2539ddae50
modsetting: also add libglx to library symbol test
...
Increase test coverage a bit.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1798 >
2025-02-17 18:42:52 +00:00
e10283d842
xfree86: os-support: unexport xf86scanpci()
...
Not used by any drivers/modules, so no need to keep it exported.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1800 >
2025-02-17 18:35:43 +00:00
b3ec0bf58a
dix: unexport MakePredeclaredAtoms()
...
Should never be called by modules/drivers.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1802 >
2025-02-17 18:13:33 +00:00
9dc26f6743
os: no need to defined PATH_MAX
...
All our supported platforms define it, so no need to do
it on our own.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1803 >
2025-02-17 16:55:13 +01:00
6c24e53a89
os: drop `upstart` specific SIGSTOP signaling logic
...
Upstart is long dead, discontinued a decade ago, so there's no
need to keep around a special signaling logic just for it.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1791 >
2025-02-14 14:25:28 +01:00
6349773d5e
mi: (re)export and document miPointerInitialize()
...
Xrdp project request exporting this function again, because (unlike the
usual xf86-video-* drivers) they need their own custom cursor handling:
RDP is designed to draw cursors on client side.
Also documenting what the function does.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1787 >
2025-02-13 23:53:39 +00:00
7933cc24d0
xvfb: Add multiple CRTC support
...
Multiple CRTCs can be added on a per-screen basis with the new -crtcs
option. Each CRTC has one associated output. Outputs beyond the first
are disabled by default and can be enabled by setting a mode. Outputs
can be disabled again by setting the associated CRTC's mode and output
to None.
Signed-off-by: Andy Myers <andy.myers@zetier.com>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1749 >
2025-02-13 23:22:49 +00:00
b61647f3a1
dix: add in-code docs for dixDestroyPixmap()
...
Give some more explaination for the dixDestroyPixmap() function.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
ee798cf212
exa: simplify CreatePixmap()/DestroyPixmap() handlers error pathes
...
Instead of complex wrap/unwrap trickery in the error path, just
protect the DestroyPixmap() handlers from half-initialized state.
Prior to this change, we always had to make sure, the we're calling
the original (screen driver's) handler directly, instead of our own
(which calls the original one, too), so it doesn't do any damage.
This isn't necessary anymore, since it finds out on it own what to do.
This not just makes the code flow simpler and easier to understand, but
also clears the road for decoupling the extension specific pixmap destructor
logic from the ScreenRec proc vectors (*1).
*1) see: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1755
https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
69837185c0
glx: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
0132baa422
xwayland: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
4d1953728e
xfree86: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
c117925ace
vfb: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
4378656cbb
kdrive: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
0a54e24721
xnest: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
c8607ca66f
render: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
d2a93d0346
randr: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
4628254698
mi: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
372a510ef0
miext: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
7a0f8301c5
glamor: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
08ec122fa7
fb: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
9ca03e6da0
exa: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
4694b8488e
dri3: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
668d9fc40e
dbe: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
eb5476381a
composite: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
5b541780c1
dix: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
c0f3b5bcef
Xext: shm: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
7ce19233bc
Xext: saver: use dixDestroyPixmap() instead of direct driver call
...
Direct calls to ScreenRec->DestroyPixmap() blocks cleaning up the wrapping
jungle, so use the proper dix function instead.
See: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1754
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:30 +01:00
984da40fbb
doc: document that ScreenRec->DestroyPixmap() shouldn't be called directly
...
Direct calls to ScreenRec->DestroyPixmap() is fragile and blocks cleaning up
the wrapping jungle, so use the proper dix function instead.
While this function originally wasn't made for that purpose (and so we have an
useless parameter here, and the naming isn't entirely correct), it's the best
effort we have right now, without breaking existing API: it's already exported
since long time, so drivers can be easily converted and still work with both
new and older Xserver ABI versions.
Retrospectively, this already should have been done 20 years ago.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1711 >
2025-02-12 17:48:26 +01:00
2e0c19b6d9
dix: add getter for display name
...
Xrdp needs to know the current display name (for setting up it's
own server sockets accordingly). Instead of exporting an internal
field, adding a little getter for this.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1783 >
2025-02-11 19:13:01 +01:00
80a3df55ae
xfree86: xf86Opt.h: fix missing include
...
<X11/Xdefs.h> is needed for `Bool` type.
