Commit graph

4516 commits

Author SHA1 Message Date
Kirill Primak
904d256581 layer-shell: don't try to unmap on destroy
A surface is guaranteed to be unmapped on destruction.
2024-01-17 16:54:37 +01:00
Simon Ser
c5fd8c050f Mark DRM lease protocol privileged
Allowing sandboxed clients to request DRM leases has security
implications.
2024-01-08 11:17:26 -05:00
Simon Ser
7e69a7076f Drop wl_drm
See https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4397
2024-01-04 08:21:33 +01:00
Simon Ser
95265fba59 input: reconfigure send_events on output hotplug
Closes: https://github.com/swaywm/sway/issues/7890
2024-01-02 14:07:35 +01:00
Billli11
0aceff7469
Remove wlr_presentation in sway_server struct
It is no longer in use.
2023-12-28 11:28:28 +01:00
Bill Li
64d644f0da Chase wlroots!4482 2023-12-27 20:57:35 +01:00
Simon Ser
22d0dd8bde Check wlr_pointer_constraint_v1_state.cursor_hint.enabled
Update for a wlroots breaking change which resets the committed
mask.

References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4478
2023-12-25 11:57:30 +01:00
Alexander Orzechowski
bbabb9aae8 output: Destroy when output layout is destroyed
Since output layout is destroyed when the wayland display is destroyed
we run into a destroy listener order problem: Either the display starts
destroying the outputs first, in which case we're good: The existing
handling will clean up. However, things go wrong if the display decides
to destroy the output layout first. In this case, sway will hold
invalid references to the output layout as part of each output so that
when it finally goes to destroy them, sway will dereference destroyed
output layout bits.

Ref: https://github.com/swaywm/sway/pull/6844#issuecomment-1843599513
2023-12-13 18:10:03 +01:00
Simon Ser
c6edbb7e5a input/seat: simplify seat_is_input_allowed()
Use an early return to make the code more readable.
2023-12-13 10:11:35 +01:00
Simon Ser
e8a0205607 input/seat: rename seat_set_exclusive_client() 2023-12-13 10:11:35 +01:00
Simon Ser
607b8aed0c input/seat: inline seat_set_exclusive_client() with NULL client 2023-12-13 10:11:35 +01:00
Simon Ser
7ad8c80bfe input/{keyboard,switch}: rename input_inhibited variable to locked
This is more descriptive now.
2023-12-13 10:11:35 +01:00
Simon Ser
dbd70faf2e input/seat: drop exclusive_client
This was a input-inhibit concept.
2023-12-13 10:11:35 +01:00
Simon Ser
a6ef12d968 Detect proprietary DisplayLink drivers
evdi is open-source, but is just some condom for their proprietary
user-space driver.
2023-12-12 10:04:14 -05:00
Simon Ser
ff07eab85b Detect Nvidia proprietary driver via drmGetVersion()
This is less punishing for users with the Nvidia driver loaded but
not used by Sway (e.g. for CUDA).
2023-12-12 10:04:14 -05:00
Simon Ser
b81c4da494 Drop fglrx detection
This ancient driver doesn't do KMS. So we were never able to run
with it anyways.
2023-12-12 10:04:14 -05:00
Tamino Bauknecht
255ff665c5 sway: raise error on non-accessible background file 2023-12-12 09:39:46 -05:00
Simon Ser
bf2b79b284 desktop/xwayland: correctly handle association on o-r change
When override-redirect changes, we need to setup/teardown listeners,
just like we do for map.

