We used to rely on the order in which the
weston_compositor::destroy_signal callbacks happened, to not access
freed memory. Don't know when, but this broke at least with ivi-shell,
which caused crashes in random places on compositor shutdown.
Valgrind found the following:
Invalid write of size 8
at 0xC2EDC69: unbind_input_panel (input-panel-ivi.c:340)
by 0x4E3B6BB: destroy_resource (wayland-server.c:537)
by 0x4E3E085: for_each_helper.isra.0 (wayland-util.c:359)
by 0x4E3E60D: wl_map_for_each (wayland-util.c:365)
by 0x4E3BEC7: wl_client_destroy (wayland-server.c:675)
by 0x4182F2: text_backend_notifier_destroy (text-backend.c:1047)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Address 0x67ea360 is 208 bytes inside a block of size 232 free'd
at 0x4C2A6BC: free (vg_replace_malloc.c:473)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Invalid write of size 8
at 0x4E3E0D7: wl_list_remove (wayland-util.c:57)
by 0xC2EDEE9: destroy_input_panel_surface (input-panel-ivi.c:191)
by 0x4E3B6BB: destroy_resource (wayland-server.c:537)
by 0x4E3BC7B: wl_resource_destroy (wayland-server.c:550)
by 0x40DB8B: wl_signal_emit (wayland-server-core.h:264)
by 0x40DB8B: weston_surface_destroy (compositor.c:1883)
by 0x40DB8B: weston_surface_destroy (compositor.c:1873)
by 0x4E3B6BB: destroy_resource (wayland-server.c:537)
by 0x4E3E085: for_each_helper.isra.0 (wayland-util.c:359)
by 0x4E3E60D: wl_map_for_each (wayland-util.c:365)
by 0x4E3BEC7: wl_client_destroy (wayland-server.c:675)
by 0x4182F2: text_backend_notifier_destroy (text-backend.c:1047)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Address 0x67ea370 is 224 bytes inside a block of size 232 free'd
at 0x4C2A6BC: free (vg_replace_malloc.c:473)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Invalid write of size 8
at 0x4E3E0E7: wl_list_remove (wayland-util.c:58)
by 0xC2EDEE9: destroy_input_panel_surface (input-panel-ivi.c:191)
by 0x4E3B6BB: destroy_resource (wayland-server.c:537)
by 0x4E3BC7B: wl_resource_destroy (wayland-server.c:550)
by 0x40DB8B: wl_signal_emit (wayland-server-core.h:264)
by 0x40DB8B: weston_surface_destroy (compositor.c:1883)
by 0x40DB8B: weston_surface_destroy (compositor.c:1873)
by 0x4E3B6BB: destroy_resource (wayland-server.c:537)
by 0x4E3E085: for_each_helper.isra.0 (wayland-util.c:359)
by 0x4E3E60D: wl_map_for_each (wayland-util.c:365)
by 0x4E3BEC7: wl_client_destroy (wayland-server.c:675)
by 0x4182F2: text_backend_notifier_destroy (text-backend.c:1047)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Address 0x67ea368 is 216 bytes inside a block of size 232 free'd
at 0x4C2A6BC: free (vg_replace_malloc.c:473)
by 0x4084FB: wl_signal_emit (wayland-server-core.h:264)
by 0x4084FB: main (compositor.c:5465)
Looking at the first of these, unbind_input_panel() gets called when the
text-backend destroys its helper client which has bound to input_panel
interface. This happens after the shell's destroy_signal callback has
been called, so the shell has already been freed.
The other two errors come from
wl_list_remove(&input_panel_surface->link);
which has gone stale when the shell was destroyed
(shell->input_panel.surfaces list).
Rather than creating even more destroy listeners and hooking them up in
spaghetti, modify text-backend to not hook up to the compositor destroy
signal. Instead, make it the text_backend_init() callers' responsibility
to also call text_backend_destroy() appropriately, before the shell goes
away.
This fixed all the above Valgrind errors, and avoid a crash with
ivi-shell when exiting Weston.
Also using desktop-shell exhibited similar Valgrind errors which are
fixed by this patch, but those didn't happen to cause any crashes AFAIK.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-By: Derek Foreman <derekf@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation,
Following features are tested,
- add notification of ivi-layer with bad parameter
- add notification of ivi-surface configure with bad parameter
- add notification of creating ivi-layer with bad parameter
- add notification of creating ivi-surface with bad parameter
- add notification of removing ivi-layer with bad parameter
- add notification of removing ivi-surface with bad parameter
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
helper client.
Following features are tested,
- add notfication of ivi-surface with bad condition
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation.
Following features are tested,
- notification of adding ivi-layer
- notification of creating ivi-layer
- notification of removing ivi-layer
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
helper client.
Following features are tested,
- notification of adding ivi-surface
- notification of ivi-surface configure
- notification of creating ivi-surface
- notification of removing ivi-surface
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
The TESTs in ivi_layout-test.c may have several server-side parts
(RUNNER_TEST in ivi_layout-test-plugin.c) each. Sometimes we need to
carry state from one RUNNER_TEST to another within one TEST, but not
across multiple TESTs. The correct lifetime of that state would be the
lifetime (and identity) of the runner_resource, as one TEST creates and
uses at most one weston_test_runner during its lifetime.
However, tests are executed one by one. Take a shortcut, and use a static
global for storing that state. This turns the test_context into a
singleton. To ensure it is not confused between multiple TESTs, add
asserts to verify its identity.
Following patches will add tests for notification callbacks. These will
be using the carried state.
[Pekka: add serialization checks, rename the global, rewrite commit message.]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation,
Following features are tested,
- ivi-screen operation with bad parameter
- render order with bad parameter
- destroy ivi-layer in the ivi-screen and call commit_changes
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation.
Following features are tested,
- ivi-screen id
- ivi-screen resolution
- render order of ivi-layers in ivi-screen
Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation.
Following features are tested for ivi-layer,
- create with bad parameter
- visibility with bad parameter
- opacity with bad parameter
- destination rectangle with bad parameter
- orientation with bad parameter
- dimension with bad parameter
- position with bad parameter
- source rectangle with bad parameter
- properties with bad parameter
- destroy ivi-layer and call set_visibility_commit_changes
- destroy ivi-layer, call set_opacity, and commit_changes
- destroy ivi-layer, call set_orientation, and commit_changes
- destroy ivi-layer, call set_dimension, and commit_changes
- call set_position, destroy ivi-layer, and commit_changes
- call set_source_rectangle, destroy ivi-layer, and commit_changes
- call set_destination_rectangle, destroy ivi-layer, and commit_changes
- create duplicate
- destroy ivi-layer and call get_layer
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
helper client.
Following features are tested,
- bad render order of ivi-surface on ivi-layer
- call commitchanges after a ivi_surface in render order is destoryed
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation.
Following features are tested for ivi-layer,
- create
- visibility
- opacity
- orientation
- dimension
- position
- destination rectangle
- source rectangle
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These test are implemented on test suite framework, which provides
helper client.
Following features are tested,
- render order of ivi-surfaces on ivi-layer
Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
internal method validation.
Following features are tested for ivi-surface,
- destination_rectangle with bad parameter
- orientation with bad parameter
- dimension with bad parameter
- position with bad parameter
- source_rectangle with bad parameter
- properties with bad parameter
Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
helper client.
Following features are tested,
- ivi_layout_runner with basic_test_names[]
- surface with bad opacity
- destroy ivi/wl_surface and call get_surface
- commit_changes_after_properties_set_surface_destroy with
surface_property_commit_changes_test_names[]
- call set_visibility, destroy ivi-surface, and commit_changes
- call set_opacity, destroy ivi-surface, and commit_changes
- call set_orientation, destroy ivi-surface, and commit_changes
- call set_dimension, destroy ivi-surface, and commit_changes
- call set_position, destroy ivi-surface, and commit_changes
- call set_source_rectangle, destroy ivi-surface, and commit_changes
- call set_destination_rectangle, destroy ivi-surface, and
commit_changes
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
These tests are implemented on test suite framework, which provides
helper client.
