On Wed, 2017-11-15 at 10:45 -0600, Bryant G. Ly wrote:
> This patch just closes the window, bad things can still happen. I wanted to
> leave it
> up to the people who actively develop in ibmveth to close the window, since
> introducing
> a lock can be expensive in tx.
You don't need to instrodu
On 11/14/17 8:47 PM, Daniel Axtens wrote:
> Hi Bryant,
>
> This looks a bit better, but...
>
>> The following patch ensures that the bounce_buffer is not null
>> prior to using it within skb_copy_from_linear_data.
> How would this occur?
>
> Looking at ibmveth.c, I see bounce_buffer being freed in
On Wed, 2017-11-15 at 13:47 +1100, Daniel Axtens wrote:
> Hi Bryant,
>
> This looks a bit better, but...
>
> > The following patch ensures that the bounce_buffer is not null
> > prior to using it within skb_copy_from_linear_data.
>
> How would this occur?
>
> Looking at ibmveth.c, I see bounce_
Hi Bryant,
This looks a bit better, but...
> The following patch ensures that the bounce_buffer is not null
> prior to using it within skb_copy_from_linear_data.
How would this occur?
Looking at ibmveth.c, I see bounce_buffer being freed in ibmveth_close()
and allocated in ibmveth_open() only.
The following patch ensures that the bounce_buffer is not null
prior to using it within skb_copy_from_linear_data.
The problem can be recreated toggling on/off Large send offload.
The following script when run (along with some iperf traffic recreates the
crash within 5-10 mins or so).
while true
On 11/13/17 7:07 PM, Daniel Axtens wrote:
> Hi Bryant,
>
> A few things:
>
> 1) The commit message could probably be trimmed, especially the stack
> traces.
Yes, I can trim it.
> 2) What exactly are you changing and why does it fix the issue? I
> couldn't figure that out from the commit message
Hi Bryant,
A few things:
1) The commit message could probably be trimmed, especially the stack
traces.
2) What exactly are you changing and why does it fix the issue? I
couldn't figure that out from the commit message.
3) Does this need a Fixes: tag?
> }
>
> - netdev->min_mtu = IB
The following script when run (along with some iperf traffic recreates the
crash within 5-10 mins or so).
while true
do
ethtool -k ibmveth0 | grep tcp-segmentation-offload
ethtool -K ibmveth0 tso off
ethtool -k ibmveth0 | grep tcp-segmentation-offload
ethtool -K ibm
