CC3220MODA: Uniflash cannot connect

Hi,

Uniflash image creator uses UART not a JTAG/SWD as CCS or IAR. Please make sure that your UART connection and SOP mode is correct.

Jan

SOP = 010 and UART seems to working fine. Most of the time uniflash fails right from the start, sometimes it goes but fails at the end with "FS_DEVELOPMENT_BOARD_WRONG_MAC". What does that even mean?

Hi,

Please use Uniflash at simple mode as is described at this article. Production line guide at chapter described how to connect UART pin properly (RX, TX, GND, RST) at chapter 5.4.

At SOP mode (2-1-0) 0-1-0 determined between UART loader and execution mode according break signal at RX line. For your first try you can use SOP mode (2-1-0) 1-0-0. After SOP mode changing is device restart reduced.

If you want,you can provide high-res photos your connection between LaunchPad and your hardware. I can check your wire connection and jumpers configuration as well.

Jan

Jan, I provided high res photos on a different occasion see here: 

 The hardware setup has not changed.  Programming the board with Uniflash did work before. I mostly am in ccs/debugger so not run Uniflash all the time. Have found Uniflash to be fairly finicky in the past and now completely unusable as stated above. Can you let me know what that MAC (address ?) error really signifies?

Hi,

Error code code -10246 (=SL_ERROR_FS_DEVELOPMENT_BOARD_WRONG_MAC) means that you have programmed image in development mode with wrong MAC address. At development mode need to be created image with exact MAC address of your CC3220SF chip. If you not do this, you will see this error and image will not be executed.

You should check that you have set proper MAC address at Uniflash field "Original Mac Address". But maybe you should create a new image in simple mode.

Jan

For some reason I am at least connecting to my board again (halleylrya !). But I continue to get this Mac address issue. I've attempted to program the board with "use device Mac address" set as well as provided the Mac address as read by uniflash - makes no difference, fails either way.

I also created a new image in simple mode, just added the service pack and this did successfully made it to the board (yay!)    So what then is wrong/different you speculate with the case where I get a wrong MAC error?

Hi,

I am not sure what can be wrong at your case because I don't know what did you exactly. But I think you have changed MAC address accidently and this caused your issues.

Jan

I appreciate you can't look over my should to see what's going here, yet can you speculate a little what may have created an incorrect MAC address?  I can assure you I have not typed in something new 'just for the heck of it'.   Hopefully you can appreciate my position of perceiving this TI setup to be a bit flakey and non-deterministic?

Hi,

The main case in which we typically see incorrect MAC addresses being used by Uniflash is when there are multiple CC32xx devices connected at the same time - in this case it is possible that Uniflash will not detect the correct device you're trying to flash. It looks like you've correctly removed the JTAG jumper connecting the XDS110 to the launchpad CC32xx, so that's probably not the cause, but still something to keep in mind.

Other than that, the MAC address detection should work correctly - I have not run into cases where Uniflash fails to detect and flash my CC32xx devices in development mode. Does this behavior only occur when you're using the off-board debugger (launchpad XDS110 connected to a custom board), or does this also occur when you use the Launchpad standalone, connecting to the on-launchpad CC32xx?

How is the reset of the custom board handled? As part of the programming procedure the CC32xx is reset a couple times, and with the launchpad Uniflash will instruct the XDS110 to assert the CC32xx reset. If this reset signal is not sent to your custom board correctly then it might result in unreliable operation.

Regards,

Michael

Reset comes in through the jtag connector and is connected to the CC3220 /RESET as well as a 10K pulp and 100nF to ground.  Pretty standard stuff really.

I use the launchpad only to connect to my custom board - it seems once I get this problem I eventually get back to normal behavior though I do not know how or why. It is unpredictable.

Hi,

Is this problem present on all of your custom boards, or on only one device?

The other day I was running some tests on a set of CC3200 devices, and one of them exhibited somewhat similar behavior in the sense that flashing through Uniflash was unreliable. The root cause of that though was thought to be a rework issue - the CC3200 populated on that board was a test chip that was hand-soldered, and the other hand-soldered test devices did not encounter the issue. Maybe there is a hardware root cause to the issue you are experiencing. Have you double-checked your custom board for possible hardware issues?

Regards,

Michael