This wraps epoll and io_uring the other way: previously, we had the
io_uring file descriptor registered in epoll and when it became ready,
we could query its completion ring. The problem is that the poll
wakeups cause considerable overhead in the Linux kernel, see
https://lore.kernel.org/io-uring/20250128133927.3989681-9-max.kellermann@ionos.com/
By wrapping epoll inside io_uring using
io_uring_prep_poll_multishot(), the "outer" loop is io_uring and
inside it, we have epoll. This adds another system call for epoll,
but that will go away as soon as most operations are going through
io_uring. Previously, io_uring had this extra system call.
This is important for operations that have complex cancellation
procedures (e.g. if they need to free buffers). They might leave an
Operation instance in the queue.
Loader converted from java to kotlin.
Instead of loading libmpd when the service thread is started,
the service will not start the the thread if libmpd failed to load.
The loader is also accessed by the view data to let
the ui adjust if failed to load, by showing the failure reason
and disabling the Start MPD button.
Android will ignore permission request and will not show the request dialog
if the user's action implies "don't ask again."
This leaves the app in a crippled state and the user confused.
Google says "don't try to convince the user", so it returns false for `shouldShowRequestPermissionRationale`.
To help the user proceed, we show the `Request permission` button only if `shouldShowRequestPermissionRationale == true`
because there's a good chance the premission request dialog will not be ignored.
If `shouldShowRequestPermissionRationale == false` we instead show the "rationale" message and a button to open
the app info dialog where the user can explicitly grand the permission.
The output gets the standard MPD log format with domain curl and a timestamp.
Using CURLOPT_DEBUGFUNCTION that is only called when CURLOPT_VERBOSE is in effect
when MPD log level is verbose.
The BlockingCall() in InitUringInputPlugin() did not work because the
EventThread was not yet started. This was never noticed until commit
e309941646 which enabled `IORING_SETUP_SINGLE_ISSUER`, and suddenly
the kernel refused to accept io_uring_submit() calls from the
io_thread because io_uring_setup() had been called from the main
thread.
