netdev - Re: [PATCH][NET_SCHED] sch_sfq: prevent unnecessary reordering

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 linux-hardening linux-cve-announce 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]

Message-ID: <20080428154146.GC2870@ami.dom.local>
Date:	Mon, 28 Apr 2008 17:41:46 +0200
From:	Jarek Poplawski <jarkao2@...il.com>
To:	lists@...yfurniss.entadsl.com
Cc:	David Miller <davem@...emloft.net>,
	Patrick McHardy <kaber@...sh.net>, netdev@...r.kernel.org
Subject: Re: [PATCH][NET_SCHED] sch_sfq: prevent unnecessary reordering

Andy Furniss wrote, On 04/28/2008 01:37 PM:

> Jarek Poplawski wrote:
>> On 27-04-2008 16:22, Jarek Poplawski wrote:
>>> [NET_SCHED] sch_sfq: fix queue limiting while enqueuing
>>>
>> David, as Patrick noticed this patch was wrong and I withdraw it.
>>
>> I still can't see why sfq should have such a low limit,
> 
> It's a useful param which I use - on slow egress I can't see the point 
> of queuing many seconds worth of traffic. It may also help keep stronger 
> connections in check.

I agree the param is useful, but I can't see any technical reason why
sfq can't do a bit more queuing sometimes - especially when we expect
more flows/users per qdisc (probably another param would be necessary
for this).

> When shaping ingress it's essential to drop early to get out of 
> slowstart before the far buffer gets too full.
> 
> 
>> [NET_SCHED] sch_sfq: prevent unnecessary reordering
>>
>> Current check of queue limit in sfq_enqueue() isn't optimal: there
>> is really not much more needed to prevent unnecessary dropping and
>> possible reordering.
> 
> Maybe it makes no difference in the long run, but if you drop and then 
> enqueue a later packet the sender will get to know sooner so congestion 
> control can do it's stuff.

Happilly(?) it looks like my patch will not change here a lot...

Thanks for comments,
Jarek P.
--
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