log
▁▁▁⏐︎▁▁▁ 3774
mircea_popescu: http://btcbase.org/log/2019-02-15#1897027 << yes, absolutely. ☝︎
a111: Logged on 2019-02-15 17:45 diana_coman: http://btcbase.org/log/2019-02-15#1896926 -> this failed precisely in the same way; need to get to the bottom of it really because Adacore's gnat otherwise IS compiled with sjlj so wtf
mircea_popescu: seems we're en route to discover that all gnat builds depend on ~this one bridge build~ that was done by a meanwhile dead graybeard sometime in 2001.
mircea_popescu: http://btcbase.org/log/2019-02-15#1897029 << myeah, eulora suddenly very fucking useful to teh broader community. i have a good mind to add a few months payroll to the goodwill line, because hory shit. ☝︎
a111: Logged on 2019-02-15 17:46 diana_coman: mircea_popescu, my current appreciation is of the sort (tower of shit inside tower of shit )^infinity
mircea_popescu: http://btcbase.org/log/2019-02-15#1897049 << pretty sure there exist aarch64 sjlj compiled kernels. ☝︎☟︎
a111: Logged on 2019-02-15 18:36 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.
mircea_popescu: http://archive.is/x8D5k << apparently they STILL have the issue in gcc 8. ☟︎
mircea_popescu: ie, gcc broken for long jump on aarch64.
mircea_popescu: "The issue is that the code generated for __builtin_longjmp reads a value for x29 (the frame pointer) from the jmp_buf, but the code generated for __builtin_setjmp doesn't actually write x29 to the jmp_buf, leading to corruption of x29 when a longjmp occurs. ☟︎
mircea_popescu: "
asciilifeform: mircea_popescu: i sank several hours into the dig and did not find whether 'worked, then broken' or 'never written'
asciilifeform: http://btcbase.org/log/2019-02-16#1897094 << if you actually found this kinda thing, plox to link ( this'd mean the former ) ☝︎
a111: Logged on 2019-02-16 00:28 mircea_popescu: http://btcbase.org/log/2019-02-15#1897049 << pretty sure there exist aarch64 sjlj compiled kernels.
mircea_popescu: i can't imagine it was never written, this long jump thing was standardf for a while neh ?
mircea_popescu: im lookin im looking
asciilifeform: mircea_popescu: sjlj ? was 'deprecated' Officially, Only A Terrorist(tm) Would Want, was the flavor i got from the http://btcbase.org/log/2019-02-15#1897058 rotters ☝︎
a111: Logged on 2019-02-15 19:15 asciilifeform: 'Jonathan Wakely 2014-12-12 15:35:00 UTC Just don't do silly things. Just because an option exists doesn't mean you need to use it.'
asciilifeform finds himself nao also wondering ~what else~ dunwork on arm
asciilifeform: ( e.g., ffa, built yes on arm64 using ave1's system, but it's spartan to the point of starvation featurewise)
mircea_popescu: yes, but also for say mid 90s to maybe 2000something it was the way to do handling neh
asciilifeform: arm64 iirc first released in late obummer reign
mircea_popescu: so you're saying "arm came too late, sjlj already indexed" ?
asciilifeform: 30 October 2012 apparently.
asciilifeform: dig through gnat/gcc sources seems to support hypothesis :
asciilifeform: gcc/common/config/arm/arm-common.c: /* Honor the --enable-sjlj-exceptions configure switch. */
asciilifeform: but nuffin re sjlj in gcc/common/config/aarch64/* .
mircea_popescu: but i mean, eg ida got support for it cca v6 ?
asciilifeform: if arm64 were a completely compat. superset of conventional 'arm7', this would make 0 diff naturally. but seems like it aint
mircea_popescu: hm.
asciilifeform: mircea_popescu: ida6 only ate 32bit arm
asciilifeform: 64 appeared iirc in 6.3 .
mircea_popescu: asciilifeform i suspect this is my problem. i don't mentally distinguish aarch64 from arm correctly.
