**** BEGIN LOGGING AT Thu Oct 08 02:59:58 2015 Oct 08 03:36:13 Hi Folks.. is anyone running the latest rcn-ee 14.04.3/ 4.1.6-ti-r15 image on a beaglebone black with working ethernet? i have rev B, can't get the ethernet to work on mine, it comes up, has link, mii-tool shows link, but i get no traffic flow. 0 rx bytes showing on eth0. i do get "net eth0: phy 4a101000.mdio:01 not found on slave 1" in dmesg but i see that in a lot of places on google and not sure if that is perhaps 'normal' for the second port th Oct 08 04:13:32 lathiat, mdio:01 is not wired... Oct 08 05:48:20 ergh forcing video mode to 720 in 4.1 has no joy Oct 08 05:55:11 that's a pain in the ass :( Oct 08 06:07:17 ok got it working had to change X to use modesetting Oct 08 06:07:23 now i can has refresh rates Oct 08 06:07:25 and res Oct 08 06:11:26 but now there's no audio Oct 08 07:13:20 rcn Oct 08 07:13:21 thanks for that link Oct 08 07:13:26 i can confirm that at least gpio stuff is working on 4.1 Oct 08 07:13:28 https://github.com/notnyt/beaglebone/commit/7caa8f19f017e4a3d489ce8a2e8526c2e36905d5 Oct 08 07:13:31 with that new dtc bin Oct 08 07:13:52 holy shit 4.1 is so nice just gotta get sound running Oct 08 07:19:47 anyone got tips? Oct 08 07:27:27 audio over hdmi that is Oct 08 09:09:13 :( Oct 08 09:42:16 found the solution for the different behavior of the one strobe signal Oct 08 09:43:33 the beaglebone seems to have some issues, testet one other that is fine Oct 08 13:05:34 Hi Oct 08 13:08:51 lo Oct 08 13:09:06 'giorno Oct 08 13:18:27 Hi, after i have create a sd card, with "Angstrom Distribution (BeagleBone, BeagleBone Black - 4GB SD) 2013-06-20", if I need to change driver for audio codec, it's necessary rebuild only the kernel. But for a correct matching, where I can found a original kernel? Oct 08 13:19:37 http://beagleboard.org/linux this procedure? Oct 08 13:30:02 ? any idea Oct 08 13:37:50 does the image come with a manifest? Oct 08 13:42:40 where is found the manifest file? Oct 08 13:43:44 where is the manifest file?* Oct 08 13:46:19 I have no idea how to dig this out. you'd need to look at git tags from years ago in the angstrom git Oct 08 13:47:38 if I'd need to do something with the BBB today, I'd probably either build an up to date angstrom myself or base off a debian console image. Also I'd probably use kernel >=4.1 instead of 3.8 Oct 08 13:47:44 The original image: "Angstrom Distribution (BeagleBone, BeagleBone Black - 4GB SD) 2013-06-20" Oct 08 13:47:57 use 3.8? Oct 08 14:27:29 Angstrom Distribution (BeagleBone, BeagleBone Black - 4GB SD) 2013-06-20 ==>> MOUNT 3.8 KERNEL? Oct 08 14:28:45 Angstrom Distribution (BeagleBone, BeagleBone Black - 4GB SD) 2013-06-20 ==>> from http://beagleboard.org/latest-images Oct 08 14:31:42 2013 Oct 08 14:31:48 pls use latest image Oct 08 14:31:52 not something 2years old Oct 08 14:32:20 Also stop using unsupported distros. ;-P Oct 08 14:36:26 @av500: okay thanks Oct 08 14:42:01 personally .. I would recommend not using smethig 6 months old either .. Oct 08 14:42:49 veremit: I write my own kernel every morning anew Oct 08 14:42:59 excellent av500 +5 Oct 08 14:42:59 so I make sure I run only the freshest SW Oct 08 14:43:39 av500 uses a freshly-forged needle to carefully poke the bits to the right values. Oct 08 14:43:42 I assume you mean from scratch .. and not simply recompile ;D Oct 08 14:43:49 av500: is it apache licensed? Oct 08 14:44:00 *shudder* Oct 08 14:44:30 veremit: no, I go down and get fresh sand from the playground Oct 08 14:44:35 then I heat up the oven Oct 08 14:44:51 by the time the chip is baked I write most of the drivers Oct 08 14:45:00 Posh, too easy. Have to grind rocks up yourself to ge the sand. Oct 08 14:45:10 agmlego: I have kids :-P Oct 08 14:45:12 to do that Oct 08 14:45:16 lol Oct 08 14:45:30 av500: You make new kids every morning too, to grind yer rocks? Oct 08 14:45:31 ;-P Oct 08 14:45:44 I grab random ones Oct 08 14:52:33 roflmao Oct 08 14:52:58 now ya jest gettin carried away :p Oct 08 15:31:14 so on the 4.1 tree, how to get sound output on hdmi? Oct 08 15:45:37 nyt, v4.1.x-ti or v4.1.x-bone? Oct 08 15:48:22 ti's been patching hdmi audio in v4.1.x-ti, so it 'should' work as long as it picks up a the correct mode from the tv.. i don't really have an hdmi-audio tv to test in th elab.. Oct 08 15:48:24 4.1.x-bone Oct 08 15:48:32 ah that's why... Oct 08 15:48:38 last thing i need to get working on 4.1 here Oct 08 15:48:40 hdmi audio is still a "todo" on mainline.. Oct 08 15:48:44 fml Oct 08 15:48:58 does 4.1.x-ti support it and cape manager? Oct 08 15:49:07 I've never bothered with hdmi audio on a desktop even yet .. Oct 08 15:49:17 although .. I don't possess a tv .. so .. Oct 08 15:49:24 nyt, the cape-manager patchset is unified on all my v4.1.x+ branches.. Oct 08 15:49:39 so 4.1.x-ti may support it? Oct 08 15:49:48 was working on 3.8 just lost it going to 4.1 Oct 08 15:50:12 nyt, should have worked on v3.14.x too.. Oct 08 15:50:27 nyt, give: sudo apt-get install linux-image-4.1.9-ti-r20 a try... Oct 08 15:50:43 will do Oct 08 15:50:58 prob a dumb question Oct 08 15:51:00 but no grub on bbb Oct 08 15:51:03 how to switch between kernels? Oct 08 15:51:17 nyt which rootfs are you using? cat /etc/dogtag ? Oct 08 15:51:43 BeagleBoard.org Debian Image 2015-03-01 Oct 08 15:52:11 btw ran you dtc script, thanks for the heads up on that, my gem works, and i pushed the changes to github and rubygems Oct 08 15:52:13 nyt, then your fine.. i have a grub inspired update-initramfs shim to update the kernel.. Oct 08 15:52:30 so whichever kernel is last apt-get installed is the one to boot Oct 08 15:52:38 nyt, correct.. ;) Oct 08 15:52:56 and you can swap between them by updating the uname_r variable in /boot/uEnv.txt Oct 08 15:53:02 ok thats it? Oct 08 15:53:04 thats easy enough Oct 08 15:53:46 christ i hope this works Oct 08 15:53:48 4.x was nice Oct 08 15:53:49 well as long as you spell the uname_r kernel value correct.. on error, it'll mount the eMMC as a flash usb device.. Oct 08 15:54:03 well, running of sd card so Oct 08 15:55:45 rcn-ee .. that's quite a nifty setup Oct 08 15:58:52 I'm trying to write some u-boot scripts in my uEnv.txt file. Is there any way to make variables multiline, ie: escape newline characters so they are ignored? Oct 08 16:00:24 rcn Oct 08 16:00:24 <3 Oct 08 16:00:34 aplroot@backupbone:~# aplay -l Oct 08 16:00:34 **** List of PLAYBACK Hardware Devices **** Oct 08 16:00:34 card 0: Black [TI BeagleBone Black], device 0: HDMI hdmi-hifi-0 [] Oct 08 16:00:34 Subdevices: 0/1 Oct 08 16:00:35 Subdevice #0: subdevice #0 Oct 08 16:01:47 i see's hdmi-audio... does it actually work on your tv thou?? Oct 08 16:01:51 abferm .. you'll hit a 255 char limit sooner or later anyhow Oct 08 16:02:00 unless that got fixed Oct 08 16:02:12 better to run sub-commands