I don't think you're "getting" my other points:
A checkpoint taken when running with N cpus **must** be restored on a system running N cpus. This is inherent in the OS state and also in the fact that the registers of N execution engines are saved / restored. Put another way, the fact that N cpus are running is "baked into" the checkpoint state. This has nothing to do with BBVs, except that you must be running with 1 cpu to collect them. Lastly, BBVs and simpoint methodology do not extend to multiple cpus. It's inherently a single cpu technique. So what you seem to want to do, to my knowledge, just can't be done. EM _______________________________________________ gem5-users mailing list -- gem5-users@gem5.org To unsubscribe send an email to gem5-users-le...@gem5.org
