Commit Graph

942 Commits

Author SHA1 Message Date
Enrico Weigelt, metux IT consult fcda278666 (!1678) 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>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult dae32195a0 (submit/rename-panoramix-sym) rename old symbol PANORAMIX to XINERAMA
PANORAMIX was the original working title of the extension, before it became
official standard. Just nobody cared about fixing the symbols to the official
naming.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult 5edcb5857a (submit/drop-cygwin) drop remains of cygwin support
Cygwin support doesn't seem to be used anymore, so it can be dropped.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult d94c68bb07 (submit/unexport-ddx-callbacks) os: rename ddx.h to ddx_priv.h
Make it clear that stuff from this file really isn't supposed to be used
by dynamically loaded modules like drivers.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult 5d107406f5 (submit/unexport-ddx-callbacks) os: unexport ddx callbacks
The DDX callbacks (where core/DIX calls into DDX) aren't supposed to be
called by drivers directly, so unexport them.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult a25156ed54 (submit/fix-char-signedness) os: utils: fix char signess mismatch
On NetBSD gives warning:

In file included from /usr/include/ctype.h:100,
                 from ../include/misc.h:174,
                 from ../os/utils.c:75:
../os/utils.c: In function ‘VerifyDisplayName’:
../os/utils.c:624:23: warning: array subscript has type ‘char’ [-Wchar-subscripts]
  624 |         if (!isdigit(d[i])) {
      |                       ^
../os/utils.c: In function ‘ProcessCommandLine’:
../os/utils.c:942:44: warning: array subscript has type ‘char’ [-Wchar-subscripts]
  942 |             if ((i + 1 < argc) && (isdigit(*argv[i + 1])))
      |                                            ^

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult be26735ec3 (submit/fix-char-signedness) os: access: fix char signess mismatch
On NetBSD gives warning:

In file included from /usr/include/ctype.h:100,
                 from ../include/misc.h:174,
                 from ../os/access.c:96:
../os/access.c: In function ‘ResetHosts’:
../os/access.c:981:49: warning: array subscript has type ‘char’ [-Wchar-subscripts]
  981 |                 lhostname[i] = tolower(ohostname[i]);
      |                                                 ^
2024-10-04 13:49:05 +02:00
Enrico Weigelt, metux IT consult 9e1a591235 (submit/drop-obsolete-have-dix-config.h) drop obsolete HAVE_DIX_CONFIG_H
The symbol controls whether to include dix-config.h, and it's always set,
thus we don't need it (and dozens of ifdef's) anymore.

This commit only removes them from our own source files, where we can
guarantee that dix-config.h is present - leaving the (potentially exported)
headers untouched.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult d2508ef824 (!1279) os: don't include client.h anymore
There's no need to include client.h anymore. But still leaving it, in case
some external consumer relying on it's presence.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 05f026ae8d (!1279) os: unexport client id retrieval functions
These aren't used by any (known) external modules, thus no need to export them.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 2f6d479dea (!1279) os: unexport DetermineClientPid() and DetermineClientCmd()
These aren't used by any drivers/modules, so no need to export them.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult e1e588114d (!1279) os: split off internal definitions from client.h
The client.h file is part of the public module API, but it also contains
definitions that aren't useful for being used in modules. Splitting them
out into their own client_priv.h file, which isn't part of the API.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 42b6f20bc0 (!1578) os: fix link failure on Illumos
On SunOS, the BSD socket API as well as hostname lookups isn't
implemented in libc, but separate libraries. We need to link them
explicitly.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 600809b13d (!1578) os: fix FTBS on Illumos due missing symbols
Several feature defines need to be set before including system headers,
otherwise build breaks:

> /usr/include/X11/Xtrans/Xtranssock.c: In function '_XSERVTransSocketRead':
> /usr/include/X11/Xtrans/Xtranssock.c:2161:14: error: 'struct msghdr' has no member named 'msg_control'
>  2161 |             .msg_control = cmsgbuf.buf,
>       |              ^~~~~~~~~~~
> /usr/include/X11/Xtrans/Xtranssock.c:2162:14: error: 'struct msghdr' has no member named 'msg_controllen'
>  2162 |             .msg_controllen = CMSG_LEN(MAX_FDS * sizeof(int))
>       |              ^~~~~~~~~~~~~~
>
> ../os/access.c:339:14: error: implicit declaration of function 'asprintf'; did you mean 'Xasprintf'? [-Werror=implicit-function-declaration]
>   339 |     length = asprintf(addr, "%s%c%s", type, delimiter, value);
>       |              ^~~~~~~~
>       |              Xasprintf

