AM335x EtherCAT Linux check vendor id failed

Dimitar Ivanov

Other Parts Discussed in Thread: SYSBIOS

Hello,

I' running into a issue with TwinCAT and Sitara recognizing the vendor and product code IDs.

The situation: We have a custom designed board based on Sitara and Linux with EtherCAT running on it. For test and learn purposes we are using the ethercat sample application provided by sysbios.

The problem: The following errors are reported from TwinCAT after the bus and during the device initialization. Here is the message text

"Error 1 29.07.2015 15:45:50 152 ms | 'Box 1 (TIESC-001)' (1001): 'INIT to PREOP' failed! Error: 'check vendor id'. Comparison '0x00000011' == '0x0000059d' failed."

"Error 3 29.07.2015 15:45:50 457 ms | 'Box 1 (TIESC-001)' (1001): 'INIT to PREOP' failed! Error: 'check product code'. Comparison '0x0000059d' == '0x54490001' failed."

I'm little bit confused, since the IDs are the same. The vendor and the product IDs in the esi-file "TIS_ESC.xml"(TwinCAT) correspond to those in the header file "ecat_def.h".

Can you please give me a hint why does this error appear?

Thank you in Advance!

over 10 years ago

0 Frank Walzer over 10 years ago

TI__Mastermind 44306 points

Dimitar,
are you sure... the values to the right are the ones from ESI file (expected). And the other value is the one read by TwinCAT I think
So there is a mismatch. Did you try to just scan the device using TwinCAT and then look at the values reported? Looks like there is an offset in the values probably due to an build issue.

Also you should start to use your own numbers now (You are not allowed to use TI vendor ID in your product). As an ETG member you should have your own ID.

Regards,

0 Dimitar Ivanov over 10 years ago in reply to Frank Walzer

Intellectual 270 points

Hello Frank,

the IDs have been replaced by ours. Just to avoid misunderstanding : it was not our attention to use in our product. It was just a short evaluation in our lab to verify if at least a device can be recognized.

Regards,
Dimitar

0 Frank Walzer over 10 years ago in reply to Dimitar Ivanov

TI__Mastermind 44306 points

Dimitar,

lab use is ok of course. Did you solve the issue?

regards,

0 Dimitar Ivanov over 10 years ago in reply to Frank Walzer

Intellectual 270 points

Frank,

I'm still searching where this mismatch is coming from.

Regards

0 Dimitar Ivanov over 10 years ago in reply to Frank Walzer

Intellectual 270 points

Hello Frank,

it seems that the issue is solved, because the failure did not occur in the last 24 hours. Probably the reason was the switch connected between the slave and the master. Therefore I'll mark the question as answered .

Thank you.

Regards,
Dimitar

0 Pototo Clark over 9 years ago in reply to Dimitar Ivanov

Expert 1220 points

How does having the switch affects reading the eeprom? I have the same problem. It does not happen when I have one slave and one master connected, but it happens when I have more than one slave connected to a master. Why would this happen when there is a network where all my slave have the same vendor ID, etc.

0 Pototo Clark over 9 years ago in reply to Frank Walzer

Expert 1220 points

Where in the code specifically are we reading from eeprom and checking for vendor ID?

Processors

Processors forum

AM335x EtherCAT Linux check vendor id failed