[llvm] r214892 - Optimize vector fabs of bitcasted constant integer values.

Sanjay Patel spatel at rotateright.com
Tue Aug 5 13:26:31 PDT 2014


Thanks all for the thorough review!
Test cases improved with r214921.


On Tue, Aug 5, 2014 at 2:18 PM, Renato Golin <renato.golin at linaro.org>
wrote:

> On 5 August 2014 20:43, Chandler Carruth <chandlerc at google.com> wrote:
> >> 2. The 'CHECK-NOT' in the ARM cases was requested by Renato. Should the
> >> equivalent also go in the X86 tests?
> >
> > It usually isn't needed when you can just check all of the instructions
> for
> > very small and focused functions (like these).
>
> Yeah, because of the ordering, the idea was to have only a check-not
> for the vabs instruction to make sure it was simplified correctly.
> Either check-dag for both movs or check-not for vabs would do.
>
> cheers,
> --renato
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140805/ad09ea4f/attachment.html>


More information about the llvm-commits mailing list