**** BEGIN LOGGING AT Thu Oct 18 03:00:02 2012
Oct 18 03:01:19 <ka6sox> morning
Oct 18 03:01:54 <emeb> ka6sox: did you teleport to a different TZ?
Oct 18 03:03:14 <ka6sox> emeb, yup
Oct 18 03:04:43 <movax|mac> alright thanks KeatonT
Oct 18 03:04:52 <movax|mac> i'll be testing it soon-ish, will report back on how it goes
Oct 18 03:05:33 <KeatonT> cool I'd love to know
Oct 18 03:09:17 <movax|mac> using the bone as a rapid proto platform before i design a sitara-a8 system from scrtach
Oct 18 03:09:20 <movax|mac> for $89 can't go wrong
Oct 18 03:11:53 <mranostay> ka6sox: can i borrow the teleporter?
Oct 18 03:16:06 <ka6sox> mranostay, you will have to fight my wife for it...and she is a Black Belt...
Oct 18 03:17:40 <ka6sox> emeb, when I get home I should have a new bcc kernel
Oct 18 03:18:01 <ds2> BCC?
Oct 18 03:18:22 <ka6sox> Blank Canvas Cape
Oct 18 03:18:52 <ds2> oh that
Oct 18 03:19:05 <ds2> what's the point of mucking with silly capes?
Oct 18 03:19:29 <ka6sox> you have something more fun to play with?
Oct 18 03:19:54 <ds2> sure... classic stuff
Oct 18 03:20:04 <ds2> sensors out the wazzo on that
Oct 18 03:20:33 <ka6sox> funzone.
Oct 18 03:21:23 <ka6sox> I've wanted to play with GPMC for a while...
Oct 18 03:23:34 <ds2> you got lots of high bandwidth devices?
Oct 18 03:24:31 <ka6sox> yep
Oct 18 03:25:22 <ds2> oh slightly different interest... I'm more into sensing the real world... accs, gyros, magnetometers, compasses, barometers, humidity, etc
Oct 18 03:27:07 <emeb> ds2: hang out on ##stm32 - those guys are all into that stuff
Oct 18 03:27:25 <emeb> they're all making ncopter navigation systems.
Oct 18 03:27:44 <ka6sox> emeb, looks like I borked it..but its a simple fix ;(
Oct 18 03:27:47 <ds2> stm32?
Oct 18 03:28:02 <emeb> ST's cortex Mx processors.
Oct 18 03:28:23 <emeb> cheap/fast/fun. I've been doing a lot of work with those lately.
Oct 18 03:28:26 <ds2> I know what they are... working on a design with them
Oct 18 03:28:34 <emeb> cool.
Oct 18 03:28:44 <ds2> i wish they would run at 3.3V
Oct 18 03:28:47 <ds2> I mean 1.8V
Oct 18 03:28:49 <ds2> not 3.3V
Oct 18 03:29:00 <emeb> hmm.
Oct 18 03:29:05 <ds2> lower spec is 2.0V
Oct 18 03:29:13 <emeb> Sounds about right.
Oct 18 03:29:23 <ds2> newer sensors are 1.8V only
Oct 18 03:29:54 <emeb> I/O not compatible?
Oct 18 03:30:04 <ds2> max Vdd IO is 1.8V
Oct 18 03:30:19 <ds2> level translation out the wazoo :(
Oct 18 03:30:24 <emeb> no kidding.
Oct 18 03:30:34 <ds2> not to mention the sea of LDOs
Oct 18 03:30:45 <emeb> heh.
Oct 18 03:31:09 <ds2> this is where you realize how big SOT-23's are
Oct 18 03:31:27 <emeb> not so tiny when you need a few dozen?
Oct 18 03:31:38 <ds2> not so tiny when you need like 2-3
Oct 18 03:31:59 * ds2 *heart* single chip power path managers
Oct 18 03:32:13 <ds2> PFETs, LDOs, etc wind up eatting a lot of space
Oct 18 03:32:17 <emeb> ka6sox: borked the kernel build?
Oct 18 03:32:48 <ka6sox> yes, I forgot a patch
Oct 18 03:32:59 <emeb> ds2: yeah - found myself wanting some 4x PFETs the other day.
Oct 18 03:33:07 <ds2> 4x? H bridge?
Oct 18 03:33:29 <emeb> ds2: no - was looking for some anode drivers for LEDs.
Oct 18 03:33:37 <ds2> ah
Oct 18 03:33:53 <ds2> think 3 SOT-23's is bigger then a 144 pin BGA :(
Oct 18 03:36:12 <emeb> world needs PFETs in 0402-sized packages. :)
Oct 18 03:36:29 <ds2> I prefer 0201 sized
Oct 18 03:37:56 <movax|mac> 0201 is hilarious...i have to send 100% overage to the assembly house for them
Oct 18 03:38:15 <ds2> assembly houses usually have them instock
Oct 18 03:38:24 <ds2> easier to pay them for it then to deal with partial tapes
Oct 18 03:39:23 <movax|mac> eh we usually send part kits, cheaper for us than paying for turn-key
Oct 18 03:39:46 <ds2> I don't get charged extra for buying parts from them if they have it stock
Oct 18 03:39:59 <ds2> so I kit the actives and criticals then buy the passives off them
Oct 18 03:40:26 <ds2> only time they charge extra is if they need to go through purchasing
Oct 18 03:40:39 <ka6sox> fleas on rats
Oct 18 03:40:50 <mranostay> hi all
Oct 18 03:41:16 <emeb> so you're saying that the pick-n-place machines have a 50% failure rate on placing 0201 parts?
Oct 18 03:41:41 <mranostay> ouch
Oct 18 03:41:54 <mranostay> emeb: btw what is the FETs for on the FPGA cape?
Oct 18 03:42:12 <emeb> mranostay: no FETs on the FPGA cape.
Oct 18 03:42:44 <movax|mac> no idea, that's what the house asks for
Oct 18 03:42:46 <ka6sox> fets on bcc?
Oct 18 03:43:05 <movax|mac> plx wants 0201s for their suggested bypassing scheme, so beit
Oct 18 03:43:20 <emeb> ka6sox: yep - no fets on bcc.
Oct 18 03:43:21 <mranostay> emeb: ok right  bottom corner. what are those?
Oct 18 03:43:36 <ka6sox> bottom left?
Oct 18 03:43:41 <ka6sox> power supply
Oct 18 03:43:41 <emeb> left bottom corner are the voltage regulators
Oct 18 03:43:56 <emeb> 3.3V and 1.2V from 5V on the 'bone.
Oct 18 03:44:27 <mranostay> er yeah left ..
Oct 18 03:44:34 <mranostay> ah makes sense
Oct 18 03:44:54 <mranostay> emeb: the 3.3V LDO isn't good enough from the bone?
Oct 18 03:45:08 <ka6sox> not enought POWER
Oct 18 03:45:18 <emeb> mranostay: it's best not to load the bone supply down too much.
Oct 18 03:45:31 <emeb> the FPGA can draw a _lot_ of current.
Oct 18 03:45:40 <movax|mac> what kind of fpga?
Oct 18 03:45:46 <emeb> Xilinx S3A
Oct 18 03:46:07 <movax|mac> probably want a dedicated switcher for it, i'd imagine
Oct 18 03:46:23 <movax|mac> i used an ltc3811 for a massive fpga project, but that's probably a bit larger than a spartan-3a
Oct 18 03:46:27 <ds2> probally need its own coal fired generator
Oct 18 03:46:29 <emeb> movax|mac: probably not needed for this smaller part.
Oct 18 03:46:49 <movax|mac> nope, but like you said, probalby not good idea to run it off the ldo on the bone
Oct 18 03:46:50 <ka6sox> we used that for a V5 project
Oct 18 03:46:51 <emeb> this is just a 200A part w/ a 50MHz clock - haven't seen it draw more than about 400ma
Oct 18 03:47:51 <ds2> a 200A part?!
Oct 18 03:47:59 <ds2> does it use miniturized tubes?
Oct 18 03:48:07 <emeb> ark ark... XC3S200A
Oct 18 03:48:14 <ds2> oh
Oct 18 03:51:26 <ka6sox> heh
Oct 18 03:52:41 <ka6sox> home time
Oct 18 03:52:42 <ka6sox> bbla
Oct 18 03:52:46 <emeb> l8r
Oct 18 04:14:19 <dezso> hello
Oct 18 04:14:28 <dezso> any wince6 guy here?
Oct 18 04:16:01 <ds2> ewwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
Oct 18 04:16:16 <movax|mac> i used to have a winmo phone
Oct 18 04:16:23 <movax|mac> then i got a 4s and life was so much better
Oct 18 04:17:18 <emeb> boy is the the wrong place to ask that question.
Oct 18 04:17:41 <ds2> it is entertaining.. wolve dens need sheeps
Oct 18 04:17:45 <dezso> me?
