**** BEGIN LOGGING AT Sun May 06 03:00:02 2012 May 06 03:10:29 hello May 06 03:11:48 anybody available for troubleshooting help? May 06 03:17:51 hello, I got beaglebone rev. A5. I installed debian.Usb to serial was working perfectly, (/dev/ttyUSB1), until I upgraded the system, so I can't serial com to the device anymore May 06 03:18:29 I booted with angstrom again, to test serial connection and neither from angstrome worked May 06 03:18:47 but the usb port supplies power without any problem May 06 04:35:09 the beagle bone can run detain! May 06 04:35:12 debain! May 06 04:46:35 err.. debian? May 06 04:48:50 yea Mac autocorrect for some reason. May 06 04:49:24 I still haven't decided if I am going to buy a beaglebone. May 06 05:40:48 agmlego, thx, i searched within angstrom, no recipe for inadyn or ez-ipupdate either May 06 06:08:42 I'm designing a cape for the beaglebone which will require a few digital pins. Any recommendations on which to pick to ensure they're least likely to conflict with other capes? May 06 08:50:48 hello May 06 08:51:33 dear all, i am having usb problem with my xM rev c (c1 says sticker) May 06 08:51:43 anyone willing to help ? May 06 08:52:49 maybe, if you would be so kind to explain your problem May 06 08:53:29 i am the one who is having usb power problem (d14 not litting) on my beagleboard xm rev c (c1 says sticker on board). When board originally came, usb was working, but i was foolish enough not to backup the original shipped image. May 06 08:53:45 <_av500_> aso download another one May 06 08:53:48 <_av500_> -a May 06 08:54:14 i tried like a 1000 different images and withouth success May 06 08:54:54 can someone link the working image for my board (i mean have the board and tried that) May 06 08:55:56 i tried everything from https://groups.google.com/forum/?fromgroups#!topic/beagleboard/wF7arvmbYaI May 06 08:55:59 <_av500_> http://www.circuitco.com/support/index.php?title=Circuitco_Support_Wiki#BEAGLEBOARD-xM_IMAGE_FILES May 06 08:56:46 i tried both 3_ and 4_ version, also withouth success May 06 08:56:58 also generally powering the board using a power supply is preferred May 06 08:57:13 very good explanation of the problem May 06 08:57:28 i have +5V, 2.25A power supply May 06 08:58:27 and what is the problem? May 06 08:59:00 i get no power to usb, d14 is not litting May 06 09:00:06 paste the full bootlog somewhere, e.g. at fpaste.org May 06 09:00:20 @av500 i tried both 4_25 and 3_30 images, not working May 06 09:01:02 i do not have serial cable... May 06 09:02:15 so how do you know that the problem is usb? May 06 09:02:21 sorry aholler, i should go buy one and paste it? May 06 09:03:05 you don't have much possibilities to debug without one May 06 09:03:20 i assume that power is not delivered to usb if d14 is not litting ? May 06 09:03:29 you might describe how you put those images onto your sd May 06 09:04:09 i used win32imager tool for img files May 06 09:04:51 and scripts that came with images under linux May 06 09:05:07 i mean tar.bz2 images under linux May 06 09:07:31 i can get into Angstrom/Android/Ubuntu but without usb May 06 09:08:01 so paste the output of dmesg May 06 09:08:06 (somewhere) May 06 09:08:41 how to get dmesg if my keyboard is not working ? May 06 09:09:58 look at /var/log on the sd. May 06 09:12:43 jkridner, koen: I assume https://fedorahosted.org/liveusb-creator/ could be a good because that downloads images by itself. Modifying that to your needs should be simple. May 06 09:14:28 ah ok, i am having now "working" gingerbread 2_3 installed on sd (where is its log ?), or i can rebuild Angstrom and paste log later May 06 09:15:26 you could also add "dmesg >/dmesg.txt" to the init-script of your distro May 06 09:16:52 i am kind a new linux, how to do that ? May 06 09:22:44 for android it would mean to andd this command to the end of init.rc ? May 06 10:16:33 please, no pm May 06 10:18:54 regarding not working USB on xm rev c (C1) i have the dmesg output: http://fpaste.org/c25W/ May 06 10:26:04 no error there, do you have connected a device? May 06 10:26:14 yes, mouse May 06 10:26:45 d14 not litting, no power to mouse (laser is off) May 06 10:33:46 the top of dmesg is missing. try without nouse, the lan should be found as it is an xm. May 06 10:35:45 this is with keyboard and mouse: http://fpaste.org/UoVS/ May 06 10:35:55 going for without anything May 06 10:39:01 without anything attached(dvi and power only): http://fpaste.org/O9TA/ May 06 10:40:48 and that is what? android? May 06 10:41:14 Angstrom May 06 10:41:29 angdroid May 06 10:41:44 angry droids? May 06 10:41:49 angst rom, the buffer is to small ;) May 06 10:42:03 4_24 version from: http://www.circuitco.com/support/index.php?title=Circuitco_Support_Wiki#BEAGLEBOARD-xM_IMAGE_FILES May 06 10:42:10 4_25 sorry May 06 10:42:42 I still thing OE should adopt "putting the angst in angstrom" as their slogan May 06 10:43:54 is that normal on a xm? [ 0.000000] Beagle cameraboard: lbcm3m1 May 06 10:44:35 did you attach a leopard camera module? May 