Ian Douglas Scott

The `poll_dispatch_pending` task does seem to work, however, if I have a task that calls `prepare_read()` and `read()`. But should it not work for reads like that done within GDK?

That's unfortunate. I guess libwayland wouldn't want to run callbacks added by a different thread since they may not be thread-safe. This behavior should be described somewhere in the documentation....

https://github.com/Smithay/wayland-rs/pull/572 has such a GTK example. So if there's a better way to do it, it can be demonstrated in that example.

It's kind of a strange situation. On the one hand, `zwp_linux_dmabuf_v1` *is* the name of the protocol. On the other hand, it *should* be named `wp_linux_dmabuf_v1` but just isn't for...

If it's buffering data and file descriptors from multiple messages, before sending on a flush, it could have more fds than the limit for a single message, right?

Ah, looking at https://www.man7.org/linux/man-pages/man7/unix.7.html: > The kernel constant SCM_MAX_FD defines a limit on the number of file descriptors in the array. Attempting to send an array larger than this limit...

Wayland uses `SOCK_STREAM` not `SOCK_DGRAM`, so even if `sendmsg`/`recvmsg` are used, I don't think there's really any concept of bytes and fds being parts of discrete datagrams/messages. But I'm not...

Yeah. A lot of the changes here could be merged without issue. With the OpenGL version currently in use. After looking into it more, it seems desktop platforms do not...

The same wording seems to be in all of them. > Calling close(2) on a file descriptor will remove any kevents that reference the descriptor. https://man.freebsd.org/cgi/man.cgi?query=kevent https://man.netbsd.org/kevent.2 https://man.openbsd.org/kqueue.2 It's caused...

My thoughts on this: * Separately from whether or not the function is marked `unsafe`, accepting `BorrowedFd` arguments makes it safer to use, and is more convenient in a project...