Hi, We are trying to use the parsec ROI hooks to create a checkpoint at the ROI and restore it later as part of a batch job with lots of simulations running in parallel.
We can successfully create a checkpoint (for fluidanimate in this example) by doing: export CHECKPOINT_NAME=whatever parsecmgmt -a run -p fluidanimate -n 1 -c gcc-hooks -i simsmall -x roi However, when we restart marss and do "loadvm whatever" in the monitor and switch back to the console, it runs the ROI in qemu emulation mode only and finishes quickly. Are we supposed to just run the simconfig command in monitor mode as part of the restore process before we do loadvm? Or is there a way to have the checkpoint restore in ptlsim simulation mode without using simconfig in the monitor mode (without having to modify the parsec ROI hooks)? Thanks, Jim Stevens University of Maryland, College Park _______________________________________________ http://www.marss86.org Marss86-Devel mailing list [email protected] https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