Oct 18 04:18:00 <emeb> ;)
Oct 18 04:18:24 <dezso> so where should I got get some WINCE help ?
Oct 18 04:18:57 <movax|mac> 2005
Oct 18 04:18:59 <emeb> MS Dev forums? just a thought. Most folks here are FOSS.
Oct 18 04:19:06 <movax|mac> (MSDN is a good place to start)
Oct 18 04:19:16 <movax|mac> what do you need specifically, programming help? should be able to use .net cf for most things
Oct 18 04:20:10 <dezso> MS forum dont care about BB dev, I'm not capable to make my own OS, just need a working WINCE7 image wath is actualy works, not like the demo loaded by mouser..
Oct 18 04:20:54 <movax|mac> uh what
Oct 18 04:21:13 <muhoo> movax|mac: +1
Oct 18 04:21:17 * mranostay opens beer
Oct 18 04:21:23 <dezso> to bad I was holding on to some hopes that some one sonner or later will release a working image or BSP, now 1 year later I have not user this board for more than 2hr
Oct 18 04:21:26 * mranostay puts popcorn in microwave
Oct 18 04:21:33 <dezso> anyone want to buy it ?
Oct 18 04:21:37 <ds2> does this beer make your counter tick, mranostay?
Oct 18 04:21:39 <movax|mac> what board
Oct 18 04:21:48 <movax|mac> a wince running board?
Oct 18 04:21:54 <dezso> BeagleBoard xm c4
Oct 18 04:22:32 <movax|mac> ahh
Oct 18 04:22:38 <movax|mac> people run windows on that?
Oct 18 04:22:52 <ds2> windows run?
Oct 18 04:22:54 <dezso> it will run android and some flawore of linux linux but I dont work onthem
Oct 18 04:23:02 <movax|mac> interesting
Oct 18 04:23:18 <dezso> I would like to run WinCE 6 or 7
Oct 18 04:23:28 <movax|mac> i'm a windows/microsoft fan...my favourite dev tools + documentation comes from them
Oct 18 04:23:39 <movax|mac> i use all sorts of tools/Oses, but my stress level is lowest when working with MS toolkits
Oct 18 04:23:52 <movax|mac> it looks like ther eis a CE6 BSP
Oct 18 04:24:00 <movax|mac> first google result: http://beagleboard.org/project/Beagleboard+Windows+CE6.0+R3/
Oct 18 04:24:15 <dezso> Have a mini6410 running on wince6 and works really good for 2 years now, finding it a bit slow so I taught will be a nice upgrade
Oct 18 04:24:58 <ds2> upgrade would be running Linux on an ARM7TDMI
Oct 18 04:25:05 <ds2> hell, an abacus would be an upgrade
Oct 18 04:25:31 <movax|mac> well not if his software stack is all windows based i guess
Oct 18 04:26:00 <Khac> 2222
Oct 18 04:27:05 <dezso> "free in a binary evaluation version" with very limited functionality, berly usable, no presistent registry no hw accel, cant even play mp3 file..
Oct 18 04:27:11 <ds2> it would still be an upgrade
Oct 18 04:27:20 <ds2> difference between running and not running at all
Oct 18 04:29:15 <dezso> Friendlyarm was very helpfull to achive what I want wint the min2440 and 6410, to bad I din't buy the Mini210 with capacitive LCD....
Oct 18 04:31:31 <dezso> is there any other support channel where wince would be more welcomed ?
Oct 18 04:31:53 <movax|mac> no idea dude
Oct 18 04:32:07 <ds2> wince is never welcomed.
Oct 18 04:32:12 <ds2> hence it is called wince ;)
Oct 18 04:32:42 <dezso> ok than, have fun macing and linuxing, ty all
Oct 18 04:33:17 * movax|mac sleep
Oct 18 04:33:19 <movax|mac> laters
Oct 18 05:21:27 <ds2> arrrgggggggggg
Oct 18 05:21:33 <ds2> emeb: @#$#@%$@#$*!@#$@#!&*$(@&!$*(#@&*(&#@$(*@
Oct 18 07:09:46 <tasslehoff> koen: what in systemd should do the job that save-rtc.sh did in a sysvinit/angstrom? I can't adjust the time backwards on my system, since it seems nothing updates /etc/timestamp on shutdown.
