**** BEGIN LOGGING AT Mon Jan 18 02:59:58 2016 Jan 18 03:58:07 Oop da loop. Got to go. See you persons-peoples later. Have a good night. Jan 18 04:06:12 * veremit shakes head(s) Jan 18 04:06:28 i'm confused about the first "export DISK=/dev/mmcblk0" here: https://eewiki.net/display/linuxonarm/BeagleBone+Black#BeagleBoneBlack-SetupmicroSDcard Jan 18 04:06:55 is mmcblk0 just the sdcard reader device? Jan 18 04:07:13 yes .. or it often appears as /dev/sdX too Jan 18 04:07:33 depends, I have USB adapters Jan 18 04:07:36 it's just that, on a typical desktop system (e.g., Fedora), they show up as /dev/sdX Jan 18 04:07:44 ah. Jan 18 04:07:45 I think native sd card readers come up as mmcblk .. Jan 18 04:07:55 you mean ones built-in to the computer? Jan 18 04:08:03 he's got two sections in that 'blob' you should see. Jan 18 04:08:07 quite possibly Jan 18 04:08:35 its an either/or for whichever comes up onyour pc Jan 18 04:08:45 my laptop's built-in card reader shows up as mmcblk Jan 18 04:09:13 so .. there you go .. mystery solved Jan 18 04:09:33 yes. Jan 18 04:09:35 one'll be mass storage .. the other .. mmc. Jan 18 04:10:30 if they appear as sd then they're imitating an SCSI disk (commonly tunneled over USB as defined by the USB mass storage class) Jan 18 04:12:05 i also don't understand the subsequent steps where he appears to be copying both MLO and u-boot.img to the sdcard. but which partition? he hasn't even formatted it yet. Jan 18 04:12:19 (partition(s)?) Jan 18 04:13:01 yates: see README of https://github.com/mvduin/bbb-asm-demo Jan 18 04:13:02 is there a good guide on how what files and partitions linux wants to see on the .img? Jan 18 04:13:07 specifically "raw mmc boot" Jan 18 04:13:07 ha ha! Jan 18 04:13:20 read my mind. Jan 18 04:13:35 ewww raw mmc .. eww eww .. I Agree with RN on this one Jan 18 04:13:56 you don't just dd random binary blobs to random places on a disk (whether they truely are random or not...) Jan 18 04:14:02 veremit: well RN does Jan 18 04:14:09 no he doesn't. Jan 18 04:14:12 he patched uboot Jan 18 04:14:17 in all recent images he does Jan 18 04:14:22 well .. ok perhaps there is still One blob Jan 18 04:14:45 who's this dude mvduin? :) Jan 18 04:15:20 lol Jan 18 04:23:49 yates: beware that u-boot produces a file named "MLO" that's actually a concatenation of the raw MMC header, padded to 512 bytes, and the *actual* MLO Jan 18 04:25:50 good to know, thanks matt. Jan 18 04:26:35 inspecting the file with hexdump -C allows one to easily tell the two apart Jan 18 04:26:54 I am new to beaglebone black.I have accidentally formatted emmc os image.When I try to reflash it through windows desktop my laptop couldnt recognize BBB Jan 18 04:27:11 I have installed all the drivers also Jan 18 04:29:12 zmatt: so you mean it has the partition table + the mlo? Jan 18 04:29:19 no Jan 18 04:29:49 although raw mmc header and partition table *can* coexist, for some reason they don't in any standard images for the bbb Jan 18 04:30:06 instead the raw header is put at the second possible location, sector 256 Jan 18 04:30:21 (hence MLO starts at sector 257) Jan 18 04:30:27 oh. is that the reason for the seek=1? Jan 18 04:31:34 beware the bs=128k ... that's the unit in which count and seek is expressed (replacing the default of 512 bytes) Jan 18 04:31:48 so count=1 actually becomes sector 256 Jan 18 04:32:11 and this "MLO" will actually be raw mmc header + MLO Jan 18 04:32:21 since otherwise these instructions would not work Jan 18 04:37:58 http://gerbil.xs4all.nl/bbb-emmc-layout.txt Jan 18 04:40:06 except for the location where ROM looks for raw mmc header + MLO, the offsets for various things are all defined in u-boot compile-time