linux-kernel - Re: [PATCH net-next] net: link_watch: prevent starvation when processing linkwatch wq

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 May 2019 16:41:26 +0800
From:   Yunsheng Lin <linyunsheng@...wei.com>
To:     Salil Mehta <salil.mehta@...wei.com>,
        Stephen Hemminger <stephen@...workplumber.org>
CC:     "davem@...emloft.net" <davem@...emloft.net>,
        "hkallweit1@...il.com" <hkallweit1@...il.com>,
        "f.fainelli@...il.com" <f.fainelli@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Linuxarm <linuxarm@...wei.com>
Subject: Re: [PATCH net-next] net: link_watch: prevent starvation when
 processing linkwatch wq

On 2019/5/29 16:12, Salil Mehta wrote:
>> From: netdev-owner@...r.kernel.org [mailto:netdev-owner@...r.kernel.org] On Behalf Of Yunsheng Lin
>> Sent: Tuesday, May 28, 2019 2:04 AM
>>
>> On 2019/5/27 22:58, Stephen Hemminger wrote:
>>> On Mon, 27 May 2019 09:47:54 +0800
>>> Yunsheng Lin <linyunsheng@...wei.com> wrote:
>>>
>>>> When user has configured a large number of virtual netdev, such
>>>> as 4K vlans, the carrier on/off operation of the real netdev
>>>> will also cause it's virtual netdev's link state to be processed
>>>> in linkwatch. Currently, the processing is done in a work queue,
>>>> which may cause worker starvation problem for other work queue.
> 
> 
> I think we had already discussed about this internally and using separate
> workqueue with WQ_UNBOUND should solve this problem. HNS3 driver was sharing
> workqueue with the system workqueue. 

Yes, using WQ_UNBOUND wq in hns3 solved the cpu starvation for hns3
workqueue.

But the rtnl_lock taken by linkwatch is still a problem for hns3's
reset workqueue to do the down operation, which need a rtnl_lock.

> 
> 
>>>> This patch releases the cpu when link watch worker has processed
>>>> a fixed number of netdev' link watch event, and schedule the
>>>> work queue again when there is still link watch event remaining.
> 
> 
> We need proper examples/use-cases because of which we require above
> kind of co-operative scheduling. Touching the common shared queue logic
> which solid argument might invite for more problem to other modules.
> 
> 
>>>> Signed-off-by: Yunsheng Lin <linyunsheng@...wei.com>
>>>
>>> Why not put link watch in its own workqueue so it is scheduled
>>> separately from the system workqueue?
>>
>> From testing and debuging, the workqueue runs on the cpu where the
>> workqueue is schedule when using normal workqueue, even using its
>> own workqueue instead of system workqueue. So if the cpu is busy
>> processing the linkwatch event, it is not able to process other
>> workqueue' work when the workqueue is scheduled on the same cpu.
>>
>> Using unbound workqueue may solve the cpu starvation problem.
> 
> [...]
> 
>> But the __linkwatch_run_queue is called with rtnl_lock, so if it
>> takes a lot time to process, other need to take the rtnl_lock may
>> not be able to move forward.
> 
> Please help me in understanding, Are you trying to pitch this patch
> to solve more general system issue OR still your argument/concern
> is related to the HNS3 driver problem mentioned in this patch?

As about.

> 
> Salil.
> 
> 
> 
> 
> 
> 
>