Merge branch 'main' into nvptx-symbol-transform

Author: vyom1611

bolt/CMakeLists.txt

@@ -82,7 +82,7 @@ endforeach()
 
 set(BOLT_ENABLE_RUNTIME_default OFF)
 if ((CMAKE_SYSTEM_PROCESSOR STREQUAL "x86_64"
-    OR CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
+    OR CMAKE_SYSTEM_PROCESSOR MATCHES "^(arm64|aarch64)$")
     AND (CMAKE_SYSTEM_NAME STREQUAL "Linux"
       OR CMAKE_SYSTEM_NAME STREQUAL "Darwin")
     AND (NOT CMAKE_CROSSCOMPILING))

bolt/include/bolt/Core/BinaryContext.h

@@ -911,7 +911,8 @@ class BinaryContext {
   /// of \p Flags.
   MCSymbol *registerNameAtAddress(StringRef Name, uint64_t Address,
                                   uint64_t Size, uint16_t Alignment,
-                                  unsigned Flags = 0);
+                                  unsigned Flags = 0,
+                                  BinarySection *Section = NULL);
 
   /// Return BinaryData registered at a given \p Address or nullptr if no
   /// global symbol was registered at the location.

bolt/lib/Core/BinaryContext.cpp

@@ -142,7 +142,6 @@ BinaryContext::BinaryContext(std::unique_ptr<MCContext> Ctx,
       InstPrinter(std::move(InstPrinter)), MIA(std::move(MIA)),
       MIB(std::move(MIB)), MRI(std::move(MRI)), DisAsm(std::move(DisAsm)),
       Logger(Logger), InitialDynoStats(isAArch64()) {
-  Relocation::Arch = this->TheTriple->getArch();
   RegularPageSize = isAArch64() ? RegularPageSizeAArch64 : RegularPageSizeX86;
   PageAlign = opts::NoHugePages ? RegularPageSize : HugePageSize;
 }
@@ -1056,18 +1055,28 @@ void BinaryContext::adjustCodePadding() {
 MCSymbol *BinaryContext::registerNameAtAddress(StringRef Name, uint64_t Address,
                                                uint64_t Size,
                                                uint16_t Alignment,
-                                               unsigned Flags) {
+                                               unsigned Flags,
+                                               BinarySection *Section) {
   // Register the name with MCContext.
   MCSymbol *Symbol = Ctx->getOrCreateSymbol(Name);
+  BinaryData *BD;
+
+  // Register out of section symbols only in GlobalSymbols map
+  if (Section && Section->getEndAddress() == Address) {
+    BD = new BinaryData(*Symbol, Address, Size, Alignment ? Alignment : 1,
+                        *Section, Flags);
+    GlobalSymbols[Name] = BD;
+    return Symbol;
+  }
 
   auto GAI = BinaryDataMap.find(Address);
-  BinaryData *BD;
   if (GAI == BinaryDataMap.end()) {
     ErrorOr<BinarySection &> SectionOrErr = getSectionForAddress(Address);
-    BinarySection &Section =
-        SectionOrErr ? SectionOrErr.get() : absoluteSection();
+    BinarySection &SectionRef = Section        ? *Section
+                                : SectionOrErr ? SectionOrErr.get()
+                                               : absoluteSection();
     BD = new BinaryData(*Symbol, Address, Size, Alignment ? Alignment : 1,
-                        Section, Flags);
+                        SectionRef, Flags);
     GAI = BinaryDataMap.emplace(Address, BD).first;
     GlobalSymbols[Name] = BD;
     updateObjectNesting(GAI);
@@ -1402,7 +1411,7 @@ void BinaryContext::postProcessSymbolTable() {
     if ((BD->getName().starts_with("SYMBOLat") ||
          BD->getName().starts_with("DATAat")) &&
         !BD->getParent() && !BD->getSize() && !BD->isAbsolute() &&
-        BD->getSection()) {
+        BD->getSection().getSize()) {
       this->errs() << "BOLT-WARNING: zero-sized top level symbol: " << *BD
                    << "\n";
       Valid = false;

