15 files changed, 138 insertions, 50 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 8639a4f9c6e8..2eca58220550 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1293,6 +1293,7 @@ void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev,
 				u32 reg, u32 v);
 struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev,
 					    struct dma_fence *gang);
+bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev);
 
 /* atpx handler */
 #if defined(CONFIG_VGA_SWITCHEROO)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 0561812aa0a4..5d9a34601a1a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -706,6 +706,13 @@ err:
 
 void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle)
 {
+	/* Temporary workaround to fix issues observed in some
+	 * compute applications when GFXOFF is enabled on GFX11.
+	 */
+	if (IP_VERSION_MAJ(adev->ip_versions[GC_HWIP][0]) == 11) {
+		pr_debug("GFXOFF is %s\n", idle ? "enabled" : "disabled");
+		amdgpu_gfx_off_ctrl(adev, idle);
+	}
 	amdgpu_dpm_switch_power_profile(adev,
 					PP_SMC_POWER_PROFILE_COMPUTE,
 					!idle);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 84f44f7e4111..6d1ff7b9258d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -171,9 +171,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	    (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
 	     kfd_mem_limit.max_ttm_mem_limit) ||
 	    (adev && adev->kfd.vram_used + vram_needed >
-	     adev->gmc.real_vram_size -
-	     atomic64_read(&adev->vram_pin_size) -
-	     reserved_for_pt)) {
+	     adev->gmc.real_vram_size - reserved_for_pt)) {
 		ret = -ENOMEM;
 		goto release;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 1bbd39b3b0fc..d038b258cc92 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -109,6 +109,7 @@ static int amdgpu_cs_p1_ib(struct amdgpu_cs_parser *p,
 		return r;
 
 	++(num_ibs[r]);
+	p->gang_leader_idx = r;
 	return 0;
 }
 
@@ -287,8 +288,10 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 		}
 	}
 
-	if (!p->gang_size)
-		return -EINVAL;
+	if (!p->gang_size) {
+		ret = -EINVAL;
+		goto free_partial_kdata;
+	}
 
 	for (i = 0; i < p->gang_size; ++i) {
 		ret = amdgpu_job_alloc(p->adev, num_ibs[i], &p->jobs[i], vm);
@@ -300,7 +303,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 		if (ret)
 			goto free_all_kdata;
 	}
-	p->gang_leader = p->jobs[p->gang_size - 1];
+	p->gang_leader = p->jobs[p->gang_leader_idx];
 
 	if (p->ctx->vram_lost_counter != p->gang_leader->vram_lost_counter) {
 		ret = -ECANCELED;
@@ -992,6 +995,7 @@ out_free_user_pages:
 		kvfree(e->user_pages);
 		e->user_pages = NULL;
 	}
+	mutex_unlock(&p->bo_list->bo_list_mutex);
 	return r;
 }
 
@@ -1194,16 +1198,18 @@ static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p)
 			return r;
 	}
 
-	for (i = 0; i < p->gang_size - 1; ++i) {
+	for (i = 0; i < p->gang_size; ++i) {
+		if (p->jobs[i] == leader)
+			continue;
+
 		r = amdgpu_sync_clone(&leader->sync, &p->jobs[i]->sync);
 		if (r)
 			return r;
 	}
 
-	r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_size - 1]);
+	r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_leader_idx]);
 	if (r && r != -ERESTARTSYS)
 		DRM_ERROR("amdgpu_ctx_wait_prev_fence failed.\n");
-
 	return r;
 }
 
@@ -1237,9 +1243,12 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	for (i = 0; i < p->gang_size; ++i)
 		drm_sched_job_arm(&p->jobs[i]->base);
 
-	for (i = 0; i < (p->gang_size - 1); ++i) {
+	for (i = 0; i < p->gang_size; ++i) {
 		struct dma_fence *fence;
 
+		if (p->jobs[i] == leader)
+			continue;
+
 		fence = &p->jobs[i]->base.s_fence->scheduled;
 		r = amdgpu_sync_fence(&leader->sync, fence);
 		if (r)
@@ -1275,7 +1284,10 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	list_for_each_entry(e, &p->validated, tv.head) {
 
 		/* Everybody except for the gang leader uses READ */
-		for (i = 0; i < (p->gang_size - 1); ++i) {
+		for (i = 0; i < p->gang_size; ++i) {
+			if (p->jobs[i] == leader)
+				continue;
+
 			dma_resv_add_fence(e->tv.bo->base.resv,
 					   &p->jobs[i]->base.s_fence->finished,
 					   DMA_RESV_USAGE_READ);
@@ -1285,7 +1297,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 		e->tv.num_shared = 0;
 	}
 
