**** BEGIN LOGGING AT Tue Mar 10 02:59:58 2015 Mar 10 03:46:11 9 Mar 10 05:58:13 hey guys i have a LCD cape that is causing a conflict, disabling my UART2 in order to give me a hardware button (an Enter key) which i do not want. is there some way to tell the beaglebone who to listen to in the event of a conflict? the dmesg line in question is as follows: bone-capemgr bone_capemgr.9: slot #8: BB-UART2 conflict P9.21 (#3:BB-BONE-LCD4-01) Mar 10 06:00:36 noob question does anyone have the username & password for a fresh debian install on my BBBlack? Mar 10 06:01:28 i forget but looks like the answer here http://elinux.org/BeagleBoardDebian is debian / temppwd Mar 10 06:01:57 or debian /debian, according to another wiki page Mar 10 06:02:07 IIRC, root has no password, just lets you in if you try to login as root Mar 10 06:02:29 thx I'll try em both Mar 10 06:05:31 or all 3 as it were :-) thanks again keyvan Mar 10 06:05:35 np Mar 10 06:07:09 u: root p: Mar 10 06:07:49 ya he left. any answer to my question though ? :] Mar 10 06:08:44 hmm Mar 10 06:09:16 all i have done is active uarts for serial breakout Mar 10 06:09:33 so alas, i have no idea Mar 10 06:09:35 yeah everything is good until you add another cape that wants one of your lines :/ Mar 10 06:09:37 can you disable it on the cape. Mar 10 06:10:10 not sure -- good question, i'll take a look in the datasheet and ask the manufacturer Mar 10 06:12:05 i have a feeling it has something to do with "device trees" Mar 10 06:12:42 looking through http://elinux.org/BeagleBone_and_the_3.8_Kernel seems to mention conflict, but seems written in a haphazard fashion -- i'll have to contribute once i figure this out Mar 10 06:21:53 found a thread discussing exact issue i am having (trying to use UART2 with LCD4) https://groups.google.com/forum/#!msg/beagleboard/O83YNfhqKNc/1tIf6RF03tAJ Mar 10 07:06:17 hii...i want to do kernel learning by making some simple application..can anyone suggest me..P.S. i have no experience in kernel..i have kinda interest in interfacing hardware and software.. Mar 10 07:06:46 so van anyone suggest me some references from where i can start learning ?? Mar 10 07:06:53 *can Mar 10 07:07:29 try the eudyptula challenge, but maybe not on a beagle but e.g. in a VM Mar 10 07:07:30 i have BBB so want to try on that.. Mar 10 07:09:15 i think in BBB i can do some stuff also with GPIO's so thought that there will be learning+application.. Mar 10 07:09:57 learn first the kernel concepts Mar 10 07:10:08 than go to beagle Mar 10 07:10:10 i ahve worked with arduino before.. Mar 10 07:10:15 lol Mar 10 07:10:30 I cooked steak before Mar 10 07:10:38 now I can cook fish Mar 10 07:12:10 haha...got it..:) Mar 10 07:13:49 will eudyptula challenge require VM ?? Mar 10 07:13:51 single tasking bare metal with direct access to all io and memory is alot simpler than multitasking os with kernel Mar 10 07:36:34 ./Nick asd Mar 10 07:37:17 The Eudyptula Challenge site says.. Mar 10 07:37:24 Email little at eudyptula-challenge.org Mar 10 07:37:46 but when i send mail it says email address ""eudyptula-challenge.org"" is not recognized. Mar 10 11:29:53 Hello everyone! Does the latest Debian image listed in http://beagleboard.org/latest-images work on beaglebone black revision B? Mar 10 11:30:31 I couldn't find any info on the wiki so thought I'd check before attempting to flash the image. Mar 10 11:41:29 its a bit big on the emmc, fine on the sd Mar 10 11:51:12 Hmmm good point: I didn't consider 2GB might not be sufficient. I suppose booting from sd would be slower? Are there any other possible issues? Mar 10 11:53:08 