06 10:45:00 or claim you did May 06 10:45:04 nothing is attaced May 06 10:45:05 don't know if those things are detected May 06 10:48:15 any clue guys ? May 06 10:49:21 Guest70386: try log_buf_len=128K so that the whole output will be visible May 06 10:49:50 where to set that May 06 10:50:02 in uEnv.txt May 06 10:50:22 add it to the end of something or use optparams May 06 10:52:24 attach device or not ? May 06 10:52:47 no, as said at least the lan should be found May 06 10:53:11 so the less you are attaching the less can brake something May 06 10:54:15 anyway, I don't know why that camera-foo is detected. May 06 10:54:33 here it is: http://fpaste.org/CH4m/ May 06 10:54:37 I don't have an xm, so I can't say more May 06 10:55:27 the top is still missing, try adding the log* after rootwait (same line, just write it at the end) May 06 10:56:12 or paste your uEnv.txt May 06 10:56:21 my uEnv May 06 10:56:36 mpurate=1000 dvimode="hd720 omapfb.vram=0:8M,1:4M,2:4M" vram=16M optargs="consoleblank=0" console="tty0 console=ttyS2,115200n8" log_buf_len=128K May 06 10:57:11 put it after the n8 (inside the quotes) May 06 10:58:08 adding debug might make sense too May 06 10:59:16 i am sorry, what to add after n8 ? May 06 10:59:34 that loglevel, move it inside the quotes May 06 10:59:41 log_buf* May 06 11:01:00 so console="tty0 console=ttyS2,115200n8 log_buf_len=128K" ? May 06 11:01:24 yes May 06 11:04:21 new: http://fpaste.org/VoTB/ May 06 11:04:59 hmm, are you sure you are pasting everything? May 06 11:05:37 ctrl A + ctrl C + ctrl V May 06 11:06:08 <_sundar_> ctrl (A + C +V) May 06 11:07:13 try 1MB instead of 128K, but I don't think that is the problem why your log is missing the top May 06 11:07:28 1M May 06 11:07:54 128k should be enough May 06 11:08:22 * mru thought it was 640k that ought to be enough May 06 11:08:39 for gates ;) May 06 11:09:02 <_sundar_> people would require no more than 640k lol May 06 11:17:05 again same with 1 M May 06 11:21:02 should i throw this bb and buy new one ? May 06 11:22:48 I see no errors in that log May 06 11:24:13 so, why then usb would not work. D14 is not litting May 06 11:25:23 it's detecting the onboard hub just fine May 06 11:28:10 what kernel is that you're running? May 06 11:29:42 it is from circuitco 4_25 demo image, 2.6.32 kernel it says May 06 11:30:57 sounds dodgy May 06 11:31:23 ? May 06 11:32:48 have you tried anything else? May 06 11:33:11 like a recent angstrom 3.x kernel May 06 11:33:13 yes, a dozens of images android/ubuntu May 06 11:33:35 i used narcissus build also, i has new kernel ? May 06 11:33:47 it* has new kernel ? May 06 11:34:13 and all of them gave me the same thing, now power to usb May 06 11:34:29 sorry, *no power to usb May 06 11:39:05 i tried also all of: http://downloads.angstrom-distribution.org/demo/beagleboard/ May 06 11:40:04 GNOME_TI iamge uses 3.x kernel, but i had no power for usb there too. May 06 11:42:08 what i really do not get it, is that a lot of people is having same issue andd there is no unique rule how to solve that problem May 06 11:42:34 maybe they are not having the same problem May 06 11:43:23 no power in usb, it might be from different causes ? May 06 11:43:47 almost everything can have different causes May 06 11:43:57 the world is not a simple place May 06 11:44:42 yea, true. I guess i dont have luck with bb ... May 06 11:45:45 if some1 else do not have any more suggestions i am going to take some sleep May 06 11:50:11 ok thank you guys for the help. I will try again later. May 06 13:28:10 Knightmare_: pong (in private window) May 06 13:28:49 Knightmare_: sorry, IRC window was hidden behind lot of windows May 06 13:38:55 Any1 know if there is some digestable information about how to use PRUSS ? I tried to toggle GPIO pins as fast as possible, but seems like max speed is about 78µs for a toggle. Thats 12.8kHz and certainly not stable, so I guess I'll need to use PRUSS May 06 13:45:51 how did you do said toggling? May 06 13:47:39 using the filesystem May 06 13:47:44 that probably why its quite slow May 06 13:48:21 yeah May 06 13:48:23 I just made a loop that writes 10000 time a 1 to the file and then a 0 May 06 13:48:30 and atached the logic analyser May 06 13:48:35 thats where I get these speeds from May 06 13:48:40 use a more direct method May 06 13:48:54 would you be so kind to help me a little ? May 06 13:49:04 just some hints or so May 06 13:49:12 I work in C by the way May 06 13:49:53 78µs resolution would be fine for my stepper motors to do normal work, but when my laser is going to engrave, I really need a lot more :( May 06 14:05:33 mru? no hints available please? May 06 14:14:00 educa: accessing gpio from user space can be limited due to overhead May 06 14:14:09 educa: try accessing it from kernel space May 06 14:18:53 I do understand what you are saying , that kernel can interrupt me at any time. I just don't know how to solve this since I don't know anything about kernels :( thats why I ask if anyone could help a little, but it looks like thats going