(struct nvkm_device *device, int index, struct nvkm_bar **pbar)
 {
-	return gf100_bar_new_(&tu104_bar, device, index, pbar);
+	return gf100_bar_new_(&tu102_bar, device, index, pbar);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c
index 3133b28f849c..b099d1209be8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/dp.c
@@ -212,7 +212,7 @@ nvbios_dpcfg_match(struct nvkm_bios *bios, u16 outp, u8 pc, u8 vs, u8 pe,
 	u16 data;
 
 	if (*ver >= 0x30) {
-		const u8 vsoff[] = { 0, 4, 7, 9 };
+		static const u8 vsoff[] = { 0, 4, 7, 9 };
 		idx = (pc * 10) + vsoff[vs] + pe;
 		if (*ver >= 0x40 && *ver <= 0x41 && *hdr >= 0x12)
 			idx += nvbios_rd08(bios, outp + 0x11) * 40;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
index 9cc10e438b3d..ec0e9f7224b5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
@@ -806,12 +806,12 @@ init_generic_condition(struct nvbios_init *init)
 	init->offset += 3;
 
 	switch (cond) {
-	case 0:
+	case 0: /* CONDITION_ID_INT_DP. */
 		if (init_conn(init) != DCB_CONNECTOR_eDP)
 			init_exec_set(init, false);
 		break;
-	case 1:
-	case 2:
+	case 1: /* CONDITION_ID_USE_SPPLL0. */
+	case 2: /* CONDITION_ID_USE_SPPLL1. */
 		if ( init->outp &&
 		    (data = nvbios_dpout_match(bios, DCB_OUTPUT_DP,
 					       (init->outp->or << 0) |
@@ -826,10 +826,13 @@ init_generic_condition(struct nvbios_init *init)
 		if (init_exec(init))
 			warn("script needs dp output table data\n");
 		break;
-	case 5:
+	case 5: /* CONDITION_ID_ASSR_SUPPORT. */
 		if (!(init_rdauxr(init, 0x0d) & 1))
 			init_exec_set(init, false);
 		break;
+	case 7: /* CONDITION_ID_NO_PANEL_SEQ_DELAYS. */
+		init_exec_set(init, false);
+		break;
 	default:
 		warn("INIT_GENERIC_CONDITON: unknown 0x%02x\n", cond);
 		init->offset += size;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
index 3ef505a5c01b..f3c388932b6f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
@@ -13,4 +13,4 @@ nvkm-y += nvkm/subdev/devinit/gf100.o
 nvkm-y += nvkm/subdev/devinit/gm107.o
 nvkm-y += nvkm/subdev/devinit/gm200.o
 nvkm-y += nvkm/subdev/devinit/gv100.o
-nvkm-y += nvkm/subdev/devinit/tu104.o
+nvkm-y += nvkm/subdev/devinit/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu102.c
index aae87b3fc429..397670e72fff 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu102.c
@@ -26,7 +26,7 @@
 #include <subdev/clk/pll.h>
 
 static int
-tu104_devinit_pll_set(struct nvkm_devinit *init, u32 type, u32 freq)
+tu102_devinit_pll_set(struct nvkm_devinit *init, u32 type, u32 freq)
 {
 	struct nvkm_subdev *subdev = &init->subdev;
 	struct nvkm_device *device = subdev->device;
@@ -66,7 +66,7 @@ tu104_devinit_pll_set(struct nvkm_devinit *init, u32 type, u32 freq)
 }
 
 static int
-tu104_devinit_post(struct nvkm_devinit *base, bool post)
+tu102_devinit_post(struct nvkm_devinit *base, bool post)
 {
 	struct nv50_devinit *init = nv50_devinit(base);
 	gm200_devinit_preos(init, post);
@@ -74,16 +74,16 @@ tu104_devinit_post(struct nvkm_devinit *base, bool post)
 }
 
 static const struct nvkm_devinit_func
-tu104_devinit = {
+tu102_devinit = {
 	.init = nv50_devinit_init,
-	.post = tu104_devinit_post,
-	.pll_set = tu104_devinit_pll_set,
+	.post = tu102_devinit_post,
+	.pll_set = tu102_devinit_pll_set,
 	.disable = gm107_devinit_disable,
 };
 
 int
-tu104_devinit_new(struct nvkm_device *device, int index,
+tu102_devinit_new(struct nvkm_device *device, int index,
 		struct nvkm_devinit **pinit)
 {
-	return nv50_devinit_new_(&tu104_devinit, device, index, pinit);
+	return nv50_devinit_new_(&tu102_devinit, device, index, pinit);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild
index 794eb1745b2f..42586267fc08 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild
@@ -1,4 +1,5 @@
 nvkm-y += nvkm/subdev/fault/base.o
+nvkm-y += nvkm/subdev/fault/user.o
 nvkm-y += nvkm/subdev/fault/gp100.o
 nvkm-y += nvkm/subdev/fault/gv100.o
-nvkm-y += nvkm/subdev/fault/tu104.o
+nvkm-y += nvkm/subdev/fault/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c
index 4ba1e21e8fda..ca251560d3e0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c
@@ -176,5 +176,7 @@ nvkm_fault_new_(const struct nvkm_fault_func *func, struct nvkm_device *device,
 		return -ENOMEM;
 	nvkm_subdev_ctor(&nvkm_fault, device, index, &fault->subdev);
 	fault->func = func;
+	fault->user.ctor = nvkm_ufault_new;
+	fault->user.base = func->user.base;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c
index 8fb96fe614f9..4f3c4e091117 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c
@@ -23,6 +23,8 @@
 
 #include <subdev/mc.h>
 
+#include <nvif/class.h>
+
 static void
 gp100_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable)
 {
@@ -69,6 +71,7 @@ gp100_fault = {
 	.buffer.init = gp100_fault_buffer_init,
 	.buffer.fini = gp100_fault_buffer_fini,
 	.buffer.intr = gp100_fault_buffer_intr,
+	.user = { { 0, 0, MAXWELL_FAULT_BUFFER_A }, 0 },
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c
index 6fc54e17c935..6747f09c2dc3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c
@@ -25,6 +25,8 @@
 #include <subdev/mmu.h>
 #include <engine/fifo.h>
 
+#include <nvif/class.h>
+
 static void
 gv100_fault_buffer_process(struct nvkm_fault_buffer *buffer)
 {
@@ -166,6 +168,13 @@ gv100_fault_intr(struct nvkm_fault *fault)
 		}
 	}
 
+	if (stat & 0x08000000) {
+		if (fault->buffer[1]) {
+			nvkm_event_send(&fault->event, 1, 1, NULL, 0);
+			stat &= ~0x08000000;
+		}
+	}
+
 	if (stat) {
 		nvkm_debug(subdev, "intr %08x\n", stat);
 	}
@@ -208,6 +217,13 @@ gv100_fault = {
 	.buffer.init = gv100_fault_buffer_init,
 	.buffer.fini = gv100_fault_buffer_fini,
 	.buffer.intr = gv100_fault_buffer_intr,
+	/*TODO: Figure out how to expose non-replayable fault buffer, which,
+	 *      for some reason, is where recoverable CE faults appear...
+	 *
+	 * 	It's a bit tricky, as both NVKM and SVM will need access to
+	 * 	the non-replayable fault buffer.
+	 */
+	.user = { { 0, 0, VOLTA_FAULT_BUFFER_A }, 1 },
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h
index 8ca8b2876dad..975e66ac6344 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h
@@ -34,7 +34,14 @@ struct nvkm_fault_func {
 		void (*fini)(struct nvkm_fault_buffer *);
 		void (*intr)(struct nvkm_fault_buffer *, bool enable);
 	} buffer;
+	struct {
+		struct nvkm_sclass base;
+		int rp;
+	} user;
 };
 
 int gv100_fault_oneinit(struct nvkm_fault *);
+
+int nvkm_ufault_new(struct nvkm_device *, const struct nvkm_oclass *,
+		    void *, u32, struct nvkm_object **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu102.c
index 9c8a3adf99d7..fa1dfe5692b0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu102.c
@@ -28,7 +28,7 @@
 #include <nvif/class.h>
 
 static void
-tu104_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable)
+tu102_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable)
 {
 	/*XXX: Earlier versions of RM touched the old regs on Turing,
 	 *     which don't appear to actually work anymore, but newer
@@ -37,7 +37,7 @@ tu104_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable)
 }
 
 static void
-tu104_fault_buffer_fini(struct nvkm_fault_buffer *buffer)
+tu102_fault_buffer_fini(struct nvkm_fault_buffer *buffer)
 {
 	struct nvkm_device *device = buffer->fault->subdev.device;
 	const u32 foff = buffer->id * 0x20;
@@ -45,7 +45,7 @@ tu104_fault_buffer_fini(struct nvkm_fault_buffer *buffer)
 }
 
 static void
-tu104_fault_buffer_init(struct nvkm_fault_buffer *buffer)
+tu102_fault_buffer_init(struct nvkm_fault_buffer *buffer)
 {
 	struct nvkm_device *device = buffer->fault->subdev.device;
 	const u32 foff = buffer->id * 0x20;
@@ -57,7 +57,7 @@ tu104_fault_buffer_init(struct nvkm_fault_buffer *buffer)
 }
 
 static void
-tu104_fault_buffer_info(struct nvkm_fault_buffer *buffer)
+tu102_fault_buffer_info(struct nvkm_fault_buffer *buffer)
 {
 	struct nvkm_device *device = buffer->fault->subdev.device;
 	const u32 foff = buffer->id * 0x20;
@@ -70,7 +70,7 @@ tu104_fault_buffer_info(struct nvkm_fault_buffer *buffer)
 }
 
 static void
-tu104_fault_intr_fault(struct nvkm_fault *fault)
+tu102_fault_intr_fault(struct nvkm_fault *fault)
 {
 	struct nvkm_subdev *subdev = &fault->subdev;
 	struct nvkm_device *device = subdev->device;
@@ -96,14 +96,14 @@ tu104_fault_intr_fault(struct nvkm_fault *fault)
 }
 
 static void
-tu104_fault_intr(struct nvkm_fault *fault)
+tu102_fault_intr(struct nvkm_fault *fault)
 {
 	struct nvkm_subdev *subdev = &fault->subdev;
 	struct nvkm_device *device = subdev->device;
 	u32 stat = nvkm_rd32(device, 0xb83094);
 
 	if (stat & 0x80000000) {
-		tu104_fault_intr_fault(fault);
+		tu102_fault_intr_fault(fault);
 		nvkm_wr32(device, 0xb83094, 0x80000000);
 		stat &= ~0x80000000;
 	}
@@ -129,7 +129,7 @@ tu104_fault_intr(struct nvkm_fault *fault)
 }
 
 static void
-tu104_fault_fini(struct nvkm_fault *fault)
+tu102_fault_fini(struct nvkm_fault *fault)
 {
 	nvkm_notify_put(&fault->nrpfb);
 	if (fault->buffer[0])
@@ -138,7 +138,7 @@ tu104_fault_fini(struct nvkm_fault *fault)
 }
 
 static void
-tu104_fault_init(struct nvkm_fault *fault)
+tu102_fault_init(struct nvkm_fault *fault)
 {
 	/*XXX: enable priv faults */
 	fault->func->buffer.init(fault->buffer[0]);
@@ -146,22 +146,23 @@ tu104_fault_init(struct nvkm_fault *fault)
 }
 
 static const struct nvkm_fault_func
-tu104_fault = {
+tu102_fault = {
 	.oneinit = gv100_fault_oneinit,
-	.init = tu104_fault_init,
-	.fini = tu104_fault_fini,
-	.intr = tu104_fault_intr,
+	.init = tu102_fault_init,
+	.fini = tu102_fault_fini,
+	.intr = tu102_fault_intr,
 	.buffer.nr = 2,
 	.buffer.entry_size = 32,
-	.buffer.info = tu104_fault_buffer_info,
-	.buffer.init = tu104_fault_buffer_init,
-	.buffer.fini = tu104_fault_buffer_fini,
-	.buffer.intr = tu104_fault_buffer_intr,
+	.buffer.info = tu102_fault_buffer_info,
+	.buffer.init = tu102_fault_buffer_init,
+	.buffer.fini = tu102_fault_buffer_fini,
+	.buffer.intr = tu102_fault_buffer_intr,
+	.user = { { 0, 0, VOLTA_FAULT_BUFFER_A }, 1 },
 };
 
 int
-tu104_fault_new(struct nvkm_device *device, int index,
+tu102_fault_new(struct nvkm_device *device, int index,
 		struct nvkm_fault **pfault)
 {
-	return nvkm_fault_new_(&tu104_fault, device, index, pfault);
+	return nvkm_fault_new_(&tu102_fault, device, index, pfault);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c
new file mode 100644
index 000000000000..ac835c9582fd
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c
@@ -0,0 +1,106 @@
+/*
+ * Copyright 2018 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "priv.h"
+
+#include <core/memory.h>
+#include <subdev/mmu.h>
+
+#include <nvif/clb069.h>
+#include <nvif/unpack.h>
+
+static int
+nvkm_ufault_map(struct nvkm_object *object, void *argv, u32 argc,
+		enum nvkm_object_map *type, u64 *addr, u64 *size)
+{
+	struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object);
+	struct nvkm_device *device = buffer->fault->subdev.device;
+	*type = NVKM_OBJECT_MAP_IO;
+	*addr = device->func->resource_addr(device, 3) + buffer->addr;
+	*size = nvkm_memory_size(buffer->mem);
+	return 0;
+}
+
+static int
+nvkm_ufault_ntfy(struct nvkm_object *object, u32 type,
+		 struct nvkm_event **pevent)
+{
+	struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object);
+	if (type == NVB069_V0_NTFY_FAULT) {
+		*pevent = &buffer->fault->event;
+		return 0;
+	}
+	return -EINVAL;
+}
+
+static int
+nvkm_ufault_fini(struct nvkm_object *object, bool suspend)
+{
+	struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object);
+	buffer->fault->func->buffer.fini(buffer);
+	return 0;
+}
+
+static int
+nvkm_ufault_init(struct nvkm_object *object)
+{
+	struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object);
+	buffer->fault->func->buffer.init(buffer);
+	return 0;
+}
+
+static void *
+nvkm_ufault_dtor(struct nvkm_object *object)
+{
+	return NULL;
+}
+
+static const struct nvkm_object_func
+nvkm_ufault = {
+	.dtor = nvkm_ufault_dtor,
+	.init = nvkm_ufault_init,
+	.fini = nvkm_ufault_fini,
+	.ntfy = nvkm_ufault_ntfy,
+	.map = nvkm_ufault_map,
+};
+
+int
+nvkm_ufault_new(struct nvkm_device *device, const struct nvkm_oclass *oclass,
+		void *argv, u32 argc, struct nvkm_object **pobject)
+{
+	union {
+		struct nvif_clb069_v0 v0;
+	} *args = argv;
+	struct nvkm_fault *fault = device->fault;
+	struct nvkm_fault_buffer *buffer = fault->buffer[fault->func->user.rp];
+	int ret = -ENOSYS;
+
+	if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) {
+		args->v0.entries = buffer->entries;
+		args->v0.get = buffer->get;
+		args->v0.put = buffer->put;
+	} else
+		return ret;
+
+	nvkm_object_ctor(&nvkm_ufault, oclass, &buffer->object);
+	*pobject = &buffer->object;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
index 60ece0a8a2e1..1d2d6bae73cd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
@@ -87,7 +87,7 @@ nvkm_gddr3_calc(struct nvkm_ram *ram)
 		WR  = (ram->next->bios.timing[2] & 0x007f0000) >> 16;
 		/* XXX: Get these values from the VBIOS instead */
 		DLL = !(ram->mr[1] & 0x1);
-		RON = !(ram->mr[1] & 0x300) >> 8;
+		RON = !((ram->mr[1] & 0x300) >> 8);
 		break;
 	default:
 		return -ENOSYS;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild
new file mode 100644
index 000000000000..26fc6feb807e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild
@@ -0,0 +1 @@
+nvkm-y += nvkm/subdev/gsp/gv100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gv100.c
new file mode 100644
index 000000000000..dccfaf1162e2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gv100.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright 2019 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <subdev/gsp.h>
+#include <subdev/top.h>
+#include <engine/falcon.h>
+
+static int
+gv100_gsp_oneinit(struct nvkm_subdev *subdev)
+{
+	struct nvkm_gsp *gsp = nvkm_gsp(subdev);
+
+	gsp->addr = nvkm_top_addr(subdev->device, subdev->index);
+	if (!gsp->addr)
+		return -EINVAL;
+
+	return nvkm_falcon_v1_new(subdev, "GSP", gsp->addr, &gsp->falcon);
+}
+
+static void *
+gv100_gsp_dtor(struct nvkm_subdev *subdev)
+{
+	struct nvkm_gsp *gsp = nvkm_gsp(subdev);
+	nvkm_falcon_del(&gsp->falcon);
+	return gsp;
+}
+
+static const struct nvkm_subdev_func
+gv100_gsp = {
+	.dtor = gv100_gsp_dtor,
+	.oneinit = gv100_gsp_oneinit,
+};
+
+int
+gv100_gsp_new(struct nvkm_device *device, int index, struct nvkm_gsp **pgsp)
+{
+	struct nvkm_gsp *gsp;
+
+	if (!(gsp = *pgsp = kzalloc(sizeof(*gsp), GFP_KERNEL)))
+		return -ENOMEM;
+
+	nvkm_subdev_ctor(&gv100_gsp, device, index, &gsp->subdev);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild
index f3b06329c338..c64e399326b3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild
@@ -12,4 +12,4 @@ nvkm-y += nvkm/subdev/mc/gk104.o
 nvkm-y += nvkm/subdev/mc/gk20a.o
 nvkm-y += nvkm/subdev/mc/gp100.o
 nvkm-y += nvkm/subdev/mc/gp10b.o
-nvkm-y += nvkm/subdev/mc/tu104.o
+nvkm-y += nvkm/subdev/mc/tu102.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu102.c
index b7165bd18999..d098c44a4fcb 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu102.c
@@ -22,7 +22,7 @@
 #include "priv.h"
 
 static void
-tu104_mc_intr_hack(struct nvkm_mc *mc, bool *handled)
+tu102_mc_intr_hack(struct nvkm_mc *mc, bool *handled)
 {
 	struct nvkm_device *device = mc->subdev.device;
 	u32 stat = nvkm_rd32(device, 0xb81010);
@@ -37,19 +37,19 @@ tu104_mc_intr_hack(struct nvkm_mc *mc, bool *handled)
 }
 
 static const struct nvkm_mc_func
-tu104_mc = {
+tu102_mc = {
 	.init = nv50_mc_init,
 	.intr = gp100_mc_intr,
 	.intr_unarm = gp100_mc_intr_unarm,
 	.intr_rearm = gp100_mc_intr_rearm,
 	.intr_mask = gp100_mc_intr_mask,
 	.intr_stat = gf100_mc_intr_stat,
-	.intr_hack = tu104_mc_intr_hack,
+	.intr_hack = tu102_mc_intr_hack,
 	.reset = gk104_mc_reset,
 };
 
 int
-tu104_mc_new(struct nvkm_device *device, int index, struct nvkm_mc **pmc)
+tu102_mc_new(struct nvkm_device *device, int index, struct nvkm_mc **pmc)
 {
-	return gp100_mc_new_(&tu104_mc, device, index, pmc);
+	return gp100_mc_new_(&tu102_mc, device, index, pmc);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
index 8966180b36cc..db9c56028f21 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
@@ -13,7 +13,7 @@ nvkm-y += nvkm/subdev/mmu/gm20b.o
 nvkm-y += nvkm/subdev/mmu/gp100.o
 nvkm-y += nvkm/subdev/mmu/gp10b.o
 nvkm-y += nvkm/subdev/mmu/gv100.o
-nvkm-y += nvkm/subdev/mmu/tu104.o
+nvkm-y += nvkm/subdev/mmu/tu102.o
 
 nvkm-y += nvkm/subdev/mmu/mem.o
 nvkm-y += nvkm/subdev/mmu/memnv04.o
@@ -34,7 +34,7 @@ nvkm-y += nvkm/subdev/mmu/vmmgm20b.o
 nvkm-y += nvkm/subdev/mmu/vmmgp100.o
 nvkm-y += nvkm/subdev/mmu/vmmgp10b.o
 nvkm-y += nvkm/subdev/mmu/vmmgv100.o
-nvkm-y += nvkm/subdev/mmu/vmmtu104.o
+nvkm-y += nvkm/subdev/mmu/vmmtu102.o
 
 nvkm-y += nvkm/subdev/mmu/umem.o
 nvkm-y += nvkm/subdev/mmu/ummu.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp100.c
index 651b8805c67c..65cb9d28e60e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp100.c
@@ -31,7 +31,7 @@ gp100_mmu = {
 	.dma_bits = 47,
 	.mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}},
 	.mem = {{ -1,  0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map },
-	.vmm = {{ -1, -1, NVIF_CLASS_VMM_GP100}, gp100_vmm_new },
+	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, gp100_vmm_new },
 	.kind = gm200_mmu_kind,
 	.kind_sys = true,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp10b.c
index 3bd3db31e0bb..0a50be9a785a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp10b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gp10b.c
@@ -31,7 +31,7 @@ gp10b_mmu = {
 	.dma_bits = 47,
 	.mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}},
 	.mem = {{ -1, -1, NVIF_CLASS_MEM_GF100}, .umap = gf100_mem_map },
-	.vmm = {{ -1, -1, NVIF_CLASS_VMM_GP100}, gp10b_vmm_new },
+	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, gp10b_vmm_new },
 	.kind = gm200_mmu_kind,
 	.kind_sys = true,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c
index f666cb57f69e..e0997eedd6d9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gv100.c
@@ -31,7 +31,7 @@ gv100_mmu = {
 	.dma_bits = 47,
 	.mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}},
 	.mem = {{ -1,  0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map },
-	.vmm = {{ -1, -1, NVIF_CLASS_VMM_GP100}, gv100_vmm_new },
+	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, gv100_vmm_new },
 	.kind = gm200_mmu_kind,
 	.kind_sys = true,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
index 948a48c21be4..2ad1102a4e31 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
@@ -28,7 +28,7 @@ struct nvkm_mmu_func {
 
 	struct {
 		struct nvkm_sclass user;
-		int (*ctor)(struct nvkm_mmu *, u64 addr, u64 size,
+		int (*ctor)(struct nvkm_mmu *, bool managed, u64 addr, u64 size,
 			    void *argv, u32 argc, struct lock_class_key *,
 			    const char *name, struct nvkm_vmm **);
 		bool global;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c
index 8e6f4096170d..c0db0ce10cba 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c
@@ -27,17 +27,17 @@
 #include <nvif/class.h>
 
 static const struct nvkm_mmu_func
-tu104_mmu = {
+tu102_mmu = {
 	.dma_bits = 47,
 	.mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}},
 	.mem = {{ -1,  0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map },
-	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, tu104_vmm_new },
+	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, tu102_vmm_new },
 	.kind = gm200_mmu_kind,
 	.kind_sys = true,
 };
 
 int
-tu104_mmu_new(struct nvkm_device *device, int index, struct nvkm_mmu **pmmu)
+tu102_mmu_new(struct nvkm_device *device, int index, struct nvkm_mmu **pmmu)
 {
-	return nvkm_mmu_new_(&tu104_mmu, device, index, pmmu);
+	return nvkm_mmu_new_(&tu102_mmu, device, index, pmmu);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c
index 6889076097ec..c43b8248c682 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c
@@ -43,6 +43,69 @@ nvkm_uvmm_search(struct nvkm_client *client, u64 handle)
 }
 
 static int
+nvkm_uvmm_mthd_pfnclr(struct nvkm_uvmm *uvmm, void *argv, u32 argc)
+{
+	struct nvkm_client *client = uvmm->object.client;
+	union {
+		struct nvif_vmm_pfnclr_v0 v0;
+	} *args = argv;
+	struct nvkm_vmm *vmm = uvmm->vmm;
+	int ret = -ENOSYS;
+	u64 addr, size;
+
+	if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) {
+		addr = args->v0.addr;
+		size = args->v0.size;
+	} else
+		return ret;
+
+	if (!client->super)
+		return -ENOENT;
+
+	if (size) {
+		mutex_lock(&vmm->mutex);
+		ret = nvkm_vmm_pfn_unmap(vmm, addr, size);
+		mutex_unlock(&vmm->mutex);
+	}
+
+	return ret;
+}
+
+static int
+nvkm_uvmm_mthd_pfnmap(struct nvkm_uvmm *uvmm, void *argv, u32 argc)
+{
+	struct nvkm_client *client = uvmm->object.client;
+	union {
+		struct nvif_vmm_pfnmap_v0 v0;
+	} *args = argv;
+	struct nvkm_vmm *vmm = uvmm->vmm;
+	int ret = -ENOSYS;
+	u64 addr, size, *phys;
+	u8  page;
+
+	if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, true))) {
+		page = args->v0.page;
+		addr = args->v0.addr;
+		size = args->v0.size;
+		phys = args->v0.phys;
+		if (argc != (size >> page) * sizeof(args->v0.phys[0]))
+			return -EINVAL;
+	} else
+		return ret;
+
+	if (!client->super)
+		return -ENOENT;
+
+	if (size) {
+		mutex_lock(&vmm->mutex);
+		ret = nvkm_vmm_pfn_map(vmm, page, addr, size, phys);
+		mutex_unlock(&vmm->mutex);
+	}
+
+	return ret;
+}
+
+static int
 nvkm_uvmm_mthd_unmap(struct nvkm_uvmm *uvmm, void *argv, u32 argc)
 {
 	struct nvkm_client *client = uvmm->object.client;
@@ -78,7 +141,7 @@ nvkm_uvmm_mthd_unmap(struct nvkm_uvmm *uvmm, void *argv, u32 argc)
 		goto done;
 	}
 
-	nvkm_vmm_unmap_locked(vmm, vma);
+	nvkm_vmm_unmap_locked(vmm, vma, false);
 	ret = 0;
 done:
 	mutex_unlock(&vmm->mutex);
@@ -124,6 +187,11 @@ nvkm_uvmm_mthd_map(struct nvkm_uvmm *uvmm, void *argv, u32 argc)
 		goto fail;
 	}
 
+	if (ret = -EINVAL, vma->mapped && !vma->memory) {
+		VMM_DEBUG(vmm, "pfnmap %016llx", addr);
+		goto fail;
+	}
+
 	if (ret = -EINVAL, vma->addr != addr || vma->size != size) {
 		if (addr + size > vma->addr + vma->size || vma->memory ||
 		    (vma->refd == NVKM_VMA_PAGE_NONE && !vma->mapref)) {
@@ -271,6 +339,15 @@ nvkm_uvmm_mthd(struct nvkm_object *object, u32 mthd, void *argv, u32 argc)
 	case NVIF_VMM_V0_PUT   : return nvkm_uvmm_mthd_put   (uvmm, argv, argc);
 	case NVIF_VMM_V0_MAP   : return nvkm_uvmm_mthd_map   (uvmm, argv, argc);
 	case NVIF_VMM_V0_UNMAP : return nvkm_uvmm_mthd_unmap (uvmm, argv, argc);
+	case NVIF_VMM_V0_PFNMAP: return nvkm_uvmm_mthd_pfnmap(uvmm, argv, argc);
+	case NVIF_VMM_V0_PFNCLR: return nvkm_uvmm_mthd_pfnclr(uvmm, argv, argc);
+	case NVIF_VMM_V0_MTHD(0x00) ... NVIF_VMM_V0_MTHD(0x7f):
+		if (uvmm->vmm->func->mthd) {
+			return uvmm->vmm->func->mthd(uvmm->vmm,
+						     uvmm->object.client,
+						     mthd, argv, argc);
+		}
+		break;
 	default:
 		break;
 	}
@@ -304,8 +381,10 @@ nvkm_uvmm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc,
 	struct nvkm_uvmm *uvmm;
 	int ret = -ENOSYS;
 	u64 addr, size;
+	bool managed;
 
 	if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, more))) {
+		managed = args->v0.managed != 0;
 		addr = args->v0.addr;
 		size = args->v0.size;
 	} else
@@ -317,7 +396,7 @@ nvkm_uvmm_new(const struct nvkm_oclass *oclass, void *argv, u32 argc,
 	*pobject = &uvmm->object;
 
 	if (!mmu->vmm) {
-		ret = mmu->func->vmm.ctor(mmu, addr, size, argv, argc,
+		ret = mmu->func->vmm.ctor(mmu, managed, addr, size, argv, argc,
 					  NULL, "user", &uvmm->vmm);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
index 6b87fff014b3..fa93f964e6a4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
@@ -255,11 +255,23 @@ nvkm_vmm_unref_sptes(struct nvkm_vmm_iter *it, struct nvkm_vmm_pt *pgt,
 }
 
 static bool
-nvkm_vmm_unref_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
+nvkm_vmm_unref_ptes(struct nvkm_vmm_iter *it, bool pfn, u32 ptei, u32 ptes)
 {
 	const struct nvkm_vmm_desc *desc = it->desc;
 	const int type = desc->type == SPT;
 	struct nvkm_vmm_pt *pgt = it->pt[0];
+	bool dma;
+
+	if (pfn) {
+		/* Need to clear PTE valid bits before we dma_unmap_page(). */
+		dma = desc->func->pfn_clear(it->vmm, pgt->pt[type], ptei, ptes);
+		if (dma) {
+			/* GPU may have cached the PT, flush before unmap. */
+			nvkm_vmm_flush_mark(it);
+			nvkm_vmm_flush(it);
+			desc->func->pfn_unmap(it->vmm, pgt->pt[type], ptei, ptes);
+		}
+	}
 
 	/* Drop PTE references. */
 	pgt->refs[type] -= ptes;
@@ -349,7 +361,7 @@ nvkm_vmm_ref_sptes(struct nvkm_vmm_iter *it, struct nvkm_vmm_pt *pgt,
 }
 
 static bool
-nvkm_vmm_ref_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
+nvkm_vmm_ref_ptes(struct nvkm_vmm_iter *it, bool pfn, u32 ptei, u32 ptes)
 {
 	const struct nvkm_vmm_desc *desc = it->desc;
 	const int type = desc->type == SPT;
@@ -379,7 +391,7 @@ nvkm_vmm_sparse_ptes(const struct nvkm_vmm_desc *desc,
 }
 
 static bool
-nvkm_vmm_sparse_unref_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
+nvkm_vmm_sparse_unref_ptes(struct nvkm_vmm_iter *it, bool pfn, u32 ptei, u32 ptes)
 {
 	struct nvkm_vmm_pt *pt = it->pt[0];
 	if (it->desc->type == PGD)
@@ -387,14 +399,14 @@ nvkm_vmm_sparse_unref_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
 	else
 	if (it->desc->type == LPT)
 		memset(&pt->pte[ptei], 0x00, sizeof(pt->pte[0]) * ptes);
-	return nvkm_vmm_unref_ptes(it, ptei, ptes);
+	return nvkm_vmm_unref_ptes(it, pfn, ptei, ptes);
 }
 
 static bool
-nvkm_vmm_sparse_ref_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
+nvkm_vmm_sparse_ref_ptes(struct nvkm_vmm_iter *it, bool pfn, u32 ptei, u32 ptes)
 {
 	nvkm_vmm_sparse_ptes(it->desc, it->pt[0], ptei, ptes);
-	return nvkm_vmm_ref_ptes(it, ptei, ptes);
+	return nvkm_vmm_ref_ptes(it, pfn, ptei, ptes);
 }
 
 static bool
@@ -487,8 +499,8 @@ nvkm_vmm_ref_swpt(struct nvkm_vmm_iter *it, struct nvkm_vmm_pt *pgd, u32 pdei)
 
 static inline u64
 nvkm_vmm_iter(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
-	      u64 addr, u64 size, const char *name, bool ref,
-	      bool (*REF_PTES)(struct nvkm_vmm_iter *, u32, u32),
+	      u64 addr, u64 size, const char *name, bool ref, bool pfn,
+	      bool (*REF_PTES)(struct nvkm_vmm_iter *, bool pfn, u32, u32),
 	      nvkm_vmm_pte_func MAP_PTES, struct nvkm_vmm_map *map,
 	      nvkm_vmm_pxe_func CLR_PTES)
 {
@@ -548,7 +560,7 @@ nvkm_vmm_iter(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 		}
 
 		/* Handle PTE updates. */
-		if (!REF_PTES || REF_PTES(&it, ptei, ptes)) {
+		if (!REF_PTES || REF_PTES(&it, pfn, ptei, ptes)) {
 			struct nvkm_mmu_pt *pt = pgt->pt[type];
 			if (MAP_PTES || CLR_PTES) {
 				if (MAP_PTES)
@@ -590,7 +602,7 @@ static void
 nvkm_vmm_ptes_sparse_put(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 			 u64 addr, u64 size)
 {
-	nvkm_vmm_iter(vmm, page, addr, size, "sparse unref", false,
+	nvkm_vmm_iter(vmm, page, addr, size, "sparse unref", false, false,
 		      nvkm_vmm_sparse_unref_ptes, NULL, NULL,
 		      page->desc->func->invalid ?
 		      page->desc->func->invalid : page->desc->func->unmap);
@@ -602,8 +614,8 @@ nvkm_vmm_ptes_sparse_get(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 {
 	if ((page->type & NVKM_VMM_PAGE_SPARSE)) {
 		u64 fail = nvkm_vmm_iter(vmm, page, addr, size, "sparse ref",
-					 true, nvkm_vmm_sparse_ref_ptes, NULL,
-					 NULL, page->desc->func->sparse);
+					 true, false, nvkm_vmm_sparse_ref_ptes,
+					 NULL, NULL, page->desc->func->sparse);
 		if (fail != ~0ULL) {
 			if ((size = fail - addr))
 				nvkm_vmm_ptes_sparse_put(vmm, page, addr, size);
@@ -666,11 +678,11 @@ nvkm_vmm_ptes_sparse(struct nvkm_vmm *vmm, u64 addr, u64 size, bool ref)
 
 static void
 nvkm_vmm_ptes_unmap_put(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
-			u64 addr, u64 size, bool sparse)
+			u64 addr, u64 size, bool sparse, bool pfn)
 {
 	const struct nvkm_vmm_desc_func *func = page->desc->func;
 	nvkm_vmm_iter(vmm, page, addr, size, "unmap + unref",
-		      false, nvkm_vmm_unref_ptes, NULL, NULL,
+		      false, pfn, nvkm_vmm_unref_ptes, NULL, NULL,
 		      sparse ? func->sparse : func->invalid ? func->invalid :
 							      func->unmap);
 }
@@ -681,10 +693,10 @@ nvkm_vmm_ptes_get_map(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 		      nvkm_vmm_pte_func func)
 {
 	u64 fail = nvkm_vmm_iter(vmm, page, addr, size, "ref + map", true,
-				 nvkm_vmm_ref_ptes, func, map, NULL);
+				 false, nvkm_vmm_ref_ptes, func, map, NULL);
 	if (fail != ~0ULL) {
 		if ((size = fail - addr))
-			nvkm_vmm_ptes_unmap_put(vmm, page, addr, size, false);
+			nvkm_vmm_ptes_unmap_put(vmm, page, addr, size, false, false);
 		return -ENOMEM;
 	}
 	return 0;
@@ -692,10 +704,11 @@ nvkm_vmm_ptes_get_map(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 
 static void
 nvkm_vmm_ptes_unmap(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
-		    u64 addr, u64 size, bool sparse)
+		    u64 addr, u64 size, bool sparse, bool pfn)
 {
 	const struct nvkm_vmm_desc_func *func = page->desc->func;
-	nvkm_vmm_iter(vmm, page, addr, size, "unmap", false, NULL, NULL, NULL,
+	nvkm_vmm_iter(vmm, page, addr, size, "unmap", false, pfn,
+		      NULL, NULL, NULL,
 		      sparse ? func->sparse : func->invalid ? func->invalid :
 							      func->unmap);
 }
@@ -705,7 +718,7 @@ nvkm_vmm_ptes_map(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 		  u64 addr, u64 size, struct nvkm_vmm_map *map,
 		  nvkm_vmm_pte_func func)
 {
-	nvkm_vmm_iter(vmm, page, addr, size, "map", false,
+	nvkm_vmm_iter(vmm, page, addr, size, "map", false, false,
 		      NULL, func, map, NULL);
 }
 
@@ -713,7 +726,7 @@ static void
 nvkm_vmm_ptes_put(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 		  u64 addr, u64 size)
 {
-	nvkm_vmm_iter(vmm, page, addr, size, "unref", false,
+	nvkm_vmm_iter(vmm, page, addr, size, "unref", false, false,
 		      nvkm_vmm_unref_ptes, NULL, NULL, NULL);
 }
 
@@ -721,7 +734,7 @@ static int
 nvkm_vmm_ptes_get(struct nvkm_vmm *vmm, const struct nvkm_vmm_page *page,
 		  u64 addr, u64 size)
 {
-	u64 fail = nvkm_vmm_iter(vmm, page, addr, size, "ref", true,
+	u64 fail = nvkm_vmm_iter(vmm, page, addr, size, "ref", true, false,
 				 nvkm_vmm_ref_ptes, NULL, NULL, NULL);
 	if (fail != ~0ULL) {
 		if (fail != addr)
@@ -763,6 +776,7 @@ nvkm_vma_tail(struct nvkm_vma *vma, u64 tail)
 	new->part = vma->part;
 	new->user = vma->user;
 	new->busy = vma->busy;
+	new->mapped = vma->mapped;
 	list_add(&new->head, &vma->head);
 	return new;
 }
@@ -935,11 +949,40 @@ nvkm_vmm_node_split(struct nvkm_vmm *vmm,
 }
 
 static void
+nvkm_vma_dump(struct nvkm_vma *vma)
+{
+	printk(KERN_ERR "%016llx %016llx %c%c%c%c%c%c%c%c%c %p\n",
+	       vma->addr, (u64)vma->size,
+	       vma->used ? '-' : 'F',
+	       vma->mapref ? 'R' : '-',
+	       vma->sparse ? 'S' : '-',
+	       vma->page != NVKM_VMA_PAGE_NONE ? '0' + vma->page : '-',
+	       vma->refd != NVKM_VMA_PAGE_NONE ? '0' + vma->refd : '-',
+	       vma->part ? 'P' : '-',
+	       vma->user ? 'U' : '-',
+	       vma->busy ? 'B' : '-',
+	       vma->mapped ? 'M' : '-',
+	       vma->memory);
+}
+
+static void
+nvkm_vmm_dump(struct nvkm_vmm *vmm)
+{
+	struct nvkm_vma *vma;
+	list_for_each_entry(vma, &vmm->list, head) {
+		nvkm_vma_dump(vma);
+	}
+}
+
+static void
 nvkm_vmm_dtor(struct nvkm_vmm *vmm)
 {
 	struct nvkm_vma *vma;
 	struct rb_node *node;
 
+	if (0)
+		nvkm_vmm_dump(vmm);
+
 	while ((node = rb_first(&vmm->root))) {
 		struct nvkm_vma *vma = rb_entry(node, typeof(*vma), tree);
 		nvkm_vmm_put(vmm, &vma);
@@ -972,16 +1015,32 @@ nvkm_vmm_dtor(struct nvkm_vmm *vmm)
 	}
 }
 
+static int
+nvkm_vmm_ctor_managed(struct nvkm_vmm *vmm, u64 addr, u64 size)
+{
+	struct nvkm_vma *vma;
+	if (!(vma = nvkm_vma_new(addr, size)))
+		return -ENOMEM;
+	vma->mapref = true;
+	vma->sparse = false;
+	vma->used = true;
+	vma->user = true;
+	nvkm_vmm_node_insert(vmm, vma);
+	list_add_tail(&vma->head, &vmm->list);
+	return 0;
+}
+
 int
 nvkm_vmm_ctor(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
-	      u32 pd_header, u64 addr, u64 size, struct lock_class_key *key,
-	      const char *name, struct nvkm_vmm *vmm)
+	      u32 pd_header, bool managed, u64 addr, u64 size,
+	      struct lock_class_key *key, const char *name,
+	      struct nvkm_vmm *vmm)
 {
 	static struct lock_class_key _key;
 	const struct nvkm_vmm_page *page = func->page;
 	const struct nvkm_vmm_desc *desc;
 	struct nvkm_vma *vma;
-	int levels, bits = 0;
+	int levels, bits = 0, ret;
 
 	vmm->func = func;
 	vmm->mmu = mmu;
@@ -1009,11 +1068,6 @@ nvkm_vmm_ctor(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
 	if (WARN_ON(levels > NVKM_VMM_LEVELS_MAX))
 		return -EINVAL;
 
-	vmm->start = addr;
-	vmm->limit = size ? (addr + size) : (1ULL << bits);
-	if (vmm->start > vmm->limit || vmm->limit > (1ULL << bits))
-		return -EINVAL;
-
 	/* Allocate top-level page table. */
 	vmm->pd = nvkm_vmm_pt_new(desc, false, NULL);
 	if (!vmm->pd)
@@ -1036,50 +1090,273 @@ nvkm_vmm_ctor(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
 	vmm->free = RB_ROOT;
 	vmm->root = RB_ROOT;
 
-	if (!(vma = nvkm_vma_new(vmm->start, vmm->limit - vmm->start)))
-		return -ENOMEM;
+	if (managed) {
+		/* Address-space will be managed by the client for the most
+		 * part, except for a specified area where NVKM allocations
+		 * are allowed to be placed.
+		 */
+		vmm->start = 0;
+		vmm->limit = 1ULL << bits;
+		if (addr + size < addr || addr + size > vmm->limit)
+			return -EINVAL;
+
+		/* Client-managed area before the NVKM-managed area. */
+		if (addr && (ret = nvkm_vmm_ctor_managed(vmm, 0, addr)))
+			return ret;
+
+		/* NVKM-managed area. */
+		if (size) {
+			if (!(vma = nvkm_vma_new(addr, size)))
+				return -ENOMEM;
+			nvkm_vmm_free_insert(vmm, vma);
+			list_add_tail(&vma->head, &vmm->list);
+		}
+
+		/* Client-managed area after the NVKM-managed area. */
+		addr = addr + size;
+		size = vmm->limit - addr;
+		if (size && (ret = nvkm_vmm_ctor_managed(vmm, addr, size)))
+			return ret;
+	} else {
+		/* Address-space fully managed by NVKM, requiring calls to
+		 * nvkm_vmm_get()/nvkm_vmm_put() to allocate address-space.
+		 */
+		vmm->start = addr;
+		vmm->limit = size ? (addr + size) : (1ULL << bits);
+		if (vmm->start > vmm->limit || vmm->limit > (1ULL << bits))
+			return -EINVAL;
+
+		if (!(vma = nvkm_vma_new(vmm->start, vmm->limit - vmm->start)))
+			return -ENOMEM;
+
+		nvkm_vmm_free_insert(vmm, vma);
+		list_add(&vma->head, &vmm->list);
+	}
 
-	nvkm_vmm_free_insert(vmm, vma);
-	list_add(&vma->head, &vmm->list);
 	return 0;
 }
 
 int
 nvkm_vmm_new_(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
-	      u32 hdr, u64 addr, u64 size, struct lock_class_key *key,
-	      const char *name, struct nvkm_vmm **pvmm)
+	      u32 hdr, bool managed, u64 addr, u64 size,
+	      struct lock_class_key *key, const char *name,
+	      struct nvkm_vmm **pvmm)
 {
 	if (!(*pvmm = kzalloc(sizeof(**pvmm), GFP_KERNEL)))
 		return -ENOMEM;
-	return nvkm_vmm_ctor(func, mmu, hdr, addr, size, key, name, *pvmm);
+	return nvkm_vmm_ctor(func, mmu, hdr, managed, addr, size, key, name, *pvmm);
+}
+
+static struct nvkm_vma *
+nvkm_vmm_pfn_split_merge(struct nvkm_vmm *vmm, struct nvkm_vma *vma,
+			 u64 addr, u64 size, u8 page, bool map)
+{
+	struct nvkm_vma *prev = NULL;
+	struct nvkm_vma *next = NULL;
+
+	if (vma->addr == addr && vma->part && (prev = node(vma, prev))) {
+		if (prev->memory || prev->mapped != map)
+			prev = NULL;
+	}
+
+	if (vma->addr + vma->size == addr + size && (next = node(vma, next))) {
+		if (!next->part ||
+		    next->memory || next->mapped != map)
+			next = NULL;
+	}
+
+	if (prev || next)
+		return nvkm_vmm_node_merge(vmm, prev, vma, next, size);
+	return nvkm_vmm_node_split(vmm, vma, addr, size);
+}
+
+int
+nvkm_vmm_pfn_unmap(struct nvkm_vmm *vmm, u64 addr, u64 size)
+{
+	struct nvkm_vma *vma = nvkm_vmm_node_search(vmm, addr);
+	struct nvkm_vma *next;
+	u64 limit = addr + size;
+	u64 start = addr;
+
+	if (!vma)
+		return -EINVAL;
+
+	do {
+		if (!vma->mapped || vma->memory)
+			continue;
+
+		size = min(limit - start, vma->size - (start - vma->addr));
+
+		nvkm_vmm_ptes_unmap_put(vmm, &vmm->func->page[vma->refd],
+					start, size, false, true);
+
+		next = nvkm_vmm_pfn_split_merge(vmm, vma, start, size, 0, false);
+		if (!WARN_ON(!next)) {
+			vma = next;
+			vma->refd = NVKM_VMA_PAGE_NONE;
+			vma->mapped = false;
+		}
+	} while ((vma = node(vma, next)) && (start = vma->addr) < limit);
+
+	return 0;
+}
+
+/*TODO:
+ * - Avoid PT readback (for dma_unmap etc), this might end up being dealt
+ *   with inside HMM, which would be a lot nicer for us to deal with.
+ * - Multiple page sizes (particularly for huge page support).
+ * - Support for systems without a 4KiB page size.
+ */
+int
+nvkm_vmm_pfn_map(struct nvkm_vmm *vmm, u8 shift, u64 addr, u64 size, u64 *pfn)
+{
+	const struct nvkm_vmm_page *page = vmm->func->page;
+	struct nvkm_vma *vma, *tmp;
+	u64 limit = addr + size;
+	u64 start = addr;
+	int pm = size >> shift;
+	int pi = 0;
+
+	/* Only support mapping where the page size of the incoming page
+	 * array matches a page size available for direct mapping.
+	 */
+	while (page->shift && page->shift != shift &&
+	       page->desc->func->pfn == NULL)
+		page++;
+
+	if (!page->shift || !IS_ALIGNED(addr, 1ULL << shift) ||
+			    !IS_ALIGNED(size, 1ULL << shift) ||
+	    addr + size < addr || addr + size > vmm->limit) {
+		VMM_DEBUG(vmm, "paged map %d %d %016llx %016llx\n",
+			  shift, page->shift, addr, size);
+		return -EINVAL;
+	}
+
+	if (!(vma = nvkm_vmm_node_search(vmm, addr)))
+		return -ENOENT;
+
+	do {
+		bool map = !!(pfn[pi] & NVKM_VMM_PFN_V);
+		bool mapped = vma->mapped;
+		u64 size = limit - start;
+		u64 addr = start;
+		int pn, ret = 0;
+
+		/* Narrow the operation window to cover a single action (page
+		 * should be mapped or not) within a single VMA.
+		 */
+		for (pn = 0; pi + pn < pm; pn++) {
+			if (map != !!(pfn[pi + pn] & NVKM_VMM_PFN_V))
+				break;
+		}
+		size = min_t(u64, size, pn << page->shift);
+		size = min_t(u64, size, vma->size + vma->addr - addr);
+
+		/* Reject any operation to unmanaged regions, and areas that
+		 * have nvkm_memory objects mapped in them already.
+		 */
+		if (!vma->mapref || vma->memory) {
+			ret = -EINVAL;
+			goto next;
+		}
+
+		/* In order to both properly refcount GPU page tables, and
+		 * prevent "normal" mappings and these direct mappings from
+		 * interfering with each other, we need to track contiguous
+		 * ranges that have been mapped with this interface.
+		 *
+		 * Here we attempt to either split an existing VMA so we're
+		 * able to flag the region as either unmapped/mapped, or to
+		 * merge with adjacent VMAs that are already compatible.
+		 *
+		 * If the region is already compatible, nothing is required.
+		 */
+		if (map != mapped) {
+			tmp = nvkm_vmm_pfn_split_merge(vmm, vma, addr, size,
+						       page -
+						       vmm->func->page, map);
+			if (WARN_ON(!tmp)) {
+				ret = -ENOMEM;
+				goto next;
+			}
+
+			if ((tmp->mapped = map))
+				tmp->refd = page - vmm->func->page;
+			else
+				tmp->refd = NVKM_VMA_PAGE_NONE;
+			vma = tmp;
+		}
+
+		/* Update HW page tables. */
+		if (map) {
+			struct nvkm_vmm_map args;
+			args.page = page;
+			args.pfn = &pfn[pi];
+
+			if (!mapped) {
+				ret = nvkm_vmm_ptes_get_map(vmm, page, addr,
+							    size, &args, page->
+							    desc->func->pfn);
+			} else {
+				nvkm_vmm_ptes_map(vmm, page, addr, size, &args,
+						  page->desc->func->pfn);
+			}
+		} else {
+			if (mapped) {
+				nvkm_vmm_ptes_unmap_put(vmm, page, addr, size,
+							false, true);
+			}
+		}
+
+next:
+		/* Iterate to next operation. */
+		if (vma->addr + vma->size == addr + size)
+			vma = node(vma, next);
+		start += size;
+
+		if (ret) {
+			/* Failure is signalled by clearing the valid bit on
+			 * any PFN that couldn't be modified as requested.
+			 */
+			while (size) {
+				pfn[pi++] = NVKM_VMM_PFN_NONE;
+				size -= 1 << page->shift;
+			}
+		} else {
+			pi += size >> page->shift;
+		}
+	} while (vma && start < limit);
+
+	return 0;
 }
 
 void
 nvkm_vmm_unmap_region(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
 {
-	struct nvkm_vma *next = node(vma, next);
 	struct nvkm_vma *prev = NULL;
+	struct nvkm_vma *next;
 
 	nvkm_memory_tags_put(vma->memory, vmm->mmu->subdev.device, &vma->tags);
 	nvkm_memory_unref(&vma->memory);
+	vma->mapped = false;
 
