Merge branch 'drm-fixes-5.0' of git://people.freedesktop.org/~agd5f/linux into drm-fixes

Fixes for 5.0:
- Fix KFD on ARM64
- Fix KFD topology with mixed APU and dGPU systems
- Powerplay fix for vega12
- DC Raven fixes
- Freesync fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190116224524.3314-1-alexander.deucher@amd.com

14 files changed, 65 insertions, 59 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index dafc645b2e4e..b083b219b1a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -531,17 +531,6 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev,
 	struct drm_gem_object *obj;
 	struct amdgpu_framebuffer *amdgpu_fb;
 	int ret;
-	int height;
-	struct amdgpu_device *adev = dev->dev_private;
-	int cpp = drm_format_plane_cpp(mode_cmd->pixel_format, 0);
-	int pitch = mode_cmd->pitches[0] / cpp;
-
-	pitch = amdgpu_align_pitch(adev, pitch, cpp, false);
-	if (mode_cmd->pitches[0] != pitch) {
-		DRM_DEBUG_KMS("Invalid pitch: expecting %d but got %d\n",
-			      pitch, mode_cmd->pitches[0]);
-		return ERR_PTR(-EINVAL);
-	}
 
 	obj = drm_gem_object_lookup(file_priv, mode_cmd->handles[0]);
 	if (obj ==  NULL) {
@@ -556,13 +545,6 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev,
 		return ERR_PTR(-EINVAL);
 	}
 
-	height = ALIGN(mode_cmd->height, 8);
-	if (obj->size < pitch * height) {
-		DRM_DEBUG_KMS("Invalid GEM size: expecting >= %d but got %zu\n",
-			      pitch * height, obj->size);
-		return ERR_PTR(-EINVAL);
-	}
-
 	amdgpu_fb = kzalloc(sizeof(*amdgpu_fb), GFP_KERNEL);
 	if (amdgpu_fb == NULL) {
 		drm_gem_object_put_unlocked(obj);
diff --git a/drivers/gpu/drm/amd/amdkfd/Kconfig b/drivers/gpu/drm/amd/amdkfd/Kconfig
index fbf0ee5201c3..c3613604a4f8 100644
--- a/drivers/gpu/drm/amd/amdkfd/Kconfig
+++ b/drivers/gpu/drm/amd/amdkfd/Kconfig
@@ -4,8 +4,8 @@
 
 config HSA_AMD
 	bool "HSA kernel driver for AMD GPU devices"
-	depends on DRM_AMDGPU && X86_64
-	imply AMD_IOMMU_V2
+	depends on DRM_AMDGPU && (X86_64 || ARM64)
+	imply AMD_IOMMU_V2 if X86_64
 	select MMU_NOTIFIER
 	help
 	  Enable this if you want to use HSA features on AMD GPU devices.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index b7bc7d7d048f..5d85ff341385 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -863,6 +863,7 @@ static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size,
 	return 0;
 }
 
+#if CONFIG_X86_64
 static int kfd_fill_iolink_info_for_cpu(int numa_node_id, int *avail_size,
 				uint32_t *num_entries,
 				struct crat_subtype_iolink *sub_type_hdr)
@@ -905,6 +906,7 @@ static int kfd_fill_iolink_info_for_cpu(int numa_node_id, int *avail_size,
 
 	return 0;
 }
+#endif
 
 /* kfd_create_vcrat_image_cpu - Create Virtual CRAT for CPU
  *
@@ -920,7 +922,9 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 	struct crat_subtype_generic *sub_type_hdr;
 	int avail_size = *size;
 	int numa_node_id;
+#ifdef CONFIG_X86_64
 	uint32_t entries = 0;
+#endif
 	int ret = 0;
 
 	if (!pcrat_image || avail_size < VCRAT_SIZE_FOR_CPU)
@@ -982,6 +986,7 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 			sub_type_hdr->length);
 
 		/* Fill in Subtype: IO Link */
+#ifdef CONFIG_X86_64
 		ret = kfd_fill_iolink_info_for_cpu(numa_node_id, &avail_size,
 				&entries,
 				(struct crat_subtype_iolink *)sub_type_hdr);
@@ -992,6 +997,9 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 
 		sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
 				sub_type_hdr->length * entries);
+#else
+		pr_info("IO link not available for non x86 platforms\n");
+#endif
 
 		crat_table->num_domains++;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 5f5b2acedbac..09da91644f9f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1093,8 +1093,6 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
  *		the GPU device is not already present in the topology device
  *		list then return NULL. This means a new topology device has to
  *		be created for this GPU.
- * TODO: Rather than assiging @gpu to first topology device withtout
- *		gpu attached, it will better to have more stringent check.
  */
 static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 {
@@ -1102,12 +1100,20 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 	struct kfd_topology_device *out_dev = NULL;
 
 	down_write(&topology_lock);
-	list_for_each_entry(dev, &topology_device_list, list)
+	list_for_each_entry(dev, &topology_device_list, list) {
+		/* Discrete GPUs need their own topology device list
+		 * entries. Don't assign them to CPU/APU nodes.
+		 */
+		if (!gpu->device_info->needs_iommu_device &&
+		    dev->node_props.cpu_cores_count)
+			continue;
+
 		if (!dev->gpu && (dev->node_props.simd_count > 0)) {
 			dev->gpu = gpu;
 			out_dev = dev;
 			break;
 		}
+	}
 	up_write(&topology_lock);
 	return out_dev;
 }
@@ -1392,7 +1398,6 @@ int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev)
 
 static int kfd_cpumask_to_apic_id(const struct cpumask *cpumask)
 {
-	const struct cpuinfo_x86 *cpuinfo;
 	int first_cpu_of_numa_node;
 
 	if (!cpumask || cpumask == cpu_none_mask)
@@ -1400,9 +1405,11 @@ static int kfd_cpumask_to_apic_id(const struct cpumask *cpumask)
 	first_cpu_of_numa_node = cpumask_first(cpumask);
 	if (first_cpu_of_numa_node >= nr_cpu_ids)
 		return -1;
-	cpuinfo = &cpu_data(first_cpu_of_numa_node);
-
-	return cpuinfo->apicid;
+#ifdef CONFIG_X86_64
+	return cpu_data(first_cpu_of_numa_node).apicid;
+#else
+	return first_cpu_of_numa_node;
+#endif
 }
 
 /* kfd_numa_node_to_apic_id - Returns the APIC ID of the first logical processor
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 34f35e9a3c46..f4fa40c387d3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1772,7 +1772,7 @@ static int amdgpu_dm_backlight_update_status(struct backlight_device *bd)
 		+ caps.min_input_signal * 0x101;
 
 	if (dc_link_set_backlight_level(dm->backlight_link,
-			brightness, 0, 0))
+			brightness, 0))
 		return 0;
 	else
 		return 1;
@@ -5933,7 +5933,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		if (!drm_atomic_crtc_needs_modeset(new_crtc_state) &&
 		    !new_crtc_state->color_mgmt_changed &&
-		    !new_crtc_state->vrr_enabled)
+		    old_crtc_state->vrr_enabled == new_crtc_state->vrr_enabled)
 			continue;
 
 		if (!new_crtc_state->enable)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
index 52deacf39841..b0265dbebd4c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -2190,8 +2190,7 @@ int dc_link_get_backlight_level(const struct dc_link *link)
 
 bool dc_link_set_backlight_level(const struct dc_link *link,
 		uint32_t backlight_pwm_u16_16,
-		uint32_t frame_ramp,
-		const struct dc_stream_state *stream)
+		uint32_t frame_ramp)
 {
 	struct dc  *core_dc = link->ctx->dc;
 	struct abm *abm = core_dc->res_pool->abm;
@@ -2206,10 +2205,6 @@ bool dc_link_set_backlight_level(const struct dc_link *link,
 		(abm->funcs->set_backlight_level_pwm == NULL))
 		return false;
 
-	if (stream)
-		((struct dc_stream_state *)stream)->bl_pwm_level =
-				backlight_pwm_u16_16;
-
 	use_smooth_brightness = dmcu->funcs->is_dmcu_initialized(dmcu);
 
 	DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n",
@@ -2637,11 +2632,6 @@ void core_link_enable_stream(
 
 		if (dc_is_dp_signal(pipe_ctx->stream->signal))
 			enable_stream_features(pipe_ctx);
-
-		dc_link_set_backlight_level(pipe_ctx->stream->sink->link,
-				pipe_ctx->stream->bl_pwm_level,
-				0,
-				pipe_ctx->stream);
 	}
 
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index 29f19d57ff7a..b2243e0dad1f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -146,8 +146,7 @@ static inline struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_
  */
 bool dc_link_set_backlight_level(const struct dc_link *dc_link,
 		uint32_t backlight_pwm_u16_16,
-		uint32_t frame_ramp,
-		const struct dc_stream_state *stream);
+		uint32_t frame_ramp);
 
 int dc_link_get_backlight_level(const struct dc_link *dc_link);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index be34d638e15d..d70c9e1cda3d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -91,7 +91,6 @@ struct dc_stream_state {
 
 	/* DMCU info */
 	unsigned int abm_level;
-	unsigned int bl_pwm_level;
 
 	/* from core_stream struct */
 	struct dc_context *ctx;
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 4bf24758217f..8f09b8625c5d 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -1000,7 +1000,7 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
 
 		pipe_ctx->stream_res.audio->funcs->az_enable(pipe_ctx->stream_res.audio);
 
