Another additional error check for invalid Mach-O files for the

obsolete load commands.

Again the philosophy of the error checking in libObject for
Mach-O files, the idea behind the checking is that we never
will return a Mach-O file out of libObject that contains unknown
things the library code can’t operate on.  So known obsolete
load commands will cause a hard error.

Also to make things clear I have added comments to the
values and structures in Support/Mach-O.h and
Support/MachO.def as to what is obsolete.

As noted in a TODO in the code, there may need to be a
non-default mode to allow some unknown values for well
structured Mach-O files with things like unknown load
load commands.  So things like using an old lldb on a newer
Mach-O file could still provide some limited functionality.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@285342 91177308-0d34-0410-b5e6-96231b3b80d8

include/llvm/Support/MachO.def

@@ -15,27 +15,37 @@
 
 HANDLE_LOAD_COMMAND(LC_SEGMENT, 0x00000001u, segment_command)
 HANDLE_LOAD_COMMAND(LC_SYMTAB, 0x00000002u, symtab_command)
+// LC_SYMSEG is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_SYMSEG, 0x00000003u, symseg_command)
 HANDLE_LOAD_COMMAND(LC_THREAD, 0x00000004u, thread_command)
 HANDLE_LOAD_COMMAND(LC_UNIXTHREAD, 0x00000005u, thread_command)
+// LC_LOADFVMLIB is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_LOADFVMLIB, 0x00000006u, fvmlib_command)
+// LC_IDFVMLIB is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_IDFVMLIB, 0x00000007u, fvmlib_command)
+// LC_IDENT is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_IDENT, 0x00000008u, ident_command)
+// LC_FVMFILE is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_FVMFILE, 0x00000009u, fvmfile_command)
+// LC_PREPAGE is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_PREPAGE, 0x0000000Au, load_command)
 HANDLE_LOAD_COMMAND(LC_DYSYMTAB, 0x0000000Bu, dysymtab_command)
 HANDLE_LOAD_COMMAND(LC_LOAD_DYLIB, 0x0000000Cu, dylib_command)
 HANDLE_LOAD_COMMAND(LC_ID_DYLIB, 0x0000000Du, dylib_command)
 HANDLE_LOAD_COMMAND(LC_LOAD_DYLINKER, 0x0000000Eu, dylinker_command)
 HANDLE_LOAD_COMMAND(LC_ID_DYLINKER, 0x0000000Fu, dylinker_command)
+// LC_PREBOUND_DYLIB is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_PREBOUND_DYLIB, 0x00000010u, prebound_dylib_command)
 HANDLE_LOAD_COMMAND(LC_ROUTINES, 0x00000011u, routines_command)
 HANDLE_LOAD_COMMAND(LC_SUB_FRAMEWORK, 0x00000012u, sub_framework_command)
 HANDLE_LOAD_COMMAND(LC_SUB_UMBRELLA, 0x00000013u, sub_umbrella_command)
 HANDLE_LOAD_COMMAND(LC_SUB_CLIENT, 0x00000014u, sub_client_command)
 HANDLE_LOAD_COMMAND(LC_SUB_LIBRARY, 0x00000015u, sub_library_command)
+// LC_TWOLEVEL_HINTS is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_TWOLEVEL_HINTS, 0x00000016u, twolevel_hints_command)
+// LC_PREBIND_CKSUM is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_PREBIND_CKSUM, 0x00000017u, prebind_cksum_command)
+// LC_LOAD_WEAK_DYLIB is obsolete and no longer supported.
 HANDLE_LOAD_COMMAND(LC_LOAD_WEAK_DYLIB, 0x80000018u, dylib_command)
 HANDLE_LOAD_COMMAND(LC_SEGMENT_64, 0x00000019u, segment_command_64)
 HANDLE_LOAD_COMMAND(LC_ROUTINES_64, 0x0000001Au, routines_command_64)

include/llvm/Support/MachO.h

@@ -578,6 +578,7 @@ namespace llvm {
       uint32_t header_addr;
     };
 
+    // The fvmlib_command is obsolete and no longer supported.
     struct fvmlib_command {
       uint32_t  cmd;
       uint32_t cmdsize;
@@ -621,6 +622,7 @@ namespace llvm {
       uint32_t sub_library;
     };
 
+    // The prebound_dylib_command is obsolete and no longer supported.
     struct prebound_dylib_command {
       uint32_t cmd;
       uint32_t cmdsize;
@@ -740,18 +742,21 @@ namespace llvm {
                flags:8;
     };
 
+    // The twolevel_hints_command is obsolete and no longer supported.
     struct twolevel_hints_command {
       uint32_t cmd;
       uint32_t cmdsize;
       uint32_t offset;
       uint32_t nhints;
     };
 
+    // The twolevel_hints_command is obsolete and no longer supported.
     struct twolevel_hint {
       uint32_t isub_image:8,
                itoc:24;
     };
 
+    // The prebind_cksum_command is obsolete and no longer supported.
     struct prebind_cksum_command {
       uint32_t cmd;
       uint32_t cmdsize;
@@ -835,18 +840,21 @@ namespace llvm {
       uint32_t count;
     };
 
+    // The symseg_command is obsolete and no longer supported.
     struct symseg_command {
       uint32_t cmd;
       uint32_t cmdsize;
       uint32_t offset;
       uint32_t size;
     };
 
+    // The ident_command is obsolete and no longer supported.
     struct ident_command {
       uint32_t cmd;
       uint32_t cmdsize;
     };
 
+    // The fvmfile_command is obsolete and no longer supported.
     struct fvmfile_command {
       uint32_t cmd;
       uint32_t cmdsize;
@@ -1268,19 +1276,22 @@ namespace llvm {
       sys::swapByteOrder(C);
     }
 
+    // The prebind_cksum_command is obsolete and no longer supported.
     inline void swapStruct(prebind_cksum_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
       sys::swapByteOrder(C.cksum);
     }
 
+    // The twolevel_hints_command is obsolete and no longer supported.
     inline void swapStruct(twolevel_hints_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
       sys::swapByteOrder(C.offset);
       sys::swapByteOrder(C.nhints);
     }
 
+    // The prebound_dylib_command is obsolete and no longer supported.
     inline void swapStruct(prebound_dylib_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
@@ -1289,20 +1300,23 @@ namespace llvm {
       sys::swapByteOrder(C.linked_modules);
     }
 
+    // The fvmfile_command is obsolete and no longer supported.
     inline void swapStruct(fvmfile_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
       sys::swapByteOrder(C.name);
       sys::swapByteOrder(C.header_addr);
     }
 
+    // The symseg_command is obsolete and no longer supported.
     inline void swapStruct(symseg_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
       sys::swapByteOrder(C.offset);
       sys::swapByteOrder(C.size);
     }
 
+    // The ident_command is obsolete and no longer supported.
     inline void swapStruct(ident_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);
@@ -1314,6 +1328,7 @@ namespace llvm {
       sys::swapByteOrder(C.header_addr);
     }
 
+    // The fvmlib_command is obsolete and no longer supported.
     inline void swapStruct(fvmlib_command &C) {
       sys::swapByteOrder(C.cmd);
       sys::swapByteOrder(C.cmdsize);

lib/Object/MachOObjectFile.cpp

@@ -912,6 +912,23 @@ static Error checkTwoLevelHintsCommand(const MachOObjectFile *Obj,
   return Error::success();
 }
 