-	if (!vma->part || ((prev = node(vma, prev)), prev->memory))
+	if (vma->part && (prev = node(vma, prev)) && prev->mapped)
 		prev = NULL;
-	if (!next->part || next->memory)
+	if ((next = node(vma, next)) && (!next->part || next->mapped))
 		next = NULL;
 	nvkm_vmm_node_merge(vmm, prev, vma, next, vma->size);
 }
 
 void
-nvkm_vmm_unmap_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
+nvkm_vmm_unmap_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma, bool pfn)
 {
 	const struct nvkm_vmm_page *page = &vmm->func->page[vma->refd];
 
 	if (vma->mapref) {
-		nvkm_vmm_ptes_unmap_put(vmm, page, vma->addr, vma->size, vma->sparse);
+		nvkm_vmm_ptes_unmap_put(vmm, page, vma->addr, vma->size, vma->sparse, pfn);
 		vma->refd = NVKM_VMA_PAGE_NONE;
 	} else {
-		nvkm_vmm_ptes_unmap(vmm, page, vma->addr, vma->size, vma->sparse);
+		nvkm_vmm_ptes_unmap(vmm, page, vma->addr, vma->size, vma->sparse, pfn);
 	}
 
 	nvkm_vmm_unmap_region(vmm, vma);
@@ -1090,7 +1367,7 @@ nvkm_vmm_unmap(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
 {
 	if (vma->memory) {
 		mutex_lock(&vmm->mutex);
-		nvkm_vmm_unmap_locked(vmm, vma);
+		nvkm_vmm_unmap_locked(vmm, vma, false);
 		mutex_unlock(&vmm->mutex);
 	}
 }
@@ -1224,6 +1501,7 @@ nvkm_vmm_map_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma,
 	nvkm_memory_tags_put(vma->memory, vmm->mmu->subdev.device, &vma->tags);
 	nvkm_memory_unref(&vma->memory);
 	vma->memory = nvkm_memory_ref(map->memory);
+	vma->mapped = true;
 	vma->tags = map->tags;
 	return 0;
 }
@@ -1269,14 +1547,16 @@ nvkm_vmm_put_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
 
 	if (vma->mapref || !vma->sparse) {
 		do {
-			const bool map = next->memory != NULL;
+			const bool mem = next->memory != NULL;
+			const bool map = next->mapped;
 			const u8  refd = next->refd;
 			const u64 addr = next->addr;
 			u64 size = next->size;
 
 			/* Merge regions that are in the same state. */
 			while ((next = node(next, next)) && next->part &&
-			       (next->memory != NULL) == map &&
+			       (next->mapped == map) &&
+			       (next->memory != NULL) == mem &&
 			       (next->refd == refd))
 				size += next->size;
 
@@ -1286,7 +1566,8 @@ nvkm_vmm_put_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
 				 * the page tree.
 				 */
 				nvkm_vmm_ptes_unmap_put(vmm, &page[refd], addr,
-							size, vma->sparse);
+							size, vma->sparse,
+							!mem);
 			} else
 			if (refd != NVKM_VMA_PAGE_NONE) {
 				/* Drop allocation-time PTE references. */
@@ -1301,7 +1582,7 @@ nvkm_vmm_put_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma)
 	 */
 	next = vma;
 	do {
-		if (next->memory)
+		if (next->mapped)
 			nvkm_vmm_unmap_region(vmm, next);
 	} while ((next = node(vma, next)) && next->part);
 
@@ -1522,7 +1803,7 @@ nvkm_vmm_join(struct nvkm_vmm *vmm, struct nvkm_memory *inst)
 }
 
 static bool
-nvkm_vmm_boot_ptes(struct nvkm_vmm_iter *it, u32 ptei, u32 ptes)
+nvkm_vmm_boot_ptes(struct nvkm_vmm_iter *it, bool pfn, u32 ptei, u32 ptes)
 {
 	const struct nvkm_vmm_desc *desc = it->desc;
 	const int type = desc->type == SPT;
@@ -1544,7 +1825,7 @@ nvkm_vmm_boot(struct nvkm_vmm *vmm)
 	if (ret)
 		return ret;
 
-	nvkm_vmm_iter(vmm, page, vmm->start, limit, "bootstrap", false,
+	nvkm_vmm_iter(vmm, page, vmm->start, limit, "bootstrap", false, false,
 		      nvkm_vmm_boot_ptes, NULL, NULL, NULL);
 	vmm->bootstrapped = true;
 	return 0;
@@ -1584,7 +1865,8 @@ nvkm_vmm_new(struct nvkm_device *device, u64 addr, u64 size, void *argv,
 	struct nvkm_mmu *mmu = device->mmu;
 	struct nvkm_vmm *vmm = NULL;
 	int ret;
-	ret = mmu->func->vmm.ctor(mmu, addr, size, argv, argc, key, name, &vmm);
+	ret = mmu->func->vmm.ctor(mmu, false, addr, size, argv, argc,
+				  key, name, &vmm);
 	if (ret)
 		nvkm_vmm_unref(&vmm);
 	*pvmm = vmm;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
index 42ad326521a3..5e55ecbd8005 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
@@ -67,6 +67,10 @@ struct nvkm_vmm_desc_func {
 	nvkm_vmm_pte_func mem;
 	nvkm_vmm_pte_func dma;
 	nvkm_vmm_pte_func sgl;
+
+	nvkm_vmm_pte_func pfn;
+	bool (*pfn_clear)(struct nvkm_vmm *, struct nvkm_mmu_pt *, u32 ptei, u32 ptes);
+	nvkm_vmm_pxe_func pfn_unmap;
 };
 
 extern const struct nvkm_vmm_desc_func gf100_vmm_pgd;
@@ -141,6 +145,11 @@ struct nvkm_vmm_func {
 		     struct nvkm_vmm_map *);
 	void (*flush)(struct nvkm_vmm *, int depth);
 
+	int (*mthd)(struct nvkm_vmm *, struct nvkm_client *,
+		    u32 mthd, void *argv, u32 argc);
+
+	void (*invalidate_pdb)(struct nvkm_vmm *, u64 addr);
+
 	u64 page_block;
 	const struct nvkm_vmm_page page[];
 };
@@ -151,11 +160,12 @@ struct nvkm_vmm_join {
 };
 
 int nvkm_vmm_new_(const struct nvkm_vmm_func *, struct nvkm_mmu *,
-		  u32 pd_header, u64 addr, u64 size, struct lock_class_key *,
-		  const char *name, struct nvkm_vmm **);
+		  u32 pd_header, bool managed, u64 addr, u64 size,
+		  struct lock_class_key *, const char *name,
+		  struct nvkm_vmm **);
 int nvkm_vmm_ctor(const struct nvkm_vmm_func *, struct nvkm_mmu *,
-		  u32 pd_header, u64 addr, u64 size, struct lock_class_key *,
-		  const char *name, struct nvkm_vmm *);
+		  u32 pd_header, bool managed, u64 addr, u64 size,
+		  struct lock_class_key *, const char *name, struct nvkm_vmm *);
 struct nvkm_vma *nvkm_vmm_node_search(struct nvkm_vmm *, u64 addr);
 struct nvkm_vma *nvkm_vmm_node_split(struct nvkm_vmm *, struct nvkm_vma *,
 				     u64 addr, u64 size);
@@ -163,13 +173,25 @@ int nvkm_vmm_get_locked(struct nvkm_vmm *, bool getref, bool mapref,
 			bool sparse, u8 page, u8 align, u64 size,
 			struct nvkm_vma **pvma);
 void nvkm_vmm_put_locked(struct nvkm_vmm *, struct nvkm_vma *);
-void nvkm_vmm_unmap_locked(struct nvkm_vmm *, struct nvkm_vma *);
-void nvkm_vmm_unmap_region(struct nvkm_vmm *vmm, struct nvkm_vma *vma);
+void nvkm_vmm_unmap_locked(struct nvkm_vmm *, struct nvkm_vma *, bool pfn);
+void nvkm_vmm_unmap_region(struct nvkm_vmm *, struct nvkm_vma *);
+
+#define NVKM_VMM_PFN_ADDR                                 0xfffffffffffff000ULL
+#define NVKM_VMM_PFN_ADDR_SHIFT                                              12
+#define NVKM_VMM_PFN_APER                                 0x00000000000000f0ULL
+#define NVKM_VMM_PFN_HOST                                 0x0000000000000000ULL
+#define NVKM_VMM_PFN_VRAM                                 0x0000000000000010ULL
+#define NVKM_VMM_PFN_W                                    0x0000000000000002ULL
+#define NVKM_VMM_PFN_V                                    0x0000000000000001ULL
+#define NVKM_VMM_PFN_NONE                                 0x0000000000000000ULL
+
+int nvkm_vmm_pfn_map(struct nvkm_vmm *, u8 page, u64 addr, u64 size, u64 *pfn);
+int nvkm_vmm_pfn_unmap(struct nvkm_vmm *, u64 addr, u64 size);
 
 struct nvkm_vma *nvkm_vma_tail(struct nvkm_vma *, u64 tail);
 
 int nv04_vmm_new_(const struct nvkm_vmm_func *, struct nvkm_mmu *, u32,
-		  u64, u64, void *, u32, struct lock_class_key *,
+		  bool, u64, u64, void *, u32, struct lock_class_key *,
 		  const char *, struct nvkm_vmm **);
 int nv04_vmm_valid(struct nvkm_vmm *, void *, u32, struct nvkm_vmm_map *);
 
@@ -179,70 +201,76 @@ int nv50_vmm_valid(struct nvkm_vmm *, void *, u32, struct nvkm_vmm_map *);
 void nv50_vmm_flush(struct nvkm_vmm *, int);
 
 int gf100_vmm_new_(const struct nvkm_vmm_func *, const struct nvkm_vmm_func *,
-		   struct nvkm_mmu *, u64, u64, void *, u32,
+		   struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		   struct lock_class_key *, const char *, struct nvkm_vmm **);
 int gf100_vmm_join_(struct nvkm_vmm *, struct nvkm_memory *, u64 base);
 int gf100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *);
 void gf100_vmm_part(struct nvkm_vmm *, struct nvkm_memory *);
 int gf100_vmm_aper(enum nvkm_memory_target);
 int gf100_vmm_valid(struct nvkm_vmm *, void *, u32, struct nvkm_vmm_map *);
-void gf100_vmm_flush_(struct nvkm_vmm *, int);
 void gf100_vmm_flush(struct nvkm_vmm *, int);
+void gf100_vmm_invalidate(struct nvkm_vmm *, u32 type);
+void gf100_vmm_invalidate_pdb(struct nvkm_vmm *, u64 addr);
 
 int gk20a_vmm_aper(enum nvkm_memory_target);
 
 int gm200_vmm_new_(const struct nvkm_vmm_func *, const struct nvkm_vmm_func *,
-		   struct nvkm_mmu *, u64, u64, void *, u32,
+		   struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		   struct lock_class_key *, const char *, struct nvkm_vmm **);
 int gm200_vmm_join_(struct nvkm_vmm *, struct nvkm_memory *, u64 base);
 int gm200_vmm_join(struct nvkm_vmm *, struct nvkm_memory *);
 
+int gp100_vmm_new_(const struct nvkm_vmm_func *,
+		   struct nvkm_mmu *, bool, u64, u64, void *, u32,
+		   struct lock_class_key *, const char *, struct nvkm_vmm **);
 int gp100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *);
 int gp100_vmm_valid(struct nvkm_vmm *, void *, u32, struct nvkm_vmm_map *);
 void gp100_vmm_flush(struct nvkm_vmm *, int);
+int gp100_vmm_mthd(struct nvkm_vmm *, struct nvkm_client *, u32, void *, u32);
+void gp100_vmm_invalidate_pdb(struct nvkm_vmm *, u64 addr);
 
 int gv100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *);
 
-int nv04_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int nv04_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		 struct lock_class_key *, const char *, struct nvkm_vmm **);
-int nv41_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int nv41_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		 struct lock_class_key *, const char *, struct nvkm_vmm **);
-int nv44_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int nv44_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		 struct lock_class_key *, const char *, struct nvkm_vmm **);
-int nv50_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int nv50_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		 struct lock_class_key *, const char *, struct nvkm_vmm **);
-int mcp77_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int mcp77_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *, struct nvkm_vmm **);
-int g84_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int g84_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		struct lock_class_key *, const char *, struct nvkm_vmm **);
-int gf100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gf100_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *, struct nvkm_vmm **);
-int gk104_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gk104_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *, struct nvkm_vmm **);
-int gk20a_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gk20a_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *, struct nvkm_vmm **);
-int gm200_vmm_new_fixed(struct nvkm_mmu *, u64, u64, void *, u32,
+int gm200_vmm_new_fixed(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 			struct lock_class_key *, const char *,
 			struct nvkm_vmm **);
-int gm200_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gm200_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
-int gm20b_vmm_new_fixed(struct nvkm_mmu *, u64, u64, void *, u32,
+int gm20b_vmm_new_fixed(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 			struct lock_class_key *, const char *,
 			struct nvkm_vmm **);
-int gm20b_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gm20b_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
-int gp100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gp100_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
-int gp10b_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gp10b_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
-int gv100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int gv100_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
-int tu104_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32,
+int tu102_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgf100.c
index faf5a7e9265e..ab6424faf84c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgf100.c
@@ -178,15 +178,19 @@ gf100_vmm_desc_16_16[] = {
 };
 
 void
-gf100_vmm_flush_(struct nvkm_vmm *vmm, int depth)
+gf100_vmm_invalidate_pdb(struct nvkm_vmm *vmm, u64 addr)
+{
+	struct nvkm_device *device = vmm->mmu->subdev.device;
+	nvkm_wr32(device, 0x100cb8, addr);
+}
+
+void
+gf100_vmm_invalidate(struct nvkm_vmm *vmm, u32 type)
 {
 	struct nvkm_subdev *subdev = &vmm->mmu->subdev;
 	struct nvkm_device *device = subdev->device;
-	u32 type = depth << 24;
-
-	type = 0x00000001; /* PAGE_ALL */
-	if (atomic_read(&vmm->engref[NVKM_SUBDEV_BAR]))
-		type |= 0x00000004; /* HUB_ONLY */
+	struct nvkm_mmu_pt *pd = vmm->pd->pt[0];
+	u64 addr = 0;
 
 	mutex_lock(&subdev->mutex);
 	/* Looks like maybe a "free flush slots" counter, the
@@ -197,7 +201,20 @@ gf100_vmm_flush_(struct nvkm_vmm *vmm, int depth)
 			break;
 	);
 
-	nvkm_wr32(device, 0x100cb8, vmm->pd->pt[0]->addr >> 8);
+	if (!(type & 0x00000002) /* ALL_PDB. */) {
+		switch (nvkm_memory_target(pd->memory)) {
+		case NVKM_MEM_TARGET_VRAM: addr |= 0x00000000; break;
+		case NVKM_MEM_TARGET_HOST: addr |= 0x00000002; break;
+		case NVKM_MEM_TARGET_NCOH: addr |= 0x00000003; break;
+		default:
+			WARN_ON(1);
+			break;
+		}
+		addr |= (vmm->pd->pt[0]->addr >> 12) << 4;
+
+		vmm->func->invalidate_pdb(vmm, addr);
+	}
+
 	nvkm_wr32(device, 0x100cbc, 0x80000000 | type);
 
 	/* Wait for flush to be queued? */
@@ -211,7 +228,10 @@ gf100_vmm_flush_(struct nvkm_vmm *vmm, int depth)
 void
 gf100_vmm_flush(struct nvkm_vmm *vmm, int depth)
 {
-	gf100_vmm_flush_(vmm, 0);
+	u32 type = 0x00000001; /* PAGE_ALL */
+	if (atomic_read(&vmm->engref[NVKM_SUBDEV_BAR]))
+		type |= 0x00000004; /* HUB_ONLY */
+	gf100_vmm_invalidate(vmm, type);
 }
 
 int
@@ -354,6 +374,7 @@ gf100_vmm_17 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 17, &gf100_vmm_desc_17_17[0], NVKM_VMM_PAGE_xVxC },
 		{ 12, &gf100_vmm_desc_17_12[0], NVKM_VMM_PAGE_xVHx },
@@ -368,6 +389,7 @@ gf100_vmm_16 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 16, &gf100_vmm_desc_16_16[0], NVKM_VMM_PAGE_xVxC },
 		{ 12, &gf100_vmm_desc_16_12[0], NVKM_VMM_PAGE_xVHx },
@@ -378,14 +400,14 @@ gf100_vmm_16 = {
 int
 gf100_vmm_new_(const struct nvkm_vmm_func *func_16,
 	       const struct nvkm_vmm_func *func_17,
-	       struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	       struct lock_class_key *key, const char *name,
-	       struct nvkm_vmm **pvmm)
+	       struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	       void *argv, u32 argc, struct lock_class_key *key,
+	       const char *name, struct nvkm_vmm **pvmm)
 {
 	switch (mmu->subdev.device->fb->page) {
-	case 16: return nv04_vmm_new_(func_16, mmu, 0, addr, size,
+	case 16: return nv04_vmm_new_(func_16, mmu, 0, managed, addr, size,
 				      argv, argc, key, name, pvmm);
-	case 17: return nv04_vmm_new_(func_17, mmu, 0, addr, size,
+	case 17: return nv04_vmm_new_(func_17, mmu, 0, managed, addr, size,
 				      argv, argc, key, name, pvmm);
 	default:
 		WARN_ON(1);
@@ -394,10 +416,10 @@ gf100_vmm_new_(const struct nvkm_vmm_func *func_16,
 }
 
 int
-gf100_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gf100_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return gf100_vmm_new_(&gf100_vmm_16, &gf100_vmm_17, mmu, addr,
+	return gf100_vmm_new_(&gf100_vmm_16, &gf100_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk104.c
index 0ebb7bccfcd2..0b59c01fd146 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk104.c
@@ -71,6 +71,7 @@ gk104_vmm_17 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 17, &gk104_vmm_desc_17_17[0], NVKM_VMM_PAGE_xVxC },
 		{ 12, &gk104_vmm_desc_17_12[0], NVKM_VMM_PAGE_xVHx },
@@ -85,6 +86,7 @@ gk104_vmm_16 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 16, &gk104_vmm_desc_16_16[0], NVKM_VMM_PAGE_xVxC },
 		{ 12, &gk104_vmm_desc_16_12[0], NVKM_VMM_PAGE_xVHx },
@@ -93,10 +95,10 @@ gk104_vmm_16 = {
 };
 
 int
-gk104_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gk104_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return gf100_vmm_new_(&gk104_vmm_16, &gk104_vmm_17, mmu, addr,
+	return gf100_vmm_new_(&gk104_vmm_16, &gk104_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
index 8086994a0446..5a9582dce970 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
@@ -40,6 +40,7 @@ gk20a_vmm_17 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 17, &gk104_vmm_desc_17_17[0], NVKM_VMM_PAGE_xxHC },
 		{ 12, &gk104_vmm_desc_17_12[0], NVKM_VMM_PAGE_xxHx },
@@ -54,6 +55,7 @@ gk20a_vmm_16 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 16, &gk104_vmm_desc_16_16[0], NVKM_VMM_PAGE_xxHC },
 		{ 12, &gk104_vmm_desc_16_12[0], NVKM_VMM_PAGE_xxHx },
@@ -62,10 +64,10 @@ gk20a_vmm_16 = {
 };
 
 int
-gk20a_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gk20a_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return gf100_vmm_new_(&gk20a_vmm_16, &gk20a_vmm_17, mmu, addr,
+	return gf100_vmm_new_(&gk20a_vmm_16, &gk20a_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm200.c
index a1676a4644fe..2e61af02d4d8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm200.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm200.c
@@ -113,6 +113,7 @@ gm200_vmm_17 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 27, &gm200_vmm_desc_17_17[1], NVKM_VMM_PAGE_Sxxx },
 		{ 17, &gm200_vmm_desc_17_17[0], NVKM_VMM_PAGE_SVxC },
@@ -128,6 +129,7 @@ gm200_vmm_16 = {
 	.aper = gf100_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 27, &gm200_vmm_desc_16_16[1], NVKM_VMM_PAGE_Sxxx },
 		{ 16, &gm200_vmm_desc_16_16[0], NVKM_VMM_PAGE_SVxC },
@@ -139,9 +141,9 @@ gm200_vmm_16 = {
 int
 gm200_vmm_new_(const struct nvkm_vmm_func *func_16,
 	       const struct nvkm_vmm_func *func_17,
-	       struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	       struct lock_class_key *key, const char *name,
-	       struct nvkm_vmm **pvmm)
+	       struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	       void *argv, u32 argc, struct lock_class_key *key,
+	       const char *name, struct nvkm_vmm **pvmm)
 {
 	const struct nvkm_vmm_func *func;
 	union {
@@ -163,23 +165,23 @@ gm200_vmm_new_(const struct nvkm_vmm_func *func_16,
 	} else
 		return ret;
 
-	return nvkm_vmm_new_(func, mmu, 0, addr, size, key, name, pvmm);
+	return nvkm_vmm_new_(func, mmu, 0, managed, addr, size, key, name, pvmm);
 }
 
 int
-gm200_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gm200_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return gm200_vmm_new_(&gm200_vmm_16, &gm200_vmm_17, mmu, addr,
+	return gm200_vmm_new_(&gm200_vmm_16, &gm200_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
 
 int
-gm200_vmm_new_fixed(struct nvkm_mmu *mmu, u64 addr, u64 size,
+gm200_vmm_new_fixed(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
 		    void *argv, u32 argc, struct lock_class_key *key,
 		    const char *name, struct nvkm_vmm **pvmm)
 {
-	return gf100_vmm_new_(&gm200_vmm_16, &gm200_vmm_17, mmu, addr,
+	return gf100_vmm_new_(&gm200_vmm_16, &gm200_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm20b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm20b.c
index 64d4b6cff8dd..96b759695dd8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm20b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgm20b.c
@@ -28,6 +28,7 @@ gm20b_vmm_17 = {
 	.aper = gk20a_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 27, &gm200_vmm_desc_17_17[1], NVKM_VMM_PAGE_Sxxx },
 		{ 17, &gm200_vmm_desc_17_17[0], NVKM_VMM_PAGE_SxHC },
@@ -43,6 +44,7 @@ gm20b_vmm_16 = {
 	.aper = gk20a_vmm_aper,
 	.valid = gf100_vmm_valid,
 	.flush = gf100_vmm_flush,
+	.invalidate_pdb = gf100_vmm_invalidate_pdb,
 	.page = {
 		{ 27, &gm200_vmm_desc_16_16[1], NVKM_VMM_PAGE_Sxxx },
 		{ 16, &gm200_vmm_desc_16_16[0], NVKM_VMM_PAGE_SxHC },
@@ -52,19 +54,19 @@ gm20b_vmm_16 = {
 };
 
 int
-gm20b_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gm20b_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return gm200_vmm_new_(&gm20b_vmm_16, &gm20b_vmm_17, mmu, addr,
+	return gm200_vmm_new_(&gm20b_vmm_16, &gm20b_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
 
 int
-gm20b_vmm_new_fixed(struct nvkm_mmu *mmu, u64 addr, u64 size,
+gm20b_vmm_new_fixed(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
 		    void *argv, u32 argc, struct lock_class_key *key,
 		    const char *name, struct nvkm_vmm **pvmm)
 {
-	return gf100_vmm_new_(&gm20b_vmm_16, &gm20b_vmm_17, mmu, addr,
+	return gf100_vmm_new_(&gm20b_vmm_16, &gm20b_vmm_17, mmu, managed, addr,
 			      size, argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
index 059fafe0e771..b4f519768d5e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
@@ -21,12 +21,90 @@
  */
 #include "vmm.h"
 
+#include <core/client.h>
 #include <subdev/fb.h>
 #include <subdev/ltc.h>
+#include <subdev/timer.h>
+#include <engine/gr.h>
 
 #include <nvif/ifc00d.h>
 #include <nvif/unpack.h>
 
+static void
+gp100_vmm_pfn_unmap(struct nvkm_vmm *vmm,
+		    struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes)
+{
+	struct device *dev = vmm->mmu->subdev.device->dev;
+	dma_addr_t addr;
+
+	nvkm_kmap(pt->memory);
+	while (ptes--) {
+		u32 datalo = nvkm_ro32(pt->memory, pt->base + ptei * 8 + 0);
+		u32 datahi = nvkm_ro32(pt->memory, pt->base + ptei * 8 + 4);
+		u64 data   = (u64)datahi << 32 | datalo;
+		if ((data & (3ULL << 1)) != 0) {
+			addr = (data >> 8) << 12;
+			dma_unmap_page(dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
+		}
+		ptei++;
+	}
+	nvkm_done(pt->memory);
+}
+
+static bool
+gp100_vmm_pfn_clear(struct nvkm_vmm *vmm,
+		    struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes)
+{
+	bool dma = false;
+	nvkm_kmap(pt->memory);
+	while (ptes--) {
+		u32 datalo = nvkm_ro32(pt->memory, pt->base + ptei * 8 + 0);
+		u32 datahi = nvkm_ro32(pt->memory, pt->base + ptei * 8 + 4);
+		u64 data   = (u64)datahi << 32 | datalo;
+		if ((data & BIT_ULL(0)) && (data & (3ULL << 1)) != 0) {
+			VMM_WO064(pt, vmm, ptei * 8, data & ~BIT_ULL(0));
+			dma = true;
+		}
+		ptei++;
+	}
+	nvkm_done(pt->memory);
+	return dma;
+}
+
+static void
+gp100_vmm_pgt_pfn(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
+		  u32 ptei, u32 ptes, struct nvkm_vmm_map *map)
+{
+	struct device *dev = vmm->mmu->subdev.device->dev;
+	dma_addr_t addr;
+
+	nvkm_kmap(pt->memory);
+	while (ptes--) {
+		u64 data = 0;
+		if (!(*map->pfn & NVKM_VMM_PFN_W))
+			data |= BIT_ULL(6); /* RO. */
+
+		if (!(*map->pfn & NVKM_VMM_PFN_VRAM)) {
+			addr = *map->pfn >> NVKM_VMM_PFN_ADDR_SHIFT;
+			addr = dma_map_page(dev, pfn_to_page(addr), 0,
+					    PAGE_SIZE, DMA_BIDIRECTIONAL);
+			if (!WARN_ON(dma_mapping_error(dev, addr))) {
+				data |= addr >> 4;
+				data |= 2ULL << 1; /* SYSTEM_COHERENT_MEMORY. */
+				data |= BIT_ULL(3); /* VOL. */
+				data |= BIT_ULL(0); /* VALID. */
+			}
+		} else {
+			data |= (*map->pfn & NVKM_VMM_PFN_ADDR) >> 4;
+			data |= BIT_ULL(0); /* VALID. */
+		}
+
+		VMM_WO064(pt, vmm, ptei++ * 8, data);
+		map->pfn++;
+	}
+	nvkm_done(pt->memory);
+}
+
 static inline void
 gp100_vmm_pgt_pte(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
 		  u32 ptei, u32 ptes, struct nvkm_vmm_map *map, u64 addr)
@@ -89,6 +167,9 @@ gp100_vmm_desc_spt = {
 	.mem = gp100_vmm_pgt_mem,
 	.dma = gp100_vmm_pgt_dma,
 	.sgl = gp100_vmm_pgt_sgl,
+	.pfn = gp100_vmm_pgt_pfn,
+	.pfn_clear = gp100_vmm_pfn_clear,
+	.pfn_unmap = gp100_vmm_pfn_unmap,
 };
 
 static void
@@ -306,16 +387,100 @@ gp100_vmm_valid(struct nvkm_vmm *vmm, void *argv, u32 argc,
 	return 0;
 }
 
+static int
+gp100_vmm_fault_cancel(struct nvkm_vmm *vmm, void *argv, u32 argc)
+{
+	struct nvkm_device *device = vmm->mmu->subdev.device;
+	union {
+		struct gp100_vmm_fault_cancel_v0 v0;
+	} *args = argv;
+	int ret = -ENOSYS;
+	u32 inst, aper;
+
+	if ((ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false)))
+		return ret;
+
+	/* Translate MaxwellFaultBufferA instance pointer to the same
+	 * format as the NV_GR_FECS_CURRENT_CTX register.
+	 */
+	aper = (args->v0.inst >> 8) & 3;
+	args->v0.inst >>= 12;
+	args->v0.inst |= aper << 28;
+	args->v0.inst |= 0x80000000;
+
+	if (!WARN_ON(nvkm_gr_ctxsw_pause(device))) {
+		if ((inst = nvkm_gr_ctxsw_inst(device)) == args->v0.inst) {
+			gf100_vmm_invalidate(vmm, 0x0000001b
+					     /* CANCEL_TARGETED. */ |
+					     (args->v0.hub    << 20) |
+					     (args->v0.gpc    << 15) |
+					     (args->v0.client << 9));
+		}
+		WARN_ON(nvkm_gr_ctxsw_resume(device));
+	}
+
+	return 0;
+}
+
+static int
+gp100_vmm_fault_replay(struct nvkm_vmm *vmm, void *argv, u32 argc)
+{
+	union {
+		struct gp100_vmm_fault_replay_vn vn;
+	} *args = argv;
+	int ret = -ENOSYS;
+
+	if (!(ret = nvif_unvers(ret, &argv, &argc, args->vn))) {
+		gf100_vmm_invalidate(vmm, 0x0000000b); /* REPLAY_GLOBAL. */
+	}
+
+	return ret;
+}
+
+int
+gp100_vmm_mthd(struct nvkm_vmm *vmm,
+	       struct nvkm_client *client, u32 mthd, void *argv, u32 argc)
+{
+	if (client->super) {
+		switch (mthd) {
+		case GP100_VMM_VN_FAULT_REPLAY:
+			return gp100_vmm_fault_replay(vmm, argv, argc);
+		case GP100_VMM_VN_FAULT_CANCEL:
+			return gp100_vmm_fault_cancel(vmm, argv, argc);
+		default:
+			break;
+		}
+	}
+	return -EINVAL;
+}
+
+void
+gp100_vmm_invalidate_pdb(struct nvkm_vmm *vmm, u64 addr)
+{
+	struct nvkm_device *device = vmm->mmu->subdev.device;
+	nvkm_wr32(device, 0x100cb8, lower_32_bits(addr));
+	nvkm_wr32(device, 0x100cec, upper_32_bits(addr));
+}
+
 void
 gp100_vmm_flush(struct nvkm_vmm *vmm, int depth)
 {
-	gf100_vmm_flush_(vmm, 5 /* CACHE_LEVEL_UP_TO_PDE3 */ - depth);
+	u32 type = (5 /* CACHE_LEVEL_UP_TO_PDE3 */ - depth) << 24;
+	type = 0; /*XXX: need to confirm stuff works with depth enabled... */
+	if (atomic_read(&vmm->engref[NVKM_SUBDEV_BAR]))
+		type |= 0x00000004; /* HUB_ONLY */
+	type |= 0x00000001; /* PAGE_ALL */
+	gf100_vmm_invalidate(vmm, type);
 }
 
 int
 gp100_vmm_join(struct nvkm_vmm *vmm, struct nvkm_memory *inst)
 {
-	const u64 base = BIT_ULL(10) /* VER2 */ | BIT_ULL(11); /* 64KiB */
+	u64 base = BIT_ULL(10) /* VER2 */ | BIT_ULL(11) /* 64KiB */;
+	if (vmm->replay) {
+		base |= BIT_ULL(4); /* FAULT_REPLAY_TEX */
+		base |= BIT_ULL(5); /* FAULT_REPLAY_GCC */
+	}
 	return gf100_vmm_join_(vmm, inst, base);
 }
 
@@ -326,6 +491,8 @@ gp100_vmm = {
 	.aper = gf100_vmm_aper,
 	.valid = gp100_vmm_valid,
 	.flush = gp100_vmm_flush,
+	.mthd = gp100_vmm_mthd,
+	.invalidate_pdb = gp100_vmm_invalidate_pdb,
 	.page = {
 		{ 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx },
 		{ 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx },
@@ -338,10 +505,39 @@ gp100_vmm = {
 };
 
 int
-gp100_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gp100_vmm_new_(const struct nvkm_vmm_func *func,
+	       struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	       void *argv, u32 argc, struct lock_class_key *key,
+	       const char *name, struct nvkm_vmm **pvmm)
+{
+	union {
+		struct gp100_vmm_vn vn;
+		struct gp100_vmm_v0 v0;
+	} *args = argv;
+	int ret = -ENOSYS;
+	bool replay;
+
+	if (!(ret = nvif_unpack(ret, &argv, &argc, args->v0, 0, 0, false))) {
+		replay = args->v0.fault_replay != 0;
+	} else
+	if (!(ret = nvif_unvers(ret, &argv, &argc, args->vn))) {
+		replay = false;
+	} else
+		return ret;
+
+	ret = nvkm_vmm_new_(func, mmu, 0, managed, addr, size, key, name, pvmm);
+	if (ret)
+		return ret;
+
+	(*pvmm)->replay = replay;
+	return 0;
+}
+
+int
+gp100_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&gp100_vmm, mmu, 0, addr, size,
-			     argv, argc, key, name, pvmm);
+	return gp100_vmm_new_(&gp100_vmm, mmu, managed, addr, size,
+			      argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp10b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp10b.c
index 3dcc6bddb32f..e081239afe58 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp10b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp10b.c
@@ -28,6 +28,8 @@ gp10b_vmm = {
 	.aper = gk20a_vmm_aper,
 	.valid = gp100_vmm_valid,
 	.flush = gp100_vmm_flush,
+	.mthd = gp100_vmm_mthd,
+	.invalidate_pdb = gp100_vmm_invalidate_pdb,
 	.page = {
 		{ 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx },
 		{ 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx },
@@ -40,10 +42,10 @@ gp10b_vmm = {
 };
 
 int
-gp10b_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gp10b_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&gp10b_vmm, mmu, 0, addr, size,
-			     argv, argc, key, name, pvmm);
+	return gp100_vmm_new_(&gp10b_vmm, mmu, managed, addr, size,
+			      argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c
index 2fa40c16e6d2..f0e21f63253a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgv100.c
@@ -66,6 +66,8 @@ gv100_vmm = {
 	.aper = gf100_vmm_aper,
 	.valid = gp100_vmm_valid,
 	.flush = gp100_vmm_flush,
+	.mthd = gp100_vmm_mthd,
+	.invalidate_pdb = gp100_vmm_invalidate_pdb,
 	.page = {
 		{ 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx },
 		{ 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx },
@@ -78,10 +80,10 @@ gv100_vmm = {
 };
 
 int
-gv100_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+gv100_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&gv100_vmm, mmu, 0, addr, size,
-			     argv, argc, key, name, pvmm);
+	return gp100_vmm_new_(&gv100_vmm, mmu, managed, addr, size,
+			      argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmmcp77.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmmcp77.c
index e63d984cbfd4..bdddd99f5877 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmmcp77.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmmcp77.c
@@ -36,10 +36,10 @@ mcp77_vmm = {
 };
 
 int
-mcp77_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+mcp77_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&mcp77_vmm, mmu, 0, addr, size,
+	return nv04_vmm_new_(&mcp77_vmm, mmu, 0, managed, addr, size,
 			     argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv04.c
index 0cab1ffc9f64..4c6b3b7d221f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv04.c
@@ -100,16 +100,17 @@ nv04_vmm = {
 
 int
 nv04_vmm_new_(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
-	      u32 pd_header, u64 addr, u64 size, void *argv, u32 argc,
-	      struct lock_class_key *key, const char *name,
-	      struct nvkm_vmm **pvmm)
+	      u32 pd_header, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
 {
 	union {
 		struct nv04_vmm_vn vn;
 	} *args = argv;
 	int ret;
 
-	ret = nvkm_vmm_new_(func, mmu, pd_header, addr, size, key, name, pvmm);
+	ret = nvkm_vmm_new_(func, mmu, pd_header, managed, addr, size,
+			    key, name, pvmm);
 	if (ret)
 		return ret;
 
@@ -117,15 +118,15 @@ nv04_vmm_new_(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu,
 }
 
 int
-nv04_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	     struct lock_class_key *key, const char *name,
+nv04_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	     void *argv, u32 argc, struct lock_class_key *key, const char *name,
 	     struct nvkm_vmm **pvmm)
 {
 	struct nvkm_memory *mem;
 	struct nvkm_vmm *vmm;
 	int ret;
 
-	ret = nv04_vmm_new_(&nv04_vmm, mmu, 8, addr, size,
+	ret = nv04_vmm_new_(&nv04_vmm, mmu, 8, managed, addr, size,
 			    argv, argc, key, name, &vmm);
 	*pvmm = vmm;
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv41.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv41.c
index b595f130e573..1d3369683a21 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv41.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv41.c
@@ -104,10 +104,10 @@ nv41_vmm = {
 };
 
 int
-nv41_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	     struct lock_class_key *key, const char *name,
+nv41_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	     void *argv, u32 argc, struct lock_class_key *key, const char *name,
 	     struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&nv41_vmm, mmu, 0, addr, size,
+	return nv04_vmm_new_(&nv41_vmm, mmu, 0, managed, addr, size,
 			     argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv44.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv44.c
index b834e4352334..a82936ba9890 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv44.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv44.c
@@ -205,15 +205,15 @@ nv44_vmm = {
 };
 
 int
-nv44_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	     struct lock_class_key *key, const char *name,
+nv44_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	     void *argv, u32 argc, struct lock_class_key *key, const char *name,
 	     struct nvkm_vmm **pvmm)
 {
 	struct nvkm_subdev *subdev = &mmu->subdev;
 	struct nvkm_vmm *vmm;
 	int ret;
 
-	ret = nv04_vmm_new_(&nv44_vmm, mmu, 0, addr, size,
+	ret = nv04_vmm_new_(&nv44_vmm, mmu, 0, managed, addr, size,
 			    argv, argc, key, name, &vmm);
 	*pvmm = vmm;
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv50.c
index 64f75d906202..c98afe3134ee 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmnv50.c
@@ -376,10 +376,10 @@ nv50_vmm = {
 };
 
 int
-nv50_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, void *argv, u32 argc,
-	     struct lock_class_key *key, const char *name,
+nv50_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	     void *argv, u32 argc, struct lock_class_key *key, const char *name,
 	     struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&nv50_vmm, mmu, 0, addr, size,
+	return nv04_vmm_new_(&nv50_vmm, mmu, 0, managed, addr, size,
 			     argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c
index adaadd92110f..be91cffc3b52 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c
@@ -24,7 +24,7 @@
 #include <subdev/timer.h>
 
 static void
-tu104_vmm_flush(struct nvkm_vmm *vmm, int depth)
+tu102_vmm_flush(struct nvkm_vmm *vmm, int depth)
 {
 	struct nvkm_subdev *subdev = &vmm->mmu->subdev;
 	struct nvkm_device *device = subdev->device;
@@ -50,12 +50,13 @@ tu104_vmm_flush(struct nvkm_vmm *vmm, int depth)
 }
 
 static const struct nvkm_vmm_func
-tu104_vmm = {
+tu102_vmm = {
 	.join = gv100_vmm_join,
 	.part = gf100_vmm_part,
 	.aper = gf100_vmm_aper,
 	.valid = gp100_vmm_valid,
-	.flush = tu104_vmm_flush,
+	.flush = tu102_vmm_flush,
+	.mthd = gp100_vmm_mthd,
 	.page = {
 		{ 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx },
 		{ 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx },
@@ -68,10 +69,10 @@ tu104_vmm = {
 };
 
 int
-tu104_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size,
+tu102_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
 	      void *argv, u32 argc, struct lock_class_key *key,
 	      const char *name, struct nvkm_vmm **pvmm)
 {
-	return nv04_vmm_new_(&tu104_vmm, mmu, 0, addr, size,
-			     argv, argc, key, name, pvmm);
+	return gp100_vmm_new_(&tu102_vmm, mmu, managed, addr, size,
+			      argv, argc, key, name, pvmm);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/memx.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/memx.c
index 11b28b086a06..7b052879af72 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/memx.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/memx.c
@@ -88,10 +88,10 @@ nvkm_memx_fini(struct nvkm_memx **pmemx, bool exec)
 	if (exec) {
 		nvkm_pmu_send(pmu, reply, PROC_MEMX, MEMX_MSG_EXEC,
 			      memx->base, finish);
+		nvkm_debug(subdev, "Exec took %uns, PMU_IN %08x\n",
+			   reply[0], reply[1]);
 	}
 
