We too used Lex and Yacc and a c compiler to create full parsers in a single term. Fascinating how much Tools solve so many problems. And BTW this is why they expect great things from LLM AI... it's just exponential growth in the power of the tools.
On Sat, Dec 20, 2025 at 1:43 PM Pete Soper <[email protected]> wrote: > This reminds me of a graduate compiler construction course I took in the > early 70s. A buddy and I wrote our compilers in a high level language. The > others used Univac 1100 assembler. Buddy and I had our compilers going > early and made them way past the specs the prof dictated, polishing them > until they glowed. With 24 starters there were eight survivers of the > course. Then in the last class the prof mentioned he'd blown his budget for > machine time. He looked at my friend and I and we went bright red as the > others sat oblivious. The host language compiler was some painful number of > times more expensive to run than the assembler and linking would have been > cheaper too. I guess this could segue into a chat about bloatware. :-) > Pete > > Dec 20, 2025 1:08:12 PM Mike Lisanke <[email protected]>: > > Pete, There are many courses for an Engineer in Computer Systems > science/engineering. And they include HLD and ISA and bit-slice and > architecture. And yes, also compiler and interpreter design and machine > language programming bootstrapping into assembly language and boot loaders. > There are as they say the Whole Nine Yards. Many of us really old timers > learnt machines from the ground up. I was younger than most when many many > CPUs MPUs were being shoveled onto the mass market for embedded systems And > the newest thing, personal computers. Most had no tools and had many of us > using one system (with BASIC and DOS) to boot assembly languages for other > cpu on bare hardware. I recall as a HS hobbyist typing in my own hex editor > into memory the boot block of DOS on an (at the time smallest) 8" Shugart > 128K floppy diskette. Those were the days. And when I got to UF their > Computer engineering lab had US play with bitslice to build our own ALU. > But fortunately, before even seeing the hardware in the lab... I simulated > it on one of my PCs in my apartment; and was done in a half hour vs 2-3 > that others had... coding by hand, pencil and paper, and using hex-keypad > to key into raw HW. That was the best fun and learning of its time; I've > often wondered how much it was repeated as "kids today" take for granted > that infrastructure is Just There. So much so, that when I ask about Cross > Platform tool development, nobody in Triangle Linux Users Group wants to > volunteer to talk,,, I would but it's gotten way more complicated to keep > Toolchains up to date (and functional) that Egg on Face is always a likely > occurrence. > > Thanks for your musing this AM and the memories it brought back. QED > > On Sat, Dec 20, 2025 at 12:33 PM Pete Soper via TriEmbed < > [email protected]> wrote: > >> The blog below caught my attention on Hacker News and got me thinking >> about bottom up understanding of computing. This is about projects relating >> to some kind of demo/contest event for FPGA's (field programmable gate >> arrays), but the author goes into some detail about how entertaining "toys" >> like a VGA graphics generator can be made. That led to thinking about >> whether learning to work with FPGAs would help somebody to understand how >> computers "really work". But a little more thought made me ask myself >> whether a modern programmer even needs to know about, let alone understand >> machine language. Not clear that this is relevant. (but I'd love it if at >> least one CS course would show how decompiling a single C++ statement >> leveraging overloading, polymorphism, grotesque layers of header >> references, etc can result in an avalanche of machine code) >> Anyway, I haven't even finished this and I'm talked out of the >> supposition. :-) >> But perhaps some of you might be interested in playing with FPGAs, as the >> hobby level hardware is very cheap and tools and examples are plentiful. >> >> https://www.a1k0n.net/2025/12/19/tiny-tapeout-demo.html >> >> Pete >> _______________________________________________ >> Triangle, NC Embedded Interest Group mailing list >> >> To post message: [email protected] >> List info: >> http://mail.triembed.org/mailman/listinfo/triembed_triembed.org >> TriEmbed web site: https://TriEmbed.org >> To unsubscribe, click link and send a blank message: mailto: >> [email protected]?subject=unsubscribe >> Searchable email archive available at >> https://www.mail-archive.com/[email protected]/ >> >> > > -- > Best regards, Mike > > -- Best regards, Mike
_______________________________________________ Triangle, NC Embedded Interest Group mailing list To post message: [email protected] List info: http://mail.triembed.org/mailman/listinfo/triembed_triembed.org TriEmbed web site: https://TriEmbed.org To unsubscribe, click link and send a blank message: mailto:[email protected]?subject=unsubscribe Searchable email archive available at https://www.mail-archive.com/[email protected]/
