Devices that succeeded during PreInit and DEVICE_INIT but failed in
DEVICE_ON would be deleted through xf86DeleteInput but not removed from the
list of input devices (and not turned off). The result was a double free on
server shutdown.
Fix this by calling RemoveDevice if EnableDevice fails.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
The nature of hotplug is that a device we enumerated might already be
gone by the time we look at it, so don't assume otherwise.
Signed-off-by: Lennart Poettering <lennart@poettering.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Don't enumerate/monitor all devices of the system (since that can be
quite a few), but limit our search to devices from the "input"
subsystem, as well as the "tty" subsystem (to cover Wacom tablets).
This should make X start up a bit faster and reduce the number of
unnecessary wake-ups of the X server.
Signed-off-by: Lennart Poettering <lennart@poettering.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
udev gives no guarantee that before each "changed" event for a device
there's an "add" event, or that before each "remove" is an "add", or
that before each "add" there was no "add" already and so on. Users can
trigger these events at any time with "udevadm trigger", and netlink is
a lossy transport, hence the events can come in unexpected ordering.
With other words: regardless which event is generated, the X server must
not choke on it and make the best of it, hence make sure that if we get
an "add" event for an existing device we don't add the device a second
time.
Signed-off-by: Lennart Poettering <lennart@poettering.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Bugfix for broken xkbcomp: if we encounter an XFree86Private action with
Any+AnyOfOrNone(All), then we skip the interp as broken. Versions of
xkbcomp below 1.2.2 had a bug where they would interpret a symbol that
couldn't be found in an interpret as Any. So, an
XF86LogWindowTree+AnyOfOrNone(All) interp that triggered the PrWins
action would make every key without an action trigger PrWins if libX11
didn't yet know about the XF86LogWindowTree keysym. None too useful.
We only do this for XFree86 actions, as the current XKB dataset relies
on Any+AnyOfOrNone(All) -> SetMods for Ctrl in particular.
See xkbcomp commits 2a473b906943ffd807ad81960c47530ee7ae9a60 and
3caab5aa37decb7b5dc1642a0452efc3e1f5100e for more details.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The previous patch accidentally introduced a hard dependency on
Composite. Sorry, OS X.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reported-by: Jeremy Huddleston <jeremyhu@apple.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The GLXversion member of the __GLXscreen struct
is just cruft since commit ad5c0d9efa,
when we started returning the minimum GLX version supported by all
of the screens on the display, rather than the maximum version supported
by the server.
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Document the -[no]wgl options in the XWin manpage
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Report Window XIDs in Window create/destroy/reparent debug messages
It's actually quite useful if you are trying to corrolate those events
with what a client is doing...
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Add a suitable cast to the generated code for glWinSetupDispatchTable()
so it doesn't generate screeds of warnings
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Handle failure to get any useful pixel formats for GLX fbconfigs
more gracefully: If we didn't get any useful pixel formats from
wglGetPixelFormatAttribivARB(), fall back to using DescribePixelFormat().
If that doesn't give us any useful pixel formats, fallback to software
rendering.
This works around a problem with Intel 845G drivers, where
wglGetPixelFormatAttribivARB() doesn't seem to work as we expect it to...
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
... instead just log if an attempt is made to call a wrapper for
a function which didn't resolve
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Propagate and report the failure if SetPixelFormat() fails
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Request the message using languageID 0 (best effort), rather than only using language neutral messages
Always report the numeric error code.
Trim any trailing \r from FormatMessage() output
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Some graphics hardware supports hundreds of pixel formats, so adjust
formatting in fbconfig dumper for 3 digit index numbers
Also report the PFD_SUPPORT_DIRECTDRAW, PFD_DIRECT3D_ACCELERATED and
PFD_SUPPORT_COMPOSITION flags introduced with aero
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
mieqFini() already does the right thing, but it needs to be called by the
various DDXs and the XTest Extension.
X.Org Bug 38634 <http://bugs.freedesktop.org/show_bug.cgi?id=38634>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Acked-by: Jeremy Huddleston <jeremyhu@apple.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
In updateSlaveDeviceCoords, pDev->last.valuators was being copied from
the master, but pDev->last.remainder wasn't. Make sure we copy both, to
avoid minor inconsistencies.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
RawEvents are supposed to be events coming from the driver. When warping the
pointer, this should not generate a raw event.
X.Org Bug 30068 <http://bugs.freedesktop.org/show_bug.cgi?id=30068>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Previously, the input tests were working up to sizeof(mask) * 8, which
could be more than the arrays; the latter only being sized as
MAX_VALUATORS.
Hypothetically, if you were switching the stored valuator values to
double instead of uint32_t, and you attempted to set the 39th member of
a MAX_VALUATORS-sized (36) array, you'd probably end up smashing
ev->key_repeat into oblivion and then tripping the check for invalid
flags because you haven't yet put XIPointerEmulated into the valid flags
for XI_Motion. Probably.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Similar to how we link Xorg, make sure that whenever any of the
component libraries changes, we relink libxservertest and the tests.