-	nvkm_debug(subdev, "Exec took %uns, PMU_IN %08x\n",
-		   reply[0], reply[1]);
 	kfree(memx);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/acr_r352.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/acr_r352.c
index 5c14d6ac855d..1df09ed6fe6d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/acr_r352.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/acr_r352.c
@@ -853,7 +853,7 @@ acr_r352_shutdown(struct acr_r352 *acr, struct nvkm_secboot *sb)
 		 * and the expected behavior on RM as well
 		 */
 		if (ret && ret != 0x1d) {
-			nvkm_error(subdev, "HS unload failed, ret 0x%08x", ret);
+			nvkm_error(subdev, "HS unload failed, ret 0x%08x\n", ret);
 			return -EINVAL;
 		}
 		nvkm_debug(subdev, "HS unload blob completed\n");
@@ -922,7 +922,7 @@ acr_r352_bootstrap(struct acr_r352 *acr, struct nvkm_secboot *sb)
 	if (ret < 0) {
 		return ret;
 	} else if (ret > 0) {
-		nvkm_error(subdev, "HS load failed, ret 0x%08x", ret);
+		nvkm_error(subdev, "HS load failed, ret 0x%08x\n", ret);
 		return -EINVAL;
 	}
 	nvkm_debug(subdev, "HS load blob completed\n");
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c
index 3695cde669f8..07914e36939e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c
@@ -132,11 +132,12 @@ nvkm_therm_update(struct nvkm_therm *therm, int mode)
 			duty = nvkm_therm_update_linear(therm);
 			break;
 		case NVBIOS_THERM_FAN_OTHER:
-			if (therm->cstate)
+			if (therm->cstate) {
 				duty = therm->cstate;
-			else
+				poll = false;
+			} else {
 				duty = nvkm_therm_update_linear_fallback(therm);
-			poll = false;
+			}
 			break;
 		}
 		immd = false;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/top/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/top/base.c
index 67ada1d9a28c..cce6e4e90ebf 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/top/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/top/base.c
@@ -41,6 +41,22 @@ nvkm_top_device_new(struct nvkm_top *top)
 }
 
 u32
+nvkm_top_addr(struct nvkm_device *device, enum nvkm_devidx index)
+{
+	struct nvkm_top *top = device->top;
+	struct nvkm_top_device *info;
+
+	if (top) {
+		list_for_each_entry(info, &top->device, head) {
+			if (info->index == index)
+				return info->addr;
+		}
+	}
+
+	return 0;
+}
+
+u32
 nvkm_top_reset(struct nvkm_device *device, enum nvkm_devidx index)
 {
 	struct nvkm_top *top = device->top;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c
index 39081eadfd84..e01746ce9fc4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c
@@ -73,6 +73,7 @@ gk104_top_oneinit(struct nvkm_top *top)
 #define A_(A) if (inst == 0) info->index = NVKM_ENGINE_##A
 #define B_(A) if (inst + NVKM_ENGINE_##A##0 < NVKM_ENGINE_##A##_LAST + 1)      \
 		info->index = NVKM_ENGINE_##A##0 + inst
+#define C_(A) if (inst == 0) info->index = NVKM_SUBDEV_##A
 		switch (type) {
 		case 0x00000000: A_(GR    ); break;
 		case 0x00000001: A_(CE0   ); break;
@@ -88,6 +89,7 @@ gk104_top_oneinit(struct nvkm_top *top)
 		case 0x0000000f: A_(NVENC1); break;
 		case 0x00000010: B_(NVDEC ); break;
 		case 0x00000013: B_(CE    ); break;
+		case 0x00000014: C_(GSP   ); break;
 			break;
 		default:
 			break;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
index bcd179ba11d0..146adcdd316a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
@@ -2,6 +2,7 @@ nvkm-y += nvkm/subdev/volt/base.o
 nvkm-y += nvkm/subdev/volt/gpio.o
 nvkm-y += nvkm/subdev/volt/nv40.o
 nvkm-y += nvkm/subdev/volt/gf100.o
+nvkm-y += nvkm/subdev/volt/gf117.o
 nvkm-y += nvkm/subdev/volt/gk104.o
 nvkm-y += nvkm/subdev/volt/gk20a.o
 nvkm-y += nvkm/subdev/volt/gm20b.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gf117.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gf117.c
new file mode 100644
index 000000000000..547a58f0aeac
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gf117.c
@@ -0,0 +1,60 @@
+/*
+ * Copyright 2019 Ilia Mirkin
+ *
+ * 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: Ilia Mirkin
+ */
+#include "priv.h"
+
+#include <subdev/fuse.h>
+
+static int
+gf117_volt_speedo_read(struct nvkm_volt *volt)
+{
+	struct nvkm_device *device = volt->subdev.device;
+	struct nvkm_fuse *fuse = device->fuse;
+
+	if (!fuse)
+		return -EINVAL;
+
+	return nvkm_fuse_read(fuse, 0x3a8);
+}
+
+static const struct nvkm_volt_func
+gf117_volt = {
+	.oneinit = gf100_volt_oneinit,
+	.vid_get = nvkm_voltgpio_get,
+	.vid_set = nvkm_voltgpio_set,
+	.speedo_read = gf117_volt_speedo_read,
+};
+
+int
+gf117_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
+{
+	struct nvkm_volt *volt;
+	int ret;
+
+	ret = nvkm_volt_new_(&gf117_volt, device, index, &volt);
+	*pvolt = volt;
+	if (ret)
+		return ret;
+
+	return nvkm_voltgpio_init(volt);
+}
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 00a9c2ab9e6c..64fb788b6647 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -1406,7 +1406,7 @@ static void dsi_pll_disable(struct dss_pll *pll)
 
 static int dsi_dump_dsi_clocks(struct seq_file *s, void *p)
 {
-	struct dsi_data *dsi = p;
+	struct dsi_data *dsi = s->private;
 	struct dss_pll_clock_info *cinfo = &dsi->pll.cinfo;
 	enum dss_clk_source dispc_clk_src, dsi_clk_src;
 	int dsi_module = dsi->module_id;
@@ -1467,7 +1467,7 @@ static int dsi_dump_dsi_clocks(struct seq_file *s, void *p)
 #ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
 static int dsi_dump_dsi_irqs(struct seq_file *s, void *p)
 {
-	struct dsi_data *dsi = p;
+	struct dsi_data *dsi = s->private;
 	unsigned long flags;
 	struct dsi_irq_stats stats;
 
@@ -1558,7 +1558,7 @@ static int dsi_dump_dsi_irqs(struct seq_file *s, void *p)
 
 static int dsi_dump_dsi_regs(struct seq_file *s, void *p)
 {
-	struct dsi_data *dsi = p;
+	struct dsi_data *dsi = s->private;
 
 	if (dsi_runtime_get(dsi))
 		return 0;
@@ -4751,6 +4751,17 @@ static int dsi_set_config(struct omap_dss_device *dssdev,
 	dsi->vm.flags |= DISPLAY_FLAGS_HSYNC_HIGH;
 	dsi->vm.flags &= ~DISPLAY_FLAGS_VSYNC_LOW;
 	dsi->vm.flags |= DISPLAY_FLAGS_VSYNC_HIGH;
+	/*
+	 * HACK: These flags should be handled through the omap_dss_device bus
+	 * flags, but this will only be possible when the DSI encoder will be
+	 * converted to the omapdrm-managed encoder model.
+	 */
+	dsi->vm.flags &= ~DISPLAY_FLAGS_PIXDATA_NEGEDGE;
+	dsi->vm.flags |= DISPLAY_FLAGS_PIXDATA_POSEDGE;
+	dsi->vm.flags &= ~DISPLAY_FLAGS_DE_LOW;
+	dsi->vm.flags |= DISPLAY_FLAGS_DE_HIGH;
+	dsi->vm.flags &= ~DISPLAY_FLAGS_SYNC_POSEDGE;
+	dsi->vm.flags |= DISPLAY_FLAGS_SYNC_NEGEDGE;
 
 	dss_mgr_set_timings(&dsi->output, &dsi->vm);
 
@@ -5083,15 +5094,15 @@ static int dsi_bind(struct device *dev, struct device *master, void *data)
 
 	snprintf(name, sizeof(name), "dsi%u_regs", dsi->module_id + 1);
 	dsi->debugfs.regs = dss_debugfs_create_file(dss, name,
-						    dsi_dump_dsi_regs, &dsi);
+						    dsi_dump_dsi_regs, dsi);
 #ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
 	snprintf(name, sizeof(name), "dsi%u_irqs", dsi->module_id + 1);
 	dsi->debugfs.irqs = dss_debugfs_create_file(dss, name,
-						    dsi_dump_dsi_irqs, &dsi);
+						    dsi_dump_dsi_irqs, dsi);
 #endif
 	snprintf(name, sizeof(name), "dsi%u_clks", dsi->module_id + 1);
 	dsi->debugfs.clks = dss_debugfs_create_file(dss, name,
-						    dsi_dump_dsi_clocks, &dsi);
+						    dsi_dump_dsi_clocks, dsi);
 
 	return 0;
 }
@@ -5104,8 +5115,6 @@ static void dsi_unbind(struct device *dev, struct device *master, void *data)
 	dss_debugfs_remove_file(dsi->debugfs.irqs);
 	dss_debugfs_remove_file(dsi->debugfs.regs);
 
-	of_platform_depopulate(dev);
-
 	WARN_ON(dsi->scp_clk_refcount > 0);
 
 	dss_pll_unregister(&dsi->pll);
@@ -5457,6 +5466,8 @@ static int dsi_remove(struct platform_device *pdev)
 
 	dsi_uninit_output(dsi);
 
+	of_platform_depopulate(&pdev->dev);
+
 	pm_runtime_disable(&pdev->dev);
 
 	if (dsi->vdds_dsi_reg != NULL && dsi->vdds_dsi_enabled) {
diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c
index 874d8f3cbff6..9da94d10782a 100644
--- a/drivers/gpu/drm/omapdrm/omap_connector.c
+++ b/drivers/gpu/drm/omapdrm/omap_connector.c
@@ -17,7 +17,7 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "omap_drv.h"
 
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 40acf4ce7c9f..d99e24dcc0bf 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -18,7 +18,6 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_mode.h>
 #include <drm/drm_plane_helper.h>
 #include <linux/math64.h>
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 5e67d58cbc28..f8292278f57d 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -21,8 +21,8 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "omap_dmm_tiler.h"
 #include "omap_drv.h"
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index bd7f2c227a25..0c57d2814c51 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -23,7 +23,6 @@
 #include <linux/workqueue.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem.h>
 #include <drm/omap_drm.h>
 
diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c
index 148b6b20274f..0d85b3a35767 100644
--- a/drivers/gpu/drm/omapdrm/omap_encoder.c
+++ b/drivers/gpu/drm/omapdrm/omap_encoder.c
@@ -18,7 +18,7 @@
 #include <linux/list.h>
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_edid.h>
 
 #include "omap_drv.h"
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index 4d264fd554d8..4f8eb9d08f99 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -18,7 +18,7 @@
 #include <linux/seq_file.h>
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 
 #include "omap_dmm_tiler.h"
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index a71f44191273..3e070153ef21 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -77,6 +77,17 @@ config DRM_PANEL_JDI_LT070ME05000
 	  The panel has a 1200(RGB)×1920 (WUXGA) resolution and uses
 	  24 bit per pixel.
 
+config DRM_PANEL_KINGDISPLAY_KD097D04
+	tristate "Kingdisplay kd097d04 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 Kingdisplay kd097d04
+	  TFT-LCD modules. The panel has a 1536x2048 resolution and uses
+	  24 bit RGB per pixel. It provides a MIPI DSI interface to
+	  the host and has a built-in LED backlight.
+
 config DRM_PANEL_SAMSUNG_LD9040
 	tristate "Samsung LD9040 RGB/SPI panel"
 	depends on OF && SPI
@@ -196,6 +207,16 @@ config DRM_PANEL_SHARP_LS043T1LE01
 	  Say Y here if you want to enable support for Sharp LS043T1LE01 qHD
 	  (540x960) DSI panel as found on the Qualcomm APQ8074 Dragonboard
 
+config DRM_PANEL_SITRONIX_ST7701
+	tristate "Sitronix ST7701 panel driver"
+	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 Sitronix
+	  ST7701 controller for 480X864 LCD panels with MIPI/RGB/SPI
+	  system interfaces.
+
 config DRM_PANEL_SITRONIX_ST7789V
 	tristate "Sitronix ST7789V panel"
 	depends on OF && SPI
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index cd14ca39c6e0..e7ab71968bbf 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_DRM_PANEL_ILITEK_IL9322) += panel-ilitek-ili9322.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_ILI9881C) += panel-ilitek-ili9881c.o
 obj-$(CONFIG_DRM_PANEL_INNOLUX_P079ZCA) += panel-innolux-p079zca.o
 obj-$(CONFIG_DRM_PANEL_JDI_LT070ME05000) += panel-jdi-lt070me05000.o
+obj-$(CONFIG_DRM_PANEL_KINGDISPLAY_KD097D04) += panel-kingdisplay-kd097d04.o
 obj-$(CONFIG_DRM_PANEL_LG_LG4573) += panel-lg-lg4573.o
 obj-$(CONFIG_DRM_PANEL_OLIMEX_LCD_OLINUXINO) += panel-olimex-lcd-olinuxino.o
 obj-$(CONFIG_DRM_PANEL_ORISETECH_OTM8009A) += panel-orisetech-otm8009a.o
@@ -20,6 +21,7 @@ obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0) += panel-samsung-s6e8aa0.o
 obj-$(CONFIG_DRM_PANEL_SEIKO_43WVF1G) += panel-seiko-43wvf1g.o
 obj-$(CONFIG_DRM_PANEL_SHARP_LQ101R1SX01) += panel-sharp-lq101r1sx01.o
 obj-$(CONFIG_DRM_PANEL_SHARP_LS043T1LE01) += panel-sharp-ls043t1le01.o
+obj-$(CONFIG_DRM_PANEL_SITRONIX_ST7701) += panel-sitronix-st7701.o
 obj-$(CONFIG_DRM_PANEL_SITRONIX_ST7789V) += panel-sitronix-st7789v.o
 obj-$(CONFIG_DRM_PANEL_TPO_TPG110) += panel-tpo-tpg110.o
 obj-$(CONFIG_DRM_PANEL_TRULY_NT35597_WQXGA) += panel-truly-nt35597.o
diff --git a/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c b/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c
new file mode 100644
index 000000000000..2a25a914d09e
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-kingdisplay-kd097d04.c
@@ -0,0 +1,473 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (c) 2017, Fuzhou Rockchip Electronics Co., Ltd
+ */
+
+#include <linux/backlight.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_panel.h>
+
+#include <video/mipi_display.h>
+
+struct kingdisplay_panel {
+	struct drm_panel base;
+	struct mipi_dsi_device *link;
+
+	struct backlight_device *backlight;
+	struct regulator *supply;
+	struct gpio_desc *enable_gpio;
+
+	bool prepared;
+	bool enabled;
+};
+
+struct kingdisplay_panel_cmd {
+	char cmd;
+	char data;
+};
+
+/*
+ * According to the discussion on
+ * https://review.coreboot.org/#/c/coreboot/+/22472/
+ * the panel init array is not part of the panels datasheet but instead
+ * just came in this form from the panel vendor.
+ */
+static const struct kingdisplay_panel_cmd init_code[] = {
+	/* voltage setting */
+	{ 0xB0, 0x00 },
+	{ 0xB2, 0x02 },
+	{ 0xB3, 0x11 },
+	{ 0xB4, 0x00 },
+	{ 0xB6, 0x80 },
+	/* VCOM disable */
+	{ 0xB7, 0x02 },
+	{ 0xB8, 0x80 },
+	{ 0xBA, 0x43 },
+	/* VCOM setting */
+	{ 0xBB, 0x53 },
+	/* VSP setting */
+	{ 0xBC, 0x0A },
+	/* VSN setting */
+	{ 0xBD, 0x4A },
+	/* VGH setting */
+	{ 0xBE, 0x2F },
+	/* VGL setting */
+	{ 0xBF, 0x1A },
+	{ 0xF0, 0x39 },
+	{ 0xF1, 0x22 },
+	/* Gamma setting */
+	{ 0xB0, 0x02 },
+	{ 0xC0, 0x00 },
+	{ 0xC1, 0x01 },
+	{ 0xC2, 0x0B },
+	{ 0xC3, 0x15 },
+	{ 0xC4, 0x22 },
+	{ 0xC5, 0x11 },
+	{ 0xC6, 0x15 },
+	{ 0xC7, 0x19 },
+	{ 0xC8, 0x1A },
+	{ 0xC9, 0x16 },
+	{ 0xCA, 0x18 },
+	{ 0xCB, 0x13 },
+	{ 0xCC, 0x18 },
+	{ 0xCD, 0x13 },
+	{ 0xCE, 0x1C },
+	{ 0xCF, 0x19 },
+	{ 0xD0, 0x21 },
+	{ 0xD1, 0x2C },
+	{ 0xD2, 0x2F },
+	{ 0xD3, 0x30 },
+	{ 0xD4, 0x19 },
+	{ 0xD5, 0x1F },
+	{ 0xD6, 0x00 },
+	{ 0xD7, 0x01 },
+	{ 0xD8, 0x0B },
+	{ 0xD9, 0x15 },
+	{ 0xDA, 0x22 },
+	{ 0xDB, 0x11 },
+	{ 0xDC, 0x15 },
+	{ 0xDD, 0x19 },
+	{ 0xDE, 0x1A },
+	{ 0xDF, 0x16 },
+	{ 0xE0, 0x18 },
+	{ 0xE1, 0x13 },
+	{ 0xE2, 0x18 },
+	{ 0xE3, 0x13 },
+	{ 0xE4, 0x1C },
+	{ 0xE5, 0x19 },
+	{ 0xE6, 0x21 },
+	{ 0xE7, 0x2C },
+	{ 0xE8, 0x2F },
+	{ 0xE9, 0x30 },
+	{ 0xEA, 0x19 },
+	{ 0xEB, 0x1F },
+	/* GOA MUX setting */
+	{ 0xB0, 0x01 },
+	{ 0xC0, 0x10 },
+	{ 0xC1, 0x0F },
+	{ 0xC2, 0x0E },
+	{ 0xC3, 0x0D },
+	{ 0xC4, 0x0C },
+	{ 0xC5, 0x0B },
+	{ 0xC6, 0x0A },
+	{ 0xC7, 0x09 },
+	{ 0xC8, 0x08 },
+	{ 0xC9, 0x07 },
+	{ 0xCA, 0x06 },
+	{ 0xCB, 0x05 },
+	{ 0xCC, 0x00 },
+	{ 0xCD, 0x01 },
+	{ 0xCE, 0x02 },
+	{ 0xCF, 0x03 },
+	{ 0xD0, 0x04 },
+	{ 0xD6, 0x10 },
+	{ 0xD7, 0x0F },
+	{ 0xD8, 0x0E },
+	{ 0xD9, 0x0D },
+	{ 0xDA, 0x0C },
+	{ 0xDB, 0x0B },
+	{ 0xDC, 0x0A },
+	{ 0xDD, 0x09 },
+	{ 0xDE, 0x08 },
+	{ 0xDF, 0x07 },
+	{ 0xE0, 0x06 },
+	{ 0xE1, 0x05 },
+	{ 0xE2, 0x00 },
+	{ 0xE3, 0x01 },
+	{ 0xE4, 0x02 },
+	{ 0xE5, 0x03 },
+	{ 0xE6, 0x04 },
+	{ 0xE7, 0x00 },
+	{ 0xEC, 0xC0 },
+	/* GOA timing setting */
+	{ 0xB0, 0x03 },
+	{ 0xC0, 0x01 },
+	{ 0xC2, 0x6F },
+	{ 0xC3, 0x6F },
+	{ 0xC5, 0x36 },
+	{ 0xC8, 0x08 },
+	{ 0xC9, 0x04 },
+	{ 0xCA, 0x41 },
+	{ 0xCC, 0x43 },
+	{ 0xCF, 0x60 },
+	{ 0xD2, 0x04 },
+	{ 0xD3, 0x04 },
+	{ 0xD4, 0x03 },
+	{ 0xD5, 0x02 },
+	{ 0xD6, 0x01 },
+	{ 0xD7, 0x00 },
+	{ 0xDB, 0x01 },
+	{ 0xDE, 0x36 },
+	{ 0xE6, 0x6F },
+	{ 0xE7, 0x6F },
+	/* GOE setting */
+	{ 0xB0, 0x06 },
+	{ 0xB8, 0xA5 },
+	{ 0xC0, 0xA5 },
+	{ 0xD5, 0x3F },
+};
+
+static inline
+struct kingdisplay_panel *to_kingdisplay_panel(struct drm_panel *panel)
+{
+	return container_of(panel, struct kingdisplay_panel, base);
+}
+
+static int kingdisplay_panel_disable(struct drm_panel *panel)
+{
+	struct kingdisplay_panel *kingdisplay = to_kingdisplay_panel(panel);
+	int err;
+
+	if (!kingdisplay->enabled)
+		return 0;
+
+	backlight_disable(kingdisplay->backlight);
+
+	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);
+
+	kingdisplay->enabled = false;
+
+	return 0;
+}
+
+static int kingdisplay_panel_unprepare(struct drm_panel *panel)
+{
+	struct kingdisplay_panel *kingdisplay = to_kingdisplay_panel(panel);
+	int err;
+
+	if (!kingdisplay->prepared)
+		return 0;
+
+	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);
+		return err;
+	}
+
+	/* T15: 120ms */
+	msleep(120);
+
+	gpiod_set_value_cansleep(kingdisplay->enable_gpio, 0);
+
+	err = regulator_disable(kingdisplay->supply);
+	if (err < 0)
+		return err;
+
+	kingdisplay->prepared = false;
+
+	return 0;
+}
+
+static int kingdisplay_panel_prepare(struct drm_panel *panel)
+{
+	struct kingdisplay_panel *kingdisplay = to_kingdisplay_panel(panel);
+	int err, regulator_err;
+	unsigned int i;
+
+	if (kingdisplay->prepared)
+		return 0;
+
+	gpiod_set_value_cansleep(kingdisplay->enable_gpio, 0);
+
+	err = regulator_enable(kingdisplay->supply);
+	if (err < 0)
+		return err;
+
+	/* T2: 15ms */
+	usleep_range(15000, 16000);
+
+	gpiod_set_value_cansleep(kingdisplay->enable_gpio, 1);
+
+	/* T4: 15ms */
+	usleep_range(15000, 16000);
+
+	for (i = 0; i < ARRAY_SIZE(init_code); i++) {
+		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);
+			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);
+		goto poweroff;
+	}
+
+	/* T6: 120ms */
+	msleep(120);
+
+	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);
+		goto poweroff;
+	}
+
+	/* T7: 10ms */
+	usleep_range(10000, 11000);
+
+	kingdisplay->prepared = true;
+
+	return 0;
+
+poweroff:
+	gpiod_set_value_cansleep(kingdisplay->enable_gpio, 0);
+
+	regulator_err = regulator_disable(kingdisplay->supply);
+	if (regulator_err)
+		DRM_DEV_ERROR(panel->dev, "failed to disable regulator: %d\n",
+			      regulator_err);
+
+	return err;
+}
+
+static int kingdisplay_panel_enable(struct drm_panel *panel)
+{
+	struct kingdisplay_panel *kingdisplay = to_kingdisplay_panel(panel);
+	int ret;
+
+	if (kingdisplay->enabled)
+		return 0;
+
+	ret = backlight_enable(kingdisplay->backlight);
+	if (ret) {
+		DRM_DEV_ERROR(panel->drm->dev,
+			      "Failed to enable backlight %d\n", ret);
+		return ret;
+	}
+
+	kingdisplay->enabled = true;
+
+	return 0;
+}
+
+static const struct drm_display_mode default_mode = {
+	.clock = 229000,
+	.hdisplay = 1536,
+	.hsync_start = 1536 + 100,
+	.hsync_end = 1536 + 100 + 24,
+	.htotal = 1536 + 100 + 24 + 100,
+	.vdisplay = 2048,
+	.vsync_start = 2048 + 95,
+	.vsync_end = 2048 + 95 + 2,
+	.vtotal = 2048 + 95 + 2 + 23,
+	.vrefresh = 60,
+};
+
+static int kingdisplay_panel_get_modes(struct drm_panel *panel)
+{
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(panel->drm, &default_mode);
+	if (!mode) {
+		DRM_DEV_ERROR(panel->drm->dev, "failed to add mode %ux%ux@%u\n",
+			      default_mode.hdisplay, default_mode.vdisplay,
+			      default_mode.vrefresh);
+		return -ENOMEM;
+	}
+
+	drm_mode_set_name(mode);
+
+	drm_mode_probed_add(panel->connector, mode);
+
+	panel->connector->display_info.width_mm = 147;
+	panel->connector->display_info.height_mm = 196;
+	panel->connector->display_info.bpc = 8;
+
+	return 1;
+}
+
+static const struct drm_panel_funcs kingdisplay_panel_funcs = {
+	.disable = kingdisplay_panel_disable,
+	.unprepare = kingdisplay_panel_unprepare,
+	.prepare = kingdisplay_panel_prepare,
+	.enable = kingdisplay_panel_enable,
+	.get_modes = kingdisplay_panel_get_modes,
+};
+
+static const struct of_device_id kingdisplay_of_match[] = {
+	{ .compatible = "kingdisplay,kd097d04", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, kingdisplay_of_match);
+
+static int kingdisplay_panel_add(struct kingdisplay_panel *kingdisplay)
+{
+	struct device *dev = &kingdisplay->link->dev;
+	int err;
+
+	kingdisplay->supply = devm_regulator_get(dev, "power");
+	if (IS_ERR(kingdisplay->supply))
+		return PTR_ERR(kingdisplay->supply);
+
+	kingdisplay->enable_gpio = devm_gpiod_get_optional(dev, "enable",
+							   GPIOD_OUT_HIGH);
+	if (IS_ERR(kingdisplay->enable_gpio)) {
+		err = PTR_ERR(kingdisplay->enable_gpio);
+		dev_dbg(dev, "failed to get enable gpio: %d\n", err);
+		kingdisplay->enable_gpio = NULL;
+	}
+
+	kingdisplay->backlight = devm_of_find_backlight(dev);
+	if (IS_ERR(kingdisplay->backlight))
+		return PTR_ERR(kingdisplay->backlight);
+
+	drm_panel_init(&kingdisplay->base);
+	kingdisplay->base.funcs = &kingdisplay_panel_funcs;
+	kingdisplay->base.dev = &kingdisplay->link->dev;
+
+	return drm_panel_add(&kingdisplay->base);
+}
+
+static void kingdisplay_panel_del(struct kingdisplay_panel *kingdisplay)
+{
+	drm_panel_remove(&kingdisplay->base);
+}
+
+static int kingdisplay_panel_probe(struct mipi_dsi_device *dsi)
+{
+	struct kingdisplay_panel *kingdisplay;
+	int err;
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+			  MIPI_DSI_MODE_LPM;
+
+	kingdisplay = devm_kzalloc(&dsi->dev, sizeof(*kingdisplay), GFP_KERNEL);
+	if (!kingdisplay)
+		return -ENOMEM;
+
+	mipi_dsi_set_drvdata(dsi, kingdisplay);
+	kingdisplay->link = dsi;
+
+	err = kingdisplay_panel_add(kingdisplay);
+	if (err < 0)
+		return err;
+
+	return mipi_dsi_attach(dsi);
+}
+
+static int kingdisplay_panel_remove(struct mipi_dsi_device *dsi)
+{
+	struct kingdisplay_panel *kingdisplay = mipi_dsi_get_drvdata(dsi);
+	int err;
+
+	err = kingdisplay_panel_unprepare(&kingdisplay->base);
+	if (err < 0)
+		DRM_DEV_ERROR(&dsi->dev, "failed to unprepare panel: %d\n",
+			      err);
+
+	err = kingdisplay_panel_disable(&kingdisplay->base);
+	if (err < 0)
+		DRM_DEV_ERROR(&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);
+
+	kingdisplay_panel_del(kingdisplay);
+
+	return 0;
+}
+
+static void kingdisplay_panel_shutdown(struct mipi_dsi_device *dsi)
+{
+	struct kingdisplay_panel *kingdisplay = mipi_dsi_get_drvdata(dsi);
+
+	kingdisplay_panel_unprepare(&kingdisplay->base);
+	kingdisplay_panel_disable(&kingdisplay->base);
+}
+
+static struct mipi_dsi_driver kingdisplay_panel_driver = {
+	.driver = {
+		.name = "panel-kingdisplay-kd097d04",
+		.of_match_table = kingdisplay_of_match,
+	},
+	.probe = kingdisplay_panel_probe,
+	.remove = kingdisplay_panel_remove,
+	.shutdown = kingdisplay_panel_shutdown,
+};
+module_mipi_dsi_driver(kingdisplay_panel_driver);
+
+MODULE_AUTHOR("Chris Zhong <zyw@rock-chips.com>");
+MODULE_AUTHOR("Nickey Yang <nickey.yang@rock-chips.com>");
+MODULE_DESCRIPTION("kingdisplay KD097D04 panel driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 9c69e739a524..9e8218f6a3f2 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -1597,6 +1597,30 @@ static const struct panel_desc kyo_tcg121xglp = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
 };
 
+static const struct drm_display_mode lemaker_bl035_rgb_002_mode = {
+	.clock = 7000,
+	.hdisplay = 320,
+	.hsync_start = 320 + 20,
+	.hsync_end = 320 + 20 + 30,
+	.htotal = 320 + 20 + 30 + 38,
+	.vdisplay = 240,
+	.vsync_start = 240 + 4,
+	.vsync_end = 240 + 4 + 3,
+	.vtotal = 240 + 4 + 3 + 15,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc lemaker_bl035_rgb_002 = {
+	.modes = &lemaker_bl035_rgb_002_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 70,
+		.height = 52,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.bus_flags = DRM_BUS_FLAG_DE_LOW,
+};
+
 static const struct drm_display_mode lg_lb070wv8_mode = {
 	.clock = 33246,
 	.hdisplay = 800,
@@ -2008,6 +2032,30 @@ static const struct panel_desc ortustech_com43h4m85ulc = {
 	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
 };
 
+static const struct drm_display_mode pda_91_00156_a0_mode = {
+	.clock = 33300,
+	.hdisplay = 800,
+	.hsync_start = 800 + 1,
+	.hsync_end = 800 + 1 + 64,
+	.htotal = 800 + 1 + 64 + 64,
+	.vdisplay = 480,
+	.vsync_start = 480 + 1,
+	.vsync_end = 480 + 1 + 23,
+	.vtotal = 480 + 1 + 23 + 22,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc pda_91_00156_a0  = {
+	.modes = &pda_91_00156_a0_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+};
+
+
 static const struct drm_display_mode qd43003c0_40_mode = {
 	.clock = 9000,
 	.hdisplay = 480,
@@ -2638,6 +2686,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "kyo,tcg121xglp",
 		.data = &kyo_tcg121xglp,
 	}, {
+		.compatible = "lemaker,bl035-rgb-002",
+		.data = &lemaker_bl035_rgb_002,
+	}, {
 		.compatible = "lg,lb070wv8",
 		.data = &lg_lb070wv8,
 	}, {
@@ -2686,6 +2737,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "ortustech,com43h4m85ulc",
 		.data = &ortustech_com43h4m85ulc,
 	}, {
+		.compatible = "pda,91-00156-a0",
+		.data = &pda_91_00156_a0,
+	}, {
 		.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
new file mode 100644
index 000000000000..63f9a1c7fb1b
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7701.c
@@ -0,0 +1,426 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2019, Amarula Solutions.
+ * Author: Jagan Teki <jagan@amarulasolutions.com>
+ */
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_print.h>
+
+#include <linux/backlight.h>
+#include <linux/gpio/consumer.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/regulator/consumer.h>
+
+#include <video/mipi_display.h>
+
+/* Command2 BKx selection command */
+#define DSI_CMD2BKX_SEL			0xFF
+
+/* Command2, BK0 commands */
+#define DSI_CMD2_BK0_PVGAMCTRL		0xB0 /* Positive Voltage Gamma Control */
+#define DSI_CMD2_BK0_NVGAMCTRL		0xB1 /* Negative Voltage Gamma Control */
+#define DSI_CMD2_BK0_LNESET		0xC0 /* Display Line setting */
+#define DSI_CMD2_BK0_PORCTRL		0xC1 /* Porch control */
+#define DSI_CMD2_BK0_INVSEL		0xC2 /* Inversion selection, Frame Rate Control */
+
+/* Command2, BK1 commands */
+#define DSI_CMD2_BK1_VRHS		0xB0 /* Vop amplitude setting */
+#define DSI_CMD2_BK1_VCOM		0xB1 /* VCOM amplitude setting */
+#define DSI_CMD2_BK1_VGHSS		0xB2 /* VGH Voltage setting */
+#define DSI_CMD2_BK1_TESTCMD		0xB3 /* TEST Command Setting */
+#define DSI_CMD2_BK1_VGLS		0xB5 /* VGL Voltage setting */
+#define DSI_CMD2_BK1_PWCTLR1		0xB7 /* Power Control 1 */
+#define DSI_CMD2_BK1_PWCTLR2		0xB8 /* Power Control 2 */
+#define DSI_CMD2_BK1_SPD1		0xC1 /* Source pre_drive timing set1 */
+#define DSI_CMD2_BK1_SPD2		0xC2 /* Source EQ2 Setting */
+#define DSI_CMD2_BK1_MIPISET1		0xD0 /* MIPI Setting 1 */
+
+/**
+ * Command2 with BK function selection.
+ *
+ * BIT[4, 0]: [CN2, BKXSEL]
+ * 10 = CMD2BK0, Command2 BK0
+ * 11 = CMD2BK1, Command2 BK1
+ * 00 = Command2 disable
+ */
+#define DSI_CMD2BK1_SEL			0x11
+#define DSI_CMD2BK0_SEL			0x10
+#define DSI_CMD2BKX_SEL_NONE		0x00
+
+/* Command2, BK0 bytes */
+#define DSI_LINESET_LINE		0x69
+#define DSI_LINESET_LDE_EN		BIT(7)
+#define DSI_LINESET_LINEDELTA		GENMASK(1, 0)
+#define DSI_CMD2_BK0_LNESET_B1		DSI_LINESET_LINEDELTA
+#define DSI_CMD2_BK0_LNESET_B0		(DSI_LINESET_LDE_EN | DSI_LINESET_LINE)
+#define DSI_INVSEL_DEFAULT		GENMASK(5, 4)
+#define DSI_INVSEL_NLINV		GENMASK(2, 0)
+#define DSI_INVSEL_RTNI			GENMASK(2, 1)
+#define DSI_CMD2_BK0_INVSEL_B1		DSI_INVSEL_RTNI
+#define DSI_CMD2_BK0_INVSEL_B0		(DSI_INVSEL_DEFAULT | DSI_INVSEL_NLINV)
+#define DSI_CMD2_BK0_PORCTRL_B0(m)	((m)->vtotal - (m)->vsync_end)
+#define DSI_CMD2_BK0_PORCTRL_B1(m)	((m)->vsync_start - (m)->vdisplay)
+
+/* Command2, BK1 bytes */
+#define DSI_CMD2_BK1_VRHA_SET		0x45
+#define DSI_CMD2_BK1_VCOM_SET		0x13
+#define DSI_CMD2_BK1_VGHSS_SET		GENMASK(2, 0)
+#define DSI_CMD2_BK1_TESTCMD_VAL	BIT(7)
+#define DSI_VGLS_DEFAULT		BIT(6)
+#define DSI_VGLS_SEL			GENMASK(2, 0)
+#define DSI_CMD2_BK1_VGLS_SET		(DSI_VGLS_DEFAULT | DSI_VGLS_SEL)
+#define DSI_PWCTLR1_AP			BIT(7) /* Gamma OP bias, max */
+#define DSI_PWCTLR1_APIS		BIT(2) /* Source OP input bias, min */
+#define DSI_PWCTLR1_APOS		BIT(0) /* Source OP output bias, min */
+#define DSI_CMD2_BK1_PWCTLR1_SET	(DSI_PWCTLR1_AP | DSI_PWCTLR1_APIS | \
+					DSI_PWCTLR1_APOS)
+#define DSI_PWCTLR2_AVDD		BIT(5) /* AVDD 6.6v */
+#define DSI_PWCTLR2_AVCL		0x0    /* AVCL -4.4v */
+#define DSI_CMD2_BK1_PWCTLR2_SET	(DSI_PWCTLR2_AVDD | DSI_PWCTLR2_AVCL)
+#define DSI_SPD1_T2D			BIT(3)
+#define DSI_CMD2_BK1_SPD1_SET		(GENMASK(6, 4) | DSI_SPD1_T2D)
+#define DSI_CMD2_BK1_SPD2_SET		DSI_CMD2_BK1_SPD1_SET
+#define DSI_MIPISET1_EOT_EN		BIT(3)
+#define DSI_CMD2_BK1_MIPISET1_SET	(BIT(7) | DSI_MIPISET1_EOT_EN)
+
+struct st7701_panel_desc {
+	const struct drm_display_mode *mode;
+	unsigned int lanes;
+	unsigned long flags;
+	enum mipi_dsi_pixel_format format;
+	const char *const *supply_names;
+	unsigned int num_supplies;
+	unsigned int panel_sleep_delay;
+};
+
+struct st7701 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	const struct st7701_panel_desc *desc;
+
+	struct backlight_device *backlight;
+	struct regulator_bulk_data *supplies;
+	struct gpio_desc *reset;
+	unsigned int sleep_delay;
+};
+
+static inline struct st7701 *panel_to_st7701(struct drm_panel *panel)
+{
+	return container_of(panel, struct st7701, panel);
+}
+
+static inline int st7701_dsi_write(struct st7701 *st7701, const void *seq,
+				   size_t len)
+{
+	return mipi_dsi_dcs_write_buffer(st7701->dsi, seq, len);
+}
+
+#define ST7701_DSI(st7701, seq...)				\
+	{							\
+		const u8 d[] = { seq };				\
+		st7701_dsi_write(st7701, d, ARRAY_SIZE(d));	\
+	}
+
+static void st7701_init_sequence(struct st7701 *st7701)
+{
+	const struct drm_display_mode *mode = st7701->desc->mode;
+
+	ST7701_DSI(st7701, MIPI_DCS_SOFT_RESET, 0x00);
+
+	/* We need to wait 5ms before sending new commands */
+	msleep(5);
+
+	ST7701_DSI(st7701, MIPI_DCS_EXIT_SLEEP_MODE, 0x00);
+
+	msleep(st7701->sleep_delay);
+
+	/* Command2, BK0 */
+	ST7701_DSI(st7701, DSI_CMD2BKX_SEL,
+		   0x77, 0x01, 0x00, 0x00, DSI_CMD2BK0_SEL);
+	ST7701_DSI(st7701, DSI_CMD2_BK0_PVGAMCTRL, 0x00, 0x0E, 0x15, 0x0F,
+		   0x11, 0x08, 0x08, 0x08, 0x08, 0x23, 0x04, 0x13, 0x12,
+		   0x2B, 0x34, 0x1F);
+	ST7701_DSI(st7701, DSI_CMD2_BK0_NVGAMCTRL, 0x00, 0x0E, 0x95, 0x0F,
+		   0x13, 0x07, 0x09, 0x08, 0x08, 0x22, 0x04, 0x10, 0x0E,
+		   0x2C, 0x34, 0x1F);
+	ST7701_DSI(st7701, DSI_CMD2_BK0_LNESET,
+		   DSI_CMD2_BK0_LNESET_B0, DSI_CMD2_BK0_LNESET_B1);
+	ST7701_DSI(st7701, DSI_CMD2_BK0_PORCTRL,
+		   DSI_CMD2_BK0_PORCTRL_B0(mode),
+		   DSI_CMD2_BK0_PORCTRL_B1(mode));
+	ST7701_DSI(st7701, DSI_CMD2_BK0_INVSEL,
+		   DSI_CMD2_BK0_INVSEL_B0, DSI_CMD2_BK0_INVSEL_B1);
+
+	/* Command2, BK1 */
+	ST7701_DSI(st7701, DSI_CMD2BKX_SEL,
+			0x77, 0x01, 0x00, 0x00, DSI_CMD2BK1_SEL);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_VRHS, DSI_CMD2_BK1_VRHA_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_VCOM, DSI_CMD2_BK1_VCOM_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_VGHSS, DSI_CMD2_BK1_VGHSS_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_TESTCMD, DSI_CMD2_BK1_TESTCMD_VAL);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_VGLS, DSI_CMD2_BK1_VGLS_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_PWCTLR1, DSI_CMD2_BK1_PWCTLR1_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_PWCTLR2, DSI_CMD2_BK1_PWCTLR2_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_SPD1, DSI_CMD2_BK1_SPD1_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_SPD2, DSI_CMD2_BK1_SPD2_SET);
+	ST7701_DSI(st7701, DSI_CMD2_BK1_MIPISET1, DSI_CMD2_BK1_MIPISET1_SET);
+
+	/**
+	 * ST7701_SPEC_V1.2 is unable to provide enough information above this
+	 * specific command sequence, so grab the same from vendor BSP driver.
+	 */
+	ST7701_DSI(st7701, 0xE0, 0x00, 0x00, 0x02);
+	ST7701_DSI(st7701, 0xE1, 0x0B, 0x00, 0x0D, 0x00, 0x0C, 0x00, 0x0E,
+		   0x00, 0x00, 0x44, 0x44);
+	ST7701_DSI(st7701, 0xE2, 0x33, 0x33, 0x44, 0x44, 0x64, 0x00, 0x66,
+		   0x00, 0x65, 0x00, 0x67, 0x00, 0x00);
+	ST7701_DSI(st7701, 0xE3, 0x00, 0x00, 0x33, 0x33);
+	ST7701_DSI(st7701, 0xE4, 0x44, 0x44);
+	ST7701_DSI(st7701, 0xE5, 0x0C, 0x78, 0x3C, 0xA0, 0x0E, 0x78, 0x3C,
+		   0xA0, 0x10, 0x78, 0x3C, 0xA0, 0x12, 0x78, 0x3C, 0xA0);
+	ST7701_DSI(st7701, 0xE6, 0x00, 0x00, 0x33, 0x33);
+	ST7701_DSI(st7701, 0xE7, 0x44, 0x44);
+	ST7701_DSI(st7701, 0xE8, 0x0D, 0x78, 0x3C, 0xA0, 0x0F, 0x78, 0x3C,
+		   0xA0, 0x11, 0x78, 0x3C, 0xA0, 0x13, 0x78, 0x3C, 0xA0);
+	ST7701_DSI(st7701, 0xEB, 0x02, 0x02, 0x39, 0x39, 0xEE, 0x44, 0x00);
+	ST7701_DSI(st7701, 0xEC, 0x00, 0x00);
+	ST7701_DSI(st7701, 0xED, 0xFF, 0xF1, 0x04, 0x56, 0x72, 0x3F, 0xFF,
+		   0xFF, 0xFF, 0xFF, 0xF3, 0x27, 0x65, 0x40, 0x1F, 0xFF);
+
+	/* disable Command2 */
+	ST7701_DSI(st7701, DSI_CMD2BKX_SEL,
+		   0x77, 0x01, 0x00, 0x00, DSI_CMD2BKX_SEL_NONE);
+}
+
+static int st7701_prepare(struct drm_panel *panel)
+{
+	struct st7701 *st7701 = panel_to_st7701(panel);
+	int ret;
+
+	gpiod_set_value(st7701->reset, 0);
+
+	ret = regulator_bulk_enable(st7701->desc->num_supplies,
+				    st7701->supplies);
+	if (ret < 0)
+		return ret;
+	msleep(20);
+
+	gpiod_set_value(st7701->reset, 1);
+	msleep(150);
+
+	st7701_init_sequence(st7701);
+
+	return 0;
+}
+
+static int st7701_enable(struct drm_panel *panel)
+{
+	struct st7701 *st7701 = panel_to_st7701(panel);
+
+	ST7701_DSI(st7701, MIPI_DCS_SET_DISPLAY_ON, 0x00);
+	backlight_enable(st7701->backlight);
+
+	return 0;
+}
+
+static int st7701_disable(struct drm_panel *panel)
+{
+	struct st7701 *st7701 = panel_to_st7701(panel);
+
+	backlight_disable(st7701->backlight);
+	ST7701_DSI(st7701, MIPI_DCS_SET_DISPLAY_OFF, 0x00);
+
+	return 0;
+}
+
+static int st7701_unprepare(struct drm_panel *panel)
+{
+	struct st7701 *st7701 = panel_to_st7701(panel);
+
+	ST7701_DSI(st7701, MIPI_DCS_ENTER_SLEEP_MODE, 0x00);
+
+	msleep(st7701->sleep_delay);
+
+	gpiod_set_value(st7701->reset, 0);
+
+	/**
+	 * During the Resetting period, the display will be blanked
+	 * (The display is entering blanking sequence, which maximum
+	 * time is 120 ms, when Reset Starts in Sleep Out –mode. The
+	 * display remains the blank state in Sleep In –mode.) and
+	 * then return to Default condition for Hardware Reset.
+	 *
+	 * So we need wait sleep_delay time to make sure reset completed.
+	 */
+	msleep(st7701->sleep_delay);
+
+	regulator_bulk_disable(st7701->desc->num_supplies, st7701->supplies);
+
+	return 0;
+}
+
+static int st7701_get_modes(struct drm_panel *panel)
+{
+	struct st7701 *st7701 = panel_to_st7701(panel);
+	const struct drm_display_mode *desc_mode = st7701->desc->mode;
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(panel->drm, desc_mode);
+	if (!mode) {
+		DRM_DEV_ERROR(&st7701->dsi->dev,
+			      "failed to add mode %ux%ux@%u\n",
+			      desc_mode->hdisplay, desc_mode->vdisplay,
+			      desc_mode->vrefresh);
+		return -ENOMEM;
+	}
+
+	drm_mode_set_name(mode);
+	drm_mode_probed_add(panel->connector, mode);
+
+	panel->connector->display_info.width_mm = desc_mode->width_mm;
+	panel->connector->display_info.height_mm = desc_mode->height_mm;
+
+	return 1;
+}
+
+static const struct drm_panel_funcs st7701_funcs = {
+	.disable	= st7701_disable,
+	.unprepare	= st7701_unprepare,
+	.prepare	= st7701_prepare,
+	.enable		= st7701_enable,
+	.get_modes	= st7701_get_modes,
+};
+
+static const struct drm_display_mode ts8550b_mode = {
+	.clock		= 27500,
+
+	.hdisplay	= 480,
+	.hsync_start	= 480 + 38,
+	.hsync_end	= 480 + 38 + 12,
+	.htotal		= 480 + 38 + 12 + 12,
+
+	.vdisplay	= 854,
+	.vsync_start	= 854 + 4,
+	.vsync_end	= 854 + 4 + 8,
+	.vtotal		= 854 + 4 + 8 + 18,
+
+	.width_mm	= 69,
+	.height_mm	= 139,
+
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static const char * const ts8550b_supply_names[] = {
+	"VCC",
+	"IOVCC",
+};
+
+static const struct st7701_panel_desc ts8550b_desc = {
+	.mode = &ts8550b_mode,
+	.lanes = 2,
+	.flags = MIPI_DSI_MODE_VIDEO,
+	.format = MIPI_DSI_FMT_RGB888,
+	.supply_names = ts8550b_supply_names,
+	.num_supplies = ARRAY_SIZE(ts8550b_supply_names),
+	.panel_sleep_delay = 80, /* panel need extra 80ms for sleep out cmd */
+};
+
+static int st7701_dsi_probe(struct mipi_dsi_device *dsi)
+{
+	const struct st7701_panel_desc *desc;
+	struct st7701 *st7701;
+	int ret, i;
+
+	st7701 = devm_kzalloc(&dsi->dev, sizeof(*st7701), GFP_KERNEL);
+	if (!st7701)
+		return -ENOMEM;
+
+	desc = of_device_get_match_data(&dsi->dev);
+	dsi->mode_flags = desc->flags;
+	dsi->format = desc->format;
+	dsi->lanes = desc->lanes;
+
+	st7701->supplies = devm_kcalloc(&dsi->dev, desc->num_supplies,
+					sizeof(*st7701->supplies),
+					GFP_KERNEL);
+	if (!st7701->supplies)
+		return -ENOMEM;
+
+	for (i = 0; i < desc->num_supplies; i++)
+		st7701->supplies[i].supply = desc->supply_names[i];
+
+	ret = devm_regulator_bulk_get(&dsi->dev, desc->num_supplies,
+				      st7701->supplies);
+	if (ret < 0)
+		return ret;
+
+	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");
+		return PTR_ERR(st7701->reset);
+	}
+
+	st7701->backlight = devm_of_find_backlight(&dsi->dev);
+	if (IS_ERR(st7701->backlight))
+		return PTR_ERR(st7701->backlight);
+
+	drm_panel_init(&st7701->panel);
+
+	/**
+	 * Once sleep out has been issued, ST7701 IC required to wait 120ms
+	 * before initiating new commands.
+	 *
+	 * On top of that some panels might need an extra delay to wait, so
+	 * add panel specific delay for those cases. As now this panel specific
+	 * delay information is referenced from those panel BSP driver, example
+	 * ts8550b and there is no valid documentation for that.
+	 */
+	st7701->sleep_delay = 120 + desc->panel_sleep_delay;
+	st7701->panel.funcs = &st7701_funcs;
+	st7701->panel.dev = &dsi->dev;
+
+	ret = drm_panel_add(&st7701->panel);
+	if (ret < 0)
+		return ret;
+
+	mipi_dsi_set_drvdata(dsi, st7701);
+	st7701->dsi = dsi;
+	st7701->desc = desc;
+
+	return mipi_dsi_attach(dsi);
+}
+
+static int st7701_dsi_remove(struct mipi_dsi_device *dsi)
+{
+	struct st7701 *st7701 = mipi_dsi_get_drvdata(dsi);
+
+	mipi_dsi_detach(dsi);
+	drm_panel_remove(&st7701->panel);
+
+	return 0;
+}
+
+static const struct of_device_id st7701_of_match[] = {
+	{ .compatible = "techstar,ts8550b", .data = &ts8550b_desc },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, st7701_of_match);
+
+static struct mipi_dsi_driver st7701_dsi_driver = {
+	.probe		= st7701_dsi_probe,
+	.remove		= st7701_dsi_remove,
+	.driver = {
+		.name		= "st7701",
+		.of_match_table	= st7701_of_match,
+	},
+};
+module_mipi_dsi_driver(st7701_dsi_driver);
+
+MODULE_AUTHOR("Jagan Teki <jagan@amarulasolutions.com>");
+MODULE_DESCRIPTION("Sitronix ST7701 LCD Panel Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/pl111/pl111_drv.c b/drivers/gpu/drm/pl111/pl111_drv.c
index 33e0483d62ae..a8958c201a88 100644
--- a/drivers/gpu/drm/pl111/pl111_drv.c
+++ b/drivers/gpu/drm/pl111/pl111_drv.c
@@ -64,14 +64,14 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_of.h>
-#include <drm/drm_bridge.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "pl111_drm.h"
 #include "pl111_versatile.h"
diff --git a/drivers/gpu/drm/qxl/Makefile b/drivers/gpu/drm/qxl/Makefile
index 33a7d0c434b7..fc59d42b31af 100644
--- a/drivers/gpu/drm/qxl/Makefile
+++ b/drivers/gpu/drm/qxl/Makefile
@@ -2,6 +2,6 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-qxl-y := qxl_drv.o qxl_kms.o qxl_display.o qxl_ttm.o qxl_fb.o qxl_object.o qxl_gem.o qxl_cmd.o qxl_image.o qxl_draw.o qxl_debugfs.o qxl_irq.o qxl_dumb.o qxl_ioctl.o qxl_release.o qxl_prime.o
+qxl-y := qxl_drv.o qxl_kms.o qxl_display.o qxl_ttm.o qxl_object.o qxl_gem.o qxl_cmd.o qxl_image.o qxl_draw.o qxl_debugfs.o qxl_irq.o qxl_dumb.o qxl_ioctl.o qxl_release.o qxl_prime.o
 
 obj-$(CONFIG_DRM_QXL)+= qxl.o
diff --git a/drivers/gpu/drm/qxl/qxl_cmd.c b/drivers/gpu/drm/qxl/qxl_cmd.c
index 2e100f644236..0a2e51af1230 100644
--- a/drivers/gpu/drm/qxl/qxl_cmd.c
+++ b/drivers/gpu/drm/qxl/qxl_cmd.c
@@ -374,25 +374,25 @@ void qxl_io_flush_surfaces(struct qxl_device *qdev)
 void qxl_io_destroy_primary(struct qxl_device *qdev)
 {
 	wait_for_io_cmd(qdev, 0, QXL_IO_DESTROY_PRIMARY_ASYNC);
-	qdev->primary_created = false;
+	qdev->primary_bo->is_primary = false;
+	drm_gem_object_put_unlocked(&qdev->primary_bo->gem_base);
+	qdev->primary_bo = NULL;
 }
 
