Hi,
I am working a project that need to measure the maximum memory bandwidth
that can achieved for CPUs. I’m using the following parameters to start
simulation:
./build/ARM/gem5.fast ./configs/example/fs.py --kernel=vmlinux
--dtb-file=armv8_gem5_v1_4cpu.dtb --machine-type=VExpress_GEM5_V1 --num-cpu=4
--cpu-clock=3GHz --caches --l2cache --cpu-type=TimingSimpleCPU --mem-size=4GB
--mem-type=DDR4_2400_8x8
However I observed that the speed measured by application running on the
simulated system is only about 300MB/s, which is much lower than the calculated
memory bandwidth for this type of memory. I’m not sure if the CPU or the cache
system is the reason of that.
_______________________________________________
gem5-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
%(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
