lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Fri, 16 Jul 2010 07:30:10 +0800
From:	Changli Gao <xiaosuo@...il.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	hadi@...erus.ca, "David S. Miller" <davem@...emloft.net>,
	Patrick McHardy <kaber@...sh.net>,
	Tom Herbert <therbert@...gle.com>, netdev@...r.kernel.org
Subject: Re: [PATCH RFC] act_cpu: packet distributing

On Fri, Jul 16, 2010 at 12:54 AM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> Le jeudi 15 juillet 2010 à 08:48 -0400, jamal a écrit :
>>
>> RCU maybe a little trickier here Eric. Actions could be shared i.e.
>> example, it is possible to have a policer action restricting rates for a
>> group of flows  across multiple netdevices etc. Since action stats get
>> written to by different CPUs concurrently. It could be probably done if
>> one was to implement per-cpu stats which get summed-up when user space
>> asks.
>
> It's certainly tricky, but is act_cpu useful in its current shape, based
> on an infrastructure that had to use a lock because of exact
> rates/accounting ?
>
> I dont understand how distributing packets to different cpus, if going
> through a central lock can be an improvement. Changli patches are most
> of the time not documented, and no performance data is provided.
>
The tcf_lock is per-instance, so I should not an issue here if the
corresponding NIC isn't an multiqueue NIC or the instance is
per-rx-queue. I agree
the performance data is necessary and I'll publish it in the formal
patch.

> Even if we solve this locking problem, using percpu variables, act_cpu
> hits another problem :
>
> The socket refcount, taken by the 'master' cpu, and released by the
> consumer cpu.
This is why I asked if I can assign sk to skb.

>
> RFS provides sort of a lazy flow-based distribution without central lock
> or cache line ping pongs. Why Changli dont use this, we dont know yet.
>
The fact is I don't know how to do this. I'll work on this issue later.

Thanks.


-- 
Regards,
Changli Gao(xiaosuo@...il.com)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists