**** BEGIN LOGGING AT Thu Jan 19 03:00:02 2017
Jan 19 04:09:18 <ericxdu> I'm having trouble finding how to turn OFF one or more of the LED. All the information is for the old way... which doesn't work with newest images.
Jan 19 04:09:21 <ericxdu> Help please. :-)
Jan 19 05:26:36 <dlech> ericxdu, check out /sys/class/leds/
Jan 19 05:27:19 <ericxdu> dlech: is that still the place to control them? I'll try again
Jan 19 05:27:27 <dlech> echo 0 > /sys/class/leds/beaglebone:green:usr1/brightness
Jan 19 05:27:30 <dlech> as root
Jan 19 05:31:28 <ericxdu> permission denied
Jan 19 05:31:42 <ericxdu> I used sudo
Jan 19 05:37:07 <ericxdu> dlech: using sudo to run that command I get "Permission denied" unfortunately
Jan 19 05:37:17 <zeekhuge> @ericxdu sure that file exists there ?
Jan 19 05:37:59 <ericxdu> zeekhuge: it does exist
Jan 19 05:41:08 <phen> <phen> Hey hey. I'm trying to compile the example here http://derekmolloy.ie/writing-a-linux-kernel-module-part-1-introduction/ on a beagle bone black running ubuntu 16.04 (rt patched kernel 4.4.31)
Jan 19 05:41:08 <phen> <phen> Unfortunately, i'm getting a "WARNING: "__aeabi_uldivmod" undefined" thrown at me by the compiler, which results in a kernel module that cant be loaded due to an unknown symbol
Jan 19 05:41:08 <phen> <phen> After a bit of googling, i found that the function is supposed to be defined in arch/arm/lib1funcs.S, but unfortunately its not actually there
Jan 19 05:41:08 <phen> <phen> Does anyone have any ideas
Jan 19 05:44:19 <phen> On another note, I also want to try my luck with just the 4.4.31 rt headers, does anyone have a link? I cant seem to find them for the life of me
Jan 19 05:52:41 <phen> also, when trying to explore the bb-kernel repo directly, i cant see any files. Does anyone know how I could browse the repo?
Jan 19 05:54:48 <zeekhuge> phen:  The kernel section of this post might help : https://www.zeekhuge.me/post/a_handfull_of_commands_and_scripts_to_get_started_with_beagleboneblack/
Jan 19 05:57:31 <Raghav> Help: I am very new to beaglebone black. The sd card is empty , which image to be flashed in order get the ubunto up
Jan 19 05:58:47 <phen> Raghav: http://elinux.org/Beagleboard:Ubuntu_On_BeagleBone_Black
Jan 19 06:07:35 <phen> zeekhuge: thanks a ton. I installed the appropriate kernel headers and just tried compiling with that. It gives me the exact same issue
Jan 19 06:14:47 <ds2> sigh... once again sudo...just have a proper root account and things will work better
Jan 19 06:18:48 <ds2> running things with the sudo requires a careful consideration of how and which shell evaluates the command.
Jan 19 06:21:34 <phen> Sorry, who are you referring to?
Jan 19 06:28:05 <phen> In any case, I found my problem. http://stackoverflow.com/questions/25623956/aeabi-ldivmod-undefined-when-compiling-kernel-module . Apperently i cant do implicit 64 bit division in kernel space
Jan 19 06:28:27 <ericxdu> ds2: I'd love to... but on bone debian I'm given sudo access and don't know the root passwd
Jan 19 06:28:54 <phen> ericxdu, do sudo su
Jan 19 06:29:02 * ericxdu speaks in old man voice "back in my day, you just logged in as root"
Jan 19 06:29:22 <phen> my capstone prof showed me that yesterday and it blew my mind
Jan 19 06:29:53 <ericxdu> will do
Jan 19 06:46:29 <ericxdu> dltech, ds2, phen: worked!
