**** BEGIN LOGGING AT Tue Jun 07 02:59:58 2016 Jun 07 07:13:45 hi! Jun 07 07:13:59 i am sourish from kolkata Jun 07 07:14:07 i am very new in linux Jun 07 07:14:38 i have written a program in python Jun 07 07:15:09 but i am unable to load this program by default in beagle bone black Jun 07 07:15:16 please help Jun 07 07:16:19 hi Jun 07 09:22:04 hello, kernel 4.6.0 doesn't have cape manager ? Jun 07 09:22:44 i need to enable spi0 and i don;t have any clue how to do that without capemanager Jun 07 09:30:41 anyone ? Jun 07 09:30:55 hello, kernel 4.6.0 doesn't have cape manager ? Jun 07 09:48:17 hello, kernel 4.6.0 doesn't have cape manager ? Jun 07 09:49:13 i don;t have a /sys/devices/bone_capemgr.* folder Jun 07 09:59:10 then use one that has capemanager Jun 07 10:11:07 ii need 4.6.x kernel Jun 07 10:11:49 this cap emanager can pe isntalled ? or is there another way to enable spi0 and get /dev/spidev0 ? Jun 07 10:21:26 I improved the readme in https://gitlab.com/nivwiz/cwvw , can anyone please comment, if it clear? Jun 07 10:21:48 this allows to setup a wireless connection without a keyboard and screen Jun 07 10:41:17 i need 4.6.x kernel. Jun 07 10:41:26 this cap emanager can pe isntalled ? or is there another way to enable spi0 and get /dev/spidev0 ? Jun 07 11:48:19 costin: device-tree Jun 07 11:48:34 cape_mgr is only a hacky wrapper Jun 07 12:46:31 in the bbgw what is used to run the softap? Jun 07 12:46:48 I mean , what software? Jun 07 13:00:42 Good morning Jun 07 13:01:27 Você quis dizer: alguém pode me ajudar Can someone help me Jun 07 13:02:57 Hi Jun 07 13:06:22 I'm trying to install debian 7 in beaglebone green and an error saying that the unit [dev / mmcblk1] does not exist Jun 07 13:10:39 download image, write image to uSD card . . boot .. voila. Jun 07 13:47:41 Hello, I would like to know if, using the ASUS USB-N13 802.11 b/g/n device, is it possible to create a peer to peer connection between two BeagleBoards with a date rate of 300Mbit/s. Jun 07 14:09:45 FNT: theoretically you might get a link speed of 300Mbit but since the USB port is only 480Mbit half-duplex you're gonna max that out pretty quickly. Jun 07 14:15:00 veremit: My communication is half duplex. One device is constantly transmitting information to the other one. So, can I reach the maximum speed in this case? Jun 07 14:15:23 ymmv, but you can certainly try Jun 07 14:20:23 thank you. What about the power consumption? On the wiki page I read that the current required for an USB dongle is about 1A, while in the faq page the recommended current is 2A Jun 07 14:21:45 depending on your peripheral use, 2A is recommended .. the board itself alone will function at ~500mA from USB, but its not by any means guaranteed, and it will de-clock if needed. Jun 07 14:21:52 FNT: to-spec USB devices shouldn't draw more than 500mA. That has never stopped them in practice from trying to draw 2+A. This change has been driven by micro-USB charging, and other equipment has followed Jun 07 14:22:59 It's also happy-fun-time if you have some USB equipment on external power that suddenly _loses_ it's power (say, due to a fuse going) and then attempting to pull all it's power via USB, causing USB overcurrent shutdown situations... Happy happy fun time to debug remotely. ;) Jun 07 14:23:01 Spidler_: that is an average value .. so short term peaks are tolerated. That's why the spec. does indeed state you should have decent bulk capacitance on each USB port, which alas the BBB lacks Jun 07 14:23:38 veremit: Yep. But devices on USB has become lazier and expect more power. Jun 07 14:23:39 now who was it here that build the BBB-usb hub .. they made quite a smart job of that. Jun 07 14:23:57 Spidler_: with usb3 ports, its less of aproblem .. but your statements are true. Jun 07 14:24:01 I've got a few USB modems that won't even boot if they can "only" get 500mA Jun 07 14:24:14 doesn't stop people abusing them.. and blowing the odd port up Jun 07 14:24:20 Might be cause they run full on-device mips based Linux systems. Jun 07 14:24:35 As I clearly stated .. "ymmv" :D Jun 07 14:25:06 If you want a funky thing to play with, the ZTE MF823 modems have 128 megs of read/Write flash, 32 megs of RAM, and runs telnet with a blank root password Jun 07 14:25:18 Don't try to use them as modems though, they do a despicable job of that. Jun 07 14:25:40 Their on-device DNS server will interrupt and rewrite DNS requests if the device gets an SMS or similar. Jun 07 14:25:50 But! it's a funky little hacky platform in a USB format :) Jun 07 15:07:48 Thank you for your help! Jun 07 15:11:17 how is responsible for the bbgw softap ? Jun 07 15:11:45 I see iw_parse , but I do not understand how the welcome page is served? Jun 07 15:13:46 citylight2: is there a different image/etc .. they're not using something derived from RN's image(s) ? Jun 07 15:14:23 which image do they use? in the rcn image I do see iw_parse Jun 07 15:15:02 citylight2: no idea :D Jun 07 15:19:43 is rcn-ee my only hope in this case? Jun 07 15:21:21 citylight2: you're the one doing all the funky wifi stuff, right? Jun 07 15:21:29 alas rcn-ee is very much MIA :/ Jun 07 15:21:56 Serious. Hope he's okay. Jun 07 15:22:18 I think he's still coding like a madman .. just doesn't have time for IRC any more :/ Jun 07 15:22:39 Cool. Long as he's okay. Jun 07 15:23:14 rcn-ee puts out more code than anyone i've ever seen Jun 07 15:23:43 like he could be two or three people for all we know Jun 07 15:24:57 lol Jun 07 16:14:31 veremit: "and it will de-clock if needed" -> no. Jun 07 16:15:32 it doesn't? what happens? it just .. draws current until it reboots? Jun 07 16:15:46 veremit: it does not have any kind of early warning that power consumption is too high for the supply, the pmic will just abruptly cut power once vsys drops below the trip point Jun 07 16:15:57 it doesn't reboot by itself Jun 07 16:17:34 TERRITORY THEN Jun 07 16:17:39 BROWN-OUT* Jun 07 16:17:45 sheesh wtf is *up* with my PC Jun 07 16:18:52 I can see how "territory" is an obvious typo for "brown-out" Jun 07 16:19:26 that was actually typed before I got to 'territory' and there should have been a 'serious' too Jun 07 16:19:36 but I seem to have major processor bottleneck issues currently .. Jun 07 16:19:56 life is hard Jun 07 16:20:23 enough to badly throttle my pitiful typing rate .. Jun 07 16:21:57 +even my Jun 07 16:36:26 hello! quick question, i need to compile a kernel module on beagleboard xm rev c. i have debian 8.3 installed, a premade image from beagleboard.org, but when i try to install the linux headers, i cant find the ones suitable for me. the output from uname -r is 4.4.1-armv7-x5, while the repo only has headers for 4.4.1-armv7-x7, is it safe to install those? Jun 07 16:37:13 generally, yes. Jun 07 16:37:36 you might be able to also update to the matching kernel... Jun 07 16:37:43 I could see if I could get rcn-ee to put the 4.4.1-armv7-x5 version into the package feeds... or you could simply update your kernel. Jun 07 16:38:31 jkridner: is rcn-ee still active? not seen on IRC for .. ages .. eons .. Jun 07 16:39:01 veremit: he's not very active on IRC, but very responsive to e-mail. We've had him pretty loaded lately, but he's still been responsive. Jun 07 16:39:03 jkridner: also... any news on the BBblue yet!? Jun 07 16:39:27 ah I figured he was probably uber short on time Jun 07 16:40:00 I published the current design and have been trying to pull in resources to accelerate it. It has been bogged down by a number of things, but the pressure to get it out is starting to break the logjam. Jun 07 16:40:37 the linux image 4.4.1-armv7-x7 is not in the repo Jun 07 16:41:07 jkridner: I can almost imagine :) x15 gettin closer too, I trust? Jun 07 16:41:37 * veremit goes looking for bbblue designs .. Jun 07 16:41:46 should have some news on that front very soon. looks like we'll start shipping a few non-FCC boards with all the delays in the FCC fixes. Jun 07 16:42:01 http://github.com/jadonk/beaglebone-blue Jun 07 16:42:25 * Ragnorok sense conflict on BBB (grin) Jun 07 16:43:39 yes, major acronym conflict :D Jun 07 16:44:13 jkridner: ty Jun 07 16:44:55 * jkridner never supported BBB, but was happy when someone donated http://bbb.io Jun 07 16:59:17 heh Jun 07 17:35:36 Maybe should have gone with Beaglebone Azure or something Jun 07 17:36:54 beaglebone cloud? Jun 07 17:37:22 Yeah I guess don't want to get sued by Microsoft Jun 07 17:37:28 But Azure is a shade of blue too Jun 07 17:37:50 I vote for the BeagleBone Teal. Jun 07 17:38:25 And thus the next great internet holy war began... Jun 07 17:39:13 fwiw I think what is labeled as teal in computer systems is actually a rather bland color :/ Jun 07 17:39:41 I avoid it when choosing syntax highlighting schemes Jun 07 17:39:44 It can be a bit washed out, but that's a saturation thing. (shrug) Either way, it's not likely to happen. Jun 07 17:40:50 incidentally I gave one of my BBB's the hostname azulbian, but that's b/c it's in a blue housing Jun 07 17:40:55 (and running Debian) Jun 07 17:41:18 azul = blue in spanish Jun 07 17:42:12 There is no BeagleBone, only Azul. Jun 07 17:42:36 haha, 10 points for Ragnorok Jun 07 18:55:09 Cannot build kernel object because the kernel-devel package for kernel version 3.8.13-bone47 is not installed. Jun 07 18:55:43 I need to find a packege Jun 07 18:56:00 But, where? Jun 07 19:00:10 Is that on Debian? Jun 07 19:02:30 Yes, in Debian!!! Jun 07 19:02:45 Yes, on Debian!!! Jun 07 19:04:05 wow thats Old. Jun 07 19:07:03 Well, assuming you're compiling on the beaglebone itself, try sudo apt-get install linux-headers-3.8.13-bone47 Jun 07 19:07:09 So, Need I use other Image? Jun 07 19:07:39 If you got the BBB from element14 and are using the stock image on the EMMC I would strongly recommend flashing a recent image to it instead Jun 07 19:08:00 I ran into so many headaches trying to make their frankenstein image work with what everyone else knows Jun 07 19:08:07 Look Jun 07 19:08:09 E: Unable to locate package linux-headers-3.8.13-bone47 E: Couldn't find any package by regex 'linux-headers-3.8.13-bone47' Jun 07 19:08:14 Consule: What MathOnNapkins said. Jun 07 19:08:30 The default image's repository listings are all messed up Jun 07 19:08:46 And upgrading the kernel on the bone47 image from element14 is borked too Jun 07 19:09:02 just get a recent one, put it on an SD card, and flash it to the eMMC Jun 07 19:10:03 Can be id bone-debian-8.4-lxqt-4gb-armhf-2016-05-13-4gb? Jun 07 19:10:16 Can be it bone-debian-8.4-lxqt-4gb-armhf-2016-05-13-4gb? Jun 07 19:10:32 sounds like a good bet Jun 07 19:11:28 Once that's flashed you can update the kernel with apt-get. Jun 07 19:11:31 I have been trying to use BB with a USB Digital persona Jun 07 19:12:18 (And install the appropriate linux-headers-* package too) Jun 07 19:13:24 Tks! Jun 07 19:14:15 Yeah good luck :0 Jun 07 20:00:37 there should be an apt-cache search linux-image* such command .. but I'm less familiar with debian's PMS Jun 07 20:01:20 tbh though, it its *that8 old .. its easier to download & flash a new image than attempt an apt-get upgrade :D Jun 07 20:02:43 Well I forgot to tell him to apt-get update, but he still would had to add the rcn repos url manually, iirc Jun 07 20:03:05 indeed Jun 07 21:01:15 hi over there Jun 07 21:02:21 I'm interesting in launch date and availability of beagle board x15 Jun 07 21:07:44 Hello, Jun 07 21:08:42 I am trying to run BBB with osx el capitan. I have read the notes that standard HoRNDIS drivers don't work with this os Jun 07 21:08:57 so I'm using the rel8-pre2-dbg driver for HoRNDIS Jun 07 21:09:05 but still, to no avail. I have it plugged into the USB Jun 07 21:09:27 I also installed the energie ftdid driver Jun 07 21:09:53 but still nothing. Is there anything I could be missing? I am able to ssh into "beaglebone.local" but not to the assigned IP Jun 07 21:10:27 could it have something to do with bonjour's mDNS bindings? Any help would be greatly appreciated. Thanks! Jun 07 21:21:43 alexhiam: saw your nick and reminded me to ask about PWM. I'm working on Linux 4.1+ PWM for Adafruit_BBIO. https://github.com/adafruit/adafruit-beaglebone-io-python/issues/103 Jun 07 21:22:57 were you thinking of something similar to this for PyBBIO? https://github.com/adafruit/adafruit-beaglebone-io-python/issues/103#issuecomment-223801820 Jun 07 21:24:21 pwm_test driver seemed to simply the process of getting the /sys entries created automatically for a given PWM pin in Linux 3.8 but that doesn't seem to apply anymore in 4.1+ Jun 07 21:25:43 pdp7: whoa, I still haven't finished 4.x PyBBIO support ( :( ) and haven't even got to PWM yet. I didn't realize it was so different! Jun 07 21:26:04 yeah, been trying to pick it apart and figure out the best approach Jun 07 21:26:15 pwm_test really did a lot to help in 3.8 Jun 07 21:27:00 as far as I can tell, the best method in 4.1+ is to figure out the pwmchip for a given PWM pin Jun 07 21:27:14 then echo to the export file Jun 07 21:27:30 for example: Jun 07 21:27:33 echo 0 > /sys/devices/platform/ocp/48302000.epwmss/48302200.ehrpwm/pwm/pwmchip2/export Jun 07 21:27:44 then you'll get Jun 07 21:27:47 /sys/devices/platform/ocp/48302000.epwmss/48302200.ehrpwm/pwm/pwmchip2/pwm0/ Jun 07 21:28:05 with the usual files: enable, period and duty_cycle Jun 07 21:29:12 anyways, just thought i'd check if you or others in the channel had a simpler method Jun 07 21:29:20 pdp7: huh, so much more like gpio. Jun 07 21:29:26 yeah Jun 07 21:29:36 I'm into that method, but of course a bummer that it's a big change :( Jun 07 21:35:54 btw, this PR added ADC support for both 3.8 and 4.1+: https://github.com/adafruit/adafruit-beaglebone-io-python/pull/96 Jun 07 21:39:05 pdp7: nice! Jun 07 21:39:10 * alexhiam is way behind Jun 07 21:39:32 have you gotten users submitting PRs? Jun 07 21:39:49 I'm starting to go