linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vlastimil Babka <vbabka@suse.cz>
To: Mel Gorman <mgorman@techsingularity.net>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linux-MM <linux-mm@kvack.org>
Cc: Rik van Riel <riel@surriel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis
Date: Wed, 15 Jun 2016 14:52:52 +0200	[thread overview]
Message-ID: <ce7c8bac-49ca-d8c0-05fd-95465dc2f65b@suse.cz> (raw)
In-Reply-To: <1465495483-11855-5-git-send-email-mgorman@techsingularity.net>

On 06/09/2016 08:04 PM, Mel Gorman wrote:
> This patch makes reclaim decisions on a per-node basis. A reclaimer knows
> what zone is required by the allocation request and skips pages from
> higher zones. In many cases this will be ok because it's a GFP_HIGHMEM
> request of some description. On 64-bit, ZONE_DMA32 requests will cause
> some problems but 32-bit devices on 64-bit platforms are increasingly
> rare. Historically it would have been a major problem on 32-bit with big
> Highmem:Lowmem ratios but such configurations are also now rare and even
> where they exist, they are not encouraged. If it really becomes a problem,
> it'll manifest as very low reclaim efficiencies.
>
> Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
> ---
>  mm/vmscan.c | 72 ++++++++++++++++++++++++++++++++++++++++---------------------
>  1 file changed, 47 insertions(+), 25 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index f87a5a0f8793..ab1b28e7e20a 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -84,6 +84,9 @@ struct scan_control {
>  	/* Scan (total_size >> priority) pages at once */
>  	int priority;
>
> +	/* The highest zone to isolate pages for reclaim from */
> +	enum zone_type reclaim_idx;
> +
>  	unsigned int may_writepage:1;
>
>  	/* Can mapped pages be reclaimed? */
> @@ -1369,6 +1372,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>  	struct list_head *src = &lruvec->lists[lru];
>  	unsigned long nr_taken = 0;
>  	unsigned long scan;
> +	LIST_HEAD(pages_skipped);
>
>  	for (scan = 0; scan < nr_to_scan && nr_taken < nr_to_scan &&
>  					!list_empty(src); scan++) {
> @@ -1379,6 +1383,11 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>
>  		VM_BUG_ON_PAGE(!PageLRU(page), page);
>
> +		if (page_zonenum(page) > sc->reclaim_idx) {
> +			list_move(&page->lru, &pages_skipped);
> +			continue;
> +		}
> +
>  		switch (__isolate_lru_page(page, mode)) {
>  		case 0:
>  			nr_taken += hpage_nr_pages(page);
> @@ -1395,6 +1404,15 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>  		}
>  	}
>
> +	/*
> +	 * Splice any skipped pages to the start of the LRU list. Note that
> +	 * this disrupts the LRU order when reclaiming for lower zones but
> +	 * we cannot splice to the tail. If we did then the SWAP_CLUSTER_MAX
> +	 * scanning would soon rescan the same pages to skip and put the
> +	 * system at risk of premature OOM.
> +	 */
> +	if (!list_empty(&pages_skipped))
> +		list_splice(&pages_skipped, src);

Hmm, that's unfortunate. But probably better than reclaiming the pages 
in the name of LRU order, even though it wouldn't help the allocation at 
hand.

[...]

> @@ -2516,14 +2535,14 @@ static inline bool compaction_ready(struct zone *zone, int order, int classzone_
>   * If a zone is deemed to be full of pinned pages then just give it a light
>   * scan then give up on it.
>   */
> -static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc)
> +static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc,
> +		enum zone_type classzone_idx)
>  {
>  	struct zoneref *z;
>  	struct zone *zone;
>  	unsigned long nr_soft_reclaimed;
>  	unsigned long nr_soft_scanned;
>  	gfp_t orig_mask;
> -	enum zone_type requested_highidx = gfp_zone(sc->gfp_mask);
>
>  	/*
>  	 * If the number of buffer_heads in the machine exceeds the maximum
> @@ -2536,15 +2555,15 @@ static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc)
>
>  	for_each_zone_zonelist_nodemask(zone, z, zonelist,
>  					gfp_zone(sc->gfp_mask), sc->nodemask) {
> -		enum zone_type classzone_idx;
> -
>  		if (!populated_zone(zone))
>  			continue;
>
> -		classzone_idx = requested_highidx;
>  		while (!populated_zone(zone->zone_pgdat->node_zones +
> -							classzone_idx))
> +							classzone_idx)) {
> +			sc->reclaim_idx--;
>  			classzone_idx--;
> +			continue;

Isn't this wrong to do this across whole zonelist which will contain 
multiple nodes? Example: a small node 0 without Normal zone will get us 
sc->reclaim_idx == classzone_idx == dma32. Node 1 won't have dma/dma32 
zones so we won't see classzone_idx populated, and the while loop will 
lead to underflow?

And sc->reclaim_idx seems to be unitialized when called via 
try_to_free_pages() -> do_try_to_free_pages() -> shrink_zones() ?
Which means it's zero and we underflow immediately?

> @@ -3207,15 +3228,14 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
>  			sc.may_writepage = 1;
>
>  		/*
> -		 * Now scan the zone in the dma->highmem direction, stopping
> -		 * at the last zone which needs scanning.
> -		 *
> -		 * We do this because the page allocator works in the opposite
> -		 * direction.  This prevents the page allocator from allocating
> -		 * pages behind kswapd's direction of progress, which would
> -		 * cause too much scanning of the lower zones.
> +		 * Continue scanning in the highmem->dma direction stopping at
> +		 * the last zone which needs scanning. This may reclaim lowmem
> +		 * pages that are not necessary for zone balancing but it
> +		 * preserves LRU ordering. It is assumed that the bulk of
> +		 * allocation requests can use arbitrary zones with the
> +		 * possible exception of big highmem:lowmem configurations.
>  		 */
> -		for (i = 0; i <= end_zone; i++) {
> +		for (i = end_zone; i >= end_zone; i--) {

				   i >= 0 ?

  reply	other threads:[~2016-06-15 12:52 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-09 18:04 [PATCH 00/27] Move LRU page reclaim from zones to nodes v6 Mel Gorman
2016-06-09 18:04 ` [PATCH 01/27] mm, vmstat: Add infrastructure for per-node vmstats Mel Gorman
2016-06-10 13:31   ` Vlastimil Babka
2016-06-10 13:47     ` Mel Gorman
2016-06-13 17:26   ` Christoph Lameter
2016-06-14 14:25     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 02/27] mm, vmscan: Move lru_lock to the node Mel Gorman
2016-06-10 16:39   ` Vlastimil Babka
2016-06-10 17:48     ` Peter Zijlstra
2016-06-14 14:41     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 03/27] mm, vmscan: Move LRU lists to node Mel Gorman
2016-06-10 18:00   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman
2016-06-15 12:52   ` Vlastimil Babka [this message]
2016-06-09 18:04 ` [PATCH 05/27] mm, vmscan: Have kswapd only scan based on the highest requested zone Mel Gorman
2016-06-15 13:13   ` Vlastimil Babka
2016-06-16  7:44     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 06/27] mm, vmscan: Make kswapd reclaim in terms of nodes Mel Gorman
2016-06-15 14:23   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 07/27] mm, vmscan: Remove balance gap Mel Gorman
2016-06-15 14:28   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 08/27] mm, vmscan: Simplify the logic deciding whether kswapd sleeps Mel Gorman
2016-06-15 15:18   ` Vlastimil Babka
2016-06-16  8:30     ` Mel Gorman
2016-06-16  9:16       ` Mel Gorman
2016-06-09 18:04 ` [PATCH 09/27] mm, vmscan: By default have direct reclaim only shrink once per node Mel Gorman
2016-06-16  8:59   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 10/27] mm, vmscan: Clear congestion, dirty and need for compaction on a per-node basis Mel Gorman
2016-06-16  9:29   ` Vlastimil Babka
2016-06-16 10:29     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 11/27] mm: vmscan: Do not reclaim from kswapd if there is any eligible zone Mel Gorman
2016-06-16 10:08   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 12/27] mm, vmscan: Make shrink_node decisions more node-centric Mel Gorman
2016-06-16 13:35   ` Vlastimil Babka
2016-06-16 14:47     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 13/27] mm, memcg: Move memcg limit enforcement from zones to nodes Mel Gorman
2016-06-16 15:06   ` Vlastimil Babka
2016-06-16 15:53     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 14/27] mm, workingset: Make working set detection node-aware Mel Gorman
2016-06-16 15:13   ` Vlastimil Babka
2016-06-16 15:56     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 15/27] mm, page_alloc: Consider dirtyable memory in terms of nodes Mel Gorman
2016-06-16 15:45   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 16/27] mm: Move page mapped accounting to the node Mel Gorman
2016-06-16 15:52   ` Vlastimil Babka
2016-06-16 16:04     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 17/27] mm: Rename NR_ANON_PAGES to NR_ANON_MAPPED Mel Gorman
2016-06-17  8:28   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 18/27] mm: Move most file-based accounting to the node Mel Gorman
2016-06-17  8:35   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 19/27] mm: Move vmscan writes and file write " Mel Gorman
2016-06-17  8:51   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 20/27] mm, vmscan: Update classzone_idx if buffer_heads_over_limit Mel Gorman
2016-06-17  8:53   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 21/27] mm, vmscan: Only wakeup kswapd once per node for the requested classzone Mel Gorman
2016-06-17 10:46   ` Vlastimil Babka
2016-06-17 12:03     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 22/27] mm: Convert zone_reclaim to node_reclaim Mel Gorman
2016-06-17 10:55   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 23/27] mm, vmscan: Add classzone information to tracepoints Mel Gorman
2016-06-17 10:57   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 24/27] mm, page_alloc: Remove fair zone allocation policy Mel Gorman
2016-06-17 11:27   ` Vlastimil Babka
2016-06-17 12:07     ` Mel Gorman
2016-06-09 18:04 ` [PATCH 25/27] mm: page_alloc: Cache the last node whose dirty limit is reached Mel Gorman
2016-06-17 11:31   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 26/27] mm: vmstat: Replace __count_zone_vm_events with a zone id equivalent Mel Gorman
2016-06-17 11:36   ` Vlastimil Babka
2016-06-09 18:04 ` [PATCH 27/27] mm: vmstat: Account per-zone stalls and pages skipped during reclaim Mel Gorman
2016-06-17 12:39   ` Vlastimil Babka
  -- strict thread matches above, loose matches on Subject: below --
2016-06-21 14:15 [PATCH 00/27] Move LRU page reclaim from zones to nodes v7 Mel Gorman
2016-06-21 14:15 ` [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman
2016-06-22 14:04   ` Vlastimil Babka
2016-06-22 16:00     ` Vlastimil Babka
2016-06-23 11:07       ` Mel Gorman
2016-06-23 11:13         ` Michal Hocko
2016-06-23 10:58     ` Mel Gorman
     [not found] <02ed01d1c47a$49fbfbc0$ddf3f340$@alibaba-inc.com>
2016-06-12  7:33 ` Hillf Danton
2016-06-14 14:47   ` Mel Gorman
2016-04-15  9:13 [PATCH 00/27] Move LRU page reclaim from zones to nodes v5 Mel Gorman
2016-04-15  9:13 ` [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ce7c8bac-49ca-d8c0-05fd-95465dc2f65b@suse.cz \
    --to=vbabka@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=riel@surriel.com \
    --subject='Re: [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).