93400+ entries in 0.052s
a111: Logged on 2016-02-24 04:23 mircea_popescu: omfg alf sees
the world like a bee-dog : in black and white and all pixelated.
mircea_popescu: then again alf almost sounds like an anime character, doesn't seem
to have hurt him any.
a111: Logged on 2018-06-11 18:33 apt-get: rude
tbh, I've been using
this one online for quite some
time
mircea_popescu: eventually went
to specialist store, bought 3 meters of double-width
towel substance, had
them rodeando it. 3 * 3500 + 3000 for
the work = ~30 bux. now i have a proper beach
towel, can seat five.
mircea_popescu: and in other fuck-this-failed-civilisation, NO SHOP in all
the fucking
town had a proper beach
towel.
the chinese overlords have decided all
towels must be up
to 1/3 size and
that's it. "i want a
towel king bed size" "you mean sheets ?" "no dood.
towel." "here's
the
towels." "these are small."
deedbot: Provide a paste URL
to
the ascii-armored GPG public key or
the full 40 character key fingerprint without spaces or dashes.
oda: asciilifeform:
thanks, will do.
oda: Just wanted
to lurk a bit and see what sort of chat goes on here
oda: Hi, just got here after reading
the cr50 article on loper-os
mircea_popescu: hand crafted wood. dood was beffudled, didn't really even want
to sell it
to me. "it's for ovens".
a111: Logged on 2018-06-11 20:35 asciilifeform: so far my only clue
that h1 actually runs
the given fw , is
that i was able
to flash in a vendor update :
http://btcbase.org/log/2018-06-08#1821699 and ended up with a slightly different, in
the ways suggested by
the src, console
a111: Logged on 2018-06-11 19:57 asciilifeform: swiftgeek: given your introduction (
http://btcbase.org/log/2018-06-11#1822589 ) i assume you may be interested in verifying fact
that cr50 is not a subfunctionality of
the ordinary (i.e. kept in winbond spi ) bootrom or
the EC controller ('nuvoton' arm , visible in right hand of photo ).
this is very simple
to do:
a111: Logged on 2016-12-16 14:31 mircea_popescu: japan got buldozed chiefly because of
the utterly immoral attitude of
thinking people at
the
time.
trinque: for one, I'd expect anyone who spent enough
time in one
to come out
the other side schizophrenic
trinque: perhaps
the compartmentalized nonsense factory is not such a strong longterm strategy
danielpbarron: crack smokers will burn any
time
they can get ahold of, granted
they generally glob onto other cracker smokers of
the have-money-to-buy-more variety
a111: Logged on 2018-06-11 22:15 BingoBoingo: Well, living in
the land of Mate crackpipes you get
to see
this behavior
taken
to extremes.
They are playing
the game where
they
TRY
to burn as many man hours as possible in an unproductive manner.
a111: Logged on 2018-06-11 21:24 hl`:
that's actually a fair point
too. as implemented in e.g. PCs nowadays, even putting
the closed firmware issues aside,
the way
they are integrated is _not_ secure.
they're just connected using open pins
to
the CPU, you could easily replay everything
BingoBoingo: Productivity is dangerous. In Uruguay
the danger is raising
the bar and
taking away from crack pipe
time. In DARPA land
the dangers being fended off are myriad.
BingoBoingo: Well, living in
the land of Mate crackpipes you get
to see
this behavior
taken
to extremes.
They are playing
the game where
they
TRY
to burn as many man hours as possible in an unproductive manner.
☟︎ a111: Logged on 2017-09-15 23:48 asciilifeform: kanzure: i spilled
the beans from a similar darpa conference
that i attended, in
the heart of
the beast itself, few yrs back ( it's in
the l0gz, spoiler : multilinear map homomorphic crypto is bunkum ) and still waiting for gasenwagen
a111: Logged on 2018-06-11 21:15 swiftgeek: DARPA was messing with
that a lot
BingoBoingo: that, or
to repeat
the "common" process
they are sure gets done all
the
time.
☟︎ BingoBoingo: And with
the read only brains, only hope
they have is
that
they are interesting enough
to an MP for
them
to recieve
their exploitable crash
that re-enables write access.
BingoBoingo: Well for most people, wife replaced while
they sleep would probably be +EV
BingoBoingo: <hl`>
that doesn't protect against physical attacks. << Buy a dog and carry a hammer
swiftgeek: anyway
that covers everything for me, i can only wait for more docs
to appear (or dead boards)
swiftgeek: anyway so far
there is no root of
trust implemented in SoC
that respect end user
swiftgeek: and
TPM implemented so poorly
that it doesn't reset x86 with it
swiftgeek: but exploiting
TPM firmware so much
that it resets
swiftgeek: yep
that makes it possible even with root of
trust
hl`: (see
TPM reset attacks.
the
TPM specification people claim
that
these were fixed with
TPM1.2,
this is not correct however)
hl`: that's actually a fair point
too. as implemented in e.g. PCs nowadays, even putting
the closed firmware issues aside,
the way
they are integrated is _not_ secure.
they're just connected using open pins
to
the CPU, you could easily replay everything
☟︎ swiftgeek: and in x86 case
that happens
to be EC
swiftgeek: otherwise i will exploit other device on
the bus and replay it remotely
swiftgeek: hl`: anyway only with proper root of
trust you can measure all stages with
tpm
swiftgeek: nah i was just referring
that qcom code is generating code
that generates
to generate code
that (....)
swiftgeek: and not at all for anyone
trying
to exploit it xD
swiftgeek: asciilifeform: no
this is qualcomm modem code
hl`: i.e., you'd have
to solve
the halting problem
to write a program which can analyse
the generated programs in
the general case, meaning
that any computational malevolence (compromised silicon, etc.) can only compute
the result of
the algorithm by executing it unless someone solves
the halting problem
swiftgeek: asciilifeform:
those packages are expensive and cute :)
hl`: general case - basically using
the halting problem as a
trapdoor function.
hl`: asciilifeform:
that's actually an interesting idea - i've
toyed with a similar idea previously,
though for different applications. basically, my idea was
to come up with some way of algorithmically generating algorithms such
that
the algorithm generator can know
the correct answer computationally easily, but where
the structure of
the algorithm is highly randomised such
that it resists analysis in
the
swiftgeek: you wouldn't go below 65nm if you are sane for
tpm