Jan 19 06:46:41 <ericxdu> I figured there was some sillyness with using sudo in that situation
Jan 19 06:46:59 <phen> yeah, sudo is weird. For example, if you do sudo pwd it returns nothing
Jan 19 06:49:49 <ericxdu> I get that it's meant for quick admin one-liners. I can imagine that use-case doesn't apply to everything you might want to do as root
Jan 19 06:50:12 <ericxdu> So, good to know how to get a proper root shell
Jan 19 06:51:00 <ericxdu> phen: sudo su IS mind-blowing
Jan 19 07:18:27 <ds2> then use a different distro that doesn't force sudo on people
Jan 19 07:19:17 <phen> the problem's been solved lol
Jan 19 07:19:19 <ds2> sudo runs another shell as root so all of a sudden you NEED to know what shell runs what or it leads to all sortsof issues
Jan 19 07:19:37 <ds2> the whole sudo mess keeps popping up
Jan 19 07:19:44 <phen> and "use a different distro" isnt necessarily always an option
Jan 19 07:21:26 <ds2> yocto, angstrom, etc
Jan 19 07:26:19 <ericxdu> sudo is useful. I will not limit myself to some obscure distro that bans sudo for the use-cases it's not meant for
Jan 19 07:28:16 <Frogging101> sudo -i
Jan 19 07:30:41 <ericxdu> ^ works
Jan 19 07:31:07 <ericxdu> Also in Debian can't you usually get root with su?
Jan 19 07:31:24 <phen> yeah
Jan 19 07:31:46 <phen> in fact by default its only su
Jan 19 07:32:03 <ericxdu> In fact, in Debian Jesse I distincly remember having to use su all the time until I specifically set myself up to use sudo
Jan 19 07:32:03 <ericxdu> So... which distro forces sudo on people?
Jan 19 07:32:14 <phen> Ubuntu
Jan 19 07:32:26 <phen> If by "forces on people" you mean "comes installed by default"
Jan 19 07:32:41 <phen> But su always works on ubuntu/debian and their derivatives
Jan 19 07:32:51 <ericxdu> ^ exactly
Jan 19 07:33:31 <ericxdu> Not on Bone Debian though, for some reason
Jan 19 07:33:38 <phen> Weird
Jan 19 07:33:54 <ericxdu> Yeah
Jan 19 07:33:57 <phen> Well if you use a distro that forces you to log into root nobody can accuse you of being a casual so there's that
Jan 19 07:34:31 <ericxdu> Hahaha
Jan 19 07:36:00 <ericxdu> phen: maybe on Bone Debian the user "debian" is set up in the sudoers file but is not an "admin" user so can't use su?
Jan 19 07:37:06 <phen> Yeah, i dont know. Ubuntu has the same problem if you try to use su
Jan 19 07:37:24 <phen> Anywho, im heading to bed, g'night, and thanks again for the help zeekhuge
Jan 19 07:43:03 <Frogging101> anyone can use su
Jan 19 07:43:15 <Frogging101> you just need the password
Jan 19 07:43:17 <Frogging101> or so I thought
Jan 19 07:43:26 <Frogging101> I think fedora has a group called "wheel" or something
Jan 19 07:44:09 <ericxdu> wheel!
Jan 19 07:51:36 <Frogging101> wheel!
Jan 19 07:58:45 <ericxdu> Not sure why that word is used
Jan 19 07:58:50 <ericxdu> for group stuff and python stuff
Jan 19 09:23:39 <ionte> Hi everyone
Jan 19 09:23:59 <ionte> So I'm designing a carrier board for BBBW
Jan 19 09:24:37 <ionte> The BBBW will be powered through the expansion ports.
Jan 19 09:25:51 <ionte> But I'm not sure what the max current of BBBW will be. Of course I understand that it depends on what subsystems will be used and if I connect anything external, USB etc.
Jan 19 09:27:38 <ionte> I've seen power supply recommendations between 1.2 and 2A, but it's also supposed to work when powered through USB (max 500 mA) and that's actually how I'm powering it right now.
Jan 19 09:30:39 <zmatt> USB might be officially max 500 mA, that doesn't mean you can't often get away with drawing more than that
Jan 19 09:31:00 <zmatt> and yes the BBB power consumption is "it depends"
Jan 19 09:33:50 <zmatt> two examples I remember from datasheets is max 100 mA for eMMC, max 100 mA for ethernet, the latter obviously not applicable for BBBW
Jan 19 09:35:58 <zmatt> ericxdu: "sudo su" is silly, use "sudo -s" or "sudo -i" depending on whether you want a clear environment
Jan 19 09:36:16 <zmatt> or just log in as root via ssh
Jan 19 09:42:10 <ionte> zmatt: thanks. suppose i will have to look into each subsystem on the BBBW to try to find a safe max current draw then...
