diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2007-12-11 01:46:18 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2007-12-11 01:46:18 +0000 |
| commit | 14b32e194152f7147d6f531346b923152e58900f (patch) | |
| tree | 1f539307e974ed81a1a9bddb9e00e990f24c1053 /lib/CodeGen/RegAllocLinearScan.cpp | |
| parent | 844e0f9def5d1caa28e1f2612cb4226e14fd22b3 (diff) | |
- Improved v8i16 shuffle lowering. It now uses pshuflw and pshufhw as much as
possible before resorting to pextrw and pinsrw.
- Better codegen for v4i32 shuffles masquerading as v8i16 or v16i8 shuffles.
- Improves (i16 extract_vector_element 0) codegen by recognizing
(i32 extract_vector_element 0) does not require a pextrw.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44836 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/CodeGen/RegAllocLinearScan.cpp')
0 files changed, 0 insertions, 0 deletions