Quoting Jason Ekstrand (2017-08-10 06:35:43)
> +igt_main
> +{
> +       int fd;
> +
> +       igt_fixture {
> +               fd = drm_open_driver(DRIVER_ANY);

DRIVER_ANY | DRIVER_VGEM

(ANY really means KMS)

> +               igt_require(has_syncobj_wait(fd));
> +               igt_require_sw_sync();

We should add something like

igt_require(has_timer_resolution(CLOCK_MONOTONIC, SHORT_TIME_NSEC / 2));

static bool has_timer_resolution(clockid_t clk, uint64_t target_ns)
{
        struct timespec res;

        if (clock_getres(clk, &res))
                return false; /* clock unavailable */

        return res.tv_sec * NSEC_PER_SEC + res.tv_nsec <= target_ns;
}

That'll rule out testing on pnv most likely.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reply via email to