[lkml]   [2011]   [Feb]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    Patch in this message
    Subject[PATCH 2/3 v2] smp_call_function_many: handle concurrent clearing of mask
    Mike Galbraith reported finding a lockup ("perma-spin bug") where the
    cpumask passed to smp_call_function_many was cleared by other cpu(s)
    while a cpu was preparing its call_data block, resulting in no cpu to
    clear the last ref and unlock the block.

    Having cpus clear their bit asynchronously could be useful on a mask of
    cpus that might have a translation context, or cpus that need a push to
    complete an rcu window.

    Instead of adding a BUG_ON and requiring yet another cpumask copy, just
    detect the race and handle it.

    Note: arch_send_call_function_ipi_mask must still handle an empty cpumask
    because the data block is globally visible before the that arch callback
    is made. And (obviously) there are no guarantees to which cpus are
    notified if the mask is changed during the call; only cpus that were online
    and had their mask bit set during the whole call are guaranteed to be called.

    Reported-by: Mike Galbraith <>
    Signed-off-by: Milton Miller <>
    v2: rediff for v2 of call_function_many: fix list delete vs add race

    The arch code not expecting the race to empty the mask is the cause
    of that Andrew pointed

    Index: linux-2.6/kernel/smp.c
    --- linux-2.6.orig/kernel/smp.c 2011-01-31 18:25:47.266755387 -0600
    +++ linux-2.6/kernel/smp.c 2011-01-31 18:46:37.848099236 -0600
    @@ -450,7 +450,7 @@ void smp_call_function_many(const struct
    struct call_function_data *data;
    unsigned long flags;
    - int cpu, next_cpu, this_cpu = smp_processor_id();
    + int refs, cpu, next_cpu, this_cpu = smp_processor_id();

    * Can deadlock when called with interrupts disabled.
    @@ -489,6 +489,13 @@ void smp_call_function_many(const struct
    data-> = info;
    cpumask_and(data->cpumask, mask, cpu_online_mask);
    cpumask_clear_cpu(this_cpu, data->cpumask);
    + refs = cpumask_weight(data->cpumask);
    + /* some callers might race with other cpus changing the mask */
    + if (unlikely(!refs)) {
    + csd_unlock(&data->csd);
    + return;
    + }

    * We reuse the call function data without waiting for any grace
    @@ -512,7 +519,7 @@ void smp_call_function_many(const struct
    * We rely on the wmb() in list_add_rcu to order the writes
    * to func, data, and cpumask before this write to refs.
    - atomic_set(&data->refs, cpumask_weight(data->cpumask));
    + atomic_set(&data->refs, refs);
    raw_spin_unlock_irqrestore(&call_function.lock, flags);


     \ /
      Last update: 2011-02-01 08:15    [W:0.020 / U:1.524 seconds]
    ©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site