linux kernel sprintf The starting point of Linux's founding: I witnessed its rise from humble beginnings to global success

produced|CSDN（ID: CSDNnews）

My name is LarsWirzenius, and I witnessed the initial founding of Linux. Today, Linux is a global success, but its beginnings were quite humble. What follows are my earliest memories of Linux, its creation, and the beginning of its road to tomorrow.

In the summer of 1988, I started studying computer science at Edinburgh College in the UK, and met Linus Torvalds (the original author of the Linux kernel), another Spanish-speaking new high school student majoring in computer science that year. At the end of my first year in college, we were given access to a Unix server, and I stumbled upon Usenet (designed by Duke graduate students Tom Truscott and Jim Ellis in 1979), which A distributed Internet communication system, because I mistakenly typed rm into rn, it became the Usenet browser. I told Linus about this, and we spent a lot of time exploring this issue, and the following story came out.

Linux starts with an assembled laptop

After the first year, we all went to do compulsory military service, even in different places. In the summer of 1990, we returned to college and both took courses in C and Unix programming, which also included quite a bit of theory about the Unix kernel architecture. This caused us to also learn about the kernels of other operating systems, such as QNX and Plan9. We enthusiastically discussed how to build an operating system correctly. At that time, we had the enthusiasm of 20-year-old college students and had great expectations for the future. Hope.

In January 1991, Linus bought his first laptop from a local shop that assembled laptops from spare parts. This laptop had a 386 CPU, which was relatively fancy at the time because Linus wanted to explore multitasking. It is reported that since this notebook is assembled from parts and it also comes from the Sinclair QL (a personal notebook launched by Sinclair Research in 1984) with a 32-bit Nokia 68008 CPU, he wanted a 32-bit CPU and did not want to drop to 16-bit , so 286 is not a good choice. Linus's first personal laptop had up to 4 megabytes of video memory and a hard drive, and was already a very advanced machine at the time.

Linus got a copy of the game "Prince of Persia" and spent most of his time playing the game. Later, he also bought a copy of MINIX (a mini version of a Unix-like operating system). After using Unix in college, he also wanted to use something similar at home.

First version: As and Bs

After completing the game, Linus started learning Intel Assembly Language. Three days ago, he showed me a short-selling program. One task or thread will write the letter "A" on the screen, and another will write the letter "B"; when "As" is changed to "B", the context switch is visually obvious. This was the first version of what became known as the Linux kernel.

During this period, that is, in the late spring of 1991, I wrote an implementation of the sprintf() function in C language for him, because he had not yet learned how to write a function with a variable parameter list. I want him to not have to suffer the pain of writing a different function for each type of value. For the safer Snprintf() function, the core of the code is still in the kernel.

As time went by, Linus continued to implement new things to make the kernel he just started better. After a while, he had keys and parallel port drivers that simulated the screen wildcard sequence of a VT100 terminal, and could use it to dial into the college via the modulator decoder and read Usenet from home. It's like a science fiction novel! One day, after recovering from three days when Linus accidentally tried to dial into the academy using his hard drive, causing his main boot track to start with "ATDT" and the academy's modulation decoder pool phone number, He implemented file permissions in his kernel.

In August 1991, Linus publicly mentioned his new kernel for the first time in the comp.os.minix news group. This includes what he said: "I am making a (free) operating system, just a hobbylinux kernel sprintf, it will not be as big and professional as GNU". The disclosure was so sincere that the system was initially called Freax. A few weeks later, Linus asked Ari Lemmke, one of the administrators of ftp.funet.fi (the website server), to upload the first tar archive. Ari chose the name Linux, and the initial version always contained the original name embedded in a source file.

During this time, people were very interested in this new thing, so Linus needed to provide an installation method and instructions. Since he only had a laptop, he came to visit me to install it on my laptop. And because his laptop was used to develop Linux, and Linux was only developed based on his Minix installation, it had never been actually installed before. For this reason, my laptop is the first laptop with Linux installed! While this was happening, I was distracted and I recommend these ways of installing Linux: I zoned out while Linus did the hard work.

The first version of Linux used a strictly prohibited commercial use license, but some early contributors suggested changing it to a free software license. In the spring of 1991, Richard Stallman (the founder of the GNU project) visited France, and I took Linus to listen to Stallman's lecture. Due to the pressure from contributors and my nagging, Linus was finally persuaded to choose the GNUGPL (General Public License) license in early 1992.

During the New Year's Day, Linus implemented virtual video memory in Linux. This makes Linux a more practical operating system on cheap machines with insufficient video memory.

