Ian Douglas Scott

Updated so the `debug` feature builds, and all the feature combinations tested on CI now work. Still needs a fix for that crash. Exactly how best to abstract over multiple...

This seems to be working correctly, including with the `debug` feature. We may want to change how renderers are abstracted dynamically in the future (and provide some helper for this...

It seems much cleaner to implement this in the `pixman` crate, if it's okay to have such a restriction there. So https://github.com/cmeissl/pixman-rs/pull/18 makes the changes I think are necessary for...

I guess the server should also remove the registry from `known_registries` (looks like currently elements are only removed their when the client is gone.) Not sure if there's anything else...

The `wayland.xml` update can just be merged now. I guess then we'd want https://github.com/Smithay/wayland-rs/pull/727, which I think is correct but @elinorbgr wanted some tests for (which definitely is a good...

Ah, I guess this (and making the socket code less redundant) is covered in https://github.com/bytecodealliance/rustix/pull/1004. I didn't really expect anyone else would have tried adding Netlink support recently. Though I...

Looks like this is failing to generate bindings for https://wayland.app/protocols/color-management-v1 since it has objects with "destructor" requests called `::create` that take a `new_id` argument, rather than no arguments like normal...

We wouldn't want to restrict it to only destructors called `::destroy`, since it should probably work with globals that have a `::release` destructor. Could restrict it to just destructors that...

https://gitlab.freedesktop.org/wlroots/wlroots/-/blob/master/backend/drm/atomic.c?ref_type=heads#L412-472 Wlroots does seem to disable planes when it sets `ACTIVE` to false. https://gitlab.gnome.org/GNOME/mutter/-/blob/main/src/backends/native/meta-kms-impl-device-atomic.c?ref_type=heads#L390-470 Mutter seems to set `MODE_ID` to `0` when it sets `ACTIVE` to `0`? I don't see...

> So naively we might want to additionally introduce an api to "turn off" the underlying DrmSurface, that just internally drops the DrmSurface inside DrmCompositor + invalidating all the state...