Re: [PATCH] xfree86: loader: Remove redundant debug command in loader.c.
Another place where I don't know the code, but the patch sure looks like a good idea. Reviewed-by: Jamey Sharp On Fri, Oct 1, 2010 at 7:06 PM, Jesse Adkins wrote: > Signed-off-by: Jesse Adkins > --- > hw/xfree86/loader/loader.c | 4 > 1 files changed, 0 insertions(+), 4 deletions(-) > > diff --git a/hw/xfree86/loader/loader.c b/hw/xfree86/loader/loader.c > index 8921831..39e0e75 100644 > --- a/hw/xfree86/loader/loader.c > +++ b/hw/xfree86/loader/loader.c > @@ -143,10 +143,6 @@ LoaderOpen(const char *module, int *errmaj, int *errmin) > { > void *ret; > > -#if defined(DEBUG) > - ErrorF("LoaderOpen(%s)\n", module); > -#endif > - > xf86Msg(X_INFO, "Loading %s\n", module); > > if (!(ret = dlopen(module, DLOPEN_LAZY | DLOPEN_GLOBAL))) { > -- > 1.7.0.4 > > ___ > xorg-devel@lists.x.org: X.Org development > Archives: http://lists.x.org/archives/xorg-devel > Info: http://lists.x.org/mailman/listinfo/xorg-devel > ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH] xfree86: loader: Remove redundant debug command in loader.c.
Signed-off-by: Jesse Adkins --- hw/xfree86/loader/loader.c |4 1 files changed, 0 insertions(+), 4 deletions(-) diff --git a/hw/xfree86/loader/loader.c b/hw/xfree86/loader/loader.c index 8921831..39e0e75 100644 --- a/hw/xfree86/loader/loader.c +++ b/hw/xfree86/loader/loader.c @@ -143,10 +143,6 @@ LoaderOpen(const char *module, int *errmaj, int *errmin) { void *ret; -#if defined(DEBUG) -ErrorF("LoaderOpen(%s)\n", module); -#endif - xf86Msg(X_INFO, "Loading %s\n", module); if (!(ret = dlopen(module, DLOPEN_LAZY | DLOPEN_GLOBAL))) { -- 1.7.0.4 ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH modular] Catch bad command-line arguments.
From: Trevor Woerner If the user specifies unknown, misspelled, or too many cmdline arguments make sure to catch them and generate an error. Signed-off-by: Trevor Woerner --- NOTE: this patch to be applied after the 10 I sent yesterday! build.sh |9 + 1 files changed, 9 insertions(+), 0 deletions(-) diff --git a/build.sh b/build.sh index 6c63cc5..6c76a5c 100755 --- a/build.sh +++ b/build.sh @@ -897,6 +897,7 @@ HAVE_ARCH="`uname -i`" DIR_ARCH="" DIR_CONFIG="." LIB_ONLY=0 +PREFIX="" # Process command line args while [ $# != 0 ] @@ -984,6 +985,14 @@ do fi ;; *) + if [ X"$PREFIX" != X ]; then + echo "unrecognized and/or too many command-line arguments" + echo " new PREFIX: $1" + echo " existing PREFIX: $PREFIX" + echo "" + usage + exit 1 + fi PREFIX=$1 ;; esac -- 1.7.3.1.45.g9855b ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH v3 1/2] dix: Add facilities for client ID tracking.
> +AC_ARG_ENABLE(clientids, AS_HELP_STRING([--disable-clientids], [Build > Xorg with client ID tracking (default: enabled)]), [CLIENTIDS=$enableval], > [CLIENTIDS=yes]) Shouldn't that be --enable-clientids instead of --disable-clientids? ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH] test: Fix make distcheck when not building unit tests
Signed-off-by: Jeremy Huddleston --- test/Makefile.am |3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/test/Makefile.am b/test/Makefile.am index 7c44671..456221e 100644 --- a/test/Makefile.am +++ b/test/Makefile.am @@ -33,10 +33,9 @@ libxservertest_la_LIBADD = \ $(top_builddir)/mi/libmi.la \ $(top_builddir)/os/libos.la \ @XORG_LIBS@ +endif CLEANFILES=libxservertest.c libxservertest.c: touch $@ - -endif -- 1.5.6.6 ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
xserver make distdir fails without doxygen because of hw/dmx/doc
make[3]: *** No rule to make target `dmx.txt', needed by `distdir'. Stop. make[2]: *** [distdir] Error 1 make[1]: *** [distdir] Error 1 make: *** [distdir] Error 1 Do we really want to ship all of the generated documentation in hw/dmx/doc? Is this the case in other modules, or is this an exception? ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH modular 05/10] Add preconditions.
