Hi,
I am trying to simulate the multi-programed workload with fast-forward,
warmup and detailed simulation. After the warmup, the gem5 ends very soon
and it does not execute the detailed simulation with the speficied number
of intructions. However, the gem5 works well for the single-programmed
workload with standard switch.
When I ran the gem5.debug for the multi-programmed wrokload and it reports:
gem5.debug: build/ALPHA/cpu/simple/timing.cc:112: virtual unsigned int
TimingSimpleCPU::drain(DrainManager*): Assertion `!fetchEvent.scheduled()'
failed.
Before updating the gem5, it worked well with the gem5 changeset 9727.
Here are my command lines for the multiple-programmed and
single-programmed cases respectively:
build/ALPHA/gem5.fast configs/example/se.py -n 4
--bench=astar_lake-astar_lake-astar_lake-astar_lake \
--caches --cacheline_size 64 --l1i_size 32kB --l1d_size 32kB --l2cache
--l2_size 2MB \
--cpu-clock 4GHz \
--mem-size 8GB --mem-type DDR3_1600_x64 \
-F 100000 -W 300000 -I 5000000 -s 1
build/ALPHA/gem5.fast configs/example/se.py -n 1 --bench=astar_lake
\
--caches --cacheline_size 64 --l1i_size 32kB --l1d_size 32kB --l2cache
--l2_size 2MB \
--cpu-clock 4GHz \
--mem-size 8GB --mem-type DDR3_1600_x64 \
-F 100000 -W 300000 -I 5000000 -s 1
Thanks.
Jianhui
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
