On 2018/9/20 22:50, Haiyang Zhang wrote:

quoted

-----Original Message-----
From: Stephen Hemminger <stephen@networkplumber.org>
Sent: Thursday, September 20, 2018 10:44 AM
To: YueHaibing <redacted>
Cc: davem at davemloft.net; dmitry.tarnyagin at lockless.no;
wg at grandegger.com; mkl at pengutronix.de; michal.simek at xilinx.com;
hsweeten at visionengravers.com; madalin.bucur at nxp.com;
pantelis.antoniou at gmail.com; claudiu.manoil at nxp.com; leoyang.li at nxp.com;
linux at armlinux.org.uk; sammy at sammy.net; ralf at linux-mips.org;
nico at fluxnic.net; steve.glendinning at shawell.net; f.fainelli at gmail.com;
grygorii.strashko at ti.com; w-kwok2 at ti.com; m-karicheri2 at ti.com;
t.sailer at alumni.ethz.ch; jreuter at yaina.de; KY Srinivasan [off-list ref];
Haiyang Zhang [off-list ref]; wei.liu2 at citrix.com;
paul.durrant at citrix.com; arvid.brodin at alten.se; pshelar at ovn.org;
dev at openvswitch.org; linux-mips at linux-mips.org; xen-
devel at lists.xenproject.org; netdev at vger.kernel.org; linux-usb at vger.kernel.org;
linux-kernel at vger.kernel.org; linux-can at vger.kernel.org;
devel at linuxdriverproject.org; linux-hams at vger.kernel.org; linux-
omap at vger.kernel.org; linuxppc-dev at lists.ozlabs.org; linux-arm-
kernel at lists.infradead.org
Subject: Re: [PATCH net-next 17/22] hv_netvsc: fix return type of
ndo_start_xmit function

On Thu, 20 Sep 2018 20:33:01 +0800
YueHaibing [off-list ref] wrote:

quoted

int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 	 */
 	vf_netdev = rcu_dereference_bh(net_device_ctx->vf_netdev);
 	if (vf_netdev && netif_running(vf_netdev) &&
-	    !netpoll_tx_running(net))
-		return netvsc_vf_xmit(net, vf_netdev, skb);
+	    !netpoll_tx_running(net)) {
+		ret = netvsc_vf_xmit(net, vf_netdev, skb);
+		if (ret)
+			return NETDEV_TX_BUSY;
+	}

Sorry, the new code is wrong. It will fall through if ret == 0 (NETDEV_TX_OK)
Please review and test your patches.

Plus consideration of -- For error case, please just return NETDEV_TX_OK. We 
are not sure if the error can go away after retrying, returning NETDEV_TX_BUSY 
may cause infinite retry from the upper layer.

So, let's just always return NETDEV_TX_OK like this:
		netvsc_vf_xmit(net, vf_netdev, skb);
		return NETDEV_TX_OK;

Thank you for review.

Will do that in v2.

Thanks,
- Haiyang

.