[Intel-gfx] [RFC 3/5] drm/i915: Concurrent context uAPI

Chris Wilson chris at chris-wilson.co.uk
Mon Nov 13 13:28:49 UTC 2017


Quoting Tvrtko Ursulin (2017-11-13 13:09:07)
> From: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> 
> When submitting work against existing contexts userspace knows that:
> 
>  a) context state can be relied upon
>  b) submitted batches will complete in order.
> 
> By adding an option to mark its context as "concurrent" we allow userspace
> to notify the kernel that it does not care about both of the above. i915
> is then at liberty to break the above guarantees if and when appropriate.

We want a short paragraph on the merits of ctx vs execbuf, and of the
nature of the different scheduling algorithms that may be enacted. i.e.
the likely direction we will be going in after this, and whether this is
temporary api or the start of grander things.
-Chris


More information about the Intel-gfx mailing list