to be difficult. I understand that you guys are all busy but I'm afraid I won't find it all by myself May 06 14:19:22 I just found: http://41j.com/blog/2011/09/beagleboard-gpio-input-driverless/ May 06 14:19:27 dunno if thats what you mean? May 06 14:19:38 some kind of direct memory access to gpio ? May 06 14:39:52 Hey May 06 15:21:03 am I right to say that if you acces gpio using direct memory access, you don't actually set 1 single pin, but you set a whole port at once? May 06 16:27:54 Has anyone ever seen this at the end of a beaglebone build? "| ERROR: ld.so: object 'libpseudo.so' from LD_PRELOAD cannot be preloaded: ignored. | Creating rootfs loopback | genext2fs: set_file_size: ftruncate: Invalid argument" May 06 16:30:35 I would suspect the real error above May 06 16:46:00 forgive the possibly stupid question, but can you mount a USB memorystick to be readable on the usbport of the beaglebone? I mean not the port connected to the pc, but the other port of course May 06 16:48:18 usually that's an important use case of usb-ports May 06 16:48:45 yes I know :) May 06 16:50:54 * mranostay scrolls up and regrets it :) May 06 18:17:00 hi koen-ster May 06 18:18:10 hey May 06 18:18:39 * koen directs some radiation at mranostay May 06 18:23:22 koen, I saw you tested LCD3 , is that going to become available soon ? May 06 18:28:28 Hi, need to update my beaglebone SDcard and have been following the recipe in the Beaglebone 101 ppt. Downloaded Angstrom-Cloud9-IDE-GNOME-eglibc-ipk-v2012.05-beaglebone-2012.04.22 and writing the image to the card using the win32 disc imager...but the board wont boot ...any hints?? May 06 20:10:07 hi there May 06 20:10:40 I'd like to know if the beagleboard-xm is h.264 capable with hardware acceleration (1080p) May 06 20:11:10 I'm trying to decide between the beagleboard and pandaboard, and I really need help May 06 20:14:47 it is not May 06 20:17:36 even with C64x-accelerated mplayer? May 06 20:17:55 no, and such a thing does not exist afaik May 06 20:19:06 maybe I'm missing the point: I saw there's a NEON accelerated version May 06 20:19:10 isn't it the same? May 06 20:19:20 no May 06 20:19:31 standard Angstrom mplayer package utilises NEON optimization May 06 20:19:38 yes May 06 20:19:49 ok, what's NEON optimization then? May 06 20:21:27 * LetoThe2nd offers mru some buzzword beverage May 06 20:21:28 I wouldn't be asking if I could find a clear answer myself May 06 20:21:51 neon is a simd instruction set on the ARM May 06 20:22:09 ah! ok, now it's clear May 06 20:22:31 it's non video-decoding specific I guess May 06 20:22:58 so, how's the building quality on the Pandaboard ES? May 06 20:23:00 no it isn't May 06 20:24:10 is it a good idea to get the Pandaboard ES instead? I know it's the beagleboard channel, but I believe the Pandaboard is the next logical step May 06 20:24:54 the only thing that keeps me away from it its the fact that it's subsidized by IT May 06 20:25:09 while the Beagleboard isn't May 06 20:25:30 what's your opinion on this? May 06 20:25:58 whats bad in getting hardware cheaper than the real market value? May 06 20:27:10 * _av500_ prefers that Knightmare_ gives the money to him May 06 20:27:40 good idea. May 06 20:27:40 Knightmare_: NEON is particularly tailored to media uses, because it isn't 100% compliant with the IEEE floating point standard May 06 20:27:57 XMPPwocky: that conclusion is rather far-fetched May 06 20:27:57 LetoThe2nd: nothing I guess: it's just that Beagleboard is kind of "standard" and it could be manufactured again in the future for the same price, but we can't guess what the real price of manufacturing the Panda in the future if TI decides to drop it May 06 20:28:20 XMPPwocky: neon floating-point doesn't have traps and all rounding modes May 06 20:28:27 that doesn't mean it's useless for non-media things May 06 20:28:30 mru: not really. it makes specific tradeoffs for speed May 06 20:28:34 and video codecs don't use floats at all May 06 20:28:59 Knightmare_: pretty short-sighted, you are not able to build a beagle also if ti decides to not sell the omap3 any more ;) May 06 20:29:12 note that I'm not a hardware maker / cloner of any kind: but I've been using pbsolete hardware for decades now (Amiga) and that's why I'm concerned May 06 20:29:48 note that neither beagle nor panda are intended to be used in production systems May 06 20:30:07 mru: ... really? May 06 20:30:14 would you happen to have Boeing's number May 06 20:30:19 uh? May 06 20:30:24 :P May 06 20:30:35 you totally lost me there May 06 20:30:36 I'm not using it in a production system: I just get attached to a given hardware and I want to use it for decades May 06 20:30:52 so if I go for a board, I'd prefer one that could be easily replicated in teh future May 06 20:31:01 with good care, a beagle or panda should last you quite some years May 06 20:31:25 I know it sounds bit obsessive-compulsive, mru :) May 06 20:31:50 don't count on anyone building beagles in 5 years time May 06 20:31:51 I also noticed the Panda