nathan binkert wrote: > > If you add a function for it I'll use it. We should be careful, though, > since when nanoseconds (or microseconds or milliseconds) spill into seconds > might get a little tricky. Actually, right now nanoseconds, microseconds, and > milliseconds all change the same data, but seconds are separate. That might > not be obvious to somebody not in the know. > > > I was more thinking of accepting Ticks directly, instead of taking > nanoseconds (i.e. dealing with Int::ns in the function). We could > make a version of set() that just takes Ticks. set() is separate from > sec() and nsec(). Though perhaps that is confusing. > > What if, instead of doing it in base/time.hh, we did in sim/core.hh? > We could just have some stand alone function that takes Ticks and > turns it into Time by properly using Int::ns. > > The alternative is to use Float::s and use the version of set() that > takes a double. I think it would simply be set(p->time_sync_period * > SimClock::Float::s). That feels a bit kludgy though. > > Nate > ------------------------------------------------------------------------ > > _______________________________________________ > m5-dev mailing list > m5-dev@m5sim.org > http://m5sim.org/mailman/listinfo/m5-dev >
Yeah, I tried to stay away from using double since it can be imprecise. To me it makes more sense to put a conversion function in base/time.hh since Ticks seem like the more fundamental thing. Maybe it's as simple as adding a setTick? Gabe _______________________________________________ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
