In g_object_unref(), we call _object_unref_clear_weak_locations() before
and after dispose() already. At those places it is necessary to do.
Calling it a third time during g_object_real_dispose() seems not useful,
has unnecessary overhead and actually undesirable.
In particular, because g_object_real_dispose() is the implementation for
the virtual function GObject.dispose(). For subclasses that override dispose(),
it's not well defined at which point they should chain up the parent
implementation (for dispose(), I'd argue that usually they chain up at
the end of their own code). If they chain up at the end, this has no
effect.
This only really matters if you try to register GWeakRef during dipose
and/or resurrect the object.
static void dispose(GObject *object)
{
g_weak_ref_set(&global_weak_ref, object);
global_ref = g_object_ref(object);
G_OBJECT_CLASS (parent_class)->dispose (object);
}
the object was resurrected, but g_object_real_dispose() would clear the
weak ref. That is not desirable, nor does it make sense.
Instead, the virtual function dispose() is called from two places, from
g_object_unref() and g_object_run_dispose(). In both cases, it is
ensured that weak locations are cleared *after* dispatching the virtual
function. Don't do it somewhere in the middle from
g_object_real_dispose().
dispose() can resurrect an object and/or register a weak-ref. After
returning from dispose(), we must check again. And we must do so in a
race-free manner, where we check that we have no more weak-locations
and the ref-count is one.
In fact, if _object_unref_clear_weak_locations() determines that the
ref-count is 1, it must also decrement the ref-count to zero while
holding the weak_locations_lock. This prevents g_weak_ref_set() to
still register a weak-pointer after the reference count dropped to zero.
Also add an assertion to g_weak_ref_set(), that the object is still
alive. The assertion is useful to finding bugs, but the change really
makes it impossible that a wrongly used g_weak_ref_set() can still
resurrect the object after finalization starts.
The final
g_datalist_id_set_data (&object->qdata, quark_weak_locations, NULL);
during finalization is no longer necessary and dropped.
We never set any data for quark_weak_refs. It's unused, drop it.
Also, fail a g_critical() assertion, if the GWeakRef is unexpectedly not
registered in the object. That really shouldn't happen.
Previously:
1. old_val = atomic_add(&object->ref_count);
2. if (old_val == 1 && OBJECT_HAS_TOGGLE_REF (object)) { toggle_notify() }
As old_val was 1, you might think that no other thread can have a valid
reference to object. However, that's not the case. For one, GWeakRef can
be used to create another strong reference. More easily, the single
reference can be shared between multiple threads (as long as the code
takes care that the object lives long enough).
That means, another thread can easily add and drop references (including
toggle references). All between step 1 and 2.
A race here might be hard to hit, and the effect might not be obviously
bad. However, consider old_val is 1 due to a normal reference, and
another thread adds a toggle ref between step 1. and 2. Then we would
notify a toggle from 1->2, although a newly added toggle ref is expected
to always start with a normal and a toggle reference. The first toggle
notification is expected to notify about the loss of other references, not
about getting a second reference.
To handle this properly, when we increase the reference count from 1 to
2, we must do so under a lock and check for the toggle notification.
As we now correctly track the toggle behavior, we can also assert in
toggle_refs_get_notify_unlocked() that n_toggle_refs agrees with the
number of references, that is, that the user did always match
g_object_add_toggle_ref() with g_object_remove_toggle_ref().
The downside is here too, that there is now a case (when increasing the
reference count from 1 to 2) where we need to take the global lock.
That performance problem should be addresses by using per-object locks
instead of a global lock.
The previous g_object_unref() was racy. There were three places where we
decremented the ref count, but still accessed the object afterwards
(while assuming that somebody else might still hold a reference). For
example:
if (!g_atomic_int_compare_and_exchange_full ((int *) &object->ref_count,
old_ref, old_ref - 1,
&old_ref))
continue;
TRACE (GOBJECT_OBJECT_UNREF (object, G_TYPE_FROM_INSTANCE (object), old_ref));
/* if we went from 2->1 we need to notify toggle refs if any */
if (old_ref == 2 && OBJECT_HAS_TOGGLE_REF (object))
{
/* The last ref being held in this case is owned by the toggle_ref */
toggle_refs_notify (object, TRUE);
}
After we decrement the reference count (and gave up our reference), we
are only allowed to access object if we know we have the only possible
reference to it. In particular, if old_ref is larger than 1, then
somebody else holds references and races against destroying object.
The object might be a dangling pointer already.
This is slightly complicated due to toggle references and clearing of
weak-locations.
