On Fri, Apr 17, 2009 at 1:43 PM, Gabriel Michael Black < gbl...@eecs.umich.edu> wrote:
> > You should also be sure to test starting from a checkpoint which I > believe is different from a switchover, although I'm not sure. It shouldn't be different from O3's perspective; we always restore from a checkpoint into a SimpleCPU model and then switch over to O3, since you can't directly restore a SimpleCPU checkpoint into and O3 model. > I've > done things in the past that seem fine and pass regressions but break > that mechanism. Speaking of which, since I and potentially others > don't know off hand how to use checkpointing (still!), we should > probably make a regression to test those. I know we've agreed on that > before, but this is another case where it would be good to have so I > thought I'd bring it up. Agreed, absolutely. Steve
_______________________________________________ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev