**** BEGIN LOGGING AT Thu Jul 05 02:59:58 2012
Jul 05 05:38:37 <mranostay> ok monday #2 gets underway
Jul 05 05:42:18 <mandar> hii guys i just recieved my beaglebone and i am connecting it to pc using screen /dev/`dmesg | grep FTDI | grep "now attached to" | tail -n 1 |awk '{ print $NF }'` 115200 command in terminal... can i redirect debug messages to serial port or how can i connect beaglebone using minicom?
Jul 05 06:51:51 <trelane> mandar: read the dmesg log find the console device that your script up there generates, and minicom -s set up the console for 115200
Jul 05 06:51:53 <trelane> go from there :)
Jul 05 07:07:58 <Tamilarasi> hi all,i trying to run SPI on my beaglebone,code has been written but 'spidev' module unable to get to run in user space please give me solution.
Jul 05 07:08:28 <aholler> modules never run in userspace
Jul 05 07:08:48 * LetoThe2nd particularly likes the information "code has been written"
Jul 05 07:09:35 <Tamilarasi> I am not running module, in /dev/spidev is not found.
Jul 05 07:10:09 <Tamilarasi> By using this spidev i want to execute my program.
Jul 05 07:10:56 <mranostay> LetoThe2nd: "it just works" as well :)
Jul 05 07:11:08 <LetoThe2nd> so you basically just miss /dev/spidev?
Jul 05 07:11:36 <Tamilarasi> yes..
Jul 05 07:12:04 <Tamilarasi> i found this following web link..http://www.kernel.org/doc/Documentation/spi/spidev_test.c
Jul 05 07:12:10 <LetoThe2nd> mranostay: in a channel i used to hang out, the bot responded to "it doesn't work" with "what do you mean by 'it doesn't work'? does it sit at home all day on the couch?"
Jul 05 07:13:15 <aholler> finally a usefull bot
Jul 05 07:13:23 <LetoThe2nd> Tamilarasi: googling "beagleboard enable spidev" seems to bring up a lot of useful information.
Jul 05 07:14:09 <LetoThe2nd> aholler: indeed.
Jul 05 08:05:02 <stuk_gen> if i use gst-launch to play a video, it play very well, very speed. But if i use qt phonon the same video is too slow. Qt phonon is not only a wrapper for gst?
Jul 05 08:52:03 <stuk_gen> running /etc/init.d/gstti start return CMEMK Error: Failed to request_mem_region(0x86300000, 16777216)
Jul 05 08:52:15 <stuk_gen> what's this mean?
Jul 05 09:06:13 <_av500_> that something wiht CMEM is not OK
Jul 05 09:06:21 <_av500_> does lsmod show cmem?
Jul 05 09:10:07 <stuk_gen> _av500_: no there isn't in lsmod
Jul 05 09:46:40 <_av500_> how does gst-launch work then?
Jul 05 09:47:02 <_av500_> to play a video very well
Jul 05 09:48:58 <stuk_gen> _av500_: i don't know
Jul 05 09:49:21 <stuk_gen> i using gst-launch-0.10 filesrc location=/usr/share/ti/video/HistoryOfTI-WVGA.m2v ! mpegvideoparse ! ffdec_mpeg2video ! ffmpegcolorspace ! fbdevsink
Jul 05 09:56:36 <_av500_> ok
Jul 05 09:56:42 <_av500_> that is not using the dsp at all
Jul 05 09:58:58 <stuk_gen> _av500_: what package i need to use dsp?
Jul 05 09:59:13 <mru> _av500_: they finally renamed ffmpegcolorspace in 1.0
Jul 05 10:12:40 <_av500_> to what?
Jul 05 10:12:45 <_av500_> slowcolorspace?
Jul 05 10:12:53 <stuk_gen> nice
Jul 05 10:15:01 <stuk_gen> the problem is this cmem
Jul 05 10:15:13 <stuk_gen> [ 1039.561981] CMEMK Error: Failed to request_mem_region(0x86300000, 16777216)
Jul 05 10:15:13 <stuk_gen> FATAL: Error inserting cmemk (/lib/modules/3.2.18/kernel/drivers/dsp/cmemk.ko): Bad address
Jul 05 10:31:38 <_av500_> yes
Jul 05 10:31:50 <_av500_> adress given to cmem load has to match kernel memory config
Jul 05 10:31:54 <_av500_> and dsp server build
Jul 05 10:33:48 <aholler> drill a hole into the memory ;)
Jul 05 10:34:47 * dm8tbr hands the TNT
Jul 05 10:34:56 <dm8tbr> or if in doubt, nuke it from orbit
Jul 05 10:36:19 <_av500_> stuk_gen: what is your mem= in kernel cmdline?
Jul 05 10:36:42 <aholler> 33mm above 0
Jul 05 10:43:17 <stuk_gen> _av500_: optargs="mem=99M@0x80000000 mem=384M@0x88000000"
Jul 05 10:48:03 <aholler> cat /proc/cmdline to verify it is there
Jul 05 10:50:48 <stuk_gen> only omapdss.def_disp=dvi root=/dev/mmcblk0p2 ro rootfstype=ext4 rootwait
Jul 05 10:51:50 <aholler> so you've found the problem
Jul 05 11:03:32 <_av500_> cat /proc/cmdline
Jul 05 11:03:36 <_av500_> ah
Jul 05 11:10:35 <mathur> i have a problem while using my beaglebone. I have used the elinux.org/BeagleBoardUbuntu page to install oneiric prebuilt image on the beaglebone...the kernel tries building for some time and then gives me the following error
Jul 05 11:10:53 <mathur>     1.569763] loop: module loaded [    1.573242] at24 1-0051: 32768 byte 24c256 EEPROM, writable, 64 bytes/write [    1.633148] No daughter card found [    1.636779] at24 1-0050: 32768 byte 24c256 EEPROM, writable, 64 bytes/write [    1.652069] AM335X: wrong header 0xffffffff, expected 0xee3355aa [    1.658416] The error message above indicates that there is an issue with [    1.658447] the EEPROM or the EEPROM contents.  After verify
Jul 05 11:12:21 <mathur> ntents, if any, refer to the am335x_evm_setup function in the [    1.658447] arch/arm/mach-omap2/board-am335xevm.c file to modify the board [    1.658477] initialization code to match the hardware configuration
Jul 05 11:31:24 <Qbert_> Has anyone here bitbaked  angstrom with aufs or unionfs root-fs?
