Re: [Xen-devel] [PATCH v2 1/1] xen-netback: process malformed sk_buff correctly to avoid BUG_ON()

[Dongli Zhang <dongli.zhang@oracle.com>]

Hi Eric,

On 03/29/2018 12:03 PM, Eric Dumazet wrote:
> 
> 
> On 03/28/2018 08:51 PM, Dongli Zhang wrote:
>> The "BUG_ON(!frag_iter)" in function xenvif_rx_next_chunk() is triggered if
>> the received sk_buff is malformed, that is, when the sk_buff has pattern
>> (skb->data_len && !skb_shinfo(skb)->nr_frags). Below is a sample call
>> stack:
>>
>> ...
> 
> 
>>
>> The issue is hit by xen-netback when there is bug with other networking
>> interface (e.g., dom0 physical NIC), who has generated and forwarded
>> malformed sk_buff to dom0 vifX.Y. It is possible to reproduce the issue on
>> purpose with below sample code in a kernel module:
>>
>> skb->dev = dev; // dev of vifX.Y
>> skb->len = 386;
>> skb->data_len = 352;
>> skb->tail = 98;
>> skb->end = 384;
>> skb_shinfo(skb)->nr_frags = 0;
>> dev->netdev_ops->ndo_start_xmit(skb, dev);
>>
> 
> This would be a serious bug in the provider of such skb.

/nods

> 
> Are you sure you do not have instead an skb with a chain of skbs ?
> 
> (skb_shinfo(skb)->frag_list would be not NULL)

I am sure the skb_shinfo(skb)->frag_list is NULL.

> 
> Maybe your driver is wrongly advertising NETIF_F_FRAGLIST
> 
> commit 2167ca029c244901831 would be the bug origin then...

Unlike the new linux version (whose BUG_ON() does not panic the server), the
BUG_ON() in prior old kernel version would panic xen dom0 server and then people
would always blame xen paravirtual driver.

Indeed, xen-netback did not process the malformed sk_buff appropriately on rx
path. The issue is not hit with old dom0 kernel, when I am running the debug
module (as shown in below link) to generate a malformed sk_buff on purpose.

https://lists.xenproject.org/archives/html/xen-devel/2018-03/msg03176.html

Dongli Zhang