On Thu, 2010-09-30 at 17:27 -0400, Trevor Woerner wrote: > From: Trevor Woerner > > Provide preconditions for some of the functions in build.sh script. > > Signed-off-by: Trevor Woerner > --- > build.sh | 12 > 1 files changed, 12 insertions(+), 0 deletions(-) > For the remainder patches up to 0010. Reviewed-by: Gaetan Nadon signature.asc Description: This is a digitally signed message part ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH modular 04/10] Document functions in build.sh.
On Thu, 2010-09-30 at 17:27 -0400, Trevor Woerner wrote: > From: Trevor Woerner > > Provide explanations of functioning and behaviour for some of the > functions > in build.sh script. > > Signed-off-by: Trevor Woerner > --- > build.sh | 31 +++ > 1 files changed, 31 insertions(+), 0 deletions(-) > Reviewed-by: Gaetan Nadon signature.asc Description: This is a digitally signed message part ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH modular 03b/10] Centralize NOQUIT build mode.
On Thu, 2010-09-30 at 22:45 -0400, Trevor Woerner wrote: > From: Trevor Woerner > > Put all the logic to handle the NOQUIT (-n) mode into one location. > > Signed-off-by: Trevor Woerner > --- > build.sh | 97 > + > 1 files changed, 78 insertions(+), 19 deletions(-) > Reviewed-by: Gaetan Nadon signature.asc Description: This is a digitally signed message part ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH modular 03a/10] Centralize build mode logic.
On Thu, 2010-09-30 at 22:44 -0400, Trevor Woerner wrote: > From: Trevor Woerner > > Place into one location the logic to decide what to do with regards to > the following build modes: LISTONLY, RESUME, and BUILD_ONE. > > Signed-off-by: Trevor Woerner > --- > build.sh | 66 > +++-- > 1 files changed, 38 insertions(+), 28 deletions(-) > Reviewed-by: Gaetan Nadon signature.asc Description: This is a digitally signed message part ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, 01 Oct 2010 12:11:49 -0400, Gaetan Nadon wrote: > Just a reminder for those reading this thread. > > Should that information (the oldest server a driver wishes to support) > become available, > it should be coded in configure.ac: > > PKG_CHECK_MODULES(XORG, [xorg-server >= 1.0.99.901 ...) Thanks for the reminder -- I'm pushing a version bump to the X server to 1.9.99.1 so that drivers can tell that they're building against the 1.10 development series. -- keith.pack...@intel.com pgpzbN26jx9o3.pgp Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fre, 2010-10-01 at 18:57 +0200, Julien Cristau wrote: > On Fri, Oct 1, 2010 at 18:29:54 +0200, Michel Dänzer wrote: > > > On Fre, 2010-10-01 at 18:27 +0200, Julien Cristau wrote: > > > #if ABI_VIDEODRV_VERSION < SET_ABI_VERSION(9, 0) > > > old code > > > #else > > > new code > > > #endif > > > > That would work perfectly, if the ABI was bumped by every change which > > breaks the ABI... > > > It'll work for this specific bump. Not when building against xserver commits between 5c6a2f93ebc16a78093782b442306de23ae94e78 and 1a9022d3289ff74043112808cc8aa315fd9c40e8 . Certainly better than merely checking the server version though. > Or you could do something like > AC_CHECK_MEMBERS([ScrnInfoRec.pixmapPrivate],[],[],[#include ]) > in configure.ac, and check HAVE_SCRNINFOREC_PIXMAPPRIVATE. Right, though that will require manually re-running configure when crossing this ABI break. -- Earthling Michel Dänzer |http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, Oct 1, 2010 at 13:14:51 -0400, Trevor Woerner wrote: > Irregardless of which solution is chosen, if someone wanted to compile > a driver for an older server, they'd still have to checkout an older > version of the server code to compile against. Wouldn't they? What are > the chances a newer driver 1 year from now will still compile against > the older server source? > Typically they'd already have an X server (because it came with their distro, or because they are the distro), and just want an updated driver. > Shouldn't a check like this be done at runtime? By making a decision > like this at build time, aren't we sentencing the compiled driver to > always only connect to either older or newer servers, but not both? > Being able to run the driver without recompiling on different X server versions only matters if you can't recompile your driver. Cheers, Julien ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
2010/10/1 Julien Cristau : > On Fri, Oct 1, 2010 at 18:29:54 +0200, Michel Dänzer wrote: > >> On Fre, 2010-10-01 at 18:27 +0200, Julien Cristau wrote: >> > #if ABI_VIDEODRV_VERSION < SET_ABI_VERSION(9, 0) >> > old code >> > #else >> > new code >> > #endif >> >> That would work perfectly, if the ABI was bumped by every change which >> breaks the ABI... >> > It'll work for this specific bump. Or you could do something like > AC_CHECK_MEMBERS([ScrnInfoRec.pixmapPrivate],[],[],[#include ]) > in configure.ac, and check HAVE_SCRNINFOREC_PIXMAPPRIVATE. Irregardless of which solution is chosen, if someone wanted to compile a driver for an older server, they'd still have to checkout an older version of the server code to compile against. Wouldn't they? What are the chances a newer driver 1 year from now will still compile against the older server source? Shouldn't a check like this be done at runtime? By making a decision like this at build time, aren't we sentencing the compiled driver to always only connect to either older or newer servers, but not both? ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, Oct 1, 2010 at 18:29:54 +0200, Michel Dänzer wrote: > On Fre, 2010-10-01 at 18:27 +0200, Julien Cristau wrote: > > #if ABI_VIDEODRV_VERSION < SET_ABI_VERSION(9, 0) > > old code > > #else > > new code > > #endif > > That would work perfectly, if the ABI was bumped by every change which > breaks the ABI... > It'll work for this specific bump. Or you could do something like AC_CHECK_MEMBERS([ScrnInfoRec.pixmapPrivate],[],[],[#include ]) in configure.ac, and check HAVE_SCRNINFOREC_PIXMAPPRIVATE. Cheers, Julien ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fre, 2010-10-01 at 18:27 +0200, Julien Cristau wrote: > On Fri, Oct 1, 2010 at 12:16:55 -0400, Trevor Woerner wrote: > > > On Fri, Oct 1, 2010 at 11:37 AM, Mark Kettenis > > wrote: > > > Also, drivers are supposed to work with some older servers as well, at > > > least some of them. So even if removing that line for the current > > > server is OK, it probably isn't for older servers. > > > > What's the trick for compiling future drivers for older servers when > > the given symbol has been removed from the server header files? > > #IFDEF's? > > #if ABI_VIDEODRV_VERSION < SET_ABI_VERSION(9, 0) > old code > #else > new code > #endif That would work perfectly, if the ABI was bumped by every change which breaks the ABI... -- Earthling Michel Dänzer |http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, Oct 1, 2010 at 12:16:55 -0400, Trevor Woerner wrote: > On Fri, Oct 1, 2010 at 11:37 AM, Mark Kettenis > wrote: > > Also, drivers are supposed to work with some older servers as well, at > > least some of them. So even if removing that line for the current > > server is OK, it probably isn't for older servers. > > What's the trick for compiling future drivers for older servers when > the given symbol has been removed from the server header files? > #IFDEF's? #if ABI_VIDEODRV_VERSION < SET_ABI_VERSION(9, 0) old code #else new code #endif Cheers, Julien ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fre, 2010-10-01 at 12:11 -0400, Gaetan Nadon wrote: > > On Fri, 2010-10-01 at 17:37 +0200, Mark Kettenis wrote: > > > Also, drivers are supposed to work with some older servers as well, > at > > least some of them. So even if removing that line for the current > > server is OK, it probably isn't for older servers. > > > Just a reminder for those reading this thread. > > Should that information (the oldest server a driver wishes to support) > become available, > it should be coded in configure.ac: > > PKG_CHECK_MODULES(XORG, [xorg-server >= 1.0.99.901 ...) > > with the appropriate rationale in the commit text. Whenever possible, it should be preferable to check for (the lack of) a feature directly rather than indirectly via a version. -- Earthling Michel Dänzer |http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, Oct 1, 2010 at 11:37 AM, Mark Kettenis wrote: > Also, drivers are supposed to work with some older servers as well, at > least some of them. So even if removing that line for the current > server is OK, it probably isn't for older servers. What's the trick for compiling future drivers for older servers when the given symbol has been removed from the server header files? #IFDEF's? ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fri, 2010-10-01 at 17:37 +0200, Mark Kettenis wrote: > Also, drivers are supposed to work with some older servers as well, at > least some of them. So even if removing that line for the current > server is OK, it probably isn't for older servers. Just a reminder for those reading this thread. Should that information (the oldest server a driver wishes to support) become available, it should be coded in configure.ac: PKG_CHECK_MODULES(XORG, [xorg-server >= 1.0.99.901 ...) with the appropriate rationale in the commit text. Gaetan signature.asc Description: This is a digitally signed message part ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
> From: Michel =?ISO-8859-1?Q?D=E4nzer?= > Date: Fri, 01 Oct 2010 17:18:05 +0200 > > On Fre, 2010-10-01 at 08:07 -0400, Trevor Woerner wrote: > > From: Trevor Woerner > > > > Signed-off-by: Trevor Woerner > > --- > > src/drmmode_display.c |1 - > > 1 files changed, 0 insertions(+), 1 deletions(-) > > > > diff --git a/src/drmmode_display.c b/src/drmmode_display.c > > index a6038b2..57a8ec6 100644 > > --- a/src/drmmode_display.c > > +++ b/src/drmmode_display.c > > @@ -1145,7 +1145,6 @@ drmmode_xf86crtc_resize (ScrnInfoPtr scrn, int width, > > int height) > >width, height, -1, -1, pitch * cpp, > >info->fb_shadow); > > } > > - scrn->pixmapPrivate.ptr = ppix->devPrivate.ptr; > > Before applying this as is, someone should confirm that this line really > isn't needed with servers that have the pixmapPrivate field. It was > added in Jerome's commit 95385f0906f371a1ed6e60eb3e597e699c7a3222 ('kms: > Fix resizing when acceleration is disabled'), which indicates it might > be necessary when resizing the screen with acceleration disabled. Also, drivers are supposed to work with some older servers as well, at least some of them. So even if removing that line for the current server is OK, it probably isn't for older servers. ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [PATCH ati] Remove reference to pixmapPrivate.
