resources: Clean-up the parsec README.md
This commit removes some of the experiment specific test in the parsec
README.md. Resources are not for specific experiment setups.
Other small cleanups have been added: typos, grammar, phrasing, etc.
Change-Id: Iabaf767c7a3cd6f708bcac8423939332c58c20bc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5-resources/+/41257
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: Bobby R. Bruce <bbruce@ucdavis.edu>
diff --git a/src/parsec/README.md b/src/parsec/README.md
index 9956be3..5090f15 100644
--- a/src/parsec/README.md
+++ b/src/parsec/README.md
@@ -68,12 +68,13 @@
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 used to run the experiments with. In these experiments we only used kernel version 4.19.83 (You can download the binary [here](http://dist.gem5.org/dist/v20-1/kernels/x86/static/vmlinux-4.19.83)).
-* disk: The path to the PARSEC disk-image (The disk-image created above will work for both set of experiments).
-* cpu: The type of cpu that is used to run the simulation with. There are two possible options: kvm (KvmCPU) and timing (TimingSimpleCPU).
-* benchmark: The workload among 13 workloads of PARSEC. They include `blackscholes`, `bodytrack`, `canneal`, `dedup`, `facesim`, `ferret`, `fluidanimate`, `freqmine`, `raytrace`, `streamcluster`, `swaptions`, `vips`, `x264`. For further information on the workloads read [here](https://parsec.cs.princeton.edu/).
-* size: The size of chosen workload. In our experiments we used only three different sizes. For the experiments with kvm size `simsmall`, `simlarge`, and `native` are used. For the experiments with timing only size `simsmall` has been used.
-* num_cpus: Number of cpus used to run the simulation with. For experiments with classic memory (located in `configs`) only valid option is `1`. However for the experiments with ruby memory (located in `configs-mesi-two-level`) the number of cores differ based on what cpu model is used the below table shows what core counts have been used with each cpu model.
+* **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/v20-1/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:
+
| CPU Model | Core Counts |
|-----------------|-------------|
@@ -83,9 +84,8 @@
Below are the examples of running an experiment with the two configurations.
```sh
-gem5/build/X86/gem5.opt configs/run_parsec.py linux-stable/vmlinux-4.19.83 disk-image/parsec/parsec-image/parsec timing bodytrack simsmall 1
-
-gem5/build/X86_MESI_Two_Level/gem5.opt configs-mesi-two-level/run_parsec_mesi_two_level.py linux-stable/vmlinux-4.19.83 disk-image/parsec/parsec-image/parsec timing raytrace simsmall 2
+<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