variables, hence could be subject to change Jan 18 04:40:41 excellent info. Jan 18 04:41:11 perhaps this will save me a few hours of scratching my head wondering why my img doesn't boot... Jan 18 04:42:10 the fact that u-boot spits out a file named "MLO" but having a raw mmc header is really foul, since for booting using a DOS partition you need a file named exactly "MLO" that does _not_ have that header Jan 18 04:42:51 no argument that this whole procedure is very confusing. Jan 18 04:44:06 what is this "falcon uImage"? never heard of "falcon" Jan 18 04:44:32 there's a README about falcon in u-boot's doc dir Jan 18 04:50:38 ok Jan 18 05:01:26 llr. download a new OS image from the official site, write a uSD card, hold the 'boot' button next to the SD card while you power up, you will re-flash your beagle Jan 18 05:02:08 I recommend this page ... http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#2015-11-03 .. the 'flasher' image will reprogram your eMMC Jan 18 05:02:35 check whether you've got a revC board then you can use the 4Gb image .. otherwise use the 2Gb one Jan 18 05:03:04 seriously av500, etc .. we need a !faq bot in here :P Jan 18 05:04:27 you may want to use the 2gb version even if you have a rev C, to flash less crap and have more free space Jan 18 05:04:42 depends if you want chromium I guess :p Jan 18 05:04:44 (also faster download, less time to flash, etc) Jan 18 05:05:15 you still on dial-up zmatt!? :P Jan 18 05:07:03 is MLO less than or equal to 512 bytes in size? Jan 18 05:08:18 otherwise the "count=1" would put stuff up above sector 257 Jan 18 05:08:50 yates: MLO is less than or equal to 109 KB Jan 18 05:08:58 oh, wait. Jan 18 05:09:58 so the raw mmc header pluse the actual MLO is less than or equal to 109 kB? Jan 18 05:10:13 no, I mean the actual MLO Jan 18 05:10:57 you have to miss the partition table Jan 18 05:11:10 and I'm not sure mmc has an MBR does it .. Jan 18 05:11:26 can't do .. Jan 18 05:11:43 specifically, the memory range into which MLO is uploaded is 0x402f0400 - 0x4030b800 Jan 18 05:11:50 thats why you have al the other crap lol Jan 18 05:12:03 veremit: what on earth are you trying to communicate? :P Jan 18 05:12:06 ok I'm gonna grab a short nap :/ catchya later Jan 18 05:12:14 zmatt: rambling out loud Jan 18 05:12:16 nite nite Jan 18 05:12:20 morning :p Jan 18 05:12:35 i think i'll call it a night too. Jan 18 05:13:15 good nite folks. Jan 18 05:13:18 * yates waves Jan 18 05:13:18 yates: to be really specific, 109 KB is the max size of the image loaded by ROM Jan 18 05:13:22 MLO = 8 bytes + that Jan 18 05:13:29 good night Jan 18 06:22:11 Hi Jan 18 06:22:25 Is there sonebody? Jan 18 07:08:51 Thanks! I was holding the wrong button <_< Jan 18 07:09:42 Now I'm trying to access the PRU, but I don't have a `/sys/devices/bone.capemgr.9` directory Jan 18 07:11:29 I'm running linux 4.4.0 Jan 18 07:11:53 Running `find /sys/devices -name 'bone_capemgr*'` only gives: Jan 18 07:11:57 `/sys/devices/platform/bone_capemgr` Jan 18 07:13:01 But if I try `echo BB-BONE-PRU-01 > /sys/devices/platform/bone_capemgr/slots` I get "echo: write error: no such file or directory" Jan 18 07:13:19 Has the way to access the PRU changed? Jan 18 07:27:45 gyscos: yes Jan 18 07:28:13 gyscos: also cape manager seems to have changed a bit Jan 18 07:29:28 I see. Do you know where I could get up-to-date documentations? Jan 18 07:29:54 Most guides/posts online seem to refer to the old methods Jan 18 07:36:37 tbr: I found http://www.thing-printer.com/cape-manager-is-back-baby/, looks promising, reading now. Jan 18 07:39:16 good Jan 18 07:39:26 I sadly don't have much info myself Jan 18 07:41:05 there also a new mechanism via