Re: pktgen and spin_lock_bh in xmit path

[Ben Greear <greearb@candelatech.com>]

On 10/20/2009 11:54 AM, Eric Dumazet wrote:
> Eric Dumazet a écrit :
>
>> David, we are changing skb->mapping while transmitting it...
>>
>> So yes, it can break pktgen if its skbs are cloned.
>>
>> Maybe pktgen should call skb_record_rx_queue() instead of skb_set_queue_mapping()
>>
>> Or maybe we should avoid this +/- 1 thing we do in skb_record_rx_queue(), since
>> we use same skb field for recording rx_queue or tx_queue :(
>>
>
> A possible fix would be :

I tried your patch.  It's not crashing, but the link is bouncing around:

Oct 20 13:13:34 localhost kernel: ixgbe 0000:05:00.0: master disable timed out
Oct 20 13:13:35 localhost kernel: ixgbe 0000:05:00.0: master disable timed out
Oct 20 13:13:35 localhost kernel: ixgbe 0000:05:00.0: master disable timed out
...

Oct 20 13:14:06 localhost kernel: ixgbe 0000:05:00.0: master disable timed out
Oct 20 13:14:06 localhost kernel: ixgbe: eth9 NIC Link is Down
Oct 20 13:14:06 localhost kernel: ixgbe: eth9 NIC Link is Up 10 Gbps, Flow Control: RX/TX
Oct 20 13:14:09 localhost kernel: ixgbe: eth8 NIC Link is Up 10 Gbps, Flow Control: RX/TX
Oct 20 13:14:12 localhost kernel: ixgbe 0000:05:00.1: master disable timed out
Oct 20 13:14:15 localhost kernel: ixgbe: eth9 NIC Link is Up 10 Gbps, Flow Control: RX/TX
Oct 20 13:14:19 localhost kernel: ixgbe: eth9 NIC Link is Down
Oct 20 13:14:19 localhost kernel: ixgbe 0000:05:00.0: master disable timed out


I'm not sure if this is another weirdness in my system or something
else...

Ben


-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com