My bad. LDI takes 1 cycle, and it's not even here. The reason is that OUT instruction enables interrupts here, and interrupt can occure before the second STS instruction.
2012/11/19 Anton Voloshin <zombi...@gmail.com> > The code was compiled on windows with -Os optimization for Atmega > 1284P. The "ldi" instruction takes 2 cycles, according to disassembly > listing. > > wdt_enable() didn't work for me until I made this correction. > > > 2012/11/19 Joerg Wunsch <invalid.nore...@gnu.org> > >> Follow-up Comment #1, bug #37772 (project avr-libc): >> >> Well, this inline assembly statement has been written that >> way for years now, and was proven to work. How did you >> verify it doesn't work, and for which controller type? >> >> The OUT instruction takes one CPU cycle, and the STS >> instruction two cycles, that's why the four-cycle requirement >> is still maintained. >> >> _______________________________________________________ >> >> Reply to this item at: >> >> <http://savannah.nongnu.org/bugs/?37772> >> >> _______________________________________________ >> Message sent via/by Savannah >> http://savannah.nongnu.org/ >> >> > > > -- > Anton Voloshin > Jabber: zombi...@bitcheese.net > -- Anton Voloshin Jabber: zombi...@bitcheese.net _______________________________________________ AVR-libc-dev mailing list AVR-libc-dev@nongnu.org https://lists.nongnu.org/mailman/listinfo/avr-libc-dev
