04:03 Tim-Rex: Wasn't quite sure where to go to ask this, but I have a question about implementing a wayland client application.. Is there a provided Wayland mechanism to generate and receive user-defined message types?
04:03 Tim-Rex: Specifically, I'm wanting to signal my wayland client application from another thread, and unsure what the recommended approach is for a well behaved Wayland application.
04:05 Tim-Rex: I could certainly use non-wayland facilities to do that (shared memory, some other signalling) but something using the wayland event loop seems perhaps a better way to go (if that's a facility that is offered)
04:19 orowith2os[m]: Tim-Rex: if it doesn't involve other Wayland clients, or the compositor, you can use your own mechanism
04:21 orowith2os[m]: Wayland doesn't really get a say in what the client does for thread communication in the client
04:22 Tim-Rex: Okay, that works. Thanks for confirming @orowith2os
09:50 wlb: weston Merge request !1471 opened by manuel alfayate (vanfanel) libweston: input: enable pointer constraints for fullscreen surfaces. https://gitlab.freedesktop.org/wayland/weston/-/merge_requests/1471
11:14 wlb: weston Issue #881 closed \o/ (app_id and title are set after setting surface id https://gitlab.freedesktop.org/wayland/weston/-/issues/881)
14:48 any1: Is di_info_get_edid supposed to return a decent error message? It looks like most failure messages just get discarded because NULL is returned, so you can't get get the failure message from the info object.
14:50 emersion: hm, i don't think we set an error message when we return NULL?
14:50 emersion: do you have an example in mind?
14:50 emersion: but yeah, that's a known defect of the API
14:51 emersion: https://gitlab.freedesktop.org/emersion/libdisplay-info/-/issues/2
14:55 any1: https://gitlab.freedesktop.org/emersion/libdisplay-info/-/blob/main/edid.c?ref_type=heads#L541 -> line 1260 -> https://gitlab.freedesktop.org/emersion/libdisplay-info/-/blob/main/info.c?ref_type=heads#L29
14:55 any1: If I'm reading the code right, this happens all over the place.
14:56 emersion: parse_byte_descriptor() returns true
14:56 emersion: as in, the error is not fatal
14:56 emersion: as opposed to an allocation failure
14:59 any1: Oh, yeah, I did an incomplete search and skipped ahead
15:00 emersion: this is not very obvious for sure
15:29 wlb: weston Merge request !1472 opened by Jordan Williams (jwillikers) Rename os_create_anonymous_file to weston_os_create_anonymous_file https://gitlab.freedesktop.org/wayland/weston/-/merge_requests/1472
15:30 emersion: classic static lib symbol conflicts :/
15:31 emersion: we have a whole wlroots discussion about how it's unfortunate to not be able to strip internal symbols from static libs
15:35 JEEB: for multiple times it has felt to me that it would be mighty nice if there was a solib-like static library format
15:35 JEEB: and not just an archive of objects
15:43 emersion: yeah…
17:10 zamundaaa[m]: Vaughn: I made a MR for the Xwayland workaround: https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1356
17:11 Vaughn: 👍
17:12 Vaughn: I was unable to make the full patch-set not crash on startup. Not digging further into that, since it won't work anyway.
17:12 Vaughn: Though I did find that, once GLFW 3.4 is out, it'll stop being an issue; at that point relevant applications run natively on Wayland instead.
17:13 Vaughn: (And promptly crash, because GLFW errors out if the icon & mouse cursor position can't be set. But that's fine.)
17:16 Vaughn: https://usercontent.irccloud-cdn.com/file/TBmr7KFv/image.png
17:16 Vaughn: Easy enough to patch around anyway.
17:17 Vaughn: ...uhm, it boots and runs much, much smoother. I won't complain, but where is the 50% FPS increase coming from I wonder?
17:19 Vaughn: (Also, GLFW 3.4 *is* out, it just isn't packaged. Time to fix that~)
17:53 wlb: wayland-protocols Merge request !287 opened by Xaver Hugl (Zamundaaa) staging: add alpha-multiplier protocol https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/287
18:41 Vaughn: While I acknowledge I should presumably be trying to debug it instead... is it possible to disable high-resolution wheel scrolling in Wayland?
18:41 Vaughn: Aka. `Option "HighResolutionWheelScrolling" "off"`
18:54 zamundaaa[m]: Vaughn: outside of libinput quirks maybe, no
19:00 Vaughn: That does in fact work. It's https://unix.stackexchange.com/questions/743820/what-could-cause-a-missing-mouse-scroll-event-just-after-reversing-scroll-direct, but I've given up hope of fixing it properly.