[lld] r297008 - [ELF] - Make Bss and BssRelRo sections to be synthetic (#2).

Rui Ueyama via llvm-commits llvm-commits at lists.llvm.org
Wed Mar 8 15:29:19 PST 2017


My wild guess is that this patch changed the way how copy relocated symbols
are represented (previously we keep shared symbols as-is, but this patch
changed that to DefinedRegular using replaceBody) and that's most likely
the cause of the issue, but I may be wrong.

On Wed, Mar 8, 2017 at 3:22 PM, Rafael Avila de Espindola <
rafael.espindola at gmail.com> wrote:

> Cool. I will check if there is an available aarch64 machine at the gcc
> farm I can use to reproduce the issue.
>
> Renato Golin via llvm-commits <llvm-commits at lists.llvm.org> writes:
>
> > On 8 March 2017 at 18:08, Rui Ueyama <ruiu at google.com> wrote:
> >> It's not clear what are broken. Looks like the failing tests are not
> LLD's
> >> tests. Could you give me a bit more info?
> >
> > I did a bisect and found that this was indeed the commit that broke
> > it, thus why I asked for reverting.
> >
> > Now that the bot is back to green, I'll reapply the commit+fix and
> > report the failures on the bug.
> >
> > Thanks!
> > --renato
> > _______________________________________________
> > llvm-commits mailing list
> > llvm-commits at lists.llvm.org
> > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170308/bed40af0/attachment.html>


More information about the llvm-commits mailing list