**** BEGIN LOGGING AT Tue Feb 23 02:59:58 2016
Feb 23 05:23:35 <gayatri> hi im using the beagleboneblack
Feb 23 05:23:53 <gayatri> i have noticed an issue which keeps recurring
Feb 23 05:25:07 <gayatri> sometimes when you power on the board the board doesnt power on and the sys_5v falls to 1.1v
Feb 23 05:25:44 <gayatri> and the board fails to work
Feb 23 05:26:02 <gayatri> what is the reason for this and how to rectify this
Feb 23 05:50:02 <zmatt> gayatri: sys_5v falls to 1.1V, or never rises above 1.1V ?
Feb 23 05:51:32 <zmatt> it does tend to hover around 1V when the system is powered off... I can think of two reasons it might not power up:  1. overcurrent detected by pmic   2. power is ramped up too slowly
Feb 23 05:52:02 <zmatt> does the system power on in thsi state if you press the power button?
Feb 23 05:52:54 <muni> hii
Feb 23 05:54:41 <muni> i am using beagle bone black  along with 4D 7" LCD cape with android os for one my project
Feb 23 05:56:37 <muni> i used beagle bone black as a master and i need to connect more than 5 slave devices which as RS485 how should i interlink slaves to master
Feb 23 06:03:37 <zmatt> muni: RS485 support multidrop (and in fact peer-to-peer) communications, if whatever protocol you're using supports it
Feb 23 06:04:04 <zmatt> with quality RS485 transceivers you can connect up to 128 devices to a single bus
Feb 23 06:04:30 <zmatt> (with standard ones 32 devices, which is still plenty)
Feb 23 06:18:30 <muni> is there any rs485 cape  to connect to 4D 7" LCD cpe
Feb 23 06:20:04 <muni> is there compulsory cape needed or external RS232 to RS485 connecter we can use for modbus communication
Feb 23 06:23:33 <gayatri> zmatt:It does not power up even after pressing the power button
Feb 23 06:24:13 <zmatt> gayatri: that suggest to me the power supply ramps up too slowly (> 50 ms)... for some annoying reason that flips out the PMIC
Feb 23 06:24:35 <gayatri> ohk
Feb 23 06:24:39 <zmatt> I've only seen that happen with one particular lab supply myself
Feb 23 06:24:49 <gayatri> so what do you suggest
Feb 23 06:25:33 <zmatt> what are you using to power it at the moment?
Feb 23 06:26:25 <ds2> bike generator?
Feb 23 06:27:05 <zmatt> morning ds2 :)
Feb 23 06:27:26 <ds2> =)
Feb 23 06:27:47 <gayatri> im using U8031A from Agilent to power it on
Feb 23 06:28:42 <zmatt> gayatri: lol, that "one particular lab supply"... also an agilent
Feb 23 06:28:54 <zmatt> try connecting the BBB after enabling the supply
Feb 23 06:29:33 <zmatt> this particular PMIC really sucks
Feb 23 06:30:21 <gayatri> ohk we even tried out aplabs power supply
Feb 23 06:30:25 <zmatt> there's another way to make this issue disappear: supply power via the battery terminals (the not-installed 2x2 header)
Feb 23 06:30:30 <gayatri> the problem still persists
Feb 23 06:30:49 <zmatt> ok, so the issue is that it needs to ramp up sufficiently fast
Feb 23 06:31:13 <gayatri> but i am goin to use it for a battery powered application
Feb 23 06:31:26 <gayatri> im using a li on battery
Feb 23 06:32:22 <zmatt> that should actually make the problem disappear.  note however that the battery-powered operation of the BBB is not *officially* supported, and requires at least one patch to avoid trouble
Feb 23 06:33:04 <zmatt> see http://elinux.org/BeagleBone_Power_Management .. the Warning near the top has a link to a forum thread with details
Feb 23 06:33:32 <zmatt> the second link on that page shows our workaround for it
Feb 23 06:34:08 <ds2> battery works fine.
Feb 23 06:34:23 <zmatt> ds2: as long as you never power off...
Feb 23 06:34:40 <ds2> powering off is overrated
Feb 23 06:38:15 <ds2> zmatt: does the page implies it is the HDMI framer's fault?
Feb 23 06:39:18 <zmatt> no?
Feb 23 06:39:37 <ds2> so no updates from your email?
Feb 23 06:39:42 <gayatri> will try it and get back to you
Feb 23 06:39:49 <zmatt> ds2: ?
Feb 23 06:39:53 <gayatri> thanks fr your help zmatt
Feb 23 06:40:00 <gayatri> hav a gr8 day
Feb 23 06:40:07 <ds2> zmatt: trying to figure out what, if any difference, is there from what you posted to the list a while back
Feb 23 06:40:25 <zmatt> ds2: the two links *are* to posts on the list
Feb 23 06:40:43 <ds2> I am looking at a elinux page
Feb 23 06:41:19 <zmatt> that's a collection of material that I've been working on since back then
Feb 23 06:41:43 <ds2> figure that...trying to figure out if there has been updates
Feb 23 06:42:49 <zmatt> you can check the page history of course
Feb 23 06:46:10 <ds2> 'k looks like no changes
Feb 23 06:46:43 <ds2> I suppose an external PFET would render that issue moot
Feb 23 06:48:29 <zmatt> ?
Feb 23 07:04:42 <ds2> instead of using the PMIC to control the vsys line, put a PFET between it and the battery
Feb 23 07:04:55 <ds2> and yank that PFET on poweroff
Feb 23 07:05:57 <zmatt> ehm, you're not making sense.  the problem isn't vsys staying powered, the problem is the 3V3B regulator staying on
Feb 23 07:06:12 <ds2> yes
Feb 23 07:06:32 <ds2> but yanking the battery power would side step the problem
Feb 23 07:06:38 <zmatt> which is also easy to fix (unlike what you're suggesting which sounds like an impossible patch)
Feb 23 07:07:02 <zmatt> oh I'm now maybe getting what you mean
Feb 23 07:07:02 <ds2> hmmm?
Feb 23 07:07:35 <ds2> screwing with the regulator requires careful rework
Feb 23 07:07:59 <ds2> putting a switch in line with the battery would keep things simple
Feb 23 07:08:34 <zmatt> it's actually a really easy patch
Feb 23 07:08:51 <zmatt> how were you imagining you'd control the switch exactly?
Feb 23 07:09:03 <ds2> gpio
Feb 23 07:13:07 <zmatt> wouldn't a pfet open when the system powers off?
Feb 23 07:13:49 <ds2> isn't that what's needed?
Feb 23 07:13:51 <zmatt> I mean, close (if thinking of it a switch)
Feb 23 07:13:58 <zmatt> I meant open as in a gate
Feb 23 07:14:10 <ds2> yes, open as in disconnect battery
Feb 23 07:14:12 <zmatt> conduct, whatever, annoying terminology :P
Feb 23 07:14:42 <ds2> the polarity can be swapped around with an NFET in front
Feb 23 07:16:34 <zmatt> well, no, though it can using an inverter :P  getting more and more complicated here... and there's still the little issue of powering the device on again
Feb 23 07:17:20 <ds2> that's harder to explain then it is to implement
Feb 23 07:17:28 <ds2> I did that for a battery power microcontroller setup before
Feb 23 07:19:35 <zmatt> it doesn't sound simple to get reliable.  you also lose some of the benefits of having a battery attached, such as getting a clean poweroff sequence without the pmic visiting FAULT state, and getting rid of some of the more obnoxious misbehaviours of the pmic such as the <50ms powerup slew requirement
Feb 23 07:20:57 <ds2> for your application where the battery is more of a UPS, sure
Feb 23 07:21:02 <ds2> for portable use...
