Quoting Tomi Sarvela (2018-07-17 14:12:27) > On 07/17/2018 03:52 PM, Chris Wilson wrote: > > The CI infrastructure is ready and is already running the kernel > > selftests in BAT, so there is no reason to blacklist them from the > > shards. > > > > Signed-off-by: Chris Wilson <[email protected]> > > Cc: Tomi Sarvela <[email protected]> > > Cc: Petri Latvala <[email protected]> > > Cc: Martin Peres <[email protected]> > > --- > > tests/intel-ci/blacklist.txt | 2 -- > > 1 file changed, 2 deletions(-) > > > > diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt > > index d65d8ff35..7faec8b5c 100644 > > --- a/tests/intel-ci/blacklist.txt > > +++ b/tests/intel-ci/blacklist.txt > > @@ -2,8 +2,6 @@ igt@meta_test(@.*)? > > ############################################### > > # GEM > > ############################################### > > -igt@drv_selftest(@.*)? > > -igt@drm_mm(@.*)? > > igt@gem_busy@.*hang.* > > igt@gem_close_race@(?!.*basic).* > > igt@gem_concurrent_blit(@.*)? > > > > From Intel-GFX-CI point of view: > > Because of the test order randomization, this blacklist is used by IGT > builder when creating the randomized testlists for shard nodes, and > builder might or might not have knowledge of i915 or selftests. > > The selftests are scraped to their own testlist in kernel build time (as > they are kernel specific) and used separately after fast-feedback or as > a last shard in a full run. > > I'd advise not to whitelist drv_selftests or drm_mm. If the selftest > coverage is considered full replacement for gem_exec_flush tests, then > remove blacklisting of igt@gem_exec_flush@(?!.*basic).* from > blacklist.txt and remove same tests from fast-feedback.testlist.
Nah, here I was just going "why blacklist these perfectly fine tests? The discrimination has to end!" I just wanted to be sure we were running them everywhere. -Chris _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
