[PATCH] D124084: [AMDGPU] Remove atomic pattern args in FLAT_[Global_]Atomic_Pseudo defs

[Jay Foad via Phabricator via llvm-commits]

foad added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/FLATInstructions.td:611
 defm FLAT_ATOMIC_CMPSWAP    : FLAT_Atomic_Pseudo <"flat_atomic_cmpswap",
-                                VGPR_32, i32, AMDGPUatomic_cmp_swap_flat_32,
                                 v2i32, VReg_64>;
----------------
Remove this argument from the definition of the multiclass, if nothing is using it?


================
Comment at: llvm/lib/Target/AMDGPU/FLATInstructions.td:761
 defm GLOBAL_ATOMIC_CMPSWAP : FLAT_Global_Atomic_Pseudo <"global_atomic_cmpswap",
-                               VGPR_32, i32, AMDGPUatomic_cmp_swap_global_32,
                                v2i32, VReg_64>;
----------------
Likewise.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D124084/new/

https://reviews.llvm.org/D124084