[lldb][MachO] Fix inspection of global variables that start with 'O'

[Michael137]

On Darwin C-symbols are prefixed with a '_'. The LLDB Macho-O parses handles Objective-C metadata symbols starting with '_OBJC' specially. Previously global symbols starting with a '_O' prefix were lost because of incorrectly scoped if-guards. This patch removes those checks.

There is more cleanup that can be done in this file because there's a bunch of duplicated checks for these ObjC symbols. I decided to leave that for an NFC follow-up.

Depends on #161520

rdar://158159242

[llvmbot]

@llvm/pr-subscribers-lldb

Author: Michael Buch (Michael137)

Changes

On Darwin C-symbols are prefixed with a '_'. The LLDB Macho-O parses handles Objective-C metadata symbols starting with '_OBJC' specially. Previously global symbols starting with a '_O' prefix were lost because of incorrectly scoped if-guards. This patch removes those checks.

There is more cleanup that can be done in this file because there's a bunch of duplicated checks for these ObjC symbols. I decided to leave that for an NFC follow-up.

Depends on #161520

rdar://158159242

---

Full diff: https://github.com/llvm/llvm-project/pull/161521.diff

3 Files Affected:

• (modified) lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp (+42-48)
• (added) lldb/test/Shell/Expr/TestGlobalSymbolObjCConflict.c (+33)
• (modified) lldb/test/Shell/lit.cfg.py (+1-1)

diff --git a/lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp b/lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
index fada1fda2b4bc..91c93be1b8cfd 100644
--- a/lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
+++ b/lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
@@ -2805,32 +2805,29 @@ void ObjectFileMachO::ParseSymtab(Symtab &symtab) {
                         is_gsym = true;
                         sym[sym_idx].SetExternal(true);
 
-                        if (symbol_name && symbol_name[0] == '_' &&
-                            symbol_name[1] == 'O') {
-                          llvm::StringRef symbol_name_ref(symbol_name);
-                          if (symbol_name_ref.starts_with(
-                                  g_objc_v2_prefix_class)) {
-                            symbol_name_non_abi_mangled = symbol_name + 1;
-                            symbol_name =
-                                symbol_name + g_objc_v2_prefix_class.size();
-                            type = eSymbolTypeObjCClass;
-                            demangled_is_synthesized = true;
-
-                          } else if (symbol_name_ref.starts_with(
-                                         g_objc_v2_prefix_metaclass)) {
-                            symbol_name_non_abi_mangled = symbol_name + 1;
-                            symbol_name =
-                                symbol_name + g_objc_v2_prefix_metaclass.size();
-                            type = eSymbolTypeObjCMetaClass;
-                            demangled_is_synthesized = true;
-                          } else if (symbol_name_ref.starts_with(
-                                         g_objc_v2_prefix_ivar)) {
-                            symbol_name_non_abi_mangled = symbol_name + 1;
-                            symbol_name =
-                                symbol_name + g_objc_v2_prefix_ivar.size();
-                            type = eSymbolTypeObjCIVar;
-                            demangled_is_synthesized = true;
-                          }
+                        llvm::StringRef symbol_name_ref(symbol_name);
+                        if (symbol_name_ref.starts_with(
+                                g_objc_v2_prefix_class)) {
+                          symbol_name_non_abi_mangled = symbol_name + 1;
+                          symbol_name =
+                              symbol_name + g_objc_v2_prefix_class.size();
+                          type = eSymbolTypeObjCClass;
+                          demangled_is_synthesized = true;
+
+                        } else if (symbol_name_ref.starts_with(
+                                       g_objc_v2_prefix_metaclass)) {
+                          symbol_name_non_abi_mangled = symbol_name + 1;
+                          symbol_name =
+                              symbol_name + g_objc_v2_prefix_metaclass.size();
+                          type = eSymbolTypeObjCMetaClass;
+                          demangled_is_synthesized = true;
+                        } else if (symbol_name_ref.starts_with(
+                                       g_objc_v2_prefix_ivar)) {
+                          symbol_name_non_abi_mangled = symbol_name + 1;
+                          symbol_name =
+                              symbol_name + g_objc_v2_prefix_ivar.size();
+                          type = eSymbolTypeObjCIVar;
+                          demangled_is_synthesized = true;
                         } else {
                           if (nlist.n_value != 0)
                             symbol_section = section_info.GetSection(
@@ -3652,7 +3649,7 @@ void ObjectFileMachO::ParseSymtab(Symtab &symtab) {
 
       if (is_debug) {
         switch (nlist.n_type) {
-        case N_GSYM:
+        case N_GSYM: {
           // global symbol: name,,NO_SECT,type,0
           // Sometimes the N_GSYM value contains the address.
 
