Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 19 May 2022 14:29:39 -0700
From:   "Paul E. McKenney" <paulmck@kernel.org>
To:     Qian Cai <quic_qiancai@quicinc.com>
Cc:     Mel Gorman <mgorman@techsingularity.net>,
        Andrew Morton <akpm@linux-foundation.org>,
        Nicolas Saenz Julienne <nsaenzju@redhat.com>,
        Marcelo Tosatti <mtosatti@redhat.com>,
        Vlastimil Babka <vbabka@suse.cz>,
        Michal Hocko <mhocko@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linux-MM <linux-mm@kvack.org>, kafai@fb.com, kpsingh@kernel.org
Subject: Re: [PATCH 0/6] Drain remote per-cpu directly v3
Message-ID: <20220519212939.GE1790663@paulmck-ThinkPad-P17-Gen-1>
Reply-To: paulmck@kernel.org
References: <20220512085043.5234-1-mgorman@techsingularity.net>
 <20220517233507.GA423@qian>
 <20220518125152.GQ3441@techsingularity.net>
 <YoUealVA1bMaSH2l@qian>
 <20220518171503.GQ1790663@paulmck-ThinkPad-P17-Gen-1>
 <YoZGSd6yQL3EP8tk@qian>
 <20220519191524.GC1790663@paulmck-ThinkPad-P17-Gen-1>
 <YoaxAMvQwHzDPxyi@qian>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <YoaxAMvQwHzDPxyi@qian>
Precedence: bulk

On Thu, May 19, 2022 at 05:05:04PM -0400, Qian Cai wrote:
> On Thu, May 19, 2022 at 12:15:24PM -0700, Paul E. McKenney wrote:
> > Is the task doing offline_pages()->synchronize_rcu() doing this
> > repeatedly?  Or is there a stalled RCU grace period?  (From what
> > I can see, offline_pages() is not doing huge numbers of calls to
> > synchronize_rcu() in any of its loops, but I freely admit that I do not
> > know this code.)
> 
> Yes, we are running into an endless loop in isolate_single_pageblock().
> There was a similar issue happened not long ago, so I am wondering if we
> did not solve it entirely then. Anyway, I will continue the thread over
> there.
> 
> https://lore.kernel.org/all/YoavU%2F+NfQIzQiDF@qian/

I do know that feeling.

> > Or is it possible that reverting those three patches simply decreases
> > the probability of failure, rather than eliminating the failure?
> > Such a decrease could be due to many things, for example, changes to
> > offsets and sizes of data structures.
> 
> Entirely possible. Sorry for the false alarm.

Not a problem!

> > Do you ever see RCU CPU stall warnings?
> 
> No.

OK, then perhaps a sequence of offline_pages() calls.

Hmmm...  The percpu_up_write() function sets ->block to zero before
awakening waiters.  Given wakeup latencies, might this allow an only
somewhat unfortunate sequence of events to allow offline_pages() to
starve readers?  Or is there something I am missing that prevents this
from happening?

							Thanx, Paul