not really Mar 10 11:54:17 awesome! good to know! I guess I'll just deal with any issues are they arise then. Mar 10 11:54:25 thanks a lot av500 and bwarff_! Mar 10 12:46:42 is there a way to hook the execution of a certain command after the loading of a specific cape overlay done by capemanager? Mar 10 15:06:13 I've got my BBB revC connected to my little 10" display and nothings showing up. I tested on my 9" with the same resolution and it works fine so it's not the cable or the hdmi port Mar 10 15:06:40 check the EDID Mar 10 15:06:55 if that's garbage you might need to force the resolution or such Mar 10 15:07:00 now my buddy has a factory BBB revC that works with my 10". uname -a seems to say that we're both the same Mar 10 15:07:37 I've been down that road. xrandr says can't open screen Mar 10 15:09:05 I have current SSH connections to both so is there something I could cat and compare for more insight Mar 10 15:11:24 lsb_release -a Mar 10 15:11:34 shows one on 7.4 the other 7.8 Mar 10 15:20:00 sounds like different images then? Mar 10 16:29:20 Computer crashed... did anyone have a suggestion about my HDMI issue when I was gone? Mar 10 16:30:52 TheBestJohn, "uname -a seems to say".... what's `uname -r` show on both? Mar 10 16:39:12 3.8.13-bone50 on the non-working 3.8.13-bone47 on the other Mar 10 16:41:04 48 tweaked the adc and added TIEQEP Mar 10 16:41:20 49: added logibone Mar 10 16:42:10 50: added CBB-Relay cape Mar 10 16:42:12 maybe try with the march 1st image or would that not change anything Mar 10 16:42:24 so... your probally either running a custom *.dtbo... or... Mar 10 16:43:54 I tried wheezy backports while I was down Mar 10 16:44:54 didn't change anything screen wise but probably effected the output of uname Mar 10 16:47:52 TheBestJohn, it wouldnt' affect uname.. nothing depends on that kernel package, i have it setup so you physcially have to choose to upgrade the kernel... Mar 10 16:48:25 I'm thinking your loading a custom dtbo... "dmesg | grep cape"... Mar 10 16:48:36 ah. ok well I'll re-flash to march 3rd and try from there Mar 10 16:49:00 errr march 1st haha Mar 10 16:49:07 i'd first analize the working image.. specially if it's a custom *.dtbo.. no point in starting from scratch.. Mar 10 16:50:03 I don't remember doing anything custom Mar 10 16:50:13 I just grabbed it from recommended build Mar 10 16:50:40 also this literally has nothing installed on it and is essentially a bare build Mar 10 16:51:16 okay, you should be safe then.. Mar 10 17:00:39 I'm killing myself finding and up to date reference for using Ubuntu 14.04 on BBB specifically accessing gpios and i2c . Does anyone have a reference they have found beneficial Mar 10 17:04:24 hmm so not even pre-boot uboot text shows up is that something that should show up regardless? Mar 10 17:04:30 just realized Mar 10 17:05:55 on your lcd? i don't have the u-boot video driver enabled... Mar 10 17:12:33 gottcha Mar 10 17:25:15 I like the light behaviour on the new flash process Mar 10 17:29:58 it helps make it obvious, the system is flashing the eMMC. ;) Mar 10 17:34:50 there's still a slight delay from flashing -> full shutdown, that i'd like to eventually get rid off.. Mar 10 17:41:51 same issue with bone70 Mar 10 17:48:27 maybe I need 3.8.13-bone47 ? Mar 10 17:58:57 hmm looks like the one that works isn't using xrandr Mar 10 17:59:07 as it also comes up as can't open display Mar 10 18:03:50 rcn-ee: also adding video=HDMI-A-1:640x480@60 Mar 10 18:04:01 to the uEnv.txt will not work Mar 10 18:04:15 errr 1280x800@60 Mar 10 18:48:04 Hello. Anyone awake? Mar 10 18:48:44 <_av500_> no