**** BEGIN LOGGING AT Wed May 13 02:59:59 2015
May 13 06:32:26 <Tlin> Hi, I updated my BBB to Debian 8.0 and now I am trying to activate CAN1. I am following this tutorial http://www.embedded-things.com/bbb/enable-canbus-on-the-beaglebone-black/ but I am unable to execute "echo BB-DCAN1 > /sys/devices/bone_capemgr.*/slots" because there is no "bone_capemgr". How can I install/run bone_capemgr on Debian 8?
May 13 06:48:41 <Tlin> Or is there another way to activate can1 on Debian 8?
May 13 07:30:41 <mrpackethead> any TI Staff here?
May 13 09:40:09 <woglinde> hm is jailhouse running on the bbb https://github.com/siemens/jailhouse?
May 13 09:44:16 <av500> jawat?
May 13 09:45:26 <av500> woglinde: my coworker says: its from Siemens, dont touch it!
May 13 09:45:56 <LetoThe2nd> i've seen it in action, but the siemens guys are heavily x86 focused in this thing
May 13 09:46:12 <av500> does the am335x a8 have virt stuffs?
May 13 09:46:26 <LetoThe2nd> my last word from them was that arm support is possible and on the roadmap, but not without VT extensions, e.g. A15
May 13 09:46:57 <woglinde> av500 I thought you are the cpu expert
May 13 09:47:32 <LetoThe2nd> av500: nope
May 13 09:47:56 <LetoThe2nd> woglinde: so in short: "no"
May 13 09:53:17 <av500> woglinde: me, I know nothing
May 13 09:54:55 <philenot1ound> av500: You are Jon Snow?
May 13 10:02:10 <av500> philenot1ound: I would not know
May 13 10:12:53 <gurpartap> how to calculate physical pin number of gpio pins. I want to read input from a switch
May 13 11:49:09 <CityLights1> tbr: hi
May 13 11:50:06 <tbr> CityLights1: yes?
May 13 11:50:19 <CityLights1> hi, I seek some advice
May 13 11:50:54 <CityLights1> I wrote a program that will serve access point on the BBB
May 13 11:51:17 <CityLights1> I packaged it so it will be easy to install
May 13 11:51:52 <CityLights1> but as I add a new wireless virtual device I add some lines in the /etc/network/interface file
May 13 11:52:15 <CityLights1> my question, what is the proper way to add five lines there?
May 13 11:52:59 <CityLights1> If I use a patch, then if the user has a different interfaces file then the default, install will fail - right?
May 13 11:53:36 <vvu> any link to it CityLights1 ?
May 13 11:53:47 <CityLights1> hi vvu
May 13 11:54:03 <CityLights1> let me post this file now:
May 13 11:55:08 <vvu> okei
May 13 11:55:11 <CityLights1> http://paste.ubuntu.com/11112426/
May 13 11:55:28 <CityLights1> I am reffering to the last five lines
May 13 11:55:41 <CityLights1> so I can append them
May 13 11:56:09 <av500> so what is the problem?
May 13 11:56:35 <av500> you dont need to patch, you can simply append
May 13 11:56:48 <av500> cat five_lines >> /etc/network/interface
May 13 11:57:06 <CityLights1> is this the professional solution?
May 13 11:57:15 <CityLights1> I am new to publishing a solution
May 13 11:57:23 <CityLights1> at work/home its easier
May 13 11:57:43 <av500> I was told I am not professional :)
May 13 11:58:05 <CityLights1> I also have a second issue. I want to overwrite the /etc/defualt/hostapd file
May 13 11:58:45 <CityLights1> if I add it to my deb file, during install the install fails as it conflicts with the hostapd install
May 13 12:00:07 <av500> CityLights1: thats more a question for #debian
May 13 12:00:27 <CityLights1> o ok
May 13 12:18:26 <ogra_> CityLights1, not sure debian supports that yet, but /etc/network/interfaces.d/ should work (just put your config snippets in there)
May 13 12:18:45 <CityLights1> sweet
May 13 14:11:43 <Posterdati> hi
May 13 14:12:40 <Posterdati> please I'm updating using opkg, but seems that it stucks on kernel_dev ipk package, any hints? Thanks
May 13 14:13:21 <vvu> update to debian :)
May 13 14:13:43 <Posterdati> vvu: shall I change the sd card to a more capable one?
May 13 14:13:51 <av500> switch to debian
May 13 14:13:53 <vvu> flash debian on it
May 13 14:14:27 <Posterdati> shall I take it from debian.org
May 13 14:14:28 <Posterdati> ?
May 13 14:14:41 <vvu> http://beagleboard.org/latest-images
May 13 14:14:46 <vvu> you have here
May 13 14:15:15 <Posterdati> ah nice, I will write this image with an sd adapter
May 13 14:16:24 <Posterdati> may I wipe off the windows partition too?
May 13 14:16:47 <vvu> use http://sourceforge.net/projects/win32diskimager/
May 13 14:16:52 <vvu> this will do the things for you
May 13 14:17:18 <Posterdati> I do not use windows here... I will dd the image on the sd card directly
May 13 14:17:28 <vvu> remove everything if so
May 13 14:22:58 <Posterdati> thanks people, I'm going to use an 8GB empty sd card
May 13 14:26:39 <Posterdati> is it possible to expand the installed partition to 8GB?
May 13 14:58:54 <Posterdati> is anyone interested in Common Lisp programming on the beagle board?
May 13 15:06:51 <thurgood_> Posterdati: I'm not sure about expanding a mounted active partition, but if you put the card in a desktop/laptop that can read the partition... yes
May 13 15:08:32 <Posterdati> USR1 and USR3 are lit
May 13 15:09:17 <Posterdati> mmmh it seems it's powercycling
May 13 15:17:08 <Posterdati> not working
May 13 15:26:28 <Posterdati> I flashed the .img file with dd and no bs=1m option
May 13 16:51:26 <zmatt> rcn-ee: crap, I had no expected that  CAPE(s)  would be parsed as formatting syntax in that bug tracker thing :/
May 13 16:51:31 <zmatt> *not expected
May 13 16:52:19 <zmatt> thurgood_: you can live-expand a mounted ext4 filesystem yes
May 13 16:53:28 <thurgood_> cool, if a bit risky
May 13 16:53:41 <zmatt> thurgood_: not afaik
May 13 16:54:03 <thurgood_> I guess it depends on how it stores the file allocation info
May 13 16:54:19 <thurgood_> which I don't know enough about in ext4
May 13 16:55:57 <zmatt> it's common to use when the backing storage uses LVM or is otherwise virtualized
May 13 16:56:49 <zmatt> apparently kernel 3.3 also added online shrinking
May 13 17:41:01 <abhi_> hi
May 13 17:42:06 <abhi_> any one is there????
