[PATCH] ARM: Form paired load after stack coloring.

Quentin Colombet qcolombet at apple.com
Thu Jun 20 09:59:02 PDT 2013


Ping?

-Quentin

On Jun 17, 2013, at 10:37 AM, Quentin Colombet <qcolombet at apple.com> wrote:

> Hi Evan,
> 
> Have you got a chance to look at that?
> 
> Just to give a bit of context, I would like to remove the check on getValue to be able to form paired loads when it is not available. Indeed, I think we should not have a dependency on that here, as we do not make any use of it (we are not checking for aliases for instance).
> But maybe I missed something obvious, what is why we would need your comment.
> 
> Thanks,
> 
> -Quentin
> 
> On Jun 11, 2013, at 3:50 PM, Renato Golin <renato.golin at linaro.org> wrote:
> 
>> Evan,
>> 
>> Can you shed some light on what the getValue() is testing in this case?
>> 
>> +  // Must sure the base address satisfies i64 ld / st alignment requirement.
>> +  if (!Op0->hasOneMemOperand() ||
>> +      !Op0->memoperands_begin()->getValue() ||
>> +      Op0->memoperands_begin()->isVolatile())
>> 
>> cheers,
>> --renato
> 
> _______________________________________________
> 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/20130620/879854be/attachment.html>


More information about the llvm-commits mailing list