On Thu, Sep 01, 2016 at 09:01:41PM +0200, Peter Zijlstra wrote: > On Fri, Aug 26, 2016 at 02:45:52PM +0200, Oleg Nesterov wrote: > > > We do not need anything tricky to avoid the race, > > The race being: > > CPU0 CPU1 CPU2 > > __wait_on_bit_lock() > bit_wait_io() > io_schedule() > > clear_bit_unlock() > __wake_up_common(.nr_exclusive=1) > list_for_each_entry() > if (curr->func() && --nr_exclusive) > break > > signal() > > if (signal_pending_state()) == TRUE > return -EINTR > > And no progress because CPU1 exits without acquiring the lock and CPU0 > thinks its done because it woke someone.
FWIW, the way the mutex code avoids this issue is by doing the signal_pending test while holding the q->lock, that way its exclusive with wakeup.