On Fre, 2010-10-01 at 08:07 -0400, Trevor Woerner wrote: > From: Trevor Woerner > > Signed-off-by: Trevor Woerner > --- > src/drmmode_display.c |1 - > 1 files changed, 0 insertions(+), 1 deletions(-) > > diff --git a/src/drmmode_display.c b/src/drmmode_display.c > index a6038b2..57a8ec6 100644 > --- a/src/drmmode_display.c > +++ b/src/drmmode_display.c > @@ -1145,7 +1145,6 @@ drmmode_xf86crtc_resize (ScrnInfoPtr scrn, int width, > int height) > width, height, -1, -1, pitch * cpp, > info->fb_shadow); > } > - scrn->pixmapPrivate.ptr = ppix->devPrivate.ptr; Before applying this as is, someone should confirm that this line really isn't needed with servers that have the pixmapPrivate field. It was added in Jerome's commit 95385f0906f371a1ed6e60eb3e597e699c7a3222 ('kms: Fix resizing when acceleration is disabled'), which indicates it might be necessary when resizing the screen with acceleration disabled. -- Earthling Michel Dänzer |http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH v3 0/2] xserver: Client ID tracking
Fixes based on comments of various reviewers. v2 -- Mark Kettenis: move /proc specific code into os treat pid -1 as error instead of 0 Tiago Vignatti: preserve copyrights amend authorship add description of the framework use doxygen style rename cmd -> cmdline in ClientIdsPrivateRec compile only if XRES or XSELINUX is enabled v3 -- Fernando Carrijo: don't use -o or -a with "test" shell command Tiago Vignatti: use doxygen @file directive move pid_t related functions into os remove unnecessary header inclusions let client ID tracking be disabled with --disable-clientids create dummy stubs for initialization functions if disabled don't make XSELinux depend on client ID tracking squash copyrights into main commit don't add copyright for restructuring of minor code segments Review comments not respected: don't install client ID tracking headers Rationale: As soon as client ID tracking is incorporated into X server, it will be used from debugging code of at least one video driver. Rami Ylimäki (2): dix: Add facilities for client ID tracking. Xext: Use general OS functions to determine client command string in SELinux. Xext/xselinux_hooks.c| 22 +-- configure.ac | 15 ++- dix/Makefile.am |1 + dix/client.c | 345 ++ dix/main.c |3 + hw/xfree86/loader/sdksyms.sh |1 + include/Makefile.am |1 + include/client.h | 60 include/dix-config.h.in |3 + include/os.h |3 + os/access.c | 76 + 11 files changed, 512 insertions(+), 18 deletions(-) create mode 100644 dix/client.c create mode 100644 include/client.h ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH v3 2/2] Xext: Use general OS functions to determine client command string in SELinux.
SELinux could be also modified to take advantage of client ID tracking but we don't introduce that dependency here. People interested in SELinux are free to add that dependency later. Signed-off-by: Rami Ylimäki --- Xext/xselinux_hooks.c | 22 +- 1 files changed, 5 insertions(+), 17 deletions(-) diff --git a/Xext/xselinux_hooks.c b/Xext/xselinux_hooks.c index 560e1e9..73a1179 100644 --- a/Xext/xselinux_hooks.c +++ b/Xext/xselinux_hooks.c @@ -129,26 +129,14 @@ SELinuxLabelClient(ClientPtr client) /* For local clients, try and determine the executable name */ if (XaceIsLocal(client)) { - struct ucred creds; - socklen_t len = sizeof(creds); - char path[PATH_MAX + 1]; - size_t bytes; + pid_t pid = GetPidFromClient(client); + const char *cmd = GetCommandFromPid(pid); - memset(&creds, 0, sizeof(creds)); - if (getsockopt(fd, SOL_SOCKET, SO_PEERCRED, &creds, &len) < 0) + if (!cmd) goto finish; - snprintf(path, PATH_MAX + 1, "/proc/%d/cmdline", creds.pid); - fd = open(path, O_RDONLY); - if (fd < 0) - goto finish; - - bytes = read(fd, path, PATH_MAX + 1); - close(fd); - if (bytes <= 0) - goto finish; - - strncpy(subj->command, path, COMMAND_LEN - 1); + strncpy(subj->command, cmd, COMMAND_LEN - 1); + free((void *) cmd); } finish: -- 1.6.3.3 ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH v3 1/2] dix: Add facilities for client ID tracking.
