Fixpoint

2024-01-02

#jwrd Logs for Jan 2024

Filed under: #jwrd logs, Logs — Jacob Welsh @ 17:30
Day changed to 2024-01-02
[17:30] jfw: whaack: re flashrom support for Thinkpad X200 variants, I tracked down in the code where it's checking the PCI IDs (board_enable.c), so if yours indeed don't match, that'd be why; and I'm not seeing any relevant additions in the upstream latest (v1.3.0). so, we can try patching that whitelist and hopefully it doesn't brick the embedded
[17:30] sourcerer: 2023-12-20 17:05:14 (#jwrd) jfw: that was http://jfxpt.com/paste/v8iqychdcn - the "lspci -nn" output on whaack's thinkpad which flashrom was complaining about being "unsupported".
[17:30] jfw: controller or something as the warnings cry about. perhaps I'll bring the offline flashing kit on my visit and we can try it then, so we'll have a recovery option.
[17:32] jfw: and happy 2024, #jwrd.
[17:35] jfw: actually, it looked like it should be possible to use the Bus Blaster or similar FTDI based gadgets to do the SPI programming, removing the need for raspberrypi with its respective care & feeding from our process
[17:35] jfw: which would be especially interesting if we had a reliable supply of such gadgets :/
[17:37] jfw: but that might actually make for an easier first hardware project compared to the TRNG. at minimum because it's easier to see if it's working correctly or not
[17:53] jfw: that one's basically still stuck on finding someone with a nice oscilloscope to borrow here in the jungle, or committing to buy one outright
Day changed to 2024-01-03
[14:52] jfw: http://jfxpt.com/2023/jwrd-logs-for-Dec-2023/#10135 - recovered, http://jfxpt.com/paste/7xwuq32cvu
[14:52] sourcerer: 2023-12-28 19:29:08 (#jwrd) jfw: meanwhile local power outages knocked out my browser tab with this one loaded, and it expired from the pastebin
Day changed to 2024-01-05
[02:14] whaack: jfw: hey I messaged you through other channels, get back to me when you get the chance. I'll be diving tomorrow and will be free after 1 PM CR time.
[03:25] jfw has done so.
Day changed to 2024-01-11
[19:38] jfw: ahoy whaack, from back home
[19:38] whaack: jfw: was the rest of your trip smooth sailing?
[19:39] whaack: Welcome back!
[19:40] jfw: SJO sucked, hefty queues at each step - checkin, migracion, and security; got to the gate 45 mins before scheduled departure. the plane got going on time but then had to wait 20 mins for takeoff.
[19:40] jfw: besides that, all fine.
Day changed to 2024-01-12
[04:07] whaack: jfw: what a pain. I guess that's part of coming during 'high season' because i typically spend <5 minutes before I'm at my gate.
[04:10] whaack: There was a lot of components I didn't fully grok wrt your lofty goals for your scheme interpreter that you explained to me during lunch. I don't know if now's the time but I would like to fill in the gaps of my understanding.
[04:12] whaack: Per our discussion it's time to get right back to business... i'll start here http://jfxpt.com/2023/jwrd-logs-for-Apr-2023/#6910
[04:12] sourcerer: 2023-04-18 23:12:26 (#jwrd) jfw: but that all looked rather messy & difficult to follow, at best; furthermore I'm not sure if it's even required if you take the route of transforming the EC operations, such as using the Montgomery form. so that was my next thing to look into.
[15:28] jfw: http://jfxpt.com/2024/jwrd-logs-for-Jan-2024/#10158 - the main idea is 1) preserve the C code as a bridge for bootstrapping but 2) transform the interpreter guts into an abstract scheme-based assembly-style language which 3) can be automatically translated back to C to continue working on current machines or 4) likewise translated to a concrete assembly language to remove the C compiler dependency
[15:28] sourcerer: 2024-01-12 04:10:11 (#jwrd) whaack: There was a lot of components I didn't fully grok wrt your lofty goals for your scheme interpreter that you explained to me during lunch. I don't know if now's the time but I would like to fill in the gaps of my understanding.
[15:28] jfw: or 5) assembled as a microcode table for a custom processor, perhaps on fpga, removing the C machine dependency.
[15:30] jfw: the last part then seems to demand either rewriting everything in scheme or else defining more narrowly what it's all for
[15:33] jfw: since the wallet is the most significant thing it's used for so far, improving that and getting more usage of it would help make the case for resuming the interpreter work.
[15:40] jfw: whaack, I haven't gone any further with Montgomery form so that's still open indeed.
[15:49] jfw: for the logs, we completed this, adding the variant PCI ID for the LPC controller, and flashing was successful. we also had to reconfigure & rebuild the coreboot image because the machine has a 4MB ROM (Atmel if I recall) rather than the more common 8MB.
[15:49] sourcerer: 2024-01-02 17:30:56 (#jwrd) jfw: whaack: re flashrom support for Thinkpad X200 variants, I tracked down in the code where it's checking the PCI IDs (board_enable.c), so if yours indeed don't match, that'd be why; and I'm not seeing any relevant additions in the upstream latest (v1.3.0). so, we can try patching that whitelist and hopefully it d
[15:56] jfw: I'm still getting in trouble here for my dainty eating habits, apparently latinos or at least some consider watermelon 100% edible, including the mature black seeds, white flesh under the rind and, if pressed, even the rind itself
[15:59] jfw: next thing I'll hear they eat lobster shells too
[16:38] jfw: whaack, I think you'd said that inner ear pressure equalization happens automatically during decompression (ascent)? why then in an ascending plane would I get discomfort in the ears unless I'm actively working the yawn muscles?
[16:41] jfw: on the descent I tried out nose pinching methods instead, both giving pressure from lungs (valsalva?) and from the mouth with closed glottis. it seemed to require significant pressure to open the tubes - with the right side usually giving way first - and easy to overdo, yet still had to be repeated quite frequently
[18:52] whaack: jfw: correct, during the ascent the pressure builds up inside the middle ear which causes pressure to the eustachian tubes, which causes them to open, resulting in automatic equalization. It is possible for mucus/ inflammation to stop this process from working properly. This it is known in freediving as a 'reverse block' and in an extremely nasty situation to be in because you have to stop/slow your ascent. in the worst case you need to rupture your air drum
[18:52] whaack: s to reach the surface
[18:53] whaack: It's not normal to need to equalize on the ascent in a plane. But I also doubt you have a reverse block. So it's possible that you're just very sensitive to the eardrum flexing at all
[18:56] whaack: That said, maybe you do have some mucus blocking your tubes if you had trouble equalizing on the descent as well. And yes you must be careful not to overdo it, the tongue and larynx are strong enough to create enough pressure to rupture your eardrum. I'm sure you can rupture your eardrum from valsalva as well.
[18:57] whaack: Proper equalization technique is a very gentle motion
[19:15] jfw: whaack: hard to say if "trouble" or just not used to it.
[19:17] jfw: as to ascent, possibly I'm just sensitive, yes, though it seemed to only get worse if I ignored it; it's always been like that for me, ascending and descending. likewise for fast elevators in tall buildings (maybe 50+ floors)
[21:39] whaack: jfw: for ascending you can try to learn the 'reverse frnzel' where instead of creating pressure in the Naval cavities, you create a vacuum.
[21:47] jfw: nasal & frenzel I assume.
[21:57] whaack: jfw: anyways... do you have any recommendations for books to get me up to speed with abstract algebra in general?
[21:57] whaack: and yes
[21:59] jfw: hm. well, in general an algebra defines a set of values and some operations on them
[22:00] jfw: then explores what follows from those definitions
[22:01] jfw: on the shelf I have an 'elements of modern algebra' by gilbert & gilbert
[22:01] jfw: 7th ed.
[22:02] jfw: I don't have much memory of it specifically
[22:03] jfw: possibly the ECC/ECDSA standards include enough background on preliminaries to get started?
[22:06] jfw: plus the ol' Handbook of Applied Cryptography
[22:07] jfw: http://jfxpt.com/codeview/tree/gbw-signer/gbw-signer_static_bit_ops_reindent/gbw-signer/library/ecdsa.scm#L238
[22:08] whaack: jfw: ty
Day changed to 2024-01-14
[18:40] whaack: jfw: I've been enjoying going through the handbook, only ~550 pages to go until I get to Montgomery forms!
Day changed to 2024-01-15
[03:58] jfw: whaack, happy hunting!
[15:25] dorion_road: whaack, how goes ? are you ready to resume the training this week ?
[21:37] whaack: I can do an afternoon session on Wednesday but not a morning one
Day changed to 2024-01-16
[02:13] jfw: whaack: aok, 4pm panama / 3pm ticotime? and did you get to pick up the remaining supplies?
Day changed to 2024-01-17
[16:36] whaack: jfw: 3pm works, I still haven't picked up the cables. I actually just got back from San Jose last night
[16:36] whaack: 3pm my time*
[21:02] whaack_: joining in 1m
Day changed to 2024-01-18
[00:46] jfw: whaack: http://dangerousprototypes.com/docs/Bus_Blaster_v4_design_overview#Reprogramming_logic - indeed the v4 changed the self-programming aspect
[00:47] jfw: that's what the extra 3-pin jumper you found is for - move it from "normal" to "update buffer" and then the "introspection" test should work as I showed.
[00:53] jfw: and I've updated the outline now for various details.
[00:56] jfw: for the log - he turned up a mini-USB cable so we got as far as building xc3sprog, tracking down USB device node numberings, and (it seems) getting xc3sprog talking to the Bus Blaster; so it just remains to wire that up to the FUCKGOATS JTAG header.
[01:06] jfw: & recording is up.
[02:21] dorion_road: jfw, whaack, nice !
[20:32] whaack: dorion_road: it feels good man, there's some parts where I feel like i am just going through the motions, but I am picking up understanding along the way.
[20:33] whaack: jfw: I picked up the cables, i'll keep you posted
[21:27] whaack: dorion_road: your blog is down for me
[21:53] jfw: working from here
[21:57] whaack: weird, seems to be a me problem. dunno why
[21:58] whaack: works for my phone on the same wifi
[21:59] jfw: what kind of machine is the problem?
Day changed to 2024-01-19
[21:58] whaack: jfw: a mac
Day changed to 2024-01-22
[18:06] dorion_road: hola el_pirata_loco, como andas ?
[18:07] dorion_road: hey, sstacksyrc, how've you been ?
[18:17] el_pirata_loco: todos bien!
[18:19] dorion_road: cool, who might you be ?
[23:55] jfw: who knew, you can deter pirates on the high seas just by asking such discomforting questions!
Day changed to 2024-01-23
[15:08] whaack: aska me my's name, and i'lls have ur walk dah plank!
[19:42] whaack: What are you up to nowadays dorian?
[20:36] dorion_road: whaack, been in nyc the past coupla weeks visiting bobbybuckets.
[20:40] dorion_road: currently seated @ the oculus.
[20:41] dorion_road: about to go see if I can find where tontine coffee house was located.
[20:42] dorion_road: whaack, had a nice call w/ Senyor Sher last week. cool guy, I suspect something will come of it.
[21:39] whaack: dorion: I am unfamiliar with 'the oculus' as well as the coffeehouse. Good to hear that the call went well
[22:58] dorion_road: whaack, teh oculus. from the inside looking up, resembles a giant cunt.
[23:01] dorion_road: the coffee house hasn't existed for a while, but it's was basically the first stock market here.
Day changed to 2024-01-24
[14:43] whaack: jfw: we are meeting in 15 minutes, correct?
[14:59] jfw: whaack: yep, I'll be there shortly
[17:40] whaack: i have a contact interested in buying 3-4 btc, if any1 is interested gpg me
[17:40] whaack: partial is ok too
[18:36] dorion_road: whaack, http://jfxpt.com/paste/fnmw84kf6e
[19:24] jfw: whaack, how/where can the contact deliver fiat?
[19:25] whaack: it would be via international wire transfer from a euro bank
[19:27] jfw: ok, don't think I can help
[19:29] jfw: I mean, I guess I can ask around if they can't fill it otherwise
Day changed to 2024-01-29
[05:27] jfw: http://jfxpt.com/2024/spreading-entropy-with-rngnet-a-first-software-edition/
[15:11] dorion_road: jfw, nice. looks good.
[17:14] jfw: whaack, how's the router install going and how's the wednesday schedule looking?
[17:15] jfw: I'll have some guests in town so I'm just as happy to push it back a week if you prefer
Day changed to 2024-01-30
[14:05] whaack: jfw: let's move to next week
[14:09] jfw: alright
Day changed to 2024-01-31
[00:45] jfw: just hit 4 different shops in El Dorado where the computer parts live, nobody has an ATX PC case of the sort that was formerly commonplace, namely 1) no stupid side glass cutout, 2) 120mm fan mount, 3) at least 3x 3.5" HDD bays.
[00:46] jfw: there's a small selection matching 1 and somewhat larger matching 2 (though more of the "plaster the whole front panel with fans and rgb leds" variety)
[00:47] jfw: I imagine it's not just a panama problem either. but I'll try mail order since surely someone in the world will have it (heh)
[00:50] jfw: for consolation I did score a line-interactive APC BackUPS BX1350 for a modest ~501k sat.
[00:56] jfw: line-interactive aka AVR is a step in between the cheaper 'standby' and significantly costlier 'online' (double conversion), it means it can rectify some input conditions, like under- or over-voltage within some range, without having to cut over to battery, thus extending battery longevity by reduced wear in routine operation.
[18:42] whaack: jfw: my 'offline' x200 does not light up at all when i plug in the battery, can't turn it on, any thoughts on where to start? i may take the battery from the other laptop and try that
[20:38] whaack: lmk if you think i should try that, im not sure if that is a safe idea
[21:16] jfw: whaack: I can't quite tell what you mean, does it work with the power cord?
[21:18] jfw: it's true that there's some digital communication between battery pack and computer, eg for make/model/serial and capacity reporting, so leaks can't be ruled out when swapping them
[21:21] whaack: jfw: does not work with the power cord
[21:22] whaack: my mac requires the battery to work to function, even with power cord, and perhaps i mistakenly thought that (dumb) design was universal
[21:30] jfw: I can test that now
[21:32] jfw: here it boots up fine without battery; and whether booted or not there's a green plug shaped LED when plugged
[21:33] jfw: plug symbol I mean
[21:33] whaack: yup, i dont get that plug symbol
[21:36] jfw: sounds not so good. I guess give it a little more time to re-check the obvious stuff like the outlet and cord connections or whatever
[21:37] jfw: around when was this one last booted?
[21:37] whaack: long time ago, not sure
[23:43] jfw: whaack, did you try also with power cord and no battery?

No Comments »

No comments yet.

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by MP-WP. Copyright Jacob Welsh.