fdt 0.3.0

Flattened Device Tree Python Module

Flattened Device Tree Python Module

This python module is usable for manipulation with Device Tree Data and primary was created for imxsb tool

Some parts in this module have been inspired from: https://github.com/superna9999/pyfdt project.

Installation

pip install fdt

To install the latest version from master branch execute in shell following command:

pip install -U https://github.com/molejar/pyFDT/archive/master.zip

In case of development, install it from cloned sources:

git clone https://github.com/molejar/pyFDT.git
cd pyFDT
pip install -U -e .

NOTE: You may run into a permissions issues running these commands. Here are a few options how to fix it:

1. Run with sudo to install fdt and dependencies globally
2. Specify the --user option to install locally into your home directory (export "~/.local/bin" into PATH variable if haven't).
3. Run the command in a virtualenv local to a specific project working set.

Usage

fdt module has intuitive and self describing API, what is presented in following example. Many of general requirements for manipulation with FDT Nodes, Properties and dts/dtb files are already implemented.

  import fdt

  #-----------------------------------------------
  # convert *.dtb to *.dts
  # ----------------------------------------------
  with open("example.dtb", "rb") as f:
      dtb_data = f.read()

  dt1 = fdt.parse_dtb(dtb_data)

  with open("example.dts", "w") as f:
      f.write(dt1.to_dts())

  #-----------------------------------------------
  # convert *.dts to *.dtb
  # ----------------------------------------------
  with open("example.dts", "r") as f:
      dts_text = f.read()

  dt2 = fdt.parse_dts(dts_text)

  with open("example.dtb", "wb") as f:
      f.write(dt2.to_dtb(version=17))

  #-----------------------------------------------
  # add new Node into dt2
  # ----------------------------------------------
  # create node instance
  node = fdt.Node('test_node1')

  # add properties
  node.append(fdt.Property('basic_property'))
  node.append(fdt.PropStrings('string_property', 'value1', 'value2'))
  node.append(fdt.PropWords('words_property', 0x80000000))
  node.append(fdt.PropBytes('bytes_property', 0x00, 0x01, 0x02))

  # PropBytes constructor take also complex data object as bytes() or bytearray()
  node.append(fdt.PropBytes('bytes_property2', data=b"\x00\x01\x02"))

  # add created node into root path of dt2
  dt2.add_item(node)

  # use set_property method to update or create new property
  dt2.set_property('words_property', [0, 1], path='/test_node1')
  dt2.set_property('bytes_property', b"\x00\x01", path='/test_node1')
  dt2.set_property('string_property', ['value1', 'value2', 'value3'], path='/test_node1')  

  # use search method for find all string properties and then update it
  items = dt2.search("", itype=fdt.ItemType.PROP_STRINGS, path="/test_node1")
  for item in items:
    item.data = ['value1', 'value2']

  #-----------------------------------------------
  # merge dt2 into dt1
  # ----------------------------------------------
  dt1.merge(dt2)

  with open("merged.dtb", "wb") as f:
      f.write(dt1.to_dtb(version=17))

  #-----------------------------------------------
  # diff two fdt objects
  # ----------------------------------------------
  out = fdt.diff(dt1, dt2)

  print(out[0]) # same in dt1 and dt2
  print(out[1]) # specific for dt1
  print(out[2]) # specific for dt2

[ pydtc ] Tool

The python device tree converter pydtc is a tool for conversion *.dts to *.dtb and vice versa. Is distributed together with fdt module. This tool can be in some cases used as replacement of device tree compiler.

  $ pydtc -h

usage: pydtc [-h] [-v] {pack,unpack,merge,diff} ...

Flat Device Tree (FDT) tool for manipulation with *.dtb and *.dts files

positional arguments:
  {pack,unpack,merge,diff}
    pack                Pack *.dts into binary blob (*.dtb)
    unpack              Unpack *.dtb into readable format (*.dts)
    merge               Merge more files in *.dtb or *.dts format
    diff                Compare two files in *.dtb or *.dts format

optional arguments:
  -h, --help            show this help message and exit
  -v, --version         show program's version number and exit

$ pydtc unpack [-h] [-s TAB_SIZE] [-o DTS_FILE] dtb_file

Unpack Device Tree from binary blob *.dtb into readable text file *.dts

dtb_file - Single DTB file with dtb extension

optional arguments:

• -h, --help - Show this help message and exit
• -s TAB_SIZE - Tabulator Size
• -o DTS_FILE - Output path/file name (*.dts)

Example:

pydtc unpack test.dtb

DTS saved as: test.dts

$ pydtc pack [-h] [-v VERSION] [-l LC_VERSION] [-c CPU_ID] [-p] [-o DTB_FILE] dts_file

Pack Device Tree from readable text file *.dts into binary blob *.dtb

dts_file - Single DTS file as *.dts

optional arguments:

• -h, --help - Show this help message and exit
• -v VERSION - DTB Version
• -l LC_VERSION - DTB Last Compatible Version
• -c CPU_ID - Boot CPU ID
• -p - Update phandle
• -o DTB_FILE - Output path/file name (*.dtb)

Example:

pydtc pack -v 17 test.dts

DTB saved as: test.dtb

$ pydtc merge [-h] [-t {auto,dts,dtb}] [-s TAB_SIZE] out_file in_files [in_files ...]

Merge two and more *.dtb or *.dts files into one *.dts file

out_file - The output file name with *.dts extension
in_files - Two or more input files with *.dtb or *.dts extension

optional arguments:

• -h, --help - Show this help message and exit
• -t {auto,dts,dtb} - Input file type: 'auto', 'dts', 'dtb' (default: auto)
• -s TAB_SIZE - Tabulator Size

Example:

pydtc merge out.dts test1.dtb test2.dtb

Output saved as: out.dts

$ pydtc diff [-h] [-t {auto,dts,dtb}] [-o OUT_DIR] in_file1 in_file2

Compare two dtb/dts files and generate 3 dts files (same in 1 and 2, specific for 1, specific for 2)

in_file1 - Input file 1
in_file2 - Input file 2

optional arguments:

• -h, --help - Show this help message and exit
• -t {auto,dts,dtb} - Input file type: 'auto', 'dts', 'dtb' (default: auto)
• -o OUT_DIR - Output directory (default: diff_out)

Example:

pydtc diff test1.dtb test2.dtb

Output saved into: diff_out