Merge tag 'amd-drm-fixes-5.11-2021-02-03' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-5.11-2021-02-03:

amdgpu:
- Fix retry in gem create
- Vangogh fixes
- Fix for display from shared buffers
- Various display fixes

amdkfd:
- Fix regression in buffer free

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210204041300.4425-1-alexander.deucher@amd.com

12 files changed, 92 insertions, 56 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 2d991da2cead..d1ed4f8df2b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -26,6 +26,7 @@
 #include <linux/sched/task.h>
 
 #include "amdgpu_object.h"
+#include "amdgpu_gem.h"
 #include "amdgpu_vm.h"
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_dma_buf.h"
@@ -1152,7 +1153,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	struct sg_table *sg = NULL;
 	uint64_t user_addr = 0;
 	struct amdgpu_bo *bo;
-	struct amdgpu_bo_param bp;
+	struct drm_gem_object *gobj;
 	u32 domain, alloc_domain;
 	u64 alloc_flags;
 	int ret;
@@ -1220,19 +1221,14 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s\n",
 			va, size, domain_string(alloc_domain));
 
-	memset(&bp, 0, sizeof(bp));
-	bp.size = size;
-	bp.byte_align = 1;
-	bp.domain = alloc_domain;
-	bp.flags = alloc_flags;
-	bp.type = bo_type;
-	bp.resv = NULL;
-	ret = amdgpu_bo_create(adev, &bp, &bo);
+	ret = amdgpu_gem_object_create(adev, size, 1, alloc_domain, alloc_flags,
+				       bo_type, NULL, &gobj);
 	if (ret) {
 		pr_debug("Failed to create BO on domain %s. ret %d\n",
-				domain_string(alloc_domain), ret);
+			 domain_string(alloc_domain), ret);
 		goto err_bo_create;
 	}
+	bo = gem_to_amdgpu_bo(gobj);
 	if (bo_type == ttm_bo_type_sg) {
 		bo->tbo.sg = sg;
 		bo->tbo.ttm->sg = sg;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index f764803c53a4..48cb33e5b382 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -926,8 +926,10 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev,
 				       struct drm_file *file_priv,
 				       const struct drm_mode_fb_cmd2 *mode_cmd)
 {
-	struct drm_gem_object *obj;
 	struct amdgpu_framebuffer *amdgpu_fb;
+	struct drm_gem_object *obj;
+	struct amdgpu_bo *bo;
+	uint32_t domains;
 	int ret;
 
 	obj = drm_gem_object_lookup(file_priv, mode_cmd->handles[0]);
@@ -938,7 +940,9 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev,
 	}
 
 	/* Handle is imported dma-buf, so cannot be migrated to VRAM for scanout */
-	if (obj->import_attach) {
+	bo = gem_to_amdgpu_bo(obj);
+	domains = amdgpu_display_supported_domains(drm_to_adev(dev), bo->flags);
+	if (obj->import_attach && !(domains & AMDGPU_GEM_DOMAIN_GTT)) {
 		drm_dbg_kms(dev, "Cannot create framebuffer from imported dma_buf\n");
 		return ERR_PTR(-EINVAL);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index d0a1fee1f5f6..174a73eb23f0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -269,8 +269,8 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 		resv = vm->root.base.bo->tbo.base.resv;
 	}
 
-retry:
 	initial_domain = (u32)(0xffffffff & args->in.domains);
+retry:
 	r = amdgpu_gem_object_create(adev, size, args->in.alignment,
 				     initial_domain,
 				     flags, ttm_bo_type_device, resv, &gobj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 25ec4d57333f..b4c8e5d5c763 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -897,7 +897,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 		return -EINVAL;
 
 	/* A shared bo cannot be migrated to VRAM */
-	if (bo->prime_shared_count) {
+	if (bo->prime_shared_count || bo->tbo.base.import_attach) {
 		if (domain & AMDGPU_GEM_DOMAIN_GTT)
 			domain = AMDGPU_GEM_DOMAIN_GTT;
 		else
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 346963e3cf73..d86b42a36560 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -99,6 +99,10 @@
 #define mmGCR_GENERAL_CNTL_Sienna_Cichlid			0x1580
 #define mmGCR_GENERAL_CNTL_Sienna_Cichlid_BASE_IDX	0
 
