Re: [PATCH v4 net-next 1/1] net_sched: Introduce skbmod action

On 16-09-12 06:26 PM, Eric Dumazet wrote:

On Mon, 2016-09-12 at 18:14 -0400, Jamal Hadi Salim wrote:

quoted

I noticed some very weird issues when I took that out.
Running sufficiently large amount of traffic (ping -f is sufficient)
I saw that when i did a dump it took anywhere between 6-15 seconds.
With the read_lock in place response was immediate.
I can go back and run things to verify - but it was very odd.

This was on uni processor ?

It was a VM.

Looks like typical starvation caused by aggressive softirq.

Well, then it is strange that in one case a tc dump of the rule
was immediate and in the other case it was consistent for 5-15
seconds.

Anyway, I suspect your kernel build has rcu_read_lock() and
rcu_read_unlock() as NOP ;)

Which doesnt give me a good feel if i tested this well ;->

I would like to try again with those two kernels just to
make sure i was not imagining this.

cheers,
jamal