configfs Jan 18 07:41:32 so you don't have to abuse capemgr anymore to load cape-unrelated overlays (hence don't have to put nonsense cape metadata in the overlay) Jan 18 07:42:57 though I never use either really, I prefer static DTs Jan 18 07:47:09 I see... Indeed, for now, I don't have any cape, and just want to use the PRU. Jan 18 07:48:08 Hmm, echo'ing something into /sys/devices/platform/bone_capemgr/slots appears to work, it's just that BB-BONE-PRU-01 is not a valid value Jan 18 07:51:22 iirc it will try to load /lib/firmware/BB-BONE-PRU-01-00A0.dtbo (the 00A0 being the revision of the "cape") Jan 18 07:53:49 Aah, so that's what it does Jan 18 07:54:46 indeed, I have nothing related to the PRU here :-/ Jan 18 08:03:13 which image are you using? Jan 18 08:03:46 actually nm since I can' Jan 18 08:04:49 t really give useful advice here regardless of which image you're using :P Jan 18 08:05:50 I'm using an Archlinux ARM image Jan 18 08:07:27 It seems the `linux-firmware` package does not include the BB-BONE-PRU dtbo file :-S Jan 18 08:08:18 scratch that, the other BB-BONE files are owned by the kernel package itself Jan 18 08:13:09 Maybe it means the PRU is enabled by default, and doesn't need a separate device tree blob? Jan 18 08:13:16 How would I check that? Jan 18 08:19:19 Hello Jan 18 08:20:03 I'm the guy having problem booting the latest image on BBW (the same SD card works perfectly on BBB) Jan 18 08:25:07 djidan: check the serial console output Jan 18 09:08:14 anyone know of a touch screen the same size and compatible with the pi zero or something of similar size?? if so let me know both please, i want to set one up with bluetooth, wifi, and maybe RFID/NFC and a touch screen so it can be a controller for a door lock - which can be interfaced to with a cellphone or NFC/RFID/bluetooth keyfob/token/beacon Jan 18 09:19:18 a) pi* is off topic here Jan 18 09:19:46 b) I hope you're not planning to put all the electronics on the outside, that would be plain stupid Jan 18 09:20:07 c) there are plenty of small SPI displays, no touch on those though Jan 18 09:26:43 tbr: Here is my boot log : http://pastebin.com/WXuv6Xzm Jan 18 09:30:36 djidan: interesting. it seems to hang for 90s and then continue Jan 18 09:30:45 djidan: does this happen with earlier images? Jan 18 09:33:39 lol @ "irq 187: nobody cared" Jan 18 09:33:52 question is, is that an intc irq number or a linux-invented one Jan 18 09:34:00 wait nm, intc doesn't go higher than 127 Jan 18 09:36:28 clearly something is going quite wrong in the tps65217 driver Jan 18 09:38:24 it happened with the november image in latest-images as well Jan 18 09:39:08 and after those messages, all the BBW's lights go off (even the ethernet ones) and the BBW goes radio silent Jan 18 09:40:03 tbr - i said anything of similar size Jan 18 09:40:13 was just using pi zero as size reference Jan 18 09:40:43 djidan: does the problem also occur with a non-rt kernel? Jan 18 09:41:11 looking for SOC package and screen arround the same size pretty much Jan 18 09:41:18 zmatt: what do you mean ? Jan 18 09:41:44 djidan: well you're using an RT kernel right now (4.1.15-ti-rt-r40) Jan 18 09:42:14 That's the one included in the image, I didn't try to change it Jan 18 09:42:33 ok weird, rcn defaulting to rt nowadays? Jan 18 09:42:46 and tbr - looking to build some sort of enclosure so it would be like a normal door lock - just with a screen and NFC/RFID area Jan 18 09:42:58 try booting it up on the BBB and apt-get install linux-image-4.1.15-ti-r42 Jan 18 09:43:04 I don't know if there are Jesse images for BBW using another kernel Jan 18 09:43:17 lilCodie: still. NEVER put the logic on the outside Jan 18 09:43:23 