May 13 17:43:10 <vagrantc> probably, but usually best to just ask a question and wait patiently
May 13 17:43:31 <abhi_> I need a help How to start the work with BeagleBoard-xM ??? and how to install iso image ???
May 13 17:45:56 <abhi_> If any videos for installing the os in BeagleBoard-xM just provide me.....
May 13 17:46:50 <abi_> I need a help How to start the work with BeagleBoard-xM ??? and how to install iso image ???
May 13 17:47:09 <abi_> If any videos for installing the os in BeagleBoard-xM just provide
May 13 17:48:54 <rcn-ee> abi_, what distrubution are you trying to install?
May 13 17:52:02 <abi_> Here one problem is it is not going in to boot also ???
May 13 17:53:55 <Posterdati> I flashed the debian image on the beaglebone rev. b and it won't work: USR1 and USR3 stay on
May 13 17:54:27 <Posterdati> I flashed it on an 8GB sd card
May 13 17:54:28 <abi_> And one think is I'm totally new to this  BeagleBoard-xM   and i want to install the Linux os in this board??
May 13 17:55:58 <abi_> Tell me how to the process for this and provide me any guide
May 13 17:55:59 <rcn-ee> Posterdati, do you have a usb-serial devie to read the boot log?
May 13 17:56:26 <abi_> No
May 13 17:57:18 <rcn-ee> abi_, sorry i don't understand, maybe english isn't your native language: ready to go image for the xM: http://elinux.org/BeagleBoardDebian#BeagleBoard_xM
May 13 17:57:54 <Posterdati> rcn-ee: an usb serial adapter...
May 13 17:59:28 <abhi_> sorry no i'm not having
May 13 18:00:25 <rcn-ee> Posterdati, if you have one, plug it in and it'll tell you over serial what's actually happening then "USR1 and USR3 stay on.."
May 13 18:00:56 <Posterdati> now USR0 and USR1
May 13 18:03:02 <rcn-ee> Posterdati, i know your trying to be helpful, that doesn't actually tell us anything.. ;sorry..
May 13 18:03:05 <abi_> ok
May 13 18:04:27 <abi_> Can you send me any video for installing process i will go with that one.........
May 13 18:05:05 <Posterdati> rcn-ee: 9600 8n1 ?
May 13 18:05:17 <rcn-ee> Posterdati, 115200 8 n1
May 13 18:05:57 <rcn-ee> abi_,  a video for 3 shell commands? wget file ; uxz file ; dd file... your kidding right?
May 13 18:06:10 <thurgood_> abi... do you have a linux distro that you want to install, or are you just looking for anything that will work?
May 13 18:07:29 <Posterdati> rcn-ee: not working
May 13 18:08:42 <thurgood_> Posterdati: you need an ftdi chip in the cable for the bbb
May 13 18:09:51 <Posterdati> Prolific Technology Inc.
May 13 18:10:26 <Posterdati> so no debug, good!
May 13 18:12:42 <mistawright> Posterdati, what kind of prolific serial cable are you using? I use one myself at home with no issues
May 13 18:13:26 <Posterdati> mistawright: XURS232_FC
May 13 18:14:30 <mistawright> I use one of these guys fine: Prolific Technologies PL2303 but it was recommended on adafruit and few other sites
May 13 18:15:40 <mistawright> so rcn-ee I installed the kernel you built yesterday. do i need to edit my uEnv.txt to enable the dtb you mentioned as well?
May 13 18:16:17 <mistawright> rcn-ee, I installed the kernel you built yesterday. do i need to edit my uEnv.txt to enable the dtb you mentioned as well?
May 13 18:21:25 <Posterdati> mistawright: mine is not working at all
May 13 18:22:15 <Posterdati> mistawright: anyway even the original sd card has a corrupted system, because I tryed to upgrade it and got no space left on device... gantastic
May 13 18:41:08 <Posterdati> nice
May 13 18:41:21 <Posterdati> crippled beaglebone...
May 13 18:47:45 <mistawright> ive got one with a borked ethernet port that I would love to use. but it doesnt get passed trying to initialize the ethernet port
May 13 18:48:30 <tbr> mistawright: tried building a kernel without the ethernet driver?
May 13 18:48:47 <tbr> or just removing ethernet from the DTB
May 13 18:51:25 <mistawright> tbr, im not able to get the board to the point of booting currently to be able to do anything. Would I need to use an emmc flasher and edit what dtb's it uses?
May 13 18:57:48 <Posterdati> is beaglebone rev. 6 not compatible with debian image on the beagleboard site?
May 13 18:58:16 <mistawright> are you using a 4gb or 2gb image
May 13 18:58:26 <Posterdati> 4
May 13 18:58:36 <mistawright> assuming you are trying to flash the emmc that wont work
May 13 18:58:47 <Posterdati> no
May 13 18:58:50 <mistawright> booting off of sd i dont believe should be an issue
May 13 18:59:04 <Posterdati> I flashed the sd card with another linux box
May 13 19:01:46 <tbr> mistawright: step one, do that with an SD  card. step two use the system on the sd to boot, mount the emmc, fix that the same way as SD.
May 13 19:02:11 <Posterdati> uh!
May 13 19:02:26 <zmatt> lol... "It’s easy to get frustrated when your program has errors. I try not to think of them as errors, though. I try to think of them as the pathetic attempts of a socially inept non-native English speaker (your computer) asking for help."
May 13 19:14:13 <mistawright> tbr, so what you are recommending is that i edit eh am33xx dts and remove the ethernet port and build and use that dtb?
May 13 19:15:21 <tbr> mistawright: that would be probably my first attempt, yes
May 13 19:20:52 <zmatt> mistawright: borked in what sense?
