tmsr - Log 2019-07-24

asciilifeform: meanwhile, in measurements : performance of http://www.loper-os.org/?p=3420 system : as measured by a) 'bogomips' (linus's benchmark) : 93.4 times slower than host cpu; b) 'dhrystone' (traditional integer benchmark) : 300.98 times slower than host cpu.

asciilifeform: (actual iron does it in O(1), emulated -- naturally in O(N), where N is up to all 16 entries in tlb table)

asciilifeform: i.e. sitting on 3GHz 'opteron', a 'm' session is approx on part with 486dx, in re ops/sec.

asciilifeform: can prolly be improved somewhat, via cycle-shaving, but not by order of magnitude.

asciilifeform: i'ma let BingoBoingo think re whether it makes sense to offer shared hosting on a thing like that. (or whether folx will simply die of laughter) ☟︎

asciilifeform predicted initially that cost would be ~40x vs host, based on avg. # of x86 instrs. taken per simulated instr.

asciilifeform: seems like ~this~ is the actual (vs. 'i-can't-believe-it's-not-butter' vmware et al) cost of 100% process isolation on pc.

asciilifeform: afaik this is the 1st attempt to measure it, too ( i.e. with max sweat over cycles, 0 c in proggy, 0 libs, etc sim. )

asciilifeform: i/o, however, is only ~4x slower than host's, so in fact one ~could~ run e.g. trb on it.

asciilifeform: to be pedantically exact : 66.35 'bogomips' (on 3GHz opteron) , and published table gives e.g. 'Pentium/166 66.36'. i was unable to find similar table of classics for 'dhrystone' .

asciilifeform: ^ figure on properly unloaded box, vs. earlier; ~= 252.3x slower than host on same benchm.

mod6: http://btcbase.org/log/2019-07-19#1923606 << This is a neat-o vpatch 'who gave', but it came in just after the 'NO NEW WORK IN SHA PLOX'; so there are a few like this that probably will go into TRB main Vtree once the Lordship reviews/audits the proposed Keccak TRB Vtree; perhaps possibly after TRB has a new home OS/environment. ☝︎☟︎

a111: Logged on 2019-07-19 13:58 asciilifeform: revisiting upstack : hey mod6 , did 'who-gave' ever make it into flagship vtree ? ( i dun see it there ... )

mod6: (My apologies for the delayed response). Anyway, long story short, we're in a bit of a holding pattern on new vpatches atm. ☟︎

mod6: thanks asciilifeform! Just trying to catch up on logs/blogs atm. Been behind all month!

mod6: In light of recent conversations re: v (large genesis), ebuilds; this needs some review still. ☝︎

a111: Logged on 2019-07-08 04:01 mp_en_viaje: meanwhile speaking of http://bvt-trace.net/2019/07/vdiff-vpatch-blockwise-keccaking/#footnote_0_42 would someone look at that ?

mod6: In a few days, once I'm caught back up, if still no one has tested this, I'll give it a try.

asciilifeform: mod6: i was gonna genesis a kernel for that thing, but nao after sad benchmark thinkin', it's of ~0 use outside of very specialized applications ☝︎☝︎☟︎

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

a111: Logged on 2019-07-18 05:03 asciilifeform: at some pt somebody's gonna have to audit a snapshot of bin gnat. which wouldja rather ?

mod6: yeah, oof. I saw your post re-'M', will try to read through here soon. I did see your benchmark comments above tho.

asciilifeform: conceivably there is a use for e.g. cuntoo that loads from memory snapshot in <3sec but runs 'like pentium 166'

asciilifeform: to round off that thread -- asciilifeform strongly suspects that kernel would run 100-200x faster on the 10 $ 'ice40' than on opteron-cum-'M'

asciilifeform: ( for that matter -- if e.g. mp_en_viaje one day bakes actual silicon cpu , would run on par with pc, with unchanged soft. )

asciilifeform might even be wrong, and it is conceivable that 'dhrystone of pentium166 but i/o of opteron' actually suffices to e.g. host blogs. ☟︎

a111: Logged on 2019-07-24 00:11 asciilifeform: i'ma let BingoBoingo think re whether it makes sense to offer shared hosting on a thing like that. (or whether folx will simply die of laughter)

a111: Logged on 2019-07-24 00:58 mod6: (My apologies for the delayed response). Anyway, long story short, we're in a bit of a holding pattern on new vpatches atm.

mp_en_viaje: i dunno how exactly you came up with some magic incantations that'll make it okay. however, to repeat this the 5th or so time : there is no magic ; it did not make it okay. ☟︎

a111: Logged on 2019-07-24 01:22 asciilifeform: mod6: i was gonna genesis a kernel for that thing, but nao after sad benchmark thinkin', it's of ~0 use outside of very specialized applications

a111: Logged on 2019-07-24 01:44 asciilifeform might even be wrong, and it is conceivable that 'dhrystone of pentium166 but i/o of opteron' actually suffices to e.g. host blogs.

