netdev - Re: [PATCH net-next-2.6] macvlan: add multiqueue capability

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: <4AA008C9.4000805@gmail.com>
Date:	Thu, 03 Sep 2009 20:19:53 +0200
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Patrick McHardy <kaber@...sh.net>
CC:	"David S. Miller" <davem@...emloft.net>,
	Linux Netdev List <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next-2.6] macvlan: add multiqueue capability

Patrick McHardy a écrit :
> Eric Dumazet wrote:
>> macvlan devices are currently not multi-queue capable.
>>
>> We can do that defining rtnl_link_ops method,
>> get_tx_queues(), called from rtnl_create_link()
>>
>> This new method gets num_tx_queues/real_num_tx_queues
>> from lower device.
>>
>> macvlan_get_tx_queues() is a copy of vlan_get_tx_queues().
>>
>> Because macvlan_start_xmit() has to update netdev_queue
>> stats only (and not dev->stats), I chose to change
>> tx_errors/tx_aborted_errors accounting to tx_dropped,
>> since netdev_queue structure doesnt define tx_errors /
>> tx_aborted_errors.
> 
> The patch looks fine, but it just occured to me that this won't
> have any effect since both VLAN and macvlan use a tx_queue_len of 0,
> so they will by default have queueing disabled. In fact this
> will increase costs for the default case since we're now hashing
> every packet.

Just read again dev_queue_xmit(), in case we have no queueing
on macvlan/vlan

Having mutiple txq should help multi flow / multi cpus setups,
since hashing will provide more chances to hit different txq/locks,
and let several cpus run concurrently, each one on a different queue.

So I dont understand why you think it'll increase costs...
--
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