diff options
author | dim <dim@FreeBSD.org> | 2012-04-14 13:54:10 +0000 |
---|---|---|
committer | dim <dim@FreeBSD.org> | 2012-04-14 13:54:10 +0000 |
commit | 1fc08f5e9ef733ef1ce6f363fecedc2260e78974 (patch) | |
tree | 19c69a04768629f2d440944b71cbe90adae0b615 /test/CodeGen/X86/blend-msb.ll | |
parent | 07637c87f826cdf411f0673595e9bc92ebd793f2 (diff) | |
download | FreeBSD-src-1fc08f5e9ef733ef1ce6f363fecedc2260e78974.zip FreeBSD-src-1fc08f5e9ef733ef1ce6f363fecedc2260e78974.tar.gz |
Vendor import of llvm trunk r154661:
http://llvm.org/svn/llvm-project/llvm/trunk@r154661
Diffstat (limited to 'test/CodeGen/X86/blend-msb.ll')
-rw-r--r-- | test/CodeGen/X86/blend-msb.ll | 37 |
1 files changed, 37 insertions, 0 deletions
diff --git a/test/CodeGen/X86/blend-msb.ll b/test/CodeGen/X86/blend-msb.ll new file mode 100644 index 0000000..3a10c70 --- /dev/null +++ b/test/CodeGen/X86/blend-msb.ll @@ -0,0 +1,37 @@ +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=corei7 -promote-elements -mattr=+sse41 | FileCheck %s + + +; In this test we check that sign-extend of the mask bit is performed by +; shifting the needed bit to the MSB, and not using shl+sra. + +;CHECK: vsel_float +;CHECK: pslld +;CHECK-NEXT: blendvps +;CHECK: ret +define <4 x float> @vsel_float(<4 x float> %v1, <4 x float> %v2) { + %vsel = select <4 x i1> <i1 true, i1 false, i1 false, i1 false>, <4 x float> %v1, <4 x float> %v2 + ret <4 x float> %vsel +} + +;CHECK: vsel_4xi8 +;CHECK: pslld +;CHECK-NEXT: blendvps +;CHECK: ret +define <4 x i8> @vsel_4xi8(<4 x i8> %v1, <4 x i8> %v2) { + %vsel = select <4 x i1> <i1 true, i1 false, i1 false, i1 false>, <4 x i8> %v1, <4 x i8> %v2 + ret <4 x i8> %vsel +} + + +; We do not have native support for v8i16 blends and we have to use the +; blendvb instruction or a sequence of NAND/OR/AND. Make sure that we do not r +; reduce the mask in this case. +;CHECK: vsel_8xi16 +;CHECK: psllw +;CHECK: psraw +;CHECK: pblendvb +;CHECK: ret +define <8 x i16> @vsel_8xi16(<8 x i16> %v1, <8 x i16> %v2) { + %vsel = select <8 x i1> <i1 true, i1 false, i1 false, i1 false, i1 true, i1 false, i1 false, i1 false>, <8 x i16> %v1, <8 x i16> %v2 + ret <8 x i16> %vsel +} |