fs/nfs: Use fatal_signal_pending instead of signal_pending

commit bb00238 upstream.

We set the state of the current process to TASK_KILLABLE via
prepare_to_wait(). Should we use fatal_signal_pending() to detect
the signal here?

Fixes: b4868b4 ("NFSv4: Wait for stateid updates after CLOSE/OPEN_DOWNGRADE")
Signed-off-by: zhouchuangao <zhouchuangao@vivo.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs/nfs/nfs4proc.c

@@ -1682,7 +1682,7 @@ static void nfs_set_open_stateid_locked(struct nfs4_state *state,
 		rcu_read_unlock();
 		trace_nfs4_open_stateid_update_wait(state->inode, stateid, 0);
 
-		if (!signal_pending(current)) {
+		if (!fatal_signal_pending(current)) {
 			if (schedule_timeout(5*HZ) == 0)
 				status = -EAGAIN;
 			else
@@ -3458,7 +3458,7 @@ static bool nfs4_refresh_open_old_stateid(nfs4_stateid *dst,
 		write_sequnlock(&state->seqlock);
 		trace_nfs4_close_stateid_update_wait(state->inode, dst, 0);
 
-		if (signal_pending(current))
+		if (fatal_signal_pending(current))
 			status = -EINTR;
 		else
 			if (schedule_timeout(5*HZ) != 0)