**** BEGIN LOGGING AT Thu Jan 08 02:59:59 2015 Jan 08 03:58:46 hello anybody there ?? Jan 08 05:11:29 I'm new to developing with hardware and want to start messing around with assembly and graphics. I don't want a fancy OS or anything, just barebones. Where would be a good place to start? I have a relatively strong c/c++ background. Jan 08 05:12:38 I have been looking at the BeagleBone Black, but don't really know where I would start with it Jan 08 05:16:00 I see a lot about bonescript, can you use c/assembly instead? Jan 08 05:33:31 grizzly: sure you can Jan 08 05:33:47 in fact you can run assembly on the PRUs without needing to remove the linux that comes pre-installed Jan 08 05:33:53 and, also Jan 08 05:34:07 i think there is now a C compiler for it Jan 08 05:34:12 for the PRU i mean Jan 08 05:34:59 however BBB is just a regular linux system, so you can of course write C and Assembly for that, too. just like you could on a laptop Jan 08 05:39:35 cool thanks for the info, any idea on some good getting started materials? Jan 08 05:47:48 eh Jan 08 05:47:51 i just googled Jan 08 05:47:57 as for arm assembly Jan 08 05:48:39 there is a good guide on thinkgeek Jan 08 05:49:39 i want to know how to program PRU Jan 08 05:49:54 arm already is fully covered by gcc Jan 08 05:51:36 ok will get my google on Jan 08 05:52:54 grizzly, shay_shay: http://mythopoeic.org/bbb-pru-minimal/ Jan 08 05:52:59 that was the guide i used to get started Jan 08 05:53:09 it's 'hello world' for the pru Jan 08 05:53:37 opcode: thanks. Jan 08 05:53:43 reading now Jan 08 05:54:52 good luck! Jan 08 05:54:57 it is fun Jan 08 05:55:59 I am going to morse code "hello world" on a LED Jan 08 06:17:52 nice Jan 08 06:41:51 anyone know where i can get bone-debian-7.6-console-armhf-2014-09-04-2gb.img Jan 08 07:58:15 hi, Jan 08 07:58:18 I followed this : Jan 08 07:58:18 https://eewiki.net/display/linuxonarm/BeagleBoard#BeagleBoard-Ubuntu14.04.1 Jan 08 07:58:33 and after booting i get this message : Jan 08 07:58:48 Kernel image @ 0x82000000 [ 0x000000 - 0x4d0058 ] Starting kernel ... Jan 08 07:58:59 kernel bootign seems to hung Jan 08 07:59:20 in the logs i get this messag eas well : Jan 08 07:59:28 unable to find omap3-beagle-xm.dtb ... Jan 08 07:59:35 is this is the cause for this ?? Jan 08 08:07:05 yes Jan 08 08:07:33 if your kernel version is not 3.2 Jan 08 08:11:00 Hi, the leds on my BBB are not working and when I plug a power source or I plug it wia USB I only have a flash from the power light. Do you have any ideas of the problem? Jan 08 08:13:20 connect a serial console Jan 08 10:06:56 hi guys Jan 08 10:07:05 is there any official Beaglebone black developer? Jan 08 10:07:39 what? Jan 08 10:13:58 who is developing the BBB kernel and so on? Jan 08 10:14:20 i'd like to ask some questions about setting "strange" speeds on BBB UARTs. Probably the kernel needs a patch Jan 08 10:16:12 hm robert c nelson maintains the kernel at least for the official debian images Jan 08 10:19:24 is he used to join here? Jan 08 10:27:28 Talorno sometimes Jan 08 10:27:52 I think he's US based, so US dailight and evening times are most likely, but he's not always here. Jan 08 10:28:05 asking on the mailing list might be more efficient Jan 08 11:12:49 Talorno: you could of course alter the uart's config regs directly after opening the dev... the kernel will never know :P Jan 08 11:13:42 hmm I just managed to wipe out my beaglebones bootable partition. Correct in thinking I'll need a micro-sd + reader in order to recover from this a situation, or is it possible to reflash emmc via a usb pendrive or similar? Jan 08 11:14:43 Talorno: or change the number of samples per bit to a custom value, that effectively rescales all bitrates and I don't think the kernel touches that setting (but I'm too lazy to check) Jan 08 11:15:12 Hello? Jan 08 11:15:58 Is there anyone here that has advanced knowledge of this? Jan 08 11:17:01 of what? Jan 08 11:21:17 rick-monster: microsd is the easiest Jan 08 11:21:42 rick-monster: anything else would probably take several magnitudes more time to set up and get working Jan 08 11:21:58 Talorno: the uart's register interface is "fun" to work with though... it's sufficiently confusing that the official examples from TI (e.g. the init procedure in the TRM, or far worse the StarterWare uart driver) contain errors ranging from minor to fatal Jan 08 11:22:12 thanks for the info tbr Jan 08 11:22:41 zmatt: it's TI after all :) Jan 08 11:23:32 Talorno: have you looked at the usual UART speed related things like multipliers etc? Jan 08 11:24:05 tbr: well yeah I'll admit every time I looked at some chunk of StarterWare so far I had a face-palm moment, but the nevertheless the uart is horrid above and beyond what is typical for peripherals Jan 08 11:24:16 can someone help me in deciding programming language to be used for my project (beagleboard first timer ) Jan 08 11:24:36 cooper: what are you trying to do in your project? Jan 08 11:24:38 haskell of course. Jan 08 11:25:00 O:) Jan 08 11:25:07 * tbr just went near PROLOG yesterday (on a shipping mobile phone) Jan 08 11:26:00 ECG measurement Jan 08 11:26:20 and waveform display Jan 08 11:26:45 quite ambitious Jan 08 11:26:47 C++ goes near prolog too, with the template abuse that's considered a perfectly normal way to do metaprogramming nowadays (sfinae...) Jan 08 11:26:50 :P Jan 08 11:27:23 lol :P Jan 08 11:27:46 zmatt: this stuff runs on actual current mobile phones: https://github.com/nemomobile/ohm/blob/master/examples/policy/prolog/audio.pl Jan 08 11:29:02 cool, just quickly scrolled through it but I can imagine prolog actually being good at configuration-stuff Jan 08 11:29:19 "find the combination of settings that gets us the results we want" Jan 08 11:29:39 it originated in Nokia Research Center, AFAIU Jan 08 11:32:40 I still want a good programming language for embedded systems... C actually really sucks when dealing with hardware... it's not expressive enough to explain the semantics or constraints of things like memory-mapped I/O so you end up having to thoroughly sabotage the optimizer (with obvious consequences for efficiency) and/or live in constant fear that the compiler is going to screw up your code :/ Jan 08 11:33:29 cooper: I'd suggest to keep it here on the channel. Jan 08 11:35:10 zmatt: C is one of the better languages when it comes to embedded stuff Jan 08 11:35:12 C++ is a bit better thanks to its metaprogramming functionality (especially in recent revisions), though it's still awful Jan 08 11:35:23 jacekowski: yes, and that's depressing Jan 08 11:35:35 :P Jan 08 11:35:41 sure Jan 08 11:42:31 I also like Forth, especially for interactively exploring a peripheral to fill the gaps in the documentation (and not uncommonly discover some documentation errors or silicon errata in the process :P ) Jan 08 11:43:27 it has the downside of being a bit difficult to read Jan 08 12:03:17 zmatt, sorry, I was away Jan 08 12:04:27 the thing is that actually i am using an external framework that has a plugin that outputs "DMX" over UART [it has to be connected to an rs-485 buffer and so on, obviously]. The point is that the plugin should set on it's own the bitrate... Jan 08 12:04:58 and actually i don't know how the board so well to set the uart register on my own to match by 250k bitrate Jan 08 12:04:58 DMX... I think you should seriously consider making that e.g. with PRUSS rather than the UART Jan 08 12:05:15 zmatt, read yesterday about it but i saw it's bitbanged! why??? Jan 08 12:06:13 it's what pruss is designed for Jan 08 12:06:29 and also... i cannot.. i mean i have to use that plugin that works directly with UART. Actually my problem is to get the underlying HW working good, not to program the whole thing. In theory all it's already