So, it looks like you can create a checkpoint using fs.py and restore using ruby_fs.py.
To reduce the simulation time of PARSEC benchmarks, the checkpoint is taken using TimingSimple CPU and restored in detailed mode. But, in the current RUBY model, one cannot restore a checkpoint in O3 if it is taken in timing mode. Am I correct in saying this? Why don't you test it out? I did test it out and only then posted the message. *http://www.mail-archive.com/gem5-users@gem5.org/msg03973.html* * * *I received a message as the above link. * * * *Created checkpoint using following command* * * ./build/ALPHA_MOESI_CMP_directory/9123moesi.opt -d 9123moesi_results/blackscholes configs/example/fs.py --cpu-type=timing -n 4 -b Blackscholes Restored using ./build/ALPHA_MOESI_CMP_directory/9123moesi.opt --debug-flags=ProtocolTrace --trace-start=8115519521000 -d 9123moesi_results/blackscholes/ configs/example/ruby_fs.py --ruby --cpu-type=detailed --restore-with-cpu=timing --caches --l2cache --checkpoint-restore=1 -n 4 Received the following message Listening for system connection on port 3456 0: system.tsunami.io.rtc: Real-time clock set to Thu Jan 1 00:00:00 2009 0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000 0: system.remote_gdb.listener: listening for remote gdb #1 on port 7001 0: system.remote_gdb.listener: listening for remote gdb #2 on port 7002 0: system.remote_gdb.listener: listening for remote gdb #3 on port 7003 fatal: Can't unserialize 'system.cpu0:locked' @ cycle 8115519521000 [paramIn:build/ALPHA_MOESI_CMP_directory/sim/serialize.cc, line 228] Memory Usage: 739476 KBytes Thanks, Anusha
_______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
