rds: fix rds_send_xmit() serialization

rds_send_xmit() was changed to hold an interrupt masking spinlock instead of a
mutex so that it could be called from the IB receive tasklet path.  This broke
the TCP transport because its xmit method can block and masks and unmasks
interrupts.

This patch serializes callers to rds_send_xmit() with a simple bit instead of
the current spinlock or previous mutex.  This enables rds_send_xmit() to be
called from any context and to call functions which block.  Getting rid of the
c_send_lock exposes the bare c_lock acquisitions which are changed to block
interrupts.

A waitqueue is added so that rds_conn_shutdown() can wait for callers to leave
rds_send_xmit() before tearing down partial send state.  This lets us get rid
of c_senders.

rds_send_xmit() is changed to check the conn state after acquiring the
RDS_IN_XMIT bit to resolve races with the shutdown path.  Previously both
worked with the conn state and then the lock in the same order, allowing them
to race and execute the paths concurrently.

rds_send_reset() isn't racing with rds_send_xmit() now that rds_conn_shutdown()
properly ensures that rds_send_xmit() can't start once the conn state has been
changed.  We can remove its previous use of the spinlock.

Finally, c_send_generation is redundant.  Callers can race to test the c_flags
bit by simply retrying instead of racing to test the c_send_generation atomic.

Signed-off-by: Zach Brown <zach.brown@oracle.com>

1 files changed, 5 insertions, 14 deletions

diff --git a/net/rds/connection.c b/net/rds/connection.c
index 5bb0eec..89871db 100644
--- a/net/rds/connection.c
+++ b/net/rds/connection.c
@@ -148,9 +148,7 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr,
 	spin_lock_init(&conn->c_lock);
 	conn->c_next_tx_seq = 1;
 
-	spin_lock_init(&conn->c_send_lock);
-	atomic_set(&conn->c_send_generation, 1);
-	atomic_set(&conn->c_senders, 0);
+	init_waitqueue_head(&conn->c_waitq);
 	INIT_LIST_HEAD(&conn->c_send_queue);
 	INIT_LIST_HEAD(&conn->c_retrans);
 
@@ -275,15 +273,8 @@ void rds_conn_shutdown(struct rds_connection *conn)
 		}
 		mutex_unlock(&conn->c_cm_lock);
 
-		/* verify everybody's out of rds_send_xmit() */
-		spin_lock_irq(&conn->c_send_lock);
-		spin_unlock_irq(&conn->c_send_lock);
-
-		while(atomic_read(&conn->c_senders)) {
-			schedule_timeout(1);
-			spin_lock_irq(&conn->c_send_lock);
-			spin_unlock_irq(&conn->c_send_lock);
-		}
+		wait_event(conn->c_waitq,
+			   !test_bit(RDS_IN_XMIT, &conn->c_flags));
 
 		conn->c_trans->conn_shutdown(conn);
 		rds_conn_reset(conn);
@@ -477,8 +468,8 @@ static int rds_conn_info_visitor(struct rds_connection *conn,
 		sizeof(cinfo->transport));
 	cinfo->flags = 0;
 
-	rds_conn_info_set(cinfo->flags,
-			  spin_is_locked(&conn->c_send_lock), SENDING);
+	rds_conn_info_set(cinfo->flags, test_bit(RDS_IN_XMIT, &conn->c_flags),
+			  SENDING);
 	/* XXX Future: return the state rather than these funky bits */
 	rds_conn_info_set(cinfo->flags,
 			  atomic_read(&conn->c_state) == RDS_CONN_CONNECTING,