AMDGPU: Temporarily stop adding AtomicExpand to new PM passes

This breaks using -passes=atomic-expand (but only sometimes?).
Somehow an AtomicExpand pass ends up running without a TargetMachine,
despite always being constructed with one.

Author: arsenm

llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp

@@ -1806,7 +1806,8 @@ void AMDGPUCodeGenPassBuilder::addIRPasses(AddIRPass &addPass) const {
       (AMDGPUAtomicOptimizerStrategy != ScanOptions::None))
     addPass(AMDGPUAtomicOptimizerPass(TM, AMDGPUAtomicOptimizerStrategy));
 
-  addPass(AtomicExpandPass());
+  // FIXME: Adding atomic-expand manages to break -passes=atomic-expand
+  // addPass(AtomicExpandPass(TM));
 
   if (TM.getOptLevel() > CodeGenOptLevel::None) {
     addPass(AMDGPUPromoteAllocaPass(TM));