Merge branch 'drm-next-4.18' of git://people.freedesktop.org/~agd5f/linux into drm-next

Fixes for 4.18. Highlights:
- Fixes for gfxoff on Raven
- Remove an ATPX quirk now that the root cause is fixed
- Runtime PM fixes
- Vega20 register header update
- Wattman fixes
- Misc bug fixes
Signed-off-by: NDave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180614141428.2909-1-alexander.deucher@amd.com

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c

@@ -342,15 +342,12 @@ void get_local_mem_info(struct kgd_dev *kgd,
 			mem_info->local_mem_size_public,
 			mem_info->local_mem_size_private);
 
-	if (amdgpu_emu_mode == 1) {
-		mem_info->mem_clk_max = 100;
-		return;
-	}
-
 	if (amdgpu_sriov_vf(adev))
 		mem_info->mem_clk_max = adev->clock.default_mclk / 100;
-	else
+	else if (adev->powerplay.pp_funcs)
 		mem_info->mem_clk_max = amdgpu_dpm_get_mclk(adev, false) / 100;
+	else
+		mem_info->mem_clk_max = 100;
 }
 
 uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
@@ -367,13 +364,12 @@ uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 
 	/* the sclk is in quantas of 10kHz */
-	if (amdgpu_emu_mode == 1)
-		return 100;
-
 	if (amdgpu_sriov_vf(adev))
 		return adev->clock.default_sclk / 100;
-
-	return amdgpu_dpm_get_sclk(adev, false) / 100;
+	else if (adev->powerplay.pp_funcs)
+		return amdgpu_dpm_get_sclk(adev, false) / 100;
+	else
+		return 100;
 }
 
 void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)

drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c

@@ -569,7 +569,6 @@ static const struct amdgpu_px_quirk amdgpu_px_quirk_list[] = {
 	{ 0x1002, 0x6900, 0x1002, 0x0124, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0x1002, 0x6900, 0x1028, 0x0812, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0x1002, 0x6900, 0x1028, 0x0813, AMDGPU_PX_QUIRK_FORCE_ATPX },
-	{ 0x1002, 0x67DF, 0x1028, 0x0774, AMDGPU_PX_QUIRK_FORCE_ATPX },
 	{ 0, 0, 0, 0, 0 },
 };
 

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -522,6 +522,9 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 	struct amdgpu_bo_list_entry *e;
 	struct list_head duplicates;
 	unsigned i, tries = 10;
+	struct amdgpu_bo *gds;
+	struct amdgpu_bo *gws;
+	struct amdgpu_bo *oa;
 	int r;
 
 	INIT_LIST_HEAD(&p->validated);
@@ -652,31 +655,36 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 
 	amdgpu_cs_report_moved_bytes(p->adev, p->bytes_moved,
 				     p->bytes_moved_vis);
+
 	if (p->bo_list) {
-		struct amdgpu_bo *gds = p->bo_list->gds_obj;
-		struct amdgpu_bo *gws = p->bo_list->gws_obj;
-		struct amdgpu_bo *oa = p->bo_list->oa_obj;
 		struct amdgpu_vm *vm = &fpriv->vm;
 		unsigned i;
 
+		gds = p->bo_list->gds_obj;
+		gws = p->bo_list->gws_obj;
+		oa = p->bo_list->oa_obj;
 		for (i = 0; i < p->bo_list->num_entries; i++) {
 			struct amdgpu_bo *bo = p->bo_list->array[i].robj;
 
 			p->bo_list->array[i].bo_va = amdgpu_vm_bo_find(vm, bo);
 		}
+	} else {
+		gds = p->adev->gds.gds_gfx_bo;
+		gws = p->adev->gds.gws_gfx_bo;
+		oa = p->adev->gds.oa_gfx_bo;
+	}
 
-		if (gds) {
-			p->job->gds_base = amdgpu_bo_gpu_offset(gds);
-			p->job->gds_size = amdgpu_bo_size(gds);
-		}
-		if (gws) {
-			p->job->gws_base = amdgpu_bo_gpu_offset(gws);
-			p->job->gws_size = amdgpu_bo_size(gws);
-		}
-		if (oa) {
-			p->job->oa_base = amdgpu_bo_gpu_offset(oa);
-			p->job->oa_size = amdgpu_bo_size(oa);
-		}
+	if (gds) {
+		p->job->gds_base = amdgpu_bo_gpu_offset(gds);
+		p->job->gds_size = amdgpu_bo_size(gds);
+	}
+	if (gws) {
+		p->job->gws_base = amdgpu_bo_gpu_offset(gws);
+		p->job->gws_size = amdgpu_bo_size(gws);
+	}
+	if (oa) {
+		p->job->oa_base = amdgpu_bo_gpu_offset(oa);
+		p->job->oa_size = amdgpu_bo_size(oa);
 	}
 
 	if (!r && p->uf_entry.robj) {

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -1730,6 +1730,18 @@ static int amdgpu_device_ip_late_set_cg_state(struct amdgpu_device *adev)
 			}
 		}
 	}
+
+	if (adev->powerplay.pp_feature & PP_GFXOFF_MASK) {
+		/* enable gfx powergating */
+		amdgpu_device_ip_set_powergating_state(adev,
+						       AMD_IP_BLOCK_TYPE_GFX,
+						       AMD_PG_STATE_GATE);
+		/* enable gfxoff */
+		amdgpu_device_ip_set_powergating_state(adev,
+						       AMD_IP_BLOCK_TYPE_SMC,
+						       AMD_PG_STATE_GATE);
+	}
+
 	return 0;
 }
 

drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c

@@ -30,6 +30,7 @@
 #include <drm/drmP.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
+#include "amdgpu_display.h"
 
 void amdgpu_gem_object_free(struct drm_gem_object *gobj)
 {
@@ -235,6 +236,13 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	/* create a gem object to contain this object in */
 	if (args->in.domains & (AMDGPU_GEM_DOMAIN_GDS |
 	    AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) {
+		if (flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) {
+			/* if gds bo is created from user space, it must be
+			 * passed to bo list
+			 */
+			DRM_ERROR("GDS bo cannot be per-vm-bo\n");
+			return -EINVAL;
+		}
 		flags |= AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
 		if (args->in.domains == AMDGPU_GEM_DOMAIN_GDS)
 			size = size << AMDGPU_GDS_SHIFT;
@@ -749,15 +757,16 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 	struct amdgpu_device *adev = dev->dev_private;
 	struct drm_gem_object *gobj;
 	uint32_t handle;
+	u32 domain;
 	int r;
 
 	args->pitch = amdgpu_align_pitch(adev, args->width,
 					 DIV_ROUND_UP(args->bpp, 8), 0);
 	args->size = (u64)args->pitch * args->height;
 	args->size = ALIGN(args->size, PAGE_SIZE);
-
-	r = amdgpu_gem_object_create(adev, args->size, 0,
-				     AMDGPU_GEM_DOMAIN_VRAM,
+	domain = amdgpu_bo_get_preferred_pin_domain(adev,
+				amdgpu_display_supported_domains(adev));
+	r = amdgpu_gem_object_create(adev, args->size, 0, domain,
 				     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED,
 				     false, NULL, &gobj);
 	if (r)

