[PATCH][LoopVectorizer] Enable fsub reductions

[James Molloy]

Hi Arnold,

The attached patch enables floating-point subtraction reductions in fast-math mode.

I'm not sure why it wasn't enabled before - the logic seems sound and I can't find any FP math pitfalls.

This, coupled with making the loop vectorizer run before SLP, should give us about 10% in sphinx3 in SPEC.

Cheers,

James

-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590
ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140807/fd4c313d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fsub-reduction.diff
Type: application/octet-stream
Size: 2193 bytes
Desc: fsub-reduction.diff
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140807/fd4c313d/attachment.obj>