[Cluster-devel] [PATCH v7 24/24] mm: Use memalloc_nofs_save in readahead path

[Matthew Wilcox <willy@infradead.org>]

From: "Matthew Wilcox (Oracle)" <willy@infradead.org>

Ensure that memory allocations in the readahead path do not attempt to
reclaim file-backed pages, which could lead to a deadlock.  It is
possible, though unlikely this is the root cause of a problem observed
by Cong Wang.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reported-by: Cong Wang <xiyou.wangcong@gmail.com>
Suggested-by: Michal Hocko <mhocko@suse.com>
---
 mm/readahead.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/mm/readahead.c b/mm/readahead.c
index bbe7208fcc2d..9fb5f77dcf69 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -22,6 +22,7 @@
 #include <linux/mm_inline.h>
 #include <linux/blk-cgroup.h>
 #include <linux/fadvise.h>
+#include <linux/sched/mm.h>
 
 #include "internal.h"
 
@@ -186,6 +187,18 @@ void page_cache_readahead_unbounded(struct address_space *mapping,
 	};
 	unsigned long i;
 
+	/*
+	 * Partway through the readahead operation, we will have added
+	 * locked pages to the page cache, but will not yet have submitted
+	 * them for I/O.  Adding another page may need to allocate memory,
+	 * which can trigger memory reclaim.  Telling the VM we're in
+	 * the middle of a filesystem operation will cause it to not
+	 * touch file-backed pages, preventing a deadlock.  Most (all?)
+	 * filesystems already specify __GFP_NOFS in their mapping's
+	 * gfp_mask, but let's be explicit here.
+	 */
+	unsigned int nofs = memalloc_nofs_save();
+
 	/*
 	 * Preallocate as many pages as we will need.
 	 */
@@ -230,6 +243,7 @@ void page_cache_readahead_unbounded(struct address_space *mapping,
 	 * will then handle the error.
 	 */
 	read_pages(&rac, &page_pool);
+	memalloc_nofs_restore(nofs);
 }
 EXPORT_SYMBOL_GPL(page_cache_readahead_unbounded);
 
-- 
2.25.0