date:20121108

Re: [PATCH v2] xf86: Fix non-PCI configuration-less setups

2012-11-08 Thread Thierry Reding

On Wed, Oct 17, 2012 at 12:06:47PM +0200, Thierry Reding wrote:
 For non-PCI video devices, such as those found on many ARM embedded
 systems, the X server currently requires the BusID option to specify the
 full path to the DRM device's sysfs node in order to properly match it
 against the probed platform devices.
 
 In order to allow X to start up properly if either the BusID option was
 omitted or no configuration is present at all, the first video device is
 used by default.
 
 Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
 ---
 Changes in v2:
 - Add additional checks for safety (I don't think numDevs will ever be 0
   since a default will be generated if no configuration is present, but
   it doesn't hurt to check anyway). Without these checks there is a
   possibility of the X server crashing if no platform devices have been
   found.
 
  hw/xfree86/common/xf86platformBus.c | 8 
  1 file changed, 8 insertions(+)
 
 diff --git a/hw/xfree86/common/xf86platformBus.c 
 b/hw/xfree86/common/xf86platformBus.c
 index 0525e39..599d84a 100644
 --- a/hw/xfree86/common/xf86platformBus.c
 +++ b/hw/xfree86/common/xf86platformBus.c
 @@ -377,6 +377,14 @@ xf86platformProbeDev(DriverPtr drvp)
  continue;
  }
  
 +/*
 + * If all of the above fails, which can happen if X was started without
 + * configuration or if BusID wasn't set for non-PCI devices, use the 
 first
 + * device by default.
 + */
 +if (!foundScreen  xf86_num_platform_devices  0  numDevs  0)
 +foundScreen = probeSingleDevice(xf86_platform_devices[0], drvp, 
 devList[0], 0);
 +
  /* if autoaddgpu devices is enabled then go find a few more and add them 
 as GPU screens */
  if (xf86Info.autoAddGPU  numDevs) {
  for (j = 0; j  xf86_num_platform_devices; j++) {

Can anyone review this patch? I think it is also a candidate for 1.13,
since it fixes a regression from the 1.12 series.

Thierry


pgpIfvRWYOWRj.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

[PATCH xf86-video-modesetting 1/2] build: Put m4 macros in m4 subdirectory

2012-11-08 Thread Thierry Reding

This silences a warning from libtoolize when running the autogen.sh
script.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 Makefile.am  | 2 ++
 autogen.sh   | 1 +
 configure.ac | 1 +
 3 files changed, 4 insertions(+)

diff --git a/Makefile.am b/Makefile.am
index 8e4dfdb..a7ab3e0 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -18,6 +18,8 @@
 #  IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 #  CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
+ACLOCAL_AMFLAGS = -I m4
+
 SUBDIRS = src man
 
 MAINTAINERCLEANFILES = ChangeLog INSTALL
diff --git a/autogen.sh b/autogen.sh
index 904cd67..f83ed27 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -5,6 +5,7 @@ test -z $srcdir  srcdir=.
 
 ORIGDIR=`pwd`
 cd $srcdir
+test -d m4 || mkdir m4
 
 autoreconf -v --install || exit 1
 cd $ORIGDIR || exit $?
diff --git a/configure.ac b/configure.ac
index b6815c4..58b1666 100644
--- a/configure.ac
+++ b/configure.ac
@@ -28,6 +28,7 @@ AC_INIT([xf86-video-modesetting],
 [xf86-video-modesetting])
 AC_CONFIG_SRCDIR([Makefile.am])
 AC_CONFIG_HEADERS([config.h])
+AC_CONFIG_MACRO_DIR([m4])
 AC_CONFIG_AUX_DIR(.)
 AC_SYS_LARGEFILE
 
-- 
1.8.0

___
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 xf86-video-modesetting 2/2] Remove call to miInitializeBackingStore()

2012-11-08 Thread Thierry Reding

Recent versions of the X server no longer provide this function, which
has been obsolete for over 2 years now.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 src/driver.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/driver.c b/src/driver.c
index 38a1c09..4866585 100644
--- a/src/driver.c
+++ b/src/driver.c
@@ -897,7 +897,6 @@ ScreenInit(SCREEN_INIT_ARGS_DECL)
 
 xf86SetBlackWhitePixels(pScreen);
 
-miInitializeBackingStore(pScreen);
 xf86SetBackingStore(pScreen);
 xf86SetSilkenMouse(pScreen);
 miDCInitialize(pScreen, xf86GetPointerScreenFuncs());
-- 
1.8.0

___
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 0/7] Various DIX patches

2012-11-08 Thread Jon TURNEY

Various DIX patches

Patches #2, #4 and #7 touch Xquartz to adjust for these changes, and I have 
tried to ensure that they do the right thing, but I am not able to test that 
Xquartz builds with those changes.

Jon TURNEY (7):
  os: Remove any old logfile before trying to write to it
  Allow DDX to provide a main()
  glx: Synchronize Xserver glx/rensize.c with mesa src/glx/compsize.c
  Move pseudoramiX code where it can be shared between Xwin and Xquartz
  Build dpmsstubs.c once as a convenience library, rather than once for
each DDX which wants to use it
  Build Xi/stubs.c once as a convenience library, rather than once for
each DDX which wants to use it
  Add a screen PaintWindow wrapper

 Makefile.am   |2 +
 Xext/Makefile.am  |4 ++-
 Xi/Makefile.am|5 ++-
 composite/compwindow.c|2 +-
 configure.ac  |3 +-
 dix/Makefile.am   |3 +-
 dix/main.c|6 +
 dix/stubmain.c|   33 +++
 dix/window.c  |3 +-
 glx/rensize.c |2 +
 hw/dmx/doc/dmx.xml|2 +-
 hw/vfb/Makefile.am|6 ++--
 hw/xnest/Makefile.am  |6 ++--
 hw/xquartz/Makefile.am|4 +-
 hw/xquartz/quartz.c   |2 +-
 hw/xwin/InitOutput.c  |6 +
 hw/xwin/Makefile.am   |9 ---
 hw/xwin/winrandr.c|2 +-
 include/scrnintstr.h  |5 
 mi/miexpose.c |   28 +-
 mi/mioverlay.c|   12 +-
 mi/miscrinit.c|1 +
 mi/miwindow.c |4 +-
 miext/rootless/rootlessCommon.h   |1 +
 miext/rootless/rootlessScreen.c   |3 +-
 miext/rootless/rootlessWindow.c   |   35 +
 miext/rootless/rootlessWindow.h   |1 +
 os/log.c  |3 ++
 pseudoramiX/Makefile.am   |7 +
 {hw/xquartz = pseudoramiX}/pseudoramiX.c |0
 {hw/xquartz = pseudoramiX}/pseudoramiX.h |0
 test/Makefile.am  |6 ++--
 32 files changed, 140 insertions(+), 66 deletions(-)
 create mode 100644 dix/stubmain.c
 create mode 100644 pseudoramiX/Makefile.am
 rename {hw/xquartz = pseudoramiX}/pseudoramiX.c (100%)
 rename {hw/xquartz = pseudoramiX}/pseudoramiX.h (100%)

-- 
1.7.9

___
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 1/7] os: Remove any old logfile before trying to write to it

2012-11-08 Thread Jon TURNEY

If we are not backing up logfiles, remove the old logfile before trying to write
a new logfile, as otherwise the operation may fail if the previous logfile was
created by a different user.

This change is useful when:
- The DDX doesn't use the logfile backup mechanism (i.e. not Xorg)
- The DDX is run by a non-root user, and then by a different non-root user
- The logfile directory doesn't have the restricted-deletion flag set

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
Acked-by: Yaakov Selkowitz yselkow...@users.sourceforge.net
Reviewed-by: Yaakov Selkowitz yselkow...@users.sourceforge.net
---
 os/log.c |3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/os/log.c b/os/log.c
index 4820e9a..15271eb 100644
--- a/os/log.c
+++ b/os/log.c
@@ -216,6 +216,9 @@ LogInit(const char *fname, const char *backup)
 free(oldLog);
 }
 }
+else {
+unlink(logFileName);
+}
 if ((logFile = fopen(logFileName, w)) == NULL)
 FatalError(Cannot open log file \%s\\n, logFileName);
 setvbuf(logFile, NULL, _IONBF, 0);
-- 
1.7.9

___
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 2/7] Allow DDX to provide a main()

2012-11-08 Thread Jon TURNEY

XQuartz already conditionally renames main() as dix_main() so it can provide
it's own main().  This isn't ideal as it prevents libdix built this way from
being useful with any other DDX.

So instead, always name that function dix_main(), and provide a stub main()
which just calls it where needed.

Add a main() to XWin.
It's no longer neccessary to link XWin and XQuartz with libmain.

Future work: prototype dix_main() somewhere, without it ending up in sdksyms...

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
 configure.ac |2 +-
 dix/Makefile.am  |3 ++-
 dix/main.c   |6 +-
 dix/stubmain.c   |   33 +
 hw/xwin/InitOutput.c |6 ++
 hw/xwin/Makefile.am  |4 ++--
 6 files changed, 45 insertions(+), 9 deletions(-)
 create mode 100644 dix/stubmain.c

diff --git a/configure.ac b/configure.ac
index 38ac240..e6f9cb7 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1960,7 +1960,7 @@ if test x$XQUARTZ = xyes; then
AC_DEFINE(XQUARTZ,1,[Have Quartz])
AC_DEFINE(ROOTLESS,1,[Build Rootless code])
 
-   XQUARTZ_LIBS=$MAIN_LIB $FB_LIB $FIXES_LIB $XEXT_LIB $DBE_LIB 
$RECORD_LIB $RANDR_LIB $RENDER_LIB $DAMAGE_LIB $MIEXT_SYNC_LIB 
$MIEXT_DAMAGE_LIB $MIEXT_SHADOW_LIB $XI_LIB $XKB_LIB $XKB_STUB_LIB $XPSTUBS_LIB
+   XQUARTZ_LIBS=$FB_LIB $FIXES_LIB $XEXT_LIB $DBE_LIB $RECORD_LIB 
$RANDR_LIB $RENDER_LIB $DAMAGE_LIB $MIEXT_SYNC_LIB $MIEXT_DAMAGE_LIB 
$MIEXT_SHADOW_LIB $XI_LIB $XKB_LIB $XKB_STUB_LIB $XPSTUBS_LIB
AC_SUBST([XQUARTZ_LIBS])
 
AC_CHECK_LIB([Xplugin],[xp_init],[:])
diff --git a/dix/Makefile.am b/dix/Makefile.am
index b7358aa..e7ca236 100644
--- a/dix/Makefile.am
+++ b/dix/Makefile.am
@@ -4,7 +4,7 @@ AM_CPPFLAGS = -I$(top_srcdir)/include
 AM_CFLAGS = $(DIX_CFLAGS)
 
 libmain_la_SOURCES =\
-   main.c
+   stubmain.c
 
 libdix_la_SOURCES =\
atom.c  \
@@ -14,6 +14,7 @@ libdix_la_SOURCES =   \
dispatch.c  \
dispatch.h  \
dixfonts.c  \
+   main.c  \
dixutils.c  \
enterleave.c\
enterleave.h\
diff --git a/dix/main.c b/dix/main.c
index fb935c9..82b5ee7 100644
--- a/dix/main.c
+++ b/dix/main.c
@@ -125,14 +125,10 @@ BOOL serverRunning = FALSE;
 pthread_mutex_t serverRunningMutex = PTHREAD_MUTEX_INITIALIZER;
 pthread_cond_t serverRunningCond = PTHREAD_COND_INITIALIZER;
 
-int dix_main(int argc, char *argv[], char *envp[]);
+#endif
 
 int
 dix_main(int argc, char *argv[], char *envp[])
-#else
-int
-main(int argc, char *argv[], char *envp[])
-#endif
 {
 int i;
 HWEventQueueType alwaysCheckForInput[2];
diff --git a/dix/stubmain.c b/dix/stubmain.c
new file mode 100644
index 000..25ef50d
--- /dev/null
+++ b/dix/stubmain.c
@@ -0,0 +1,33 @@
+/***
+
+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 (including the next
+paragraph) 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 SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.
+
+**/
+
+int dix_main(int argc, char *argv[], char *envp[]);
+
+/*
+  A default implementation of main, which can be overridden by the DDX
+ */
+int
+main(int argc, char *argv[], char *envp[])
+{
+return dix_main(argc, argv, envp);
+}
diff --git a/hw/xwin/InitOutput.c b/hw/xwin/InitOutput.c
index 4bcd3a0..fe8dcd4 100644
--- a/hw/xwin/InitOutput.c
+++ b/hw/xwin/InitOutput.c
@@ -186,6 +186,12 @@ ddxBeforeReset(void)
 }
 #endif
 
+int
+main(int argc, char *argv[], char *envp[])
+{
+return dix_main(argc, argv, envp);
+}
+
 /* See Porting Layer Definition - p. 57 */
 void
 ddxGiveUp(enum ExitCode error)
diff --git a/hw/xwin/Makefile.am b/hw/xwin/Makefile.am
index 3fcaf9d..6aac6a9 100644
--- a/hw/xwin/Makefile.am
+++ b/hw/xwin/Makefile.am
@@ -149,8 +149,8 @@ INCLUDES = -I$(top_srcdir)/miext/rootless
 
 XWIN_SYS_LIBS += -ldxguid
 
-XWin_DEPENDENCIES = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_LIBS) 
$(MAIN_LIB) $(XSERVER_LIBS)
-XWin_LDADD = $(MULTIWINDOWEXTWM_LIBS)

[PATCH 3/7] glx: Synchronize Xserver glx/rensize.c with mesa src/glx/compsize.c

2012-11-08 Thread Jon TURNEY

https://bugs.freedesktop.org/show_bug.cgi?id=30102

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
Reviewed-by: Colin Harrison colin.harri...@virgin.net
---
 glx/rensize.c |2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/glx/rensize.c b/glx/rensize.c
index bcc3a53..a78e60c 100644
--- a/glx/rensize.c
+++ b/glx/rensize.c
@@ -226,12 +226,14 @@ __glXImageSize(GLenum format, GLenum type, GLenum target,
 case GL_INTENSITY:
 elementsPerGroup = 1;
 break;
+case GL_RG:
 case GL_422_EXT:
 case GL_422_REV_EXT:
 case GL_422_AVERAGE_EXT:
 case GL_422_REV_AVERAGE_EXT:
 case GL_DEPTH_STENCIL_NV:
 case GL_DEPTH_STENCIL_MESA:
+case GL_YCBCR_422_APPLE:
 case GL_YCBCR_MESA:
 case GL_LUMINANCE_ALPHA:
 elementsPerGroup = 2;
-- 
1.7.9

___
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 4/7] Move pseudoramiX code where it can be shared between Xwin and Xquartz

2012-11-08 Thread Jon TURNEY

Move pseudoramiX code to a separate top-level directory.  Link Xwin and Xquartz
with libPseudoramiX

I'm not sure moving this to a top-level directory is appropriate, but I'm not
sure where else it fits.

Future work: pseudoramiX can probably be consolidated with the rrxinerama code
(which I think provides fake xinerama data when real XINERAMA is disabled and we
only have one screen)

v2: fix distcheck

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
 Makefile.am   |2 ++
 configure.ac  |1 +
 hw/xquartz/Makefile.am|4 ++--
 hw/xwin/Makefile.am   |1 +
 pseudoramiX/Makefile.am   |7 +++
 {hw/xquartz = pseudoramiX}/pseudoramiX.c |0
 {hw/xquartz = pseudoramiX}/pseudoramiX.h |0
 7 files changed, 13 insertions(+), 2 deletions(-)
 create mode 100644 pseudoramiX/Makefile.am
 rename {hw/xquartz = pseudoramiX}/pseudoramiX.c (100%)
 rename {hw/xquartz = pseudoramiX}/pseudoramiX.h (100%)

diff --git a/Makefile.am b/Makefile.am
index 9a62853..7be4271 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -27,6 +27,7 @@ SUBDIRS = \
Xext \
miext \
os \
+   pseudoramiX \
randr \
render  \
Xi \
@@ -77,6 +78,7 @@ DIST_SUBDIRS = \
Xext \
miext \
os \
+   pseudoramiX \
randr \
render  \
Xi \
diff --git a/configure.ac b/configure.ac
index e6f9cb7..3ad9ca5 100644
--- a/configure.ac
+++ b/configure.ac
@@ -2237,6 +2237,7 @@ miext/damage/Makefile
 miext/shadow/Makefile
 miext/rootless/Makefile
 os/Makefile
+pseudoramiX/Makefile
 randr/Makefile
 render/Makefile
 xkb/Makefile
diff --git a/hw/xquartz/Makefile.am b/hw/xquartz/Makefile.am
index a7cc012..2e894d7 100644
--- a/hw/xquartz/Makefile.am
+++ b/hw/xquartz/Makefile.am
@@ -28,7 +28,6 @@ libXquartz_la_SOURCES = \
darwinEvents.c \
darwinXinput.c \
keysym2ucs.c \
-   pseudoramiX.c \
quartz.c \
quartzCocoa.m \
quartzKeyboard.c \
@@ -36,6 +35,8 @@ libXquartz_la_SOURCES = \
quartzRandR.c \
console_redirect.c
 
+libXquartz_la_LIBADD = $(top_builddir)/pseudoramiX/libPseudoramiX.la
+
 EXTRA_DIST = \
X11Application.h \
X11Controller.h \
@@ -44,7 +45,6 @@ EXTRA_DIST = \
darwinfb.h \
darwinEvents.h \
keysym2ucs.h \
-   pseudoramiX.h \
quartz.h \
quartzCommon.h \
quartzKeyboard.h \
diff --git a/hw/xwin/Makefile.am b/hw/xwin/Makefile.am
index 6aac6a9..e00434c 100644
--- a/hw/xwin/Makefile.am
+++ b/hw/xwin/Makefile.am
@@ -148,6 +148,7 @@ XWin_SOURCES = $(SRCS)
 INCLUDES = -I$(top_srcdir)/miext/rootless
 
 XWIN_SYS_LIBS += -ldxguid
+XWIN_LIBS +=   $(top_builddir)/pseudoramiX/libPseudoramiX.la
 
 XWin_DEPENDENCIES = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_LIBS) 
$(XSERVER_LIBS)
 XWin_LDADD = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_GLX_LINK_FLAGS) 
$(XWIN_LIBS) $(XSERVER_LIBS) $(XSERVER_SYS_LIBS) $(XWIN_SYS_LIBS)
diff --git a/pseudoramiX/Makefile.am b/pseudoramiX/Makefile.am
new file mode 100644
index 000..17b664b
--- /dev/null
+++ b/pseudoramiX/Makefile.am
@@ -0,0 +1,7 @@
+# Fake Xinerama extension
+
+AM_CFLAGS = $(DIX_CFLAGS)
+
+noinst_LTLIBRARIES = libPseudoramiX.la
+
+libPseudoramiX_la_SOURCES = pseudoramiX.c pseudoramiX.h
diff --git a/hw/xquartz/pseudoramiX.c b/pseudoramiX/pseudoramiX.c
similarity index 100%
rename from hw/xquartz/pseudoramiX.c
rename to pseudoramiX/pseudoramiX.c
diff --git a/hw/xquartz/pseudoramiX.h b/pseudoramiX/pseudoramiX.h
similarity index 100%
rename from hw/xquartz/pseudoramiX.h
rename to pseudoramiX/pseudoramiX.h
-- 
1.7.9

___
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 5/7] Build dpmsstubs.c once as a convenience library, rather than once for each DDX which wants to use it

