Re: [PATCH iwl] idpf: fix corrupted frames and skb leaks in singleq mode

[Alexander Lobakin]

From: Eric Dumazet <edumazet@xxxxxxxxxx>
Date: Thu, 7 Dec 2023 13:48:52 +0100

> On Thu, Dec 7, 2023 at 12:58 PM Simon Horman <horms@xxxxxxxxxx> wrote:
>>
>> On Fri, Dec 01, 2023 at 03:38:21PM +0100, Alexander Lobakin wrote:
>>> idpf_ring::skb serves only for keeping an incomplete frame between
>>> several NAPI Rx polling cycles, as one cycle may end up before
>>> processing the end of packet descriptor. The pointer is taken from
>>> the ring onto the stack before entering the loop and gets written
>>> there after the loop exits. When inside the loop, only the onstack
>>> pointer is used.
>>> For some reason, the logics is broken in the singleq mode, where the
>>> pointer is taken from the ring each iteration. This means that if a
>>> frame got fragmented into several descriptors, each fragment will have
>>> its own skb, but only the last one will be passed up the stack
>>> (containing garbage), leaving the rest leaked.
>>> Just don't touch the ring skb field inside the polling loop, letting
>>> the onstack skb pointer work as expected: build a new skb if it's the
>>> first frame descriptor and attach a frag otherwise.
>>>
>>> Fixes: a5ab9ee0df0b ("idpf: add singleq start_xmit and napi poll")
>>> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@xxxxxxxxx>
>>> Reviewed-by: Michal Kubiak <michal.kubiak@xxxxxxxxx>
>>> Signed-off-by: Alexander Lobakin <aleksander.lobakin@xxxxxxxxx>
>>
>> Reviewed-by: Simon Horman <horms@xxxxxxxxxx>
> 
> It seems singlequeue mode is not really used on idpf :)

>From what I know, there's currently no hardware supporting singleq mode.
I'd remove it completely, but seems like it was decided to keep it in
case someone would like to support it one day...

> 
> Reviewed-by: Eric Dumazet <edumazet@xxxxxxxxxx>

Thanks,
Olek