AVR stack problem

for AVRstudio did you select the divice?

Yes. The correct device appears to be selected.

I am using avrstudio 3.53 and 4.0x. If you have your code set up correctly you should see the Stackpointer being set. should be one of the 1st instructions.

It looks like everything is set up correctly. The grayed out fields on in AVRStudio might be that limits are disabled. In any case, when I step through my code everything seems to work, the stack pointer changes correctly but on a return from subroutine I head off into the woods.

I am not using external SRAM and it almost looks like the code is trying to use it.

Any help would be greatly appreciated.

skrev i meddelandet news: snipped-for-privacy@z14g2000cwz.googlegroups.com...

If you use a link file which assumes external RAM the stack might be situated in external RAM.

I am using this asm code to set up the stackpointer. ramend is at 0FFF'h for atmega 103 and 128 reset: ; ************* Stack Pointer Setup Code ldi wl,high(ramend-5) ; Stack Pointer Setup out SPH,wl ; Stack Pointer High Byte ldi wl,low(ramend-5) ; Stack Pointer Setup out SPL,wl ; Stack Pointer Low Byte

this should work to init the SP

I figured out that my stack pointer is being initializaed outside of my internal SRAM space. That should fix it. Thanks.

Hi,

Have a look at:

It appears that your fuse bits are still in compatibility mode i.e. pretending that you are using an Atmega 103.

...Well, certain odd problems arise out of the situation that the AVR devices as shipped by Atmel often come with a default fuse bit configuration that doesn't match the user's expectations. Here is a list of things to care for:

.... . The ATmega128 ships with the fuse enabled that turns this device into ATmega103 compatibility mode. This means that some ports are not fully usable, and in particular that the internal SRAM is located at lower addresses. Since by default, the stack is located at the top of internal SRAM, a program compiled for an ATmega128 running on such a device will immediately crash upon the first function call (or rather, upon the first function return). .... /Roland