[llvm] Check for overflows when computing load-command's addresses.

[oontvoo]

Summary

PR/56746

[alexander-shaposhnikov]

llvm/lib/Object/MachOObjectFile.cpp

210

nit: I'd probably report a separate message for the overflow case (cause it'd make debugging a tiny bit easier)

[tschuett]

llvm/lib/Object/MachOObjectFile.cpp

210

nit: could you instead use an integer with a defined bit width: uint64_t?

[oontvoo]

llvm/lib/Object/MachOObjectFile.cpp

210

nit: could you instead use an integer with a defined bit width: uint64_t?

How about uintptr_t ?

[tschuett]

Sounds good.

[oontvoo]

• updated type
• separate overflow case to different err msg to make debugging easier

[jyknight]

llvm/lib/Object/MachOObjectFile.cpp

210

long unsigned int should be uintptr_t here.
And I think the comparison should be >=?

258

Probably all of the checks here can go away, except to check that L.Ptr + L.C.cmdsize doesn't overflow.

All the range checks are done by getLoadCommandInfo anyways?

[tschuett]

llvm/lib/Object/MachOObjectFile.cpp

260

I am known to be stupid. You do saturating add with uintptr_t and claim errors on uint32_t?

Would it help to make the asserts into real errors? I like to be informed about malformed files in release builds.

[oontvoo]

addressed review comment:

• refactor code so that the overflow checks can be done in place
• remove redundant check
• fix off-by-one

[oontvoo]

llvm/lib/Object/MachOObjectFile.cpp

260

oops, sorry - forgot to update the error messages when I updated the types.
yes - can make these errors instead of asserts.

[oontvoo]

updated err msgs and removed a few asserts

[alexander-shaposhnikov]

llvm/lib/Object/MachOObjectFile.cpp

137

A slightly more general version:

template <class T, class... Ts>
std::enable_if_t<std::is_unsigned<T>::value, T>
SaturatingAdd(T X, T Y, T Z, Ts... args) {
  bool Overflowed = false;
  T XY = SaturatingAdd(X, Y, &Overflowed);
  if (Overflowed)
    return SaturatingAdd(std::numeric_limits<T>::max(), 1, args...);
  return SaturatingAdd(XY, Z, args...);
}

(p.s. if you like - i can send a small patch with this addition to MathExtras)

[oontvoo]

llvm/lib/Object/MachOObjectFile.cpp

137

sure! Thanks!

llvm/lib/Object/MachOObjectFile.cpp

 #include "llvm/Support/Debug.h"
 #include "llvm/Support/Errc.h"
 #include "llvm/Support/Error.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Format.h"
 #include "llvm/Support/Host.h"
 #include "llvm/Support/LEB128.h"
+#include "llvm/Support/MathExtras.h"
 #include "llvm/Support/MemoryBufferRef.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/SwapByteOrder.h"
 #include "llvm/Support/raw_ostream.h"
 #include <algorithm>
 #include <cassert>
 #include <cstddef>
 #include <cstdint>
static T getStruct(const MachOObjectFile &O, const char *P) {
   if (O.isLittleEndian() != sys::IsLittleEndianHost)
     MachO::swapStruct(Cmd);
   return Cmd;
 }
 
 template <typename T>
 static Expected<T> getStructOrErr(const MachOObjectFile &O, const char *P) {
   // Don't read before the beginning or past the end of the file
-  if (P < O.getData().begin() || P + sizeof(T) > O.getData().end())
+  bool Overflowed = false;
+  uintptr_t Ret = llvm::SaturatingAdd<uintptr_t>(reinterpret_cast<uintptr_t>(P),
+                                                 sizeof(T), &Overflowed);
+  if (P < O.getData().begin() || Overflowed ||
+      Ret > reinterpret_cast<uintptr_t>(O.getData().end())) {
     return malformedError("Structure read out-of-range");
+  }
 
   T Cmd;
   memcpy(&Cmd, P, sizeof(T));
   if (O.isLittleEndian() != sys::IsLittleEndianHost)
     MachO::swapStruct(Cmd);
   return Cmd;
 }
 

 static StringRef parseSegmentOrSectionName(const char *P) {
   if (P[15] == 0)
     // Null terminated.
     return P;
   // Not null terminated, so this is a 16 char string.
   return StringRef(P, 16);
 }
 
+static uintptr_t SaturatingAddHelper(uintptr_t Arg1, uintptr_t Arg2,
+                                     uintptr_t Arg3, bool *Overflowed) {
+  uintptr_t Ret = llvm::SaturatingAdd<uintptr_t>(Arg1, Arg2, Overflowed);
+  if (*Overflowed)
+    return Ret;
+  return llvm::SaturatingAdd<uintptr_t>(Ret, Arg3, Overflowed);
+}
+
 static unsigned getCPUType(const MachOObjectFile &O) {
   return O.getHeader().cputype;
 }
 
 static unsigned getCPUSubType(const MachOObjectFile &O) {
   return O.getHeader().cpusubtype;
 }
 
if (O.is64Bit()) {
     MachO::section_64 Sect = O.getSection64(Sec);
     return Sect.flags;
   }
   MachO::section Sect = O.getSection(Sec);
   return Sect.flags;
 }
 