bolt/lib/Rewrite/MachORewriteInstance.cpp

@@ -72,6 +72,7 @@ MachORewriteInstance::MachORewriteInstance(object::MachOObjectFile *InputFile,
                                            StringRef ToolPath, Error &Err)
     : InputFile(InputFile), ToolPath(ToolPath) {
   ErrorAsOutParameter EAO(&Err);
+  Relocation::Arch = InputFile->makeTriple().getArch();
   auto BCOrErr = BinaryContext::createBinaryContext(
       InputFile->makeTriple(), InputFile->getFileName(), nullptr,
       /* IsPIC */ true, DWARFContext::create(*InputFile),

bolt/lib/Rewrite/RewriteInstance.cpp

@@ -354,6 +354,7 @@ RewriteInstance::RewriteInstance(ELFObjectFileBase *File, const int Argc,
     }
   }
 
+  Relocation::Arch = TheTriple.getArch();
   auto BCOrErr = BinaryContext::createBinaryContext(
       TheTriple, File->getFileName(), Features.get(), IsPIC,
       DWARFContext::create(*File, DWARFContext::ProcessDebugRelocations::Ignore,
@@ -955,13 +956,13 @@ void RewriteInstance::discoverFileObjects() {
     uint64_t SymbolSize = ELFSymbolRef(Symbol).getSize();
     uint64_t SymbolAlignment = Symbol.getAlignment();
 
-    auto registerName = [&](uint64_t FinalSize) {
+    auto registerName = [&](uint64_t FinalSize, BinarySection *Section = NULL) {
       // Register names even if it's not a function, e.g. for an entry point.
       BC->registerNameAtAddress(UniqueName, SymbolAddress, FinalSize,
-                                SymbolAlignment, SymbolFlags);
+                                SymbolAlignment, SymbolFlags, Section);
       if (!AlternativeName.empty())
         BC->registerNameAtAddress(AlternativeName, SymbolAddress, FinalSize,
-                                  SymbolAlignment, SymbolFlags);
+                                  SymbolAlignment, SymbolFlags, Section);
     };
 
     section_iterator Section =
@@ -986,12 +987,25 @@ void RewriteInstance::discoverFileObjects() {
                       << " for function\n");
 
     if (SymbolAddress == Section->getAddress() + Section->getSize()) {
+      ErrorOr<BinarySection &> SectionOrError =
+          BC->getSectionForAddress(Section->getAddress());
+
+      // Skip symbols from invalid sections
+      if (!SectionOrError) {
+        BC->errs() << "BOLT-WARNING: " << UniqueName << " (0x"
+                   << Twine::utohexstr(SymbolAddress)
+                   << ") does not have any section\n";
+        continue;
+      }
+
       assert(SymbolSize == 0 &&
              "unexpect non-zero sized symbol at end of section");
-      LLVM_DEBUG(
-          dbgs()
-          << "BOLT-DEBUG: rejecting as symbol points to end of its section\n");
-      registerName(SymbolSize);
+      LLVM_DEBUG({
+        dbgs() << "BOLT-DEBUG: rejecting as symbol " << UniqueName
+               << " points to end of " << SectionOrError->getName()
+               << " section\n";
+      });
+      registerName(SymbolSize, &SectionOrError.get());
       continue;
     }
 
@@ -2143,6 +2157,14 @@ bool RewriteInstance::analyzeRelocation(
   if (!Relocation::isSupported(RType))
     return false;
 
+  auto IsWeakReference = [](const SymbolRef &Symbol) {
+    Expected<uint32_t> SymFlagsOrErr = Symbol.getFlags();
+    if (!SymFlagsOrErr)
+      return false;
+    return (*SymFlagsOrErr & SymbolRef::SF_Undefined) &&
+           (*SymFlagsOrErr & SymbolRef::SF_Weak);
+  };
+
   const bool IsAArch64 = BC->isAArch64();
 
