**** BEGIN LOGGING AT Fri May 11 02:59:59 2012 May 11 03:42:21 lots of parts and joins May 11 03:43:20 Welcome to quiet channels on freenode. May 11 03:43:48 agmlego where here. May 11 03:43:54 we're* May 11 03:44:09 Yes, here. Did you think I was talking about somewhere else? May 11 03:44:12 Ah. May 11 03:44:17 Sure, and so are... May 11 03:44:28 ...169 others. May 11 03:44:34 Still a quiet channel. May 11 03:44:52 lol May 11 03:44:59 So you have a beaglebone May 11 03:45:59 No. May 11 03:46:04 board? May 11 03:46:28 I have an ex-xM, and will soon have a PandaBoard ES. May 11 03:53:22 so I just built MLO, uboot, and the kernel but when i power on the BB-XM only the power LED comes on and nothing happens May 11 03:53:53 any advice on troubleshooting? May 11 03:54:29 Try the card that came with it? May 11 03:54:37 yeah, the board works May 11 03:54:51 and I can make an image with OE that works May 11 03:55:14 but I am trying to use rcn-ee's stable-kernel tools May 11 03:55:41 and when I say the image with OE works, I mean it boots May 11 04:03:55 serekson_: OK, so what did you forget to include? May 11 04:04:07 Or, what software is a different version and breaks things? May 11 04:13:44 well, mlo and uboot are different May 11 04:13:50 I am guessign those are the problem May 11 04:13:58 since only the power led comes on May 11 04:14:38 but I am not sure how to figure out what is wrong with them May 11 04:22:37 well, I am pretty sure it is uboot May 11 04:22:45 but i just built it from mainline source May 11 04:48:15 why is there no syslog? none installed no available angstrom package :/ May 11 04:59:03 do I have to have a serial cable to get a syslog? May 11 05:06:49 you could probably build syslog fairly easily May 11 05:07:48 surprising though, pretty basic software May 11 06:27:29 ls -l /dev/root May 11 06:27:32 uups May 11 06:36:52 hmm, my beagle c4 only uses 4W (including hd) whereas my dockstar needs 8W. May 11 06:38:09 s/4W/5.5W/ ;) May 11 06:38:58 ok, between 4W and 6W ;) May 11 06:40:57 too bad the beagle doesn't have a gbit-if. May 11 06:41:25 Hi how can I please load u-boot through JTag from CCSv5. I was trying to but there are some memory problem during run. I was trying also load at first MLO and then u-boot. But withouth succes. Everythings is ok if i place them on SD card, but i want to load it through JTag. I know i shoul initialize sdram at first but please tell how. May 11 06:42:08 someone already explained it to you. May 11 06:42:32 most people here don't use jtag and almost no one uses ccs May 11 06:43:27 have a look in the irc-logs at beagleboard.org May 11 06:43:37 (when you've asked it the last time) May 11 06:46:10 or try to ask that in ti's forum where they support ccs. May 11 06:46:43 or maybe on the ml May 11 06:51:18 hmm, in a week I will get free amens 4 days long delivered right through my window :/ May 11 06:51:33 free amends? May 11 06:52:06 what did they do that this is necessary? ;) May 11 06:52:12 the catholics will overrun mannheim May 11 06:52:21 oh, christian marines. May 11 06:52:53 And I'm not far away from their main place and will hear everything May 11 06:53:04 its the MCI then.. Mannem Catholic Invasion. May 11 06:53:15 http://www.katholikentag.de/ (german) May 11 06:53:50 please, think of the children May 11 06:53:59 Hart du sein musst, junger Padawan. May 11 06:55:14 they made a test run some weeks ago, it was just horribly (loud) May 11 06:55:38 I am doing it the way it was written. Load an ELF spl file, execute it, and then load u-boot, but i get on ADRESS: 00020008C with instruction EAFFFFFE B 0x2008C. Where is the problem? May 11 06:55:42 hm wage loudness war on them? May 11 06:56:48 I think the monks from the monastery around the corner would come after me ;) May 11 06:57:14 aholler: pie jesu domine, dona eis requiem. *klonk* May 11 06:58:04 hmm, some eis could be required May 11 06:58:11 and pie too May 11 06:58:24 matus_: sorry, but not idea May 11 06:58:34 CSS is a TI-supported product May 11 06:58:50 did you ask in TI e2e? May 11 06:59:00 all the pilgrims will get free eat and drinks May 11 06:59:25 they even make a large table (several hundred meters) on one day May 11 06:59:28 my pilgrims usually come for free spice. May 11 06:59:51 mine come for free sarcasm May 11 07:00:16 But how elseway this proccess could be done? May 11 07:00:24 Which products do u use? May 11 07:00:40 av500: troll communities usually don't count as religions... May 11 07:01:06 they (catholics) have a very long experience in organizing such events. I think they also learned how to handle trolls May 11 07:01:36 aholler: bribing them with gob(e)lin art? May 11 07:02:01 ban trolling and then have an inquisition May 11 07:02:18 at least they have refrained from burning trolls in the last few years. May 11 07:02:44 matus_: you did that? http://www.beagleboard.org/irclogs/index.php?date=2012-05-09#T11:03:31 May 11 07:05:06 matus_: you might have a look at what people do to use openocd. I would expect openocd has something which initializes the memory May 11 07:05:22 on a beagle May 11 07:07:28 I was doing it exactly that way. But during debuging spl ELF file which is loaded correct way to SRAM. The procesor gets to bad adresses. May 11 07:08:25 Ok i wil strart with OPEN OCD at first May 11 07:09:40 usually the initialization of the memory in openocd can be found in some init-script (where they write some values to registers) May 11 07:14:43 matus_: if you load and run MLO via CCS, it should init the memory May 11 07:14:51 it should also init the UART and print something May 11 07:14:59 these things are easily checked May 11 07:15:14 once you did that, move to uboot May 11 07:17:01 matus_: if you need help to operate CCS, it has help files and TI support May 11 07:20:29 I understand. But it does print nothing. I konw it is loaded correctly to SRAM, but it seems it does not correctly initialize SDRAM. May 11 07:20:39 Only some instructions run correct way: May 11 07:20:42 SBCVCS R7, PC, #43520 May 11 07:20:50 CDPLE P3, #1, C5, C2, C9, #1 May 11 07:20:58 BPL 0x3F83F06C May 11 07:21:17 and now it jumped heer : 0002000c: E59FF018 LDR PC, 0x2002C May 11 07:21:25 I really dont know why May 11 07:21:56 the first instr. is loaded at 0x402f0400 May 11 07:23:36 matus_: caching on / off / cleared ? May 11 07:23:44 I would assume starterware has some instructions on that topic May 11 07:24:59 matus_: after you load MLO, you can inspect sram May 11 07:25:03 and then single step it May 11 07:25:11 the above instructions look like garbage May 11 07:27:16 are the down triangles here http://elinux.org/images/6/6b/Trainer_Rev-B_schematic.pdf ground? May 11 07:27:25 when i load u-boot-spl - which is ELF MLO file: the instructions in SRAM: May 11 07:27:40 IE TP5 May 11 07:28:01 402f0400: 72DF7CAA SBCVCS R7, PC, #43520 402f0404: DE125329 CDPLE P3, #1, C5, C2, C9, #1 402f0408: 5AD53B17 BPL 0x3F83F06C 402f040c: 2E726ECF CDPCS P14, #7, C6, C2, C15, #6 402f0410: 00D57B16 SBCEQS R7, R5, R6, LSL R11 402f0414: 41286E74 BKPTMI #0x86E4 May 11 07:28:06 sorry May 11 07:28:10 402f0400: 72DF7CAA SBCVCS R7, PC, #43520 May 11 07:28:14 402f0404: DE125329 CDPLE P3, #1, C5, C2, C9, #1 May 11 07:28:18 if someone wants to add a signed-off: or tested-by: http://lists.denx.de/pipermail/u-boot/2012-May/124165.html May 11 07:28:19 matus_ May 11 07:28:20 402f0408: 5AD53B17 BPL 0x3F83F06C May 11 07:28:22 dude, there's pastebin May 11 07:28:23 use pastebin May 11 07:28:27 LOL May 11 07:28:31 sorry May 11 07:28:49 fpaste.org has less advertising ;) May 11 07:28:58 