Mity CPU Platforms » MityDSP-L138 (ARM9 Based Platforms)

Jacek,

Sorry to hear that you're having an issue with one of the modules.

Can you please provide the serial number of the module which is located on a label on the back/bottom side of the module and is 8 numeric characters in length?

The "BUSY" LED is tied to the FPGA DOUT_BUSY signal which during initial boot into UBoot should not be lit under normal operation. Even when programmed the BUSY LED is typically not utilized unless programmed to do so in the FPGA.

What carrier board is this module being used in, custom or a MityDSP-L138 Industrial IO Dev Board? Additionally any history about the modules prior operation/history would be helpful.

Thank you,

Alex

The "BUSY" LED is tied to the FPGA DOUT_BUSY signal which during initial boot into UBoot should not be lit under normal operation. Even when programmed the BUSY LED is typically not utilized unless programmed to do so in the FPGA.

And what about there is no correct uboot image in L138? Because I wonder if I can't flash L138 because BUSY is ON or I can't flash L138 despite the fact that BUSY is ON

Additionally any history about the modules prior operation/history would be helpful

We had to replace OMAPL138*B* to L138*E*. We were sure that they compatible, but what is your opinion? Is the problem connected with poor soldering or incopatibility (or something else)

Is the "E" chip the secure boot option? That would explain why it was not booting up.

Unfortunately we need more clarification on the part number of the processor that was installed. The "B" would presumably be in reference to the revision field (Rev 2.0 or 2.1) and an "E" for that field would be Rev 2.3. However the "Secure Boot" enabled processor is denoted by an "E" at the end of processor part number. Please provide the part number of the processor installed (i.e. OMAPL138EZWTD4E).

In either case it appears that the FPGA became damaged during the removal and/or re-installation process of the OMAP_L138 processor as again we do not expect that LED to be lit under normal circumstances unless programmed to do so.

If the OMAP-L138 processor that was originally installed on our module has been removed and replaced with the "Secure Boot" enabled version we would not expect you to be able to boot the module unless you have developed a compatible secure boot enabled bootloader.

At this time Critical Link does not offer support for the Secure Boot enabled processors.

Alex

I have got OMAPL138EZWT64ALVIW GI D450 527 ZWT

Thank you for confirming the part markings on the OMAP-L138 processor. Best I can tell mean this is likely not a secure boot enabled processor. Unfortunately I could not find a document from TI that covers the package markings enough to confirm how/where the marking to show Secure Boot would be shown. To confirm if this were a secure boot enabled processor I would need the part number that was ordered.

Taking a step back have you confirmed that even the "BOOTME" message is not shown by the OMAP-L138 boot-rom?
1) Connect your PC serial port to the Industrial IO Dev Board (115200 8-n-1) A NULL modem adapter typically is needed, be sure to confirm this with the working board you have to ensure the setup is correct.
2) With the problem module installed hold the "BOOT" button the dev board
3) Apply power to the dev board
4) You shout see a single "BOOTME" printed on the serial port

However again based on the operation of the BUSY led and inability to boot it sounds like the module has become damaged during the removal and re-installation of the processor. At this time we would recommend you purchase a new module from the same distributor used to obtain the original. Can you share with me what you were trying to accomplish by replacing the processor, if not for secure boot?

Thank you,

Alex

This processor is without secure-boot. Secure-boot has E following ZWT

There is no BOOTME printed. Other modules print BOOTME but not this.

This module had broken MCBSP pin. Other modules was working correctly but not this. (and under power-off this pin (DOUT) was shorted to ground by 150ohm). So the processor was replaced by equvalent in production (newer silicon mask) https://www.digikey.com/product-detail/en/texas-instruments/OMAPL138EZWTD4/296-42793-ND/5019337

Jacek,

Thank you for confirming which processor was used in the replacement. Based on the lack of the "BOOTME" from the processor it certainly sounds like the module is unrecoverable.

In the future please note that if we are notified about a suspect module failure you could have worked with the supplier where the module/dev kit was purchased from to get the unit replaced under our standard 1 year warranty. However replacing the processor yourself voids any warranty that Critical Link would have provided.

Our recommendation is to obtain another module from one of our distribution partners and to "scrap" this module.

Please let me know if you require any assistance in purchasing a replacement.

Thank you,

Alex