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]

Re: [PATCH] A new meta intrinsic header file for current and future x86 instrinsics.


Guo, Xuepeng wrote:

This patch will replace gmmintrin.h with avxintrin.h. And implement a new file named immintrin.h. And from now on never use avxintrin.h directly, include immintrin.h instead. OK for mainline?

Thanks,
Xuepeng Guo

2008-11-05 H.J. Lu <hongjiu.lu@intel.com>
Xuepeng Guo <xuepeng.guo@intel.com>
* config.gcc (extra_headers): For x86 and x86-64, remove
gmmintrin.h, add immintrin.h and avxintrin.h.
* config/i386/avxintrin.h: New.
* config/i386/immintrin.h: Likewise.
* config/i386/gmmintrin.h: Removed.
* config/i386/avxintrin.h: Don't include intrinsics if
_IMMINTRIN_H_INCLUDED is undedined.

testsuite/ChangeLog:


* gcc.target/i386/avx-1.c: Include <immintrin.h> instead of
<gmmintrin.h>.
* g++.dg/other/i386-5.C: Likewise.
* g++.dg/other/i386-6.C: Likewise.
* gcc.target/i386/avx-2.c: Likewise.
* gcc.target/i386/m256-check.h: Likewise.

If there are no further comments or objections from AMD people in the next 48 hours, and since this header file is intended for _Intel_ intrinsics, it is OK for mainline.


BTW: I guess there will be a header file (ammintrin.h ?) introduced to include AMD defined intrinsics, so gcc can still add x86intrin.h that will include both immintrin.h and eventual ammintrin.h. Hopefully, these two intrinsics won't step onto each other toes.

I hope this eventual solution is acceptable to both contributors.

Thanks,
Uros.


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