AMDGPU: Avoid constructing new std::vector in initCandidate

[arsenm]

Summary

Approximately 30% of the time was spent in the std::vector
constructor. In one testcase this pushes the scheduler to being the
second slowest pass.

I'm not sure I understand why these vector are necessary. The default
scheduler initCandidate seems to use some pre-existing vectors for the
pressure.

[rampitec]

LGTM

[arsenm]

r371136

lib/Target/AMDGPU/GCNSchedStrategy.h

void pickNodeFromQueue(SchedBoundary &Zone, const CandPolicy &ZonePolicy,
                          const RegPressureTracker &RPTracker,
                          SchedCandidate &Cand);
 
   void initCandidate(SchedCandidate &Cand, SUnit *SU,
                      bool AtTop, const RegPressureTracker &RPTracker,
                      const SIRegisterInfo *SRI,
                      unsigned SGPRPressure, unsigned VGPRPressure);
 
+  std::vector<unsigned> Pressure;
+  std::vector<unsigned> MaxPressure;
+
   unsigned SGPRExcessLimit;
   unsigned VGPRExcessLimit;
   unsigned SGPRCriticalLimit;
   unsigned VGPRCriticalLimit;
 
   unsigned TargetOccupancy;
 
   MachineFunction *MF;

lib/Target/AMDGPU/GCNSchedStrategy.cpp

void GCNMaxOccupancySchedStrategy::initCandidate(SchedCandidate &Cand, SUnit *SU,
 
   Cand.SU = SU;
   Cand.AtTop = AtTop;
 
   // getDownwardPressure() and getUpwardPressure() make temporary changes to
   // the tracker, so we need to pass those function a non-const copy.
   RegPressureTracker &TempTracker = const_cast<RegPressureTracker&>(RPTracker);
 
-  std::vector<unsigned> Pressure;
-  std::vector<unsigned> MaxPressure;
+  Pressure.clear();
+  MaxPressure.clear();
 
   if (AtTop)
     TempTracker.getDownwardPressure(SU->getInstr(), Pressure, MaxPressure);
   else {
     // FIXME: I think for bottom up scheduling, the register pressure is cached
     // and can be retrieved by DAG->getPressureDif(SU).
     TempTracker.getUpwardPressure(SU->getInstr(), Pressure, MaxPressure);
   }