through all the issues and PRs for the Adafruit lib Jun 07 21:40:40 I'm hoping to borrow as much logic as I can from jkridner's bonescript 0.5.x Jun 07 21:41:38 pdp7: I don't get many PRs, I think PyBBIO is a bit overshadowed by Adafruit_BBIO ;) Jun 07 21:41:56 ah, personally, I think your semantics are better. Jun 07 21:42:21 but I guesss the RPi.GPIO compatibilty is why Adafruit_BBIO is used for Adafruit tutorials Jun 07 21:43:21 yeah, that's a huge plus Jun 07 21:43:29 I actually think GPIO Zero from Ben Nutall looks pretty nice https://www.raspberrypi.org/blog/gpio-zero-a-friendly-python-api-for-physical-computing/ Jun 07 21:43:55 its unforunate so many people did projects with RPi.GPIO as it was never meant to be used by the world Jun 07 21:58:17 ah, here's the link I was thinking of http://bennuttall.com/gpio-zero-developing-a-new-friendly-python-api-for-physical-computing/ Jun 07 21:59:10 as you said at #BringAHack, it would be nice to align PyBBIO and Adafruit_BBIO so that we don't have to duplicate efforts Jun 07 22:00:21 pdp7: yeah, that does look nice. I originally designed the PyBBIO API to be as resource light as possible to try and keep it fast, but userspace IO control is never going to be good for production anyway, so why not abstract more and make the API super friendly Jun 07 22:01:11 userspace IO vs. kernel driver? Jun 07 22:01:16 yeah Jun 07 22:01:19 ah, ok Jun 07 22:01:36 the userspace stuff is all for playing and rapid prototyping really Jun 07 22:01:54 that makes sense Jun 07 22:02:09 and learning of course, and the API makes a big difference there Jun 07 22:03:53 pdp7: I'm so down to align the libraries. Not sure what that would look like though Jun 07 22:08:32 yeah... i wonder if it is possible to maintain both sets of semantics Jun 07 22:08:51 i think for Adafruit_BBIO, compat with existing tutorials is importnat Jun 07 22:10:11 pdp7: yeah, for sure. Might just be a matter of some fancy import stuff and a few wrapper functions... Jun 07 22:11:29 ah ok, that's encouraging Jun 07 22:12:03 so we need a common abstration then? Jun 07 22:13:32 though I guess right off the bat there's incompatibility because the packages have different names :/ Jun 07 22:14:01 rigt Jun 07 22:14:04 *right Jun 07 22:14:27 jkridner: any thoughts about beaglebone python library unification? Jun 07 22:18:05 pdp7: yeah, it could be setup where GPIO.input() just calls pinMode(), or vice versa Jun 07 22:18:22 ah, ok Jun 07 22:18:48 I'm sure there'd be some messy corner cases, but I think it's certainly doable Jun 07 22:19:53 great... i'll look more closely once I've got something working for PWM in 4.1+ Jun 07 22:21:49 pdp7: you have gpio working on 4? Jun 07 22:22:20 hmm, i thought so but I need to check that Jun 07 22:23:35 I got that working in PyBBIO at least, though haven't pushed yet Jun 07 22:24:02 ah ok Jun 07 22:24:13 yeah, i 'd be interested to take a look Jun 07 22:24:53 I'll push it up to a new branch once I have a chance to dig up the SD card it's on... Jun 07 22:26:30 pdp7: I gotta run to dinner. Should we start an email thread about this? Jun 07 23:09:07 alexhiam: yeah, that is good idea Jun 08 00:06:11 alexhiam: "userspace IO control is never going to be good for production anyway" ... I beg to differ Jun 08 00:06:53 but then, my idea of userspace IO is not wading through the thick layers of kernel abstraction to begin with :P Jun 08 01:32:04 helo **** ENDING LOGGING AT Wed Jun 08 02:59:59 2016