log☇︎
1200+ entries in 0.18s
asciilifeform: he's a very clever maths d00d. logs in every other week or so, nitpicks re fine points of ffa.
asciilifeform: for single-author, e.g. ffa, nothing really changes.
asciilifeform: ftr several different items in ffa seemed to me to be 'five-angled heptagons' (starting with how to compute the asm-less addition carries) until i solved'em
asciilifeform: the other open q is where to draw the line re 'somebody, some day, with mutilated ffa' .
a111: Logged on 2018-01-17 16:43 asciilifeform: concretely, e.g., Dividend_Index in http://btcbase.org/patches/ffa_ch7_turbo_egyptians#L31
asciilifeform: concretely, e.g., Dividend_Index in http://btcbase.org/patches/ffa_ch7_turbo_egyptians#L31 ☟︎
asciilifeform: ffa itself is a sort of tightrope walk, an attempt to 'и рыбку съесть и нахуй сесть' . it isn't actually possible to make all of the routines able to take every conceivable kind of compile-time abuse ( which i'd argue overlap of arguments, is ) without conditional jumps.
asciilifeform: i never proclaimed, fwiw, that all ffa routines must be able to cleanly walk over own inputs.
asciilifeform: (spoiler : ^ == ffa ch4 )
asciilifeform: the whole point of ffa is that it is to contain NO such possibility of NO such thing.
a111: Logged on 2018-01-16 06:15 apeloyee: FFA homework is fun! I'd like to present a homework problem of my own, the following Perfectly Innocent (tm) patch: http://p.bvulpes.com/pastes/tcvvO/?raw=true . The task is to find a bug WITHOUT running the code (that's cheating, per Dijkstra: " we see to it that the programming language in question has not been implemented on campus so that students are protected from the temptation to...
apeloyee: FFA homework is fun! I'd like to present a homework problem of my own, the following Perfectly Innocent (tm) patch: http://p.bvulpes.com/pastes/tcvvO/?raw=true . The task is to find a bug WITHOUT running the code (that's cheating, per Dijkstra: " we see to it that the programming language in question has not been implemented on campus so that students are protected from the temptation to... ☟︎
asciilifeform: though i'll point out that ffa as seen today, does not use the cpu carry flag ( recall, no-inline-asmolade rule )
a111: Logged on 2018-01-15 19:45 BingoBoingo: phf: If I remember right Elbrus was a SPARC64 y entonces likely missing an ASM instruction useful for FFA with speed
BingoBoingo: phf: If I remember right Elbrus was a SPARC64 y entonces likely missing an ASM instruction useful for FFA with speed ☟︎
mircea_popescu: dja happen upon either ffa or eucrypt write-ups, as it happens ?
asciilifeform turns head and looks at the literal wall of ffa
mod6: from_ffa: sorry I missed ya on here. but for now, yeah, that behavior is intended.
a111: Logged on 2018-01-10 20:17 from_ffa: at any rate, if I run "v.pl p ~/workspace ffa_ch1_genesis.vpatch" I expect to find ffa alongside all other projects. not an empty workspace dir with ffa in it
from_ffa: at any rate, if I run "v.pl p ~/workspace ffa_ch1_genesis.vpatch" I expect to find ffa alongside all other projects. not an empty workspace dir with ffa in it ☟︎
asciilifeform: from_ffa: so please consider logging in
asciilifeform: from_ffa: and please consider getting a name, and registering with deedbot,
asciilifeform: from_ffa: post it, he'll answer when he wakes up
from_ffa: a regular lurker of the chan. just started to muck around with your ffa series
asciilifeform: from_ffa: who might you be ?
asciilifeform: from_ffa: hello ?
deedbot: from_ffa voiced for 30 minutes.
asciilifeform: !!up from_ffa
asciilifeform: meanwhile, in other ffa noose, http://www.loper-os.org/?p=2105&cpage=1#comment-18610
mircea_popescu: asciilifeform so are you implementing maurer in ffa ?
asciilifeform: it is worth to keep in mind that ffa is , to steal from knuth, 'for readers to read first, and for machine to execute second'
asciilifeform: ( ffa specifically. in ffa_calc, very obviously they do , in the stack mechanism . however the inner loops, i.e. 99.9999...% of the time is spent, in ffa proper. )
asciilifeform: ffa is one of the very few ( i dun expect to see another ) proggies where you can prove bounds. because the array indices do not EVER depend on the inputs.
asciilifeform: http://btcbase.org/patches/ffa_ch4_ffacalc#L194 << from here , to end of ffa_calc.adb, whole mechanism.
asciilifeform: it does need the fast ffa, among other items.
asciilifeform: http://btcbase.org/log/2018-01-07#1766206 << after ffa i'ma necessarily move on to the chore of nailing down gnat. ☝︎
asciilifeform: http://btcbase.org/patches/ffa_ch1_genesis#L161 << the specific item that seems to fail on his box
esthlos: uh, just used ./v.pl p press_output ffa_ch1_genesis.vpatch, if that's what you mean
asciilifeform: the only data-dependent branches are in ffa_calc , the calculator.
asciilifeform: ffa is possibly one of the few nontrivial proggies where you can actually do this. because there are no data-dependent branches (aside from the 'data' which is the bitness you run under -- this determines the loopness of the loops)
asciilifeform: we're approx. at level of a ( much cleaner ) circa- june 2017 ffa nao.
mircea_popescu: "Let’s verify the RSA seal of ffa_ch6_simplest_rsa.vpatch, the Chapter 6 code itself, using itself". epic.
asciilifeform: and now it takes a lifetime to simply juggle the v-mechanics, leaving ~0 time for any of the 'irrelevant' detail like, say, writing ffa.
asciilifeform: this is when asciilifeform admits that he has a small demo of this. but is being saved for after ffa, cuz Finishing Things Properly (tm)
asciilifeform: if there's anybody else who signed , but not currently visible in http://btcbase.org/patches/ffa_ch5_egypt in appropriate spot -- plox to write in.
asciilifeform: !~later tell phf plox to snarf http://www.loper-os.org/pub/ffa/ffa_ch3_shifts.vpatch.benvulpes.sig .
asciilifeform: !~later tell phf plox to snarf http://www.loper-os.org/pub/ffa/ffa_ch4_ffacalc.vpatch.benvulpes.sig , ty
asciilifeform: http://btcbase.org/patches/ffa_ch5_egypt/tree/ffa/ffacalc/ffa_calc.adb#L232 << for reference
asciilifeform: point being, if you patch on an existing ffa_calc, it'll have to be reground
asciilifeform: so far the folx who code on paper, wrote ffa, and the folx who wrote on display -- wrote what.
asciilifeform: and yes this is what asciilifeform does when he wakes up : rereads ffa. all of it.
asciilifeform: >> http://btcbase.org/patches/ffa_ch5_egypt#L63 << concretely.
asciilifeform: there's a stray MustNotZero(Stack(SP)); in ffa_calc.
asciilifeform: that being said, gprbuild ( what ffa uses instead of gnumake ) seems to make very effective use of a kind of interpreted subset of ada
asciilifeform: diana_coman: as a side effect, you get an easily formatted code-box for blogging, as in ffa.
asciilifeform: e.g. http://btcbase.org/patches/ffa_ch4_ffacalc#L205
asciilifeform: ben_vulpes: observe again style in ffa
asciilifeform: meanwhile observe asciilifeform's formatting in ffa
asciilifeform: and speaking of this, asciilifeform is almost certainly doomed to properly texize ffa in the end. there are so few things more loathesome than reading MECHANICALLY REFLOWED code. whether line flow or page breaks.
asciilifeform: ~all ffa ops can be reduced to 'add-with-possible-2scomplement' and 'shift'.
phf: asciilifeform: http://btcbase.org/patches/ffa_ch5_egypt
a111: Logged on 2018-01-01 20:07 ben_vulpes: pretty good, listening to raffi, working through ffa_ch4, tidying up some stuff for $work
ben_vulpes: pretty good, listening to raffi, working through ffa_ch4, tidying up some stuff for $work ☟︎
asciilifeform: fortunately there is nothing nontrivial in ffa.
asciilifeform: to round off thread -- asciilifeform very much enjoying rewriting ( and it is , yes , a total rewrite ) ffa ☟︎
asciilifeform: ( incidentally gprbuild, which is used in ffa, is quite interesting. )
asciilifeform: all of them but the FFA_IO items and the hex_digit thing.
mircea_popescu: and why the fuck would you maintain it in a place which has no need to maintain it. ffa is free from said flesh.
asciilifeform: consider the approach in ffa. only critical bottlenecks, that soak up 90+% of cycles and create an impractically-slow rsa op , get massaged ~to the extent necessary~ , and without compromising type safety
asciilifeform: trinque: orig stimulus for ffa was gossipd's rsa-in-realtime req.
