GDK has a lock to mark critical sections inside the backends.
Additionally, code that would re-enter into the GTK main loop was
supposed to hold the lock.
Back in the Good Old Days™ this was guaranteed to kind of work only on
the X11 backend, and would cause a neat explosion on any other GDK
backend.
During GTK+ 3.x we deprecated the API to enter and leave the critical
sections, and now we can remove all the internal uses of the lock, since
external API that uses GTK+ 4.x won't be able to hold the GDK lock.
https://bugzilla.gnome.org/show_bug.cgi?id=793124
The g_print documentation explicitly says not to do this, since
g_print is meant to be redirected by applications. Instead use
g_message for logging that can be triggered via GTK_DEBUG.
Remove checks for NULL before g_free() and g_clear_object().
Merge check for NULL, freeing of pointer and its setting
to NULL by g_clear_pointer().
https://bugzilla.gnome.org/show_bug.cgi?id=733157
If a motion event handler (or other handler running from the flush-events
phase of the frame clock) recursed the main loop then flushing wouldn't
complete until after the recursed main loop returned, and various aspects
of the state would get out of sync.
To fix this, change flushing of the event queue to simply mark events as
ready to flush, and let normal event delivery handle the rest.
https://bugzilla.gnome.org/show_bug.cgi?id=705176
When events are paused, we should not return TRUE from prepare() or check().
GTK+ handles this for events that are already in the GTK+ queue, but
we also need suppress checks for events that are in the system queue - if we
return TRUE indicating that there are events in the system queue, then we'll
call dispatch(), and do nothing. The event source will spin, and will never
run the other phases of the paint clock.
(Broadway doesn't have a window system queue separate from the GDK event queue,
but we write the function the same way for consistency.)
https://bugzilla.gnome.org/show_bug.cgi?id=694274
Apply patch from Kristian Rietveld which addresses two issues
in gdkeventloop-quartz.c:
This patch moves the autorelease pool drain and introduces protection against
the invalidated ufds. Basically, when we suspect ufds has been invalidated by a
recursive main loop instance, we refrain from calling the collect function.
(cherry picked from commit 79b3326eaa)
Wake up the run loop unconditionally (don't check if it is waiting) because
it might go into waiting state right after we checked for it. Fixes GIMP
startup (which has a lot of GIOChannel I/O but zero NSEvents) from several
minutes to a few seconds.
(cherry picked from commit 0729cdc9a1)
Fixes e.g. crashs when dropping from finder.
Turn the "getting_events" boolean into a counter to handle poll_func()
being called recursively, and track the loop depth correctly by
changing its counter before bailing out in run_loop_observer_callback().
This way we reallocate our autorelease pool at the right time, and
don't kill memory that is still in use by outer run loops.
Also drain, not release the pool, just for some defensive forward
compatibility.
(cherry picked from commit ef9a92d225)
2008-11-12 Richard Hult <richard@imendio.com>
Bug 550942 – [patch] Rework of gdkeventloop-quartz.c
* gdk/gdk.c:
* gdk/gdkinternals.h: Add eventloop debug facility.
* gdk/quartz/gdkeventloop-quartz.c: Big rework of the quartz
mainloop integration, patch from Owen Taylor. See bug #550942 for
the details.
svn path=/trunk/; revision=21783
2008-05-08 Richard Hult <richard@imendio.com>
* gdk/quartz/gdkeventloop-quartz.c (poll_func): Only set the fake
fd's revents, not the first one. Fixes bug #531056, spotted by
Yevgen Muntyan.
svn path=/trunk/; revision=20086
2008-02-24 Richard Hult <richard@imendio.com>
* gdk/quartz/gdkprivate-quartz.h:
* gdk/quartz/gdkeventloop-quartz.c: (got_fd_activity), (poll_func):
Use the subtype field for the custom event that is used to wake up
the mainloop so we can have other custom event types.
* gdk/quartz/gdkevents-quartz.c:
(_gdk_quartz_events_trigger_crossing_events):
* gdk/quartz/gdkwindow-quartz.c: (show_window_internal): Create
crossing events after showing a window if necessary, to work
around problems with the tracking rect API.
svn path=/trunk/; revision=19637
2007-12-10 Richard Hult <richard@imendio.com>
* gdk/quartz/gdkdrawable-quartz.c:
(gdk_quartz_drawable_get_context),
(gdk_quartz_drawable_release_context):
* gdk/quartz/gdkeventloop-quartz.c: (gdk_event_prepare),
(gdk_event_check), (gdk_event_dispatch), (poll_func):
* gdk/quartz/gdkwindow-quartz.h: Replace the autorelease pools
used for each drawing context and in prepare, dispatch and poll
with one that exists across each main loop iteration. Fixes leaks
on leopard and protects against future leaks introduce when the
underlying system changes again (bug #492977).
svn path=/trunk/; revision=19149
2007-12-10 Richard Hult <richard@imendio.com>
* gdk/quartz/gdkprivate-quartz.h:
* gdk/quartz/gdkeventloop-quartz.c:
(_gdk_quartz_event_loop_get_pending),
(_gdk_quartz_event_loop_check_pending),
(_gdk_quartz_event_loop_release_event), (gdk_event_prepare),
(gdk_event_check), (gdk_event_dispatch):
* gdk/quartz/gdkevents-quartz.c: (gdk_events_pending)
(_gdk_events_queue): Fix a bug where we could end up trying to
handle the same event more than once. Based on patch from Paul
Davis.
svn path=/trunk/; revision=19143
2007-04-28 Ronald Bultje <rbultje@ronald.bitfreak.net>
* gdk/quartz/gdkeventloop-quartz.c: (select_thread_func),
(poll_func): fix two more potential races that could happen when
an application is polling in the mainloop and a separate thread
tries to wake it up using g_idle_add(). Fixes#425271 comment 5.
svn path=/trunk/; revision=17680
* gdk/quartz/gdkeventloop-quartz.c: Protect the polling thread
setup and shutdown function by mutexes and read the wakeup
pipe unconditionally and unblocking. This should make the main
loop always be woken up when using g_idle_add() from another
thread in the Quartz backend (#425271).
svn path=/trunk/; revision=17595
2007-02-18 Richard Hult <richard@imendio.com>
* gdk/quartz/GdkQuartzView.h:
* gdk/quartz/GdkQuartzWindow.h:
* gdk/quartz/gdkquartz.h: Import AppKit.h instead of Quartz.h.
* gdk/quartz/gdkscreen-quartz.c: (get_mm_from_pixels): Don't use
userSpaceScaleFactor for 10.3. Those changes bring us closer to working
on panther, patch from Mathias Hasselmann.
* gdk/quartz/gdkeventloop-quartz.c: Add includes to fix build warnings.
svn path=/trunk/; revision=17323
2006-12-28 Mikael Hallendal <micke@imendio.com>
* gdk/quartz/gdkevents-quartz.c: Factored out the event loop
integration into gdkeventloop-quartz.c.
* gdk/quartz/Makefile.am: Added gdkeventloop-quartz.c
* gdk/quartz/gdkeventloop-quartz.c: New file containing the event loop
integration.
