Hi,

I have Splash-2 compiled with m5threads for the X86 architecture. I'm
trying to run it with the classic memory system, in SE mode, with the
TimingSimpleCPU. I'm not sure these runs are making forward progress. They
have been running for over 2 days. Looking at a trace I see a lot of
repetitive addresses being generated.

Here is the exact command I
used: ~/gem5-stable-6a043adb1e8d/build/X86_MESI_CMP_directory/gem5.opt
~/gem5-stable-6a043adb1e8d/configs/example/se.py --l2cache --l1d_size=32kB
--l1d_assoc=4 --l1i_size=32kB --l1i_assoc=4 --l2_size=256kB --l2_assoc=8
--cacheline_size=64 --num-cpus=4
--cmd=/home/DREXEL/sn446/ECEC414/X86/splash2/splash2/codes/kernels/lu/non_contiguous_blocks/LU
--options=-p4 -t --cpu-type=timing

Has anyone encountered this issue before? Has anyone run this particular
configuration?

When I switch to the AtomicSimpleCPU all my runs finish very quickly.

Regards,
Sid
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to