Hi,
I am trying to run spec benchmarks on different cores. I use se.py where my
command line is:
build/ALPHA/gem5.opt --stats-file=mcf-4-identical.txt configs/example/se.py
--benchmark "mcf;mcf;mcf;mcf" --fast-forward 1000000000 --maxinsts
100000000 --cpu-type detailed --caches --l2cache --num-cpus 4 > mcf.ou
where I run 4 identical copies of mcf.
However, for some of the benchmarks I get segfault like mcf. For other
benchmarks I get: Exiting @ tick 1825633000 because target called exit().
Note that a single core works.
Here is significant part of a script:
if options.cpu_type == "detailed" or options.cpu_type == "inorder":
#check for SMT workload
workloads = options.benchmark.split(';')
if len(workloads) > 1:
#process = []
smt_idx = 0
for wrkld in workloads:
print wrkld
if wrkld == 'bzip2':
smt_process = alphaspec2006.bzip2_bench()
elif wrkld == 'gcc':
smt_process = alphaspec2006.gcc_bench()
elif wrkld == 'mcf':
print "in mcf"
smt_process = alphaspec2006.mcf_bench()
elif wrkld == 'gobmk':
smt_process = alphaspec2006.gobmk_bench()
elif wrkld == 'hmmer':
smt_process = alphaspec2006.hmmer_bench()
elif wrkld == 'sjeng':
smt_process = alphaspec2006.sjeng_bench()
elif wrkld == 'libquantum':
smt_process = alphaspec2006.libquantum_bench()
elif wrkld == 'h264ref':
smt_process = alphaspec2006.h264ref_bench()
elif wrkld == 'omnetpp':
smt_process = alphaspec2006.omnetpp_bench()
elif wrkld == 'astar':
smt_process = alphaspec2006.astar_bench()
elif wrkld == 'xalancbmk':
smt_process = alphaspec2006.xalancbmk_bench()
#elif wrkld == 'specrand_i':
# smt_process = alphaspec2006.specrand_i_bench()
#process += [smt_process, ]
process.append(smt_process)
smt_idx += 1
...........
for i in xrange(np):
#system.cpu[i].workload = multiprocesses[i]
print process
system.cpu[i].workload = process[i]
Does anyone has clue?
Thanks,
--
Heba
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
