This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[AVX]: Correct AVX integer vector patterns


Hi,

AVX doesn't have 256bit operation on integer vectors.
We use floating point ones for OR, XOR, NAND and
AND. I am checking in this patch to generate 128bit
integer instructions for AVX.


-- 
H.J.
---
2008-06-09  H.J. Lu  <hongjiu.lu@intel.com>

	* config/i386/sse.md (AVXMODEI): Removed.
	(*avx_nand<mode>3): Replace AVXMODEI with AVX256MODEI.
	(*avx_<code><mode>3): Likewise.
	(*avx_<code><mode>3): New.

Attachment: a.txt
Description: Text document


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]