[PATCH net, 0/2] hv_netvsc: Some fixes for the select_queue

STALE2217d

4 messages, 2 authors, 2020-08-20 · open the first message on its own page

[PATCH net, 0/2] hv_netvsc: Some fixes for the select_queue

From: Haiyang Zhang <haiyangz@microsoft.com>
Date: 2020-08-20 21:53:52

This patch set includes two fixes for the select_queue process.

Haiyang Zhang (2):
  hv_netvsc: Remove "unlikely" from netvsc_select_queue
  hv_netvsc: Fix the queue_mapping in netvsc_vf_xmit()

 drivers/net/hyperv/netvsc_drv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 
2.25.1

[PATCH net, 1/2] hv_netvsc: Remove "unlikely" from netvsc_select_queue

From: Haiyang Zhang <haiyangz@microsoft.com>
Date: 2020-08-20 21:54:22

When using vf_ops->ndo_select_queue, the number of queues of VF is
usually bigger than the synthetic NIC. This condition may happen
often.
Remove "unlikely" from the comparison of ndev->real_num_tx_queues.

Fixes: b3bf5666a510 ("hv_netvsc: defer queue selection to VF")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
---
 drivers/net/hyperv/netvsc_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 787f17e2a971..0029292cdb9f 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -367,7 +367,7 @@ static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb,
 	}
 	rcu_read_unlock();
 
-	while (unlikely(txq >= ndev->real_num_tx_queues))
+	while (txq >= ndev->real_num_tx_queues)
 		txq -= ndev->real_num_tx_queues;
 
 	return txq;
-- 
2.25.1

[PATCH net, 2/2] hv_netvsc: Fix the queue_mapping in netvsc_vf_xmit()

From: Haiyang Zhang <haiyangz@microsoft.com>
Date: 2020-08-20 21:54:29

netvsc_vf_xmit() / dev_queue_xmit() will call VF NIC’s ndo_select_queue
or netdev_pick_tx() again. They will use skb_get_rx_queue() to get the
queue number, so the “skb->queue_mapping - 1” will be used. This may
cause the last queue of VF not been used.

Use skb_record_rx_queue() here, so that the skb_get_rx_queue() called
later will get the correct queue number, and VF will be able to use
all queues.

Fixes: b3bf5666a510 ("hv_netvsc: defer queue selection to VF")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
---
 drivers/net/hyperv/netvsc_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 0029292cdb9f..64b0a74c1523 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -502,7 +502,7 @@ static int netvsc_vf_xmit(struct net_device *net, struct net_device *vf_netdev,
 	int rc;
 
 	skb->dev = vf_netdev;
-	skb->queue_mapping = qdisc_skb_cb(skb)->slave_dev_queue_mapping;
+	skb_record_rx_queue(skb, qdisc_skb_cb(skb)->slave_dev_queue_mapping);
 
 	rc = dev_queue_xmit(skb);
 	if (likely(rc == NET_XMIT_SUCCESS || rc == NET_XMIT_CN)) {
-- 
2.25.1

Re: [PATCH net, 0/2] hv_netvsc: Some fixes for the select_queue

From: David Miller <davem@davemloft.net>
Date: 2020-08-20 23:24:37

From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Thu, 20 Aug 2020 14:53:13 -0700
This patch set includes two fixes for the select_queue process.
Series applied, thank you.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help