2012-11-08 Thread Jon TURNEY

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
 Xext/Makefile.am |4 +++-
 hw/vfb/Makefile.am   |4 ++--
 hw/xnest/Makefile.am |4 ++--
 hw/xwin/Makefile.am  |4 ++--
 test/Makefile.am |2 +-
 5 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/Xext/Makefile.am b/Xext/Makefile.am
index 8e31f13..a9a4468 100644
--- a/Xext/Makefile.am
+++ b/Xext/Makefile.am
@@ -1,4 +1,4 @@
-noinst_LTLIBRARIES = libXext.la
+noinst_LTLIBRARIES = libXext.la libXextdpmsstubs.la
 
 AM_CFLAGS = $(DIX_CFLAGS)
 
@@ -96,6 +96,8 @@ endif
 libXext_la_SOURCES =   $(BUILTIN_SRCS)
 libXext_la_LIBADD =$(BUILTIN_LIBS)
 
+libXextdpmsstubs_la_SOURCES = dpmsstubs.c
+
 EXTRA_DIST = \
$(MITSHM_SRCS) \
$(XV_SRCS) \
diff --git a/hw/vfb/Makefile.am b/hw/vfb/Makefile.am
index 9f4992c..c9f6957 100644
--- a/hw/vfb/Makefile.am
+++ b/hw/vfb/Makefile.am
@@ -9,7 +9,6 @@ AM_CFLAGS = -DHAVE_DIX_CONFIG_H \
 
 SRCS = InitInput.c \
InitOutput.c \
-   $(top_srcdir)/Xext/dpmsstubs.c \
$(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c
 
@@ -22,7 +21,8 @@ XVFB_LIBS = \
 @XVFB_LIBS@ \
libfbcmap.a \
$(MAIN_LIB) \
-   $(XSERVER_LIBS)
+   $(XSERVER_LIBS) \
+   $(top_builddir)/Xext/libXextdpmsstubs.la
 
 Xvfb_LDADD = $(XVFB_LIBS) $(XVFB_SYS_LIBS) $(XSERVER_SYS_LIBS)
 Xvfb_DEPENDENCIES = $(XVFB_LIBS)
diff --git a/hw/xnest/Makefile.am b/hw/xnest/Makefile.am
index 3c099cd..c5560fd 100644
--- a/hw/xnest/Makefile.am
+++ b/hw/xnest/Makefile.am
@@ -42,7 +42,6 @@ SRCS =Args.c \
XNPixmap.h \
XNWindow.h \
xnest-config.h \
-   $(top_srcdir)/Xext/dpmsstubs.c \
$(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c
 
@@ -51,7 +50,8 @@ libfbcmap_a_CFLAGS = $(AM_CFLAGS)
 
 XNEST_LIBS = \
 @XNEST_LIBS@ \
-   libfbcmap.a
+   libfbcmap.a \
+   $(top_builddir)/Xext/libXextdpmsstubs.la
 
 Xnest_SOURCES = $(SRCS)
 
diff --git a/hw/xwin/Makefile.am b/hw/xwin/Makefile.am
index e00434c..a5f2348 100644
--- a/hw/xwin/Makefile.am
+++ b/hw/xwin/Makefile.am
@@ -122,7 +122,6 @@ SRCS =  InitInput.c \
winresource.h \
winwindow.h \
XWin.rc \
-   $(top_srcdir)/Xext/dpmsstubs.c \
$(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c \
$(top_srcdir)/fb/fbcmap_mi.c \
@@ -148,7 +147,8 @@ XWin_SOURCES = $(SRCS)
 INCLUDES = -I$(top_srcdir)/miext/rootless
 
 XWIN_SYS_LIBS += -ldxguid
-XWIN_LIBS +=   $(top_builddir)/pseudoramiX/libPseudoramiX.la
+XWIN_LIBS +=   $(top_builddir)/pseudoramiX/libPseudoramiX.la \
+$(top_builddir)/Xext/libXextdpmsstubs.la
 
 XWin_DEPENDENCIES = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_LIBS) 
$(XSERVER_LIBS)
 XWin_LDADD = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_GLX_LINK_FLAGS) 
$(XWIN_LIBS) $(XSERVER_LIBS) $(XSERVER_SYS_LIBS) $(XWIN_SYS_LIBS)
diff --git a/test/Makefile.am b/test/Makefile.am
index 34f53fc..0c1e595 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -68,7 +68,6 @@ else
 nodist_libxservertest_la_SOURCES = \
 ddxstubs.c \
 $(top_srcdir)/mi/miinitext.c \
-$(top_srcdir)/Xext/dpmsstubs.c \
 $(top_srcdir)/Xi/stubs.c
 
 libxservertest_la_LIBADD += \
@@ -80,6 +79,7 @@ libxservertest_la_LIBADD += \
 $(top_builddir)/randr/librandr.la \
 $(top_builddir)/render/librender.la \
 $(top_builddir)/Xext/libXext.la \
+$(top_builddir)/Xext/libXextdpmsstubs.la \
 $(top_builddir)/Xi/libXi.la \
 $(top_builddir)/xfixes/libxfixes.la \
 $(top_builddir)/xkb/libxkb.la \
-- 
1.7.9

___
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 6/7] Build Xi/stubs.c once as a convenience library, rather than once for each DDX which wants to use it

2012-11-08 Thread Jon TURNEY

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
 Xi/Makefile.am   |5 +++--
 hw/vfb/Makefile.am   |4 ++--
 hw/xnest/Makefile.am |4 ++--
 hw/xwin/Makefile.am  |4 ++--
 test/Makefile.am |4 ++--
 5 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/Xi/Makefile.am b/Xi/Makefile.am
index 69c7886..026824b 100644
--- a/Xi/Makefile.am
+++ b/Xi/Makefile.am
@@ -1,4 +1,4 @@
-noinst_LTLIBRARIES = libXi.la
+noinst_LTLIBRARIES = libXi.la libXistubs.la
 
 AM_CFLAGS = $(DIX_CFLAGS)
 
@@ -105,4 +105,5 @@ libXi_la_SOURCES =  \
xiwarppointer.c \
xiwarppointer.h
 
-EXTRA_DIST = stubs.c
+libXistubs_la_SOURCES =\
+   stubs.c
diff --git a/hw/vfb/Makefile.am b/hw/vfb/Makefile.am
index c9f6957..a02a438 100644
--- a/hw/vfb/Makefile.am
+++ b/hw/vfb/Makefile.am
@@ -9,7 +9,6 @@ AM_CFLAGS = -DHAVE_DIX_CONFIG_H \
 
 SRCS = InitInput.c \
InitOutput.c \
-   $(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c
 
 libfbcmap_a_CFLAGS = $(AM_CFLAGS)
@@ -22,7 +21,8 @@ XVFB_LIBS = \
libfbcmap.a \
$(MAIN_LIB) \
$(XSERVER_LIBS) \
-   $(top_builddir)/Xext/libXextdpmsstubs.la
+   $(top_builddir)/Xext/libXextdpmsstubs.la \
+   $(top_builddir)/Xi/libXistubs.la
 
 Xvfb_LDADD = $(XVFB_LIBS) $(XVFB_SYS_LIBS) $(XSERVER_SYS_LIBS)
 Xvfb_DEPENDENCIES = $(XVFB_LIBS)
diff --git a/hw/xnest/Makefile.am b/hw/xnest/Makefile.am
index c5560fd..bb8d3d8 100644
--- a/hw/xnest/Makefile.am
+++ b/hw/xnest/Makefile.am
@@ -42,7 +42,6 @@ SRCS =Args.c \
XNPixmap.h \
XNWindow.h \
xnest-config.h \
-   $(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c
 
 libfbcmap_a_SOURCES = $(top_srcdir)/fb/fbcmap_mi.c
@@ -51,7 +50,8 @@ libfbcmap_a_CFLAGS = $(AM_CFLAGS)
 XNEST_LIBS = \
 @XNEST_LIBS@ \
libfbcmap.a \
-   $(top_builddir)/Xext/libXextdpmsstubs.la
+   $(top_builddir)/Xext/libXextdpmsstubs.la \
+   $(top_builddir)/Xi/libXistubs.la
 
 Xnest_SOURCES = $(SRCS)
 
diff --git a/hw/xwin/Makefile.am b/hw/xwin/Makefile.am
index a5f2348..2e45c57 100644
--- a/hw/xwin/Makefile.am
+++ b/hw/xwin/Makefile.am
@@ -122,7 +122,6 @@ SRCS =  InitInput.c \
winresource.h \
winwindow.h \
XWin.rc \
-   $(top_srcdir)/Xi/stubs.c \
$(top_srcdir)/mi/miinitext.c \
$(top_srcdir)/fb/fbcmap_mi.c \
$(SRCS_CLIPBOARD) \
@@ -148,7 +147,8 @@ INCLUDES = -I$(top_srcdir)/miext/rootless
 
 XWIN_SYS_LIBS += -ldxguid
 XWIN_LIBS +=   $(top_builddir)/pseudoramiX/libPseudoramiX.la \
-$(top_builddir)/Xext/libXextdpmsstubs.la
+$(top_builddir)/Xext/libXextdpmsstubs.la \
+$(top_builddir)/Xi/libXistubs.la
 
 XWin_DEPENDENCIES = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_LIBS) 
$(XSERVER_LIBS)
 XWin_LDADD = $(MULTIWINDOWEXTWM_LIBS) $(XWIN_GLX_LIBS) $(XWIN_GLX_LINK_FLAGS) 
$(XWIN_LIBS) $(XSERVER_LIBS) $(XSERVER_SYS_LIBS) $(XWIN_SYS_LIBS)
diff --git a/test/Makefile.am b/test/Makefile.am
index 0c1e595..06e5538 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -67,8 +67,7 @@ endif
 else
 nodist_libxservertest_la_SOURCES = \
 ddxstubs.c \
-$(top_srcdir)/mi/miinitext.c \
-$(top_srcdir)/Xi/stubs.c
+$(top_srcdir)/mi/miinitext.c
 
 libxservertest_la_LIBADD += \
 $(top_builddir)/damageext/libdamageext.la \
@@ -81,6 +80,7 @@ libxservertest_la_LIBADD += \
 $(top_builddir)/Xext/libXext.la \
 $(top_builddir)/Xext/libXextdpmsstubs.la \
 $(top_builddir)/Xi/libXi.la \
+$(top_builddir)/Xi/libXistubs.la \
 $(top_builddir)/xfixes/libxfixes.la \
 $(top_builddir)/xkb/libxkb.la \
 $(top_builddir)/xkb/libxkbstubs.la
-- 
1.7.9

___
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 7/7] Add a screen PaintWindow wrapper

2012-11-08 Thread Jon TURNEY

Revisit b3415187e92960cbff784108b5a3a8d130dc34c5, which shoves some rootless
specific code directly into miPaintWindow.

Unfortunately as written, this means that dix mi/libmi (when built with ROOTLESS
defined) must be linked into a ddx which links with miext/rootless/librootless,
which makes it impossible to build rootless supporting DDX (XWin and XQuartz)
and non-rootless supporting DDX (the rest) at the same time.  This is a pain
when developing for XWin, as you have to re-./configure to check if changes have
broken other DDXs, and is also a bit inconvenient for packaging and
tinderboxing.

Instead, add a way to wrap the PaintWindow function, so DDXs which have the
rootless extension can install a wrapper which does the the neccesary checking
which rootless needs to do.  DDXs which don't support rootless won't install the
hook, but the DIX remains compatible with them, so they can be built at the same
time.

This is kind of reverting part of what e4d11e58ce349dfe6af2f73ff341317f9b39684c
Remove the PaintWindow optimization did in removing the PaintWindowBackground
and PaintWindowBorder hooks, but it doesn't resurrect the problematic
optimization which made use of those hooks.

See [1] for admission that code as it stands is a horrible hack :-). But some
comments in that email make me wonder if this code is in the right place at all.

[1] http://lists.macosforge.org/pipermail/xquartz-dev/2009-September/002551.html

Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
 composite/compwindow.c  |2 +-
 dix/window.c|3 +--
 hw/dmx/doc/dmx.xml  |2 +-
 hw/xquartz/quartz.c |2 +-
 hw/xwin/winrandr.c  |2 +-
 include/scrnintstr.h|5 +
 mi/miexpose.c   |   28 ++--
 mi/mioverlay.c  |   12 ++--
 mi/miscrinit.c  |1 +
 mi/miwindow.c   |4 ++--
 miext/rootless/rootlessCommon.h |1 +
 miext/rootless/rootlessScreen.c |3 ++-
 miext/rootless/rootlessWindow.c |   35 +++
 miext/rootless/rootlessWindow.h |1 +
 14 files changed, 60 insertions(+), 41 deletions(-)

diff --git a/composite/compwindow.c b/composite/compwindow.c
index 0be7a1b..f61600e 100644
--- a/composite/compwindow.c
+++ b/composite/compwindow.c
@@ -104,7 +104,7 @@ compRepaintBorder(ClientPtr pClient, pointer closure)
 
 RegionNull(exposed);
 RegionSubtract(exposed, pWindow-borderClip, pWindow-winSize);
-miPaintWindow(pWindow, exposed, PW_BORDER);
+(*pWindow-drawable.pScreen-PaintWindow)(pWindow, exposed, 
PW_BORDER);
 RegionUninit(exposed);
 }
 return TRUE;
diff --git a/dix/window.c b/dix/window.c
index e70531a..c78d171 100644
--- a/dix/window.c
+++ b/dix/window.c
@@ -122,7 +122,6 @@ Equipment Corporation.
 #endif
 #include dixevents.h
 #include globals.h
-#include mi.h /* miPaintWindow */
 #ifdef COMPOSITE
 #include compint.h
 #endif
@@ -1466,7 +1465,7 @@ ChangeWindowAttributes(WindowPtr pWin, Mask vmask, XID 
*vlist, ClientPtr client)
 
 RegionNull(exposed);
 RegionSubtract(exposed, pWin-borderClip, pWin-winSize);
-miPaintWindow(pWin, exposed, PW_BORDER);
+(*pWin-drawable.pScreen-PaintWindow)(pWin, exposed, PW_BORDER);
 RegionUninit(exposed);
 }
 return error;
diff --git a/hw/dmx/doc/dmx.xml b/hw/dmx/doc/dmx.xml
index 6d7df41..c6c8544 100644
--- a/hw/dmx/doc/dmx.xml
+++ b/hw/dmx/doc/dmx.xml
@@ -1268,7 +1268,7 @@ default routine is chosen during GC validation.
 
 paraNote that some pointers to functions that draw to the screen are
 stored in the Screen structure.  They include GetImage(), GetSpans(),
-CopyWindow() and RestoreAreas().
+PaintWindow(), CopyWindow() and RestoreAreas().
 /para
 /sect3
 
diff --git a/hw/xquartz/quartz.c b/hw/xquartz/quartz.c
index 31b0e61..e7f30e6 100644
--- a/hw/xquartz/quartz.c
+++ b/hw/xquartz/quartz.c
@@ -300,7 +300,7 @@ QuartzUpdateScreens(void)
 quartzProcs-UpdateScreen(pScreen);
 
 /* miPaintWindow needs to be called after RootlessUpdateScreenPixmap (from 
xprUpdateScreen) */
-miPaintWindow(pRoot, pRoot-borderClip, PW_BACKGROUND);
+pScreen-PaintWindow(pRoot, pRoot-borderClip, PW_BACKGROUND);
 
 /* Tell RandR about the new size, so new connections get the correct info 
*/
 RRScreenSizeNotify(pScreen);
diff --git a/hw/xwin/winrandr.c b/hw/xwin/winrandr.c
index 1b34085..5624e75 100644
--- a/hw/xwin/winrandr.c
+++ b/hw/xwin/winrandr.c
@@ -104,7 +104,7 @@ winDoRandRScreenSetSize(ScreenPtr pScreen,
 SetRootClip(pScreen, TRUE);
 
 // and arrange for it to be repainted
-miPaintWindow(pRoot, pRoot-borderClip, PW_BACKGROUND);
+pScreen-PaintWindow(pRoot, pRoot-borderClip, PW_BACKGROUND);
 
 /* Indicate that a screen size change took place */
 RRScreenSizeNotify(pScreen);
diff --git a/include/scrnintstr.h b/include/scrnintstr.h
index

[PATCH 1/4] xfree86: Remove some unused variables

2012-11-08 Thread Thierry Reding

GCC complains about these variables being set but not used, so they can
just as well be removed.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 hw/xfree86/common/xf86Helper.c |  4 +---
 hw/xfree86/modes/xf86RandR12.c | 14 --
 2 files changed, 1 insertion(+), 17 deletions(-)

diff --git a/hw/xfree86/common/xf86Helper.c b/hw/xfree86/common/xf86Helper.c
index 721159d..841b8d5 100644
--- a/hw/xfree86/common/xf86Helper.c
+++ b/hw/xfree86/common/xf86Helper.c
@@ -421,7 +421,6 @@ xf86SetDepthBpp(ScrnInfoPtr scrp, int depth, int dummy, int 
fbbpp,
  * Check for DefaultDepth and DefaultFbBpp options in the
  * Device sections.
  */
-int i;
 GDevPtr device;
 Bool found = FALSE;
 
@@ -1068,9 +1067,8 @@ void
 xf86EnableDisableFBAccess(ScrnInfoPtr pScrnInfo, Bool enable)
 {
 ScreenPtr pScreen = pScrnInfo-pScreen;
-PixmapPtr pspix;
 
-pspix = (*pScreen-GetScreenPixmap) (pScreen);
+(*pScreen-GetScreenPixmap) (pScreen);
 if (enable) {
 /*
  * Restore all of the clip lists on the screen
diff --git a/hw/xfree86/modes/xf86RandR12.c b/hw/xfree86/modes/xf86RandR12.c
index 3530abf..6f487bb 100644
--- a/hw/xfree86/modes/xf86RandR12.c
+++ b/hw/xfree86/modes/xf86RandR12.c
@@ -453,7 +453,6 @@ xf86RandR12GetInfo(ScreenPtr pScreen, Rotation * rotations)
 ScrnInfoPtr scrp = xf86ScreenToScrn(pScreen);
 XF86RandRInfoPtr randrp = XF86RANDRINFO(pScreen);
 DisplayModePtr mode;
-int refresh0 = 60;
 int maxX = 0, maxY = 0;
 
 *rotations = randrp-supported_rotations;
@@ -478,8 +477,6 @@ xf86RandR12GetInfo(ScreenPtr pScreen, Rotation * rotations)
 if (maxY  mode-VDisplay)
 maxY = mode-VDisplay;
 }
-if (mode == scrp-modes)
-refresh0 = refresh;
 pSize = RRRegisterSize(pScreen,
mode-HDisplay, mode-VDisplay,
randrp-mmWidth, randrp-mmHeight);
@@ -683,7 +680,6 @@ xf86RandR12ScreenSetSize(ScreenPtr pScreen,
 WindowPtr pRoot = pScreen-root;
 PixmapPtr pScrnPix;
 Bool ret = FALSE;
-Bool panning = FALSE;
 int c;
 
 if (xf86RandR12Key) {
@@ -715,7 +711,6 @@ xf86RandR12ScreenSetSize(ScreenPtr pScreen,
 crtc-panningTrackingArea.y2 += height - pScreen-height;
 xf86RandR13VerifyPanningArea(crtc, width, height);
 xf86RandR13Pan(crtc, randrp-pointerX, randrp-pointerY);
-   panning = TRUE;
 }
 }
 
@@ -945,8 +940,6 @@ xf86RandR12SetRotations(ScreenPtr pScreen, Rotation 
rotations)
 void
 xf86RandR12SetTransformSupport(ScreenPtr pScreen, Bool transforms)
 {
-XF86RandRInfoPtr randrp;
-
 #if RANDR_13_INTERFACE
 ScrnInfoPtr pScrn = xf86ScreenToScrn(pScreen);
 int c;
@@ -956,7 +949,6 @@ xf86RandR12SetTransformSupport(ScreenPtr pScreen, Bool 
transforms)
 if (xf86RandR12Key == NULL)
 return;
 
-randrp = XF86RANDRINFO(pScreen);
 #if RANDR_13_INTERFACE
 for (c = 0; c  config-num_crtc; c++) {
 xf86CrtcPtr crtc = config-crtc[c];
@@ -1454,7 +1446,6 @@ xf86RandR12SetInfo12(ScreenPtr pScreen)
 RRCrtcPtr *crtcs;
 int ncrtc;
 int o, c, l;
-RRCrtcPtr randr_crtc;
 int nclone;
 
 clones = malloc(config-num_output * sizeof(RROutputPtr));
@@ -1467,11 +1458,6 @@ xf86RandR12SetInfo12(ScreenPtr pScreen)
 if (output-possible_crtcs  (1  c))
 crtcs[ncrtc++] = config-crtc[c]-randr_crtc;
 
-if (output-crtc)
-randr_crtc = output-crtc-randr_crtc;
-else
-randr_crtc = NULL;
-
 if (!RROutputSetCrtcs(output-randr_output, crtcs, ncrtc)) {
 free(crtcs);
 free(clones);
-- 
1.8.0

___
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 2/4] xfree86: Override physical output size only if EDID is available

2012-11-08 Thread Thierry Reding

The output's physical dimensions may be set by the driver, which is the
case at least for xf86-video-modesetting. It makes no sense to overwrite
it with 0 if EDID is unavailable because it will cause the only source
of this information to be lost.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 hw/xfree86/modes/xf86Crtc.c | 6 --
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/hw/xfree86/modes/xf86Crtc.c b/hw/xfree86/modes/xf86Crtc.c
index 177f7ac..93c741c 100644
--- a/hw/xfree86/modes/xf86Crtc.c
+++ b/hw/xfree86/modes/xf86Crtc.c
@@ -2960,8 +2960,10 @@ xf86OutputSetEDID(xf86OutputPtr output, xf86MonPtr 
edid_mon)
 free(output-MonInfo);
 
 output-MonInfo = edid_mon;
-output-mm_width = 0;
-output-mm_height = 0;
+if (edid_mon) {
+output-mm_width = 0;
+output-mm_height = 0;
+}
 
 if (debug_modes) {
 xf86DrvMsg(scrn-scrnIndex, X_INFO, EDID for output %s\n,
-- 
1.8.0

___
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 4/4] HACK: xfree86: modes: Recompute DPI when the screen size changes

2012-11-08 Thread Thierry Reding

The X server start up with the assumption that the display has 96 DPI
unless configured otherwise. However under the proper circumstances, it
can compute the correct value from the information provided by output
resources.

This commit adds this computation directly in xf86RandR12ScreenSetSize()
because it is the only place where enough information is available. The
xf86SetDpi() helper cannot be reused because it is based on the
assumption that the DPI is either configured (on the command-line or via
the configuration file) or computes it based on EDID data only.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 hw/xfree86/modes/xf86RandR12.c | 15 +++
 1 file changed, 15 insertions(+)

diff --git a/hw/xfree86/modes/xf86RandR12.c b/hw/xfree86/modes/xf86RandR12.c
index e18f357..dfbeb8d 100644
--- a/hw/xfree86/modes/xf86RandR12.c
+++ b/hw/xfree86/modes/xf86RandR12.c
@@ -723,6 +723,21 @@ xf86RandR12ScreenSetSize(ScreenPtr pScreen,
 xf86SetViewport(pScreen, pScreen-width - 1, pScreen-height - 1);
 xf86SetViewport(pScreen, 0, 0);
 
+/* HACK: recompute DPI */
+if (mmWidth  0  mmHeight  0) {
+pScrn-xDpi = (width * 254) / (mmWidth * 10);
+pScrn-yDpi = (height * 254) / (mmHeight * 10);
+
+if (pScrn-xDpi  0  pScrn-yDpi = 0)
+pScrn-yDpi = pScrn-xDpi;
+
+if (pScrn-yDpi  0  pScrn-xDpi = 0)
+pScrn-xDpi = pScrn-yDpi;
+
+xf86DrvMsg(pScrn-scrnIndex, X_INFO, DPI set to (%d, %d)\n,
+   pScrn-xDpi, pScrn-yDpi);
+}
+
  finish:
 update_desktop_dimensions();
 
-- 
1.8.0

___
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 3/4] xfree86: Use physical dimensions provided by output

2012-11-08 Thread Thierry Reding

If an output provides its physical dimensions, they should be used in
preference to computing them on the assumption that the display has a
default DPI setting.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
 hw/xfree86/modes/xf86RandR12.c | 12 +---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/hw/xfree86/modes/xf86RandR12.c b/hw/xfree86/modes/xf86RandR12.c
index 6f487bb..e18f357 100644
--- a/hw/xfree86/modes/xf86RandR12.c
+++ b/hw/xfree86/modes/xf86RandR12.c
@@ -806,10 +806,16 @@ xf86RandR12CreateScreenResources(ScreenPtr pScreen)
 }
 else {
 /*
- * Otherwise, just set the screen to DEFAULT_DPI
+ * Otherwise, just set the screen to DEFAULT_DPI unless the
+ * output knows its physical dimensions.
  */
-mmWidth = width * 25.4 / DEFAULT_DPI;
-mmHeight = height * 25.4 / DEFAULT_DPI;
+if (output-mm_width == 0 || output-mm_height == 0) {
+mmWidth = width * 25.4 / DEFAULT_DPI;
+mmHeight = height * 25.4 / DEFAULT_DPI;
+} else {
+mmWidth = output-mm_width;
+mmHeight = output-mm_height;
+}
 }
 }
 xf86DrvMsg(pScrn-scrnIndex, X_INFO,
-- 
1.8.0

___
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 1/7] os: Remove any old logfile before trying to write to it

2012-11-08 Thread walter harms



Am 08.11.2012 14:41, schrieb Jon TURNEY:
 If we are not backing up logfiles, remove the old logfile before trying to 
 write
 a new logfile, as otherwise the operation may fail if the previous logfile was
 created by a different user.
 

what are the permissions of that file ?
normally you can  not remove any file that is not worn by your.
I would print an advive to use rm /logfile if the creation failed.

just my 2 cents

re,
 wh


 This change is useful when:
 - The DDX doesn't use the logfile backup mechanism (i.e. not Xorg)
 - The DDX is run by a non-root user, and then by a different non-root user
 - The logfile directory doesn't have the restricted-deletion flag set
 
 Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
 Acked-by: Yaakov Selkowitz yselkow...@users.sourceforge.net
 Reviewed-by: Yaakov Selkowitz yselkow...@users.sourceforge.net
 ---
  os/log.c |3 +++
  1 files changed, 3 insertions(+), 0 deletions(-)
 
 diff --git a/os/log.c b/os/log.c
 index 4820e9a..15271eb 100644
 --- a/os/log.c
 +++ b/os/log.c
 @@ -216,6 +216,9 @@ LogInit(const char *fname, const char *backup)
  free(oldLog);
  }
  }
 +else {
 +unlink(logFileName);
 +}
  if ((logFile = fopen(logFileName, w)) == NULL)
  FatalError(Cannot open log file \%s\\n, logFileName);
  setvbuf(logFile, NULL, _IONBF, 0);
___
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 1/7] os: Remove any old logfile before trying to write to it

2012-11-08 Thread Ángel González

On 08/11/12 18:21, walter harms wrote:
 Am 08.11.2012 14:41, schrieb Jon TURNEY:
 If we are not backing up logfiles, remove the old logfile before trying to 
 write
 a new logfile, as otherwise the operation may fail if the previous logfile 
 was
 created by a different user.

 
 what are the permissions of that file ?
 normally you can  not remove any file that is not worn by your.
 I would print an advive to use rm /logfile if the creation failed.
 
 just my 2 cents
 
 re,
  wh

It's not the permissions of the file what matters, but those of the
folder where it lives.
If you can write in that folder, and it doesn't have the sticky bit set,
you can also delete files.
Extended attributes and security modules could change that, obviously.

I find the commit message quite clear in what it expects and its
limitations.
I tried to make up a use case where the logfile was being used by a user
and removing it actually made it fail because it couldn't be recreated,
but then, the it's unlikely that unlink() would have worked.
It would have to be a setup with multiple users on a sticky folder.

___
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

Possible regression with GLX+Xvfb after 5f5bbbe?

2012-11-08 Thread Raphael Kubo da Costa

Hey there, everyone.

I've recently had to test some code which relies on GLX using Xvfb on
xorg-server 1.13.0, but it looks like it is not loaded anymore after
commit 5f5bbbe543f65c48ecbb5cce80116a86ca3fbe86.

It turns out that this commit removed a duplicate code path in
miinitext.c which was used by Xvfb and was responsible for calling
GlxExtensionInit(). Right now, even passing `+extension GLX' does not
help, since the contents of ExtensionToggleList are not read in
InitExtensions().

Furthermore, commit aad428b8e21c77397c623b78706eb64b1fea77c9, makes
attempts to call GlxExtensionInit() in miinitext.c fail when linking the
unit tests.

Before working on a patch (or hoping someone else does that :-), I'd
like to clarify whether this side-effect was not really intentional.

___
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 1/4] xfree86: Remove some unused variables

2012-11-08 Thread Aaron Plattner


These all seem fine to me.  For the series,

Reviewed-by: Aaron Plattner aplatt...@nvidia.com

-- Aaron

On 11/08/2012 05:50 AM, Thierry Reding wrote:

GCC complains about these variables being set but not used, so they can
just as well be removed.

