linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@suse.de>
To: Linux-MM <linux-mm@kvack.org>
Cc: Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Michal Hocko <mhocko@suse.cz>,
	LKML <linux-kernel@vger.kernel.org>, Mel Gorman <mgorman@suse.de>
Subject: [PATCH 25/25] mm: page_alloc: Take fewer passes when allocating to the low watermark
Date: Mon,  8 Jun 2015 14:56:31 +0100	[thread overview]
Message-ID: <1433771791-30567-26-git-send-email-mgorman@suse.de> (raw)
In-Reply-To: <1433771791-30567-1-git-send-email-mgorman@suse.de>

Signed-off-by: Mel Gorman <mgorman@suse.de>
---
 mm/page_alloc.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 886102cc9b09..58f6330ec3e2 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1927,6 +1927,27 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			if (alloc_flags & ALLOC_NO_WATERMARKS)
 				goto try_this_zone;
 
+			/*
+			 * If checking the low watermark, see if we meet the
+			 * min watermark and if so, try the zone and wake
+			 * kswapd instead of falling back to a remote zone
+			 * or having to take a second pass
+			 */
+			if (alloc_flags & ALLOC_WMARK_LOW) {
+				int min_flags = alloc_flags;
+
+				min_flags &= ~ALLOC_WMARK_LOW;
+				min_flags |= ALLOC_WMARK_MIN;
+
+				if (zone_watermark_ok(zone, order,
+						zone->watermark[min_flags & ALLOC_WMARK_MASK],
+						ac->classzone_idx,
+						min_flags)) {
+					wakeup_kswapd(zone, order, ac->classzone_idx);
+					goto try_this_zone;
+				}
+			}
+
 			if (node_reclaim_mode == 0 ||
 			    !zone_allows_reclaim(ac->preferred_zone, zone))
 				goto this_zone_full;
-- 
2.3.5

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2015-06-08 13:57 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-08 13:56 [RFC PATCH 00/25] Move LRU page reclaim from zones to nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 01/25] mm, vmstat: Add infrastructure for per-node vmstats Mel Gorman
2015-06-08 13:56 ` [PATCH 02/25] mm, vmscan: Move lru_lock to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 03/25] mm, vmscan: Move LRU lists to node Mel Gorman
2015-06-08 13:56 ` [PATCH 04/25] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman
2015-06-08 13:56 ` [PATCH 05/25] mm, vmscan: Have kswapd only scan based on the highest requested zone Mel Gorman
2015-06-08 13:56 ` [PATCH 06/25] mm, vmscan: Avoid a second search through zones checking if compaction is required Mel Gorman
2015-06-08 13:56 ` [PATCH 07/25] mm, vmscan: Make kswapd think of reclaim in terms of nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 08/25] mm, vmscan: By default have direct reclaim only shrink once per node Mel Gorman
2015-06-08 13:56 ` [PATCH 09/25] mm, vmscan: Clear congestion, dirty and need for compaction on a per-node basis Mel Gorman
2015-06-08 13:56 ` [PATCH 10/25] mm, vmscan: Make shrink_node decisions more node-centric Mel Gorman
2015-06-08 13:56 ` [PATCH 11/25] mm, workingset: Make working set detection node-aware Mel Gorman
2015-06-08 13:56 ` [PATCH 12/25] mm, page_alloc: Consider dirtyable memory in terms of nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 13/25] mm: Move NR_FILE_MAPPED accounting to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 14/25] mm: Rename NR_ANON_PAGES to NR_ANON_MAPPED Mel Gorman
2015-06-08 13:56 ` [PATCH 15/25] mm: Move most file-based accounting to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 16/25] mm, vmscan: Update classzone_idx if buffer_heads_over_limit Mel Gorman
2015-06-08 13:56 ` [PATCH 17/25] mm, vmscan: Check if cpusets are enabled during direct reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 18/25] mm, vmscan: Only wakeup kswapd once per node for the requested classzone Mel Gorman
2015-06-08 13:56 ` [PATCH 19/25] mm, vmscan: Account in vmstat for pages skipped during reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 20/25] mm, page_alloc: Remove fair zone allocation policy Mel Gorman
2015-06-08 13:56 ` [PATCH 21/25] mm, page_alloc: Defer zlc_setup until it is known it is required Mel Gorman
2015-06-08 13:56 ` [PATCH 22/25] mm: Convert zone_reclaim to node_reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 23/25] mm, page_alloc: Delete the zonelist_cache Mel Gorman
2015-06-08 13:56 ` [PATCH 24/25] mm, page_alloc: Use ac->classzone_idx instead of zone_idx(preferred_zone) Mel Gorman
2015-06-08 13:56 ` Mel Gorman [this message]
2015-06-19 17:01 ` [RFC PATCH 00/25] Move LRU page reclaim from zones to nodes Johannes Weiner
2015-06-21 14:04   ` 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=1433771791-30567-26-git-send-email-mgorman@suse.de \
    --to=mgorman@suse.de \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.cz \
    --cc=riel@redhat.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox