[PATCH] Cost model support for lowered math builtins.

[Nadav Rotem]

Thanks :) 

On Feb 28, 2013, at 11:34 AM, Benjamin Kramer <benny.kra at gmail.com> wrote:

> 
> On 28.02.2013, at 20:26, Arnold Schwaighofer <aschwaighofer at apple.com> wrote:
> 
>> This is nice feature, can you brag about this on in docs/Vectorizer.rst?
>> 
>> We also support vectorization of libm functions that can be lowered to intrinsic math functions.
>> 
>> void foo(float *f) {
>> for (unsigned i = 0; i != 1024; ++i)
>> f[i] = floorf(f[i]);
>> }
>> 
>> Or something along those lines.
> 
> Done in r176292 :)
> 
> - Ben
>> 
>> 
>> Thanks,
>> Arnold
>> 
>> On Feb 28, 2013, at 1:11 PM, Benjamin Kramer <benny.kra at gmail.com> wrote:
>> 
>>> 
>>> Closed by commit rL176287 (authored by d0k).
>>> 
>>> CHANGED PRIOR TO COMMIT
>>> http://llvm-reviews.chandlerc.com/D466?vs=1138&id=1154#toc
>>> 
>>> http://llvm-reviews.chandlerc.com/D466
>>> 
>>> COMMIT
>>> http://llvm-reviews.chandlerc.com/rL176287
>>> _______________________________________________
>>> 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/20130228/3261d5c8/attachment.html>