-	seq = amdgpu_ctx_add_fence(p->ctx, p->entities[p->gang_size - 1],
+	seq = amdgpu_ctx_add_fence(p->ctx, p->entities[p->gang_leader_idx],
 				   p->fence);
 	amdgpu_cs_post_dependencies(p);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h
index cbaa19b2b8a3..f80adf9069ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h
@@ -54,6 +54,7 @@ struct amdgpu_cs_parser {
 
 	/* scheduler job objects */
 	unsigned int		gang_size;
+	unsigned int		gang_leader_idx;
 	struct drm_sched_entity	*entities[AMDGPU_CS_GANG_SIZE];
 	struct amdgpu_job	*jobs[AMDGPU_CS_GANG_SIZE];
 	struct amdgpu_job	*gang_leader;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index ddaecb2610c9..f1e9663b4051 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -4060,15 +4060,18 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev)
  * at suspend time.
  *
  */
-static void amdgpu_device_evict_resources(struct amdgpu_device *adev)
+static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
 {
+	int ret;
+
 	/* No need to evict vram on APUs for suspend to ram or s2idle */
 	if ((adev->in_s3 || adev->in_s0ix) && (adev->flags & AMD_IS_APU))
-		return;
+		return 0;
 
-	if (amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM))
+	ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM);
+	if (ret)
 		DRM_WARN("evicting device resources failed\n");
-
+	return ret;
 }
 
 /*
@@ -4118,7 +4121,9 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 	if (!adev->in_s0ix)
 		amdgpu_amdkfd_suspend(adev, adev->in_runpm);
 
-	amdgpu_device_evict_resources(adev);
+	r = amdgpu_device_evict_resources(adev);
+	if (r)
+		return r;
 
 	amdgpu_fence_driver_hw_fini(adev);
 
@@ -6039,3 +6044,44 @@ struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev,
 	dma_fence_put(old);
 	return NULL;
 }
+
+bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
+#ifdef CONFIG_DRM_AMDGPU_SI
+	case CHIP_HAINAN:
+#endif
+	case CHIP_TOPAZ:
+		/* chips with no display hardware */
+		return false;
+#ifdef CONFIG_DRM_AMDGPU_SI
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+#endif
+#ifdef CONFIG_DRM_AMDGPU_CIK
+	case CHIP_BONAIRE:
+	case CHIP_HAWAII:
+	case CHIP_KAVERI:
+	case CHIP_KABINI:
+	case CHIP_MULLINS:
+#endif
+	case CHIP_TONGA:
+	case CHIP_FIJI:
+	case CHIP_POLARIS10:
+	case CHIP_POLARIS11:
+	case CHIP_POLARIS12:
+	case CHIP_VEGAM:
+	case CHIP_CARRIZO:
+	case CHIP_STONEY:
+		/* chips with display hardware */
+		return true;
+	default:
+		/* IP discovery */
+		if (!adev->ip_versions[DCE_HWIP][0] ||
+		    (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK))
+			return false;
+		return true;
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 3c9fecdd6b2f..bf2d50c8c92a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2201,7 +2201,8 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 		pm_runtime_forbid(dev->dev);
 	}
 
-	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2)) {
+	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) &&
+	    !amdgpu_sriov_vf(adev)) {
 		bool need_to_reset_gpu = false;
 
 		if (adev->gmc.xgmi.num_physical_nodes > 1) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 34233a74248c..b06cb0bb166c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -656,7 +656,7 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_sriov_vf(adev) ||
-	    !amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_DCE)) {
+	    !amdgpu_device_has_display_hardware(adev)) {
 		size = 0;
 	} else {
 		size = amdgpu_gmc_get_vbios_fb_size(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index bf1ff8f0e712..4e42dcb1950f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -337,12 +337,14 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info,
 		fw_info->feature = adev->psp.cap_feature_version;
 		break;
 	case AMDGPU_INFO_FW_MES_KIQ:
-		fw_info->ver = adev->mes.ucode_fw_version[0];
-		fw_info->feature = 0;
+		fw_info->ver = adev->mes.kiq_version & AMDGPU_MES_VERSION_MASK;
+		fw_info->feature = (adev->mes.kiq_version & AMDGPU_MES_FEAT_VERSION_MASK)
+					>> AMDGPU_MES_FEAT_VERSION_SHIFT;
 		break;
 	case AMDGPU_INFO_FW_MES:
-		fw_info->ver = adev->mes.ucode_fw_version[1];
-		fw_info->feature = 0;
+		fw_info->ver = adev->mes.sched_version & AMDGPU_MES_VERSION_MASK;
+		fw_info->feature = (adev->mes.sched_version & AMDGPU_MES_FEAT_VERSION_MASK)
+					>> AMDGPU_MES_FEAT_VERSION_SHIFT;
 		break;
 	case AMDGPU_INFO_FW_IMU:
 		fw_info->ver = adev->gfx.imu_fw_version;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
index f4b5301ea2a0..500a1dc4fe02 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
@@ -500,6 +500,8 @@ static int amdgpu_vkms_sw_init(void *handle)
 
