xserver/miext
Enrico Weigelt, metux IT consult ef62929f58 treewide: NULL-protect ScreenRec->DestroyPixmap() calls
Right now, we're assuming that even when deep nesting involved, the proc
vector is always set to a valid function. One the one hand it requires
extra dummy procs in some cases, OTOH it's making upcoming refactoring
of the code flow unnecessarily complex.

The big plot (of subsequent commits) is splitting out the extension's
(and possibly subsystem's) special logic out of the wrapping chain and
let them be executed independently from the DDX/drivers - when applicable
even only when the pixmap is really destroyed (not just unref'ed).
(At some later point, it might even become be actually a valid situation
that DestroyPixmap vector really being NULL.)

See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1755

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1709>
2025-02-06 23:02:06 +00:00
..
damage treewide: NULL-protect ScreenRec->DestroyPixmap() calls 2025-02-06 23:02:06 +00:00
rootless mi: unexport miShapedWindowIn() 2025-02-06 16:45:20 +02:00
shadow drop obsolete HAVE_DIX_CONFIG_H 2024-10-10 13:38:31 +00:00
sync drop obsolete HAVE_DIX_CONFIG_H 2024-10-10 13:38:31 +00:00