1 files changed, 157 insertions, 124 deletions

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 7f77f2338fa7..77277d90b6e2 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -146,14 +146,6 @@ MODULE_FIRMWARE(FIRMWARE_NAVI12_DMCU);
 /* Number of bytes in PSP footer for firmware. */
 #define PSP_FOOTER_BYTES 0x100
 
-/*
- * DMUB Async to Sync Mechanism Status
- */
-#define DMUB_ASYNC_TO_SYNC_ACCESS_FAIL 1
-#define DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT 2
-#define DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS 3
-#define DMUB_ASYNC_TO_SYNC_ACCESS_INVALID 4
-
 /**
  * DOC: overview
  *
@@ -1104,7 +1096,7 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 	/* Initialize hardware. */
 	memset(&hw_params, 0, sizeof(hw_params));
 	hw_params.fb_base = adev->gmc.fb_start;
-	hw_params.fb_offset = adev->gmc.aper_base;
+	hw_params.fb_offset = adev->vm_manager.vram_base_offset;
 
 	/* backdoor load firmware and trigger dmub running */
 	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
@@ -1226,7 +1218,7 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_
 	pa_config->system_aperture.agp_top = (uint64_t)agp_top << 24;
 
 	pa_config->system_aperture.fb_base = adev->gmc.fb_start;
-	pa_config->system_aperture.fb_offset = adev->gmc.aper_base;
+	pa_config->system_aperture.fb_offset = adev->vm_manager.vram_base_offset;
 	pa_config->system_aperture.fb_top = adev->gmc.fb_end;
 
 	pa_config->gart_config.page_table_start_addr = page_table_start.quad_part << 12;
@@ -1441,6 +1433,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	memset(&init_params, 0, sizeof(init_params));
 #endif
 
+	mutex_init(&adev->dm.dpia_aux_lock);
 	mutex_init(&adev->dm.dc_lock);
 	mutex_init(&adev->dm.audio_lock);
 
@@ -1805,6 +1798,7 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
 
 	mutex_destroy(&adev->dm.audio_lock);
 	mutex_destroy(&adev->dm.dc_lock);
+	mutex_destroy(&adev->dm.dpia_aux_lock);
 
 	return;
 }
@@ -4875,6 +4869,35 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev,
 	return 0;
 }
 
+static inline void fill_dc_dirty_rect(struct drm_plane *plane,
+				      struct rect *dirty_rect, int32_t x,
+				      int32_t y, int32_t width, int32_t height,
+				      int *i, bool ffu)
+{
+	if (*i > DC_MAX_DIRTY_RECTS)
+		return;
+
+	if (*i == DC_MAX_DIRTY_RECTS)
+		goto out;
+
+	dirty_rect->x = x;
+	dirty_rect->y = y;
+	dirty_rect->width = width;
+	dirty_rect->height = height;
+
+	if (ffu)
+		drm_dbg(plane->dev,
+			"[PLANE:%d] PSR FFU dirty rect size (%d, %d)\n",
+			plane->base.id, width, height);
+	else
+		drm_dbg(plane->dev,
+			"[PLANE:%d] PSR SU dirty rect at (%d, %d) size (%d, %d)",
+			plane->base.id, x, y, width, height);
+
+out:
+	(*i)++;
+}
+
 /**
  * fill_dc_dirty_rects() - Fill DC dirty regions for PSR selective updates
  *
@@ -4895,10 +4918,6 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev,
  * addition, certain use cases - such as cursor and multi-plane overlay (MPO) -
  * implicitly provide damage clips without any client support via the plane
  * bounds.
- *
- * Today, amdgpu_dm only supports the MPO and cursor usecase.
- *
- * TODO: Also enable for FB_DAMAGE_CLIPS
  */
 static void fill_dc_dirty_rects(struct drm_plane *plane,
 				struct drm_plane_state *old_plane_state,
@@ -4909,12 +4928,11 @@ static void fill_dc_dirty_rects(struct drm_plane *plane,
 	struct dm_crtc_state *dm_crtc_state = to_dm_crtc_state(crtc_state);
 	struct rect *dirty_rects = flip_addrs->dirty_rects;
 	uint32_t num_clips;
+	struct drm_mode_rect *clips;
 	bool bb_changed;
 	bool fb_changed;
 	uint32_t i = 0;
 