   const size_t RelSize = Relocation::getSizeForType(RType);
@@ -2174,7 +2196,8 @@ bool RewriteInstance::analyzeRelocation(
     // Section symbols are marked as ST_Debug.
     IsSectionRelocation = (cantFail(Symbol.getType()) == SymbolRef::ST_Debug);
     // Check for PLT entry registered with symbol name
-    if (!SymbolAddress && (IsAArch64 || BC->isRISCV())) {
+    if (!SymbolAddress && !IsWeakReference(Symbol) &&
+        (IsAArch64 || BC->isRISCV())) {
       const BinaryData *BD = BC->getPLTBinaryDataByName(SymbolName);
       SymbolAddress = BD ? BD->getAddress() : 0;
     }
@@ -2603,7 +2626,7 @@ void RewriteInstance::handleRelocation(const SectionRef &RelocatedSection,
       Expected<StringRef> SectionName = Section->getName();
       if (SectionName && !SectionName->empty())
         ReferencedSection = BC->getUniqueSectionByName(*SectionName);
-    } else if (ReferencedSymbol && ContainingBF &&
+    } else if (BC->isRISCV() && ReferencedSymbol && ContainingBF &&
                (cantFail(Symbol.getFlags()) & SymbolRef::SF_Absolute)) {
       // This might be a relocation for an ABS symbols like __global_pointer$ on
       // RISC-V
@@ -2614,6 +2637,30 @@ void RewriteInstance::handleRelocation(const SectionRef &RelocatedSection,
     }
   }
 
+  if (Relocation::isGOT(RType) && !Relocation::isTLS(RType)) {
+    auto exitOnGotEndSymol = [&](StringRef Name) {
+      BC->errs() << "BOLT-ERROR: GOT table contains currently unsupported "
+                    "section end symbol "
+                 << Name << "\n";
+      exit(1);
+    };
+
+    if (SymbolIter != InputFile->symbol_end() && ReferencedSection) {
+      if (cantFail(SymbolIter->getAddress()) ==
+          ReferencedSection->getEndAddress())
+        exitOnGotEndSymol(cantFail(SymbolIter->getName()));
+    } else {
+      // If no section and symbol are provided by relocation, try to find the
+      // symbol by its name, including the possibility that the symbol is local.
+      BinaryData *BD = BC->getBinaryDataByName(SymbolName);
+      if (!BD && NR.getUniquifiedNameCount(SymbolName) == 1)
+        BD = BC->getBinaryDataByName(NR.getUniqueName(SymbolName, 1));
+
+      if ((BD && BD->getAddress() == BD->getSection().getEndAddress()))
+        exitOnGotEndSymol(BD->getName());
+    }
+  }
+
   if (!ReferencedSection)
     ReferencedSection = BC->getSectionForAddress(SymbolAddress);
 
@@ -5509,6 +5556,14 @@ uint64_t RewriteInstance::getNewFunctionOrDataAddress(uint64_t OldAddress) {
   if (const BinaryFunction *BF =
           BC->getBinaryFunctionContainingAddress(OldAddress)) {
     if (BF->isEmitted()) {
+      // If OldAddress is the another entry point of
+      // the function, then BOLT could get the new address.
+      if (BF->isMultiEntry()) {
+        for (const BinaryBasicBlock &BB : *BF)
+          if (BB.isEntryPoint() &&
+              (BF->getAddress() + BB.getOffset()) == OldAddress)
+            return BF->getOutputAddress() + BB.getOffset();
+      }
       BC->errs() << "BOLT-ERROR: unable to get new address corresponding to "
                     "input address 0x"
                  << Twine::utohexstr(OldAddress) << " in function " << *BF

bolt/test/AArch64/Inputs/build_id.ldscript

@@ -0,0 +1,9 @@
+SECTIONS
+{
+  PROVIDE (__executable_start = SEGMENT_START("text-segment", 0x400000)); . = SEGMENT_START("text-segment", 0x400000) + SIZEOF_HEADERS;
+  .note.gnu.build-id (0x400400):
+   {
+    build_id_note = ABSOLUTE(.);
+    *(.note.gnu.build-id)
+   }
+}