zmatt: OK, I'll try that Jan 18 09:43:38 the logic wouldnt be - it would be on the inside of the door with cables running through a hole Jan 18 09:44:15 basically want it to be able to drive a solenoid or door strike or something similar Jan 18 09:45:01 querying the web search engine of my least distrust for "spi touchscreen" immediately yields: https://learn.adafruit.com/adafruit-2-dot-8-color-tft-touchscreen-breakout-v2/spi-wiring Jan 18 09:47:36 tbr- still too big of a screen IMO - want something to be able to display some symbols/numbers/letters and a red or green background to display authentication status - that way the symbols/or characters can change order so someone cant tell what area of the screen corosponds to what symbol as it randomizes each time its lit up Jan 18 09:50:13 lilCodie: if you go much smaller it becomes pretty much impossible to operate by humans. you should try and print some mockups to get a feeling for size Jan 18 09:51:19 basically want it 4 keyboard keys wide - by 1.5 tall approx - if i had to use an arbitrary means of measurment i have on hand Jan 18 09:51:38 (standard keyboard) Jan 18 09:52:04 basically small rectangular display Jan 18 09:54:04 is there a way to get decimal numbers with i2cget(1) readings? I can't find it Jan 18 09:57:41 43.2×57.6 mm is already pretty damn small in my book, especially if you draw keys on it. Jan 18 10:03:40 i know - just not looking for more then 4-6~ or so symbols - when they put the 1st 4+in the right order can always load another set of them - want the door lock enclosure and screen to be as small as possible so its not something the size of a pi2 logic board ontop of the handle since im guessing between the display, and NFC board, plastic/metal enclosure + Jan 18 10:03:40 wires itll be pretty big allready Jan 18 10:03:46 there are smaller SPI screens. not sure if touchscreens Jan 18 10:04:35 if you are looking for something elongated, try one of those matrix LCD screens Jan 18 10:04:54 but you'll need to place physical buttons next to that Jan 18 10:05:14 or figure out a way to overlay a custom touch surface Jan 18 10:05:31 debating just going with LEDs w/ a difuser and buttons since itll probably handle the cold better Jan 18 10:06:31 or something like the skywriter gesture sensor pi hat thingy Jan 18 10:06:41 or go for buttons with integrated OLED displays Jan 18 10:06:58 think OPTIMUS by art lebedev Jan 18 10:08:30 http://www.artlebedev.com/everything/optimus/mini-three/ could work Jan 18 10:11:03 https://www.adafruit.com/products/661 and some capacitive touch surfaces Jan 18 10:13:30 that intrigues me Jan 18 10:17:45 since its ok to be stored at 0c~ should be ok if its properly sealed for condensation - could also put a tiny resistive heater cloth in it to keep everything a lil warm when its cold out Jan 18 10:20:38 ipod nano 5th gen screen looks ok too would just need buttons of some sort or a capacitive touch layer Jan 18 10:21:41 its 2.2-inch Jan 18 10:24:07 http://www.aliexpress.com/wholesale?SearchText=capacitive+button Jan 18 10:25:15 I have one of those TTP224 things for playing around. you could obviously use metal foil pads inside a plastic casing to create touch surfaces Jan 18 10:25:38 zmatt : that seems to work ! I'll run some more tests, but thanks a lot Jan 18 10:44:28 djidan: ok, that probably means the tps65217 driver is badly written Jan 18 10:44:31 which makes sense, since it is Jan 18 10:45:20 RT kernels have a tendency to expose sloppy driver programming Jan 18 11:01:38 why would a RT TI kernel be default? Jan 18 11:01:51 ask rcn Jan 18 11:02:12 yeah Jan 18 15:37:32 Hi all, my beaglebone black has just arrived, and i never worked