Re: [PATCH] lib: kunit: add bitfield test conversion to KUnit

From: Brendan Higgins
Date: Sat Oct 10 2020 - 01:47:43 EST


On Thu, Sep 17, 2020 at 3:00 AM Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote:
>
> On Wed, 2020-08-19 at 14:10 -0700, Brendan Higgins wrote:
> > On Wed, Jul 29, 2020 at 10:58 AM Vitor Massaru Iha <vitor@xxxxxxxxxxx> wrote:
> > > This adds the conversion of the runtime tests of test_bitfield,
> > > from `lib/test_bitfield.c` to KUnit tests.
> > >
> > > Please apply this commit first (linux-kselftest/kunit-fixes):
> > > 3f37d14b8a3152441f36b6bc74000996679f0998 kunit: kunit_config: Fix parsing of CONFIG options with space
> > >
> > > Code Style Documentation: [0]
> > >
> > > Signed-off-by: Vitor Massaru Iha <vitor@xxxxxxxxxxx>
> > > Link: [0] https://lore.kernel.org/linux-kselftest/20200620054944.167330-1-davidgow@xxxxxxxxxx/T/#u
> >
> > Reviewed-by: Brendan Higgins <brendanhiggins@xxxxxxxxxx>
> >
> > Probably still want a review from Johannes though.
>
> Huh, sorry, this slipped through the cracks.
>
> Yeah, don't really care, looks fine to me? I'm not familiar with the
> kunit infrastructure much yet.

Cool, well in that case I will apply it.

> Not sure I see much value in converting TEST_BITFIELD_COMPILE to a
> KUNIT_CASE though, because anyway it will not compile if you enable
> that? IOW, just leaving the function there without any KUNIT_CASE()
> reference to it should be fine and saves you an ifdef ...

Well I think it is also the case that we only want to count the test
case if it actually has everything to run; that is a point that is
somewhat up in the air. David is exploring adding the concept of
"skipped" tests to KUnit, but we don't have that yet.

Cheers!