gmane.comp.freedesktop.xorg.devel

[PATCH:libX11] Make sure internal headers include required headers

Alan Coopersmith — 2013-05-17T22:28:04

From: Niveditha Rau <Niveditha.Rau-veTT2BtV2gBXrIkS9f7CXA< at >public.gmane.org>

Fixes builds with Solaris Studio 12.3 when lint is enabled, since it no
longer ignores *.h files, but complains when they reference undefined
typedefs or macros.

Signed-off-by: Niveditha Rau <Niveditha.Rau-veTT2BtV2gBXrIkS9f7CXA< at >public.gmane.org>
Signed-off-by: Alan Coopersmith <alan.coopersmith-QHcLZuEGTsvQT0dZR+AlfA< at >public.gmane.org>
---
 src/Cmap.h         |    2 ++
 src/Cr.h           |    2 ++
 src/Key.h          |    3 +++
 src/Xintatom.h     |    1 +
 src/Xintconn.h     |    1 +
 src/Xprivate.h     |    2 ++
 src/Xresinternal.h |    2 ++
 src/locking.h      |    2 ++
 8 files changed, 15 insertions(+)

diff --git a/src/Cmap.h b/src/Cmap.h
index 062b538..78cc3ea 100644
--- a/src/Cmap.h
+++ b/src/Cmap.h
< at >< at > -2,6 +2,8 < at >< at >
 #ifndef _CMAP_H_
 #define _CMAP_H_
 
+#include <X11/Xlib.h>
+
 extern void
 _XcmsDeleteCmapRec(
     Display *dpy,
diff --git a/src/Cr.h b/src/Cr.h
index 800c9ab..635e9e4 100644
--- a/src/Cr.h
+++ b/src/Cr.h
< at >< at >

X11 status Capture for an arbitary application

Dibyayan Chakraborty — 2013-05-17T14:31:30

Hi,
I am currently challenged with the following problem.Any help would be
appreciated.

For some x11 application at some particular moment i would like to access
all its component's status ( window positions , child windows ,menu lists ,
event queue status and everything else) .

What are the functions that can be used for this purpose ?

[PATCH libXi] Copy the sequence number into the target event too(#64687)

Peter Hutterer — 2013-05-16T23:09:58

X.Org Bug 64687 <http://bugs.freedesktop.org/show_bug.cgi?id=64687>

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
 src/XExtInt.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/src/XExtInt.c b/src/XExtInt.c
index 507573b..2139c6a 100644
--- a/src/XExtInt.c
+++ b/src/XExtInt.c
< at >< at > -915,6 +915,7 < at >< at > static void xge_copy_to_cookie(xGenericEvent* ev,
     cookie->type = ev->type;
     cookie->evtype = ev->evtype;
     cookie->extension = ev->extension;
+    cookie->serial = ev->sequenceNumber;
 }
 
 static Bool
< at >< at > -1521,6 +1522,7 < at >< at > wireToDeviceEvent(xXIDeviceEvent *in, XGenericEventCookie* cookie)
     out = next_block(&ptr_lib, sizeof(XIDeviceEvent));
     out->display = cookie->display;
     out->type = in->type;
+    out->serial = in->sequenceNumber;
     out->extension = in->extension;
     out->evtype = in->evtype;
     out->send_event = ((in->type & 0x80) != 0);
< at >< at > -1793,6 +1795,7 < at >< at > wireToDeviceChangedEvent(xXIDeviceChangedEvent *in, XGenericEventC

[PATCH 1/2] ephyr: fix typo in man page

Peter Hutterer — 2013-05-14T23:17:11

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
 hw/kdrive/ephyr/man/Xephyr.man | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/kdrive/ephyr/man/Xephyr.man b/hw/kdrive/ephyr/man/Xephyr.man
index ca88ab2..792511b 100644
--- a/hw/kdrive/ephyr/man/Xephyr.man
+++ b/hw/kdrive/ephyr/man/Xephyr.man
< at >< at > -46,7 +46,7 < at >< at > and the following additional options:
 sets the screen size.
 .TP 8
 .BI -parent " id"
-uses exiting window
+uses existing window
 .I id .
 If a
 .BI -screen

[PULL 1.14] ephyr crash, pnprintf fixes

Peter Hutterer — 2013-05-14T22:41:46

The ephyr fix is something useful since I just ran into it myself yesterday.
pnprintf fixes are largely to prevent weird messages on FatalError.


PS: I accidentally pushed this to origin first but reset the repo a minute
later. just in case you saw that float past on the commit list.

