log☇︎
▁▄▄▁▁▁▁▁▁▁▁▁⏐︎▂▄▂▁▂█▂▁▂▂▁▁ 2240
mod6: trinque: http://p.bvulpes.com/pastes/4YG2X/?raw=true << Here is my **experimental only** ebuild, 'ave1_musltronic_tools_x86_64-20180924.ebuild'. I've been placing this file, while testing in a directory ' /var/db/repos/mod6/app-tmsr/ave1_musltronic_tools_x86_64'.
mod6: To make this work properly, I built all of the ave1 musltronic tools on a gentoo instance with a 2016 AdaCore GNAT, then bundled up all of the binaries myself, and placed 'em on mod6.net for testing purposes.
mod6: With everything in the right place... (I even needed a '/var/db/repos/mod6/metadata' directory with one file in it, 'layout.conf', that contains one single line: masters = cuntoo) then I was able to run a `ebuild ave1_musltronic_tools_x86_64-20180924.ebuild clean manifest install merge` and end up with the extracted contents in '/ave1_musltronic_tools_x86_64-20180924'.
mod6: Anyway, I'm only just getting started with these. If there are glaring mistakes or feedback otherwise, please write in.
trinque: fuck yes mod6 !
trinque: I will soon give this a try. ☟︎
trinque: hm, in reading it, it looks like this pulls a mysterymeat tarball from your server and unpacks it, but doesn't build anything
trinque: the idea here would be to rebuild it from source, not just use the bins you built
trinque: ah maybe it's all covered by ebuild defaults, I'm perhaps mistaken
trinque: another thought, might make sense to call this sys-devel/gcc-9999, so one might perma-satisfy w/e caltrops demand gcc of a particular (and always higher) version
mod6: hey, sorry for the delay.
mod6: so I never was able to build the ave1 musltronic tools on cuntoo, because no working gnat, I just bundled them up after I built them on a gentoo machine that had a working AdaCore 2016.
mod6: If the thinking is that it ~can~ be built on cuntoo, I'd much prefer an ebuild that would do that dance indeed.
trinque: yep, gcc can be built on a musl system
mod6: alright, I'll have to go back and try to get this to build on here.
trinque: there are likely challenges around getting a gnat on there with which to build itself.
mod6: yeah, i think this was the issue.
mod6: http://p.bvulpes.com/pastes/IY76z/?raw=true << this is what I end up with, fwiw.
mod6: For completeness, this is what happens when you try to install AdaCore 2016 (binaries) on Cuntoo (keep in mind that the install actually does place the binaries in the expected places...): http://p.bvulpes.com/pastes/vX16f/?raw=true
trinque: mod6: this is expected; they're linked with glibc
trinque: oddly the OS reports a missing libc as "file not found" when executing the bin that wants it
trinque: actually makes sense, I guess it means the libc wasn't found, but it's unclear when it happens
mod6: Anyway, if I could work around this issue, then I can make an ebuild that'll build ave1s stuff with this.
trinque: mod6: https://www.openwall.com/lists/musl/2017/11/05/4 << possible line of inquiry
mod6 looks
mod6: hmm.
ave1: trinque, doesn't cuntoo with default gcc also build it with some mystery meat? it probably starts out on a host with gcc? ☟︎
ave1: btw I've *not* used the ada core version as a starting point for about a year now
danielpbarron: !Q later tell mircea_popescu http://wotpaste.cascadianhacker.com/pastes/QcDk9/?raw=true
lobbesbot: danielpbarron: The operation succeeded.
mp_en_viaje: hello from europe's worst shithole -- norway.
mp_en_viaje: i don't know that i can explain what an utter piece of shit oslo is.
mp_en_viaje: http://btcbase.org/log/2019-06-24#1919763 << so what, a new era of redistributing binaries dawns ? ☝︎☟︎☟︎
a111: Logged on 2019-06-24 01:44 trinque: I will soon give this a try.
diana_coman: mp_en_viaje: lolz, what does Norway/Oslo offer so much worse than all the rest?
diana_coman: at least it seems to have working internet unlike cluj...
mp_en_viaje: it's pretty much exactly argentina.
mp_en_viaje: imagine this : i get out of the airport, and there is a queue, i kid you not 1k+ people long
mp_en_viaje: THEY ARE WAITING FOR CABS
mp_en_viaje: fucktarded morons will queue for a cab.
