linux — Linux source inclusions (v3.16)

Description

A directed network of the inclusions between the source code files of the Linux kernel, version 3.16 (released in August 2014), extracted from the kernel's source tarball following the method of J. Kunegis. Each directed edge points from a C source or header file to a file it includes with an #include directive (repeated inclusions are counted once). The included files were located heuristically: an #include was resolved by looking for the file relative to the directory of the including file and its parent directories (and their include subdirectories), then relative to the top directory, then by a unique match of the file name anywhere in the tree; system headers (e.g. sys/... or endian.h) and the generated linux/version.h were ignored, as were includes that could not be resolved. The paths were then simplified by removing architecture-, platform- and API-specific directory components (e.g. arch/<arch>/, include/, asm/, uapi/, linux/, mach-*/), so that several files with the same simplified path, such as the variants of the same header for different architectures, are merged into one node. Only files that include or are included by some file are part of the network. Since a file and a file it includes can be mapped to the same node, 530 nodes have self-loops.

The network has the following properties:

(Node) name: the simplified path of the file(s), which identifies the node.
(Node) files: the list of the paths, relative to the top directory of the source tree, of the source files whose simplified path is that of the node (empty only for config.h, which is generated during the build).

Tags

Technological Software Unweighted Metadata

Added

2020-08-11

Citation

J. Kunegis, "Linux." KONECT, the Koblenz Network Collection (2016) 10.1145/2487788.2488173 Sci-Hub

Upstream URL OK

https://www.kernel.org/

Upstream license

GPL-2.0, https://www.kernel.org/pub/linux/kernel/COPYING

Networks

Tip: hover over a column header for a description.

Column legend

$\left<k\right>$: Mean degree
$\sigma_k$: Standard deviation of the degrees
$\lambda_h$: Largest eigenvalue of the Hashimoto (non-backtracking) matrix
$\tau$: Random walk mixing time
$r$: Degree assortativity
$c$: Global clustering coefficient
$\oslash$: (Pseudo-) diameter
$S$: Relative size of the largest component
NPs: Node properties
EPs: Edge properties

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
linux	30,837	213,954	6.94	107.79	155.62	67.60	-0.17	0.00	12	1.00	Directed	Unipartite			674 KiB	1.3 MiB	1.3 MiB	1022 KiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.