LLVM and SPIRV-LLVM-Translator pulldown (WW41)

LLVM: llvm/llvm-project@894927b491b7
SPIRV-LLVM-Translator: KhronosGroup/SPIRV-LLVM-Translator@e68ddf7

Author: bb-sycl

bolt/include/bolt/Core/MCPlus.h

@@ -66,6 +66,7 @@ class MCAnnotation {
     kTailCall,            /// Tail call.
     kConditionalTailCall, /// CTC.
     kOffset,              /// Offset in the function.
+    kLabel,               /// MCSymbol pointing to this instruction.
     kGeneric              /// First generic annotation.
   };
 

bolt/include/bolt/Core/MCPlusBuilder.h

@@ -160,6 +160,7 @@ class MCPlusBuilder {
   const MCInstrAnalysis *Analysis;
   const MCInstrInfo *Info;
   const MCRegisterInfo *RegInfo;
+  const MCSubtargetInfo *STI;
 
   /// Map annotation name into an annotation index.
   StringMap<uint64_t> AnnotationNameIndexMap;
@@ -331,8 +332,8 @@ class MCPlusBuilder {
 
 public:
   MCPlusBuilder(const MCInstrAnalysis *Analysis, const MCInstrInfo *Info,
-                const MCRegisterInfo *RegInfo)
-      : Analysis(Analysis), Info(Info), RegInfo(RegInfo) {
+                const MCRegisterInfo *RegInfo, const MCSubtargetInfo *STI)
+      : Analysis(Analysis), Info(Info), RegInfo(RegInfo), STI(STI) {
     // Initialize the default annotation allocator with id 0
     AnnotationAllocators.emplace(0, AnnotationAllocator());
     MaxAllocatorId++;
@@ -1179,6 +1180,13 @@ class MCPlusBuilder {
   /// Remove offset annotation.
   bool clearOffset(MCInst &Inst);
 
+  /// Return the label of \p Inst, if available.
+  std::optional<MCSymbol *> getLabel(const MCInst &Inst) const;
+
+  /// Set the label of \p Inst. This label will be emitted right before \p Inst
+  /// is emitted to MCStreamer.
+  bool setLabel(MCInst &Inst, MCSymbol *Label, AllocatorIdTy AllocatorId = 0);
+
   /// Return MCSymbol that represents a target of this instruction at a given
   /// operand number \p OpNum. If there's no symbol associated with
   /// the operand - return nullptr.
@@ -2079,15 +2087,18 @@ class MCPlusBuilder {
 
 MCPlusBuilder *createX86MCPlusBuilder(const MCInstrAnalysis *,
                                       const MCInstrInfo *,
-                                      const MCRegisterInfo *);
+                                      const MCRegisterInfo *,
+                                      const MCSubtargetInfo *);
 
 MCPlusBuilder *createAArch64MCPlusBuilder(const MCInstrAnalysis *,
                                           const MCInstrInfo *,
-                                          const MCRegisterInfo *);
+                                          const MCRegisterInfo *,
+                                          const MCSubtargetInfo *);
 
 MCPlusBuilder *createRISCVMCPlusBuilder(const MCInstrAnalysis *,
                                         const MCInstrInfo *,
-                                        const MCRegisterInfo *);
+                                        const MCRegisterInfo *,
+                                        const MCSubtargetInfo *);
 
 } // namespace bolt
 } // namespace llvm

bolt/include/bolt/Core/Relocation.h

@@ -98,6 +98,10 @@ struct Relocation {
   /// Return true if relocation type is for thread local storage.
   static bool isTLS(uint64_t Type);
 
+  /// Return true of relocation type is for referencing a specific instruction
+  /// (as opposed to a function, basic block, etc).
+  static bool isInstructionReference(uint64_t Type);
+
   /// Return code for a NONE relocation
   static uint64_t getNone();
 

bolt/include/bolt/Rewrite/RewriteInstance.h

@@ -584,7 +584,8 @@ class RewriteInstance {
 MCPlusBuilder *createMCPlusBuilder(const Triple::ArchType Arch,
                                    const MCInstrAnalysis *Analysis,
                                    const MCInstrInfo *Info,
-                                   const MCRegisterInfo *RegInfo);
+                                   const MCRegisterInfo *RegInfo,
+                                   const MCSubtargetInfo *STI);
 
 } // namespace bolt
 } // namespace llvm

bolt/lib/Core/BinaryContext.cpp

@@ -1888,6 +1888,8 @@ void BinaryContext::printInstruction(raw_ostream &OS, const MCInst &Instruction,
   }
   if (std::optional<uint32_t> Offset = MIB->getOffset(Instruction))
     OS << " # Offset: " << *Offset;
+  if (auto Label = MIB->getLabel(Instruction))
+    OS << " # Label: " << **Label;
 
