[PATCH] D130100: [AArch64] Combine a load into GPR followed by a copy to FPR to a load into FPR directly through MIPeepholeOpt

[Mingming Liu via Phabricator via llvm-commits]

mingmingl added a comment.

An update:

1. For the motivating use case (around pmull intrinsic and instruction), I sent out D130548 <https://reviews.llvm.org/D130548>.
2. The general issue (missed combination of {load + copy -> load} exists.

For 2), `peephole-opt` should have folded the load (code <https://github.com/llvm/llvm-project/blob/809855b56f06dd7182685f88fbbc64111df9339a/llvm/lib/CodeGen/PeepholeOptimizer.cpp#L1785>) given AArch64InstrInfo.cpp considers this pattern (mentioned in comment <https://github.com/llvm/llvm-project/blob/809855b56f06dd7182685f88fbbc64111df9339a/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp#L4455-L4467>). Going to see why it's not executed and update back.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130100/new/

https://reviews.llvm.org/D130100