mp_en_viaje: the exchange offices are out and out scams, trading insane nonsense like 9.2 (while advertising 9.4) to buy and 10.5 to sell (their shitty local coin is about 9.6 irl)
mp_en_viaje: imagine the insane cheek required to even print this spread
diana_coman: some will queue for anything; and anyways, not sure if "piling for cabs" is much better otherwise (e.g. Rome) though indeed I can't quite see why would there be such a lack of cabs in Oslo
diana_coman: ahaha, are those exchange offices in town ?
diana_coman: i.e. not at airport I gather
diana_coman: I have to admit I have a hard time doing argentina ~= norway mainly because of the whole south/north thing; that being said, from socialist pov sure, can very well be indeed.
mp_en_viaje: yes, in town
diana_coman: nuts
mp_en_viaje: and mind you, they make like e150 to take you the 30km from airport to oslo
mp_en_viaje: but yes, they're exactly the same thing : hanger ons, marginal shitholes where people wanna be polite and everyone to be equal
mp_en_viaje: result ? FUCKING SAME
mp_en_viaje: i hope the arabs drive them into the sea.
diana_coman: meanwhile I'm trying to disentangle here the cs/ps mess and I'm scratching my head at various bits: the CS engine works everywhere with names (strings) as ids (i.e. there are FindX functions only with string param, not with some numeric id despite the fact that internally it DOES HAVE them ffs) BUT then in some places it keeps various hashmaps with numeric ids for "common strings"
diana_coman: I suspect it's still xml-effect because "oh noez, needs name of object in xml not id"
asciilifeform: BingoBoingo: you may be interested to know that my shell to dulap bumped after 132d. ( for no detectable reason, either on this end or the other )
asciilifeform: mp_en_viaje: i can't resist to ask, why went to oslo ? what's there, other than tomb of naggum ? ( i dun even know where it is, cannot help to either piss on it or to put flower )
asciilifeform: afaik it was 'europe's worst shithole' even in 19th c. competed with bucharest.
asciilifeform: meanwhile in heathendom, 'Canonical's decision to cease development of 32-bit libraries in Ubuntu 19.10 "eoan" means it won't support Steam gaming runtime and devs say the Wine compatibility layer for running Windows apps will be little use.'
trinque: http://btcbase.org/log/2019-06-24#1919790 << I caught it a few bars later. ☝︎
a111: Logged on 2019-06-24 12:27 mp_en_viaje: http://btcbase.org/log/2019-06-24#1919763 << so what, a new era of redistributing binaries dawns ?
mp_en_viaje: asciilifeform, i dunno... it was there
mp_en_viaje: trinque, word
mp_en_viaje: and, obviously, https://i2.wp.com/norwaytoday.info/wp-content/uploads/2016/09/tb5e0db0.jpg?resize=777%2C437&ssl=1
asciilifeform: ugh they have their own merkel?!
asciilifeform: http://btcbase.org/log/2019-06-24#1919790 << old era. afaik no one yet knows even a theoretical cure for ye olde 'to build gnat, 1st find a working gnat' headache ☝︎
a111: Logged on 2019-06-24 12:27 mp_en_viaje: http://btcbase.org/log/2019-06-24#1919763 << so what, a new era of redistributing binaries dawns ?
asciilifeform: thinking about it, i still dunget why there are ~2~ binaryturd gcc's in the mix.
asciilifeform: ( 1 cuntoo's, 1 ave1's )
BingoBoingo: asciilifeform: ty for the heads up
BingoBoingo: Meanwhile in local happenings https://www.subrayado.com.uy/escapo-carcel-central-rocco-morabito-exjefe-la-mafia-calabresa-n537098
asciilifeform: http://btcbase.org/log/2019-06-23#1919694 << prolly i oughta elaborate. in case wasn't clear, i've a ~12kB x64 proggy that boots a mostly-stock linux 3.x kernel, with uart as only device. which means that applications like this are theoretically possible, when i get back to this subj. ☝︎☝︎
a111: Logged on 2019-06-23 21:27 asciilifeform: http://btcbase.org/log/2019-06-22#1919567 << interestingly, i got his to boot, simply required xor eax, 0x3 where byte addressing used, 1 opcode. (not needed for generic linux built by sane people, but i dun have one yet for mips )
a111: Logged on 2019-06-04 00:51 asciilifeform: incidentally, i generate these by machine, and it takes about 3sec per. would have put it as a net-connected hopper thing aeons ago, BUT it of course uses a heathen render (there are no 'demonstrably electrically correct' pdf eaters, and i dun expect one to exist) and suffers from the obvious problem
asciilifeform: and incidentally, ones like this also ☝︎
a111: Logged on 2017-11-19 05:19 lobbes: Currently going down the headless-browser path ben_vulpes suggested. Looking into phantomjs atm, which seems like it could do the job. I have an old craptop I'm thinking of using as the proverbial 'public toilet' to house it on. (This is the same craptop I was planning to put a trb node on. I put a spanking-new ssd in there but then realized that the ethernet dun work anymore; wifi only. May be a good use for the thing to just be a turd server in
asciilifeform: aaand pretty much 90% of mircea_popescu's 'uci' ( i.e. you can plant this wherever, and get artificial 'box' w/ specified os & proggies... )
asciilifeform: ( the remaining 10%, is, of course, the fast rsatron !! )
asciilifeform: it'll defo need review by literate folx before deployment to such things as piz, tho. ☟︎
asciilifeform: ( possibly after ffa, we do a 'mips in 1 instruction a day' likbez series...? )
asciilifeform: ( ftr there is not yet any support for 'artificial disks'. but the mmap'd ram image can be made persistent by setting the 'write' flag, and when running from ramdisk you get in effect actual disk. this imho suffices for some applications. )
asciilifeform: item is nao in a condition where i can safely unload it from head & return to wrapping up ch20.