   MIB->printAnnotations(Instruction, OS);
 

bolt/lib/Core/BinaryEmitter.cpp

@@ -498,6 +498,9 @@ void BinaryEmitter::emitFunctionBody(BinaryFunction &BF, FunctionFragment &FF,
         BB->getLocSyms().emplace_back(Offset, LocSym);
       }
 
+      if (auto Label = BC.MIB->getLabel(Instr))
+        Streamer.emitLabel(*Label);
+
       Streamer.emitInstruction(Instr, *BC.STI);
       LastIsPrefix = BC.MIB->isPrefix(Instr);
     }

bolt/lib/Core/BinaryFunction.cpp

@@ -1173,6 +1173,13 @@ bool BinaryFunction::disassemble() {
   // basic block.
   Labels[0] = Ctx->createNamedTempSymbol("BB0");
 
+  // Map offsets in the function to a label that should always point to the
+  // corresponding instruction. This is used for labels that shouldn't point to
+  // the start of a basic block but always to a specific instruction. This is
+  // used, for example, on RISC-V where %pcrel_lo relocations point to the
+  // corresponding %pcrel_hi.
+  LabelsMapType InstructionLabels;
+
   uint64_t Size = 0; // instruction size
   for (uint64_t Offset = 0; Offset < getSize(); Offset += Size) {
     MCInst Instruction;
@@ -1329,9 +1336,23 @@ bool BinaryFunction::disassemble() {
                 ItrE = Relocations.lower_bound(Offset + Size);
            Itr != ItrE; ++Itr) {
         const Relocation &Relocation = Itr->second;
+        MCSymbol *Symbol = Relocation.Symbol;
+
+        if (Relocation::isInstructionReference(Relocation.Type)) {
+          uint64_t RefOffset = Relocation.Value - getAddress();
+          LabelsMapType::iterator LI = InstructionLabels.find(RefOffset);
+
+          if (LI == InstructionLabels.end()) {
+            Symbol = BC.Ctx->createNamedTempSymbol();
+            InstructionLabels.emplace(RefOffset, Symbol);
+          } else {
+            Symbol = LI->second;
+          }
+        }
+
         int64_t Value = Relocation.Value;
         const bool Result = BC.MIB->replaceImmWithSymbolRef(
-            Instruction, Relocation.Symbol, Relocation.Addend, Ctx.get(), Value,
+            Instruction, Symbol, Relocation.Addend, Ctx.get(), Value,
             Relocation.Type);
         (void)Result;
         assert(Result && "cannot replace immediate with relocation");
@@ -1366,6 +1387,13 @@ bool BinaryFunction::disassemble() {
     addInstruction(Offset, std::move(Instruction));
   }
 
+  for (auto [Offset, Label] : InstructionLabels) {
+    InstrMapType::iterator II = Instructions.find(Offset);
+    assert(II != Instructions.end() && "reference to non-existing instruction");
+
+    BC.MIB->setLabel(II->second, Label);
+  }
+
   // Reset symbolizer for the disassembler.
   BC.SymbolicDisAsm->setSymbolizer(nullptr);
 
@@ -4489,7 +4517,7 @@ void BinaryFunction::addRelocation(uint64_t Address, MCSymbol *Symbol,
   uint64_t Offset = Address - getAddress();
   LLVM_DEBUG(dbgs() << "BOLT-DEBUG: addRelocation in "
                     << formatv("{0}@{1:x} against {2}\n", *this, Offset,
-                               Symbol->getName()));
+                               (Symbol ? Symbol->getName() : "<undef>")));
   bool IsCI = BC.isAArch64() && isInConstantIsland(Address);
   std::map<uint64_t, Relocation> &Rels =
       IsCI ? Islands->Relocations : Relocations;

bolt/lib/Core/MCPlusBuilder.cpp

@@ -268,6 +268,19 @@ bool MCPlusBuilder::clearOffset(MCInst &Inst) {
   return true;
 }
 
