[MISched] Add statistics to quantify scheduling (llvm#138090)

When diagnosing scheduler issues it can be useful to know how scheduling
changes the order of instructions, particularly for large functions when
it's not trivial to figure out from the debug output by looking at the
scheduling unit (SU) IDs.

This adds pre-RA and post-RA statistics to track 1) the number of
instructions that remain in source order after scheduling and 2) the
total number of instructions scheduled, to compare 1) against.

Author: c-rhodes

llvm/include/llvm/CodeGen/MachineScheduler.h

@@ -1196,6 +1196,9 @@ class GenericSchedulerBase : public MachineSchedStrategy {
   const MachineSchedContext *Context;
   const TargetSchedModel *SchedModel = nullptr;
   const TargetRegisterInfo *TRI = nullptr;
+  unsigned TopIdx = 0;
+  unsigned BotIdx = 0;
+  unsigned NumRegionInstrs = 0;
 
   MachineSchedPolicy RegionPolicy;
 

llvm/lib/CodeGen/MachineScheduler.cpp

@@ -74,6 +74,14 @@ using namespace llvm;
 
 #define DEBUG_TYPE "machine-scheduler"
 
+STATISTIC(NumInstrsInSourceOrderPreRA,
+          "Number of instructions in source order after pre-RA scheduling");
+STATISTIC(NumInstrsInSourceOrderPostRA,
+          "Number of instructions in source order after post-RA scheduling");
+STATISTIC(NumInstrsScheduledPreRA,
+          "Number of instructions scheduled by pre-RA scheduler");
+STATISTIC(NumInstrsScheduledPostRA,
+          "Number of instructions scheduled by post-RA scheduler");
 STATISTIC(NumClustered, "Number of load/store pairs clustered");
 
 namespace llvm {
@@ -3505,6 +3513,9 @@ void GenericScheduler::initPolicy(MachineBasicBlock::iterator Begin,
     RegionPolicy.OnlyBottomUp = false;
     RegionPolicy.OnlyTopDown = false;
   }
+
+  BotIdx = NumRegionInstrs - 1;
+  this->NumRegionInstrs = NumRegionInstrs;
 }
 
 void GenericScheduler::dumpPolicy() const {
@@ -3981,6 +3992,18 @@ SUnit *GenericScheduler::pickNode(bool &IsTopNode) {
 
   LLVM_DEBUG(dbgs() << "Scheduling SU(" << SU->NodeNum << ") "
                     << *SU->getInstr());
+
+  if (IsTopNode) {
+    if (SU->NodeNum == TopIdx++)
+      ++NumInstrsInSourceOrderPreRA;
+  } else {
+    assert(BotIdx < NumRegionInstrs && "out of bounds");
+    if (SU->NodeNum == BotIdx--)
+      ++NumInstrsInSourceOrderPreRA;
+  }
+
+  NumInstrsScheduledPreRA += 1;
+
   return SU;
 }
 
@@ -4104,6 +4127,9 @@ void PostGenericScheduler::initPolicy(MachineBasicBlock::iterator Begin,
     RegionPolicy.OnlyBottomUp = false;
     RegionPolicy.OnlyTopDown = false;
   }
+
+  BotIdx = NumRegionInstrs - 1;
+  this->NumRegionInstrs = NumRegionInstrs;
 }
 
 void PostGenericScheduler::registerRoots() {
@@ -4323,6 +4349,18 @@ SUnit *PostGenericScheduler::pickNode(bool &IsTopNode) {
 
   LLVM_DEBUG(dbgs() << "Scheduling SU(" << SU->NodeNum << ") "
                     << *SU->getInstr());
+
+  if (IsTopNode) {
+    if (SU->NodeNum == TopIdx++)
+      ++NumInstrsInSourceOrderPostRA;
+  } else {
+    assert(BotIdx < NumRegionInstrs && "out of bounds");
+    if (SU->NodeNum == BotIdx--)
+      ++NumInstrsInSourceOrderPostRA;
+  }
+
+  NumInstrsScheduledPostRA += 1;
+
   return SU;
 }