RE: [PATCH V3,net-next, 1/2] hv_netvsc: Add XDP support

[Haiyang Zhang]

> -----Original Message-----
> From: Jakub Kicinski <kuba@xxxxxxxxxx>
> Sent: Thursday, January 23, 2020 11:59 AM
> To: Haiyang Zhang <haiyangz@xxxxxxxxxxxxx>
> Cc: sashal@xxxxxxxxxx; linux-hyperv@xxxxxxxxxxxxxxx; netdev@xxxxxxxxxxxxxxx;
> KY Srinivasan <kys@xxxxxxxxxxxxx>; Stephen Hemminger
> <sthemmin@xxxxxxxxxxxxx>; olaf@xxxxxxxxx; vkuznets
> <vkuznets@xxxxxxxxxx>; davem@xxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH V3,net-next, 1/2] hv_netvsc: Add XDP support
> 
> On Wed, 22 Jan 2020 09:23:33 -0800, Haiyang Zhang wrote:
> > This patch adds support of XDP in native mode for hv_netvsc driver,
> > and transparently sets the XDP program on the associated VF NIC as well.
> >
> > Setting / unsetting XDP program on synthetic NIC (netvsc) propagates
> > to VF NIC automatically. Setting / unsetting XDP program on VF NIC
> > directly is not recommended, also not propagated to synthetic NIC, and
> > may be overwritten by setting of synthetic NIC.
> >
> > The Azure/Hyper-V synthetic NIC receive buffer doesn't provide
> > headroom for XDP. We thought about re-use the RNDIS header space, but
> > it's too small. So we decided to copy the packets to a page buffer for
> > XDP. And, most of our VMs on Azure have Accelerated  Network (SRIOV)
> > enabled, so most of the packets run on VF NIC. The synthetic NIC is
> > considered as a fallback data-path. So the data copy on netvsc won't
> > impact performance significantly.
> >
> > XDP program cannot run with LRO (RSC) enabled, so you need to disable
> > LRO before running XDP:
> >         ethtool -K eth0 lro off
> >
> > XDP actions not yet supported:
> >         XDP_REDIRECT
> >
> > Signed-off-by: Haiyang Zhang <haiyangz@xxxxxxxxxxxxx>
> >
> > ---
> > Changes:
> > 	v3: Minor code and comment updates.
> >         v2: Added XDP_TX support. Addressed review comments.
> 
> How does the locking of the TX path work? You seem to be just calling the
> normal xmit method, but you don't hold the xmit queue lock, so the stack can
> start xmit concurrently, no?

The netvsc and vmbus can handle concurrent transmits, except the msd 
(Multi-Send Data) field which can only be used by one queue. 

I already added a new flag to netvsc_send(), so packets from XDP_TX won't use 
the msd.

Thanks,
- Haiyang