The changes made in commit bc59e28bf6b0f70ff345aef80356d0076f44a0e7
(issue #3399) fixed introspection of the GThread API. However, they
introduced a trampoline in every threading function. So with those
changes applied, the disassembly of `g_mutex_lock()` (for example) was:
```
0x7ffff7f038b0 <g_mutex_lock> jmp 0x7ffff7f2f440 <g_mutex_lock_impl>
0x7ffff7f038b5 data16 cs nopw 0x0(%rax,%rax,1)
```
i.e. It jumps straight to the `_impl` function, even with an optimised
build. Since `g_mutex_lock()` (and various other GThread functions) are
frequently run hot paths, this additional `jmp` to a function which has
ended up in a different code page is a slowdown which we’d rather avoid.
So, this commit reworks things to define all the `_impl` functions as
`G_ALWAYS_INLINE static inline` (which typically expands to
`__attribute__((__always_inline__)) static inline`), and to move them
into the same compilation unit as `gthread.c` so that they can be
inlined without the need for link-time optimisation to be enabled.
It makes the code a little less readable, but not much worse than what
commit bc59e28bf6b0f70ff345aef80356d0076f44a0e7 already did. And perhaps
the addition of the `inline` decorations to all the `_impl` functions
will make it a bit clearer what their intended purpose is
(platform-specific implementations).
After applying this commit, the disassembly of `g_mutex_lock()`
successfully contains the inlining for me:
```
=> 0x00007ffff7f03d80 <+0>: xor %eax,%eax
0x00007ffff7f03d82 <+2>: mov $0x1,%edx
0x00007ffff7f03d87 <+7>: lock cmpxchg %edx,(%rdi)
0x00007ffff7f03d8b <+11>: jne 0x7ffff7f03d8e <g_mutex_lock+14>
0x00007ffff7f03d8d <+13>: ret
0x00007ffff7f03d8e <+14>: jmp 0x7ffff7f03610 <g_mutex_lock_slowpath>
```
I considered making a similar change to the other APIs touched in #3399
(GContentType, GAppInfo, GSpawn), but they are all much less performance
critical, so it’s probably not worth making their code more complex for
that sake.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Fixes: #3417
Move various doc/introspection comments from `gthread-posix.c` (which is
platform-specific) to `gthread.c` (which is not). Having the
introspection annotations and doc comments in a platform-independent
file means that they are seen by the build process on all platforms, and
we don’t end up with unintrospectable APIs on some platforms, or
platform-specific annotation differences.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: #3399
Setting the main thread's scheduler settings is not reliably possible,
especially not if SELinux or similar mechanisms are used to limit what
can be done.
As such, get rid of all the complicated code that tried to do this
better and use a separate thread for spawning threads for the global
shared thread pool. These will always inherit the priority of the main
thread (or rather the thread that created the first shared thread pool).
Fixes https://gitlab.gnome.org/GNOME/glib/-/issues/2769
Add SPDX license (but not copyright) headers to all files which follow a
certain pattern in their existing non-machine-readable header comment.
This commit was entirely generated using the command:
```
git ls-files glib/*.[ch] | xargs perl -0777 -pi -e 's/\n \*\n \* This library is free software; you can redistribute it and\/or\n \* modify it under the terms of the GNU Lesser General Public/\n \*\n \* SPDX-License-Identifier: LGPL-2.1-or-later\n \*\n \* This library is free software; you can redistribute it and\/or\n \* modify it under the terms of the GNU Lesser General Public/igs'
```
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Helps: #1415
Glib cannot be built statically on Windows because glib, gobject and gio
modules need to perform specific initialization when DLL are loaded and
cleanup when unloaded. Those initializations and cleanups are performed
using the DllMain function which is not called with static builds.
Issue is known for a while and solutions were already proposed but never
merged (see: https://gitlab.gnome.org/GNOME/glib/-/issues/692). Last
patch is from version 2.36.x and since then the
"constructor/destructor" mechanism has been implemented and used in
other part of the system.
This patch takes back the old idea and updates it to the last version of
glib to allow static compilation on Windows.
WARNING: because DllMain doesn't exist anymore in static compilation
mode, there is no easy way of knowing when a Windows thread finishes.
This patch implements a workaround for glib threads created by calling
g_thread_new(), so all glib threads created through glib API will behave
exactly the same way in static and dynamic compilation modes.
Unfortunately, Windows threads created by using CreateThread() or
_beginthread/ex() will not work with glib TLS functions. If users need
absolutely to use a thread NOT created with glib API under Windows and
in static compilation mode, they should not use glib functions within
their thread or they may encounter memory leaks when the thread finishes.
This should not be an issue as users should use exclusively the glib API
to manipulate threads in order to be cross-platform compatible and this
would be very unlikely and cumbersome that they may mix up Windows native
threads API with glib one.
Closes#692
glib/gthread-win32.c: In function 'g_system_thread_new':
glib/gthread-win32.c:507:37: warning: comparison of integer expressions of different signedness: 'DWORD' {aka 'long unsigned int'} and 'int'
if (ResumeThread (thread->handle) == -1)
^~
glib/gthread-win32.c: In function 'g_private_get_impl':
glib/gthread-win32.c:310:16: warning: cast from pointer to integer of different size
DWORD impl = (DWORD) key->p;
^
glib/gthread-win32.c:315:14: warning: cast from pointer to integer of different size
impl = (DWORD) key->p;
^
glib/gthread-win32.c: In function 'SetThreadName':
glib/gthread-win32.c:596:60: warning: passing argument 4 of 'RaiseException' from incompatible pointer type
RaiseException (EXCEPTION_SET_THREAD_NAME, 0, infosize, (DWORD *) &info);
^~~~~~~~~~~~~~~
Zero is a valid TLS index so it needs to be checked. It’s also the
integer used to indicate that no TLS has been allocated yet, so it can’t
be used as a TLS identifier.
Incorporates changes from Philip Withnall.
Fixes: #2058
Since Windows 10 1607, we can make use of SetThreadDescription() API
for setting thread name. Unlike previously used exception based
method, this API will preserve configured thread name on dump file.
In the Windows destructor list, consistently access
`g_private_destructors` using atomic primitives.
`g_atomic_pointer_compare_and_exchange()` should be equivalent to
`InterlockedCompareExchangePointer()`, but is a bit more understandable
in a general GLib context, and pairs with `g_atomic_pointer_get()`. (I
can’t find a Windows API equivalent for that.)
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Helps: #600
If the old value is destroyed before updating the TLS value in pthreads
(or the Windows equivalent) then there’s a risk of infinite recursion if
`g_private_replace()` is called from within the `GDestroyNotify`.
Avoid that by destroying the old value after doing the TLS update.
Thanks to Matthias Clasen for diagnosing the issue.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Fixes: #2210
This was mostly machine generated with the following command:
```
codespell \
--builtin clear,rare,usage \
--skip './po/*' --skip './.git/*' --skip './NEWS*' \
--write-changes .
```
using the latest git version of `codespell` as per [these
instructions](https://github.com/codespell-project/codespell#user-content-updating).
Then I manually checked each change using `git add -p`, made a few
manual fixups and dropped a load of incorrect changes.
There are still some outdated or loaded terms used in GLib, mostly to do
with git branch terminology. They will need to be changed later as part
of a wider migration of git terminology.
If I’ve missed anything, please file an issue!
Signed-off-by: Philip Withnall <withnall@endlessm.com>
On Linux the sched_getattr syscall might be available at compile-time
but not actually work at runtime (e.g. because an older kernel is
running or valgrind is used). Instead of killing the process, return
FALSE and handle this gracefully at runtime with some fallback code.
Fixes https://gitlab.gnome.org/GNOME/glib/issues/2007
By default (on POSIX) we would be inheriting thread priorities from the
thread that pushed a new task on non-exclusive thread pools and causes a
new thread to be created. This can cause any non-exclusive thread pool
to accidentally contain threads of different priorities, or e.g. threads
with real-time priority.
To prevent this, custom handling for setting the scheduler settings for
Linux and Windows is added and as a fallback for other platforms a new
thread is added that is responsible for spawning threads for
non-exclusive thread pools.
Fixes https://gitlab.gnome.org/GNOME/glib/issues/1834
This code was a persistent source of `-fsanitize=thread` errors
when I was trying to use it on OSTree.
The problem is that while I think this code is functionally correct,
we hold a mutex during the writes, but not the reads, and TSAN (IMO
correctly) flags that.
Reading this, I don't see a reason we need a mutex at all. At the
cost of some small code duplication between posix/win32, we can just
pass the data we need down into each implementation. This ends up
being notably cleaner I think than the awkward "lock/unlock to
serialize" dance.
(Minor review changes made by Philip Withnall <withnall@endlessm.com>.)
https://gitlab.gnome.org/GNOME/glib/issues/1224
The tests in test_async_queue_timed() assume that g_async_queue_timeout_pop()
and in turn g_cond_wait_until() wait at least until end_time
before returning, i.e. calling g_get_monotonic_time() after the timeout should result
in a value equal or larger than the timeout end time.
For the win32 implementation of g_cond_wait_until() this isn't the case which
makes those tests fail.
There are three reasons why the function returns early:
1) The underlying API works with milliseconds and the timeout gets rounded
down, resulting in a too small timeout value.
2) In case the timeout is too large to be passed to the API it gets limited
(there is also a bug because it converts INFINITE to milliseconds while
they already are, but using INFINITE would be wrong as well, as passing
a large timeout is not the same as blocking forever really)
3) Even with the rounding changed the underlying API still returns a bit early
sometimes on my machine (relative to g_get_monotonic_time())
This changes the implementation to round up to the next millisecond (fixing 1)
and to wait again in case a timeout occurs but the end time hasn't been
reached yet (fixing 2 and 3).
This makes the test_async_queue_timed() tests pass.
https://bugzilla.gnome.org/show_bug.cgi?id=795569
All glib/*.{c,h} files have been processed, as well as gtester-report.
12 of those files are not licensed under LGPL:
gbsearcharray.h
gconstructor.h
glibintl.h
gmirroringtable.h
gscripttable.h
gtranslit-data.h
gunibreak.h
gunichartables.h
gunicomp.h
gunidecomp.h
valgrind.h
win_iconv.c
Some of them are generated files, some are licensed under a BSD-style
license and win_iconv.c is in the public domain.
Sub-directories inside glib/:
deprecated/: processed in a previous commit
glib-mirroring-tab/: already LGPLv2.1+
gnulib/: not modified, the code is copied from gnulib
libcharset/: a copy
pcre/: a copy
tests/: processed in a previous commit
https://bugzilla.gnome.org/show_bug.cgi?id=776504
This works by using semi-documented[1] exception to tell the debugger
that a thread needs to have its name changed.
If this exception is not caught and handled by something, it will crash
the process, so we need to set up our own handler in case there's no
debugger attached or the debugger can't handle this type of exception.
Since SEH is not supported by gcc on i686 (at the moment), we need to use VEH
instead. For completeness the MSVC-oriented code still uses SEH, although
there is no reason why it shouldn't work with the VEH variant used by MinGW.
VEH handler has to be set up somewhere (g_thread_win32_init () works nicely)
and removed once it's not needed (g_thread_win32_process_detach () is added
expressly for that purpose). Note that g_thread_win32_process_detach() is
only called when glib is unloaded by FreeLibrary(), not when glib-using
process is terminating.
This exception is known to work with WinDbg, and adding support for it into
GDB proved to be feasible (GDB patch will be sent upstream, eventually).
[1] https://msdn.microsoft.com/en-us/library/xcb2z8hs%28v=vs.71%29.aspxhttps://bugzilla.gnome.org/show_bug.cgi?id=747478
This reverts commits dfbac178bdcbcb042be928dcbeaa87742b511836 and
56348210f38209a11902e54b2db41833aab36414.
These two commits introduce undesirable behaviour and were made with no
apparent approval from anybody at all, and without reference to a bug or
mailing list discussion.
When removing an item from the list, check the next one's my_owner,
and fix it accordingly. And take this case into account when last
of the list is deleted.
Also, assign NULL to 'my_owner' in g_thread_xp_WakeConditionVariable.
There are some races in the condition variable emulation code for
Windows XP with respect to timeouts while waiting.
First, in the event of a timeout, we never remove the waiter from the
condition variable. This can cause crashes later. That problem was
found by Rodrigo Rivas Costa.
Second, if the waiting thread times out and exits just as we were about
to call SetEvent() on its waiter event, we could end up trying to access
the waiter after it was closed/freed. We need to hold on to the lock a
little bit longer to ensure that that's not possible.
https://bugzilla.gnome.org/show_bug.cgi?id=666551
This was turning all the GLIB_VARs in the glib headers into
dllexports on windows, causing all sort of nastiness. libgthread is
mostly empty now anyway, so we don't need any GLIB_COMPILATION like
flag.
Switch GCond to using monotonic time for timed waits by introducing a
new API based on monotonic time in a gint64: g_cond_wait_until().
Deprecate the old API based on wallclock time in a GTimeVal.
Fix up the gtk-doc for GCond while we're at it: update the examples to
use static-allocated GCond and GMutex and clarify some things a bit.
Also explain the rationale behind using an absolute time instead of a
relative time.
Merge the GThreadData with the GThreadWin32 struct. Drop the extra TLS
variable.
Close the handle on _free(), which means that there is no leak if
g_system_thread_join() isn't called.
Remove all internal concept of joinability.
