**** BEGIN LOGGING AT Fri Apr 27 03:00:00 2012 Apr 27 03:57:56 hi there, has anyone had problems with the BeagleBone transmitting serial data at > 115.2k baud? Apr 27 03:58:21 It seems to receive OK at higher speeds (500k) but not transmit. Apr 27 03:58:44 I have scoped the signal and it seems to be OK, however it cannot be successfully read by my PC Apr 27 04:04:39 never seen a uart run at more than 115k Apr 27 04:05:31 some usb-dongles are able to do more Apr 27 04:06:03 serial is 12v right? lots of reflection afaik Apr 27 04:08:06 * dm8tbr regularly uses OMAP3 with 1MBaud Apr 27 04:08:28 jay6981, you're thinking of rs-232 most likely at that level Apr 27 04:08:41 serial doesn't mean rs-232 Apr 27 04:08:56 usb dongles are rs-232 right? Apr 27 04:09:25 only at the DTE end Apr 27 04:09:45 Dan211: are you sure the pc-side is able to do more? Apr 27 04:09:57 he said he was talking to a PC so I figured 232 Apr 27 04:10:26 for example, beaglebone uarts are 3.3v i/o… Apr 27 04:10:56 jay6981: hrm, most pcs don't have rs-232 ports now though Apr 27 04:11:22 the problem beyond 115k2 is that some chipsets don't cope well with them Apr 27 04:11:25 12v to -12v is a "typical" voltage swing for rs232 Apr 27 04:11:36 but it could be different depending on hardware Apr 27 04:11:37 with beaglebone's ftdi…the serial comm never is translated into rs-232 for transmission to the pc Apr 27 04:11:38 hm, TIL Apr 27 04:11:54 Russ, up to +-15V per spec, in fact, iirc Apr 27 04:12:15 hmm, the ftdi should be able to do more Apr 27 04:12:20 e.g. pl2303 won't do 500 and 1000k but something 'close' to that Apr 27 04:12:41 I had forgotten the bone has an ftdi on board Apr 27 04:12:53 usb2serial ftdi I have regularly used with 1MBaud Apr 27 04:29:22 hey guys sorry for the slow reply. Some additional information - I am trying to communicate between an Arduino and BeagleBone using serial. I am able to send data Arduino -> Beagle at 500kbps but unable to go above 115.2k the other way. If I connect tx to rx on the Arduino, it can handle 500kbps, likewise with the Beagle but not when connected together. I am doing some level conversion to account for 3.3V <-> 5V Apr 27 04:30:12 5V -> 3.3V uses a voltage divider, 3.3V to 5V uses a transistor but should work without (and does at <= 115.2k) Apr 27 04:31:09 some unimportant information ;) Apr 27 04:31:27 additionally, I cannot communicate with my PC at > 115.2k (using ttl -> RS232 converter) Apr 27 04:31:50 But my Arduino can communicate with my PC at > 115.2k Apr 27 04:32:16 furthermore, it looks like someone else is having this problem - https://groups.google.com/forum/?fromgroups#!topic/beagleboard/8MspWJgfFAI Apr 27 04:33:45 Dan211: which converter? the one on the board itself? Apr 27 04:34:53 mh, two things I wonder: a) is the level shifter fast enough / does it distort the signal b) does the UART timing differ somehow on both sides Apr 27 04:35:17 some people have a very short attention span Apr 27 04:36:07 even without the conversion from 3.3 -> 5V it should still work but it doesn't (only above 115.2k) Apr 27 04:36:15 I have also scoped the signal and it doesn Apr 27 04:36:18 * Doesn Apr 27 04:36:25 *doesn't look distorted Apr 27 04:37:00 I think it is most likely a timing issue or a bug in the UART configuration? Apr 27 04:42:46 UPDATE: It now looks as if I can get it working at 500kbps for Beagle -> Arduino if I put a slight delay between writing characters and I change the data bits to 7, i.e. 7N1???? Apr 27 04:45:16 does anyone know how to access the low level microcontroller registers from the standard Angstrom distribution? Apr 27 04:53:15 you do it through a program called linux Apr 27 04:54:08 look at the omap serial driver Apr 27 05:06:48 Dan211, do you have a ground wire? Apr 27 05:07:02 Yip Apr 27 05:07:07 ground? we don't need no stinking ground! Apr 27 05:07:13 everything levitates here! Apr 27 05:07:26 the same setup is able to communicate at <= 115.2k Apr 27 05:08:47 Dan211: you might try to raise the baudrate of the console (using uEnv.txt) Apr 27 05:09:25 Ok I will try that aholler, thank you! Apr 27 05:10:14 so you could be sure it isn't your cabling or whatever because the onboard-ftdi is likely wired correctly Apr 27 05:11:46 a good symptom of not having a good return path is 'I measured it and it's fine' Apr 27 05:14:21 we have a german saying which is "Wer misst misst Mist". That translates to "who measures measures bullshit". Apr 27 05:14:34 what a horrible saying Apr 27 05:14:44 its like the asian saying that the nail that sticks up gets hammered down Apr 27 05:16:06 the saying proved too often to be sure ;) Apr 27 05:16:21 you can't get very far with circuitry without measuring :) Apr 27 05:17:30 aholler, uEnv.txt only seems to contain "optargs=run_hardware_tests quiet" Apr 27 05:20:09 was that a mao quote? Apr 27 05:20:17 you can add "console=ttyO2,115200n8" Apr 27 05:21:05 or use another baudrate. I'm not sure what exactly the fdti does support Apr 27 05:21:24 but I know he can more than 500k Apr 27 05:21:40 ok thanks, Linux is still quite new for me Apr 27 05:22:03 cat /proc/cmdline to verify what you have done Apr 27 05:22:36 (and be prepared to use ssh over ethernet) ;) Apr 27 05:24:49 i have to say it's just theory, I'm not sure if the console-driver as the baudrates hardwired. never tried it myself ;) Apr 27 05:24:57 s/as/has/ Apr 27 05:26:49 and I don't know if ttyO2 is used on the bone. I don't have one. Apr 27 05:28:28 full disclosure there :) Apr 27 05:28:48 I changed it to ttyO0 since uart0 is used for the console on the bone. It did not work unfortunately (still communicated at 115.2k) Apr 27 05:29:57 how do you know? Apr 27 05:30:42 how do I know it uses uart0 or how do I know it still communicated at 115.2k? Apr 27 05:30:59 115.2k Apr 27 05:31:49 I was able to see the boot screen on my PC terminal using 115.2k Apr 27 05:32:07 at 500k, it did not work (I had set uEnv.txt to 500k) Apr 27 05:33:24 cat /proc/cmdline and have a look at dmesg Apr 27 05:34:21 Dan211: try 576000 ? Apr 27 05:34:25 a more standard rate Apr 27 05:36:29 the console baud has not changed from 115200 Apr 27 05:36:30 * mranostay waves and yawns Apr 27 05:40:33 Documentation/serial-console.txt says "The maximum baudrate is 115200." Apr 27 05:40:59 so there is the limit (for the console). Apr 27 05:41:31 but if you have a ssh-connection, you can just use a terminal programm like minicom Apr 27 05:41:31 you might be able to push it higher in non-obvious non-straight-forward ways though :) Apr 27 05:41:46 dm8tbr: why? :-/ Apr 27 05:41:55 * dm8tbr has used kernel console at 1MBaud Apr 27 05:42:04 see extrabaud option or what it was Apr 27 05:44:26 dm8tbr: yeah i know it can be done but are you using xmodem with it? Apr 27 05:48:22 mranostay: the origin for using 1MBaud is in AVLO, dunno if av500 still calls it that way, but basically serial kernel upload, then console at same speed so no switching Apr 27 05:48:35 <_av500_> AWOL Apr 27 06:09:59 * _tasslehoff_ waits for AMOL Apr 27 06:11:11 * Russ waits for 2 Apr 27 06:11:35 what's amol? Apr 27 06:14:25 ah, wheel of time Apr 27 06:15:13 thanks for everyones help! Apr 27 06:15:40 was it just that the baud rate was set wrong? Apr 27 06:27:27 <_av500_> dm8tbr: no need for extrabaud any more Apr 27 06:27:40 <_av500_> just set console=1000000 etc.. Apr 27 06:27:53 ah, nice :) Apr 27 06:35:32 Hi, I am trying to setup my ubuntu(11.10) machine for some qt development for BB. while trying to install tslib, i got this error. Apr 27 06:35:32 libtool: install: error: cannot install `linear.la' to a directory not ending in /home/administrator/apps/tslib_arm/lib/ts/ Apr 27 06:35:32 can anybody help me? Apr 27 07:48:15 I have trouble build a (loadable) kernel module. I have a Ångström image and download the kernel source as the kernel-dev and kernel-headers packages didn't installed anything. I get a "Invalid module format" when I try to insmod it. Apr 27 07:51:18 module version: "3.2.14 mod_unload modversions ARMv7 thumb2 p2v8", uname: "Linux beaglebone3 3.2.14 #1 Sat Apr 21 10:00:05 CEST 2012 armv7l GNU/Linux" error reported in dmesg: "no symbol version for module_layout " Apr 27 09:19:47 Hi, when programming the beaglebone from CSS in Win7 I often get into a state which requires to power down the beaglebone and connect to target again. I wonder if my system will turn stable if I somehow get to completely remove Linux from the SD card? If so how would I do this? Apr 27 09:25:35 by just removing the card? Apr 27 09:26:26 but I think writting error-free code is a better approach Apr 27 09:27:02 pfft, thats boring ... and makes all the bug triagers jobless Apr 27 09:27:14 Hi there, is this tutorial http://elinux.org/BeagleBoardPinMux is up to date concerning u-boot muxing? Apr 27 09:42:18 http://olimex.wordpress.com/2012/04/24/cortex-a8-in-tqfp-sure-allwinner-a13/ Apr 27 09:45:33 isn't the allwinner a10 better than the a13? Apr 27 09:45:39 and what a silly name Apr 27 09:45:59 a13 is a cost down a10 Apr 27 09:46:04 probably looks nice in chinese letters Apr 27 09:46:23 and calling an arm chip a10/a13 is just asking for confusion Apr 27 09:46:31 like A4 and A5 Apr 27 09:46:35 is this an a8 or an a10? Apr 27 09:46:42 and what's inside the a5? Apr 27 09:46:48 A9 Apr 27 09:46:52 iKnow Apr 27 09:47:00 9 is 5 and 10 is 8 Apr 27 09:47:02 iKuK Apr 27 09:47:07 no, 5 is 9 Apr 27 09:47:11 5of9 Apr 27 09:47:21 * mru demands a 7 Apr 27 09:47:27 oh right, there is a7 Apr 27 09:47:28 arm7 Apr 27 09:48:07 mru: hey, add our product names and you get a5 and a7 as well Apr 27 09:48:16 and A9 Apr 27 09:48:32 and a10_1 Apr 27 09:48:39 yup Apr 27 09:49:52 hey, I don't have an a7! Apr 27 09:50:08 and no documentation => useless Apr 27 09:50:10 what do I need to optimise to get one? Apr 27 09:51:03 osk olimex, they have a sponsor program Apr 27 09:51:43 http://olimex.com/dev/projects.html Apr 27 09:52:53 aholler: no documentation is pretty standard Apr 27 09:53:05 and even the docs you could get are in fact "none" Apr 27 09:53:27 at least A10 kernel is open source already Apr 27 09:58:46 Aholler: Tried to remove the SD card but when loading the out file onto the board I get errors: (CortxA8: File Loader: Data verification failed at address 0x80000000 Please verify target memory and memory map.;CortxA8: Unable to terminate memory download: NULL buffer pointer at 0x3a9f; CortxA8: GEL: File: C:\ti\AM335X_StarterWare_02_00_00_06\binary\armv7a\cgt_ccs\am335x\beaglebone\gpio\gpioLEDBlink.out: a data verification error occu Apr 27 10:00:06 vigoo: sorry, I have no knowledge about css Apr 27 10:00:33 hehe! ok! Apr 27 10:02:51 but maybe following what the error message says will help Apr 27 10:03:23 I'm trying to compile u-boot, but I can't find the good sources for an Beagleboard xM c1. Any clues? Apr 27 10:03:42 google meta-ti :-D Apr 27 10:04:17 lpi: mainline Apr 27 10:05:08 so denx.de Apr 27 10:05:44 http://www.pineriver.cn/eshowProDetail.asp?ProID=1531 Apr 27 10:05:47 a10 inside Apr 27 10:07:07 aholler: the good git is git://git.denx.de/u-boot.git? Apr 27 10:07:53 reads like that is the mainline one Apr 27 10:08:11 ok thanks :D Apr 27 10:08:39 I used an outdated one I guess: http://gitorious.org/beagleboard-validation/u-boot Apr 27 10:10:41 koen: if you need a build server for angstrom: http://www.hetzner.de/en/hosting/produkte_rootserver/ex10 Apr 27 10:59:27 note to all: it's friday so act like it Apr 27 10:59:58 I have doubt about Friday Apr 27 11:02:18 please tell me the exact steps Apr 27 11:02:31 sir! Apr 27 11:02:39 is that a homework assignment ? Apr 27 11:06:04 can we go to PM? Apr 27 11:07:06 only if you enable smartreflex Apr 27 11:07:25 I want something more proactive Apr 27 11:07:50 I have trouble when compiling u-boot, I have "arm-ctng-linux-gnueabi-gcc not found" error when I'm entering "make CROSS_COMPILE=arm-none-linux-gnueabi- ". According to http://elinux.org/BeagleBoard#U-Boot Apr 27 11:09:21 read it again Apr 27 11:11:03 ctng? Apr 27 11:11:05 Done. I guess the arm compiler isn't correctly installed. But I copied all files contained in the toolchain provided by narcissus online builder Apr 27 11:11:11 compiler tools for the next generatnion? Apr 27 11:12:29 don't know :( Apr 27 11:13:32 the toolchain tar.gz provided by narcissus builder isn't enough? Apr 27 11:14:52 av500, that's a toolchain I can embrace Apr 27 11:21:45 doesn't help... Apr 27 11:22:35 \me is going out for lunch Apr 27 11:22:43 * lpi is going out for lunch Apr 27 11:22:47 try CROSS_COMPILE=arm-none-linux-gnueabi- make Apr 27 11:24:53 * mdp wonders which channel has the steps to make this crypto driver work Apr 27 11:25:06 mdp: but its a "free" toolchain which according to you cannot be good :) Apr 27 11:25:08 the exact steps? Apr 27 11:25:15 ah, yes, I wanted to try that too Apr 27 11:25:25 av500, "Freedom!" Apr 27 11:25:36 mdp: you need a special openssl to test it, the angstrom one should work with that ocf abomination Apr 27 11:26:28 no,this is just the crypto api stuff with the test module Apr 27 11:26:58 supposedly there's some forward port issue Apr 27 11:27:09 I'm waiting for the exact steps from my colleague Apr 27 11:27:35 executing holding pattern until peoples arrive at the mothership Apr 27 11:29:25 koen, I did hear about that stuff, but I believe that's for the legacy parts Apr 27 11:29:43 write a(nother) book about patterns Apr 27 11:30:30 av500, free beer is acceptable :P Apr 27 11:30:34 mdp: it would be nice to dump that ocf stuff and use the in-kernel crypto stuff Apr 27 11:30:47 koen, that's what this driver is Apr 27 11:30:53 sweet Apr 27 11:31:20 he's supposed to "mainline it" Apr 27 11:31:34 on forum? Apr 27 11:31:39 *ahem* Apr 27 11:31:44 that's ph0rum to you Apr 27 11:32:49 can't have forum with rum you know Apr 27 11:33:02 it can't go mainline until dmaengine edma driver is written…another blocker Apr 27 11:33:07 is foo-rum better than bar-rum? Apr 27 11:33:19 nobody likes to hear that stuff when their task is to mainline just that driver Apr 27 11:33:30 use PIO then Apr 27 11:33:37 yeah Apr 27 11:33:40 "fridays patterns" Apr 27 11:33:53 nobody likes to submit anything but a completely "done" driver either Apr 27 11:34:01 * av500 patents "Friday" Apr 27 11:34:01 lalalala Apr 27 11:34:18 * koen wants to go back to sleep Apr 27 11:34:27 * mdp circles Friday in a patented Friday holding pattern Apr 27 11:34:29 it's 13.34, so it's almost weekend! Apr 27 11:34:38 mmmmm 13.34 Apr 27 11:35:18 anything but 1664... Apr 27 11:35:24 LOL Apr 27 11:36:42 aka: the budweiser of .fr Apr 27 11:39:02 our hotel bar in velizy once ran out of real beer, leaving only the option of 1664…had to call it a night Apr 27 11:39:46 mdp: next time you are in velizy, call ahead! Apr 27 11:39:57 i'll bring beer from .de along! Apr 27 11:40:03 mmmm Apr 27 11:40:20 that'll educate you higher! Apr 27 11:40:36 tbh, I hope not to have to go back to velizy :) Apr 27 11:40:48 altenmünster Apr 27 11:40:57 * av500 calls his higher up TI contacts Apr 27 11:41:02 aholler: aw Apr 27 11:41:07 av500, edumucate at a higher level Apr 27 11:41:15 (mind you, not all german beer is good ... ) Apr 27 11:41:17 * av500 goes to get a Worscht Apr 27 11:41:20 av500, oh and thanks for the "final troll" on g+ Apr 27 11:41:39 mdp: your pass was perfect, I had to score Apr 27 11:42:55 av500, go team! I exist but to assist Apr 27 11:45:28 "higher ups"..hehe Apr 27 11:46:21 are those the same ones monitoring our illicit use of github? Apr 27 11:48:28 av500, I'm thankful we have easy access to quality domestic beer here now. it used to be that you had to go to a couple import stores Apr 27 11:48:46 now the problem is sifting through the zillion breweries Apr 27 11:49:16 I love such a problems Apr 27 11:50:21 yes Apr 27 11:50:48 mdp: where is "here"? Apr 27 11:50:58 anywhere in .us now Apr 27 11:51:18 versus 20 years ago Apr 27 11:51:44 yeah, there are loads of nice microbrews Apr 27 11:52:17 20 years ago I was not allowed to drink anyway... Apr 27 11:52:31 I think many people have discovered the fun of what some of us have enjoyed with traveling .eu…hitting those regional beers at pubs Apr 27 11:55:32 plus it's way cheaper to get a style of beer you like…the cost of say, a Reissdorf Kolsch is way high here and there's some outstanding micro kolschs being made here Apr 27 11:56:44 mru, I am old :) Apr 27 11:57:34 * mdp notes that we are doing pretty good with lack of topic and friday Apr 27 11:58:54 koen, fwiw, I'm sure this driver will end up molding in your tree for a year while mainline gets sorted out Apr 27 11:58:55 we need old people Apr 27 11:59:01 to make fun of Apr 27 11:59:15 indeed Apr 27 12:01:05 you guys have 1s and 0s…when I was a kid we only had 0s. Apr 27 12:01:32 kölsch is just good to rinse Apr 27 12:01:44 yep Apr 27 12:05:14 cheese eating surrender monkeys Apr 27 12:05:18 LOL Apr 27 12:05:44 * mdp successfully baits the private bashing to this channel Apr 27 12:06:41 http://www.youtube.com/watch?v=4rJAw-fuYHk Apr 27 12:07:59 :) Apr 27 12:08:10 koen->meanie++ Apr 27 12:08:19 lol Apr 27 12:09:23 aholler: I got System not configured - see README make: *** [all] Erreur 1 Apr 27 12:13:58 panto, imagine the response lag during august :) Apr 27 12:14:34 oh god NO! Apr 27 12:14:40 * panto starts saving like mad Apr 27 12:15:27 it's a good time for you to take off and enroll in mime school Apr 27 12:26:03 mdp: is that where you learn to send email attachments? Apr 27 12:27:27 mru: imagine all these mime types would actually talk to you... Apr 27 12:27:44 constant murmuring in the inbox... Apr 27 12:29:27 heh Apr 27 12:29:39 av500, mimes don't talk :) Apr 27 12:30:10 they could make gestures about trying to get out of your (in)box. Apr 27 12:30:30 panto: well, they sould not be mime types then of course Apr 27 12:30:36 would* Apr 27 12:30:40 * koen wonders what a multipart mime looks like Apr 27 12:30:48 splatter... Apr 27 12:30:59 koen: it seems TI is supporting the SDK right on the beagle ML! Apr 27 12:31:06 koen: can we sue them? Apr 27 12:32:13 anybody can file a lawsuit…easy Apr 27 12:38:52 av500, which list? Apr 27 12:39:47 av500: I don't think they really *support* it Apr 27 12:40:03 av500: if they do, I'd urge people to ask them for 1GHz support on xM :) Apr 27 12:40:26 koen, only support what is supported Apr 27 12:40:48 and supportable Apr 27 12:40:54 mdp: it's the other side of the coin Apr 27 12:41:02 supportify? Apr 27 12:41:11 mdp: TI pretended for a long, long time that 1ghz mode worked and was supported Apr 27 12:41:26 the upcoming SDK will officially not support it :) Apr 27 12:41:45 but I suspect TI wants to have it both ways: claim it works and not support it Apr 27 12:41:52 of course Apr 27 12:42:21 there's lots of stuff on the datasheet that's not supported in software Apr 27 12:42:41 some not even in hardware Apr 27 12:42:49 absolutely Apr 27 12:42:53 er, wait! Apr 27 12:43:04 those must be typos Apr 27 12:43:19 or your higher education reading comprehension failed you :) Apr 27 12:43:49 mdp: my education was free, so it could not have been good.... Apr 27 12:43:58 "this diagram has pins labeled 'TSIF', but the datasheet makes no mention of the transport stream interface" Apr 27 12:44:35 * av500 sends black helicopter after koen Apr 27 12:44:50 av500, not even sure how you can hope to draw a logical conclusion after your free education Apr 27 12:45:01 mdp: I use google Apr 27 12:45:07 wait, thats free too.... Apr 27 12:45:12 "free" Apr 27 12:46:16 I wonder what's wrong with all the drive-by people on this channel that didn't learn to use google for free Apr 27 12:46:39 sometimes they need baidu Apr 27 12:47:03 there's a whole subculture on Ask Jeeves Apr 27 12:47:25 ....Then we turned it on, and watched with glee as lines of computer code scrolled up the screen .... Apr 27 12:47:49 koen, I can see that Apr 27 12:48:18 av500, is this an R-Pi review? Apr 27 12:48:55 "We were overcome with giddiness as the device filled us with hope . . ." Apr 27 12:49:11 that, or somebody watched a dot matrix printer at work... Apr 27 12:49:26 mmm Apr 27 12:50:05 omg, run, the brits are building a rocket Apr 27 12:50:43 going to attack the iron sky moon base? Apr 27 12:52:56 or putting Lord British into orbit permanently? Apr 27 12:53:35 http://news.bbcimg.co.uk/media/images/59872000/jpg/_59872925__52929340_a23e7a2b-3f0c-48a9-b66a-f2c599c2bd85-1.jpg Apr 27 12:53:45 looks like an attack craft to me Apr 27 12:53:52 * mdp couldn't find google to look it up Apr 27 12:54:39 what are the difference from linux kernel and linux omap kernel? Apr 27 12:55:25 omap stuff might reach mainline one day Apr 27 12:56:49 av500: i see omap kernel is 2.6.x now there but the last is 3.3... Apr 27 12:57:05 but can i use a normal linux 3.3 with beagleboard? or i have to apply patch? Apr 27 12:57:28 it depends Apr 27 12:57:53 mdp: from whats? Apr 27 12:58:09 your expectations of functionality Apr 27 12:58:40 it boots…if you need some other functionality YMMV Apr 27 12:59:05 I don't always patch my kernels, but when I do... Apr 27 12:59:17 mdp: what i need is a simply way to build a kernel image to run with my beagleboard xM with qt powervr and qws. I try to build all packed manually but its a very wasting time...now i try buildroot Apr 27 12:59:50 well, "simply" is one constraint too many Apr 27 13:00:08 if you want qt and pvr, use angstrom Apr 27 13:00:34 cheap, fast, good, pick any 2 Apr 27 13:01:17 i try angstrom...but i need a tlib...and this install only on 3.x kernel and i have 2.6.x Apr 27 13:01:35 touchscreensupport* Apr 27 13:01:46 you "have"? Apr 27 13:02:05 i download angstrom and this have 2.6.x kernel Apr 27 13:02:34 angstrom has 3.x as well Apr 27 13:02:55 but i don't know how to upgrade kernel...i try opkg update upgrade...but nothing Apr 27 13:02:59 http://www.angstrom-distribution.org/kernel-30-hx4700 Apr 27 13:03:57 yes...but if you try online building...you select 2011.04 package and this have 2.6.x kernel Apr 27 13:04:14 so maybe 2011.04 is not the most recent one Apr 27 13:04:35 but is the only package you can use if you select beagleboard Apr 27 13:05:17 koen, didn't you tell me the narcissus feeds were a bit out of date due to the transition? Apr 27 13:05:51 oe-core package name changes, etc. etc. Apr 27 13:06:07 he fed the trolls instead Apr 27 13:08:08 ok that i'm new...but its incredible that for running a qt project with dbus pvr opengl phonon gstreamer i have to download qt with pacthec for pvr, plus download every single packaging...build...ecc.. Apr 27 13:08:18 packages* Apr 27 13:08:20 mdp: narcissus doesn't expose all the new options Apr 27 13:08:35 stuk_gen: you don't Apr 27 13:08:54 stuk_gen: you can use an angstrom demo release of use the angstrom buildsystem to automate the buid Apr 27 13:08:56 koen: this is the reason that i try buildroot... Apr 27 13:09:47 koen: angstrom buildsystem have a something like xconfig to select package kernel version ecc? Apr 27 13:10:03 please document your findings on elinux when you see how buildroot works Apr 27 13:10:14 no need Apr 27 13:10:34 av500: what's you mean? Apr 27 13:11:07 stuk_gen: I was not replying to you Apr 27 13:11:14 av500: ok Apr 27 13:12:22 then the next person can baidu his buildroot success from the documentation on elinux Apr 27 13:13:25 mdp: is there a people that use buildroot on beagleboard xM Apr 27 13:14:27 I don't know Apr 27 13:14:58 * av500 uses buildroot in production on omap3+4 products Apr 27 13:14:58 try it and find out Apr 27 13:15:16 stuk_gen: do as koen said, install latest BB demo images Apr 27 13:15:23 than add the packages you need Apr 27 13:16:25 av500: ok...i need build angstrom because narcissus is out date Apr 27 13:16:51 can you read? Apr 27 13:17:22 what i dont understand is what kernel install this command 'bitbake virtual/kernel' Apr 27 13:18:19 it depends Apr 27 13:18:19 the Mechanical Turk API comes to mind now Apr 27 13:19:26 av500: ok... how can i specific the kernel ? Apr 27 13:19:36 why do you need to specify the kernel? Apr 27 13:20:16 beacuse i need a kernel >3.0.x Apr 27 13:20:21 not a 2.6.x Apr 27 13:20:30 15:19 < koen> why do you need to specify the kernel? Apr 27 13:20:44 -.- Apr 27 13:21:00 http://www.igloocommunity.org/support/File:Tizen_on_Snowball2.jpg Apr 27 13:21:11 have you looked at what kernel it will build? Apr 27 13:21:22 If no, why don't you try that first? Apr 27 13:21:30 koen: what kernel does it build? Apr 27 13:22:27 because i don't want a surprise...if after much hours i have a kernel < 3.x... Apr 27 13:22:40 av500, although nice, that handset display is too unwieldy to carry in my pocket Apr 27 13:23:11 but if you can says 'don't worry the bitbake use the latest kernel testing and this is sure >3.x' i will trying... Apr 27 13:23:23 stuk_gen: you could also have a peek inside the kernel recipe Apr 27 13:25:53 ok...so the ways are 2... or use buildroot that download build and create image for package that i need... ore use angstomm and install after on the targer via opkg the package that i need Apr 27 13:27:08 ...last question...how many times and size require the build of angstrom? Apr 27 13:27:36 you dont need to build angstrom Apr 27 13:27:45 you can just install the packages you need Apr 27 13:27:50 -.