board has a LOT more resistors on board May 06 20:31:52 let alone 10 May 06 20:32:01 the panda has a lot more everything May 06 20:32:45 yes, I know it has a lot more systems on-board and that implies a lot more components May 06 20:32:49 * mranostay waves May 06 20:33:13 but that's also added complexity May 06 20:33:34 so it could cause more failures May 06 20:33:55 is the building quality of the Panda ES good? May 06 20:34:05 or could we call it "cheap hardware"? May 06 20:34:13 Hello folks, just short survey: What are you using your beaglebone or -board for? May 06 20:34:21 mlbox: OS development May 06 20:34:46 which OS? May 06 20:34:51 mine :P May 06 20:35:25 katari? May 06 20:35:46 yeah May 06 20:36:50 XMPPwocky interesting May 06 20:36:57 sounds like we have a r-pi person in here :) May 06 20:36:59 yes, I am :3 May 06 20:38:06 you divide people in r-pi and beagle-people? May 06 20:39:11 well, the former don't quite qualify for the term 'people' May 06 20:39:47 usually we divide between rpi fanboys and mankind ;) May 06 20:39:49 May 06 20:40:12 raspi is an incredibly awesome thing, and I totally support it. it's made to get kids interested in coding May 06 20:40:31 and I think it'll probably be great at that May 06 20:40:34 as long as people don't expect it to run ubuntu... May 06 20:40:40 Knightmare_: unless there are sensitive private details involved, please keep discussions here May 06 20:40:56 ok, thanks :) May 06 20:41:08 sorry to bother you May 06 20:41:12 I also think both are nice. no need to flame. (pssst.. I ordered one too) May 06 20:41:55 I was asking for the build quality of the Pandaboard ES: is it populated with good qualiity components? May 06 20:42:15 I saw it has very few (if any) capacitors May 06 20:42:17 * mranostay goes outside to get some of something called fresh air May 06 20:43:26 In case I'm sounding strange, it's my first venture from the full-blown pc world into the elegant ARM development boards, and I'd like to know what ground am I jumping into May 06 20:43:47 Knightmare_: i think the question is mostly pointless. nobody would probably say: "sorry,but the quality is crap." futhermore i see no relation between capacitor count and build quality, an for the count itself: the schematics are freely available, feel free to count them at your leisure. May 06 20:44:19 Knightmare_: are you writing an OS? May 06 20:44:41 Knightmare_: feel free to join us in #pandaboard of course if you want to poke that more. May 06 20:45:14 capacitors are a problem after some time: it's the main failure point in any logic board I've had May 06 20:45:25 that's why I prefer hardware with less caps :) May 06 20:45:40 Knightmare_: to be honest: sounds like nonsense. May 06 20:46:04 ask in any Amiga forum... May 06 20:46:17 yeah, that's BS May 06 20:46:19 Knightmare_: sounds like total buzzwording, unaware of the various types of capacitors that are there and their specific characteristics. May 06 20:46:45 there was a problem with a few bad batches of electrolytic capacitors May 06 20:46:51 yep May 06 20:46:59 Knightmare_: i *guess* you are just generalizing some bad experiences with electrolytic caps. May 06 20:47:17 LetoThe2nd: I admit I'm May 06 20:47:33 the ones that blew up May 06 20:47:47 Knightmare_: hint then: better stop it. :) May 06 20:47:48 well, in the Amiga boards they also "dry-up" May 06 20:48:08 ok, then May 06 20:48:27 well, let's go for another question May 06 20:48:41 Knightmare_: and stop relating everything to amigas. most people here know what elelctrolytic caps are and how they fail. as i said, feel free to check the panda schematics and count them. May 06 20:48:46 Knightmare_: sure, and that's like saying that after the Pentium FDIV bug, processors with less FLOPS are more reliable May 06 20:49:14 ok, ok, sorry about that May 06 20:49:32 forget about the capacitors problem May 06 20:49:51 as a bonus, capacitors fail pretty obviously May 06 20:50:07 it's my fault and ib fact it's what you call "fuzzwording" (it's the first time I red that words but it seems adequate) May 06 20:50:19 I have also a question: whats the fastest way to control a 12v motor via a beaglebone? May 06 20:50:28 mlbox: h-bridge May 06 20:50:33 mlbox, gpio May 06 20:50:55 L293D chip May 06 20:50:59 yeah, GPIO to an h-bridge IC May 06 20:52:08 as for the Pandaboard ES hardware: does it have fully open-source Linux drivers or are there "binary blobs" you've to deal with in Nvidia style? May 06 20:52:14 maybe unrelated question, but I don't find answer on google after lot of searching. I want to check if a buffer declared as char buffer[256] contains "123" in the first 3 characters but until now no succes May 06 20:52:22 Knightmare_: binary blob. May 06 20:52:52 using ansi C May 06 20:52:58 LetoThe2nd: and what abou the Beagleboard hardware? May 06 20:53:15 *about May 06 20:53:24 Knightmare_: at least for some parts of the video codec things. afaik the same is true for the beagle, but not sure there. May 06 20:53:33 thanx, googling.. May 06 20:53:55 thanks, Leto May 06 20:54:37 so, if someone can provide an answer for the Beagleboard.. I'd prefer fully open-source drivers May 06 20:54:47 educa: er... May 06 20:54:54 xmp, its ok May 06 20:54:57 stupid by me May 06 20:55:04 array of CHAR is ' ' and not " " May 06 20:55:08 educa: strncmp May 06 20:56:01 is that also ansi C and not only c++ ? May 06 20:56:52 ... yeah. it's in string.h May 06 20:58:03 ok, looking now, tnx May 06 20:58:38 might I suggest K&R, second edition May 06 20:58:55 ok, for what I see in google, it seems the main "closed source" part is the SGX 3d unit May 06 20:59:28 but back in 2010, TI expressed their interest in providing Open Source drivers May 06 20:59:39 K&R second edition = ? May 06 21:00:34 educa: http://lmgtfy.com/?q=K%26R%2C+second+edition May 06 21:01:23 it's the C book May 06 21:01:47 if you like open source, the rpi is not for you... May 06 21:02:07 hehe LOL May 06 21:02:14 lmgtfy :) May 06 21:02:20 thanks anyway, will check out May 06 21:02:39 mru: I don't get the near-religious open-source stuff May 06 21:02:44 me neither May 06 21:02:52 but I don't like the idea of a binary blob boot loader May 06 21:02:56 If something's polished and I like it, I don't care if it's open-source May 06 21:03:19 mru: yeah, if it's in ROM that can kinda suck May 06 21:03:26 no, not rom May 06 21:03:27 (at least it's not U-Boot, though :P) May 06 21:03:39 the rpi needs a gigantic binary blob to do _anything_ May 06 21:03:54 the beagle has a boot rom, which is nice May 06 21:03:56 that's the same I think: it's not religious but I prefer compilationg all my controllers from the linux kernel sources and not having to mess with binary blobs May 06 21:04:18 mru: yeah, and it's, you know, actually documented May 06 21:04:21 that's why I'm asking May 06 21:04:46 the beagle boot rom is not fully documented, not even within TI iiuc May 06 21:05:01 but everything you need to boot an OS is public May 06 21:05:32 mru: exactly May 06 21:05:52 stick the header on (for some devices), and the ROM jumps to your entry point May 06 21:07:08 mru & XMPPwocky: in a Mac Mini, I can also boot Linux with Open Source software only. But if I want accelerated 2d & 3d, I need a binary blob from Nvidia. Is this the same case? May 06 21:07:22 I know it's not the same computer: it's just an example May 06 21:07:44 yes, the gpu drivers are closed May 06 21:08:29 so they are released for a given kernel version May 06 21:08:34 <_av500_> no May 06 21:08:35 aren't they? May 06 21:08:41 <_av500_> no May 06 21:08:48 <_av500_> they do not depend on kernel May 06 21:08:51 they are released for a random kernel version May 06 21:09:09 mru: yeah, that's actually the main reason video cards are expensive May 06 21:09:24 nonsense May 06 21:09:29 nvidia has to have a big compile farm, constantly rebuilding their drivers until they get one with the right kernel version May 06 21:09:30 <_av500_> they are expensive because they waste a lot of silicon May 06 21:09:33 wake up sheeple May 06 21:09:33 goodnight everybody May 06 21:09:39 video cards cost exactly what the market is willing to pay for them May 06 21:10:13 <_av500_> mru: i NEED to see my own reflection in the back of the bullets from my BFG.... May 06 21:10:32 Nvidia drivers get "broken" with new kernels often May 06 21:10:34 enhance May 06 21:10:47 so I'm not talking nonsense I think May 06 21:10:49 <_av500_> and the back of the bullet reflected in my sunglasses May 06 21:10:56 I lived that in Gentoo several times May 06 21:11:07 Knightmare_: nvidia drivers have worked fine on every kernel I've put them on for many years May 06 21:11:36 10 years ago they could be a bit flaky May 06 21:11:54 but I don't recall having any problems after 2005 May 06 21:12:28 if you rebuild your system and compile an new kernel, then emerge a new Nvidia drivre, it will probably be allrigh. But if you just update your kernel and then reboot your system without updating the Nvidia drivers, they often fail May 06 21:12:39 *driver May 06 21:12:44 <_av500_> like any other module May 06 21:12:50 yes. of course May 06 21:12:53 <_av500_> so? May 06 21:13:03 <_av500_> linux has no stable binary API May 06 21:13:08 <_av500_> or ABI May 06 21:13:13 but Nvidia drivers have been incompatible with "newer" kernels among the years May 06 21:13:30 <_av500_> dont they have an open glue layer? May 06 21:13:31 and only the fact that nvidia releases new updated blobs keep my hardware working May 06 21:14:01 yes. they have May 06 21:14:15 <_av500_> so the glue breaks May 06 21:14:29 it's rebult agains the current kernel when you emerge the new drivers May 06 21:14:34 <_av500_> yes May 06 21:14:37 <_av500_> like any other module May 06 21:14:59 yes, but other modules are matching the kernel version they're shipped with May 06 21:15:06 so they always should work May 06 21:15:10 <_av500_> compiled with May 06 21:15:13 Knightmare_: it might just be re-linking it May 06 21:15:42 <_av500_> Knightmare_: nvidia does not ship kernels May 06 21:15:48 <_av500_> so it cannot ship matching modules May 06 21:15:54 ... I know Nvidia doesn't... May 06 21:15:57 oh well May 06 21:16:01 nvidia is not a problem May 06 21:16:29 their drivers work fine across a wide range of kernel versions May 06 21:16:53 their drivers don't support KMS May 06 21:17:04 who cares? May 06 21:17:05 is this still going on? :) May 06 21:17:25 their closed-source policy doesn't allow proped x-less 2d and/or 3d acceleration on the hardware I paid for May 06 21:17:26 they pick the native resolution of my monitors automatically May 06 21:17:26 etc May 06 21:17:28 that's all I need May 06 21:18:13 I'm curious about using the beaglebone as the brains for an advanced home monitoring/automation/security system. May 06 21:18:22 <_av500_> KeatonT: sure May 06 21:19:04 I have plenty of experience in a wide variety of languages and actually already has a custom home security system setup using an arduino. May 06 21:19:30 My eventual goal is to have all of the necessary code on a single device. May 06 21:19:37 ok, my last question: have you tried Risc OS on the Panda? How good is the port? Is it stable? May 06 21:19:49 <_av500_> no idea May 06 21:19:51 Basically why I'm looking at the beaglebone. May 06 21:19:52 KeatonT: using GPIO lines? May 06 21:20:15 <_av500_> Knightmare_: I doubt its enterprise quality.... May 06 21:20:17 mranostay yep. May 06 21:20:31 Basically just reading the values off the magnetic door and window sensors. May 06 21:20:38 (currently) May 06 21:20:49 _av500_: is the Beagleboard port more mature or stable? May 06 21:20:52 <_av500_> and I guess it works as good as on any other modern ARM SoC May 06 21:20:55 <_av500_> Knightmare_: no idea May 06 21:21:04 ok bbs from the coffee shop with okay coffee and kinda cute baristas May 06 21:21:13 ok, thanks May 06 21:21:13 <_av500_> its not really the OS that most people here talk about May 06 21:21:20 <_av500_> in fact, only few do May 06 21:21:25 <_av500_> or rather did May 06 21:21:52 most people here talk about the beer May 06 21:21:52 <_av500_> mranostay: POIDNH May 06 21:22:29 mranostay: as good as "my" barmaids? May 06 21:22:30 mru: too early May 06 21:23:17 Looking into the beaglebone it seems a good platform for what I'd like to do as long as a suitable way exists to work with the GPIO pins on the beaglebone. May 06 21:23:29 I just keep getting unclear information from internet searches. May 06 21:23:44 I have a beagleboard C4, where can i find the latest u-boot and mlo? May 06 21:24:25 denx.de? May 06 21:25:03 <_av500_> KeatonT: the pins can be used May 06 21:25:21 yea, looks like using the filesystem seems the easiest way. May 06 21:25:52 But I've heard mentions of other ways of interacting with the GPIOs May 06 21:26:38 <_av500_> yes, a kernel driver May 06 21:29:16 any suggestions? May 06 21:29:28 filesystem or kernel driver? May 06 21:30:00 <_av500_> depends May 06 21:30:45 on? May 06 21:30:55 things May 06 21:31:30 such as? May 06 21:31:38 phase of moon May 06 21:32:07 <_av500_> as always, the "use case" May 06 21:32:38 KeatonT: it should probably be compiled May 06 21:32:39 mru: always the dark side. May 06 21:32:40 <_av500_> detect a broken window, or detect it 1000x per second May 06 21:33:03 LetoThe2nd: where the moon nazis are May 06 21:33:06 Definitely not 1000x per second. May 06 21:33:13 I just want to detect state changes. May 06 21:33:52 mru: totally yes. May 06 21:34:02 Preferably with some type of interrupts or events May 06 21:34:17 LetoThe2nd: got to see that film when it comes out here May 06 21:34:19 <_av500_> jkridner__ knows all about gpio events May 06 21:34:43 hopefully the moon nazis won't get some beaglebones into their hands... May 06 21:34:55 mru: it is out here, but ATM i totally lack time for things like these. May 06 21:35:08 and when it does come out here, I'll be in germany... May 06 21:35:29 <_av500_> mru: comming to visit? May 06 21:35:39 _av500_: berlin May 06 21:35:51 you said you would not come there May 06 21:35:57 afraid of lydia or something? May 06 21:36:02 <_av500_> totally May 06 21:36:39 ah its linuxtag isnt it? May 06 21:36:48 <_av500_> yes May 06 21:36:52 KeatonT I'd try the easier way... May 06 21:36:53 that's the general idea May 06 21:37:05 and week after hong kong May 06 21:37:14 <_av500_> its an excuse to drink beer, the tag itself I find more and more useless May 06 21:37:24 certainly May 06 21:37:41 it's gone downhill since they moved it to berlin May 06 21:37:44 mlbox I think I May 06 21:37:49 I'd try the filesystem way first. May 06 21:37:57 Then move on to trying to get kernel support. May 06 21:38:09 I'd like if at all possible to prevent from running my code in a continual loop. May 06 21:38:13 KeatonT, so how many sensors etc. will you connect to it? May 06 21:38:19 13 May 06 21:38:37 so far but that could go up considerably depending on how I do the home automation. May 06 21:38:48 only sensors or also switches and other stuff May 06 21:38:57 <_av500_> the same May 06 21:39:20 ^^ May 06 21:39:34 ? May 06 21:39:49 The sensors are simple magnetic switches. May 06 21:41:37 <_av500_> AIUI you can poll() on gpios via sysfs May 06 21:42:13 Yea. May 06 21:42:26 All the specifics can be worked out. May 06 21:42:35 I'm just still debating on if I should buy one. May 06 21:42:47 I suppose it can always be used for other projects. May 06 21:43:31 <_av500_> and its just $99 May 06 21:44:15 * mranostay hates dollar bills May 06 21:44:18 sure.. I am currently using one as a Squeezebox server May 06 21:44:19 but I think it's a shame not to make use of the gpios May 06 21:45:05 <_av500_> The imx6 has been delayed until at least August due to needing another May 06 21:45:05 <_av500_> respin of the silicon. May 06 21:45:05 <_av500_> The imx6 has been delayed until at least August due to needing another May 06 21:45:05 <_av500_> respin of the silicon. May 06 21:45:08 <_av500_> oops May 06 21:45:09 anyone know how many GPIO pins can be used for detecting simple IO May 06 21:45:10 <_av500_> sorry May 06 21:45:16 <_av500_> was not meant for here May 06 21:47:02 Wonder if it would even be worth trying to install Debian* May 06 21:48:14 <_av500_> sure May 06 21:48:23 <_av500_> debain or any other linux distro May 06 21:48:50 <_av500_> all blink the cursor at the same speed May 06 21:48:57 I'm far more used to Debian but I'll give the default OS a try May 06 21:50:08 _av500 is it difficult to access the gpios under debian or ubuntu? May 06 21:51:21 mlbox filesystem wise I doubt it. May 06 21:51:22 lol May 06 21:51:33 kernel wise… maybeeeeee. May 06 21:51:56 mlbox: some would argue it is too easy May 06 21:52:52 lol mranostay I think I would agree with that statement. May 06 21:53:41 lucky those pins can source much current May 06 21:53:48 *can't May 06 21:53:50 <_av500_> mlbox: how hard do the two make it to call open() et al? May 06 21:54:40 _av500_: just wait until lennart replaces libc May 06 21:55:13 Seems simple enough to create a class to handle talking to the FS. May 06 21:55:24 <_av500_> mru: I wait for the day that systemd becomes selfaware.... May 06 21:55:41 * _av500_ sends jonmasters_ as message, just because he can May 06 21:55:54 hmm... my experiance with these things... have to read a bit more.. May 06 21:56:45 how does the system know that it has these gpio-pins? May 06 21:57:09 _av500_: systemd, systend, systenk, sysnetk, skysnet, skynet May 06 21:57:28 <_av500_> mru: skynet is just the decoy May 06 21:58:00 does jonmasters_ just lurk in here? May 06 21:58:33 <_av500_> apaprently mentioning him triggers a note to be printed and courier-delivered to him May 06 21:59:31 <_av500_> https://plus.google.com/u/0/106265217227408958782/posts/KnkzUM61B19 May 06 22:00:05 _av500_: memoserv? May 06 22:00:16 or just saying jonmasters_ spam him? :) May 06 22:01:08 * _av500_ has no idea if saying jonmasters_ will spam jonmasters_, so let's try that: jonmasters_ May 06 22:04:50 lol May 06 22:05:38 so no one has tried RiscOS on the Pandaboard? May 06 22:05:45 I mean, no one here May 06 22:05:54 <_av500_> Knightmare_: there were people May 06 22:05:58 <_av500_> in beagle or panda May 06 22:06:00 <_av500_> i forgot May 06 22:06:23 ok, I'll try in #panda May 06 22:07:16 <_av500_> http://www.riscosopen.org/content/sales/cmos-widgets May 06 22:07:56 <_av500_> in fact: 16:42 < av500 > wtf? http://www.riscosopen.org/content/sales/cmos-widgets May 06 22:10:06 _av500_: why? why? May 06 22:10:09 thanks, _av500_ May 06 22:10:20 it seems to be an officialy supported platform May 06 22:10:34 <_av500_> maybe its a scam page May 06 22:10:57 5 pounds for that? May 06 22:11:04 you must be mad May 06 22:11:09 <_av500_> mranostay: its for riscos people May 06 22:11:18 <_av500_> no risc, no OS May 06 22:11:38 scam? hahaha! that's a good one: who'd try to scam Risc OS users nowadays? It's also in the openriscos domain May 06 22:12:32 maybe riscos is the scam May 06 22:12:57 maybe the scam is the sacam! May 06 22:13:01 scam May 06 22:13:13 secam sure was a scam May 06 22:13:17 <_av500_> so much sacasm May 06 22:13:37 hey, you don't want to offend poor french people, do you? May 06 22:13:53 <_av500_> I do it every day May 06 22:14:04 <_av500_> and thta was sarkosm May 06 22:14:29 being mock-british is probably an offence to the french, so why bother trying to be nice? May 06 22:15:14 mru: being nice is so overrated :) May 06 22:15:26 because french people have the ability to invent unheard english insults, just like those castle guys in "Knights of the square table" May 06 22:15:38 mranostay: Nice is french, after all May 06 22:15:56 hey, people are here! May 06 22:16:00 omg! May 06 22:16:10 do you all show up about this time every week? May 06 22:16:19 only on fridays May 06 22:16:23 <_av500_> only when there is a super moon May 06 22:16:49 ozzloy: we're not people but scammers May 06 22:17:55 ozzloy: just trolls we barely count :) May 06 22:18:09 wtf is a super moon? May 06 22:18:43 mranostay: one of those things you have in the sky right now May 06 22:18:50 heh May 06 22:19:06 * mranostay needs to get out more May 06 22:19:09 some friends and i are working with a beagle May 06 22:19:53 mranostay: forgetting about the moon is probably a sign of that, yes May 06 22:20:52 <_av500_> mranostay: dont look into that other thing in the sky May 06 22:21:03 _av500_: what other thing? May 06 22:22:35 <_av500_> mru: the SDI laser May 06 22:23:46 I've heard of a pie in the sky May 06 22:23:53 is it a raspberry pie? May 06 22:23:55 Hey I guess if things don't work out with using the GPIO pins on the beaglebone I could always connect an arduino up to the USB host and get sensor inputs from serial connections. May 06 22:23:58 lol May 06 22:24:04 <_av500_> sure May 06 22:24:14 <_av500_> you could have the arduino morse it May 06 22:24:16 rpi sure does seem like pie in the sky to me May 06 22:24:17 <_av500_> and detect that May 06 22:25:23 yea many many ways. May 06 22:25:38 I can't see why using the GPIO on the beaglebone would be so hard I'd have to do that. May 06 22:25:44 (Hopefully) May 06 22:26:39 <_av500_> its not hard at all May 06 22:28:34 Rather code in C/C++ but python wouldn't be too bad. May 06 22:32:26 mru: i rather buy pie than an rpi :) May 06 22:33:18 <_av500_> I'd rather buy beer May 06 22:33:42 ok good point May 06 22:34:57 _av500_: agreed May 06 22:47:48 prpplague: _av500_: do i see a #beagle pub crawl in the future? May 06 22:48:07 mranostay: hehe, we have them regularly at the conferences May 06 22:48:23 <_av500_> prpplague outcrawls everybody May 06 22:49:06 _av500_: i have to be great at something...... May 06 22:49:20 alcoholism it is :P May 06 22:50:11 mranostay: :) May 06 22:54:40 i think we all should message jonmasters_ May 06 22:57:27 <_av500_> even jonmasters_ should message jonmasters_ May 06 22:59:27 isn't that a sign of mental illness? May 06 23:24:34 anyone in here used a beagle with a kinect? May 06 23:54:52 lol May 06 23:55:10 _av500_: look, it worked! May 06 23:55:27 o_O May 07 00:05:33 jonmasters_: so you get a flurry of texts? :) May 07 00:07:14 it's rate limited :) May 07 00:07:30 I have an algorithm for that May 07 00:08:04 let's test it: jonmasters_ jonmasters_ jonmasters_ jonmasters_ jonmasters_ jonmasters_ May 07 00:08:29 doesn't work like that :) May 07 00:08:49 I have to be idle, and it'll screen on nick, and lots of other things. I call it "antiping" May 07 00:09:01 it's designed to train people never to do: "jonmasters: ping?" May 07 00:09:14 instead, it's to train them to do "jonmasters: here is exactly what I need from you" May 07 00:09:52 in that it will, under certain circumstances where you do ping me, reply to you with a polite explanation that "ping" is uselessly wasting my time...except it won't use quite those words May 07 00:10:19 jonmasters_: I need you to pack all the money in a brown paper bag and place it to the _left_ of the bench by the bus stop at exactly 4:13pm tomorrow May 07 00:11:01 well yes, that would be appropriate, if we were engaged in international espionage. But let's talk about that after the white dove flies over the moon's pyjamas. May 07 00:11:11 ok? May 07 00:11:54 lol May 07 00:12:05 sounds like a nice away script. May 07 00:13:17 mru: reminds me of my 'friends' that like to send me texts like 'all praise to allah!' when i'm flying May 07 00:13:39 that's a good idea, I should try that some time May 07 00:14:18 I'll remember not to log into IRC while using the in-flight wifi. May 07 00:14:19 :P May 07 00:14:22 all fun and games till someone gets strip searched May 07 00:14:43 strip searches can be fun and games too May 07 00:15:20 are these the kind where stripper and strippee are distinct persons? May 07 00:15:45 heh yes May 07 00:16:08 which can be fun in some cases as well May 07 00:16:14 oh yes May 07 00:25:15 * mru attempts to build latest gcc 4.8 snapshot May 07 00:27:30 ICE May 07 00:32:14 asdf May 07 00:32:24 lag May 07 00:34:09 jonmasters_, here is exactly what i need from you: tips and tricks for working with the kinect and beagle{bone,board} May 07 00:37:24 is the beaglebone rev3 the same as rev5 in terms of hardware? May 07 00:37:36 there was a bug fix with an ethernet LED May 07 00:37:43 but other than that, they're identical May 07 00:37:53 i thought that was rev4 May 07 00:38:16 or do you mean a software bug fix? May 07 00:38:24 hm, not sure May 07 00:38:36 i know 4 broke something badly and 5 fixed it May 07 00:38:37 so the rev3 might not blink the ethernet led? May 07 00:38:47 i can live with that May 07 00:39:33 XMPPwocky, i just put together your name and rolled my eyes at myself for taking so long to put together your name May 07 00:39:40 hahaha, yeah May 07 00:40:13 * XMPPwocky resets the "days since someone's gotten his name" sign May 07 01:12:39 :-O May 07 02:11:47 where are the analog in/digital out pins in /dev (or /sys) May 07 02:11:52 and how do I read/set them? May 07 02:11:56 assuming there's a howto somewhere May 07 02:18:45 http://www.elinux.org/EBC_Exercise_02_Flashing_an_LED **** ENDING LOGGING AT Mon May 07 02:59:59 2012