Our layer shell implementation assigns every layer surface to an output
on creation. It tracks this output using the output field on the
underlying wlr_layer_surface_v1 structure. As such, much of the existing
code assumes that output is always non-NULL and omits NULL checks
accordingly.
However, there are currently two cases where we destroy a
sway_layer_surface and output is NULL. The first is when we can't find
an output to assign the surface to and destroy it immediately after
creation. The second is when we destroy a surface in response to its
output getting destroyed, as we set output to NULL in
handle_output_destroy() before we call wlr_layer_surface_v1_destroy(),
which is what calls the appropriate unmap and destroy callbacks.
The former case doesn't cause any problems, since we haven't even
allocated a sway_layer_surface at that point or registered any
callbacks. The latter case, however, currently triggers a crash (#6120)
if a popup is visible, since our popup_handle_unmap() implementation
can't handle a NULL output.
To fix this issue, keep output set until right before we free the
sway_layer_surface. All we need to do is remove some of the cleanup
logic from handle_output_destroy(), since as of commit c9060bcc12
("layer-shell: replace close() with destroy()") that same logic is
guaranteed to be happen later when wlroots calls handle_destroy() as
part of wlr_layer_surface_v1_destroy().
This lets us remove some NULL checks from other unmap/destroy callbacks,
which is nice. We also don't need to check that the wlr_output points to
a valid sway_output anymore, since we unset that pointer after disabling
the output as of commit a0bbe67076 ("Address emersions comments on
output re-enabling") Just to be safe, I've added assertions that the
wlr_output is non-NULL wherever we use it.
Fixes#6120.
This commit makes sure the extents are kept up-to-date, fixes not
damaging the surface if its layer shell-specific state didn't change,
and adds a check if the layer shell-specific state didn't change but the
surface got mapped/unmapped, which could affect keyboard focus.
When a layer surface shrinks we need to damage the area it previously
occupied, but we don't know the location of all its subsurfaces in the
previous state, so instead damage a rectangle that encloses the entire
previous extent.
With these changes, sway will respect positive exclusive zones of layer
surfaces anchored to one or three sides.
This matches the protocol, which states that a positive exclusive zone
should be respected, "if the surface is anchored to one edge or an
edge and both perpendicular edges". If the surfaces is "anchored to
only two perpendicular edges (a corner), anchored to only two
parallel edges or anchored to all edges a positive value will be
treated the same as zero".
This fixes the following crash:
#0 0x00007f7daac3af25 in raise () at /usr/lib/libc.so.6
#1 0x00007f7daac24897 in abort () at /usr/lib/libc.so.6
#2 0x00007f7daac24767 in _nl_load_domain.cold () at /usr/lib/libc.so.6
#3 0x00007f7daac33526 in () at /usr/lib/libc.so.6
#4 0x0000555bfbc35029 in seat_set_focus_layer (seat=0x555bfd76d860, layer=0x555bfdda2ff0)
at ../sway/input/seat.c:1164
__PRETTY_FUNCTION__ = "seat_set_focus_layer"
__PRETTY_FUNCTION__ = "seat_set_focus_layer"
#5 0x0000555bfbc35029 in seat_set_focus_layer (seat=seat@entry=0x555bfd76d860, layer=0x555bfdda2ff0)
at ../sway/input/seat.c:1164
__PRETTY_FUNCTION__ = "seat_set_focus_layer"
#6 0x0000555bfbc25899 in handle_output_destroy (listener=0x555bfdb90688, data=<optimized out>)
at ../sway/desktop/layer_shell.c:263
layer = 0x555bfdd6b040
sway_layer = 0x555bfdb90610
seat = 0x555bfd76d860
client = 0x555bfdb76d70
set_focus = <optimized out>
#7 0x0000555bfbc5b669 in wl_signal_emit (data=0x555bfd795930, signal=0x555bfd795ae0)
at /usr/include/wayland-server-core.h:472
l = <optimized out>
next = 0x555bfdb6a3e8
__PRETTY_FUNCTION__ = "output_disable"
index = <optimized out>
#8 0x0000555bfbc5b669 in output_disable (output=output@entry=0x555bfd795930)
at ../sway/tree/output.c:263
__PRETTY_FUNCTION__ = "output_disable"
index = <optimized out>
#9 0x0000555bfbc3b890 in apply_output_config (oc=0x555bfd7d73d0, output=output@entry=0x555bfd795930)
at ../sway/config/output.c:321
wlr_output = 0x555bfd7afaf0
scale = <optimized out>
output_box = <optimized out>
#10 0x0000555bfbc28309 in handle_output_manager_apply
(listener=0x555bfbc7f148 <server+488>, data=0x555bfdca6eb0) at ../sway/desktop/output.c:936
wlr_output = <optimized out>
output = 0x555bfd795930
oc = <optimized out>
server = 0x555bfbc7ef60 <server>
config = 0x555bfdca6eb0
config_head = 0x555bfdb79350
ok = true
#11 0x00007f7dab4fbf7c in wlr_signal_emit_safe (signal=<optimized out>, data=0x555bfdca6eb0)
