Hi Ivan, I believe the correct way to do this is fastforward individual benchmark to the beginning of its ROI, and take a checkpoint there for single core. And then aggregate multiple checkpoints to form one checkpoint for multi-programmed case. There is a checkpoint aggregator at /utils/checkpoint_aggregator.py.
I did not get it working though. Bonzi On Fri, Feb 7, 2014 at 12:28 PM, Ivan Stalev <[email protected]> wrote: > Hi, > > For SPEC2006, I have starting instruction numbers for regions of interest > (ROI) for individual benchmarks obtained from simpoint. Now I want to run a > multi-programmed workload. Unfortunately, each benchmark has a different > ROI and there does not seem to be an obvious way to make them > simultaneously execute in their respective (ROIs). For example, benchmark > X's ROI may be after 4B instructions, while benchmark Y might start from 8B > instructions. > > Does anyone have a solution for this? > > Thank you, > > Ivan > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