Feb 23 07:21:33 <zmatt> for either of those
Feb 23 07:23:29 <zmatt> ohey it's morning
Feb 23 07:24:06 <zmatt> (I mean, it already was a little bit morning before, but now it's getting very morning)
Feb 23 07:24:22 <ds2> :)
Feb 23 08:59:48 <SAK> hey guys. i am building kernel in my BBB's sd card. I am following this link http://kernelnewbies.org/KernelBuild to build my kernel. when i tried to install module i got depmod error. here is the actual error http://pastebin.com/GBafs35J. to fix this we ran "sudo apt-get upgrade sudo apt-get install linux-generic linux-headers-generic linux-image-generic" it failed because process was already working
Feb 23 09:00:09 <SAK> we tried to restrt BBB but it is not starting up now
Feb 23 09:00:41 <SAK> all 4 LED's are lit solid. neither of them are blinking
Feb 23 09:00:59 <SAK> we tried to debug using serial cable its ouput is
Feb 23 09:02:45 <SAK> http://pastebin.com/1ZE4ijFb
Feb 23 09:03:09 <SAK> i am unable to understand what is wrong with BBB
Feb 23 09:06:50 <veremit> SAK: if you want to roll your own kernel .. i suggest following these instructions.. https://eewiki.net/display/linuxonarm/BeagleBone+Black
Feb 23 09:07:26 <veremit> you're unlikely to succeed using standard mainline howto's as the latest beagle specific patches are unlikely to be included
Feb 23 09:07:52 <veremit> not to mention lots of gotchas around archs and toolchains :)
Feb 23 09:08:08 <philenotfound> before digging through code, does anyone know the lower limits of the gpu, like if it's possible to output 240p at 15khz?
Feb 23 09:08:29 <veremit> if you are running one of the debian builds, there are a number of kernels you can simply apt-get .. see http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Kernel_Options
Feb 23 09:08:46 <veremit> philenotfound: you're trying to do what?
Feb 23 09:09:17 <philenotfound> hooking it up to a 15khz crt
Feb 23 09:09:38 <veremit> I doubt you want a 15khz refresh rate .. :)
Feb 23 09:09:40 <philenotfound> running mame at native resolution and refresh
Feb 23 09:09:59 <veremit> I highly doubt the gpu will generate anything like that too
Feb 23 09:10:31 <tbr> also it can't do analog output
Feb 23 09:10:46 <veremit> tbr++
Feb 23 09:10:52 <SAK> what should i do right now. remove SD card and flash new debian image in BBB
Feb 23 09:10:53 <SAK> ??
Feb 23 09:11:07 <tbr> and it's not the GPU, but the HDMI framer OR the framebuffer (depending on which route you take)
Feb 23 09:11:29 <philenotfound> tbr: i was thinking of an r2r ladder network and trying VGA first
Feb 23 09:11:53 <veremit> probably do better with a hdmi->vga converter
Feb 23 09:12:11 <tbr> an _active_ converter
Feb 23 09:12:22 <veremit> indeed
Feb 23 09:12:47 <philenotfound> haven't seen any that are capable of 240p
Feb 23 09:12:49 <veremit> SAK .. the latest debian images are at http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#2015-11-03
Feb 23 09:13:03 <veremit> yes, that is a rather ... "poor" resolution by modern standards
Feb 23 09:13:24 <veremit> however, the BBB will _do_ low resolutions
Feb 23 09:13:40 <SAK> so i have to start over all again. Right @veremit
Feb 23 09:13:58 <veremit> SAK: what have you done so far?
Feb 23 09:14:46 <SAK> well. after installation od debian. i loaded SD Card and git linux4.1 in SD Card
Feb 23 09:14:55 <SAK> ran config and make commands
Feb 23 09:15:10 <SAK> means i have done installations only
Feb 23 09:15:44 <philenotfound> veremit: thanks, I'll give it a try. i'll use a jpac in between to avoid damaging the crt by feeding it with 31khz hrefresh
Feb 23 09:16:10 <veremit> ouchie
Feb 23 09:16:25 <philenotfound> veremit: to give a little context: https://www.ultimarc.com/monfaq.html
Feb 23 09:16:27 <veremit> SAK .. 4.1 is the 'stock' kernel as far as I'm aware .. what do you need?