a111: Logged on 2017-12-29 03:38 asciilifeform: http://www.loper-os.org/pub/ffa/pehbot.jpg << for posterity.
mircea_popescu: and is it getting updated in step with the ffa publishings ?
asciilifeform: http://www.loper-os.org/pub/ffa/pehbot.jpg << for posterity. ☟︎
asciilifeform: for that matter, what's ffa, lol
asciilifeform: http://www.loper-os.org/?p=1913&cpage=1#comment-18457 << congrats to spyked , graduate of ffa ch1.
asciilifeform: !~later tell phf plox to snarf : http://www.loper-os.org/pub/ffa/ffa_ch1_genesis.vpatch.spyked.sig
asciilifeform: as in, ffa-width cpu.
asciilifeform: !~later tell phf plox to snarf: 3) http://www.loper-os.org/pub/ffa/ffa_ch4_ffacalc.vpatch.peterl.sig . ty
asciilifeform: !~later tell phf plox to snarf: 1) http://www.loper-os.org/pub/ffa/ffa_ch2_logicals.vpatch.peterl.sig 2) http://www.loper-os.org/pub/ffa/ffa_ch3_shifts.vpatch.peterl.sig
asciilifeform: trinque: i actually put a good bit of thought into the vtronic shape of ffa, while rewriting it ( current-day ffa , observe, is a rewrite, largely by hand, of the previous )
asciilifeform: ffa is, among other things, an experiment with minimizing theugly.
phf: asciilifeform: also http://btcbase.org/patches/ffa_ch4_ffacalc
a111: Logged on 2017-12-23 03:53 asciilifeform: speaking of http://btcbase.org/log/2017-12-23#1757328 , today's ffa-series item will appear tomorrow, apparently was too big a bite for asciilifeform's mere 1wk
asciilifeform: speaking of http://btcbase.org/log/2017-12-23#1757328 , today's ffa-series item will appear tomorrow, apparently was too big a bite for asciilifeform's mere 1wk ☝︎☟︎
asciilifeform working on unexpectedly massive ffa-series piece
asciilifeform: btw, ftr asciilifeform considered f77 as close runner-up for 'in what to write ffa and all from nao on'
asciilifeform actually has the db item, based on 'horsecocks' routine posted earlier ; but not ready for primetime and on hold pending ffa/p release
asciilifeform: ftr i dun plan to include 'secure memory'ism in ffa.
asciilifeform: ( e.g., gprclean actually cleans ffa . )
asciilifeform: btw if esthlos ( or anyone else, anywhere ) knows how to aks on 2048b primes in nongeological time, i'll happily port their algo to ffa, even if original is in cobol
mircea_popescu: esthlos i'd love for you to be able to jump straight into this ; however there's some groundwork to be laid. look into the V system, because ideally you'd be presenting the finished item as a patch on diana's eucrypt lib. and asciilifeform is working on and publishing a final FFA which is what we intend to use here.
mircea_popescu: rather than being all negative!!!, got a review copy of ffa somewhere to pass to esthlos ?
mircea_popescu: esthlos look through the various FFA items as pasted by asciilifeform first ; to get an idea what the item is like. so you don't have to wait till he's finished publishing it all
mircea_popescu: esthlos ada. see ffa for a basis
phf: asciilifeform: http://btcbase.org/patches/ffa_ch3_shifts in press order you can see all the mentioned sigs included
asciilifeform: mirrored at http://www.loper-os.org/pub/ffa/ffa_ch1_genesis.vpatch.diana_coman.sig , http://www.loper-os.org/pub/ffa/ffa_ch2_logicals.vpatch.diana_coman.sig , http://www.loper-os.org/pub/ffa/ffa_ch3_shifts.vpatch.diana_coman.sig . congrats to diana_coman , first graduate of ch1-ch3 !
asciilifeform: ( ada params are a moar complicated matter than ffa might suggest: when reading heathen proggies you will encounter such things as 'aliased' i.e. pointerola , and related items. )
asciilifeform: ben_vulpes: there is no 'constructor' in ffa . and FZ_Clear does exactly as printed on the box
asciilifeform: !~later tell phf plox to intern http://www.loper-os.org/pub/mpi/ffa_ch1_genesis.vpatch.benvulpes.sig
asciilifeform: in other noose, congrats to ben_vulpes , http://www.loper-os.org/?p=1913&cpage=1#comment-18420 << 2nd ffa ch1 graduate