diff options
| author | Adam Nemet <anemet@apple.com> | 2014-06-18 16:51:07 +0000 |
|---|---|---|
| committer | Adam Nemet <anemet@apple.com> | 2014-06-18 16:51:07 +0000 |
| commit | ded81a810c93b94ee4041b0409daf4553e947e9f (patch) | |
| tree | 6cf795edc87444488c1c688da4b9eb60a7b32e97 /llvm/lib/Target/X86/X86InstrAVX512.td | |
| parent | 9aa09ef30f0bc65aeb848f0cdc6b5987fc884d16 (diff) | |
| download | bcm5719-llvm-ded81a810c93b94ee4041b0409daf4553e947e9f.tar.gz bcm5719-llvm-ded81a810c93b94ee4041b0409daf4553e947e9f.zip | |
[X86] AVX512: Specify compressed displacement for vmovntdqa
Use the max 64-bit element size with EVEX_CD8. This should work since element
size is ignored for a full-vector access (FVM).
llvm-svn: 211175
Diffstat (limited to 'llvm/lib/Target/X86/X86InstrAVX512.td')
| -rw-r--r-- | llvm/lib/Target/X86/X86InstrAVX512.td | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/llvm/lib/Target/X86/X86InstrAVX512.td b/llvm/lib/Target/X86/X86InstrAVX512.td index 4b9f560597f..618f42abe94 100644 --- a/llvm/lib/Target/X86/X86InstrAVX512.td +++ b/llvm/lib/Target/X86/X86InstrAVX512.td @@ -1798,7 +1798,7 @@ def VMOVNTDQAZrm : AVX5128I<0x2A, MRMSrcMem, (outs VR512:$dst), "vmovntdqa\t{$src, $dst|$dst, $src}", [(set VR512:$dst, (int_x86_avx512_movntdqa addr:$src))]>, - EVEX, EVEX_V512; + EVEX, EVEX_V512, EVEX_CD8<64, CD8VF>; //===----------------------------------------------------------------------===// // AVX-512 - Integer arithmetic |
