linux-kernel - Re: [PATCH v2] rcu/nocb: Avoid polling when myrdp->nocb_head

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:   Tue, 7 Jun 2022 21:06:27 -0700
From:   "Paul E. McKenney" <paulmck@...nel.org>
To:     "Zhang, Qiang1" <qiang1.zhang@...el.com>
Cc:     "frederic@...nel.org" <frederic@...nel.org>,
        "rcu@...r.kernel.org" <rcu@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] rcu/nocb: Avoid polling when myrdp->nocb_head_rdp
 list is empty

On Wed, Jun 08, 2022 at 01:01:31AM +0000, Zhang, Qiang1 wrote:
> On Wed, Jun 08, 2022 at 12:41:28AM +0000, Zhang, Qiang1 wrote:
> > On Tue, Jun 07, 2022 at 03:50:57PM +0800, Zqiang wrote:
> > > Currently, If the 'rcu_nocb_poll' bootargs is enable, all rcuog 
> > > kthreads enter polling mode. however, due to only insert CPU's rdp 
> > > which belong to rcu_nocb_mask to 'nocb_head_rdp' list or all CPU's 
> > > rdp served by rcuog kthread have been de-offloaded, these cause the 
> > > 'nocb_head_rdp' list served by rcuog kthread is empty, when the 
> > > 'nocb_head_rdp' is empty, the rcuog kthread in polling mode not 
> > > actually do anything. fix it by exiting polling mode when the 
> > > 'nocb_head_rdp'list is empty, otherwise entering polling mode.
> > > 
> > > Signed-off-by: Zqiang <qiang1.zhang@...el.com>
> > >
> > >This looks a bit more plausible, but what have you done to test this?
> > 
> > Yes , I have only tested as follows and I added two trace events.
> > 
> > +                      trace_rcu_nocb_wake(rcu_state.name, cpu, 
> > + TPS("EnterNoPoll"));
> >                         
> > rcu_wait(READ_ONCE(my_rdp->nocb_toggling_rdp));
> > +                       trace_rcu_nocb_wake(rcu_state.name, cpu, 
> > + TPS("ExitNoPoll"));
> > 
> > runqemu kvm slirp nographic qemuparams="-m 2048 -smp 4" 
> > bootparams="rcu_nocbs=2,3 rcutree.dump_tree=1 rcu_nocb_poll 
> > rcutorture.nocbs_nthreads=4 rcutorture.test_boost=0" -d
> 
> >To exercise your change, could you please also add an appropriate value for the rcutorture.nocbs_nthreads kernel boot parameter?  Without that boot parameter, your kernel will not actually offload or deoffload any CPUs.
> 
> I have set ' rcutorture.nocbs_nthreads=4 ' in bootargs.

I see it now, apologies for my confusion!

							Thanx, Paul

> >An alternative approach is to run rcutorture scenario TREE01.
> 
> I will also use TREE01 for testing
> 
> Thanks
> Zqiang
> 
> >
> >							Thanx, Paul
> 
> > Thanks
> > Zqiang
> > 
> > >
> > >							Thanx, Paul
> > >
> > > ---
> > >  v1->v2:
> > >  Move rcu_nocb_poll flags check from rdp_offload_toggle() to 
> > > rcu_nocb_rdp_offload/deoffload(), avoid unnecessary setting of 
> > > rdp_gp->nocb_gp_sleep flags, because when rcu_nocb_poll is set  the 
> > > rdp_gp->nocb_gp_sleep is not used.
> > > 
> > >  kernel/rcu/tree_nocb.h | 16 ++++++++++------
> > >  1 file changed, 10 insertions(+), 6 deletions(-)
> > > 
> > > diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index
> > > fa8e4f82e60c..2a52c9abc681 100644
> > > --- a/kernel/rcu/tree_nocb.h
> > > +++ b/kernel/rcu/tree_nocb.h
> > > @@ -698,10 +698,14 @@ static void nocb_gp_wait(struct rcu_data *my_rdp)
> > >  				   TPS("WakeBypassIsDeferred"));
> > >  	}
> > >  	if (rcu_nocb_poll) {
> > > -		/* Polling, so trace if first poll in the series. */
> > > -		if (gotcbs)
> > > -			trace_rcu_nocb_wake(rcu_state.name, cpu, TPS("Poll"));
> > > -		schedule_timeout_idle(1);
> > > +		if (list_empty(&my_rdp->nocb_head_rdp)) {
> > > +			rcu_wait(READ_ONCE(my_rdp->nocb_toggling_rdp));
> > > +		} else {
> > > +			/* Polling, so trace if first poll in the series. */
> > > +			if (gotcbs)
> > > +				trace_rcu_nocb_wake(rcu_state.name, cpu, TPS("Poll"));
> > > +			schedule_timeout_idle(1);
> > > +		}
> > >  	} else if (!needwait_gp) {
> > >  		/* Wait for callbacks to appear. */
> > >  		trace_rcu_nocb_wake(rcu_state.name, cpu, TPS("Sleep")); @@ 
> > > -1030,7
> > > +1034,7 @@ static long rcu_nocb_rdp_deoffload(void *arg)
> > >  
> > >  	mutex_lock(&rdp_gp->nocb_gp_kthread_mutex);
> > >  	if (rdp_gp->nocb_gp_kthread) {
> > > -		if (wake_gp)
> > > +		if (wake_gp || rcu_nocb_poll)
> > >  			wake_up_process(rdp_gp->nocb_gp_kthread);
> > >  
> > >  		/*
> > > @@ -1152,7 +1156,7 @@ static long rcu_nocb_rdp_offload(void *arg)
> > >  	 *      rcu_nocb_unlock()         rcu_nocb_unlock()
> > >  	 */
> > >  	wake_gp = rdp_offload_toggle(rdp, true, flags);
> > > -	if (wake_gp)
> > > +	if (wake_gp || rcu_nocb_poll)
> > >  		wake_up_process(rdp_gp->nocb_gp_kthread);
> > >  	swait_event_exclusive(rdp->nocb_state_wq,
> > >  			      rcu_segcblist_test_flags(cblist, SEGCBLIST_KTHREAD_CB) &&
> > > --
> > > 2.25.1
> > >