May 13 19:22:11 <rcn-ee> zmatt, he has one of those boards that requires the cspw driver to "find" the port used.. the hack we've been using, has issues with the new "overlay" stuff, i need to take a moment and fix it.. (hard kernel crash)
May 13 19:22:31 <zmatt> that's still a problem that puzzles me greatly
May 13 19:23:02 <zmatt> like, the phy address strapping option has  1. internal pull-down in the phy  2. internal pull-down in the am335x  3. external pull-down
May 13 19:23:11 <zmatt> how the fuck can the phy ever manage to sample it high
May 13 19:23:23 <zmatt> regardless of precise reset timing
May 13 19:23:30 <rcn-ee> it's so odd.. the random moving strapping address... it seems to happen more often when booting with microsd..
May 13 19:24:07 <zmatt> in any case, the mdio controller already scans for phy address, and you can reconfigure the strapping options via mdio
May 13 19:24:14 <zmatt> so u-boot could easily fix it
May 13 19:24:50 <zmatt> but I'm still wondering what the root cause is here
May 13 19:25:20 <rcn-ee> the last guy i was helping started down the u-boot fix route, then he found my kernel patch, and yeah... (losts a chance there..)
May 13 19:25:37 <zmatt> kernel shouldn't need any patching
May 13 19:25:49 <zmatt> (it could use a rewrite of the cpsw driver, but for other reasons :P )
May 13 19:26:09 <rcn-ee> correct. ;)  it's caused by butterflies in the upper atmosphere, reflecting photons...
May 13 19:26:29 <ds2> maybe that PHY is just plain F'ed?
May 13 19:26:41 <zmatt> ds2: that sounds like a reasonable option
May 13 19:26:53 <zmatt> but it's also fussy about its power/reset timing, no idea if its constraints are actually met
May 13 19:26:57 <ds2> it seems to be a non issue on other PHY's
May 13 19:27:09 <zmatt> its reset should have been on a GPIO, not connected to the system reset pin
May 13 19:27:29 <ds2> how often do you need to reset a PHY?
May 13 19:27:34 <zmatt> normally, never
May 13 19:27:38 <zmatt> this phy requires it however
May 13 19:27:47 <zmatt> (per datasheet)
May 13 19:28:03 <rcn-ee> it is a smsc part, wonder if it's pre / post microchip... and what that means for micrel phy parts..
May 13 19:28:25 <zmatt> I'm used to PHYs simply working directly from power-up
May 13 19:28:42 <ds2> so putting on a GPIO would seem like a waste
May 13 19:28:58 <zmatt> ds2: if you switch to a different PHY...
May 13 19:29:02 <ds2> you got route out that line from the nasty BGA and spend time in SW to toggle it...
May 13 19:29:08 <ds2> zmatt: :)
May 13 19:29:46 <zmatt> ds2: but hooking it to nreset is also annoying, there's no need to reset the phy if you reboot the system
May 13 19:29:49 <zmatt> in fact I'd rather it not
May 13 19:30:16 <zmatt> this will become even more relevant on the x15 which has both ports of the switch subsystem
May 13 19:30:30 <ds2> both ports?
May 13 19:30:30 <zmatt> then you definitely do _not_ want the phy to be reset during a warm system reset
May 13 19:31:07 <zmatt> ds2: yes, many recent SoCs (centaurus, subarctic, aegis, vayu) have an ethernet switch subsystem with two external gigabit ports
May 13 19:31:13 <mistawright> zmatt, that when i attempt to boot the board it stops booting when it begins to initialize the ethernet port
May 13 19:31:15 <zmatt> beaglebone unfortunately only makes one usable
May 13 19:31:33 <zmatt> mistawright: on many distros it has a very long timeout on getting an IP
May 13 19:31:48 <zmatt> mistawright: I actually dumped ifupdown and switched to systemd-networkd
May 13 19:31:53 <mrpackethead> does anyone else find Ti's E2E to be the most fustrating "support" system
May 13 19:31:59 <ds2> zmatt: the AM335x has that
May 13 19:32:06 <zmatt> ds2: yes
May 13 19:32:34 <zmatt> ds2: in addition to the ethercat support in the am3359 (which uses different pins however)
May 13 19:32:48 <ds2> that's a little different
May 13 19:32:59 <zmatt> yes, that goes via the PRUs
May 13 19:33:08 <zmatt> the switch subsystem is however autonomous
May 13 19:33:31 <rcn-ee> mistawright, 4.1.0-rc3-bone1 has the disabled eMMC/hdmi overlay: am335x-boneblack-overlay.dtb and a very important fix from zmatt, been using it test lcd3/4/7's on the bbb today..
May 13 19:33:46 <zmatt> once initialized it can do its job without software looking after it
May 13 19:33:56 <zmatt> and even remain operational during warm resets
May 13 19:34:23 <ds2> let's do a kickstarter for a BeagleBone/Fixed then :D
May 13 19:34:36 <zmatt> (though there's an erratum there which means you do need to "unplug" the internal host port before performing a system reset if you want to use reset isolation)
May 13 19:34:47 <zmatt> which is not actually a big burden though
May 13 19:35:19 <zmatt> the linux driver makes all this totally unusable though, since the switch driver and host ethernet driver are completely interwoven for no good reason
May 13 19:35:31 <zmatt> so it ends up reseting the switch anyway
May 13 19:35:46 <zmatt> ditto for the driver in u-boot
May 13 19:36:40 <ds2> doesn't this make the USB stuff look better now?
May 13 19:36:41 <ds2> :D
May 13 19:36:59 <zmatt> but I have a dm814x board with baremetal code where it works fine... switch is initialized during early boot, and the host ethernet driver can be (re)initialized at any later time without affecting the switch
May 13 19:37:06 <zmatt> no, since USB is broken in hardware
May 13 19:37:07 <zmatt> :P
May 13 19:37:30 <ds2> your claim is the ehternet stuff is broken in hw on the BBB ;)
May 13 19:37:33 <zmatt> and as a protocol in general
May 13 19:38:22 <zmatt> that's choice of PHY and board design... and perhaps more "suboptimal" than "broken" (except for that weird strapping option thing)
May 13 19:38:34 <zmatt> I'd still much rather deal with all that than with usb
May 13 19:38:56 <zmatt> reading the usb chapter of the TRM makes me want to jump off a cliff
May 13 19:40:01 <zmatt> and the problem is that two gigabit ethernet ports obviously take up more space and add to the cost... the idea of the BBB was to make it low-cost, and sacrifices were made
May 13 19:41:38 <zmatt> (btw, in general, whatever you do: use MII or RGMII, avoid RMII and GMII
May 13 19:41:40 <zmatt> )
May 13 19:46:23 <mistawright> zmatt, the timeout i have definitely noticed. It takes my beaglebone almost a minute or two to boot and connect to wifi
May 13 19:46:32 <mistawright> rcn-ee, will check it out now
May 13 19:56:52 <stt_michael> zmatt, I saw all the salient signals are brought out of the am3359 for rgmii for GB eth. Its just the PHY side that would need bodging :)
May 13 19:57:23 <stt_michael> zmatt, the difference in cost for the GB lan chip vs the 100mb is insane.
