log☇︎
15200+ entries in 0.108s
mircea_popescu: ie, bitcoin is 0.02% complete. yet something tells me the next ten years are going to see a lot more completion than the first ten.
mircea_popescu: consequently bitcoin is merely using 0.02 to 0.01% of world energy generation, less than the 50%+1 it's supposed to use by a margin of say 5000.
mircea_popescu: to continue the http://btcbase.org/log/2019-02-10#1894616 discussion / produce a benchmark for the republic's development : taking mining efficiency at 20 GH/joule (slightly above the antiminer s9) 40 exahash would be ~2 Gjoules. ☝︎
asciilifeform: btw , to go with http://trilema.com/2019/so-what-is-the-man-saying , really oughta disasm a zcx variant and longjmp side by side and see what actually changes. ( when diana_coman comes back with working bins, i'ma set this up , for thread-co) ☟︎
mircea_popescu: but the calls thing can be made any arbitrary size with a switch. you want it 16777216 rather than 65536 is the idea ?
a111: Logged on 2019-02-14 07:49 diana_coman: according to docs, the mere presence of a handler of exception slows the whole things down when lj
mircea_popescu: (honestly i never heard of a program that properly used 64k stack frames ; seems if truly one needs such depths, one's welcome to fucking rewrite something, recursion be damned.)
mircea_popescu: IF indeed there's a significant difference between call and loop re that cost, this'll bring it out.
mircea_popescu: diana_coman : btw, here's my current model for the calling timing harness : write three procedures, A B C. have each of these 1. increment a global counter, X ; 2. check if X is over a max value ; 3. if it is not, have each call either one or the other of the other two randomly ; 4. if X is over max value, have them simply return. ☟︎
mircea_popescu: but there is no such thing as a TRULY unhandled exception. either it hoses the box or else it goes to the default handler.
diana_coman: being starter, I preferred not to force a choice there; but at any rate, if the previous node is basically broken as I gather that's certainly a problem
mircea_popescu: but i mean, your code would have handled exceptions if they arose, yes ? if a for looped out of bounds, or whatever. isn't it so ?
diana_coman: according to docs, the mere presence of a handler of exception slows the whole things down when lj ☟︎
diana_coman: http://btcbase.org/log/2019-02-14#1896320 -> hm, trinque, do you suspect it's really just down to V version? I can easily re-run the thing with a V pressed to same node as yours to rule that out, if that's the case ☝︎☟︎
BingoBoingo: mod6 seems to be in an uncomfortable place resembling where I was exactly a year ago except the people around him speak his mom's tongue. I hope he can pupate in *spite* of that
mircea_popescu: anyway, this trinque - mod6 exchange's gonna be a thing for the ages.
trinque cheats and just boots a moduletronic kernel, then "make localyesconfig" and rebuilds
a111: Logged on 2019-02-09 19:58 asciilifeform: ( change all yer 'M' to a 'Y' .. )
trinque: mod6: a/profiles/home/mod6/cuntoo/cuntoo/build/usr/portage/profiles/releases/17.0/package.use.mask << this for example is not a valid path for anything
trinque: I suspect I have a different vdiff than both of you
mod6: ok so next marching orders are to not do the USB Stick, and `lsmod -v` compare; but instead, will take a few days to look into initramfs, and then try to build another?
mircea_popescu: autoimmune encephalitis ; it actually is a thing.
mircea_popescu: im sure he fucked a bunch of seattle cokewhores without asking them.
mircea_popescu: they do a lot of this http://trilema.com/2018/the-night-of-the-hunter/#selection-133.0-133.542 lately, don't they.
mod6: hmm, nope didn't verify. I threw in a sha512 of each sigfile just to ensure that I have the correct ones: http://p.bvulpes.com/pastes/CCU2l/?raw=true
a111: Logged on 2019-02-14 03:04 trinque: I still want to know whether your genesis.vpatch matches mine, and this is at least as important as whether it produced a bootable drive
mircea_popescu: in any case, on a sane system that's stable there's ~0 reason to have the CODE on disk.
mircea_popescu: but a lot of this before we care about that.
mod6: I suppose there's some good wisdom in that. I wouldn't want mircea_popescu torch his trb if it does weird shit on a 'getinfo' either.
trinque is quite familiar with the impulse to torch the thing when it misbehaves, but it's not going to get us a distro we own.
mod6: you know? I think learning about it is a fine thing, probably would make me understand the errors of my ways.
mod6: you know, at one point i did actually handroll an initramfs on cuntoo, at which point I updated the /etc/lilo.conf and when I booted, and it panic'd, it did drop me into a shell.
