On Thu, 3 Mar 2011, Erik Christiansen wrote:
A distrust of possible regressions in optimisation between compiler versions (and these have occurred in our toolchain) is one of my motivations for writing ISRs in assembler to start with. Being able to upgrade gcc to gain bugfixes or size optimisations, with zero risk of ISR timing change, leaves the developer with more time to build a good product, I think.
Agreed. If one needs cycle-accurate timing, write in assembler. I have some rather lengthy ISRs written in assembler for just that reason. If faster is better, C is often good enough, but not always. -- Michael [email protected] "Pessimist: The glass is half empty. Optimist: The glass is half full. Engineer: The glass is twice as big as it needs to be." _______________________________________________ AVR-chat mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/avr-chat