May 13 19:57:45 <stt_michael> zmatt, but there could be a good alternative to the uChip one chosen.
May 13 19:59:11 <mrpackethead> stt_michael: what is your definition of insane
May 13 20:07:13 <zmatt> stt_michael: all the signals are there so it would be relatively minor surgery on the pcb design... mainly some extra checking that the more stringent requirements for rgmii are met
May 13 20:07:52 <mrpackethead> why do you need gig ethernet?
May 13 20:09:34 <mistawright> rcn-ee, with the kernel update should I be able to echo my dtbo into slots?
May 13 20:11:15 <zmatt> mrpackethead: because speedier?  tbh I'm not sure I need it, 100 Mbit may be fine.  when using the switch subsystem however you'll more likely want Gb ethernet since the ports may be carrying traffic for multiple hosts
May 13 20:11:34 <zmatt> but 100 Mbit was probably the right choice for the BBB
May 13 20:11:57 <zmatt> especially if the price difference in PHYs is indeed large (I have no idea myself)
May 13 20:12:24 <mrpackethead> why not 10GB
May 13 20:12:32 <mrpackethead> lol.
May 13 20:12:39 <zmatt> because it's not supported by the am335x
May 13 20:12:44 <zmatt> :P
May 13 20:12:52 <mrpackethead> then you should replace that too
May 13 20:13:07 <mrpackethead> and put a faster one in.
May 13 20:13:16 <_av500_> zmatt: there was some hw issue with gigE on the BBB
May 13 20:13:19 <mrpackethead> and more storage and more memory while your at it.
May 13 20:13:26 <_av500_> see the mailinmg list for geralds explanation
May 13 20:13:29 <rcn-ee> mistawright, haven't figured that out yet..
May 13 20:13:30 <zmatt> _av500_: external reference clock may be needed
May 13 20:13:36 <_av500_> i forgot
May 13 20:13:39 <_av500_> whatever it was
May 13 20:13:52 <_av500_> it did not justifiy the expense for a $45 hobby board
May 13 20:14:08 <zmatt> they hooked up the internal reference clock to the core PLL, which is not a low-jitter PLL
May 13 20:14:27 <zmatt> in particular, it doesn't have sufficiently low jitter for anything but MII
May 13 20:14:56 <_av500_> as said, it does not matter
May 13 20:15:05 <_av500_> I see now point in gigE on the BBB
May 13 20:15:17 <_av500_> what what you do with all the data anyway?
May 13 20:15:19 <zmatt> I agree
May 13 20:15:26 <_av500_> pipe it to /dev/null faster?
May 13 20:15:32 <zmatt> yeah!
May 13 20:15:44 <_av500_> its like blinking the cursor faster
May 13 20:15:50 <_av500_> impressive :)
May 13 20:18:21 <zmatt> there are other boards with an am335x anyway, if you're willing to spend a bit more... like this thing => http://www.ti.com/tool/tmdssk3358
May 13 20:18:27 <_av500_> sure
May 13 20:18:40 <_av500_> but now that $9 is the new benchmark ....
May 13 20:18:57 <zmatt> what's $9 ?
May 13 20:19:03 <_av500_> a computer
May 13 20:19:05 <_av500_> THE computer
May 13 20:19:12 <_av500_> you live under a rock?
May 13 20:19:17 <zmatt> mostly yes
May 13 20:19:20 <_av500_> ah
May 13 20:19:32 <_av500_> google "$9 computer" :)
May 13 20:20:27 <mrpackethead> a pic10f is cheaper than that, abou $0.80
May 13 20:20:34 <zmatt> rpi based?  no thx
May 13 20:20:35 <mrpackethead> and a low end MSP430 is sub $1
May 13 20:21:06 <_av500_> mrpackethead: stm32f030 is 50¢ :)
May 13 20:21:26 <_av500_> zmatt: rpi? mo
May 13 20:21:27 <_av500_> no
May 13 20:21:51 <_av500_> http://www.usatoday.com/story/tech/2015/05/11/9-computer-killing-it-on-kickstarter/27129135/
May 13 20:22:02 <zmatt> "CHIP, a Raspberry Pi-based motherboard"
May 13 20:22:15 <_av500_> its not pi based
May 13 20:22:20 <Humpelstilzchen> allwinner...
May 13 20:22:23 <zmatt> that's what your link said
May 13 20:22:24 <_av500_> stupid journalists are stupid
May 13 20:22:24 <mrpackethead> wow, they are doing well
May 13 20:22:27 <zmatt> ah
May 13 20:22:49 <mrpackethead> good luck to them
May 13 20:22:55 <_av500_> yup
May 13 20:23:09 <mrpackethead> will transform their life.
May 13 20:23:19 <_av500_> handling all that money for 0 margin must be fun
May 13 20:23:52 <mrpackethead> you are assuming it is $0
May 13 20:23:55 <mrpackethead> margin
May 13 20:24:09 <zmatt> so then you have a $9 computer which is actually useless unless you spend much more money than that on external stuff
May 13 20:24:25 <zmatt> but we'll see :P
May 13 20:24:37 <_av500_> zmatt: the average backer is spending $50
May 13 20:24:45 <_av500_> psych 101 at work
May 13 20:25:08 <_av500_> mrpackethead: even if they take home 20%, is that worth it?
May 13 20:25:21 <_av500_> with full risk
May 13 20:25:24 <mrpackethead> without knowing al the numbers, you just dont' know.
May 13 20:25:35 <_av500_> I know some of the numbers
May 13 20:27:57 <mrpackethead> I've thought ( not so seriously ) about running a KS project
May 13 20:28:29 <_av500_> go for izt
May 13 20:28:32 <_av500_> its a sellers market
May 13 20:28:49 <_av500_> do it like pebble, use it get garanteed preorders
May 13 20:30:25 <_av500_> and if you feel like it, halfway through blame some supplier and fail to deliver
May 13 20:30:38 <_av500_> thats the "risk" in the boilerplate
May 13 20:30:49 <_av500_> that everybody signed off on
May 13 20:31:01 <_av500_> brillant
May 13 20:31:16 <mrpackethead> i'd only actually go to sell somethign i had finished.
