Merge tag 'drm-fixes-2023-01-06' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Daniel Vetter:
 "Still not much, but more than last week. Dave should be back next week
  from the beaching.

  drivers:
   - i915-gvt fixes
   - amdgpu/kfd fixes
   - panfrost bo refcounting fix
   - meson afbc corruption fix
   - imx plane width fix

  core:
   - drm/sched fixes
   - drm/mm kunit test fix
   - dma-buf export error handling fixes"

* tag 'drm-fixes-2023-01-06' of git://anongit.freedesktop.org/drm/drm:
  Revert "drm/amd/display: Enable Freesync Video Mode by default"
  drm/i915/gvt: fix double free bug in split_2MB_gtt_entry
  drm/i915/gvt: use atomic operations to change the vGPU status
  drm/i915/gvt: fix vgpu debugfs clean in remove
  drm/i915/gvt: fix gvt debugfs destroy
  drm/i915: unpin on error in intel_vgpu_shadow_mm_pin()
  drm/amd/display: Uninitialized variables causing 4k60 UCLK to stay at DPM1 and not DPM0
  drm/amdkfd: Fix kernel warning during topology setup
  drm/scheduler: Fix lockup in drm_sched_entity_kill()
  drm/imx: ipuv3-plane: Fix overlay plane width
  drm/scheduler: Fix lockup in drm_sched_entity_kill()
  drm/virtio: Fix memory leak in virtio_gpu_object_create()
  drm/meson: Reduce the FIFO lines held when AFBC is not used
  drm/tests: reduce drm_mm_test stack usage
  drm/panfrost: Fix GEM handle creation ref-counting
  drm/plane-helper: Add the missing declaration of drm_atomic_state
  dma-buf: fix dma_buf_export init order v2

27 files changed, 204 insertions, 153 deletions

diff --git a/drivers/dma-buf/dma-buf-sysfs-stats.c b/drivers/dma-buf/dma-buf-sysfs-stats.c
index f69d68122b9b..fbf725fae7c1 100644
--- a/drivers/dma-buf/dma-buf-sysfs-stats.c
+++ b/drivers/dma-buf/dma-buf-sysfs-stats.c
@@ -168,14 +168,11 @@ void dma_buf_uninit_sysfs_statistics(void)
 	kset_unregister(dma_buf_stats_kset);
 }
 
-int dma_buf_stats_setup(struct dma_buf *dmabuf)
+int dma_buf_stats_setup(struct dma_buf *dmabuf, struct file *file)
 {
 	struct dma_buf_sysfs_entry *sysfs_entry;
 	int ret;
 
-	if (!dmabuf || !dmabuf->file)
-		return -EINVAL;
-
 	if (!dmabuf->exp_name) {
 		pr_err("exporter name must not be empty if stats needed\n");
 		return -EINVAL;
@@ -192,7 +189,7 @@ int dma_buf_stats_setup(struct dma_buf *dmabuf)
 
 	/* create the directory for buffer stats */
 	ret = kobject_init_and_add(&sysfs_entry->kobj, &dma_buf_ktype, NULL,
-				   "%lu", file_inode(dmabuf->file)->i_ino);
+				   "%lu", file_inode(file)->i_ino);
 	if (ret)
 		goto err_sysfs_dmabuf;
 
diff --git a/drivers/dma-buf/dma-buf-sysfs-stats.h b/drivers/dma-buf/dma-buf-sysfs-stats.h
index a49c6e2650cc..7a8a995b75ba 100644
--- a/drivers/dma-buf/dma-buf-sysfs-stats.h
+++ b/drivers/dma-buf/dma-buf-sysfs-stats.h
@@ -13,7 +13,7 @@
 int dma_buf_init_sysfs_statistics(void);
 void dma_buf_uninit_sysfs_statistics(void);
 
