Crash after plugging in USB graphics tablet multiple times #6836
Labels
bug
Not working as intended
Comments
|
Does the following diff fix the issue? diff --git a/sway/input/text_input.c b/sway/input/text_input.c
index b8c19c1..58911c2 100644
--- a/sway/input/text_input.c
+++ b/sway/input/text_input.c
@@ -77,8 +77,6 @@ static void handle_im_grab_keyboard(struct wl_listener *listener, void *data) {
struct wlr_keyboard *active_keyboard = wlr_seat_get_keyboard(relay->seat->wlr_seat);
wlr_input_method_keyboard_grab_v2_set_keyboard(keyboard_grab,
active_keyboard);
- wlr_input_method_keyboard_grab_v2_send_modifiers(keyboard_grab,
- &active_keyboard->modifiers);
wl_signal_add(&keyboard_grab->events.destroy,
&relay->input_method_keyboard_grab_destroy);I'm not sure why it's not crashing in EDIT: Right, the |
|
thanks @issue-dispenser , I compiled with that patch, and I'll get back you in a few days, given I don't have a way to trigger the crash quickly. |
|
@issue-dispenser sway has been running continuously with no crashes since I built it with this patch, seems promising! |
issue-dispenser
pushed a commit
to issue-dispenser/sway
that referenced
this issue
Mar 4, 2022
active_keyboard may be NULL, in which case an invalid pointer could be passed to wlr_input_method_keyboard_grab_v2_send_modifiers. This procedure call is unnecessary since wlroots commit 372a52ec "input method: send modifiers in set_keyboard", so the call can simply be removed. Fixes swaywm#6836.
emersion
pushed a commit
that referenced
this issue
Mar 4, 2022
active_keyboard may be NULL, in which case an invalid pointer could be passed to wlr_input_method_keyboard_grab_v2_send_modifiers. This procedure call is unnecessary since wlroots commit 372a52ec "input method: send modifiers in set_keyboard", so the call can simply be removed. Fixes #6836.
WillPower3309
referenced
this issue
in WillPower3309/swayfx
Jan 4, 2023
* build: bump wlroots dependency to 0.16.0 * swaymsg: replace if with switch in pretty_print * swaymsg: add GET_TREE pretty-printing * swaybar: fix errno handling in status_handle_readable If getline fails once, it was not reset before the next getline call. errno is only overwritten by getline on error. * Add cairo_image_surface_create error handling cairo_image_surface_create can fail, e.g. when running out of memory or when the size is too big. Avoid crashing in this case. Closes: swaywm/sway#6531 * build: bump version to 1.8-dev Historically we've been sticking with the last release number in the master branch. However that's a bit confusing, people can't easily figure out whether they're using a release or a work-in-progress snapshot. Only the commit hash appended to the version number may help, but that's not very explicit and disappears when using a tarball. We could bump the version in master to the next release number. However during the RC cycle there would be a downgrade from 1.8 to 1.8-rc1. Also it would be hard to tell the difference between a stable release and an old snapshot. This patch introduces a new pre-release identifier, "dev". It's alphabetically before "rc" so it should be correctly sorted by semver comparisons. "dev" is upgraded to "rc" (and then to stable) when doing a release. The master branch always uses a "dev" version, only release branches use "rc" or stable versions. * [IPC] Add repeat delay/rate info to keyboard Closes #6735 wlroots already has the info in the struct so let's access it and print it out. * input/seat: unset has_focus when focus_stack becomes empty We currently track the focus of a seat in two ways: we use a list called focus_stack to track the order in which nodes have been focused, with the first node representing what's currently focused, and we use a variable called has_focus to indicate whether anything has focus--i.e. whether we should actually treat that first node as focused at any given time. In a number of places, we treat has_focus as implying that a focused node exists. If it's true, we attempt to dereference the return value of seat_get_focus(), our helper function for getting the first node in focus_list, with no further checks. But this isn't quite correct with the current implementation of seat_get_focus(): not only does it return NULL when has_focus is false, it also returns NULL when focus_stack contains no items. In most cases, focus_stack never becomes empty and so this doesn't matter at all. Since focus_stack stores a history of focused nodes, we rarely remove nodes from it. The exception to this is when a node itself goes away. In that case, we call seat_node_destroy() to remove it from focus_stack and free it. But we don't unset has_focus if we've removed the final node! This lets us get into a state where has_focus is true but seat_get_focus() returns