OSDN Git Service

InstCombine now optimizes vector udiv by power of 2 to shifts
authorPete Cooper <peter_cooper@apple.com>
Mon, 7 Nov 2011 23:04:49 +0000 (23:04 +0000)
committerStephen Hines <srhines@google.com>
Mon, 14 Nov 2011 17:11:57 +0000 (09:11 -0800)
commit5f77a98a2be9fa01f3284f323e32a94ca6c1ce6d
tree6e05d7850dbe43a590aa609996a18a451aac8ace
parent2278edcb4a5a89eabc1bbe5a54535c7848e4f61a
InstCombine now optimizes vector udiv by power of 2 to shifts

Fixes r8429

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144036 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/InstCombine/InstCombineMulDivRem.cpp
test/CodeGen/X86/vec_udiv_to_shift.ll [new file with mode: 0644]