-void qxl_io_create_primary(struct qxl_device *qdev,
-			   unsigned int offset, struct qxl_bo *bo)
+void qxl_io_create_primary(struct qxl_device *qdev, struct qxl_bo *bo)
 {
 	struct qxl_surface_create *create;
 
+	if (WARN_ON(qdev->primary_bo))
+		return;
+
 	DRM_DEBUG_DRIVER("qdev %p, ram_header %p\n", qdev, qdev->ram_header);
 	create = &qdev->ram_header->create_surface;
 	create->format = bo->surf.format;
 	create->width = bo->surf.width;
 	create->height = bo->surf.height;
 	create->stride = bo->surf.stride;
-	if (bo->shadow) {
-		create->mem = qxl_bo_physical_address(qdev, bo->shadow, offset);
-	} else {
-		create->mem = qxl_bo_physical_address(qdev, bo, offset);
-	}
+	create->mem = qxl_bo_physical_address(qdev, bo, 0);
 
 	DRM_DEBUG_DRIVER("mem = %llx, from %p\n", create->mem, bo->kptr);
 
@@ -400,7 +400,9 @@ void qxl_io_create_primary(struct qxl_device *qdev,
 	create->type = QXL_SURF_TYPE_PRIMARY;
 
 	wait_for_io_cmd(qdev, 0, QXL_IO_CREATE_PRIMARY_ASYNC);
-	qdev->primary_created = true;
+	qdev->primary_bo = bo;
+	qdev->primary_bo->is_primary = true;
+	drm_gem_object_get(&qdev->primary_bo->gem_base);
 }
 
 void qxl_io_memslot_add(struct qxl_device *qdev, uint8_t id)
@@ -460,8 +462,7 @@ void qxl_surface_id_dealloc(struct qxl_device *qdev,
 }
 
 int qxl_hw_surface_alloc(struct qxl_device *qdev,
-			 struct qxl_bo *surf,
-			 struct ttm_mem_reg *new_mem)
+			 struct qxl_bo *surf)
 {
 	struct qxl_surface_cmd *cmd;
 	struct qxl_release *release;
@@ -487,16 +488,7 @@ int qxl_hw_surface_alloc(struct qxl_device *qdev,
 	cmd->u.surface_create.width = surf->surf.width;
 	cmd->u.surface_create.height = surf->surf.height;
 	cmd->u.surface_create.stride = surf->surf.stride;
-	if (new_mem) {
-		int slot_id = surf->type == QXL_GEM_DOMAIN_VRAM ? qdev->main_mem_slot : qdev->surfaces_mem_slot;
-		struct qxl_memslot *slot = &(qdev->mem_slots[slot_id]);
-
-		/* TODO - need to hold one of the locks to read tbo.offset */
-		cmd->u.surface_create.data = slot->high_bits;
-
-		cmd->u.surface_create.data |= (new_mem->start << PAGE_SHIFT) + surf->tbo.bdev->man[new_mem->mem_type].gpu_offset;
-	} else
-		cmd->u.surface_create.data = qxl_bo_physical_address(qdev, surf, 0);
+	cmd->u.surface_create.data = qxl_bo_physical_address(qdev, surf, 0);
 	cmd->surface_id = surf->surface_id;
 	qxl_release_unmap(qdev, release, &cmd->release_info);
 
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 1f8fddcc34d6..08c725544a2f 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -24,11 +24,11 @@
  */
 
 #include <linux/crc32.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_atomic_helper.h>
 #include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "qxl_drv.h"
 #include "qxl_object.h"
@@ -80,10 +80,10 @@ static int qxl_display_copy_rom_client_monitors_config(struct qxl_device *qdev)
 		DRM_DEBUG_KMS("no client monitors configured\n");
 		return status;
 	}
-	if (num_monitors > qdev->monitors_config->max_allowed) {
+	if (num_monitors > qxl_num_crtc) {
 		DRM_DEBUG_KMS("client monitors list will be truncated: %d < %d\n",
-			      qdev->monitors_config->max_allowed, num_monitors);
-		num_monitors = qdev->monitors_config->max_allowed;
+			      qxl_num_crtc, num_monitors);
+		num_monitors = qxl_num_crtc;
 	} else {
 		num_monitors = qdev->rom->client_monitors_config.count;
 	}
@@ -96,8 +96,7 @@ static int qxl_display_copy_rom_client_monitors_config(struct qxl_device *qdev)
 		return status;
 	}
 	/* we copy max from the client but it isn't used */
-	qdev->client_monitors_config->max_allowed =
-				qdev->monitors_config->max_allowed;
+	qdev->client_monitors_config->max_allowed = qxl_num_crtc;
 	for (i = 0 ; i < qdev->client_monitors_config->count ; ++i) {
 		struct qxl_urect *c_rect =
 			&qdev->rom->client_monitors_config.heads[i];
@@ -191,20 +190,63 @@ void qxl_display_read_client_monitors_config(struct qxl_device *qdev)
 	}
 }
 
-static int qxl_add_monitors_config_modes(struct drm_connector *connector,
-                                         unsigned *pwidth,
-                                         unsigned *pheight)
+static int qxl_check_mode(struct qxl_device *qdev,
+			  unsigned int width,
+			  unsigned int height)
+{
+	unsigned int stride;
+	unsigned int size;
+
+	if (check_mul_overflow(width, 4u, &stride))
+		return -EINVAL;
+	if (check_mul_overflow(stride, height, &size))
+		return -EINVAL;
+	if (size > qdev->vram_size)
+		return -ENOMEM;
+	return 0;
+}
+
+static int qxl_check_framebuffer(struct qxl_device *qdev,
+				 struct qxl_bo *bo)
+{
+	return qxl_check_mode(qdev, bo->surf.width, bo->surf.height);
+}
+
+static int qxl_add_mode(struct drm_connector *connector,
+			unsigned int width,
+			unsigned int height,
+			bool preferred)
+{
+	struct drm_device *dev = connector->dev;
+	struct qxl_device *qdev = dev->dev_private;
+	struct drm_display_mode *mode = NULL;
+	int rc;
+
+	rc = qxl_check_mode(qdev, width, height);
+	if (rc != 0)
+		return 0;
+
+	mode = drm_cvt_mode(dev, width, height, 60, false, false, false);
+	if (preferred)
+		mode->type |= DRM_MODE_TYPE_PREFERRED;
+	mode->hdisplay = width;
+	mode->vdisplay = height;
+	drm_mode_set_name(mode);
+	drm_mode_probed_add(connector, mode);
+	return 1;
+}
+
+static int qxl_add_monitors_config_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
 	struct qxl_device *qdev = dev->dev_private;
 	struct qxl_output *output = drm_connector_to_qxl_output(connector);
 	int h = output->index;
-	struct drm_display_mode *mode = NULL;
 	struct qxl_head *head;
 
 	if (!qdev->monitors_config)
 		return 0;
-	if (h >= qdev->monitors_config->max_allowed)
+	if (h >= qxl_num_crtc)
 		return 0;
 	if (!qdev->client_monitors_config)
 		return 0;
@@ -214,60 +256,28 @@ static int qxl_add_monitors_config_modes(struct drm_connector *connector,
 	head = &qdev->client_monitors_config->heads[h];
 	DRM_DEBUG_KMS("head %d is %dx%d\n", h, head->width, head->height);
 
-	mode = drm_cvt_mode(dev, head->width, head->height, 60, false, false,
-			    false);
-	mode->type |= DRM_MODE_TYPE_PREFERRED;
-	mode->hdisplay = head->width;
-	mode->vdisplay = head->height;
-	drm_mode_set_name(mode);
-	*pwidth = head->width;
-	*pheight = head->height;
-	drm_mode_probed_add(connector, mode);
-	/* remember the last custom size for mode validation */
-	qdev->monitors_config_width = mode->hdisplay;
-	qdev->monitors_config_height = mode->vdisplay;
-	return 1;
+	return qxl_add_mode(connector, head->width, head->height, true);
 }
 
 static struct mode_size {
 	int w;
 	int h;
-} common_modes[] = {
-	{ 640,  480},
+} extra_modes[] = {
 	{ 720,  480},
-	{ 800,  600},
-	{ 848,  480},
-	{1024,  768},
 	{1152,  768},
-	{1280,  720},
-	{1280,  800},
 	{1280,  854},
-	{1280,  960},
-	{1280, 1024},
-	{1440,  900},
-	{1400, 1050},
-	{1680, 1050},
-	{1600, 1200},
-	{1920, 1080},
-	{1920, 1200}
 };
 
-static int qxl_add_common_modes(struct drm_connector *connector,
-                                unsigned int pwidth,
-                                unsigned int pheight)
+static int qxl_add_extra_modes(struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-	struct drm_display_mode *mode = NULL;
-	int i;
+	int i, ret = 0;
 
-	for (i = 0; i < ARRAY_SIZE(common_modes); i++) {
-		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h,
-				    60, false, false, false);
-		if (common_modes[i].w == pwidth && common_modes[i].h == pheight)
-			mode->type |= DRM_MODE_TYPE_PREFERRED;
-		drm_mode_probed_add(connector, mode);
-	}
-	return i - 1;
+	for (i = 0; i < ARRAY_SIZE(extra_modes); i++)
+		ret += qxl_add_mode(connector,
+				    extra_modes[i].w,
+				    extra_modes[i].h,
+				    false);
+	return ret;
 }
 
 static void qxl_send_monitors_config(struct qxl_device *qdev)
@@ -302,13 +312,12 @@ static void qxl_crtc_update_monitors_config(struct drm_crtc *crtc,
 	struct qxl_head head;
 	int oldcount, i = qcrtc->index;
 
-	if (!qdev->primary_created) {
+	if (!qdev->primary_bo) {
 		DRM_DEBUG_KMS("no primary surface, skip (%s)\n", reason);
 		return;
 	}
 
-	if (!qdev->monitors_config ||
-	    qdev->monitors_config->max_allowed <= i)
+	if (!qdev->monitors_config || qxl_num_crtc <= i)
 		return;
 
 	head.id = i;
@@ -323,6 +332,8 @@ static void qxl_crtc_update_monitors_config(struct drm_crtc *crtc,
 		head.y = crtc->y;
 		if (qdev->monitors_config->count < i + 1)
 			qdev->monitors_config->count = i + 1;
+		if (qdev->primary_bo == qdev->dumb_shadow_bo)
+			head.x += qdev->dumb_heads[i].x;
 	} else if (i > 0) {
 		head.width = 0;
 		head.height = 0;
@@ -348,9 +359,10 @@ static void qxl_crtc_update_monitors_config(struct drm_crtc *crtc,
 	if (oldcount != qdev->monitors_config->count)
 		DRM_DEBUG_KMS("active heads %d -> %d (%d total)\n",
 			      oldcount, qdev->monitors_config->count,
-			      qdev->monitors_config->max_allowed);
+			      qxl_num_crtc);
 
 	qdev->monitors_config->heads[i] = head;
+	qdev->monitors_config->max_allowed = qxl_num_crtc;
 	qxl_send_monitors_config(qdev);
 }
 
@@ -401,13 +413,15 @@ static int qxl_framebuffer_surface_dirty(struct drm_framebuffer *fb,
 	struct qxl_device *qdev = fb->dev->dev_private;
 	struct drm_clip_rect norect;
 	struct qxl_bo *qobj;
+	bool is_primary;
 	int inc = 1;
 
 	drm_modeset_lock_all(fb->dev);
 
 	qobj = gem_to_qxl_bo(fb->obj[0]);
 	/* if we aren't primary surface ignore this */
-	if (!qobj->is_primary) {
+	is_primary = qobj->shadow ? qobj->shadow->is_primary : qobj->is_primary;
+	if (!is_primary) {
 		drm_modeset_unlock_all(fb->dev);
 		return 0;
 	}
@@ -424,7 +438,7 @@ static int qxl_framebuffer_surface_dirty(struct drm_framebuffer *fb,
 	}
 
 	qxl_draw_dirty_fb(qdev, fb, qobj, flags, color,
-			  clips, num_clips, inc);
+			  clips, num_clips, inc, 0);
 
 	drm_modeset_unlock_all(fb->dev);
 
@@ -466,12 +480,7 @@ static int qxl_primary_atomic_check(struct drm_plane *plane,
 
 	bo = gem_to_qxl_bo(state->fb->obj[0]);
 
-	if (bo->surf.stride * bo->surf.height > qdev->vram_size) {
-		DRM_ERROR("Mode doesn't fit in vram size (vgamem)");
-		return -EINVAL;
-	}
-
-	return 0;
+	return qxl_check_framebuffer(qdev, bo);
 }
 
 static int qxl_primary_apply_cursor(struct drm_plane *plane)
@@ -526,15 +535,14 @@ static void qxl_primary_atomic_update(struct drm_plane *plane,
 {
 	struct qxl_device *qdev = plane->dev->dev_private;
 	struct qxl_bo *bo = gem_to_qxl_bo(plane->state->fb->obj[0]);
-	struct qxl_bo *bo_old;
+	struct qxl_bo *bo_old, *primary;
 	struct drm_clip_rect norect = {
 	    .x1 = 0,
 	    .y1 = 0,
 	    .x2 = plane->state->fb->width,
 	    .y2 = plane->state->fb->height
 	};
-	int ret;
-	bool same_shadow = false;
+	uint32_t dumb_shadow_offset = 0;
 
 	if (old_state->fb) {
 		bo_old = gem_to_qxl_bo(old_state->fb->obj[0]);
@@ -542,32 +550,21 @@ static void qxl_primary_atomic_update(struct drm_plane *plane,
 		bo_old = NULL;
 	}
 
-	if (bo == bo_old)
-		return;
+	primary = bo->shadow ? bo->shadow : bo;
 
-	if (bo_old && bo_old->shadow && bo->shadow &&
-	    bo_old->shadow == bo->shadow) {
-		same_shadow = true;
-	}
-
-	if (bo_old && bo_old->is_primary) {
-		if (!same_shadow)
+	if (!primary->is_primary) {
+		if (qdev->primary_bo)
 			qxl_io_destroy_primary(qdev);
-		bo_old->is_primary = false;
-
-		ret = qxl_primary_apply_cursor(plane);
-		if (ret)
-			DRM_ERROR(
-			"could not set cursor after creating primary");
+		qxl_io_create_primary(qdev, primary);
+		qxl_primary_apply_cursor(plane);
 	}
 
-	if (!bo->is_primary) {
-		if (!same_shadow)
-			qxl_io_create_primary(qdev, 0, bo);
-		bo->is_primary = true;
-	}
+	if (bo->is_dumb)
+		dumb_shadow_offset =
+			qdev->dumb_heads[plane->state->crtc->index].x;
 
-	qxl_draw_dirty_fb(qdev, plane->state->fb, bo, 0, 0, &norect, 1, 1);
+	qxl_draw_dirty_fb(qdev, plane->state->fb, bo, 0, 0, &norect, 1, 1,
+			  dumb_shadow_offset);
 }
 
 static void qxl_primary_atomic_disable(struct drm_plane *plane,
@@ -723,12 +720,68 @@ static void qxl_cursor_atomic_disable(struct drm_plane *plane,
 	qxl_release_fence_buffer_objects(release);
 }
 
+static void qxl_update_dumb_head(struct qxl_device *qdev,
+				 int index, struct qxl_bo *bo)
+{
+	uint32_t width, height;
+
+	if (index >= qdev->monitors_config->max_allowed)
+		return;
+
+	if (bo && bo->is_dumb) {
+		width = bo->surf.width;
+		height = bo->surf.height;
+	} else {
+		width = 0;
+		height = 0;
+	}
+
+	if (qdev->dumb_heads[index].width == width &&
+	    qdev->dumb_heads[index].height == height)
+		return;
+
+	DRM_DEBUG("#%d: %dx%d -> %dx%d\n", index,
+		  qdev->dumb_heads[index].width,
+		  qdev->dumb_heads[index].height,
+		  width, height);
+	qdev->dumb_heads[index].width = width;
+	qdev->dumb_heads[index].height = height;
+}
+
+static void qxl_calc_dumb_shadow(struct qxl_device *qdev,
+				 struct qxl_surface *surf)
+{
+	struct qxl_head *head;
+	int i;
+
+	memset(surf, 0, sizeof(*surf));
+	for (i = 0; i < qdev->monitors_config->max_allowed; i++) {
+		head = qdev->dumb_heads + i;
+		head->x = surf->width;
+		surf->width += head->width;
+		if (surf->height < head->height)
+			surf->height = head->height;
+	}
+	if (surf->width < 64)
+		surf->width = 64;
+	if (surf->height < 64)
+		surf->height = 64;
+	surf->format = SPICE_SURFACE_FMT_32_xRGB;
+	surf->stride = surf->width * 4;
+
+	if (!qdev->dumb_shadow_bo ||
+	    qdev->dumb_shadow_bo->surf.width != surf->width ||
+	    qdev->dumb_shadow_bo->surf.height != surf->height)
+		DRM_DEBUG("%dx%d\n", surf->width, surf->height);
+}
+
 static int qxl_plane_prepare_fb(struct drm_plane *plane,
 				struct drm_plane_state *new_state)
 {
 	struct qxl_device *qdev = plane->dev->dev_private;
 	struct drm_gem_object *obj;
-	struct qxl_bo *user_bo, *old_bo = NULL;
+	struct qxl_bo *user_bo;
+	struct qxl_surface surf;
 	int ret;
 
 	if (!new_state->fb)
@@ -738,28 +791,30 @@ static int qxl_plane_prepare_fb(struct drm_plane *plane,
 	user_bo = gem_to_qxl_bo(obj);
 
 	if (plane->type == DRM_PLANE_TYPE_PRIMARY &&
-	    user_bo->is_dumb && !user_bo->shadow) {
-		if (plane->state->fb) {
-			obj = plane->state->fb->obj[0];
-			old_bo = gem_to_qxl_bo(obj);
+	    user_bo->is_dumb) {
+		qxl_update_dumb_head(qdev, new_state->crtc->index,
+				     user_bo);
+		qxl_calc_dumb_shadow(qdev, &surf);
+		if (!qdev->dumb_shadow_bo ||
+		    qdev->dumb_shadow_bo->surf.width  != surf.width ||
+		    qdev->dumb_shadow_bo->surf.height != surf.height) {
+			if (qdev->dumb_shadow_bo) {
+				drm_gem_object_put_unlocked
+					(&qdev->dumb_shadow_bo->gem_base);
+				qdev->dumb_shadow_bo = NULL;
+			}
+			qxl_bo_create(qdev, surf.height * surf.stride,
+				      true, true, QXL_GEM_DOMAIN_SURFACE, &surf,
+				      &qdev->dumb_shadow_bo);
 		}
-		if (old_bo && old_bo->shadow &&
-		    user_bo->gem_base.size == old_bo->gem_base.size &&
-		    plane->state->crtc     == new_state->crtc &&
-		    plane->state->crtc_w   == new_state->crtc_w &&
-		    plane->state->crtc_h   == new_state->crtc_h &&
-		    plane->state->src_x    == new_state->src_x &&
-		    plane->state->src_y    == new_state->src_y &&
-		    plane->state->src_w    == new_state->src_w &&
-		    plane->state->src_h    == new_state->src_h &&
-		    plane->state->rotation == new_state->rotation &&
-		    plane->state->zpos     == new_state->zpos) {
-			drm_gem_object_get(&old_bo->shadow->gem_base);
-			user_bo->shadow = old_bo->shadow;
-		} else {
-			qxl_bo_create(qdev, user_bo->gem_base.size,
-				      true, true, QXL_GEM_DOMAIN_VRAM, NULL,
-				      &user_bo->shadow);
+		if (user_bo->shadow != qdev->dumb_shadow_bo) {
+			if (user_bo->shadow) {
+				drm_gem_object_put_unlocked
+					(&user_bo->shadow->gem_base);
+				user_bo->shadow = NULL;
+			}
+			drm_gem_object_get(&qdev->dumb_shadow_bo->gem_base);
+			user_bo->shadow = qdev->dumb_shadow_bo;
 		}
 	}
 
@@ -788,7 +843,7 @@ static void qxl_plane_cleanup_fb(struct drm_plane *plane,
 	user_bo = gem_to_qxl_bo(obj);
 	qxl_bo_unpin(user_bo);
 
-	if (user_bo->shadow && !user_bo->is_primary) {
+	if (old_state->fb != plane->state->fb && user_bo->shadow) {
 		drm_gem_object_put_unlocked(&user_bo->shadow->gem_base);
 		user_bo->shadow = NULL;
 	}
@@ -925,14 +980,26 @@ free_mem:
 
 static int qxl_conn_get_modes(struct drm_connector *connector)
 {
+	struct drm_device *dev = connector->dev;
+	struct qxl_device *qdev = dev->dev_private;
+	struct qxl_output *output = drm_connector_to_qxl_output(connector);
 	unsigned int pwidth = 1024;
 	unsigned int pheight = 768;
 	int ret = 0;
 
-	ret = qxl_add_monitors_config_modes(connector, &pwidth, &pheight);
-	if (ret < 0)
-		return ret;
-	ret += qxl_add_common_modes(connector, pwidth, pheight);
+	if (qdev->client_monitors_config) {
+		struct qxl_head *head;
+		head = &qdev->client_monitors_config->heads[output->index];
+		if (head->width)
+			pwidth = head->width;
+		if (head->height)
+			pheight = head->height;
+	}
+
+	ret += drm_add_modes_noedid(connector, 8192, 8192);
+	ret += qxl_add_extra_modes(connector);
+	ret += qxl_add_monitors_config_modes(connector);
+	drm_set_preferred_mode(connector, pwidth, pheight);
 	return ret;
 }
 
@@ -941,20 +1008,11 @@ static enum drm_mode_status qxl_conn_mode_valid(struct drm_connector *connector,
 {
 	struct drm_device *ddev = connector->dev;
 	struct qxl_device *qdev = ddev->dev_private;
-	int i;
-
-	/* TODO: is this called for user defined modes? (xrandr --add-mode)
-	 * TODO: check that the mode fits in the framebuffer */
 
-	if (qdev->monitors_config_width == mode->hdisplay &&
-	    qdev->monitors_config_height == mode->vdisplay)
-		return MODE_OK;
+	if (qxl_check_mode(qdev, mode->hdisplay, mode->vdisplay) != 0)
+		return MODE_BAD;
 
-	for (i = 0; i < ARRAY_SIZE(common_modes); i++) {
-		if (common_modes[i].w == mode->hdisplay && common_modes[i].h == mode->vdisplay)
-			return MODE_OK;
-	}
-	return MODE_BAD;
+	return MODE_OK;
 }
 
 static struct drm_encoder *qxl_best_encoder(struct drm_connector *connector)
@@ -1096,9 +1154,8 @@ int qxl_create_monitors_object(struct qxl_device *qdev)
 {
 	int ret;
 	struct drm_gem_object *gobj;
-	int max_allowed = qxl_num_crtc;
 	int monitors_config_size = sizeof(struct qxl_monitors_config) +
-		max_allowed * sizeof(struct qxl_head);
+		qxl_num_crtc * sizeof(struct qxl_head);
 
 	ret = qxl_gem_object_create(qdev, monitors_config_size, 0,
 				    QXL_GEM_DOMAIN_VRAM,
@@ -1120,7 +1177,12 @@ int qxl_create_monitors_object(struct qxl_device *qdev)
 		qxl_bo_physical_address(qdev, qdev->monitors_config_bo, 0);
 
 	memset(qdev->monitors_config, 0, monitors_config_size);
-	qdev->monitors_config->max_allowed = max_allowed;
+	qdev->dumb_heads = kcalloc(qxl_num_crtc, sizeof(qdev->dumb_heads[0]),
+				   GFP_KERNEL);
+	if (!qdev->dumb_heads) {
+		qxl_destroy_monitors_object(qdev);
+		return -ENOMEM;
+	}
 	return 0;
 }
 
@@ -1172,18 +1234,11 @@ int qxl_modeset_init(struct qxl_device *qdev)
 	qxl_display_read_client_monitors_config(qdev);
 
 	drm_mode_config_reset(&qdev->ddev);
-
-	/* primary surface must be created by this point, to allow
-	 * issuing command queue commands and having them read by
-	 * spice server. */
-	qxl_fbdev_init(qdev);
 	return 0;
 }
 
 void qxl_modeset_fini(struct qxl_device *qdev)
 {
-	qxl_fbdev_fini(qdev);
-
 	qxl_destroy_monitors_object(qdev);
 	drm_mode_config_cleanup(&qdev->ddev);
 }
diff --git a/drivers/gpu/drm/qxl/qxl_draw.c b/drivers/gpu/drm/qxl/qxl_draw.c
index c408bb83c7a9..97c3f1a95a32 100644
--- a/drivers/gpu/drm/qxl/qxl_draw.c
+++ b/drivers/gpu/drm/qxl/qxl_draw.c
@@ -109,152 +109,6 @@ make_drawable(struct qxl_device *qdev, int surface, uint8_t type,
 	return 0;
 }
 
-static int alloc_palette_object(struct qxl_device *qdev,
-				struct qxl_release *release,
-				struct qxl_bo **palette_bo)
-{
-	return qxl_alloc_bo_reserved(qdev, release,
-				     sizeof(struct qxl_palette) + sizeof(uint32_t) * 2,
-				     palette_bo);
-}
-
-static int qxl_palette_create_1bit(struct qxl_bo *palette_bo,
-				   struct qxl_release *release,
-				   const struct qxl_fb_image *qxl_fb_image)
-{
-	const struct fb_image *fb_image = &qxl_fb_image->fb_image;
-	uint32_t visual = qxl_fb_image->visual;
-	const uint32_t *pseudo_palette = qxl_fb_image->pseudo_palette;
-	struct qxl_palette *pal;
-	int ret;
-	uint32_t fgcolor, bgcolor;
-	static uint64_t unique; /* we make no attempt to actually set this
-				 * correctly globaly, since that would require
-				 * tracking all of our palettes. */
-	ret = qxl_bo_kmap(palette_bo, (void **)&pal);
-	if (ret)
-		return ret;
-	pal->num_ents = 2;
-	pal->unique = unique++;
-	if (visual == FB_VISUAL_TRUECOLOR || visual == FB_VISUAL_DIRECTCOLOR) {
-		/* NB: this is the only used branch currently. */
-		fgcolor = pseudo_palette[fb_image->fg_color];
-		bgcolor = pseudo_palette[fb_image->bg_color];
-	} else {
-		fgcolor = fb_image->fg_color;
-		bgcolor = fb_image->bg_color;
-	}
-	pal->ents[0] = bgcolor;
-	pal->ents[1] = fgcolor;
-	qxl_bo_kunmap(palette_bo);
-	return 0;
-}
-
-void qxl_draw_opaque_fb(const struct qxl_fb_image *qxl_fb_image,
-			int stride /* filled in if 0 */)
-{
-	struct qxl_device *qdev = qxl_fb_image->qdev;
-	struct qxl_drawable *drawable;
-	struct qxl_rect rect;
-	const struct fb_image *fb_image = &qxl_fb_image->fb_image;
-	int x = fb_image->dx;
-	int y = fb_image->dy;
-	int width = fb_image->width;
-	int height = fb_image->height;
-	const char *src = fb_image->data;
-	int depth = fb_image->depth;
-	struct qxl_release *release;
-	struct qxl_image *image;
-	int ret;
-	struct qxl_drm_image *dimage;
-	struct qxl_bo *palette_bo = NULL;
-
-	if (stride == 0)
-		stride = depth * width / 8;
-
-	ret = alloc_drawable(qdev, &release);
-	if (ret)
-		return;
-
-	ret = qxl_image_alloc_objects(qdev, release,
-				      &dimage,
-				      height, stride);
-	if (ret)
-		goto out_free_drawable;
-
-	if (depth == 1) {
-		ret = alloc_palette_object(qdev, release, &palette_bo);
-		if (ret)
-			goto out_free_image;
-	}
-
-	/* do a reservation run over all the objects we just allocated */
-	ret = qxl_release_reserve_list(release, true);
-	if (ret)
-		goto out_free_palette;
-
-	rect.left = x;
-	rect.right = x + width;
-	rect.top = y;
-	rect.bottom = y + height;
-
-	ret = make_drawable(qdev, 0, QXL_DRAW_COPY, &rect, release);
-	if (ret) {
-		qxl_release_backoff_reserve_list(release);
-		goto out_free_palette;
-	}
-
-	ret = qxl_image_init(qdev, release, dimage,
-			     (const uint8_t *)src, 0, 0,
-			     width, height, depth, stride);
-	if (ret) {
-		qxl_release_backoff_reserve_list(release);
-		qxl_release_free(qdev, release);
-		return;
-	}
-
-	if (depth == 1) {
-		void *ptr;
-
-		ret = qxl_palette_create_1bit(palette_bo, release, qxl_fb_image);
-
-		ptr = qxl_bo_kmap_atomic_page(qdev, dimage->bo, 0);
-		image = ptr;
-		image->u.bitmap.palette =
-			qxl_bo_physical_address(qdev, palette_bo, 0);
-		qxl_bo_kunmap_atomic_page(qdev, dimage->bo, ptr);
-	}
-
-	drawable = (struct qxl_drawable *)qxl_release_map(qdev, release);
-
-	drawable->u.copy.src_area.top = 0;
-	drawable->u.copy.src_area.bottom = height;
-	drawable->u.copy.src_area.left = 0;
-	drawable->u.copy.src_area.right = width;
-
-	drawable->u.copy.rop_descriptor = SPICE_ROPD_OP_PUT;
-	drawable->u.copy.scale_mode = 0;
-	drawable->u.copy.mask.flags = 0;
-	drawable->u.copy.mask.pos.x = 0;
-	drawable->u.copy.mask.pos.y = 0;
-	drawable->u.copy.mask.bitmap = 0;
-
-	drawable->u.copy.src_bitmap =
-		qxl_bo_physical_address(qdev, dimage->bo, 0);
-	qxl_release_unmap(qdev, release, &drawable->release_info);
-
-	qxl_push_command_ring_release(qdev, release, QXL_CMD_DRAW, false);
-	qxl_release_fence_buffer_objects(release);
-
-out_free_palette:
-	qxl_bo_unref(&palette_bo);
-out_free_image:
-	qxl_image_free_objects(qdev, dimage);
-out_free_drawable:
-	if (ret)
-		free_drawable(qdev, release);
-}
-
 /* push a draw command using the given clipping rectangles as
  * the sources from the shadow framebuffer.
  *
@@ -267,7 +121,8 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev,
 		       struct qxl_bo *bo,
 		       unsigned int flags, unsigned int color,
 		       struct drm_clip_rect *clips,
-		       unsigned int num_clips, int inc)
+		       unsigned int num_clips, int inc,
+		       uint32_t dumb_shadow_offset)
 {
 	/*
 	 * TODO: if flags & DRM_MODE_FB_DIRTY_ANNOTATE_FILL then we should
@@ -295,6 +150,9 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev,
 	if (ret)
 		return;
 
+	clips->x1 += dumb_shadow_offset;
+	clips->x2 += dumb_shadow_offset;
+
 	left = clips->x1;
 	right = clips->x2;
 	top = clips->y1;
@@ -342,7 +200,8 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev,
 		goto out_release_backoff;
 
 	ret = qxl_image_init(qdev, release, dimage, surface_base,
-			     left, top, width, height, depth, stride);
+			     left - dumb_shadow_offset,
+			     top, width, height, depth, stride);
 	qxl_bo_kunmap(bo);
 	if (ret)
 		goto out_release_backoff;
@@ -397,89 +256,3 @@ out_free_drawable:
 		free_drawable(qdev, release);
 
 }
-
-void qxl_draw_copyarea(struct qxl_device *qdev,
-		       u32 width, u32 height,
-		       u32 sx, u32 sy,
-		       u32 dx, u32 dy)
-{
-	struct qxl_drawable *drawable;
-	struct qxl_rect rect;
-	struct qxl_release *release;
-	int ret;
-
-	ret = alloc_drawable(qdev, &release);
-	if (ret)
-		return;
-
-	/* do a reservation run over all the objects we just allocated */
-	ret = qxl_release_reserve_list(release, true);
-	if (ret)
-		goto out_free_release;
-
-	rect.left = dx;
-	rect.top = dy;
-	rect.right = dx + width;
-	rect.bottom = dy + height;
-	ret = make_drawable(qdev, 0, QXL_COPY_BITS, &rect, release);
-	if (ret) {
-		qxl_release_backoff_reserve_list(release);
-		goto out_free_release;
-	}
-
-	drawable = (struct qxl_drawable *)qxl_release_map(qdev, release);
-	drawable->u.copy_bits.src_pos.x = sx;
-	drawable->u.copy_bits.src_pos.y = sy;
-	qxl_release_unmap(qdev, release, &drawable->release_info);
-
-	qxl_push_command_ring_release(qdev, release, QXL_CMD_DRAW, false);
-	qxl_release_fence_buffer_objects(release);
-
-out_free_release:
-	if (ret)
-		free_drawable(qdev, release);
-}
-
-void qxl_draw_fill(struct qxl_draw_fill *qxl_draw_fill_rec)
-{
-	struct qxl_device *qdev = qxl_draw_fill_rec->qdev;
-	struct qxl_rect rect = qxl_draw_fill_rec->rect;
-	uint32_t color = qxl_draw_fill_rec->color;
-	uint16_t rop = qxl_draw_fill_rec->rop;
-	struct qxl_drawable *drawable;
-	struct qxl_release *release;
-	int ret;
-
-	ret = alloc_drawable(qdev, &release);
-	if (ret)
-		return;
-
-	/* do a reservation run over all the objects we just allocated */
-	ret = qxl_release_reserve_list(release, true);
-	if (ret)
-		goto out_free_release;
-
-	ret = make_drawable(qdev, 0, QXL_DRAW_FILL, &rect, release);
-	if (ret) {
-		qxl_release_backoff_reserve_list(release);
-		goto out_free_release;
-	}
-
-	drawable = (struct qxl_drawable *)qxl_release_map(qdev, release);
-	drawable->u.fill.brush.type = SPICE_BRUSH_TYPE_SOLID;
-	drawable->u.fill.brush.u.color = color;
-	drawable->u.fill.rop_descriptor = rop;
-	drawable->u.fill.mask.flags = 0;
-	drawable->u.fill.mask.pos.x = 0;
-	drawable->u.fill.mask.pos.y = 0;
-	drawable->u.fill.mask.bitmap = 0;
-
-	qxl_release_unmap(qdev, release, &drawable->release_info);
-
-	qxl_push_command_ring_release(qdev, release, QXL_CMD_DRAW, false);
-	qxl_release_fence_buffer_objects(release);
-
-out_free_release:
-	if (ret)
-		free_drawable(qdev, release);
-}
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index 13c8a662f9b4..bb81e310eb6d 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -33,7 +33,8 @@
 
 #include <drm/drmP.h>
 #include <drm/drm.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "qxl_drv.h"
 #include "qxl_object.h"
 
@@ -93,6 +94,8 @@ qxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto modeset_cleanup;
 
+	drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, 0, "qxl");
+	drm_fbdev_generic_setup(&qdev->ddev, 32);
 	return 0;
 
 modeset_cleanup:
@@ -242,7 +245,6 @@ static struct pci_driver qxl_pci_driver = {
 
 static struct drm_driver qxl_driver = {
 	.driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
-			   DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED |
 			   DRIVER_ATOMIC,
 
 	.dumb_create = qxl_mode_dumb_create,
@@ -250,14 +252,10 @@ static struct drm_driver qxl_driver = {
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = qxl_debugfs_init,
 #endif
-	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
-	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export = drm_gem_prime_export,
 	.gem_prime_import = drm_gem_prime_import,
 	.gem_prime_pin = qxl_gem_prime_pin,
 	.gem_prime_unpin = qxl_gem_prime_unpin,
-	.gem_prime_get_sg_table = qxl_gem_prime_get_sg_table,
-	.gem_prime_import_sg_table = qxl_gem_prime_import_sg_table,
 	.gem_prime_vmap = qxl_gem_prime_vmap,
 	.gem_prime_vunmap = qxl_gem_prime_vunmap,
 	.gem_prime_mmap = qxl_gem_prime_mmap,
diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h
index 13a0254b59a1..4a0331b3ff7d 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.h
+++ b/drivers/gpu/drm/qxl/qxl_drv.h
@@ -84,6 +84,7 @@ struct qxl_bo {
 	struct ttm_bo_kmap_obj		kmap;
 	unsigned int pin_count;
 	void				*kptr;
+	unsigned int                    map_count;
 	int                             type;
 
 	/* Constant after initialization */
@@ -130,10 +131,13 @@ struct qxl_mman {
 };
 
 struct qxl_memslot {
+	int             index;
+	const char      *name;
 	uint8_t		generation;
 	uint64_t	start_phys_addr;
-	uint64_t	end_phys_addr;
+	uint64_t	size;
 	uint64_t	high_bits;
+	uint64_t        gpu_offset;
 };
 
 enum {
@@ -216,8 +220,6 @@ struct qxl_device {
 	struct qxl_mman		mman;
 	struct qxl_gem		gem;
 
-	struct drm_fb_helper	fb_helper;
-
 	void *ram_physical;
 
 	struct qxl_ring *release_ring;
@@ -226,16 +228,12 @@ struct qxl_device {
 
 	struct qxl_ram_header *ram_header;
 
-	unsigned int primary_created:1;
-
-	struct qxl_memslot	*mem_slots;
-	uint8_t		n_mem_slots;
+	struct qxl_bo *primary_bo;
+	struct qxl_bo *dumb_shadow_bo;
+	struct qxl_head *dumb_heads;
 
-	uint8_t		main_mem_slot;
-	uint8_t		surfaces_mem_slot;
-	uint8_t		slot_id_bits;
-	uint8_t		slot_gen_bits;
-	uint64_t	va_slot_mask;
+	struct qxl_memslot main_slot;
+	struct qxl_memslot surfaces_slot;
 
 	spinlock_t	release_lock;
 	struct idr	release_idr;
@@ -308,30 +306,20 @@ void qxl_ring_free(struct qxl_ring *ring);
 void qxl_ring_init_hdr(struct qxl_ring *ring);
 int qxl_check_idle(struct qxl_ring *ring);
 
-static inline void *
-qxl_fb_virtual_address(struct qxl_device *qdev, unsigned long physical)
-{
-	DRM_DEBUG_DRIVER("not implemented (%lu)\n", physical);
-	return 0;
-}
-
 static inline uint64_t
 qxl_bo_physical_address(struct qxl_device *qdev, struct qxl_bo *bo,
 			unsigned long offset)
 {
-	int slot_id = bo->type == QXL_GEM_DOMAIN_VRAM ? qdev->main_mem_slot : qdev->surfaces_mem_slot;
-	struct qxl_memslot *slot = &(qdev->mem_slots[slot_id]);
+	struct qxl_memslot *slot =
+		(bo->tbo.mem.mem_type == TTM_PL_VRAM)
+		? &qdev->main_slot : &qdev->surfaces_slot;
+
+	WARN_ON_ONCE((bo->tbo.offset & slot->gpu_offset) != slot->gpu_offset);
 
 	/* TODO - need to hold one of the locks to read tbo.offset */
-	return slot->high_bits | (bo->tbo.offset + offset);
+	return slot->high_bits | (bo->tbo.offset - slot->gpu_offset + offset);
 }
 
-/* qxl_fb.c */
-#define QXLFB_CONN_LIMIT 1
-
-int qxl_fbdev_init(struct qxl_device *qdev);
-void qxl_fbdev_fini(struct qxl_device *qdev);
-
 /* qxl_display.c */
 void qxl_display_read_client_monitors_config(struct qxl_device *qdev);
 int qxl_create_monitors_object(struct qxl_device *qdev);
@@ -392,7 +380,6 @@ void qxl_update_screen(struct qxl_device *qxl);
 /* qxl io operations (qxl_cmd.c) */
 
 void qxl_io_create_primary(struct qxl_device *qdev,
-			   unsigned int offset,
 			   struct qxl_bo *bo);
 void qxl_io_destroy_primary(struct qxl_device *qdev);
 void qxl_io_memslot_add(struct qxl_device *qdev, uint8_t id);
@@ -437,22 +424,13 @@ int qxl_alloc_bo_reserved(struct qxl_device *qdev,
 			  struct qxl_bo **_bo);
 /* qxl drawing commands */
 
-void qxl_draw_opaque_fb(const struct qxl_fb_image *qxl_fb_image,
-			int stride /* filled in if 0 */);
-
 void qxl_draw_dirty_fb(struct qxl_device *qdev,
 		       struct drm_framebuffer *fb,
 		       struct qxl_bo *bo,
 		       unsigned int flags, unsigned int color,
 		       struct drm_clip_rect *clips,
-		       unsigned int num_clips, int inc);
-
-void qxl_draw_fill(struct qxl_draw_fill *qxl_draw_fill_rec);
-
-void qxl_draw_copyarea(struct qxl_device *qdev,
-		       u32 width, u32 height,
-		       u32 sx, u32 sy,
-		       u32 dx, u32 dy);
+		       unsigned int num_clips, int inc,
+		       uint32_t dumb_shadow_offset);
 
 void qxl_release_free(struct qxl_device *qdev,
 		      struct qxl_release *release);
@@ -485,9 +463,6 @@ int qxl_gem_prime_mmap(struct drm_gem_object *obj,
 int qxl_irq_init(struct qxl_device *qdev);
 irqreturn_t qxl_irq_handler(int irq, void *arg);
 
