Skip to main content

Analysing code footprint on embedded microcontrollers using GCC generated Map files

Project description

https://img.shields.io/pypi/v/fpvgcc.svg?logo=pypi https://img.shields.io/pypi/pyversions/fpvgcc.svg?logo=pypi https://img.shields.io/travis/ebs-universe/fpv-gcc.svg?logo=travis https://img.shields.io/coveralls/github/ebs-universe/fpv-gcc.svg?logo=coveralls https://img.shields.io/requires/github/ebs-universe/fpv-gcc.svg https://img.shields.io/pypi/l/fpvgcc.svg

Deprecation Notice

This (v1.1.3) will be the final release of the 1.x series of fpvgcc. The tool has evolved to the extent possible within the naive framework it was originally built on, and a number of open issues cannot be resolved without refactoring big chunks of the code. fpvgcc v1.1.2 is essentially functional and will remain available in its present form.

Though I have been unable to find time to maintain and build up the tool after the original releases, I hope it has been useful to the few people who’ve used it.

A number of other tools are now available which do a similar job. While I have minor qualms about each of those tools, such qualms are mostly subjective and aesthetic objections. I am forced to admit that fpvgcc would never have been written if some of those tools had existed at the time of its writing.

I am considering starting work on a 2.x series of fpvgcc, which will likely be a page one rewrite. The following are the main planned changes:

  • Reduce the number of assumptions made when parsing map files. For example, allow multiple object files with the same name.

  • Use polars or something suitably performant to store the parsed map information.

  • Add a minimal fastapi interface to be able to provide a GUI using a webview or similar.

  • Remove python 2 support. Minimum python version required will probably be py37 or py38.

I expect it will be some time before 2.x will be usable. If and when development starts, it will be available in the v2 branch of this repository. When v2.x reaches feature parity with v1.1.2, it will be merged in to main and python packages for fpvgcc 2.x will be published to pypi.

Please feel free to write to me at shashank at chintal dot in if any of the following apply to you:

  • you have been using fpvgcc or are maintaining a public or private fork and have any concerns regarding the deprecation or plans for 2.x

  • you wish to take over the 1.x codebase and keep it functional and available in the future

  • you have any suggestions for features to include in 2.x

  • you wish to contribute to the development of 2.x

Introduction

fpvgcc is a python script/package to help analyse code footprint on embedded microcontrollers using GCC generated Map files.

This module uses information contained within .map files generated by gcc (when invoked with -Wl,-Map,out.map), to provide easily readable summaries of static memory usage at various levels of the code hierarchy. This package generates no information that isn’t already contained within the .map file.

The provided outputs can be used to gain insight into the relative sizes of included code, and aid in prioritizing static memory optimization for very low memory platforms. Some provided functionality may also deliver minor usability improvements to the workflow involved in parsing though generated assembly listings.

Known Issues

This script was first written based on the format of mapfiles generated by msp430-elf-gcc, v4.9.1. Over time, it was modifed to accept elements found in mapfiles generated by later versions and gcc-based toolchains for other platforms.

Still, remember that the file parsing was implemented by observing the content of real mapfiles, and not based on a file format specification. Even with toolchains it was written to support, there are large sections of the file that are not actually used. Due to this, the outputs generated are not always accurate. Various boundary conditions result in minor errors in size reporting.

The following more serious issues are known. They should be fixed at some point, but for the moment I’ve chosen to work around them :

  • Having two C filenames with the same name (or generating the same obj name) in your tree will cause parsing to break on some platforms / toolchains.

Project Information

The latest version of the documentation, including installation, usage, and API/developer notes can be found at ReadTheDocs.

The latest version of the sources can be found at GitHub. Please use GitHub’s features to report bugs, request features, or submit pull/merge requests.

The principle author for fpvgcc is Chintalagiri Shashank. The author can be contacted if necessary via the information on the author’s github profile . See the AUTHORS file for a full list of collaborators and/or contributing authors, if any.

fpvgcc is distributed under the terms of the GPLv3 license . A copy of the text of the license is included along with the sources.

Project details


Download files

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

Source Distribution

fpvgcc-1.1.2.tar.gz (211.5 kB view details)

Uploaded Source

File details

Details for the file fpvgcc-1.1.2.tar.gz.

File metadata

  • Download URL: fpvgcc-1.1.2.tar.gz
  • Upload date:
  • Size: 211.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.3

File hashes

Hashes for fpvgcc-1.1.2.tar.gz
Algorithm Hash digest
SHA256 942c1a5f0d21fa8202a5fadb053f0464de1daf5d2edd2b107b2c399e267d3c5a
MD5 833223b2117609ff434f38f237a9a96a
BLAKE2b-256 e503bbe89be88b39606895d567109ba98e0be4bffea36007b3028e131af5b614

See more details on using hashes here.

Supported by

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