Re: [PATCH net] hv_netvsc: Restore needed_headroom request

[Vitaly Kuznetsov]

Vitaly Kuznetsov <vkuznets@xxxxxxxxxx> writes:

> Commit c0eb454034aa ("hv_netvsc: Don't ask for additional head room in the
> skb") got rid of needed_headroom setting for the driver. With the change I
> hit the following issue trying to use ptkgen module:
>
> [   57.522021] kernel BUG at net/core/skbuff.c:1128!
> [   57.522021] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC
> ...
> [   58.721068] Call Trace:
> [   58.721068]  [<ffffffffa0144e86>] netvsc_start_xmit+0x4c6/0x8e0 [hv_netvsc]
> ...
> [   58.721068]  [<ffffffffa02f87fc>] ? pktgen_finalize_skb+0x25c/0x2a0 [pktgen]
> [   58.721068]  [<ffffffff814f5760>] ? __netdev_alloc_skb+0xc0/0x100
> [   58.721068]  [<ffffffffa02f9907>] pktgen_thread_worker+0x257/0x1920 [pktgen]
>
> Basically, we're calling skb_cow_head(skb, RNDIS_AND_PPI_SIZE) and crash on
>     if (skb_shared(skb))
>         BUG();
>
> We probably need to restore needed_headroom setting (but shrunk to
> RNDIS_AND_PPI_SIZE as we don't need more) to request the required headroom
> space. In theory, it should not give us performance penalty.

I'm sorry for the ping but this is kind of a regression and it would be
nice to have it fixed in 4.5.

Here is a script to trigger kernel crash:
#! /bin/sh
modprobe pktgen

echo "rem_device_all" > /proc/net/pktgen/kpktgend_0
echo "add_device eth0" > /proc/net/pktgen/kpktgend_0
echo "pkt_size 60" > /proc/net/pktgen/eth0
echo "clone_skb 1000000" > /proc/net/pktgen/eth0
echo "count 100000" > /proc/net/pktgen/eth0
echo "dst <SOME_IP>" > /proc/net/pktgen/eth0
echo "dst_mac <SOME_MAC>" > /proc/net/pktgen/eth0
echo "start" > /proc/net/pktgen/pgctrl
cat /proc/net/pktgen/eth0

Please review.

>
> Signed-off-by: Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>
> ---
>  drivers/net/hyperv/netvsc_drv.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
> index 1d3a665..98e34fe 100644
> --- a/drivers/net/hyperv/netvsc_drv.c
> +++ b/drivers/net/hyperv/netvsc_drv.c
> @@ -1089,6 +1089,9 @@ static int netvsc_probe(struct hv_device *dev,
>  	net->ethtool_ops = &ethtool_ops;
>  	SET_NETDEV_DEV(net, &dev->device);
>
> +	/* We always need headroom for rndis header */
> +	net->needed_headroom = RNDIS_AND_PPI_SIZE;
> +
>  	/* Notify the netvsc driver of the new device */
>  	memset(&device_info, 0, sizeof(device_info));
>  	device_info.ring_size = ring_size;

-- 
  Vitaly