AM6412: Clarification needed for Errata i2409

Koichiro Tashiro

Part Number: AM6412

Hi,

AM64x Errata document was updated in Dec. 2023 and i2409 was added.

My customer has some questions.

Q1) When the workaround will be implemented in TI SDK if not already available ?
Q2) There is no description for "suspend_residency_enable" nor "SUSP_CTRL" in TRM. What are they?
Q3) It is not clear when USB entering suspend or when a USB wake-up event is triggered. Please explain the mechanism.

Thanks and regards,
Koichiro Tashiro

over 1 year ago

0 Bin Liu over 1 year ago

TI__Guru**** 168501 points

Hi Tashiro-san,

Koichiro Tashiro said:
Q3) It is not clear when USB entering suspend or when a USB wake-up event is triggered. Please explain the mechanism.

The suspend/resume is done between the USB controller and the PHY automatically according to the definition in the USB Specifications. Most of the details are transparent to software.

Koichiro Tashiro said:

Q1) When the workaround will be implemented in TI SDK if not already available ?

I assume you meant the MCU+ SDK. I am routing your query to our MCU+ SDK expert for comments.

0 Koichiro Tashiro over 1 year ago in reply to Bin Liu

TI__Mastermind 22945 points

Hi Bin,

Bin Liu said:
The suspend/resume is done between the USB controller and the PHY automatically according to the definition in the USB Specifications. Most of the details are transparent to software.

Do you mean the suspend/resume mechanism is something related to USB specification. Not AM64x suspend/resume features, correct?

Bin Liu said:
I assume you meant the MCU+ SDK. I am routing your query to our MCU+ SDK expert for comments.

I meant Linux SDK.

Thanks and regards,
Koichiro Tashiro

0 Bin Liu over 1 year ago in reply to Koichiro Tashiro

TI__Guru**** 168501 points

Hi Tashiro-san,

Koichiro Tashiro said:
Do you mean the suspend/resume mechanism is something related to USB specification. Not AM64x suspend/resume features, correct?

Correct.

Koichiro Tashiro said:
I meant Linux SDK.

I just checked the internal tracking The schedule of the fix in kernel has not been planned yet.

0 Koichiro Tashiro over 1 year ago in reply to Bin Liu

TI__Mastermind 22945 points

Hi Bin,

I guess i2409 impacts to all customers who use USB, correct?
If so, the workaround is needed ASAP.

Thanks and regards,
Koichiro Tashiro

0 Bin Liu over 1 year ago in reply to Koichiro Tashiro

TI__Guru**** 168501 points

Hi Tashiro-san,

Koichiro Tashiro said:
I guess i2409 impacts to all customers who use USB, correct?

Yes.

Koichiro Tashiro said:
If so, the workaround is needed ASAP.

I have pinged the sw dev team, the work is started. The implementation will be done for AM62x first, then followed by AM64x. I will keep you posted.

0 liu yuchen over 1 year ago in reply to Bin Liu

Prodigy 40 points

Hello Bin,

Do you mind to provide when will the dev team publish the fix? Or do you mind to provide the fix code of the workaround? Because of you provide no information about "SUSP_CTRL" in TRM, it is not possibile for us to implement the work around by ourselves.

BR,

Yuchen

0 Bin Liu over 1 year ago in reply to liu yuchen

TI__Guru**** 168501 points

Hi Yuchen,

Are you working on the same project with Tashiro-san? If not, please create a new E2E thread for your questions.

0 liu yuchen over 1 year ago in reply to Bin Liu

Prodigy 40 points

Hi Bin，

Im not working on the same project with Tashiro-san. And I post a new E2E thread as following：e2e.ti.com/.../am6422-need-support-for-usb-2-0-phy-errata-i2409

0 Bin Liu over 1 year ago in reply to liu yuchen

TI__Guru**** 168501 points

Thanks Yuchen. You question will be answered in the new thread.

+1 Bin Liu over 1 year ago in reply to Koichiro Tashiro

TI__Guru**** 168501 points

Hi Tashiro-san,

Koichiro Tashiro said:
If so, the workaround is needed ASAP.

The workaround is available now on https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-6.1.y-cicd with the following two kernel patches:

https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-6.1.y-cicd&id=6f5c812761b52b4af0df503bd7b8fcddf47cda8d

https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-6.1.y-cicd&id=b9e5d91ad98df1996983104cd2c8e535240ea02c

0 Koichiro Tashiro over 1 year ago in reply to Bin Liu

TI__Mastermind 22945 points

Hi Bin,

Thanks for patches.

The customer wants to confirm the workaround works properly, but they did not see the errata behavior before.
Is there any way to reproduce the issue easily?

Thanks and regards,
Koichiro Tashiro

+1 Mukul Bhatnagar over 1 year ago in reply to Koichiro Tashiro

TI__Guru* 83975 points

No , there is no easy way to reproduce this. The issue was first identified on AM62x and then based on design and IP vendor review applicability to other devices was confirmed. and workaround specific to AM62 and AM64 family were provided.

Customer should incorporate these patches w/o needing to reproduce the failure.

Processors

Processors forum

AM6412: Clarification needed for Errata i2409