References: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3773
2023-12-06 23:24:44 +03:00
mrusme
ae3acf26f8 Update sway.5.scd
Make it extra extra extra clear, because I had to stumbled upon https://github.com/swaywm/sway/issues/3292 to understand this.
2023-12-04 16:30:20 +01:00
Alexander Orzechowski
f12023b1a2 Don't destroy output layout on exit
wlroots will destroy this object itself.
2023-12-01 09:11:20 +01:00
Sergei Trofimovich
2cd73a33c2 sway/config.c: use memcpy() for known buffer size
`gcc-14` added a new warning around dangerous use of `strncpy()` withi
known overflow:

    ../sway/config.c: In function 'do_var_replacement':
    ../sway/config.c:983:33: error: '__builtin___strncpy_chk' specified bound depends on the length of the source argument [-Werror=stringop-truncation]
      983 |                                 strncpy(newptr, var->value, vvlen);
          |                                 ^
    ../sway/config.c:971:45: note: length computed here
      971 |                                 int vvlen = strlen(var->value);
          |                                             ^~~~~~~~~~~~~~~~~~

It's a bit fishy to rely on truncating behaviour of `strncpy()`. The
change uses `memcpy()` as more explicit way to express copy of `vvlen`
bytes.
2023-11-29 10:30:17 +01:00
Kirill Primak
47e6a1164c xdg-shell: chase events update 2023-11-23 19:41:57 +01:00
Simon Ser
128b6253a9 Pass wl_display to wlr_output_layout
References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4310
2023-11-23 16:13:19 +03:00
Simon Ser
fd6d6f1d97 Add wlr/util/transform.h includes
References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4097
Closes: https://github.com/swaywm/sway/issues/7830
2023-11-23 16:13:19 +03:00
llyyr
a946b1aecf Chase wlroots!4440
References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4440
2023-11-22 00:42:55 +03:00
Simon Ser
9d666a08e1 Fix reference to wlr_output_event_commit.committed
This has been dropped from wlroots. Previous commit missed that.
2023-11-21 16:52:19 +01:00
Simon Ser
072fa60cb4 Add support for security-context-v1
As a first step, deny access to privileged protocols to sandboxed
apps.

References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3589
2023-11-21 15:12:42 +01:00
Simon Ser
5bdd608514 Apply gamma LUT when an output re-enabled
References: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3679
2023-11-21 15:05:58 +01:00
EBADBEEF
86f96a786b view: re-apply criteria when window gets unmapped
Remove any existing executed criteria items at unmap time. If a window
gets unmapped but not destroyed, we want to reapply 'for_window'
criteria. Fixes #6905.
2023-11-21 14:57:58 +01:00
NemuiBanila
c3ec141b34 seat: Fix cursor theme reload for tablet tool 2023-11-21 11:35:49 +01:00
Lina Banik
f7a40cfa2f seat: Fix reloading cursor theme after change
This reverts commit afde6369
"seat: avoid unneeded reloading xcursor theme".
Always avoiding to reload the xcursor theme prevents reloading the
cursor even when this is desired. Instead seat_configure_xcursor
can determine whether a full reload is necessary.

To stay with the spirit of the reverted change, cursors are only fully
reloaded, if the theme has changed.

Fixes #6931
2023-11-21 11:35:49 +01:00
Simon Ser
5c99b98805 input/input-manager: drop input_manager_configure_all_inputs
This is now unused.
2023-11-16 16:29:07 +01:00
Simon Ser
7036769bea Only reconfigure input mappings on output change
Fully reconfiguring all input devices on output change takes a
loooong time. Let's just reconfigure what we need: only mappings
depend on outputs.
2023-11-16 16:29:07 +01:00
Simon Ser
caa92ddee8 input/seat: rename seat_apply_input_config
seat_apply_input_mapping is a lot more descriptive.
2023-11-16 16:29:07 +01:00
Simon Ser
01dcad80e1 input/seat: don't configure mappings for switches
Switch devices cannot be mapped to an output/region, stop trying
to do so.
2023-11-16 16:29:07 +01:00
Bill Li
6f6b82793d chase wlroots!4411
References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4411
    fix #7802
2023-10-31 09:51:19 +01:00
Kirill Primak
f2425b4fd6 xdg-shell: send maximized if tiled isn't supported
wlroots doesn't do it automatically anymore.
2023-10-29 17:21:18 +01:00
Dudemanguy
647521244a xdg_shell: don't update wlr_toplevel if the container has no size yet
3d5ae9813d added logic to change the
underlying wlr_toplevel size for floating containers, but it does it
even if the container has no actual coordinates yet. This doesn't really
make sense to update the toplevel size in this case since there's many
things that could affect the initial coordinates (sway commands,
fullscreen state, etc.). Skip this by doing a crude check to see if the
current container state has any width.
2023-10-27 17:36:32 -07:00
Kirill Primak
47263aca28 view: check if the buffer was uploaded on save
wlr_surface_has_buffer() is insufficient; if a client has committed a
buffer but it couldn't be applied, NULL deref happens in the next line.
2023-10-24 14:35:57 +02:00
Tamino Bauknecht
0dfaf7ea63 config/output: Allow approximation of output refresh rate
Previous behavior was that only if resolution and refresh rate match
exactly, the mode was accepted. As fallback, the mode with the highest
refresh rate and the same resolution was chosen.

