[AMDGPU] Disable FillMFMAShadowMutation by default

[kerbowa]

Summary

Disable amdgpu mfma power sched.

llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp

 #define DEBUG_TYPE "amdgpu-subtarget"
 
 #define GET_SUBTARGETINFO_TARGET_DESC
 #define GET_SUBTARGETINFO_CTOR
 #define AMDGPUSubtarget GCNSubtarget
 #include "AMDGPUGenSubtargetInfo.inc"
 #undef AMDGPUSubtarget
 
-static cl::opt<bool> DisablePowerSched(
-  "amdgpu-disable-power-sched",
-  cl::desc("Disable scheduling to minimize mAI power bursts"),
+static cl::opt<bool> EnablePowerSched(
+  "amdgpu-enable-power-sched",
+  cl::desc("Enable scheduling to minimize mAI power bursts"),
   cl::init(false));
 
 static cl::opt<bool> EnableVGPRIndexMode(
   "amdgpu-vgpr-index-mode",
   cl::desc("Use GPR indexing mode instead of movrel for vector indexing"),
   cl::init(false));
 
 static cl::opt<bool> UseAA("amdgpu-use-aa-in-codegen",
while (!Worklist.empty() && MaxChain-- > 0) {
       }
     }
 
     return Linked;
   }
 
   void apply(ScheduleDAGInstrs *DAGInstrs) override {
     const GCNSubtarget &ST = DAGInstrs->MF.getSubtarget<GCNSubtarget>();
-    if (!ST.hasMAIInsts() || DisablePowerSched)
+    if (!ST.hasMAIInsts())
       return;
     DAG = static_cast<ScheduleDAGMI*>(DAGInstrs);
     const TargetSchedModel *TSchedModel = DAGInstrs->getSchedModel();
     if (!TSchedModel || DAG->SUnits.empty())
       return;
 
     // Scan for MFMA long latency instructions and try to add a dependency
     // of available SALU instructions to give them a chance to fill MFMA

 
 void GCNSubtarget::getPostRAMutations(
     std::vector<std::unique_ptr<ScheduleDAGMutation>> &Mutations) const {
   Mutations.push_back(std::make_unique<FillMFMAShadowMutation>(&InstrInfo));
 }
 
 std::unique_ptr<ScheduleDAGMutation>
 GCNSubtarget::createFillMFMAShadowMutation(const TargetInstrInfo *TII) const {
-  return std::make_unique<FillMFMAShadowMutation>(&InstrInfo);
+  return EnablePowerSched ? std::make_unique<FillMFMAShadowMutation>(&InstrInfo)
+                          : nullptr;
 }
 
 const AMDGPUSubtarget &AMDGPUSubtarget::get(const MachineFunction &MF) {
   if (MF.getTarget().getTargetTriple().getArch() == Triple::amdgcn)
     return static_cast<const AMDGPUSubtarget&>(MF.getSubtarget<GCNSubtarget>());
   else
     return static_cast<const AMDGPUSubtarget&>(MF.getSubtarget<R600Subtarget>());
 }
 
 const AMDGPUSubtarget &AMDGPUSubtarget::get(const TargetMachine &TM, const Function &F) {
   if (TM.getTargetTriple().getArch() == Triple::amdgcn)
     return static_cast<const AMDGPUSubtarget&>(TM.getSubtarget<GCNSubtarget>(F));
   else
     return static_cast<const AMDGPUSubtarget&>(TM.getSubtarget<R600Subtarget>(F));
 }

llvm/test/CodeGen/AMDGPU/sched-prefer-non-mfma.mir

-# RUN: llc -march=amdgcn -mcpu=gfx908 -run-pass post-RA-sched -amdgpu-disable-power-sched -verify-machineinstrs -o - %s | FileCheck -check-prefix=GCN %s
+# RUN: llc -march=amdgcn -mcpu=gfx908 -run-pass post-RA-sched -verify-machineinstrs -o - %s | FileCheck -check-prefix=GCN %s
 
 # GCN-LABEL: name: unrelated_mfma
 # GCN: V_MFMA_F32_32X32X1F32
 # GCN: S_CMP_LG_U32
 # GCN: V_MFMA_F32_32X32X1F32
 
 ---
 name:            unrelated_mfma