Robert Soubie wrote :

> I have browsed the whole assembler file compild fo my "big"
> (8k) application, and found no trace of interrupt stuff
> (enabling or disabling) by the compiler. This is what I
> thought I would find, because that compiler has a special
> system for interrupts, that are delayed under the current
> basic instruction has been executed; in other words, basic
> routines cannot be iterrupted in the immediate sens of the
> term.

It's very possible that the Basic compiler has something called
"interrupts" that isn't the same thing as what a ASM programmer
would call an "interrupt".  That is, it's not unlikely that
the hardware (PIC) always has interrupts enabled, but the
Basic RT (Run Time) environment "hides" this from
the higher level code and also, as you point out, delays
the execution of the *Basic* ISR until the current Basic
"statement" has finished.

Just speculation, but seems as a reasonable way to build
an Basic runtime environment...

Now, the generated ASM code might be hard to read, but
one could just see what's going on around (or after) the
PIC interrupt vector at program address h'04'.

Regards
Jan-Erik.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.