asciilifeform: funnily enuff, i think that 12kB actually weighs less than the human-readable text i've shat into the log re subj !
asciilifeform: roughly 30 x86 instrs are executed for erry mips instr.
asciilifeform: ( interestingly, this does ~not~ work out to a 'runs at effectively 1/30th the native clock', on acct of this effect. but i won't spoil this yet.) ☝︎
a111: Logged on 2019-06-05 16:47 asciilifeform: cuz yer 'amd f' spends 95% of its cycles waiting for ram.
bvt: asciilifeform: myeah, i was too fast with sse -- simdifying such loops would be ~impossible.
asciilifeform: bvt: afaik sse offers only the very same 64x64bit multiplier as ordinary x64.
asciilifeform: ( nor does it appear to give a 128bit shift/add even )
asciilifeform: so, entirely useless for the given problem, near as i can see
bvt: i agree. i played a bit with the code, arrived at slightly cleaner (and sub-1% faster) code for multiplication, seems that the performance is close to the hw limit (at least at 32 Indices Comba-Karatsuba threshold)
bvt: re ~2.5x speedup: for 8-indeces-threshold asm comba makes up only ~25% of runtime, rest is spent in karatsuba squaring
asciilifeform: bvt: it isn't actually a given that comba is even needed at all, with asmistic 64x64 mul
bvt: you mean, can use just karatsuba?
asciilifeform: aha
asciilifeform: bvt: what i meant is, the optimal asmistic base case is not necessarily comba. could be 'micro' karatsuba, or even straight schoolbook mul. for so long as asm.
asciilifeform: ( ideally -- would be a 256bit iron mul. but it dun seem to exist on any known irons today. )
asciilifeform: http://btcbase.org/log/2019-06-24#1919784 << wb ave1 . ☝︎
a111: Logged on 2019-06-24 03:55 ave1: trinque, doesn't cuntoo with default gcc also build it with some mystery meat? it probably starts out on a host with gcc?
asciilifeform: ave1: what have you been up to ?
bvt: in case only asm is used, plain comba may be fast enough as well
asciilifeform: for the base case -- entirely possib
mp_en_viaje: would be kinda interesting if comba ends up snipped.
asciilifeform: mp_en_viaje: i 99% expect it ~will~ get snipped from the asm branch.
mp_en_viaje: we snip everything (tm)
asciilifeform: ( properly speaking, comba aint even a mult algo in the usual mathematical sense, but merely a 'sideways' restatement of the traditional schoolboy O(n^2) mult algo to behave smoother on pc and other irons where cache is in play )
mp_en_viaje to bed
mp_en_viaje: where oh where does he pop up from tomorrow ?! noooobody knows....
asciilifeform: in the flagship branch, it comes into play as base case when 8 or fewer words in multiplicand
asciilifeform: nighty night mp_en_viaje
BingoBoingo: So with the Party Internal elections coming up Sunday, guess what totally not a campaign event the local Unions are holding tomorrow... A Work Stoppage!
BingoBoingo: Or as PIT-CNT is calling it "paro general de 24 horas"
BingoBoingo: Note the "paro general de 24 horas" is scheduled to cover a full day's hangover after today's Uruguay-Chile futbol appointment
feedbot: http://qntra.net/2019/06/rubio-files-amendment-to-end-us-patent-recourse-for-firms-on-usg-badlists/ << Qntra -- Rubio Files Amendment To End US Patent Recourse For Firms On USG Badlists
trinque: epic, literally going to fasttrack "IP theft" ~from~ the chinese. Huehue Corp must have good shit. ☟︎
trinque: this is good for the death of IP