+std::optional<MCSymbol *> MCPlusBuilder::getLabel(const MCInst &Inst) const {
+  if (auto Label = tryGetAnnotationAs<MCSymbol *>(Inst, MCAnnotation::kLabel))
+    return *Label;
+  return std::nullopt;
+}
+
+bool MCPlusBuilder::setLabel(MCInst &Inst, MCSymbol *Label,
+                             AllocatorIdTy AllocatorId) {
+  getOrCreateAnnotationAs<MCSymbol *>(Inst, MCAnnotation::kLabel, AllocatorId) =
+      Label;
+  return true;
+}
+
 bool MCPlusBuilder::hasAnnotation(const MCInst &Inst, unsigned Index) const {
   const MCInst *AnnotationInst = getAnnotationInst(Inst);
   if (!AnnotationInst)

bolt/lib/Core/Relocation.cpp

@@ -858,6 +858,19 @@ bool Relocation::isTLS(uint64_t Type) {
   return isTLSX86(Type);
 }
 
+bool Relocation::isInstructionReference(uint64_t Type) {
+  if (Arch != Triple::riscv64)
+    return false;
+
+  switch (Type) {
+  default:
+    return false;
+  case ELF::R_RISCV_PCREL_LO12_I:
+  case ELF::R_RISCV_PCREL_LO12_S:
+    return true;
+  }
+}
+
 uint64_t Relocation::getNone() {
   if (Arch == Triple::aarch64)
     return ELF::R_AARCH64_NONE;

bolt/lib/Passes/BinaryPasses.cpp

@@ -575,6 +575,7 @@ bool CheckLargeFunctions::shouldOptimize(const BinaryFunction &BF) const {
 
 void LowerAnnotations::runOnFunctions(BinaryContext &BC) {
   std::vector<std::pair<MCInst *, uint32_t>> PreservedOffsetAnnotations;
+  std::vector<std::pair<MCInst *, MCSymbol *>> PreservedLabelAnnotations;
 
   for (auto &It : BC.getBinaryFunctions()) {
     BinaryFunction &BF = It.second;
@@ -609,22 +610,29 @@ void LowerAnnotations::runOnFunctions(BinaryContext &BC) {
           if (BF.requiresAddressTranslation() && BC.MIB->getOffset(*II))
             PreservedOffsetAnnotations.emplace_back(&(*II),
                                                     *BC.MIB->getOffset(*II));
+          if (auto Label = BC.MIB->getLabel(*II))
+            PreservedLabelAnnotations.emplace_back(&*II, *Label);
           BC.MIB->stripAnnotations(*II);
         }
       }
     }
   }
   for (BinaryFunction *BF : BC.getInjectedBinaryFunctions())
     for (BinaryBasicBlock &BB : *BF)
-      for (MCInst &Instruction : BB)
+      for (MCInst &Instruction : BB) {
+        if (auto Label = BC.MIB->getLabel(Instruction))
+          PreservedLabelAnnotations.emplace_back(&Instruction, *Label);
         BC.MIB->stripAnnotations(Instruction);
+      }
 
   // Release all memory taken by annotations
   BC.MIB->freeAnnotations();
 
   // Reinsert preserved annotations we need during code emission.
   for (const std::pair<MCInst *, uint32_t> &Item : PreservedOffsetAnnotations)
     BC.MIB->setOffset(*Item.first, Item.second);
+  for (auto [Instr, Label] : PreservedLabelAnnotations)
+    BC.MIB->setLabel(*Instr, Label);
 }
 
 // Check for dirty state in MCSymbol objects that might be a consequence

bolt/lib/Passes/LongJmp.cpp

@@ -138,18 +138,21 @@ BinaryBasicBlock *LongJmpPass::lookupStubFromGroup(
       Cand = LeftCand;
   }
   int BitsAvail = BC.MIB->getPCRelEncodingSize(Inst) - 1;
-  uint64_t Mask = ~((1ULL << BitsAvail) - 1);
+  assert(BitsAvail < 63 && "PCRelEncodingSize is too large to use int64_t to"
+                           "check for out-of-bounds.");
+  int64_t MaxVal = (1ULL << BitsAvail) - 1;
+  int64_t MinVal = -(1ULL << BitsAvail);
   uint64_t PCRelTgtAddress = Cand->first;
-  PCRelTgtAddress = DotAddress > PCRelTgtAddress ? DotAddress - PCRelTgtAddress
-                                                 : PCRelTgtAddress - DotAddress;
+  int64_t PCOffset = (int64_t)(PCRelTgtAddress - DotAddress);
+
   LLVM_DEBUG({
     if (Candidates.size() > 1)
       dbgs() << "Considering stub group with " << Candidates.size()
              << " candidates. DotAddress is " << Twine::utohexstr(DotAddress)
              << ", chosen candidate address is "
              << Twine::utohexstr(Cand->first) << "\n";
   });
-  return PCRelTgtAddress & Mask ? nullptr : Cand->second;
+  return (PCOffset < MinVal || PCOffset > MaxVal) ? nullptr : Cand->second;
 }
 