you're debugging some JPEG picture? :p May 11 07:30:08 here: http://pastebin.com/P0YLuykb May 11 07:30:36 SRAM after load uboot-spl May 11 07:30:44 MLO is not in elf-format May 11 07:30:56 but how can it be exectued then? May 11 07:31:46 try file u-boot-spl (or whatever you are loading) May 11 07:31:55 this is some school project? May 11 07:33:06 u-boot-spl: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, not stripped May 11 07:33:25 ah, ok, so it isn't MLO. May 11 07:34:39 I think this file is correct. I should at first load this file to through JTag, shouldnt i? May 11 07:36:08 do you have tried to load and execute it? May 11 07:36:13 (without debugging) May 11 07:37:15 Not yet. May 11 07:38:02 maybe the debugger needs sdram too, don't know. May 11 07:38:32 no May 11 07:41:16 who was it that ran tincantools, I thought he was on here ? May 11 07:41:49 we belive in privacy ;) May 11 07:42:08 The same result May 11 07:43:29 aholler ok May 11 07:43:30 but if you use the MLO (which should be generated from the spl) it works? May 11 07:43:59 matus_: the MLO should spit out something on the serial May 11 07:46:01 Borillion: he is not here May 11 07:46:07 kk May 11 07:46:10 you can try prpplague once he is awake May 11 07:46:14 I figured May 11 07:46:28 prpplague, knows everything :P May 11 07:46:38 Yes it works as expected. May 11 07:46:56 when used from SD carad and with u-boot.bin May 11 07:47:01 SD card * May 11 07:47:34 hi all May 11 07:47:59 I'm still trying to build cloud9-gnome-image for beagleboard May 11 07:49:13 now it fails at task 2227: the package ti-cgt6x-1_7_2_7-r4 doesn't have a URL to fetch... May 11 07:51:06 WARNING: Failed to fetch URL http://install.source.dir.local/ti_cgt_c6000_7.2.7_setup_linux_x86.bin;name=cgt6xbin May 11 07:51:47 I am really lost :( May 11 07:52:03 koen: is this a known issue? May 11 07:52:26 sp00k: have you read the README in the cgt directory? May 11 07:52:36 sp00k: yes known May 11 07:52:45 is it possible to connect a ATmega328 to a pci card? May 11 07:52:58 with superglue? May 11 07:53:12 sure its possible May 11 07:53:19 men have walked on the moon May 11 07:53:20 ROFL May 11 07:53:27 * sp00k is reading May 11 07:54:01 Borillion, depends on how pedantic you want to be May 11 07:54:10 yes its possible if you use a pci bridge chip May 11 07:54:24 just wondering, I was trying to find an easy way to connect a am/fm tuner or something like that May 11 07:54:31 or yes its possible if you use the i2c signals defined in the spec (pcie) May 11 07:54:35 ? May 11 07:54:42 tuner? pci? May 11 07:54:48 you mean SPI? May 11 07:58:17 no like this http://www.drs.com/Products/ISR/PDF/SI9135.pdf May 11 07:58:22 tuner :P May 11 07:58:36 its pci May 11 07:58:41 it's i2c May 11 07:59:51 its cPCI May 11 07:59:52 except thats a defense grade product, Im talking consumer stuff same connector tho May 11 08:00:02 get a nice conpact PCI rack May 11 08:00:06 the beagle has pci? May 11 08:00:06 compact* May 11 08:00:10 aholler: no May 11 08:00:14 but the atmega328 May 11 08:00:18 oh missed that, I mean just pci May 11 08:00:28 koen: looking at board-am335xevm.c I see many capes with their own initialization code. Has initialization based on EEPROM contents been abandoned? May 11 08:00:37 an atmega haspci? May 11 08:00:37 so what exactly are you connecting to what? May 11 08:00:58 just asking if something like that would be feasible, why not use something thats off the shelf May 11 08:01:05 so what exactly are you connecting to what? May 11 08:01:25 ncbas: hasn't been written yet May 11 08:02:35 av500: pci is just a matter of an i2c gpio expander, /dev/mem and some python, no? May 11 08:03:13 you forgot duct tape May 11 08:03:57 behold the dragon! May 11 08:08:53 koen: sorry to bother you again, but I don't get it: The only reference to the missing url is in setup-scripts/sources/meta-ti/recipes-ti/devtools/ti-xdctools. Is there another place for the README you mentioned? May 11 08:11:19 av500, a card like this http://shopper.cnet.com/tv-tuner-adapters/asus-tv-fm-card/4027-9332_9-31205080.html May 11 08:11:42 just use the fm tuner off of it May 11 08:12:59 I figured it had everything you need ready so why not find a way to interface it May 11 08:14:07 ah May 11 08:14:21 Russ was right, thanks russ May 11 08:14:30 and you would input the FM signal into your atmega328 how? May 11 08:14:38 and what would you do with it? May 11 08:15:07 another one who forgets that writing sw needs time too ;) May 11 08:16:49 Borillion: http://www.sparkfun.com/products/10342 May 11 08:17:06 koen: I guess I have to download the cgt-package myself (as a registration is required) and place it in my filesystem an reference to it in the ti-cgt6x.inc. Is this the right path? May 11 08:17:42 http://www.sparkfun.com/products/10344 May 11 08:17:46 you'll get usb-receivers too, and some of them will work with an beagle (if the kernel has drivers) May 11 08:17:48 sp00k: yes May 11 08:18:46 can I reference to a file on my local filesystem in the ti-cgt6x.inc file (SRC_URI = "/path/to/xy.bin")? May 11 08:21:35 i guess May 11 08:22:05 thanks May 11 08:30:22 sp00k: just put it in the dowload dir May 11 08:36:00 ok, that was too easy ;) May 11 08:37:57 koen: I'm looking into that, but find the scheme a bit strange: seems to require a hughe table with ascii pin-names. May 11 08:38:26 and part seems to be duplication (ascii name and int mode)? May 11 08:47:56 ncbas: it should be completely binary May 11 08:48:21 ncbas: https://github.com/jadonk/bonescript/blob/master/bonescript/eeprom.js May 11 08:48:24 DT May 11 08:48:32 systemDT May 11 08:48:52 ??? the eeprom contents: yes. But the initialisation code in the kernel uses ascii strings. May 11 08:49:12 and need pin number lookup to ascii string. May 11 08:50:05 Just noticed: the pinmux array datasize in the kernel differs from that in the bone manual. Which is one is correct? May 11 08:51:14 koen: kernel uses 170 bytes, manual says 74 pins -> 148 bytes... May 11 08:57:14 koen: I already got my eeprom init'ed and it's been recognized by the kernel. But the pin settings aren't used. So that's what I'd like to implement. May 11 09:07:37 ncbas: the javascript implementation is what jkridner and I use, if the kernel is wrong we should fix it May 11 09:14:34 koen: Ok. If I start implementing this. I'm not duplicating someone elses work? May 11 09:15:34 ncbas: nope May 11 09:17:25 Hi everybody, I'd this when I boot on my bb: http://pastebin.com/8xxB5nLm May 11 09:18:03 koen: Then I'll work on it (as far as time permits). Do you have the contact to get this upstream? I't been a while for me ;-) May 11 09:18:54 http://pastebin.com/43H3miPP May 11 09:20:06 ncbas: cute May 11 09:20:15 ncbas: there is no am335x support upstream May 11 09:20:19 ncbas: much less beaglebone May 11 09:20:45 ncbas: and when I asked how to get it into the TI vendor tree I got a bullshit answer May 11 09:21:03 ncbas: so the best we can do is to get it onto the factory sd card May 11 09:25:25 koen: bummer. But we can get it into a tree you're maintaining? I don't want to maintain a patchset, done that for a while in the past and it eats up too much time. May 11 09:27:22 koen: bullshit - sounds like what ti is producing for the pruss right now. Thought they were into silicon and calculators (got my first TI-58 in Almelo!). May 11 09:41:17 I'd this error when I'm booting on my beagle board..; Any Idea? http://pastebin.com/4S91p0zW May 11 09:42:14 card seems fubar May 11 09:42:21 or not done correctly May 11 09:42:28 try a known good image