OSDN Git Service

[X86][SSE] Prefer to combine shuffles to VZEXT over VZEXT_MOVL.
authorSimon Pilgrim <llvm-dev@redking.me.uk>
Tue, 21 Feb 2017 15:09:00 +0000 (15:09 +0000)
committerSimon Pilgrim <llvm-dev@redking.me.uk>
Tue, 21 Feb 2017 15:09:00 +0000 (15:09 +0000)
commitd14347a186058f1105e87b13f6dd7704ade9687e
treea04551c70a5161ac135b89a56b1e6ac9894709f3
parent70dcac1118ac77bd9690a62e106d2117dc4754d7
[X86][SSE] Prefer to combine shuffles to VZEXT over VZEXT_MOVL.

This matches what is already done during shuffle lowering and helps prevent the need for a zero-vector in cases where shuffles match both patterns.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@295723 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp
test/CodeGen/X86/vector-shuffle-combining-sse41.ll