Messages in this thread | | | Date | Fri, 26 Jun 2009 16:50:27 +0200 | From | Oleg Nesterov <> | Subject | Re: [PATCH] net: fix race in the receive/select |
| |
On 06/26, Davide Libenzi wrote: > > On Fri, 26 Jun 2009, Oleg Nesterov wrote: > > > And if we remove waitqueue_active() in xxx_update(), then lock/unlock is > > not needed too. > > > > If xxx_poll() takes q->lock first, it can safely miss the changes in ->status > > and schedule(): xxx_update() will take q->lock, notice the sleeper and wake > > it up (ok, it will set ->triggered but this doesn't matter). > > > > If xxx_update() takes q->lock first, xxx_poll() must see the changes in > > status after poll_wait()->unlock(&q->lock) (in fact, after lock, not unlock). > > Sure. The snippet above was just to show what typically the code does, not > a suggestion on how to solve the socket case.
Yes, yes. I just meant you are right imho, we shouldn't add mb() into add_wait_queue().
> But yeah, the problem in this case is the waitqueue_active() call. Without > that, the wait queue lock/unlock in poll_wait() and the one in wake_up() > guarantees the necessary barriers. > Some might argue the costs of the lock/unlock of q->lock, and wonder if > MBs are a more efficient solution. This is something I'm not going into. > To me, it just looked not right having cross-matching MB in different > subsystems.
This is subjective and thus up to maintainers, but personally I think you are very, very right.
Perhaps we can add
void sock_poll_wait(struct file *file, struct sock *sk, poll_table *pt) { if (pt) { poll_wait(file, sk->sk_sleep, pt); /* * fat comment */ smp_mb(); // or smp_mb__after_unlock(); } }
Oleg.
| |