**** BEGIN LOGGING AT Thu Nov 19 02:59:58 2015
Nov 19 03:13:59 <Set_> https://packages.debian.org/jessie/utils/mtp-tools
Nov 19 03:15:55 <Set_> I just searched the entire catalog. It says, "Sorry, your search gave no results," this for mt7601-sta-dkms.
Nov 19 03:19:47 <Set_> I typed "mt" only and found many items.
Nov 19 03:21:37 <Elray> i think it's a pakege of ubuntu
Nov 19 03:21:45 <Elray> pakage
Nov 19 03:25:01 <Set_> Oh...okay. I am out then.
Nov 19 03:25:29 <Set_> I guess going to Ubuntu's wiki would get some good background on the issue.
Nov 19 03:26:59 <Set_> https://packages.debian.org/jessie/broadcom-sta-dkms
Nov 19 03:38:39 <Set_> Elray...try this: https://packages.debian.org/jessie/wireless-tools
Nov 19 05:47:02 <Set_> I need to reboot. Enjoy...I should be back with more, more, more.
Nov 19 08:32:52 <palehorse> hey
Nov 19 08:32:56 <palehorse> anyone here?
Nov 19 08:38:12 <palehorse> ?
Nov 19 08:41:32 <av500> no
Nov 19 08:41:34 <GrumpeiYokoi> I'm here.... and he's gone
Nov 19 10:48:39 <kartik> can anyone plz help me to know how to write drivers for beaglebone..
Nov 19 10:48:49 <kartik> im a newbie.. thanx in advance
Nov 19 10:49:43 <av500> https://lwn.net/Kernel/LDD3/
Nov 19 10:49:54 <av500> assuming you mean kernel drivers
Nov 19 10:53:59 <kartik> sry for asking a silly question .. but how much time it will take me to be a contributor like u guyz
Nov 19 10:54:17 <kartik> i know basics of c and VCS for now
Nov 19 10:57:59 <kartik> and thanx for the link which gives necessary guide to linux kernel driver making
Nov 19 10:58:02 <GrumpeiYokoi> As long as it takes? :)
Nov 19 10:58:11 <GrumpeiYokoi> I wouldn't rush anything.
Nov 19 10:58:51 <GrumpeiYokoi> Or rather, be concerned about some kind of timeframe for contributing.
Nov 19 11:01:01 <kartik> thanx.. i love making projects using avr, arduino and have got a strong hold onto it. that made me interested into making drivers for beaglebone
Nov 19 11:01:13 <av500> kartik: there is no "u guyz"
Nov 19 11:01:18 <kartik> can u plz help me by tellling me what all i should learn
Nov 19 11:01:33 <av500> kartik: you are asking the wrong question
Nov 19 11:01:41 <av500> there is no recipe
Nov 19 11:01:46 <av500> just dive in and get going
Nov 19 11:01:59 <av500> and there is the whole internet that you can use
Nov 19 11:02:07 <kartik> sry av500.. but how to dive in .. whr to start from
Nov 19 11:02:24 <av500> why do you want to "write a driver"
Nov 19 11:02:28 <av500> ?
Nov 19 11:02:39 <av500> are you sure you know what that means?
Nov 19 11:02:49 <GrumpeiYokoi> It begs the question, "a driver for what?"
Nov 19 11:02:57 <av500> if you mean a linux driver, then read https://lwn.net/Kernel/LDD3/
Nov 19 11:03:19 <av500> if its an "arduino type of driver", you mean a library
Nov 19 11:03:23 <GrumpeiYokoi> I've been itching to write a kernel mode driver for a device we make at work, but I have stalled on that.
Nov 19 11:03:43 <av500> GrumpeiYokoi: the trick is to hand out the device to somebody else
Nov 19 11:03:47 <av500> free driver... :)
Nov 19 11:03:49 <GrumpeiYokoi> The impression I get is that the Linux kernel has a lot more flexibility on implementing drivers than Windows
Nov 19 11:04:23 <GrumpeiYokoi> like, our Windows driver uses pretty much just pipes and ioctl
Nov 19 11:04:43 <GrumpeiYokoi> I suppose there is the capability to do more in Windows, I don't really know.
Nov 19 11:04:47 <kartik> coz i would love to integrate different peripherals and making them work. so thanx for the link.. and then what next..
Nov 19 11:05:03 <GrumpeiYokoi> (By that I mean creating files and character devices)
Nov 19 11:05:16 <GrumpeiYokoi> (It's a USB device btw)
Nov 19 11:05:31 <av500> kartik: I dont know what next
Nov 19 11:05:39 <av500> you read that thing pretty fast
Nov 19 11:05:56 <GrumpeiYokoi> av500: Actually yeah my boss told me that one of our customers did a large chunk of our kernel mode driver for ISA devices
Nov 19 11:05:59 <av500> also you did not answer my questions
Nov 19 11:06:16 <av500> GrumpeiYokoi: your devices secret?
