IRC channel logs

<siraben>started fuzzing hex2, blood-elf, m1, kaem. Good news is that hex2, blood-elf and m1 don't have crashes. But kaem...

<stikonas>I think oriansj fuzzied hex2, blood-elf and m1 a lot

<stikonas>well, M2-Planet too

<stikonas>but you found some more

<stikonas>though I guess those were recent changes

<siraben>I'm fuzzing cc_x86

<siraben>Switched to grammar-directed fuzzing after black box stopped returning problems

<siraben>What do people think about switching to gas syntax instead of m1?

<matrix_bridge><wildwestrom> Isn't the reason we have M1 because it's simple to compose macros rather than parse gas-like syntax?

<matrix_bridge><wildwestrom> This was mostly generated code, but this is a rough example of what that could look like: https://github.com/wildwestrom/stage0-riscv64-baremetal/blob/main/baremetal/GAS/as0.s

<matrix_bridge>It's a GAS-like assembler in GAS.

<siraben>prs up for cc written in m1/gas

<siraben>Also finally generalizing hcc to not be overfit to tcc heh

<roconnor>Oh I didn't realize that nix merged its minimal bootstrap.

<roconnor>siraben: are you going to PR the M2-Mesoplanet fix?

<siraben>Yeah will do

<matrix_bridge><Andrius Štikonas> wildwestrom: yes, but gtker is working on gnu syntax assembler

<siraben>Adding a lightweight make to be used, at the same level as kaem (compiled by cc_*)

<siraben>roconnor: yeah they merged it in since 4 months ago

<siraben>pretty neat! considering a project of this size

<siraben>and see channel log for the various prs on stage0-posix-*

<stikonas>yeah, ok, I'll take a look

<stikonas>those asm programs are definitely written with the assumption of a valid input

2026-05-13.log