Merge tag 'amd-drm-next-6.2-2022-11-25' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.2-2022-11-25: amdgpu: - Old GCC fix - GFX11 fixes - PSP suspend/resume fix - PCI ref count fix - DC KASAN fix - DCN 3.2.x fixes - Dell platform suspend/resume fixes - DCN 3.1.4 fixes - RAS fixes - SMU 13.x fixes - Flex array changes - VCN 4.0 RAS updates - Add missing licsense to some files - Documentation updates - SR-IOV fixes - DP MST DSC fix amdkfd: - Fix topology locking in error case Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221125180519.6389-1-alexander.deucher@amd.com
2022-11-30 15:58:56 +10:00
parent 02339a83dc 10d2d1fc05
commit f513a19a60
65 changed files with 896 additions and 216 deletions
@@ -327,7 +327,6 @@ static void amdgpu_connector_free_edid(struct drm_connector *connector)

 	kfree(amdgpu_connector->edid);
 	amdgpu_connector->edid = NULL;
-	drm_connector_update_edid_property(connector, NULL);
 }

 static int amdgpu_connector_ddc_get_modes(struct drm_connector *connector)
@@ -4181,21 +4181,15 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)

 	r = amdgpu_device_ip_resume(adev);

-	/* no matter what r is, always need to properly release full GPU */
-	if (amdgpu_sriov_vf(adev)) {
-		amdgpu_virt_init_data_exchange(adev);
-		amdgpu_virt_release_full_gpu(adev, true);
-	}
-
 	if (r) {
 		dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r);
-		return r;
+		goto exit;
 	}
 	amdgpu_fence_driver_hw_init(adev);

 	r = amdgpu_device_ip_late_init(adev);
 	if (r)
-		return r;
+		goto exit;

 	queue_delayed_work(system_wq, &adev->delayed_init_work,
 			   msecs_to_jiffies(AMDGPU_RESUME_MS));
@@ -4203,12 +4197,19 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	if (!adev->in_s0ix) {
 		r = amdgpu_amdkfd_resume(adev, adev->in_runpm);
 		if (r)
-			return r;
+			goto exit;
 	}

+exit:
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_init_data_exchange(adev);
+		amdgpu_virt_release_full_gpu(adev, true);
+	}
+
+	if (r)
+		return r;
+
 	/* Make sure IB tests flushed */
-	if (amdgpu_sriov_vf(adev))
-		amdgpu_irq_gpu_reset_resume_helper(adev);
 	flush_delayed_work(&adev->delayed_init_work);

 	if (adev->in_s0ix) {
@@ -5045,6 +5046,8 @@ static void amdgpu_device_resume_display_audio(struct amdgpu_device *adev)
 		pm_runtime_enable(&(p->dev));
 		pm_runtime_resume(&(p->dev));
 	}
+
+	pci_dev_put(p);
 }

 static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev)
@@ -5083,6 +5086,7 @@ static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev)

 		if (expires < ktime_get_mono_fast_ns()) {
 			dev_warn(adev->dev, "failed to suspend display audio\n");
+			pci_dev_put(p);
 			/* TODO: abort the succeeding gpu reset? */
 			return -ETIMEDOUT;
 		}
@@ -5090,6 +5094,7 @@ static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev)

 	pm_runtime_disable(&(p->dev));

+	pci_dev_put(p);
 	return 0;
 }

@@ -1702,11 +1702,13 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev)
 	return 0;
 }

+#if defined(CONFIG_DRM_AMD_DC)
 static void amdgpu_discovery_set_sriov_display(struct amdgpu_device *adev)
 {
 	amdgpu_device_set_sriov_virtual_display(adev);
 	amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block);
 }
+#endif

 static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev)
 {
@@ -238,8 +238,7 @@ module_param_named(gartsize, amdgpu_gart_size, uint, 0600);

 /**
 * DOC: gttsize (int)
- * Restrict the size of GTT domain in MiB for testing. The default is -1 (It's VRAM size if 3GB < VRAM < 3/4 RAM,
- * otherwise 3/4 RAM size).
+ * Restrict the size of GTT domain in MiB for testing. The default is -1 (Use 1/2 RAM, minimum value is 3GB).
 */
 MODULE_PARM_DESC(gttsize, "Size of the GTT domain in megabytes (-1 = auto)");
 module_param_named(gttsize, amdgpu_gtt_size, int, 0600);
@@ -479,6 +479,12 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
 	unsigned i;
 	unsigned vmhub, inv_eng;

+	if (adev->enable_mes) {
+		/* reserve engine 5 for firmware */
+		for (vmhub = 0; vmhub < AMDGPU_MAX_VMHUBS; vmhub++)
+			vm_inv_engs[vmhub] &= ~(1 << 5);
+	}
+
 	for (i = 0; i < adev->num_rings; ++i) {
 		ring = adev->rings[i];
 		vmhub = ring->funcs->vmhub;
@@ -174,7 +174,12 @@ static void amdgpu_job_free_cb(struct drm_sched_job *s_job)
 	drm_sched_job_cleanup(s_job);

 	amdgpu_sync_free(&job->explicit_sync);
-	dma_fence_put(&job->hw_fence);
+
+	/* only put the hw fence if has embedded fence */
+	if (!job->hw_fence.ops)
+		kfree(job);
+	else
+		dma_fence_put(&job->hw_fence);
 }

 void amdgpu_job_set_gang_leader(struct amdgpu_job *job,
@@ -198,6 +198,7 @@ void psp_ta_free_shared_buf(struct ta_mem_context *mem_ctx)
 {
 	amdgpu_bo_free_kernel(&mem_ctx->shared_bo, &mem_ctx->shared_mc_addr,
 			      &mem_ctx->shared_buf);
+	mem_ctx->shared_bo = NULL;
 }

 static void psp_free_shared_bufs(struct psp_context *psp)
@@ -208,6 +209,7 @@ static void psp_free_shared_bufs(struct psp_context *psp)
 	/* free TMR memory buffer */
 	pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
 	amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, pptr);
+	psp->tmr_bo = NULL;

 	/* free xgmi shared memory */
 	psp_ta_free_shared_buf(&psp->xgmi_context.context.mem_context);
@@ -769,7 +771,7 @@ static int psp_load_toc(struct psp_context *psp,
 /* Set up Trusted Memory Region */
 static int psp_tmr_init(struct psp_context *psp)
 {
-	int ret;
+	int ret = 0;
 	int tmr_size;
 	void *tmr_buf;
 	void **pptr;
@@ -796,10 +798,12 @@ static int psp_tmr_init(struct psp_context *psp)
 		}
 	}

-	pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
-	ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_ALIGNMENT,
-				      AMDGPU_GEM_DOMAIN_VRAM,
-				      &psp->tmr_bo, &psp->tmr_mc_addr, pptr);
+	if (!psp->tmr_bo) {
+		pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
+		ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_ALIGNMENT,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &psp->tmr_bo, &psp->tmr_mc_addr, pptr);
+	}

 	return ret;
 }
@@ -2727,8 +2731,6 @@ static int psp_suspend(void *handle)
 	}

 out:
-	psp_free_shared_bufs(psp);
-
 	return ret;
 }

@@ -32,7 +32,6 @@

 #define RB_ENABLED (1 << 0)
 #define RB4_ENABLED (1 << 1)
-#define MMSCH_DOORBELL_OFFSET 0x8

 #define MMSCH_VF_ENGINE_STATUS__PASS 0x1

@@ -770,6 +770,33 @@ static void vcn_v2_5_enable_clock_gating(struct amdgpu_device *adev)
 	}
 }

+static void vcn_v2_6_enable_ras(struct amdgpu_device *adev, int inst_idx,
+				bool indirect)
+{
+	uint32_t tmp;
+
+	if (adev->ip_versions[UVD_HWIP][0] != IP_VERSION(2, 6, 0))
+		return;
+
+	tmp = VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK;
+	WREG32_SOC15_DPG_MODE(inst_idx,
+			      SOC15_DPG_MODE_OFFSET(VCN, 0, mmVCN_RAS_CNTL),
+			      tmp, 0, indirect);
+
+	tmp = UVD_VCPU_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK;
+	WREG32_SOC15_DPG_MODE(inst_idx,
+			      SOC15_DPG_MODE_OFFSET(VCN, 0, mmUVD_VCPU_INT_EN),
+			      tmp, 0, indirect);
+
+	tmp = UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK;
+	WREG32_SOC15_DPG_MODE(inst_idx,
+			      SOC15_DPG_MODE_OFFSET(VCN, 0, mmUVD_SYS_INT_EN),
+			      tmp, 0, indirect);
+}
+
 static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect)
 {
 	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
@@ -849,6 +876,8 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, mmUVD_LMI_CTRL2), 0, 0, indirect);

+	vcn_v2_6_enable_ras(adev, inst_idx, indirect);
+
 	/* unblock VCPU register access */
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, mmUVD_RB_ARB_CTRL), 0, 0, indirect);
@@ -103,7 +103,6 @@ static int vcn_v4_0_sw_init(void *handle)
 	struct amdgpu_ring *ring;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int i, r;
-	int vcn_doorbell_index = 0;

 	r = amdgpu_vcn_sw_init(adev);
 	if (r)
@@ -115,12 +114,6 @@ static int vcn_v4_0_sw_init(void *handle)
 	if (r)
 		return r;

-	if (amdgpu_sriov_vf(adev)) {
-		vcn_doorbell_index = adev->doorbell_index.vcn.vcn_ring0_1 - MMSCH_DOORBELL_OFFSET;
-		/* get DWORD offset */
-		vcn_doorbell_index = vcn_doorbell_index << 1;
-	}
-
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		volatile struct amdgpu_vcn4_fw_shared *fw_shared;

@@ -144,7 +137,7 @@ static int vcn_v4_0_sw_init(void *handle)
 		ring = &adev->vcn.inst[i].ring_enc[0];
 		ring->use_doorbell = true;
 		if (amdgpu_sriov_vf(adev))
-			ring->doorbell_index = vcn_doorbell_index + i * (adev->vcn.num_enc_rings + 1) + 1;
+			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + i * (adev->vcn.num_enc_rings + 1) + 1;
 		else
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + 8 * i;

@@ -1805,21 +1805,75 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct
 	pr_debug("Added [%d] GPU cache entries\n", num_of_entries);
 }

+static int kfd_topology_add_device_locked(struct kfd_dev *gpu, uint32_t gpu_id,
+					  struct kfd_topology_device **dev)
+{
+	int proximity_domain = ++topology_crat_proximity_domain;
+	struct list_head temp_topology_device_list;
+	void *crat_image = NULL;
+	size_t image_size = 0;
+	int res;
+
+	res = kfd_create_crat_image_virtual(&crat_image, &image_size,
+					    COMPUTE_UNIT_GPU, gpu,
+					    proximity_domain);
+	if (res) {
+		pr_err("Error creating VCRAT for GPU (ID: 0x%x)\n",
+		       gpu_id);
+		topology_crat_proximity_domain--;
+		goto err;
+	}
+
+	INIT_LIST_HEAD(&temp_topology_device_list);
+
+	res = kfd_parse_crat_table(crat_image,
+				   &temp_topology_device_list,
+				   proximity_domain);
+	if (res) {
+		pr_err("Error parsing VCRAT for GPU (ID: 0x%x)\n",
+		       gpu_id);
+		topology_crat_proximity_domain--;
+		goto err;
+	}
+
+	kfd_topology_update_device_list(&temp_topology_device_list,
+					&topology_device_list);
+
+	*dev = kfd_assign_gpu(gpu);
+	if (WARN_ON(!*dev)) {
+		res = -ENODEV;
+		goto err;
+	}
+
+	/* Fill the cache affinity information here for the GPUs
+	 * using VCRAT
+	 */
+	kfd_fill_cache_non_crat_info(*dev, gpu);
+
+	/* Update the SYSFS tree, since we added another topology
+	 * device
+	 */
+	res = kfd_topology_update_sysfs();
+	if (!res)
+		sys_props.generation_count++;
+	else
+		pr_err("Failed to update GPU (ID: 0x%x) to sysfs topology. res=%d\n",
+		       gpu_id, res);
+
+err:
+	kfd_destroy_crat_image(crat_image);
+	return res;
+}
+
 int kfd_topology_add_device(struct kfd_dev *gpu)
 {
 	uint32_t gpu_id;
 	struct kfd_topology_device *dev;
 	struct kfd_cu_info cu_info;
 	int res = 0;
-	struct list_head temp_topology_device_list;
-	void *crat_image = NULL;
-	size_t image_size = 0;
-	int proximity_domain;
 	int i;
 	const char *asic_name = amdgpu_asic_name[gpu->adev->asic_type];

-	INIT_LIST_HEAD(&temp_topology_device_list);
-
 	gpu_id = kfd_generate_gpu_id(gpu);
 	pr_debug("Adding new GPU (ID: 0x%x) to topology\n", gpu_id);

@@ -1831,54 +1885,11 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	 */
 	down_write(&topology_lock);
 	dev = kfd_assign_gpu(gpu);
-	if (!dev) {
-		proximity_domain = ++topology_crat_proximity_domain;
-
-		res = kfd_create_crat_image_virtual(&crat_image, &image_size,
-						    COMPUTE_UNIT_GPU, gpu,
-						    proximity_domain);
-		if (res) {
-			pr_err("Error creating VCRAT for GPU (ID: 0x%x)\n",
-			       gpu_id);
-			topology_crat_proximity_domain--;
-			return res;
-		}
-
-		res = kfd_parse_crat_table(crat_image,
-					   &temp_topology_device_list,
-					   proximity_domain);
-		if (res) {
-			pr_err("Error parsing VCRAT for GPU (ID: 0x%x)\n",
-			       gpu_id);
-			topology_crat_proximity_domain--;
-			goto err;
-		}
-
-		kfd_topology_update_device_list(&temp_topology_device_list,
-			&topology_device_list);
-
-		dev = kfd_assign_gpu(gpu);
-		if (WARN_ON(!dev)) {
-			res = -ENODEV;
-			goto err;
-		}
-
-		/* Fill the cache affinity information here for the GPUs
-		 * using VCRAT
-		 */
-		kfd_fill_cache_non_crat_info(dev, gpu);
-
-		/* Update the SYSFS tree, since we added another topology
-		 * device
-		 */
-		res = kfd_topology_update_sysfs();
-		if (!res)
-			sys_props.generation_count++;
-		else
-			pr_err("Failed to update GPU (ID: 0x%x) to sysfs topology. res=%d\n",
-						gpu_id, res);
-	}
+	if (!dev)
+		res = kfd_topology_add_device_locked(gpu, gpu_id, &dev);
 	up_write(&topology_lock);
+	if (res)
+		return res;

 	dev->gpu_id = gpu_id;
 	gpu->id = gpu_id;
@@ -2003,8 +2014,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)

 	if (!res)
 		kfd_notify_gpu_change(gpu_id, 1);
-err:
-	kfd_destroy_crat_image(crat_image);
+
 	return res;
 }

@@ -1371,7 +1371,44 @@ static const struct dmi_system_id hpd_disconnect_quirk_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3460"),
 		},
 	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower Plus 7010"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower 7010"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF Plus 7010"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF 7010"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro Plus 7010"),
+		},
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro 7010"),
+		},
+	},
 	{}
+	/* TODO: refactor this from a fixed table to a dynamic option */
 };

 static void retrieve_dmi_info(struct amdgpu_display_manager *dm)
@@ -1187,7 +1187,7 @@ static int pre_compute_mst_dsc_configs_for_state(struct drm_atomic_state *state,
 	struct amdgpu_dm_connector *aconnector;
 	struct drm_dp_mst_topology_mgr *mst_mgr;
 	int link_vars_start_index = 0;
-	int ret;
+	int ret = 0;

 	for (i = 0; i < dc_state->stream_count; i++)
 		computed_streams[i] = false;
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 // This is a stripped-down version of the smu11_driver_if.h file for the relevant DAL interfaces.

 #define SMU11_DRIVER_IF_VERSION 0x40
@@ -333,8 +333,8 @@ void dcn31_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zst
 			(support == DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY))
 		support = DCN_ZSTATE_SUPPORT_DISALLOW;

-
-	if (support == DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY)
+	if (support == DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY ||
+	    support == DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY)
 		param = 1;
 	else
 		param = 0;
@@ -346,8 +346,6 @@ void dcn314_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zs
 	if (!clk_mgr->smu_present)
 		return;

-	// Arg[15:0] = 8/9/0 for Z8/Z9/disallow -> existing bits
-	// Arg[16] = Disallow Z9 -> new bit
 	switch (support) {

 	case DCN_ZSTATE_SUPPORT_ALLOW:
@@ -366,6 +364,16 @@ void dcn314_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zs
 		param = (1 << 10);
 		break;

+	case DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY:
+		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
+		param = (1 << 10) | (1 << 8);
+		break;
+
+	case DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY:
+		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
+		param = (1 << 8);
+		break;
+
 	default: //DCN_ZSTATE_SUPPORT_UNKNOWN
 		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
 		param = 0;
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 // This is a stripped-down version of the smu13_driver_if.h file for the relevant DAL interfaces.

 #define SMU13_DRIVER_IF_VERSION  0x18
@@ -1192,7 +1192,7 @@ static void disable_vbios_mode_if_required(

 					if (pix_clk_100hz != requested_pix_clk_100hz) {
 						core_link_disable_stream(pipe);
-						pipe->stream->dpms_off = true;
+						pipe->stream->dpms_off = false;
 					}
 				}
 			}
@@ -3098,6 +3098,19 @@ static bool update_planes_and_stream_state(struct dc *dc,

 	if (update_type == UPDATE_TYPE_FULL) {
 		if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) {
+			/* For phantom pipes we remove and create a new set of phantom pipes
+			 * for each full update (because we don't know if we'll need phantom
+			 * pipes until after the first round of validation). However, if validation
+			 * fails we need to keep the existing phantom pipes (because we don't update
+			 * the dc->current_state).
+			 *
+			 * The phantom stream/plane refcount is decremented for validation because
+			 * we assume it'll be removed (the free comes when the dc_state is freed),
+			 * but if validation fails we have to increment back the refcount so it's
+			 * consistent.
+			 */
+			if (dc->res_pool->funcs->retain_phantom_pipes)
+				dc->res_pool->funcs->retain_phantom_pipes(dc, dc->current_state);
 			BREAK_TO_DEBUGGER();
 			goto fail;
 		}
@@ -3740,6 +3753,8 @@ static bool could_mpcc_tree_change_for_active_pipes(struct dc *dc,

 	struct dc_stream_status *cur_stream_status = stream_get_status(dc->current_state, stream);
 	bool force_minimal_pipe_splitting = false;
+	bool subvp_active = false;
+	uint32_t i;

 	*is_plane_addition = false;

@@ -3771,11 +3786,25 @@ static bool could_mpcc_tree_change_for_active_pipes(struct dc *dc,
 		}
 	}

+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_NONE) {
+			subvp_active = true;
+			break;
+		}
+	}
+
 	/* For SubVP when adding or removing planes we need to add a minimal transition
 	 * (even when disabling all planes). Whenever disabling a phantom pipe, we
 	 * must use the minimal transition path to disable the pipe correctly.
+	 *
+	 * We want to use the minimal transition whenever subvp is active, not only if
+	 * a plane is being added / removed from a subvp stream (MPO plane can be added
+	 * to a DRR pipe of SubVP + DRR config, in which case we still want to run through
+	 * a min transition to disable subvp.
 	 */
-	if (cur_stream_status && stream->mall_stream_config.type == SUBVP_MAIN) {
+	if (cur_stream_status && subvp_active) {
 		/* determine if minimal transition is required due to SubVP*/
 		if (cur_stream_status->plane_count > surface_count) {
 			force_minimal_pipe_splitting = true;
@@ -332,9 +332,21 @@ bool dc_stream_set_cursor_attributes(

 	dc = stream->ctx->dc;

-	if (dc->debug.allow_sw_cursor_fallback && attributes->height * attributes->width * 4 > 16384)
-		if (stream->mall_stream_config.type == SUBVP_MAIN)
+	/* SubVP is not compatible with HW cursor larger than 64 x 64 x 4.
+	 * Therefore, if cursor is greater than 64 x 64 x 4, fallback to SW cursor in the following case:
+	 * 1. For single display cases, if resolution is >= 5K and refresh rate < 120hz
+	 * 2. For multi display cases, if resolution is >= 4K and refresh rate < 120hz
+	 *
+	 * [< 120hz is a requirement for SubVP configs]
+	 */
+	if (dc->debug.allow_sw_cursor_fallback && attributes->height * attributes->width * 4 > 16384) {
+		if (dc->current_state->stream_count == 1 && stream->timing.v_addressable >= 2880 &&
+				((stream->timing.pix_clk_100hz * 100) / stream->timing.v_total / stream->timing.h_total) < 120)
 			return false;
+		else if (dc->current_state->stream_count > 1 && stream->timing.v_addressable >= 2160 &&
+				((stream->timing.pix_clk_100hz * 100) / stream->timing.v_total / stream->timing.h_total) < 120)
+			return false;
+	}

 	stream->cursor_attributes = *attributes;

@@ -47,7 +47,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;

-#define DC_VER "3.2.212"
+#define DC_VER "3.2.213"

 #define MAX_SURFACES 3
 #define MAX_PLANES 6
@@ -267,6 +267,7 @@ struct dc_caps {
 	uint16_t subvp_pstate_allow_width_us;
 	uint16_t subvp_vertical_int_margin_us;
 	bool seamless_odm;
+	uint8_t subvp_drr_vblank_start_margin_us;
 };

 struct dc_bug_wa {
@@ -492,6 +493,8 @@ enum dcn_pwr_state {
 enum dcn_zstate_support_state {
 	DCN_ZSTATE_SUPPORT_UNKNOWN,
 	DCN_ZSTATE_SUPPORT_ALLOW,
+	DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY,
+	DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY,
 	DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY,
 	DCN_ZSTATE_SUPPORT_DISALLOW,
 };
@@ -849,6 +852,7 @@ struct dc_debug_options {
 	unsigned int force_subvp_num_ways;
 	unsigned int force_mall_ss_num_ways;
 	bool alloc_extra_way_for_cursor;
+	uint32_t subvp_extra_lines;
 	bool force_usr_allow;
 	/* uses value at boot and disables switch */
 	bool disable_dtb_ref_clk_switch;
@@ -493,6 +493,7 @@ static void populate_subvp_cmd_drr_info(struct dc *dc,

 	pipe_data->pipe_config.vblank_data.drr_info.min_vtotal_supported = min_vtotal_supported;
 	pipe_data->pipe_config.vblank_data.drr_info.max_vtotal_supported = max_vtotal_supported;
+	pipe_data->pipe_config.vblank_data.drr_info.drr_vblank_start_margin = dc->caps.subvp_drr_vblank_start_margin_us;
 }