I found Oct 08 16:03:13 yes Oct 08 16:03:26 sweet!! thanks for verifing nyt! Oct 08 16:03:51 also Oct 08 16:04:11 nyt Oct 08 16:04:12 ack Oct 08 16:04:17 root@backupbone:~# irb Oct 08 16:04:17 irb(main):001:0> require 'beaglebone' Oct 08 16:04:17 => true Oct 08 16:04:17 irb(main):002:0> include Beaglebone Oct 08 16:04:17 => Object Oct 08 16:04:18 irb(main):003:0> p9_11 = GPIOPin.new(:P9_11, :IN) Oct 08 16:04:20 => # Oct 08 16:04:22 irb(main):004:0> callback = lambda { |pin,edge,count| puts "[#{count}] #{pin} #{edge}"} Oct 08 16:04:24 => # Oct 08 16:04:26 irb(main):005:0> p9_11.run_on_edge(callback, :BOTH) Oct 08 16:04:28 => # Oct 08 16:04:30 irb(main):006:0> Oct 08 16:04:32 irb(main):007:0* [0] P9_11 LOW Oct 08 16:04:36 [1] P9_11 HIGH Oct 08 16:04:38 [2] P9_11 HIGH Oct 08 16:04:40 [3] P9_11 LOW Oct 08 16:04:42 [4] P9_11 HIGH Oct 08 16:04:48 so that works ;) Oct 08 16:05:16 ive only tested the gpio stuff from https://github.com/notnyt/beaglebone Oct 08 16:05:24 i'll get to the rest of it eventually Oct 08 16:05:26 *ahem* pastebin :p Oct 08 16:05:36 or alternatives :) Oct 08 16:08:56 yeah sorry ;) Oct 08 16:09:03 it looked small on my console lol Oct 08 16:09:55 veremit: Still, it would be nice to be able to visualize conditionals, etc. Oct 08 16:10:35 mm uboot is a bit of a pain for actual versatile scripting Oct 08 16:11:04 rcn ty again Oct 08 16:11:21 and for all the works as always Oct 08 16:11:57 so, how far of are we from having X with accelerated graphics and not just fb writes? Oct 08 16:13:25 nyt, the plan is egl + wayland... X is done. ;) Oct 08 16:15:58 ah Oct 08 16:16:04 i hate X Oct 08 16:16:08 ive hated it for decades Oct 08 16:16:33 my arm boards are currently all headless Oct 08 16:17:22 yeah well this one is hooked up to a 60" display in an installation lol Oct 08 16:17:23 mine are too Oct 08 16:17:31 and i think one of my h-bridges roasted my bbb the other day :( Oct 08 16:17:33 but i havent gotten to check it Oct 08 16:17:47 kewl .. that's one application .. but I'd probably use a pi .. *cough* lol Oct 08 16:17:59 yeah but its also hooked up to lots of io Oct 08 16:18:07 ah beagle ftw Oct 08 16:18:09 yes Oct 08 16:18:11 indeed Oct 08 16:18:17 :) Oct 08 16:18:26 if only the video was a tooooooouch better, but i cant complain Oct 08 16:18:38 my ruby lib makes scripting io stuff such cake Oct 08 16:18:40 Pi cape! Oct 08 16:18:41 with threading and everything Oct 08 16:18:43 *eww* lol Oct 08 16:18:49 nah its awesome Oct 08 16:18:53 can dev shit so fast Oct 08 16:19:08 pi2 .. quad core :p Oct 08 16:19:16 same insane price Oct 08 16:19:18 i know :( Oct 08 16:19:19 i have one Oct 08 16:19:26 im using one as a bluetooth audio sink in my kitchen Oct 08 16:19:31 but no emmc/nand/etc Oct 08 16:19:33 have it hooked up to an emotiva a100 and some speakers i built Oct 08 16:19:41 cool ^_^ Oct 08 16:19:43 i dont even use the emmc on the bbb Oct 08 16:20:17 arm chips are still quite application specific Oct 08 16:20:33 rcn-ee.. did the 485 serial patches make it in yet?! :) Oct 08 16:20:57 veremit, haven't seen anything yet.. i thought i had someone working on it... Oct 08 16:21:23 might get to use the beagle as an ethernet io board :D Oct 08 16:21:26 it should be easy with all the helpers.. just need somone with a board.. Oct 08 16:21:44 well .. if I buy one .. I can sort that .. Oct 08 16:22:02 but I'da thought you had contacts .. lol :P Oct 08 