a111: Logged on 2016-09-23 13:59 mircea_popescu: they ddosed trilema, qntra etc - nothing happened ; they had their agent ddosed - that's the end of the road for him.

mp_en_viaje: asciilifeform, don't feel too bad, this isn't "sad" just because not obviously useful ; it's glorious, because correct. id much rather read this than w/e, say yest's attempt at the claiming of sleeping beauty.

mp_en_viaje: (what happened was that 1. i was looking through search results for "daughter encounter", the why we'll not include as we can't regress this storytelling forever into endless past ; and then 2. ran into dude killed for being dude, which i suppose is kinda sad, but w/e, america rite ; and then 3. fell on "Take It For Daddy: A Rough an

mp_en_viaje: d Reluctant Bondage-Sex Encounter Between Stepdad and Daughter", The world’s greatest erotica writer, Kelly Leigh proudly presents "Take It For Daddy" – Incredibly hot sex stories featuring stepdaughters and their stepfathers!

mp_en_viaje: which 4. made me exclaim "what the fuck is wrong with these idiots, ~I~ am the world's greatest erotica writer, and by such a fat margin that if i were the mvp of the winning team, nobody else writing erotica would be even good enough to be accepted as last choice in one of the feeder clubs of the conference's weakest team. ☟︎

mp_en_viaje: which in turn 5. provoked girly to mention anne rice, because whatever, as a virgin tween she got really hot and bothered over the princess getting sold by the prince to some merchants who stuffed her cunt fulla figs and such things.

mp_en_viaje: so she found me a copy, and 6. i proceeded to read. and i ran into bounty aplenty, such as :

mp_en_viaje: 6.a. "And approaching her, he gave a soft gasp as he touched her cheek, and her teeth through her parted lips, and then her tender rounded eyelids." sparked protest because ~everyone else~ is fucking decrepit and gaunt and dusted over in spider webs, so the disparity provokes in me the expectation a spider's jumping right out, antisexual 100%

mp_en_viaje: 6.b. "Her face was perfect to him" provoked protest because it's a cop out : the woman "kills two rabbits with one stone", by avoiding BOTH the work of describing her face, AND the work of explaining credibly why these two like each other. yet the job of the writer is 100% and precisely NOT to make this deal ; because if they make it, even once, then their story inescapably becomes "this is a good^H^H^H^Hgreat story'

mp_en_viaje: 6.c. "her embroidered gown had fallen deep into the crease between her legs so that he could see he shape of her sex beneath it" produced the protestation that this woman has never seen embroidery (or, i guess, never seen girls in work out pants without panties).

mp_en_viaje: 6.d. "Her ni**les were a rosy pink as were her lips, and the hair between her legs was darkly yellow and curlier than the long straight hair of her head, which covered her arms almost down to her hips on either side of her." actually killed the read, because 6.d.I. the rosy pink of the lips is NEVER THE SAME as the hue of the lips, this is 9yo coloring ; yet I have seen four digit young cunts and the two hues are alw

mp_en_viaje: 6.d.II. wtf is "darkly yellow" ; 6.d.III. wtf is "curlier than perfectly straight" ; 6.d.IV. you DO NOT wish to have her arms "covered in hair" ; then he proceeds to 6.d.V. "cut her sleeves" in the very next line, notwithstanding she was saying "arms" here, and well, i can not read this.

mp_en_viaje: this read mechanically failed ; i have no qualms with whoever managed to jack off to anne rice nevertheless -- but it is NOT good. even if it works for you. ☟︎

mp_en_viaje: because "works for me" ain't the fucking criteria of goodness. mp being able to read through the thing, ie, CORRECTNESS, is the criteria of goodness.

mp_en_viaje: so there you go, i'd rather read you than A. N. Roquelaure ; and this isn't a rhetorical turn of phrase, it is literal description of actual historical occurrence.

mp_en_viaje: and no -- no apologies for log clutter ; this imo is the exact opposite of log clutter. this is because there are no spurious details included -- if it's long, it's because the point is intricate.

mp_en_viaje: http://btcbase.org/log/2019-07-24#1924476 << this, btw, includes all, living and dead, from d. h. lawrence to whoever the fuck you wish to pick, salo 0.1 marquis de sade, jane austen, WHOEVER wilkes, beerbohm, ☝︎

a111: Logged on 2019-07-24 08:48 mp_en_viaje: which 4. made me exclaim "what the fuck is wrong with these idiots, ~I~ am the world's greatest erotica writer, and by such a fat margin that if i were the mvp of the winning team, nobody else writing erotica would be even good enough to be accepted as last choice in one of the feeder clubs of the conference's weakest team.

