log☇︎
42400+ entries in 0.026s
asciilifeform: http://www.loper-os.org/pub/amd64_mulx.jpg << amd seems to support ~currently~, but dun say from what date of chip specifically
asciilifeform: bvt: the only new instrs that seem to be even theoretically of use, are 'mulx' and 'adcx' -- but i dun have any iron that supports these atm, and cannot even begin to say whether constant time etc
asciilifeform: where yes it'll do a e.g. 512-bit add, but -- evidently -- using ~existing~ regs, and putting out same (or greater) amt of heat, and locking up the pipe
asciilifeform: i.e. they can't stuff any moar transistors in there, and end up offering the equiv. of ye olde 'winmodem'
asciilifeform: in lulz inspired by bvt's article, asciilifeform went and dug re 'modern' cpu arithm instructions, and found https://lemire.me/blog/2018/04/19/by-how-much-does-avx-512-slow-down-your-cpu-a-first-experiment/ << intel's crud apparently ~drops frequency~ if you use'em , ultimately nuking all gains from doing so ( they want you to use, so as to shit out binaries that crash on amd, but really gains 0 )
a111: Logged on 2018-12-03 21:48 diana_coman: hm, if it's indeed the tmp thing, it might be worth a try to press vtools to current leaf (i.e. vtools_tempfile_standalone or _notmp) and see if that cures it; my archive contains pressed vtools to ksum patch only, not further
trinque: http://btcbase.org/log/2018-12-03#1878057 << I am currently running a build with a vdiff pressed to same. The only difference is that I have altered the gpr files to statically link. ☝︎
trinque: the remainder of work here is resolving this issue (I have not had) with paths, after which we can start producing ebuilds for novel republican work atop the genesis. ☟︎
trinque: the paths ending up in your genesis vpatches are hard to blame on my script, rather than a difference between our vdiff executables.
asciilifeform: it is such a retarded design that even intel and microshit ~tried to escape~, in 1990s. but nsa decided that it ~likes~ x86 , and for it to remain cemented standard , on acct of http://www.loper-os.org/?p=1299 .
trinque: diana_coman, other folks that have cuntooed, can y'all confirm that the paths that ended up in your genesis.vpatch do not in fact exist? I'd like you to reproduce the commands starting at line 114 of scripts/make_portage_tree.sh in your build directories, i.e. cd ~/src/cuntoo/build/cuntoo and then run them, as root ☟︎
asciilifeform: aaand this is not even to mention their seekrit 'optimizations' behind the scenes.
asciilifeform: when you build 1 of these things, there's a set of decisions that end up determining shape of whole thing; and it so happens that intel made ~all~ of the most retarded possible choices. ☟︎☟︎
asciilifeform: ( and this is not even touching the subj of the tlb cache, which is ~1/3 to half of those transistors , which we have on acct of the idjit paging scheme )
asciilifeform: if you ever wonder why your x64 iron draws 50x the wattage to do same thing as e.g. rk, wonder no longer -- the insanity where shit gets moved around to accomodate idjit instructions with fixed in/out hoppers, the insanity where you gotta set prefixes to specify what width ~each operand~ is (why this is needed ? srsly) , all of this adds up to 3bil transistors that heat the room ☟︎
asciilifeform: btw, bvt , rax etc. ~are~ encoded as 1-8, the iron dun see reg names at all, the classic names are a convention of the asmers and the vendor docs. and imho remains on acct of the asinine x86isms like MUL which use fixed input and output regs, makes'em slightly easier to remember. ☟︎
mircea_popescu: ftr, warlock has the stronger troops for large maps ; sorceress for medium. if you're looking for a shortcut to beating his ass.
lobbesbot: asciilifeform: The operation succeeded.