Mar 10 18:48:45 <_av500_> zzzzz Mar 10 18:49:26 I have a Beaglebone Black that seems dead. I have never used one before. But I have looked at the LDO U4 (TL5209), but I see no voltage on pin three, which I read is the output of the regulator. Mar 10 18:49:32 _av500_: nice Mar 10 18:49:50 So i suspect that this "dead" boad could be fixable. Mar 10 18:50:32 <_av500_> there is an RMA program Mar 10 18:50:37 <_av500_> for broken boards Mar 10 18:52:00 _av500_: Ohh well, I guess I can not use that, since I did not buy it myself. I was warned it was dead, when I got it. But I figured, that I might just have a look. Mar 10 18:52:13 the PWR led does not light. I measure 5V on the adapter. Mar 10 18:53:25 When I measure pin 2, I have about 1V, that is a bit strange, since that is the input to the regulator. Mar 10 18:57:57 I don't feel anything hot... so where are those four volts going? Mar 10 19:14:29 ok anyone know where to get the factory revC image? Mar 10 19:15:02 TheBestJohn of huluboard or seagull? Mar 10 19:15:04 TheBestJohn, well the 3-01 is the "new" factory image.. Mar 10 19:15:17 looking for bone 47 Mar 10 19:15:34 TheBestJohn, you can grab the old one from here: http://rcn-ee.net/rootfs/bb.org/release/2014-05-14/ Mar 10 19:16:07 see if this allows my display to show up Mar 10 19:16:16 also thank you Mar 10 19:16:22 Didn't you keep the old image? Mar 10 19:17:14 Mine was updated far before I had this screen Mar 10 19:17:24 it's my buddies board that works with it Mar 10 19:17:38 and at this point all I can assume is that it's the os Mar 10 19:18:51 he just purchased his but it had bone47 on it. both mine and his some back with can't open screen on xrandr but his still shows up on my screen Mar 10 19:20:17 and it's on emmc so unless I can make an image from there Mar 10 19:20:58 which it appears I can Mar 10 19:41:01 hmm Mar 10 19:41:09 that's the bone50 kernel Mar 10 19:41:22 I'll try an emmc copy Mar 10 19:44:54 TheBestJohn, bone50 is in the repo.. Mar 10 19:45:34 wget http://rcn-ee.net/deb/wheezy-armhf/v3.8.13-bone50/linux-image-3.8.13-bone50_1.0wheezy_armhf.deb ; sudo dpkg -i linux-image*.deb ; sudo reboot Mar 10 19:46:21 just make sure to copy the dtb's: http://rcn-ee.net/deb/wheezy-armhf/v3.8.13-bone50/3.8.13-bone50-dtbs.tar.gz to /boot/dtbs/3.8.13-bone50/*.dtb as that kernel was before i intergrated the *.dtbs in the deb... Mar 10 19:47:39 Hey all--we're having an issue with our BBBs out in the field we've so far been unable to reproduce in the lab. They will run great for a few days, and then freeze up in a state where only the power LED is lit—all user LEDs are dim. Mar 10 19:48:07 Logs suddenly stop, showing no signs of a kernel panic or other software-related issue. Mar 10 19:48:46 And we know it's not a uboot console issue, since we've already dealt with that and now have a patched version of uboot running. Mar 10 19:49:04 thisisbrians, uname -r ? Mar 10 19:49:45 just a sec... Mar 10 19:50:53 `uname -r`: 3.14.4.1-bone-armhf.com Mar 10 19:51:23 we'd updated kernel in the past to solve what we thought was a USB-related issue Mar 10 19:52:33 your going to have to bug armhf.com on that one, i don't know what he setup for config/patches.. for 3.14.x make sure you atleast have: https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=65ed9236011acdec3b491087ba447e7ef6838230 as that was casuing: http://bugs.elinux.org/issues/127 Mar 10 19:53:37 Our best guess was that it was an issue with a voltage sag or slow voltage ramp up causing the PMIC to prevent a boot, but we can't reproduce it. Mar 10 19:54:13 thanks for the info Mar 10 20:02:21 