Skip to content

Latest commit

 

History

History
40 lines (34 loc) · 2.2 KB

README.md

File metadata and controls

40 lines (34 loc) · 2.2 KB
Raw

Cache Simulator

Introduction

A simple cache simulator written in Rust programming language.

Steps to run

Run immediately:

  • trace 1: make run config_path=./configs/test.json trace_path=./trace/trace1.txt
  • trace 2: make run config_path=./configs/test.json trace_path=./trace/trace2.txt

Detailed Steps:

  1. Get Rust toolchain and make sure you could compile Rust codes with cargo.
  2. Clone the repository: git clone https://github.com/xuehaonan27/cache-simulator.git.
  3. Enter the repository: cd cache-simulator.
  4. Prepare a configuration file as described below.
  5. Prepare a trace file.
  6. Run test with make run config_path=</path/to/your/config/file> trace_path=</path/to/your/trace/file>.
  7. The results will be printed to stdout with pretty format (as least I think it's pretty).
  8. Add > /path/to/your/save/file to save the results to a file.

Configuration file

You can see example configuration files in configs directory, e.g. test.json.

A configuration file that could be parsed by the simulator must be a json file, containing a single object with four required fields:

  • CPU_GHZ: CPU frequency in Ghz.
  • using_latency_as_cycles:
    • When set to true: "hit_latency": 1 or "bus_latency": 1 means 1 cycle.
    • When set to false: "hit_latency": 1 or "bus_latency": 1 means 1 nanosecond.
  • memory: configuration describing memory.
  • caches: a list of configuration objects describing caches. The layer fields must be continuous integers like 1, 2, 3, ..., which mean L1, L2, L3, ... caches.

Raw data for Lab3-1

  • configs/group_1 contains configuration files for question 1, and so on and so forth. These configuration files are generated by build_scripts.py.
  • sim_results/group_1 contains results for question 1, and so on and so forth. These results are generated by the simulator, and the auto script is run_scripts.py.
  • figures/group_1 contains pictures visualizing the results for question 1 and 2. The figures are plotted by with plot_scripts.py.
  • L1s64Kw8bs64_L2s512Kw8bs64 means:
    • L1 cache total size 64K, ways 8 (associativity), block size 64B.
    • L2 cache total size 512K, ways 8 (associativity), block size 64B.