Extract an utility for computing symbol sizes on MachO and COFF.

I will add a second user in the next commit.

llvm-svn: 240366

llvm/include/llvm/Object/SymbolSize.h

@@ -0,0 +1,26 @@
+//===- SymbolSize.h ---------------------------------------------*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// Declares the writeArchive function for writing an archive file.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_SYMBOLSIZE_H
+#define LLVM_OBJECT_SYMBOLSIZE_H
+
+#include "llvm/Object/ObjectFile.h"
+
+namespace llvm {
+namespace object {
+ErrorOr<std::vector<std::pair<SymbolRef, uint64_t>>>
+computeSymbolSizes(const ObjectFile &O);
+}
+} // namespace llvm
+
+#endif

llvm/lib/Object/CMakeLists.txt

@@ -16,6 +16,7 @@ add_llvm_library(LLVMObject
   ObjectFile.cpp
   RecordStreamer.cpp
   SymbolicFile.cpp
+  SymbolSize.cpp
 
   ADDITIONAL_HEADER_DIRS
   ${LLVM_MAIN_INCLUDE_DIR}/llvm/Object

llvm/lib/Object/SymbolSize.cpp

@@ -0,0 +1,88 @@
+//===- SymbolSize.cpp -----------------------------------------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Object/SymbolSize.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/Object/ELFObjectFile.h"
+
+using namespace llvm;
+using namespace object;
+
+namespace {
+struct SymEntry {
+  symbol_iterator I;
+  uint64_t Address;
+  unsigned Number;
+  SectionRef Section;
+};
+}
+
+static int compareAddress(const SymEntry *A, const SymEntry *B) {
+  if (A->Section == B->Section)
+    return A->Address - B->Address;
+  if (A->Section < B->Section)
+    return -1;
+  return 1;
+}
+
+static int compareNumber(const SymEntry *A, const SymEntry *B) {
+  return A->Number - B->Number;
+}
+
+ErrorOr<std::vector<std::pair<SymbolRef, uint64_t>>>
+llvm::object::computeSymbolSizes(const ObjectFile &O) {
+  std::vector<std::pair<SymbolRef, uint64_t>> Ret;
+
+  if (isa<ELFObjectFileBase>(&O)) {
+    for (SymbolRef Sym : O.symbols()) {
+      Ret.push_back({Sym, Sym.getSize()});
+    }
+    return Ret;
+  }
+
+  // Collect sorted symbol addresses. Include dummy addresses for the end
+  // of each section.
+  std::vector<SymEntry> Addresses;
+  unsigned SymNum = 0;
+  for (symbol_iterator I = O.symbol_begin(), E = O.symbol_end(); I != E; ++I) {
+    SymbolRef Sym = *I;
+    uint64_t Address;
+    if (std::error_code EC = Sym.getAddress(Address))
+      return EC;
+    section_iterator SecI = O.section_end();
+    if (std::error_code EC = Sym.getSection(SecI))
+      return EC;
+    Addresses.push_back({I, Address, SymNum, *SecI});
+    ++SymNum;
+  }
+  for (const SectionRef Sec : O.sections()) {
+    uint64_t Address = Sec.getAddress();
+    uint64_t Size = Sec.getSize();
+    Addresses.push_back({O.symbol_end(), Address + Size, 0, Sec});
+  }
+  array_pod_sort(Addresses.begin(), Addresses.end(), compareAddress);
+
+  // Compute the size as the gap to the next symbol
+  for (unsigned I = 0, N = Addresses.size() - 1; I < N; ++I) {
+    auto &P = Addresses[I];
+    if (P.I == O.symbol_end())
+      continue;
+    uint64_t Size = Addresses[I + 1].Address - P.Address;
+    P.Address = Size;
+  }
+
+  // Put back in the original order and copy the result
+  array_pod_sort(Addresses.begin(), Addresses.end(), compareNumber);
+  for (SymEntry &P : Addresses) {
+    if (P.I == O.symbol_end())
+      continue;
+    Ret.push_back({*P.I, P.Address});
+  }
+  return Ret;
+}

llvm/tools/llvm-rtdyld/llvm-rtdyld.cpp

@@ -25,6 +25,7 @@
 #include "llvm/MC/MCRegisterInfo.h"
 #include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/Object/MachO.h"
+#include "llvm/Object/SymbolSize.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/DynamicLibrary.h"
 #include "llvm/Support/ManagedStatic.h"
@@ -252,38 +253,14 @@ static int printLineInfoForInput(bool LoadObjects, bool UseDebugObj) {
     std::unique_ptr<DIContext> Context(
       new DWARFContextInMemory(*SymbolObj,LoadedObjInfo.get()));
 
-    // FIXME: This is generally useful. Figure out a place in lib/Object to
-    // put utility functions.
-    std::map<object::SectionRef, std::vector<uint64_t>> FuncAddresses;
-    if (!isa<ELFObjectFileBase>(SymbolObj)) {
-      for (object::SymbolRef Sym : SymbolObj->symbols()) {
-        object::SymbolRef::Type SymType;
-        if (Sym.getType(SymType))
-          continue;
-        if (SymType != object::SymbolRef::ST_Function)
-          continue;
-        uint64_t Addr;
-        if (Sym.getAddress(Addr))
-          continue;
-        object::section_iterator Sec = SymbolObj->section_end();
-        if (Sym.getSection(Sec))
-          continue;
-        std::vector<uint64_t> &Addrs = FuncAddresses[*Sec];
-        if (Addrs.empty()) {
-          uint64_t SecAddr = Sec->getAddress();
-          uint64_t SecSize = Sec->getSize();
-          Addrs.push_back(SecAddr + SecSize);
-        }
-        Addrs.push_back(Addr);
-      }
-      for (auto &Pair : FuncAddresses) {
-        std::vector<uint64_t> &Addrs = Pair.second;
-        array_pod_sort(Addrs.begin(), Addrs.end());
-      }
-    }
+    ErrorOr<std::vector<std::pair<SymbolRef, uint64_t>>> SymAddrOrErr =
+        object::computeSymbolSizes(*SymbolObj);
+    if (std::error_code EC = SymAddrOrErr.getError())
+      return Error(EC.message());
 
     // Use symbol info to iterate functions in the object.
-    for (object::SymbolRef Sym : SymbolObj->symbols()) {
+    for (const auto &P : *SymAddrOrErr) {
+      object::SymbolRef Sym = P.first;
       object::SymbolRef::Type SymType;
       if (Sym.getType(SymType))
         continue;
@@ -295,20 +272,7 @@ static int printLineInfoForInput(bool LoadObjects, bool UseDebugObj) {
         if (Sym.getAddress(Addr))
           continue;
 
-        uint64_t Size;
-        if (isa<ELFObjectFileBase>(SymbolObj)) {
-          Size = Sym.getSize();
-        } else {
-          object::section_iterator Sec = SymbolObj->section_end();
-          if (Sym.getSection(Sec))
-            continue;
-          const std::vector<uint64_t> &Addrs = FuncAddresses[*Sec];
-          auto AddrI = std::find(Addrs.begin(), Addrs.end(), Addr);
-          assert(AddrI != Addrs.end() && (AddrI + 1) != Addrs.end());
-          assert(*AddrI == Addr);
-          Size = *(AddrI + 1) - Addr;
-        }
-
+        uint64_t Size = P.second;
         // If we're not using the debug object, compute the address of the
         // symbol in memory (rather than that in the unrelocated object file)
         // and use that to query the DWARFContext.