If we end up reading all pending X events in the course of other server
execution, then our notify FD callback won't get invoked and we won't
process them. Fix this by noting that there are queued events in the
block handler, setting the poll timeout to zero and queuing a work
proc to clear the event queue.
v2: use a work proc to clear the event queue rather than doing it in
the block handler directly.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Under glamor, we need to re-create the screen pixmap at the new size
so that we can ask glamor for the associated texture. Fortunately, we
can simply use ephyr_glamor_create_screen_resources to create the new
pixmap.
Because this is being done after the server has started, we need to
walk the window heirarchy and reset any windows pointing at the old
pixmap. I could easily be convinced that this TraverseTree should be
moved to miSetScreenPixmap.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
The screen block and wakeup handlers are the only ones which provide a
well known ordering between the wrapping layers; placing these as
close as possible to the server blocking provides a way for the driver
to control the flow of execution correctly.
Switch the shadow code to run in the screen block handler so that it
now occurrs just before the server goes to sleep.
Switch glamor to call down to the driver after it has executed its own
block handler piece, in case the driver needs to perform additional
flushing work after glamor has called glFlush.
These changes ensure that the following modules update the screen in
the correct order:
animated cursors (uses RegisterBlockAndWakeupHandlers dynamically)
composite (dynamic wrapping)
misprite (dynamic wrapping)
shadow (static wrapping)
glamor (static wrapping)
driver (static wrapping)
It looks like there's still a bit of confusion between composite and
misprite; if composite updates after misprite, then it's possible
you'd exit the block handler chain with the cursor left hidden. To fix
that, misprite should be wrapping during ScreenInit time and not
unwrapping. And composite might as well join in that fun, just to make
things consistent.
[v2] Unwrap BlockHandler in shadowCloseScreen (ajax)
[v3] ephyr: Use screen block handler for flushing changes
ephyr needs to make sure it calls glXSwapBuffers after glamor finishes
its rendering. As the screen block handler is now called last, we have
to use that instead of a registered block/wakeup handler to make sure
the GL rendering is done before we copy it to the front buffer.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
When there aren't any devices, the input thread is going to be pretty
lonely, so don't bother to even start it.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
As the man page for the latter states:
The effects of signal() in a multithreaded process are unspecified.
We already have an interface to call sigaction() instead, use it.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
We won't need these locks with the new threaded input code as it holds
the input lock across all of the input device I/O operations.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
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>
kdrive uses the NotifyFd interface, which handles all of the necessary
fd configuration in the OS layer. Having it also use the old
EnableDevice interfaces is incorrect.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
gcc6 says:
keyboard.c:46:21: warning: ‘linux_to_x’ defined but not used
Only referenced by a bunch of long if-0'd code, so chuck it all out.
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Currently Xephyr doesn't inherit host X server's keymap, which
may lead to keymap mismatches when using a non-US keyboard in a
window inside Xephyr. This patch makes Xephyr change its keymap
to match host X server's one (unless XKB support is disabled),
using xcb-xkb to retrieve the needed XKB controls.
This implementation is analogous to Xnest one at commit 83fef4235.
Supersedes: https://patchwork.freedesktop.org/patch/67504
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
This patch introduces a new flag ATTR_KEY for hotplugged input devices,
so we can better distinguish between real keyboards (i.e. devices with
udev property ID_INPUT_KEYBOARD="1") and other key input devices like
lid switches, power buttons, etc.
All supported hotplug backends (udev, hal, and wscons) will set both
flags ATTR_KEY and ATTR_KEYBOARD for real keyboards, but udev backend
will set ATTR_KEY, but not ATTR_KEYBOARD, for non-keyboard key input
devices (hal and wscons will set both flags in any case). With this
distinction, kdrive input hotplugging mechanism will be allowed to only
grab real keyboards, as other key input devices are currently not
supported.
In order to don't break current behaviour, this patch will replace all
ATTR_KEYBOARD occurrences with ATTR_KEY in hw/xfree86/common/xf86Xinput.c.
[ajax: Just add ATTR_KEY, don't re-number the other attributes]
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
Option -seat passed to Xephyr requires -sw-cursor to be passed as well,
otherwise the mouse cursor will remain invisible for the given seat.
This patch takes care of enabling -sw-cursor if -seat is passed.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
This patch introduces convenient command-line options -xkb-rules,
-xkb-model, -xkb-layout, -xkb-variant, and -xkb-options, to set default
values for these properties.
These options can be handful for cases in which compile-time default
values don't match user locale, since kdrive doesn't support InputClass
matching rules yet and not all Linux distros provide default rules to
store these values in udev properties (which by the way is a discouraged
practice).
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
This patch introduces input hot-plugging support for kdrive-based
applications in multi-seat context. This feature is enabled by passing
-seat option with desired seat name. All keyboard/mouse devices assigned
to that seat will be automatically grabbed by kdrive.
It supports udev and hal backends for input hot-plugging support.
Another patches may be required for wscons backend.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=33140
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
Implement missing parts in kdrive evdev driver for
correct update of evdev keyboard LEDs.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=22302
[ajax: Fixed deref-before-null-check bug]
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
216bdbc735 removed the SetRootClip call in the XWayland output-hotplug
handler when running rootless (e.g. as a part of Weston/Mutter), since
the root window has no storage, so generating exposures will result in
writes to invalid memory.
Unfortunately, preventing the segfault also breaks sprite confinement.
SetRootClip updates winSize and borderSize for the root window, which
when combined with RRScreenSizeChanged calling ScreenRestructured,
generates a new sprite-confinment area to update it to the whole screen.
Removing this call results in the window geometry being reported
correctly, but winSize/borderSize never changing from their values at
startup, i.e. out of sync with the root window geometry / screen
information in the connection info / XRandR.
