Skip to main content

Cortex-M debugger for Python

Project description

pyOCD

[pyocd.io] [Docs] [Slack] [Mailing list] [CI results]

News

  • The v0.40 release introduces support for multiple concurrent GDB client connections to the same GDB server.

See the wiki news page for all recent news.

pyOCD is an open source Python based tool and package for programming and debugging Arm Cortex-M microcontrollers with a wide range of debug probes. It is fully cross-platform, with support for Linux, macOS, Windows, and FreeBSD.

A command line tool is provided that covers most use cases, or you can make use of the Python API to facilitate custom target control. A common use for the Python API is to run and control CI tests.

Support for more than 70 popular MCUs is built-in. In addition, through the use of CMSIS Device Family Packs, nearly every Cortex-M device on the market is supported.

The pyocd command line tool gives you total control over your device with these subcommands:

  • gdbserver: GDB remote server allows you to debug using gdb via either the console or several GUI debugger options.
  • load: Program files of various formats into flash or RAM.
  • erase: Erase part or all of an MCU's flash memory.
  • pack: Manage CMSIS Device Family Packs that provide additional target device support.
  • commander: Interactive REPL control and inspection of the MCU.
  • server: Share a debug probe with a TCP/IP server.
  • reset: Hardware or software reset of a device.
  • rtt: Stream Segger RTT IO with any debug probe.
  • list: Show connected devices.

The API and tools provide these features:

  • halt, step, resume control
  • read/write memory
  • read/write core registers
  • set/remove hardware and software breakpoints
  • set/remove watchpoints
  • write to flash memory
  • load binary, hex, or ELF files into flash
  • reset control
  • access CoreSight DP and APs
  • SWO and SWV
  • and more!

Configuration and customization is supported through config files, user scripts, and the Python API.

Requirements

  • Python 3.8.0 or later.†
  • macOS, Linux, Windows 7 or newer, or FreeBSD
  • A recent version of libusb. See libusb installation for details.
  • Microcontroller with an Arm Cortex-M CPU
  • Supported debug probe
    • CMSIS-DAP v1 (HID) or v2 (WinUSB), including:
      • Atmel EDBG/nEDBG
      • Atmel-ICE
      • Cypress KitProg3 or MiniProg4
      • DAPLink based debug probe, either on-board or standalone
      • Keil ULINKplus
      • NXP LPC-LinkII
      • NXP MCU-Link
      • WCH-Link (1a86:8011, 2a86:8011 and others)
    • PE Micro Cyclone and Multilink
    • Raspberry Pi Picoprobe
    • SEGGER J-Link
    • STLinkV2 or STLinkV3, either on-board or the standalone versions

† Version 0.29 is the last version to support Python 2.

Status

PyOCD is beta quality.

The Python API is considered stable for version 0.x, but will be changed in version 1.0.

Documentation

The pyOCD documentation is available on the pyocd.io website.

In addition to user guides, you can generate reference documentation using Doxygen with the supplied config file.

Installing

The full installation guide is available in the documentation.

For notes about installing and using on non-x86 systems such as Raspberry Pi, see the relevant documentation.

The latest stable version of pyOCD may be installed via pip as follows:

$ python3 -mpip install -U pyocd

Note: depending on your system, you may need to use python instead of python3.

The latest pyOCD package is available on PyPI as well as on GitHub.

To install the latest prerelease version from the HEAD of the develop branch, you can do the following:

$ python3 -mpip install --pre -U git+https://github.com/pyocd/pyOCD.git@develop

You can also install directly from the source by cloning the git repository and running:

$ python3 pip install .

Note that, depending on your operating system, you may run into permissions issues running these commands. You have a few options here:

  1. Under Linux, run with sudo -H to install pyOCD and dependencies globally. On macOS, installing with sudo should never be required, although sometimes permissions can become modified such that installing without using sudo fails.
  2. Specify the --user option to install local to your user account.
  3. Run the command in a virtualenv local to a specific project working set.

udev rules on Linux

On Linux, particularly Ubuntu 16.04+, you must configure udev rules to allow pyOCD to access debug probes from user space. Otherwise you will need to run pyOCD as root, using sudo, which is very highly discouraged. (You should never run pyOCD as root on any OS.)

To help with this, example udev rules files are included with pyOCD in the udev folder. The readme in this folder has detailed instructions.

Target support

See the target support documentation for information on how to check if the MCU(s) you are using have built-in support, and how to install support for additional MCUs via CMSIS-Packs.

Using GDB

See the GDB setup documentation for a guide for setting up and using pyocd with gdb and IDEs.

Community resources

Join the pyOCD community!

pyocd.io website
Documentation
Issues
Discussions
Wiki
Mailing list for announcements

In order to foster a healthy and safe environment, we expect contributors and all members of the community to follow our Code of Conduct.

Contributions

We welcome contributions in any area, even if you just create an issue. If you would like to get involved but aren't sure what to start with, just ask on Slack or GitHub discussions and we'll be happy to help you. Or you can look for an open issue. Any work on major changes should be discussed with the maintainers to make everyone is aligned.

Please see the contribution guidelines for detailed requirements. The developers' Guide has instructions on how to set up a development environment for pyOCD.

New pull requests should be created against the develop branch. (You have to change the "base" to develop.)

License

PyOCD is licensed with the permissive Apache 2.0 license. See the LICENSE file for the full text of the license. All documentation and the website are licensed with CC BY 4.0.

Copyright © 2006-2025 PyOCD Authors

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyocd-0.41.0.tar.gz (16.1 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pyocd-0.41.0-py3-none-any.whl (14.5 MB view details)

Uploaded Python 3

File details

Details for the file pyocd-0.41.0.tar.gz.

File metadata

  • Download URL: pyocd-0.41.0.tar.gz
  • Upload date:
  • Size: 16.1 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.1

File hashes

Hashes for pyocd-0.41.0.tar.gz
Algorithm Hash digest
SHA256 a43c22d9ba6e96ec4447fbd4d5ab3963a339f8f753b785773c155944b6785c91
MD5 7f6e7e92fe001097d649425a83b17404
BLAKE2b-256 b06cccbd3035abbdc1d9b7215b8ef55189bae78d710f801f42520a7677d128f8

See more details on using hashes here.

File details

Details for the file pyocd-0.41.0-py3-none-any.whl.

File metadata

  • Download URL: pyocd-0.41.0-py3-none-any.whl
  • Upload date:
  • Size: 14.5 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.1

File hashes

Hashes for pyocd-0.41.0-py3-none-any.whl
Algorithm Hash digest
SHA256 7b9f0131cea3ff6b8bab906c90f4795f2deb563b0150fcaaa39f994af23e84d2
MD5 de92ffff564e0b4cb4b5660acc2dd02d
BLAKE2b-256 d79f6abfa347e033488f8c708a6c20f37b487eb98991f2ab8225c162c0aa8a01

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page