- Apr 27 13:27:54 I CAN'T!!!! Apr 27 13:27:56 once if you do it right the first time Apr 27 13:28:13 1) install latest demo image Apr 27 13:28:19 2) make you have network Apr 27 13:28:23 3) opkg update Apr 27 13:28:27 4) opkg install Apr 27 13:28:33 make sure* Apr 27 13:28:59 av500: hoe can you get latest angstrom demo? via narcissus or other? Apr 27 13:29:01 then, if you need a new kernel, yes clone the git and bitbake a kernel Apr 27 13:29:02 how* Apr 27 13:29:24 https://www.google.com/search?q=angstrom+demo Apr 27 13:29:42 av500: let there be network? Apr 27 13:30:13 koen: wtf happened to that pge? Apr 27 13:30:15 page Apr 27 13:30:19 where are the file dates? Apr 27 13:30:56 it's service by nginx now instead of apache, so we lost the fancy dirindex Apr 27 13:30:56 koen: also, there used to be 2012 stuff, no? Apr 27 13:31:19 uimage, size 0 Apr 27 13:31:22 the stuff with v2013 in the name? Apr 27 13:31:25 where inspiring Apr 27 13:31:35 yes, symlinks have no size Apr 27 13:31:54 koen: MMe Michu does not know that Apr 27 13:31:56 or care Apr 27 13:33:15 madame michu likely doesn't care about the seperate files Apr 27 13:33:20 but yes, fixing that is on the list Apr 27 13:34:06 koen: you mean she would use the ancient sd-images from 2010? Apr 27 13:34:19 why are they even on that page? Apr 27 13:35:00 ah, nicer: http://www.beagleboard.org/angstrom-mirror/www.angstrom-distribution.org/demo/beagleboard/ Apr 27 13:35:10 somebody teach google to return that one first... Apr 27 13:37:13 old kernel :P Apr 27 13:39:02 can you teach an old kernel new hacks? Apr 27 13:40:32 * mru watches idiots queue by the airport gate Apr 27 13:45:13 that image have a problem with pvr-init Apr 27 13:45:27 which one? Apr 27 13:45:29 they using &cputype Apr 27 13:45:32 $* Apr 27 13:46:03 and the scripa have a switch from OMAP* and TI*... Apr 27 13:46:06 script* Apr 27 13:46:33 so for run pvrInit i need change manually that scring with $cputype="OMAP3530" Apr 27 13:46:47 script* (sorry bad keyboard) Apr 27 13:47:47 maybe because that image is for beagleboard c4 not xM...i don't know Apr 27 13:48:36 right Apr 27 13:48:39 you don't know Apr 27 13:49:31 koen: and you know from a name of image....if is a image for beagleboard c4 or xM? Apr 27 13:50:09 both Apr 27 13:51:48 ok...but if i run $cputype in xM return 0x00000E00 not a string Apr 27 13:54:15 and that script have a case on "OMAP3530" or "TI816x" Apr 27 14:25:23 mru: what is thos opcode? e3001be0 Apr 27 14:25:26 this Apr 27 14:25:32 ah, he's on a plane Apr 27 16:18:10 it's always an adventure to see how patches are walking around until they finally end up somewhere or nowhere. the panic-heartbeat now is in -mm from which it seems to diffuse linux-next Apr 27 16:18:27 +to Apr 27 16:18:55 s/+to/+into/ Apr 27 16:19:02 ;) Apr 27 16:20:06 some needs to make that into a schoolhouse rock video Apr 27 16:20:16 haha Apr 27 16:20:33 hello Apr 27 16:20:35 that would be fun Apr 27 16:20:48 http://www.youtube.com/watch?v=tyeJ55o3El0 Apr 27 16:20:53 (original, bill to law) Apr 27 16:21:11 I just built a working version from angstrom repository for angstrom-v2012.5 Apr 27 16:21:12 (for those who have no clue what I'm talking about) Apr 27 16:21:37 however this version has very slow to boot up for first time Apr 27 16:21:44 I boot, load a module, do stuff, unload the module, try to load the very same module again and it says the module_layout is incorrect, does that mean I messed up some place in the kernel memory? Apr 27 16:21:49 i don't know why? any one has any clue? Apr 27 16:22:59 goodpeer, pastebin bootlog? Apr 27 16:23:02 another question is how to block a module Apr 27 16:23:09 ok Apr 27 16:23:30 watch it boot and see what steps takes a long time? Apr 27 16:25:46 jsabeaudry: does it work ok with a hello world module? Apr 27 16:28:16 jay6981, yes and it also works with that module usually Apr 27 16:29:01 and that module does some dma and stuff, so I suspect that I might be breaking stuff during the execution of the module :) Apr 27 16:29:24 seems reasonable Apr 27 16:29:26 hello jsabeaudry Apr 27 16:29:44 which file you stands for bootlog? Apr 27 16:30:51 goodpeer, the output of the `dmesg` command Apr 27 16:31:08 ok. thanks Apr 27 16:31:10 goodpeer, and by wathcing it boot I mean look at the serial output while it boots Apr 27 16:34:06 s Apr 27 16:34:28 [ 0.000000] Initializing cgroup subsys cpuset [ 0.000000] Initializing cgroup subsys cpu [ 0.000000] Linux version 3.0.28 (koen@revo.dominion.void) (gcc version 4.5.4 20120305 (prerelease) (GCC) ) #1 Sun Apr 22 21:45:11 CEST 2012 [ 0.000000] CPU: ARMv7 Processor [413fc082] revision 2 (ARMv7), cr=10c53c7d [ 0.000000] CPU: VIPT nonaliasing data cache, VIPT aliasing instruction cache [ 0.000000] Machine: OMAP3 Beagle B Apr 27 16:34:38 hello Apr 27 16:34:42 can you read? Apr 27 16:34:45 goodpeer, pastebin.ca Apr 27 16:34:52 hi Apr 27 16:34:53 asl? Apr 27 16:34:57 hi Apr 27 16:35:07 or your favorite paste tool Apr 27 16:35:47 i am not family with IRC so probably this time I CAN use paste Apr 27 16:35:58 [ 0.000000] Beagle second expansionboard: none [ 0.000000] Reserving 12582912 bytes SDRAM for VRAM [ 0.000000] Memory policy: ECC disabled, Data cache writeback [ 0.000000] OMAP3630 ES1.2 (l2cache iva sgx neon isp 192mhz_clk ) [ 0.000000] SRAM: Mapped pa 0x40200000 to va 0xfe400000 size: 0x10000 [ 0.000000] On node 0 totalpages: 128000 [ 0.000000] free_area_init_node: node 0, pgdat c068f344, node_mem_map c06c9000 Apr 27 16:36:04 :D Apr 27 16:36:10 [ 0.000000] Normal zone: 1024 pages used for memmap [ 0.000000] Normal zone: 0 pages reserved [ 0.000000] Normal zone: 126976 pages, LIFO batch:31 [ 0.000000] Clocking rate (Crystal/Core/MPU): 26.0/400/600 MHz [ 0.000000] Reprogramming SDRC clock to 400000000 Hz Apr 27 16:36:11 rofl Apr 27 16:36:15 ... Apr 27 16:36:33 [ 0.000000] dpll3_m2_clk rate change failed: -22 [ 0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768 [ 0.000000] pcpu-alloc: [0] 0 [ 0.000000] Built 1 zonelists in Zone order, mobility grouping on. Total pages: 126976 [ 0.000000] Kernel command line: console=tty0 console=ttyS2,115200n8 consoleblank=0 mpurate=1000 buddy=none buddy2=none camera=none vram=12M omapfb.mode=dvi:640x480MR-16@60 omapdss.def_disp=dvi roo Apr 27 16:36:41 goodpeer: go to www.pastebin.com Apr 27 16:36:42 gah, I really need ops in this channel Apr 27 16:36:46 goodpeer use pastebin or you'll get kicked out Apr 27 16:37:06 Russ: yes, ask jkridner_ Apr 27 16:37:08 goodpeer: once there paste all this junk in there Apr 27 16:37:29 goodpeer: once pasted send us the small http:// link ok ? yes then we will family your request Apr 27 16:37:59 sorry for the paste. I will try to do it. Apr 27 16:38:13 goodpeer: now you are acting like a goodpeer Apr 27 16:38:31 family your request? lol Apr 27 16:38:39 yes Apr 27 16:38:44 that what he said Apr 27 16:38:50 so its some code word language i dont get Apr 27 16:38:58 yeah he's got a very odd dialect Apr 27 16:39:03 so its best to speak to him in what he gets Apr 27 16:39:38 machine translated? Apr 27 16:39:46 sorry. here is http://pastebin.com/XKExm0F5 Apr 27 16:39:51 im guessing.... Apr 27 16:39:59 I hope I am doing the right Apr 27 16:40:10 or worse.... chinese robot translations Apr 27 16:40:10 it's ok now Apr 27 16:40:14 goodpeer: you are doing fine now Apr 27 16:40:43 thanks. I have several questions about beagleboard openembedded linux Apr 27 16:41:01 apart from the slow boot up issue, Apr 27 16:41:33 how can I block the modules at startup Apr 27 16:43:37 I guess the problem of my angstrom version is something wrong in clock? Apr 27 16:46:24 I use angstrom-gnome-image, anyone got clue why this image was so slow at the first boot Apr 27 16:48:30 like a kid it needs to grow up Apr 27 16:49:25 wooo, this is amazing,isn't it? Apr 27 16:49:46 I bet you don't have a kid-:) Apr 27 16:50:21 angstrom does a lot of setup and configuration the first boot Apr 27 16:50:58 so you mean this is the main reason why it is so slow at the first boot Apr 27 16:51:08 yes Apr 27 16:51:40 there could be other factors, but that is what makes the first boot slower than subsequent boots Apr 27 16:51:51 then why koen kooi's kernel was very fast to boot Apr 27 16:52:18 there is a lot more to booting than just the kernel Apr 27 16:52:48 This line is your pastebin: Apr 27 16:52:51 [ 2.642944] VFS: Mounted root (ext3 filesystem) readonly on device 179:2. Apr 27 16:53:06 indicates the kernel is done booting Apr 27 16:53:09 i know. I mean the beagle test imag was faster Apr 27 16:53:15 yes. Apr 27 16:53:22 for that you need bootchart Apr 27 16:53:52 dmesg just tells you the timing for the kernel, bootchart tells you the timing for everything else Apr 27 16:54:01 http://www.bootchart.org/ Apr 27 16:54:10 thanks I will do this later Apr 27 16:54:12 I've never looked, but angstrom probably has support Apr 27 16:54:45 one question about how to block modules at the booting stage Apr 27 16:55:22 I tried to write a config file and put it into modules-load.d Apr 27 16:55:39 then use update-modules to generate a file called modules at /etc Apr 27 16:56:03 but it was not working to block some modules that I don't want to be loaded Apr 27 16:56:55 and once my angstrom version booted, there was no network service Apr 27 16:57:14 i had to use /etc/init.d/networking restart to get network up Apr 27 16:58:38 the third question is everytime the board was booted, the date and time automatically reset back to 1970 Apr 27 16:59:13 i doubt the timer code was not working Apr 27 17:00:32 do you have seen a battery on the board? Apr 27 17:01:50 no. there is no battery Apr 27 17:02:08 the srm describes what to do Apr 27 17:03:16 ok. thaks Apr 27 17:03:20 thanks Apr 27 17:04:17 any advice for how to block modules? Apr 27 17:05:15 maybe you want to describe what your config file contains. Apr 27 17:05:19 jsabeaudry: remind me, it was edma fifo mode that didn't work for you, right? so you used increment mode with 0? Apr 27 17:07:51 local.conf contains DISTRO = "angstrom-v2012.05" Apr 27 17:08:19 then nothing more special. Apr 27 17:08:26 I mean the one you've created to block modules Apr 27 17:08:46 fwiw, angstrom only has bootchart2, not bootchart Apr 27 17:09:49 i try to block angstrom default usb wifi dongle, blacklist rt2800usb blacklist rt2x00usb blacklist rt2x00lib blacklist rt2870lib Apr 27 17:10:13 and I want to use rt3572 WUSB600N v2 Apr 27 17:12:02 I saw the previous version openembedded-release-2011.03 has rt3572 BB file Apr 27 17:12:42 but the new one koen maintained git tree, rt3572 was no long existed. Apr 27 17:27:21 jsabeaudry, nm, got crazy and searched the logs to confirm Apr 27 17:44:45 apologize to everyone, I didn't use this channel before, I post junks sorry Apr 27 18:00:29 gm :) Apr 27 18:03:22 gm djlewis Apr 27 18:09:02 mdp, well it's not really that it didnt work but that the "requirements" are so prohibitive Apr 27 18:09:49 mdp, the fact what you need to be 256 bits aligned is a deal breaker for me Apr 27 18:10:32 hrm, ok..ahh, that explains that comment Apr 27 18:10:51 may make sense when you have a 256bit wide fifo bu mine is only 16 bits Apr 27 18:11:00 sure Apr 27 18:11:01 I have 16 of them Apr 27 18:11:29 Have you found anything new regarding lockups with the edma test app? Apr 27 18:11:35 I was given a drive to review/debug that has non-working edma support Apr 27 18:11:47 oh, I followed up on that, probably when you were away Apr 27 18:11:59 it's a TI documentation issue Apr 27 18:12:00 the 3 vs 4 TC thing? Apr 27 18:12:04 yes Apr 27 18:12:08 whats a TC? Apr 27 18:12:18 linking to the 81xx specific sample app Apr 27 18:12:21 transfer controller Apr 27 18:12:47 edma is composed on a channel controller block and 1…N transfer controllers that actually do the work Apr 27 18:13:22 oh yes and you can link the TCs Apr 27 18:13:26 the CC block does all the management of "channels", priority, queuing based on all the triggering methods and param definitions Apr 27 18:14:06 then that is used to hand everything off to the TCs..they have more on 