with this Jan 18 15:38:23 i need to change the debian image to angstrom image, i followed the steps but it didn`t worked Jan 18 15:38:50 why do you need to change the image to angstrom - it's no longer supported .. Jan 18 15:39:29 Even if it was .. its much harder to work with, unless you're familiar with embedded tools Jan 18 15:40:35 i have a project done with the comands for angstron Jan 18 15:40:42 just because it Jan 18 15:41:06 i don`t know how change the comands, this is my first project Jan 18 15:41:42 have some way to change it Jan 18 15:41:43 ? Jan 18 15:43:29 well.. the old images are at http://downloads.angstrom-distribution.org/demo/beaglebone/ if that's the way you wanna go Jan 18 15:45:59 I thought angstrom still had support, since they do not have, you know where I can find a way to change the language? Jan 18 15:46:24 by the way thanks for the link Jan 18 15:46:59 I've never used angstrom myself .. but a few others here have Jan 18 15:47:50 I've always taken the debian route on all my arm boards, chiefly thanks to RN's eewiki "linuxonarm" site Jan 18 15:49:57 depends what you're used to I guess Jan 18 15:50:01 Veremit, thanks for your help!!! i will try to do my project on debian Jan 18 15:50:31 its quite easy to use :) Jan 18 15:51:08 i think that isn`t hard to change the language, now that i know that don`t have support, i will do this Jan 18 15:51:19 thank you! Jan 18 19:14:38 Hi Jan 18 20:07:14 hello, any news on x15? Jan 18 20:10:29 I think not yet....:( Jan 18 20:12:10 It is sad( Jan 18 20:40:22 that's what happens when regulation governs what you do ... ;P its not all fun and games .. Jan 18 22:29:58 Hi.. does anyone know if beaagleboard offers "industrial temp certified" board or "industrial temp tested" or both? Jan 18 22:33:33 matrixboy: i vaguely recall one of the variants generally targetting that sort of thing... Jan 18 22:37:51 yes there is one . can't remember the vendor currently .. Jan 18 22:39:00 LMGTFY : https://specialcomp.com/beaglebone/ Jan 19 00:19:39 hmm I didn't see him slip in Jan 19 00:32:45 I have a beaglebone black running latest debian from 11-03-15 kernel 3.8.13-bone79 and I'm able to activate spi1 with echoing BB-SPIDEV1 into slots and I'm using the SPI just fine but what I find interesting is that there is no /lib/firemware/BB-SPIDEV1-00A0.dtbo and simply no BB-SPIDEV1-00A0.dtbo anywhere on the filesystem... but yet the capemgr thinks it has a BB-SPIDEV1-00A0.dtbo somewhere but dmesg doesn't show the path to where Jan 19 00:32:55 it is as if all .dtbo files are just inside the kernel Jan 19 00:33:41 So now I have no idea what .dtbo files that aren't even in /lib/firmware may be available for me to use pre-built. Anyone know how to find out what .dtbo setups are just "built in" to the kernel? Jan 19 00:34:52 I've got an overlay I made to use the eqep2B encoder hardware and I made the .dts and built it into a .dtbo and copied to /lib/firmware to use that hardware no problem. But for the life of me I don't see the spidev .dtbo file anywhere on the system. Used the find command everywhere and even mounted and looked in the FAT partition that isn't mounted because I disabled my HDMI. Jan 19 00:35:48 Hoping to know what .dtbo setups may be available without having to get the kernel source or some other source from Robert Nelson's git repository... I'm quite preplexed here hoping someone knows a quite answer. Haven't been able to find much w/ googling. Jan 19 00:36:14 i think spi-dev disappeared .. Jan 19 00:36:28 but I have no dt on my live beagle atm Jan 19 00:37:26 * GenTooMan just looks confused of course that could be normal for him. **** ENDING LOGGING AT Tue Jan 19 02:59:58 2016