at ../subprojects/wlroots/util/signal.c:29
pos = 0x555bfbc7f148 <server+488>
l = 0x555bfbc7f148 <server+488>
cursor =
{link = {prev = 0x555bfbc7f148 <server+488>, next = 0x7fff238a8390}, notify = 0x7f7dab4fbef0 <handle_noop>}
end =
{link = {prev = 0x7fff238a8370, next = 0x555bfd7419f8}, notify = 0x7f7dab4fbef0 <handle_noop>}
#12 0x00007f7daa45469a in ffi_call_unix64 () at /usr/lib/libffi.so.6
#13 0x00007f7daa453fb6 in ffi_call () at /usr/lib/libffi.so.6
#14 0x00007f7daae6f82f in () at /usr/lib/libwayland-server.so.0
#15 0x00007f7daae6c193 in () at /usr/lib/libwayland-server.so.0
#16 0x00007f7daae6d7f2 in wl_event_loop_dispatch () at /usr/lib/libwayland-server.so.0
#17 0x00007f7daae6c39c in wl_display_run () at /usr/lib/libwayland-server.so.0
This crash happens because focus can only be set on mapped surfaces.
When arranging layer-shell layers, verify that the currently focused
layer, if any, for each seat is still keyboard interactive. If the layer
is no longer keyboard interactive and there is not a keyboard
interactive overlay or top layer to change the focus to, refocus the
focus inactive node for the seat.
If a layer is focused by any seat, it needs to be unfocused on unmap. If
the unmap was due to an output being disabled, there would not be a
sway_output and unmap would do an early return. This results in a
use-after-free if the layer was focused by any seat prior to being
unmapped. This change just moves the refocusing code above the early
returns.
Popups are positioned relative to local surface coordinates of the
parent surface. There's no need to consider values set with
xdg_surface.set_window_geometry for parent surfaces.
Focused layers are not cleared when destroyed, they are cleared on unmap.
Giving focus to an unmapped layer surface is (1) incorrect and (2) triggers a
use-after-free.
Closes: https://github.com/swaywm/sway/issues/4517
The documentation for wayland-server.h says:
> Use of this header file is discouraged. Prefer including
> wayland-server-core.h instead, which does not include the server protocol
> header and as such only defines the library PI, excluding the deprecated API
> below.
Replacing wayland-server.h with wayland-server-core.h allows us to drop the
WL_HIDE_DEPRECATED declaration.
This commit si similar to wlroots' ca45f4490ccc ("Remove all wayland-server.h
includes").
If the noop output is focused (all other outputs disabled/disconnected),
do not auto assign a layer surface to it. The noop output is not enabled
and does not have the `output->layers` list initialized. It also does
not make sense to map the layer surfaces to something that is not
visible.
In addition to removing unused code, two minor problems are fixed:
(1) `resize set` and `resize adjust` did not error when given
too many arguments.
(2) `orientation` was incorrectly overridden to be 'U' for
scroll events in the swaybar tray `handle_click` function.
This commit mostly duplicates the wlr_log functions, although
with a sway_* prefix. (This is very similar to PR #2009.)
However, the logging function no longer needs to be replaceable,
so sway_log_init's second argument is used to set the exit
callback for sway_abort.
wlr_log_init is still invoked in sway/main.c
This commit makes it easier to remove the wlroots dependency for
the helper programs swaymsg, swaybg, swaybar, and swaynag.
The input manager is a singleton object. Passing the sway_input_manager
argument to each of its functions is unnecessary, while removing the
argument makes it obvious to the caller that it's a singleton. This
patch removes the argument and makes the input manager use server.input
instead.
On a similar note:
* sway_input_manager.server is removed in favour of using the server
global.
* seat.input is removed because it can get it from server.input.
Due to a circular dependency, creating seat0 is now done directly in
server_init rather than in input_manager_create. This is because
creating seats must be done after server.input is set.
Lastly, it now stores the default seat name using a constant and removes
a second reference to seat0 (in input_manager_get_default_seat).
* Have multiple outputs
* Launch swaylock
* Unplug an output (possibly has to be the last "connected" one)
* The swaylock surface on the remaining output would not respond to key
events
This was happening because when the output destroys, focus was not given
to the other swaylock surface.
This patch makes focus be transferred to another surface owned by the
same Wayland client, but only if input was inhibited by the surface
being destroyed, and only if it's in the overlay layer. I figure it's
best to be overly specific and relax the requirements later if needed.
