[dpdk-stable] [dpdk-dev] [PATCH] kni: fix crash for KNI interface remove

[Ferruh Yigit]

On 9/26/2016 1:15 PM, Yuanhan Liu wrote:
> On Mon, Sep 26, 2016 at 10:21:12AM +0100, Ferruh Yigit wrote:
>> On 9/26/2016 9:34 AM, Yuanhan Liu wrote:
>>> On Fri, Sep 09, 2016 at 03:46:01PM +0200, Thomas Monjalon wrote:
>>>> 2016-08-31 14:06, Ferruh Yigit:
>>>>> Removing KNI interface that has no PCI driver for ethtool support cause
>>>>> kernel crash.
>>>>>
>>>>> Fixes: 109febfe58f9 ("net/igb: move PCI device IDs from EAL")
>>>>> Fixes: 221fba3b987c ("net/ixgbe: move PCI device IDs from EAL")
>>>>>
>>>>> Signed-off-by: Ferruh Yigit <ferruh.yigit at intel.com>
>>>>
>>>> Applied, thanks
>>>
>>> FYI, this patch has been applied to stable branch v16.07.
>>>
>>> It hasn't been pushed to http://dpdk.org/browse/dpdk-stable/ yet.
>>> It will be pushed if I get no objections in around TWO days. So 
>>> please shutout if you have objections.
>>
>> If the patches mentioned in Fixes line applied to stable branch, this
>> one also should be applied.
> 
> Thanks for the response.
> 
> It might address the conflict issue, 

Which conflict issue?

> but with a simple glimpse, the
> build issue might still remain. Say, commit 109febfe58f9 ("net/igb: move
> PCI device IDs from EAL") will not be applied (as there is no fixline),

109febfe58f9 & 221fba3b987c are in fixes line of this patch.

> but it is required by commit 5544a453b4ac ("kni: fix crash when removing
> interface"), if I'm not mistaken.

I am not clear with question but, the issue is:
109febfe58f9 & 221fba3b987c creates a defect and this patch fixes it.

If both 109febfe58f9 or 221fba3b987c get into stable tree, this patch
needs to get in to fix crash. If you prefer to get one of them (although
that doesn't make much sense), a backported version of this patch needs
to be applied.

> 
> Or, I'm wrong?
> 
> 	--yliu
>