linux-kernel - Re: [PATCH 2/2] mm: get_scan_count consider reclaimable lru pages

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:	Wed, 27 Jul 2016 16:13:33 +0100
From:	Mel Gorman <mgorman@...e.de>
To:	Minchan Kim <minchan@...nel.org>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Johannes Weiner <hannes@...xchg.org>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] mm: get_scan_count consider reclaimable lru pages

On Wed, Jul 27, 2016 at 03:22:26PM +0100, Mel Gorman wrote:
> ---8<---
> From: Mel Gorman <mgorman@...hsingularity.net>
> Subject: [PATCH] mm, vmscan: Wait on a waitqueue when too many pages are
>  isolated
> 

This is potentially a much better version as it avoids wakeup storms and
do a better job of handling the case where pages could not be reclaimed.

---8<---
mm, vmscan: Wait on a waitqueue when too many pages are isolated

When too many pages are isolated, direct reclaim waits on congestion to
clear for up to a tenth of a second. There is no reason to believe that too
many pages are isolated due to dirty pages, reclaim efficiency or congestion.
It may simply be because an extremely large number of processes have entered
direct reclaim at the same time.

This patch has processes wait on a waitqueue when too many pages are
isolated.  When parallel reclaimers finish shrink_page_list, they wake the
waiters to recheck whether too many pages are isolated. While it is difficult
to trigger this corner case, it's possible by lauching an extremely large
number of hackbench processes on a 32-bit system with limited memory. Without
the patch, a large number of processes wait uselessly and with the patch
applied, I was unable to stall the system.

Signed-off-by: Mel Gorman <mgorman@...hsingularity.net>
---
 include/linux/mmzone.h |  1 +
 mm/page_alloc.c        |  1 +
 mm/vmscan.c            | 24 +++++++++++++++---------
 3 files changed, 17 insertions(+), 9 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index d572b78b65e1..467878d7af33 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -653,6 +653,7 @@ typedef struct pglist_data {
 	int node_id;
 	wait_queue_head_t kswapd_wait;
 	wait_queue_head_t pfmemalloc_wait;
+	wait_queue_head_t isolated_wait;
 	struct task_struct *kswapd;	/* Protected by
 					   mem_hotplug_begin/end() */
 	int kswapd_order;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index fbd329e61bf6..3800972f240e 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5859,6 +5859,7 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat)
 #endif
 	init_waitqueue_head(&pgdat->kswapd_wait);
 	init_waitqueue_head(&pgdat->pfmemalloc_wait);
+	init_waitqueue_head(&pgdat->isolated_wait);
 #ifdef CONFIG_COMPACTION
 	init_waitqueue_head(&pgdat->kcompactd_wait);
 #endif
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 9c0b2b0fc164..e264fcb7556b 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1554,16 +1554,16 @@ int isolate_lru_page(struct page *page)
  * the LRU list will go small and be scanned faster than necessary, leading to
  * unnecessary swapping, thrashing and OOM.
  */
-static int too_many_isolated(struct pglist_data *pgdat, int file,
+static bool safe_to_isolate(struct pglist_data *pgdat, int file,
 		struct scan_control *sc)
 {
 	unsigned long inactive, isolated;
 
 	if (current_is_kswapd())
-		return 0;
+		return true;
 
-	if (!sane_reclaim(sc))
-		return 0;
+	if (sane_reclaim(sc))
+		return true;
 
 	if (file) {
 		inactive = node_page_state(pgdat, NR_INACTIVE_FILE);
@@ -1581,7 +1581,7 @@ static int too_many_isolated(struct pglist_data *pgdat, int file,
 	if ((sc->gfp_mask & (__GFP_IO | __GFP_FS)) == (__GFP_IO | __GFP_FS))
 		inactive >>= 3;
 
-	return isolated > inactive;
+	return isolated < inactive;
 }
 
 static noinline_for_stack void
@@ -1701,12 +1701,15 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
 	if (!inactive_reclaimable_pages(lruvec, sc, lru))
 		return 0;
 
-	while (unlikely(too_many_isolated(pgdat, file, sc))) {
-		congestion_wait(BLK_RW_ASYNC, HZ/10);
+	if (!safe_to_isolate(pgdat, file, sc)) {
+		wait_event_killable(pgdat->isolated_wait,
+			safe_to_isolate(pgdat, file, sc));
 
 		/* We are about to die and free our memory. Return now. */
-		if (fatal_signal_pending(current))
-			return SWAP_CLUSTER_MAX;
+		if (fatal_signal_pending(current)) {
+			nr_reclaimed = SWAP_CLUSTER_MAX;
+			goto out;
+		}
 	}
 
 	lru_add_drain();
@@ -1819,6 +1822,9 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
 	trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
 			nr_scanned, nr_reclaimed,
 			sc->priority, file);
+
+out:
+	wake_up(&pgdat->isolated_wait);
 	return nr_reclaimed;
 }