Without a multiqueue this is not possible, when you send() the packet to the 
first queue, your actual buffer handle will point to a new buffer (a swap 
operation takes place, this is needed for zero-copy), thus with the second 
send() you are actually sending uninitialised data, instead sending to a 
multiqueue the same packet reference is passed to both queues.

Alfredo

> On 11 Aug 2015, at 16:20, Michael Nicolazzo <[email protected]> wrote:
> 
> Alfredo,
> 
> Thanks. One other question - what are the consequences if instead of a 
> multi-queue I queue the same packet to two different individual queues?
> 
> Thanks again,
> 
> Mike
> 
>> On Aug 11, 2015, at 10:16 AM, Alfredo Cardigliano <[email protected]> 
>> wrote:
>> 
>> 
>>> On 11 Aug 2015, at 15:45, Michael Nicolazzo <[email protected]> wrote:
>>> 
>>> Hi,
>>> 
>>> I have a question about the behavior of pfring_zc multi_queue. Suppose I 
>>> bundle two queues into a multi_queue. What happens if I send a buffer to 
>>> both queues, but one of the queues has an error, such as full? Will 
>>> send_pkt_multi queue one and and fail the other?
>> 
>> Correct
>> 
>>> If so, how can I tell which queue had the problem?
>> 
>> With the current API the only way is checking queue stats.
>> 
>> Alfredo
>> 
>>> 
>>> Thanks,
>>> 
>>> Mike
>>> _______________________________________________
>>> Ntop-misc mailing list
>>> [email protected]
>>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc
>> 
>> _______________________________________________
>> Ntop-misc mailing list
>> [email protected]
>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc
> 
> _______________________________________________
> Ntop-misc mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
Ntop-misc mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop-misc

Reply via email to