linux-kernel - Re: [RFC][PATCH] page reclaim throttle take2

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]

Date:	Tue, 26 Feb 2008 22:18:38 +0100
From:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
Cc:	linux-kernel@...r.kernel.org, linux-mm@...ck.org,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	Balbir Singh <balbir@...ux.vnet.ibm.com>,
	Rik van Riel <riel@...hat.com>,
	Lee Schermerhorn <Lee.Schermerhorn@...com>,
	Nick Piggin <npiggin@...e.de>
Subject: Re: [RFC][PATCH] page reclaim throttle take2


On Tue, 2008-02-26 at 11:32 +0900, KOSAKI Motohiro wrote:

> Index: b/include/linux/mmzone.h
> ===================================================================
> --- a/include/linux/mmzone.h	2008-02-25 21:37:49.000000000 +0900
> +++ b/include/linux/mmzone.h	2008-02-26 10:12:12.000000000 +0900
> @@ -335,6 +335,9 @@ struct zone {
>  	unsigned long		spanned_pages;	/* total size, including holes */
>  	unsigned long		present_pages;	/* amount of memory (excluding holes) */
>  
> +
> +	atomic_t		nr_reclaimers;
> +	wait_queue_head_t	reclaim_throttle_waitq;
>  	/*
>  	 * rarely used fields:
>  	 */

Small nit, that extra blank line seems at the wrong end of the text
block :-)

> Index: b/mm/vmscan.c
> ===================================================================
> --- a/mm/vmscan.c	2008-02-25 21:37:49.000000000 +0900
> +++ b/mm/vmscan.c	2008-02-26 10:59:38.000000000 +0900
> @@ -1252,6 +1252,55 @@ static unsigned long shrink_zone(int pri
>  	return nr_reclaimed;
>  }
>  
> +
> +#define RECLAIM_LIMIT (3)
> +
> +static int do_shrink_zone_throttled(int priority, struct zone *zone,
> +				    struct scan_control *sc,
> +				    unsigned long *ret_reclaimed)
> +{
> +	u64 start_time;
> +	int ret = 0;
> +
> +	start_time = jiffies_64;
> +
> +	wait_event(zone->reclaim_throttle_waitq,
> +		   atomic_add_unless(&zone->nr_reclaimers, 1, RECLAIM_LIMIT));
> +
> +	/* more reclaim until needed? */
> +	if (scan_global_lru(sc) &&
> +	    !(current->flags & PF_KSWAPD) &&
> +	    time_after64(jiffies, start_time + HZ/10)) {
> +		if (zone_watermark_ok(zone, sc->order, 4*zone->pages_high,
> +				      MAX_NR_ZONES-1, 0)) {
> +			ret = -EAGAIN;
> +			goto out;
> +		}
> +	}
> +
> +	*ret_reclaimed += shrink_zone(priority, zone, sc);
> +
> +out:
> +	atomic_dec(&zone->nr_reclaimers);
> +	wake_up_all(&zone->reclaim_throttle_waitq);
> +
> +	return ret;
> +}

Would it be possible - and worthwhile - to make this FIFO fair?

--
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/