This is an archive of the discontinued LLVM Phabricator instance.

[ARM]: Extend -mfpu options for half-precision and vfpv3xd
ClosedPublic

Authored by javed.absar on Jun 26 2015, 2:34 AM.

Details

Summary

Some of the the permissible ARM -mfpu (floating-point hardware) options, which are already supported in GCC, are currently not present in llvm/clang. This patch adds, namely, the options 'neon-fp16', 'vfpv3-fp16', 'vfpv3-d16-fp16', 'vfpv3xd' and 'vfpv3xd-fp16. These options are related to half-precision floating-point and single precision.

Diff Detail

Repository
rL LLVM

Event Timeline

javed.absar retitled this revision from to [ARM]: Extend -mfpu options for half-precision and vfpv3xd.
javed.absar updated this object.
javed.absar edited the test plan for this revision. (Show Details)
javed.absar set the repository for this revision to rL LLVM.
javed.absar added a subscriber: Unknown Object (MLST).

Hi Rentao:

Could you please review/accept this patch. You have reviewed it earlier. This is clang tests related to the changes in http://reviews.llvm.org/D10645, which you okay-ed this morning.

Thanks and Best Regards
Javed

rengolin accepted this revision.Jun 26 2015, 5:51 AM
rengolin added a reviewer: rengolin.

LGTM. Thanks!

This revision is now accepted and ready to land.Jun 26 2015, 5:51 AM
This revision was automatically updated to reflect the committed changes.