Appendix E Contributors to the GNU C Library

The GNU C library was written originally by Roland McGrath, and is currently maintained by Ulrich Drepper. Some parts of the library were contributed or worked on by other people.

• The getopt function and related code was written by Richard Stallman, David J. MacKenzie, and Roland McGrath.
• The merge sort function qsort was written by Michael J. Haertel.
• The quick sort function used as a fallback by qsort was written by Douglas C. Schmidt.
• The memory allocation functions malloc, realloc and free and related code were written by Michael J. Haertel, Wolfram Gloger, and Doug Lea.
• Fast implementations of many of the string functions (memcpy, strlen, etc.) were written by Torbjörn Granlund.
• The tar.h header file was written by David J. MacKenzie.
• The port to the MIPS DECStation running Ultrix 4 (mips-dec-ultrix4) was contributed by Brendan Kehoe and Ian Lance Taylor.
• The DES encryption function crypt and related functions were contributed by Michael Glad.
• The ftw and nftw functions were contributed by Ulrich Drepper.
• The startup code to support SunOS shared libraries was contributed by Tom Quinn.
• The mktime function was contributed by Paul Eggert.
• The port to the Sequent Symmetry running Dynix version 3 (i386-sequent-bsd) was contributed by Jason Merrill.
• The timezone support code is derived from the public-domain timezone package by Arthur David Olson and his many contributors.
• The port to the DEC Alpha running OSF/1 (alpha-dec-osf1) was contributed by Brendan Kehoe, using some code written by Roland McGrath.
• The port to SGI machines running Irix 4 (mips-sgi-irix4) was contributed by Tom Quinn.
• The port of the Mach and Hurd code to the MIPS architecture (mips-anything-gnu) was contributed by Kazumoto Kojima.
• The floating-point printing function used by printf and friends and the floating-point reading function used by scanf, strtod and friends were written by Ulrich Drepper. The multi-precision integer functions used in those functions are taken from GNU MP, which was contributed by Torbjörn Granlund.
• The internationalization support in the library, and the support programs locale and localedef, were written by Ulrich Drepper. Ulrich Drepper adapted the support code for message catalogs (libintl.h, etc.) from the GNU gettext package, which he also wrote. He also contributed the catgets support and the entire suite of multi-byte and wide-character support functions (wctype.h, wchar.h, etc.).
• The implementations of the nsswitch.conf mechanism and the files and DNS backends for it were designed and written by Ulrich Drepper and Roland McGrath, based on a backend interface defined by Peter Eriksson.
• The port to Linux i386/ELF (i386-anything-linux) was contributed by Ulrich Drepper, based in large part on work done in Hongjiu Lu's Linux version of the GNU C Library.
• The port to Linux/m68k (m68k-anything-linux) was contributed by Andreas Schwab.
• The ports to Linux/ARM (arm-ANYTHING-linuxaout) and ARM standalone (arm-ANYTHING-none), as well as parts of the IPv6 support code, were contributed by Philip Blundell.
• Richard Henderson contributed the ELF dynamic linking code and other support for the Alpha processor.
• David Mosberger-Tang contributed the port to Linux/Alpha (alpha-anything-linux).
• The port to Linux on PowerPC (powerpc-anything-linux) was contributed by Geoffrey Keating.
• Miles Bader wrote the argp argument-parsing package, and the argz/envz interfaces.
• Stephen R. van den Berg contributed a highly-optimized strstr function.
• Ulrich Drepper contributed the hsearch and drand48 families of functions; reentrant ‘..._r’ versions of the random family; System V shared memory and IPC support code; and several highly-optimized string functions for ix86 processors.
• The math functions are taken from fdlibm-5.1 by Sun Microsystems, as modified by J.T. Conklin, Ian Lance Taylor, Ulrich Drepper, Andreas Schwab, and Roland McGrath.
• The libio library used to implement stdio functions on some platforms was written by Per Bothner and modified by Ulrich Drepper.
• Eric Youngdale and Ulrich Drepper implemented versioning of objects on the symbol level.
• Thorsten Kukuk provided an implementation for NIS (YP) and NIS+, securelevel 0, 1 and 2.
• Andreas Jaeger provided a test suite for the math library.
• Mark Kettenis implemented the utmpx interface and an utmp daemon.
• Ulrich Drepper added character conversion functions (iconv).
• Thorsten Kukuk provided an implementation for a caching daemon for NSS (nscd).
• Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
• Mark Kettenis provided a Hesiod NSS module.
• The Internet-related code (most of the inet subdirectory) and several other miscellaneous functions and header files have been included from 4.4 BSD with little or no modification. The copying permission notice for this code can be found in the file LICENSES in the source distribution.
• The random number generation functions random, srandom, setstate and initstate, which are also the basis for the rand and srand functions, were written by Earl T. Cohen for the University of California at Berkeley and are copyrighted by the Regents of the University of California. They have undergone minor changes to fit into the GNU C library and to fit the ISO C standard, but the functional code is Berkeley's.
• The DNS resolver code is taken directly from BIND 4.9.5, which includes copyrighted code from UC Berkeley and from Digital Equipment Corporation. See the file LICENSES for the text of the DEC license.
• The code to support Sun RPC is taken verbatim from Sun's rpcsrc-4.0 distribution; see the file LICENSES for the text of the license.
• Some of the support code for Mach is taken from Mach 3.0 by CMU; the file if_ppp.h is also copyright by CMU, but under a different license; see the file LICENSES for the text of the licenses.
• Many of the IA64 math functions are taken from a collection of “Highly Optimized Mathematical Functions for Itanium” that Intel makes available under a free license; see the file LICENSES for details.
• The getaddrinfo and getnameinfo functions and supporting code were written by Craig Metz; see the file LICENSES for details on their licensing.
• Many of the IEEE 64-bit double precision math functions (in the sysdeps/ieee754/dbl-64 subdirectory) come from the IBM Accurate Mathematical Library, contributed by IBM.