[PATCH] D135525: [LoongArch] Add codegen support for atomicrmw umin/umax operation on LA64

[Gong LingQin via Phabricator via llvm-commits]

gonglingqin added a comment.

In D135525#3845403 <https://reviews.llvm.org/D135525#3845403>, @xry111 wrote:

> Do we need dbar instructions around the ll/sc loop?

Yes. Thanks for the checking, I will fix this.



================
Comment at: llvm/lib/Target/LoongArch/LoongArchExpandAtomicPseudoInsts.cpp:417
+      .addImm(0);
+  BuildMI(LoopTailMBB, DL, TII->get(LoongArch::BEQ))
+      .addReg(Scratch1Reg)
----------------
xry111 wrote:
> I think it's better to use BEQZ here.
Good suggestion! I will modify it. Thanks.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D135525/new/

https://reviews.llvm.org/D135525