[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20081219181815.39dfb565@extreme>
Date: Fri, 19 Dec 2008 18:18:15 -0800
From: Stephen Hemminger <shemminger@...tta.com>
To: Eric Dumazet <dada1@...mosbay.com>
Cc: "David S. Miller" <davem@...emloft.net>,
Linux Netdev List <netdev@...r.kernel.org>
Subject: Re: [net-next-2.6 BUG] VLAN out of service ?
On Fri, 19 Dec 2008 20:46:18 +0100
Eric Dumazet <dada1@...mosbay.com> wrote:
> I just tried some VLAN settings on my machine and got an immediate crash on latest net-next-2.6
>
> ip link set eth2 up
>
> ip link add link eth2 eth2.825 type vlan id 825
> ip addr add 10.170.73.104/25 dev eth2.825
> ip link set eth2.825 up
>
> ip route add 10.170.72.0/24 via 10.170.73.126
>
> ping 10.170.73.126
> <<crash>>
> Code: Bad EIP value.
> EIP: [<00000000>] 0x0 SS:ESP 0068:f707bdb4
> Kernel panic - not syncing: Fatal exception in interrupt
>
> crash while jumping to ops->ndo_start_xmit(skb, dev);
>
> CONFIG_COMPAT_NET_DEV_OPS is set in my .config, eth2 is a tg3
>
> static const struct net_device_ops vlan_netdev_ops
> has no ndo_start_xmit initializer.
>
> Thanks
>
> [PATCH] vlan: fix convertion to net_device_ops
>
> commit 656299f706e52e0409733d704c2761f1b12d6954
> (vlan: convert to net_device_ops) added a net_device_ops
> with a NULL ndo_start_xmit field.
>
> This gives a crash in dev_hard_start_xmit()
>
> Fix it using two net_device_ops structures, one for hwaccel vlan,
> one for non hwaccel vlan.
>
> Signed-off-by: Eric Dumazet <dada1@...mosbay.com>
> ---
> diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
> index 71193a6..89a3bbd 100644
> --- a/net/8021q/vlan_dev.c
> +++ b/net/8021q/vlan_dev.c
> @@ -593,6 +593,8 @@ static const struct header_ops vlan_header_ops = {
> .parse = eth_header_parse,
> };
>
> +static const struct net_device_ops vlan_netdev_ops, vlan_netdev_accel_ops;
> +
> static int vlan_dev_init(struct net_device *dev)
> {
> struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
> @@ -619,11 +621,11 @@ static int vlan_dev_init(struct net_device *dev)
> if (real_dev->features & NETIF_F_HW_VLAN_TX) {
> dev->header_ops = real_dev->header_ops;
> dev->hard_header_len = real_dev->hard_header_len;
> - dev->hard_start_xmit = vlan_dev_hwaccel_hard_start_xmit;
> + dev->netdev_ops = &vlan_netdev_accel_ops;
> } else {
> dev->header_ops = &vlan_header_ops;
> dev->hard_header_len = real_dev->hard_header_len + VLAN_HLEN;
> - dev->hard_start_xmit = vlan_dev_hard_start_xmit;
> + dev->netdev_ops = &vlan_netdev_ops;
> }
>
> if (is_vlan_dev(real_dev))
> @@ -704,6 +706,22 @@ static const struct net_device_ops vlan_netdev_ops = {
> .ndo_uninit = vlan_dev_uninit,
> .ndo_open = vlan_dev_open,
> .ndo_stop = vlan_dev_stop,
> + .ndo_start_xmit = vlan_dev_hard_start_xmit,
> + .ndo_validate_addr = eth_validate_addr,
> + .ndo_set_mac_address = vlan_dev_set_mac_address,
> + .ndo_set_rx_mode = vlan_dev_set_rx_mode,
> + .ndo_set_multicast_list = vlan_dev_set_rx_mode,
> + .ndo_change_rx_flags = vlan_dev_change_rx_flags,
> + .ndo_do_ioctl = vlan_dev_ioctl,
> +};
> +
> +static const struct net_device_ops vlan_netdev_accel_ops = {
> + .ndo_change_mtu = vlan_dev_change_mtu,
> + .ndo_init = vlan_dev_init,
> + .ndo_uninit = vlan_dev_uninit,
> + .ndo_open = vlan_dev_open,
> + .ndo_stop = vlan_dev_stop,
> + .ndo_start_xmit = vlan_dev_hwaccel_hard_start_xmit,
> .ndo_validate_addr = eth_validate_addr,
> .ndo_set_mac_address = vlan_dev_set_mac_address,
> .ndo_set_rx_mode = vlan_dev_set_rx_mode,
That works (or there could just be one vlan_dev_start_xmit) that handled
both cases.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists