aboutsummaryrefslogtreecommitdiff
path: root/nx-X11/programs
Commit message (Collapse)AuthorAgeFilesLines
* sendexev.c: Fix return of unitialized variableUlrich Sibiller2016-10-201-3/+7
| | | | | Well, kinda fix. Upstream handles that differently here, we need a thorough check! Add another FIXME comment because of this.
* xfixes: Remove unused variable.Cyril Brulebois2016-10-201-2/+0
| | | | | | | | | The last use of pScreen in ProcXFixesSetPictureClipRegion was removed in aa7096ca6f108e399d9916639cf20c57f9776305 so remove it entirely. Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Signed-off-by: Cyril Brulebois <kibi@debian.org> Backported-to-NX-by: Ulrich Sibiller <uli42@gmx.de>
* Remove useless line of code that contained a bug and triggered a gcc ↵Benjamin Herrenschmidt2016-10-201-2/+0
| | | | warning. This variable will be overriden before being used anyway. (Bugzilla #5595)
* dix: fix cast from pointer to integerJulien Cristau2016-10-201-1/+1
|
* dbe: Call to DDX SwapBuffers requires address of int, not unsigned int ↵Keith Packard2016-10-201-3/+4
| | | | | | | | | | | | | | | | | | [CVE-2014-8097 pt. 2] When the local types used to walk the DBE request were changed, this changed the type of the parameter passed to the DDX SwapBuffers API, but there wasn't a matching change in the API definition. At this point, with the API frozen, I just stuck a new variable in with the correct type. Because we've already bounds-checked nStuff to be smaller than UINT32_MAX / sizeof(DbeSwapInfoRec), we know it will fit in a signed int without overflow. Signed-off-by: Keith Packard <keithp@keithp.com Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Backported-to-NX-by: Ulrich Sibiller <uli42@gmx.de>
* xkb.c: fix 'assignment makes integer from pointer without a cast'Ulrich Sibiller2016-10-201-1/+1
| | | | | | xkb.c: In function ‘_GetCountedString’: xkb.c:4405:8: warning: assignment makes integer from pointer without a cast [enabled by default] len= (CARD16 *)wire;
* xkmread.c: use _X_UNUSED to avoid compiler warning when result is not neededUlrich Sibiller2016-10-201-4/+7
|
* xcmisc.c: remove HAVE_STDINTUlrich Sibiller2016-10-201-4/+0
| | | | | This eliminates a warning since we do not have this define in our build environment.
* security.c: fix compiler warningsUlrich Sibiller2016-10-201-2/+1
|
* Window.c: Fix function assignment warningUlrich Sibiller2016-10-201-1/+1
| | | | | | Window.c: In function ‘nxagentFrameBufferPaintWindow’: Window.c:1968:31: warning: ISO C forbids assignment between function pointer and ‘void *’ [-Wpedantic] PaintWindowBackgroundBackup = pWin->drawable.pScreen -> PaintWindowBackground;
* Display.c: remove unused variableUlrich Sibiller2016-10-201-2/+0
| | | | | | | Display.c: In function ‘nxagentCheckForPixmapFormatsCompatibility’: Display.c:2471:8: warning: variable ‘one_match’ set but not used [-Wunused-but-set-variable] bool one_match = false; ^
* Colormap.c: add cast to avoid warningUlrich Sibiller2016-10-201-1/+1
| | | | | | | Colormap.c: In function ‘nxagentSetInstalledColormapWindows’: Colormap.c:314:27: warning: format ‘%p’ expects argument of type ‘void *’, but argument 3 has type ‘WindowPtr’ [-Wformat=] pWin, pWin -> drawable.class); ^
* Args.c: fix string format warningUlrich Sibiller2016-10-201-2/+2
| | | | warning: format ‘%u’ expects argument of type ‘unsigned int’, but argument 5 has type ‘long int’ [-Wformat=]
* rename xf86bigfstr.h to xf86bigfproto.hUlrich Sibiller2016-10-192-2/+2
|
* nx-X11/programs/Xserver/GL/glx/Imakefile: typo fix, leading to ↵Mihai Moldovan2016-10-141-1/+1
| | | | | | glcontextmodes.c not being compiled. The typo didn't cause immediate problems.
* hw/nxagent/{Clipboard|Events}.c: Typo fix in error messages.Mike Gabriel2016-10-132-3/+3
|
* Remove CRAY supportUlrich Sibiller2016-10-101-6/+1
| | | | | This basically repeats commit bad67799229b94ea2ba0174319949766ad1c2fc6 after the upgrade to libX11 1.3.4.
* xserver: add missing xkbfile.hUlrich Sibiller2016-10-081-0/+484
| | | | | This should have been part of commit 2c31ce2 which also happens to have a wrong description... Sigh.
