**** BEGIN LOGGING AT Thu Sep 12 02:59:59 2013 Sep 12 03:00:46 I have tried the procedure documented here: http://www.michaelhleonard.com/enable-bluetooth-on-beaglebone-black/ Sep 12 03:01:42 However, when I plug the dongle in, it doesn't even light up. Nor does it show up when I "lsusb" Sep 12 03:02:33 The BBB is being powered by a 2.5A 5V adapter, so power should not be an issue Sep 12 03:10:24 so confirmed sd performance fix on the gorups awesome find Sep 12 03:10:36 over tripled my sd card read / write speeds Sep 12 03:12:49 nice Sep 12 03:15:45 err sorry only reads i have not confirmed any write differences yet Sep 12 03:23:19 could any one provide me a link to a image for beaglebone which supports Camera cape? I tested the latest image (06-20-1013), the OS recognize the camera cape as "video0", but when I capture images, I always got an error: "Inappropriate ioctl for device" Sep 12 04:35:04 this does seem slower then OE classic Sep 12 05:05:44 hello? Sep 12 05:07:54 I'm very rusty on my linux… I'm trying to create a ubuntu boot SD Card for my beaglebone black and the image I create is coming out as 2Gb.. Is there any good tutorial that will help me expand the partition to use the full 16Gb card that I have? Sep 12 05:11:04 GryphMD: you would have to ask the ubuntu folks as the official supported distro for black is angstrom Sep 12 05:11:09 at least for now Sep 12 05:11:40 angstrom doesn't seem to have drivers for the USB hardware I am running... Sep 12 05:11:42 :( Sep 12 05:12:23 I'd like to sample multiple inputs from a piezo at about 2kHz that is attached to a vibrating machine -- is there an obvious method for this, analog pins would require protection, probably op-amp, and shifting the zero-point, is an ADC an option, if so, what's the easiest transport? Sep 12 05:13:19 wondering whether the simplest is usb-audio with a hub to give multiple imputs? Sep 12 05:14:23 Thanks for the guidance prpplague. Sep 12 05:15:19 anyone on here successfully run bfgminer with multiple asic based bitcoin miners on a beaglebone black? Sep 12 05:19:19 ha Sep 12 05:21:08 LetoThe2nd: the black must flow! Sep 12 05:21:19 LetoThe2nd: package picked up? Sep 12 06:16:17 is it normal for there to be so many /dev/mmc partitions? Sep 12 06:16:22 6 of them I think.. Sep 12 06:20:26 <_av500_> they come for free Sep 12 07:37:12 a wonderfully, rainy JIHAD everyone. Sep 12 07:37:28 i hope you didnt overdo the 9/11 celebrations Sep 12 07:38:12 Sep 12 07:44:20 ah, completly missed 9/11 :p canceling greader was an amazing idea Sep 12 07:47:50 hello Sep 12 07:48:28 is it possible to use hdmi and lcd cape at the same time on bbb ? Sep 12 07:49:11 hello Sep 12 08:01:50 Hi! When I try MACHINE=beaglebone ./oebb.sh config beaglebone the script hangs when trying to clone meta-smartphone :( Sep 12 08:02:18 koen: your customer ^^^^ Sep 12 08:03:08 sorry :( Sep 12 08:06:22 ask the meta-smartphone people what is up with their git server Sep 12 08:06:31 contrary to popular belief, I don't control the internet Sep 12 08:07:30 ok, sorry again, just wanted a confirmation that it may be some server problem Sep 12 08:08:14 mru: thats one for 26+26 Sep 12 08:09:38 koen: you dont?!? Sep 12 08:09:50 nope Sep 12 08:09:53 koen: you disapoint me! Sep 12 08:09:59 i expected more of you Sep 12 08:10:08 we're not on the golden path yet Sep 12 08:10:27 so LetoThe2nd controls our destiny as well as the internet Sep 12 08:10:54 :) Sep 12 08:11:04 all hail me. and bring beer. Sep 12 08:11:05 *lightbulb* Sep 12 08:11:13 * koen loads dune on his kindle for the trip Sep 12 08:13:00 btw, if I want to change some kernel files (dts) and rebuild the kernel, is it better to use these instructions http://www.angstrom-distribution.org/building-angstrom or these instructions or these https://github.com/beagleboard/kernel/tree/3.8 ? Sep 12 08:13:49 ooops, was a "or these instructions" too much :) Sep 12 08:14:39 depends on your skill level Sep 12 08:14:49 the first does everything for you Sep 12 08:15:18 the second leaves some room in between the instructions Sep 12 08:19:24 do I need to install a cross-toolchain for the first or is it handled by the scripts? Sep 12 08:19:37 it is handled by the scripts Sep 12 08:19:50 wow :) Sep 12 08:20:26 LetoThe2nd: beer? i dont support alcoholism. but i can bring you chocolate. that's way more healthy too :-) Sep 12 08:22:33 I guess I will get a chance to modify the kernel files after the " ./oebb.sh update " step i.e. before the bitbake step? Sep 12 08:45:08 KotH: hm. Sep 12 08:53:09 Anguel: go to