Quoting nathan binkert <[email protected]>: > I have a couple of comments on this changeset. First, I think this > has the potential to affect alpha, so you should run regressions. > (MC146818 is used by alpha for the clock).
That's right, and I did that before I committed. I was a little surprised, but unless I did something wrong this had no impact at all on the ALPHA_FS regressions. I'll comment on the rest of this discussion once I've had a chance to read through it and think about it. My immediate thought is that while a simplified manual roll over sort of mechanism would avoid using the library functions and their potential system dependence, even something simple like keeping track of when to roll the month over would be not hard but non-trivial. Then there are all the weird, fancy leap seconds and spring forwards and fall backwards, etc., that are possibly not very important but would be missing. I was trying to avoid having to figure out (and potentially screwing up) those sorts of considerations when I went with the library functions. Gabe _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
