When the system supports it (as all Linux kernels ≥ 5.3 should), it’s
preferable to use `pidfd_open()` and `waitid()` to be notified of
child processes exiting or being signalled, rather than installing a
default `SIGCHLD` handler.
A default `SIGCHLD` handler is global, and can never interact well with
other code (from the application or other libraries) which also wants to
install a `SIGCHLD` handler.
This use of `pidfd_open()` is racy (the PID may be reused between
`g_child_watch_source_new()` being called and `pidfd_open()` being
called), so it doesn’t improve behaviour there. For that, we’d need
continuous use of pidfds throughout GLib, from fork/spawn time until
here. See #1866 for that.
The use of `waitid()` to get the process exit status could be expanded
in future to also work for stopped or continued processes (as per #175)
by adding `WSTOPPED | WCONTINUED` into the flags. That’s a behaviour
change which is outside the strict scope of adding pidfd support,
though.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Helps: #1866Fixes: #2216
On win32, WaitForSingleObject may return before the timeout is
dispatched, as it doesn't have a resolution higher than the system tick.
Wait for ~50ms before checking the callback changes.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Do not wakeup immediately for our own context wakeup poll registration.
g_main_context_add_poll_unlocked() will call g_wakeup_signal() to wake
up the loop waiting in poll(). Doing so during context creation will
create an extra wakeup for the first poll().
Skip the wakeup call if it is the wake_up_rec registration. No other
sources/caller should ever reach that condition.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
In the case strerror_r returns an error (both in the char* variant and
in the int variant) we should not try to proceed converting the message
and adding to the errors maps, as that's likely causing errors.
So, let's just return a null string in case this happens
We had gcc-only implementations for them while both can be used in all
the supported platforms we have.
So let's just provide generic definitions, while we keep the old ones
for both consistency and retro-compatibility.
There’s a precondition assertion on the function which checks that
`group_name != NULL`.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Coverity CID: #1474780
It’s failed on a CI machine
(https://gitlab.gnome.org/GNOME/glib/-/jobs/2006694) with log:
```
[unused] => pushed 100 threads onto the idle pool
[unused] stopping unused threads
[unused] waiting ONE second for threads to die
[unused] waiting ONE second for threads to die
[unused] waiting ONE second for threads to die
[unused] waiting ONE second for threads to die
[unused] waiting ONE second for threads to die
[unused] stopped idle threads, 9 remain
Bail out! GLib:ERROR:../glib/tests/thread-pool-slow.c:113:test_thread_stop_unused: assertion failed (g_thread_pool_get_num_unused_threads () == 0): (9 == 0)
```
I can’t reproduce this locally, but it’s possible that it failed because
a one second wait was not enough to enqueue all the jobs (on a heavily
loaded machine), and hence some jobs spawned thread pool threads after
calling `g_thread_pool_stop_unused_threads()`, leading to the assertion
failure.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
We have fallback in places for GNU's variadic arguments in macros, and
for static inline functions with variadic arguments as an fallback of
last resort, but going forward we are going to depend on `__VA_ARGS__`
for macros that cannot be re-implemented using a static inline function.
Fixes: #2681
Some of the CI runners disable ptrace, as they’re running inside a
container and ptrace would allow container escape. On those systems, gdb
can’t work, so skip the gdb test.
Add some additional stderr debug output to make it clear that’s the
case.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
This part of test_non_utf8_printf() was missed while converting other
test expectations to look for a figure space, probably because it is
only run if a legacy ja_JP.eucjp locale exists. Debian generates a fairly
comprehensive set of locales (including some that are non-UTF-8) while
running GLib's installed-tests, so this was caught by CI.
Fixes: 7074122 "gdatetime: Pad numbers with numeric space"
Signed-off-by: Simon McVittie <smcv@debian.org>
In `g_data_remove_internal()`, call the `GDataElt:destroy` functions in the
order that they appear in `keys`, instead of the order that they are found in
`datalist`.
Fixes https://gitlab.gnome.org/GNOME/glib/-/issues/2672
Clang seems to apply more rigorous type checks across the different
arguments to the atomic builtins. Fix various
`-Wincompatible-pointer-types`, `-Wpointer-sign` and `-Wint-conversion`
warnings.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
