On Mar 3, 2010, at 11:08 PM, Chris Lattner wrote:

>> +/* SSE4 Rounding Intrinsics.  */
>> +static inline __m128 __attribute__((__always_inline__, __nodebug__))
>> +_mm_round_ps (__m128 __V, const int __M)
>> +{
>> +  return (__m128) __builtin_ia32_roundps ((__v4sf)__V, __M);
>> +}
> 
> 'const int __M' is the same as 'int __M'.  Do these intrinsics require a 
> literal integer value?  If so, they may not be working, they might need to be 
> written as a macro.

*grumble* You're right, they aren't. :)

Want them as a macro always or iirc some platforms do the ifdef OPTIMIZE trick 
to leave it as a function most of the time.

-eric
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to