[PATCH] D84639: AMDGPU: Add type mangling to llvm.amdgcn.readfirstlane

[Jay Foad via Phabricator via llvm-commits]

foad added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp:609
           B.CreateTrunc(B.CreateLShr(PHI, 32), B.getInt32Ty());
-      CallInst *const ReadFirstLaneLo =
-          B.CreateIntrinsic(Intrinsic::amdgcn_readfirstlane, {}, ExtractLo);
-      CallInst *const ReadFirstLaneHi =
-          B.CreateIntrinsic(Intrinsic::amdgcn_readfirstlane, {}, ExtractHi);
+      Function *Readfirstlane = getReadfirstlane(M);
+
----------------
Hoist this above the "if" on line 605?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84639/new/

https://reviews.llvm.org/D84639