Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge isSHUFPMask and isCommutedSHUFPMask into single function that can do ↵ | Craig Topper | 2011-12-06 | 1 | -1/+8 |
| | | | | | | both. Do the same for the 256-bit version. Use loops to reduce size of isVSHUFPYMask. Fix test cases that were incorrectly passing due to isCommutedSHUFPMask not checking for the vector being 128-bit. This caused some 256-bit shuffles to be incorrectly commuted. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145921 91177308-0d34-0410-b5e6-96231b3b80d8 | ||||
* | The VPERM2F128 is a AVX instruction which permutes between two 256-bit | Bruno Cardoso Lopes | 2011-08-12 | 1 | -0/+62 |
vectors. It operates on 128-bit elements instead of regular scalar types. Recognize shuffles that are suitable for VPERM2F128 and teach the x86 legalizer how to handle them. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137519 91177308-0d34-0410-b5e6-96231b3b80d8 |