drivers/gpu/drm/amd/amdgpu/amdgpu_object.c

@@ -703,11 +703,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 	/* This assumes only APU display buffers are pinned with (VRAM|GTT).
 	 * See function amdgpu_display_supported_domains()
 	 */
-	if (domain == (AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT)) {
-		domain = AMDGPU_GEM_DOMAIN_VRAM;
-		if (adev->gmc.real_vram_size <= AMDGPU_SG_THRESHOLD)
-			domain = AMDGPU_GEM_DOMAIN_GTT;
-	}
+	domain = amdgpu_bo_get_preferred_pin_domain(adev, domain);
 
 	if (bo->pin_count) {
 		uint32_t mem_type = bo->tbo.mem.mem_type;
@@ -1066,3 +1062,14 @@ u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo)
 
 	return bo->tbo.offset;
 }
+
+uint32_t amdgpu_bo_get_preferred_pin_domain(struct amdgpu_device *adev,
+					    uint32_t domain)
+{
+	if (domain == (AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT)) {
+		domain = AMDGPU_GEM_DOMAIN_VRAM;
+		if (adev->gmc.real_vram_size <= AMDGPU_SG_THRESHOLD)
+			domain = AMDGPU_GEM_DOMAIN_GTT;
+	}
+	return domain;
+}

drivers/gpu/drm/amd/amdgpu/amdgpu_object.h

@@ -289,7 +289,8 @@ int amdgpu_bo_restore_from_shadow(struct amdgpu_device *adev,
 				  struct reservation_object *resv,
 				  struct dma_fence **fence,
 				  bool direct);
-
+uint32_t amdgpu_bo_get_preferred_pin_domain(struct amdgpu_device *adev,
+					    uint32_t domain);
 
 /*
  * sub allocation

drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c

@@ -49,8 +49,6 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work);
 
 int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring;
-	struct drm_sched_rq *rq;
 	unsigned long bo_size;
 	const char *fw_name;
 	const struct common_firmware_header *hdr;
@@ -84,6 +82,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	}
 
 	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
+	adev->vcn.fw_version = le32_to_cpu(hdr->ucode_version);
 	family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
 	version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
 	version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
@@ -102,24 +101,6 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 		return r;
 	}
 
-	ring = &adev->vcn.ring_dec;
-	rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL];
-	r = drm_sched_entity_init(&ring->sched, &adev->vcn.entity_dec,
-				  rq, NULL);
-	if (r != 0) {
-		DRM_ERROR("Failed setting up VCN dec run queue.\n");
-		return r;
-	}
-
-	ring = &adev->vcn.ring_enc[0];
-	rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL];
-	r = drm_sched_entity_init(&ring->sched, &adev->vcn.entity_enc,
-				  rq, NULL);
-	if (r != 0) {
-		DRM_ERROR("Failed setting up VCN enc run queue.\n");
-		return r;
-	}
-
 	return 0;
 }
 
@@ -129,10 +110,6 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev)
 
 	kfree(adev->vcn.saved_bo);
 
-	drm_sched_entity_fini(&adev->vcn.ring_dec.sched, &adev->vcn.entity_dec);
-
-	drm_sched_entity_fini(&adev->vcn.ring_enc[0].sched, &adev->vcn.entity_enc);
-
 	amdgpu_bo_free_kernel(&adev->vcn.vcpu_bo,
 			      &adev->vcn.gpu_addr,
 			      (void **)&adev->vcn.cpu_addr);
@@ -278,7 +255,7 @@ int amdgpu_vcn_dec_ring_test_ring(struct amdgpu_ring *ring)
 }
 
 static int amdgpu_vcn_dec_send_msg(struct amdgpu_ring *ring,
-				   struct amdgpu_bo *bo, bool direct,
+				   struct amdgpu_bo *bo,
 				   struct dma_fence **fence)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -306,19 +283,12 @@ static int amdgpu_vcn_dec_send_msg(struct amdgpu_ring *ring,
 	}
 	ib->length_dw = 16;
 
-	if (direct) {
-		r = amdgpu_ib_schedule(ring, 1, ib, NULL, &f);
-		job->fence = dma_fence_get(f);
-		if (r)
-			goto err_free;
+	r = amdgpu_ib_schedule(ring, 1, ib, NULL, &f);
+	job->fence = dma_fence_get(f);
+	if (r)
+		goto err_free;
 
-		amdgpu_job_free(job);
-	} else {
-		r = amdgpu_job_submit(job, ring, &adev->vcn.entity_dec,
-				      AMDGPU_FENCE_OWNER_UNDEFINED, &f);
-		if (r)
-			goto err_free;
-	}
+	amdgpu_job_free(job);
 
 	amdgpu_bo_fence(bo, f, false);
 	amdgpu_bo_unreserve(bo);
@@ -370,11 +340,11 @@ static int amdgpu_vcn_dec_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 	for (i = 14; i < 1024; ++i)
 		msg[i] = cpu_to_le32(0x0);
 
-	return amdgpu_vcn_dec_send_msg(ring, bo, true, fence);
+	return amdgpu_vcn_dec_send_msg(ring, bo, fence);
 }
 
 static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
-			       bool direct, struct dma_fence **fence)
+			       struct dma_fence **fence)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_bo *bo = NULL;
@@ -396,7 +366,7 @@ static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 	for (i = 6; i < 1024; ++i)
 		msg[i] = cpu_to_le32(0x0);
 
-	return amdgpu_vcn_dec_send_msg(ring, bo, direct, fence);
+	return amdgpu_vcn_dec_send_msg(ring, bo, fence);
 }
 
 int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
@@ -410,7 +380,7 @@ int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto error;
 	}
 
-	r = amdgpu_vcn_dec_get_destroy_msg(ring, 1, true, &fence);
+	r = amdgpu_vcn_dec_get_destroy_msg(ring, 1, &fence);
 	if (r) {
 		DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r);
 		goto error;

drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h

@@ -67,8 +67,6 @@ struct amdgpu_vcn {
 	struct amdgpu_ring	ring_dec;
 	struct amdgpu_ring	ring_enc[AMDGPU_VCN_MAX_ENC_RINGS];
 	struct amdgpu_irq_src	irq;
-	struct drm_sched_entity entity_dec;
-	struct drm_sched_entity entity_enc;
 	unsigned		num_enc_rings;
 };
 

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -2123,7 +2123,8 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
 			before->last = saddr - 1;
 			before->offset = tmp->offset;
 			before->flags = tmp->flags;
-			list_add(&before->list, &tmp->list);
+			before->bo_va = tmp->bo_va;
+			list_add(&before->list, &tmp->bo_va->invalids);
 		}
 
 		/* Remember mapping split at the end */
@@ -2133,7 +2134,8 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
 			after->offset = tmp->offset;
 			after->offset += after->start - tmp->start;
 			after->flags = tmp->flags;
-			list_add(&after->list, &tmp->list);
+			after->bo_va = tmp->bo_va;
+			list_add(&after->list, &tmp->bo_va->invalids);
 		}
 
 		list_del(&tmp->list);

drivers/gpu/drm/amd/amdgpu/df_v3_6.c

@@ -64,7 +64,7 @@ static u32 df_v3_6_get_hbm_channel_number(struct amdgpu_device *adev)
 	int fb_channel_number;
 
 	fb_channel_number = adev->df_funcs->get_fb_channel_number(adev);
-	if (fb_channel_number > ARRAY_SIZE(df_v3_6_channel_number))
+	if (fb_channel_number >= ARRAY_SIZE(df_v3_6_channel_number))
 		fb_channel_number = 0;
 
 	return df_v3_6_channel_number[fb_channel_number];

drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c

@@ -111,6 +111,7 @@ static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] =
 
 static const struct soc15_reg_golden golden_settings_gc_9_0_vg20[] =
 {
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_DCC_CONFIG, 0x0f000080, 0x04000080),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_2, 0x0f000000, 0x0a000000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_3, 0x30000000, 0x10000000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0xf3e777ff, 0x22014042),
@@ -1837,13 +1838,15 @@ static void gfx_v9_1_parse_ind_reg_list(int *register_list_format,
 				int indirect_offset,
 				int list_size,
 				int *unique_indirect_regs,
-				int *unique_indirect_reg_count,
+				int unique_indirect_reg_count,
 				int *indirect_start_offsets,
-				int *indirect_start_offsets_count)
+				int *indirect_start_offsets_count,
+				int max_start_offsets_count)
 {
 	int idx;
 
 	for (; indirect_offset < list_size; indirect_offset++) {
+		WARN_ON(*indirect_start_offsets_count >= max_start_offsets_count);
 		indirect_start_offsets[*indirect_start_offsets_count] = indirect_offset;
 		*indirect_start_offsets_count = *indirect_start_offsets_count + 1;
 
