mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2025-02-01 05:01:59 +01:00
e1bca7c7bb
The generated sequence with whilelo is unintuitive, but it's the best I could come up with given the limited number of SVE instructions that interact with scalar registers. The other sequence I was considering was something like dup+cmpne, but an extra scalar instruction seems better than an extra vector instruction. Differential Revision: https://reviews.llvm.org/D71160