Merge tag 'amd-drm-next-6.3-2023-01-06' of https://gitlab.freedesktop…

….org/agd5f/linux into drm-next

amd-drm-next-6.3-2023-01-06:

amdgpu:
- secure display support for multiple displays
- DML optimizations
- DCN 3.2 updates
- PSR updates
- DP 2.1 updates
- SR-IOV RAS updates
- VCN RAS support
- SMU 13.x updates
- Switch 1 element arrays to flexible arrays
- Add RAS support for DF 4.3
- Stack size improvements
- S0ix rework
- Soft reset fix
- Allow 0 as a vram limit on APUs
- Display fixes
- Misc code cleanups
- Documentation fixes
- Handle profiling modes for SMU13.x

amdkfd:
- Error handling fixes
- PASID fixes

radeon:
- Switch 1 element arrays to flexible arrays

drm:
- Add DP adaptive sync DPCD definitions

UAPI:
- Add new INFO queries for peak and min sclk/mclk for profile modes on newer chips
  Proposed mesa patch: https://gitlab.freedesktop.org/mesa/drm/-/merge_requests/278

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230106222037.7870-1-alexander.deucher@amd.com

drivers/gpu/drm/amd/amdgpu/Makefile

@@ -81,7 +81,8 @@ amdgpu-y += \
 # add DF block
 amdgpu-y += \
 	df_v1_7.o \
-	df_v3_6.o
+	df_v3_6.o \
+	df_v4_3.o
 
 # add GMC block
 amdgpu-y += \

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

@@ -149,7 +149,7 @@ struct amdgpu_watchdog_timer
  * Modules parameters.
  */
 extern int amdgpu_modeset;
-extern int amdgpu_vram_limit;
+extern unsigned int amdgpu_vram_limit;
 extern int amdgpu_vis_vram_limit;
 extern int amdgpu_gart_size;
 extern int amdgpu_gtt_size;
@@ -194,6 +194,7 @@ extern int amdgpu_emu_mode;
 extern uint amdgpu_smu_memory_pool_size;
 extern int amdgpu_smu_pptable_id;
 extern uint amdgpu_dc_feature_mask;
+extern uint amdgpu_freesync_vid_mode;
 extern uint amdgpu_dc_debug_mask;
 extern uint amdgpu_dc_visual_confirm;
 extern uint amdgpu_dm_abm_level;
@@ -607,7 +608,7 @@ int amdgpu_cs_wait_fences_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp);
 
 /* VRAM scratch page for HDP bug, default vram page */
-struct amdgpu_vram_scratch {
+struct amdgpu_mem_scratch {
 	struct amdgpu_bo		*robj;
 	volatile uint32_t		*ptr;
 	u64				gpu_addr;
@@ -754,6 +755,11 @@ struct amdgpu_mqd {
 #define AMDGPU_PRODUCT_NAME_LEN 64
 struct amdgpu_reset_domain;
 
+/*
+ * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise.
+ */
+#define AMDGPU_HAS_VRAM(_adev) ((_adev)->gmc.real_vram_size)
+
 struct amdgpu_device {
 	struct device			*dev;
 	struct pci_dev			*pdev;
@@ -847,7 +853,7 @@ struct amdgpu_device {
 
 	/* memory management */
 	struct amdgpu_mman		mman;
-	struct amdgpu_vram_scratch	vram_scratch;
+	struct amdgpu_mem_scratch	mem_scratch;
 	struct amdgpu_wb		wb;
 	atomic64_t			num_bytes_moved;
 	atomic64_t			num_evictions;
@@ -869,7 +875,7 @@ struct amdgpu_device {
 	struct amdgpu_vkms_output       *amdgpu_vkms_output;
 	struct amdgpu_mode_info		mode_info;
 	/* For pre-DCE11. DCE11 and later are in "struct amdgpu_device->dm" */
-	struct work_struct		hotplug_work;
+	struct delayed_work         hotplug_work;
 	struct amdgpu_irq_src		crtc_irq;
 	struct amdgpu_irq_src		vline0_irq;
 	struct amdgpu_irq_src		vupdate_irq;

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

@@ -996,13 +996,33 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force)
 		}
 	}
 
+	if (amdgpu_connector->detected_hpd_without_ddc) {
+		force = true;
+		amdgpu_connector->detected_hpd_without_ddc = false;
+	}
+
 	if (!force && amdgpu_connector_check_hpd_status_unchanged(connector)) {
 		ret = connector->status;
 		goto exit;
 	}
 
-	if (amdgpu_connector->ddc_bus)
+	if (amdgpu_connector->ddc_bus) {
 		dret = amdgpu_display_ddc_probe(amdgpu_connector, false);
+
+		/* Sometimes the pins required for the DDC probe on DVI
+		 * connectors don't make contact at the same time that the ones
+		 * for HPD do. If the DDC probe fails even though we had an HPD
+		 * signal, try again later
+		 */
+		if (!dret && !force &&
+		    amdgpu_display_hpd_sense(adev, amdgpu_connector->hpd.hpd)) {
+			DRM_DEBUG_KMS("hpd detected without ddc, retrying in 1 second\n");
+			amdgpu_connector->detected_hpd_without_ddc = true;
+			schedule_delayed_work(&adev->hotplug_work,
+					      msecs_to_jiffies(1000));
+			goto exit;
+		}
+	}
 	if (dret) {
 		amdgpu_connector->detected_by_load = false;
 		amdgpu_connector_free_edid(connector);

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

@@ -924,32 +924,33 @@ static int amdgpu_device_asic_init(struct amdgpu_device *adev)
 }
 
 /**
- * amdgpu_device_vram_scratch_init - allocate the VRAM scratch page
+ * amdgpu_device_mem_scratch_init - allocate the VRAM scratch page
  *
  * @adev: amdgpu_device pointer
  *
  * Allocates a scratch page of VRAM for use by various things in the
  * driver.
  */
-static int amdgpu_device_vram_scratch_init(struct amdgpu_device *adev)
+static int amdgpu_device_mem_scratch_init(struct amdgpu_device *adev)
 {
-	return amdgpu_bo_create_kernel(adev, AMDGPU_GPU_PAGE_SIZE,
-				       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
-				       &adev->vram_scratch.robj,
-				       &adev->vram_scratch.gpu_addr,
-				       (void **)&adev->vram_scratch.ptr);
+	return amdgpu_bo_create_kernel(adev, AMDGPU_GPU_PAGE_SIZE, PAGE_SIZE,
+				       AMDGPU_GEM_DOMAIN_VRAM |
+				       AMDGPU_GEM_DOMAIN_GTT,
+				       &adev->mem_scratch.robj,
+				       &adev->mem_scratch.gpu_addr,
+				       (void **)&adev->mem_scratch.ptr);
 }
 
 /**
- * amdgpu_device_vram_scratch_fini - Free the VRAM scratch page
+ * amdgpu_device_mem_scratch_fini - Free the VRAM scratch page
  *
  * @adev: amdgpu_device pointer
  *
  * Frees the VRAM scratch page.
  */
-static void amdgpu_device_vram_scratch_fini(struct amdgpu_device *adev)
+static void amdgpu_device_mem_scratch_fini(struct amdgpu_device *adev)
 {
-	amdgpu_bo_free_kernel(&adev->vram_scratch.robj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->mem_scratch.robj, NULL, NULL);
 }
 
 /**
@@ -2390,9 +2391,9 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			if (amdgpu_sriov_vf(adev))
 				amdgpu_virt_exchange_data(adev);
 
-			r = amdgpu_device_vram_scratch_init(adev);
+			r = amdgpu_device_mem_scratch_init(adev);
 			if (r) {
-				DRM_ERROR("amdgpu_vram_scratch_init failed %d\n", r);
+				DRM_ERROR("amdgpu_mem_scratch_init failed %d\n", r);
 				goto init_failed;
 			}
 			r = adev->ip_blocks[i].version->funcs->hw_init((void *)adev);
@@ -2410,8 +2411,9 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			/* right after GMC hw init, we create CSA */
 			if (amdgpu_mcbp) {
 				r = amdgpu_allocate_static_csa(adev, &adev->virt.csa_obj,
-								AMDGPU_GEM_DOMAIN_VRAM,
-								AMDGPU_CSA_SIZE);
+							       AMDGPU_GEM_DOMAIN_VRAM |
+							       AMDGPU_GEM_DOMAIN_GTT,
+							       AMDGPU_CSA_SIZE);
 				if (r) {
 					DRM_ERROR("allocate CSA failed %d\n", r);
 					goto init_failed;
@@ -2581,9 +2583,10 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev,
 		i = state == AMD_CG_STATE_GATE ? j : adev->num_ip_blocks - j - 1;
 		if (!adev->ip_blocks[i].status.late_initialized)
 			continue;
-		/* skip CG for GFX on S0ix */
+		/* skip CG for GFX, SDMA on S0ix */
 		if (adev->in_s0ix &&
-		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX)
+		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX ||
+		     adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA))
 			continue;
 		/* skip CG for VCE/UVD, it's handled specially */
 		if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_UVD &&
@@ -2617,9 +2620,10 @@ int amdgpu_device_set_pg_state(struct amdgpu_device *adev,
 		i = state == AMD_PG_STATE_GATE ? j : adev->num_ip_blocks - j - 1;
 		if (!adev->ip_blocks[i].status.late_initialized)
 			continue;
-		/* skip PG for GFX on S0ix */
+		/* skip PG for GFX, SDMA on S0ix */
 		if (adev->in_s0ix &&
-		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX)
+		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX ||
+		     adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA))
 			continue;
 		/* skip CG for VCE/UVD, it's handled specially */
 		if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_UVD &&
@@ -2871,7 +2875,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 			amdgpu_ucode_free_bo(adev);
 			amdgpu_free_static_csa(&adev->virt.csa_obj);
 			amdgpu_device_wb_fini(adev);
-			amdgpu_device_vram_scratch_fini(adev);
+			amdgpu_device_mem_scratch_fini(adev);
 			amdgpu_ib_pool_fini(adev);
 		}
 
