Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 23 Oct 2019 09:46:22 -0400
From:   Rafael Aquini <aquini@redhat.com>
To:     Vlastimil Babka <vbabka@suse.cz>
Cc:     Michal Hocko <mhocko@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Mel Gorman <mgorman@suse.de>, Waiman Long <longman@redhat.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Roman Gushchin <guro@fb.com>,
        Konstantin Khlebnikov <khlebnikov@yandex-team.ru>,
        Jann Horn <jannh@google.com>, Song Liu <songliubraving@fb.com>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>,
        Michal Hocko <mhocko@suse.com>
Subject: Re: [RFC PATCH 2/2] mm, vmstat: reduce zone->lock holding time by
 /proc/pagetypeinfo
Message-ID: <20191023134622.GA22601@optiplex-lnx>
References: <20191023095607.GE3016@techsingularity.net>
 <20191023102737.32274-1-mhocko@kernel.org>
 <20191023102737.32274-3-mhocko@kernel.org>
 <30211965-8ad0-416d-0fe1-113270bd1ea8@suse.cz>
MIME-Version: 1.0
In-Reply-To: <30211965-8ad0-416d-0fe1-113270bd1ea8@suse.cz>
User-Agent: Mutt/1.12.1 (2019-06-15)
Content-Type: text/plain; charset=WINDOWS-1252
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Oct 23, 2019 at 03:32:05PM +0200, Vlastimil Babka wrote:
> On 10/23/19 12:27 PM, Michal Hocko wrote:
> > From: Michal Hocko <mhocko@suse.com>
> >=20
> > pagetypeinfo_showfree_print is called by zone->lock held in irq mode.
> > This is not really nice because it blocks both any interrupts on that
> > cpu and the page allocator. On large machines this might even trigger
> > the hard lockup detector.
> >=20
> > Considering the pagetypeinfo is a debugging tool we do not really need
> > exact numbers here. The primary reason to look at the outuput is to see
> > how pageblocks are spread among different migratetypes therefore puttin=
g
> > a bound on the number of pages on the free_list sounds like a reasonabl=
e
> > tradeoff.
> >=20
> > The new output will simply tell
> > [...]
> > Node    6, zone   Normal, type      Movable >100000 >100000 >100000 >10=
0000  41019  31560  23996  10054   3229    983    648
> >=20
> > instead of
> > Node    6, zone   Normal, type      Movable 399568 294127 221558 102119=
  41019  31560  23996  10054   3229    983    648
> >=20
> > The limit has been chosen arbitrary and it is a subject of a future
> > change should there be a need for that.
> >=20
> > Suggested-by: Andrew Morton <akpm@linux-foundation.org>
> > Signed-off-by: Michal Hocko <mhocko@suse.com>
>=20
> Hmm dunno, I would rather e.g. hide the file behind some config or boot
> option than do this. Or move it to /sys/kernel/debug ?
>

You beat me to it. I was going to suggest moving it to debug, as well.


=20
> > ---
> >  mm/vmstat.c | 19 ++++++++++++++++++-
> >  1 file changed, 18 insertions(+), 1 deletion(-)
> >=20
> > diff --git a/mm/vmstat.c b/mm/vmstat.c
> > index 4e885ecd44d1..762034fc3b83 100644
> > --- a/mm/vmstat.c
> > +++ b/mm/vmstat.c
> > @@ -1386,8 +1386,25 @@ static void pagetypeinfo_showfree_print(struct s=
eq_file *m,
> > =20
> >  =09=09=09area =3D &(zone->free_area[order]);
> > =20
> > -=09=09=09list_for_each(curr, &area->free_list[mtype])
> > +=09=09=09list_for_each(curr, &area->free_list[mtype]) {
> >  =09=09=09=09freecount++;
> > +=09=09=09=09/*
> > +=09=09=09=09 * Cap the free_list iteration because it might
> > +=09=09=09=09 * be really large and we are under a spinlock
> > +=09=09=09=09 * so a long time spent here could trigger a
> > +=09=09=09=09 * hard lockup detector. Anyway this is a
> > +=09=09=09=09 * debugging tool so knowing there is a handful
> > +=09=09=09=09 * of pages in this order should be more than
> > +=09=09=09=09 * sufficient
> > +=09=09=09=09 */
> > +=09=09=09=09if (freecount > 100000) {
> > +=09=09=09=09=09seq_printf(m, ">%6lu ", freecount);
> > +=09=09=09=09=09spin_unlock_irq(&zone->lock);
> > +=09=09=09=09=09cond_resched();
> > +=09=09=09=09=09spin_lock_irq(&zone->lock);
> > +=09=09=09=09=09continue;
> > +=09=09=09=09}
> > +=09=09=09}
> >  =09=09=09seq_printf(m, "%6lu ", freecount);
> >  =09=09}
> >  =09=09seq_putc(m, '\n');
> >=20
>=20