diff options
| author | Bruno Cardoso Lopes <bruno.cardoso@gmail.com> | 2011-07-27 00:56:34 +0000 |
|---|---|---|
| committer | Bruno Cardoso Lopes <bruno.cardoso@gmail.com> | 2011-07-27 00:56:34 +0000 |
| commit | cea34e41fa3bde8a7271c725abc6bb79c31377c9 (patch) | |
| tree | 8f5f96c85a01817d810253ccd1f2c276c47bbc2b /test/CodeGen/X86 | |
| parent | cd9e5aed53987b70fa19396424b913f2e9d50f60 (diff) | |
| download | external_llvm-cea34e41fa3bde8a7271c725abc6bb79c31377c9.zip external_llvm-cea34e41fa3bde8a7271c725abc6bb79c31377c9.tar.gz external_llvm-cea34e41fa3bde8a7271c725abc6bb79c31377c9.tar.bz2 | |
The vpermilps and vpermilpd have different behaviour regarding the
usage of the shuffle bitmask. Both work in 128-bit lanes without
crossing, but in the former the mask of the high part is the same
used by the low part while in the later both lanes have independent
masks. Handle this properly and and add support for vpermilpd.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136200 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86')
| -rw-r--r-- | test/CodeGen/X86/avx-vpermil.ll | 16 |
1 files changed, 16 insertions, 0 deletions
diff --git a/test/CodeGen/X86/avx-vpermil.ll b/test/CodeGen/X86/avx-vpermil.ll new file mode 100644 index 0000000..d57c173 --- /dev/null +++ b/test/CodeGen/X86/avx-vpermil.ll @@ -0,0 +1,16 @@ +; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=corei7-avx -mattr=+avx | FileCheck %s + +; CHECK: vpermilps +define <8 x float> @funcA(<8 x float> %a) nounwind uwtable readnone ssp { +entry: + %shuffle = shufflevector <8 x float> %a, <8 x float> undef, <8 x i32> <i32 1, i32 2, i32 3, i32 1, i32 5, i32 6, i32 7, i32 5> + ret <8 x float> %shuffle +} + +; CHECK: vpermilpd +define <4 x double> @funcB(<4 x double> %a) nounwind uwtable readnone ssp { +entry: + %shuffle = shufflevector <4 x double> %a, <4 x double> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 3> + ret <4 x double> %shuffle +} + |