On Wed, Nov 11, 2015 at 02:40:45PM +0800, Jijiang Liu wrote:
> Add virtio guest offload capabilities.
>
> Signed-off-by: Jijiang Liu <jijiang.liu at intel.com>
> ---
> lib/librte_vhost/virtio-net.c | 5 ++++-
> 1 files changed, 4 insertions(+), 1 deletions(-)
>
> diff --git a/lib/librte_vhost/virtio-net.c b/lib/librte_vhost/virtio-net.c
> index 81bd309..839a333 100644
> --- a/lib/librte_vhost/virtio-net.c
> +++ b/lib/librte_vhost/virtio-net.c
> @@ -80,7 +80,10 @@ static struct virtio_net_config_ll *ll_root;
> (1ULL << VHOST_USER_F_PROTOCOL_FEATURES) | \
> (1ULL << VIRTIO_NET_F_HOST_TSO4) | \
> (1ULL << VIRTIO_NET_F_HOST_TSO6) | \
> - (1ULL << VIRTIO_NET_F_CSUM))
> + (1ULL << VIRTIO_NET_F_CSUM) | \
> + (1ULL << VIRTIO_NET_F_GUEST_CSUM) | \
> + (1ULL << VIRTIO_NET_F_GUEST_TSO4) | \
> + (1ULL << VIRTIO_NET_F_GUEST_TSO6))
I don't think we need that, and it might be wrong to set those fields at
vhost.
And, TBH, I am not 100% sure that I understand what those flags truely are
and for. All I know is that they seem have something to do with QEMU/TAP
device.
Hopefully the virtio expert, Michael, could shine some lights on that.
--yliu
>
> static uint64_t VHOST_FEATURES = VHOST_SUPPORTED_FEATURES;
>
> --
> 1.7.7.6
