Rewrite shift/and/compare sequences to promote better licm of the RHS.
authorChris Lattner <sabre@nondot.org>
Mon, 18 Sep 2006 04:22:48 +0000 (04:22 +0000)
committerChris Lattner <sabre@nondot.org>
Mon, 18 Sep 2006 04:22:48 +0000 (04:22 +0000)
commit65b72baa595052c2179d36a584a24cd293e983d0
treee0f3f106622d1a140ca647f55b67e6419f122227
parent20fdab6fbf582116a99a185afbcba3ca34655fe1
Rewrite shift/and/compare sequences to promote better licm of the RHS.
Use isLogicalShift/isArithmeticShift to simplify code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@30448 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/Scalar/InstructionCombining.cpp