projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: cc0fbdb) | patch
[AArch64] Improve single vector lane unscaled stores
authorEvandro Menezes <e.menezes@samsung.com>
Tue, 15 May 2018 20:41:12 +0000 (20:41 +0000)
committerEvandro Menezes <e.menezes@samsung.com>
Tue, 15 May 2018 20:41:12 +0000 (20:41 +0000)
commit8d522d811a6d8ec72b10dbd76ba670678bc4251b
treebd40d941d5078ae2f4933c19de4667335c0c2276tree | snapshot
parentcc0fbdb6e4e7dee89b2b83916d53aa82fa9863c1commit | diff
[AArch64] Improve single vector lane unscaled stores

When storing the 0th lane of a vector, use a simpler and usually more
efficient scalar store instead.  In this case, also using the unscaled
offset.

Differential revision: https://reviews.llvm.org/D46762

llvm-svn: 332394
llvm/lib/Target/AArch64/AArch64InstrInfo.td diff | blob | history
llvm/test/CodeGen/AArch64/arm64-st1.ll diff | blob | history
llvm/test/CodeGen/AArch64/fp16-vector-load-store.ll diff | blob | history
Domain: System / Toolchain;