Quoting Mika Kuoppala (2018-09-20 11:38:08)
> Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> writes:
> > + end_time = jiffies + i915_selftest.timeout_jiffies;
> > + for_each_prime_number_from(prime, 2, 8192) {
> > + times[1] = ktime_get_raw();
> > +
> > + for (n = 0; n < prime; n++) {
> > + rq = i915_request_alloc(engine, ctx[n % nctx]);
> > + if (IS_ERR(rq)) {
> > + err = PTR_ERR(rq);
> > + goto out_unlock;
> > + }
> > +
> > + /*
> > + * This space is left intentionally blank.
> > + *
> > + * We do not actually want to perform any
> > + * action with this request, we just want
> > + * to measure the latency in allocation
> > + * and submission of our breadcrumbs -
> > + * ensuring that the bare request is sufficient
> > + * for the system to work (i.e. proper HEAD
> > + * tracking of the rings, interrupt handling,
> > + * etc). It also gives us the lowest bounds
> > + * for latency.
> > + */
> > +
> > + i915_request_add(rq);
>
> We get the breadcrumbs in place, with the wa_tails and so on so this
> should work without any nops nor bb ends.
Right, not one batch.
> > + }
> > + i915_request_wait(rq,
> > + I915_WAIT_LOCKED,
> > + MAX_SCHEDULE_TIMEOUT);
Fwiw, a lesson learnt from buggy code is never to use
MAX_SCHEDULE_TIMEOUT in tests :)
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