This patch introduces a hybrid mode, where we update winSize and
borderSize for the root window, enabling sprite confinement to work
correctly, but keep the clip emptied so exposures are never generated.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Tested-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
When used for single-GPU multi-seat purposes, there's no need to enable
ephyr virtual input devices, since Xephyr is supposed to handle its own
hardware devices.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
Multi-seat-capable display managers commonly pass command-line options
like "-novtswitch", "-sharevts", or "-layout seatXXXX" to Xorg server,
but Xephyr currently refuses to start if these options are passed to it,
which may break Xephyr-based single-GPU multiseat setups.
[ajax: shortened summary]
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
KDrive evdev driver deliberately name grabbed devices as "Evdev mouse"
or "Evdev keyboard". This patch will make it skip this step if grabbed
devices are already named (i.e. from udev).
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
If kdrive input driver "evdev" is available, no other driver was
explicitly set for a given input device, and its kernel device node is
/dev/input/event*, this patch will make kdrive set "evdev" driver
automatically for such device.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
This patch simplifies NewInputDeviceRequest() implementation in
kinput.c, making use of improved KdParseKbdOptions() /
KdParsePointerOptions() and merging several "if (ki)"/"if (pi)" clauses.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
This should fix aborts()s from epoxy on old software stacks.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This only worked if the backend server supported DRI1, which is
stunningly unlikely these days.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
On desktop GL, ask for a 3.1 core profile context if that's available,
otherwise create a generic context.
v2: tell glamor the profile is a core one.
v2.1: add/use GL version defines
v3: let glamor work out core itself
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Core contexts require the use of vertex array objects, so switch both glamor
and ephyr/glamor over.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Eliminates polling every 20ms for device input.
v2: rename ephyrPoll to ephyrXcbNotify and fix the API so it can be
used directly for SetNotifyFd. Thanks to Daniel Martin
<consume.noise@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Cc: Daniel Martin <consume.noise@gmail.com>
This switches the kdrive code to use FD notification for input
devices, rather than the block and wakeup handlers.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Replace the block/wakeup handlers with a NotifyFd callback
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Globally replace #ifdef and #if defined usage of 'sun' with '__sun'
such that strict ISO compiler modes such as -ansi or -std=c99 can be used.
Signed-off-by: Richard PALO <richard@NetBSD.org>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
The number of FDs has been decremented already, therefore the
number contained the index of the top one that is to me moved down.
This problem was introduced by:
commit 1110b71e36
Author: Chris Clayton <chris2553@googlemail.com>
kdrive: fix build error on gcc 4.8 for out-of-bounds array access
The reason for the warning was likely a confused compiler.
Hoping to reduce the confusion by moving the decrement behind the end
if the copy loop.
Signed-off-by: Egbert Eich <eich@suse.de>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
There's nothing in configure to enable this, and KdTsPhyScreen isn't
defined anywhere.
[ajax: Rebase, also clean up Xfbdev]
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Jamey Sharp <jamey@minilop.net>
These have never been used in the history of the tree, and were
producing string literal const loss warnings.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
This improves the case for when we paint an area without SHM.
xcb_image_subimage() is used to create a subimage for the damaged area, which
is converted to native format if necessary.
Signed-off-by: Ian Scott <ian.scott@arteris.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
This is a trivial patch that moves host_has_extension() implementation
from ephyr.c to hostx.c so that it can be called by hostx.c internal
functions. Also rename function to hostx_has_extension() for consistency.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Laércio de Sousa <laerciosousa@sme-mogidascruzes.sp.gov.br>
xorg/xserver/hw/kdrive/ephyr/ephyr.c:979:9: error: format ‘%d’ expects argument of type ‘int’, but argument 4 has type ‘ScreenPtr’ [-Werror=format=]
This looks like a genuine bug, and ephyrCursorScreen->myNum was meant here
rather than ephyrCursorScreen
v2:
Insert a ":" as well
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
I doubt anyone builds with this turned off or has done for a long
time.
It helps my eyes bleed slightly less when reading the code, I've left
the define in place as some drivers use it.
Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The image is created in the native byte order of the machine Xephyr is
rendered on however drawn in the image byte order of the Xephyr server.
Correct byte order in the xcb_image_t structure and convert to native
before updating the window.
If depths of Xephyr and host server differ this is already taken care of
by the depth conversion routine.
It is a terrible wase to always convert and transmit the entire image
no matter of the size of the damaged area. One should probably use
sub-images here. For now we leave this as an exercise.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
xcb_image_put() prints the entire image, therefore don't use an offset.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
The DDX specific command line parsing function only gets called
if command line arguments are present. Therefore this function
is not suitable to initialize mandatory global variables.
Replace main() instead.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
ephyr_glamor_connect() returns NULL if we failed, but applying
xcb_connection_has_error() to NULL is not permitted.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
By dropping the unconditional logic op disable at the end of
rendering, this fixes GL errors being thrown in GLES2 contexts (which
don't have logic ops). On desktop, this also means a little less
overhead per draw call from taking one less trip through the
glEnable/glDisable switch statement of doom in Mesa.
The exchange here is that we end up taking a trip through it in the
XV, Render, and gradient-generation paths. If the glEnable() is
actually costly, we should probably cache our logic op state in our
screen, since there's no way the GL could make that switch statement
as cheap as the caller caching it would be.
v2: Don't forget to set the logic op in Xephyr's drawing.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Remove these defines as we start to remove support for non-standard
glamor layering as used by the intel driver.
v2: Rebase on the blockhandler change and the Xephyr init failure
change (by anholt), fix stray NO_DRI3 addition to xwayland.
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Eric Anholt <eric@anholt.net>