| Message ID | 20200205095441.1769599-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/2] drm/i915: Flush execution tasklets before checking request status | expand |
Chris Wilson <chris@chris-wilson.co.uk> writes: > Rather than flushing the submission tasklets just before we sleep, flush > before we check the request status. Ideally this gives us a moment to > process the tasklets after sleeping just before we timeout. Makes sense to check the possibly most recent state. Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > --- > drivers/gpu/drm/i915/i915_request.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c > index f56b046a32de..5c2bb0b9478b 100644 > --- a/drivers/gpu/drm/i915/i915_request.c > +++ b/drivers/gpu/drm/i915/i915_request.c > @@ -1564,6 +1564,7 @@ long i915_request_wait(struct i915_request *rq, > goto out; > > for (;;) { > + intel_engine_flush_submission(rq->engine); > set_current_state(state); > > if (i915_request_completed(rq)) { > @@ -1581,7 +1582,6 @@ long i915_request_wait(struct i915_request *rq, > break; > } > > - intel_engine_flush_submission(rq->engine); > timeout = io_schedule_timeout(timeout); > } > __set_current_state(TASK_RUNNING); > -- > 2.25.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoting Mika Kuoppala (2020-02-05 14:42:16) > Chris Wilson <chris@chris-wilson.co.uk> writes: > > > Rather than flushing the submission tasklets just before we sleep, flush > > before we check the request status. Ideally this gives us a moment to > > process the tasklets after sleeping just before we timeout. > > Makes sense to check the possibly most recent state. > > Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> I suppose I should mention a small counter argument is that by kicking ksoftirqd before checking completed, we may add a small amount of latency to the ready client. A compromise is to kick before timeout? -Chris
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index f56b046a32de..5c2bb0b9478b 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -1564,6 +1564,7 @@ long i915_request_wait(struct i915_request *rq, goto out; for (;;) { + intel_engine_flush_submission(rq->engine); set_current_state(state); if (i915_request_completed(rq)) { @@ -1581,7 +1582,6 @@ long i915_request_wait(struct i915_request *rq, break; } - intel_engine_flush_submission(rq->engine); timeout = io_schedule_timeout(timeout); } __set_current_state(TASK_RUNNING);
Rather than flushing the submission tasklets just before we sleep, flush before we check the request status. Ideally this gives us a moment to process the tasklets after sleeping just before we timeout. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> --- drivers/gpu/drm/i915/i915_request.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)