-/* qxl_fb.c */
-bool qxl_fbdev_qobj_is_fb(struct qxl_device *qdev, struct qxl_bo *qobj);
-
 int qxl_debugfs_add_files(struct qxl_device *qdev,
 			  struct drm_info_list *files,
 			  unsigned int nfiles);
@@ -497,8 +472,7 @@ int qxl_surface_id_alloc(struct qxl_device *qdev,
 void qxl_surface_id_dealloc(struct qxl_device *qdev,
 			    uint32_t surface_id);
 int qxl_hw_surface_alloc(struct qxl_device *qdev,
-			 struct qxl_bo *surf,
-			 struct ttm_mem_reg *mem);
+			 struct qxl_bo *surf);
 int qxl_hw_surface_dealloc(struct qxl_device *qdev,
 			   struct qxl_bo *surf);
 
diff --git a/drivers/gpu/drm/qxl/qxl_dumb.c b/drivers/gpu/drm/qxl/qxl_dumb.c
index e3765739c396..272d19b677d8 100644
--- a/drivers/gpu/drm/qxl/qxl_dumb.c
+++ b/drivers/gpu/drm/qxl/qxl_dumb.c
@@ -59,7 +59,7 @@ int qxl_mode_dumb_create(struct drm_file *file_priv,
 	surf.stride = pitch;
 	surf.format = format;
 	r = qxl_gem_object_create_with_handle(qdev, file_priv,
-					      QXL_GEM_DOMAIN_VRAM,
+					      QXL_GEM_DOMAIN_SURFACE,
 					      args->size, &surf, &qobj,
 					      &handle);
 	if (r)
diff --git a/drivers/gpu/drm/qxl/qxl_fb.c b/drivers/gpu/drm/qxl/qxl_fb.c
deleted file mode 100644
index a819d24225d2..000000000000
--- a/drivers/gpu/drm/qxl/qxl_fb.c
+++ /dev/null
@@ -1,300 +0,0 @@
-/*
- * Copyright © 2013 Red Hat
- *
- * 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 (including the next
- * paragraph) 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 AUTHORS OR COPYRIGHT HOLDERS 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:
- *     David Airlie
- */
-#include <linux/module.h>
-
-#include <drm/drmP.h>
-#include <drm/drm.h>
-#include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_gem_framebuffer_helper.h>
-
-#include "qxl_drv.h"
-
-#include "qxl_object.h"
-
-static void qxl_fb_image_init(struct qxl_fb_image *qxl_fb_image,
-			      struct qxl_device *qdev, struct fb_info *info,
-			      const struct fb_image *image)
-{
-	qxl_fb_image->qdev = qdev;
-	if (info) {
-		qxl_fb_image->visual = info->fix.visual;
-		if (qxl_fb_image->visual == FB_VISUAL_TRUECOLOR ||
-		    qxl_fb_image->visual == FB_VISUAL_DIRECTCOLOR)
-			memcpy(&qxl_fb_image->pseudo_palette,
-			       info->pseudo_palette,
-			       sizeof(qxl_fb_image->pseudo_palette));
-	} else {
-		 /* fallback */
-		if (image->depth == 1)
-			qxl_fb_image->visual = FB_VISUAL_MONO10;
-		else
-			qxl_fb_image->visual = FB_VISUAL_DIRECTCOLOR;
-	}
-	if (image) {
-		memcpy(&qxl_fb_image->fb_image, image,
-		       sizeof(qxl_fb_image->fb_image));
-	}
-}
-
-static struct fb_ops qxlfb_ops = {
-	.owner = THIS_MODULE,
-	DRM_FB_HELPER_DEFAULT_OPS,
-	.fb_fillrect = drm_fb_helper_sys_fillrect,
-	.fb_copyarea = drm_fb_helper_sys_copyarea,
-	.fb_imageblit = drm_fb_helper_sys_imageblit,
-};
-
-static void qxlfb_destroy_pinned_object(struct drm_gem_object *gobj)
-{
-	struct qxl_bo *qbo = gem_to_qxl_bo(gobj);
-
-	qxl_bo_kunmap(qbo);
-	qxl_bo_unpin(qbo);
-
-	drm_gem_object_put_unlocked(gobj);
-}
-
-static int qxlfb_create_pinned_object(struct qxl_device *qdev,
-				      const struct drm_mode_fb_cmd2 *mode_cmd,
-				      struct drm_gem_object **gobj_p)
-{
-	struct drm_gem_object *gobj = NULL;
-	struct qxl_bo *qbo = NULL;
-	int ret;
-	int aligned_size, size;
-	int height = mode_cmd->height;
-
-	size = mode_cmd->pitches[0] * height;
-	aligned_size = ALIGN(size, PAGE_SIZE);
-	/* TODO: unallocate and reallocate surface0 for real. Hack to just
-	 * have a large enough surface0 for 1024x768 Xorg 32bpp mode */
-	ret = qxl_gem_object_create(qdev, aligned_size, 0,
-				    QXL_GEM_DOMAIN_SURFACE,
-				    false, /* is discardable */
-				    false, /* is kernel (false means device) */
-				    NULL,
-				    &gobj);
-	if (ret) {
-		pr_err("failed to allocate framebuffer (%d)\n",
-		       aligned_size);
-		return -ENOMEM;
-	}
-	qbo = gem_to_qxl_bo(gobj);
-
-	qbo->surf.width = mode_cmd->width;
-	qbo->surf.height = mode_cmd->height;
-	qbo->surf.stride = mode_cmd->pitches[0];
-	qbo->surf.format = SPICE_SURFACE_FMT_32_xRGB;
-
-	ret = qxl_bo_pin(qbo);
-	if (ret) {
-		goto out_unref;
-	}
-	ret = qxl_bo_kmap(qbo, NULL);
-
-	if (ret)
-		goto out_unref;
-
-	*gobj_p = gobj;
-	return 0;
-out_unref:
-	qxlfb_destroy_pinned_object(gobj);
-	*gobj_p = NULL;
-	return ret;
-}
-
-/*
- * FIXME
- * It should not be necessary to have a special dirty() callback for fbdev.
- */
-static int qxlfb_framebuffer_dirty(struct drm_framebuffer *fb,
-				   struct drm_file *file_priv,
-				   unsigned int flags, unsigned int color,
-				   struct drm_clip_rect *clips,
-				   unsigned int num_clips)
-{
-	struct qxl_device *qdev = fb->dev->dev_private;
-	struct fb_info *info = qdev->fb_helper.fbdev;
-	struct qxl_fb_image qxl_fb_image;
-	struct fb_image *image = &qxl_fb_image.fb_image;
-
-	/* TODO: hard coding 32 bpp */
-	int stride = fb->pitches[0];
-
-	/*
-	 * we are using a shadow draw buffer, at qdev->surface0_shadow
-	 */
-	image->dx = clips->x1;
-	image->dy = clips->y1;
-	image->width = clips->x2 - clips->x1;
-	image->height = clips->y2 - clips->y1;
-	image->fg_color = 0xffffffff; /* unused, just to avoid uninitialized
-					 warnings */
-	image->bg_color = 0;
-	image->depth = 32;	     /* TODO: take from somewhere? */
-	image->cmap.start = 0;
-	image->cmap.len = 0;
-	image->cmap.red = NULL;
-	image->cmap.green = NULL;
-	image->cmap.blue = NULL;
-	image->cmap.transp = NULL;
-	image->data = info->screen_base + (clips->x1 * 4) + (stride * clips->y1);
-
-	qxl_fb_image_init(&qxl_fb_image, qdev, info, NULL);
-	qxl_draw_opaque_fb(&qxl_fb_image, stride);
-
-	return 0;
-}
-
-static const struct drm_framebuffer_funcs qxlfb_fb_funcs = {
-	.destroy = drm_gem_fb_destroy,
-	.create_handle = drm_gem_fb_create_handle,
-	.dirty = qxlfb_framebuffer_dirty,
-};
-
-static int qxlfb_create(struct drm_fb_helper *helper,
-			struct drm_fb_helper_surface_size *sizes)
-{
-	struct qxl_device *qdev =
-		container_of(helper, struct qxl_device, fb_helper);
-	struct fb_info *info;
-	struct drm_framebuffer *fb = NULL;
-	struct drm_mode_fb_cmd2 mode_cmd;
-	struct drm_gem_object *gobj = NULL;
-	struct qxl_bo *qbo = NULL;
-	int ret;
-	int bpp = sizes->surface_bpp;
-	int depth = sizes->surface_depth;
-	void *shadow;
-
-	mode_cmd.width = sizes->surface_width;
-	mode_cmd.height = sizes->surface_height;
-
-	mode_cmd.pitches[0] = ALIGN(mode_cmd.width * ((bpp + 1) / 8), 64);
-	mode_cmd.pixel_format = drm_mode_legacy_fb_format(bpp, depth);
-
-	ret = qxlfb_create_pinned_object(qdev, &mode_cmd, &gobj);
-	if (ret < 0)
-		return ret;
-
-	qbo = gem_to_qxl_bo(gobj);
-	DRM_DEBUG_DRIVER("%dx%d %d\n", mode_cmd.width,
-			 mode_cmd.height, mode_cmd.pitches[0]);
-
-	shadow = vmalloc(array_size(mode_cmd.pitches[0], mode_cmd.height));
-	/* TODO: what's the usual response to memory allocation errors? */
-	BUG_ON(!shadow);
-	DRM_DEBUG_DRIVER("surface0 at gpu offset %lld, mmap_offset %lld (virt %p, shadow %p)\n",
-			 qxl_bo_gpu_offset(qbo), qxl_bo_mmap_offset(qbo),
-			 qbo->kptr, shadow);
-
-	info = drm_fb_helper_alloc_fbi(helper);
-	if (IS_ERR(info)) {
-		ret = PTR_ERR(info);
-		goto out_unref;
-	}
-
-	info->par = helper;
-
-	fb = drm_gem_fbdev_fb_create(&qdev->ddev, sizes, 64, gobj,
-				     &qxlfb_fb_funcs);
-	if (IS_ERR(fb)) {
-		DRM_ERROR("Failed to create framebuffer: %ld\n", PTR_ERR(fb));
-		ret = PTR_ERR(fb);
-		goto out_unref;
-	}
-
-	/* setup helper with fb data */
-	qdev->fb_helper.fb = fb;
-
-	strcpy(info->fix.id, "qxldrmfb");
-
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-
-	info->fbops = &qxlfb_ops;
-
-	/*
-	 * TODO: using gobj->size in various places in this function. Not sure
-	 * what the difference between the different sizes is.
-	 */
-	info->fix.smem_start = qdev->vram_base; /* TODO - correct? */
-	info->fix.smem_len = gobj->size;
-	info->screen_base = shadow;
-	info->screen_size = gobj->size;
-
-	drm_fb_helper_fill_var(info, &qdev->fb_helper, sizes->fb_width,
-			       sizes->fb_height);
-
-	/* setup aperture base/size for vesafb takeover */
-	info->apertures->ranges[0].base = qdev->ddev.mode_config.fb_base;
-	info->apertures->ranges[0].size = qdev->vram_size;
-
-	info->fix.mmio_start = 0;
-	info->fix.mmio_len = 0;
-
-	if (info->screen_base == NULL) {
-		ret = -ENOSPC;
-		goto out_unref;
-	}
-
-	/* XXX error handling. */
-	drm_fb_helper_defio_init(helper);
-
-	DRM_INFO("fb mappable at 0x%lX, size %lu\n",  info->fix.smem_start, (unsigned long)info->screen_size);
-	DRM_INFO("fb: depth %d, pitch %d, width %d, height %d\n",
-		 fb->format->depth, fb->pitches[0], fb->width, fb->height);
-	return 0;
-
-out_unref:
-	if (qbo) {
-		qxl_bo_kunmap(qbo);
-		qxl_bo_unpin(qbo);
-	}
-	drm_gem_object_put_unlocked(gobj);
-	return ret;
-}
-
-static const struct drm_fb_helper_funcs qxl_fb_helper_funcs = {
-	.fb_probe = qxlfb_create,
-};
-
-int qxl_fbdev_init(struct qxl_device *qdev)
-{
-	return drm_fb_helper_fbdev_setup(&qdev->ddev, &qdev->fb_helper,
-					 &qxl_fb_helper_funcs, 32,
-					 QXLFB_CONN_LIMIT);
-}
-
-void qxl_fbdev_fini(struct qxl_device *qdev)
-{
-	struct fb_info *fbi = qdev->fb_helper.fbdev;
-	void *shadow = fbi ? fbi->screen_buffer : NULL;
-
-	drm_fb_helper_fbdev_teardown(&qdev->ddev);
-	vfree(shadow);
-}
diff --git a/drivers/gpu/drm/qxl/qxl_kms.c b/drivers/gpu/drm/qxl/qxl_kms.c
index 15238a413f9d..bee61fa2c9bc 100644
--- a/drivers/gpu/drm/qxl/qxl_kms.c
+++ b/drivers/gpu/drm/qxl/qxl_kms.c
@@ -26,7 +26,7 @@
 #include "qxl_drv.h"
 #include "qxl_object.h"
 
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/io-mapping.h>
 
 int qxl_log_level;
@@ -53,40 +53,47 @@ static bool qxl_check_device(struct qxl_device *qdev)
 	return true;
 }
 
-static void setup_hw_slot(struct qxl_device *qdev, int slot_index,
-			  struct qxl_memslot *slot)
+static void setup_hw_slot(struct qxl_device *qdev, struct qxl_memslot *slot)
 {
 	qdev->ram_header->mem_slot.mem_start = slot->start_phys_addr;
-	qdev->ram_header->mem_slot.mem_end = slot->end_phys_addr;
-	qxl_io_memslot_add(qdev, slot_index);
+	qdev->ram_header->mem_slot.mem_end = slot->start_phys_addr + slot->size;
+	qxl_io_memslot_add(qdev, qdev->rom->slots_start + slot->index);
 }
 
-static uint8_t setup_slot(struct qxl_device *qdev, uint8_t slot_index_offset,
-	unsigned long start_phys_addr, unsigned long end_phys_addr)
+static void setup_slot(struct qxl_device *qdev,
+		       struct qxl_memslot *slot,
+		       unsigned int slot_index,
+		       const char *slot_name,
+		       unsigned long start_phys_addr,
+		       unsigned long size)
 {
 	uint64_t high_bits;
-	struct qxl_memslot *slot;
-	uint8_t slot_index;
 
-	slot_index = qdev->rom->slots_start + slot_index_offset;
-	slot = &qdev->mem_slots[slot_index];
+	slot->index = slot_index;
+	slot->name = slot_name;
 	slot->start_phys_addr = start_phys_addr;
-	slot->end_phys_addr = end_phys_addr;
+	slot->size = size;
 
-	setup_hw_slot(qdev, slot_index, slot);
+	setup_hw_slot(qdev, slot);
 
 	slot->generation = qdev->rom->slot_generation;
-	high_bits = slot_index << qdev->slot_gen_bits;
+	high_bits = (qdev->rom->slots_start + slot->index)
+		<< qdev->rom->slot_gen_bits;
 	high_bits |= slot->generation;
-	high_bits <<= (64 - (qdev->slot_gen_bits + qdev->slot_id_bits));
+	high_bits <<= (64 - (qdev->rom->slot_gen_bits + qdev->rom->slot_id_bits));
 	slot->high_bits = high_bits;
-	return slot_index;
+
+	DRM_INFO("slot %d (%s): base 0x%08lx, size 0x%08lx, gpu_offset 0x%lx\n",
+		 slot->index, slot->name,
+		 (unsigned long)slot->start_phys_addr,
+		 (unsigned long)slot->size,
+		 (unsigned long)slot->gpu_offset);
 }
 
 void qxl_reinit_memslots(struct qxl_device *qdev)
 {
-	setup_hw_slot(qdev, qdev->main_mem_slot, &qdev->mem_slots[qdev->main_mem_slot]);
-	setup_hw_slot(qdev, qdev->surfaces_mem_slot, &qdev->mem_slots[qdev->surfaces_mem_slot]);
+	setup_hw_slot(qdev, &qdev->main_slot);
+	setup_hw_slot(qdev, &qdev->surfaces_slot);
 }
 
 static void qxl_gc_work(struct work_struct *work)
@@ -229,23 +236,6 @@ int qxl_device_init(struct qxl_device *qdev,
 		r = -ENOMEM;
 		goto cursor_ring_free;
 	}
-	/* TODO - slot initialization should happen on reset. where is our
-	 * reset handler? */
-	qdev->n_mem_slots = qdev->rom->slots_end;
-	qdev->slot_gen_bits = qdev->rom->slot_gen_bits;
-	qdev->slot_id_bits = qdev->rom->slot_id_bits;
-	qdev->va_slot_mask =
-		(~(uint64_t)0) >> (qdev->slot_id_bits + qdev->slot_gen_bits);
-
-	qdev->mem_slots =
-		kmalloc_array(qdev->n_mem_slots, sizeof(struct qxl_memslot),
-			      GFP_KERNEL);
-
-	if (!qdev->mem_slots) {
-		DRM_ERROR("Unable to alloc mem slots\n");
-		r = -ENOMEM;
-		goto release_ring_free;
-	}
 
 	idr_init(&qdev->release_idr);
 	spin_lock_init(&qdev->release_idr_lock);
@@ -264,33 +254,24 @@ int qxl_device_init(struct qxl_device *qdev,
 	r = qxl_irq_init(qdev);
 	if (r) {
 		DRM_ERROR("Unable to init qxl irq\n");
-		goto mem_slots_free;
+		goto release_ring_free;
 	}
 
 	/*
 	 * Note that virtual is surface0. We rely on the single ioremap done
 	 * before.
 	 */
-	qdev->main_mem_slot = setup_slot(qdev, 0,
-		(unsigned long)qdev->vram_base,
-		(unsigned long)qdev->vram_base + qdev->rom->ram_header_offset);
-	qdev->surfaces_mem_slot = setup_slot(qdev, 1,
-		(unsigned long)qdev->surfaceram_base,
-		(unsigned long)qdev->surfaceram_base + qdev->surfaceram_size);
-	DRM_INFO("main mem slot %d [%lx,%x]\n",
-		 qdev->main_mem_slot,
-		 (unsigned long)qdev->vram_base, qdev->rom->ram_header_offset);
-	DRM_INFO("surface mem slot %d [%lx,%lx]\n",
-		 qdev->surfaces_mem_slot,
-		 (unsigned long)qdev->surfaceram_base,
-		 (unsigned long)qdev->surfaceram_size);
+	setup_slot(qdev, &qdev->main_slot, 0, "main",
+		   (unsigned long)qdev->vram_base,
+		   (unsigned long)qdev->rom->ram_header_offset);
+	setup_slot(qdev, &qdev->surfaces_slot, 1, "surfaces",
+		   (unsigned long)qdev->surfaceram_base,
+		   (unsigned long)qdev->surfaceram_size);
 
 	INIT_WORK(&qdev->gc_work, qxl_gc_work);
 
 	return 0;
 
-mem_slots_free:
-	kfree(qdev->mem_slots);
 release_ring_free:
 	qxl_ring_free(qdev->release_ring);
 cursor_ring_free:
diff --git a/drivers/gpu/drm/qxl/qxl_object.c b/drivers/gpu/drm/qxl/qxl_object.c
index 91f3bbc73ecc..4928fa602944 100644
--- a/drivers/gpu/drm/qxl/qxl_object.c
+++ b/drivers/gpu/drm/qxl/qxl_object.c
@@ -36,6 +36,7 @@ static void qxl_ttm_bo_destroy(struct ttm_buffer_object *tbo)
 	qdev = (struct qxl_device *)bo->gem_base.dev->dev_private;
 
 	qxl_surface_evict(qdev, bo, false);
+	WARN_ON_ONCE(bo->map_count > 0);
 	mutex_lock(&qdev->gem.mutex);
 	list_del_init(&bo->list);
 	mutex_unlock(&qdev->gem.mutex);
@@ -60,8 +61,10 @@ void qxl_ttm_placement_from_domain(struct qxl_bo *qbo, u32 domain, bool pinned)
 	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_SURFACE)
+	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;
+	}
 	if (domain == QXL_GEM_DOMAIN_CPU)
 		qbo->placements[c++].flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM | pflag;
 	if (!c)
@@ -129,6 +132,7 @@ int qxl_bo_kmap(struct qxl_bo *bo, void **ptr)
 	if (bo->kptr) {
 		if (ptr)
 			*ptr = bo->kptr;
+		bo->map_count++;
 		return 0;
 	}
 	r = ttm_bo_kmap(&bo->tbo, 0, bo->tbo.num_pages, &bo->kmap);
@@ -137,6 +141,7 @@ int qxl_bo_kmap(struct qxl_bo *bo, void **ptr)
 	bo->kptr = ttm_kmap_obj_virtual(&bo->kmap, &is_iomem);
 	if (ptr)
 		*ptr = bo->kptr;
+	bo->map_count = 1;
 	return 0;
 }
 
@@ -178,6 +183,9 @@ void qxl_bo_kunmap(struct qxl_bo *bo)
 {
 	if (bo->kptr == NULL)
 		return;
+	bo->map_count--;
+	if (bo->map_count > 0)
+		return;
 	bo->kptr = NULL;
 	ttm_bo_kunmap(&bo->kmap);
 }
@@ -332,7 +340,7 @@ int qxl_bo_check_id(struct qxl_device *qdev, struct qxl_bo *bo)
 		if (ret)
 			return ret;
 
-		ret = qxl_hw_surface_alloc(qdev, bo, NULL);
+		ret = qxl_hw_surface_alloc(qdev, bo);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/gpu/drm/qxl/qxl_prime.c b/drivers/gpu/drm/qxl/qxl_prime.c
index a55dece118b2..8b448eca1cd9 100644
--- a/drivers/gpu/drm/qxl/qxl_prime.c
+++ b/drivers/gpu/drm/qxl/qxl_prime.c
@@ -23,44 +23,43 @@
  */
 
 #include "qxl_drv.h"
+#include "qxl_object.h"
 
 /* Empty Implementations as there should not be any other driver for a virtual
  * device that might share buffers with qxl */
 
 int qxl_gem_prime_pin(struct drm_gem_object *obj)
 {
-	WARN_ONCE(1, "not implemented");
-	return -ENOSYS;
-}
+	struct qxl_bo *bo = gem_to_qxl_bo(obj);
 
-void qxl_gem_prime_unpin(struct drm_gem_object *obj)
-{
-	WARN_ONCE(1, "not implemented");
+	return qxl_bo_pin(bo);
 }
 
-struct sg_table *qxl_gem_prime_get_sg_table(struct drm_gem_object *obj)
+void qxl_gem_prime_unpin(struct drm_gem_object *obj)
 {
-	WARN_ONCE(1, "not implemented");
-	return ERR_PTR(-ENOSYS);
-}
+	struct qxl_bo *bo = gem_to_qxl_bo(obj);
 
-struct drm_gem_object *qxl_gem_prime_import_sg_table(
-	struct drm_device *dev, struct dma_buf_attachment *attach,
-	struct sg_table *table)
-{
-	WARN_ONCE(1, "not implemented");
-	return ERR_PTR(-ENOSYS);
+	qxl_bo_unpin(bo);
 }
 
 void *qxl_gem_prime_vmap(struct drm_gem_object *obj)
 {
-	WARN_ONCE(1, "not implemented");
-	return ERR_PTR(-ENOSYS);
+	struct qxl_bo *bo = gem_to_qxl_bo(obj);
+	void *ptr;
+	int ret;
+
+	ret = qxl_bo_kmap(bo, &ptr);
+	if (ret < 0)
+		return ERR_PTR(ret);
+
+	return ptr;
 }
 
 void qxl_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
 {
-	WARN_ONCE(1, "not implemented");
+	struct qxl_bo *bo = gem_to_qxl_bo(obj);
+
+	qxl_bo_kunmap(bo);
 }
 
 int qxl_gem_prime_mmap(struct drm_gem_object *obj,
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index 886f61e94f24..92f5db5b296f 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -100,6 +100,11 @@ static int qxl_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
 static int qxl_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 			     struct ttm_mem_type_manager *man)
 {
+	struct qxl_device *qdev = qxl_get_qdev(bdev);
+	unsigned int gpu_offset_shift =
+		64 - (qdev->rom->slot_gen_bits + qdev->rom->slot_id_bits + 8);
+	struct qxl_memslot *slot;
+
 	switch (type) {
 	case TTM_PL_SYSTEM:
 		/* System memory */
@@ -110,8 +115,11 @@ static int qxl_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 	case TTM_PL_VRAM:
 	case TTM_PL_PRIV:
 		/* "On-card" video ram */
+		slot = (type == TTM_PL_VRAM) ?
+			&qdev->main_slot : &qdev->surfaces_slot;
+		slot->gpu_offset = (uint64_t)type << gpu_offset_shift;
 		man->func = &ttm_bo_manager_func;
-		man->gpu_offset = 0;
+		man->gpu_offset = slot->gpu_offset;
 		man->flags = TTM_MEMTYPE_FLAG_FIXED |
 			     TTM_MEMTYPE_FLAG_MAPPABLE;
 		man->available_caching = TTM_PL_MASK_CACHING;
@@ -196,7 +204,7 @@ static void qxl_ttm_io_mem_free(struct ttm_bo_device *bdev,
  * TTM backend functions.
  */
 struct qxl_ttm_tt {
-	struct ttm_dma_tt		ttm;
+	struct ttm_tt		        ttm;
 	struct qxl_device		*qdev;
 	u64				offset;
 };
@@ -225,7 +233,7 @@ static void qxl_ttm_backend_destroy(struct ttm_tt *ttm)
 {
 	struct qxl_ttm_tt *gtt = (void *)ttm;
 
-	ttm_dma_tt_fini(&gtt->ttm);
+	ttm_tt_fini(&gtt->ttm);
 	kfree(gtt);
 }
 
@@ -245,13 +253,13 @@ 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.ttm.func = &qxl_backend_func;
+	gtt->ttm.func = &qxl_backend_func;
 	gtt->qdev = qdev;
-	if (ttm_dma_tt_init(&gtt->ttm, bo, page_flags)) {
+	if (ttm_tt_init(&gtt->ttm, bo, page_flags)) {
 		kfree(gtt);
 		return NULL;
 	}
-	return &gtt->ttm.ttm;
+	return &gtt->ttm;
 }
 
 static void qxl_move_null(struct ttm_buffer_object *bo,
diff --git a/drivers/gpu/drm/r128/r128_drv.c b/drivers/gpu/drm/r128/r128_drv.c
index 0d2b7e42b3a7..4b1a505ab353 100644
--- a/drivers/gpu/drm/r128/r128_drv.c
+++ b/drivers/gpu/drm/r128/r128_drv.c
@@ -57,7 +57,7 @@ static const struct file_operations r128_driver_fops = {
 static struct drm_driver driver = {
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_PCI_DMA | DRIVER_SG | DRIVER_LEGACY |
-	    DRIVER_HAVE_DMA | DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED,
+	    DRIVER_HAVE_DMA | DRIVER_HAVE_IRQ,
 	.dev_priv_size = sizeof(drm_r128_buf_priv_t),
 	.load = r128_driver_load,
 	.preclose = r128_driver_preclose,
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index d587779a80b4..a97294ac96d5 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -5676,7 +5676,7 @@ int ci_dpm_init(struct radeon_device *rdev)
 	u16 data_offset, size;
 	u8 frev, crev;
 	struct ci_power_info *pi;
-	enum pci_bus_speed speed_cap;
+	enum pci_bus_speed speed_cap = PCI_SPEED_UNKNOWN;
 	struct pci_dev *root = rdev->pdev->bus->self;
 	int ret;
 
@@ -5685,7 +5685,8 @@ int ci_dpm_init(struct radeon_device *rdev)
 		return -ENOMEM;
 	rdev->pm.dpm.priv = pi;
 
-	speed_cap = pcie_get_speed_cap(root);
+	if (!pci_is_root_bus(rdev->pdev->bus))
+		speed_cap = pcie_get_speed_cap(root);
 	if (speed_cap == PCI_SPEED_UNKNOWN) {
 		pi->sys_pcie_mask = 0;
 	} else {
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c
index 8d3251a10cd4..224cc21bbe38 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -29,6 +29,7 @@
 #include <acpi/video.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "radeon.h"
 #include "radeon_acpi.h"
 #include "atom.h"
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 414642e5b7a3..de1745adcccc 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -28,6 +28,7 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_dp_mst_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/radeon_drm.h>
 #include "radeon.h"
 #include "radeon_audio.h"
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 59c8a6647ff2..53f29a115104 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -29,6 +29,7 @@
 #include <linux/slab.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_cache.h>
 #include <drm/radeon_drm.h>
 #include <linux/pm_runtime.h>
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index b190478ad087..aa898c699101 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -35,6 +35,7 @@
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_edid.h>
 
 #include <linux/gcd.h>
diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c
index a0c70e27ab65..8d85540bbb43 100644
--- a/drivers/gpu/drm/radeon/radeon_dp_mst.c
+++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c
@@ -3,6 +3,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_dp_mst_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "radeon.h"
 #include "atom.h"
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 99c63eeb2866..2e96c886392b 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -43,6 +43,7 @@
 #include <drm/drm_fb_helper.h>
 
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 /*
  * KMS wrapper.
@@ -533,9 +534,7 @@ radeon_get_crtc_scanout_position(struct drm_device *dev, unsigned int pipe,
 
 static struct drm_driver kms_driver = {
 	.driver_features =
-	    DRIVER_USE_AGP |
-	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
-	    DRIVER_PRIME | DRIVER_RENDER,
+	    DRIVER_USE_AGP | DRIVER_GEM | DRIVER_PRIME | DRIVER_RENDER,
 	.load = radeon_driver_load_kms,
 	.open = radeon_driver_open_kms,
 	.postclose = radeon_driver_postclose_kms,
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index afaf10db47cc..1d5e3ba7383e 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -27,6 +27,7 @@
  */
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/radeon_drm.h>
 #include "radeon_reg.h"
 #include "radeon.h"
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 8fb60b3af015..0a785ef0ab66 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -6899,7 +6899,7 @@ int si_dpm_init(struct radeon_device *rdev)
 	struct ni_power_info *ni_pi;
 	struct si_power_info *si_pi;
 	struct atom_clock_dividers dividers;
-	enum pci_bus_speed speed_cap;
+	enum pci_bus_speed speed_cap = PCI_SPEED_UNKNOWN;
 	struct pci_dev *root = rdev->pdev->bus->self;
 	int ret;
 
@@ -6911,7 +6911,8 @@ int si_dpm_init(struct radeon_device *rdev)
 	eg_pi = &ni_pi->eg;
 	pi = &eg_pi->rv7xx;
 
-	speed_cap = pcie_get_speed_cap(root);
+	if (!pci_is_root_bus(rdev->pdev->bus))
+		speed_cap = pcie_get_speed_cap(root);
 	if (speed_cap == PCI_SPEED_UNKNOWN) {
 		si_pi->sys_pcie_mask = 0;
 	} else {
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index 225141656e19..7c36e2777a15 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -4,6 +4,7 @@ config DRM_RCAR_DU
 	depends on DRM && OF
 	depends on ARM || ARM64
 	depends on ARCH_RENESAS || COMPILE_TEST
+	imply DRM_RCAR_LVDS
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
index 771b460c7216..4cdea14d552f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
@@ -9,15 +9,17 @@
 
 #include <linux/clk.h>
 #include <linux/mutex.h>
+#include <linux/platform_device.h>
 #include <linux/sys_soc.h>
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_vblank.h>
 
 #include "rcar_du_crtc.h"
 #include "rcar_du_drv.h"
@@ -26,6 +28,7 @@
 #include "rcar_du_plane.h"
 #include "rcar_du_regs.h"
 #include "rcar_du_vsp.h"
+#include "rcar_lvds.h"
 
 static u32 rcar_du_crtc_read(struct rcar_du_crtc *rcrtc, u32 reg)
 {
@@ -657,8 +660,27 @@ static void rcar_du_crtc_atomic_enable(struct drm_crtc *crtc,
 				       struct drm_crtc_state *old_state)
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
+	struct rcar_du_crtc_state *rstate = to_rcar_crtc_state(crtc->state);
+	struct rcar_du_device *rcdu = rcrtc->group->dev;
 
 	rcar_du_crtc_get(rcrtc);
+
+	/*
+	 * On D3/E3 the dot clock is provided by the LVDS encoder attached to
+	 * the DU channel. We need to enable its clock output explicitly if
+	 * the LVDS output is disabled.
+	 */
+	if (rcdu->info->lvds_clk_mask & BIT(rcrtc->index) &&
+	    rstate->outputs == BIT(RCAR_DU_OUTPUT_DPAD0)) {
+		struct rcar_du_encoder *encoder =
+			rcdu->encoders[RCAR_DU_OUTPUT_LVDS0 + rcrtc->index];
+		const struct drm_display_mode *mode =
+			&crtc->state->adjusted_mode;
+
+		rcar_lvds_clk_enable(encoder->base.bridge,
+				     mode->clock * 1000);
+	}
+
 	rcar_du_crtc_start(rcrtc);
 }
 
@@ -666,10 +688,24 @@ static void rcar_du_crtc_atomic_disable(struct drm_crtc *crtc,
 					struct drm_crtc_state *old_state)
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
+	struct rcar_du_crtc_state *rstate = to_rcar_crtc_state(old_state);
+	struct rcar_du_device *rcdu = rcrtc->group->dev;
 
 	rcar_du_crtc_stop(rcrtc);
 	rcar_du_crtc_put(rcrtc);
 
+	if (rcdu->info->lvds_clk_mask & BIT(rcrtc->index) &&
+	    rstate->outputs == BIT(RCAR_DU_OUTPUT_DPAD0)) {
+		struct rcar_du_encoder *encoder =
+			rcdu->encoders[RCAR_DU_OUTPUT_LVDS0 + rcrtc->index];
+
+		/*
+		 * Disable the LVDS clock output, see
+		 * rcar_du_crtc_atomic_enable().
+		 */
+		rcar_lvds_clk_disable(encoder->base.bridge);
+	}
+
 	spin_lock_irq(&crtc->dev->event_lock);
 	if (crtc->state->event) {
 		drm_crtc_send_vblank_event(crtc, crtc->state->event);
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index d1f305694367..75ab17af13a9 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -18,10 +18,11 @@
 #include <linux/wait.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "rcar_du_drv.h"
 #include "rcar_du_kms.h"
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.h b/drivers/gpu/drm/rcar-du/rcar_du_drv.h
index 6c187d0bf7c2..1327cd0df90a 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.h
@@ -22,6 +22,7 @@ struct device;
 struct drm_device;
 struct drm_property;
 struct rcar_du_device;
+struct rcar_du_encoder;
 
 #define RCAR_DU_FEATURE_CRTC_IRQ_CLOCK	BIT(0)	/* Per-CRTC IRQ and clock */
 #define RCAR_DU_FEATURE_VSP1_SOURCE	BIT(1)	/* Has inputs from VSP1 */
@@ -81,6 +82,8 @@ struct rcar_du_device {
 	struct rcar_du_crtc crtcs[RCAR_DU_MAX_CRTCS];
 	unsigned int num_crtcs;
 
+	struct rcar_du_encoder *encoders[RCAR_DU_OUTPUT_MAX];
+
 	struct rcar_du_group groups[RCAR_DU_MAX_GROUPS];
 	struct rcar_du_vsp vsps[RCAR_DU_MAX_VSPS];
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
index f16209499117..8ee4e762f4e5 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
@@ -10,7 +10,7 @@
 #include <linux/export.h>
 
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_panel.h>
 
 #include "rcar_du_drv.h"
@@ -30,8 +30,7 @@ static const struct drm_encoder_funcs encoder_funcs = {
 
 int rcar_du_encoder_init(struct rcar_du_device *rcdu,
 			 enum rcar_du_output output,
-			 struct device_node *enc_node,
-			 struct device_node *con_node)
+			 struct device_node *enc_node)
 {
 	struct rcar_du_encoder *renc;
 	struct drm_encoder *encoder;
@@ -42,6 +41,7 @@ int rcar_du_encoder_init(struct rcar_du_device *rcdu,
 	if (renc == NULL)
 		return -ENOMEM;
 
+	rcdu->encoders[output] = renc;
 	renc->output = output;
 	encoder = rcar_encoder_to_drm_encoder(renc);
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.h b/drivers/gpu/drm/rcar-du/rcar_du_encoder.h
index 552f2a02e5b5..df9be4524301 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.h
@@ -26,7 +26,6 @@ struct rcar_du_encoder {
 
 int rcar_du_encoder_init(struct rcar_du_device *rcdu,
 			 enum rcar_du_output output,
-			 struct device_node *enc_node,
-			 struct device_node *con_node);
+			 struct device_node *enc_node);
 
 #endif /* __RCAR_DU_ENCODER_H__ */
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
index e4b248e368d6..3b7d50a8fb9b 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
@@ -10,10 +10,12 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_vblank.h>
 
 #include <linux/of_graph.h>
 #include <linux/wait.h>
@@ -330,17 +332,10 @@ static int rcar_du_encoders_init_one(struct rcar_du_device *rcdu,
 				     enum rcar_du_output output,
 				     struct of_endpoint *ep)
 {
-	struct device_node *connector = NULL;
-	struct device_node *encoder = NULL;
-	struct device_node *ep_node = NULL;
-	struct device_node *entity_ep_node;
 	struct device_node *entity;
 	int ret;
 
-	/*
-	 * Locate the connected entity and infer its type from the number of
-	 * endpoints.
-	 */
+	/* Locate the connected entity and initialize the encoder. */
 	entity = of_graph_get_remote_port_parent(ep->local_node);
 	if (!entity) {
 		dev_dbg(rcdu->dev, "unconnected endpoint %pOF, skipping\n",
@@ -352,52 +347,17 @@ static int rcar_du_encoders_init_one(struct rcar_du_device *rcdu,
 		dev_dbg(rcdu->dev,
 			"connected entity %pOF is disabled, skipping\n",
 			entity);
+		of_node_put(entity);
 		return -ENODEV;
 	}
 
-	entity_ep_node = of_graph_get_remote_endpoint(ep->local_node);
-
-	for_each_endpoint_of_node(entity, ep_node) {
-		if (ep_node == entity_ep_node)
-			continue;
-
-		/*
-		 * We've found one endpoint other than the input, this must
-		 * be an encoder. Locate the connector.
-		 */
-		encoder = entity;
-		connector = of_graph_get_remote_port_parent(ep_node);
-		of_node_put(ep_node);
-
-		if (!connector) {
-			dev_warn(rcdu->dev,
-				 "no connector for encoder %pOF, skipping\n",
-				 encoder);
-			of_node_put(entity_ep_node);
-			of_node_put(encoder);
-			return -ENODEV;
-		}
-
-		break;
-	}
-
-	of_node_put(entity_ep_node);
-
-	if (!encoder) {
-		dev_warn(rcdu->dev,
-			 "no encoder found for endpoint %pOF, skipping\n",
-			 ep->local_node);
-		return -ENODEV;
-	}
-
-	ret = rcar_du_encoder_init(rcdu, output, encoder, connector);
+	ret = rcar_du_encoder_init(rcdu, output, entity);
 	if (ret && ret != -EPROBE_DEFER)
 		dev_warn(rcdu->dev,
 			 "failed to initialize encoder %pOF on output %u (%d), skipping\n",
-			 encoder, output, ret);
+			 entity, output, ret);
 
-	of_node_put(encoder);
-	of_node_put(connector);
+	of_node_put(entity);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.c b/drivers/gpu/drm/rcar-du/rcar_du_plane.c
index fa6b9aabc832..c6430027169f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_plane.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_plane.c
@@ -10,8 +10,9 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
index dec314a687e0..0878accbd134 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
@@ -9,11 +9,11 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_vblank.h>
 
 #include <linux/bitops.h>
 #include <linux/dma-mapping.h>
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c
index 96d749a35b25..7ef97b2a6eda 100644
--- a/drivers/gpu/drm/rcar-du/rcar_lvds.c
+++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c
@@ -10,6 +10,7 @@
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/io.h>
+#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/of_graph.h>
@@ -19,9 +20,10 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
+#include "rcar_lvds.h"
 #include "rcar_lvds_regs.h"
 
 struct rcar_lvds;
@@ -182,8 +184,9 @@ struct pll_info {
 
 static void rcar_lvds_d3_e3_pll_calc(struct rcar_lvds *lvds, struct clk *clk,
 				     unsigned long target, struct pll_info *pll,
-				     u32 clksel)
+				     u32 clksel, bool dot_clock_only)
 {
+	unsigned int div7 = dot_clock_only ? 1 : 7;
 	unsigned long output;
 	unsigned long fin;
 	unsigned int m_min;
@@ -217,9 +220,9 @@ static void rcar_lvds_d3_e3_pll_calc(struct rcar_lvds *lvds, struct clk *clk,
 	 *                     `------------> | |
 	 *                                    |/
 	 *
-	 * The /7 divider is optional when the LVDS PLL is used to generate a
-	 * dot clock for the DU RGB output, without using the LVDS encoder. We
-	 * don't support this configuration yet.
+	 * The /7 divider is optional, it is enabled when the LVDS PLL is used
+	 * to drive the LVDS encoder, and disabled when  used to generate a dot
+	 * clock for the DU RGB output, without using the LVDS encoder.
 	 *
 	 * The PLL allowed input frequency range is 12 MHz to 192 MHz.
 	 */
@@ -279,7 +282,7 @@ static void rcar_lvds_d3_e3_pll_calc(struct rcar_lvds *lvds, struct clk *clk,
 				 * the PLL, followed by a an optional fixed /7
 				 * divider.
 				 */
-				fout = fvco / (1 << e) / 7;
+				fout = fvco / (1 << e) / div7;
 				div = DIV_ROUND_CLOSEST(fout, target);
 				diff = abs(fout / div - target);
 
@@ -300,7 +303,7 @@ static void rcar_lvds_d3_e3_pll_calc(struct rcar_lvds *lvds, struct clk *clk,
 
 done:
 	output = fin * pll->pll_n / pll->pll_m / (1 << pll->pll_e)
-	       / 7 / pll->div;
+	       / div7 / pll->div;
 	error = (long)(output - target) * 10000 / (long)target;
 
 	dev_dbg(lvds->dev,
@@ -310,17 +313,18 @@ done:
 		pll->pll_m, pll->pll_n, pll->pll_e, pll->div);
 }
 
-static void rcar_lvds_pll_setup_d3_e3(struct rcar_lvds *lvds, unsigned int freq)
+static void __rcar_lvds_pll_setup_d3_e3(struct rcar_lvds *lvds,
+					unsigned int freq, bool dot_clock_only)
 {
 	struct pll_info pll = { .diff = (unsigned long)-1 };
 	u32 lvdpllcr;
 
 	rcar_lvds_d3_e3_pll_calc(lvds, lvds->clocks.dotclkin[0], freq, &pll,
-				 LVDPLLCR_CKSEL_DU_DOTCLKIN(0));
+				 LVDPLLCR_CKSEL_DU_DOTCLKIN(0), dot_clock_only);
 	rcar_lvds_d3_e3_pll_calc(lvds, lvds->clocks.dotclkin[1], freq, &pll,
-				 LVDPLLCR_CKSEL_DU_DOTCLKIN(1));
+				 LVDPLLCR_CKSEL_DU_DOTCLKIN(1), dot_clock_only);
 	rcar_lvds_d3_e3_pll_calc(lvds, lvds->clocks.extal, freq, &pll,
-				 LVDPLLCR_CKSEL_EXTAL);
+				 LVDPLLCR_CKSEL_EXTAL, dot_clock_only);
 
 	lvdpllcr = LVDPLLCR_PLLON | pll.clksel | LVDPLLCR_CLKOUT
 		 | LVDPLLCR_PLLN(pll.pll_n - 1) | LVDPLLCR_PLLM(pll.pll_m - 1);
@@ -329,6 +333,9 @@ static void rcar_lvds_pll_setup_d3_e3(struct rcar_lvds *lvds, unsigned int freq)
 		lvdpllcr |= LVDPLLCR_STP_CLKOUTE | LVDPLLCR_OUTCLKSEL
 			 |  LVDPLLCR_PLLE(pll.pll_e - 1);
 
+	if (dot_clock_only)
+		lvdpllcr |= LVDPLLCR_OCKSEL;
+
 	rcar_lvds_write(lvds, LVDPLLCR, lvdpllcr);
 
 	if (pll.div > 1)
@@ -342,6 +349,57 @@ static void rcar_lvds_pll_setup_d3_e3(struct rcar_lvds *lvds, unsigned int freq)
 		rcar_lvds_write(lvds, LVDDIV, 0);
 }
 
+static void rcar_lvds_pll_setup_d3_e3(struct rcar_lvds *lvds, unsigned int freq)
+{
+	__rcar_lvds_pll_setup_d3_e3(lvds, freq, false);
+}
+
+/* -----------------------------------------------------------------------------
+ * Clock - D3/E3 only
+ */
+
+int rcar_lvds_clk_enable(struct drm_bridge *bridge, unsigned long freq)
+{
+	struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge);
+	int ret;
+
+	if (WARN_ON(!(lvds->info->quirks & RCAR_LVDS_QUIRK_EXT_PLL)))
+		return -ENODEV;
+
+	dev_dbg(lvds->dev, "enabling LVDS PLL, freq=%luHz\n", freq);
+
+	WARN_ON(lvds->enabled);
+
+	ret = clk_prepare_enable(lvds->clocks.mod);
+	if (ret < 0)
+		return ret;
+
+	__rcar_lvds_pll_setup_d3_e3(lvds, freq, true);
+
+	lvds->enabled = true;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(rcar_lvds_clk_enable);
+
+void rcar_lvds_clk_disable(struct drm_bridge *bridge)
+{
+	struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge);
+
+	if (WARN_ON(!(lvds->info->quirks & RCAR_LVDS_QUIRK_EXT_PLL)))
+		return;
+
+	dev_dbg(lvds->dev, "disabling LVDS PLL\n");
+
+	WARN_ON(!lvds->enabled);
+
+	rcar_lvds_write(lvds, LVDPLLCR, 0);
+
+	clk_disable_unprepare(lvds->clocks.mod);
+
+	lvds->enabled = false;
+}
+EXPORT_SYMBOL_GPL(rcar_lvds_clk_disable);
+
 /* -----------------------------------------------------------------------------
  * Bridge
  */
@@ -544,7 +602,10 @@ static int rcar_lvds_attach(struct drm_bridge *bridge)
 		return drm_bridge_attach(bridge->encoder, lvds->next_bridge,
 					 bridge);
 
-	/* Otherwise we have a panel, create a connector. */
+	/* Otherwise if we have a panel, create a connector. */
+	if (!lvds->panel)
+		return 0;
+
 	ret = drm_connector_init(bridge->dev, connector, &rcar_lvds_conn_funcs,
 				 DRM_MODE_CONNECTOR_LVDS);
 	if (ret < 0)
@@ -592,7 +653,8 @@ static int rcar_lvds_parse_dt(struct rcar_lvds *lvds)
 	local_output = of_graph_get_endpoint_by_regs(lvds->dev->of_node, 1, 0);
 	if (!local_output) {
 		dev_dbg(lvds->dev, "unconnected port@1\n");
-		return -ENODEV;
+		ret = -ENODEV;
+		goto done;
 	}
 
 	/*
@@ -642,6 +704,15 @@ done:
 	of_node_put(remote_input);
 	of_node_put(remote);
 
+	/*
+	 * On D3/E3 the LVDS encoder provides a clock to the DU, which can be
+	 * used for the DPAD output even when the LVDS output is not connected.
+	 * Don't fail probe in that case as the DU will need the bridge to
+	 * control the clock.
+	 */
+	if (lvds->info->quirks & RCAR_LVDS_QUIRK_EXT_PLL)
+		return ret == -ENODEV ? 0 : ret;
+
 	return ret;
 }
 
@@ -785,6 +856,7 @@ static const struct rcar_lvds_device_info rcar_lvds_r8a77995_info = {
 
 static const struct of_device_id rcar_lvds_of_table[] = {
 	{ .compatible = "renesas,r8a7743-lvds", .data = &rcar_lvds_gen2_info },
+	{ .compatible = "renesas,r8a7744-lvds", .data = &rcar_lvds_gen2_info },
 	{ .compatible = "renesas,r8a774c0-lvds", .data = &rcar_lvds_r8a77990_info },
 	{ .compatible = "renesas,r8a7790-lvds", .data = &rcar_lvds_r8a7790_info },
 	{ .compatible = "renesas,r8a7791-lvds", .data = &rcar_lvds_gen2_info },
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.h b/drivers/gpu/drm/rcar-du/rcar_lvds.h
new file mode 100644
index 000000000000..a709cae1bc32
--- /dev/null
+++ b/drivers/gpu/drm/rcar-du/rcar_lvds.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * rcar_lvds.h  --  R-Car LVDS Encoder
+ *
+ * Copyright (C) 2013-2018 Renesas Electronics Corporation
+ *
+ * Contact: Laurent Pinchart (laurent.pinchart@ideasonboard.com)
+ */
+
+#ifndef __RCAR_LVDS_H__
+#define __RCAR_LVDS_H__
+
+struct drm_bridge;
+
+#if IS_ENABLED(CONFIG_DRM_RCAR_LVDS)
+int rcar_lvds_clk_enable(struct drm_bridge *bridge, unsigned long freq);
+void rcar_lvds_clk_disable(struct drm_bridge *bridge);
+#else
+static inline int rcar_lvds_clk_enable(struct drm_bridge *bridge,
+				       unsigned long freq)
+{
+	return -ENOSYS;
+}
+static inline void rcar_lvds_clk_disable(struct drm_bridge *bridge) { }
+#endif /* CONFIG_DRM_RCAR_LVDS */
+
+#endif /* __RCAR_LVDS_H__ */
diff --git a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
index 080f05352195..bc4423624209 100644
--- a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
+++ b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
@@ -21,10 +21,10 @@
 #include <linux/clk.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include <video/of_videomode.h>
 #include <video/videomode.h>
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
index 8ad0d773dc33..f7b9d45aa1d6 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
@@ -14,10 +14,10 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/clk.h>
 #include <linux/component.h>
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.h b/drivers/gpu/drm/rockchip/cdn-dp-core.h
index f57e296401b8..48fef95cb3c6 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.h
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.h
@@ -16,9 +16,9 @@
 #define _CDN_DP_CORE_H
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include "rockchip_drm_drv.h"
 
 #define MAX_PHY		2
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-reg.c b/drivers/gpu/drm/rockchip/cdn-dp-reg.c
index 5a485489a1e2..6c8b14fb1d2f 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-reg.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-reg.c
@@ -113,7 +113,7 @@ static int cdp_dp_mailbox_write(struct cdn_dp_device *dp, u8 val)
 
 static int cdn_dp_mailbox_validate_receive(struct cdn_dp_device *dp,
 					   u8 module_id, u8 opcode,
-					   u8 req_size)
+					   u16 req_size)
 {
 	u32 mbox_size, i;
 	u8 header[4];
diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
index 89c63cfde5c8..4cdc9f86c2e5 100644
--- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
@@ -16,8 +16,8 @@
 
 #include <drm/drm_of.h>
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/bridge/dw_hdmi.h>
 
 #include "rockchip_drm_drv.h"
diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.c b/drivers/gpu/drm/rockchip/inno_hdmi.c
index 27c945e030a0..ce1545862b6c 100644
--- a/drivers/gpu/drm/rockchip/inno_hdmi.c
+++ b/drivers/gpu/drm/rockchip/inno_hdmi.c
@@ -26,8 +26,8 @@
 #include <drm/drm_of.h>
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include "rockchip_drm_drv.h"
 #include "rockchip_drm_vop.h"
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index be6c2573039a..d7fa17f12769 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -15,10 +15,10 @@
  */
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/dma-mapping.h>
 #include <linux/dma-iommu.h>
 #include <linux/pm_runtime.h>
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
index 08bec50d9c5d..97438bbbe389 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c
@@ -17,8 +17,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "rockchip_drm_drv.h"
 #include "rockchip_drm_fb.h"
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
index e6650553f5d6..8ce68bd508be 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
@@ -15,7 +15,7 @@
 #include <drm/drm.h>
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "rockchip_drm_drv.h"
 #include "rockchip_drm_gem.h"
@@ -91,7 +91,6 @@ static int rockchip_drm_fbdev_create(struct drm_fb_helper *helper,
 	}
 
 	fbi->par = helper;
-	fbi->flags = FBINFO_FLAG_DEFAULT;
 	fbi->fbops = &rockchip_drm_fbdev_ops;
 
 	fb = helper->fb;
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_psr.c b/drivers/gpu/drm/rockchip/rockchip_drm_psr.c
index 22a70ab6e214..a0c8bd235b67 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_psr.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_psr.c
@@ -14,7 +14,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "rockchip_drm_drv.h"
 #include "rockchip_drm_psr.h"
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index 619b6db05d58..c7d4c6073ea5 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -17,10 +17,10 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_uapi.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_flip_work.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #ifdef CONFIG_DRM_ANALOGIX_DP
 #include <drm/bridge/analogix_dp.h>
 #endif
@@ -1675,7 +1675,9 @@ static void vop_win_init(struct vop *vop)
 
 		vop_win->data = win_data;
 		vop_win->vop = vop;
-		vop_win->yuv2yuv_data = &vop_data->win_yuv2yuv[i];
+
+		if (vop_data->win_yuv2yuv)
+			vop_win->yuv2yuv_data = &vop_data->win_yuv2yuv[i];
 	}
 }
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c
index 456bd9f13bae..e52dd5a8529e 100644
--- a/drivers/gpu/drm/rockchip/rockchip_lvds.c
+++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c
@@ -16,10 +16,10 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/component.h>
 #include <linux/clk.h>
diff --git a/drivers/gpu/drm/rockchip/rockchip_rgb.c b/drivers/gpu/drm/rockchip/rockchip_rgb.c
index 96ac1458a59c..ce4d82d293e4 100644
--- a/drivers/gpu/drm/rockchip/rockchip_rgb.c
+++ b/drivers/gpu/drm/rockchip/rockchip_rgb.c
@@ -1,25 +1,16 @@
-//SPDX-License-Identifier: GPL-2.0+
+// SPDX-License-Identifier: GPL-2.0
 /*
  * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd
  * Author:
  *      Sandy Huang <hjc@rock-chips.com>
- *
- * This software is licensed under the terms of the GNU General Public
- * License version 2, as published by the Free Software Foundation, and
- * may be copied, distributed, and modified under those terms.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
  */
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/component.h>
 #include <linux/of_graph.h>
@@ -113,8 +104,10 @@ struct rockchip_rgb *rockchip_rgb_init(struct device *dev,
 		child_count++;
 		ret = drm_of_find_panel_or_bridge(dev->of_node, 0, endpoint_id,
 						  &panel, &bridge);
-		if (!ret)
+		if (!ret) {
+			of_node_put(endpoint);
 			break;
+		}
 	}
 
 	of_node_put(port);
diff --git a/drivers/gpu/drm/rockchip/rockchip_rgb.h b/drivers/gpu/drm/rockchip/rockchip_rgb.h
index 38b52e63b2b0..27b9635124bc 100644
--- a/drivers/gpu/drm/rockchip/rockchip_rgb.h
+++ b/drivers/gpu/drm/rockchip/rockchip_rgb.h
@@ -1,17 +1,8 @@
-//SPDX-License-Identifier: GPL-2.0+
+/* SPDX-License-Identifier: GPL-2.0 */
 /*
  * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd
  * Author:
  *      Sandy Huang <hjc@rock-chips.com>
- *
- * This software is licensed under the terms of the GNU General Public
- * License version 2, as published by the Free Software Foundation, and
- * may be copied, distributed, and modified under those terms.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
  */
 
 #ifdef CONFIG_ROCKCHIP_RGB
diff --git a/drivers/gpu/drm/savage/savage_state.c b/drivers/gpu/drm/savage/savage_state.c
index 7559a820bd43..ebb8b7d32b33 100644
--- a/drivers/gpu/drm/savage/savage_state.c
+++ b/drivers/gpu/drm/savage/savage_state.c
@@ -299,6 +299,7 @@ static int savage_dispatch_dma_prim(drm_savage_private_t * dev_priv,
 	case SAVAGE_PRIM_TRILIST_201:
 		reorder = 1;
 		prim = SAVAGE_PRIM_TRILIST;
+		/* fall through */
 	case SAVAGE_PRIM_TRILIST:
 		if (n % 3 != 0) {
 			DRM_ERROR("wrong number of vertices %u in TRILIST\n",
@@ -436,6 +437,7 @@ static int savage_dispatch_vb_prim(drm_savage_private_t * dev_priv,
 	case SAVAGE_PRIM_TRILIST_201:
 		reorder = 1;
 		prim = SAVAGE_PRIM_TRILIST;
+		/* fall through */
 	case SAVAGE_PRIM_TRILIST:
 		if (n % 3 != 0) {
 			DRM_ERROR("wrong number of vertices %u in TRILIST\n",
@@ -557,6 +559,7 @@ static int savage_dispatch_dma_idx(drm_savage_private_t * dev_priv,
 	case SAVAGE_PRIM_TRILIST_201:
 		reorder = 1;
 		prim = SAVAGE_PRIM_TRILIST;
+		/* fall through */
 	case SAVAGE_PRIM_TRILIST:
 		if (n % 3 != 0) {
 			DRM_ERROR("wrong number of indices %u in TRILIST\n", n);
@@ -695,6 +698,7 @@ static int savage_dispatch_vb_idx(drm_savage_private_t * dev_priv,
 	case SAVAGE_PRIM_TRILIST_201:
 		reorder = 1;
 		prim = SAVAGE_PRIM_TRILIST;
+		/* fall through */
 	case SAVAGE_PRIM_TRILIST:
 		if (n % 3 != 0) {
 			DRM_ERROR("wrong number of indices %u in TRILIST\n", n);
diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 4463d3826ecb..e2942c9a11a7 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -440,13 +440,10 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 
 	while ((entity->dependency =
 			sched->ops->dependency(sched_job, entity))) {
+		trace_drm_sched_job_wait_dep(sched_job, entity->dependency);
 
-		if (drm_sched_entity_add_dependency_cb(entity)) {
-
-			trace_drm_sched_job_wait_dep(sched_job,
-						     entity->dependency);
+		if (drm_sched_entity_add_dependency_cb(entity))
 			return NULL;
-		}
 	}
 
 	/* skip jobs from entity that marked guilty */
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index dbb69063b3d5..19fc601c9eeb 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -60,8 +60,6 @@
 
 static void drm_sched_process_job(struct dma_fence *f, struct dma_fence_cb *cb);
 
-static void drm_sched_expel_job_unlocked(struct drm_sched_job *s_job);
-
 /**
  * drm_sched_rq_init - initialize a given run queue struct
  *
@@ -286,8 +284,6 @@ static void drm_sched_job_finish(struct work_struct *work)
 	cancel_delayed_work_sync(&sched->work_tdr);
 
 	spin_lock_irqsave(&sched->job_list_lock, flags);
-	/* remove job from ring_mirror_list */
-	list_del_init(&s_job->node);
 	/* queue TDR for next job */
 	drm_sched_start_timeout(sched);
 	spin_unlock_irqrestore(&sched->job_list_lock, flags);
@@ -295,22 +291,11 @@ static void drm_sched_job_finish(struct work_struct *work)
 	sched->ops->free_job(s_job);
 }
 
-static void drm_sched_job_finish_cb(struct dma_fence *f,
-				    struct dma_fence_cb *cb)
-{
-	struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
-						 finish_cb);
-	schedule_work(&job->finish_work);
-}
-
 static void drm_sched_job_begin(struct drm_sched_job *s_job)
 {
 	struct drm_gpu_scheduler *sched = s_job->sched;
 	unsigned long flags;
 
-	dma_fence_add_callback(&s_job->s_fence->finished, &s_job->finish_cb,
-			       drm_sched_job_finish_cb);
-
 	spin_lock_irqsave(&sched->job_list_lock, flags);
 	list_add_tail(&s_job->node, &sched->ring_mirror_list);
 	drm_sched_start_timeout(sched);
@@ -335,6 +320,51 @@ static void drm_sched_job_timedout(struct work_struct *work)
 	spin_unlock_irqrestore(&sched->job_list_lock, flags);
 }
 
+ /**
+  * drm_sched_increase_karma - Update sched_entity guilty flag
+  *
+  * @bad: The job guilty of time out
+  *
+  * Increment on every hang caused by the 'bad' job. If this exceeds the hang
+  * limit of the scheduler then the respective sched entity is marked guilty and
+  * jobs from it will not be scheduled further
+  */
+void drm_sched_increase_karma(struct drm_sched_job *bad)
+{
+	int i;
+	struct drm_sched_entity *tmp;
+	struct drm_sched_entity *entity;
+	struct drm_gpu_scheduler *sched = bad->sched;
+
+	/* don't increase @bad's karma if it's from KERNEL RQ,
+	 * because sometimes GPU hang would cause kernel jobs (like VM updating jobs)
+	 * corrupt but keep in mind that kernel jobs always considered good.
+	 */
+	if (bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) {
+		atomic_inc(&bad->karma);
+		for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_KERNEL;
+		     i++) {
+			struct drm_sched_rq *rq = &sched->sched_rq[i];
+
+			spin_lock(&rq->lock);
+			list_for_each_entry_safe(entity, tmp, &rq->entities, list) {
+				if (bad->s_fence->scheduled.context ==
+				    entity->fence_context) {
+					if (atomic_read(&bad->karma) >
+					    bad->sched->hang_limit)
+						if (entity->guilty)
+							atomic_set(entity->guilty, 1);
+					break;
+				}
+			}
+			spin_unlock(&rq->lock);
+			if (&entity->list != &rq->entities)
+				break;
+		}
+	}
+}
+EXPORT_SYMBOL(drm_sched_increase_karma);
+
 /**
  * drm_sched_hw_job_reset - stop the scheduler if it contains the bad job
  *
@@ -342,50 +372,42 @@ static void drm_sched_job_timedout(struct work_struct *work)
  * @bad: bad scheduler job
  *
  */
-void drm_sched_hw_job_reset(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad)
+void drm_sched_stop(struct drm_gpu_scheduler *sched)
 {
 	struct drm_sched_job *s_job;
-	struct drm_sched_entity *entity, *tmp;
 	unsigned long flags;
-	int i;
+	struct dma_fence *last_fence =  NULL;
 
+	kthread_park(sched->thread);
+
+	/*
+	 * Verify all the signaled jobs in mirror list are removed from the ring
+	 * by waiting for the latest job to enter the list. This should insure that
+	 * also all the previous jobs that were in flight also already singaled
+	 * and removed from the list.
+	 */
 	spin_lock_irqsave(&sched->job_list_lock, flags);
 	list_for_each_entry_reverse(s_job, &sched->ring_mirror_list, node) {
 		if (s_job->s_fence->parent &&
 		    dma_fence_remove_callback(s_job->s_fence->parent,
-					      &s_job->s_fence->cb)) {
+					      &s_job->cb)) {
 			dma_fence_put(s_job->s_fence->parent);
 			s_job->s_fence->parent = NULL;
 			atomic_dec(&sched->hw_rq_count);
+		} else {
+			 last_fence = dma_fence_get(&s_job->s_fence->finished);
+			 break;
 		}
 	}
 	spin_unlock_irqrestore(&sched->job_list_lock, flags);
 
-	if (bad && bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) {
-		atomic_inc(&bad->karma);
-		/* don't increase @bad's karma if it's from KERNEL RQ,
-		 * becuase sometimes GPU hang would cause kernel jobs (like VM updating jobs)
-		 * corrupt but keep in mind that kernel jobs always considered good.
-		 */
-		for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_KERNEL; i++ ) {
-			struct drm_sched_rq *rq = &sched->sched_rq[i];
-
-			spin_lock(&rq->lock);
-			list_for_each_entry_safe(entity, tmp, &rq->entities, list) {
-				if (bad->s_fence->scheduled.context == entity->fence_context) {
-				    if (atomic_read(&bad->karma) > bad->sched->hang_limit)
-						if (entity->guilty)
-							atomic_set(entity->guilty, 1);
-					break;
-				}
-			}
-			spin_unlock(&rq->lock);
-			if (&entity->list != &rq->entities)
-				break;
-		}
+	if (last_fence) {
+		dma_fence_wait(last_fence, false);
+		dma_fence_put(last_fence);
 	}
 }
-EXPORT_SYMBOL(drm_sched_hw_job_reset);
+
+EXPORT_SYMBOL(drm_sched_stop);
 
 /**
  * drm_sched_job_recovery - recover jobs after a reset
@@ -393,18 +415,58 @@ EXPORT_SYMBOL(drm_sched_hw_job_reset);
  * @sched: scheduler instance
  *
  */
-void drm_sched_job_recovery(struct drm_gpu_scheduler *sched)
+void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery)
 {
 	struct drm_sched_job *s_job, *tmp;
-	bool found_guilty = false;
-	unsigned long flags;
 	int r;
 
-	spin_lock_irqsave(&sched->job_list_lock, flags);
+	if (!full_recovery)
+		goto unpark;
+
+	/*
+	 * Locking the list is not required here as the sched thread is parked
+	 * so no new jobs are being pushed in to HW and in drm_sched_stop we
+	 * flushed all the jobs who were still in mirror list but who already
+	 * signaled and removed them self from the list. Also concurrent
+	 * GPU recovers can't run in parallel.
+	 */
+	list_for_each_entry_safe(s_job, tmp, &sched->ring_mirror_list, node) {
+		struct dma_fence *fence = s_job->s_fence->parent;
+
+		if (fence) {
+			r = dma_fence_add_callback(fence, &s_job->cb,
+						   drm_sched_process_job);
+			if (r == -ENOENT)
+				drm_sched_process_job(fence, &s_job->cb);
+			else if (r)
+				DRM_ERROR("fence add callback failed (%d)\n",
+					  r);
+		} else
+			drm_sched_process_job(NULL, &s_job->cb);
+	}
+
+	drm_sched_start_timeout(sched);
+
+unpark:
+	kthread_unpark(sched->thread);
+}
+EXPORT_SYMBOL(drm_sched_start);
+
+/**
+ * drm_sched_resubmit_jobs - helper to relunch job from mirror ring list
+ *
+ * @sched: scheduler instance
+ *
+ */
+void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched)
+{
+	struct drm_sched_job *s_job, *tmp;
+	uint64_t guilty_context;
+	bool found_guilty = false;
+
+	/*TODO DO we need spinlock here ? */
 	list_for_each_entry_safe(s_job, tmp, &sched->ring_mirror_list, node) {
 		struct drm_sched_fence *s_fence = s_job->s_fence;
-		struct dma_fence *fence;
-		uint64_t guilty_context;
 
 		if (!found_guilty && atomic_read(&s_job->karma) > sched->hang_limit) {
 			found_guilty = true;
@@ -414,31 +476,11 @@ void drm_sched_job_recovery(struct drm_gpu_scheduler *sched)
 		if (found_guilty && s_job->s_fence->scheduled.context == guilty_context)
 			dma_fence_set_error(&s_fence->finished, -ECANCELED);
 
-		spin_unlock_irqrestore(&sched->job_list_lock, flags);
-		fence = sched->ops->run_job(s_job);
+		s_job->s_fence->parent = sched->ops->run_job(s_job);
 		atomic_inc(&sched->hw_rq_count);
-
-		if (fence) {
-			s_fence->parent = dma_fence_get(fence);
-			r = dma_fence_add_callback(fence, &s_fence->cb,
-						   drm_sched_process_job);
-			if (r == -ENOENT)
-				drm_sched_process_job(fence, &s_fence->cb);
-			else if (r)
-				DRM_ERROR("fence add callback failed (%d)\n",
-					  r);
-			dma_fence_put(fence);
-		} else {
-			if (s_fence->finished.error < 0)
-				drm_sched_expel_job_unlocked(s_job);
-			drm_sched_process_job(NULL, &s_fence->cb);
-		}
-		spin_lock_irqsave(&sched->job_list_lock, flags);
 	}
-	drm_sched_start_timeout(sched);
-	spin_unlock_irqrestore(&sched->job_list_lock, flags);
 }
-EXPORT_SYMBOL(drm_sched_job_recovery);
+EXPORT_SYMBOL(drm_sched_resubmit_jobs);
 
 /**
  * drm_sched_job_init - init a scheduler job
@@ -552,18 +594,27 @@ drm_sched_select_entity(struct drm_gpu_scheduler *sched)
  */
 static void drm_sched_process_job(struct dma_fence *f, struct dma_fence_cb *cb)
 {
-	struct drm_sched_fence *s_fence =
-		container_of(cb, struct drm_sched_fence, cb);
+	struct drm_sched_job *s_job = container_of(cb, struct drm_sched_job, cb);
+	struct drm_sched_fence *s_fence = s_job->s_fence;
 	struct drm_gpu_scheduler *sched = s_fence->sched;
+	unsigned long flags;
+
+	cancel_delayed_work(&sched->work_tdr);
 
-	dma_fence_get(&s_fence->finished);
 	atomic_dec(&sched->hw_rq_count);
 	atomic_dec(&sched->num_jobs);
+
+	spin_lock_irqsave(&sched->job_list_lock, flags);
+	/* remove job from ring_mirror_list */
+	list_del_init(&s_job->node);
+	spin_unlock_irqrestore(&sched->job_list_lock, flags);
+
 	drm_sched_fence_finished(s_fence);
 
 	trace_drm_sched_process_job(s_fence);
-	dma_fence_put(&s_fence->finished);
 	wake_up_interruptible(&sched->wake_up_worker);
+
+	schedule_work(&s_job->finish_work);
 }
 
 /**
@@ -626,34 +677,22 @@ static int drm_sched_main(void *param)
 
 		if (fence) {
 			s_fence->parent = dma_fence_get(fence);
-			r = dma_fence_add_callback(fence, &s_fence->cb,
+			r = dma_fence_add_callback(fence, &sched_job->cb,
 						   drm_sched_process_job);
 			if (r == -ENOENT)
-				drm_sched_process_job(fence, &s_fence->cb);
+				drm_sched_process_job(fence, &sched_job->cb);
 			else if (r)
 				DRM_ERROR("fence add callback failed (%d)\n",
 					  r);
 			dma_fence_put(fence);
-		} else {
-			if (s_fence->finished.error < 0)
-				drm_sched_expel_job_unlocked(sched_job);
-			drm_sched_process_job(NULL, &s_fence->cb);
-		}
+		} else
+			drm_sched_process_job(NULL, &sched_job->cb);
 
 		wake_up(&sched->job_scheduled);
 	}
 	return 0;
 }
 
-static void drm_sched_expel_job_unlocked(struct drm_sched_job *s_job)
-{
-	struct drm_gpu_scheduler *sched = s_job->sched;
-
-	spin_lock(&sched->job_list_lock);
-	list_del_init(&s_job->node);
-	spin_unlock(&sched->job_list_lock);
-}
-
 /**
  * drm_sched_init - Init a gpu scheduler instance
  *
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index 499b5fdb869f..b6988a6d698e 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -16,6 +16,7 @@
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "shmob_drm_backlight.h"
 #include "shmob_drm_crtc.h"
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index f2cfd1698b78..cb821adfc321 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -18,6 +18,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "shmob_drm_drv.h"
 #include "shmob_drm_kms.h"
@@ -126,7 +127,7 @@ static irqreturn_t shmob_drm_irq(int irq, void *arg)
 DEFINE_DRM_GEM_CMA_FOPS(shmob_drm_fops);
 
 static struct drm_driver shmob_drm_driver = {
-	.driver_features	= DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET
+	.driver_features	= DRIVER_GEM | DRIVER_MODESET
 				| DRIVER_PRIME,
 	.irq_handler		= shmob_drm_irq,
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
index a17268444c6d..2e08bc203bf9 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_kms.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
@@ -13,6 +13,7 @@
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "shmob_drm_crtc.h"
 #include "shmob_drm_drv.h"
diff --git a/drivers/gpu/drm/sti/sti_crtc.c b/drivers/gpu/drm/sti/sti_crtc.c
index ec9f87483e39..387f0bed6c1c 100644
--- a/drivers/gpu/drm/sti/sti_crtc.c
+++ b/drivers/gpu/drm/sti/sti_crtc.c
@@ -11,8 +11,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sti_compositor.h"
 #include "sti_crtc.h"
diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/gpu/drm/sti/sti_drv.c
index ac54e0f9caea..a525fd899f68 100644
--- a/drivers/gpu/drm/sti/sti_drv.c
+++ b/drivers/gpu/drm/sti/sti_drv.c
@@ -14,12 +14,12 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sti_crtc.h"
 #include "sti_drv.h"
diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index 6d33772f77eb..b31cc2672d36 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -13,8 +13,8 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sti_awg_utils.h"
 #include "sti_drv.h"
diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
index a63dd5eb7081..ff9256673fc8 100644
--- a/drivers/gpu/drm/sti/sti_hda.c
+++ b/drivers/gpu/drm/sti/sti_hda.c
@@ -12,7 +12,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 /* HDformatter registers */
 #define HDA_ANA_CFG                     0x0000
diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index 458fcb5a93f2..6000df624980 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -15,8 +15,8 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 
 #include <sound/hdmi-codec.h>
 
diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c
index ea4a3b87fa55..c42f2fa7053c 100644
--- a/drivers/gpu/drm/sti/sti_tvout.c
+++ b/drivers/gpu/drm/sti/sti_tvout.c
@@ -15,7 +15,7 @@
 #include <linux/seq_file.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_atomic_helper.h>
 
 #include "sti_crtc.h"
 #include "sti_drv.h"
diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c
index 8dec001b9d37..0a7f933ab007 100644
--- a/drivers/gpu/drm/stm/drv.c
+++ b/drivers/gpu/drm/stm/drv.c
@@ -9,15 +9,19 @@
  */
 
 #include <linux/component.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
 #include <linux/of_platform.h>
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_fb_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_vblank.h>
 
 #include "ltdc.h"
 
diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index a91e04139595..b1741a9d5be2 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -10,18 +10,25 @@
 
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
 #include <linux/of_address.h>
 #include <linux/of_graph.h>
+#include <linux/platform_device.h>
 #include <linux/reset.h>
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_device.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_of.h>
-#include <drm/drm_bridge.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_vblank.h>
 
 #include <video/videomode.h>
 
diff --git a/drivers/gpu/drm/sun4i/Kconfig b/drivers/gpu/drm/sun4i/Kconfig
index c2c042287c19..1dbbc3a1b763 100644
--- a/drivers/gpu/drm/sun4i/Kconfig
+++ b/drivers/gpu/drm/sun4i/Kconfig
@@ -45,10 +45,11 @@ config DRM_SUN6I_DSI
 	default MACH_SUN8I
 	select CRC_CCITT
 	select DRM_MIPI_DSI
+	select PHY_SUN6I_MIPI_DPHY
 	help
 	  Choose this option if you want have an Allwinner SoC with
 	  MIPI-DSI support. If M is selected the module will be called
-	  sun6i-dsi
+	  sun6i_mipi_dsi.
 
 config DRM_SUN8I_DW_HDMI
 	tristate "Support for Allwinner version of DesignWare HDMI"
diff --git a/drivers/gpu/drm/sun4i/Makefile b/drivers/gpu/drm/sun4i/Makefile
index 0eb38ac8e86e..0d04f2447b01 100644
--- a/drivers/gpu/drm/sun4i/Makefile
+++ b/drivers/gpu/drm/sun4i/Makefile
@@ -24,9 +24,6 @@ sun4i-tcon-y			+= sun4i_lvds.o
 sun4i-tcon-y			+= sun4i_tcon.o
 sun4i-tcon-y			+= sun4i_rgb.o
 
-sun6i-dsi-y			+= sun6i_mipi_dphy.o
-sun6i-dsi-y			+= sun6i_mipi_dsi.o
-
 obj-$(CONFIG_DRM_SUN4I)		+= sun4i-drm.o
 obj-$(CONFIG_DRM_SUN4I)		+= sun4i-tcon.o
 obj-$(CONFIG_DRM_SUN4I)		+= sun4i_tv.o
@@ -37,7 +34,7 @@ ifdef CONFIG_DRM_SUN4I_BACKEND
 obj-$(CONFIG_DRM_SUN4I)		+= sun4i-frontend.o
 endif
 obj-$(CONFIG_DRM_SUN4I_HDMI)	+= sun4i-drm-hdmi.o
-obj-$(CONFIG_DRM_SUN6I_DSI)	+= sun6i-dsi.o
+obj-$(CONFIG_DRM_SUN6I_DSI)	+= sun6i_mipi_dsi.o
 obj-$(CONFIG_DRM_SUN8I_DW_HDMI)	+= sun8i-drm-hdmi.o
 obj-$(CONFIG_DRM_SUN8I_MIXER)	+= sun8i-mixer.o
 obj-$(CONFIG_DRM_SUN8I_TCON_TOP) += sun8i_tcon_top.o
diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c
index 892197f52557..4c0d51f73237 100644
--- a/drivers/gpu/drm/sun4i/sun4i_backend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_backend.c
@@ -14,10 +14,10 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/component.h>
 #include <linux/list.h>
@@ -141,7 +141,6 @@ static const uint32_t sun4i_backend_formats[] = {
 	DRM_FORMAT_ARGB1555,
 	DRM_FORMAT_ARGB4444,
 	DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_BGRX8888,
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_RGB888,
 	DRM_FORMAT_RGBA4444,
@@ -765,17 +764,18 @@ static struct sun4i_frontend *sun4i_backend_find_frontend(struct sun4i_drv *drv,
 		remote = of_graph_get_remote_port_parent(ep);
 		if (!remote)
 			continue;
+		of_node_put(remote);
 
 		/* does this node match any registered engines? */
 		list_for_each_entry(frontend, &drv->frontend_list, list) {
 			if (remote == frontend->node) {
-				of_node_put(remote);
 				of_node_put(port);
+				of_node_put(ep);
 				return frontend;
 			}
 		}
 	}
-
+	of_node_put(port);
 	return ERR_PTR(-EINVAL);
 }
 
@@ -1013,6 +1013,10 @@ static const struct of_device_id sun4i_backend_of_table[] = {
 		.data = &sun7i_backend_quirks,
 	},
 	{
+		.compatible = "allwinner,sun8i-a23-display-backend",
+		.data = &sun8i_a33_backend_quirks,
+	},
+	{
 		.compatible = "allwinner,sun8i-a33-display-backend",
 		.data = &sun8i_a33_backend_quirks,
 	},
diff --git a/drivers/gpu/drm/sun4i/sun4i_crtc.c b/drivers/gpu/drm/sun4i/sun4i_crtc.c
index 3eedf335a935..cdb881e34470 100644
--- a/drivers/gpu/drm/sun4i/sun4i_crtc.c
+++ b/drivers/gpu/drm/sun4i/sun4i_crtc.c
@@ -13,8 +13,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_modes.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/clk-provider.h>
 #include <linux/ioport.h>
diff --git a/drivers/gpu/drm/sun4i/sun4i_drv.c b/drivers/gpu/drm/sun4i/sun4i_drv.c
index 93da837194cf..3ebd9f5e2719 100644
--- a/drivers/gpu/drm/sun4i/sun4i_drv.c
+++ b/drivers/gpu/drm/sun4i/sun4i_drv.c
@@ -16,11 +16,11 @@
 #include <linux/of_reserved_mem.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sun4i_drv.h"
 #include "sun4i_frontend.h"
@@ -165,6 +165,7 @@ static bool sun4i_drv_node_is_frontend(struct device_node *node)
 		of_device_is_compatible(node, "allwinner,sun5i-a13-display-frontend") ||
 		of_device_is_compatible(node, "allwinner,sun6i-a31-display-frontend") ||
 		of_device_is_compatible(node, "allwinner,sun7i-a20-display-frontend") ||
+		of_device_is_compatible(node, "allwinner,sun8i-a23-display-frontend") ||
 		of_device_is_compatible(node, "allwinner,sun8i-a33-display-frontend") ||
 		of_device_is_compatible(node, "allwinner,sun9i-a80-display-frontend");
 }
@@ -404,6 +405,7 @@ static const struct of_device_id sun4i_drv_of_table[] = {
 	{ .compatible = "allwinner,sun6i-a31-display-engine" },
 	{ .compatible = "allwinner,sun6i-a31s-display-engine" },
 	{ .compatible = "allwinner,sun7i-a20-display-engine" },
+	{ .compatible = "allwinner,sun8i-a23-display-engine" },
 	{ .compatible = "allwinner,sun8i-a33-display-engine" },
 	{ .compatible = "allwinner,sun8i-a83t-display-engine" },
 	{ .compatible = "allwinner,sun8i-h3-display-engine" },
diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.c b/drivers/gpu/drm/sun4i/sun4i_frontend.c
index e8239d4d4dd5..346c8071bd38 100644
--- a/drivers/gpu/drm/sun4i/sun4i_frontend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_frontend.c
@@ -720,6 +720,10 @@ const struct of_device_id sun4i_frontend_of_table[] = {
 		.data = &sun4i_a10_frontend
 	},
 	{
+		.compatible = "allwinner,sun8i-a23-display-frontend",
+		.data = &sun8i_a33_frontend
+	},
+	{
 		.compatible = "allwinner,sun8i-a33-display-frontend",
 		.data = &sun8i_a33_frontend
 	},
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
index 554a6f4561f3..d18862629301 100644
--- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
@@ -11,7 +11,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_of.h>
@@ -93,6 +93,8 @@ static void sun4i_hdmi_disable(struct drm_encoder *encoder)
 	val = readl(hdmi->base + SUN4I_HDMI_VID_CTRL_REG);
 	val &= ~SUN4I_HDMI_VID_CTRL_ENABLE;
 	writel(val, hdmi->base + SUN4I_HDMI_VID_CTRL_REG);
+
+	clk_disable_unprepare(hdmi->tmds_clk);
 }
 
 static void sun4i_hdmi_enable(struct drm_encoder *encoder)
@@ -103,6 +105,8 @@ static void sun4i_hdmi_enable(struct drm_encoder *encoder)
 
 	DRM_DEBUG_DRIVER("Enabling the HDMI Output\n");
 
+	clk_prepare_enable(hdmi->tmds_clk);
+
 	sun4i_hdmi_setup_avi_infoframes(hdmi, mode);
 	val |= SUN4I_HDMI_PKT_CTRL_TYPE(0, SUN4I_HDMI_PKT_AVI);
 	val |= SUN4I_HDMI_PKT_CTRL_TYPE(1, SUN4I_HDMI_PKT_END);
diff --git a/drivers/gpu/drm/sun4i/sun4i_layer.c b/drivers/gpu/drm/sun4i/sun4i_layer.c
index c5a999ca1d72..a514fe88d441 100644
--- a/drivers/gpu/drm/sun4i/sun4i_layer.c
+++ b/drivers/gpu/drm/sun4i/sun4i_layer.c
@@ -117,6 +117,11 @@ static void sun4i_backend_layer_atomic_update(struct drm_plane *plane,
 static bool sun4i_layer_format_mod_supported(struct drm_plane *plane,
 					     uint32_t format, uint64_t modifier)
 {
+	struct sun4i_layer *layer = plane_to_sun4i_layer(plane);
+
+	if (IS_ERR_OR_NULL(layer->backend->frontend))
+		sun4i_backend_format_is_supported(format, modifier);
+
 	return sun4i_backend_format_is_supported(format, modifier) ||
 	       sun4i_frontend_format_is_supported(format, modifier);
 }
@@ -165,6 +170,21 @@ static const uint32_t sun4i_layer_formats[] = {
 	DRM_FORMAT_YVYU,
 };
 
+static const uint32_t sun4i_backend_layer_formats[] = {
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_RGBA5551,
+	DRM_FORMAT_RGBA4444,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+};
+
 static const uint64_t sun4i_layer_modifiers[] = {
 	DRM_FORMAT_MOD_LINEAR,
 	DRM_FORMAT_MOD_ALLWINNER_TILED,
@@ -175,6 +195,9 @@ static struct sun4i_layer *sun4i_layer_init_one(struct drm_device *drm,
 						struct sun4i_backend *backend,
 						enum drm_plane_type type)
 {
+	const uint64_t *modifiers = sun4i_layer_modifiers;
+	const uint32_t *formats = sun4i_layer_formats;
+	unsigned int formats_len = ARRAY_SIZE(sun4i_layer_formats);
 	struct sun4i_layer *layer;
 	int ret;
 
@@ -182,12 +205,19 @@ static struct sun4i_layer *sun4i_layer_init_one(struct drm_device *drm,
 	if (!layer)
 		return ERR_PTR(-ENOMEM);
 
+	layer->backend = backend;
+
+	if (IS_ERR_OR_NULL(backend->frontend)) {
+		formats = sun4i_backend_layer_formats;
+		formats_len = ARRAY_SIZE(sun4i_backend_layer_formats);
+		modifiers = NULL;
+	}
+
 	/* possible crtcs are set later */
 	ret = drm_universal_plane_init(drm, &layer->plane, 0,
 				       &sun4i_backend_layer_funcs,
-				       sun4i_layer_formats,
-				       ARRAY_SIZE(sun4i_layer_formats),
-				       sun4i_layer_modifiers, type, NULL);
+				       formats, formats_len,
+				       modifiers, type, NULL);
 	if (ret) {
 		dev_err(drm->dev, "Couldn't initialize layer\n");
 		return ERR_PTR(ret);
@@ -195,7 +225,6 @@ static struct sun4i_layer *sun4i_layer_init_one(struct drm_device *drm,
 
 	drm_plane_helper_add(&layer->plane,
 			     &sun4i_backend_layer_helper_funcs);
-	layer->backend = backend;
 
 	drm_plane_create_alpha_property(&layer->plane);
 	drm_plane_create_zpos_property(&layer->plane, 0, 0,
diff --git a/drivers/gpu/drm/sun4i/sun4i_lvds.c b/drivers/gpu/drm/sun4i/sun4i_lvds.c
index e7eb0d1e17be..147b97ed1a09 100644
--- a/drivers/gpu/drm/sun4i/sun4i_lvds.c
+++ b/drivers/gpu/drm/sun4i/sun4i_lvds.c
@@ -8,9 +8,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sun4i_crtc.h"
 #include "sun4i_tcon.h"
diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c
index f4a22689eb54..cae19e7bbeaa 100644
--- a/drivers/gpu/drm/sun4i/sun4i_rgb.c
+++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c
@@ -14,9 +14,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sun4i_crtc.h"
 #include "sun4i_tcon.h"
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index 0420f5c978b9..7136fc91c603 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -14,11 +14,11 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_connector.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_modes.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include <uapi/drm/drm_mode.h>
 
@@ -761,6 +761,7 @@ static int sun4i_tcon_init_clocks(struct device *dev,
 			return PTR_ERR(tcon->sclk0);
 		}
 	}
+	clk_prepare_enable(tcon->sclk0);
 
 	if (tcon->quirks->has_channel_1) {
 		tcon->sclk1 = devm_clk_get(dev, "tcon-ch1");
@@ -775,6 +776,7 @@ static int sun4i_tcon_init_clocks(struct device *dev,
 
 static void sun4i_tcon_free_clocks(struct sun4i_tcon *tcon)
 {
+	clk_disable_unprepare(tcon->sclk0);
 	clk_disable_unprepare(tcon->clk);
 }
 
@@ -1494,6 +1496,7 @@ const struct of_device_id sun4i_tcon_of_table[] = {
 	{ .compatible = "allwinner,sun6i-a31-tcon", .data = &sun6i_a31_quirks },
 	{ .compatible = "allwinner,sun6i-a31s-tcon", .data = &sun6i_a31s_quirks },
 	{ .compatible = "allwinner,sun7i-a20-tcon", .data = &sun7i_a20_quirks },
+	{ .compatible = "allwinner,sun8i-a23-tcon", .data = &sun8i_a33_quirks },
 	{ .compatible = "allwinner,sun8i-a33-tcon", .data = &sun8i_a33_quirks },
 	{ .compatible = "allwinner,sun8i-a83t-tcon-lcd", .data = &sun8i_a83t_lcd_quirks },
 	{ .compatible = "allwinner,sun8i-a83t-tcon-tv", .data = &sun8i_a83t_tv_quirks },
diff --git a/drivers/gpu/drm/sun4i/sun4i_tv.c b/drivers/gpu/drm/sun4i/sun4i_tv.c
index 1a838d208211..e8700a362064 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tv.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tv.c
@@ -18,9 +18,9 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sun4i_crtc.h"
 #include "sun4i_drv.h"
diff --git a/drivers/gpu/drm/sun4i/sun6i_drc.c b/drivers/gpu/drm/sun4i/sun6i_drc.c
index 88eb268fdf73..442094a4af7a 100644
--- a/drivers/gpu/drm/sun4i/sun6i_drc.c
+++ b/drivers/gpu/drm/sun4i/sun6i_drc.c
@@ -101,6 +101,7 @@ static int sun6i_drc_remove(struct platform_device *pdev)
 static const struct of_device_id sun6i_drc_of_table[] = {
 	{ .compatible = "allwinner,sun6i-a31-drc" },
 	{ .compatible = "allwinner,sun6i-a31s-drc" },
+	{ .compatible = "allwinner,sun8i-a23-drc" },
 	{ .compatible = "allwinner,sun8i-a33-drc" },
 	{ .compatible = "allwinner,sun9i-a80-drc" },
 	{ }
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dphy.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dphy.c
deleted file mode 100644
index e4d19431fa0e..000000000000
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dphy.c
+++ /dev/null
@@ -1,292 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0+
-/*
- * Copyright (c) 2016 Allwinnertech Co., Ltd.
- * Copyright (C) 2017-2018 Bootlin
- *
- * Maxime Ripard <maxime.ripard@free-electrons.com>
- */
-
-#include <linux/bitops.h>
-#include <linux/clk.h>
-#include <linux/of_address.h>
-#include <linux/regmap.h>
-#include <linux/reset.h>
-
-#include "sun6i_mipi_dsi.h"
-
-#define SUN6I_DPHY_GCTL_REG		0x00
-#define SUN6I_DPHY_GCTL_LANE_NUM(n)		((((n) - 1) & 3) << 4)
-#define SUN6I_DPHY_GCTL_EN			BIT(0)
-
-#define SUN6I_DPHY_TX_CTL_REG		0x04
-#define SUN6I_DPHY_TX_CTL_HS_TX_CLK_CONT	BIT(28)
-
-#define SUN6I_DPHY_TX_TIME0_REG		0x10
-#define SUN6I_DPHY_TX_TIME0_HS_TRAIL(n)		(((n) & 0xff) << 24)
-#define SUN6I_DPHY_TX_TIME0_HS_PREPARE(n)	(((n) & 0xff) << 16)
-#define SUN6I_DPHY_TX_TIME0_LP_CLK_DIV(n)	((n) & 0xff)
-
-#define SUN6I_DPHY_TX_TIME1_REG		0x14
-#define SUN6I_DPHY_TX_TIME1_CLK_POST(n)		(((n) & 0xff) << 24)
-#define SUN6I_DPHY_TX_TIME1_CLK_PRE(n)		(((n) & 0xff) << 16)
-#define SUN6I_DPHY_TX_TIME1_CLK_ZERO(n)		(((n) & 0xff) << 8)
-#define SUN6I_DPHY_TX_TIME1_CLK_PREPARE(n)	((n) & 0xff)
-
-#define SUN6I_DPHY_TX_TIME2_REG		0x18
-#define SUN6I_DPHY_TX_TIME2_CLK_TRAIL(n)	((n) & 0xff)
-
-#define SUN6I_DPHY_TX_TIME3_REG		0x1c
-
-#define SUN6I_DPHY_TX_TIME4_REG		0x20
-#define SUN6I_DPHY_TX_TIME4_HS_TX_ANA1(n)	(((n) & 0xff) << 8)
-#define SUN6I_DPHY_TX_TIME4_HS_TX_ANA0(n)	((n) & 0xff)
-
-#define SUN6I_DPHY_ANA0_REG		0x4c
-#define SUN6I_DPHY_ANA0_REG_PWS			BIT(31)
-#define SUN6I_DPHY_ANA0_REG_DMPC		BIT(28)
-#define SUN6I_DPHY_ANA0_REG_DMPD(n)		(((n) & 0xf) << 24)
-#define SUN6I_DPHY_ANA0_REG_SLV(n)		(((n) & 7) << 12)
-#define SUN6I_DPHY_ANA0_REG_DEN(n)		(((n) & 0xf) << 8)
-
-#define SUN6I_DPHY_ANA1_REG		0x50
-#define SUN6I_DPHY_ANA1_REG_VTTMODE		BIT(31)
-#define SUN6I_DPHY_ANA1_REG_CSMPS(n)		(((n) & 3) << 28)
-#define SUN6I_DPHY_ANA1_REG_SVTT(n)		(((n) & 0xf) << 24)
-
-#define SUN6I_DPHY_ANA2_REG		0x54
-#define SUN6I_DPHY_ANA2_EN_P2S_CPU(n)		(((n) & 0xf) << 24)
-#define SUN6I_DPHY_ANA2_EN_P2S_CPU_MASK		GENMASK(27, 24)
-#define SUN6I_DPHY_ANA2_EN_CK_CPU		BIT(4)
-#define SUN6I_DPHY_ANA2_REG_ENIB		BIT(1)
-
-#define SUN6I_DPHY_ANA3_REG		0x58
-#define SUN6I_DPHY_ANA3_EN_VTTD(n)		(((n) & 0xf) << 28)
-#define SUN6I_DPHY_ANA3_EN_VTTD_MASK		GENMASK(31, 28)
-#define SUN6I_DPHY_ANA3_EN_VTTC			BIT(27)
-#define SUN6I_DPHY_ANA3_EN_DIV			BIT(26)
-#define SUN6I_DPHY_ANA3_EN_LDOC			BIT(25)
-#define SUN6I_DPHY_ANA3_EN_LDOD			BIT(24)
-#define SUN6I_DPHY_ANA3_EN_LDOR			BIT(18)
-
-#define SUN6I_DPHY_ANA4_REG		0x5c
-#define SUN6I_DPHY_ANA4_REG_DMPLVC		BIT(24)
-#define SUN6I_DPHY_ANA4_REG_DMPLVD(n)		(((n) & 0xf) << 20)
-#define SUN6I_DPHY_ANA4_REG_CKDV(n)		(((n) & 0x1f) << 12)
-#define SUN6I_DPHY_ANA4_REG_TMSC(n)		(((n) & 3) << 10)
-#define SUN6I_DPHY_ANA4_REG_TMSD(n)		(((n) & 3) << 8)
-#define SUN6I_DPHY_ANA4_REG_TXDNSC(n)		(((n) & 3) << 6)
-#define SUN6I_DPHY_ANA4_REG_TXDNSD(n)		(((n) & 3) << 4)
-#define SUN6I_DPHY_ANA4_REG_TXPUSC(n)		(((n) & 3) << 2)
-#define SUN6I_DPHY_ANA4_REG_TXPUSD(n)		((n) & 3)
-
-#define SUN6I_DPHY_DBG5_REG		0xf4
-
-int sun6i_dphy_init(struct sun6i_dphy *dphy, unsigned int lanes)
-{
-	reset_control_deassert(dphy->reset);
-	clk_prepare_enable(dphy->mod_clk);
-	clk_set_rate_exclusive(dphy->mod_clk, 150000000);
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_CTL_REG,
-		     SUN6I_DPHY_TX_CTL_HS_TX_CLK_CONT);
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_TIME0_REG,
-		     SUN6I_DPHY_TX_TIME0_LP_CLK_DIV(14) |
-		     SUN6I_DPHY_TX_TIME0_HS_PREPARE(6) |
-		     SUN6I_DPHY_TX_TIME0_HS_TRAIL(10));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_TIME1_REG,
-		     SUN6I_DPHY_TX_TIME1_CLK_PREPARE(7) |
-		     SUN6I_DPHY_TX_TIME1_CLK_ZERO(50) |
-		     SUN6I_DPHY_TX_TIME1_CLK_PRE(3) |
-		     SUN6I_DPHY_TX_TIME1_CLK_POST(10));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_TIME2_REG,
-		     SUN6I_DPHY_TX_TIME2_CLK_TRAIL(30));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_TIME3_REG, 0);
-
-	regmap_write(dphy->regs, SUN6I_DPHY_TX_TIME4_REG,
-		     SUN6I_DPHY_TX_TIME4_HS_TX_ANA0(3) |
-		     SUN6I_DPHY_TX_TIME4_HS_TX_ANA1(3));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_GCTL_REG,
-		     SUN6I_DPHY_GCTL_LANE_NUM(lanes) |
-		     SUN6I_DPHY_GCTL_EN);
-
-	return 0;
-}
-
-int sun6i_dphy_power_on(struct sun6i_dphy *dphy, unsigned int lanes)
-{
-	u8 lanes_mask = GENMASK(lanes - 1, 0);
-
-	regmap_write(dphy->regs, SUN6I_DPHY_ANA0_REG,
-		     SUN6I_DPHY_ANA0_REG_PWS |
-		     SUN6I_DPHY_ANA0_REG_DMPC |
-		     SUN6I_DPHY_ANA0_REG_SLV(7) |
-		     SUN6I_DPHY_ANA0_REG_DMPD(lanes_mask) |
-		     SUN6I_DPHY_ANA0_REG_DEN(lanes_mask));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_ANA1_REG,
-		     SUN6I_DPHY_ANA1_REG_CSMPS(1) |
-		     SUN6I_DPHY_ANA1_REG_SVTT(7));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_ANA4_REG,
-		     SUN6I_DPHY_ANA4_REG_CKDV(1) |
-		     SUN6I_DPHY_ANA4_REG_TMSC(1) |
-		     SUN6I_DPHY_ANA4_REG_TMSD(1) |
-		     SUN6I_DPHY_ANA4_REG_TXDNSC(1) |
-		     SUN6I_DPHY_ANA4_REG_TXDNSD(1) |
-		     SUN6I_DPHY_ANA4_REG_TXPUSC(1) |
-		     SUN6I_DPHY_ANA4_REG_TXPUSD(1) |
-		     SUN6I_DPHY_ANA4_REG_DMPLVC |
-		     SUN6I_DPHY_ANA4_REG_DMPLVD(lanes_mask));
-
-	regmap_write(dphy->regs, SUN6I_DPHY_ANA2_REG,
-		     SUN6I_DPHY_ANA2_REG_ENIB);
-	udelay(5);
-
-	regmap_write(dphy->regs, SUN6I_DPHY_ANA3_REG,
-		     SUN6I_DPHY_ANA3_EN_LDOR |
-		     SUN6I_DPHY_ANA3_EN_LDOC |
-		     SUN6I_DPHY_ANA3_EN_LDOD);
-	udelay(1);
-
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA3_REG,
-			   SUN6I_DPHY_ANA3_EN_VTTC |
-			   SUN6I_DPHY_ANA3_EN_VTTD_MASK,
-			   SUN6I_DPHY_ANA3_EN_VTTC |
-			   SUN6I_DPHY_ANA3_EN_VTTD(lanes_mask));
-	udelay(1);
-
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA3_REG,
-			   SUN6I_DPHY_ANA3_EN_DIV,
-			   SUN6I_DPHY_ANA3_EN_DIV);
-	udelay(1);
-
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA2_REG,
-			   SUN6I_DPHY_ANA2_EN_CK_CPU,
-			   SUN6I_DPHY_ANA2_EN_CK_CPU);
-	udelay(1);
-
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA1_REG,
-			   SUN6I_DPHY_ANA1_REG_VTTMODE,
-			   SUN6I_DPHY_ANA1_REG_VTTMODE);
-
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA2_REG,
-			   SUN6I_DPHY_ANA2_EN_P2S_CPU_MASK,
-			   SUN6I_DPHY_ANA2_EN_P2S_CPU(lanes_mask));
-
-	return 0;
-}
-
-int sun6i_dphy_power_off(struct sun6i_dphy *dphy)
-{
-	regmap_update_bits(dphy->regs, SUN6I_DPHY_ANA1_REG,
-			   SUN6I_DPHY_ANA1_REG_VTTMODE, 0);
-
-	return 0;
-}
-
-int sun6i_dphy_exit(struct sun6i_dphy *dphy)
-{
-	clk_rate_exclusive_put(dphy->mod_clk);
-	clk_disable_unprepare(dphy->mod_clk);
-	reset_control_assert(dphy->reset);
-
-	return 0;
-}
-
-static struct regmap_config sun6i_dphy_regmap_config = {
-	.reg_bits	= 32,
-	.val_bits	= 32,
-	.reg_stride	= 4,
-	.max_register	= SUN6I_DPHY_DBG5_REG,
-	.name		= "mipi-dphy",
-};
-
-static const struct of_device_id sun6i_dphy_of_table[] = {
-	{ .compatible = "allwinner,sun6i-a31-mipi-dphy" },
-	{ }
-};
-
-int sun6i_dphy_probe(struct sun6i_dsi *dsi, struct device_node *node)
-{
-	struct sun6i_dphy *dphy;
-	struct resource res;
-	void __iomem *regs;
-	int ret;
-
-	if (!of_match_node(sun6i_dphy_of_table, node)) {
-		dev_err(dsi->dev, "Incompatible D-PHY\n");
-		return -EINVAL;
-	}
-
-	dphy = devm_kzalloc(dsi->dev, sizeof(*dphy), GFP_KERNEL);
-	if (!dphy)
-		return -ENOMEM;
-
-	ret = of_address_to_resource(node, 0, &res);
-	if (ret) {
-		dev_err(dsi->dev, "phy: Couldn't get our resources\n");
-		return ret;
-	}
-
-	regs = devm_ioremap_resource(dsi->dev, &res);
-	if (IS_ERR(regs)) {
-		dev_err(dsi->dev, "Couldn't map the DPHY encoder registers\n");
-		return PTR_ERR(regs);
-	}
-
-	dphy->regs = devm_regmap_init_mmio(dsi->dev, regs,
-					   &sun6i_dphy_regmap_config);
-	if (IS_ERR(dphy->regs)) {
-		dev_err(dsi->dev, "Couldn't create the DPHY encoder regmap\n");
-		return PTR_ERR(dphy->regs);
-	}
-
-	dphy->reset = of_reset_control_get_shared(node, NULL);
-	if (IS_ERR(dphy->reset)) {
-		dev_err(dsi->dev, "Couldn't get our reset line\n");
-		return PTR_ERR(dphy->reset);
-	}
-
-	dphy->bus_clk = of_clk_get_by_name(node, "bus");
-	if (IS_ERR(dphy->bus_clk)) {
-		dev_err(dsi->dev, "Couldn't get the DPHY bus clock\n");
-		ret = PTR_ERR(dphy->bus_clk);
-		goto err_free_reset;
-	}
-	regmap_mmio_attach_clk(dphy->regs, dphy->bus_clk);
-
-	dphy->mod_clk = of_clk_get_by_name(node, "mod");
-	if (IS_ERR(dphy->mod_clk)) {
-		dev_err(dsi->dev, "Couldn't get the DPHY mod clock\n");
-		ret = PTR_ERR(dphy->mod_clk);
-		goto err_free_bus;
-	}
-
-	dsi->dphy = dphy;
-
-	return 0;
-
-err_free_bus:
-	regmap_mmio_detach_clk(dphy->regs);
-	clk_put(dphy->bus_clk);
-err_free_reset:
-	reset_control_put(dphy->reset);
-	return ret;
-}
-
-int sun6i_dphy_remove(struct sun6i_dsi *dsi)
-{
-	struct sun6i_dphy *dphy = dsi->dphy;
-
-	regmap_mmio_detach_clk(dphy->regs);
-	clk_put(dphy->mod_clk);
-	clk_put(dphy->bus_clk);
-	reset_control_put(dphy->reset);
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
index e3b34a345546..318994cd1b85 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
@@ -16,12 +16,13 @@
 #include <linux/slab.h>
 