An interface is provided for figuring out the PID and process name of a client. Make some existing functionality from SELinux and IA extensions available for general use. Signed-off-by: Rami Ylimäki --- configure.ac | 15 ++- dix/Makefile.am |1 + dix/client.c | 345 ++ dix/main.c |3 + hw/xfree86/loader/sdksyms.sh |1 + include/Makefile.am |1 + include/client.h | 60 include/dix-config.h.in |3 + include/os.h |3 + os/access.c | 76 + 10 files changed, 507 insertions(+), 1 deletions(-) create mode 100644 dix/client.c create mode 100644 include/client.h diff --git a/configure.ac b/configure.ac index 1a1f2d3..056d68f 100644 --- a/configure.ac +++ b/configure.ac @@ -645,6 +645,7 @@ AC_ARG_ENABLE(vbe, AS_HELP_STRING([--enable-vbe], [Build Xorg with VB AC_ARG_ENABLE(int10-module, AS_HELP_STRING([--enable-int10-module], [Build Xorg with int10 module (default: enabled)]), [INT10MODULE=$enableval], [INT10MODULE=yes]) AC_ARG_ENABLE(windowswm, AS_HELP_STRING([--enable-windowswm], [Build XWin with WindowsWM extension (default: no)]), [WINDOWSWM=$enableval], [WINDOWSWM=no]) AC_ARG_ENABLE(libdrm, AS_HELP_STRING([--enable-libdrm], [Build Xorg with libdrm support (default: enabled)]), [DRM=$enableval],[DRM=yes]) +AC_ARG_ENABLE(clientids, AS_HELP_STRING([--disable-clientids], [Build Xorg with client ID tracking (default: enabled)]), [CLIENTIDS=$enableval], [CLIENTIDS=yes]) dnl DDXes. AC_ARG_ENABLE(xorg, AS_HELP_STRING([--enable-xorg], [Build Xorg server (default: auto)]), [XORG=$enableval], [XORG=auto]) @@ -999,6 +1000,18 @@ if test "x$RES" = xyes; then REQUIRED_MODULES="$REQUIRED_MODULES $RESOURCEPROTO" fi +# The XRes extension may support client ID tracking only if it has +# been specifically enabled. Client ID tracking is implicitly not +# supported if XRes extension is disabled. +AC_MSG_CHECKING([whether to track client ids]) +if test "x$RES" = xyes && test "x$CLIENTIDS" = xyes; then + AC_DEFINE(CLIENTIDS, 1, [Support client ID tracking]) +else + CLIENTIDS=no +fi +AC_MSG_RESULT([$CLIENTIDS]) +AM_CONDITIONAL(CLIENTIDS, [test "x$CLIENTIDS" = xyes]) + if test "x$GLX" = xyes; then PKG_CHECK_MODULES([XLIB], [x11]) PKG_CHECK_MODULES([GL], $GLPROTO $LIBGL) @@ -1527,7 +1540,7 @@ if test "x$XNEST" = xyes; then if test "x$have_xnest" = xno; then AC_MSG_ERROR([Xnest build explicitly requested, but required modules not found.]) fi - XNEST_LIBS="$FB_LIB $FIXES_LIB $MI_LIB $XEXT_LIB $DBE_LIB $RECORD_LIB $GLX_LIBS $RANDR_LIB $RENDER_LIB $DAMAGE_LIB $MIEXT_DAMAGE_LIB $MIEXT_SHADOW_LIB $XI_LIB $XKB_LIB $XKB_STUB_LIB $COMPOSITE_LIB $DIX_LIB $MAIN_LIB $OS_LIB" + XNEST_LIBS="$FB_LIB $FIXES_LIB $MI_LIB $XEXT_LIB $DBE_LIB $RECORD_LIB $GLX_LIBS $RANDR_LIB $RENDER_LIB $DAMAGE_LIB $MIEXT_DAMAGE_LIB $MIEXT_SHADOW_LIB $XI_LIB $XKB_LIB $XKB_STUB_LIB $COMPOSITE_LIB $MAIN_LIB $DIX_LIB $OS_LIB" XNEST_SYS_LIBS="$XNESTMODULES_LIBS $GLX_SYS_LIBS" AC_SUBST([XNEST_LIBS]) AC_SUBST([XNEST_SYS_LIBS]) diff --git a/dix/Makefile.am b/dix/Makefile.am index 5e2dad7..49e41d0 100644 --- a/dix/Makefile.am +++ b/dix/Makefile.am @@ -8,6 +8,7 @@ libmain_la_SOURCES =\ libdix_la_SOURCES =\ atom.c \ colormap.c \ + client.c\ cursor.c\ deprecated.c\ devices.c \ diff --git a/dix/client.c b/dix/client.c new file mode 100644 index 000..1016a56 --- /dev/null +++ b/dix/client.c @@ -0,0 +1,345 @@ +/* + * Copyright (C) 2010 Nokia Corporation and/or its subsidiary(-ies). All + * rights reserved. + * Copyright (c) 1993, 2010, Oracle and/or its affiliates. All rights reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE
Re: [DOC] Re: java OutOfMemory building documentation
