projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 504b3fe) | patch
[X86][AVX] lowerShuffleAsLanePermuteAndShuffle - consistently normalize multi-input...
authorSimon Pilgrim <llvm-dev@redking.me.uk>
Fri, 10 Jan 2020 14:55:00 +0000 (14:55 +0000)
committerSimon Pilgrim <llvm-dev@redking.me.uk>
Fri, 10 Jan 2020 17:21:20 +0000 (17:21 +0000)
commita5bdada09defc15d2b009314306f4fcb8fa8458d
tree8e6d25e888fad8fe77a612cd34fbe1414a24fa27tree | snapshot
parent504b3fe5bfed7ea24c7c74f563ef6a8214e24223commit | diff
[X86][AVX] lowerShuffleAsLanePermuteAndShuffle - consistently normalize multi-input shuffle elements

We only use lowerShuffleAsLanePermuteAndShuffle for unary shuffles at the moment, but we should consistently handle lane index calculations for multiple inputs in both the AVX1 and AVX2 paths.

Minor (almost NFC) tidyup as I'm hoping to use lowerShuffleAsLanePermuteAndShuffle for binary shuffles soon.
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
Domain: System / Toolchain;