DS560DF810: DS560DF810 Bring up

Anthony Ting

Intellectual 2381 points

Part Number: DS560DF810

We are repling your question in e2e here, let's keep discuss following question, thanks.

The DS560DF810 final version (RTM) is PG1.2 right ? -->Yes, RTM version is PG1.2 right.
Could your release all related final documents of DS560DF810 to us, include Latte tool and Lib file of PG1.2 ? --> please find latest version of documnet, Latte, and C API on mysecure. Let us know if you have any question about these.
Attached Hex file is generated by EEPROM_ConfigUpdatedSingle_PG12 to build FW file for EEporm mode , could you help to review ?
The 0~7 TX channels setting( enable/disable, preset….) can not adjust independent in PG1.2 excel config file , is it correct ?

Regarding questions 3-4, I will investigate and get back to you on these early next week.

over 1 year ago

0 Wang Kevin over 1 year ago

Prodigy 90 points

Hi Sir,

I could not login my previous ti account (Kevin26_Wang@asrockrack.com),I do not know why.

BTW, I have register again the same account, please help to release again Doc to my account.

thanks.

0 Michael Lu (Santa Clara) over 1 year ago in reply to Wang Kevin

TI__Mastermind 29655 points

Hi Kevin,

I noticed a few duplicate entries in mySecure, so I removed all prior ones under your e-mail and granted new access to the documentation again under your e-mail address. Please let us know if you still have trouble accessing (please wait at least 30min from the time this message is sent).

Michael

0 Wang Kevin over 1 year ago in reply to Michael Lu (Santa Clara)

Prodigy 90 points

Hi Michael,

Why all documents of mySecure were disappear in my account ? (ID = Kevin26_Wang@asrockrack.com )

The documents were still there on last week (11/3 ).

The original files inside are as below.

0 Drew Miller1 over 1 year ago in reply to Wang Kevin

TI__Mastermind 36143 points

Hi Kevin,

Apologies for the mysecure issues. I think I re-granted mysecure access. Please let me know if you are still having issues.

Regarding questions 3-4, I'm still investigating these and plan to have an update in the next couple days.

Thanks,

Drew

0 Drew Miller1 over 1 year ago

TI__Mastermind 36143 points

Hi Kevin,

Apologies for the delay.

Anthony Ting said:
Attached Hex file is generated by EEPROM_ConfigUpdatedSingle_PG12 to build FW file for EEporm mode , could you help to review ?

The 0~7 TX channels setting( enable/disable, preset….) can not adjust independent in PG1.2 excel config file , is it correct ?

3) I don't see any issues with the EEPROM excel file you shared.

4) Using the excel config file, there is not a way to independently change TX FFE settings. However, since the EEPROM file just contains device level macros, it would be possible to generate an EEPROM that independently configures TX FFE for each channel.

Is your use case for the retimer just a single data rate in straight mode?

Thanks,

Drew

0 Wang Kevin over 1 year ago in reply to Drew Miller1

Prodigy 90 points

Hi Miller,

However, since the EEPROM file just contains device level macros, it would be possible to generate an EEPROM that independently configures TX FFE for each channel.

1. How to config it ?

Is your use case for the retimer just a single data rate in straight mode?

2. Yes, we only use 53.125 Gb/s data rate.

Could you help review our layout ?

thanks.

0 Drew Miller1 over 1 year ago in reply to Wang Kevin

TI__Mastermind 36143 points

Hi Kevin,

Yes, I can help review layout. You can share it with me over E2E private message.

Thanks,

Drew

0 Wang Kevin over 1 year ago in reply to Drew Miller1

Prodigy 90 points

Hi Miller,

The location U13 is DS560DF810 in the brd file. please help to review.

thanks.

0 Wang Kevin over 1 year ago in reply to Wang Kevin

Prodigy 90 points

add layout brd file again

asr2_n19_ipb2v4s56_r100__review.brd

0 David Waier over 1 year ago in reply to Wang Kevin

TI__Mastermind 20901 points

Hi Kevin,

Please allow us to get back to you shortly while Drew is out.

Best,
David

0 Lucas Wolter over 1 year ago in reply to David Waier

TI__Mastermind 27770 points

Hi Kevin,

Please allow us some extra time to review the layout due to US holidays this week. We will aim to have feedback by EOD 12/1.

Best,

Lucas

0 Wang Kevin over 1 year ago in reply to Lucas Wolter

Prodigy 90 points

Hi Sir,

Because we need gerber out on this week, could you help to provide layout review result today ?

Attached is our final brd file currently.

asr2_n19_ipb2v4s56_r100_1129.1508.brd

Thanks.

0 Drew Miller1 over 1 year ago in reply to Wang Kevin

TI__Mastermind 36143 points

Hi Kevin,

Apologies for the delay, I reviewed your layout and found a couple areas for improvement.

ASRock Layout Review.pptx

Also, I apologize for the late notice on this as I understand you are in your layout phase, but we recently identified some bugs with EEPROM configuration for our retimer that we don't yet have a resolution for. Do you have any I2C controller on your system? I'm wondering if it might be possible to configure the device through an I2C controller instead of EEPROM.

Thanks,

Drew

0 Wang Kevin over 1 year ago in reply to Drew Miller1

Prodigy 90 points

Hi Drew,

Thanks for your support. We will follow your recommend to modify our layout.

Yes, we have I2C controller on our system but usinig EEprom configuration is more easy for our application.

What are bugs with EEPROM configuration ? Maybe these bugs are fine for our application.

thanks.

0 Drew Miller1 over 1 year ago in reply to Wang Kevin

TI__Mastermind 36143 points

Hi Kevin,

There are some fixes we made related to retimer behavior since device release, even in the analog low latency straight mode you're planning on using. When the retimer is used with an I2C controller, the controller loads a patch file onto the retimer on each power up. This patch file is responsible for any improvements to retimer behavior beyond what is "frozen" in silicon. TI has a C API to help with loading the patch file and device configuration.

We currently don't have a solution for loading this patch file onto the retimer via EEPROM. We are looking into how we will handle this, but currently there is not a plan to address this gap.

At this point, I'd recommend connecting an I2C controller to the retimer. Apologies that this is more complicated, but I think it will help de-risk your design.

Thanks,

Drew

Interface

Interface forum

DS560DF810: DS560DF810 Bring up