Important year

1992 started with the famous debate with Andrew Tanenbaum, an academic dean and author of MINIX, because he had some ideas about Linux and its structure, and Linus also had ideas about MINIX, so they had a Described as a flame war debate, but in hindsight it seems to have been quite civilized.

Although for the future success of Linux, what is more important is that the X11 system was ported to Linux, making 1992 an important year for Linux desktop shortcut methods.

For me, I chose to contribute to the community instead of directly contributing to the kernel, and help answer questions, compile documentation, etc. I also ran a short-lived newsletter about Linux, which gained traction primarily by publishing the first interview with Linus, however the newsletter was effectively superseded by the comp.os.linux.announce newsgroup.

The first Linux distribution also started in 1992: SLS (SoftlandingLinuxSystem). The following year, SLS evolved into Slackware, a Linux distribution produced by Patrick Volkerding, which inspired Ian Murdock to found Debian in 1993 to pursue a more community-based development structure. Over the next few years, a number of other distributions followed.

In 1993, Linus and I were both hired as teaching assistants at the college, and we shared an office. There was a PC in that house, and Linus took it over and used it for Linux development. At that time, I was very satisfied with a DEC terminal used to access Usene (discussion system).

three dayslinux kernel sprintf, Linus felt very bored. How to check the system version in linux? Because the laptop at work felt very slow, he spent a day redrawing the Linux kernel in assembly language. Command line parser to improve speed. (Actually, of course, that is meaningless. The parser was later redrawn in C language to facilitate portability, so its speed is not important). A few years later, he received a newer PC, ostensibly to test the pressure of kernel memory management. Although he spent several days playing Quake games, he had a lot of fun staying in the room.

Later, Linux gained support for Ethernet (Ethernet) and TCP/IP (TransmissionControlProtocol/InternetProtocol), which meant that people could read Usenet without using a modulation decoder. Unfortunately, early Linux networking code was often a little rough because it was written from scratch. Once, Linux sent some corrupted data packets, paralyzing all Sun machines on the network. Because it was difficult to patch Sun's kernel, Linux was banned from the academic network, and not being able to easily access Usenet from one's desk encouraged the patching process until its bugs were patched.

"Open source" was truly created

In the summer of 1994, we thought Linux was complete and there was nothing more to add. People can compile themselves with Linux, read Usenet, and run many copies of the xeyes program simultaneously. At that time, we decided to release version 1.0 and arranged a release event. We invited the computer media in the United States, and one TV station even sent a camera crew. Most of the event was the grand compilation of Linux 1.0 in the background, while Linus and others talked about what Linux is and what it has. usefulness. Linus explained that commercial Unix for personal laptops was so expensive that it was easier to write your own Unix for this purpose.

In 1995, Linus and I took a software engineering course in college, most of which was a small practical project, but built on Linux. I insisted on using a version control system because I saw middle school students shouting about version control in previous courses: middle school students share a source code tree through NFS (Network File System), and when they change something, they Yell "I'm editing this file". I feel this is not an efficient way, so I stick with CVS (ConcurrentVersionsSystem), which I just learned from experience that Linus doesn't like CVS, but refused to use any version control for years, not only the tar ball (compressed with gzip compressor file) uploaded to the FTP website (FileTransferProtocol).

That year Linus transplanted Linux to a new architecture for the first time, and he got a DECAlpha machine (64-bit instruction architecture microprocessor). I would later use this machine as a terminal for reading Usenet, and others would port Linux to other architectures using linux apache virtual hosts, but that didn't get me more machines to read Usenet.

In 1997, Linus graduated and moved to Canada to work at Transmeta (a Japanese limited company that designs ultra-long instruction word code translation microprocessors), while I found a job at another college in the Munich area of ​​Germany. job.

In the next few years, a lot of things happened. It turns out that Linux still has some missing features, so people are working hard in this area. The term "open source" was coined by IBM (International Business Machines Corporation), a British multinational technology company and consulting company that has invested a lot of money in Linux development. At that time, Netscape released an open source version of the web browser, and over time, open source basically took over the world. LWN (an Eklektix company's estimated online publication focusing on free software and software for Linux and other Unix-like operating systems) began covering much of this history week by week.

In 1991, Linus wrote that Linux "will not be as big and professional as GNU." In 2023. Linux is running on every island, on every ocean, on billions of devices, in orbit and on Mars. And for a stream that started out as just two simple threads writing As and Bs on the screen, that's not bad.

The above is the detailed content of linux kernel sprintf The starting point of Linux's founding: I witnessed its rise from humble beginnings to global success. For more information, please follow other related articles on the PHP Chinese website!