[PATCH] fix for PR19431 - x86 BMI intrinsics are broken

[Sanjay Patel]

Hi Richard,

Please review this revised patch. In this version, I've used #define for
all duplicates except 'bextr' - that intrinsic is actually different
between AMD and Intel. Test cases also updated to account for that
difference.

Thanks!


On Wed, Apr 30, 2014 at 8:39 PM, Richard Smith <richard at metafoo.co.uk>wrote:

> What a mess :(
>
> Can we define one in terms of the other, rather than duplicating the
> definitions of all of these? (Maybe even #define one to the other?)
>
>
> On Wed, Apr 30, 2014 at 3:14 PM, Sanjay Patel <spatel at rotateright.com>wrote:
>
>> Please review (and commit if ok) the attached patch which duplicates all
>> of the existing double-underscore BMI intrinsics with single-underscore
>> variants.
>>
>> More details here:
>> http://llvm.org/bugs/show_bug.cgi?id=19431
>>
>> The equivalent bug in GCC is already fixed:
>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60847
>>
>> _______________________________________________
>> cfe-commits mailing list
>> cfe-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>>
>>
>


-- 
Sanjay Patel
RotateRight, LLC
http://www.rotateright.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140502/a51d3c19/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bmi_rev2.patch
Type: application/octet-stream
Size: 6124 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140502/a51d3c19/attachment.obj>