linux-kernel - Re: [PATCH 2/3] task: RCU protect tasks on the runqueue

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: <20190903194218.GU2349@hirez.programming.kicks-ass.net>
Date:   Tue, 3 Sep 2019 21:42:18 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Eric W. Biederman" <ebiederm@...ssion.com>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Oleg Nesterov <oleg@...hat.com>,
        Russell King - ARM Linux admin <linux@...linux.org.uk>,
        Chris Metcalf <cmetcalf@...hip.com>,
        Christoph Lameter <cl@...ux.com>,
        Kirill Tkhai <tkhai@...dex.ru>, Mike Galbraith <efault@....de>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...nel.org>,
        Linux List Kernel Mailing <linux-kernel@...r.kernel.org>,
        Davidlohr Bueso <dave@...olabs.net>,
        "Paul E. McKenney" <paulmck@...ux.ibm.com>
Subject: Re: [PATCH 2/3] task: RCU protect tasks on the runqueue

On Tue, Sep 03, 2019 at 01:13:22PM -0500, Eric W. Biederman wrote:

> I think this is where I am looking a things differently than you and
> Peter.  Why does it have to be ___schedule() that changes the value
> in the task_struct?  Why can't it be something else that changes the
> value and then proceeds to call schedule()?

If you call schedule() you will pass through plenty that already implies
smp_mb() before writing the ->curr pointer. If you care about that case,
adding RELEASE semantics to that store gains you absolutely nothing
except a marginally slower kernel.

> If we use RCU_INIT_POINTER if there was something that changed
> task_struct and then called schedule() what ensures that a remote cpu
> that has a stale copy of task_struct cached will update it's cache
> after following the new value rq->curr?  Don't we need
> rcu_assign_pointer to get that guarantee?

That whole construct doesn't really make sense: one it is very rare to
change task_struct content for !current tasks (and if we do, it must be
with atomic ops, because then there can be concurrency), secondly when
calling schedule() there is no guarantee on what @next will be.