Re: [Intel-gfx] [PATCH 1/5] drm/i915/execlists: Skip redundant resubmission

On 03/10/2019 22:00, Chris Wilson wrote:

If we unwind the active requests, and on resubmission discover that we intend to preempt the active context with itself, simply skip the ELSP submission.

Signed-off-by: Chris Wilson chris@chris-wilson.co.uk

drivers/gpu/drm/i915/gt/intel_lrc.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 431d3b8c3371..3cfea1758fd2 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -1739,11 +1739,26 @@ static void execlists_dequeue(struct intel_engine_cs *engine)

if (submit) { *port = execlists_schedule_in(last, port - execlists->pending);

• memset(port + 1, 0, (last_port - port) * sizeof(*port));
  

  execlists->switch_priority_hint = switch_prio(engine, *execlists->pending);

• /*
  

•  * Skip if we ended up with exactly the same set of requests,
  

•  * e.g. trying to timeslice a pair of ordered contexts
  

•  */
  

• if (!memcmp(execlists->active, execlists->pending,
  

• 	    (port - execlists->pending + 1) * sizeof(*port))) {
  

• 	do
  

• 		execlists_schedule_out(fetch_and_zero(port));
  

• 	while (port-- != execlists->pending);
  

• 	goto skip_submit;
  

• }
  

• memset(port + 1, 0, (last_port - port) * sizeof(*port));
  

  execlists_submit_ports(engine); } else {

+skip_submit: ring_set_paused(engine, 0); } }

A little bit of tracepoint noise with in/out but looks correct after I read the new code. And I wonder if trace.pl still works after last couple months of refactors.. :)

Reviewed-by: Tvrtko Ursulin tvrtko.ursulin@intel.com

Regards,

Tvrtko