[MachineCombiner] Support for floating-point FMA on ARM64

Evaluates fmul+fadd -> fmadd combines and similar code sequences in the
machine combiner. It adds support for float and double similar to the existing
integer implementation. The key features are:

- DAGCombiner checks whether it should combine greedily or let the machine
combiner do the evaluation. This is only supported on ARM64.
- It gives preference to throughput over latency: the heuristic used is
to combine always in loops. The targets decides whether the machine
combiner should optimize for throughput or latency.
- Supports for fmadd, f(n)msub, fmla, fmls patterns
- On by default at O3 ffast-math

llvm-svn: 267098

1 files changed, 11 insertions, 1 deletions

diff --git a/llvm/lib/CodeGen/MachineCombiner.cpp b/llvm/lib/CodeGen/MachineCombiner.cpp
index 44601d5..6b5c6ba 100644
--- a/llvm/lib/CodeGen/MachineCombiner.cpp
+++ b/llvm/lib/CodeGen/MachineCombiner.cpp
@@ -40,6 +40,7 @@ class MachineCombiner : public MachineFunctionPass {
   const TargetRegisterInfo *TRI;
   MCSchedModel SchedModel;
   MachineRegisterInfo *MRI;
+  MachineLoopInfo *MLI; // Current MachineLoopInfo
   MachineTraceMetrics *Traces;
   MachineTraceMetrics::Ensemble *MinInstr;
 
@@ -86,6 +87,7 @@ char &llvm::MachineCombinerID = MachineCombiner::ID;
 
 INITIALIZE_PASS_BEGIN(MachineCombiner, "machine-combiner",
                       "Machine InstCombiner", false, false)
+INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
 INITIALIZE_PASS_DEPENDENCY(MachineTraceMetrics)
 INITIALIZE_PASS_END(MachineCombiner, "machine-combiner", "Machine InstCombiner",
                     false, false)
@@ -93,6 +95,7 @@ INITIALIZE_PASS_END(MachineCombiner, "machine-combiner", "Machine InstCombiner",
 void MachineCombiner::getAnalysisUsage(AnalysisUsage &AU) const {
   AU.setPreservesCFG();
   AU.addPreserved<MachineDominatorTree>();
+  AU.addRequired<MachineLoopInfo>();
   AU.addPreserved<MachineLoopInfo>();
   AU.addRequired<MachineTraceMetrics>();
   AU.addPreserved<MachineTraceMetrics>();
@@ -354,6 +357,8 @@ bool MachineCombiner::combineInstructions(MachineBasicBlock *MBB) {
   DEBUG(dbgs() << "Combining MBB " << MBB->getName() << "\n");
 
   auto BlockIter = MBB->begin();
+  // Check if the block is in a loop.
+  const MachineLoop *ML = MLI->getLoopFor(MBB);
 
   while (BlockIter != MBB->end()) {
     auto &MI = *BlockIter++;
@@ -406,11 +411,15 @@ bool MachineCombiner::combineInstructions(MachineBasicBlock *MBB) {
       if (!NewInstCount)
         continue;
 
+      bool SubstituteAlways = false;
+      if (ML && TII->isThroughputPattern(P))
+        SubstituteAlways = true;
+
       // Substitute when we optimize for codesize and the new sequence has
       // fewer instructions OR
       // the new sequence neither lengthens the critical path nor increases
       // resource pressure.
-      if (doSubstitute(NewInstCount, OldInstCount) ||
+      if (SubstituteAlways || doSubstitute(NewInstCount, OldInstCount) ||
           (improvesCriticalPathLen(MBB, &MI, BlockTrace, InsInstrs,
                                    InstrIdxForVirtReg, P) &&
            preservesResourceLen(MBB, BlockTrace, InsInstrs, DelInstrs))) {
@@ -447,6 +456,7 @@ bool MachineCombiner::runOnMachineFunction(MachineFunction &MF) {
   SchedModel = STI.getSchedModel();
   TSchedModel.init(SchedModel, &STI, TII);
   MRI = &MF.getRegInfo();
+  MLI = &getAnalysis<MachineLoopInfo>();
   Traces = &getAnalysis<MachineTraceMetrics>();
   MinInstr = nullptr;
   OptSize = MF.getFunction()->optForSize();