log☇︎
514 entries in 0.657s
diana_coman: paging ave1 - do you have any idea re compiling GNAT with musl WITH sjlj? see thread above
a111: Logged on 2019-02-15 16:13 diana_coman: anyways, I'll try presently ave1's previous version i.e. http://ave1.org/2018/building-gnat-on-musl-now-with-partial-and-parallel-build-support/
diana_coman: anyways, I'll try presently ave1's previous version i.e. http://ave1.org/2018/building-gnat-on-musl-now-with-partial-and-parallel-build-support/ ☟︎
diana_coman: hm, full replica might require I upload somewhere the tarballs too since at least 1 link was broken ; anyways, it's with http://ave1.org/2018/building-gnat-on-musl-now-always-static/ + taken out the download script for ada2016 because broken link + added GCC_CONFFLAGS="--enable-sjlj-exceptions" to extraconfig.sh ; set path to point to existing and working adacore 2016 gnat + put all tarballs in their place ☟︎
mod6: I'm not asking you to add my posbox foibles into your config. trb/ada/musl that I want to test on there doesn't care what kernel modules are loaded.
asciilifeform: the only hit is x86_64-linux-musl/share/info/mpc.info -- a readme
diana_coman: ave1 or anyone else more experienced in rebuilding ave1's gnat with a previous incarnation of same: I'm trying to build using the scripts in ada-musl-cross-2018-09-24.tgz on a machine that has as only existing and perfectly working !) gnat a previous ave1 gnat version; I ran as the readme says simply ./build-ada.sh absolute_path_to_dir but the whole thing fails because it doesn't find some libs such as libgmp.la; a closer look at the outpu
bvt: musl-cross-make
bvt: built myself, based on ave1s patches and musl-cross-make (https://github.com/richfelker/musl-cross-make/ , i had some experience with it so did the work rather quickly).
asciilifeform: maybe musl implements some functionality is a way unexpected << where didja get a gcc5istic gnat built on musl, bvt ??
trinque: x11 runs fine on musl
mircea_popescu: trinque "it's a test of musl" in the sense that if something breaks it'll likely be that, not reasonable to expect that because you took out whatever solitaire version gets bundled with gentoobuntu then therefore your build crashes.
trinque: yes, snapshot-and-fork of a musl-gentoo, but particularly that the snapshotting produces a genesis.vpatch
diana_coman: aha; well, mine is stock gentoo rather than musl so yes
a111: Logged on 2019-02-03 04:03 trinque: GNU bash, version 4.4.12(1)-release (x86_64-gentoo-linux-musl)
a111: Logged on 2019-02-03 00:35 trinque: ave1: is it your experience that your gcc+gnat builds on musl? builds for musl, yes, but on?
ave1: trinque, http://btcbase.org/log/2019-02-03#1891961, I've build it with itself (i.e. musl). But the ultimate start has always been the gcc version from AdaCore. I could not get it to work with an FSF start. ☝︎
trinque: GNU bash, version 4.4.12(1)-release (x86_64-gentoo-linux-musl) ☟︎
trinque: ave1: is it your experience that your gcc+gnat builds on musl? builds for musl, yes, but on? ☟︎
shinohai: Not bad, just noticed last nite ave1's site back up so in process of building musl gnat today then on to revisiting ffa stuffs
asciilifeform: given as it comes already with sane gcc and musl
mod6: I'm currently building out a Cuntoo (have had a minor issue with my kernel not booting correctly, but will work on that soon), and will use that + ave1's musl to build the keccak v tooling, and then TRB. Which will also go into the upcoming changes to the HOWTO Guide.
asciilifeform: ( 'rotor' & descendants, on musl )
trinque: asciilifeform: I failed repeatedly to build chrome on musl, said "fuck it", dustbinned.
trinque: asciilifeform: I've built ff on musl with various rando's patches, have one running currently
trinque: asciilifeform: we had a thread on the stepping down before, jumping directly to 4.9.4 does not work on musl. not a clue why.
asciilifeform: http://btcbase.org/log/2018-11-24#1874369 << is this also true in musl , trinque ? i haven't looked closely yet ☝︎
trinque: I'll need to read up on how DNS resolution works in musl to recommend something strongly, but might be as simple as having it read a second hosts file. I don't know of any support for including one hosts file in another in either musl or glibc.
deedbot: http://bvt-trace.net/2018/10/linux-portability-part-2-exploring-musl-ifdefs-or-define-pdp_endian-3412/ << bvt's backtrace - Linux Portability, Part 2: Exploring musl #ifdefs, or #define PDP_ENDIAN 3412
deedbot: http://bvt-trace.net/2018/10/linux-portability-part-1-exploring-musl-architecture-specific-headers/ << bvt's backtrace - Linux Portability, Part 1: Exploring musl Architecture-Specific Headers
billymg: is this the right place to start reading to get gnat installed? http://ave1.org/2018/building-gnat-on-musl-no-more-usrincludex86_64-linux-gnu/
bvt: http://bvt-trace.net/2018/10/linux-portability-part-2-exploring-musl-ifdefs-or-define-pdp_endian-3412/
bvt: hi, i have made some exploration of linux syscall interface (using musl) http://bvt-trace.net/2018/10/linux-portability-part-1-exploring-musl-architecture-specific-headers/
bvt: well, http://git.musl-libc.org/cgit/musl/tree/src/network/connect.c -- the structure is the same; digging is rather easy, if you don't look into glibc
bvt: the thing is, structure definitions and all sort of flag numbers appear in the libc via magic. having all this things in ada is possible, and would involve exactly same work that e.g. musl people are doing today
bvt: but so far it's not clear what the subset should be, what to do about C structures (with problems like e.g. https://www.openwall.com/lists/musl/2017/01/25/1)
phf: well, it's also a reason why bug wasn't caught in development. a sequence of wtfs: linux man page says mktemp should be 3 or more X's, so project builds on a non-musl build. meanwhile POSIX mandates there to be exactly 6 X's, so a musl build fails to produce a random string, returning instead a blank one, which is when gnat decides to not only generate a temp file but also do cleanup.
bvt: 3. Musl calculates a simple hash over current time, address of variable on stack, and address of template to generate the random characters for mk*temp family of functions: http://git.musl-libc.org/cgit/musl/tree/src/temp/__randname.c#n6
phf: asciilifeform: to be fair musl's version is naive, it generates/stats in a loop until there's no collision, but theoretically you can still get a race condition. glibc i believe does all kinds of smart tricks to ensure that doesn't happen, but it's all magic
bvt: Reapplied ave1's patches over gnat17, and musl-cross-make (https://github.com/richfelker/musl-cross-make/)
bvt: musl wants 6 template chars: http://git.musl-libc.org/cgit/musl/tree/src/temp/mktemp.c#n13
asciilifeform: and the orig 'rotor', i baked for tbf, and from it we have the musl thing, etc
deedbot: http://ave1.org/2018/building-gnat-on-musl-no-more-usrincludex86_64-linux-gnu/ << ave1 - Building GNAT on MUSL, no more /usr/include/x86_64-linux-gnu
trinque: quick update on cuntoo before I depart. the fully machine-driven transformation from snapshotted gentoo to cuntoo genesis.vpatch works, and successfully rebuilt itself whole. I've also got the classical gentoo repo acting as a subordinate repository, such that porting ebuilds will be extremely easy (i.e. gentoo repo is now an overlay, just like musl overlay, which can be used or not as decreed by operator,
a111: Logged on 2018-09-22 18:46 phf: asciilifeform: that i can see, but i'm not sure why it's picking up system wide includes. there's an isystem there that points to correct musl include tree, that has the sys/types.h file in it
phf: asciilifeform: i'm not sure but i don't think the problem is environment variables. the failing command has an -isystem flag that points to correct musl header tree, and some of the files are being picked up from there, but not all (e.g. errno.h is picked from musl, but sys/types.h from global)
phf: asciilifeform: that i can see, but i'm not sure why it's picking up system wide includes. there's an isystem there that points to correct musl include tree, that has the sys/types.h file in it ☟︎
a111: Logged on 2018-09-22 03:05 phf: ave1: i'm getting the following error http://p.bvulpes.com/pastes/49Tie/?raw=true when building ada-musl-cross-2018-06-01. i feel like it came up before, but i'm having hard time finding the thread
phf: ave1: i'm getting the following error http://p.bvulpes.com/pastes/49Tie/?raw=true when building ada-musl-cross-2018-06-01. i feel like it came up before, but i'm having hard time finding the thread ☟︎
phf: is ada-musl-cross-2018-06-01 the latest version of ave1's ada?
