I’ve spent months working on this program. I’ve long thought that “how fast can you make a FizzBuzz” would be a really interesting question for learning about high-performance programming, and when I subsequently saw this question posted on CGCC, I pretty much had to try.
This is so beautiful 🥹
edit: outjerked yet again https://codegolf.stackexchange.com/a/269772
This isn’t even really C, it’s a bunch of assembly instructions jammed into C
Not that it’s not impressive… but still.
Correction: This is not C. It may seem so because of the #include’s and #define’s but this is just standard gnu assembler syntax. You can notice that the included files are all in asm/ and the macros are just valid assembly.
It’s built with GCC instead of as because gcc links the standard library (which include those asm/ files) for you. Though you can build it normally with as, too.
Heavy, iirc. I haven’t done this stuff in a while so please correct me if i’m wrong.
Why did you post that implementation and not David Frank’s with 1.7 Terrabit/s ?
Can we get a tldr for the approach used?
hand-optimized SIMD-instructions probably
I saw it before that one, and by that time i already posted. I think deleting and reposting is a little weird so I guess I’ll have to make due X)
I’ll mention it in the post body though, thanks lol
I already have a master’s thesis. This was harder.
Lmao
God I miss PCJ 😭