Reapply r185393.

Original commit message:

Remove floating point computations from SpillPlacement.cpp.

Patch by Benjamin Kramer!

Use the BlockFrequency class instead of floats in the Hopfield network
computations. This rescales the node Bias field from a [-2;2] float
range to two block frequencies BiasN and BiasP pulling in opposite
directions. This construct has a more predictable behavior when block
frequencies saturate.

The per-node scaling factors are no longer necessary, assuming the block
frequencies around a bundle are consistent.

This patch can cause the register allocator to make different spilling
decisions. The differences should be small.

llvm-svn: 186434

llvm/lib/CodeGen/SpillPlacement.cpp

@@ -59,6 +59,10 @@ void SpillPlacement::getAnalysisUsage(AnalysisUsage &AU) const {
   MachineFunctionPass::getAnalysisUsage(AU);
 }
 
+/// Decision threshold. A node gets the output value 0 if the weighted sum of
+/// its inputs falls in the open interval (-Threshold;Threshold).
+static const BlockFrequency Threshold = 2;
+
 /// Node - Each edge bundle corresponds to a Hopfield node.
 ///
 /// The node contains precomputed frequency data that only depends on the CFG,
@@ -69,31 +73,25 @@ void SpillPlacement::getAnalysisUsage(AnalysisUsage &AU) const {
 /// because all weights are positive.
 ///
 struct SpillPlacement::Node {
-  /// Scale - Inverse block frequency feeding into[0] or out of[1] the bundle.
+  /// BiasN - Sum of blocks that prefer a spill.
-  /// Ideally, these two numbers should be identical, but inaccuracies in the
+  BlockFrequency BiasN;
-  /// block frequency estimates means that we need to normalize ingoing and
+  /// BiasP - Sum of blocks that prefer a register.
-  /// outgoing frequencies separately so they are commensurate.
+  BlockFrequency BiasP;
-  float Scale[2];
-
-  /// Bias - Normalized contributions from non-transparent blocks.
-  /// A bundle connected to a MustSpill block has a huge negative bias,
-  /// otherwise it is a number in the range [-2;2].
-  float Bias;
 
   /// Value - Output value of this node computed from the Bias and links.
-  /// This is always in the range [-1;1]. A positive number means the variable
+  /// This is always on of the values {-1, 0, 1}. A positive number means the
-  /// should go in a register through this bundle.
+  /// variable should go in a register through this bundle.
-  float Value;
+  int Value;
 
-  typedef SmallVector<std::pair<float, unsigned>, 4> LinkVector;
+  typedef SmallVector<std::pair<BlockFrequency, unsigned>, 4> LinkVector;
 
   /// Links - (Weight, BundleNo) for all transparent blocks connecting to other
-  /// bundles. The weights are all positive and add up to at most 2, weights
+  /// bundles. The weights are all positive block frequencies.
-  /// from ingoing and outgoing nodes separately add up to a most 1. The weight
-  /// sum can be less than 2 when the variable is not live into / out of some
-  /// connected basic blocks.
   LinkVector Links;
 
+  /// SumLinkWeights - Cached sum of the weights of all links + ThresHold.
+  BlockFrequency SumLinkWeights;
+
   /// preferReg - Return true when this node prefers to be in a register.
   bool preferReg() const {
     // Undecided nodes (Value==0) go on the stack.
@@ -102,28 +100,24 @@ struct SpillPlacement::Node {
 
   /// mustSpill - Return True if this node is so biased that it must spill.
   bool mustSpill() const {
-    // Actually, we must spill if Bias < sum(weights).
+    // We must spill if Bias < -sum(weights) or the MustSpill flag was set.
-    // It may be worth it to compute the weight sum here?
+    // BiasN is saturated when MustSpill is set, make sure this still returns
-    return Bias < -2.0f;
+    // true when the RHS saturates. Note that SumLinkWeights includes Threshold.
-  }
+    return BiasN >= BiasP + SumLinkWeights;
-
-  /// Node - Create a blank Node.
-  Node() {
-    Scale[0] = Scale[1] = 0;
   }
 
   /// clear - Reset per-query data, but preserve frequencies that only depend on
   // the CFG.
   void clear() {
-    Bias = Value = 0;
+    BiasN = BiasP = Value = 0;
+    SumLinkWeights = Threshold;
     Links.clear();
   }
 
   /// addLink - Add a link to bundle b with weight w.
-  /// out=0 for an ingoing link, and 1 for an outgoing link.
+  void addLink(unsigned b, BlockFrequency w) {
-  void addLink(unsigned b, float w, bool out) {
+    // Update cached sum.
-    // Normalize w relative to all connected blocks from that direction.
+    SumLinkWeights += w;
-    w *= Scale[out];
 
     // There can be multiple links to the same bundle, add them up.
     for (LinkVector::iterator I = Links.begin(), E = Links.end(); I != E; ++I)
@@ -135,33 +129,48 @@ struct SpillPlacement::Node {
     Links.push_back(std::make_pair(w, b));
   }
 
-  /// addBias - Bias this node from an ingoing[0] or outgoing[1] link.
+  /// addBias - Bias this node.
-  /// Return the change to the total number of positive biases.
+  void addBias(BlockFrequency freq, BorderConstraint direction) {
-  void addBias(float w, bool out) {
+    switch (direction) {
-    // Normalize w relative to all connected blocks from that direction.
+    default:
-    w *= Scale[out];
+      break;
-    Bias += w;
+    case PrefReg:
+      BiasP += freq;
+      break;
+    case PrefSpill:
+      BiasN += freq;
+      break;
+    case MustSpill:
+      BiasN = BlockFrequency::getMaxFrequency();
+      break;
+    }
   }
 
   /// update - Recompute Value from Bias and Links. Return true when node
   /// preference changes.
   bool update(const Node nodes[]) {
     // Compute the weighted sum of inputs.
-    float Sum = Bias;
+    BlockFrequency SumN = BiasN;
-    for (LinkVector::iterator I = Links.begin(), E = Links.end(); I != E; ++I)
+    BlockFrequency SumP = BiasP;
-      Sum += I->first * nodes[I->second].Value;
+    for (LinkVector::iterator I = Links.begin(), E = Links.end(); I != E; ++I) {
+      if (nodes[I->second].Value == -1)
+        SumN += I->first;
+      else if (nodes[I->second].Value == 1)
+        SumP += I->first;
+    }
 
-    // The weighted sum is going to be in the range [-2;2]. Ideally, we should
+    // Each weighted sum is going to be less than the total frequency of the
-    // simply set Value = sign(Sum), but we will add a dead zone around 0 for
+    // bundle. Ideally, we should simply set Value = sign(SumP - SumN), but we
-    // two reasons:
+    // will add a dead zone around 0 for two reasons:
+    //
     //  1. It avoids arbitrary bias when all links are 0 as is possible during
     //     initial iterations.
     //  2. It helps tame rounding errors when the links nominally sum to 0.
-    const float Thres = 1e-4f;
+    //
     bool Before = preferReg();
-    if (Sum < -Thres)
+    if (SumN >= SumP + Threshold)
       Value = -1;
-    else if (Sum > Thres)
+    else if (SumP >= SumN + Threshold)
       Value = 1;
     else
       Value = 0;
@@ -178,23 +187,13 @@ bool SpillPlacement::runOnMachineFunction(MachineFunction &mf) {
   nodes = new Node[bundles->getNumBundles()];
 
   // Compute total ingoing and outgoing block frequencies for all bundles.
-  BlockFrequency.resize(mf.getNumBlockIDs());
+  BlockFrequencies.resize(mf.getNumBlockIDs());
   MachineBlockFrequencyInfo &MBFI = getAnalysis<MachineBlockFrequencyInfo>();
-  float EntryFreq = BlockFrequency::getEntryFrequency();
   for (MachineFunction::iterator I = mf.begin(), E = mf.end(); I != E; ++I) {
-    float Freq = MBFI.getBlockFreq(I).getFrequency() / EntryFreq;
     unsigned Num = I->getNumber();
-    BlockFrequency[Num] = Freq;
+    BlockFrequencies[Num] = MBFI.getBlockFreq(I);
-    nodes[bundles->getBundle(Num, 1)].Scale[0] += Freq;
-    nodes[bundles->getBundle(Num, 0)].Scale[1] += Freq;
   }
 
-  // Scales are reciprocal frequencies.
-  for (unsigned i = 0, e = bundles->getNumBundles(); i != e; ++i)
-    for (unsigned d = 0; d != 2; ++d)
-      if (nodes[i].Scale[d] > 0)
-        nodes[i].Scale[d] = 1 / nodes[i].Scale[d];
-
   // We never change the function.
   return false;
 }
@@ -215,12 +214,15 @@ void SpillPlacement::activate(unsigned n) {
   // landing pads, or loops with many 'continue' statements. It is difficult to
   // allocate registers when so many different blocks are involved.
   //
-  // Give a small negative bias to large bundles such that 1/32 of the
+  // Give a small negative bias to large bundles such that a substantial
-  // connected blocks need to be interested before we consider expanding the
+  // fraction of the connected blocks need to be interested before we consider
-  // region through the bundle. This helps compile time by limiting the number
+  // expanding the region through the bundle. This helps compile time by
-  // of blocks visited and the number of links in the Hopfield network.
+  // limiting the number of blocks visited and the number of links in the
-  if (bundles->getBlocks(n).size() > 100)
+  // Hopfield network.
-    nodes[n].Bias = -0.0625f;
+  if (bundles->getBlocks(n).size() > 100) {
+    nodes[n].BiasP = 0;
+    nodes[n].BiasN = (BlockFrequency::getEntryFrequency() / 16);
+  }
 }
 
