asciilifeform: BingoBoingo: ketamine, lol! when asciilifeform was research slave for us army, they were offering comparatively impressive sums (coupla thou) to anyone willing to take ketamine and run labyrinths / do various puzzlers
asciilifeform: i'm still waking up, coming up empty when digging
asciilifeform: incidentally, has anyone yet published a ~complete~ log bot kit ? last i knew , all that was trooly genesis'd is http://btcbase.org/patches/logbot-genesis , which is one-half of a log bot ( the rear half )
asciilifeform: unless it was the cpp components which did
asciilifeform: (i.e. quite unlike ida, they actually hired 9000 grunts to try & reimplement... )
asciilifeform: mircea_popescu: my entire setup runs on surplus gear, which turns up at reasonable price point erry odd year. so not precisely 'impulse'.
asciilifeform: i still have difficulty swallowing that they can make anyffin moar complicated than camel saddle.
asciilifeform: BingoBoingo: hooka worx best with ~servant~ who fills/empties, i suspect
asciilifeform: ( iirc we had thread, re how ~100% of usg.psychopharma 100y consists of 'let's reimplement 2000 y.o. plant dopes in patentable form' )
asciilifeform: 'no one has yet created body armor with air conditioning built in' << iirc this was actually commissioned during usg's iraq adventure. ( or at the very least, a golden toilet contract was issued for it, i have nfi what the accompanying theatrical prop looked like, and whether anyone bothered to produce )
asciilifeform at one time owned 'mp3 player', was this brick that connected to parallel port...
asciilifeform: world's saddest mp3 player, possibly second only to sony's
asciilifeform: trinque: i put it in its coffin in 2015, with 'rotor' , and wasn't even aware that there was contemplation of keeping it alive until mircea_popescu asked last mo
asciilifeform: mircea_popescu: re adacorpse's standard & compiler, the marvel aint that it's broken in places, but that it was built at all, and is even largely usable
asciilifeform: ( and that the can per se was added in 'production' model to improve thermal transfer )
asciilifeform: for thread-completeness, http://btcbase.org/log/2018-11-30#1876504 << bottom side of the 'tin can' variant.
asciilifeform: compare with the fg xilinx die , the latter has no fancy package, and is entirely homogeneous at 35kV.
asciilifeform: http://www.loper-os.org/pub/amberglint_bolix_1.jpg ( via http://btcbase.org/log/2018-11-30#1876538 ) for comparison.
asciilifeform: i dun know ~anyffin about it, either. q is where the fuck standards-compliant soft are to come from, to begin with.
asciilifeform: cuz that'd be interesting. i initially supposed that it was built as a 'lowest common denominator', i.e. without presumption that os supports preemption
asciilifeform: ( ye olde msdos did not claim to implement multitasking at all, so dun really belong in this comparison, but the various bolt-on tsr systems for same also had this property )
asciilifeform: btw there were a great many 1990s os that behaved exactly like zcx model -- no preemption, tasks run until they yield or self-terminate. e.g. win9x, crapple os <= 9, etc
asciilifeform: bvt: i started with the supposition that it was built for irons where there is no time slicer , but currently nfi whether this is so, or whether was simply a kludge for no particular reason
asciilifeform: bvt: zcx is broken ~by design~, it doesn't attempt to implement preemptive multitasking at all.
asciilifeform aint about to use a closed turd compiler , regardless, for any practical work. but still finds the q of 'is there a standard-compliant adatron somewhere' not wholly uninteresting
asciilifeform: diana_coman: 'compliance tools' aha. but , if not with the standard... then with what.
asciilifeform: commercial gnat complies
asciilifeform: good % of the riotous complexity of the sadkernel, once you subtract deviceisms, is the slicer
asciilifeform: btw when we plant gnatism on bare irons, will have to implement a scheduler, and it is 'black art' of sorts ( how big to make the quantum ? how to apportion slices to cores ? etc )
asciilifeform: for compactness
asciilifeform went to compute mircea_popescu's q, 'just how much did it eat', apparently it's a bitch : absorption constant varies by chip, and in heavy industry folx mostly gave up , they stick dosimetric film underneath the board. i'ma get an upper bound tho, it's important q when we do the bitflip thing.
asciilifeform thought 'oughta transparent..' and closed hole in the sample shelf with paper.