Nov 19 11:06:18 <GrumpeiYokoi> The customer needed to get something working badly enough. So it was a free driver
Nov 19 11:06:36 <GrumpeiYokoi> But it was also a very complicated piece of code, for kernel stuff (I think, anyway)
Nov 19 11:06:56 <GrumpeiYokoi> av500: The secret is love
Nov 19 11:07:12 <kartik> sry av500.. that link was very helpfull and i am very eager to learn it. afterwards what am i supposed to do
Nov 19 11:07:22 <GrumpeiYokoi> The USB driver is just hacked off of the bulkusb example code in the WinDDK (WDK now)
Nov 19 11:07:35 <GrumpeiYokoi> My boss is very pragmatic <___<
Nov 19 11:07:37 <av500> kartik: I am sorry, thats not how it works
Nov 19 11:07:57 <av500> thats not a "learn how to write a driver in a week" link
Nov 19 11:08:09 <av500> [12:02] <av500> why do you want to "write a driver"
Nov 19 11:08:18 <av500> do you know what a driver is?
Nov 19 11:08:46 <kartik> coz i  love embedded systems and writing drivers for beagleboard will be fun
Nov 19 11:09:02 <kartik> yes i know..
Nov 19 11:12:20 <av500> le sigh
Nov 19 11:13:11 <GrumpeiYokoi> oui
Nov 19 11:13:14 <GrumpeiYokoi> *le oui
Nov 19 11:13:56 <GrumpeiYokoi> I almost told him to write my driver but I feel bad just foisting work on people. But he seemed so eager...
Nov 19 11:31:04 <Guest72843> Hi - anyone know how to get /dev/spidev1.0?
Nov 19 12:09:25 <zmatt> GrumpeiYokoi: I would have proposed rewriting musb, but this time make it work *well*
Nov 19 12:10:17 <GrumpeiYokoi> zmatt: It... seems to work okay in 4.2.5 for me. I haven't had a chance to diff out what voodoo might have changed.
Nov 19 12:11:02 <GrumpeiYokoi> I mainly was looking at the 3.8.13-* progression a couple weeks ago, but bone79 didn't work anyway, so it was kind of a pointless exercise
Nov 19 12:11:18 <GrumpeiYokoi> Rather, it didn't work for the hotplugging issues I was having in particular.
Nov 19 12:11:21 <zmatt> well someone recently observed that enabling DMA support currently makes it *slower*
Nov 19 12:11:31 <zmatt> that's not usually the desired outcome
Nov 19 12:16:04 <GrumpeiYokoi> Yeah, that's not good. But for my application it's probably not worth worrying about.
Nov 19 12:16:54 <GrumpeiYokoi> Certainly I feel like it's a wrong that should be righted though :/
Nov 19 12:18:33 <zmatt> I was about to say, how egotistical for you to only consider your application! haven't you thought for a moment about how much fun it would be for kartik to fix it!
Nov 19 12:18:36 <zmatt> ;)
Nov 19 12:19:53 <GrumpeiYokoi> LoL
Nov 19 12:20:11 <GrumpeiYokoi> Well he was pretty psyched up
Nov 19 12:24:21 <GrumpeiYokoi> Btw I pretty much gave up trying to figure out what was wrong with the element14 images. I tried frankensteining them to be more like the mainline images but I couldn't seem to find what curse was keeping them from working.
Nov 19 12:25:21 <zmatt> kinda "funny" mmc 2 is the only mmc interface which is directly hooked up to the L3 interconnect rather than the L4LS, yet also the mmc interface least likely to be used for anything on a BBB
Nov 19 12:25:29 <zmatt> the element14 images?
Nov 19 12:25:34 <GrumpeiYokoi> Well, gave up for now. It still pisses me off and I'll probably have another crack at it at a later date.
Nov 19 12:26:11 <GrumpeiYokoi> Oh, sorry I guess I was mainly talking to vere_mit about that (don't want to needlessly beep him).
Nov 19 12:26:22 <GrumpeiYokoi>  / them
Nov 19 12:27:13 * zmatt scrolls up
Nov 19 12:27:49 <GrumpeiYokoi> The element14 BBBs out of the box are pretty much not compatible with rcn's linux-image-* packages
Nov 19 12:27:52 * zmatt doesn't see it within the first few pages of scrollback and is too lazy to search further
Nov 19 12:27:58 <GrumpeiYokoi> zmatt: This was from days ago. Probably a week or more ago.
