Build ELF linux/dump_syms everywhere (also for macOS and for Windows with MinGW) by illwieckz · Pull Request #20 · DaemonEngine/breakpad

illwieckz · 2026-07-14T16:40:49Z

Build ELF linux/dump_syms everywhere (also for macOS and for Windows with MinGW).

It builds and works for both Windows and macOS.

How to build:

autoreconf -fi && ./configure --host=x86_64-w64-mingw32 && make -j"$(nproc)"

You then get:

src/tools/linux/dump_syms/dump_syms.exe

It makes possible to symbolize NaCl nexe binaries on Windows and on macOS.

Disclaimer: I was assisted by ChatGPT when I was writing and debugging the posix_compat.h file, this was done through a discussion with the bot, by explaining the various shortcomings I was facing in my trial and error attempts. For example when I pushed the branch at first there was still a mistake and the tool didn't do anything despite building successfully. After extensive debugging (I even found a bug unrelated to this patch, I'll report) ChatGPT identified the mistake in the mmap() compatibility wrapper. The bot also suggested API conversion advises and short snippets I had to adapt in the making of those wrappers. The chat bot didn't have access to the repository (neither read, neither write) and only answered my questions based on my descriptions and its general knowledge of upstream breakpad and APIs from Linux, Windows and MinGW. Once the patch was ready I pasted the diff in the bot conversation and the chat bot identified some shortcomings like redundant things I cleaned-up.

…with MinGW)

illwieckz · 2026-07-14T17:46:04Z

When running it on Wine I get:

Failed to write symbol file.

So maybe it fails on some /dev/stdout equivalent for Windows.

illwieckz · 2026-07-14T19:28:07Z

It now works. It was a mistake with flags in the mmap() wrapper.

When generating symbol files with the Linux native dump_syms or the Windows dump_syms_elf built with MinGW and running on Linux, they are the same, except the end line (DOS or Unix style). When converting symbol files with dos2unix the files are identical bit to bit. I tested by running the symbolizers on the Linux daemon engine client binary.

illwieckz · 2026-07-14T23:33:02Z

So, after having rebased the code over:

dwarf2reader: support NaCl Saigo CFA pseudo-register -2 #19

I symbolized a cgame-amd64.nexe built with Saigo using the MinGW-build dump_syms_elf running on Wine and it produced the exact same symbol file as the native Linux dump_syms (once the file produced by the Windows binary is converted to linux line ending it's the same).

illwieckz · 2026-07-16T10:51:39Z

Actually merging this would make far easier to solve some merge conflicts from:

Merging upstream? (2026-07) #21

Because then we can revert 4678e20:

    Disable DwpReader for MinGW
    
    elf_reader.cc uses mmap(), which isn't available
    
    Perhaps we could change to using MmapWrapper(), which I've already fixed for
    Windows, but this code isn't doing anything for us anyhow

So even if “this code isn't doing anything for us anyhow”, the specialization of the code is generating annoying conflicts.

illwieckz · 2026-07-16T10:57:25Z

Anyway, since that's just a copy of the Linux dump_syms (without stabs), we may just enable stabs for it and don't duplicate the Linux dump_syms. Then we would have to rename Windows dump_syms to something like dump_syms_pecoff because dump_syms would be the ELF one. We may also rename everything to avoid any confusion (only rename the build artifact to not introduce merge conflicts):

dump_syms_elf
dump_syms_macho
dump_syms_pdb
dump_syms_dwarf

illwieckz · 2026-07-16T11:45:02Z

MinGW is now building linux_dump/syms/dump_syms.exe with nothing disabled compared to the Linux build (so that includes stab support) The _SOURCES is duplicated because of the way the Makefile.amis written. Upstream breakpad reworked the Makefile.am in a way we will be able to deduplicate that later.

illwieckz · 2026-07-16T13:16:39Z

Today I learnt that we cannot even build Windows' dump_syms with MinGW…

illwieckz · 2026-07-16T17:20:39Z

It now builds on macOS too.

It builds and runs. The symbol files may have some difference, but that is not a bug:

1599c1599
< FUNC cc0 300 0 Cvar::Cvar<float>::Cvar(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, int, float)
---
> FUNC cc0 300 0 Cvar::Cvar<float>::Cvar(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, int, float)
…
592676c592676
< PUBLIC 660140 0 std::exception_ptr std::make_exception_ptr[abi:ne210000]<std::__1::future_error>(std::__1::future_error)::'lambda'(void*)::__invoke(void*)
---
> PUBLIC 660140 0 std::make_exception_ptr[abi:ne210000]<std::__1::future_error>(std::__1::future_error)::{lambda(void*)#1}::__invoke(void*)

The _SOURCES is duplicated because of the way the Makefile.amis written. Upstream breakpad reworked the Makefile.am in a way we will be able to deduplicate that later.

