DS90UB936-Q1: About Lock status

Chris Liu

Hi,

Recently, I got some logs from some fielded devices.

The logs show that the lock status (0x4D[0]) was changed to 0 from 1 and changed to 1 in a very short time.

The register reading is triggered by GPIO3/INTB. Please see the following logs, the value in the brackets[] is in seconds:

2023-03-15T09:38:18.659823-05:00 kernel: [ 839.914369] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:38:18.663845-05:00 kernel: [ 839.917600] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:38:36.383821-05:00 kernel: [ 857.637384] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:38:36.387848-05:00 kernel: [ 857.640330] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:15.231825-05:00 kernel: [ 956.484223] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:15.236126-05:00 kernel: [ 956.487832] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:28.184029-05:00 kernel: [ 969.437112] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:28.187829-05:00 kernel: [ 969.440443] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:33.647832-05:00 kernel: [ 974.897651] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:33.647845-05:00 kernel: [ 974.901067] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:34.947852-05:00 kernel: [ 976.199564] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:34.951831-05:00 kernel: [ 976.203683] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:34.983827-05:00 kernel: [ 976.236640] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:34.987820-05:00 kernel: [ 976.240530] ds90ub954 2-0030: rx0: LOCKED
2023-03-15T09:40:35.195880-05:00 kernel: [ 976.446342] ds90ub954 2-0030: rx0: NOT LOCKED
2023-03-15T09:40:35.199835-05:00 kernel: [ 976.449555] ds90ub954 2-0030: rx0: LOCKED

We think it's because the quality of FPD-Link is not good, but we don't have any strategy to solve that.

The device worked well before this happened.

Could you help to give the possible reasons for the sympton?

Any suggestion is appreciated.

Thanks.

over 3 years ago

0 Glenn Uranga over 3 years ago

TI__Genius 16220 points

Hello,

Thank you for your inquiry.

Can you also collect the following registers 0x4D,0x4E,0x55,0x56,0x7A,0x7B? Also, have you tried completing MAP analysis to see if the link status?

FPD-Link Margin Analysis Program (MAP) user's guide (ti.com)

Some clarification questions:

Is this isolated to one or multiple systems?
Is this repeatable at every power up?
Have you tried different cable, Ser, and Des configuration?

Glenn

0 Chris Liu over 3 years ago in reply to Glenn Uranga

Prodigy 210 points

Hi Glenn,

Thanks for your quick reply.

Since the sympton can't be reproduced in lab, it only occurs on fielded devices, I'll request customer to do the MAP once the sympton occurs.

I'll also try to modify my code to collect the register info.

To answer your question:

1. It happens on multiple fielded devices.

2. not sure.

3. Yes, but it seems it doesn't help.

0 Glenn Uranga over 3 years ago in reply to Chris Liu

TI__Genius 16220 points

Hi Chris,

If this is hard to reproduce. Let's get a full register dump of all registers then. That way we don't have to keep going back and forth. Provide the register in a good state and failing state.

Glenn

Interface

Interface forum

DS90UB936-Q1: About Lock status