From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail203.messagelabs.com (mail203.messagelabs.com [216.82.254.243]) by kanga.kvack.org (Postfix) with SMTP id 347736B00BA for ; Wed, 1 Dec 2010 20:36:30 -0500 (EST) Received: by iwn41 with SMTP id 41so690322iwn.14 for ; Wed, 01 Dec 2010 17:36:28 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: <1291172911.12777.58.camel@sli10-conroe> <20101201132730.ABC2.A69D9226@jp.fujitsu.com> <20101201155854.GA3372@barrios-desktop> <1291249749.12777.86.camel@sli10-conroe> <1291251908.12777.94.camel@sli10-conroe> Date: Thu, 2 Dec 2010 10:36:27 +0900 Message-ID: Subject: Re: [patch]vmscan: make kswapd use a correct order From: Minchan Kim Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-linux-mm@kvack.org To: Shaohua Li Cc: KOSAKI Motohiro , linux-mm , Andrew Morton , Mel Gorman List-ID: Where is my mail? I will resend lost content. On Thu, Dec 2, 2010 at 10:23 AM, Minchan Kim wrote: > On Thu, Dec 2, 2010 at 10:05 AM, Shaohua Li wrote: >> On Thu, 2010-12-02 at 08:54 +0800, Minchan Kim wrote: >>> On Thu, Dec 2, 2010 at 9:29 AM, Shaohua Li wrote= : >>> > On Wed, 2010-12-01 at 23:58 +0800, Minchan Kim wrote: >>> >> On Wed, Dec 01, 2010 at 06:44:27PM +0900, KOSAKI Motohiro wrote: >>> >> > > T0: Task1 wakeup_kswapd(order=3D3) >>> >> > > T1: kswapd enters balance_pgdat >>> >> > > T2: Task2 wakeup_kswapd(order=3D2), because pages reclaimed by k= swapd are used >>> >> > > quickly >>> >> > > T3: kswapd exits balance_pgdat. kswapd will do check. Now new or= der=3D2, >>> >> > > pgdat->kswapd_max_order will become 0, but order=3D3, if sleepin= g_prematurely, >>> >> > > then order will become pgdat->kswapd_max_order(0), while at this= time the >>> >> > > order should 2 >>> >> > > This isn't a big deal, but we do have a small window the order i= s wrong. >>> >> > > >>> >> > > Signed-off-by: Shaohua Li >>> >> > > >>> >> > > diff --git a/mm/vmscan.c b/mm/vmscan.c >>> >> > > index d31d7ce..15cd0d2 100644 >>> >> > > --- a/mm/vmscan.c >>> >> > > +++ b/mm/vmscan.c >>> >> > > @@ -2450,7 +2450,7 @@ static int kswapd(void *p) >>> >> > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 } >>> >> > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 } >>> >> > > >>> >> > > - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order =3D pgdat->kswapd_max_or= der; >>> >> > > + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order =3D max_t(unsigned long,= new_order, pgdat->kswapd_max_order); >>> >> > > =A0 =A0 =A0 =A0 =A0 } >>> >> > > =A0 =A0 =A0 =A0 =A0 finish_wait(&pgdat->kswapd_wait, &wait); >>> >> > >>> >> > Good catch! >>> >> > >>> >> > But unfortunatelly, the code is not correct. At least, don't fit c= orrent >>> >> > design. >>> >> > >>> >> > 1) if "order < new_order" condition is false, we already decided t= o don't >>> >> > =A0 =A0use new_order. So, we shouldn't use new_order after kswapd_= try_to_sleep() >>> >> > 2) if sleeping_prematurely() return false, it probably mean >>> >> > =A0 =A0zone_watermark_ok_safe(zone, order, high_wmark) return fals= e. >>> >> > =A0 =A0therefore, we have to retry reclaim by using old 'order' pa= rameter. >>> >> >>> >> Good catch, too. >>> >> >>> >> In Shaohua's scenario, if Task1 gets the order-3 page after kswapd's= reclaiming, >>> >> it's no problem. >>> >> But if Task1 doesn't get the order-3 page and others used the order-= 3 page for Task1, >>> >> Kswapd have to reclaim order-3 for Task1, again. >>> > why? it's just a possibility. Task1 might get its pages too. If Task1 >>> > doesn't get its pages, it will wakeup kswapd too with its order. >>> > >>> >> In addtion, new order is always less than old order in that context. >>> >> so big order page reclaim makes much safe for low order pages. >>> > big order page reclaim makes we have more chances to reclaim useful >>> > pages by lumpy, why it's safe? >>> >>> For example, It assume tat Task1 continues to fail get the order-3 >>> page of GFP_ATOMIC since other tasks continues to allocate order-2 >>> pages so that they steal pages. >> but even you reclaim order-3, you can't guarantee task1 can get the >> pages too. order-3 page can be steal by order-2 allocation > > But at least, it has a high possibility to allocate order-3 page than > reclaim order-2 pages. > >> >>> Then, your patch makes continue to >>> reclaim order-2 page in this scenario. Task1 never get the order-3 >>> pages if it doesn't have a merge luck. >> Task1 will wakeup kswapd again for order-3, so kswapd will reclaim >> order-3 very soon after the order-2 reclaim. > > GFP_ATOMIC case doesn't wakeup kswapd. > When kswapd wakeup by order-3 depends on caller's retry. > And this situation can be repeated in next turn. > > We can't guarantee 100% Task-1's order-3 pages so I hope we should go > way to reduce the problem as possible as we can. > >> >> > > > > -- > Kind regards, > Minchan Kim > --=20 Kind regards, Minchan Kim -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/ Don't email: email@kvack.org