-int dma_buf_stats_setup(struct dma_buf *dmabuf);
+int dma_buf_stats_setup(struct dma_buf *dmabuf, struct file *file);
 
 void dma_buf_stats_teardown(struct dma_buf *dmabuf);
 #else
@@ -25,7 +25,7 @@ static inline int dma_buf_init_sysfs_statistics(void)
 
 static inline void dma_buf_uninit_sysfs_statistics(void) {}
 
-static inline int dma_buf_stats_setup(struct dma_buf *dmabuf)
+static inline int dma_buf_stats_setup(struct dma_buf *dmabuf, struct file *file)
 {
 	return 0;
 }
diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
index b6c36914e7c6..e6528767efc7 100644
--- a/drivers/dma-buf/dma-buf.c
+++ b/drivers/dma-buf/dma-buf.c
@@ -95,10 +95,11 @@ static int dma_buf_file_release(struct inode *inode, struct file *file)
 		return -EINVAL;
 
 	dmabuf = file->private_data;
-
-	mutex_lock(&db_list.lock);
-	list_del(&dmabuf->list_node);
-	mutex_unlock(&db_list.lock);
+	if (dmabuf) {
+		mutex_lock(&db_list.lock);
+		list_del(&dmabuf->list_node);
+		mutex_unlock(&db_list.lock);
+	}
 
 	return 0;
 }
@@ -528,17 +529,17 @@ static inline int is_dma_buf_file(struct file *file)
 	return file->f_op == &dma_buf_fops;
 }
 
-static struct file *dma_buf_getfile(struct dma_buf *dmabuf, int flags)
+static struct file *dma_buf_getfile(size_t size, int flags)
 {
 	static atomic64_t dmabuf_inode = ATOMIC64_INIT(0);
-	struct file *file;
 	struct inode *inode = alloc_anon_inode(dma_buf_mnt->mnt_sb);
+	struct file *file;
 
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 
-	inode->i_size = dmabuf->size;
-	inode_set_bytes(inode, dmabuf->size);
+	inode->i_size = size;
+	inode_set_bytes(inode, size);
 
 	/*
 	 * The ->i_ino acquired from get_next_ino() is not unique thus
@@ -552,8 +553,6 @@ static struct file *dma_buf_getfile(struct dma_buf *dmabuf, int flags)
 				 flags, &dma_buf_fops);
 	if (IS_ERR(file))
 		goto err_alloc_file;
-	file->private_data = dmabuf;
-	file->f_path.dentry->d_fsdata = dmabuf;
 
 	return file;
 
@@ -619,19 +618,11 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info)
 	size_t alloc_size = sizeof(struct dma_buf);
 	int ret;
 
-	if (!exp_info->resv)
-		alloc_size += sizeof(struct dma_resv);
-	else
-		/* prevent &dma_buf[1] == dma_buf->resv */
-		alloc_size += 1;
-
-	if (WARN_ON(!exp_info->priv
-			  || !exp_info->ops
-			  || !exp_info->ops->map_dma_buf
-			  || !exp_info->ops->unmap_dma_buf
-			  || !exp_info->ops->release)) {
+	if (WARN_ON(!exp_info->priv || !exp_info->ops
+		    || !exp_info->ops->map_dma_buf
+		    || !exp_info->ops->unmap_dma_buf
+		    || !exp_info->ops->release))
 		return ERR_PTR(-EINVAL);
-	}
 
 	if (WARN_ON(exp_info->ops->cache_sgt_mapping &&
 		    (exp_info->ops->pin || exp_info->ops->unpin)))
@@ -643,10 +634,21 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info)
 	if (!try_module_get(exp_info->owner))
 		return ERR_PTR(-ENOENT);
 
+	file = dma_buf_getfile(exp_info->size, exp_info->flags);
+	if (IS_ERR(file)) {
+		ret = PTR_ERR(file);
+		goto err_module;
+	}
+
+	if (!exp_info->resv)
+		alloc_size += sizeof(struct dma_resv);
+	else
+		/* prevent &dma_buf[1] == dma_buf->resv */
+		alloc_size += 1;
 	dmabuf = kzalloc(alloc_size, GFP_KERNEL);
 	if (!dmabuf) {
 		ret = -ENOMEM;
-		goto err_module;
+		goto err_file;
 	}
 
 	dmabuf->priv = exp_info->priv;
@@ -658,43 +660,35 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info)
 	init_waitqueue_head(&dmabuf->poll);
 	dmabuf->cb_in.poll = dmabuf->cb_out.poll = &dmabuf->poll;
 	dmabuf->cb_in.active = dmabuf->cb_out.active = 0;
+	INIT_LIST_HEAD(&dmabuf->attachments);
 
 	if (!resv) {
-		resv = (struct dma_resv *)&dmabuf[1];
-		dma_resv_init(resv);
+		dmabuf->resv = (struct dma_resv *)&dmabuf[1];
+		dma_resv_init(dmabuf->resv);
+	} else {
+		dmabuf->resv = resv;
 	}
-	dmabuf->resv = resv;
 
