From mboxrd@z Thu Jan 1 00:00:00 1970 From: Li Zefan Subject: Re: [PATCH 2/2] cpuset: remove redundant checks in file write functions Date: Wed, 14 Aug 2013 10:01:32 +0800 Message-ID: <520AE4FC.5090702@huawei.com> References: <5209892D.8070502@huawei.com> <52098941.4070600@huawei.com> <20130813150549.GF26596@mtj.dyndns.org> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20130813150549.GF26596-9pTldWuhBndy/B6EtB590w@public.gmane.org> Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Tejun Heo Cc: LKML , cgroups On 2013/8/13 23:05, Tejun Heo wrote: > On Tue, Aug 13, 2013 at 09:17:53AM +0800, Li Zefan wrote: >> Now cgroup core gets a reference to the css when a cgroup file is >> opened(), and the reference is dropped at file release. so it's >> guaranteed the cpuset is online during the write function. > > Hmmm... it doesn't really guarantee that as css's can be offlined with > residual css refcnts, os the css may well be offlined by the time it > reaches the rw functions. What's guaranteed is that their refcnts > wouldn't be zero. Oh, right. But most controllers don't check this in those read/write functions. It shoudn't do any harm not checking online/offline status. > Eventually we need to implement proper sever > semantics (probably by replacing the custom fs implementation with > sysfs) but right now controllers still need to deal with offline > css's. > > Thanks. >