asciilifeform too lazy atm to compute approx dose from this, in civilized units
asciilifeform: normally the killing dose for $chip is determined empirically.
asciilifeform: it's not even clear that this dance is req'd, of yet, seems like there's a lid which simply needs to come off, on these
asciilifeform: ( ftr, companion pieces, http://www.loper-os.org/?p=46 & http://www.loper-os.org/?p=448 )
asciilifeform: meanwhile, in x86 lulz, http://www.loper-os.org/pub/wampler/index.html
asciilifeform: ( tho it doesn't impact asciilifeform's current item , or afaik anything else just nao )
asciilifeform no longer specifically stampcollects these, but occasionally turns up by chance
asciilifeform: e.g. how come erry man of fighting age in vz aint on the boarder, pumping lead into the 'humanitarian convoy', manning ack-ack batteries, etc ?
asciilifeform: ( or perhaps could, but would have to station an army around it, and it begins to look economical in comparison to put it on ocean floor )
asciilifeform: 3-4 , i suppose, if it's going to junkyard instead of marseilles and you can compact it a bit
asciilifeform: cuz impedance-tolerance is tight.
asciilifeform: ftr there is a (theoretical) 'innocent' explanation for 'nuffin socketed' -- at current-day freqs, sockets actually cost comparably with the chip
asciilifeform: ( not simply arm, either, but ~anyffin~ non-x86 )
asciilifeform: rright. but must observe, that folx who dun grasp http://btcbase.org/log/2018-06-12#1823933 , ~will~ build a gnu. regardless of what they're nominally attempting to build.
asciilifeform: the pov where 'esr quisling, and if not for him, linux would work' is imho simplistic -- there were 9000 quislings, and maybe dozen frustrated honest folx . what could that possibly have added up to.
asciilifeform: even without any poetterings -- the gnutards just about make republican spain look like unified front in comparison ( e.g. what to make of http://btcbase.org/log/2019-02-23#1898976 ?! )
asciilifeform: mircea_popescu: fella & his worx certainly excellent demonstration of 'what happens to 'technically competent' without http://btcbase.org/log/2019-02-20#1898175 ' imho
asciilifeform: hanbot: orig 'v' was an asciilifeform torture room item, but it was other folx who made it actually useful ( mircea_popescu -- conceived a philosophical foundation for it, conceptually; ben_vulpes -- documented; mod6 et al -- reimplemented , and filled missing pheature holes ; trb users -- battlefield-tested; phf -- keccakized; etc )
asciilifeform: http://btcbase.org/log/2019-02-24#1899025 << imho trb ml contains massive record of ' asciilifeform's initial failed attempts at v ' : the sequence of 'determine flow with bare hands' pre-v patches. ( signed, too )
asciilifeform: has plenty of folx with primo employment quals to be fulltime furnace stokers, tho
asciilifeform: 'November 2, 2008 Cross compiling gcc's c++ support doesn't work because even though I'm saying --enable-threads=posix the various subdirectories are trying to run the compiler it built to grep out the "Thread model: " line to set the config entry target_thread_file (which gets used to set glibcxx_thread_h, which is breaking) and you CAN'T RUN IT WHEN WE'RE CROSS COMPILING YOU TWIT!'
asciilifeform: 'November 9, 2008 And Morris got cracked. Great. Both busybox.net and uclibc.org may be compromised.'
asciilifeform: 'A sad day; the McDonald's double cheeseburger is no longer a dollar menu item. They now have the "McDouble" (which is the same thing but only one slice of cheese), but it's a sign that the whole thing is likely to drop off soon. Apparently, just because the economy has imploded doesn't mean we haven't still got galloping inflation.' << there's 9000 of these, i'ma leave off , so as not to ruin appetites
asciilifeform: ilar. If it's not electronically searchable, it won't matter... ...Project Gutenberg will scan in and OCR the public domain books, but in 100 years will anybody remember that UPI (a competitor to AP that went belly up a few years ago) even existed' << poor bugger
asciilifeform: ^ the other attempt at 'alt gcc'
asciilifeform: ( ftr linus is pretty lulzy example of 'uses gpg' -- erry single time signed MB-weighing ball of ??? that in no conceivable way he could have read or fit in head )
asciilifeform: 'To that end, the compromise may actually be a good thing in the long run. Kuhn said that they have contacted the CVS maintainers and have offered to pay for development of features that would allow GPG signing of commits through CVS -- making it much more difficult for changes to be inserted unnoticed into code held in a CVS repository. He said that they have also contacted the GNU Arch maintainer about adding GPG signing. Though it
asciilifeform: quite 'effectual', turns out, simply not to any effect you'd want.
asciilifeform: d00d was entirely happy to exercise authority . simply not to any sane end, but as strictly reflex to butthurts, 1:1.
asciilifeform: afaik rms is best-known example of this, his www is fulla 'free palestine' etc
asciilifeform: he attempted this -- 'hurd'. but was unable to commission a working kernel.
asciilifeform recalls also pete_d's attempts to sell a 'rebrand' of fuckgoats somewhere
asciilifeform: one might naively picture heathens would jump at 'i bring martian tech and +9000 strength of will' etc. but nope. 'we built a++ straw plane, cargo when'
asciilifeform: it is ! it's the turd the gimp folx replaced ye olde tinyscheme with
asciilifeform spend 2d digging in tcc and a (sad) attempt at crystallography
asciilifeform: in entirely other olds, https://repo.or.cz/tinycc.git/blob_plain/HEAD:/arm64-gen.c << current (?) arm64 backend of tcc. ( nfi whether worx -- but pretty compact , as these go. )
asciilifeform: via bvt's link : '...versions of GCC prior to 4.8 also allow bootstrapping with a ISO C89 compiler and versions of GCC prior to 3.4 also allow bootstrapping with a traditional (K&R) C compiler...'
asciilifeform: other interesting claims, in https://bootstrappable.org/projects.html we find that 'The C and C++ compilers of the GNU Compiler Collection make up the foundation of many free software distributions. Current versions of GCC are written in C++, which means that a C++ compiler is needed to build it from source. GCC 4.7 was the last version of the collection that could be built with a plain C compiler'
asciilifeform: but won't complain, pleasant surprises are few in life
asciilifeform: Mocky: they forgo threading cuz on c machine you have exactly 3 options : a) forgo threading (i'ma not detail python's 'forgo threading and then lie about it' as separate variant, it is beneath contempt) b) sit down on unix's threads (pthread) c) implement own scheduler
asciilifeform: some things are simply steaming piles of shit, from which there is nuffin to say, and the only use of which is that one can make a few bux cleaning'em up
asciilifeform: Mocky: i sympathize re that you toiled in it for 20y; for my part , i work with... microshit kernel diddles etc, for bread. but i'll be the last to say 'here's what we can learn from microshit' (afaik there is ~0 to be learned from microshit, other than how to run the hell away)
asciilifeform: afaik java simply aint 1 of'em tho.
asciilifeform: ( ever wonder wai fortran still in use at '9000 cpus' massively parallel comps ? this is why )
asciilifeform: hanbot: v.py actually exists in modern ( keccakistic ) incarnation -- there's 2 variants, largely identical , diana_coman's (the 1 i use now) and phf's (he had some incompatibility in his python iirc which required own variant)
asciilifeform: i find it hard to picture how a sane cpu, where optimizing compiler is 10k loc and 'fits in head', could have no market. but then again the bolix people proved that it ~is~ possible to go broke with one.
asciilifeform: you can't, for instance, buy a x64 that ~only runs in 64 mode~ : no, you gotta have the (broken) msdos compat, inner 386, etc. cuz winblowz.
asciilifeform: the worst thing, imho, is that the 'baseless distinction' is quite often 'in one strand' with actual distinction -- e.g. the 'speshul olympics' of x86 vs arm : both are shit archs, but 1 has 35 yrs of compat-crud, and eats 100x the wattage vs other
asciilifeform: writing sane compiler targeting e.g. x86 is not entirely unlike dancing ballet while carrying 50kg weights
asciilifeform: this is quite typical example.
asciilifeform: the tard arch people passed the 'register killing' ball to compiler, as result bloating ~all~ compiler with massive gnarl, ~just for reg killing logic~
asciilifeform: i confess, did not follow the thing's trajectory into the ground, do not know which fashionable insanity in particular it followed as it died of immunocompromise
asciilifeform: he had a kernel-building compiler in coupla 1000 loc .
asciilifeform: btw do you recall fabrice bellard's compiler ?
asciilifeform: a la mpi etc
asciilifeform: it's at the mpi confiscation stage atm.
asciilifeform: mircea_popescu: asciilifeform's observation is that all historic attempts at 'pure (b)' were gimped by low expectations from iron, on acct of author's lifelong crippling experience with shit irons
asciilifeform: mircea_popescu: in pctardation land, it's a consequence of the (massively failed, historically) attempt to hack around the fact that erry peripheral has free run of entire bus
asciilifeform: in asciilifefor's cosmography, sane machine has neither 'linker' nor 'compiler' in the customary sense
asciilifeform: moar concretely, e.g. memory allocator -- attempts to put sane one on top of linux's, all caught fire in very similar way, i.e. impedance mismatch ( whether libc's or java's or commonlisp's )
asciilifeform: it's a forced failure, in the sense that it is imposed by trying to be a layer on top of os, rather than os per se
asciilifeform: i suspect not only cheaper in manpower, but faster in wall clock. cuz can hand-massage emulator, whereas compiler barf , even with peepholing etc, is invariably bloated.
asciilifeform put ~decade into various attempts at $subj, but ~0 forward motion until made the sage breakthrough and gained ability to debug bios. but has been on back burner in preference to moar urgent tmsrisms ever since
asciilifeform: i suspect also that erry sargeant in erry unit on planet, also could summarize compactly.
asciilifeform: re the penicillin -- also compactly summarized in http://trilema.com/2014/the-all-american-asshole-in-his-own-words-with-my-own-notes/#selection-3501.0-3501.226
asciilifeform: or the deathbed bolix people, who did ~99% of the job but sank on the titanic of the dec alpha, the only arch they managed to massage into compliance before ran out of dough
asciilifeform: various folx attempted something roughly like this, but succumbed to own idiocy -- e.g. the java people, who 1) wrapped their 'vm' around the nonsensical shape of their tardlang 2) stopped short of 'and no this doesn't run on your tard os, is IS the os, fuckyou' ;
asciilifeform: yer still stuck with certain idjicies of iron (e.g. dma) but then when finally bake new iron, you have a ready compiler & userland to plant on it, that actually behaves.
asciilifeform: really oughta have ~1~ compiler, and for a sane arch; and if yet cannot afford to actually siliconize the sane arch, then the sad iron oughta be booting straight in bios to a compact (asmistic) emulator of same
asciilifeform: i also suspect that on modern irons, 'retargetable compiler' is plugging the wrong end of the funnel.
asciilifeform: the most rampant inconsistency, btw, is not even in the userland abi, but in driver (module) end
asciilifeform: 1 slightly moar practical variant of this seal would be if could make the 'glitter' in the epoxy from kcl or other commonplace mild radioactive. then tape instant film over the chassis and develop/swap erry coupla months, and compare the liquishit to previous photo.
asciilifeform: http://www.loper-os.org/pub/misc/xray/dead_samsung_35kv_100s.jpg << to complete yest. thrd : the least rubbishy shot of the dead samsung. steel shell prolly calls for a scintillator screen , and large film, to actually get decent shot. but even here usbistic contacts sorta visible.
asciilifeform: ( it may be possible to pack'em tighter, but currently imho reliability trumps max utilization of cubic metrage )
asciilifeform: what we have this wk, is a 'sjlj-gnat dun go on arm64'. i dun expect this is a permanent disease, but i do expect it'll take time to fix, esp. if asciilifeform has to do with own 2hands. until then, rk cluster is stuck with the obsolete compiler ( which runs e.g. 'lamp stack' and even zcx-gnat , but obv. won't run the emerging standard sjljistic one until fix. )
asciilifeform: meanwhile, in world of ancient warez finds : http://nosuchlabs.com/pub/warez/artek.zip << ada-83 compiler for msdos, circa 1987.
asciilifeform: i can see that it might have to let go of any locks it holds, but what if holds none ? this is easily shown at compile time
asciilifeform expects that it'll fit, compressed, on 1 cd
asciilifeform: ( this is where i point out, that the fabled 'sane iron' isn't simply a purely aesthetic win to ticke asciilifeform's aestheticles, but in fact substantially cuts down on the complexity of ~all other sane items~ that are to stand on top of it )
asciilifeform: ( for n00bz: ) writing a compiler back-end aint actually hard. asciilifeform & many many other folx, did it ~as homework~ , at school. the hard thing is writing a ~decent~ optimizing backend.
asciilifeform: at the very least, theirs is 50x moar compact than gcc.