-		if (num_audio == 1 && pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+		if (num_audio >= 1 && pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
 			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
 			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
 		/* un-mute audio */
@@ -1017,6 +1017,8 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 	pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control(
 			pipe_ctx->stream_res.stream_enc, true);
 	if (pipe_ctx->stream_res.audio) {
+		struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
+
 		if (option != KEEP_ACQUIRED_RESOURCE ||
 				!dc->debug.az_endpoint_mute_only) {
 			/*only disalbe az_endpoint if power down or free*/
@@ -1036,6 +1038,9 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false);
 			pipe_ctx->stream_res.audio = NULL;
 		}
+		if (pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
+			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
 
 		/* TODO: notify audio driver for if audio modes list changed
 		 * add audio mode list change flag */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
index dcb3c5530236..cd1ebe57ed59 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
@@ -463,7 +463,7 @@ void dpp1_set_cursor_position(
 	if (src_y_offset >= (int)param->viewport.height)
 		cur_en = 0;  /* not visible beyond bottom edge*/
 
-	if (src_y_offset < 0)
+	if (src_y_offset + (int)height <= 0)
 		cur_en = 0;  /* not visible beyond top edge*/
 
 	REG_UPDATE(CURSOR0_CONTROL,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
index 345af015d061..d1acd7165bc8 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
@@ -1140,7 +1140,7 @@ void hubp1_cursor_set_position(
 	if (src_y_offset >= (int)param->viewport.height)
 		cur_en = 0;  /* not visible beyond bottom edge*/
 
-	if (src_y_offset < 0) //+ (int)hubp->curs_attr.height
+	if (src_y_offset + (int)hubp->curs_attr.height <= 0)
 		cur_en = 0;  /* not visible beyond top edge*/
 
 	if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 91e015e14355..58a12ddf12f3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -2355,29 +2355,22 @@ static void dcn10_apply_ctx_for_surface(
 			top_pipe_to_program->plane_state->update_flags.bits.full_update)
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
+			tg = pipe_ctx->stream_res.tg;
 			/* Skip inactive pipes and ones already updated */
 			if (!pipe_ctx->stream || pipe_ctx->stream == stream
-					|| !pipe_ctx->plane_state)
+					|| !pipe_ctx->plane_state
+					|| !tg->funcs->is_tg_enabled(tg))
 				continue;
 
-			pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg);
+			tg->funcs->lock(tg);
 
 			pipe_ctx->plane_res.hubp->funcs->hubp_setup_interdependent(
 				pipe_ctx->plane_res.hubp,
 				&pipe_ctx->dlg_regs,
 				&pipe_ctx->ttu_regs);
-		}
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
-		if (!pipe_ctx->stream || pipe_ctx->stream == stream
-				|| !pipe_ctx->plane_state)
-			continue;
-
-		dcn10_pipe_control_lock(dc, pipe_ctx, false);
-	}
+			tg->funcs->unlock(tg);
+		}
 
 	if (num_planes == 0)
 		false_optc_underflow_wa(dc, stream, tg);
diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 00f63b7dd32f..c11a443dcbc8 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -57,6 +57,7 @@ static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_le
 #define NUM_POWER_FN_SEGS 8
 #define NUM_BL_CURVE_SEGS 16
 
+#pragma pack(push, 1)
 /* NOTE: iRAM is 256B in size */
 struct iram_table_v_2 {
 	/* flags                      */
@@ -100,6 +101,7 @@ struct iram_table_v_2 {
 	uint8_t dummy8;							/* 0xfe       */
 	uint8_t dummy9;							/* 0xff       */
 };
+#pragma pack(pop)
 
 static uint16_t backlight_8_to_16(unsigned int backlight_8bit)
 {
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
index 54364444ecd1..0c8212902275 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
@@ -753,6 +753,22 @@ static int vega12_init_smc_table(struct pp_hwmgr *hwmgr)
 	return 0;
 }
 
+static int vega12_run_acg_btc(struct pp_hwmgr *hwmgr)
+{
+	uint32_t result;
+
+	PP_ASSERT_WITH_CODE(
+		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc) == 0,
+		"[Run_ACG_BTC] Attempt to run ACG BTC failed!",
+		return -EINVAL);
+
+	result = smum_get_argument(hwmgr);
+	PP_ASSERT_WITH_CODE(result == 1,
+			"Failed to run ACG BTC!", return -EINVAL);
+
+	return 0;
+}
+
 static int vega12_set_allowed_featuresmask(struct pp_hwmgr *hwmgr)
 {
 	struct vega12_hwmgr *data =
@@ -931,6 +947,11 @@ static int vega12_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
 			"Failed to initialize SMC table!",
 			result = tmp_result);
 
+	tmp_result = vega12_run_acg_btc(hwmgr);
+	PP_ASSERT_WITH_CODE(!tmp_result,
+			"Failed to run ACG BTC!",
+			result = tmp_result);
+
 	result = vega12_enable_all_smu_features(hwmgr);
 	PP_ASSERT_WITH_CODE(!result,
 			"Failed to enable all smu features!",