On 14/04/14 17:51, David Miller wrote:
> From: David Vrabel <david.vrabel@citrix.com>
> Date: Mon, 14 Apr 2014 10:42:20 +0100
> 
>> On 12/04/14 21:51, David Miller wrote:
>>> From: Vincenzo Maffione <v.maffione@gmail.com>
>>> Date: Sat, 12 Apr 2014 11:55:40 +0200
>>>
>>>> This patch fixes the initialization of an array used in the TX
>>>> datapath that was mistakenly initialized together with the
>>>> RX datapath arrays. An out of range array access could happen
>>>> when RX and TX rings had different sizes.
>>>>
>>>> Signed-off-by: Vincenzo Maffione <v.maffione@gmail.com>
>>>
>>> Good catch, applied, thanks.
>>
>> Thanks.  You can queue this for net-next since the Tx and Rx rings are
>> the same constant size.
> 
> I was able to determine when I reviewed this patch that the size in bytes
> of the rings are the same (PAGE_SIZE), but I couldn't ascertain whether
> the individual ring entries in the TX ring and RX ring are the same size.
> 
> Are they?

Yes.  It's not at all obvious, but each ends up with 256 entries.

Tx entries are 12 bytes and Rx entries are 8 bytes.  The ring macros
reserve some space in the shared page for the producer/consumer indexes
/and/ round down the number to the next power of two.

David