Hello,
On Mon, May 11, 2009 at 10:56 PM, Raphael Neider <rnei...@web.de> wrote:
> Dear Néstor,
>
>> your answer is very appreciated. I have refactored the code as you
>> told, I put all the global variables in a header file, called
>> "defs.h". I have also taken out all the calls to functions from the
>> interrupt. Now, interrupt handler routine is a lot simpler. Just one
>> thing: I haven't been able to disable the interrupt and save the stack
>> with the code you gave me. Apparently, it should work, since I'm using
>> a PIC 16F877, which seems to have 16 stack bytes. Am I missing
>> something from your explanation?
>
>
> What do you mean with you "haven't been able to disable interrupts and
> save the stack" --- could you do either but not both, or none? How did
> you find out that you did not succeed in performing the task?!?
I'm able to disable interrupts with GIE=0, but when I try to save the
stack, the interrupt just hangs. Look at the commented out code.
>
> Do you get compile time errors?
No, no. It compiles smoothly.
> Run-time errors?
Yes, it hangs when I enable the timer to trigger the interrupt.
> What precisely are the symptoms/error indications?
This application draws a countdown. Well, this countdown actually
stops when I enable the timer to trigger the interrupt which *should*
refresh the countdown onscreen. Look at the commented out code, which
is causing this behaviour. I'm maybe doing something wrong.
Thanks.
>
>
>
> To make stack saving a bit more efficient, this afternoon I tried
> (something like)
>
> static void
> Intr(void) __interrupt(0)
> {
> char int_status;
> char int_stack[16];
> __data char *src;
> __data char *dst;
> unsigned char u;
>
> // disable interrupts (same as the macro)
> int_status = INTCON & 0xC0;
> INTCON &= 0x3F; // GIE = PIE = 0;
>
> // save pseudo stack
> src = (__data char *)0x70; // &STK00
> dst = &int_stack[0];
> for (u = 0; u < 16; u++)
> {
> *dst++ = *src++;
> }
>
> // handle timer interrupt
> if (T0IF)
> {
> T0IF = 0; // re-arm for next interrupt
> // --- more handler code ---
> }
>
> end:
> // restore stack
> src = &int_stack[0];
> dst = (__data char *)0x70; // &STK00
> for (u = 0; u < 16; u++)
> {
> *dst++ = *src++;
> }
>
> // restore/re-enable interrupts
> INTCON |= int_status; // restore GIE and PIE
> }
>
> ... and the .asm looks good (did not test on real hardware, though).
>
> No meddling with GIE/PIE or T0IF/T0IE should be required in the
> interrupt handler iff INTCON is masked and restored as shown.
> Of course, you will have to enable GIE/PIE and T0IE at some point
> in your program to enter the interrupt handler, but this seems to
> be done in the menu handling code (if I recall correctly).
>
> Rather puzzled,
> Raphael
>
> ------------------------------------------------------------------------------
> The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
> production scanning environment may not be a perfect world - but thanks to
> Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
> Series Scanner you'll get full speed at 300 dpi even with all image
> processing features enabled. http://p.sf.net/sfu/kodak-com
> _______________________________________________
> Sdcc-user mailing list
> Sdcc-user@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/sdcc-user
>
--
Ricardo L. Febrero
------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image
processing features enabled. http://p.sf.net/sfu/kodak-com
_______________________________________________
Sdcc-user mailing list
Sdcc-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sdcc-user