@@ -1851,14 +1854,14 @@ static void gfx_v9_1_parse_ind_reg_list(int *register_list_format,
 			indirect_offset += 2;
 
 			/* look for the matching indice */
-			for (idx = 0; idx < *unique_indirect_reg_count; idx++) {
+			for (idx = 0; idx < unique_indirect_reg_count; idx++) {
 				if (unique_indirect_regs[idx] ==
 					register_list_format[indirect_offset] ||
 					!unique_indirect_regs[idx])
 					break;
 			}
 
-			BUG_ON(idx >= *unique_indirect_reg_count);
+			BUG_ON(idx >= unique_indirect_reg_count);
 
 			if (!unique_indirect_regs[idx])
 				unique_indirect_regs[idx] = register_list_format[indirect_offset];
@@ -1893,9 +1896,10 @@ static int gfx_v9_1_init_rlc_save_restore_list(struct amdgpu_device *adev)
 				    adev->gfx.rlc.reg_list_format_direct_reg_list_length,
 				    adev->gfx.rlc.reg_list_format_size_bytes >> 2,
 				    unique_indirect_regs,
-				    &unique_indirect_reg_count,
+				    unique_indirect_reg_count,
 				    indirect_start_offsets,
-				    &indirect_start_offsets_count);
+				    &indirect_start_offsets_count,
+				    ARRAY_SIZE(indirect_start_offsets));
 
 	/* enable auto inc in case it is disabled */
 	tmp = RREG32(SOC15_REG_OFFSET(GC, 0, mmRLC_SRM_CNTL));
@@ -3404,11 +3408,6 @@ static int gfx_v9_0_late_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_GFX,
-						   AMD_PG_STATE_GATE);
-	if (r)
-		return r;
-
 	return 0;
 }
 

drivers/gpu/drm/amd/amdgpu/psp_v3_1.c

@@ -47,6 +47,8 @@ MODULE_FIRMWARE("amdgpu/vega20_asd.bin");
 
 #define smnMP1_FIRMWARE_FLAGS 0x3010028
 
+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)
 {
@@ -210,12 +212,31 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp)
 	return ret;
 }
 
+static bool psp_v3_1_match_version(struct amdgpu_device *adev, uint32_t ver)
+{
+	int i;
+
+	if (ver == adev->psp.sos_fw_version)
+		return true;
+
+	/*
+	 * Double check if the latest four legacy versions.
+	 * If yes, it is still the right version.
+	 */
+	for (i = 0; i < sizeof(sos_old_versions) / sizeof(uint32_t); i++) {
+		if (sos_old_versions[i] == adev->psp.sos_fw_version)
+			return true;
+	}
+
+	return false;
+}
+
 static int psp_v3_1_bootloader_load_sos(struct psp_context *psp)
 {
 	int ret;
 	unsigned int psp_gfxdrv_command_reg = 0;
 	struct amdgpu_device *adev = psp->adev;
-	uint32_t sol_reg;
+	uint32_t sol_reg, ver;
 
 	/* Check sOS sign of life register to confirm sys driver and sOS
 	 * are already been loaded.
@@ -248,6 +269,10 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp)
 			   RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81),
 			   0, true);
 
+	ver = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58);
+	if (!psp_v3_1_match_version(adev, ver))
+		DRM_WARN("SOS version doesn't match\n");
+
 	return ret;
 }
 

drivers/gpu/drm/amd/amdgpu/soc15.c

@@ -685,6 +685,7 @@ static int soc15_common_early_init(void *handle)
 			AMD_CG_SUPPORT_BIF_MGCG |
 			AMD_CG_SUPPORT_BIF_LS |
 			AMD_CG_SUPPORT_HDP_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
 			AMD_CG_SUPPORT_ROM_MGCG |
 			AMD_CG_SUPPORT_VCE_MGCG |
 			AMD_CG_SUPPORT_UVD_MGCG;

drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c

@@ -769,14 +769,14 @@ static int vcn_v1_0_stop(struct amdgpu_device *adev)
 	return 0;
 }
 
-bool vcn_v1_0_is_idle(void *handle)
+static bool vcn_v1_0_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	return (RREG32_SOC15(VCN, 0, mmUVD_STATUS) == 0x2);
 }
 
-int vcn_v1_0_wait_for_idle(void *handle)
+static int vcn_v1_0_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int ret = 0;

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

@@ -46,6 +46,7 @@
 #include <linux/moduleparam.h>
 #include <linux/version.h>
 #include <linux/types.h>
+#include <linux/pm_runtime.h>
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
@@ -2095,12 +2096,6 @@ convert_color_depth_from_display_info(const struct drm_connector *connector)
 {
 	uint32_t bpc = connector->display_info.bpc;
 
-	/* Limited color depth to 8bit
-	 * TODO: Still need to handle deep color
-	 */
-	if (bpc > 8)
-		bpc = 8;
-
 	switch (bpc) {
 	case 0:
 		/* Temporary Work around, DRM don't parse color depth for
@@ -2316,27 +2311,22 @@ decide_crtc_timing_for_drm_display_mode(struct drm_display_mode *drm_mode,
 	}
 }
 
-static int create_fake_sink(struct amdgpu_dm_connector *aconnector)
+static struct dc_sink *
+create_fake_sink(struct amdgpu_dm_connector *aconnector)
 {
-	struct dc_sink *sink = NULL;
 	struct dc_sink_init_data sink_init_data = { 0 };
-
+	struct dc_sink *sink = NULL;
 	sink_init_data.link = aconnector->dc_link;
 	sink_init_data.sink_signal = aconnector->dc_link->connector_signal;
 
 	sink = dc_sink_create(&sink_init_data);
 	if (!sink) {
 		DRM_ERROR("Failed to create sink!\n");
-		return -ENOMEM;
+		return NULL;
 	}
-
 	sink->sink_signal = SIGNAL_TYPE_VIRTUAL;
-	aconnector->fake_enable = true;
-
-	aconnector->dc_sink = sink;
-	aconnector->dc_link->local_sink = sink;
 
-	return 0;
+	return sink;
 }
 
 static void set_multisync_trigger_params(
@@ -2399,7 +2389,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 	struct dc_stream_state *stream = NULL;
 	struct drm_display_mode mode = *drm_mode;
 	bool native_mode_found = false;
-
+	struct dc_sink *sink = NULL;
 	if (aconnector == NULL) {
 		DRM_ERROR("aconnector is NULL!\n");
 		return stream;
@@ -2417,15 +2407,18 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 			return stream;
 		}
 