-	file = dma_buf_getfile(dmabuf, exp_info->flags);
-	if (IS_ERR(file)) {
-		ret = PTR_ERR(file);
+	ret = dma_buf_stats_setup(dmabuf, file);
+	if (ret)
 		goto err_dmabuf;
-	}
 
+	file->private_data = dmabuf;
+	file->f_path.dentry->d_fsdata = dmabuf;
 	dmabuf->file = file;
 
-	INIT_LIST_HEAD(&dmabuf->attachments);
-
 	mutex_lock(&db_list.lock);
 	list_add(&dmabuf->list_node, &db_list.head);
 	mutex_unlock(&db_list.lock);
 
-	ret = dma_buf_stats_setup(dmabuf);
-	if (ret)
-		goto err_sysfs;
-
 	return dmabuf;
 
-err_sysfs:
-	/*
-	 * Set file->f_path.dentry->d_fsdata to NULL so that when
-	 * dma_buf_release() gets invoked by dentry_ops, it exits
-	 * early before calling the release() dma_buf op.
-	 */
-	file->f_path.dentry->d_fsdata = NULL;
-	fput(file);
 err_dmabuf:
+	if (!resv)
+		dma_resv_fini(dmabuf->resv);
 	kfree(dmabuf);
+err_file:
+	fput(file);
 err_module:
 	module_put(exp_info->owner);
 	return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6b74df446694..e3e2e6e3b485 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -195,6 +195,7 @@ 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;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b4f2d61ea0d5..1353ffd08988 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -181,6 +181,7 @@ 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;
@@ -880,6 +881,32 @@ 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)
  */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index bceb1a5b2518..3fdaba56be6f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -801,7 +801,7 @@ static int kfd_build_sysfs_node_entry(struct kfd_topology_device *dev,
 
 		p2plink->attr.name = "properties";
 		p2plink->attr.mode = KFD_SYSFS_FILE_MODE;
-		sysfs_attr_init(&iolink->attr);
+		sysfs_attr_init(&p2plink->attr);
 		ret = sysfs_create_file(p2plink->kobj, &p2plink->attr);
 		if (ret < 0)
 			return ret;
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 86bc23a67d97..1b7f20a9d4ae 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -5835,7 +5835,8 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 		 */
 		DRM_DEBUG_DRIVER("No preferred mode found\n");
 	} else {
-		recalculate_timing = is_freesync_video_mode(&mode, aconnector);
+		recalculate_timing = amdgpu_freesync_vid_mode &&
+				 is_freesync_video_mode(&mode, aconnector);
 		if (recalculate_timing) {
 			freesync_mode = get_highest_refresh_rate_mode(aconnector, false);
 			drm_mode_copy(&saved_mode, &mode);
@@ -6986,7 +6987,7 @@ static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connect
 	struct amdgpu_dm_connector *amdgpu_dm_connector =
 		to_amdgpu_dm_connector(connector);
 
-	if (!edid)
+	if (!(amdgpu_freesync_vid_mode && edid))
 		return;
 
 	if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10)
@@ -8850,7 +8851,8 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		 * TODO: Refactor this function to allow this check to work
 		 * in all conditions.
 		 */
-		if (dm_new_crtc_state->stream &&
+		if (amdgpu_freesync_vid_mode &&
+		    dm_new_crtc_state->stream &&
 		    is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state))
 			goto skip_modeset;
 
@@ -8885,7 +8887,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		if (!dm_old_crtc_state->stream)
 			goto skip_modeset;
 
-		if (dm_new_crtc_state->stream &&
+		if (amdgpu_freesync_vid_mode && dm_new_crtc_state->stream &&
 		    is_timing_unchanged_for_freesync(new_crtc_state,
 						     old_crtc_state)) {
 			new_crtc_state->mode_changed = false;
@@ -8897,7 +8899,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 			set_freesync_fixed_config(dm_new_crtc_state);
 
 			goto skip_modeset;
-		} else if (aconnector &&
+		} else if (amdgpu_freesync_vid_mode && aconnector &&
 			   is_freesync_video_mode(&new_crtc_state->mode,
 						  aconnector)) {
 			struct drm_display_mode *high_mode;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index 5af601cff1a0..b53feeaf5cf1 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -6257,12 +6257,12 @@ bool dml32_CalculateDETSwathFillLatencyHiding(unsigned int NumberOfActiveSurface
 	double SwathSizePerSurfaceC[DC__NUM_DPP__MAX];
 	bool NotEnoughDETSwathFillLatencyHiding = false;
 
-	/* calculate sum of single swath size for all pipes in bytes*/
+	/* calculate sum of single swath size for all pipes in bytes */
 	for (k = 0; k < NumberOfActiveSurfaces; k++) {
-		SwathSizePerSurfaceY[k] += SwathHeightY[k] * SwathWidthY[k] * BytePerPixelInDETY[k] * NumOfDPP[k];
+		SwathSizePerSurfaceY[k] = SwathHeightY[k] * SwathWidthY[k] * BytePerPixelInDETY[k] * NumOfDPP[k];
 
 		if (SwathHeightC[k] != 0)
-			SwathSizePerSurfaceC[k] += SwathHeightC[k] * SwathWidthC[k] * BytePerPixelInDETC[k] * NumOfDPP[k];
+			SwathSizePerSurfaceC[k] = SwathHeightC[k] * SwathWidthC[k] * BytePerPixelInDETC[k] * NumOfDPP[k];
 		else
 			SwathSizePerSurfaceC[k] = 0;
 
diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c
index 9f1c209d9251..0616b73175f3 100644
--- a/drivers/gpu/drm/i915/gvt/debugfs.c
+++ b/drivers/gpu/drm/i915/gvt/debugfs.c
@@ -151,6 +151,22 @@ DEFINE_SIMPLE_ATTRIBUTE(vgpu_scan_nonprivbb_fops,
 			vgpu_scan_nonprivbb_get, vgpu_scan_nonprivbb_set,
 			"0x%llx\n");
 
+static int vgpu_status_get(void *data, u64 *val)
+{
+	struct intel_vgpu *vgpu = (struct intel_vgpu *)data;
+
+	*val = 0;
+
+	if (test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
+		*val |= (1 << INTEL_VGPU_STATUS_ATTACHED);
+	if (test_bit(INTEL_VGPU_STATUS_ACTIVE, vgpu->status))
+		*val |= (1 << INTEL_VGPU_STATUS_ACTIVE);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(vgpu_status_fops, vgpu_status_get, NULL, "0x%llx\n");
+
 /**
  * intel_gvt_debugfs_add_vgpu - register debugfs entries for a vGPU
  * @vgpu: a vGPU
@@ -162,11 +178,12 @@ void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu)
 	snprintf(name, 16, "vgpu%d", vgpu->id);
 	vgpu->debugfs = debugfs_create_dir(name, vgpu->gvt->debugfs_root);
 
-	debugfs_create_bool("active", 0444, vgpu->debugfs, &vgpu->active);
 	debugfs_create_file("mmio_diff", 0444, vgpu->debugfs, vgpu,
 			    &vgpu_mmio_diff_fops);
 	debugfs_create_file("scan_nonprivbb", 0644, vgpu->debugfs, vgpu,
 			    &vgpu_scan_nonprivbb_fops);
+	debugfs_create_file("status", 0644, vgpu->debugfs, vgpu,
+			    &vgpu_status_fops);
 }
 
 /**
@@ -175,8 +192,13 @@ void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu)
  */
 void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu)
 {
-	debugfs_remove_recursive(vgpu->debugfs);
-	vgpu->debugfs = NULL;
+	struct intel_gvt *gvt = vgpu->gvt;
+	struct drm_minor *minor = gvt->gt->i915->drm.primary;
+
+	if (minor->debugfs_root && gvt->debugfs_root) {
+		debugfs_remove_recursive(vgpu->debugfs);
+		vgpu->debugfs = NULL;
+	}
 }
 