Consumers shouldn't have to rely on Xdefs.h being accidentally included
by something else.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1782 >
2025-02-11 14:52:15 +00:00
5f1bdd0aa3
ci: add intel driver to build matrix
...
Add the intel driver to our CI, in order to prevent API breaks going
unnoticed again.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
d4f60b5462
ci: workaround for building xf86-video-intel via autotools
...
This driver's meson build scripts are currently broken, but autotools
still works fine. So add a little workaround for forcing it to be built
via autotools, even if meson.build file is there.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
9b3b938ede
os: re-add System() function for backwards compat
...
This function had been unexported long ago, but the Intel driver
still needs it. Adding a tiny temporary compat wrapper, so Intel
team has time to keep up with us.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
5b6f2292a3
os: drop win32-only System() function
...
Not used anymore, so it can be dropped.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
f837dee79d
win32: use lib system() instead of our own function
...
mingw32 has system() function, so we don't need our own implementation
anymore.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
6789b911c4
render: picturestr.h: re-add compat macros for intel driver
...
The ancient pict_f_vector and pixman_f_vector macros have been a fallout
from long ago transition to pixman and had been removed quite a while ago.
Unfortunately, Intel driver (which is not under our reign) hasn't been
kept up to date yet. So re-adding the compat macro, until Intel driver
team had a chance to keep up.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1778 >
2025-02-11 09:53:59 +01:00
3562298068
xquartz: fix inverted tablet pen Y tilt on macOS
...
On macOS, the y tilt behavior is inverted; an increase in the tilt
value corresponds to tilting the device away from the user.
see https://chromium-review.googlesource.com/c/chromium/src/+/2348544
see 0f128fd7c5/src/plugins/platforms/cocoa/qnsview_tablet.mm (L63)Fixes #792
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1779 >
2025-02-10 22:02:16 +08:00
da0de21b72
xquartz: fix compilation
...
Fixes #1788
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1777 >
2025-02-10 19:37:18 +08:00
2631cad85d
No longer leak FDs on VT switch.
...
Modified-by: Povilas Kanapickas <povilas@radix.lt>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/997 >
2025-02-08 17:38:20 +02:00
5677de586f
os/osinit: fix build when execinfo.h is missing
...
HAVE_BACKTRACE gets set on systems with libunwind however
execinfo.h may be missing (on *BSD or musl for example).
HAVE_EXECINFO_H has been defined by the build system for a long
time but previously not used.
Signed-off-by: Randy Palamar <randy@rnpnr.xyz>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1207 >
2025-02-08 15:22:29 +00:00
143a6a21e1
miext: rootless: fix unused variables
...
Compiling w/ -Werror (which our CI does) breaks compile due a buch of
unused variables.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1703 >
2025-02-08 15:15:57 +00:00
6ee163cb94
mi: miexpose: fix FTBS w/ rootless helper
...
FTBS when rootless subsys enabled:
> ../mi/miexpose.c: In function ‘miPaintWindow’:
> ../mi/miexpose.c:411:15: error: unused variable ‘orig_pWin’ [-Werror=unused-variable]
> 411 | WindowPtr orig_pWin = pWin;
> | ^~~~~~~~~
> cc1: all warnings being treated as errors
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1703 >
2025-02-08 15:15:57 +00:00
bcbc7479f3
xfree86: common: use LogMessageVerb() instead of xf86Msg()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
5fd918421a
xfree86: fbdevhw: use LogMessageVerb() instead of xf86Msg()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
4ba0cf1f55
xfree86: os-support: use LogMessageVerb() instead of xf86Msg()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
1f93ec5c33
xfree86: i2c: use LogMessageVerb() instead of xf86Msg()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
76874498be
glx: use LogMessageVerb() instead of xf86Msg()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
6fc4f35f62
xfree86: use LogMessageVerb() instead of xf86MsgVerb()
...
Both are doing same job, so no need to keep using an duplicated implementation.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1681 >
2025-02-07 22:40:06 +00:00
8e8bf63908
os: move AbortServer() to os/utils.c
...
This function doesn't much to do with logging, except for being
called once by FatalError(). It's better placed in utils.c
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1678 >
2025-02-07 18:45:35 +01:00
Enrico Weigelt, metux IT consult
Andy Myers
Aki Sakurai
Spiky Caterpillar
Randy Palamar