-		if (create_fake_sink(aconnector))
+		sink = create_fake_sink(aconnector);
+		if (!sink)
 			return stream;
+	} else {
+		sink = aconnector->dc_sink;
 	}
 
-	stream = dc_create_stream_for_sink(aconnector->dc_sink);
+	stream = dc_create_stream_for_sink(sink);
 
 	if (stream == NULL) {
 		DRM_ERROR("Failed to create stream for sink!\n");
-		return stream;
+		goto finish;
 	}
 
 	list_for_each_entry(preferred_mode, &aconnector->base.modes, head) {
@@ -2464,12 +2457,15 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 	fill_audio_info(
 		&stream->audio_info,
 		drm_connector,
-		aconnector->dc_sink);
+		sink);
 
 	update_stream_signal(stream);
 
 	if (dm_state && dm_state->freesync_capable)
 		stream->ignore_msa_timing_param = true;
+finish:
+	if (sink && sink->sink_signal == SIGNAL_TYPE_VIRTUAL)
+		dc_sink_release(sink);
 
 	return stream;
 }
@@ -2714,6 +2710,9 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector)
 	struct dm_connector_state *state =
 		to_dm_connector_state(connector->state);
 
+	if (connector->state)
+		__drm_atomic_helper_connector_destroy_state(connector->state);
+
 	kfree(state);
 
 	state = kzalloc(sizeof(*state), GFP_KERNEL);
@@ -2724,8 +2723,7 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector)
 		state->underscan_hborder = 0;
 		state->underscan_vborder = 0;
 
-		connector->state = &state->base;
-		connector->state->connector = connector;
+		__drm_atomic_helper_connector_reset(connector, &state->base);
 	}
 }
 
@@ -3083,17 +3081,6 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 		}
 	}
 
-	/* It's a hack for s3 since in 4.9 kernel filter out cursor buffer
-	 * prepare and cleanup in drm_atomic_helper_prepare_planes
-	 * and drm_atomic_helper_cleanup_planes because fb doens't in s3.
-	 * IN 4.10 kernel this code should be removed and amdgpu_device_suspend
-	 * code touching fram buffers should be avoided for DC.
-	 */
-	if (plane->type == DRM_PLANE_TYPE_CURSOR) {
-		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(new_state->crtc);
-
-		acrtc->cursor_bo = obj;
-	}
 	return 0;
 }
 
@@ -4281,6 +4268,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 			if (dm_old_crtc_state->stream)
 				remove_stream(adev, acrtc, dm_old_crtc_state->stream);
 
+			pm_runtime_get_noresume(dev->dev);
+
 			acrtc->enabled = true;
 			acrtc->hw_mode = new_crtc_state->mode;
 			crtc->hwmode = new_crtc_state->mode;
@@ -4469,6 +4458,16 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		drm_atomic_helper_wait_for_flip_done(dev, state);
 
 	drm_atomic_helper_cleanup_planes(dev, state);
+
+	/* Finally, drop a runtime PM reference for each newly disabled CRTC,
+	 * so we can put the GPU into runtime suspend if we're not driving any
+	 * displays anymore
+	 */
+	pm_runtime_mark_last_busy(dev->dev);
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		if (old_crtc_state->active && !new_crtc_state->active)
+			pm_runtime_put_autosuspend(dev->dev);
+	}
 }
 
 
@@ -4768,15 +4767,16 @@ static int dm_update_crtcs_state(struct dc *dc,
 		 * We want to do dc stream updates that do not require a
 		 * full modeset below.
 		 */
-		if (!enable || !aconnector || modereset_required(new_crtc_state))
+		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 attempting to enable a CRTC. Which has a...
-		 * 2. Valid connector attached, and
-		 * 3. User does not want to reset it (disable or mark inactive,
-		 *    which can happen on a CRTC that's already disabled).
-		 * => It currently exists.
+		 * 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);
 

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c

@@ -555,6 +555,9 @@ static inline int dm_irq_state(struct amdgpu_device *adev,
 		return 0;
 	}
 
+	if (acrtc->otg_inst == -1)
+		return 0;
+
 	irq_source = dal_irq_type + acrtc->otg_inst;
 
 	st = (state == AMDGPU_IRQ_STATE_ENABLE);

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c

@@ -234,6 +234,33 @@ static void pp_to_dc_clock_levels(
 	}
 }
 
+static void pp_to_dc_clock_levels_with_latency(
+		const struct pp_clock_levels_with_latency *pp_clks,
+		struct dm_pp_clock_levels_with_latency *clk_level_info,
+		enum dm_pp_clock_type dc_clk_type)
+{
+	uint32_t i;
+
+	if (pp_clks->num_levels > DM_PP_MAX_CLOCK_LEVELS) {
+		DRM_INFO("DM_PPLIB: Warning: %s clock: number of levels %d exceeds maximum of %d!\n",
+				DC_DECODE_PP_CLOCK_TYPE(dc_clk_type),
+				pp_clks->num_levels,
+				DM_PP_MAX_CLOCK_LEVELS);
+
+		clk_level_info->num_levels = DM_PP_MAX_CLOCK_LEVELS;
+	} else
+		clk_level_info->num_levels = pp_clks->num_levels;
+
+	DRM_DEBUG("DM_PPLIB: values for %s clock\n",
+			DC_DECODE_PP_CLOCK_TYPE(dc_clk_type));
+
+	for (i = 0; i < clk_level_info->num_levels; i++) {
+		DRM_DEBUG("DM_PPLIB:\t %d\n", pp_clks->data[i].clocks_in_khz);
+		clk_level_info->data[i].clocks_in_khz = pp_clks->data[i].clocks_in_khz;
+		clk_level_info->data[i].latency_in_us = pp_clks->data[i].latency_in_us;
+	}
+}
+
 bool dm_pp_get_clock_levels_by_type(
 		const struct dc_context *ctx,
 		enum dm_pp_clock_type clk_type,
@@ -311,8 +338,22 @@ bool dm_pp_get_clock_levels_by_type_with_latency(
 	enum dm_pp_clock_type clk_type,
 	struct dm_pp_clock_levels_with_latency *clk_level_info)
 {
-	/* TODO: to be implemented */
-	return false;
+	struct amdgpu_device *adev = ctx->driver_context;
+	void *pp_handle = adev->powerplay.pp_handle;
+	struct pp_clock_levels_with_latency pp_clks = { 0 };
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+
+	if (!pp_funcs || !pp_funcs->get_clock_by_type_with_latency)
+		return false;
+
+	if (pp_funcs->get_clock_by_type_with_latency(pp_handle,
+						     dc_to_pp_clock_type(clk_type),
+						     &pp_clks))
+		return false;
+
+	pp_to_dc_clock_levels_with_latency(&pp_clks, clk_level_info, clk_type);
+
+	return true;
 }
 
 bool dm_pp_get_clock_levels_by_type_with_voltage(

drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c

@@ -449,6 +449,11 @@ static inline unsigned int clamp_ux_dy(
 		return min_clamp;
 }
 
+unsigned int dc_fixpt_u3d19(struct fixed31_32 arg)
+{
+	return ux_dy(arg.value, 3, 19);
+}
+
 unsigned int dc_fixpt_u2d19(struct fixed31_32 arg)
 {
 	return ux_dy(arg.value, 2, 19);

drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c

@@ -1630,17 +1630,42 @@ static enum dc_status read_hpd_rx_irq_data(
 	struct dc_link *link,
 	union hpd_irq_data *irq_data)
 {
+	static enum dc_status retval;
+
 	/* The HW reads 16 bytes from 200h on HPD,
 	 * but if we get an AUX_DEFER, the HW cannot retry
 	 * and this causes the CTS tests 4.3.2.1 - 3.2.4 to
 	 * fail, so we now explicitly read 6 bytes which is
 	 * the req from the above mentioned test cases.
+	 *
+	 * For DP 1.4 we need to read those from 2002h range.
 	 */
-	return core_link_read_dpcd(
-	link,
-	DP_SINK_COUNT,
-	irq_data->raw,
-	sizeof(union hpd_irq_data));
+	if (link->dpcd_caps.dpcd_rev.raw < DPCD_REV_14)
+		retval = core_link_read_dpcd(
+			link,
+			DP_SINK_COUNT,
+			irq_data->raw,
+			sizeof(union hpd_irq_data));
+	else {
+		/* Read 2 bytes at this location,... */
+		retval = core_link_read_dpcd(
+			link,
+			DP_SINK_COUNT_ESI,
+			irq_data->raw,
+			2);
+
+		if (retval != DC_OK)
+			return retval;
+
+		/* ... then read remaining 4 at the other location */
+		retval = core_link_read_dpcd(
+			link,
+			DP_LANE0_1_STATUS_ESI,
+			&irq_data->raw[2],
+			4);
+	}
+
+	return retval;
 }
 
 static bool allow_hpd_rx_irq(const struct dc_link *link)
@@ -2278,7 +2303,7 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data,
 
 static bool retrieve_link_cap(struct dc_link *link)
 {
-	uint8_t dpcd_data[DP_TRAINING_AUX_RD_INTERVAL - DP_DPCD_REV + 1];
+	uint8_t dpcd_data[DP_ADAPTER_CAP - DP_DPCD_REV + 1];
 
 	union down_stream_port_count down_strm_port_count;
 	union edp_configuration_cap edp_config_cap;

drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c

@@ -72,7 +72,8 @@ static void dce110_update_generic_info_packet(
 	uint32_t max_retries = 50;
 
 	/*we need turn on clock before programming AFMT block*/
-	REG_UPDATE(AFMT_CNTL, AFMT_AUDIO_CLOCK_EN, 1);
+	if (REG(AFMT_CNTL))
+		REG_UPDATE(AFMT_CNTL, AFMT_AUDIO_CLOCK_EN, 1);
 
 	if (REG(AFMT_VBI_PACKET_CONTROL1)) {
 		if (packet_index >= 8)
@@ -719,7 +720,8 @@ static void dce110_stream_encoder_update_hdmi_info_packets(
 			const uint32_t *content =
 				(const uint32_t *) &info_frame->avi.sb[0];
 			/*we need turn on clock before programming AFMT block*/
-			REG_UPDATE(AFMT_CNTL, AFMT_AUDIO_CLOCK_EN, 1);
+			if (REG(AFMT_CNTL))
+				REG_UPDATE(AFMT_CNTL, AFMT_AUDIO_CLOCK_EN, 1);
 
 			REG_WRITE(AFMT_AVI_INFO0, content[0]);
 

drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c

@@ -121,10 +121,10 @@ static void reset_lb_on_vblank(struct dc_context *ctx)
 		frame_count = dm_read_reg(ctx, mmCRTC_STATUS_FRAME_COUNT);
 
 
-		for (retry = 100; retry > 0; retry--) {
+		for (retry = 10000; retry > 0; retry--) {
 			if (frame_count != dm_read_reg(ctx, mmCRTC_STATUS_FRAME_COUNT))
 				break;
-			msleep(1);
+			udelay(10);
 		}
 		if (!retry)
 			dm_error("Frame count did not increase for 100ms.\n");
@@ -147,14 +147,14 @@ static void wait_for_fbc_state_changed(
 	uint32_t addr = mmFBC_STATUS;
 	uint32_t value;
 
-	while (counter < 10) {
+	while (counter < 1000) {
 		value = dm_read_reg(cp110->base.ctx, addr);
 		if (get_reg_field_value(
 			value,
 			FBC_STATUS,
 			FBC_ENABLE_STATUS) == enabled)
 			break;
-		msleep(10);
+		udelay(100);
 		counter++;
 	}
 

drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c

@@ -1004,9 +1004,9 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option)
 		/*don't free audio if it is from retrain or internal disable stream*/
 		if (option == FREE_ACQUIRED_RESOURCE && dc->caps.dynamic_audio == true) {
 			/*we have to dynamic arbitrate the audio endpoints*/
-			pipe_ctx->stream_res.audio = NULL;
 			/*we free the resource, need reset is_audio_acquired*/
 			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false);
+			pipe_ctx->stream_res.audio = NULL;
 		}
 
 		/* TODO: notify audio driver for if audio modes list changed

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c

@@ -132,8 +132,7 @@ void dpp_set_gamut_remap_bypass(struct dcn10_dpp *dpp)
 
 #define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19))
 
-
-bool dpp_get_optimal_number_of_taps(
+static bool dpp_get_optimal_number_of_taps(
 		struct dpp *dpp,
 		struct scaler_data *scl_data,
 		const struct scaling_taps *in_taps)

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h

@@ -1424,12 +1424,8 @@ void dpp1_set_degamma(
 		enum ipp_degamma_mode mode);
 
 void dpp1_set_degamma_pwl(struct dpp *dpp_base,
-								 const struct pwl_params *params);
+		const struct pwl_params *params);
 
-bool dpp_get_optimal_number_of_taps(
-		struct dpp *dpp,
-		struct scaler_data *scl_data,
-		const struct scaling_taps *in_taps);
 
 void dpp_read_state(struct dpp *dpp_base,
 		struct dcn_dpp_state *s);

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c

@@ -565,16 +565,16 @@ static void dpp1_dscl_set_manual_ratio_init(
 	uint32_t init_int = 0;
 
 	REG_SET(SCL_HORZ_FILTER_SCALE_RATIO, 0,
-			SCL_H_SCALE_RATIO, dc_fixpt_u2d19(data->ratios.horz) << 5);
+			SCL_H_SCALE_RATIO, dc_fixpt_u3d19(data->ratios.horz) << 5);
 
 	REG_SET(SCL_VERT_FILTER_SCALE_RATIO, 0,
-			SCL_V_SCALE_RATIO, dc_fixpt_u2d19(data->ratios.vert) << 5);
+			SCL_V_SCALE_RATIO, dc_fixpt_u3d19(data->ratios.vert) << 5);
 
 	REG_SET(SCL_HORZ_FILTER_SCALE_RATIO_C, 0,
-			SCL_H_SCALE_RATIO_C, dc_fixpt_u2d19(data->ratios.horz_c) << 5);
+			SCL_H_SCALE_RATIO_C, dc_fixpt_u3d19(data->ratios.horz_c) << 5);
 
 	REG_SET(SCL_VERT_FILTER_SCALE_RATIO_C, 0,
-			SCL_V_SCALE_RATIO_C, dc_fixpt_u2d19(data->ratios.vert_c) << 5);
+			SCL_V_SCALE_RATIO_C, dc_fixpt_u3d19(data->ratios.vert_c) << 5);
 
 	/*
 	 * 0.24 format for fraction, first five bits zeroed

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c

@@ -396,11 +396,15 @@ bool hubp1_program_surface_flip_and_addr(
 		if (address->grph_stereo.right_addr.quad_part == 0)
 			break;
 
