lyrl/palanteer

High performance visual profiler, debugger, tests enabler for C++ and Python

Look into Palanteer and get an omniscient view of your program

Palanteer is a set of lean and efficient tools to improve the quality of software, for C++ and Python programs.

Simple code instrumentation, mostly automatic in Python, delivers powerful features:

• Collection of meaningful atomic events on timings, memory, locks wait and usage, context switches, data values..
• Visual and interactive observation of record: timeline, plot, histograms, flame graph...
• Remote command call and events observation can be scripted in Python: deep testing has never been simpler
• C++:
  • ultra-light single-header cross-platform instrumentation library
  • compile-time selection of groups of instrumentation
  • compile-time hashing of static string to minimize their cost
  • compile-time striping of all instrumentation static strings
  • enhanced assertions, stack trace dump...
• Python:
  • Automatic instrumentation of functions enter/leave, memory allocations, raised exceptions, garbage collection runs

Palanteer is an efficient, lean and comprehensive solution for better and enjoyable software development!

C++ instrumentation example

Below is a simple example of a C++ program instrumented with Palanteer and generating 100 000 random integers. The range can be remotely configured with a user-defined CLI.

The Python scripting module can control this program, in particular:

• call the setBoundsCliHandler to change the configuration
• temporarily stop the program at the freeze point
• see all "random data" values and the timing of the scope event "Generate some random values"

See C++ example code

// File: example.cpp
// On Linux, build with:  g++ -DUSE_PL=1 -I <palanteer C++ instrumentation folder> example.cpp -lpthread -o example
#include <stdlib.h>          // For "rand"
#define PL_IMPLEMENTATION 1  // The instrumentation library shall be "implemented" once
#include "palanteer.h"

int globalMinValue = 0, globalMaxValue = 10;

// Handler (=user implementation) of the example CLI, which sets the range
void setBoundsCliHandler(plCliIo& cio)             // 'cio' is a communication helper passed to each C++ CLI handler
{
    int minValue = cio.getParamInt(0);             // Get the 2 CLI parameters as integers (as declared)
    int maxValue = cio.getParamInt(1);
    if(minValue>maxValue) {                        // Case where the CLI execution fails. The text answer contains some information about it
        cio.setErrorState("Minimum value (%d) shall be lower than the maximum value (%d)", minValue, maxValue);
        return;
    }

    // Modify the state of the program. No care about thread-safety here, to keep the example simple
    globalMinValue = minValue;
    globalMaxValue = maxValue;
    // CLI execution was successful (because no call to cio.setErrorState())
}


int main(int argc, char** argv)
{
    plInitAndStart("example");              // Start the instrumentation, for the program named "example"
    plDeclareThread("Main");                // Declare the current thread as "Main" so that it can be identified more easily in the script
    plRegisterCli(setBoundsCliHandler, "config:setRange", "min=int max=int", "Sets the value bounds of the random generator");  // Declare our CLI
    plFreezePoint();                        // Add a freeze point here to be able to configure the program at a controlled moment

    plBegin("Generate some random values");
    for(int i=0; i<100000; ++i) {
        int value = globalMinValue + rand()%(globalMaxValue+1-globalMinValue);
        plData("random data", value);       // Here are the "useful" values
    }
    plEnd("");                              // Shortcut for plEnd("Generate some random values")

    plStopAndUninit();                      // Stop and uninitialize the instrumentation
    return 0;
}

Some C++ performance figures (see here for more details):

• nanosecond resolution and ~25 nanoseconds cost per event on a standard x64 machine
• up to ~3 millions events per second when recording, the bottleneck being on the server processing side
• up to ~150 000 events per second when processing the flow through a Python script, the bottleneck being on the Python script side

More details and an example of remote script is provided here

Python instrumentation example

Execution of unmodified Python programs can be analyzed directly with a syntax similar to the one of cProfile, as a large part of the instrumentation is automated by default:

• Functions enter/leave
• Interpreter memory allocations
• All raised exceptions
• Garbage collection runs

In some cases, a manual instrumentation which enhances or replaces the automatic one is desired.
The example below is such an equivalent of the C++ one above but in Python:

See Python manual instrumentation example code

#! /usr/bin/env python3
import sys
import random
from palanteer import *

globalMinValue, globalMaxValue =  0, 10

# Handler (=implementation) of the example CLI, which sets the range
def setBoundsCliHandler(minValue, maxValue):              # 2 parameters (both integer) as declared
    global globalMinValue, globalMaxValue
    if minValue>maxValue:                                 # Case where the CLI execution fails (non null status). The text answer contains some information about it
        return 1, "Minimum value (%d) shall be lower than the maximum value (%d)" % (minValue, maxValue)

