linux-kernel - Re: [PATCH tip/core/rcu 01/22] sched: Make non-production PREEMPT cond

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: <20180423191227.GA26819@linux.vnet.ibm.com>
Date:   Mon, 23 Apr 2018 12:12:27 -0700
From:   "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        linux-kernel@...r.kernel.org, mingo@...nel.org,
        jiangshanlai@...il.com, dipankar@...ibm.com,
        akpm@...ux-foundation.org, mathieu.desnoyers@...icios.com,
        josh@...htriplett.org, tglx@...utronix.de, dhowells@...hat.com,
        edumazet@...gle.com, fweisbec@...il.com, oleg@...hat.com,
        joel.opensrc@...il.com
Subject: Re: [PATCH tip/core/rcu 01/22] sched: Make non-production PREEMPT
 cond_resched() help Tasks RCU

On Mon, Apr 23, 2018 at 08:47:07AM -0700, Paul E. McKenney wrote:
> On Mon, Apr 23, 2018 at 10:35:29AM -0400, Steven Rostedt wrote:
> > On Mon, 23 Apr 2018 16:10:38 +0200
> > Peter Zijlstra <peterz@...radead.org> wrote:
> > 
> > > >  http://lkml.kernel.org/r/20180227153646.GD3777@linux.vnet.ibm.com  
> > > 
> > > That thread using cond_resched_task_rcu_qs() seems like a _lot_ better
> > > than having cond_resched() semantics change depending on random
> > > !scheduler config parameters.
> > 
> > Yeah, I agree. Not sure why Paul didn't push it. Maybe because I never
> > replied to that final email and he forgot?
> > 
> > Paul?
> 
> Yeah, I have been a bit event-driven of late.  So the thought is to keep
> cond_resched() as-is and use cond_resched_task_rcu_qs(), that is after
> the rename, for the stress tests instead of the current cond_resched().
> Or did I lose the thread?

Actually, I already use cond_resched_tasks_rcu_qs() in
benchmark_event_kthread(), so all I need to do is drop this commit.

Which I have done, thanks everyone for the review!

							Thanx, Paul