On 02/16/2018 11:43 AM, Sergei Shtylyov wrote:
> Hello!
>
> On 02/16/2018 10:42 PM, Florian Fainelli wrote:
>
>>> Allow for changing the MTU within the limit of the maximum size of a
>>> descriptor (2048 bytes). Add the callback to change MTU from user-space
>>> and take the configurable MTU into account when configuring the
>>> hardware.
>>>
>>> Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@xxxxxxxxxxxx>
>>> ---
>>
>>>
>>> +static int ravb_change_mtu(struct net_device *ndev, int new_mtu)
>>> +{
>>> + if (netif_running(ndev))
>>> + return -EBUSY;
>>> +
>>> + ndev->mtu = new_mtu;
>>> + netdev_update_features(ndev);
>>
>> Don't you somehow need to quiesce the RX DMA and make sure you that you
>> re-allocate all RX buffers within priv->rx_skb[q][entry] such that they
>> will be able to accept a larger buffer size?
>>
>> If we put the ravb interface under high RX load and we change the MTU on
>> the fly, can we crash the kernel?
>
> I thought the netif_running() check guards against that. Does it not?
It does, completely missed that, thanks!
--
Florian