Not much use testing anything other than the actual source in your tree.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
miPointerUpdateSprite is already called from mieqProcessInputEvents, so
calling it by hand immediately after isn't massively helpful.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Cyril Brulebois <kibi@debian.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
miPointerUpdateSprite is already called from mieqProcessInputEvents, so
calling it by hand immediately after isn't massively helpful.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Cyril Brulebois <kibi@debian.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Headers don't really need to be mode 0755.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Cyril Brulebois <kibi@debian.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Add four new private XKB actions for debugging:
* PrGrbs: print active grabs to the log file
* Ungrab: ungrab all currently active grabs
* ClsGrb: kill clients with active grabs
* PrWins: dump the current window tree to the log file
To use these, you need to modify your XKB maps, e.g. the following to
have Ctrl+Alt+(F9-F12) mapped to the above:
- compat/xfree86:
interpret XF86LogGrabInfo {
action = Private(type=0x86, data="PrGrbs");
};
interpret XF86Ungrab {
action = Private(type=0x86, data="Ungrab");
}
interpret XF86ClearGrab {
action = Private(type=0x86, data="ClsGrb");
}
interpret XF86LogWindowTree {
action = Private(type=0x86, data="PrWins");
}
- symbols/pc:
key <FK09> { type="CTRL+ALT", [ Return, XF86LogGrabInfo ] };
key <FK10> { type="CTRL+ALT", [ Return, XF86Ungrab ] };
key <FK11> { type="CTRL+ALT", [ Return, XF86ClearGrab ] };
key <FK12> { type="CTRL+ALT", [ Return, XF86LogWindowTree ] };
At the moment, this only works if the grabbing client continues to call
AllowEvents, as the server does no event processing at all when a device
is frozen.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Rewrite PrintWindowTree to make it actually tell you what you want to
know.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Report to find out all non-UTF-8 files created by
cat extensions |xargs -I XXXX find . -name \*.XXXX |while read FILE ; do
if ( iconv -f utf8 -t ucs2 $FILE >/dev/null 2>/dev/null ) ; then
/bin/true
else
echo $FILE
fi
done >>report
Signed-off-by: Matěj Cepl <mcepl@redhat.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
[Daniel: git am failed for me, so I redid it. The method listed in the
commit message also failed, so I just used file/grep/iconv. The
results are the same though.]
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
BlockHandler and WakeupHandlers may be removed within a different
BlockHandler or WakeupHandler, especially since config/udev uses
these and removes devices.
Calling the deleted handlers and passing potentially freed data
can result in the X server segfaulting after device removal, or
events that result in device removal such as undocking or suspend/
resume.
Signed-off-by: Scott James Remnant <scott@netsplit.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
ConfigNotify is set by DRI2ScreenInit, but not restored to
previous state on close. Fix it.
(I'm preparing a patch for xf86-video-nouveau which detects GPU lockup
after dri2 init and it needs to reinitialize dri2)
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Unable to find any use of the PERL Automake variable.
It was used in hw/xfree86/scanpci around 2005.
Should it ever be needed, use XORG_WITH_PERL macro.
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
Signed-off-by: Keith Packard <keithp@keithp.com>
Left-justify website link in About box. This is a cosmetic fix to make
the About box display correctly when Windows is configured with a
non-default DPI value
Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Tested-by: Colin Harrison <colin.harrison@virgin.net>
When the style changes, adjust the window size so the client area remains the same.
Otherwise the window size may change when sizing is reflected from Windows to X, and
some windows are drawn expecting them to be exactly the requested size (e.g. the
gmplayer control window)
Use DeferWindowPos to delay the resize to preserve client area on WM_STYLECHANGING
until after the style change has actually happened in WM_STYLECHANGED
As a consquence of this, we need to be more careful to create windows with exactly
the requested placement and client area initially, so the client area matches what
the X client requested
Also synchronize the X windows idea of the placement of a window which Windows is
allowed to place
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Tested-by: Colin Harrison <colin.harrison@virgin.net>
Move GLX extension string utility functions from the DRI1 convenience library
into the GLX extension convenience library, so other DDX which don't have DRI
can use them.
This is probably also needed if anyone actually tries to build an Xorg DDX
with only DRI2 support...
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Use the new event types so we can pass a valid SBC value to clients.
Fix up the completion calls to use CARD32 instead of CARD64 to match
the new field size.
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
There are, in general, more fbconfig depths than visual depths.
fbconfigs need not support Window rendering, however any that do must
have an associated visual ID (which we got right), and any that do not
must not claim GLX_WINDOW_BIT in GLX_DRAWABLE_TYPE (which we got
wrong).
Fixes piglit/glx-fbconfig-sanity, assuming you have a sufficiently
recent piglit, as that test formerly wrongly required pixmap-capable
fbconfigs to have a visual.
v2: Additional check for fbconfigs that didn't have GLX_WINDOW_BIT in
the first place, from previous patch by Jon TURNEY; also, also clear
->visualID.
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Guard the use of backtrace() with HAVE_BACKTRACE, since we already
have the autoconf machinery for setting that.
For the moment, assume dladdr() is available when backtrace() is
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
automake generates _DEPENDENCIES from _LIBADD, but it strips out variables.
This resulted in not relinking if some components were rebuilt (like
libdix, libos, etc).
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
In addition, this change will not call into the X11 activation unless an X11
window was active when we deactivated. We can't rely on the event and current
key windows because the key window will be nil until activated, and the event
will only reference the window if the window was clicked (whereas it will be
nil if we activated via dock or cmd-tab).
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
sdksyms moved from hw/xfree86/loader to hw/xfree86, so the
configure-time create of sdksyms.dep needs to reflect that
change. Otherwise, make might be confused by a missing file and (more
importantly to me) hw/xfree86/loader/sdksyms.dep will be left around
after 'make clean' causing 'make distcheck' to fail.
Introduced in e3f296d91d, when the ifdef DEBUG
around the whole block was removed, but only two of the three ErrorF
switched to DebugF.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