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult d96caa88b1 (submit/miext-extinit) Xext: geext: move out noGEExtension to corresponding extension
The OS abstraction isn't really the right place for those flags,
they are're probably better off in their corresponding extensions.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 1981b00776 (submit/miext-extinit) xfree86: vidmode: unexport noXFree86VidModeExtension and move into extension
This isn't needed by any external module, so no need to export it.
And those flags are better off in the corresponding extension,
instead of the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 8a4ce974fb (submit/miext-extinit) xfree86: dga: unexport noXFree86DGAExtension and move into extension
This isn't needed by any external module, so no need to export it.
And those flags are better off in the corresponding extension,
instead of the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 26fa974ed7 (submit/miext-extinit) xfree86: dri2: unexport noXFree86DRI2Extension and move into extension
This isn't needed by any external module, so no need to export it.
And those flags are better off in the corresponding extension,
instead of the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 83095294e7 (submit/miext-extinit) xfree86: dri: unexport noXFree86DRIExtension and move into extension
This isn't needed by any external module, so no need to export it.
And those flags are better off in the corresponding extension,
instead of the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 3aca7710d4 (submit/miext-extinit) miext: move over extinit_priv.h from include
Since most of the extension init logic (and on/off switches for them)
is driven from miext, this seems the appropriate place for the header.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 29c2aefae9 (submit/miext-extinit) composite: move noCompositeExtension into the extension code
This flag is better off in the extension code instead of the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult f891f802c5 (submit/miext-extinit) Xext: shape: unexport noXFixesExtension
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult d94808e639 (submit/miext-extinit) Xext: xf86bigfont: unexport noXFree86BigfontExtension
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult c204fb44a1 (submit/miext-extinit) Xext: security: unexport noSecurityExtension
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult afff450524 (submit/miext-extinit) Xext: shm: unexport noMITShmExtension
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 63dd127157 (submit/miext-extinit) Xext: saver: unexport noScreenSaverExtension
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult c64602f252 (submit/miext-extinit) Xext: shape: unexport noShapeExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 88a305bb78 (submit/miext-extinit) res: unexport noResExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult f00014b01e (submit/miext-extinit) render: unexport noRenderExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 69b8c84d2e (submit/miext-extinit) randr: unexport noRRExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 3568fa3dd3 (submit/miext-extinit) panoramix: move noPanoramiXExtension field into extension.
This field is better of in the corresponding extension,
than in the OS layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 9b9c59c4d9 (submit/miext-extinit) xtest: unexport noTestExtensions field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 738edafc74 (submit/miext-extinit) glx: unexport noGlxExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 206ac3f069 (submit/miext-extinit) dpms: unexport noDPMSExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult 6003879200 (submit/miext-extinit) dbe: unexport noDbeExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult d9cb9b2543 (submit/miext-extinit) damageext: unexport noDamageExtension field
This field isn't used by any drivers, and also better belongs into
the corresponding extension instead of OS adaption layer.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:04 +02:00
Enrico Weigelt, metux IT consult a60613aefb (submit/recv-fds) os: read file descriptors into client struct at once
Instead of having the request handler ask for fd's one by one, just read them
all into a little array in ClientRec struct. And also automatically clean up
after request had been handled.

Request handlers need to set the entries to -1 if they shouldn't be closed
automatically.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:03 +02:00
Enrico Weigelt, metux IT consult a04f251618 (!1677) os: drop obsolete LogHdrMessage()
This function isn't used anywhere, so no need to keep it around.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:02 +02:00
Enrico Weigelt, metux IT consult 3d6de2d17b (!1682) os: log: consolidate OS specific fsync() call into helper
Instead of having lots of #ifdef's, consolidating the conditionally
compiled fsync() call into a tiny inline helper.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult 0c2fd28cc6 (!1682) os: log: drop now meaningless XLOG_FLUSH option
Since we're not indirectly writing via FILE anymore, this option has
become meaningless: it meant flushing out our in-process buffer to
the kernel, but we're now doing direct write() calls anyways.

xf86 still accepts the "flush" config file flag for backwards compatibility,
but it hasn't any practical meaning anymore.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult e3839084b6 (!1682) os: log via fd instead of FILE
Instead of maintaining both the logfile fd, as well as ANSI FILE pointer,
simplify it to just a fd.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult bbedc2b4c9 (!1683) os: replace LogVWrite() by LogVMessageVerb()
It's just a wrapper around LogVMessageVerb() and no external module
using it, so can easily be optimized-away.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult 348f0c3997 (!1683) os: replace LogWrite() by LogMessageVerb()
It's nothing but a wrapper, doing the same as LogMessageVerb(X_NONE, ...),
and no external module / driver needs it, so can be easily optimized away.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult 2af0764558 (!1685) os: log: use localtime_r() on mingw builds
MingW also provides localtime_r(), but needs _POSIX_THREAD_SAFE_FUNCTIONS
symbol before including anything, in order for the prototype being defined.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Enrico Weigelt, metux IT consult 52c3796122 (!1686) os: drop redefining getpid() on mingw32
mingw32 does does have getpid() function, so conditionally redefining
it to _getpid() isn't necessary.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2024-10-04 13:49:01 +02:00
Olivier Fourdan 57a446c0f9 os: Fix NULL pointer dereference
RemoveHost() can be called from DisableLocalHost() with a NULL client,
but doesn't actually check whether the given client pointer is valid on
error and assigns the error value unconditionally, leading to a possible
NULL pointer dereference and a crash of the Xserver.

To avoid the issue, simply check whether the client pointer is not NULL
prior to assign the errorValue.

Closes: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1752
See-also: https://bugzilla.redhat.com/2313799
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1701>
2024-09-23 09:36:52 +02:00
Enrico Weigelt, metux IT consult 1d7cb4b2c3 os: unexport CloseDownConnection()
Not used by any drivers, so no need to export it.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1582>
2024-09-02 16:02:39 +00:00
Matthieu Herrb 59f5445a7f Return NULL in *cmdname if the client argv or argv[0] is NULL
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/995>
2024-09-01 22:59:26 +00:00
Matthieu Herrb a8512146ba Don't crash if the client argv or argv[0] is NULL.
Report from  bauerm at pestilenz dot org.

Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/995>
2024-09-01 22:59:26 +00:00
Enrico Weigelt, metux IT consult 811977228b os: unexport screen saver timer functions
These functions aren't supposed to be used by drivers, so move them
out of the public API.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1583>
2024-09-01 20:23:10 +00:00