+#define mmCGTS_TCC_DISABLE_Vangogh                0x5006
+#define mmCGTS_TCC_DISABLE_Vangogh_BASE_IDX       1
+#define mmCGTS_USER_TCC_DISABLE_Vangogh                0x5007
+#define mmCGTS_USER_TCC_DISABLE_Vangogh_BASE_IDX       1
 #define mmGOLDEN_TSC_COUNT_UPPER_Vangogh                0x0025
 #define mmGOLDEN_TSC_COUNT_UPPER_Vangogh_BASE_IDX       1
 #define mmGOLDEN_TSC_COUNT_LOWER_Vangogh                0x0026
@@ -4936,8 +4940,18 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev)
 static void gfx_v10_0_get_tcc_info(struct amdgpu_device *adev)
 {
 	/* TCCs are global (not instanced). */
-	uint32_t tcc_disable = RREG32_SOC15(GC, 0, mmCGTS_TCC_DISABLE) |
-			       RREG32_SOC15(GC, 0, mmCGTS_USER_TCC_DISABLE);
+	uint32_t tcc_disable;
+
+	switch (adev->asic_type) {
+	case CHIP_VANGOGH:
+		tcc_disable = RREG32_SOC15(GC, 0, mmCGTS_TCC_DISABLE_Vangogh) |
+				RREG32_SOC15(GC, 0, mmCGTS_USER_TCC_DISABLE_Vangogh);
+		break;
+	default:
+		tcc_disable = RREG32_SOC15(GC, 0, mmCGTS_TCC_DISABLE) |
+				RREG32_SOC15(GC, 0, mmCGTS_USER_TCC_DISABLE);
+		break;
+	}
 
 	adev->gfx.config.tcc_disabled_mask =
 		REG_GET_FIELD(tcc_disable, CGTS_TCC_DISABLE, TCC_DISABLE) |
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 c6da89df055d..961abf1cf040 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1833,8 +1833,8 @@ static void emulated_link_detect(struct dc_link *link)
 	link->type = dc_connection_none;
 	prev_sink = link->local_sink;
 
-	if (prev_sink != NULL)
-		dc_sink_retain(prev_sink);
+	if (prev_sink)
+		dc_sink_release(prev_sink);
 
 	switch (link->connector_signal) {
 	case SIGNAL_TYPE_HDMI_TYPE_A: {
@@ -1934,7 +1934,7 @@ static void dm_gpureset_commit_state(struct dc_state *dc_state,
 		dc_commit_updates_for_stream(
 			dm->dc, bundle->surface_updates,
 			dc_state->stream_status->plane_count,
-			dc_state->streams[k], &bundle->stream_update, dc_state);
+			dc_state->streams[k], &bundle->stream_update);
 	}
 
 cleanup:
@@ -1965,8 +1965,7 @@ static void dm_set_dpms_off(struct dc_link *link)
 
 	stream_update.stream = stream_state;
 	dc_commit_updates_for_stream(stream_state->ctx->dc, NULL, 0,
-				     stream_state, &stream_update,
-				     stream_state->ctx->dc->current_state);
+				     stream_state, &stream_update);
 	mutex_unlock(&adev->dm.dc_lock);
 }
 
@@ -2330,8 +2329,10 @@ void amdgpu_dm_update_connector_after_detect(
 		 * TODO: check if we still need the S3 mode update workaround.
 		 * If yes, put it here.
 		 */
-		if (aconnector->dc_sink)
+		if (aconnector->dc_sink) {
 			amdgpu_dm_update_freesync_caps(connector, NULL);
+			dc_sink_release(aconnector->dc_sink);
+		}
 
 		aconnector->dc_sink = sink;
 		dc_sink_retain(aconnector->dc_sink);
@@ -2347,8 +2348,6 @@ void amdgpu_dm_update_connector_after_detect(
 
 			drm_connector_update_edid_property(connector,
 							   aconnector->edid);
-			drm_add_edid_modes(connector, aconnector->edid);
-
 			if (aconnector->dc_link->aux_mode)
 				drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux,
 						    aconnector->edid);
@@ -7549,7 +7548,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 				    struct drm_crtc *pcrtc,
 				    bool wait_for_vblank)
 {
-	uint32_t i;
+	int i;
 	uint64_t timestamp_ns;
 	struct drm_plane *plane;
 	struct drm_plane_state *old_plane_state, *new_plane_state;
@@ -7590,7 +7589,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		amdgpu_dm_commit_cursors(state);
 
 	/* update planes when needed */
-	for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) {
+	for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) {
 		struct drm_crtc *crtc = new_plane_state->crtc;
 		struct drm_crtc_state *new_crtc_state;
 		struct drm_framebuffer *fb = new_plane_state->fb;
@@ -7813,8 +7812,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 						     bundle->surface_updates,
 						     planes_count,
 						     acrtc_state->stream,
-						     &bundle->stream_update,
-						     dc_state);
+						     &bundle->stream_update);
 
