linux-kernel - [PATCH v1 4/4] mm/compaction: optimize >0 order folio compaction by sorting source 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]

Message-ID: <20231113170157.280181-5-zi.yan@sent.com>
Date:   Mon, 13 Nov 2023 12:01:57 -0500
From:   Zi Yan <zi.yan@...t.com>
To:     linux-mm@...ck.org, linux-kernel@...r.kernel.org
Cc:     Zi Yan <ziy@...dia.com>, "Huang, Ying" <ying.huang@...el.com>,
        Ryan Roberts <ryan.roberts@....com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        David Hildenbrand <david@...hat.com>,
        "Yin, Fengwei" <fengwei.yin@...el.com>,
        Yu Zhao <yuzhao@...gle.com>, Vlastimil Babka <vbabka@...e.cz>,
        "Kirill A . Shutemov" <kirill.shutemov@...ux.intel.com>,
        Johannes Weiner <hannes@...xchg.org>,
        Baolin Wang <baolin.wang@...ux.alibaba.com>,
        Kemeng Shi <shikemeng@...weicloud.com>,
        Mel Gorman <mgorman@...hsingularity.net>,
        Rohan Puri <rohan.puri15@...il.com>,
        Mcgrof Chamberlain <mcgrof@...nel.org>,
        Adam Manzanares <a.manzanares@...sung.com>,
        "Vishal Moola (Oracle)" <vishal.moola@...il.com>
Subject: [PATCH v1 4/4] mm/compaction: optimize >0 order folio compaction by sorting source pages.

From: Zi Yan <ziy@...dia.com>

It should maximize high order free page use and minimize free page splits.
It might be useful before free page merging is implemented.

Signed-off-by: Zi Yan <ziy@...dia.com>
---
 mm/compaction.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/mm/compaction.c b/mm/compaction.c
index 9c083e6b399a..91809bee5422 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -127,6 +127,37 @@ static unsigned long release_free_list(struct page_list *freepages)
 }
 
 #ifdef CONFIG_COMPACTION
+
+static void sort_folios_by_order(struct list_head *pages)
+{
+	struct page_list page_list[MAX_ORDER + 1];
+	int order;
+	struct folio *folio, *next;
+
+	for (order = 0; order <= MAX_ORDER; order++)
+		init_page_list(&page_list[order]);
+
+	list_for_each_entry_safe(folio, next, pages, lru) {
+		order = folio_order(folio);
+
+		if (order > MAX_ORDER)
+			continue;
+
+		list_move(&folio->lru, &page_list[order].pages);
+		page_list[order].nr_pages++;
+	}
+
+	for (order = MAX_ORDER; order >= 0; order--) {
+		if (page_list[order].nr_pages) {
+
+			list_for_each_entry_safe(folio, next,
+						 &page_list[order].pages, lru) {
+				list_move_tail(&folio->lru, pages);
+			}
+		}
+	}
+}
+
 bool PageMovable(struct page *page)
 {
 	const struct movable_operations *mops;
@@ -2639,6 +2670,8 @@ compact_zone(struct compact_control *cc, struct capture_control *capc)
 				pageblock_start_pfn(cc->migrate_pfn - 1));
 		}
 
+		sort_folios_by_order(&cc->migratepages);
+
 		err = migrate_pages(&cc->migratepages, compaction_alloc,
 				compaction_free, (unsigned long)cc, cc->mode,
 				MR_COMPACTION, &nr_succeeded);
-- 
2.42.0