asciilifeform: !Q later tell bvt http://bvt-trace.net/2019/03/ffa-chapter-9-homework-comba-in-x86_64-assembly/comment-page-1/#comment-12
asciilifeform: spyked et al : http://nosuchlabs.com/pub/gutentext.tar.xz << mirror (replaced the old tarball) ☟︎
a111: Logged on 2019-02-07 19:12 asciilifeform: diana_coman: np. lemme know which flavours the kid ended up liking, i may have moar in the depths of the warez chests along those lines.
diana_coman: http://btcbase.org/log/2019-02-07#1893431 -> kid settled atm on heroes of might and magic II and can think of nothing else! he choose Necromancer ("which one has stronger troops?") and so he is working out ...percentages to know how many skeletons he should get. ☝︎
mircea_popescu: spyked prolly should redirect to explanation tho. 404 is ambiguous here ☟︎
asciilifeform: ( why ? asciilifeform , for instance, sometimes gotta switch gnats , and the emacsen dun see it )
asciilifeform: meanwhile, in unixtardation : apparently there is no known way to change PATH envir. var. for ~running~ processes globally
asciilifeform: spyked: aaa, seems to work, ty
spyked: asciilifeform, I applied mircea_popescu's pill to block spam-linkers. pl0x to set referrer to http://thetarpit.org or http://lucian.mogosanu.ro
asciilifeform: spyked: download link is broken, redirects to 404
feedbot: http://thetarpit.org/posts/y05/088-gutenberg-iv.html << The Tar Pit -- Gutenberg ASCII archive updated, now with 0.5% less junk
trinque: this some serious dick-cheese and cum-socks tier c, gotos and all.
hanbot: BingoBoingo tyvm!
trinque: your asking did however send me on a spelunk through scripts/kconfig/conf.c by way of ./scripts/kconfig/Makefile
trinque: asciilifeform: that gentoo-specific patchset question is because these patches are present in "gentoo sources"
lobbesbot: BingoBoingo: The operation succeeded.
BingoBoingo: !Qlater tell hanbot http://p.bvulpes.com/pastes/yRAfP/?raw=true reset
asciilifeform: trinque: outta curiosity, hiw didja deduce this ? afaik barf is same in both scenarios?
a111: Logged on 2019-03-08 18:01 asciilifeform: hanbot: that answers it then. that one contains some flag that trinque's kernel has nfi what to do with.
trinque: http://btcbase.org/log/2019-03-08#1900931 << it's actualy the opposite. kernel makefile found that .config lacked a value for a *new* key present in the trinque-supplied kernel, so asked for a value, suggesting default. ☝︎
asciilifeform: gotta try this
a111: Logged on 2019-03-08 18:15 asciilifeform: why linus et al did not supply a ready-built http://btcbase.org/log/2019-03-08#1900938 util, remains a mystery to asciilifeform , just as he found it mystery in '90s
BingoBoingo: Eh, I'm still up. Will put the shoes back on and remember to open the terminal before I take shoes off text walk
BingoBoingo: hanbot: Is it alright if the reboot waits until morning?
hanbot: BingoBoingo please reboot me yetagain, with my compliments to your soles.
asciilifeform: it was 'designed' like dog 'designs' turd as it emerges from arse.
asciilifeform: mircea_popescu: if some part of the shitstack 'was designed', i cannot say what it was
asciilifeform: 'ok, ctrl-c? i'll assume you meant the last of the 9000 jobs you just spawned, right'
asciilifeform: e.g. you ctrl-c and it nukes ~child~ of process however many layers deep, as in hanbot's example, and instead of propagating the kill up the tree, 'returns' barf uptree
asciilifeform: whole concept of 'current task' is perlesque liquishit
mircea_popescu: time and again we're back to ye olde "oh, line, you mean paragraph" pile of accidentally-grown-never-designed festering sores.
mircea_popescu: this is fucking stupid all through.
