4.3BSD

The History of the Berkeley Software Distribution begins in the 1970s.

1BSD (PDP-11)

The earliest distributions of Unix from Bell Labs in the 1970s included the source code to the operating system, allowing researchers at universities to modify and extend Unix. The operating system arrived at Berkeley in 1974, at the request of computer science professor Bob Fabry who had been on the program committee for the Symposium on Operating Systems Principles where Unix was first presented. A PDP-11/45 was bought to run the system, but for budgetary reasons, this machine was shared with the mathematics and statistics groups at Berkeley, who used RSTS, so that Unix only ran on the machine eight hours per day (sometimes during the day, sometimes during the night). A larger PDP-11/70 was installed at Berkeley the following year, using money from the Ingres database project.

Also in 1975, Ken Thompson took a sabbatical from Bell Labs and came to Berkeley as a visiting professor. He helped to install Version 6 Unix and started working on a Pascal implementation for the system. Graduate students Chuck Haley and Bill Joy improved Thompson's Pascal and implemented an improved text editor, ex. Other universities became interested in the software at Berkeley, and so in 1977 Joy started compiling the first Berkeley Software Distribution (1BSD), which was released on March 9, 1978. 1BSD was an add-on to Version 6 Unix rather than a complete operating system in its own right. Some thirty copies were sent out.

2BSD (PDP-11)

The Second Berkeley Software Distribution (2BSD), released in May 1979, included updated versions of the 1BSD software as well as two new programs by Joy that persist on Unix systems to this day: the vi text editor (a visual version of ex) and the C shell. Some 75 copies of 2BSD were sent out by Bill Joy. A further feature was a networking package called Berknet, developed by Eric Schmidt as part of his master's thesis work, that could connect up to twenty-six computers and provided email and file transfer.

After 3BSD (see below) had come out for the VAX line of computers, new releases of 2BSD for the PDP-11 were still issued and distributed through USENIX; for example, 1982's 2.8.1BSD included a collection of fixes for performance problems in Version 7 Unix, and later releases contained ports of changes from the VAX-based releases of BSD back to the PDP-11 architecture. 2.9BSD from 1983 included code from 4.1cBSD, and was the first release that was a full OS (a modified V7 Unix) rather than a set of applications and patches.

The most recent release, 2.11BSD, was first issued in 1991. In the 21st Century, maintenance updates from volunteers continued: patch #473 was released on January 25, 2022.

3BSD

A VAX computer was installed at Berkeley in 1978, but the port of Unix to the VAX architecture, UNIX/32V, did not take advantage of the VAX's virtual memory capabilities. The kernel of 32V was largely rewritten by Berkeley graduate student Özalp Babaoğlu to include a virtual memory implementation, and a complete operating system including the new kernel, ports of the 2BSD utilities to the VAX, and the utilities from 32V was released as 3BSD at the end of 1979. 3BSD was also alternatively called Virtual VAX/UNIX or VMUNIX (for Virtual Memory Unix), and BSD kernel images were normally called /vmunix until 4.4BSD.

The success of 3BSD was a major factor in the Defense Advanced Research Projects Agency's (DARPA) decision to fund Berkeley's Computer Systems Research Group (CSRG), which would develop a standard Unix platform for future DARPA research in the VLSI Project.

4BSD

4BSD (November 1980) offered a number of enhancements over 3BSD, notably job control in the previously released csh, delivermail (the antecedent of sendmail), "reliable" signals, and the Curses programming library. In a 1985 review of BSD releases, John Quarterman ''et al.'', wrote:

4.1BSD

4.1BSD (June 1981) was a response to criticisms of BSD's performance relative to the dominant VAX operating system, VMS. The 4.1BSD kernel was systematically tuned up by Bill Joy until it could perform as well as VMS on several benchmarks. The release would have been called ''5BSD'', but after objections from AT&T the name was changed; AT&T feared confusion with AT&T's UNIX System V.

4.2BSD

4.2BSD (August 1983) would take over two years to implement and contained several major overhauls. Before its official release came three intermediate versions: 4.1a from April 1982 incorporated a modified version of BBN's preliminary TCP/IP implementation; 4.1b from June 1982 included the new Berkeley Fast File System, implemented by Marshall Kirk McKusick; and 4.1c in April 1983 was an interim release during the last few months of 4.2BSD's development. Back at Bell Labs, 4.1cBSD became the basis of the 8th Edition of Research Unix, and a commercially supported version was available from mtXinu.

