[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20170831011024.GC491396@devbig577.frc2.facebook.com>
Date: Wed, 30 Aug 2017 18:10:24 -0700
From: Tejun Heo <tj@...nel.org>
To: Neeraj Upadhyay <neeraju@...eaurora.org>
Cc: lizefan@...wei.com, mingo@...nel.org, longman@...hat.com,
akpm@...ux-foundation.org, linux-kernel@...r.kernel.org,
sramana@...eaurora.org, prsood@...eaurora.org
Subject: Re: [PATCH] cgroup: Fix potential race between cgroup_exit and
cpuset_attach
Hello,
On Wed, Aug 30, 2017 at 06:23:50PM +0530, Neeraj Upadhyay wrote:
> cpuset_hotplug_workfn()
> cgroup_transfer_tasks()
> cgroup_migrate()
> cgroup_migrate_execute()
> <TASK EXIT>
> list_del_init(&task->cg_list)
> cpuset_attach()
> cgroup_taskset_first(tset, &css) // css is not set
> guarantee_online_mems(cs, ...) // data abort
I don't think this can happen anymore. We lock out task exits during
migration and don't call migration callbacks if there's no tasks to
migrate.
Thanks.
--
tejun
Powered by blists - more mailing lists