 static Expected<MachOObjectFile::LoadCommandInfo>
-getLoadCommandInfo(const MachOObjectFile &Obj, const char *Ptr,
-                   uint32_t LoadCommandIndex) {
-  if (auto CmdOrErr = getStructOrErr<MachO::load_command>(Obj, Ptr)) {
-    if (CmdOrErr->cmdsize + Ptr > Obj.getData().end())
+getLoadCommandInfo(const MachOObjectFile &Obj, const char *BasePtr,
+                   uintptr_t Offset, uint32_t LoadCommandIndex,
+                   bool CheckLoadCmdEnd = false) {
+  bool Overflowed = false;
+  uintptr_t Ptr = llvm::SaturatingAdd<uintptr_t>(
+      reinterpret_cast<uintptr_t>(BasePtr), Offset, &Overflowed);
+  if (Overflowed)
+    return malformedError("load command " + Twine(LoadCommandIndex) +
+                          " starts at an overflowed address");
+  if (auto CmdOrErr = getStructOrErr<MachO::load_command>(
+          Obj, reinterpret_cast<const char *>(Ptr))) {
+    const auto HeaderSize = Obj.is64Bit() ? sizeof(MachO::mach_header_64)
+                                          : sizeof(MachO::mach_header);
+
+    if (Ptr >= reinterpret_cast<uintptr_t>(Obj.getData().end()))
       return malformedError("load command " + Twine(LoadCommandIndex) +
                             " extends past end of file");
+
+    if (CheckLoadCmdEnd) {
+      uintptr_t CmdEnd = llvm::SaturatingAdd<uintptr_t>(
+          Ptr, sizeof(MachO::load_command), &Overflowed);
+      if (Overflowed)
+        return malformedError("load command " + Twine(LoadCommandIndex) +
+                              "extends past uintptr_t type limit");
+      uintptr_t LoadCmdsEnd = SaturatingAddHelper(
+          reinterpret_cast<uintptr_t>(Obj.getData().begin()), HeaderSize,
+          Obj.getHeader().sizeofcmds, &Overflowed);
+      if (CmdEnd > LoadCmdsEnd)
+        return malformedError(
+            "load command " + Twine(LoadCommandIndex + 1) +
+            " extends past the end of all load commands in the file");
+    }
+
     if (CmdOrErr->cmdsize < 8)
       return malformedError("load command " + Twine(LoadCommandIndex) +
                             " with size less than 8 bytes");
-    return MachOObjectFile::LoadCommandInfo({Ptr, *CmdOrErr});
+    return MachOObjectFile::LoadCommandInfo(
+        {reinterpret_cast<const char *>(Ptr), *CmdOrErr});
   } else
     return CmdOrErr.takeError();
 }
 
 static Expected<MachOObjectFile::LoadCommandInfo>
 getFirstLoadCommandInfo(const MachOObjectFile &Obj) {
   unsigned HeaderSize = Obj.is64Bit() ? sizeof(MachO::mach_header_64)
                                       : sizeof(MachO::mach_header);
-  if (sizeof(MachO::load_command) > Obj.getHeader().sizeofcmds)
-    return malformedError("load command 0 extends past the end all load "
-                          "commands in the file");
-  return getLoadCommandInfo(Obj, getPtr(Obj, HeaderSize), 0);
+  return getLoadCommandInfo(Obj, getPtr(Obj, HeaderSize), 0, 0);
 }
 
 static Expected<MachOObjectFile::LoadCommandInfo>
 getNextLoadCommandInfo(const MachOObjectFile &Obj, uint32_t LoadCommandIndex,
                        const MachOObjectFile::LoadCommandInfo &L) {
-  unsigned HeaderSize = Obj.is64Bit() ? sizeof(MachO::mach_header_64)
-                                      : sizeof(MachO::mach_header);
-  if (L.Ptr + L.C.cmdsize + sizeof(MachO::load_command) >
-      Obj.getData().data() + HeaderSize + Obj.getHeader().sizeofcmds)
-    return malformedError("load command " + Twine(LoadCommandIndex + 1) +
-                          " extends past the end all load commands in the file");
-  return getLoadCommandInfo(Obj, L.Ptr + L.C.cmdsize, LoadCommandIndex + 1);
+  return getLoadCommandInfo(Obj, L.Ptr, L.C.cmdsize, LoadCommandIndex + 1,
+                            true);
 }
 
 template <typename T>
 static void parseHeader(const MachOObjectFile &Obj, T &Header,
                         Error &Err) {
   if (sizeof(T) > Obj.getData().size()) {
     Err = malformedError("the mach header extends past the end of the "
                          "file");
     return;
   }
MachOObjectFile::MachOObjectFile(MemoryBufferRef Object, bool IsLittleEndian,
   } else {
     parseHeader(*this, Header, Err);
     SizeOfHeaders = sizeof(MachO::mach_header);
     cputype = Header.cputype;
   }
   if (Err)
     return;
   SizeOfHeaders += getHeader().sizeofcmds;
-  if (getData().data() + SizeOfHeaders > getData().end()) {
+  bool Overflowed = false;
+  uintptr_t LoadCmdsStart = llvm::SaturatingAdd<uintptr_t>(
+      SizeOfHeaders, reinterpret_cast<uintptr_t>(getData().data()),
+      &Overflowed);
+  if (Overflowed) {
+    Err = malformedError("load commands extends past uintptr_t type limit");
+    return;
+  }
+  if (LoadCmdsStart > reinterpret_cast<uintptr_t>(getData().end())) {
     Err = malformedError("load commands extend past the end of the file");
     return;
   }
   if (UniversalCputype != 0 && cputype != UniversalCputype) {
     Err = malformedError("universal header architecture: " +
                          Twine(UniversalIndex) + "'s cputype does not match "
                          "object file's mach header");
     return;