New behavior is that the mode with the closest match for the refresh
rate is used with a limit of up to 1Hz. The fallback behavior stays the same.

Additionally, the logging was made more verbose.
2023-10-23 11:05:08 +02:00
Alexander Orzechowski
4326a26ad6 Chase wlroots!4331 2023-10-06 14:39:12 +03:00
iff
8a8fb76ec1 Update sway-input.5.scd 2023-10-05 10:04:32 +02:00
Simon Ser
7cf4e1d5c6 Drop support for KDE's idle protocol
We support the standard idle-notify protocol since Sway 1.8.
2023-10-05 00:01:54 +02:00
Manuel Stoeckl
8edc48f6dd sway/server: Fix null pointer crash when Xwayland disabled 2023-09-10 08:33:10 +09:00
Johan Sköld
d952ce403e i3 compat: Adding support for the all criteria
Matches all views. i3 PR: https://github.com/i3/i3/pull/4460

Fixes #7110
2023-09-07 10:42:54 +09:00
iff
3dd2f4a67f
commands/input/map_from_region: don't treat 0x prefix as hex identifier
When using the `map_from_region` for pen tables, we will usually make
the available area as big as possible while maintaining the proportions
with the screen.

As most of the tablets uses a 16:10 ratios while the most popular screen
ratios is still 16:9, the argument for most people should be `0x0 1x0.9`
to have the maximum effective area.

However, the argument above won't work because the current code will
treat `0x...` as a hexadecimal number, instead of setting both `x` and
`y` to `0`.

This fix allows the use of the following syntax:

```
input type:tablet_tool {
	map_from_region 0x0 1x0.9
}
```
2023-09-07 01:31:35 +00:00
Dudemanguy
3d5ae9813d xdg_shell: update wlr_toplevel size on client resizes
If a floating client resizes itself, sway updates several of its
internal dimensions to match but not wlr_toplevel. This means that the
next time wlroots sends a toplevel configure event, it can have wrong
coordinates that resize the client back to its old size. To fix this,
let's just use wlr_xdg_toplevel_set_size so the wlr_toplevel has the
same dimensions as sway.

Exactly the same as 0183b9d35d0ce750588e700e536b7d2e367b0d0a but the
logic is onlly applied to xdg_shell and not xwayland.
2023-08-28 17:04:53 -07:00
Dudemanguy
60a94b4916 Revert "view: update wlr_toplevel size on client resizes"
This isn't the right fix for this issue because the xwayland code also
uses this function and updating the wlr_toplevel there doesn't make
sense and also causes problems. Fixes #7722.

This reverts commit bf44690ee8.
2023-08-28 17:04:53 -07:00
Dudemanguy
bf44690ee8 view: update wlr_toplevel size on client resizes
If a floating client resizes itself, sway updates several of its
internal dimensions to match but not wlr_toplevel. This means that the
next time wlroots sends a toplevel configure event, it can have wrong
coordinates that resize the client back to its old size. To fix this,
let's just use wlr_xdg_toplevel_set_size so the wlr_toplevel has the
same dimensions as sway. Fixes #5266.
2023-08-26 13:44:37 -07:00
Simon Ser
4a2210577c Hide xwayland_shell_v1 from regular clients
Regular clients are not allowed to use this interface. wlroots
already sends a protocol error if a non-Xwayland client tries to
use this interface, but let's remove all temptation by hiding it
completely.
2023-08-24 18:27:22 +02:00