Jan 19 11:48:04 <CarlIsMyName> fds
Jan 19 11:48:10 <CarlIsMyName> help
Jan 19 11:48:11 <CarlIsMyName> ?
Jan 19 11:49:56 <CarlIsMyNAme> index
Jan 19 11:50:32 <rah> lol
Jan 19 11:59:59 <DaStoned> Wow, there's a lot of people here
Jan 19 12:00:48 <DaStoned> Anyway, does anybody know what startup script allocates /var/swap.img and installs it into fstab?
Jan 19 12:01:41 <DaStoned> On first boot from an rcn-ee image there is no swap, and it gets automatically added on second boot.
Jan 19 12:02:07 <DaStoned> I would really like to snip that
Jan 19 14:15:50 <gquere> say I measure 0 Ohms between the VCC and VSS of a flash memory chip, it means it's burnt inside right?
Jan 19 14:21:14 <ericxdu> zmatt: thanks.
Jan 19 14:31:05 <lore4> anyone can help me to understand the OPC addressing please? In the wiki, it is wrote that 0x80000 is the offset from PRU to access the host memory map ( http://elinux.org/PRUSSv2_Memory_Map ). However, in this example ( https://github.com/pgmmpk/beaglebone_pru_adc/blob/master/src/firmware.p ) it is used directly the address 0x44e0d000 to access ADC memory
Jan 19 14:44:18 <eballetbo> by chance, anyone had success trying to wake-up from supspend state via a GPIO?
Jan 19 14:51:03 <lore4> eballetbo: sorry, never tried : \
Jan 19 15:03:55 <gquere> zmatt: got my BBB replacement, i doesn't behave as the BBW did though. when I use your showpin utility I have the same result whether the SPI DTS is activated or not (echo BB-SPI1DEV > /sys/devices/platform/bone_capemgr/slots) although it does create /dev/spidev1.0
Jan 19 15:04:30 <gquere> it doesn't*
Jan 19 15:17:14 <eballetbo> lore4: np, thanks to answer, asking just in case anyone tested, because I think this doesn't work on any kernel
Jan 19 15:18:29 <lore4> eballetbo: welcome, I am sorry that I cannot help : \. BTW, I recall that some days ago someone else was having some issue with SPI.
Jan 19 15:27:09 <gquere> prob me ?
Jan 19 15:27:34 <lore4> anyone has experience of accessing external peripherals from PRU by OCP? I have some doubt about the addressing.
Jan 19 15:27:56 <lore4> gquere: could it be ;)
Jan 19 16:10:56 <zeekhuge> lore4: go ahead and ask, and see if someone knows.
Jan 19 16:18:18 <lore4> zeekhuge: ok, ty
Jan 19 16:21:33 <lore4> I would like to access address 0x44e0d000 which corresponds to a certain  peripheral. Via assembly, it is done by using that address into a register and then using that register as a pointer by instruction LBBO. How can I do the same via C?
Jan 19 16:23:56 <lore4> reading some examples for AM335X (i.e. beaglebone black) and AM57XX (i.e. dunno), I believe I found two ways of doing that but for the AM335X case (i.e. "volatile __far sysPwmss PWMSS0 __attribute__((cregister("PWMSS0", far), peripheral));"), I do not know where the address is taken from as I cannot find PWMSS0 cosntant defined anywhere.
Jan 19 16:24:28 <lore4> Instead, for AM57XX it is used a more straightforward syntax: #define PWMSS0 (*((volatile sysPwmss*)0x4843E000))
Jan 19 16:26:42 <zeekhuge> the former is the better way
Jan 19 16:27:04 <zeekhuge> the address is defined in  the .cmd file
Jan 19 16:27:25 <zeekhuge> like AM335x_PRU.cmd
Jan 19 16:27:35 <zeekhuge> lore4: ^^
Jan 19 16:28:14 <lore4> zeekhuge: thank you! there, there is PWMSS0
Jan 19 16:28:33 <lore4> hence in order to use that address, I should:
Jan 19 16:29:29 <lore4> 1. add something like "TSDADC			: org = 0x44e0d000 len = 0x00000XXX	CREGISTER=??";
Jan 19 16:30:00 <lore4> in am335x_pru.cmd
Jan 19 16:30:45 <lore4> 2. add "volatile __far sysPwmss TSCADR __attribute__((cregister("TSCADR", far), peripheral));" into my header or .c file
Jan 19 16:30:54 <lore4> zeekhuge: am I right?
Jan 19 16:33:14 <lore4> 3. add a sysADC structure like the one defined in sys_pwmss.h in a sys_adcss.h; this structure should also be a map of ADC registers. Right?
Jan 19 16:34:36 <zeekhuge> yes
Jan 19 16:34:39 <zeekhuge> I think so
Jan 19 16:36:01 <lore4> zeekhuge: that's a very good news, zeek!
Jan 19 16:36:12 <lore4> zeekhuge: thanks a lot, man ;)
Jan 19 16:37:08 <zeekhuge> :)
Jan 19 16:37:48 <lore4> tomorrow, I'll try it as soon as possible!!
Jan 19 16:38:09 <lore4> btw
Jan 19 16:38:15 <lore4> what's CREGISTER?
Jan 19 16:39:57 <lore4> I believe it stands for the constant register index...
Jan 19 16:40:50 <lore4> got to go ;) thanks again zeek
Jan 19 16:42:58 * zeekhuge opening up the docs
Jan 19 16:45:13 <zeekhuge> lore4: there ?
Jan 19 16:45:44 <zeekhuge> thats actually the index of the register in the constant table
Jan 19 16:45:58 <zeekhuge> oh ... anyway ... you already got that.
Jan 19 18:06:15 <JHU_roboticist> has anybody else experienced difficulty booting from uSD image on the BBBWireless? Despite using most recent stable image, ensuring bootable flag set on uSD card, pressing and holding boot while powering I cannot seem to boot off the SD card. After initial USR light activity, seeming goes into a static state.
Jan 19 18:16:29 <jkridner_> You should use etcher.io to perform the flashing these days.
Jan 19 18:16:38 <jkridner_> Instructions on the website are unfortunately out-of-date.
Jan 19 18:20:35 <JHU_roboticist> good 'ol dd doesn't cut it these days 'eh? OK, I'll give that a go. (makes me feel like a dinosaur)
Jan 19 18:28:30 <MathOnNapkins> Wait... is that a website?
Jan 19 18:28:40 <MathOnNapkins> crazy days we're living in if so
Jan 19 18:29:05 <MathOnNapkins> (using javascript on a website to write an image to a disk)
Jan 19 18:29:13 <JHU_roboticist> it's a downloadable app
Jan 19 18:29:18 <JHU_roboticist> ~MB
Jan 19 18:29:21 <JHU_roboticist> ~50MB
Jan 19 18:29:26 <tbr> wtf...
Jan 19 18:29:28 <MathOnNapkins> Ah, ok
Jan 19 18:29:39 <tbr> comes with nodejs bundled?
Jan 19 18:35:32 <jkridner> MathOnNapkins: yeah, it is a downloadable app with nodejs bundled.... using a tool called Electron.
Jan 19 18:35:48 <jkridner> bundles a browser and a server in one.
Jan 19 18:36:07 <MathOnNapkins> Yeah I've heard of them, but thanks for giving a little more detail (nodejs and Electron)
Jan 19 18:36:07 <jkridner> a web app, hosted and executed locally. :-_
Jan 19 18:36:10 <jkridner> :-)
Jan 19 18:36:35 * jkridner wants to see BBBlfs ported to Electron and bundled with etcher.io
Jan 19 18:36:49 <jkridner> GSoC students don't seem to be able to figure it out.
Jan 19 18:37:07 <MathOnNapkins> Isn't BBBlfs.... where you flash over the network?
Jan 19 18:37:39 <jkridner> it boots using the on-board ROM over USB.
Jan 19 18:37:49 <MathOnNapkins> iirc I could never quite figure out how to use it or how it worked. The BBB has built in USB devices (part of the SoC) that handled it right?
Jan 19 18:37:54 <MathOnNapkins> * , right?
Jan 19 18:38:01 <jkridner> it then loads a Linux image that serves up the disk as UMS (USB Mass Storage)
Jan 19 18:38:11 <MathOnNapkins> oh right.... it's coming back to me a bit
Jan 19 18:38:14 <Frogging101> that sounds horrible
Jan 19 18:38:36 <jkridner> MathOnNapkins: yes, the ROM acts like a USB RNDIS device and you have to handle several protocols to get it to boot.
Jan 19 18:38:39 <Frogging101> >nodejs and electron to flash an image
Jan 19 18:38:43 <Frogging101> I threw up in my mouth a bit
Jan 19 18:38:51 <jkridner> tftp.
Jan 19 18:38:56 <jkridner> Frogging101: :-D
Jan 19 18:39:05 <MathOnNapkins> It's nice to have a backup method Frogging...
Jan 19 18:39:16 <jkridner> Well, etcher.io is a pretty nice cross-platform app.
Jan 19 18:39:21 <MathOnNapkins> On Linux distro I was on, the built in image writing support was busted
Jan 19 18:39:29 <MathOnNapkins> And Win32DiskImager always felt kind of sketchy
Jan 19 18:39:30 <JHU_roboticist> Flash via etcher.IO app worked first go - cheers.
Jan 19 18:39:35 <Frogging101> what about dd
Jan 19 18:39:43 <jkridner> MathOnNapkins: it is silly to have a whole Linux distro image just to enter UMS mode.... since u-boot does that on its own.
Jan 19 18:39:54 <jkridner> Frogging101: works great if  you know how.
Jan 19 18:40:05 <MathOnNapkins> works bad if you don't
Jan 19 18:40:09 <Frogging101> it's so easy though
Jan 19 18:40:10 <jkridner> Frogging101: I use etcher.io and dd about 50% of the time each.
Jan 19 18:40:11 <MathOnNapkins> reel bad
Jan 19 18:40:24 <Frogging101> dd if=image.img of=/dev/blah
Jan 19 18:40:30 <Frogging101> maybe some other options if there's weirdness
Jan 19 18:40:39 <Frogging101> like bs=
Jan 19 18:40:42 <jkridner> Frogging101: yeah, don't forget 'sudo' and don't ever do /dev/sda1 :-)
Jan 19 18:40:49 <Frogging101> well, duh
Jan 19 18:40:51 <Frogging101> :p
Jan 19 18:40:55 <JHU_roboticist> sure. I usually do bs=1M
Jan 19 18:40:59 * jkridner hardly tries to optimize with bs= anymore.
Jan 19 18:41:22 <jkridner> the /dev/sda1 issue is reason enough to never point anyone to 'dd' on IRC.
Jan 19 18:41:24 <JHU_roboticist> perhaps I could have uncompressed the image first, instead of piping xzcat into dd, but that method has worked for me in the past
Jan 19 18:41:40 <Frogging101> wait, you weren't joking about [13:16:29] <@jkridner_> You should use etcher.io to perform the flashing these days.
Jan 19 18:41:41 <MathOnNapkins> You madman you
Jan 19 18:41:43 <jkridner> etcher.io will unxz for you.
Jan 19 18:41:49 <JHU_roboticist> sure.
Jan 19 18:41:56 <JHU_roboticist> but i'm speaking of dd.
Jan 19 18:42:04 <JHU_roboticist> i'm just curious as to why it didn't work
Jan 19 18:42:16 <jkridner> Frogging101: indeed, incredible lack of irony/sarcasm in that suggestion.
Jan 19 18:42:22 <MathOnNapkins> Who knows. I always decompress images I'm going to write
Jan 19 18:42:27 <Frogging101> but this is just so absurd and cancerous
Jan 19 18:42:29 <MathOnNapkins> before writing them I mean
Jan 19 18:42:42 <MathOnNapkins> Just b/c I don't want to chance something going wrong. It's probably not warranted paranoia though
Jan 19 18:42:42 <Frogging101> it's like that microcontroller with built-in javascript
Jan 19 18:42:48 <Frogging101> like, why
Jan 19 18:43:13 <MathOnNapkins> I have a reflexive dislike of things that use javascript, but this kind of thing is useful enough to win me over
Jan 19 18:43:34 <MathOnNapkins> Much like the Cloud 9 IDE. I was shocked at how useful it was once I gave it a chance
Jan 19 18:45:32 <JHU_roboticist> incorporates git, no?
