linux-kernel - [PATCH v2] ext4: make dioread_nolock consistent in each mapping round

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20230207124136.88222-1-hanjinke.666@bytedance.com>
Date:   Tue,  7 Feb 2023 20:41:36 +0800
From:   Jinke Han <hanjinke.666@...edance.com>
To:     tytso@....edu, adilger.kernel@...ger.ca
Cc:     linux-ext4@...r.kernel.org, linux-kernel@...r.kernel.org,
        yi.zhang@...wei.com, Jinke Han <hanjinke.666@...edance.com>
Subject: [PATCH v2] ext4: make dioread_nolock consistent in each mapping round

From: Jinke Han <hanjinke.666@...edance.com>

When disable and enable dioread_nolock by remount, we may see
dioread_lock in ext4_do_writepages while see dioread_nolock in
mpage_map_one_extent. This inconsistency may triger the warning
in ext4_add_complete_io when the io_end->handle is NULL. Although
this warning is harmless in most cases, there is still a risk of
insufficient log reservation in conversion of unwritten extents.

Signed-off-by: Jinke Han <hanjinke.666@...edance.com>
Signed-off-by: Zhang Yi <yi.zhang@...wei.com>
---
 fs/ext4/inode.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 40579ef513b7..122a22ccddb3 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1568,6 +1568,7 @@ struct mpage_da_data {
 	struct ext4_io_submit io_submit;	/* IO submission data */
 	unsigned int do_map:1;
 	unsigned int scanned_until_end:1;
+	unsigned int dioread_nolock:1;
 };
 
 static void mpage_release_unused_pages(struct mpage_da_data *mpd,
@@ -2391,7 +2392,7 @@ static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
 	struct inode *inode = mpd->inode;
 	struct ext4_map_blocks *map = &mpd->map;
 	int get_blocks_flags;
-	int err, dioread_nolock;
+	int err, dioread_nolock = mpd->dioread_nolock;
 
 	trace_ext4_da_write_pages_extent(inode, map);
 	/*
@@ -2412,7 +2413,6 @@ static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
 	get_blocks_flags = EXT4_GET_BLOCKS_CREATE |
 			   EXT4_GET_BLOCKS_METADATA_NOFAIL |
 			   EXT4_GET_BLOCKS_IO_SUBMIT;
-	dioread_nolock = ext4_should_dioread_nolock(inode);
 	if (dioread_nolock)
 		get_blocks_flags |= EXT4_GET_BLOCKS_IO_CREATE_EXT;
 	if (map->m_flags & BIT(BH_Delay))
@@ -2727,10 +2727,11 @@ static int ext4_do_writepages(struct mpage_da_data *mpd)
 	handle_t *handle = NULL;
 	struct inode *inode = mpd->inode;
 	struct address_space *mapping = inode->i_mapping;
-	int needed_blocks, rsv_blocks = 0, ret = 0;
+	int needed_blocks, rsv_blocks = 0, rsv = 0, ret = 0;
 	struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb);
 	struct blk_plug plug;
 	bool give_up_on_write = false;
+	bool dioread_nolock;
 
 	trace_ext4_writepages(inode, wbc);
 
@@ -2783,15 +2784,6 @@ static int ext4_do_writepages(struct mpage_da_data *mpd)
 		ext4_journal_stop(handle);
 	}
 
-	if (ext4_should_dioread_nolock(inode)) {
-		/*
-		 * We may need to convert up to one extent per block in
-		 * the page and we may dirty the inode.
-		 */
-		rsv_blocks = 1 + ext4_chunk_trans_blocks(inode,
-						PAGE_SIZE >> inode->i_blkbits);
-	}
-
 	if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
 		range_whole = 1;
 
@@ -2837,6 +2829,18 @@ static int ext4_do_writepages(struct mpage_da_data *mpd)
 		goto unplug;
 
 	while (!mpd->scanned_until_end && wbc->nr_to_write > 0) {
+		dioread_nolock = ext4_should_dioread_nolock(inode);
+		if (!rsv && dioread_nolock) {
+			/*
+			 * We may need to convert up to one extent per block in
+			 * the page and we may dirty the inode.
+			 */
+			rsv = 1 + ext4_chunk_trans_blocks(inode,
+							PAGE_SIZE >> inode->i_blkbits);
+		}
+		rsv_blocks = dioread_nolock ? rsv : 0;
+		mpd->dioread_nolock = dioread_nolock;
+
 		/* For each extent of pages we use new io_end */
 		mpd->io_submit.io_end = ext4_init_io_end(inode, GFP_KERNEL);
 		if (!mpd->io_submit.io_end) {
-- 
2.20.1