i210 failed to program MAC address and checksum update

[PGeni]

Hi,

we've designed a new board based on ApolloLake that include two WGI210AT chips for network interfaces. Both i210 are unprogrammed version and used without external flash.

At boot they are correctly recognised with 0x1531 PCI ID (unprogrammed version). Using stuff from Technical Library items 572162, 348742 and 513655 i've programmed them with

./eeupdate64e /NIC=1 /INVMUPDATE /FILE=I210_Invm_Copper_NoAPM_v0.6.txt

and I got

Update INVM content ... Verify autoload configuration ... done.

After a reboot the PCI ID is changed to 0x157B, so supposedly correctly programmed.

But when I try to configure MAC address with

./eeupdate64e /NIC=1 /MAC=xxxxxxxxxxxx

I got

1: Updating Mac Address to xxxxxxxxxxxx...Failed!

as well when I try to update the checksum with

./eeupdate64e /NIC=1 /CALCCHKSUM

I got

1: Updating Checksum and CRCs...Failed.

Using

./eeupdate64e /NIC=1 /INVMISLOCKED

NVM seems not locked

At linux boot igb driver is not probed:

[   3.504251] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.4.0-k

[   3.504254] igb: Copyright (c) 2007-2014 Intel Corporation.

[   3.516047] igb 0000:02:00.0: The NVM Checksum Is Not Valid

[   3.824168] igb: probe of 0000:02:00.0 failed with error -5

[   3.836254] igb 0000:03:00.0: The NVM Checksum Is Not Valid

[   4.116297] igb: probe of 0000:03:00.0 failed with error -5

I've also faced anothed strange output using EepromAccessTool

inside package Id 572162. The output is:

Intel(R) EEPROM Access Tool NVM/OTP Programming Example Tool  Version 0.7.9

Provided under the terms of a CNDA. Do Not Distribute.

Copyright(C) 2017-2019 by Intel(R) Corporation

NIC   BUS   DEV   FUN   Silicon   Memory Type Present

===   ===   ===   ===   =====   ======================

Updating command word..... Done!

 1   2    0    0   I210       INVM+FLASH

Updating command word..... Done!

 2   3    0    0   I210       INVM+FLASH

But I really do not have FLASH chip on board… according to design guide pin 14 has been pulled up.

Am I wrong somewhere? How can I get i210 chips correctly programmed?

Thanks in advance

[CarlosAM_INTEL]

Hello, @PGeni​:

Thank you for contacting Intel Embedded Community.

In order to be on the same page, could you please let us know the sources (documents, schematics, and others) that you have used to develop the affected design? Please include document number, revision, and version of the sources.

Could you please inform us if this design has been reviewed by Intel?

Please give us the information that should answer these questions.

Best regards,

@Mæcenas_INTEL​.

[PGeni]

Hello @Mæcenas_INTEL​ ,

sure...

we've used Oxbow Hill CRB schematics as reference for the overall design, including I210 section (Technical Library #562639)

and specifically for I210 design and programming:

• 513305-i210-dual-design-guide-v-1-1.pdf
• 348742-quartzville-tools-563038.zip
• 572162-eepromaccesstool-0-7-9.zip
• ethernet-controller-i210-i211-faq.pdf

Yes, the design has been reviewed by Intel; I can give you the name of Intel Technical Account Manager that supported us during initial design. I tried to contact him on this topic few weeks ago without success (we've some problems here in Italy now…)

Kind regards,

Paolo

[CarlosAM_INTEL]

Hello, @PGeni​:

Thanks for your update.

Based on your previous communications, could you please verify that the affected design is unrelated to the condition 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?This document can be found at:

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

We are waiting for your verification.

Best regards,

@Mæcenas_INTEL​.

[PGeni]

Hello, @Mæcenas_INTEL​ 

thanks for your reply.

Yes, I've double checked right now question 2.23 on pointed FAQ document, but it can't help.

I've tried with a 1kohm pull-down applied to pin 12, but i still got a failure both on MAC programming

./eeupdate64e /NIC=1 /MAC=xxxxxxxxxxxxx

and checksum calculation

./eeupdate64e /NIC=1 /CALCCHKSUM

I've also had a look to question 2.24, and referred section 3.2 of i211 datasheet, but I do not find new ideas.

May be that my problem is related to iNVM security feature, but:

• with ./eeupdate64e /NIC=1 /INVMISLOCKED i got "INVM content is UNLOCKED !"
• question 2.23 of FAQ document states that MAC address field can be updated also when security feature is enabled

Best regards

Paolo

[CarlosAM_INTEL]

Hello, @PGeni​:

Thanks for your reply.

We want more information to have a better idea of this situation, we need to address the following questions: Could you please let us know how many units of the project related to this circumstance have been manufactured? How many are affected? Could you please give the failure rate?

Could you please give pictures of the top side markings of the affected processors and Ethernet controllers?

We are waiting for your reply to these questions.

Best regards,

@Mæcenas_INTEL​.

[PGeni]

Hello @Mæcenas_INTEL​ 

we' re in prototype phase of a new board design, so I've not so many working samples, but I have two boards with two i210 chip each exactly in same situation (and of course I've programmed them with same procedure described into original post).

This is really the last open point that currently avoid us to deliver the prototype samples for next production step.

Attached you can find a hi-res photo of the full board (U43 and U44 are i210 chips)

Hope this can help you to give us some help...

Kind regards

Paolo

[CarlosAM_INTEL]

Hello, @PGeni​:

Thanks for your update.

Reviewing the picture provided in your previous message, we found based on the top side markings specifically the S Spec information of the processor that it is not an Apollo Lake as you may notice in the following screenshot:

Due to this fact, we suggest you review again your implementation.

Best regards,

@Mæcenas_INTEL​.

[PGeni]

Hello @Mæcenas_INTEL​,

yes… we've different board versions and yesterday i've posted a picture of a different one.

Let me re-attach the picture of the board that is currently on my lab desk… this should be really an ApolloLake….

Sorry for this inconvenience

Hoping that you can investigate now my i210 programming issue...

Paolo

[CarlosAM_INTEL]

Hello, @PGeni​:

Thanks for your reply.

In order to have more information related to this situation, could you please clarify if you use Intel(R) LANConf tool to program the MAC address? The information that may help you is stated in Table 8, on page 22 of the Intel(R) LANConf User Manual included in the DOCS file of the Intel(R) Network Connections Tools 24.0 PV LAN Software Tools document # 348742, which is also where may find the cited tool. It can be found when you are logged into your Resource & Design Center (RDC) privileged account on the following website:

 http://www.intel.com/cd/edesign/library/asmo-na/eng/348742.htm

The RDC Account Support form is the channel to process your account update request. It can be found at:

https://www.intel.com/content/www/us/en/forms/support/my-intel-sign-on-support.html

We are waiting for your answer to this question.

Best regards,

@Mæcenas_INTEL​.

[PGeni]

Hello @Mæcenas_INTEL​ ,

Yes.. I already have privileged account for Intel Technical Library access and I already have downloaded and used document #348742 (see my post on Apr 21).

Here some output from lanconf tool used in text mode:

Enter command> MACADDR 821122334401 ERROR: MAC Address write failed Enter command> RESET Successful adapter reset Enter command> MACADDR 821122334401 ERROR: MAC Address write failed Enter command> DISPLAY Adapter:........... Intel(R) I210 Gigabit Network Connection MAC Address:....... 00-00-00-00-00-00 Media Type:........ Copper Bus Location:...... 3:00.0 Link Type:......... Not Linked Enter command> SCAN 01) Intel(R) I210 Gigabit Network Connection 8086-157B 2:00.0 02) Intel(R) I210 Gigabit Network Connection 8086-157B 3:00.0 Enter command> VER LANConf v1.34.22.5 Enter command> EEPROM Test started. Press ESC or 'x' to stop current test...PASSED

As you can see MACADDR update still fails; tested both with and without pulldown on i210 pin 12.

Hope that this can help you to drive us to a solution.

Kind regards,

Paolo

[CarlosAM_INTEL]

Hello, @PGeni​:

Thanks for your reply.

Based on your communications, could you please let us know if you are seeing the same issue when you try the latest IGB driver?

We are waiting for your answer.

Best regards,

@Mæcenas_INTEL​.