 #include <linux/phy/phy.h>
+#include <linux/phy/phy-mipi-dphy.h>
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "sun4i_drv.h"
 #include "sun6i_mipi_dsi.h"
@@ -616,6 +617,8 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder)
 	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
 	struct sun6i_dsi *dsi = encoder_to_sun6i_dsi(encoder);
 	struct mipi_dsi_device *device = dsi->device;
+	union phy_configure_opts opts = { 0 };
+	struct phy_configure_opts_mipi_dphy *cfg = &opts.mipi_dphy;
 	u16 delay;
 
 	DRM_DEBUG_DRIVER("Enabling DSI output\n");
@@ -634,8 +637,15 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder)
 	sun6i_dsi_setup_format(dsi, mode);
 	sun6i_dsi_setup_timings(dsi, mode);
 
-	sun6i_dphy_init(dsi->dphy, device->lanes);
-	sun6i_dphy_power_on(dsi->dphy, device->lanes);
+	phy_init(dsi->dphy);
+
+	phy_mipi_dphy_get_default_config(mode->clock * 1000,
+					 mipi_dsi_pixel_format_to_bpp(device->format),
+					 device->lanes, cfg);
+
+	phy_set_mode(dsi->dphy, PHY_MODE_MIPI_DPHY);
+	phy_configure(dsi->dphy, &opts);
+	phy_power_on(dsi->dphy);
 
 	if (!IS_ERR(dsi->panel))
 		drm_panel_prepare(dsi->panel);
@@ -673,8 +683,8 @@ static void sun6i_dsi_encoder_disable(struct drm_encoder *encoder)
 		drm_panel_unprepare(dsi->panel);
 	}
 
-	sun6i_dphy_power_off(dsi->dphy);
-	sun6i_dphy_exit(dsi->dphy);
+	phy_power_off(dsi->dphy);
+	phy_exit(dsi->dphy);
 
 	pm_runtime_put(dsi->dev);
 }
@@ -967,7 +977,6 @@ static const struct component_ops sun6i_dsi_ops = {
 static int sun6i_dsi_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	struct device_node *dphy_node;
 	struct sun6i_dsi *dsi;
 	struct resource *res;
 	void __iomem *base;
@@ -1013,11 +1022,10 @@ static int sun6i_dsi_probe(struct platform_device *pdev)
 	 */
 	clk_set_rate_exclusive(dsi->mod_clk, 297000000);
 
-	dphy_node = of_parse_phandle(dev->of_node, "phys", 0);
-	ret = sun6i_dphy_probe(dsi, dphy_node);
-	of_node_put(dphy_node);
-	if (ret) {
+	dsi->dphy = devm_phy_get(dev, "dphy");
+	if (IS_ERR(dsi->dphy)) {
 		dev_err(dev, "Couldn't get the MIPI D-PHY\n");
+		ret = PTR_ERR(dsi->dphy);
 		goto err_unprotect_clk;
 	}
 
@@ -1026,7 +1034,7 @@ static int sun6i_dsi_probe(struct platform_device *pdev)
 	ret = mipi_dsi_host_register(&dsi->host);
 	if (ret) {
 		dev_err(dev, "Couldn't register MIPI-DSI host\n");
-		goto err_remove_phy;
+		goto err_pm_disable;
 	}
 
 	ret = component_add(&pdev->dev, &sun6i_dsi_ops);
@@ -1039,9 +1047,8 @@ static int sun6i_dsi_probe(struct platform_device *pdev)
 
 err_remove_dsi_host:
 	mipi_dsi_host_unregister(&dsi->host);
-err_remove_phy:
+err_pm_disable:
 	pm_runtime_disable(dev);
-	sun6i_dphy_remove(dsi);
 err_unprotect_clk:
 	clk_rate_exclusive_put(dsi->mod_clk);
 	return ret;
@@ -1055,7 +1062,6 @@ static int sun6i_dsi_remove(struct platform_device *pdev)
 	component_del(&pdev->dev, &sun6i_dsi_ops);
 	mipi_dsi_host_unregister(&dsi->host);
 	pm_runtime_disable(dev);
-	sun6i_dphy_remove(dsi);
 	clk_rate_exclusive_put(dsi->mod_clk);
 
 	return 0;
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
index dbbc5b3ecbda..a07090579f84 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
@@ -13,13 +13,6 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_mipi_dsi.h>
 
-struct sun6i_dphy {
-	struct clk		*bus_clk;
-	struct clk		*mod_clk;
-	struct regmap		*regs;
-	struct reset_control	*reset;
-};
-
 struct sun6i_dsi {
 	struct drm_connector	connector;
 	struct drm_encoder	encoder;
@@ -29,7 +22,7 @@ struct sun6i_dsi {
 	struct clk		*mod_clk;
 	struct regmap		*regs;
 	struct reset_control	*reset;
-	struct sun6i_dphy	*dphy;
+	struct phy		*dphy;
 
 	struct device		*dev;
 	struct sun4i_drv	*drv;
@@ -52,12 +45,4 @@ static inline struct sun6i_dsi *encoder_to_sun6i_dsi(const struct drm_encoder *e
 	return container_of(encoder, struct sun6i_dsi, encoder);
 };
 
-int sun6i_dphy_probe(struct sun6i_dsi *dsi, struct device_node *node);
-int sun6i_dphy_remove(struct sun6i_dsi *dsi);
-
-int sun6i_dphy_init(struct sun6i_dphy *dphy, unsigned int lanes);
-int sun6i_dphy_power_on(struct sun6i_dphy *dphy, unsigned int lanes);
-int sun6i_dphy_power_off(struct sun6i_dphy *dphy);
-int sun6i_dphy_exit(struct sun6i_dphy *dphy);
-
 #endif /* _SUN6I_MIPI_DSI_H_ */
diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c
index 44a9ba7d8433..30a2eff55687 100644
--- a/drivers/gpu/drm/sun4i/sun8i_mixer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c
@@ -14,10 +14,10 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/component.h>
 #include <linux/dma-mapping.h>
diff --git a/drivers/gpu/drm/sun4i/sun8i_ui_layer.c b/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
index 18534263a05d..a342ec8b131e 100644
--- a/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_ui_layer.c
@@ -16,11 +16,11 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include "sun8i_ui_layer.h"
diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
index 87be898f9b7a..8a0616238467 100644
--- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
@@ -10,11 +10,11 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include "sun8i_vi_layer.h"
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
index 2e0d6213f6bc..33c463e8d49f 100644
--- a/drivers/gpu/drm/tegra/Makefile
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -10,6 +10,7 @@ tegra-drm-y := \
 	dc.o \
 	output.o \
 	rgb.o \
+	hda.o \
 	hdmi.o \
 	mipi-phy.o \
 	dsi.o \
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 4b70ce664c41..0c5f1e6a0446 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -92,10 +92,6 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 		return -ENOMEM;
 
 	if (iommu_present(&platform_bus_type)) {
-		u64 carveout_start, carveout_end, gem_start, gem_end;
-		struct iommu_domain_geometry *geometry;
-		unsigned long order;
-
 		tegra->domain = iommu_domain_alloc(&platform_bus_type);
 		if (!tegra->domain) {
 			err = -ENOMEM;
@@ -105,27 +101,6 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 		err = iova_cache_get();
 		if (err < 0)
 			goto domain;
-
-		geometry = &tegra->domain->geometry;
-		gem_start = geometry->aperture_start;
-		gem_end = geometry->aperture_end - CARVEOUT_SZ;
-		carveout_start = gem_end + 1;
-		carveout_end = geometry->aperture_end;
-
-		order = __ffs(tegra->domain->pgsize_bitmap);
-		init_iova_domain(&tegra->carveout.domain, 1UL << order,
-				 carveout_start >> order);
-
-		tegra->carveout.shift = iova_shift(&tegra->carveout.domain);
-		tegra->carveout.limit = carveout_end >> tegra->carveout.shift;
-
-		drm_mm_init(&tegra->mm, gem_start, gem_end - gem_start + 1);
-		mutex_init(&tegra->mm_lock);
-
-		DRM_DEBUG("IOMMU apertures:\n");
-		DRM_DEBUG("  GEM: %#llx-%#llx\n", gem_start, gem_end);
-		DRM_DEBUG("  Carveout: %#llx-%#llx\n", carveout_start,
-			  carveout_end);
 	}
 
 	mutex_init(&tegra->clients_lock);
@@ -159,6 +134,36 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 	if (err < 0)
 		goto fbdev;
 
+	if (tegra->domain) {
+		u64 carveout_start, carveout_end, gem_start, gem_end;
+		u64 dma_mask = dma_get_mask(&device->dev);
+		dma_addr_t start, end;
+		unsigned long order;
+
+		start = tegra->domain->geometry.aperture_start & dma_mask;
+		end = tegra->domain->geometry.aperture_end & dma_mask;
+
+		gem_start = start;
+		gem_end = end - CARVEOUT_SZ;
+		carveout_start = gem_end + 1;
+		carveout_end = end;
+
+		order = __ffs(tegra->domain->pgsize_bitmap);
+		init_iova_domain(&tegra->carveout.domain, 1UL << order,
+				 carveout_start >> order);
+
+		tegra->carveout.shift = iova_shift(&tegra->carveout.domain);
+		tegra->carveout.limit = carveout_end >> tegra->carveout.shift;
+
+		drm_mm_init(&tegra->mm, gem_start, gem_end - gem_start + 1);
+		mutex_init(&tegra->mm_lock);
+
+		DRM_DEBUG("IOMMU apertures:\n");
+		DRM_DEBUG("  GEM: %#llx-%#llx\n", gem_start, gem_end);
+		DRM_DEBUG("  Carveout: %#llx-%#llx\n", carveout_start,
+			  carveout_end);
+	}
+
 	if (tegra->hub) {
 		err = tegra_display_hub_prepare(tegra->hub);
 		if (err < 0)
@@ -1041,6 +1046,7 @@ int tegra_drm_register_client(struct tegra_drm *tegra,
 {
 	mutex_lock(&tegra->clients_lock);
 	list_add_tail(&client->list, &tegra->clients);
+	client->drm = tegra;
 	mutex_unlock(&tegra->clients_lock);
 
 	return 0;
@@ -1051,6 +1057,7 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
 {
 	mutex_lock(&tegra->clients_lock);
 	list_del_init(&client->list);
+	client->drm = NULL;
 	mutex_unlock(&tegra->clients_lock);
 
 	return 0;
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 1012335bb489..70154c253d45 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -17,11 +17,11 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_probe_helper.h>
 
 #include "gem.h"
 #include "hub.h"
@@ -88,6 +88,7 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 struct tegra_drm_client {
 	struct host1x_client base;
 	struct list_head list;
+	struct tegra_drm *drm;
 
 	unsigned int version;
 	const struct tegra_drm_client_ops *ops;
@@ -124,7 +125,7 @@ struct tegra_output {
 	struct drm_panel *panel;
 	struct i2c_adapter *ddc;
 	const struct edid *edid;
-	struct cec_notifier *notifier;
+	struct cec_notifier *cec;
 	unsigned int hpd_irq;
 	int hpd_gpio;
 	enum of_gpio_flags hpd_gpio_flags;
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index b947e82bbeb1..0a4ce05e00ab 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -15,6 +15,7 @@
 #include "drm.h"
 #include "gem.h"
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_modeset_helper.h>
 
 #ifdef CONFIG_DRM_FBDEV_EMULATION
 static inline struct tegra_fbdev *to_tegra_fbdev(struct drm_fb_helper *helper)
@@ -255,7 +256,6 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 	helper->fbdev = info;
 
 	info->par = helper;
-	info->flags = FBINFO_FLAG_DEFAULT;
 	info->fbops = &tegra_fb_ops;
 
 	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
diff --git a/drivers/gpu/drm/tegra/hda.c b/drivers/gpu/drm/tegra/hda.c
new file mode 100644
index 000000000000..94245a18a043
--- /dev/null
+++ b/drivers/gpu/drm/tegra/hda.c
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright (C) 2019 NVIDIA Corporation
+ */
+
+#include <linux/bug.h>
+
+#include <sound/hda_verbs.h>
+
+#include "hda.h"
+
+void tegra_hda_parse_format(unsigned int format, struct tegra_hda_format *fmt)
+{
+	unsigned int mul, div, bits, channels;
+
+	if (format & AC_FMT_TYPE_NON_PCM)
+		fmt->pcm = false;
+	else
+		fmt->pcm = true;
+
+	if (format & AC_FMT_BASE_44K)
+		fmt->sample_rate = 44100;
+	else
+		fmt->sample_rate = 48000;
+
+	mul = (format & AC_FMT_MULT_MASK) >> AC_FMT_MULT_SHIFT;
+	div = (format & AC_FMT_DIV_MASK) >> AC_FMT_DIV_SHIFT;
+
+	fmt->sample_rate *= (mul + 1) / (div + 1);
+
+	switch (format & AC_FMT_BITS_MASK) {
+	case AC_FMT_BITS_8:
+		fmt->bits = 8;
+		break;
+
+	case AC_FMT_BITS_16:
+		fmt->bits = 16;
+		break;
+
+	case AC_FMT_BITS_20:
+		fmt->bits = 20;
+		break;
+
+	case AC_FMT_BITS_24:
+		fmt->bits = 24;
+		break;
+
+	case AC_FMT_BITS_32:
+		fmt->bits = 32;
+		break;
+
+	default:
+		bits = (format & AC_FMT_BITS_MASK) >> AC_FMT_BITS_SHIFT;
+		WARN(1, "invalid number of bits: %#x\n", bits);
+		fmt->bits = 8;
+		break;
+	}
+
+	channels = (format & AC_FMT_CHAN_MASK) >> AC_FMT_CHAN_SHIFT;
+
+	/* channels are encoded as n - 1 */
+	fmt->channels = channels + 1;
+}
diff --git a/drivers/gpu/drm/tegra/hda.h b/drivers/gpu/drm/tegra/hda.h
new file mode 100644
index 000000000000..77269955a4f2
--- /dev/null
+++ b/drivers/gpu/drm/tegra/hda.h
@@ -0,0 +1,20 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright (C) 2019 NVIDIA Corporation
+ */
+
+#ifndef DRM_TEGRA_HDA_H
+#define DRM_TEGRA_HDA_H 1
+
+#include <linux/types.h>
+
+struct tegra_hda_format {
+	unsigned int sample_rate;
+	unsigned int channels;
+	unsigned int bits;
+	bool pcm;
+};
+
+void tegra_hda_parse_format(unsigned int format, struct tegra_hda_format *fmt);
+
+#endif
diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index a7566c67bfb0..47c55974756d 100644
--- a/drivers/gpu/drm/tegra/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -11,6 +11,7 @@
 #include <linux/debugfs.h>
 #include <linux/gpio.h>
 #include <linux/hdmi.h>
+#include <linux/math64.h>
 #include <linux/of_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/regulator/consumer.h>
@@ -18,12 +19,9 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
-
-#include <sound/hda_verbs.h>
-
-#include <media/cec-notifier.h>
+#include <drm/drm_probe_helper.h>
 
+#include "hda.h"
 #include "hdmi.h"
 #include "drm.h"
 #include "dc.h"
@@ -71,8 +69,7 @@ struct tegra_hdmi {
 	const struct tegra_hdmi_config *config;
 
 	unsigned int audio_source;
-	unsigned int audio_sample_rate;
-	unsigned int audio_channels;
+	struct tegra_hda_format format;
 
 	unsigned int pixel_clock;
 	bool stereo;
@@ -119,68 +116,11 @@ static inline void tegra_hdmi_writel(struct tegra_hdmi *hdmi, u32 value,
 }
 
 struct tegra_hdmi_audio_config {
-	unsigned int pclk;
 	unsigned int n;
 	unsigned int cts;
 	unsigned int aval;
 };
 
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_32k[] = {
-	{  25200000, 4096,  25200, 24000 },
-	{  27000000, 4096,  27000, 24000 },
-	{  74250000, 4096,  74250, 24000 },
-	{ 148500000, 4096, 148500, 24000 },
-	{         0,    0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_44_1k[] = {
-	{  25200000, 5880,  26250, 25000 },
-	{  27000000, 5880,  28125, 25000 },
-	{  74250000, 4704,  61875, 20000 },
-	{ 148500000, 4704, 123750, 20000 },
-	{         0,    0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_48k[] = {
-	{  25200000, 6144,  25200, 24000 },
-	{  27000000, 6144,  27000, 24000 },
-	{  74250000, 6144,  74250, 24000 },
-	{ 148500000, 6144, 148500, 24000 },
-	{         0,    0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_88_2k[] = {
-	{  25200000, 11760,  26250, 25000 },
-	{  27000000, 11760,  28125, 25000 },
-	{  74250000,  9408,  61875, 20000 },
-	{ 148500000,  9408, 123750, 20000 },
-	{         0,     0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_96k[] = {
-	{  25200000, 12288,  25200, 24000 },
-	{  27000000, 12288,  27000, 24000 },
-	{  74250000, 12288,  74250, 24000 },
-	{ 148500000, 12288, 148500, 24000 },
-	{         0,     0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_176_4k[] = {
-	{  25200000, 23520,  26250, 25000 },
-	{  27000000, 23520,  28125, 25000 },
-	{  74250000, 18816,  61875, 20000 },
-	{ 148500000, 18816, 123750, 20000 },
-	{         0,     0,      0,     0 },
-};
-
-static const struct tegra_hdmi_audio_config tegra_hdmi_audio_192k[] = {
-	{  25200000, 24576,  25200, 24000 },
-	{  27000000, 24576,  27000, 24000 },
-	{  74250000, 24576,  74250, 24000 },
-	{ 148500000, 24576, 148500, 24000 },
-	{         0,     0,      0,     0 },
-};
-
 static const struct tmds_config tegra20_tmds_config[] = {
 	{ /* slow pixel clock modes */
 		.pclk = 27000000,
@@ -418,52 +358,53 @@ static const struct tmds_config tegra124_tmds_config[] = {
 	},
 };
 
-static const struct tegra_hdmi_audio_config *
-tegra_hdmi_get_audio_config(unsigned int sample_rate, unsigned int pclk)
+static int
+tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pix_clock,
+			    struct tegra_hdmi_audio_config *config)
 {
-	const struct tegra_hdmi_audio_config *table;
-
-	switch (sample_rate) {
-	case 32000:
-		table = tegra_hdmi_audio_32k;
-		break;
-
-	case 44100:
-		table = tegra_hdmi_audio_44_1k;
-		break;
-
-	case 48000:
-		table = tegra_hdmi_audio_48k;
-		break;
-
-	case 88200:
-		table = tegra_hdmi_audio_88_2k;
-		break;
-
-	case 96000:
-		table = tegra_hdmi_audio_96k;
-		break;
-
-	case 176400:
-		table = tegra_hdmi_audio_176_4k;
-		break;
-
-	case 192000:
-		table = tegra_hdmi_audio_192k;
-		break;
-
-	default:
-		return NULL;
-	}
-
-	while (table->pclk) {
-		if (table->pclk == pclk)
-			return table;
-
-		table++;
+	const unsigned int afreq = 128 * audio_freq;
+	const unsigned int min_n = afreq / 1500;
+	const unsigned int max_n = afreq / 300;
+	const unsigned int ideal_n = afreq / 1000;
+	int64_t min_err = (uint64_t)-1 >> 1;
+	unsigned int min_delta = -1;
+	int n;
+
+	memset(config, 0, sizeof(*config));
+	config->n = -1;
+
+	for (n = min_n; n <= max_n; n++) {
+		uint64_t cts_f, aval_f;
+		unsigned int delta;
+		int64_t cts, err;
+
+		/* compute aval in 48.16 fixed point */
+		aval_f = ((int64_t)24000000 << 16) * n;
+		do_div(aval_f, afreq);
+		/* It should round without any rest */
+		if (aval_f & 0xFFFF)
+			continue;
+
+		/* Compute cts in 48.16 fixed point */
+		cts_f = ((int64_t)pix_clock << 16) * n;
+		do_div(cts_f, afreq);
+		/* Round it to the nearest integer */
+		cts = (cts_f & ~0xFFFF) + ((cts_f & BIT(15)) << 1);
+
+		delta = abs(n - ideal_n);
+
+		/* Compute the absolute error */
+		err = abs((int64_t)cts_f - cts);
+		if (err < min_err || (err == min_err && delta < min_delta)) {
+			config->n = n;
+			config->cts = cts >> 16;
+			config->aval = aval_f >> 16;
+			min_delta = delta;
+			min_err = err;
+		}
 	}
 
-	return NULL;
+	return config->n != -1 ? 0 : -EINVAL;
 }
 
 static void tegra_hdmi_setup_audio_fs_tables(struct tegra_hdmi *hdmi)
@@ -510,7 +451,7 @@ static void tegra_hdmi_write_aval(struct tegra_hdmi *hdmi, u32 value)
 	unsigned int i;
 
 	for (i = 0; i < ARRAY_SIZE(regs); i++) {
-		if (regs[i].sample_rate == hdmi->audio_sample_rate) {
+		if (regs[i].sample_rate == hdmi->format.sample_rate) {
 			tegra_hdmi_writel(hdmi, value, regs[i].offset);
 			break;
 		}
@@ -519,8 +460,9 @@ static void tegra_hdmi_write_aval(struct tegra_hdmi *hdmi, u32 value)
 
 static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi)
 {
-	const struct tegra_hdmi_audio_config *config;
+	struct tegra_hdmi_audio_config config;
 	u32 source, value;
+	int err;
 
 	switch (hdmi->audio_source) {
 	case HDA:
@@ -564,7 +506,7 @@ static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi)
 		 * play back system startup sounds early. It is possibly not
 		 * needed on Linux at all.
 		 */
-		if (hdmi->audio_channels == 2)
+		if (hdmi->format.channels == 2)
 			value = SOR_AUDIO_CNTRL0_INJECT_NULLSMPL;
 		else
 			value = 0;
@@ -595,25 +537,28 @@ static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi)
 		tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_AUDIO_SPARE0);
 	}
 
-	config = tegra_hdmi_get_audio_config(hdmi->audio_sample_rate,
-					     hdmi->pixel_clock);
-	if (!config) {
+	err = tegra_hdmi_get_audio_config(hdmi->format.sample_rate,
+					  hdmi->pixel_clock, &config);
+	if (err < 0) {
 		dev_err(hdmi->dev,
 			"cannot set audio to %u Hz at %u Hz pixel clock\n",
-			hdmi->audio_sample_rate, hdmi->pixel_clock);
-		return -EINVAL;
+			hdmi->format.sample_rate, hdmi->pixel_clock);
+		return err;
 	}
 
+	dev_dbg(hdmi->dev, "audio: pixclk=%u, n=%u, cts=%u, aval=%u\n",
+		hdmi->pixel_clock, config.n, config.cts, config.aval);
+
 	tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_HDMI_ACR_CTRL);
 
 	value = AUDIO_N_RESETF | AUDIO_N_GENERATE_ALTERNATE |
-		AUDIO_N_VALUE(config->n - 1);
+		AUDIO_N_VALUE(config.n - 1);
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_N);
 
-	tegra_hdmi_writel(hdmi, ACR_SUBPACK_N(config->n) | ACR_ENABLE,
+	tegra_hdmi_writel(hdmi, ACR_SUBPACK_N(config.n) | ACR_ENABLE,
 			  HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_HIGH);
 
-	tegra_hdmi_writel(hdmi, ACR_SUBPACK_CTS(config->cts),
+	tegra_hdmi_writel(hdmi, ACR_SUBPACK_CTS(config.cts),
 			  HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_LOW);
 
 	value = SPARE_HW_CTS | SPARE_FORCE_SW_CTS | SPARE_CTS_RESET_VAL(1);
@@ -624,7 +569,7 @@ static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi)
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_N);
 
 	if (hdmi->config->has_hda)
-		tegra_hdmi_write_aval(hdmi, config->aval);
+		tegra_hdmi_write_aval(hdmi, config.aval);
 
 	tegra_hdmi_setup_audio_fs_tables(hdmi);
 
@@ -788,7 +733,7 @@ static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 		return;
 	}
 
-	frame.channels = hdmi->audio_channels;
+	frame.channels = hdmi->format.channels;
 
 	err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer));
 	if (err < 0) {
@@ -1590,24 +1535,6 @@ static const struct of_device_id tegra_hdmi_of_match[] = {
 };
 MODULE_DEVICE_TABLE(of, tegra_hdmi_of_match);
 
-static void hda_format_parse(unsigned int format, unsigned int *rate,
-			     unsigned int *channels)
-{
-	unsigned int mul, div;
-
-	if (format & AC_FMT_BASE_44K)
-		*rate = 44100;
-	else
-		*rate = 48000;
-
-	mul = (format & AC_FMT_MULT_MASK) >> AC_FMT_MULT_SHIFT;
-	div = (format & AC_FMT_DIV_MASK) >> AC_FMT_DIV_SHIFT;
-
-	*rate = *rate * (mul + 1) / (div + 1);
-
-	*channels = (format & AC_FMT_CHAN_MASK) >> AC_FMT_CHAN_SHIFT;
-}
-
 static irqreturn_t tegra_hdmi_irq(int irq, void *data)
 {
 	struct tegra_hdmi *hdmi = data;
@@ -1624,14 +1551,9 @@ static irqreturn_t tegra_hdmi_irq(int irq, void *data)
 		value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_AUDIO_HDA_CODEC_SCRATCH0);
 
 		if (value & SOR_AUDIO_HDA_CODEC_SCRATCH0_VALID) {
-			unsigned int sample_rate, channels;
-
 			format = value & SOR_AUDIO_HDA_CODEC_SCRATCH0_FMT_MASK;
 
-			hda_format_parse(format, &sample_rate, &channels);
-
-			hdmi->audio_sample_rate = sample_rate;
-			hdmi->audio_channels = channels;
+			tegra_hda_parse_format(format, &hdmi->format);
 
 			err = tegra_hdmi_setup_audio(hdmi);
 			if (err < 0) {
@@ -1665,8 +1587,6 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 	hdmi->dev = &pdev->dev;
 
 	hdmi->audio_source = AUTO;
-	hdmi->audio_sample_rate = 48000;
-	hdmi->audio_channels = 2;
 	hdmi->stereo = false;
 	hdmi->dvi = false;
 
@@ -1710,10 +1630,6 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 		return PTR_ERR(hdmi->vdd);
 	}
 
-	hdmi->output.notifier = cec_notifier_get(&pdev->dev);
-	if (hdmi->output.notifier == NULL)
-		return -ENOMEM;
-
 	hdmi->output.dev = &pdev->dev;
 
 	err = tegra_output_probe(&hdmi->output);
@@ -1772,9 +1688,6 @@ static int tegra_hdmi_remove(struct platform_device *pdev)
 
 	tegra_output_remove(&hdmi->output);
 
-	if (hdmi->output.notifier)
-		cec_notifier_put(hdmi->output.notifier);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/tegra/hub.c b/drivers/gpu/drm/tegra/hub.c
index 71cc3cf60066..ba9b3cfb8c3d 100644
--- a/drivers/gpu/drm/tegra/hub.c
+++ b/drivers/gpu/drm/tegra/hub.c
@@ -19,7 +19,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "drm.h"
 #include "dc.h"
diff --git a/drivers/gpu/drm/tegra/output.c b/drivers/gpu/drm/tegra/output.c
index c662efc7e413..9c2b9dad55c3 100644
--- a/drivers/gpu/drm/tegra/output.c
+++ b/drivers/gpu/drm/tegra/output.c
@@ -36,7 +36,7 @@ int tegra_output_connector_get_modes(struct drm_connector *connector)
 	else if (output->ddc)
 		edid = drm_get_edid(connector, output->ddc);
 
-	cec_notifier_set_phys_addr_from_edid(output->notifier, edid);
+	cec_notifier_set_phys_addr_from_edid(output->cec, edid);
 	drm_connector_update_edid_property(connector, edid);
 
 	if (edid) {
@@ -73,7 +73,7 @@ tegra_output_connector_detect(struct drm_connector *connector, bool force)
 	}
 
 	if (status != connector_status_connected)
-		cec_notifier_phys_addr_invalidate(output->notifier);
+		cec_notifier_phys_addr_invalidate(output->cec);
 
 	return status;
 }
@@ -174,11 +174,18 @@ int tegra_output_probe(struct tegra_output *output)
 		disable_irq(output->hpd_irq);
 	}
 
+	output->cec = cec_notifier_get(output->dev);
+	if (!output->cec)
+		return -ENOMEM;
+
 	return 0;
 }
 
 void tegra_output_remove(struct tegra_output *output)
 {
+	if (output->cec)
+		cec_notifier_put(output->cec);
+
 	if (gpio_is_valid(output->hpd_gpio)) {
 		free_irq(output->hpd_irq, output);
 		gpio_free(output->hpd_gpio);
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index 44feac2a0359..40057106f5f3 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -19,8 +19,6 @@
 
 #include <soc/tegra/pmc.h>
 
-#include <sound/hda_verbs.h>
-
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
@@ -28,6 +26,7 @@
 
 #include "dc.h"
 #include "drm.h"
+#include "hda.h"
 #include "sor.h"
 #include "trace.h"
 
@@ -411,6 +410,8 @@ struct tegra_sor {
 	struct clk *clk_dp;
 	struct clk *clk;
 
+	u8 xbar_cfg[5];
+
 	struct drm_dp_aux *aux;
 
 	struct drm_info_list *debugfs_files;
@@ -429,10 +430,7 @@ struct tegra_sor {
 	struct delayed_work scdc;
 	bool scdc_enabled;
 
-	struct {
-		unsigned int sample_rate;
-		unsigned int channels;
-	} audio;
+	struct tegra_hda_format format;
 };
 
 struct tegra_sor_state {
@@ -1818,7 +1816,7 @@ static void tegra_sor_edp_enable(struct drm_encoder *encoder)
 
 	/* XXX not in TRM */
 	for (value = 0, i = 0; i < 5; i++)
-		value |= SOR_XBAR_CTRL_LINK0_XSEL(i, sor->soc->xbar_cfg[i]) |
+		value |= SOR_XBAR_CTRL_LINK0_XSEL(i, sor->xbar_cfg[i]) |
 			 SOR_XBAR_CTRL_LINK1_XSEL(i, i);
 
 	tegra_sor_writel(sor, 0x00000000, SOR_XBAR_POL);
@@ -2186,7 +2184,7 @@ static int tegra_sor_hdmi_enable_audio_infoframe(struct tegra_sor *sor)
 		return err;
 	}
 
-	frame.channels = sor->audio.channels;
+	frame.channels = sor->format.channels;
 
 	err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer));
 	if (err < 0) {
@@ -2215,7 +2213,7 @@ static void tegra_sor_hdmi_audio_enable(struct tegra_sor *sor)
 	value |= SOR_AUDIO_CNTRL_SOURCE_SELECT(SOURCE_SELECT_HDA);
 
 	/* inject null samples */
-	if (sor->audio.channels != 2)
+	if (sor->format.channels != 2)
 		value &= ~SOR_AUDIO_CNTRL_INJECT_NULLSMPL;
 	else
 		value |= SOR_AUDIO_CNTRL_INJECT_NULLSMPL;
@@ -2246,7 +2244,7 @@ static void tegra_sor_hdmi_audio_enable(struct tegra_sor *sor)
 	value = SOR_HDMI_AUDIO_N_RESET | SOR_HDMI_AUDIO_N_LOOKUP;
 	tegra_sor_writel(sor, value, SOR_HDMI_AUDIO_N);
 
-	value = (24000 * 4096) / (128 * sor->audio.sample_rate / 1000);
+	value = (24000 * 4096) / (128 * sor->format.sample_rate / 1000);
 	tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0320);
 	tegra_sor_writel(sor, 4096, SOR_AUDIO_NVAL_0320);
 
@@ -2259,15 +2257,15 @@ static void tegra_sor_hdmi_audio_enable(struct tegra_sor *sor)
 	tegra_sor_writel(sor, 20000, SOR_AUDIO_AVAL_1764);
 	tegra_sor_writel(sor, 18816, SOR_AUDIO_NVAL_1764);
 
-	value = (24000 * 6144) / (128 * sor->audio.sample_rate / 1000);
+	value = (24000 * 6144) / (128 * sor->format.sample_rate / 1000);
 	tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0480);
 	tegra_sor_writel(sor, 6144, SOR_AUDIO_NVAL_0480);
 
-	value = (24000 * 12288) / (128 * sor->audio.sample_rate / 1000);
+	value = (24000 * 12288) / (128 * sor->format.sample_rate / 1000);
 	tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0960);
 	tegra_sor_writel(sor, 12288, SOR_AUDIO_NVAL_0960);
 
-	value = (24000 * 24576) / (128 * sor->audio.sample_rate / 1000);
+	value = (24000 * 24576) / (128 * sor->format.sample_rate / 1000);
 	tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_1920);
 	tegra_sor_writel(sor, 24576, SOR_AUDIO_NVAL_1920);
 
@@ -2555,7 +2553,7 @@ static void tegra_sor_hdmi_enable(struct drm_encoder *encoder)
 
 	/* XXX not in TRM */
 	for (value = 0, i = 0; i < 5; i++)
-		value |= SOR_XBAR_CTRL_LINK0_XSEL(i, sor->soc->xbar_cfg[i]) |
+		value |= SOR_XBAR_CTRL_LINK0_XSEL(i, sor->xbar_cfg[i]) |
 			 SOR_XBAR_CTRL_LINK1_XSEL(i, i);
 
 	tegra_sor_writel(sor, 0x00000000, SOR_XBAR_POL);
@@ -3176,6 +3174,8 @@ MODULE_DEVICE_TABLE(of, tegra_sor_of_match);
 static int tegra_sor_parse_dt(struct tegra_sor *sor)
 {
 	struct device_node *np = sor->dev->of_node;
+	u32 xbar_cfg[5];
+	unsigned int i;
 	u32 value;
 	int err;
 
@@ -3193,25 +3193,18 @@ static int tegra_sor_parse_dt(struct tegra_sor *sor)
 		sor->pad = TEGRA_IO_PAD_HDMI_DP0 + sor->index;
 	}
 
-	return 0;
-}
-
-static void tegra_hda_parse_format(unsigned int format, unsigned int *rate,
-				   unsigned int *channels)
-{
-	unsigned int mul, div;
-
-	if (format & AC_FMT_BASE_44K)
-		*rate = 44100;
-	else
-		*rate = 48000;
-
-	mul = (format & AC_FMT_MULT_MASK) >> AC_FMT_MULT_SHIFT;
-	div = (format & AC_FMT_DIV_MASK) >> AC_FMT_DIV_SHIFT;
-
-	*rate = *rate * (mul + 1) / (div + 1);
+	err = of_property_read_u32_array(np, "nvidia,xbar-cfg", xbar_cfg, 5);
+	if (err < 0) {
+		/* fall back to default per-SoC XBAR configuration */
+		for (i = 0; i < 5; i++)
+			sor->xbar_cfg[i] = sor->soc->xbar_cfg[i];
+	} else {
+		/* copy cells to SOR XBAR configuration */
+		for (i = 0; i < 5; i++)
+			sor->xbar_cfg[i] = xbar_cfg[i];
+	}
 
-	*channels = (format & AC_FMT_CHAN_MASK) >> AC_FMT_CHAN_SHIFT;
+	return 0;
 }
 
 static irqreturn_t tegra_sor_irq(int irq, void *data)
@@ -3226,14 +3219,11 @@ static irqreturn_t tegra_sor_irq(int irq, void *data)
 		value = tegra_sor_readl(sor, SOR_AUDIO_HDA_CODEC_SCRATCH0);
 
 		if (value & SOR_AUDIO_HDA_CODEC_SCRATCH0_VALID) {
-			unsigned int format, sample_rate, channels;
+			unsigned int format;
 
 			format = value & SOR_AUDIO_HDA_CODEC_SCRATCH0_FMT_MASK;
 
-			tegra_hda_parse_format(format, &sample_rate, &channels);
-
-			sor->audio.sample_rate = sample_rate;
-			sor->audio.channels = channels;
+			tegra_hda_parse_format(format, &sor->format);
 
 			tegra_sor_hdmi_audio_enable(sor);
 		} else {
diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c
index d47983deb1cf..39bfed9623de 100644
--- a/drivers/gpu/drm/tegra/vic.c
+++ b/drivers/gpu/drm/tegra/vic.c
@@ -26,6 +26,7 @@
 struct vic_config {
 	const char *firmware;
 	unsigned int version;
+	bool supports_sid;
 };
 
 struct vic {
@@ -105,6 +106,22 @@ static int vic_boot(struct vic *vic)
 	if (vic->booted)
 		return 0;
 
+	if (vic->config->supports_sid) {
+		struct iommu_fwspec *spec = dev_iommu_fwspec_get(vic->dev);
+		u32 value;
+
+		value = TRANSCFG_ATT(1, TRANSCFG_SID_FALCON) |
+			TRANSCFG_ATT(0, TRANSCFG_SID_HW);
+		vic_writel(vic, value, VIC_TFBIF_TRANSCFG);
+
+		if (spec && spec->num_ids > 0) {
+			value = spec->ids[0] & 0xffff;
+
+			vic_writel(vic, value, VIC_THI_STREAMID0);
+			vic_writel(vic, value, VIC_THI_STREAMID1);
+		}
+	}
+
 	/* setup clockgating registers */
 	vic_writel(vic, CG_IDLE_CG_DLY_CNT(4) |
 			CG_IDLE_CG_EN |
@@ -181,13 +198,6 @@ static int vic_init(struct host1x_client *client)
 		vic->domain = tegra->domain;
 	}
 
-	if (!vic->falcon.data) {
-		vic->falcon.data = tegra;
-		err = falcon_load_firmware(&vic->falcon);
-		if (err < 0)
-			goto detach;
-	}
-
 	vic->channel = host1x_channel_request(client->dev);
 	if (!vic->channel) {
 		err = -ENOMEM;
@@ -246,6 +256,30 @@ static const struct host1x_client_ops vic_client_ops = {
 	.exit = vic_exit,
 };
 
+static int vic_load_firmware(struct vic *vic)
+{
+	int err;
+
+	if (vic->falcon.data)
+		return 0;
+
+	vic->falcon.data = vic->client.drm;
+
+	err = falcon_read_firmware(&vic->falcon, vic->config->firmware);
+	if (err < 0)
+		goto cleanup;
+
+	err = falcon_load_firmware(&vic->falcon);
+	if (err < 0)
+		goto cleanup;
+
+	return 0;
+
+cleanup:
+	vic->falcon.data = NULL;
+	return err;
+}
+
 static int vic_open_channel(struct tegra_drm_client *client,
 			    struct tegra_drm_context *context)
 {
@@ -256,19 +290,25 @@ static int vic_open_channel(struct tegra_drm_client *client,
 	if (err < 0)
 		return err;
 