Feb 23 09:16:51 <philenotfound> Basically i want to achieve the same results as with an arcade vga card
Feb 23 09:17:25 <veremit> philenotfound: classic! just don't burn yourself on the HT :P
Feb 23 09:18:02 <SAK> yep. actually i have to run opencv in BBB. to run we are reqquired o do some cmemk.ko. which is done from kernel build. if i am correct. these are the steps
Feb 23 09:18:27 <veremit> SAK: ah oh joy .. opencv and cmem lol
Feb 23 09:19:53 <SAK> you are loling and i am stuck lol... well thanks for guide. but......
Feb 23 09:19:59 <SAK> anyways i am starting over again
Feb 23 09:21:03 <veremit> SAK: its been done ;)
Feb 23 09:21:20 <SAK> done?? seriously??
Feb 23 09:22:17 <veremit> philenotfound: the BBB works with small TFT displays, so the resolution isn't a problem .. but you're going to want to convert straight from digital rgb .. that hdmi framer won't play nice I don't suspect
Feb 23 09:22:49 <veremit> SAK: http://derekmolloy.ie/beaglebone/beaglebone-video-capture-and-image-processing-on-embedded-linux-using-opencv/ -- https://solarianprogrammer.com/2014/04/21/opencv-beaglebone-black-ubuntu/
Feb 23 09:24:55 <veremit> SAK: its quite a popular project
Feb 23 09:24:58 <philenotfound> veremit: you don't think a simple r2r dac is enough?
Feb 23 09:25:18 <veremit> philenotfound: ymmv, it could work .. worth a try ..
Feb 23 09:25:52 <SAK> veremit: yep i followed it.but when i tried to run i got stuck with cmemk.ko no found. upon googling i was told to build kernel. which i was doing
Feb 23 09:26:04 <veremit> philenotfound: just beware the bbb io pins are quite .. sensitive.. and you can fairly easily blow the chip up :/
Feb 23 09:26:33 <veremit> SAK: ah .. let me have another look .. it should be possible to find the sources ..
Feb 23 09:27:27 <SAK> well guide me. what should i do with my current beagle bone black :p its all LEDs are solid lit
Feb 23 09:30:45 <philenotfound> veremit: i'm not feeding any signals into them..
Feb 23 09:31:08 <veremit> philenotfound: no problem then .. hopefully :) just a friendly warning!
Feb 23 09:31:17 <veremit> SAK: sounds like its not happy
Feb 23 09:31:38 <SAK> exactly
Feb 23 09:32:36 <SAK> we debugged it using FTDI and here is the output http://pastebin.com/1ZE4ijFb
Feb 23 09:35:27 <veremit> SAK: ok best to start with a fresh debian install image
Feb 23 09:35:57 <SAK> thanks again buddy
Feb 23 09:36:19 <veremit> the opencv should be built ready .. just looking for the debian package...
Feb 23 09:37:02 <SAK> sure
Feb 23 09:37:06 <SAK> i will be waiting here
Feb 23 09:37:07 <SAK> thanks
Feb 23 09:40:19 <gayatri> hi again
Feb 23 09:41:07 <gayatri> i am working on the beaglebone black and am facing an issue where it looks like d tps powering ic has entered sleep mode
Feb 23 09:41:18 <gayatri> the sys_5v is fixed at 1.1v
Feb 23 09:41:59 <gayatri> in the forum we had a discussion that it could be a rampup problem with the power supply
Feb 23 09:42:29 <gayatri> so i directly powered it on a dc jack with a 5V/3A adapter
Feb 23 09:42:42 <veremit> gayatri: you most likely need the wisdom of zmatt .. oh zmatt ..
Feb 23 09:43:41 <gayatri> yess precisely
Feb 23 09:43:52 <gayatri> zmatt was d one who gave me this idea
Feb 23 09:44:10 <gayatri> it still doesnt power up
Feb 23 09:44:45 <gayatri> the sys_5V still stays at 1.1V inspite of a power button
Feb 23 09:45:15 <gayatri> to check it seperately i also borrowed a tps65217c evm from ti
Feb 23 09:45:27 <gayatri> but the evm also shows the same condition
Feb 23 09:46:28 <gayatri> Any other things that i could try?
Feb 23 09:46:41 <veremit> gayatri: not my area of expertise :/
Feb 23 09:47:07 <gayatri> ohk veremit
Feb 23 09:47:25 <gayatri> :)
Feb 23 09:51:51 <veremit> SAK: urgh .. opencv is new to me .. sounds like debian may/may not have the opencv ready-to-fly :/
Feb 23 09:53:50 <SAK> veremit Thanks anyways man. i think i must switch to other BBB OS
Feb 23 09:55:21 <veremit> SAK: angstrom images are at http://downloads.angstrom-distribution.org/demo/beaglebone/ fyi
Feb 23 10:01:19 <SAK> thanks again
Feb 23 10:21:42 <zmatt> gayatri: do you have a scope to check what's happening on the sys_5v ?
Feb 23 10:24:07 <philenotfound> veremit: thanks, i'll try to to fry the bbb
Feb 23 10:27:59 <gayatri> Yess zmatt i have a scope
Feb 23 10:29:02 <gayatri> will powering it with usb help??
Feb 23 10:30:03 <zmatt> gayatri: maybe check what the 5v supply and/or sys_5v are doing (both can be easily measured via P9)
Feb 23 10:30:31 <gayatri> ohk
Feb 23 10:30:36 <zmatt> usb is not really treated very different from the 5v dc supply
Feb 23 10:30:49 <gayatri> ohk
Feb 23 10:36:09 <zmatt> the "power path" section of http://elinux.org/BeagleBone_Power_Management has an example of a power up... note that in this case the 5v dc (pink) ramps up barely fast enough, I probably used that agilent supply but got lucky
Feb 23 10:37:20 <gayatri> haha good for you
Feb 23 10:37:31 <zmatt> if it's too slow then sys_5v (green) simply never gets switched on
Feb 23 10:38:50 <gayatri> ya bt how can you relate this problem to the tps65217c evm
Feb 23 10:39:09 <gayatri> how does the same problem persist even there
Feb 23 10:39:32 <zmatt> the ramp-up issue *is* an issue of the tps65217 in general, not of the bbb specifically
Feb 23 10:40:24 <gayatri> i hav tried powering that on also with the same dc jack 5V/3A
Feb 23 10:41:51 <gayatri> but the sys_5v is 1.1 even there
Feb 23 10:42:07 <zmatt> I don't know, my suggestion is to check with a scope to see what's going on
Feb 23 10:42:23 <gayatri> yup wil do
Feb 23 10:43:07 <gayatri> if this wouldnt work den ill hav to buy a new bbb
Feb 23 10:43:09 <gayatri> :P
Feb 23 10:43:13 <gayatri> :D
Feb 23 10:43:47 <gayatri> thanks a lot zmatt
Feb 23 10:43:56 <gayatri> wil check and let you knw
Feb 23 10:43:57 <zmatt> the only reasons I know of for failure to power up are  1. slow ramp up  2. a short or other overcurrent fault  3. failure to request power-up (can't happen on a bbb unless the cpu is damaged, might be possible on that evm perhaps)
Feb 23 10:45:03 <muni> hi please any one give me details about modbus communication implementation through beagle bone black with android kitkat
Feb 23 10:45:45 <muni> i am  using beagle bone black as a master and connect more than 5 slaves to beagle bone black
Feb 23 11:51:48 <mnnikhil92> Can anybody please tell me how to port FreeRTOS to Beaglebone Black??
Feb 23 12:04:30 <zmatt> try pulling hard
Feb 23 12:15:05 <veremit> o,O
Feb 23 12:40:11 <zmatt> well that should be a fine way to port something right?
Feb 23 12:45:19 <veremit> sledgehammer might work?
Feb 23 13:01:37 <jkridner> gm all
Feb 23 13:03:19 <veremit> afternoon jkridner - what news?!
Feb 23 13:04:21 <jkridner> Looking at making some more hacks on the BeagleBone Blue design today... Looking forward to the meet-up Thursday.
Feb 23 13:04:47 <jkridner> wondering if http://bit.ly/bbgsocideas has any updates.