 
@@ -229,27 +231,20 @@ void SpillPlacement::activate(unsigned n) {
 void SpillPlacement::addConstraints(ArrayRef<BlockConstraint> LiveBlocks) {
   for (ArrayRef<BlockConstraint>::iterator I = LiveBlocks.begin(),
        E = LiveBlocks.end(); I != E; ++I) {
-    float Freq = getBlockFrequency(I->Number);
+    BlockFrequency Freq = BlockFrequencies[I->Number];
-    const float Bias[] = {
-      0,           // DontCare,
-      1,           // PrefReg,
-      -1,          // PrefSpill
-      0,           // PrefBoth
-      -HUGE_VALF   // MustSpill
-    };
 
     // Live-in to block?
     if (I->Entry != DontCare) {
       unsigned ib = bundles->getBundle(I->Number, 0);
       activate(ib);
-      nodes[ib].addBias(Freq * Bias[I->Entry], 1);
+      nodes[ib].addBias(Freq, I->Entry);
     }
 
     // Live-out from block?
     if (I->Exit != DontCare) {
       unsigned ob = bundles->getBundle(I->Number, 1);
       activate(ob);
-      nodes[ob].addBias(Freq * Bias[I->Exit], 0);
+      nodes[ob].addBias(Freq, I->Exit);
     }
   }
 }
@@ -258,15 +253,15 @@ void SpillPlacement::addConstraints(ArrayRef<BlockConstraint> LiveBlocks) {
 void SpillPlacement::addPrefSpill(ArrayRef<unsigned> Blocks, bool Strong) {
   for (ArrayRef<unsigned>::iterator I = Blocks.begin(), E = Blocks.end();
        I != E; ++I) {
-    float Freq = getBlockFrequency(*I);
+    BlockFrequency Freq = BlockFrequencies[*I];
     if (Strong)
       Freq += Freq;
     unsigned ib = bundles->getBundle(*I, 0);
     unsigned ob = bundles->getBundle(*I, 1);
     activate(ib);
     activate(ob);
-    nodes[ib].addBias(-Freq, 1);
+    nodes[ib].addBias(Freq, PrefSpill);
-    nodes[ob].addBias(-Freq, 0);
+    nodes[ob].addBias(Freq, PrefSpill);
   }
 }
 
@@ -286,9 +281,9 @@ void SpillPlacement::addLinks(ArrayRef<unsigned> Links) {
       Linked.push_back(ib);
     if (nodes[ob].Links.empty() && !nodes[ob].mustSpill())
       Linked.push_back(ob);
-    float Freq = getBlockFrequency(Number);
+    BlockFrequency Freq = BlockFrequencies[Number];
-    nodes[ib].addLink(ob, Freq, 1);
+    nodes[ib].addLink(ob, Freq);
-    nodes[ob].addLink(ib, Freq, 0);
+    nodes[ob].addLink(ib, Freq);
   }
 }
 

llvm/lib/CodeGen/SpillPlacement.h

@@ -30,6 +30,7 @@
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/CodeGen/MachineFunctionPass.h"
+#include "llvm/Support/BlockFrequency.h"
 
 namespace llvm {
 
@@ -57,7 +58,7 @@ class SpillPlacement  : public MachineFunctionPass {
   SmallVector<unsigned, 8> RecentPositive;
 
   // Block frequencies are computed once. Indexed by block number.
-  SmallVector<float, 4> BlockFrequency;
+  SmallVector<BlockFrequency, 4> BlockFrequencies;
 
 public:
   static char ID; // Pass identification, replacement for typeid.
@@ -140,7 +141,9 @@ public:
   /// getBlockFrequency - Return the estimated block execution frequency per
   /// function invocation.
   float getBlockFrequency(unsigned Number) const {
-    return BlockFrequency[Number];
+    // FIXME: Return the BlockFrequency directly.
+    const float Scale = 1.0f / BlockFrequency::getEntryFrequency();
+    return BlockFrequencies[Number].getFrequency() * Scale;
   }
 
 private: