Try changing mode on shm segments to 0600. We'll see who complains.
authorOwen Taylor <otaylor@redhat.com>
Wed, 18 Apr 2001 18:28:19 +0000 (18:28 +0000)
committerOwen Taylor <otaylor@src.gnome.org>
Wed, 18 Apr 2001 18:28:19 +0000 (18:28 +0000)
commit6cff7051b16d99178c5320592e96c757597835e4
tree6aabc47ef5228d44c3a95b3a55cee06281200ac7
parent60b6a010e931aaaf97d723c893068381a421f0a0
Try changing mode on shm segments to 0600. We'll see who complains.

Wed Apr 18 14:23:14 2001  Owen Taylor  <otaylor@redhat.com>

* gdk/x11/gdkimage-x11.c (gdk_image_new): Try changing
mode on shm segments to 0600. We'll see who complains.

* gdk/gdkwindow.c (_gdk_window_destroy_hierarchy): Call
_gdk_windowing_window_destroy() AFTER recursing through
children.

* tests/Makefile.am (noinst_PROGRAMS): Build testsocket,
testsocket_child on X.

* tests/testsocket[_child].c: Fix uses of
gtk_window_get_default_accel_group().

[ Merge patch from Ramiro Estrugo  <ramiro@eazel.com> from gtk-1-2 ]

* gdk/gdkimage.c: (gdk_image_get):
Deal with the possibility that XGetImage() might return NULL.
Allocate the GdkImagePrivate structure only after XGetImage()
succeeds in order not to dereference a NULL ximage pointer.  This
prevents a core dump when XGetImage() fails - which is unlikely,
but can happen due to race conditions accessing the geometries of
drawables.  An x error will still be triggered, but the gdk image
wrapper at least wont seg fault.
13 files changed:
ChangeLog
ChangeLog.pre-2-0
ChangeLog.pre-2-10
ChangeLog.pre-2-2
ChangeLog.pre-2-4
ChangeLog.pre-2-6
ChangeLog.pre-2-8
gdk/gdkwindow.c
gdk/gdkwindow.h
gdk/x11/gdkimage-x11.c
tests/Makefile.am
tests/testsocket.c
tests/testsocket_child.c