hmm Mar 10 20:04:41 thisisbrians, compare you config with: https://github.com/RobertCNelson/ti-linux-kernel-dev/blob/ti-linux-3.14.y/patches/defconfig it's what i'm shipping to bbb/bbx15 users... Mar 10 20:05:30 i'll take a look Mar 10 20:15:04 I think I need to give up today Mar 10 20:19:53 rcn-ee: it looks like we're using the vanilla binary from http://www.armhf.com/boards/beaglebone-black/#precise Mar 10 20:20:35 so i don't know how the kernel was built. Mar 10 20:22:03 me neither... hopefully you can just dump "zcat /proc/config.gz" to atleast get his config.. Mar 10 20:22:46 ahh, let me try that. Mar 10 20:23:29 alright, i'll get a diff Mar 10 20:24:25 based on the name, he might got it from arch too: https://github.com/archlinuxarm/PKGBUILDs/tree/50ee29ff7218708631816ac37a0e971ac0687281/core/linux-am33x Mar 10 20:29:10 I'm using pins 31-38 on the P8 header (BB Black) and need to have them pulled down. I have disabled the eMMC and HDMI capes in uEnv.txt (booting from SD card). The board gets stuck during boot however as I think it is still trying to access the pulled-down pins. Has anyone encountered this problem? Mar 10 20:32:45 ovali, does the board still stick if you hold down the boot button before applying power? (the bootrom might be confused with your io restrictions..) Mar 10 20:34:20 rcn-ee: the diff is pretty massive. any interest in taking a look? Mar 10 20:34:47 thisisbrians, stick the config on pastebin.com, i'll look at it to see if anything stick's out.. Mar 10 20:34:58 you got it Mar 10 20:34:59 ovali: are you pulling the pins down externally? Mar 10 20:37:21 ovali: header P8 pins 31-46 are also SYSBOOT pins and must not be driven at boot time Mar 10 20:38:09 see pages 96, 97 and 106 on BBB SRM Mar 10 20:39:03 rcn-ee: http://pastebin.com/e3MPWMyd Mar 10 20:44:30 thisisbrians, pmic is setup correctly... nothing that really stands out.. Mar 10 20:45:00 yes I am pulling down the pins externally Mar 10 20:45:27 oh I see, so there is no way I can use those pins at boot time? Mar 10 20:46:21 rcn-ee: gotcha. we battered the power supply in the lab, and it recovered from every reasonable thing we threw at it, so we're leaning away from that theory now. Mar 10 20:47:42 is it the same power supply that is in the field? Mar 10 20:47:57 yes Mar 10 20:48:51 strange, this looks like a fun one for debugging.. Mar 10 20:50:07 it would appear so... Mar 10 20:52:50 if we are susceptible to the 'random reboots' you linked me to earlier, do you think there's any chance that's leading into a hang? Mar 10 20:56:34 i believe it could be causing part of it.. Mar 10 20:57:07 it's also about the easiest thing to test, as you can use dtc to decompile the *.dtb add that line and rebuild.. Mar 10 20:57:44 a little bit of new info: i just shelled into a BBB with the same kernel that is plugged into a giant UPS, and uptime is telling me 3 days. Mar 10 20:57:48 so i think you're right. Mar 10 20:58:06 (it's been out there for weeks) Mar 10 20:58:47 but, you should see that reboot too... the random reboot would happen on good 5.0 volt power.. if it's up 3 days... Mar 10 20:58:59 maybe another 4 days in the test chambor? Mar 10 21:01:35 oh, i'm saying it has already rebooted. it's been out in the field (plugged into that UPS) for a few weeks already. Mar 10 21:04:44 anyway i can simulate that reboot to see if it causes a hang in the lab? Mar 10 21:05:46 maybe `echo b > /proc/sysrq-trigger`? Mar 10 21:27:45 is there a way to just send HDMI down the pipe without rebooting? Mar 10 21:28:46 I know I said I'd give up for the day but this is so stupid. I don't understand