On 03/02/16 11:28, Bruce Richardson wrote:

On Fri, Dec 04, 2015 at 02:18:34PM -0500, Eric Kinzie wrote:

quoted

On Fri Dec 04 19:36:09 +0100 2015, Andriy Berestovskyy wrote:

quoted

Hi guys,
I'm not quite sure if we can support less TX queues on a slave that easy:

quoted

queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id);
num_tx_slave = rte_eth_tx_burst(slaves[i], queue_id,
      slave_bufs[i], slave_nb_pkts[i]);

It seems that two different lcores might end up writing to the same
slave queue at the same time, isn't it?

Regards,
Andriy

Andriy, I think you're probably right about this.  Perhaps it should
instead refuse to add or refuse to activate a slave with too few
tx queues.  Could probably fix this with another layer of buffering
so that an lcore with a valid tx queue could pick up the mbufs later,
but this doesn't seem very appealing.

Eric

Hi Eric, Stephen, Declan,

all patches of the set apart from this one and the next (nos 6 & 7) have no
comments and have been acked. Is there a resolution on these two patches, so they
can be acked and merged?

Regards,
/Bruce

Hey Bruce, Eric, Stephen, sorry about leaving this patchset hanging 
around. Can you apply patches 1-5 & patch 8 in this patch set. I've 
reviewed and acked all of those patches and I believe they are good tof 
go. I need to give further feedback on patches 6 and 7, as I would like 
to avoid bring further rte_ring buffering into the bonded device if 
possible and I think this should be possible but I haven't had time to 
prototype any alternatives but that shouldn't stop the other patches 
being applied.

Thanks
Declan