I can’t get the Flasher ARM to program Spansion S29NS128P NOR.
If I manually set the NOR programming sequence by entering the commands in the CPU tab of the project setting window everything works as expected. When I let J-Flash ARM perform the task programming fails.
In looking at the output log I see that J-Flash uses 0xAAAA and 0x5554 as the address offsets in the command sequence. According to the Spansion documentation the address offsets should be 0xAAA and 0x554.
According to the Spansion S29NS128P data sheet Amax – A14 are don’t cares, so the 0xAAAA and 0x5554 offsets would not be the same as 0xAAA and 0x554. Many of the NOR flash parts are Amax – A11 are don’t cares, and for these parts the two address offsets would be equivalent.

So, is there some way to force J-Flash to only use 12bit addresses for the command sequence? Or, am I possibly missing something?
I have verified that for example the following sequence will work in the CPU tab if entered manually:

The following does not work:

Thanks,
Ossian

Hello Alex,

Thank you for posting the beta update V4.15d of the J-Flash application. Following your instructions I'm still not able to get programing to work. Using the "User target RAM" setting it is difficult to tell from the log out put what is happening. Any suggestions on how to make this more readable?
With "User target RAM" disabled it is easy to see from the log where things are going wrong, but as you indicated with the fix in V4.15d not using the "target RAM" option will not work.
Is it possible to get a version that works without enabling the "User target RAM" option?

I have attached the Spansion S29NS-P data sheet, hopefully this will help provide a better understanding of the issues. Note the notes on pages 79 and 81. Also by default DYB is set (protected) for every sector on startup, so all sectors must have DYB cleared. My plan was to perform this task for every sector through the J-Flash scripting, but it would be nice if this happened automatically.
The J-Flash application indicates that the Spansion S29WS128P is supported. Although I don't have the S29WS128P, looking through its data sheet it looks like the S29WS128P would have similar addressing issues as the S29NS128. I have tried manually forcing J-Flash to use the S29WS128P option, but this is not working and when I look at the output log it still looks like it is using 16-bit offsets?

Please let me know if any of the output logs from the J-Flash would be helpful in resolving this issue. Also let me know if you have any other versions of J-Flash that I can try. Is it possible to get a version of J-Flash with 12-bit offsets that does not require "User target RAM"? Obviously once everything is working I will want to switch to using the "User target RAM" option since everything programs much faster under this configuration.

Thanks for all you help,
Ossian

It looks like I can't post the Spansion S29NS-P data sheet because it is too large.
Let me know if the data sheet would be helpful and how best I can get this to you.

Thanks,
Ossian

Alex,

Thank you for the response. Are you saying that not using the "Use Target RAM" should still produce the correct programming algorithm?
Is there any way for me to verify from the log outputs that Segger's AMD programming algorithm is correctly switching to 12-bit offsets?

Thanks,
Ossian

Hello Alex,

I think we will go with another programming option.
Thank you for your time.

Ossian