On Fri, Oct 16, 2015 at 4:08 AM, Emil Velikov <[email protected]>
wrote:

> On 15 October 2015 at 19:25, Dylan Baker <[email protected]> wrote:
> > Yes please. Not running all of the subtests (or at least returning
> > values for them) breaks assumptions made by the framework.
> >
> How exactly doesn't the framework count the subtests ? Does it
> detect/parse piglit_report_subtest_result() usage or does it use some
> other method ? Afaict the former is not that commonly used as it
> should have :'(
>
> Thanks
> Emil
>

Well, the framework doesn't consider a test a subtest unless it uses the
piglit_report_subtest_result helper (or rolls the same functionality by
hand). Basically the framework treats tests with subtests as groups, and
the subtests as full tests, so leaving out subtests will push a crash into
the enabled/disabled page instead of the regressions/fixes page.
_______________________________________________
Piglit mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/piglit

Reply via email to