TMDSIDK574: TMDSIDK574

Hi Anders,

The prints you are seeing are from the MLO i.e SPL.
For some reason mmc init has failed. I will try to get more information on the rev B at my end and get back to you.


Regards,
Keerthy

Hi Anders,

Could you help me with the exact SDK version link?

Thanks,
Keerthy

Hi!

I have been trying to boot with both the U-Boot/MLO that was present on the microSD card that came with the IDK (Linux SDK SPRW271) and with one that I have built using the downloaded SDK (am57xx-evm-06.02.00.81) with which I built U-Boot SRCREV update to 01.00.00.15A 

(Git hash: 28a2f128fbadd37463060eb5ee40c7c9a47cc530)

I saw the same problems with both of these. They work fine on the IDK rev 1.0A of the board though. Is it correct then that the processor ROM code manages to initialze the mmc and read the MLO from card, load it in RAM and start it. But the MLO fails to init (re-initialize?) the mmc to read the U-Boot proper from the card? 

The documentation says that the board is configured to have the CPU first try to boot from the MMC and then from SPI memory. Is it possible to prepare and use a SPI memory device to get the board to boot from that instead? Would it be through port J45 or J46? 

Regards, 
  /Anders 

Hi Anders,

Thanks for the details. 

"Is it correct then that the processor ROM code manages to initialze the mmc and read the MLO from card, load it in RAM and start it. But the MLO fails to init (re-initialize?) the mmc to read the U-Boot proper from the card? "

To be precise the ROM loads MLO from card to internal SRAM.

MLO fails to init that is right.

Since the SD boot is not working even to flash the images to spi we need SD card working. I am trying
to reproduce the issue at my end. I will get back to you as soon as i reproduce the issue at my end.
I will try to figure out if there is any other to flash the images on to SPI.

Thanks,
Keerthy

Anders,

One of our engineer had seen the same issue on another Rev 1.0B sample and
has shared the boot loader images that potentially fixes the issue.

I am sharing the u-boot.img & MLO in a zipped folder. Could you please try them
and let me know if you are able to boot to u-boot prompt with them?1134.u-boot.zip

Thanks,
Keerthy

Hi!

Thanks for the new boot software which I have tried on our rev. B card. It does not improve things, but our IDK was in a different state now in which it had stopped endlessly trying to load u-boot so it is possible that it might have solved the problem if the IDK was still in the same state. While it was endlessly trying to load u-boot I noticed that sometimes it seemed to go a little further so maybe after many attempts it managed to load some corrupt firmware and execute it. After this it has stopped printing anything on the console after reset or power on, but it does light up 2 of the Ethernet interfaces, J5 and J8 instead of all 4 as on our Rev. A card, and a blue and a green LED.

We have been able to access the card using JTAG and the debugger software and seen that the cores are alive so it is not completely gone. Perhaps some persistent setting has been altered in the boot process by corrupt code.  

I'm not sure our card is Rev. 1.0B because there's no Rev. sticker on it but the silkscreen print says Rev B. 

I can confirm that the updated MLO and u-boot image work well on the Rev 1.0A which is now the IDK we have to work with.

Best Regards, 

  /Anders 

Hi Anders,

Thanks for trying them. So do you want me to try anything from my end.
Or we should close this issue?

- Keerthy

Hi!

Thanks for your help but we feel that this board is not working normally (certainly not compared to Rev. 1.0A), so we will likely have to make a RMR request.

Best Regards, 

  /Anders