1 files changed, 36 insertions, 8 deletions

diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index c34433432d47..ce2bf9d74224 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1041,8 +1041,8 @@ static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
 	/* For undo access buffer must have data copied */
 	if (undo && !jh->b_committed_data)
 		goto out;
-	if (jh->b_transaction != handle->h_transaction &&
-	    jh->b_next_transaction != handle->h_transaction)
+	if (READ_ONCE(jh->b_transaction) != handle->h_transaction &&
+	    READ_ONCE(jh->b_next_transaction) != handle->h_transaction)
 		goto out;
 	/*
 	 * There are two reasons for the barrier here:
@@ -1906,6 +1906,9 @@ static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
  */
 static void __jbd2_journal_unfile_buffer(struct journal_head *jh)
 {
+	J_ASSERT_JH(jh, jh->b_transaction != NULL);
+	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+
 	__jbd2_journal_temp_unlink_buffer(jh);
 	jh->b_transaction = NULL;
 	jbd2_journal_put_journal_head(jh);
@@ -1997,6 +2000,7 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal,
 {
 	struct buffer_head *head;
 	struct buffer_head *bh;
+	bool has_write_io_error = false;
 	int ret = 0;
 
 	J_ASSERT(PageLocked(page));
@@ -2021,11 +2025,26 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal,
 		jbd_unlock_bh_state(bh);
 		if (buffer_jbd(bh))
 			goto busy;
+
+		/*
+		 * If we free a metadata buffer which has been failed to
+		 * write out, the jbd2 checkpoint procedure will not detect
+		 * this failure and may lead to filesystem inconsistency
+		 * after cleanup journal tail.
+		 */
+		if (buffer_write_io_error(bh)) {
+			pr_err("JBD2: Error while async write back metadata bh %llu.",
+			       (unsigned long long)bh->b_blocknr);
+			has_write_io_error = true;
+		}
 	} while ((bh = bh->b_this_page) != head);
 
 	ret = try_to_free_buffers(page);
 
 busy:
+	if (has_write_io_error)
+		jbd2_journal_abort(journal, -EIO);
+
 	return ret;
 }
 
@@ -2223,14 +2242,16 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
 			return -EBUSY;
 		}
 		/*
-		 * OK, buffer won't be reachable after truncate. We just set
-		 * j_next_transaction to the running transaction (if there is
-		 * one) and mark buffer as freed so that commit code knows it
-		 * should clear dirty bits when it is done with the buffer.
+		 * OK, buffer won't be reachable after truncate. We just clear
+		 * b_modified to not confuse transaction credit accounting, and
+		 * set j_next_transaction to the running transaction (if there
+		 * is one) and mark buffer as freed so that commit code knows
+		 * it should clear dirty bits when it is done with the buffer.
 		 */
 		set_buffer_freed(bh);
 		if (journal->j_running_transaction && buffer_jbddirty(bh))
 			jh->b_next_transaction = journal->j_running_transaction;
+		jh->b_modified = 0;
 		jbd2_journal_put_journal_head(jh);
 		spin_unlock(&journal->j_list_lock);
 		jbd_unlock_bh_state(bh);
@@ -2451,13 +2472,20 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh)
 
 	was_dirty = test_clear_buffer_jbddirty(bh);
 	__jbd2_journal_temp_unlink_buffer(jh);
+
+	/*
+	 * b_transaction must be set, otherwise the new b_transaction won't
+	 * be holding jh reference
+	 */
+	J_ASSERT_JH(jh, jh->b_transaction != NULL);
+
 	/*
 	 * We set b_transaction here because b_next_transaction will inherit
 	 * our jh reference and thus __jbd2_journal_file_buffer() must not
 	 * take a new one.
 	 */
-	jh->b_transaction = jh->b_next_transaction;
-	jh->b_next_transaction = NULL;
+	WRITE_ONCE(jh->b_transaction, jh->b_next_transaction);
+	WRITE_ONCE(jh->b_next_transaction, NULL);
 	if (buffer_freed(bh))
 		jlist = BJ_Forget;
 	else if (jh->b_modified)