This patch introduces a mechanism to periodically check and update
the superblock within the ext4 file system. The main purpose of this
patch is to keep the disk superblock up to date. The update will be
performed if more than one hour has passed since the last update, and
if more than 16MB of data have been written to disk.
This check and update is performed within the ext4_journal_commit_callback
function, ensuring that the superblock is written while the disk is
active, rather than based on a timer that may trigger during disk idle
periods.
Discussion https://www.spinics.net/lists/linux-ext4/msg85865.html
Signed-off-by: Vitaliy Kuznetsov <[email protected]>
---
fs/ext4/super.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++++
fs/ext4/sysfs.c | 4 ++--
2 files changed, 54 insertions(+), 2 deletions(-)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index c94ebf704616..2159e9705404 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -433,6 +433,57 @@ static time64_t __ext4_get_tstamp(__le32 *lo, __u8 *hi)
#define ext4_get_tstamp(es, tstamp) \
__ext4_get_tstamp(&(es)->tstamp, &(es)->tstamp ## _hi)
+#define EXT4_SB_REFRESH_INTERVAL_SEC (3600) /* seconds (1 hour) */
+#define EXT4_SB_REFRESH_INTERVAL_KB (16384) /* kilobytes (16MB) */
+
+/*
+ * The ext4_maybe_update_superblock() function checks and updates the
+ * superblock if needed.
+ *
+ * This function is designed to update the on-disk superblock only under
+ * certain conditions to prevent excessive disk writes and unnecessary
+ * waking of the disk from sleep. The superblock will be updated if:
+ * 1. More than an hour has passed since the last superblock update, and
+ * 2. More than 16MB have been written since the last superblock update.
+ *
+ * @sb: The superblock
+ */
+static void ext4_maybe_update_superblock(struct super_block *sb)
+{
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct ext4_super_block *es = sbi->s_es;
+ journal_t *journal = sbi->s_journal;
+ time64_t now;
+ __u64 last_update;
+ __u64 lifetime_write_kbytes;
+ __u64 diff_size;
+
+ if (sb_rdonly(sb) || !(sb->s_flags & SB_ACTIVE) ||
+ !journal || (journal->j_flags & JBD2_UNMOUNT))
+ return;
+
+ now = ktime_get_real_seconds();
+ last_update = ext4_get_tstamp(es, s_wtime);
+
+ if (likely(now - last_update < EXT4_SB_REFRESH_INTERVAL_SEC))
+ return;
+
+ lifetime_write_kbytes = sbi->s_kbytes_written +
+ ((part_stat_read(sb->s_bdev, sectors[STAT_WRITE]) -
+ sbi->s_sectors_written_start) >> 1);
+
+ /* Get the number of kilobytes not written to disk to account
+ * for statistics and compare with a multiple of 16 MB. This
+ * is used to determine when the next superblock commit should
+ * occur (i.e. not more often than once per 16MB if there was
+ * less written in an hour).
+ */
+ diff_size = lifetime_write_kbytes - le64_to_cpu(es->s_kbytes_written);
+
+ if (diff_size > EXT4_SB_REFRESH_INTERVAL_KB)
+ schedule_work(&EXT4_SB(sb)->s_error_work);
+}
+
/*
* The del_gendisk() function uninitializes the disk-specific data
* structures, including the bdi structure, without telling anyone
@@ -459,6 +510,7 @@ static void ext4_journal_commit_callback(journal_t *journal, transaction_t *txn)
BUG_ON(txn->t_state == T_FINISHED);
ext4_process_freed_data(sb, txn->t_tid);
+ ext4_maybe_update_superblock(sb);
spin_lock(&sbi->s_md_lock);
while (!list_empty(&txn->t_private_list)) {
diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
index 6d332dff79dd..9f334de4f636 100644
--- a/fs/ext4/sysfs.c
+++ b/fs/ext4/sysfs.c
@@ -515,7 +515,8 @@ static const struct kobj_type ext4_feat_ktype = {
void ext4_notify_error_sysfs(struct ext4_sb_info *sbi)
{
- sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
+ if (sbi->s_add_error_count > 0)
+ sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
}
static struct kobject *ext4_root;
@@ -605,4 +606,3 @@ void ext4_exit_sysfs(void)
remove_proc_entry(proc_dirname, NULL);
ext4_proc_root = NULL;
}
-
--
On Jul 31, 2023, at 6:25 AM, Vitaliy Kuznetsov <[email protected]> wrote:
>
> This patch introduces a mechanism to periodically check and update
> the superblock within the ext4 file system. The main purpose of this
> patch is to keep the disk superblock up to date. The update will be
> performed if more than one hour has passed since the last update, and
> if more than 16MB of data have been written to disk.
>
> This check and update is performed within the ext4_journal_commit_callback
> function, ensuring that the superblock is written while the disk is
> active, rather than based on a timer that may trigger during disk idle
> periods.
>
> Discussion https://www.spinics.net/lists/linux-ext4/msg85865.html
>
> Signed-off-by: Vitaliy Kuznetsov <[email protected]>
Thanks for sending the patch.
Reviewed-by: Andreas Dilger <[email protected]>
> ---
> fs/ext4/super.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++++
> fs/ext4/sysfs.c | 4 ++--
> 2 files changed, 54 insertions(+), 2 deletions(-)
>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index c94ebf704616..2159e9705404 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -433,6 +433,57 @@ static time64_t __ext4_get_tstamp(__le32 *lo, __u8 *hi)
> #define ext4_get_tstamp(es, tstamp) \
> __ext4_get_tstamp(&(es)->tstamp, &(es)->tstamp ## _hi)
>
> +#define EXT4_SB_REFRESH_INTERVAL_SEC (3600) /* seconds (1 hour) */
> +#define EXT4_SB_REFRESH_INTERVAL_KB (16384) /* kilobytes (16MB) */
> +
> +/*
> + * The ext4_maybe_update_superblock() function checks and updates the
> + * superblock if needed.
> + *
> + * This function is designed to update the on-disk superblock only under
> + * certain conditions to prevent excessive disk writes and unnecessary
> + * waking of the disk from sleep. The superblock will be updated if:
> + * 1. More than an hour has passed since the last superblock update, and
> + * 2. More than 16MB have been written since the last superblock update.
> + *
> + * @sb: The superblock
> + */
> +static void ext4_maybe_update_superblock(struct super_block *sb)
> +{
> + struct ext4_sb_info *sbi = EXT4_SB(sb);
> + struct ext4_super_block *es = sbi->s_es;
> + journal_t *journal = sbi->s_journal;
> + time64_t now;
> + __u64 last_update;
> + __u64 lifetime_write_kbytes;
> + __u64 diff_size;
> +
> + if (sb_rdonly(sb) || !(sb->s_flags & SB_ACTIVE) ||
> + !journal || (journal->j_flags & JBD2_UNMOUNT))
> + return;
> +
> + now = ktime_get_real_seconds();
> + last_update = ext4_get_tstamp(es, s_wtime);
> +
> + if (likely(now - last_update < EXT4_SB_REFRESH_INTERVAL_SEC))
> + return;
> +
> + lifetime_write_kbytes = sbi->s_kbytes_written +
> + ((part_stat_read(sb->s_bdev, sectors[STAT_WRITE]) -
> + sbi->s_sectors_written_start) >> 1);
> +
> + /* Get the number of kilobytes not written to disk to account
> + * for statistics and compare with a multiple of 16 MB. This
> + * is used to determine when the next superblock commit should
> + * occur (i.e. not more often than once per 16MB if there was
> + * less written in an hour).
> + */
> + diff_size = lifetime_write_kbytes - le64_to_cpu(es->s_kbytes_written);
> +
> + if (diff_size > EXT4_SB_REFRESH_INTERVAL_KB)
> + schedule_work(&EXT4_SB(sb)->s_error_work);
> +}
> +
> /*
> * The del_gendisk() function uninitializes the disk-specific data
> * structures, including the bdi structure, without telling anyone
> @@ -459,6 +510,7 @@ static void ext4_journal_commit_callback(journal_t *journal, transaction_t *txn)
> BUG_ON(txn->t_state == T_FINISHED);
>
> ext4_process_freed_data(sb, txn->t_tid);
> + ext4_maybe_update_superblock(sb);
>
> spin_lock(&sbi->s_md_lock);
> while (!list_empty(&txn->t_private_list)) {
> diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
> index 6d332dff79dd..9f334de4f636 100644
> --- a/fs/ext4/sysfs.c
> +++ b/fs/ext4/sysfs.c
> @@ -515,7 +515,8 @@ static const struct kobj_type ext4_feat_ktype = {
>
> void ext4_notify_error_sysfs(struct ext4_sb_info *sbi)
> {
> - sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
> + if (sbi->s_add_error_count > 0)
> + sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
> }
>
> static struct kobject *ext4_root;
> @@ -605,4 +606,3 @@ void ext4_exit_sysfs(void)
> remove_proc_entry(proc_dirname, NULL);
> ext4_proc_root = NULL;
> }
> -
> --
Cheers, Andreas
On Mon, Jul 31, 2023 at 04:25:26PM +0400, Vitaliy Kuznetsov wrote:
> diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
> index 6d332dff79dd..9f334de4f636 100644
> --- a/fs/ext4/sysfs.c
> +++ b/fs/ext4/sysfs.c
> @@ -515,7 +515,8 @@ static const struct kobj_type ext4_feat_ktype = {
>
> void ext4_notify_error_sysfs(struct ext4_sb_info *sbi)
> {
> - sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
> + if (sbi->s_add_error_count > 0)
> + sysfs_notify(&sbi->s_kobj, NULL, "errors_count");
> }
The problem is that ext4_notify_error_sysfs() is called in
flush_stashed_error_work() **after** that function calls
ext4_update_super() --- and ext4_update_super will zero out
s_add_error_count. So this will result in the sysfs_notify call
*never* getting called, which would be a regression. So
unfortunately, I can't accept this patch as currently written.