The following changes since commit a11cf8dd68275ba4676888e5957bff13efbfbb3d:

  dix: copy event in TouchConvertToPointerEvent correctly (2013-05-06 10:20:18 +1000)

are available in the git repository at:

  git://people.freedesktop.org/~whot/xserver server-1.14-branch

for you to fetch changes up to e90beb18000cf49b9108c4f977abfd111ed908ad:

  os: Use ErrorFSigSafe from FatalError and it's friends (2013-05-14 16:37:32 +1000)

----------------------------------------------------------------
Daniel Martin (1):
      ephyr: Fix crash on 24bpp host framebuffer

Peter Hutterer (8):
      dix: send the current axis value in DeviceChangedEvents (#62321)
      If neither HAL nor udev backends are enabled, warn the user
      os: docu

[PATCH] Shadow: Switch the Amiga/Atari bitplane code to the canonicalX.Org license

Geert Uytterhoeven — 2013-05-14T11:15:23

Also add my name to the list of authors in COPYING

Suggested-by: Alan Coopersmith <alan.coopersmith-QHcLZuEGTsvQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Geert Uytterhoeven <geert-Td1EMuHUCqxL1ZNQvxDV9g< at >public.gmane.org>
Cc: Keith Packard <keithp-aN4HjG94KOLQT0dZR+AlfA< at >public.gmane.org>
---
Alternative pull request:

The following changes since commit 2f1aedcaed8fd99b823d451bf1fb02330c078f67:
  Peter Hutterer (1):
        input: print warnings if drivers don't initialize properly

are available in the git repository at:

  https://github.com/geertu/xorg-xserver.git master

Geert Uytterhoeven (1):
      Shadow: Switch the Amiga/Atari bitplane code to the canonical X.Org license

 COPYING                   |    1 +
 miext/shadow/c2p_core.h   |   37 +++++++++++++++++++------------------
 miext/shadow/shafb4.c     |   36 ++++++++++++++++++------------------
 miext/shadow/shafb8.c     |   36 ++++++++++++++++++------------------
 miext/shadow/shiplan2p4.c |   36 ++++++++++++++++++------------------
 miext/shado

[PATCH] dix: devices must have valuators before touch is initialized

Peter Hutterer — 2013-05-13T22:31:19

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
This used to return False before 2f1aedcaed8fd99b823d451bf1fb02330c078f67
but I misssed changing that condition. So now it breaks make check
because InitTouchClassDeviceStruct requires the valuators to be set up.

 dix/devices.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/dix/devices.c b/dix/devices.c
index c514d77..c514cfe 100644
--- a/dix/devices.c
+++ b/dix/devices.c
< at >< at > -1647,6 +1647,7 < at >< at > InitTouchClassDeviceStruct(DeviceIntPtr device, unsigned int max_touches,
 
     BUG_RETURN_VAL(device == NULL, FALSE);
     BUG_RETURN_VAL(device->touch != NULL, FALSE);
+    BUG_RETURN_VAL(device->valuator == NULL, FALSE);
 
     /* Check the mode is valid, and at least X and Y axes. */
     BUG_RETURN_VAL(mode != XIDirectTouch && mode != XIDependentTouch, FALSE);

EXA composite ops don't report damage

Daniel Drake — 2013-05-11T17:19:16

Hi,

I am investigating a bug in AbiWord where when the screen is rotated,
newly inputted text does not appear in the document. This is affecting
OLPC XO.

The graphics driver
(http://dev.laptop.org/git/projects/xf86-video-dove/) implements
screen rotation via EXA, i.e. it supports composite operations with a
rotation transform applied.

I audited the EXA composite requests from the driver level, dumping
images for the source and dest pixmaps before executing the composite
op, and also the dest pixmap after compositing. My conclusion is that
the driver is doing what it is told.

I can see that the system maintains two big pixmaps - one of the
unrotated screen, and one of the rotated screen (which is shown on the
display). When the unrotated screen is changed, the changed areas are
composited onto the rotated screen. However, in the case of new text
input, the new characters are composited onto the unrotated screen,
but no corresponding composite operation happens in order to update
the rotated screen pixmap.

[PULL] plenty of touch fixes, mainly for #56578

Peter Hutterer — 2013-05-10T05:30:35

I'm not going to even pretend this is understandable, given that the first
patches to fix this issue go back to November last year. And touch grab
handling and pointer emulation is such a beast that I can't get it all in my
head anymore. I've got a bunch of test cases that show the various bugs that
have been fixed in this series and we've had some reasonable testing on this
branch. 

