Merge tag 'drm-fixes-2020-12-11' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Last week of fixes, just amdgpu and i915 collections. We had a i915
  regression reported by HJ Lu reported this morning, and this contains
  a fix for that he has tested.

  There are a fair few other fixes, but they are spread across the two
  drivers, and all fairly self contained.

  amdgpu:
   - Fan fix for CI asics
   - Fix a warning in possible_crtcs
   - Build fix for when debugfs is disabled
   - Display overflow fix
   - Display watermark fixes for Renoir
   - SDMA 5.2 fix
   - Stolen vga memory regression fix
   - Power profile fixes
   - Fix a regression from removal of GEM and PRIME callbacks

  amdkfd:
   - Fix a memory leak in dmabuf import

  i915:
   - rc7 regression fix for modesetting
   - vdsc/dp slice fixes
   - gen9 mocs entries fix
   - preemption timeout fix
   - unsigned compare against 0 fix
   - selftest fix
   - submission error propogatig fix
   - request flow suspend fix"

* tag 'drm-fixes-2020-12-11' of git://anongit.freedesktop.org/drm/drm:
  drm/i915/display: Go softly softly on initial modeset failure
  drm/amd/pm: typo fix (CUSTOM -> COMPUTE)
  drm/amdgpu: Initialise drm_gem_object_funcs for imported BOs
  drm/amdgpu: fix size calculation with stolen vga memory
  drm/amd/pm: update smu10.h WORKLOAD_PPLIB setting for raven
  drm/amdkfd: Fix leak in dmabuf import
  drm/amdgpu: fix sdma instance fw version and feature version init
  drm/amd/display: Add wm table for Renoir
  drm/amd/display: Prevent bandwidth overflow
  drm/amdgpu: fix debugfs creation/removal, again
  drm/amdgpu/disply: set num_crtc earlier
  drm/amdgpu/powerplay: parse fan table for CI asics
  drm/i915/gt: Declare gen9 has 64 mocs entries!
  drm/i915/display/dp: Compute the correct slice count for VDSC on DP
  drm/i915: fix size_t greater or equal to zero comparison
  drm/i915/gt: Cancel the preemption timeout on responding to it
  drm/i915/gt: Ignore repeated attempts to suspend request flow across reset
  drm/i915/gem: Propagate error from cancelled submit due to context closure
  drm/i915/gem: Check the correct variable in selftest

Author: torvalds

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

@@ -459,6 +459,7 @@ amdgpu_dma_buf_create_obj(struct drm_device *dev, struct dma_buf *dma_buf)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_bo *bo;
 	struct amdgpu_bo_param bp;
+	struct drm_gem_object *gobj;
 	int ret;
 
 	memset(&bp, 0, sizeof(bp));
@@ -469,17 +470,20 @@ amdgpu_dma_buf_create_obj(struct drm_device *dev, struct dma_buf *dma_buf)
 	bp.type = ttm_bo_type_sg;
 	bp.resv = resv;
 	dma_resv_lock(resv, NULL);
-	ret = amdgpu_bo_create(adev, &bp, &bo);
+	ret = amdgpu_gem_object_create(adev, dma_buf->size, PAGE_SIZE,
+			AMDGPU_GEM_DOMAIN_CPU,
+			0, ttm_bo_type_sg, resv, &gobj);
 	if (ret)
 		goto error;
 
+	bo = gem_to_amdgpu_bo(gobj);
 	bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
 	bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
 	if (dma_buf->ops != &amdgpu_dmabuf_ops)
 		bo->prime_shared_count = 1;
 
 	dma_resv_unlock(resv);
-	return &bo->tbo.base;
+	return gobj;
 
 error:
 	dma_resv_unlock(resv);

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

@@ -66,26 +66,12 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 	bp.type = type;
 	bp.resv = resv;
 	bp.preferred_domain = initial_domain;
-retry:
 	bp.flags = flags;
 	bp.domain = initial_domain;
 	r = amdgpu_bo_create(adev, &bp, &bo);
-	if (r) {
-		if (r != -ERESTARTSYS) {
-			if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) {
-				flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
-				goto retry;
-			}
-
-			if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) {
-				initial_domain |= AMDGPU_GEM_DOMAIN_GTT;
-				goto retry;
-			}
-			DRM_DEBUG("Failed to allocate GEM object (%ld, %d, %u, %d)\n",
-				  size, initial_domain, alignment, r);
-		}
+	if (r)
 		return r;
-	}
+
 	*obj = &bo->tbo.base;
 
 	return 0;
@@ -225,7 +211,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	uint64_t size = args->in.bo_size;
 	struct dma_resv *resv = NULL;
 	struct drm_gem_object *gobj;
-	uint32_t handle;
+	uint32_t handle, initial_domain;
 	int r;
 
