• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 99 subscribers
  • Views 420 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

PROCESSOR-SDK-AM437X: QSPI flash device data

Thomas Kuhnt
Intellectual 510 points

Part Number: PROCESSOR-SDK-AM437X

Hi experts,

I am referring to pdk_am437x_1_0_15, subfolder .\packages\ti\starterware\device.
Using qspi_flash.c for an alternate board with Micronix N25Q256A flash device,
I called the function QSPIFlashGetDeviceData(). I always get an error caused
by NULL pointer returning from that function.

While debugging I found that variable devIndex is incremented double
(once within FOR iteration, and in addition after comparing values).
This can't be correct - right?

Regards, Thomas

  • 0
    TI__Mastermind 30098 points

    Hi Thomas,

    I need to investigate this, I'll get back with you shortly.

    Regards,
    Frank

  • 0 in reply to Frank Livingston
    TI__Mastermind 30098 points

    Hi Thomas,

    Can you please confirm you're referring to the code below?

    qspiFlashDeviceData_t *QSPIFlashGetDeviceData(uint32_t devId)
{
    uint32_t numDev = sizeof(gQspiFlashDeviceData) / sizeof(qspiFlashDeviceData_t);
    uint32_t devIndex = 0U;
    qspiFlashDeviceData_t *flashDeviceData = NULL;

    for (devIndex = 0U; devIndex < numDev; devIndex++)
    {
        if (devId == gQspiFlashDeviceData[devIndex].devId)
        {
            /* Device Found */
            flashDeviceData = &gQspiFlashDeviceData[devIndex];
            break;
        }
        devIndex++;
    }
    return flashDeviceData;
}

    Yes, the increment of devIndex inside the for() loop seems like a bug. I'll take this up with the development team and get back with you concerning the response I get.

    Regards,
    Frank

  • 0 in reply to Frank Livingston
    Intellectual 510 points

    Hi Frank,

    Yes, it's the code part from your screenshot.

    The device identification works if devIndex++ is removed but programming the N25Q256A QSPI flash fails.

    It looks like some commands are different (e.g. FULL_ERASE, read MFG_ID) between MX66L512 and N25Q256A.

    And I am wondering if access to lock registers or unlocking sectors has to be added prior to programming!?

    Regards, Thomas

  • 0 in reply to Thomas Kuhnt
    TI__Mastermind 30098 points

    Hi Thomas,

    I filed a bug report for the extra index increment in the for() loop. Thanks much for pointing this out.

    I suppose access to lock registers & sectors depends on the device.

    Regards,
    Frank