This is a note to let you know that I've just added the patch titled
jbd2: increase the journal IO's priority
to the 6.6-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
jbd2-increase-the-journal-io-s-priority.patch
and it can be found in the queue-6.6 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
commit 1a364cd42d150cfc4f6647d0a0c29f55a652bcab
Author: Zhang Yi <yi.zhang@xxxxxxxxxx>
Date: Wed Nov 29 19:47:40 2023 +0800
jbd2: increase the journal IO's priority
[ Upstream commit 6a3afb6ac6dfab158ebdd4b87941178f58c8939f ]
Current jbd2 only add REQ_SYNC for descriptor block, metadata log
buffer, commit buffer and superblock buffer, the submitted IO could be
throttled by writeback throttle in block layer, that could lead to
priority inversion in some cases. The log IO looks like a kind of high
priority metadata IO, so it should not be throttled by WBT like QOS
policies in block layer, let's add REQ_SYNC | REQ_IDLE to exempt from
writeback throttle, and also add REQ_META together indicates it's a
metadata IO.
Signed-off-by: Zhang Yi <yi.zhang@xxxxxxxxxx>
Reviewed-by: Jan Kara <jack@xxxxxxx>
Link: https://lore.kernel.org/r/20231129114740.2686201-2-yi.zhang@xxxxxxxxxxxxxxx
Signed-off-by: Theodore Ts'o <tytso@xxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 8d6f934c3d95..9bdb377a348f 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -119,7 +119,7 @@ static int journal_submit_commit_record(journal_t *journal,
struct commit_header *tmp;
struct buffer_head *bh;
struct timespec64 now;
- blk_opf_t write_flags = REQ_OP_WRITE | REQ_SYNC;
+ blk_opf_t write_flags = REQ_OP_WRITE | JBD2_JOURNAL_REQ_FLAGS;
*cbh = NULL;
@@ -395,8 +395,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
*/
jbd2_journal_update_sb_log_tail(journal,
journal->j_tail_sequence,
- journal->j_tail,
- REQ_SYNC);
+ journal->j_tail, 0);
mutex_unlock(&journal->j_checkpoint_mutex);
} else {
jbd2_debug(3, "superblock not updated\n");
@@ -715,6 +714,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
for (i = 0; i < bufs; i++) {
struct buffer_head *bh = wbuf[i];
+
/*
* Compute checksum.
*/
@@ -727,7 +727,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
clear_buffer_dirty(bh);
set_buffer_uptodate(bh);
bh->b_end_io = journal_end_buffer_io_sync;
- submit_bh(REQ_OP_WRITE | REQ_SYNC, bh);
+ submit_bh(REQ_OP_WRITE | JBD2_JOURNAL_REQ_FLAGS,
+ bh);
}
cond_resched();
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index e7aa47a02d4d..19c69229ac6e 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1100,8 +1100,7 @@ int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
* space and if we lose sb update during power failure we'd replay
* old transaction with possibly newly overwritten data.
*/
- ret = jbd2_journal_update_sb_log_tail(journal, tid, block,
- REQ_SYNC | REQ_FUA);
+ ret = jbd2_journal_update_sb_log_tail(journal, tid, block, REQ_FUA);
if (ret)
goto out;
@@ -1768,8 +1767,7 @@ static int journal_reset(journal_t *journal)
*/
jbd2_journal_update_sb_log_tail(journal,
journal->j_tail_sequence,
- journal->j_tail,
- REQ_SYNC | REQ_FUA);
+ journal->j_tail, REQ_FUA);
mutex_unlock(&journal->j_checkpoint_mutex);
}
return jbd2_journal_start_thread(journal);
@@ -1791,6 +1789,11 @@ static int jbd2_write_superblock(journal_t *journal, blk_opf_t write_flags)
return -EIO;
}
+ /*
+ * Always set high priority flags to exempt from block layer's
+ * QOS policies, e.g. writeback throttle.
+ */
+ write_flags |= JBD2_JOURNAL_REQ_FLAGS;
if (!(journal->j_flags & JBD2_BARRIER))
write_flags &= ~(REQ_FUA | REQ_PREFLUSH);
@@ -2045,7 +2048,7 @@ void jbd2_journal_update_sb_errno(journal_t *journal)
jbd2_debug(1, "JBD2: updating superblock error (errno %d)\n", errcode);
sb->s_errno = cpu_to_be32(errcode);
- jbd2_write_superblock(journal, REQ_SYNC | REQ_FUA);
+ jbd2_write_superblock(journal, REQ_FUA);
}
EXPORT_SYMBOL(jbd2_journal_update_sb_errno);
@@ -2166,8 +2169,7 @@ int jbd2_journal_destroy(journal_t *journal)
++journal->j_transaction_sequence;
write_unlock(&journal->j_state_lock);
- jbd2_mark_journal_empty(journal,
- REQ_SYNC | REQ_PREFLUSH | REQ_FUA);
+ jbd2_mark_journal_empty(journal, REQ_PREFLUSH | REQ_FUA);
mutex_unlock(&journal->j_checkpoint_mutex);
} else
err = -EIO;
@@ -2468,7 +2470,7 @@ int jbd2_journal_flush(journal_t *journal, unsigned int flags)
* the magic code for a fully-recovered superblock. Any future
* commits of data to the journal will restore the current
* s_start value. */
- jbd2_mark_journal_empty(journal, REQ_SYNC | REQ_FUA);
+ jbd2_mark_journal_empty(journal, REQ_FUA);
if (flags)
err = __jbd2_journal_erase(journal, flags);
@@ -2514,7 +2516,7 @@ int jbd2_journal_wipe(journal_t *journal, int write)
if (write) {
/* Lock to make assertions happy... */
mutex_lock_io(&journal->j_checkpoint_mutex);
- jbd2_mark_journal_empty(journal, REQ_SYNC | REQ_FUA);
+ jbd2_mark_journal_empty(journal, REQ_FUA);
mutex_unlock(&journal->j_checkpoint_mutex);
}
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 52772c826c86..0fc6c1f51262 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -1374,6 +1374,9 @@ JBD2_FEATURE_INCOMPAT_FUNCS(csum2, CSUM_V2)
JBD2_FEATURE_INCOMPAT_FUNCS(csum3, CSUM_V3)
JBD2_FEATURE_INCOMPAT_FUNCS(fast_commit, FAST_COMMIT)
+/* Journal high priority write IO operation flags */
+#define JBD2_JOURNAL_REQ_FLAGS (REQ_META | REQ_SYNC | REQ_IDLE)
+
/*
* Journal flag definitions
*/