 BinaryBasicBlock *
@@ -512,13 +515,15 @@ bool LongJmpPass::needsStub(const BinaryBasicBlock &BB, const MCInst &Inst,
   }
 
   int BitsAvail = BC.MIB->getPCRelEncodingSize(Inst) - 1;
-  uint64_t Mask = ~((1ULL << BitsAvail) - 1);
+  assert(BitsAvail < 63 && "PCRelEncodingSize is too large to use int64_t to"
+                           "check for out-of-bounds.");
+  int64_t MaxVal = (1ULL << BitsAvail) - 1;
+  int64_t MinVal = -(1ULL << BitsAvail);
 
   uint64_t PCRelTgtAddress = getSymbolAddress(BC, TgtSym, TgtBB);
-  PCRelTgtAddress = DotAddress > PCRelTgtAddress ? DotAddress - PCRelTgtAddress
-                                                 : PCRelTgtAddress - DotAddress;
+  int64_t PCOffset = (int64_t)(PCRelTgtAddress - DotAddress);
 
-  return PCRelTgtAddress & Mask;
+  return PCOffset < MinVal || PCOffset > MaxVal;
 }
 
 bool LongJmpPass::relax(BinaryFunction &Func) {

bolt/lib/Rewrite/JITLinkLinker.cpp

@@ -192,6 +192,13 @@ void JITLinkLinker::loadObject(MemoryBufferRef Obj,
     exit(1);
   }
 
+  if ((*LG)->getTargetTriple().getArch() != BC.TheTriple->getArch()) {
+    errs() << "BOLT-ERROR: linking object with arch "
+           << (*LG)->getTargetTriple().getArchName()
+           << " into context with arch " << BC.TheTriple->getArchName() << "\n";
+    exit(1);
+  }
+
   auto Ctx = std::make_unique<Context>(*this, MapSections);
   jitlink::link(std::move(*LG), std::move(Ctx));
 }

bolt/lib/Rewrite/MachORewriteInstance.cpp

@@ -56,25 +56,28 @@ namespace bolt {
 
 extern MCPlusBuilder *createX86MCPlusBuilder(const MCInstrAnalysis *,
                                              const MCInstrInfo *,
-                                             const MCRegisterInfo *);
+                                             const MCRegisterInfo *,
+                                             const MCSubtargetInfo *);
 extern MCPlusBuilder *createAArch64MCPlusBuilder(const MCInstrAnalysis *,
                                                  const MCInstrInfo *,
-                                                 const MCRegisterInfo *);
+                                                 const MCRegisterInfo *,
+                                                 const MCSubtargetInfo *);
 
 namespace {
 
 MCPlusBuilder *createMCPlusBuilder(const Triple::ArchType Arch,
                                    const MCInstrAnalysis *Analysis,
                                    const MCInstrInfo *Info,
-                                   const MCRegisterInfo *RegInfo) {
+                                   const MCRegisterInfo *RegInfo,
+                                   const MCSubtargetInfo *STI) {
 #ifdef X86_AVAILABLE
   if (Arch == Triple::x86_64)
-    return createX86MCPlusBuilder(Analysis, Info, RegInfo);
+    return createX86MCPlusBuilder(Analysis, Info, RegInfo, STI);
 #endif
 
 #ifdef AARCH64_AVAILABLE
   if (Arch == Triple::aarch64)
-    return createAArch64MCPlusBuilder(Analysis, Info, RegInfo);
+    return createAArch64MCPlusBuilder(Analysis, Info, RegInfo, STI);
 #endif
 
   llvm_unreachable("architecture unsupported by MCPlusBuilder");
@@ -106,8 +109,9 @@ MachORewriteInstance::MachORewriteInstance(object::MachOObjectFile *InputFile,
     return;
   }
   BC = std::move(BCOrErr.get());
-  BC->initializeTarget(std::unique_ptr<MCPlusBuilder>(createMCPlusBuilder(
-      BC->TheTriple->getArch(), BC->MIA.get(), BC->MII.get(), BC->MRI.get())));
+  BC->initializeTarget(std::unique_ptr<MCPlusBuilder>(
+      createMCPlusBuilder(BC->TheTriple->getArch(), BC->MIA.get(),
+                          BC->MII.get(), BC->MRI.get(), BC->STI.get())));
   if (opts::Instrument)
     BC->setRuntimeLibrary(std::make_unique<InstrumentationRuntimeLibrary>());
 }