diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2018-02-11 10:52:37 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2018-02-11 10:52:37 +0000 |
| commit | 0be5567a891c192582ea87d67b58f18f00dcf038 (patch) | |
| tree | a27f38c984d237bdb85019bed0ff7b62f1273485 /llvm/test/CodeGen/X86/combine-smax.ll | |
| parent | 91e2b9d081829af95a7b84ec5393e33539e9130e (diff) | |
| download | bcm5719-llvm-0be5567a891c192582ea87d67b58f18f00dcf038.tar.gz bcm5719-llvm-0be5567a891c192582ea87d67b58f18f00dcf038.zip | |
[X86][SSE] Enable SMIN/SMAX/UMIN/UMAX custom lowering for all legal types
This allows us to recognise more saturation patterns and also simplify some MINMAX codegen that was failing to combine CMPGE comparisons to a legal CMPGT.
Differential Revision: https://reviews.llvm.org/D43014
llvm-svn: 324837
Diffstat (limited to 'llvm/test/CodeGen/X86/combine-smax.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/combine-smax.ll | 7 |
1 files changed, 1 insertions, 6 deletions
diff --git a/llvm/test/CodeGen/X86/combine-smax.ll b/llvm/test/CodeGen/X86/combine-smax.ll index a7f339dba5d..0c0ffc642e6 100644 --- a/llvm/test/CodeGen/X86/combine-smax.ll +++ b/llvm/test/CodeGen/X86/combine-smax.ll @@ -13,12 +13,7 @@ define <16 x i8> @test_v16i8_nosignbit(<16 x i8> %a, <16 x i8> %b) { ; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15] ; SSE2-NEXT: pand %xmm2, %xmm0 ; SSE2-NEXT: pand %xmm2, %xmm1 -; SSE2-NEXT: movdqa %xmm0, %xmm2 -; SSE2-NEXT: pcmpgtb %xmm1, %xmm2 -; SSE2-NEXT: pand %xmm2, %xmm0 -; SSE2-NEXT: pandn %xmm1, %xmm2 -; SSE2-NEXT: por %xmm0, %xmm2 -; SSE2-NEXT: movdqa %xmm2, %xmm0 +; SSE2-NEXT: pmaxub %xmm1, %xmm0 ; SSE2-NEXT: retq ; ; SSE41-LABEL: test_v16i8_nosignbit: |
