• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 39 subscribers
  • Views 1151 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.

Running CC3200 from SFlash and NWP version

AQ
Expert 1580 points
Other Parts Discussed in Thread: CC3200, UNIFLASH

I have been having problems getting the CC3200 Launchpad board to run from SFlash for our customized application.  It works when I program a .bin file from the SDK example folders, but when I try to program the bin file compiled by CCS for our project (which was originally modified from an example application anyway), it doesn't start.


I've seen various threads on this forum related to the same problem, but none of them helped me fix this. 

1. I formatted the SFlash without the secure option

2. I tried with and without the service pack ( and I tried both 1.0.0.1.1 and 1.0.0.1.2 although I know there is no difference ).

3. I removed the SOP2 jumper when resetting.

4. I even changed my Windows language settings as one poster mentioned this fixed something for him.

Note: I am using a rev 3.2 Launchpad board, which has the XCC3200HZ pre-production silicon.

I wrote my steps and Uniflash log on this topic: http://e2e.ti.com/support/wireless_connectivity/f/968/p/355615/1403076#1403076

Then I started looking into the NWP version number and started to wonder if something was wrong with that.

According to the Servicepack_1.0.0.1.1_Release_Notes.pdf (and the 1.0.0.1.2 release notes show the same thing,) I should expect to see 2.2.7.1.31.1.2.4.2.1.5.3.23 for the firmware version. When I program the service pack I get the following messages in Uniflash:

[17:31:25] Begin ServicePackUpdate operation.
[17:31:25] INFO: > Executing Operation: Connect
[17:31:27] INFO: setting break signal
[17:31:27] INFO: --- please restart the device ---
[17:31:27] INFO: connection succeeded
[17:31:27] INFO: getting storage list
[17:31:27] INFO: > Executing Operation: Init
[17:31:27] INFO: reading version info
[17:31:27] INFO: DEVICE CC3200 ES1.32
[17:31:27] INFO: reading version info
[17:31:29] INFO: reading version info
[17:31:31] INFO: > Executing Operation: ServicePackUpdate
[17:31:31] INFO: Path to the service pack file: C:\TI\CC31xx_CC32xx_ServicePack_1.0.0.1.2\servicepack_1.0.0.1.2.bin
[17:31:31] INFO: reading version info
[17:31:31] INFO: Flash Device detected.
[17:31:31] INFO: Service pack version information:
[17:31:31] INFO: NWP version: 2.2.7.1
[17:31:31] INFO: MAC version: 1.2.4.2
[17:31:31] INFO: PHY version: 1.5.3.23
[17:31:31] INFO: NWP Patch version: 2.2.0.1
[17:31:31] INFO: MAC Patch version: 1.2.0.2
[17:31:31] INFO: PHY Patch version: 1.0.3.23
[17:31:31] INFO: downloading file [NWP_CODE]
[17:31:31] INFO: Erase storage FLASH
[17:31:31] INFO: erase storage succeeded
[17:31:31] INFO: erase storage completed
[17:31:55] INFO: Verifying Data...
[17:32:12] INFO:

Verification OK
[17:32:13] INFO: Downloading file "/sys/servicepack.ucf" with size 245288
[17:32:23] INFO:

New Token is 0x0
[17:32:23] INFO: Download complete
[17:32:23] INFO: > Executing Operation: Disconnect
[17:32:23] Operation ServicePackUpdate returned.

But when the program runs (which again only works when programming through CCS directly to RAM), this is the version number reported:

2.2.7.1.31.0.0.3.0.1.1.8.8

This appears to me that the NWP version was updated to the new service pack (2.2.7.1 for pre-production), but the MAC and PHY were not updated.  Am I understanding this incorrectly?  Could this be causing my application to not run when the SOP2 jumper is removed?  Is this expected behaviour or should I be doing something differently?

Thanks,

Angelo

  • 0
    TI__Guru 64405 points

    Hi,

    Your understanding is correct. NWP part seems to be flshed OK (NWP code is not residing on SFLASH) but MAC/PHY which resides on SFLASH is not.

    On the other hand, Uniflash log looks OK and you do say that other examples from SDK do seem to be programmed and executed.

    Maybe it is something on your application.  Can you take a look at the following E2E: http://e2e.ti.com/support/wireless_connectivity/f/968/p/386460/1372676#1372676 ?

    Shlomi

  • 0 in reply to Shlomi Itzhak
    Expert 1580 points

    Thank you Shlomi, that was it.  There was a Report() call before InitTerm().  I actually noticed this earlier but never imagined it would cause the application to stop running completely, (and in SFlash mode only,) so didn't try removing it.  (I also don't know I missed that post after searching this forum for so long.)

    Regarding the service pack drivers, you said the NWP does not reside in SFlash.  So does that mean these patches are applied directly to the NWP firmware in ROM (which I suppose means it's actually Flash and not ROM)?

    Thanks again,

    Angelo

  • 0 in reply to AQ
    TI__Guru 64405 points

    Good to know this was the issue.

    Regarding NWP, this is true only for the pre-production devices like you have (I can see ES1.32 from Uniflash log). In this case it is stored on an internal non volatile memory and not the SFLASH. For Rom devices (ES1.33), all patches are on SFLASH.

    Regards,

    Shlomi