Re: [PATCH] POSIX message queues - syscalls & SIGEV_THREAD

[Jamie Lokier]

Krzysztof Benedyczak wrote:
> Intuitive
> solution is with FUTEX_FD & poll but this will have synchronization
> problems. The solution with one futex and multiple values would be very
> complicated (we need mechanism for cancellation of notification and of
> course information which queue(s) produced event(s)). On the another hand
> I can think about signals doing all the work - using thread sig mask we
> have synchronization and signals can carry quite a lot information. Of
> course this are only suggestions and I can miss something about futexes.

Please can you describe your "intuitive solution" using FUTEX_FD more clearly?

I don't quite understand what you wrote, but there are flaws(*) in the
current FUTEX_FD implementation which I would like to fix anyway.

Perhaps we can improve async futexes in a way which is useful for you?

Thanks,
-- Jamie

(*) FUTEX_FD cannot be used as a drop-in replacement for synchronous
futexes, due to a race condition in cancellation.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/