On 2021/5/11 17:04, Yunsheng Lin wrote:
> On 2021/5/11 12:22, Jakub Kicinski wrote:
>> On Mon, 10 May 2021 09:42:36 +0800 Yunsheng Lin wrote:
>>> The netdev qeueue might be stopped when byte queue limit has
>>> reached or tx hw ring is full, net_tx_action() may still be
>>> rescheduled endlessly if STATE_MISSED is set, which consumes
>>> a lot of cpu without dequeuing and transmiting any skb because
>>> the netdev queue is stopped, see qdisc_run_end().
>>>
>>> This patch fixes it by checking the netdev queue state before
>>> calling qdisc_run() and clearing STATE_MISSED if netdev queue is
>>> stopped during qdisc_run(), the net_tx_action() is recheduled
>>> again when netdev qeueue is restarted, see netif_tx_wake_queue().
>>>
>>> Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking")
>>> Reported-by: Michal Kubecek <mkubecek@xxxxxxx>
>>> Signed-off-by: Yunsheng Lin <linyunsheng@xxxxxxxxxx>
>>
>> Patches 1 and 2 look good to me but this one I'm not 100% sure.
>>
>>> @@ -251,8 +253,10 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate,
>>> *validate = true;
>>>
>>> if ((q->flags & TCQ_F_ONETXQUEUE) &&
>>> - netif_xmit_frozen_or_stopped(txq))
>>> + netif_xmit_frozen_or_stopped(txq)) {
>>> + clear_bit(__QDISC_STATE_MISSED, &q->state);
>>> return skb;
>>> + }
>>
>> The queues are woken asynchronously without holding any locks via
>> netif_tx_wake_queue(). Theoretically we can have a situation where:
>>
>> CPU 0 CPU 1
>> . .
>> dequeue_skb() .
>> netif_xmit_frozen..() # true .
>> . [IRQ]
>> . netif_tx_wake_queue()
>> . <end of IRQ>
>> . netif_tx_action()
>> . set MISSED
>> clear MISSED
>> return NULL
>> ret from qdisc_restart()
>> ret from __qdisc_run()
>> qdisc_run_end()
>> -> MISSED not set
>
> Yes, the above does seems to have the above data race.
>
> As my understanding, there is two ways to fix the above data race:
> 1. do not clear the STATE_MISSED for netif_xmit_frozen_or_stopped()
> case, just check the netif_xmit_frozen_or_stopped() before
> calling __netif_schedule() at the end of qdisc_run_end(). This seems
> to only work with qdisc with TCQ_F_ONETXQUEUE flag because it seems
> we can only check the netif_xmit_frozen_or_stopped() with q->dev_queue,
> I am not sure q->dev_queue is pointint to which netdev queue when qdisc
> is not set with TCQ_F_ONETXQUEUE flag.
>
> 2. clearing the STATE_MISSED for netif_xmit_frozen_or_stopped() case
> as this patch does, and protect the __netif_schedule() with q->seqlock
> for netif_tx_wake_queue(), which might bring unnecessary overhead for
> non-stopped queue case
>
> Any better idea?
3. Or check the netif_xmit_frozen_or_stopped() again after clearing
STATE_MISSED, like below:
if (netif_xmit_frozen_or_stopped(txq)) {
clear_bit(__QDISC_STATE_MISSED, &q->state);
/* Make sure the below netif_xmit_frozen_or_stopped()
* checking happens after clearing STATE_MISSED.
*/
smp_mb__after_atomic();
/* Checking netif_xmit_frozen_or_stopped() again to
* make sure __QDISC_STATE_MISSED is set if the
* __QDISC_STATE_MISSED set by netif_tx_wake_queue()'s
* rescheduling of net_tx_action() is cleared by the
* above clear_bit().
*/
if (!netif_xmit_frozen_or_stopped(txq))
set_bit(__QDISC_STATE_MISSED, &q->state);
}
It is kind of ugly, but it does seem to fix the above data race too.
And it seems like a common pattern to deal with the concurrency between
xmit and NAPI polling, as below:
https://elixir.bootlin.com/linux/v5.12-rc2/source/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c#L1409
>
>>
>> .
>>
> _______________________________________________
> Linuxarm mailing list -- linuxarm@xxxxxxxxxxxxx
> To unsubscribe send an email to linuxarm-leave@xxxxxxxxxxxxx
>