done, and it should work if the UART works at the right bitrate Jan 08 12:06:31 it can access its gpio directly as a register, and it has extremely simple instruction timing Jan 08 12:06:48 you can't make reliable break timing with the uart, and DMX needs that Jan 08 12:07:13 also, some lower-end DMX fixtures need more than 2 stop bits to keep up Jan 08 12:07:19 the uart doesn't support that Jan 08 12:07:53 ...so.. fast and rough solution? Using an array of USB FTDI? Jan 08 12:08:08 i need 4 Universes at least [4 is better] Jan 08 12:08:20 *8 is better Jan 08 12:08:48 many options... pruss can do it easily, but you might also be able to abuse the LCDC for it Jan 08 12:09:21 zmatt i don't want / don't need to write "my own" code. I already have a framework that does ARTNET to DMX [it's OLA ] Jan 08 12:10:13 basically i want to have an Artnet node and i want to do it using BBB and OLA as a lot of people has already done Jan 08 12:10:32 a framework doesn't magically give you a DMX hardware peripheral, and most UARTs are simply not able to do the job properly Jan 08 12:10:57 zmatt, it's not the first time i use it. It works. But other times i used it mostly with USB FTDIs Jan 08 12:11:08 it's not a framework problem/fault Jan 08 12:11:36 I "just" should give it a working hardware...and i do not have time/resources to do it writing some low-level code Jan 08 12:12:06 if the crap you'd get out of an FTDI usb-to-serial converter suffices for the job then the internal UART will indeed work too Jan 08 12:12:11 so what, if not UARTs? Some USB FTDIs? Jan 08 12:12:45 no way it'll conform to DMX specs, but if you're lucky it'll work most of the time, depending on how tolerant your fixtures are Jan 08 12:12:48 Good morning! Jan 08 12:13:02 mmm. Good to know. Jan 08 12:13:20 what would you do? Jan 08 12:14:57 (for background: I've worked on the firmware of a standalone DMX controller product) Jan 08 12:15:37 zmatt can i ask you which one? i am a lighting designer and lighting console operator :) Jan 08 12:16:53 a few days ago I would have said PRUSS, but now I'd actually first check if the LCD raster controller could be abused for it ... there'd be some annoying bit twiddling involved though, so PRUSS may still be the better choice Jan 08 12:17:00 the LanBox-LCX Jan 08 12:18:31 it's too much work for me.. Jan 08 12:18:33 thing is, you *may* be able to get away with garbage timing, but it's very much dependent on the fixtures... some are very fussy about it Jan 08 12:19:14 so all there is to it is to make sure the UART is set to 250kbit (instead of 230400)? Jan 08 12:19:21 that should be easy Jan 08 12:19:33 zmatt, yep i know :( Jan 08 12:19:37 tbr, yup Jan 08 12:19:46 tbr: and you need to generate a break of a specific length Jan 08 12:19:55 I know people run their OMAP UARTs at e.g. 1MBit instead of that multiple value Jan 08 12:19:59 at least i would know that the board it's "probably" outputting the right dmx signal :) Jan 08 12:20:10 and then? Jan 08 12:21:11 tbr, not getting you, i think Jan 08 12:21:26 I don't think the am335x prcm gives you a choice in the clock for the uart does it? I'd need to check Jan 08 12:22:16 but otherwise changing number of samples per bit works too... that's actually used on omaps to compensate for weird functional clock (instead of having to change the divider values used by the driver) Jan 08 12:22:50 ok ok.. wait, i'm starting to get lost! you obviously are more "electronic" than me :P Jan 08 12:23:16 Talorno: you tried the obvious thing of using setserial and calculating a divisor? Jan 08 12:23:52 ? Jan 08 12:24:17 I'm actually not familiar with the linux omap-uart driver, does it allow setting an arbitrary divisor value? Jan 08 12:24:29 since then you're indeed done... divisor = 12 Jan 08 12:24:31 no idea, but I'd at least try it Jan 08 12:25:07 (48 MHz / 12 / 16 samples-per-bit = 250 kbit) Jan 08 