 /**
@@ -151,6 +151,20 @@ struct dc_panel_config {
 		bool optimize_edp_link_rate; /* eDP ILR */
 	} ilr;
 };
+
+/*
+ *  USB4 DPIA BW ALLOCATION STRUCTS
+ */
+struct dc_dpia_bw_alloc {
+	int sink_verified_bw;  // The Verified BW that sink can allocated and use that has been verified already
+	int sink_allocated_bw; // The Actual Allocated BW that sink currently allocated
+	int padding_bw;        // The Padding "Un-used" BW allocated by CM for padding reasons
+	int sink_max_bw;       // The Max BW that sink can require/support
+	int estimated_bw;      // The estimated available BW for this DPIA
+	int bw_granularity;    // BW Granularity
+	bool bw_alloc_enabled; // The BW Alloc Mode Support is turned ON for all 3:  DP-Tx & Dpia & CM
+};
+
 /*
 * A link contains one or more sinks and their connected status.
 * The currently active signal type (HDMI, DP-SST, DP-MST) is also reported.
@@ -144,7 +144,7 @@ struct test_pattern {
 	unsigned int cust_pattern_size;
 };

-#define SUBVP_DRR_MARGIN_US 500 // 500us for DRR margin (SubVP + DRR)
+#define SUBVP_DRR_MARGIN_US 600 // 600us for DRR margin (SubVP + DRR)

 enum mall_stream_type {
 	SUBVP_NONE, // subvp not in use
@@ -1561,8 +1561,13 @@ static enum dc_status apply_single_controller_ctx_to_hw(

 	pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != NULL;

-	pipe_ctx->stream->link->psr_settings.psr_feature_enabled = false;
-
+	/* Phantom and main stream share the same link (because the stream
+	 * is constructed with the same sink). Make sure not to override
+	 * and link programming on the main.
+	 */
+	if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) {
+		pipe_ctx->stream->link->psr_settings.psr_feature_enabled = false;
+	}
 	return DC_OK;
 }

@@ -359,7 +359,8 @@ static const struct dce_audio_registers audio_regs[] = {
 	audio_regs(2),
 	audio_regs(3),
 	audio_regs(4),
-	audio_regs(5)
+	audio_regs(5),
+	audio_regs(6),
 };

 #define DCE120_AUD_COMMON_MASK_SH_LIST(mask_sh)\
@@ -436,34 +436,48 @@ void dpp1_set_cursor_position(
 		uint32_t height)
 {
 	struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base);
-	int src_x_offset = pos->x - pos->x_hotspot - param->viewport.x;
-	int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y;
+	int x_pos = pos->x - param->viewport.x;
+	int y_pos = pos->y - param->viewport.y;
+	int x_hotspot = pos->x_hotspot;
+	int y_hotspot = pos->y_hotspot;
+	int src_x_offset = x_pos - pos->x_hotspot;
+	int src_y_offset = y_pos - pos->y_hotspot;
+	int cursor_height = (int)height;
+	int cursor_width = (int)width;
 	uint32_t cur_en = pos->enable ? 1 : 0;

-	// Cursor width/height and hotspots need to be rotated for offset calculation
+	// Transform cursor width / height and hotspots for offset calculations
 	if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) {
-		swap(width, height);
+		swap(cursor_height, cursor_width);
+		swap(x_hotspot, y_hotspot);
+
 		if (param->rotation == ROTATION_ANGLE_90) {
-			src_x_offset = pos->x - pos->y_hotspot - param->viewport.x;
-			src_y_offset = pos->y - pos->x_hotspot - param->viewport.y;
+			// hotspot = (-y, x)
+			src_x_offset = x_pos - (cursor_width - x_hotspot);
+			src_y_offset = y_pos - y_hotspot;
+		} else if (param->rotation == ROTATION_ANGLE_270) {
+			// hotspot = (y, -x)
+			src_x_offset = x_pos - x_hotspot;
+			src_y_offset = y_pos - (cursor_height - y_hotspot);
 		}
 	} else if (param->rotation == ROTATION_ANGLE_180) {
+		// hotspot = (-x, -y)
 		if (!param->mirror)
-			src_x_offset = pos->x - param->viewport.x;
+			src_x_offset = x_pos - (cursor_width - x_hotspot);

-		src_y_offset = pos->y - param->viewport.y;
+		src_y_offset = y_pos - (cursor_height - y_hotspot);
 	}

 	if (src_x_offset >= (int)param->viewport.width)
 		cur_en = 0;  /* not visible beyond right edge*/

-	if (src_x_offset + (int)width <= 0)
+	if (src_x_offset + cursor_width <= 0)
 		cur_en = 0;  /* not visible beyond left edge*/

 	if (src_y_offset >= (int)param->viewport.height)
 		cur_en = 0;  /* not visible beyond bottom edge*/

-	if (src_y_offset + (int)height <= 0)
+	if (src_y_offset + cursor_height <= 0)
 		cur_en = 0;  /* not visible beyond top edge*/

 	REG_UPDATE(CURSOR0_CONTROL,
@@ -1179,10 +1179,12 @@ void hubp1_cursor_set_position(
 		const struct dc_cursor_mi_param *param)
 {
 	struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp);
-	int src_x_offset = pos->x - pos->x_hotspot - param->viewport.x;
-	int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y;
+	int x_pos = pos->x - param->viewport.x;
+	int y_pos = pos->y - param->viewport.y;
 	int x_hotspot = pos->x_hotspot;
 	int y_hotspot = pos->y_hotspot;
+	int src_x_offset = x_pos - pos->x_hotspot;
+	int src_y_offset = y_pos - pos->y_hotspot;
 	int cursor_height = (int)hubp->curs_attr.height;
 	int cursor_width = (int)hubp->curs_attr.width;
 	uint32_t dst_x_offset;
@@ -1200,18 +1202,26 @@ void hubp1_cursor_set_position(
 	if (hubp->curs_attr.address.quad_part == 0)
 		return;

-	// Rotated cursor width/height and hotspots tweaks for offset calculation
+	// Transform cursor width / height and hotspots for offset calculations
 	if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) {
 		swap(cursor_height, cursor_width);
+		swap(x_hotspot, y_hotspot);
+
 		if (param->rotation == ROTATION_ANGLE_90) {
-			src_x_offset = pos->x - pos->y_hotspot - param->viewport.x;
-			src_y_offset = pos->y - pos->x_hotspot - param->viewport.y;
+			// hotspot = (-y, x)
+			src_x_offset = x_pos - (cursor_width - x_hotspot);
+			src_y_offset = y_pos - y_hotspot;
+		} else if (param->rotation == ROTATION_ANGLE_270) {
+			// hotspot = (y, -x)
+			src_x_offset = x_pos - x_hotspot;
+			src_y_offset = y_pos - (cursor_height - y_hotspot);
 		}
 	} else if (param->rotation == ROTATION_ANGLE_180) {
+		// hotspot = (-x, -y)
 		if (!param->mirror)
-			src_x_offset = pos->x - param->viewport.x;
+			src_x_offset = x_pos - (cursor_width - x_hotspot);

-		src_y_offset = pos->y - param->viewport.y;
+		src_y_offset = y_pos - (cursor_height - y_hotspot);
 	}

 	dst_x_offset = (src_x_offset >= 0) ? src_x_offset : 0;
@@ -1248,8 +1258,8 @@ void hubp1_cursor_set_position(
 			CURSOR_Y_POSITION, pos->y);

 	REG_SET_2(CURSOR_HOT_SPOT, 0,
-			CURSOR_HOT_SPOT_X, x_hotspot,
-			CURSOR_HOT_SPOT_Y, y_hotspot);
+			CURSOR_HOT_SPOT_X, pos->x_hotspot,
+			CURSOR_HOT_SPOT_Y, pos->y_hotspot);

 	REG_SET(CURSOR_DST_OFFSET, 0,
 			CURSOR_DST_X_OFFSET, dst_x_offset);
@@ -2252,6 +2252,12 @@ void dcn10_enable_timing_synchronization(
 		opp = grouped_pipes[i]->stream_res.opp;
 		tg = grouped_pipes[i]->stream_res.tg;
 		tg->funcs->get_otg_active_size(tg, &width, &height);
+
+		if (!tg->funcs->is_tg_enabled(tg)) {
+			DC_SYNC_INFO("Skipping timing sync on disabled OTG\n");
+			return;
+		}
+
 		if (opp->funcs->opp_program_dpg_dimensions)
 			opp->funcs->opp_program_dpg_dimensions(opp, width, 2*(height) + 1);
 	}
@@ -973,10 +973,12 @@ void hubp2_cursor_set_position(
 		const struct dc_cursor_mi_param *param)
 {
 	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
-	int src_x_offset = pos->x - pos->x_hotspot - param->viewport.x;
-	int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y;
+	int x_pos = pos->x - param->viewport.x;
+	int y_pos = pos->y - param->viewport.y;
 	int x_hotspot = pos->x_hotspot;
 	int y_hotspot = pos->y_hotspot;
+	int src_x_offset = x_pos - pos->x_hotspot;
+	int src_y_offset = y_pos - pos->y_hotspot;
 	int cursor_height = (int)hubp->curs_attr.height;
 	int cursor_width = (int)hubp->curs_attr.width;
 	uint32_t dst_x_offset;
@@ -994,18 +996,26 @@ void hubp2_cursor_set_position(
 	if (hubp->curs_attr.address.quad_part == 0)
 		return;

-	// Rotated cursor width/height and hotspots tweaks for offset calculation
+	// Transform cursor width / height and hotspots for offset calculations
 	if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) {
 		swap(cursor_height, cursor_width);
+		swap(x_hotspot, y_hotspot);
+
 		if (param->rotation == ROTATION_ANGLE_90) {
-			src_x_offset = pos->x - pos->y_hotspot - param->viewport.x;
-			src_y_offset = pos->y - pos->x_hotspot - param->viewport.y;
+			// hotspot = (-y, x)
+			src_x_offset = x_pos - (cursor_width - x_hotspot);
+			src_y_offset = y_pos - y_hotspot;
+		} else if (param->rotation == ROTATION_ANGLE_270) {
+			// hotspot = (y, -x)
+			src_x_offset = x_pos - x_hotspot;
+			src_y_offset = y_pos - (cursor_height - y_hotspot);
 		}
 	} else if (param->rotation == ROTATION_ANGLE_180) {
+		// hotspot = (-x, -y)
 		if (!param->mirror)
-			src_x_offset = pos->x - param->viewport.x;
+			src_x_offset = x_pos - (cursor_width - x_hotspot);

-		src_y_offset = pos->y - param->viewport.y;
+		src_y_offset = y_pos - (cursor_height - y_hotspot);
 	}

 	dst_x_offset = (src_x_offset >= 0) ? src_x_offset : 0;
@@ -1042,8 +1052,8 @@ void hubp2_cursor_set_position(
 			CURSOR_Y_POSITION, pos->y);

 	REG_SET_2(CURSOR_HOT_SPOT, 0,
-			CURSOR_HOT_SPOT_X, x_hotspot,
-			CURSOR_HOT_SPOT_Y, y_hotspot);
+			CURSOR_HOT_SPOT_X, pos->x_hotspot,
+			CURSOR_HOT_SPOT_Y, pos->y_hotspot);

 	REG_SET(CURSOR_DST_OFFSET, 0,
 			CURSOR_DST_X_OFFSET, dst_x_offset);
@@ -1052,8 +1062,8 @@ void hubp2_cursor_set_position(
 	hubp->pos.cur_ctl.bits.cur_enable = cur_en;
 	hubp->pos.position.bits.x_pos = pos->x;
 	hubp->pos.position.bits.y_pos = pos->y;
-	hubp->pos.hot_spot.bits.x_hot = x_hotspot;
-	hubp->pos.hot_spot.bits.y_hot = y_hotspot;
+	hubp->pos.hot_spot.bits.x_hot = pos->x_hotspot;
+	hubp->pos.hot_spot.bits.y_hot = pos->y_hotspot;
 	hubp->pos.dst_offset.bits.dst_x_offset = dst_x_offset;
 	/* Cursor Rectangle Cache
 	 * Cursor bitmaps have different hotspot values
@@ -96,6 +96,13 @@ static void dccg314_set_pixel_rate_div(
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 	enum pixel_rate_div cur_k1 = PIXEL_RATE_DIV_NA, cur_k2 = PIXEL_RATE_DIV_NA;

+	// Don't program 0xF into the register field. Not valid since
+	// K1 / K2 field is only 1 / 2 bits wide
+	if (k1 == PIXEL_RATE_DIV_NA || k2 == PIXEL_RATE_DIV_NA) {
+		BREAK_TO_DEBUGGER();
+		return;
+	}
+
 	dccg314_get_pixel_rate_div(dccg, otg_inst, &cur_k1, &cur_k2);
 	if (k1 == cur_k1 && k2 == cur_k2)
 		return;
@@ -348,10 +348,8 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 	two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing);
 	odm_combine_factor = get_odm_config(pipe_ctx, NULL);

-	if (pipe_ctx->stream->signal == SIGNAL_TYPE_VIRTUAL)
-		return odm_combine_factor;
-
 	if (is_dp_128b_132b_signal(pipe_ctx)) {
+		*k1_div = PIXEL_RATE_DIV_BY_1;
 		*k2_div = PIXEL_RATE_DIV_BY_1;
 	} else if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal) || dc_is_dvi_signal(pipe_ctx->stream->signal)) {
 		*k1_div = PIXEL_RATE_DIV_BY_1;
@@ -359,7 +357,7 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 			*k2_div = PIXEL_RATE_DIV_BY_2;
 		else
 			*k2_div = PIXEL_RATE_DIV_BY_4;
-	} else if (dc_is_dp_signal(pipe_ctx->stream->signal)) {
+	} else if (dc_is_dp_signal(pipe_ctx->stream->signal) || dc_is_virtual_signal(pipe_ctx->stream->signal)) {
 		if (two_pix_per_container) {
 			*k1_div = PIXEL_RATE_DIV_BY_1;
 			*k2_div = PIXEL_RATE_DIV_BY_2;
@@ -96,8 +96,10 @@ static void dccg32_set_pixel_rate_div(

 	// Don't program 0xF into the register field. Not valid since
 	// K1 / K2 field is only 1 / 2 bits wide
-	if (k1 == PIXEL_RATE_DIV_NA || k2 == PIXEL_RATE_DIV_NA)
+	if (k1 == PIXEL_RATE_DIV_NA || k2 == PIXEL_RATE_DIV_NA) {
+		BREAK_TO_DEBUGGER();
 		return;
+	}

 	dccg32_get_pixel_rate_div(dccg, otg_inst, &cur_k1, &cur_k2);
 	if (k1 == cur_k1 && k2 == cur_k2)
@@ -227,8 +227,13 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];

+		/* If PSR is supported on an eDP panel that's connected, but that panel is
+		 * not in PSR at the time of trying to enter MALL SS, we have to include it
+		 * in the static screen CAB calculation
+		 */
 		if (!pipe->stream || !pipe->plane_state ||
-				pipe->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED ||
+				(pipe->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED &&
+				pipe->stream->link->psr_settings.psr_allow_active) ||
 				pipe->stream->mall_stream_config.type == SUBVP_PHANTOM)
 			continue;

@@ -1170,10 +1175,8 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 	two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing);
 	odm_combine_factor = get_odm_config(pipe_ctx, NULL);

-	if (pipe_ctx->stream->signal == SIGNAL_TYPE_VIRTUAL)
-		return odm_combine_factor;
-
 	if (is_dp_128b_132b_signal(pipe_ctx)) {
+		*k1_div = PIXEL_RATE_DIV_BY_1;
 		*k2_div = PIXEL_RATE_DIV_BY_1;
 	} else if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal) || dc_is_dvi_signal(pipe_ctx->stream->signal)) {
 		*k1_div = PIXEL_RATE_DIV_BY_1;
@@ -723,7 +723,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	/* Must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions*/
 	.enable_double_buffered_dsc_pg_support = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
-	.allow_sw_cursor_fallback = false,
+	.allow_sw_cursor_fallback = false, // Linux can't do SW cursor "fallback"
 	.alloc_extra_way_for_cursor = true,
 	.min_prefetch_in_strobe_ns = 60000, // 60us
 };
@@ -1721,6 +1721,27 @@ static struct dc_stream_state *dcn32_enable_phantom_stream(struct dc *dc,
 	return phantom_stream;
 }

+void dcn32_retain_phantom_pipes(struct dc *dc, struct dc_state *context)
+{
+	int i;
+	struct dc_plane_state *phantom_plane = NULL;
+	struct dc_stream_state *phantom_stream = NULL;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->top_pipe && !pipe->prev_odm_pipe &&
+				pipe->plane_state && pipe->stream &&
+				pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) {
+			phantom_plane = pipe->plane_state;
+			phantom_stream = pipe->stream;
+
+			dc_plane_state_retain(phantom_plane);
+			dc_stream_retain(phantom_stream);
+		}
+	}
+}
+
 // return true if removed piped from ctx, false otherwise
 bool dcn32_remove_phantom_pipes(struct dc *dc, struct dc_state *context)
 {
@@ -1903,7 +1924,7 @@ int dcn32_populate_dml_pipes_from_context(

 		pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal;
 		if (context->stream_count == 1 &&
-				context->stream_status[0].plane_count <= 1 &&
+				context->stream_status[0].plane_count == 1 &&
 				!dc_is_hdmi_signal(res_ctx->pipe_ctx[i].stream->signal) &&
 				is_h_timing_divisible_by_2(res_ctx->pipe_ctx[i].stream) &&
 				pipe->stream->timing.pix_clk_100hz * 100 > DCN3_2_VMIN_DISPCLK_HZ &&
@@ -2033,6 +2054,7 @@ static struct resource_funcs dcn32_res_pool_funcs = {
 	.update_soc_for_wm_a = dcn30_update_soc_for_wm_a,
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.remove_phantom_pipes = dcn32_remove_phantom_pipes,
+	.retain_phantom_pipes = dcn32_retain_phantom_pipes,
 };


@@ -2124,6 +2146,7 @@ static bool dcn32_resource_construct(
 	dc->caps.subvp_swath_height_margin_lines = 16;
 	dc->caps.subvp_pstate_allow_width_us = 20;
 	dc->caps.subvp_vertical_int_margin_us = 30;
+	dc->caps.subvp_drr_vblank_start_margin_us = 100; // 100us margin

 	dc->caps.max_slave_planes = 2;
 	dc->caps.max_slave_yuv_planes = 2;
@@ -83,6 +83,9 @@ bool dcn32_release_post_bldn_3dlut(
 bool dcn32_remove_phantom_pipes(struct dc *dc,
 		struct dc_state *context);

+void dcn32_retain_phantom_pipes(struct dc *dc,
+		struct dc_state *context);
+
 void dcn32_add_phantom_pipes(struct dc *dc,
 		struct dc_state *context,
 		display_e2e_pipe_params_st *pipes,
@@ -111,7 +111,7 @@ uint32_t dcn32_helper_calculate_num_ways_for_subvp(struct dc *dc, struct dc_stat
 			mall_alloc_width_blk_aligned = full_vp_width_blk_aligned;

 			/* mall_alloc_height_blk_aligned_l/c = CEILING(sub_vp_height_l/c - 1, blk_height_l/c) + blk_height_l/c */
-			mall_alloc_height_blk_aligned = (pipe->stream->timing.v_addressable - 1 + mblk_height - 1) /
+			mall_alloc_height_blk_aligned = (pipe->plane_res.scl_data.viewport.height - 1 + mblk_height - 1) /
 					mblk_height * mblk_height + mblk_height;

 			/* full_mblk_width_ub_l/c = mall_alloc_width_blk_aligned_l/c;
@@ -721,7 +721,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	/*must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions*/
 	.enable_double_buffered_dsc_pg_support = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
-	.allow_sw_cursor_fallback = false,
+	.allow_sw_cursor_fallback = false, // Linux can't do SW cursor "fallback"
 	.alloc_extra_way_for_cursor = true,
 	.min_prefetch_in_strobe_ns = 60000, // 60us
 };
@@ -1621,6 +1621,7 @@ static struct resource_funcs dcn321_res_pool_funcs = {
 	.update_soc_for_wm_a = dcn30_update_soc_for_wm_a,
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.remove_phantom_pipes = dcn32_remove_phantom_pipes,
+	.retain_phantom_pipes = dcn32_retain_phantom_pipes,
 };


@@ -1711,6 +1712,7 @@ static bool dcn321_resource_construct(
 	dc->caps.subvp_swath_height_margin_lines = 16;
 	dc->caps.subvp_pstate_allow_width_us = 20;
 	dc->caps.subvp_vertical_int_margin_us = 30;
+	dc->caps.subvp_drr_vblank_start_margin_us = 100; // 100us margin
 	dc->caps.max_slave_planes = 1;
 	dc->caps.max_slave_yuv_planes = 1;
 	dc->caps.max_slave_rgb_planes = 1;
@@ -949,6 +949,7 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 	int plane_count;
 	int i;
 	unsigned int optimized_min_dst_y_next_start_us;
+	bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > 1000.0;

 	plane_count = 0;
 	optimized_min_dst_y_next_start_us = 0;
@@ -963,6 +964,8 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 	 * 	2. single eDP, on link 0, 1 plane and stutter period > 5ms
 	 * Z10 only cases:
 	 * 	1. single eDP, on link 0, 1 plane and stutter period >= 5ms
+	 * Z8 cases:
+	 * 	1. stutter period sufficient
 	 * Zstate not allowed cases:
 	 * 	1. Everything else
 	 */
@@ -990,11 +993,14 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 		if (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || optimized_min_dst_y_next_start_us > 5000)
 			return DCN_ZSTATE_SUPPORT_ALLOW;
 		else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr)
-			return DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY;
+			return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY;
 		else
-			return DCN_ZSTATE_SUPPORT_DISALLOW;
-	} else
+			return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY : DCN_ZSTATE_SUPPORT_DISALLOW;
+	} else if (allow_z8) {
+		return DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY;
+	} else {
 		return DCN_ZSTATE_SUPPORT_DISALLOW;
+	}
 }

 void dcn20_calculate_dlg_params(
@@ -149,8 +149,8 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 	.num_states = 5,
 	.sr_exit_time_us = 16.5,
 	.sr_enter_plus_exit_time_us = 18.5,
-	.sr_exit_z8_time_us = 442.0,
-	.sr_enter_plus_exit_z8_time_us = 560.0,
+	.sr_exit_z8_time_us = 280.0,
+	.sr_enter_plus_exit_z8_time_us = 350.0,
 	.writeback_latency_us = 12.0,
 	.dram_channel_width_bytes = 4,
 	.round_trip_ping_latency_dcfclk_cycles = 106,
@@ -157,7 +157,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = {
 	.dispclk_dppclk_vco_speed_mhz = 4300.0,
 	.do_urgent_latency_adjustment = true,
 	.urgent_latency_adjustment_fabric_clock_component_us = 1.0,
-	.urgent_latency_adjustment_fabric_clock_reference_mhz = 1000,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 3000,
 };

 void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr)
@@ -211,7 +211,7 @@ void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr)
 	/* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */
 	if (clk_mgr->base.ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) {
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].valid = true;
-		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 38;
+		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 50;
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.fclk_change_latency_us = fclk_change_latency_us;
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us = sr_exit_time_us;
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us;
@@ -221,7 +221,7 @@ void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr)
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_uclk = min_uclk_mhz;
 		clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_uclk = 0xFFFF;
 		clk_mgr->base.bw_params->dummy_pstate_table[0].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[0].memclk_mhz * 16;
-		clk_mgr->base.bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 38;
+		clk_mgr->base.bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 50;
 		clk_mgr->base.bw_params->dummy_pstate_table[1].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[1].memclk_mhz * 16;
 		clk_mgr->base.bw_params->dummy_pstate_table[1].dummy_pstate_latency_us = 9;
 		clk_mgr->base.bw_params->dummy_pstate_table[2].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[2].memclk_mhz * 16;
@@ -256,16 +256,24 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
 							    int vlevel)
 {
 	const int max_latency_table_entries = 4;
-	const struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 	int dummy_latency_index = 0;
+	enum clock_change_support temp_clock_change_support = vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb];

 	dc_assert_fp_enabled();

 	while (dummy_latency_index < max_latency_table_entries) {
+		if (temp_clock_change_support != dm_dram_clock_change_unsupported)
+			vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support;
 		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
 				dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;
 		dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false);

