-
No functional change here. Part-of: <!1568>
baef6377 -
Instead of selectively undefining individual symbols with platform specific hacks, just rename the enum values. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1548>
83c94b73 -
Since kdrive is only working on Unix'es, we don't need to care whether that file might not exist on Windows, neither any need for doing so exlusively on Solaris. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1548>
1b49f6dd -
All supported platforms, except Linux, have have to call xf86InitVidMem() from xf86scanpci(), so we can make the code much simpler. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1548>
fd6fe366 -
The header exists on all supported platform, maybe some might still need to define _XOPEN_SOURCE to get pow() defined. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1548>
0a073172 -
Instead of keeping lots of special hacks for Solaris, just rename the field of struct i386_segment_regs, so we don't need special care anymore. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1548>
229a6953 -
DES isn't considered secure anymore for long time now, more and more platforms dropping DES from their RPC implementations, and even the one where it came from (Solaris) disabled it for a decade now. We have much better alternatives (eg. passing creds via Unix socket for local connections, ssh tunneling, MIT-MAGIC-COOKIE-1, ...), so it's unlikely anybody still really relying on it. Therefore, sweep it out. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1453>
71b207a2 -
It hasn't been used for really long time, the history even doesn't tell whether it ever had been used. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1500>
62a0a587 -
Leftover of dropping mfb about 1.5 decades ago. Fixes: f31bd087 Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1499>
98104c96 -
These are only needed inside one .c source, so no need to keep thm exported. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1498>
27babe03 -
The args have become dummies almost two decades ago, so no need to keep them any longer. Fixes: 0afeb024 Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1522>
4a27ff7a -
No reason to explicitly force it off on Windows - if dependencies are present, we can build it. Even though it will be rarely needed. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1534>
1ead6704 -
The current implementation silently disables Xnest, even if requested explcitly. That's a bit unstable behaviour - it's better to break the build in those cases, just disable silently in auto mode. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1534>
842e8661 -
Conditional on ipv6 around including <arpa/inet.h> isn't necessary. The only reason target not having it is win32, but it doesn't enable SIOCGLIFCONF anyways. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1539>
9d6a1d11 -
This #ifdef is always true, since two lines above it's already checked that TCPCONN is defined. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1539>
88b2257c -
We're currently not checking whether IPv6 is available before build, so it just gets silently disabled if AF_INET6 is missing - even when user wants to enable it explicitly. Adding explicit check into meson.build, so failing when ipv6 option enabled, but no IPv6 supported. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1539>
bc55a98d -
Since IPv6 is now only defined if IPv6 is really supported, we don't need to ifdef on AF_INET6 anymore. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1539>
f8057261 -
That's the only place where it is used, so no need to keep it in a public header. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1462>
cdb4d564 -
With removal of Xwin's NATIVEGDI (back a decade ago), the last caller is gone, and it also doesn't seem to be called by any driver. Fixes: 8465ee78 Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1504>
245f77cf -
This header isn't public and holds defines for code in os/ directory, so no need to keep it in the global header dir - it's probably better off in os/ directory - just like we already have with many others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1389>
487eb468 -
The comment about ClientPtr also being defined in misc.h is outdated for about 20 years now: XFree86 4.3.0.1 dropped the redundant definition over there. It's still also in Xdefs.h, so the guard is still needed. Fixes: d5682217 Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1514>
224fa46f -
The secur.h doesn't need the symbols defined here (eg. Status or Display) anymore, so no need to keep it around anymore. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1550>
c77bcb9f -
This used to be for panoramiXproto.h, but it doesn't look at it anymore, so no need to keep it around. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1551>
9c953472 -
This could potentially be security related or crash the server: XFixesSelectSelectionInput() calls the XACE_SELECTION_ACCESS hook with wrong parameter type: XID instead of pointer to Selection struct. It seems that it hadn't been kept up in XACE changed to polyinstantiation. When XACE is used (eg. Security or SELinux extension enabled), this can easily lead to memory corruptions at attacker-controlled locations, since the client-given XID is interpreted as the memory location of Selection structure. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
601fd0fd -
These aren't used by any drivers, so no need to keep them installed. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
22a3014c -
These aren't used by any drivers, so no need to keep them exported. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
3e0a4ab7 -
Overloading real function name with macro makes it unnecessarily complicated. Just rename the real function. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
632a48a0 -
The generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1556>
ae3c5733 -
The generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
9524ffee -
The generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
67e468c8 -
The generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
3dfe00d5 -
The generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
922b7685 -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
0f6bb23b -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
47d6c3ad -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1556>
51d8bcfc -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
305f2d59 -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
591d95c7 -
he generic XaceHook() call isn't typesafe (und unnecessarily slow). Better add an explicit function, just like we already have for others. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1556>
facdaae4 -
As soon as winapi headers are included, we're running into a name clash on UpdateColors(), since winapi has a function by the same name. Trivial fix simply renaming our own UpdateColors() function. ../dix/colormap.c:110:13: error: conflicting types for ‘UpdateColors’ 110 | static void UpdateColors(ColormapPtr /*pmap */ | ^~~~~~~~~~~~ In file included from /usr/share/mingw-w64/include/windows.h:71, from /usr/share/mingw-w64/include/winsock2.h:23, from /usr/i686-w64-mingw32/include/X11/Xwinsock.h:57, from ../os/osdep.h:138, from ../dix/colormap.c:57: /usr/share/mingw-w64/include/wingdi.h:3202:28: note: previous declaration of ‘UpdateColors’ was here 3202 | WINGDIAPI WINBOOL WINAPI UpdateColors(HDC hdc); | ^~~~~~~~~~~~ Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1351>
d8cbcfd6 -
Internal symbol, no need to have it in public header. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1351>
53e8ba3a -
Lots of fields from opaque.h aren't used by any drivers/modules and thus don't need to be exported at all. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <xorg/xserver!1351>
06b599ed -
There're lots of field that aren't used by any modules or drivers, thus no need to keep them exported. Signed-off-by:
Enrico Weigelt, metux IT consult <info@metux.net> Part-of: <!1351>
34662f15 -
struct hostent->h_addr_list is of type char**, not const char**. GCC considers this an error when in C99 mode or later. Signed-off-by:
Joaquim Monteiro <joaquim.monteiro@protonmail.com> Part-of: <!1572>
0ddcd878 -
When IPv6 support isn't enabled, and h_addr isn't defined, there is no for loop, so the break statement is invalid. Signed-off-by:
Joaquim Monteiro <joaquim.monteiro@protonmail.com> Part-of: <!1572>
a6a993f9 -
Never used. Part-of: <!1570>
ed575eee -
If the manual redirection was done by xwl_present_maybe_redirect_window, we are in control. This allows xwl_present_maybe_redirect_window to work as intended again. Previously, xwl_window_update_surface_window would ignore the window redirected by xwl_present_maybe_redirect_window, which would result in xwl_present_maybe_redirect_window undoing the redirection and bailing. Part-of: <!1570>
e1b8a121 -
If the surface window already uses automatic redirection, we can upgrade to manual redirection and save some blits with common use cases. This fixes a minor performance regression from a65bb848 ('Revert "xwayland/glamor: Avoid implicit redirection with depth 32 parent windows"') with mutter >= 44. Part-of: <!1570>
8e8bc0a1 -
nerdopolis authored
This fixes a hang on simpledrm where min_cursor_height and min_cursor_width is never established, and drmmode_load_cursor_argb_check would infinitely when the minimum values where 0 or less.
05c63d2a
- .gitlab-ci.yml 4 additions, 5 deletions.gitlab-ci.yml
- Xext/meson.build 0 additions, 1 deletionXext/meson.build
- Xext/panoramiX.h 1 addition, 2 deletionsXext/panoramiX.h
- Xext/saver.c 3 additions, 6 deletionsXext/saver.c
- Xext/securitysrv.h 0 additions, 19 deletionsXext/securitysrv.h
- Xext/shm.c 2 additions, 2 deletionsXext/shm.c
- Xext/sync.c 0 additions, 1 deletionXext/sync.c
- Xext/xace.c 78 additions, 124 deletionsXext/xace.c
- Xext/xace.h 37 additions, 22 deletionsXext/xace.h
- Xi/exevents.c 8 additions, 8 deletionsXi/exevents.c
- Xi/grabdevb.c 1 addition, 1 deletionXi/grabdevb.c
- Xi/grabdevk.c 1 addition, 1 deletionXi/grabdevk.c
- Xi/listdev.c 1 addition, 1 deletionXi/listdev.c
- Xi/xiquerydevice.c 2 additions, 2 deletionsXi/xiquerydevice.c
- composite/compext.c 3 additions, 3 deletionscomposite/compext.c
- composite/compinit.c 2 additions, 0 deletionscomposite/compinit.c
- dbe/dbe.c 1 addition, 1 deletiondbe/dbe.c
- dbe/midbe.c 3 additions, 3 deletionsdbe/midbe.c
- dix/colormap.c 6 additions, 6 deletionsdix/colormap.c
- dix/cursor.c 3 additions, 3 deletionsdix/cursor.c