title: PARSEC tags: - x86 - fullsystem layout: default permalink: resources/parsec shortdoc: > Resources to build a disk image with the parsec workloads. author: [“Mahyar Samani”] license: BSD-3-Clause

This document includes instructions on how to create an Ubuntu 18.04 disk-image with PARSEC benchmark installed. The disk-image will be compatible with the gem5 simulator.

This is how the src/parsec-tests/ directory will look like if all the artifacts are created correctly.

parsec/
  |___ gem5/                                   # gem5 folder
  |
  |___ disk-image/
  |      |___ shared/
  |      |___ parsec/
  |             |___ parsec-image/
  |             |      |___ parsec             # the disk image will be here
  |             |___ parsec.json               # the Packer script
  |             |___ parsec-install.sh         # the script to install PARSEC
  |             |___ post-installation.sh      # the script to install m5
  |             |___ runscript.sh              # script to run each workload
  |             |___ parsec-benchmark          # the parsec benchmark suite
  |
  |___ configs
  |      |___ system                           # system config directory
  |      |___ run_parsec.py                    # gem5 run script
  |
  |___ configs-mesi-two-level
  |      |___ system                           # system config directory
  |      |___ run_parsec_mesi_two_level.py     # gem5 run script
  |
  |___ README.md

Notice that there are two sets of system configuration directories and run scripts. For further detail on the config files look here.

Building the disk image

In order to build the disk-image for PARSEC tests with gem5, build the m5 utility in src/parsec-tests/ using the following:

git clone https://gem5.googlesource.com/public/gem5
cd gem5/util/m5
scons build/x86/out/m5

We use packer to create our disk-image. The instructions on how to install packer is shown below:

cd disk-image
wget https://releases.hashicorp.com/packer/1.6.0/packer_1.6.0_linux_amd64.zip
unzip packer_1.6.0_linux_amd64.zip

In order to build the disk-image first the script needs to be validated. Run the following command to validate disk-image/parsec/parsec.json.

./packer validate parsec/parsec.json

After the script has been successfuly validated you can create the disk-image by runnning:

./packer build parsec/parsec.json

You can find the disk-image in parsec/parsec-image/parsec.

gem5 run scripts

There are two sets of run scripts and system configuration files in the directory. The scripts found in configs use the classic memory system while the scripts in configs-mesi-two-level use the ruby memory system with MESI_Two_Level cache coherency protocol. The parameters used in the both sets of experiments are explained below:

• kernel: The path to the linux kernel. We have verified capatibility with kernel version 4.19.83 which you can download at http://dist.gem5.org/dist/develop/kernels/x86/static/vmlinux-4.19.83. More information on building kernels for gem5 can be around in src/linux-kernel.
• disk: The path to the PARSEC disk-image.
• cpu: The type of cpu to use. There are two supported options: kvm (KvmCPU) and timing (TimingSimpleCPU).
• benchmark: The PARSEC workload to run. They include blackscholes, bodytrack, canneal, dedup, facesim, ferret, fluidanimate, freqmine, raytrace, streamcluster, swaptions, vips, x264. For more information on the workloads can be found at https://parsec.cs.princeton.edu/.
• size: The size of the chosen workload. Valid sizes are simsmall, simmedium, and simlarge.
• num_cpus: The number of cpus to simulate. When using configs, the only valid option is 1. When using configs-mesi-two-level the number of supported cpus is show in the table below:

Below are the examples of running an experiment with the two configurations.

<gem5 X86 binary> configs/run_parsec.py <kernel> <disk> <cpu> <benchmark> <size> <num_cpus>
<gem5 X86_MESI_Two_Level binary> configs-mesi-two-level/run_parsec.py <kernel> <disk> <cpu> <benchmark> <size> <num_cpus>

Working Status

The working status of PARSEC runs for gem5-20 has been documented here.