[igt-dev] [PATCH v5 i-g-t 0/8] Introduce KUnit

[Piatkowski, Dominik Karol]

Hi Janusz,

> -----Original Message-----
> From: Janusz Krzysztofik <janusz.krzysztofik at linux.intel.com>
> Sent: wtorek, 6 czerwca 2023 09:47
> To: igt-dev at lists.freedesktop.org; Piatkowski, Dominik Karol
> <dominik.karol.piatkowski at intel.com>; Mauro Carvalho Chehab
> <mauro.chehab at linux.intel.com>
> Cc: Janusz Krzysztofik <janusz.krzysztofik at linux.intel.com>
> Subject: Re: [PATCH v5 i-g-t 0/8] Introduce KUnit
> 
> On Monday, 5 June 2023 12:47:08 CEST Dominik Karol Piatkowski wrote:
> > This series is a continuation of Isabella's work on introducing KUnit
> > to IGT.
> >
> > Sample drm_buddy output:
> > 	Starting subtest: all-tests
> > 	[thread:8003] TAP version 1
> > 	[thread:8003] Executing 6 tests in: drm_buddy
> > 	Starting dynamic subtest: drm_buddy__drm_test_buddy_alloc_limit
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_limit:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest:
> drm_buddy__drm_test_buddy_alloc_range
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_range:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest:
> drm_buddy__drm_test_buddy_alloc_optimistic
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_optimistic:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest:
> drm_buddy__drm_test_buddy_alloc_pessimistic
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_pessimistic:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest:
> drm_buddy__drm_test_buddy_alloc_smoke
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_smoke:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest:
> drm_buddy__drm_test_buddy_alloc_pathological
> > 	Dynamic subtest drm_buddy__drm_test_buddy_alloc_pathological:
> SUCCESS (0.000s)
> > 	Starting dynamic subtest: drm_buddy
> > 	Dynamic subtest drm_buddy: SUCCESS (0.000s)
> > 	Subtest all-tests: SUCCESS (0.707s)
> 
> It would be nice if we could see CI results from the new kunit subtests
> before accepting the series.  Unfortunately, none of the tests seems to have
> been executed within the pre-merge run.  Can we fix that?

AFAIK, Kconfig used by CI does not have required params set (CONFIG_KUNIT and CONFIG_DRM_I915_KUNIT_SELFTESTS). I submitted a merge request to enable these params, but it was not merged yet. The best option I can think of right now, is a trybot run tested with Kconfig patch enabling these params. Would that be ok with you?

Thanks,
Dominik

> 
> Thanks,
> Janusz
> 
> >
> > The issue of possibility of too many sublevels occurrence is solved by
> > name concatenation.
> >
> > Cc: Janusz Krzysztofik <janusz.krzysztofik at linux.intel.com>
> > Cc: Mauro Carvalho Chehab <mauro.chehab at linux.intel.com>
> >
> > Dominik Karol Piatkowski (2):
> >   Change logic of ktap parser to run on a thread
> >   lib/igt_kmod: fix nesting igt_fixture in igt_subtest
> >
> > Isabella Basso (4):
> >   lib/igt_kmod: rename kselftest functions to ktest
> >   lib/igt_kmod.c: check if module is builtin before attempting to unload
> >     it
> >   lib/igt_kmod: add compatibility for KUnit
> >   tests: DRM selftests: switch to KUnit
> >
> > Mauro Carvalho Chehab (2):
> >   lib/igt_kmod: place KUnit tests on a subtest
> >   kunit tests: add an optional name for the selftests
> >
> >  lib/igt_kmod.c       | 146 +++++++++-
> >  lib/igt_kmod.h       |  14 +-
> >  lib/igt_ktap.c       | 615
> +++++++++++++++++++++++++++++++++++++++++++
> >  lib/igt_ktap.h       |  50 ++++
> >  lib/meson.build      |   1 +
> >  tests/drm_buddy.c    |   4 +-
> >  tests/drm_mm.c       |   5 +-
> >  tests/kms_selftest.c |  19 ++
> >  8 files changed, 835 insertions(+), 19 deletions(-)  create mode
> > 100644 lib/igt_ktap.c  create mode 100644 lib/igt_ktap.h
> >
> >
> 
> 
>