how the BB lost functionality with a kernel update Mar 10 21:37:23 TheBestJohn .. since the kernel is the pivotal configuration and driver code for any linux system .. its quite likely that if there is a subtle difference .. this will highlight any issue Mar 10 21:38:21 particularly .. as kernel major and minor version changes .. the code and methodology can change Mar 10 21:41:06 I see. What I can't seem to find out is the steps the beaglebone takes to figure out if it can display on a screen Mar 10 21:41:20 xrandr fails on both boards I have available Mar 10 21:41:25 it probes via i2c on the hdmi Mar 10 21:41:43 however one board sends video just fine Mar 10 21:42:31 I don't think (could be wrong) xrandr works on the beagle Mar 10 21:42:53 I thought you had to configure it at boot time (again, could be wrong) Mar 10 21:43:16 do you know if theres a way to query the hdmi for resolutions? Mar 10 21:43:37 TheBestJohn .. there may be some basic tools I think that do edid/etc. Mar 10 21:44:57 apt-get install read-edid Mar 10 21:45:02 I'm pretty sure xrandr doesn't work on the BeagleBone Mar 10 21:45:09 and I thought read-edid only worked on x86? Mar 10 21:45:30 https://packages.debian.org/wheezy/read-edid Mar 10 21:45:43 I see arm packages .. Mar 10 21:45:47 guess I was wrong Mar 10 21:47:40 EDID works on whatever can use ACCESS bus so you can use a msp430G2413 to read the EDID data and dump it to a serial port. Mar 10 21:48:45 and it looks like read-edid is installed on base image too Mar 10 21:49:17 a What, GenTooMan !? :p Mar 10 21:50:17 parse-edid works but it looks like I can't get-edid Mar 10 21:52:00 http://www.ti.com/product/MSP430G2413?keyMatch=msp430g2413&tisearch=Search-EN-Everything Mar 10 21:52:36 found this http://beagleboard.org/support/BoneScript/demo_readEDID/ Mar 10 21:53:00 GenTooMan .. I assume that's present on the Beaglebone?! or is it a standard shelf item?! ;) Mar 10 21:53:16 eg. @walmart/target/etc Mar 10 21:53:46 so it sees my screen Mar 10 21:54:26 http://www.ti.com/tool/msp-exp430g2?keyMatch=launchpad%20g2&tisearch=Search-EN-Everything Mar 10 21:56:11 or this for more fun http://43oh.com/ Mar 10 21:57:34 GenTooMan : heh, cool .. about to start work with the stm32 cortex m3/4's shortly Mar 10 21:58:10 ( I need 3x pwm servo drives with a bit of associated processing and qei, etc) Mar 10 21:58:21 Good for small stuff like EDID reader to a serial port kind of stuff. Mar 10 21:58:37 GenTooMan : arduino :) Mar 10 21:59:09 PSoC! Mar 10 22:01:24 arduino is how much? That dev kit is $10 including entire developement software and in situ debugging. Mar 10 22:03:03 GenTooMan .. depends whether you get the cheap clones or not :p Mar 10 22:03:41 PSoC is cool used that before. However it's a bit complicated at times because of the "programmable" hardware. About 12% of the flash is used to configure the hardware I found. Not a big deal but could be frustrating if you need to do more than just simple stuff. Mar 10 22:05:01 i'm searching for a guide that will help me modify the LCD4 dts to disable its conflict with UART2, compile, and install into my beaglebone. anyone know of a guide? or perhaps an easier way to do pin conflict resolution ? thanks Mar 10 22:06:18 You can spin a board for pretty cheap then. Look at 43oh.com you'll find a lot of crazy stuff including USB LS support by bit banging. Mar 10 22:06:26 the goal with the PSoC is to do it in HW Mar 10 22:06:28 nice Mar 10 22:07:28 I don't believe it miracst using the BBB works Mar 10 22:07:39 https://github.com/01org/wysiwidi Mar 10 22:07:47 it actually works Mar 10 22:08:02 bye