Oct 18 07:10:38 <koen> tasslehoff: opkg install timestamp-service
Oct 18 07:14:24 <woglinde> gm tasslehoff and koen
Oct 18 07:14:40 <tasslehoff> koen: just grepped my way to it on my target. now on to figuring out why it doesn't work for me
Oct 18 07:14:42 <tasslehoff> morning woglinde
Oct 18 07:15:53 <koen> tasslehoff: try 'load-timestamp --save'
Oct 18 07:15:59 <koen> and see if it updated something
Oct 18 07:17:47 <tasslehoff> koen: it did. here be dragons.
Oct 18 07:52:48 <_tasslehoff_> koen: I think there's a bug in load-timestamp.sh. If the date has been set back, it will not work.
Oct 18 07:53:11 <_tasslehoff_> it will read the current /etc/timestamp and write it back again :)
Oct 18 08:01:20 <koen> right
Oct 18 08:01:24 <koen> it doesn't allow going back in time
Oct 18 08:01:43 <koen> you're the one with the rtc that initializes in the future?
Oct 18 08:02:28 <_tasslehoff_> koen: yes and no. think I have seen some messages about it, but I have never complained to anyone :)
Oct 18 08:03:47 <koen> I was planning to deprecate timestamp-service till I found out some "features" in ext4
Oct 18 08:04:16 <koen> short version: fs created in the "future" triggers all kind of nasty issues
Oct 18 08:04:51 <Russ> yup, and stupid fsck, stop throwing warnings just because rtc is out of whack
Oct 18 08:21:07 <_tasslehoff_> koen: haven't moved to ext4 yet. plan is to do it after the current release
Oct 18 08:21:36 <_tasslehoff_> koen: will you fix, or should I send a patch?
Oct 18 08:33:28 <koen> _tasslehoff_: send a patch to angstrom-devel and I will have a look at it
Oct 18 12:09:22 <Russ> 'Unknown individuals are contacting Southern California Gas Company (SoCalGas®) customers claiming President Barack Obama will directly pay the utility bills or provide credits.'
Oct 18 12:09:53 <Russ> *sigh*, really people?
Oct 18 12:44:16 <jkridner_> gm all
Oct 18 12:47:05 <keesj> Hi
Oct 18 12:49:41 <jkridner_> hello keesj
Oct 18 12:52:56 <woglinde> hi jkrdiner
Oct 18 13:01:23 <jkridner_> hi woglinde
Oct 18 13:29:40 <mdp> Russ, sounds similar to the obamaphone craze...that video was shot not too far from me ;)
Oct 18 13:42:02 <georgem> has anyone managed to get 3.7rc1 to boot on the beaglebone with any gcc 4.6.0 through 4.7.2?
Oct 18 13:42:55 <mru> I believe koen has
Oct 18 13:49:27 <georgem> I talked to him a couple of days ago and he said to use 4.5.4.
Oct 18 13:52:44 <mru> what happens with 4.7?
Oct 18 13:56:53 <georgem> I haven't tried 4.7. With 4.6.2 it just reboots as soon as I do bootm
Oct 18 13:57:52 <koen> mru: nothing happens, that's the problem
Oct 18 13:57:58 <koen> mru: hang after "starting kernel"
Oct 18 13:58:26 <mru> unaligned crap?
Oct 18 13:58:35 <georgem> I was just wondering if anyone else happened to build it with a newer toolchain and actually had any luck.
Oct 18 13:58:36 <mru> I'm so tired of that bullshit
Oct 18 13:59:48 <LetoThe2nd> mru: shullbit?
Oct 18 14:00:19 <georgem> or if anyone has managed to get anything from mach-omap2 to boot with 4.6+
Oct 18 14:07:27 <stahl> how can i access cloud9 from ssh? where are the files located?
Oct 18 14:09:19 <LetoThe2nd> georgem: certainly.
Oct 18 14:10:00 <LetoThe2nd> georgem: i'm pretty sure i have rebuilt vanilla and preempt rt for pandaboard with 4.6something
Oct 18 14:10:18 <georgem> LetoThe2nd: 3.7rc1?
Oct 18 14:10:27 <mdp> georgem: I use gcc version 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu5) on 3.7-rc1
Oct 18 14:10:35 <mdp> the ubuntu packaged version
Oct 18 14:10:45 <LetoThe2nd> georgem: only tested up to 3.4.something
Oct 18 14:11:19 <mdp> georgem, that's on beaglebone and am335x-evm, fwiw
Oct 18 14:11:37 <georgem> mdp: and they booted? interesting
Oct 18 14:11:53 <mdp> georgem, I've been using it for months for my daily devel on those platforms
Oct 18 14:12:20 <mdp> so I'll have to say it boots quite well
Oct 18 14:12:56 <georgem> mdp: ok, well I'll newer 4.6.x linaro then and see what I get with that.
Oct 18 14:13:00 <georgem> try*
Oct 18 14:13:35 <mdp> georgem, I know somebody using the 4.7.x linaro that's tested some of my patches that confirms it working as well on the same am335x stuff
Oct 18 14:14:11 <mdp> georgem, one thing, when you say 3.7-rc1...are you referring to the stock mainline version...or is this the beaglebone-3.7 tree/branch?
Oct 18 14:15:20 <georgem> I'm currently trying the beaglebone-3.7 tree. But if they managed to get it to work on stock mainline I could certainly try that out too and see if I get different results
Oct 18 14:16:45 <mdp> ok, unless you are targeting inclusion of a patch in mainline, probably no need to try to use stock 3.7-rc1...but I wanted to make sure because irregardless of toolchain...stock 3.7-rc1 won't boot on beaglebone due to bug...that's fixed for queued for -rc2
Oct 18 14:46:11 <alan_o> so if I said that doing GPIO using mmap() isn't really the "right" way to do it, would you be offended?
Oct 18 14:46:54 <alan_o> I think I offended an OEM rep by saying that yesterday.
Oct 18 14:49:17 <ka6sox> that sounds like bitbanging, writ large...
Oct 18 14:51:22 <alan_o> don't worry, I'm not bit-banging :)
Oct 18 14:51:34 <alan_o> It's not an RPi
Oct 18 14:59:50 <arcanescu> well on ubuntu this the armhf toolchain is missing gcc-4.7-arm-linux-gnueabihf has it been replaced by something else or what exactlyu?
Oct 18 15:06:51 <mdp> alan_o, people live to be offended by little things, no worries
Oct 18 15:41:13 <alan_o> #beagle poll: My toolchain of choice comes from .... ?
Oct 18 15:42:17 <alan_o> (I've been using ct-ng toolchains, but had to switch to a codesourcery one for a project after I had some non-bootable kernels yesterday)
Oct 18 15:42:30 <alan_o> haven't investigated what's wrong with mine :(
Oct 18 16:23:37 <Cuong> hello everyone, I got a problem when build my QT project from Ubuntu Host for beagleboard xM target
Oct 18 16:23:51 <Cuong> this is my make file:
Oct 18 16:24:02 <Cuong> #!/bin/bash /opt/qt-dm3730/bin/qmake -spec /opt/qt-dm3730/mkspecs/qws/linux-dm3730-g++ -o Makefile *.pro  make clean make
Oct 18 16:24:33 <Cuong> and I got this problem: http://pastebin.com/5kHzAkWf
Oct 18 16:24:44 <Cuong> Thank you so much
Oct 18 16:34:03 <Cuong> Anyone here?
Oct 18 17:11:05 <mranostay> nobody here
Oct 18 17:11:46 <mdp> yesterday there was nobody here that could solve my problem, I had to figure it out myself.
Oct 18 17:11:51 <mdp> useless channel
Oct 18 17:24:10 <Cuong> anyone hear?
Oct 18 17:24:18 <woglinde> no
Oct 18 17:24:35 <woglinde> I  cant hear
Oct 18 17:25:24 <Cuong> yes, I got a prob. when build for BBxm target
Oct 18 17:27:12 <woglinde> building what?
Oct 18 17:28:09 <Cuong> yes, please: http://pastebin.com/Hu8zcfzS
Oct 18 17:28:43 <Cuong> I am build a short program using openCV for my BB-xM running Angstrom from my Ubuntu host
Oct 18 17:29:21 <woglinde> you can read nor?
Oct 18 17:29:23 <woglinde> could not read symbols: File in wrong format
Oct 18 17:30:01 <Cuong> yes, please show me why
Oct 18 17:30:07 <Cuong> Im very new
Oct 18 17:30:29 <Cuong> because when I use g++ build for Ubuntu host is no prob.
Oct 18 17:30:47 <woglinde> you are mixing two archs
Oct 18 17:30:59 <woglinde> host proably 64bit intel stuff
Oct 18 17:31:00 <woglinde> and arm#
Oct 18 17:31:06 <woglinde> that do not work
Oct 18 17:31:26 <Cuong> yes, so I using cross compiler for Arm
Oct 18 17:31:40 <woglinde> what OS do you run on the beagle?