-	flip_addrs->dirty_rect_count = 0;
-
 	/*
 	 * Cursor plane has it's own dirty rect update interface. See
 	 * dcn10_dmub_update_cursor_data and dmub_cmd_update_cursor_info_data
@@ -4922,20 +4940,20 @@ static void fill_dc_dirty_rects(struct drm_plane *plane,
 	if (plane->type == DRM_PLANE_TYPE_CURSOR)
 		return;
 
-	/*
-	 * Today, we only consider MPO use-case for PSR SU. If MPO not
-	 * requested, and there is a plane update, do FFU.
-	 */
+	num_clips = drm_plane_get_damage_clips_count(new_plane_state);
+	clips = drm_plane_get_damage_clips(new_plane_state);
+
 	if (!dm_crtc_state->mpo_requested) {
-		dirty_rects[0].x = 0;
-		dirty_rects[0].y = 0;
-		dirty_rects[0].width = dm_crtc_state->base.mode.crtc_hdisplay;
-		dirty_rects[0].height = dm_crtc_state->base.mode.crtc_vdisplay;
-		flip_addrs->dirty_rect_count = 1;
-		DRM_DEBUG_DRIVER("[PLANE:%d] PSR FFU dirty rect size (%d, %d)\n",
-				 new_plane_state->plane->base.id,
-				 dm_crtc_state->base.mode.crtc_hdisplay,
-				 dm_crtc_state->base.mode.crtc_vdisplay);
+		if (!num_clips || num_clips > DC_MAX_DIRTY_RECTS)
+			goto ffu;
+
+		for (; flip_addrs->dirty_rect_count < num_clips; clips++)
+			fill_dc_dirty_rect(new_plane_state->plane,
+					   &dirty_rects[i], clips->x1,
+					   clips->y1, clips->x2 - clips->x1,
+					   clips->y2 - clips->y1,
+					   &flip_addrs->dirty_rect_count,
+					   false);
 		return;
 	}
 
@@ -4946,7 +4964,6 @@ static void fill_dc_dirty_rects(struct drm_plane *plane,
 	 * If plane is moved or resized, also add old bounding box to dirty
 	 * rects.
 	 */
-	num_clips = drm_plane_get_damage_clips_count(new_plane_state);
 	fb_changed = old_plane_state->fb->base.id !=
 		     new_plane_state->fb->base.id;
 	bb_changed = (old_plane_state->crtc_x != new_plane_state->crtc_x ||
@@ -4954,36 +4971,51 @@ static void fill_dc_dirty_rects(struct drm_plane *plane,
 		      old_plane_state->crtc_w != new_plane_state->crtc_w ||
 		      old_plane_state->crtc_h != new_plane_state->crtc_h);
 
-	DRM_DEBUG_DRIVER("[PLANE:%d] PSR bb_changed:%d fb_changed:%d num_clips:%d\n",
-			 new_plane_state->plane->base.id,
-			 bb_changed, fb_changed, num_clips);
+	drm_dbg(plane->dev,
+		"[PLANE:%d] PSR bb_changed:%d fb_changed:%d num_clips:%d\n",
+		new_plane_state->plane->base.id,
+		bb_changed, fb_changed, num_clips);
 
-	if (num_clips || fb_changed || bb_changed) {
-		dirty_rects[i].x = new_plane_state->crtc_x;
-		dirty_rects[i].y = new_plane_state->crtc_y;
-		dirty_rects[i].width = new_plane_state->crtc_w;
-		dirty_rects[i].height = new_plane_state->crtc_h;
-		DRM_DEBUG_DRIVER("[PLANE:%d] PSR SU dirty rect at (%d, %d) size (%d, %d)\n",
-				 new_plane_state->plane->base.id,
-				 dirty_rects[i].x, dirty_rects[i].y,
-				 dirty_rects[i].width, dirty_rects[i].height);
-		i += 1;
-	}
-
-	/* Add old plane bounding-box if plane is moved or resized */
 	if (bb_changed) {
-		dirty_rects[i].x = old_plane_state->crtc_x;
-		dirty_rects[i].y = old_plane_state->crtc_y;
-		dirty_rects[i].width = old_plane_state->crtc_w;
-		dirty_rects[i].height = old_plane_state->crtc_h;
-		DRM_DEBUG_DRIVER("[PLANE:%d] PSR SU dirty rect at (%d, %d) size (%d, %d)\n",
-				old_plane_state->plane->base.id,
-				dirty_rects[i].x, dirty_rects[i].y,
-				dirty_rects[i].width, dirty_rects[i].height);
-		i += 1;
-	}
+		fill_dc_dirty_rect(new_plane_state->plane, &dirty_rects[i],
+				   new_plane_state->crtc_x,
+				   new_plane_state->crtc_y,
+				   new_plane_state->crtc_w,
+				   new_plane_state->crtc_h, &i, false);
+
+		/* Add old plane bounding-box if plane is moved or resized */
+		fill_dc_dirty_rect(new_plane_state->plane, &dirty_rects[i],
+				   old_plane_state->crtc_x,
+				   old_plane_state->crtc_y,
+				   old_plane_state->crtc_w,
+				   old_plane_state->crtc_h, &i, false);
+	}
+
+	if (num_clips) {
+		for (; i < num_clips; clips++)
+			fill_dc_dirty_rect(new_plane_state->plane,
+					   &dirty_rects[i], clips->x1,
+					   clips->y1, clips->x2 - clips->x1,
+					   clips->y2 - clips->y1, &i, false);
+	} else if (fb_changed && !bb_changed) {
+		fill_dc_dirty_rect(new_plane_state->plane, &dirty_rects[i],
+				   new_plane_state->crtc_x,
+				   new_plane_state->crtc_y,
+				   new_plane_state->crtc_w,
+				   new_plane_state->crtc_h, &i, false);
+	}
+
+	if (i > DC_MAX_DIRTY_RECTS)
+		goto ffu;
 
 	flip_addrs->dirty_rect_count = i;
+	return;
+
+ffu:
+	fill_dc_dirty_rect(new_plane_state->plane, &dirty_rects[0], 0, 0,
+			   dm_crtc_state->base.mode.crtc_hdisplay,
+			   dm_crtc_state->base.mode.crtc_vdisplay,
+			   &flip_addrs->dirty_rect_count, true);
 }
 
 static void update_stream_scaling_settings(const struct drm_display_mode *mode,
@@ -10204,91 +10236,92 @@ uint32_t dm_read_reg_func(const struct dc_context *ctx, uint32_t address,
 	return value;
 }
 
-static int amdgpu_dm_set_dmub_async_sync_status(bool is_cmd_aux,
-						struct dc_context *ctx,
-						uint8_t status_type,
-						uint32_t *operation_result)
+int amdgpu_dm_process_dmub_aux_transfer_sync(
+		struct dc_context *ctx,
+		unsigned int link_index,
+		struct aux_payload *payload,
+		enum aux_return_code_type *operation_result)
 {
 	struct amdgpu_device *adev = ctx->driver_context;
-	int return_status = -1;
 	struct dmub_notification *p_notify = adev->dm.dmub_notify;
+	int ret = -1;
 
-	if (is_cmd_aux) {
-		if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS) {
-			return_status = p_notify->aux_reply.length;
-			*operation_result = p_notify->result;
-		} else if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT) {
-			*operation_result = AUX_RET_ERROR_TIMEOUT;
-		} else if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_FAIL) {
-			*operation_result = AUX_RET_ERROR_ENGINE_ACQUIRE;
-		} else if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_INVALID) {
-			*operation_result = AUX_RET_ERROR_INVALID_REPLY;
-		} else {
-			*operation_result = AUX_RET_ERROR_UNKNOWN;
+	mutex_lock(&adev->dm.dpia_aux_lock);
+	if (!dc_process_dmub_aux_transfer_async(ctx->dc, link_index, payload)) {
+		*operation_result = AUX_RET_ERROR_ENGINE_ACQUIRE;
+		goto out;
+ 	}
+
+	if (!wait_for_completion_timeout(&adev->dm.dmub_aux_transfer_done, 10 * HZ)) {
+		DRM_ERROR("wait_for_completion_timeout timeout!");
+		*operation_result = AUX_RET_ERROR_TIMEOUT;
+		goto out;
+	}
+
+	if (p_notify->result != AUX_RET_SUCCESS) {
+		/*
+		 * Transient states before tunneling is enabled could
+		 * lead to this error. We can ignore this for now.
+		 */
+		if (p_notify->result != AUX_RET_ERROR_PROTOCOL_ERROR) {
+			DRM_WARN("DPIA AUX failed on 0x%x(%d), error %d\n",
+					payload->address, payload->length,
+					p_notify->result);
 		}
-	} else {
-		if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS) {
-			return_status = 0;
-			*operation_result = p_notify->sc_status;
-		} else {
-			*operation_result = SET_CONFIG_UNKNOWN_ERROR;
+		*operation_result = AUX_RET_ERROR_INVALID_REPLY;
+		goto out;
+	}
+
+
+	payload->reply[0] = adev->dm.dmub_notify->aux_reply.command;
+	if (!payload->write && p_notify->aux_reply.length &&
+			(payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) {
+
+		if (payload->length != p_notify->aux_reply.length) {
+			DRM_WARN("invalid read length %d from DPIA AUX 0x%x(%d)!\n",
+				p_notify->aux_reply.length,
+					payload->address, payload->length);
+			*operation_result = AUX_RET_ERROR_INVALID_REPLY;
+			goto out;
 		}
+
+		memcpy(payload->data, p_notify->aux_reply.data,
+				p_notify->aux_reply.length);
 	}
 