asciilifeform: diana_coman: libc, specifically ( i like musl, but it doesn't belong on the fyootoor all-adatronic box )
ave1: this builds with the musl gnat, the binary from adacore and the latests zfp.
asciilifeform: in other noose, trad-gentoo's crossdev is completely rotten nao, tries to build glibc-2.27 (which is broken, as it won't build with any version of 'ld' i was able to find, full stop) EVEN WHEN YOU CROSSDEV ....linux-musl !!
asciilifeform: Mocky: cpp proggy always rides on libc. witness trb, the orig experiment with musl here.
Mocky: so in reading the logs I see that musl is a libc which is smaller and stricter than glibc. is there such a thing for c++ standard library or is it not needed?
diana_coman: I almost hoped to read "diana has trouble fucking the goats with eucrypt on musl cuntoo"
asciilifeform: the possible suspects , so far -- 1) ttl dongles 2) hub 3) musl userland , somehow
diana_coman: well, 1st test: changed to 2nd FG on musl-system; checked and it reads fine with dd from console; ran the tests of eucrypt , it read 4 and then it got stuck
diana_coman: asciilifeform, fwiw I am trying to get it on my local box (non-musl) and so far it's working perfectly fine
asciilifeform: and , while you're at it, from the other FG in the musl box.
asciilifeform: diana_coman: see if you get same horror on the old, non-musl box
phf: some libc's (specifically in glibc, musl, there's also a custom one in busybox. presumably cygwin/mingw comes with a glibc derivative), but that shouldn't be used directly. PeterL's patch is not needed on linux/freebsd/apple.
a111: Logged on 2018-07-18 13:46 phf: or is tmsr ada whatever ave1 put into his musl build, which is, worse, a political situation. diana_coman can argue for her ffi stuff to be included, should i be arguing for my get/put stuff to be included?
ave1: for overview there will probably will be about 6 flavours of zfp I think based on: MUSL C, X86_64 ASM + linux, X86_64 ASM + no os, AARCH64 + linux, AARCH64 + no os.
asciilifeform: phf: are we speaking of http://ave1.org/2018/building-gnat-on-musl-updated-tar-line/ or http://ave1.org/2018/gnat-zero-foot-print-take-2-no-c/ ? these are separate items
phf: or is tmsr ada whatever ave1 put into his musl build, which is, worse, a political situation. diana_coman can argue for her ffi stuff to be included, should i be arguing for my get/put stuff to be included? ☟︎
a111: Logged on 2018-07-15 16:22 asciilifeform: http://btcbase.org/log/2018-07-15#1834764 << trinque do you have a list of items you observed to be broken under musl ? or is this still in the works
BingoBoingo: <trinque> some of this work won't have to be done completely ab initio; there are musltronic distros out there that can at least act as source material for research, alpine and void linux for example. <<As far as I can tell Alpine is a musl build of gentoo, comparatively clean. Void on the other hand is more like a tampon. Bloody inside and out
asciilifeform: i meant 'any'. afaik the only remaining graphical www thing that builds even on asciilifeform's ban-flagged gentoo ( not even speaking of musl ) is old firefox.
a111: Logged on 2018-07-12 11:18 asciilifeform: trinque's gcc, btw, is exactly as was printed on the crate, x86_64-gentoo-linux-musl 4.9.4
a111: Logged on 2018-07-15 15:45 trinque: musl will "break" where it refused to implement glibc feature-creep outside the posix libc standard. it's a stricter implementation.
asciilifeform: http://btcbase.org/log/2018-07-15#1834764 << trinque do you have a list of items you observed to be broken under musl ? or is this still in the works ☝︎☟︎
a111: Logged on 2018-07-15 15:45 trinque: musl will "break" where it refused to implement glibc feature-creep outside the posix libc standard. it's a stricter implementation.
