[Intel-gfx] [PATCH i-g-t] i915/gem_eio: Assert the hanging request is correctly identified

Chris Wilson chris at chris-wilson.co.uk
Fri Jul 5 08:01:32 UTC 2019


Quoting Chris Wilson (2019-07-02 15:23:39)
> When forcing a reset, it is crucial that the kernel correctly identifies
> the injected hang. Verify this is the case for reset-stress.
> 
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> ---
> Keep the igt_spin_free at the end to avoid issues where we fail to
> bypass the guilty batch.

Ping?

> ---
>  tests/i915/gem_eio.c | 17 +++++++++--------
>  1 file changed, 9 insertions(+), 8 deletions(-)
> 
> diff --git a/tests/i915/gem_eio.c b/tests/i915/gem_eio.c
> index 5396a04e2..dc7afb0fd 100644
> --- a/tests/i915/gem_eio.c
> +++ b/tests/i915/gem_eio.c
> @@ -175,7 +175,7 @@ static igt_spin_t * __spin_poll(int fd, uint32_t ctx, unsigned long flags)
>         struct igt_spin_factory opts = {
>                 .ctx = ctx,
>                 .engine = flags,
> -               .flags = IGT_SPIN_FAST,
> +               .flags = IGT_SPIN_FAST | IGT_SPIN_FENCE_OUT,
>         };
>  
>         if (gem_can_store_dword(fd, opts.engine))
> @@ -270,12 +270,12 @@ static void check_wait(int fd, uint32_t bo, unsigned int wait, igt_stats_t *st)
>                 igt_stats_push(st, igt_nsec_elapsed(&ts));
>  }
>  
> -static void check_wait_elapsed(int fd, igt_stats_t *st)
> +static void check_wait_elapsed(const char *prefix, int fd, igt_stats_t *st)
>  {
>         double med, max, limit;
>  
> -       igt_info("Completed %d resets, wakeups took %.3f+-%.3fms (min:%.3fms, median:%.3fms, max:%.3fms)\n",
> -                st->n_values,
> +       igt_info("%s: completed %d resets, wakeups took %.3f+-%.3fms (min:%.3fms, median:%.3fms, max:%.3fms)\n",
> +                prefix, st->n_values,
>                  igt_stats_get_mean(st)*1e-6,
>                  igt_stats_get_std_deviation(st)*1e-6,
>                  igt_stats_get_min(st)*1e-6,
> @@ -715,8 +715,8 @@ static void test_inflight_internal(int fd, unsigned int wait)
>         close(fd);
>  }
>  
> -static void reset_stress(int fd,
> -                        uint32_t ctx0, unsigned int engine,
> +static void reset_stress(int fd, uint32_t ctx0,
> +                        const char *name, unsigned int engine,
>                          unsigned int flags)
>  {
>         const uint32_t bbe = MI_BATCH_BUFFER_END;
> @@ -759,6 +759,7 @@ static void reset_stress(int fd,
>  
>                 /* Wedge after a small delay. */
>                 check_wait(fd, obj.handle, 100e3, &stats);
> +               igt_assert_eq(sync_fence_status(hang->out_fence), -EIO);
>  
>                 /* Unwedge by forcing a reset. */
>                 igt_assert(i915_reset_control(true));
> @@ -782,7 +783,7 @@ static void reset_stress(int fd,
>                 igt_spin_free(fd, hang);
>                 gem_context_destroy(fd, ctx);
>         }
> -       check_wait_elapsed(fd, &stats);
> +       check_wait_elapsed(name, fd, &stats);
>         igt_stats_fini(&stats);
>  
>         gem_close(fd, obj.handle);
> @@ -797,7 +798,7 @@ static void test_reset_stress(int fd, unsigned int flags)
>         unsigned int engine;
>  
>         for_each_engine(fd, engine)
> -               reset_stress(fd, ctx0, engine, flags);
> +               reset_stress(fd, ctx0, e__->name, engine, flags);
>  
>         gem_context_destroy(fd, ctx0);
>  }
> -- 
> 2.20.1
> 


More information about the Intel-gfx mailing list