Hello,
I hope all is well. I am using gem5 full system simulation. I want to generate
checkpoint after boot of the image. then restore this checkpoint and take
checkpoint after 1000000 instructions. The problem is the generated checkpoint
after booting is marked by tick number. the python command i am using for
restoring is:
$GEM5_PATH/build/X86/gem5.opt --verbose --outdir=$OUTPUT_DIR -r -e
--stdout-file=stdout.txt --stderr-file=stderr.txt $CONFIG_FILE --os-type=linux
--disk-image=$IMG_DIR --caches --cpu-type=AtomicSimpleCPU
--restore-with-cpu=AtomicSimpleCPU --kernel=./vmlinux-5.4.49 -r 1
--checkpoint-dir=${BM_DIR}/chkps/ --script=${BM_DIR}/my_fs.rcS -n 2
--mem-type=SimpleMemory --mem-size=64GB --at-instruction --take-checkpoints
10000000 --max-checkpoints 10
I tried to rename the checkpoint folder to match the expected folder name
(cpt.None.1) but i got the following error:
Global frequency set at 1000000000000 ticks per second
0: system.pc.south_bridge.cmos.rtc: Real-time clock set to Sun Jan 1
00:00:00 2012
Creating checkpoint at inst:10000001
120260822403500: system.pc.com_1.device: attach terminal 0
exit cause = a thread reached the max instruction count
Writing checkpoint
Checkpoint written.
Exiting @ tick 120278584296500 because a thread reached the max instruction
count
i think the error is in the changing of the format. how can i change the format
as appropriate (how can i map the tick number to instruction number)?. also,
how can i generate checkpoint using m5 checkpoint but with instruction number
instead of tick number?
Thanks in advance,
Bishoy Attia
_______________________________________________
gem5-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