May 11 09:42:35 just done May 11 09:42:51 it's a brand new sdcard with new image May 11 09:43:26 I heard about a switch code with "RESET" and "USER" button May 11 09:50:57 ncbas: yes, we can get it into my tree May 11 09:53:46 any idea May 11 09:53:47 ? May 11 09:54:51 fix the sd, u-boot doesn't find uImage May 11 09:59:32 aholler: fix the sd? you mean physically, or partition is wrong? May 11 09:59:44 how could I know? May 11 10:01:16 mount the card on your PC and try to read it May 11 10:01:30 it works May 11 10:01:53 i refreshly reformated, and recopied u-boot, uimage, and all the fs on the sdcard May 11 10:03:45 koen: Do you have any idea what causes ? May 11 10:03:49 is it possible that it's a hardware failure? May 11 10:04:12 koen: garbage following instruction -- `dmb ish' May 11 10:05:03 PaulePanter: my guess is some binutils versions are buggy May 11 10:05:14 PaulePanter: I know the ios binutils is oooooooooooooooold May 11 10:05:29 PaulePanter: maybe with binutils 2.22 we don't need those reverts anymore May 11 10:06:20 koen: Thanks. I will update the ticket with that information. May 11 10:09:29 this is the output of printenv: http://pastebin.com/FZ71Yui5 May 11 10:15:06 try fatload by hand May 11 10:18:26 any1 has the experience in combining boeegleboard xm it vitrolight LCD's May 11 10:22:17 aholler: not working http://pastebin.com/srNzGRBZ May 11 10:28:26 PaulePanter: attending linuxtag? May 11 10:30:09 very pleased with archlinux/ARM on beaglebone. so far it's been very nice to me :) May 11 10:32:00 koen: Probably. Expecially if you will be there. ;-) May 11 10:33:23 koen: Unfortunately I did not do a lot with OE after the OE-Core/meta-oe migration. That means nothing. :/ May 11 10:33:51 koen: Are you going to give a talk? May 11 10:52:05 Is it possible that my SD card reader is broken? http://pastebin.com/78PxSuhf May 11 10:52:37 it reads u-boot, no? May 11 10:53:13 yes, but it cannot launch uImage May 11 10:53:35 on BB xM, the bootloaders are only stored on sdcard? May 11 10:53:41 so you think the reader breaks inbetween loading u-boot and uIamge? May 11 10:54:17 I did a fatls mmc 0:1, and I had ** Unable to use mmc 0:1 for fatls ** May 11 10:54:19 lpi: where else? May 11 10:54:29 does your xm has nand? May 11 10:54:47 Everybody said on th net that there is no NAND May 11 10:55:04 so where should u-boot come from? May 11 10:55:10 if not from the card? May 11 10:55:14 yes, from the sdcard .. May 11 10:55:43 normally I could read on the partition where uboot is stored with fatls, right? May 11 10:56:12 yes, if u-boot isn't broken May 11 10:56:31 so is the command I typed above is right? May 11 10:57:04 try mmc reset before May 11 10:57:37 and check your partitions. cfdisk can do that May 11 10:58:23 we don't know what you've done with your sd. but when it loads mlo and u-boot it's unlikely that the hw is broken May 11 11:01:45 aholler: cfdisk shows clean partition. Concerning SD, I put the originally SDCARD that have never been modified, and it doesn't work May 11 11:03:54 aholler: and the kernel is on the fat partition (uImage) so I don't know why it doesn't boot May 11 11:51:05 lpi: still having problems? May 11 12:07:22 damian0815: yes May 11 12:07:41 lpi: you have a beagle Xm? which revision? May 11 12:07:50 rev C May 11 12:08:00 lpi: try mmcinit, mmc init, mmc init 0 May 11 12:08:26 lpi: you need to initialise the mmc reader somehow, but it's not clear exactly how sometimes May 11 12:08:43 lpi: or mmc listdev, or something May 11 12:10:02 mmcinit isn't recognized (it's normal), i did mmc init (seems to work), and mmc init 0 return me mmc help May 11 12:10:22 mmc list return: "OMAP SD/MMC: 0" May 11 12:12:03 damian0815/lpi, the command for u-boot changed over time, so it depends on your version of u-boot May 11 12:12:30 if you do a "help mmc" you'll see the possible arguments to the mmc command May 11 12:12:56 on current versions, "mmc rescan" is happy May 11 12:13:51 mdp/lpi: yeah mmc rescan sounds familiar May 11 12:14:28 damian0815: I played for 2 hours with mmc: tried to see partition with "mmc part 0", it returns Unknown partition table May 11 12:14:51 mmc rescan do nothing, but I guess it simply reload sdcard tpart table May 11 12:15:42 otherwise, Audio processor on the board let me cook my bacon-eggs dish on it May 11 12:15:50 ohh yeah May 11 12:15:52 too warm May 11 12:15:59 lpi: you need to turn down the volume on one of the channels May 11 12:16:02 So, I'm trying to design my cape so it's compatible with the widest possible variety of existing capes May 11 12:16:18 lpi: there's a power amplifier somewhere in there that's expected a speaker, and it's poorly wired May 11 12:16:22 I need a UART with a RTS pin; it looks like that's impossible to square with also not conflicting with the DVI cape? May 11 12:16:29 lpi: so when you turn it on it heats up to like a hundred degrees May 11 12:16:31 alsamixer May 11 12:16:42 turn everything off/down except the headphone out May 11 12:17:05 nothing is loaded on the board, kernel not loaded, so sound is off no? May 11 12:17:23 lpi: i think the sound mixer has a register that saves the mixer state May 11 12:17:28 the chip i mean May 11 12:17:28 The DVI cape uses pins required for UARTs 5, 3, and 4; UART 2's RTS pin is on the same pin as UART5_TX, and UART1's RTS pin is on the same pin as I2C2_SCL May 11 12:17:29 maybe? May 11 12:17:53 no nand memort on teh BB xM, but maybe in the audio processor May 11 12:17:55 i've only had that with a running linux system. but i've found the audio mixer settings persist between different MMC cards with different viersions of linux May 11 12:18:09 anyway May 11 12:18:12 yeah i think in the audio chip, the twl4030, which is also the power chip :-) May 11 12:18:13 anyway May 11 12:18:18 trying to boot on linux first time ^ May 11 12:18:20 ^^ May 11 12:18:34 mmc rescan didn't throw an error? May 11 12:18:51 if you mmc fatload now, does it look like it's working? May 11 12:18:58 (have you done this before?) May 11 12:19:10 lpi, mmc rescan runs mmc_init() on current u-boot so it doesn't do nothing…though you might have an unrelated failure :) May 11 12:19:25 * mdp toys with double negatives to add confusion May 11 12:19:31 mdp: i think by 'does nothing' lpi might mean 'prints nothing' May 11 12:19:42 yes ^^ May 11 12:19:43 aha, ok, yeah, that's expected, lpi May 11 12:19:59 no error = everything is going well :) May 11 12:20:10 thx damian0815, little slow..only on second coffee May 11 12:20:12 So I guess I have to decide if I should break compatibility, or find a way to void needing the rts pin May 11 12:20:21 well I can't see if all part are well recognized with uboot commands? May 11 12:20:41 not so much no... May 11 12:20:51 uboot only knows how to read FAT, not ext2/3 May 11 12:21:01 ok May 11 12:21:19 so... what can I do now? uboot is loaded into the memory, but cant load linux kernel image uImage May 11 12:21:22 what image are you using to boot? May 11 12:21:28 uImage is on the fat partition right? May 11 12:21:32 yes May 11 12:21:46 i mean what file did you download + burn to your sdcard? May 11 12:21:49 angstrom-narcissus builded image May 11 12:22:03 I extracted all files from May 11 12:22:21 copied the 3 files from ./boot/ folder from the tar.gz May 11 12:22:35 extracted all others files into the ext3 partition May 11 12:22:55 I did it something like 10 times, and now it doesn't work May 11 12:23:06 oh it worked before? May 11 12:23:15 note that I shorted something yesterday on this board, so maybe there is a HW failure somewahre May 11 12:23:19 yep May 11 12:23:22 ohhhh May 11 12:23:26 ^^ May 11 12:23:27 shorted sounds bad. May 11 12:23:44 yep... green screen for 2 secondes on DVI, and reboot May 11 12:23:50 ohhhhh May 11 12:23:54 sounds really quite bad May 11 12:23:59 yes I know May 11 12:24:10 although odd that it's showing up in u-boot May 11 12:24:26 but can't understand why I can have access to SDCARD and UBOOT with fried OMAP May 11 12:24:36 yeah, if you're printing stuff then your CPU is ok May 11 12:24:47 all parts of the CPU? May 11 12:24:48 more likely it's a power management thing.. perhaps ? May 11 12:24:56 depends on how fried May 11 12:25:02 all is powered up May 11 12:25:05 yep May 11 12:25:06 stir fried, french fried.... May 11 12:25:13 deep fried... May 11 12:25:15 so... now it's random isn't it? May 11 12:25:22 i've never had a partially fried logic component before. binary, either it works completely or it doesn't at all :-) May 11 12:25:36 I like when it does that May 11 12:25:48 lee searching, just buying another one =D May 11 12:25:53 less May 11 12:26:14 maybe you blew up parts of your sdcard? May 11 12:26:20 try another card, with a fresh install May 11 12:26:29 already done :s May 11 12:26:38 for simplicity i'd recommend one of the Ubuntu preinstalled images, or rcn-ee's minimal systems May 11 12:26:41 with the sdcard that they give with the board May 11 12:26:50 oh and that doesn't boot? May 11 12:26:51 link? May 11 12:26:57 no doesn't boot May 11 12:27:03 so I think it's game over May 11 12:27:12 http://elinux.org/BeagleBoardUbuntu#Oneiric_11.10 May 11 12:27:40 seems that way. although i'm highly suspected a fried TWL4030 May 11 12:28:01 especially if it's running hot. that's the chip that also manages distributing power to the various components, like the mmc reader. May 11 12:28:31 does it change something for booting? you said it was audio. I don't care about audio for this baord May 11 12:28:46 i've had it run really hot with audio May 11 12:28:49 damian0815: fwiw, u-boot groks ext2/3…patch posted for ext4 May 11 12:28:59 mdp: orly? May 11 12:29:36 lpi: but if it's running really hot for you, and it's failing to read the mmc, then you could have a power issue May 11 12:29:38 the angstrom uboot for beagleboard and beaglebone handles ext2/3/4 May 11 12:29:52 lpi: likely a short is going to blow up the power management circuitry before anything else. May 11 12:30:49 damian0815: ok, it was a lose fight anyway. I will try your oneiric solution and see what's hapen May 11 12:31:00 k, good luck! May 11 12:31:21 koen, I was hoping you'd chime in :) I just knew ext2/3 definitely for your distro since the images actually use it May 11 12:33:06 lpi: did those pastebins come from the serial console? May 11 12:33:12 yep May 11 12:35:03 serekson: any clues? May 11 12:36:09 no, I am pretty new to this. I was just wondering because the MLO and uboot I build dont boot but I don't have a serial cable to troubleshoot them yet May 11 12:37:54 serekson: hard to debug then... Maybe you can build on if you have a serial port on your PC May 11 12:40:11 lpi: yeah, I am goign to try and track one down today. Or if I can only find a serial connector I can hack my Arduino USB adapter to work May 11 12:41:12 which beagle do you have? May 11 12:41:44 xm rev C May 11 12:42:00 ok, no pins for serials available ;( May 11 12:42:45 hence the need for the connector May 11 12:43:58 damian0815: BB xM is laughing at me ^^ http://pastebin.com/ca8AcM9Q May 11 12:44:33 damian0815: and now: http://pastebin.com/0kA7QGtv May 11 12:44:40 something going wrong with power May 11 12:44:46 lpi: :-( May 11 12:45:44 I have a different error at each reboot xD May 11 12:46:12 lpi: I don't know how many voltage regulators there are but you could try checking their outputs if there aren't too many May 11 12:48:20 it shoudl be obvious if you damaged one May 11 12:48:48 ok will see. Are they in the datasheet? May 11 12:50:03 lpi: sadly the twl4030 is a BGA chip so no easy way to replace it.. May 11 12:51:09 lpi: it's dead, Jim May 11 12:54:05 whoa, are BGA power management ICs common on embedded devices? May 11 12:54:38 I have never seen one before, but this insn't really my field May 11 12:54:47 yes May 11 12:55:07 on devices that care about PM May 11 12:55:08 at least on mobile phones and the likes May 11 12:55:32 * av500 wants a BGA 7805 May 11 12:55:56 a 7805 with balls, indeed May 11 13:18:14 Hi, does anyone know what software the schematics were drawn in the reference manual? May 11 13:19:05 yes May 11 13:19:31 I assume the one who drew them knows May 11 13:26:05 niro: ask Gerald on the mailing list May 11 13:26:15 cheers May 11 13:30:12 on those lines. is there an authoritative physical layout drawing anywhere, for beagle/panda/beaglexm? whenever i've made enclosures i've aways ended up measuring the positions of the mounting holes by hand, and it always leads to woe. May 11 13:30:32 or do the mounting holes follow an industry standard? May 11 13:30:40 beagleboard.org/hardware/design May 11 13:33:03 thanks, but the 'schematics' PDF files don't have any physical design information, and i've never had success reading the gerber files. May 11 13:33:47 i understand orcad and allegro are commercial software, windows-based? May 11 13:34:26 http://gerbv.sourceforge.net/ May 11 13:34:38 mdp: tried that, failed. May 11 13:34:49 odd, works for me May 11 13:34:51 :( May 11 13:35:03 i mean i can get the images out, but have NO IDEA how to read mm measurements off the files. May 11 13:35:59 basically i just want to know the width/height of the board in mm and the x/y positions of the mounting holes in offsets from the sides. May 11 13:37:37 iirc the failure of gerbv was due to Xquartz failing to render the views properly May 11 13:37:52 there are Mac users among us. May 11 13:38:11 ESPECIALLY in the design/arts/object-building field.. May 11 13:38:54 ok i stop now. consider this a polite request for gerbers exported as PDF or at least PS documents. May 11 13:39:10 (with measurements.) May 11 13:39:15 ask Gerald on the mailing list May 11 13:39:24 kk will do May 11 13:39:32 what, mac users? run! May 11 13:51:07 Hi, my Bone only has a I2C SCLK running 25khz and not 100/400 khz. Anyone knows how to correct this? May 11 13:57:19 how do you measure that? May 11 13:57:34 Zollstock? May 11 13:58:16 I have a scope probing the SCLK I2C pin May 11 13:58:46 and you are counting the spikes or messages? May 11 13:59:06 I'm looking a the clock period (50% duty cycle width) May 11 13:59:45 there was one here a week ago or so, who said the same, he just was unable to interpret correctly what he saw on his scope May 11 14:00:13 thanks, I'm certain I'm reading the scope correctly May 11 14:02:19 I'm working with the 3.2.0 kernel and wonder if the problem has been solved already on the latest kernel? May 11 14:03:07 which problem? May 11 14:03:22 i2c speed is not a problem May 11 14:03:37 my scope allows me to assign veritcal bars between for example rising edges and then computes the frequency (I don't have to do any math :)) May 11 14:03:48 ok May 11 14:03:57 its maybe set too low in the kernel May 11 14:04:11 there was some discussion about that on the ML recently May 11 14:04:55 ok, yes I'm looking at a related thread on this topic but don't know yet how to correct the kernel May 11 14:05:48 I'll figure it out over time...just wanted to hear if anyone here had resolved such problem on the Bone May 11 14:06:26 as said, it was on the ML May 11 14:06:33 as said, the only who said such had it's i2c running at 400khz and thought it was 25khz May 11 14:08:46 so without pics people might not trust you ;) May 11 14:09:56 hehe! ok...it seems like the problem is well known here (https://groups.google.com/forum/?fromgroups#!searchin/beagleboard/beaglebone$2025khz/beagleboard/Q1pDr1lT7Gk/WouGuOyZWMgJ) May 11 14:10:51 aholler: there was one guy who did not manage to pass the boot arg May 11 14:10:53 I remember that May 11 14:11:00 I blame you for uenv layout... May 11 14:11:19 setting i2c-speed by boot-arg is just stupid May 11 14:11:56 and I haven't invented the uEnv-layout, it was already there. ;) May 11 14:12:09 I blame you for not ignoring that :) May 11 14:12:22 done by the kindly people from denx, in fact the master himself ;) May 11 14:13:32 why would setting it by boot arg be stupid? May 11 14:15:03 because it depends on the device behind? so you'll need something which knows how to talk with the stuff behind, and that thing could change the speed himself. May 11 14:17:03 isn't there an ioctl or similiar to change the speed? I've never played with i2c-tools in userland May 11 14:19:42 thurbad: btw, is your stuff done? May 11 14:21:24 define done? May 11 14:21:30 :) May 11 14:22:38 Guys, Changed the I2C clocks in uBoot - set all I2C periphs {0,1,2} to 400khz and it turned out 1 of them was running 100khz and another one 25khz (the third one I don't have access to on the Bone) May 11 14:23:23 I don't understand why they dont have the same SCLK clock though May 11 14:23:53 now trying with the 3.2.14 and crossing my fingers May 11 14:24:07 thurbad: right, it was "art", no? May 11 14:24:15 3.2.14 works....yahoooooo! May 11 14:24:43 default i2c clk freq = 100khz...nice! will stop making noise now! May 11 14:27:51 no, we have video content in our application but it's not art May 11 14:37:42 is it a good choice to switch to beaglebone? May 11 14:38:07 just need SPI, GPIO, and LCD plug possibility May 11 14:41:15 book_: should a patch go into the default kernel? May 11 14:46:51 koen: what version of node are we up to in the demo image? I could really use the 'fork' command. May 11 14:49:06 I see 0.6 in the feeds, so I'm going to try that.... May 11 14:49:13 not sure how you kept cloud9 working. May 11 14:54:12 jkridner: 'node' is 0.6.x, 'node4' is 0.4.x May 11 14:54:28 jkridner: cloud9 is run as 'node4 cloud9' and will use 0.4.x from the gui May 11 14:55:10 does that mean apps invoked from the gui will be run in 0.4.x? May 11 14:55:25 that pretty much makes writing bonescript to use 0.6.x useless. May 11 14:55:34 right May 11 14:55:47 you're welcome to fix the node version selector in cloud9 May 11 15:14:59 koen: just seen your picture album on flickr... Amazing all the things you'd done with the Beagle May 11 15:22:01 jkridner: bone101 runs in 0.6.x fwiw May 11 15:25:13 jkridner: wanna try a quick hack to use 0.6.x? May 11 15:25:41 sure... I need to build an SD card with the latest demo image again. I lost the last one. May 11 15:25:53 I'm verifying my code works with node 0.6.x now.... May 11 15:25:59 jkridner: apply http://dominion.thruhere.net/koen/angstrom/0001-ide-use-node-as-interpreter-for-sketches-instead-of-.patch to /usr/share/cloud9 May 11 15:26:06 I'm going to put in a fall-back to keep the existing functionality with 0.4.x May 11 15:26:26 k May 11 15:26:47 then restart c9 or reboot May 11 15:37:25 jkridner: socket.io api changed May 11 15:37:35 ugh. May 11 15:37:38 lient connected: undefined May 11 15:37:38 /var/lib/cloud9/bonescript/index.js:297 socket.broadcast('connect', 'connected: ' + sessionId); May 11 15:38:55 I might have already updated the code to deal with that on my wip-20120419 branch. May 11 15:39:20 i no longer have socket.broadcast May 11 15:39:41 I do have socket.sessionId though, so hopefully that is still around. May 11 15:41:33 http://www.reghardware.com/2012/05/10/product_round_up_arm_mini_computers_the_best_and_the_rest/ May 11 15:41:44 "As Upton puts it: "It’s not entirely clear to me why the Beagleboard is so expensive. Somebody in that Beagleboard value chain has got to be making a pile of money – I mean, $175 for a Pandaboard or $100 for a Beagleboard? Somebody’s got to be amassing a pile of cash there, because that’s a $10 chip in that device. I don’t know why they’re so expensive."" May 11 15:41:57 ugh. FUD. May 11 15:42:04 subsidies, how do they work May 11 15:42:30 BeagleBone processor can be bought as low as $5. May 11 15:42:32 boris houndleroy lives in an executive dog house from that pile of money :) May 11 15:42:41 jkridner: how did you test that wip branch? May 11 15:42:54 communists ;) May 11 15:43:01 killing distributors isn't a good way to make a sustainable business. May 11 15:43:10 jkridner: http://pastebin.com/Cred207W May 11 15:43:17 aholler, upton is representing that 99% May 11 15:43:41 koen: must not have pushed the update that includes the binary for misc.node. May 11 15:43:43 i know May 11 15:43:52 you can build it locally with the commands given. May 11 15:43:56 jkridner: distributor blood is the secret ingredient in the AM335x May 11 15:44:04 aholler, a true man of the people May 11 15:44:12 jkridner: I need this to be rcranetx proof May 11 15:44:29 for sure.... that's why it is still on the 'wip' branch. May 11 15:45:32 * koen runs said commands May 11 15:45:34 I've probably broken weatherstation.js from here to Sunday because I did several socket interface changes. :-/ May 11 15:45:51 but, bone101 was debugged. May 11 15:46:03 jkridner: http://pastebin.com/2NpFwEQ5 May 11 15:46:05 and I can fix weatherstation quickly. May 11 15:46:49 odd.... not sure why it wouldn't load if it built. May 11 15:47:17 guess the file bonescript/misc.node is there, right? May 11 15:47:19 mdp: everyone has to live and when upton wants that everyone in the western world earns the same as the chinese, it's his opinion, which doesn't have to be mine. May 11 15:48:08 jkridner: root@beaglebone:/var/lib/cloud9/bonescript# cp build/Release/misc* . May 11 15:48:43 XMPPwocky: I recommend reading http://www.longtail.com/the_long_tail/2009/01/a-business-mode.html from Chris Anderson of Wired fame and advocate for open source hardware. May 11 15:49:00 aholler, never confuse marketing with the truth..that's all I can say :) May 11 15:49:20 jkridner: oh, i know the Register article's BS May 11 15:49:26 mdp: I don't do that. May 11 15:49:38 k. May 11 15:49:58 koen: odd... I thought the copy happened automatically. May 11 15:50:49 mdp: I already said yesterday that I don't qualify for a sales job, because I don't want to lie ;) May 11 15:51:27 crap, I wonder if I've been testing an old binary. :( May 11 15:53:07 as someone writing a kernel in assembly, rpi's being ARMv6 instead of v7 takes it off the table May 11 15:54:37 mdp: lying isn't very target-oriented while working in the development business. So I'm kind of bad in that through experience. May 11 15:54:42 that article in the register is all that bad apart from the nonsense quote about beagle raking in money May 11 15:55:09 jkridner: someone should contact them and point out the inaccuracy of that claim May 11 15:55:36 let them get happy with r-pi's May 11 15:55:48 I don't care if they're happy with rpi May 11 15:55:56 but I don't like people bad-mouthing beagle May 11 15:56:37 in fact the beagle was the first cheap dev-board out there. May 11 15:56:53 yes, and that should be emphasised May 11 15:57:18 if nobody else want to I can shoot them an email from by @bb.org address May 11 15:57:32 you can't argue with such people like upton. May 11 15:57:43 funny thing