aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-08-24xwayland: fix some shutdown casesDominique Martinet
2017-08-24xwayland: fix shutdown caused by Xwayland/client stopDominique Martinet
2017-08-23xwm: remove xcb-util dependencyDominique Martinet
This hardcodes a define, but other projects (wlc/weston) either use the raw number or just redefine it with another name anyway... This should fix travis build.
2017-08-23xwayland: fix style issuesDominique Martinet
2017-08-23xwm: reply to configure_requestsDominique Martinet
2017-08-23example compositor: loop through xwayland surfacesDominique Martinet
2017-08-23xwm: handle some more eventsDominique Martinet
handle map/unmap_notify, property_notify (just logging for now), and client_message for wl_surface_id. We almost can display an X window now, just need the compositor to iterate over them
2017-08-23wlr surface/shell types: add message on creationDominique Martinet
Not sure if we really want to keep that, would be helpful to get more logging levels at some point
2017-08-23xwayland: move & split internal header fileDominique Martinet
2017-08-23xwm: fix minimal init, this gets us some events on window creationsDominique Martinet
2017-08-23init xwm only once Xwayland server is readyDominique Martinet
2017-08-23wlr_compositor: add signal for create_surfaceDominique Martinet
2017-08-23xwayland: split xwm structure out, initial xwm.cDominique Martinet
2017-08-23start xwayland in example compositorDominique Martinet
2017-08-23Xwayland: first draft, just start server for nowDominique Martinet
2017-08-22Merge pull request #121 from martinetd/wlr_seat_destroyDrew DeVault
wlr_seat: fix destroy with multiple handles
2017-08-21wlr_seat: fix destroy with multiple handlesDominique Martinet
Need to use for_each_safe because wl_resource_destroy's callback will remove the handle from the list itself
2017-08-20Merge pull request #118 from martinetd/delay_handle_keyboard_cbDrew DeVault
example compositor: handle keyboard after we're done with init
2017-08-20Merge pull request #117 from ascent12/meson-optionsDrew DeVault
Change meson to be more configurable
2017-08-20example compositor: handle keyboard after we're done with initDominique Martinet
the libinput backend does wl_display roundtrips, during which there is a small window where keystrokes can be handled before the rest of the example compositor is ready. Setting the callback later ensures we're not called at this point
2017-08-20Change meson to be more configurableScott Anderson
We move the warning options into the project's default_options, meaning a user can configure them with `meson configure` and not need to edit the file. We also make it possible to disable logind/libcap even if they're present.
2017-08-19Merge pull request #115 from martinetd/move_wl_compositorDrew DeVault
move wl_compositor into wlroots as wlr_compositor
2017-08-19wlr_compositor: change _init/finish to _create/destroyDominique Martinet
2017-08-19move wl_compositor into wlroots as wlr_compositorDominique Martinet
2017-08-19Merge pull request #114 from martinetd/cleanupsDrew DeVault
Cleanups
2017-08-19wlr_surface: change update_size/damage to boolDominique Martinet
2017-08-19wayland backend: wl_event_source_remove on destroyDominique Martinet
2017-08-19examples: separate compositor_fini from runDominique Martinet
compositor_fini destroys the display, but it is an error to destroy it before e.g. wlr_seat that references it. This lets us order destroy calls properly, following first-in-last-out logic.
2017-08-19wlr_surface: small refactoringDominique Martinet
- split surface damage and buffer damage - the pending buffers are always valid, so can be 'fini'shed - consider opaque buffer and input buffer in commit, just to free - only update size if scale/transform/buffer changed
2017-08-19wlr region/surface: explicitely select interface fieldsDominique Martinet
2017-08-19wlr_surface: clear/fini buffers when appropriateDominique Martinet
2017-08-19wlr renderer/texture: rename init to create when it does allocDominique Martinet
2017-08-19libinput: wl_event_source_remove on destroyDominique Martinet
2017-08-19wayland backend: remove explicit scale = 1Dominique Martinet
This is now set correctly in the wlr_output init
2017-08-18Merge pull request #113 from acrisci/feature/rename-wlr-cursorDrew DeVault
rename wlr_cursor to wlr_xcursor
2017-08-18rename wlr_cursor.c to wlr_xcursor.cTony Crisci
2017-08-18rename wlr_cursor to wlr_xcursorTony Crisci
This is for the implementation of another type that should be called wlr_cursor.
2017-08-18Merge pull request #110 from acrisci/feature/refactor-example-configDrew DeVault
Refactor example output config
2017-08-18bring output configuration into shared.hTony Crisci
2017-08-18Refactor example output configTony Crisci
Put all the config parsing into shared.h so it is shared among the examples.
2017-08-18Merge pull request #107 from nyorain/wlr_data_device_managerDrew DeVault
Implement wlr_data_{source,device,device_manager}
2017-08-18Fix wlr_data_device_managernyorain
2017-08-18Merge pull request #109 from martinetd/post_eventDrew DeVault
wlr_surface: use post_event instead of queue
2017-08-18wlr_surface: use post_event instead of queueDominique Martinet
This fixes not sending anything back to the client if the only event it's waiting for is a buffer release, and nothing else is in the pipe. Workaround for #102 - there probably is a better solution as weston and wlc do not need to use post_event here.
2017-08-18Merge pull request #108 from martinetd/realloc_crtcs_indexDrew DeVault
drm realloc_crtcs: check we found an output
2017-08-18Merge pull request #106 from acrisci/bug/fix-wild-bufferDrew DeVault
Fix wild pointer to buffer
2017-08-18Fix data_device destructionnyorain
2017-08-18Implement wlr_data_{source,device,device_manager}nyorain
2017-08-18Fix wild pointer to bufferTony Crisci
After the buffer is released from the release event, it should no longer be accessed. Fixes a GTK crash.
2017-08-18drm realloc_crtcs: check we found an outputDominique Martinet
This follows-up on #105 that just initialized to 0. Instead, make sure we actually found an output that matches