 	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
+	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
+
 	r = amdgpu_display_modeset_create_props(adev);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 2291aa14d888..003aa9e47085 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -143,32 +143,6 @@ int amdgpu_vm_set_pasid(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	return 0;
 }
 
-/*
- * vm eviction_lock can be taken in MMU notifiers. Make sure no reclaim-FS
- * happens while holding this lock anywhere to prevent deadlocks when
- * an MMU notifier runs in reclaim-FS context.
- */
-static inline void amdgpu_vm_eviction_lock(struct amdgpu_vm *vm)
-{
-	mutex_lock(&vm->eviction_lock);
-	vm->saved_flags = memalloc_noreclaim_save();
-}
-
-static inline int amdgpu_vm_eviction_trylock(struct amdgpu_vm *vm)
-{
-	if (mutex_trylock(&vm->eviction_lock)) {
-		vm->saved_flags = memalloc_noreclaim_save();
-		return 1;
-	}
-	return 0;
-}
-
-static inline void amdgpu_vm_eviction_unlock(struct amdgpu_vm *vm)
-{
-	memalloc_noreclaim_restore(vm->saved_flags);
-	mutex_unlock(&vm->eviction_lock);
-}
-
 /**
  * amdgpu_vm_bo_evicted - vm_bo is evicted
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 83acb7bd80fe..6546e786bf00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -492,7 +492,48 @@ void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m);
  */
 static inline uint64_t amdgpu_vm_tlb_seq(struct amdgpu_vm *vm)
 {
+	unsigned long flags;
+	spinlock_t *lock;
+
+	/*
+	 * Workaround to stop racing between the fence signaling and handling
+	 * the cb. The lock is static after initially setting it up, just make
+	 * sure that the dma_fence structure isn't freed up.
+	 */
+	rcu_read_lock();
+	lock = vm->last_tlb_flush->lock;
+	rcu_read_unlock();
+
+	spin_lock_irqsave(lock, flags);
+	spin_unlock_irqrestore(lock, flags);
+
 	return atomic64_read(&vm->tlb_seq);
 }
 
+/*
+ * vm eviction_lock can be taken in MMU notifiers. Make sure no reclaim-FS
+ * happens while holding this lock anywhere to prevent deadlocks when
+ * an MMU notifier runs in reclaim-FS context.
+ */
+static inline void amdgpu_vm_eviction_lock(struct amdgpu_vm *vm)
+{
+	mutex_lock(&vm->eviction_lock);
+	vm->saved_flags = memalloc_noreclaim_save();
+}
+
+static inline bool amdgpu_vm_eviction_trylock(struct amdgpu_vm *vm)
+{
+	if (mutex_trylock(&vm->eviction_lock)) {
+		vm->saved_flags = memalloc_noreclaim_save();
+		return true;
+	}
+	return false;
+}
+
+static inline void amdgpu_vm_eviction_unlock(struct amdgpu_vm *vm)
+{
+	memalloc_noreclaim_restore(vm->saved_flags);
+	mutex_unlock(&vm->eviction_lock);
+}
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index 358b91243e37..b5f3bba851db 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -597,7 +597,9 @@ static int amdgpu_vm_pt_alloc(struct amdgpu_device *adev,
 	if (entry->bo)
 		return 0;
 
+	amdgpu_vm_eviction_unlock(vm);
 	r = amdgpu_vm_pt_create(adev, vm, cursor->level, immediate, &pt);
+	amdgpu_vm_eviction_lock(vm);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 73a517bcf5c1..80dd1343594c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -435,7 +435,7 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man,
 	if (place->flags & TTM_PL_FLAG_TOPDOWN)
 		vres->flags |= DRM_BUDDY_TOPDOWN_ALLOCATION;
 
-	if (fpfn || lpfn != man->size)
+	if (fpfn || lpfn != mgr->mm.size)
 		/* Allocate blocks in desired range */
 		vres->flags |= DRM_BUDDY_RANGE_ALLOCATION;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
index 21d822b1d589..88f9b327183a 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
@@ -45,6 +45,7 @@ MODULE_FIRMWARE("amdgpu/psp_13_0_0_ta.bin");
 MODULE_FIRMWARE("amdgpu/psp_13_0_7_sos.bin");
 MODULE_FIRMWARE("amdgpu/psp_13_0_7_ta.bin");
 MODULE_FIRMWARE("amdgpu/psp_13_0_10_sos.bin");
+MODULE_FIRMWARE("amdgpu/psp_13_0_10_ta.bin");
 
 /* For large FW files the time to complete can be very long */
 #define USBC_PD_POLLING_LIMIT_S 240