Following features are tested for ivi-surface
- orientation
- dimension
- position
- destination rectangle
- source rectangle
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
"remove" is not proper name beacause it destorys a layer. The name of
the api is changed from layer_remove to layer_destroy.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
In previous code, it sends notification whenever setter calls. This
patch fixs that notification will not happens if there is no change of
properties by setter.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Previous code sends notification without event mask. So the notification
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This interface is required for user to manage screens per IDs.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
To get property changes of ivi_layer, callbacks can be registered.
However there was no API to remove a callback rather than removing all
callbacks by using layer_remove_notification.
layer_remove_notification_by_callback can do it.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
To get property changes of ivi_surface, callbacks can be registered.
However there was no API to remove a callback rather than remove all
callbacks by using surface_remove_notification.
surface_remove_notification_by_callback can do it.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
if a controller calls ivi_layout_layer_create_with_demenstion with a ID
which is already created before, the API returns exist ivi_layer.
However addtionally, it shall count up ref count to destroy ivi_layer
when ref count is 0. The previous code will destroy ivi_layer
immediately even if the other code still refers ivi_layer.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
ivi_shell_surface lifetime shall follow the ivi_surface protocol object
lifetime, and frees the ivi-id by destroying the ivi_layout_surface
from both wl_surface and ivi_surface destruction as the protocol specifies.
Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
AM_CFLAGS is the default for any target that doesn't specify its
own cflags. We should use AM_CFLAGS in preference to GCC_CFLAGS
so we can change AM_CFLAGS and get all targets.
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Jon A. Cruz <jonc@osg.samsung.com>
[pekka: fixed patch conflicts]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This is a follow-up for the patch that removed weston-screensaver. The
aim is to clean up shell.c a little by removing non-essential
components. Vanilla Weston desktop is only a demo, external projects are
encouraged to create user-friendly desktop environments.
The support for launching a screensaver client and the protocol bindings
are removed. With them, all related configuration options are removed,
and the manuals are updated accordingly.
The screensaver protocol definition is left in desktop-shell.xml for
posterity.
This does not affect Weston's or desktop-shells ability to put screens
to sleep after inactivity. The inactivity timer continues to operate as
before. Also screen locking is unaffected.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Giulio Camuffo <giuliocamuffo@gmail.com>
This removes the weston-screensaver client.
Screensavers are not so useful, DPMS is much better. This example has
existed here for a good while, and things that we could learn from it
have been learnt.
Nowadays this is just dead weigth, which is usually not even compiled,
because it depends on both cairo-gl and GLU. Removing it removes the
only possible dependency to GLU and one user of cairo-gl. Now the last
user of cairo-gl is gears (clients/nested.c uses cairo-glesv2).
Support for screensavers is still left in desktop-shell, so external
projects can still have their screensavers if they want.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Giulio Camuffo <giuliocamuffo@gmail.com>
The headless-backend.so was missing in available backend list
Signed-off-by: JoonCheol Park <jooncheol@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Help messages were missing for some command line options.
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
We don't want to use installed binaries when running tests.
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
EGLGetDisplay() doesn't generate a GL error, so we shouldn't print one.
I've also renamed the goto labels so it's a little clearer when to use them.
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Removed duplicate definitions of the container_of() macro and
refactored sources to use the single implementation.
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Removed multiple definitions of the MIN() macro from existing
locations and unified with a single definition. Updated sources
to use the shared version.
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
To help reduce code duplication and also 'kitchen-sink' includes
the ARRAY_LENGTH macro was moved to a stand-alone file and
referenced from the sources consuming it. Other macros will be
added in subsequent passes.
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Using the parent '../' path component in #include statements makes
the codebase more rigid and is redundant due to proper -I use.
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
