linux-kernel - Re: [PATCH v2 rcu 13/16] workqueue: Make queue_rcu_work() use call_rcu

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:   Mon, 21 Nov 2022 15:37:56 -1000
From:   Tejun Heo <tj@...nel.org>
To:     "Paul E. McKenney" <paulmck@...nel.org>
Cc:     rcu@...r.kernel.org, linux-kernel@...r.kernel.org,
        kernel-team@...a.com, rostedt@...dmis.org,
        Uladzislau Rezki <urezki@...il.com>,
        Joel Fernandes <joel@...lfernandes.org>,
        Lai Jiangshan <jiangshanlai@...il.com>
Subject: Re: [PATCH v2 rcu 13/16] workqueue: Make queue_rcu_work() use
 call_rcu_flush()

On Mon, Nov 21, 2022 at 05:23:57PM -0800, Paul E. McKenney wrote:
> On Mon, Nov 21, 2022 at 03:09:29PM -1000, Tejun Heo wrote:
> > On Mon, Nov 21, 2022 at 05:04:18PM -0800, Paul E. McKenney wrote:
> > > And another call_rcu() instance that cannot be lazy is the one
> > > in queue_rcu_work(), given that callers to queue_rcu_work() are
> > > not necessarily OK with long delays.
> > 
> > So, this is fine but another thing we can do is propagating the distinction
> > through the workqueue interface so that the the choice can be made by
> > workqueue users. Would that make sense?
> 
> It might well!  My thought was to wait to suggest that until we found a
> real-life case where this was needed, but I have no objection to being
> proactive here.

Oh yeah, I'm completely fine either way too.

> But the hard part...  Thought for a good name?  ;-)

If we go with a separate interface, yeah, _flush would be confusing for
workqueue. Maybe _quick or _hurry? Hmm... it'd be nice to keep the suffix
consistent with RCU. What's the relationship with
synchronize_rcu_expedited()? Would using _expedited be confusing?

Thanks.

-- 
tejun