[AMDGPU] Ignore wavefront barrier latency during scheduling DAG mutation (#168500)

Do not add latency for wavefront and singlethread scope fences during
barrier latency DAG mutation.
These scopes do not typically introduce any latency and adjusting
schedules based on them significantly impacts latency hiding.

1 files changed, 3 insertions, 3 deletions

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
index 5ff16e29bbbb..0346580ffa68 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -647,7 +647,7 @@ createGCNMaxOccupancyMachineScheduler(MachineSchedContext *C) {
   DAG->addMutation(createIGroupLPDAGMutation(AMDGPU::SchedulingPhase::Initial));
   DAG->addMutation(createAMDGPUMacroFusionDAGMutation());
   DAG->addMutation(createAMDGPUExportClusteringDAGMutation());
-  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation());
+  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation(C->MF));
   return DAG;
 }
 
@@ -668,7 +668,7 @@ createGCNMaxMemoryClauseMachineScheduler(MachineSchedContext *C) {
   if (ST.shouldClusterStores())
     DAG->addMutation(createStoreClusterDAGMutation(DAG->TII, DAG->TRI));
   DAG->addMutation(createAMDGPUExportClusteringDAGMutation());
-  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation());
+  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation(C->MF));
   return DAG;
 }
 
@@ -1209,7 +1209,7 @@ GCNTargetMachine::createPostMachineScheduler(MachineSchedContext *C) const {
       EnableVOPD)
     DAG->addMutation(createVOPDPairingMutation());
   DAG->addMutation(createAMDGPUExportClusteringDAGMutation());
-  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation());
+  DAG->addMutation(createAMDGPUBarrierLatencyDAGMutation(C->MF));
   return DAG;
 }
 //===----------------------------------------------------------------------===//