netdev - Re: [PATCH 0/2] limit xen vnic max queues number to online cpu number

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:	Fri, 23 Oct 2015 17:05:21 +0800
From:	Joe Jin <joe.jin@...cle.com>
To:	Paul Durrant <Paul.Durrant@...rix.com>,
	Wei Liu <wei.liu2@...rix.com>,
	Ian Campbell <Ian.Campbell@...rix.com>,
	Boris Ostrovsky <boris.ostrovsky@...cle.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	"David S. Miller" <davem@...emloft.net>
Cc:	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"xen-devel@...ts.xenproject.org" <xen-devel@...ts.xenproject.org>
Subject: Re: [PATCH 0/2] limit xen vnic max queues number to online cpu number

On 10/23/2015 04:47 PM, Paul Durrant wrote:
>> -----Original Message-----
>> From: netdev-owner@...r.kernel.org [mailto:netdev-
>> owner@...r.kernel.org] On Behalf Of Joe Jin
>> Sent: 23 October 2015 08:54
>> To: Wei Liu; Ian Campbell; Boris Ostrovsky; Konrad Rzeszutek Wilk; David S.
>> Miller
>> Cc: netdev@...r.kernel.org; xen-devel@...ts.xenproject.org
>> Subject: [PATCH 0/2] limit xen vnic max queues number to online cpu
>> number
>>
>> Currently xen vnic allowed to create lots of queues by set module param
>> max_queues(both netback and netfront), when queues number larger than
>> cpu number, it does not help for performance but need more cpu time.
>>
> 
> But it's an override, so why would you want to limit it? The parameter should not be set in the common case.

Always we can not stop people use it because we provided it :)

If queues number is larger than cpu number, with heavy network load,
cpus have to take more time for interrupt, this lead others less
chance to be scheduled.
Imaging dom0 have 64 cpus, and assigned 4 vcpus to the guest, if
set max_queues to 64 on guest, it will consumed more cpu times
and bandwidth on backend, I think this is not we expected?

Thanks,
Joe
> 
>   Paul
> 
>> This patchset limit netback and netfront max queues number to online
>> cpus number.
>>
>> Joe Jin (2):
>>   xen-netback: limit xen vif max queues number to online cpus
>>   xen-front: limit vnic max_queues number to online cpus
>>
>>  drivers/net/xen-netback/netback.c |   28 ++++++++++++++++++++++------
>>  drivers/net/xen-netfront.c        |   27 +++++++++++++++++++++------
>>  2 files changed, 43 insertions(+), 12 deletions(-)
>>
>> --
>> 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
--
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