http://www.angstrom-distribution.org/ and search the frontpage for "kernel workflow" Sep 12 08:53:28 Anguel: the BBW (but strangely not the BBB) SRM also has instructions for that Sep 12 08:54:51 koen: thank you very much! Sep 12 09:06:33 LetoThe2nd: dont tell me you dont want any chocolate Sep 12 09:09:03 KotH: i like chocolate, but $SO likes it more :) Sep 12 09:09:32 particularly swiss chocolate, that is Sep 12 09:10:29 Hi, i've a problem booting my beagleboard Sep 12 09:10:44 here is the output: Hit any key to stop autoboot: 0 mmc_send_cmd : timeout: No status update Card did not respond to voltage select! mmc1(part 0) is current device Card did not respond to voltage select! FAIL:Init of MMC card Sep 12 09:12:35 i'm using rowboat android on mmc and i'm using the same procedure that i've used on beagleboard-xm and beagleboneblack Sep 12 09:12:48 but on beaglebone white doesn't work, the mmc won't boot. Sep 12 09:17:46 LetoThe2nd: bring her with you Sep 12 09:17:59 LetoThe2nd: i make sure i have enough chocolate with me to feed $SO's :) Sep 12 09:18:20 LetoThe2nd: and now that ka6sox is not comming, that shouldnt be a problem ;-) Sep 12 09:20:02 KotH: hehe, will be alone this time. but if you can bring along a pack, will certainly pass it to her grateful hands. Sep 12 09:24:11 koen: do you know if i can skip the meta-smartphone? is it used for BBB at all? Sep 12 09:24:37 [kernel] koenkooi pushed 2 new commits to 3.8: http://git.io/kcsK6g Sep 12 09:24:37 kernel/3.8 da6ea70 Koen Kooi: 3.8: add replicape support... Sep 12 09:24:37 kernel/3.8 9fdb452 Koen Kooi: 3.8: switch beaglebone mmc1 to 4-bit mode... Sep 12 09:25:23 [kernel] koenkooi pushed 2 new commits to 3.12: http://git.io/krH2Fw Sep 12 09:25:23 kernel/3.12 9390cff Koen Kooi: 3.12: update to latest mainline, switch mmc1 to 4-bit mode... Sep 12 09:25:23 kernel/3.12 f05e7b9 Koen Kooi: 3.12: make LED triggers behave like 3.8... Sep 12 09:44:59 Having a bit of issue with Beaglebone Black ethernet. It comes up fine on boot, but if I then do an ifdown eht0; ifup eth0; then the NIC is dead. Only a reboot gets it back. Sep 12 09:57:08 sounds like a driver bug Sep 12 10:03:16 hno: what kernel do you use? Sep 12 10:05:23 hno: Are you doing it over a ssh-connection to eth0? Sep 12 10:05:44 Try doing it in screen or tmux. Sep 12 10:07:16 KotH, have tried both with the image that was on the beaglebone from factory (2012.12), and a freshly build one from 2013.06 angstrom branch. 3.8.13 something in both cases. Sep 12 10:08:10 hno: is it attached to a "green" switch? Sep 12 10:08:28 plundra, I am doing it over serial console. It's defenitely a driver/hardware issue. Verified even with manual config using ip link down/up. ip addresses etc is fine, but after a link down/up it stops sending any packets. Sep 12 10:08:31 there's a bug in the cpsw driver that makes it not wake up for powersave mode Sep 12 10:08:39 the "green" switches trigger that bug Sep 12 10:08:51 tried with two different netgear switches. Not sure if they are "green". Sep 12 10:08:53 (among other things) Sep 12 10:09:08 GS108 and GS108T Sep 12 10:09:29 koen, where can I find more information on that bug? Sep 12 10:09:53 not sure if it's documented somewhere Sep 12 10:10:05 TI was supposed to send me a patch months ago for that Sep 12 10:10:10 but as usual, not patch Sep 12 10:10:22 Ok. Sep 12 10:10:30 make some noise on the beagle mailinglist Sep 12 10:10:47 hopefully gerald will bitchslap someone at TI and the patch will magically appear in my inbox Sep 12 10:11:10 ok. Sep 12 10:12:12 koen: you should collect the paths their daughters take to school and back Sep 12 10:12:21 koen: it would simplify such things greatly Sep 12 10:23:54 qmake2: error while loading shared libraries: __vdso_time: invalid mode for dlopen(): Invalid argument Sep 12 10:25:04 using angstrom-v2013.06-yocto1.4 meta-toolchain-qte MACHINE=beaglebone Sep 12 10:25:28 any idea what is wrong with the toolchain? Sep 12 10:26:06 koen, tried with an old Cisco switch now and same issue. Don't think that one is "green". Sep 12 10:56:28 hno: ok, I still suspect it's teh same bug, though Sep 12 10:56:39 qt-x: stop using ubuntu Sep 12 10:58:55 koen what should I use ? Sep 12 11:02:43 openbsd, and get your hack on Sep 12 11:03:19 qt-x: anything but ubuntu, so debian, fedora, gentoo, arch, opensuse, mint Sep 12 11:03:21 wait Sep 12 11:03:28 mint is ubuntu based, right? Sep 12 11:03:32 if so , scratch that Sep 12 11:03:33 yea Sep 12 11:04:52 dose it happens to know what is the major problem in ubuntu ? Sep 12 11:07:50 yes, canonical Sep 12 11:45:04 koen, agreed. Sep 12 12:30:30 <_vicash_> Hello. I am using the UART 6-pin connector along with a logic analyzer to view any kind of print statements from MLO but I only see the data sent from the uImage. Does anyone know or has anyone tried to view output from the MLO and u-boot blocks ? Sep 12 12:31:32 <_vicash_> Or is there a way to force MLO to run in debug mode such as through the command line either in uEnv.txt or through recompiling of MLO and u-boot ? Sep 12 12:32:06 * LetoThe2nd thinks that there's just nothing that the MLO outputs, probably due to space restrictions Sep 12 12:32:29 but one could probably check that in the source. Sep 12 12:33:24 tinkering with MLO would at least certainly require recompilation, as it gets executed waaay earlier than uEnv gets loaded. Sep 12 12:33:24 * _vicash_ agrees with LetoThe2nd regarding size restrictions Sep 12 12:33:52 <_vicash_> ok Sep 12 12:34:07 <_vicash_> I was wondering if there was a way out of recompiling it Sep 12 12:34:20 i'm pretty sure that not. Sep 12 12:34:34 <_vicash_> as in any kind of runtime configurations. Sep 12 12:34:35 <_vicash_> ok Sep 12 12:35:16 <_vicash_> What other ways of debugging MLO do you recommend ? Do I really have to use TI's Code Composer to actually debug MLO or can I do it in qemu ? Sep 12 12:35:27 huh? Sep 12 12:35:50 <_vicash_> well suppose I want to make changes to MLO and stuff doesn't boot after that. how do I debug the MLO ? Sep 12 12:36:04 <_vicash_> that's my main question. I assumed the UART was giving outputs but that's only after Linux loads Sep 12 12:36:06 MLO can be built as part of u-boot, just have a look at that. check if you can add some kind of output. Sep 12 12:36:31 <_vicash_> yea i will try and see... Sep 12 12:36:51 why tinker with MLO anyways, in the end its just first stage bootstrapping loader. Sep 12 12:37:03 we have a working implementation, thats about all that counts. Sep 12 12:37:16 <_vicash_> well.. for a side project i want to write to the EEPROM in the board Sep 12 12:37:33 so what? Sep 12 12:37:37 <_vicash_> the EEPROM in a BBB is a 4KB EEPROM of which only 78 bytes are used by the MLO boot loader to load the header Sep 12 12:38:10 <_vicash_> the rest is empty. i want to store a key on it... Sep 12 12:38:11 still, so what? Sep 12 12:38:20 how to do it vicash? Sep 12 12:38:21 Hi Sep 12 12:38:33 <_vicash_> and then modify MLO to boot only when the key matches its signature in the MLO Sep 12 12:38:38 u-boot has an spi to do all that. Sep 12 12:38:41 <_vicash_> some sort of signature checking Sep 12 12:38:43 s/spi/api/ Sep 12 12:39:47 <_vicash_> ok... i have to look to find the API.. i went through the MLO source code and it does read the EEPROM so I figured it can read more than what it reads and can modify it accordingly Sep 12 12:39:52 ty LetoThe2nd Sep 12 12:40:01 stored in the old memory banks Sep 12 12:40:05 for when i might need i Sep 12 12:40:06 t Sep 12 12:40:34 Anyone using bbb? Sep 12 12:40:46 heff: nah, just www and sometimes irc. Sep 12 12:40:59 but we do love tla :) Sep 12 12:41:21 Doh, Beaglebone Black... Sep 12 12:41:48 heff: well, just ask your real question. Sep 12 12:41:53 <_vicash_> LetoThe2nd: if you could point me to the file that has the API in the u-boot source that will be great because i couldnt find any API except for the commands for the Hush shell Sep 12 12:43:29 Ok Kwisatz Haderach, I was looking for an update on the SGX drivers, are we still waiting on TI? Sep 12 12:43:31 _vicash_, make sure you look at mainline U-Boot Sep 12 12:43:45 the crypto signature stuff came in, in v2013.07 Sep 12 12:43:57 <_vicash_> Tartarus: ok. i was looking at v13.04 which is what the Angstrom build system uses Sep 12 12:44:27 heff: no idea bout sgx. but i'll let my father know you messed us up ;) Sep 12 12:48:25 <_vicash_> emocakes: the files in board/ti/am335x/ have code demonstrating how to read from EEPROM on BBB Sep 12 12:48:38 <_vicash_> emocakes: in the u-boot git source Sep 12 12:48:45 i'll check it out Sep 12 12:48:48 thanks :) Sep 12 12:48:52 time to zzz though Sep 12 12:48:54 night all Sep 12 12:53:37 I'm getting grumpy about this SGX stuff, starting to feel like I've bought a lemon, again... Sep 12 12:55:00 Caipirinha FTW Sep 12 13:34:56 Hi Everyone. I see instructions for how to update to the latest version of Angstrom in windows, but I don't see where the instructions are for linux users. Sep 12 13:36:10 http://downloads.angstrom-distribution.org/demo/beaglebone/ has linux instructions Sep 12 13:36:16 but basically Sep 12 