This is useful for mock input drivers that control the server in
integration tests. Given that input submission happens on a different
thread than processing, it's otherwise impossible for the driver to
synchronize with the completion of the processing of submitted events.
Signed-off-by: Povilas Kanapickas <povilas@radix.lt>
This allows us to remove darwinEvents_lock() and darwinEvents_unlock()
and remove the serverRunning hack from dix
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
This call wasn't converted to 'input_unlock()' when the SIGIO code was
removed from the server, and so when the queue growing was reworked to
be done from the input thread, it got left sitting here. As the caller
now manages the lock, we don't need to switch this to input_unlock at
this point.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Now that events are read at normal process time, we can use malloc to
grow the event queue instead of discarding events.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
The current SIGIO signal handler method, used at generation of input events,
has a bunch of oddities. This patch introduces an alternative way using a
thread, which is used to select() all input device file descriptors.
A mutex was used to control the access to input structures by the main and input
threads. Two pipes to emit alert events (such hotplug ones) and guarantee the
proper communication between them was also used.
Co-authored-by: Fernando Carrijo <fcarrijo@freedesktop.org>
Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
v2: Fix non-Xorg link. Enable where supported by default.
This also splits out the actual enabling of input threads to
DDX-specific patches which follow
v3: Make the input lock recursive
v4: Use regular RECURSIVE_MUTEXes instead of rolling our own
Respect the --disable-input-thread configuration option by
providing stubs that expose the same API/ABI.
Respond to style comments from Peter Hutterer.
v5: use __func__ in inputthread debug and error mesages.
Respond to style comments from Peter Hutterer.
v6: use AX_PTHREAD instead of inlining pthread tests.
Suggested by Emil Velikov <emil.l.velikov@gmail.com>
v7: Use pthread_sigmask instead of sigprocmask when using threads
Suggested by Adam Jackson <ajax@redhat.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
This removes all of the SIGIO handling support used for input
throughout the X server, preparing the way for using threads for input
handling instead.
Places calling OsBlockSIGIO and OsReleaseSIGIO are marked with calls
to stub functions input_lock/input_unlock so that we don't lose this
information.
xfree86 SIGIO support is reworked to use internal versions of
OsBlockSIGIO and OsReleaseSIGIO.
v2: Don't change locking order (Peter Hutterer)
v3: Comment weird && FALSE in xf86Helper.c
Leave errno save/restore in xf86ReadInput
Squash with stub adding patch (Peter Hutterer)
v4: Leave UseSIGIO config parameter so that
existing config files don't break (Peter Hutterer)
v5: Split a couple of independent patch bits out
of kinput.c (Peter Hutterer)
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
v2:
Uses BUG_WARN_MSG to also provide a stack trace. (Peter Hutterer)
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Andy Ritger <aritger@nvidia.com>
mieq.c:520:9: error: void function 'mieqProcessDeviceEvent' should not return a value [-Wreturn-type,Semantic Issue]
return 0;
^ ~
1 error generated.
Regression-from: 9fb08310b5
Found-by: Tinderbox
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Once a device is disabled, it doesn't have a sprite pointer anymore. If an
event is still in the queue and processed after DisableDevice finished, a
dereference causes a crash. Example backtrace (crash forced by injecting an
event at the right time):
(EE) 0: /opt/xorg/bin/Xorg (OsSigHandler+0x3c) [0x48d334]
(EE) 1: /lib64/libpthread.so.0 (__restore_rt+0x0) [0x37fcc0f74f]
(EE) 2: /opt/xorg/bin/Xorg (mieqMoveToNewScreen+0x38) [0x609240]
(EE) 3: /opt/xorg/bin/Xorg (mieqProcessDeviceEvent+0xd4) [0x609389]
(EE) 4: /opt/xorg/bin/Xorg (mieqProcessInputEvents+0x206) [0x609720]
(EE) 5: /opt/xorg/bin/Xorg (ProcessInputEvents+0xd) [0x4aeb58]
(EE) 6: /opt/xorg/bin/Xorg (xf86VTSwitch+0x1a6) [0x4af457]
(EE) 7: /opt/xorg/bin/Xorg (xf86Wakeup+0x2bf) [0x4af0a7]
(EE) 8: /opt/xorg/bin/Xorg (WakeupHandler+0x83) [0x4445cb]
(EE) 9: /opt/xorg/bin/Xorg (WaitForSomething+0x3fe) [0x491bf6]
(EE) 10: /opt/xorg/bin/Xorg (Dispatch+0x97) [0x435748]
(EE) 11: /opt/xorg/bin/Xorg (dix_main+0x61d) [0x4438a9]
(EE) 12: /opt/xorg/bin/Xorg (main+0x28) [0x49ba28]
(EE) 13: /lib64/libc.so.6 (__libc_start_main+0xf5) [0x37fc821d65]
(EE) 14: /opt/xorg/bin/Xorg (_start+0x29) [0x425e69]
(EE) 15: ? (?+0x29) [0x29]
xf86VTSwitch() calls ProcessInputEvents() before disabling a device, and
DisableDevice() calls mieqProcessInputEvents() again when flushing touches and
button events. Between that and disabling the device (which causes new events
to be refused) there is a window where events may be triggered and enqueued.
On the next call to PIE that event is processed on a now defunct device,
causing the crash.
The simplest fix to this is to discard events from disabled devices. We flush
the queue often enough before disabling that when we get here, we really don't
care about the events from this device.
X.Org Bug 77884 <http://bugs.freedesktop.org/show_bug.cgi?id=77884>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reported-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Tested-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
mieq.c:290:26: warning: format '%u' expects argument of type 'unsigned int',
but argument 2 has type 'size_t' [-Wformat=]
pnprintf supports size_t since 5ea21560dd
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
A coupel of unused variables, and some debug code with mis-matching
printf format and variable types.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Based on some bugzilla scraping I did around November 2012. Of xserver
bugs in Red Hat bugzilla with an EQ size message in the log, the
distribution looked like:
String | Matches
-------------------------------------
Increasing EQ size to 512 | 460
Increasing EQ size to 1024 | 52
Increasing EQ size to 2048 | 6
Increasing EQ size to 4096 | 0
Most of the "512" ones appear to be mostly harmless, some relatively
expensive path in either rendering or resource destruction simply taking
too long due to external pressures like paging or CPU contention. So
let's raise the initial queue size, both to reduce the number of
spurious abrt reports and to drop fewer events in all but the most
pathological cases.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Different miPointerSpriteFuncRec implementations do a varying
business at ultimately calling miPointerUpdateSprite(), this
particularly fails when using the plain mi sprite on touch events,
where the sprite is just moved/updated on cursor changes.
So, ensure miPointerUpdateSprite() is called generically for
pointer emulating touch events as with regular motion events.
Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This adds support for clients that would like to get a notification
every time a barrier is hit, and allows clients to temporarily release
a barrier so that pointers can go through them, without having to
destroy and recreate barriers.
Based on work by Chris Halse Rogers <chris.halse.rogers@canonical.com>
Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This is strictly the application of the script 'x-indent-all.sh'
from util/modular. Compared to the patch that Daniel posted in
January, I've added a few indent flags:
-bap
-psl
-T PrivatePtr
-T pmWait
-T _XFUNCPROTOBEGIN
-T _XFUNCPROTOEND
-T _X_EXPORT
The typedefs were needed to make the output of sdksyms.sh match the
previous output, otherwise, the code is formatted badly enough that
sdksyms.sh generates incorrect output.
The generated code was compared with the previous version and found to
be essentially identical -- "assert" line numbers and BUILD_TIME were
the only differences found.
The comparison was done with this script:
dir1=$1
dir2=$2
for dir in $dir1 $dir2; do
(cd $dir && find . -name '*.o' | while read file; do
dir=`dirname $file`
base=`basename $file .o`
dump=$dir/$base.dump
objdump -d $file > $dump
done)
done
find $dir1 -name '*.dump' | while read dump; do
otherdump=`echo $dump | sed "s;$dir1;$dir2;"`
diff -u $dump $otherdump
done
Signed-off-by: Keith Packard <keithp@keithp.com>
Acked-by: Daniel Stone <daniel@fooishbar.org>
Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
This patch applies most of the protocol conversions and the internal event
type for ownership events.
Note that ownership events are generated by the DIX only, they do not pass
through the event queue.
Co-authored-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
No-one can generated them yet, but if they could, we'd be processing them
like there was no tomorrow.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
The are the same as device events internally but require the touch ID
separately from the detail.button field (the protocol uses the detail field
for the touch id).
For simpler integration of pointer emulation we need to set the
detail.button field while keeping the touchid around.
Add the three new touch event types to the various places in the server
where they need to be handled. The actual handling of the events is somewhat
more complicated in most places.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
The last 64 events in the event queue will be reserved for release
events in order to help return the system to a cleaner state when
it comes back from a soft wedge.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
This patch changes from a static length event queue (512) to one that
starts at 128 and grows to 4096 as it overflows, logging each time it
grows.
This change also allows for multiple backtraces to be printed when the
server is wedged rather than just one. This increased sampling should
help identify the true hog in cases where one backtrace might be
insufficient.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
fromDIX is neither exactly true nor particularly helpful in understanding
what this parameter triggers. Rename to set_dequeue_screen, because that's
exactly what happens.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Fixes regression introduced by 5690199802
mieq.c:159:5: error: implicit declaration of function 'verify_internal_event' is invalid in C99 [-Wimplicit-function-declaration,Semantic Issue]
verify_internal_event(e);
^
1 error generated.
Also includes some other warning cleanups in events.c we're there.
events.c:2198:24: warning: equality comparison with extraneous parentheses [-Wparentheses,Semantic Issue]
else if ((type == MotionNotify))
~~~~~^~~~~~~~~~~~~~~
events.c:2198:24: note: remove extraneous parentheses around the comparison to silence this warning [Semantic Issue]
else if ((type == MotionNotify))
~ ^ ~
events.c:2198:24: note: use '=' to turn this equality comparison into an assignment [Semantic Issue]
else if ((type == MotionNotify))
^~
=
events.c:2487:5: error: implicit declaration of function 'verify_internal_event' is invalid in C99 [-Wimplicit-function-declaration,Semantic Issue]
verify_internal_event(event);
^
events.c:5909:22: warning: declaration shadows a local variable [-Wshadow,Semantic Issue]
DeviceIntPtr it = inputInfo.devices;
^
events.c:5893:18: note: previous declaration is here
DeviceIntPtr it = inputInfo.devices;
^
3 warnings and 1 error generated.
events.c:2836:27: warning: incompatible pointer types passing 'DeviceEvent *' (aka 'struct _DeviceEvent *') to parameter of type
'const InternalEvent *' (aka 'const union _InternalEvent *')
verify_internal_event(ev);
^~
../include/inpututils.h:40:56: note: passing argument to parameter 'ev' here
extern void verify_internal_event(const InternalEvent *ev);
^
1 warning generated.
Found-by: yuffie tinderbox (-Werror=implicit)
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
EventListPtr is a relic from pre-1.6, when we had protocol events in the
event queue and thus events of varying size.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
The macro is sufficient if called during a development cycle, but not
sufficient information when triggered by a user (e.g.
https://bugzilla.redhat.com/show_bug.cgi?id=688693).
Expand what this does to print the event content and a backtrace, so at
least we know where we're coming from. Only the first 32 bytes are printed
since if something goes wrong, the event we have is almost certainly an
xEvent or xError, both restricted to 32 bytes.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
The removal of the double-use will cause some suble bugs as some conditions
to check for the dev->u.master case were broken and also evaluated as true
if lastSlave was set (instead of master).
Also breaks the input ABI.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
This is not a straightforward search/replacement due to a long-standing
issue.
dev->u.master is the same field as dev->u.lastSlave. Thus, if dev is a master
device, a check for dev->u.master may give us false positives and false
negatives.
The switch to IsFloating() spells out these cases and modifies the
conditions accordingly to cover both cases.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
These two were sideeffects of lastSlave being in the same field as the
master. For devices generated by the master device directly, lastSlave was 0
and the device would (with the old checks) be interpreted as floating.
Add the required checks to safeguard against master devices.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
The subtype in the DGA event is the core type and all ET_ event types (where
applicable) are identical to the core types. Thus the switch statement below
will work as required and assign the right master device.
Fixes a crasher bug on keyboard devices with valuators. If a device sends a
motion event while grabbed and a DGA client is active (but has not selected
input through DGA), the valuator event is posted through the VCK and
eventually results in a NULL-pointer dereference on dev->valuator.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
event_size is never assigned to in mieqProcessInputEvents(), so realloc()
is always called. This is benign, but I'm guessing not intended.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
The only remaining X-functions used in server are XNF*, the rest is converted to
plain alloc/calloc/realloc/free/strdup.
X* functions are still exported from server and x* macros are still defined in
header file, so both ABI and API are not affected by this change.
Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Now that all event queues hold internal events only, they never need
to be resized. Resizing them led to memory corruption as they would
get sized for an appropriate xEvent, not an internal event.
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The server was processing ET_RawMotion type when the cursor was wrapping to
another screen and getting wrong valuator values. This fix such issue
considering only ET_Motion, ET_KeyPress, ET_KeyRelease, ET_ButtonPress and
ET_ButtonRelease types when the cursor detects a new screen, keeping the
"normal" processing of device events.
Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This is long overdue. The device events are InternalEvents now (and only one
at a time), diminishing the need for an EventList for the master event.
Furthermore, don't make masterEvent a static since this will interfere if
mieqProcessDeviceEvent is called from somewhere else (e.g. XKB actions).
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
InternalEvents shouldn't be used anywhere outside the X server itself. Split
up into events.h for opaque typedefs for the events needed by various
headers and eventstr.h for the actual struct definitions.
eventstr.h must only be included by code that requires internal events and
is not part of the SDK.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Povilas Kanapickas
Adam Jackson
Laurent Carlier
Jeremy Huddleston Sequoia
Keith Packard
Andy Ritger
Julien Cristau
Peter Hutterer
Carlos Garnacho
Jasper St. Pierre
Chase Douglas
Jon TURNEY
Mikhail Gusarov
Jeremy Huddleston
Tiago Vignatti