Signed-off-by: Thierry Reding thierry.red...@avionic-design.de
---
  hw/xfree86/common/xf86Helper.c |  4 +---
  hw/xfree86/modes/xf86RandR12.c | 14 --
  2 files changed, 1 insertion(+), 17 deletions(-)

diff --git a/hw/xfree86/common/xf86Helper.c b/hw/xfree86/common/xf86Helper.c
index 721159d..841b8d5 100644
--- a/hw/xfree86/common/xf86Helper.c
+++ b/hw/xfree86/common/xf86Helper.c
@@ -421,7 +421,6 @@ xf86SetDepthBpp(ScrnInfoPtr scrp, int depth, int dummy, int 
fbbpp,
   * Check for DefaultDepth and DefaultFbBpp options in the
   * Device sections.
   */
-int i;
  GDevPtr device;
  Bool found = FALSE;

@@ -1068,9 +1067,8 @@ void
  xf86EnableDisableFBAccess(ScrnInfoPtr pScrnInfo, Bool enable)
  {
  ScreenPtr pScreen = pScrnInfo-pScreen;
-PixmapPtr pspix;

-pspix = (*pScreen-GetScreenPixmap) (pScreen);
+(*pScreen-GetScreenPixmap) (pScreen);
  if (enable) {
  /*
   * Restore all of the clip lists on the screen
diff --git a/hw/xfree86/modes/xf86RandR12.c b/hw/xfree86/modes/xf86RandR12.c
index 3530abf..6f487bb 100644
--- a/hw/xfree86/modes/xf86RandR12.c
+++ b/hw/xfree86/modes/xf86RandR12.c
@@ -453,7 +453,6 @@ xf86RandR12GetInfo(ScreenPtr pScreen, Rotation * rotations)
  ScrnInfoPtr scrp = xf86ScreenToScrn(pScreen);
  XF86RandRInfoPtr randrp = XF86RANDRINFO(pScreen);
  DisplayModePtr mode;
-int refresh0 = 60;
  int maxX = 0, maxY = 0;

  *rotations = randrp-supported_rotations;
@@ -478,8 +477,6 @@ xf86RandR12GetInfo(ScreenPtr pScreen, Rotation * rotations)
  if (maxY  mode-VDisplay)
  maxY = mode-VDisplay;
  }
-if (mode == scrp-modes)
-refresh0 = refresh;
  pSize = RRRegisterSize(pScreen,
 mode-HDisplay, mode-VDisplay,
 randrp-mmWidth, randrp-mmHeight);
@@ -683,7 +680,6 @@ xf86RandR12ScreenSetSize(ScreenPtr pScreen,
  WindowPtr pRoot = pScreen-root;
  PixmapPtr pScrnPix;
  Bool ret = FALSE;
-Bool panning = FALSE;
  int c;

  if (xf86RandR12Key) {
@@ -715,7 +711,6 @@ xf86RandR12ScreenSetSize(ScreenPtr pScreen,
  crtc-panningTrackingArea.y2 += height - pScreen-height;
  xf86RandR13VerifyPanningArea(crtc, width, height);
  xf86RandR13Pan(crtc, randrp-pointerX, randrp-pointerY);
-   panning = TRUE;
  }
  }

@@ -945,8 +940,6 @@ xf86RandR12SetRotations(ScreenPtr pScreen, Rotation 
rotations)
  void
  xf86RandR12SetTransformSupport(ScreenPtr pScreen, Bool transforms)
  {
-XF86RandRInfoPtr randrp;
-
  #if RANDR_13_INTERFACE
  ScrnInfoPtr pScrn = xf86ScreenToScrn(pScreen);
  int c;
@@ -956,7 +949,6 @@ xf86RandR12SetTransformSupport(ScreenPtr pScreen, Bool 
transforms)
  if (xf86RandR12Key == NULL)
  return;

-randrp = XF86RANDRINFO(pScreen);
  #if RANDR_13_INTERFACE
  for (c = 0; c  config-num_crtc; c++) {
  xf86CrtcPtr crtc = config-crtc[c];
@@ -1454,7 +1446,6 @@ xf86RandR12SetInfo12(ScreenPtr pScreen)
  RRCrtcPtr *crtcs;
  int ncrtc;
  int o, c, l;
-RRCrtcPtr randr_crtc;
  int nclone;

  clones = malloc(config-num_output * sizeof(RROutputPtr));
@@ -1467,11 +1458,6 @@ xf86RandR12SetInfo12(ScreenPtr pScreen)
  if (output-possible_crtcs  (1  c))
  crtcs[ncrtc++] = config-crtc[c]-randr_crtc;

-if (output-crtc)
-randr_crtc = output-crtc-randr_crtc;
-else
-randr_crtc = NULL;
-
  if (!RROutputSetCrtcs(output-randr_output, crtcs, ncrtc)) {
  free(crtcs);
  free(clones);



___
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 xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

2012-11-08 Thread Chase Douglas

On Mon, Oct 29, 2012 at 6:38 PM, Peter Hutterer peter.hutte...@who-t.netwrote:

 Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
 ---
  configure.ac | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

 diff --git a/configure.ac b/configure.ac
 index 59fd496..e3afde3 100644
 --- a/configure.ac
 +++ b/configure.ac
 @@ -8,7 +8,7 @@ AC_CONFIG_SRCDIR([Makefile.am])
  AC_CONFIG_MACRO_DIR([m4])

  # Initialize Automake
 -AM_INIT_AUTOMAKE([foreign dist-bzip2])
 +AM_INIT_AUTOMAKE([foreign dist-bzip2 no-dist-gzip])
  AM_MAINTAINER_MODE([enable])

  # Checks for programs.


Sorry for the delay in reviewing these. Ubuntu developer summit and time
off got in the way...

All the patches in this series look reasonable to me.

Reviewed-by: Chase Douglas chase.doug...@ubuntu.com
___
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 0/3] Fix pointer emulation on non-ownership windows

2012-11-08 Thread Chase Douglas

On Thu, Nov 1, 2012 at 10:17 PM, Peter Hutterer peter.hutte...@who-t.netwrote:


 https://bugs.freedesktop.org/show_bug.cgi?id=56557

 Pointer emulation on the root window does not work correctly. We currently
 only deliver to window owners, but no client owns the root window. The real
 bug here is that even though the events are registered for, the listener
 lookup fails and we never send emulated pointer events to the client.

 Fix this by storing the resource type of the listener with the touch so we
 know what to actually look up. And to make this work we need a resource
 type
 for active grabs, and we need to store implicit passive grabs in the
 resource system as well.

 oh, and having the second argument to FreeResource() be don't free the
 resource data if the resource type to remove equals this type is an
 interesting choice of an API.


I'm slowly losing my memory of the internals of X :(. While I can't make
out exactly how this works immediately, it looks like a proper solution all
around.

Reviewed-by: Chase Douglas chase.doug...@ubuntu.com
___
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 xorg-gtest] Expand on the default 'failed to open connection to display' error

2012-11-08 Thread Chase Douglas

On Mon, Nov 5, 2012 at 5:28 PM, Peter Hutterer peter.hutte...@who-t.netwrote:

 Virtually everyone trying the tests the first time will run into this issue
 since we cannot check if the ABI for dummy and whatever else is needed
 actually matches the server (well, we have a test for that, but unless you
 run that one first...)

 So be at least more verbose about the potential reasons and provide some
 basic debugging help.

 Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
 ---
  src/test.cpp | 11 +--
  1 file changed, 9 insertions(+), 2 deletions(-)

 diff --git a/src/test.cpp b/src/test.cpp
 index c7b69bc..a484787 100644
 --- a/src/test.cpp
 +++ b/src/test.cpp
 @@ -50,8 +50,15 @@ void xorg::testing::Test::SetUp() {
  dpy = d_-display_string.c_str();

d_-display = XOpenDisplay(dpy);
 -  if (!d_-display)
 -throw std::runtime_error(Failed to open connection to display);
 +  if (!d_-display) {
 +std::stringstream ss;
 +ss  Failed to open connection to display;
 +if (dpy != NULL) ss dpy;
 +ss  .\nThis usually means that your X server did not start
 properly.\n;
 +ss  Check the log file, or set XORG_GTEST_CHILD_STDOUT to see the
 server's\n
 +  error messages when starting.;
 +throw std::runtime_error(ss.str());
 +  }
  }

  void xorg::testing::Test::TearDown() {


This should be very helpful for newbies :).

Reviewed-by: Chase Douglas chase.doug...@ubuntu.com
___
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 libXi] man: add generation of missing man pages for XIGrabTouchBegin

2012-11-08 Thread Chase Douglas

On Tue, Nov 6, 2012 at 2:52 PM, Peter Hutterer peter.hutte...@who-t.netwrote:

 The man page itself already contained the description, but it was missing
 from NAME so the shadow man pages were not generated.

 Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
 ---
  man/Makefile.am  | 6 --
  man/XIGrabButton.txt | 3 +--
  2 files changed, 5 insertions(+), 4 deletions(-)

 diff --git a/man/Makefile.am b/man/Makefile.am
 index 28211a1..af63952 100644
 --- a/man/Makefile.am
 +++ b/man/Makefile.am
 @@ -89,7 +89,9 @@ XI2_shadows = \
 XIGetSelectedEvents.man \
 XIDeleteProperty.man\
 XIGetProperty.man   \
 -   XIFreeDeviceInfo.man
 +   XIFreeDeviceInfo.man\
 +   XIGrabTouchBegin.man\
 +   XIUngrabTouchBegin.man

  libman_shadows =   \
 XGetDeviceDontPropagateList.man \
 @@ -141,7 +143,7 @@ XDeleteDeviceProperty.man: XGetDeviceProperty.man

  # Dependencies for X Input v2 shadow pages
  XIUndefineCursor.man: XIDefineCursor.man
 -XIUngrabButton.man XIGrabKeycode.man XIUngrabKeycode.man: XIGrabButton.man
 +XIUngrabButton.man XIGrabKeycode.man XIUngrabKeycode.man
 XIGrabTouchBegin.man XIUngrabTouchBegin.man: XIGrabButton.man
  XIGetClientPointer.man: XISetClientPointer.man
  XIGetFocus.man: XISetFocus.man
  XIUngrabDevice.man: XIGrabDevice.man
 diff --git a/man/XIGrabButton.txt b/man/XIGrabButton.txt
 index a046ac7..c5100fb 100644
 --- a/man/XIGrabButton.txt
 +++ b/man/XIGrabButton.txt
 @@ -4,8 +4,7 @@ XIGRABBUTTON(libmansuffix)
  NAME
  

 -   XIGrabButton, XIUngrabButton, XIGrabKeycode, XIUngrabKeycode -
 -   grab/ungrab buttons or keys
 +   XIGrabButton, XIUngrabButton, XIGrabKeycode, XIUngrabKeycode,
 XIGrabTouchBegin, XIUngrabTouchBegin - grab/ungrab buttons or keys

  SYNOPSIS
  


Looks right.

