Re: [PATCH] net/skbuff: silence warnings under memory pressure

[Sergey Senozhatsky]

On (09/04/19 08:54), Michal Hocko wrote:
> I am sorry, I could have been more explicit when CCing you.

Oh, sorry! My bad!

> Sure the ratelimit is part of the problem. But I was more interested
> in the potential livelock (infinite loop) mentioned by Qian Cai. It
> is not important whether we generate one or more lines of output from
> the softirq context as long as the printk generates more irq processing
> which might end up doing the same. Is this really possible?

Hmm. I need to look at this more... wake_up_klogd() queues work only once
on particular CPU: irq_work_queue(this_cpu_ptr(&wake_up_klogd_work));

bool irq_work_queue()
{
	/* Only queue if not already pending */
	if (!irq_work_claim(work))
		return false;

	 __irq_work_queue_local(work);
}

softirqs are processed in batches, right? The softirq batch can add XXXX
lines to printk logbuf, but there will be only one PRINTK_PENDING_WAKEUP
queued. Qian Cai mentioned that "net_rx_action softirqs again which are
plenty due to connected via ssh etc." so the proportion still seems to be
N:1 - we process N softirqs, add 1 printk irq_work.

But need to think more.
Interesting question.

	-ss