linux-kernel - [RFC PATCH 3/3] mm, page_alloc: optimize batch count in free_pcppages

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] [day] [month] [year] [list]

Message-Id: <1366619590-31526-3-git-send-email-iamjoonsoo.kim@lge.com>
Date:	Mon, 22 Apr 2013 17:33:10 +0900
From:	Joonsoo Kim <iamjoonsoo.kim@....com>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	linux-kernel@...r.kernel.org, linux-mm@...ck.org,
	Minchan Kim <minchan@...nel.org>, Mel Gorman <mgorman@...e.de>,
	Joonsoo Kim <iamjoonsoo.kim@....com>
Subject: [RFC PATCH 3/3] mm, page_alloc: optimize batch count in free_pcppages_bulk()

If we use a division operation, we can compute a batch count more closed
to ideal value. With this value, we can remove at least one list in each
iteration. So we end the logic within MIGRATE_PCPTYPES iteration.

This makes logic more simple and understandable. In addition,
batching to free more pages may be helpful to cache usage.

Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@....com>

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index b212554..2632131 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -633,53 +633,71 @@ static inline int free_pages_check(struct page *page)
 static void free_pcppages_bulk(struct zone *zone, int count,
 					struct per_cpu_pages *pcp)
 {
-	int migratetype = 0;
-	int batch_free = 0;
+	struct page *page;
+	struct list_head *list;
 	int to_free = count;
+	int batch_free;
+	int mt, page_mt;
+	int nr_list;
+	int i;
+	bool all = false;
+
+	if (pcp->count == to_free)
+		all = true;
 
 	spin_lock(&zone->lock);
 	zone->all_unreclaimable = 0;
 	zone->pages_scanned = 0;
 
-	while (to_free) {
-		struct page *page;
-		struct list_head *list;
+redo:
+	/* Count non-empty list */
+	nr_list = 0;
+	for (mt = 0; mt < MIGRATE_PCPTYPES; mt++) {
+		list = &pcp->lists[mt];
+		if (!list_empty(list))
+			nr_list++;
+	}
 
-		/*
-		 * Remove pages from lists in a round-robin fashion. A
-		 * batch_free count is maintained that is incremented when an
-		 * empty list is encountered.  This is so more pages are freed
-		 * off fuller lists instead of spinning excessively around empty
-		 * lists
-		 */
-		do {
-			batch_free++;
-			if (++migratetype == MIGRATE_PCPTYPES)
-				migratetype = 0;
-			list = &pcp->lists[migratetype];
-		} while (list_empty(list));
+	/*
+	 * If there is only one non-empty list, free them all.
+	 * Otherwise, remove pages from lists in a round-robin fashion.
+	 * batch_free is set to remove at least one list.
+	 */
+	if (all || nr_list == 1)
+		batch_free = to_free;
+	else
+		batch_free = to_free / nr_list;
 
-		/* This is the only non-empty list. Free them all. */
-		if (batch_free == MIGRATE_PCPTYPES)
-			batch_free = to_free;
+	for (mt = 0; mt < MIGRATE_PCPTYPES; mt++) {
+		list = &pcp->lists[mt];
 
-		do {
-			int mt;	/* migratetype of the to-be-freed page */
+		for (i = 0; i < batch_free; i++) {
+			if (list_empty(list))
+				break;
 
+			to_free--;
 			page = list_entry(list->prev, struct page, lru);
+
 			/* must delete as __free_one_page list manipulates */
 			list_del(&page->lru);
-			mt = get_freepage_migratetype(page);
+			page_mt = get_freepage_migratetype(page);
 			/* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */
-			__free_one_page(page, zone, 0, mt);
-			trace_mm_page_pcpu_drain(page, 0, mt);
-			if (likely(!is_migrate_isolate_page(page))) {
-				__mod_zone_page_state(zone, NR_FREE_PAGES, 1);
-				if (is_migrate_cma(mt))
-					__mod_zone_page_state(zone, NR_FREE_CMA_PAGES, 1);
-			}
-		} while (--to_free && --batch_free && !list_empty(list));
+			__free_one_page(page, zone, 0, page_mt);
+			trace_mm_page_pcpu_drain(page, 0, page_mt);
+
+			if (unlikely(is_migrate_isolate_page(page)))
+				continue;
+
+			__mod_zone_page_state(zone, NR_FREE_PAGES, 1);
+			if (is_migrate_cma(page_mt))
+				__mod_zone_page_state(zone,
+						NR_FREE_CMA_PAGES, 1);
+		}
 	}
+
+	if (to_free)
+		goto redo;
+
 	spin_unlock(&zone->lock);
 }
 
-- 
1.7.9.5

--
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/