linux-kernel - Re: [PATCH 7/7] vmscan: isolated_lru_pages() stop neighbor search if neighbor can't be isolated

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]

Message-ID: <20100805154045.GE2985@barrios-desktop>
Date:	Fri, 6 Aug 2010 00:40:45 +0900
From:	Minchan Kim <minchan.kim@...il.com>
To:	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
Cc:	LKML <linux-kernel@...r.kernel.org>, linux-mm <linux-mm@...ck.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Johannes Weiner <hannes@...xchg.org>,
	Mel Gorman <mel@....ul.ie>,
	Wu Fengguang <fengguang.wu@...el.com>,
	Rik van Riel <riel@...hat.com>
Subject: Re: [PATCH 7/7] vmscan: isolated_lru_pages() stop neighbor search
 if neighbor can't be isolated

On Thu, Aug 05, 2010 at 03:16:06PM +0900, KOSAKI Motohiro wrote:
> isolate_lru_pages() doesn't only isolate LRU tail pages, but also
> isolate neighbor pages of the eviction page.
> 
> Now, the neighbor search don't stop even if neighbors can't be isolated.
> It is silly. successful higher order allocation need full contenious
> memory, even though only one page reclaim failure mean to fail making
> enough contenious memory.
> 
> Then, isolate_lru_pages() should stop to search PFN neighbor pages and
> try to search next page on LRU soon. This patch does it. Also all of
> lumpy reclaim failure account nr_lumpy_failed.
> 
> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>

I agree this patch. 
But I have a one question. 

> ---
>  mm/vmscan.c |   24 ++++++++++++++++--------
>  1 files changed, 16 insertions(+), 8 deletions(-)
> 
<snip>

  
>  			if (__isolate_lru_page(cursor_page, mode, file) == 0) {
>  				list_move(&cursor_page->lru, dst);
> @@ -1074,9 +1080,11 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>  					nr_lumpy_dirty++;
>  				scan++;
>  			} else {
> -				if (mode == ISOLATE_BOTH &&
> -						page_count(cursor_page))
> -					nr_lumpy_failed++;

sc->order = 1;
shrink_inactive_list;
isolate_pages_global with ISOLATE_INACTIVE(I mean no lumpy relcaim mode);
lumpy relcaim in inactive list in isolate_lru_pages;
(But I am not sure we can call it as lumpy reclaim. but at lesat I think 
it a part of lumpy reclaim)
I mean it can reclaim physical pfn order not LRU order in inactive list since
it only consider sc->order.  Is it a intentional?

I guess it's intentional since we care of ISOLATE_BOTH when we increase nr_lumpy_failed. 
If it is, Shouldn't we care of ISOLATE_BOTH still?


> +				/* the page is freed already. */
> +				if (!page_count(cursor_page))
> +					continue;
> +				nr_lumpy_failed++;
> +				break;
>  			}
>  		}
>  	}
> -- 
> 1.6.5.2
> 


-- 
Kind regards,
Minchan Kim
--
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/