Feb 23 13:13:10 <veremit> jkridner: how's that going .. are we to understand correctly that its a single-board .. and when do we get to find out hardware specs?!
Feb 23 13:15:12 <jkridner> yeah, it's a single board. combines the cape and BeagleBone in one.
Feb 23 13:15:38 <veremit> zmatt and I were trying to figure out what that massive (size, not pins) bga was on it .. :p
Feb 23 13:15:45 <jkridner> Well, I'm hoping I can do a blog post about the software and consumer partnerships first.
Feb 23 13:16:04 <jkridner> still trying to get clarity if I can blog about those details or not.
Feb 23 13:16:28 <veremit> how's the timeline looking to get info out?
Feb 23 13:16:31 <jkridner> veremit: yeah, I published that with the intent to create some speculation. :-)
Feb 23 13:16:51 <jkridner> Hardware details might be as late as April.
Feb 23 13:17:13 <jkridner> But, a couple of the partners should be announced this week.
Feb 23 13:17:26 <veremit> besides UCSD
Feb 23 13:17:37 <veremit> and TI :P and Beagle
Feb 23 13:17:37 <jkridner> Christi hates drip campaigns, but that's what we seem to be doing.
Feb 23 13:17:45 <jkridner> yeah, a couple more. :-)
Feb 23 13:17:51 <veremit> don't blame 'em
Feb 23 13:17:56 <jkridner> besides BeagleBoard.org, TI and UCSD.
Feb 23 13:18:08 <veremit> Cco doin manufacture again?
Feb 23 13:18:16 <veremit> *duck*
Feb 23 13:20:21 <zmatt> veremit: well it also has more balls than the am335x (400 vs 324)
Feb 23 13:20:33 <veremit> zmatt: we covered that :p
Feb 23 13:20:40 <veremit> any further detective work/
Feb 23 13:21:39 <zmatt> no :P
Feb 23 13:22:03 <veremit> slacker lol
Feb 23 13:23:18 <veremit> jkridner: any updates on the x15 yet? or still "in the works" too ...
Feb 23 13:23:34 <jkridner> "Any day now"(TM)
Feb 23 13:23:39 <zmatt> veremit: yeah, you know how it is... if you want to get something done, you need to do it yourself...
Feb 23 13:23:45 <zmatt> let me know if you find anything ;)
Feb 23 13:23:50 <veremit> jkridner: so .. september be about right?! ;P
Feb 23 13:24:00 <veremit> zmatt: gawd not you too :/
Feb 23 13:24:48 <jkridner> I give up trying to speculate. Gerald is working it with FCC/CCo. Blue's destiny is in my hands.
Feb 23 13:25:19 <veremit> jkridner: board layout done, fcc sorted, then?! ;)
Feb 23 13:25:37 <jkridner> I don't know if they've retested or not.
Feb 23 13:25:44 <jkridner> about time that they should have
Feb 23 13:25:55 <veremit> *chuckle*
Feb 23 13:26:00 * jkridner would check the latest mailing list responses from Gerald. :-)
Feb 23 13:26:14 <veremit> ugh not another ML of spam and abuse .. naa, thanks :)
Feb 23 13:26:18 <jkridner> I spoke to him yesterday and he just told me he's bugging CCo about it non-stop.
Feb 23 13:26:32 <veremit> lol, nice.
Feb 23 13:26:59 <zmatt> veremit: yeah I've been busy catching up with my backlog of doctor who episodes... gotto prioritize
Feb 23 13:27:13 <veremit> zmatt: ah, I'll let it slide then.
Feb 23 13:27:33 <veremit> jkridner: so .. who knows(tm) on x15 .. but you got sign-off on blue yet?
Feb 23 13:27:46 <jkridner> how do you mean sign-off?
Feb 23 13:27:52 <jkridner> I haven't released the final schematic.
Feb 23 13:27:57 <jkridner> I've got a first prototype.
Feb 23 13:27:57 <veremit> test, fcc appro.... ah lol
Feb 23 13:28:05 <veremit> loooooooooooooooong way off that then lol
Feb 23 13:28:12 <jkridner> but, it is missing the WiFi/BT and Flash on this prototype.
