46000+ entries in 0.026s
mircea_popescu: alright. and if
there's 22 for loops,
this means
the correct count of "how many
times serpent is run" is 50 ^ 22
then ?
mircea_popescu: do me a favour first and let's nail
the numbers down. so, for 1
to 100 if mod 2 results in
the loop being run 50
times ? or 49 ?
mircea_popescu: asciilifeform basically, we found i can't math ;
that aside, we found
that in one context serpent
takes ~3us, and in another ~0.3us.
a111: Logged on 2019-01-10 14:08 asciilifeform: aaand
to round off : it vanished on
the
test box also. culprit appears
to have been a running raid-verify job...
mircea_popescu: we are currently entering
the loop
twice, and we enter a
total of 22 loops.
therefore
the number of
times serpent is run is 2 ^ 22.
diana_coman will need
to leave for about 1 hr in ~5min
time but will read
a111: Logged on 2019-02-13 14:24 diana_coman: asciilifeform, one! and look here at
times: 1 loop -> 0.000168893 s ; 2 loops -> 0.007213758 s ; 3 loops -> 0.351611073 s ; 4 loops -> 17.74 s ; 5 loops-> 879.95 s
mircea_popescu: except
the other set converged, look : 3.37786e-06 2.8855032e-06 2.812888584e-06 2.8384e-06 2.81584e-06
diana_coman: and both of
them are likely
to be more due
to measurement error
mircea_popescu: this is degree of magnitude off, somehow. wanna walk
through
the math with me ?
mircea_popescu: on
the basis of
the previous
timings you reported, we estimated it ~2.8 microseconds
mircea_popescu: diana_coman
the problem is
this : on
the basis of
this last run, we're estimating serpent
to
take 0.3 microseconds.
diana_coman: because
the clock
thing is not very precise; on 4 runs of same
thing I got: 1.25, 1.27, 1.29, 1.28
diana_coman: given
that it's still not
that long, I'd do at least an avg
mircea_popescu: asciilifeform i
think her work above provides us with some useful data : it
turns out it is reasonable
to expect
timings converge for measured intervals of at least 1/3 of a second, on
the basis of
the above.
mircea_popescu: diana_coman
try whole 23 loops, same 1
to 10 mod 4 plox ?
mircea_popescu: diana_coman seems it stabilizes after 3 loops or so, if you look, it's within a few % with .007s and pretty much
there at 0.35s
diana_coman: trouble is
that it might be
too fast for
the sort of
timing precision we have
diana_coman: a
to j aka 10 loops from 1
to 10 only and
then with if mod 4 -> 0.000855 s (no long jmp)
mircea_popescu: diana_coman idea was
to go
twice per loop, rather
than 1nce. 2 will go in
too many
times still
mircea_popescu: i might be
the weakest person in
the world for basic arithmetics.
diana_coman: see
the
timings above for 1-4 loops for some concrete idea
mircea_popescu: diana_coman you'll have
to abort it, reduce
the loops significantly. sorry bout
that.
lobbesbot: mircea_popescu: Error: Something in
there wasn't a valid number.
diana_coman: it's of course not exactly surprising, given
that one goes from 50 serpent executions
to 50mn rather quickly
diana_coman: 5 loops is already 15 minutes and ftr 6 loops is still running on
that machine as we speak - by
the looks of it, it'll be more
than 1 hour
diana_coman: for completeness:
the set of data above is from a different machine hence a bit faster
than
those of yest
diana_coman: hence my original only 3 loops really; but mircea_popescu asked for
the full set of 23 loops so
there it is, still running
diana_coman: asciilifeform, one! and look here at
times: 1 loop -> 0.000168893 s ; 2 loops -> 0.007213758 s ; 3 loops -> 0.351611073 s ; 4 loops -> 17.74 s ; 5 loops-> 879.95 s
☟︎ diana_coman: basic
test including serpent +
test project with full set of loops : ossasepia.com/available_resources/ljmp_test.tar
diana_coman: so I'll post
the
testing code with full set of loops in a bit and
then go and set up another run
too
diana_coman: I
think a gradual approach (i.e. run it & record results gradually increasing
the number of loops) might be useful meanwhile
a111: Logged on 2019-02-12 23:41 mircea_popescu: i'm still waiting for diana_coman
to return.
a111: Logged on 2015-02-02 06:39 mircea_popescu: Daryush Valizadeh,[1] (born June 14, 1979)[2] also known as Roosh V and Roosh Vorek, is an American[3] writer, pick-up artist
5 and self-styled "love
tourist"[6] of Iranian and Armenian descent,[7] known for his writings on seduction and antifeminism.
mircea_popescu: BingoBoingo do you happen
to recall a
trilema article about some moronic "pick up artist" and his
tedious adventure getting some 17yo i
think it was
to sorta-take it, after half a day of pigeoning ?
mircea_popescu: in ro
the expression'd be rather negative, "i got nothing". cuz "o pula" works also as universal negation, "do you have any money ?" "i got dick"
a111: Logged on 2016-06-21 01:18 mircea_popescu: "mult stimate ion caciula, om al muncii fara scula : cit ai degete si limba, legea
tarii nu se schimba."
a111: Logged on 2018-10-25 19:10 asciilifeform: when you add compatibility spackle, serious reader is not saved from reading
the
thing you spackled over -- on
the contrary nao he has
to read
the ~original~ rubbish ~plus~ your spackle, however much it weighs.
a111: Logged on 2015-03-12 01:50 asciilifeform: 'mil-std-1750a' for
the curious.
a111: Logged on 2019-02-12 23:03 bvt: asciilifeform: sure, cpp code won't be instrumented.
the 'polled' mode was iirc in
the guts of
the ancient linux
threads implementation.
a111: Logged on 2019-02-12 16:10 asciilifeform: in e.g. airplane, 'exception' should mean
that
the entire comp shuts down and
transfers control
to 1 of
the hot spare duplicates.
mircea_popescu: if
there's a mechanism in computing
that's supposed
to be push not pull,
that's exceptions. can't fucking have a pull mechanism for exception handling, and i see 0 gaisn from moving
the honest braindamage of "if ZCX_By_Default
then return;" into an elaborately & contrivedly hidden same exact
thing.
bvt: it won't we poll-killable if
the whole loop is in cpp/asm. if just a linear code - should not be
too bad; sjlj should still work better.