**** BEGIN LOGGING AT Tue Dec 31 02:59:58 2013 Dec 31 05:04:31 i am not able to ssh my BBB Dec 31 05:08:56 Where can I buy a BBB? Seems to be on back order everywhere. Dec 31 05:09:41 <_av500_> indeed Dec 31 05:09:43 <_av500_> order Dec 31 05:09:44 <_av500_> wait Dec 31 05:09:45 <_av500_> receive Dec 31 05:10:12 <_av500_> back ordered mean exactly that Dec 31 05:13:41 Nobody will even accept an order. Dec 31 05:13:43 or, buy a refurb: http://boardzoo.com/index.php/clearance/beaglebone-black.html#.UsJSe2RDvmY Dec 31 05:14:00 Jameco accepted my order, then cancelled after a couple weeks. Dec 31 05:14:12 Didn't know about refurm, thanks! Dec 31 05:14:44 I just put a fresh copy of Anstrom on my Black's eMMC, booting it from wall wart and connected to ethernet cable. It doesn't seem to have any ports open. I've been stuck in a loop for the last day or so. Any hints a t which way to go to troubleshoot? I don't have a console to usb thingy, so I can't see that. I could plug the screen in tho, i guess. Dec 31 05:18:52 Thanks, emeb_mac! Dec 31 05:31:00 np Dec 31 05:43:41 Did the web-based ssh client go away with the latest Angstrom release or did I muck something up? Dec 31 05:48:48 should be still there Dec 31 07:42:21 I am getting a white screen when connecting 4D system LCD with BBB Dec 31 07:42:32 Any one felt before Dec 31 10:50:21 anybody knows if you enable UART1 the UART1RTS it's enabled by default? Dec 31 11:10:32 * N2TOH has about 18 hours to go Dec 31 12:16:07 salaet: not sure (i'm never sure these days) but I never had to disable rts, i think should be disbaled by default Dec 31 12:17:28 check this Dec 31 12:17:34 is my beagbleboneblack Dec 31 12:17:34 http://pastebin.com/gjNBvq61 Dec 31 12:18:18 uart rts semms to be nothing Dec 31 12:18:21 uart1 Dec 31 12:19:15 I need to use rts for max485 enable/disble Dec 31 12:26:59 ok, did you try with minicom? Dec 31 12:27:13 minicom has easy settings for enabling hardware/software flow control Dec 31 12:27:22 if you enable hardware one, that should help Dec 31 12:27:28 i guess Dec 31 12:27:36 yes Dec 31 12:28:31 I configured minicom with /dev/ttyO1 with hardware flow control Dec 31 12:30:05 still didn't work ? :( Dec 31 12:30:12 I'm mesuring with oscilloscope rts signal Dec 31 12:30:17 always 3.3v Dec 31 12:30:25 aprox Dec 31 12:30:41 3v Dec 31 12:30:49 always Dec 31 12:30:52 never changes Dec 31 12:31:21 an tx transmit when I hit the keyboard Dec 31 12:31:27 sorry for my bad english Dec 31 12:31:51 your english is fine (mine is not good anyway) :) Dec 31 12:34:36 ok but, again, not sure but, rts/cts are used on demand right? Dec 31 12:35:26 i mean, when fifo of one end gets full, and it wants data flow to stop until it deals with the buffer i guess Dec 31 12:36:08 if you want to see it works, you need something else rather than just transmitting some data Dec 31 12:36:27 i found this; might not be that handy though; http://man7.org/linux/man-pages/man4/tty_ioctl.4.html Dec 31 12:39:08 also see this; http://unix.stackexchange.com/questions/75983/uart0-access-in-am1808-using-embedded-linux Dec 31 12:47:26 sorry I go to read Dec 31 13:38:47 greetings Dec 31 13:39:28 * Tenkawa has his openbsd bbb running -current now successfully.. Dec 31 13:39:54 that was an extensive recompile Dec 31 13:43:13 but it worked yay Dec 31 13:43:14 heeey Dec 31 13:43:18 congrats Dec 31 13:44:09 currently its building xenocars so i can actually install ports Dec 31 13:44:19 musufu: thanks Dec 31 13:44:57 er xenocara Dec 31 13:45:04 darn osd Dec 31 13:45:27 rather osk Dec 31 13:50:43 are you gonna share builds / sourcetrees / patches anything? Dec 31 13:52:01 