13:36:21 xzcat img.xz > /dev/sdX Sep 12 13:43:40 koen, I'm wondering if the sdk issues we see are due to Angstrom pinning toolchain versions Sep 12 13:43:55 it would appear the oe-core default does not have the issue Sep 12 13:44:16 in the 1.3 release? Sep 12 13:45:07 Crofton|work: people that care can compare the linaro gcc recipes to the oe-core ones Sep 12 13:45:20 <_vicash_> Can the UART 6-pin header be used for debugging the MLO/u-boot portions of the boot process ? Or is it activated only after the uImage is loaded ? Sep 12 13:45:31 Keon: Thanks for the linux instructions link Sep 12 13:45:38 obviously, we care because it is impacting Angstrom users Sep 12 14:01:42 when I do "MACHINE=beaglebone ./oebb.sh bitbake virtual/kernel" I get "ERROR: Fetcher failure: Unable to find file file://arm/0005-ARM-DTS-AM33XX-Add-PMU-support.patch anywhere." Sep 12 14:03:21 Summary: 1 task failed: /home/anguel/Angstrom-Build/setup-scripts/sources/meta-beagleboard/common-bsp/recipes-kernel/linux/linux-mainline_3.8.bb, do_fetch Sep 12 14:04:29 Anguel: there you go https://github.com/beagleboard/meta-beagleboard/commit/136a28e4d16a03ee9a5e51d6e378f1b80c618b28 Sep 12 14:04:39 * koen wonders how that escaped notice for so long Sep 12 14:07:38 ok, how can I make the script use that? Sep 12 14:14:27 Hi, Can anyone explain me how can i do the following? https://groups.google.com/d/msg/beagleboard/nuyyVDhU6bw/CfoDppz35DUJ Sep 12 14:21:31 .o0(sie m�ssen nur den nippel durch die lasche ziehn....) Sep 12 14:22:12 bbbusb: do wahat? Sep 12 14:22:16 bbbusb: do what? Sep 12 14:22:29 bbbusb: ground your board properly? Sep 12 14:22:59 fine ground or coarse ground? Sep 12 14:26:48 ground the board properly Sep 12 14:27:35 get yourself a copy of "the circuit designer's companion" by Williams. it has a hole section about wiring and grounding Sep 12 14:27:51 read that, understand it, and live happily ever after Sep 12 14:28:00 KotH: sounds awful if it has a hole where the section on grounding is suppose3d to be. Sep 12 14:28:02 good answer Sep 12 14:28:46 LetoThe2nd: well, getting a proper copy without holes is the key here Sep 12 14:29:11 KotH: the key hole? Sep 12 14:29:47 do you think the beagleboneblack has gounding problems? Sep 12 14:30:10 why me and other people are gettings problem using usb devices? Sep 12 14:30:20 bbbusb: no Sep 12 14:30:28 bbbusb: the board itself has no grounding problems Sep 12 14:30:43 bbbusb: what you connect to it, on the other hand, is totally up to you Sep 12 14:30:57 bbbusb: and if you dont do that right (or actually very wrong), then you run into grounding problems Sep 12 14:31:09 i'm using the same usb device that i've usen the last year with beaglebone-xm Sep 12 14:31:31 bbbusb: also, using cheap power supplies is a problem, because they are noisy, wobble around when power changes, etc pp Sep 12 14:31:57 bbbusb: whether one board works better or the other is mostly a function of luck Sep 12 14:32:02 i've only upgraded my hardware switching from beagleboard-xm to beagleboneblack Sep 12 14:32:22 bbbusb: some compoents have more headroom than others Sep 12 14:32:35 ie they tollerate more abuse than others Sep 12 14:59:18 ok who made the arm debian img Sep 12 15:00:20 Is there some Linux driver that can monitor the power button in beaglebone black? Sep 12 15:03:27 <_av500_> is it not mapped to some input event? Sep 12 15:06:34 _av500_, if it is then that's excellent. Sep 12 15:09:43 wonder how to tap into that event.. Sep 12 15:09:58 Hi everybody Sep 12 15:10:17 judging from the snippets in the SRM, the power button works directly on the logic in the PMIC Sep 12 15:11:04 LetoThe2nd, yes, and the PMIC can generate interrupt to the CPU. Question is how to get this event in a Linux application. Sep 12 15:11:13 Koen : I just have recreated a new VM : the new Angstrom is working ok now, no more Zlib issue Sep 12 15:11:57 hno: the question is more like: is the interrupt just generated, but you can't really do much about it then, or can you actually dealy/cancel it Sep 12 15:12:43 Besides, from a comment from here: http://blog.qt.digia.com/blog/2009/09/10/cross-compiling-qtx11/ I'd like to knw if with the BeagleBoard Black it's really better to bitbake qt4-embedded instead of qt4-x11 ? Sep 12 15:13:15 here the comment ", I would recommend (without hesitation) that you use Qt Embedded with DirectFB (The newer Qt, the better as DirectFB is being heavily worked on) for a single process media center-ish application. If the vendor provided a 2D accelerated X11 driver (or an X11 integrated