2bd6fbf20e
Fix that major(st_rdev) have no meaning on FreeBSD
...
The major device number does not indicate the device type on FreeBSD,
and AFAIK the only way to differentiate between DRM, input, and other
devices is checking the fd path. This commit implements that.
The drmDropmaster and drmSetmaster calls are necessary, because the
implicit drop (that should occur when the DRM fd is closed) seems not
to be working in some scenarios (e.g. if you have a tmux session
running - maybe the fd is retained somehow by tmux?). This is a
problem, because once you exit the compositor, you can't start it (or
any other program that wants to be DRM master) again until you close
all your tmux sessions.
2018-08-30 11:15:39 +02:00
633663cdde
Use legacy modesetting for gamma control
...
It doesn't look like there's any motion on a fix from the AMDGPU side,
and using the legacy interface for this isn't so bad.
2018-08-29 23:05:38 +02:00
939a4ea60f
Merge pull request #1196 from dcz-purism/surface_move
...
rootston/surface_layers: fix clicking after surface moved
2018-08-28 22:15:50 -04:00
d11dac6a64
Merge pull request #1202 from alexbakker/destroy-signal-fix
...
Init the new destroy signals added by #1200
2018-08-27 12:41:10 -04:00
221d412824
Init the new destroy signals added by #1200
2018-08-27 18:21:36 +02:00
792f98c2f0
Merge pull request #1200 from alexbakker/destroy-signal
...
Add destroy signals to types that are destroyed by wl_display_destroy
2018-08-27 08:16:49 -04:00
20db29779e
Add destroy signals to types that are destroyed by wl_display_destroy
2018-08-26 23:23:12 +02:00
c6955fa89c
Merge pull request #1199 from apreiml/farthest_output
...
introduce wlr_output_layout_farthest_output
2018-08-25 11:00:25 -04:00
5617d2bd60
introduce wlr_output_layout_farthest_output
...
Similar to wlr_output_layout_adjacent_output but will return the
one that is the farthest away from the reference in given direction.
2018-08-25 09:14:55 +02:00
6025fcaa09
Update required meson version to 0.47.1
2018-08-24 09:20:25 -04:00
8ecf9224f5
Merge pull request #1198 from ascent12/meson_feature
...
Fix meson changes
2018-08-24 08:19:10 -04:00
dc37194291
Change how glgen.sh outputs files
...
This fixes building wlroots as a subproject.
2018-08-24 19:35:21 +12:00
fd3fa760d3
Revert "Revert "Merge pull request #1194 from ascent12/meson_feature""
...
This reverts commit 9c886f20b9
2018-08-24 19:35:02 +12:00
9c886f20b9
Revert "Merge pull request #1194 from ascent12/meson_feature"
...
This breaks wlroots when used as a meson subproject.
This reverts commit dea311992e6db9c4b746
2018-08-23 21:00:58 -04:00
dea311992e
Merge pull request #1194 from ascent12/meson_feature
...
Build system changes
2018-08-23 20:41:07 -04:00
f4ae9824f7
rootston/layer_shell: fix clicking after surface moved/resized
...
Layer surfaces are not notified of cursor position changes if the surface moves, only if the cursor moves. This workaround emits a cursor position event every time a cursor ends up over a newly resized layer surface to make sure the following clicks land in the right place.
This change doesn't address sending leave events when a cursor previously present over the surface becomes away.
There are 2 separate mechanisms in play, because a layer surface gets resized in 2 steps:
1. Layer surface resize & rearrange.
2. Underlying surface resize.
The first step may affect all layer surfaces. The cursor events are sent to cursors placed over all layer surfaces which have moved (not been resized). The second step affects any layer surface whose surface changed size. The cursor event is sent only to that surface.
Together, these events cover all surfaces: those which moves, and those which changed size, as long as each layer surface resize is accompanied by an immediate surface resize.
2018-08-23 11:29:03 +02:00
39ac21401f
Replace generator with custom_target
...
This stops the protocol header constantly being regenerated for every
target using them.
2018-08-23 14:21:41 +12:00
f41c411980
Use input file for configure_file
...
This is so we can potentially add comments to it, and so if a user looks
at the installed header, they can see the /* #undef WLR_HAS_FEATURE */
line to see every option, even if not available.
2018-08-23 14:19:59 +12:00
653ea512c2
Use current_source_dir instead of source_root
...
This is so that log messages are printed properly when wlroots is build
as a subproject.
2018-08-23 14:19:59 +12:00
b1ed61b649
Use pkg-config to find wayland-scanner
2018-08-23 14:19:56 +12:00
6db9c4b746
wlr_virtual_keyboard: use wlr_input_device_destroy
2018-08-22 14:43:29 +01:00
2929f0f8dd
wlr_input_device: fix keyboard destruction
...