May 13 20:32:01 <_av500_> yes
May 13 20:32:24 <zmatt> mrpackethead: the approach suggested by _av500_ works better if you first have your sense of ethics surgically removed
May 13 20:32:29 <_av500_> free publicity and no-risk preorders
May 13 20:32:41 <_av500_> zmatt: lets say thats optional
May 13 20:33:37 <mrpackethead> Risk#1. I am using TI parts, and there is no real documentation
May 13 20:33:38 <_av500_> zmatt: I like the ks idea for people that produce stuff like comic books or music
May 13 20:33:52 <_av500_> for HW projects, not so much
May 13 20:34:15 <_av500_> mrpackethead: :)
May 13 20:34:27 <_av500_> risk #1: MUSB
May 13 20:34:28 <zmatt> mrpackethead: lol waht
May 13 20:34:44 <zmatt> TI is about the only SoC vendor who publishes real docs
May 13 20:35:01 <zmatt> although allwinner seems to be doing okayish now
May 13 20:35:10 <_av500_> lolwat?
May 13 20:35:15 <zmatt> broadcom.. ehm... *chuckles*
May 13 20:35:49 <_av500_> the difference is, BCM has data sheets
May 13 20:35:55 <_av500_> not so sure AW has some at al
May 13 20:36:27 <zmatt> _av500_: they have 'em... not available, but I suspect they do exist yes
May 13 20:36:32 <zmatt> possibly
May 13 20:36:39 <_av500_> maybe
May 13 20:37:00 <zmatt> GvL seems to have access to some
May 13 20:37:04 <_av500_> we neve got them from AW or RK
May 13 20:37:09 <_av500_> and we worked damn close with them
May 13 20:37:22 <zmatt> RK ?
May 13 20:37:27 <_av500_> rockchip
May 13 20:37:53 <zmatt> ah
May 13 20:38:07 <_av500_> sure they have docs that name registers and bits in them
May 13 20:38:13 <_av500_> but that seems about all
May 13 20:38:24 <zmatt> yeah I can complain plenty about TI docs, but I still know they're epicly awesome compared to the other guys
May 13 20:38:27 <_av500_> its not like a TI TRM
May 13 20:38:42 <_av500_> TI docs are good
May 13 20:38:59 <zmatt> the am335x docs actually sub-par compared to various other SoCs
May 13 20:39:10 <zmatt> am335x in general seems to have been done in a hurry or something
May 13 20:39:34 <_av500_> I guess TI has more ppl doc writing than AW chip designing
May 13 20:39:37 <mrpackethead> yeah, i just dont' like E2E
May 13 20:39:50 <zmatt> mrpackethead: I do have custom CSS rules that make it a bit more bearable
May 13 20:40:05 <zmatt> (since the site layout change)
May 13 20:40:34 <mrpackethead> Custom CSS rules?
May 13 20:40:38 <mrpackethead> As in ??
May 13 20:40:44 <_av500_> mrpackethead: html
May 13 20:41:01 <mrpackethead> for using E2E?
May 13 20:41:04 <zmatt> _av500_: I often consult the docs of related SoCs for more info... but that does require having a some intuitive feeling which parts are applicable and which aren't
May 13 20:41:58 <zmatt> mrpackethead: as in https://chrome.google.com/webstore/detail/user-css/okpjlejfhacmgjkmknjhadmkdbcldfcb
May 13 20:42:14 <mrpackethead> I just dont' like the fact that the only way you can get in support, is to ask in an open forum
May 13 20:42:43 <zmatt> mrpackethead: that's definitely not the only way, but it is if you're not a sufficiently big customer
May 13 20:42:45 <mrpackethead> and then you get somone answering who actually is only intent on closign teh thread and not answering the question
May 13 20:43:00 <zmatt> hmm, that seems to vary
May 13 20:43:11 <mrpackethead> i only seem to spend $40-50k each month with TI
May 13 20:43:19 <mrpackethead> so its probalby not big enough to get them too excited.
May 13 20:43:36 <zmatt> there seem to be some decently knowledgable people in the Sitara forum
May 13 20:43:45 <mrpackethead> the items that i buy, you dont' have to program
May 13 20:43:46 <mrpackethead> :-(
May 13 20:43:57 <zmatt> otoh the DM814x support guys are... well-intentioned
May 13 20:44:01 <zmatt> but that's about it
May 13 20:44:03 <zmatt> :/
May 13 20:44:42 <zmatt> like... http://i1.kym-cdn.com/photos/images/facebook/000/325/934/060.png
May 13 20:44:53 <_av500_> lol
May 13 20:46:45 <zmatt> so it's not like I can get angry with them... they don't seem to have access to decent resources themselves
May 13 20:47:39 <zmatt> (not that it matters now that our dm814x board has been indefinitely shelved)
May 13 21:09:43 <zmatt> rcn-ee: any idea if the first x15 boards will have experimental cpu versions? (like the one on the elinux pictures)
May 13 21:09:48 <zmatt> *first released
May 13 21:10:20 <zmatt> since the chip pictured most likely has the whole feature set, including EVEs and stuff
May 13 21:11:17 <rcn-ee> zmatt, well the first alpha board, said 'am57xyz', not the usual, 'xamxxx'..
May 13 21:11:27 <rcn-ee> but now they glued a heatsink to it..
May 13 21:11:54 <zmatt> ah, too bad
May 13 21:12:41 <zmatt> the photo still showed an x5777
May 13 21:12:57 <zmatt> (http://elinux.org/File:X15_TOP_SIDE.jpg )
May 13 21:14:13 <zmatt> the whole "AM" designation makes no sense to me anyway, based on feature set it should have been a DM
May 13 21:14:46 <Posterdati> usb over ethernet stop working, great!
May 13 21:15:05 <rcn-ee> zmatt, which should have just been omapxyz ;)
May 13 21:15:28 <rcn-ee> silly marketting..
May 13 21:16:04 <zmatt> well omap = dead, also I'm not sure how much sense Vayu makes as an omap
May 13 21:16:12 <zmatt> even though it's clearly omap5-derived
May 13 21:16:54 <rcn-ee> it only died because of marketting decided to spin it out as "sitara" and the "amXyz.."
May 13 21:17:13 <zmatt> well no, the omap division has been killed off a while ago already
May 13 21:17:26 <zmatt> also, the am572x is the light-version of the cpu
May 13 21:17:45 <rcn-ee> so with omapconf: DRA75X ES1.1 GP Device (STANDARD performance (1.0GHz))
May 13 21:18:16 <zmatt> yeah, dra75x/dra74x are other partcodes for Vayu
May 13 21:18:25 <rcn-ee> it's running 1.5Ghz.. so ignore that last spot. .;)
May 13 21:18:38 <zmatt> the real ones are tda2x though
May 13 21:18:46 <zmatt> like if all the EVEs pass factory test
May 13 21:19:03 <zmatt> it's designed for ADAS
May 13 21:19:10 <Posterdati> please, how can I setup ethernet over usb on beagle bone rev. 6?
May 13 21:19:17 <Posterdati> please, how can I setup ethernet over usb on beagle bone rev. B?
May 13 21:21:10 <rcn-ee> Posterdati, it's amazing what you can find on this new thing called google!!! https://askubuntu.com/questions/380810/internet-over-usb-on-beaglebone-black
May 13 21:21:43 <rcn-ee> i think google's going to get big someday.. just watch..
May 13 21:21:43 <Posterdati> rcn-ee: this post won't work
May 13 21:21:51 <rcn-ee> Posterdati, why not?
May 13 21:22:06 <Posterdati> ifconfig usb0 192.168.7.2
May 13 21:22:11 <Posterdati> ifconfig: SIOCSIFADDR: No such device
May 13 21:22:21 <Posterdati> uname -a
May 13 21:22:32 <Posterdati> Linux beaglebone 3.2.42 #1 Thu Mar 28 11:28:20 CET 2013 armv7l GNU/Linux
May 13 21:22:44 <rcn-ee> ahh 3.2.x good luck with that!
May 13 21:22:52 <rcn-ee> your on your own now..
May 13 21:23:02 <rcn-ee> http://beagleboard.org/latest-images
May 13 21:23:35 <Posterdati> what?!?! This is crazy! I just did opkg upgrade to upgrade old packages
May 13 21:24:07 <rcn-ee> bb.org hasn't had an Angstrom maintainer for almost 1.5 years...
May 13 21:24:16 <Posterdati> anyway
May 13 21:24:40 <Posterdati> I flashed a debian too from that page and it won't work either
May 13 21:25:10 <rcn-ee> "won't work" please pastebin.com your serial bootlog..
May 13 21:25:19 <Posterdati> no serial bootlog
May 13 21:25:38 <rcn-ee> okay, so how did you "flash" your microSD card?
May 13 21:25:57 <veremit> opkg rofl
May 13 21:25:57 <rcn-ee> dd? etc?
May 13 21:26:08 <veremit> apt-get I suggest
May 13 21:26:16 <Posterdati> downloaded that image, xz -d ... dd if=file.img of=/dev/sde bs=1M
May 13 21:26:33 <rcn-ee> veremit, do you know where that got's picture with stark "one doesn't upgrade angstrom"..
May 13 21:26:42 <zmatt> rcn-ee: I stumbled across something curious btw: the control module has DDR PHY registers for 5 data macros... suggesting at one point in the design its memory bus was 32-bit + ECC
May 13 21:27:01 <zmatt> one of many things that got axed in an effort to make it smaller/cheaper I guess
May 13 21:27:05 <veremit> rcn-ee .. alas no. I feel deprived lol
May 13 21:27:45 <Posterdati> rcn-ee: the board boots, led went all on and then only USR1 and USR3
May 13 21:28:09 <rcn-ee> usr1/usr3 is cpu and mmc, so what's the issue?
May 13 21:28:30 <zmatt> sounds like the flasher in action?
May 13 21:28:33 <rcn-ee> plug in serial, log in..
May 13 21:28:45 <rcn-ee> he's got the bone, so no eMMC. ;)
May 13 21:28:49 <zmatt> ah
May 13 21:29:27 <rcn-ee> and i think the flasher 'should' shutdown... with the bone white... never bothered to check it..
May 13 21:30:28 <Posterdati> ok
May 13 21:30:32 <Posterdati> I've got the console
May 13 21:30:39 <Posterdati> uboot seems to boot the kernel
May 13 21:30:48 <Posterdati> Uncompressing Linux... done, booting the kernel.
May 13 21:30:55 <Posterdati> and then reboot
May 13 21:31:27 <rcn-ee> what was the exact file name you flashed to the microSD?
May 13 21:31:48 <Posterdati> bone-debian-7.8-lxde-4gb-armhf-2015-03-01-4gb.img
May 13 21:32:55 <Posterdati> dd if=bone-debian-7.8-lxde-4gb-armhf-2015-03-01-4gb.img of=/dev/sde bs=1M
May 13 21:33:26 <veremit> . && sync
May 13 21:34:24 <bahar_> hello
May 13 21:34:29 <rcn-ee> okay, dd'ing a card, 5mins/GB.. ;)
May 13 21:34:46 <zmatt> isn't cat likely to be faster?
May 13 21:34:53 <zmatt> or at least a bigger blocksize
May 13 21:34:57 <Posterdati> veremit: never used sync on dd images
May 13 21:35:27 <bahar_> anyone know when the new beagleboard X15 will come out?
May 13 21:35:40 <rcn-ee> the day after it's ready.
May 13 21:36:59 <bahar_> I only ask because http://www.reddit.com/r/BeagleBone/comments/2sfd4e/new_beagleboardx15_due_out_february_dualcore_2g/ mentions it was to be out a couple months ago, so I'm curious if there are updates
May 13 21:37:05 <zmatt> I think the last estimate I saw was june?
May 13 21:37:10 <zmatt> check the forum thread
May 13 21:37:54 * veremit checks the calendar. Yup, still May .. *phew*
May 13 21:38:28 <rcn-ee> bahar, it 'was', then i accidentlly reported a failure to software reset problem to ti, then they went down that nasty rabbit hole.. (pmic, core, etc..)
May 13 21:39:15 <Posterdati> rcn-ee: I see you created a debian 8 image
May 13 21:39:16 <rcn-ee> plus it was their first "22nm" process...
May 13 21:39:52 <zmatt> rcn-ee: so YOU did this!!! ;)
May 13 21:39:53 <Posterdati> rcn-ee: could be a wrong uEnv.txt file?
