i210 eeupdate64e programming alongside a similar ComExpress solution

[JMerc4]

I've got a ComExpress carrier board that accommodates a board with either a i210 or i211.

On the carrier board itself, I also have a i210 with its own flash.

Programming my own i210 using EEupdate64e works well, specifying bus# and device # . Everything works on the network with both network devices, until I replace the ComExpress board with another new card.

Then it seems that my on-carrier i210 loses its programming, and I need to reprogram the carrier card 's i210 mac address.

I noticed a reference to "shared flash" during the EEupdate64e process.

Is the flash program updating a single flash on the ComExpress and not the i210's flash on the carrier?

If so, is Linux and the i210 driver somehow taking the responsibility of doing register setup instead of the power-up flash read, and is there a way to prevent this?

[CarlosAM_INTEL]

Hello, JeffMercure:

Thank you for contacting Intel Embedded Community.

Based on your previous communication, could you please clarify if the affected project has been manufactured by you or by a third-party company? In case that it is a third-party device, could you please give us the part number, model, and where to find its documentation?

Thanks in advance for your cooperation.

Best regards,

Carlos_A.

[JMerc4]

The base card has been manufactured by us, the Com Express board is a common Type-6. Any manufacturer like Kontron or Adlink shows the same oddity when it is replaced.

Kontron COMe-mAL10 E2 E3930 4E/8S shows it for instance. It is a E3930 card. ( https://www.kontron.com/products/boards-and-standard-form-factors/com-express/com-express-mini/come-mal10-e2-.html https://www.kontron.com/products/boards-and-standard-form-factors/com-express/com-express-mini/come-mal10-e2-.html )

For clarification, when the Com Express board is replaced with a new one, the base card's i210 MAC address reverts to something default, and the PCI device id reverts back to 1531 rather than the correct 1533 denoting a programmed device.

[CarlosAM_INTEL]

Hello, JeffMercure :

Thanks for your update.

Based on your previous communication, could you please tell us if the cited issue is related to a specific COM Express board or the problem persists with boards manufactured by different developers? Please give us all the information related to the affected boards.

Thanks in advance for your clarification.

Best regards,

Carlos_A.

[JMerc4]

Any manufacturer like Kontron or Adlink shows the same oddity when it is replaced.

Maybe a simpler question is: What is the implication of the "updating shared flash" message coming from the EEupdate program during a MAC address programming?

It seems to imply exactly what I am worried about: A single flash location governing both Ethernet controllers, instead of each controller acting on its own with its own flash...

[MJeff]

Hi Jeff

What is the command line you are using for eeupdate.

You need to program the appropriate binary file from the production images package into a blank i210 before it will work.

[JMerc4]

My problem is not programming and operating the i210. We have been using the same i210 design and flash file & mac address programming process for several years with the same process. (Eeupdate64e /BUS=1 /DEV=0 /MAC=xxyyzzqqwwvv & Eeupdate64e /BUS=1 /DEV=0 /DATA filename.BIN) I have no problem using the i210.

My problem is that after this is done, if we replace the Com Express board in the system (the one that has a PCIe channel to the i210, and that has its own i210 or i211 Ethernet controller on it), I lose my base board's MAC address and programming, and it shows up as a 1531 device in PCI space rather than the programmed 1533 device.

This leads me to believe that when I execute EEUPDATE64E, I am really programming a shared device somehow on the ComExpress assembly and NOT on my i210 base board design. The extra evidence that I have is that during the programming process, there is a "programming shared flash" notice. I am trying to get confirmation that that is indeed the issue, and if so, is there a way around it, so that replacing the ComExpress module doesn't require reprogramming my base card's mac address and config.

[CarlosAM_INTEL]

Hello, JeffMercure :

Thanks for your updates.

Based on your previous communications, could you please confirm us that you are following the guidelines stated in the answers to questions 2.15 and 2.25, on pages 8 and 9 of the https://www.intel.com/content/dam/www/public/us/en/documents/faqs/ethernet-controller-i210-i211-faq.pdf Intel(R) Ethernet Controller I210/I211 Frequently Asked Questions [FAQs] document # 335346?

Waiting for your reply.

Best regards,

Carlos_A.

[CarlosAM_INTEL]

Hello, JeffMercure:

By the way, could you please clarify the carrier board has a flash and this second card does?

Also, after writing with EEupdate, could you please tell us if the power is being cycled?

Waiting for your answer to these and previous questions.

Best regards,

Carlos_A.