NULL, leading to a segfault when we try to dereference it. Fix the issue both by updating has_focus in seat_node_destroy() and by adding an assertion in seat_get_focus() that ensures focus_stack and has_focus are in sync, which will make it easier to track down similar issues in the future. Fixes #6395. [1] There's some discussion in #1585 from when this was implemented about whether has_focus is actually necessary; it's possible we could remove it entirely, but for the moment this is the architecture we have. * swaybar: fix tray_padding vs min-height re: scale Co-authored-by: xdavidwu <xdavidwuph@gmail.com> * swaybar: fix tray item icon scaling, positioning * container: Fix crash when view unmaps + maps quickly Followup on 4e4898e. If a view quickly maps and unmaps repeatedly, there will be multiple destroyed containers with same view in a single transaction. Each of these containers will then try to destroy this view, resulting in use after free. The container should only destroy the view if the view still belongs to the container. Simple reproducer: couple XMapWindow + XUnmapWindow in a loop followed by XDestroyWindow. See #6605 * commands/move: Fix crash when pos_y is omitted Fixes #6737 * Destroy sub-surfaces with parent layer-shell surface Closes: swaywm/sway#6337 * Add safety assert in parse_movement_unit Let's add this just in case a caller passes argc == 0. References: swaywm/sway#6737 (comment) * meson: check: false on run_command Future meson releases will change the default and warns when the implicit default is used, breaking builds. Explicitly set check: false to maintain behavior and silence warnings. * Print deprecation notice when running SUID SUID privilege drop is needed for the "builtin"-backend of libseat, which copied our old "direct" backend behavior for the sake of compatibility and ease of transition. libseat now has a better alternative in the form of seatd-launch. It uses the normal seatd daemon and libseat backend and takes care of SUID for us. Add a soft deprecation warning to highlight our future intent of removing this code. The deprecation cycle is needed to avoid surprises when sway no longer drops privileges. * xdg-shell: use toplevel geometry to adjust the popup box `popup_unconstrain` uses view coordinates to init the output box for popups. However wlroots expects the box to be set in a toplevel surface coordinate system, which is not always equal to view. The difference between those is a window geometry set via xdg-shell. GTK4 reserves some space for client-side decoration and thus has a window with top left corner not matching to (0, 0) of a surface. The box calculated without taking that into account was slightly shifted compared to the actual output and allowed to position part of the popup off screen. * build: fix building with basu 02b412a introduced the use of list for sdbus deps, however it was assuming that all packages which were in a list has a version higher than 239. That is true for libsystemd and libelogind, since they use the same versions, however basu is using version numbers which are way lower than what libsystemd/libelogind are using, so basu only build is failing. * Upgrade for wlroots surface refactoring See [1] for details. [1]: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3412 * commands/focus: drop trailing whitespace * input/cursor: count pointer gestures as idle activity Fixes swaywm/sway#6765. * input/cursor: treat swipe begin as idle activity too Accidentally overlooked in fd53f80. * treat fullscreen windows as 'tiled' for commands/focus * transaction: destroying nodes aren't hidden Commit 37d7bc6 ("transaction: Only wait for ack from visible views") introduced a check which uses view_is_visible() to check if a view is still visible on the screen. However view_is_visible() will early return in case the node is in the destroying state. This is incorrect for transactions, since a destroying view which is visible will trigger configure events for other clients. This bug was visible when repeatedly opening and closing two views side by side, since we ignore the destroying node we get a frame where the still open view is shown with the old configure values and the rest is the desktop background. The next frame is than correct again. Fix this by considering destroying views as visible, we correctly wait for them and send the configure events to other views in time, fixing the background flicker. Fixes #6473 * build: execute wlroots subproject before finding deps wlroots often requires dependencies more recent than Sway's. Executing the wlroots subproject first will give Meson a chance to find these newer dependencies, possibly via subprojects. The subproject will override the "wlroots" dependency when