[libcxx-commits] [PATCH] D99095: [libcxx] [test] Add XFAIL LIBCXX-WINDOWS-FIXME in 124 tests that fail in the CI configuration

Louis Dionne via Phabricator via libcxx-commits libcxx-commits at lists.llvm.org
Thu Mar 25 11:42:09 PDT 2021

ldionne added inline comments.

Comment at: libcxx/test/std/language.support/support.limits/support.limits.general/complex.version.pass.cpp:16
mstorsjo wrote:
> ldionne wrote:
> > mstorsjo wrote:
> > > Mordante wrote:
> > > > mstorsjo wrote:
> > > > > Mordante wrote:
> > > > > > This file and the other `*.version.pass.cpp` are automatically generated by `utils/generate_feature_test_macro_components.py`. If a developer runs this script these changes will accidentally be reverted. I'm not sure what would be the best fix for now.
> > > > > Those particular xfails can be removed with D99213
> > > > Since D99213 has landed, could you commit a fix for them?
> > > Didn't D99213 already remove them as part of that commit? At least the xfail in std/language.support/support.limits/support.limits.general/complex.version.pass.cpp should have been removed now. Or did I leave stray whitespace changes behind compared with how they're autogenerated?
> > @Mordante  Thanks for noticing this. I missed that during my review because I glanced at the changes and didn't think about the auto-generated files.
> > 
> > @mstorsjo As long as you don't see any diff when you run the script, all's good. If you do see a diff, then we need to fix that.
> > 
> > I'll work on adding a job that catches exactly these issues.
> CI for enforcing that would be good yeah. And nice that it turned out to be easily fixable...
> I don't see any diff after running that script now, so everything should be good now?
No diff = everything's good. Thanks!

  rG LLVM Github Monorepo



More information about the libcxx-commits mailing list