-	return return_status;
+	/* success */
+	ret = p_notify->aux_reply.length;
+	*operation_result = p_notify->result;
+out:
+	mutex_unlock(&adev->dm.dpia_aux_lock);
+	return ret;
 }
 
-int amdgpu_dm_process_dmub_aux_transfer_sync(bool is_cmd_aux, struct dc_context *ctx,
-	unsigned int link_index, void *cmd_payload, void *operation_result)
+int amdgpu_dm_process_dmub_set_config_sync(
+		struct dc_context *ctx,
+		unsigned int link_index,
+		struct set_config_cmd_payload *payload,
+		enum set_config_status *operation_result)
 {
 	struct amdgpu_device *adev = ctx->driver_context;
-	int ret = 0;
+	bool is_cmd_complete;
+	int ret;
 
-	if (is_cmd_aux) {
-		dc_process_dmub_aux_transfer_async(ctx->dc,
-			link_index, (struct aux_payload *)cmd_payload);
-	} else if (dc_process_dmub_set_config_async(ctx->dc, link_index,
-					(struct set_config_cmd_payload *)cmd_payload,
-					adev->dm.dmub_notify)) {
-		return amdgpu_dm_set_dmub_async_sync_status(is_cmd_aux,
-					ctx, DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS,
-					(uint32_t *)operation_result);
-	}
+	mutex_lock(&adev->dm.dpia_aux_lock);
+	is_cmd_complete = dc_process_dmub_set_config_async(ctx->dc,
+			link_index, payload, adev->dm.dmub_notify);
 
-	ret = wait_for_completion_timeout(&adev->dm.dmub_aux_transfer_done, 10 * HZ);
-	if (ret == 0) {
+	if (is_cmd_complete || wait_for_completion_timeout(&adev->dm.dmub_aux_transfer_done, 10 * HZ)) {
+		ret = 0;
+		*operation_result = adev->dm.dmub_notify->sc_status;
+	} else {
 		DRM_ERROR("wait_for_completion_timeout timeout!");
-		return amdgpu_dm_set_dmub_async_sync_status(is_cmd_aux,
-				ctx, DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT,
-				(uint32_t *)operation_result);
-	}
-
-	if (is_cmd_aux) {
-		if (adev->dm.dmub_notify->result == AUX_RET_SUCCESS) {
-			struct aux_payload *payload = (struct aux_payload *)cmd_payload;
-
-			payload->reply[0] = adev->dm.dmub_notify->aux_reply.command;
-			if (!payload->write && adev->dm.dmub_notify->aux_reply.length &&
-			    payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK) {
-
-				if (payload->length != adev->dm.dmub_notify->aux_reply.length) {
-					DRM_WARN("invalid read from DPIA AUX %x(%d) got length %d!\n",
-							payload->address, payload->length,
-							adev->dm.dmub_notify->aux_reply.length);
-					return amdgpu_dm_set_dmub_async_sync_status(is_cmd_aux, ctx,
-							DMUB_ASYNC_TO_SYNC_ACCESS_INVALID,
-							(uint32_t *)operation_result);
-				}
-
-				memcpy(payload->data, adev->dm.dmub_notify->aux_reply.data,
-				       adev->dm.dmub_notify->aux_reply.length);
-			}
-		}
+		ret = -1;
+		*operation_result = SET_CONFIG_UNKNOWN_ERROR;
 	}
 
-	return amdgpu_dm_set_dmub_async_sync_status(is_cmd_aux,
-			ctx, DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS,
-			(uint32_t *)operation_result);
+	mutex_unlock(&adev->dm.dpia_aux_lock);
+	return ret;
 }
 
 /*