 	/* reject invalid gem flags */
@@ -269,9 +255,28 @@ 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);
 	r = amdgpu_gem_object_create(adev, size, args->in.alignment,
-				     (u32)(0xffffffff & args->in.domains),
+				     initial_domain,
 				     flags, ttm_bo_type_device, resv, &gobj);
+	if (r) {
+		if (r != -ERESTARTSYS) {
+			if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) {
+				flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
+				goto retry;
+			}
+
+			if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) {
+				initial_domain |= AMDGPU_GEM_DOMAIN_GTT;
+				goto retry;
+			}
+			DRM_DEBUG("Failed to allocate GEM object (%llu, %d, %llu, %d)\n",
+				  size, initial_domain, args->in.alignment, r);
+		}
+		return r;
+	}
+
 	if (flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) {
 		if (!r) {
 			struct amdgpu_bo *abo = gem_to_amdgpu_bo(gobj);

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

@@ -499,6 +499,9 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev)
 	else
 		size = amdgpu_gmc_get_vbios_fb_size(adev);
 
+	if (adev->mman.keep_stolen_vga_memory)
+		size = max(size, (unsigned)AMDGPU_VBIOS_VGA_ALLOCATION);
+
 	/* set to 0 if the pre-OS buffer uses up most of vram */
 	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
 		size = 0;

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

@@ -1172,7 +1172,7 @@ static void amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
 			con->dir, &con->disable_ras_err_cnt_harvest);
 }
 
-void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
+static void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
 		struct ras_fs_if *head)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
@@ -1194,7 +1194,6 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
 
 void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
 {
-#if defined(CONFIG_DEBUG_FS)
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_manager *obj;
 	struct ras_fs_if fs_info;
@@ -1203,7 +1202,7 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
 	 * it won't be called in resume path, no need to check
 	 * suspend and gpu reset status
 	 */
-	if (!con)
+	if (!IS_ENABLED(CONFIG_DEBUG_FS) || !con)
 		return;
 
 	amdgpu_ras_debugfs_create_ctrl_node(adev);
@@ -1217,10 +1216,9 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
 			amdgpu_ras_debugfs_create(adev, &fs_info);
 		}
 	}
-#endif
 }
 
-void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
+static void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
 		struct ras_common_if *head)
 {
 	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);
@@ -1234,7 +1232,6 @@ void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
 
 static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
 {
-#if defined(CONFIG_DEBUG_FS)
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_manager *obj, *tmp;
 
@@ -1243,7 +1240,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
 	}
 
 	con->dir = NULL;
-#endif
 }
 /* debugfs end */
 
@@ -1291,7 +1287,8 @@ static int amdgpu_ras_fs_init(struct amdgpu_device *adev)
 
 static int amdgpu_ras_fs_fini(struct amdgpu_device *adev)
 {
-	amdgpu_ras_debugfs_remove_all(adev);
+	if (IS_ENABLED(CONFIG_DEBUG_FS))
+		amdgpu_ras_debugfs_remove_all(adev);
 	amdgpu_ras_sysfs_remove_all(adev);
 	return 0;
 }

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

@@ -607,14 +607,8 @@ int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
 int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
 		struct ras_common_if *head);
 
-void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
-		struct ras_fs_if *head);
-
 void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev);
 
-void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
-		struct ras_common_if *head);
-
 int amdgpu_ras_error_query(struct amdgpu_device *adev,
 		struct ras_query_if *info);
 

drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c

@@ -186,7 +186,7 @@ static int sdma_v5_2_init_microcode(struct amdgpu_device *adev)
 			if (err)
 				goto out;
 
-			err = sdma_v5_2_init_inst_ctx(&adev->sdma.instance[0]);
+			err = sdma_v5_2_init_inst_ctx(&adev->sdma.instance[i]);
 			if (err)
 				goto out;
 		}

drivers/gpu/drm/amd/amdkfd/kfd_chardev.c

@@ -1736,6 +1736,7 @@ static int kfd_ioctl_import_dmabuf(struct file *filep,
 	}
 
 	mutex_unlock(&p->mutex);
+	dma_buf_put(dmabuf);
 
 	args->handle = MAKE_HANDLE(args->gpu_id, idr_handle);
 
@@ -1745,6 +1746,7 @@ static int kfd_ioctl_import_dmabuf(struct file *filep,
 	amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem, NULL);
 err_unlock:
 	mutex_unlock(&p->mutex);
+	dma_buf_put(dmabuf);
 	return r;
 }
 

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

@@ -1058,9 +1058,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		goto error;
 	}
 
-	/* Update the actual used number of crtc */
-	adev->mode_info.num_crtc = adev->dm.display_indexes_num;
-
 	/* create fake encoders for MST */
 	dm_dp_create_fake_mst_encoders(adev);
 