for now Mar 10 22:11:00 Yes but not everything can be done in hardware. Especially things involving text LOL. Mar 10 22:12:17 Technically speaking it probably isn't impossible to make a simple LCD controller in the PSoC hardware that does 800x480 not sure how fancy one could get but you probably could. Mar 10 22:12:59 everything is HW. SW is imaginary Mar 10 22:13:14 lol Mar 10 22:13:16 Heh programmers these days think the other way around :D Mar 10 22:13:16 I'd disagree Mar 10 22:13:25 800x480 is easily doable on the PSoC Mar 10 22:13:30 main limit is memory Mar 10 22:13:50 LCD controllers can be done with a few GAL chips or a GAL chip + some 7400 parts Mar 10 22:14:12 did a 640x240 one as a class project before Mar 10 22:14:14 Prefer using a CPLD or FPGA myself. Mar 10 22:14:40 when those guys have sane tool license.... Mar 10 22:35:00 Yeah that's the big issue. they use JTAG to program it and they keep most of the programming process proprietary. Altera is nasty that way. Mar 10 22:35:29 Only a handful of devices are close to open. Mar 10 22:42:33 Not sure if it's practical to even have open programmable hardware that would be the first issue. Mar 10 22:56:17 you got that and then there is the so called "free" tools Mar 10 22:56:48 they are more accurately - time limited trials that can be renewed at their descretion Mar 10 22:59:57 GenTooMan: it is practical, but no money in it Mar 10 23:00:37 GenTooMan: you need lots of money to produce millions of chips and be able to wait out until sales ramp up to a sufficient value Mar 10 23:01:02 GenTooMan: nothing you can do with just OSS/OH entusiasts Mar 10 23:01:05 Well you have icharus verilog Mar 10 23:01:18 that's a simulator, that's not hardware Mar 10 23:01:29 and it's verilog, which is an abomination by and in itself Mar 10 23:01:38 Heh :D Mar 10 23:02:15 Hmmm RTL layer might be interesting. Mar 10 23:02:22 RTL layer? Mar 10 23:03:08 REgister Transfer Logic. Essentially one step after the HAL/ hardware abstraction layer. Mar 10 23:03:16 icarus can't synthesize for real HW Mar 10 23:03:40 GenTooMan: i know what RTL is, i just dont understand what you mean Mar 10 23:04:33 Hmmm maybe I am misusing the term? I am sometimes happy to get the right words for what I'm thinking (sad but true). Mar 10 23:05:54 Anyhow you have to create some sort of HAL layer. I have seen ABLE used on a number of synthesis tools however I am not sure what licensing might be require too use that as an intermediate. Mar 10 23:06:30 ABLE is a bus interface, that's something totally different from what RTL is Mar 10 23:06:41 RTL is an abstraction level of how to design logic Mar 10 23:07:19 and HAL is a software layer that unifies access to hardware (which the software people confusingly call an abstraction layer) Mar 10 23:07:36 or in other words, HAL is a library Mar 10 23:07:39 like libc Mar 10 23:12:38 abstraction of hardware so the application layer is the same correct. Mar 10 23:14:05 and sorry I meant ABEL not ABLE (don't kill me .. yet) Mar 10 23:14:37 * KotH calls O.A.D.S to stop the "delivery" Mar 10 23:16:49 Well anyhow yes it is a bit of a problem having a way to go from VHDL/Verilog to (PLT?) Mar 10 23:19:05 Making something to generate something below HL HDL is probably not "hard" but going from that to the target requires some knowledge of the hardware. That is the point of breakdown in such an idea. Mar 10 23:22:57 seems to be an issue of resource planning and allocation then connection. In the end one is still mostly dependent on vendor provided tools it would seem. **** ENDING LOGGING AT Wed Mar 11 02:59:58 2015