[igt-dev] [PATCH i-g-t] i915/selftest: Allow filtering of individual subtests

Katarzyna Dec katarzyna.dec at intel.com
Thu Jan 24 14:28:27 UTC 2019


On Tue, Jan 22, 2019 at 06:35:21PM +0000, Chris Wilson wrote:
> Take an environment variable, SELFTESTS=foo,bar, and pass that along to
> the kernel (as i915.st_filter=foo,bar) to provide fine grained test
> selection. This can be either as an exact match to select only that
> test, or to exclude only test. For example,
> 
> SELFTESTS=igt_vma_create,igt_vma_pin1 i915_selftest --run mock_vma
> SELFTESTS=!igt_vma_create i915_selftest --run mock_vma
> 
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>

If I understand correctly we can choose specific subtest or exclude specific
subtest from binary?

That is a really nice feature :) I didn't have chance to use it, but LGTM.

Kasia :)
> ---
>  tests/i915/selftest.c | 18 ++++++++++++------
>  1 file changed, 12 insertions(+), 6 deletions(-)
> 
> diff --git a/tests/i915/selftest.c b/tests/i915/selftest.c
> index 80e515c61..b8d7f0af1 100644
> --- a/tests/i915/selftest.c
> +++ b/tests/i915/selftest.c
> @@ -28,10 +28,16 @@ IGT_TEST_DESCRIPTION("Basic unit tests for i915.ko");
>  
>  igt_main
>  {
> -	igt_kselftests("i915",
> -		       "mock_selftests=-1 disable_display=1",
> -		       NULL, "mock");
> -	igt_kselftests("i915",
> -		       "live_selftests=-1 disable_display=1",
> -		       "live_selftests", "live");
> +	const char *env = getenv("SELFTESTS");
> +	char opts[1024];
> +
> +	igt_assert(snprintf(opts, sizeof(opts),
> +			    "mock_selftests=-1 disable_display=1 st_filter=%s",
> +			    env) < sizeof(opts));
> +	igt_kselftests("i915", opts, NULL, "mock");
> +
> +	igt_assert(snprintf(opts, sizeof(opts),
> +			    "live_selftests=-1 disable_display=1 st_filter=%s",
> +			    env) < sizeof(opts));
> +	igt_kselftests("i915", opts, "live_selftests", "live");
>  }
> -- 
> 2.20.1
> 
> _______________________________________________
> igt-dev mailing list
> igt-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/igt-dev


More information about the igt-dev mailing list