mp_en_viaje: there just is nobody the fuck else, not even fucking close, nor ever was. and we're not even discussing ~range~ here, the fact that i'm one and these are legion. we're simply saying that out of all humans who ever lived, nobody came even close in any way and on any direction. i own erotica like zeus owns olympus.

a111: Logged on 2019-07-24 08:25 mp_en_viaje: http://btcbase.org/log/2019-07-24#1924425 << it doesn't seem practical.

asciilifeform: http://btcbase.org/log/2019-07-24#1924470 << the sheer waste of current aside -- not sure that ~this~ : let's gedankenexperiment, and picture a 'M' running apache, under ddos: ☝︎

asciilifeform: the way i was gonna implement the nic, is to pipe packets via linux's 'tap/tun' into a slot in the 'm' mmio space (see bus.asm) and generate slave irq (see irq.asm) . if these arrive faster than the irq can be eaten, they simply get dropped.

asciilifeform: grr phf is it gonna take 2 months for the thing to be in btcbase.org/patches so i can link to it ?! ☟︎

asciilifeform: http://btcbase.org/log/2019-07-24#1924463 << on the actual iron, it happens in 1 clock cycle, cuz entirely parallel. in 'm', ends up iterating through all 16 tlb entries. i looked into using 'simd' instructions to do it, but they are ~artfully~ useless , ended up using the xmm regs strictly as fast temp storage; ☝︎

asciilifeform: originally was at least gonna park the tlb ~in~ the xmm regs. but guesswat, there aint even a way, in 1 clock cycle, to get anything other than low 64bit out of'em !

asciilifeform: http://btcbase.org/log/2019-07-24#1924495 << for comparison, btw, 2m 23.2s for same 1000000000 runs on host box. ☝︎

asciilifeform: mp_en_viaje: well, on my opteron, they're 128bit regs (on moar recent -- 256 and even 512!) but only the lower 64 is directly addressable. and gives buncha 'simd' ops, e.g. 'shuffles' where 'take erry 7th bit and add'em' etc. but looked and looked an' found nothing that'd correspond to table lookup

asciilifeform: kernel-mode coad btw bypasses the tlb, so runs with the expected ~40x slowdown.

asciilifeform: ( for n00bz / tlbism (aka mmu) is how os gives individual userland process the illusion of 'infinite memory, starting from addr 0', paging, etc )

asciilifeform: to add insult to injury, asciilifeform knows how to represent 'x bit lookup from x*k-bit string representing table' via arithmetic methods -- but! amd64 dun let you do full arithmetic on xmm either ☟︎

asciilifeform: in the beginning, thought i could at least implement the mips instruction operand decode using xmm. but ha! not that either.

asciilifeform: ( at least not on my opteron, on 2014+ theoretically could, but i dun buy those )

asciilifeform: best i could think of , was to arrange the mov.../and.../shr... etc so as to occupy the amd64 pipeline properly, in each inst. form.

asciilifeform: seems like there is only so much 'silk purse' that can be made from this particular sow's ear.

asciilifeform: in the actual instrs. per se (see 'mipsinst' dir.) , used conditional mov's wherever could think of how, so as to also keep pipeline full

asciilifeform: they used 'dynamic recompile' method, where the sim simulates a kind of cache. when sim-instruction fetched, it looks whether found in this cache, and if not, emits a chunk of x86olade into said segment of cache, which corresponds to 'compiled' ver. of that instr.

asciilifeform: and as it goes, tries to keep cache full, does equiv. of 'prefetch' (if this were a normal iron cache), etc

asciilifeform: all of this is 1) exceedingly gnarly , from 'fits in head' pov 2) requires a writable-and-executable memory segment. i.e. massive open wound.

lobbes: http://btcbase.org/log/2019-07-22#1924301 << roger that. I'll aim for this weekend to get it published ☝︎

a111: Logged on 2019-07-22 20:12 mp_en_viaje: lobbes, you'll live. anyway, get that thing published so others can patch on it and thus lighten your own workload!

lobbes: http://btcbase.org/log/2019-07-22#1924303 << I appreciate this offer, ty. I've got my living arrangements and job situation under control for the short term, BUT once this is done I may indeed take you up on your offer re: finding girls. Been poking around the internet and well, it is still a shannonized ocean of wtf in this regard. ☝︎

a111: Logged on 2019-07-22 22:56 Mocky: lobbes if you want to relocate 3 hours east, 90% chance I can get you a job in Ral, I have a spare room you can stay in while you look for a place, ditch the dysfunction, we'll find new girls

asciilifeform: mp_en_viaje: thinking about it moar, i suspect that it is possible to bake a less screamingly retarded version of bellardism , where, e.g., the 'cache' is prefilled with unconditional jumps to the sim-instructions, eliminating the decode hit

asciilifeform: the published item is imho at the point where ~other~ folx could saw on it, run the dhrystone, etc.

asciilifeform: also imho it is potentially foundation for a http://btcbase.org/log/2019-02-05#1892603 , likewise if the hands can be found. ☝︎

a111: Logged on 2019-02-05 01:10 asciilifeform described, in the past, the general shape of the recipe for this -- you write e.g. small lisp in coupla kb of asm, and large 1 on that, and an adatron on the latter... etc

a111: Logged on 2019-07-20 10:03 mp_en_viaje: http://btcbase.org/log/2019-07-20#1923930 << maybe worth going over "what would this solution look like" one more time.

BingoBoingo: http://btcbase.org/log/2019-07-24#1924486 << Yet another plague wrought by open access to the internet. Most published sex writing done either by or for bipeds that don't have sex. ☝︎☟︎

a111: Logged on 2019-07-24 08:48 mp_en_viaje: this read mechanically failed ; i have no qualms with whoever managed to jack off to anne rice nevertheless -- but it is NOT good. even if it works for you.

BingoBoingo: http://btcbase.org/log/2019-07-24#1924470 << M seems very useful to keep on hand for ice40 or having a base for porting Cuntoo to printed MIPS in 3-7 years ☝︎

asciilifeform: BingoBoingo: it aint as if there were no support for mips in the classic kernel tho. ( i only had to add 100ln or so specifically to bring up the simplified sim-devices )

asciilifeform: relatedly, i can't fathom why bellard et al decided to simulate gnarly existing physical irons , rather than patching kernel

BingoBoingo: asciilifeform: Sure, but if we print a MIPS down the line sim-devices may be the way to do it instead of DMA everywhere

asciilifeform: BingoBoingo: principal reason why i went awol from ffa to shit out this horror , was the http://btcbase.org/log/2019-07-19#1923477 burning q ☝︎

a111: Logged on 2019-07-19 03:23 asciilifeform: would like to offer heathens a) fully isolated linuxen b) at competitive cost. so far erryone in meat wot to whom showed piz rate sheet, replied something in vein of http://btcbase.org/log/2019-07-16#1922858 .

asciilifeform: somehow gotta find method to house coupla 100 heathens in actual isolation from ea. other

asciilifeform: and preferably in such a way that BingoBoingo dun have to dig through rat's nest of wires erry time 1 subscribes or unsubscribes

asciilifeform: grr i even had 'disk' driver mostly written, was gonna build bootable simdisk w/ cuntoo...

BingoBoingo: In other news, fever gone, gut still a bit tender, brown girl on bleach detail

a111: Logged on 2019-07-24 14:14 asciilifeform: to add insult to injury, asciilifeform knows how to represent 'x bit lookup from x*k-bit string representing table' via arithmetic methods -- but! amd64 dun let you do full arithmetic on xmm either

asciilifeform: mp_en_viaje: x86 is a mountain of sad of which most folx (even such as write proggies) see only the very tip. but goes all the way down into magma.

a111: Logged on 2019-07-24 15:45 asciilifeform: thing's 50-60% comment by weight, imho amply documented.

mp_en_viaje: http://btcbase.org/log/2019-07-24#1924548 << this is true. and before that, most erotica written by little bois, oh so impressed with selves for having broken an ear or somesuch off inca statue. ☝︎

a111: Logged on 2019-07-24 15:55 BingoBoingo: http://btcbase.org/log/2019-07-24#1924486 << Yet another plague wrought by open access to the internet. Most published sex writing done either by or for bipeds that don't have sex.

mp_en_viaje: no systematically anti-stupid , ideologically integrated, philosophically superior efforts.

mp_en_viaje: asciilifeform, http://trilema.com/2012/filosofia-din-budoar-dragi-libertini/ << there's a ro series, directly comparable and imo far superior.

BingoBoingo: related to the incels doing erotica discussion: https://archive.is/NXNBC "As part of an ongoing military conflict, an attack against a USN carrier would not necessarily represent a legal challenge; aircraft carriers are weapons of war, after all, and they are just as vulnerable to attack as any other weapon. But as military theorists have pointed out for at least two centuries, states choose their levels of escalation very carefully.

BingoBoingo: Most wars are limited wars, and in limited wars, generals, admirals, and politicians are aware of the political import of the targets they select. Consequently, some targets remain off-limits for states that want to keep a war limited, even if those targets make a material contribution to the conduct of the conflict."

BingoBoingo: And they stooge writes this as though no one learned since 2003 that when the USG comes knocking you can't play into their escalation moderating game or a "Shock And Awe" happens

mp_en_viaje: the ugly fat kid nobody likes similarly goes through school "believing nobody will hit him". what other options does he have ?