After destroying a keyboard input device, seat's listeners could still be pointing to destroyed wlr_input_device signals. This patch makes sure the references are removed while the input device is being destroyed.
2018-08-22 14:43:29 +01:00
51892e0d74
Install headers explicitly
2018-08-22 21:16:16 +12:00
a2bbf2c1f7
Make style more consistent
2018-08-22 21:16:16 +12:00
869dca1bce
Remove unnecessary project arguments
...
Meson handles rpath properly, and we don't need to use -I ourselves by
using include_directories with '.'.
2018-08-22 21:16:16 +12:00
1358d80ab9
Add link dependency on symbols file
...
So the library is rebuilt properly when it's updated.
2018-08-22 21:16:16 +12:00
584a3f8b6e
Update examples to use new option
...
We use a dictionary to remove some code duplication.
2018-08-22 21:15:34 +12:00
560e96aa57
Update rootston to use new option
2018-08-22 20:41:25 +12:00
784c20c82f
Use new options for X11 backend and Xwayland
2018-08-22 20:27:37 +12:00
94ed2fc7bb
Use feature options for libcap/logind
...
We now use a combo choice between systemd/elogind as they are mutually
exclusive anyway.
2018-08-22 19:53:00 +12:00
3fa3dca6d5
Add new feature options
...
Bump minimum meson version to 0.47 for the feature.
2018-08-22 19:41:53 +12:00
66064a07a6
Remove unneeded mesa-specific header
2018-08-22 19:39:15 +12:00
d7c027d3e3
Merge pull request #1191 from emersion/drm-conn-dpi
...
backend/drm: add support for DRM_MODE_CONNECTOR_DPI
2018-08-21 08:16:02 -04:00
7a3f7b3c8a
backend/drm: add support for DRM_MODE_CONNECTOR_DPI
2018-08-21 13:05:38 +01:00
24212df830
rootston: fix again FreeBSD build
...
Had an outdated copy of wlroots, so missed that one.
2018-08-16 11:22:42 +01:00
56600804ff
Merge pull request #1189 from agx/ctags-f
...
ctags: use -f instead of -o
2018-08-16 11:11:08 +01:00
d66d33f5a2
Merge pull request #1188 from emersion/freebsd-fixes
...
Fix build on FreeBSD
2018-08-16 11:02:21 +01:00
570970db64
ctags: use -f instead of -o
...
The later does not exist on FreeBSD
See #725
2018-08-16 12:01:06 +02:00
f86f1daf9a
Fix build on FreeBSD
2018-08-16 10:54:45 +01:00
1a73baac94
Merge pull request #1187 from ammen99/master
...
move wlr_drm_connector_add_mode to wlr/backend/drm.h
2018-08-13 14:43:57 +01:00
4486b52aa8
move wlr_drm_connector_add_mode to wlr/backend/drm.h
2018-08-13 10:55:42 +03:00
4ed6ee0a4d
Merge pull request #1184 from sdilts/add-default-output-size
...
Keep X11 backend windows from having an invalid size
2018-08-11 08:41:29 +01:00
8e7df5eb88
Fix xcb_create_window parameters
...
Set the window width and height, not the location of the window.
2018-08-10 18:05:34 -06:00
11d440972d
rootston: Double fork for keyboard bindings
...
This avoids leaving around zombies, without having to setup SIGCHLD
handler (which interferes with other fork/waitpid calls).
2018-08-10 10:59:50 -04:00
da79fef5f7
Add guard for changing the size of X11 backend windows
...
Events that set the window to either a width or height of zero are now
ignored and logged.
2018-08-09 22:09:48 -06:00
55cca6deaa
Set default output size for X11 backend
...
Fixes issue #1170
Also set the created window size to match the output size.
2018-08-09 22:05:18 -06:00
d2814c1795
Merge pull request #1183 from ammen99/master
...
remove dangling declaration of wlr_xdg_surface_popup_get_position()
2018-08-09 12:56:49 +01:00
88e9dbe32e
remove dangling declaration of wlr_xdg_surface_popup_get_position()
2018-08-09 07:55:20 +03:00
ce0ab4d4b5
Merge pull request #1182 from ammen99/master
...
session: load GPU devices even if they have zero connectors
2018-08-09 08:37:05 +12:00
cb42e16f64
session: load GPU devices even if they have zero crtcs/connectors/encoders
...
On some systems (most notably laptops with two GPUs) there are GPUs that
don't have attached outputs. However, we still want to load those GPUs
because they could still be used by the compositor for rendering.
2018-08-08 07:34:12 +03:00
sghctoma
Drew DeVault
Brian Ashworth
Alexander Bakker
Armin Preiml
Scott Anderson
Dorota Czaplejewicz
emersion
Guido Günther
Ilia Bozhinov
sdilts
Genki Sky
Scott Anderson