With the recent universal plane and atomic modeset / nuclear pageflip
development in the kernel, cursor content updates on Intel are currently causing
an extra wait for vblank. This drops Weston's framerate to a fraction by
2 when cursor contents update. This combined with the damage tracking
bug in Weston which causes cursor content updates on every frame the
cursor moves makes using hw cursors really bad.

It is possible that the Intel DRM driver will get fixed and cursor
updates there revert to their old behaviour on the contemporary KMS API.
However, it is hardware dependant whether cursor updates can happen
immediately.  Some other hardware, especially ARM-related, may not be
able to do immediate updates. Therefore it is better to just not even
try - we should rely only on the lowest common denominator behaviour
between hardware and drivers as there is no and will not be any way to
reliably detect it.

Note, that while having different drivers do different things (immediate
update vs. update that gets latched on the next vblank), we cannot
rearrange the contemporary KMS API calls such that it would always work
fine. Either some hardware would update the cursor too early, or other
hardware would update the cursor too late and perhaps cause the
framerate decimation.

Mark hardware cursors broken by default. This avoids using them, and
works around the immediate problem of framerate issues in Weston. This
follows the same reasoning why hardware overlay planes have been
disabled by default for a long time.

This disablement will be removed once the current code for hardware
planes and cursors is replaced with code using the atomic KMS API.

The Intel driver change that exposed this problem is
https://github.com/torvalds/linux/commit/38f3ce3af5742eb5a3e9b01997f5ab85109c5762


which is first included in Linux 4.0-rc1.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Cc: nerdopolis <bluescreen_avenger@verizon.net>
Cc: Daniel Stone <daniel@fooishbar.org>
Cc: Giulio Camuffo <giuliocamuffo@gmail.com>
Reviewed-By: David FORT <contact@hardening-consulting.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Derek Foreman authored 9 years ago and Pekka Paalanen committed 9 years ago

If the GL implementation doesn't provide an XRGB visual we may still be
able to proceed with an ARGB one. Since we're not changing the scanout
buffer format, and our current rendering loop always results in saturated
alpha in the frame buffer, it should be Just Fine(tm) - and probably better
than just exiting.

This is a workaround for https://bugs.freedesktop.org/show_bug.cgi?id=89689



Reviewed-By: Bryce Harrington <bryce@osg.samsung.com>
Tested-By: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-By: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

Derek Foreman authored 9 years ago and Pekka Paalanen committed 9 years ago

Currently we pass either a single format or no formats to the gl renderer
create and output_create functions.  We extend this to any number of
formats so we can allow fallback formats if we don't get our first pick.

Reviewed-By: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-By: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

These always bite me, so while probably uncommon may as well document
them...

Remove trailing whitespaces because they are not needed and jumping to
the end of al ine should do just that and not jump to the whitespace.

Derek Foreman authored 9 years ago and Pekka Paalanen committed 9 years ago

Really, there's pretty much no time we'd ever want O_CLOEXEC unset,
as it will likely result in leaking fds to processes that aren't
interested in them or shouldn't have them.

This also removes the (now unused) code from weston_logind_open() that
could drop O_CLOEXEC.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Derek Foreman authored 9 years ago and Pekka Paalanen committed 9 years ago

You had one job...

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Acked-by: David Herrmann <dh.herrmann@gmail.com> [implicit from v1
comment]
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Enable all hints by default. This fixes the "Maximize" button in apps that
don't set any hints - e.g., xclock or Firefox. (There's still a problem, though:
"decorate" is sometimes treated as a boolean, sometimes as a bitmask.)

Handle MWM_DECOR_ALL correctly. It looks like it's supposed to invert the values
of the rest of the flags.

Signed-off-by: Dima Ryazanov <dima@gmail.com>
Reviewed-By: Giulio Camuffo <giuliocamuffo@gmail.com>

This patch fixes the problem reported on the mailing list
(http://lists.freedesktop.org/archives/wayland-devel/2015-January/019575.html

).
All certificate and key paths were not copied when given to FreeRDP, so they
were freed when the peer was disconnecting. And so the next connection was failing.

All the initialization stuffs have been moved to the activate callback, as when it is
called the peer is ready for graphics.

We also differ the creation of the seat, so that a seat is initialized only the
peer really do the activation sequence. That helps when mstsc just connects to see
the certificate, ask if the certificate should be trusted, and then reconnects.

This patch also adds configuration settings for recent versions of FreeRDP that
comes with everything disabled. This makes remoteFx functionnal again.

The patch also handles the skipCompression flag for last FreeRDP versions, that
allows to skip bulk compression for surfaces that have been already compressed by
the remoteFx or NS codec.

This also fixes the compilation against FreeRDP master with callback that now return
BOOL.

Signed-off-by: Hardening <rdp.effort@gmail.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

To make it more readable, add an empty line between each request and
event.

Also comes with a bonus indentation fix.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

commit 78d00e45 renamed text_model to text_input

This cleans up remaining uses of the word "model"

Reviewed-by: Jan Arne Petersen <janarne@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

Multi-seat configurations currently break the text-backend, crashing
weston.  This is an attempt to clean up any crashes and have somewhat
sensible input panel behavior with multi-seat.

Store a link to the manager that created a text_input, use this to
ensure that only a single panel gets popped up at a time, since there
is only one manager.

Replace deactivate_text_input with deactivate_input_method: multiple
input methods may focus the same text_input, so deactivating a text_input
is weird in multi-seat and confusing to perform.

In destroy_input_method_context set the context's input_method's context
pointer to NULL to prevent a dangling pointer.

Reviewed-by: Jan Arne Petersen <janarne@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

Skype's popup notifications use this type.

