Change how symbol sizes are handled in lib/Object.

COFF and MachO only define symbol sizes for common symbols. Reflect that
in the class hierarchy by having a method for common symbols only in the base
and a general one in ELF.

This avoids the need of using a magic value for the size, which had a few
problems
* Most callers didn't check for it.
* The ones that did could not tell the magic value from a file actually having
  that value.

llvm-svn: 240529

5 files changed, 32 insertions, 36 deletions

diff --git a/llvm/tools/dsymutil/MachODebugMapParser.cpp b/llvm/tools/dsymutil/MachODebugMapParser.cpp
index b803e410199..b1e6abcefa8 100644
--- a/llvm/tools/dsymutil/MachODebugMapParser.cpp
+++ b/llvm/tools/dsymutil/MachODebugMapParser.cpp
@@ -160,7 +160,7 @@ void MachODebugMapParser::handleStabSymbolTableEntry(uint32_t StringIndex,
     // symbol table to find its address as it might not be in the
     // debug map (for common symbols).
     Value = getMainBinarySymbolAddress(Name);
-    if (Value == UnknownAddressOrSize)
+    if (Value == UnknownAddress)
       return;
     break;
   case MachO::N_FUN:
@@ -199,8 +199,7 @@ void MachODebugMapParser::loadCurrentObjectFileSymbols() {
   for (auto Sym : CurrentObjectHolder.Get().symbols()) {
     StringRef Name;
     uint64_t Addr;
-    if (Sym.getAddress(Addr) || Addr == UnknownAddressOrSize ||
-        Sym.getName(Name))
+    if (Sym.getAddress(Addr) || Addr == UnknownAddress || Sym.getName(Name))
       continue;
     CurrentObjectAddresses[Name] = Addr;
   }
@@ -212,7 +211,7 @@ void MachODebugMapParser::loadCurrentObjectFileSymbols() {
 uint64_t MachODebugMapParser::getMainBinarySymbolAddress(StringRef Name) {
   auto Sym = MainBinarySymbolAddresses.find(Name);
   if (Sym == MainBinarySymbolAddresses.end())
-    return UnknownAddressOrSize;
+    return UnknownAddress;
   return Sym->second;
 }
 
@@ -233,7 +232,7 @@ void MachODebugMapParser::loadMainBinarySymbols() {
     // are the only ones that need to be queried because the address
     // of common data won't be described in the debug map. All other
     // addresses should be fetched for the debug map.
-    if (Sym.getAddress(Addr) || Addr == UnknownAddressOrSize ||
+    if (Sym.getAddress(Addr) || Addr == UnknownAddress ||
         !(Sym.getFlags() & SymbolRef::SF_Global) || Sym.getSection(Section) ||
         Section->isText() || Sym.getName(Name) || Name.size() == 0 ||
         Name[0] == '\0')
diff --git a/llvm/tools/llvm-nm/llvm-nm.cpp b/llvm/tools/llvm-nm/llvm-nm.cpp
index 8013f584239..b96d152f783 100644
--- a/llvm/tools/llvm-nm/llvm-nm.cpp
+++ b/llvm/tools/llvm-nm/llvm-nm.cpp
@@ -569,7 +569,7 @@ static void sortAndPrintSymbolList(SymbolicFile &Obj, bool printName,
       continue;
     if ((I->TypeChar == 'U') && DefinedOnly)
       continue;
-    if (SizeSort && !PrintAddress && I->Size == UnknownAddressOrSize)
+    if (SizeSort && !PrintAddress)
       continue;
     if (PrintFileName) {
       if (!ArchitectureName.empty())
@@ -586,16 +586,15 @@ static void sortAndPrintSymbolList(SymbolicFile &Obj, bool printName,
     char SymbolAddrStr[18] = "";
     char SymbolSizeStr[18] = "";
 