executed, so we don't need to use get_variable anymore. References: swaywm/sway#6498 (comment) * tray: do not render passive items https://www.freedesktop.org/wiki/Specifications/StatusNotifierItem/StatusNotifierItem/#org.freedesktop.statusnotifieritem.status * cmd/swap: error on swapping a container with itself * input/cursor: pass through pointer hold gestures This just follows swaywm/wlroots#3047, so `wl_pointer_gestures_v1` clients can be notified of these events. * swaynag: remove buffer destruction condition An address of a variable can never be NULL, so checking it doesn't make sense; and `destroy_buffer()` can operate on already destroyed buffers anyway. Fixes #6780 * Use bools for CLI flags * xwayland: listen to `request_activate` event When REAPER submenu is closed `XCB_CLIENT_MESSAGE` with type `NET_ACTIVE_WINDOW` is sent to set focus to parent menu. Closes: swaywm/sway#6324 * chore: chase wlr_output_layout_get_box() update https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3439 * Chase wlroots xdg-shell refactor * Minor update to focus_on_window_activation Removed xwayland limitation since wayland clients are supported via xdg-activation. * Translated README into Italian * readme: add link to Italian translation * readme: sort language list alphabetically * readme: use relative links for translations * xdg-shell: use wlr_xdg_toplevel in sway_view Improved type safety. Closes: swaywm/sway#6813 * xdg-shell: use wlr_xdg_popup in sway_xdg_popup Improved type safety. * Fix snprintf compiler warning * Remove all sprintf calls Replace them with snprintf, which ensures buffer overflows won't happen. * sway/commands: add missing wlr_keyboard interface include in xkb_switch_layout * sway/input: use wlr_input_device from input device base * Remove some erroneous apostrophes in comments * Don't enter seatop_move_floating when fullscreen Currently, a floating window that's been fullscreened can send us xdg_toplevel::move, and we'll enter seatop_move_floating, which lets us drag the surface around while it's fullscreen. We don't want this--fullscreen surfaces should always be aligned to the screen--so add the same check that seatop_default already does when entering this mode. Tested with Weston's weston-fullscreen demo, which sends a move request if you click anywhere on its surface. * swaynag: die on all allocation failures * sway/input: don't pass possibly invalid modifiers pointer active_keyboard may be NULL, in which case an invalid pointer could be passed to wlr_input_method_keyboard_grab_v2_send_modifiers. This procedure call is unnecessary since wlroots commit 372a52ec "input method: send modifiers in set_keyboard", so the call can simply be removed. Fixes #6836. * sway/input: destroy sway_switch properly Fix: #6861 Added seat_device_destroy function to seat_device_destroy function. * commands/focus: fix segfault when no container is already focused. Fixes #6690. * Remove WLR_SWITCH_STATE_TOGGLE usage Ref [1]. [1]: https://gitlab.freedesktop.org/wlroots/wlroots/-/commit/4792446ee8f50104bd207d9ccd8558a7e4eb4514 * Replace pcre with pcre2 Closes: swaywm/sway#6838 * swaybar: remove swaybar_output.input_region No need to keep the region around, we can immediately destroy it after the wl_surface.set_input_region request. * swaybar: set opaque region When the background color is fully opaque, set the surface's opaque region to the whole surface. * Updating criteria checking with PCRE2 * swaynag: allocate button_details with details They are used together, so it doesn't make sense to allocate them separately. * swaynag: statically allocate button_close, and move declaration Every swaynag has a close button, so it doesn't make sense to allocate it dynamically. The declaration is moved later to when it is actually needed. * swaynag: remove unnecessary zero of swaynag struct Global variables are initialized to 0. * swaynag: remove redundant status variables in main Instead, we just use `status` for all failures. * remove unnecessary strlen call * sway/input/cursor: take device mm size from wlr_tablet * sway/input/seat: take output name from specialized input device * sway/input: follow up wlroots input device events renaming * sway/input: fix bad position of wlr_drag * sway/input: wlr_seat_keyboard() now takes wlr_keyboard * bash-completion: localize variables * sway/main: move constants off the stack This makes stack traces from gdb slightly easier to read. * Fix farsi label * Avoid format-truncation warning The existing code gives this error when compiled with GCC 12: ../sway/server.c: In function ‘server_init’: ../sway/server.c:217:75: error: ‘%d’ directive output may be truncated writing between 1 and 11 bytes into a region of size 8 [-Werror=format-truncation=] 217 | snprintf(name_candidate, sizeof(name_candidate), "wayland-%d", i); | ^~ ../sway/server.c:217:66: note: directive argument in the range [-2147483647, 32] 217 | snprintf(name_candidate, sizeof(name_candidate), "wayland-%d", i); | ^~~~~~~~~~~~ ../sway/server.c:217:17: note: ‘snprintf’ output between 10 and 20 bytes into a destination of size 16 217 | snprintf(name_candidate, sizeof(name_candidate), "wayland-%d", i); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Because i is never negative, this is a false positive, but it is easy to change i to unsigned to silence the error. * Shuffle variables to satisfy -Werror=restrict This also fixes an invalid strlen invocation on uninitialized memory. * layer_shell: keep output non-NULL wherever possible 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 c9060bc ("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 a0bbe67 ("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. * Chase wlroots X11 hints update * Add Swedish README * Support cursor capture in grimshot Refactor argument parser Bring back `sh` compatibility Default to NOTIFY=no * Update grimshot.1.scd Fixed typo. The object is **files**, which is plural. **image** modifies files; it's not countable. * xkb_switch_layout: fix relative layout switches Fixes #6011 * Implement ext-session-lock-v1 * Avoid inspecting a NULL view in seat_set_focus Fixes #6968 * swaynag: do error checking and rename read_from_stdin read_from_stdin not only read from stdin, but trimming trailing newlines, so rename it to reflect this. * swaynag: improve robustness when loading config * swaynag: combine consecutive declaration/assignments * config: Remove unused mouse binding structure Mouse bindings are handled alongside normal bindings. Remove the unused separate data structure definition to avoid confusion. Signed-off-by: Michael Weiser <michael.weiser@gmx.de> * Replace strncpy with memcpy strncpy is useless here, is dangerous because it doesn't guarantee that the string is NUL-terminated and causes the following warning: ../sway/criteria.c: In function ‘criteria_parse’: ../sway/criteria.c:712:25: error: ‘strncpy’ destination unchanged after copying no bytes [-Werror=stringop-truncation] 712 | strncpy(value, valuestart, head - valuestart); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * Add descriptions for `stacking` and `tabbed` layouts Resolves #5918 * man: Fix trailing spaces * server: request xdg-shell v2 Wlroots does not yet support the newer xdg-shell versions and now requires the compositor to set the supported xdg-shell version during creation. Set this to v2 for sway as well. Fixes swaywm/sway#7001 * realtime: request SCHED_RR using CAP_SYS_NICE Try to gain SCHED_RR (round-robin) realtime scheduling privileges before starting the server. This requires CAP_SYS_NICE on Linux systems. We additionally register a pthread_atfork callback which resets the scheduling class back to SCHED_OTHER (the Linux system default). Due to CAP_SYS_NICE, setting RLIMIT_RTPRIO has no effect on the process as documented within man 7 sched (from Linux): Privileged (CAP_SYS_NICE) threads ignore the RLIMIT_RTPRIO limit; as with older kernels, they can make arbitrary changes to scheduling policy and priority. See getrlimit(2) for further information on RLIMIT_RTPRIO Note that this requires the sway distribution packagers to set the CAP_SYS_NICE capability on the sway binary. Supersedes #6992 * ext-session-lock: disable direct scan-out when locked * Polish the language in README.zh-CN.md & sync with English one Co-Authored-By: Urey. Xue <urey.s.knowledge@gmail.com> * De-duplicate IPC output descriptions * Handle NULL output make/model/serial * chore: chase wlroots xdg-shell update * xdg-shell: schedule a configure on maximize request This commit reverts 03879290dbee26127f6867ef60bc2a7f9a6c8c5f and fc84bcb7fb0ffa29b1f9bed287762241a3473803. * Add a Hindi (हिन्दी) translation to the README Hindi is one of the most prominent languages of the Indian Subcontinent. This commit adds the translation of the README into the Hindi language. Some of the words are still written in English because there wasn't an appropriate technical term of the word in the language. Co-authored-by: Surendrajat <surendrajat@protonmail.com> * sway: add bindgesture command Co-authored-by: Michael Weiser <michael.weiser@gmx.de> * build: link with -pthread Fixes the following FreeBSD error: ld: error: undefined symbol: pthread_getschedparam >>> referenced by realtime.c:25 (../sway/realtime.c:25) >>> sway/sway.p/realtime.c.o:(set_rr_scheduling) Fixes: a3a82ef ("realtime: request SCHED_RR using CAP_SYS_NICE") * ipc: remove chatty debug log messages These aren't particularly useful, and clobber the debug logs. * Refuse to start when SUID is detected