search for: getshufflemask

...s an enum, and then wrap up the entire description of a fixed or scalable shuffle into a class "ShuffleMask". This would allow checking whether an operation matches one of the above patterns, and can be converted to the existing ArrayRef<int> for fixed shuffles. ShuffleVectorInst::getShuffleMask would then return a ShuffleMask, I think. Then we could add an alternate API getFixedShuffleMask() that only works for fixed shuffles, and just returns the fixed mask as an ArrayRef<int>. I'm working on refactoring the existing shufflevector handling in LLVM to allow making these changes...