May 13 21:40:15 <rcn-ee> did I? no... the omap5 board in my basement did it.. not me...
May 13 21:41:04 <zmatt> you buried it in your basement?  note to self: don't make rcn-ee angry
May 13 21:41:16 <rcn-ee> Posterdati, uEnv.txt doesn't matter on the original BeagleBone... Now if you have a Beaglebone Black, uEnv.txt could be an issue..
May 13 21:41:32 <rcn-ee> it's locked in a box, don't want those omap5's getting lose...
May 13 21:42:51 <Posterdati> ok, so shall I buy a raspberry-pi?
May 13 21:42:52 <veremit> down in the dungeon .. mwhahaha
May 13 21:42:58 <veremit> a What?
May 13 21:43:42 <rcn-ee> Posterdati, just verifing the image here on my bone classic, it takes around 5mins per GB, so another 10 minutes... yayayayay..
May 13 21:44:02 <Posterdati> ok thanks
May 13 21:45:56 <rcn-ee> little odd it being a B/6...  There's usually a stick on the eth,, BeagleBone Rev: (value)...
May 13 21:46:47 <zmatt> ARGH, just packed my stuff, ready to catch last train... did a quick check online, it's delayed and will arrive too late to make the last connection -.-
May 13 21:47:21 <thurgood_> eek that blows
May 13 21:47:25 <veremit> w00 public transportatoin ;D
May 13 21:47:33 <thurgood_> maybe your connection will be late too
May 13 21:47:40 <zmatt> thurgood_: very unlikely
May 13 21:47:48 <thurgood_> ;/
May 13 21:50:45 <rcn-ee> Posterdati, wonders if your "B" is the new classic BeagleBone with black's am3358 (es2.x)..
May 13 21:52:23 <Posterdati> yes AM3358B
May 13 21:53:01 <zmatt> didn't know there were BBWs with an am3358b/c
May 13 21:53:28 <rcn-ee> CircuitCo used up their reserver of ES1.x am3359's.. ;)
May 13 21:53:42 <zmatt> ah
May 13 21:54:15 <Posterdati> on the back of the box there's "Circuitco"
May 13 21:54:17 <rcn-ee> Posterdati, when and where did you get this board? (we don't have stock, so i'm trying to figure where to grab one.)
May 13 21:54:51 <rcn-ee> ah crap.. this usb 3.0 controller crapped out again.. startting dd on another box...
May 13 21:54:51 <Posterdati> rcn-ee: was a mouser board almos 11 months ago
May 13 21:54:52 <bahar_> rcn-ee - i'm interested in hearing more about that... got a link or...?
May 13 21:59:20 <rcn-ee> Posterdati, https://paste.debian.net/176563/
May 13 21:59:56 <rcn-ee> bahar, sorry no link... some of it is on the x15-beagle mailing list, the rest was over the phone.. ;)
May 13 22:00:14 <Posterdati> rcn-ee: so?
May 13 22:01:12 <Posterdati> what is the problem then?
May 13 22:01:30 <rcn-ee> the image looks fine here, mouser doesn't have stock, (so can't test one of theirs..) i'd say reflash the microSD, otherwise you should see exactly what i showed on that paste.debian.net link..
May 13 22:02:04 <Posterdati> rcn-ee: this is the third time I will reflash it, I think without success
May 13 22:02:14 <rcn-ee> different microSD card?
May 13 22:02:23 <rcn-ee> use a different power supply?
May 13 22:04:07 <Posterdati> rcn-ee: different ps same problem
May 13 22:04:48 <ds2> *yawn*
May 13 22:05:16 <rcn-ee> okay'd i give this console image a try: (it's 3.14.x vs 3.8.x) if that still fails rma it.. : http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Jessie_Snapshot_console
May 13 22:05:55 <Posterdati> you uEnv.txt has got a different size from my
May 13 22:06:56 <Posterdati> rcn-ee: could you post it please?
May 13 22:08:03 <rcn-ee> http://paste.debian.net/176564/
May 13 22:10:36 <Posterdati> same crap
May 13 22:11:57 <Posterdati> nice I burnt 71 euros!
May 13 22:12:50 <thurgood_> rma it...
May 13 22:12:55 <rcn-ee> just rma it.. circuitco is quick..
May 13 22:13:08 <Posterdati> and spent other money on it?
May 13 22:13:12 <thurgood_> they'll make it good, you'll just burn some shipping
May 13 22:13:44 <Posterdati> the board worked with their sd card
May 13 22:14:20 <rcn-ee> yet it fails with the factory image on: http://beagleboard.org/latest-images
May 13 22:14:27 <thurgood_> wait 70? are you using a white or a black?
May 13 22:14:39 <zmatt> white
May 13 22:14:40 <rcn-ee> yeah the white costs that much..
May 13 22:14:55 <thurgood_> ah.. I thought it was black previously
May 13 22:14:56 <rcn-ee> (it was before that rasberry pi came out)
May 13 22:15:08 <Posterdati> it is a white board
May 13 22:16:28 <thurgood_> what happened to the original card?
May 13 22:16:59 <Posterdati> anyway this forces me to choose mips card
May 13 22:19:44 <Posterdati> I need a nice board to deploy a small project on it in less than one month and I haven't got time for this... So texas and beaglebone won't be my next choose
May 13 22:22:54 <Posterdati> I must say that ti gave me always problems :) never been lucky with that
May 13 22:38:51 <Posterdati> reflashed... not working
May 13 22:49:00 <Posterdati> the jessie distro seems to work! LOL
May 13 22:49:12 <Posterdati> Error: Driver 'tfp410' is already registered, aborting...
May 13 22:50:10 <Posterdati> it stucks in this error
May 13 22:56:33 <Posterdati> another crap board
May 13 23:01:07 <Posterdati> basically they stole me 71 euros
May 13 23:01:23 <ds2> if I haven't said it before, I HATE VIM
May 13 23:04:14 <Posterdati> I hope ti to fail with pain
May 13 23:04:16 <Posterdati> :) lol
May 13 23:08:35 <zmatt> ds2: vim <3
May 14 00:41:20 <JParisse> Can anybody suggest a BegleBone tutor for a Linux noob in LA (near LAX)? I need help getting started... Thanks in advance
May 14 00:42:15 <JParisse> Um.... BeagleBone.... Debian Linux....
