jriwanek / vctools

Some tools usable for reengineering the VideoCore IV processor found on the Raspberry Pi.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

VideoCore Tools

This repository contains a number of tools usable for reverse engineering the VideoCore CPU or for writing software for it.

Register Database

In the directory vcdb there is a database containing documentation about most known hardware registers in the area starting at 0x7e000000. The database is in yaml format, with one file for every register region. The central file which includes all other files is vcdb/_regdb.yaml.

Documentation Generator

The directory docgen contains a program which creates a PDF or markdown file from the register database (docgen/docgen.py). Use it to create PDF documentation like this (which will place the output files in .):

python docgen/docgen.py --latex vcdb .
pdflatex vcregs.tex

In order to run pdflatex to generate the PDF, you need the bytefield package from http://www.ctan.org/pkg/bytefield!

If you want to generate markdown, call the program like this:

python docgen/docgen.py --md vcdb .

Header Generator

The directory docgen also contains a program which creates a C header file containing all documented registers and register groups. To generate the file, call it like this:

python docgen/headergen.py vcdb .

The last parameter again specifies the directory in which the header file (called vcregs.h) is placed.

If the -a option is passed headergen will also produce a file useable by the as assembler.

Register Trace Decoder

The directory trace_decoder contains a program which processes register traces produced by resim and replaces addresses/values by their names if they are documented in the register database.

To compile the program, simply execute compile.sh from within the trace_decoder directory. To convert a trace, pipe it into the program, then the converted data is printed on stdout:

./trace_decoder < vc4emul.log > decoded.log

About

Some tools usable for reengineering the VideoCore IV processor found on the Raspberry Pi.


Languages

Language:C 50.6%Language:Assembly 23.6%Language:Python 23.0%Language:Makefile 1.7%Language:C++ 1.1%