[llvm] AMDGPU: Avoid creating unnecessary block split in atomic expansion (PR #102440)

[Stanislav Mekhanoshin via llvm-commits]

================
@@ -16595,8 +16595,6 @@ void SITargetLowering::emitExpandAtomicRMW(AtomicRMWInst *AI) const {
   Function *F = BB->getParent();
   BasicBlock *ExitBB =
       BB->splitBasicBlock(Builder.GetInsertPoint(), "atomicrmw.end");
-  BasicBlock *CheckSharedBB =
----------------
rampitec wrote:

Update the comment above (line 16553)?

https://github.com/llvm/llvm-project/pull/102440