is I spoke to Eben about it, told him I thought it was inaccurate and misleading, and he claimed he wouldn't repeat it. Guess The Register didn't contact him for an updated comment, since that was his exact quote from an earlier article. May 11 15:57:46 no, but I can point out the facts to the register May 11 15:58:28 jkridner: you should put that in an email to the register May 11 15:58:32 for the register nuclear power is something clean, cheap and not all dangerous. May 11 15:58:47 you can't argue about cheap May 11 15:58:52 they print everything someone pays for. May 11 15:58:53 and it's not particularly dangerous May 11 15:59:09 no risks involved May 11 15:59:39 much more people have died from bursting water dams and such things May 11 15:59:49 nuclear power is only cheap for the companies, because the tax-payer and future generations will pay a large part May 11 15:59:50 than from nuclear power accidents May 11 16:01:32 and regardless of who pays for content in the register, it's stupid to just let comments like that stand May 11 16:01:36 because people do read it May 11 16:01:40 to become back to the topic, just write an article about the beagle-series and give it free to the register. May 11 16:01:42 mru, does that include dams that were built by beavers? May 11 16:02:09 jsabeaudry: I was talking about dams used for power generation May 11 16:02:23 ;P May 11 16:02:29 the bone hits an interesting niche May 11 16:03:34 a sort of super-Arduino May 11 16:04:17 super-duper May 11 16:04:27 that and I don't need to learn wiring, I can just use BASH May 11 16:04:30 because I'm LAZY! May 11 16:04:32 suparduino May 11 16:05:48 you mean processing and not wiring? May 11 16:06:27 powerful enough to do abstractions; low-level enough to be... low-level. May 11 16:06:34 btw., why does the bone cost 70$? ;) May 11 16:06:44 and if you mean wiring for i2c .... definately not a steep curve May 11 16:06:48 its not a curve at all :/ May 11 16:09:19 it's like with all those app-devs. They are luckily give apple 30% and don't earn enough to make a living. Don't know whats the reasoning behind that. May 11 16:10:07 yet they still end up becoming millionaires :/ May 11 16:10:35 and most apps are just crap.... May 11 16:11:00 * XMPPwocky really hopes the raspi takes off in schools May 11 16:11:21 everyone will get a job then May 11 16:11:27 two or three are making a hit, but most of them just learn that they should develop opensource instead of closed source apps ;) May 11 16:11:46 most apps lack vision May 11 16:12:22 I hate that it's so hard to find good apps doing _one_ thing May 11 16:12:23 they go for the fastest money making route..... proper versioning /free/lite version and all that can prove to be more fruitful i beleive May 11 16:12:44 mru one thing? i find it hard to find an app that does anything..... May 11 16:12:55 i was looking for a morse code practice app of some sort.... May 11 16:13:00 all sorts of junk showed up May 11 16:13:22 most of them do several things poorly May 11 16:13:22 mru: on iOS maybe; if there's one thing Android does well, it's jinter-app communication May 11 16:13:24 and then i gave up May 11 16:13:33 flashlight app is pretty useful, works flawlessly May 11 16:13:59 ahhh the flashlight app i saw a freind using it .... pretty handy May 11 16:15:06 and ios ridiculus..... May 11 16:15:31 i hate appliances May 11 16:16:11 if you're going to put a cpu in something, you'd better give me a way to make it run my code May 11 16:16:36 I don't really care that I can't reprogram the washing machine May 11 16:17:13 mru, I wish I could change the ringtone on mine May 11 16:17:31 your washing machine has a ring tone? May 11 16:17:32 mru mb you can.... May 11 16:17:42 arcanescu: maybe, but I haven't tried May 11 16:17:44 check what its running maybe if its a pic or some atmel chip... throw it out May 11 16:17:52 then use a logic probe May 11 16:17:58 to see what its communicating to and what May 11 16:18:04 mru, ya, all the recent samsung washing machines and dryers have a tune when the cycle completes May 11 16:18:15 how annoying May 11 16:18:30 tune..... ryan air does that May 11 16:18:49 if the flight lands on time they play a victory theme music May 11 16:18:56 with the words " ryan air always on time" May 11 16:19:10 :/ May 11 16:19:16 one more reason not to fly with them May 11 16:19:32 arcanescu: i want to replace that sound with panicked screwams May 11 16:19:34 wait until washing machines are in the net, twitter will become flooted ;) May 11 16:19:39 I've flown with ryanair exactly once in 2004 May 11 16:19:41 ugh. node4 is giving me seg faults on input.js May 11 16:19:49 it was horrible then and from what I've heard it's even worse now May 11 16:20:26 mru .... definately is.... its more like a bus worse than a bus really May 11 16:20:37 i could hear the pilot going shut the door lets go go go go go May 11 16:20:42 * mru _hates_ buses May 11 16:20:48 whilst the passengers were still boarding May 11 16:21:07 they even introduced a scheme for pilots saying well we wont pay you ... you pay us May 11 16:21:12 to fly our planes get some expereince May 11 16:21:18 then maybe just maybe we will hire you May 11 16:21:26 amazing business model May 11 16:22:41 many companies try it that way. even in the it-sector. May 11 16:23:17 i get it for airlines but it-sector tell me more May 11 16:23:20 s/sector/business/ May 11 16:23:23 so i pay them to work? May 11 16:23:30 jkridner: so, 1GHz on xM: discuss! May 11 16:23:48 Hi guys I am new to beagleboard. I just got my board board -xM rev C1. I still don't have DVI cable and I am eager to export the board X over shh. I have Xming sever runing on my host PC but no luck. Should I do something on the beagle board ? May 11 16:24:01 ? May 11 16:24:05 koen: ? May 11 16:24:08 lolwut May 11 16:24:11 jkridner: it seems a hot topic lately May 11 16:24:20 yeah, it seems. May 11 16:24:30 arcanescu: what do you think are all those contests for? May 11 16:24:46 jkridner: so when are our pals from india going to fix it in mainline? May 11 16:24:49 aholler: oh.... hmmmm never thought of it that way May 11 16:25:00 dpenev: have you done X over ssh before? should be the same as with other Linux machines. May 11 16:25:06 koen: no idea. May 11 16:25:26 koen: I forgot about a bug in node-waf. May 11 16:25:44 aholler: but its not exactly give us the money!... its a more suttle way approaching it May 11 16:25:59 but hmm i see what you maen May 11 16:26:02 *mean May 11 16:27:07 koen: fyi: https://github.com/joyent/node/issues/1229 May 11 16:27:40 dpenev: export the board x over shh :/ .... what? May 11 16:28:01 dpenev: what do you want to do? May 11 16:28:42 jkridner: yes I ddi it in the past but I ma not expert. On the xM I have X proces running so I assume I need to see the X display exported on my host if I enable the X forwarding on my ssh terminal but it doesn't happen. I can dig some more myself but I think that someone can just point me on the steps necesery May 11 16:29:14 koen: node 0.4 seems to require the: conf.env.append_value('CPPFLAGS_NODE', '-DEV_MULTIPLICITY=0') May 11 16:29:24 ok I want to see the X display of -xM on my host PC over ssh May 11 16:29:26 jkridner: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/771537 May 11 16:29:42 oh that! May 11 16:29:45 dpenev: you might try starting with VNC and then try to debug X remote. May 11 16:30:21 jkridner: do you know what the proposed fix is? (the one they refer to in the April 26 status update) May 11 16:31:32 odd. no idea why the status