[Intel-gfx] [PATCH 14/17] drm/i915: Nonblocking request submission

Joonas Lahtinen joonas.lahtinen at linux.intel.com
Tue Aug 30 08:35:14 UTC 2016


On su, 2016-08-28 at 21:46 +0100, Chris Wilson wrote:
> @@ -442,6 +442,18 @@ i915_gem_request_alloc(struct intel_engine_cs *engine,
>  	 */
>  	req->head = req->ring->tail;
>  
> +	prev = i915_gem_active_peek(&engine->last_request,
> +				    &req->i915->drm.struct_mutex);
> +	if (prev) {
> +		ret = i915_sw_fence_await_sw_fence(&req->submit,
> +						   &prev->submit,
> +						   GFP_KERNEL);
> +		if (ret < 0) {
> +			i915_add_request(req);

As discussed in IRC, this should not be necessary at all. We're still
allocating the request, and the fence_await call failed (not setting up
any dependencies to our request implied) so nobody should know of our
request yet.

That fixed,

Reviewed-by: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>

Regards, Joonas
-- 
Joonas Lahtinen
Open Source Technology Center
Intel Corporation


More information about the Intel-gfx mailing list