16:22:25 the user who fixed in in 3.8.x was interested.. Oct 08 16:23:46 well .. I got like $800 of IO stuck in these machines we build ... could replace it easily with one $40 beagle .. Oct 08 16:24:27 What is being patched for 485? I'm very interested since I use 485 to talk to pretty much everything. Oct 08 16:24:55 abferm .. half duplex needs TXEN signal 'fudging' Oct 08 16:25:29 We just use the rts pin for TXEN. Oct 08 16:25:31 abferm, 8250_omap needs to be extended for 485, the 8250 has a gpio_485 helper... but it needs a user with the connectors to test. ;) Oct 08 16:26:12 rcn-ee. you know if there's a proto cape? Oct 08 16:26:49 rcn-ee: s/connectors/drivers/ Oct 08 16:27:13 zmatt .. the s/w is out there .. and I use rs485 hardware almost daily Oct 08 16:27:15 (in the hardware sense of the word) Oct 08 16:27:25 I use a software rts on my 485 to toggle it, I haven't had any issues, and I've driven it up to 1Mb/s Oct 08 16:27:40 what kernel version? must be old ... Oct 08 16:28:05 I think the omap driver got patched .. Oct 08 16:28:40 I was using my own build of 3.19.0 Oct 08 16:28:41 but normally the txen signal gets squished one end or the other .. I've never tried it .. but I've seen arm implementations Oct 08 16:28:59 abferm: my only concern would be whether the kernel drives makes sure the tx fifo is empty before honoring your rts change Oct 08 16:29:20 got access to a decent DSO .. Oct 08 16:29:35 I let the kernel drivers manage rts, they are just set in the device tree. Oct 08 16:29:37 so .. I'm game .. just need a beagle and headers, and I can crowbar an IO board Oct 08 16:30:17 abferm ,, I thought RTS timed out prematurely .. but I'd be interested to see your setup/code/etc if that's available :p Oct 08 16:33:31 timing may also depend on the specifig 485 transceiver used... We are using TI's SN65HVD11DR Oct 08 16:34:18 bog standard maxim 485 drivers and derivatives here Oct 08 16:34:55 Maxim makes one with auto-TXEN, but we were having issues with it. Oct 08 16:35:16 yea we're using FTDI drivers off a USB connection .. cos they generate the tXEN internally Oct 08 16:36:24 Yeah, FTDIs are awesome in everything but price Oct 08 16:36:34 Is it possible to find an img of a Beaglebone Black with rtai installed or do i have to build the kernel and install it my self? Oct 08 16:37:39 the ft230x's are much cheaper than the 232's Oct 08 16:37:48 and have adequate functionality Oct 08 16:38:13 same driver ofc Oct 08 16:38:17 software-wise Oct 08 16:38:57 nothin special about that TI 485 chip btw :) Oct 08 16:39:14 I think we're on max491's last I saw Oct 08 19:54:55 rcn you here? Oct 08 19:55:00 p8 dtc stuff not working Oct 08 19:55:02 there a trick for it? Oct 08 19:55:21 nyt what you getting? Oct 08 19:55:26 dmesg | grep bone Oct 08 19:55:42 nm Oct 08 19:55:46 my code reuses the dts files Oct 08 19:55:50 and dtb if its compiled already Oct 08 19:55:51 i have old ones Oct 08 19:55:52 need to remove Oct 08 19:56:13 oops .. :p Oct 08 19:57:30 cool all good sorry Oct 08 20:00:01 nyt, you got everything working? ;) Oct 08 20:00:35 not Everything, rcn .. surely thats not possible :p Oct 08 20:01:31 there's always hope! Oct 08 20:01:48 :D I guess .. Oct 08 20:14:07 rcn-ee: appears so ;) Oct 08 20:14:58 nyt, btw, do you detect the patch dtc with your project? or should i move after i build/install the patch dtc in the image build script? Oct 08 20:15:24 i just call 'dtc' Oct 08 20:15:25 so Oct 08 20:15:27 whichever is in the path Oct 08 20:15:30 yours won via local Oct 08 20:15:36 and all worked Oct 08 20:15:40 cool... Oct 08 20:15:53 https://github.com/notnyt/beaglebone/commit/7caa8f19f017e4a3d489ce8a2e8526c2e36905d5 Oct 08 20:15:54 that was it Oct 08 22:55:06 Hey guys, I get this error from zz-uent_txt: Oct 08 22:55:11 sed: -e expression #1, char 24: unterminated `s' command Oct 08 22:55:23 abferm, what system? Oct 08 22:55:55 I've seen that before ... Oct 08 22:55:58 abferm, it's this line i belive: https://github.com/rcn-ee/repos/blob/master/bb-customizations/suite/jessie/debian/zz-uenv_txt#L19 Oct 08 22:56:10 what version of sed? Oct 08 22:56:22 hmmm Oct 08 22:56:25 gnu sed? Oct 08 22:56:42 sed (GNU sed) 4.2.2 Oct 08 22:57:24 humm it's ben 4.2.x for awhile in debian: https://packages.qa.debian.org/s/sed.html Oct 08 22:57:25 rcn-ee: yep, that's the line. The error dissappears when I comment it out. Oct 08 22:57:32 did you switch bash? Oct 08 22:57:39 ? Oct 08 22:58:30 abferm, did you switch /bin/sh to /bin/dash? Oct 08 22:58:41 nope Oct 08 22:59:16 didn't switch it to /bin/bash either ;P Oct 08 22:59:28 default shell is dash sometimes .. Oct 08 22:59:33 i usually assume /bin/bash. ;) Oct 08 22:59:41 can't remember which distro .. but I hit that before .. Oct 08 22:59:46 ubuntu... Oct 08 22:59:55 bloody ubuntu .. Oct 08 23:00:00 I build my own debian images, the default is definately bash Oct 08 23:00:40 guys any ideas on a replacment sed line? ;) Oct 08 23:01:01 just swapping uname_r to a the newly insalled kenrle.. Oct 08 23:01:50 change the : for ... erm, anything? Oct 08 23:02:20 assuming its the delimiter its whining about ... Oct 08 23:03:13 shellcheck.net doesn't complain...... Oct 08 23:03:37 try a ~ or a ^ .. something the shell is gonna ignore Oct 08 23:04:14 + .. can be almost anything in sed .. Oct 08 23:04:32 but / is traditional .. and Oct 08 23:04:41 # is my 2nd choice when there's / involved Oct 08 23:05:11 bash doesn't have an issue with it... I prefer the traditional, and it shouldn't be involved with a uname. Oct 08 23:06:30 s/krenle/kernel Oct 08 23:07:05 veremit: don't forget your /g Oct 08 23:07:22 abferm, https://gist.github.com/RobertCNelson/6e305e67562b15a6f63b Oct 08 23:07:41 don't need to be greedy :P Oct 08 23:07:49 can you try that.. Oct 08 23:08:19 under: /etc/kernel/postinst.d/ Oct 08 23:10:47 working on it... dpkg-reconfigure takes a while Oct 08 23:11:00 same thing Oct 08 23:12:22 pastebin is broken... Oct 08 23:12:41 abferm, yeap... Oct 08 23:12:47 bugger Oct 08 23:12:58 it's funny as it's broken on paste.debian.net and there's a debian bug report... Oct 08 23:13:14 nice Oct 08 23:14:05 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=799580 Oct 08 23:14:05 update-initramfs: Generating /boot/initrd.img-3.19.0-armv7-1 zz-uenv_txt: Updating /boot/uEnv.txt [uname_r=3.19.0-armv7-1] sed: -e expression #1, char 24: unterminated `s' command run-parts: /etc/kernel/postinst.d/zz-uenv_txt exited with return code 1 Oct 08 23:14:30 which is char 24 ... Oct 08 23:15:41 oh .. it is the delimiter .. is $older_kernel bein eval'd correctly? Oct 08 23:16:17 veremit, "correctly"... i'm just checking if it's valid... https://github.com/rcn-ee/repos/blob/master/bb-customizations/suite/jessie/debian/zz-uenv_txt#L15 Oct 08 23:16:28 probally should safe guard it more.. Oct 08 23:17:05 abferm, can you change the "#!/bin/sh -e" -> "#!/bin/sh -ex" so we can see how badly i mis evaluated it? ;) Oct 08 23:17:07 The echo works fine though... Oct 08 23:17:09 abferm.. what does the previous echo say .. does it look right? Oct 08 23:17:16 nvm lol Oct 08 23:17:55 yea -x should tell us what the interpreter is 'seein'g Oct 08 23:19:35 What part of this do you need? Oct 08 23:20:09 abferm everything involving the "older_kernel" value.. Oct 08 23:20:20 it's probally an extra line return... Oct 08 23:20:29 nvm patebin.com is up again, I'll just give you everything http://pastebin.com/Q7x9mLaK Oct 08 23:21:27 lmao good one .. hehehe Oct 08 23:21:35 don't think it should contain al -that- crap lol Oct 08 23:21:37 humm... you have /uEnv.txt and /boot/uEnv.txt to combine. ;) Oct 08 23:21:42 just the 3.1x .. bit Oct 08 23:21:47 the x made it very verbose, err... just ignore my scripts in there Oct 08 23:21:48 how'd that happen. ;) Oct 08 23:22:18 i'll fix the older_kernel to protech against that. ;) Oct 08 23:22:54 i remember this bug from a few months back, we never got down to the issue.. now it make sense.. Oct 08 23:26:24 I still can't qiute see how that arose though .. Oct 08 23:28:54 something to do with the awk bit .. which is where I'm lost .. Oct 08 23:28:54 veremit, it looks like /uEnv.txt and /boot/uEnv.txt got cat'ed together... Oct 08 23:29:34 rcn-ee .. well I'm guessing something -did- that .. Oct 08 23:29:47 yeah but what... Oct 08 23:30:40 ah I can see what you're trying to do .. lol Oct 08 23:31:24 why don't you set the kernel version as a uboot variable ... Oct 08 23:31:35 and use substitution there Oct 08 23:31:55 i'm thinking windows line feeds.. Oct 08 23:32:02 but .. nvm that .. Oct 08 23:32:23 how did you get uenv+uenv :P window's doesn't run that .. (or shouldn't ...!!) Oct 08 23:35:33 rcn-ee: its easy enough to filter out any rogue /r #s from ya /n #s Oct 08 23:35:56 and that be CR not LF :P Oct 08 23:37:15 abferm, can you test this.. https://gist.github.com/RobertCNelson/6e305e67562b15a6f63b Oct 08 23:37:33 replace the older_kernel, i've converted spaces to line feeds to seperate that tail stuff.. Oct 08 23:38:35 actually should grep for "uname_r=" as that should knock out the rest too.. Oct 08 23:40:00 yyyessss... that would do it .. Oct 08 23:40:39 what's nice that's in a deb package, so i can push out an update toight. ;) Oct 08 23:40:58 or just use an option to grep to pull only the first match .. Oct 08 23:41:49 i tried that with tail, while playing with a temp script, that failed too. ;) Oct 08 23:42:16 so i went with a classic swith blanks to line feeds... Oct 08 23:44:13 what are you separating with spaces and linefeeds?? Oct 08 23:44:29 I can only see half the picture here lol Oct 08 23:45:12 nothing wrong with [grep 'uname_r=' /boot/uEnv.txt] surely Oct 08 23:46:02 it looks like it's picking up the next line after that variable.. Oct 08 23:46:48 hmm .. its not *nix in origin? Oct 08 23:47:35 it should be on extX... but it might have got copied from fatx.. u-boot will read it either way.. Oct 09 00:37:33 abferm .. did you sort it? Oct 09 00:38:54 i pushed it out, so a "sudo apt-get update ; sudo apt-get upgrade" will pull in the fix.. and i found a pastebin fix.. so that's buildign right now.. Oct 09 00:39:09 heres something interesting Oct 09 00:39:21 i had a rev b booting no problem with 4.1, gui and everything Oct 09 00:39:24 moved sd card into a rev c Oct 09 00:39:25 no more graphics Oct 09 00:39:37 need to recompile dtb? Oct 09 00:39:42 odd... nope.. Oct 09 00:39:55 yeah no idea Oct 09 00:39:57 maybe something was set in the eMMC of hte b.. that the c doesn't.. Oct 09 00:40:04 oh Oct 09 00:40:07 i updated the bootloader of the b Oct 09 00:40:11 i guess i should do so with the c? Oct 09 00:40:20 give it a shot... Oct 09 00:40:28 kk Oct 09 00:40:47 i did push an update to the bootloader today for zmatt, i2c1 is no longer randomly muxed as i2c... Oct 09 00:41:07 it's used for spidev0, or gpio normally.. but u-boot was setting it for i2c.. Oct 09 00:42:20 rcn-ee: you've probably forgotten already but turns out my ethernet issue yesterday was a switch fault of all things.. it's stopped programming ports despite giving link and thus no traffic flow.. go figure. thanks for confirming the mdio:1 thing was as i thought though, and for the second unwired port. Oct 09 00:42:43 lathiat, good to hear! btw, i've hidden mdio.01 in later kernels.. ;) Oct 09 00:42:48 ah :) Oct 09 00:43:05 now i just need to get a new device tree with adc support.. assuming this random guide i found is still correct for 4.1 Oct 09 00:43:07 we've always had it due to another bug, but that got fixed.. .;) Oct 09 00:43:31 lathiat, https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-ADC-00A0.dts Oct 09 00:43:34 so happy about the 4.x native device tree stuff Oct 09 00:43:37 ah, thanks :) Oct 09 00:43:49 so odd Oct 09 00:43:51 ok trying this now then Oct 09 00:45:04 neg :( Oct 09 00:45:05 any idea? Oct 09 00:45:12 updated bootloader, its running the -ti kernel Oct 09 00:45:16 i can see the screen with edid Oct 09 00:45:20 serial grounding? Oct 09 00:45:20 but no display Oct 09 00:45:37 i assume the same power supply, you just swapped boards.. Oct 09 00:45:45 right Oct 09 00:45:48 everything same Oct 09 00:45:51 just dif board Oct 09 00:45:56 and the board boots with gfx off emmc Oct 09 00:45:57 but not sd Oct 09 00:45:59 (has weird gnd loops at my desk) Oct 09 00:46:32 very odd.. Oct 09 00:47:41 v Oct 09 00:47:42 debian@backupbone:~$ xrandr Oct 09 00:47:43 Screen 0: minimum 320 x 200, current 1280 x 720, maximum 2048 x 2048 Oct 09 00:47:43 HDMI-0 connected 1280x720+0+0 (normal left inverted right x axis y axis) 1209mm x 680mm Oct 09 00:47:43 1280x720 60.0*+ 50.0 59.9 Oct 09 00:47:45 x is running :/ Oct 09 00:47:46 just cant see Oct 09 00:48:35 changing back to bone kernel to see if it changes Oct 09 00:48:53 reboot bug still exists lol Oct 09 00:49:44 yay, pastebinit fix pushed out, but everyone will have to add "s" to http: .pastebinit.xml: http://paste.debian.net/315050/ Oct 09 00:50:06 new images will have the fix auto.. Oct 09 00:51:01 4.1.10-bone16 no joy either Oct 09 00:51:02 * nyt boggles Oct 09 00:51:03 this is so odd Oct 09 00:51:07 and if i take the SD out its fine Oct 09 00:53:07 and x log shows the display Oct 09 00:53:10 and all the edid and modelines Oct 09 00:53:14 and its all there Oct 09 00:53:15 just no output Oct 09 00:54:07 hw mods between rev b and c ? Oct 09 00:54:36 on the sd card that was in the b Oct 09 00:54:44 i updated the kernel, rebuilt dtcs Oct 09 00:54:48 there was something once about display going into standby .. Oct 09 00:55:02 if not connected or something? Oct 09 00:55:43 its connected Oct 09 00:55:46 x sees it Oct 09 00:55:50 shows up in sys Oct 09 00:55:56