drm/i915: Move user fault tracking to a separate list

We want to decouple RPM and struct_mutex, but currently RPM has to walk
the list of bound objects and remove userspace mmapping before we
suspend (otherwise userspace may continue to access the GTT whilst it is
powered down). This currently requires the struct_mutex to walk the
bound_list, but if we move that to a separate list and lock we can take
the first step towards removing the struct_mutex.

v2: Split runtime suspend unmapping vs regular unmapping, to make the
locking (and barriers) clearer. Add the object to the userfault_list
prior to inserting the first PTE, the race between add/revoke depends
upon struct_mutex for regular unmappings and rpm for runtime-suspend.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> #v1
Link: http://patchwork.freedesktop.org/patch/msgid/20161024124218.18252-1-chris@chris-wilson.co.uk

drivers/gpu/drm/i915/i915_debugfs.c

@@ -107,7 +107,7 @@ static char get_tiling_flag(struct drm_i915_gem_object *obj)
 
 static char get_global_flag(struct drm_i915_gem_object *obj)
 {
-	return obj->fault_mappable ? 'g' : ' ';
+	return !list_empty(&obj->userfault_link) ? 'g' : ' ';
 }
 
 static char get_pin_mapped_flag(struct drm_i915_gem_object *obj)

drivers/gpu/drm/i915/i915_drv.h

@@ -1361,6 +1361,14 @@ struct i915_gem_mm {
 	 */
 	struct list_head unbound_list;
 
+	/** Protects access to the userfault_list */
+	spinlock_t userfault_lock;
+
+	/** List of all objects in gtt_space, currently mmaped by userspace.
+	 * All objects within this list must also be on bound_list.
+	 */
+	struct list_head userfault_list;
+
 	/** Usable portion of the GTT for GEM */
 	unsigned long stolen_base; /* limited to low memory (32-bit) */
 
@@ -2205,6 +2213,11 @@ struct drm_i915_gem_object {
 	struct drm_mm_node *stolen;
 	struct list_head global_list;
 
+	/**
+	 * Whether the object is currently in the GGTT mmap.
+	 */
+	struct list_head userfault_link;
+
 	/** Used in execbuf to temporarily hold a ref */
 	struct list_head obj_exec_link;
 
@@ -2232,13 +2245,6 @@ struct drm_i915_gem_object {
 	 */
 	unsigned int madv:2;
 
-	/**
-	 * Whether the current gtt mapping needs to be mappable (and isn't just
-	 * mappable by accident). Track pin and fault separate for a more
-	 * accurate mappable working set.
-	 */
-	unsigned int fault_mappable:1;
-
 	/*
 	 * Is the object to be mapped as read-only to the GPU
 	 * Only honoured if hardware has relevant pte bit

drivers/gpu/drm/i915/i915_gem.c

@@ -1839,16 +1839,19 @@ int i915_gem_fault(struct vm_area_struct *area, struct vm_fault *vmf)
 	if (ret)
 		goto err_unpin;
 
+	/* Mark as being mmapped into userspace for later revocation */
+	spin_lock(&dev_priv->mm.userfault_lock);
+	if (list_empty(&obj->userfault_link))
+		list_add(&obj->userfault_link, &dev_priv->mm.userfault_list);
+	spin_unlock(&dev_priv->mm.userfault_lock);
+
 	/* Finally, remap it using the new GTT offset */
 	ret = remap_io_mapping(area,
 			       area->vm_start + (vma->ggtt_view.params.partial.offset << PAGE_SHIFT),
 			       (ggtt->mappable_base + vma->node.start) >> PAGE_SHIFT,
 			       min_t(u64, vma->size, area->vm_end - area->vm_start),
 			       &ggtt->mappable);
-	if (ret)
-		goto err_unpin;
 
-	obj->fault_mappable = true;
 err_unpin:
 	__i915_vma_unpin(vma);
 err_unlock:
@@ -1916,13 +1919,22 @@ int i915_gem_fault(struct vm_area_struct *area, struct vm_fault *vmf)
 void
 i915_gem_release_mmap(struct drm_i915_gem_object *obj)
 {
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+	bool zap = false;
+
 	/* Serialisation between user GTT access and our code depends upon
 	 * revoking the CPU's PTE whilst the mutex is held. The next user
 	 * pagefault then has to wait until we release the mutex.
 	 */
-	lockdep_assert_held(&obj->base.dev->struct_mutex);
+	lockdep_assert_held(&i915->drm.struct_mutex);
 
-	if (!obj->fault_mappable)
+	spin_lock(&i915->mm.userfault_lock);
+	if (!list_empty(&obj->userfault_link)) {
+		list_del_init(&obj->userfault_link);
+		zap = true;
+	}
+	spin_unlock(&i915->mm.userfault_lock);
+	if (!zap)
 		return;
 
 	drm_vma_node_unmap(&obj->base.vma_node,
@@ -1936,17 +1948,26 @@ i915_gem_release_mmap(struct drm_i915_gem_object *obj)
 	 * memory writes before touching registers / GSM.
 	 */
 	wmb();
-
-	obj->fault_mappable = false;
 }
 
 void
 i915_gem_release_all_mmaps(struct drm_i915_private *dev_priv)
 {
 	struct drm_i915_gem_object *obj;
 
-	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list)
-		i915_gem_release_mmap(obj);
+	spin_lock(&dev_priv->mm.userfault_lock);
+	while ((obj = list_first_entry_or_null(&dev_priv->mm.userfault_list,
+					       struct drm_i915_gem_object,
+					       userfault_link))) {
+		list_del_init(&obj->userfault_link);
+		spin_unlock(&dev_priv->mm.userfault_lock);
+
+		drm_vma_node_unmap(&obj->base.vma_node,
+				   obj->base.dev->anon_inode->i_mapping);
+
+		spin_lock(&dev_priv->mm.userfault_lock);
+	}
+	spin_unlock(&dev_priv->mm.userfault_lock);
 }
 
 /**
@@ -4108,6 +4129,7 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj,
 	int i;
 
 	INIT_LIST_HEAD(&obj->global_list);
+	INIT_LIST_HEAD(&obj->userfault_link);
 	for (i = 0; i < I915_NUM_ENGINES; i++)
 		init_request_active(&obj->last_read[i],
 				    i915_gem_object_retire__read);
@@ -4521,6 +4543,7 @@ int i915_gem_init(struct drm_device *dev)
 	int ret;
 
 	mutex_lock(&dev->struct_mutex);
+	spin_lock_init(&dev_priv->mm.userfault_lock);
 
 	if (!i915.enable_execlists) {
 		dev_priv->gt.resume = intel_legacy_submission_resume;
@@ -4640,6 +4663,7 @@ i915_gem_load_init(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.bound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.fence_list);
+	INIT_LIST_HEAD(&dev_priv->mm.userfault_list);
 	INIT_DELAYED_WORK(&dev_priv->gt.retire_work,
 			  i915_gem_retire_work_handler);
 	INIT_DELAYED_WORK(&dev_priv->gt.idle_work,

drivers/gpu/drm/i915/i915_gem_evict.c

@@ -56,7 +56,7 @@ mark_free(struct i915_vma *vma, unsigned int flags, struct list_head *unwind)
 	if (WARN_ON(!list_empty(&vma->exec_list)))
 		return false;
 
-	if (flags & PIN_NONFAULT && vma->obj->fault_mappable)
+	if (flags & PIN_NONFAULT && !list_empty(&vma->obj->userfault_link))
 		return false;
 
 	list_add(&vma->exec_list, unwind);

drivers/gpu/drm/i915/i915_gem_fence.c

@@ -391,7 +391,7 @@ void i915_gem_restore_fences(struct drm_device *dev)
 		 */
 		if (vma && !i915_gem_object_is_tiled(vma->obj)) {
 			GEM_BUG_ON(!reg->dirty);
-			GEM_BUG_ON(vma->obj->fault_mappable);
+			GEM_BUG_ON(!list_empty(&vma->obj->userfault_link));
 
 			list_move(&reg->link, &dev_priv->mm.fence_list);
 			vma->fence = NULL;