+	err = vic_load_firmware(vic);
+	if (err < 0)
+		goto rpm_put;
+
 	err = vic_boot(vic);
-	if (err < 0) {
-		pm_runtime_put(vic->dev);
-		return err;
-	}
+	if (err < 0)
+		goto rpm_put;
 
 	context->channel = host1x_channel_get(vic->channel);
 	if (!context->channel) {
-		pm_runtime_put(vic->dev);
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto rpm_put;
 	}
 
 	return 0;
+
+rpm_put:
+	pm_runtime_put(vic->dev);
+	return err;
 }
 
 static void vic_close_channel(struct tegra_drm_context *context)
@@ -291,6 +331,7 @@ static const struct tegra_drm_client_ops vic_ops = {
 static const struct vic_config vic_t124_config = {
 	.firmware = NVIDIA_TEGRA_124_VIC_FIRMWARE,
 	.version = 0x40,
+	.supports_sid = false,
 };
 
 #define NVIDIA_TEGRA_210_VIC_FIRMWARE "nvidia/tegra210/vic04_ucode.bin"
@@ -298,6 +339,7 @@ static const struct vic_config vic_t124_config = {
 static const struct vic_config vic_t210_config = {
 	.firmware = NVIDIA_TEGRA_210_VIC_FIRMWARE,
 	.version = 0x21,
+	.supports_sid = false,
 };
 
 #define NVIDIA_TEGRA_186_VIC_FIRMWARE "nvidia/tegra186/vic04_ucode.bin"
@@ -305,6 +347,7 @@ static const struct vic_config vic_t210_config = {
 static const struct vic_config vic_t186_config = {
 	.firmware = NVIDIA_TEGRA_186_VIC_FIRMWARE,
 	.version = 0x18,
+	.supports_sid = true,
 };
 
 #define NVIDIA_TEGRA_194_VIC_FIRMWARE "nvidia/tegra194/vic.bin"
@@ -312,6 +355,7 @@ static const struct vic_config vic_t186_config = {
 static const struct vic_config vic_t194_config = {
 	.firmware = NVIDIA_TEGRA_194_VIC_FIRMWARE,
 	.version = 0x19,
+	.supports_sid = true,
 };
 
 static const struct of_device_id vic_match[] = {
@@ -372,10 +416,6 @@ static int vic_probe(struct platform_device *pdev)
 	if (err < 0)
 		return err;
 
-	err = falcon_read_firmware(&vic->falcon, vic->config->firmware);
-	if (err < 0)
-		goto exit_falcon;
-
 	platform_set_drvdata(pdev, vic);
 
 	INIT_LIST_HEAD(&vic->client.base.list);
@@ -393,7 +433,6 @@ static int vic_probe(struct platform_device *pdev)
 	err = host1x_client_register(&vic->client.base);
 	if (err < 0) {
 		dev_err(dev, "failed to register host1x client: %d\n", err);
-		platform_set_drvdata(pdev, NULL);
 		goto exit_falcon;
 	}
 
diff --git a/drivers/gpu/drm/tegra/vic.h b/drivers/gpu/drm/tegra/vic.h
index 21844817a7e1..017584340dd6 100644
--- a/drivers/gpu/drm/tegra/vic.h
+++ b/drivers/gpu/drm/tegra/vic.h
@@ -17,11 +17,20 @@
 
 /* VIC registers */
 
+#define VIC_THI_STREAMID0	0x00000030
+#define VIC_THI_STREAMID1	0x00000034
+
 #define NV_PVIC_MISC_PRI_VIC_CG			0x000016d0
 #define CG_IDLE_CG_DLY_CNT(val)			((val & 0x3f) << 0)
 #define CG_IDLE_CG_EN				(1 << 6)
 #define CG_WAKEUP_DLY_CNT(val)			((val & 0xf) << 16)
 
+#define VIC_TFBIF_TRANSCFG	0x00002044
+#define  TRANSCFG_ATT(i, v)	(((v) & 0x3) << (i * 4))
+#define  TRANSCFG_SID_HW	0
+#define  TRANSCFG_SID_PHY	1
+#define  TRANSCFG_SID_FALCON	2
+
 /* Firmware offsets */
 
 #define VIC_UCODE_FCE_HEADER_OFFSET		(6*4)
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index 337e86a1d5ea..3030af9e7b35 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -24,6 +24,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "tilcdc_drv.h"
 #include "tilcdc_regs.h"
@@ -511,7 +512,7 @@ static int tilcdc_debugfs_init(struct drm_minor *minor)
 DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver tilcdc_driver = {
-	.driver_features    = (DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET |
+	.driver_features    = (DRIVER_GEM | DRIVER_MODESET |
 			       DRIVER_PRIME | DRIVER_ATOMIC),
 	.irq_handler        = tilcdc_irq,
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.h b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
index 62cea5ff5558..d86397da12a9 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.h
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
@@ -30,10 +30,9 @@
 #include <linux/list.h>
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_gem_cma_helper.h>
-#include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_bridge.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_gem_cma_helper.h>
 
 /* Defaulting to pixel clock defined on AM335x */
 #define TILCDC_DEFAULT_MAX_PIXELCLOCK  126000
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_external.c b/drivers/gpu/drm/tilcdc/tilcdc_external.c
index b4eaf9bc87f8..e9969cd36610 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_external.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_external.c
@@ -10,6 +10,7 @@
 
 #include <linux/component.h>
 #include <linux/of_graph.h>
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_of.h>
 
 #include "tilcdc_drv.h"
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_panel.c b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
index a1acab39d87f..5d532a596e1e 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_panel.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
@@ -23,6 +23,7 @@
 #include <video/of_display_timing.h>
 #include <video/videomode.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "tilcdc_drv.h"
 #include "tilcdc_panel.h"
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
index daebf1aa6b0a..fe59fbfdde69 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
@@ -21,6 +21,7 @@
 #include <linux/pinctrl/pinmux.h>
 #include <linux/pinctrl/consumer.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "tilcdc_drv.h"
 #include "tilcdc_tfp410.h"
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-core.c b/drivers/gpu/drm/tinydrm/core/tinydrm-core.c
index 614f532ea89f..554abd5d3b53 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-core.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-core.c
@@ -9,10 +9,10 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_print.h>
 #include <drm/tinydrm/tinydrm.h>
 #include <linux/device.h>
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
index 323564329535..bb5b1c1e21ba 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
@@ -8,10 +8,10 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_modes.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_print.h>
 #include <drm/tinydrm/tinydrm.h>
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 0ec08394e17a..3f56647cdb35 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -198,19 +198,22 @@ static void ttm_bo_ref_bug(struct kref *list_kref)
 
 void ttm_bo_del_from_lru(struct ttm_buffer_object *bo)
 {
+	struct ttm_bo_device *bdev = bo->bdev;
+	bool notify = false;
+
 	if (!list_empty(&bo->swap)) {
 		list_del_init(&bo->swap);
 		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+		notify = true;
 	}
 	if (!list_empty(&bo->lru)) {
 		list_del_init(&bo->lru);
 		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+		notify = true;
 	}
 
-	/*
-	 * TODO: Add a driver hook to delete from
-	 * driver-specific LRU's here.
-	 */
+	if (notify && bdev->driver->del_from_lru_notify)
+		bdev->driver->del_from_lru_notify(bo);
 }
 
 void ttm_bo_del_sub_from_lru(struct ttm_buffer_object *bo)
@@ -676,15 +679,6 @@ void ttm_bo_put(struct ttm_buffer_object *bo)
 }
 EXPORT_SYMBOL(ttm_bo_put);
 
-void ttm_bo_unref(struct ttm_buffer_object **p_bo)
-{
-	struct ttm_buffer_object *bo = *p_bo;
-
-	*p_bo = NULL;
-	ttm_bo_put(bo);
-}
-EXPORT_SYMBOL(ttm_bo_unref);
-
 int ttm_bo_lock_delayed_workqueue(struct ttm_bo_device *bdev)
 {
 	return cancel_delayed_work_sync(&bdev->wq);
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index a1d977fbade5..e86a29a1e51f 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -71,7 +71,7 @@ static vm_fault_t ttm_bo_vm_fault_idle(struct ttm_buffer_object *bo,
 		ttm_bo_get(bo);
 		up_read(&vmf->vma->vm_mm->mmap_sem);
 		(void) dma_fence_wait(bo->moving, true);
-		ttm_bo_unreserve(bo);
+		reservation_object_unlock(bo->resv);
 		ttm_bo_put(bo);
 		goto out_unlock;
 	}
@@ -131,11 +131,7 @@ static vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf)
 	 * for reserve, and if it fails, retry the fault after waiting
 	 * for the buffer to become unreserved.
 	 */
-	err = ttm_bo_reserve(bo, true, true, NULL);
-	if (unlikely(err != 0)) {
-		if (err != -EBUSY)
-			return VM_FAULT_NOPAGE;
-
+	if (unlikely(!reservation_object_trylock(bo->resv))) {
 		if (vmf->flags & FAULT_FLAG_ALLOW_RETRY) {
 			if (!(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) {
 				ttm_bo_get(bo);
@@ -165,6 +161,8 @@ static vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf)
 	}
 
 	if (bdev->driver->fault_reserve_notify) {
+		struct dma_fence *moving = dma_fence_get(bo->moving);
+
 		err = bdev->driver->fault_reserve_notify(bo);
 		switch (err) {
 		case 0:
@@ -177,6 +175,13 @@ static vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf)
 			ret = VM_FAULT_SIGBUS;
 			goto out_unlock;
 		}
+
+		if (bo->moving != moving) {
+			spin_lock(&bdev->glob->lru_lock);
+			ttm_bo_move_to_lru_tail(bo, NULL);
+			spin_unlock(&bdev->glob->lru_lock);
+		}
+		dma_fence_put(moving);
 	}
 
 	/*
@@ -291,7 +296,7 @@ static vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf)
 out_io_unlock:
 	ttm_mem_io_unlock(man);
 out_unlock:
-	ttm_bo_unreserve(bo);
+	reservation_object_unlock(bo->resv);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/tve200/tve200_drv.c b/drivers/gpu/drm/tve200/tve200_drv.c
index 28e2d03c0ccf..d5c6a7ecf232 100644
--- a/drivers/gpu/drm/tve200/tve200_drv.c
+++ b/drivers/gpu/drm/tve200/tve200_drv.c
@@ -43,14 +43,14 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_panel.h>
 #include <drm/drm_of.h>
-#include <drm/drm_bridge.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 
 #include "tve200_drm.h"
 
diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c
index 68e88bed77ca..66885c24590f 100644
--- a/drivers/gpu/drm/udl/udl_connector.c
+++ b/drivers/gpu/drm/udl/udl_connector.c
@@ -14,6 +14,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "udl_connector.h"
 #include "udl_drv.h"
 
diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c
index a63e3011e971..22cd2d13e272 100644
--- a/drivers/gpu/drm/udl/udl_drv.c
+++ b/drivers/gpu/drm/udl/udl_drv.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "udl_drv.h"
 
 static int udl_usb_suspend(struct usb_interface *interface,
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index 1b014d92855b..9086d0d1b880 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -12,6 +12,7 @@
  */
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "udl_drv.h"
 
 /* -BULK_SIZE as per usb-skeleton. Can we get full page and avoid overhead? */
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index f7508e907536..4704b2df3688 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -234,18 +234,21 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 	for (q = 0; q < V3D_MAX_QUEUES; q++) {
 		struct drm_gpu_scheduler *sched = &v3d->queue[q].sched;
 
-		kthread_park(sched->thread);
-		drm_sched_hw_job_reset(sched, (sched_job->sched == sched ?
-					       sched_job : NULL));
+		drm_sched_stop(sched);
+
+		if(sched_job)
+			drm_sched_increase_karma(sched_job);
 	}
 
 	/* get the GPU back into the init state */
 	v3d_reset(v3d);
 
+	for (q = 0; q < V3D_MAX_QUEUES; q++)
+		drm_sched_resubmit_jobs(sched_job->sched);
+
 	/* Unblock schedulers and restart their jobs. */
 	for (q = 0; q < V3D_MAX_QUEUES; q++) {
-		drm_sched_job_recovery(&v3d->queue[q].sched);
-		kthread_unpark(v3d->queue[q].sched.thread);
+		drm_sched_start(&v3d->queue[q].sched, true);
 	}
 
 	mutex_unlock(&v3d->reset_lock);
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 97caf1671dd0..730008d3da76 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -34,8 +34,8 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_uapi.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <linux/component.h>
diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c
index f185812970da..169521e547ba 100644
--- a/drivers/gpu/drm/vc4/vc4_dpi.c
+++ b/drivers/gpu/drm/vc4/vc4_dpi.c
@@ -24,10 +24,10 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/of_graph.h>
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index f6f5cd80c04d..5fcd2f0da7f7 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -175,7 +175,6 @@ static struct drm_driver vc4_drm_driver = {
 	.driver_features = (DRIVER_MODESET |
 			    DRIVER_ATOMIC |
 			    DRIVER_GEM |
-			    DRIVER_HAVE_IRQ |
 			    DRIVER_RENDER |
 			    DRIVER_PRIME |
 			    DRIVER_SYNCOBJ),
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 0c607eb33d7e..11702e1d9011 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -30,11 +30,11 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_of.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/clk-provider.h>
 #include <linux/completion.h>
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 051b61b62541..88fd5df7e7dc 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -43,8 +43,8 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/i2c.h>
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index 0490edb192a1..91b8c72ff361 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -17,9 +17,9 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_plane_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_plane_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
diff --git a/drivers/gpu/drm/vc4/vc4_perfmon.c b/drivers/gpu/drm/vc4/vc4_perfmon.c
index 437e7a27f21d..495150415020 100644
--- a/drivers/gpu/drm/vc4/vc4_perfmon.c
+++ b/drivers/gpu/drm/vc4/vc4_perfmon.c
@@ -117,7 +117,7 @@ int vc4_perfmon_create_ioctl(struct drm_device *dev, void *data,
 			return -EINVAL;
 	}
 
-	perfmon = kzalloc(sizeof(*perfmon) + (req->ncounters * sizeof(u64)),
+	perfmon = kzalloc(struct_size(perfmon, counters, req->ncounters),
 			  GFP_KERNEL);
 	if (!perfmon)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index 6e23c50168f9..aa279b5b0de7 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -9,9 +9,9 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_cma_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_writeback.h>
 #include <linux/clk.h>
 #include <linux/component.h>
diff --git a/drivers/gpu/drm/vc4/vc4_vec.c b/drivers/gpu/drm/vc4/vc4_vec.c
index 8e7facb6514e..858c3a483229 100644
--- a/drivers/gpu/drm/vc4/vc4_vec.c
+++ b/drivers/gpu/drm/vc4/vc4_vec.c
@@ -25,9 +25,9 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/of_graph.h>
diff --git a/drivers/gpu/drm/via/via_dmablit.c b/drivers/gpu/drm/via/via_dmablit.c
index 345bda4494e1..8bf3a7c23ed3 100644
--- a/drivers/gpu/drm/via/via_dmablit.c
+++ b/drivers/gpu/drm/via/via_dmablit.c
@@ -177,12 +177,14 @@ via_free_sg_info(struct pci_dev *pdev, drm_via_sg_info_t *vsg)
 	switch (vsg->state) {
 	case dr_via_device_mapped:
 		via_unmap_blit_from_device(pdev, vsg);
+		/* fall through */
 	case dr_via_desc_pages_alloc:
 		for (i = 0; i < vsg->num_desc_pages; ++i) {
 			if (vsg->desc_pages[i] != NULL)
 				free_page((unsigned long)vsg->desc_pages[i]);
 		}
 		kfree(vsg->desc_pages);
+		/* fall through */
 	case dr_via_pages_locked:
 		for (i = 0; i < vsg->num_pages; ++i) {
 			if (NULL != (page = vsg->pages[i])) {
@@ -191,8 +193,10 @@ via_free_sg_info(struct pci_dev *pdev, drm_via_sg_info_t *vsg)
 				put_page(page);
 			}
 		}
+		/* fall through */
 	case dr_via_pages_alloc:
 		vfree(vsg->pages);
+		/* fall through */
 	default:
 		vsg->state = dr_via_sg_init;
 	}
diff --git a/drivers/gpu/drm/via/via_drv.c b/drivers/gpu/drm/via/via_drv.c
index aaf766f7cca2..af6a12d3c058 100644
--- a/drivers/gpu/drm/via/via_drv.c
+++ b/drivers/gpu/drm/via/via_drv.c
@@ -70,8 +70,7 @@ static const struct file_operations via_driver_fops = {
 
 static struct drm_driver driver = {
 	.driver_features =
-	    DRIVER_USE_AGP | DRIVER_HAVE_IRQ | DRIVER_LEGACY |
-	    DRIVER_IRQ_SHARED,
+	    DRIVER_USE_AGP | DRIVER_HAVE_IRQ | DRIVER_LEGACY,
 	.load = via_driver_load,
 	.unload = via_driver_unload,
 	.open = via_driver_open,
diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index 87d7c49cf057..653ec7d0bf4d 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -26,9 +26,9 @@
  */
 
 #include "virtgpu_drv.h"
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #define XRES_MIN    32
 #define YRES_MIN    32
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index af92964b6889..b996ac1d4fcc 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -205,14 +205,10 @@ static struct drm_driver driver = {
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = virtio_gpu_debugfs_init,
 #endif
-	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
-	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export = drm_gem_prime_export,
 	.gem_prime_import = drm_gem_prime_import,
 	.gem_prime_pin = virtgpu_gem_prime_pin,
 	.gem_prime_unpin = virtgpu_gem_prime_unpin,
-	.gem_prime_get_sg_table = virtgpu_gem_prime_get_sg_table,
-	.gem_prime_import_sg_table = virtgpu_gem_prime_import_sg_table,
 	.gem_prime_vmap = virtgpu_gem_prime_vmap,
 	.gem_prime_vunmap = virtgpu_gem_prime_vunmap,
 	.gem_prime_mmap = virtgpu_gem_prime_mmap,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index cf896d879382..3238fdf58eb4 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -34,9 +34,9 @@
 #include <drm/drmP.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_atomic.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/ttm/ttm_bo_api.h>
 #include <drm/ttm/ttm_bo_driver.h>
 #include <drm/ttm/ttm_placement.h>
@@ -354,10 +354,6 @@ int virtio_gpu_object_wait(struct virtio_gpu_object *bo, bool no_wait);
 /* virtgpu_prime.c */
 int virtgpu_gem_prime_pin(struct drm_gem_object *obj);
 void virtgpu_gem_prime_unpin(struct drm_gem_object *obj);
-struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj);
-struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
-	struct drm_device *dev, struct dma_buf_attachment *attach,
-	struct sg_table *sgt);
 void *virtgpu_gem_prime_vmap(struct drm_gem_object *obj);
 void virtgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
 int virtgpu_gem_prime_mmap(struct drm_gem_object *obj,
diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
index f39a183d59c2..e7e946035027 100644
--- a/drivers/gpu/drm/virtio/virtgpu_object.c
+++ b/drivers/gpu/drm/virtio/virtgpu_object.c
@@ -28,10 +28,21 @@
 static int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev,
 				       uint32_t *resid)
 {
+#if 0
 	int handle = ida_alloc(&vgdev->resource_ida, GFP_KERNEL);
 
 	if (handle < 0)
 		return handle;
+#else
+	static int handle;
+
+	/*
+	 * FIXME: dirty hack to avoid re-using IDs, virglrenderer
+	 * can't deal with that.  Needs fixing in virglrenderer, also
+	 * should figure a better way to handle that in the guest.
+	 */
+	handle++;
+#endif
 
 	*resid = handle + 1;
 	return 0;
@@ -39,7 +50,9 @@ static int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev,
 
 static void virtio_gpu_resource_id_put(struct virtio_gpu_device *vgdev, uint32_t id)
 {
+#if 0
 	ida_free(&vgdev->resource_ida, id - 1);
+#endif
 }
 
 static void virtio_gpu_ttm_bo_destroy(struct ttm_buffer_object *tbo)
diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c
index 86ce0ae93f59..c59ec34c80a5 100644
--- a/drivers/gpu/drm/virtio/virtgpu_prime.c
+++ b/drivers/gpu/drm/virtio/virtgpu_prime.c
@@ -39,20 +39,6 @@ void virtgpu_gem_prime_unpin(struct drm_gem_object *obj)
 	WARN_ONCE(1, "not implemented");
 }
 
-struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj)
-{
-	WARN_ONCE(1, "not implemented");
-	return ERR_PTR(-ENODEV);
-}
-
-struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
-	struct drm_device *dev, struct dma_buf_attachment *attach,
-	struct sg_table *table)
-{
-	WARN_ONCE(1, "not implemented");
-	return ERR_PTR(-ENODEV);
-}
-
 void *virtgpu_gem_prime_vmap(struct drm_gem_object *obj)
 {
 	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
diff --git a/drivers/gpu/drm/vkms/vkms_crc.c b/drivers/gpu/drm/vkms/vkms_crc.c
index 9d9e8146db90..d7b409a3c0f8 100644
--- a/drivers/gpu/drm/vkms/vkms_crc.c
+++ b/drivers/gpu/drm/vkms/vkms_crc.c
@@ -1,4 +1,5 @@
-// SPDX-License-Identifier: GPL-2.0
+// SPDX-License-Identifier: GPL-2.0+
+
 #include "vkms_drv.h"
 #include <linux/crc32.h>
 #include <drm/drm_atomic.h>
diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c
index 177bbcb38306..8a9aeb0a9ea8 100644
--- a/drivers/gpu/drm/vkms/vkms_crtc.c
+++ b/drivers/gpu/drm/vkms/vkms_crtc.c
@@ -1,22 +1,20 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
+// SPDX-License-Identifier: GPL-2.0+
 
 #include "vkms_drv.h"
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
-static void _vblank_handle(struct vkms_output *output)
+static enum hrtimer_restart vkms_vblank_simulate(struct hrtimer *timer)
 {
+	struct vkms_output *output = container_of(timer, struct vkms_output,
+						  vblank_hrtimer);
 	struct drm_crtc *crtc = &output->crtc;
 	struct vkms_crtc_state *state = to_vkms_crtc_state(crtc->state);
+	u64 ret_overrun;
 	bool ret;
 
 	spin_lock(&output->lock);
+
 	ret = drm_crtc_handle_vblank(crtc);
 	if (!ret)
 		DRM_ERROR("vkms failure on handling vblank");
@@ -37,19 +35,11 @@ static void _vblank_handle(struct vkms_output *output)
 			DRM_WARN("failed to queue vkms_crc_work_handle");
 	}
 
-	spin_unlock(&output->lock);
-}
-
-static enum hrtimer_restart vkms_vblank_simulate(struct hrtimer *timer)
-{
-	struct vkms_output *output = container_of(timer, struct vkms_output,
-						  vblank_hrtimer);
-	int ret_overrun;
-
-	_vblank_handle(output);
-
 	ret_overrun = hrtimer_forward_now(&output->vblank_hrtimer,
 					  output->period_ns);
+	WARN_ON(ret_overrun != 1);
+
+	spin_unlock(&output->lock);
 
 	return HRTIMER_RESTART;
 }
@@ -87,6 +77,9 @@ bool vkms_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe,
 
 	*vblank_time = output->vblank_hrtimer.node.expires;
 
+	if (!in_vblank_irq)
+		*vblank_time -= output->period_ns;
+
 	return true;
 }
 
@@ -104,6 +97,7 @@ static void vkms_atomic_crtc_reset(struct drm_crtc *crtc)
 	vkms_state = kzalloc(sizeof(*vkms_state), GFP_KERNEL);
 	if (!vkms_state)
 		return;
+	INIT_WORK(&vkms_state->crc_work, vkms_crc_work_handle);
 
 	crtc->state = &vkms_state->base;
 	crtc->state->crtc = crtc;
diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c
index 2a16b86196dc..738dd6206d85 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.c
+++ b/drivers/gpu/drm/vkms/vkms_drv.c
@@ -1,9 +1,4 @@
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
+// SPDX-License-Identifier: GPL-2.0+
 
 /**
  * DOC: vkms (Virtual Kernel Modesetting)
@@ -16,10 +11,10 @@
 
 #include <linux/module.h>
 #include <drm/drm_gem.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_probe_helper.h>
 #include "vkms_drv.h"
 
 #define DRIVER_NAME	"vkms"
diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h
index e4469cd3d254..81f1cfbeb936 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.h
+++ b/drivers/gpu/drm/vkms/vkms_drv.h
@@ -1,3 +1,5 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+
 #ifndef _VKMS_DRV_H_
 #define _VKMS_DRV_H_
 
diff --git a/drivers/gpu/drm/vkms/vkms_gem.c b/drivers/gpu/drm/vkms/vkms_gem.c
index 80311daed47a..138b0bb325cf 100644
--- a/drivers/gpu/drm/vkms/vkms_gem.c
+++ b/drivers/gpu/drm/vkms/vkms_gem.c
@@ -1,10 +1,4 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
+// SPDX-License-Identifier: GPL-2.0+
 
 #include <linux/shmem_fs.h>
 
diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c
index 271a0eb9042c..3b162b25312e 100644
--- a/drivers/gpu/drm/vkms/vkms_output.c
+++ b/drivers/gpu/drm/vkms/vkms_output.c
@@ -1,14 +1,8 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
+// SPDX-License-Identifier: GPL-2.0+
 
 #include "vkms_drv.h"
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
 
 static void vkms_connector_destroy(struct drm_connector *connector)
 {
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 418817600ad1..0e67d2d42f0c 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -1,10 +1,4 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
+// SPDX-License-Identifier: GPL-2.0+
 
 #include "vkms_drv.h"
 #include <drm/drm_plane_helper.h>
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
index 7ce1c2f87d9a..5d5c2bce01f3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
@@ -534,7 +534,6 @@ static void vmw_user_bo_release(struct ttm_base_object **p_base)
 {
 	struct vmw_user_buffer_object *vmw_user_bo;
 	struct ttm_base_object *base = *p_base;
-	struct ttm_buffer_object *bo;
 
 	*p_base = NULL;
 
@@ -543,8 +542,7 @@ static void vmw_user_bo_release(struct ttm_base_object **p_base)
 
 	vmw_user_bo = container_of(base, struct vmw_user_buffer_object,
 				   prime.base);
-	bo = &vmw_user_bo->vbo.base;
-	ttm_bo_unref(&bo);
+	ttm_bo_put(&vmw_user_bo->vbo.base);
 }
 
 
@@ -597,7 +595,6 @@ int vmw_user_bo_alloc(struct vmw_private *dev_priv,
 		      struct ttm_base_object **p_base)
 {
 	struct vmw_user_buffer_object *user_bo;
-	struct ttm_buffer_object *tmp;
 	int ret;
 
 	user_bo = kzalloc(sizeof(*user_bo), GFP_KERNEL);
@@ -614,7 +611,7 @@ int vmw_user_bo_alloc(struct vmw_private *dev_priv,
 	if (unlikely(ret != 0))
 		return ret;
 
-	tmp = ttm_bo_reference(&user_bo->vbo.base);
+	ttm_bo_get(&user_bo->vbo.base);
 	ret = ttm_prime_object_init(tfile,
 				    size,
 				    &user_bo->prime,
@@ -623,7 +620,7 @@ int vmw_user_bo_alloc(struct vmw_private *dev_priv,
 				    &vmw_user_bo_release,
 				    &vmw_user_bo_ref_obj_release);
 	if (unlikely(ret != 0)) {
-		ttm_bo_unref(&tmp);
+		ttm_bo_put(&user_bo->vbo.base);
 		goto out_no_base_object;
 	}
 
@@ -911,7 +908,7 @@ int vmw_user_bo_lookup(struct ttm_object_file *tfile,
 
 	vmw_user_bo = container_of(base, struct vmw_user_buffer_object,
 				   prime.base);
-	(void)ttm_bo_reference(&vmw_user_bo->vbo.base);
+	ttm_bo_get(&vmw_user_bo->vbo.base);
 	if (p_base)
 		*p_base = base;
 	else
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
index 48d1380a952e..70dab55e7888 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
@@ -765,7 +765,7 @@ static bool vmw_cmdbuf_try_alloc(struct vmw_cmdbuf_man *man,
 
 	if (info->done)
 		return true;
- 
+
 	memset(info->node, 0, sizeof(*info->node));
 	spin_lock(&man->lock);
 	ret = drm_mm_insert_node(&man->mm, info->node, info->page_size);
@@ -1276,8 +1276,10 @@ int vmw_cmdbuf_set_pool_size(struct vmw_cmdbuf_man *man,
 	return 0;
 
 out_no_map:
-	if (man->using_mob)
-		ttm_bo_unref(&man->cmd_space);
+	if (man->using_mob) {
+		ttm_bo_put(man->cmd_space);
+		man->cmd_space = NULL;
+	}
 
 	return ret;
 }
@@ -1380,7 +1382,8 @@ void vmw_cmdbuf_remove_pool(struct vmw_cmdbuf_man *man)
 	(void) vmw_cmdbuf_idle(man, false, 10*HZ);
 	if (man->using_mob) {
 		(void) ttm_bo_kunmap(&man->map_obj);
-		ttm_bo_unref(&man->cmd_space);
+		ttm_bo_put(man->cmd_space);
+		man->cmd_space = NULL;
 	} else {
 		dma_free_coherent(&man->dev_priv->dev->pdev->dev,
 				  man->size, man->map, man->handle);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 25afb1d594e3..6165fe2c4504 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -26,6 +26,7 @@
  **************************************************************************/
 #include <linux/module.h>
 #include <linux/console.h>
+#include <linux/dma-mapping.h>
 
 #include <drm/drmP.h>
 #include "vmwgfx_drv.h"
@@ -34,7 +35,6 @@
 #include <drm/ttm/ttm_placement.h>
 #include <drm/ttm/ttm_bo_driver.h>
 #include <drm/ttm/ttm_module.h>
-#include <linux/intel-iommu.h>
 
 #define VMWGFX_DRIVER_DESC "Linux drm driver for VMware graphics devices"
 #define VMWGFX_CHIP_SVGAII 0
@@ -546,6 +546,21 @@ static void vmw_get_initial_size(struct vmw_private *dev_priv)
 }
 
 /**
+ * vmw_assume_iommu - Figure out whether coherent dma-remapping might be
+ * taking place.
+ * @dev: Pointer to the struct drm_device.
+ *
+ * Return: true if iommu present, false otherwise.
+ */
+static bool vmw_assume_iommu(struct drm_device *dev)
+{
+	const struct dma_map_ops *ops = get_dma_ops(dev->dev);
+
+	return !dma_is_direct(ops) && ops &&
+		ops->map_page != dma_direct_map_page;
+}
+
+/**
  * vmw_dma_select_mode - Determine how DMA mappings should be set up for this
  * system.
  *
@@ -565,55 +580,27 @@ static int vmw_dma_select_mode(struct vmw_private *dev_priv)
 		[vmw_dma_alloc_coherent] = "Using coherent TTM pages.",
 		[vmw_dma_map_populate] = "Keeping DMA mappings.",
 		[vmw_dma_map_bind] = "Giving up DMA mappings early."};
-#ifdef CONFIG_X86
-	const struct dma_map_ops *dma_ops = get_dma_ops(dev_priv->dev->dev);
 
-#ifdef CONFIG_INTEL_IOMMU
-	if (intel_iommu_enabled) {
+	if (vmw_force_coherent)
+		dev_priv->map_mode = vmw_dma_alloc_coherent;
+	else if (vmw_assume_iommu(dev_priv->dev))
 		dev_priv->map_mode = vmw_dma_map_populate;
-		goto out_fixup;
-	}
-#endif
-
-	if (!(vmw_force_iommu || vmw_force_coherent)) {
+	else if (!vmw_force_iommu)
 		dev_priv->map_mode = vmw_dma_phys;
-		DRM_INFO("DMA map mode: %s\n", names[dev_priv->map_mode]);
-		return 0;
-	}
-
-	dev_priv->map_mode = vmw_dma_map_populate;
-
-	if (dma_ops && dma_ops->sync_single_for_cpu)
+	else if (IS_ENABLED(CONFIG_SWIOTLB) && swiotlb_nr_tbl())
 		dev_priv->map_mode = vmw_dma_alloc_coherent;
-#ifdef CONFIG_SWIOTLB
-	if (swiotlb_nr_tbl() == 0)
+	else
 		dev_priv->map_mode = vmw_dma_map_populate;
-#endif
 
-#ifdef CONFIG_INTEL_IOMMU
-out_fixup:
-#endif
-	if (dev_priv->map_mode == vmw_dma_map_populate &&
-	    vmw_restrict_iommu)
+	if (dev_priv->map_mode == vmw_dma_map_populate && vmw_restrict_iommu)
 		dev_priv->map_mode = vmw_dma_map_bind;
 
-	if (vmw_force_coherent)
-		dev_priv->map_mode = vmw_dma_alloc_coherent;
-
-#if !defined(CONFIG_SWIOTLB) && !defined(CONFIG_INTEL_IOMMU)
-	/*
-	 * No coherent page pool
-	 */
-	if (dev_priv->map_mode == vmw_dma_alloc_coherent)
+	/* No TTM coherent page pool? FIXME: Ask TTM instead! */
+        if (!(IS_ENABLED(CONFIG_SWIOTLB) || IS_ENABLED(CONFIG_INTEL_IOMMU)) &&
+	    (dev_priv->map_mode == vmw_dma_alloc_coherent))
 		return -EINVAL;
-#endif
-
-#else /* CONFIG_X86 */
-	dev_priv->map_mode = vmw_dma_map_populate;
-#endif /* CONFIG_X86 */
 
 	DRM_INFO("DMA map mode: %s\n", names[dev_priv->map_mode]);
-
 	return 0;
 }
 
@@ -625,24 +612,20 @@ out_fixup:
  * With 32-bit we can only handle 32 bit PFNs. Optionally set that
  * restriction also for 64-bit systems.
  */
-#ifdef CONFIG_INTEL_IOMMU
 static int vmw_dma_masks(struct vmw_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
+	int ret = 0;
 
-	if (intel_iommu_enabled &&
+	ret = dma_set_mask_and_coherent(dev->dev, DMA_BIT_MASK(64));
+	if (dev_priv->map_mode != vmw_dma_phys &&
 	    (sizeof(unsigned long) == 4 || vmw_restrict_dma_mask)) {
 		DRM_INFO("Restricting DMA addresses to 44 bits.\n");
-		return dma_set_mask(dev->dev, DMA_BIT_MASK(44));
+		return dma_set_mask_and_coherent(dev->dev, DMA_BIT_MASK(44));
 	}
-	return 0;
-}
-#else
-static int vmw_dma_masks(struct vmw_private *dev_priv)
-{
-	return 0;
+
+	return ret;
 }
-#endif
 
 static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 {
@@ -1582,7 +1565,7 @@ static const struct file_operations vmwgfx_driver_fops = {
 };
 
 static struct drm_driver driver = {
-	.driver_features = DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED |
+	.driver_features =
 	DRIVER_MODESET | DRIVER_PRIME | DRIVER_RENDER | DRIVER_ATOMIC,
 	.load = vmw_driver_load,
 	.unload = vmw_driver_unload,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index cd607ba9c2fe..accb2fafe2f1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -1337,18 +1337,15 @@ static inline void vmw_bo_unreference(struct vmw_buffer_object **buf)
 
 	*buf = NULL;
 	if (tmp_buf != NULL) {
-		struct ttm_buffer_object *bo = &tmp_buf->base;
-
-		ttm_bo_unref(&bo);
+		ttm_bo_put(&tmp_buf->base);
 	}
 }
 
 static inline struct vmw_buffer_object *
 vmw_bo_reference(struct vmw_buffer_object *buf)
 {
-	if (ttm_bo_reference(&buf->base))
-		return buf;
-	return NULL;
+	ttm_bo_get(&buf->base);
+	return buf;
 }
 
 static inline struct ttm_mem_global *vmw_mem_glob(struct vmw_private *dev_priv)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index f2d13a72c05d..88b8178d4687 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -3570,7 +3570,7 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv,
 		*p_fence = NULL;
 	}
 
-	return 0;
+	return ret;
 }
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index b351fb5214d3..ed2f67822f45 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -1646,7 +1646,7 @@ static int vmw_kms_check_topology(struct drm_device *dev,
 		struct drm_connector_state *conn_state;
 		struct vmw_connector_state *vmw_conn_state;
 
-		if (!du->pref_active) {
+		if (!du->pref_active && new_crtc_state->enable) {
 			ret = -EINVAL;
 			goto clean;
 		}
@@ -2554,8 +2554,8 @@ void vmw_kms_helper_validation_finish(struct vmw_private *dev_priv,
 				      user_fence_rep)
 {
 	struct vmw_fence_obj *fence = NULL;
-	uint32_t handle;
-	int ret;
+	uint32_t handle = 0;
+	int ret = 0;
 
 	if (file_priv || user_fence_rep || vmw_validation_has_bos(ctx) ||
 	    out_fence)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index 655abbcd4058..535b03599e55 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -29,8 +29,8 @@
 #define VMWGFX_KMS_H_
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
+#include <drm/drm_probe_helper.h>
 #include "vmwgfx_drv.h"
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
index 7ed179d30ec5..d83cc66e1210 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
@@ -300,7 +300,8 @@ out_no_setup:
 						 &batch->otables[i]);
 	}
 
-	ttm_bo_unref(&batch->otable_bo);
+	ttm_bo_put(batch->otable_bo);
+	batch->otable_bo = NULL;
 out_no_bo:
 	return ret;
 }
@@ -365,7 +366,8 @@ static void vmw_otable_batch_takedown(struct vmw_private *dev_priv,
 	vmw_bo_fence_single(bo, NULL);
 	ttm_bo_unreserve(bo);
 
-	ttm_bo_unref(&batch->otable_bo);
+	ttm_bo_put(batch->otable_bo);
+	batch->otable_bo = NULL;
 }
 
 /*
@@ -463,7 +465,8 @@ static int vmw_mob_pt_populate(struct vmw_private *dev_priv,
 
 out_unreserve:
 	ttm_bo_unreserve(mob->pt_bo);
-	ttm_bo_unref(&mob->pt_bo);
+	ttm_bo_put(mob->pt_bo);
+	mob->pt_bo = NULL;
 
 	return ret;
 }
@@ -580,8 +583,10 @@ static void vmw_mob_pt_setup(struct vmw_mob *mob,
  */
 void vmw_mob_destroy(struct vmw_mob *mob)
 {
-	if (mob->pt_bo)
-		ttm_bo_unref(&mob->pt_bo);
+	if (mob->pt_bo) {
+		ttm_bo_put(mob->pt_bo);
+		mob->pt_bo = NULL;
+	}
 	kfree(mob);
 }
 
@@ -698,8 +703,10 @@ int vmw_mob_bind(struct vmw_private *dev_priv,
 
 out_no_cmd_space:
 	vmw_fifo_resource_dec(dev_priv);
-	if (pt_set_up)
-		ttm_bo_unref(&mob->pt_bo);
+	if (pt_set_up) {
+		ttm_bo_put(mob->pt_bo);
+		mob->pt_bo = NULL;
+	}
 
 	return -ENOMEM;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index 3025bfc001a1..a7c30e567f09 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -461,7 +461,8 @@ vmw_resource_check_buffer(struct ww_acquire_ctx *ticket,
 	}
 
 	INIT_LIST_HEAD(&val_list);
-	val_buf->bo = ttm_bo_reference(&res->backup->base);
+	ttm_bo_get(&res->backup->base);
+	val_buf->bo = &res->backup->base;
 	val_buf->num_shared = 0;
 	list_add_tail(&val_buf->head, &val_list);
 	ret = ttm_eu_reserve_buffers(ticket, &val_list, interruptible, NULL);
@@ -484,7 +485,8 @@ vmw_resource_check_buffer(struct ww_acquire_ctx *ticket,
 out_no_validate:
 	ttm_eu_backoff_reservation(ticket, &val_list);
 out_no_reserve:
-	ttm_bo_unref(&val_buf->bo);
+	ttm_bo_put(val_buf->bo);
+	val_buf->bo = NULL;
 	if (backup_dirty)
 		vmw_bo_unreference(&res->backup);
 
@@ -544,7 +546,8 @@ vmw_resource_backoff_reservation(struct ww_acquire_ctx *ticket,
 	INIT_LIST_HEAD(&val_list);
 	list_add_tail(&val_buf->head, &val_list);
 	ttm_eu_backoff_reservation(ticket, &val_list);
-	ttm_bo_unref(&val_buf->bo);
+	ttm_bo_put(val_buf->bo);
+	val_buf->bo = NULL;
 }
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
index b3f547fc5d3d..e9944ac2e057 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
@@ -628,8 +628,10 @@ void vmw_validation_unref_lists(struct vmw_validation_context *ctx)
 	struct vmw_validation_bo_node *entry;
 	struct vmw_validation_res_node *val;
 
-	list_for_each_entry(entry, &ctx->bo_list, base.head)
-		ttm_bo_unref(&entry->base.bo);
+	list_for_each_entry(entry, &ctx->bo_list, base.head) {
+		ttm_bo_put(entry->base.bo);
+		entry->base.bo = NULL;
+	}
 
 	list_splice_init(&ctx->resource_ctx_list, &ctx->resource_list);
 	list_for_each_entry(val, &ctx->resource_list, head)
diff --git a/drivers/gpu/drm/xen/xen_drm_front.c b/drivers/gpu/drm/xen/xen_drm_front.c
index 4d3d36fc3a5d..3e78a832d7f9 100644
--- a/drivers/gpu/drm/xen/xen_drm_front.c
+++ b/drivers/gpu/drm/xen/xen_drm_front.c
@@ -10,7 +10,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drm_gem.h>
 
 #include <linux/of_device.h>
diff --git a/drivers/gpu/drm/xen/xen_drm_front_conn.c b/drivers/gpu/drm/xen/xen_drm_front_conn.c
index 54af2669b1b3..9f5f31f77f1e 100644
--- a/drivers/gpu/drm/xen/xen_drm_front_conn.c
+++ b/drivers/gpu/drm/xen/xen_drm_front_conn.c
@@ -9,7 +9,7 @@
  */
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include <video/videomode.h>
 
diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c
index 28bc501af450..53c376d55fcf 100644
--- a/drivers/gpu/drm/xen/xen_drm_front_gem.c
+++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c
@@ -11,9 +11,9 @@
 #include "xen_drm_front_gem.h"
 
 #include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem.h>
+#include <drm/drm_probe_helper.h>
 
 #include <linux/dma-buf.h>
 #include <linux/scatterlist.h>
@@ -235,8 +235,14 @@ static int gem_mmap_obj(struct xen_gem_object *xen_obj,
 	vma->vm_flags &= ~VM_PFNMAP;
 	vma->vm_flags |= VM_MIXEDMAP;
 	vma->vm_pgoff = 0;
-	vma->vm_page_prot =
-			pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
+	/*
+	 * According to Xen on ARM ABI (xen/include/public/arch-arm.h):
+	 * all memory which is shared with other entities in the system
+	 * (including the hypervisor and other guests) must reside in memory
+	 * which is mapped as Normal Inner Write-Back Outer Write-Back
+	 * Inner-Shareable.
+	 */
+	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
 
 	/*
 	 * vm_operations_struct.fault handler will be called if CPU access
@@ -282,8 +288,9 @@ void *xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj)
 	if (!xen_obj->pages)
 		return NULL;
 
+	/* Please see comment in gem_mmap_obj on mapping and attributes. */
 	return vmap(xen_obj->pages, xen_obj->num_pages,
-		    VM_MAP, pgprot_writecombine(PAGE_KERNEL));
+		    VM_MAP, PAGE_KERNEL);
 }
 
 void xen_drm_front_gem_prime_vunmap(struct drm_gem_object *gem_obj,
diff --git a/drivers/gpu/drm/xen/xen_drm_front_kms.c b/drivers/gpu/drm/xen/xen_drm_front_kms.c
index a3479eb72d79..c2955d375394 100644
--- a/drivers/gpu/drm/xen/xen_drm_front_kms.c
+++ b/drivers/gpu/drm/xen/xen_drm_front_kms.c
@@ -13,9 +13,9 @@
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
 
 #include "xen_drm_front.h"
 #include "xen_drm_front_conn.h"
@@ -54,7 +54,7 @@ fb_create(struct drm_device *dev, struct drm_file *filp,
 	  const struct drm_mode_fb_cmd2 *mode_cmd)
 {
 	struct xen_drm_front_drm_info *drm_info = dev->dev_private;
-	static struct drm_framebuffer *fb;
+	struct drm_framebuffer *fb;
 	struct drm_gem_object *gem_obj;
 	int ret;
 
diff --git a/drivers/gpu/drm/zte/zx_drm_drv.c b/drivers/gpu/drm/zte/zx_drm_drv.c
index f5ea32ae8600..28e8d6072910 100644
--- a/drivers/gpu/drm/zte/zx_drm_drv.c
+++ b/drivers/gpu/drm/zte/zx_drm_drv.c
@@ -18,12 +18,12 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_cma_helper.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/drmP.h>
 
 #include "zx_drm_drv.h"
diff --git a/drivers/gpu/drm/zte/zx_hdmi.c b/drivers/gpu/drm/zte/zx_hdmi.c
index 9fc98bb4f3d9..df522d74bebf 100644
--- a/drivers/gpu/drm/zte/zx_hdmi.c
+++ b/drivers/gpu/drm/zte/zx_hdmi.c
@@ -20,9 +20,9 @@
 #include <linux/of_device.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include <sound/hdmi-codec.h>
diff --git a/drivers/gpu/drm/zte/zx_tvenc.c b/drivers/gpu/drm/zte/zx_tvenc.c
index b73afb212fb2..87b5d86413d2 100644
--- a/drivers/gpu/drm/zte/zx_tvenc.c
+++ b/drivers/gpu/drm/zte/zx_tvenc.c
@@ -14,7 +14,7 @@
 #include <linux/regmap.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include "zx_drm_drv.h"
diff --git a/drivers/gpu/drm/zte/zx_vga.c b/drivers/gpu/drm/zte/zx_vga.c
index 23d1ff4355a0..e14c1d709740 100644
--- a/drivers/gpu/drm/zte/zx_vga.c
+++ b/drivers/gpu/drm/zte/zx_vga.c
@@ -13,7 +13,7 @@
 #include <linux/regmap.h>
 
 #include <drm/drm_atomic_helper.h>
-#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
 #include <drm/drmP.h>
 
 #include "zx_drm_drv.h"
diff --git a/drivers/gpu/drm/zte/zx_vou.c b/drivers/gpu/drm/zte/zx_vou.c
index 442311d31110..15400ffb1d22 100644
--- a/drivers/gpu/drm/zte/zx_vou.c
+++ b/drivers/gpu/drm/zte/zx_vou.c
@@ -15,12 +15,12 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_crtc_helper.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/drmP.h>
 
 #include "zx_common_regs.h"