bolt/test/AArch64/Inputs/got_end_of_section_symbol.lld_script

@@ -0,0 +1,6 @@
+SECTIONS {
+  PROVIDE (__executable_start = SEGMENT_START("text-segment", 0x400000)); . = SEGMENT_START("text-segment", 0x400000) + SIZEOF_HEADERS;
+  .data : { *(.data)  *(.array) }
+  .text : { *(.text) }
+  .got  : { *(.got) *(.igot) }
+}

bolt/test/AArch64/build_id.c

@@ -0,0 +1,25 @@
+// This test checks that referencing build_id through GOT table
+// would result in GOT access after disassembly, not directly
+// to build_id address.
+
+// RUN: %clang %cflags -fuse-ld=lld -Wl,-T,%S/Inputs/build_id.ldscript -Wl,-q \
+// RUN:   -Wl,--no-relax -Wl,--build-id=sha1 %s -o %t.exe
+// RUN: llvm-bolt -print-disasm --print-only=get_build_id %t.exe -o %t.bolt | \
+// RUN:   FileCheck %s
+
+// CHECK: adrp	[[REG:x[0-28]+]], __BOLT_got_zero
+// CHECK: ldr x{{.*}}, [[[REG]], :lo12:__BOLT_got_zero{{.*}}]
+
+struct build_id_note {
+  char pad[16];
+  char hash[20];
+};
+
+extern const struct build_id_note build_id_note;
+
+__attribute__((noinline)) char get_build_id() { return build_id_note.hash[0]; }
+
+int main() {
+  get_build_id();
+  return 0;
+}

bolt/test/AArch64/got_end_of_section_symbol.s

@@ -0,0 +1,28 @@
+# RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown \
+# RUN:   %s -o %t.o
+# RUN: %clang %cflags  -nostartfiles -nodefaultlibs -static -Wl,--no-relax \
+# RUN:   -Wl,-q -Wl,-T %S/Inputs/got_end_of_section_symbol.lld_script  \
+# RUN:   %t.o -o %t.exe
+# RUN: not llvm-bolt %t.exe -o %t.bolt 2>&1 | FileCheck %s
+
+# CHECK: BOLT-ERROR: GOT table contains currently unsupported section end
+# CHECK-SAME: symbol array_end
+
+.section .array, "a", @progbits
+.globl array_start
+.globl array_end
+array_start:
+  .word 0
+array_end:
+
+.section .text
+.globl _start
+.type _start, %function
+_start:
+  adrp x1, #:got:array_start
+  ldr x1, [x1, #:got_lo12:array_start]
+  adrp x0, #:got:array_end
+  ldr x0, [x0, #:got_lo12:array_end]
+  adrp x2, #:got:_start
+  ldr x2, [x2, #:got_lo12:_start]
+  ret

bolt/test/AArch64/update-weak-reference-symbol.s

@@ -0,0 +1,34 @@
+// This test checks whether BOLT can correctly handle relocations against weak symbols.
+
+// RUN: %clang %cflags -Wl,-z,notext -shared -Wl,-q %s -o %t.so
+// RUN: llvm-bolt %t.so -o %t.so.bolt
+// RUN: llvm-nm -n %t.so.bolt > %t.out.txt
+// RUN: llvm-objdump -dj .rodata %t.so.bolt >> %t.out.txt
+// RUN: FileCheck %s --input-file=%t.out.txt
+
+# CHECK: w func_1
+# CHECK: {{0+}}[[#%x,ADDR:]] W func_2
+
+# CHECK: {{.*}} <.rodata>:
+# CHECK-NEXT: {{.*}} .word 0x00000000
+# CHECK-NEXT: {{.*}} .word 0x00000000
+# CHECK-NEXT: {{.*}} .word 0x{{[0]+}}[[#ADDR]]
+# CHECK-NEXT: {{.*}} .word 0x00000000
+
+	.text
+	.weak	func_2
+	.weak	func_1
+	.global	wow
+	.type	wow, %function
+wow:
+	bl func_1
+	bl func_2
+	ret
+	.type   func_2, %function
+func_2:
+	ret
+	.section	.rodata
+.LC0:
+	.xword	func_1
+.LC1:
+	.xword	func_2