xfs: fix iclog release error check race with shutdown

Prior to commit df732b29 ("xfs: call xlog_state_release_iclog with
l_icloglock held"), xlog_state_release_iclog() always performed a
locked check of the iclog error state before proceeding into the
sync state processing code. As of this commit, part of
xlog_state_release_iclog() was open-coded into
xfs_log_release_iclog() and as a result the locked error state check
was lost.

The lockless check still exists, but this doesn't account for the
possibility of a race with a shutdown being performed by another
task causing the iclog state to change while the original task waits
on ->l_icloglock. This has reproduced very rarely via generic/475
and manifests as an assert failure in __xlog_state_release_iclog()
due to an unexpected iclog state.

Restore the locked error state check in xlog_state_release_iclog()
to ensure that an iclog state update via shutdown doesn't race with
the iclog release state processing code.

Fixes: df732b29 ("xfs: call xlog_state_release_iclog with l_icloglock held")
Reported-by: Zorro Lang <zlang@redhat.com>
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>

fs/xfs/xfs_log.c

@@ -605,18 +605,23 @@ xfs_log_release_iclog(
 	struct xlog		*log = mp->m_log;
 	bool			sync;
 
-	if (iclog->ic_state == XLOG_STATE_IOERROR) {
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		return -EIO;
-	}
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
+		goto error;
 
 	if (atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock)) {
+		if (iclog->ic_state == XLOG_STATE_IOERROR) {
+			spin_unlock(&log->l_icloglock);
+			goto error;
+		}
 		sync = __xlog_state_release_iclog(log, iclog);
 		spin_unlock(&log->l_icloglock);
 		if (sync)
 			xlog_sync(log, iclog);
 	}
 	return 0;
+error:
+	xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
+	return -EIO;
 }
 
 /*