[igt-dev] [PATCH i-g-t] tests/perf: Fix timeout for RC6 re-enabling
Lionel Landwerlin
lionel.g.landwerlin at intel.com
Tue Apr 16 14:03:45 UTC 2019
On 16/04/2019 14:51, Imre Deak wrote:
> After closing the perf stream the parking the GPU engines may easily
> take more than 1 second: releasing the FD itself results in a new
> request submission via i915_perf_release()->i915_oa_stream_destroy()->
> gen8_disable_metric_set(). That means a >1sec delay for the delayed
> unpark to be called due to the delay from
>
> queue_delayed_work(retire_work, round_jiffies_up_relative(HZ))
>
> + the delay from
>
> mod_delayed_work(idle_work, msecs_to_jiffies(100))
>
> Scheduling may push this delay even further, I measured >2sec delays on
> my GLK. Based on this give some slack for the RC6 residency counter to
> change.
Thanks a lot for looking into this.
Whatever you do it will probably be better than the current state :)
Adding a couple of people who are also looking at fixing this test.
-Lionel
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103179
> Cc: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
> Cc: Chris Wilson <chris at chris-wilson.co.uk>
> Signed-off-by: Imre Deak <imre.deak at intel.com>
> ---
> tests/perf.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/tests/perf.c b/tests/perf.c
> index 4307d36a..5dc10108 100644
> --- a/tests/perf.c
> +++ b/tests/perf.c
> @@ -3591,6 +3591,7 @@ test_rc6_disable(void)
> };
> unsigned long n_events_start, n_events_end;
> unsigned long rc6_enabled;
> + int i;
>
> rc6_enabled = 0;
> igt_sysfs_scanf(sysfs, "power/rc6_enable", "%lu", &rc6_enabled);
> @@ -3606,8 +3607,12 @@ test_rc6_disable(void)
> __perf_close(stream_fd);
>
> n_events_start = rc6_residency_ms();
> - nanosleep(&(struct timespec){ .tv_sec = 1, .tv_nsec = 0 }, NULL);
> - n_events_end = rc6_residency_ms();
> + for (i = 0; i < 5; i++) {
> + nanosleep(&(struct timespec){ .tv_sec = 1, .tv_nsec = 0 }, NULL);
> + n_events_end = rc6_residency_ms();
> + if (n_events_end != n_events_start)
> + break;
> + }
> igt_assert_neq(n_events_end - n_events_start, 0);
> }
>
More information about the igt-dev
mailing list