12:27:37 "setserial -a /dev/ttyO0" confirms "Baud_base: 3000000", so a div of 12 would indeed yield 250k Jan 08 12:31:17 it's normally always given an 48 MHz functional clock, and it looks like the omap-uart driver automagicaly selects between 16 vs 13 samples per bit based on which results in the best approximation of requested baudrate Jan 08 12:32:16 Talorno: so looks like you can simply set the serial port to 250 kbit, no magic needed Jan 08 12:33:14 the thing is that actually i did it Jan 08 12:33:27 but the fixtures i tried with, they doesn't respond well Jan 08 12:33:34 and it looks like a timing problem Jan 08 12:33:45 break timing will be a big problem Jan 08 12:33:55 get a scope, look at the waveform Jan 08 12:34:05 did it, some time ago Jan 08 12:34:24 i can remember there was something wrong Jan 08 12:34:28 with timing Jan 08 12:36:23 and of course you need a decent RS485 driver hooked up to the uart Jan 08 12:36:30 yep Jan 08 12:36:50 essentially i put that BBB board inside a DMX splitter Jan 08 12:36:58 and hooked up the UART output to the 485 driver input Jan 08 12:36:59 but the break timing is a real issue, since the uart just has a bit you set and clear in software Jan 08 12:40:12 the current standard require a 176 - 352 microsecond break, and then 12 - 88 microseconds later the start byte Jan 08 12:43:06 I don't think fixtures are *that* fussy about it though... but like I said, some cheaper fixtures can't keep up with the data rate unless you increase the interbyte gap to something (much) larger than 1 or 2 stop bits Jan 08 12:43:58 some fancier fixtures however demand low-jitter frame timing Jan 08 12:44:42 because they derive their own internal timing to interpolate fades and make them smoother than the literal interpretation of the dmx data Jan 08 12:45:43 got it... Jan 08 12:46:15 so what if i'd need to have my own made artnet to dmx node? Jan 08 12:47:13 it's just the step of transmitting the dmx frame buffer that's fussy... it's should basically be treated as hard real-time Jan 08 12:47:17 still, many options available Jan 08 12:47:53 you can format the bitstream yourself and then shove it out of an SPI port or even a digital audio port Jan 08 12:48:13 I think I've actually already seen a DMX transmitter project for PRUSS Jan 08 12:48:20 i seen it too Jan 08 12:48:29 i don't know.... Jan 08 12:48:48 probaly i would go for the Microcontroller approach, using some pic16 and ethernet interface Jan 08 12:48:53 the LCD controller may actually also be able to do a good job at transmitting data, but you'd still need to pre-format the bitstream(s) yourself in software Jan 08 12:48:56 probably Jan 08 12:49:08 the LCD controller would give you up to 24 universes :) Jan 08 12:49:25 even if i'm not so good with micros and i would have to learn all -.- Jan 08 12:49:29 haha 24 is enough :D Jan 08 12:50:22 you'd basically make an image where every pixel is one bit-time (4 us) and every bit of the color data is a universe Jan 08 12:53:12 making pre-formatted bitstreams is annoying, but quite doable, and then there are many choices of peripheral to abuse as "gpio player" of that bitstream Jan 08 12:53:35 and you do get absolute control over the bit timing Jan 08 12:54:14 sure....but actually it's too much for me, i think Jan 08 12:54:46 * zmatt shrugs Jan 08 12:55:29 :/ Jan 08 12:55:44 there are plenty of options in any case Jan 08 12:56:01 choosing which is easiest is up to you Jan 08 12:57:11 mmmm Jan 08 12:57:18 generating a bitstream and shoving it out some port is probably the easiest since it avoids dealing directly with real-time constraints Jan 08 12:57:54 can i ask you what architecture the lanbox is? Jan 08 12:58:09 an ancient one Jan 08 12:58:30 i mean.. a micro? Jan 08 12:58:37 or something more "embedded" Jan 08 12:58:38 ? Jan 08 12:58:43 large part of the firmware is written in assembly to get the necessary performance (especially for the faders and mixer) Jan 08 12:58:54 16-bit microcontroller Jan 08 12:59:23 http://www.keil.com/dd/docs/datashts/infineon/c165utah_ds.pdf that thing iirc Jan 08 13:00:09 and how are you outputting dmx on this? bitbanging? Jan 08 13:02:50 not sure really, I don't think I ever touched that part of the firmware... I think it depends on whether it's in slow-dmx mode or not: probably normally uart, and synchronous serial port for slow-dmx (large interbyte gap) Jan 08 13:03:09 beatbanging Jan 08 13:03:43 but the whole firmware basically runs synced to the dmx frame timer Jan 08 13:05:20 got it, thanks Jan 08 13:06:02 anyhow, I'm off for a bit, good luck with your project Jan 08 13:06:11 thank you A LOT for your infos Jan 08 13:06:27 i think i'm gonna eat something... Jan 08 15:01:35 hey everyone! Jan 08 15:02:39 I am a newbie trying to run my blink.js in cloud9 on a mac Jan 08 15:03:15 the debugger highlights the line >>> var b = require('bonescript'); Jan 08 15:03:24 what am I doing wrong? Jan 08 15:21:37 where are the latest, prebuilt debian images for the bbb? Jan 08 15:24:23 there is this place: http://beagleboard.org/latest-images Jan 08 15:24:43 but there are much newer builds available. Jan 08 15:26:17 yates: you can have a look at http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Debian_Image_Testing_Snapshots Jan 08 15:30:22 LetoThe2nd: thank you Jan 08 15:37:51 what about, e.g., these: https://rcn-ee.net/rootfs/2015-01-06/microsd/ Jan 08 15:40:22 what's the difference there between bone-... and bb-... ? Jan 08 15:40:51 which one is for the beaglebone black? i'm thinking bb, but.. Jan 08 17:07:40 hello, i was wondering why does the website say that the BBB has a Processor: AM335x 1GHz ARMĀ® Cortex-A8, but the linux distro says its an ARM A7? Jan 08 17:09:02 ed_: where does the distro say that? Jan 08 17:17:35 ed_: do you mean ARMv7? Jan 08 17:37:54 ddrown: are you using debian on your bbb? Jan 08 17:38:07 is so, which incarnation? (link?) Jan 08 17:38:38 i've tried two of rcn's experimental versions, neither of which boot. Jan 08 17:39:57 i.e., from https://rcn-ee.net/rootfs/... Jan 08 17:43:18 yates: yes, I'm using the image from http://beagleboard.org/latest-images Jan 08 17:43:42 yates: "Debian (BeagleBone Black - 2GB eMMC) 2014-05-14" Jan 08 17:44:01 right. Jan 08 17:44:27 how do you flash your sdcard? Jan 08 17:44:36 do you do it under Windows? Jan 08 17:44:59 I actually flashed my sdcard from the BBB itself Jan 08 17:45:05 i'm doing it under linux using "dd if=file.img of=/dev/sdd" Jan 08 17:45:27 ddrown: how do you do that? Jan 08 17:45:35 with dd, like you did Jan 08 17:45:47 except I didn't store it somewhere first Jan 08 17:46:06 hi! Jan 08 17:46:14 how do you flash something that's not on your local drive? Jan 08 17:46:28 in very tricky ways, I wouldn't recommend it Jan 08 17:46:35 are you somehow using ssh/scp? Jan 08 17:46:53 I don't see it in my command history anymore, so all I have is a vague memory Jan 08 17:47:05 really? hmm. Jan 08 17:47:27 jaybird_: wish i was. Jan 08 17:47:44 yates: I probably streamed it over the network using socat or nc and piped that to dd Jan 08 17:48:02 are you running from internal eMMC? Jan 08 17:48:09 yes Jan 08 17:48:14 and then flashing the flasher version? Jan 08 17:48:27 yup Jan 08 17:48:31 right. Jan 08 17:48:36 I ran into what I think were out of memory crashes Jan 08 17:49:02 I think I telling dd to use direct (non cached) access fixed that Jan 08 17:50:49 Background: I want to use my beaglebone black with simulink/matlab. Matlab only has support for beaglebone-xM or beaglebone Bx/Cx. Question: Can I use their image for the beaglebone xM on my beaglebone black? In other words, can a xM image be used on the black? Jan 08 17:52:06 I didn't quit... I don't know why it's saying I did... Jan 08 17:55:01 Does anyone know if a