[Intel-gfx] [PATCH 6/6] drm/i915: Track the last-active inside the i915_vma

[Tvrtko Ursulin]

On 06/07/2018 11:39, Chris Wilson wrote:
> Using a VMA on more than one timeline concurrently is the exception
> rather than the rule (using it concurrently on multiple engines). As we
> expect to only use one active tracker, store the most recently used
> tracker inside the i915_vma itself and only fallback to the rbtree if
> we need a second or more concurrent active trackers.
> 
> v2: Comments on how we overwrite any existing last_active cache.
> v3: __list_del_entry() before list_replace_init() is confusing and, much
> more important, entirely redundant.
> v4: Note that both last_active and the rbtree may be simultaneously
> tracking this timeline, albeit with different requests, and so the vma
> may be retired twice for the same timeline.
> 
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> ---
>   drivers/gpu/drm/i915/i915_vma.c | 56 +++++++++++++++++++++++++++++++--
>   drivers/gpu/drm/i915/i915_vma.h |  1 +
>   2 files changed, 55 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index b4cc98330225..6fbd09d6af28 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -119,6 +119,12 @@ i915_vma_retire(struct i915_gem_active *base, struct i915_request *rq)
>   	__i915_vma_retire(active->vma, rq);
>   }
>   
> +static void
> +i915_vma_last_retire(struct i915_gem_active *base, struct i915_request *rq)
> +{
> +	__i915_vma_retire(container_of(base, struct i915_vma, last_active), rq);
> +}
> +
>   static struct i915_vma *
>   vma_create(struct drm_i915_gem_object *obj,
>   	   struct i915_address_space *vm,
> @@ -136,6 +142,7 @@ vma_create(struct drm_i915_gem_object *obj,
>   
>   	vma->active = RB_ROOT;
>   
> +	init_request_active(&vma->last_active, i915_vma_last_retire);
>   	init_request_active(&vma->last_fence, NULL);
>   	vma->vm = vm;
>   	vma->ops = &vm->vma_ops;
> @@ -895,6 +902,29 @@ static struct i915_gem_active *active_instance(struct i915_vma *vma, u64 idx)
>   {
>   	struct i915_vma_active *active;
>   	struct rb_node **p, *parent;
> +	struct i915_request *old;
> +
> +	/*
> +	 * We track the most recently used timeline to skip a rbtree search
> +	 * for the common case, under typical loads we never need the rbtree
> +	 * at all. We can reuse the last_active slot if it is empty, that is
> +	 * after the previous activity has been retired, or if the active
> +	 * matches the current timeline.
> +	 *
> +	 * Note that we allow the timeline to be active simultaneously in
> +	 * the rbtree and the last_active cache. We do this to avoid having
> +	 * to search and replace the rbtree element for a new timeline, with
> +	 * the cost being that we must be aware that the vma may be retired
> +	 * twice for the same timeline (as the older rbtree element will be
> +	 * retired before the new request added to last_active).
> +	 */
> +	old = i915_gem_active_raw(&vma->last_active,
> +				  &vma->vm->i915->drm.struct_mutex);
> +	if (!old || old->fence.context == idx)
> +		goto out;
> +
> +	/* Move the currently active fence into the rbtree */
> +	idx = old->fence.context;
>   
>   	parent = NULL;
>   	p = &vma->active.rb_node;
> @@ -903,7 +933,7 @@ static struct i915_gem_active *active_instance(struct i915_vma *vma, u64 idx)
>   
>   		active = rb_entry(parent, struct i915_vma_active, node);
>   		if (active->timeline == idx)
> -			return &active->base;
> +			goto replace;
>   
>   		if (active->timeline < idx)
>   			p = &parent->rb_right;
> @@ -922,7 +952,24 @@ static struct i915_gem_active *active_instance(struct i915_vma *vma, u64 idx)
>   	rb_link_node(&active->node, parent, p);
>   	rb_insert_color(&active->node, &vma->active);
>   
> -	return &active->base;
> +replace:
> +	/*
> +	 * Overwrite the previous active slot in the rbtree with last_active,
> +	 * leaving last_active zeroed. If the previous slot is still active,
> +	 * we must be careful as we now only expect to recieve one retire
> +	 * callback not two, and so much undo the active counting for the
> +	 * overwritten slot.
> +	 */
> +	if (i915_gem_active_isset(&active->base)) {
> +		vma->active_count--;
> +		GEM_BUG_ON(!vma->active_count);
> +	}
> +	GEM_BUG_ON(list_empty(&vma->last_active.link));
> +	list_replace_init(&vma->last_active.link, &active->base.link);
> +	active->base.request = fetch_and_zero(&vma->last_active.request);
> +
> +out:
> +	return &vma->last_active;
>   }
>   
>   int i915_vma_move_to_active(struct i915_vma *vma,
> @@ -1002,6 +1049,11 @@ int i915_vma_unbind(struct i915_vma *vma)
>   		 */
>   		__i915_vma_pin(vma);
>   
> +		ret = i915_gem_active_retire(&vma->last_active,
> +					     &vma->vm->i915->drm.struct_mutex);
> +		if (ret)
> +			goto unpin;
> +
>   		rbtree_postorder_for_each_entry_safe(active, n,
>   						     &vma->active, node) {
>   			ret = i915_gem_active_retire(&active->base,
> diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
> index c297b0a0dc47..f06d66377107 100644
> --- a/drivers/gpu/drm/i915/i915_vma.h
> +++ b/drivers/gpu/drm/i915/i915_vma.h
> @@ -97,6 +97,7 @@ struct i915_vma {
>   
>   	unsigned int active_count;
>   	struct rb_root active;
> +	struct i915_gem_active last_active;
>   	struct i915_gem_active last_fence;
>   
>   	/**
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>

Regards,

Tvrtko