On 1 July 2014 14:56, Gerolf Hoflehner <[email protected]> wrote:
> I filed rdar://17523868 for clang. I think the FE needs to be the first
> phase to look into why 32b and 64b initialization code is different.

Why an internal bug? The test that regressed is in the open source
three, so we should  have a bug about tracking fixing it in
llvm.org/bugs.

If the tests needs to be in clang, that is fine IMHO. We try to have
as few integration tests as possible, but not fewer :-)

Cheers,
Rafael
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to