[PGO][PGSO] DAG.shouldOptForSize part.

Summary:
(Split of off D67120)

SelectionDAG::shouldOptForSize changes for profile guided size optimization.

Reviewers: davidxl

Subscribers: hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D70095

9 files changed, 48 insertions, 34 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 94e98cd8065..fb7ddf5b233 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -217,7 +217,7 @@ namespace {
     DAGCombiner(SelectionDAG &D, AliasAnalysis *AA, CodeGenOpt::Level OL)
         : DAG(D), TLI(D.getTargetLoweringInfo()), Level(BeforeLegalizeTypes),
           OptLevel(OL), AA(AA) {
-      ForCodeSize = DAG.getMachineFunction().getFunction().hasOptSize();
+      ForCodeSize = DAG.shouldOptForSize();
 
       MaximumLegalStoreInBits = 0;
       // We use the minimum store size here, since that's all we can guarantee
@@ -12885,7 +12885,7 @@ SDValue DAGCombiner::visitFPOW(SDNode *N) {
 
     // Assume that libcalls are the smallest code.
     // TODO: This restriction should probably be lifted for vectors.
-    if (DAG.getMachineFunction().getFunction().hasOptSize())
+    if (ForCodeSize)
       return SDValue();
 
     // pow(X, 0.25) --> sqrt(sqrt(X))
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index 7cfd8ade954..f1b88d80f43 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -24,6 +24,8 @@
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/Triple.h"
 #include "llvm/ADT/Twine.h"
+#include "llvm/Analysis/BlockFrequencyInfo.h"
+#include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/Analysis/ValueTracking.h"
 #include "llvm/CodeGen/ISDOpcodes.h"
 #include "llvm/CodeGen/MachineBasicBlock.h"
@@ -63,6 +65,7 @@
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/Target/TargetMachine.h"
 #include "llvm/Target/TargetOptions.h"
+#include "llvm/Transforms/Utils/SizeOpts.h"
 #include <algorithm>
 #include <cassert>
 #include <cstdint>
@@ -1005,7 +1008,9 @@ SelectionDAG::SelectionDAG(const TargetMachine &tm, CodeGenOpt::Level OL)
 void SelectionDAG::init(MachineFunction &NewMF,
                         OptimizationRemarkEmitter &NewORE,
                         Pass *PassPtr, const TargetLibraryInfo *LibraryInfo,
-                        LegacyDivergenceAnalysis * Divergence) {
+                        LegacyDivergenceAnalysis * Divergence,
+                        ProfileSummaryInfo *PSIin,
+                        BlockFrequencyInfo *BFIin) {
   MF = &NewMF;
   SDAGISelPass = PassPtr;
   ORE = &NewORE;
@@ -1014,6 +1019,8 @@ void SelectionDAG::init(MachineFunction &NewMF,
   LibInfo = LibraryInfo;
   Context = &MF->getFunction().getContext();
   DA = Divergence;
+  PSI = PSIin;
+  BFI = BFIin;
 }
 
 SelectionDAG::~SelectionDAG() {
@@ -1023,6 +1030,11 @@ SelectionDAG::~SelectionDAG() {
   delete DbgInfo;
 }
 
+bool SelectionDAG::shouldOptForSize() const {
+  return MF->getFunction().hasOptSize() ||
+      llvm::shouldOptimizeForSize(FLI->MBB->getBasicBlock(), PSI, BFI);
+}
+
 void SelectionDAG::allnodes_clear() {
   assert(&*AllNodes.begin() == &EntryNode);
   AllNodes.remove(AllNodes.begin());
@@ -1427,7 +1439,7 @@ SDValue SelectionDAG::getConstantPool(const Constant *C, EVT VT,
   assert((TargetFlags == 0 || isTarget) &&
          "Cannot set target flags on target-independent globals");
   if (Alignment == 0)
-    Alignment = MF->getFunction().hasOptSize()
+    Alignment = shouldOptForSize()
                     ? getDataLayout().getABITypeAlignment(C->getType())
                     : getDataLayout().getPrefTypeAlignment(C->getType());
   unsigned Opc = isTarget ? ISD::TargetConstantPool : ISD::ConstantPool;
@@ -5733,12 +5745,13 @@ static bool isMemSrcFromConstant(SDValue Src, ConstantDataArraySlice &Slice) {
                                   SrcDelta + G->getOffset());
 }
 
