https://github.com/eronnen created
https://github.com/llvm/llvm-project/pull/138416
* Changes the default synthetic symbol names to contain their file address
This is a new PR after the first PR (#137512) was reverted because it didn't
update the way unnamed symbols were searched in the symbol table, which relied
on the index being in the name.
This time also added extra test to make sure the symbol is found as expected
>From 3ec9e1962940799ab291aaa8455e6f76da02af0f Mon Sep 17 00:00:00 2001
From: Ely Ronnen <elyron...@gmail.com>
Date: Sun, 27 Apr 2025 13:48:45 +0200
Subject: [PATCH 1/3] Change ___lldb_unnamed_symbol generated names to have the
file address
---
lldb/source/Symbol/Symbol.cpp | 4 +++-
lldb/test/Shell/ObjectFile/ELF/eh_frame-symbols.yaml | 4 ++--
.../test/Shell/SymbolFile/Breakpad/symtab-sorted-by-size.test | 2 +-
lldb/test/Shell/SymbolFile/Breakpad/symtab.test | 2 +-
4 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/lldb/source/Symbol/Symbol.cpp b/lldb/source/Symbol/Symbol.cpp
index 4828de4fdfa37..da74707c75e13 100644
--- a/lldb/source/Symbol/Symbol.cpp
+++ b/lldb/source/Symbol/Symbol.cpp
@@ -639,7 +639,9 @@ void Symbol::SynthesizeNameIfNeeded() const {
// breakpoints on them.
llvm::SmallString<256> name;
llvm::raw_svector_ostream os(name);
- os << GetSyntheticSymbolPrefix() << GetID();
+ os << GetSyntheticSymbolPrefix() << "_"
+ << llvm::format_hex_no_prefix(
+ m_addr_range.GetBaseAddress().GetFileAddress(), 0);
m_mangled.SetDemangledName(ConstString(os.str()));
}
}
diff --git a/lldb/test/Shell/ObjectFile/ELF/eh_frame-symbols.yaml
b/lldb/test/Shell/ObjectFile/ELF/eh_frame-symbols.yaml
index 0dcc9fb76bd4f..709c37e79d878 100644
--- a/lldb/test/Shell/ObjectFile/ELF/eh_frame-symbols.yaml
+++ b/lldb/test/Shell/ObjectFile/ELF/eh_frame-symbols.yaml
@@ -3,8 +3,8 @@
# CHECK: Index UserID DSX Type File Address/Value Load Address
Size Flags Name
# CHECK: [ 0] 1 SourceFile 0x0000000000000000
0x0000000000000000 0x00000004 -
-# CHECK: [ 1] 2 SX Code 0x0000000000201180
0x0000000000000010 0x00000000 ___lldb_unnamed_symbol{{[0-9]*}}
-# CHECK: [ 2] 3 SX Code 0x0000000000201190
0x0000000000000006 0x00000000 ___lldb_unnamed_symbol{{[0-9]*}}
+# CHECK: [ 1] 2 SX Code 0x0000000000201180
0x0000000000000010 0x00000000 ___lldb_unnamed_symbol_{{[0-9a-f]*}}
+# CHECK: [ 2] 3 SX Code 0x0000000000201190
0x0000000000000006 0x00000000 ___lldb_unnamed_symbol_{{[0-9a-f]*}}
--- !ELF
FileHeader:
diff --git a/lldb/test/Shell/SymbolFile/Breakpad/symtab-sorted-by-size.test
b/lldb/test/Shell/SymbolFile/Breakpad/symtab-sorted-by-size.test
index 98052ea20bedd..00e04eb39a98e 100644
--- a/lldb/test/Shell/SymbolFile/Breakpad/symtab-sorted-by-size.test
+++ b/lldb/test/Shell/SymbolFile/Breakpad/symtab-sorted-by-size.test
@@ -3,7 +3,7 @@
# RUN: -s %s | FileCheck %s
# CHECK: num_symbols = 4 (sorted by size):
-# CHECK: [ 0] 0 SX Code 0x0000000000400000
0x00000000000000b0 0x00000000 ___lldb_unnamed_symbol0
+# CHECK: [ 0] 0 SX Code 0x0000000000400000
0x00000000000000b0 0x00000000 ___lldb_unnamed_symbol_400000
# CHECK: [ 1] 0 X Code 0x00000000004000d0
0x0000000000000022 0x00000000 _start
# CHECK: [ 2] 0 X Code 0x00000000004000b0
0x0000000000000010 0x00000000 f1
# CHECK: [ 3] 0 X Code 0x00000000004000c0
0x0000000000000010 0x00000000 f2
diff --git a/lldb/test/Shell/SymbolFile/Breakpad/symtab.test
b/lldb/test/Shell/SymbolFile/Breakpad/symtab.test
index ef41bb3bea955..a32eb5808426f 100644
--- a/lldb/test/Shell/SymbolFile/Breakpad/symtab.test
+++ b/lldb/test/Shell/SymbolFile/Breakpad/symtab.test
@@ -5,7 +5,7 @@
# CHECK-LABEL: (lldb) image dump symtab symtab.out
# CHECK: Symtab, file = {{.*}}symtab.out, num_symbols = 4:
# CHECK: Index UserID DSX Type File Address/Value Load Address
Size Flags Name
-# CHECK: [ 0] 0 SX Code 0x0000000000400000
0x00000000000000b0 0x00000000 ___lldb_unnamed_symbol{{[0-9]*}}
+# CHECK: [ 0] 0 SX Code 0x0000000000400000
0x00000000000000b0 0x00000000 ___lldb_unnamed_symbol_{{[0-9a-f]*}}
# CHECK: [ 1] 0 X Code 0x00000000004000b0
0x0000000000000010 0x00000000 f1
# CHECK: [ 2] 0 X Code 0x00000000004000c0
0x0000000000000010 0x00000000 f2
# CHECK: [ 3] 0 X Code 0x00000000004000d0
0x0000000000000022 0x00000000 _start
>From 3798aff067f8facbb386a3b3282975b23680447f Mon Sep 17 00:00:00 2001
From: Ely Ronnen <elyron...@gmail.com>
Date: Sat, 3 May 2025 18:55:50 +0200
Subject: [PATCH 2/3] add test for looking up unnamed symbols
---
.../python_api/unnamed_symbol_lookup/Makefile | 12 ++++++
.../TestUnnamedSymbolLookup.py | 40 +++++++++++++++++++
.../python_api/unnamed_symbol_lookup/main.c | 10 +++++
3 files changed, 62 insertions(+)
create mode 100644 lldb/test/API/python_api/unnamed_symbol_lookup/Makefile
create mode 100644
lldb/test/API/python_api/unnamed_symbol_lookup/TestUnnamedSymbolLookup.py
create mode 100644 lldb/test/API/python_api/unnamed_symbol_lookup/main.c
diff --git a/lldb/test/API/python_api/unnamed_symbol_lookup/Makefile
b/lldb/test/API/python_api/unnamed_symbol_lookup/Makefile
new file mode 100644
index 0000000000000..9ba76898b61ba
--- /dev/null
+++ b/lldb/test/API/python_api/unnamed_symbol_lookup/Makefile
@@ -0,0 +1,12 @@
+C_SOURCES := main.c
+
+include Makefile.rules
+
+all: a.out.stripped
+
+a.out.stripped:
+ $(STRIP) --keep-symbol=main -o a.out.stripped a.out
+
+ifneq "$(CODESIGN)" ""
+ $(CODESIGN) -fs - a.out.stripped
+endif
diff --git
a/lldb/test/API/python_api/unnamed_symbol_lookup/TestUnnamedSymbolLookup.py
b/lldb/test/API/python_api/unnamed_symbol_lookup/TestUnnamedSymbolLookup.py
new file mode 100644
index 0000000000000..09d43a34c7e30
--- /dev/null
+++ b/lldb/test/API/python_api/unnamed_symbol_lookup/TestUnnamedSymbolLookup.py
@@ -0,0 +1,40 @@
+"""
+Test lookup unnamed symbols.
+"""
+
+
+import lldb
+from lldbsuite.test.decorators import *
+from lldbsuite.test.lldbtest import *
+from lldbsuite.test import lldbutil
+
+
+class TestUnnamedSymbolLookup(TestBase):
+ def test_unnamed_symbol_lookup(self):
+ """Test looking up unnamed symbol synthetic name"""
+ self.build()
+ (target, process, thread, bkpt) = lldbutil.run_to_name_breakpoint(
+ self, "main", exe_name="a.out.stripped"
+ )
+
+ main_frame = thread.GetFrameAtIndex(0)
+
+ # Step until reaching the unnamed symbol called from main
+ for _ in range(100):
+ thread.StepInto()
+ if thread.GetFrameAtIndex(0) != main_frame:
+ break
+
+ thread.StepInto()
+
+ self.assertEqual(
+ main_frame, thread.GetFrameAtIndex(1), "Expected to be called from
main"
+ )
+ symbol = thread.GetFrameAtIndex(0).GetSymbol()
+ self.assertIsNotNone(symbol, "unnamed symbol called from main not
reached")
+ self.assertTrue(symbol.name.startswith("___lldb_unnamed_symbol"))
+
+ exe_module = symbol.GetStartAddress().GetModule()
+ found_symbols = exe_module.FindSymbols(symbol.name)
+ self.assertIsNotNone(found_symbols)
+ self.assertEqual(found_symbols.GetSize(), 1)
diff --git a/lldb/test/API/python_api/unnamed_symbol_lookup/main.c
b/lldb/test/API/python_api/unnamed_symbol_lookup/main.c
new file mode 100644
index 0000000000000..1c9ce8e0e6697
--- /dev/null
+++ b/lldb/test/API/python_api/unnamed_symbol_lookup/main.c
@@ -0,0 +1,10 @@
+__attribute__((nodebug)) int stripped_function(int val)
+{
+ return val * val;
+}
+
+int main (void)
+{
+ stripped_function(10);
+ return 0;
+}
>From 1ef05ed18e225388019d6008934f8d51fb8012bc Mon Sep 17 00:00:00 2001
From: Ely Ronnen <elyron...@gmail.com>
Date: Sat, 3 May 2025 19:09:46 +0200
Subject: [PATCH 3/3] correctly finding unnamed symbols in symbol table with
file address
---
lldb/include/lldb/Symbol/Symbol.h | 2 +-
lldb/source/Symbol/Symbol.cpp | 2 +-
lldb/source/Symbol/Symtab.cpp | 11 ++++++-----
3 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/lldb/include/lldb/Symbol/Symbol.h
b/lldb/include/lldb/Symbol/Symbol.h
index e05c845a69f3e..688c8a5931feb 100644
--- a/lldb/include/lldb/Symbol/Symbol.h
+++ b/lldb/include/lldb/Symbol/Symbol.h
@@ -258,7 +258,7 @@ class Symbol : public SymbolContextScope {
bool ContainsFileAddress(lldb::addr_t file_addr) const;
static llvm::StringRef GetSyntheticSymbolPrefix() {
- return "___lldb_unnamed_symbol";
+ return "___lldb_unnamed_symbol_";
}
/// Decode a serialized version of this object from data.
diff --git a/lldb/source/Symbol/Symbol.cpp b/lldb/source/Symbol/Symbol.cpp
index da74707c75e13..d6689a647062a 100644
--- a/lldb/source/Symbol/Symbol.cpp
+++ b/lldb/source/Symbol/Symbol.cpp
@@ -639,7 +639,7 @@ void Symbol::SynthesizeNameIfNeeded() const {
// breakpoints on them.
llvm::SmallString<256> name;
llvm::raw_svector_ostream os(name);
- os << GetSyntheticSymbolPrefix() << "_"
+ os << GetSyntheticSymbolPrefix()
<< llvm::format_hex_no_prefix(
m_addr_range.GetBaseAddress().GetFileAddress(), 0);
m_mangled.SetDemangledName(ConstString(os.str()));
diff --git a/lldb/source/Symbol/Symtab.cpp b/lldb/source/Symbol/Symtab.cpp
index 9aee5d3e813d8..52545f22322d2 100644
--- a/lldb/source/Symbol/Symtab.cpp
+++ b/lldb/source/Symbol/Symtab.cpp
@@ -654,7 +654,7 @@ uint32_t Symtab::GetNameIndexes(ConstString symbol_name,
if (count)
return count;
// Synthetic symbol names are not added to the name indexes, but they start
- // with a prefix and end with a the symbol UserID. This allows users to find
+ // with a prefix and end with the symbol file address. This allows users to
find
// these symbols without having to add them to the name indexes. These
// queries will not happen very often since the names don't mean anything, so
// performance is not paramount in this case.
@@ -663,11 +663,12 @@ uint32_t Symtab::GetNameIndexes(ConstString symbol_name,
if (!name.consume_front(Symbol::GetSyntheticSymbolPrefix()))
return 0; // Not a synthetic symbol name
- // Extract the user ID from the symbol name
- unsigned long long uid = 0;
- if (getAsUnsignedInteger(name, /*Radix=*/10, uid))
+ // Extract the file address from the symbol name
+ unsigned long long file_address = 0;
+ if (getAsUnsignedInteger(name, /*Radix=*/16, file_address))
return 0; // Failed to extract the user ID as an integer
- Symbol *symbol = FindSymbolByID(uid);
+
+ Symbol *symbol = FindSymbolAtFileAddress(static_cast<addr_t>(file_address));
if (symbol == nullptr)
return 0;
const uint32_t symbol_idx = GetIndexForSymbol(symbol);
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
