[<prev] [next>] [day] [month] [year] [list]
Message-Id: <20240920122030.159751-1-gxxa03070307@gmail.com>
Date: Fri, 20 Sep 2024 20:20:30 +0800
From: Xiang Gao <gxxa03070307@...il.com>
To: akpm@...ux-foundation.org
Cc: linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
gaoxiang17 <gaoxiang17@...omi.com>
Subject: [PATCH] mm/page_alloc: Add some detailed comments in can_steal_fallback
From: gaoxiang17 <gaoxiang17@...omi.com>
Signed-off-by: gaoxiang17 <gaoxiang17@...omi.com>
---
mm/page_alloc.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 0aefae4a26b2..8b0634d1728b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1818,6 +1818,13 @@ static bool can_steal_fallback(unsigned int order, int start_mt)
if (order >= pageblock_order)
return true;
+ /*
+ * movable pages won't cause permanent fragmentation, so when you alloc small pages,
+ * you just need to temporarily steal unmovable or reclaimable pages that are closest
+ * to the request size. After a while, memory compact may occur to form large contiguous
+ * pages, and the next movable allocation may not need to steal. Unmovable and reclaimable
+ * allocation need to actually steal pages.
+ */
if (order >= pageblock_order / 2 ||
start_mt == MIGRATE_RECLAIMABLE ||
start_mt == MIGRATE_UNMOVABLE ||
--
2.34.1
Powered by blists - more mailing lists