asciilifeform: they're as diff as x86 an' x64 ☟︎
asciilifeform: lotsa subtle breakage, cuz the orig wasn't built by thinkin' folx and hence not designed to be 'stretched'
asciilifeform: this might also explain why adacorpse Officially distributes arm7 gnat but not arm8 (64)
asciilifeform: i.e. the latter is pre-alpha quality
mircea_popescu: myeah.
mircea_popescu: http://btcbase.org/log/2019-02-15#1897069 << keks. ☝︎
a111: Logged on 2019-02-15 20:02 diana_coman: darn, ofc there are TWO flags to set to the same thing i.e. one for bootstrapper and one for the resulting gnat
asciilifeform wonders why arm64 even exists, not to mention displaced arm7, cuz afaik you can't actually ~buy~ an arm8 box with >4G of ram in it
asciilifeform: for any price
asciilifeform: iguess you get 64-bit arithmetics.
asciilifeform: but that's pretty much all
asciilifeform: ( arm7 == last 32bit , arm8 == arm64 aka aarch64 in gnutardspeak )
mircea_popescu: yet as per http://btcbase.org/log/2019-02-06#1893295 ... ☝︎
a111: Logged on 2019-02-06 15:44 asciilifeform: mircea_popescu: there's not so much win from 64 on lappies where max ram e.g. 3GB.
asciilifeform: mircea_popescu: thread was re mid-2000s lappies. ( i have moar recent ones, which have e.g. 32GB in'em etc )
mircea_popescu believes the tb ram arm is one of them http://btcbase.org/log/2019-02-01#1891303 ☝︎
a111: Logged on 2019-02-01 02:08 asciilifeform: or the pumping stations ( unmarked on maps, cuz 'strategic', but quite visible from street )
mircea_popescu: asciilifeform point being, small ram 64 bit not particularly useful.
asciilifeform: mircea_popescu: not even 'tb', but go an' try to find on market arm box of ~any~ description that eats ordinary socketed dimms ☟︎
asciilifeform: i found ~one~, chinese, for ~more~ than the cost of serious opteron with complete filling.
mircea_popescu: right ?
asciilifeform: it's outrageous
asciilifeform: afaik nuffin keeps standard dimms from working with arm8, any of'em
asciilifeform: the electricals are exactly same as for the soldered-down crap
asciilifeform: mircea_popescu: i would not go so far as to say 'entirely useless', 2ce the bitness gets you 2ce the rsa speed for same clock.
asciilifeform: but still frustrating.
mircea_popescu: cursory familiarity with the state of the computing stack (and, implicitly, the utter ridiculousness of any wasp-remnant notions of "the public oppionion" and "wouldn't permit" etc) would have saved that white haired tard a lot of unplesantness down in ecuador embassy.
mircea_popescu: because FUCKING OBVIOUSLY it dun work worth a shit nor does it do anything, practically speaking. and if it doesn't here, it doesn't re "education" or "politics" or "civil society" or what have you.
asciilifeform: recall http://btcbase.org/log/2017-03-10#1624158 + http://btcbase.org/log/2017-10-02#1719717 ☝︎☝︎
a111: Logged on 2017-03-10 14:33 asciilifeform: http://btcbase.org/log/2017-03-10#1624146 << alpha was already half-dead when hp swallowed compaq, in '02. iirc compaq shitburied it in '01 and sold all rights to intel, who proclaimed the arch now known as 'itanic' as its replacement.
a111: Logged on 2017-10-02 19:29 mircea_popescu: o, speaking of "women in tech" : after thoroughly wrecking yahoo (eerily reminiscent of carly fiorina's "performance" at meanwhile disappeared compaq/hp), melissa meyer disappeared from view ; leaving verizon to pick up teh pieces.
asciilifeform: usg.dept-of-microshit systematically exterminated all the sane archs in late '90s. arm was introduced (from pile of shit 'acorn' in uk) as an 'unprincipled exception' , for pnojeism, which -- try as one might -- dunwork so much on x86 w/ any battery anybody's yet devised
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: ( leaving aside the fact that it is a shit arch on multiple levels of design, much like x86, only real diff is that it dun come with 35 yrs of legacy crapola )
asciilifeform: for history aficionados, the opening shot of that slaughter was when sgi was usgized and permitted to swallow cray, and shortly after proclaimed that 'crays' henceforth would be made of pentiums
asciilifeform: promptly folded, after pumping remaining moneys into intel pocket
asciilifeform: this killed the (very reasonable) 64bit mips.
asciilifeform: then they proceeded to wipe out alpha , then hp ( also quite reasonable 'pa-risc' 64-bit irons, i had one ) .
asciilifeform: all to try an' prop up the wintel nonsense hegemony.
asciilifeform: note that at the time of death, each and erry one of these archs wiped the floor with intel's offering, esp. per-watt
asciilifeform: usg as early as in 1980s very muchly disliked the existence of multi-vendor high-perf cpu. was too hard to keep from being sold to untermenschen, see.
asciilifeform: phunnily enuff, there is record that chairman of nsa filed an Official protest during the destruction of orig. cray co.
asciilifeform: ( at the time nsa was single largest operator of crays )
asciilifeform: phat lotta good it did, tho.
asciilifeform: idjit squid swallows own tentacle.
asciilifeform brb,meat
mircea_popescu: asciilifeform see, the cardboard "entrepreneurs" don't exist in the us like in the orcistans, as named individuals. they're named... legal persons.
mircea_popescu: but everything works exacrtly the same outside of this one detail, specifically as in http://trilema.com/2018/you-know-who-the-best-us-president-was-how-about-andrew-johnson/#selection-317.267-317.1139
mircea_popescu: so yeah, much like retarded russkis go "oh, we too businessmen, gave $dork state income, now he can buy suit" ( http://btcbase.org/log/2018-03-21#1788430 finest example), the ustards are all "oh, gave computing to wintel, can has tech!!!" ☝︎☟︎
a111: Logged on 2018-03-21 16:44 mircea_popescu: "Like a bad ass she is, she proved to be wise beyond her years by standing head and shoulders (Or is it ass-tall) with the boys." doesn't even INTEND irony, you realise.
mircea_popescu: meanwhile -- no they don't have any tech, much like the zimbuzz dun have music.
mircea_popescu: hey ave1
ave1: goodmorning, I was digesting te logs
ave1: and read all the gnat problems
ave1: I was working on getting a cuntoo up (which is going slow, I need to relearn to build a kernel, this used to be a lot easier 20 years ago)
ave1: I'll park that and start running the gnat build script.
mircea_popescu: ave1 did you ever run a sjlj version ?
ave1: not yet, I'm now trying to get the build up and running
mircea_popescu: ah. ok.
mircea_popescu: well, if you don't have experience with it or anything, it's not urgent -- can continue with cuntoo, can switch to this, either way.
