[PATCH] X86: VZeroUpperInserter: shortcut should not trigger if we have any function live-ins.

[Lang Hames]

Sounds good to me. Thanks for fixing this. :)

On Wed, Jan 7, 2015 at 4:27 PM, Matthias Braun <matze at braunis.de> wrote:

> Thanks for the review. The testcase (CodeGen/X86/vec_extract-avx.ll) that
> made me create this patch is already in the repository and will break if
> subreg liveness is enabled for X86.
>
> Greetings
>     Matthias
>
>
> On Jan 7, 2015, at 4:18 PM, Lang Hames <lhames at gmail.com> wrote:
>
> LGTM. Do you have a test case for this?
>
> - Lang.
>
> On Tue, Dec 23, 2014 at 5:58 PM, Matthias Braun <matze at braunis.de> wrote:
>
>> The later code adds a vzeroupper if a ymm register is live-in to the
>> function, so the shortcut should needs to check for that as well before
>> deciding to bail out early.
>>
>> Greetings
>>     Matthias
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>
>>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150107/48df9f31/attachment.html>