Merge tag 'amd-drm-fixes-6.9-2024-05-01' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.9-2024-05-01:

amdgpu:
- Fix VRAM memory accounting
- DCN 3.1 fixes
- DCN 2.0 fix
- DCN 3.1.5 fix
- DCN 3.5 fix
- DCN 3.2.1 fix
- DP fixes
- Seamless boot fix
- Fix call order in amdgpu_ttm_move()
- Fix doorbell regression
- Disable panel replay temporarily

amdkfd:
- Flush wq before creating kfd process

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240501135054.1919108-1-alexander.deucher@amd.com

15 files changed, 138 insertions, 55 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 2131de36e3da..e4d4e55c08ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -220,7 +220,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	    (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
 	     kfd_mem_limit.max_ttm_mem_limit) ||
 	    (adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] + vram_needed >
-	     vram_size - reserved_for_pt)) {
+	     vram_size - reserved_for_pt - atomic64_read(&adev->vram_pin_size))) {
 		ret = -ENOMEM;
 		goto release;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index ce733e3cb35d..f6d503432a9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1243,14 +1243,18 @@ int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
  * amdgpu_bo_move_notify - notification about a memory move
  * @bo: pointer to a buffer object
  * @evict: if this move is evicting the buffer from the graphics address space
+ * @new_mem: new resource for backing the BO
  *
  * Marks the corresponding &amdgpu_bo buffer object as invalid, also performs
  * bookkeeping.
  * TTM driver callback which is called when ttm moves a buffer.
  */
-void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict)
+void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
+			   bool evict,
+			   struct ttm_resource *new_mem)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
+	struct ttm_resource *old_mem = bo->resource;
 	struct amdgpu_bo *abo;
 
 	if (!amdgpu_bo_is_amdgpu_bo(bo))
@@ -1262,12 +1266,12 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict)
 	amdgpu_bo_kunmap(abo);
 
 	if (abo->tbo.base.dma_buf && !abo->tbo.base.import_attach &&
-	    bo->resource->mem_type != TTM_PL_SYSTEM)
+	    old_mem && old_mem->mem_type != TTM_PL_SYSTEM)
 		dma_buf_move_notify(abo->tbo.base.dma_buf);
 
-	/* remember the eviction */
-	if (evict)
-		atomic64_inc(&adev->num_evictions);
+	/* move_notify is called before move happens */
+	trace_amdgpu_bo_move(abo, new_mem ? new_mem->mem_type : -1,
+			     old_mem ? old_mem->mem_type : -1);
 }
 
 void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index fa03d9e4874c..bc42ccbde659 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -328,7 +328,9 @@ int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata,
 int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
 			   size_t buffer_size, uint32_t *metadata_size,
 			   uint64_t *flags);
-void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict);
+void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
+			   bool evict,
+			   struct ttm_resource *new_mem);
 void amdgpu_bo_release_notify(struct ttm_buffer_object *bo);
 vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 1d71729e3f6b..109fe557a02b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -419,7 +419,7 @@ bool amdgpu_res_cpu_visible(struct amdgpu_device *adev,
 		return false;
 
 	if (res->mem_type == TTM_PL_SYSTEM || res->mem_type == TTM_PL_TT ||
-	    res->mem_type == AMDGPU_PL_PREEMPT)
+	    res->mem_type == AMDGPU_PL_PREEMPT || res->mem_type == AMDGPU_PL_DOORBELL)
 		return true;
 
 	if (res->mem_type != TTM_PL_VRAM)
@@ -481,14 +481,16 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 
 	if (!old_mem || (old_mem->mem_type == TTM_PL_SYSTEM &&
 			 bo->ttm == NULL)) {
+		amdgpu_bo_move_notify(bo, evict, new_mem);
 		ttm_bo_move_null(bo, new_mem);
-		goto out;
+		return 0;
 	}
 	if (old_mem->mem_type == TTM_PL_SYSTEM &&
 	    (new_mem->mem_type == TTM_PL_TT ||
 	     new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
+		amdgpu_bo_move_notify(bo, evict, new_mem);
 		ttm_bo_move_null(bo, new_mem);
-		goto out;
+		return 0;
 	}
 	if ((old_mem->mem_type == TTM_PL_TT ||
 	     old_mem->mem_type == AMDGPU_PL_PREEMPT) &&
@@ -498,9 +500,10 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 			return r;
 
