Commit Graph

20106 Commits

Author SHA1 Message Date
Enrico Weigelt, metux IT consult 08c6bbf32b (1881) Xace: remove remains of XACE_CORE_DISPATCH
There's no actual caller of this hook - removed almost a decade ago
(see commit 6cb34816af), but some remains
had been forgotten to clean up.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 7e7ca1b5ae (1882) xwin: drop unused variable WindowsDRIReqCode
This variable is assigned once, but never used.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 673dc8fb18 (1883) Xquartz: drop unused variable DRIReqCode
This variable is assigned once, but never used.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 8c2b016ae9 (1884) xfree86: dri: drop unused variable DRIReqCode
This variable is assigned once, but never used.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 88304840c1 (1886) XI: drop now obsolete SReplyIDispatch()
nothing else to do for it anymore, since all previous callers now have been
migrated to do the byte-swapping directly, so it can be dropped entirely.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 3d2d9c4495 (1886) XI: directly write out X_XIGetFocus reply
Write out the X_XIGetFocus reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 9b80c56198 (1886) XI: directly write out X_XIGetSelectedEvents reply
Write out the X_XIGetSelectedEvents reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 5a5dddbdec (1886) XI: directly write out X_XIGetProperty reply
Write out the X_XIGetProperty reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult f968077100 (1886) XI: directly write out X_XIListProperties reply
Write out the X_XIListProperties reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult d0de2cdf4d (1886) XI: directly write out X_XIPassiveGrabDevice reply
Write out the X_XIPassiveGrabDevice reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 2dbe9854e9 (1886) XI: directly write out X_XIGrabDevice reply
Write out the X_XIGrabDevice reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 02f9aef5ee (1886) XI: directly write out X_XIQueryDevice reply
Write out the X_XIQueryDevice  reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 9aca1a404e (1886) XI: directly write out X_XIQueryVersion reply
Write out the X_XIQueryVersion reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult ab84bed41c (1886) XI: directly write out X_XIGetClientPointer reply
Write out the X_XIGetClientPointer reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 0ac34c0c44 (1886) XI: directly write out X_XIQueryPointer reply
Write out the X_XIQueryPointer reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult b294a16845 (1886) XI: directly write out X_GetDeviceProperty reply
Write out the X_GetDeviceProperty reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult dc58917c5c (1886) XI: directly write out X_ListDeviceProperties reply
Write out the X_ListDeviceProperties reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult ce001bc635 (1886) XI: directly write out X_ChangeDeviceControl reply
Write out the X_ChangeDeviceControl reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult f0b358a3b9 (1886) XI: directly write out X_GetDeviceControl reply
Write out the X_GetDeviceControl reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult e87f7b4c7a (1886) XI: directly write out X_SetDeviceValuators reply
Write out the X_SetDeviceValuators reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 2f66205552 (1886) XI: directly write out X_QueryDeviceState reply
Write out the X_QueryDeviceState reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 6ac6ace79d (1886) XI: directly write out X_SetDeviceButtonMapping reply
Write out the X_SetDeviceButtonMapping reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 2d9b3175c9 (1886) XI: directly write out X_GetDeviceButtonMapping reply
Write out the X_GetDeviceButtonMapping reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 2b790ff842 (1886) XI: directly write out X_SetDeviceModifierMapping reply
Write out X_SetDeviceModifierMapping the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 356e0a5f03 (1886) XI: directly write out X_GetDeviceModifierMapping reply
Write out X_GetDeviceModifierMapping the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 1a3dd73155 (1886) XI: directly write out X_GetDeviceKeyMapping reply
Write out X_GetDeviceKeyMapping the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 81b5b6f379 (1886) XI: directly write out X_GetFeedbackControl reply
Write out X_GetFeedbackControl the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 690dcd25e7 (1886) XI: directly write out X_GetDeviceFocus reply
Write out X_GetDeviceFocus the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult ad8333683f (1886) XI: directly write out X_GrabDevice reply
Write out X_GrabDevice the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult ccca3bd2bf (1886) XI: directly write out X_GetDeviceMotionEvents reply
Write out X_GetDeviceMotionEvents the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 4bc7f7d953 (1886) Xi: declare variables where needed in ProcXGetDeviceMotionEvents()
Improve code readability by moving variable declarations to where
they're actually needed.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 6355930427 (1886) XI: drop extra temp variable in ProcXGetDeviceMotionEvents()
The value is only checked once, so no need to copy it to a temporary
variable first.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 504b7cb86a (1886) XI: use SwapLongs() in ProcXGetDeviceMotionEvents()
We already have a standard function for swapping 32 bit ints,
so let's use this one instead of rolling our own loop.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 7afacdf1b2 (1886) XI: directly write out X_GetDeviceDontPropagateList reply
Write out X_GetDeviceDontPropagateList the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 4d16420692 (1886) XI: directly write out X_GetSelectedExtensionEvents reply
Write out X_GetSelectedExtensionEvents the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 0c5c4f9f89 (1886) XI: directly write out X_SetDeviceMode reply
Write out X_SetDeviceMode the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 91bfcb98fd (1886) XI: directly write out X_OpenDevice reply
Write out X_OpenDevice the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 5f94953789 (1886) XI: directly write out X_ListInputDevices reply
Write out X_ListInputDevices the reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 476ef09ce8 (1886) XI: directly write out X_GetExtensionVersion reply
Write out the X_GetExtensionVersion reply directly (and do the swapping
within the request handler) instead of going through separate callback
that's having demux the replies again.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult ece1ac6a9a (1886) XI: canonicalize reply struct declarations and naming
* name all of them "rep", as throughout most of the Xserver codebase
* always declare them where initialized

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult d82380613e (1886) dix: let change_modmap() return BadValue instead of -1
The function returns X result codes, but -1 isn't a valid value here.
Therefore all callers explicitly translate -1 to BadValue, so we can return
that directly instead of -1.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult dda861fb39 (1888) present: simplify dispatching
Using simple case statements with actual opcode define's instead
of call vector arrays where the exact order matters.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 233203007b (1890) xfixes: write out GetCursorName reply directly
The request handler already does byte-swapping on its own, and
there's no extra reply-swap handler for it, so no need to call
into callbacks here.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 547d38448d (1889) xwin: fix int size mismatch
> ../hw/xwin/winshadgdi.c: In function ‘winBltExposedWindowRegionShadowGDI’:
> ../hw/xwin/winshadgdi.c:866:78: warning: format ‘%x’ expects argument of type ‘unsigned int’, but argument 2 has type ‘DWORD’ {aka ‘long unsigned int’} [-Wformat=]
>   866 |             ErrorF("winBltExposedWindowRegionShadowGDI - BitBlt failed: 0x%08x\n",
>       |                                                                           ~~~^
>       |                                                                              |
>       |                                                                              unsigned int
>       |                                                                           %08lx
>   867 |                    GetLastError());
>       |                    ~~~~~~~~~~~~~~
>       |                    |
>       |                    DWORD {aka long unsigned int}

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 11c1cbcb23 (1889) xwin: fix name clash between Xserver and Windows headers
> ../hw/xwin/winscrinit.c: In function ‘winFinishScreenInitFB’:
> ../hw/xwin/winscrinit.c:381:18: error: ‘struct _Screen’ has no member named ‘CreateWindowA’; did you mean ‘CreateWindow’?
>   381 |         pScreen->CreateWindow = winCreateWindowRootless;
>       |                  ^~~~~~~~~~~~
> ../hw/xwin/winscrinit.c:405:18: error: ‘struct _Screen’ has no member named ‘CreateWindowA’; did you mean ‘CreateWindow’?
>   405 |         pScreen->CreateWindow = winCreateWindowMultiWindow;
>       |                  ^~~~~~~~~~~~

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 2e77f0fc41 (1889) xwin: fix unitialized variables
> ../hw/xwin/winclipboard/xevents.c: In function ‘winClipboardSelectionNotifyData.constprop’:
> ../hw/xwin/winclipboard/xevents.c:313:23: warning: ‘codepage’ may be used uninitialized in this function [-Wmaybe-uninitialized]
>   313 |     int iUnicodeLen = MultiByteToWideChar(codepage, 0,
>       |                       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>   314 |                                           pszReturnData, -1, NULL, 0);
>       |                                           ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> ../hw/xwin/winclipboard/xevents.c: In function ‘winClipboardFlushXEvents’:
> ../hw/xwin/winclipboard/xevents.c:550:35: warning: ‘codepage’ may be used uninitialized in this function [-Wmaybe-uninitialized]
>   550 |             int iConvertDataLen = WideCharToMultiByte(codepage, 0,
>       |                                   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>   551 |                                                       (LPCWSTR) pszGlobalData, -1,
>       |                                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>   552 |                                                       NULL, 0, NULL, NULL);
>       |                                                       ~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 086c081375 (1889) xkb: fix printf conversion error on Windows
> ../xkb/xkb.c: In function ‘_XkbSetMapCheckLength’:
> ../xkb/xkb.c:2440:53: warning: unknown conversion type character ‘z’ in format [-Wformat=]
>  2440 |     ErrorF("[xkb] BOGUS LENGTH in SetMap: expected %zd got %zd\n", len, req_len);
>       |                                                     ^
> ../xkb/xkb.c:2440:61: warning: unknown conversion type character ‘z’ in format [-Wformat=]
>  2440 |     ErrorF("[xkb] BOGUS LENGTH in SetMap: expected %zd got %zd\n", len, req_len);
>       |                                                             ^
> ../xkb/xkb.c:2440:12: warning: too many arguments for format [-Wformat-extra-args]
>  2440 |     ErrorF("[xkb] BOGUS LENGTH in SetMap: expected %zd got %zd\n", len, req_len);
>       |            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 7728c50050 (1889) xwin: fix missing include of windsock2.h
Windows headers are pretty nitpicking about include order:

> In file included from /usr/i686-w64-mingw32/include/X11/Xwinsock.h:57,
>                  from /usr/i686-w64-mingw32/include/xcb/xcb_windefs.h:34,
>                  from /usr/i686-w64-mingw32/include/xcb/xcb.h:41,
>                  from ../hw/xwin/winmultiwindowicons.c:43:
> /usr/share/mingw-w64/include/winsock2.h:15:2: warning: #warning Please include winsock2.h before windows.h [-Wcpp]

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult f1bc7745ab (1892) xkb: XkbInitRules() don't hard-crash the server on strdup() fail
No need to hard-crash the whole server if some strdup() calls failing,
those fields already may be NULL, so consumers can handle that situation.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00
Enrico Weigelt, metux IT consult 9284d82d03 (1892) xkb: ddxLoad: don't crash the server when strcpy() fails
No need for RunXkbComp() to hard-crash (by calling XNFstrdup()) the server
if strdup() fails to allocate more memory - it's callers already handling
the situation gracefully.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
2025-06-03 11:37:35 +02:00