RISC-V Application Profiler

The RISC-V Application Profiler is a Python-based tool designed to help software developers optimize the performance of their applications on RISC-V hardware. It works by parsing execution logs and providing insights about the application's behavior. The tool has a modular design, where performance metrics are added as plugins, allowing developers to customize the profiler to their specific needs. The profiler is highly accessible, easy to use, and can be adapted to capture different types of performance metrics. Overall, the RISC-V Application Profiler is a flexible and customizable solution for software developers who want to ensure optimal performance of their applications on RISC-V platforms.

Detailed documentation can be found here.

Installation

Install pycachesim. This is a requirement to use the caches plugin in the profiler.

git clone https://github.com/mahendraVamshi/pycachesim.git
cd pycachesim
pip install -e .
cd ..

Install riscv_isac. This is a development version of isac.

git clone https://github.com/mahendraVamshi/riscv-isac.git
cd riscv-isac
pip install -e .
cd ..

Finally, install the profiler itself.

git clone https://github.com/mahendraVamshi/riscv-application-profiler.git
cd riscv-application-profiler
pip install -e .

Usage

To display the help message, run:

riscv_application_profiler --help
riscv_application_profiler profile --help

To generate a log file, run:

spike --log-commits <path-to-binary>

NOTE: You need to use --enable-commitlog while configuring spike.

To profile an application, run:

riscv_application_profiler profile --log <path-to-log> --output <path-to-output-directory> --config <path-to-config-file> config.yaml

To profile an application with cycle accurate simulation, run:

riscv_application_profiler profile --log <path-to-log> --output <path-to-output-directory> --config <path-to-config-file> config.yaml --cycle_accurate_config <path-to-config-file> config.yaml

Info:

Path to the log file is mandatory. Example log files can be found in the sample_artifacts/logs directory.

Path to the output directory is optional. If not provided, the profiler will create a directory named build in the current working directory.

Path to the config file is mandatory. Example config.yaml is located in sample_configs/profiler_config directory. L2 cache config files are located in sample_configs/profiler_config/L2_configs directory.

Path to the cycle accurate config file is optional. Example config.yaml is located in sample_configs/cycle_accurate directory. Use this option only if you want to profile an application with cycle accurate simulation. L2 cache config files are located in sample_configs/cycle_accurate/L2_configs directory.

Command line options to the profile command:

Options:
  -l, --log TEXT                  This option expects the path to an execution
                                  log.  [required]
  -o, --output TEXT               Path to the output file.  [default: ./build]
  -c, --config TEXT               Path to the YAML configuration file.
                                  [required]
  -ca, --cycle_accurate_config TEXT
                                  Path to the YAML cycle accurate
                                  configuration file.
  -v, --verbose [info|error|debug]
                                  Set verbose level
  --help                          Show this message and exit.

Example:

To profile an application, run:

riscv_application_profiler profile --log ./sample_artifacts/logs/hello.log --output ./build --config ./sample_configs/profiler_config/config.yaml   

To profile an application with cycle accurate simulation, run:

riscv_application_profiler profile --log ./sample_artifacts/logs/hello.log --output ./build --config ./sample_configs/profiler_config/L2_configs/config.yaml --cycle_accurate_config ./sample_configs/cycle_accurate/L2_configs/config.yaml 

Note: The log file should be an execution log generated using spike as of today. Support for configuring log formats will be added in the future.

Note: Metrics such as grouping instructs by operation and privledge mode are hard coded in the profiler.py file as the input to these functions is the commit log. However, the metrics such as grouping instructs by CSRs and cache computation is mandatory for a cycle accurate profiling.

Features

The profiler supports the following list of features as plugins:

Grouping instructions by:

• Type of operation performed.
• Privilege mode used for execution.
• Directions and Sizes (for jumps/branches).

Lists:

• Presence of Nested Loops.
• Store-Load bypass.
• Presence of RAW dependencies.
• Pattern detection for custom instructions.

Histogram for:

• RegisterFile (XRF/FRF) usage.
• CSR accesses.
• D$/I$ Hits/Misses/Usage/Utilization.
• Unifed L2 Cache Hits/Misses/Usage/Utilization.