OpenGL ES driver) I would tend to argue in favor of X11," Sep 12 15:13:27 LetoThe2nd, unless you hold it for 8 seconds it's just an event handled by Linux somehow. The PMIC don't do anything. Sep 12 15:13:36 Mouha: qt-e is a massive pain in the ass Sep 12 15:13:42 hno: ah ok Sep 12 15:13:48 Mouha: you have to provide your own windowing server Sep 12 15:14:01 Koen : qwt I suppose ? Sep 12 15:14:12 qwt is a graphing/plotting toolkit Sep 12 15:14:31 there are no generic qws's available Sep 12 15:14:41 and the fullscreen option is hit and miss Sep 12 15:14:54 koen : qws not qwt yes Sep 12 15:15:11 qws is a bait and switch Sep 12 15:15:22 when you complain about it the qt guys will suggest a paid alternative Sep 12 15:15:42 koen : ohhh great to know that ! Sep 12 15:15:59 koen: so obviously qt4-x11 remains the best to build Sep 12 15:17:16 yeah Sep 12 15:17:35 both qt/e and qt/x will use software rendering Sep 12 15:17:39 Ok, so As soon my build is done I will test all of this and tell Anguel about the 4D-Cape Sep 12 15:17:49 with qt/x you get decent input device handling for free Sep 12 15:18:03 got it Sep 12 15:18:25 Fortunatelly I've started with -x11 :) Sep 12 15:18:26 and a proper qt app needs zero code changes to get built against qt/e or qt/x11 Sep 12 15:18:51 as it is with mobiles/desktops Sep 12 15:18:58 if you're using OE, it's a matter of replacing 'inherit qt4x11' with 'inherit qt4e' Sep 12 15:19:02 the debian wheeze arm img for bbb needs fixing Sep 12 15:19:06 grrr Sep 12 15:19:19 Koen: nope, I m using Angstrom of course .... Sep 12 15:19:27 they did not set a generic fqdn in the /etc/hosts file Sep 12 15:19:33 using OE to build angstrom :) Sep 12 15:20:02 <_vicash_> Hi, Does anyone know if the linux kernel uImage can be built in a FIT format ? Sep 12 15:20:28 so if you try to install qmail it bombs Sep 12 15:20:31 grrr Sep 12 15:20:48 Koen : ah ! ok, no I've followed all from : http://www.angstrom-distribution.org/building-angstrom Sep 12 15:24:34 Mouha: that's what I meant, you're good :) Sep 12 15:24:49 Mouha: you can test with ts_calibrate and then ts_test, no need for Qt Sep 12 15:26:12 any idea if i can remove the LCD dts files just by deleting them from the kernel sources? Sep 12 15:26:43 i want to load them dynamically outside the kernel, from /lib/firmware if I am not mistaken Sep 12 15:28:51 <_vicash_> How does one access the u-boot prompt ? I keep reading in the docs in the u-boot source code that some commands can be given at the prompt yet there is no way to access it on the BBB.. Sep 12 15:29:58 _vicash_: of course there is, the uart on the header Sep 12 15:30:43 <_vicash_> LetoThe2nd: i connected the UART to Hyperterminal on Windows 7 but got no response ..on the other hand I connected it to a logic analyzer and got the outputs to the screen. Sep 12 15:31:16 _vicash_: well so guess what is the culprit? ;) Sep 12 15:31:24 <_vicash_> LetoThe2nd: have you had any success with Hyperterminal or should I use putty or something like that ? Sep 12 15:31:49 <_vicash_> I was using a straight through serial cable , do i need a null modem cable instead ? Sep 12 15:32:19 _vicash_: did you connect it straight to your computers rs2323? Sep 12 15:32:26 <_vicash_> yes Sep 12 15:32:46 _vicash_: and yes, i'm often using it succesfully. i use an os though, not windows/hyperterminal. Sep 12 15:33:00 _vicash_: hope you haven't fired anything then, the uart is 3,3v only. Sep 12 15:33:17 _vicash_: manuals are there for reading, you know? *SCNR* Sep 12 15:33:46 <_vicash_> LetoThe2nd: i read the BBB_SRM manual but it did not say much except for what the correct pins to use were Sep 12 15:33:58 _vicash_: BBB_SRM 5.6 (pg.34) Sep 12 15:34:18 _vicash_: it says "TTL" Sep 12 15:34:31 <_vicash_> ok. Sep 12 15:34:39 <_vicash_> so i need to buy that then Sep 12 15:34:47 yep. Sep 12 15:35:25 <_vicash_> thanks. that seems like an inevitable purchase now :( Sep 12 15:35:31 and sacrifice something cool, like a bunch of old floppies or what ever and pray to $DEITY that your 15v out of the rs232 haven't fried the RX port. Sep 12 15:35:53 well I got arm pkgs build and I have a basic start to a pbx on bbb Sep 12 15:36:02 this rocks Sep 12 15:36:11 <_vicash_> LetoThe2nd: yes i shall pray and maybe buy a second BBB as a backup Sep 12 15:44:25 is the floating point calculation hard or soft on the beagle board? Sep 12 16:30:44 check chip specs.. Sep 12 16:41:47 <_av500_> C2H6O_: hard Sep 12 16:45:53 Hi, Beaglepeople Sep 12 16:45:57 LetoThe2nd, TTL is NOT 3.3V. Sep 12 16:46:37 How much analog input and analog