Jul 05 11:31:42 <SmashLink> Hey, does somebody know if there's a way of booting from a device other than the SD card in the BeagleBone?
Jul 05 11:32:18 <LetoThe2nd> SmashLink: none supported. you already asked that yesterday, BTW.
Jul 05 11:32:47 <mru> can't it do serial boot?
Jul 05 11:32:48 <aholler> boot from the serial should be possible
Jul 05 11:36:59 <dm8tbr> check the TRM and look for SYSBOOT
Jul 05 11:57:46 <SmashLink> I'd like to be able to fix a corrupted SD card without taking it out and re-writing the image inside... maybe this is possible with a cape
Jul 05 11:58:32 <SmashLink> ssh wouldnt work since you need to fix it even if the board already turned off
Jul 05 12:03:13 <aholler> boot from serial and load a kernel which supports nfs and use nfsroot
Jul 05 12:03:39 <aholler> or include a root in the kernel
Jul 05 12:04:44 <woglinde> aholler booting via serial cable is tricky
Jul 05 12:05:04 <aholler> nobody said it will be easy ;)
Jul 05 12:05:38 <aholler> one-click-solutions aren't given here ;)
Jul 05 12:05:46 <LetoThe2nd> judging from our conversation yesterday, i still stick with: use sd card.
Jul 05 12:06:04 <stuk_gen> boot from eth
Jul 05 12:06:12 <woglinde> LetoThe2nd who told you?
Jul 05 12:06:15 <SmashLink> as i've seen, you can't boot from serial in BeagleBone
Jul 05 12:06:27 <woglinde> stuk_gen without mlo(hint the sdcard is broken)
Jul 05 12:06:41 <woglinde> smashlink hm okay
Jul 05 12:07:11 <stuk_gen> woglinde: how can you start u-Boot whitout a sd card?
Jul 05 12:07:13 <LetoThe2nd> woglinde: nobody told me - the OP was already here yesterday, asking very similar things. and judging from that, i think that anything beside sd card is out of the needed skill set, at least for now.
Jul 05 12:07:54 <woglinde> stuk_gen on omap3530 you could boot a kernel via serial cable
Jul 05 12:08:07 <woglinde> but it had to be very small
Jul 05 12:08:30 <stuk_gen> woglinde: ok but his is not done by u-Boot?
Jul 05 12:08:34 <SmashLink> yeah but i'm not talking now about using a SD card or not (I believe there's no way of booting without it); i'm talking about fixing a corrupted SD card
Jul 05 12:08:43 <_av500_> take it out
Jul 05 12:08:50 <_av500_> put a new one
Jul 05 12:08:55 <woglinde> *g+
Jul 05 12:08:58 <stuk_gen> SmashLink: buy another sd?
Jul 05 12:09:02 <woglinde> hi av500
Jul 05 12:09:17 <SmashLink> well, it's not like the sd card broke; obviously you should buy a new one in this case
Jul 05 12:09:45 <_av500_> well, remove it and write a new image
Jul 05 12:09:51 <SmashLink> i mean the card's files got corrupted. So you have to write a new image
Jul 05 12:10:29 <LetoThe2nd> SmashLink: why would a card only supplying a simple loader get corrupted, except for misuse?
Jul 05 12:10:33 <SmashLink> yeah, but how can you do it without removing the sd card? let's say you haven't access to it, and you can only opperate via the internet
Jul 05 12:10:46 <_av500_> how would you boot from serial then?
Jul 05 12:10:48 <_av500_> or usb?
Jul 05 12:10:51 <_av500_> or anything?
Jul 05 12:10:56 <woglinde> lol
Jul 05 12:11:09 <woglinde> make a pig fly and you have nice bacon
Jul 05 12:11:44 <SmashLink> i don't know, maybe it can't get corrupted if you only have the boot and the kernel inside, but... who knows, i guess it could happen
Jul 05 12:12:20 <SmashLink> an overvoltage can make that kind of things happen
Jul 05 12:13:06 <LetoThe2nd> SmashLink: anything can happen, but you should not try to make the beaglebone behave like something it is not. and it is not a boot-under-any-circumstance-fix-anything-remote-server. its a kit for playing, tinkering, developing.
Jul 05 12:13:52 <woglinde> LetoThe2nd as the other TI would say, "It is a toy"
Jul 05 12:13:58 <LetoThe2nd> SmashLink: if you need something more reliable, you need to come up with better flasha nd bootlogic integrated into board. and that is not there.
Jul 05 12:14:03 <LetoThe2nd> woglinde: exactly.
Jul 05 12:14:22 * LetoThe2nd is not affiliated with TI in any way, just for the protocol.
Jul 05 12:14:23 <_av500_> SmashLink: not what I meant
Jul 05 12:15:14 <_av500_> are you asking a hypothetical question?
Jul 05 12:15:23 <_av500_> or do you have a remote unit that failed?