* xserver: adapt xkbsrc includes syntax to match upstreamUlrich Sibiller2016-10-065-5/+5
|
* os/access: fix regression in server interpreted authDave Airlie2016-10-061-1/+1
| | | | | | | | | | | | | | | | This was reported on irc on Fedora when rawhide went to 1.17.1. regression occured in: 2566835b4374edb3e5a8353d4f7c9e7ec4851c57 os: Eliminate uninitialized value warnings from access.c siAddrMatch doesn't need addr to be a useful value, it checks some things like localuser without having an address at all. Signed-off-by: Dave Airlie <airlied@redhat.com> Tested-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> v2: backport to nx-libs 3.6.x (Ulrich Sibiller)
* os: Eliminate uninitialized value warnings from access.cKeith Packard2016-10-061-8/+8
| | | | | | | | | | | | The ConvertAddr function doesn't reliably set the 'addr' return value, and so callers are getting flagged for using potentially uninitialized values. Initialize the value in the callers to NULL and then go ahead and check for NULL values before using them. Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> v2: backport to nx-libs 3.6.x (Ulrich Sibiller)
* dix: GetHosts bounds check using wrong pointer value [CVE-2014-8092 pt. 6]Keith Packard2016-10-061-1/+1
| | | | | | | | | | | | | | | GetHosts saves the pointer to allocated memory in *data, and then wants to bounds-check writes to that region, but was mistakenly using a bare 'data' instead of '*data'. Also, data is declared as void **, so we need a cast to turn it into a byte pointer so we can actually do pointer comparisons. Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> v1: Keith Packard v2: backport to nx-libs 3.6.x (Ulrich Sibiller)
* Fix multiple warnings in os/xdmauth.c.Eamon Walsh2016-10-061-10/+12
| | | | | v1: Eamon Walsh v2: backport to nx-libs 3.6.x (Ulrich Sibiller)
* replace transport.c link by xstrans.cUlrich Sibiller2016-10-068-14/+39
| | | | Just like upstream does
* xserver: drop XKBSRV_NEED_FILE_FUNCSUlrich Sibiller2016-10-0617-20/+5
| | | | just as upstream did
* xserver: remove define XKB_IN_SERVERUlrich Sibiller2016-10-0615-370/+1
| | | | it is no longer needed because the code here is only used in the server.
* xserver: use own copy of XKBgeom.h headerUlrich Sibiller2016-10-0611-16/+671
| | | | just like upstream does
* xserver: use own copy of XKBrules.h headerUlrich Sibiller2016-10-064-5/+3
| | | | just like upstream does
* xserver: use own copy of XKBrules.h headerUlrich Sibiller2016-10-062-1/+200
| | | | just like upstream does
* xserver: use own copy of XKBstr.h headerUlrich Sibiller2016-10-069-8/+615
| | | | just like upstream does
* xserver: use own copy of XKBsrv.h headerUlrich Sibiller2016-10-0640-38/+1228
| | | | just like upstream does
* copy files from libX11 to xserver dirUlrich Sibiller2016-10-065-6/+3506
| | | | | | | | Before those files where symlinked in the makefile. While trying to upgrade libX11 to 1.3.4 it turned out that the symlinked did not work in the xserver anymore. Upstream has copies of those files, too.
* randrproto: It was discovered that nx-libs shipped two different version of ↵Mike Gabriel2016-10-014-1371/+3
| | | | randr.h and randrproto.h. This commit leaves the newer file versions in the package and let's the build process use those everywhere.
* hw/nxagent/X11/include/Xrandr_nxagent.h: Reduce injected libXrandr(_nxagent) ↵Mike Gabriel2016-10-011-22/+0
| | | | header file, so that it only contains the XRRScreenChangeNotifyEvent typedef.
* Fix FTBFS with gcc 4.4Ulrich Sibiller2016-10-011-2/+0
| | | | | | | This fixes ArcticaProject/nx-libs#204. However, this fix does not address the root of the problem, it simply avoids it. We could run into it again... See bugreport for an explanation.
* Revert "Fix wrong include"Ulrich Sibiller2016-09-241-7/+0
| | | | | | | After commit cbf3ef0f219a7d5c94a7f5d5b46c9a0e26cc6588 this no longer necessary. This reverts commit 9a4bb50af54ced8d53112b8fa3b5b9dc3b481224.
* remove misc dirUlrich Sibiller2016-09-213-5/+117
| | | | xserver upstream has included the required files. So do we.
* move xkblib to xserverUlrich Sibiller2016-09-218-7/+7399
| | | | | Upstream does not use xkbfile but has own copies of the required files. It was not used elsewhere.
* hw/nxagent/Init.c: Fix copy+paste flaw when re-arranging copyright ↵Mike Gabriel2016-08-241-2/+2
| | | | information as display by nxagent.
* Fix pasting the clipboard directly after session start.Ulrich Sibiller2016-08-241-0/+9
|
* nxagent.1 man page: Some layout fixes for string options with well-defined ↵Mike Gabriel2016-07-281-0/+5
| | | | expected selections of values.
* hw/nxagent/Display.c: Fix comment where the meaning of the "bypass" ↵Mike Gabriel2016-07-281-3/+2
| | | | reconnectchecks parameter value is explained.
* nxagent.1 man page: Add explanation about the new reconnectchecks nx/nx option.Mike Gabriel2016-07-281-0/+26
|
* hw/nxagent/Args.c: Follow-up fix for PR #183: expect reconnectchecks=bypass ↵Mike Gabriel2016-07-281-1/+1
| | | | instead of =none.
* nx-X11/programs/Xserver/hw/nxagent/Display.c: add FIXME comment to ↵Mihai Moldovan2016-07-281-0/+1
| | | | | | nxagentCheckForColormapsCompatibility(). Maybe needs special strictness handling as well.
* nx-X11/programs/Xserver/hw/nxagent/Display.c: add FIXME comment to ↵Mihai Moldovan2016-07-281-0/+1
| | | | | | nxagentInitAndCheckVisuals(). Maybe needs special strictness handling as well.
* nx-X11/programs/Xserver/hw/nxagent/Display.c: split up ↵Mihai Moldovan2016-07-281-5/+9
| | | | | | | | | | | | | | | nxagentCheckForPixmapFormatsCompatibility() and nxagentInitPixmapFormats(). Don't implicitly call the checking function in the init function (and throw away the checking functions result...) Instead, explicitly use the checking function after the init function in other parts of the code and throw away the checking functions return value selectively or actually use it while reconnecting. This is totally different behavior compared to the original one. Hopefully it doesn't cause any problems (i.e., hopefully there was no good reason for never using the checking functions return value, but an oversight.)
* nx-X11/programs/Xserver/hw/nxagent/Display.c: use new ReconnectTolerance ↵Mihai Moldovan2016-07-281-23/+105
| | | | | | | | | | | | | | | | | | | | | | | | | | | | nxagentOption value in nxagentCheckForPixmapFormatsCompatibility() and modify behavior based on this value. Recognized values: - Strict means that the number of internal and external pixmap formats must match exactly and every internal pixmap format must be available in the external pixmap format array. - Safe means that the number of pixmap formats might diverge, but all internal pixmap formats must also be included in the external pixmap formats array. This is recommended, because it allows clients with more pixmap formats to still connect, but not lose functionality. - Risky means that the internal pixmap formats array is allowed to be smaller than the external pixmap formats array, but at least one pixmap format must be included in both. This is potentially unsafe. - Bypass or higher means that all of these checks are essentially deactivated. This is a very bad idea. Note that the default ReconnectTolerance value is still Strict. Also, the return value of this function is currently not used or passed through to other functions at all. An upcoming commit will change this behavior and actually enable the checks.
* nx-X11/programs/Xserver/hw/nxagent/Display.c: use new ReconnectTolerance ↵Mihai Moldovan2016-07-281-24/+118
| | | | | | | | | | | | | | | | | | | | | | nxagentOption value in nxagentCheckForDepthsCompatibility() and modify behavior based on this value. Recognized values: - Strict means that the number of old and new depths must match exactly and every old depth value must be available in the new depth array. - Safe means that the number of depths might diverge, but all former depth must also be included in the new depth array. This is recommended, because it allows clients with more depths to still connect, but not lose functionality. - Risky means that the new depths array is allowed to be smaller than the old depths array, but at least one depth value must be included in both. This is potentially unsafe. - Bypass or higher means that all of these checks are essentially deactivated. This is a very bad idea. Note that the default ReconnectTolerance value is still Strict.
* nx-X11/programs/Xserver/hw/nxagent/Display.c: use new ReconnectTolerance ↵Mihai Moldovan2016-07-281-1/+42
| | | | | | | | | | | | | | | | | | | nxagentOption in nxagentCheckForDefaultDepthCompatibility() and modify behavior based on this value. Recognized values: - Strict means that the old and new default depth values must match exactly. - Safe or Risky means that the default depth values might differ, but the new default depth value must be at least as high as the former default depth value. This is recommended, because it allows clients with a higher default depth value to still connect, but not lose functionality. - Bypass or higher means that all of these checks are essentially deactivated. This is probably a very bad idea. Note that the default ReconnectTolerance value is still Strict.
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-03 20:41:08 +0000