MSPM0G3507: SWD Programming Problem

Hi,

I also wonder if the read exception problem is related to ECC?

If so, how do I use the ECC function? And is there any information for reference on the ECC programming process?

Thanks!

Hi,

It could related to the ECC function. Well, we have little information on the ECC programming. I think you can refer to the flashctl_program_with_ecc code example in the SDK first. 

Have you tried using Uniflash tool to erase and then read out the data? Does it also have the same issue as yours?

Best regards,

Cash Hao

Hi Hao,

After turning on ECC, the IC can indeed be programmed normally! ! Thanks! ! !

But I later discovered that the IC seemed to be able to only program the first 64KByte of data, and could not program data into the last 64KByte.

If I confirm correctly, this IC should be 128KByte.

Also, are there any sample code for erasing?

Since I didn't have Uniflash tool at hand to test the IC, I used GPIO to simulate SWD communication with the IC.

If possible, can you help me analyze the slightly exposed problems? Thank you.

Hi,

About the higher 64k Bytes programming issue. Have you tried to use the CCS v12.4? The latest CCS version should solved this issue.

Best regards,

Cash Hao

Hi Hao,

I have installed the latest version of CCS, but I can't seem to find the relevant programming process.

Could you please tell me where to put it?

Or you can provide me with relevant screenshot information. Thanks! !

Hi Hao,

Also I just encountered a problem while testing.

I wrote CMDWEPROTNM to all 0's and performed a Mass erase.

As a result, the IC seems to be no longer able to communicate through SWD.

Although the SWD ID can still be read, the IC's ID can no longer be read.

Is there any solution to this problem? Thanks.

Hi,

It looks like the NONMAIN got erased so that you could not connect to the MCU anymore. Some useful links that may help you.

https://software-dl.ti.com/msp430/esd/MSPM0-SDK/latest/docs/english/tools/ccs_ide_guide/doc_guide/doc_guide-srcs/ccs_ide_guide.html#dssm-commands

https://e2e.ti.com/support/microcontrollers/arm-based-microcontrollers-group/arm-based-microcontrollers/f/arm-based-microcontrollers-forum/1237262/mspm0l1306-implementation-of-dssm-factory-reset-over-swd

Best regards,

Cash Hao

Hi Hao,

Okay, thank you for the information you provided.

I will try to unlock the locked IC at hand, thank you!

But I currently want to solve the problem of only burning the first 64KByte.

Is there any relevant program code that I can refer to?

Hi,

Could you send out the exact detailed procedure how you program the MSPM0? I can ask our software team for help to check whether there is any issues.

Best regards,

Cash Hao

Hi Hao,

I found that in addition to ECC, the problems I encountered were that the flash read after erasing would not be all 0xFF and the last 64KByte could not be programmed.

These problems seem to be related to problems with the IC itself.

The left side of the picture is an IC borrowed from the client for testing, and the right side is an IC obtained from TI sales in the United States for another project.

The IC on the right can be programmed without any problems.

After erasing, the contents of the IC read are all 0xFF, and all 128KByte flash can be programmed normally.

The IC on the left will cause the problem I encountered...

Hi,

The left one mark with TI logo is the mass production version. And the right one is the pre-release sample. 

If the mass erase can not fully erase the higher 64kB region. Have you try to use sector erase for instead? 

Best regards,

Cash Hao

Hi Hao,

Yes, I have tried using sector erase, but it still cannot be completely erased.

In fact, I think this may not mean that it cannot be completely erased.

The state after erasing is that there will be more than 10 scattered bits of 0 in the entire flash.

But after erasing it should be all 1.

Blocks higher than 64Kbyte cannot be written to in any case.

These problems will occur in the mass production version I got, which is the IC on the left in the picture.

The IC on the right in the picture does not have these problems at all.

Hi,

I get your information and sent it with tools team about your issue. Let's wait for their comments on this one.

Best regards,

Cash Hao

Hi Hao,

Are there any updates?

Thanks!

Hi,

No feedbacks yet. Most of the team member are on vocation since the thank giving. it will take some time to get feedbacks from them. 

Best regards,

Cash Hao

Hi,

The software team feedback that they would like to check the code you are using for programming/erasing. Could you share it here?

Best regards,

Cash Hao

Hi Cho,

The program flow of programming and erasing is as follows.

Hi,

Get feedback from the team. 

"I see customer is using register level programming. Can you recommend the customer to start from the DriverLib example flashctl_program_with_ecc and simply modify the example to write to upper 64kB?

It might also be good to have the customer send a screenshot confirming their support package version."

You can find the support package version through CCS->Help->About Code Composer Studio->Installation details

Best regards,

Cash Hao 

Hi Cho,

I need to spend some time modifying and testing my program.

But I have seen the contents of flashctl_program_with_ecc before.

According to my understanding, the content of the program should not be much different, and the execution content is almost the same.

But he also said that it shouldn't be difficult to modify, it just takes a little time.

Hi,

Get it. Please let us know when you have updates.

Best regards,

Cash Hao

Hi Hao,

I have modified and tested the programming process according to DriverLib, but the result is still incomplete erasure.