Since the tool now builds everywhere, I could deduplicate this easily.

illwieckz · 2026-07-16T22:32:49Z

In fact, now that I did nice compatibility layers, adding support for FreeBSD was straightforward… The ELF dump_syms is now very portable, not only it now runs on many systems, but adding a new system would be very easy.

slipher · 2026-07-22T01:40:06Z

I checked the master branch on Mac and it already builds a working dump_syms for ELF. So a port is only needed for Windows.

illwieckz · 2026-07-22T02:48:28Z

Do you mean our master branch or Google upstream?

slipher · 2026-07-22T04:11:57Z

Never mind, I confused myself by using the release mode of crash_test.py which doesn't dump syms. That only proves that minidump_stackwalk is working. But the good news is that NaCl crash dumps work perfectly
under Rosetta emulation!

I will try to build this branch and use the script's non-release mode to test it.

slipher · 2026-07-22T04:23:01Z

It works with crash_test.py (on Mac). Yay!

slipher · 2026-07-22T04:48:15Z

+static inline char* breakpad_realpath(const char* path,
+                                      char* resolved)
+{
+  DWORD len = GetFullPathNameA(path, MAX_PATH, resolved, nullptr);


The size constant is supposed to be PATH_MAX.

slipher · 2026-07-22T05:06:08Z

+  return resolved;
+}
+
+#ifndef realpath


Is there some reason to assume this function would be defined as a macro when it is present?

slipher · 2026-07-22T05:08:14Z

 Configuration ?= Debug
 all-local:
-	cd $(top_srcdir)/src ; ./tools/gyp/gyp --no-circular-check --no-parallel --depth=. -f make client/windows/breakpad_client.gyp
+	cd $(top_srcdir)/src ; GYP_DEFINES='OS=win' ./tools/gyp/gyp --no-circular-check --no-parallel --depth=. -f make client/windows/breakpad_client.gyp


Where does GYP_DEFINES come from?

$ grep -r GYP_DEFINES src/tools/gyp/ src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = \ src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: os.environ['GYP_DEFINES'] = ( src/tools/gyp/test/defines-escaping/gyptest-defines-escaping.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines/gyptest-define-override.py:os.environ['GYP_DEFINES'] = 'OS=fakeos' src/tools/gyp/test/defines/gyptest-defines-env-regyp.py: os.environ['GYP_DEFINES'] = 'value=50' src/tools/gyp/test/defines/gyptest-defines-env-regyp.py: os.environ['GYP_DEFINES'] = '' src/tools/gyp/test/defines/gyptest-defines-env-regyp.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines/gyptest-defines-env.py: os.environ['GYP_DEFINES'] = 'value=10' src/tools/gyp/test/defines/gyptest-defines-env.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines/gyptest-defines-env.py: os.environ['GYP_DEFINES'] = 'value=20' src/tools/gyp/test/defines/gyptest-defines-env.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines/gyptest-defines-env.py: os.environ['GYP_DEFINES'] = 'value=30' src/tools/gyp/test/defines/gyptest-defines-env.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/defines/gyptest-defines-env.py: os.environ['GYP_DEFINES'] = 'value=40' src/tools/gyp/test/defines/gyptest-defines-env.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/gyp-defines/gyptest-multiple-values.py:os.environ['GYP_DEFINES'] = 'key=value1 key=value2 key=value3' src/tools/gyp/test/gyp-defines/gyptest-multiple-values.py:os.environ['GYP_DEFINES'] = 'key=repeated_value key=value1 key=repeated_value' src/tools/gyp/test/gyp-defines/gyptest-regyp.py:os.environ['GYP_DEFINES'] = 'key=repeated_value key=value1 key=repeated_value' src/tools/gyp/test/lib/TestGyp.py: # Clear $GYP_DEFINES for the same reason. src/tools/gyp/test/lib/TestGyp.py: if 'GYP_DEFINES' in os.environ: src/tools/gyp/test/lib/TestGyp.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/msvs/external_builder/gyptest-all.py: os.environ['GYP_DEFINES'] = 'use_external_builder=1' src/tools/gyp/test/msvs/external_builder/gyptest-all.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/msvs/external_builder/gyptest-all.py: os.environ['GYP_DEFINES'] = 'use_external_builder=1' src/tools/gyp/test/msvs/external_builder/gyptest-all.py: del os.environ['GYP_DEFINES'] src/tools/gyp/test/variables/commands/gyptest-commands-ignore-env.py:os.environ['GYP_DEFINES'] = 'FOO=BAR'

Oh, damn, those are fetched with ./fetch-externals… We should really get rid of gyp!

slipher · 2026-07-22T05:15:26Z

 endif

+if MACOS_HOST
+AM_CPPFLAGS += -Wno-deprecated-declarations


Warnings should just be non-fatal rather than disabled completely. See #22 (this warning is already there on master).