 /**
@@ -199,6 +221,10 @@ void intel_gvt_debugfs_init(struct intel_gvt *gvt)
  */
 void intel_gvt_debugfs_clean(struct intel_gvt *gvt)
 {
-	debugfs_remove_recursive(gvt->debugfs_root);
-	gvt->debugfs_root = NULL;
+	struct drm_minor *minor = gvt->gt->i915->drm.primary;
+
+	if (minor->debugfs_root) {
+		debugfs_remove_recursive(gvt->debugfs_root);
+		gvt->debugfs_root = NULL;
+	}
 }
diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c
index 355f1c0e8664..ffe41e9be04f 100644
--- a/drivers/gpu/drm/i915/gvt/dmabuf.c
+++ b/drivers/gpu/drm/i915/gvt/dmabuf.c
@@ -134,7 +134,8 @@ static void dmabuf_gem_object_free(struct kref *kref)
 	struct list_head *pos;
 	struct intel_vgpu_dmabuf_obj *dmabuf_obj;
 
-	if (vgpu && vgpu->active && !list_empty(&vgpu->dmabuf_obj_list_head)) {
+	if (vgpu && test_bit(INTEL_VGPU_STATUS_ACTIVE, vgpu->status) &&
+	    !list_empty(&vgpu->dmabuf_obj_list_head)) {
 		list_for_each(pos, &vgpu->dmabuf_obj_list_head) {
 			dmabuf_obj = list_entry(pos, struct intel_vgpu_dmabuf_obj, list);
 			if (dmabuf_obj == obj) {
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 51e5e8fb505b..4ec85308379a 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -55,7 +55,7 @@ static bool intel_gvt_is_valid_gfn(struct intel_vgpu *vgpu, unsigned long gfn)
 	int idx;
 	bool ret;
 
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return false;
 
 	idx = srcu_read_lock(&kvm->srcu);
@@ -1178,7 +1178,7 @@ static int is_2MB_gtt_possible(struct intel_vgpu *vgpu,
 	if (!HAS_PAGE_SIZES(vgpu->gvt->gt->i915, I915_GTT_PAGE_SIZE_2M))
 		return 0;
 
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return -EINVAL;
 	pfn = gfn_to_pfn(vgpu->vfio_device.kvm, ops->get_pfn(entry));
 	if (is_error_noslot_pfn(pfn))
@@ -1209,10 +1209,8 @@ static int split_2MB_gtt_entry(struct intel_vgpu *vgpu,
 	for_each_shadow_entry(sub_spt, &sub_se, sub_index) {
 		ret = intel_gvt_dma_map_guest_page(vgpu, start_gfn + sub_index,
 						   PAGE_SIZE, &dma_addr);
-		if (ret) {
-			ppgtt_invalidate_spt(spt);
-			return ret;
-		}
+		if (ret)
+			goto err;
 		sub_se.val64 = se->val64;
 
 		/* Copy the PAT field from PDE. */
@@ -1231,6 +1229,17 @@ static int split_2MB_gtt_entry(struct intel_vgpu *vgpu,
 	ops->set_pfn(se, sub_spt->shadow_page.mfn);
 	ppgtt_set_shadow_entry(spt, se, index);
 	return 0;
+err:
+	/* Cancel the existing addess mappings of DMA addr. */
+	for_each_present_shadow_entry(sub_spt, &sub_se, sub_index) {
+		gvt_vdbg_mm("invalidate 4K entry\n");
+		ppgtt_invalidate_pte(sub_spt, &sub_se);
+	}
+	/* Release the new allocated spt. */
+	trace_spt_change(sub_spt->vgpu->id, "release", sub_spt,
+		sub_spt->guest_page.gfn, sub_spt->shadow_page.type);
+	ppgtt_free_spt(sub_spt);
+	return ret;
 }
 
 static int split_64KB_gtt_entry(struct intel_vgpu *vgpu,
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index 62823c0e13ab..2d65800d8e93 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -172,13 +172,18 @@ struct intel_vgpu_submission {
 
 #define KVMGT_DEBUGFS_FILENAME		"kvmgt_nr_cache_entries"
 
+enum {
+	INTEL_VGPU_STATUS_ATTACHED = 0,
+	INTEL_VGPU_STATUS_ACTIVE,
+	INTEL_VGPU_STATUS_NR_BITS,
+};
+
 struct intel_vgpu {
 	struct vfio_device vfio_device;
 	struct intel_gvt *gvt;
 	struct mutex vgpu_lock;
 	int id;
-	bool active;
-	bool attached;
+	DECLARE_BITMAP(status, INTEL_VGPU_STATUS_NR_BITS);
 	bool pv_notified;
 	bool failsafe;
 	unsigned int resetting_eng;
@@ -467,7 +472,7 @@ void intel_vgpu_write_fence(struct intel_vgpu *vgpu,
 