This patch removes a check in seat_set_focus_surface which was
preventing focus from being passed from a layer surface to any other
surface. I don't know of a use case for this check, but it's possible
that this change could produce issues.
This commit changes the meaning of sway_container so that it only refers
to layout containers and view containers. Workspaces, outputs and the
root are no longer known as containers. Instead, root, outputs,
workspaces and containers are all a type of node, and containers come in
two types: layout containers and view containers.
In addition to the above, this implements type safe variables. This
means we use specific types such as sway_output and sway_workspace
instead of generic containers or nodes. However, it's worth noting that
in a few places places (eg. seat focus and transactions) referring to
them in a generic way is unavoidable which is why we still use nodes in
some places.
If you want a TL;DR, look at node.h, as well as the struct definitions
for root, output, workspace and container. Note that sway_output now
contains a workspaces list, and workspaces now contain a tiling and
floating list, and containers now contain a pointer back to the
workspace.
There are now functions for seat_get_focused_workspace and
seat_get_focused_container. The latter will return NULL if a workspace
itself is focused. Most other seat functions like seat_get_focus and
seat_set_focus now accept and return nodes.
In the config->handler_context struct, current_container has been
replaced with three pointers: node, container and workspace. node is the
same as what current_container was, while workspace is the workspace
that the node resides on and container is the actual container, which
may be NULL if a workspace itself is focused.
The global root_container variable has been replaced with one simply
called root, which is a pointer to the sway_root instance.
The way outputs are created, enabled, disabled and destroyed has
changed. Previously we'd wrap the sway_output in a container when it is
enabled, but as we don't have containers any more it needs a different
approach. The output_create and output_destroy functions previously
created/destroyed the container, but now they create/destroy the
sway_output. There is a new function output_disable to disable an output
without destroying it.
Containers have a new view property. If this is populated then the
container is a view container, otherwise it's a layout container. Like
before, this property is immutable for the life of the container.
Containers have both a `sway_container *parent` and
`sway_workspace *workspace`. As we use specific types now, parent cannot
point to a workspace so it'll be NULL for containers which are direct
children of the workspace. The workspace property is set for all
containers, except those which are hidden in the scratchpad as they have
no workspace.
In some cases we need to refer to workspaces in a container-like way.
For example, workspaces have layout and children, but when using
specific types this makes it difficult. Likewise, it's difficult for a
container to get its parent's layout when the parent could be another
container or a workspace. To make it easier, some helper functions have
been created: container_parent_layout and container_get_siblings.
container_remove_child has been renamed to container_detach and
container_replace_child has been renamed to container_replace.
`container_handle_fullscreen_reparent(con, old_parent)` has had the
old_parent removed. We now unfullscreen the workspace when detaching the
container, so this function is simplified and only needs one argument
now.
container_notify_subtree_changed has been renamed to
container_update_representation. This is more descriptive of its
purpose. I also wanted to be able to call it with whatever container was
changed rather than the container's parent, which makes bubbling up to
the workspace easier.
There are now state structs per node thing. ie. sway_output_state,
sway_workspace_state and sway_container_state.
The focus, move and layout commands have been completely refactored to
work with the specific types. I considered making these a separate PR,
but I'd be backporting my changes only to replace them again, and it's
easier just to test everything at once.
This moves the arrange_windows call into the arrange_layers function,
where we know the output actually needs to be arranged.
Additionally, we shouldn't set focus to the parent of an unknown
container type, because the parent may be an output and this causes a
crash because outputs can't have direct focus.
Fixes#2543
When we have type safety we'll need to have functions for
workspace_add_tiling and so on. This means the existing container
functions will be just for containers, so they are being moved to
container.c. At this point layout.c doesn't contain much else, so I've
relocated everything and removed the file.
* container_swap and its static functions have been moved to the swap
command and made static.
* container_recursive_resize has been moved to the resize command and
made static.
* The following have been moved to container.c:
* container_handle_fullscreen_reparent
* container_insert_child
* container_add_sibling
* container_add_child
* container_remove_child
* container_replace_child
* container_split
* enum movement_direction and sway_dir_to_wlr have been moved to util.c.
Side note: Several commands included layout.h which then included
root.h. With layout.h gone, root.h has to be included by those commands.
This PR changes the way we handle transactions to a more simple method.
The new method is to mark containers as dirty from low level code
(eg. arranging, or container_destroy, and eventually seat_set_focus),
then call transaction_commit_dirty which picks up those containers and
runs them through a transaction. The old methods of using transactions
(arrange_and_commit, or creating one manually) are now no longer
possible.
The highest-level code (execute_command and view implementation
handlers) will call transaction_commit_dirty, so most other code just
needs to set containers as dirty. This is done by arranging, but can
also be done by calling container_set_dirty.
