From 4ac646153cb468a5128332a70d715cfea093e397 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Wed, 27 Sep 2017 05:45:36 -0600
Subject: [PATCH] buffer: grow_dev_page() should use __GFP_NOFAIL for all cases
References: bnc#971975 VM performance -- memory usage during reclaim
Patch-mainline: v4.15-rc1
Git-commit: 94dc24c0c59a224a093f110060d01c2c620f275a
We currently use it for find_or_create_page(), which means that it
cannot fail. Ensure we also pass in 'retry == true' to
alloc_page_buffers(), which also ensure that it cannot fail.
After this, there are no failure cases in grow_dev_page() that
occur because of a failed memory allocation.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
fs/buffer.c | 6 +-----
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/fs/buffer.c b/fs/buffer.c
index 497eb36d7346..c4d9d65ebe77 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -991,8 +991,6 @@ grow_dev_page(struct block_device *bdev, sector_t block,
gfp_mask |= __GFP_NOFAIL;
page = find_or_create_page(inode->i_mapping, index, gfp_mask);
- if (!page)
- return ret;
BUG_ON(!PageLocked(page));
@@ -1011,9 +1009,7 @@ grow_dev_page(struct block_device *bdev, sector_t block,
/*
* Allocate some buffers for this page
*/
- bh = alloc_page_buffers(page, size, false);
- if (!bh)
- goto failed;
+ bh = alloc_page_buffers(page, size, true);
/*
* Link the page to the buffers and initialise them. Take the