tmsr - Search: from:asciilifeform

asciilifeform: 1ad4391c46ab3520e9aaf8dce2d48d4788fd3752e93c70fabf0dc66f770628d835e764855ab933905a4152524b100358c82b27ca0471a77af02836740e9c41be piece_of_shit.tar.gz

asciilifeform: (srsly, there is an 'atari' port.)

asciilifeform: 'NetBSD: running on a dead squirrel in a cardboard box since '03'

asciilifeform: that's really all.

asciilifeform: don't change.

asciilifeform: 0x800000 - if it wasn't obvious - is where uboot parks the kernel prior to jumping in

asciilifeform: incidentally it is possible to write the eeprom from uboot (you gotta pump in the new payloads via tftp in that case, but you save time by not needing to wait for an os - such as the terrifying bloat of systemdized-archlinux - to load)

asciilifeform: (the value is simply the sum of the lengths of mtd0..3, naturally, and these numbers are available in the uboot console if that's all you've got)

asciilifeform: if you monkeyed with your pogo's mtd mappings, you will have to recalculate

asciilifeform: the 0x1000000 is the offset of the default mtd4 eeprom partition

asciilifeform: i possibly ought to discuss the constants

asciilifeform: will post patch set once i can be bothered to crap one out

asciilifeform: for whatever that's worth.

asciilifeform: i have demonstrated that netbsd kernel (of my custom trimmed build) will correctly init the periphs on pogo

asciilifeform: ben_vulpes, mod6, jurov, mircea_popescu, et al ^^^

asciilifeform: and possibly in case someone wishes to install on larger disk and boot from it, for purposes of dev work.

asciilifeform: the only conceivable use of the above sample is to demonstrate that netbsd will function correctly on 'pogo'.

asciilifeform: (throw it up on ftp on your own box)

asciilifeform: can find binary sets for arm at http://ftp.twaren.net/NetBSD/NetBSD-6.1.5/evbarm/binary/sets or your local sets build

asciilifeform: use as described above. will boot into a netbsd installer.

asciilifeform: i will not sign this. if it annihilates your home planet, i am not responsible.

asciilifeform: ^^^^^^

asciilifeform: http://www.loper-os.org/pub/piece_of_shit.tar.gz

asciilifeform forgot that not everyone's pogo has the serial jack

asciilifeform: unless you have it configured to do the udp console thing

asciilifeform: ideally

asciilifeform: (i just eyeballed it)

asciilifeform: the 0xc00000 is a round number here, you want it to encompass your binary's size

asciilifeform: (for manual testing, can make this default behaviour if want)

asciilifeform: then you halt uboot on the next powerup and 'run boot_bsd'

asciilifeform: fw_setenv boot_bsd 'nand read.e 0x800000 0x1000000 0xc00000; bootm 0x800000'

asciilifeform: nandwrite -p /dev/mtd4 netbsd.ub

asciilifeform: flash_erase /dev/mtd4 0 0

asciilifeform: basically, used like this,

asciilifeform: that results in a netbsd.ub which can be used straight

asciilifeform: danielpbarron: i will post a patch to the netbsd src tree

asciilifeform: for what it's worth

asciilifeform: decimation: i forget if mentioned earlier: got the whole build to work

asciilifeform: decimation: at the moment, yes

asciilifeform: butugychag.

asciilifeform read that tlp article, wakes up to having known precisely the type of animal described therein - in real life

asciilifeform: having actual support for the nand would make it possible to eliminate the 'ramdisk' and shave a few MB off the memory footprint.

asciilifeform: (why do i need a conventional install? because want to test my attempt at driver for the nand as a kernel mod)

asciilifeform proceeded to stuff in a drive and... fire it

asciilifeform realized that he built the fscking netbsd -installer- into the eeprom

asciilifeform: mircea_popescu: i do recall this, now that you mentioned it

asciilifeform: ^ true at time of writing

asciilifeform: ^ anyone -not- recognize source ?

asciilifeform: l0l

asciilifeform: but long prior

asciilifeform: well yes

asciilifeform: interestingly, hitler's 'table talk' contains a few instances of whining that he'd like xxxxx but the orders don't get carried out

asciilifeform: mircea_popescu suggests that such an order would not be obeyed? probably true

asciilifeform: ?

asciilifeform: but doesn't

asciilifeform: sooo >> usg could mandate pgp for letters of credit just as easily as it mandates the garbage we see today

asciilifeform: but none of this is so much as a mild annoyance in Castle Mircea_popescustein proper

asciilifeform: anywhere it matters

asciilifeform: is it possible for someone to steal the identity of mircea_popescu? if not, why not ?

asciilifeform: http://log.bitcoin-assets.com//?date=06-02-2015#1009868 << see also ☝︎

asciilifeform: ?

asciilifeform: how is the above picture mistaken

asciilifeform: l0l

asciilifeform: - usg.

asciilifeform: why is it possible to borrow $maxint in mr. chump's name without his pgp key

asciilifeform: aha

asciilifeform: any single usg employee's security, in all respects, continues at the MERCY of unidentified party. in no sense and in no manner does it flow from usg-anything << to be fair, the very reason why 'identity theft' is a real problem for anybody is usg

asciilifeform: http://www.loper-os.org/?p=41 << obligatory

asciilifeform: these two 'engineerings'

asciilifeform: check back on it in a decade, they'll look rather similar

asciilifeform: it's approximately how software 'engineering' is done now

asciilifeform: that's the 'big secret', really

asciilifeform: roi - terrible

asciilifeform: by and large none of it works

asciilifeform: not same principle. snippets of synthetic nucleic acid are used, it goes (or more often, doesn't..) to a specific place, there is a (usually bad) protein model to predict effect, etc.

asciilifeform: nope.

asciilifeform: who and where is still doing this, i cannot say

asciilifeform: when folks would 'let that thing eat gamma until almost dead, then plant it'

asciilifeform: that style of 'engineering' is 1940s state of the art