81xx, cause there's more "stuff" on it :) Apr 27 18:14:12 for lack of a better description Apr 27 18:14:55 and the TRM is misleading as usual…shows a block diagram right up front for am335x indicating that there are 4 TCs…there is not Apr 27 18:15:29 I submitted that issue to the TRM folks Apr 27 18:21:22 jsabeaudry, perhaps you'll find this amusing…this driver is checking for 32-bit alignment on the fast path and then using fifo mode. Apr 27 18:22:05 hmm, where did you get this three points from? ;) Apr 27 18:22:12 32 bits or bytes? Apr 27 18:22:48 … Apr 27 18:22:51 ;) Apr 27 18:23:09 mac users alert ? Apr 27 18:23:09 enough learned for today ;) Apr 27 18:23:19 bits brother! did I stutter? :) Apr 27 18:23:44 but good question :) Apr 27 18:23:55 nah its just that it should be looking for 32 bytes alignment Apr 27 18:24:18 right, that's why I saw it was a good clarifying question :) Apr 27 18:24:23 s/saw/say Apr 27 18:25:03 will triple check and dig deeper…but when I first started looking, "FIFO" jumped out and triggered that memory Apr 27 18:25:19 I'll try to get that heartbeat led to stop blinking while playing with my driver to try to have some repro steps Apr 27 18:25:50 panic Apr 27 18:26:07 collegue of mine informed me that on some beefed up version of the msp430 you couldnt use dma at the same time as something else Apr 27 18:26:29 sounds like a feature Apr 27 18:40:27 That is similar to the problem on the am33x. Some sort of interference between PIO and DMA USB operations Apr 27 18:43:19 um, the am33x issues with cppi 4.1 dma , pio, and musb is very specific to that IP Apr 27 18:44:00 but also found on am180x due to the same musb/cppi4.1 dma combination Apr 27 18:44:04 <_av500_> ti creates very specific issue for every IP Apr 27 18:44:04 That's what I thought Apr 27 18:44:23 <_av500_> they dont want to be caught doing copy&paste issues Apr 27 18:44:35 _av500_, job security Apr 27 18:44:40 <_av500_> except for EHCI Apr 27 18:44:49 <_av500_> that is broken now on 3 generatins Apr 27 18:45:00 <_av500_> it runs in the family Apr 27 18:45:01 _av500_, and we don't want your stinking s/w fixes :) Apr 27 18:45:28 <_av500_> phase 2) denial.... Apr 27 18:45:41 Somebody upstream recently fixed the kworker CPU hog problem, which makes my BBone much more usable even with DMA disabled Apr 27 18:45:50 * mdp listens to an nfc presentation and keep wanting to shout out, "that's not what it stands for!" Apr 27 18:46:14 <_av500_> mdp: the NFC people have NFC! Apr 27 18:46:21 indeed Apr 27 18:46:27 non-functioning cortex? Apr 27 18:46:48 <_av500_> no f'ing clue Apr 27 18:56:21 <_cjp> Hi. Anyone played with USB cameras on BB? We've got an issue of not being able to capture beyond VGA size. Just wondering if this is an inherent problem that everyone knows about; except me OR we've messed something up. Thoughts/feedback? Apr 27 18:59:41 av500: did you figure out that opcode? Apr 27 18:59:46 mdp: hey, do you do anything with the stellaris? Apr 27 19:00:07 no sir Apr 27 19:00:14 above my pay grade Apr 27 19:00:14 'k Apr 27 19:00:39 wrong star huh Apr 27 19:01:01 stellarium? Apr 27 19:01:55 guess ST it is Apr 27 19:03:56 shopping for an Mn n<=3 vendor? Apr 27 19:04:00 "roadmap from $1 to 1 GHz." Apr 27 19:04:20 av500: in case you didn't, it's movw r1, #0xbe0 Apr 27 19:04:27 indeed Apr 27 19:04:51 since the >=8 offers are no longer pop'ing... Apr 27 19:05:59 n=7 a thought? Apr 27 19:06:18 == Apr 27 19:06:21 I'd give anything a thought if they are available in a small enough package Apr 27 19:06:43 Mn n<=4 seems to be the most likely given the package constraints Apr 27 19:07:04 curious…what is the package constraint…edumucate me Apr 27 19:07:26 foot print area Apr 27 19:07:50 non pop'ing packages double the effective footprint area Apr 27 19:09:06 so since I don't know anything about the m0/m3 market in terms of packages…are they just very small discrete sizes or are they also available in pop? Apr 27 19:09:18 they are just smaller Apr 27 19:09:20 fewer pins Apr 27 19:09:24 sure Apr 27 19:09:42 there is a budgetary issue too...blind vias are not in the cards Apr 27 19:09:46 I wasn't sure if there was pop too Apr 27 19:09:52 ok Apr 27 19:10:28 am33x in POP with a 0.8mm bottom... would be nice Apr 27 19:10:39 w/o pop, there is 3 chip footprints min :( Apr 27 19:10:55 the m's w/o pop lets me do it with just 2 footprints Apr 27 19:12:24 actually... there might be an A vendor that doesn't require 3 chips...vague remember the atlas being absorbed Apr 27 19:13:38 can we pop the power management chip on top of the stack too? Apr 27 19:16:21 that isn't an A Apr 27 19:18:35 Russ: then it's no longer pop, but a steak sammich, and I think _av500_ has patented that ;) Apr 27 19:19:18 I'm getting hungry Apr 27 19:20:33 sammich... Apr 27 19:20:53 I know, its just questions now, so its probably good time to bail Apr 27 19:21:13 not very good question either Apr 27 19:22:33 maybe it's a day for rudy's Apr 27 19:23:33 I have only 1.52 hours left…but who's counting Apr 27 19:28:30 * djlewis is into a turkey sandwich.. Apr 27 19:29:52 Has anyone here experience with adjusting the alsa soc driver for the beaglebone ? Apr 27 19:32:27 UlrikKj: check the mailinglist, I posted a patch last week Apr 27 19:34:42 hi UlrikKj Apr 27 19:35:02 * jkridner__ looks at http://e2e.ti.com/support/embedded/linux/f/354/t/158851.aspx?pi239031352=2 Apr 27 19:35:17 koen: are you still having issues? Apr 27 19:35:44 I thought that driver wasn't *quite* working. Apr 27 19:36:32 pinmux Apr 27 19:36:53 jkridner__: what do you think? Apr 27 19:37:04 jkridner__: of course it's still fucking broken Apr 27 19:37:20 * mdp throws koen a sammich Apr 27 19:44:03 koen: ok are you referring to the mux33xx.c patch? Apr 27 19:44:20 I am trying this now Apr 27 19:44:30 koen, you need a team of trained driver monkeys Apr 27 19:45:35 I am trying to get the alsa driver register the mcasp as a sound device on the beagleboard even when there is no codec attached on the i2c Apr 27 19:46:10 Do you know if this is possible w the current code? Apr 27 19:46:37 I have had difficulties getting the codec to register so the alsa driver wont register as a result Apr 27 19:50:29 you do mean beaglebone, right? Apr 27 19:51:04 yes, beaglebone! Apr 27 19:52:07 I don' t know of anyone who's got it working yet, but I think these are the right people to help get next steps to debug. Apr 27 19:52:58 do you have the patch you've attempted to apply? is that the one at the above link? Apr 27 19:53:53 Well, I suppose I need all the changes in the thread Apr 27 19:54:05 mdp: I know, I have been trying to avoid hiring driver monkeys, but I guess I'll step up those efforts Apr 27 19:54:12 not just the patch to the mux Apr 27 19:54:19 http://dominion.thruhere.net/koen/angstrom/beaglebone/dvi-audio.diff is what was mailed to the list Apr 27 19:54:50 As far as I'm concerned: if a bone related patch wasn't mailed to the beagle list, it doesn't exist Apr 27 19:54:53 koen, they are kept in a different cage from the u-boot monkeys Apr 27 19:55:24 mdp: yes, the uboot monkeys tend to fling more feces Apr 27 19:55:28 (so I've heard) Apr 27 20:01:39 heh Apr 27 20:11:21 when I apply the diff patch to my linux-04.06.00.03.tar.gz tree the patch fails,- how do I apply this patch? Apr 27 20:14:35 you apply it to the beagelbone kernel Apr 27 20:14:45 not some TI kernel that lacks beaglebone support Apr 27 20:14:52 why would you do that? Apr 27 20:15:06 * koen now really walks away from computer Apr 27 20:17:48 So how do I get the beaglebone kernel source code with the changes for the pinmux? Apr 27 20:30:52 UlrikKj: am335x SDK has it Apr 27 20:32:11 is there a zoo somewhere? what are all these monkeys? Apr 27 20:33:02 aholler, sometimes one can't tell they are in a cage Apr 27 20:34:04 woops << Apr 27 20:52:31 i wanted the changes to the kernel tree from the thread http://e2e.ti.com/support/embedded/linux/f/354/t/158851.aspx?pi239031352=2 which apparantly is in http://dominion.thruhere.net/koen/angstrom/beaglebone/dvi-audio.diff Apr 27 20:53:23 so my q is: do I need to setup a git clone of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap.git to be able to apply this diff? Apr 27 20:57:12 UlrikKj: you can apply a functional am335x patch to an upstream tree Apr 27 20:57:35 you must use a TI or community fork for am335x support Apr 27 20:58:16 *sigh* Apr 27 20:58:18 friday... Apr 27 20:58:24 s/can/can't/ Apr 27 21:09:43 ok .. so how do I get the community fork of the tree? Apr 27 21:10:17 https://github.com/koenkooi/linux Apr 27 21:11:43 you probably want the r11 branch or so.. Apr 27 21:12:19 github is unhappy for me atm so you'll have to poke down at the branches link Apr 27 21:12:53 first time here-- any tips on debugging a beagleboard that prints a 40W during boot and does nothing else? was working fine earlier today. Apr 27 21:13:47 seems like a perfectly fine board Apr 27 21:18:12 this one ? https://github.com/koenkooi/linux/tree/linux-ti33x-psp-3.2-r10d+gitre8004dad869568692ca2a45c04e8464ce48cc4d7 Apr 27 21:18:26 can only see 10d, no r11 Apr 27 21:18:44 only getting 1.196 on the VIO_1V8 rail, and the tps65950 seems hot =X Apr 27 21:19:15 UlrikKj: probably whatever has the latest commits…my own stuff is still on r6 atm Apr 27 21:27:08 mdp: sorry for asking,- so to get that branch I should do a "git clone https://github.com/koenkooi/linux.git" Apr 27 21:27:56 and then "git checkout ...what" Apr 27 21:27:59 ? Apr 27 21:47:10 google or check github help for basic git branch workflow Apr 27 21:47:19 * mdp heads out camping..bbl Apr 27 22:07:08 mdp: thanks Apr 27 22:13:10 sigh, does "(EE) sgxAllocateDeviceMem: PVR2D_Ext_AllocExpMem failed" look familiar to anyone? Apr 27 22:17:58 does 0006-Increase-SGX-s-general-heap-size-with-48MB.patch from http://people.linaro.org/~rsalveti/sgx/kernel/patches/ do anything useful? Apr 27 22:27:09 <_av500_> wtf v4l2-gfx? anybody uses that at all? Apr 27 22:27:58 <_av500_> ah ok, 1y old... Apr 27 22:28:00 _av500_: these are old patches Apr 27 22:28:03 :) Apr 27 22:28:09 <_av500_> yes, I just realized Apr 27 22:30:33 still the 0006-Increase-SGX-s-general-heap-size-with-48MB.patch applies to the "latest and greatest" http://tigraphics.blogspot.com/2012/04/1q-sgx-driver-update-package-available.html Apr 27 22:31:07 <_av500_> well, its just a config patch Apr 27 22:31:51 <_av500_> fudging a few values Apr 27 22:33:14 well, I think I need some config tweaks, because this SGX junk is apparently running out of memory Apr 27 22:37:17 _av500_: I guess you could not care less about SGX integration with DRI/DRM (Xorg), right? Apr 27 22:38:04 <_av500_> right :) Apr 27 22:40:01 <_av500_> I will stand at the side lines watching sgx code being merged into systemd... Apr 27 22:40:22 lol Apr 27 22:42:09 <_av500_> of course by then 3 of the 11 remaining "linux on the desktop" users will protest Apr 27 22:44:12 _av500_: isn't it the year of the linux desktop? Apr 27 22:44:36 <_av500_> its the groundhog year of it Apr 27 22:50:30 have a good weekend guys :) Apr 28 00:10:02 is the beaglebone rev3 the same as rev5 in terms of hardware? Apr 28 00:18:32 Is anyone able to provide advice on what hardware to purchase? I would like to have a free hardware system (with associated CAD/circuit diagrams), running Linux Libre if possible? Wulgulmerang Apr 28 00:33:30 wulgulmerang: beagleboard could work Apr 28 00:34:19 or ask RMS what he uses Apr 28 00:43:33 is there anyone around with beaglebone experience? Apr 28 01:04:47 god that is a new one for the folder Apr 28 02:03:21 I thought this beagle thing was supposed to be low power. now it's asking for 40W and refuses to boot Apr 28 02:04:21 "Asking?" Apr 28 02:05:24 yes, on the serial port Apr 28 02:07:50 40w? are you trolling? Apr 28 02:08:32 * mru trolling? never! Apr 28 02:08:45 :P Apr 28 02:08:46 If it didn't boot, what was asking? Apr 28 02:09:53 it prints 40W on the serial port, then does nothing Apr 28 02:10:12 what else could it mean? Apr 28 02:13:31 Oil is often measured in such terms. 10W30 etc. Maybe it needs oil? Apr 28 02:17:09 I have some olive oil, would that work? Apr 28 02:17:31 that only works for arduinos Apr 28 02:18:19 40W? must be a cold weather beagle Apr 28 02:21:23 Since a Beagle max power is less than a tenth of that amount, i doubt it needs 40 watts Apr 28 02:22:40 BeagleWhat? bone? board? boardxm? Apr 28 02:22:47 borg **** ENDING LOGGING AT Sat Apr 28 03:00:00 2012