[PATCH] [AArch64] Simplify scalar dup by lane0 for fp

Jiangning Liu liujiangning1 at gmail.com
Thu Dec 19 18:15:21 PST 2013


Hi,

For scalar fp data type, we use s0/d0, which is actually a shared register
with VFP v0, so dup by lane0 for fp should not generate redundant scalar
dup instruction.

-- 
Thanks,
-Jiangning
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131220/38452b88/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dup_fp_lane0_clang.patch
Type: text/x-patch
Size: 1558 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131220/38452b88/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: add_missing_test_llvm.patch
Type: text/x-patch
Size: 16111 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131220/38452b88/attachment-0001.bin>


More information about the llvm-commits mailing list