AMDGPU: Run SIInsertWaits as pre-emit pass

Running this after the scheduler enables scheduling
waits later so other ALU instructions can run while
this would be waiting.

When combined with enabling the post-RA scheduler, this
gives about a ~20% improvement on sgemm.

llvm-svn: 241473

llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp

@@ -280,10 +280,10 @@ void GCNPassConfig::addPostRegAlloc() {
 }
 
 void GCNPassConfig::addPreSched2() {
-  addPass(createSIInsertWaits(*TM), false);
 }
 
 void GCNPassConfig::addPreEmitPass() {
+  addPass(createSIInsertWaits(*TM), false);
   addPass(createSILowerControlFlowPass(*TM), false);
 }