vdpa/mlx5: should exclude header length and fcs from mtu
authorSi-Wei Liu <si-wei.liu@oracle.com>
Thu, 8 Apr 2021 09:10:43 +0000 (12:10 +0300)
committerMichael S. Tsirkin <mst@redhat.com>
Fri, 9 Apr 2021 16:08:27 +0000 (12:08 -0400)
commitd084d996aaf53c0cc583dc75a4fc2a67fe485846
treec2abe0c58fa5b3acfc05bae0bd22a0910ffc514f
parente49d033bddf5b565044e2abe4241353959bc9120
vdpa/mlx5: should exclude header length and fcs from mtu

When feature VIRTIO_NET_F_MTU is negotiated on mlx5_vdpa,
22 extra bytes worth of MTU length is shown in guest.
This is because the mlx5_query_port_max_mtu API returns
the "hardware" MTU value, which does not just contain the
 Ethernet payload, but includes extra lengths starting
from the Ethernet header up to the FCS altogether.

Fix the MTU so packets won't get dropped silently.

Fixes: 1a86b377aa21 ("vdpa/mlx5: Add VDPA driver for supported mlx5 devices")
Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eli Cohen <elic@nvidia.com>
Link: https://lore.kernel.org/r/20210408091047.4269-2-elic@nvidia.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
drivers/vdpa/mlx5/core/mlx5_vdpa.h
drivers/vdpa/mlx5/net/mlx5_vnet.c