This document provides a guide on how to use Google Performance Tools to test performance on a Raspberry Pi and generate performance analysis reports.

1. Download and Install Google Performance ToolsCopy Anchor Link

wget https://github.com/gperftools/gperftools/releases/download/gperftools-2.15/gperftools-2.15.tar.gz
tar -xvf ./gperftools-2.15.tar.gz
cd gperftools-2.15
./configure
make
sudo make install

The LD_PRELOAD environment variable is set correctly to automatically load the performance analysis library when running the program.

export LD_PRELOAD="/usr/local/lib/libprofiler.so"

2. Install Necessary DependenciesCopy Anchor Link

The necessary dependencies to generate PDF reports:

sudo apt-get install ghostscript graphviz

3. Run the Program and Generate Performance DataCopy Anchor Link

Execute program and output the performance data to a specified file, for example:

CPUPROFILE=./prof.out HEAPPROFILE=heap CPUPROFILESIGNAL=12 ./pi-webrtc --camera=v4l2:0 --fps=30 --width=1280 --height=960 --v4l2-format=h264 --hw-accel --uid=home-pi-3b

Send a signal to the process to start/stop collect performance data:

killall -12 pi-webrtc

4. Use pprof Tool to Generate Performance ReportCopy Anchor Link

Convert the performance data into a PDF report using the pprof tool:

pprof /home/pi/IoT/pi-webrtc/build/pi-webrtc prof.out.0 --pdf > prof_0.pdf
pprof --pdf --base=heap.0001.heap ./test_recorder heap.0003.heap > heap_diff.pdf

ReferenceCopy Anchor Link

• gperftools