> sim_insts is the total across all cores, so we do need a per-core total. sim_insts is there so we can understand the performance of the simulator itself. It could certainly be a formula (if it's not already)
> Why is there an "SMT side of things?" I don't think we need two sets of > stats (one for SMT and one not), we just need per-thread stats and a > per-core total. So I can see having two committed inst counts for > system.cpu (one that's clearly labeled as thread 0's and one that's the core > total), but not four. Vector stats were designed for this sort of thing. > If we want to get fancy and suppress printing the per-thread stats when > there's only one thread, I'm OK with that, though I don't think it's > necessary, and sometimes it's nice to be consistent as you go from 1 to N > threads. I'm interested in what people thing on that. Vector stats can probably already deal with this suppression. > I would really love it if O3 could use the same names as SimpleCPU for the > stats that correspond. I also don't like the all-caps bits like "COM:"... > those are from waaay back in SimpleScalar days, and they're totally > redundant wrt the "cpu.commit." prefix. Renaming stats is a pretty big > change that could break people's analysis scripts so it's not something we > want to do lightly, but if we're going to rename them at all, I'd say we > should try and go all the way to where we really want to be in one shot. I'd go further and say that we should move some stats into the CPU base class. At least the declaration and regStats bits. Nate _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