 		/**
 		 * Enable or disable the interrupts on the backend.
@@ -8150,13 +8148,13 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state);
 		struct dm_connector_state *dm_old_con_state = to_dm_connector_state(old_con_state);
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
-		struct dc_surface_update dummy_updates[MAX_SURFACES];
+		struct dc_surface_update surface_updates[MAX_SURFACES];
 		struct dc_stream_update stream_update;
 		struct dc_info_packet hdr_packet;
 		struct dc_stream_status *status = NULL;
 		bool abm_changed, hdr_changed, scaling_changed;
 
-		memset(&dummy_updates, 0, sizeof(dummy_updates));
+		memset(&surface_updates, 0, sizeof(surface_updates));
 		memset(&stream_update, 0, sizeof(stream_update));
 
 		if (acrtc) {
@@ -8213,16 +8211,15 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		 * To fix this, DC should permit updating only stream properties.
 		 */
 		for (j = 0; j < status->plane_count; j++)
-			dummy_updates[j].surface = status->plane_states[0];
+			surface_updates[j].surface = status->plane_states[j];
 
 
 		mutex_lock(&dm->dc_lock);
 		dc_commit_updates_for_stream(dm->dc,
-						     dummy_updates,
+						surface_updates,
 						     status->plane_count,
 						     dm_new_crtc_state->stream,
-						     &stream_update,
-						     dc_state);
+						     &stream_update);
 		mutex_unlock(&dm->dc_lock);
 	}
 
@@ -8359,14 +8356,14 @@ static int dm_force_atomic_commit(struct drm_connector *connector)
 
 	ret = PTR_ERR_OR_ZERO(conn_state);
 	if (ret)
-		goto err;
+		goto out;
 
 	/* Attach crtc to drm_atomic_state*/
 	crtc_state = drm_atomic_get_crtc_state(state, &disconnected_acrtc->base);
 
 	ret = PTR_ERR_OR_ZERO(crtc_state);
 	if (ret)
-		goto err;
+		goto out;
 
 	/* force a restore */
 	crtc_state->mode_changed = true;
@@ -8376,17 +8373,15 @@ static int dm_force_atomic_commit(struct drm_connector *connector)
 
 	ret = PTR_ERR_OR_ZERO(plane_state);
 	if (ret)
-		goto err;
-
+		goto out;
 
 	/* Call commit internally with the state we just constructed */
 	ret = drm_atomic_commit(state);
-	if (!ret)
-		return 0;
 
-err:
-	DRM_ERROR("Restoring old state failed with %i\n", ret);
+out:
 	drm_atomic_state_put(state);
+	if (ret)
+		DRM_ERROR("Restoring old state failed with %i\n", ret);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 8ab0b9060d2b..f2d8cf34be46 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -833,6 +833,9 @@ bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state,
 		if (computed_streams[i])
 			continue;
 
+		if (dcn20_remove_stream_from_ctx(stream->ctx->dc, dc_state, stream) != DC_OK)
+			return false;
+
 		mutex_lock(&aconnector->mst_mgr.lock);
 		if (!compute_mst_dsc_configs_for_link(state, dc_state, stream->link)) {
 			mutex_unlock(&aconnector->mst_mgr.lock);
@@ -850,7 +853,8 @@ bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state,
 		stream = dc_state->streams[i];
 
 		if (stream->timing.flags.DSC == 1)
-			dc_stream_add_dsc_to_resource(stream->ctx->dc, dc_state, stream);
+			if (dc_stream_add_dsc_to_resource(stream->ctx->dc, dc_state, stream) != DC_OK)
+				return false;
 	}
 
 	return true;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 58eb0d69873a..6cf1a5a2a5ec 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2679,8 +2679,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 		struct dc_surface_update *srf_updates,
 		int surface_count,
 		struct dc_stream_state *stream,
-		struct dc_stream_update *stream_update,
-		struct dc_state *state)
+		struct dc_stream_update *stream_update)
 {
 	const struct dc_stream_status *stream_status;
 	enum surface_update_type update_type;
@@ -2699,6 +2698,12 @@ void dc_commit_updates_for_stream(struct dc *dc,
 
 
 	if (update_type >= UPDATE_TYPE_FULL) {
+		struct dc_plane_state *new_planes[MAX_SURFACES];
+
+		memset(new_planes, 0, sizeof(new_planes));
+
+		for (i = 0; i < surface_count; i++)
+			new_planes[i] = srf_updates[i].surface;
 
 		/* initialize scratch memory for building context */
 		context = dc_create_state(dc);
@@ -2707,15 +2712,21 @@ void dc_commit_updates_for_stream(struct dc *dc,
 			return;
 		}
 
-		dc_resource_state_copy_construct(state, context);
+		dc_resource_state_copy_construct(
+				dc->current_state, context);
 
-		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i];
-			struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+		/*remove old surfaces from context */
+		if (!dc_rem_all_planes_for_stream(dc, stream, context)) {
+			DC_ERROR("Failed to remove streams for new validate context!\n");
+			return;
+		}
 
-			if (new_pipe->plane_state && new_pipe->plane_state != old_pipe->plane_state)
-				new_pipe->plane_state->force_full_update = true;
+		/* add surface to context */
+		if (!dc_add_all_planes_for_stream(dc, stream, new_planes, surface_count, context)) {
+			DC_ERROR("Failed to add streams for new validate context!\n");
+			return;
 		}
+
 	}
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
index f95bade59624..1e4794e2825c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
@@ -892,13 +892,13 @@ static uint32_t translate_training_aux_read_interval(uint32_t dpcd_aux_read_inte
 
 	switch (dpcd_aux_read_interval) {
 	case 0x01:
-		aux_rd_interval_us = 400;
+		aux_rd_interval_us = 4000;
 		break;
 	case 0x02:
-		aux_rd_interval_us = 4000;
+		aux_rd_interval_us = 8000;
 		break;
 	case 0x03:
-		aux_rd_interval_us = 8000;
+		aux_rd_interval_us = 12000;
 		break;
 	case 0x04:
 		aux_rd_interval_us = 16000;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index b7910976b81a..e243c01b9672 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -283,8 +283,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 		struct dc_surface_update *srf_updates,
 		int surface_count,
 		struct dc_stream_state *stream,
-		struct dc_stream_update *stream_update,
-		struct dc_state *state);
+		struct dc_stream_update *stream_update);
 /*
  * Log the current stream state.
  */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index b000b43a820d..674376428916 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -906,6 +906,8 @@ enum dcn20_clk_src_array_id {
 	DCN20_CLK_SRC_PLL0,
 	DCN20_CLK_SRC_PLL1,
 	DCN20_CLK_SRC_PLL2,
+	DCN20_CLK_SRC_PLL3,
+	DCN20_CLK_SRC_PLL4,
 	DCN20_CLK_SRC_TOTAL_DCN21
 };
 
@@ -2030,6 +2032,14 @@ static bool dcn21_resource_construct(
 			dcn21_clock_source_create(ctx, ctx->dc_bios,
 				CLOCK_SOURCE_COMBO_PHY_PLL2,
 				&clk_src_regs[2], false);
+	pool->base.clock_sources[DCN20_CLK_SRC_PLL3] =
+			dcn21_clock_source_create(ctx, ctx->dc_bios,
+				CLOCK_SOURCE_COMBO_PHY_PLL3,
+				&clk_src_regs[3], false);
+	pool->base.clock_sources[DCN20_CLK_SRC_PLL4] =
+			dcn21_clock_source_create(ctx, ctx->dc_bios,
+				CLOCK_SOURCE_COMBO_PHY_PLL4,
+				&clk_src_regs[4], false);
 
 	pool->base.clk_src_count = DCN20_CLK_SRC_TOTAL_DCN21;
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
index 5c1482d4ca43..92ad2cdbae10 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
@@ -591,14 +591,17 @@ static ssize_t vangogh_get_gpu_metrics(struct smu_context *smu,
 	gpu_metrics->average_socket_power = metrics.CurrentSocketPower;
 	gpu_metrics->average_cpu_power = metrics.Power[0];
 	gpu_metrics->average_soc_power = metrics.Power[1];
+	gpu_metrics->average_gfx_power = metrics.Power[2];
 	memcpy(&gpu_metrics->average_core_power[0],
 		&metrics.CorePower[0],
 		sizeof(uint16_t) * 8);
 
 	gpu_metrics->average_gfxclk_frequency = metrics.GfxclkFrequency;
 	gpu_metrics->average_socclk_frequency = metrics.SocclkFrequency;
+	gpu_metrics->average_uclk_frequency = metrics.MemclkFrequency;
 	gpu_metrics->average_fclk_frequency = metrics.MemclkFrequency;
 	gpu_metrics->average_vclk_frequency = metrics.VclkFrequency;
+	gpu_metrics->average_dclk_frequency = metrics.DclkFrequency;
 
 	memcpy(&gpu_metrics->current_coreclk[0],
 		&metrics.CoreFrequency[0],