@@ -3027,6 +3031,12 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
 		     adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_MES))
 			continue;
 
+		/* SDMA 5.x+ is part of GFX power domain so it's covered by GFXOFF */
+		if (adev->in_s0ix &&
+		    (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(5, 0, 0)) &&
+		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA))
+			continue;
+
 		/* XXX handle errors */
 		r = adev->ip_blocks[i].version->funcs->suspend(adev);
 		/* XXX handle errors */
@@ -3227,15 +3237,6 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev)
 			return r;
 		}
 		adev->ip_blocks[i].status.hw = true;
-
-		if (adev->in_s0ix && adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
-			/* disable gfxoff for IP resume. The gfxoff will be re-enabled in
-			 * amdgpu_device_resume() after IP resume.
-			 */
-			amdgpu_gfx_off_ctrl(adev, false);
-			DRM_DEBUG("will disable gfxoff for re-initializing other blocks\n");
-		}
-
 	}
 
 	return 0;
@@ -4221,13 +4222,6 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	/* Make sure IB tests flushed */
 	flush_delayed_work(&adev->delayed_init_work);
 
-	if (adev->in_s0ix) {
-		/* re-enable gfxoff after IP resume. This re-enables gfxoff after
-		 * it was disabled for IP resume in amdgpu_device_ip_resume_phase2().
-		 */
-		amdgpu_gfx_off_ctrl(adev, true);
-		DRM_DEBUG("will enable gfxoff for the mission mode\n");
-	}
 	if (fbcon)
 		drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false);
 
@@ -4608,11 +4602,6 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev)
 	if (!amdgpu_ras_is_poison_mode_supported(adev))
 		return true;
 
-	if (!amdgpu_device_ip_check_soft_reset(adev)) {
-		dev_info(adev->dev,"Timeout, but no hardware hang detected.\n");
-		return false;
-	}
-
 	if (amdgpu_sriov_vf(adev))
 		return true;
 
@@ -4737,7 +4726,8 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 		if (!need_full_reset)
 			need_full_reset = amdgpu_device_ip_need_full_reset(adev);
 
-		if (!need_full_reset && amdgpu_gpu_recovery) {
+		if (!need_full_reset && amdgpu_gpu_recovery &&
+		    amdgpu_device_ip_check_soft_reset(adev)) {
 			amdgpu_device_ip_pre_soft_reset(adev);
 			r = amdgpu_device_ip_soft_reset(adev);
 			amdgpu_device_ip_post_soft_reset(adev);

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

@@ -33,6 +33,7 @@
 #include "gmc_v9_0.h"
 #include "df_v1_7.h"
 #include "df_v3_6.h"
+#include "df_v4_3.h"
 #include "nbio_v6_1.h"
 #include "nbio_v7_0.h"
 #include "nbio_v7_4.h"
@@ -2329,6 +2330,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(3, 5, 2):
 		adev->df.funcs = &df_v1_7_funcs;
 		break;
+	case IP_VERSION(4, 3, 0):
+		adev->df.funcs = &df_v4_3_funcs;
+		break;
 	default:
 		break;
 	}

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

@@ -63,7 +63,7 @@
 void amdgpu_display_hotplug_work_func(struct work_struct *work)
 {
 	struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
-						  hotplug_work);
+						  hotplug_work.work);
 	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	struct drm_connector *connector;

