----- On Jul 28, 2017, at 1:15 PM, Andrew Hunter a...@google.com wrote:
> On Thu, Jul 27, 2017 at 12:43 PM, Paul E. McKenney
> <paul...@linux.vnet.ibm.com> wrote:
>> On Thu, Jul 27, 2017 at 10:20:14PM +0300, Avi Kivity wrote:
>>> IPIing only running threads of my process would be perfect. In fact
>>> I might even be able to make use of "membarrier these threads
>>> please" to reduce IPIs, when I change the topology from fully
>>> connected to something more sparse, on larger machines.
>>>
>
> We do this as well--sometimes we only need RSEQ fences against
> specific CPU(s), and thus pass a subset.
>
>> +static void membarrier_private_expedited_ipi_each(void)
>> +{
>> + int cpu;
>> +
>> + for_each_online_cpu(cpu) {
>> + struct task_struct *p;
>> +
>> + rcu_read_lock();
>> + p = task_rcu_dereference(&cpu_rq(cpu)->curr);
>> + if (p && p->mm == current->mm)
>> + smp_call_function_single(cpu, ipi_mb, NULL, 1);
>> + rcu_read_unlock();
>> + }
>> +}
>> +
>
> We have the (simpler imho)
>
> const struct cpumask *mask = mm_cpumask(mm);
> /* possibly AND it with a user requested mask */
> smp_call_function_many(mask, ipi_func, ....);
>
> which I think will be faster on some archs (that support broadcast)
> and have fewer problems with out of sync values (though we do have to
> check in our IPI function that we haven't context switched out.
>
> Am I missing why this won't work?
The mm cpumask is not populated on all architectures, unfortunately, so
we need to do the generic implementation without it. Moreover, I recall
that using this in addition to the rq->curr checks adds extra complexity
wrt memory barriers vs updates of the mm_cpumask.
The ipi_each loop you refer to here is only for the fallback case.
The common case allocates a cpumask, populates it by looking at
each rq->curr, and uses smp_call_function_many on that cpumask.
Thanks,
Mathieu
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
