| 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
| ||
|
Message-ID: <20110819013628.GA7925@localhost>
Date: Fri, 19 Aug 2011 09:36:28 +0800
From: Wu Fengguang <fengguang.wu@...el.com>
To: Jan Kara <jack@...e.cz>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
Greg Thelen <gthelen@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-mm@...ck.org" <linux-mm@...ck.org>,
"containers@...ts.osdl.org" <containers@...ts.osdl.org>,
"linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
Balbir Singh <bsingharora@...il.com>,
Daisuke Nishimura <nishimura@....nes.nec.co.jp>,
Minchan Kim <minchan.kim@...il.com>,
Johannes Weiner <hannes@...xchg.org>,
Dave Chinner <david@...morbit.com>,
Vivek Goyal <vgoyal@...hat.com>,
Andrea Righi <andrea@...terlinux.com>,
Ciju Rajan K <ciju@...ux.vnet.ibm.com>,
David Rientjes <rientjes@...gle.com>,
"Li, Shaohua" <shaohua.li@...el.com>,
"Shi, Alex" <alex.shi@...el.com>,
"Chen, Tim C" <tim.c.chen@...el.com>
Subject: Re: [PATCH v9 12/13] memcg: create support routines for page
writeback
On Fri, Aug 19, 2011 at 04:08:56AM +0800, Jan Kara wrote:
> On Thu 18-08-11 20:17:14, Wu Fengguang wrote:
> > On Thu, Aug 18, 2011 at 06:12:48PM +0800, Jan Kara wrote:
> > > On Thu 18-08-11 10:36:10, Wu Fengguang wrote:
> > > > Subject: squeeze max-pause area and drop pass-good area
> > > > Date: Tue Aug 16 13:37:14 CST 2011
> > > >
> > > > Remove the pass-good area introduced in ffd1f609ab10 ("writeback:
> > > > introduce max-pause and pass-good dirty limits") and make the
> > > > max-pause area smaller and safe.
> > > >
> > > > This fixes ~30% performance regression in the ext3 data=writeback
> > > > fio_mmap_randwrite_64k/fio_mmap_randrw_64k test cases, where there are
> > > > 12 JBOD disks, on each disk runs 8 concurrent tasks doing reads+writes.
> > > >
> > > > Using deadline scheduler also has a regression, but not that big as
> > > > CFQ, so this suggests we have some write starvation.
> > > >
> > > > The test logs show that
> > > >
> > > > - the disks are sometimes under utilized
> > > >
> > > > - global dirty pages sometimes rush high to the pass-good area for
> > > > several hundred seconds, while in the mean time some bdi dirty pages
> > > > drop to very low value (bdi_dirty << bdi_thresh).
> > > > Then suddenly the global dirty pages dropped under global dirty
> > > > threshold and bdi_dirty rush very high (for example, 2 times higher
> > > > than bdi_thresh). During which time balance_dirty_pages() is not
> > > > called at all.
> > > >
> > > > So the problems are
> > > >
> > > > 1) The random writes progress so slow that they break the assumption of
> > > > the max-pause logic that "8 pages per 200ms is typically more than
> > > > enough to curb heavy dirtiers".
> > > >
> > > > 2) The max-pause logic ignored task_bdi_thresh and thus opens the
> > > > possibility for some bdi's to over dirty pages, leading to
> > > > (bdi_dirty >> bdi_thresh) and then (bdi_thresh >> bdi_dirty) for others.
> > > >
> > > > 3) The higher max-pause/pass-good thresholds somehow leads to some bad
> > > > swing of dirty pages.
> > > >
> > > > The fix is to allow the task to slightly dirty over task_bdi_thresh, but
> > > > no way to exceed bdi_dirty and/or global dirty_thresh.
> > > >
> > > > Tests show that it fixed the JBOD regression completely (both behavior
> > > > and performance), while still being able to cut down large pause times
> > > > in balance_dirty_pages() for single-disk cases.
> > > >
> > > > Reported-by: Li Shaohua <shaohua.li@...el.com>
> > > > Tested-by: Li Shaohua <shaohua.li@...el.com>
> > > > Signed-off-by: Wu Fengguang <fengguang.wu@...el.com>
> > > > ---
> > > > include/linux/writeback.h | 11 -----------
> > > > mm/page-writeback.c | 15 ++-------------
> > > > 2 files changed, 2 insertions(+), 24 deletions(-)
> > > >
> > > > --- linux.orig/mm/page-writeback.c 2011-08-18 09:52:59.000000000 +0800
> > > > +++ linux/mm/page-writeback.c 2011-08-18 10:28:57.000000000 +0800
> > > > @@ -786,21 +786,10 @@ static void balance_dirty_pages(struct a
> > > > * 200ms is typically more than enough to curb heavy dirtiers;
> > > > * (b) the pause time limit makes the dirtiers more responsive.
> > > > */
> > > > - if (nr_dirty < dirty_thresh +
> > > > - dirty_thresh / DIRTY_MAXPAUSE_AREA &&
> > > > + if (nr_dirty < dirty_thresh &&
> > > > + bdi_dirty < (task_bdi_thresh + bdi_thresh) / 2 &&
> > > > time_after(jiffies, start_time + MAX_PAUSE))
> > > > break;
> > > This looks definitely much safer than the original patch since we now
> > > always observe global dirty limit.
> >
> > Yeah.
> >
> > > I just wonder: We have throttled the
> > > task because bdi_nr_reclaimable > task_bdi_thresh.
> >
> > Not necessarily. It's possible (bdi_nr_reclaimable < task_bdi_thresh)
> > for the whole loop. And the 200ms pause that trigger the above test
> > may totally come from the io_schedule_timeout() calls.
> >
> > > Now in practice there
> > > should be some pages under writeback and this task should have submitted
> > > even more just a while ago. So the condition
> > > bdi_dirty < (task_bdi_thresh + bdi_thresh) / 2
> >
> > I guess the writeback_inodes_wb() call is irrelevant for the above
> > test, because writeback_inodes_wb() transfers reclaimable pages to
> > writeback pages, with the total bdi_dirty value staying the same.
> > Not to mention the fact that both the bdi_dirty and bdi_nr_reclaimable
> > variables have not been updated between writeback_inodes_wb() and the
> > max-pause test.
> Right, that comment was a bit off.
>
> > > looks still relatively weak. Shouldn't there be
> > > bdi_nr_reclaimable < (task_bdi_thresh + bdi_thresh) / 2?
> >
> > That's much easier condition to satisfy..
> Argh, sorry. I was mistaken by the name of the variable - I though it
> contains only dirty pages on the bdi but it also contains pages under
> writeback and bdi_nr_reclaimable is the one that contains only dirty pages.
Yeah the name may be a bit confusing.. but we'll soon get rid of bdi_nr_reclaimable :)
> So your patch does exactly what I had in mind. You can add:
> Acked-by: Jan Kara <jack@...e.cz>
Thanks! I'll test it in linux-next for a week and then send to Linus.
Thanks,
Fengguang
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists