Summary:
ecore_event_window_unregister(ee->prop.window) is actually *exactly*
the same as ecore_evas_input_event_unregister(ee)
So this sequence just uselessly tries to remove something from an empty
hash table.
Reviewers: devilhorns
Reviewed By: devilhorns
Subscribers: cedric, #committers, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6521
Summary:
A reference was taken on these when they were added, so by passing NULL
here we're not allowing that reference to be dropped. This results in
not only leaking the device forever, but also potentially leaving it as
a default device after it's removed so that no new attach will replace it.
Under weston all devices are removed on a VT switch, and when you switch
back the default device is wrong, which leads to problems with events
like "mouse in" which, for legacy reasons, don't take a seat, but
instead look up the default seat.
This allows the delete callback to be fired for the first time ever,
and while I've tried to fix some bugs this has revealed, I'm not
actually sure I've caught them all.
Depends on D6182
Reviewers: zmike
Reviewed By: zmike
Subscribers: cedric, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6183
Summary:
We were creating seats with silly arbitray names like seat-11 when
creating a new canvas.
Depends on D6130
Reviewers: zmike, cedric
Reviewed By: zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6131
If we don't ack_configure here, it'll just automatically happen in
ecore_wl2_commit() next time we render anyway.
If we do ack_configure here, we can totally screw up E's internal window
handling.
Removing it seems like a win.
Instead of having elm handle all this stuff, handle it in ee. This fixes
a loooong standing bug where we would always use an alpha enabled surface
because CSD needs shadows - even if we aren't drawing CSD.
Someone could NOT use elementary and use ecore_evas only. In this case, content
size which is defined by elementary is 0x0. If content size is 0x0, then frame
size is equal to window size. But the frame size is defined by elementary as
well. So if there is not a content, then the frame size should be 0.
It's possible that we unregister then reregister an animator so quickly
that the ecore_evas idle enter/exiters haven't had a chance to run yet.
In this case a render will come shortly anyway and we shouldn't use a
'false' commit to kick off a frame callback.
Silences an ERR and unmeasurably improves protocol utilization.
If we have a blacklisted gl implementation then we'll create a canvas,
it'll fail to be useful, and we'll free it - which calls
ecore_wl2_shutdown(), but then we'll also call ecore_wl2_shutdown()
a second time further down the failure path.
Take a bonus reference before freeing the failed canvas to keep refcounts
sane.
Summary: This patch replaces ecore_wl_dpi_get with ecore_wl2_output_dpi_get using Ecore_Wl2_Output
Test Plan: Execute test suite
Reviewers: cedric, raster, jpeg, stefan_schmidt, Jaehyun_Cho, devilhorns
Reviewed By: devilhorns
Differential Revision: https://phab.enlightenment.org/D5804
The same pointer is valid post recovery, and anything that used to
check this for NULL now checks that the window's display is NULL
which is handled in ecore_wl2 now.
This isn't meant to be installed. The canvas API in EO is based around
the interfaces Efl.Canvas and the widget Efl.Ui.Win. Anything else is
not EO (eg: ecore_evas, evas, ...)
Note: evas_canvas3d is the last remaining thing that is installed along
EO files, but those are all beta APIs.
Turns out when apps reconnect to the compositor they don't always
realize they need to redraw themselves. Force a manual render
at startup if we end up in a state where an update is needed but
has probably been dropped on the floor.
This is really several inseparable commits mashed together, as doing this
a piece at a time would introduce broken intermediate revisions.
Double buffer incoming "configure" state from the compositor so it's held
back during asynchronous render and processed at frame completion.
Hold off on certain requests if their API has been invoked during async
render.
This should fix a lot of races, cosmetic issues, issues where weston can
kill our clients for acking configure (or not) at bad times, etc.
with deferred surface creation the first canvas change of hints may not
be able to trigger protocol methods for size hints, so ensure that hints are
set
Small patch to add a handler for catching Window Iconify State Change
events
'#divergence'
@feature
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Derek Foreman
YeongJong Lee
Shinwoo Kim
Cedric Bail
Myoungwoon Roy, Kim
Mike Blumenkrantz
Jean-Philippe Andre
Carsten Haitzler
Chris Michael