Nov 19 12:27:58 <zmatt> ah
Nov 19 12:28:18 <zmatt> so... reflash them with the latest jessie snapshot, problem solved
Nov 19 12:28:36 <zmatt> why are you trying to animate the living dead?
Nov 19 12:28:39 <GrumpeiYokoi> I didn't know we had a working jessie snapshot
Nov 19 12:28:43 <GrumpeiYokoi> I've been all wheezy
Nov 19 12:28:50 <zmatt> ... poor you
Nov 19 12:29:05 <GrumpeiYokoi> zmatt: To understand what is wrong with them I think is informative
Nov 19 12:29:16 <zmatt> diff?
Nov 19 12:29:42 <GrumpeiYokoi> I have been using the BBBs for about a year and a half but I've mostly been just porting our device code to it
Nov 19 12:29:57 <GrumpeiYokoi> so I wanted to understand the whole system a bit better, and it seemed like a good opportunity
Nov 19 12:30:00 <zmatt> https://rcn-ee.com/rootfs/bb.org/testing/2015-11-15/
Nov 19 12:30:07 <zmatt> latest jessie snapshot
Nov 19 12:30:21 <GrumpeiYokoi> diff of what?
Nov 19 12:30:37 <zmatt> the filesystem with a more recent one I guess
Nov 19 12:30:38 <zmatt> dunno
Nov 19 12:30:58 <GrumpeiYokoi> I did what I could, mainly diffing /boot/* and /etc/kernel/*
Nov 19 12:31:14 <GrumpeiYokoi> And learning as much as I could about uEnv.txt and uboot, etc.
Nov 19 12:31:18 <zmatt> kernel incompatibility is pretty weird
Nov 19 12:31:26 <zmatt> since it doesn't rely on much userland at all
Nov 19 12:31:33 <GrumpeiYokoi> Well, what would happen is you'd try to upgrade the kernel
Nov 19 12:31:43 <GrumpeiYokoi> But they'd be stuck at 3.8.13-bone47 on reboot
Nov 19 12:31:50 <GrumpeiYokoi> Which is what it was shipped with
Nov 19 12:31:56 <zmatt> ohh
Nov 19 12:32:01 <zmatt> they still have a separate boot partition
Nov 19 12:32:03 <zmatt> which isn't /boot
Nov 19 12:32:17 <GrumpeiYokoi> Yeah it's the fat partition
Nov 19 12:32:24 <zmatt> so the linux-image- package doesn't actually install the kernel in the right place for those systems
Nov 19 12:32:38 <zmatt> iirc it used to use some scripts for updating the kernel instead of actual packages
Nov 19 12:32:59 <zmatt> I think the image probably ships with no linux-image- package installed at all
Nov 19 12:33:00 <zmatt> ?
Nov 19 12:33:01 <GrumpeiYokoi> That may be it, now that you've pointed it out
Nov 19 12:33:09 <GrumpeiYokoi> Well, part of it
Nov 19 12:33:25 <zmatt> a package can't be partly installed
Nov 19 12:33:25 <GrumpeiYokoi> I'd have to go back and check, but I'm so sick of debugging that that I've moved on for the time being
Nov 19 12:33:33 <GrumpeiYokoi> The package installed successfully
Nov 19 12:33:35 <zmatt> I mean installed as in, according to dpkg
Nov 19 12:33:51 <GrumpeiYokoi> But the initrd and vmlinuz and yadda yadda just didn't take effect on reboot
Nov 19 12:34:04 <zmatt> yeah no I meant, as delivered it probably has no linux-image- package installed whatsoever
Nov 19 12:34:06 <GrumpeiYokoi> Well... I dunno it seemed to me that it installed with no errors
Nov 19 12:34:30 <zmatt> which would be a very clear sign that the kernel lives outside package management
Nov 19 12:34:33 <GrumpeiYokoi> The only element14 BBB I haven't reflashed is at work
Nov 19 12:34:43 <GrumpeiYokoi> sort of...
