Received-SPF: pass (google.com: domain of linux-kernel+bounces-93317-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
From: Johannes Weiner <hannes@cmpxchg.org>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Vlastimil Babka <vbabka@suse.cz>,
	Mel Gorman <mgorman@techsingularity.net>,
	Zi Yan <ziy@nvidia.com>,
	Mike Kravetz <mike.kravetz@oracle.com>,
	"Huang, Ying" <ying.huang@intel.com>,
	David Hildenbrand <david@redhat.com>,
	linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH 05/10] mm: page_alloc: fix move_freepages_block() range error
Date: Tue,  5 Mar 2024 23:08:36 -0500
Message-ID: <20240306041526.892167-6-hannes@cmpxchg.org>
In-Reply-To: <20240306041526.892167-1-hannes@cmpxchg.org>
References: <20240306041526.892167-1-hannes@cmpxchg.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

When a block is partially outside the zone of the cursor page, the
function cuts the range to the pivot page instead of the zone
start. This can leave large parts of the block behind, which
encourages incompatible page mixing down the line (ask for one type,
get another), and thus long-term fragmentation.

This triggers reliably on the first block in the DMA zone, whose
start_pfn is 1. The block is stolen, but everything before the pivot
page (which was often hundreds of pages) is left on the old list.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
---
 mm/page_alloc.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 0f385531e130..787ae3f0ac06 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1645,9 +1645,15 @@ int move_freepages_block(struct zone *zone, struct page *page,
 	start_pfn = pageblock_start_pfn(pfn);
 	end_pfn = pageblock_end_pfn(pfn) - 1;
 
-	/* Do not cross zone boundaries */
+	/*
+	 * The caller only has the lock for @zone, don't touch ranges
+	 * that straddle into other zones. While we could move part of
+	 * the range that's inside the zone, this call is usually
+	 * accompanied by other operations such as migratetype updates
+	 * which also should be locked.
+	 */
 	if (!zone_spans_pfn(zone, start_pfn))
-		start_pfn = pfn;
+		return 0;
 	if (!zone_spans_pfn(zone, end_pfn))
 		return 0;
 
-- 
2.44.0