Reviewed-by: Chase Douglas chase.doug...@ubuntu.com
___
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 xorg-gtest 2/2] xserver: install default X error handler

2012-11-08 Thread Chase Douglas

On Tue, Nov 6, 2012 at 7:57 PM, Peter Hutterer peter.hutte...@who-t.netwrote:

 Xlib's default error handler prints the error and calls exit(1). Tests that
 accidentally trigger an error thus quit without cleaning up properly.

 Install a default error handler that prints the basic info and continue
 with
 the test. Clients that expect to trigger errors should set a custom error
 handler.

 Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
 ---
  include/xorg/gtest/xorg-gtest-xserver.h | 13 ++
  src/xserver.cpp | 54 +++-
  test/xserver-test.cpp   | 73
 +
  3 files changed, 139 insertions(+), 1 deletion(-)

 diff --git a/include/xorg/gtest/xorg-gtest-xserver.h
 b/include/xorg/gtest/xorg-gtest-xserver.h
 index 8bf7996..11fc93d 100644
 --- a/include/xorg/gtest/xorg-gtest-xserver.h
 +++ b/include/xorg/gtest/xorg-gtest-xserver.h
 @@ -265,6 +265,19 @@ class XServer : public xorg::testing::Process {
   */
  static void RegisterXIOErrorHandler();

 +/**
 + * Install a default XErrorHandler. That error handler will cause a
 test
 + * failure if called.
 + *
 + * This function is called automatically by XServer::Start(). Usually,
 + * you will not need to call this function unless your test does not
 + * instantiate and Start() an XServer object.
 + *
 + * This function will only install a new error handler if the
 currently
 + * installed XErrorHandler is not the default handler used by Xlib.
 + */
 +static void RegisterXErrorHandler();
 +
private:
  struct Private;
  std::auto_ptrPrivate d_;
 diff --git a/src/xserver.cpp b/src/xserver.cpp
 index ad018a1..4faa8e9 100644
 --- a/src/xserver.cpp
 +++ b/src/xserver.cpp
 @@ -394,6 +394,40 @@ const std::string
 xorg::testing::XServer::GetVersion(void) {
return d_-version;
  }

 +static int _x_error_handler(Display *dpy, XErrorEvent *err)
 +{
 +  std::stringstream error;
 +  switch(err-error_code) {
 +case BadRequest: error  BadRequest; break;
 +case BadValue: error  BadValue; break;
 +case BadWindow: error  BadWindow; break;
 +case BadPixmap: error  BadPixmap; break;
 +case BadAtom: error  BadAtom; break;
 +case BadCursor: error  BadCursor; break;
 +case BadFont: error  BadFont; break;
 +case BadMatch: error  BadMatch; break;
 +case BadDrawable: error  BadDrawable; break;
 +case BadAccess: error  BadAccess; break;
 +case BadAlloc: error  BadAlloc; break;
 +case BadColor: error  BadColor; break;
 +case BadGC: error  BadGC; break;
 +case BadIDChoice: error  BadIDChoice; break;
 +case BadName: error  BadName; break;
 +case BadLength: error  BadLength; break;
 +case BadImplementation: error  BadImplementation; break;
 +default:
 +  error  err-error_code;
 +  break;
 +  }
 +
 +  ADD_FAILURE()  XError received:   error.str()  , request  
 +(int)err-request_code  (  (int)err-minor_code  ), detail: 
 + err-resourceid  \nThis error handler is likely to be triggered
 
 +more than once.\nCheck the first error for the real error;
 +  return 0;
 +}
 +
 +
  static int _x_io_error_handler(Display *dpy) _X_NORETURN;
  static int _x_io_error_handler(Display *dpy)
  {
 @@ -409,6 +443,15 @@ void xorg::testing::XServer::RegisterXIOErrorHandler()
  XSetIOErrorHandler(old_handler);
  }

 +void xorg::testing::XServer::RegisterXErrorHandler()
 +{
 +  XErrorHandler old_handler;
 +  old_handler = XSetErrorHandler(_x_error_handler);
 +
 +  if (old_handler != _XDefaultError)
 +XSetErrorHandler(old_handler);
 +}
 +
  void xorg::testing::XServer::Start(const std::string program) {
TestStartup();

 @@ -464,7 +507,15 @@ void xorg::testing::XServer::Start(const std::string
 program) {
  args.push_back(it-second);
  }

 -Process::Start(program.empty() ? d_-path_to_server : program, args);
 +std::string server_binary = program.empty() ? d_-path_to_server :
 program;
 +
 +if (getenv(XORG_GTEST_XSERVER_USE_VALGRIND)) {
 +  args.insert(args.begin(), server_binary);
 +  server_binary = valgrind;
 +  args.insert(args.begin(), --leak-check=full);
 +}


While this looks totally cool, it probably belongs in a separate commit :).

Without this hunk:

Reviewed-by: Chase Douglas chase.doug...@ubuntu.com

You can also add my Reviewed-by tag to a separate commit for just this
hunk. You might want to consider allowing for some standard options like
--show-reachable

-- Chase
___
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 xorg-gtest 1/2] gtest: add gtest-spi.h header

2012-11-08 Thread Chase Douglas

On Tue, Nov 6, 2012 at 7:57 PM, Peter Hutterer peter.hutte...@who-t.netwrote:

 This header is for self-testing, specifically EXPECT_FATAL_FAILURE

 Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
 ---
  gtest/include/Makefile.am   |   2 +-
  gtest/include/gtest/gtest-spi.h | 232
 
  2 files changed, 233 insertions(+), 1 deletion(-)
  create mode 100644 gtest/include/gtest/gtest-spi.h

 diff --git a/gtest/include/Makefile.am b/gtest/include/Makefile.am
 index 54d3088..6456b01 100644
 --- a/gtest/include/Makefile.am
 +++ b/gtest/include/Makefile.am
 @@ -22,4 +22,4 @@
  # SOFTWARE.
  #

 -nobase_include_HEADERS = gtest/gtest.h
 +nobase_include_HEADERS = gtest/gtest.h gtest/gtest-spi.h
 diff --git a/gtest/include/gtest/gtest-spi.h
 b/gtest/include/gtest/gtest-spi.h
 new file mode 100644
 index 000..f63fa9a
 --- /dev/null
 +++ b/gtest/include/gtest/gtest-spi.h
 @@ -0,0 +1,232 @@
 +// Copyright 2007, Google Inc.
 +// All rights reserved.
 +//
 +// Redistribution and use in source and binary forms, with or without
 +// modification, are permitted provided that the following conditions are
 +// met:
 +//
 +// * Redistributions of source code must retain the above copyright
 +// notice, this list of conditions and the following disclaimer.
 +// * Redistributions in binary form must reproduce the above
 +// copyright notice, this list of conditions and the following disclaimer
 +// in the documentation and/or other materials provided with the
 +// distribution.
 +// * Neither the name of Google Inc. nor the names of its
 +// contributors may be used to endorse or promote products derived from
 +// this software without specific prior written permission.
 +//
 +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 +// AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
 +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 +//
 +// Author: w...@google.com (Zhanyong Wan)
 +//
 +// Utilities for testing Google Test itself and code that uses Google Test
 +// (e.g. frameworks built on top of Google Test).
 +
 +#ifndef GTEST_INCLUDE_GTEST_GTEST_SPI_H_
 +#define GTEST_INCLUDE_GTEST_GTEST_SPI_H_
 +
 +#include gtest/gtest.h
 +
 +namespace testing {
 +
 +// This helper class can be used to mock out Google Test failure reporting
 +// so that we can test Google Test or code that builds on Google Test.
 +//
 +// An object of this class appends a TestPartResult object to the
 +// TestPartResultArray object given in the constructor whenever a Google
 Test
 +// failure is reported. It can either intercept only failures that are
 +// generated in the same thread that created this object or it can
 intercept
 +// all generated failures. The scope of this mock object can be
 controlled with
 +// the second argument to the two arguments constructor.
 +class GTEST_API_ ScopedFakeTestPartResultReporter
 +: public TestPartResultReporterInterface {
 + public:
 +  // The two possible mocking modes of this object.
 +  enum InterceptMode {
 +INTERCEPT_ONLY_CURRENT_THREAD,  // Intercepts only thread local
 failures.
 +INTERCEPT_ALL_THREADS   // Intercepts all failures.
 +  };
 +
 +  // The c'tor sets this object as the test part result reporter used
 +  // by Google Test.  The 'result' parameter specifies where to report the
 +  // results. This reporter will only catch failures generated in the
 current
 +  // thread. DEPRECATED
 +  explicit ScopedFakeTestPartResultReporter(TestPartResultArray* result);
 +
 +  // Same as above, but you can choose the interception scope of this
 object.
 +  ScopedFakeTestPartResultReporter(InterceptMode intercept_mode,
 +   TestPartResultArray* result);
 +
 +  // The d'tor restores the previous test part result reporter.
 +  virtual ~ScopedFakeTestPartResultReporter();
 +
 +  // Appends the TestPartResult object to the TestPartResultArray
 +  // received in the constructor.
 +  //
 +  // This method is from the TestPartResultReporterInterface
 +  // interface.
 +  virtual void ReportTestPartResult(const TestPartResult result);
 + private:
 +  void Init();
 +
 +  const InterceptMode intercept_mode_;
 +  TestPartResultReporterInterface* old_reporter_;
 +  TestPartResultArray* const result_;
 +
 +  GTEST_DISALLOW_COPY_AND_ASSIGN_(ScopedFakeTestPartResultReporter);

Re: [PATCH v2] xf86: Fix non-PCI configuration-less setups

2012-11-08 Thread Dave Airlie

On Wed, Oct 17, 2012 at 8:06 PM, Thierry Reding
thierry.red...@avionic-design.de wrote:
 For non-PCI video devices, such as those found on many ARM embedded
 systems, the X server currently requires the BusID option to specify the
 full path to the DRM device's sysfs node in order to properly match it
 against the probed platform devices.

 In order to allow X to start up properly if either the BusID option was
 omitted or no configuration is present at all, the first video device is
 used by default.

 Signed-off-by: Thierry Reding thierry.red...@avionic-design.de

Looks fine to me!

Reviewed-by: Dave Airlie airl...@redhat.com

Dave.
___
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 xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

2012-11-08 Thread Daniel Stone

On 30 October 2012 12:38, Peter Hutterer peter.hutte...@who-t.net wrote:

 --- a/configure.ac
 +++ b/configure.ac
 @@ -8,7 +8,7 @@ AC_CONFIG_SRCDIR([Makefile.am])
  AC_CONFIG_MACRO_DIR([m4])

  # Initialize Automake
 -AM_INIT_AUTOMAKE([foreign dist-bzip2])
 +AM_INIT_AUTOMAKE([foreign dist-bzip2 no-dist-gzip])


How about enabling .xz too?

Cheers,
Daniel
___
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 xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

2012-11-08 Thread Peter Hutterer

On Fri, Nov 09, 2012 at 01:44:16PM +1100, Daniel Stone wrote:
 On 30 October 2012 12:38, Peter Hutterer peter.hutte...@who-t.net wrote:
 
  --- a/configure.ac
  +++ b/configure.ac
  @@ -8,7 +8,7 @@ AC_CONFIG_SRCDIR([Makefile.am])
   AC_CONFIG_MACRO_DIR([m4])
 
   # Initialize Automake
  -AM_INIT_AUTOMAKE([foreign dist-bzip2])
  +AM_INIT_AUTOMAKE([foreign dist-bzip2 no-dist-gzip])
 
 
 How about enabling .xz too?

I disabled .gz because IMO one tarball is enough, we don't need two if one
is likely never used anyway. A similar situation would apply to xz, we'd
have two tarballs but no real need. And the space savings aren't that huge
that it switching from bz2 to xz alone really makes a difference.

365522 xorg-gtest-0.6.0.tar.bz2
308152 xorg-gtest-0.6.0.tar.xz

Cheers,
   Peter
___
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 xorg-gtest 2/2] xserver: install default X error handler

2012-11-08 Thread Peter Hutterer

On Thu, Nov 08, 2012 at 05:46:12PM -0800, Chase Douglas wrote:
 On Tue, Nov 6, 2012 at 7:57 PM, Peter Hutterer 
 peter.hutte...@who-t.netwrote:
 
  Xlib's default error handler prints the error and calls exit(1). Tests that
  accidentally trigger an error thus quit without cleaning up properly.
 
  Install a default error handler that prints the basic info and continue
  with
  the test. Clients that expect to trigger errors should set a custom error
  handler.
 
  Signed-off-by: Peter Hutterer peter.hutte...@who-t.net
  ---
   include/xorg/gtest/xorg-gtest-xserver.h | 13 ++
   src/xserver.cpp | 54 +++-
   test/xserver-test.cpp   | 73
  +
   3 files changed, 139 insertions(+), 1 deletion(-)
 
  diff --git a/include/xorg/gtest/xorg-gtest-xserver.h
  b/include/xorg/gtest/xorg-gtest-xserver.h
  index 8bf7996..11fc93d 100644
  --- a/include/xorg/gtest/xorg-gtest-xserver.h
  +++ b/include/xorg/gtest/xorg-gtest-xserver.h
  @@ -265,6 +265,19 @@ class XServer : public xorg::testing::Process {
*/
   static void RegisterXIOErrorHandler();
 
  +/**
  + * Install a default XErrorHandler. That error handler will cause a
  test
  + * failure if called.
  + *
  + * This function is called automatically by XServer::Start(). Usually,
  + * you will not need to call this function unless your test does not
  + * instantiate and Start() an XServer object.
  + *
  + * This function will only install a new error handler if the
  currently
  + * installed XErrorHandler is not the default handler used by Xlib.
  + */
  +static void RegisterXErrorHandler();
  +
 private:
   struct Private;
   std::auto_ptrPrivate d_;
  diff --git a/src/xserver.cpp b/src/xserver.cpp
  index ad018a1..4faa8e9 100644
  --- a/src/xserver.cpp
  +++ b/src/xserver.cpp
  @@ -394,6 +394,40 @@ const std::string
  xorg::testing::XServer::GetVersion(void) {
 return d_-version;
   }
 
  +static int _x_error_handler(Display *dpy, XErrorEvent *err)
  +{
  +  std::stringstream error;
  +  switch(err-error_code) {
  +case BadRequest: error  BadRequest; break;
  +case BadValue: error  BadValue; break;
  +case BadWindow: error  BadWindow; break;
  +case BadPixmap: error  BadPixmap; break;
  +case BadAtom: error  BadAtom; break;
  +case BadCursor: error  BadCursor; break;
  +case BadFont: error  BadFont; break;
  +case BadMatch: error  BadMatch; break;
  +case BadDrawable: error  BadDrawable; break;
  +case BadAccess: error  BadAccess; break;
  +case BadAlloc: error  BadAlloc; break;
  +case BadColor: error  BadColor; break;
  +case BadGC: error  BadGC; break;
  +case BadIDChoice: error  BadIDChoice; break;
  +case BadName: error  BadName; break;
  +case BadLength: error  BadLength; break;
  +case BadImplementation: error  BadImplementation; break;
  +default:
  +  error  err-error_code;
  +  break;
  +  }
  +
  +  ADD_FAILURE()  XError received:   error.str()  , request  
  +(int)err-request_code  (  (int)err-minor_code  ), detail: 
  + err-resourceid  \nThis error handler is likely to be triggered
  
  +more than once.\nCheck the first error for the real error;
  +  return 0;
  +}
  +
  +
   static int _x_io_error_handler(Display *dpy) _X_NORETURN;
   static int _x_io_error_handler(Display *dpy)
   {
  @@ -409,6 +443,15 @@ void xorg::testing::XServer::RegisterXIOErrorHandler()
   XSetIOErrorHandler(old_handler);
   }
 
  +void xorg::testing::XServer::RegisterXErrorHandler()
  +{
  +  XErrorHandler old_handler;
  +  old_handler = XSetErrorHandler(_x_error_handler);
  +
  +  if (old_handler != _XDefaultError)
  +XSetErrorHandler(old_handler);
  +}
  +
   void xorg::testing::XServer::Start(const std::string program) {
 TestStartup();
 
  @@ -464,7 +507,15 @@ void xorg::testing::XServer::Start(const std::string
  program) {
   args.push_back(it-second);
   }
 
  -Process::Start(program.empty() ? d_-path_to_server : program, args);
  +std::string server_binary = program.empty() ? d_-path_to_server :
  program;
  +
  +if (getenv(XORG_GTEST_XSERVER_USE_VALGRIND)) {
  +  args.insert(args.begin(), server_binary);
  +  server_binary = valgrind;
  +  args.insert(args.begin(), --leak-check=full);
  +}
 
 
 While this looks totally cool, it probably belongs in a separate commit :).

oh, that's where that hunk ended up :) I was wondering which branch it was
on, and could not find the git commit referring to it. got committed by
accident.

 Without this hunk:
 
 Reviewed-by: Chase Douglas chase.doug...@ubuntu.com
 
 You can also add my Reviewed-by tag to a separate commit for just this
 hunk. You might want to consider allowing for some standard options like
 --show-reachable

yeah, it's not complete yet, that's why I hadn't intended to send this one
out just yet :)

Re: [PATCH v2] xf86: Fix non-PCI configuration-less setups

[PATCH xf86-video-modesetting 1/2] build: Put m4 macros in m4 subdirectory

[PATCH xf86-video-modesetting 2/2] Remove call to miInitializeBackingStore()

[PATCH 0/7] Various DIX patches

[PATCH 1/7] os: Remove any old logfile before trying to write to it

[PATCH 2/7] Allow DDX to provide a main()

[PATCH 3/7] glx: Synchronize Xserver glx/rensize.c with mesa src/glx/compsize.c

[PATCH 4/7] Move pseudoramiX code where it can be shared between Xwin and Xquartz

[PATCH 5/7] Build dpmsstubs.c once as a convenience library, rather than once for each DDX which wants to use it

[PATCH 6/7] Build Xi/stubs.c once as a convenience library, rather than once for each DDX which wants to use it

[PATCH 7/7] Add a screen PaintWindow wrapper

[PATCH 1/4] xfree86: Remove some unused variables

[PATCH 2/4] xfree86: Override physical output size only if EDID is available

[PATCH 4/4] HACK: xfree86: modes: Recompute DPI when the screen size changes

[PATCH 3/4] xfree86: Use physical dimensions provided by output

Re: [PATCH 1/7] os: Remove any old logfile before trying to write to it

Re: [PATCH 1/7] os: Remove any old logfile before trying to write to it

Possible regression with GLX+Xvfb after 5f5bbbe?

Re: [PATCH 1/4] xfree86: Remove some unused variables

Re: [PATCH xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

Re: [PATCH 0/3] Fix pointer emulation on non-ownership windows

Re: [PATCH xorg-gtest] Expand on the default 'failed to open connection to display' error

Re: [PATCH libXi] man: add generation of missing man pages for XIGrabTouchBegin

Re: [PATCH xorg-gtest 2/2] xserver: install default X error handler

Re: [PATCH xorg-gtest 1/2] gtest: add gtest-spi.h header

Re: [PATCH v2] xf86: Fix non-PCI configuration-less setups

Re: [PATCH xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

Re: [PATCH xorg-gtest 01/11] Drop .gz tarballs, bz2 is enough

Re: [PATCH xorg-gtest 2/2] xserver: install default X error handler

29 matches

Site Navigation

Mail list logo

Footer information