@@ -3251,6 +3248,10 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	enum dc_connection_type new_connection_type = dc_connection_none;
 	const struct dc_plane_cap *plane;
 
+	dm->display_indexes_num = dm->dc->caps.max_streams;
+	/* Update the actual used number of crtc */
+	adev->mode_info.num_crtc = adev->dm.display_indexes_num;
+
 	link_cnt = dm->dc->caps.max_links;
 	if (amdgpu_dm_mode_config_init(dm->adev)) {
 		DRM_ERROR("DM: Failed to initialize mode config\n");
@@ -3312,8 +3313,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 			goto fail;
 		}
 
-	dm->display_indexes_num = dm->dc->caps.max_streams;
-
 	/* loops over all connectors on the board */
 	for (i = 0; i < link_cnt; i++) {
 		struct dc_link *link = NULL;

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c

@@ -579,7 +579,7 @@ static struct clk_bw_params rn_bw_params = {
 
 };
 
-static struct wm_table ddr4_wm_table = {
+static struct wm_table ddr4_wm_table_gs = {
 	.entries = {
 		{
 			.wm_inst = WM_A,
@@ -616,7 +616,7 @@ static struct wm_table ddr4_wm_table = {
 	}
 };
 
-static struct wm_table lpddr4_wm_table = {
+static struct wm_table lpddr4_wm_table_gs = {
 	.entries = {
 		{
 			.wm_inst = WM_A,
@@ -690,6 +690,80 @@ static struct wm_table lpddr4_wm_table_with_disabled_ppt = {
 	}
 };
 
+static struct wm_table ddr4_wm_table_rn = {
+	.entries = {
+		{
+			.wm_inst = WM_A,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 9.09,
+			.sr_enter_plus_exit_time_us = 10.14,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_B,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_C,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_D,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+	}
+};
+
+static struct wm_table lpddr4_wm_table_rn = {
+	.entries = {
+		{
+			.wm_inst = WM_A,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 7.32,
+			.sr_enter_plus_exit_time_us = 8.38,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_B,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 9.82,
+			.sr_enter_plus_exit_time_us = 11.196,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_C,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 9.89,
+			.sr_enter_plus_exit_time_us = 11.24,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_D,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 9.748,
+			.sr_enter_plus_exit_time_us = 11.102,
+			.valid = true,
+		},
+	}
+};
+
 static unsigned int find_dcfclk_for_voltage(struct dpm_clocks *clock_table, unsigned int voltage)
 {
 	int i;
@@ -771,6 +845,11 @@ void rn_clk_mgr_construct(
 	struct dc_debug_options *debug = &ctx->dc->debug;
 	struct dpm_clocks clock_table = { 0 };
 	enum pp_smu_status status = 0;
+	int is_green_sardine = 0;
+
+#if defined(CONFIG_DRM_AMD_DC_DCN)
+	is_green_sardine = ASICREV_IS_GREEN_SARDINE(ctx->asic_id.hw_internal_rev);
+#endif
 
 	clk_mgr->base.ctx = ctx;
 	clk_mgr->base.funcs = &dcn21_funcs;
@@ -811,10 +890,16 @@ void rn_clk_mgr_construct(
 			if (clk_mgr->periodic_retraining_disabled) {
 				rn_bw_params.wm_table = lpddr4_wm_table_with_disabled_ppt;
 			} else {
-				rn_bw_params.wm_table = lpddr4_wm_table;
+				if (is_green_sardine)
+					rn_bw_params.wm_table = lpddr4_wm_table_gs;
+				else
+					rn_bw_params.wm_table = lpddr4_wm_table_rn;
 			}
 		} else {
-			rn_bw_params.wm_table = ddr4_wm_table;
+			if (is_green_sardine)
+				rn_bw_params.wm_table = ddr4_wm_table_gs;
+			else
+				rn_bw_params.wm_table = ddr4_wm_table_rn;
 		}
 		/* Saved clocks configured at boot for debug purposes */
 		rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info);

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

@@ -3394,10 +3394,13 @@ uint32_t dc_bandwidth_in_kbps_from_timing(
 {
 	uint32_t bits_per_channel = 0;
 	uint32_t kbps;
+	struct fixed31_32 link_bw_kbps;
 
 	if (timing->flags.DSC) {
-		kbps = (timing->pix_clk_100hz * timing->dsc_cfg.bits_per_pixel);
-		kbps = kbps / 160 + ((kbps % 160) ? 1 : 0);
+		link_bw_kbps = dc_fixpt_from_int(timing->pix_clk_100hz);
+		link_bw_kbps = dc_fixpt_div_int(link_bw_kbps, 160);
+		link_bw_kbps = dc_fixpt_mul_int(link_bw_kbps, timing->dsc_cfg.bits_per_pixel);
+		kbps = dc_fixpt_ceil(link_bw_kbps);
 		return kbps;
 	}