trinque: larger context here is I'm not eating a linux distribution by myself
trinque: yeah, comparing a boot with initramfs to one without would be a fine thing; even better would be for you to bake your own initramfs sometime with say busybox, and have the init script drop to a shell so you can see what it is
mod6: and if alf's way seems like a sane thing to try, that's what I'm working on now. otherwise, I'm all ears and can adjust as you see fit.
mod6: i suspect that if we did have a beer irl, some of this would be cleared up.
mod6: Anything that even remotely looks like the finish line seems like a win -- even if it's not.
mod6: I'll try to do this the way that #t sees fit. but, on my own, left to my own decision making (as I don't know much about these things) might be a bit askew.
trinque: I very much want you on an actual cuntoo because one of the immediate needs is a trb ebuild.
trinque: I don't even think your box is a piece of shit.
trinque: recall the work with trb, and how much better a patch that removes is than one that adds.
mod6: But alf helped me to realize that there is a way we can still try to diagnose this issue by trying to boot off of USB stick, so I'm going to attempt that first before any more initramfs stuff.
trinque: so then. a core claim I'm making is *this is all I have ever needed to boot a sane linux on commodity hardware for years*
mod6: I posted a jpg of the beginning of my genkernel. Basically there is another similar command to make the initramfs. Both end in failure that I spoke about above.
trinque: since I have a fetish for futility, lemme press on
trinque: so your goal is to have a computer for mod6 on the back of my work, and if it incidentally helps get the cuntoo thing done, w/e
mod6: But not matter what happens, I'll be sure to provide you with any information that I can provide. Who knows, maybe this is a shit system or some such.
mod6: well, that's a win for me, the only thing that sucks is you don't get to make your improvements from my own experience.
mod6: Maybe we're talking past eachother a bit here. Anyway, I don't know much about these things. I'm kinda learning bit by bit as I go... it all certainly doesn't "fit in head" or whatever yet.
trinque: eh dude, explaining to you what's an initramfs is not calling you a fool.
mod6: Sometimes that means taking a shot in the dark, even if it's misguided from the point of view of someone who knows better.
mod6: I'm just a guy, trying to boot the thing so I can test something totally not related to kernel mods.
mod6: I'm not a "kernel hacker guy", so I've been down these roads based on trying to emulate what I see working in my other machines, and various online posts in teh gentoo forums and others.
trinque: I still want to know whether your genesis.vpatch matches mine, and this is at least as important as whether it produced a bootable drive ☟︎
trinque: initramfs is used when the kernel can't mount the root fs without using a program to do something complicated first, i.e. loading firmware, unpacking a squashfs in an embedded device, etc
mod6: Might take a bit, but I'll begin work on that tomorrow-ish. Should have something to look at hopefully by Friday evening-ish. I'll inflate an entirely fresh cuntoo.
asciilifeform: i expect you'll need a root=/dev/sda3 in the kernel cmdline and similar in fstab.
mod6: so, `dd` the entire cuntoo ssd disk onto a usb, then attempt to boot the USB, and then do the lsmod diff between the USB cuntoo and the SSD cuntoo?
asciilifeform: mod6: quickest way to learn wtf, i suspect, is to dump the thing bitwise onto a usb, and boot that, then lsmod -v and diff with your known working set's lsmod -v output.
mod6: So next mission is to use a clean USB drive, and inflate cuntoo onto that, and then try?
mircea_popescu: this is a good point ; basically i misread that line.
asciilifeform: all he's got, is a [ 0.971934] SCSI subsystem initialized , which happens when kernel brought up regardless of whether any module matches up to working disk hanger
asciilifeform: ^ when finds a sata .
asciilifeform: here, have a line from a working box : [ 3.129745] sd 0:0:0:0: Attached scsi generic sg0 type 0
mircea_popescu: ie, from a cursory look at his published logs, my impression was that the kernel has sata just fine, but the disk's not plugged in the config-set hole or somesuch.
mircea_popescu: a. hm.
a111: Logged on 2019-02-14 01:56 mod6: it's like, fuck this, I want a "GIVE_ME_EVERYTHING_YOUVE_GOT_WHORE" flag that build every driver known.
asciilifeform: it dun give a damn
asciilifeform: mod6: you get a 500MB kernel then, lol
mod6: it's like, fuck this, I want a "GIVE_ME_EVERYTHING_YOUVE_GOT_WHORE" flag that build every driver known. ☟︎
asciilifeform: ( linux is notorious for 'oh hey we added a new required flag and ha even tho you set x, y, z, device d no longer gets its module because fuckyou ' )
mod6: Could be, a solid way to try to prove that is the USB Stick route. Because, I'm not super well versed in SATA kernel drivers.
mod6: Which is why, in the first place, I thought I had a bad SSD. Which is why I ended up buying a second one just to be sure.
mircea_popescu: BingoBoingo gotta have a great hook.
mod6: I did look through the makemenuconfig, there are a lot of added in USB drivers. however, there was one option that I noticed that I should look hard at (all when trying to do the genkernel for different reasons), which is the Intel USB Drivers.
asciilifeform: supposing, that is, you didn't also manage to produce a kernel that dun see yer usb chip
mod6: Ok, so you're saying: Do all of the above steps, but instead of using a SATA SSD, inflate cuntoo onto a USB stick?
asciilifeform: mod6: if you were to take this exact thing, and stick it onto a usb , and then it boots when '/dev/sda3' -- then you will know that the above is it
mod6: Right, well, cuntoo expands all of those things and build it, you just feed it a config. So if there is something that is "off" about my config with some other slightly different source version, then that might be a part of it.
mircea_popescu: asciilifeform it'd be a lot more possible if the same kernel on same machine didn't boot before.
asciilifeform: mod6: bring up a working barbarian linux on that box and lsmod -v
asciilifeform: mod6: seems like you may have build a kernel that dun see your sata chipset.
mod6: So perhaps im kindof on the right track - use UUIDs where I can, and build an initramfs... however, haven't had a lot of success with building the initramfs yet. I may have to fight through that. But that's the latest update.
mod6: my disk is a WD 250Gb SSD SATA
mod6: I think that I might be on the right track here though, and I did try a few other things after reading some documentation. For instance, after the above, I changed /etc/fstab to use only UUIDs instead of '/dev/sda{1,2,3}', and then tried that. Same problem essentialy. I also found about 'PARTUUID', which is supposed to help in certain circumstances. Nothing has worked yet...
mod6: http://p.bvulpes.com/pastes/BoTML/?raw=true << Ok, so after installing Cuntoo, I did what I said I'd do, which was test editing the append section and throwing a UUID in there instead of 'root=/dev/sda3'. It didn't work, I did get a kern dump.
mircea_popescu: i'd like a purpose made item, fit in head as such, can keep for later.
asciilifeform: ( otherwise it aint a call, but a 1way ticket, i.e. jump )
mircea_popescu: i've been thinking about how to correctly construct a calling test for this purpose, but i confess nothing i have yet is passing muster. if anyone wants to step in.
mircea_popescu: asciilifeform yes, but it's the job of "optimizing" compiler to keep all that to a low roar.
a111: Logged on 2019-02-12 18:59 mircea_popescu: basically idea is, a markov chain of callings.
mircea_popescu: should possibly also do the http://btcbase.org/log/2019-02-12#1895520 test tho (even though on a good compiler, there really shouldn't be much difference between a call and a loop) ☝︎
asciilifeform: mircea_popescu: if you give a damn re exception-propagation speed, prolly oughta measure that. otherwise seems like the docs didn't lie, it dun do anyffin to ordinary jump
mircea_popescu: in fact in some cases it could be the case IT IS ACTUALLY FASTER (to a very small degree)
a111: Logged on 2019-02-13 21:40 diana_coman: since it seems it'll take a while until I can add to my data the numbers for sjlj on ave1's gnat as well, here's what I have so far: http://p.bvulpes.com/pastes/1esL2/?raw=true
mircea_popescu: http://btcbase.org/log/2019-02-13#1895836 << so on the basis of this better table neatly presenting data i'm concluding that a) serpent run indeed takes 2.8 us or so ; b) timing data converges within 1/3 s test runs or so ; c) these statements equal to foregoing earlier items which are thus retrospectively deemed correct and finally, and most importantly d) tentatively it seems sjlj adds no measurable time delay on running co ☝︎
asciilifeform: ( ~= 'a bullet hit stierlitz in his head. 'a dumdum', he deduced, scatter-brained.' )
mircea_popescu: "what goes through a fly's mind as it crashes on windshield ?" "its arse".
mircea_popescu: it's a convenient fiction trope, this, that a merciful god on the moment of closing accounts points out the hole.
a111: Logged on 2019-02-13 23:43 asciilifeform: mircea_popescu: these folx seem to have a psychiatrically-classical 'царь-батюшка' imaginary friendship w/ trump