How to debug unsupported flash attached to i210-iT?

[Jdu_P]

We're trying to set up our Copper i210 (-iT) with an SST25PF040C 4 Mb flash.

1. Flash is detected and accessible in boundary scan.
2. LanConf utility does not detect flash. SST25VF040 is on the supported device list, but SST25PF040 is not, is this the problem?
3. Controller enumerates correctly in Linux (with lspci).
4. Can the i210 be operated in flash-less mode? What steps should be taken to do this?

[Caguicla_Intel]

Hello Jdu P,

Thank you for posting in Intel Ethernet Communities.

Your query will be best answered by our Embedded Support team, we will help you to move this post to the designated team. Please feel free to contact us if you need assistance from Ethernet support team.

Best regards,

Crisselle C

Intel® Customer Support

[CarlosAM_INTEL]

Hello, @Jdu P​:

Thank you for contacting Intel Embedded Community.

We suggest verify that the affected designs fulfill the requirements stated in the answer to the question 2.23, on page 9 of the Intel(R) Ethernet Controller I210/I211 Frequently Asked Questions [FAQs] document # 335346. Also, the information that may reply your last consultation is stated in the respond to interrogation 2.9, on page 6 of the cited FAQs document. This document can be found at:

https://www.intel.com/content/dam/www/public/us/en/documents/faqs/ethernet-controller-i210-i211-faq.pdf

Best regards,

@Mæcenas_INTEL​.

[Jdu_P]

Thanks for the feedback @Mæcenas_INTEL​ 

We use the i210-iT in a copper-only (1000BASE-T) application and we only program our MAC address to the device.

1. I have been able to write the iNVM image supplied in the package 513655-i210-i211-production-images (specifically, I210_Invm_Copper_APM_v0.6.txt) with LanConf.

2. After doing this, the device shows up with the correct device ID that corresponds to an iNVM-only configuration.

3. Rebooting Linux and inspecting the kernel output shows that the iNVM checksum is incorrect and the device does not initialise.

4. I used the eeupdate utility to calculate and program the checksum.

5. After another reboot, the kernel output shows that the MAC address is invalid. I can then assign a MAC address with eeupdate using the /mac command.

6. Reloading the igb driver (with rmmod and then insmod) is then successful and the controller can connect to our network at 1 Gbps.

7. However, doing a complete power cycle of the board seems to erase the iNVM checksum, since rebooting into Linux reports an invalid iNVM checksum again. After running the eeupdate GUI to re-calculate the checksum and reloading the kernel driver fixes the problem.

What should be done to fix the iNVM checksum problem? Why does it reset after power cycle?

Kind regards

[CarlosAM_INTEL]

Hello, @Jdu P​:

Thank you for contacting Intel Embedded Community.

Could you please clarify if you verify our previous suggestions?

we are waiting for your reply.

Best regards,

@Mæcenas_INTEL​.

[Jdu_P]

Hi,

We've decided to modify the PCBs by pulling the NVM SI pin low during boot. This solves the checksum problem.

Regards,

J