This ensures that those surprised by the deprecation of SUID operation receive an error rather than accidentally having sway run as root. This detection will be removed in a future release. * swaynag: move close_button up to fix SIGSEGV When swaynag_parse_options encounters '--dismiss-button' (or its shorthand '-s'), it sets the text of the first button in the swaynag.buttons list, which is expected to exist and to be the dismiss button, to the one passed by the user. Commit 4780afb ("swaynag: statically allocate button_close, and move declaration") moved the list initialization to after swaynag_parse_options is called which made that code fail. For example, the command 'swaynag --dismiss-button Dismiss' crashes and 'swaynag --message Message --button Yes "" --dismiss-button Dismiss' shows the wrong buttons. Move it back to before swaynag_parse_options is called. * config/output: use wlr_output_commit_state This makes the code more robust because we don't potentially leave bad state in wlr_output.pending behind anymore. This also fixes a bug. Closes: swaywm/sway#7043 References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3610 * Allocate enough space for `cmd_results->error` * Remove access to wlr_input_device union References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3626 Closes: swaywm/sway#7077 * Rename dpms output command to power The "dpms" command refers to VESA Display Power Management Signaling, a deprecated standard. It's superseded by VESA DPM. Instead of tying out command name to a particular standard, use the neutral term "power". * Strip quotes in bindsym --input-device=... If the input device is quoted, which is common when using variables in the config file, those quotes must be ignored here, or the input device will be ignored. Fixes #7029. * Avoid unecessary string copy * Reject font values that are invalid for pango Use pango to parse font configuration early, and reject the command as invalid if the value is invalid for pango. Since we're already parsing the font into a `PangoFontDescription`, keep that instance around and avoid re-parsing the font each time we render text. Fixes: swaywm/sway#6805 * Reuse parsed PangoFontDescription Avoids parsing the configured font each time text is rendered. * ipc: add "power" to output reply * config.in: switch to `output power` * Remove internal references to DPMS While at it, use an int for the config field, just like we do for all other fields. * fix: remove redundant empty statement in main.c This semi-colon looks like a typo. Luckily, it has no effect on the code as it's treated as an empty statement leading the switch case. Really straightforward nitpick change, was just something I was confused by when reading over the code. * input: chase delta_discrete semantics change * swaymsg: fix floating_nodes being ignored Fix floating_nodes being ignored in pretty_print_tree. * ipc: make get_deco_rect check config->hide_lone_tab Without this, the `IPC_GET_TREE` ipc call would return false information about the container's `deco_rect` and `rect` properties if `hide_edge_borders --i3` was in effect. * grimshot: fix tilde expansion within quotes * Enable single-pixel-buffer-v1 References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3428 * sway-output.5: improve display of parameter Since "width" and "height" are separate parameters, show them as such. * man: sway(5) move fixes * ipc: drop WLR_OUTPUT_ADAPTIVE_SYNC_UNKNOWN case This has been removed from wlroots. * config/output: test adaptive sync Required for [1]. [1]: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3637 * Fix crash in xdg_activation_v1.c wlr_xdg_surface_from_wlr_surface() can return a NULL pointer, so check for NULL before dereferencing it. * sway: Add non-desktop-output type Currently, when encountering a non-desktop display, sway offers the output for leasing and returns without storing it in a sway specific output type like `struct sway_output`. Additionally, running `swaymsg -t get_outputs` doesn't show non-desktop outputs. This commit stores the non-desktop outputs into a struct called `sway_output_non_desktop`, and adds them to a list on `sway_root` * sway: add non-desktop outputs to json when running `swaymsg -t get_outputs` * swaymsg: show non-desktop property when pretty printing outputs * man: Add XWayland information * ipc: expose mode picture aspect ratio * swaymsg: show mode picture aspect ratio * build: simplify protocol paths No need for arrays here. * sway/commands/output: Add command for unplugging non-physical outputs * Improve Japanese translation * allow pointer_constraints on layer_shell surfaces * check for NULL * use seat directly * Use keyboard_state.focused_surface directly * input: focus floating container when clicked on border Fixes #7209. * input: focus container when scrolling on titlebar Fixes #6503. * Fix leaks in criteria_destroy() * Avoid double free in criteria_destroy() * Add support for ext-idle-notify-v1 References: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/3753 * input: tweak focus behavior to allow focusing parent containers Sway focuses the inactive child when focusing split containers. However, there is currently no way to focus the parent container itself by mouse. A user must use the keyboard to do so. This commit maintains the current behavior, but makes it such that a second click on the split container titlebar (i.e., after its children are visible) focuses the split container itself. * Fix keymap being NULL and segfaulting on dev add Moved `libinput_config` to the callers of `sway_input_configure_libinput_device` so that we send the event after the added event. * Rework session lock keyboard focus handling When removing outputs, it is possible to end up in a situation where none of the session lock client's surfaces have keyboard focus, resulting in it not receiving keyboard events. Track the focused surface and update it as needed on surface destroy. * Fix focus tracking when session lock is active Remove the incorrect attempt to block focus changes when an input grab is present and replace it with the same logic used for layer_shell-based screen lockers: restore the focus after changing it. This fixes a use-after-free of seat->workspace if outputs are destroyed while a screen lock is enabled. * container_floating_set_default_size: Store workspace size box on the stack * Support libinput's 1.21 new dwtp option Support the new dwtp (disable while trackpointing) option introduced in libinput 1.21, allowing users to control whether the trackpoint (like those in Thinkpads, but not only) should be disabled while using the keyboard/touchpad. See: https://gitlab.freedesktop.org/libinput/libinput/-/issues/731 * tree: support formatting null titles Any windows that have never had a title set visually behave closer to that of an empty title, but are unformattable, as the code bails out early on a NULL title. * criteria: allow matching on empty (NULL) titles * criteria: allow matching for NULL string criteria * ci: install hwdata * Use wl_signal_emit_mutable() This function fixes segfaults when emitting a signal potentially removes arbitrary listeners. * Use wlr_damage_ring wlr_output_damage is to be replaced with wlr_damage_ring, so use that. * lock: fix crash on output destroy Closes: swaywm/sway#7120 * container_get_siblings: handle NULL workspace * ci: checkout wlroots 0.16.0 * workspace_create: Don't allow NULL name (cherry picked from commit 34933bb) * output: set damage ring bounds to pixel values Fixes: swaywm/sway#7254 (cherry picked from commit 85005b5) * Use shm_open instead of mkstemp shm_open is more reliable because it does not require a writeable filesystem folder, unlike mkstemp. (cherry picked from commit e2bc886) * build: drop wayland-scanner fallback (cherry picked from commit 366f6ef) * build: unify server & client protocol generation No need to make a difference here, let's just generate header files for both. (cherry picked from commit 5be5a03) * build: drop "server" from target name for protocol code (cherry picked from commit e5475d9) * build: drop intermediate libraries for protocols (cherry picked from commit af8a5a8) * root: move the workspace matching code to its own file This removes the pid_workspace bits from tree/root before it gets too interesting. No functional change. (cherry picked from commit eb5021e) * node: prettify node type names (cherry picked from commit 1c4b94a) * launcher: track workspaces by node This removes the need to rename the pid_workspaces when a workspace is renamed. It also opens the possibility of tracking other node types. Tracking containers would allow application to be placed correctly in the container tree even if the user has moved their focus elsewhere since it was launched. (cherry picked from commit 3b49f27) * launcher: use xdga tokens This reuses wlroots token tracking for workspace matching. It doesn't export any xdga tokens for clients yet. (cherry picked from commit bd66f49) * launcher: rename pid_workspace to launcher_ctx Soon we will match views with more than just a pid. (cherry picked from commit d75c9f9) * view: associate launch contexts with views Views now maintain a reference to a launch context which, as a last resort, is populated at map time with a context associated with its pid. This opens the possibility of populating it before map via another source, e.g. xdga-tokens or configuration. (cherry picked from commit 864b3a9) * launcher: initialize launcher_ctxs once on startup (cherry