netdev - Re: tx_queue

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] [day] [month] [year] [list]

Message-ID: <20090401111633.GD5970@ff.dom.local>
Date:	Wed, 1 Apr 2009 11:16:33 +0000
From:	Jarek Poplawski <jarkao2@...il.com>
To:	=?ISO-8859-2?Q?Miguel_=C1ngel_=C1lvarez?= 
	<gotzoncabanes@...il.com>
Cc:	netdev@...r.kernel.org, Krzysztof Halasa <khc@...waw.pl>
Subject: Re: tx_queue_len question

On 30-03-2009 18:49, Miguel Ángel Álvarez wrote:
> Hi
> 
> Could anyone please tell me how tx_queue_len works? (Or where to take
> a look to?).

Maybe net/sched/sch_generic.c: pfifo_fast_enqueue()?

> 
> I am facing two problems with it:
> - Using ixp4xx_hss driver, I am not in the same situation if I let the
> driver chose tx_queue_len in the initialization, or if I use ifconfig
> with the same value (100). With the last it seems to be capable of
> storing more frames.

This shouldn't matter unless I miss something.

> - When I increment it to 300 using ifconfig, the behaviour seems to be
> worse than in the case of 100 or 200. Which are the litis of the
> value?

Except some special cases there is used a common value for a similar
class of devices, like in ether_setup(). The behaviour should depend
on the kind of traffic.

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