[etherlab-dev] r1713 - Distributed Clocks IOCTL functions

[Florian Pose]

Hi,

On Fri, Apr 24, 2009 at 12:37:42PM +0200, Erwin Burgstaller wrote:
> kernel: [ 2096.611309] EtherCAT ERROR: Failed to get DC times of slave 0: No response.
> kernel: [ 2096.743109] EtherCAT ERROR: Failed to get DC times of slave 13: No response.
> 
> slave 0 is an Beckhoff EK1100 and 13 is our own device. Both devices
> were stuck in state 'INIT E', which is not a real problem with the
> EK1100, 'cause I've nothing to talk with, but obviously a serious one
> for the other device.

strange. Reading the DC system time is only done, if the slave tells the master,
that it supports DC. An EK1100 is a topology device, the knowledge base
tell us:

2. Slaves supporting only propagation delay measurement:
Mandatory for ESCs with 3 or more ports (topology devices like EK1100
and ET1100). Local clock and receive time stamps are supported.

The EK1100's I have, do not show this behavior. Anyway, it wouldn't be
nice not supporting older devices, so let's try to find a solution. ;-)

The datagram that is not responded, reads the registers 0x910, 0x918 and
0x920 (0x910-0x927, "System Time", "Receive Time ECAT Processing Unit"
and "System Time Offset"). Maybe it is a problem with reading all three
parameters together, because the slave does not support the "Receive
Time ECAT Processing Unit" parameter... We don't need this one anyway,
so I'll try to split the reading in two datagrams. I'm preparing a new
tarball...

-- 
Best regards,
Florian Pose

http://etherlab.org