-static bool shouldLowerMemFuncForSize(const MachineFunction &MF) {
+static bool shouldLowerMemFuncForSize(const MachineFunction &MF,
+                                      SelectionDAG &DAG) {
   // On Darwin, -Os means optimize for size without hurting performance, so
   // only really optimize for size when -Oz (MinSize) is used.
   if (MF.getTarget().getTargetTriple().isOSDarwin())
     return MF.getFunction().hasMinSize();
-  return MF.getFunction().hasOptSize();
+  return DAG.shouldOptForSize();
 }
 
 static void chainLoadsAndStoresForMemcpy(SelectionDAG &DAG, const SDLoc &dl,
@@ -5788,7 +5801,7 @@ static SDValue getMemcpyLoadsAndStores(SelectionDAG &DAG, const SDLoc &dl,
   bool DstAlignCanChange = false;
   MachineFunction &MF = DAG.getMachineFunction();
   MachineFrameInfo &MFI = MF.getFrameInfo();
-  bool OptSize = shouldLowerMemFuncForSize(MF);
+  bool OptSize = shouldLowerMemFuncForSize(MF, DAG);
   FrameIndexSDNode *FI = dyn_cast<FrameIndexSDNode>(Dst);
   if (FI && !MFI.isFixedObjectIndex(FI->getIndex()))
     DstAlignCanChange = true;
@@ -5971,7 +5984,7 @@ static SDValue getMemmoveLoadsAndStores(SelectionDAG &DAG, const SDLoc &dl,
   bool DstAlignCanChange = false;
   MachineFunction &MF = DAG.getMachineFunction();
   MachineFrameInfo &MFI = MF.getFrameInfo();
-  bool OptSize = shouldLowerMemFuncForSize(MF);
+  bool OptSize = shouldLowerMemFuncForSize(MF, DAG);
   FrameIndexSDNode *FI = dyn_cast<FrameIndexSDNode>(Dst);
   if (FI && !MFI.isFixedObjectIndex(FI->getIndex()))
     DstAlignCanChange = true;
@@ -6077,7 +6090,7 @@ static SDValue getMemsetStores(SelectionDAG &DAG, const SDLoc &dl,
   bool DstAlignCanChange = false;
   MachineFunction &MF = DAG.getMachineFunction();
   MachineFrameInfo &MFI = MF.getFrameInfo();
-  bool OptSize = shouldLowerMemFuncForSize(MF);
+  bool OptSize = shouldLowerMemFuncForSize(MF, DAG);
   FrameIndexSDNode *FI = dyn_cast<FrameIndexSDNode>(Dst);
   if (FI && !MFI.isFixedObjectIndex(FI->getIndex()))
     DstAlignCanChange = true;
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index 4fa907a4dcf..1ed0dc2c979 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -28,10 +28,12 @@
 #include "llvm/ADT/Twine.h"
 #include "llvm/Analysis/AliasAnalysis.h"
 #include "llvm/Analysis/BranchProbabilityInfo.h"
+#include "llvm/Analysis/BlockFrequencyInfo.h"
 #include "llvm/Analysis/ConstantFolding.h"
 #include "llvm/Analysis/EHPersonalities.h"
 #include "llvm/Analysis/Loads.h"
 #include "llvm/Analysis/MemoryLocation.h"
+#include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
 #include "llvm/Analysis/ValueTracking.h"
 #include "llvm/Analysis/VectorUtils.h"
@@ -5360,8 +5362,8 @@ static SDValue ExpandPowI(const SDLoc &DL, SDValue LHS, SDValue RHS,
     if (Val == 0)
       return DAG.getConstantFP(1.0, DL, LHS.getValueType());
 
-    const Function &F = DAG.getMachineFunction().getFunction();
-    if (!F.hasOptSize() ||
+    bool OptForSize = DAG.shouldOptForSize();
+    if (!OptForSize ||
         // If optimizing for size, don't insert too many multiplies.
         // This inserts up to 5 multiplies.
         countPopulation(Val) + Log2_32(Val) < 7) {
@@ -10441,7 +10443,7 @@ void SelectionDAGBuilder::visitSwitch(const SwitchInst &SI) {
     return;
   }
 
-  SL->findJumpTables(Clusters, &SI, DefaultMBB, nullptr, nullptr);
+  SL->findJumpTables(Clusters, &SI, DefaultMBB, DAG.getPSI(), DAG.getBFI());
   SL->findBitTestClusters(Clusters, &SI);
 
   LLVM_DEBUG({
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
index 2757e1cf015..c6a2bfd31fe 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
@@ -444,7 +444,8 @@ bool SelectionDAGISel::runOnMachineFunction(MachineFunction &mf) {
   SplitCriticalSideEffectEdges(const_cast<Function &>(Fn), DT, LI);
 