trinque: to take a specific example, my mention of musl distros isn't an excuse for anyone to blindly pilfer patches and then sign because "built!!"
trinque: musl will "break" where it refused to implement glibc feature-creep outside the posix libc standard. it's a stricter implementation. ☟︎☟︎
trinque: http://btcbase.org/log/2018-07-14#1834507 << read and enjoyed your porting process for mysql. this will be how things go. many items already have patched ebuilds in the musl-overlay, which your system is using atop traditional portage. some wont, and will need to be fixed. ☝︎
asciilifeform: diana_coman: was mysql the only eulora component that misbehaved on musl ?
deedbot: http://www.dianacoman.com/2018/07/14/cutting-mysql-into-musl-shape/ << Ossasepia - Cutting MySQL into Musl Shape
ave1: btw the most recent version will only build static executables! so no support for shared anything. Does the musl / gentoo overlay depend on shared libs? (Python usually builds statically and depends for installable modules on it, etc.)
asciilifeform: trinque's gcc, btw, is exactly as was printed on the crate, x86_64-gentoo-linux-musl 4.9.4 ☟︎
hobotrinque: removing swap is totally fine by me. musl gcc-4 would not build on gcc-6
asciilifeform: ( not accident, but per trinque's script, chroot . emerge '=gcc-5.4.0-r4::musl' )
asciilifeform: phf: as discussed in the recent diana_coman thread, if you want systemwide musl-anythings, gotta have a pure musl box.
asciilifeform: phf: only on a system where the systemwide crapola is traditional (glibc) and your musl item is a homedir-local build, a la rotor.
phf: asciilifeform: is it possible to combine musl and libc on a same gentoo system (gentoo insists i use something they call crossdev, i haven't yet looked into it further) (i'm using your aarch64 gentoo root file system)?
phf: i very briefly looked at ave1's script, and i'm guessing that's what it's supposed to do (besides musl related): build a x86 hosted aarch64 crosscompiler, build an aarch64 build using that compiler?
asciilifeform: mircea_popescu: whole thrust behind asciilifeform's initial colonization of musl was to be rid of dlopen() etc
trinque: anybody that wants to help verify the portage genesis.vpatch later oughta try building the musl gentoo in my post.
ave1: Then I decided I could better spend my time on musl than trying to salvage anything from gnu libc
mircea_popescu: look, the bootloader thing is either musl or isn't. if it isn't, i expect trinque wants to hear about it ; whether he wants to or not i still want it said.
a111: Logged on 2018-06-21 12:34 asciilifeform: http://btcbase.org/log/2018-06-21#1827977 << 're-emerge' seems to imply systemwide ? you're more or less guaranteed a borked box, muslism has to be done either rotor-style (i.e. 100% user-local build of 1 proggy at a time) or systemwide ( trinque's cuntoo ), on account of the impossibility of cleanly linking glibc libs to musl proggy or vice versa
mircea_popescu: the "a full gnat musl thing" portion seems to be satisfied by trinque 's currently published bootloader.
mimisbrunnr: Logged on 2018-06-20 16:30 mircea_popescu: which brings us to : ben_vulpes would pizarro be amenable to bring up the spare sometime so she makes it a full gnat musl thing, test whether we can move everything there, and either move it (so basically, moving servers) or else backing down (so basically i guess either renting both for a while or powerting back down the spare) ?
mircea_popescu: i dunno why ; cuz i said the idea is to ~test a musl process~ ; without hosing the machine currently running it.
mircea_popescu: so if i use it, the resulting system will in fact permit my testing of a whole set of musl-compiled deps, system wide, such as mysql and whatever other libs eulora pulls.
mircea_popescu: trinque ok, help me understand this. so you have a script that a) works, in the sense of bootloading a b) gentoo which is musl-compiled but c) doesn't have a proper v tree.
trinque: mircea_popescu: it is musl; it is just not using a v-tree of ebuilds for its portage
mircea_popescu: trinque ok, but it's not musl.
asciilifeform: diana_coman: as i understand, it would not be a wasted effort , given as if a musl-incompatibility is revealed , diana_coman can go straight to patching it, working in parallel with trinque's finishing touches on cuntoo
asciilifeform: this will reveal where, if anywhere, your orchestra breaks on musl