Jul 05 12:15:36 <woglinde> trolling at the second level
Jul 05 12:15:44 <stuk_gen> :D
Jul 05 12:15:52 <LetoThe2nd> remote trolling?
Jul 05 12:15:54 <_av500_> if you need a reliable system, there are ways
Jul 05 12:16:00 <SmashLink> the beaglebone has enough power to do real good things, and it is very low-cost, besides the documentation is great, so... making it a bit more realiable, like you say, would be great
Jul 05 12:16:12 <_av500_> like adding hardware watchdog
Jul 05 12:16:21 <_av500_> and maybe a 2nd card
Jul 05 12:16:32 <mru> isn't there an onchip watchdog?
Jul 05 12:16:37 <stuk_gen> _av500_: 2nd card for what? if first fails run second?
Jul 05 12:16:37 <woglinde> hm cape with second card?
Jul 05 12:16:37 <_av500_> or the 1st stage checking some gpio to know what image to boot
Jul 05 12:16:47 <_av500_> stuk_gen: for example
Jul 05 12:16:49 * LetoThe2nd suggests a mask programmed rom containing the uboot.
Jul 05 12:16:52 <woglinde> but can the onboard bl read the second card
Jul 05 12:16:59 <_av500_> woglinde: I said adding hardware
Jul 05 12:17:07 <_av500_> sccard mux
Jul 05 12:17:10 <_av500_> sdcard
Jul 05 12:17:21 <_av500_> triggered on boot fail due to hw watchdog timeout
Jul 05 12:17:23 <woglinde> raid on sdcaed?
Jul 05 12:17:25 <_av500_> something like that
Jul 05 12:17:27 <woglinde> sdcard
Jul 05 12:17:36 <_av500_> or add a micro that feeds serial boot to the chip
Jul 05 12:17:48 <mru> lunchpad!
Jul 05 12:17:48 <_av500_> many ways
Jul 05 12:17:59 <_av500_> whether one needs to do that is another question
Jul 05 12:18:04 <woglinde> mirco?
Jul 05 12:18:21 <stuk_gen> mirco?
Jul 05 12:18:30 <mru> keep in mind that adding stuff increases price
Jul 05 12:18:32 <LetoThe2nd> a mircophone, even?
Jul 05 12:18:32 <_av500_> yes, you can have Mirco sit next to it as well
Jul 05 12:18:43 <woglinde> "pleas bone boot from the other sdacrd"
Jul 05 12:18:46 <_av500_> if it fails to boot, give him a call
Jul 05 12:19:30 <aholler> build a cape with a good old eprom ;)
Jul 05 12:19:47 <aholler> the bone really doesn't support boot by serial?
Jul 05 12:19:51 <woglinde> aholler coreboot?
Jul 05 12:20:24 <SmashLink> a cape wouldnt be possible since the board probably went off, so the cape have stopped working
Jul 05 12:20:34 <LetoThe2nd> with mirco, it would be more a co-rebott
Jul 05 12:21:33 <mru> is mirco wearing a cape?
Jul 05 12:21:59 <LetoThe2nd> a batcape? or at least a batcape equation?
Jul 05 12:22:03 <_av500_> SmashLink: easy, 2 boards
Jul 05 12:22:08 <_av500_> one checking the other
Jul 05 12:22:24 <_av500_> a cluster
Jul 05 12:22:27 <mru> and a third for tie-breaking
Jul 05 12:22:48 <LetoThe2nd> make a bone team.
Jul 05 12:22:55 <stuk_gen> now you have a powerfull system that cost only $$$$$ dollars :D
Jul 05 12:22:57 <SmashLink> _av500_ that's increasing the price by two...
Jul 05 12:23:06 <mru> LetoThe2nd: that's called a skeleton
Jul 05 12:23:07 <LetoThe2nd> one doing the work, the others drinking beer and shooting snide comments.
Jul 05 12:23:24 <LetoThe2nd> mru: not if every bone is accompanied by its dedicated mirco
Jul 05 12:23:31 <SmashLink> XD
Jul 05 12:23:32 <_av500_> SmashLink: yes
Jul 05 12:23:36 <stuk_gen> lol
Jul 05 12:23:36 <_av500_> still a bone is cheap
Jul 05 12:23:52 <_av500_> we are just listing options, your call
Jul 05 12:24:16 <SmashLink> yeah and i'm very thankful, this is a hard question, i guess
Jul 05 12:24:50 <_av500_> would you have a way to remotely trigger a reboot?
Jul 05 12:25:16 <stuk_gen> ssh
Jul 05 12:25:26 <_av500_> on crashed system, not
Jul 05 12:25:56 <stuk_gen> on crash...not
Jul 05 12:26:30 <stuk_gen> without watchdog is impossible
Jul 05 12:26:43 <stuk_gen> make this simply...
Jul 05 12:27:11 <SmashLink> but wouldnt it be easy to build a board that only connects to internet, recieves the image and copies it to the SD card?
Jul 05 12:27:28 <stuk_gen> boot from eth0
Jul 05 12:27:33 <stuk_gen> but if crash...
Jul 05 12:27:38 <stuk_gen> you have to reboot that
Jul 05 12:27:46 <SmashLink> oh, i guess that's impossible if you don't extract the sd card -.-
Jul 05 12:28:03 <_av500_> if you have a 2nd board, you can use that to supply the boot image
Jul 05 12:28:14 <_av500_> but you jsut said 2 boards are expensive
Jul 05 12:28:28 <SmashLink> yep that looks like a solution
Jul 05 12:28:41 <SmashLink> no, it's not that expensive... i was just thinking of a cheaper solution
Jul 05 12:29:03 <_av500_> you can pray nothing goes bad, cost: 0$
Jul 05 12:29:17 <aholler> stuk_gen: again, you can't boot from ethernet without some sw which does that
Jul 05 12:29:35 <SmashLink> but anyway, using a second board only for replacing the first one if it breaks... doesn't look very good
Jul 05 12:29:50 <aholler> get something which has nand, e.g. the classic beagleboard
Jul 05 12:30:48 <SmashLink> yeah i know, in the beagleboard is way easier
Jul 05 12:31:03 <stuk_gen> but its cheaper XD
Jul 05 12:31:42 <SmashLink> you can't boot from ethernet in the beaglebone, can't you?
