jbd2: Make state lock a spinlock

Bit-spinlocks are problematic on PREEMPT_RT if functions which might sleep
on RT, e.g. spin_lock(), alloc/free(), are invoked inside the lock held
region because bit spinlocks disable preemption even on RT.

A first attempt was to replace state lock with a spinlock placed in struct
buffer_head and make the locking conditional on PREEMPT_RT and
DEBUG_BIT_SPINLOCKS.

Jan pointed out that there is a 4 byte hole in struct journal_head where a
regular spinlock fits in and he would not object to convert the state lock
to a spinlock unconditionally.

Aside of solving the RT problem, this also gains lockdep coverage for the
journal head state lock (bit-spinlocks are not covered by lockdep as it's
hard to fit a lockdep map into a single bit).

The trivial change would have been to convert the jbd_*lock_bh_state()
inlines, but that comes with the downside that these functions take a
buffer head pointer which needs to be converted to a journal head pointer
which adds another level of indirection.

As almost all functions which use this lock have a journal head pointer
readily available, it makes more sense to remove the lock helper inlines
and write out spin_*lock() at all call sites.

Fixup all locking comments as well.
Suggested-by: Jan Kara <jack@suse.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Jan Kara <jack@suse.com>
Cc: linux-ext4@vger.kernel.org
Link: https://lore.kernel.org/r/20190809124233.13277-7-jack@suse.czSigned-off-by: Theodore Ts'o <tytso@mit.edu>

fs/jbd2/commit.c

@@ -482,10 +482,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		if (jh->b_committed_data) {
 			struct buffer_head *bh = jh2bh(jh);
 
-			jbd_lock_bh_state(bh);
+			spin_lock(&jh->b_state_lock);
 			jbd2_free(jh->b_committed_data, bh->b_size);
 			jh->b_committed_data = NULL;
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 		}
 		jbd2_journal_refile_buffer(journal, jh);
 	}
@@ -928,7 +928,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		 * done with it.
 		 */
 		get_bh(bh);
-		jbd_lock_bh_state(bh);
+		spin_lock(&jh->b_state_lock);
 		J_ASSERT_JH(jh,	jh->b_transaction == commit_transaction);
 
 		/*
@@ -1024,7 +1024,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		}
 		JBUFFER_TRACE(jh, "refile or unfile buffer");
 		drop_ref = __jbd2_journal_refile_buffer(jh);
-		jbd_unlock_bh_state(bh);
+		spin_unlock(&jh->b_state_lock);
 		if (drop_ref)
 			jbd2_journal_put_journal_head(jh);
 		if (try_to_free)

fs/jbd2/journal.c

@@ -363,7 +363,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	/* keep subsequent assertions sane */
 	atomic_set(&new_bh->b_count, 1);
 
-	jbd_lock_bh_state(bh_in);
+	spin_lock(&jh_in->b_state_lock);
 repeat:
 	/*
 	 * If a new transaction has already done a buffer copy-out, then
@@ -405,13 +405,13 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	if (need_copy_out && !done_copy_out) {
 		char *tmp;
 
-		jbd_unlock_bh_state(bh_in);
+		spin_unlock(&jh_in->b_state_lock);
 		tmp = jbd2_alloc(bh_in->b_size, GFP_NOFS);
 		if (!tmp) {
 			brelse(new_bh);
 			return -ENOMEM;
 		}
-		jbd_lock_bh_state(bh_in);
+		spin_lock(&jh_in->b_state_lock);
 		if (jh_in->b_frozen_data) {
 			jbd2_free(tmp, bh_in->b_size);
 			goto repeat;
@@ -464,7 +464,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	__jbd2_journal_file_buffer(jh_in, transaction, BJ_Shadow);
 	spin_unlock(&journal->j_list_lock);
 	set_buffer_shadow(bh_in);
-	jbd_unlock_bh_state(bh_in);
+	spin_unlock(&jh_in->b_state_lock);
 
