vendor/llvm/llvm-release_601-r335540vendor/llvm-60

5 files changed, 66 insertions, 6 deletions

diff --git a/lib/Target/AArch64/AArch64AsmPrinter.cpp b/lib/Target/AArch64/AArch64AsmPrinter.cpp
index 2ff2ee347f560..6704fa27c86ee 100644
--- a/lib/Target/AArch64/AArch64AsmPrinter.cpp
+++ b/lib/Target/AArch64/AArch64AsmPrinter.cpp
@@ -299,6 +299,11 @@ void AArch64AsmPrinter::printOperand(const MachineInstr *MI, unsigned OpNum,
     printOffset(MO.getOffset(), O);
     break;
   }
+  case MachineOperand::MO_BlockAddress: {
+    MCSymbol *Sym = GetBlockAddressSymbol(MO.getBlockAddress());
+    Sym->print(O, MAI);
+    break;
+  }
   }
 }
 
diff --git a/lib/Target/AArch64/AArch64FalkorHWPFFix.cpp b/lib/Target/AArch64/AArch64FalkorHWPFFix.cpp
index d1ddb2e3ef703..0d00dab598d57 100644
--- a/lib/Target/AArch64/AArch64FalkorHWPFFix.cpp
+++ b/lib/Target/AArch64/AArch64FalkorHWPFFix.cpp
@@ -46,6 +46,7 @@
 #include "llvm/Pass.h"
 #include "llvm/Support/Casting.h"
 #include "llvm/Support/Debug.h"
+#include "llvm/Support/DebugCounter.h"
 #include "llvm/Support/raw_ostream.h"
 #include <cassert>
 #include <iterator>
@@ -60,6 +61,8 @@ STATISTIC(NumCollisionsAvoided,
           "Number of HW prefetch tag collisions avoided");
 STATISTIC(NumCollisionsNotAvoided,
           "Number of HW prefetch tag collisions not avoided due to lack of regsiters");
+DEBUG_COUNTER(FixCounter, "falkor-hwpf",
+              "Controls which tag collisions are avoided");
 
 namespace {
 
@@ -729,6 +732,21 @@ void FalkorHWPFFix::runOnLoop(MachineLoop &L, MachineFunction &Fn) {
       bool Fixed = false;
       DEBUG(dbgs() << "Attempting to fix tag collision: " << MI);
 
+      if (!DebugCounter::shouldExecute(FixCounter)) {
+        DEBUG(dbgs() << "Skipping fix due to debug counter:\n  " << MI);
+        continue;
+      }
+
+      // Add the non-base registers of MI as live so we don't use them as
+      // scratch registers.
+      for (unsigned OpI = 0, OpE = MI.getNumOperands(); OpI < OpE; ++OpI) {
+        if (OpI == static_cast<unsigned>(LdI.BaseRegIdx))
+          continue;
+        MachineOperand &MO = MI.getOperand(OpI);
+        if (MO.isReg() && MO.readsReg())
+          LR.addReg(MO.getReg());
+      }
+
       for (unsigned ScratchReg : AArch64::GPR64RegClass) {
         if (!LR.available(ScratchReg) || MRI.isReserved(ScratchReg))
           continue;
diff --git a/lib/Target/AArch64/AArch64FrameLowering.cpp b/lib/Target/AArch64/AArch64FrameLowering.cpp
index d66f7b59a4b52..789200b28445e 100644
--- a/lib/Target/AArch64/AArch64FrameLowering.cpp
+++ b/lib/Target/AArch64/AArch64FrameLowering.cpp
@@ -917,6 +917,8 @@ int AArch64FrameLowering::resolveFrameIndexReference(const MachineFunction &MF,
   int FPOffset = MFI.getObjectOffset(FI) + FixedObject + 16;
   int Offset = MFI.getObjectOffset(FI) + MFI.getStackSize();
   bool isFixed = MFI.isFixedObjectIndex(FI);
+  bool isCSR = !isFixed && MFI.getObjectOffset(FI) >=
+                               -((int)AFI->getCalleeSavedStackSize());
 
   // Use frame pointer to reference fixed objects. Use it for locals if
   // there are VLAs or a dynamically realigned SP (and thus the SP isn't
@@ -930,6 +932,12 @@ int AArch64FrameLowering::resolveFrameIndexReference(const MachineFunction &MF,
     // Argument access should always use the FP.
     if (isFixed) {
       UseFP = hasFP(MF);
+    } else if (isCSR && RegInfo->needsStackRealignment(MF)) {
+      // References to the CSR area must use FP if we're re-aligning the stack
+      // since the dynamically-sized alignment padding is between the SP/BP and
+      // the CSR area.
+      assert(hasFP(MF) && "Re-aligned stack must have frame pointer");
+      UseFP = true;
     } else if (hasFP(MF) && !RegInfo->hasBasePointer(MF) &&
                !RegInfo->needsStackRealignment(MF)) {
       // Use SP or FP, whichever gives us the best chance of the offset
@@ -947,9 +955,9 @@ int AArch64FrameLowering::resolveFrameIndexReference(const MachineFunction &MF,
     }
   }
 
-  assert((isFixed || !RegInfo->needsStackRealignment(MF) || !UseFP) &&
+  assert(((isFixed || isCSR) || !RegInfo->needsStackRealignment(MF) || !UseFP) &&
          "In the presence of dynamic stack pointer realignment, "
-         "non-argument objects cannot be accessed through the frame pointer");
+         "non-argument/CSR objects cannot be accessed through the frame pointer");
 
   if (UseFP) {
     FrameReg = RegInfo->getFrameRegister(MF);
diff --git a/lib/Target/AArch64/AArch64ISelLowering.cpp b/lib/Target/AArch64/AArch64ISelLowering.cpp
index 41ed24c329ef0..233d6be247c2c 100644
--- a/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -4930,7 +4930,8 @@ bool AArch64TargetLowering::isOffsetFoldingLegal(
 bool AArch64TargetLowering::isFPImmLegal(const APFloat &Imm, EVT VT) const {
   // We can materialize #0.0 as fmov $Rd, XZR for 64-bit and 32-bit cases.
   // FIXME: We should be able to handle f128 as well with a clever lowering.
-  if (Imm.isPosZero() && (VT == MVT::f16 || VT == MVT::f64 || VT == MVT::f32)) {
+  if (Imm.isPosZero() && (VT == MVT::f64 || VT == MVT::f32 ||
+                          (VT == MVT::f16 && Subtarget->hasFullFP16()))) {
     DEBUG(dbgs() << "Legal fp imm: materialize 0 using the zero register\n");
     return true;
   }
@@ -5066,7 +5067,7 @@ SDValue AArch64TargetLowering::getRecipEstimate(SDValue Operand,
 
 // Table of Constraints
 // TODO: This is the current set of constraints supported by ARM for the
-// compiler, not all of them may make sense, e.g. S may be difficult to support.
+// compiler, not all of them may make sense.
 //
 // r - A general register
 // w - An FP/SIMD register of some size in the range v0-v31
@@ -5126,6 +5127,8 @@ AArch64TargetLowering::getConstraintType(StringRef Constraint) const {
     // currently handle addresses it is the same as 'r'.
     case 'Q':
       return C_Memory;
+    case 'S': // A symbolic address
+      return C_Other;
     }
   }
   return TargetLowering::getConstraintType(Constraint);
@@ -5250,6 +5253,23 @@ void AArch64TargetLowering::LowerAsmOperandForConstraint(
       Result = DAG.getRegister(AArch64::WZR, MVT::i32);
     break;
   }
+  case 'S': {
+    // An absolute symbolic address or label reference.
+    if (const GlobalAddressSDNode *GA = dyn_cast<GlobalAddressSDNode>(Op)) {
+      Result = DAG.getTargetGlobalAddress(GA->getGlobal(), SDLoc(Op),
+                                          GA->getValueType(0));
+    } else if (const BlockAddressSDNode *BA =
+                   dyn_cast<BlockAddressSDNode>(Op)) {
+      Result =
+          DAG.getTargetBlockAddress(BA->getBlockAddress(), BA->getValueType(0));
+    } else if (const ExternalSymbolSDNode *ES =
+                   dyn_cast<ExternalSymbolSDNode>(Op)) {
+      Result =
+          DAG.getTargetExternalSymbol(ES->getSymbol(), ES->getValueType(0));
+    } else
+      return;
+    break;
+  }
 
   case 'I':
   case 'J':
@@ -9637,6 +9657,15 @@ static SDValue performPostLD1Combine(SDNode *N,
   if (LD->getOpcode() != ISD::LOAD)
     return SDValue();
 
+  // The vector lane must be a constant in the LD1LANE opcode.
+  SDValue Lane;
+  if (IsLaneOp) {
+    Lane = N->getOperand(2);
+    auto *LaneC = dyn_cast<ConstantSDNode>(Lane);
+    if (!LaneC || LaneC->getZExtValue() >= VT.getVectorNumElements())
+      return SDValue();
+  }
+
   LoadSDNode *LoadSDN = cast<LoadSDNode>(LD);
   EVT MemVT = LoadSDN->getMemoryVT();
   // Check if memory operand is the same type as the vector element.
@@ -9693,7 +9722,7 @@ static SDValue performPostLD1Combine(SDNode *N,
     Ops.push_back(LD->getOperand(0));  // Chain
     if (IsLaneOp) {
       Ops.push_back(Vector);           // The vector to be inserted
-      Ops.push_back(N->getOperand(2)); // The lane to be inserted in the vector
+      Ops.push_back(Lane);             // The lane to be inserted in the vector
     }
     Ops.push_back(Addr);
     Ops.push_back(Inc);
diff --git a/lib/Target/AArch64/AArch64InstrInfo.td b/lib/Target/AArch64/AArch64InstrInfo.td
index 79826ca2ed8dc..040011d858e77 100644
--- a/lib/Target/AArch64/AArch64InstrInfo.td
+++ b/lib/Target/AArch64/AArch64InstrInfo.td
@@ -2713,7 +2713,7 @@ defm FMOV : UnscaledConversion<"fmov">;
 // Add pseudo ops for FMOV 0 so we can mark them as isReMaterializable
 let isReMaterializable = 1, isCodeGenOnly = 1, isAsCheapAsAMove = 1 in {
 def FMOVH0 : Pseudo<(outs FPR16:$Rd), (ins), [(set f16:$Rd, (fpimm0))]>,
-    Sched<[WriteF]>;
+    Sched<[WriteF]>, Requires<[HasFullFP16]>;
 def FMOVS0 : Pseudo<(outs FPR32:$Rd), (ins), [(set f32:$Rd, (fpimm0))]>,
     Sched<[WriteF]>;
 def FMOVD0 : Pseudo<(outs FPR64:$Rd), (ins), [(set f64:$Rd, (fpimm0))]>,