This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Missing C6748 boot ROM?

Hi

We have just received the first prototypes for a new product that we are developing fitted with a C6748 DSP.  So far the device seems to power up successfully and we can connect to the device using our emulator.  However we have noticed that there does not appear to be any boot ROM code; starting at address 0x00700000, there appears to be a single NOP followed by a bunch of IDLE instructions.  We were not expecting this! 

Have we managed to source parts without boot code or is there a simple explanation (eg, something we have to set up in our gel file) as to why Is there appears to be no boot code in our devices?  The latest Silicon errata does not seem to suggest that there are parts in existence without any boot code. 

Device markings are as below:

TMX320

C6748ZWT

$7-9CAGI4W

GI

Thanks in advance

Tim

  • Hi Tim

    Can you re-confirm that when you follow the steps listed in the boot loader documentation, you are not able to read out a valid romid value

     

    The C6748/C6746/C6742 bootloader has undergone multiple revisions. To check the version of your
    device, perform the following steps.
    1. Connect to the device in the Code Composer Studio™ software.
    2. Select View → Memory.
    3. Enter the address of the beginning of the ROM, 0x11700000, at the top of the memory window.
    4. Select Character mode at the bottom.

     

    The text d800k006 should appear in the memory window at offset 0x08. For earlier ROM revisions, the text could also appear as d800k002 or d800k004.

    If you are not able to read out the correct value, can you you confirm the following on your protype board design

    1) Confirm that you are hooking up the SATA_VDD pins ( M2, P1, P2, N4) to a 1.2V supply

    2) Confirm that the RVDD pins (E5, H14, N7) are hooked up properly to a 1.2V supply

    3) Confirm that USB_CVDD (M12) is hooked up properly to a 1.2V supply.

    Note 1 & 3 are required for PG1.x silicon, and can be left a NC on PG2.x silicon. 2 is required always, have seen customers miss hooking this up correctly.

    From your case marking it would look like you have PG1.0 silicon

    Details on case marking in the errata:  http://focus.ti.com/lit/er/sprz303b/sprz303b.pdf

    Regards

    Mukul

  • Hi Mukul

    Thank you for your prompt response.  I have tried following the steps at the start of your email can can confirm that I still don't see anything.  However I suspected that this might be the case as soon as I read the end of your email.

    We do have RVDD and USB_CVDD hooked up to 1.2V (as per the datasheet).  However, given that ultimately we wish to target the C6746 we did NOT hook up the SATA_VDD pins (the C6746 datasheet shows these as NC and I do not recall reading anything in the C6748 documentation to suggest that this was necessary/advisable).

    I take it from your final comment that there is no work around to this (other than to source and refit PG2 silicon)?

    Thanks

    Tim

  • Hi Tim

    Tim James said:
    I take it from your final comment that there is no work around to this (other than to source and refit PG2 silicon)?

    Unfortunately that is true. If you have no provision to blue wire the pins M2, P1, P2, N4,  to a 1.2V supply then there is no work around for PG1.x silicon. The issue is fixed on PG2 silicon and I would expect your prototypes to work with PG2.0 silicon.

    Regret the inconvenience, we need to do a better job highlighting this in the c6746 collateral to highlight PG1.x vs PG2 differences for this issue.

    Regards

    Mukul