 #define for_each_active_vgpu(gvt, vgpu, id) \
 	idr_for_each_entry((&(gvt)->vgpu_idr), (vgpu), (id)) \
-		for_each_if(vgpu->active)
+		for_each_if(test_bit(INTEL_VGPU_STATUS_ACTIVE, vgpu->status))
 
 static inline void intel_vgpu_write_pci_bar(struct intel_vgpu *vgpu,
 					    u32 offset, u32 val, bool low)
@@ -725,7 +730,7 @@ static inline bool intel_gvt_mmio_is_cmd_write_patch(
 static inline int intel_gvt_read_gpa(struct intel_vgpu *vgpu, unsigned long gpa,
 		void *buf, unsigned long len)
 {
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return -ESRCH;
 	return vfio_dma_rw(&vgpu->vfio_device, gpa, buf, len, false);
 }
@@ -743,7 +748,7 @@ static inline int intel_gvt_read_gpa(struct intel_vgpu *vgpu, unsigned long gpa,
 static inline int intel_gvt_write_gpa(struct intel_vgpu *vgpu,
 		unsigned long gpa, void *buf, unsigned long len)
 {
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return -ESRCH;
 	return vfio_dma_rw(&vgpu->vfio_device, gpa, buf, len, true);
 }
diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c
index a6b2021b665f..68eca023bbc6 100644
--- a/drivers/gpu/drm/i915/gvt/interrupt.c
+++ b/drivers/gpu/drm/i915/gvt/interrupt.c
@@ -433,7 +433,7 @@ static int inject_virtual_interrupt(struct intel_vgpu *vgpu)
 	 * enabled by guest. so if msi_trigger is null, success is still
 	 * returned and don't inject interrupt into guest.
 	 */
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return -ESRCH;
 	if (vgpu->msi_trigger && eventfd_signal(vgpu->msi_trigger, 1) != 1)
 		return -EFAULT;
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index f5451adcd489..8ae7039b3683 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -638,7 +638,7 @@ static bool __kvmgt_vgpu_exist(struct intel_vgpu *vgpu)
 
 	mutex_lock(&vgpu->gvt->lock);
 	for_each_active_vgpu(vgpu->gvt, itr, id) {
-		if (!itr->attached)
+		if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, itr->status))
 			continue;
 
 		if (vgpu->vfio_device.kvm == itr->vfio_device.kvm) {
@@ -655,9 +655,6 @@ static int intel_vgpu_open_device(struct vfio_device *vfio_dev)
 {
 	struct intel_vgpu *vgpu = vfio_dev_to_vgpu(vfio_dev);
 
-	if (vgpu->attached)
-		return -EEXIST;
-
 	if (!vgpu->vfio_device.kvm ||
 	    vgpu->vfio_device.kvm->mm != current->mm) {
 		gvt_vgpu_err("KVM is required to use Intel vGPU\n");
@@ -667,14 +664,14 @@ static int intel_vgpu_open_device(struct vfio_device *vfio_dev)
 	if (__kvmgt_vgpu_exist(vgpu))
 		return -EEXIST;
 
-	vgpu->attached = true;
-
 	vgpu->track_node.track_write = kvmgt_page_track_write;
 	vgpu->track_node.track_flush_slot = kvmgt_page_track_flush_slot;
 	kvm_get_kvm(vgpu->vfio_device.kvm);
 	kvm_page_track_register_notifier(vgpu->vfio_device.kvm,
 					 &vgpu->track_node);
 
+	set_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status);
+
 	debugfs_create_ulong(KVMGT_DEBUGFS_FILENAME, 0444, vgpu->debugfs,
 			     &vgpu->nr_cache_entries);
 
@@ -698,11 +695,10 @@ static void intel_vgpu_close_device(struct vfio_device *vfio_dev)
 {
 	struct intel_vgpu *vgpu = vfio_dev_to_vgpu(vfio_dev);
 
-	if (!vgpu->attached)
-		return;
-
 	intel_gvt_release_vgpu(vgpu);
 