Can replace that with -Wno-error=deprecated-declarations instead.

slipher · 2026-07-22T05:17:36Z

+  return -1;
+}
+
+#define sys_mmap mmap


The function could be named sys_mmap directly instead of having a function and a macro.

Yes but there is still original code calling mmap() without sys_ prefix. I preferred to keep the patch smaller and requiring less rewrite.

slipher · 2026-07-22T06:45:03Z

With the addition of #23, the full crash_test.py works on MinGW too.

illwieckz added 11 commits July 14, 2026 15:00

Add stab.h from Debian Trixie libc6-dev 2.41-12+deb13u3

1f1ce2b

Add compat/elf.h

ea41ecc

Add compat/mman.h

6f5fdb5

Add compat/link.h

cb101c6

Add compat/syscall.h

deae767

Add compat/inet.h

1b6a34a

Add compat/guiddef.h

b90b71c

Add compat/paths.h

dcfda26

Add compat/linux.h

9a1cb9f

Add macOS support to configure.ac

5e05321

Build ELF linux/dump_syms everywhere (also for macOS and for Windows …

afe3f3a

…with MinGW)

illwieckz mentioned this pull request Jul 14, 2026

Stop shipping NaCl IRT and build it alongside gamelogic instead? DaemonEngine/Daemon#1993

Open

illwieckz force-pushed the illwieckz/dump-syms-elf branch 4 times, most recently from 78ae084 to 05f7742 Compare July 14, 2026 19:22

illwieckz force-pushed the illwieckz/dump-syms-elf branch from 05f7742 to dffe3ab Compare July 14, 2026 20:59

illwieckz commented Jul 14, 2026

View reviewed changes

Comment thread src/common/windows/posix_compat.h Outdated

Base automatically changed from illwieckz/saigo-nacl to master July 14, 2026 23:26

illwieckz force-pushed the illwieckz/dump-syms-elf branch from dffe3ab to 0ba693b Compare July 14, 2026 23:26

illwieckz force-pushed the illwieckz/dump-syms-elf branch from 0ba693b to 6cd2cb2 Compare July 14, 2026 23:34

illwieckz marked this pull request as ready for review July 14, 2026 23:34

illwieckz force-pushed the illwieckz/dump-syms-elf branch from 6cd2cb2 to 48d9427 Compare July 16, 2026 11:43

illwieckz changed the title ~~Build a Windows dump_syms_elf with MinGW~~ Build the Linux dump_syms for Windows with MinGW Jul 16, 2026

illwieckz force-pushed the illwieckz/dump-syms-elf branch from 48d9427 to 5a6e226 Compare July 16, 2026 16:41

illwieckz changed the title ~~Build the Linux dump_syms for Windows with MinGW~~ Build ELF linux/dump_syms everywhere (also for macOS and for Windows with MinGW) Jul 16, 2026

illwieckz force-pushed the illwieckz/dump-syms-elf branch 2 times, most recently from d78f4ac to 8a9ff22 Compare July 16, 2026 17:20

illwieckz force-pushed the illwieckz/dump-syms-elf branch 9 times, most recently from 33c22d8 to f19bed4 Compare July 16, 2026 22:29

illwieckz force-pushed the illwieckz/dump-syms-elf branch from f19bed4 to afe3f3a Compare July 16, 2026 23:05

slipher reviewed Jul 22, 2026

View reviewed changes

Uh oh!

Conversation

illwieckz commented Jul 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

illwieckz commented Jul 14, 2026

Uh oh!

illwieckz commented Jul 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

illwieckz commented Jul 14, 2026

Uh oh!

illwieckz commented Jul 16, 2026

Uh oh!

illwieckz commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

illwieckz commented Jul 16, 2026

Uh oh!

illwieckz commented Jul 16, 2026

Uh oh!

illwieckz commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

illwieckz commented Jul 16, 2026

Uh oh!

slipher commented Jul 22, 2026

Uh oh!

illwieckz commented Jul 22, 2026

Uh oh!

slipher commented Jul 22, 2026

Uh oh!

slipher commented Jul 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

slipher commented Jul 22, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

illwieckz commented Jul 14, 2026 •

edited

Loading

illwieckz commented Jul 14, 2026 •

edited

Loading

illwieckz commented Jul 16, 2026 •

edited

Loading

illwieckz commented Jul 16, 2026 •

edited

Loading

slipher commented Jul 22, 2026 •

edited

Loading