Nov 19 12:34:53 <zmatt> but I'm pretty sure of the above
Nov 19 12:34:56 <GrumpeiYokoi> The uEnv.txt called out zImage explicitly
Nov 19 12:35:06 <GrumpeiYokoi> kernel_image=zImage
Nov 19 12:35:32 <zmatt> separate boot partition not mounted at /boot (but /boot/uboot or something like that iirc)
Nov 19 12:35:32 <GrumpeiYokoi> So when uboot sees that it doesn't go looking for anything else. And zImage is in /boot/uboot
Nov 19 12:35:44 <zmatt> exactly
Nov 19 12:35:57 <GrumpeiYokoi> I was able to update the kernel manually by overwriting the zImage
Nov 19 12:36:05 <zmatt> kernel lives in /boot/uboot outside package management
Nov 19 12:36:05 <GrumpeiYokoi> and the initrd.img file
Nov 19 12:36:17 <zmatt> and iirc it used to be updated using scripts living somewhere in /opt or something
Nov 19 12:36:32 <GrumpeiYokoi> Yeah
Nov 19 12:36:42 <zmatt> there ya go, problem solved
Nov 19 12:36:48 <zmatt> now go install jessie
Nov 19 12:36:50 <zmatt> :)
Nov 19 12:36:53 <GrumpeiYokoi> Those scripts don't work anymore
Nov 19 12:36:59 <GrumpeiYokoi> I'm afraid of jessie <___<
Nov 19 12:37:03 <GrumpeiYokoi> I just got used to wheezy
Nov 19 12:37:04 <zmatt> that's hardly surprising
Nov 19 12:37:09 <zmatt> wheezy has been officially obsoleted
Nov 19 12:37:33 <zmatt> hell I think my own makefiles don't even work on wheezy because it's "make" is too old
Nov 19 12:37:37 <GrumpeiYokoi> I will do what my customer tells me to do, unless they're being complete idiots
Nov 19 12:37:52 <GrumpeiYokoi> They wanted to stick with 3.8.13-bone74 for no apparent reason
Nov 19 12:37:55 <zmatt> just remind them that wheezy is declared obsolete by the debian project
Nov 19 12:38:16 <GrumpeiYokoi> this is a company that to this day still buys ISA card devices from us
Nov 19 12:38:24 <GrumpeiYokoi> Do you think they care about obsolete? :D
Nov 19 12:38:36 <zmatt> and rcn-ee actually said out loud he no longer really cares about anything older than 4.1
Nov 19 12:38:57 <GrumpeiYokoi> Well I am not having problems with kernel version 4.2.5-bone2 so far
Nov 19 12:38:59 <zmatt> if they care about security updates
Nov 19 12:39:10 <GrumpeiYokoi> But, jessie vs. wheezy has more to do with userland packages, right?
Nov 19 12:39:16 <zmatt> yes
Nov 19 12:39:23 <GrumpeiYokoi> So... I'm not sure if i care about moving to jessie
Nov 19 12:39:25 <zmatt> I run stretch actually
Nov 19 12:39:27 <GrumpeiYokoi> unless it becomes a problem
Nov 19 12:39:42 <zmatt> wheezy no longer receives security updates
Nov 19 12:39:47 <zmatt> or, updates period
Nov 19 12:39:50 <GrumpeiYokoi> It does seem like some of the packages I use are wildly out of date, like svn though
Nov 19 12:40:05 <GrumpeiYokoi> It's in 1.6 era? That seems ancient
Nov 19 12:40:12 <zmatt> hello, it's debian
Nov 19 12:40:23 <zmatt> debian stable normally already has wildly out of date packages
Nov 19 12:40:37 <zmatt> if debian declares something obsolete, you can be pretty sure it *really* is
Nov 19 12:40:39 <GrumpeiYokoi> you know, I guess I'd heard that, but it didn't really click until now
Nov 19 12:40:46 <GrumpeiYokoi> I mainly use Fedora at home
Nov 19 12:40:55 <zmatt> look at your systemd version
Nov 19 12:40:57 <GrumpeiYokoi> And it seems... mostly up to date
Nov 19 12:42:05 <zmatt> your fedora presumably uses systemd-networkd instead of.. whatever it used to use before that?
Nov 19 12:43:13 <zmatt> on debian you need stretch for that since even jessie ships with too ancient systemd
Nov 19 12:43:24 <zmatt> (well, it sort-of-works on jessie but you don't even have "networkctl")
Nov 19 12:43:53 <GrumpeiYokoi> systemd --version returns "unknown option?" in red
Nov 19 12:43:56 <GrumpeiYokoi> on this BBB
Nov 19 12:44:08 <zmatt> lol
Nov 19 12:44:16 <GrumpeiYokoi> yeah...
Nov 19 12:44:23 <zmatt> try apt-cache policy systemd
Nov 19 12:45:43 <GrumpeiYokoi> 44-11+deb7u4 ?
Nov 19 12:45:51 <zmatt> LOL
Nov 19 12:45:55 <zmatt> systemd 44 ?
Nov 19 12:45:56 <zmatt> what?