Weston has sent the first 'resizing' configure event with width=height=0.
But resizing to that size doesn't make sense.
Instead, we now send the current width and height of the surface at the beginning
of resizing.

Signed-off-by: Ryo Munakata <ryomnktml@gmail.com>

The compositor's output_created signal used to be sent in weston_output_init()
which the backend call before putting the output in the output_list.
This caused problems when creating a new view in a listener to that signal,
because weston_view_assign_output() doesn't yet know the new output exists.
To fix this add a new weston_composito_add_output() func which adds the
output in the list and later sends the signal, and make the backends call
that.

weston-editor is the only stock client spawning the virtual
keyboard ; which means it may be the only client able to
obtain some special characters (depending on the user's
keyboard layout).

If we implement Cut, Copy and Paste, the user has now a way
to copy such characters to other useful clients (such as
weston-terminal). Plus, it demonstrates text data exchange
between two clients of different nature.

Functionality is implemented in a right-click menu and the
Ctrl+Shift+X/C/V bindings, just as in weston-terminal.

Signed-off-by: Manuel Bachmann <manuel.bachmann@open.eurogiciel.org>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Tested-by: Bryce Harrington <bryce@osg.samsung.com>

We cannot rely on the client to provide a surface filling the output so
we must specify what happens with the outputs area that is not covered
completely.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Some times the compositor needs to send a configure request but without
having any clue about what size the surface should have. Examples
include unmaximizing a surface that was mapped as maximized, or an
initial state which doesn't have any size expectations.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Mention set_window_geometry in configure documentation.

Add a strategic "For instance" to clarify what is just an example.

Clarify that the arguments of set_window_geometry are in the surface
local coordinate space.

Point out that the client needs to destroy a dismissed popup.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Require the client to have attached (either previously committed, or
newly) a buffer to the corresponding wl_surface, and that the window
will not be potentially mapped until calling wl_surface.commit after
having created the window. This is required to make valid double
buffered xdg_surface state possible when creating a window.

Currently there is no double buffered state in xdg_popup, but it should
behave the same as xdg_surface, and for making it future proof in case
we want to add double buffered state to xdg_popup.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

They are errors that may be as a result of calling get_xdg_popup on an
xdg_shell, not a result of calling a request on xdg_popup.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Require all child objects to be destroyed before the parent. In other
words, all popups and surfaces created by one xdg_shell instance needs
to be destroyed before the xdg_shell object, otherwise a protocol error
is raised.

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

Giulio Camuffo authored 9 years ago and Daniel Stone committed 9 years ago

As we do for the input interfaces such as wl_pointer, we must send the
selection event to all the wl_data_device resources the client created for
a specified seat.

Reviewed-by: Daniel Stone <daniels@collabora.com>

Giulio Camuffo authored 9 years ago and Daniel Stone committed 9 years ago

This commit adds a new exported function, weston_seat_send_selection(),
which sends the current selection to a specified client. This is
useful e.g. to implement a clipboard manager as a special client.

Reviewed-by: Daniel Stone <daniels@collabora.com>

Giulio Camuffo authored 9 years ago and Daniel Stone committed 9 years ago

Reviewed-by: Daniel Stone <daniels@collabora.com>

Derek Foreman authored 9 years ago and Daniel Stone committed 9 years ago

The other set_focus() functions take the relevant type instead of a seat
already, so this is consistent.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Pass 'this' weston_surface as the data argument to
weston_surface::destroy_signal listeners. The old &surface->resource was
really just an offsetted pointer to the weston_surface anyway. And,
because 'resource' happened to be the first member in struct weston_surface,
it was actually 'this' weston_surface.

The argument type was accidentally changed in commit
26ed73ce from wl_resource* to
wl_resource**.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

There is no valid case, where you would actually destroy a
weston_surface, while leaving the wl_surface protocol object in
existence. Therefore, inert wl_surface objects do not exist, except
because of bugs.

To catch such bugs, check that the resource is really NULL before
actually destroying the weston_surface.

We actually used to have this check, but it was removed by:

	commit 9dadfb53
	Author: Kristian Høgsberg <krh@bitplanet.net>
	Date:   Mon Jul 8 13:49:36 2013 -0400

	    compositor: Eliminate marshalling warning for leave events

However, the invariant was put back in:

	commit 0d379744
	Author: Giulio Camuffo <giuliocamuffo@gmail.com>
	Date:   Fri Nov 15 22:06:15 2013 +0100

	    compositor: set weston_surface:resource to NULL when destroyed

So apparently the issue fixed by 9dadfb53 was fixed another way later.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Nobuhiko Tanibata authored 9 years ago and Pekka Paalanen committed 9 years ago

Especially, the size of destination rectagle is set to (1,1).
This is because the size will be used for caluculating scale and not to
be 0 to avoid 0 dividing.

I also remark this as FIXME. This shall be fixed at ivi-layout-tansition.c.
In new invoded application, its property is initilized by (0,0)
destination rectangle. So transition fucntion always calculates its scale
as inf at first frame of fade-in with new invoked application. To fix this,
restructing transition function is ideally needed.

Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Nobuhiko Tanibata authored 9 years ago and Pekka Paalanen committed 9 years ago

This method should not update weston_view directly. This shall be done by
controller via ivi_layout_*.

Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Nobuhiko Tanibata authored 9 years ago and Pekka Paalanen committed 9 years ago

When application changes the size of its content, UI shall fit the source
rectangle, view area, to the size of its content to show whole content.

Signed-off-by: Nobuhiko Tanibata <nobuhiko_tanibata@xddp.denso.co.jp>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>