picked from commit 6656850) * launcher: fudge the interface a bit We want to create a context before knowing the pid it will match with. (cherry picked from commit bdeb9f9) * launcher: export xdga tokens and use them for workspace matching (cherry picked from commit 30ad4dc) * launcher: export X startup ids and use them for workspace matching (cherry picked from commit 28fda4c) * swaybar: Prioritize hotspot events to bar bindings This is consistent with i3bar's behaviour, and for example, allows binding a command to button1, while still being able to click on tray icons or other zones on the bar's status line which may have their own bindings. E.g., in Sway, without this commit, this config. makes tray icons unclickable: bar { # ... bindsym button1 exec swaynag -m You_clicked_the_tray._Want_some_help? } But the same configuration in i3 (with i3-nagbar) keeps tray items clickable. Signed-off-by: Joan Bruguera <joanbrugueram@gmail.com> (cherry picked from commit 53f9dbd) * swaybar: Make hotspots block bar release bindings The previous commit prioritized hotspots before bar bindings for press events, which matches i3's behaviour. However, since hotspots don't need to do any processing on release events, those were not handled, and simply fell through to `bindsym --release` bar bindings (if any). This is counter-intuitive, and doesn't match i3's behaviour. Instead in case a hotspot handles the press event, it should also handle the release event, doing nothing, but blocking the event from triggering a --release bar binding. E.g., in Sway, without this commit, this config. shows a text on tray clicks: bar { # ... bindsym --release button1 exec swaynag -m I_got_the_release_event. } But the same configuration in i3 (with i3-nagbar) doesn't show the text. Signed-off-by: Joan Bruguera <joanbrugueram@gmail.com> (cherry picked from commit 94b69ac) * build: drop unused wayland-egl dependency (cherry picked from commit 37e4a3d) * build: bump version to 1.8-rc1 * Fix build on Debian Stable (cherry picked from commit dca0bb5) * build: fix have_xwayland when xcb-icccm is not found xcb-icccm is required to build Xwayland support. Backported from commit d41f11e. * build: bump version to 1.8-rc2 * seat: Avoid sending redundant keymaps on reload When we reload the config, we reset every input device and re-apply configuration from the config file. This means that the keyboard keymap is updated at least once during config reload, more if the config file contains keyboard configuration. When they keyboard keymap changes and is updated through wlr_seat, the keymap ends up sent to every keyboard bound in every client, seemingly multiple times. On an x230 of mine with a keyboard layout set in the config file, I see 42 keymap events sent to foot on config reload. Reduce events from keyboard configurations by skipping all but the currently active keyboard for the seat, and by clearing the active keyboard during input manager device reset. After this change, I only see a single just-in-time keymap event. Fixes: swaywm/sway#6654 * criteria: be lenient on window_role and instance too * build: bump version to 1.8-rc3 * commands/move: Warp cursor after moving workspace to another output This makes sway's behavior consistent with i3 when `mouse_warping` is set to any value besides `none`. Fixes #7027. (cherry picked from commit e3c63bf) * seat: Set keyboard if seat keyboard is NULL sway sends wl_keyboard.enter on seat focus change and when a keyboard active on a seat is configured. If all keyboards are removed and a keyboard is added back without changing the focused client, no new notify event would be sent despite having keyboard focus. This could lead to key events without notify, which is a protocol violation. As a quick fix, when configuring a keyboard on a seat where no keyboard is currently active, activate the keyboard so that a focused surface will receive a notify event. Regressed by: e1b268a Closes: swaywm/sway#7330 (cherry picked from commit 1ade0ce) * build: bump version to 1.8-rc4 * swaynag: fix NULL font description The font description was only set if provided on the CLI. It was left NULL for the defaults and when reading from the config file. Closes: swaywm/sway#7186 (cherry picked from commit fd0af78) * build: bump version to 1.8 * Removed other README languages * Fixed build issues * Removed alpha from render_data struct * Updated PKGBUILDs and COPR spec * Update sway/desktop/render.c Co-authored-by: Alexis Tacnet <alexistacnet@gmail.com> * Fixed deco_data not being initialized properly * Replaced wlr_egl_(make|unset)_current with eglMakeCurrent * Added matrix_projection into fx_renderer Signed-off-by: Michael Weiser <michael.weiser@gmx.de> Co-authored-by: Sefa Eyeoglu <contact@scrumplex.net> Co-authored-by: Simon Ser <contact@emersion.fr> Co-authored-by: Seth Barberee <seth.barberee@gmail.com> Co-authored-by: Thomas Hebb <tommyhebb@gmail.com> Co-authored-by: Nathan Schulte <nmschulte@gmail.com> Co-authored-by: xdavidwu <xdavidwuph@gmail.com> Co-authored-by: David Rosca <nowrep@gmail.com> Co-authored-by: David96 <david@hameipe.de> Co-authored-by: Kenny Levinsen <kl@kl.wtf> Co-authored-by: Aleksei Bavshin <alebastr89@gmail.com> Co-authored-by: Kirill Chibisov <contact@kchibisov.com> Co-authored-by: Tudor Brindus <me@tbrindus.ca> Co-authored-by: Patrick Hilhorst <git@hilhorst.be> Co-authored-by: Rouven Czerwinski <rouven@czerwinskis.de> Co-authored-by: Tobias Bengfort <tobias.bengfort@posteo.de> Co-authored-by: Ronan Pigott <rpigott@berkeley.edu> Co-authored-by: Kirill Primak <vyivel@eclair.cafe> Co-authored-by: Tuomas Yrjölä <mail@yrhki.fi> Co-authored-by: Kirill Primak <vyivel@posteo.net> Co-authored-by: Alexander Browne <elcste@users.noreply.github.com> Co-authored-by: Marco Rubin <20150305+Rubo3@users.noreply.github.com> Co-authored-by: Muhamed Hobi <woohoomoo2u@gmail.com> Co-authored-by: Simon Zeni <simon@bl4ckb0ne.ca> Co-authored-by: Nihal Jere <nihal@nihaljere.xyz> Co-authored-by: Alexander Gramiak <agrambot@gmail.com> Co-authored-by: Moon Sungjoon <sumoon@seoulsaram.org> Co-authored-by: Nicolas Avrutin <nicolas@avrutin.net> Co-authored-by: ndren <andreien@ctemplar.com> Co-authored-by: Bill Li <billli11hkb@gmail.com> Co-authored-by: Leonardo Hernández Hernández <leohdz172@protonmail.com> Co-authored-by: Oğuz Ersen <oguz@ersen.moe> Co-authored-by: Manuel Stoeckl <code@mstoeckl.com> Co-authored-by: Yasin Silavi <59373143+sttatusx@users.noreply.github.com> Co-authored-by: Daniel De Graaf <code@danieldg.net> Co-authored-by: kraftwerk28 <kefirchik3@gmail.com> Co-authored-by: Eskil <67291226+eschillus@users.noreply.github.com> Co-authored-by: Alice Carroll <git@alice-carroll.pet> Co-authored-by: Alan <51193876+Pound-Hash@users.noreply.github.com> Co-authored-by: Victor Makarov <vitja.makarov@gmail.com> Co-authored-by: Michael Weiser <michael.weiser@gmx.de> Co-authored-by: -k <slowdive@me.com> Co-authored-by: Hongyi <61831273+FrozenArcher@users.noreply.github.com> Co-authored-by: Urey. Xue <urey.s.knowledge@gmail.com> Co-authored-by: LordRishav <75823494+LordRishav@users.noreply.github.com> Co-authored-by: Surendrajat <surendrajat@protonmail.com> Co-authored-by: Florian Franzen <Florian.Franzen@gmail.com> Co-authored-by: Greg Depoire--Ferrer <greg@gregdf.com> Co-authored-by: Thomas Jost <schnouki@schnouki.net> Co-authored-by: Hugo Osvaldo Barrera <hugo@barrera.io> Co-authored-by: zkldi <ktchidev@gmail.com> Co-authored-by: llyyr <llyyr.public@gmail.com> Co-authored-by: Baltazár Radics <baltazar.radics@gmail.com> Co-authored-by: Martin Michlmayr <tbm@cyrius.com> Co-authored-by: Filip Szczepański <jazz2rulez@gmail.com> Co-authored-by: Alex Maese <memaese@hotmail.com> Co-authored-by: マリウス <marius@xn--gckvb8fzb.com> Co-authored-by: Andri Yngvason <andri@yngvason.is> Co-authored-by: ohno418 <yutaro.ono.418@gmail.com> Co-authored-by: Ferdinand Schober <ferdinand.schober@fau.de> Co-authored-by: cparm <armelcadetpetit@gmail.com> Co-authored-by: Yaroslav de la Peña Smirnov <yps@yaroslavps.com> Co-authored-by: Alexander Orzechowski <orzechowski.alexander@gmail.com> Co-authored-by: pudiva chip líquida <pudiva@skylittlesystem.org> Co-authored-by: Puck Meerburg <puck@puckipedia.com> Co-authored-by: Callum Andrew <calcium@mailbox.org> Co-authored-by: Ronan Pigott <ronan@rjp.ie> Co-authored-by: Joan Bruguera <joanbrugueram@gmail.com> Co-authored-by: nerdopolis <bluescreen_avenger@verizon.net> Co-authored-by: Ankit Pandey <anpandey@protonmail.com> Co-authored-by: Alexis Tacnet <alexistacnet@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Sway Version:
sway version 1.8-dev-f707f583 (Feb 13 2022, branch 'master')Debug Log:
Because I had sway running for almost 2 days before the crash, my debug log is 50MB. gist.github.com timed out trying to upload it. I uploaded a compressed version of it in the message below.
Configuration File:
This is the config file I'm using. Because of how long it takes to reproduce this bug, I didn't try to reproduce with the default config yet.
configuration file:
stack trace:
When I plug in my HUION H610 Pro USB graphics tablet, sometimes sway crashes. I have never had this happen the first time I plug the tablet in, however. I have to wait multiple hours or days of sway running, and several plug/unplug cycles of the tablet, before it finally crashes. When it does crash, it happens instantly, as soon as I plug the tablet in.
Similar behavior to #6502, but the backtrace looks pretty dissimilar.
The text was updated successfully, but these errors were encountered: