[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <581803e21c35041f077f178b25621132bac98aec.1554892007.git.pabeni@redhat.com>
Date: Wed, 10 Apr 2019 14:32:37 +0200
From: Paolo Abeni <pabeni@...hat.com>
To: netdev@...r.kernel.org
Cc: "David S. Miller" <davem@...emloft.net>,
Jamal Hadi Salim <jhs@...atatu.com>,
Cong Wang <xiyou.wangcong@...il.com>,
Jiri Pirko <jiri@...nulli.us>,
Eric Dumazet <edumazet@...gle.com>,
Ivan Vecera <ivecera@...hat.com>
Subject: [PATCH net-next v2 1/5] net: caif: avoid using qdisc_qlen()
Such helper does not cope correctly with NOLOCK qdiscs.
In the following patches we will move back qlen to per CPU
values for such qdiscs, so qdisc_qlen_sum() is not an option,
too.
Instead, use qlen only for lock qdiscs, and always set
flow off for NOLOCK qdiscs with a not empty tx queue.
Signed-off-by: Paolo Abeni <pabeni@...hat.com>
---
net/caif/caif_dev.c | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c
index 711d7156efd8..6c6e01963aac 100644
--- a/net/caif/caif_dev.c
+++ b/net/caif/caif_dev.c
@@ -186,15 +186,19 @@ static int transmit(struct cflayer *layer, struct cfpkt *pkt)
goto noxoff;
if (likely(!netif_queue_stopped(caifd->netdev))) {
+ struct Qdisc *sch;
+
/* If we run with a TX queue, check if the queue is too long*/
txq = netdev_get_tx_queue(skb->dev, 0);
- qlen = qdisc_qlen(rcu_dereference_bh(txq->qdisc));
-
- if (likely(qlen == 0))
+ sch = rcu_dereference_bh(txq->qdisc);
+ if (likely(qdisc_is_empty(sch)))
goto noxoff;
+ /* can check for explicit qdisc len value only !NOLOCK,
+ * always set flow off otherwise
+ */
high = (caifd->netdev->tx_queue_len * q_high) / 100;
- if (likely(qlen < high))
+ if (!(sch->flags & TCQ_F_NOLOCK) && likely(sch->q.qlen < high))
goto noxoff;
}
--
2.20.1
Powered by blists - more mailing lists