   CurDAG->init(*MF, *ORE, this, LibInfo,
-   getAnalysisIfAvailable<LegacyDivergenceAnalysis>());
+               getAnalysisIfAvailable<LegacyDivergenceAnalysis>(),
+               nullptr, nullptr);
   FuncInfo->set(Fn, *MF, CurDAG);
   SwiftError->setFunction(*MF);
 
diff --git a/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp b/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
index 4d4f76e8e91..ae6c1429e56 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
@@ -39,20 +39,16 @@ class AArch64DAGToDAGISel : public SelectionDAGISel {
   /// make the right decision when generating code for different targets.
   const AArch64Subtarget *Subtarget;
 
-  bool ForCodeSize;
-
 public:
   explicit AArch64DAGToDAGISel(AArch64TargetMachine &tm,
                                CodeGenOpt::Level OptLevel)
-      : SelectionDAGISel(tm, OptLevel), Subtarget(nullptr),
-        ForCodeSize(false) {}
+      : SelectionDAGISel(tm, OptLevel), Subtarget(nullptr) {}
 
   StringRef getPassName() const override {
     return "AArch64 Instruction Selection";
   }
 
   bool runOnMachineFunction(MachineFunction &MF) override {
-    ForCodeSize = MF.getFunction().hasOptSize();
     Subtarget = &MF.getSubtarget<AArch64Subtarget>();
     return SelectionDAGISel::runOnMachineFunction(MF);
   }
@@ -399,7 +395,7 @@ static bool isWorthFoldingSHL(SDValue V) {
 bool AArch64DAGToDAGISel::isWorthFolding(SDValue V) const {
   // Trivial if we are optimizing for code size or if there is only
   // one use of the value.
-  if (ForCodeSize || V.hasOneUse())
+  if (CurDAG->shouldOptForSize() || V.hasOneUse())
     return true;
   // If a subtarget has a fastpath LSL we can fold a logical shift into
   // the addressing mode and save a cycle.
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.td b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
index 7ae5b5dd6a8..80cf31ff3d5 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
@@ -518,10 +518,10 @@ def AArch64uunpklo : SDNode<"AArch64ISD::UUNPKLO", SDT_AArch64unpk>;
 // the Function object through the <Target>Subtarget and objections were raised
 // to that (see post-commit review comments for r301750).
 let RecomputePerFunction = 1 in {
-  def ForCodeSize   : Predicate<"MF->getFunction().hasOptSize()">;
-  def NotForCodeSize   : Predicate<"!MF->getFunction().hasOptSize()">;
+  def ForCodeSize   : Predicate<"shouldOptForSize(MF)">;
+  def NotForCodeSize   : Predicate<"!shouldOptForSize(MF)">;
   // Avoid generating STRQro if it is slow, unless we're optimizing for code size.
-  def UseSTRQro : Predicate<"!Subtarget->isSTRQroSlow() || MF->getFunction().hasOptSize()">;
+  def UseSTRQro : Predicate<"!Subtarget->isSTRQroSlow() || shouldOptForSize(MF)">;
 
   def UseBTI : Predicate<[{ MF->getFunction().hasFnAttribute("branch-target-enforcement") }]>;
   def NotUseBTI : Predicate<[{ !MF->getFunction().hasFnAttribute("branch-target-enforcement") }]>;
diff --git a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
index a91b23a3ce5..56508fc8840 100644
--- a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
@@ -335,7 +335,7 @@ namespace {
       // Do not want to hoist if we're not optimizing for size.
       // TODO: We'd like to remove this restriction.
       // See the comment in X86InstrInfo.td for more info.
-      if (!OptForSize)
+      if (!CurDAG->shouldOptForSize())
         return false;
 
       // Walk all the users of the immediate.
@@ -3019,7 +3019,7 @@ bool X86DAGToDAGISel::foldLoadStoreIntoMemOperand(SDNode *Node) {
    LLVM_FALLTHROUGH;
   case X86ISD::ADD:
     // Try to match inc/dec.
-    if (!Subtarget->slowIncDec() || OptForSize) {
+    if (!Subtarget->slowIncDec() || CurDAG->shouldOptForSize()) {
       bool IsOne = isOneConstant(StoredVal.getOperand(1));
       bool IsNegOne = isAllOnesConstant(StoredVal.getOperand(1));
       // ADD/SUB with 1/-1 and carry flag isn't used can use inc/dec.
diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 2f9a714f92a..07d77485254 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -25,7 +25,9 @@
 #include "llvm/ADT/Statistic.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringSwitch.h"
+#include "llvm/Analysis/BlockFrequencyInfo.h"
 #include "llvm/Analysis/EHPersonalities.h"
+#include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/CodeGen/IntrinsicLowering.h"
 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineFunction.h"
@@ -8397,7 +8399,7 @@ static SDValue lowerBuildVectorAsBroadcast(BuildVectorSDNode *BVOp,
   // TODO: If multiple splats are generated to load the same constant,
   // it may be detrimental to overall size. There needs to be a way to detect
   // that condition to know if this is truly a size win.
-  bool OptForSize = DAG.getMachineFunction().getFunction().hasOptSize();
+  bool OptForSize = DAG.shouldOptForSize();
 
