linux-kernel - Re: [PATCH v3] exfat: check disk status during buffer write

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: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <036c4601c7684a67b8caedef9917b929@BJMBX02.spreadtrum.com>
Date: Wed, 31 Jul 2024 02:31:27 +0000
From: 崔东亮 (Dongliang Cui) <Dongliang.Cui@...soc.com>
To: "linkinjeon@...nel.org" <linkinjeon@...nel.org>,
        "sj1557.seo@...sung.com"
	<sj1557.seo@...sung.com>,
        "hch@...radead.org" <hch@...radead.org>,
        "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
CC: "niuzhiguo84@...il.com" <niuzhiguo84@...il.com>,
        王皓 (Hao_hao Wang) <Hao_hao.Wang@...soc.com>,
        王科 (Ke Wang) <Ke.Wang@...soc.com>,
        牛志国 (Zhiguo Niu) <Zhiguo.Niu@...soc.com>,
        "；cuidongliang390@...il.com
	<o"<.cuidongliang390@...il.com>
Subject: Re: [PATCH v3] exfat: check disk status during buffer write

Loop more

-----邮件原件-----
发件人: 崔东亮 (Dongliang Cui) <Dongliang.Cui@...soc.com> 
发送时间: 2024年7月31日 10:27
收件人: linkinjeon@...nel.org; sj1557.seo@...sung.com; hch@...radead.org; linux-fsdevel@...r.kernel.org; linux-kernel@...r.kernel.org
抄送: niuzhiguo84@...il.com; 王皓 (Hao_hao Wang) <Hao_hao.Wang@...soc.com>; 王科 (Ke Wang) <Ke.Wang@...soc.com>; 崔东亮 (Dongliang Cui) <Dongliang.Cui@...soc.com>; 牛志国 (Zhiguo Niu) <Zhiguo.Niu@...soc.com>
主题: [PATCH v3] exfat: check disk status during buffer write

We found that when writing a large file through buffer write, if the disk is inaccessible, exFAT does not return an error normally, which leads to the writing process not stopping properly.

To easily reproduce this issue, you can follow the steps below:

1. format a device to exFAT and then mount (with a full disk erase) 2. dd if=/dev/zero of=/exfat_mount/test.img bs=1M count=8192 3. eject the device

You may find that the dd process does not stop immediately and may continue for a long time.

The root cause of this issue is that during buffer write process, exFAT does not need to access the disk to look up directory entries or the FAT table (whereas FAT would do) every time data is written.
Instead, exFAT simply marks the buffer as dirty and returns, delegating the writeback operation to the writeback process.

If the disk cannot be accessed at this time, the error will only be returned to the writeback process, and the original process will not receive the error, so it cannot be returned to the user side.

When the disk cannot be accessed normally, an error should be returned to stop the writing process.

Signed-off-by: Dongliang Cui <dongliang.cui@...soc.com>
Signed-off-by: Zhiguo Niu <zhiguo.niu@...soc.com>
---
Changes in v3:
 - Implement .shutdown to monitor disk status.
---
 fs/exfat/exfat_fs.h | 10 ++++++++++
 fs/exfat/inode.c    |  3 +++
 fs/exfat/super.c    | 11 +++++++++++
 3 files changed, 24 insertions(+)

diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index ecc5db952deb..c6cf36070aa3 100644
--- a/fs/exfat/exfat_fs.h
+++ b/fs/exfat/exfat_fs.h
@@ -148,6 +148,9 @@ enum {
 #define DIR_CACHE_SIZE		\
 	(DIV_ROUND_UP(EXFAT_DEN_TO_B(ES_MAX_ENTRY_NUM), SECTOR_SIZE) + 1)
 
+/* Superblock flags */
+#define EXFAT_FLAGS_SHUTDOWN	1
+
 struct exfat_dentry_namebuf {
 	char *lfn;
 	int lfnbuf_len; /* usually MAX_UNINAME_BUF_SIZE */ @@ -267,6 +270,8 @@ struct exfat_sb_info {
 	unsigned int clu_srch_ptr; /* cluster search pointer */
 	unsigned int used_clusters; /* number of used clusters */
 
+	unsigned long s_exfat_flags; /* Exfat superblock flags */
+
 	struct mutex s_lock; /* superblock lock */
 	struct mutex bitmap_lock; /* bitmap lock */
 	struct exfat_mount_options options;
@@ -338,6 +343,11 @@ static inline struct exfat_inode_info *EXFAT_I(struct inode *inode)
 	return container_of(inode, struct exfat_inode_info, vfs_inode);  }
 
+static inline int exfat_forced_shutdown(struct super_block *sb) {
+	return test_bit(EXFAT_FLAGS_SHUTDOWN, &EXFAT_SB(sb)->s_exfat_flags); }
+
 /*
  * If ->i_mode can't hold 0222 (i.e. ATTR_RO), we use ->i_attrs to
  * save ATTR_RO instead of ->i_mode.
diff --git a/fs/exfat/inode.c b/fs/exfat/inode.c index dd894e558c91..b1b814183494 100644
--- a/fs/exfat/inode.c
+++ b/fs/exfat/inode.c
@@ -452,6 +452,9 @@ static int exfat_write_begin(struct file *file, struct address_space *mapping,  {
 	int ret;
 
+	if (unlikely(exfat_forced_shutdown(mapping->host->i_sb)))
+		return -EIO;
+
 	*pagep = NULL;
 	ret = block_write_begin(mapping, pos, len, pagep, exfat_get_block);
 
diff --git a/fs/exfat/super.c b/fs/exfat/super.c index 323ecebe6f0e..9d7d9c4ba55a 100644
--- a/fs/exfat/super.c
+++ b/fs/exfat/super.c
@@ -167,6 +167,16 @@ static int exfat_show_options(struct seq_file *m, struct dentry *root)
 	return 0;
 }
 
+static void exfat_shutdown(struct super_block *sb) {
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	if (exfat_forced_shutdown(sb))
+		return;
+
+	set_bit(EXFAT_FLAGS_SHUTDOWN, &sbi->s_exfat_flags); }
+
 static struct inode *exfat_alloc_inode(struct super_block *sb)  {
 	struct exfat_inode_info *ei;
@@ -193,6 +203,7 @@ static const struct super_operations exfat_sops = {
 	.sync_fs	= exfat_sync_fs,
 	.statfs		= exfat_statfs,
 	.show_options	= exfat_show_options,
+	.shutdown	= exfat_shutdown,
 };
 
 enum {
--
2.25.1