 	return do_escape | (done_copy_out << 1);
 }
@@ -2410,6 +2410,8 @@ static struct journal_head *journal_alloc_journal_head(void)
 		ret = kmem_cache_zalloc(jbd2_journal_head_cache,
 				GFP_NOFS | __GFP_NOFAIL);
 	}
+	if (ret)
+		spin_lock_init(&ret->b_state_lock);
 	return ret;
 }
 

fs/ocfs2/suballoc.c

@@ -1252,6 +1252,7 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
 					 int nr)
 {
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
+	struct journal_head *jh;
 	int ret;
 
 	if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
@@ -1260,13 +1261,14 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
 	if (!buffer_jbd(bg_bh))
 		return 1;
 
-	jbd_lock_bh_state(bg_bh);
-	bg = (struct ocfs2_group_desc *) bh2jh(bg_bh)->b_committed_data;
+	jh = bh2jh(bg_bh);
+	spin_lock(&jh->b_state_lock);
+	bg = (struct ocfs2_group_desc *) jh->b_committed_data;
 	if (bg)
 		ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
 	else
 		ret = 1;
-	jbd_unlock_bh_state(bg_bh);
+	spin_unlock(&jh->b_state_lock);
 
 	return ret;
 }
@@ -2387,6 +2389,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	int status;
 	unsigned int tmp;
 	struct ocfs2_group_desc *undo_bg = NULL;
+	struct journal_head *jh;
 
 	/* The caller got this descriptor from
 	 * ocfs2_read_group_descriptor().  Any corruption is a code bug. */
@@ -2405,10 +2408,10 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 		goto bail;
 	}
 
+	jh = bh2jh(group_bh);
 	if (undo_fn) {
-		jbd_lock_bh_state(group_bh);
-		undo_bg = (struct ocfs2_group_desc *)
-					bh2jh(group_bh)->b_committed_data;
+		spin_lock(&jh->b_state_lock);
+		undo_bg = (struct ocfs2_group_desc *) jh->b_committed_data;
 		BUG_ON(!undo_bg);
 	}
 
@@ -2423,7 +2426,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	le16_add_cpu(&bg->bg_free_bits_count, num_bits);
 	if (le16_to_cpu(bg->bg_free_bits_count) > le16_to_cpu(bg->bg_bits)) {
 		if (undo_fn)
-			jbd_unlock_bh_state(group_bh);
+			spin_unlock(&jh->b_state_lock);
 		return ocfs2_error(alloc_inode->i_sb, "Group descriptor # %llu has bit count %u but claims %u are freed. num_bits %d\n",
 				   (unsigned long long)le64_to_cpu(bg->bg_blkno),
 				   le16_to_cpu(bg->bg_bits),
@@ -2432,7 +2435,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	}
 
 	if (undo_fn)
-		jbd_unlock_bh_state(group_bh);
+		spin_unlock(&jh->b_state_lock);
 
 	ocfs2_journal_dirty(handle, group_bh);
 bail:

include/linux/jbd2.h

@@ -313,7 +313,6 @@ enum jbd_state_bits {
 	BH_Revoked,		/* Has been revoked from the log */
 	BH_RevokeValid,		/* Revoked flag is valid */
 	BH_JBDDirty,		/* Is dirty but journaled */
-	BH_State,		/* Pins most journal_head state */
 	BH_JournalHead,		/* Pins bh->b_private and jh->b_bh */
 	BH_Shadow,		/* IO on shadow buffer is running */
 	BH_Verified,		/* Metadata block has been verified ok */
@@ -342,21 +341,6 @@ static inline struct journal_head *bh2jh(struct buffer_head *bh)
 	return bh->b_private;
 }
 
