netdev - Re: [net-next PATCH 0/3] Backwards-compatible noqueue in virtual interfaces

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]

Date:	Wed, 29 Jul 2015 16:10:03 -0700
From:	Cong Wang <cwang@...pensource.com>
To:	Phil Sutter <phil@....cc>
Cc:	netdev <netdev@...r.kernel.org>,
	Jesper Dangaard Brouer <brouer@...hat.com>,
	Eric Dumazet <eric.dumazet@...il.com>,
	David Miller <davem@...emloft.net>
Subject: Re: [net-next PATCH 0/3] Backwards-compatible noqueue in virtual interfaces

On Wed, Jul 29, 2015 at 1:51 PM, Phil Sutter <phil@....cc> wrote:
> This series is inspired by a patch sent[1] in from Jesper Brouer and the
> discussion it started. Basically it tries to provide a solution which is
> backwards compatible while still enabling the veth performance
> improvement by default.

You can specify the tx_queue_len when you create the veth device,
if that can be your "default".

tx_queue_len ties too much with qdisc probably due to historical
reasons, ideally we should only specify the queue len via qdisc
interface since at least a per-device tx_queue_len doesn't make
much sense for those multiqueue or hierarchical qdisc's. It's hard
to change without breaking existing scripts/code.

Like David said, maybe you can add a new netdev flag saying
not to have any qdisc even when tx_queue_len !=0, or let each
device choose its default qdisc...
--
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