Jul 05 12:32:01 <SmashLink> isn't that called pxe?
Jul 05 12:32:05 <stuk_gen> aholler: you can boot from uBoot from eth0...you have to set up manually some parameters...but i think there's a method to build uBoot with this parameter set up correctly
Jul 05 12:32:31 <aholler> stuk_gen: and from where do you get u-boot?
Jul 05 12:32:47 <xxiao> i need use 2.6.32 for beagleboard, undere angstrom, there is reciples-kernel/linux/linux-omap-psp_2.6.32.bb
Jul 05 12:32:48 <stuk_gen> from the boot of sd card
Jul 05 12:33:07 <SmashLink> btw, what's u-boot exactly? i've been hearing of this all these days
Jul 05 12:33:08 <xxiao> however that recipe is not downloading psp2.6.32 kernel, it's checking out the top of linux-omap3.git
Jul 05 12:33:08 <stuk_gen> but i boot only the rootfs from eth0
Jul 05 12:33:13 <xxiao> why is that?
Jul 05 12:33:42 <xxiao> i inherited some stuff which are only verified to build against 2.6.32 psp kernel
Jul 05 12:33:42 <aholler> stuk_gen: if you have a working sd-card, you can boot a kernel from it too
Jul 05 12:34:18 <stuk_gen> aholler: oh yes you're rught...
Jul 05 12:34:20 <stuk_gen> right...
Jul 05 12:34:48 <stuk_gen> so...bone haven't nand...so if sd card is break...is break and of the game...
Jul 05 12:35:02 <xxiao> SRC_URI += "git://arago-project.org/git/projects/linux-omap3.git;protocol=git;branch=master
Jul 05 12:35:11 <stuk_gen> you can use 2 boards...but its the same...using bone there's not a perfect solution
Jul 05 12:35:12 <xxiao> why is this used for psp2.6.32 kernel?
Jul 05 12:37:03 <stuk_gen> _av500_: return to mem problem in my uEnv.txt i've _av500_: i set in my uEnvt.txt optargs="mem=99M@0x80000000 mem=384M@0x88000000" so its correct. its not correct?
Jul 05 12:37:55 <SmashLink> oh, ok, i found information about the u-boot http://en.wikipedia.org/wiki/Das_U-Boot
Jul 05 12:39:31 <_av500_> stuk_gen: if you see it in /proc/cmdline its ok
Jul 05 12:40:14 <stuk_gen> _av500_: ok, but i've no idea how to setup if uEnv.txt not works
Jul 05 12:40:42 <_av500_> add it to some of the other lines
Jul 05 12:40:48 <_av500_> aholler wrote uenv, he knows
Jul 05 12:40:52 <aholler> see the link SmashLink posted
Jul 05 12:41:24 <aholler> printenv in u-boot
Jul 05 12:41:50 <aholler> maybe your u-boot doesn't support optargs
Jul 05 12:42:20 <stuk_gen> aholler: i'm using angstrom
Jul 05 12:42:28 <aholler> or even uEnv.txt, who knows…
Jul 05 12:42:51 <stuk_gen> no uEnv.txt works its change the resolution correctly
Jul 05 12:43:08 <aholler> try without the quotes, they shouldn't be needed
Jul 05 12:43:44 <aholler> and be sure not to use a windows-editor
Jul 05 12:44:17 <stuk_gen> aholler: i'm using scite
Jul 05 12:44:36 <stuk_gen> i try optargs=mem=99M@0x80000000 mem=384M@0x88000000 consoleblank=0
Jul 05 12:44:49 <aholler> scite exist for almost every os
Jul 05 12:44:58 <stuk_gen> scite for windows :D
Jul 05 12:45:13 <aholler> ask ti-people
Jul 05 12:45:43 <aholler> they didn't want to support that
Jul 05 12:46:24 <stuk_gen> uff don't work
Jul 05 12:47:11 <SmashLink> and why does the beaglebone only boot from sd card?
Jul 05 12:47:32 <stuk_gen> because haven't nand
Jul 05 12:48:37 <SmashLink> but if it was supported, you could boot from an external flash drive, from ethernet...
Jul 05 12:49:06 <stuk_gen> its support
Jul 05 12:49:11 <stuk_gen> but after reading uBoot
Jul 05 12:49:17 <stuk_gen> and uBoot is in sd card :)
Jul 05 12:49:28 <SmashLink> and why not reading uBoot from an external source?
Jul 05 12:49:46 <stuk_gen> tell to TI people
Jul 05 12:49:55 <aholler> I've talked about support for windows-editors
Jul 05 12:50:06 <aholler> just to not been misunderstood
Jul 05 12:51:53 <stuk_gen> aholler: i've try setenv optargs 'mem=99M@0x80000000 mem=384M@0x88000000'
Jul 05 12:51:55 <stuk_gen> and boot
Jul 05 12:52:05 <stuk_gen> but in cmdline there isn't mem...
Jul 05 12:52:46 <aholler> as said, ask the the-ti-maintainer for u-boot for further help
Jul 05 12:52:46 <woglinde> stuk_gem then something else is booted
Jul 05 12:53:00 <woglinde> use the evn.txt
Jul 05 12:53:04 <woglinde> or what it was called
Jul 05 12:53:10 <stuk_gen> woglinde: not works
Jul 05 12:53:21 <stuk_gen> woglinde: i've just try using optargs="mem=99M@0x80000000 mem=384M@0x88000000"
Jul 05 12:53:27 <stuk_gen> in uEnv.txt but not works
Jul 05 12:53:59 <_av500_> pastebin your uenv.txt
Jul 05 12:54:39 <stuk_gen> _av500_: http://pastebin.com/pFDim8jY
Jul 05 12:54:53 <stuk_gen> no
Jul 05 12:55:14 <stuk_gen> _av500_: http://pastebin.com/PieU28kR
Jul 05 12:55:18 <stuk_gen> this is correct
Jul 05 13:09:21 <xxiao> koen: why is psp-omap2.6.32 kernel recipe is SRC_URI=linux-omap3.git?
Jul 05 13:09:47 <xxiao> ls
Jul 05 13:12:53 <koen> xxiao: stop trolling
Jul 05 13:12:54 <koen> SRC_URI += "git://arago-project.org/git/projects/linux-omap3.git;protocol=git;branch=master \
Jul 05 13:13:01 <koen> that's the psp kernel
Jul 05 13:13:23 <xxiao> but that's the head instead of 2.6.32
Jul 05 13:13:32 <xxiao> the new linux-omap3 are on 2.6.37?
Jul 05 13:14:18 <koen> it checks out a specific revision
Jul 05 13:14:29 <koen> I don't know about 'new' linux-omap3, ask TI
Jul 05 13:16:12 <xxiao> thanks. i'm remote update arago linux-omap3 and check if that sha is for 2.6.32
Jul 05 13:17:10 <koen> next time start with that instead of trolling here
Jul 05 13:17:40 <xxiao> koen: i did not intend to troll, just asking you, what make you think i'm trolling
Jul 05 13:17:52 <xxiao> if so sorry for that
Jul 05 13:20:01 <woglinde> xxiao why you need .32 at all?
Jul 05 13:20:34 <xxiao> woglinde: want to add bluetooth to an existing product which runs 2.6.32
Jul 05 13:20:49 <xxiao> and TI's bluetooth code is geared towards that version as well
Jul 05 13:21:03 <woglinde> poor customer
Jul 05 13:21:17 <xxiao> ls
Jul 05 13:21:34 <woglinde> command not found
Jul 05 13:24:22 <xxiao> :)
Jul 05 13:24:45 <stuk_gen> koen: have you the recipes for build from oe the demoimage of angstrom?
Jul 05 13:37:04 <joelagnel> on stack traces, I might be missing a kernel config option here, but
Jul 05 13:37:13 <joelagnel> /proc/../stack for any process just gives me a "0xFFFFFFFF" without a stack trace
Jul 05 13:37:21 <joelagnel> CONFIG_STACKTRACE=y
Jul 05 13:37:44 <joelagnel> any clue?
Jul 05 13:42:32 <joelagnel> hrmph
Jul 05 13:51:46 <joelagnel> and, /proc/kallsyms shows me all 0s for memory addresses:
Jul 05 13:51:51 <joelagnel> 0000000000000000 t pci_mmcfg_check_hostbridge
Jul 05 13:51:51 <joelagnel> 0000000000000000 t pci_mmcfg_amd_fam10h
Jul 05 13:51:53 <joelagnel> ..etc
Jul 05 13:52:30 <aholler> zeroMAP
Jul 05 13:52:47 <aholler> so it already does exist ;)
Jul 05 13:53:00 <joelagnel>  :)
Jul 05 14:12:46 <koen> stuk_gen: go to the angstrom website, click 'developers', follow instructions
Jul 05 14:12:55 <koen> (I remember saying that before)
Jul 05 14:13:10 <stuk_gen> koen: i just build the systemd-GNOME-image
Jul 05 14:13:24 <stuk_gen> koen: but this haven't the package for dsp and all stuff
Jul 05 14:13:45 <stuk_gen> koen: i have to manually set up theme...but someone said to me that you recently have build the demo image
Jul 05 14:14:03 <stuk_gen> koen: that have all feature support dsp,powervr ecc..
Jul 05 14:14:23 <stuk_gen> koen: so what recipes have you use to support this?
Jul 05 14:24:46 * [IDC]Dragon moves a DMAI discussion from #panda to here
Jul 05 14:25:27 * _av500_ grabs beer and follows
Jul 05 14:25:28 <[IDC]Dragon> av500, I'm here
Jul 05 14:26:01 <_av500_> so, the underlying CE allows you to attach an ID to a decode frame and get the reordered ID back
Jul 05 14:26:18 <[IDC]Dragon> looking at DMAI buffers, I see no timestamps
Jul 05 14:26:30 <_av500_> but you get the reordered buffer anyway, so you can use the buffer address as well
Jul 05 14:26:32 <[IDC]Dragon> or something for private data
Jul 05 14:26:42 <_av500_> [IDC]Dragon: url for me?
Jul 05 14:27:02 <[IDC]Dragon> I have an url to my HD
Jul 05 14:27:13 <_av500_> well, keep a reorder hashlist and assign timestams to buffer addrs
Jul 05 14:27:32 <[IDC]Dragon> I had that idea too...  ;-)
Jul 05 14:27:53 <[IDC]Dragon> but if there's a less clumsy way, I woudn't mind
Jul 05 14:30:10 <_av500_> I do it this way
Jul 05 14:30:23 <_av500_> since I also keep the frame type
Jul 05 14:31:01 <_av500_> and can attach more data if I need to
Jul 05 14:31:51 <[IDC]Dragon> with CE, the attached ID could be a pointer, or index to your extensive metadata, right?
Jul 05 14:32:01 <_av500_> its a uint32
Jul 05 14:32:09 <[IDC]Dragon> see
Jul 05 14:32:33 <_av500_> but there are some codecs that e.g. do not allow id==0
Jul 05 14:32:47 <_av500_> so I just use the address as ID (yes, useless) and my hashlist
Jul 05 14:33:12 <[IDC]Dragon> ok, anyway
Jul 05 14:33:12 <angs> is the maximum current sink for beagleboard-xm's GPIO 8mA?
Jul 05 14:33:44 <_av500_> [IDC]Dragon: but gst-ti does not handle that already?
Jul 05 14:33:58 <_av500_> since its on top of dmai
Jul 05 14:34:13 <[IDC]Dragon> it handles it in a stupid way
Jul 05 14:34:39 <[IDC]Dragon> it adds a fixed frame duration for every output frame
Jul 05 14:34:52 <_av500_> o_O
Jul 05 14:35:18 <[IDC]Dragon> my input is already timestamped, so all I need is to safely get this across
Jul 05 14:35:32 <_av500_> right
Jul 05 14:35:36 <[IDC]Dragon> I have a working patch, but it fails for reordering
Jul 05 14:35:56 <_av500_> extra fun when the timestamps are either PTS or DTS
Jul 05 14:35:57 <[IDC]Dragon> (so, better not call it a working patch)
Jul 05 14:36:09 <[IDC]Dragon> they are pts
Jul 05 14:36:40 <_av500_> lucky you
Jul 05 14:37:37 <[IDC]Dragon> is it safe to assume that an output buffer passed to a DMAI Vdec2_process() will actually be used for *that* output data?
Jul 05 14:38:02 <[IDC]Dragon> it could as well be just added to an internal pool
Jul 05 14:38:30 <mru> you might get another buffer back
Jul 05 14:38:37 <mru> decode order != display order
Jul 05 14:39:06 <[IDC]Dragon> I may get an earlier buffer back first
Jul 05 14:39:20 <[IDC]Dragon> that's expected
Jul 05 14:39:45 <_av500_> yes
Jul 05 14:40:10 <SmashLink> in this thread i am reading there might be various boot modes on the beaglebone if you configure SYSBOOT correctly http://e2e.ti.com/support/dsp/sitara_arm174_microprocessors/f/791/t/154192.aspx
Jul 05 14:40:50 <SmashLink> does anyone know any source about how to configure this?
Jul 05 14:40:57 <[IDC]Dragon> my question is, would the mapping of input and output buffers be still the same as in the call?
Jul 05 14:41:04 <_av500_> yes
Jul 05 14:41:24 <_av500_> at least for all recent codecs that I know
Jul 05 14:41:29 <[IDC]Dragon> ok
Jul 05 14:42:03 <_av500_> I had codecs in the past that kept internal ref buffers
Jul 05 14:42:21 <_av500_> so you wouldget the reordered frames but not reordered buffers
Jul 05 14:42:27 <_av500_> but that was davinci times
Jul 05 14:50:55 <SmashLink> i don't see anything about boot modes on the beaglebone SRM... does anyone know which trm are they talking about in the thread?
Jul 05 14:51:35 <mru> the am3359 trm
Jul 05 14:51:45 <_av500_> not SRM, TRM
Jul 05 14:53:44 <SmashLink> what's the difference?
Jul 05 14:53:48 <mru> nor URM
Jul 05 14:53:55 <mru> VRM is right out
Jul 05 15:06:25 <LetoThe2nd> mru: don't count to 2, nor to 4...
Jul 05 15:06:59 <mru> +1 point to LetoThe2nd
Jul 05 15:11:11 <dm8tbr> count two times to 26
Jul 05 15:11:29 <mru> 54!
Jul 05 16:53:15 <may> I have a gpio question.
Jul 05 16:53:22 <may> cd /sys/class/gpio
Jul 05 16:53:32 <may> echo 4 > export
Jul 05 16:53:38 <may> Results in
Jul 05 16:53:49 <may> bash: echo: write error: Device or resource busy
Jul 05 16:54:05 <may> So someone else is using gpio.
Jul 05 16:54:17 <may> The question is, how do I discover who is using it?
Jul 05 16:54:51 <agmlego> may: Might be a protocol driver, like I2C or SPI, or somesuch. I believe there are tutorials on how to discover that online.
Jul 05 16:55:55 <niro> try lsof
Jul 05 16:57:02 <may> I'm installing lsof to try it out.
Jul 05 16:57:16 <agmlego> Should just be there already...
Jul 05 16:57:27 <agmlego> What distro/image are you using, on what device?
Jul 05 16:57:40 <may> I've been poking around netland, but haven't come across any tutorials.
Jul 05 16:57:55 <may> I'm running on a Beagle Xm.
Jul 05 16:58:08 <agmlego> What are you running on it?
Jul 05 16:58:19 <may> I'm upgrading from a 2.6.32 kernel to 3.0.14, both Angstrom
Jul 05 16:58:23 <niro> I don't have lsof on my angstrom
Jul 05 16:58:28 <agmlego> Interesting.
Jul 05 16:58:46 <agmlego> Oh, right. Angstrom uses busybox.
Jul 05 16:59:27 <niro> And that ofcourse...
Jul 05 16:59:44 <may> opkg install lsof works
Jul 05 16:59:53 <agmlego> Yeah.
Jul 05 17:00:05 <may> lsof returns some 5,000 files.  What should I be looking for?
Jul 05 17:00:10 <agmlego> Just took me by surprise that something that I think of as basic is not there.
Jul 05 17:00:20 <agmlego> may: Drivers that use GPIO.
Jul 05 17:00:42 <niro> export?
Jul 05 17:00:50 <niro> lsof |grep export
Jul 05 17:01:20 <may> niro: That returns nother.
Jul 05 17:01:38 <may> lsof | grep gpio returns /sys/class/gpio
Jul 05 17:01:55 <may> Oops, try again, lsof | grep export returns nothing.
Jul 05 17:02:24 <may> cat /sys/kernel/debug/gpio returns
Jul 05 17:02:56 <may> GPIOs 0-31, gpio:  gpio-4   (user                ) in  lo
Jul 05 17:03:09 <may> The value tracks the state of the switch.
Jul 05 17:03:31 <may> How does one got about reading the USER switch on an xM?
Jul 05 17:09:36 <aholler> lsof doesn't help
Jul 05 17:11:43 <Steve123> hello I'm using the beagleboard with the 5MP but the image comes out dark, how do I get the white balance to work? is there a patch?
Jul 05 17:12:21 <agmlego> The 5MP what?
Jul 05 17:12:48 <Steve123> LI-5M03 (mt9p031)
Jul 05 17:12:59 <Steve123> the camera :)
Jul 05 17:13:48 <Steve123> the webcam adjust by itself depending on the light but the 5MP camera does not
Jul 05 17:16:08 <agmlego> Steve123: What does it say to do in the camera manual?
Jul 05 17:16:26 <buq2> MT9P031 does not have auto exposure
Jul 05 17:17:04 <buq2> Steve123: you can probably control the exposure V4L2
Jul 05 17:17:10 <buq2> using V4L2
Jul 05 17:17:26 <Steve123> thanks buq2, how can I do that?
Jul 05 17:17:39 <Steve123> I been looking but I didn't see anything
Jul 05 17:17:50 <Steve123> I might been searching with the wrong keywords
Jul 05 17:19:01 <buq2> I used LI-5M03 with custom/modified sensor driver, so I'm not sure if the version which you are using supports it
Jul 05 17:20:44 <Steve123> where did you got the driver?
Jul 05 17:21:01 <buq2> I can not currently access my source codes, but I could probably send the driver for you
Jul 05 17:21:27 <buq2> I think that it requires kernel 2.6.32 or 2.6.34
Jul 05 17:21:50 <Steve123> i'm using kernel 2.6.34
Jul 05 17:22:13 <Steve123> where did you found this information to adjust the exposure?
Jul 05 17:22:56 <buq2> From the MT9P031 datasheet
Jul 05 17:23:43 <buq2> http://www.aptina.com/assets/downloadDocument.do?id=865
Jul 05 17:28:32 <angs> what is the current sink range for beagleboard-xm's GPIO?
Jul 05 17:29:23 <agmlego> Probably around 20mA or less.
Jul 05 17:29:30 <agmlego> It ought to say in the manual.
Jul 05 17:35:31 <angs> agmlego: thank you, I saw voltage info on the manual, but could not see the current range
Jul 05 17:35:38 <angs> I will read it again
Jul 05 17:35:59 <Echilon> I am thinking about building a custom musical box, I have looked at the BeagleBoard and have consulted another online forum and have determined that the BeagleBoard might be abit too much of what I need
Jul 05 17:36:02 <agmlego> angs: Note that the manual itself *may* not have that. YOu might have to go into the datasheet for the processor itself.
Jul 05 17:36:15 <Echilon> any suggestions of what other boards to look at?
Jul 05 17:36:29 <agmlego> Echilon: Define "bit too much"
Jul 05 17:36:44 <agmlego> Echilon: There are lots and lots and lots of processors out there capable of playing music.
Jul 05 17:36:57 <angs> agmlego: does GPIO directly connected to the processor?
Jul 05 17:37:05 <djlewis> Echilon: can you be more specific?
Jul 05 17:37:20 <Echilon> agmlego: abit too much as in I do not need something that is running at 1GHZ
Jul 05 17:37:35 <agmlego> angs: Uh, yeah.
Jul 05 17:37:45 <agmlego> Echilon: So get something slower.
Jul 05 17:37:55 <angs> agmlego: thanks
Jul 05 17:38:21 <Echilon> djlewis: I would like a board that has stereo out, enough rom to store a few 10sec clips and probably about 100MHZ in speed
Jul 05 17:38:29 <jacekowski> Echilon: bb doesn't run at 1gh
Jul 05 17:38:35 <agmlego> angs: GPIO in this (and most other cases) is provided by physical pins on the processor itself.
Jul 05 17:38:42 <Echilon> agmlego: any suggestions as to which boards to look at?
Jul 05 17:38:47 <jacekowski> Echilon: beaglebone
Jul 05 17:38:52 <jacekowski> Echilon: raspberry pi
Jul 05 17:39:08 <agmlego> Echilon: Sounds like you are looking for something like the Stellaris or other Cortex-M series, not the A-series the BeagleBoard has.
Jul 05 17:39:14 <jacekowski> netduino may cut it
Jul 05 17:39:28 <jacekowski> arduino with mp3 shield
Jul 05 17:39:29 <Echilon> jacekowski: I looked at that as well, but I don't need something to run a full out OS on it...
Jul 05 17:39:29 <agmlego> Echilon: Maple, any of the STM boards, etc. and so on.
Jul 05 17:40:04 <Echilon> okay writing these down...
Jul 05 17:40:11 <jacekowski> Echilon: raspberry may be the best thing really
Jul 05 17:40:18 <jacekowski> Echilon: you won't get anything else that cheap
Jul 05 17:40:27 <jacekowski> Echilon: when they finally start selling them
Jul 05 17:40:47 <Echilon> jacekowski: yes...when they start selling them
Jul 05 17:41:04 <jacekowski> well, it's meant to be august
Jul 05 17:42:18 <Echilon> agmlego: STMicro was another one I was looking at as well, though work has been occupying my time for the last few weeks
Jul 05 17:42:33 <Echilon> I will do more research on these suggestions, thank you all so much ^^
Jul 05 17:46:29 <jacekowski> hmm, cheap mp3 player
Jul 05 17:46:32 <jacekowski> that can run rockbox
Jul 05 17:46:40 <agmlego> jacekowski: Sansa.
Jul 05 17:46:56 <jacekowski> nah, i'm just sayign that would be good idea
Jul 05 17:47:03 <jacekowski> but it's bit too late
Jul 05 17:47:04 <agmlego> Yeah, but he left.
Jul 05 17:47:54 <niro> http://www.rikomagic.co.uk/ this looks like good fun
Jul 05 17:56:45 <djlewis> sounds like one of those cheapy mp3 players would work.
Jul 05 17:56:56 <djlewis> take the board out and pretend it is an embedded project
Jul 05 18:17:21 <xxiao> oebb.sh config will erase all my local committed changes?
Jul 05 18:17:36 <xxiao> i wasted 2 hours, it's all gone
Jul 05 18:18:48 <xxiao> added a new board, MACHINE=newboard ./oebb.sh config newboard
Jul 05 18:19:10 <xxiao> after that it seems all my local git-commit disappeared
Jul 05 18:19:17 <xxiao> which are under meta-ti
Jul 05 18:20:20 <aholler> try git log -g
Jul 05 18:21:29 <xxiao> aholler: it shows my commit between two top commits from angstrom
Jul 05 18:21:41 <xxiao> that two commits from angstrom are exactly the same
Jul 05 18:21:47 <aholler> so just checkout the commit-number
Jul 05 18:22:16 <xxiao> git checkout mycommitsha?
Jul 05 18:22:45 <aholler> yes, then git format-patch HEAD~N afterwards, and save your patches elsewhere
Jul 05 18:22:48 <xxiao> why does that happen? or did i do something wrong?
Jul 05 18:22:58 <aholler> I don't know
Jul 05 18:23:21 <aholler> look at the oe-script about what it does
Jul 05 18:24:48 <aholler> but as long as git gc wasn't run (either manually or automatically) you can still checkout your old commits, even if they don't appear in the tree anymore
Jul 05 18:25:07 <xxiao> aholler: thanks. never used that feature, time-saver
Jul 05 18:25:59 <aholler> lifesave if you've done a git reset --hard through a mistake
Jul 05 18:26:17 <aholler> or similiar
Jul 05 18:27:22 <xxiao> just tried it again, adding a new platform, do oebb.sh config, conf/auto.conf etc are updated
Jul 05 18:27:37 <xxiao> afterwards my new-platform-commit in meta-ti is totally gone
Jul 05 18:27:50 <xxiao> so bitbake complains it can not compile the new platform
Jul 05 18:40:23 <koen> xxiao: try reading the output on your screen
Jul 05 18:40:56 <koen> xxiao: I can't decide if you're trolling or just not making an effort to find out for yourself
Jul 05 18:41:44 <mranostay> koen: Friday mode already? :)
Jul 05 18:43:39 <xxiao> koen: i never intended to troll
Jul 05 18:43:43 <koen> mranostay: i've lost track of time lately, moving house and all
Jul 05 18:44:06 <koen> xxiao: then start with simple things like reading the output on your screen before asking here
Jul 05 18:46:15 <xxiao> MACHINE=newboard ./oebb.sh config newboard, it outputs lots of repos, and i did not spot anything useful
Jul 05 18:46:30 <xxiao> to the point it tells me local changes are going to disappear
Jul 05 18:46:55 <xxiao> i might be missing something and asking stupid questions, but i did try, just may not as smart as you
Jul 05 18:49:02 <xxiao> and i don't know what made you think i'm always trolling here, next time please just ignore my question
Jul 05 18:53:32 * mranostay yawns
Jul 05 18:54:13 * djlewis tosses mranostay a jolt cola
Jul 05 18:55:09 <mranostay> already a few cups away from the LD50 for caffeine
Jul 05 19:48:07 <kris92_> Hi ! do you know how to retrieve u-boot version ?
Jul 05 19:49:10 <mranostay> boot up the board?
Jul 05 19:49:44 <kris92_> once logged in ?
Jul 05 19:54:25 <mranostay> strings /boot/u-boot.img | grep U-boot
Jul 05 19:56:51 <kris92_> thks a lot ! U-Boot 2012.04.01-00005-g45939b4
Jul 05 20:01:26 <jacekowski> mranostay: how is that meant to work
Jul 05 20:01:55 <jacekowski> x-loader can read ext2/3/4?
Jul 05 20:02:23 <mranostay> ah yes
Jul 06 01:29:41 <romaxa> does anyone has expirience with android base DisplayHardware,FramebufferNativeWindow and HWComposer apis and and specifically making it work on device with automatic FB updates disabled...
Jul 06 01:39:56 <mranostay> have nfc what those are
Jul 06 01:42:42 <romaxa> rcn-ee: ping
Jul 06 02:03:34 <prp^2> koen: ping
Jul 06 02:14:43 <rcn-ee> romaxa, pong
Jul 06 02:44:01 <romaxa> rcn-ee: are you familiar with SGX android drivers?
Jul 06 02:44:23 <romaxa> rcn-ee: or maybe you know who is familiar?
Jul 06 02:44:58 <romaxa> rcn-ee: I'm trying to run simple test examples but nothing get rendered because automatic display updates are disabled on that device
Jul 06 02:45:24 <rcn-ee> sorry, not really.. haven't touched android devel since gingerbread..
**** ENDING LOGGING AT Fri Jul 06 02:59:58 2012