Hi guys, I think this is a known issue with Ruby checkpointing right now. Ruby checkpointing may not be able to restore the event queue back to its state prior to taking the checkpoint, because it uses a complicated process to flush caches and create a cache trace during the checkpointing process. As a result, trying to continue execution past the checkpoint may not work in the original simulation (There's a warning about this in recent gem5 versions: "warn: Ruby memory writeback is experimental. Continuing simulation afterwards may not always work as intended.").
HOWEVER: based on the output that Prateek copied into his original email, it looks like the checkpoint writing process finished correctly (indicated by the simulation trying to reenter the event queue after "Writing checkpoint"). It should be possible to restore the simulation from that checkpoint, so I'd recommend testing the checkpoint and see what happens. Joel On Wed, Jun 15, 2016 at 8:42 AM, Anirudh <[email protected]> wrote: > Anouk <a.laer <at> ee.ucl.ac.uk> writes: > > > > > Hi Prateek, > > > > I have come across the exact same problem as you and tried some earlier > > versions of the repository but the problem persists. > > Did you perhaps ever find out what was causing it? > > > > Thank you in advance > > Kind regards > > Anouk > > > > _______________________________________________ > > gem5-users mailing list > > gem5-users <at> gem5.org > > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > Hi, > > I am facing the same issue as well. Curious if you found a solution to > this. > Thanks > > > > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > -- Joel Hestness PhD Candidate, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