@@ -3668,33 +3665,30 @@ void ObjectFileMachO::ParseSymtab(Symtab &symtab) {
           is_gsym = true;
           sym[sym_idx].SetExternal(true);
 
-          if (symbol_name && symbol_name[0] == '_' && symbol_name[1] == 'O') {
-            llvm::StringRef symbol_name_ref(symbol_name);
-            if (symbol_name_ref.starts_with(g_objc_v2_prefix_class)) {
-              symbol_name_non_abi_mangled = symbol_name + 1;
-              symbol_name = symbol_name + g_objc_v2_prefix_class.size();
-              type = eSymbolTypeObjCClass;
-              demangled_is_synthesized = true;
-
-            } else if (symbol_name_ref.starts_with(
-                           g_objc_v2_prefix_metaclass)) {
-              symbol_name_non_abi_mangled = symbol_name + 1;
-              symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size();
-              type = eSymbolTypeObjCMetaClass;
-              demangled_is_synthesized = true;
-            } else if (symbol_name_ref.starts_with(g_objc_v2_prefix_ivar)) {
-              symbol_name_non_abi_mangled = symbol_name + 1;
-              symbol_name = symbol_name + g_objc_v2_prefix_ivar.size();
-              type = eSymbolTypeObjCIVar;
-              demangled_is_synthesized = true;
-            }
+          llvm::StringRef symbol_name_ref(symbol_name);
+          if (symbol_name_ref.starts_with(g_objc_v2_prefix_class)) {
+            symbol_name_non_abi_mangled = symbol_name + 1;
+            symbol_name = symbol_name + g_objc_v2_prefix_class.size();
+            type = eSymbolTypeObjCClass;
+            demangled_is_synthesized = true;
+
+          } else if (symbol_name_ref.starts_with(g_objc_v2_prefix_metaclass)) {
+            symbol_name_non_abi_mangled = symbol_name + 1;
+            symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size();
+            type = eSymbolTypeObjCMetaClass;
+            demangled_is_synthesized = true;
+          } else if (symbol_name_ref.starts_with(g_objc_v2_prefix_ivar)) {
+            symbol_name_non_abi_mangled = symbol_name + 1;
+            symbol_name = symbol_name + g_objc_v2_prefix_ivar.size();
+            type = eSymbolTypeObjCIVar;
+            demangled_is_synthesized = true;
           } else {
             if (nlist.n_value != 0)
               symbol_section =
                   section_info.GetSection(nlist.n_sect, nlist.n_value);
             type = eSymbolTypeData;
           }
-          break;
+        } break;
 
         case N_FNAME:
           // procedure name (f77 kludge): name,,NO_SECT,0,0
diff --git a/lldb/test/Shell/Expr/TestGlobalSymbolObjCConflict.c b/lldb/test/Shell/Expr/TestGlobalSymbolObjCConflict.c
new file mode 100644
index 0000000000000..da7263648776f
--- /dev/null
+++ b/lldb/test/Shell/Expr/TestGlobalSymbolObjCConflict.c
@@ -0,0 +1,33 @@
+// Tests that LLDB correctly parses global symbols
+// starting with 'O'. On some platforms (e.g., Darwin)
+// C-symbols are prefixed with a '_'. The LLDB Macho-O
+// parses handles Objective-C metadata symbols starting
+// with '_OBJC' specially. This test ensures that we don't
+// lose track of regular global symbols with a '_O' prefix
+// in this.
+
+// RUN: %clang_host -c -g -fno-common %s -o %t.o
+// RUN: %clang_host %t.o -o %t.out
+// RUN: %lldb -b -x %t.out \
+// RUN:       -o "b 17" \
+// RUN:       -o "run" \
+// RUN:       -o "p OglobalVar" \
+// RUN:       -o "p Oabc" | FileCheck %s
+
+typedef struct {
+  int a;
+} Oabc_t;
+
+Oabc_t Oabc;
+int OglobalVar;
+
+int main(int argc, const char *argv[]) {
+  Oabc.a = 15;
+  OglobalVar = 10;
+  return OglobalVar + Oabc.a;
+}
+
+// CHECK: (lldb) p OglobalVar
+// CHECK: (int) 10
+// CHECK: (lldb) p Oabc
+// CHECK: (Oabc_t) (a = 15)
diff --git a/lldb/test/Shell/lit.cfg.py b/lldb/test/Shell/lit.cfg.py
index 505847fb763e0..cdc0cfe51f7c6 100644
--- a/lldb/test/Shell/lit.cfg.py
+++ b/lldb/test/Shell/lit.cfg.py
@@ -33,7 +33,7 @@
 
 # suffixes: A list of file extensions to treat as test files. This is overriden
 # by individual lit.local.cfg files in the test subdirectories.
-config.suffixes = [".test", ".cpp", ".s", ".m", ".ll"]
+config.suffixes = [".test", ".cpp", ".s", ".m", ".ll", ".c"]
 
 # excludes: A list of directories to exclude from the testsuite. The 'Inputs'
 # subdirectories contain auxiliary inputs for various tests in their parent