May 14 00:43:38 <ds2> in person tutor?
May 14 00:46:05 <JParisse> In person... yes... I have a work project that needs porting over to BeagleBone... I just need an hour or two to get started... I have been "stuck" for a couple of days trying to get past the basics...
May 14 00:49:44 <ds2> The last time I was at scale, I didn't come across too many folks doing the BB stuff (I'm up in SJC)...closest was some folks out in orange county doing a distro
May 14 00:49:55 <ds2> but I can't recall the name of the distro
May 14 00:53:01 <JParisse> The root of my problem is Linux... perhaps a Linux tutor could help me with getting started...
May 14 00:53:56 <veremit> JParisse .. there's loads of videos/etc tutorials online .. have you really tried everything?
May 14 00:54:05 <erakis> Hi, I'm running Qt5 application over EGLFS and once I touch the screen I'm seeing the console flashing in the background. How can I stop it ?
May 14 00:56:17 <JParisse> It is not that I have tried everything... I am in a time crunch and I want to get productive quickly....  Even the BeagleBone/Support/Bone101 page gets me hung up...
May 14 00:56:46 <JParisse> An hour of one on one time with an expert is worth a week of hitting my head against a desk... :)
May 14 00:59:10 <ds2> the problem is you need it in person
May 14 00:59:49 <veremit> you'd be better handing the project off ..
May 14 00:59:59 <JParisse> exactly....
May 14 01:00:17 <veremit> what are you trying to do, briefly?
May 14 01:00:35 <JParisse> no... I want to learn this... just want to get a head start... but thanks for the suggestion
May 14 01:01:04 <JParisse> a simple project... a scanning Lidar rangefinder
May 14 01:01:16 <veremit> cool
May 14 01:01:28 <ds2> really?
May 14 01:01:48 <ds2> i am hacking something like that together as a side project...
May 14 01:02:04 <JParisse> woks great in Arduino... my TD wants us to start using BBB
May 14 01:02:41 <JParisse>  great = not that great... :)
May 14 01:02:43 <ds2> who's light engine are you using?
May 14 01:03:05 <JParisse> Lidar Lite by PulsedLight 3D
May 14 01:04:00 <ds2> what kind of range are you seeing? I am using the same thing
May 14 01:04:20 <veremit> well we got some common ground already :D
May 14 01:04:48 <ds2> wish someone made one that can do much much longer ranges
May 14 01:05:16 <JParisse> :)   40m is pretty long....
May 14 01:06:20 <ds2> that's less then 120ft
May 14 01:06:24 <JParisse> so you guys can appreciate why I want to DYI.... I just need some nomenclature help...
May 14 01:06:42 <ds2> the entry level bushnell stuff can do 400ft or so
May 14 01:06:51 <ds2> 1000ft would nice
May 14 01:07:22 <JParisse> I am doing multi copter sense and avoid so 100' is perfect... but yeah... 1000' would be... :)
May 14 01:08:14 <ds2> isn't this too narrow of a beam?
May 14 01:08:15 <zmatt> my first thought would be: emit a pseudo-random sequence, correlate measured reflection with a delayed version of same sequence... would be relatively slow to "home in" on the right value, but should be rather sensitive (and insensitive to noise)
May 14 01:08:29 <ds2> or do you have a mirror somewhere?
May 14 01:08:30 <zmatt> no idea how those things normally work though ;)
May 14 01:08:59 <ds2> zmatt: and what kind of PN code would you recommend?
May 14 01:09:27 <JParisse> the hardware I am using can make a burst of measurements and average them... convenient ...
May 14 01:09:52 <zmatt> ds2: not sure it matters much... anything that has basically 0 auto-correlation with shifted versions of itself
May 14 01:10:01 <zmatt> they're used a lot, also in things like GPS
May 14 01:10:04 <ds2> you must not be using the stock firmware on your Lidar module
May 14 01:10:34 <zmatt> you could also do it with simple pwm, but at various frequencies in succession
May 14 01:10:52 <zmatt> (benefit of the BBB eHRPWM module is that it's adjustable in ~200 ps steps)
May 14 01:11:16 <ds2> zmatt: but then your bit duration becomes your measurement resolution
May 14 01:11:26 <JParisse> yep... check the example code... it's there....
May 14 01:12:07 <ds2> I am getting my inertial sensors math done first
May 14 01:12:23 <zmatt> ds2: hmm, I guess 200 ps is not fine enough for these purposes... how the fuck do those things perform their measurement usually them?
May 14 01:12:26 <zmatt> *then
May 14 01:13:19 <ds2> zmatt: not sure... need to read it a few more time... I was thinking of modifying one of those TDR chips to do it
May 14 01:13:34 <JParisse> well off to eat dinner... gonna look for a local Linux guru...  thanks all !!!
May 14 01:13:43 <zmatt> still, generating a carefully timed signal intuitively seems easier than carefully measuring the timing of a signal, hence my intuition towards correlating against a locally generated signal
May 14 01:13:49 <zmatt> but see also "rake search"
May 14 01:14:10 <ds2> got enough things to chew off at once
May 14 01:14:10 <zmatt> (what cellphones to do lock onto CDMA signals)
May 14 01:14:29 <ds2> yes but knowning when the lock occurs isn't critical in CDMA
May 14 01:14:29 <zmatt> if you have a TDR chip, that sounds like very much want you'd want
May 14 01:14:43 <ds2> zmatt: there are many TDR chips around
May 14 01:15:15 <zmatt> nor is it in your case... but the "lock" consists of determining a delay value which *is* what you're looking for
May 14 01:15:49 <ds2> that delay is on the granularity of a bit boundary
May 14 01:16:22 <zmatt> no, it's much finer than that
May 14 01:17:18 <zmatt> (in case of CDMA it actually attempts to locate multiple reflections, since these often occur in urban environments and detecting them allows one to use them for improved reception instead of being interference as they would be in previous systems)
May 14 01:17:57 <zmatt> but I have no idea how all that stuff works
May 14 01:18:25 <zmatt> some keystone chips have hw accelerators for it :P
May 14 01:20:04 <zmatt> they actually hooked up the rake search thing to the C66x DSP's coprocessor interface (EFI) rather than as a memory-mapped peripheral... I guess for lower latency
May 14 01:20:29 <zmatt> (yes, I read way too much random docs)
**** ENDING LOGGING AT Thu May 14 02:59:58 2015