asciilifeform: could propose that there is a (3) , if one of the closed $maxint winshit adas actually implemented own, rather than stealing gcc's
asciilifeform: funnily enuff, i suspect there are a grand total of ~two~ ada back ends in existence : 1) the Official adacorpse one , sewed out of gcc ( the 'fsf gnat' is simply old copy of same )
asciilifeform: concretely -- on sane iron, cpu do not share memory, but instead implement exactly diana_coman's work-queue mechanism.
asciilifeform: all of this might seem uninteresting until you realize that this is what sits under all threading, no matter how implemented on os side.
asciilifeform: this was 'fixed' by intel (with obscene amt of direct standard authorship by microshit) by making the controller 9000x moar complex
asciilifeform: old msdos hands will recall setting jumpers for e.g. 'soundblaster' irq etc
asciilifeform: ( incidentally, when asciilifeform speaks of 'iron babel', interrupts are a screamingly concrete example : there is ~no uniformity b/w archs re how they're implemented (does it save regs ? which ? what happens if two interrupts temporally near ? ) or how many , or for what devices, etc )
asciilifeform: ( when build on unix -- it gets implemented as unix signals )
asciilifeform: diana_coman: the 1 use case i can picture for zcx, is on ultracompact irons. but even there, really, are we gonna use a http://btcbase.org/log/2019-02-12#1895612 somewhere ?!
asciilifeform: i dun see where drepperism wins in ~any~ version, vs. the ab initio and 10x moar compact musl.
asciilifeform: mircea_popescu: near as i can tell , zcx is a http://btcbase.org/log/2019-02-17#1897558 , it attempts to implement multiprocess on ~all~ machines, incl. ones that dun have any support for interrupts (e.g http://btcbase.org/log/2019-02-12#1895612 )
asciilifeform: mircea_popescu: until diana_coman's test battery, i never even attempted to use the tasking system.
asciilifeform: glibc is simply poison.
asciilifeform: ( half the reason why asciilifeform dug out musl, is that it's compact enuff to be fixable, at least in principle )
asciilifeform: this btw is how asciilifeform implemented phuctor backend.
asciilifeform: ( with the bonus lul, that sometimes somebody walks in with a http://www.loper-os.org/?p=2906 , and then simply has to 'unhappen' )
asciilifeform: mircea_popescu: rright, but that's ~per build~, and ea. build destroys symbols ( in the sense that they got statically pumped into the bin, and no longer 'exported' )
asciilifeform: ( spoiler : heapism per se is a solution to a problem that is ill-posed , and therefore impossible for malloc to be anyffin but an elaborate http://btcbase.org/log/2016-01-21#1379603 )
asciilifeform: ( this is where asciilifeform cannot resist to 'on bolix, you simply ask the box what all symbols are, and can get source for whichever you point at' etc )
asciilifeform: objdump --syms eulora | wc -l
asciilifeform: can be computed at runtime.
asciilifeform: the infestation is deep enuff that you quite obv could not make above example actually build reliably on any extant unix. the entire notion of 'portability' as pictured by unixtards (i.e. powered by #ifdefism) precludes it.
asciilifeform: even e.g. #include <foo.h:e432850de89226c6745301a5932e30c5b09f260b9c850a5e76e8119f66b2f06f1798156138a1741aeff9c46ab90ff1d8ad97d9c089c7d76991a8b7ea8b104bdf> would've been improvement
asciilifeform: from rampant 'substitute unlikes as likes' and onward
asciilifeform: not simply 'fully qualified' , but with knowledge of wtf the item ~is~
asciilifeform still not fully made sense of franco -- d00d was 'fuhrer' by sheer accident, after mola's plane crash , and possibly only won war cuz other side resembled gavin & co, 9000 rando 'hasta y per siempre' derpfactions
asciilifeform: to compare.
asciilifeform: bvt: compiled into noops << where ? in stdlib ? ( or can point in http://btcbase.org/log/2019-02-16#1897234 ?? )
asciilifeform: http://www.loper-os.org/pub/misc/feb_16_adatests_o.html << for utter thread-completeness, the main unit.
asciilifeform: incidentally, re the posted htm, no yer box aint broken, there really aint any clickable links therein (why not? in what's undisputably world's best disasmer ? cuz world's best disasmer is actually not so great , compared to what's obviously possible )
asciilifeform: nao this is not a complete answr to 'what does it do, EXACTLY' cuz that also ropes in the ~2MB of standard lib
asciilifeform: diana_coman: are we sawing'em open because it dun actually abort, or simply to compare with old gnat's sjlj , tho ?
asciilifeform: diana_coman: if , like asciilifeform , you grew up on intel asm syntax, you can objdump -M intel -d proc.o
asciilifeform: ( 1 of those things for which there aint a complete substitute, sadly, if yer doing serious reversing )
asciilifeform: diana_coman: ida, but objdump -d proc.o will do the job likewise ( in slightly gnarlier format )
asciilifeform: diana_coman: could plz tar up that built gnat and post ? i have a suspicion ( to get ljmp with old gnat, i had to set include paths on my box )
asciilifeform: but as i undertand, arm is permitted to 'just barely work', it Would Not Do (tm) for it to compete with intelism on heavy irons , somehow
asciilifeform: i found ~one~, chinese, for ~more~ than the cost of serious opteron with complete filling.
asciilifeform: if arm64 were a completely compat. superset of conventional 'arm7', this would make 0 diff naturally. but seems like it aint
asciilifeform: 'Fabian Vogt 2014-12-12 15:30:53 UTC If sjlj exceptions are not supported for ARM, shouldn't the configure option be invalid for ARM or at least print a warning? If an option does exist and it simply crashes the compiler during make, it's definitely broken and must be removed.'
asciilifeform: this, note, is ~still~ a bug ( or , worse, aarch64 dun know how to longjmp-task ? i.e. not implemented ? ) and will have to be cured. but as i undestand diana_coman is currently interested in x64.
asciilifeform: diana_coman: were you attempting to build ~x64~ gnat on this ? ftr i was never able to ( with classic zxc variant, this was long before the current thread )
asciilifeform empties head.
asciilifeform: for all i know, not even a computable function, lol
asciilifeform: i was answering q of 'why does gcc put out this odd form for a computed goto'.
asciilifeform: mircea_popescu: the 'test' vs 'cmp' thing is microscopic , so arguably red herring here
asciilifeform: ( zcx on left hand, ljmp on right )
asciilifeform: ( btw the obj reason why none of this is visible in ffa -- there aint any tasks! so even in ljmp mode it shits out same coad )
asciilifeform: think, cmp can only work by subtracting
asciilifeform: test simply ORs the bits in register together (and this happens by default when you load it, cuz it's costless electrically)
asciilifeform: mircea_popescu: opposite. cmp is ~3x slower than 'test'
asciilifeform: i sawed it open to try an' see how thefuq the longjmp thing actually worx
asciilifeform: zcx hands-down moar compact
asciilifeform: it's 'optimal' in the sense that it fucks the branch predictor less than an always-computed-jump
asciilifeform: mircea_popescu: it's what gcc does for small (i dun recall threshhold) computed switch
asciilifeform: aa ljmp
asciilifeform: it is only in ljmp-variant
asciilifeform: also loox like the ljmp variant puts abortism stub in erry proc (that appears in a task, that is)
asciilifeform: http://www.loper-os.org/pub/misc/zcx_procs.asm http://www.loper-os.org/pub/misc/ljmp_procs.asm
asciilifeform: cuz it keeps the where to longjmp in'ere.
asciilifeform: i was discussing diana_coman's much earlier empirical find, that on ljmp stack fills faster per same # of calls. this here is why.
asciilifeform: but indeed the ljmp variant craps out slightly bulkier coad across the board
asciilifeform: for thread-completeness : http://www.loper-os.org/pub/misc/zcx_proc_flow.png http://www.loper-os.org/pub/misc/ljmp_proc_flow.png
asciilifeform: longjmpistic:
asciilifeform: diana_coman: btw your 'lick the 9v' intuitive observation earlier was correct, on ljmp variant the default stack frame indeed longer, 184byte vs 40
asciilifeform: diana_coman: the 'zcx' tarball contains a 'ljmp_calls' dir, same as other 1. which is correct ?
asciilifeform: diana_coman: fwiw the noise floor on e.g my test box, is 0.003 (i.e. a proggy with empty main)
asciilifeform: BingoBoingo: as i understand , BingoBoingostan is theoretically idea heatpump country
asciilifeform: electric resistance heat is -ev insanity on pretty much 100% of planet, moar or less 100% of the time -- heat pump (if >0C) or gas (if <0C and there's ~any~ gas to be had , at just about any historical price) wins by fat margin
asciilifeform: ( i'ma take mircea_popescu's word for 'they are available to konsoomer' , evidently currently worth slightly moar as chump bait than as ore )
asciilifeform: and same ones used in nuke and in bk0010 schoolboy comp.
asciilifeform: ( it never wins on cost, in climate where heat pumping worx (avg. >0c) it is used, otherwise fossils win)
asciilifeform: i suspect that 'heat house' aint happening, no matter what level of 'important'. obsolete miners even nao aint worth the cost of transport , floor space, or fan dusting / noise isolation, even to use as heater. whereas 'current' iron is ~unobtainable to commoner on acct of being a strategic good. and can't picture how this would change as 'important' goes up.
asciilifeform: mpleteness
asciilifeform: btw , to go with http://trilema.com/2019/so-what-is-the-man-saying , really oughta disasm a zcx variant and longjmp side by side and see what actually changes. ( when diana_coman comes back with working bins, i'ma set this up , for thread-co)
asciilifeform: http://btcbase.org/log/2019-02-14#1896364 << possibly pertinent detail : on modern irons, long jmp within page seems to take same time as short. so the toy tester may not reveal diff.
asciilifeform: mod6: nope , that line gets printed if kernel compiled with scsi option set, regardless of whether any card.
asciilifeform: mod6: quickest way to learn wtf, i suspect, is to dump the thing bitwise onto a usb, and boot that, then lsmod -v and diff with your known working set's lsmod -v output.
asciilifeform: it comes, in this particular case, from the people who actually do http://btcbase.org/log/2019-02-14#1896130 -- e.g. shituntu -- whose boot log would be 9000km long if ~every~ device that was simply not found, produced an eggog
asciilifeform: simply barfs like depicted in mod6's paste -- 'where the fuck is / ? halted'
asciilifeform: ( otherwise it aint a call, but a 1way ticket, i.e. jump )
asciilifeform: the compiler can slim it down to just the return addr, but the latter goes always
asciilifeform: mircea_popescu: calls dump register snapshot liquishit on the stack, and inevitably ding the cache, whereas loops not
asciilifeform will re-play ffa benchmarks on the longjmp gnat, once the latter's built, but doesn't expect to find any measurable diff
asciilifeform: mircea_popescu: if you give a damn re exception-propagation speed, prolly oughta measure that. otherwise seems like the docs didn't lie, it dun do anyffin to ordinary jump
asciilifeform: the sorta thing that makes e.g. bolixologies, look quite trivial in comparison. ( sad old crate, fits on desk; but try and fit the finns )
asciilifeform: iirc we even had an actual finn here, for maybe 1d. but then promptly sank back to the bottom of the sea .
asciilifeform wouldn't be terribly surprised if one day mircea_popescu digs up the wreckage of an attempted 'trilema' that was run in e.g. classical greek , tho
asciilifeform: ( i suppose could empty it of chix, if there were 9000 mircea_popescu's , but there's exactly 1, at least in my light cone )
asciilifeform: mircea_popescu: these folx seem to have a psychiatrically-classical 'царь-батюшка' imaginary friendship w/ trump
asciilifeform: the only hit is x86_64-linux-musl/share/info/mpc.info -- a readme
asciilifeform: BingoBoingo: after mircea_popescu nitpicks, perhaps this oughta be used as a template for others
asciilifeform not picking on diana_coman out of sadism, simply had nfi how to parse that line
asciilifeform: why would you even compare b/w diff machines.
asciilifeform: whythefuq do they even include, the gimped 'zcx' variant, then.. ( aside from http://btcbase.org/log/2019-02-12#1895611 isms)
asciilifeform spend yrs in the unfortunate company of erry known species of usgtard , and came to conclusion that the principal 'limiting reactant' for defections is... they ain't got where to. ~nobody outside the amerigulag needs ~any of'em for anyffin.
asciilifeform: 'the Cyber Conspirators created a Facebook account that purported to belong to a USIC employee and former colleague of Witt, and which utilized legitimate information and photos from the USIC employee’s actual Facebook account. This particular fake account caused several of Witt’s former colleagues to accept “friend” requests' << lol! defectors aint what they usedto be...
asciilifeform: 'Tarrio believes that unless President Donald Trump steps in, the de-platforming and dehumanizing of conservatives will continue. “He needs to step in, not only because if he doesn’t he will lose in 2020 with all of his supporters being kicked off social media, but because it’s the right thing to do,” Tarrio finishes.' << lol!!
asciilifeform: diana_coman, mircea_popescu : all of above is still just the baseline (no longjmpism) case ?
asciilifeform: what's hard to compute tho. you put sumthing on www, erry 2bit spamplagiarismbot mirrors it in 10sec, not even speaking of actual humans laffing
asciilifeform: (this was prior to mold going public re his 'kompyooting' )
asciilifeform remembers the roosh-v d00d, but for something quite else, iirc it was him who at one pt had asciilifeform's www linked as -- insultingly but lulzily -- 'the mencius moldbug of computing'
asciilifeform: ( so far impression is, roughly similar to '80s su, tho with ~less~ obv. leakage of suppressed laffter through the pg , oddly , i expected moar )
asciilifeform: asciilifeform often points out that we have buffer overflows on acct of the iron having nfi what 'integer', 'character', 'string', 'code' are , just pushes untagged words around; but it is also the reason why we have e.g. thread deadlocks (on an iron which 'knows' when yer e.g. assigning a variable, deadlock is ~impossible)
asciilifeform: mircea_popescu: thinking about it -- 'zxc' strikes me as a classic case of http://btcbase.org/log/2018-10-25#1866191 and impedence mismatch generally. it was clearly written as attempt to 'deterministic scheduler on ??? iron/os', but fails, cuz you can't actually spackle away impedence mismatch b/w the underlying platform and the proggy
asciilifeform: bvt: my reasoning is based on the docs description of how implemented. but certainly worth a test, if anyone can be arsed ( given the 'all loops gotta be doctored' thing, my understanding is that 'polled' proggy would choke if wedges inside a cppism )
asciilifeform: currently i suspect that polling (even if can be made to work errywhere) will be 9000x moar expensive than longjmpism ( the latter only adds to cost of exceptions, the former -- to ~all~ loops )
asciilifeform: maybe musl implements some functionality is a way unexpected << where didja get a gcc5istic gnat built on musl, bvt ??
asciilifeform: BingoBoingo: http://qntra.net/2019/02/madame-secretary-mike-pompeo-declares-huawei-deployments-obstacle-to-us-partnership/#comment-124253
asciilifeform: zcx ? or longjmp ?
asciilifeform: i.e. only affects cpu cost of exceptions, rather than all jmp.
asciilifeform: http://btcbase.org/log/2019-02-12#1895366 << hm what was the runtime ~with~ longjmp ? (or still going ? )
asciilifeform: in e.g. airplane, 'exception' should mean that the entire comp shuts down and transfers control to 1 of the hot spare duplicates.
asciilifeform: i use'em as strictly as equivalent of the red lamp
asciilifeform: ( in ffa, exceptions are a 'catch fire' condition, and drop into the last-chance handler, but in moar complicated proggy, with, say, devices, you may want to actually handle and keep working )
asciilifeform: i.e. try with simply the wrappings.
asciilifeform: mircea_popescu: in that sense ideal test is empty loop, rather than loop containing an item that is simply heavy but not on acct of jumpism
asciilifeform: http://btcbase.org/log/2019-02-12#1895272 << serpent dun have many jumps in it, recall, it's unrolled, even fewer jumps (conditional or otherwise) than ffa
asciilifeform: http://btcbase.org/log/2019-02-12#1895234 << ffa uses exceptions strictly as 'fucking stop whole program (and if it's running on a micro, whole machine, and flash 'dead!' lamp) right nao!' , so won't impact. my understanding is that it'd impact only speed of the ~exceptions~, longjump is slower cuz it crosses pages -- cachaistically.
asciilifeform: for total thread-completeness, there is a workaround that works, but it is imho ugly ( ada 'interrupts' package ), i'ma post example if mircea_popescu et al ask.
asciilifeform: nao ideally would find where the fuck they broke the (implemented!!! in gnat) annex D Right Thing
asciilifeform: and in particular, ln. 1079 of s-taprop-linux.adb , which implements the pthread mechanics behind the gnat scheduler
asciilifeform: turns out that yes implements:
asciilifeform: so then went to the standard docs. annex D ~in fact does~ mandate support for mircea_popescu-style 'halon switch, motherfucker' aborts. so then went to find , does gnat implement ?