 		amdgpu_ttm_backend_unbind(bo->bdev, bo->ttm);
+		amdgpu_bo_move_notify(bo, evict, new_mem);
 		ttm_resource_free(bo, &bo->resource);
 		ttm_bo_assign_mem(bo, new_mem);
-		goto out;
+		return 0;
 	}
 
 	if (old_mem->mem_type == AMDGPU_PL_GDS ||
@@ -512,8 +515,9 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 	    new_mem->mem_type == AMDGPU_PL_OA ||
 	    new_mem->mem_type == AMDGPU_PL_DOORBELL) {
 		/* Nothing to save here */
+		amdgpu_bo_move_notify(bo, evict, new_mem);
 		ttm_bo_move_null(bo, new_mem);
-		goto out;
+		return 0;
 	}
 
 	if (bo->type == ttm_bo_type_device &&
@@ -525,22 +529,23 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 		abo->flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 	}
 
-	if (adev->mman.buffer_funcs_enabled) {
-		if (((old_mem->mem_type == TTM_PL_SYSTEM &&
-		      new_mem->mem_type == TTM_PL_VRAM) ||
-		     (old_mem->mem_type == TTM_PL_VRAM &&
-		      new_mem->mem_type == TTM_PL_SYSTEM))) {
-			hop->fpfn = 0;
-			hop->lpfn = 0;
-			hop->mem_type = TTM_PL_TT;
-			hop->flags = TTM_PL_FLAG_TEMPORARY;
-			return -EMULTIHOP;
-		}
+	if (adev->mman.buffer_funcs_enabled &&
+	    ((old_mem->mem_type == TTM_PL_SYSTEM &&
+	      new_mem->mem_type == TTM_PL_VRAM) ||
+	     (old_mem->mem_type == TTM_PL_VRAM &&
+	      new_mem->mem_type == TTM_PL_SYSTEM))) {
+		hop->fpfn = 0;
+		hop->lpfn = 0;
+		hop->mem_type = TTM_PL_TT;
+		hop->flags = TTM_PL_FLAG_TEMPORARY;
+		return -EMULTIHOP;
+	}
 
+	amdgpu_bo_move_notify(bo, evict, new_mem);
+	if (adev->mman.buffer_funcs_enabled)
 		r = amdgpu_move_blit(bo, evict, new_mem, old_mem);
-	} else {
+	else
 		r = -ENODEV;
-	}
 
 	if (r) {
 		/* Check that all memory is CPU accessible */
@@ -555,11 +560,10 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 			return r;
 	}
 
-	trace_amdgpu_bo_move(abo, new_mem->mem_type, old_mem->mem_type);
-out:
-	/* update statistics */
+	/* update statistics after the move */
+	if (evict)
+		atomic64_inc(&adev->num_evictions);
 	atomic64_add(bo->base.size, &adev->num_bytes_moved);
-	amdgpu_bo_move_notify(bo, evict);
 	return 0;
 }
 
@@ -1559,7 +1563,7 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo,
 static void
 amdgpu_bo_delete_mem_notify(struct ttm_buffer_object *bo)
 {
-	amdgpu_bo_move_notify(bo, false);
+	amdgpu_bo_move_notify(bo, false, NULL);
 }
 
