RE: [PATCH V3 net-next 1/1] hv_netvsc: Use the xmit_more skb flag to optimize signaling the host
From: KY Srinivasan
Date: Wed May 06 2015 - 14:29:05 EST
> -----Original Message-----
> From: Eric Dumazet [mailto:eric.dumazet@xxxxxxxxx]
> Sent: Wednesday, May 6, 2015 11:09 AM
> To: KY Srinivasan
> Cc: davem@xxxxxxxxxxxxx; netdev@xxxxxxxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx; devel@xxxxxxxxxxxxxxxxxxxxxx; olaf@xxxxxxxxx;
> apw@xxxxxxxxxxxxx; jasowang@xxxxxxxxxx
> Subject: Re: [PATCH V3 net-next 1/1] hv_netvsc: Use the xmit_more skb flag
> to optimize signaling the host
>
> On Wed, 2015-05-06 at 11:20 -0700, K. Y. Srinivasan wrote:
> > Based on the information given to this driver (via the xmit_more skb flag),
> > we can defer signaling the host if more packets are on the way. This will
> help
> > make the host more efficient since it can potentially process a larger batch
> of
> > packets. Implement this optimization.
>
>
> > diff --git a/drivers/net/hyperv/netvsc_drv.c
> b/drivers/net/hyperv/netvsc_drv.c
> > index 5993c7e..4efaa6e 100644
> > --- a/drivers/net/hyperv/netvsc_drv.c
> > +++ b/drivers/net/hyperv/netvsc_drv.c
> > @@ -435,6 +435,8 @@ check_size:
> > packet->page_buf = page_buf;
> >
> > packet->q_idx = skb_get_queue_mapping(skb);
> > + if (netif_tx_queue_stopped(netdev_get_tx_queue(net, packet-
> >q_idx)))
> > + packet->xmit_more = false;
>
> Have you tested this condition ever triggers ?
>
> It should not.
>
> netvsc_start_xmit() should not be called if the queue is stopped.
>
> The problem is the following :
>
> netvsc_start_xmit() is called with the packet that is going to stop the
> queue, filling last slot, but with skb->xmit_more = 1;
>
> So you need to do something about xmit_more, after calling
> netif_tx_stop_queue(), not before.
Ah! I too was wondering how we could get into this situation. The condition you mention
is already handled in the lower level - if the attempt to put the last packet on vmbus were to
fail because the ring is full, we will notify the host independent of kick_q parameter - look
at the function vmbus_sendpacket_ctl() in drivers/hv/channel.c
I will resend the patch by reverting this to version 2.
Regards,
K. Y
>
>