   // Handle broadcasting a single constant scalar from the constant pool
   // into a vector.
@@ -11174,7 +11176,7 @@ static SDValue lowerShuffleAsBlend(const SDLoc &DL, MVT VT, SDValue V1,
   case MVT::v32i16:
   case MVT::v64i8: {
     // Attempt to lower to a bitmask if we can. Only if not optimizing for size.
-    bool OptForSize = DAG.getMachineFunction().getFunction().hasOptSize();
+    bool OptForSize = DAG.shouldOptForSize();
     if (!OptForSize) {
       if (SDValue Masked = lowerShuffleAsBitMask(DL, VT, V1, V2, Mask, Zeroable,
                                                  Subtarget, DAG))
@@ -18315,7 +18317,7 @@ static SDValue LowerFunnelShift(SDValue Op, const X86Subtarget &Subtarget,
          "Unexpected funnel shift type!");
 
   // Expand slow SHLD/SHRD cases if we are not optimizing for size.
-  bool OptForSize = DAG.getMachineFunction().getFunction().hasOptSize();
+  bool OptForSize = DAG.shouldOptForSize();
   if (!OptForSize && Subtarget.isSHLDSlow())
     return SDValue();
 
@@ -18547,7 +18549,7 @@ SDValue X86TargetLowering::BuildFILD(SDValue Op, EVT SrcVT, SDValue Chain,
 /// implementation, and likely shuffle complexity of the alternate sequence.
 static bool shouldUseHorizontalOp(bool IsSingleSource, SelectionDAG &DAG,
                                   const X86Subtarget &Subtarget) {
-  bool IsOptimizingSize = DAG.getMachineFunction().getFunction().hasOptSize();
+  bool IsOptimizingSize = DAG.shouldOptForSize();
   bool HasFastHOps = Subtarget.hasFastHorizontalOps();
   return !IsSingleSource || IsOptimizingSize || HasFastHOps;
 }
@@ -20473,7 +20475,7 @@ static SDValue LowerAndToBT(SDValue And, ISD::CondCode CC,
     } else {
       // Use BT if the immediate can't be encoded in a TEST instruction or we
       // are optimizing for size and the immedaite won't fit in a byte.
-      bool OptForSize = DAG.getMachineFunction().getFunction().hasOptSize();
+      bool OptForSize = DAG.shouldOptForSize();
       if ((!isUInt<32>(AndRHSVal) || (OptForSize && !isUInt<8>(AndRHSVal))) &&
           isPowerOf2_64(AndRHSVal)) {
         Src = AndLHS;
@@ -39645,7 +39647,7 @@ static SDValue combineOrShiftToFunnelShift(SDNode *N, SelectionDAG &DAG,
     return SDValue();
 
   // fold (or (x << c) | (y >> (64 - c))) ==> (shld64 x, y, c)
-  bool OptForSize = DAG.getMachineFunction().getFunction().hasOptSize();
+  bool OptForSize = DAG.shouldOptForSize();
   unsigned Bits = VT.getScalarSizeInBits();
 
   // SHLD/SHRD instructions have lower register pressure, but on some
diff --git a/llvm/lib/Target/X86/X86InstrInfo.td b/llvm/lib/Target/X86/X86InstrInfo.td
index e452145f3b6..3ab45ba833f 100644
--- a/llvm/lib/Target/X86/X86InstrInfo.td
+++ b/llvm/lib/Target/X86/X86InstrInfo.td
@@ -983,12 +983,12 @@ def IsNotPIC     : Predicate<"!TM.isPositionIndependent()">;
 // the Function object through the <Target>Subtarget and objections were raised
 // to that (see post-commit review comments for r301750).
 let RecomputePerFunction = 1 in {
-  def OptForSize   : Predicate<"MF->getFunction().hasOptSize()">;
+  def OptForSize   : Predicate<"shouldOptForSize(MF)">;
   def OptForMinSize : Predicate<"MF->getFunction().hasMinSize()">;
-  def OptForSpeed  : Predicate<"!MF->getFunction().hasOptSize()">;
+  def OptForSpeed  : Predicate<"!shouldOptForSize(MF)">;
   def UseIncDec : Predicate<"!Subtarget->slowIncDec() || "
-                            "MF->getFunction().hasOptSize()">;
-  def NoSSE41_Or_OptForSize : Predicate<"MF->getFunction().hasOptSize() || "
+                            "shouldOptForSize(MF)">;
+  def NoSSE41_Or_OptForSize : Predicate<"shouldOptForSize(MF) || "
                                         "!Subtarget->hasSSE41()">;
 }