Nov 19 12:46:09 <GrumpeiYokoi> Yeah I'm not sure what any of this is
Nov 19 12:46:11 <zmatt> hahahaha
Nov 19 12:46:35 <GrumpeiYokoi> http://pastebin.com/BLYHqPDp
Nov 19 12:46:56 <zmatt> http://www.freedesktop.org/software/systemd/   look at the date for systemd-44
Nov 19 12:47:22 <zmatt> (also, yay for a directory listing that actually correctly sorts numbers)
Nov 19 12:47:31 <GrumpeiYokoi> heh
Nov 19 12:47:49 <GrumpeiYokoi> See I thought by your reaction that the version number seemed way too high
Nov 19 12:47:59 <GrumpeiYokoi> But yeah, party like it's 2012 I guess
Nov 19 12:48:26 <zmatt> mostly because it was a 2-digit number while I haven't seen anything below 200 in quite a while
Nov 19 12:49:25 <zmatt> it felt a bit like reporting a 2.6.* kernel or something
Nov 19 12:49:57 <GrumpeiYokoi> Wonder if anyone is pushing a devuan BBB image out there
Nov 19 12:50:27 <zmatt> why on earth would you want to
Nov 19 12:50:51 <GrumpeiYokoi> I don't care, but systemd is divisive, last I checked
Nov 19 12:51:15 <GrumpeiYokoi> In fact, it is every time I check
Nov 19 12:51:23 <GrumpeiYokoi> People can't seem to stop arguing about it
Nov 19 12:52:06 <zmatt> plenty of complainy people yeah, but for me, every time I replaced some crufty bunch of old scripts by a systemd equivalent the result was a system that worked better in some way (often by reducing boot time)
Nov 19 12:53:25 <zmatt> package.debian.org is useful btw for getting an overview of where the various releases are w.r.t. package versions, e.g. https://packages.debian.org/search?arch=armhf&keywords=systemd
Nov 19 12:54:17 <GrumpeiYokoi> Cool, thanks. I have scoured repository sites before in moments of desperation :)
Nov 19 12:54:36 <zmatt> just install jessie :P
Nov 19 12:54:47 <GrumpeiYokoi> Moreso when I had problems with Angstrom, though.
Nov 19 12:54:58 <GrumpeiYokoi> Can you migrate from wheezy to jessie without reflashing?
Nov 19 12:55:17 <zmatt> if you like a trainwreck
Nov 19 12:55:25 <zmatt> the issue isn't so much wheezy/jessie
Nov 19 12:55:34 <zmatt> but the non-package-managed-stuff of rcn-ee
Nov 19 12:55:53 <GrumpeiYokoi> like?
Nov 19 12:56:03 <zmatt> oh wait, you're on a recent wheezy image
Nov 19 12:56:09 <zmatt> so those should be in sync on that part
Nov 19 12:56:13 <GrumpeiYokoi> Yeah it's from July I believe
Nov 19 12:56:23 <GrumpeiYokoi> I know there's a newer one
Nov 19 12:56:32 <zmatt> ok, not-too-ancient at least
Nov 19 12:56:35 <GrumpeiYokoi> For wheezy anyway
Nov 19 12:56:46 <zmatt> latest is november yeah
Nov 19 12:57:07 <zmatt> then dunno, I've never tried it on a BBB
Nov 19 12:57:19 <zmatt> though I've done dist-upgrades on my laptop often enough
Nov 19 12:57:29 <zmatt> well, I've done dist-upgrades
Nov 19 12:57:38 <zmatt> not that often since I usually very quickly end up using sid
Nov 19 12:57:57 <zmatt> after finding out even testing had a too-ancient version of some package
Nov 19 12:58:51 <GrumpeiYokoi> Migrating major versions of Fedora is kind of a pain imo
Nov 19 12:59:10 <zmatt> for our BBBs I started out with a jessie console image, removed unnecessary cruft, upgraded to stretch, and then started installing packages we needed
Nov 19 12:59:11 <GrumpeiYokoi> Ubuntu makes it a snap *shakes fist at their aptitude*
Nov 19 12:59:38 <zmatt> too bad aptitude is reeaaaaaaly slow on a BBB
Nov 19 12:59:44 <GrumpeiYokoi> *shakes fist at my own pun*
Nov 19 12:59:54 <zmatt> I still use it though
Nov 19 13:00:15 <GrumpeiYokoi> Why would aptitude be any slower on a BBB vs. desktop?
Nov 19 13:00:28 <GrumpeiYokoi> or laptop, etc
Nov 19 13:00:49 <GrumpeiYokoi> Unless there was some kind of bug that only shows up on maybe arm machines
Nov 19 13:01:33 <GrumpeiYokoi> Would be nice if I could just have another computer download all the packages and then just stream them over my LAN for install or something
Nov 19 13:01:37 <zmatt> I suspect both the slowness of a cortex-a8 compared to a quad-core intel core i5-2520M @ 2.5 GHz
Nov 19 13:01:47 <zmatt> and the slowness of eMMC
Nov 19 13:02:11 <zmatt> aptitude is already slow on my laptop
Nov 19 13:02:49 <GrumpeiYokoi> Well, I always figured that package managers were as slow as the servers they're talking to
Nov 19 13:02:56 <zmatt> no
Nov 19 13:03:15 <zmatt> aptitude's UI is so slow on a BBB it's unbearable
Nov 19 13:03:31 <GrumpeiYokoi> I'm doing everything in terminal
Nov 19 13:03:42 <GrumpeiYokoi> I've barely touched an X session on these things
Nov 19 13:03:50 <zmatt> curses ui
Nov 19 13:04:00 <GrumpeiYokoi> ok
Nov 19 13:04:07 <zmatt> I don't even have any X installed or display attached on most BBBs
Nov 19 13:04:12 <zmatt> (in fact, all but one)
Nov 19 13:04:59 <GrumpeiYokoi> Come to think of it, apt-cache has been slow for me sometimes
Nov 19 13:05:09 <GrumpeiYokoi> maybe slow search algorithms or something
Nov 19 13:05:42 <zmatt> keep an eye on the CPU and MMC leds, that should tell you which is the limiting factor
Nov 19 13:05:46 <GrumpeiYokoi> Or bad code optimization *shrug*
Nov 19 13:09:21 <zmatt> even if a dist-upgrade works, reflashing may be a lot faster, especially if you have lots of crap installed
Nov 19 13:09:44 <zmatt> also, keep in mind you'd need to have enough free MMC to first download the package before they're installed
Nov 19 13:10:04 <zmatt> if you have a console image it may work, a GUI image... phat chance
Nov 19 13:11:01 <GrumpeiYokoi> I suppose I could do it on an SD card
Nov 19 13:11:03 <zmatt> I also try to purge accumulated cruft now and then
Nov 19 13:11:16 <GrumpeiYokoi> (If larger than 4 GB)
Nov 19 13:11:23 <zmatt> nfs
Nov 19 13:11:42 <GrumpeiYokoi> I have no idea how to do that
Nov 19 13:11:53 <GrumpeiYokoi> I looked, it looked like Greek to me
Nov 19 13:12:24 <GrumpeiYokoi> Well, I actually know some ancient Greek. Let's say it looked like Korean, of which I know nothing.
Nov 19 13:12:28 <zmatt> fair enough.  or just on a big host machine inside a systemd-nspawn ... with a bit of luck the speediness of a big cpu may even outweigh the overhead of qemu
Nov 19 13:12:57 <zmatt> (yes, you can enter a BBB image using something like chroot or nspawn)
Nov 19 13:13:00 <GrumpeiYokoi> Yeah I was thinking earlier that doing it in an emulator would be nice if I could swing it
Nov 19 13:13:17 <zmatt> at least on debian it's trivially easy
Nov 19 13:13:32 <GrumpeiYokoi> I read about it a  bit in Derek Molloy's book
Nov 19 13:13:55 <GrumpeiYokoi> I had other things on my mind at the time though, so didn't really get my hands too dirty
Nov 19 13:15:31 <zmatt> install qemu-user-static, copy /usr/bin/qemu-arm-static to $targetfs/usr/bin/
Nov 19 13:15:40 <zmatt> systemd-nspawn -D $targetfs
Nov 19 13:16:01 <zmatt> and you got a shell inside your targetfs, even though it's an armhf one
Nov 19 13:16:32 <zmatt> very useful for repairing a borked BBB
Nov 19 13:18:41 <zmatt> and at least *recent* systemd version also seem quite aware of the fact you're inside an nspawn and don't try to do anything too silly
Nov 19 13:20:23 <Bobodk> Is there a new revision of the beaglebone black, since it's out of stock everywhere
Nov 19 13:27:51 <zmatt> dx.com has 'em
Nov 19 13:28:35 <zmatt> http://www.dx.com/p/-318741
Nov 19 13:30:07 <zmatt> there's no new revision
Nov 19 13:30:44 <GrumpeiYokoi> element14 (newark) seems to have 4000 of them
Nov 19 13:31:12 <zmatt> oh he left already
Nov 19 13:31:40 <zmatt> argh @ people with hit-and-run questions
Nov 19 13:31:42 <GrumpeiYokoi> I turned off conference mode in this channel, and still missed it XD
Nov 19 13:32:12 <zmatt> yeah I hide all join/part/quit noise since there's just too much of it here
Nov 19 13:32:40 <zmatt> I just add exceptions if I need to pounce someone on sight
Nov 19 13:32:41 <GrumpeiYokoi> Um..... it's not really that noisey here in that regard. Everyone seems to lurk a lot here
Nov 19 13:33:00 <GrumpeiYokoi> I used to be on the llvm channel and that was pretty bad so i kept conference mode on
Nov 19 13:33:14 <GrumpeiYokoi> Or when someone is having ping out problems it can get pretty annoying
Nov 19 13:35:01 <GrumpeiYokoi> zmatt: Do you happen to know what controls auto mounting of removable media on these things? Is it just a bunch of udev rules? I hope not, b/c I hate reading udev rules files. I'm not having a problem, everything seems to work great, just trying to learn or at least get an idea of where to look.