Dan Nicholson wrote: > On Thu, Sep 30, 2010 at 6:56 PM, Jeremy Huddleston wrote: >> Also of note... My G5 took an hour and a half just to 'make install' with >> these changes. It looks like it's redoing some things in 'make install' >> that it already did in 'make': >> >> http://tinderbox.x.org/builds/2010-09-30-0022/ > > One guess: don't make nobase_x11locale_DATA depend on builddirs in > nls/Makefile.am. Since builddirs is essentially a phony target and > doesn't produce a target with a timestamp, nobase_x11locale_DATA will > always be considered out of date and hence get rebuilt. > > I think there are two ways to handle this: > > 1. Change all the .pre rules (including the one in cpprules.in) to > create the directory they're being copied to if necessary. Something > like: > > .pre: > @dir=`echo $@ | sed 's,\(.*\)/.*,\1,'`; $(MKDIR_P) $$dir Shouldn't this be able to be simplified to just: @$(MKDIR_P) $(@D) I thought we previously determined that was portable to all the versions of make people used. -- -Alan Coopersmith-alan.coopersm...@oracle.com Oracle Solaris Platform Engineering: X Window System ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
patch:libXt-1.0.8 fix possible NULL access
Hi List, while testing a static checker, i got warnings about a possible NULL access. looking into the code i found that GetClassActions() may return NULL causing table->signature to ask for trouble. re, wh Honor that GetClassActions() may return NULL. Signed-off-by: Walter --- libXt-1.0.8/src/TMaction.c.org 2010-09-29 20:37:48.0 +0200 +++ libXt-1.0.8/src/TMaction.c 2010-09-29 20:39:56.0 +0200 @@ -807,10 +807,12 @@ if (*num_actions_return) { list = *actions_return = (XtActionList) __XtMalloc(*num_actions_return * sizeof(XtActionsRec)); - table = GetClassActions(widget_class); - for (i= (*num_actions_return); --i >= 0; list++, table++) { - list->string = XrmQuarkToString(table->signature); - list->proc = table->proc; + table = GetClassActions(widget_class); + if (table != NULL ) { + for (i= (*num_actions_return); --i >= 0; list++, table++) { + list->string = XrmQuarkToString(table->signature); + list->proc = table->proc; + } } } UNLOCK_PROCESS; --- libXt-1.0.8/src/TMaction.c.org 2010-09-29 20:37:48.0 +0200 +++ libXt-1.0.8/src/TMaction.c 2010-09-29 20:39:56.0 +0200 @@ -807,10 +807,12 @@ if (*num_actions_return) { list = *actions_return = (XtActionList) __XtMalloc(*num_actions_return * sizeof(XtActionsRec)); - table = GetClassActions(widget_class); /* FIXME: can be NULL */ - for (i= (*num_actions_return); --i >= 0; list++, table++) { - list->string = XrmQuarkToString(table->signature); - list->proc = table->proc; + table = GetClassActions(widget_class); + if (table != NULL ) { + for (i= (*num_actions_return); --i >= 0; list++, table++) { + list->string = XrmQuarkToString(table->signature); + list->proc = table->proc; + } } } UNLOCK_PROCESS; ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: [DOC] Re: java OutOfMemory building documentation
On Thu, Sep 30, 2010 at 6:56 PM, Jeremy Huddleston wrote: > Also of note... My G5 took an hour and a half just to 'make install' with > these changes. It looks like it's redoing some things in 'make install' > that it already did in 'make': > > http://tinderbox.x.org/builds/2010-09-30-0022/ One guess: don't make nobase_x11locale_DATA depend on builddirs in nls/Makefile.am. Since builddirs is essentially a phony target and doesn't produce a target with a timestamp, nobase_x11locale_DATA will always be considered out of date and hence get rebuilt. I think there are two ways to handle this: 1. Change all the .pre rules (including the one in cpprules.in) to create the directory they're being copied to if necessary. Something like: .pre: @dir=`echo $@ | sed 's,\(.*\)/.*,\1,'`; $(MKDIR_P) $$dir $(AM_V_GEN)$(RAWCPP) $(RAWCPPFLAGS) $(CPP_FILES_FLAGS) < $< | $(CPP_SED_MAGIC) > $@ 2. Create a stamp file when you make the directories and clean it up builddirs: $(MKDIR_P) $(locales) @touch $@ CLEANFILES += builddirs If you build with "make --debug=b", you can see info about what prereqs are considered out of date. -- Dan ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH siliconmotion] Remove reference to pixmapPrivate.
From: Trevor Woerner Signed-off-by: Trevor Woerner --- src/smi_crtc.c |5 - src/smi_driver.c |1 - 2 files changed, 0 insertions(+), 6 deletions(-) diff --git a/src/smi_crtc.c b/src/smi_crtc.c index bccf9b2..36bd085 100644 --- a/src/smi_crtc.c +++ b/src/smi_crtc.c @@ -210,11 +210,6 @@ SMI_CrtcConfigResize(ScrnInfoPtr pScrn, pScrn->pScreen->ModifyPixmapHeader(pScrn->pScreen->GetScreenPixmap(pScrn->pScreen), -1,-1,-1,-1,-1, pSmi->FBBase + pSmi->FBOffset); - if(pScrn->pixmapPrivate.ptr) - /* The pixmap devPrivate just set may be overwritten by - xf86EnableDisableFBAccess */ - pScrn->pixmapPrivate.ptr = pSmi->FBBase + pSmi->FBOffset; - /* Modify the screen pitch */ pScrn->displayWidth = aligned_pitch / pSmi->Bpp; pScrn->pScreen->ModifyPixmapHeader(pScrn->pScreen->GetScreenPixmap(pScrn->pScreen), diff --git a/src/smi_driver.c b/src/smi_driver.c index 391efbc..45d19a9 100644 --- a/src/smi_driver.c +++ b/src/smi_driver.c @@ -927,7 +927,6 @@ SMI_EnterVT(int scrnIndex, int flags) /* FBBase may have changed after remapping the memory */ pScrn->pScreen->ModifyPixmapHeader(pScrn->pScreen->GetScreenPixmap(pScrn->pScreen), -1,-1,-1,-1,-1, pSmi->FBBase + pSmi->FBOffset); -pScrn->pixmapPrivate.ptr=pSmi->FBBase + pSmi->FBOffset; if(pSmi->useEXA) pSmi->EXADriverPtr->memoryBase=pSmi->FBBase; -- 1.7.3.1.45.g9855b ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH mach64] Remove reference to pixmapPrivate.
From: Trevor Woerner Signed-off-by: Trevor Woerner --- src/aticonsole.c |3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/src/aticonsole.c b/src/aticonsole.c index 1be147e..902e260 100644 --- a/src/aticonsole.c +++ b/src/aticonsole.c @@ -715,8 +715,6 @@ ATIEnterVT pScreenPixmap = (*pScreen->GetScreenPixmap)(pScreen); PixmapPrivate = pScreenPixmap->devPrivate; -if (!PixmapPrivate.ptr) -pScreenPixmap->devPrivate = pScreenInfo->pixmapPrivate; /* Tell framebuffer about remapped aperture */ Entered = (*pScreen->ModifyPixmapHeader)(pScreenPixmap, @@ -724,7 +722,6 @@ ATIEnterVT if (!PixmapPrivate.ptr) { -pScreenInfo->pixmapPrivate = pScreenPixmap->devPrivate; pScreenPixmap->devPrivate.ptr = NULL; } -- 1.7.3.1.45.g9855b ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH ati] Remove reference to pixmapPrivate.
From: Trevor Woerner Signed-off-by: Trevor Woerner --- src/drmmode_display.c |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/src/drmmode_display.c b/src/drmmode_display.c index a6038b2..57a8ec6 100644 --- a/src/drmmode_display.c +++ b/src/drmmode_display.c @@ -1145,7 +1145,6 @@ drmmode_xf86crtc_resize (ScrnInfoPtr scrn, int width, int height) width, height, -1, -1, pitch * cpp, info->fb_shadow); } - scrn->pixmapPrivate.ptr = ppix->devPrivate.ptr; // xf86DrvMsg(scrn->scrnIndex, X_INFO, "New front buffer at 0x%lx\n", // info->front_bo-); -- 1.7.3.1.45.g9855b ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
[PATCH geode] Remove reference to pixmapPrivate.
From: Trevor Woerner Signed-off-by: Trevor Woerner --- src/gx_randr.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/src/gx_randr.c b/src/gx_randr.c index 1593c97..2deac14 100644 --- a/src/gx_randr.c +++ b/src/gx_randr.c @@ -218,8 +218,6 @@ GXRandRSetMode(ScreenPtr pScreen, */ pspix = (*pScreen->GetScreenPixmap) (pScreen); -if (pspix->devPrivate.ptr) - pScrni->pixmapPrivate = pspix->devPrivate; xf86ReconfigureLayout(); -- 1.7.3.1.45.g9855b ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
Re: Build error: pixmapPrivate
The information in bug report 30451 [https://bugs.freedesktop.org/show_bug.cgi?id=30451] seems to suggest simply removing any references to pixmapPrivate in the code is enough. Can anyone confirm that? I can provide patch attempts, but have no way of testing them. ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
(RFC) siliconmotion: Offscreen Pixmaps not properly initialized?
We have a MPC5200 PowerPC @ 400 MHz with a SM501 (TQM5200 board). With XAA enabled, the xf86-video-siliconmotion works nicely except for screen-to-screen-copies (?). The symptom is that x11perf -copywinpix100 x11perf -copypixwin100 are 10x slower than with the following hack: smi_xaa.c, SMI_XAAInit: infoPtr->offscreenDepths = (1 << (24 - 1)); // depth is usually 24 infoPtr->offscreenDepthsInitialized = 1; The xserver doesn't set bit 23 in offscreenDepths (in xaaInit.c, XAAInitializeOffscreenDepths) because it appears to be unable to create a pixmap with depth 24 in XAAPixmapBPP. However, with the hack, the display in x11perf is the same as without. Below is part of the log. The reason I cannot live with the hack any more is that, in recent xserver, it causes parts of the screen to be not properly displayed (e.g. no fonts are displayed), even if I set all other relevant bits in offscreenDepths. In a xserver from 2009-08 it was working. On the other hand, without the hack, the CPU-load increases from 80% to 100% in our application, which is also bad. I've spent a week fulltime on this without success. So really any sort of hint is appreciated. Marc (II) SMI(0): Creating default Display subsection in Screen section (==) SMI(0): Depth 24, (--) framebuffer bpp 32 (==) SMI(0): RGB weight 888 (==) SMI(0): Default visual is TrueColor (**) SMI(0): Option "pci_burst" "0" (**) SMI(0): Option "NoAccel" "0" (**) SMI(0): Option "HWCursor" "1" (**) SMI(0): Option "ByteSwap" "0" (**) SMI(0): Option "UseBIOS" "0" (**) SMI(0): Option "Dualhead" "0" (**) SMI(0): Option "AccelMethod" "XAA" (**) SMI(0): Option "PanelSize" "800x600" (**) SMI(0): Option "UseFBDev" "0" (**) SMI(0): Option "CSCVideo" "1" (**) SMI(0): PCI Burst disabled (==) SMI(0): PCI Retry disabled (**) SMI(0): UseFBDev disabled. (**) SMI(0): Using Hardware Cursor (**) SMI(0): Chipset: "MSOC" (**) SMI(0): Dual head disabled (**) SMI(0): Using XAA acceleration architecture (**) SMI(0): CSC Video enabled (--) SMI(0): videoram: 7680kB (--) SMI(0): videoram: 7680kB (II) SMI(0): Reserved: 0077F800 (II) SMI(0): TFT Panel Size = 800x600 (==) SMI(0): Using gamma correction (1.0, 1.0, 1.0) (II) SMI(0): MCLK = 24.000 (II) SMI(0): MXCLK = 144.000 (II) SMI(0): Output LVDS using monitor section Panel (II) SMI(0): Not using default mode "640x350" (vrefresh out of range) ... (II) SMI(0): Printing probed modes for output LVDS (II) SMI(0): Modeline "800x600"x59.9 38.25 800 832 912 1024 600 603 607 624 -hsync +vsync (37.4 kHz) (II) SMI(0): Output LVDS connected (II) SMI(0): Using fuzzy aspect match for initial modes (II) SMI(0): Output LVDS using initial mode 800x600 (II) SMI(0): Using default gamma of (1.0, 1.0, 1.0) unless otherwise stated. (**) SMI(0): Display dimensions: (215, 161) mm (**) SMI(0): DPI set to (94, 126) (II) SMI(0): Reserved: 0077F800 (II) SMI(0): Register dump (Before Save) (II) SMI(0): FrameBuffer Box: 0,0 - 800,2456 (II) SMI(0): Using XFree86 Acceleration Architecture (XAA) (II) SMI(0): Clock request 38250.00 (max_divider 5) (==) SMI(0): DPMS enabled (II) SMI(0): RandR 1.2 enabled, ignore the following RandR disabled message. (II) SMI(0): Setting screen physical size to 215 x 161 while(!solved) { boss->patience >> 1; work->time++; wife |= UNHAPPY; } ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel