How I developed an Am29000 C compiler and web browser

How I developed an Am29000 C compiler and web browser(nanochess.org)

92 points by nanochess 38 days ago | 22 comments

jacquesm 37 days ago |

Fascinating. DJ Delorie rates a mention, I wonder how he's doing, I owe him a good chunk of my career.

This reads very much like the same journey I made recently when reviving my old OS, only Oscar went much further than I did and wrote a working browser as well, which - even though the browsers back then were much simpler - is extremely impressive. I still have a stack of disks from those days, I should probably find a way to read them to see what's on there.

edit: I sent DJ Delorie a short note, I should have done that years ago, without him I would have never managed.

nazgulsenpai 37 days ago | |

I remember being probably 13-14 in my bedroom late at night with nothing but the glow of a CRT while poking around with Allegro and DJGPP. I could boot into Mandrake Linux and compile the same code I wrote in Windows. It felt like anything was possible! It is one of those core memories that sticks with me.

I love the website still looks exactly like I remember it (even though I'm sure it isn't) https://www.delorie.com/djgpp/

nsxwolf 37 days ago | |

DJGPP was a revelation. Suddenly I had real build tools, for free, and I was linking to Glide libraries and thinking I was going to be the next John Carmack.

jll29 37 days ago |

Well done. I agree with the credits to DJ Delorie (and to Stallman, for the original GCC!).

The code of the O.P. is actually not so bad for a teenager; however, the variables in Spanish do not help: "[isq]" and "[der]" probably stood for "isqierda" and "derecha", so the index of the left and right subtree of an expression.

Speaking of Transputers, Dr. Tim King, founder of Perihelion, was another big name of the Transputer times; he sadly passed away last month (he also wrote Tripos, which turned into AmigaDOS).

jz391 36 days ago | |

That is sad to hear. I had the pleasure of meeting him once, working with Perihelion as a student. Perihelion were a smart bunch and Helios (quasi UNIX distributed OS for Transputers) was incredibly usable and powerful for its time and hardware (Transputers lacked MMU).

dboreham 37 days ago |

This part brought back memories from an even older time:

"Worst, it required two more programs: Flex and Bison."

In the early 80s I had the idea to port a C compiler to a new machine (I forget if this was K&R or pcc, but those were the only two C compilers you could get the source for at the time, legal issues aside...). At some point I had that same realization: uhh...this thing needs two other pretty big things: yacc and lex. The "first catch your fish" of compilers: "first port the parser-generators".

Joker_vD 37 days ago | |

"Can't you just ship the generated .h/.c files with the source distribution?!"

"Those take less space than lex/yacc so of course not".

compiler-guy 37 days ago | | |

GCC at the time was under GPL v2, which includes.

"The source code for a work means the preferred form of the work for making modifications to it."

For lex and yacc, that means the original .y and .l files. You don't have a choice if you want to be compliant with the terms. Also, shipping the .h and .c files in addition also ends up confusing the build systems of the time.

chasil 37 days ago |

How closely related is this CPU to SPARC?

https://en.wikipedia.org/wiki/AMD_Am29000

"Based on the seminal Berkeley RISC, the 29k added a number of significant improvements... One design element used in some of the Berkeley RISC-derived designs is the concept of register windows, a technique used to speed up procedure calls significantly."

Taniwha 37 days ago | |

They're from roughly the same era (and the same as 68020s/396s/88ks/mips) - 29ks came out of AMD and sadly didn't fly as high as we hoped at the time - their unique stack management may have been part of that

p_l 36 days ago | | |

I'd say the insane bus arrangements on released chips (unrelated to ISA itself) had more to say there - the 3 & 4 bus configurations only make sense if you're throwing a bunch of extra chips onto it to make a proper CPU module that will expose sensible single bus to the world. Instead you effectively got a harvard architecture design in way that is least helpful.

p_l 36 days ago | |

Both are based on RISC-2 design, yes.

dmitrygr 37 days ago |

Oscar Toledo, you are simply insane (in a good day). I bow my head in respect at the completely pointless and insanely impressive result.

anthk 37 days ago |

>Ciega, sordomuda", "Amor de papel", “Laura no esta” and “Barbie girl”.

Oh poor soul...

https://github.com/nanochess/Am29000

This is cool, it's very Motif like.

iberator 37 days ago |

Amazing and super detailed.

Worth to read

kreelman 37 days ago | |

Amazing, you can play chess on an emulated system built in the '90s from technological "mud and straw".

It's an amazing feat of teenage engineering.

I hope you have a job working with code somewhere!

Gracias!

c-parse.o : $(srcdir)/c-parse.c $(CONFIG_H) $(TREE_H) c-lex.h $(GGC_H) intl.h \ $(C_TREE_H) input.h flags.h system.h toplev.h output.h cpplib.h $(CC) $(ALL_CFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) -c $(srcdir)/c-parse.c $(srcdir)/c-parse.c: $(srcdir)/c-parse.y (cd $(srcdir) && $(BISON) $(BISONFLAGS) -o c-p$$$$.c c-parse.y && \ mv -f c-p$$$$.c c-parse.c) $(srcdir)/c-parse.y: c-parse.in echo '/*WARNING: This file is automatically generated!*/' >tmp-c-parse.y sed -e "/^ifobjc$$/,/^end ifobjc$$/d" \ -e "/^ifc$$/d" -e "/^end ifc$$/d" \ $(srcdir)/c-parse.in >>tmp-c-parse.y $(SHELL) $(srcdir)/move-if-change tmp-c-parse.y $(srcdir)/c-parse.y $(srcdir)/tradcif.c: $(srcdir)/tradcif.y (cd $(srcdir) && $(BISON) $(BISONFLAGS) -o tr$$$$.c tradcif.y && \ mv -f tr$$$$.c tradcif.c)

srcdir=. top_srcdir=../.. cgram.c: $(srcdir)/cgram.y $(YACC) $(YFLAGS) -d $(srcdir)/cgram.y mv -f y.tab.c cgram.c mv -f y.tab.h cgram.h scan.c: $(srcdir)/scan.l $(LEX) $(LFLAGS) $(srcdir)/scan.l mv -f $(LEX_OUTPUT_ROOT).c scan.c $(OBJS): $(HDRS) external.c cgram.c cgram.o: cgram.c $(CC) $(CF0) $(CFLAGS) $(CPPFLAGS) -c -o $@ cgram.c scan.o: scan.c $(CC) $(CF0) $(CFLAGS) $(CPPFLAGS) -c -o $@ scan.c