-		REG_UPDATE_4(DCSURF_SURFACE_CONTROL,
+		REG_UPDATE_8(DCSURF_SURFACE_CONTROL,
 				PRIMARY_SURFACE_TMZ, address->tmz_surface,
 				PRIMARY_SURFACE_TMZ_C, address->tmz_surface,
 				PRIMARY_META_SURFACE_TMZ, address->tmz_surface,
-				PRIMARY_META_SURFACE_TMZ_C, address->tmz_surface);
+				PRIMARY_META_SURFACE_TMZ_C, address->tmz_surface,
+				SECONDARY_SURFACE_TMZ, address->tmz_surface,
+				SECONDARY_SURFACE_TMZ_C, address->tmz_surface,
+				SECONDARY_META_SURFACE_TMZ, address->tmz_surface,
+				SECONDARY_META_SURFACE_TMZ_C, address->tmz_surface);
 
 		if (address->grph_stereo.right_meta_addr.quad_part != 0) {
 
@@ -459,9 +463,11 @@ void hubp1_dcc_control(struct hubp *hubp, bool enable,
 	uint32_t dcc_ind_64b_blk = independent_64b_blks ? 1 : 0;
 	struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp);
 
-	REG_UPDATE_2(DCSURF_SURFACE_CONTROL,
+	REG_UPDATE_4(DCSURF_SURFACE_CONTROL,
 			PRIMARY_SURFACE_DCC_EN, dcc_en,
-			PRIMARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk);
+			PRIMARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk,
+			SECONDARY_SURFACE_DCC_EN, dcc_en,
+			SECONDARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk);
 }
 
 void hubp1_program_surface_config(

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h

@@ -312,6 +312,12 @@
 	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, PRIMARY_META_SURFACE_TMZ_C, mask_sh),\
 	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, PRIMARY_SURFACE_DCC_EN, mask_sh),\
 	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, PRIMARY_SURFACE_DCC_IND_64B_BLK, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_SURFACE_TMZ, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_SURFACE_TMZ_C, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_META_SURFACE_TMZ, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_META_SURFACE_TMZ_C, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_SURFACE_DCC_EN, mask_sh),\
+	HUBP_SF(HUBPREQ0_DCSURF_SURFACE_CONTROL, SECONDARY_SURFACE_DCC_IND_64B_BLK, mask_sh),\
 	HUBP_SF(HUBPRET0_HUBPRET_CONTROL, DET_BUF_PLANE1_BASE_ADDRESS, mask_sh),\
 	HUBP_SF(HUBPRET0_HUBPRET_CONTROL, CROSSBAR_SRC_CB_B, mask_sh),\
 	HUBP_SF(HUBPRET0_HUBPRET_CONTROL, CROSSBAR_SRC_CR_R, mask_sh),\
@@ -489,6 +495,8 @@
 	type SECONDARY_META_SURFACE_TMZ_C;\
 	type PRIMARY_SURFACE_DCC_EN;\
 	type PRIMARY_SURFACE_DCC_IND_64B_BLK;\
+	type SECONDARY_SURFACE_DCC_EN;\
+	type SECONDARY_SURFACE_DCC_IND_64B_BLK;\
 	type DET_BUF_PLANE1_BASE_ADDRESS;\
 	type CROSSBAR_SRC_CB_B;\
 	type CROSSBAR_SRC_CR_R;\

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c

@@ -319,6 +319,10 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 		REG_UPDATE(DP_PIXEL_FORMAT, DP_COMPONENT_DEPTH,
 				DP_COMPONENT_PIXEL_DEPTH_12BPC);
 		break;
+	case COLOR_DEPTH_161616:
+		REG_UPDATE(DP_PIXEL_FORMAT, DP_COMPONENT_DEPTH,
+				DP_COMPONENT_PIXEL_DEPTH_16BPC);
+		break;
 	default:
 		REG_UPDATE(DP_PIXEL_FORMAT, DP_COMPONENT_DEPTH,
 				DP_COMPONENT_PIXEL_DEPTH_6BPC);

drivers/gpu/drm/amd/display/include/fixed31_32.h

@@ -496,6 +496,8 @@ static inline int dc_fixpt_ceil(struct fixed31_32 arg)
  * fractional
  */
 
+unsigned int dc_fixpt_u3d19(struct fixed31_32 arg);
+
 unsigned int dc_fixpt_u2d19(struct fixed31_32 arg);
 
 unsigned int dc_fixpt_u0d19(struct fixed31_32 arg);

drivers/gpu/drm/amd/include/asic_reg/df/df_3_6_sh_mask.h

@@ -36,13 +36,13 @@
 /* DF_CS_AON0_DramBaseAddress0 */
 #define DF_CS_UMC_AON0_DramBaseAddress0__AddrRngVal__SHIFT						0x0
 #define DF_CS_UMC_AON0_DramBaseAddress0__LgcyMmioHoleEn__SHIFT						0x1
-#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan__SHIFT						0x4
-#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel__SHIFT						0x8
+#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan__SHIFT						0x2
+#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel__SHIFT						0x9
 #define DF_CS_UMC_AON0_DramBaseAddress0__DramBaseAddr__SHIFT						0xc
 #define DF_CS_UMC_AON0_DramBaseAddress0__AddrRngVal_MASK						0x00000001L
 #define DF_CS_UMC_AON0_DramBaseAddress0__LgcyMmioHoleEn_MASK						0x00000002L
-#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan_MASK						0x000000F0L
-#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel_MASK						0x00000700L
+#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvNumChan_MASK						0x0000003CL
+#define DF_CS_UMC_AON0_DramBaseAddress0__IntLvAddrSel_MASK						0x00000E00L
 #define DF_CS_UMC_AON0_DramBaseAddress0__DramBaseAddr_MASK						0xFFFFF000L
 
 #endif

drivers/gpu/drm/amd/include/atomfirmware.h

@@ -2026,17 +2026,15 @@ enum atom_smu11_syspll_id {
   SMU11_SYSPLL3_1_ID          = 6,
 };
 
-
 enum atom_smu11_syspll0_clock_id {
-  SMU11_SYSPLL0_SOCCLK_ID   = 0,       //	SOCCLK
-  SMU11_SYSPLL0_MP0CLK_ID   = 1,       //	MP0CLK
-  SMU11_SYSPLL0_DCLK_ID     = 2,       //	DCLK
-  SMU11_SYSPLL0_VCLK_ID     = 3,       //	VCLK
-  SMU11_SYSPLL0_ECLK_ID     = 4,       //	ECLK
+  SMU11_SYSPLL0_ECLK_ID     = 0,       //	ECLK
+  SMU11_SYSPLL0_SOCCLK_ID   = 1,       //	SOCCLK
+  SMU11_SYSPLL0_MP0CLK_ID   = 2,       //	MP0CLK
+  SMU11_SYSPLL0_DCLK_ID     = 3,       //	DCLK
+  SMU11_SYSPLL0_VCLK_ID     = 4,       //	VCLK
   SMU11_SYSPLL0_DCEFCLK_ID  = 5,       //	DCEFCLK
 };
 
-
 enum atom_smu11_syspll1_0_clock_id {
   SMU11_SYSPLL1_0_UCLKA_ID   = 0,       // UCLK_a
 };

drivers/gpu/drm/amd/powerplay/amd_powerplay.c

@@ -180,7 +180,6 @@ static int pp_late_init(void *handle)
 {
 	struct amdgpu_device *adev = handle;
 	struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
-	int ret;
 
 	if (hwmgr && hwmgr->pm_en) {
 		mutex_lock(&hwmgr->smu_lock);
@@ -191,13 +190,6 @@ static int pp_late_init(void *handle)
 	if (adev->pm.smu_prv_buffer_size != 0)
 		pp_reserve_vram_for_smu(adev);
 
-	if (hwmgr->hwmgr_func->gfx_off_control &&
-	    (hwmgr->feature_mask & PP_GFXOFF_MASK)) {
-		ret = hwmgr->hwmgr_func->gfx_off_control(hwmgr, true);
-		if (ret)
-			pr_err("gfx off enabling failed!\n");
-	}
-
 	return 0;
 }
 
@@ -245,7 +237,7 @@ static int pp_set_powergating_state(void *handle,
 	}
 
 	if (hwmgr->hwmgr_func->enable_per_cu_power_gating == NULL) {
-		pr_info("%s was not implemented.\n", __func__);
+		pr_debug("%s was not implemented.\n", __func__);
 		return 0;
 	}
 

drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c

@@ -265,19 +265,18 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip,
 	if (skip)
 		return 0;
 
-	if (!hwmgr->ps)
-		/*
-		 * for vega12/vega20 which does not support power state manager
-		 * DAL clock limits should also be honoured
-		 */
-		phm_apply_clock_adjust_rules(hwmgr);
-
 	phm_pre_display_configuration_changed(hwmgr);
 
 	phm_display_configuration_changed(hwmgr);
 
 	if (hwmgr->ps)
 		power_state_management(hwmgr, new_ps);
+	else
+		/*
+		 * for vega12/vega20 which does not support power state manager
+		 * DAL clock limits should also be honoured
+		 */
+		phm_apply_clock_adjust_rules(hwmgr);
 
 	phm_notify_smc_display_config_after_ps_adjustment(hwmgr);
 

drivers/gpu/drm/amd/powerplay/hwmgr/ppatomfwctrl.c

@@ -496,7 +496,9 @@ int pp_atomfwctrl_get_clk_information_by_clkid(struct pp_hwmgr *hwmgr, BIOS_CLKI
 	uint32_t ix;
 
 	parameters.clk_id = id;
+	parameters.syspll_id = 0;
 	parameters.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	parameters.dfsdid = 0;
 
 	ix = GetIndexIntoMasterCmdTable(getsmuclockinfo);
 
@@ -505,7 +507,7 @@ int pp_atomfwctrl_get_clk_information_by_clkid(struct pp_hwmgr *hwmgr, BIOS_CLKI
 		return -EINVAL;
 
 	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&parameters;
-	*frequency = output->atom_smu_outputclkfreq.smu_clock_freq_hz / 10000;
+	*frequency = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
 
 	return 0;
 }

drivers/gpu/drm/amd/powerplay/hwmgr/process_pptables_v1_0.c

@@ -870,12 +870,6 @@ static int init_over_drive_limits(
 	hwmgr->platform_descriptor.maxOverdriveVDDC = 0;
 	hwmgr->platform_descriptor.overdriveVDDCStep = 0;
 
-	if (hwmgr->platform_descriptor.overdriveLimit.engineClock == 0 \
-		|| hwmgr->platform_descriptor.overdriveLimit.memoryClock == 0) {
-		hwmgr->od_enabled = false;
-		pr_debug("OverDrive feature not support by VBIOS\n");
-	}
-
 	return 0;
 }
 

drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c

@@ -1074,12 +1074,6 @@ static int init_overdrive_limits(struct pp_hwmgr *hwmgr,
 				powerplay_table,
 				(const ATOM_FIRMWARE_INFO_V2_1 *)fw_info);
 
-	if (hwmgr->platform_descriptor.overdriveLimit.engineClock == 0
-		&& hwmgr->platform_descriptor.overdriveLimit.memoryClock == 0) {
-		hwmgr->od_enabled = false;
-		pr_debug("OverDrive feature not support by VBIOS\n");
-	}
-
 	return result;
 }
 

drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c

@@ -53,8 +53,37 @@ static const unsigned long SMU10_Magic = (unsigned long) PHM_Rv_Magic;
 
 
 static int smu10_display_clock_voltage_request(struct pp_hwmgr *hwmgr,
-		struct pp_display_clock_request *clock_req);
+		struct pp_display_clock_request *clock_req)
+{
+	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
+	enum amd_pp_clock_type clk_type = clock_req->clock_type;
+	uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000;
+	PPSMC_Msg        msg;
 
+	switch (clk_type) {
+	case amd_pp_dcf_clock:
+		if (clk_freq == smu10_data->dcf_actual_hard_min_freq)
+			return 0;
+		msg =  PPSMC_MSG_SetHardMinDcefclkByFreq;
+		smu10_data->dcf_actual_hard_min_freq = clk_freq;
+		break;
+	case amd_pp_soc_clock:
+		 msg = PPSMC_MSG_SetHardMinSocclkByFreq;
+		break;
+	case amd_pp_f_clock:
+		if (clk_freq == smu10_data->f_actual_hard_min_freq)
+			return 0;
+		smu10_data->f_actual_hard_min_freq = clk_freq;
+		msg = PPSMC_MSG_SetHardMinFclkByFreq;
+		break;
+	default:
+		pr_info("[DisplayClockVoltageRequest]Invalid Clock Type!");
+		return -EINVAL;
+	}
+	smum_send_msg_to_smc_with_parameter(hwmgr, msg, clk_freq);
+
+	return 0;
+}
 
 static struct smu10_power_state *cast_smu10_ps(struct pp_hw_power_state *hw_ps)
 {
@@ -284,7 +313,7 @@ static int smu10_disable_gfx_off(struct pp_hwmgr *hwmgr)
 
 static int smu10_disable_dpm_tasks(struct pp_hwmgr *hwmgr)
 {
-	return smu10_disable_gfx_off(hwmgr);
+	return 0;
 }
 
 static int smu10_enable_gfx_off(struct pp_hwmgr *hwmgr)
@@ -299,7 +328,7 @@ static int smu10_enable_gfx_off(struct pp_hwmgr *hwmgr)
 
 static int smu10_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
 {
-	return smu10_enable_gfx_off(hwmgr);
+	return 0;
 }
 
 static int smu10_gfx_off_control(struct pp_hwmgr *hwmgr, bool enable)
@@ -1000,6 +1029,12 @@ static int smu10_get_clock_by_type_with_voltage(struct pp_hwmgr *hwmgr,
 	case amd_pp_soc_clock:
 		pclk_vol_table = pinfo->vdd_dep_on_socclk;
 		break;
+	case amd_pp_disp_clock:
+		pclk_vol_table = pinfo->vdd_dep_on_dispclk;
+		break;
+	case amd_pp_phy_clock:
+		pclk_vol_table = pinfo->vdd_dep_on_phyclk;
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -1017,39 +1052,7 @@ static int smu10_get_clock_by_type_with_voltage(struct pp_hwmgr *hwmgr,
 	return 0;
 }
 
-static int smu10_display_clock_voltage_request(struct pp_hwmgr *hwmgr,
-		struct pp_display_clock_request *clock_req)
-{
-	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
-	enum amd_pp_clock_type clk_type = clock_req->clock_type;
-	uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000;
-	PPSMC_Msg        msg;
 