output have the new beaglebone ? Sep 12 16:46:46 hno: thats right. Sep 12 16:46:54 <_av500_> beagle: that is documented Sep 12 16:47:08 hno: however, the "TTL" should at least ring the bell that its not standard rs232 Sep 12 16:47:54 yes. Sep 12 16:48:24 Someone here who use a DOGXL160-104 Display with i2c ? Sep 12 16:49:48 _av500_: where please ... Sep 12 17:04:39 <_av500_> the SRM for example Sep 12 17:13:33 Hi all. I was wondering if anyone has had trouble with SSH for BeagleBone Black rev 0A5C Sep 12 17:14:32 <_av500_> updated to the latest SW? Sep 12 17:14:36 <_av500_> please do that first Sep 12 17:14:53 I haven't, though wasn't sure it was necessary Sep 12 17:15:12 did anyone ever successfully connect BDI3000 to BBB ? Sep 12 17:55:28 hm whats the trick mentioned here in adding that gpio overlay? http://elinux.org/Bacon_Cape#Features Sep 12 17:57:49 add them in /lib/firmware/BB-BONE-BACONE-00A0.dts ? Sep 12 18:00:22 <_av500_> LetoThe2nd: google it dammit Sep 12 18:00:26 <_av500_> :) Sep 12 18:00:45 _av500_: tried, but my goog-efu seems to be broken today Sep 12 18:03:19 the 7segment is always on and blinding me :( Sep 12 18:05:06 ah, manually setting gpio48 out seems to do the trick. Sep 12 18:05:16 LetoThe2nd: i need to post some updated info Sep 12 18:05:26 LetoThe2nd: i'll try to get it posted this afternoon Sep 12 18:05:44 prpplague: no worries. just tinkering :) and as you might have guessed, things have arrived Sep 12 18:05:49 yay \o/ Sep 12 18:05:52 indeed Sep 12 18:05:57 LetoThe2nd: no issues with customs? Sep 12 18:07:13 see msg :) Sep 12 18:35:12 anyone here using the angstrom-distribution setup-scripts? specifically the console-image, and if so are you using it with wifi? Sep 12 18:36:56 I'm having trouble with the console-image not being able to use the wifi (and yes... I did pull the ethernet cable). I was able to get it working buy doing opkg update; opkg upgrade and then updating the config files for connman. Sep 12 18:37:41 but I'd rather not have to go through those steps. I'd rather just be able to build and image and have it work. Anyone? Sep 12 18:39:31 btw I'm using a beaglebone black rev. A5C Sep 12 18:41:17 so works from feeds, but not from build? Sep 12 18:42:07 <_troll_> feed the troll Sep 12 18:42:52 <_av500_> +1 Sep 12 18:43:20 Crofton|work, yes Sep 12 18:44:30 * Crofton|work slaps the trolls Sep 12 18:44:47 can you compare dmesg for success and fialure? Sep 12 18:46:22 Crofton|work, i'm trying to do a SD dupe to eliminate the SD card itself. Sep 12 18:46:38 heh Sep 12 18:46:49 It would be good to compare the kernel log Sep 12 18:46:56 I'm copying my working SD to my NON working sd card. Sep 12 18:47:19 hmm, the intel spy is back Sep 12 18:47:36 if the problem goes away. then at lease I know it's not the SD card some how... This probably only makes sense to me... Sep 12 18:53:23 Crofton|work, it still doesn't address the root-cause of the defect(?) Sep 12 18:53:39 the big bang? Sep 12 18:55:07 I've also tried with stock console-image and a different rtl8192cu driver file. That didn't help either. Sep 12 18:56:30 if i can get an image working on my RPi i will abandon the Beagle Black. I'm just that fed up with it. Sep 12 18:58:15 I'm adding a wifi card for crying out loud. I'm not trying to get it talking to some undesigned, alpha-code usb widget. Sep 12 18:58:52 did you istall the firmware Sep 12 18:59:02 like I found out I had to for my usb Sep 12 18:59:15 wich I still have issues with Sep 12 18:59:26 yeah, well duh, of course I installed the firmware... It's in there, checked and double checked. Sep 12 18:59:32 I can get it to link andget a ip but I cant surf Sep 12 18:59:41 ok Sep 12 19:00:22 I want to know what the usb in the /etc/network/interfaces is for Sep 12 19:00:22 it needs better documenting Sep 12 19:01:00 rneese, I have one SD card that works... stupid me, forgot how I got it working. All I remember is doing opkg update; opkg upgrade and then adding the wpa_supplicant.conf and wifi.config files. Sep 12 19:01:29 rneese, a different connection manager... Sep 12 19:01:47 ok Sep 12 19:02:09 well I use the debiam img system to build my sd cards Sep 12 19:02:34 and I am now using the same system to then build pbx boxies easier Sep 12 19:02:36 brb Sep 12 19:02:42 Angstrom uses connmann (barely documented). Are you using a stock image or building your own? Sep 12 19:03:54 * koen wonders why people still assume USB stuff will actually work on a beaglebone Sep 12 19:04:30 s/on a beaglebone// Sep 12 19:04:52 and s/usb// Sep 12 19:05:04 because it should! Sep 12 19:05:20 I am using the beagle bone debian img Sep 12 19:05:28 its just a base install Sep 12 19:06:05 with the proper kernel modules Sep 12 19:06:18 fixing people's expectations is easier than fixing the product Sep 12 19:06:23 I'm trying to use koen's setup-scripts to build a customized Angstrom console-image. But he doesn't seem to see the importance of WiFi enablement on the beaglebone. Sep 12 19:06:54 Let's lower our expectations then, and we'll all be happier. Sep 12 19:07:55 * vvu|Log sees that a war is incoming here. back to my anti-nuclear bunker Sep 12 19:08:48 cfo215: the USB hardware is broken in silicon Sep 12 19:08:52 Hello Sep 12 19:09:12 * mru thought usb by broken already in the committee Sep 12 19:09:28 I have an issue ssh Sep 12 19:09:36 cfo215: I do seem the importance of wifi on the bone, but that doesn't mean I can't point out that USB things are *$(@*$@( broken on the bone Sep 12 19:09:38 Can anyone help me with that? Sep 12 19:11:25 koen, but works for one method but not another Sep 12 19:11:33 definition of lazy nerd? I am stirring my gin & tonic with a pci slot blank because I'm too lazy to get off my ass and get something else Sep 12 19:11:35 koen, who do we put a fire under to get them corrected? Circuitco? Angstrom-Distro? Ti? Sep 12 19:11:37 cfo215, pastebin th elogs of success and failure Sep 12 19:12:07 Crofton|work, will do. Sep 12 19:12:39 hi Sep 12 19:13:01 cfo215: TI of course Sep 12 19:13:07 noone else can fix the silicon Sep 12 19:13:25 koen, true that! Sep 12 19:13:42 if someone speaks with TI tell them about the OTG problem Sep 12 19:13:46 i'm looking at beaglebone black. Is there an IDE that comes with it so that I can develop code? Sep 12 19:14:14 mikeB_: think of it like developing code for a linux PC. use whatever IDE you want Sep 12 19:14:45 * Crofton|work goes to hacking on a zynq device tree Sep 12 19:14:51 then download your code into linux? Sep 12 19:14:53 because all Si vendors suck Sep 12 19:15:49 can you get at the nuts and bolts of the processor? Sep 12 19:16:12 maybe if you had small enough tweezers Sep 12 19:17:16 so you can't directly write code for the processor, like access the I/O ports etc? Sep 12 19:25:52 <_av500_> sure Sep 12 19:25:56 <_av500_> CPU runs SW Sep 12 19:26:03 <_av500_> even on the beagle Sep 12 19:26:48 <_av500_> mikeB_: it is not special in that regard Sep 12 19:26:55 <_av500_> you can write code on it or on the PC Sep 12 19:27:01 <_av500_> compile with fancy IDE or command like Sep 12 19:27:13 <_av500_> use python or assembly Sep 12 19:27:20 <_av500_> run and OS or run your own Sep 12 19:27:38 <_av500_> its just an ARM core with some periphery around Sep 12 19:30:37 is there any sort of debugger? Sep 12 19:31:15 well I got freeswitch running and now to start devel the new gui Sep 12 19:31:24 this is going to be great Sep 12 19:32:51 mikeB_: gdb works fine, or you can remote debug with an IDE Sep 12 19:33:08 there's a jtag header? Sep 12 19:33:59 If I configure a pin as a gpio output, then as a pwm , then back to a gpio I can no longer set its value Sep 12 19:34:06 mikeB_: don't need jtag for gdb unless you are debugging something in the bootloader Sep 12 19:34:35 anybody use the camera module before? Sep 12 19:34:41 The first time works fine, but the second time it doesn't. I am using the cape manager and the gpio device driver Sep 12 19:42:29 <_av500_> mikeB_: unless you write your own OS, gdb and printf are fine Sep 12 19:42:33 <_av500_> and printk Sep 12 19:59:15 _av500_: dont forget the osciloscope, DMM and logic analyzer Sep 12 19:59:36 <_av500_> anda wet sponge Sep 12 20:01:00 sometimes, a slave to beat up when nothing works, helps too Sep 12 20:18:35 <_SY_> hello all, anyone here? Sep 12 20:18:54 <_av500_> no Sep 12 20:18:59 <_av500_> all gone home Sep 12 20:19:06 <_av500_> just the cleaning crew Sep 12 20:19:29 _SY_: check the topic for the channel Sep 12 20:19:58 <_SY_> hello, can you help me? do you know if there is a zig bee board for the beaglebone? Sep 12 20:21:20 <_av500_> https://github.com/raul7/ZigBeagle Sep 12 20:21:32 <_SY_> ty, will take a look now Sep 12 20:21:34 <_av500_> xbbe are SPI Sep 12 20:21:37 <_av500_> xbee Sep 12 20:21:43 <_av500_> hooking them up should be easy Sep 12 20:22:01 <_av500_> get a protocape and solder one Sep 12 20:22:17 or get a usb zigbee widget Sep 12 20:22:49 sadly zwave seems to have stolen most of the zigbee functionality, gone proprietary, and greatly reduced the market for zigbee widgets Sep 12 20:23:30 I was hoping to get zigbee lightbulbs, light switches, and outlets to use as repeaters Sep 12 20:23:39 then zigbee er, non-repeaters for sensors Sep 12 20:23:53 doesn't seem much out there that is turnkey Sep 12 20:24:47 <_av500_> true Sep 12 20:24:55 <_av500_> there's BLE stuff coming Sep 12 20:24:59 <_av500_> eventually Sep 12 20:25:01 <_SY_> I have a colleage who is exploring using zigbee on a platform offshore. Sep 12 20:25:05 <_av500_> but that does not mesh Sep 12 20:26:18 <_SY_> MESH is exactly what he wants Sep 12 20:26:30 <_av500_> sure, zigbee does that Sep 12 20:26:40 <_av500_> get a few xbees and hook them up Sep 12 20:26:53 <_av500_> there are also $5 zigbee modules on ebay Sep 12 20:27:11 <_SY_> Thank you, if he is interested in pursuing it further I will get some and hook them up. Sep 12 20:27:46 does ble do mesh? Sep 12 20:29:23 the paypal ble widget looks pretty cool, but isn't obviously mesh. Sep 12 20:29:51 <_SY_> g2g, thanks for the information. Sep 12 20:29:53 <_SY_> bye Sep 12 20:34:18 <_av500_> spikebike: nope Sep 12 20:34:23 <_av500_> BLE is for short range Sep 12 20:35:13 short range seems compatible with mesh Sep 12 20:35:18 but ok Sep 12 20:35:21 <_av500_> no Sep 12 20:35:29 saw the paypall ble widget, looks pretty slick Sep 12 20:35:38 futuristic even ;-) Sep 12 20:35:41 <_av500_> I dont think even std BLE does meshing in practice Sep 12 20:35:59 <_av500_> its in the std, but I doubt anybody implemented it Sep 12 20:39:53 Can someone help? Sep 12 20:40:00 I have a BBB which refuses to boot Sep 12 20:40:12 <_av500_> ask Sep 12 20:40:15 keep going . . . Sep 12 20:40:28 No LED is lit after power up. Sep 12 20:40:39 not even power led ? Sep 12 20:40:47 Connected to the serial debug console Sep 12 20:41:13 Got: Sep 12 20:41:15 U-Boot SPL 2013.07-00013-g469f02f (Jul 30 2013 - 12:27:41) musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn) musb-hdrc: MHDRC RTL version 2.0 musb-hdrc: setup fifo_mode 4 musb-hdrc: 28/31 max ep, 16384/16384 memory USB Peripheral mode controller at 47401000 using PIO, IRQ 0 : 0, U-Boot SPL 2013.07-00013-g469f02f (Jul 30 2013 - 12:27:41) Sep 12 20:41:45 Nothing else Sep 12 20:41:59 Power led is lit. Sep 12 20:42:02 naturally. Sep 12 20:42:06 maor what did you do to it ? Sep 12 20:42:12 no idea Sep 12 20:42:13 no not naturally Sep 12 20:42:33 you didnt install a new distro or something ? reflash the EMMC ? Sep 12 20:43:03 first thign that goes through my mind is that MLO isbad Sep 12 20:43:17 I have another BBB that works. Sep 12 20:43:30 I have a MicroSD with angstrom on it. Sep 12 20:43:37 The other BBB works of it. Sep 12 20:43:50 but the problem one wont ? Sep 12 20:43:52 When I put it in the dead BBB it doesn't boot from it. Sep 12 20:44:08 no status LEDs Sep 12 20:44:14 you might want to consider requesting an RMA Sep 12 20:44:26 get the same result from the serial console Sep 12 20:44:27 Anyone familiar enough with bluetooth/zwave to commend on the security for things like deadbolt on a front door? Sep 12 20:44:43 comment rather Sep 12 20:45:14 <_av500_> I commend security Sep 12 20:45:56 spikebike probably a bad idea Sep 12 20:46:27 * XorA is amused by a house you cant get out of when your battery is dead :-D Sep 12 20:46:42 Or get it... Sep 12 20:46:47 get in... Sep 12 20:46:48 im more amused at all the linux hackers who could get in ;) Sep 12 20:46:54 <_av500_> battery can be outside Sep 12 20:47:13 m_billybob: yeah, thought so. Seems like many assume because popular chips don't let you fake the mac address or equiv that it's somehow impossible to send the same radio signal Sep 12 20:47:35 <_av500_> if your security is only mac, you are doing it wrong Sep 12 20:47:50 av500: agreed, seems common with bluetooth/zwave though Sep 12 20:47:50 <_av500_> the transport itself should not be the security components Sep 12 20:47:57 you can fake mac addresses Sep 12 20:48:01 with BT Sep 12 20:48:08 just use broadcast mac everywhere Sep 12 20:48:10 m_billybob: many chips do not suppor tthat I believe. Sep 12 20:48:23 spikebike it can be done all in software Sep 12 20:49:02 The reason I bring this up is I want to write something for android that will securely encrypted a command and send it to a beaglebone to securely control the opening/closing of a door. Sep 12 20:49:10 <_av500_> do it like car keys Sep 12 20:49:31 spikebike erm heh use the internet Sep 12 20:49:33 I wouldn't mind wireless from BBB -> door, but sounds like I should just run some copper Sep 12 20:50:39 spikebike, but it could be done with bt, but you'd have to think about it carefully Sep 12 20:50:56