Feb 23 13:28:24 <veremit> holy .. but you've done component choice?!
Feb 23 13:28:24 <jkridner> yeah, I'm not expecting to ship this guy until May/June.
Feb 23 13:28:29 <jkridner> yeah...
Feb 23 13:28:31 * veremit reads Jan2015
Feb 23 13:28:33 <veremit> 16**
Feb 23 13:28:35 <veremit> no
Feb 23 13:28:36 <veremit> f***
Feb 23 13:28:40 <veremit> its Jan2017now llol
Feb 23 13:28:52 <veremit> where does the time go?!
Feb 23 13:28:53 <zmatt> veremit: time flies
Feb 23 13:29:04 <veremit> zmatt: yeah, wish I was
Feb 23 13:29:35 <zmatt> really, all these time flies...
Feb 23 13:29:40 * zmatt gets a fly swatter
Feb 23 13:30:24 <zmatt> (badum, tss)
Feb 23 13:33:29 <veremit> jkridner: well, all sounds promising. Hopefully there'll be something juicy for April :D
Feb 23 13:39:35 <fhyfing> hi there. we like to use openvpn on our devices (am335x based). yet the throughput stays @ 15Mbit/s (3.18 stock kernel) Any ideas?
Feb 23 13:41:51 <veremit> why still a 3.18.. thought that was 3.8 for a minute.
Feb 23 13:47:46 <veremit> fhyfing: have you tried other kernels?
Feb 23 13:48:12 <veremit> what distro/sw version(s) are you using?
Feb 23 13:50:38 <fhyfing> dependencies... we tested w/ bbb for comparison (debian images) Debian8, 4.1 22Mbit Debian7, 3.8 ~70Mbit.
Feb 23 13:51:16 <fhyfing> our distros: buildroot and openwrt
Feb 23 13:52:27 <fhyfing> buildroot 4.1 stock on our system also 15Mbit/s
Feb 23 13:53:47 <zmatt> http://bugs.elinux.org/issues/137
Feb 23 13:53:51 <veremit> hmm wild variations :/
Feb 23 13:56:30 <zmatt> that one was declared "fixed" 9 months ago though, so any recent kernel from rcn-ee shouldn't exhibit that particular issue anymore
Feb 23 13:57:25 <fhyfing> openwrt is the one w/ 3.18 max 15Mbit/s
Feb 23 14:04:19 <fhyfing> should we use the stock kernels or prefer the ti kernels for this problem?
Feb 23 16:02:05 <Catslab> re
Feb 23 16:02:13 <Catslab> saleté de mac qui m'a planté dans la figure
Feb 23 16:02:14 <Catslab> lol
Feb 23 16:02:21 <Catslab> pire que windows xp parfois
Feb 23 16:02:28 <Catslab> mm sorry
Feb 23 16:02:33 <Catslab> wrong channel :)
**** BEGIN LOGGING AT Tue Feb 23 23:00:36 2016
Feb 23 23:57:16 <ozzloy> so i'm finding that maybe beagleboard green DMA is just plain broken and that means most webcams are not going to work.  does that sound right to people in here?
Feb 23 23:58:12 <ozzloy> i'm not really sure what "DMA" is.  just that it stands for direct memory access and that it's coming up a lot in my troubleshooting
Feb 24 01:23:14 <opcode> DMA is the idea that peripherals can access the RAM without going through the processor
Feb 24 01:23:18 <opcode> which is how firewire works
Feb 24 01:23:51 <opcode> it's faster because the CPU is free to do other things while the memory access is happening
Feb 24 01:43:47 <zmatt> someone fairly recently came to the conclusion that usb was actually faster *without* DMA
Feb 24 01:44:13 <zmatt> and afaik it's actually disabled in all kernels from rcn-ee
Feb 24 01:44:22 <zmatt> (for usb)
Feb 24 01:51:07 <opcode> should have said "it can be faster because.."
Feb 24 01:51:18 <opcode> :)
**** ENDING LOGGING AT Wed Feb 24 02:59:58 2016