-    if (OutputFormat == sysv || I->Address == UnknownAddressOrSize)
+    if (OutputFormat == sysv || I->Address == UnknownAddress)
       strcpy(SymbolAddrStr, printBlanks);
     if (OutputFormat == sysv)
       strcpy(SymbolSizeStr, printBlanks);
 
-    if (I->Address != UnknownAddressOrSize)
+    if (I->Address != UnknownAddress)
       format(printFormat, I->Address)
           .print(SymbolAddrStr, sizeof(SymbolAddrStr));
-    if (I->Size != UnknownAddressOrSize)
-      format(printFormat, I->Size).print(SymbolSizeStr, sizeof(SymbolSizeStr));
+    format(printFormat, I->Size).print(SymbolSizeStr, sizeof(SymbolSizeStr));
 
     // If OutputFormat is darwin or we are printing Mach-O symbols in hex and
     // we have a MachOObjectFile, call darwinPrintSymbol to print as darwin's
@@ -613,8 +612,7 @@ static void sortAndPrintSymbolList(SymbolicFile &Obj, bool printName,
         outs() << SymbolAddrStr << ' ';
       if (PrintSize) {
         outs() << SymbolSizeStr;
-        if (I->Size != UnknownAddressOrSize)
-          outs() << ' ';
+        outs() << ' ';
       }
       outs() << I->TypeChar;
       if (I->TypeChar == '-' && MachO)
@@ -930,11 +928,13 @@ static void dumpSymbolNamesFromObject(SymbolicFile &Obj, bool printName,
     if (Nsect && Nsect != getNsectInMachO(*MachO, I))
       continue;
     NMSymbol S;
-    S.Size = UnknownAddressOrSize;
-    S.Address = UnknownAddressOrSize;
-    if (PrintSize && isa<ELFObjectFileBase>(Obj)) {
-      symbol_iterator SymI = I;
-      S.Size = SymI->getSize();
+    S.Size = 0;
+    S.Address = UnknownAddress;
+    if (PrintSize) {
+      if (auto *E = dyn_cast<ELFObjectFileBase>(&Obj)) {
+        symbol_iterator SymI = I;
+        S.Size = E->getSymbolSize(*SymI);
+      }
     }
     if (PrintAddress && isa<ObjectFile>(Obj))
       if (error(symbol_iterator(I)->getAddress(S.Address)))
diff --git a/llvm/tools/llvm-objdump/MachODump.cpp b/llvm/tools/llvm-objdump/MachODump.cpp
index 1730bf3859f..bc743bcd067 100644
--- a/llvm/tools/llvm-objdump/MachODump.cpp
+++ b/llvm/tools/llvm-objdump/MachODump.cpp
@@ -2417,10 +2417,9 @@ static const char *get_pointer_32(uint32_t Address, uint32_t &offset,
 // for the specified section offset in the specified section reference.
 // If no relocation information is found and a non-zero ReferenceValue for the
 // symbol is passed, look up that address in the info's AddrMap.
-static const char *
-get_symbol_64(uint32_t sect_offset, SectionRef S, DisassembleInfo *info,
-              uint64_t &n_value,
-              uint64_t ReferenceValue = UnknownAddressOrSize) {
+static const char *get_symbol_64(uint32_t sect_offset, SectionRef S,
+                                 DisassembleInfo *info, uint64_t &n_value,
+                                 uint64_t ReferenceValue = UnknownAddress) {
   n_value = 0;
   if (!info->verbose)
     return nullptr;
@@ -2454,7 +2453,7 @@ get_symbol_64(uint32_t sect_offset, SectionRef S, DisassembleInfo *info,
   const char *SymbolName = nullptr;
   if (reloc_found && isExtern) {
     Symbol.getAddress(n_value);
-    if (n_value == UnknownAddressOrSize)
+    if (n_value == UnknownAddress)
       n_value = 0;
     StringRef name;
     Symbol.getName(name);
@@ -2475,7 +2474,7 @@ get_symbol_64(uint32_t sect_offset, SectionRef S, DisassembleInfo *info,
 
   // We did not find an external relocation entry so look up the ReferenceValue
   // as an address of a symbol and if found return that symbol's name.
-  if (ReferenceValue != UnknownAddressOrSize)
+  if (ReferenceValue != UnknownAddress)
     SymbolName = GuessSymbolName(ReferenceValue, info->AddrMap);
 
   return SymbolName;
diff --git a/llvm/tools/llvm-objdump/llvm-objdump.cpp b/llvm/tools/llvm-objdump/llvm-objdump.cpp
index 8adbc70afc7..518d6de34fa 100644
--- a/llvm/tools/llvm-objdump/llvm-objdump.cpp
+++ b/llvm/tools/llvm-objdump/llvm-objdump.cpp
@@ -784,7 +784,7 @@ static void DisassembleObject(const ObjectFile *Obj, bool InlineRelocs) {
         uint64_t Address;
         if (error(Symbol.getAddress(Address)))
           break;
-        if (Address == UnknownAddressOrSize)
+        if (Address == UnknownAddress)
           continue;
         Address -= SectionAddr;
         if (Address >= SectSize)
@@ -1101,9 +1101,9 @@ void llvm::PrintSymbolTable(const ObjectFile *o) {
     bool Hidden = Flags & SymbolRef::SF_Hidden;
 
     if (Common)
-      Address = Symbol.getSize();
+      Address = Symbol.getCommonSize();
 
-    if (Address == UnknownAddressOrSize)
+    if (Address == UnknownAddress)
       Address = 0;
     char GlobLoc = ' ';
     if (Type != SymbolRef::ST_Unknown)
@@ -1149,7 +1149,8 @@ void llvm::PrintSymbolTable(const ObjectFile *o) {
 
     outs() << '\t';
     if (Common || isa<ELFObjectFileBase>(o)) {
-      uint64_t Val = Common ? Symbol.getAlignment() :  Symbol.getSize();
+      uint64_t Val = Common ? Symbol.getAlignment()
+                            : cast<ELFObjectFileBase>(o)->getSymbolSize(Symbol);
       outs() << format("\t %08" PRIx64 " ", Val);
     }
 
diff --git a/llvm/tools/llvm-symbolizer/LLVMSymbolize.cpp b/llvm/tools/llvm-symbolizer/LLVMSymbolize.cpp
index 448cfe99ddd..b6af342a0bc 100644
--- a/llvm/tools/llvm-symbolizer/LLVMSymbolize.cpp
+++ b/llvm/tools/llvm-symbolizer/LLVMSymbolize.cpp
@@ -94,7 +94,7 @@ void ModuleInfo::addSymbol(const SymbolRef &Symbol, DataExtractor *OpdExtractor,
     return;
   uint64_t SymbolAddress;
   if (error(Symbol.getAddress(SymbolAddress)) ||
-      SymbolAddress == UnknownAddressOrSize)
+      SymbolAddress == UnknownAddress)
     return;
   if (OpdExtractor) {
     // For big-endian PowerPC64 ELF, symbols in the .opd section refer to
@@ -109,15 +109,12 @@ void ModuleInfo::addSymbol(const SymbolRef &Symbol, DataExtractor *OpdExtractor,
       SymbolAddress = OpdExtractor->getAddress(&OpdOffset32);
   }
   uint64_t SymbolSize;
-  // Getting symbol size is linear for Mach-O files, so assume that symbol
-  // occupies the memory range up to the following symbol.
-  if (isa<MachOObjectFile>(Module))
+  // Onyl ELF has a size for every symbol so assume that symbol occupies the
+  // memory range up to the following symbol.
+  if (auto *E = dyn_cast<ELFObjectFileBase>(Module))
+    SymbolSize = E->getSymbolSize(Symbol);
+  else
     SymbolSize = 0;
-  else {
-    SymbolSize = Symbol.getSize();
-    if (SymbolSize == UnknownAddressOrSize)
-      return;
-  }
   StringRef SymbolName;
   if (error(Symbol.getName(SymbolName)))
     return;