-	switch (clk_type) {
-	case amd_pp_dcf_clock:
-		if (clk_freq == smu10_data->dcf_actual_hard_min_freq)
-			return 0;
-		msg =  PPSMC_MSG_SetHardMinDcefclkByFreq;
-		smu10_data->dcf_actual_hard_min_freq = clk_freq;
-		break;
-	case amd_pp_soc_clock:
-		 msg = PPSMC_MSG_SetHardMinSocclkByFreq;
-		break;
-	case amd_pp_f_clock:
-		if (clk_freq == smu10_data->f_actual_hard_min_freq)
-			return 0;
-		smu10_data->f_actual_hard_min_freq = clk_freq;
-		msg = PPSMC_MSG_SetHardMinFclkByFreq;
-		break;
-	default:
-		pr_info("[DisplayClockVoltageRequest]Invalid Clock Type!");
-		return -EINVAL;
-	}
-
-	smum_send_msg_to_smc_with_parameter(hwmgr, msg, clk_freq);
-
-	return 0;
-}
 
 static int smu10_get_max_high_clocks(struct pp_hwmgr *hwmgr, struct amd_pp_simple_clock_info *clocks)
 {
@@ -1182,6 +1185,7 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = {
 	.set_mmhub_powergating_by_smu = smu10_set_mmhub_powergating_by_smu,
 	.smus_notify_pwe = smu10_smus_notify_pwe,
 	.gfx_off_control = smu10_gfx_off_control,
+	.display_clock_voltage_request = smu10_display_clock_voltage_request,
 };
 
 int smu10_init_function_pointers(struct pp_hwmgr *hwmgr)

drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c

@@ -791,7 +791,8 @@ static int smu7_setup_dpm_tables_v1(struct pp_hwmgr *hwmgr)
 			data->dpm_table.sclk_table.count++;
 		}
 	}
-
+	if (hwmgr->platform_descriptor.overdriveLimit.engineClock == 0)
+		hwmgr->platform_descriptor.overdriveLimit.engineClock = dep_sclk_table->entries[i-1].clk;
 	/* Initialize Mclk DPM table based on allow Mclk values */
 	data->dpm_table.mclk_table.count = 0;
 	for (i = 0; i < dep_mclk_table->count; i++) {
@@ -806,6 +807,8 @@ static int smu7_setup_dpm_tables_v1(struct pp_hwmgr *hwmgr)
 		}
 	}
 
+	if (hwmgr->platform_descriptor.overdriveLimit.memoryClock == 0)
+		hwmgr->platform_descriptor.overdriveLimit.memoryClock = dep_mclk_table->entries[i-1].clk;
 	return 0;
 }
 
@@ -3752,14 +3755,17 @@ static int smu7_trim_dpm_states(struct pp_hwmgr *hwmgr,
 static int smu7_generate_dpm_level_enable_mask(
 		struct pp_hwmgr *hwmgr, const void *input)
 {
-	int result;
+	int result = 0;
 	const struct phm_set_power_state_input *states =
 			(const struct phm_set_power_state_input *)input;
 	struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend);
 	const struct smu7_power_state *smu7_ps =
 			cast_const_phw_smu7_power_state(states->pnew_state);
 
-	result = smu7_trim_dpm_states(hwmgr, smu7_ps);
+	/*skip the trim if od is enabled*/
+	if (!hwmgr->od_enabled)
+		result = smu7_trim_dpm_states(hwmgr, smu7_ps);
+
 	if (result)
 		return result;
 

drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c

@@ -321,8 +321,12 @@ static int vega10_odn_initial_default_setting(struct pp_hwmgr *hwmgr)
 		odn_table->min_vddc = dep_table[0]->entries[0].vddc;
 
 	i = od_table[2]->count - 1;
-	od_table[2]->entries[i].clk = hwmgr->platform_descriptor.overdriveLimit.memoryClock;
-	od_table[2]->entries[i].vddc = odn_table->max_vddc;
+	od_table[2]->entries[i].clk = hwmgr->platform_descriptor.overdriveLimit.memoryClock > od_table[2]->entries[i].clk ?
+					hwmgr->platform_descriptor.overdriveLimit.memoryClock :
+					od_table[2]->entries[i].clk;
+	od_table[2]->entries[i].vddc = odn_table->max_vddc > od_table[2]->entries[i].vddc ?
+					odn_table->max_vddc :
+					od_table[2]->entries[i].vddc;
 
 	return 0;
 }
@@ -1311,6 +1315,9 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr)
 	vega10_setup_default_single_dpm_table(hwmgr,
 			dpm_table,
 			dep_gfx_table);
+	if (hwmgr->platform_descriptor.overdriveLimit.engineClock == 0)
+		hwmgr->platform_descriptor.overdriveLimit.engineClock =
+					dpm_table->dpm_levels[dpm_table->count-1].value;
 	vega10_init_dpm_state(&(dpm_table->dpm_state));
 
 	/* Initialize Mclk DPM table based on allow Mclk values */
@@ -1319,6 +1326,10 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr)
 	vega10_setup_default_single_dpm_table(hwmgr,
 			dpm_table,
 			dep_mclk_table);
+	if (hwmgr->platform_descriptor.overdriveLimit.memoryClock == 0)
+		hwmgr->platform_descriptor.overdriveLimit.memoryClock =
+					dpm_table->dpm_levels[dpm_table->count-1].value;
+
 	vega10_init_dpm_state(&(dpm_table->dpm_state));
 
 	data->dpm_table.eclk_table.count = 0;

drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c

@@ -1104,7 +1104,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
 	for (count = 0; count < num_se; count++) {
 		data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | ( count << GRBM_GFX_INDEX__SE_INDEX__SHIFT);
 		WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, data);
-		result |= vega10_program_didt_config_registers(hwmgr, PSMSEEDCStallPatternConfig_Vega10, VEGA10_CONFIGREG_DIDT);
+		result = vega10_program_didt_config_registers(hwmgr, PSMSEEDCStallPatternConfig_Vega10, VEGA10_CONFIGREG_DIDT);
 		result |= vega10_program_didt_config_registers(hwmgr, PSMSEEDCStallDelayConfig_Vega10, VEGA10_CONFIGREG_DIDT);
 		result |= vega10_program_didt_config_registers(hwmgr, PSMSEEDCCtrlResetConfig_Vega10, VEGA10_CONFIGREG_DIDT);
 		result |= vega10_program_didt_config_registers(hwmgr, PSMSEEDCCtrlConfig_Vega10, VEGA10_CONFIGREG_DIDT);

drivers/gpu/drm/amd/powerplay/hwmgr/vega10_processpptables.c

@@ -267,12 +267,6 @@ static int init_over_drive_limits(
 	hwmgr->platform_descriptor.maxOverdriveVDDC = 0;
 	hwmgr->platform_descriptor.overdriveVDDCStep = 0;
 
-	if (hwmgr->platform_descriptor.overdriveLimit.engineClock == 0 ||
-		hwmgr->platform_descriptor.overdriveLimit.memoryClock == 0) {
-		hwmgr->od_enabled = false;
-		pr_debug("OverDrive feature not support by VBIOS\n");
-	}
-
 	return 0;
 }
 

drivers/gpu/drm/scheduler/gpu_scheduler.c

@@ -349,8 +349,13 @@ static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity)
 	struct dma_fence * fence = entity->dependency;
 	struct drm_sched_fence *s_fence;
 
-	if (fence->context == entity->fence_context) {
-		/* We can ignore fences from ourself */
+	if (fence->context == entity->fence_context ||
+            fence->context == entity->fence_context + 1) {
+                /*
+                 * Fence is a scheduled/finished fence from a job
+                 * which belongs to the same entity, we can ignore
+                 * fences from ourself
+                 */
 		dma_fence_put(entity->dependency);
 		return false;
 	}