    # Modify the state of the program
    globalMinValue, globalMaxValue = minValue, maxValue
    # CLI execution was successful (null status)
    return 0, ""


def main(argv):
    global globalMinValue, globalMaxValue

    plInitAndStart("example")                             # Start the instrumentation
    plDeclareThread("Main")                               # Declare the current thread as "Main", so that it can be identified more easily in the script
    plRegisterCli(setBoundsCliHandler, "config:setRange", "min=int max=int", "Sets the value bounds of the random generator")  # Declare the CLI
    plFreezePoint()                                       # Add a freeze point here to be able to configure the program at a controlled moment

    plBegin("Generate some random values")
    for i in range(100000):
        value = int(globalMinValue + random.random()*(globalMaxValue+1-globalMinValue))
        plData("random data", value)                      # Here are the "useful" values
    plEnd("")                                             # Shortcut for plEnd("Generate some random values")

    plStopAndUninit()                                     # Stop and uninitialize the instrumentation

# Bootstrap
if __name__ == "__main__":
    main(sys.argv)

Scripting example

Both examples above (C++ and Python) can be remotely controlled with a simple Python script.

Typical usage is:

• Tests based on stimulations/configuration with CLI and event observation, as data can also be logged
• Evaluation of the program performance
• Monitoring
• ...

See a scripting example code (Python)

#! /usr/bin/env python3
import sys
import palanteer_scripting as ps

def main(argv):
    if len(sys.argv)<2:
        print("Error: missing parameters (the program to launch)")
        sys.exit(1)

    # Initialize the scripting module
    ps.initialize_scripting()

    # Enable the freeze mode so that we can safely configure the program once stopped on its freeze point
    ps.program_set_freeze_mode(True)

    # Launch the program under test
    ps.process_launch(sys.argv[1], args=sys.argv[2:])
    # From here, we are connected to the remote program

    # Configure the selection of events to receive
    my_selection = ps.EvtSpec(thread="Main", events=["random data"]) # Thread "Main", only the event "random data"
    ps.data_configure_events(my_selection)

    # Configure the program
    status, response = ps.program_cli("config:setRange min=300 max=500")
    if status!=0:
        print("Error when configuring: %s\nKeeping original settings." % response)

    # Disable the freeze mode so that the program resumes its execution
    ps.program_set_freeze_mode(False)

    # Collect the events as long as the program is alive or we got some events in the last round
    qty, sum_values, min_value, max_value, has_worked = 0, 0, 1e9, 0, True
    while ps.process_is_running() or has_worked:
        has_worked = False
        for e in ps.data_collect_events(timeout_sec=1.):  # Loop on received events, per batch
            has_worked, qty, sum_values, min_value, max_value = True, qty+1, sum_values+e.value, min(min_value, e.value), max(max_value, e.value)

    # Display the result of the processed collection of data
    print("Quantity: %d\nMinimum : %d\nAverage : %d\nMaximum : %d" % (qty, min_value, sum_values/max(qty,1), max_value))

    # Cleaning
    ps.process_stop()            # Kills the launched process, if still running
    ps.uninitialize_scripting()  # Uninitialize the scripting module


# Bootstrap
if __name__ == "__main__":
    main(sys.argv)

The execution of this script (on Linux) gives the following output:

> time ./remoteScript.py example
Quantity: 100000
Minimum : 300
Average : 400
Maximum : 500
./remoteScript.py example  0.62s user 0.02s system 24% cpu 2.587 total  

Details can be found here.

Documentation

The complete documentation is accessible inside the repository, and online:

• Introduction
• Getting started
• Base concepts
• C++ instrumentation API
• C++ instrumentation configuration
• Python instrumentation API
• Scripting API
• More

OS Support

Viewer and scripting library:

• Linux 64 bits
• Windows 10

Instrumentation libraries:

• Linux 32 or 64 bits (tested on PC and armv7l)
• Windows 10
• Support for virtual threads (userland threads, like fibers)

Requirements

Palanteer is lean, its full installation requires only usual components:

• a C++14+ compiler (gcc, clang or MSVC) in Windows 10 or Linux 64 bits for the viewer and scripting module
• a C++11+ compiler (tested with gcc, clang and MSVC) 32 or 64 bits for the C++ instrumentation library
• CPython 3.7+
• OpenGL 3.3+

In particular, the C++ single-header instrumentation library requires only C++11 or above.

See here for more details on the requirements per component.

Other dependencies are snapshotted inside this repository, so for information only:

Dependency name	License type	URL
Khronos OpenGL API and Extension	MIT	https://www.khronos.org/registry/OpenGL/api/GL
Dear ImGui	MIT	https://github.com/ocornut/imgui
stb_image	Public domain	https://github.com/nothings/stb
Fonts 'Roboto-Medium.ttf'	Apache License, Version 2.0	https://fonts.google.com/specimen/Roboto
ZStandard	BSD	https://facebook.github.io/zstd
Markdeep	BSD	https://casual-effects.com/markdeep

License

The instrumentation libraries are under the MIT license.

The viewer and the Python scripting module are under the AGPLv3+.

See LICENSE.md for details.

Warning: Beta state

Even if no major bugs are known and a special care has been taken to test as many cases as possible, this project is young and in beta state.

Your feedback and raised issues are warmly welcome to improve its quality, especially as it aims at improving software quality...