flash memory

[samo]

Hi,

I am using the flash memory for a few variables and it works very well. Now I have one flag which is defined by #define statement and if it is enable then the initial flash space together with initial values of variables are defined. If flag is not defined then the program knows that it can read values of variables from flash and they were defined in past. This procedure works very good but the problem is that in first step I have to load the program with #define flag statement and subsequently the same program without #define flag statement. Of course it is really time consuming procedure because of that I would like to know if it is possible to do in one step that I can load only one program on nrf chip.

Best regards

Samo

[Roger Clark]

Converted my comment to an answer (see below)

[Roger Clark]

This sounds like programming 101 and not really a specific Nordic device question.

Why not just erase all the flash. Then in your program check if a magic number is in a specific location in flash.

If the Flash location is blank (0xFF ) the magic number is not present, so your software must copy the data (and the magic number) to the flash.

Then the next time you run the program, it checks the Flash it will find the magic number, so you can read the data from flash as you know it has been initialised

[samo]

Thanks for help Roger:)

[samo]

At this point I have one question as well. It is possible to add some instruction in "Options of Target" that when the program is loaded on chip the address where my magic number is located will be erased automatically.

[Roger Clark]

Do you mean something like a compiler define

I'm not sure what IDE or toolchain you are using but with gcc arm you can use -DOPTIONS_FOR_TARGET SOME_VALUE

e.g. its equivalent to doing

#define OPTIONS_FOR_TARGET SOME_VALUE

in your code