This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

uVision and scratch pad flash area

The uVision3 IDE erases the scratchpad flash when I re-program a Silabs C8051F120.

I want to use the scratch pad area for serial numbers and other similar data. Is there a way to stop the scratchpad flash being erase on a code upload when using the uVision IDE for debugging?

I previously asked this question here:
www.cygnal.org/.../000206.html

There has been a suggestions to use v:0x0000 to access the scratch pad memory - but tis did not work for me (I'm using uVision v3.33).

And a suggestion to Write and use an AGSI-DLL - which is well beyond what I'm able, or want to do.

Do you have suggestions as to how I load and debug using uVision3 without loosing my scratch pad flash content?

Regards,
Steven

Parents
  • I do not 'know' about scratchpad erase, I do not use the scratchpad.
    However I do, in one instance, write parametres to the 64-128k flash in the f120 and THAT does not get wiped when downloading code (using SILabs).

    Every 'problem' posted from those that program/debug SILabs from Keil has been resolved by using Keil (which is the best) for compiling/linking and SILabs (which is the best) for loading and debugging.
    It is not realistic to expect Keil to immediately react to every little change SILabs make and equally unrealistic to expect SILabs to immediately react to every little change Keil makes.

    Some will find it admirably that the two try to integrate, but expecting such to be seemless in all respects is not realistic.

    Erik

Reply
  • I do not 'know' about scratchpad erase, I do not use the scratchpad.
    However I do, in one instance, write parametres to the 64-128k flash in the f120 and THAT does not get wiped when downloading code (using SILabs).

    Every 'problem' posted from those that program/debug SILabs from Keil has been resolved by using Keil (which is the best) for compiling/linking and SILabs (which is the best) for loading and debugging.
    It is not realistic to expect Keil to immediately react to every little change SILabs make and equally unrealistic to expect SILabs to immediately react to every little change Keil makes.

    Some will find it admirably that the two try to integrate, but expecting such to be seemless in all respects is not realistic.

    Erik

Children
No data