[lkml]   [2007]   [May]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: [PATCH] make cancel_rearming_delayed_work() reliable
    On 05/15, Tejun Heo wrote:
    > Oleg Nesterov wrote:
    > > So, try_to_grab_pending() should check "VALID && pointers equal" atomically.
    > > We can't do "if (VALID && cwq == get_wq_data(work))". We should do something
    > > like this
    > >
    > > (((long)cwq) | VALID | PENDING) == atomic_long_read(&work->data)
    > >
    > > Yes? I need to think more about this.
    > I don't think the test for PENDING should be atomic too. cwq pointer
    > and VALID is one package. PENDING lives its own life as a atomic bit
    > switch.

    Yes sure, it should not be atomic. But (VALID && !PENDING) == BUG, so we
    can't just "kill" PENDING form the check above.

    > > BTW, in _theory_, spinlock() is not a read barrier, yes?
    > It actually is.
    > > From Documentation/memory-barriers.txt
    > >
    > > Memory operations that occur before a LOCK operation may appear to happen
    > > after it completes.
    > Which means that spin_lock() isn't a write barrier.

    yes, it is not very clear which "Memory operations" memory-barriers.txt

    > lock is read
    > barrier, unlock is write barrier.

    (To avoid a possible confusion: I am not arguing, I am trying to understand,
    and please also note "in _theory_" above)

    Is it so? Shoudn't we document this if it is true?

    > Otherwise, locking doesn't make much
    > sense.

    Why? Could you please give a code example we have which relies on this?

    > If we're going the barrier way, I think we're better off with
    > explicit smp_wmb(). It's only barrier() on x86/64.

    Yes. But note that we don't have any reason to do set_wq_data() under
    cwq->lock (this is also true for wake_up(->more_work) btw), so it makes
    sense to do this change anyway. And "wmb + spin_lock" looks a bit strange,
    I _suspect_ spin_lock() means a full barrier on most platforms.

    Could you also look at

    and, in particular,



    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to
    More majordomo info at
    Please read the FAQ at

     \ /
      Last update: 2007-05-16 00:03    [W:0.024 / U:0.320 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site