For toggle references, we must take a lock on the mutex. Luckily, that
is only necessary, when the current reference count is exactly 2.
Note that we emit the TRACE() after the ref count was already decreased.
If another thread unrefs the object, inside the TRACE() we might have a
dangling pointer. That would only be fixable, by emitting the TRACE()
before the actual unref (which has its own problems). This problem
already existed previously.
The change to the test is necessary and correct. Before this patch,
g_object_unref() would call dispose() and decrement the reference count
right after.
In the test case at gobject/tests/reference.c:1108, the reference count
after dispose and decrement is 1. Then it thaws the queue notification,
which emits a property changed signal. The test then proceeds to
reference the object again and notifying the toggle reference.
Previously, the toggle reference was notified 3 times.
After this change, the property changed signal is emitted before
decreasing the reference count. Taking a reference then does not cause
an additional toggle on+off, so in total only one toggle happens.
That accounts for the change in the test. The new behavior is
correct.
The indentation level in g_object_unref() is wrong. Fix it by reformatting
the function with clang-format. That makes follow up patches easier to adhere
a consistent style.
No other changes.
Toggle refs are seldom used, and when they are, it makes mostly sense that
there is only one of them. Thus, when removing the last toggle ref, also
remove the associated data.
Previously:
- if the object is currently not frozen, we called
g_object_notify_queue_freeze() once. Afterwards dispatch the event
directly. This is probably the common case, and requires one
notify_lock lock.
- if the object is currently frozen, we call
g_object_notify_queue_freeze(), g_object_notify_queue_add().
g_object_notify_queue_thaw().
This required taking the notify_lock three times.
- if the object is currently not frozen and in_init, then we called
g_object_notify_queue_freeze(), g_object_notify_queue_freeze(),
g_object_notify_queue_add(). This also required to take
the lock three times. There is another thaw at the end of
object initialization.
That was because we first call g_object_notify_queue_freeze() to see
whether we are frozen. And depending on that, queue the event (and thaw
again).
Instead, g_object_notify_queue_add() can do the check and queueing in
one step. There is no need to call a freeze() to (conditionally) enqueue
a notification. Now only one lock is taken in all cases.
Also, g_object_notify_queue_freeze() and g_object_notify_queue_thaw()
both call g_datalist_id_get_data() (which also take a bit lock). As the
thaw is no longer necessary, the second lock is also saved.
Before dispatching signals (and calling out to user code), we want to
take a reference and ensure that the object stays alive.
However, a thaw may not decrease the freeze_count to zero, or there may
be no properties to notify. Avoid taking a reference in those cases.
This was done since the beginning (commit e773d7dba6 ('fixed dealing
with collection/lcopy of NULL values.'). But it's not clear, why we
would need to take a reference on the calling object.
Freeze does not emit any signals/callbacks and does not call back to the
user. It just sets up some internal state.
This doesn't require to take a reference. The caller must hold a valid
reference to being with, but if that's given, there is no need to
acquire another reference.
g_atomic_int_get() returns a signed int. While we don't expect this to be ever
negative, a negative value would also indicate a bug. Adjust the check to assert
against negative ref-count too.
Generating gir and typelib files has inter-dependencies that may depend
on other elements.
For example, glib requires gobject and gdump generated files require
gmodule, so we've a cyclic dependency because gmodule requires gobject,
that requires glib.
To prevent this, let's just generate the introspection files at once in
a different meson file so that we don't have to deal with this.
As per this we could even revert commit fa37ab6d0 since gio is now
compiled before the gir files.
GLib gir requires glib-types.h that also includes gobject-visibility.h
that needs to be generated in order to be able to generate the GLib gir,
so explicitly add it to the sources.
Also, in order to link the typelib we need the definitions of the *_get_type()
functions that are defined in the gboxed.c file, so this should be part
of the gir sources or we'll have linking issues.
Aside from checking that we're accessing the global GParamSpecPool
without necessarily initializing GObjectClass, we should also verify
that we're doing so safely without the class init lock.
Ensure that the fix in commit af024b6d7e7d3fbef23c1f7d1f7704fc90ac4fb1
works, by replicating what gobject-introspection does:
- initialise the default GTypeInterface from a GType without also
initialising GObjectClass
- install a property for the interface
- find the GParamSpec using g_object_interface_find_property()
Right now, we're assuming that GObjectClass will be initialised first
and under a lock, but that's not always the case: when traversing a list
of type, the first one might be a GTypeInterface, and if we initialise
an interface that installs a property, the whole thing comes crashing
down because the global GParamSpecPool is not initialised.
Instead of taking a lock everywhere, we can use an atomic compare and
swap; the first thread that installs a property wins the race, as any
other access to the GParamSpecPool is performed under a lock.
While GParamSpecPool should never be used by newly written code, having
the ability to free the associated memory is a good idea. The only
reason why this hasn't been necessary until now is that we assume base
classes are going to keep their GParamSpecPool around forever.
Symptom:
20 0x00007ffff756337b in gtk_widget_size_allocate_with_baseline (widget=0xd7a0c0 [DirectWindow], allocation=, baseline=) at
gdb.error: value has been optimized out
Nicks and blurbs don't have any practical use for gio/gobject libraries.
Leaving tests untouched since this features is still used by other libraries.
Closes#2991
Signals (as opposed to signal connections) can never be unregistered, so
these closures have to be around forever.
Fixes some g-ir-scanner warnings.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: #3037
Boxed types can never be unregistered, so these closures have to be
around forever.
Fixes some g-ir-scanner warnings.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: #3037
This allows the methods for `GDir` to be introspected. While we don’t
expect languages which use the introspection bindings to use `GDir`,
full introspection support is necessary for the `GDir` documentation to
be correctly built with gi-docgen.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: #3037
Otherwise the installed one will be used, which will mean that new API
will not be available when linking.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Modify all the similar Python test wrappers to set
`G_DEBUG=fatal-warnings` in the environment of the program being tested,
so we can catch unexpected warnings/criticals.
Adding this because I noticed it was missing, not because I noticed a
warning/critical was being ignored.
Signed-off-by: Philip Withnall <philip@tecnocode.co.uk>
On CHERI-enabled systems we use uintptr_t as the underlying storage for
GType and therefore casting to gsize strips the upper bits from a pointer.
Fix this by casting via uintptr_t instead and introduce a new set of
macros to convert between GType and pointers.
This is required for CHERI systems such as Arm Morello, where gsize is
not large enough to hold a pointer. As GType can contain pointers, we
have to use guintptr instead.
It needs to be in a separate page because it’s all macros and they have
no type/class associated with them.
Signed-off-by: Philip Withnall <philip@tecnocode.co.uk>
Helps: #3037
Currently, the introspection data for GLib and its sub-libraries is
generated by gobject-introspection, to avoid the cyclic dependency
between the two projects.
Since gobject-introspection is generally available on installed systems,
we can check for its presence, and generate the introspection data
directly from GLib.
This does introduce a cyclic dependency, which is why it's possible to
build GLib without introspection, then build gobject-introspection, and
finally rebuild GLib.
By having introspection data available during the GLib build, we can do
things like generating documentation; validating newly added API; and
close the loop between adding new API and it becoming available to non-C
consumers of the C ABI (i.e. language bindings).
Round-tripping pointers via gsize is not guaranteed to work (the C standard
only requires this for (u)inptr_t) and in fact breaks on CHERI-enabled
systems such as Arm Morello where pointers are 128-bits but size_t is 64.
This means the current casts would strip the high bits of the pointer and
return a non-dereferenceable value. Fix this by casting the operand that
holds the pointer to guintptr instead of gsize.
Helps: https://gitlab.gnome.org/GNOME/glib/-/issues/2842
This should not result in any functional changes, but will eventually
allow glib to be functional on CHERI-enabled systems such as Morello.
Helps: https://gitlab.gnome.org/GNOME/glib/-/issues/2842
If you take a release build (--buildtype=release) previously of GLib,
functions such as g_type_create_instance() would call out to
g_type_test_flags() which you can see by disassembling and looking for the
call instruction to <g_type_test_flags> via the library ABI.
Now that the previous commit allows checking abstract and deprecated flags
it makes sense to let the compiler optimize those checks into a single
pass. This is only possible if the functions themselves are inlined.
Additionally, any time we have the same TypeNode we would want to be able
to reuse that node rather than re-locate it.
Every call to g_type_create_instance() currently will incur a RWLock at
least once, but usually twice to test for both G_TYPE_FLAG_ABSTRACT and
G_TYPE_FLAG_DEPRECATED.
Additionally, each call to g_type_instance_free() also checks for these.
That results in a synchronization of GTypeInstance creation across all
threads as well as being a huge amount of overhead when creating instances
like GskRenderNode.
With this patch in place, the next two biggest issues are
g_type_class_ref() and g_type_test_flags() not getting inlined within
gtype.c in release builds. We can address that separately though.