+		/* for subvp + DRR case, if subvp pipes are still present we support pstate */
+		if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported &&
+				dcn32_subvp_in_use(dc, context))
+			vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support;
+
 		if (vlevel < context->bw_ctx.dml.vba.soc.num_states &&
 				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] != dm_dram_clock_change_unsupported)
 			break;
@@ -577,6 +585,9 @@ void dcn32_set_phantom_stream_timing(struct dc *dc,
 	num_dpp = vba->NoOfDPP[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]];
 	phantom_vactive += num_dpp > 1 ? vba->meta_row_height[vba->pipe_plane[pipe_idx]] : 0;

+	/* dc->debug.subvp_extra_lines 0 by default*/
+	phantom_vactive += dc->debug.subvp_extra_lines;
+
 	// For backporch of phantom pipe, use vstartup of the main pipe
 	phantom_bp = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);

@@ -1141,7 +1152,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 				context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final ==
 					dm_prefetch_support_uclk_fclk_and_stutter) {
 				context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
-								dm_prefetch_support_stutter;
+								dm_prefetch_support_fclk_and_stutter;
 				/* There are params (such as FabricClock) that need to be recalculated
 				 * after validation fails (otherwise it will be 0). Calculation for
 				 * phantom vactive requires call into DML, so we must ensure all the
@@ -1756,6 +1767,7 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 	if (repopulate_pipes) {
 		int flag_max_mpc_comb = vba->maxMpcComb;
 		int flag_vlevel = vlevel;
+		int i;

 		pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate);

@@ -1774,7 +1786,7 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				flag_max_mpc_comb != context->bw_ctx.dml.vba.maxMpcComb) {
 			/* check the context constructed with pipe split flags is still valid*/
 			bool flags_valid = false;
-			for (int i = flag_vlevel; i < context->bw_ctx.dml.soc.num_states; i++) {
+			for (i = flag_vlevel; i < context->bw_ctx.dml.soc.num_states; i++) {
 				if (vba->ModeSupport[i][flag_max_mpc_comb]) {
 					vba->maxMpcComb = flag_max_mpc_comb;
 					vba->VoltageLevel = i;
@@ -1816,14 +1828,38 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 	unsigned int dummy_latency_index = 0;
 	int maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb;
 	unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed;
+	bool subvp_in_use = dcn32_subvp_in_use(dc, context);
 	unsigned int min_dram_speed_mts_margin;
+	bool need_fclk_lat_as_dummy = false;
+	bool is_subvp_p_drr = true;

 	dc_assert_fp_enabled();

-	// Override DRAMClockChangeSupport for SubVP + DRR case where the DRR cannot switch without stretching it's VBLANK
-	if (!pstate_en && dcn32_subvp_in_use(dc, context)) {
-		context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp;
-		pstate_en = true;
+	/* need to find dummy latency index for subvp */
+	if (subvp_in_use) {
+		/* Override DRAMClockChangeSupport for SubVP + DRR case where the DRR cannot switch without stretching it's VBLANK */
+		if (!pstate_en) {
+			context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp;
+			pstate_en = true;
+			is_subvp_p_drr = true;
+		}
+		dummy_latency_index = dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(dc,
+						context, pipes, pipe_cnt, vlevel);
+
+		/* For DCN32/321 need to validate with fclk pstate change latency equal to dummy so prefetch is
+		 * scheduled correctly to account for dummy pstate.
+		 */
+		if (context->bw_ctx.dml.soc.fclk_change_latency_us < dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us) {
+			need_fclk_lat_as_dummy = true;
+			context->bw_ctx.dml.soc.fclk_change_latency_us =
+					dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;
+		}
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+							dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
+		dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false);
+		if (is_subvp_p_drr) {
+			context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp;
+		}
 	}

 	context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false;
@@ -1847,9 +1883,11 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 			/* For DCN32/321 need to validate with fclk pstate change latency equal to dummy so
 			 * prefetch is scheduled correctly to account for dummy pstate.
 			 */
-			if (dummy_latency_index == 0)
+			if (context->bw_ctx.dml.soc.fclk_change_latency_us < dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us) {
+				need_fclk_lat_as_dummy = true;
 				context->bw_ctx.dml.soc.fclk_change_latency_us =
 						dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;
+			}
 			dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false);
 			maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb;
 			dcfclk_from_fw_based_mclk_switching = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
@@ -1951,13 +1989,13 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,

 		if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] ==
 			dm_dram_clock_change_unsupported) {
-			int min_dram_speed_mts_offset = dc->clk_mgr->bw_params->clk_table.num_entries - 1;
+			int min_dram_speed_mts_offset = dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels - 1;

 			min_dram_speed_mts =
 				dc->clk_mgr->bw_params->clk_table.entries[min_dram_speed_mts_offset].memclk_mhz * 16;
 		}

-		if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+		if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching && !subvp_in_use) {
 			/* find largest table entry that is lower than dram speed,
 			 * but lower than DPM0 still uses DPM0
 			 */
@@ -2037,7 +2075,8 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,

 	context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod;

-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching && dummy_latency_index == 0)
+	/* for proper prefetch calculations, if dummy lat > fclk lat, use fclk lat = dummy lat */
+	if (need_fclk_lat_as_dummy)
 		context->bw_ctx.dml.soc.fclk_change_latency_us =
 				dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;

@@ -2050,10 +2089,12 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,

 	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
 		dcn30_setup_mclk_switch_using_fw_based_vblank_stretch(dc, context);
-		if (dummy_latency_index == 0)
-			context->bw_ctx.dml.soc.fclk_change_latency_us =
-					dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us;
 	}
+
+	/* revert fclk lat changes if required */
+	if (need_fclk_lat_as_dummy)
+		context->bw_ctx.dml.soc.fclk_change_latency_us =
+				dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us;
 }

 static void dcn32_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts,
@@ -670,6 +670,25 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 		v->cursor_bw[k] = mode_lib->vba.NumberOfCursors[k] * mode_lib->vba.CursorWidth[k][0] * mode_lib->vba.CursorBPP[k][0] / 8 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k];
 	}

+	v->NotEnoughDETSwathFillLatencyHiding = dml32_CalculateDETSwathFillLatencyHiding(
+						mode_lib->vba.NumberOfActiveSurfaces,
+						mode_lib->vba.ReturnBW,
+						v->UrgentLatency,
+						mode_lib->vba.SwathHeightY,
+						mode_lib->vba.SwathHeightC,
+						v->swath_width_luma_ub,
+						v->swath_width_chroma_ub,
+						v->BytePerPixelDETY,
+						v->BytePerPixelDETC,
+						mode_lib->vba.DETBufferSizeY,
+						mode_lib->vba.DETBufferSizeC,
+						mode_lib->vba.DPPPerPlane,
+						mode_lib->vba.HTotal,
+						mode_lib->vba.PixelClock,
+						mode_lib->vba.VRatio,
+						mode_lib->vba.VRatioChroma,
+						mode_lib->vba.UsesMALLForPStateChange);
+
 	for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) {
 		v->MaxVStartupLines[k] = ((mode_lib->vba.Interlace[k] &&
 				!mode_lib->vba.ProgressiveToInterlaceUnitInOPP) ?
@@ -1664,6 +1683,7 @@ static void mode_support_configuration(struct vba_vars_st *v,
 				&& mode_lib->vba.PTEBufferSizeNotExceeded[i][j] == true
 				&& mode_lib->vba.DCCMetaBufferSizeNotExceeded[i][j] == true
 				&& mode_lib->vba.NonsupportedDSCInputBPC == false
+				&& mode_lib->vba.NotEnoughDETSwathFillLatencyHidingPerState[i][j] == false
 				&& !mode_lib->vba.ExceededMALLSize
 				&& ((mode_lib->vba.HostVMEnable == false
 				&& !mode_lib->vba.ImmediateFlipRequiredFinal)
@@ -2286,7 +2306,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 					&& (mode_lib->vba.Output[k] == dm_dp || mode_lib->vba.Output[k] == dm_dp2p0
 							|| mode_lib->vba.Output[k] == dm_edp
 							|| mode_lib->vba.Output[k] == dm_hdmi)
-					&& mode_lib->vba.OutputBppPerState[i][k] == 0 && (mode_lib->vba.UsesMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe)) {
+					&& mode_lib->vba.OutputBppPerState[i][k] == 0) {
 				mode_lib->vba.LinkCapacitySupport[i] = false;
 			}
 		}
@@ -3158,6 +3178,25 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 					mode_lib->vba.UrgentBurstFactorChroma,
 					mode_lib->vba.UrgentBurstFactorCursor);

+			mode_lib->vba.NotEnoughDETSwathFillLatencyHidingPerState[i][j] = dml32_CalculateDETSwathFillLatencyHiding(
+					mode_lib->vba.NumberOfActiveSurfaces,
+					mode_lib->vba.ReturnBWPerState[i][j],
+					mode_lib->vba.UrgLatency[i],
+					mode_lib->vba.SwathHeightYThisState,
+					mode_lib->vba.SwathHeightCThisState,
+					mode_lib->vba.swath_width_luma_ub_this_state,
+					mode_lib->vba.swath_width_chroma_ub_this_state,
+					mode_lib->vba.BytePerPixelInDETY,
+					mode_lib->vba.BytePerPixelInDETC,
+					mode_lib->vba.DETBufferSizeYThisState,
+					mode_lib->vba.DETBufferSizeCThisState,
+					mode_lib->vba.NoOfDPPThisState,
+					mode_lib->vba.HTotal,
+					mode_lib->vba.PixelClock,
+					mode_lib->vba.VRatio,
+					mode_lib->vba.VRatioChroma,
+					mode_lib->vba.UsesMALLForPStateChange);
+
 			v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.VMDataOnlyReturnBWPerState = dml32_get_return_bw_mbps_vm_only(&mode_lib->vba.soc, i,
 					mode_lib->vba.DCFCLKState[i][j], mode_lib->vba.FabricClockPerState[i],
 					mode_lib->vba.DRAMSpeedPerState[i]);
@@ -6228,3 +6228,72 @@ void dml32_CalculateImmediateFlipBandwithSupport(unsigned int NumberOfActiveSurf
 	*ImmediateFlipBandwidthSupport = (*TotalBandwidth <= ReturnBW);
 	*FractionOfUrgentBandwidth = *TotalBandwidth / ReturnBW;
 }
+
+bool dml32_CalculateDETSwathFillLatencyHiding(unsigned int NumberOfActiveSurfaces,
+		double ReturnBW,
+		double UrgentLatency,
+		unsigned int SwathHeightY[],
+		unsigned int SwathHeightC[],
+		unsigned int SwathWidthY[],
+		unsigned int SwathWidthC[],
+		double  BytePerPixelInDETY[],
+		double  BytePerPixelInDETC[],
+		unsigned int    DETBufferSizeY[],
+		unsigned int    DETBufferSizeC[],
+		unsigned int	NumOfDPP[],
+		unsigned int	HTotal[],
+		double	PixelClock[],
+		double	VRatioY[],
+		double	VRatioC[],
+		enum dm_use_mall_for_pstate_change_mode UsesMALLForPStateChange[DC__NUM_DPP__MAX])
+{
+	int k;
+	double SwathSizeAllSurfaces = 0;
+	double SwathSizeAllSurfacesInFetchTimeUs;
+	double DETSwathLatencyHidingUs;
+	double DETSwathLatencyHidingYUs;
+	double DETSwathLatencyHidingCUs;
+	double SwathSizePerSurfaceY[DC__NUM_DPP__MAX];
+	double SwathSizePerSurfaceC[DC__NUM_DPP__MAX];
+	bool NotEnoughDETSwathFillLatencyHiding = false;
+
+	/* calculate sum of single swath size for all pipes in bytes*/
+	for (k = 0; k < NumberOfActiveSurfaces; k++) {
+		SwathSizePerSurfaceY[k] += SwathHeightY[k] * SwathWidthY[k] * BytePerPixelInDETY[k] * NumOfDPP[k];
+
+		if (SwathHeightC[k] != 0)
+			SwathSizePerSurfaceC[k] += SwathHeightC[k] * SwathWidthC[k] * BytePerPixelInDETC[k] * NumOfDPP[k];
+		else
+			SwathSizePerSurfaceC[k] = 0;
+
+		SwathSizeAllSurfaces += SwathSizePerSurfaceY[k] + SwathSizePerSurfaceC[k];
+	}
+
+	SwathSizeAllSurfacesInFetchTimeUs = SwathSizeAllSurfaces / ReturnBW + UrgentLatency;
+
+	/* ensure all DET - 1 swath can hide a fetch for all surfaces */
+	for (k = 0; k < NumberOfActiveSurfaces; k++) {
+		double LineTime = HTotal[k] / PixelClock[k];
+
+		/* only care if surface is not phantom */
+		if (UsesMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe) {
+			DETSwathLatencyHidingYUs = (dml_floor(DETBufferSizeY[k] / BytePerPixelInDETY[k] / SwathWidthY[k], 1.0) - SwathHeightY[k]) / VRatioY[k] * LineTime;
+
+			if (SwathHeightC[k] != 0) {
+				DETSwathLatencyHidingCUs = (dml_floor(DETBufferSizeC[k] / BytePerPixelInDETC[k] / SwathWidthC[k], 1.0) - SwathHeightC[k]) / VRatioC[k] * LineTime;
+
+				DETSwathLatencyHidingUs = dml_min(DETSwathLatencyHidingYUs, DETSwathLatencyHidingCUs);
+			} else {
+				DETSwathLatencyHidingUs = DETSwathLatencyHidingYUs;
+			}
+
+			/* DET must be able to hide time to fetch 1 swath for each surface */
+			if (DETSwathLatencyHidingUs < SwathSizeAllSurfacesInFetchTimeUs) {
+				NotEnoughDETSwathFillLatencyHiding = true;
+				break;
+			}
+		}
+	}
+
+	return NotEnoughDETSwathFillLatencyHiding;
+}
@@ -1141,4 +1141,22 @@ void dml32_CalculateImmediateFlipBandwithSupport(unsigned int NumberOfActiveSurf
 		double  *FractionOfUrgentBandwidth,
 		bool *ImmediateFlipBandwidthSupport);

+bool dml32_CalculateDETSwathFillLatencyHiding(unsigned int NumberOfActiveSurfaces,
+		double ReturnBW,
+		double UrgentLatency,
+		unsigned int SwathHeightY[],
+		unsigned int SwathHeightC[],
+		unsigned int SwathWidthY[],
+		unsigned int SwathWidthC[],
+		double  BytePerPixelInDETY[],
+		double  BytePerPixelInDETC[],
+		unsigned int    DETBufferSizeY[],
+		unsigned int    DETBufferSizeC[],
+		unsigned int	NumOfDPP[],
+		unsigned int	HTotal[],
+		double	PixelClock[],
+		double	VRatioY[],
+		double	VRatioC[],
+		enum dm_use_mall_for_pstate_change_mode UsesMALLForPStateChange[DC__NUM_DPP__MAX]);
+
 #endif
@@ -126,9 +126,9 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = {
 	.sr_enter_plus_exit_z8_time_us = 320,
 	.writeback_latency_us = 12.0,
 	.round_trip_ping_latency_dcfclk_cycles = 263,
-	.urgent_latency_pixel_data_only_us = 9.35,
-	.urgent_latency_pixel_mixed_with_vm_data_us = 9.35,
-	.urgent_latency_vm_data_only_us = 9.35,
+	.urgent_latency_pixel_data_only_us = 4,
+	.urgent_latency_pixel_mixed_with_vm_data_us = 4,
+	.urgent_latency_vm_data_only_us = 4,
 	.fclk_change_latency_us = 20,
 	.usr_retraining_latency_us = 2,
 	.smn_latency_us = 2,
@@ -156,7 +156,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = {
 	.dispclk_dppclk_vco_speed_mhz = 4300.0,
 	.do_urgent_latency_adjustment = true,
 	.urgent_latency_adjustment_fabric_clock_component_us = 1.0,
-	.urgent_latency_adjustment_fabric_clock_reference_mhz = 1000,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 3000,
 };

 static void get_optimal_ntuple(struct _vcs_dpi_voltage_scaling_st *entry)
@@ -1050,6 +1050,7 @@ struct vba_vars_st {
 	double MinFullDETBufferingTime;
 	double AverageReadBandwidthGBytePerSecond;
 	bool   FirstMainPlane;
+	bool NotEnoughDETSwathFillLatencyHiding;

 	unsigned int ViewportWidthChroma[DC__NUM_DPP__MAX];
 	unsigned int ViewportHeightChroma[DC__NUM_DPP__MAX];
@@ -1233,6 +1234,7 @@ struct vba_vars_st {
 	unsigned int BlockWidthC[DC__NUM_DPP__MAX];
 	unsigned int SubViewportLinesNeededInMALL[DC__NUM_DPP__MAX];
 	bool VActiveBandwithSupport[DC__VOLTAGE_STATES][2];
+	bool NotEnoughDETSwathFillLatencyHidingPerState[DC__VOLTAGE_STATES][2];
 	struct dummy_vars dummy_vars;
 };

@@ -241,6 +241,7 @@ struct resource_funcs {
            unsigned int index);

 	bool (*remove_phantom_pipes)(struct dc *dc, struct dc_state *context);
+	void (*retain_phantom_pipes)(struct dc *dc, struct dc_state *context);
 	void (*get_panel_config_defaults)(struct dc_panel_config *panel_config);
 };

@@ -83,10 +83,15 @@ static const struct dpp_input_csc_matrix __maybe_unused dpp_input_csc_matrix[] =
 	{COLOR_SPACE_YCBCR709,
 		{0x3265, 0x2000, 0, 0xe6ce, 0xf105, 0x2000, 0xfa01, 0xa7d, 0,
 						0x2000, 0x3b61, 0xe24f} },
-
 	{COLOR_SPACE_YCBCR709_LIMITED,
 		{0x39a6, 0x2568, 0, 0xe0d6, 0xeedd, 0x2568, 0xf925, 0x9a8, 0,
-						0x2568, 0x43ee, 0xdbb2} }
+						0x2568, 0x43ee, 0xdbb2} },
+	{COLOR_SPACE_2020_YCBCR,
+		{0x2F30, 0x2000, 0, 0xE869, 0xEDB7, 0x2000, 0xFABC, 0xBC6, 0,
+						0x2000, 0x3C34, 0xE1E6} },
+	{COLOR_SPACE_2020_RGB_LIMITEDRANGE,
+		{0x35E0, 0x255F, 0, 0xE2B3, 0xEB20, 0x255F, 0xF9FD, 0xB1E, 0,
+						0x255F, 0x44BD, 0xDB43} }
 };

 struct dpp_grph_csc_adjustment {
@@ -0,0 +1,28 @@
+
+/*
+ * Copyright 2022 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
+ *
+ */
+/*********************************************************************/
+//				USB4 DPIA BANDWIDTH ALLOCATION LOGIC
+/*********************************************************************/
@@ -0,0 +1,69 @@
+/*
+ * Copyright 2021 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 DC_INC_LINK_DP_DPIA_BW_H_
+#define DC_INC_LINK_DP_DPIA_BW_H_
+
+// XXX: TODO: Re-add for Phase 2
+/* Number of Host Routers per motherboard is 2 and 2 DPIA per host router */
+#define MAX_HR_NUM 2
+
+struct dc_host_router_bw_alloc {
+	int max_bw[MAX_HR_NUM];             // The Max BW that each Host Router has available to be shared btw DPIAs
+	int total_estimated_bw[MAX_HR_NUM]; // The Total Verified and available BW that Host Router has
+};
+
+/*
+ * Enable BW Allocation Mode Support from the DP-Tx side
+ *
+ * @link: pointer to the dc_link struct instance
+ *
+ * return: SUCCESS or FAILURE
+ */
+bool set_dptx_usb4_bw_alloc_support(struct dc_link *link);
+
+/*
+ * Send a request from DP-Tx requesting to allocate BW remotely after
+ * allocating it locally. This will get processed by CM and a CB function
+ * will be called.
+ *
+ * @link: pointer to the dc_link struct instance
+ * @req_bw: The requested bw in Kbyte to allocated
+ *
+ * return: none
+ */
+void set_usb4_req_bw_req(struct dc_link *link, int req_bw);
+
+/*
+ * CB function for when the status of the Req above is complete. We will
+ * find out the result of allocating on CM and update structs accordingly
+ *
+ * @link: pointer to the dc_link struct instance
+ *
+ * return: none
+ */
+void get_usb4_req_bw_resp(struct dc_link *link);
+
+#endif /* DC_INC_LINK_DP_DPIA_BW_H_ */
@@ -1029,13 +1029,14 @@ struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 {
 			uint16_t vtotal;
 			uint16_t htotal;
 			uint8_t vblank_pipe_index;
-			uint8_t padding[2];
+			uint8_t padding[1];
 			struct {
 				uint8_t drr_in_use;
 				uint8_t drr_window_size_ms;	// Indicates largest VMIN/VMAX adjustment per frame
 				uint16_t min_vtotal_supported;	// Min VTOTAL that supports switching in VBLANK
 				uint16_t max_vtotal_supported;	// Max VTOTAL that can support SubVP static scheduling
 				uint8_t use_ramping;		// Use ramping or not
+				uint8_t drr_vblank_start_margin;
 			} drr_info;				// DRR considered as part of SubVP + VBLANK case
 		} vblank_data;
 	} pipe_config;
@@ -102,9 +102,18 @@ static const struct abm_parameters abm_settings_config1[abm_defines_max_level] =
 	{0x82,   0x4d,    0x20,       0x00,     0x00,        0xff,     0xb3, 0x70,     0x70,     0xcccc,  0xcccc},
 };

+static const struct abm_parameters abm_settings_config2[abm_defines_max_level] = {
+//  min_red  max_red  bright_pos  dark_pos  bright_gain  contrast  dev   min_knee  max_knee  blRed    blStart
+	{0xf0,   0xbf,    0x20,       0x00,     0x88,        0x99,     0xb3, 0x40,     0xe0,    0x0000,  0xcccc},
+	{0xd8,   0x85,    0x20,       0x00,     0x70,        0x90,     0xa8, 0x40,     0xc8,    0x0700,  0xb333},
+	{0xb8,   0x58,    0x20,       0x00,     0x64,        0x88,     0x78, 0x70,     0xa0,    0x7000,  0x9999},
+	{0x82,   0x40,    0x20,       0x00,     0x00,        0xb8,     0xb3, 0x70,     0x70,    0xc333,  0xb333},
+};
+
 static const struct abm_parameters * const abm_settings[] = {
 	abm_settings_config0,
 	abm_settings_config1,
+	abm_settings_config2,
 };

 #define NUM_AMBI_LEVEL    5
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 #ifndef _dcn_3_0_0_OFFSET_HEADER
 #define _dcn_3_0_0_OFFSET_HEADER

@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 #ifndef _dcn_3_0_0_SH_MASK_HEADER
 #define _dcn_3_0_0_SH_MASK_HEADER

@@ -993,7 +993,8 @@
 #define mmUVD_RAS_VCPU_VCODEC_STATUS_BASE_IDX                  1
 #define mmUVD_RAS_MMSCH_FATAL_ERROR                            0x0058
 #define mmUVD_RAS_MMSCH_FATAL_ERROR_BASE_IDX                   1
-
+#define mmVCN_RAS_CNTL                                                                                 0x04b9
+#define mmVCN_RAS_CNTL_BASE_IDX                                                                        1

 /* JPEG 2_6_0 regs */
 #define mmUVD_RAS_JPEG0_STATUS                                 0x0059
@@ -3618,6 +3618,33 @@
 #define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_VF_MASK             0x7FFFFFFFL
 #define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_PF_MASK             0x80000000L

+//VCN 2_6_0 VCN_RAS_CNTL
+#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN__SHIFT                                                                0x0
+#define VCN_RAS_CNTL__MMSCH_FATAL_ERROR_EN__SHIFT                                                             0x1
+#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN__SHIFT                                                               0x4
+#define VCN_RAS_CNTL__MMSCH_PMI_EN__SHIFT                                                                     0x5
+#define VCN_RAS_CNTL__VCPU_VCODEC_REARM__SHIFT                                                                0x8
+#define VCN_RAS_CNTL__MMSCH_REARM__SHIFT                                                                      0x9
+#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN__SHIFT                                                             0xc
+#define VCN_RAS_CNTL__VCPU_VCODEC_READY__SHIFT                                                                0x10
+#define VCN_RAS_CNTL__MMSCH_READY__SHIFT                                                                      0x11
+#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK                                                                  0x00000001L
+#define VCN_RAS_CNTL__MMSCH_FATAL_ERROR_EN_MASK                                                               0x00000002L
+#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK                                                                 0x00000010L
+#define VCN_RAS_CNTL__MMSCH_PMI_EN_MASK                                                                       0x00000020L
+#define VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK                                                                  0x00000100L
+#define VCN_RAS_CNTL__MMSCH_REARM_MASK                                                                        0x00000200L
+#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK                                                               0x00001000L
+#define VCN_RAS_CNTL__VCPU_VCODEC_READY_MASK                                                                  0x00010000L
+#define VCN_RAS_CNTL__MMSCH_READY_MASK                                                                        0x00020000L
+
+//VCN 2_6_0 UVD_VCPU_INT_EN
+#define UVD_VCPU_INT_EN__RASCNTL_VCPU_VCODEC_EN__SHIFT                                                        0x16
+#define UVD_VCPU_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK                                                          0x00400000L
+
+//VCN 2_6_0 UVD_SYS_INT_EN
+#define UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK                                                           0x04000000L
+
 /* JPEG 2_6_0 UVD_RAS_JPEG0_STATUS */
 #define UVD_RAS_JPEG0_STATUS__POISONED_VF__SHIFT                0x0
 #define UVD_RAS_JPEG0_STATUS__POISONED_PF__SHIFT                0x1f
@@ -9292,7 +9292,7 @@ typedef struct {

 typedef struct {
  VFCT_IMAGE_HEADER   VbiosHeader;
-  UCHAR   VbiosContent[1];
+  UCHAR   VbiosContent[];
 }GOP_VBIOS_CONTENT;

 typedef struct {
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 #ifndef YELLOW_CARP_OFFSET_H
 #define YELLOW_CARP_OFFSET_H

@@ -168,6 +168,7 @@ struct smu_temperature_range {
 	int mem_crit_max;
 	int mem_emergency_max;
 	int software_shutdown_temp;
+	int software_shutdown_temp_offset;
 };

 struct smu_state_validation_block {
@@ -25,10 +25,10 @@

 // *** IMPORTANT ***
 // PMFW TEAM: Always increment the interface version on any change to this file
-#define SMU13_DRIVER_IF_VERSION  0x2C
+#define SMU13_DRIVER_IF_VERSION  0x35

 //Increment this version if SkuTable_t or BoardTable_t change
-#define PPTABLE_VERSION 0x20
+#define PPTABLE_VERSION 0x27

 #define NUM_GFXCLK_DPM_LEVELS    16
 #define NUM_SOCCLK_DPM_LEVELS    8
@@ -96,7 +96,7 @@
 #define FEATURE_MEM_TEMP_READ_BIT             47
 #define FEATURE_ATHUB_MMHUB_PG_BIT            48
 #define FEATURE_SOC_PCC_BIT                   49
-#define FEATURE_SPARE_50_BIT                  50
+#define FEATURE_EDC_PWRBRK_BIT                50
 #define FEATURE_SPARE_51_BIT                  51
 #define FEATURE_SPARE_52_BIT                  52
 #define FEATURE_SPARE_53_BIT                  53
@@ -282,15 +282,15 @@ typedef enum {
 } I2cControllerPort_e;

 typedef enum {
-  I2C_CONTROLLER_NAME_VR_GFX = 0,
-  I2C_CONTROLLER_NAME_VR_SOC,
-  I2C_CONTROLLER_NAME_VR_VMEMP,
-  I2C_CONTROLLER_NAME_VR_VDDIO,
-  I2C_CONTROLLER_NAME_LIQUID0,
-  I2C_CONTROLLER_NAME_LIQUID1,
-  I2C_CONTROLLER_NAME_PLX,
-  I2C_CONTROLLER_NAME_OTHER,
-  I2C_CONTROLLER_NAME_COUNT,
+	I2C_CONTROLLER_NAME_VR_GFX = 0,
+	I2C_CONTROLLER_NAME_VR_SOC,
+	I2C_CONTROLLER_NAME_VR_VMEMP,
+	I2C_CONTROLLER_NAME_VR_VDDIO,
+	I2C_CONTROLLER_NAME_LIQUID0,
+	I2C_CONTROLLER_NAME_LIQUID1,
+	I2C_CONTROLLER_NAME_PLX,
+	I2C_CONTROLLER_NAME_FAN_INTAKE,
+	I2C_CONTROLLER_NAME_COUNT,
 } I2cControllerName_e;

 typedef enum {
@@ -302,6 +302,7 @@ typedef enum {
  I2C_CONTROLLER_THROTTLER_LIQUID0,
  I2C_CONTROLLER_THROTTLER_LIQUID1,
  I2C_CONTROLLER_THROTTLER_PLX,
+  I2C_CONTROLLER_THROTTLER_FAN_INTAKE,
  I2C_CONTROLLER_THROTTLER_INA3221,
  I2C_CONTROLLER_THROTTLER_COUNT,
 } I2cControllerThrottler_e;
@@ -309,8 +310,9 @@ typedef enum {
 typedef enum {
  I2C_CONTROLLER_PROTOCOL_VR_XPDE132G5,
  I2C_CONTROLLER_PROTOCOL_VR_IR35217,
-  I2C_CONTROLLER_PROTOCOL_TMP_TMP102A,
+  I2C_CONTROLLER_PROTOCOL_TMP_MAX31875,
  I2C_CONTROLLER_PROTOCOL_INA3221,
+  I2C_CONTROLLER_PROTOCOL_TMP_MAX6604,
  I2C_CONTROLLER_PROTOCOL_COUNT,
 } I2cControllerProtocol_e;

@@ -690,6 +692,9 @@ typedef struct {
 #define PP_OD_FEATURE_UCLK_BIT      8
 #define PP_OD_FEATURE_ZERO_FAN_BIT      9
 #define PP_OD_FEATURE_TEMPERATURE_BIT 10
+#define PP_OD_FEATURE_POWER_FEATURE_CTRL_BIT 11
+#define PP_OD_FEATURE_ASIC_TDC_BIT 12
+#define PP_OD_FEATURE_COUNT 13

 typedef enum {
  PP_OD_POWER_FEATURE_ALWAYS_ENABLED,
@@ -697,6 +702,11 @@ typedef enum {
  PP_OD_POWER_FEATURE_ALWAYS_DISABLED,
 } PP_OD_POWER_FEATURE_e;

+typedef enum {
+  FAN_MODE_AUTO = 0,
+  FAN_MODE_MANUAL_LINEAR,
+} FanMode_e;
+
 typedef struct {
  uint32_t FeatureCtrlMask;

@@ -708,8 +718,8 @@ typedef struct {
  uint8_t                RuntimePwrSavingFeaturesCtrl;

  //Frequency changes
-  int16_t               GfxclkFmin;           // MHz
-  int16_t               GfxclkFmax;           // MHz
+  int16_t                GfxclkFmin;           // MHz
+  int16_t                GfxclkFmax;           // MHz
  uint16_t               UclkFmin;             // MHz
  uint16_t               UclkFmax;             // MHz

@@ -730,7 +740,12 @@ typedef struct {
  uint8_t                MaxOpTemp;
  uint8_t                Padding[4];

-  uint32_t               Spare[12];
+  uint16_t               GfxVoltageFullCtrlMode;
+  uint16_t               GfxclkFullCtrlMode;
+  uint16_t               UclkFullCtrlMode;
+  int16_t                AsicTdc;
+
+  uint32_t               Spare[10];
  uint32_t               MmHubPadding[8]; // SMU internal use. Adding here instead of external as a workaround
 } OverDriveTable_t;

@@ -748,8 +763,8 @@ typedef struct {
  uint8_t                IdlePwrSavingFeaturesCtrl;
  uint8_t                RuntimePwrSavingFeaturesCtrl;

-  uint16_t               GfxclkFmin;           // MHz
-  uint16_t               GfxclkFmax;           // MHz
+  int16_t                GfxclkFmin;           // MHz
+  int16_t                GfxclkFmax;           // MHz
  uint16_t               UclkFmin;             // MHz
  uint16_t               UclkFmax;             // MHz

@@ -769,7 +784,12 @@ typedef struct {
  uint8_t                MaxOpTemp;
  uint8_t                Padding[4];

-  uint32_t               Spare[12];
+  uint16_t               GfxVoltageFullCtrlMode;
+  uint16_t               GfxclkFullCtrlMode;
+  uint16_t               UclkFullCtrlMode;
+  int16_t                AsicTdc;
+
+  uint32_t               Spare[10];

 } OverDriveLimits_t;

@@ -903,7 +923,8 @@ typedef struct {
  uint16_t  FanStartTempMin;
  uint16_t  FanStartTempMax;

-  uint32_t Spare[12];
+  uint16_t  PowerMinPpt0[POWER_SOURCE_COUNT];
+  uint32_t  Spare[11];

 } MsgLimits_t;

@@ -1086,11 +1107,13 @@ typedef struct {
  uint32_t        GfxoffSpare[15];

  // GFX GPO
-  float           DfllBtcMasterScalerM;
+  uint32_t        DfllBtcMasterScalerM;
  int32_t         DfllBtcMasterScalerB;
-  float           DfllBtcSlaveScalerM;
+  uint32_t        DfllBtcSlaveScalerM;
  int32_t         DfllBtcSlaveScalerB;
-  uint32_t        GfxGpoSpare[12];
+  uint32_t        DfllPccAsWaitCtrl; //GDFLL_AS_WAIT_CTRL_PCC register value to be passed to RLC msg
+  uint32_t        DfllPccAsStepCtrl; //GDFLL_AS_STEP_CTRL_PCC register value to be passed to RLC msg
+  uint32_t        GfxGpoSpare[10];

  // GFX DCS

@@ -1106,7 +1129,10 @@ typedef struct {
  uint16_t        DcsTimeout;           //This is the amount of time SMU FW waits for RLC to put GFX into GFXOFF before reverting to the fallback mechanism of throttling GFXCLK to Fmin.


-  uint32_t        DcsSpare[16];
+  uint32_t        DcsSpare[14];
+
+  // UCLK section
+  uint16_t     ShadowFreqTableUclk[NUM_UCLK_DPM_LEVELS];     // In MHz

  // UCLK section
  uint8_t      UseStrobeModeOptimizations; //Set to indicate that FW should use strobe mode optimizations
@@ -1163,13 +1189,14 @@ typedef struct {
  uint16_t IntakeTempHighIntakeAcousticLimit;
  uint16_t IntakeTempAcouticLimitReleaseRate;

-  uint16_t FanStalledTempLimitOffset;
+  int16_t FanAbnormalTempLimitOffset;
  uint16_t FanStalledTriggerRpm;
-  uint16_t FanAbnormalTriggerRpm;
-  uint16_t FanPadding;
-
-  uint32_t     FanSpare[14];
+  uint16_t FanAbnormalTriggerRpmCoeff;
+  uint16_t FanAbnormalDetectionEnable;

+  uint8_t      FanIntakeSensorSupport;
+  uint8_t      FanIntakePadding[3];
+  uint32_t     FanSpare[13];
  // SECTION: VDD_GFX AVFS

  uint8_t      OverrideGfxAvfsFuses;
@@ -1193,7 +1220,6 @@ typedef struct {
  uint32_t   dGbV_dT_vmin;
  uint32_t   dGbV_dT_vmax;

-  //Unused: PMFW-9370
  uint32_t   V2F_vmin_range_low;
  uint32_t   V2F_vmin_range_high;
  uint32_t   V2F_vmax_range_low;
@@ -1238,8 +1264,21 @@ typedef struct {
  // SECTION: Advanced Options
  uint32_t          DebugOverrides;

+  // Section: Total Board Power idle vs active coefficients
+  uint8_t     TotalBoardPowerSupport;
+  uint8_t     TotalBoardPowerPadding[3];
+
+  int16_t     TotalIdleBoardPowerM;
+  int16_t     TotalIdleBoardPowerB;
+  int16_t     TotalBoardPowerM;
+  int16_t     TotalBoardPowerB;
+
+  QuadraticInt_t qFeffCoeffGameClock[POWER_SOURCE_COUNT];
+  QuadraticInt_t qFeffCoeffBaseClock[POWER_SOURCE_COUNT];
+  QuadraticInt_t qFeffCoeffBoostClock[POWER_SOURCE_COUNT];
+
  // SECTION: Sku Reserved
-  uint32_t         Spare[64];
+  uint32_t         Spare[43];

  // Padding for MMHUB - do not modify this
  uint32_t     MmHubPadding[8];
@@ -1304,7 +1343,8 @@ typedef struct {
  // SECTION: Clock Spread Spectrum

  // UCLK Spread Spectrum
-  uint16_t     UclkSpreadPadding;
+  uint8_t      UclkTrainingModeSpreadPercent; // Q4.4
+  uint8_t      UclkSpreadPadding;
  uint16_t     UclkSpreadFreq;      // kHz

  // UCLK Spread Spectrum
@@ -1317,11 +1357,7 @@ typedef struct {

  // Section: Memory Config
  uint8_t      DramWidth; // Width of interface to the channel for each DRAM module. See DRAM_BIT_WIDTH_TYPE_e
-  uint8_t      PaddingMem1[3];
-
-  // Section: Total Board Power
-  uint16_t     TotalBoardPower;     //Only needed for TCP Estimated case, where TCP = TGP+Total Board Power
-  uint16_t     BoardPowerPadding;
+  uint8_t      PaddingMem1[7];

  // SECTION: UMC feature flags
  uint8_t      HsrEnabled;
@@ -1423,8 +1459,11 @@ typedef struct {
  uint16_t Vcn1ActivityPercentage  ;

  uint32_t EnergyAccumulator;
-  uint16_t AverageSocketPower    ;
+  uint16_t AverageSocketPower;
+  uint16_t AverageTotalBoardPower;
+
  uint16_t AvgTemperature[TEMP_COUNT];
+  uint16_t AvgTemperatureFanIntake;

  uint8_t  PcieRate               ;
  uint8_t  PcieWidth              ;
@@ -1592,5 +1631,7 @@ typedef struct {
 #define IH_INTERRUPT_CONTEXT_ID_AUDIO_D0            0x5
 #define IH_INTERRUPT_CONTEXT_ID_AUDIO_D3            0x6
 #define IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING  0x7
+#define IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL        0x8
+#define IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY        0x9

 #endif
@@ -131,7 +131,13 @@
 #define PPSMC_MSG_EnableAudioStutterWA           0x44
 #define PPSMC_MSG_PowerUpUmsch                   0x45
 #define PPSMC_MSG_PowerDownUmsch                 0x46
-#define PPSMC_Message_Count                      0x47
+#define PPSMC_MSG_SetDcsArch                     0x47
+#define PPSMC_MSG_TriggerVFFLR                   0x48
+#define PPSMC_MSG_SetNumBadMemoryPagesRetired    0x49
+#define PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel 0x4A
+#define PPSMC_MSG_SetPriorityDeltaGain           0x4B
+#define PPSMC_MSG_AllowIHHostInterrupt           0x4C
+#define PPSMC_Message_Count                      0x4D

 //Debug Dump Message
 #define DEBUGSMC_MSG_TestMessage                    0x1
@@ -239,7 +239,9 @@
 	__SMU_DUMMY_MAP(DriverMode2Reset), \
 	__SMU_DUMMY_MAP(GetGfxOffStatus),		 \
 	__SMU_DUMMY_MAP(GetGfxOffEntryCount),		 \
-	__SMU_DUMMY_MAP(LogGfxOffResidency),
+	__SMU_DUMMY_MAP(LogGfxOffResidency),			\
+	__SMU_DUMMY_MAP(SetNumBadMemoryPagesRetired),		\
+	__SMU_DUMMY_MAP(SetBadMemoryPagesRetiredFlagsPerChannel),

 #undef __SMU_DUMMY_MAP
 #define __SMU_DUMMY_MAP(type)	SMU_MSG_##type
@@ -31,7 +31,7 @@
 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x07
 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04
 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10 0x32
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x2C
+#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x35
 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_10 0x1D

 #define SMU13_MODE1_RESET_WAIT_TIME_IN_MS 500  //500ms
@@ -1376,6 +1376,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 	 */
 	uint32_t ctxid = entry->src_data[0];
 	uint32_t data;
+	uint32_t high;

 	if (client_id == SOC15_IH_CLIENTID_THM) {
 		switch (src_id) {
@@ -1432,6 +1433,36 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 					schedule_work(&smu->throttling_logging_work);

 				break;
+			case 0x8:
+				high = smu->thermal_range.software_shutdown_temp +
+					smu->thermal_range.software_shutdown_temp_offset;
+				high = min_t(typeof(high),
+					     SMU_THERMAL_MAXIMUM_ALERT_TEMP,
+					     high);
+				dev_emerg(adev->dev, "Reduce soft CTF limit to %d (by an offset %d)\n",
+							high,
+							smu->thermal_range.software_shutdown_temp_offset);
+
+				data = RREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL);
+				data = REG_SET_FIELD(data, THM_THERMAL_INT_CTRL,
+							DIG_THERM_INTH,
+							(high & 0xff));
+				data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK);
+				WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data);
+				break;
+			case 0x9:
+				high = min_t(typeof(high),
+					     SMU_THERMAL_MAXIMUM_ALERT_TEMP,
+					     smu->thermal_range.software_shutdown_temp);
+				dev_emerg(adev->dev, "Recover soft CTF limit to %d\n", high);
+
+				data = RREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL);
+				data = REG_SET_FIELD(data, THM_THERMAL_INT_CTRL,
+							DIG_THERM_INTH,
+							(high & 0xff));
+				data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK);
+				WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data);
+				break;
 			}
 		}
 	}
@@ -141,6 +141,9 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(PrepareMp1ForUnload,		PPSMC_MSG_PrepareMp1ForUnload,         0),
 	MSG_MAP(DFCstateControl,		PPSMC_MSG_SetExternalClientDfCstateAllow, 0),
 	MSG_MAP(ArmD3,				PPSMC_MSG_ArmD3,                       0),
+	MSG_MAP(SetNumBadMemoryPagesRetired,	PPSMC_MSG_SetNumBadMemoryPagesRetired,   0),
+	MSG_MAP(SetBadMemoryPagesRetiredFlagsPerChannel,
+			    PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel,   0),
 };

 static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = {
@@ -1838,6 +1841,40 @@ static void smu_v13_0_0_set_smu_mailbox_registers(struct smu_context *smu)
 	smu->debug_resp_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_SMN_C2PMSG_54);
 }

+static int smu_v13_0_0_smu_send_bad_mem_page_num(struct smu_context *smu,
+		uint32_t size)
+{
+	int ret = 0;
+
+	/* message SMU to update the bad page number on SMUBUS */
+	ret = smu_cmn_send_smc_msg_with_param(smu,
+					  SMU_MSG_SetNumBadMemoryPagesRetired,
+					  size, NULL);
+	if (ret)
+		dev_err(smu->adev->dev,
+			  "[%s] failed to message SMU to update bad memory pages number\n",
+			  __func__);
+
+	return ret;
+}
+
+static int smu_v13_0_0_send_bad_mem_channel_flag(struct smu_context *smu,
+		uint32_t size)
+{
+	int ret = 0;
+
+	/* message SMU to update the bad channel info on SMUBUS */
+	ret = smu_cmn_send_smc_msg_with_param(smu,
+				  SMU_MSG_SetBadMemoryPagesRetiredFlagsPerChannel,
+				  size, NULL);
+	if (ret)
+		dev_err(smu->adev->dev,
+			  "[%s] failed to message SMU to update bad memory pages channel info\n",
+			  __func__);
+
+	return ret;
+}
+
 static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask,
 	.set_default_dpm_table = smu_v13_0_0_set_default_dpm_table,
@@ -1908,6 +1945,8 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.mode1_reset = smu_v13_0_0_mode1_reset,
 	.set_mp1_state = smu_v13_0_0_set_mp1_state,
 	.set_df_cstate = smu_v13_0_0_set_df_cstate,
+	.send_hbm_bad_pages_num = smu_v13_0_0_smu_send_bad_mem_page_num,
+	.send_hbm_bad_channel_flag = smu_v13_0_0_send_bad_mem_channel_flag,
 };

 void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu)
@@ -1223,6 +1223,7 @@ static int smu_v13_0_7_get_thermal_temperature_range(struct smu_context *smu,
 	range->mem_emergency_max = (pptable->SkuTable.TemperatureLimit[TEMP_MEM] + CTF_OFFSET_MEM)*
 		SMU_TEMPERATURE_UNITS_PER_CENTIGRADES;
 	range->software_shutdown_temp = powerplay_table->software_shutdown_temp;
+	range->software_shutdown_temp_offset = pptable->SkuTable.FanAbnormalTempLimitOffset;

 	return 0;
 }