2016-11-21 0:38 GMT+08:00 Anders K. Pedersen | Cohaesio :
> From: Anders K. Pedersen
>
> As Liping Zhang reports, after commit a8b1e36d0d1d ("netfilter: nft_dynset:
> fix element timeout for HZ != 1000"), priv->timeout was stored in jiffies,
> while
On Sun, 2016-11-20 at 09:31 -0800, Eric Dumazet wrote:
> Thanks Eric, I will test the patch myself, because I believe we need it
> asap ;)
Current net-next without Florian patch :
lpaa24:~# time for f in `seq 1 2000` ; do iptables -A FORWARD ; done
real0m12.856s
user0m0.590s
sys
On Sun, 2016-11-20 at 12:22 -0500, Eric D wrote:
> I'm currently abroad for work and will come back home soon. I will
> test the solution and provide feedback to Florian by end of week.
>
> Thanks for jumping on this quickly.
>
> Eric
>
>
> On Nov 20, 2016 7:33 AM, "Eric Dumazet"
From: Anders K. Pedersen
As Liping Zhang reports, after commit a8b1e36d0d1d ("netfilter: nft_dynset:
fix element timeout for HZ != 1000"), priv->timeout was stored in jiffies,
while set->timeout was stored in milliseconds. This is inconsistent and
incorrect.
Firstly, we
2016-11-20 17:38 GMT+08:00 Anders K. Pedersen | Cohaesio :
[...]
> I believe that updating elements in dynsets can happen much more
> frequently than interactions with userspace. One of my own use cases is
Make sense. Considering the performance cost, your patch looks better:)
Hi Liping,
On søn, 2016-11-20 at 14:18 +0800, Liping Zhang wrote:
> After commit a8b1e36d0d1d ("netfilter: nft_dynset: fix element
> timeout
> for HZ != 1000"), priv->timeout was stored in jiffies, while
> set->timeout was stored in milliseconds. This is inconsistent and
> incorrect.
>
>