beagleboard-xM SD card image works on the beagleboard black? Jan 08 17:55:13 jaybird__: i believe so Jan 08 17:56:35 So if a program (matlab) has support for the beagleboard-xM, I can just tread my beaglebone black like an xM with the Ubuntu image that matlab provides? Jan 08 17:57:36 beaglebone ideal for git server / backup server / light duty smb server? Jan 08 17:57:47 i think thatis the case, yes Jan 08 17:58:54 Thanks!!! Jan 08 18:05:28 i retract that. sorry jaybird__ (who is now gone..) Jan 08 18:32:21 ok, i've tried about 4 different images now, one of which worked. Jan 08 18:32:52 the one which worked proves to me a) the sdcard is good, b) my sdcard flash process is good, and c) my beaglebone-black board is good Jan 08 18:32:59 but the other three won't boot Jan 08 18:34:07 "won't boot" means three of the four blue leds right under the usb connector come on solid and stay solid, and no network connection (neither ndis/usb nor the ethernet) is alive Jan 08 18:34:24 any ideas what's going on? Jan 08 18:34:30 connecting to the TTL serial would give more info on what's going on Jan 08 18:35:52 ddrown: ok, let me look "TTL serial" up in the srm. Jan 08 18:39:05 ddrown: do you mean using a usb/serial adapter plugging into the same usb you can power the board from? Jan 08 18:39:26 yates: something like this https://codechief.wordpress.com/2013/11/11/beaglebone-black-serial-debug-connection/ Jan 08 18:40:01 USB/serial adapter yes, but not plugged into the BBB USB Jan 08 18:40:59 ok, thanks ddrown Jan 08 18:41:00 J1 has the debug serial (at TTL levels) which will have the bootloader messages and other boot messages Jan 08 18:42:04 elinux has more USB serial cables and their connection info - http://elinux.org/Beagleboard:BeagleBone_Black_Serial Jan 08 18:43:40 great! Jan 08 18:46:48 you can just plug the "standard FTDI cable" shown in that last link directly into the serial header? Jan 08 18:47:29 looks that way. I don't have one to confirm Jan 08 18:51:11 if you have the right cable then yeah you can just plug it directly into the header (you do have to pay attention to insert it correctly since the ftdi cable has a 5V supply on one of its pins... I actually removed that dangerous wire from the plug to avoid accidents) Jan 08 18:51:42 thanks zmatt Jan 08 18:59:03 (you can remove individual pins from that SIL connector without damaging them, so I just put a little "sock" of heat-shrink tubing on it in case it ever ends up being useful again, but 5V and modern chips do not mix very well) Jan 08 19:54:11 hi Jan 08 19:56:41 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 19:57:58 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 19:58:52 are you there Jan 08 20:00:33 Stoob are you there Jan 08 20:02:30 I need help Jan 08 20:03:39 can you help me Jan 08 20:03:50 please Jan 08 20:04:02 ddrown: btw, it's funny they call 3.3V "TTL levels" - they're not. Jan 08 20:04:23 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 20:04:24 TTL was 0-0.7V low, 3.7-5V high Jan 08 20:04:32 or so. Jan 08 20:05:11 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 20:06:01 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 20:06:40 yates: yeah, it's not actually TTL voltage levels, but "TTL Serial"/"TTL Uart" seems to be used to describe it Jan 08 20:07:16 how would set the GREEN_LED and the RED_LED to both of the bottonStates Jan 08 20:08:27 Are you there? Jan 08 20:10:20 Can you help me please Jan 08 20:10:47 Can you help me please Jan 08 20:13:53 please before you leave can you help me Jan 08 20:14:18 kr_: please don't spam the channel Jan 08 20:14:27 if nobody answers, asking 10 more times is not going to help Jan 08 20:14:38 ok Jan 08 20:14:44 try asking more clearly, I have no idea what you are talking about Jan 08 20:15:25 provide code for context, otherwise what you are asking is not clear Jan 08 20:17:37 hey just struggling to