ave1: diana_coman, re http://btcbase.org/log/2019-02-15#1896922, all the links worked this morning, could you tell me which one failed for you? ☝︎
a111: Logged on 2019-02-15 16:12 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
ave1: O wait, I tried with: http://ave1.org/2018/building-gnat-on-musl-no-more-usrincludex86_64-linux-gnu/
diana_coman: ave1, it was the link that downloaded the gcc src and stuff for adacore2016 ; after that first fail I just copied my own stuff into tarballs/ , packed properly and never really bothered with the rest i.e. it builds all offline now anyway
diana_coman: update re builds: it built fine with --enable-sjlj-exceptions in place, checked it in the log and yes, it's set; but the result still seems to be build ada code with zcx in fact (i.e. my test code with tasks is STILL hung waiting for them to abort) and if I specify --rts=sjlj to gprbuild it complains that there is no native compiler for ada and so can't do anything; ftr I compared the dirs of my adacore install and it has this specific dir
diana_coman: rts-sjlj with its own adainclude/ and so on; so I am back to my previous obv that this flag per se is apparently not enough
diana_coman: when I say "it built" above I mean specifically those versions: http://ave1.org/2018/building-gnat-on-musl-no-more-usrincludex86_64-linux-gnu/ and http://ave1.org/2018/building-gnat-on-musl-now-with-partial-and-parallel-build-support/ i.e. both static only and previous
ave1: diana_coman, in system-linux-*.ads (in gcc/ada directory) you'll find the line: ZCX_By_Default : constant Boolean := False; ☟︎
ave1: I'm currenlty building with it set to False, but it is True in the distributed files
bvt: hello. ave1: doesn't makefile.in in gcc-interface tweak it with sed to the necessary value? ☟︎
diana_coman: ave1, ah, thanks! I'll set it to false and try again then
spyked: http://btcbase.org/log/2019-02-16#1897122 <-- the armv7-armv8 gap is even bigger than x86-x64. armv8 is an entirely new arch, also has 32-bit mode (called aarch32), but that's completely backward incompatible with armv7 ☝︎
a111: Logged on 2019-02-16 00:53 asciilifeform: they're as diff as x86 an' x64
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 bbl in 2-3h
mircea_popescu: http://btcbase.org/log/2019-02-16#1897191 << it should, yes. ☝︎
a111: Logged on 2019-02-16 10:07 bvt: hello. ave1: doesn't makefile.in in gcc-interface tweak it with sed to the necessary value?
diana_coman: oh hey, apparently ave1's find was spot on!
mircea_popescu: nice.
diana_coman: after all the struggle I'm still looking suspiciously at it, lol
asciilifeform: guten morgen diana_coman , mircea_popescu , et al
asciilifeform: diana_coman: oh hey it runs ?
BingoBoingo: Buenas tardes asciilifeform
mircea_popescu: diana_coman does it actually produce sjlj code ?
asciilifeform: http://btcbase.org/log/2019-02-16#1897170 << they have both types ( bezos et al ) neh ☝︎
a111: Logged on 2019-02-16 02:32 mircea_popescu: so yeah, much like retarded russkis go "oh, we too businessmen, gave $dork state income, now he can buy suit" ( http://btcbase.org/log/2018-03-21#1788430 finest example), the ustards are all "oh, gave computing to wintel, can has tech!!!"
diana_coman: asciilifeform, how did you vivisect that pair of binaries? I'd rather do it to see what came out since otherwise I have only this "oh, now it DOES kill tasks!"
asciilifeform: diana_coman: ida, but objdump -d proc.o will do the job likewise ( in slightly gnarlier format )
mircea_popescu: BingoBoingo do you by any chance recall the princeton "trained psychologist" on okcupid who couldn't spell but was nevertheless doing diagnosing over the phone ?
BingoBoingo: mircea_popescu: I'll have to dig, but it sounds very familiar
asciilifeform: diana_coman: if they dun abort and you want another detailed vivisect, feel free to post'em, i'ma rip'em open after tea
asciilifeform: ( or even if they work )
mircea_popescu: asciilifeform i think she means to learn how it's done.
asciilifeform: mircea_popescu: i hesitate to addict folx to ida, it's a gnarly ball of proprietary liquishit, and needs a toilet box with 'wine' to run etc
asciilifeform: ( 1 of those things for which there aint a complete substitute, sadly, if yer doing serious reversing )
mircea_popescu: diana_coman it's of course cheaper to just ship the object files to him
diana_coman: mircea_popescu yes + I must say I like the idea of "rip'em open after tea" - sounds great; but other than that: I'll try objdump too since after this gnat-adventure I'd rather have some way to look inside it
asciilifeform: diana_coman: if , like asciilifeform , you grew up on intel asm syntax, you can objdump -M intel -d proc.o
asciilifeform: diana_coman: are we sawing'em open because it dun actually abort, or simply to compare with old gnat's sjlj , tho ? ☟︎
asciilifeform brb,tea
diana_coman: asciilifeform, they abort; we are sawing them open because I don't trust it anymore re "what does it actually do exactly"
diana_coman: i.e I want to see what did it do
mircea_popescu: it has lied before...
BingoBoingo: mircea_popescu: http://trilema.com/wp-content/uploads/2016/01/how-do-you-spel-illiterate.png linked from http://trilema.com/2017/what-are-you-being-distracted-from/#selection-387.0-411.1 ??? ☟︎☟︎
mircea_popescu: BingoBoingo found it http://trilema.com/2016/usg-in-btc-a-history-of-constant-failure/#selection-39.0-39.51 << ☟︎
mircea_popescu: o damn.
mircea_popescu: she was from harvard not princeton! my bad.
BingoBoingo: AHA, the perils of searching for keywords so strongly associated with constant failure
mircea_popescu: you know ? but ty!
BingoBoingo: Princevard, harMIT, they all blend together
diana_coman: asciilifeform, here it is: ossasepia.com/available_resources/sjlj_testtasks.tar.gz
diana_coman: I added a delay in there before the final check of tasks
asciilifeform: diana_coman et al : http://www.loper-os.org/pub/misc/feb_16_tasks_o.html ☟︎☟︎
asciilifeform: ^ near as i can tell, it uses exactly same scheme as the old gnat's sjlj-mode handlers
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: but the proggy .o seems to be exactly as expected ( if diana_coman has time, can try with the old gnat and see if in fact bitwise-same or not )
asciilifeform still wakin' up, misread http://btcbase.org/log/2019-02-16#1897225 as 'illiterature' ☝︎
a111: Logged on 2019-02-16 17:10 BingoBoingo: mircea_popescu: http://trilema.com/wp-content/uploads/2016/01/how-do-you-spel-illiterate.png linked from http://trilema.com/2017/what-are-you-being-distracted-from/#selection-387.0-411.1 ???
mircea_popescu: just about.
mircea_popescu: the "textbooks" this generation of mostly female, mostly useless college kids don't read are very fucking different from the textbooks we didn't read twenty years ago.
asciilifeform still got good # of the latter
mircea_popescu: as per the "this is the shop where we don't have talapia. the shop where they don't have salmon is down the street"
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: ( see also e.g. http://btcbase.org/log/2016-04-22#1455082 thrd ) ☝︎
a111: Logged on 2016-04-22 01:10 asciilifeform: ida is a particularly interesting case because it is a TOTAL monopoly
asciilifeform: ( to add insult to injury, erry time i post 1 of these, i gotta manually switch font to monospace and remove the winturd line endings, cuz the joys of closedsores proggy )
asciilifeform: http://www.loper-os.org/pub/misc/feb_16_adatests_o.html << for utter thread-completeness, the main unit.
asciilifeform: diana_coman et al : plox to lemme know if yer interested in magnifying glass treatment of some particular aspect.
asciilifeform: .text:00000000000005B8 call tasks__abort_tasks
asciilifeform: ^ the magic moment.
asciilifeform: ^ btw diana_coman if you build with -d , the var_xxx liquishit can be made into varnames.
asciilifeform: ( ida eats 'dwarf' debugolade )
asciilifeform: phunphakt : ada std includes a pragma that manually throws a string for bin auditor to read, in selected spot.
asciilifeform: ( i dun currently use, but it's definitely going into the final revs of ffa )
feedbot: http://bimbo.club/2019/02/philosophical-transactions-for-the-months-of-july-august-and-september-1716-part-v/ << Bimbo.Club -- Philosophical Transactions. For the months of July, August and September, 1716. - Part V.
asciilifeform: http://btcbase.org/log/2019-02-16#1897225 << the former link eggogs, but i think this is by design, on acct of mircea_popescu's anti-hotlinker pill ☝︎
a111: Logged on 2019-02-16 17:10 BingoBoingo: mircea_popescu: http://trilema.com/wp-content/uploads/2016/01/how-do-you-spel-illiterate.png linked from http://trilema.com/2017/what-are-you-being-distracted-from/#selection-387.0-411.1 ???
mircea_popescu: asciilifeform shouldn't eggog, should redirect you to explainy.
asciilifeform: mircea_popescu: seems to redirect to random article
asciilifeform: ( try it, by clicking from phf's log )
mircea_popescu: well, i did what the article proposes, set my referrer to trilema permanently, so i always see trilema pics.
mircea_popescu: asciilifeform check it out, my htaccess got fucked meanwhile.
mircea_popescu: should be fixed now.
asciilifeform: loads via the article, as intended
mircea_popescu: yeah, but if you try it solo should take you to http://trilema.com/2019/trilema-images-no-longer-showing/
mircea_popescu: not to a random article.
asciilifeform: nao worx
diana_coman: asciilifeform, thank you! and yes, I know it's not enough to know what it does exactly but it's still useful.
asciilifeform: yw diana_coman
feedbot: http://trilema.com/2019/get-lost-dumbo/ << Trilema -- Get lost, dumbo!
mircea_popescu: asciilifeform http://p.bvulpes.com/pastes/o3YKv/?raw=true
asciilifeform: mircea_popescu: got it ( and iirc was linked recently in trilema )
asciilifeform bbl, winching xray box onto pedestal and etc
asciilifeform: http://trilema.com/2019/get-lost-dumbo/#selection-1777.110-1777.165 << wat's with all the 'engaged' chix still hanging on fatlife, gotta wonder
mircea_popescu: they lie. part and parcel of http://trilema.com/2017/the-boundless-burden/ is that she ain't ever gonna be as engaged as you think she is, nor as on fetlife as you think she is, nor etc.
bvt: http://btcbase.org/log/2019-02-12#1895598 << i'd like to briefly report on this: at this point i'm sure that there is a race condition. i'll do a post tomorrow morning on it, but short summary is ☝︎
a111: Logged on 2019-02-12 23:36 mircea_popescu: but seems he ALSO found a race condition in the handlers ?
bvt: unwinding stack for exceptions requires taking locks https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=libgcc/unwind-dw2-fde.c;h=24b4ecee68c17e1701c4482580e449b03a4e6fe9;hb=HEAD#l1046
bvt: which is broken with static linking due to usual 'creativeness' of gnu folx https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=libgcc/gthr-posix.h;h=88cbc23937ec20b15b35c5adb7f9983282c6f084;hb=HEAD#l247
mircea_popescu: cool.
bvt: http://bvt-trace.net/src/test_task_exceptions.tgz - test source for those who want to test. to be run as 'while true; do echo -n .; ./adatests >/dev/null || break; done' -- should not take too long to have it segfault.
bvt: (still tested only with gnat2017, but this is a different story; i see no reason to believe that ave1gnat does not have the same issue)
mircea_popescu: me either.
bvt: i.e. with static linking, all locks are compiled into noops.
bvt: i still have no solution for this, afaik musl authors solved the problem for fortran and c++, but gnat seems to lack equivalent knob
mircea_popescu: how did they solve it for c++ ?
bvt: they disabled weak symbols for c++/fortran components, http://port70.net/~nsz/musl/gcc-5.3.0/0004-gthr.patch
bvt: however i don't understand if this is possible only to achieve only for gnat components, and when i do it globally, i get 'undefined symbol' when linking with libgcc
asciilifeform: bvt: compiled into noops << where ? in stdlib ? ( or can point in http://btcbase.org/log/2019-02-16#1897234 ?? ) ☝︎
a111: Logged on 2019-02-16 17:53 asciilifeform: diana_coman et al : http://www.loper-os.org/pub/misc/feb_16_tasks_o.html
asciilifeform suspects this is a gcc5ism