diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-05-26 11:07:22 +0100 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-05-26 11:14:21 +0100 |
| commit | 21aec4fdc5de213a1009bc684adff988c290c6b3 (patch) | |
| tree | 93157ee18b99af134b6bfa6ee37b6a3bcaa70f9d /lldb/source/Interpreter/CommandObject.cpp | |
| parent | 2a41d702be478e3975f49bffec0672c991225786 (diff) | |
| download | llvm-21aec4fdc5de213a1009bc684adff988c290c6b3.zip llvm-21aec4fdc5de213a1009bc684adff988c290c6b3.tar.gz llvm-21aec4fdc5de213a1009bc684adff988c290c6b3.tar.bz2 | |
[X86][SLM] Fix vector PSHUFB + variable shift resource/throughputs
Match whats documented in the Intel AOM (+Agner) - PSHUFB xmm is really slow, and mmx/xmm vector shifts are half rate.
Noticed while working to get the cost tables to more closely match llvm-mca analysis, in this case for shifts and truncations.
Diffstat (limited to 'lldb/source/Interpreter/CommandObject.cpp')
0 files changed, 0 insertions, 0 deletions