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

@@ -105,13 +105,15 @@
  * - 3.46.0 - To enable hot plug amdgpu tests in libdrm
  * - 3.47.0 - Add AMDGPU_GEM_CREATE_DISCARDABLE and AMDGPU_VM_NOALLOC flags
  * - 3.48.0 - Add IP discovery version info to HW INFO
- *   3.49.0 - Add gang submit into CS IOCTL
+ * - 3.49.0 - Add gang submit into CS IOCTL
+ * - 3.50.0 - Update AMDGPU_INFO_DEV_INFO IOCTL for minimum engine and memory clock
+ *            Update AMDGPU_INFO_SENSOR IOCTL for PEAK_PSTATE engine and memory clock
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	49
+#define KMS_DRIVER_MINOR	50
 #define KMS_DRIVER_PATCHLEVEL	0
 
-int amdgpu_vram_limit;
+unsigned int amdgpu_vram_limit = UINT_MAX;
 int amdgpu_vis_vram_limit;
 int amdgpu_gart_size = -1; /* auto */
 int amdgpu_gtt_size = -1; /* auto */
@@ -181,6 +183,7 @@ int amdgpu_mes_kiq;
 int amdgpu_noretry = -1;
 int amdgpu_force_asic_type = -1;
 int amdgpu_tmz = -1; /* auto */
+uint amdgpu_freesync_vid_mode;
 int amdgpu_reset_method = -1; /* auto */
 int amdgpu_num_kcq = -1;
 int amdgpu_smartshift_bias;
@@ -879,6 +882,32 @@ module_param_named(backlight, amdgpu_backlight, bint, 0444);
 MODULE_PARM_DESC(tmz, "Enable TMZ feature (-1 = auto (default), 0 = off, 1 = on)");
 module_param_named(tmz, amdgpu_tmz, int, 0444);
 
+/**
+ * DOC: freesync_video (uint)
+ * Enable the optimization to adjust front porch timing to achieve seamless
+ * mode change experience when setting a freesync supported mode for which full
+ * modeset is not needed.
+ *
+ * The Display Core will add a set of modes derived from the base FreeSync
+ * video mode into the corresponding connector's mode list based on commonly
+ * used refresh rates and VRR range of the connected display, when users enable
+ * this feature. From the userspace perspective, they can see a seamless mode
+ * change experience when the change between different refresh rates under the
+ * same resolution. Additionally, userspace applications such as Video playback
+ * can read this modeset list and change the refresh rate based on the video
+ * frame rate. Finally, the userspace can also derive an appropriate mode for a
+ * particular refresh rate based on the FreeSync Mode and add it to the
+ * connector's mode list.
+ *
+ * Note: This is an experimental feature.
+ *
+ * The default value: 0 (off).
+ */
+MODULE_PARM_DESC(
+	freesync_video,
+	"Enable freesync modesetting optimization feature (0 = off (default), 1 = on)");
+module_param_named(freesync_video, amdgpu_freesync_vid_mode, uint, 0444);
+
 /**
  * DOC: reset_method (int)
  * GPU reset method (-1 = auto (default), 0 = legacy, 1 = mode0, 2 = mode1, 3 = mode2, 4 = baco)

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

@@ -372,8 +372,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 		 * KIQ MQD no matter SRIOV or Bare-metal
 		 */
 		r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
-					    AMDGPU_GEM_DOMAIN_VRAM, &ring->mqd_obj,
-					    &ring->mqd_gpu_addr, &ring->mqd_ptr);
+					    AMDGPU_GEM_DOMAIN_VRAM |
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &ring->mqd_obj,
+					    &ring->mqd_gpu_addr,
+					    &ring->mqd_ptr);
 		if (r) {
 			dev_warn(adev->dev, "failed to create ring mqd ob (%d)", r);
 			return r;

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

@@ -202,13 +202,20 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo)
 void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
 			      u64 base)
 {
+	uint64_t vis_limit = (uint64_t)amdgpu_vis_vram_limit << 20;
 	uint64_t limit = (uint64_t)amdgpu_vram_limit << 20;
 
 	mc->vram_start = base;
 	mc->vram_end = mc->vram_start + mc->mc_vram_size - 1;
-	if (limit && limit < mc->real_vram_size)
+	if (limit < mc->real_vram_size)
 		mc->real_vram_size = limit;
 
+	if (vis_limit && vis_limit < mc->visible_vram_size)
+		mc->visible_vram_size = vis_limit;
+
+	if (mc->real_vram_size < mc->visible_vram_size)
+		mc->visible_vram_size = mc->real_vram_size;
+
 	if (mc->xgmi.num_physical_nodes == 0) {
 		mc->fb_start = mc->vram_start;
 		mc->fb_end = mc->vram_end;