[CodeGen][MachinePipeliner] Limit register pressure when scheduling (#…

…74807)

In software pipelining, when searching for the Initiation Interval (II),
`MachinePipeliner` tries to reduce register pressure, but doesn't check
how many variables can actually be alive at the same time. As a result,
a lot of register spills/fills can be generated after register
allocation, which might cause performance degradation. To prevent such
cases, this patch adds a check phase that calculates the maximum
register pressure of the scheduled loop and reject it if the pressure is
too high. This can be enabled this by specifying
`pipeliner-register-pressure`. Additionally, an II search range is
currently fixed at 10, which is too small to find a schedule when the
above algorithm is applied. Therefore this patch also adds a new option
`pipeliner-ii-search-range` to specify the length of the range to
search. There is one more new option
`pipeliner-register-pressure-margin`, which can be used to estimate a
register pressure limit less than actual for conservative analysis.

Discourse thread:
https://discourse.llvm.org/t/considering-register-pressure-when-deciding-initiation-interval-in-machinepipeliner/74725

llvm/include/llvm/CodeGen/MachinePipeliner.h

@@ -273,8 +273,8 @@ class SwingSchedulerDAG : public ScheduleDAGInstrs {
 
   /// Return the new base register that was stored away for the changed
   /// instruction.
-  unsigned getInstrBaseReg(SUnit *SU) {
-    DenseMap<SUnit *, std::pair<unsigned, int64_t>>::iterator It =
+  unsigned getInstrBaseReg(SUnit *SU) const {
+    DenseMap<SUnit *, std::pair<unsigned, int64_t>>::const_iterator It =
         InstrChanges.find(SU);
     if (It != InstrChanges.end())
       return It->second.first;
@@ -639,16 +639,20 @@ class SMSchedule {
   computeUnpipelineableNodes(SwingSchedulerDAG *SSD,
                              TargetInstrInfo::PipelinerLoopInfo *PLI);
 
+  std::deque<SUnit *>
+  reorderInstructions(const SwingSchedulerDAG *SSD,
+                      const std::deque<SUnit *> &Instrs) const;
+
   bool
   normalizeNonPipelinedInstructions(SwingSchedulerDAG *SSD,
                                     TargetInstrInfo::PipelinerLoopInfo *PLI);
   bool isValidSchedule(SwingSchedulerDAG *SSD);
   void finalizeSchedule(SwingSchedulerDAG *SSD);
-  void orderDependence(SwingSchedulerDAG *SSD, SUnit *SU,
-                       std::deque<SUnit *> &Insts);
-  bool isLoopCarried(SwingSchedulerDAG *SSD, MachineInstr &Phi);
-  bool isLoopCarriedDefOfUse(SwingSchedulerDAG *SSD, MachineInstr *Def,
-                             MachineOperand &MO);
+  void orderDependence(const SwingSchedulerDAG *SSD, SUnit *SU,
+                       std::deque<SUnit *> &Insts) const;
+  bool isLoopCarried(const SwingSchedulerDAG *SSD, MachineInstr &Phi) const;
+  bool isLoopCarriedDefOfUse(const SwingSchedulerDAG *SSD, MachineInstr *Def,
+                             MachineOperand &MO) const;
   void print(raw_ostream &os) const;
   void dump() const;
 };