[etherlab-dev] EtherCAT WARNING: 1 datagram UNMATCHED!

Florian Pose fp at igh-essen.com
Wed Jun 18 08:43:59 CEST 2008


Hi Olav,

On Mon, Jun 16, 2008 at 06:29:24PM +0200, Zarges, Olav wrote:
> No, but I reduced the packet-size which cures the problem. But instead
> of reducing ethernet utilization I even thought of sending 800 bytes every
> 100 us (~63 MBit/s) which should (?) be possible on a 1.6GHz system,
> shouldn't it?

it should.

It could be a DMA problem. Please try to increase the TX_RING_SIZE in
master/device.c from 2 to a higher value (10?).

>> Moreover I wonder why datagram 142 is in the queue before 141...
>
> You're right! I didn't notice in the first place.
>
> What can cause such a behaviour? Any race condition which scrambles the
> queue(s)? I guess I need to have a closer look to 
> ec_master_send_datagrams().

The datagram queue is a kernel list; insertions an deletions are
critical sections that have to be protected of parallel access. Do you
have multiple threads that call master functions?

-- 
Best regards,
Florian Pose

http://etherlab.org



More information about the Etherlab-dev mailing list