would go to fixed. May 11 16:31:52 rsalveti: do you know why ^^^^ ? May 11 16:31:52 oh VNC? my host is Windows will this suits? May 11 16:32:06 jkridner, cool, somebody wrote the code? May 11 16:32:12 dpenev: sure. many VNC clients for Windows. May 11 16:32:20 got me. May 11 16:32:38 sakoman_: I heard through the grapevine that you had a nice meeting about ABB May 11 16:33:24 indeed, looking for support for the published spec using mainline kernels May 11 16:33:26 ok jkridner coul dyou please explain shortly what should I do after installing VNC? should I do something on the BB ? May 11 16:33:50 you need to invoke a vnc server on the beagleboard. May 11 16:34:12 http://elinux.org/BeagleBoardFAQ#Accessing_the_BeagleBoard_using_VNC May 11 16:36:04 sakoman_: fwiw, those working on am/dm37x in mainline are consumed on more fundamentally broken things atm..one reason the support story might continue to be a bit weak May 11 16:36:12 just imho May 11 16:36:31 jkridner: opkg install angstrom-x11vnc-xinit now :) May 11 16:37:37 fundamentally broken things ;) May 11 16:39:00 btw. why doesn't u-boot-spl support usb? May 11 16:39:18 aholler, nobody has implemented and submitted it May 11 16:39:25 because u-boot is terrible May 11 16:39:34 same reason u-boot-spl used to not support serial May 11 16:39:45 x-loader had it. May 11 16:39:49 * mdp states the obvious May 11 16:40:06 spl is not x-loader May 11 16:40:18 true May 11 16:40:26 * mdp continues to state the obvious May 11 16:40:34 x-loader wasn't maintained ;) May 11 16:40:43 that made it easy to add features May 11 16:40:54 yes May 11 16:41:32 there's a lot of demand for it…I'd expect to see submission any time now May 11 16:41:43 it's talked about daily here it seems like :) May 11 16:42:01 usb-support in spl? May 11 16:42:32 yes, depends though on what flavor of that you mean May 11 16:43:22 maybe I will have a look at it. I would like it, if it would be able to load u-boot from usb. May 11 16:43:28 usb in x-loader?? don't recall ever seeing a version with that! May 11 16:43:39 one flavor is standard dfu for spl May 11 16:43:45 the boot rom certainly supports it. is that what you are thinking of? May 11 16:44:18 sakoman_, depends on which part May 11 16:44:59 I suspect the reason it isn't in SPL is size related May 11 16:45:16 sakoman_: there was a x-loader which could load u-boot from usb. But I'm out in the nature and can't look at my beagle. May 11 16:45:28 sakoman_: generally people want to do factory load via it..which is not just rom code…but having spl get u-boot in…and then in turn all the images to write May 11 16:45:31 in a chain May 11 16:45:46 sakoman_: gadget side is small, not a problem May 11 16:46:36 omap4 spl just barely fits -- not even enough room to include the dram autoconfigure code! May 11 16:46:41 aholler, I suggest you poke Tartarus and see if he's up to anything in this area before you start May 11 16:47:03 mdp: I know the code, as I've added it to u-boot ;) May 11 16:47:09 sakoman_: thumb makes that better May 11 16:47:15 should be merged for the next release May 11 16:47:32 sakoman_: bah, omap4 May 11 16:47:43 mdp: so it't would be that hard to try if the usb-stuff fits into spl. May 11 16:48:02 aholler: iirc the usb boot stuff was more about mangling x-loader/u-boot for the format the ROM wants May 11 16:48:13 It should be adaptable, but I have not tried May 11 16:48:54 sakoman_, with what drivers loaded? May 11 16:49:00 mmc/fat? May 11 16:49:13 mdp: yes May 11 16:49:27 in my use case, I would leave those out May 11 16:49:58 then it might fit May 11 16:50:02 right May 11 16:50:32 I first have to check if u-boot-spl works in nand. I've build it the first time this morning and only tested it by booting from mmc. currently I still have x-loader and some 2011-u-boot in nand. May 11 16:51:02 Ok I have isntalled vnc server on the BB and I installed TightVNC on my windows host. I try to connect to the BB from the VNC Viewer but get ".. the remote machine actively refuse it" any suggestions? May 11 16:51:24 aholler: yes, SPL works from NAND May 11 16:51:53 sakoman_: cherry-pick the thumb patches from u-boot-arm or just grab 'em off of u-boots patchwork May 11 16:52:02 It's both a size and speed (due to redueced size) win May 11 16:52:36 sakoman_, if I had my design choice though, it would be usb serial @ ~1Mbaud and uart spl for factory load May 11 16:52:53 but that's not always an option on the final product :( May 11 16:52:55 Tartarus: I'll give that a try for omap4, it would be really nice to re-enable to ram autoconfigure May 11 16:53:08 I have proto boards with multiple ram types May 11 16:53:17 Aneesh did it for omap4/5 May 11 16:53:26 I think I tried just for fun on beagle May 11 16:53:36 ti should exchange the rom with some flash. May 11 16:53:45 Tartarus: thumb + ram autoconfigure? May 11 16:54:06 thumb dropped the size a good bit May 11 16:54:12 so autoconfigure could come back on i suspect May 11 16:55:11 aholler, can't have you overwriting the "security through obscurity" bits in there now May 11 16:55:48 I don't discuss drm ;) May 11 16:55:57 I just flame about it ;) May 11 16:56:17 I might have hinted a bit at my feelings May 11 16:56:25 just a hint May 11 16:57:06 ti might save some of the sram when adding flash ;) May 11 16:57:23 not sure if that could be an argument May 11 16:58:05 jkridner: how did that cloud9 patch work for you? May 11 16:58:20 I can't imagine anybody with enough political capital to drive a change from ROM :) May 11 16:58:29 got caught in e-mail and debugging code on node 0.6/0.4. May 11 16:58:54 still need to get a board booted with the latest demo image to apply the patch. May 11 16:58:55 mdp: atmel has ;) May 11 16:59:16 well, I mean inside the mothership here :) May 11 17:00:59 mdp: and making rom and sram smaller inf favour of flash? that would still left room for some obscurity in rom ;) May 11 17:01:16 and rom just had to boot from falsh May 11 17:01:32 that would even free some gpios May 11 17:01:52 thos sys-boot-pins May 11 17:04:07 but enough. Just my 2¢ ;) May 11 17:06:09 it'll cost you more than that May 11 17:08:48 I know nothing about what flash, sram and rom does cost (inside a chip) ;) May 11 17:14:22 doesn't matter…just the change costs across the board May 11 17:37:57 do you take a signicant perforamance hit if you virtualize your cross compiling machine? May 11 17:38:27 like any virtualisation May 11 17:39:32 that is what I figured but wanted to make sure May 11 17:59:23 Guess my beaglebone will never arrive. May 11 18:00:56 no, probably won't; best order an rpi instead May 11 18:03:34 mru: :-) May 11 18:06:06 lol mru having trouble getting yours too? May 11 18:06:15 not at all May 11 18:06:18 in fact, I have two May 11 18:06:27 lol May 11 18:06:35 then why do you say I won't get mine? May 11 18:06:47 have you heard of a thing called sarcasm? May 11 18:07:00 it's the best thing since sliced bread May 11 18:07:15 lol ok May 11 18:09:56 the mailman decided to keep it for himself May 11 18:10:22 his child needed an educational computer May 11 18:10:46 i thought he might use it for a brick May 11 18:11:03 and because those rpi just aren't shipping he had to steal a bone instead May 11 18:12:01 it's for the chilluns May 11 18:12:32 KeatonT: see, you are making a fine contribution May 11 18:12:45 kids these days... May 11 18:13:01 when I was a child, computers were operated with a hand crank May 11 18:13:29 "back in the day" May 11 18:13:31 when I was a child I had to walk 5 miles in ice and snow to get to the hand crank