Nov 19 13:37:10 <GrumpeiYokoi> e.g. /media/BEAGLEBONE from the SD card shows up, as does its rootfs; my USB flash drive shows up as /media/its_proper_tag_name. It's all lovely.
Nov 19 13:37:43 <zmatt> probably some daemon doing that, udisks or something
Nov 19 13:38:09 <zmatt> it won't be done for you on a bare console image
Nov 19 13:38:21 <GrumpeiYokoi> Really? That's interesting.
Nov 19 13:38:45 <zmatt> definitely not merely a udev rule
Nov 19 13:38:58 <GrumpeiYokoi> Is udisks a daemon that is only installed for systems that use X?
Nov 19 13:39:35 <GrumpeiYokoi> I've written one udev file and it was to allow our company's USB device to be accessible without root access
Nov 19 13:40:16 <GrumpeiYokoi> And I felt like I was taking a lot of leaps of faith in writing it. The documentation seemed vague to me at the time. Maybe I know more now...
Nov 19 13:40:36 <zmatt> it's installed as part of some desktop environments... my laptop's spartan installation doesn't have it even though I have X
Nov 19 13:40:51 <GrumpeiYokoi> And it would auto allocate 'files' under /dev/ but not /media, certaintly
Nov 19 13:41:15 <zmatt> indeed udev is responsible for /dev
Nov 19 13:43:00 <GrumpeiYokoi> Yeah certainly that makes perfect sense. Earlier today in my mind I was conflating the automatic naming of files in /dev with the automatic naming and mounting of devices in /media. I kind of realized this error as I was talking to you. Sorry about that :x
Nov 19 13:44:13 <zmatt> it is raining...
Nov 19 13:44:16 <zmatt> I do not approve
Nov 19 13:45:04 <GrumpeiYokoi> I'll trade you :3
Nov 19 13:45:18 <GrumpeiYokoi> It's all warm and lightly sunny here
Nov 19 13:45:54 <GrumpeiYokoi> In spite of being nearly December in the northern hemisphere
Nov 19 13:49:42 <zmatt> did you know braids are orderable?
Nov 19 13:49:48 <zmatt> https://perso.univ-rennes1.fr/bertold.wiest/bouquin.pdf
Nov 19 13:52:56 <GrumpeiYokoi> I did not
Nov 19 13:52:58 <av500> zmatt: I ordered some from China
Nov 19 13:54:03 <GrumpeiYokoi> On a side note, I was reading yesterday that a professor in Nigeria is claiming to have cracked the Riemann Zeta Hypothesis
Nov 19 13:55:16 <GrumpeiYokoi> He said it took 7 years, but I wonder if he is just trying to copy Andrew Wiles because that's how long he said it took him to do Fermat's Last.
Nov 19 13:56:00 <zmatt> it's just called the "Riemann Hypothesis" ... and I'll wait for the reviews of his publication :P
Nov 19 13:56:33 <GrumpeiYokoi> Now you're being pedantic :)
Nov 19 13:56:33 <zmatt> don't forget Wiles also screwed up the first attempt and needed whole different tooling to repair it
Nov 19 13:56:41 <GrumpeiYokoi> It's for the Zeta function or whatever
Nov 19 13:57:01 <GrumpeiYokoi> It's been a long time since I looked at it. Took a crack at it myself but I was a noob.
Nov 19 13:57:05 <zmatt> hehe
Nov 19 13:57:33 <zmatt> like Fermat's Last, you can be reasonably sure the proof won't fit in the margin here either
Nov 19 13:58:21 <GrumpeiYokoi> I am kneejerk suspicious of any mathematical proof that is like 100 pages
Nov 19 13:58:49 <GrumpeiYokoi> But I guess that might be the reality for some of these 'simple' conjectures
Nov 19 13:59:56 <zmatt> which is why the Clay Mathematics Institute requires the proof to withstand a year of scrutiny
Nov 19 14:00:18 <GrumpeiYokoi> I have definitely looked at some of the proofs on blog for fermat's last theorem and I have seen the guy on there use some very roundabout methods to arrive at conclusions that would have taken 20 less steps, for example
Nov 19 14:00:27 <GrumpeiYokoi> *on a blog for ...
Nov 19 14:00:44 <zmatt> I know the proof has been simplified since the original version
Nov 19 14:00:56 <GrumpeiYokoi> The article I was reading said two years of scrutiny in a major journal, iirc
Nov 19 14:01:01 <zmatt> it's easier once you know the route to the solution to start optimizing that route
Nov 19 14:01:15 <zmatt> could be, something like that in any case
Nov 19 14:01:47 <zmatt> if you don't yet quite know the tools you need, it makes sense to construct overpowered tools
Nov 19 14:02:09 <GrumpeiYokoi> Yeah, certainly. I had that happen to me recently even.
Nov 19 14:02:49 <zmatt> Perelman also didn't prove Poincaré conjecture... he proved something vastly more general which just had the Poincaré conjecture as corollary
Nov 19 14:03:13 <GrumpeiYokoi> Right on Perelman
Nov 19 14:03:18 <GrumpeiYokoi> good game
Nov 19 14:03:46 <GrumpeiYokoi> I didn't realize until recently that there's this ABC conjecture that sits above Fermat's Last Theorem which is still unproven
Nov 19 14:04:05 <GrumpeiYokoi> These damn prime numbers are the root of all mathematical suffering, it seems
Nov 19 14:04:31 * GrumpeiYokoi writes to congress to ban all prime numbers
Nov 19 14:04:49 <zmatt> we need 'em for crypto
Nov 19 14:05:04 <av500> all they do is to help terrorists
Nov 19 14:05:07 <GrumpeiYokoi> If you're not guilty of anything, you don't need crypto
Nov 19 14:05:23 <GrumpeiYokoi> We'll have no secrets, the world will be filled with peace and harmony :3
Nov 19 14:05:27 <av500> if you use it, you are guilty of using it
Nov 19 14:06:18 <GrumpeiYokoi> I live in a state that at one time rounded pi to 3.2 by law. I think we can make a prime number ban happen.
Nov 19 14:06:51 <zmatt> it would also adversely impact the maintainability of INTERCAL code, since the Google INTERCAL Style Guide advises:
Nov 19 14:06:55 <zmatt> "In addition to the above rules, you should always use composite numbers when writing new code. Prime numbers are reserved for use by other parts of the code to fall back on when they run out of numbers in their designated block."
Nov 19 14:09:10 <philenotfound> while you are at it, also ban NULL, only slows down economy growth by crashing apps pointing to it.
Nov 19 14:10:07 <philenotfound> and whitespace, bringing down enterprise-level business logic shell scripts
Nov 19 14:10:16 <zmatt> you can turn them into NullPointerExceptions though, now also on ARM thanks to this little snippet of code I wrote -> https://github.com/mvduin/arm-signal-unwind
Nov 19 14:14:30 <philenotfound> zmatt: that's just lipstick on a pig
Nov 19 14:17:11 <zmatt> actually I needed it to catch bus errors when probing peripherals
Nov 19 15:08:27 <GrumpeiYokoi> zmatt: You were right, it was the udisks service / daemon
Nov 19 15:09:13 <GrumpeiYokoi> So I owe you a coke
Nov 19 15:25:45 <zmatt> lol
Nov 19 16:12:42 <niro> Anyone have any experience with javame?
Nov 19 23:03:24 <jkridner> niro: why bother with javame?
Nov 19 23:03:30 <jkridner> why not use the full thing?
Nov 19 23:09:38 <jkridner> ikmaak: ask me, but ask me here!
Nov 19 23:09:44 <jkridner> that way it goes in the logs, etc.
Nov 19 23:09:56 <jkridner> Seeed BeagleBone Green support went mainline for both kernel and u-boot.
Nov 19 23:09:59 <jkridner> u-boot: http://git.denx.de/cgi-bin/gitweb.cgi?p=u-boot.git;a=commit;h=dfd1bb4ec89e8b1e87f0605af2345ad6dcf777da
Nov 19 23:10:38 <ikmaak> thank you
Nov 19 23:11:01 <jkridner> that includes a link to the patch in the kernel mainline: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=79a4e64c679d8a0b1037da174e4aea578c80c4e6
Nov 19 23:36:43 <ds2> project "moldy"? :D
Nov 19 23:37:03 <GenTooMan> maybe it's in the spore stage?
Nov 20 00:40:15 <veremit> it'll .grow. on ya ...
Nov 20 00:43:43 <zmatt> badum, tsss!
**** ENDING LOGGING AT Fri Nov 20 02:59:58 2015