Sysprof shows that the RWLock, with this patch in place, falls off the
profiles.
The introspection parser isn't good enough to expand the shift symbol,
which means G_TYPE_FUNDAMENTAL_MAX is evaluated as (255 << 0).
We can use the `(value)` annotation to force the symbol value in the
introspection data.
See: GNOME/gobject-introspection#473
When building for CHERI with additional warning flags, implicitly
converting uintptr_t to an integer type that can't store a pointer
results in a compiler warnings. Silence two of these by adding
explicit casts.
This patch is based upon Garrett Regier's work from 2015 to provide
some reliability and predictability to how disposal handles weak
reference state.
A primary problem is that GWeakRef and GWeakNotify state is shared and
therefore you cannot rely on GWeakRef status due to a GWeakNotify
calling into second-degree code.
It's important to ensure that both weak pointer locations and GWeakRef
will do the proper thing before user callbacks are executed during
disposal. Otherwise, user callbacks cannot rely on the status of their
weak pointers. That would be mostly okay but becomes an issue when
second degree objects are then disposed before any notification of
dependent object disposal.
Consider objects A and B.
`A` contains a reference to `B` and `B` contains a `GWeakRef` to `A`.
When `A` is disposed, `B` may be disposed as a consequence but has not
yet been notified that `A` has been disposed. It's `GWeakRef` may also
cause liveness issues if `GWeakNotify` on `A` result in tertiary code
running which wants to interact with `B`.
This example is analagous to how `GtkTextView` and `GtkTextBuffer` work
in text editing applications.
To provide application and libraries the ability to handle this using
already existing API, `GWeakRef` is separated into it's own GData quark
so that weak locations and `GWeakRef` are cleared before user code is
executed as a consequence of `GData` cleanup.
# Conflicts:
# gobject/tests/signals.c
There’s no reason that anyone can think of that this should be
disallowed. It’s useful for language runtimes like GJS to be able to
find out the allocation size of dynamic GObjects.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Fixes: #623
UAC will terminate this test program from running in 32-bit x86 builds as
it believes that it will alter Windows. In order to make this run, we
create a manifest file for 32-bit Windows builds in order to tell UAC
that this program should not need admin privileges.
This will allow the entire test suite for GLib to run on 32-bit Windows
builds.
Add to script the new functions added in
commit 2368187e and commit e5ee6e14 which are:
signal_emitv_unlocked()
signal_emit_valist_unlocked()
so that the "<emit signal 'blabla'>" line keeps
showing after them.
The compiler annotations are mainly useful for people using the symbols
directly.
To avoid getting compiler warnings for the GTypeValueTable definition
itself, we need to wrap the structure with
G_GNUC_BEGIN_IGNORE_DEPRECATIONS and G_GNUC_END_IGNORE_DEPRECATIONS.
The introspection scanner cannot deal very well with function pointers
into a plain structure. In order to document the various function
pointers in GTypeValueTable we need to create typed callbacks, and
use them to replace the anonymous function pointers inside the
structure. This not only allows us to properly document the function
pointers, but it also allows us to annotate the arguments and return
value of those function pointers.
See also: https://gitlab.gnome.org/GNOME/gobject-introspection/-/merge_requests/400#note_1721707
For now, the function parse_trigraph() defined in gobject/glib-mkenums
script was not taking double-quotes characters into account:
>>> parse_trigraph('name="eek, a comma"')
{'name': '"eek', 'a': None}
This patch take double-quotes characters into account:
>>> parse_trigraph('name="eek, a comma"')
{'name': 'eek, a comma'}
Closes issue #65
Normally we don't really have emission hooks around, so try to allocate
only tiny array to contain a few of them and in case we exceed that limit,
we go back to use allocated ones.
We used to call this function as unlocked, with a node value that
could be invalid at the point of the call, so let's ensure that when
we call such function it's defined, and then reduce the access to the
signal node members when we're unlocked or after a lock/unlock operation
that may have changed it.
As per this, add more tests handling multiple signal hooks cases that we
did not cover before.
In g_signal_emit_valist() we used to access to param types array and
n_params values after unlocking the mutex, and this might have lead to
making such values unreliable for the current call.
So let's keep them around until we're done with the function call
Since we're locking and unlocking once we've found the signal ID, we
might have performed calls to g_signal_emit_valist() with a signal id
that was already been removed, and thus failing later.
This is not really an issue as inside g_signal_emit_valist() we were
re-checking for the signal id, but we can make this more reliable so
that the first thread that acquires the lock can also be sure to emit.