+	clear_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status);
+
 	debugfs_remove(debugfs_lookup(KVMGT_DEBUGFS_FILENAME, vgpu->debugfs));
 
 	kvm_page_track_unregister_notifier(vgpu->vfio_device.kvm,
@@ -718,8 +714,6 @@ static void intel_vgpu_close_device(struct vfio_device *vfio_dev)
 	vgpu->dma_addr_cache = RB_ROOT;
 
 	intel_vgpu_release_msi_eventfd_ctx(vgpu);
-
-	vgpu->attached = false;
 }
 
 static u64 intel_vgpu_get_bar_addr(struct intel_vgpu *vgpu, int bar)
@@ -1512,9 +1506,6 @@ static void intel_vgpu_remove(struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = dev_get_drvdata(&mdev->dev);
 
-	if (WARN_ON_ONCE(vgpu->attached))
-		return;
-
 	vfio_unregister_group_dev(&vgpu->vfio_device);
 	vfio_put_device(&vgpu->vfio_device);
 }
@@ -1559,7 +1550,7 @@ int intel_gvt_page_track_add(struct intel_vgpu *info, u64 gfn)
 	struct kvm_memory_slot *slot;
 	int idx;
 
-	if (!info->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, info->status))
 		return -ESRCH;
 
 	idx = srcu_read_lock(&kvm->srcu);
@@ -1589,8 +1580,8 @@ int intel_gvt_page_track_remove(struct intel_vgpu *info, u64 gfn)
 	struct kvm_memory_slot *slot;
 	int idx;
 
-	if (!info->attached)
-		return 0;
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, info->status))
+		return -ESRCH;
 
 	idx = srcu_read_lock(&kvm->srcu);
 	slot = gfn_to_memslot(kvm, gfn);
@@ -1668,7 +1659,7 @@ int intel_gvt_dma_map_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 	struct gvt_dma *entry;
 	int ret;
 
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return -EINVAL;
 
 	mutex_lock(&vgpu->cache_lock);
@@ -1714,8 +1705,8 @@ int intel_gvt_dma_pin_guest_page(struct intel_vgpu *vgpu, dma_addr_t dma_addr)
 	struct gvt_dma *entry;
 	int ret = 0;
 
-	if (!vgpu->attached)
-		return -ENODEV;
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
+		return -EINVAL;
 
 	mutex_lock(&vgpu->cache_lock);
 	entry = __gvt_cache_find_dma_addr(vgpu, dma_addr);
@@ -1742,7 +1733,7 @@ void intel_gvt_dma_unmap_guest_page(struct intel_vgpu *vgpu,
 {
 	struct gvt_dma *entry;
 
-	if (!vgpu->attached)
+	if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status))
 		return;
 
 	mutex_lock(&vgpu->cache_lock);
@@ -1778,7 +1769,7 @@ static void intel_gvt_test_and_emulate_vblank(struct intel_gvt *gvt)
 	idr_for_each_entry((&(gvt)->vgpu_idr), (vgpu), (id)) {
 		if (test_and_clear_bit(INTEL_GVT_REQUEST_EMULATE_VBLANK + id,
 				       (void *)&gvt->service_request)) {
-			if (vgpu->active)
+			if (test_bit(INTEL_VGPU_STATUS_ACTIVE, vgpu->status))
 				intel_vgpu_emulate_vblank(vgpu);
 		}
 	}
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 9cd8fcbf7cad..f4055804aad1 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -695,6 +695,7 @@ intel_vgpu_shadow_mm_pin(struct intel_vgpu_workload *workload)
 
 	if (workload->shadow_mm->type != INTEL_GVT_MM_PPGTT ||
 	    !workload->shadow_mm->ppgtt_mm.shadowed) {
+		intel_vgpu_unpin_mm(workload->shadow_mm);
 		gvt_vgpu_err("workload shadow ppgtt isn't ready\n");
 		return -EINVAL;
 	}
@@ -865,7 +866,8 @@ pick_next_workload(struct intel_gvt *gvt, struct intel_engine_cs *engine)
 		goto out;
 	}
 
-	if (!scheduler->current_vgpu->active ||
+	if (!test_bit(INTEL_VGPU_STATUS_ACTIVE,
+		      scheduler->current_vgpu->status) ||
 	    list_empty(workload_q_head(scheduler->current_vgpu, engine)))
 		goto out;
 
diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c
index 3c529c2705dd..a5497440484f 100644
--- a/drivers/gpu/drm/i915/gvt/vgpu.c
+++ b/