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

Lu Weining via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Oct 12 17:40:24 PDT 2022


SixWeining added a comment.

Some inline comments. Thanks.



================
Comment at: llvm/test/CodeGen/LoongArch/ir-instruction/atomicrmw-minmax.ll:76
+; LA64:       # %bb.0:
+; LA64-NEXT:    ammax_db.wu $a0, $a1, $a0
+; LA64-NEXT:    ret
----------------
rd == rj ? Maybe we miss `earlyclobber`?


================
Comment at: llvm/test/CodeGen/LoongArch/ir-instruction/atomicrmw-minmax.ll:85
+; LA64:       # %bb.0:
+; LA64-NEXT:    ammax_db.du $a0, $a1, $a0
+; LA64-NEXT:    ret
----------------
Ditto.


================
Comment at: llvm/test/CodeGen/LoongArch/ir-instruction/atomicrmw-minmax.ll:161
+; LA64:       # %bb.0:
+; LA64-NEXT:    ammin_db.wu $a0, $a1, $a0
+; LA64-NEXT:    ret
----------------
Ditto.


================
Comment at: llvm/test/CodeGen/LoongArch/ir-instruction/atomicrmw-minmax.ll:170
+; LA64:       # %bb.0:
+; LA64-NEXT:    ammin_db.du $a0, $a1, $a0
+; LA64-NEXT:    ret
----------------
Ditto.


================
Comment at: llvm/test/CodeGen/LoongArch/ir-instruction/atomicrmw.ll:23
 ; LA32-NEXT:    sc.w $a5, $a2, 0
-; LA32-NEXT:    beq $a5, $zero, .LBB0_1
+; LA32-NEXT:    beqz $a5, .LBB0_1
 ; LA32-NEXT:  # %bb.2:
----------------
Had better mention this change in summary.


Repository:
  rG LLVM Github Monorepo

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

https://reviews.llvm.org/D135525



More information about the llvm-commits mailing list