Kristian Høgsberg [Wed, 30 Oct 2013 23:27:16 +0000 (16:27 -0700)]
launcher: Make sure we drop drm master before we allow VT switches
On shutdown, we can risk having a pending vt switch that we normally
handle in the vt signal handler. However as we put the vt back in VT_AUTO
mode, the pending VT switch will go through and if we haven't dropped
drm master at that point, we could switch to another display server
without dropping drm master. That will typically crash the other server,
so let's try to make sure we don't do that.
https://bugs.freedesktop.org/show_bug.cgi?id=70877
Kristian Høgsberg [Wed, 30 Oct 2013 22:49:45 +0000 (15:49 -0700)]
shell: Create view for lock surface
Nothing created the view for the lock surface and weston would crash
when trying to display it.
https://bugs.freedesktop.org/show_bug.cgi?id=70873
Kristian Høgsberg [Wed, 30 Oct 2013 20:53:01 +0000 (13:53 -0700)]
logind: Return proper error value when tty or seat doesn't match
If the seat or tty doesn't match we return with r == 0, which looks like
success to weston_launcher_connect(), which then fails to fall back
to the legacy path.
https://bugs.freedesktop.org/show_bug.cgi?id=70876
Kristian Høgsberg [Wed, 30 Oct 2013 16:07:04 +0000 (09:07 -0700)]
animation.c: Schedule a repaint when the animation is done
Otherwise we don't repaint with the final state of the surface and
we're stuck with the second-to-last frame of the animation until
something else (moving the mouse or such) triggers a redraw.
https://bugs.freedesktop.org/show_bug.cgi?id=70930
Kristian Høgsberg [Wed, 30 Oct 2013 15:59:55 +0000 (08:59 -0700)]
Fall back to ANIMATION_NONE if startup-animation is zoom
get_animation_type() parses "none", "zoom" and "fade" but for the
startup animation, we only support "none" and "fade". If we get "zoom"
just fall back to "none" like we do for all unrecognized strings.
https://bugs.freedesktop.org/show_bug.cgi?id=71006
Jason Ekstrand [Mon, 28 Oct 2013 02:32:54 +0000 (21:32 -0500)]
toytoolkit: Don't segfault on window close
Without this commit, the client will segmentation fault due to accessing
the frame after everything has been destroyed.
Kristian Høgsberg [Tue, 29 Oct 2013 05:06:11 +0000 (22:06 -0700)]
window.c: Don't ungrab input after displaying the menu
We used to ungrab first to stop any existing grab and then grab after
showing the menu. That was broken in
c680e904897dfb19331f8f296ab, which
moved the ungrab down below the grab, and as a result menus are now
shown without a grab. This commit moves the grab back up.
Kristian Høgsberg [Mon, 28 Oct 2013 22:32:02 +0000 (15:32 -0700)]
input.c: Rename default grab functions to be consistent
Jonas recent patches made the inconsistency obvious, this patch fixes it.
Jonas Ådahl [Fri, 25 Oct 2013 21:18:05 +0000 (23:18 +0200)]
Add cancel function to grab interfaces
A grab can potentially allocate memory and would normally end the grab
itself, freeing the allocated memory in the process. However at in some
situations the compositor may want to abort a grab. The grab owner still
needs to free some memory and abort the grab properly. To do this a new
function 'cancel' is introduced in all the grab interfaces instructing
the grabs owner to abort the grab.
This patch also hooks up grab cancelling to seat device releasing and
when the compositor looses focus, which would potentially leak memory
before.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Fri, 25 Oct 2013 21:18:04 +0000 (23:18 +0200)]
shell: Don't leak weston_touch_move_grab structs
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Fri, 25 Oct 2013 21:18:03 +0000 (23:18 +0200)]
input: Remove unused variable
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Fri, 25 Oct 2013 21:18:02 +0000 (23:18 +0200)]
configure.ac: Report if built with XCB-XKB or not
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Fri, 25 Oct 2013 21:18:01 +0000 (23:18 +0200)]
compositor-x11: Hide update_xkb_keymap behind #ifdef HAVE_XCB_XKB
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Fri, 25 Oct 2013 21:18:00 +0000 (23:18 +0200)]
tests/.gitignore: Add *.trs
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Tomeu Vizoso [Mon, 28 Oct 2013 09:17:45 +0000 (10:17 +0100)]
rpi: Remove create_view and destroy_view implementations
And also remove the vfunc definitions from the compositor as they
are unused now.
Kristian Høgsberg [Fri, 25 Oct 2013 23:18:17 +0000 (16:18 -0700)]
configure.ac: Require wayland >= 1.3.91
Kristian Høgsberg [Fri, 25 Oct 2013 20:34:58 +0000 (13:34 -0700)]
weston-launch: Revoke and close input fds when we vt switch away
We used to leak the input fds, as weston would reopen all fds on vt
enter. We could just close them after sending the open fd through the
socket, but this patch also adds support for the new EVIOCREVOKE evdev
ioctl, that revokes the fd in question (including the copy that we
sent to the compositor).
https://bugs.freedesktop.org/show_bug.cgi?id=70849
Kristian Høgsberg [Fri, 25 Oct 2013 19:49:00 +0000 (12:49 -0700)]
configure.ac: Require at least version 0.3.0 of xkbcommon
https://bugs.freedesktop.org/show_bug.cgi?id=70870
Kristian Høgsberg [Fri, 25 Oct 2013 19:15:42 +0000 (12:15 -0700)]
.gitignore: Add .valid files
Ander Conselvan de Oliveira [Fri, 25 Oct 2013 13:26:34 +0000 (16:26 +0300)]
pixman-renderer, gl-renderer: Destroy surface state with the renderer
Previously the renderers destroy function assumed they are only called
when the compositor is shutting down and that the compositor had
already destroyed all the surfaces. However, if a runtime renderer
switch would be done, the surface state would be leaked.
This patch adds a destroy_signal to the pixman and gl renderers. The
surface state objects will listen for that signal and destroy
themselves if needed.
This is a step towards runtime switchable renderers.
Ander Conselvan de Oliveira [Fri, 25 Oct 2013 13:26:33 +0000 (16:26 +0300)]
compositor: Let renderers create and destroy surface state on their own
Remove create_surface() and destroy_surface() from the renderer
interface and change the renderers to create surface state on demand
and destroy it using the weston_surface's destroy signal.
Also make sure the surfaces' renderer state is reset to NULL on
destruction.
This is a step towards runtime switchable renderers.
(rpi-renderer changes are only compile-tested)
Ander Conselvan de Oliveira [Fri, 25 Oct 2013 13:26:32 +0000 (16:26 +0300)]
pixman-renderer, gl-renderer: Destroy debug bindings on clean up
Also make sure backends destroy the renderers before shutting down the
compositor to avoid a double call to weston_binding_destroy().
This is a step towards making renderers switchable during runtime.
Ander Conselvan de Oliveira [Fri, 25 Oct 2013 13:26:31 +0000 (16:26 +0300)]
pixman-renderer: Initialize pixman-renderer struct with zeros
Othrewise a crash may happen because of an unitialized value of the
create_view field.
Tomeu Vizoso [Fri, 25 Oct 2013 08:34:38 +0000 (10:34 +0200)]
rpi: Protect in-use DispmanX resources from premature deletion
The compositor will check if the client destroyed the wl_buffer
while it was in use in a display update, and delete the resource
itself once the update has finished.
Tomeu Vizoso [Fri, 25 Oct 2013 08:34:37 +0000 (10:34 +0200)]
rpi: Remove dangling reference to weston_view
This also marks the rpir_view for destroy in rpi_renderer_finish_frame()
Tomeu Vizoso [Thu, 24 Oct 2013 13:38:33 +0000 (15:38 +0200)]
rpi: Only check for prematurely destroyed wl_buffers in the EGL case
Tomeu Vizoso [Thu, 24 Oct 2013 13:38:32 +0000 (15:38 +0200)]
rpi: EGL surfaces need to be swapped always that we have an incoming back buffer
Tomeu Vizoso [Thu, 24 Oct 2013 13:38:31 +0000 (15:38 +0200)]
rpi: Fix logging of buffer swaps for the EGL case
Tomeu Vizoso [Thu, 24 Oct 2013 13:38:30 +0000 (15:38 +0200)]
rpi: Initialize surface's list of views
Tomeu Vizoso [Thu, 24 Oct 2013 13:38:29 +0000 (15:38 +0200)]
rpi: Link with EGL if ENABLE_EGL
Peter Hutterer [Wed, 23 Oct 2013 04:57:02 +0000 (14:57 +1000)]
protocol: hook up xmllint to check validity
Kristian Høgsberg [Thu, 24 Oct 2013 23:52:00 +0000 (16:52 -0700)]
shell.c: Prevent moving a window from more than one source
It was possible to start a move or resize with a pointer and then
start another move for the same surface with touch or a pointer from
another seat, which would make the window flicker around. With this
commit, we now reject any attempts to move or resize a surface that is
currently grabbed.
Kristian Høgsberg [Thu, 24 Oct 2013 23:29:51 +0000 (16:29 -0700)]
shell.c: Don't overwrite focus surface argument in activate_binding()
We need either the pointer or touch focus, which is passed in to
activate_binding(). Don't overwrite this with the pointer focus.
Kristian Høgsberg [Thu, 24 Oct 2013 23:21:39 +0000 (16:21 -0700)]
shell: Don't cast weston_view to surface when we just need view->surface
Most of the shell.c bindings take effect on the current pointer or
touch focus. That's now a view, but we need a surface in these cases.
Just get the surface from view->surface instead of trying to cast a
view to a surface.
Kristian Høgsberg [Thu, 24 Oct 2013 22:54:49 +0000 (15:54 -0700)]
compositor: Remove two unused variables
Kristian Høgsberg [Thu, 24 Oct 2013 22:51:35 +0000 (15:51 -0700)]
shell: Create a view for the grab_surface
Without the view, we can't set pointer and touch focus to the grab_surface
and touch moving fails.
Kristian Høgsberg [Thu, 24 Oct 2013 21:21:53 +0000 (14:21 -0700)]
input: Properly handle setting touch focus to NULL
This happens on TOUCH_UP, and we unconditionally dereference view->surface.
Kristian Høgsberg [Thu, 24 Oct 2013 06:38:50 +0000 (23:38 -0700)]
compositor-x11: Use the fullscreen size provided by the configure event
Make sure we initialize the x11 shm state using the right window size.
https://bugs.freedesktop.org/show_bug.cgi?id=70590
Kristian Høgsberg [Thu, 24 Oct 2013 06:36:04 +0000 (23:36 -0700)]
shell: Add the right output object to the shell output_list
Error on my part while renaming the shell_output variable.
Xiong Zhang [Wed, 23 Oct 2013 05:58:32 +0000 (13:58 +0800)]
shell: register output->destroy_signal handler
setup_output_destroy_handler() deal with output created at
drm backend initialize time.
handle_output_create() deal with output created by hot plug handler
output_destroy_handler is removed when output was unplugged or
shell is destroyed.
Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com>
Xiong Zhang [Wed, 23 Oct 2013 05:58:35 +0000 (13:58 +0800)]
window, desktop-shell: deal with output unplug on client side
when output is removed, weston-desktop-shell should destroy panel
and background surface on destroyed output.
Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com>
Xiong Zhang [Wed, 23 Oct 2013 05:58:31 +0000 (13:58 +0800)]
compositor: set surface->plane from destroyed plane to NULL
In drm backend, the cursor_surface->plane point to
drm_output->cursor_plane.when this output is removed,
drm_output->cursor_plane is destroyed, butcursor_surface->plane
still point to destroyed plane. So once mouse move to this
cursor_surface and system will repaint this cursor_surface,
segment fault will generate in weston_surface_damage_below() function.
V2:
-set surface->plane to NULL whose plane point to unplugged output,
then change weston_surface_damage_below() to do nothing if
surface->plane is NULL (Kristian)
-set surface->plane to NULL in weston_surface_unmap(),
so that all surfaces that have a non-NULL plane pointer wil be
on compositor->surface_list (Kristian).
bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69777
Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com>
Kristian Høgsberg [Thu, 24 Oct 2013 05:12:13 +0000 (22:12 -0700)]
window.c: Don't put titlebars on menu windows
Kristian Høgsberg [Thu, 24 Oct 2013 04:49:30 +0000 (21:49 -0700)]
window.c: Use frame code for drawing menus
This gives us a nice frame and drop shadows for the menus.
Kristian Høgsberg [Thu, 24 Oct 2013 03:47:35 +0000 (20:47 -0700)]
terminal: Show character grid size in header bar when resizing
We don't have a reliable way to know when to clear this indicator.
Typically the pointer will still be over the window when the resize is
done and we'll get an enter event, but if the window sets a max size
the pointer may be over another window when the resize is done.
We'll need a new wl_shell (or more likely xdg_shell) event for this.
Kristian Høgsberg [Wed, 23 Oct 2013 23:52:05 +0000 (16:52 -0700)]
terminal: Add context menu with new terminal/copy/paste
Kristian Høgsberg [Wed, 23 Oct 2013 23:29:14 +0000 (16:29 -0700)]
terminal: Add scroll-back history
Ctrl-Shift-UP/DOWN to scroll through the history.
Kristian Høgsberg [Wed, 23 Oct 2013 16:30:48 +0000 (09:30 -0700)]
configure.ac: Bump version to 1.3.90 for the master branch
Rui Matos [Thu, 10 Oct 2013 17:44:22 +0000 (19:44 +0200)]
clients/window: Dispose of previous keymap and state on keymap change
Rui Matos [Thu, 10 Oct 2013 17:44:21 +0000 (19:44 +0200)]
compositor-wayland: Handle keymap changes
Rui Matos [Thu, 10 Oct 2013 17:44:20 +0000 (19:44 +0200)]
compositor-x11: Update keymap when XKB keymap changes
Rui Matos [Thu, 10 Oct 2013 17:44:19 +0000 (19:44 +0200)]
input: Add core API to update the keymap
How and when to update the keymap is left to each backend.
The new keymap only becomes effective when no keys are pressed and we
keep latched and locked modifiers from the previous state.
Jason Ekstrand [Sun, 13 Oct 2013 03:38:11 +0000 (22:38 -0500)]
Split the geometry information from weston_surface out into weston_view
The weston_surface structure is split into two structures:
* The weston_surface structure storres everything required for a
client-side or server-side surface. This includes buffers; callbacks;
backend private data; input, damage, and opaque regions; and a few other
bookkeeping bits.
* The weston_view structure represents an entity in the scenegraph and
storres all of the geometry information. This includes clip region,
alpha, position, and the transformation list as well as all of the
temporary information derived from the geometry state. Because a view,
and not a surface, is a scenegraph element, the view is what is placed
in layers and planes.
There are a few things worth noting about the surface/view split:
1. This is *not* a modification to the protocol. It is, instead, a
modification to Weston's internal scenegraph to allow a single surface
to exist in multiple places at a time. Clients are completely unaware
of how many views to a particular surface exist.
2. A view is considered a direct child of a surface and is destroyed when
the surface is destroyed. Because of this, the view.surface pointer is
always valid and non-null.
3. The compositor's surface_list is replaced with a view_list. Due to
subsurfaces, building the view list is a little more complicated than
it used to be and involves building a tree of views on the fly whenever
subsurfaces are used. However, this means that backends can remain
completely subsurface-agnostic.
4. Surfaces and views both keep track of which outputs they are on.
5. The weston_surface structure now has width and height fields. These
are populated when a new buffer is attached before surface.configure
is called. This is because there are many surface-based operations
that really require the width and height and digging through the views
didn't work well.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Kristian Høgsberg [Tue, 22 Oct 2013 20:19:23 +0000 (13:19 -0700)]
logind: Remove missing 'else'
Only fall through to cleaning up weston-launch or standalone state
if we didn't use logind.
Kristian Høgsberg [Tue, 22 Oct 2013 19:40:54 +0000 (12:40 -0700)]
compositor: Remove redundant and not well-defined focus field
It was never clear what this field really did.
Kristian Høgsberg [Tue, 22 Oct 2013 19:27:46 +0000 (12:27 -0700)]
compositor-drm: Don't call drop/set master anymore
This is done by either laucnher-util.c directly, weston-launch or
logind, depending on how we're launched.
David Herrmann [Tue, 22 Oct 2013 15:11:26 +0000 (17:11 +0200)]
compositor: finish frame if redraw fails
If we are about to finish a frame, but a redraw is pending and we let the
compositor redraw, we need to check for errors. If the redraw fails and
the backend cannot schedule a page-flip, we need to finish the frame,
anyway.
All backends except DRM use a timer to schedule frames. Hence, they cannot
fail. But for DRM, we need to be able to handle drmModePageFlip() failures
in case access got revoked.
