diff options
| author | Nemanja Ivanovic <nemanja.i.ibm@gmail.com> | 2016-07-12 12:16:27 +0000 |
|---|---|---|
| committer | Nemanja Ivanovic <nemanja.i.ibm@gmail.com> | 2016-07-12 12:16:27 +0000 |
| commit | eebbcb6d5722dc976e68afe0a1cdd4a18c40bbf0 (patch) | |
| tree | 441ef6a4cff93d8a40690f79caeadf307ccb0c8f /llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll | |
| parent | acee56854533780d94b1aa38c5aaa33720b3a596 (diff) | |
| download | bcm5719-llvm-eebbcb6d5722dc976e68afe0a1cdd4a18c40bbf0.tar.gz bcm5719-llvm-eebbcb6d5722dc976e68afe0a1cdd4a18c40bbf0.zip | |
[PowerPC] Cannonicalize applicable vector shift immediates as swaps
This patch corresponds to review:
http://reviews.llvm.org/D21358
Vector shifts that have the same semantics as a vector swap are cannonicalized
as such to provide additional opportunities for swap removal optimization to
remove unnecessary swaps.
llvm-svn: 275168
Diffstat (limited to 'llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll')
| -rw-r--r-- | llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll b/llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll index 3610c044b94..74dc1561d9f 100644 --- a/llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll +++ b/llvm/test/CodeGen/PowerPC/p8altivec-shuffles-pred.ll @@ -9,8 +9,8 @@ entry: ret <2 x i32> %strided.vec ; CHECK-LABEL: @test1 -; CHECK: vsldoi [[TGT:[0-9]+]], 2, 2, 8 -; CHECK: vmrghw 2, 2, [[TGT]] +; CHECK: xxswapd 35, 34 +; CHECK: vmrghw 2, 2, 3 ; CHECK: blr } |
