IRC channel logs
2023-05-16.log
back to list of logs
<oriansj>Let us hope there is a lambda core in Ada, with which one would be able to bootstrap the language. <mihi>Also i noticed, both for mescc-tools and for M2-Planet, when I run "make install" while any of the binaries is not built, make's VPATH magic seems to fail and install that file from the wrong directory. I am not familar enough with VPATH to fix it, though. <mihi>that was quick, thanks :) <mihi>never let your guards down. even I am not infallible. <Irvise_>theruran: the magic of binary releases :))))) <theruran>Irvise_: it's some Wizard of Oz shit. making the release process a mystery, and "just trust us!" <Irvise_>theruran: tell that to charlie5 over at #ada x) he has been packaging most if not all Ada packages for Archlinux/AUR. <h01ger>ACTION finds it ironic, that https://www.pdc.kth.se/about/what-is-pdc-1.736889 does not explain what PDC stands for. "The PDC Center for High Performance Computing at the KTH Royal Institute of Technology is a member branch of the National Academic Infrastructure for Supercomputing in Sweden (NAISS), which is the leading provider of HPC services for academic research in Sweden." - fine. and what <h01ger>ACTION do the letters PDC stand for? <janneke>it seems the bootstrap path to that abbreviation was lost <[exa]>h01ger: the history page looks like it's "centre for parallel computers" but in swedish I'd say <[exa]>ah yes, PDC Parallelldatorcentrum <h01ger>even works in engrish, parallel data center <oriansj>theruran: the latest merged change of M1 in mescc-tools; I always fuzz after code changes as I want mescc-tools and M2-Planet to be as robust as possible in the handling of bad input <oriansj>and if I knew of a way to make the generated output of M2-Planet more robust, I would leverage the hell out of that too <oriansj>currently after 17 hours and 33 minutes of fuzzing; no crashes found yet <rickmasters>oriansj: I hate to bring this up as it may be pedantic nitpicking you've already thought of but: <rickmasters>stage0-posix-x86 seems more accurately described as stage0-posix-linux-x86 <rickmasters>as the raw interrupts being used are not defined by POSIX - they only match their semantics and are specific to linux <oriansj>indeed and work would need to be done to natively port it to say MacOSX syscalls or NetBSD <rickmasters>oriansj: yeah, I was thinking about your repo/directory structure and wondering where you would put something like those... <oriansj>well right now only M2libc has that distinction <oriansj>and in the future if someone does the work of porting the bits to the new POSIX bases, then I'll be more than happy to update the names of the repos <oriansj>but as currently Linux binaries can run on FreeBSD and NetBSD; Traditional POSIX systems don't seem to have any resistance to breaking binary compatibility without such much as notice of having done so. It is effectively the only stable POSIX ABI we have. <rickmasters>regarding your directory structure perhaps stage0-posix-x86/linux works if the distinction is needed. <rickmasters>anyway not a big issue right now so we can resume regularly scheduled programming, so to speak <oriansj>or if we wanted to order the names in importance: x86-linux-stage0 <rickmasters>that latter one seems like a bigger change as you've nested under stage0 already, but we're just spitballing... <oriansj>true, honestly the name never mattered to me. All that mattered is that the work is done and we had fun doing it together. <janneke>rickmasters: that's a greand read, thanks!