mircea_popescu: which is JUST AS BROKEN BY DESIGN. ctrl-d will end a file AND kill a terminal, but not touch a task. ctrl-c will kill n levels of task depth, as in her example, FOUR. ☟︎
mircea_popescu: in more practical sense : ctrl-C is broken-by-design, in that it is ambiguously used to either mean "kill this task" or "kill all tasks". note how ctrl-c doesn't kill THE TERMINAL. it needs ctrl-d for that.
mircea_popescu: fuck that dumb shit.
mircea_popescu: dude, this "x distance apart" is like the fucking speed bumps. brilliant fucking idea, "how about everyone pays a tax so that some edge cases benefit"
asciilifeform: errything else seems to require fiddling.
asciilifeform: most complicated mechanism i ever used that 'works' , in light of mircea_popescu's concept thereof, is.. idk, kalash
asciilifeform: ( the subj of microshit, i'll spare, tho could expound.. )
asciilifeform: tho it aint clear to me than the e.g. crapple rituals, make any moar sense
asciilifeform: hanbot: mine are min. of 1m apart always, for this reason. this dun mean that mircea_popescu is wrong tho, re opensores.
hanbot: i dunno, shit's supposed to fail if you do the wrong thing. possibly i'm a moron for making it too easy to do the wrong thing, keyboards oughta be 10ft apart, give second to surgical assistant, whatever. but it is fucking frustrating.
mircea_popescu: in this very specific sense, "free and open source" is WAY worse than microsoft apple etc.
mircea_popescu: there's a difference between selecting for intelligent users and selecting for inept hoop jumping acrobatic users -- and "foss" managed to implement the 2nd kind while loudly making postureclaims as to everything under the sun.
mircea_popescu: hanbot this sounds like EXACTLY what the problem has been, all these years.
hanbot: which means i have yet another incomplete cuntoo install with no vpatch and yet no way to start again without another freakin' reboot. it's like walking a tightrope, here i was worried the power'd go out or w/e, but no, fingers got confused after 30m of back and forth, fuck me.
a111: Logged on 2019-03-08 18:06 asciilifeform: hanbot: see if you can build using http://nosuchlabs.com/pub/conf_current.conf , it is an asciilifeform-sewn config for the specific iron you happen to be on
hanbot: so there i am merrily entering choices into the kernel config's endless querying on the tail of my cuntoo installation attempt, and i'm noting down all the modules it asks about that don't appear in http://btcbase.org/log/2019-03-08#1900940 for posterity, during which i ctrl + c on the wrong of the two keyboards i'm handling while trying to copy a module name and bam, make dies, which means kernel configuring dies, which means bootstrap.sh dies, ☝︎☟︎
asciilifeform: prior to usg-engineered 'international' blockage, ~north~ was the industrial kr.
BingoBoingo: They ate the battery potatoes?
BingoBoingo: Well, when you put a third of your generation capacity in the same damn dam...
BingoBoingo: Well, she had about 45 minutes of mains current today
BingoBoingo: asciilifeform: She's still there. Can't help people with things they refuse help for.
BingoBoingo: Venezolana today during spanish class logistics talk made of a point of how maduro was blaming the "Imperialists" for the blackout. I offered that it was almost certainly the imperialists, breaking shit is part of their regime change recipe.
feedbot: http://bingology.net/2019/03/08/ceviche-theory-and-practice/ << Bingology - BingoBoingo's Blog -- Ceviche Theory And Practice
a111: Logged on 2019-03-08 22:10 mircea_popescu: http://btcbase.org/log/2019-03-08#1900912 << check it out, bolivar socialism and roosevelt socialism competing on ye olde marotes, "who brings true light" etcetera.
asciilifeform: http://btcbase.org/log/2019-03-08#1900966 << not clear to me that 'isms' are in play : afaik it's moar of the thing where the anglos have same relation to other folx as worm to apple ☝︎
asciilifeform: http://btcbase.org/log/2019-03-08#1900962 << near as i could tell, subj is a kind of http://btcbase.org/log/2017-08-03#1693150 , i.e. junkyard dog ☝︎☝︎
asciilifeform: guten tag mircea_popescu !
a111: Logged on 2019-03-08 18:16 asciilifeform: the obv hypothesis is that the gnutards ended up happy with the position where 'priesthood' builds kernels and users blindly eat the binaries, etc
mircea_popescu: http://btcbase.org/log/2019-03-08#1900954 << seems quite evidently a case of "here in orc cvasirepublic, we rely on implicits to do the job where understanding wtf's going on would have been useful". ☝︎
a111: Logged on 2019-03-08 18:03 asciilifeform: this is prolly not an answr that will make hanbot happy. the only good news is, this only gotta be done 1ce per kernel-iron combo.
mircea_popescu: most amusing element -- neither of the "parties" involved read as much as a line of ye olde menshevik-bolshevik disputes.
a111: Logged on 2019-03-08 16:44 asciilifeform: 'US Secretary of State Mike Pompeo said Maduro’s government only has itself to blame for power shortages in Venezuela, noting sarcastically that “Maduro’s policies bring nothing but darkness.”' << lol!!
mircea_popescu: http://btcbase.org/log/2019-03-08#1900912 << check it out, bolivar socialism and roosevelt socialism competing on ye olde marotes, "who brings true light" etcetera. ☝︎☟︎
a111: Logged on 2019-03-08 10:50 spyked: moreover, the same item will be useful for adding comments to thetarpit's cl blogotron, implementing mp-wp-anti-spam for the same, etc.
mircea_popescu: http://btcbase.org/log/2019-03-07#1900888 << very best korea flavour to the whole thing. ☝︎
mod6: hanbot: I did see these questions pop up too while doing the cuntoo bootstrap. This in particular happened to me after I copied my working gentoo kernel from /usr/src/linux/.config to cuntoo/myCopiedConfig, then replaced all of the =m with =y ( sed -i 's/=m/=y/' myCopiedConfig).
asciilifeform: the obv hypothesis is that the gnutards ended up happy with the position where 'priesthood' builds kernels and users blindly eat the binaries, etc ☟︎
a111: Logged on 2019-03-08 18:05 asciilifeform: iirc trinque had an experimental proggy that takes a running kernel's lspci / lsmod output, and diddles an existing config so as to give a kernel that boots on $iron . but afaik not yet published.
asciilifeform: why linus et al did not supply a ready-built http://btcbase.org/log/2019-03-08#1900938 util, remains a mystery to asciilifeform , just as he found it mystery in '90s ☝︎☟︎
asciilifeform: sadly it's how the current ball o'crapola worx.
asciilifeform: since mircea_popescu is asleep, i'ma 'fill in' for him and note ftr that the necessity for this process is indeed a warcrime of the gnutards, and really not how sane configurator of any kind oughta work
lobbes_field: ^^ was the approach I took. Accepted defaults n' manually walked it after I confirmed it booted
asciilifeform: ( i.e. it ~may~ end up containing useless modules, that you will later want to cut out, but ~will~ contain the necessaries given in the supplied config )
asciilifeform: hanbot: if it ends up asking you q's , like prev. time, feel free to press 'enter' and let it give the default answers -- they come from the existing config (when available) and from default of the given kernel's (if not) . this generally gives bootable kernel, that you can trim later at your leisure.
asciilifeform: ( and of course the snsa-operated dulap )
asciilifeform: all pizarro-owned x86 boxen in the rack, are currently using this config
asciilifeform: ( trimmed manually )
asciilifeform: hanbot: see if you can build using http://nosuchlabs.com/pub/conf_current.conf , it is an asciilifeform-sewn config for the specific iron you happen to be on ☟︎
asciilifeform: ( if you were to take trinque's example config, your kernel will boot but will panic when looks for root partition, as it won't find that iron's raid card )
asciilifeform: iirc trinque had an experimental proggy that takes a running kernel's lspci / lsmod output, and diddles an existing config so as to give a kernel that boots on $iron . but afaik not yet published. ☟︎