Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Tiny-C Compiler (2001)

iro.umontreal.ca

Post reply on HN
Print view View original

61–70 of 82 posts

Re: Tiny-C Compiler (2001)

#61

Post by Mike_12345 »

masklinn wrote: post #48
Mike_12345 wrote: post #44

Earlier quoted context omitted.

That is definitely a compiler and anyone with a CS degree would call it that if they were discussing its functionality, because that's technically what it is. (Referring specifically to the part which compiles Python to bytecode) Your SQL database also has a compiler. SQL is compiled to an execution plan. Compile doesn't only mean "create a machine code executable file".

> That is definitely a compiler and anyone with a CS degree would call it that if they were discussing its functionality because that's technically what it is. None of these assertions is correct. > (Referring specifically to the part which compiles Python to bytecode) So referring specifically to something different than what I explicitly specified, it's called something else. By that reasoning, a cow is a muscle an…

> None of these assertions is correct.

You should fix the Wikipedia article:

https://en.wikipedia.org/wiki/CPython

"CPython can be defined as both an interpreter and a compiler as it compiles Python code into bytecode before interpreting it."

Re: Tiny-C Compiler (2001)

#62

Post by WoodenChair »

benj111 wrote: post #38
WoodenChair wrote: post #5

This is an interpreter for a super restricted subset of C and it looks well written from a pedagogical standpoint (keeps thing pretty simple, fairly easy to read). But it's slightly awk to strip-down a language (what features do you keep, what do you lose?). I think it's more fun to build an interpreter for an actual tiny language. In my next book I have interpreters for Brainfuck [0], an obfuscated kind of joke of a…

While I appreciate your point. 1. You use the example of a tiny basic of a 'real language' and I don't see how tiny basic is a 'real language', but tiny C is a stripped down language. 2. You can build on this to make a full c implementation. A minimal c implementation that can potentially bootstrap a full c environment is more useful than a brainfuck interpreter.

1. I think I explained why. You can Google real programs people used to do their day to day computing tasks and games and run them unmodified. You can’t do that with this dialect of C. It’s cool to run real programs.

2. Good point, but it’s several orders of magnitude more work to go from this to C, so it depends if you’re doing a one off project or have much bigger ambitions. I think you get more bang for the buck out of Tiny BASIC if it’s a one-off project is my point.

Re: Tiny-C Compiler (2001)

#63

Post by userbinator »

masklinn wrote: post #13
userbinator wrote: post #10

Earlier quoted context omitted.

It is a compiler rather than a direct evaluator, since it generates bytecode for a stack VM --- and also includes the interpreter for that (look at the bottom).

That’s more or less every interpreter. CPython compiles to bytecode before interpreting that, yet nobody would call it a compiler.

I think this is a question of interface vs. implementation.

Python, JavaScript, and other languages which are traditionally considered interpreted but may do (JIT) compilation in their implementation are used as if they were interpreters: to the user, there's no separate compilation step. You run python somefile.py or node somefile.js (or refresh a browser holding a page), and editing the source code causes the next invocation to take those changes immediately. Contrast this with C/C++ and Java where there is definitely an explicit compilation step in nearly all implementations.

The program in this article thus is an implementation of a compiler, but has the interface of an interpreter.

Re: Tiny-C Compiler (2001)

#64

Post by sitkack »

feeley wrote: post #40

Author here. Just for context tinyc.c was created in 2000 (I found the file in my archives and the last modification date is January 12, 2001). I was not aware at the time of Fabrice Bellard's work which after all won the IOCCC in 2001, so the confusion with TCC was not intentional. My tinyc.c was meant to teach the basics of compilers in a relatively accessible way, from parsing to AST to code generation to bytecode…

I want to thank you for this pedagogical tool, I have used a couple times to learn a new language by porting this exercise.

How did you use this in your teaching? It seems like it could the basis for a longer term project that students could take in many directions.

Re: Tiny-C Compiler (2001)

#66

Post by t-3 »

circuit10 wrote: post #52
Laaas wrote: post #17

Earlier quoted context omitted.

I use tcc for all of my small C "scripts" for doing ioctls, etc. Less bloat, suckless. I imagine most software would be better off using tcc than gcc/clang. Performance isn't that important in most cases.

I feel like a lot of software written in C is written in C for performance reasons. Obviously that’s not always the case and TCC is useful but I wouldn’t say that that most software should use it

C is usually the path of least resistance on *nix, if you're trying to interact with the OS. The docs give you C prototypes and C examples, and the source is mostly C, several standard directories are filled with C code and libraries compiled from C.

Re: Tiny-C Compiler (2001)

#67

Post by vidarh »

LukeShu wrote: post #51
vidarh wrote: post #20

Earlier quoted context omitted.

Since they say "scripts", note that tcc supports being invoked in the shebang line. E.g. #!/usr/bin/tcc -run You can do that with gcc/clang too (e.g. #if 0, #endif to wrap a block of shell script to compile the current file and execute the result) but a primary value of tcc is that it compiles fast . On a more philosophical note, the suckless approach is to optimise for simplicity not storage. It's perfectly valid to…

You can only sort of do that with gcc/clang. The #if 0 trick relies on funny behavior that is in a few common shells. When you try to execve(2) a script without a proper #! shebang, the kernel will return ENOEXEC. Bash will check for ENOEXEC then check a few heuristics to see if it looks like a text file, and if it does, then it will try to run it as a shell script. This means that your script will work when run from…

That makes sense. When I checked the tcc command line I was mildly surprised to see you could do the #if 0 hack at all, so not surprised to hear it has severe limitations.

Re: Tiny-C Compiler (2001)

#68

Post by feeley »

sitkack wrote: post #64
feeley wrote: post #40

Author here. Just for context tinyc.c was created in 2000 (I found the file in my archives and the last modification date is January 12, 2001). I was not aware at the time of Fabrice Bellard's work which after all won the IOCCC in 2001, so the confusion with TCC was not intentional. My tinyc.c was meant to teach the basics of compilers in a relatively accessible way, from parsing to AST to code generation to bytecode…

I want to thank you for this pedagogical tool, I have used a couple times to learn a new language by porting this exercise. How did you use this in your teaching? It seems like it could the basis for a longer term project that students could take in many directions.

tinyc.c was designed to illustrate three things at once in a second year course on concepts of programming languages. The course had 4 parts: imperative programming (using C), functional programming (using Scheme), and logic programming (using Prolog), and programming language implementation. For the imperative programming part it was important to show enough of the C language for the following operating systems course, so we needed to show C manual memory management and pointers in a rather detailed way. So tinyc.c was principally an example of programming with pointers, including pointer arithmetic such as *pc++. It was indirectly an example of compiler and interpreter, subjects we also covered in the course. I have also used parts of the compiler in a third year compiler course but not as the basis of a project. I have always asked (forced?) my students to use Scheme to implement their compiler projects... a much friendlier language than C for compiler writing.

Re: Tiny-C Compiler (2001)

#69

Post by feeley »

stevefan1999 wrote: post #58
feeley wrote: post #40

Author here. Just for context tinyc.c was created in 2000 (I found the file in my archives and the last modification date is January 12, 2001). I was not aware at the time of Fabrice Bellard's work which after all won the IOCCC in 2001, so the confusion with TCC was not intentional. My tinyc.c was meant to teach the basics of compilers in a relatively accessible way, from parsing to AST to code generation to bytecode…

Would you think rewriting this to generate a minimal set of instructions could be benefitial to the compiler bootstrapping?

I'm not sure what you mean by "minimal set of instructions". What is more important for bootstrapping is to restrict the programming style used to write the compiler. So if that was the goal I would implement pointers and maybe arrays, and not implement other types such as structs (which can be useful but add complexity to the compiler, i.e. it is no longer the case that a value fits in a machine word). Functions and function call would obviously be useful to add. But that's about it.

What I'd love to do someday is to write a compiler for a fairly complete C subset in POSIX shell. The goal would be to use only a POSIX shell and this compiler to compile TCC, and then use TCC to bootstrap gcc, all from source. This would be a great tool for reproducible builds from source. If someone here finds this interesting and would like to help out, please reach out to me.

Post reply on HN