[PATCH] D130594: [RISCV] Inline ceil/floor/trunc for float and double

[Alex Bradbury via Phabricator via llvm-commits]

asb added a comment.

I'd wondered about FROUND too (I'd looked a little at this chatting to an intern implementing the equivalent operation in JSC). Given we don't have an instruction for fround like AArch64 does, it's not an obvious win unless the gains from avoiding the libcall (and perhaps strict fp correctness) are worth it, or if inlining exposes more opportunities for optimisation.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130594/new/

https://reviews.llvm.org/D130594