May 11 18:14:09 well, you all have zeroes and ones, when I was kid we only had zeroes May 11 18:20:28 We had to make our own zeroes... May 11 18:21:41 my grandfather actually had one of these: http://www.rechenmaschinen-illustrated.com/images/Odhner%20227_sn911052_1916_st.jpg May 11 18:22:37 I remember playing with it when I was 4 or 5 May 11 18:23:21 I'm sure one of my ancestors has had an abacus ;) May 11 18:27:45 i am certain some of mine had an abcess May 11 18:27:57 heck, I have :P May 11 18:28:01 but I remember those rechenmaschine too May 11 18:28:42 when I was a young fart working for an electronics parts store, I used a mechanical cash register :) May 11 18:29:23 * djlewis misses local electronics parts stores May 11 18:30:37 djlewis: not too many in AR? :) May 11 18:30:53 zero, nadda, zapp May 11 18:31:26 well, I take that back, at a recent ham fest I met a woman that does a little of it from Harrison Ar of all places May 11 18:32:40 dad had one of these in use until we switched his union over to using visicalc and fanfold checks :) http://img3.etsystatic.com/il_fullxfull.303980603.jpg May 11 18:32:50 those were fun to play with May 11 18:35:57 hey, my parents have one of those May 11 18:37:15 heh May 11 18:37:47 I've grewn up with those electronic sets from philips with springs like that: http://norbert.old.no/kits/ee1000/ee1050.html May 11 18:37:50 Russ, I was looking in his old basement office a couple months ago and was reminded of it May 11 18:38:02 my father bought them for himself ;) May 11 18:38:34 aholler, mmm….radioshack 250 in 1 kit here…found it in the closet before it was given to me and played with it for two weeks secretly May 11 18:39:32 those springy ones were nice May 11 18:41:08 building flipflops to blink lamps ;) May 11 18:41:31 hehe May 11 18:42:02 * djlewis gave one to his niece when she was 9. She built every project. May 11 18:42:31 Man, I'd love to have had one when I was 10 May 11 18:42:37 now it's better to give R-Pi :) May 11 18:43:04 its give beagles to mail man week ;) May 11 18:43:39 which voltage does the r-pi gpios have? I bet they have no protection too ;) May 11 18:43:42 i remember those radioshack 250 kits May 11 18:44:19 mranostay, I lived on that and the vic-20 in grade school…the good 'ol days May 11 18:44:22 * mranostay wonders how many batteries terminals he shorted on that thing May 11 18:44:41 this is back when i had no concept of resistors May 11 18:44:58 storage on cassette tape, they way computers were intended to store data May 11 18:45:30 30 minutes to load a programm May 11 18:45:40 * kkeller thanks mdp for making him feel old again... I was in my twenties when vic-20 came out… May 11 18:45:48 kkeller: you are old :) May 11 18:45:51 mranostay, when you thought, "these wires are like a series of tubes…." May 11 18:46:09 kkeller, that's why I want you around here :) May 11 18:46:47 * kkeller mutters… I remember when… ;) May 11 18:46:54 my atari 800 had tape, what a pain to load startrek if it hiccuped May 11 18:47:25 we use a big drawer of parts, breadboards, and tutorials on the tubes to replace the old 250-in-1 here May 11 18:48:01 kkeller, if you were rvinson you'd tell analog computer stories May 11 18:48:10 breadboads are the best thing since the transistor was developed ;) May 11 18:48:15 * kkeller LOL! May 11 18:48:33 er i think you have that backwards :) May 11 18:49:13 might depend on the boards. as said, I've ruined my fingers with those springs ;) May 11 18:50:00 mdp - looked it up I was 35 when the vic-20 was announced (1980) :( May 11 18:50:18 no comment May 11 18:50:50 at least I got to participate in this industry growing up… May 11 18:51:05 and down May 11 18:51:38 evolving yourself into a javascript programmer :) May 11 18:52:07 just one of the many… osX now too May 11 18:56:14 prpplague: hey purple May 11 18:56:19 hey May 11 18:57:03 mdp: you forget the 'd' in front of that word :) May 11 18:57:48 prpplague: you just reminded me to resubscribe for the list change :) May 11 18:57:55 * mranostay wishes he hadn't tossed his student id May 11 18:58:05 prpplague: that's how commanding your presence is May 11 18:58:09 hehe May 11 18:59:02 mdp: should i be bowing down right now? May 11 19:00:20 you better, or prpplague might throw prototype cTI adapter at you May 11 19:00:45 hehe May 11 19:00:59 his skills at throwing electronics parts like ninja stars are legen- May 11 19:01:06 …dary May 11 19:01:10 i want a cTI adapter May 11 19:01:18 * mranostay doesn't bow May 11 19:01:22 just don't lose a hand catching it May 11 19:01:57 would be an interesting workers comp claim May 11 19:08:51 <_av500_> gee May 11 19:09:13 enn May 11 19:09:35 <_av500_> eye May 11 19:10:13 hen May 11 19:10:29 mdp: bazinga May 11 19:12:26 mranostay: speaking of interesting workers comp, think of how you would be on a first date and the woman asks what you do for a living and you have to say: well, i am an actor, i currently play the character Tinky Winky on the series Teletubbies...... May 11 19:12:45 .... what did I just read May 11 19:13:44 <_av500_> prpplague: childhood trauma? May 11 19:13:48 ah pass the joint, please? :) May 11 19:13:49 _av500_: hehe May 11 19:20:47 can someone tell how to use this command properly in beagle board to increase any real time priority :chrt -p -f 50??? May 11 19:23:28 koen, ;) May 11 19:24:10 lol May 11 19:30:00 you can look up nice, not sure about realtime being supported May 11 19:30:49 thurbad: that's what chrt is for May 11 19:31:34 does angstrom support realtime priority? May 11 19:31:43 it's a standard linux kernel May 11 19:31:52 didn't think it did forsome reason May 11 19:32:09 linux is never a proper rtos of course May 11 19:32:39 but sometimes the pseudo-realtime scheduling policies are good enough May 11 19:39:04 we have a big chicken hanging out at the bank across the street. May 11 19:40:05 er? May 11 19:40:12 is everyone high today? May 11 19:40:32 frhiDAY May 11 19:40:56 aeh, frHIday I menat May 11 19:41:07 hmpf, too late ;) May 11 19:41:26 aholler confirms mranostay 's suspicions May 11 19:41:28 * mdp notes that aholler confirms mranostay's theory May 11 19:41:33 heh May 11 19:41:43 beat ya by 2 seconds May 11 19:41:53 story of my life May 11 19:42:04 beeing hi? May 11 19:42:16 bumblebeeing May 11 19:42:19 hi! being :) May 11 20:05:28 evolving into a javascript programmer, that will give bad dreams… May 11 20:08:11 aholler remember HTML5 is the future of windows. May 11 20:08:28 :P May 11 20:10:15 m$ will get bought by mozilla? May 11 20:11:02 doubtful. May 11 20:11:22 Just I find it saddening that Microsoft is encouraging HTML5 development for windows 8 May 11 20:11:31 and also pushing C# May 11 20:11:35 with XAML. May 11 20:11:38 I hate XAML. May 11 20:42:48 what is this Windows i hear about May 11 20:47:55 <_av500_> a new tablet OS May 11 20:48:07 <_av500_> one that saves us from browserbloat May 11 20:49:26 hey #beagle, is there a timer_add equivalent for struct timespec? May 11 20:49:50 google turns up lots of patches, including something called timespec_add_safe, but my man doesn't know anything about it May 11 20:51:56 ok forget it, i have found timespec_add and (handily) timespec_add_ns in linux/time.h May 11 21:03:06 oh crap but it's inside #ifdef __KERNEL__ May 12 02:29:56 Looks like I'll be using the pyBBIO May 12 02:30:01 libraries. May 12 02:30:31 Might be making a C++ version of it soon. May 12 02:54:55 KeatonT: you'll need some alcohol May 12 02:55:05 lol what makes you say that? **** ENDING LOGGING AT Sat May 12 02:59:58 2012