The biggest issue fixed here (aside from Ubuntu having a stuck grab
approximately as soon as you touch the dash) is a fix for
xorg-server-1.13.99.901-2-g9ad0fdb. That sticks the grab pointer into the
touch listener, but that grab may get free'd at any time (ProcUngrab*),
leading to dangling pointers and crashes whenever the touch continues.

One remaining issue: the Nexus 7 still seems to trigger some bug, but I've
been unable to reproduce this and in the meantime having these see wider
testing will be useful.


The following changes since commit 2f1aedcaed8fd99b823d451bf1fb02330c078f67:

  input: print warnings if drivers don'

[PULL] memleaks, BigRequest fix, more warnings

Peter Hutterer — 2013-05-10T01:53:59

The following changes since commit 6993f8b459e25b1a7a0a03e209688a28ce6c2c56:

  Xi: free barrier code at reset time (2013-05-07 09:41:19 +1000)

are available in the git repository at:

  git://people.freedesktop.org/~whot/xserver for-keith

for you to fetch changes up to 2f1aedcaed8fd99b823d451bf1fb02330c078f67:

  input: print warnings if drivers don't initialize properly (2013-05-10 11:05:00 +1000)

----------------------------------------------------------------
Peter Hutterer (4):
      dix: delete all callbacks before reset
      dix: reset the OsBuffers after killing all clients
      dix: don't overwrite proximity/focus classes
      input: print warnings if drivers don't initialize properly

Robert Morell (1):
      os: Reset input buffer's 'ignoreBytes' field

 Xi/exevents.c      | 16 ++++++++++------
 dix/devices.c      | 41 ++++++++++++++++++++++++++++++++---------
 dix/dispatch.c     |  1 +
 dix/dixutils.c     |  8 +++++++-
 dix/main.c         |  2 ++
 include/callback.h |  1 +
 os/io.c

[PATCH] dix: reset the OsBuffers after killing all clients

Peter Hutterer — 2013-05-09T23:44:29

==21860== 24 bytes in 1 blocks are still reachable in loss record 85 of 397
==21860==    at 0x4C2B3F8: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==21860==    by 0x61ED93: AllocateOutputBuffer (io.c:1037)
==21860==    by 0x61E15A: WriteToClient (io.c:764)
==21860==    by 0x457B30: ProcQueryExtension (extension.c:275)
==21860==    by 0x43596B: Dispatch (dispatch.c:432)
==21860==    by 0x425DAB: main (main.c:295)

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
makes the valgrind output a lot less noisy.

 dix/dispatch.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/dix/dispatch.c b/dix/dispatch.c
index 20f2414..398c0bf 100644
--- a/dix/dispatch.c
+++ b/dix/dispatch.c
< at >< at > -465,6 +465,7 < at >< at > Dispatch(void)
     free(clientReady);
     dispatchException &= ~DE_RESET;
     SmartScheduleLatencyLimited = 0;
+    ResetOsBuffers();
 }
 
 static int VendorRelease = VENDOR_RELEASE;

[PATCH] os: Reset input buffer's 'ignoreBytes' field

Robert Morell — 2013-05-09T20:09:02

If a client sends a request larger than maxBigRequestSize, the server is
supposed to ignore it.

Before commit cf88363d, the server would simply disconnect the client.  After
that commit, it attempts to gracefully ignore the request by remembering how
long the client specified the request to be, and ignoring that many bytes.
However, if a client sends a BigReq header with a large size and disconnects
before actually sending the rest of the specified request, the server will
reuse the ConnectionInput buffer without resetting the ignoreBytes field.  This
makes the server ignore new X clients' requests.

This fixes that behavior by resetting the ignoreBytes field when putting the
ConnectionInput buffer back on the FreeInputs list.

Signed-off-by: Robert Morell <rmorell-DDmLM1+adcrQT0dZR+AlfA< at >public.gmane.org>
---
 os/io.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/os/io.c b/os/io.c
index 2f091c4..0d980ab 100644
--- a/os/io.c
+++ b/os/io.c
< at >< at > -1063,6 +1063,7 < at >< at > FreeOsBuffers(OsCommPtr oc)
             o

[PATCH] Filter out -l parameters when setting dependencies

Gustavo Noronha Silva — 2013-05-08T22:57:42

Newer make (Fedora 19) gets confused when it finds a -l parameter in a
dependency, and tries to make it as a target, causing the build to fail.

Signed-off-by: Gustavo Noronha Silva <gustavo.noronha-ZGY8ohtN/8qB+jHODAdFcQ< at >public.gmane.org>
---
 configure.ac                        |   2 +-
 fb/fb.h                             |   3 +
 fb/fbpict.c                         | 149 +++++++++++++++++++++++++++++++++++-
 fb/fbscreen.c                       |   1 +
 hw/dmx/Makefile.am                  |   2 +-
 hw/vfb/Makefile.am                  |   2 +-
 hw/xfree86/Makefile.am              |   2 +-
 hw/xnest/Makefile.am                |   2 +-
 hw/xquartz/mach-startup/Makefile.am |  11 +--
 hw/xwin/Makefile.am                 |   2 +-
 test/Makefile.am                    |   2 +-
 11 files changed, 165 insertions(+), 13 deletions(-)

diff --git a/configure.ac b/configure.ac
index 5f8dd56..89a7a9d 100644
--- a/configure.ac
+++ b/configure.ac
< at >< at > -818,7 +818,7 < at >< at > LIBPCIACCESS="pciaccess >= 0.12.901"
 LIBUDEV="libudev >

[PATCH 1/2] dix: don't overwrite proximity/focus classes

Peter Hutterer — 2013-05-09T04:30:49

InitPointerClassDeviceStruct/InitKeyboardDeviceStruct allocate a
proximity/focus class, respectively. If a driver calls
InitFocusClassDeviceStruct or InitProximityClassDeviceStruct beforehand,
the previously allocated class is overwritten, leaking the memory.

Neither takes a parameter other than the device, so we can simply skip
initialising it if we already have one.

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
 dix/devices.c | 2 +-
 xkb/xkbInit.c | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/dix/devices.c b/dix/devices.c
index 138ed25..3188fb7 100644
--- a/dix/devices.c
+++ b/dix/devices.c
< at >< at > -1366,7 +1366,7 < at >< at > InitValuatorClassDeviceStruct(DeviceIntPtr dev, int numAxes, Atom *labels,
     valc->numMotionEvents = numMotionEvents;
     valc->motionHintWindow = NullWindow;
 
-    if (mode & OutOfProximity)
+    if ((mode & OutOfProximity) && !dev->proximity)
         InitProximityClassDeviceStruct(dev);
 
     dev->valuator = valc;
dif

[PATCH synaptics 1/2] Plug memory leak for option_string

Peter Hutterer — 2013-05-09T01:14:06

xf86SetStrOption() returns a strdup'd string that must be freed after use.

Signed-off-by: Peter Hutterer <peter.hutterer-Pf4JEFdB4epeoWH0uzbU5w< at >public.gmane.org>
---
 src/synaptics.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/synaptics.c b/src/synaptics.c
index 2de05d4..d4b21a9 100644
--- a/src/synaptics.c
+++ b/src/synaptics.c
< at >< at > -513,12 +513,15 < at >< at > set_softbutton_areas_option(InputInfoPtr pInfo)
     memcpy(pars->softbutton_areas[0], values, 4 * sizeof(int));
     memcpy(pars->softbutton_areas[1], values + 4, 4 * sizeof(int));
 
+    free(option_string);
+
     return;
 
  fail:
     xf86IDrvMsg(pInfo, X_ERROR,
                 "invalid SoftButtonAreas value '%s', keeping defaults\n",
                 option_string);
+    free(option_string);
 }
 
 static void

[PULL unreviewed] two scaling patches

Peter Hutterer — 2013-05-07T23:43:29

Second one is the v2 of the one keith reviewed.

The following changes since commit 6993f8b459e25b1a7a0a03e209688a28ce6c2c56:

  Xi: free barrier code at reset time (2013-05-07 09:41:19 +1000)

are available in the git repository at:

  git://people.freedesktop.org/~whot/xserver unreviewed

for you to fetch changes up to 6e4459dc5cb691c1f378ca2efeeee74238549d5f:

  dix: fix device scaling to use a [min,max[ range. (2013-05-08 09:26:55 +1000)

----------------------------------------------------------------
Peter Hutterer (2):
      dix: pre-scale x by the screen:device:resolution ratio
      dix: fix device scaling to use a [min,max[ range.

 dix/devices.c   |  4 ++--
 dix/getevents.c | 60 ++++++++++++++++++++++++++++++++++++---------------------
 2 files changed, 40 insertions(+), 24 deletions(-)
_______________________________________________
xorg-devel-go0+a7rfsptAfugRpC6u6w< at >public.gmane.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-

[PULL] misc input fixes, pnprintf improvements

Peter Hutterer — 2013-05-06T23:45:24

The following changes since commit 2746c681639f9512e6e45fb8d0151b996b6aff7f:

  Merge remote-tracking branch 'alanc/master' (2013-05-06 10:52:40 -0700)

are available in the git repository at:


  git://people.freedesktop.org/~whot/xserver for-keith

for you to fetch changes up to 6993f8b459e25b1a7a0a03e209688a28ce6c2c56:

  Xi: free barrier code at reset time (2013-05-07 09:41:19 +1000)

----------------------------------------------------------------
Peter Hutterer (12):
      Xi: fix comment - XI2 grabs aren't keysym grabs
      dix: send the current axis value in DeviceChangedEvents (#62321)
      Xi: always return BadMatch for XTest devices ChangeDeviceControl requests
      If neither HAL nor udev backends are enabled, warn the user
      xkb: free XkbRulesUsed and XkbRulesDflt on extension cleanup
      dix: reset the registry before quitting
      os: support pnprintf length modifiers for integers
      os: support %% in pnprintf
      os: support %c in pnprintf
      os: complain about unsupported p

[PATCH xaw3d 0/2] Xaw3d build system

Javier Pello — 2013-05-06T11:08:36

I am attaching a couple of patches to the Xaw3d build system.

The first patch deals with a bug when running configure. Xaw3d has
several optional features (multiplane bitmaps, gray stipples, arrow
scrollbars), disabled by default, that can be selected at configure
time via --enable-foo switches. However, if you pass an explicit
--disable-foo switch, configure will actually *enable* the feature,
which, in my opinion, breaks the principle of least astonishment.
The first patch corrects this.

The second patch restores recording of such compilation options in
Xaw3dP.h. In 1.5, features selected at configure time (multiplane
bitmaps, etc.) were recorded as defines in Xaw3dP.h, so packages
compiling against Xaw3d could check those defines by including
Xaw3dP.h to see what features were enabled. This has now changed,
and, as I understand it, detection of optional features should be
done by the additional -Dfoo flags reported by `pkgconfig --cflags
xaw3d`. This, however, breaks compilation of old packages against

(Crowd funded) fix for SiS 671/771 video cards

Pander — 2013-05-06T09:07:44

Hi all,

Since a few months the driver for SiS 671/771 video cards is no longer
working in main stream distributions such as Ubuntu. This is affecting
many users and I would like to ask the development community of Xorg who
would like to fix this issue.

For the details please see these bug reports:
  https://bugs.freedesktop.org/show_bug.cgi?id=15190
and:
 https://bugs.launchpad.net/debian/+source/xserver-xorg-video-sis/+bug/301958

Since the number of users waiting for a fix is in the thousands, I am sure
that some sort of crowd funding such as:
  https://www.catincan.com/proposal/altdrag/altdrag-10
might be possible.

I am more than happy to make a financial contribution in order to not
having to write off some still good hardware. I am sure many others are
willing to chip in. Who would be interested in taking this on?

Regards,

Pander

_______________________________________________
xorg-devel-go0+a7rfsptAfugRpC6u6w< at >public.gmane.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Inf

MultiThreaded Xorg Server?

小飞珑 — 2013-05-06T08:36:46

I am concentrating on performance of Xorg server for several months. I found that it seemed Dispatch() in dix/dispatch.c could be multi threaded, say, the event and request handling might be handled by several threads. Was it possible? Or it might not be as easy as it looks like? Thanks for your attention._______________________________________________
xorg-devel-go0+a7rfsptAfugRpC6u6w< at >public.gmane.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel

[PULL 1.14] input fixes for 1.14.2

Peter Hutterer — 2013-05-06T00:42:48

The following changes since commit f5796f98dadccf67c04f601178966614dd51a1b4:

  xserver 1.14.1 (2013-04-17 16:12:00 +1000)

are available in the git repository at:

  git://people.freedesktop.org/~whot/xserver server-1.14-branch

for you to fetch changes up to a11cf8dd68275ba4676888e5957bff13efbfbb3d:

  dix: copy event in TouchConvertToPointerEvent correctly (2013-05-06 10:20:18 +1000)

----------------------------------------------------------------
Keith Packard (1):
      Xi: Use correct destination when swapping barrier events

Knut Petersen (1):
      Never try to execute BellProcPtr NULL.

Maarten Lankhorst (2):
      Xi: Do not handle ET_TouchOwnership in ProcessTouchEvent
      dix: copy event in TouchConvertToPointerEvent correctly

Peter Hutterer (2):
      dix: don't set non-exisiting flags on touch events
      xfree86: change a log message

Rui Matos (1):
      Xi: fix swapped grab mode args for keyboard devices in XIGrabDevice

 Xi/devbell.c                               |  3 ++-
 Xi/exevents.