354367414b3a1b04ae2ab92805e4d780546d669b - platform_external_llvm

commit	354367414b3a1b04ae2ab92805e4d780546d669b	[log] [tgz]
author	Simon Pilgrim <llvm-dev@redking.me.uk>	Fri Jul 20 16:55:18 2018 +0000
committer	Simon Pilgrim <llvm-dev@redking.me.uk>	Fri Jul 20 16:55:18 2018 +0000
tree	19fa846f034af61fc951e99f9abf589c55c2eec9
parent	86a77cda68e60807a00ccbf80a5af45a60af34e2 [diff]

[X86][XOP] Fix SUB constant folding for VPSHA/VPSHL shift lowering

We can safely use getConstant here as we're still lowering, which allows constant folding to kick in and simplify the vector shift codegen.

Noticed while working on D49562.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@337578 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Target/X86/X86ISelLowering.cpp[diff]
test/CodeGen/X86/combine-sdiv.ll[diff]
test/CodeGen/X86/vector-shift-ashr-128.ll[diff]
test/CodeGen/X86/vector-shift-ashr-256.ll[diff]
test/CodeGen/X86/vector-shift-lshr-128.ll[diff]
test/CodeGen/X86/vector-shift-lshr-256.ll[diff]

6 files changed

tree: 19fa846f034af61fc951e99f9abf589c55c2eec9