On 20/02/2017 16:07, Chris Wilson wrote:
On Mon, Feb 20, 2017 at 03:58:12PM +0000, Tvrtko Ursulin wrote:
On 30/01/2017 19:44, Chris Wilson wrote:
On Mon, Jan 30, 2017 at 06:07:29PM +0000, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
i915_gem_ring_notify is more appropriate since we do not have
the request information at this point, but it is simply a
signal from the engine that some request has been completed.
v2:
* Always trace and log if there were any waiters.
Agreed, I'd sketched that change, well to always emit the tracepoint when
the interrupt fired. Whether it was in vain is icing on the cake - I've
some sketches of keeping the user-interrupt alive until the following rq
to short-circuit some ping-ponging or irq enable state.
Shall we progress this series or leave it for some future convenient time?
iirc, it marries well with the patch (on list now) to keep the interrupt
alive for an extra tick. Now's as good as time as any to push the
improvements to rq tracing.
Cool, patches just need some r-b's then. :) Let me see if they still
apply and resend the series.
Regards,
Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