Oct 18 17:31:43 <Cuong> you can see here: /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/bin/g++
Oct 18 17:31:49 <Cuong> Angstrom
Oct 18 17:31:50 <woglinde> yes
Oct 18 17:32:27 <woglinde> for your little programm I would compile on the board it self
Oct 18 17:33:05 <Cuong> Yes, but can I build via host?
Oct 18 17:33:20 <woglinde> if you want to stay with crosscompiling check the output of "file /usr/local/lib/libopencv_calib3d.so"
Oct 18 17:33:23 <woglinde> sure
Oct 18 17:33:28 <woglinde> opkg install g++
Oct 18 17:33:31 <woglinde> opkg update
Oct 18 17:33:39 <woglinde> opkg list | grep opencv
Oct 18 17:33:50 <Cuong> yes, I do it rightaway
Oct 18 17:33:51 <woglinde> opkg install all_opencv_dev_packages_i_need
Oct 18 17:33:59 <Cuong> thank you so much.
Oct 18 17:34:06 <woglinde> no prob
Oct 18 17:34:15 <Cuong> yes, one more plz
Oct 18 17:34:35 <Cuong> how can I fix the above bug when build via host
Oct 18 17:34:55 <woglinde> set the right search path for opecv
Oct 18 17:34:59 <woglinde> with -L
Oct 18 17:35:26 <woglinde> proably -L  /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/lib
Oct 18 17:36:04 <Cuong> Add it to the command?
Oct 18 17:37:10 <woglinde> yes
Oct 18 17:37:19 <woglinde> hm oh mom
Oct 18 17:37:28 <Cuong> where can I add to: /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/bin/g++ -ggdb `pkg-config --cflags opencv` main.cpp -o main `pkg-config --libs opencv`
Oct 18 17:37:53 <Cuong> Please, Im new linux. :D
Oct 18 17:37:54 <woglinde> whats the output of pkg-config --libs opencv ?
Oct 18 17:38:18 <woglinde> that has nothing todo with linux really
Oct 18 17:38:24 <woglinde> its compiling
Oct 18 17:38:35 <woglinde> would be the same under macos or bsd
Oct 18 17:39:16 <woglinde> your problem is pkgconfig searches under /usr/local
Oct 18 17:39:30 <woglinde> not /usr/local/arm
Oct 18 17:39:51 <Cuong> yes, brighter
Oct 18 17:40:03 <Cuong> thank you so much.
Oct 18 17:41:17 <woglinde> ypu know shell enviroment variables?
Oct 18 17:41:20 <woglinde> you need
Oct 18 17:41:22 <woglinde> PKG_CONFIG_SYSROOT_DIR
Oct 18 17:41:27 <Cuong> yes, please
Oct 18 17:41:37 <woglinde> so pkgconfig searches the arm dir
Oct 18 17:41:54 <Cuong> yes, so exciting than windows
Oct 18 17:42:16 <Cuong> If you have free time, please show me anything..
Oct 18 17:42:56 <woglinde> this article should help you
Oct 18 17:42:58 <woglinde> http://www.flameeyes.eu/autotools-mythbuster/pkgconfig/cross-compiling.html
Oct 18 17:43:28 <Cuong> yes..
Oct 18 17:43:42 <Cuong> a bunch of thannks
Oct 18 17:44:58 <woglinde> no prob again
Oct 18 17:50:35 <georgem> beaglebone-3.7 didn't boot when compiled with linaro-4.7-2012.10. My binunits is pretty old though maybe it needs to be updated.
Oct 18 18:26:19 <SilicaGel> xbee programmable: nifty.
Oct 18 18:26:23 * SilicaGel Do not Eat
Oct 18 18:34:27 <mranostay> someone is still drunk it seems
Oct 18 18:44:34 <woglinde> hi silicalgel
Oct 18 18:45:51 <ds2> it is such a pleasant surprise to find people making connectors with pin 1 on the other end
Oct 18 18:46:45 <SilicaGel> hi woglinde, how goes the war
Oct 18 18:47:14 <woglinde> which war?
Oct 18 18:49:07 <SilicaGel> hm, let's say, the make openjdk fast and reliable in arm war?
Oct 18 18:49:19 <mdp> mranostay, remember "beer lunch"
Oct 18 18:51:59 <woglinde> silicalgel haha
Oct 18 18:52:48 <Russ> ds2, before or after spin?
Oct 18 19:03:37 <ds2> Russ: board has wrong layout but I found another source for a connector that matches board
Oct 18 19:03:41 <ds2> hence pleasant surprise
Oct 18 19:03:52 <ds2> of course, connector is about 2x as much
Oct 18 19:20:50 <jsabeaudry> Hardware question for FPGA gods
Oct 18 19:21:36 <jsabeaudry> Will the absence of coupling capacitors on a certain io bank have an effet if the io bank is configured in input?
Oct 18 19:27:03 <Russ> jsabeaudry, hmm
Oct 18 19:28:13 <Russ> jsabeaudry, it might if the caps are used for any internal IO switching, or if the signal rings and goes beyond the rails
Oct 18 19:28:29 <Russ> (return current though clamping diodes)
Oct 18 19:33:13 <jsabeaudry> Russ, a ringing signal is exactly what I have
Oct 18 19:33:21 <jsabeaudry> Russ, thank you!
Oct 18 19:35:12 <jsabeaudry> I'm doing fpga-beaglebone communication through gpmc but that thing shines (EMI wise) in the powercord!
Oct 18 19:36:30 <jsabeaudry> Any hunch on what could be an issue?
Oct 18 19:38:40 <Russ> I'm not an emi expert, but you could put a cheater coil on it
Oct 18 19:39:11 <Russ> squelching emi from cables leaving a device is hard
Oct 18 19:43:15 <djlewis> ferrite is your friend
Oct 18 19:44:55 <jsabeaudry> I'm thinking that logical transitions on the gpmc should not shine so much in the power cord, and that something must be very wrong
Oct 18 19:45:07 <jsabeaudry> We tried ferrite, little to no effect
Oct 18 19:45:10 <stahl> i updated bonescript via ssh (git pull), now i can't run blinkled.js anymore. Error: Cannot find module 'cluster'. please help..
Oct 18 19:45:20 <jsabeaudry> I'll try to see what cheater coils are
Oct 18 19:47:54 <jsabeaudry> Russ, is there another name for "cheater coils"?
Oct 18 19:51:08 <XorA|gone> is the FPGA on a different PSU?
Oct 18 19:52:18 <jsabeaudry> XorA|gone, Hmmm, I think it might be on a different psu, I think it's running on 3.3V and not the 5 like the bb
Oct 18 19:52:32 <Russ> sorry, coil is the wrong name for it
Oct 18 19:53:08 <XorA|gone> jsabeaudry: you might need a choke on the ground line then, as the potential of gnd on FPGA board and beagle will tend to vary with the relative loads causing noise
Oct 18 19:53:41 <XorA|gone> its called a current compensated choke or something like that
Oct 18 19:54:06 <Russ> you might just try a snap on filter
Oct 18 19:54:31 <jsabeaudry> XorA|gone, excellent, I will look into that option too
Oct 18 19:55:09 <Russ> if you have communication board to board, I'd be wary of putting a choke inline with the return (ground)
Oct 18 19:55:21 <Russ> so long as it's the right frequency range, you should be ok
Oct 18 19:55:51 <Russ> hmm..suppose it isn't the return, there are probably a bunch of ground signals already in the gpmc bus, no?
Oct 18 19:56:04 * XorA|gone would normally in this situation grab the nearest EE :-)
Oct 18 19:57:44 <mranostay> and beat them?
Oct 18 19:57:51 <jsabeaudry> heh
Oct 18 19:57:53 <Russ> is the power cord going between the boards, or from the boards to something else?
Oct 18 19:57:54 * mru hides
Oct 18 19:58:28 * mranostay drags mru out
Oct 18 19:58:30 <emeb> jsabeaudry: what's the length of the connection from bb to fpga board?
Oct 18 19:58:44 <Crofton> jsabeaudry, what is the actual problem you are trying to solve
Oct 18 19:58:47 <jsabeaudry> Russ, the power cord to the massive power adapter that plug in the wall socket, that cord is far from the boards
Oct 18 19:58:47 * mru kicks and screams
Oct 18 19:59:00 <jsabeaudry> Crofton, emi
Oct 18 19:59:07 <Crofton> define emi
Oct 18 19:59:28 <Russ> are there any other cords leaving the board?
Oct 18 19:59:43 <jsabeaudry> emeb, probably close to 5 cm chip to chip
Oct 18 20:00:00 <emeb> jsabeaudry: that's not too awful.
Oct 18 20:00:19 <emeb> One thing about the bb I/O connectors though is that there aren't many ground lines.
Oct 18 20:00:40 <emeb> Not good for decent signal integrity - you'll want to keep things as short as possible.
Oct 18 20:01:51 <Russ> and terminate well :)
Oct 18 20:02:33 <jsabeaudry> Crofton, electromagnetic emissions, I'm not an expert at this, but it's part of some kind of certification
Oct 18 20:02:47 <emeb> ideally, yes. Unfortunately the bb has no source terminations. FPGA might support load termination, but usually not for single-ended signals as found on bb.
Oct 18 20:03:30 <Russ> it may not help that much anyway if the fpga doesn't have any decoupling caps on that io bank
Oct 18 20:03:43 <emeb> that's not good.
Oct 18 20:03:47 <jsabeaudry> emeb, it does, but the terminations supported have a too low impedence and the bb does not have enough output current to drive them
Oct 18 20:04:26 <emeb> jsabeaudry: yeah - that's a problem. the I/O specs on the FPGAs are made for things like LVDS which is not what bb outputs do.
Oct 18 20:04:29 <jsabeaudry> Russ, would you say coupling caps are more important than line terminations?
Oct 18 20:04:30 <Crofton> but communciation is fine, just radiated is the issue?
Oct 18 20:05:02 <jsabeaudry> Crofton, precisely
Oct 18 20:05:06 <Russ> jsabeaudry, if the terminations are in the fpga, without decoupling caps, they can't really do their job
Oct 18 20:05:17 <Russ> the worse your SI, the worse your EMI
Oct 18 20:05:33 <Crofton> and you have it in a matel box?
Oct 18 20:05:34 <jsabeaudry> SI ?
Oct 18 20:05:46 <Russ> signal integrity, rining, overshoot, etc
Oct 18 20:05:49 <Crofton> metal box
Oct 18 20:06:15 <Crofton> basically, it seems like all the things you want to do, will degrade actual interface performance
Oct 18 20:06:19 <Crofton> which is bad
Oct 18 20:06:23 <jsabeaudry> Crofton, yes, tests revealed that the source of the emi is really the powercoard that sticks out of the box
Oct 18 20:06:41 <emeb> jsabeaudry: does that power cord supply both the bb and fpga boards?
Oct 18 20:06:42 <Crofton> ok, so filter there
Oct 18 20:06:42 <Russ> can you tie the power cord ground to the metal box at point of entry?
Oct 18 20:07:01 <Crofton> start with a ferrite
Oct 18 20:07:10 <Russ> he apparently tried a ferrite
Oct 18 20:07:15 <Crofton> also, if this is CE testing, laod on image that turns everything off
Oct 18 20:07:33 * Crofton is not making a joke
Oct 18 20:07:35 <jsabeaudry> Crofton, isn't that cheating?
Oct 18 20:07:48 <jsabeaudry> Not exactly "normal operation" to me
Oct 18 20:08:10 <emeb> http://www.corcom.com/Series/PEM/C/
Oct 18 20:08:16 <Russ> Crofton, I think you were supposed to make that suggestion in the form of a wink wink nudge nudge
Oct 18 20:09:06 <XorA|gone> a lot of CE testing is done in "normal" operation, why do you think so many decives first thing that happens is firmware upgrades?
Oct 18 20:09:09 <jsabeaudry> I'm pretty sure one of the hw guys messed up really bad for this kind of situation to happen, no coupling caps on io bank seems pretty bad, looking at other options too, a brainstorm if you will
Oct 18 20:09:49 * XorA|gone sees so many firmwares release a week after device that jacks up clock speeds
Oct 18 20:13:12 <jsabeaudry> hehe
Oct 18 20:14:37 <XorA|gone> if device is not normally supplied with a PSU, run it on battery for the CE test
Oct 18 20:15:19 <ds2> test it with a foil case ;)
Oct 18 20:17:22 <jsabeaudry> ds2, yes a good dose of foil around the power cord right up to the wall socket :P
Oct 18 20:20:57 <jsabeaudry> Thank you so much for all the ideas Russ, emeb, XorA|gone, Crofton, djlewis, ds2, it is much appreciated!
Oct 18 20:36:31 <ds2> geez github is slow today
Oct 18 20:38:08 <Crofton> ds2,  GitHub ‏@github
Oct 18 20:38:08 <Crofton> We are experiencing issues due to a DDOS attack, working hard to restore service...
Oct 18 20:38:13 <Crofton> from twitter
Oct 18 20:40:35 <ds2> oh doh
Oct 18 20:40:46 <ds2> Crofton: thanks
Oct 18 21:10:45 <mranostay> Crofton: nice..
Oct 18 21:14:46 <woglinde> hm github was fast here
**** ENDING LOGGING AT Fri Oct 19 03:00:01 2012