This fixes a bug where logind+drm caused keyboard input to be missed as we
didn't reenable it after a failed page-flip during deactivation.
David Herrmann [Tue, 22 Oct 2013 15:11:25 +0000 (17:11 +0200)]
compositor-drm: finish frame if initial page-flip fails
If the initial page-flip fails, immediately finish the frame to avoid
being stuck in the given frame. We already do this if we have no fbo
available. Now we do the same if the page-flip fails.
Cameron Stewart [Fri, 18 Oct 2013 01:56:41 +0000 (01:56 +0000)]
Rename invocations of frame_create that were missed by
ee7fefcf
For some reason it not only compiled with different parameters but the
method succeeded and just failed later.
David Herrmann [Mon, 21 Oct 2013 22:28:09 +0000 (00:28 +0200)]
launcher: add logind backend
Instead of connecting to weston-launch from launcher-util, we now try to
connect to logind first. If logind provides session-devices, we use them.
If not, we fall back to the old weston-launch facility.
David Herrmann [Mon, 21 Oct 2013 22:28:08 +0000 (00:28 +0200)]
launcher: add weston_launcher_close() dummy
If you request a device via weston_launcher_open(), you should now release
it via weston_launcher_close() instead of close(). This is currently not
needed but will be required for logind devices.
David Herrmann [Mon, 21 Oct 2013 22:28:07 +0000 (00:28 +0200)]
dbus: add dbus-match helpers
These helpers simplify adding dbus-matches by allowing var-arg arguments
to assemble the matching rules.
David Herrmann [Mon, 21 Oct 2013 22:28:06 +0000 (00:28 +0200)]
Add optional dbus helpers
This adds optional libdbus integration for weston. If libdbus is available
and not disabled via --disable-dbus during weston build, we now provide
basic DBusConnection main-loop integration for weston.
The dbus.c file provides a new helper to integrate any DBusConnection
object into a wl_event_loop object. This avoids any glib/qt/..
dependencies but instead only uses the low-level libdbus library.
Note that we do not provide dummy fallbacks for dbus helpers in case
dbus-support is disabled. The reason for that is that you need dbus/dbus.h
for nearly any operation you want to do via dbus. Therefore, only the most
basic helpers which can be used independently provide a "static inline"
dummy fallback to avoid #ifdef all over the code.
Kristian Høgsberg [Mon, 21 Oct 2013 22:23:17 +0000 (15:23 -0700)]
window: Only clamp min size for windows with frame
Also fix width/height typo.
Kristian Høgsberg [Mon, 21 Oct 2013 22:14:44 +0000 (15:14 -0700)]
weston-launch: Use /bin/sh for launching weston
https://bugs.freedesktop.org/show_bug.cgi?id=68739
Kristian Høgsberg [Mon, 21 Oct 2013 22:05:49 +0000 (15:05 -0700)]
window: Set a minimum size for toytoolkit windows
The decorations tiles start to overlap and look weird if we go below
200x200 size windows. Just set that as a minimum size if the app
doesn't provide a bigger minimum size.
https://bugs.freedesktop.org/show_bug.cgi?id=66794
Jonas Ådahl [Thu, 17 Oct 2013 21:04:08 +0000 (23:04 +0200)]
udev-seat: Repick seat after a new device was added
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Thu, 17 Oct 2013 21:04:07 +0000 (23:04 +0200)]
input: Unmap pointer sprite when no more pointer devices are connected
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Thu, 17 Oct 2013 21:04:06 +0000 (23:04 +0200)]
input: Unset focus of seat device when releasing last reference
When the last input device with a certain capability is removed, unset
the focus of the seat device associated with the capability.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Jonas Ådahl [Thu, 17 Oct 2013 21:04:05 +0000 (23:04 +0200)]
evdev: Reference count input device's seat capabilities
When the only input device of a certain seat capability is unplugged,
stop advertising the capability.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Kristian Høgsberg [Wed, 16 Oct 2013 23:31:42 +0000 (16:31 -0700)]
compositor-drm: Support configuring the framebuffer format from weston.ini
This patch adds a new weston.ini key, gbm-format, to the [core] section.
This new key can be rgb565, xrgb8888 or xrgb2101010, and makes the
compositor use the corresponding GBM format for the framebuffer.
Kristian Høgsberg [Wed, 16 Oct 2013 23:15:11 +0000 (16:15 -0700)]
compositor-drm: Use a param struct for initialization like other backends
Kristian Høgsberg [Wed, 16 Oct 2013 23:08:57 +0000 (16:08 -0700)]
compositor-drm: Make drm_output_check_scanout_format() work for all formats
This used to work only for ARGB8888 and XRGB8888 buffers, but this lets
us support pageflipping to any client buffer as long as it matches the
framebuffer format.
Kristian Høgsberg [Wed, 16 Oct 2013 23:06:18 +0000 (16:06 -0700)]
window: Set opaque region to window size if we're fullscreen
Kristian Høgsberg [Wed, 16 Oct 2013 18:38:24 +0000 (11:38 -0700)]
shell: Add a new weston.ini to control the startup animation
The new key startup-animation in the [shell] section lets you
control the startup animation. Default is fade, but with this patch
we can also do none, which just show the desktop as fast as possible.
Kristian Høgsberg [Wed, 16 Oct 2013 18:10:12 +0000 (11:10 -0700)]
compositor: Stop the repaint loop if the compositor went to sleep
We check the state when we schedule a repaint, but we will still repaint
in weston_output_finish_frame() if a repaint is needed. Now we check
whether we went to sleep while waiting for the page flip and stop repainting
in that case.
https://bugs.freedesktop.org/show_bug.cgi?id=65802
Kristian Høgsberg [Wed, 16 Oct 2013 05:03:31 +0000 (22:03 -0700)]
simple-egl: Add support for using a 16 bpp EGLConfig
David Herrmann [Tue, 15 Oct 2013 12:29:57 +0000 (14:29 +0200)]
fbdev: open launcher only once
We currently call launcher_connect() twice, which is redundant and
amazingly works (ugh?). Fix this and connect only once to the launcher.
David Herrmann [Tue, 15 Oct 2013 12:29:56 +0000 (14:29 +0200)]
evdev: release devices on read() error
If read() fails without EAGAIN/EINTR, the device is very likely dead.
However, we must not remove the device as it might be muted/revoked. So we
simply remove the event-source to avoid polling the device and simply wait
for the udev-remove signal now.
Note that we cannot call evdev_device_destroy() as the caller created the
FD and might need custom code to close it (like weston_launcher_close()).
Kristian Høgsberg [Mon, 14 Oct 2013 22:46:13 +0000 (15:46 -0700)]
evdev: Clear touch capability if we see a mouse-type button
If an input device has BTN_LEFT (typically) it's not a touch screen but
a touch pad.
Kristian Høgsberg [Mon, 14 Oct 2013 22:32:08 +0000 (15:32 -0700)]
evdev: Only init a pointer if the evdev device has a button
We used to test for abs | rel | button, which inits a pointer device for
a device with just rel or abs capabilities. We now make sure we have either
rel or abs as well as button.
Kristian Høgsberg [Mon, 14 Oct 2013 22:28:01 +0000 (15:28 -0700)]
evdev: Stop looking for pointer buttons when we get to BTN_JOYSTICK
We don't want to mark a touchscreen as a button device just because it
exposes the BTN_TOUCH and BTN_TOOL buttons.
Ander Conselvan de Oliveira [Mon, 14 Oct 2013 12:57:11 +0000 (15:57 +0300)]
gl-renderer: Build as a loadable module
The time spent loading EGL and GLES libraries from disk can be a
considerable hit in some embedded use cases. If Weston is compiled
with EGL support, the binary will depend on those libraries, even if
a software renderer is in use.
This patch splits the GL renderer into a separate loadable module,
and moves the dependency on EGL and GLES to it. The backends still
need the EGL headers for the native types and EGLint. The function
load_module() is renamed to weston_load_module() and exported, so
that it can be used by the backends.
The gl renderer interface is changed so that there is only one symbol
that needs to be dlsym()'d. This symbol contains pointers to all the
functions and data necessary to interact with the renderer. As a side
effect, this change simplifies gl-renderer.h a great deal.
Ander Conselvan de Oliveira [Mon, 14 Oct 2013 12:57:10 +0000 (15:57 +0300)]
build: Add dependency between libdrm and compositor for drm backend
The compositor uses libdrm in launcher-util.c if the drm backend is
built, but there was no explicit requirement in the build. egl brings
libdrm implicity so the build doesn't fail.
This patch adds an explicit dependency between the compositor and
libdrm if the drm backend is built, so that changes to the compositor
modules don't cause build failures.
Kristian Høgsberg [Mon, 14 Oct 2013 20:59:53 +0000 (13:59 -0700)]
launcher: Drop unecessary checks for drm_fd == -1
We never get into this case, and if we do, we just want to pass the invalid
fd through to the underlying ioctl and get the error that way."
Kristian Høgsberg [Mon, 14 Oct 2013 20:57:44 +0000 (13:57 -0700)]
launcher: Fix reverted sense of drm_check_master() wrapper
Also rename to drm_is_master() to make it clearer what we're checking.
https://bugs.freedesktop.org/show_bug.cgi?id=70459
Neil Roberts [Thu, 3 Oct 2013 15:43:07 +0000 (16:43 +0100)]
Don't remove the touch grab until the last touch point is removed
Previously if you move a window around and temporarily add a second
finger then it will cancel the grab even though the original finger is
still held on the screen. It seems more robust to avoid cancelling the
grab until all fingers have been removed.
Neil Roberts [Thu, 3 Oct 2013 15:43:06 +0000 (16:43 +0100)]
Only update the touch grab position for the first finger
Previously if you add a second finger while moving a window with a
touch grab then the position will keep jumping between the position of
each finger as you move them around. This patch changes it so that it
keeps track of the first touch id that starts the grab and only
updates the grab position when that finger moves.
Neil Roberts [Thu, 3 Oct 2013 15:43:05 +0000 (16:43 +0100)]
Add a touch move binding
When holding the compositor super key the touch events can now be used
to move a window.
Neil Roberts [Thu, 3 Oct 2013 15:43:04 +0000 (16:43 +0100)]
Add a touch binding to activate a surface
Adds a new binding type for touch events via the new function
weston_compositor_add_touch_binding. The binding can only be added for
a touch down with the first finger. The shell now uses this to install
a binding to activate the current surface.
Kristian Høgsberg [Mon, 14 Oct 2013 05:11:07 +0000 (22:11 -0700)]
xwm: Fall back to xcb_kill_client() if window does support WM_DELETE_WINDOW
Jason Ekstrand [Mon, 14 Oct 2013 00:08:41 +0000 (19:08 -0500)]
xwayland: Use decoration frame from cairo-util
In particular, this gives us a close button on xwayland windows.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Mon, 14 Oct 2013 00:08:40 +0000 (19:08 -0500)]
Use cairo-util frame in tinytoolkit
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Mon, 14 Oct 2013 00:08:39 +0000 (19:08 -0500)]
Add decoration frame support to cairo-util
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Mon, 14 Oct 2013 00:08:38 +0000 (19:08 -0500)]
Rename frame_create and frame_set_child_size with a window_ prefix
This is to avoid a collision with an actual frame structure.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Kristian Høgsberg [Fri, 11 Oct 2013 23:48:19 +0000 (16:48 -0700)]
xwm: Redirect root window subwindows
xwayland no longer does this for us, we have to request it now
Kristian Høgsberg [Fri, 11 Oct 2013 22:25:42 +0000 (15:25 -0700)]
fbdev: Fix compilation without EGL
We have to duplicate #defines and typedefs for the EGL types and constants
we use in gl-renderer.h so we can compile the fbdev backend without EGL.
https://bugs.freedesktop.org/show_bug.cgi?id=70392
Alexandru DAMIAN [Wed, 2 Oct 2013 16:51:05 +0000 (17:51 +0100)]
compositor-drm: fix EGL format type
EGLInt is not always uint32_t so we need
to make sure we use the right int size for the format.
Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Stefan Schmidt [Thu, 10 Oct 2013 13:28:30 +0000 (14:28 +0100)]
clients: Update .gitignore for weston-multi-resource
Signed-off-by: Stefan Schmidt <s.schmidt@samsung.com>
Stefan Schmidt [Thu, 10 Oct 2013 13:28:29 +0000 (14:28 +0100)]
xwayland: Remove unused call weston_wm_handle_configure_notify.
We get the child position but never use this information here. Just remove it.
Spotted by Christopher Michael.
Signed-off-by: Stefan Schmidt <s.schmidt@samsung.com>