-static inline void jbd_lock_bh_state(struct buffer_head *bh)
-{
-	bit_spin_lock(BH_State, &bh->b_state);
-}
-
-static inline int jbd_is_locked_bh_state(struct buffer_head *bh)
-{
-	return bit_spin_is_locked(BH_State, &bh->b_state);
-}
-
-static inline void jbd_unlock_bh_state(struct buffer_head *bh)
-{
-	bit_spin_unlock(BH_State, &bh->b_state);
-}
-
 static inline void jbd_lock_bh_journal_head(struct buffer_head *bh)
 {
 	bit_spin_lock(BH_JournalHead, &bh->b_state);
@@ -551,9 +535,9 @@ struct transaction_chp_stats_s {
  *      ->jbd_lock_bh_journal_head()	(This is "innermost")
  *
  *    j_state_lock
- *    ->jbd_lock_bh_state()
+ *    ->b_state_lock
  *
- *    jbd_lock_bh_state()
+ *    b_state_lock
  *    ->j_list_lock
  *
  *    j_state_lock

include/linux/journal-head.h

@@ -11,6 +11,8 @@
 #ifndef JOURNAL_HEAD_H_INCLUDED
 #define JOURNAL_HEAD_H_INCLUDED
 
+#include <linux/spinlock.h>
+
 typedef unsigned int		tid_t;		/* Unique transaction ID */
 typedef struct transaction_s	transaction_t;	/* Compound transaction type */
 
@@ -23,6 +25,11 @@ struct journal_head {
 	 */
 	struct buffer_head *b_bh;
 
+	/*
+	 * Protect the buffer head state
+	 */
+	spinlock_t b_state_lock;
+
 	/*
 	 * Reference count - see description in journal.c
 	 * [jbd_lock_bh_journal_head()]
@@ -30,7 +37,7 @@ struct journal_head {
 	int b_jcount;
 
 	/*
-	 * Journalling list for this buffer [jbd_lock_bh_state()]
+	 * Journalling list for this buffer [b_state_lock]
 	 * NOTE: We *cannot* combine this with b_modified into a bitfield
 	 * as gcc would then (which the C standard allows but which is
 	 * very unuseful) make 64-bit accesses to the bitfield and clobber
@@ -41,20 +48,20 @@ struct journal_head {
 	/*
 	 * This flag signals the buffer has been modified by
 	 * the currently running transaction
-	 * [jbd_lock_bh_state()]
+	 * [b_state_lock]
 	 */
 	unsigned b_modified;
 
 	/*
 	 * Copy of the buffer data frozen for writing to the log.
-	 * [jbd_lock_bh_state()]
+	 * [b_state_lock]
 	 */
 	char *b_frozen_data;
 
 	/*
 	 * Pointer to a saved copy of the buffer containing no uncommitted
 	 * deallocation references, so that allocations can avoid overwriting
-	 * uncommitted deletes. [jbd_lock_bh_state()]
+	 * uncommitted deletes. [b_state_lock]
 	 */
 	char *b_committed_data;
 
@@ -63,7 +70,7 @@ struct journal_head {
 	 * metadata: either the running transaction or the committing
 	 * transaction (if there is one).  Only applies to buffers on a
 	 * transaction's data or metadata journaling list.
-	 * [j_list_lock] [jbd_lock_bh_state()]
+	 * [j_list_lock] [b_state_lock]
 	 * Either of these locks is enough for reading, both are needed for
 	 * changes.
 	 */
@@ -73,13 +80,13 @@ struct journal_head {
 	 * Pointer to the running compound transaction which is currently
 	 * modifying the buffer's metadata, if there was already a transaction
 	 * committing it when the new transaction touched it.
-	 * [t_list_lock] [jbd_lock_bh_state()]
+	 * [t_list_lock] [b_state_lock]
 	 */
 	transaction_t *b_next_transaction;
 
 	/*
 	 * Doubly-linked list of buffers on a transaction's data, metadata or
-	 * forget queue. [t_list_lock] [jbd_lock_bh_state()]
+	 * forget queue. [t_list_lock] [b_state_lock]
 	 */
 	struct journal_head *b_tnext, *b_tprev;