get the debian wheezy 7.7 image working ( http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Kernel_Upgrade the one called Console: Flasher) - anyone had any luck with that one? Jan 08 20:18:26 I am try to get help to make the Launch pad turn on the green led when I press on one botton and turn on the red led when I press the other botten Jan 08 20:19:44 I thought the flashing stage had worked correctly but can't connect via ssh through usb and can't see the device using lsusb on the host machine Jan 08 20:20:04 ok bya Jan 08 20:24:19 so just attempting to flash it again - single LED pattern dances side-to-side for a minute or so. Now I thought once all 4 LEDs light, this indicates that the process is finished. Is that correct? Or should I wait longer? Jan 08 21:18:09 kind of a networking question plz. I would like to turn my bbb into a wired router with an 8 port ethernet switch behind it. I would have to use an ethernet to usb adapter, would the usb/ethernet adapter bottleneck the transfer rates? Jan 08 21:32:03 aikidouke: the USB adapter probably wouldn't, but the BBB's USB hasn't been reliable for me Jan 08 21:32:43 Hi everyone Jan 08 21:32:45 Do anyone know is it possible to access LCD timings(hsync etc) from userspace? i'm using am3352 cpu and da8xx_fb kernel module. Jan 08 21:56:42 spaszkoPL, if you dont mind me asking what lcd are you using? Jan 08 21:59:19 ddrown:what problems have you had/distro? Jan 08 21:59:26 Winstar WF57DTI. Jan 08 21:59:36 It's 320x200 CGA display Jan 08 22:00:11 aikidouke: I'm using the Debian distro, and the USB port powers off every few days with the message "CAUTION: musb: Babble Interrupt Occurred" Jan 08 22:00:14 i've used values from datasheet, screen shows but it's moved 4 px right and about 100px down Jan 08 22:01:19 So i thought that maybe changing this values from userspace and testing hsync and vsync will give me an answer why it's not working properly. Jan 08 22:01:39 mistawright : Jan 08 22:02:00 ddrown: thanks Jan 08 22:10:47 rick-monster``: i'm also have trouble with a 7.7 img but it's the non-flasher version. Jan 08 22:11:39 neither bb-debian-7.7-console-armhf-2015-01-06-2gb.img or bone-debian-7.7-console-armhf-2014-10-29-2gb.img work Jan 08 22:12:23 but one of my other images works, which worked proves to me a) the sdcard is good, b) my sdcard flash process is good, and c) my beaglebone-black board is good Jan 08 22:12:30 hmmm - just about to try flashing BBB-eMMC-flasher-debian-7.7-lxde-armhf-2014-12-19-2gb.img.xz ... Jan 08 22:12:47 where are you getting it from? Jan 08 22:12:52 I was able to flash and boot BBB-eMMC-flasher-debian-7.5-2014-05-14-2gb.img Jan 08 22:13:07 rick-monster``: could you ssh in? Jan 08 22:13:08 using microsd adaptor and the sd slot on this intel chromebook Jan 08 22:13:14 yes I ssh-ed in Jan 08 22:13:58 but wasn't able to upgrade the kernel, using I think the same steps as I used on the factory-installed image Jan 08 22:14:23 kept running out of space on the partition mounted on /boot... Jan 08 22:16:00 yates the one that worked up to ssh over usb was downloaded here http://beagleboard.org/latest-images Jan 08 22:21:22 ~seen rcn-ee Jan 08 22:24:21 .seen rcn-ee Jan 08 22:24:24 ,seen rcn-ee Jan 08 22:24:44 whois rcn-ee Jan 08 22:24:52 vacation. Jan 08 22:26:42 ok, can someone remind me please how to use the new device tree in 3.19-rc3 ? Jan 08 22:26:54 I want to set gpio48 to pull up Jan 08 23:10:49 ok yates I was also able to flash BBB-eMMC-flasher-debian-7.7-lxde-armhf-2014-12-19-2gb.img.xz which I downloaded here http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Kernel_Upgrade Jan 08 23:11:58 ssh over usb is working - now going to try and install a newer kernel, which I need in order to get usb midi devices working properly... Jan 09 02:43:37 * ds2 tosses scooby snacks around the channel **** ENDING LOGGING AT Fri Jan 09 02:59:58 2015