Jan 19 18:46:12 <JHU_roboticist> (as in you can commit changes right from the IDE)
Jan 19 18:46:15 <MathOnNapkins> Cloud 9? I don't use git so I wasn't looking for that specifically but I wouldn't be surprised
Jan 19 18:52:24 <jkridner> JHU_roboticist: I haven't noticed git integration. I will edit with it often and just use the built-in command-line for git.
Jan 19 18:52:53 <jkridner> JHU_roboticist: however, I'm just as likely to use 'vi' within the command-line window and make the command-line window full screen within the browser.
Jan 19 18:53:14 <jkridner> Frogging101: yes, I could use 'ssh' for that. :-)
Jan 19 18:53:49 <Frogging101> oh these people are responsible https://resin.io/
Jan 19 18:53:52 <Frogging101> "Built for IoT"
Jan 19 18:53:54 <Frogging101> well that explains everything
Jan 19 18:53:56 <Frogging101> :)
Jan 19 19:06:06 <JHU_roboticist> @jkridner: thanks for the help, appreaciate it.
Jan 19 19:24:31 <gregoryfenton> Hi all, I have an issue running apt-get upgrade on bbb latest debian. All goes fine but it errors out at node-gyp rebuild. Error output at http://pastebin.com/jDkAKypT
Jan 19 19:26:04 <jkridner> gregoryfenton: a longer back-log would be useful. I suspect you'll need to reach Robert on https://beagleboard.org/discuss as he worked on the c9-core-installer package.
Jan 19 19:27:58 <gregoryfenton> Thanks, I can reflash and re-run the apt-get upgrade and tee the complete output to a file if that would help.
Jan 19 21:21:11 <GregoryFenton> Did my last message saying thanks go through a moment ago? If not, thanks, Robert did indeed provide the solution. As I suspected one of the packages had gone end of life. Much appreciated all.
Jan 20 02:05:15 <diji> Would anyone here be able to help me troubleshoot display issues with a BBBW?
Jan 20 02:09:06 <diji> hey zmatt would you be able to help me out more?
Jan 20 02:14:36 <zmatt> I'm not a personal assistant.  just ask your question and with some luck someone may be able to help you, perhaps even me
Jan 20 02:15:22 <diji> heh sorry, i just know that you know your stuff
Jan 20 02:15:46 <diji> i found a fix for a samsung monitor like mine
Jan 20 02:16:02 <diji> http://elinux.org/Beagleboard_talk:BeagleBoneBlack_HDMI
Jan 20 02:16:20 <diji> but i don't have a /var/run/gdm/ directory
Jan 20 02:19:50 <zmatt> you tested that resolution?
Jan 20 02:20:45 <diji> no, but i figured I could force another resolution in its place
Jan 20 02:21:01 <omar> hello there i need help with a robot
Jan 20 02:21:21 <zmatt> I think last time I mentioned you can force resolution via a video= setting in /boot/uEnv.txt
Jan 20 02:21:27 <zmatt> there's even an example for that in there iirc
Jan 20 02:21:48 <omar> can u help me
Jan 20 02:21:50 <zmatt> but it makes sense to first test resolutions
Jan 20 02:23:23 <zmatt> omar: saying "i need help" is not likely to get a useful response.  try to ask a concrete question (and be patient, it can take a while for someone who might know to respond)
Jan 20 02:23:58 <diji> zmatt: My uEnv.txt http://pastebin.com/39mP0uPe
Jan 20 02:24:09 <diji> i have tried a few resolutions
Jan 20 02:24:53 <diji> I also use "fbset" after i reboot to see if it changed
Jan 20 02:25:14 <diji> Don't know if that is showing me the right info though
Jan 20 02:25:34 <zmatt> I'm fairly certain fbset does nothing useful with the tilcdc kernel driver
Jan 20 02:26:02 <diji> i don't know, i just know it shows a resolution
Jan 20 02:26:02 <zmatt> not 100% sure though
Jan 20 02:26:12 <zmatt> yeah it can see that
Jan 20 02:27:52 <diji> see in uEnv.txt i have the resolution set to 640x480@75e, but fbset shows 1280x720
Jan 20 02:29:11 <diji> i would like to use xrandr, but it only throw an error
Jan 20 02:29:29 <diji> "Can't open display"
**** ENDING LOGGING AT Fri Jan 20 03:00:02 2017