i'm just using stock ports atm. Dec 31 13:53:13 if i need to make changes for them to work i'll try to update the maintainer of each one Dec 31 13:57:58 hmm, how did it work on beaglebone without any changes? did somebody already made a port for it? Dec 31 13:58:13 yes Dec 31 13:58:24 rmv7 Dec 31 13:58:26 er Dec 31 13:58:28 armv7 Dec 31 13:59:11 its an official branch i think Dec 31 13:59:13 including omap335x soc support too i guess then Dec 31 15:25:59 musufu: confirmed, BBB Uart1 rts dons't work Dec 31 15:26:50 oscilloscope connected directly to the pin Dec 31 15:27:03 :[ strange, thinking that also might be related with pinmux, Dec 31 15:27:08 let me check Dec 31 15:27:12 which pin is it? Dec 31 15:27:27 P9_19 Dec 31 15:30:00 hmm looks like that pin is rtsn in default Dec 31 15:30:08 but Dec 31 15:30:14 says uart1_rtsn Dec 31 15:30:29 probably meaning inverted maybe? the 'n' at the end Dec 31 15:31:43 yes Dec 31 15:31:57 sorry what does it mean rtsn? Dec 31 15:32:00 null? Dec 31 15:32:04 nope Dec 31 15:32:05 inverted Dec 31 15:32:08 ok Dec 31 15:32:23 means; it's gonna be HIGH (3.3) when you expect a binary 0 Dec 31 15:32:37 and it's gonna be LOW (0v) when you expect a binary 1 Dec 31 15:32:42 yes its Dec 31 15:33:14 ok Dec 31 15:33:43 it's possible to invert without hardware Dec 31 15:33:51 ? Dec 31 15:34:40 i'm using max485 Dec 31 15:34:47 1st, I'm not sure if it's really inverted, just guessing from the 'n' in the name Dec 31 15:34:49 for RE/DE Dec 31 15:35:23 btw, I found someone else trying to use beaglebone for rs485 Dec 31 15:35:26 see this link; https://groups.google.com/forum/#!topic/beagleboard/-qnUNuiY_AI Dec 31 15:36:32 thansk musufu: but I'know this post but its using rs485 cape Dec 31 15:36:59 hmm :[ Dec 31 15:37:25 keep forgetting i have this irc client Dec 31 15:37:36 ah, see this link, http://support.criticallink.com/redmine/projects/armc8-platforms/wiki/Driving_RS-485_transceivers Dec 31 15:37:54 it just disables rts :D Dec 31 15:37:56 whats up, Dec 31 15:37:58 er ? Dec 31 15:38:17 and manually enables TXE pin with a gpio Dec 31 15:38:34 musufu : o_O Dec 31 15:38:38 :)รง Dec 31 15:39:45 you can even make txe pin always high, probably no need to use a gpio for it Dec 31 15:40:09 what was your 485 IC? Dec 31 15:40:14 no replies from the remote devices? Dec 31 15:41:16 zmatt: not sure about rs485 but, rx and tx lines are separate, and both of them are active always like this, isn't it? Dec 31 15:41:53 A and B in that diagram are a differential pair, not tx/rx lines Dec 31 15:41:58 aaa Dec 31 15:41:58 485 is half-duplex Dec 31 15:42:14 and you need to enable/disable TXE pin to control it? Dec 31 15:42:23 normally you'd tie TXE and nRXE together to switch direction Dec 31 15:42:39 with nRXE tied to ground like this, you'll receive an echo of your own signal Dec 31 15:42:52 yes Dec 31 15:43:05 ah didn't know and went a bit overboard :D sorry Dec 31 15:44:17 ah, you mean connect RXE and TXE to the same gpio? (since RXE is inverted), and control it like that? :) that makes sense Dec 31 15:44:28 yes Dec 31 15:44:52 ah, ok thanks :) enlightened a bit more Dec 31 15:44:54 that's the normal use, hence they conveniently inverted one of the signals Dec 31 15:46:59 though half-duplex, 485 is still nice though... with proper quality cable it reached like >1000 meters Dec 31 15:47:06 try that with rs-232 ;) Dec 31 15:48:18 (don't forget the bus terminator with such lengths though) Dec 31 15:51:10 eps Dec 31 15:51:16 sorry I read now Dec 31 15:51:21 I'm triying :) Dec 31 15:52:01 good luck :) Dec 31 15:52:44 zmatt, more like 1000 < meters with RS485 Dec 31 15:53:17 DMX (which uses it) has official max cable length of 1200m with a bus load of 32 devices Dec 31 15:53:51 for other protocols it will depend on the bitrate though Dec 31 15:55:16 zmatt:Thanks for your notes Dec 31 15:55:38 yes the bit rate has alot to do with the limitation Dec 31 15:55:42 zmatt : yes you are correct but it's my first time with beaglebone black Dec 31 15:55:47 with rs485 Dec 31 15:55:51 of course even a single shitty device can ruin things... we've seen DMX fixtures which simply connected (-) to its GND and (+) to a TTL input Dec 31 15:56:03 bye bye bus Dec 31 15:56:24 (and with a bit of luck, hello ground-loop) Dec 31 15:56:50 :D Dec 31 15:56:52 brilliant Dec 31 15:57:02 improvising technician Dec 31 15:57:43 "it worked fine when we tried it" Dec 31 15:59:35 zmatt: with BBB? Dec 31 16:00:33 we actually had to add a special mode to our dmx controller to enable some enourmous inter-byte gap because some cheapass fixtures couldn't cope with it otherwise -.- Dec 31 16:00:40 salaet: no Dec 31 16:01:02 salaet: not related :) Dec 31 16:01:14 other than though the use of rs-485 ;-) Dec 31 16:02:10 hehe Dec 31 16:04:47 improvising technicians are hazardous though... we had an RMA where the varistors protecting the DMX line from transient overvoltage were literally *gone*, nothing left but a scorch mark.... apparently someone though it was convenient to use the same XLR connectors for 230V Dec 31 16:05:11 so Dec 31 16:05:19 i asked yesterday, I'll ask again Dec 31 16:05:34 is it possible to use netflix with a BBB + android ? Dec 31 16:05:53 doesn't netflex use some DRM shizzle? Dec 31 16:05:56 basically I want to have a ton of expiramental fun but after it's all said and done that would be a nice final resting place for it Dec 31 16:06:12 *netflix Dec 31 16:06:22 yeah i assume. still not possible on linux afaik Dec 31 16:06:29 whatever possible in that topic, probably implemented in xmbc project i guess Dec 31 16:06:30 but android phones have a netflix app... Dec 31 16:07:03 android phones usually have magic stuff going on behind linux' back Dec 31 16:08:38 "TI OMAP4 First to be Awarded Netflix HD DRM Certification" Dec 31 16:08:48 I'm very sad with rts of BBB not runs with max485 Dec 31 16:08:48 :( Dec 31 16:08:53 so you're looking at HS devices there Dec 31 16:09:00 not GP devices like the beagle Dec 31 16:09:30 salaet: i will work, if you do these; Dec 31 16:09:38 connect rt & tx as usual Dec 31 16:10:12 select any gpio for controlling sending/receiving from 485, connect that pin to TXE and /RXE of your 485 IC Dec 31 16:10:20 btw, what is your 485 IC ? Dec 31 16:11:01 nezZario: effectively the CPU is partitioned into two VMs, the "public" VM runs linux, the "secure" VM contains stuff like netflix decryption, which then streams its content to peripherals whose firewalls are configured to only allow access from secure world Dec 31 16:11:43 annoying copyright/drm world :-/ Dec 31 16:12:45 on the bright side, the ROM seems to be full of bugs... I've browsed around a bit in the "secure ROM" of the dm814x and it's a mess Dec 31 16:12:57 so there may be workarounds ;) Dec 31 16:15:14 anyhow, need to do some shopping, afk Dec 31 16:15:15 well Dec 31 16:16:11 where can I get a good microSD ? Dec 31 16:16:54 I asked a lot of unrelated peoples this questions a while back... if there was a such thing as high quality microSD / USB storage .. Never got a good answer .. it amazes me most retailers and manufactors don't give read/write specs on flash storage Dec 31 16:23:47 wich is tha max voltage that accept the pins of BBB? Dec 31 16:23:47 I'm thinking to bridge with 3.3 v to 5v Dec 31 16:23:48 nezZario I dont have idea Dec 31 16:25:58 nope don't use 5v Dec 31 16:26:27 from beagle bone ----to---> 5v circuit is ok Dec 31 16:26:47 reverse Dec 31 16:26:56 from 5v circuit ---to---> beaglebone, use a level-shifter, or simple voltage divier Dec 31 16:27:19 for ex: if you have 2.2k and 3.3k resistors, you can easily make one Dec 31 16:27:54 what is the threshold for the 3v3 logic? Dec 31 16:28:46 salaet: this should help: http://i.stack.imgur.com/8GIWf.png Dec 31 16:29:14 musufu:max485 Dec 31 16:29:16 typical Dec 31 16:29:27 musufu:thanks Dec 31 16:29:41 i have a level converter Dec 31 16:29:56 ah thats nice Dec 31 16:29:58 just use that :D Dec 31 16:30:59 nezZario: there are high quality sd cards and usb storage Dec 31 16:31:09 nezZario: google for "industrial grade sd cards" Dec 31 16:31:12 people Dec 31 16:31:19 good year to all Dec 31 16:31:23 bye Dec 31 16:31:27 nezZario: their price is usually 10 times higher than consumer grade Dec 31 16:31:28 thanks musufu Dec 31 16:31:34 well that's not worth it ;) Dec 31 16:31:36 thank you too :) Dec 31 16:31:39 and zmatt Dec 31 16:31:47 bye Dec 31 16:31:53 KotH: any idea if it's worth it... do they have significantly more read/write cycles or anything? Dec 31 16:31:55 nezZario: well, but they have exact specs for their read write endurance :) Dec 31 16:32:05 nezZario: depends Dec 31 16:32:37 nezZario: there are MLC and SLC sd cards, the MLC have about the same endurance as consumer grades, while SLC have 5 to 10 times the endurance Dec 31 16:33:17 nezZario: but the most important feature is: you will be notified when something (ie anything) on the sd card changes Dec 31 16:33:50 nezZario: unlike consumer grade, where you can buy 3 of the same sd cards (same manufacturer and model) and get 4 different cards Dec 31 16:34:00 nezZario: and all behaving slightly different Dec 31 16:35:16 nezZario: you can also try swissbit, they have non-industrial grade sd cards with SLC Dec 31 16:35:25 nezZario: don't know about their prices though Dec 31 16:36:35 i'll check them out Dec 31 16:37:01 But basically, yeah.. 'consumer grade' 'SLC' is what I'd like.. if that isn't an oxymoron :) Dec 31 16:37:41 really appreciate it.. i was asking other communities (hardware chan?) a few months ago for unrealted reasons and never got an answer... thanks a ton! Dec 31 16:39:07 nezZario: i didnt say consumer grade :) Dec 31 16:39:14 nezZario: there is no such thing as consumer grade SLC Dec 31 16:39:21 nezZario: consumer grade always means MLC Dec 31 16:40:13 nezZario: and i can understand that you dont get an answer to this, it's something you only know if you have to deal with such stuff at work. hobbists never get access to the real data of sd cards Dec 31 16:41:18 nezZario: oh.. and btw: if you want to hack sd cards (as seen on hackaday) the swissbit cards contain a 32bit risc controller, not an 8051 derivative Dec 31 17:12:27 * SpeedEvil would be most amused if the answer to 'what is the best card' became 'shittycard 32g with open source firmware, and ubifs' Dec 31 17:18:57 SpeedEvil: lol Dec 31 17:19:07 SpeedEvil: there aint no such thing as oss firmware for sd cards Dec 31 17:19:34 In principle, with reverse-engineering, and flashing, I mean. Dec 31 17:19:40 SpeedEvil: most (all?) of these have a custom build controller. and the FTL is a close guarded secret Dec 31 17:19:58 KotH: You've not seen the recent things about hacking them then? Dec 31 17:20:16 i've seen it Dec 31 17:20:26 but that's just one of them Dec 31 17:20:40 and the controller might change in next revision Dec 31 17:20:46 Sure - which is why I mentioned a specific (though imaginary) brand. Dec 31 17:20:48 And of course. Dec 31 17:20:51 just because the competitor is 0.00001usd cheaper Dec 31 17:21:07 would make more sense to use a flash filesystem tuned to the properties of NAND flash, rather than emulate a behaviour that isn't native to flash Dec 31 17:21:36 oh.. and even if there is such a major change, unless there is a big change in specs (ie storage capacity) the marking will be exactly the same and you will not know what type you buy until you pry the card open Dec 31 17:21:49 (though an issue is apparently that a lot of techniques for dealing with MLC flash are patented -.- ) Dec 31 17:22:26 zmatt: emulating standard random access memory makes the most sense Dec 31 17:22:58 having flash emulate RAM makes even less sense than making it emulate a block device Dec 31 17:23:46 any emulation pays a price, and the bigger the mismatch between the interface being emulated and the native one, the bigger the cost Dec 31 17:25:00 KotH: A single byte write on a raw NAND can mean reading, erasing, and rewriting several hundred K. Dec 31 17:25:27 Write 50kb, and you've just destroyed a block of a few hundred K. Dec 31 17:25:40 SpeedEvil: nope Dec 31 17:25:59 SpeedEvil: nah, you'd of course use a translation layer that also does wear-levelling Dec 31 17:26:09 SpeedEvil: yes, you might need to start a new page, but you will not write a complete page until you have to Dec 31 17:26:23 SpeedEvil: FTLs are really sofisticated to prevent that scenario Dec 31 17:26:24 but doing that for byte-size writes will be definitely much more costly than for block-writes Dec 31 17:26:33 Sure - but RAM is even more hideous to emulate than a block device, because the blocks are 512-4096 times smaller Dec 31 17:26:54 SpeedEvil: an sd card is a 512byte addressed block device Dec 31 17:26:57 STT-MRAM should just hurry up Dec 31 17:27:11 and get bigger and cheaper :P Dec 31 17:29:15 it would be funny if, after a long detour, we'd end up back at really fancy core-memory Dec 31 17:34:01 Rod logic. :) Dec 31 17:34:30 * SpeedEvil wants 'A young gentlemans illustrated primer'. Dec 31 17:50:09 <_av500_> +1 Dec 31 17:54:36 <_av500_> Dec 31 18:05:40 hi everyone Dec 31 19:21:43 hi Dec 31 19:22:29 Hello Dec 31 19:57:55 openbsd-current is running pretty well on my bbb so far Dec 31 20:09:29 usb working? Dec 31 20:10:17 somazero: not yet Dec 31 20:10:43 trying to figure out if anyone is making any progress on thar Dec 31 20:10:46 er that Dec 31 20:11:03 Darn! Once USB is functional, there's a lot I want to try. Dec 31 20:11:27 i'd like to use my usb wifi Dec 31 20:12:16 Yeah, I don't want to do anything crazy...want to try to replace a pfsense atom box, for fun. will need to throw a usb nic on there, though. Dec 31 20:12:25 does freebsd work on the bbb yet? Dec 31 20:12:50 I saw some words about the original, but not the BBB. I admit, though, I haven't looked too hard. Dec 31 20:13:02 Spent way too much time with Gentoo...:) Dec 31 20:13:33 i'd convert my other bbb to openbsd if the usb support gets added (currently slackware) Dec 31 20:13:58 i have 4 rpi's running slack that i can leave running linux Dec 31 20:15:59 I've got 4xBBB and 2xRPIs. Lots to play with, although, the RPIs all run Openelec as XBMC frontends. I have a BBB running a small samba file server with some USB drives. Dec 31 20:16:09 ah Dec 31 20:16:57 somazero: working on compiling some ports now after that long upgrade to current Dec 31 20:18:08 Tenkawa, Yeah, my openbsd-fu skills aren't that's finely honed in order for me to attempt something like that. I'll need to brush up on my pf knowledge again. Dec 31 20:18:28 Tenkawa, as well as learn how to track current... Dec 31 20:18:48 Tenkawa, Although, it does seem as though openbsd is putting more into the BBB than freebsd Dec 31 20:19:58 I want to try the 3.12 image on one of my wheezy BBBs to see if there's a dramatic performance increase. Dec 31 20:20:24 ah Dec 31 20:23:09 somazero, use 3.13.. :) i just pushed 3.13 over 3.12 last night in my repo.. Dec 31 20:23:44 rcn-ee, Nice! I'll give it a go! Thanks again! Dec 31 20:26:09 so I' Dec 31 20:26:14 ve had a beagle bone for a while .. Dec 31 20:26:41 when I first got it, I spent a bunch of time playing with angstrom Dec 31 20:26:57 it felt like it needed to percolate .. so I let it Dec 31 20:27:03 is it time to drink yet? Dec 31 20:28:19 or is the debian stuff becoming more standard? Dec 31 20:28:42 rcn-ee: hey, congrats for 3.13 Dec 31 20:28:55 I also managed to get most of the stuff running Dec 31 20:28:59 including multipath-tcp Dec 31 20:29:06 Rickta59, I don't have a lot of experience with angstrom. I've been bouncing between debian and gentoo. Although, it appears debian is much better supported. You can, however, run the same kernels on gentoo. Dec 31 20:29:08 that's awesome Dec 31 20:29:15 but somehow disabled capemgr :[ Dec 31 20:29:42 that needs to be re-written. ;) Dec 31 20:29:44 so camemgr is built-in to the kernel, but not showing up at all, do you know any possible reasons for it? Dec 31 20:30:00 aa? is it something about 3.13 then? Dec 31 20:31:23 it just needs to be re-ported by someone who is more familar with that interface.. i got bored with it so i started my own this morning.. Dec 31 20:32:08 ah okay Dec 31 20:32:16 hdmi output is working somehow without it Dec 31 20:32:23 i'm not gonna use any capes at all anyway Dec 31 20:32:26 dumping it :D Dec 31 20:32:31 thanks :] Dec 31 20:32:34 hdmi/eMMC works out of the box. ;) Dec 31 20:32:52 couldn't manage to get suspend working too Dec 31 20:32:58 for anything else, i'm working on a simple dtb patcher: https://github.com/RobertCNelson/rscm Dec 31 20:33:01 when i do; echo mem > /sys/power/state Dec 31 20:33:12 it works, but resumes right away, Dec 31 20:33:21 might be ethernet interrupts though, not sure Dec 31 20:33:25 there's a lot of crap i need from the ti-3.12.x kernel for suspend to work.. so maybe in 3.14/3.15 Dec 31 20:34:04 so trying to catch up with dev that has happened over the last 6 months .. is there better support for the video card in the BBB? Dec 31 20:34:17 video chip Dec 31 20:35:03 Rickta59, I'm not sure, I run all mine headless. Dec 31 20:35:06 read john's qt 5.1.1 thread on beagleboard groups.. Dec 31 20:35:12 thanks Dec 31 20:37:38 john syne? Dec 31 20:37:56 the long qt 5.1.1 thread.. Dec 31 20:43:14 found this for suppend, pm functions, https://gitorious.org/am335x/linux-am33x/commit/d7f801cce3c9ae5d1617bc80bc170bed0271fb77#arch/arm/mach-omap2/pm33xx.c Dec 31 20:43:42 but was too troublesome to do in midnight :/ i'm already tired, working on this for 3~4 days straight Dec 31 20:45:13 yeap, that file and all it's dependices.. it blows up pretty quick.. best to wait for atleast the clk framework to hit mainline before you really try to port it.. Dec 31 20:45:43 :] ok, anyway got to rest a bit Dec 31 20:45:56 see you all later Dec 31 20:46:23 * musufu goes to eat fake breakfast, and to the hatsumoude afterwards Dec 31 20:55:10 thats where the xorg-server-xvfb-1.14.3-x86_64-2 slackware pkg puts it Dec 31 21:04:09 oops Dec 31 21:31:56 Has anyone ever gotten a build of Chromium OS working on a BBB? Curious. Jan 01 00:39:10 hello Jan 01 02:52:51 greetings all **** ENDING LOGGING AT Wed Jan 01 02:59:58 2014