+// Returns true if the libObject code does not support the load command and its
+// contents.  The cmd value it is treated as an unknown load command but with
+// an error message that says the cmd value is obsolete.
+static bool isLoadCommandObsolete(uint32_t cmd) {
+  if (cmd == MachO::LC_SYMSEG ||
+      cmd == MachO::LC_LOADFVMLIB ||
+      cmd == MachO::LC_IDFVMLIB ||
+      cmd == MachO::LC_IDENT ||
+      cmd == MachO::LC_FVMFILE ||
+      cmd == MachO::LC_PREPAGE ||
+      cmd == MachO::LC_PREBOUND_DYLIB ||
+      cmd == MachO::LC_TWOLEVEL_HINTS ||
+      cmd == MachO::LC_PREBIND_CKSUM)
+    return true;
+  return false;
+}
+
 Expected<std::unique_ptr<MachOObjectFile>>
 MachOObjectFile::create(MemoryBufferRef Object, bool IsLittleEndian,
                         bool Is64Bits, uint32_t UniversalCputype,
@@ -1250,11 +1267,20 @@ MachOObjectFile::MachOObjectFile(MemoryBufferRef Object, bool IsLittleEndian,
     } else if (Load.C.cmd == MachO::LC_THREAD) {
       if ((Err = checkThreadCommand(this, Load, I, "LC_THREAD")))
         return;
+    // Note: LC_TWOLEVEL_HINTS is really obsolete and is not supported.
     } else if (Load.C.cmd == MachO::LC_TWOLEVEL_HINTS) {
        if ((Err = checkTwoLevelHintsCommand(this, Load, I,
                                             &TwoLevelHintsLoadCmd)))
          return;
+    } else if (isLoadCommandObsolete(Load.C.cmd)) {
+      Err = malformedError("load command " + Twine(I) + " for cmd value of: " +
+                           Twine(Load.C.cmd) + " is obsolete and not "
+                           "supported");
+      return;
     }
+    // TODO: generate a error for unknown load commands by default.  But still
+    // need work out an approach to allow or not allow unknown values like this
+    // as an option for some uses like lldb.
     if (I < LoadCommandCount - 1) {
       if (auto LoadOrErr = getNextLoadCommandInfo(this, I, Load))
         Load = *LoadOrErr;

test/Object/macho-invalid.test

@@ -409,3 +409,27 @@ INVALID-TWOLEVELHINTS-OFFSET-HNINTS: macho-invalid-twolevelhints-offset-nhints':
 
 RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-fat_cputype 2>&1 | FileCheck -check-prefix INVALID-FAT-CPUTYPE %s
 INVALID-FAT-CPUTYPE: macho-invalid-fat_cputype': truncated or malformed object (universal header architecture: 0's cputype does not match object file's mach header)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-prebind_cksum-obsolete 2>&1 | FileCheck -check-prefix INVALID-PREBIND_CKSUM-OBSOLETE %s
+INVALID-PREBIND_CKSUM-OBSOLETE: macho-invalid-prebind_cksum-obsolete': truncated or malformed object (load command 0 for cmd value of: 23 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-symseg-obsolete 2>&1 | FileCheck -check-prefix INVALID-SYMSEG-OBSOLETE %s
+INVALID-SYMSEG-OBSOLETE: macho-invalid-symseg-obsolete': truncated or malformed object (load command 0 for cmd value of: 3 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-idfvmlib-obsolete 2>&1 | FileCheck -check-prefix INVALID-IDFVMLIB-OBSOLETE %s
+INVALID-IDFVMLIB-OBSOLETE: macho-invalid-idfvmlib-obsolete': truncated or malformed object (load command 0 for cmd value of: 7 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-loadfvmlib-obsolete 2>&1 | FileCheck -check-prefix INVALID-LOADFVMLIB-OBSOLETE %s
+INVALID-LOADFVMLIB-OBSOLETE: macho-invalid-loadfvmlib-obsolete': truncated or malformed object (load command 0 for cmd value of: 6 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-prebound_dylib-obsolete 2>&1 | FileCheck -check-prefix INVALID-PREBOUND_DYLIB-OBSOLETE %s
+INVALID-PREBOUND_DYLIB-OBSOLETE: macho-invalid-prebound_dylib-obsolete': truncated or malformed object (load command 0 for cmd value of: 16 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-ident-obsolete 2>&1 | FileCheck -check-prefix INVALID-IDENT-OBSOLETE %s
+INVALID-IDENT-OBSOLETE: macho-invalid-ident-obsolete': truncated or malformed object (load command 0 for cmd value of: 8 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-fvmfile-obsolete 2>&1 | FileCheck -check-prefix INVALID-FVMFILE-OBSOLETE %s
+INVALID-FVMFILE-OBSOLETE: macho-invalid-fvmfile-obsolete': truncated or malformed object (load command 0 for cmd value of: 9 is obsolete and not supported)
+
+RUN: not llvm-objdump -macho -private-headers %p/Inputs/macho-invalid-prepage-obsolete 2>&1 | FileCheck -check-prefix INVALID-PREPAGE-OBSOLETE %s
+INVALID-PREPAGE-OBSOLETE: macho-invalid-prepage-obsolete': truncated or malformed object (load command 0 for cmd value of: 10 is obsolete and not supported)