Forum Discussion

NanjundaM's avatar
NanjundaM
Icon for New Contributor rankNew Contributor
7 years ago

Error flashing the QSPI NOR Flash connected to HPS with Command shell

Hi,

We are seeing few errors flashing the QSPI NOR Flash connected to HPS with Command shell as per the link: https://rocketboards.org/foswiki/Documentation/A10Gsrd161QspiBoot#Programming_QSPI

Below are the errors occurred:

Sector Erase Quad SPI flash at 0x060A0000

Program Quad SPI flash ...

Verify Quad SPI flash ...

Error: Fail to match data at flash address 0x000003FC with file address 0x000003FC. Found 0x34 but expect 0x00

Error: Quartus Prime Programmer was unsuccessful. 0 errors, 0 warnings

Error: Peak virtual memory: 4269 megabytes

Error: Processing ended: Sun Sep 09 03:29:56 2018

Error: Elapsed time: 04:30:47

Error: Total CPU time (on all processors): 00:46:45

We are using 1Gb QSPI with minimal file systems for arria10 in our case.

Has anyone come across these errors, kindly help us resolve this issue.

Thanks & Regards,

Nanjunda M

7 Replies

  • Daixiwen's avatar
    Daixiwen
    Icon for Occasional Contributor rankOccasional Contributor

    Is it on a development kit or a custom PCB? If custom, which flash are you using?

    On our custom PCB we have the same problem when using a Micron MT25QL512ABB8E12-0SIT but it works fine with a Cypress S25FL512SAGBHIC1.

    I didn't investigate that much because both the flash are working with our own software, that uses a lower frequency than the quartus_hps command line tool. It could be signal integrity problems but we were careful when routing the board, and adding 33 ohm resistors on the lines didn't change anything.

    I haven't found a way to reduce the clock frequency quartus_hps uses either, and if anyone knows I'd be interested.

  • NanjundaM's avatar
    NanjundaM
    Icon for New Contributor rankNew Contributor

    Hi Daixiwen,

    Its on the custom PCB. We are using the Micron's 1GB flash part(MT25QU01GBBB8E12-0SIT) with Arria10 SOC. We do have a series 33E resistors on board for the flash interface signals.

    If anyone can help us a way to understand & resolve this error, it would be grateful.

    Thanks & Regards,

    Nanjunda M

  • Daixiwen's avatar
    Daixiwen
    Icon for Occasional Contributor rankOccasional Contributor

    Then you could have the same problem than me. I don't know if quartus_hps has an option to reduce the SPI frequency, but if it does it is not documented so only Intel can answer that question.

    My current solution is to upload the preloader and our flashing software (based on Redboot) through JTAG first and then use our software to write to the flash.

    You could also try a Cypress/Spansion chip but it's not that easy to change a BGA chip on a PCB without specialized equipment.

  • MLamb4's avatar
    MLamb4
    Icon for New Contributor rankNew Contributor

    Were any of you able to find a solution to this?

    We are also developing a custom board and are having the exact same issue, at the exact same address 0x000003fc when verifying flash programming.

    Using a Micron - MT25QL02GCBB8E12-0SIT with a Cyclone V SoC 5CSEMA5F31I7N.

    Thanks

  • FawazJ_Altera's avatar
    FawazJ_Altera
    Icon for Frequent Contributor rankFrequent Contributor

    Hello,

    I did file a bug case to our internal engineering team. Please note that it might take some time to find a fix for this issue.

    Thanks

  • MLamb4's avatar
    MLamb4
    Icon for New Contributor rankNew Contributor

    Thank you.

    I forgot to mention that we have only USB Blaster 1 models and thus cannot alter the JTAG frequency.