Implement Code level Performance Analysis using the PMUv3 plugin
Introduction
PMUv3 plugin features
Download and build the PMUv3 plugin
Instrument one section of code
Plot, visualize, and analyze the results
Instrument multiple sections of code
Next Steps

  Log an issue

  Fork and edit

  Discuss on Discord

Visualize the results

A Python program is available to help you plot, visualize, and analyze the results collected with the PMUv3 plugin.

You need python3 and a number of Python packages.

If you are running on Ubuntu, install the following packages:

    

        
        
sudo apt install python-is-python3 python3-pip python3-venv -y

Create and activate a Python virtual environment:

    

        
        
python3 -m venv venv
source venv/bin/activate

Next, use Pip to install the required packages:

    

        
        
pip install pandas pyyaml matplotlib PyPDF2

Download the Python application code to plot and analyze results:

    

        
        
git clone https://github.com/GayathriNarayana19/Performance_Analysis_Backend.git

Copy the code below into a file named config.yaml in your test/ directory which contains your CSV files:

    

        
        
base_dirs:
  - path: '.'
    output_file: 'metrics.csv'
output_dir: './test_plotting/'
base_filename: 'bundle{}.csv'
num_bundles: 15
scenarios:
  - "test1: section1"
title: 'Section1'

#########DO NOT MODIFY BELOW THIS LINE##########
kpi_metrics:
  - - ['L1_I-cache_MPKI', ['L1I_CACHE_REFILL', 'INST_RETIRED']]
    - ['I-side_page_table_MPKI', ['ITLB_WALK', 'INST_RETIRED']]
    - ['L2_cache_MPKI', ['L2D_CACHE_REFILL', 'INST_RETIRED']]
    - ['Branch_MPKI', ['BR_MIS_PRED_RETIRED', 'INST_RETIRED']]
    - ['D-side_page_table_MPKI', ['DTLB_WALK', 'INST_RETIRED']]
    - ['L1_D-cache_MPKI', ['L1D_CACHE_REFILL', 'INST_RETIRED']]
    - ['LLC_cache_MPKI', ['LL_CACHE_MISS_RD', 'INST_RETIRED']]
  - - ['L1_data_TLB_read_miss_rate', ['L1D_TLB_REFILL_RD', 'L1D_TLB_RD']]
    - ['L2_TLB_miss_rate', ['L2D_TLB_REFILL', 'L2D_TLB']]
    - ['L2_TLB_write_miss_rate', ['L2D_TLB_REFILL_WR', 'L2D_TLB_WR']]
    - ['L2_TLB_read_miss_rate', ['L2D_TLB_REFILL_RD', 'L2D_TLB_RD']]
    - ['L1_data_TLB_miss_rate', ['L1D_TLB_REFILL', 'L1D_TLB']]
    - ['L1_instruction_TLB_miss_rate', ['L1I_TLB_REFILL', 'L1I_TLB']]
    - ['L1_data_TLB_write_miss_rate', ['L1D_TLB_REFILL_WR', 'L1D_TLB_WR']]
  - - ['L1_D-cache_read_miss_rate', ['L1D_CACHE_REFILL_RD', 'L1D_CACHE_RD']]
    - ['L1_D-cache_write_miss_rate', ['L1D_CACHE_REFILL_WR', 'L1D_CACHE_WR']]
    - ['L1_D-cache_miss_rate', ['L1D_CACHE_REFILL', 'L1D_CACHE']]
    - ['L1_I-cache_miss_rate', ['L1I_CACHE_REFILL', 'L1I_CACHE']]
    - ['L2_cache_miss_rate', ['L2D_CACHE_REFILL', 'L2D_CACHE']]
    - ['L1_D-cache_rate_of_cache_misses_in_L1_and_L2', ['L1D_CACHE_REFILL_OUTER', 'L1D_CACHE_REFILL']]
  - - ['Front_end_stall_rate', ['STALL_FRONTEND', 'CPU_CYCLES']]
    - ['Back_end_stall_rate', ['STALL_BACKEND', 'CPU_CYCLES']]
  - - ['Speculatively_executed_IPC', ['INST_SPEC', 'CPU_CYCLES']]
    - ['Architecturally_executed_IPC', ['INST_RETIRED', 'CPU_CYCLES']]
  - - ['VFP_instruction_rate_per_instructions', ['VFP_SPEC', 'INST_SPEC']]
    - ['DMB_rate_per_instructions', ['DMB_SPEC', 'INST_SPEC']]
    - ['DP_instruction_rate_per_instructions', ['DP_SPEC', 'INST_SPEC']]
    - ['ISB_rate_per_instructions', ['ISB_SPEC', 'INST_SPEC']]
    - ['CRYPTO_instruction_rate_per_instructions', ['CRYPTO_SPEC', 'INST_SPEC']]
    - ['PC_WRITE_instruction_rate_per_instructions', ['PC_WRITE_SPEC', 'INST_SPEC']]
    - ['SIMD_instruction_rate_per_instructions', ['ASE_SPEC', 'INST_SPEC']]
    - ['BR_IMMED_instruction_rate_per_instructions', ['BR_IMMED_SPEC', 'INST_SPEC']]
    - ['ST_instruction_rate_per_instructions', ['ST_SPEC', 'INST_SPEC']]
    - ['BR_RETURN_instruction_rate_per_instructions', ['BR_RETURN_SPEC', 'INST_SPEC']]
    - ['DSB_rate_per_instructions', ['DSB_SPEC', 'INST_SPEC']]
    - ['LD_instruction_rate_per_instructions', ['LD_SPEC', 'INST_SPEC']]
    - ['BR_INDIRECT_instruction_rate_per_instructions', ['BR_INDIRECT_SPEC', 'INST_SPEC']]
    - ['Exception_rate_per_instructions', ['EXC_TAKEN', 'INST_RETIRED']]

kpi_file_groups:
  - ["bundle11.csv", "bundle12.csv"]
  - ["bundle0.csv", "bundle1.csv", "bundle7.csv", "bundle10.csv"]
  - ["bundle5.csv", "bundle6.csv", "bundle13.csv"]
  - ["bundle4.csv"]
  - ["bundle8.csv"]
  - ["bundle8.csv", "bundle9.csv", "bundle14.csv"]

Run the Python application to create the performance plots as follows:

    

        
        
python3 Performance_Analysis_Backend/PMUv3_Backend/pmuv3_plotting.py -config config.yaml

Look in the test_plotting/ directory for a CSV file and the PDF files with the results.

The next section explains how to instrument multiple sections of code.

Back
Next