To guide the design of 4.2BSD, Duane Adams of DARPA formed a "steering committee" consisting of Bob Fabry, Bill Joy and Sam Leffler from UCB, Alan Nemeth and Rob Gurwitz from BBN, Dennis Ritchie from Bell Labs, Keith Lantz from Stanford, Rick Rashid from Carnegie Mellon, Bert Halstead from MIT, Dan Lynch from ISI, and Gerald J. Popek of UCLA. The committee met from April 1981 to June 1983.

Apart from the Fast File System, several features from outside contributors were accepted, including disk quotas and job control. Sun Microsystems provided testing on its Motorola 68000 machines prior to release, ensuring portability of the system.

The official 4.2BSD release came in August 1983. It was notable as the first version released after the 1982 departure of Bill Joy to co-found Sun Microsystems; Mike Karels and Marshall Kirk McKusick took on leadership roles within the project from that point forward. On a lighter note, it also marked the debut of BSD's daemon mascot in a drawing by John Lasseter that appeared on the cover of the printed manuals distributed by USENIX.

4.3BSD

4.3BSD was released in June 1986. Its main changes were to improve the performance of many of the new contributions of 4.2BSD that had not been as heavily tuned as the 4.1BSD code. Prior to the release, BSD's implementation of TCP/IP had diverged considerably from BBN's official implementation. After several months of testing, DARPA determined that the 4.2BSD version was superior and would remain in 4.3BSD. (See also History of the Internet.)

After 4.3BSD, it was determined that BSD would move away from the aging VAX platform. The Power 6/32 platform (codenamed "Tahoe") developed by Computer Consoles Inc. seemed promising at the time, but was abandoned by its developers shortly thereafter. Nonetheless, the 4.3BSD-Tahoe port (June 1988) proved valuable, as it led to a separation of machine-dependent and machine-independent code in BSD which would improve the system's future portability.

Apart from portability, the CSRG worked on an implementation of the OSI network protocol stack, improvements to the kernel virtual memory system and (with Van Jacobson of LBL) new TCP/IP algorithms to accommodate the growth of the Internet.M.K. McKusick, M.J. Karels, Keith Sklower, Kevin Fall, Marc Teitelbaum and Keith Bostic (1989). Current Research by The Computer Systems Research Group of Berkeley. Proc. European Unix Users Group.

Until then, all versions of BSD incorporated proprietary AT&T Unix code and were, therefore, subject to an AT&T software license. Source code licenses had become very expensive and several outside parties had expressed interest in a separate release of the networking code, which had been developed entirely outside AT&T and would not be subject to the licensing requirement. This led to Networking Release 1 (Net/1), which was made available to non-licensees of AT&T code and was freely redistributable under the terms of the BSD license. It was released in June 1989.

4.3BSD-Reno came in early 1990. It was an interim release during the early development of 4.4BSD, and its use was considered a "gamble", hence the naming after the gambling center of Reno, Nevada. This release was explicitly moving towards POSIX compliance, and, according to some, away from the BSD philosophy (as POSIX is very much based on System V, and Reno was quite bloated compared to previous releases). Among the new features were an NFS implementation from the University of Guelph, a status key ("Ctrl-T") and support for the HP 9000 range of computers, originating in the University of Utah's "HPBSD" port.

In August 2006, ''InformationWeek'' magazine rated 4.3BSD as the "Greatest Software Ever Written". They commented: "BSD 4.3 represents the single biggest theoretical undergirder of the Internet."

Net/2 and legal troubles

After Net/1, BSD developer Keith Bostic proposed that more non-AT&T sections of the BSD system be released under the same license as Net/1. To this end, he started a project to reimplement most of the standard Unix utilities without using the AT&T code. For example, vi, which had been based on the original Unix version of ed, was rewritten as nvi (new vi). Within eighteen months, all of the AT&T utilities had been replaced, and it was determined that only a few AT&T files remained in the kernel. These files were removed, and the result was the June 1991 release of Networking Release 2 (Net/2), a nearly complete operating system that was freely distributable.

Net/2 was the basis for two separate ports of BSD to the Intel 80386