[PATCHv2 net] xen-netfront: request Tx response events more often
From: David Vrabel <hidden>
Date: 2016-01-26 17:13:09
From: Malcolm Crossley <redacted> Trying to batch Tx response events results in poor performance because this delays freeing the transmitted skbs. Instead use the standard RING_FINAL_CHECK_FOR_RESPONSES() macro to be notified once the next Tx response is placed on the ring. Signed-off-by: Malcolm Crossley <redacted> Signed-off-by: David Vrabel <redacted> --- v2: - Use bool. --- drivers/net/xen-netfront.c | 15 +++------------ 1 file changed, 3 insertions(+), 12 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index d6abf19..96ccd4e 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c@@ -364,6 +364,7 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) RING_IDX cons, prod; unsigned short id; struct sk_buff *skb; + bool more_to_do; BUG_ON(!netif_carrier_ok(queue->info->netdev));
@@ -398,18 +399,8 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) queue->tx.rsp_cons = prod; - /* - * Set a new event, then check for race with update of tx_cons. - * Note that it is essential to schedule a callback, no matter - * how few buffers are pending. Even if there is space in the - * transmit ring, higher layers may be blocked because too much - * data is outstanding: in such cases notification from Xen is - * likely to be the only kick that we'll get. - */ - queue->tx.sring->rsp_event = - prod + ((queue->tx.sring->req_prod - prod) >> 1) + 1; - mb(); /* update shared area */ - } while ((cons == prod) && (prod != queue->tx.sring->rsp_prod)); + RING_FINAL_CHECK_FOR_RESPONSES(&queue->tx, more_to_do); + } while (more_to_do); xennet_maybe_wake_tx(queue); }
--
2.1.4