Fortunately, only flush_stashed_error_work() calls
ext4_notify_error_sysfs(), so it should be easy enough to sample
s_add_error_count before calling ext4_update_super(), and then
conditionally call sysfs_notify() if it is non-zero.
- Ted
This patch introduces a mechanism to periodically check and update
the superblock within the ext4 file system. The main purpose of this
patch is to keep the disk superblock up to date. The update will be
performed if more than one hour has passed since the last update, and
if more than 16MB of data have been written to disk.
This check and update is performed within the ext4_journal_commit_callback
function, ensuring that the superblock is written while the disk is
active, rather than based on a timer that may trigger during disk idle
periods.
Discussion https://www.spinics.net/lists/linux-ext4/msg85865.html
Signed-off-by: Vitaliy Kuznetsov <[email protected]>
---
fs/ext4/super.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 61 insertions(+), 1 deletion(-)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index c94ebf704616..8bee05118c7a 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -433,6 +433,57 @@ static time64_t __ext4_get_tstamp(__le32 *lo, __u8 *hi)
#define ext4_get_tstamp(es, tstamp) \
__ext4_get_tstamp(&(es)->tstamp, &(es)->tstamp ## _hi)
+#define EXT4_SB_REFRESH_INTERVAL_SEC (3600) /* seconds (1 hour) */
+#define EXT4_SB_REFRESH_INTERVAL_KB (16384) /* kilobytes (16MB) */
+
+/*
+ * The ext4_maybe_update_superblock() function checks and updates the
+ * superblock if needed.
+ *
+ * This function is designed to update the on-disk superblock only under
+ * certain conditions to prevent excessive disk writes and unnecessary
+ * waking of the disk from sleep. The superblock will be updated if:
+ * 1. More than an hour has passed since the last superblock update, and
+ * 2. More than 16MB have been written since the last superblock update.
+ *
+ * @sb: The superblock
+ */
+static void ext4_maybe_update_superblock(struct super_block *sb)
+{
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct ext4_super_block *es = sbi->s_es;
+ journal_t *journal = sbi->s_journal;
+ time64_t now;
+ __u64 last_update;
+ __u64 lifetime_write_kbytes;
+ __u64 diff_size;
+
+ if (sb_rdonly(sb) || !(sb->s_flags & SB_ACTIVE) ||
+ !journal || (journal->j_flags & JBD2_UNMOUNT))
+ return;
+
+ now = ktime_get_real_seconds();
+ last_update = ext4_get_tstamp(es, s_wtime);
+
+ if (likely(now - last_update < EXT4_SB_REFRESH_INTERVAL_SEC))
+ return;
+
+ lifetime_write_kbytes = sbi->s_kbytes_written +
+ ((part_stat_read(sb->s_bdev, sectors[STAT_WRITE]) -
+ sbi->s_sectors_written_start) >> 1);
+
+ /* Get the number of kilobytes not written to disk to account
+ * for statistics and compare with a multiple of 16 MB. This
+ * is used to determine when the next superblock commit should
+ * occur (i.e. not more often than once per 16MB if there was
+ * less written in an hour).
+ */
+ diff_size = lifetime_write_kbytes - le64_to_cpu(es->s_kbytes_written);
+
+ if (diff_size > EXT4_SB_REFRESH_INTERVAL_KB)
+ schedule_work(&EXT4_SB(sb)->s_error_work);
+}
+
/*
* The del_gendisk() function uninitializes the disk-specific data
* structures, including the bdi structure, without telling anyone
@@ -459,6 +510,7 @@ static void ext4_journal_commit_callback(journal_t *journal, transaction_t *txn)
BUG_ON(txn->t_state == T_FINISHED);
ext4_process_freed_data(sb, txn->t_tid);
+ ext4_maybe_update_superblock(sb);
spin_lock(&sbi->s_md_lock);
while (!list_empty(&txn->t_private_list)) {
@@ -715,6 +767,7 @@ static void flush_stashed_error_work(struct work_struct *work)
*/
if (!sb_rdonly(sbi->s_sb) && journal) {
struct buffer_head *sbh = sbi->s_sbh;
+ bool call_notify_err;
handle = jbd2_journal_start(journal, 1);
if (IS_ERR(handle))
goto write_directly;
@@ -722,6 +775,10 @@ static void flush_stashed_error_work(struct work_struct *work)
jbd2_journal_stop(handle);
goto write_directly;
}
+
+ if (sbi->s_add_error_count > 0)
+ call_notify_err = true;
+
ext4_update_super(sbi->s_sb);
if (buffer_write_io_error(sbh) || !buffer_uptodate(sbh)) {
ext4_msg(sbi->s_sb, KERN_ERR, "previous I/O error to "
@@ -735,7 +792,10 @@ static void flush_stashed_error_work(struct work_struct *work)
goto write_directly;
}
jbd2_journal_stop(handle);
- ext4_notify_error_sysfs(sbi);
+
+ if (call_notify_err)
+ ext4_notify_error_sysfs(sbi);
+
return;
}
write_directly:
--
2.39.2