 static struct ttm_device_funcs amdgpu_bo_driver = {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 58c1fe542193..451bb058cc62 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -829,6 +829,14 @@ struct kfd_process *kfd_create_process(struct task_struct *thread)
 	if (process) {
 		pr_debug("Process already found\n");
 	} else {
+		/* If the process just called exec(3), it is possible that the
+		 * cleanup of the kfd_process (following the release of the mm
+		 * of the old process image) is still in the cleanup work queue.
+		 * Make sure to drain any job before trying to recreate any
+		 * resource for this process.
+		 */
+		flush_workqueue(kfd_process_wq);
+
 		process = create_process(thread);
 		if (IS_ERR(process))
 			goto out;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index f3f94d109726..d6e71aa808d8 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -4537,15 +4537,18 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	/* Determine whether to enable Replay support by default. */
 	if (!(amdgpu_dc_debug_mask & DC_DISABLE_REPLAY)) {
 		switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) {
-		case IP_VERSION(3, 1, 4):
-		case IP_VERSION(3, 1, 5):
-		case IP_VERSION(3, 1, 6):
-		case IP_VERSION(3, 2, 0):
-		case IP_VERSION(3, 2, 1):
-		case IP_VERSION(3, 5, 0):
-		case IP_VERSION(3, 5, 1):
-			replay_feature_enabled = true;
-			break;
+/*
+ * Disabled by default due to https://gitlab.freedesktop.org/drm/amd/-/issues/3344
+ *		case IP_VERSION(3, 1, 4):
+ *		case IP_VERSION(3, 1, 5):
+ *		case IP_VERSION(3, 1, 6):
+ *		case IP_VERSION(3, 2, 0):
+ *		case IP_VERSION(3, 2, 1):
+ *		case IP_VERSION(3, 5, 0):
+ *		case IP_VERSION(3, 5, 1):
+ *			replay_feature_enabled = true;
+ *			break;
+ */
 		default:
 			replay_feature_enabled = amdgpu_dc_feature_mask & DC_REPLAY_MASK;
 			break;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index eee4945653e2..c7715a17f388 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -1495,7 +1495,9 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -1596,7 +1598,9 @@ static ssize_t dp_dsc_clock_en_write(struct file *f, const char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -1681,7 +1685,9 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -1780,7 +1786,9 @@ static ssize_t dp_dsc_slice_width_write(struct file *f, const char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -1865,7 +1873,9 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -1964,7 +1974,9 @@ static ssize_t dp_dsc_slice_height_write(struct file *f, const char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2045,7 +2057,9 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2141,7 +2155,9 @@ static ssize_t dp_dsc_bits_per_pixel_write(struct file *f, const char __user *bu
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2220,7 +2236,9 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2276,7 +2294,9 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2347,7 +2367,9 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
@@ -2418,7 +2440,9 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf,
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe_ctx->stream &&
-		    pipe_ctx->stream->link == aconnector->dc_link)
+		    pipe_ctx->stream->link == aconnector->dc_link &&
+		    pipe_ctx->stream->sink &&
+		    pipe_ctx->stream->sink == aconnector->dc_sink)
 			break;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
index 05f392501c0a..ab31643b1096 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -2948,6 +2948,7 @@ static enum bp_result construct_integrated_info(
 				result = get_integrated_info_v2_1(bp, info);
 				break;
 			case 2:
+			case 3:
 				result = get_integrated_info_v2_2(bp, info);
 				break;
 			default:
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
index 644da4637320..5506cf9b3672 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
@@ -145,6 +145,10 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base,
 	 */
 	clk_mgr_base->clks.zstate_support = new_clocks->zstate_support;
 	if (safe_to_lower) {
+		if (clk_mgr_base->clks.dtbclk_en && !new_clocks->dtbclk_en) {
+			dcn315_smu_set_dtbclk(clk_mgr, false);
+			clk_mgr_base->clks.dtbclk_en = new_clocks->dtbclk_en;
+		}
 		/* check that we're not already in lower */
 		if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_LOW_POWER) {
 			display_count = dcn315_get_active_display_cnt_wa(dc, context);
@@ -160,6 +164,10 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base,
 			}
 		}
 	} else {
+		if (!clk_mgr_base->clks.dtbclk_en && new_clocks->dtbclk_en) {
+			dcn315_smu_set_dtbclk(clk_mgr, true);
+			clk_mgr_base->clks.dtbclk_en = new_clocks->dtbclk_en;
+		}
 		/* check that we're not already in D0 */
 		if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_MISSION_MODE) {
 			union display_idle_optimization_u idle_info = { 0 };
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index bec252e1dd27..e506e4f969ca 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -712,8 +712,12 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 					 * since we calculate mode support based on softmax being the max UCLK
 					 * frequency.
 					 */
-					dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK,
-							dc->clk_mgr->bw_params->dc_mode_softmax_memclk);
+					if (dc->debug.disable_dc_mode_overwrite) {
+						dcn30_smu_set_hard_max_by_freq(clk_mgr, PPCLK_UCLK, dc->clk_mgr->bw_params->max_memclk_mhz);
+						dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, dc->clk_mgr->bw_params->max_memclk_mhz);
+					} else
+						dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK,
+								dc->clk_mgr->bw_params->dc_mode_softmax_memclk);
 				} else {
 					dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, dc->clk_mgr->bw_params->max_memclk_mhz);
 				}
@@ -746,8 +750,13 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 		/* set UCLK to requested value if P-State switching is supported, or to re-enable P-State switching */
 		if (clk_mgr_base->clks.p_state_change_support &&
 				(update_uclk || !clk_mgr_base->clks.prev_p_state_change_support) &&
-				!dc->work_arounds.clock_update_disable_mask.uclk)
+				!dc->work_arounds.clock_update_disable_mask.uclk) {
+			if (dc->clk_mgr->dc_mode_softmax_enabled && dc->debug.disable_dc_mode_overwrite)
+				dcn30_smu_set_hard_max_by_freq(clk_mgr, PPCLK_UCLK,
+						max((int)dc->clk_mgr->bw_params->dc_mode_softmax_memclk, khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz)));
+
 			dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz));
+		}
 
 		if (clk_mgr_base->clks.num_ways != new_clocks->num_ways &&
 				clk_mgr_base->clks.num_ways > new_clocks->num_ways) {
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 03b554e912a2..d68c83e40d4d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1801,6 +1801,9 @@ bool dc_validate_boot_timing(const struct dc *dc,
 		return false;
 	}
 
+	if (link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED)
+		return false;
+
 	if (dc->link_srv->edp_is_ilr_optimization_required(link, crtc_timing)) {
 		DC_LOG_EVENT_LINK_TRAINING("Seamless boot disabled to optimize eDP link rate\n");
 		return false;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_link_encoder.c
index 5b7ad38f85e0..65e45a0b4ff3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_link_encoder.c
@@ -395,6 +395,12 @@ void dcn31_hpo_dp_link_enc_set_throttled_vcp_size(
 				x),
 			25));
 
+	// If y rounds up to integer, carry it over to x.
+	if (y >> 25) {
+		x += 1;
+		y = 0;
+	}
+
 	switch (stream_encoder_inst) {
 	case 0:
 		REG_SET_2(DP_DPHY_SYM32_VC_RATE_CNTL0, 0,
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
index deb6d162a2d5..7307b7b8d8ad 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
@@ -291,6 +291,7 @@ static struct _vcs_dpi_soc_bounding_box_st dcn3_15_soc = {
 	.do_urgent_latency_adjustment = false,
 	.urgent_latency_adjustment_fabric_clock_component_us = 0,
 	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+	.dispclk_dppclk_vco_speed_mhz = 2400.0,
 	.num_chans = 4,
 	.dummy_pstate_latency_us = 10.0
 };
@@ -438,6 +439,7 @@ static struct _vcs_dpi_soc_bounding_box_st dcn3_16_soc = {
 	.do_urgent_latency_adjustment = false,
 	.urgent_latency_adjustment_fabric_clock_component_us = 0,
 	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+	.dispclk_dppclk_vco_speed_mhz = 2500.0,
 };
 
 void dcn31_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
index 5491b707cec8..5a965c26bf20 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
@@ -270,7 +270,7 @@ static void set_usb4_req_bw_req(struct dc_link *link, int req_bw)
 
 	/* Error check whether requested and allocated are equal */
 	req_bw = requested_bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity);
-	if (req_bw == link->dpia_bw_alloc_config.allocated_bw) {
+	if (req_bw && (req_bw == link->dpia_bw_alloc_config.allocated_bw)) {
 		DC_LOG_ERROR("%s: Request bw equals to allocated bw for link(%d)\n",
 			__func__, link->link_index);
 	}
@@ -341,6 +341,14 @@ bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link)
 			ret = true;
 			init_usb4_bw_struct(link);
 			link->dpia_bw_alloc_config.bw_alloc_enabled = true;
+
+			/*
+			 * During DP tunnel creation, CM preallocates BW and reduces estimated BW of other
+			 * DPIA. CM release preallocation only when allocation is complete. Do zero alloc
+			 * to make the CM to release preallocation and update estimated BW correctly for
+			 * all DPIAs per host router
+			 */
+			link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0);
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
index a2387cea1af9..622214b365a2 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
@@ -2449,6 +2449,7 @@ static bool dcn20_resource_construct(
 	dc->caps.post_blend_color_processing = true;
 	dc->caps.force_dp_tps4_for_cp2520 = true;
 	dc->caps.extended_aux_timeout_support = true;
+	dc->caps.dmcub_support = true;
 
 	/* Color pipeline capabilities */
 	dc->caps.color.dpp.dcn_arch = 1;